**Meet the editors**

Dr. Likun Pan received his PhD in 2005 at Nanyang Technological University, Singapore, and currently works at the Department of Physics, East China Normal University as a professor. His research interest includes the synthesis and properties of nanomaterials and their applications in energy and environmental fields. He has published more than 180 journal articles with over 4000

citations and his current H-index is 36. Dr. Pan has severed as editorial board member of several international journals and as technical chair or general secretary in several international conferences.

Prof. Guang Zhu received his PhD in 2012 at East China Normal University, China. He was a postdoctoral fellow at Northwestern University, USA, and later joined Suzhou University as a associate professor, China. His current research interests include materials synthesis, characterization, and applications in photocatalysis, electrosorption, and solar cells.

## Contents

## **Preface XIII**


Heng Wu and Xinhua Zhu

Chapter 6 **Synthesis, Crystal Structure, and Physical Properties of the Perovskite Iridates 185** Yunqi Cai, Yan Li and Jinguang Cheng

## X Contents

## **Section 2 Perovskite Materials: Properties 219**


## Chapter 15 **Numerical Simulations on Perovskite Photovoltaic Devices 445** Bernabé Marí Soucase, Inmaculada Guaita Pradas and Krishna R. Adhikari

**Section 2 Perovskite Materials: Properties 219**

**5d Perovskite Iridates 221**

W.J. James

**VI** Contents

Hitoshi Ohsato

Carlito S. Ponseca Jr.

**Trihalide Perovskites 377**

**Section 3 Perovskite Materials: Applications 401**

Shuzi

Kuen-Feng Lin

Chapter 7 **Metal–Insulator Transitions and Non-Fermi Liquid Behaviors in**

J.B. Yang, M.S. Kim, T. F. Creel, H. Zhao, X.G. Chen, W.B. Yelon and

Abhijit Biswas, Ki-Seok Kim and Yoon Hee Jeong

Chapter 9 **Microwave Dielectrics with Perovskite-Type Structure 281**

Chapter 11 **Charge Carrier Dynamics in Organometal Halide Perovskite**

Chapter 12 **Photoexcitations and Emission Processes in Organometal**

Chapter 13 **Optical Absorption, Charge Separation and Recombination**

Chapter 14 **Optical, Excitonic, and Electronic Properties of CH3NH3PbI3**

**Thin Films and Their Application in Photovoltaics 423** Sheng Hsiung Chang, Hsin-Ming Cheng, Sheng-Hui Chen and

Andrea Mura and Giovanni Bongiovanni

**Probed by Time-Resolved Electrical Measurements 355**

Michele Cadelano, Michele Saba, Nicola Sestu, Valerio Sarritzu, Daniela Marongiu, Feipeng Chen, Roberto Piras, Francesco Quochi,

**Dynamics in Pb and Sn/Pb Cocktail Perovskite Solar Cells and Their Relationships to the Photovoltaic Properties 403** Shen Qing, Ogomi Yuhei, Toyoda Taro, Yoshino Kenji and Hayase

Chapter 10 **ESR and Magnetization Studies of Bi-manganites 331**

Rajender Singh and Ramesh Ade

Chapter 8 **Structural, Magnetic and Transport Properties of B-Site Substituted Perovskite La0.7Sr0.3MnO3 261**


## Preface

Perovskite materials have attracted much attention because of their promising physical properties in colossal magnetoresistance, superconductivity, multi-ferroelectricity, metal-in‐ sulator transition, charge/orbital ordering, photoelectricity, etc. Over the past two decades, remarkable efforts have been conducted worldwide to explore various techniques to pre‐ pare perovskite materials for many applications. Difference in properties of perovskite structured materials has provided new insight and direction for researchers to meet the dif‐ ferent applications. Herein, this book was written in part as a survey of the state of the art research activities that focus on the rational synthesis, structural characterization, proper‐ ties, and applications of perovskite materials. In part, it is also an effort to bridge this knowl‐ edge gap and to better understand perovskite materials characterization and their applications.

The book with 21 chapters has been organized to accommodate the vision outlined above via summarizing the current state of the perovskite materials: synthesis, characterization, properties, and applications. Most chapters have been structured to include (1) a review on the actual knowledge and (2) cutting-edge research results. Thus this book is an essential source of reference for scientists with research fields in energy, physics, chemistry and mate‐ rials. It is also suitable reading matter for graduate students.

The Book Editors would like to thank all who kindly contributed their high quality chapters for this book and to InTech staff for their kind help and cooperation. We are also indebted to the InTech editorial and production team.

## **Likun Pan**

Engineering Research Center for Nanophotonics and Advanced Instrument, Ministry of Education, Department of Physics East China Normal University, P. R. China

## **Guang Zhu**

Anhui Key Laboratory of Spin Electron and Nanomaterials Suzhou University, Suzhou, P. R. China

**Perovskite Materials: Synthesis and Characterization**

## **Solid-State Mechanochemical Syntheses of Perovskites**

## Piotr Dulian

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61521

## **Abstract**

The chapter presents the possibility of applying high-energy ball milling techniques to car‐ ry out the synthesis of ceramics with perovskite structure, thereby eliminating prolonged use of high temperatures in their preparation.

On the examples of alkaline-earth metal perovskites, an influence of the most significant parameters of mechanochemical treatment on their forming process and product quality was illustrated. For the first time, it was done that the contamination of the product de‐ rived from the attrition of the grinding media and internal parts of the vial can constitute modifiers of the functional properties of produced electroceramics.

Dielectric characteristics of mechanochemically produced materials as well as using hightemperature solid synthesis were compared.

**Keywords:** Mechanochemistry, High-energy Ball Milling, Perovskites, Complex Oxides, Electroceramics

## **1. Introduction**

The family of chemical compounds with perovskite-type structure due to the unique electrical properties comprises a broad range of electrotechnical materials – dielectrics by semiconduc‐ tors, superionic conductors, conductors with combined ionic and electron conductivity to high-temperature superconductors [1, 2].

Moreover, as is known, these compounds in certain temperature ranges have piezoelectric, pyroelectric, ferroelectric, antiferroelectric, paraelectric, ferromagnetic, or paramagnetic properties [3–6]. Also important is the simplicity of their crystalline structure, chemical composition, and the synthesis of these compounds in monocrystalline or polycrystalline form. It is easy to modify the structure and thus the properties of perovskites. Even a slight change of their ideal crystal structure and chemical composition may result in the appearance of new,

not only electrical but also other, properties such as catalytic or mechanical. Therefore, it is very important to select the method of their production. The synthesis of polycrystalline titanates with perovskite structure, due to the fact that these compounds are hardly fusible materials, is carried out at high temperatures by solid-phase reaction. However, such a temperature causes the appearance of sinters and agglomerates what hinder fine crystalline product formation. This is extremely important because the morphology and grain size in electroceramics directly affects on their properties [7–12]. This problem can be solved by different ways, e.g., using a sol–gel method [13]. This technique owing to the thorough mixing of the precursor materials in solution and the relatively lower temperature of crystallization (1,200 K) allows to obtain a homogenous material with small, fine crystals and excellent chemical stoichiometry. Unfortunately, the sol–gel method is complex and requires advanced, very clean equipment, and organometallic reagents that are not only expensive but also environmentally hazardous. There are also known other synthesis techniques such as coprecipitation or hydrothermal methods. However, many of these methods enable to synthesize perovskite-type ceramics with fine crystals; they are unpopular because their complexities and costs preclude their use in a large-scale industrial fabrication [14–20]. Alternatively, this kind of ceramics can be produced by high-energy ball milling at room temperature [21–28]. This technique leads to the activation and/or synthesis of new compounds. Activation of solid powders in this case is based on the high degree of fragmentation and a large number of structural defects. However, in order to obtain the crystalline milling products, sometimes the subsequent heat treatment is needed. Then, much lower temperature may be used than in conventional methods. Moreover, using the mechanochemical method, it is possible to control the grain morphology of ceramics by selection of appropriate process parameters. Such a simplicity and large control of process parameters make it an excellent alternative to expensive and complex manufacturing techniques of advanced ceramic materials.

## **2. General aspects of mechanochemistry**

The essence of mechanical treatment is the impact of moving grinding media with grains of ground material and the interaction between the grains of powder. During these processes depending on the type of the mill and the applied milling parameters, the energy supplied to the material is in the range of 0.1 to 100 MJ/kg. Transfer of this energy is precisely localized in the collision zone at the moment of the collision between the grinding media. During the collision, the kinetic energy of the grinding media is absorbed by a small volume (approx. 1 mm3 of ground powder) and is immediately converted into elastic energy. Resulting stresses cause the destruction of the ground material. Depending on the physical nature of the ground powder, mainly hardness and thermodynamic conditions, the cracks of crystals occur, resulting in the reduction of the grain size, and/or a mutual merging of the particles.

All these phenomena intensify the diffusion processes in solids accelerating the chemical reactions. It is caused by the forces of collisions, strike/hit, compression, and friction occurring between the grinding media and the ground material and between the grinding media and the walls of the reactor [29–35]. In consequence, the reactions take place without the need of diffusion of substrates through the product layer because interfacial contact of ground materials is periodically renewed.

## **2.1. Process parameters of mechanochemical treatment**

not only electrical but also other, properties such as catalytic or mechanical. Therefore, it is very important to select the method of their production. The synthesis of polycrystalline titanates with perovskite structure, due to the fact that these compounds are hardly fusible materials, is carried out at high temperatures by solid-phase reaction. However, such a temperature causes the appearance of sinters and agglomerates what hinder fine crystalline product formation. This is extremely important because the morphology and grain size in electroceramics directly affects on their properties [7–12]. This problem can be solved by different ways, e.g., using a sol–gel method [13]. This technique owing to the thorough mixing of the precursor materials in solution and the relatively lower temperature of crystallization (1,200 K) allows to obtain a homogenous material with small, fine crystals and excellent chemical stoichiometry. Unfortunately, the sol–gel method is complex and requires advanced, very clean equipment, and organometallic reagents that are not only expensive but also environmentally hazardous. There are also known other synthesis techniques such as coprecipitation or hydrothermal methods. However, many of these methods enable to synthesize perovskite-type ceramics with fine crystals; they are unpopular because their complexities and costs preclude their use in a large-scale industrial fabrication [14–20]. Alternatively, this kind of ceramics can be produced by high-energy ball milling at room temperature [21–28]. This technique leads to the activation and/or synthesis of new compounds. Activation of solid powders in this case is based on the high degree of fragmentation and a large number of structural defects. However, in order to obtain the crystalline milling products, sometimes the subsequent heat treatment is needed. Then, much lower temperature may be used than in conventional methods. Moreover, using the mechanochemical method, it is possible to control the grain morphology of ceramics by selection of appropriate process parameters. Such a simplicity and large control of process parameters make it an excellent alternative to expensive

4 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

and complex manufacturing techniques of advanced ceramic materials.

The essence of mechanical treatment is the impact of moving grinding media with grains of ground material and the interaction between the grains of powder. During these processes depending on the type of the mill and the applied milling parameters, the energy supplied to the material is in the range of 0.1 to 100 MJ/kg. Transfer of this energy is precisely localized in the collision zone at the moment of the collision between the grinding media. During the collision, the kinetic energy of the grinding media is absorbed by a small volume (approx. 1 mm3 of ground powder) and is immediately converted into elastic energy. Resulting stresses cause the destruction of the ground material. Depending on the physical nature of the ground powder, mainly hardness and thermodynamic conditions, the cracks of crystals occur,

resulting in the reduction of the grain size, and/or a mutual merging of the particles.

All these phenomena intensify the diffusion processes in solids accelerating the chemical reactions. It is caused by the forces of collisions, strike/hit, compression, and friction occurring between the grinding media and the ground material and between the grinding media and the walls of the reactor [29–35]. In consequence, the reactions take place without the need of

**2. General aspects of mechanochemistry**

Fabrication of different materials by a mechanochemical treatment is a complex process because it is influenced by many factors. Generally, they are connected with energetics and/or the environment of milling. The amount of energy supplied to the material during mechanical treatment mainly depends on the type of mill. There are various kinds of con‐ struction solutions of the mills [31, 36–37]. The type of mill should be chosen taking into account the advantages and disadvantages of each device. The decisive parameter for the kinetics of the mechanochemical processes is the rotation speed of the reactor [38]. The rotation speed or impact, in the case of vibratory mills, transfer directly into interaction frequency of grinding media with powder particles and their speed inside the reactor. Milling energy highly depends on the BPR factor (*ball to powder ratio*), which expresses the ratio (e.g., mass) of grinding media to the ground material. Although this relationship is not linear, this is related to the degree of the reactor filling with balls (a large number of them makes the movement of the balls more complex) [29, 39]. Grinding media lose their energy due to frequent collisions among them‐ selves. Too small or too large volume of ball in the vial reduces the efficiency of the milling process. It should be pointed that both high speed and the large number of balls cause increase in the reactor temperature. Higher temperature can be beneficial to the phenomenon, stimu‐ lates the diffusion of the atoms in the solids, and also increases the degree of ductility of the steel, which leads to a faster wear. All parameters relating to energetics of the grinding unit should be taken into account when planning the mechanical treatment process.

The selection of the process parameters directly affects the properties of the obtained phases. Depending on the desired final effect of high-energy, ball milling can/must be used different time, the atmosphere, and the medium of mechanochemical treatment. Determination of mechanochemical synthesis duration is rather simple – depends on the time required to form the desired phases. The processes associated with mechanical activation of solids, such as structure modification, deposition of active catalytic phase on carrier or simple communition need own individual milling time, therefore it must be selected experimentally.

Mechanochemical processes are often carried out under a protective atmosphere or in vacuum. This prevents milled material from the additional reactions with air components such as oxygen or nitrogen. Negative phenomenon of agglomeration of grains as a result of highenergy milling process can be reduced by using water or alcohol as a medium [21, 40–52].

The kinetics of mechanochemical reactions can be also controlled through various types of precursors. This is particularly important in the case of exothermic reactions, which very often have an explosive nature (SHS reaction) as in the case of metallothermic processes [53–55]. The use one of the substrates not in the form of oxide (i.e., as a ready reactant) but as hydroxosalt or carbonate needs heat to its stepwise decomposition into the oxide (endothermic process), increasing the time of availability of the reagent for the synthesis. This type of process is called by Avvakumow [56] as "soft mechanochemistry".

## **3. Direct mechanochemical syntheses of titanates**

The continuous desire to limit great inconvenience to the natural environment and the cost of preparation of a variety of functional materials and the need for materials with more "sophis‐ ticated" properties makes the mechanochemical treatment an interesting alternative technique to the conventional ones. The possibility of applying this method for the preparation of highquality ceramic materials with perovskite structure is presented below. In Figures 1 and 2 are shown the influence of various process parameters of high-energy ball milling and the nature of the reactants on the dynamics and properties of the obtained products, respectively.

**Figure 1.** Various milling conditions in mechanochemical syntheses of perovskites

## **4. Syntheses of perovskites of alkaline earth metals – MTiO3 (M = Ca, Sr, Ba)**

## **Mechanochemical synthesis of calcium titanate**

Synthesis of compounds with perovskite structure in the system CaO-TiO2 is carried out in the solid phase without the need of high-temperature processing (for processing details – see:

**Figure 2.** Different oxides' precursors to syntheses of complex oxides (CaTiO3)

**3. Direct mechanochemical syntheses of titanates**

6 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 1.** Various milling conditions in mechanochemical syntheses of perovskites

**Mechanochemical synthesis of calcium titanate**

**Ba)**

**4. Syntheses of perovskites of alkaline earth metals – MTiO3 (M = Ca, Sr,**

Synthesis of compounds with perovskite structure in the system CaO-TiO2 is carried out in the solid phase without the need of high-temperature processing (for processing details – see:

The continuous desire to limit great inconvenience to the natural environment and the cost of preparation of a variety of functional materials and the need for materials with more "sophis‐ ticated" properties makes the mechanochemical treatment an interesting alternative technique to the conventional ones. The possibility of applying this method for the preparation of highquality ceramic materials with perovskite structure is presented below. In Figures 1 and 2 are shown the influence of various process parameters of high-energy ball milling and the nature of the reactants on the dynamics and properties of the obtained products, respectively.

> Appendix A1). However, the formation of a crystalline product is highly conditioned by several milling process parameters (see: Section 2.1). The influence of parameters relating to milling energy, such as the BPR value and rotation speed on the product formation time, is illustrated in Figure 3. It is worth to pay attention to the synthesis time, which using the appropriate conditions can be only about 1 h, i.e., substantially shorter than the other methods of synthesis.

> Decrease in the values of rotation speed or BPR parameter increases the synthesis time. The type of TiO2 (rutile or anatase) is also important for the kinetics of process. Mechanochemically obtained calcium titanate is characterized by grain size less than 100 nm and good homogeneity in terms of particle size. In this case, it is not necessary to make the subsequent high-temper‐ ature treatment process, which eliminates the problem of agglomerates formation and an excessive non-uniform grain growth (Figure 4). This has a meaning for properties (e.g., a ferroelectric or catalytic) of materials.

> Commonly used wet environment, during the high-energy ball milling (ethanol or water), in order to reduce the negative phenomenon of agglomeration acts negatively on the course of the synthesis reaction because of hindering the reactants' phase contact. However, high-energy milling activates the reactants by particles size reduction and creation of the crystal defects which facilitates the diffusion of atoms, lowering the temperature of subsequent calcination process (see Figure 5).

**Figure 3.** XRD patterns of calcium titanate – illustration of milling conditions (BPR = 20:1 and 40:1; rpm = 500 and 1000) and two forms of TiO2 for CaTiO3 synthesis

**Figure 4.** SEM images of mechanochemically synthesized CaTiO3; (a) and (b) – different magnification

In the systems of CaCO3-TiO2 and Ca(OH)2-TiO2, synthesis is much more difficult. It is a consequence of two reactions: (1) and (2) or (1\*) and (2). Synthesis is limited by the decompo‐ sition rate of carbonate or calcium hydroxide to the oxide.

**Figure 5.** XRD patterns of CaO-TiO2 system after mechanochemical activation in different media (water and alcohol) and subsequent calcination

$$\begin{aligned} \text{CaCO}\_{3(s)} &\leftrightarrow \text{CaO}\_{(s)} + \text{CO}\_{2(g)} & \qquad \text{AH} \gg 0\\ \text{Ca(OH)}\_{2(s)} &\leftrightarrow \text{CaO}\_{(s)} + \text{H}\_2\text{O}\_{(g)} & \qquad \text{AH} \gg 0^\ast \end{aligned} \tag{1}$$

$$\text{CaO}\_{\text{(s)}} + \text{TiO}\_{\text{2(s)}} \rightarrow \text{CaTiO}\_{\text{3(s)}} \qquad \qquad \qquad \Lambda \text{H} \lhd 0 \tag{2}$$

The example of mechanochemical synthesis of calcium titanate in CaCO3-TiO2 system is presented in Figure 6.

As can be seen, in order to obtain a monophase product after 4 h of milling, a subsequent heat treatment at 800°C was necessary to use [40].

The preparation of other alkaline-earth metal titanates, e.g., barium and strontium, by a mechanochemical synthesis proceeds in analogy to the above-mentioned CaTiO3 example.

The abovementioned experimental results indicate that the method of mechanochemical synthesis may be used to produce high-quality perovskite ceramics. Thus, short time of the synthesis, the use of substrates in the oxide form, and a lack of using the high-temperature treatment significantly reduce both manufacturing costs and negative influence on environ‐ ment.

In the systems of CaCO3-TiO2 and Ca(OH)2-TiO2, synthesis is much more difficult. It is a consequence of two reactions: (1) and (2) or (1\*) and (2). Synthesis is limited by the decompo‐

**Figure 4.** SEM images of mechanochemically synthesized CaTiO3; (a) and (b) – different magnification

**Figure 3.** XRD patterns of calcium titanate – illustration of milling conditions (BPR = 20:1 and 40:1; rpm = 500 and 1000)

sition rate of carbonate or calcium hydroxide to the oxide.

8 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

and two forms of TiO2 for CaTiO3 synthesis

**Figure 6.** XRD patterns of calcium titanate formation: (a) up to 4 h of milling, (b) subsequent calcination at different temperatures

## **5. Modification of perovskite-type electroceramics properties**

Apart from the rapid and relatively simple synthesis of perovskite compounds, mechano‐ chemical method also allows to modify their chemical and physical properties. Such modifi‐ cations can be made by changing the chemical composition of perovskites, production of ceramics of an adequate morphology (grain size, defected, etc.), and also using as additives the contaminations originating from wear down of the grinding media (Examples 1–3).

## **5.1. Formation of solid solution by doping various cations**

## **Example 1. Ba1–***x***Sr***x***TiO3 (0.0 ≤** *x* **≤ 0.4)**

XRD patterns of the synthesized compounds are shown in Figure 7. They indicate the influence of the presence of Sr2+ on the crystallinity of products. Any phase of strontium oxide was found even in the case of strontium concentration equal to *x* = 0.4 in Ba1–xSrxTiO3.

Different ionic radii of Ba2+ (1.36 Å) and Sr2+ (1.16 Å) induce distortion of lattice. A close look at a slow scan of reflection (e.g., from 31.0° to 32.5°) shown as an inset which indicates that there is a slight shift of this reflection to higher 2θ angles. This confirms the substitution of Sr2+ ions in the BaTiO3 lattice. It might be concluded that single phases of BaTiO3, Ba0.8Sr0.2TiO3, and Ba0.6Sr0.4TiO3 can be successfully prepared by the high-energy ball milling process. (Synthesis conditions – see: Appendix A2)

Below it is shown an additional excellent example that, in the case of mechanochemical synthesis, should not be based on results from only one technique. The negative phenomenon of agglomeration, often occurring during milling, may affect the interpretation of the particle size and specific surface area of the material. Performed for all solid solutions, the analysis of the of particle size distribution by laser diffraction method showed that in each case about 10 microns is the dominant fraction. The specific surface area measurements (BET method) of each solid solutions clearly indicated that this area is growing with the increasing concentra‐

**Figure 7.** XRD patterns of mechanochemically synthesized BaTiO3 (BT) and Ba0.8Sr0.2TiO3 (BST-0.2), Ba0.6Sr0.4TiO3 (BST-0.4) [59]

tion of modifier – from 7.16 m2 /g for pure BaTiO3 to 34.99 m2 /g for Ba0,6Sr0,4TiO3 indicating on diminishing grains size. This was also confirmed by microscopic observation (SEM). The size of grains varies from 500 to approx. 100 nm.

## *Dielectric properties of Ba1–xSrxTiO3*

**5. Modification of perovskite-type electroceramics properties**

even in the case of strontium concentration equal to *x* = 0.4 in Ba1–xSrxTiO3.

**5.1. Formation of solid solution by doping various cations**

10 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

process. (Synthesis conditions – see: Appendix A2)

**Example 1. Ba1–***x***Sr***x***TiO3 (0.0 ≤** *x* **≤ 0.4)**

temperatures

Apart from the rapid and relatively simple synthesis of perovskite compounds, mechano‐ chemical method also allows to modify their chemical and physical properties. Such modifi‐ cations can be made by changing the chemical composition of perovskites, production of ceramics of an adequate morphology (grain size, defected, etc.), and also using as additives the contaminations originating from wear down of the grinding media (Examples 1–3).

**Figure 6.** XRD patterns of calcium titanate formation: (a) up to 4 h of milling, (b) subsequent calcination at different

XRD patterns of the synthesized compounds are shown in Figure 7. They indicate the influence of the presence of Sr2+ on the crystallinity of products. Any phase of strontium oxide was found

Different ionic radii of Ba2+ (1.36 Å) and Sr2+ (1.16 Å) induce distortion of lattice. A close look at a slow scan of reflection (e.g., from 31.0° to 32.5°) shown as an inset which indicates that there is a slight shift of this reflection to higher 2θ angles. This confirms the substitution of Sr2+ ions in the BaTiO3 lattice. It might be concluded that single phases of BaTiO3, Ba0.8Sr0.2TiO3, and Ba0.6Sr0.4TiO3 can be successfully prepared by the high-energy ball milling

Below it is shown an additional excellent example that, in the case of mechanochemical synthesis, should not be based on results from only one technique. The negative phenomenon of agglomeration, often occurring during milling, may affect the interpretation of the particle size and specific surface area of the material. Performed for all solid solutions, the analysis of the of particle size distribution by laser diffraction method showed that in each case about 10 microns is the dominant fraction. The specific surface area measurements (BET method) of each solid solutions clearly indicated that this area is growing with the increasing concentra‐

The dielectric properties of BaTiO3 (BT), Ba0.8Sr0.2TiO3 (BST-0.2), and Ba0.6Sr0.4TiO3 (BST-0.4) ceramics are described by the temperature dependence of the real (ε') and imaginary (ε") parts of electric permittivity. These properties were determined at selected frequencies of the electric field (1 MHz).

For the ε'/T dependence (Figure 8a), with increase in the strontium concentration, the *Curie* temperature *T*<sup>C</sup> gradually shifts toward lower value, and the peak of this transition becomes broader.

For BaTiO3 (BT) sample a classic paraelectric–ferroelectric (PE–FE) phase transition at 368 K occurs simultaneously with the change from a cubic to tetragonal structure. At lower temper‐ ature, the peak has diffused character which can be explained by the presence of small amount of impurities (ZrO2) from the reaction vial and grinding media. For the samples of solid solution BST-0.2 and BST-0.4, the ε'(T) plots show a diffusion nature of PE–FE phase transitions. The value of ε' maximum for BST-0.2 ceramics is about three times smaller and for BST-0.4

**Figure 8.** The temperature dependence of real part of dielectric permittivity (ε') a) and imaginary part of dielectric per‐ mittivity (ε") b) for BaTiO3 (BT), Ba0.8Sr0.2TiO3 (BST-0.2), and Ba0.6Sr0.4TiO3 (BST-0.4) samples

four times than in the case of BT sample. The phase transitions for Ba0.8Sr0.2TiO3 (BST-0.2) and Ba0.6Sr0.4TiO3 (BST-0.4) samples occur at temperature 343 and 288 K, respectively.

The energy loss of the electric field represented by imaginary part of electrical permittivity (ε") is tied to a structural phase change (Figure 8b). The temperature of maximum ε"(T) correlates with the temperature of maximum ε'(T).

## **Example 2. (Ba1–***x***Na***x***)(Ti1–***x***Nb***x***)O3** *(0.0 ≤ x ≤ 0.15)*

Figure 9 shows a comparison of X-ray powder diffraction patterns of BaTiO3 ceramics and (Ba1–*x*Na*x*)(Ti1–*x*Nb*x*)O3 for: *x* = 0.01; *x* = 0.04; *x* = 0.15 obtained by mechanochemical method. Visible shifts of the main diffraction reflections confirm the formation of appropriate solid solutions. In addition, these materials are characterized by a uniform grain size, approx. 500 nm. All are characterized by clearly defined grain boundaries and the lack of sinters. The example of morphology images of these materials in comparison with ceramics synthesized by conventional high-temperature method is presented in Figure 10.

Preparation of *(Ba*1–x*Na*x*)(Ti*1-x*Nb*x*)O*3 solid solution by conventional high-temperature solidphase synthesis requires a long-term heating of the mixture starting materials at high temper‐ ature. Using mechanochemical method significantly reduces the synthesis time. The monophase product was obtained after 1.5 h of high-energy milling and as in the previous example without the need for subsequent calcination.

## *Dielectric properties of (Ba1–xNax)(Ti1–xNbx)O3*

Comparing the dielectric properties of the same materials produced by two methods (Figure 11), i.e., mechanochemical and high-temperature syntheses, can draw the following conclu‐ sions:

**•** BaTiO3 and the ceramic solid solution of BNTN*x* for the composition of *x* = 0.01 are classical ferroelectrics with a sharp PE–FE phase transition.

four times than in the case of BT sample. The phase transitions for Ba0.8Sr0.2TiO3 (BST-0.2) and

**Figure 8.** The temperature dependence of real part of dielectric permittivity (ε') a) and imaginary part of dielectric per‐

The energy loss of the electric field represented by imaginary part of electrical permittivity (ε") is tied to a structural phase change (Figure 8b). The temperature of maximum ε"(T) correlates

Figure 9 shows a comparison of X-ray powder diffraction patterns of BaTiO3 ceramics and (Ba1–*x*Na*x*)(Ti1–*x*Nb*x*)O3 for: *x* = 0.01; *x* = 0.04; *x* = 0.15 obtained by mechanochemical method. Visible shifts of the main diffraction reflections confirm the formation of appropriate solid solutions. In addition, these materials are characterized by a uniform grain size, approx. 500 nm. All are characterized by clearly defined grain boundaries and the lack of sinters. The example of morphology images of these materials in comparison with ceramics synthesized

Preparation of *(Ba*1–x*Na*x*)(Ti*1-x*Nb*x*)O*3 solid solution by conventional high-temperature solidphase synthesis requires a long-term heating of the mixture starting materials at high temper‐ ature. Using mechanochemical method significantly reduces the synthesis time. The monophase product was obtained after 1.5 h of high-energy milling and as in the previous

Comparing the dielectric properties of the same materials produced by two methods (Figure 11), i.e., mechanochemical and high-temperature syntheses, can draw the following conclu‐

**•** BaTiO3 and the ceramic solid solution of BNTN*x* for the composition of *x* = 0.01 are classical

Ba0.6Sr0.4TiO3 (BST-0.4) samples occur at temperature 343 and 288 K, respectively.

mittivity (ε") b) for BaTiO3 (BT), Ba0.8Sr0.2TiO3 (BST-0.2), and Ba0.6Sr0.4TiO3 (BST-0.4) samples

12 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

by conventional high-temperature method is presented in Figure 10.

with the temperature of maximum ε'(T).

**Example 2. (Ba1–***x***Na***x***)(Ti1–***x***Nb***x***)O3** *(0.0 ≤ x ≤ 0.15)*

example without the need for subsequent calcination.

ferroelectrics with a sharp PE–FE phase transition.

*Dielectric properties of (Ba1–xNax)(Ti1–xNbx)O3*

sions:

**Figure 9.** XRD patterns of mechanochemically synthesized (Ba1–xNax)(Ti1–xNbx)O3 for: *x* = 0.01; *x* = 0.04; *x* = 0.15 solid solutions

**Figure 10.** The SEM micrographs of microstructure of BNTN*x* (for *x* = 0.04) sample surface, (a) mechanochemical, (b) high temperature

**•** The increase in the value of *x* in the BNTN*x* samples causes a diffuseness of the phase transition. Such behavior is a result of the different valency of substituted ions in both cationic sublattices.


The quick and simple synthesis without any thermal operation and better functional properties of products show the advantages of mechanochemistry. More information can be found in the works [57, 58, 61].

**Figure 11.** The temperature dependence of the real part of the complex dielectric permittivity for (a) BT, (b) BNTN*x*; *x* = 0.01, (c) *x* = 0.04, (d) *x* = 0.15 samples obtained by conventional (solid line) and mechanochemical (open symbol) meth‐ ods [61]

## **Example 3. Influence of impurities from the milling process on the properties of BaTiO3 (A) and CaCu3Ti4O12 (B)**

As mentioned in the chapter introduction, negative phenomenon of all processes associated with the high-energy milling is the attrition of some construction elements such as grinding media and/or inner coating of vial. This is a significant problem to technologists because such impurities are difficult to remove from milling products.

However, this effect can be used to modify the properties of materials. By selecting a suitable material of grinding media and grinding vial, it is possible to modify for example catalytic or electrical properties of ceramics.

Possibilities of the use of such impurities from the milling process are illustrated on perovskite compounds with high technological importance, i.e., BaTiO3 and CaCu3Ti4O12. The results are presented for comparison with analogous materials produced by high-temperature solid phase synthesis.

**a.** *BaTiO3*

**•** For composition of *x* = 0.15, a strong dispersion of the dielectric permittivity maximum is observed. The obtained results clearly indicate that BNTN*x* (*x* = 0.15) sample is relaxor

**•** Materials produced by mechanochemical synthesis method is characterized by better parameters, e.g., higher value of dielectric permittivity, less diffused character of para–

The quick and simple synthesis without any thermal operation and better functional properties of products show the advantages of mechanochemistry. More information can be found in the

**Figure 11.** The temperature dependence of the real part of the complex dielectric permittivity for (a) BT, (b) BNTN*x*; *x* = 0.01, (c) *x* = 0.04, (d) *x* = 0.15 samples obtained by conventional (solid line) and mechanochemical (open symbol) meth‐

**Example 3. Influence of impurities from the milling process on the properties of BaTiO3 (A)**

As mentioned in the chapter introduction, negative phenomenon of all processes associated with the high-energy milling is the attrition of some construction elements such as grinding

ferroelectric.

works [57, 58, 61].

ods [61]

**and CaCu3Ti4O12 (B)**

ferroelectric phase transition.

14 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Barium titanate was prepared mechanochemically using grinding media and grinding vial made of steel and zirconium oxide. Conditions of synthesis – see: appendix A3.

In Figure 12 is presented room temperature X-ray powder diffraction for mechanochemically and high-temperature produced BaTiO3. There are considerable differences in the intensities and in half-width of the diffraction reflexes of ceramics and indicating on the various mor‐ phologies of the grains. This is confirmed by SEM microscopic observation (Figure 13). The samples after mechanical treatment have much smaller particles (about 500 nm) in comparison to conventionally produced BaTiO3, which is characterized by large, uneven in size and shape of the grains.

Materials prepared by these two methods also differ in terms of size BET surface area. The mechanochemically obtained powders were characterized by surface area of approx. 7 m2 /g, BaTiO3 prepared by high-temperature synthesis due to the presence, among others, of sinters have a much smaller specific surface area of 0.25 m2 /g.

BaTiO3 synthesized using the high-temperature method had tetragonal symmetry at room temperature, in contrast to that obtained mechanochemically, which was cubic. The high degree of communition and the large number of defects caused by high-energy ball milling limits long-range order in the crystallographic structure, and this prevents phase transitions. Calcination of the powder for 1 h at 1,373 K eliminates this problem [21].

The amount of impurities – in both cases the iron or zirconium oxide – in samples prepared mechanochemically was approx. 1% wt. Because the barium titanate is a model ferroelectrics, below is illustrated the effect of the presence of these impurities on these properties. Figure 14 presents the temperature dependence of the dielectric permittivity for all ceramics. Samples are marked on the system as: BaTiO3/T – high-temperature synthesis, BaTiO3/Zr and Ba‐ TiO3/Fe – mechanochemical synthesis.

In the case of BaTiO3/T, the temperature at which ε' is the highest represents the paraelectric– ferroelectric (PE–FE) phase transition. At all frequencies of the electric field, a classical, sharp transition can be seen at 403 K, which corresponds to a structural shift between cubic and tetragonal phases. At 288 K there is another maximum, however its value is around fourtimes smaller than the transition at 403 K. The observed temperature of this phase transition for

**Figure 12.** Comparison of X-ray powder patterns of BaTiO3 obtained by two different synthesis methods [21]

**Figure 13.** SEM photomicrographs of BaTiO3 powders obtained by (a) mechanochemical and (b) high-temperature sol‐ id-state syntheses [21]

BaTiO3 varies from the literature by 5–20 K [8, 16, 17, 20]. It is possible to surmise that transition at 288 K which corresponds to a change from a tetragonal structure to an orthorhombic one.

For BaTiO3/Zr ceramics, the PE–FE transition occurs at 368 K. This transition is somewhat diffused, and the frequency of the electrical field a little bit changes the behavior of the material.

**Figure 14.** The temperature dependence of real part of dielectric permittivity (ε') for BaTiO3/T, BaTiO3/Fe, and Ba‐ TiO3/Zr samples [21]

A further lowering of the temperature causes greater peak diffusing compared to that observed in "pure" BaTiO3. There is no maximum observed at 288 K.

The BaTiO3/Fe product has a characteristically diffused peak at the PE–FE transition, the most diffused of any of the samples tested. The observed maximum of ε' approaches a value of 3,000 at a temperature of 303 K, nearly 100 K lower than the same transition in BaTiO3/T. Measure‐ ments at different frequencies of electrical field show that this material does not have the properties of a relaxor.

It is possible to postulate that the diffused character of the phase transition in BaTiO3/Zr and BaTiO3/Fe ceramics is caused by the presence of the impurities from the milling process.

## **b.** *CaCu3Ti4O12*

BaTiO3 varies from the literature by 5–20 K [8, 16, 17, 20]. It is possible to surmise that transition at 288 K which corresponds to a change from a tetragonal structure to an orthorhombic one.

**Figure 13.** SEM photomicrographs of BaTiO3 powders obtained by (a) mechanochemical and (b) high-temperature sol‐

id-state syntheses [21]

**Figure 12.** Comparison of X-ray powder patterns of BaTiO3 obtained by two different synthesis methods [21]

16 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

For BaTiO3/Zr ceramics, the PE–FE transition occurs at 368 K. This transition is somewhat diffused, and the frequency of the electrical field a little bit changes the behavior of the material. Figure 15 shows the XRD patterns for the mechanochemically synthesized CaCu3Ti4O12 (CCTO/Fe and CCTO/Zr, reactor and balls of steel, and ZrO2, respectively) and by hightemperature treatment (CCTO/T). As can be seen, all samples are monophase – CCTO. Syntheses details are given in Appendix A3.

Significant differences in the intensities and widths of diffraction reflexes of materials produced by high-temperature and mechanochemical route indicate different degrees of crystallinity and grain morphology. These observations were confirmed by SEM studies (Figure 16a, b).

High-temperature synthesis method results in grain growth (1.5–2.5 μm) and sintering of grains giving uneven distribution of particle size (Figure 16a). In addition, in material at the grain boundaries are visible places with high concentration of copper caused by the presence of sinters and a change in the oxidation state of copper ions during calcination at high temperature. This phenomenon is well known and this is the main problem during production of this material by methods where high-temperature processing is used. Such a ceramic is chemically inhomogeneous in volume of the particles and at their boundaries. CaCu3Ti4O12 prepared mechanochemically is characterized by a uniform size distribution in the range of

**Figure 15.** XRD patterns of mechanochemically synthesized CaCu3Ti4O12 (CCTO/Fe and CCTO/Zr, reactor and balls of steel and ZrO2, respectively) and by high-temperature treatment (CCTO/T) [60]

**Figure 16.** SEM images of CaCu3Ti4O12 : (a) high-temperature treatment (CCTO/T), (b) mechanochemical synthesis (CCTO/Zr) [60]

100–500nm and the lack of the high-temperature processing during synthesis eliminates the problem of chemical inhomogeneity of the material (Figure 16b). Grain boundary problem in ceramics CaCu3Ti4O12 and their influence on its dielectric properties are widely discussed in the papers [62–65]. Dielectric properties of studied perovskite-related CaCu3Ti4O12 compound synthesized under different conditions as a temperature relationship of the real component of dielectric permittivity (ε') and dielectric loss (tan δ) at the field frequency of 1 kHz are presented in Figure 17a and b, respectively.

**Figure 17.** Temperature relationship of the real component of (a) dielectric permittivity (ε') and (b) dielectric loss (tan δ) at the field frequency of 1 kHz for CaCu3Ti4O12 synthesized under different conditions [60]

The presence of small amount of zirconia (CCTO/Zr) causes the smallest changes in the electric permittivity (ε') versus temperature, similar to CCTO/T nature. From the practical point of view, such effect as well as very low and stable value of dielectric loss (tan δ) in the temperature range of –50–50o C is very convenient. Material behaves differently with the presence of metallic iron (CCTO/Fe). From the ambient temperature, a significant increase in the value of ε' is observed. At 200°C, ε' reaches a value of about 3,000. However, high values of dielectric loss (tan δ), indicating the conversion of electrical energy into heat, disqualify this ceramics for practical applications, e.g., as capacitor material.

## **6. Summary**

100–500nm and the lack of the high-temperature processing during synthesis eliminates the problem of chemical inhomogeneity of the material (Figure 16b). Grain boundary problem in ceramics CaCu3Ti4O12 and their influence on its dielectric properties are widely discussed in the papers [62–65]. Dielectric properties of studied perovskite-related CaCu3Ti4O12 compound

**Figure 16.** SEM images of CaCu3Ti4O12 : (a) high-temperature treatment (CCTO/T), (b) mechanochemical synthesis

**Figure 15.** XRD patterns of mechanochemically synthesized CaCu3Ti4O12 (CCTO/Fe and CCTO/Zr, reactor and balls of

steel and ZrO2, respectively) and by high-temperature treatment (CCTO/T) [60]

18 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

(CCTO/Zr) [60]

Mechanochemistry being one of the easiest and cheapest methods of producing nanomaterials also enables the synthesis of advanced ceramics with perovskite structure. It is an interesting alternative to other methods to produce such compounds. As is shown in the presented examples, the reactions between the substrates in the form of oxide are going exclusively by high-energy milling without requiring long-term calcination at high temperature. This helps to avoid many technological difficulties and problems related to powder morphology and properties. Using mechanochemical synthesis of CaCu3Ti4O12 eliminates problems associated with inhomogeneity of grains and intergranular boundaries. Consequently, the product prepared mechanochemically has much better functional properties than the same obtained by conventional high-temperature solid phase synthesis. The same is the case of the popular electroceramics PZT (PbZr-TiO3), due to the high volatility of lead at high temperature. Using other synthesis techniques, it is difficult to maintain the stoichiometry of the compound. Applying for this purpose, the mechanochemical treatment not only omitted this problem but it is also able to produce nanometric ceramics.

In spite of the synthesis of perovskite compounds, mechanochemical technique can be applied also to modification of their properties. Selection of appropriate conditions for high-energy ball milling process is very important.

Thus, mechanochemistry realized by high-energy ball milling becomes recently the more environmentally acceptable for perovskite processing mainly for the following reasons:


## **7. Appendix**

## **A1.**

Mechanochemical treatment was provided using two different high-energy laboratory planetary mills with vial (250 ml) and balls (10 mm diameter) made of Ni-Cr steel:


The subsequent calcination processes of powders were carried out in Nabertherm HTC 03/15 laboratory furnace in air atmosphere.

## **A2.**

Mechanochemical treatment was provided using Fritsch *Pulverisette-6* planetary ball mill with vial (250 ml) and balls (10 mm diameter) made of ZrO2. Milling parameters: rpm = 500; ball to powder mass ratio: BPR = 20:1; milling time: 1.5 h; atmosphere: air.

Preparation of reference samples by high-temperature solid phase reaction. The reactants powders were ground together in a stoichiometric ratio in an agate mortar, pressed into pellet discs 4 mm thick and of 12 mm in diameter and sintered in air with the use of a Nabertherm HTC 03/15 laboratory furnace for 12 h at the temperature of 1,350o C.

## **A3.**

Mechanochemical treatment was provided using Fritsch *Pulverisette-6* planetary ball mill with vial (250 ml) and balls (10 mm diameter) made of ZrO2 and steel. Milling parameters: rpm = 500; ball to powder mass ratio: BPR = 20:1; milling time: 1.5 h; atmosphere: air.

Preparation of reference samples by high-temperature solid phase reaction – as in the A2.

The phase identification and physicochemical characteristics of milling products were determined using the following methods:


## **Acknowledgements**

other synthesis techniques, it is difficult to maintain the stoichiometry of the compound. Applying for this purpose, the mechanochemical treatment not only omitted this problem but

In spite of the synthesis of perovskite compounds, mechanochemical technique can be applied also to modification of their properties. Selection of appropriate conditions for high-energy

Thus, mechanochemistry realized by high-energy ball milling becomes recently the more environmentally acceptable for perovskite processing mainly for the following reasons:

Mechanochemical treatment was provided using two different high-energy laboratory

The subsequent calcination processes of powders were carried out in Nabertherm HTC 03/15

Mechanochemical treatment was provided using Fritsch *Pulverisette-6* planetary ball mill with vial (250 ml) and balls (10 mm diameter) made of ZrO2. Milling parameters: rpm = 500; ball to

Preparation of reference samples by high-temperature solid phase reaction. The reactants powders were ground together in a stoichiometric ratio in an agate mortar, pressed into pellet discs 4 mm thick and of 12 mm in diameter and sintered in air with the use of a Nabertherm

Mechanochemical treatment was provided using Fritsch *Pulverisette-6* planetary ball mill with vial (250 ml) and balls (10 mm diameter) made of ZrO2 and steel. Milling parameters: rpm =

Preparation of reference samples by high-temperature solid phase reaction – as in the A2.

500; ball to powder mass ratio: BPR = 20:1; milling time: 1.5 h; atmosphere: air.

C.

planetary mills with vial (250 ml) and balls (10 mm diameter) made of Ni-Cr steel:

**2.** *Activator-2S* (by Activator Corp. Novosibirsk) a rotation speed 1,000 rpm.

it is also able to produce nanometric ceramics.

20 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

ball milling process is very important.

**•** Give the micro/nanosized products.

laboratory furnace in air atmosphere.

**7. Appendix**

**A1.**

**A2.**

**A3.**

**•** Simplify the synthesis process to one step

**•** Reduce the cost of chemicals and/or heat used in a traditional way

**1.** *Pulverisette-6* (by Fritsch GmbH) with a rotation speed 500 rpm.

powder mass ratio: BPR = 20:1; milling time: 1.5 h; atmosphere: air.

HTC 03/15 laboratory furnace for 12 h at the temperature of 1,350o

The author would like to thank Professor Krystyna Wieczorek-Ciurowa of the Cracow University of Technology, Poland, for introduction into mechanochemical science and many hours of valuable discussion helpful during the preparation of this manuscript. For scientific collaboration, thanks also to Professor Czesław Kajtoch and Dr Wojciech Bąk of the Pedagog‐ ical University of Cracow.

This study was supported by the National Science Centre Poland, Project DEC-2012/05/N/ ST8/03764.

## **Author details**

Piotr Dulian\*

Address all correspondence to: piotrdulian@chemia.pk.edu.pl

Faculty of Chemical Engineering and Technology, Cracow University of Technology, Poland

## **References**


[16] Bocquet JF, Chhor K, Pommier C. Barium titanate powders synthesis from solvother‐ mal reaction and supercritical treatment. Mater Chem Phys 1999;57(3):272–80. DOI: 10.1016/S0254-0584(98)00233-8

**References**

10.1016/S1359-6454(99)00293-1

22 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

10.1126/science.1129564

80. DOI: 10.1039/c3nr02543h

10.1016/0375-9601(96)00077-1

8. DOI: 10.1038/358136a0

ma7096502

1995.1283

1999;82(4):797–818.

10.1007/s10832-007-9008-7

[1] Setter N, Wasser R. Electroceramic materials. Acta Mater 2000;48(1):151–78. DOI:

[2] Waser R. Modeling of electroceramics – applications and prospects. J Eur Ceramic

[3] Scott JF. Applications of modern ferroelectrics. Science 2007;315(5814):954–9. DOI:

[4] Eerestein W, Mathur ND, Scott JF. Multiferroic and magnetoelectric materials. Na‐

[5] Nuraje N, Su K. Perovskite ferroelectric nanomaterials. Nanoscale 2013;5(19):8752–

[6] Cohen RE. Origin of ferroelectricity in perovskite oxides. Nature 1992;358(6382):136–

[7] Li S, Eastman JA, Li Z, Foster CM, Newnham RE, Cross LE. Size effects in nanostruc‐ tured ferroelectrics. Phys Lett A: Gen Atom Solid State Phys 1996;212(6):341–6. DOI:

[8] Kretschmer R, Binder K. Surface effects on phase transitions in ferroelectrics and di‐ polar magnets. Phys Rev B 1979;20(3):1065–76. DOI: 10.1103/PhysRevB.20.1065

[9] Cottam MG, Tilley DR, Zeks B. Theory of surface modes in ferroelectrics. J Phys C:

[10] Binder K. Surface effects on phase transitions in ferroelectrics and antiferroelectrics.

[11] Liu J, Chen W, Wang B, Zheng Y. Theoretical methods of domain structures in ultra‐ thin ferroelectric films: a review. Materials 2014;6(9):6502–68. DOI: 10.3390/

[12] Haertling GH. Ferroelectric ceramics: history and technology. J Am Ceramic Soc

[13] Somani V, Kalita SJ. Synthesis and characterization of nanocrystalline barium stronti‐ um titanate powder via sol-gel processing. J Electroceramics 2007;18(1):57–65. DOI:

[14] Senna M. A straight way toward phase pure complex oxides. J Eur Ceramic Soc

[15] Barboux P, Griesmar P, Ribot F, Mazerolles L. Homogeneity-related problems in sol‐ ution derived powders. J Solid State Chem 1995;117(2):343–50. DOI: 10.1006/jssc.

Solid State Phys 1984;17(10):1793–823. DOI: 10.1088/0022-3719/17/10/022

Ferroelectrics 1981;35(1):99–104. DOI: 10.1080/00150198108017670

2005;25(12):1977–84. DOI: 10.1016/j.jeurceramsoc.2005.03.002

Soc 1999;19(6):655–64. DOI: 10.1016/S0955-2219(98)00293-3

ture 2006;442(7104):759–65. DOI: 10.1038/nature05023


[43] Koch CC. The synthesis and structure of nanocrystalline materials produced by me‐ chanical attrition: a review. Nanostructured Mater 1993;2(2):109–29. DOI: 10.1016/0965-9773(93)90016-5

[29] Venkataraman KS, Narayanan KS. Energetics of collision between grinding media in ball mills and mechanochemical effects. Powder Technol 1998;96(3):190–201. DOI:

[30] Menyhart M, Miskiewicz L. Communition and structural changes in a jet mill. Pow‐

[31] Baláž P, Achimovičová M, Baláž M, Billik P, Cherkezova-Zheleva Z, Criado JM, De‐ logu F, Dutková E, Gaffet E, Gotor FJ, Kumar R, Mitov I, Rojac T, Senna M, Streletskii A, Wieczorek-Ciurowa K. Hallmarks of mechanochemistry: from nanoparticles to

[32] Erdemolu M, Baláž P. An overview of surface analysis techniques for characteriza‐ tion of mechanically activated minerals. Mineral Process Extract Metallurgy Rev

[33] Beyer MK, Clausen-Schaumann H. Mechanochemistry: The mechanical activation of

[34] James SL, Adams CJ, Bolm C, Braga D, Collier P, Friić T, Grepioni F, Harris KDM, Hyett G, Jones W, Krebs A, MacK J, Maini L, Orpen AG, Parkin IP, Shearouse WC, Steed JW, Waddell DC. Mechanochemistry: opportunities for new and cleaner syn‐

[35] Boldyreva E. Mechanochemistry of inorganic and organic systems: what is similar, what is different? Chem Soc Rev 2013;42(18):7719–38. DOI: 10.1039/c3cs60052a

[36] Suryanarayana C. Mechanical alloying and milling. Prog Mater Sci 2001;46(1):1–184.

[37] Ivanov E, Suryanarayana C. Materials and process design through mechanochemical routes. J Mater Syn Process 2000;8(3):235–44. DOI: 10.1023/A:1011372312172

[38] Sopicka-Lizer M, editor. High-energy ball milling: mechanochemical processing of nanopowders. 1st Editionst ed. Woodhead Publishing Limited; 2010. 440 p. DOI:

[39] Rumpf H. Physical aspects of communition and new formulation of a law of commu‐ nition. Powder Technol 1973;7(3):145–59. DOI: 10.1016/0032-5910(73)80021-X

[40] Wieczorek-Ciurowa K, Dulian P, Nosal A, Domagała J. Effects of reagents' nature on mechanochemical synthesis of calcium titanate. J Thermal Analysis Calorimetry

[41] Boldyrev VV. Mechanochemistry and mechanical activation of solids. Solid State Ion‐

[42] Hainovsky NG, Pavlyukhin YuT, Boldyrev VV. Application of mechanical activation to the synthesis of high temperature superconductors. Mater Sci Eng B 1991;8(4):283–

technology. Chem Soc Rev 2013;42(18):7571–637. DOI: 10.1039/c3cs35468g

covalent bonds. Chem Rev 2005;105(8):2921–48. DOI: 10.1021/cr030697h

thesis. Chem Soc Rev 2012;41(1):413–47. DOI: 10.1039/c1cs15171a

der Technol 1976;15(2):261–6. DOI: 10.1016/0032-5910(76)80055-1

2012;33(1):65–88. DOI: 10.1080/08827508.2010.542582

DOI: 10.1016/S0079-6425(99)00010-9

2010;101(2):471–7. DOI: 10.1007/s10973-010-0802-0

6. DOI: 10.1016/0921-5107(91)90048-Z

ics 1993;63–65(C):537–43. DOI: 10.1016/0167-2738(93)90157-X

10.1533/9781845699444

10.1016/S0032-5910(97)03368-8

24 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications


## **Synthesis of Perovskite Oxides by Hydrothermal Processing – From Thermodynamic Modelling to Practical Processing Approaches**

Juan Carlos Rendón-Angeles, Zully Matamoros-Veloza, Karla Lorena Montoya-Cisneros, Jorge López Cuevas and Kazumichi Yanagisawa

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61568

## **Abstract**

[56] Avvakumov EG, Karakchiev LG, Gusev AA, Vinokurova OB. Preparation of disperse ceramic materials by soft mechanochemical synthesis. Mater Sci Forum 2002;386:245–

[57] Bak, W, Dulian P, Sitko D, Garbarz-Glos B, Kajtoch C, Wieczorek-Ciurowa K, Smel‐ tere I. Effect of variable valence ion doping on the dielectric properties of BaTiO3 based materials. Ferroelectrics 2014;464(1):35–41. DOI: 10.1080/00150193.2014.892810

[58] Dulian P, Bak W, Wieczorek-Ciurowa K, Kajtoch C. Dielectric properties of vanadi‐ um doped barium titanate synthesized via high-energy ball milling. Mater Sci Po‐

[59] Dulian P, Bak WB, Wieczoek-Ciurowa K, Kajtoch C. Dielectric behaviour of BaTiO3- SrTiO3 solid solutions fabricated by high-energy ball milling. Key Eng Mater

[60] Dulian P, Bak W, Wieczorek-Ciurowa K, Kajtoch C. Comparative studies of dielectric properties of Ca0.25Cu 0.75TiO3 ceramics produced by mechanochemical synthesis and in the way of high-temperature treatment. Key Eng Mater 2013;543:326–9. DOI:

[61] Bak W, Dulian P, Grabarz-Glos B, Kajtoch Cz, Wieczorek-Ciurowa K. Dielectric be‐ haviour of (Ba1-xNax)(Ti1-xNbx)O3 ceramics obtained by conventional and mecha‐

[62] Dulian P, Bak W, Kajtoch Cz, Wieczorek-Ciurowa K. A high-energy ball milling as a useful technique for the synthesis of CaCu3Ti4O12 electroceramics. Acta Phys Polon

[63] Lunkenheimer P, Fichtl R, Ebbinghaus SG, Loidl A. Nonintrinsic origin of the colos‐ sal dielectric constants in CaCu3Ti4O12. Phys Rev B Condens Matter Mater Phys

[64] Adams TB, Sinclair DC, West AR. Giant barrier layer capacitance effects in Ca‐ Cu3Ti4O12 ceramics. Adv Mater 2002;14(18):1321–3. DOI:

[65] Bender BA, Pan M-J. The effect of processing on the giant dielectric properties of Ca‐ Cu3Ti4O12. Mater Sci Eng B: Solid-State Mater Adv Technol 2005;117(3):339–47. DOI:

10.1002/1521-4095(20020916)14:18<1321::AID-ADMA1321>3.0.CO;2-P

land 2014;32(2):257–63. DOI: 10.2478/S13536-013-0185-6

10.4028/www.scientific.net/KEM.543.326

26 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

nochemical synthesis. Ferroelectrics. Forthcoming.

2004;70(17):1–4. DOI: 10.1103/PhysRevB.70.172102

10.1016/j.mseb.2004.11.019

A 2014;126(4):931–7. DOI: 10.12693/APhysPolA.126.931

2014;605:63–66. DOI: 10.4028/www.scientific.net/KEM.605.63

50.

The present chapter aims to provide a conscious review of the principles associated with the chemical reactions and the control of the parameters related to the hydrothermal processing of perovskite-structured compounds. Highlights on fundamental principles of the thermodynamic modelling coupled with the relevant technical expertise gained during the past two decades are discussed. Achievements conducted in the early 1990s on thermodynamic modelling of hydro‐ thermal reactions, leading to the estimation of the chemical reaction equilibrium occurring under specific conditions, i.e. above 100°C and 0.1 MPa, are discussed. Additional efforts resulted in dif‐ ferent thermodynamic models that predict crystal growth kinetics and the stability for particle nu‐ cleation; the models based on chemical population balance approaches are also considered. However, these models do not apply for perovskite compounds containing rare earth elements that crystallize under hydrothermal conditions above 250°C, i.e. orthorhombic lanthanum chro‐ mite perovskite. Hence, the final part comprises a literature survey for the experimental research work conducted on various perovskite species produced via hydrothermal treatments, emphasiz‐ ing the relevant conditions that led to the stoichiometric single-phase crystallization.

**Keywords:** Perovskite materials, Hydrothermal synthesis, Crystallization, Thermody‐ namic modelling, Solid-Liquid Equilibrium

## **1. Introduction**

Over the past two decades, remarkable efforts have been conducted worldwide to explore various techniques to prepare perovskite-structured ceramic oxide materials. Perovskite

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

oxides exhibit a broad variety of functional properties, namely ferroelectricity, piezoelectricity, pyroelectricity and non-linear dielectric behaviour, among others. The functionality of perovskite-structured compounds depends on the relationship established between the specific crystalline structure and the composition of its major constituents. This particular family of inorganic compounds has a large range of interesting physical properties applicable to the design and preparation of different electronic devices with applications in charge storage, non-volatile memories, transducers, actuators and infrared detection. Particular efforts focused on optimizing the physical properties of perovskite-structured compounds have recently taken place by numerous research groups worldwide. Recently, most studies are based on establishing a correlation between the crystalline structure and the chemical stoichiometry of the major constituents. These have led to improvements in the functional properties of the ideal ABX3 compound with cubic structure (space group).

Another relevant subject that attracted the interest of various research groups concerns with the development of environmentally friendly chemical processes for producing perovskite compounds at laboratory and large scales. The hydrothermal technique has become one of the most suitable chemical processing routes in terms of energy consumption and environmental friendliness for the preparation of a vast number of perovskite compounds with specific crystalline structures and chemical compositions. In the present review, the state-of–the-art associated with this technique is discussed in terms of processing aspects involved in the crystallization of perovskite compounds. In particular for the case of alkaline earth titanate perovskite oxides, these compounds had been used as a standard for developing thermody‐ namic modelling approaches. These models proposed consider the particular chemical reaction equilibriums that occur under hydrothermal conditions and lead the control of perovskite particle crystallization. Various stability crystallization diagrams were calculated by simulating a complex reaction system containing solid and ionic species in the aqueous phase employing the thermodynamic models. Likewise, a comprehensive analysis is based on the expertise gained during the past two decades by several research groups. With regards to the correlation of the fundamentals principles and the technical aspects involved in the hydrothermal synthesis, a vast number of perovskite compounds are addressed. The present review intends to give guidance for the researchers, as well as to encourage the newcomers from research fields such as Material Science, Chemistry and Physics, by directing their attention towards the key points for conducting hydrothermal reactions for preparing perovskite compounds.

## **2. Hydrothermal synthesis of perovskite materials**

### **2.1. Structure of perovskite materials, applications and synthesis processes**

### *2.1.1. The crystalline structure of mineral perovskite*

The perovskite oxide family has been studied extensively because of the diverse properties exhibited by a material when a considerable number of different atoms produce the typical

atomic arrangement associated with this crystalline structure. Gustav Rose discovered the CaTiO3 mineral in 1839 at the Ural Mountains in Russia, and the mineral was named by the Russian mineralogist Lev Aleksevich Perovski [1]. The "perovskite" term applied to a partic‐ ular group of inorganic compounds having similar crystalline structure, and the ABX3 chemical stoichiometry, the major constituent of the compound is CaTiO3. This compound is constituted by divalent (A2+) and tetravalent (B4+) cations. The small B4+ cation is corner-sharing bonded with six X2– anions, resulting in an octahedral BX6 unit forms the ideal face-centered cubic close packed unit cell. Likewise, the large A cation located at the 12-fold coordination sites produces eight octahedral points in the corners of the cube (Figure 1a). In contrast, the simplest atomic arrangement of the ideal cubic cell unit involves a different cation distribution. In this structure the A cations are in the corners of the cube, while the B cation located in the center of the cube is coordinated with six anions in the face-centred positions of the cubic unit cell (Figure 1b). This ideal cubic perovskite is not very commonly found in the mineral reservoirs, because even the mineral species has slight distortions in the atomic ordering in the cubic structure. The study of the perovskite crystalline structure firstly conducted by Goldsmith in 1920 led to propose various fundamental aspects that correlate the crystalline structure and chemical composition parameters [1,2]. One of the most important principles proposed was the tolerance factor. The tolerance factor is one tool normally used for predicting the structural arrangement and stability of a particular perovskite composition, either from the chemical and physical points of view. This factor is evaluated before selecting the adequate processing route for perovskite preparation. B atom is located at 0, 0, 0 (Figure 1b), and the X atoms in 3d are located at spatial positions ½, 0, 0; 0, ½, 0, and 0, 0, ½. The lattice parameter "a0" of this perovskite structure is 3.905 Å. This crystalline structure undergoes a series of distortions; one of this is caused by a deficiency in the Sr2+ ions (A) in the framework, producing a ReO3‐type structure [2]. Another distortion occurs when the ReO3 structure transforms into dense packing due to the octahedral unit rotation, producing a structural geometry resembling the hexagonal close packing type, i.e. RhF3. The void at the centre is in an octahedral surrounding coordination; when this octahedral hole is occupied, the ilmenite structure (FeTiO3) is obtained. The slight distortions attainable in the cubic structure are due to the displacement of ions from the ideal positions, producing a variation of a few tenths of an Å. Therefore, the final symmetry varies considerably between different materials. A mechanism that likely promotes these distortions is related to the capability of perovskite to accommodate a great variety of atoms, due to the flexibility of its crystalline structure [1–3]. An additional mechanism is the Jahn– Teller effect.

oxides exhibit a broad variety of functional properties, namely ferroelectricity, piezoelectricity, pyroelectricity and non-linear dielectric behaviour, among others. The functionality of perovskite-structured compounds depends on the relationship established between the specific crystalline structure and the composition of its major constituents. This particular family of inorganic compounds has a large range of interesting physical properties applicable to the design and preparation of different electronic devices with applications in charge storage, non-volatile memories, transducers, actuators and infrared detection. Particular efforts focused on optimizing the physical properties of perovskite-structured compounds have recently taken place by numerous research groups worldwide. Recently, most studies are based on establishing a correlation between the crystalline structure and the chemical stoichiometry of the major constituents. These have led to improvements in the functional

Another relevant subject that attracted the interest of various research groups concerns with the development of environmentally friendly chemical processes for producing perovskite compounds at laboratory and large scales. The hydrothermal technique has become one of the most suitable chemical processing routes in terms of energy consumption and environmental friendliness for the preparation of a vast number of perovskite compounds with specific crystalline structures and chemical compositions. In the present review, the state-of–the-art associated with this technique is discussed in terms of processing aspects involved in the crystallization of perovskite compounds. In particular for the case of alkaline earth titanate perovskite oxides, these compounds had been used as a standard for developing thermody‐ namic modelling approaches. These models proposed consider the particular chemical reaction equilibriums that occur under hydrothermal conditions and lead the control of perovskite particle crystallization. Various stability crystallization diagrams were calculated by simulating a complex reaction system containing solid and ionic species in the aqueous phase employing the thermodynamic models. Likewise, a comprehensive analysis is based on the expertise gained during the past two decades by several research groups. With regards to the correlation of the fundamentals principles and the technical aspects involved in the hydrothermal synthesis, a vast number of perovskite compounds are addressed. The present review intends to give guidance for the researchers, as well as to encourage the newcomers from research fields such as Material Science, Chemistry and Physics, by directing their attention towards the key points for conducting hydrothermal reactions for preparing

properties of the ideal ABX3 compound with cubic structure (space group).

28 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**2. Hydrothermal synthesis of perovskite materials**

*2.1.1. The crystalline structure of mineral perovskite*

**2.1. Structure of perovskite materials, applications and synthesis processes**

The perovskite oxide family has been studied extensively because of the diverse properties exhibited by a material when a considerable number of different atoms produce the typical

perovskite compounds.

Figure 1 Structural representation of the ideal cubic perovskite showing the (a) cubic A unit cell and (b) cubic B unit cell. **Figure 1.** Structural representation of the ideal cubic perovskite showing the (a) cubic A unit cell and (b) cubic B unit cell.

**2.1.2. Tolerance factor, orthorhombic and hexagonal crystalline unit cells** The ionic radii differences produce the structural distortions on perovskite structure, as was determined in the pioneering research work conducted by Goldschimidt [1–3]. The equation that geometrically correlates the ionic radii (*rA*, *rB* and *rO*) for the ideal cubic cell with the The compounds that belong to the idealized cubic (fcc) structure is SrTiO3 because the strontium titanate oxide exhibits the atomic packing shown in Figure 1b. The ideal cubic structure has a space group. In agreement with the Wyckoff positions, the atomic distribution is as follows: the A atoms are in x, y, z coordinates ½, ½, ½ (Figure 1b) while the B atom is located at 0, 0, 0 (Figure 1b), and the X atoms in 3d are located at spatial positions ½, 0, 0; 0, ½, 0, and 0, 0, ½. The lattice parameter "a0" of this perovskite structure is 3.905 Å. This crystalline

*a rr rr* 2 2 *<sup>A</sup> O AO* (1)

The expression that involves the unit cell length ratio is known as the Goldschmidt´s *tolerance factor t*, which is employed to estimate the distortion level attained by a particular perovskite‐structured compound. Fundamentally, it considers the ionic radii of the

lattice parameter *a0* is given as:

structure undergoes a series of distortions; one of this is caused by a deficiency in the Sr2+ ions (A) in the framework, producing a ReO3-type structure [2]. Another distortion occurs when the ReO3 structure transforms into dense packing due to the octahedral unit rotation, produc‐ ing a structural geometry resembling the hexagonal close packing type, i.e. RhF3. The void at the centre is in an octahedral surrounding coordination; when this octahedral hole is occupied, the ilmenite structure (FeTiO3) is obtained. The slight distortions attainable in the cubic structure are due to the displacement of ions from the ideal positions, producing a variation of a few tenths of an Å. Therefore, the final symmetry varies considerably between different materials. A mechanism that likely promotes these distortions is related to the capability of perovskite to accommodate a great variety of atoms, due to the flexibility of its crystalline structure [1–3]. An additional mechanism is the Jahn–Teller effect.

### *2.1.2. Tolerance factor, orthorhombic and hexagonal crystalline unit cells*

The ionic radii differences produce the structural distortions on perovskite structure, as was determined in the pioneering research work conducted by Goldschimidt [1–3]. The equation that geometrically correlates the ionic radii **(***rA*, *rB* and *rO*) for the ideal cubic cell with the lattice parameter *a0* is given as:

$$a = \sqrt{2\left(r\_A + r\_O\right)} = \mathcal{D}\left(r\_A + r\_O\right) \tag{1}$$

The expression that involves the unit cell length ratio is known as the Goldschmidt´s **tolerance factor t**, which is employed to estimate the distortion level attained by a particular perovskitestructured compound. Fundamentally, it considers the ionic radii of the constituent atoms that fit the chemical stoichiometry and enhances a pure ionic bonding between them; the mathe‐ matical expression for the tolerance factor is given in Eq. (2).

$$t = \frac{(r\_A + r\_O)}{\sqrt{2}(r\_A + r\_O)}\tag{2}$$

According to Eq. (2), the ideal perovskite cubic structure has a "t" value equal to 1, which can be calculated for SrTiO3 with ionic radii of *rA* = 1.44 Å, *rB* = 0.605 Å and *rO* = 1.40 Å. The value of *t* is less than 1 when the A ionic radius is slightly small, and structurally the octahedral unit [BO6] tilts forward filling the additional space. From Eq. (2), the grade of tolerance where the ideal cubic structure is attainable in perovskite compounds is 0.89 < *t* < 1 [2–4].

Orthorhombic ABO3 perovskites are among the most important constituents of the Earth´s crust. These compounds have been under exhaustive study due to the wide variety of their functional properties. The distortion of the cubic cell results in the formation of the ortho‐ rhombic structure. This process occurs by the tilting of the BO6 octahedra, but this distortion is not detectable as the temperature increases because the tilt angle decreases [5]. A typical orthorhombic structure that doubles in dimensions the cubic one is shown in Figure 2. Orthorhombic RMO3 perovskites (where R = rare earth element or Y, M = 3d-block transition metal) manifests a high intrinsic orthorhombic distortion when the R3+ ionic radius is approx‐ imately 1.11 Å, and this distortion decreases when the radii of R3+ is greater than 1.11 Å. Among the major constituents of the orthorhombic-structured perovskites that have been under exhaustive crystallography studies are RFeO3, RTiO3, RVO3, RMnO3 and RNiO3. Complemen‐ tary studies were conducted to determine the correlation between the Jahn–Teller cooperative orbital orderings of M cations with the usual site distortions [6].

structure undergoes a series of distortions; one of this is caused by a deficiency in the Sr2+ ions (A) in the framework, producing a ReO3-type structure [2]. Another distortion occurs when the ReO3 structure transforms into dense packing due to the octahedral unit rotation, produc‐ ing a structural geometry resembling the hexagonal close packing type, i.e. RhF3. The void at the centre is in an octahedral surrounding coordination; when this octahedral hole is occupied, the ilmenite structure (FeTiO3) is obtained. The slight distortions attainable in the cubic structure are due to the displacement of ions from the ideal positions, producing a variation of a few tenths of an Å. Therefore, the final symmetry varies considerably between different materials. A mechanism that likely promotes these distortions is related to the capability of perovskite to accommodate a great variety of atoms, due to the flexibility of its crystalline

The ionic radii differences produce the structural distortions on perovskite structure, as was determined in the pioneering research work conducted by Goldschimidt [1–3]. The equation that geometrically correlates the ionic radii **(***rA*, *rB* and *rO*) for the ideal cubic cell with the lattice

The expression that involves the unit cell length ratio is known as the Goldschmidt´s **tolerance factor t**, which is employed to estimate the distortion level attained by a particular perovskitestructured compound. Fundamentally, it considers the ionic radii of the constituent atoms that fit the chemical stoichiometry and enhances a pure ionic bonding between them; the mathe‐

> ( ) 2( ) *A O A O*

*r r*

According to Eq. (2), the ideal perovskite cubic structure has a "t" value equal to 1, which can be calculated for SrTiO3 with ionic radii of *rA* = 1.44 Å, *rB* = 0.605 Å and *rO* = 1.40 Å. The value of *t* is less than 1 when the A ionic radius is slightly small, and structurally the octahedral unit [BO6] tilts forward filling the additional space. From Eq. (2), the grade of tolerance where the

Orthorhombic ABO3 perovskites are among the most important constituents of the Earth´s crust. These compounds have been under exhaustive study due to the wide variety of their functional properties. The distortion of the cubic cell results in the formation of the ortho‐ rhombic structure. This process occurs by the tilting of the BO6 octahedra, but this distortion is not detectable as the temperature increases because the tilt angle decreases [5]. A typical orthorhombic structure that doubles in dimensions the cubic one is shown in Figure 2.

*r r*

*t*

ideal cubic structure is attainable in perovskite compounds is 0.89 < *t* < 1 [2–4].

*a rr rr* = += + 2 2 ( *AO AO* ) ( ) (1)

<sup>+</sup> <sup>=</sup> <sup>+</sup> (2)

structure [1–3]. An additional mechanism is the Jahn–Teller effect.

30 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

*2.1.2. Tolerance factor, orthorhombic and hexagonal crystalline unit cells*

matical expression for the tolerance factor is given in Eq. (2).

parameter *a0* is given as:

The hexagonal-structured perovskite compounds have tolerance factor values above 1. In particular, this occurs when either the A-cation or the B-cation are either too large or too small. The differences on the atomic radii distort the perovskite cubic structure forming the hexagonal perovskites, as it is shown in Figure 2b. In this structure, closely packed layers constituted by octahedra units bonded by face sharing promote the hexagonal structure. The stability of the hexagonal crystalline structure constituted by face-sharing octahedra is lower than that for the one formed by corner-sharing octahedra. However, some hexagonal perovskites belonging to this group overcome this restriction due to the metal–metal bonding between the B ions corresponding to the BX6 octahedra; these bonds are strong enough so that the metal–metal repulsion is overcome [1]. As a consequence of these stability restrictions, hexagonal perov‐ skites are less commonly found compared with the cubic perovskites. BaNiO3 is one of the hexagonal perovskites that adopt the space group *P63/mm*. The representation of the unit cell structure is shown in Figure 2b. This shows the chains of face-sharing BX6 octahedra orientated along the **c**-axis. Just like the cubic perovskites, hexagonal perovskites can also undergo distortions leading to a variety of structures.

On the other hand, the cubic cell is also susceptible to undergo a small structural deformation producing rhombohedral symmetry, but the deformation does not produce a marked enlarge‐ ment of the unit cell. Hence, the unit cell likely contains at least one or two rhombohedral polyhedral units with angles of α ~ 90° or α ~ 60°. However, the anions are displaced provoking the formation of a large unit cell with α ~ 60°.

The tetragonal-structured perovskite, BaTiO3, is probably one known example of a ferroelectric perovskite that is a stable phase at room temperature. In this structure, the TiO6 octahedra are slightly distorted (one Ti-O bond at 1.86 Å, four at 2.00 Å and a longer one at 2.17 Å). The barium atom is in coordination with four oxygen atoms at 2.80 Å, four at 2.83 Å and four more at 2.88 Å. Another tetragonal perovskites (PbHfO3, SrPbO3, SrZrO3, AgTaO3, etc.) are isotypic with BaTiO3 and possess single-molecular cells. However, a vast number of these materials exhibit the tetragonal structure at elevated temperatures, which makes crystallographic analyses difficult to conduct. In general, a significant number of perovskite-like materials have several polymorphic transformations. Some of these are important regarding their physical properties and applications. The compounds that exhibit this behaviour are BaTiO3 and KNbO3; the following transformation sequence was determined to occur by increasing temperature: rhombohedral → orthorhombic → tetragonal → cubic.

The lowest temperature crystalline compounds (orthorhombic, tetragonal and cubic) have ferroelectric properties. These particular phase transitions are reversible in nature, and all the polymorphic forms exhibit a pseudo-cubic unit cell with *a0* ~ 4 Å. Consequently, the polymor‐ phic variation on the perovskite structure further affects the crystal chemistry of this group of materials [4–6]. ferroelectric properties. These particular phase transitions are reversible in nature, and all the polymorphic forms exhibit a pseudo‐cubic unit cell with *a0* ~ 4 Å. Consequently, the polymorphic variation on the perovskite structure further affects the crystal chemistry of

The lowest temperature crystalline compounds (orthorhombic, tetragonal and cubic) have

2.17 Å). The barium atom is in coordination with four oxygen atoms at 2.80 Å, four at 2.83 Å and four more at 2.88 Å. Another tetragonal perovskites (PbHfO3, SrPbO3, SrZrO3, AgTaO3, etc.) are isotypic with BaTiO3 and possess single‐molecular cells. However, a vast number of these materials exhibit the tetragonal structure at elevated temperatures, which makes crystallographic analyses difficult to conduct. In general, a significant number of perovskite‐ like materials have several polymorphic transformations. Some of these are important regarding their physical properties and applications. The compounds that exhibit this behaviour are BaTiO3 and KNbO3; the following transformation sequence was determined

Figure 2 Typical perovskite (a) orthorhombic and (b) hexagonal structural unit cells. **Figure 2.** Typical perovskite (a) orthorhombic and (b) hexagonal structural unit cells.

#### **2.1.3. Applications of perovskite materials** *2.1.3. Applications of perovskite materials*

this group of materials [4–6].

Although the physical properties of perovskites‐structured materials are not the primary concern of the present review, some relevant ones are discussed below. Since the functionality of perovskites materials was discovered over five decades ago, hundreds of research works have directed forward elucidating the physical and chemical properties of perovskites. These studies have provided pertinent information regarding the fundamentals of the chemical and physical aspects that enhance the structural distortions in ABO3 materials. The former literature also suggests that there are still surprises to discover for this particular group of compounds, in particular, the perovskites with low tolerance factors; for instance, those comprised in the phase stability diagram FeTiO3‐LiNbO3‐*Pnma* [7]. A vast number of elements in the periodic table are likely located at either A or B unit cell sites. This fact provides an enormous range of compounds with structural similarity and a variety Although the physical properties of perovskites-structured materials are not the primary concern of the present review, some relevant ones are discussed below. Since the functionality of perovskites materials was discovered over five decades ago, hundreds of research works have directed forward elucidating the physical and chemical properties of perovskites. These studies have provided pertinent information regarding the fundamentals of the chemical and physical aspects that enhance the structural distortions in ABO3 materials. The former literature also suggests that there are still surprises to discover for this particular group of compounds, in particular, the perovskites with low tolerance factors; for instance, those comprised in the phase stability diagram FeTiO3-LiNbO3-*Pnma* [7]. A vast number of elements in the periodic table are likely located at either A or B unit cell sites. This fact provides an enormous range of compounds with structural similarity and a variety of properties. Among the most important properties are ferroelectricity (BaTiO3), ferromagnetism (Sr2FeRuO6), weak ferromagnetism (LaFeO3), colossal magnetoresistance [8–12], superconductivity (Ba0.6K0.4BiO3) [13] and large thermal conductivity (LaCoO3). Insulating to metallic transitions have a particular interest in the design of devices for thermistor applications (LaCoO3), the fluores‐ cence is applicable for laser devices (LaAlO3:Nd), and transport properties have attracted the attention of research for the development of high-temperature thermoelectric power devices (La2CuO4) [1,5,7–14].

Perovskite materials have been investigated for applications involving the preparation of solid electrolytes. The compounds that have been used in various electrochemical devices due to their high performance are barium cerate (BaCeO3) and barium zirconate (BaZrO3). Solid electrolyte performs three essential functions: (1) separates the anode from the cathode in the electrochemical cell (oxidizing and reducing sides), (2) it can operate as electronic insulator enhancing the flow of electric current through an external circuit and (3) high ionic conduction coefficient, required to provide the control of the electric current flow in the external circuit [15]. Likewise, the proton-conducting ceramics applications are classified according to two basic functions: (1) the material is capable to generate an electromotive force when undergoing a chemical potential gradient and (2) capability for electrochemical ion transport (hydrogen or oxygen) enhanced by an external power source. The employment of an ionic proton conductor as an electrolyte in devices operating under chemical potential gradient provides a device capable of producing electric energy [18,19]. The development of the solid oxide fuel cell (SOFC) electrical power sources was derived from this electrochemical principle. The SOFC is a conversion energy device that generates electricity via an electrochemical reaction occurring at temperatures above 800°C. The chemical reaction that takes place between the fuel (methane, hydrogen, natural gas) and the oxidizing agent (oxygen from air) is the motion to produce the electricity [15,18]. In this particular electrochemical converter device, the principal components responsible for the redox reaction are the electrodes; the ion transport takes place in all the cell constituents (solid electrolyte and electrodes). In contrast with the rechargeable battery devices, the SOFC does not need to be recharged, and it only requires to be continuously fed with a particular fuel for electricity generation [15–20].

phic variation on the perovskite structure further affects the crystal chemistry of this group of

Figure 2 Typical perovskite (a) orthorhombic and (b) hexagonal structural unit cells.

Although the physical properties of perovskites‐structured materials are not the primary concern of the present review, some relevant ones are discussed below. Since the functionality of perovskites materials was discovered over five decades ago, hundreds of research works have directed forward elucidating the physical and chemical properties of perovskites. These studies have provided pertinent information regarding the fundamentals of the chemical and physical aspects that enhance the structural distortions in ABO3 materials. The former literature also suggests that there are still surprises to discover for this particular group of compounds, in particular, the perovskites with low tolerance factors; for instance, those comprised in the phase stability diagram FeTiO3‐LiNbO3‐*Pnma* [7]. A vast number of elements in the periodic table are likely located at either A or B unit cell sites. This fact provides an enormous range of compounds with structural similarity and a variety

Although the physical properties of perovskites-structured materials are not the primary concern of the present review, some relevant ones are discussed below. Since the functionality of perovskites materials was discovered over five decades ago, hundreds of research works have directed forward elucidating the physical and chemical properties of perovskites. These studies have provided pertinent information regarding the fundamentals of the chemical and physical aspects that enhance the structural distortions in ABO3 materials. The former literature also suggests that there are still surprises to discover for this particular group of compounds, in particular, the perovskites with low tolerance factors; for instance, those comprised in the phase stability diagram FeTiO3-LiNbO3-*Pnma* [7]. A vast number of elements in the periodic table are likely located at either A or B unit cell sites. This fact provides an enormous range of compounds with structural similarity and a variety of properties. Among the most important properties are ferroelectricity (BaTiO3), ferromagnetism (Sr2FeRuO6), weak ferromagnetism (LaFeO3), colossal magnetoresistance [8–12], superconductivity (Ba0.6K0.4BiO3) [13] and large thermal conductivity (LaCoO3). Insulating to metallic transitions have a particular interest in the design of devices for thermistor applications (LaCoO3), the fluores‐ cence is applicable for laser devices (LaAlO3:Nd), and transport properties have attracted the attention of research for the development of high-temperature thermoelectric power devices

Perovskite materials have been investigated for applications involving the preparation of solid electrolytes. The compounds that have been used in various electrochemical devices due to their high performance are barium cerate (BaCeO3) and barium zirconate (BaZrO3). Solid electrolyte performs three essential functions: (1) separates the anode from the cathode in the

**2.1.3. Applications of perovskite materials**

*2.1.3. Applications of perovskite materials*

(La2CuO4) [1,5,7–14].

**Figure 2.** Typical perovskite (a) orthorhombic and (b) hexagonal structural unit cells.

The lowest temperature crystalline compounds (orthorhombic, tetragonal and cubic) have ferroelectric properties. These particular phase transitions are reversible in nature, and all the polymorphic forms exhibit a pseudo‐cubic unit cell with *a0* ~ 4 Å. Consequently, the polymorphic variation on the perovskite structure further affects the crystal chemistry of

2.17 Å). The barium atom is in coordination with four oxygen atoms at 2.80 Å, four at 2.83 Å and four more at 2.88 Å. Another tetragonal perovskites (PbHfO3, SrPbO3, SrZrO3, AgTaO3, etc.) are isotypic with BaTiO3 and possess single‐molecular cells. However, a vast number of these materials exhibit the tetragonal structure at elevated temperatures, which makes crystallographic analyses difficult to conduct. In general, a significant number of perovskite‐ like materials have several polymorphic transformations. Some of these are important regarding their physical properties and applications. The compounds that exhibit this behaviour are BaTiO3 and KNbO3; the following transformation sequence was determined to occur by increasing temperature: rhombohedral � orthorhombic � tetragonal � cubic.

materials [4–6].

this group of materials [4–6].

32 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Among the miscellaneous applications, such as photocatalytic reactions, some perovskites have been employed. Photocatalytic oxidation was extensively investigated by using various titanates and cobaltites. Titanate perovskite materials (SrTiO3 and BaTiO3) were found to exhibit a strong photocatalytic effect, in contrast with that determined for the easily reducible LaCoO3. Perovskite materials have demonstrated a notable performance as hosts in laser devices because in the A site of the structure some rare earth elements (Nd3+ or Sm3+) are incorporated. These elements are some of the major constituents employed to produce laser ion devices. Moreover, various perovskite oxides possess high electrical resistivity, which makes them useful for the preparation of dielectric materials that are poor conductors of electricity, which are widely used for electrical insulation in electrical power systems. Addi‐ tionally, these materials can be used for energy storage purposes as in capacitors. Typical conductor and semiconductor oxides are characterized to contain either B-ions with a net charge lower than the stable B4+, or B-ions having two different net charges. Typical materials that have been considered to be good conductors or semiconductors are LaVO3, LaTiO3 and SrMoO3. Miscellaneous applications of dielectric perovskites include mechanical actuation, sound generation, in materials subjected to dimensional changes as a result of an applied voltage (piezoelectrics), for transducers as in condensers, for piezoelectric microphones, for detecting changes in temperature as in pyroelectrics, and as liquid crystals employed for alphanumeric displays.

Perovskite oxides materials are also used as oxygen sensors. The mechanism involved in this type of sensors is electrical conductivity in nature, which enhances the oxygen adsorption in the crystalline structure. The electrical conductivity is proportional to oxygen partial pressure and to the concentration of vacant sites in the X site of ABX3. One compound that exhibits a high sensitivity to oxygen motion is SrTiO3. Furthermore, SrSnO3 is a promising material for combustion monitoring-sensors. The design of functional materials requires an understanding of the relationship between chemical composition and crystalline structure [15–21].

## *2.1.4. Synthesis of perovskite materials*

The preparation of perovskite oxides has been the subject of considerable research because the design and development of new technological properties require the preparation of materials with special physical morphology. Some examples of the new developments include thin films, porous solids, monodispersed powders with nanometric size, among others. In addition to this, the interest for specific structural ordering exhibiting chains that enhances unidirectional properties has prompted various research groups to pursue the preparation of new perovskite oxides with specific compositions [22,23]. The conventional method of solid-state reaction (ceramic method) was broadly used to produce perovskite oxides powders during the early 1940s. An analogous method, the flux method, which is also based on solid-state reaction, has been employed for the preparation of single crystals of various perovskite oxides [23]. Both methods are adequate to synthesize oxide compounds because they allow achieving the proper conditions in an easy way, due to the reaction proceeding under air atmosphere and ambient pressure [24]. Common steps involved in the solid-state reaction methods are homogeneous grinding and mixing. Oxide and carbonate solid reactants are among the most employed. These are mixed in stoichiometric amounts prior to the conduction of a heat treatment at elevated temperatures (< 1000°C) for long periods of time (several hours or days). The grinding and heating cycles are repeated until the desired pure phase is successfully produced. Any doubt exists in regards of the usefulness and easiness of the universal method, but the achievement of the conditions for atomic diffusion could be a problem. Hence, to accelerate the atomic diffusion for preventing the formation of metastable phases, the process must proceed at high temperatures. However, this has the disadvantage of promoting the volatility of reactants such as lead oxide (in the case of compounds containing Pb2+ as in the PbTiO3). In addition, the need of regrinding the materials several times involves high energy and time consumption, as these parameters have a strong influence on achieving the desired product functionality [22–24].

Recently, the solid-state metathesis (SSM) reaction method has emerged as an efficient processing route for synthesizing a broad range of non-oxide compounds. The process occurring under thermodynamic equilibrium conditions is more effective than the conven‐ tional solid-state reaction method [22,25]. The SSM reaction method is a variant of the SHS process in which the chemical reactions are conducted very rapidly. A typical SSM reaction involves an ionic exchange taking place between the reactants to produce thermodynamically stable perovskites. Along with the reaction, a remarkable enthalpy change occurs, and high adiabatic reaction temperatures are reached. The solid-state metathesis reaction is conducted using alkaline precursors, chalcogenide, silicide or boride; the salt is reacted with a metal halide according to the general [reaction 3]

$$A\_p B\_q + CX\_p \to BX\_q + pAX \tag{3}$$

where A = Li, Na, K, Mg, Ca, Sr, or Ba; B = B, Si, N, P, As, Sb, Bi, O, S, Se, or Te; C = transition group, principal group or actinide metal, and X = halogen.

combustion monitoring-sensors. The design of functional materials requires an understanding

The preparation of perovskite oxides has been the subject of considerable research because the design and development of new technological properties require the preparation of materials with special physical morphology. Some examples of the new developments include thin films, porous solids, monodispersed powders with nanometric size, among others. In addition to this, the interest for specific structural ordering exhibiting chains that enhances unidirectional properties has prompted various research groups to pursue the preparation of new perovskite oxides with specific compositions [22,23]. The conventional method of solid-state reaction (ceramic method) was broadly used to produce perovskite oxides powders during the early 1940s. An analogous method, the flux method, which is also based on solid-state reaction, has been employed for the preparation of single crystals of various perovskite oxides [23]. Both methods are adequate to synthesize oxide compounds because they allow achieving the proper conditions in an easy way, due to the reaction proceeding under air atmosphere and ambient pressure [24]. Common steps involved in the solid-state reaction methods are homogeneous grinding and mixing. Oxide and carbonate solid reactants are among the most employed. These are mixed in stoichiometric amounts prior to the conduction of a heat treatment at elevated temperatures (< 1000°C) for long periods of time (several hours or days). The grinding and heating cycles are repeated until the desired pure phase is successfully produced. Any doubt exists in regards of the usefulness and easiness of the universal method, but the achievement of the conditions for atomic diffusion could be a problem. Hence, to accelerate the atomic diffusion for preventing the formation of metastable phases, the process must proceed at high temperatures. However, this has the disadvantage of promoting the volatility of reactants such as lead oxide (in the case of compounds containing Pb2+ as in the PbTiO3). In addition, the need of regrinding the materials several times involves high energy and time consumption, as these parameters have a strong influence on achieving the desired product

Recently, the solid-state metathesis (SSM) reaction method has emerged as an efficient processing route for synthesizing a broad range of non-oxide compounds. The process occurring under thermodynamic equilibrium conditions is more effective than the conven‐ tional solid-state reaction method [22,25]. The SSM reaction method is a variant of the SHS process in which the chemical reactions are conducted very rapidly. A typical SSM reaction involves an ionic exchange taking place between the reactants to produce thermodynamically stable perovskites. Along with the reaction, a remarkable enthalpy change occurs, and high adiabatic reaction temperatures are reached. The solid-state metathesis reaction is conducted using alkaline precursors, chalcogenide, silicide or boride; the salt is reacted with a metal halide

*A B CX BX pAX pq p q* + ®+ (3)

of the relationship between chemical composition and crystalline structure [15–21].

*2.1.4. Synthesis of perovskite materials*

34 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

functionality [22–24].

according to the general [reaction 3]

The effectiveness of transferring the heat required achieving the chemical reactions under solid-state conditions for the methods mentioned above constitutes one of their main disadvantages from the point of view of the energy cost. One of the new processing techniques that have overcome this problem is the one known as "microwave irradia‐ tion". During the microwave processing conditions, the heating is directly supplied to the solid reactants, because it proceeds from the interaction at a molecular level between the electromagnetic field and some of the elements constituting the solid. This interaction varies depending on factors such as the dielectric and magnetic properties of the solid reactants, frequency and microwave power generation, microwave permeability and size and density of the materials [24]. The microwave heating improves the reaction kinetics in a range between 10 and 1,000 folds in comparison with the conventional and the SSM routes, resulting in significant differences associated with the crystalline structure and properties of the reaction products produced. Recently, the synthesis of various perovskite-struc‐ tured compounds has been under extensive study by microwave processing. Table 1 summarizes the processing details regarding the synthesis of some single-phase perov‐ skite materials produced under microwave irradiation [24].

Chemists in material research are constantly seeking to develop new processing methods with lower costs of energy consumption and that are more environmentally friendly. The novel methods are remarkable functional, allowing us to synthesize perovskite oxides under milder conditions, in comparison with the conventional solid-state process [26–28]. The ceramic materials produced by non-conventional techniques, such as soft chemistry routes, have a more homogeneous distribution of constituents, are purer, can be produced with a wide range of partial dopant substitutions and with various shapes (i.e. coatings and fibers), which are not really obtainable via conventional methods [29]. Soft chemical processing routes also allow the preparation of bulk monoliths without the use of intermediate powders, which occurs by drying a precursor gel at supercritical conditions [29].

Coprecipitation and sol–gel routes emerged since the early 1960s as processing alternatives for ceramic powders, including some perovskite compounds. Certain novelty aspects are inherent in these techniques, which are associated with the control of physical aspects such as size, the state of aggregation and purity of the produced compounds. The functionality of these processes is based on the principles related to the mechanism of formation of colloids in aqueous media at very low temperatures. In the aqueous "sol–gel" processing, the starting chemical reagents, i.e. salt solutions containing specific cation species, undergo chemical treatments at mild temperatures (< 100°C) to form stable colloid dispersions, or sols, when mixed with an acid or basic solution. The particle size of the sols varies between 1 nm and 1 μm. The sols can be prepared from highly hydrolyzable salt reagents such as ZrCl4 or TiCl4, i.e. in the case of perovskite zirconate or titanate. In the "sol–gel" processing, the crystallization step occurs at calcination temperatures much lower than those associated with the preparation of the same crystalline phase via the solid-state process. Prior to this step, the separation of the gel from the sol is conducted by a preliminary dehydration stage [29,30]. Enormous efforts have been carried out over the past three decades aiming to develop new alternative soft chemistry routes that allow us to reduce the sol–gel crystallization temperature and to control the stoichiometry of the reaction products. One of the processing routes recently proposed involves the reaction of a hydroxide reagent precursor, either NaOH or KOH, with Nb(OH)5⋅xH2O. The dissolution of Nb2O5 in an HF solution was the preliminary step to prepare the precursor of niobium. The hydroxide powder mixture containing stoichiometric amounts of the precursor reagents (sources of K and Nb) was heat-treated at different temperatures ranging from 200 to 700°C for 6 h in air. The orthorhombic KNbO3 perovskite phase was produced without the formation of any by-products at a temperature of 700°C while the NaNbO3 powders were prepared at a temperature as low as 500°C [30].


\*Note: Time not specified, the product was obtained after several periods of ball milling.

**Table 1.** Summary of the processing conditions employed for the preparation of niobate and titanate perovskites via microwave and conventional solid-state processing methods (data taken from reference [24]).

In approaches based on the preparation of polymer precursors using organic solvent solutions, namely the Pechini method and its alternative process polymerized complex method (PC). A polymer containing the cation precursors of the perovskite compound is produced via a chemical reaction occurring between suitable organometallic precursors and the liquid solvent. The polymer precursor is produced by a treatment conducted at low temperature (200–400°C), leading to a compositional homogenization of the elements allocated in the polymer structure [31–33]. The amorphous polymer precursors are usually calcinated to promote the crystallization of the stable KNbO3 [31,32] and BaTiO3 [33] ceramic phases. Furthermore, the polymer precursors can be designed, in terms of their constituents and composition, so as to improve their combustion capability. Based on this principle, a combus‐ tion technique has been proposed as a novel soft chemical route. This technique is of industrial interest because it allows processing large batch volumes of powder. Furthermore, the production of nanocrystalline oxide ceramics powders proceeds at lower calcination temper‐ ature in a very short time. The produced particles have the highest degree of purity and characteristics such as narrow particle size distribution, higher surface area that enhances its sinterability [32,33]. The combustion technique is based on the exothermic decomposition of a fuel-oxidant precursor. The reaction promotes the formation of a fine monodispersed powder with perovskite structure, or a partially decomposed precursor containing considerable amounts of carbon traces. The results after processing depend on the precursor and in the fuelto-oxidant ratio used to conduct the combustion. During the decomposition, the decomposi‐ tion of the organic compound facilities the rapid increase of the temperature coupled with gas production, resulting in the coalescence of particles inconsequence short diffusion pathways as well. Hence, a foamed porous aggregate formed by a pure-phase nanoparticles agglomer‐ ated can be obtained at low temperatures [32,33].

Miscellaneous processing techniques involving non-equilibrium reaction conditions are also of industrial interest to produce perovskite oxides. The glycothermal method is a novel technique involving crystallization of oxide particles, i.e., KNbO3, and which is conducted at supercritical conditions, depending on the chosen organic solvent. At supercritical conditions, the organic solvent capability is similar to that of the normal polar liquids, but it exhibits better transport properties (viscosity, diffusivity and thermal conductivity, among others) [34]. This technique is reliable in terms of reproducibility and environmental aspects. However, the high cost of the organometallic reagent precursors and the restricted types of organic solvents are some disadvantages associated with this technique. An analogous technique that employs only water as the solvent, hydrothermal processing has been widely used for more than three decades to synthesize a numerous variety of perovskite materials. This technology was explored in the engineering fields of crystal growth and metal leaching in the middle of the 20th century. Hitherto, the hydrothermal technology broadly covers various interdisciplinary fields of materials science. In the solid-state chemistry field, the hydrothermal media have recently been exploited for preparing vast type of ceramic compounds. This technology provides an efficient reaction environment for synthesizing perovskite powders, due to the effect of a combination of parameters, such as solvent media, temperature and pressure, on the ionic reaction equilibrium. The conventional hydrothermal (CH) method is an efficient route that enhances the crystallization of micro/nanometric morphology controlled and crystal growth-oriented particles. In addition to that, this method depends on the inorganic salts solubility in water under variable temperature and pressure conditions. Another fundamental factor that has a marked influence enhancing the heterogeneous reactions in this process is the vapour pressure. Hence, a detailed state-of-the-art regarding this technique is discussed in the next sections. A particular emphasis that considers the theoretical principles associated with thermodynamic modelling of the heterogeneous chemical reaction equilibrium associated with a particular hydrothermal environment is addressed. Additionally, the link between the theoretical principles and the extensive practical expertise gained over the past decades regarding the hydrothermal synthesis of the most representative perovskite compounds is further discussed in the final section of the present review.

## *2.1.5. Definition of hydrothermal synthesis*

have been carried out over the past three decades aiming to develop new alternative soft chemistry routes that allow us to reduce the sol–gel crystallization temperature and to control the stoichiometry of the reaction products. One of the processing routes recently proposed involves the reaction of a hydroxide reagent precursor, either NaOH or KOH, with Nb(OH)5⋅xH2O. The dissolution of Nb2O5 in an HF solution was the preliminary step to prepare the precursor of niobium. The hydroxide powder mixture containing stoichiometric amounts of the precursor reagents (sources of K and Nb) was heat-treated at different temperatures ranging from 200 to 700°C for 6 h in air. The orthorhombic KNbO3 perovskite phase was produced without the formation of any by-products at a temperature of 700°C while

Na2CO3 NaNbO3 800 17 1,250 \* K2CO3 KNbO3 800 12 1,000 1,800

PbNO3 PbTiO3 600 9 360 480

**Table 1.** Summary of the processing conditions employed for the preparation of niobate and titanate perovskites via

In approaches based on the preparation of polymer precursors using organic solvent solutions, namely the Pechini method and its alternative process polymerized complex method (PC). A polymer containing the cation precursors of the perovskite compound is produced via a chemical reaction occurring between suitable organometallic precursors and the liquid solvent. The polymer precursor is produced by a treatment conducted at low temperature (200–400°C), leading to a compositional homogenization of the elements allocated in the polymer structure [31–33]. The amorphous polymer precursors are usually calcinated to promote the crystallization of the stable KNbO3 [31,32] and BaTiO3 [33] ceramic phases. Furthermore, the polymer precursors can be designed, in terms of their constituents and composition, so as to improve their combustion capability. Based on this principle, a combus‐ tion technique has been proposed as a novel soft chemical route. This technique is of industrial interest because it allows processing large batch volumes of powder. Furthermore, the production of nanocrystalline oxide ceramics powders proceeds at lower calcination temper‐ ature in a very short time. The produced particles have the highest degree of purity and characteristics such as narrow particle size distribution, higher surface area that enhances its sinterability [32,33]. The combustion technique is based on the exothermic decomposition of

**Microwave Conventional**

**Temperature (°C)**

**Time (min)**

**Time (min)**

LiNbO3 800 15 500 720

BaTiO3 1,000 25 1,400 \*

the NaNbO3 powders were prepared at a temperature as low as 500°C [30].

**product**

\*Note: Time not specified, the product was obtained after several periods of ball milling.

microwave and conventional solid-state processing methods (data taken from reference [24]).

**Chemical reagents supplying A and B**

Li2CO3

BaCO3

**cations Reaction**

36 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**A cation supplier B cation supplier Power (W)**

Nb2O5

TiO2

The word "hydrothermal" comes from the etymological root of the Greek word "hydrous" that means water while "thermal" means heat. The "hydrothermal" term has a purely geological origin. In the middle of the 18th century, the Scottish Geologist Sir Roderick Murchinson introduced this term to the Scientific Society. In practical terms, he described that the formation of mineral species that occurred on Earth's crust is due to the reaction of water exposed to conditions of elevated temperature and pressures. Another technical definition accepted by the Scientific Society refers the "hydrothermal" term as any heterogeneous chemical reaction, occurring in the presence of a solvent media. The reactions occur inside a hermetical sealed vessel system at temperatures above 25°C and pressure of 0.1 MPa. Under hydrothermal conditions, it does not matter whether the solvent is aqueous or non-aqueous. The crystallization process of solid phases under hydrothermal conditions is carried out at autogenous pressure, achieved by the saturated vapour pressure of the fluid at the specified temperature and composition of the hydrothermal solution. In this concern, in terms of industrial and commercial processing, mild operating conditions are preferred, for example, temperatures of treatment below 350°C and pressures less than 50 MPa [35, 36]. The processing parameter that allows the transition from mild to severe reaction conditions during a hydro‐ thermal treatment is the lining material strength of the autoclave vessel. At severe treatment conditions in highly concentrated acidic or basic solutions, the lining material might undergo a rapid corrosion process. The progress in the experimental work research in this area of investigation has allowed improving the understanding of how is the behaviour of the chemical reactions generated in the hydrothermal media. The crystallization of several materials of oxide and non-oxide has been made with adequate optimization of the experi‐ mental parameters during the hydrothermal treatment (T < 200°C and *P* < 1.5 MPa). The recent scientific and technological achievements have made the hydrothermal synthesis more economical. One example is the synthesis of ceramic particles, which can be prepared in a single step, using advanced pressure reactor technology coupled with processing methodol‐ ogies proposed for a wide number of inorganic compounds [37].

## **2.2. Thermodynamic modelling associated with the hydrothermal synthesis of perovskite oxides**

In the early 1990s, a great interest to prepare monodispersed perovskite oxide fine particles with controlled morphology was the main concern of the chemical community. One chemical technique that provides an adequate environment to accomplish the preparation of a substan‐ tial number of ceramic materials is the hydrothermal processing. Because this method combines the dynamic interaction of processing parameters such as solvent type, temperature and pressure that governs the ionic mobility, this particular technique involves numerous simultaneous chemical reactions, which normally take place in an aqueous system comprising the interactions of dissolved and solid species. The reaction product obtained consists of an anhydrous single-phase crystalline oxide or multicomponent phases. The huge diversity of precursor chemical reactants (i.e. water, soluble salts, hydroxides and oxides, among others) allows preparing solvents that can be employed as hydrothermal media. This processing parameter promotes a broad range of reaction pathways to achieve the crystallization of a solid in a particular hydrothermal reaction system. In general, phase-pure oxides (perovskitestructured) with specific stoichiometry, particle size and morphology can be hydrothermally produced in one step processing, even using low-cost reagents at mild temperatures and pressures, five-fold lower than those required for conventional processing techniques.

Hitherto, the practical fundamentals derived from the expertise gained over the past five decades. Regarding the processing of inorganic compounds by the hydrothermal technology, to take an adequate advantage of the technology´s novelty, one must bear in mind some critical key processing factors such as the selection of a suitable precursor system (highly reactive and cost-effective) required for the optimization of the chemical equilibrium that enhances the crystallization of the desired phase. The effectiveness of the selected experimental approach can be evaluated selecting the suitable chemical precursor concentration and mixing ratios, pH of the hydrothermal media, temperature and pressure level [38,39]. This procedure is relatively complex and time-consuming due to the numerous variables involved. Hence, to determine the effectiveness of the precursor hydrothermal system, an approach based on thermodynamic modelling to simulate the hydrothermal reactions has been developed since the early 1990s [40]. OLI Systems Inc. (USA) developed an algorithm that included the thermodynamic basis to simulate the chemical reactions. The model is capable to calculate thermochemical data with high consistency and accuracy for a broad number of perovskite compounds, including the single-phase CaTiO3, SrTiO3, SrZrO3, PbTiO3, BaTiO3, as well as some selected solid solutions of Ba1–*x*Sr*x*TiO3 and PbZr1–*x*Ti*x*O3 [41–49]. In addition to that, these studies have produced relevant information regarding the behaviour of solutions under modifying pressure and temperature conditions. The most features investigated under hydrothermal conditions are solubility, stability and yield product amount, reaction mecha‐ nisms, among others.

## *2.2.1. Thermodynamic model*

geological origin. In the middle of the 18th century, the Scottish Geologist Sir Roderick Murchinson introduced this term to the Scientific Society. In practical terms, he described that the formation of mineral species that occurred on Earth's crust is due to the reaction of water exposed to conditions of elevated temperature and pressures. Another technical definition accepted by the Scientific Society refers the "hydrothermal" term as any heterogeneous chemical reaction, occurring in the presence of a solvent media. The reactions occur inside a hermetical sealed vessel system at temperatures above 25°C and pressure of 0.1 MPa. Under hydrothermal conditions, it does not matter whether the solvent is aqueous or non-aqueous. The crystallization process of solid phases under hydrothermal conditions is carried out at autogenous pressure, achieved by the saturated vapour pressure of the fluid at the specified temperature and composition of the hydrothermal solution. In this concern, in terms of industrial and commercial processing, mild operating conditions are preferred, for example, temperatures of treatment below 350°C and pressures less than 50 MPa [35, 36]. The processing parameter that allows the transition from mild to severe reaction conditions during a hydro‐ thermal treatment is the lining material strength of the autoclave vessel. At severe treatment conditions in highly concentrated acidic or basic solutions, the lining material might undergo a rapid corrosion process. The progress in the experimental work research in this area of investigation has allowed improving the understanding of how is the behaviour of the chemical reactions generated in the hydrothermal media. The crystallization of several materials of oxide and non-oxide has been made with adequate optimization of the experi‐ mental parameters during the hydrothermal treatment (T < 200°C and *P* < 1.5 MPa). The recent scientific and technological achievements have made the hydrothermal synthesis more economical. One example is the synthesis of ceramic particles, which can be prepared in a single step, using advanced pressure reactor technology coupled with processing methodol‐

38 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

ogies proposed for a wide number of inorganic compounds [37].

**oxides**

**2.2. Thermodynamic modelling associated with the hydrothermal synthesis of perovskite**

In the early 1990s, a great interest to prepare monodispersed perovskite oxide fine particles with controlled morphology was the main concern of the chemical community. One chemical technique that provides an adequate environment to accomplish the preparation of a substan‐ tial number of ceramic materials is the hydrothermal processing. Because this method combines the dynamic interaction of processing parameters such as solvent type, temperature and pressure that governs the ionic mobility, this particular technique involves numerous simultaneous chemical reactions, which normally take place in an aqueous system comprising the interactions of dissolved and solid species. The reaction product obtained consists of an anhydrous single-phase crystalline oxide or multicomponent phases. The huge diversity of precursor chemical reactants (i.e. water, soluble salts, hydroxides and oxides, among others) allows preparing solvents that can be employed as hydrothermal media. This processing parameter promotes a broad range of reaction pathways to achieve the crystallization of a solid in a particular hydrothermal reaction system. In general, phase-pure oxides (perovskitestructured) with specific stoichiometry, particle size and morphology can be hydrothermally

Today, our knowledge of the physical chemistry of the hydrothermal processes is en‐ riched by the experience acquisition in the area of hydrothermal solution chemistry. The role of the solvent during the hydrothermal treatment needs to be understood because it depends on the experimental variables temperature and pressure. Also, the solvent reactivity is affected under hydrothermal conditions by some interrelated parameters such as the structure at critical, supercritical and subcritical conditions, dielectric constant, pH variation, viscosity, coefficient of expansion and density. In general, the thermodynamic modelling applied to a hydrothermal system constitutes a potential tool for predicting the concentration and activities of ionic or neutral substances used as solvents. These include a reaction system constituted for some solid, vapour and non-aqueous phases. A particu‐ lar methodology involving a model based on the theory of electrolyte systems is establish‐ ed; the model enables a quantitative description of both the phase and the ionic equilibrium and provides values of the activity coefficients. Furthermore, the models provide data about the standard-state properties of all the substances involved in the reaction model, cou‐ pled with the Gibbs energy excess (referred to a non-ideal solution). Rafal and Zemaitis have given a complete interpretation of the experimental design fundamentals for thermo‐ dynamic modelling of the hydrothermal systems, as described in Ref. [39].

The hydrothermal medium is an environment for the inorganic particle synthesis. The chemical interactions between solid and fluid phases are affected mainly by variables such as solvent liquid, temperature, reaction interval, pH of the solvent fluid, the initial concentration of the precursor feedstock and occasionally pressure. The correct selection of these parameters is crucial to determinate the principal production conditions that allow the formation of singlephase stoichiometric compounds, namely ABX3. Pioneering hydrothermal research investi‐ gations were conducted empirically, establishing the experimental parameters in a trial-anderror fashion, resulting in an inaccurate approximation for a practical particle synthesis. The hydrothermal reaction system becomes complex when highly concentrated solutions rich in ions are employed, or if multiple heterogeneous chemical reactions alternatively proceed in the hydrothermal media. In this particular case, the ionic concentration related to the chemical equilibrium promoted in the hydrothermal fluid is strongly dependent on ionic species activity coefficients [38]. At the early 1990s, Lenka and Riman reported the first rigorous approach that is in essential a practical thermodynamic model of ionic species hydrolyzed in a hydrothermal reaction system [40]. The model is capable of predicting or optimizing the experimental conditions, namely feedstock composition, solution pH, temperature and pressure, to mini‐ mize the Edisonial trial and error design. Additionally, this model produces a series of phase stability diagrams based on the interaction of the main parameters, such as feedstock concen‐ tration, solvent pH and its concentration.

The parameters required to obtain a hydrothermal phase stability diagram are the equilibrium concentrations of the ionic species added in the system as a function of temperature, pressure and the initial content of the precursor feedstock. Initially, the number of independent chemical reaction equilibria produced under hydrothermal conditions is denoted by "*k*". The specific jth reaction (*j* = 1,..., *k*) imply *nj* different chemical substances; these are represented by *Ai (j)* (*i* = 1,..., *k*). Thus, the number of possible reactions occurring in an infinitesimal point in the hydrothermal system can be given by the following expression 4.

$$\sum\_{l=1}^{n\_j} \mathbf{v}\_l^{(j)} A\_i^{(j)} = \mathbf{0}, \qquad j = 1, \ldots, k. \tag{4}$$

While the equilibrium state of any *j*th reaction is determined using the expression for the variation in the standard Gibbs energy

$$
\Delta \mathbf{G}\_{\rangle}^{0} = \sum\_{i=1}^{n\_{\neq}} \mathbf{v}\_{i}^{(\prime)} \Delta \mathbf{G}\_{\neq}^{0} \left( \mathbf{A}\_{i}^{(\prime)} \right) = -RT \ln \mathbf{K}\_{\neq} \left( T, P \right), \qquad j = 1, \ldots, k. \tag{5}
$$

Where *G* <sup>0</sup> (*Ai* ( *j*) ) corresponds to the standard Gibbs energy for the substances formation (*Ai* ( *j*) ) and *Kj* represents the equilibrium constant of the *j*th reaction.

The equilibrium constant of any reaction is related to the molality "*m*", which is the used concentration unit. This can be expressed as

Synthesis of Perovskite Oxides by Hydrothermal Processing – From Thermodynamic Modelling to Practical... http://dx.doi.org/10.5772/61568 41

$$K\_{\boldsymbol{j}}\left(\boldsymbol{T},\boldsymbol{P}\right) = \prod\_{l=1}^{n\_{\boldsymbol{j}}} \left(m\_{\boldsymbol{A}\_{l}^{(l)}} \boldsymbol{\mathcal{Y}}\_{\boldsymbol{A}\_{l}^{(l)}}\right)^{\boldsymbol{v}\_{l}^{(l)}},\tag{6}$$

The hydrothermal medium is an environment for the inorganic particle synthesis. The chemical interactions between solid and fluid phases are affected mainly by variables such as solvent liquid, temperature, reaction interval, pH of the solvent fluid, the initial concentration of the precursor feedstock and occasionally pressure. The correct selection of these parameters is crucial to determinate the principal production conditions that allow the formation of singlephase stoichiometric compounds, namely ABX3. Pioneering hydrothermal research investi‐ gations were conducted empirically, establishing the experimental parameters in a trial-anderror fashion, resulting in an inaccurate approximation for a practical particle synthesis. The hydrothermal reaction system becomes complex when highly concentrated solutions rich in ions are employed, or if multiple heterogeneous chemical reactions alternatively proceed in the hydrothermal media. In this particular case, the ionic concentration related to the chemical equilibrium promoted in the hydrothermal fluid is strongly dependent on ionic species activity coefficients [38]. At the early 1990s, Lenka and Riman reported the first rigorous approach that is in essential a practical thermodynamic model of ionic species hydrolyzed in a hydrothermal reaction system [40]. The model is capable of predicting or optimizing the experimental conditions, namely feedstock composition, solution pH, temperature and pressure, to mini‐ mize the Edisonial trial and error design. Additionally, this model produces a series of phase stability diagrams based on the interaction of the main parameters, such as feedstock concen‐

40 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The parameters required to obtain a hydrothermal phase stability diagram are the equilibrium concentrations of the ionic species added in the system as a function of temperature, pressure and the initial content of the precursor feedstock. Initially, the number of independent chemical reaction equilibria produced under hydrothermal conditions is denoted by "*k*". The specific jth reaction (*j* = 1,..., *k*) imply *nj* different chemical substances; these are represented by *Ai*

1,..., *k*). Thus, the number of possible reactions occurring in an infinitesimal point in the

0, 1, .

While the equilibrium state of any *j*th reaction is determined using the expression for the

, , 1, .

The equilibrium constant of any reaction is related to the molality "*m*", which is the used

<sup>D</sup> <sup>=</sup> å =- <sup>=</sup> ¼D (5)

) corresponds to the standard Gibbs energy for the substances formation (*Ai*

*G v G A RTlnK T P j k*

å = = ¼¼ (4)

*vA j k*

hydrothermal system can be given by the following expression 4.

() ()

*j j i i*

( ) ( ) 0 0 () ()

1

*j j j i fi j*

represents the equilibrium constant of the *j*th reaction.

*i*

=

*j n*

*(j)* (*i* =

( *j*) )

tration, solvent pH and its concentration.

variation in the standard Gibbs energy

Where *G* <sup>0</sup>

and *Kj*

(*Ai* ( *j*) 1

*i*

concentration unit. This can be expressed as

=

*j n*

where *γAi* ( *<sup>j</sup>*) represents the activity coefficient of the ionic species *Ai* ( *j*) . To solve Eqs. 4 and 5, the mass and electro neutrality equilibria must be established, coupled with the standard Gibbs energy values associated to the formation and activity coefficients. In parallel to the solution of these equations, the Helgeson-Kirkham-Flowers-Tanger (HKFT) equation of state must be taken into account. The HKFT equation provides the standard-state thermodynamic functions of aqueous, ionic and neutral species as a function of both temperature and pressure up to 1000°C and 500 MPa, respectively. In addition, the values of the standard Gibbs energy of formation are calculated from the standard Gibbs energy, Δ*Gf* 0 enthalpy Δ*Hf* 0 of formation and entropy *So* at a reference temperature (usually, 25°C, 298.15 K). Furthermore, parameters that must be considered are the partial volume Vo and the heat capacity *Cp* <sup>0</sup> as functions of tem‐ perature. The algorithm OLI Systems Inc employs a Debye-Hükel term for net ionic species interactions, a modified Bromley expression (Bromley-Zemaitis) for a short-range ionic interactions and the Pitzer expression of ion-neutral molecule interactions (as described in detail in Ref. [38]). Some restrictions in the use of the Pitzer term are taken into account in a simplified version of the model algorithm, and also without considering three-body interac‐ tions. The OLI software data bank contains vast information associated with the standard thermodynamic values for several liquids, gaseous and solid substances; data from other sources should strictly be evaluated before using them. The information of solid substances is obtained from the JANAF/NBS databases, as well as from some other particular references such as Medvedev [43] and Robie [44] in Ref. [38]. For the case of ceramic materials, especially perovskite oxides, i.e. lead and alkaline-earth titanates, thermochemical data of pure substan‐ ces have been reported by Brain [40].

Solubility databases of solid species in pure water and other solutions, namely alkaline and/or acidic solutions, as a function of temperature, are the best reference data to obtain the standard-state properties. These data can be taken from the books Thermochemical Data of Pure Substances and Solubilities of Inorganics and Metal Organic Compounds (Refs. 45 and 46 in Ref. [38]). However, precise and accordant thermochemical data regarding the solubility of solid compounds can be obtained by regression fitting performed using the OLI software [38,40].

The methodology proposed by Lenka and Riman [38] to determine the chemical reaction equilibria is associated with a set of feedstock precursors. For example, an organic salt (barium acetate) or an inorganic salt (barium nitrate) and an oxide containing a tetravalent cation are used for the preparation of a double cation perovskite oxide (ABX3). The set of chemical reactions that are likely to occur in the hydrothermal process are summarized in Table 2. This table contains the equilibria equations of 37 aqueous and solid substances that might prevail in the Ba-Ti-H2O system. One fact that deserves emphasis is related to the usage of HNO3 and KOH because the addition of acidic or alkaline reagents is required to control the pH of the hydrothermal fluid. The presence of these solutions strongly affects the crystallization of the oxide species. The most common reagents employed to adjust the pH of the hydrothermal reaction media are alkaline solutions, such as hydroxide solutions and ammonia. In order to complete the set of modelling data, the gaseous species produced during the hydrothermal treatment needs to be calculated as well. In general, for the synthesis of perovskite compounds under specific T–P conditions, the gaseous species play a secondary role in their formation. It is necessary to remark that even a relatively simple hydrothermal system (Ba-Ti-H2O) includes a vast variety of species, which lead to the formation of a lot of equations for resolving.

The estimated entropies corresponding to the possible reactions in the hydrothermal medium are given in Table 3. Once thermodynamic parameters are determined at room temperature condition, their values at different conditions must be determined using the approach described in Ref. [40]. The complete set of data referring to the ionic species standard-state properties are used to obtain the Ba-Ti system stability diagram. The data show the predom‐ inant phases at the conditions of T and pH as a function of the net content of Ba (*mBaT* ) or Ti (*mT <sup>i</sup> <sup>T</sup>* ). The standard-state values of all the species involved in the reaction system are calculated assuming that inner pressure in the hydrothermal vessels is autogenously controlled.

The iterative practice of the model using different A and B site feedstock concentrations, in conjunction with the addition of pH adjusting agents (mineralizer or solvent), allows to estimate diagrams that correlates product yield and crystalline phase stability.


**Table 2.** Reaction equilibria in the Ba-Ti-H2O hydrothermal system using TiO2, Ba(CH3COO)2 and/or Ba(NO3)2 as precursor feedstock [38,40].

Synthesis of Perovskite Oxides by Hydrothermal Processing – From Thermodynamic Modelling to Practical... http://dx.doi.org/10.5772/61568

hydrothermal fluid. The presence of these solutions strongly affects the crystallization of the oxide species. The most common reagents employed to adjust the pH of the hydrothermal reaction media are alkaline solutions, such as hydroxide solutions and ammonia. In order to complete the set of modelling data, the gaseous species produced during the hydrothermal treatment needs to be calculated as well. In general, for the synthesis of perovskite compounds under specific T–P conditions, the gaseous species play a secondary role in their formation. It is necessary to remark that even a relatively simple hydrothermal system (Ba-Ti-H2O) includes a vast variety of species, which lead to the formation of a lot of equations for resolving.

42 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The estimated entropies corresponding to the possible reactions in the hydrothermal medium are given in Table 3. Once thermodynamic parameters are determined at room temperature condition, their values at different conditions must be determined using the approach described in Ref. [40]. The complete set of data referring to the ionic species standard-state properties are used to obtain the Ba-Ti system stability diagram. The data show the predom‐ inant phases at the conditions of T and pH as a function of the net content of Ba (*mBaT* ) or Ti

*<sup>T</sup>* ). The standard-state values of all the species involved in the reaction system are calculated

Ba(CH3COO)+

BaNO3 <sup>+</sup>

= Ba2+ + (CH3COO)-

+ NO3 -

+ (CH3COO)-

+ (CH3COO)-

Ba(CH3COO)2(aq) = Ba(CH3COO)+

 = Ba2+ + NO3 - Ba(NO3)2(aq) = BaNO3 <sup>+</sup>

(CH3COO)2(aq) = 2HCH3COO(aq) (CH3COO)2(vap) = (CH3COO)2(aq)

+ NO3 -

+ NO3 -

+ OH-

+ 2H2O

+ Ti(OH)4(aq) + 2OH-

+ (CH3COO)-

+ OH-

Ba(NO3)2(s) = Ba2+ + 2NO3 - HCH3COO(g) = HCH3COO(aq)

HCH3COO(aq) = H+

HNO3(g) = HNO3(aq) HNO3(aq) = H+

KOH(s)⋅2H2O = K+

KCH3COO(aq) = K+

KTiO3(s) + 3H2O = 2K+

KOH(s) = K+

KNO3(aq) = K+

**Table 2.** Reaction equilibria in the Ba-Ti-H2O hydrothermal system using TiO2, Ba(CH3COO)2 and/or Ba(NO3)2 as

Ba(CH3COO)3 2- = Ba2+ + 3(CH3COO)- Ba(CH3COO)2(s) = Ba2+ + 2(CH3COO)-

The iterative practice of the model using different A and B site feedstock concentrations, in conjunction with the addition of pH adjusting agents (mineralizer or solvent), allows to

assuming that inner pressure in the hydrothermal vessels is autogenously controlled.

estimate diagrams that correlates product yield and crystalline phase stability.

(*mT <sup>i</sup>*

H2O = H+

H2O(g) = H+

TiO2(s) + OH-

BaO(s) + 2H+

HCO3- = H+

CO2(g) = CO2(aq) CO2(aq) + H2O = H+

+ OH-

Ti4+ + H2O = TiOH3+ + H+ TiOH3+ + H2O = Ti(OH)2 2+ + H+ Ti(OH)2 2+ + H2O = Ti(OH)3+ + H+ Ti(OH)3+ + H2O = Ti(OH)4(aq) + H+ Ti(OH)4(aq) = TiO2(s) + 2H2O TiO2(rutile) + 2H2O = Ti(OH)4(aq) TiO2(anatase) + 2H2O = Ti(OH)4(aq) Ba(OH)2(s) = Ba2+ + OH-

+ OH-

= TiOH3+ + H+

= Ba2+ + H2O

 + CO3 2- BaCO3(s) = Ba2+ + CO3 2- BaHCO3+ = Ba2+ + HCO3 - BaTiO3(s) + H2O = Ba2+ + 2OH-

Ba2TiO4(s) + 2H2O = 2Ba2+ + 4OH-

precursor feedstock [38,40].

+ HCO3-

+ TiO2(s)

+ TiO2(s)


43


**Table 3.** Relevant species in the Ba-Ti hydrothermal systems and their standard state properties at 25°C (298.15 K).

## *2.2.2. Stability diagrams calculated from the thermodynamic model*

The stability diagrams have been developed for a vast number of hydrothermal systems including those research works related to the preparation of double component oxides. The stability diagram of perovskite-structured compounds, single-phase stoichiometry ABX3 [41– 47] and some relevant solid solutions [48, 49] are the most investigated applying thermody‐ namic modelling under hydrothermal conditions. Stability diagrams supply pertinent information regarding the ranges of equilibrium conditions at which several aqueous and/or solid species are stable in the hydrothermal reaction system. The molality of the aqueous ionic precursor includes the sum of equilibrium concentrations corresponding to dissolved ionic species in the hydrothermal media, but this term does not take into account the ionic species that precipitate from the solution. The stability diagrams also provide details related to the optimum conditions enabling incipient crystallization of the solid phases involved in the reaction. This process occurs when the ionic constituents that form the solid phase reach the supersaturation conditions in the hydrothermal fluid promoting its nucleation. However, the determination of the experimental reaction conditions required for an assumed yield of a perovskite reaction product from these diagrams is not straightforward.

The diagrams are constituted for two fields, as can be seen in Figure 3a: the solid lines indicate the boundary of incipient solid crystallization while the dashed lines denote the location where two ionic species in the liquid have equal concentrations. The stability phase diagrams are usually calculated solving the equilibrium and balance equations for the compositions of starting feedstock that accounts for the total Ba (*mBaT* ) as function of pH or Ti (*mT <sup>i</sup> T* ) as function of pH, respectively. Typical examples of these diagrams calculated for the Ba-Ti-H2O system were produced at two different temperatures, 25 and 90°C. The latter temperature has been empirically determined to correspond with optimum for the hydrothermal processing of perovskite particles of BaTiO3(s) [40]. In this typical example, the stability diagrams were calculated at 25°C (lines 1) and 90°C (line 2) by using an ideal approximation, the approach

Synthesis of Perovskite Oxides by Hydrothermal Processing – From Thermodynamic Modelling to Practical... http://dx.doi.org/10.5772/61568 45

**Solid species**

/(kJmol-1) -2779.9 -2132.9 -890.70 -883.27

/(kJmol-1) -3328.4 -2243.0 -946.01 -938.92

/(Jmol-1K-1) 422.6 196.6 50.3 49.9

 /(Jmol-1K-1) - -152.6 55.1 -55.3 *a* /(Jmol-1K-1) - 179.9 62.85 75.04 103 *b*/(Jmol-1K-2) - 6.694 11.38 0.0 10-5 *c*/(Jmol-1K) - -29.12 -9.897 -17.63

mol-1) 144.7 - 18.82 20.52 lit. 14, 45 15, 41 16 15, 42, 44

*2.2.2. Stability diagrams calculated from the thermodynamic model*

44 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

perovskite reaction product from these diagrams is not straightforward.

starting feedstock that accounts for the total Ba (*mBaT*

**Table 3.** Relevant species in the Ba-Ti hydrothermal systems and their standard state properties at 25°C (298.15 K).

The stability diagrams have been developed for a vast number of hydrothermal systems including those research works related to the preparation of double component oxides. The stability diagram of perovskite-structured compounds, single-phase stoichiometry ABX3 [41– 47] and some relevant solid solutions [48, 49] are the most investigated applying thermody‐ namic modelling under hydrothermal conditions. Stability diagrams supply pertinent information regarding the ranges of equilibrium conditions at which several aqueous and/or solid species are stable in the hydrothermal reaction system. The molality of the aqueous ionic precursor includes the sum of equilibrium concentrations corresponding to dissolved ionic species in the hydrothermal media, but this term does not take into account the ionic species that precipitate from the solution. The stability diagrams also provide details related to the optimum conditions enabling incipient crystallization of the solid phases involved in the reaction. This process occurs when the ionic constituents that form the solid phase reach the supersaturation conditions in the hydrothermal fluid promoting its nucleation. However, the determination of the experimental reaction conditions required for an assumed yield of a

The diagrams are constituted for two fields, as can be seen in Figure 3a: the solid lines indicate the boundary of incipient solid crystallization while the dashed lines denote the location where two ionic species in the liquid have equal concentrations. The stability phase diagrams are usually calculated solving the equilibrium and balance equations for the compositions of

of pH, respectively. Typical examples of these diagrams calculated for the Ba-Ti-H2O system were produced at two different temperatures, 25 and 90°C. The latter temperature has been empirically determined to correspond with optimum for the hydrothermal processing of perovskite particles of BaTiO3(s) [40]. In this typical example, the stability diagrams were calculated at 25°C (lines 1) and 90°C (line 2) by using an ideal approximation, the approach

) as function of pH or Ti (*mT <sup>i</sup>*

*T*

) as function

Δ*Gf* 0

Δ*Hf* 0

*S* 0

*Cp* 0

106 *V0* /(m3 *Ba*(*OH* )2•8*H*2*O Ba*2*TiO*<sup>4</sup> *TiO*2(*rutile*) *TiO*2(*anatase*)

**Figure 3.** Ba-Ti hydrothermal system stability diagrams calculated (a) using an ideal-solution approximation at 25°C (1) and 90°C (2); the solid and dashed lines denote the results calculated using data from Barin and Naumove, Refs. 15 and 37 in Ref. [40], respectively. And (b) at 25°C using modelled activity coefficients; state standard data were taken from Barin; the dashed line corresponds to the ideal-solution results taken from Lenka and Riman [40].

assumes that all the activity coefficients of the reaction are equal to one (Figure 3a). The variation in the boundaries of stability for the phases contained in the diagram are due to the data set of standard state properties employed to solve the equilibrium of the related species. This particular diagram illustrates the effect of deviation with respect to the ideal-solution boundaries, which were calculated using data for standard state properties from two different sources. Differences on the standard data might markedly shift the boundaries between the stability regions, as is shown in Figure 3a.

Figure 3b gives the results of the complete thermodynamic modelling conducted at 25°C, in which solid lines represent the boundaries of ionic and solid species. For comparison purposes, this diagram includes the boundary (dashed line) calculated with the aid of the ideal-solution approximation. It deserves to be emphasized that significant differences existing between the phase limits calculated using both complete and simplified (ideal solution) models. However, for the ideal-solution model, the bound between ionic species of Ba2+ and solid BaTiO3(s), as well as between ionic Ba2+–BaOH+ , and ionic BaOH+ –solid BaTiO3(s), are well defined. In contrast, the phase bound determined from the complete model do not follow straight-line functions. The curvature trend is significantly marked at higher concentrations of the aqueous species (Ba (*mBaT* ) >10–4). Furthermore, when the solution non-ideality is considered to carry out the modelling, the bounds shifted towards higher pH values. In particular, the phase boundary between Ba2+–BaOH+ is shifted by approximately 2 pH units, while that for Ba2+– BaTiO3(s) is changed approximately in 1 pH unit. This behaviour also occurs at higher temper‐ atures between 100 and 200°C.

The stability diagrams do not supply information related to the experimental conditions required to promote the total reaction of the precursors. Assuming that under certain hydro‐ thermal conditions, a total 100% yield of the crystallized reaction product must be produced. Analogous yield diagrams can establish a more practical insight of the hydrothermal process‐ ing. This tool has a great potential for establishing the optimum conditions required to achieve a total consumption of the precursors to produce high yields of the desired compound (e.g. ABX3). Therefore, the yield analysis takes into account the initial input concentration of the precursors, because the equilibrium concentration of species in a saturated solution is of primary concern for the hydrothermal process, because the feedstock must be transformed into a phase-pure product. The yield diagrams consist of sections where a specific amount of the desired product crystallizes involving the total reaction of the precursors. In other words, where the yield product is at least equal to an assumed value, namely 99%, at the stability boundary, the product yield is very small, because only an incipient nucleation of the product occurs at this point. The yield increases as the hydrothermal reaction process is driven beyond the solubility curve into the solid–liquid region. Hence, the product yield is determined by dividing the number of moles of the product by the total number of moles corresponding to the input metal precursors [38, 41–43].

Complex phase diagrams derived by thermodynamic modelling can be developed following the theory and methodology described above. These diagrams include both stability and yield fields and are calculated using the ideal solution and complete models. However, single stability and yield diagrams are portrayed against the solution pH, because this is an inde‐ pendent variable associated with the hydrothermal processing. The concentration of a pHadjusting solution required to achieve the crystallization of the desired product is not indicated in the diagram. One typical example was derived for the Sr–Zr system, and it is shown in Figure 4. The stability and yield diagrams were calculated for the preparation of SrZrO3 employing strontium hydroxide and zirconium oxide as precursors at 200°C. The diagram was determined at a fixed mixing Sr/Zr ratio of 1 which is plotted as a variable in the y coordinate axis, while the molality of the alkaline (KOH) is portrayed in the x coordinate axis. The solid line in the diagram (Figure 4) indicates the starting point of the crystallization of the SrZrO3 perovskite phase, although SrZrO3 and ZrO2 might coexist in the field above the solid line as the reaction tends to move closer to the shaded area. The remaining content of ZrO2 is gradually consumed leading to an increased amount of SrZrO3 produced. The shaded area is related to the optimum conditions, in terms of precursor feedstock: molality of Sr(OH)2 and KOH required to obtain the pure-phase SrZrO3 at a yield above 99% at the treatment temperature of 200°C [40–46].

The thermodynamic model of heterogeneous aqueous ionic liquid is a practical tool to prepare stability and yield diagrams. This powerful tool capable of estimating the effect of various processing conditions is likely towards to conduct the crystallization of smart perovskite ceramic materials in a more cost-effective way. In particular, it is possible to determine adequate conditions such as reaction temperature, pH, input precursor concentration and mixing ratios. The theoretical predictions allow researchers to formulate processing guidelines for finding optimum synthetic conditions to produce perovskite-type multicomponent compounds.

The stability diagrams do not supply information related to the experimental conditions required to promote the total reaction of the precursors. Assuming that under certain hydro‐ thermal conditions, a total 100% yield of the crystallized reaction product must be produced. Analogous yield diagrams can establish a more practical insight of the hydrothermal process‐ ing. This tool has a great potential for establishing the optimum conditions required to achieve a total consumption of the precursors to produce high yields of the desired compound (e.g. ABX3). Therefore, the yield analysis takes into account the initial input concentration of the precursors, because the equilibrium concentration of species in a saturated solution is of primary concern for the hydrothermal process, because the feedstock must be transformed into a phase-pure product. The yield diagrams consist of sections where a specific amount of the desired product crystallizes involving the total reaction of the precursors. In other words, where the yield product is at least equal to an assumed value, namely 99%, at the stability boundary, the product yield is very small, because only an incipient nucleation of the product occurs at this point. The yield increases as the hydrothermal reaction process is driven beyond the solubility curve into the solid–liquid region. Hence, the product yield is determined by dividing the number of moles of the product by the total number of moles corresponding to

46 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Complex phase diagrams derived by thermodynamic modelling can be developed following the theory and methodology described above. These diagrams include both stability and yield fields and are calculated using the ideal solution and complete models. However, single stability and yield diagrams are portrayed against the solution pH, because this is an inde‐ pendent variable associated with the hydrothermal processing. The concentration of a pHadjusting solution required to achieve the crystallization of the desired product is not indicated in the diagram. One typical example was derived for the Sr–Zr system, and it is shown in Figure 4. The stability and yield diagrams were calculated for the preparation of SrZrO3 employing strontium hydroxide and zirconium oxide as precursors at 200°C. The diagram was determined at a fixed mixing Sr/Zr ratio of 1 which is plotted as a variable in the y coordinate axis, while the molality of the alkaline (KOH) is portrayed in the x coordinate axis. The solid line in the diagram (Figure 4) indicates the starting point of the crystallization of the SrZrO3 perovskite phase, although SrZrO3 and ZrO2 might coexist in the field above the solid line as the reaction tends to move closer to the shaded area. The remaining content of ZrO2 is gradually consumed leading to an increased amount of SrZrO3 produced. The shaded area is related to the optimum conditions, in terms of precursor feedstock: molality of Sr(OH)2 and KOH required to obtain the pure-phase SrZrO3 at a yield above 99% at the treatment temperature

The thermodynamic model of heterogeneous aqueous ionic liquid is a practical tool to prepare stability and yield diagrams. This powerful tool capable of estimating the effect of various processing conditions is likely towards to conduct the crystallization of smart perovskite ceramic materials in a more cost-effective way. In particular, it is possible to determine adequate conditions such as reaction temperature, pH, input precursor concentration and mixing ratios. The theoretical predictions allow researchers to formulate processing guidelines

the input metal precursors [38, 41–43].

of 200°C [40–46].

**Figure 4.** Sr-Zr system phase stability and yield diagram calculated as a function of the solvent content ([KOH]) at 200°C for a precursor mixing ratio Sr/Zr = 1; the precursor source of strontium selected was Sr(OH)2 [40].

## **2.3. New model based on kinetic precipitation and balance population approaches**

## *2.3.1. Hydrothermal crystallization of ABX3 based on kinetic and population balance equation*

New models based on kinetic and population balance equations have received a tremendous attention. In the past decade, academic and industrial communities have devoted efforts to propose the models and evaluate their functionality to estimate the factors affecting the particle formation in a wide variety of processes. Additionally, the control of particle size of the final particles prepared via hydrothermal crystallization constitutes a potential advantage in comparison with other synthesis processes. Under hydrothermal conditions, the particle size distribution that grows from a solution strongly depends on the nuclei formation rates and subsequent particle growth. At an accelerated nucleation rate, the total number of particles produced is large while their size is relatively small. In a particular hydrothermal reaction system, the rates of nucleation and growth depend on supersaturation. Processing parameters such as precursor feedstock concentrations, temperature, and mixing conditions typically influence the supersaturation. Therefore, the final particle size distribution can be tailored by adjusting the hydrothermal synthesis conditions.

In the past decade, various efforts were conducted in order to present different ap‐ proaches accounting for the phenomena related with the particle formation of perovskitestructured compounds. Testino et al. have proposed the first meticulous kinetic approach devoted to the formation of BaTiO3 particles, the particle formation is based on a solution precipitation reaction [50–51]. The new model derived for the crystallization of BaTiO3 particles in an aqueous system is based on equations derived from basic nucleation, growth and aggregation fundamentals. This approach does not take into account empirical relationships such as those assumed in the thermodynamic equilibrium models. In contrast, the new model is based on reaction kinetics that correlates a mass balance coupled with a population balance. The population balance concerns the evaluation of some entities in a reaction system, i.e. solid particles or, events that might dictate the particle crystallization behaviour in the hydrothermal system under study. The performance of individual entities depends on variables associated with an appropriate reaction environment; consequently, the population balance equation might be correlated with balance equations that include the environmental variables [50, 52, 53].

The theoretical framework of the kinetic model was conceived by the exhaustive preliminary investigation of the aspects associated with the formation of BaTiO3 particles in an aqueous system [51]. The concentration of barium in the hydrothermal system promotes a variation in the rate of formation of BaTiO3 particles. When the content of Ba2+ increases, it decreases the particle size of the crystallized particles, and it also affects the chemical reaction when the temperature is varied; this effect has not been clarified. The model was developed taking into consideration the formation kinetics of perovskite BaTiO3 particles from dilute solutions of BaCl2 and TiCl4 (≤0.1 M) at a value of pH = 14 between 80–90°C. The overall chemical reaction experimentally studied is as follows:

$$\text{BaCl}\_{2(aq)} + \text{TiCl}\_{4(aq)} + \text{6NaOH}\_{(aq)} \rightarrow \text{BaTiO}\_{3(s)} + \text{6NaCl}\_{(aq)} + \text{3H}\_2\text{O}\_{(l)}\tag{7}$$

where (aq) denotes the hydrolyzed salt in water. The new kinetic model that is applicable to the preparation of different perovskite compounds was correlated with experimental data collected at various alkaline media concentration, temperature and Ba/Ti ratio. The prelimi‐ nary experimental analysis allowed the researchers to establish the stages and details of the mechanism associated with the nucleation, growth and particle aggregation [50, 51]. In this reaction system, the influence of non-ideality of the aqueous solution, net particle superficial area and thermodynamic properties were not considered.

In agreement with the analysis conducted by Testino et al., chemical reaction 7 occurs in two steps. (i) In the first step, the formation of a titanium hydroxide gel (THG) phase proceeds rapidly. Consequently, (ii) a slower reaction takes place between the THG phase and the Ba2+ ions dissolved in solution. When the supersaturation state of solute has reached under the second step of the reaction, the BaTiO3 nuclei spontaneously precipitate. The BaTiO3 crystal‐ lization mechanism controls the overall kinetics involved in the hydrothermal synthesis process. The kinetic study of reaction 7 determined that the evolution resembles sigmoidal curve behaviour, and the particle growth progress observations strongly support a nucleation and growth mechanism occurring in a second step. The global reaction proceeds via a twostep reaction mechanism, which achieves the formation of BaTiO3 via solution precipitation in mild alkaline conditions [50, 51]. Furthermore, the global mechanism also applies to the hydrothermal synthesis of barium titanate at temperatures above 100°C. When the THG suspension obtained by adding an alkaline media (NaOH or KOH) to the aqueous solution containing both BaCl2 and TiCl4 is used as the precursor; the same process is likely to occur when amorphous titanium hydroxide is produced by hydrolyzing a titanium (TiO2) precursor in a Ba(OH)2 solution. Hence, the global reaction involving both nucleation and growth of the perovskite is given by the following reaction

$$\text{Ti(OH)}\_{4(aq)} + \text{Ba}^{2+} \_{(aq)} + \text{OH}^- \_{(aq)} \rightarrow \text{BaTiO}\_{3(s)} + \text{3H}\_2\text{O}\_{(l)} \tag{8}$$

It can be assumed that local equilibrium conditions are carried on at the surface of the growing crystal, involving the equilibrium between the Ba2+ and BaOH– ionic species, which are correlated by Eq. (9)

$$\text{Ba}^{2+}\_{\text{(aq)}} + \text{OH}^-\_{\text{(aq)}} \rightarrow \text{BaOH}^-\_{\text{(aq)}} \tag{9}$$

The supersaturation state associated with reaction 8 is interpreted as

In the past decade, various efforts were conducted in order to present different ap‐ proaches accounting for the phenomena related with the particle formation of perovskitestructured compounds. Testino et al. have proposed the first meticulous kinetic approach devoted to the formation of BaTiO3 particles, the particle formation is based on a solution precipitation reaction [50–51]. The new model derived for the crystallization of BaTiO3 particles in an aqueous system is based on equations derived from basic nucleation, growth and aggregation fundamentals. This approach does not take into account empirical relationships such as those assumed in the thermodynamic equilibrium models. In contrast, the new model is based on reaction kinetics that correlates a mass balance coupled with a population balance. The population balance concerns the evaluation of some entities in a reaction system, i.e. solid particles or, events that might dictate the particle crystallization behaviour in the hydrothermal system under study. The performance of individual entities depends on variables associated with an appropriate reaction environment; consequently, the population balance equation might be correlated with balance equations that include

The theoretical framework of the kinetic model was conceived by the exhaustive preliminary investigation of the aspects associated with the formation of BaTiO3 particles in an aqueous system [51]. The concentration of barium in the hydrothermal system promotes a variation in the rate of formation of BaTiO3 particles. When the content of Ba2+ increases, it decreases the particle size of the crystallized particles, and it also affects the chemical reaction when the temperature is varied; this effect has not been clarified. The model was developed taking into consideration the formation kinetics of perovskite BaTiO3 particles from dilute solutions of BaCl2 and TiCl4 (≤0.1 M) at a value of pH = 14 between 80–90°C. The overall chemical reaction

2() () <sup>4</sup> ( ) 3( ) ( ) <sup>2</sup> ( ) 6 6 3 *aq aq aq <sup>s</sup> aq <sup>l</sup> BaCl TiCl NaOH BaTiO NaCl H O* + + +® + (7)

where (aq) denotes the hydrolyzed salt in water. The new kinetic model that is applicable to the preparation of different perovskite compounds was correlated with experimental data collected at various alkaline media concentration, temperature and Ba/Ti ratio. The prelimi‐ nary experimental analysis allowed the researchers to establish the stages and details of the mechanism associated with the nucleation, growth and particle aggregation [50, 51]. In this reaction system, the influence of non-ideality of the aqueous solution, net particle superficial

In agreement with the analysis conducted by Testino et al., chemical reaction 7 occurs in two steps. (i) In the first step, the formation of a titanium hydroxide gel (THG) phase proceeds rapidly. Consequently, (ii) a slower reaction takes place between the THG phase and the Ba2+ ions dissolved in solution. When the supersaturation state of solute has reached under the second step of the reaction, the BaTiO3 nuclei spontaneously precipitate. The BaTiO3 crystal‐ lization mechanism controls the overall kinetics involved in the hydrothermal synthesis process. The kinetic study of reaction 7 determined that the evolution resembles sigmoidal

the environmental variables [50, 52, 53].

48 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

experimentally studied is as follows:

area and thermodynamic properties were not considered.

$$S = \frac{\left[\text{Ti}\left(\text{OH}\right)\_4\right]\left[\text{Ba}^{2+}\right]\left[\text{OH}^-\right]^2}{K\_s} \tag{10}$$

The supersaturation solution grade is independent of the solvent species involved in the precipitation reaction; therefore, the equilibrium conditions are kept at the solid/liquid interface for an extended period. The Ti(OH)4 species are provided by the THG precursor phase dissolution. A global mechanism that proceeds in two steps conducts this process. At the first step, the disruption of the polymeric network of Ti(OH)4 takes place by the nucleophilic attack of the OH– ions occurring at Ti atom positions (denoted as Ti\*, in Eq. 11). The second step of reaction comprises the reaction between oxygen ions (O\*) produced during the decomposition of Ti(OH)4 with water molecules, restoring the hydroxyl ions. The dissolution rate, which is reversible and rate controlled at the first step, can be expressed by

$$\frac{d\left\lfloor \begin{array}{c} Ti\_{gd} \end{array} \right\rfloor}{dt} = k\_{\text{i}} \left\lceil Ti^{\*} \right\rceil \left\lceil OH^{-} \right\rceil - k\_{-\text{i}} \left\lceil O^{\*} \right\rceil \left\lceil Ti \left( OH \right)\_{4} \right\rceil \tag{11}$$

where the term [Tigel] corresponds to the number of moles of titanium contained in the gel precursor in the reacting suspension. Additionally, *k1* and *k-1* are correlated to the kinetic constants of the direct and reverse process, respectively. When [Ti\*] and [O\*] are proportional to [Tigel], Eq. (8) can be rewritten as

$$-\frac{d\left[\overline{\boldsymbol{T}}\overline{\boldsymbol{I}}\_{\rm gel}\right]}{dt} = k\_{\rm i}\left[\overline{\boldsymbol{T}}\overline{\boldsymbol{I}}\_{\rm gel}\right]\left[\boldsymbol{\Theta}\boldsymbol{H}^{-}\right]\left\{\boldsymbol{1} - \frac{\boldsymbol{K}}{\boldsymbol{K}\_{\rm s}}\left[\overline{\boldsymbol{T}}\overline{\boldsymbol{\Omega}}\boldsymbol{\Theta}\boldsymbol{H}\right]\_{4}\right\}\tag{12}$$

where *k* is the rate constant for THG dissolution, and *K* is the reciprocal of the equilibrium constant overall reaction

$$\text{TiO}(\text{OH})\_{2(\text{gal})} + \text{Ba}^{2+} \_{\text{(aq)}} + \text{2OH}^{-} \_{\text{(aq)}} \rightarrow \text{BaTiO}\_{3(s)} + 2H\_2O \_{(l)} \tag{13}$$

This general approach likely describes the mechanism applied to the classical hydrothermal process, in which crystalline titanium oxide particles (TiO2) are suspended in aqueous Ba(OH)2. In this reaction system, the net surface area of the titanium oxide particles affects Ti4+ ionic species produced via the dissolution of TiO2 powder and its dissolution rate.

### *2.3.2. Mass balance expression*

The complete model expression is constituted by two additional terms; the first one takes into account a global mass balance of the ionic species involved in the formation of the ABX3 single phase. The mathematical expression that involves the mass balance of one of the precursors, i.e. titanium ions, in the hydrothermal system can be written as [50]

$$-\frac{d\left[\overline{\boldsymbol{T}}\overline{\boldsymbol{i}}\_{\boldsymbol{aq}}\right]}{dt} = \frac{d\left[\overline{\boldsymbol{T}}\overline{\boldsymbol{i}}\_{\boldsymbol{BT}}\right]}{dt} + \frac{d\left[\overline{\boldsymbol{T}}\overline{\boldsymbol{i}}\_{\boldsymbol{gcl}}\right]}{dt} \tag{14}$$

In this differential equation, the term [Tiaq] indicates the concentration of the aqueous titanium species present in the reaction system, which typically is assigned to the [Ti(OH)4]. Likewise, the content of the precipitated BaTiO3 at a certain time of reaction is [TiBT]. Thus, the first term in Eq. (14) is correlated with the BaTiO3 particle formation rate promoted by nucleation and growth events.

$$\frac{d\left\lfloor T\hat{\imath}\_{\rm BT}\right\rfloor}{dt} = \left\{\hat{B} + \hat{G}\right\}\tag{15}$$

Eq. (16) corresponds to the global equation that combines the nucleation and growth with the dissolution rate of the THG given by Eq. (12). This expression involves the crystallization rate of BaTiO3 by nucleation and growth events coupled with the simultaneous consumption of the THG phase, which is analogous to the yield product term in the thermodynamic models. If the Ba/Ti molar ration in the THG is denoted by *R*gel, the mass balance can be given by Eq. (17), in which [Baaq] is the absolute concentration of Ba ions corresponding to the sum of [Ba2+] and [BaOH+ ] species.

Synthesis of Perovskite Oxides by Hydrothermal Processing – From Thermodynamic Modelling to Practical... http://dx.doi.org/10.5772/61568 51

$$\frac{d\left[\overline{T}\overline{\mathbf{I}}\_{\text{eq}}\right]}{dt} = \mathbf{k}\left[\overline{T}\overline{\mathbf{I}}\_{\text{gs}}\right]\left[OH^{-}\right]\left(1 - \frac{\mathbf{K}}{K\_{\text{S}}}\left[\overline{T}\overline{\mathbf{u}}(OH)\_{4}\right]\right) - \left\{\hat{\mathcal{B}} + \hat{\mathcal{G}}\right\}\tag{16}$$

$$\frac{d\left[\left.B a\_{\rm aq}\right]}{dt}\right] = R\_{gcl}\frac{d\left[\left.T\dot{I}\_{gcl}\right]}{dt}\right] - \left\{\hat{B} + \hat{G}\right\}\tag{17}$$

## *2.3.3. Population balance*

<sup>1</sup> ( )<sup>4</sup> <sup>1</sup> *gel*

where *k* is the rate constant for THG dissolution, and *K* is the reciprocal of the equilibrium

3) ( ) ( ) ( ) <sup>2</sup> <sup>2</sup> 2 2

This general approach likely describes the mechanism applied to the classical hydrothermal process, in which crystalline titanium oxide particles (TiO2) are suspended in aqueous Ba(OH)2. In this reaction system, the net surface area of the titanium oxide particles affects Ti4+ ionic species produced via the dissolution of TiO2 powder and its dissolution rate.

The complete model expression is constituted by two additional terms; the first one takes into account a global mass balance of the ionic species involved in the formation of the ABX3 single phase. The mathematical expression that involves the mass balance of one of the precursors,

In this differential equation, the term [Tiaq] indicates the concentration of the aqueous titanium species present in the reaction system, which typically is assigned to the [Ti(OH)4]. Likewise, the content of the precipitated BaTiO3 at a certain time of reaction is [TiBT]. Thus, the first term in Eq. (14) is correlated with the BaTiO3 particle formation rate promoted by nucleation and

*aq BT gel d Ti d Ti d Ti dt dt dt*

> { }<sup>ˆ</sup> <sup>ˆ</sup> *BT d Ti B G*

Eq. (16) corresponds to the global equation that combines the nucleation and growth with the dissolution rate of the THG given by Eq. (12). This expression involves the crystallization rate of BaTiO3 by nucleation and growth events coupled with the simultaneous consumption of the THG phase, which is analogous to the yield product term in the thermodynamic models. If the Ba/Ti molar ration in the THG is denoted by *R*gel, the mass balance can be given by Eq. (17), in which [Baaq] is the absolute concentration of Ba ions corresponding to the sum of [Ba2+]

*dt* é ù -

*d Ti <sup>K</sup> k Ti OH Ti OH dt K*

ë û ì ü ï ï -= - éù é ù é ù í ý ëû ë û ë û ï ï î þ

*S*

*aq aq <sup>s</sup> <sup>l</sup> gel TiO OH Ba OH BaTiO H O* + - <sup>+</sup> ®+ <sup>+</sup> (13)

éù é ù é ù ëû ë û ë û -= + (14)

ë û = + (15)

(12)

*gel*

( ) ( ( ) <sup>2</sup>

i.e. titanium ions, in the hydrothermal system can be written as [50]

é ù

50 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

constant overall reaction

*2.3.2. Mass balance expression*

growth events.

and [BaOH+

] species.

The population balance equation (PBE) is related to net particle number continuity. It was the concept introduced to complete the kinetic model. The last term takes into account the evolution of the particle size distribution for a particular cluster of particles subjected to nucleation, growth, coalescence and disruption processes. According to Testino et al. [50], the expression can be given as

$$\frac{\partial \mathfrak{n}(L,t)}{\partial t} + \frac{\partial}{\partial L} \Big[ \mathbb{G}(L,t)\mathfrak{n}(L,t) \Big] = \mathcal{B}\left(n\right) - D\left(n\right) \tag{18}$$

In the PBE equation, *n*(*L,t*) corresponds to the density of particles at a particular time *t* in the particle size range *L* to *L*+ d*L*. While *G* represents the isotropic growth rate, *B* and *D* are the particle rates of genesis and extinction, respectively. The net particles number per unit volume is defined at an infinitesimal period as d*N*, and the density function is *n* = d*N*/d*L*. Equation 15 is related to the rate of particles accumulation in the range *L* to *L*+d*L*. This range is a function of the particle growth rate and specifically of the speed at which particles of that size are directly created or removed. In order to solve this equation, the kernel aggregation function coupling with the PBE and four particle size distribution moments were considered. Because these are the quantities used to solve PBE problems [50, 52–54].

The new kinetic model derived for BaTiO3 particles is likely applicable to a broad number of perovskite systems, following the methodology described above. In either case, a raw dried titanium hydroxide or Ti-gel formed as an intermediate product are used as a precursor reactant. The nucleation and growth theory for supersaturated solutions correlates the rates of nucleation and growth with some physical parameters such as temperature, the surface tension of the solid phase and diffusion coefficient. In the model, nucleation is considered to occur in two steps, in the initial stage a homogeneous nucleation proceeds in the supersatu‐ rated solution. After the conclusion of the preliminary step, a secondary nucleation is then enhanced promoted by the variation of solute saturation in the hydrothermal media. The secondary nucleation controls the creation of new embryos on the surface of BaTiO3 particles, which are already present in the suspension as a result of nucleation and growth events. The secondary nucleation promotes the rapid increase in the BaTiO3 formation rate after the initial period dominated by primary nucleation. Moreover, secondary nucleation leads the poly‐ crystalline nature of the final particles. Diffusion-controlled growth is the process involved in the coarsening of the elementary crystals that constitute the polycrystalline particles. The new algorithm has a potential for calculating the net number of events (moments) under particular reaction conditions. These events are of particular concern because these define the particles' aspect such as total number, dimension, surface and volume of particles per liquid suspension unit volume. The new model based on a PBE analysis yields an uncertain approximation of the real process, but hinders the introduction of completely empirical relationships between growth rate and supersaturation state.

Ingeneral,onelimitationofthekineticmodelis theuseofPBEsconsideringoneinternalvariable, i.e. the size of the primary crystallites. Consequently, a more meticulous approach proposed takes into account the use of PBEs with at least two internal coordinates, for example, the size of the primary particles and the size of the polycrystalline particles. New approaches might provide a better description by taking into account the interactions between the elementary crystallites affecting the overall precipitation process. However, the application of PBEs with more than one internal coordinate to precipitation problems is still at an early stage of develop‐ ment. Recently, Marchisio has presented a detailed mathematical model involving the precipitation of BaTiO3 in aqueous solution [54]. This particular model is based on a bivariate population balance equation. The new approach overcomes the limitations consid‐ ered for a mono-variate population balance equation, namely the crystallite size and particle size. The new improved model (not reported here) shows qualitative results similar to those previously discussed for the mono-variate model. However, quantitatively the modelling results are very different from the predictions obtained with the mono-variate PBE. This complete model can be used to identify new model parameters associated with the mecha‐ nism involved in the nucleation, growth and aggregation stages of BaTiO3 hydrothermal synthesis [54].

In the past decades, an increasing interest in the production of submicron powders with uniform particle size distribution (PSD) through solution precipitation methods, including the hydrothermal powder processing, has been the main concern in the chemical engineering field, particularly, the search for the adequate operating conditions for preparing perovskite nanoparticles, similar to those examples described in the following section. However, model‐ ling works on thermodynamics coupled with the kinetic aspects of particle formation and its evolution have also been conducted by various researcher groups in different areas [38–42, 50– 54]. The motivations for developing new optimized mathematical models are multiples. Indeed, the mathematical models described in the present review can be used to estimate the hydrothermal conditions for particle formation and the evolution of mechanisms associated with this process. Also, these can be used to set up the optimal operating conditions to prepare a powder with desired characteristics, following a product engineering philosophy. Finally, there is an enormous potential to employ both thermodynamic and kinetic models, by implementing them in computational fluid dynamics models, to scale up the precipitation processes from laboratory to an industrial scale. However, new researchers in the hydrother‐ mal field must bear in mind that some caution must be exercised, when applying the proposed models to diverse perovskite systems, because modelling results might vary depending on the reaction temperature and concentration of precursors. These factors might lead to erroneous conclusions associated with the reaction mechanism operating for a particular reaction system. Moreover, the models assume that during the nucleation and growth events, the hydrothermal system is perfectly mixed, leading to a precipitation rate being controlled by the chemical reaction kinetics. Although these inferences are acceptable in small laboratory vessels, these might not operate for larger industrial reactors, where mixing problems might frequently rise due to slow and less-efficient agitation systems. The precipitation rate is one of the problems that commonly meets in the scale-up of precipitation processes.

## **2.4. Hydrothermal processing of perovskite-structured ceramic oxides**

the coarsening of the elementary crystals that constitute the polycrystalline particles. The new algorithm has a potential for calculating the net number of events (moments) under particular reaction conditions. These events are of particular concern because these define the particles' aspect such as total number, dimension, surface and volume of particles per liquid suspension unit volume. The new model based on a PBE analysis yields an uncertain approximation of the real process, but hinders the introduction of completely empirical relationships between

Ingeneral,onelimitationofthekineticmodelis theuseofPBEsconsideringoneinternalvariable, i.e. the size of the primary crystallites. Consequently, a more meticulous approach proposed takes into account the use of PBEs with at least two internal coordinates, for example, the size of the primary particles and the size of the polycrystalline particles. New approaches might provide a better description by taking into account the interactions between the elementary crystallites affecting the overall precipitation process. However, the application of PBEs with more than one internal coordinate to precipitation problems is still at an early stage of develop‐ ment. Recently, Marchisio has presented a detailed mathematical model involving the precipitation of BaTiO3 in aqueous solution [54]. This particular model is based on a bivariate population balance equation. The new approach overcomes the limitations consid‐ ered for a mono-variate population balance equation, namely the crystallite size and particle size. The new improved model (not reported here) shows qualitative results similar to those previously discussed for the mono-variate model. However, quantitatively the modelling results are very different from the predictions obtained with the mono-variate PBE. This complete model can be used to identify new model parameters associated with the mecha‐ nism involved in the nucleation, growth and aggregation stages of BaTiO3 hydrothermal

In the past decades, an increasing interest in the production of submicron powders with uniform particle size distribution (PSD) through solution precipitation methods, including the hydrothermal powder processing, has been the main concern in the chemical engineering field, particularly, the search for the adequate operating conditions for preparing perovskite nanoparticles, similar to those examples described in the following section. However, model‐ ling works on thermodynamics coupled with the kinetic aspects of particle formation and its evolution have also been conducted by various researcher groups in different areas [38–42, 50– 54]. The motivations for developing new optimized mathematical models are multiples. Indeed, the mathematical models described in the present review can be used to estimate the hydrothermal conditions for particle formation and the evolution of mechanisms associated with this process. Also, these can be used to set up the optimal operating conditions to prepare a powder with desired characteristics, following a product engineering philosophy. Finally, there is an enormous potential to employ both thermodynamic and kinetic models, by implementing them in computational fluid dynamics models, to scale up the precipitation processes from laboratory to an industrial scale. However, new researchers in the hydrother‐ mal field must bear in mind that some caution must be exercised, when applying the proposed models to diverse perovskite systems, because modelling results might vary depending on the reaction temperature and concentration of precursors. These factors might lead to erroneous conclusions associated with the reaction mechanism operating for a particular reaction system.

growth rate and supersaturation state.

52 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

synthesis [54].

According to the literature survey, the synthesis under hydrothermal conditions of different types of perovskite compounds was triggered during the 20th century. Nowadays, this synthesis method constitutes an important tool for materials processing. Because hydrother‐ mal synthesis offers some processing advantages, which enhances the preparation of mono‐ dispersed nanoparticles with controlled size and morphology. In the present review, the experimental details for the synthesis of a broad number of perovskites are addressed based on their compositional aspects, namely single substitution in either A or B sites. Some double perovskites and their related solid solutions are also included.

The synthesis of single-phase stoichiometric perovskite ABX3 has been possible under hydrothermal conditions by the employment of a broad number of atomic elements. Among the elements incorporated at the A site of the perovskite structure are the elements of IA (K, Ag, Na) and IIA (Ca, Sr, Ba) groups of the periodic table. Some bivalent elements (Bi, Pb, Rb) have been incorporated in this site as well. Experimental results demonstrated that trivalent rare earth elements, namely La, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu or Y, are able to be located at the A site, while the B site could be occupied by stable transition metals elements (Ti, Fe, Zr, Ta, Nb, Bi, Sn, Co, Ru, Cr and Mn). The next sections of the present review give an account of the main experimental aspects related to the hydrothermal processing of perovskitestructured ceramic oxides. Particular emphasis is placed on the practical aspects associated with the selection of mineralizers and their concentrations, solution pH, choice of precursors, and reaction intervals and temperatures. This review intends to present a more realistic and practical state-of-the-art on the optimization of the synthesis of the major perovskite-struc‐ tured compounds that have been extensively produced via hydrothermal processing. It deserves emphasis that succeeding on the crystallization of pure single phase perovskite oxides under hydrothermal conditions depends on major functional parameters such as the nature of precursors and mineralizer. The main aspects related to the hydrothermal processing of perovskites are discussed in terms of the type of the site substitution, focusing only on singlephase perovskite oxides. Starting from the synthesis of major A-site-ordered perovskitestructured oxides (ATiO3, AFeO3, AZrO3, etc.) and a few solid solutions, and subsequently proceeding to the synthesis of B-site-ordered perovskites and, finally, to the synthesis of A and/ B-site-ordered perovskite oxides.

## *2.4.1. Synthesis of A-site-ordered perovskite titanates (ATiO3)*

In recent years, much progress has been made in the synthesis of A-site-ordered perovskite titanates (ATiO3; A= Ba, Pb, Ca and Sr; hereafter the compounds containing the latter cations are referred as BT, PT, CT and ST, respectively) under hydrothermal conditions. In the synthesis of BT particles, various authors have employed different methodologies to prepare powders with perovskite structure under hydrothermal conditions. These methods, in general, enable us to synthesize this kind of perovskite oxides in various aqueous alkaline solutions using various mineralizers. One experimental procedure developed for the synthesis of BT powders use Ba(OH)2 as the precursor of barium and pH adjustment agent, instead of using a strong alkaline mineralizer (KOH or NaOH). The BT particles synthesized at a low temper‐ ature of 80°C in a solution with a pH of 8 exhibited a spherical morphology and cubic ABX3 structure. The chemical reaction promoted under hydrothermal conditions involved the use of chemically modified Ti-peroxo-hydroxide and Ba(OH)2 precursors. The Ti-peroxo-complex precursor was rapidly dissolved in the reaction media (Ba(OH)2), subsequently enhancing the crystallization of BT particles, providing fast and mild powder synthesis conditions [55].

The crystallization of BT particles was found to be carried out with similar morphologic and structural aspects when BaCl2 is employed as a precursor and using NaOH as mineralizer [56]. The optimal concentration of the NaOH mineralizer to conduct the synthesis was 6 M; it allowed to obtain the BT perovskite single phase between the range of 120 and 200°C for reaction periods of 1–4 h. Reaction kinetics influences the crystallization of BT at low temper‐ atures and short reaction times. However, the thermal equilibrium plays a crucial role at elevated temperatures (200°C) and/or prolonged reaction times (4 h) that control the particle crystallization. In other study conducted under similar conditions of synthesis [57], different metal titanates (ATiO3, A= Ba, Sr, Ca), among them BT particles, were also prepared. For this case, the reaction synthesis was developed at highly alkaline conditions with a value of pH > 12. The excess amount of NaOH provided an alkaline environment for the dissolution of TiO2 and its subsequent reaction to produce BT. They found that the formed Ti-OH complexes reacted at 160°C for 72 h with the Ba2+ ions in the solution to form the single-phase BT perovskite oxide.

Other methodologies have been employed to prepare BT powders exhibiting morphologies and particle size, using KOH as a mineralizer. Recently, nanocrystalline particles with a cubic shape of BT were prepared using organic compounds; these promoted the control of the particle size during the crystallization step. In this case, the hydrothermal media consisted in a KOH solution prepared with 11 mmol of reagent (2.2 M) [58]. In contrast, different perovskitestructured BT oxides were prepared from altered titanium isopropoxide and barium acetate salts controlling the pH of the suspension adding a 2.0 M KOH solution [59]. Under these conditions, phase-pure perovskite BT powders were successfully synthetized at 150°C for 18 h in an extremely alkaline feedstock media. The pH of the suspension was 13; it was concluded from this study that a strong alkaline solvent is necessary for the production of pure-phase perovskites under hydrothermal conditions. In a subsequent study, the effect of varying the molar [KOH]/[Ti-isopropoxide] ratio during the reaction synthesis was investigated. The principal aim was to determine the impact of this processing factor in the crystallinity and particle size of the BT particles [60]. In this particular reaction system, the nucleation and growth processes are affected by the electric charge micelles produced on the surface of the BT particles during the hydrothermal process. The increment in OH– concentration caused by dissolution of KOH in the reaction medium improves the tendency to BT particles separation, because the generation of electrostatic repulsion forces acting in-between negatively charged surfaces hinders the formation of BT agglomerates [60].

synthesis of BT particles, various authors have employed different methodologies to prepare powders with perovskite structure under hydrothermal conditions. These methods, in general, enable us to synthesize this kind of perovskite oxides in various aqueous alkaline solutions using various mineralizers. One experimental procedure developed for the synthesis of BT powders use Ba(OH)2 as the precursor of barium and pH adjustment agent, instead of using a strong alkaline mineralizer (KOH or NaOH). The BT particles synthesized at a low temper‐ ature of 80°C in a solution with a pH of 8 exhibited a spherical morphology and cubic ABX3 structure. The chemical reaction promoted under hydrothermal conditions involved the use of chemically modified Ti-peroxo-hydroxide and Ba(OH)2 precursors. The Ti-peroxo-complex precursor was rapidly dissolved in the reaction media (Ba(OH)2), subsequently enhancing the crystallization of BT particles, providing fast and mild powder synthesis conditions [55].

54 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The crystallization of BT particles was found to be carried out with similar morphologic and structural aspects when BaCl2 is employed as a precursor and using NaOH as mineralizer [56]. The optimal concentration of the NaOH mineralizer to conduct the synthesis was 6 M; it allowed to obtain the BT perovskite single phase between the range of 120 and 200°C for reaction periods of 1–4 h. Reaction kinetics influences the crystallization of BT at low temper‐ atures and short reaction times. However, the thermal equilibrium plays a crucial role at elevated temperatures (200°C) and/or prolonged reaction times (4 h) that control the particle crystallization. In other study conducted under similar conditions of synthesis [57], different metal titanates (ATiO3, A= Ba, Sr, Ca), among them BT particles, were also prepared. For this case, the reaction synthesis was developed at highly alkaline conditions with a value of pH > 12. The excess amount of NaOH provided an alkaline environment for the dissolution of TiO2 and its subsequent reaction to produce BT. They found that the formed Ti-OH complexes reacted at 160°C for 72 h with the Ba2+ ions in the solution to form the single-phase BT perovskite

Other methodologies have been employed to prepare BT powders exhibiting morphologies and particle size, using KOH as a mineralizer. Recently, nanocrystalline particles with a cubic shape of BT were prepared using organic compounds; these promoted the control of the particle size during the crystallization step. In this case, the hydrothermal media consisted in a KOH solution prepared with 11 mmol of reagent (2.2 M) [58]. In contrast, different perovskitestructured BT oxides were prepared from altered titanium isopropoxide and barium acetate salts controlling the pH of the suspension adding a 2.0 M KOH solution [59]. Under these conditions, phase-pure perovskite BT powders were successfully synthetized at 150°C for 18 h in an extremely alkaline feedstock media. The pH of the suspension was 13; it was concluded from this study that a strong alkaline solvent is necessary for the production of pure-phase perovskites under hydrothermal conditions. In a subsequent study, the effect of varying the molar [KOH]/[Ti-isopropoxide] ratio during the reaction synthesis was investigated. The principal aim was to determine the impact of this processing factor in the crystallinity and particle size of the BT particles [60]. In this particular reaction system, the nucleation and growth processes are affected by the electric charge micelles produced on the surface of the BT particles during the hydrothermal process. The increment in OH– concentration caused by dissolution of KOH in the reaction medium improves the tendency to BT particles separation,

oxide.

Later on, Xiao and co-workers [61] developed another methodology to synthesize BT powders with spherical morphology. In this approach was involved the usage of a surfactant agent such as ethylene glycol in a 1 M KOH solution. After 200°C for 12 h, the experimental results showed evidence of the formation reaction by-product, namely traces of BaCO3, which was removed by acid washing. The synthesis of BT with dendritic shapes without any surfactants was conducted using KOH solutions with various concentrations of 0.1, 0.3, 0.7 and 1 M, keeping other reaction conditions unchanged (200°C, 12 h). It was found that the KOH concentration was a crucial parameter that enhances the formation of the BT dendrites [62]. When the concentration of KOH increased, the particle morphology changed from dendritic to a spherelike shape. This phenomenon was explained based on the fact that dendritic growth tends to take place as the system is driven farther away from equilibrium. Therefore, when low KOH concentrations were used as solvent, the alkalinity of the media did not allow the system to reach the equilibrium conditions. This fact promoted the formation of the BT dendrites rather that the sphere-shaped particles [62].

Under microwave-hydrothermal conditions, the synthesis of BT powders was accelerated in comparison with conventional hydrothermal synthesis. Zhu et al. [63] satisfactorily synthe‐ sized perovskite BT nanowires in different reaction media using pure water and mixtures of water–ethylene glycol solutions. The pH value of the resultant mixture was adjusted to 14.0 adding 1.1217 M of KOH before the hydrothermal treatment. A substantial reduction of the reaction time was achieved using a highly concentrated alkaline solvent, allowing the synthesis of pure-phase BT particles to proceed for 50 min at 150°C. The alkalinity of the hydrothermal media had a significant influence on the crystal structure of the single-phase BT and affected the morphology of the BT powders produced by hydrothermal-microwave-assisted treat‐ ments [64]. The BT nanoparticles prepared under microwave-assisted sol-hydrothermal conditions exhibited a varied morphology accounting from elongated particles to a ring-like shape. The differences in particle morphology occurred by changing the KOH concentration from 0.25 to 5 M. The reaction involves the metal hydrous complex gel formation that is promoted by dissolution of the precursors, and the subsequent recrystallization when the solvent reaches the supersaturated state. The nuclei were rapidly transformed into BT nanoparticles because of the high pH and high reaction temperature. The nanoparticles that resulted from the high saturation of the reaction solution dissolved in the solution forming additional nuclei avoiding a marked particle growth [64].

On the other hand, the synthesis of PT particles has been extensively investigated under different hydrothermal conditions. Ohara et al. [65] conducted one of the pioneering research works for preparing PT perovskites. The optimum conditions for preparing pure PT singlephase fine fibers were 150°C for variable reaction times (24 to 72 h). The PT fibers were produced selecting a molar Pb/Ti ratio = 1, and Ti ions were supplied by adding potassium titanate 2K2O⋅11TiO2⋅3H2O as the precursor material. These authors also demonstrate that the hydrothermal technique provides the possibility to prepare PT fine powders with spherical morphology. The morphological particle transition from fibrous to spherical was obtained via an ion-exchange reaction between Pb2+ ions and K+ ions. This phenomenon occurs under hydrothermal conditions at high pH, 150°C, and pressure (0.4–0.5 MPa). The formation of the PT phase requires a Pb2+ concentration higher than that of K+ in the system. Likewise, by employing, additionally to the KOH concentration, other mineralizer salts, such as NaNO3, KNO3 and LiNO3 in the reaction media, it was possible to obtain tetragonal perovskite PT nanosheets [66]. The synthesis of pure perovskite PT with tetragonal structure was carried out in a suspension with KOH and NaNO3 concentrations of 1 and 4 M, respectively. According to these studies, it was found that the PT synthesis preferentially occurred under stoichiometric equilibrium conditions.

Miscellaneous research works were carried out to investigate the favourable conditions for synthesizing CT and single-phase RCT doped with rare earth elements (R = Eu) perovskites. These were compounds prepared using the soft chemical process microwave-hydrothermal method without the usage of organic surfactants [67, 68]. The synthesis of pure-phase ortho‐ rhombic CT and RCT was successfully carried out in a reaction media with a pH value of 14 using KOH as the mineralizer. The presence of the alkaline solution promoted the co-precip‐ itation of a complex hydroxide constituted by the TiO(OH)2-Ca(OH)2-Eu(OH)3 during the early stage of the reaction. During the formation of either CT or RCT compounds, the direct rotational water excitation that is achieved by microwave radiation produced the release of uncoupling OH– groups from the complex hydroxide. Consequently, calcium and Ti(OH)4 clusters rapidly interact themselves in the aqueous media as a result of the preliminary OH removal process. Therefore, at a pH ≥ 9, the heating promoted by microwave radiation achieved the CT and RCT nucleation, respectively. Under these conditions, the crystallization kinetics of both perovskite-structured compounds was accelerated one to two orders of magnitude. Thus, it was presumed that the diffusion of calcium and titanium clusters is higher under hydrothermal conditions than at ambient pressure and temperature. This phenomenon was explained to proceed by effective particle collisions producing irreversible oriented attachments that offer favourable thermodynamic and kinetics conditions for the crystalliza‐ tion of CT and RCT, in addition to the particle shape control.

Regarding the synthesis of ST perovskite, it was conducted in KOH solutions with concentra‐ tions as low as 0.1 M. The ST nuclei precipitated incorporating mild concentrated Sr(OH)2 solutions in the hydrothermal system, and many Sr2+ ions remained in solution after the crystallization process. However, a large amount of Sr(OH)2 remained when the KOH concentration was increased up to 1 M, and the mass transport rate of ions that feed the growing crystals was accelerated. Therefore, a variety of crystal surfaces have the opportunity to grow due to the relatively larger rate of ionic mass transport, which allowed the formation of cubiclike-shaped ST aggregated crystals [69]. Figure 5 gives the typical micrographs of ST perovskite powders hydrothermally prepared at 250°C for various reaction intervals [70]. Finally, the hydrothermal synthesis of some solid solutions was satisfactorily carried out under conditions similar to those mentioned above. Wei et al. [71] prepared pure-phase Pb0.70La0.30TiO3 fine powders with cubic structure by employing co-precipitated Pb-La-Ti-OHX hydroxide and KOH solutions with different concentrations of 2, 4 and 6 M at 220°C for 36 h. Additionally, in another set of experiments, the synthesis of Bi0.5Na0.5TiO3 (BNT) was found to occur in highly

concentrated alkaline hydrothermal media of 12 M NaOH. This concentration is the minimum required to achieve the crystallization of single-phase BNT particles at a lower temperature as 160°C for a short time of 3 h [72]. precipitated Pb‐La‐Ti‐OHX hydroxide and KOH solutions with different concentrations of 2, 4 and 6 M at 220°C for 36 h. Additionally, in another set of experiments, the synthesis of Bi0.5Na0.5TiO3 (BNT) was found to occur in highly concentrated alkaline hydrothermal media of 12 M NaOH. This concentration is the minimum required to achieve the crystallization of

single‐phase BNT particles at a lower temperature as 160°C for a short time of 3 h [72].

the crystallization process. However, a large amount of Sr(OH)2 remained when the KOH concentration was increased up to 1 M, and the mass transport rate of ions that feed the growing crystals was accelerated. Therefore, a variety of crystal surfaces have the opportunity to grow due to the relatively larger rate of ionic mass transport, which allowed the formation of cubic‐like‐shaped ST aggregated crystals [69]. Figure 5 gives the typical micrographs of ST perovskite powders hydrothermally prepared at 250°C for various

Figure 5. SEM Micrographs of ST perovskite obtained under hydrothermal conditions at 250C in a KOH solution (5 M) after (a) 0.08, (b) 1 and (c) 24 h [70]. **Figure 5.** SEM Micrographs of ST perovskite obtained under hydrothermal conditions at 250°C in a KOH solution (5 M) after (a) 0.08, (b) 1 and (c) 24 h [70].

#### **2.4.2. Synthesis of ferrite type perovskite (AFeO3)** *2.4.2. Synthesis of ferrite type perovskite (AFeO3)*

an ion-exchange reaction between Pb2+ ions and K+

56 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

equilibrium conditions.

uncoupling OH–

PT phase requires a Pb2+ concentration higher than that of K+

tion of CT and RCT, in addition to the particle shape control.

hydrothermal conditions at high pH, 150°C, and pressure (0.4–0.5 MPa). The formation of the

employing, additionally to the KOH concentration, other mineralizer salts, such as NaNO3, KNO3 and LiNO3 in the reaction media, it was possible to obtain tetragonal perovskite PT nanosheets [66]. The synthesis of pure perovskite PT with tetragonal structure was carried out in a suspension with KOH and NaNO3 concentrations of 1 and 4 M, respectively. According to these studies, it was found that the PT synthesis preferentially occurred under stoichiometric

Miscellaneous research works were carried out to investigate the favourable conditions for synthesizing CT and single-phase RCT doped with rare earth elements (R = Eu) perovskites. These were compounds prepared using the soft chemical process microwave-hydrothermal method without the usage of organic surfactants [67, 68]. The synthesis of pure-phase ortho‐ rhombic CT and RCT was successfully carried out in a reaction media with a pH value of 14 using KOH as the mineralizer. The presence of the alkaline solution promoted the co-precip‐ itation of a complex hydroxide constituted by the TiO(OH)2-Ca(OH)2-Eu(OH)3 during the early stage of the reaction. During the formation of either CT or RCT compounds, the direct rotational water excitation that is achieved by microwave radiation produced the release of

clusters rapidly interact themselves in the aqueous media as a result of the preliminary OH removal process. Therefore, at a pH ≥ 9, the heating promoted by microwave radiation achieved the CT and RCT nucleation, respectively. Under these conditions, the crystallization kinetics of both perovskite-structured compounds was accelerated one to two orders of magnitude. Thus, it was presumed that the diffusion of calcium and titanium clusters is higher under hydrothermal conditions than at ambient pressure and temperature. This phenomenon was explained to proceed by effective particle collisions producing irreversible oriented attachments that offer favourable thermodynamic and kinetics conditions for the crystalliza‐

Regarding the synthesis of ST perovskite, it was conducted in KOH solutions with concentra‐ tions as low as 0.1 M. The ST nuclei precipitated incorporating mild concentrated Sr(OH)2 solutions in the hydrothermal system, and many Sr2+ ions remained in solution after the crystallization process. However, a large amount of Sr(OH)2 remained when the KOH concentration was increased up to 1 M, and the mass transport rate of ions that feed the growing crystals was accelerated. Therefore, a variety of crystal surfaces have the opportunity to grow due to the relatively larger rate of ionic mass transport, which allowed the formation of cubiclike-shaped ST aggregated crystals [69]. Figure 5 gives the typical micrographs of ST perovskite powders hydrothermally prepared at 250°C for various reaction intervals [70]. Finally, the hydrothermal synthesis of some solid solutions was satisfactorily carried out under conditions similar to those mentioned above. Wei et al. [71] prepared pure-phase Pb0.70La0.30TiO3 fine powders with cubic structure by employing co-precipitated Pb-La-Ti-OHX hydroxide and KOH solutions with different concentrations of 2, 4 and 6 M at 220°C for 36 h. Additionally, in another set of experiments, the synthesis of Bi0.5Na0.5TiO3 (BNT) was found to occur in highly

groups from the complex hydroxide. Consequently, calcium and Ti(OH)4

ions. This phenomenon occurs under

in the system. Likewise, by

Recently, the synthesis of multiferroic materials such as perovskite BF has received intensive scientific attention because of its magnetic and ferroelectric properties. A particular interest has been paid for preparing submicron BF powders with the assistance of NaOH mineralizer at lower temperatures (150–190 °C). The reactant reagents used were Bi(NO3)3 and FeCl3 [73]. In this particular case, the crystallization of the perovskite‐structured BF Recently, the synthesis of multiferroic materials such as perovskite BF has received intensive scientific attention because of its magnetic and ferroelectric properties. A particular interest has been paid for preparing submicron BF powders with the assistance of NaOH mineralizer at lower temperatures (150–190 °C). The reactant reagents used were Bi(NO3)3 and FeCl3 [73]. In this particular case, the crystallization of the perovskite-structured BF powders with hexagonal structure occurred in low concentrated NaOH solutions, 0.03–0.12 M, at 170°C for 16 h. The BT perovskite phase crystallization and particle morphology were affected by processing parameters such as solvent solution concentration, reaction temperature and time. The parameter that strongly affects the crystallization step was the concentration of the solvent solution of NaOH. At a low concentration of NaOH media, below 0.03 M, a large number of water molecules and a few [Na(H2O)n)]+ cations coexist in the aqueous media. The presence of these species reduces the rate of the crystallization of the amorphous precursor colloid phase that contains Bi3+ and Fe3+ ions. Therefore, the principal mechanism associated with the nucleation and growth stages involves an in-situ gel transformation process. The amorphous colloid gel transformation occurs under early and intermediate step of the crystallization process. This fact is likely to proceed when water molecules are removed from the structural gel network enabling the nucleation of small embryos. The gradual increase of temperature and solvent concentration promotes the presence of a few number molecules of water together with a large number of [Na(H2O)n]+ cations. These species accelerate the dissolution rate of the amorphous precursor, enhancing a rapid dissolution–recrystallization process. Once the solution was supersaturated, nucleation and crystallization took place faster in the saturated solvent solution [73].

The control of the BF particle morphology can be achieved under hydrothermal conditions. BF pure-phase microplates with rhombohedral structure were preferentially synthesized at 200°C for 8 h, using C6H10BiNO8 as a Bi precursor reactant and particle surface modifier [74]. The synthesis was carried out in a suspension obtained adding a 0.4 M KOH solution. The formation of the perovskite BF rhombohedral structure was achieved by a self-assembly process coupled with coarsening stage promoted by the Ostwald ripening growth mechanism. A process comprising different steps explained the production of the BF pure phase. The nuclei are generated in the supersaturated solution yielding the growth of Bi25FeO40 spherical-shaped nanoparticles during the first of the reaction. Then, these spherical nanoparticles underwent an oriented aggregation due to the preferential 2D plane growth that transformed the spheres into BF nanosheet particles. The adsorption of the C6H10BiNO8 might modify the formation mechanism because this organic compound prevents the contact between the facets on which the adsorption selectively occurred. Finally, the gradual particle growth produced the formation of microplates that occurred via a repeated dissolution–crystallization mechanism.

The hydrothermal synthesis of BF has also been performed using highly concentrated alkaline solutions. A study directed towards to produce large-scale polyhedral BF particles was conducted at 200°C for 12 h using KOH solutions with concentrations varying between 1 to 9 M [75]. The total OH<sup>−</sup> ions concentration affected the agglomeration rate and the particle morphology control. The variation of KOH solvent concentration, reaction time, heating and cooling rates were considered to determine the particle morphology changes of the BF submicron particles. The formation of the large-scale BF polyhedron occurred due to the selective particle aggregation that was promoted via the dissolution–crystallization mecha‐ nism. When the concentration of solute reached the critical supersaturation required to enhance the embryo nucleation, the hydrothermal conditions simultaneously preceded the nucleation and growth of small crystallites and subsequently these BT particles nanoparticles underwent a marked agglomeration. Likewise, at low KOH concentrations, the solvent is not capable of controlling the particle surface energy, in consequence the BF nanoparticles underwent a marked agglomeration forming isotropic sphere aggregates. However, when saturated KOH solutions were used, the KOH favoured the control of the particle surface energy, and the particle growth was dominated via an oriented attachment process, which conducted the formation of euhedral-shaped particles.

In a different set of hydrothermal experiments, the crystallization of BF particles was obtained 220°C for 6 h in a 4 M KOH solution [76]. In contrast, above a critical KOH concentration, the formation of secondary impurity phases was favoured. Likewise, the formation of pure BF was strongly dependent on the KOH solvent concentration. During the progress of the hydrother‐ mal treatment, the hydroxides species of Bi(OH)3 and Fe(OH)3 were further dissolved in the solvent (KOH), and those reacted at high temperatures and pressures. The solvent saturation provoked the precipitation of the chemically stable ceramic oxide particles. The particle coarsening continues gradually as far as the hydrothermal system maintains a steady super‐ saturated state. Therefore, the dissolution and crystallization process progressed in the supersaturated fluid because the system tends to reach stability by itself. It was argued that the dissociation of bismuth and iron hydroxide coupled with the formation of complexes ionic could hinder the growth of BF crystallites and limit the size of BF particles to the submicron range. Other studies related to the synthesis of BF where performed for Jiang et al. [77]. These authors carried out the hydrothermal synthesis of BF single crystals using hydrate sodium bismuth oxide (NaBiO3 nH2O), Fe(NO3)3 and KOH. Rhombohedral-structured perovskite was obtained with a K/Bi molar ratio of 180 at 180°C for 7 days. However, chemical analyses showed that the molar ratio of the reaction product Bi:K:Fe was equal to 0.96:0.03:1.00, also detecting a small amount of potassium. From these results, the chemical composition of BF large crystals could be deduced to be Bi0.96K0.03FeO3. Taking into account the starting compound with Bi5+, the charge balance of this compound may be obtained by the presence of a mixed valence Bi atoms, resulting in the formation of a compound of Bi3+0.915Bi5+0.045K+ 0.03Fe3+O3.

process. This fact is likely to proceed when water molecules are removed from the structural gel network enabling the nucleation of small embryos. The gradual increase of temperature and solvent concentration promotes the presence of a few number molecules of water together with a large number of [Na(H2O)n]+ cations. These species accelerate the dissolution rate of the amorphous precursor, enhancing a rapid dissolution–recrystallization process. Once the solution was supersaturated, nucleation and crystallization took place faster in the saturated

58 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The control of the BF particle morphology can be achieved under hydrothermal conditions. BF pure-phase microplates with rhombohedral structure were preferentially synthesized at 200°C for 8 h, using C6H10BiNO8 as a Bi precursor reactant and particle surface modifier [74]. The synthesis was carried out in a suspension obtained adding a 0.4 M KOH solution. The formation of the perovskite BF rhombohedral structure was achieved by a self-assembly process coupled with coarsening stage promoted by the Ostwald ripening growth mechanism. A process comprising different steps explained the production of the BF pure phase. The nuclei are generated in the supersaturated solution yielding the growth of Bi25FeO40 spherical-shaped nanoparticles during the first of the reaction. Then, these spherical nanoparticles underwent an oriented aggregation due to the preferential 2D plane growth that transformed the spheres into BF nanosheet particles. The adsorption of the C6H10BiNO8 might modify the formation mechanism because this organic compound prevents the contact between the facets on which the adsorption selectively occurred. Finally, the gradual particle growth produced the formation of microplates that occurred via a repeated dissolution–crystallization mechanism.

The hydrothermal synthesis of BF has also been performed using highly concentrated alkaline solutions. A study directed towards to produce large-scale polyhedral BF particles was conducted at 200°C for 12 h using KOH solutions with concentrations varying between 1 to 9

morphology control. The variation of KOH solvent concentration, reaction time, heating and cooling rates were considered to determine the particle morphology changes of the BF submicron particles. The formation of the large-scale BF polyhedron occurred due to the selective particle aggregation that was promoted via the dissolution–crystallization mecha‐ nism. When the concentration of solute reached the critical supersaturation required to enhance the embryo nucleation, the hydrothermal conditions simultaneously preceded the nucleation and growth of small crystallites and subsequently these BT particles nanoparticles underwent a marked agglomeration. Likewise, at low KOH concentrations, the solvent is not capable of controlling the particle surface energy, in consequence the BF nanoparticles underwent a marked agglomeration forming isotropic sphere aggregates. However, when saturated KOH solutions were used, the KOH favoured the control of the particle surface energy, and the particle growth was dominated via an oriented attachment process, which

In a different set of hydrothermal experiments, the crystallization of BF particles was obtained 220°C for 6 h in a 4 M KOH solution [76]. In contrast, above a critical KOH concentration, the formation of secondary impurity phases was favoured. Likewise, the formation of pure BF was strongly dependent on the KOH solvent concentration. During the progress of the hydrother‐

ions concentration affected the agglomeration rate and the particle

solvent solution [73].

M [75]. The total OH<sup>−</sup>

conducted the formation of euhedral-shaped particles.

On the other hand, the synthesis of BF particles has been possible to carry out at low temper‐ atures with the aid of polymers and surfactants such as PVA [78]. The nanoparticles with spherical morphology have an average diameter of about 10 nm; these were satisfactorily synthesized at 160°C for 9 h in a suspension containing 15 ml of PVA. In a second run of experiments a KOH solution without surfactant addition was used as solvent. The pH value of the residual solution after the hydrothermal treatment was 8; this value is high because a highly concentrated KOH (12 M) was added prior the treatment. The polymer decreased the growing speed of BF nuclei enhancing the formation of nanoparticles. In accordance with the former results, the authors concluded that the polymer addition limited the crystal growth, and the crystal growth rate must be slower than the rate associated with the BF embryo nucleation [78]. Furthermore, experimental works were conducted to study the particle morphology control of BF by using BiCl3, FeCl3 and KOH [79]. In the method investigated, the concentration of the KOH solutions varied in the range of 0.2 to 0.7 M, and the hydrothermal synthesis was carried out at standard conditions at 180°C for 24 h. The control of particle morphology was investigated employing organic surfactant with a concentration of 0.2 M (PEG, EDTA, CTAB and PVP). The surfactant was added to the KOH solution prior the mixing of the Bi and Fe precursors. The usage of concentrated KOH solutions produced a marked agglomeration of particles, which gradually dispersed at long reaction intervals. The mor‐ phology of the particles also gradually becomes more regular and homogeneous. A mechanism based on the presence of electrostatic charge micelles is likely to promote the dispersion and morphology control of the BF particles. The micelles formed when a large amount of OH<sup>−</sup> cover the surface of the growing BF particles, generates a network of negatively charged nuclei. The repulsive forces between the nuclei are strong, and after the repulsive force reaches its maximum level, it conduces to particle isolation. Therefore, the decrease of KOH concentration in the hydrothermal system provokes the loss of the repulsive force that causes the marked particle aggregation [78].

Other studies focusing on the synthesis of BF particles via hydrothermal processing were performed in reaction media containing organic solvents, namely ethanol or acetone. Recently, Chen et al. [80–81] studied the crystallization of BF particles at low temperature using these organic solvents, which were mixed with the 4 M KOH solution. Well-crystallized pure rhombohedral structure BF nanoparticles were prepared at 120°C for 16 h with an ethanol/ water ratio of 4:3 [80]. In this typical hydrothermal process, the ratio ethanol/water had a significant role in the formation of pure BF. Later on, BF powders were synthesized with the aid of acetone, using bismuth and iron nitrates as reactants and a KOH solution concentration of 7 M. The study was directed towards to investigate the effect of solvent (KOH/acetone) on the chemical stability of the reaction product. The BF powders were prepared by using acetone as a solvent to dissolve the nitrate. Then, the suspension was hydrothermally treated at 130°C for 12 h; these conditions yield to the formation of nanometric-sized cubic-shaped particles of BF [81]. The preferential formation of the BF powders occurring at very low temperature was explained as follows: (1) the metal elements have different electronegativities, 2.02 for Bi and 1.83 for Fe. This difference might cause a variation in the hydrolysis rates of the salts employed; in practice bismuth salts hydrolyze more easily in water in comparison with the iron salt. The hydrolysis of Bi(NO3)3 rapidly occurred resulting in the subsequent formation of a white insoluble precipitate as it is expressed by chemical reaction 19.

$$\text{Bi(NO}\_3\text{)}\_{3(aq)} + \text{H}\_2\text{O}\_{(aq)} \rightarrow \text{BiONO}\_{3(s)} + \text{2HNO}\_{3(aq)}\tag{19}$$

The usage of acetone as solvent might limit the hydrolysis of bismuth salt, keeping the element as ionic species with their different electronegativities. Therefore, it might reduce the energy required for successful dissolution and precipitation to form the BF phase during the hydro‐ thermal process. Because the presence of insoluble BiONO3 does not proceed in the system, in contrast only Bi(OH)3 and Fe(OH)3 hydroxides (acting as precursors) precipitates. (2) The increase of the vapour pressure in the vessels is likely to proceed due to the evaporation temperature of acetone, which is 56.05°C. The excess of pressure derived from the evaporation of acetone might accelerate the processes of dissolution and crystallization, and, finally, it leads to the formation of pure BF powders at low temperature. (3) A surface tension phenomena produced by the liquid phase can control the reaction mechanism associated with the crystal‐ lization of BF particles. The mechanism is likely to proceed due to the surface tension differ‐ ences existing between acetone, water and absolute ethanol (23.02, 77.82 and 22.39 m/Nm, respectively). Under hydrothermal conditions, if the surface tension of solvent is high leading to hydroxide precipitation, these intermediate reaction products subsequently are likely to undergo into agglomeration. Therefore, longer reaction intervals are required to transform the hydroxides into BF via a dehydration process. When a small volume of an organic compound is mixed with water, i.e. acetone, the surface tension of the solution must decrease, leading to a rapid hydroxide precipitation achieved by the surrounded solvent. The precipitates had a homogeneous dispersion and were easily dehydrated to form BF powders. Another statement proposed to explain the BF synthesis indicates that, (4) the dielectric constant "ε" of the solvent could be lowered due to the organic solvent addition in water to form the hydrothermal media. Thereby, the reduction on "ε" causes the decrease of the BF solubility, promoting a highsaturation Bi and Fe in the solvent media. When this state is reached, it provides the optimum conditions for a rapid nucleation of the BF phase. The presence of acetone accelerates the dissolution of the reactants, enabling the crystallization to occur in the supersaturated fluid, finally leading to the achievement of pure-phase BF powders at a very low temperature [81].

Other studies focusing on the synthesis of BF particles via hydrothermal processing were performed in reaction media containing organic solvents, namely ethanol or acetone. Recently, Chen et al. [80–81] studied the crystallization of BF particles at low temperature using these organic solvents, which were mixed with the 4 M KOH solution. Well-crystallized pure rhombohedral structure BF nanoparticles were prepared at 120°C for 16 h with an ethanol/ water ratio of 4:3 [80]. In this typical hydrothermal process, the ratio ethanol/water had a significant role in the formation of pure BF. Later on, BF powders were synthesized with the aid of acetone, using bismuth and iron nitrates as reactants and a KOH solution concentration of 7 M. The study was directed towards to investigate the effect of solvent (KOH/acetone) on the chemical stability of the reaction product. The BF powders were prepared by using acetone as a solvent to dissolve the nitrate. Then, the suspension was hydrothermally treated at 130°C for 12 h; these conditions yield to the formation of nanometric-sized cubic-shaped particles of BF [81]. The preferential formation of the BF powders occurring at very low temperature was explained as follows: (1) the metal elements have different electronegativities, 2.02 for Bi and 1.83 for Fe. This difference might cause a variation in the hydrolysis rates of the salts employed; in practice bismuth salts hydrolyze more easily in water in comparison with the iron salt. The hydrolysis of Bi(NO3)3 rapidly occurred resulting in the subsequent formation of a white

insoluble precipitate as it is expressed by chemical reaction 19.

60 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

( ) <sup>3</sup> <sup>2</sup> ( ) 3( ) 3( ) ( )3 2 *aq s aq aq*

The usage of acetone as solvent might limit the hydrolysis of bismuth salt, keeping the element as ionic species with their different electronegativities. Therefore, it might reduce the energy required for successful dissolution and precipitation to form the BF phase during the hydro‐ thermal process. Because the presence of insoluble BiONO3 does not proceed in the system, in contrast only Bi(OH)3 and Fe(OH)3 hydroxides (acting as precursors) precipitates. (2) The increase of the vapour pressure in the vessels is likely to proceed due to the evaporation temperature of acetone, which is 56.05°C. The excess of pressure derived from the evaporation of acetone might accelerate the processes of dissolution and crystallization, and, finally, it leads to the formation of pure BF powders at low temperature. (3) A surface tension phenomena produced by the liquid phase can control the reaction mechanism associated with the crystal‐ lization of BF particles. The mechanism is likely to proceed due to the surface tension differ‐ ences existing between acetone, water and absolute ethanol (23.02, 77.82 and 22.39 m/Nm, respectively). Under hydrothermal conditions, if the surface tension of solvent is high leading to hydroxide precipitation, these intermediate reaction products subsequently are likely to undergo into agglomeration. Therefore, longer reaction intervals are required to transform the hydroxides into BF via a dehydration process. When a small volume of an organic compound is mixed with water, i.e. acetone, the surface tension of the solution must decrease, leading to a rapid hydroxide precipitation achieved by the surrounded solvent. The precipitates had a homogeneous dispersion and were easily dehydrated to form BF powders. Another statement proposed to explain the BF synthesis indicates that, (4) the dielectric constant "ε" of the solvent could be lowered due to the organic solvent addition in water to form the hydrothermal media.

*Bi NO H O BiONO HNO* ®+ + (19)

The synthesis of a perovskite composite with a composition of 0.5BF-0.5BT was satisfactorily obtained at 200°C employing a KOH concentration of 4 M for 6 h [82]. The crystallization of perovskite 0.5BF-0.5BT is achieved by a "dissolution and crystallization process"; this theory has been widely applied to explain the hydrothermal reaction pathways. The relevant chemical reactions that occur during the synthesis that are correlated with the crystallization of BF-BT are as follows (reactions 20 to 24):

$$2\text{Bi}(\text{NO}\_3\text{)}\_{3(aq)} + 6\text{KOH}\_{(aq)} = \text{Bi}\_2\text{O}\_{3(s)} + \text{ } 6\text{K}^\* + \text{ } 6\text{NO}^-\text{)}\_3 + \text{ } 3\text{H}\_2\text{O}\_{(l)}\tag{20}$$

$$\text{Fe} \text{(NO}\_3\text{)}\_3 + \text{3KOH}^- = \text{Fe} \text{(OH)}\_3 + \text{3K}^+ + \text{3NO}^-\_3 \tag{21}$$

$$\text{TiCl}\_4 + \text{H}\_2\text{O} \rightleftharpoons \text{TiOCl}\_2 + \text{2HCl} \tag{22}$$

$$\text{TiOCl}\_2 + 2\text{HCl} + 4\text{KOH} = \text{Ti(OH)}\_{4(aq)} + 4\text{K}^\circ + 4\text{Cl}^- + H\_2\text{O}\_{(l)}\tag{23}$$

$$\text{Ti(OH)}\_{4} + \text{2OH}^{-} = \left[ \text{Ti(OH)}\_{6} \right]^{2-} \tag{24}$$

When the hydrothermal synthesis takes place, several reactions occurred between the reactants of bismuth oxide, iron III hydroxide, barium hydroxide and titanium hydroxide. To produce the desired crystalline phases following the chemical reactions 21 and 22 must occur in the hydrothermal system at particular conditions of temperature, time and pressure. Additionally, the results demonstrated that the reaction temperature and the KOH concentration must be optimized to limit the presence of reaction by-products. Under these conditions, the formation of perovskite 0.5BF-0.5BT is likely to be conducted following a single-step reaction.

$$\text{Bi}\_2\text{O}\_3 + \text{2Fe} \{ \text{OH} \}\_3 = \text{2BiFeO}\_3 + \text{3H}\_2\text{O} \tag{25}$$

$$\left[Ba^{2+} + \left[Ti\left(\Theta H\right)\_6\right]\right]^{2-} = BaTiO\_3 + \ 3H\_2O \tag{26}$$

A reduced number of researchers have carried out the synthesis of RF (R= rare earth elements group or Y) under hydrothermal conditions using rare-earth nitrates as precursor feedstock. The hydrothermal synthesis of RF particles was conducted in highly concentrated alkaline solvents between 2 and 44 M for 48 h, whereas the reaction temperature required to promote the crystallization of the single ABX3 phase varied between the range of 230 up to 240°C [83– 85]. The formation mechanism of the RF crystals was associated with a precipitation process promoted by the precursor dissolution and nucleation processes. The role of alkalinity was a crucial parameter to control the dissolution, nucleation and growth processes because the chemical reagent precursors are chemically stable even in strong alkaline solvents such as NaOH or KOH.

Similar to the BT powders, the microwave hydrothermal synthesis has been used to synthesize perovskite BF to achieve fast crystallization kinetics and control microstructural aspects of the particles. The hydrothermal crystallization process assisted by microwave radiation accom‐ plishes the formation of BF single phase with rhombohedral structure, for very short reaction intervals of 1 h. The precursor selected to perform this reaction were the reagent-grade salts of bismuth (Bi(NO3)3) and iron (FeCl3, Fe(NO3)3). The synthesis process can occur at tempera‐ tures in the range of 150–200°C in KOH solutions with concentrations varying between 0.05 and 4 M [86, 87]. Experimental results indicate that crystallization of pure-phase BF powders is possible to take place at 200°C for 1 h with a KOH concentration of 0.05 M, due to the fast heating supplied by the microwave radiation with the aid of the precursor feedstock selected [86]. One point that deserves emphasize is related to the abnormal growth of the BF particles; this causes marked agglomeration of the BT particle. This fact is proposed to be the primary reason for the occurrence of a change in the crystal structure because a high OH– concentration favours the nucleation process. Moreover, Wang et al. [88] studied the synthesis of BF by using a polyanion, poly(methyl vinil ether-alt-maleic acid) (PMVEMA), KOH and Na2CO3 solutions with concentrations of 1 and 8 M, respectively. The BF pure phase was synthesized under the same conditions determined elsewhere [87], and a simple ultrasonic purification method was developed to obtain the pure phase. In particular, PMVEMA was selected because it contains a large number of –COOH groups that operate as anions, which avoid bringing other metal ions. Additionally, the presence of these ions also can control the concentration change of metal ions such as Na+ and K+ in the hydrothermal reaction system [88].

## *2.4.3. Synthesis of zirconite-type perovskite (AZrO3)*

Another oxide compound of interest due to its potential applications is the perovskite BZ. Continuous synthesis technologies that allowed the researchers to produce tens of grams of powder products per production hour are devoted to developing at an industrial level. The perovskite BZ phase has been obtained by a continuous hydrothermal synthesis proceeding in a single step under supercritical conditions [89, 90]. The cubic BZ pure-phase particles have been produced at higher temperatures around 450–500°C using NaOH as the solvent media. A high temperature is necessary to complete the release of OH– ions coupled with the neutralization of protons, because at severe supercritical conditions the reaction vessel can suffer a rapid acidic corrosion. The reactions that occur in the autoclave vessel are described by chemical reactions 27 to 29. Before the occurrence of these reactions, the dissolution of the precursors must take place, due to rapid ionic mobility enhanced by the dielectric constant of the solvent "ε":

A reduced number of researchers have carried out the synthesis of RF (R= rare earth elements group or Y) under hydrothermal conditions using rare-earth nitrates as precursor feedstock. The hydrothermal synthesis of RF particles was conducted in highly concentrated alkaline solvents between 2 and 44 M for 48 h, whereas the reaction temperature required to promote the crystallization of the single ABX3 phase varied between the range of 230 up to 240°C [83– 85]. The formation mechanism of the RF crystals was associated with a precipitation process promoted by the precursor dissolution and nucleation processes. The role of alkalinity was a crucial parameter to control the dissolution, nucleation and growth processes because the chemical reagent precursors are chemically stable even in strong alkaline solvents such as

62 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Similar to the BT powders, the microwave hydrothermal synthesis has been used to synthesize perovskite BF to achieve fast crystallization kinetics and control microstructural aspects of the particles. The hydrothermal crystallization process assisted by microwave radiation accom‐ plishes the formation of BF single phase with rhombohedral structure, for very short reaction intervals of 1 h. The precursor selected to perform this reaction were the reagent-grade salts of bismuth (Bi(NO3)3) and iron (FeCl3, Fe(NO3)3). The synthesis process can occur at tempera‐ tures in the range of 150–200°C in KOH solutions with concentrations varying between 0.05 and 4 M [86, 87]. Experimental results indicate that crystallization of pure-phase BF powders is possible to take place at 200°C for 1 h with a KOH concentration of 0.05 M, due to the fast heating supplied by the microwave radiation with the aid of the precursor feedstock selected [86]. One point that deserves emphasize is related to the abnormal growth of the BF particles; this causes marked agglomeration of the BT particle. This fact is proposed to be the primary

reason for the occurrence of a change in the crystal structure because a high OH–

favours the nucleation process. Moreover, Wang et al. [88] studied the synthesis of BF by using a polyanion, poly(methyl vinil ether-alt-maleic acid) (PMVEMA), KOH and Na2CO3 solutions with concentrations of 1 and 8 M, respectively. The BF pure phase was synthesized under the same conditions determined elsewhere [87], and a simple ultrasonic purification method was developed to obtain the pure phase. In particular, PMVEMA was selected because it contains a large number of –COOH groups that operate as anions, which avoid bringing other metal ions. Additionally, the presence of these ions also can control the concentration change of metal

in the hydrothermal reaction system [88].

Another oxide compound of interest due to its potential applications is the perovskite BZ. Continuous synthesis technologies that allowed the researchers to produce tens of grams of powder products per production hour are devoted to developing at an industrial level. The perovskite BZ phase has been obtained by a continuous hydrothermal synthesis proceeding in a single step under supercritical conditions [89, 90]. The cubic BZ pure-phase particles have been produced at higher temperatures around 450–500°C using NaOH as the solvent media. A high temperature is necessary to complete the release of OH– ions coupled with the neutralization of protons, because at severe supercritical conditions the reaction vessel can suffer a rapid acidic corrosion. The reactions that occur in the autoclave vessel are described

concentration

NaOH or KOH.

ions such as Na+

and K+

*2.4.3. Synthesis of zirconite-type perovskite (AZrO3)*

$$\text{Ba} \left( \text{NO}\_3 \right)\_2 \to \text{Ba}^{2\*} + \text{ 2NO}\_3^- \tag{27}$$

$$\text{NaOH}^- \rightarrow \text{Na}^+ + \text{OH}^- \tag{28}$$

$$\text{ZrO(NO}\_3\text{)}\_2 \rightarrow \text{ZrO}^{2\*} + \text{2NO}\_3 \tag{29}$$

The crystallization goes further under supercritical when the precursors are mixed because of the high ionic mobility in the solvent phase due to the low dielectric constant. The reaction that occurs is described as follows:

$$2\text{ZrO}^{2\*} + \text{Ba}^{2\*} + \text{2OH}^{-} \rightarrow \text{BaZrO}\_{3(s)} + \text{2H}^{+} \tag{30}$$

In contrast, the possibility of preparing BZ powders with cubic structure was investigated under hydrothermal conditions at 130°C for 24 h. In these experiments, the adjustment of hydrothermal media feedstock to a value of 13 occurs by adding a KOH solution with a concentration of 0.5 M [91]. Furthermore, the formation of RBZ perovskite solid solutions with cubic structure required a minimum reaction temperature of 200°C for 24 h. The reaction was conducted using a highly concentrated KOH of 16 M as a solvent; the strong alkaline solutions rapidly dissolved the reactant precursor [92]. The morphology of the RBZ particles obtained under the severe alkaline hydrothermal conditions resembled hollow nanospheres. However, the particle sizes were simply controlled varying the Rb content in the precursor and keeping the KOH concentration constant. It was confirmed that adjusting the KOH concentration would change the inter-particle force, varying the size of the aggregates.

## *2.4.4. Synthesis of lanthanum chromites and manganites-type perovskites and their solid solutions (LaCrO3, LaMnO3)*

In recent years, much work has been conducted for synthesizing perovskite-type LaCrO3 and LaMnO3 (LC and LM) under mild hydrothermal conditions. There are several reports on the preparation of nanoparticles under hydrothermal conditions. Zheng et al. [93] studied the effect of solvent alkalinity by employing KOH with concentrations between 2 and 16 M. Another parameter evaluated in this work was the molar ratio of KOH/Cr = 10–80, the hydrothermal treatments were carried out at a 260°C for a fixed time of 7 days. These authors found that a highly concentrated alkaline solvent is required when the selected B-cation has an amphoteric behaviour, as it is the particular case of Cr3+. Therefore, the formation of pure LC required ≥8 M of KOH in the reaction medium.

A series of solid solutions based on the perovskite-type structure has been hydrothermally prepared. The synthesis of La1–*x*Sr*x*CrO3 (LSC) with orthorhombic structure has been carried out using different reaction media to prepare in a preliminary stage the co-precipitated gel precursor. Perovskite LSC powders were produced using Triton non-ionic surfactant and NH4OH as co-precipitation medium; the pH of the starting aqueous media was 8. The reaction rate associated with this process was affected by the common ion effect caused by the metal precursor as well as by the steric barrier due to the presence of the surfactant micelles [94]. Also, the synthesis of LSC fine powders in a media with a NaOH concentration of 0.5 M was achieved by the dissolution–crystallization mechanism. The solvent even at the concentration selected enhances a homogeneous nucleation leading to the formation of dispersed particles. The hydrothermal solvent chemically reacts with the solid species incorporated into the vessel, producing the dissolution of all solids and the ionic saturation of the aqueous solvent media. In consequence, a spontaneous precipitation of LC and LSC powders occurs [95]. The synthesis of these powders also proceeded in a reaction medium of KOH solution. In this procedure, the KOH solution was mixed in two steps. Firstly, KOH was mixed with CrCl3 and to produce Cr(OH)3, while in the second step KOH was used to adjust the alkalinity of the reaction medium after the sources of Sr and La were added. Moreover, a higher alkalinity is necessary for the preparation of LSC because Cr3+ behaves as an amphoteric element. In fact, the optimum alkalinity range that promotes the crystallization of LSC solid solutions is from 5 up to 8 M [96]. A similar reaction pathway was determined for the synthesis of rare-earth and yttrium orthochromite perovskites. Because when a rare earth element partially substitutes one of the major constituents in the perovskite structure, it provokes that a concentrated alkaline solution must be used as a precursor solvent (i.e. 10–12 M KOH) [97].

Several types of lanthanum manganite solid solutions have been synthesized under hydro‐ thermal conditions. The synthesis of LBM nanowires (La0.5Ba0.5MnO3) with cubic perovskite structure was produced at 270°C for 25 h with a net [OH–] = 10 M; using La(NO3)3, Ba(OH)2, KMnO4 and MnCl4 as precursors [98]. Likewise, at the similar conditions of synthesis of LCM (La0.5Ca0.5MnO3) solid solution with orthorhombic structure was produced. The reaction associated with the hydrothermal processing investigated is given as follows [99].

$$\begin{aligned} &5\text{KMnO}\_{4(aq)} + 7\text{MnOCl}\_{2(aq)} + 5\text{La} \text{(NO}\_3\text{)}\_{3(aq)} + 5\text{Ca} \text{(NO}\_3\text{)}\_{2(aq)} + 36\text{KOH}\_{(aq)} \rightarrow \\ &10\text{La}\_{0.5}\text{Ca}\_{0.5}\text{MnO}\_{3(s)} + 18\text{H}\_2\text{O}\_{(aq)} + 25\text{KNO}\_{3(aq)} + 14\text{KCl}\_{(aq)} \end{aligned} \tag{31}$$

Recently, perovskite LSM (La1-xSrxMnO3+δ) were obtained at 150°C for 20 h by adjusting the suspension (hydrothermal media) pH to a value of 9 with an ammonia solution. Based on the results of preliminary runs, a systematic study was conducted aiming to obtain the desired chemical compositions of solid solutions of manganite [100]. A mechanism of dendrite nucleation leads to the synthesis of this particular perovskite compound. This fact was established based on the crystalline phase evolution analyses of the reaction product produced after each treatment, which was conducted in a hydrothermal media of 4 M NaOH. It was found that the SLM solid solution with hexagonal perovskite structure and with particle morphology of hexagonal platelets was formed at temperatures below 220°C. In contrast, the mechanism of embryo nucleation differs when the reaction temperature was 240°C. At this temperature, the LSM particles were formed via the epitaxial nucleation; this mechanism acts preferentially at the edges of the hexagonal platelet. In consequence, the growth of "tree-like" LCM dendrites is promoted under the hydrothermal conditions investigated [101]. In recent years, Spooren et al. [102] studied the synthesis of LCM, LSM and LBM solid solutions using a KOH reaction media; the use of other solvents was also investigated. However, in all the new solvents tested, the formation of reaction by-products was not avoided. The hydrothermal synthesis of LBM, LSM and LCM were achieved in the KOH solvent solution for different molar compositional ratios of:

Mn2+:MnO4 2-:Ba2+:La3+:KOH:H2O equivalent to 7:3:5:5:1250:3256 at 240°C for 24 h.

A series of solid solutions based on the perovskite-type structure has been hydrothermally prepared. The synthesis of La1–*x*Sr*x*CrO3 (LSC) with orthorhombic structure has been carried out using different reaction media to prepare in a preliminary stage the co-precipitated gel precursor. Perovskite LSC powders were produced using Triton non-ionic surfactant and NH4OH as co-precipitation medium; the pH of the starting aqueous media was 8. The reaction rate associated with this process was affected by the common ion effect caused by the metal precursor as well as by the steric barrier due to the presence of the surfactant micelles [94]. Also, the synthesis of LSC fine powders in a media with a NaOH concentration of 0.5 M was achieved by the dissolution–crystallization mechanism. The solvent even at the concentration selected enhances a homogeneous nucleation leading to the formation of dispersed particles. The hydrothermal solvent chemically reacts with the solid species incorporated into the vessel, producing the dissolution of all solids and the ionic saturation of the aqueous solvent media. In consequence, a spontaneous precipitation of LC and LSC powders occurs [95]. The synthesis of these powders also proceeded in a reaction medium of KOH solution. In this procedure, the KOH solution was mixed in two steps. Firstly, KOH was mixed with CrCl3 and to produce Cr(OH)3, while in the second step KOH was used to adjust the alkalinity of the reaction medium after the sources of Sr and La were added. Moreover, a higher alkalinity is necessary for the preparation of LSC because Cr3+ behaves as an amphoteric element. In fact, the optimum alkalinity range that promotes the crystallization of LSC solid solutions is from 5 up to 8 M [96]. A similar reaction pathway was determined for the synthesis of rare-earth and yttrium orthochromite perovskites. Because when a rare earth element partially substitutes one of the major constituents in the perovskite structure, it provokes that a concentrated alkaline solution

Several types of lanthanum manganite solid solutions have been synthesized under hydro‐ thermal conditions. The synthesis of LBM nanowires (La0.5Ba0.5MnO3) with cubic perovskite structure was produced at 270°C for 25 h with a net [OH–] = 10 M; using La(NO3)3, Ba(OH)2, KMnO4 and MnCl4 as precursors [98]. Likewise, at the similar conditions of synthesis of LCM (La0.5Ca0.5MnO3) solid solution with orthorhombic structure was produced. The reaction

( ) ( ) ( ) ( ) ( ) 42 3 3 3 2 ( )

*aq aq aq aq aq*

Recently, perovskite LSM (La1-xSrxMnO3+δ) were obtained at 150°C for 20 h by adjusting the suspension (hydrothermal media) pH to a value of 9 with an ammonia solution. Based on the results of preliminary runs, a systematic study was conducted aiming to obtain the desired chemical compositions of solid solutions of manganite [100]. A mechanism of dendrite nucleation leads to the synthesis of this particular perovskite compound. This fact was established based on the crystalline phase evolution analyses of the reaction product produced after each treatment, which was conducted in a hydrothermal media of 4 M NaOH. It was found that the SLM solid solution with hexagonal perovskite structure and with particle

( )

®

(31)

associated with the hydrothermal processing investigated is given as follows [99].

*KMnO MnCl La NO Ca NO KOH*

++ + +

.5 3( ) 2 ( ) 3( ) ( )

++ +

*s aq aq aq*

3 7 5 5 36

10 18 25 14

*La Ca MnO H O KNO KCl*

must be used as a precursor solvent (i.e. 10–12 M KOH) [97].

64 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

0.5 0

Mn2+:MnO4 2-:Sr2+ :La3+:KOH:H2O equivalent to 7:3:5:5:1250:3256 at 240°C for 24 h.

Mn2+:MnO4 2-:Ca2+:La3+:KOH:H2O equivalent to 7:3:5:5:1250:3256 at 270°C for 24 h.

At present, there are several authors interested in the study of perovskite solid solutions. [103– 104]. The investigations were directed towards to develop a simple route of synthesis for LCSM (La1–*x*Ca*x*Sr*x*MnO3) solid solutions with orthorhombic structure and cuboidal morphology. The synthesis was conducted at 240°C for 72 h. The preparation of perovskite LAM (La1– *<sup>x</sup>*Ag*x*MnO3) occurred employing the conventional hydrothermal and microwave-assisted hydrothermal method with a reaction media saturated with NH4OH solution to adjust the solution pH at 8.5 [105]. Furthermore, orthorhombic perovskite manganates YCM (YMnO3, and Y1–xCaxMnO3, x = 0.07, 0.55, 0.65) were prepared in one step by a mild hydrothermal synthesis. The optimum concentration of the alkaline solvent media employed to conduct the hydrothermal crystallization 23 M to stabilize the single phase of YCM [106]. The pH of the solvent solution is a critical parameter to achieve the crystallization of PSM and NSM man‐ ganites. The adequate KOH molar concentration that favours reaching the equilibrium between the dissolution crystallization events are between 5.0 and 10 M. Therefore, the synthesis of solid solutions PSM (Pr0.5Sr0.5MnO3), NSM (Nd0.5Sr0.5MnO3) and alkali-earth manganese oxides such as 4H-SrMnO3 and 2H-BaMnO3 were conducted using KOH with concentrations within the abovementioned range [107]. As for the case of perovskite RMnO3 (R = Dy, Ho, La, Pr, Nd, Tb and Bi), a common feature of the mild hydrothermal synthesis is the high alkalinity of the solutions. Such high alkalinity provides a critical condition, which considerably influences the crystallization and composition of the RMnO3 product in the hydrothermal synthesis [108–109]. The high alkalinity of the reaction system also constitutes an optimal condition that considerably influences the crystallization of perovskite manganites. To maintain the suitable alkalinity required for the hydrothermal reaction, it is necessary to use a strong solvent such as KOH [110].

*2.4.5. Synthesis of miscellaneous type perovskites and their solid solutions (ANbO3, ATaO3)*

The solvent concentration also has an effect on the nucleation and crystal growth of other perovskite compounds, namely NN and KN (NaNbO3, KNbO3). The formation of this particular group of perovskites has been tailored in a media composed of a mixture of KOH and NaOH solutions. The solvent has a composition assigned by the K+ :Na+ molar ratio of 1:1, while the alkaline solvent solutions concentrations (NaOH or KOH) were 0.6, 0.8, 1.0 and 1.6 M. In the hydrothermal environment, OH– ions exhibit a catalytic behaviour assisting the mass transport of the ionic solute species during the crystallization process. Also, these ions contribute to the particle nucleation and acceleration of the reaction [111]. However, when only NaOH was used as mineralizer, it was found that the reaction took place by an in situ transformation mechanism. The reaction speed was increased at low concentration of [OH– ] with increasing temperature. A high [OH– ] is not favourable to prepare perovskite NN, and there is an optimum [OH– ]. The concentration of [OH– ] and the reaction speed were correlated, which indicated that a high alkalinity is not useful for the synthesis of niobate [112]. In contrast, when a particular polymorph of Nb2O5 was selected as a precursor, the stability of the hexaniobate ion in solution was greater for lower KOH concentration and low temperature. In other words, the induction period related to the conversion of the hexaniobate ion to perovskite powder varied changing the concentration of KOH for a given temperature. Hence, it is possible to produce orthorhombic NN powders at 200°C using solutions of 8.4 M NaOH, 0.25 M Nb2O5 and KN powders; the reaction between solutions of different Nb2O5 concentra‐ tions (0.0015–0.38 M) and KOH (6.7–15 M) was only investigated at 200°C [113]. The reaction mechanism associated with the synthesis of AN is markedly different from that of NN reported in the literature. The heterogeneous chemical reaction between NH4HF2, Ag2O, and Nb2O5 gives perovskite AN at pH = 3, adjusting the reaction media adding either an acidic solution (HF) or mild basic solution (NH4OH). In the hydrothermal processing, water is not only used as the solvent for the crystallization of inorganic substances from solution, but other chemical processes also participate in the energy transfer, catalysis, dissolution and hydrolysis. The experimental results indicate that water solvent, as a solvent, provides a more powerful environment for the hydrothermal reactions for the crystallization of AN perovskites [114].

Cubic perovskite NT (NaTaO3) powders were hydrothermally produced at 200°C for 24 h, employing a reaction medium with a concentration of 7 M and using KOH as mineralizer [115]. Due to the higher OH– concentration contained in the hydrothermal system, a reaction byproduct (pyrochlore) was obtained after the treatment. The pyrochlore crystalline phase subsequently reacted with the solvent media (KOH) to form a perovskite non-stoichiometric phase. In contrast, it was found that the reaction between Ta2O5 and NaOH for the synthesis of perovskite NT (NaTaO3) powders follows the dissolution–recrystallization mechanism under microwave-assisted hydrothermal conditions. This reaction occurred without the addition of organic templates compounds and catalysts to the reaction medium. The thermo‐ dynamic energy barrier associated with the phase conversions, in this case, was estimated to be in the following order: Ta2O5/Na2Ta2O6 < NaTaO3/Na2Ta2O6 < Ta2O5/NaTaO3. A pretreatment of the Ta2O5 raw material by ball milling is a crucial step to obtain pure perovskite NT phase [116].

## *2.4.6. Synthesis of B site perovskite (MeBO3)*

At present, there are a few studies focused on the synthesis of ordered perovskites partially substituted at the B site. The synthesis of a few perovskite MeBO3 group solid solutions have been carried out under mild hydrothermal conditions, where the elements chosen were Me = Pb or Ba; B = Ti, Zr, Zn, or Ta, among others (hereafter these solid solutions are denominated as PZT, BTZ, BZT, respectively). These studies have been oriented to analyze the consolidation and functional properties of the synthesized materials, but any details were considered to clarify the chemical reaction pathways associated synthesis reaction that promotes the crystallization of these perovskite compounds. The present section of the review describes the conditions at which the synthesis of the perovskite solid solutions is possible to progress in a hydrothermal environment.

and NaOH solutions. The solvent has a composition assigned by the K+

66 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

]. The concentration of [OH–

M. In the hydrothermal environment, OH–

with increasing temperature. A high [OH–

there is an optimum [OH–

Due to the higher OH–

NT phase [116].

*2.4.6. Synthesis of B site perovskite (MeBO3)*

while the alkaline solvent solutions concentrations (NaOH or KOH) were 0.6, 0.8, 1.0 and 1.6

transport of the ionic solute species during the crystallization process. Also, these ions contribute to the particle nucleation and acceleration of the reaction [111]. However, when only NaOH was used as mineralizer, it was found that the reaction took place by an in situ transformation mechanism. The reaction speed was increased at low concentration of [OH–

which indicated that a high alkalinity is not useful for the synthesis of niobate [112]. In contrast, when a particular polymorph of Nb2O5 was selected as a precursor, the stability of the hexaniobate ion in solution was greater for lower KOH concentration and low temperature. In other words, the induction period related to the conversion of the hexaniobate ion to perovskite powder varied changing the concentration of KOH for a given temperature. Hence, it is possible to produce orthorhombic NN powders at 200°C using solutions of 8.4 M NaOH, 0.25 M Nb2O5 and KN powders; the reaction between solutions of different Nb2O5 concentra‐ tions (0.0015–0.38 M) and KOH (6.7–15 M) was only investigated at 200°C [113]. The reaction mechanism associated with the synthesis of AN is markedly different from that of NN reported in the literature. The heterogeneous chemical reaction between NH4HF2, Ag2O, and Nb2O5 gives perovskite AN at pH = 3, adjusting the reaction media adding either an acidic solution (HF) or mild basic solution (NH4OH). In the hydrothermal processing, water is not only used as the solvent for the crystallization of inorganic substances from solution, but other chemical processes also participate in the energy transfer, catalysis, dissolution and hydrolysis. The experimental results indicate that water solvent, as a solvent, provides a more powerful environment for the hydrothermal reactions for the crystallization of AN perovskites [114].

Cubic perovskite NT (NaTaO3) powders were hydrothermally produced at 200°C for 24 h, employing a reaction medium with a concentration of 7 M and using KOH as mineralizer [115].

product (pyrochlore) was obtained after the treatment. The pyrochlore crystalline phase subsequently reacted with the solvent media (KOH) to form a perovskite non-stoichiometric phase. In contrast, it was found that the reaction between Ta2O5 and NaOH for the synthesis of perovskite NT (NaTaO3) powders follows the dissolution–recrystallization mechanism under microwave-assisted hydrothermal conditions. This reaction occurred without the addition of organic templates compounds and catalysts to the reaction medium. The thermo‐ dynamic energy barrier associated with the phase conversions, in this case, was estimated to be in the following order: Ta2O5/Na2Ta2O6 < NaTaO3/Na2Ta2O6 < Ta2O5/NaTaO3. A pretreatment of the Ta2O5 raw material by ball milling is a crucial step to obtain pure perovskite

At present, there are a few studies focused on the synthesis of ordered perovskites partially substituted at the B site. The synthesis of a few perovskite MeBO3 group solid solutions have

concentration contained in the hydrothermal system, a reaction by-

:Na+ molar ratio of 1:1,

]

ions exhibit a catalytic behaviour assisting the mass

] is not favourable to prepare perovskite NN, and

] and the reaction speed were correlated,

The hydrothermal processing procedure established for the synthesis of PZT perovskites involves the preparation of a hydrated mixed gel containing the main elements that constitute this compound. In this study, the factor exhaustively investigated was the molar Zr/Ti ratio. The effect of this parameter was considered to determine the optimum molar Zr/Ti ratio that chemically stabilizes the perovskite crystalline structure in different hydrothermal environ‐ ments. As a consequence, the PZT single phase must belong to one of the three different crystalline structures, namely tetragonal, rhombohedral or pseudo-tetragonal [117]. The synthesis of PZT might occur in the sequence depicted by the chemical reactions that occur during the hydrothermal crystallization of PZT, represented by reactions 32 to 34:

$$\text{PbO} \, + \, 0.1 \, \text{TiO}\_2 \to \Big(\text{PbO} \, : 0.1 \, \text{TiO}\_2\big)\_{\text{(s)}}\tag{32}$$

$$2\left(PbO: 0.1 \text{TiO}\_2\right)\_{\text{ss}} + 0.8 \text{TiO}\_{2(\text{gal})} + ZrO\_{2(\text{gal})} \rightarrow 2Pb\left(Zr\_{0.5} \text{Ti}\_{0.5}\right)O\_{3(amorphous)}\tag{33}$$

$$2Pb\left(Zr\_{0.5}Ti\_{0.5}\right)\mathcal{O}\_{3\text{(amorphous)}} \rightarrow 2Pb\left(Zr\_{0.5}Ti\_{0.5}\right)\mathcal{O}\_{3\text{(s)}}\tag{34}$$

The presence of H2O under pressure (40–50 MPa) enhanced the solubility of TiO2 in PbO, due to rapid transport of the ionic species in the supercritical fluid. In the KOH mild alkaline fluid between 160 and 350°C, PbO partially substituted with Ti rapidly reacts with the homogene‐ ously dispersed precursor gel. The reaction accompanied by a progressive splitting off even large PbO particles (6 μm) occurred nearly the supercritical state of the hydrothermal media at 350°C by adjusting the solvent pH between 12 and 14. At this condition, the crystallization of the PZT tetragonal perovskite preferentially proceeded according to the chemical equili‐ brium that is given by Eq. (33). Therefore, the successful reactant mixing ratio Pb:Zr:Ti was 1:0.5:0.5, at which thermodynamically the hydrothermal media reach the equilibrium to achieve the crystallization of the tetragonal unit cell. The increase of the reaction time during the hydrothermal treatment provokes an increase in the crystallinity of the PZT product. It is well known that under hydrothermal treatment in an aqueous medium with basic pH (13.5), the hydroxide cation complexes of Pb, Zr and Ti undergoes a transition into PZT [118]. The crystallization of a nanocrystalline perovskite precursor phase was promoted using colloidal wet chemical processing and low-temperature crystallization stages. The common chemistry preparative procedure involves the co-precipitation of the precursor perovskite constituents from a saturated solution. The co-precipitated colloid gel is obtained using an alkaline KOH solution with a concentration of 4 M producing a colloid suspension. The suspension contain‐ ing the co-precipitated colloid (Pb-Zr-Ti-hydroxide) hydrothermally treated in an autoclave at 250°C for 5 h leads to prepare the tetragonal structured PZT particles exhibiting a cubic morphology and particle size of 600 nm [119]. On the other hand, the production of single crystal tetragonal perovskite PZT (PbZr0.52Ti0.48O3) has been conducted adding a polymer compound. The process denominated as polymer-assisted hydrothermal synthesis was carried out at 200°C for 12 h. Among the polymer additives selected to control the micelles creation were PVA and PAA. The polymer plays an important role in the growth of PZT single crystals with tetragonal structure. The micelles tend to cover the faceted surfaces of the PZT particles, and this event produces a reduction in the superficial energy of the growing particles. Consequently, the PZT particles exhibit a preferential growth along the c axis of the tetragonal structure [120]. Besides the employment of water-soluble organic compounds aiming to control the particle crystal growth. A variety of organic bases provided another advantage for synthesizing PZT perovskites. The possibility to apply the organic bases as potential solid solvents in hydrothermal media was recently investigated in the preparation of PZT perovskite powders. The experiments were carried out in an alkali-free reaction media at 160°C for 72 h. The organic alkaline media selected to conduct the hydrothermal treatments was the tetrame‐ thylammonium hydroxide (TMAOH), this also operated as a pH-adjusting agent [121].

Other solid solutions of interest are the BTZ (BaTi0.8Zr0.2O3) and the BZT (BaZr1–*x*Ti*x*O3). The synthesis of these compounds at mild conditions has not been studied in detail yet. In one study conducted to prepare BTZ perovskite, a stoichiometric powdered mixture of Ti and Zr hydroxides, which was added to a 6 M Ba(OH)2 solution. The barium saturated alkaline Ba(OH)2 solution was selected as a solvent for the hydrothermal treatments to adjust the slurry before the hydrothermal synthesis. In this reaction media it was possible to carry out the crystallization of BZT particles with cubic structure, spherical morphology and molecular level homogeneous composition [122]. Likewise, BZT hollow nanospheres with cubic structure were synthesized at 200°C for 24 h using a KOH solution with a concentration of 16 M. The formation of the particles occurred due to Ostwald ripening mechanism, and the excess of Ti in the fluid may change the reaction mechanism that favours the hollowing process [123]. Finally, the synthesis reaction for the formation of BZnT (BaZn1/3Ta1/3O3) was conducted in a reaction medium with a strong organic base (TMAOH), and the raw materials used were barium and zinc acetate, and tantalum oxalate solution. The reactions involved in the hydro‐ thermal synthesis of the BZnT perovskite particles are reactions 35 to 37. Initially, in the first stage of the reaction occurred the decomposition of the organic alkaline produces a reduction in the total concentration of OH– species at temperatures up to 150°C, simultaneously the nucleation of BZnT embryos occurred in the solution according to the chemical reaction Eq. (35). The decomposition of the TMAH to trimethylamine and methanol (Eq. 36) takes place at temperatures above 220°C in the hydrothermal vessel. The product trimethylamine is a weak alkali, and it has the same chemical reactivity such as that of the ammonia in an aqueous solution (Eq. 37), [124]. Hence, the low chemical stability of the organic alkaline might hinder the crystallization of the desired BZnT perovskite phase because the supersaturation state is difficult to reach in the reaction system.

Synthesis of Perovskite Oxides by Hydrothermal Processing – From Thermodynamic Modelling to Practical... http://dx.doi.org/10.5772/61568 69

$$\begin{aligned} \text{GBa} \text{(MeCO}\_2\text{)}\_{2(aq)} + \text{Zn} \text{(MeCO}\_2\text{)}\_{2(aq)} + \text{Ta}\_2 \text{O}\_3 \text{x} \text{H}\_2\text{O} &+ 8 \text{NMe}\_4 \text{^\circ OH}\_{(aq)} \rightarrow\\ \text{GBaZn}\_{1/3} \text{Ta}\_{2/3} \text{O}\_{3(s)} &+ 8 \text{(MeCO}\_2\text{)}^\cdot \text{NMe}\_4 \text{\text{(aq)}} + \text{ (4+x)} \text{H}\_2\text{O}\_{(l)} \end{aligned} \tag{35}$$

$$\mathrm{NMe}\_{4}\mathrm{"OH}^{-}\_{\mathrm{(aq)}} \rightarrow \mathrm{NMe}\_{\mathrm{3(aq)}} + \mathrm{MeOH}\_{\mathrm{(aq)}}\tag{36}$$

$$\text{NHMe}\_{3(aq)} + \text{H}\_2\text{O}\_{(l)} \Leftrightarrow \text{NHMe}\_3^{\cdot \text{ } \text{ } (aq)} + \text{OH}^-\_{(aq)} \tag{37}$$

## *2.4.7. Hydrothermal synthesis of AB site perovskites*

The study of the synthesis of solid solutions with simultaneous substitutions in both locations (A and B sites) has been conducted under combined hydrothermal conditions. In a preliminary study, the synthesis of complex perovskite compounds (PBSZ and PBZT) was investigated using a KOH solution with concentrations varying in the range of 1.0–3.0 M. The elements that were partially incorporated into the ABX3 structure were Pb, Ba and Sr added as A ions and Zr as B ion [125]. In the reaction media of KOH with concentrations of 2.0 and 3.0 M, most of the mixtures containing different contents of Ba and Sr were successfully crystallized and correspond to the tetragonal perovskite pure phase. However, the single perovskite phase was difficult to conduct in the alkaline solution with a concentration below 1 M. In conclusion, a hydrothermal media saturated with OH– ions is usually required to prepare perovskite ternary compounds under hydrothermal conditions. On the other hand, the synthesis of modified lead titanate powder such as (Pb0.88Sm0.08)(Ti0.99Mn0.01)O3 has been conducted at pH between 8 and 12 using NaOH as mineralizer and metal nitrates precursor solutions. The conditions that favour the crystallization of the perovskite structure are 290°C for 10 h at a pH = 10 [126]. In general, the vast number of studies discussed in the preliminary sections, regarding the synthesis of several groups of perovskite compounds, indicate that the processing parameters (precursor metal salts reagents, solvent media, pH of the precursor solution, reaction time, and temperature) must be carefully selected to carry on the processing perovskite compounds either by hydrothermal microwave-assisted synthesis, hydrothermal surfactant-assisted synthesis or supercritical water conditions.

### **2.5. Summary**

from a saturated solution. The co-precipitated colloid gel is obtained using an alkaline KOH solution with a concentration of 4 M producing a colloid suspension. The suspension contain‐ ing the co-precipitated colloid (Pb-Zr-Ti-hydroxide) hydrothermally treated in an autoclave at 250°C for 5 h leads to prepare the tetragonal structured PZT particles exhibiting a cubic morphology and particle size of 600 nm [119]. On the other hand, the production of single crystal tetragonal perovskite PZT (PbZr0.52Ti0.48O3) has been conducted adding a polymer compound. The process denominated as polymer-assisted hydrothermal synthesis was carried out at 200°C for 12 h. Among the polymer additives selected to control the micelles creation were PVA and PAA. The polymer plays an important role in the growth of PZT single crystals with tetragonal structure. The micelles tend to cover the faceted surfaces of the PZT particles, and this event produces a reduction in the superficial energy of the growing particles. Consequently, the PZT particles exhibit a preferential growth along the c axis of the tetragonal structure [120]. Besides the employment of water-soluble organic compounds aiming to control the particle crystal growth. A variety of organic bases provided another advantage for synthesizing PZT perovskites. The possibility to apply the organic bases as potential solid solvents in hydrothermal media was recently investigated in the preparation of PZT perovskite powders. The experiments were carried out in an alkali-free reaction media at 160°C for 72 h. The organic alkaline media selected to conduct the hydrothermal treatments was the tetrame‐ thylammonium hydroxide (TMAOH), this also operated as a pH-adjusting agent [121].

68 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Other solid solutions of interest are the BTZ (BaTi0.8Zr0.2O3) and the BZT (BaZr1–*x*Ti*x*O3). The synthesis of these compounds at mild conditions has not been studied in detail yet. In one study conducted to prepare BTZ perovskite, a stoichiometric powdered mixture of Ti and Zr hydroxides, which was added to a 6 M Ba(OH)2 solution. The barium saturated alkaline Ba(OH)2 solution was selected as a solvent for the hydrothermal treatments to adjust the slurry before the hydrothermal synthesis. In this reaction media it was possible to carry out the crystallization of BZT particles with cubic structure, spherical morphology and molecular level homogeneous composition [122]. Likewise, BZT hollow nanospheres with cubic structure were synthesized at 200°C for 24 h using a KOH solution with a concentration of 16 M. The formation of the particles occurred due to Ostwald ripening mechanism, and the excess of Ti in the fluid may change the reaction mechanism that favours the hollowing process [123]. Finally, the synthesis reaction for the formation of BZnT (BaZn1/3Ta1/3O3) was conducted in a reaction medium with a strong organic base (TMAOH), and the raw materials used were barium and zinc acetate, and tantalum oxalate solution. The reactions involved in the hydro‐ thermal synthesis of the BZnT perovskite particles are reactions 35 to 37. Initially, in the first stage of the reaction occurred the decomposition of the organic alkaline produces a reduction

nucleation of BZnT embryos occurred in the solution according to the chemical reaction Eq. (35). The decomposition of the TMAH to trimethylamine and methanol (Eq. 36) takes place at temperatures above 220°C in the hydrothermal vessel. The product trimethylamine is a weak alkali, and it has the same chemical reactivity such as that of the ammonia in an aqueous solution (Eq. 37), [124]. Hence, the low chemical stability of the organic alkaline might hinder the crystallization of the desired BZnT perovskite phase because the supersaturation state is

species at temperatures up to 150°C, simultaneously the

in the total concentration of OH–

difficult to reach in the reaction system.

This review describes in detail the state-of-the-art regarding the theoretical and practical aspects associated with the synthesis of a wide number of perovskite-structured compounds. The use of the thermodynamic and kinetic modelling has proved to be a powerful tool to optimize the conditions for perovskite compounds under hydrothermal conditions. These tools allow in a rational way to engineer the processing of smart perovskite compounds, in the most cost-effective approach and an environmental friendly way as well. One of the particular advantages of hydrothermal processing is associated with the preparation of monodispersed nanometer-sized particles exhibiting a control over their shape and size in addition to their chemical homogeneity. However, the models discussed in the present review are only applicable to a very narrow boundary of experimental conditions in which the heterogeneous hydrothermal reaction equilibriums are achieved, namely in the pH range from 0 to 14. The vast practical expertise gained from the past three decades in the synthesis of a wide number of perovskite-structured compounds using the hydrothermal technique indicates that some caution must be executed, when applying the proposed models to diverse perovskite systems. The results derived from the kinetics and thermodynamical models might vary depending on the reaction temperature, concentration of precursors and the alkalinity of the hydrothermal media. Therefore, more conscious analyses based on inorganic chemistry fundamentals coupled with chemical engineering and kinetic modelling might help to derive the adequate conditions to achieve controlled crystallization of particles with specific particle size and morphology. Although the present information regarding the hydrothermal synthesis of perovskite compounds, in terms of the theoretical and practical approaches, has been evalu‐ ated with a high statistical reproducibility in small-scale laboratory reactors, there is a potential field to explore when scaling up the process to larger industrial reactors, where mixing problems might frequently raise due to slow and less-efficient agitation systems. Mixinglimited precipitation rate is one of the problems commonly encountered in the scale-up of precipitation processes.

## **Author details**

Juan Carlos Rendón-Angeles1\*, Zully Matamoros-Veloza2 , Karla Lorena Montoya-Cisneros1 , Jorge López Cuevas1 and Kazumichi Yanagisawa3

\*Address all correspondence to: jcarlos.rendon@cinvestav.edu.mx

1 Research Institute for Advanced Studies of the NPI, Ceramic Department, Ramos Arizpe, Coahuila, México

2 Technological Institute of Saltillo, Metal-Mechanics Department, Saltillo, Coahuila, México

3 Research Laboratory of Hydrothermal Chemistry, Kochi University, Kochi, Japan

## **References**


[3] Kay HF, Bailey PC. Structure and properties of CaTiO3. Acta Crystallogr 1957;10:219– 26. DOI:10.1107/S0365110X57000675

applicable to a very narrow boundary of experimental conditions in which the heterogeneous hydrothermal reaction equilibriums are achieved, namely in the pH range from 0 to 14. The vast practical expertise gained from the past three decades in the synthesis of a wide number of perovskite-structured compounds using the hydrothermal technique indicates that some caution must be executed, when applying the proposed models to diverse perovskite systems. The results derived from the kinetics and thermodynamical models might vary depending on the reaction temperature, concentration of precursors and the alkalinity of the hydrothermal media. Therefore, more conscious analyses based on inorganic chemistry fundamentals coupled with chemical engineering and kinetic modelling might help to derive the adequate conditions to achieve controlled crystallization of particles with specific particle size and morphology. Although the present information regarding the hydrothermal synthesis of perovskite compounds, in terms of the theoretical and practical approaches, has been evalu‐ ated with a high statistical reproducibility in small-scale laboratory reactors, there is a potential field to explore when scaling up the process to larger industrial reactors, where mixing problems might frequently raise due to slow and less-efficient agitation systems. Mixinglimited precipitation rate is one of the problems commonly encountered in the scale-up of

, Karla Lorena Montoya-Cisneros1

,

precipitation processes.

Juan Carlos Rendón-Angeles1\*, Zully Matamoros-Veloza2

70 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

and Kazumichi Yanagisawa3

1 Research Institute for Advanced Studies of the NPI, Ceramic Department, Ramos Arizpe,

2 Technological Institute of Saltillo, Metal-Mechanics Department, Saltillo, Coahuila, México

[1] Mohammad HH. Characterisation of Mixed-Metal Oxides Prepared by Hydrother‐ mal Synthesis. [thesis]. Coventry CV4 7AL, United Kingdom: The University of War‐

[2] Mats J, Peter L. Crystallography and Chemistry of Perovskites. Handbook of Mag‐ netism and Advanced Magnetic Materials. p. 2007. DOI:

3 Research Laboratory of Hydrothermal Chemistry, Kochi University, Kochi, Japan

\*Address all correspondence to: jcarlos.rendon@cinvestav.edu.mx

**Author details**

Jorge López Cuevas1

Coahuila, México

**References**

wick; 2013.

10.1002/9780470022184.hmm411


vestigations of physical properties. Prog Solid State Chem 2009;37:173–205. DOI: 10.1016/j.progsolidstchem.2009.11.003

[29] David S. Related chemical themes in the synthesis of advanced ceramic materials. Mater Sci Eng 1989;A109:261–4. ISBN: 9780521424189

[15] Eduardo CCdeS, Reginaldo M. Properties and applications of perovskite proton con‐ ductors. Mater Res 2010;13(3):385–94. http://dx.doi.org/10.1590/

[16] Peña MA, Fierro JLG. Chemical structures and performance of perovskite oxides.

[17] Bonanos N, Knight KS, Ellis B. Perovskite solid electrolytes: structure, transport properties and fuel cell applications. Solid State Ionics 1995;79:161–70. DOI

[18] Stephen JS. Recent advances in perovskite-type materials for SOFC cathodes. Fuel

[19] Stephen JS. Recent advances in Perovskite-type materials for solid oxide fuel cell cathodes. Int J Inorg Mater 2001;3:113–21. DOI:10.1016/S1466-6049(01)00004-6

[20] Harlan UA. Review of p-type doped perovskite materials for SOFC and other appli‐ cations. Solid State Ionics 1992;52(1–3):33–41. DOI: 10.1016/0167-2738(92)90089-8

[21] Bangerta U, Falke U, Weidenkaff A. Ultra-high resolution EEL studies of domains in Perovskite. J Phys Conf Ser 2006;26:17–20. DOI:10.1088/1742-6596/26/1/004.

[22] Tapas KM, Gopalakrishnan J. From rocksalt to perovskite: a metathesis route for the synthesis of perovskite oxides of current interest. J Mater Chem 2004;14:1273–80.

[23] Hans-Conrad zL, Qingbiao Z, Daniel EB, Michael WC. 2H-perovskite related oxides: Synthesis, structures, and predictions. Cryst Eng Comm 2012;14:23–39. DOI: 10.1039/

[24] Jesús P-G, Rainer S, Emilio M. Microwave-assisted synthesis and characterization of perovskite oxides. In: Jinghua Z, Huan L, eds. Perovskite: Crystallography, Chemis‐ try and Catalytic Performance. Novascience Publishers, Hauppauge (USA) 2013. pp.

[25] Parkin IP, Solid state metathesis reaction for metal borides, silicides, pnictides and chalcogenides: ionic or elemental pathways. Chem Soc Rev 1996;25:199–207, DOI:

[26] Stanley MW. Hydrothermal synthesis of transition metal oxides under mild condi‐ tions. Solid State Mater Sci 1996;1[2]:227–32. DOI: 10.1016/S1359-0286(96)80089-1

[27] Thierry D, Ali A-Z, Berger M-H, Xavier B, Hochepied J-E. Hydrothermal synthesis of ferroelectric mixed potassium niobate–lead titanate nanoparticles. J Am Ceram Soc

[28] Stefan GE, Abicht H-P, Dronskowski R, Müller T, Reller A, Weidenkaff A. Perov‐ skite-related oxynitrides – recent developments in synthesis, characterisation and in‐

Chem Rev 2001;101:1981–2017. DOI: 10.1021/cr980129f

72 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Cells Bull 2001;4(33):6–12. DOI: 10.1016/S1464-2859(01)80254-6

S1516-14392010000300018

10.1016/0167-2738(95)00056-C

DOI: 10.1039/b315263d

10.1039/CS9962500199

117–140. DOI: 10.1002/chin.201424225

2014;97[5]:1456–64. DOI: 10.1111/jace.12778

c1ce05788j


[53] Ouchiyama N, Rough SL, Bridgwater J. Apopulation balance approach to describing bulk attrition. Chem Eng Sci 2005;60:1429–40. DOI:10.1016/j.ces.2004.08.037

[41] Malgorzata ML, Richard ER. Thermodynamics of the hydrothermal synthesis of cal‐ cium titanate with reference to other alkaline-earth titanates. Chem Mater 1995;7[1]:

[42] Malgorzata ML, Richard ER. Hydrothermal synthesis of Perovskite materials: Ther‐ modynamic modeling and experimental verification. Ferroelectrics 1994;151:159–64.

[43] Malgorzata ML, Erik N, Andrzej A, Richard ER. Hydrothermal synthesis of carbo‐ nate-free strontium zirconate: thermodynamic modeling and experimental verifica‐

[44] Malgorzata ML, Richard ER. Synthesis of lead titanate: thermodynamic modeling and experimental verification. J Am Ceram Soc 1993;76[10]:2649–59. DOI: 10.1111/j.

[45] Cho S-B, Noh J-S, Malgorzata ML, Richard ER. Low temperature hydrothermal syn‐ thesis and formation mechanisms of lead titanate (PbTiO3) particles using tetrame‐ thylammonium hydroxide: thermodynamic modelling and experimental verification.

[46] Gelabert MC, Laudise RA, Riman RE. Phase stability, solubility and hydrothermal crystal growth of PbTiO3. J Cryst Growth 1999;197[1–2]:195–203. DOI: 10.1016/

[47] James OE, Jr, Catherine CH-H, Bonnie LG, Malgorzata ML, Richard R. Kinectics and mechanism of hydrothermal synthesis of barium titanate. J Am Ceram Soc

[48] Bonnie LG, Malgorzata ML, Richard ER. Low-temperature hydrothermal synthesis of phase-pure (Ba,Sr)TiO3 perovskite using EDTA. J Am Ceram Soc 2004;87[11]:2025–

[49] Malgorzata ML, Andrzej A, Richard ER. Hydrothermal precipitation of lead zirco‐ nate titanate solid solutions: thermodynamic modeling and experimental synthesis. J

[50] Andrea T, Vincenzo B, Maria TB, Massimo V, Paolo N. Kinetic modeling of aqueous and hydrothermal synthesis of barium titanate (BaTiO3). Chem Mater 2005;17[21]:

[51] Andrea T, Maria TB, Vincenzo B, Massimo V, Carlo B, Paolo N. Kinetics and mecha‐ nism of aqueous chemical synthesis of BaTiO3 particles. Chem Mater 2004;16[8]:1536–

[52] Daniele LM. On the use of bi-variate population balance equations for modelling ba‐ rium titanate nanoparticle precipitation. Chem Eng Sci 2009;64:697–708. DOI:10.1016/

Am Ceram Soc 1995;78[10]:2609–18. DOI: 10.1111/j.1151-2916.1995.tb08030.x

J Eur Ceram Soc 2003;23:2323–35. DOI:10.1016/S0955-2219(03)00085-2

1996;79[11]:2929–39. DOI: 10.1111/j.1151-2916.1996.tb08728.x

32. DOI: 10.1111/j.1151-2916.2004.tb06355.x

5346–56. DOI: 10.1021/cm051119f

43. DOI: 10.1021/cm031130k

j.ces.2008.04.052

tion. Chem Mater 1997;9[5]:1116–25 DOI: 10.1021/cm960444n

18–25. DOI: 10.1021/cm00049a006

74 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

DOI: 10.1080/00150199408244737

1151-2916.1993.tb03994.x

S0022-0248(98)00914-2


[75] Xian-Zhi C, Zhong-Cheng Q, Jian-Ping Z, Gangqiang Z, Xiao-Bing B, Peng L. Largescale growth and shape evolution of bismuth ferrite particles with a hydrothermal method. Mater Chem Phys 2011;126:560–7. DOI:10.1016/j.matchemphys.2011.01.027

[65] Ohara Y, Koumoto K, Shimizu T, Yanagida H. Hydrotherrnal synthesis of fibrous lead titanate Powders. J Mater Sci 1995;30:263–6. DOI 10.1007/BF00352160

[66] Gang X, Xiaoqiang H, Vladimir K, Shuquan C, Xin Y, Chunying C, Ge S, Gaorong H. Hydrothermal synthesis of single-crystalline tetragonal perovskite PbTiO3 nano‐ sheets with dominant (001) or (111) facets. Cryst Eng Comm 2014;16:4373–6. DOI:

[67] Mario LM, Elaine CP, Gabriela SdN, Valeria ML, Julio RS, Valmor RM, Maria IBB, Juan A, José AV, Elson L. Structural and optical properties of CaTiO3 perovskitebased materials obtained by microwave-assisted hydrothermal synthesis: an experi‐ mental and theoretical insight. Acta Mater 2009;57:5174–85. DOI:10.1016/j.actamat.

[68] Tatiana MM, Mario LM, Ivo MP, Francini CP, Edson RL, Ieda LVR, José AV, Leinig AP, Elson L. CaTiO3:Eu3+ obtained by microwave assisted hydrothermal method: A photoluminescent approach. Opt Mater 2010;32:990–7. DOI:10.1016/j.optmat.

[69] Yonggang W, Gang X, Linlin Y, Zhaohui R, Xiao W, Wenjian W, Piyi D, Ge S, Gaor‐ ong H. Formation of single-crystal SrTiO3 dendritic nanostructures via a simple hy‐ drothermal method. J Cryst Growth 2009;311:2519–23. DOI:10.1016/j.jcrysgro.

[70] Rangel-Hernandez YM, Rendón-Angeles JC, Matamoros-Veloza Z, Pech-Canul MI, Diaz-de la ST, Yanagisawa K. One-step synthesis of fine SrTiO3 particles using SrSO4 ore under alkaline hydrothermal conditions. Chem Eng 2009;155:483–92. DOI:

[71] Xiao W, Gang X, Zhaohui R, Ge S, Gaorong H. Effect of KOH concentration on the phase and morphology of hydrothermally synthesized Pb0.70La0.30TiO3 fine powders.

[72] Jean-François T, Sophie d'A, Christian C, Philippe C, Mohamed R, Anne L. Influence of hydrothermal synthesis conditions on BNT-based piezoceramics. J Eur Ceram Soc

[73] Jianhong P, Mirabbos H, Baowei C, Juan W, Yunhua X. Surfactant-free hydrothermal synthesis of submicron BiFeO3 powders. Appl Phys A 2011;103:511–6. DOI 10.1007/

[74] Xin Y, Gang X, Zhaohui R, Xiao W, Chunying C, Siyu G, Ge S, Gaorong H. The hy‐ drothermal synthesis and formation mechanism of single-crystalline perovskite Bi‐ FeO3 microplates with dominant (012) facets. Cryst Eng Comm 2014;16:4176–82. DOI:

Mater Lett 2008;62:3719–21. DOI:10.1016/j.matlet.2008.04.039

2011;31:1997–2004. DOI:10.1016/j.jeurceramsoc.2011.04.025

10.1039/c4ce00234b

76 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

2009.07.019

2010.01.039

2009.01.103

10.1016/j.cej.2009.07.024

s00339-010-6024-2

10.1039/c3ce42488j


[99] Zhang T, Jin CG, Qian T, Lu XL, Bai JM, Li XG. Hydrothermal synthesis of singlecrystalline La0.5Ca0.5MnO3 nanowires at low temperature. J Mater Chem 2004;14:2787– 9. DOI: 10.1039/b405288a

[87] Biasotto GB, Simoes AZ, Foschini CR, Zaghete MA, Varela JA, Longo E. Microwavehydrothermal synthesis of perovskite bismuth ferrite nanoparticles. Mater Res Bull

[88] Zhi W, Wenfei X, Hui P(a), Xiaodong T(b). Polyanion modulated evolution of perov‐ skite BiFeO3 microspheres to microcubes by a microwave assisted hydrothermal

[89] Millot N, Xin B, Pighini C, Aymes D. Hydrothermal synthesis of nanostructured in‐ organic powders by a continuous process under supercritical conditions. J Eur Ce‐

[90] Aimable A, Xin B, Millot N, Aymes D. Continuous hydrothermal synthesis of nano‐ metric BaZrO3 in supercritical water. J Solid State Chem 2008;181:183–9. DOI:

[91] Zhouguang L, Yougen T, Limiao C, Yadong L. Shape-controlled synthesis and char‐ acterization of BaZrO3 microcrystals. J Cryst Growth 2004;266:539–44. DOI: 10.1016/

[92] Jia G, Guo Z, Jianguo Y, Yongcun Z, Guodong L, Guangshan Z, Yongnan Z. Fabrica‐ tion and luminescence properties of RbxBa1-xZrO3 hollow nanospheres. Mater Lett

[93] Wenjun Z, Wenqin P, Guangyao M, Dingkun P. Hydrothermal synthesis and charac‐ terization of LaCrO3. J Mater Chem 1999;9:2833–6. DOI: 10.1039/A904399C

[94] Bo HC, Shin-Ae P, Bong KP, Ho HC, Yong-Tae K. Controlled synthesis of La1 xSrxCrO3 nanoparticles by hydrothermal method with nonionic surfactant and their ORR activity in alkaline médium. Mater Res Bull 2013;48:3651–6. http://dx.doi.org/

[95] Rendón-Angelesa JC, Yanagisawa K, Matamoros-Veloza Z, Pech-Canula MI, Men‐ dez-Nonell J, Diaz-de la ST. Hydrothermal synthesis of perovskite strontium doped lanthanum chromite fine powders and its sintering. J Alloys Compd 2010;504:251–6.

[96] Shan W, Keke H, Beining Z, Jiaqi Z, Shouhua F. Mild hydrothermal synthesis and physical property of perovskite Sr doped LaCrO3. Mater Lett 2013;101:86–9. http://

[97] Kripasindhu S, Martin RL, Reza JK, Jeremy S, Richard IW. Direct hydrothermal syn‐ thesis and physical properties of rare-earth and yttrium orthochromite perovskites.

[98] Deliang Z, Hong Z, Yuheng Z. Hydrothermal synthesis of La0.5Ba0.5MnO3 nanowires.

method. J Mater Res 2013;28[11]:1498–504. DOI: 10.1557/jmr.2013.130

ram Soc 2005;25:2013–6. DOI: 10.1016/j.jeurceramsoc.2005.03.202

2014;116:164–6. http://dx.doi.org/10.1016/j.matlet.2013.11.002

2011;46:2543–7. DOI:10.1016/j.materresbull.2011.08.010

78 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

10.1016/j.jssc.2007.11.015

j.jcrysgro.2004.02.107

10.1016/j.materresbull.2013.04.084

DOI: 10.1016/j.jallcom.2010.05.103

dx.doi.org/10.1016/j.matlet.2013.03.083

Chem Mater 2011;23:48–56. DOI: 10.1021/cm102925z

Appl Phys Lett 2002;80[9]:1634–6. DOI: 10.1063/1.1455690#


synthesis. J Cryst Growth 2007;305[1]:242–8. http://dx.doi.org/10.1016/j.jcrysgro. 2007.03.052


[123] Xiaojuan W, Juan C, Zhiqiang S, Hua H, Lang L, Jianguo Y, Yongcun Z, Guodong L, Yongnan Z. Hydrothermal synthesis and photoluminescence properties of BaZr1 xTixO3 hollow nanospheres. Mater Lett 2012;86:21–4. http://dx.doi.org/10.1016/ j.matlet.2012.07.035

synthesis. J Cryst Growth 2007;305[1]:242–8. http://dx.doi.org/10.1016/j.jcrysgro.

[111] Kongjun Z, Yang C, Xiaohui W, Lin B, Jinhao Q, Hongli J. Hydrothermal synthesis of sodium niobate with controllable shape and structure. Cryst Eng Comm 2012;14:411–

[112] Wu SY, Zhang W, Chen XM. Formation mechanism of NaNbO3 powders during hy‐ drothermal synthesis. J Mater Sci Mater Electronics 2010;21[5]:450–5. DOI: 10.1007/

[113] Gregory KLG, Fred FL, Sossina MH, Carlos GL, Gregory KLG, Fred FL, Sossina MH, Carlos GL. Hydrothermal synthesis of KNbO3 and NaNbO3 powders. J Mater Res

[114] Haibo C, Mingyu S, Chenyang Z, Hongming Y, Shouhua F. Hydrothermal syntheses and structural phase transitions of AgNbO3. J Am Ceram Soc 2012;95[11]:3673–77.

[115] Gregory KLG, Sossina MH, Carlos GL, Fred FL. Hydrothermal synthesis of perov‐ skite and pyrochlore powders of potassium tantalate. J Mater Res 2002;17:3168–76.

[116] Jingying S, Guiji L, Nan W, Can L. Microwave-assisted hydrothermal synthesis of perovskite NaTaO3 nanocrystals and their photocatalytic properties. J Mater Chem

[117] Kutty TRN, Balachandran R. Direct precipitation of lead zirconate titanate by the hy‐

[118] Anjali AA, Malini SB. Synthesis of nanocrystalline PZT by hydrothermal method.

[119] Ralph N, Norbert BH. Hydrothermal synthesis of nanocrystalline perovskite powder

[120] Gang X, Zhaohui R, Piyi D, Wenjian WE, Ge S, Gaoirong H. Polymer-assisted hydro‐ thermal synthesis of single-crystalline tetragonal perovskite PbZr0.52Ti0.48O3 nano‐

[121] Ivan P, Malgorzata ML, Andrzej A, Richard ER. Hydrothermal synthesis of lead zirc‐ onate titanate (PbZr0.52Ti0.48O3) using organic mineralizers. In: Applications of Ferro‐ electrics, 1996. ISAF '96, Proceedings of the Tenth IEEE International Symposium on (Volume:2); 18–21 Aug 1996; East Brunswick, NJ: ieee; DOI: 10.1109/ISAF.

[122] Fukai K, Hidaka K, Aoki M, Abe K. Preparation and properties of uniform fine per‐ ovskite powders by hydrothermal synthesis. Ceram Int 1990;16:285–90. DOI:

wires. Adv Mater 2005;17[7]:907–10. DOI: 10.1002/adma.200400998

2007.03.052

6. DOI: 10.1039/C1CE06100C

2003;18[2]:338–45. DOI: 10.1557/JMR.2003.0044

80 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

DOI: 10.1111/j.1551-2916.2012.05392.x

2012;22:18808–13. DOI: 10.1039/c2jm33470d

systems. Mol Cyst Liq Cfysr 2000;353:329–40.

drothermal method. Mat Res Bull 1984;19:1479–88.

DOI: 10.1557/JMR.2002.0458

Defence Sci J 2007;57[1]:35–9.

10.1016/0272-8842(90)90041-D

1996.598129

s10854-009-9937-9


## **Fabrication of Yttrium-Doped Barium Zirconate for High Performance Protonic Ceramic Membranes**

W. Grover Coors, Anthony Manerbino, David Martinefski and Sandrine Ricote

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61660

## **Abstract**

Barium zirconate has emerged as the leading candidate material for fabricating dense ce‐ ramic membranes for hydrogen separation. B-sites in the ABO3 perovskite are acceptordoped with a +3 cation – most commonly yttrium – charge-compensated by the formation of oxygen ion vacancies in the lattice. A minor fraction of B-sites can be filled with cerium to give BaZr0.9-xCexY0.1O3-d, x ≤ 0.2. Upon hydration at elevated temperatures, weaklybound protons are formed in the lattice. This produces a cubic perovskite ceramic proton conductor useful in diverse applications, such as protonic ceramic fuel cells, electrolysers, and catalytic membrane reactors operating at temperatures between 600 and 800 °C. A necessary requirement for fabricating thin ceramic membranes for proton diffusion is to maximize grain size while eliminating percolating porosity. However, high-density, large-grained barium zirconate is a very difficult material to prepare by traditional pow‐ der sintering methods. This chapter describes a new methodology for making protonic ceramic membranes with large grains and virtually no residual porosity. This discovery has the potential to have a profound impact on energy conversion efficiency of the vari‐ ous membrane devices envisioned for the coming hydrogen energy economy.

**Keywords:** Protonic ceramics, Yttrium-doped barium zirconate, Hydrogen separation membranes, Solid-state reactive sintering

## **1. Introduction**

Almost thirty-five years ago, Hiroyasu Iwahara reported high temperature proton conduction in strontium cerate perovskite [1]. In that same year, 1981, there was sufficient interest in the potential for proton conductors to warrant the first international symposium on solid-state proton conductors in Paris (SSPC-I). Interestingly, the possibility of proton transport in

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

perovskite ceramics was not widely recognized at this time. Several more years would pass before Iwahara's ground-breaking discoveries would be appreciated. Even at the fourth international symposium in 1989 (SSPC-IV) there was still general scepticism about proton conduction in perovskites [2]. It was not really until the early 1990's that the commercial importance of ceramic proton conductors sparked widespread interest among researchers. Acceptor-doped barium cerate soon became the model ceramic proton conductor. It exhibited high proton conductivity and was easy to fabricate into testable specimens, but before too long it was realized that barium cerate lacked the necessary chemical stability in the hot, moist, carbon dioxide-containing atmospheres of the typical use environment.

The first significant advancement into the modern era of ceramic proton conductors came from Klaus-Dieter Kreuer at the Max Planck Institute in 1999 [3] with the identification of the ceramic perovskite, barium zirconate, as a better candidate material due to its much greater chemical stability. However, unlike barium cerate, barium zirconate proved to be difficult to sinter by traditional ceramic methods. Some successes were achieved with solid solutions of barium cerate and zirconate [4], but still, it remained a challenge to make dense ceramic with more than 50% zirconium on B-sites, and any less than about 60% zirconium was still chemically unstable.

The second major breakthrough for barium zirconate ceramic processing came from Babilo and Haile at Caltech in 2005 [5] with the discovery of solid-state reactive sintering. Rather than using pre-reacted calcine powder, fully dense, large-grained barium zirconate ceramic membranes were fabricated directly from precursor powders with a small addition of NiO as a sintering additive. In this process, unreacted Ba from BaSO4 and NiO form a binary eutectic glass that simultaneously promotes solid-state reaction and sintering. As the solid-state reaction proceeds, barium is extracted from the glass phase to take up A-sites in the more stable ABO3 perovskite. Ultimately, the BaO-NiO glass becomes sufficiently barium deficient that the glass freezes out, leaving behind a thin film of NiO, which coats the grain boundaries [6].

In bulk ceramic material prepared this way, a small amount of NiO as a second phase has practically no impact, but when exposed to reducing atmosphere at high temperatures, the NiO nucleates to form Ni nanoparticles that decorate the dihedrals and grain facets. Under extreme operating conditions, depending on size and density, these nanoparticles can result in mechanical strain and fractures along grain boundaries. The formation of Ni nanoparticles hinders the use of membranes made by solid state reactive sintering in reducing atmosphere. Other sintering additives, which do not form nanoparticles, have been tried without much success. The eutectic temperature of BaO-NiO turns out to be almost perfect for the simulta‐ neous solid-state reaction and liquid phase sintering process.

It became necessary to discover a way to remove the residual NiO from ceramic specimens after sintering. Such a process was discovered only last year, and will be described in what follows for barium zirconate-based materials with the nominal formula, BaZr0.9-xCexY0.1O3-d, where x ≤ 0.2. The term BZY will be used to refer to the ceramic formulation with more than 70 mol % Zr.

## **2. Background on barium zirconate**

perovskite ceramics was not widely recognized at this time. Several more years would pass before Iwahara's ground-breaking discoveries would be appreciated. Even at the fourth international symposium in 1989 (SSPC-IV) there was still general scepticism about proton conduction in perovskites [2]. It was not really until the early 1990's that the commercial importance of ceramic proton conductors sparked widespread interest among researchers. Acceptor-doped barium cerate soon became the model ceramic proton conductor. It exhibited high proton conductivity and was easy to fabricate into testable specimens, but before too long it was realized that barium cerate lacked the necessary chemical stability in the hot, moist,

The first significant advancement into the modern era of ceramic proton conductors came from Klaus-Dieter Kreuer at the Max Planck Institute in 1999 [3] with the identification of the ceramic perovskite, barium zirconate, as a better candidate material due to its much greater chemical stability. However, unlike barium cerate, barium zirconate proved to be difficult to sinter by traditional ceramic methods. Some successes were achieved with solid solutions of barium cerate and zirconate [4], but still, it remained a challenge to make dense ceramic with more than 50% zirconium on B-sites, and any less than about 60% zirconium was still chemically

The second major breakthrough for barium zirconate ceramic processing came from Babilo and Haile at Caltech in 2005 [5] with the discovery of solid-state reactive sintering. Rather than using pre-reacted calcine powder, fully dense, large-grained barium zirconate ceramic membranes were fabricated directly from precursor powders with a small addition of NiO as a sintering additive. In this process, unreacted Ba from BaSO4 and NiO form a binary eutectic glass that simultaneously promotes solid-state reaction and sintering. As the solid-state reaction proceeds, barium is extracted from the glass phase to take up A-sites in the more stable ABO3 perovskite. Ultimately, the BaO-NiO glass becomes sufficiently barium deficient that the glass freezes out, leaving behind a thin film of NiO, which coats the grain boundaries [6].

In bulk ceramic material prepared this way, a small amount of NiO as a second phase has practically no impact, but when exposed to reducing atmosphere at high temperatures, the NiO nucleates to form Ni nanoparticles that decorate the dihedrals and grain facets. Under extreme operating conditions, depending on size and density, these nanoparticles can result in mechanical strain and fractures along grain boundaries. The formation of Ni nanoparticles hinders the use of membranes made by solid state reactive sintering in reducing atmosphere. Other sintering additives, which do not form nanoparticles, have been tried without much success. The eutectic temperature of BaO-NiO turns out to be almost perfect for the simulta‐

It became necessary to discover a way to remove the residual NiO from ceramic specimens after sintering. Such a process was discovered only last year, and will be described in what follows for barium zirconate-based materials with the nominal formula, BaZr0.9-xCexY0.1O3-d, where x ≤ 0.2. The term BZY will be used to refer to the ceramic formulation with more than

neous solid-state reaction and liquid phase sintering process.

carbon dioxide-containing atmospheres of the typical use environment.

84 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

unstable.

70 mol % Zr.

Over the past ten years, the perovskite ceramic, yttrium-doped barium zirconate has emerged as the most promising ceramic proton conductor [7-14]. In the ABO3 structure, barium occupies A-sites and zirconium occupies B-sites. Yttrium (Y+3) substitutes for Zr+4 on B-sites as an acceptor dopant that is compensated by the formation of oxygen ion vacancies. These vacancies on the oxygen ion sublattice have the remarkable property that water from the gas phase enters the oxygen vacancies at elevated temperatures, donating two quasi-free protons to the lattice (Eq 1) [15-18]. This behaviour requires just the right oxygen ion separation, and is unique to certain oxide ceramics, such as perovskites, and does not occur in more densely packed structures like yttrium-doped zirconia. Hydrogen transport through the ceramic is by means of protons rather than gaseous diffusion. Protons are introduced into the lattice when water vapour reacts with an oxygen ion vacancy at the surface by the hydration reaction (In Kröger-Vink notation),

$$2H\_2O\text{(g)} + V\_{\text{O}}^{\bullet \bullet} + O\_{\text{O}}^{\times} \rightarrow 2OH\_{\text{O}}^{\bullet} \tag{1}$$

One of the necessary requirements for ceramic proton conductors is gas-tightness against hydrogen. This places a new set of constraints on polycrystalline ceramic fabrication that goes well beyond the simple idea of "density". Ceramics processed by conventional means are often called dense, but fail the requirement for gas-tightness, particularly when thin membranes are used at elevated temperatures. This is because traditional sintering involves removal of the pores that form between adjacent ceramic grains during sintering. During intermediate-stage sintering, the dihedral pores actually form a series of interconnected channels that become easy paths for diffusion of molecular hydrogen and other gases. It is only during the final stage of sintering that the continuous pore channels are pinched off leaving a dihedral pore at each vertex of adjoining grains. These ideas are well-treated in many textbooks on ceramic sintering such as Barsoum [19]. Depending on the refractoriness of the ceramic material, these dihedral pores are nearly impossible to eliminate entirely, so the matter of ceramic density is more a qualitative determination of when the ceramic becomes "dense enough". It is the scattering of light from these pores that is the primary reason that oxide ceramics are typically opaque. Ceramic materials, such as magnesium aluminate spinel, can be made transparent with the use of sintering aids and high sintering pressures and temperatures, which effectively squeeze out the dihedral pores [20]. Transparency is a very good indication of high density. Even though "dense enough" in the case of ceramic membranes typically means that there are no pore channels that percolate through the matrix, the best evidence for density is transparency.

Barium zirconate is a very difficult-to-sinter material at reasonable temperatures. The refrac‐ toriness of barium zirconate is the main reason why adoption of the material as a ceramic proton conductor has been so slow. Techniques, such as hot isostatic pressing [21] and spark plasma sintering [21-23] have been explored with limited success. The most common approach to processing such materials is to begin with very high purity nanopowders prepared by solgel or spray pyrolysis methods [24-25]. This approach has worked well for fabricating membranes dense enough for hydrogen separation applications, but fine-grained ceramic has more grain boundaries perpendicular to the direction of proton transport than coarse-grained ceramic. The core region between ceramic oxide grains in polycrystalline ceramics tend to become positively charged. This is not a problem for anion conductors, such as ZrO2 and CeO2 [26], but for proton conductors, the positive grain boundary charge serves to repel the positively charged protons, resulting in significant intrinsic grain boundary impedance [27-30]. Large grain size, on the other hand, extends the low temperature range where suitable operation is possible by decreasing the total grain boundary resistance.

For optimal proton conductivity it is desirable to obtain a ceramic microstructure with the largest possible grain size to reduce the effect of grain boundary impedance. Grain growth in barium zirconate during final-stage sintering is very slow, so simply sintering nanosized material for a longer time is not a viable option. Using micron-size calcined powders is also not an option. Traditionally, ceramic powder with the desired phase is prepared beforehand by solid-state reaction of the precursor oxides. This is how the ceramic powders were prepared for the sintered specimen shown in Fig. 1. Stoichiometric quantities of ZrO2, Y2O3, and BaSO4 were blended and reacted together in a covered crucible to make phase-pure BaZr0.9Y0.1O3-d (BZY10) powder, which was subsequently milled to about one micron size following the traditional method for preparing ceramic materials. Organic binder was added to the powder, pressed into a pellet under high pressure to ensure high green bulk density, and then sintered in air at 1650 °C for up to 24 hours. It is obvious that ceramic prepared this way is not suitable for application as a hydrogen separation membrane.

**Figure 1.** Poor sinterability of micron-size pre-calcined BZY10 powder at 1650 °C for 24 hours in air (Courtesy of Wade Rosensteel, Colorado School of Mines).

An entirely new approach to fabrication of membrane quality barium zirconate ceramic, called solid-state reactive sintering (SSRS), has been developed [5-6, 31-33]. In this method the precursor powders are reacted and sintered in the same operation so that the perovskite phase is produced *in situ* and grain boundary movement never occurs. This results in large grain size and the absence of dihedral pores as shown in Fig. 2. This process will be explained in detail in Section 3.

**Figure 2.** Low magnification bright-field TEM of dihedral in as-fired BaZr0.7Ce0.2Y0.1O3-d (Courtesy of Daniel Clark, Col‐ orado School of Mines).

## **3. Solid state reactive sintering**

membranes dense enough for hydrogen separation applications, but fine-grained ceramic has more grain boundaries perpendicular to the direction of proton transport than coarse-grained ceramic. The core region between ceramic oxide grains in polycrystalline ceramics tend to become positively charged. This is not a problem for anion conductors, such as ZrO2 and CeO2 [26], but for proton conductors, the positive grain boundary charge serves to repel the positively charged protons, resulting in significant intrinsic grain boundary impedance [27-30]. Large grain size, on the other hand, extends the low temperature range where suitable

For optimal proton conductivity it is desirable to obtain a ceramic microstructure with the largest possible grain size to reduce the effect of grain boundary impedance. Grain growth in barium zirconate during final-stage sintering is very slow, so simply sintering nanosized material for a longer time is not a viable option. Using micron-size calcined powders is also not an option. Traditionally, ceramic powder with the desired phase is prepared beforehand by solid-state reaction of the precursor oxides. This is how the ceramic powders were prepared for the sintered specimen shown in Fig. 1. Stoichiometric quantities of ZrO2, Y2O3, and BaSO4 were blended and reacted together in a covered crucible to make phase-pure BaZr0.9Y0.1O3-d (BZY10) powder, which was subsequently milled to about one micron size following the traditional method for preparing ceramic materials. Organic binder was added to the powder, pressed into a pellet under high pressure to ensure high green bulk density, and then sintered in air at 1650 °C for up to 24 hours. It is obvious that ceramic prepared this way is not suitable

**Figure 1.** Poor sinterability of micron-size pre-calcined BZY10 powder at 1650 °C for 24 hours in air (Courtesy of Wade

operation is possible by decreasing the total grain boundary resistance.

86 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

for application as a hydrogen separation membrane.

Rosensteel, Colorado School of Mines).

## **3.1. Mechanism of the solid-state reactive sintering**

The idea behind reactive sintering is quite simple. However, the ability to execute SSRS depends on identifying a glassy phase that consists of one of the components of the final desired ceramic phase. In the case of BZY, the A-site constituent, BaO, forms a binary eutectic glass with nickel oxide with a low melting point of 1125 °C. The BaO-NiO binary phase diagram is shown in Fig. 3. In solid-state reactive sintering, stoichiometric quantities of BaSO4, ZrO2, CeO2 and Y2O3 are mixed together with a small amount of NiO – typically 0.1 wt% with organic binders to form a green body. During the initial heating cycle, NiO reacts with BaSO4, releasing SO2, and forming a liquid at the eutectic composition. As the temperature increases, ZrO2 and Y2O3 begin to react together with Ba from the glass to form the more stable BZY perovskite phase. As more and more of the BZY phase forms, the BaO-NiO glass becomes increasingly NiO rich, following the composition of the liquidus upwards to the right. Once the final sintering temperature is reached at about 1600 °C, the glass freezes at a composition of about 80-90% NiO and liquid phase sintering ceases. The last remaining Ba is extracted more slowly from the glass by solid-state diffusion and only NiO is left behind as a second phase [6].

As discussed in more details in Section 4.1., BaSO4 is used instead of BaCO3 because it is insoluble in water. Solid state reactive sintering can also be performed with BaCO3.

**Figure 3.** Binary phase diagram for BaO-NiO. (TDnucl – Thermodata nuclear database, FactSage).

The resulting BZY is phase pure, but the amount of residual NiO second phase, depends on the starting amount. Too much NiO leads to inclusions as shown in Fig. 4, but too little leads to incomplete BZY sintering. Ideally, there should be just enough NiO to coat the grain boundaries of the BZY phase with no more than about one monolayer of Ni atoms. Face centered cubic (FCC) NiO has a lattice parameter of 4.218 Å, which is a good match for BZY, 4.240 Å. It is straightforward to estimate the optimal amount of NiO if the BZY grains are treated as volume-filling truncated octahedra. The volume (V) of a 5 μm truncated octahedron is 8√2 *s*<sup>3</sup> , where *s* is the side length (3*s* is the side length of the regular octahedron). The distance between opposing square faces is 2√2*s* ≈ 5 μm. Therefore, V = 62.5 μm3 . The surface area of the same truncated octahedron is A = (6+12√2) *s*<sup>2</sup> = 83.7 μm2 . The thickness of one monolayer of NiO (body diagonal of the fcc lattice) is 0.6 nm, giving a volume of NiO of 0.05 μm3 per 5 μm BZY grain. The volume ratio is, thus, 0.08%. The density ratio of NiO to BZY is 6.67/6.16 = 1.08, so the amount of NiO required to coat 5 μm BZY gains with a monolayer of NiO is slightly less than 0.1 wt%. Any more than this leads to inclusions, as shown in Fig. 4, and any less is not sufficient to promote complete sintering. This calculation assumes that no NiO dissolves in the BZY crystallites. There is some evidence suggesting that NiO may enter the lattice as interstitial ions or even substitute on B-sites, but the amounts are negligible. Most of the NiO remains at grain boundaries. When just the right amount of NiO is used, the grain boundaries appear very clean, as shown in Fig. 5.

**Figure 4.** Back-scattered electron micrograph of polished BZY10 prepared by SSRS with 1 wt% NiO after specimen re‐ duction. Dark regions are metallic nickel inclusions.

## **3.2. Ni nanoparticles**

NiO rich, following the composition of the liquidus upwards to the right. Once the final sintering temperature is reached at about 1600 °C, the glass freezes at a composition of about 80-90% NiO and liquid phase sintering ceases. The last remaining Ba is extracted more slowly from the glass by solid-state diffusion and only NiO is left behind as a second phase [6].

As discussed in more details in Section 4.1., BaSO4 is used instead of BaCO3 because it is

insoluble in water. Solid state reactive sintering can also be performed with BaCO3.

88 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 3.** Binary phase diagram for BaO-NiO. (TDnucl – Thermodata nuclear database, FactSage).

between opposing square faces is 2√2*s* ≈ 5 μm. Therefore, V = 62.5 μm3

NiO (body diagonal of the fcc lattice) is 0.6 nm, giving a volume of NiO of 0.05 μm3

BZY grain. The volume ratio is, thus, 0.08%. The density ratio of NiO to BZY is 6.67/6.16 = 1.08, so the amount of NiO required to coat 5 μm BZY gains with a monolayer of NiO is slightly less than 0.1 wt%. Any more than this leads to inclusions, as shown in Fig. 4, and any less is

same truncated octahedron is A = (6+12√2) *s*<sup>2</sup>

is 8√2 *s*<sup>3</sup>

The resulting BZY is phase pure, but the amount of residual NiO second phase, depends on the starting amount. Too much NiO leads to inclusions as shown in Fig. 4, but too little leads to incomplete BZY sintering. Ideally, there should be just enough NiO to coat the grain boundaries of the BZY phase with no more than about one monolayer of Ni atoms. Face centered cubic (FCC) NiO has a lattice parameter of 4.218 Å, which is a good match for BZY, 4.240 Å. It is straightforward to estimate the optimal amount of NiO if the BZY grains are treated as volume-filling truncated octahedra. The volume (V) of a 5 μm truncated octahedron

, where *s* is the side length (3*s* is the side length of the regular octahedron). The distance

= 83.7 μm2

. The surface area of the

per 5 μm

. The thickness of one monolayer of

The small amount of NiO left behind at grain boundaries from SSRS has no practical effect as long as the ceramic is only exposed to oxidizing environments. But this defeats the purpose of a ceramic membrane intended for use in reducing atmosphere. When exposed to hydrogen at elevated temperatures, the NiO gets reduced. Upon reduction, metallic nickel nucleates along the grain boundaries and coarsens into nanoparticles that range in size from a few nanometers to 100 nm, depending on time and temperature. Fig. 6 shows the Ni-nanoparticles that form when a specimen of BZY prepared by SSRS with only 0.1 wt% NiO is annealed in hydrogen

**Figure 5.** High resolution bright field TEM of as-fired BaZr0.7Ce0.2Y0.1O3-d prepared by SSRS with 0.1 wt% NiO. Grain boundary region ≈ 5.8 Å (Courtesy of Daniel Clark, Colorado School of Mines).

at 1000 °C for 24 hours. It is interesting to note that the Ni-nanoparticles that grow on the crystal facets are all about the same size and uniformly distributed. This is reminiscent of star formation in a galaxy and provides clear evidence that the Ni-nanoparticles nucleated from a uniform coating of NiO between BZY grains. Using the calculation above for the volume of NiO per square micron of grain facet - 0.6 x 106 nm3 – the volume of equivalent Ni metal is about 75% or about 0.45x106 nm3 /μm2 . A spherical 50 nm Ni particle has a volume of about 65,000 nm3 , so about 7 Ni nanoparticles per square micron of grain facet would be expected, which is about what is observed in Fig. 6.

One of the complications that arises in these protonic ceramic membranes is that the formation of Ni nanoparticles is irreversible. When the ceramic is exposed again to oxidizing conditions, rather than recoating the grain boundaries with a monolayer of NiO, the nanoparticles try to reoxidize in place, with a corresponding volume increase. This introduces local strain, which can lead to catastrophic failure. Fig. 7 shows the fracture that is characteristic of this condition. It is necessary to extract the NiO from the ceramic prior to reduction to levels below which fracture does not occur.

## **4. Nickel extraction**

## **4.1. Specimen preparation**

Barium zirconate and barium cerate can be reacted to make a complete solid solution over the entire composition range [4,6,9,14]. The most common nomenclature in use in the literature is Fabrication of Yttrium-Doped Barium Zirconate for High Performance Protonic Ceramic Membranes http://dx.doi.org/10.5772/61660 91

**Figure 6.** FESEM of Ni nanoparticles decorating grain boundary after reduction of BaZr0.7Ce0.2Y0.1O3-d without NiO ex‐ traction heat-treatment.

at 1000 °C for 24 hours. It is interesting to note that the Ni-nanoparticles that grow on the crystal facets are all about the same size and uniformly distributed. This is reminiscent of star formation in a galaxy and provides clear evidence that the Ni-nanoparticles nucleated from a uniform coating of NiO between BZY grains. Using the calculation above for the volume of

**Figure 5.** High resolution bright field TEM of as-fired BaZr0.7Ce0.2Y0.1O3-d prepared by SSRS with 0.1 wt% NiO. Grain

One of the complications that arises in these protonic ceramic membranes is that the formation of Ni nanoparticles is irreversible. When the ceramic is exposed again to oxidizing conditions, rather than recoating the grain boundaries with a monolayer of NiO, the nanoparticles try to reoxidize in place, with a corresponding volume increase. This introduces local strain, which can lead to catastrophic failure. Fig. 7 shows the fracture that is characteristic of this condition. It is necessary to extract the NiO from the ceramic prior to reduction to levels below which

Barium zirconate and barium cerate can be reacted to make a complete solid solution over the entire composition range [4,6,9,14]. The most common nomenclature in use in the literature is

nm3

, so about 7 Ni nanoparticles per square micron of grain facet would be expected,

– the volume of equivalent Ni metal is

. A spherical 50 nm Ni particle has a volume of about

NiO per square micron of grain facet - 0.6 x 106

which is about what is observed in Fig. 6.

nm3

boundary region ≈ 5.8 Å (Courtesy of Daniel Clark, Colorado School of Mines).

90 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

/μm2

about 75% or about 0.45x106

fracture does not occur.

**4. Nickel extraction**

**4.1. Specimen preparation**

65,000 nm3

BZCY*mn*, where *m* and *n* refer to the mole fractions of Zr and Ce on B-sites (times 10), respectively. For example, the formula for BZCY72 is BaZr0.7Ce0.2Y0.1O3-d. When there is no ceria, the common designation is BZY10, signifying 10 mol% yttrium. A small amount of ceria - up to 20 mol% - is advantageous in some cases. The material becomes chemically unstable in harsh environments for Ce > 20 mol% [34-35]. Furthermore, the electronic conductivity in dry hydrogen can be significant for Ce > 10 mol% [36]. In this monograph, the term, BZY, will be used for any ceramic formulation with more than 70 mol% Zr since structurally, there is practically no difference.

Solid state reactive sintering works with almost any ceramic green forming technique, so it is suitable for making everything from extruded tubes to cast tape. The method described here is for making 20 mm diameter by 1 mm thick discs by the time-honored method of slip casting. Slip casting is very easy and does not require expensive tooling. The discs are cast by simply filling PVC rings, arranged on a plaster-of-paris plate, with slip. The slip is prepared by mixing stoichiometric amounts of BaSO4, ZrO2, CeO2, and Y2O3 with additional 0.1 wt% NiO, DI water and conventional water-soluble, slip-casting binders. The reason for using BaSO4 rather than BaCO3 is that barium sulfate is insoluble in water. It is important in slip casting that no barium (or any cation) diffuse from the slip into the plaster-of-paris mold or base, thus altering the stoichiometry. Also, for health and safety reasons, it is inadvisable to use a water-soluble form of barium due to its toxicity. The slip is mixed in a Nalgene bottle on a jar roller for 2 hours with 10 mm zirconia balls. In reactive sintering, there is no need to adjust the particle size of

**Figure 7.** FESEM of intragranular fracture due to reoxidation of Ni nanoparticles in BZY10 (Courtesy of Thomas McGilvray, UCSD).

the precursor powders. It is sufficient just to obtain a homogeneous slip. The slip is filtered through a 325 mesh screen to remove the zirconia balls and any large agglomerates or undissolved binder, and continuously stirred prior to use to prevent settling.

Discs are cast by injecting a pre-calibrated amount of slip from a syringe into rings placed on a plaster-of-paris plate, and allowed to dry for several hours. Once dry and hard, the top surfaces of the cast discs are sanded flat to a thickness of about 3.5 mm. The discs are then stacked on top of BZY setters inside of a refractory enclosure to prevent excessive NiO loss during sintering. The initial sintering is carried out in air at 1600 °C for 16 hours. After sintering, the surfaces of the discs are ground with a diamond wheel on both faces to an overall thickness of 1.2 mm.

At this point in the process, the ceramic discs are fully dense and black. A second high temperature heat treatment is required to extract the NiO to a tolerable level. This is done by packing the ceramic in a bed of BZY powder inside of a ceramic retaining ring as shown in Fig. 8. A second firing cycle is carried out, also at 1600 °C. NiO extraction relies on relatively fast diffusion of NiO along grain boundaries until it reaches the surface of the specimen where it sublimes and is captured by the surrounding sacrificial BZY powder. The length of time required depends on the degree of extraction desired and the dimensions of the specimen. In the case of 1.2 mm thick discs, 16 hours is sufficient for achieving NiO level below 100 ppm. Fig. 9 shows a comparison of grain size distribution for the as-fired ceramic and after NiO extraction at 1600 °C for 16 hours. The grain-size measurements were carried out on powder obtained by crushing the specimens, assuming that fracture occurred predominantly at grain boundaries. This is supported by the observation that the typical fracture surfaces of this material are predominantly inter-granular. The as-fired specimen experienced 16 hours at 1600 °C, whereas the NiO-extracted specimen effectively experienced 32 hours at this temperature. It may be observed that the average grain size increased only slightly, from 5 μm to 6.5 μm, demonstrating that grain growth during sintering in BZY is rather slow.

**Figure 8.** Set up for NiO extraction process. Center photo shows the black sintered disc placed on top of the BZY pow‐ der bed. Each part is covered with powder.

the precursor powders. It is sufficient just to obtain a homogeneous slip. The slip is filtered through a 325 mesh screen to remove the zirconia balls and any large agglomerates or

**Figure 7.** FESEM of intragranular fracture due to reoxidation of Ni nanoparticles in BZY10 (Courtesy of Thomas

Discs are cast by injecting a pre-calibrated amount of slip from a syringe into rings placed on a plaster-of-paris plate, and allowed to dry for several hours. Once dry and hard, the top surfaces of the cast discs are sanded flat to a thickness of about 3.5 mm. The discs are then stacked on top of BZY setters inside of a refractory enclosure to prevent excessive NiO loss during sintering. The initial sintering is carried out in air at 1600 °C for 16 hours. After sintering, the surfaces of the discs are ground with a diamond wheel on both faces to an overall thickness

At this point in the process, the ceramic discs are fully dense and black. A second high temperature heat treatment is required to extract the NiO to a tolerable level. This is done by packing the ceramic in a bed of BZY powder inside of a ceramic retaining ring as shown in Fig. 8. A second firing cycle is carried out, also at 1600 °C. NiO extraction relies on relatively fast diffusion of NiO along grain boundaries until it reaches the surface of the specimen where it sublimes and is captured by the surrounding sacrificial BZY powder. The length of time required depends on the degree of extraction desired and the dimensions of the specimen. In the case of 1.2 mm thick discs, 16 hours is sufficient for achieving NiO level below 100 ppm. Fig. 9 shows a comparison of grain size distribution for the as-fired ceramic and after NiO extraction at 1600 °C for 16 hours. The grain-size measurements were carried out on powder

undissolved binder, and continuously stirred prior to use to prevent settling.

92 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

of 1.2 mm.

McGilvray, UCSD).

**Figure 9.** Grain size distributions of BZCY72 for as-fired (black) and after NiO extraction at 1600 °C for an additional 16 hours followed by an annealing treatment (grey) (Obtained with a Horiba Particle Size analyzer).

## **4.2. Characterization of the specimens after Ni extraction**

## *4.2.1. Structural characterization*

After the NiO extraction step, the discs are ground once more on both surfaces to the final 1.0 mm thickness, and ground on the outside diameter if necessary. A final annealing step in 4% H2-bal Ar at 800 °C for 24 hours is carried out to condition the ceramic discs for the testing environment. Fig. 10 shows the appearance of two specimens, BZCY81 and BZCY72, after NiO extraction at 1600 °C, followed by annealing in H2. The two specimens are transparent amber in color, and virtually Ni-free. Specimens with nickel extraction at only 1425 °C are dark and almost metallic-looking, demonstrating incomplete NiO extraction.

**Figure 10.** BZCY81 and BZCY72 specimens after NiO extraction at 1600 °C, followed by annealing in H2. At 800 °C for 24 hrs.

XRD was used to ensure that the process of NiO extraction and subsequent annealing did not modify the BZY phase in any way. Five batches of specimens were prepared for this purpose: 1) As-fired, without NiO extraction or H2 anneal, 2) NiO extraction only, 1425 °C for 16 hours, 3) NiO extraction, 1425 °C for 16 hours, plus H2 anneal, 4) NiO extraction only, 1600 °C for 16 hours, and 5) NiO extraction, 1600 °C for 16 hours, plus H2 anneal. The XRD patterns for 2θ = 20 to 120 degrees are shown in Fig. 11a. It may be observed that all cubic BZY peaks are present. The extra peak at 2θ =26.8, present on all diffractograms, comes from small barium peroxide (ICDD file 00-007-0233). More significant is the finding in Fig. 11b where the (321) peaks are magnified. It may be observed that there is a very slight increase in the peak position with increasing NiO extraction. The cubic lattice parameter for as-fired material is 4.248 Å and that for the material after NiO extraction for 16 hours at 1600 °C is 4.245 Å, or 0.07% contraction. This could suggest that at least some of the NiO may have originated from the perovskite lattice, but it should be kept in mind that this change is much less than the change typically observed due to hydration and dehydration, so this conclusion is tenuous.

**4.2. Characterization of the specimens after Ni extraction**

94 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

almost metallic-looking, demonstrating incomplete NiO extraction.

After the NiO extraction step, the discs are ground once more on both surfaces to the final 1.0 mm thickness, and ground on the outside diameter if necessary. A final annealing step in 4% H2-bal Ar at 800 °C for 24 hours is carried out to condition the ceramic discs for the testing environment. Fig. 10 shows the appearance of two specimens, BZCY81 and BZCY72, after NiO extraction at 1600 °C, followed by annealing in H2. The two specimens are transparent amber in color, and virtually Ni-free. Specimens with nickel extraction at only 1425 °C are dark and

**Figure 10.** BZCY81 and BZCY72 specimens after NiO extraction at 1600 °C, followed by annealing in H2. At 800 °C for

XRD was used to ensure that the process of NiO extraction and subsequent annealing did not modify the BZY phase in any way. Five batches of specimens were prepared for this purpose: 1) As-fired, without NiO extraction or H2 anneal, 2) NiO extraction only, 1425 °C for 16 hours, 3) NiO extraction, 1425 °C for 16 hours, plus H2 anneal, 4) NiO extraction only, 1600 °C for 16 hours, and 5) NiO extraction, 1600 °C for 16 hours, plus H2 anneal. The XRD patterns for 2θ = 20 to 120 degrees are shown in Fig. 11a. It may be observed that all cubic BZY peaks are present. The extra peak at 2θ =26.8, present on all diffractograms, comes from small barium peroxide (ICDD file 00-007-0233). More significant is the finding in Fig. 11b where the (321) peaks are magnified. It may be observed that there is a very slight increase in the peak position with increasing NiO extraction. The cubic lattice parameter for as-fired material is 4.248 Å and that for the material after NiO extraction for 16 hours at 1600 °C is 4.245 Å, or 0.07% contraction. This could suggest that at least some of the NiO may have originated from the perovskite lattice, but it should be kept in mind that this change is much less than the change typically

observed due to hydration and dehydration, so this conclusion is tenuous.

*4.2.1. Structural characterization*

24 hrs.

extraction 16h 1475 °C and annealed, (d) Ni extraction 16h 1600 °C, (e) Ni extraction 16h 1600 °C and annealed. The black bars correspond to the 04‐011‐7317 ICDD file. **Figure 11.** XRD of BZCY72 specimens prepared with different NiO extraction temperature. Full pattern (upper) and magnified (321) peak (lower). (a) as fired, (b) Ni extraction 16h 1475 °C, (c) Ni extraction 16h 1475 °C and annealed, (d) Ni extraction 16h 1600 °C, (e) Ni extraction 16h 1600 °C and annealed. The black bars correspond to the 04-011-7317 ICDD file.

Figure 11. XRD of BZCY72 specimens prepared with different NiO extraction temperature. Full pattern (upper) and magnified (321) peak (lower). (a) as fired, (b) Ni extraction 16h 1475 °C, (c) Ni

After reduction in H2 nearly all of the nickel may be assumed to be metallic, and ferromag‐ netic. This provides a convenient method for quantitative analysis of residual nickel content. It is extremely difficult to measure low levels of NiO reliably in sintered ceramics when the NiO is not uniformly distributed. By reducing the specimens, only metallic nickel remains, which can be quantitatively determined by magnetometry down to the parts per million lev‐ el. A Quatum Design SQUID-based VSM with EverCool™ at 1.8 K to 400 K, oven temp of 1000 K, uniform field of 0.05 Oe to 70,000 Oe and magnetic moment, 10-8 emu to 10 emu, AC frequency: 0.1 Hz to 1000 H was used. Fig. 12 shows the magnetization plots after reduc‐ tion in H2 as a function of NiO extraction temperature (each for 16 hours); 1) 1425 °C, 2) 1550 °C and 3) 1600 °C. The estimate of the amount of Ni in the as-fired specimens was about 0.1 wt%. The NIST standard of 54.888 emu/g-Ni was used to determine the amount of Ni in the specimens at 5000 Oe. For example, the specimen with NiO extraction at 1475 °C had a mag‐ netic moment of 0.022 emu/g-sample. Dividing this by 54.888 emu/g-Ni gives 0.039 wt % (times 58.69 g/mol-Ni/248.06 g/mol BZCY27 = 0.0092 mol%, or 92 ppm). Even at 92 ppm, the specimen was black and opaque. It was necessary to reduce the nickel to less than 0.002%, or about 4 ppm, in order to achieve transparency. It is seen that the extent of nickel extraction is highly dependent on NiO extraction temperature. Longer extraction times may be neces‐ sary for thicker specimens, but 1 mm is a good thickness for test specimens hydrogen flux measurements and electrode development.

**Figure 12.** Magnetization curves for different NiO extraction temperatures. Ni concentrations calculated at 5000 Oe (0.5 Tesla) using NIST standard of 54.888 emu/g-Ni (courtesy of Jim O'Brien, UCSD).

## *4.2.2. Microstructural characterization*

Back-scattered electron images on a polished BZCY72 specimen after Ni extraction and annealing in reducing atmosphere reveal the optimal microstructure: large grains (from 2 to 10 microns) with no dihedral pores and no Ni accumulation. An example is shown in Fig. 13. The grain boundaries of the specimen after nickel extraction and annealing in reducing atmosphere are free from nickel nano-particles (Fig. 14).

el. A Quatum Design SQUID-based VSM with EverCool™ at 1.8 K to 400 K, oven temp of 1000 K, uniform field of 0.05 Oe to 70,000 Oe and magnetic moment, 10-8 emu to 10 emu, AC frequency: 0.1 Hz to 1000 H was used. Fig. 12 shows the magnetization plots after reduc‐ tion in H2 as a function of NiO extraction temperature (each for 16 hours); 1) 1425 °C, 2) 1550 °C and 3) 1600 °C. The estimate of the amount of Ni in the as-fired specimens was about 0.1 wt%. The NIST standard of 54.888 emu/g-Ni was used to determine the amount of Ni in the specimens at 5000 Oe. For example, the specimen with NiO extraction at 1475 °C had a mag‐ netic moment of 0.022 emu/g-sample. Dividing this by 54.888 emu/g-Ni gives 0.039 wt % (times 58.69 g/mol-Ni/248.06 g/mol BZCY27 = 0.0092 mol%, or 92 ppm). Even at 92 ppm, the specimen was black and opaque. It was necessary to reduce the nickel to less than 0.002%, or about 4 ppm, in order to achieve transparency. It is seen that the extent of nickel extraction is highly dependent on NiO extraction temperature. Longer extraction times may be neces‐ sary for thicker specimens, but 1 mm is a good thickness for test specimens hydrogen flux

**Figure 12.** Magnetization curves for different NiO extraction temperatures. Ni concentrations calculated at 5000 Oe (0.5

Back-scattered electron images on a polished BZCY72 specimen after Ni extraction and annealing in reducing atmosphere reveal the optimal microstructure: large grains (from 2 to 10 microns) with no dihedral pores and no Ni accumulation. An example is shown in Fig. 13. The grain boundaries of the specimen after nickel extraction and annealing in reducing

Tesla) using NIST standard of 54.888 emu/g-Ni (courtesy of Jim O'Brien, UCSD).

atmosphere are free from nickel nano-particles (Fig. 14).

*4.2.2. Microstructural characterization*

measurements and electrode development.

96 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 13.** Back-scattered electron micrograph of a polished cross section of a BZCY72 (BaZr0.7Ce0.2Y0.1O3-d) after Ni ex‐ traction (16h 1600 °C) and annealing in reducing atmosphere.

**Figure 14.** Secondary electron micrographs of a fractured cross-section of a BZCY72 (BaZr0.7Ce0.2Y0.1O3-d) after Ni extrac‐ tion (16h 1600 °C) and annealing in reducing atmosphere.

## *4.2.3. Electronic characterization*

In moist atmosphere, BZY materials incorporate protonic defects (*OHO* •) in their lattice according to Eq. 1. For such a reaction to happen, oxygen vacancies are necessary. Some ceramic proton-conductors, such as pyrochlores [37-38], have intrinsic oxygen vacancies. For barium zirconate based materials, the oxygen vacancies are extrinsically created by substitut‐ ing the B-site of the perovskite (4+) by a trivalent cation (most commonly yttrium). Once in the lattice, the protonic defects diffuse according to the Grotthuss mechanism [39]. The protons are the only mobile species while the oxygen is localized in the vicinity of its crystallographic position [3].

In oxidizing atmosphere, oxygen dissociates in the oxygen vacancies. The corresponding reaction is most of the time reported as equation (2) [40-43], which can be misleading because the electron holes are not valence holes. Indeed, in models, these electron holes are treated as localized, since the Nernst Einstein equation is used [44]. However, no localization is defined. To avoid confusion, equation 2 can be rewritten as equation 3, assuming the electron holes are localized on the O-site [45-46].

$$\rm{^1\_2O\_2}(g) + V\_{\rm{O}}^{\bullet\bullet} \rightarrow O\_{\rm{O}}^{\times} + 2h^{\bullet} \tag{2}$$

$$\text{O}\_2^\text{-} \text{O}\_2\text{(g)} + \text{V}\_\text{O}^{\*\*} + \text{O}\_\text{O}^\* \rightarrow 2\text{O}\_\text{O}^\* \tag{3}$$

The conductivity was measured on 1 mm-thick pellets prepared as described in section 3.1 but with a smaller diameter (13 mm). Pt electrodes were painted on both sides of the pellet and fired in air for 30 min at 1000 °C. Impedance spectra were recorded from 20 Hz to 1 MHz using Hewlett Packard 4284A Precision LCR Meter interfaced with Labview and were fitted using the Zsimpwin software. The oxygen partial pressure dependences of the conductivity, with an example for BZCY72 in Fig. 15, exhibit the typical behavior for a BZY material in moist atmosphere:


According to Fig. 15, the protonic conductivity of the BZCY72 specimens after Ni extraction is 2.5 and 1.3 mS.cm-1 at 700 and 600 °C respectively. These values are consistent with those reported in the literature for BZCY72 [9,32,34-35,49], indicating that the extraction process did not deteriorate the specimens.

Some electronic conductivity in reducing atmosphere was measured in barium cerate based materials [50-51], generating from the reduction of the cerium cations and the formation of small polarons. In the present work, the conditions were not reducing enough to generate cerium small polarons. However, this phenomenon has been observed on thin BZCY72 membranes [36]. Depending on the desired operation (gas composition, humidification rate and temperature), the percentage of cerium needs to be adjusted.

*4.2.3. Electronic characterization*

localized on the O-site [45-46].

position [3].

atmosphere:

temperature below 700 °C [47-48],

not deteriorate the specimens.

from the incorporation of the oxygen (Eq. 3).

In moist atmosphere, BZY materials incorporate protonic defects (*OHO*

98 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

( ) <sup>1</sup>

( ) <sup>1</sup>

<sup>2</sup> <sup>2</sup> 2 *Og V O h O O*

<sup>2</sup> <sup>2</sup> 2 *Og V O O OO O*

The conductivity was measured on 1 mm-thick pellets prepared as described in section 3.1 but with a smaller diameter (13 mm). Pt electrodes were painted on both sides of the pellet and fired in air for 30 min at 1000 °C. Impedance spectra were recorded from 20 Hz to 1 MHz using Hewlett Packard 4284A Precision LCR Meter interfaced with Labview and were fitted using the Zsimpwin software. The oxygen partial pressure dependences of the conductivity, with an example for BZCY72 in Fig. 15, exhibit the typical behavior for a BZY material in moist

**•** A plateau at oxygen partial pressure below 10-5 atm, which corresponds to the ionic conductivity. It is agreed that for BZY material, the ionic conductivity is mainly protonic at

**•** An increase of the conductivity with the oxygen partial pressure in oxidizing atmosphere,

According to Fig. 15, the protonic conductivity of the BZCY72 specimens after Ni extraction is 2.5 and 1.3 mS.cm-1 at 700 and 600 °C respectively. These values are consistent with those reported in the literature for BZCY72 [9,32,34-35,49], indicating that the extraction process did

Some electronic conductivity in reducing atmosphere was measured in barium cerate based materials [50-51], generating from the reduction of the cerium cations and the formation of

·· ´ · + ®+ (2)

·· ´ · + +® (3)

according to Eq. 1. For such a reaction to happen, oxygen vacancies are necessary. Some ceramic proton-conductors, such as pyrochlores [37-38], have intrinsic oxygen vacancies. For barium zirconate based materials, the oxygen vacancies are extrinsically created by substitut‐ ing the B-site of the perovskite (4+) by a trivalent cation (most commonly yttrium). Once in the lattice, the protonic defects diffuse according to the Grotthuss mechanism [39]. The protons are the only mobile species while the oxygen is localized in the vicinity of its crystallographic

In oxidizing atmosphere, oxygen dissociates in the oxygen vacancies. The corresponding reaction is most of the time reported as equation (2) [40-43], which can be misleading because the electron holes are not valence holes. Indeed, in models, these electron holes are treated as localized, since the Nernst Einstein equation is used [44]. However, no localization is defined. To avoid confusion, equation 2 can be rewritten as equation 3, assuming the electron holes are

•) in their lattice

**Figure 15.** pO2 dependence of the conductivity measured on a BZCY27 pellet after Ni extraction.

As mentioned in the introduction, the high grain-boundary resistivity in the BZY materials has been explained by the presence of a space-charge layer [27-30]. The width and the magnitude of the space-charge layer depends on the synthesis/sintering processes [52]. The grain-boundary contribution to the conductivity can be decreased by having a large grain microstructure. Indeed, the larger the grains, the smaller the grain boundary density. But other factors, such as the dopant accumulation, the barium evaporation or the segregation of impurities at the grain-boundary, play an important role in the space-charge layer.

The conductivity was recorded as a function of temperature in 3% moist 5% H2, balance Ar. At high temperature, the spectra were fitted with 2 (RQ) elements in series, with R, a resistance and Q, a constant phase element. The different processes could be analyzed using the pseudocapacitance. The first (RQ) was assigned to the bulk (~10-11 F) and the second one to the Pt electrodes (~10-6 F). With decreasing temperature, the grain-boundary contribution becomes visible. Because of the frequency range used for the measurement, the electrode contribution does not appear anymore at low temperature (lower frequencies would be needed). Therefore, at low temperature, the spectra were also fitted with two (RQ) in series, the first one for the bulk and the second one for the grain boundaries (~10-9 F). Example of the fits of the impedance spectra at 600 and 400 °C are displayed in Fig. 16. The Arrhenius plot of the conductivity in 3% moist 5% H2, balance Ar, cf. Fig. 17, gives an activation energy for proton diffusion of 0.47 eV for the bulk and 0.78 eV for the grain-boundaries. Similar values were reported in the literature [13,28,30,52]. plot of the conductivity in 3% moist 5% H2, balance Ar, cf. Fig. 17, gives an activation energy for proton diffusion of 0.47 eV for the bulk and 0.78 eV for the grain‐boundaries. Similar

values were reported in the literature [13,28,30,52].

**Figure 16.** Impedance spectra of BZCY72 after Ni extraction in 3% moist 5% H 600 °C and (lower) 400 °C. 2, balance Ar. (upper) 600 °C and (lower) 400 °C. GB stands for grain boundary

Figure 16. Impedance spectra of BZCY72 after Ni extraction in 3% moist 5% H2, balance Ar. (upper)

## **5. Summary**

A process has been described for making dense, large-grained BZY suitable for use in hydrogen diffusion membranes. It is nearly impossible to make meaningful measure‐ ments on membranes films, themselves, on the order of tens of microns thick. The thick specimens shown in Fig. 10 are intended for testing purposes, where it is desirable to have

3% moist 5% H2, balance Ar, cf. Fig. 17, gives an activation energy for proton diffusion of 0.47 eV for the bulk and 0.78 eV for the grain-boundaries. Similar values were reported in the literature [13,28,30,52]. plot of the conductivity in 3% moist 5% H2, balance Ar, cf. Fig. 17, gives an activation energy

values were reported in the literature [13,28,30,52].

100 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

for proton diffusion of 0.47 eV for the bulk and 0.78 eV for the grain‐boundaries. Similar

Figure 16. Impedance spectra of BZCY72 after Ni extraction in 3% moist 5% H2, balance Ar. (upper)

**Figure 16.** Impedance spectra of BZCY72 after Ni extraction in 3% moist 5% H 600 °C and (lower) 400 °C. 2, balance Ar. (upper) 600 °C and (lower)

A process has been described for making dense, large-grained BZY suitable for use in hydrogen diffusion membranes. It is nearly impossible to make meaningful measure‐ ments on membranes films, themselves, on the order of tens of microns thick. The thick specimens shown in Fig. 10 are intended for testing purposes, where it is desirable to have

400 °C. GB stands for grain boundary

**5. Summary**

**Figure 17.** Arrhenius plot for BZCY72 after Ni extraction in 3% moist 5% H2, balance Ar. The specific grain-boundary (GB) conductivity is obtained by multiplying the grain-boundary conductivity by the ratio of the pseudo-capacitance of the bulk over the pseudo-capacitance of the grain boundaries [53].

ceramic material that is representative of thin membranes for such things as electrode development, measurement of bulk mechanical and electrical properties, etc. The fact that the specimens are transparent is incidental, except that this unique property makes it possible to measure bulk optical properties, such as band gap, by transmission spectra (see Chapter 16 of Fundamentals of Ceramics [19]).

Practical hydrogen diffusion membranes must be as thin as possible in order to maximize the hydrogen flux, but still remain impervious to all other gas species. Fig. 18 shows the crosssection of a typical BZCY72 membrane coating on a porous support tube for a protonic ceramic fuel cell or membrane reactor from a 10 mm diameter production tube from CoorsTek Membranes Sciences. The 20 μm-thick membrane is shown in the upper part of the image coating a 0.8 mm thick Ni/BZCY72 anode-support. The image was obtained on a polished specimen that highlights the microstructure. The fully dense and large-grained microstructure of the membrane is clearly seen. Such a membrane is capable of passing a hydrogen flux of about 5 μmol/cm2 .s (7.3 nml/min/cm2 ) from a H2/He permeant stream at 700 °C while passing negligible detectable helium, as determined by analysing the permeate by mass spectrometry.

Solid-state reactive sintering is a very robust and cost effective method for making such membranes on the scale of hundreds of millions of square meters that will be required for commercialization of this technology.

**Figure 18.** Polished cross-section of BZCY72 membrane on porous Ni/BZCY72 cermet support.

## **Acknowledgements**

The authors acknowledge Daniel Clark from Colorado School of Mines for the transmission electron micrograph, Dr. Jim O' Brien for providing the Squid Data, Mc Gilvray at UCSD for FESEM and Dr. Grant Hudish at CoorsTek for collecting the XRD data.

## **Author details**

W. Grover Coors1 , Anthony Manerbino1 , David Martinefski1 and Sandrine Ricote2

1 CoorsTek, Inc. Golden, Colorado, USA

2 Colorado School of Mines, Golden, Colorado, USA

## **References**

**Figure 18.** Polished cross-section of BZCY72 membrane on porous Ni/BZCY72 cermet support.

FESEM and Dr. Grant Hudish at CoorsTek for collecting the XRD data.

, Anthony Manerbino1

102 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

2 Colorado School of Mines, Golden, Colorado, USA

1 CoorsTek, Inc. Golden, Colorado, USA

The authors acknowledge Daniel Clark from Colorado School of Mines for the transmission electron micrograph, Dr. Jim O' Brien for providing the Squid Data, Mc Gilvray at UCSD for

, David Martinefski1

and Sandrine Ricote2

**Acknowledgements**

**Author details**

W. Grover Coors1


[29] Chen CT, Danel CE, Kim S (2011) On the origin of the blocking effect of grain-boun‐ daries on proton transport in yttrium-doped barium zirconates. J. Mater. Chem. 21:5435-5442.

[16] Bonanos N (1992) Transport properties and conduction mechanism in high-tempera‐

[17] Kosacki I, Tuller HL (1995) Mixed conductivity in SrCe0.95Yb0.05O3, protonic conduc‐

[18] Krug F, Schober T, Springer T (1995) In situ measurements of the water uptake in Yb

[19] Barsoum MW (2003) Sintering and grain growth. In: Taylor & Francis Group. Funda‐

[20] Meir S, Kalabukhov S, Froumin N, Dariel MP, Frage N (2009) Synthesis and Densifi‐ cation of Transparent Magnesium Aluminate Spinel by SPS Processing. J. Am. Ce‐

[21] Dahl PI, Lein HL, Yu Y, Tolchard J, Grande T, Einarsrud M-A, Kjølseth C, Norby T, Haugsrud R (2011) Microstructural characterization and electrical properties of spray pyrolyzed conventionally sintered or hot-pressed BaZrO3 and BaZr0.9Y0.1O3− δ. Solid

[22] Anselmi-Tamburini U, Buscaglia MT, Viviani M, Bassoli M, Bottino C, Buscaglia V, Nanni P, Munir ZA (2006) Solid-state synthesis and spark plasma sintering of submi‐ cron BaYxZr1−xO3−x/2 (*x* = 0, 0.08 and 0.16) ceramics. J. European Ceram. Soc.

[23] Ricote S, Bonanos N, Wang HJ, Boukamp BA (2012) Conductivity study of dense BaZr0.9Y0.1O3-δ obtained by spark plasma sintering. Solid State Ionics 213:36-41.

[24] Barison S, Battagliarin M, Cavallin T, Doubova L, Doubova M, Mortalo C, Boldrini S, Malavasi L, Gerbasi R (2008) High conductivity and chemical stability of BaCe1-xyZrxYyO3-d proton conductors prepared by a sol–gel method. J. Mater. Chem.

[25] Stuart PA, Unno T, Ayres-Rocha R, Djurado E, Skinner SJ (2009) Synthesis and sinter‐ ing behaviour of BaZrYO3 powders prepared by spray pyrolysis. J. European Ceram.

[26] Kim S, Fleig J, Maier J: Space charge conduction (2003) Simple analytical solutions for ionic and mixed conductors and application to nanocrystalline ceria. Phys. Chem.

[27] Iguchi F, Sata N, Yugami H (2010) Proton transport properties at the grain boundary of barium zirconate based proton conductors for intermediate temperature operating

[28] Kjølseth C, Fjeld H, Prytz Ø, Dahl PI, Estournes C, Haugsrud R, Norby T (2010) Space–charge theory applied to the grain boundary impedance of proton conducting

ture protonic conductors. Solid State Ionics. 53-56:967-974.

tors. Solid State Ionics. 80:223-229.

doped SrCeO3. Solid State Ionics 81:111-118.

104 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

mentals of Ceramics. LLC. pp. 302-354.

ram. Soc. 92:358-364.

State Ionics 182:32-40.

26:2313-2318.

18:5120-5128.

Soc. 29:697-702.

Chem. Phys. 5:2268-2273.

SOFC. J. Mater. Chem. 20:6265-6270.

BaZr0.9Y0.1O3−δ. Solid State Ionics 181:268-275.


The electrical conductivity and diffusion coefficient of BaCe0.9Y0.1O3–δ. Solid State Ion‐ ics 179:2240-2247.


## **Perovskite Nanomaterials – Synthesis, Characterization, and Applications**

Nada F. Atta, Ahmed Galal and Ekram H. El-Ads

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61280

## **Abstract**

The electrical conductivity and diffusion coefficient of BaCe0.9Y0.1O3–δ. Solid State Ion‐

[45] Stokes SJ, Islam MS (2010) Defect chemistry and proton-dopant association in BaZrO3

[46] Hermet J, Bottin F, Dezanneau G, Geneste G (2012) Thermodynamics of hydration and oxidation in the proton conductor Gd-doped barium cerate from density func‐

[47] Iwahara H (1995) Technological challenges in the application of proton conducting

[48] Ricote S, Bonanos N, Wang HJ, Haugsrud R (2011) Conductivity, transport number measurements and hydration thermodynamics of BaCe0.2Zr0.7Y(0.1−ξ)NiξO(3−δ). Solid

[49] Fish JS, Ricote S, O'Hayre R, Bonanos N (2014) Electrical properties and flux per‐ formance of composite ceramic hydrogen separation membranes. J. Mater. Chem. A.

[50] He T, Kreuer KD, Baikov Yu.M, Maier J (1997) Impedance spectroscopic study of thermodynamics and kinetics of a Gd-doped BaCeO3 single crystal. Solid State Ionics

[51] Bonanos N (1993) Transport study of the solid electrolyte BaCe0.9Gd0.1O2.95 at high

[52] Ricote S, Bonanos N, Manerbino A, Sullivan NP, Coors WG (2014) Effects of the fab‐ rication process on the grain-boundary resistance in BaZr0.9Y0.1O3-d. J. Mater. Chem.

[53] Haile SM, Staneff G, Ryu KH (2001) Non-stoichiometry, grain-boundary transport and chemical stability of proton conducting perovskites. J. Materials Science

ics 179:2240-2247.

State Ionics 185:11-17.

3:5392-5401.

95:301-308.

A. 2:16107-16115.

36:1149-1160.

and BaPrO3. J. Mater. Chem. 20:6258-6264.

106 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

ceramics. Solid State Ionics 77:289-298.

temperatures. J. Phys. Chem. Solids 54:867-870.

tional theory calculations. Physical review B 85:205137 1-14

Inorganic perovskite-type oxides are fascinating nanomaterials for wide applications in catalysis, fuel cells, and electrochemical sensing. Perovskites prepared in the nanoscale have recently received extensive attention due to their catalytic nature when used as electrode modifiers. The catalytic activity of these oxides is higher than that of many transition metals compounds and even some precious metal oxides. They exhibit attractive physical and chemical characteristics such as electronic conductivity, electrically active structure, the oxide ions mobility through the crystal lattice, variations on the content of the oxygen, thermal and chemical stability, and super‐ magnetic, photocatalytic, thermoelectric, and dielectric properties.

Nanoperovskites have been utilized as catalysts in oxygen reduction and hydrogen evolution reactions exhibiting high electrocatalytic activity, lower activation energy, and high electron transfer kinetics. In addition, some perovskites are promising candidates for the development of effective anodic catalysts for direct fuel cells showing high catalytic performance. Moreover, they are recently utilized in electro‐ chemical sensing of alcohols, gases, glucose, H2O2, and neurotransmitters. They can enhance the catalytic performance in terms of selectivity, sensitivity, unique long-term stability, excellent reproducibility, and anti-interference ability. In addition, organo‐ metallic halide perovskites exhibited efficient intrinsic properties for photovoltaic solar cells exhibiting good stability and high efficiency.

This chapter introduces a comprehensive coverage of the progress in perovskites research and their applications. Emphasis is given toward several intrinsic properties of perovskites, namely, electronic conductivity, electrically active structure, and electrochemical performance in terms of synthesis routes and stability. The different synthesis methods of the perovskites (coprecipitation, sol-gel, microwave, citrate/ nitrate, etc.) will be summarized in this chapter. The synthesis method affected

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

structural, surface, and catalytic properties of the prepared perovskites to a great extent. Also, this chapter will update the reader with the various applications of nanoperovskites particularly in fuel cells, catalysis, electrochemical sensing, and solar cells.

**Keywords:** Nanomaterials, Perovskites, Sensors, Catalysis, Fuel cells

## **1. Introduction**

## **1.1. General introduction to perovskites**

The mineral CaTiO3 was discovered by Geologist Gustav Rose in the Ural Mountains in 1839, and it was named perovskite in recognition beholden to Count Lev Alexevich von Perovski, an eminent Russian mineralogist [1–5]. The name perovskite represented any compound that has ABC3 formula where an octahedron of C ions surrounded the B ion. The Earth's crust contains various types of perovskites and the most abundant ones are MgSiO3 and FeSiO3. Perovskite family includes several types of oxides like transition metal oxides with the formula ABO3. Some examples of ABO3 perovskites and their corresponding properties are summar‐ ized in Table 1 [1, 2, 5].

Perovskite oxides exhibit an array of electrical properties and a variety of solid-state phenom‐ ena from insulating, semiconducting, metallic, and superconducting characters; therefore, they are very fascinating to be studied and applied in a large scale. Many of ABO3 perovskites are cubic or nearly cubic in structure in their ideal form; however, one or more phase transitions may be achieved particularly at low temperature. In addition, many of them showed magnetic ordering and as a result, large variety of magnetic structures can be found. Some perovskites contained localized electrons, some contained delocalized energy-band states, and the behavior of other perovskites was a transition between these two types. The perovskite structures can incorporate ions of various size and charge showing great flexibility of compo‐ sition. Moreover, substitutions of ions into the A- and/or B-sites or deviation from ideal stoichiometry resulted in altering the electronic properties of the perovskites. Perovskites exhibit atomic arrangement in the form of 3-dimensional array of corner sharing octahedra. On the other hand, layered perovskites included 2-dimensional layers of corner sharing octahedral separated by cations layers. As a result, the electronic energy bands of perovskites and layered perovskites are very unusual and their structure is unique in properties [1, 2]. Perovskites displayed diversity of electric, optical, and magnetic properties because of the fact that 90% of the elements in the periodic table can be stable in the perovskite structure and the feasibility of partial substitutions of cations in A- and B-sites forming A1-*x*A′*x*B1-*y*B′*y*O3 [6]. Perovskites showed great interest in several applications due to their wide various and useful properties in photochromic, electrochromic, image storage, switching, filtering, and surface acoustic wave signal processing devices. They were utilized as catalytically active catalyst for several reactions like carbon monoxide and hydrocarbons oxidation, hydrogen evolution reaction and nitrogen oxides, and oxygen reduction reactions. They also have a good impact in many electrochemical applications like sensing, biosensing, photoelectrolysis of waterproducing hydrogen, and fuel cells [1, 2].


**Table 1.** Some perovskites and corresponding properties [1].

structural, surface, and catalytic properties of the prepared perovskites to a great extent. Also, this chapter will update the reader with the various applications of nanoperovskites particularly in fuel cells, catalysis, electrochemical sensing, and solar

The mineral CaTiO3 was discovered by Geologist Gustav Rose in the Ural Mountains in 1839, and it was named perovskite in recognition beholden to Count Lev Alexevich von Perovski, an eminent Russian mineralogist [1–5]. The name perovskite represented any compound that has ABC3 formula where an octahedron of C ions surrounded the B ion. The Earth's crust contains various types of perovskites and the most abundant ones are MgSiO3 and FeSiO3. Perovskite family includes several types of oxides like transition metal oxides with the formula ABO3. Some examples of ABO3 perovskites and their corresponding properties are summar‐

Perovskite oxides exhibit an array of electrical properties and a variety of solid-state phenom‐ ena from insulating, semiconducting, metallic, and superconducting characters; therefore, they are very fascinating to be studied and applied in a large scale. Many of ABO3 perovskites are cubic or nearly cubic in structure in their ideal form; however, one or more phase transitions may be achieved particularly at low temperature. In addition, many of them showed magnetic ordering and as a result, large variety of magnetic structures can be found. Some perovskites contained localized electrons, some contained delocalized energy-band states, and the behavior of other perovskites was a transition between these two types. The perovskite structures can incorporate ions of various size and charge showing great flexibility of compo‐ sition. Moreover, substitutions of ions into the A- and/or B-sites or deviation from ideal stoichiometry resulted in altering the electronic properties of the perovskites. Perovskites exhibit atomic arrangement in the form of 3-dimensional array of corner sharing octahedra. On the other hand, layered perovskites included 2-dimensional layers of corner sharing octahedral separated by cations layers. As a result, the electronic energy bands of perovskites and layered perovskites are very unusual and their structure is unique in properties [1, 2]. Perovskites displayed diversity of electric, optical, and magnetic properties because of the fact that 90% of the elements in the periodic table can be stable in the perovskite structure and the feasibility of partial substitutions of cations in A- and B-sites forming A1-*x*A′*x*B1-*y*B′*y*O3 [6]. Perovskites showed great interest in several applications due to their wide various and useful properties in photochromic, electrochromic, image storage, switching, filtering, and surface acoustic wave signal processing devices. They were utilized as catalytically active catalyst for several reactions like carbon monoxide and hydrocarbons oxidation, hydrogen evolution

**Keywords:** Nanomaterials, Perovskites, Sensors, Catalysis, Fuel cells

108 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

cells.

**1. Introduction**

ized in Table 1 [1, 2, 5].

**1.1. General introduction to perovskites**

## **1.2. Crystallography of the perovskite structure**

In the ABO3 form, B is a transition metal ion with small radius, larger A ion is an alkali earth metals or lanthanides with larger radius, and O is the oxygen ion with the ratio of 1:1:3. In the cubic unit cell of ABO3 perovskite, atom A is located at the body center, atom B is located at the cube corner position, and oxygen atoms are located at face-centered positions (Figure 1). The 6-fold coordination of B cation (octahedron) and the 12-fold coordination of the A cation resulted in the stabilization of the perovskite structure. The perfect perovskite structure was described by Hines et al. as corner linked BO6 octahedra with interstitial A cations [1–10]. Some distortions may exist in the ideal cubic form of perovskite resulted in orthorhombic, rhombo‐ hedral, hexagonal, and tetragonal forms (Figure 1) [3–7]. Figure 2 represented the distortion from cubic perovskite to orthorhombic one. In general, all perovskite distortions maintaining the A- and the B-site oxygen coordination was achieved by the tilting of the BO6 octahedra and an associated displacement of the A cation [4].

V.M. Goldschmidt presented much of the early work on the synthetic perovskites and developed the principle of the tolerance factor *t*, which is applicable to the empirical ionic radii at room temperature [2–9]:

$$t = \left(r\_A + r\_O\right) \;/\; \left[\mathcal{B}^{1/2}\left(r\_B + r\_O\right)\right] \;/\;$$

where *r*A is the radius of the A-site cation, *r*B is the radius of the B-site cation, and *r*O is the radius of oxygen ion O2–. The tolerance factor can be used to estimate the suitability of the combination of cations for the perovskite structure [2]. It is a real measure of the degree of the distortion of perovskite from the ideal cubic structure so the value of *t* tends to unity as the structure approaches the cubic form [4]. From the equation, the tolerance factor will decrease when *r*<sup>A</sup> decreases and/o*rr*B increases. Based on the analysis of tolerance factor value, Hines et al. solely suggested that the perovskite structure can be estimated. For 1.00 < *t* < 1.13, 0.9 < *t* < 1.0, and 0.75 < *t* <0.9, the perovskite structure is hexagonal, cubic, and orthorhombic, respectively. For *t* < 0.75, the structure was adopted to hexagonal ilmenite structure (FeTiO3) [4].

Generally, two requirements should be fulfilled for perovskite formation:


*Figure\_1* 

*Figure 1: Different perovskite unit cells. Blue spheres represent the A cations, yellow spheres represent the B cations and red spheres represent oxygen anions forming an octahedra [4].* **Figure 1.** Different perovskite unit cells. Blue spheres represent the A cations, yellow spheres represent the B cations and red spheres represent oxygen anions forming an octahedra [4].

*Figure\_2* 

1

*Figure\_3* 

(a) cubic (b) orthorhombic

*Figure 2: Perovskite distortion from (a) cubic to (b) orthorhombic [4].*

Rhombohedral Hexagonal

*Figure\_2* 

*Figure\_1* 

Cubic Orthorhombic

*Figure\_3* **Figure 2.** Perovskite distortion from (a) cubic to (b) orthorhombic [4].

## **1.3. Typical properties of perovskites**

perovskite from the ideal cubic structure so the value of *t* tends to unity as the structure approaches the cubic form [4]. From the equation, the tolerance factor will decrease when *r*<sup>A</sup> decreases and/o*rr*B increases. Based on the analysis of tolerance factor value, Hines et al. solely suggested that the perovskite structure can be estimated. For 1.00 < *t* < 1.13, 0.9 < *t* < 1.0, and 0.75 < *t* <0.9, the perovskite structure is hexagonal, cubic, and orthorhombic, respectively. For

**1.** Electroneutrality; the perovskite formula must have neutral balanced charge therefore the product of the addition of the charges of A and B ions should be equivalent to the whole charge of the oxygen ions. An appropriate charge distribution should be attained in the

**2.** Ionic radii requirements; *r*A > 0.090 nm and *r*B > 0.051 nm, and the tolerance factor must

*Figure\_1* 

Cubic Orthorhombic

Rhombohedral Hexagonal

*B cations and red spheres represent oxygen anions forming an octahedra [4].*

and red spheres represent oxygen anions forming an octahedra [4].

*Figure 2: Perovskite distortion from (a) cubic to (b) orthorhombic [4].*

*Figure 1: Different perovskite unit cells. Blue spheres represent the A cations, yellow spheres represent the* 

**Figure 1.** Different perovskite unit cells. Blue spheres represent the A cations, yellow spheres represent the B cations

*Figure\_2* 

1

*Figure\_3* 

(a) cubic (b) orthorhombic

*t* < 0.75, the structure was adopted to hexagonal ilmenite structure (FeTiO3) [4].

Generally, two requirements should be fulfilled for perovskite formation:

forms of A1+B5+O3, A4+B2+O3 or A3+B3+O3.

have values within the range 0.8 < *t <* 1.0 [2–8].

110 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

1 Perovskite exhibited a variety of fascinating properties like ferroelectricity as in case of BaTiO3 and superconductivity as in case of Ba2YCu3O7. They exhibited good electrical con‐ ductivity close to metals, ionic conductivity and mixed ionic and electronic conductivity. In addition, several perovskites exhibited high catalytic activity toward various reactions. Table 2 contains a summary of typical properties of perovskites. Several typical properties will be discussed in this section like ferroelectricity, magnetism, superconductivity, and catalytic activity [9].


**Table 2.** Typical properties of perovskite oxides [9].

## *1.3.1. Dielectric properties*

There are some properties inherent to dielectric materials like ferroelectricity, piezoelectricity, electrostriction, and pyroelectricity. One of the important characteristic of perovskites is ferroelectric behavior, which is obvious in BaTiO3, PdZrO3, and their doped compounds. The ferroelectric behavior of BaTiO3 was strongly related to its crystal structure. BaTiO3 was subjected to three phase transitions; as the temperature increases, it was converted from monoclinic to tetragonal then to cubic. At temperature higher than 303 K, BaTiO3 does not show any ferroelectric behavior as it crystallizes into cubic structure. BaTiO3 showed high dielectric constant based on the anisotropy of its crystal structure resulting in large dipole moment generation in BaTiO3 [9].

## *1.3.2. Electrical conductivity and superconductivity*

One of the obvious properties of perovskites is superconductivity. Cu-based perovskites act as high-temperature superconductors, and La-Ba-Cu-O perovskite was first reported. The presence of Cu in B-site is essential for the superconductivity and various superconducting oxides can be manufactured with different A-site ions. Some examples of high temperature superconductors are YBa2Cu3O7, Bi2Sr2Ca2Cu3O10, and HgBa2Ca2Cu3O8+*<sup>δ</sup>* with critical temper‐ ature of superconducting transition (Tc) of 130–155 K. The superconductivity is associated with the layers of Cu-O in Cu-based perovskites, and the Tc value is associated with the Cu-O layers number in the crystal lattice. The synthesis of 5 or more Cu-O-layered perovskites did not achieve successfully due to the low chemical stability. One of the highly significant supercon‐ ductors with great Tc value is YBa2Cu3O7. In addition, the oxygen nonstiochiometry is one of the most significant reasons for the high value of Tc. In YBa2Cu3O7–*δ*, when the value of *δ* < 0.5, it crystallizes into orthorhombic structure, which is superconductive. For *δ* > 0.5, it showed a tetragonal structure that does not show any superconductivity. The crystal structure affected greatly the superconductivity in high Tc oxides, and as a result, high Tc values can be achieved by improving the chemical stability of the perovskite crystal structure. Furthermore, some perovskites exhibited great electronic conductivity similar to that of metals like Cu. LaCoO3 and LaMnO3 are examples of perovskites exhibiting high electronic conductivity, and therefore they are utilized as cathodes in solid oxide fuel cells displaying superior hole conductivity of 100 S/cm. The electronic conductivity of the perovskites can be enhanced by doping the A-site with another cation, which resulted in increasing the quantity of the mobile charge carriers created by the reparations of charge [9].

## *1.3.3. Catalytic activity*

Perovskites showed excellent catalytic activity and high chemical stability; therefore, they were studied in a wide range in the catalysis of different reactions. Perovskites can be described as a model of active sites and as an oxidation or oxygen-activated catalyst. The stability of the perovskite structure allowed the compounds preparation from elements with unusual valence states or a high extent of oxygen deficiency. Perovskites exhibited high catalytic activity, which is partially associated with the high surface activity to oxygen reduction ratio or oxygen activation that resulted from the large number of oxygen vacancies. Perovskites can act as automobile exhaust gas catalyst, intelligent automobile catalyst and cleaning catalyst, etc., for various catalytic environmental reactions. It was reported in the literature that perovskites containing Cu, Co, Mn, or Fe showed excellent catalytic activity toward the direct decompo‐ sition of NO at high temperature, which is considered one of the difficult reactions in the catalysis (2NO → N2+O2). Perovskites showed superior activity for this reaction at high temperatures because of the presence of oxygen deficiency and the simple elimination of the surface oxygen in the form of a reaction product. NO decomposition activity was enhanced upon doping. Also, under an atmosphere that is rich with oxygen up to 5%, Ba(La)Mn(Mg)O3 perovskite exhibited superior activity toward the decomposition of NO [9].

Perovskite showed a great impact as an automobile catalyst; intelligent catalyst. Pd-Rh-Pt catalysts was utilized as an effective catalyst for the removal of NO, CO and uncombusted hydrocarbons. There is another catalyst that consists of fine particles, with high surface-tovolume ratio, and can be utilized to reduce the amount of precious metals used. However, these fine particles exhibited very bad stability under the operation conditions leading to catalyst deactivation. Therefore, perovskite oxides can be used showing redox properties to preserve a great dispersion state. Upon oxidation, Pd is oxidized in the form of LaFe0.57Co0.38Pd0.05O3 and upon reduction; fine metallic particles of Pd were produced with radius of 1–3 nm. This cycle resulted in partial replacement of Pd into and sedimentation from the framework of the perovskite under oxidizing and reducing conditions, respectively, displaying a great dispersion state of Pd. Also, this cycle improved the excellent long-term stability of Pd during the pollutants removal from the exhaust gas. Exposing the catalyst to oxidizing and reducing atmosphere resulted in the recovery of the high dispersion state of Pd. This catalyst is known as intelligent catalyst because of the great dispersion state of Pd and the excellent stability of the perovskite structure [9].

## **2. Methods of perovskite synthesis**

## **2.1. Solid-state reactions**

subjected to three phase transitions; as the temperature increases, it was converted from monoclinic to tetragonal then to cubic. At temperature higher than 303 K, BaTiO3 does not show any ferroelectric behavior as it crystallizes into cubic structure. BaTiO3 showed high dielectric constant based on the anisotropy of its crystal structure resulting in large dipole

One of the obvious properties of perovskites is superconductivity. Cu-based perovskites act as high-temperature superconductors, and La-Ba-Cu-O perovskite was first reported. The presence of Cu in B-site is essential for the superconductivity and various superconducting oxides can be manufactured with different A-site ions. Some examples of high temperature superconductors are YBa2Cu3O7, Bi2Sr2Ca2Cu3O10, and HgBa2Ca2Cu3O8+*<sup>δ</sup>* with critical temper‐ ature of superconducting transition (Tc) of 130–155 K. The superconductivity is associated with the layers of Cu-O in Cu-based perovskites, and the Tc value is associated with the Cu-O layers number in the crystal lattice. The synthesis of 5 or more Cu-O-layered perovskites did not achieve successfully due to the low chemical stability. One of the highly significant supercon‐ ductors with great Tc value is YBa2Cu3O7. In addition, the oxygen nonstiochiometry is one of the most significant reasons for the high value of Tc. In YBa2Cu3O7–*δ*, when the value of *δ* < 0.5, it crystallizes into orthorhombic structure, which is superconductive. For *δ* > 0.5, it showed a tetragonal structure that does not show any superconductivity. The crystal structure affected greatly the superconductivity in high Tc oxides, and as a result, high Tc values can be achieved by improving the chemical stability of the perovskite crystal structure. Furthermore, some perovskites exhibited great electronic conductivity similar to that of metals like Cu. LaCoO3 and LaMnO3 are examples of perovskites exhibiting high electronic conductivity, and therefore they are utilized as cathodes in solid oxide fuel cells displaying superior hole conductivity of 100 S/cm. The electronic conductivity of the perovskites can be enhanced by doping the A-site with another cation, which resulted in increasing the quantity of the mobile charge carriers

Perovskites showed excellent catalytic activity and high chemical stability; therefore, they were studied in a wide range in the catalysis of different reactions. Perovskites can be described as a model of active sites and as an oxidation or oxygen-activated catalyst. The stability of the perovskite structure allowed the compounds preparation from elements with unusual valence states or a high extent of oxygen deficiency. Perovskites exhibited high catalytic activity, which is partially associated with the high surface activity to oxygen reduction ratio or oxygen activation that resulted from the large number of oxygen vacancies. Perovskites can act as automobile exhaust gas catalyst, intelligent automobile catalyst and cleaning catalyst, etc., for various catalytic environmental reactions. It was reported in the literature that perovskites containing Cu, Co, Mn, or Fe showed excellent catalytic activity toward the direct decompo‐ sition of NO at high temperature, which is considered one of the difficult reactions in the catalysis (2NO → N2+O2). Perovskites showed superior activity for this reaction at high

moment generation in BaTiO3 [9].

*1.3.2. Electrical conductivity and superconductivity*

112 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

created by the reparations of charge [9].

*1.3.3. Catalytic activity*

In solid-state reactions, the raw materials and the final products are in the solid-state therefore nitrates, carbonates, oxides, and others can be mixed with the stoichiometric ratios. Perovskites can be synthesized via solid-state reactions by mixing carbonates or oxides of the A- and Bsite metal ions corresponding to the perovskite formula ABO3 in the required proportion to obtain the final product with the desired composition. They are ball milling effectively in an appropriate milling media of acetone or isopropanol [11, 12]. Then the obtained product is dried at 100 °C and calcined in air at 600 °C for 4–8 h under heating/cooling rates of 2 °C/min. After that, the calcined samples are ground well and sieved. Then it was calcined again at 1300– 1600 °C for 5–15 h under the heating/cooling rate of 2 °C/min to confirm the formation of single phase of perovskite. Again grinding and sieving was carried out for the calcined samples [11, 13, 14]. The synthesis of BaCeO3-based proton conductor perovskites [13] and BaCe0.95Yb0.05O3−*<sup>δ</sup>* [11] was achieved through the previous methodology using BaCO3, CeO2, and Yb2O3 as the starting materials and isopropanol as the milling media [10].

## **2.2. Gas phase preparations**

Gas phase reaction or transport can be used for the deposition of perovskite films with a specific thickness and composition. Laser ablation [15], molecular beam epitaxy [16], dc sputtering [17], magnetron sputtering [18], electron beam evaporation [19], and thermal evaporation [20] techniques were developed for gas phase deposition. Gas phase deposition can be categorized into three types: (i) deposition at a low substrate temperature then postannealing at high temperature, (ii) deposition at an intermediate temperature of 873 to 1,073 K then postanneal‐ ing treatment, and (iii) deposition at the crystallization temperature under suitable atmos‐ phere. YBa2Cu3O7 films can be synthesized by the coevaporation of Y, Cu, and BaF2 then annealing at high temperatures in O2 atmosphere wet with water vapor to reduce the annealing time and substrate interaction [20].

## **2.3. Wet chemical methods (solution preparation)**

These methods included the sol-gel preparation, coprecipitation of metal ions using precipi‐ tating agents like cyanide, oxalate, carbonate, citrate, hydroxide ions, etc., and thermal treatment [21], which resulted in a single-phase material with large surface area and high homogeneity. These methods presented good advantages such as lower temperature com‐ pared to the solid-state reactions, better homogeneity, greater flexibility in forming thin films, improved reactivity and new compositions and better control of stoichiometry, particle size, and purity. Therefore, they opened new directions for molecular architecture in the synthesis of perovskites. Solution methods were classified based on the means used for solvent removal. Two classes were identified: (i) precipitation followed by filtration, centrifugation, etc., for the separation of the solid and liquid phases and (ii) thermal treatment such as evaporation, sublimation, combustion, etc., for solvent removal. There are several factors must be taken in consideration in solution methods like solubility, solvent compatibility, cost, purity, toxicity, and choice of presumably inert anions [10].

## *2.3.1. Precipitation*

## *2.3.1.1. Oxalate-based preparation*

This method is built on the assimilation of oxalic acid with carbonates, hydroxides, or oxides producing metal oxalates, water, and carbon dioxide as products [22]. The solubility problem is minimized as the pH of the resulting solution is close to 7. An oxidizing atmosphere like oxygen was used during calcination to avoid the formation of carbide and carbon residues [23]. Clabaugh et al. utilized an aqueous chloride solution with oxalic acid to obtain unique and novel complex compound of BaTiO(C2O4)2 4H2O as a precursor for the preparation of finely divided and stoichiometric BaTiO3 [24].

## *2.3.1.2. Hydroxide-based preparation*

This method is often used due to its low solubility and the possible variety of precipitation schemes. The sol-gel process can be used to produce a wide range of new materials and improve their properties. It presented some advantages over the other traditional methods like chemical homogeneity, low calcination temperature, room temperature deposition, and controlled hydrolysis for thin film formation. BaZrO3 powders in its pure crystalline form can be prepared by the precipitation in aqueous solution of high basicity [25]. LaCoO3 was prepared by the simultaneous oxidation and coprecipitation of a mixture containing equimolar amounts of La(III) and Co(II) nitrates producing a gel containing hydroxide then calcination at 600 °C [26].

## *2.3.1.3. Acetate-based preparation*

techniques were developed for gas phase deposition. Gas phase deposition can be categorized into three types: (i) deposition at a low substrate temperature then postannealing at high temperature, (ii) deposition at an intermediate temperature of 873 to 1,073 K then postanneal‐ ing treatment, and (iii) deposition at the crystallization temperature under suitable atmos‐ phere. YBa2Cu3O7 films can be synthesized by the coevaporation of Y, Cu, and BaF2 then annealing at high temperatures in O2 atmosphere wet with water vapor to reduce the annealing

These methods included the sol-gel preparation, coprecipitation of metal ions using precipi‐ tating agents like cyanide, oxalate, carbonate, citrate, hydroxide ions, etc., and thermal treatment [21], which resulted in a single-phase material with large surface area and high homogeneity. These methods presented good advantages such as lower temperature com‐ pared to the solid-state reactions, better homogeneity, greater flexibility in forming thin films, improved reactivity and new compositions and better control of stoichiometry, particle size, and purity. Therefore, they opened new directions for molecular architecture in the synthesis of perovskites. Solution methods were classified based on the means used for solvent removal. Two classes were identified: (i) precipitation followed by filtration, centrifugation, etc., for the separation of the solid and liquid phases and (ii) thermal treatment such as evaporation, sublimation, combustion, etc., for solvent removal. There are several factors must be taken in consideration in solution methods like solubility, solvent compatibility, cost, purity, toxicity,

This method is built on the assimilation of oxalic acid with carbonates, hydroxides, or oxides producing metal oxalates, water, and carbon dioxide as products [22]. The solubility problem is minimized as the pH of the resulting solution is close to 7. An oxidizing atmosphere like oxygen was used during calcination to avoid the formation of carbide and carbon residues [23]. Clabaugh et al. utilized an aqueous chloride solution with oxalic acid to obtain unique and novel complex compound of BaTiO(C2O4)2 4H2O as a precursor for the preparation of finely

This method is often used due to its low solubility and the possible variety of precipitation schemes. The sol-gel process can be used to produce a wide range of new materials and improve their properties. It presented some advantages over the other traditional methods like chemical homogeneity, low calcination temperature, room temperature deposition, and controlled hydrolysis for thin film formation. BaZrO3 powders in its pure crystalline form can be prepared by the precipitation in aqueous solution of high basicity [25]. LaCoO3 was prepared by the simultaneous oxidation and coprecipitation of a mixture containing equimolar

time and substrate interaction [20].

**2.3. Wet chemical methods (solution preparation)**

114 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

and choice of presumably inert anions [10].

*2.3.1. Precipitation*

*2.3.1.1. Oxalate-based preparation*

divided and stoichiometric BaTiO3 [24].

*2.3.1.2. Hydroxide-based preparation*

Different perovskites were prepared by mixing acetate ions alone or together with nitrate ions with the metal ions salts. La1-*x*Sr*x*CoO3 with *x* = 0, 0.2, 0.4, 0.6 [27] was prepared using acetate precursors then calcination at 1,123 K in air for 5 h. La1-*x*Sr*x*Co1-*y*Fe*y*O3 [28] was prepared using iron nitrate and strontium, cobalt, and lanthanum acetates then calcination at 1,123 K in air between 5 and 10 h.

## *2.3.1.4. Citrate-based preparation*

Citrate precursors can be used and undergo several decomposition steps in the synthesis of perovskite [29]. These steps included the decomposition of citrate complexes and removal of CO3 2– and NO3 ¯ ions. LaCo0.4Fe0.6O3 can be prepared by this method, and the mechanism was investigated by thermogravimetry, XRD, and IR spectroscopy.

## *2.3.1.5. Cyanide-based preparation*

Rare earth orthoferrites (REFeO3) and cobalt compounds (RECoO3) were prepared using cyanides complexes via thermal decomposition of the rare earth ferricyanide and cobalticya‐ nide compounds [30]. LaFe(CN)6 6H2O, LaCo(CN)6 5H2O, and even ferrocyanides such as NH4LaFe(CN)6 5H2O are precipitated from the aqueous solution. This method presented some advantages like control of stoichiometry and low calcination temperature. The same method was used for the preparation of europium and other rare earth hexacyanoferrate compounds [31].

## *2.3.2. Thermal treatment*

## *2.3.2.1. Freeze-drying*

The freeze-drying method can be achieved through the following steps: (i) dissolution of the starting salts in the suitable solvent, water in most cases; (ii) freezing the solution very fast to keep its chemical homogeneity; (iii) freeze-drying the frozen solution to get the dehydrated salts without passing through the liquid phase; and (iv) decomposition of the dehydrated salts to give the desired perovskite powder. The rate of heat loss from the solution is the most important characteristic for the freezing step. This rate should be as high as possible to decrease the segregation of ice-salt. Also, in case of multicomponent solutions, the heat loss rate should be high to prevent the large-scale segregation of the cation components [10, 12, 21].

## *2.3.2.2. Plasma spray-drying*

This method was applicable to various precursors, including gaseous, liquid, and solid materials. It was applied for the preparation of various ceramic, electronic, and catalytic materials. It presented many advantages in terms of economy, purity, particle size distribution, and reactivity. This method was achieved through two steps: (i) injection of the reactants and (ii) generation and interaction of the molten droplets (with substrate or with the previously generated droplets). The thick film of YBa2Cu3O*<sup>x</sup>* covering large areas was prepared via this approach, and the optimum superconducting oxide phase was obtained by varying the preparation conditions like plasma parameters, substrate temperatures, and film postdeposi‐ tion treatment [32].

## *2.3.2.3. Combustion*

A redox reaction, which is thermally induced, occurs between the oxidant and fuel. A homo‐ genous, highly reactive, and nanosized powder was prepared by this method. When compared with the other traditional methods, a single-phase perovskite powder can be obtained at lower calcination temperatures or shorter reaction times. One of the most popular solution combus‐ tion methods is citrate/nitrate combustion, where citric acid is the fuel and metal nitrates are used as the source of metal and oxidant. It is similar to the Pechini process "sol-gel combustion method" to a large extent, but in citrate/nitrate combustion, ethylene glycol or other polyhy‐ droxy alcohols are not used. In addition, in citrate/nitrate combustion, the nitrates are not eliminated in the form of NO*x*, but they remain in the mixture with the metal-citrate complex facilitating the auto-combustion. Iron, cobalt, and cerium-perovskite can be prepared via citrate/nitrate combustion synthesis [12, 33]. In addition, uniform nanopowder of La0.6Sr0.4CoO3−*<sup>δ</sup>* was prepared by the combined citrate–EDTA method, where the precursor solution was made of metal nitrates, citric acid, and EDTA under controlled pH with ammonia [34]. La0.8Sr0.2Co0.2Fe0.8O3−*<sup>δ</sup>* [35] and Sr- or Ce-doped La1−*<sup>x</sup>*M*x*CrO3 catalysts [36] were prepared by citrate/nitrate combustion method. Furthermore, the Pechini "citrate gel" process includes two stages: (i) a complex was formed between the metal ions and citric acid, then (ii) the produced complex was polyesterified with ethylene glycol to maintain the metal salt solution in a gel in a homogenous state. This approach presented some advantages like high purity, minimized segregation, and good monitoring of the resulting perovskite composition. LaMnO3 [37, 38], LaCoO3 [39–41], and LaNiO3 [42] were prepared by citric acid gel process producing nanophasic thin films [10].

## *2.3.2.4. Microwave synthesis*

The microwave irradiation process (MIP), evolving from microwave sintering, was applied widely in food drying, inorganic/organic synthesis, plasma chemistry, and microwaveinduced catalysis. MIP showed fascinating advantages: (i) fast reaction rate, (ii) regular heating, and (iii) efficient and clean energy. The microwave preparations were achieved in domestic microwave oven at frequency of 2.45 GHz with 1 kW as the maximum output power. Dielectric materials absorbed microwave energy converted directly into heat energy through the polarization and dielectric loss in the interior of materials [43]. The energy efficiency reached 80–90% which is much higher than the conventional routes. MIP was recently utilized to prepare perovskites nanomaterials reducing both the high temperature of calcination (higher than 700 °C) and long time (greater than 3 h) required for pretreatment or sintering [10]. GaAlO3 and LaCrO3 perovskites with ferroelectric, superconductive, high-temperature ionic conductive and magnetic ordering properties, faster lattice diffusion, and grain size with smaller size were prepared in MIP [44–47]. CaTiO3 powders prepared in MIP presented a fast structural ordering than powders dealt in ordinary furnace [48]. Hydrothermal conventional and dielectric heating were utilized to prepare La–Ce–Mn–O catalysts. Hydrothermal MIP leads to formation of La1−*<sup>x</sup>*Ce*x*MnO3+*ε*CeO2 (*x* + *ε* = 0.2) with enhanced catalytic activity [49] while using the conventional heating methods lead to formation of LaMnO3 + CeO2. Moreover, nanosized single-phase perovskite-type LaFeO3 [50], SmFeO3, NdFeO3, GdFeO3, barium iron niobate powders [51], KNbO3 [52], PbWO4 [53], CaMoO4 [54] and MWO4 (M: Ca, Ni) [55], strontium hexaferrite [56], and SrRuO3 [57] were prepared in MIP showing finer particles, higher specific surface areas and shorter time for synthesis of single crystalline powders.

## **3. Doping of perovskites**

and reactivity. This method was achieved through two steps: (i) injection of the reactants and (ii) generation and interaction of the molten droplets (with substrate or with the previously generated droplets). The thick film of YBa2Cu3O*<sup>x</sup>* covering large areas was prepared via this approach, and the optimum superconducting oxide phase was obtained by varying the preparation conditions like plasma parameters, substrate temperatures, and film postdeposi‐

116 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

A redox reaction, which is thermally induced, occurs between the oxidant and fuel. A homo‐ genous, highly reactive, and nanosized powder was prepared by this method. When compared with the other traditional methods, a single-phase perovskite powder can be obtained at lower calcination temperatures or shorter reaction times. One of the most popular solution combus‐ tion methods is citrate/nitrate combustion, where citric acid is the fuel and metal nitrates are used as the source of metal and oxidant. It is similar to the Pechini process "sol-gel combustion method" to a large extent, but in citrate/nitrate combustion, ethylene glycol or other polyhy‐ droxy alcohols are not used. In addition, in citrate/nitrate combustion, the nitrates are not eliminated in the form of NO*x*, but they remain in the mixture with the metal-citrate complex facilitating the auto-combustion. Iron, cobalt, and cerium-perovskite can be prepared via citrate/nitrate combustion synthesis [12, 33]. In addition, uniform nanopowder of La0.6Sr0.4CoO3−*<sup>δ</sup>* was prepared by the combined citrate–EDTA method, where the precursor solution was made of metal nitrates, citric acid, and EDTA under controlled pH with ammonia [34]. La0.8Sr0.2Co0.2Fe0.8O3−*<sup>δ</sup>* [35] and Sr- or Ce-doped La1−*<sup>x</sup>*M*x*CrO3 catalysts [36] were prepared by citrate/nitrate combustion method. Furthermore, the Pechini "citrate gel" process includes two stages: (i) a complex was formed between the metal ions and citric acid, then (ii) the produced complex was polyesterified with ethylene glycol to maintain the metal salt solution in a gel in a homogenous state. This approach presented some advantages like high purity, minimized segregation, and good monitoring of the resulting perovskite composition. LaMnO3 [37, 38], LaCoO3 [39–41], and LaNiO3 [42] were prepared by citric acid gel process

The microwave irradiation process (MIP), evolving from microwave sintering, was applied widely in food drying, inorganic/organic synthesis, plasma chemistry, and microwaveinduced catalysis. MIP showed fascinating advantages: (i) fast reaction rate, (ii) regular heating, and (iii) efficient and clean energy. The microwave preparations were achieved in domestic microwave oven at frequency of 2.45 GHz with 1 kW as the maximum output power. Dielectric materials absorbed microwave energy converted directly into heat energy through the polarization and dielectric loss in the interior of materials [43]. The energy efficiency reached 80–90% which is much higher than the conventional routes. MIP was recently utilized to prepare perovskites nanomaterials reducing both the high temperature of calcination (higher than 700 °C) and long time (greater than 3 h) required for pretreatment or sintering [10]. GaAlO3 and LaCrO3 perovskites with ferroelectric, superconductive, high-temperature

tion treatment [32].

*2.3.2.3. Combustion*

producing nanophasic thin films [10].

*2.3.2.4. Microwave synthesis*

The different properties of perovskites and their catalytic activity are highly affected by the method of synthesis, conditions of calcination (time, atmosphere, fuel, temperature, etc.), and A- and/or B-site substitutions. The catalytic activity of the perovskite is highly affected by partial or total substitutions on A- and/or B-site cations because of the oxidation state modifi‐ cation, the variation of the chemical state of the elements at A- and/or B-site, the generation of oxygen vacancies, the mobility of oxygen lattice, and the formation of structural defects [58– 60]. The powerful bond between the B-site metal ions and the oxygen ions can be used to determine the basic characters of perovskites, and as a result, the B-site cation is responsible for the perovskite catalytic activity [61, 62]. Therefore, partial substitution of B-site cation with other metals M in AB1-*y*M*y*O3 will display the properties of both metals: the main metal B and the dopant one M [62]. On the other hand, the cation A can stabilize the unusual oxidation states of B-site cations by the controlled formation of crystal lattice vacancies, which lead to different catalytic performances [61].

Upon doping A- and/or B-sites in ABO3 perovskite oxides, the catalytic activity, ionic and electronic conductivity, and flexible physical and chemical properties can be altered for utilization in various applications [63–66]. Different cations with different sizes and charges can be hosted in the A- and B-sites of these perovskites; thus, many studies can be performed to utilize doped perovskites in various applications. Multiple cationic substitutions can be accepted in the stable perovskite lattice provided that Goldschmidt tolerance factor ranged between 0.75 and 1 and electroneutrality are preserved [59, 67, 68]. Therefore, variable amounts of different structural and electronic lattice defects can be accommodated in the perovskite structure as a result of their nonstoichiometry. This will further affect the activity of the perovskite and stabilize the unusual valence states of different metal ions [61, 67]. Some physical characteristics of perovskite-type oxides seriously associated with structural charac‐ ters were affected greatly by the structural deformations from the ideal cubic structure of the perovskite [69].

The type of the metal ion at the B-sites and their partial substitutions can be used to determine the catalytic activity of perovskites. The substitution of B-site metal ion with various metal ions M in the doped perovskite AB1-*y*M*y*O3 showed a vast spectrum for the alteration of the catalytic and physical properties of the prepared perovskite [64, 67]. There is an effective synergism between the crystal lattice of the perovskite and the metal ions dissolved in the lattice upon doping. This synergism resulted in enhanced redox reaction and better catalytic activity of the prepared perovskite [70]. As well, a dramatic change in the transport and magnetic properties of the ABO3 perovskite can be achieved upon doping the B-site due to an ionic valence effect and/or an ionic size effect [64]. Furthermore, upon doping the B-site in ABO3 perovskites with transition metals especially noble metals, the stability of the perovskite was improved and the catalytic activity was enhanced greatly [71]. In addition, the incorporation of two different B ions with appropriate various charge and size may be altered the simple perovskite structure. If the two ions in B-site were used with equimolar amounts, the resulted perovskite can be represented as AB0.5B′0.5O3 with unit cell appearing as doubled along the three axes. In addition, if B and B′ have different charges, there is a slight shift of the oxygen toward the highly charged ion in the ordered structure although the octahedral symmetry of B and B′ cations is main‐ tained [5]. Different B-sites doped perovskites were mentioned in the literature showing enhanced catalytic properties like LaNi*x*Co1-*x*O3 [72], LaB0.9Pd0.1O3 [73], LaMn1−*<sup>x</sup>*PdxO3 [74], BaFe1−*<sup>x</sup>*Y*x*O3−δ [75], BaFe0.85Cu0.15O3−*<sup>δ</sup>* [76], LaNi1-*x*Fe*x*O3 [77], LaFe0.95-*x*Co*x*Pd0.05O3 [78] and LaCo0.95Pd0.05O3 [79]. On the other hand, (La1-*x*Sr*x*)*y*MnO3±*δ* [80], La1-*x*Ce*x*GaO3, La1-*x*Pr*x*GaO3 and La1-*x*Nd*x*GaO3 [81], La1-*x*Ca*x*MnO3 [82], La1-*x*Na*x*MnO3+*δ*, La1-*x*Ca*x*MnO3+*<sup>δ</sup>* [83], (Ba0.93Fe0.07)TiO3 [84], La1-*x*Sr*x*NiO3, and La1-*x*Sr*x*MnO3 [85] are examples of A-site doped perovskites.

## **4. Characterization of perovskites**

X-ray powder diffraction (XRD) can be used to differentiate the different phases of the prepared perovskites. Single-crystal XRD is another analysis used to characterize the structure of the perovskite. Thermal stability of the prepared perovskites can be tested using thermal analysis techniques like TGA, DTA, and DSC. On the other hand, scanning (SEM) and transmission (TEM) electron microscopies can be utilized to identify the different morphological and surface characteristics of the prepared perovskites. Also, BET can be utilized for surface area measurement. In addition, Fourier transform infrared spectrosco‐ py (FTIR) and X-ray photoelectron spectroscopy (XPS) can be used to completely identify the formed phases [10, 86–88].

## **4.1. XRD**

XRD can be used for the phase identification and the relative percents of different phases of the prepared materials. Also, some structural parameters like particle size, lattice parameters (a, b, and c), lattice volume, and theoretical density can be calculated from the XRD data. Also, XRD can be used to optimize the preparation conditions of the different perovskites [3, 87– 89]. Galal et al. prepared SrPdO3 by citrate/nitrate combustion method at different pH values; 2, 7, and 10 at calcination temperature 750 °C for 3 h and the XRD patterns of these samples were shown in Figure 3A. The XRD data were compared with the ICCD card of SrPdO3 (card number 00-025-0908). For pH 2, the experimental data and the theoretical one are well matched and supported the formation of the primary orthorhombic perovskite phase of SrPdO3 (the major diffraction peak 110) and the appearance of secondary phase SrPd3O4 (210). Only SrPd3O4 phase appeared in case of samples prepared at pH 7 and 10 as (110) peak disappeared. Therefore, pH 2 was the optimal pH for SrPdO3 preparation. Also, the type of fuel (citric acid, urea, and glycine) used in the preparation of SrPdO3 can be optimized using XRD (Figure 3B). SrPdO3 was the primary phase in all cases but with different percents of SrPdO3 (110) with respect to SrPd3O4 (210). The high percent was in case of urea and the small one in case of citric acid. Some structural parameters were calculated and summarized in Table 3 with good matching with theoretical data.

M in the doped perovskite AB1-*y*M*y*O3 showed a vast spectrum for the alteration of the catalytic and physical properties of the prepared perovskite [64, 67]. There is an effective synergism between the crystal lattice of the perovskite and the metal ions dissolved in the lattice upon doping. This synergism resulted in enhanced redox reaction and better catalytic activity of the prepared perovskite [70]. As well, a dramatic change in the transport and magnetic properties of the ABO3 perovskite can be achieved upon doping the B-site due to an ionic valence effect and/or an ionic size effect [64]. Furthermore, upon doping the B-site in ABO3 perovskites with transition metals especially noble metals, the stability of the perovskite was improved and the catalytic activity was enhanced greatly [71]. In addition, the incorporation of two different B ions with appropriate various charge and size may be altered the simple perovskite structure. If the two ions in B-site were used with equimolar amounts, the resulted perovskite can be represented as AB0.5B′0.5O3 with unit cell appearing as doubled along the three axes. In addition, if B and B′ have different charges, there is a slight shift of the oxygen toward the highly charged ion in the ordered structure although the octahedral symmetry of B and B′ cations is main‐ tained [5]. Different B-sites doped perovskites were mentioned in the literature showing enhanced catalytic properties like LaNi*x*Co1-*x*O3 [72], LaB0.9Pd0.1O3 [73], LaMn1−*<sup>x</sup>*PdxO3 [74], BaFe1−*<sup>x</sup>*Y*x*O3−δ [75], BaFe0.85Cu0.15O3−*<sup>δ</sup>* [76], LaNi1-*x*Fe*x*O3 [77], LaFe0.95-*x*Co*x*Pd0.05O3 [78] and LaCo0.95Pd0.05O3 [79]. On the other hand, (La1-*x*Sr*x*)*y*MnO3±*δ* [80], La1-*x*Ce*x*GaO3, La1-*x*Pr*x*GaO3 and La1-*x*Nd*x*GaO3 [81], La1-*x*Ca*x*MnO3 [82], La1-*x*Na*x*MnO3+*δ*, La1-*x*Ca*x*MnO3+*<sup>δ</sup>* [83], (Ba0.93Fe0.07)TiO3

118 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

[84], La1-*x*Sr*x*NiO3, and La1-*x*Sr*x*MnO3 [85] are examples of A-site doped perovskites.

X-ray powder diffraction (XRD) can be used to differentiate the different phases of the prepared perovskites. Single-crystal XRD is another analysis used to characterize the structure of the perovskite. Thermal stability of the prepared perovskites can be tested using thermal analysis techniques like TGA, DTA, and DSC. On the other hand, scanning (SEM) and transmission (TEM) electron microscopies can be utilized to identify the different morphological and surface characteristics of the prepared perovskites. Also, BET can be utilized for surface area measurement. In addition, Fourier transform infrared spectrosco‐ py (FTIR) and X-ray photoelectron spectroscopy (XPS) can be used to completely identify

XRD can be used for the phase identification and the relative percents of different phases of the prepared materials. Also, some structural parameters like particle size, lattice parameters (a, b, and c), lattice volume, and theoretical density can be calculated from the XRD data. Also, XRD can be used to optimize the preparation conditions of the different perovskites [3, 87– 89]. Galal et al. prepared SrPdO3 by citrate/nitrate combustion method at different pH values; 2, 7, and 10 at calcination temperature 750 °C for 3 h and the XRD patterns of these samples were shown in Figure 3A. The XRD data were compared with the ICCD card of SrPdO3 (card number 00-025-0908). For pH 2, the experimental data and the theoretical one are well matched

**4. Characterization of perovskites**

the formed phases [10, 86–88].

**4.1. XRD**

*Figure 3: XRD patterns of SrPdO3 prepared by combustion method at different pH values (A) and different fuels (B). Miller indices (h, l,k) are written in black line for SrPdO3, red line for SrPd3O4 and the symbol (\*) for SrCl2. 6H2O [89].* **Figure 3.** XRD patterns of SrPdO3 prepared by combustion method at different pH values (A) and different fuels (B). Miller indices (h, l,k) are written in black line for SrPdO3, red line for SrPd3O4 and the symbol (\*) for SrCl2. 6H2O [89].


2

*Figure 4: SEM micrographs of (A) LaNiO3, (B) LaCoO3, (C) LaFeO3 and (D) LaMnO3 prepared by the* 

*microwave-assisted citrate method at 720 W for 30 min, with a magnification of 35,000 times [90].*

**Table 3.** Structural parameters calculated from XRD data [89].

## **4.2. SEM and TEM**

SEM and TEM can be used to study the morphology and surface characteristics of the perovskite nanomaterials. The preparation conditions, synthesis method, type of A- and B-site metal ions, and doping A- and/or B-sites affected greatly the SEM of the prepared perovskites [88–94]. Galal et al. prepared LaNiO3, LaCoO3, LaFeO3, and LaMnO3 by the microwave-assisted citrate method at 720 W as operating power for 30 min under microwave irradiation. The SEM images for the different perovskites were shown in Figure 4 presenting different morphologies depending on the kind of metal ion at B-site, respectively. LaNiO3 showed compact surface with high degree of ordering while LaCoO3 and LaMnO3 showed spherical grains agglomer‐ ations with smaller grain size in case of LaMnO3. LaFeO3 showed dissimilar morphology with a porous surface containing particles with bonelike shape. In addition, LaFeO3 presented greater electrocatalytic activity toward hydrogen evolution reaction compared to other types of perovskites [90].

Furthermore, the high-resolution TEM (HRTEM) can be used to show the different morphol‐ ogies and particle characteristics of the different perovskites [86, 88, 95]. HRTEM images for LaNiO3, LaCoO3, LaFeO3, and LaMnO3 by the microwave-assisted citrate method were shown in Figure 5, respectively. The HRTEM images clearly showed orthorhombic phase with high crystallinity in case of LaFeO3, while HRTEM images of LaNiO3, LaCoO3, and LaMnO3 showed hexagonal distorted rhombohedral phases. The diffraction patterns obtained via HRTEM for the different perovskites were comparable with the XRD data [86, 95].

**Figure 4.** SEM micrographs of (A) LaNiO3, (B) LaCoO3, (C) LaFeO3 and (D) LaMnO3 prepared by the microwave-assist‐ ed citrate method at 720 W for 30 min, with a magnification of 35,000 times [90].

*Figure 5: HRTEM micrographs of (A) LaNiO3, (B) LaCoO3, (C) LaFeO3 and (D) LaMnO3 prepared by the microwave-assisted citrate method at 720Wfor 30 min [86]. Figure\_6*  **Figure 5.** HRTEM micrographs of (A) LaNiO3, (B) LaCoO3, (C) LaFeO3 and (D) LaMnO3 prepared by the microwaveassisted citrate method at 720Wfor 30 min [86].

(A) (B)

## **4.3. BET**

*sol–gel method [87].*

**4.2. SEM and TEM**

120 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

of perovskites [90].

SEM and TEM can be used to study the morphology and surface characteristics of the perovskite nanomaterials. The preparation conditions, synthesis method, type of A- and B-site metal ions, and doping A- and/or B-sites affected greatly the SEM of the prepared perovskites [88–94]. Galal et al. prepared LaNiO3, LaCoO3, LaFeO3, and LaMnO3 by the microwave-assisted citrate method at 720 W as operating power for 30 min under microwave irradiation. The SEM images for the different perovskites were shown in Figure 4 presenting different morphologies depending on the kind of metal ion at B-site, respectively. LaNiO3 showed compact surface with high degree of ordering while LaCoO3 and LaMnO3 showed spherical grains agglomer‐ ations with smaller grain size in case of LaMnO3. LaFeO3 showed dissimilar morphology with a porous surface containing particles with bonelike shape. In addition, LaFeO3 presented greater electrocatalytic activity toward hydrogen evolution reaction compared to other types

Furthermore, the high-resolution TEM (HRTEM) can be used to show the different morphol‐ ogies and particle characteristics of the different perovskites [86, 88, 95]. HRTEM images for LaNiO3, LaCoO3, LaFeO3, and LaMnO3 by the microwave-assisted citrate method were shown in Figure 5, respectively. The HRTEM images clearly showed orthorhombic phase with high crystallinity in case of LaFeO3, while HRTEM images of LaNiO3, LaCoO3, and LaMnO3 showed hexagonal distorted rhombohedral phases. The diffraction patterns obtained via HRTEM for

**Figure 4.** SEM micrographs of (A) LaNiO3, (B) LaCoO3, (C) LaFeO3 and (D) LaMnO3 prepared by the microwave-assist‐

ed citrate method at 720 W for 30 min, with a magnification of 35,000 times [90].

the different perovskites were comparable with the XRD data [86, 95].

The electrochemical performance and electrocatalytic activity of the perovskites are greatly associated with the specific surface area of the materials; therefore, it is necessary to measure the specific surface area of the prepared materials. The surface area values of different perovskites can be measured by Brunauer–Emmett–Teller (BET) nitrogen adsorption. The preparation conditions, synthesis method, type of A- and B-site metals, and presence of different dopants can greatly affect the surface area of the prepared perovskites [86, 87, 95]. Biniwale et al. prepared LaFeO3 via different methods, namely, sol-gel, combustion, and coprecipitation and measured the corresponding surface area and the average pore diameter of the prepared perovskites. The order of decreasing the surface area of the prepared LaFeO3 was sol-gel (16.5 m2 g–1) > combustion (9.3 m2 g–1) > coprecipitation method (5.4 m2 g–1). The order of increasing the average pore diameter of the prepared LaFeO3 was sol-gel (119 °A) <

3

*FTIR spectra for LaFeO3 synthesized using (a) co-precipitation method, (b) combustion method and (c)* 

 *[27] and (B)* 

*Figure 6: (A) TG spectrum of Sr and Pd mixed citrate complex, heating rate was 10 °C min<sup>−</sup><sup>1</sup>*

coprecipitation method (140 °A) < combustion (205 °A). Sol-gel and combustion methods resulted in porous surface with internal pores contributing to higher surface area, while coprecipitation method resulted in less internal pores and lower surface area due to longer calcination time [87]. (C) (D)

*Figure\_5*  (A) (B)

## **4.4. Thermal analysis**

Thermal analysis can be utilized to identify the thermal stability and the decomposition temperature of the prepared perovskites. The optimum calcination temperature of any perovskite can be identified using thermal analysis [88, 96, 97]. Galal et al. prepared SrPdO3 for the first time, and its optimum calcination temperature was investigated using TGA and DTG of the citrate complex of Sr and Pd (Figure 6A). The breakdown of the citrate complex occurs at ~330 °C through a smooth weight loss step. SrPdO3 formation was achieved at ~750 °C through a sharp weight loss step [96]. *Figure 5: HRTEM micrographs of (A) LaNiO3, (B) LaCoO3, (C) LaFeO3 and (D) LaMnO3 prepared by the microwave-assisted citrate method at 720Wfor 30 min [86].*

*FTIR spectra for LaFeO3 synthesized using (a) co-precipitation method, (b) combustion method and (c) sol–gel method [87].* **Figure 6.** (A) TG spectrum of Sr and Pd mixed citrate complex, heating rate was 10 °C min−1 [27] and (B) FTIR spectra for LaFeO3 synthesized using (a) co-precipitation method, (b) combustion method and (c) sol–gel method [87].

 *[27] and (B)* 

3

## **4.5. FTIR**

The chemical bonding and chemical structure of the prepared perovskites can be investigated via FTIR. The FTIR can give structural confirmation similar to that obtained via XRD [87, 97– 100]. Biniwale et al. prepared LaFeO3 via different methods: sol-gel, combustion, and copre‐ cipitation, and the FTIR for them was shown in Figure 6B [87]. The FTIR of LaFeO3 showed an absorption band at 558 cm–1 related to the stretching vibration mode of Fe-O. Another band appeared at 430 cm–1 was related to the deformation vibration mode of O-Fe-O. LaFeO3 prepared via coprecipitation method showed a sharp band at 3609 cm–1, which is related to La-O in lanthanum oxide. In case of the other two methods, the band at 3600 cm–1 disappeared indicating the formation of relatively pure perovskite phase. Other bands appeared at 1360 and 1480 cm–1, indicating other phases in case of coprecipitation method. As a result and as mentioned in literature, the absorption peak around 558 cm–1 was related to the stretching modes of metallic oxygen bond [87, 97–100].

## **4.6. XPS**

coprecipitation method (140 °A) < combustion (205 °A). Sol-gel and combustion methods resulted in porous surface with internal pores contributing to higher surface area, while coprecipitation method resulted in less internal pores and lower surface area due to longer

(C) (D)

*Figure\_5*  (A) (B)

Thermal analysis can be utilized to identify the thermal stability and the decomposition temperature of the prepared perovskites. The optimum calcination temperature of any perovskite can be identified using thermal analysis [88, 96, 97]. Galal et al. prepared SrPdO3 for the first time, and its optimum calcination temperature was investigated using TGA and DTG of the citrate complex of Sr and Pd (Figure 6A). The breakdown of the citrate complex occurs at ~330 °C through a smooth weight loss step. SrPdO3 formation was achieved at ~750

*Figure 5: HRTEM micrographs of (A) LaNiO3, (B) LaCoO3, (C) LaFeO3 and (D) LaMnO3 prepared by the* 

*Figure\_6*  (A) (B)

3

The chemical bonding and chemical structure of the prepared perovskites can be investigated via FTIR. The FTIR can give structural confirmation similar to that obtained via XRD [87, 97– 100]. Biniwale et al. prepared LaFeO3 via different methods: sol-gel, combustion, and copre‐ cipitation, and the FTIR for them was shown in Figure 6B [87]. The FTIR of LaFeO3 showed an absorption band at 558 cm–1 related to the stretching vibration mode of Fe-O. Another band appeared at 430 cm–1 was related to the deformation vibration mode of O-Fe-O. LaFeO3

*FTIR spectra for LaFeO3 synthesized using (a) co-precipitation method, (b) combustion method and (c)* 

**Figure 6.** (A) TG spectrum of Sr and Pd mixed citrate complex, heating rate was 10 °C min−1 [27] and (B) FTIR spectra for LaFeO3 synthesized using (a) co-precipitation method, (b) combustion method and (c) sol–gel method [87].

 *[27] and (B)* 

*Figure 6: (A) TG spectrum of Sr and Pd mixed citrate complex, heating rate was 10 °C min<sup>−</sup><sup>1</sup>*

calcination time [87].

**4.4. Thermal analysis**

*sol–gel method [87].*

**4.5. FTIR**

°C through a sharp weight loss step [96].

*microwave-assisted citrate method at 720Wfor 30 min [86].*

122 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The surface compositions of the various components of the prepared perovskites can be identified via XPS [101–106]. Lee et al. prepared La0.9FeO3 and LaFeO3 samples and identify their structural composition via XPS analysis [102]. Figure 7 showed the XPS spectra of La (3d), Fe (2p), and O (1s) in La0.9FeO3 and LaFeO3 samples. The binding energy of La (3d5/2) was 833.5 eV and 833.8 eV in case of LaFeO3 and La0.9FeO3, respectively, corresponding to the La3+ ions in the form of oxide. By contrast, the binding energy of Fe (2p3/2) was 710.2 eV for both samples corresponding to Fe3+ ions in the form of oxide. The Fe (2p) XPS signal cannot distinguish between Fe3+ and Fe4+. The XPS signal of O (1s) was divided into two peaks in case of La0.9FeO3 appearing at 529.9 and 532.1 eV. While for LaFeO3, O (1s) XPS signal was divided into three peaks appearing at 529.4, 531.9, and 534.4 eV. The O (1s) binding energy values at 529.9 and 529.4 eV in both samples are attributed to lattice oxygen species. The peaks at 532.1 and 531.9 eV are ascribed to the chemisorbed oxygen species as OH¯ or O¯. The chemisorbed oxygen species appeared at binding energy higher than that of lattice oxygen species by 2.1– 2.5 eV. The peak appeared at 534.4 eV in case of LaFeO3 was ascribed to the adsorbed water species associated with the surface lanthanum oxide which is highly hygroscopic [102].

**Figure 7.** XPS spectra of La(3d), Fe(2p) and O(1s) in La0.9FeO3 and LaFeO3 samples [102].

## **5. Applications of perovskites**

Inorganic perovskite-type oxides exhibited attractive physical and chemical characteristics such as electronic conductivity, electrically active structure, the oxide ions mobility through the crystal lattice, variations on the content of the oxygen, thermal and chemical stability, and supermagnetic, photocatalytic, thermoelectric, and dielectric properties. They are fascinating nanomaterials for wide applications in catalysis, fuel cells, and electrochemical sensing. The catalytic activity of these oxides is higher than that of many transition metals compounds and even some precious metal oxides.

Nanoperovskites are recently utilized in electrochemical sensing of alcohols, gases, amino acids, acetone, glucose, H2O2, and neurotransmitters exhibiting good selectivity, sensitivity, unique long-term stability, excellent reproducibility, and anti-interference ability. Moreover, they have been utilized as catalysts in oxygen reduction and hydrogen evolution reactions exhibiting high electrocatalytic activity, lower activation energy, and high electron transfer kinetics. In addition, some perovskites are promising candidates for the development of effective anodic catalysts for direct fuel cells showing high catalytic performance.

## **5.1. Sensors and biosensors**

## *5.1.1. Gas sensors*

There are a number of requirements that the materials utilized as gas sensors must satisfy, namely, good resemblance with the target gases, manufacturability, hydrothermal stability, convenient electronic structure, resistance to poisoning, and adaptation with existing technol‐ ogies. There is a wide variety of materials that can be used as gas sensors like semiconductors, namely, SnO2, In2O3, and WO3, and perovskite oxides, namely, LaFeO3 and SrTiO3. Perovskite oxides are fascinating materials as gas sensors due to their perfect thermal stability, ideal band gap "3–4 eV," and difference in size between the cations of A- and B-sites, allowing different dopants addition for controlling semiconducting properties and their catalytic properties. Perovskites including titanates, ferrites, and cobaltates were utilized as gas sensors for detecting CO, NO2, methanol, ethanol, and hydrocarbons [107–110]. LaCoO3 prepared via high-energy ball milling exhibited the highest amount of grain boundaries, the best CO gas sensing properties, and the smallest crystallite size of 11 nm compared to that prepared via solid-state and sol-gel reactions. The maximum response ratio increased to 26% in case of milling method with maximum response ratio temperature of 125 °C compared to 7% and 17% in case of solid-state reaction and sol-gel method. In addition, the specific surface area increased greatly from 4 m2 g–1 to 66 m2 g–1 by extra milling step, and the mobility of the oxygen was enhanced by growing the extra milling step and surface area [107]. A summary of various perovskite oxides for different gas sensing was given in Table 4.

## *5.1.2. Glucose sensor*

It is very important to analytically determine H2O2 and glucose in many fields like food, clinic, and pharmaceutical analyses. H2O2 is considered one of the most important oxidizing agents


**Table 4.** A summary of different perovskites for gas sensing.

**5. Applications of perovskites**

124 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

even some precious metal oxides.

**5.1. Sensors and biosensors**

*5.1.1. Gas sensors*

*5.1.2. Glucose sensor*

Inorganic perovskite-type oxides exhibited attractive physical and chemical characteristics such as electronic conductivity, electrically active structure, the oxide ions mobility through the crystal lattice, variations on the content of the oxygen, thermal and chemical stability, and supermagnetic, photocatalytic, thermoelectric, and dielectric properties. They are fascinating nanomaterials for wide applications in catalysis, fuel cells, and electrochemical sensing. The catalytic activity of these oxides is higher than that of many transition metals compounds and

Nanoperovskites are recently utilized in electrochemical sensing of alcohols, gases, amino acids, acetone, glucose, H2O2, and neurotransmitters exhibiting good selectivity, sensitivity, unique long-term stability, excellent reproducibility, and anti-interference ability. Moreover, they have been utilized as catalysts in oxygen reduction and hydrogen evolution reactions exhibiting high electrocatalytic activity, lower activation energy, and high electron transfer kinetics. In addition, some perovskites are promising candidates for the development of

There are a number of requirements that the materials utilized as gas sensors must satisfy, namely, good resemblance with the target gases, manufacturability, hydrothermal stability, convenient electronic structure, resistance to poisoning, and adaptation with existing technol‐ ogies. There is a wide variety of materials that can be used as gas sensors like semiconductors, namely, SnO2, In2O3, and WO3, and perovskite oxides, namely, LaFeO3 and SrTiO3. Perovskite oxides are fascinating materials as gas sensors due to their perfect thermal stability, ideal band gap "3–4 eV," and difference in size between the cations of A- and B-sites, allowing different dopants addition for controlling semiconducting properties and their catalytic properties. Perovskites including titanates, ferrites, and cobaltates were utilized as gas sensors for detecting CO, NO2, methanol, ethanol, and hydrocarbons [107–110]. LaCoO3 prepared via high-energy ball milling exhibited the highest amount of grain boundaries, the best CO gas sensing properties, and the smallest crystallite size of 11 nm compared to that prepared via solid-state and sol-gel reactions. The maximum response ratio increased to 26% in case of milling method with maximum response ratio temperature of 125 °C compared to 7% and 17% in case of solid-state reaction and sol-gel method. In addition, the specific surface area increased greatly from 4 m2 g–1 to 66 m2 g–1 by extra milling step, and the mobility of the oxygen was enhanced by growing the extra milling step and surface area [107]. A summary of various

It is very important to analytically determine H2O2 and glucose in many fields like food, clinic, and pharmaceutical analyses. H2O2 is considered one of the most important oxidizing agents

effective anodic catalysts for direct fuel cells showing high catalytic performance.

perovskite oxides for different gas sensing was given in Table 4.

in chemical and food industries. Glucose is a fundamental metabolite for most of the living organisms and for the clinical examination of diabetes mellitus, a worldwide health problem. As a result, it is very important to construct biosensors for the sensitive determination of H2O2 and glucose [128–137]. Different types of enzymatic glucose sensors were constructed and used in the literature exhibiting the advantages of simplicity and sensitivity. However, enzymatic glucose sensors suffered from the lack of stability and the difficult procedures required for the effective immobilization of enzyme on the electrode surface. The lack of enzyme stability was attributed to its intrinsic nature because the enzyme activity was highly affected by poisonous chemicals, pH, temperature, humidity, etc. As a result, most attention was given for sensitive, simple, stable, and selective nonenzymatic glucose sensor. Different novel materials were proposed for the electrocatalytic oxidation of glucose like noble nano‐ metals, nanoalloys, metal oxides, and inorganic perovskite oxides. Inorganic perovskite oxides as nanomaterials exhibited fascinating properties for glucose sensing like ferroelectricity, superconductivity, charge ordering, high thermopower, good biocompatibility, catalytic activity, and the ability of the perovskite structure to accommodate different metallic ions [128–137]. Zhen Zhang et al. utilized carbon paste electrode modified with Co0.4Fe0.6LaO3 as a promising nonenzymatic H2O2 and glucose sensor. This sensor displayed perfect electrocata‐ lytic activity toward H2O2 and glucose oxidation in alkaline medium due to the presence of large amount of active sites in the modifier. The linear dynamic range for H2O2 at this surface was 0.01 to 800 μM with low detection limit of 2.0 nM. For glucose, two ranges were obtained from 0.05 to 5 μM and from 5 to 500 μM with detection limit of 10 nM. The proposed sensor exhibited rapid response, excellent long-term stability, and anti-interference ability toward ascorbic acid, uric acid, and dopamine [132]. Furthermore, Atta et al. have recently modified SrPdO3 perovskite with a film of gold nanoparticles to be utilized as a nonenzymatic voltam‐ metric glucose sensor. The studied sensor exhibited high electrocatalytic activity toward glucose oxidation exploring the effective synergism between SrPdO3 and gold nanoparticles. SrPdO3 perovskite facilitated the charge transfer process and acted as an effective supporting substrate for gold nanoparticles. The catalytic activity of SrPdO3 was attributed to the defi‐ ciency of the surface for oxygen which resulted in enhanced intrinsic reactivity toward glucose oxidation. Another reason for the catalytic activity of SrPdO3 was the matrix effect induced by the stable crystal structure of the perovskite where there is a homogenous distribution of Pd4+ cations in the inert matrix of the perovskite during the reaction. This nanocomposite showed good performance toward glucose sensing in terms of highly reproducible response, high sensitivity, wide linearity, low detection limit, good selectivity, long-term stability, and applicability in real urine samples and blood serum [137]. A summary of different types of perovskites used for enzymatic and nonenzymatic H2O2 and glucose sensing was given in Table 5, exhibiting high sensitivity, wide linear range, low detection limit, anti-interference ability, applicability in real samples, and long-term stability.

## *5.1.3. Neurotransmitters sensor*

Dopamine (DA) is an essential catecholamine neurotransmitter that exists in the mammalian central nervous system. The depletion of DA can lead to Parkinson's disease; therefore, its determination is very crucial. The interference of ascorbic acid (AA) and uric acid (UA) with DA is the major problem in DA detection [89, 140–143]. Therefore, it is very important to present a modified surface which can be sensitively and selectively detect DA even in presence


**Table 5.** A summary of different perovskites for H2O2 and glucose sensing.

As a result, it is very important to construct biosensors for the sensitive determination of H2O2 and glucose [128–137]. Different types of enzymatic glucose sensors were constructed and used in the literature exhibiting the advantages of simplicity and sensitivity. However, enzymatic glucose sensors suffered from the lack of stability and the difficult procedures required for the effective immobilization of enzyme on the electrode surface. The lack of enzyme stability was attributed to its intrinsic nature because the enzyme activity was highly affected by poisonous chemicals, pH, temperature, humidity, etc. As a result, most attention was given for sensitive, simple, stable, and selective nonenzymatic glucose sensor. Different novel materials were proposed for the electrocatalytic oxidation of glucose like noble nano‐ metals, nanoalloys, metal oxides, and inorganic perovskite oxides. Inorganic perovskite oxides as nanomaterials exhibited fascinating properties for glucose sensing like ferroelectricity, superconductivity, charge ordering, high thermopower, good biocompatibility, catalytic activity, and the ability of the perovskite structure to accommodate different metallic ions [128–137]. Zhen Zhang et al. utilized carbon paste electrode modified with Co0.4Fe0.6LaO3 as a promising nonenzymatic H2O2 and glucose sensor. This sensor displayed perfect electrocata‐ lytic activity toward H2O2 and glucose oxidation in alkaline medium due to the presence of large amount of active sites in the modifier. The linear dynamic range for H2O2 at this surface was 0.01 to 800 μM with low detection limit of 2.0 nM. For glucose, two ranges were obtained from 0.05 to 5 μM and from 5 to 500 μM with detection limit of 10 nM. The proposed sensor exhibited rapid response, excellent long-term stability, and anti-interference ability toward ascorbic acid, uric acid, and dopamine [132]. Furthermore, Atta et al. have recently modified SrPdO3 perovskite with a film of gold nanoparticles to be utilized as a nonenzymatic voltam‐ metric glucose sensor. The studied sensor exhibited high electrocatalytic activity toward glucose oxidation exploring the effective synergism between SrPdO3 and gold nanoparticles. SrPdO3 perovskite facilitated the charge transfer process and acted as an effective supporting substrate for gold nanoparticles. The catalytic activity of SrPdO3 was attributed to the defi‐ ciency of the surface for oxygen which resulted in enhanced intrinsic reactivity toward glucose oxidation. Another reason for the catalytic activity of SrPdO3 was the matrix effect induced by the stable crystal structure of the perovskite where there is a homogenous distribution of Pd4+ cations in the inert matrix of the perovskite during the reaction. This nanocomposite showed good performance toward glucose sensing in terms of highly reproducible response, high sensitivity, wide linearity, low detection limit, good selectivity, long-term stability, and applicability in real urine samples and blood serum [137]. A summary of different types of perovskites used for enzymatic and nonenzymatic H2O2 and glucose sensing was given in Table 5, exhibiting high sensitivity, wide linear range, low detection limit, anti-interference

126 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

ability, applicability in real samples, and long-term stability.

Dopamine (DA) is an essential catecholamine neurotransmitter that exists in the mammalian central nervous system. The depletion of DA can lead to Parkinson's disease; therefore, its determination is very crucial. The interference of ascorbic acid (AA) and uric acid (UA) with DA is the major problem in DA detection [89, 140–143]. Therefore, it is very important to present a modified surface which can be sensitively and selectively detect DA even in presence

*5.1.3. Neurotransmitters sensor*

of high concentration of AA and UA. Atta et al. presented carbon paste electrode modified with SrPdO3 (CpE/SrPdO3) as a promising electrochemical DA sensor in biological fluids with unique long-term stability and low detection limit of 9.3 nmol L−1. CpE/SrPdO3 can simulta‐ neously detect DA in the presence of high concentrations of AA and UA and can successfully detect DA in human urine samples with excellent recovery results in terms of selectivity, accuracy, precision, and detection limit. The proposed sensor showed high sensitivity, good selectivity, and anti-interference ability [89]. Moreover, higher response toward DA oxidation was achieved at CpE/SrPdO3 compared to electrodeposited palladium nanoparticles modified CpE (CpE/Pd) with equivalent loading of Pd4+ salt. The higher catalytic activity at CpE/ SrPdO3 was explained in terms of the oxygen–surface interaction between the oxygen atoms of the hydroxyl groups and the transition element in the perovskite. One characteristic of perovskite is the deficiency of its surface for oxygen. As a result, the "dihydroxy"-oxygen adsorb onto perovskite surface with the formation of a "moderate" bond between oxygen atoms and the transition element in the oxide [89]. Furthermore, the descriptor that controls the catalytic process in perovskites is the type of transition metal in the perovskite, which is related to the number of occupied d orbital states of a specific symmetry, for example, of the active metal. This is associated with the surface ability to bond oxygen on the basis of the calculations of the density functional theory. Therefore, the oxygen adsorption energy represented a perfect descriptor for the oxidation of DA at CpE/SrPdO3. A summary of different perovskites used for neurotransmitters sensing was given in Table 6.


**Table 6.** A summary of different perovskites for neurotransmitters sensing.

## **5.2. Solid oxide fuel cells**

Fuel cells have come into view as efficient alternatives to combustion engines due to their potential to minimize the environmental influence of the use of fossil fuels. A fuel cell uses certain type of chemical fuel as its energy source, and there is a direct transformation of chemical energy into electrical energy like a battery. Fuel cells are attractive because of their great efficiency, modular and distributed nature, low emissions, zero noise pollution, and role in any future hydrogen fuel economy. There are several types of fuel cells depending on operating temperature, fuel type, electrolyte type, and mobile ions. Polymer electrolyte membrane fuel cells, molten carbonate, phosphoric acid or alkali fuel cells, and solid oxide fuel cell are the most common examples of fuel cells [144]. Table 7 contained some fuel cells types and some selected features [144]. Here we will concern on solid oxide fuel cells. Solid oxide fuel cells (SOFCs), based on conducting electrolyte in the form of an oxide-ion, can generate electricity and heat and they are considered as energy-saving, environment-friendly, and effective energy conversion devices. SOFCs exhibited several features compared to the other types of fuel cells like high-energy conversion efficiency, cheap materials, low sensitivity to the fuel impurities, low pollution emissions, environmental compatibility, and excellent fuel flexibility [145–160]. Figure 8 showed the working principle of a solid oxide fuel cell [157]. The high temperature of SOFC operation resulted in the difficult choice of the proper materials and the decreased cell durability. Thus, providing materials for SOFCs with good performance at intermediate temperatures (500–800 °C) is very essential so that the cell cost and the startup and shut down time can be reduced and its long-term stability can be improved [148, 155]. Perovskite oxides exhibited fascinating properties like good electrical conductivity similar to that of metals, high ionic conductivity, and perfect mixed ionic and electronic conductivity. Depending on the differences in the electrical conductive characteristics of perovskites, they are chosen as an effective component in SOFC [9]. In addition, mixed-conduction perovskite oxides possess beneficial electrochemical reaction; structural, thermal, and chemical stabilities; high electrical conductivity; high catalytic activity toward the oxygen reduction; and ideal mixed electronic and ionic conductivities to be used as effective component for intermediate temperatures SOFC (IT-SOFC) [147–160]. Shao and Haile utilized Ba0.5Sr0.5Co0.8Fe0.2O3-*δ* as an effective cathode for intermediate SOFC with the fuel of humidified H2 and the cathode gas of air. This cathode exhibited 1010 and 402 mW cm–2 as the maximum power density at 600 and 500 °C, respectively [145]. On the other hand, Goodenough reported the use of double perovskite Sr2MgMnMoO6-*<sup>δ</sup>* as an anode material for SOFC with dry methane as the fuel and 438 mW cm–2 as the maximum power density at 800 °C. This anode material showed long-term stability, stability in reducing atmosphere, tolerance to sulfur, and characteristic of oxygen deficiency [161]. Table 8 contained a summary of different perovskites used as anode and cathode for SOFC illustrating the fuel type, the operating temperature, and the maximum power density.

atoms and the transition element in the oxide [89]. Furthermore, the descriptor that controls the catalytic process in perovskites is the type of transition metal in the perovskite, which is related to the number of occupied d orbital states of a specific symmetry, for example, of the active metal. This is associated with the surface ability to bond oxygen on the basis of the calculations of the density functional theory. Therefore, the oxygen adsorption energy represented a perfect descriptor for the oxidation of DA at CpE/SrPdO3. A summary of

CpE/SrPdO3 DA 7 – 70 μmol L−1 9.3 [89]

LaFeO3 nanoparticles DA 1.5×10−7 – 8.0×10−4 M 30 [143]

Fuel cells have come into view as efficient alternatives to combustion engines due to their potential to minimize the environmental influence of the use of fossil fuels. A fuel cell uses certain type of chemical fuel as its energy source, and there is a direct transformation of chemical energy into electrical energy like a battery. Fuel cells are attractive because of their great efficiency, modular and distributed nature, low emissions, zero noise pollution, and role in any future hydrogen fuel economy. There are several types of fuel cells depending on operating temperature, fuel type, electrolyte type, and mobile ions. Polymer electrolyte membrane fuel cells, molten carbonate, phosphoric acid or alkali fuel cells, and solid oxide fuel cell are the most common examples of fuel cells [144]. Table 7 contained some fuel cells types and some selected features [144]. Here we will concern on solid oxide fuel cells. Solid oxide fuel cells (SOFCs), based on conducting electrolyte in the form of an oxide-ion, can generate electricity and heat and they are considered as energy-saving, environment-friendly, and effective energy conversion devices. SOFCs exhibited several features compared to the other types of fuel cells like high-energy conversion efficiency, cheap materials, low sensitivity to the fuel impurities, low pollution emissions, environmental compatibility, and excellent fuel flexibility [145–160]. Figure 8 showed the working principle of a solid oxide fuel cell [157]. The high temperature of SOFC operation resulted in the difficult choice of the proper materials

**Detection limit (nM)**

0.6 – 9 μmol L−1 1.63 [141]

DA 2 ×10-8 – 1.6 × 10-6 M 59 [140]

DA 8.2 × 10−8 – 1.6 × 10−7 M 62 [142]

**Reference**

different perovskites used for neurotransmitters sensing was given in Table 6.

**Perovskite Sensing for Linear dynamic range**

128 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

L-dopa EP, NE, DA, DOPAC and ST

**Table 6.** A summary of different perovskites for neurotransmitters sensing.

LaFeO3 microspheres made up of nanospheres

Graphite/SrPdO3

LaFeO3 nanostructure dendrites

**5.2. Solid oxide fuel cells**

**Figure 8.** Schematic diagram showing the working principle of a solid oxide fuel cell [157].


**Table 7.** Fuel cell types and selected features [144].


**Table 8.** A summary of different perovskites for SOFC.

## **5.3. Catalyst**

**Type Temperature °C Fuel Electrolyte Mobile ion**

AFC: Alkali fuel cell 100–250 H2 Aqueous KOH OH¯

70–110 H2, methane Sulphonated polymer

150–250 H2 H3PO4 H+

500–700 Hydrocarbons, CO (Na, K)2CO3 CO3 2-

700–1000 Hydrocarbons, CO (Zr, Y)O2-δ O2-

(~3% H2O)

Ba0.5Sr0.5Co0.2Fe0.8O3-δ Cathode Humidified H2 800 266 [147] La0.7Sr0.3Co0.5Fe0.5O3 Cathode Not reported Not reported Not reported [148] La0.6Sr0.4Fe0.8Co0.2O3 Cathode glycerol 800 Not reported [149]

Y0.8Ca0.2BaCoFeO5+δ Cathode Humidified H2 650 426 [151]

SO2

Sm0.5Sr0.5CoO3-δ cathode Not reported 700 936 [158] Sr2Fe1.4Mo0.6O6−ı cathode H2 or methanol Not reported Not reported [159] La0.75Sr0.25Cr0.5Mn0.5O3 anode methane Not reported Not reported [162] Sr2MgMnMoO6-δ anode dry methane 800 438 [161] La0.8Sr0.2Cr0.97V0.03O3 anode dry methane 800 Not reported [163]

**Fuel used Operating**

600 500

550

550

620 650

**temperature °C**

**Maximum power density mW cm-2**

1010 402

327 105

432 171

0.154 0.265 **Reference**

[145]

[150]

[150]

[153]

(Nafion)

(H2O)nH+

PEM: polymer electrolyte membrane or proton exchange membrane

PAFC: Phosphoric acid

SOFC: Solid oxide fuel

**Table 7.** Fuel cell types and selected features [144].

**Perovskite Anode/cathode in**

**the cell**

130 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Ba0.5Sr0.5Co0.8Fe0.2O3-δ Cathode Humidified H2

LaBaCuFeO5+x Cathode Humidified H2 700

LaBaCuCoO5+x Cathode Humidified H2 700

NdFeO3 anode sulfur vapor or

**Table 8.** A summary of different perovskites for SOFC.

fuel cell

cell

MCFC: Molten carbonate fuel cell Perovskite oxides can be widely used as catalyst in modern chemical industry, exhibiting appropriate solid-state, surface, and morphological properties [6]. Several perovskites exhibited enhanced catalytic activity toward different reactions like hydrogen evolution and oxygen evolution and reduction reactions [9].

## *5.3.1. Hydrogen evolution reaction*

Because of the advantages of high heat of combustion, abundant sources, and no pollution, hydrogen is considered as an ideal fuel. Hydrogen evolution reaction (HER) is a fascinating reaction in the renewable energy field. This reaction is very important in (i) metal electrode‐ position and corrosion in acids, (ii) storage of energy through production of hydrogen, and (iii) as the hydrogen oxidation reaction reverse in low-temperature fuel cells. One of the most studied reactions in electrochemistry is the electrocatalysis in HER. The material used for HER should have (i) intrinsic electrocatalytic activity, (ii) considerable active surface area per unit volume, and (iii) good stability. To reduce the cost of electrolytic HER, the overpotential required for the operation of the electrolyzer at considerable current densities should be reduced. The overpotential reduction can be achieved through the electrode active surface area enhancement or by the selection of electrode materials of high catalytic activity. The steps of the reaction in acidic solutions are as follows:

$$\rm{M} + \rm{H}\_{3}\rm{O}^{+} + \rm{e}^{-} \leftrightarrow \rm{MH}\_{\rm{ads}} + \rm{H}\_{2}\rm{O} \tag{1}$$

$$\text{MH}\_{\text{ads}} + \text{H}\_{\text{3}}\text{O}^{+} + \text{e}^{-} \leftrightarrow \text{H}\_{2} + \text{M} + \text{H}\_{2}\text{O} \tag{2}$$

$$\text{MH}\_{\text{ads}} + \text{MH}\_{\text{ads}} \leftrightarrow \text{H}\_2 + 2\text{M} \tag{3}$$

The first step in HER is the proton discharge (volume reaction, Eq. (1)), which is followed by electrodesorption step (Heyrovsky reaction, Eq. (2)) or proton recombination step, physical desorption, (Tafel reaction, Eq. (3)) [93]. Galal et al. confirmed the high catalytic activity of different perovskite oxides toward hydrogen evolution reaction [90–94, 96]. LnFeO3 perov‐ skites (Ln= Gd, La, Sm, and Nd) were prepared by the microwave assistant-citrate method, and single-phase perovskites were formed with uniform distribution of small average particle size. Tafel and electrochemical impedance measurements were used to study the catalytic activity of LnFeO3 toward HER showing the effect of the type of the lanthanide ion on HER and the partial substitution effect at the La-site in La1-*y*Sm*y*FeO3. The order of decreasing the catalytic activity toward HER was NdFeO3 > LaFeO3 > SmFeO3 > GdFeO3 based on activation energies calculations and the strength of Fe-O bond, which is related to A-type metal ion. Furthermore, the order of decreasing the catalytic activity in case of doped samples was La0.75Sm0.25FeO3 > La0.5Sm0.5FeO3 > La0.25Sm0.75FeO3 > LaFeO3 > SmFeO3 displaying greater catalytic activity of ternary perovskites compared to that of binary ones [94]. On the other hand, Galal et al. prepared SrPdO3 by the citrate method for the first time showing enhanced catalytic activity toward HER up to 100 times with respect to the unmodified surface with 27.9 kJ mol −1 as the calculated activation energy. The rate-determining step was the hydrogen adsorption on the catalyst and the order of the reaction at the catalyst surface was 0.86 [96]. Table 9 contained a summary of different perovskites used as catalysts for HER with the values of exchange current density at constant overpotential, activation energy, reaction order, and the rate-determining step.


**Table 9.** A summary of different perovskites for HER catalysis.

## *5.3.2. Oxygen reduction and oxygen evolution reactions*

Galal et al. prepared SrPdO3 by the citrate method for the first time showing enhanced catalytic activity toward HER up to 100 times with respect to the unmodified surface with 27.9 kJ mol −1 as the calculated activation energy. The rate-determining step was the hydrogen adsorption on the catalyst and the order of the reaction at the catalyst surface was 0.86 [96]. Table 9 contained a summary of different perovskites used as catalysts for HER with the values of exchange current density at constant overpotential, activation energy, reaction order, and the

**Reaction**

**order Rate determining step Reference**

the catalyst [94]

the catalyst [94]

the catalyst [94]

the catalyst [94]

the catalyst [94]

[96]

adsorption of hydrogen on the

catalyst

rate-determining step.

**Exchange current**

132 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

SrPdO3 −946.3 27.9 0.86

**Table 9.** A summary of different perovskites for HER catalysis.

**Activation energy kJ mol-1**

NdFeO3 -107.82 24.68 0.72 adsorption of hydrogen on

LaFeO3 -367.37 41.15 0.37 adsorption of hydrogen on

SmFeO3 -147.23 57.89 0.41 adsorption of hydrogen on

GdFeO3 -451.86 81.37 0.72 adsorption of hydrogen on

La0.75Sm0.25FeO3 Not reported Not reported Not reported adsorption of hydrogen on

LaFeO3 −105.6 51.61 0.37 adsorption of hydrogen [90] LaCoO3 −348.9 45.37 0.74 adsorption of hydrogen [90] LaNiO3 −158.5 41.15 0.94 adsorption of hydrogen [90] LaMnO3 −41.9 55.05 0.87 adsorption of hydrogen [90] SrRuO3 (microwave) −5326.0 6.67 1.14 adsorption of hydrogen [92]

−425.5 49.45 0.98 adsorption of hydrogen [92]

−298.0 86.32 0.88 adsorption of hydrogen [92]



**density j/ μA.cm-2**

**Perovskite**

SrRuO3 (solcombustion)

SrRuO3 (coprecipitation)

CaRuO3 (coprecipitation)

BaRuO3 (coprecipitation)

Oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) are considered one of the most important electrode reactions in many industrial processes like fuel cells, metal electrowinning,water electrolysis, electro-organic synthesis, cathodicprotection,andrecharge‐ able metal air batteries [95, 164–166]. Platinum-based catalysts and precious metal oxides are the most common catalysts for ORR or OER, but they are expensive and scarce. Therefore, it is very important to develop other catalysts for ORR or OER. Mixed metal perovskite oxides of transition and rare earth metals are promising low-cost alternatives to precious metal cata‐ lysts for both ORR and OER [165]. Perovskite oxide exhibited unique electronic and magnetic properties, defective structure, and good cation ordering resulting in disorder-free channels of oxygen vacancies and enhanced mobility of oxygen ions [164]. Some perovskite oxides were reportedas electrocatalysts forORRandOERandsummarizedinTable 10 [95, 164–166].Ruizhi Yangetal.preparedBa0.5Sr0.5Co0.8Fe0.2O3bysol-gelmethodandutilizeditasORRorOERcatalyst in basic medium of KOH. The proposed catalyst exhibited higher catalytic activity toward OER than the unmodified electrode [164]. Furthermore, Galal et al. prepared LaFeO3 by microwaveassisted citrate method and studied its catalytic activity toward OER in acid medium of HClO4. LaFeO3 exhibited greater electrocatalytic activity toward OER by about 100-folds compared to the unmodified electrode. The current density at 1.5 V increased from 3.6 × 10−5 in case of unmodifiedelectrodeto1.2×10−3A/cm2 incaseofmodifiedone.Thecalculatedactivationenergy was 20 kJ/mol, which is much lower than that reported for other iron compounds and even some precious metal oxides like RuO2. This was attributed to the matrix effect induced by the stable crystal structure of the perovskite [95]. In addition, La0.6Ca0.4CoO3, prepared by the solgel method, showed high catalytic activity and relative stability toward oxygen electrochemis‐ try in basic medium of KOH. La0.6Ca0.4CoO3 exhibited single-phase perovskite structure, high conductivity, and large surface area [165, 166].


**Table 10.** A summary of different perovskites for ORR or OER catalysis.

## **5.4. Solar cells**

Solar energy is a green source of energy that can be used instead of energy sources based on fossil fuels. Solar radiation can be directly converted into electrical energy in a suitable way creating various applications for solar energy. Solar energy can be efficiently converted into electricity using photovoltaic solar cells based on silicon. The obvious disadvantage of siliconbased solar cell is the high price of electricity generated from it so that there is a potential need to develop solar cell with low cost. Recently, attention was paid to solar cells based on organic/ inorganic solid-state methylammonium lead halide (CH3NH3PbX3, X=Br, I) hybrid perov‐ skite. This type of solar cells presented effective points such as a conversion efficiency of about 20%; its cost is lower than that of conventional silicon solar cells and the availability of the raw materials. These 3D organometal halide perovskite exhibited excellent intrinsic properties for photovoltaic applications like excellent stability, appropriate band gap (~1.55 eV), high absorption coefficient (1.5 × 104 cm–1 at 550 nm), long hole-electron diffusion length (~100 nm for CH3NH3PbI3 and ~1 μm for CH3NH3PbI3-*x*Cl*x*), high carrier mobility and transport, charge carriers with small effective mass, low temperature of processing, and easy processing steps [167–179]. Figure 9A showed the structure of ABX3 perovskite (X = oxygen, carbon, nitrogen or halogen), where A and B cations are placed in a cubo-octahedral and an octahedral site, respectively. A is usually divalent and B is tetravalent when O2– anion is used. On the other hand, Figure 9B showed the structure of CH3NH3PbI3 halide perovskite where the A-site is occupied by CH3NH3 + (organic component) and B-site cation is occupied by Pb2+. As indicated by the example, halide perovskites contain monovalent and divalent cations in A- and B-sites, respectively, to maintain electrical neutrality [169, 171]. Like oxide perovskite, the tolerance factor of halide perovskite should be as close to one to maintain a stable and symmetrical crystal structure [169].

**Figure 9.** (a) ABX3 perovskite structure showing BX6 octahedral and larger A cation occupied in cubo-octahedral site. (b) Unit cell of cubic CH3NH3PbI3 perovskite [171].

The quality of the perovskite film is very crucial for solar cells. Several methods have been used to form perovskite films with high quality such as single step solution method, vapor assistant solution process, sequential deposition of inorganic and organic precursor, and coevaporation of the precursors [167]. CH3NH3PbI3 perovskite film was prepared with high quality by adding small amounts of N-methyl-2-pyrrolidone and a mixture of g-butyrolactone and dimethylsulfoxide via a solution method. A power conversion efficiency of 11.77% with fill factor of 80.52% was obtained based on the structure of ITO/PEDOT:PSS (poly(3,4ethylenedioxythiophene):polystyrenesulfonate)/perovskite/ PCBM (fullerene-derivative phenyl-C61-butyric acid methyl ester)/Ca/Al under one sun illumination (100 mW cm–2) [167]. Table 11 contained a summary of different models based on perovskites used for solar cells applications with the values of power conversion efficiency, fill factor, method of perovskite formation, solar cell composition, cost, and stability.


**Table 11.** A summary of different models based perovskites for Solar cells applications

## **6. Conclusions**

creating various applications for solar energy. Solar energy can be efficiently converted into electricity using photovoltaic solar cells based on silicon. The obvious disadvantage of siliconbased solar cell is the high price of electricity generated from it so that there is a potential need to develop solar cell with low cost. Recently, attention was paid to solar cells based on organic/ inorganic solid-state methylammonium lead halide (CH3NH3PbX3, X=Br, I) hybrid perov‐ skite. This type of solar cells presented effective points such as a conversion efficiency of about 20%; its cost is lower than that of conventional silicon solar cells and the availability of the raw materials. These 3D organometal halide perovskite exhibited excellent intrinsic properties for photovoltaic applications like excellent stability, appropriate band gap (~1.55 eV), high

for CH3NH3PbI3 and ~1 μm for CH3NH3PbI3-*x*Cl*x*), high carrier mobility and transport, charge carriers with small effective mass, low temperature of processing, and easy processing steps [167–179]. Figure 9A showed the structure of ABX3 perovskite (X = oxygen, carbon, nitrogen or halogen), where A and B cations are placed in a cubo-octahedral and an octahedral site, respectively. A is usually divalent and B is tetravalent when O2– anion is used. On the other hand, Figure 9B showed the structure of CH3NH3PbI3 halide perovskite where the A-site is

by the example, halide perovskites contain monovalent and divalent cations in A- and B-sites, respectively, to maintain electrical neutrality [169, 171]. Like oxide perovskite, the tolerance factor of halide perovskite should be as close to one to maintain a stable and symmetrical crystal

**Figure 9.** (a) ABX3 perovskite structure showing BX6 octahedral and larger A cation occupied in cubo-octahedral site.

The quality of the perovskite film is very crucial for solar cells. Several methods have been used to form perovskite films with high quality such as single step solution method, vapor assistant solution process, sequential deposition of inorganic and organic precursor, and coevaporation of the precursors [167]. CH3NH3PbI3 perovskite film was prepared with high quality by adding small amounts of N-methyl-2-pyrrolidone and a mixture of g-butyrolactone and dimethylsulfoxide via a solution method. A power conversion efficiency of 11.77% with fill factor of 80.52% was obtained based on the structure of ITO/PEDOT:PSS (poly(3,4-

cm–1 at 550 nm), long hole-electron diffusion length (~100 nm

(organic component) and B-site cation is occupied by Pb2+. As indicated

absorption coefficient (1.5 × 104

+

134 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

(b) Unit cell of cubic CH3NH3PbI3 perovskite [171].

occupied by CH3NH3

structure [169].

Inorganic perovskite-type oxides are excellent nanomaterials for wide applications in catalysis, fuel cells, and electrochemical sensing, exhibiting attractive physical and chemical character‐ istics. They showed electronic conductivity, electrically active structure, the oxide ions mobility through the crystal lattice, variations on the content of the oxygen, thermal and chemical stability and supermagnetic, photocatalytic, thermoelectric, and dielectric properties. Nanoperovskites have been utilized as catalysts in oxygen reduction and hydrogen evolution reactions exhibiting high electrocatalytic activity, lower activation energy and high electron transfer kinetics. In addition, some perovskites are promising candidates for the development of effective anodic catalysts for direct fuel cells showing high catalytic performance. Moreover, they are recently utilized in electrochemical sensing of alcohols, gases, glucose, H2O2, and neurotransmitters. They can enhance the catalytic performance in terms of unique long-term stability, sensitivity, excellent reproducibility, selectivity, and anti-interference ability. In addition, organometallic halide perovskites exhibited efficient intrinsic properties to be utilized as a photovoltaic solar cell with good stability and high efficiency.

## **Acknowledgements**

The authors would like to acknowledge the financial support from Cairo University through the Vice President Office for Research Funds.

## **Author details**

Nada F. Atta\* , Ahmed Galal and Ekram H. El-Ads

\*Address all correspondence to: Nada\_fah1@yahoo.com

Department of Chemistry, Faculty of Science, Cairo University, Giza, Egypt

## **References**


[7] Fu1 D, Itoh M. Ferroelectrics—Material Aspects. In: Lallart M, editor. InTech: Shizuo‐ ka University, Tokyo Institute of Technology, Japan; 2011. p. 413–442. Chapter 20.

reactions exhibiting high electrocatalytic activity, lower activation energy and high electron transfer kinetics. In addition, some perovskites are promising candidates for the development of effective anodic catalysts for direct fuel cells showing high catalytic performance. Moreover, they are recently utilized in electrochemical sensing of alcohols, gases, glucose, H2O2, and neurotransmitters. They can enhance the catalytic performance in terms of unique long-term stability, sensitivity, excellent reproducibility, selectivity, and anti-interference ability. In addition, organometallic halide perovskites exhibited efficient intrinsic properties to be

The authors would like to acknowledge the financial support from Cairo University through

[1] Wolfram T, Ellialtioğlu S. Electronic and Optical Properties of d-band Perovskites.

[2] Galasso FS. Structure, Properties and Preparation of Perovskite-Type Compounds. In: Smoluchowski R, Kurti N, editors. 1st ed. Pergamon Press: New York; 1969. p. 3–

[3] Ali SMM. Synthesis of Nano-particles Using Microwave Technique, the Study of their Physical Properties and Some Applications [PhD thesis]. Faculty of Science

[5] Johnsson M, Lemmens P. Crystallography and Chemistry of Perovskites. 2005. p. 1–

[6] Peňa MA, Fierro JLG. Chemical structures and performance of perovskite oxides.

Chemical Reviews. 2001;101:1981–2017. DOI: 10.1016/j.apsusc.2014.10.168.

utilized as a photovoltaic solar cell with good stability and high efficiency.

**Acknowledgements**

**Author details**

Nada F. Atta\*

**References**

49. Chapter 2.

11.

Cairo University; 2009.

[4] Perovskite Perfect Lattice. Chapter 3. p. 79–114.

the Vice President Office for Research Funds.

, Ahmed Galal and Ekram H. El-Ads

Department of Chemistry, Faculty of Science, Cairo University, Giza, Egypt

1st ed. Cambridge University Press: New York; 2006. 1 p.

\*Address all correspondence to: Nada\_fah1@yahoo.com

136 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications


ductor yttrium barium copper oxide (YBa2Cu3O7) by electron-beam codeposition. Journal of Materials Research. 1987;2:713–725. DOI: 10.1557/JMR.1987.0713.


[32] Cuomo JJ, Guarnieri C, Richard M, Shivashankar SA, Roy RA, Yee DS, Rosenberg R. Large area plasma spray deposited superconducting YBa2Cu3O7 thick films. Ad‐ vanced Ceramic Materials. 1989;2(3B):422–429.

ductor yttrium barium copper oxide (YBa2Cu3O7) by electron-beam codeposition.

Journal of Materials Research. 1987;2:713–725. DOI: 10.1557/JMR.1987.0713.

conductors. Applied Physics Letters. 1987;51(21):1753–1755.

of Materials. 1989;1:659–664. DOI: 10.1021/cm00006a019.

Dekker Inc. Ed. Chemical Industries. 1993;50.

138 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

National Bureau of Standards. 1956;56:289–291.

265–275. DOI: 10.1016/0021-9517(90)90302-Z.

10.1016/0040-6031(70)85017-1.

j.jeurceramsoc.2008.09.001.

[20] Mankiewich PM, Scofield JH, Skocpol WJ, Howard RE, Dayem AH. Good E., Repro‐ ducible technique for fabrication of thin films of high transition temperature super‐

[21] Tejuca LG, Fiero JLG. Properties and Applications of Perovskite Type-Oxides. Marcel

[22] Gallagher PK, Fleming DA. Influence of oxygen partial pressure on the synthesis of barium yttrium copper oxide (Ba2YCu3O7) from a novel oxalate precursor. Chemistry

[23] Gallagher PK, Schrey F. The thermal decomposition of freeze-dried tantalum and mixed lithium–niobium oxalate. Thermochimica Acta. 1970;1:465–476. DOI:

[24] Clabaugh WS, Swiggard EM, Gilchrist RJ. Preparation of barium titanyl oxalate tetra‐ hydrate for conversion to barium titanate of highly purity. Journal of Research of the

[25] Boschini F, Rulmont A, Cloots R, Vertruyen B. Rapid synthesis of submicron crystal‐ line barium zirconate BaZrO3 by precipitation in aqueous basic solution below 100 °C. Journal of the European Ceramic Society 2009;29(8):1457–1462. DOI: 10.1016/

[26] Barnard KR, Foger K, Turney TW, Williams RD. Lanthanum cobalt oxide oxidation catalysts derived from mixed hydroxide precursors. Journal of Catalysis. 1990;125(2):

[27] Nakamura T, Misono M, Yoneda Y. Reduction and oxidation and catalytic properties

[28] Zhang HM, Shimizu Y, Teraoka Y, Miura N, Yamazoe N. Oxygen sorption and cata‐ lytic properties of La1−xSrxCo1−yFeyO3 Perovskite-type oxides. Journal of Catalysis.

[29] Zhang HM, Teraoka Y, Yamazoe N. Preparation of perovskite type oxides with large

[30] Gallagher PK. A simple method for the preparation of (RE)FeO3 and (RE)CoO3. Mate‐ rials Research Bulletin. 1967; 3,225–232. DOI: 10.1016/0025-5408(68)90123-2.

[31] Gallagher PK, Prescott B. Further studies of the thermal decomposition of europium hexacyanoferrate(III) and ammonium europium hexacyanoferrate(II). Inorganic

of perovskite type mixed oxide catalysts. Chemistry Letters. 1981;1589.

1990;121(2):432–440. DOI: 10.1016/0021-9517(90)90251-E.

Chemistry. 1970;9:2510–2512. DOI: 10.1021/ic50093a026.

surface area by citrate process. Chemistry Letters. 1987:665–668.


[56] Fu Y, Lin C. Fe/Sr ratio effect on magnetic properties of strontium ferrite powders synthesized by microwave-induced combustion process. Journal of Alloys and Com‐ pounds 2005;386(1–2):222–227. DOI: 10.1016/j.jallcom.2004.04.148.

[45] Gibbons KE, Blundell SJ, Mihut AI, Gameson I, Edwards PP, Miyazaki Y, Hyatt NC, Jones MO, Porch A. Rapid synthesis of colossal magnetoresistance manganites by mi‐ crowave dielectric heating. Chemical Communications. 2000;1:159–160. DOI: 10.1039/

[46] Selvam MP, Rao KJ. Microwave preparation and sintering of industrially important perovskite oxides: LaMO3 (M = Cr, Co, Ni). Journal of Materials and Chemistry.

[47] Yan H, Huang X, Lu Z, Hu H, Xue R, Chen L. Microwave synthesis of LiCoO2 cath‐ ode materials. Journal of Power Sources. 1997;68:530–532. DOI: 10.1016/

[48] Cavalcante LS, Marques VS, Sczancoski JC, Escote MT, Joya MR, Varela JA, Santos MRMC, Pizani PS, Longo E. Synthesis, structural refinement and optical behavior of CaTiO3 powders: a comparative study of processing in different furnaces. Chemical

[49] Kaddouri A, Gelin P, Dupont N. Methane catalytic combustion over La–Ce–Mn–O– perovskite prepared using dielectric heating. Catalysis Communications. 2009;10(7):

[50] Farhadi S, Momeni Z, Taherimehr M. Rapid synthesis of perovskite-type LaFeO3 nanoparticles by microwave-assisted decomposition of bimetallic La[Fe(CN)6] 5H2O compound. Journal of Alloys and Compounds. 2009;471(1–2):L5–L8. DOI: 10.1016/

[51] Charoenthai N, Traiphol R, Rujijanagul G. Microwave synthesis of barium iron nio‐ bate and dielectric properties. Materials Letters. 2008;62(29):4446–4448. DOI: 10.1016/

[52] Paula AJ, Parra R, Zaghete MA, Varela JA. Synthesis of KNbO3 nanostructures by a microwave assisted hydrothermal method. Materials Letters. 2008;62(17–18):2581–

[53] Ryu JH, Koo S, Chang DS, Yoon J, Lim CS, Shim KB. Microwave-assisted synthesis of PbWO4 nano-powders via a citrate complex precursor and its photoluminescence. Ceramics International. 2006;32(6):647–652. DOI: 10.1016/j.ceramint.2005.04.025. [54] Ryu JH, Koo S, Chang DS, Yoon J, Lim CS, Shim KB. Microwave-assisted synthesis of CaMoO4 nano-powders by a citrate complex method and its photoluminescence property Journal of Alloys and Compounds. 2005;390(1–2);245–249. DOI: 10.1016/

[55] Ryu JH, Yoon J, Lim CS, Oh W, Shim KB. Microwave-assisted synthesis of nanocrys‐ talline MWO4 (M: Ca, Ni) via water-based citrate complex precursor. Ceramics Inter‐

national. 2005;31(6):883–888. DOI: 10.1016/j.ceramint.2004.09.015.

A907677H.

S0378-7753(96)02565-7.

j.jallcom.2008.03.113.

j.matlet.2008.07.047.

j.jallcom.2004.07.064.

2003;13:596–601. DOI: 10.1039/B211602B.

140 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Engineering Journal. 2008;143(1–3):299–301.

1085–1089. DOI: 10.1016/j.catcom.2008.12.063.

2584. DOI: 10.1016/j.matlet.2007.12.059.


[79] Ke-bin Z, Hong-de C, Qun T, Bao-wei Z, Di-xin S, Xiao-bai X. Synergistic effect of palladium and oxygen vacancies in the Pd/perovskites catalysts synthesized by SPC method. Journal of environmental sciences. 2005;17(1):19–24. PMID: 15900751.

[67] Alifanti M, Auer R, Kirchnerova J, Thyrion F, Grange P, Delmon B. Activity in meth‐ ane combustion and sensitivity to sulfur poisoning of La1−xCexMn1−yCoyO3 perovskite oxides. Applied Catalysis, B. 2003;41:71–81. DOI: 10.1016/S0926-3373(02)00194-7.

[68] Kida T, Yamasaki A, Watanabe K, Yamazoe N, Shimanoe K. Oxygen-permeable membranes based on partially B-site substituted BaFe1−yMyO3−δ (M=Cu or Ni). Journal

[69] Sora IN, Caronna T, Fontana F, Fernàndez CdJ, Caneschi A, Green M. Crystal struc‐ tures and magnetic properties of strontium and copper doped lanthanum ferrites. Journal of Solid State Chemistry. 2012;191:33–39. DOI: 10.1016/j.jssc.2012.02.020.

[70] Rodrίguez GCM, Saruhan B. Effect of Fe/Co-ratio on the phase composition of Pdintegrated perovskites and its H2-SCR of NOx performance. Applied Catalysis, B.

[71] Ke-bin Z, Hong-de C, Qun T, Bao-wei Z, Di-xin S, Xiao-bai X. Synergistic effect of palladium and oxygen vacancies in the Pd/perovskite catalysts synthesized by the SPC method. Journal of Environmental Sciences. 2005;17(1):19–24. PMID: 15900751.

[72] Silva GRO, Santos JC, Martinelli DMH, Pedrosa AMG, de Souza MJB, Melo DMA. Synthesis and characterization of LaNixCo1-xO3 perovskites via complex precursor methods. Materials Sciences and Applications, 2010;1:39–45. DOI: 10.4236/msa.

[73] Russo N, Palmisano P, Fino D. Pd substitution effects on perovskite catalyst activity for methane emission control. Chemical Engineering Journal. 2009;154:137–141. DOI:

[74] Li C, Jiang B, Fanchiang W, Lin Y. The effect of Pd content in LaMnO3 for methanol partial oxidation. Catalysis Communications. 2011;16:165–169. DOI: 10.1016/j.catcom.

[75] Liu X, Zhao H, Yang J, Li Y, Chen T, Lu X, Ding W, Li F. Lattice characteristics, struc‐ ture stability and oxygen permeability of BaFe1−xYxO3−δ ceramic membranes. Journal

[76] Kida T, Yamasaki A, Watanabe K, Yamazoe N, Shimanoe K. Oxygen-permeable membranes based on partially B-site substituted BaFe1–yMyO3– δ (M= Cu or Ni). Jour‐ nal of Solid State Chemistry. 2010;183:2426–2431. DOI: 10.1016/j.jssc.2010.08.002.

[77] Jahangiri A, Aghabozorg H, Pahlavanzadeh H. Effects of Fe substitutions by Ni in LaNiO perovskite-type oxides in reforming of methane with CO2 and O2. Interna‐ tional of Hydrogen Energy. 2013;38:10407–10416. DOI: 10.1016/j.ijhydene.2013.05.080.

[78] Rodrìguez GCM, Saruhan B. Effect of Fe/Co-ratio on the phase composition of Pdintegrated perovskites and its H2-SCR of NOx performance. Applied Catalysis B: En‐

vironmental. 2010;93:304–313. DOI: 10.1016/j.apcatb.2009.10.004.

of Membrane Science. 2011;383:235– 240. DOI: 10.1016/j.memsci.2011.08.059.

2010;93:304–313. DOI: 10.1016/j.apcatb.2009.10.004.

142 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

2010.12008.

2011.09.028.

10.1016/j.cej.2009.05.015.

of Solid State Chemistry. 2010;183:2426–2431. DOI: 10.1016/j.jssc.2010.08.002.


[103] Pawlak DA, Ito M, Oku M, Shimamura K, Fukuda T. Interpretation of XPS O (1s) in mixed oxides proved on mixed perovskite crystals. Journal of Physical Chemistry B. 2002;106:504–507. DOI: 10.1021/jp012040a.

[91] Galal A, Atta NF, Ali SM. Optimization of the synthesis conditions for LaNiO3 cata‐ lyst by microwave assisted citrate method for hydrogen production. Applied Cataly‐

[92] Galal A, Darwish SA, Atta NF, Ali SM, Abd El Fatah AA. Synthesis, structure and catalytic activity of nano-structured Sr–Ru–O type perovskite for hydrogen produc‐ tion. Applied Catalysis A: General. 2010;378:151–159. DOI: 10.1016/j.apcata.

[93] Atta NF, Galal A, Ali SM. The catalytic activity of ruthenates ARuO3 (A= Ca, Sr or Ba) for the hydrogen evolution reaction in acidic medium. International Journal of

[94] Atta NF, Galal A, Ali SM. The effect of the lanthanide ion-type in LnFeO3 on the cata‐ lytic activity for the hydrogen evolution in acidic medium. International Journal of

[95] Ali SM, Abd Al-Rahman YM, Galal A. Catalytic activity toward oxygen evolution of LaFeO3 prepared by the microwave assisted citrate method. Journal of Electrochemi‐

[96] Galal A, Atta NF, Darwish SA, Abd El Fatah AA, Ali SM. Electrocatalytic evolution of hydrogen on a novel SrPdO3 perovskite electrode. Journal of Power Sources.

[97] Ghosh S, Dasgupta S. Synthesis, characterization and properties of nanocrystalline perovskite cathode materials. Materials Science—Poland. 2010;28(2):427–438.

[98] Viruthagiri G, Praveen P, Mugundan S, Gopinathan E. Synthesis and characteriza‐ tion of pure and nickel doped SrTiO3 nanoparticles via solid state reaction route. In‐

[99] Li S. Preparation and characterization of perovskite structure lanthanum gallate and lanthanum aluminate based oxides [PhD thesis]. Royal Institute of Technology Stock‐

[100] Pecchi G, Campos CM, Jiliberto MG, Delgado EJ, Fierro JLG. Effect of additive Ag on the physicochemical and catalytic properties of LaMn0.9Co0.1O3.5 perovskite. Applied

[101] Lakshminarayanan N, Choi H, Kuhn1 JN, Ozkan US. Effect of additional B-site tran‐ sition metal doping on oxygen transport and activation characteristics in La0.6Sr0.4(Co0.18Fe0.72X0.1)O3−δ (where X= Zn, Ni or Cu) perovskite oxides. Applied Cat‐

alysis B: Environmental. 2011;103:318–325. DOI: 10.1016/j.apcatb.2011.01.038.

[102] Cho Y, Choi K, Kim Y, Jung J, Lee S. Characterization and catalytic properties of sur‐ face La-rich LaFeO3 perovskite. Bulletin of the Korean Chemical Society. 2009;30(6):

Catalysis A: General. 2009;371:78–84. DOI: 10.1016/j.apcata.2009.09.031.

sis A: General. 2011;409–410:202–208. DOI: 10.1016/j.apcata.2011.10.005.

Electrochemical Science. 2012;7:725–746. DOI: 71772857.

144 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Electrochemical Science. 2014;9:2132–2148. DOI: 94967196.

cal Society. 2012;159(9):F600–F605. DOI: 10.1149/2.063209jes.

2010;195:3806–3809. DOI: 10.1016/j.jpowsour.2009.12.091.

dian Journal of Advances in Chemical Science. 2013;1(3):132–138.

2010.02.015.

holm, Sweden; 2009.

1368–1372.


[128] Jia F, Zhong H, Zhang W, Li X, Wanga G, Songa J, Cheng Z, Yin J, Guo L. A novel nonenzymatic ECL glucose sensor based on perovskite LaTiO3-Ag0.1 nanomaterials. Sensors and Actuators B. 2015;212:174–182. DOI: 10.1016/j.snb.2015.02.011.

[116] Liu X, Hu J, Cheng B, Qin H, Jiang M. Preparation and gas sensing characteristics of p-type semiconducting LnFe0.9Mg0.1O3 (Ln = Nd, Sm, Gd and Dy) materials. Current

[117] Najjar H, Batis H. La–Mn perovskite-type oxide prepared by combustion method: Catalytic activity in ethanol oxidation. Applied Catalysis A: General. 2010;383:192–

[118] Sun L, Qin HW, Cao E, Zhao M, Gao F, Hu J. Gas-sensing properties of perovskite La0.875Ba0.125FeO3 nanocrystallinepowders. Journal of Physics and Chemistry of Sol‐

[119] Kong L, Shen Y. Gas-sensing property and mechanism of CaxLa1-xFeO3 ceramics. Sen‐ sors and Actuators B. 1996;30:217–221. DOI: 10.1016/0925-4005(96)80052-9.

[120] Feng C, Ruan S, Li J, Zou B, Luo J, Chen W, Dong W, Wu F. Ethanol sensing proper‐ ties of LaCoxFe1−xO3 nanoparticles: effects of calcination temperature, Co-doping, and carbon nanotube-treatment. Sensors and Actuators B. 2011;155:232–238. DOI:

[121] Khetre SM. Ethanol gas sensing properties of nano-porous LaFeO3 thick film. Sensors

[122] Zhang L, Hu J, Song P, Qin H, Jiang M. Electrical properties and ethanol-sensing characteristics of perovskite La1−xPbxFeO3. Sensors and Actuators B. 2006;114:836–840.

[123] Martinelli G, Carotta MC, Ferroni M, Sadaoka Y, Traversa E. Screen-printed perov‐ skite-type thick films as gas sensors for environmental monitoring. Sensors and Ac‐

[124] Carotta MC, Butturi MA, Martinelli G, Sadaoka Y, Nunziante P, Traversa E. Micro‐ structural evolution of nanosized LaFeO3 powders from the thermal decomposition of a cyano-complex for thick film gas sensors. Sensors and Actuators B. 1997;44:590–

[125] Di Bartolomeo E, Grilli ML, Yoon JW, Traversa E. NOx sensors based on interfacing nano-sized LaFeO3 perovskite-type oxide and ionic conductors. Project of Special Materials for Advanced Technologies—MSTA II. Via della Ricerca Scientifica, 00133

[126] Sahner K, Moos R, Matam M, Tunney JJ, Post M. Hydrocarbon sensing with thick and thin film p-type conducting perovskite materials. Sensors and Actuators B.

[127] Giang HT, Duy HT, Ngan PQ, Thai GH, Thu DTA, Thu DT, Toan NN. Hydrocarbon gas sensing of nano-crystalline perovskite oxides LnFeO3 (Ln = La, Nd and Sm). Sen‐

sors and Actuators B. 2011;158:246– 251. DOI: 10.1016/j.snb.2011.06.013.

tuators B. 1999;55:99–110. DOI: 10.1016/S0925-4005(99)00054-4.

Applied Physics. 2009;9:613–617. DOI: 10.1016/j.cap.2008.05.014.

201. DOI: 10.1016/j.apcata.2010.05.048.

146 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

10.1016/j.snb.2010.11.053.

and Transducers. 2013;149(2):13–19.

594. DOI: 10.1016/S0925-4005(97)00177-9.

2005;108:102–112. DOI: 10.1016/j.snb.2004.12.104.

DOI: 10.1016/j.snb.2005.08.002.

Rome, Italy.

ids. 2011;72:29–33. DOI: 10.1016/j.jpcs.2010.10.074.


mediate-temperature solid oxide fuel cells. Journal of Rare Earths. 2015;33(5):519– 523. DOI: 10.1016/S1002-0721(14)60450-6.

[152] Chen Y, Wei Y, Zhong H, Gao J, Liu X, Meng G. Synthesis and electrical properties of Ln0.6Ca0.4FeO3–δ (Ln Pr, Nd, Sm) as cathode materials for IT-SOFC. Ceramics Interna‐ tional. 2007;33:1237–1241. DOI: 10.1016/j.ceramint.2006.03.035.

[140] Thirumalairajan S, Girija K, Mastelaro VR, Ganesh V, Ponpandian N. Detection of the neurotransmitter dopamine by a glassy carbon electrode modified with self assem‐ bled perovskite LaFeO3 microspheres made up of nanospheres. RSC Advances.

[141] Atta NF, Ali SM, El-Ads EH, Galal A. The electrochemistry and determination of some neurotransmitters at SrPdO3 modified graphite electrode. Journal of the Elec‐

[142] Thirumalairajan S, Girija K, Ganesh V, Mangalaraj D, Viswanathan C, Ponpandian N. Novel synthesis of LaFeO3 nanostructure dendrites: a systematic investigation of growth mechanism, properties, and biosensing for highly selective determination of neurotransmitter compounds. Crystal Growth and Design. 2013;13:291−302. DOI:

[143] Wang G, Sun J, Zhang W, Jiao S, Fang B. Simultaneous determination of dopamine, uric acid and ascorbic acid with LaFeO3 nanoparticles modified electrode. Microchim

[144] Haile SM. Fuel cell materials and components. Acta Materialia. 2003;51:5981–6000.

[145] Shao Z, Haile SM. A high-performance cathode for the next generation of solid-oxide

[146] Ding X, Kong X, Wang X, Jiang J, Cui C. Characterization and optimization of Ln1.7Sr0.3CuO4 (Ln = La, Nd)-based cathodes for intermediate temperature solid oxide fuel cells. Journal of Alloys and Compounds. 2010;502:472–476. DOI: 10.1016/j.jall‐

[147] Song K, Lee K. Characterization of Ba0.5Sr0.5M1–xFexO3–δ (M = Co and Cu) perovskite oxide cathode materials for intermediate temperature solid oxide fuel cells. Ceramics

[148] Conceiçăo LD, Silva AM, Ribeiro NFP, Souza MMVM. Combustion synthesis of La0.7Sr0.3Co0.5Fe0.5O3 (LSCF) porous materials for application as cathode in IT-SOFC. Materials Research Bulletin. 2011;46:308–314. DOI: 10.1016/j.materresbull.2010.10.009.

[149] Faro ML, Minutoli M, Monforte G, Antonucci V, Aricó AS. Glycerol oxidation in sol‐ id oxide fuel cells based on a Ni-perovskite electrocatalyst. Biomass and Bioenergy.

[150] Ling Y, Lin B, Zhao L, Zhang X, Yu J, Peng R, Meng G, Liu X. Layered perovskite LaBaCuMO5+x (M= Fe, Co) cathodes for intermediate-temperature protonic ceramic membrane fuel cells. Journal of Alloys and Compounds. 2010;493:252–255. DOI:

[151] Lianghao Y, Yonghong C, Qingwen G, Dong T, Xiaoyong L, Guangyao M, Bin L. Layered perovskite oxide Y0.8Ca0.2BaCoFeO5+δ as a novel cathode material for inter‐

International. 2012;38:5123–5131. DOI: 10.1016/j.ceramint.2012.03.015.

2011;35:1075–1084. DOI: 10.1016/j.biombioe.2010.11.018.

trochemical Society. 2013;160(7):G3144-G3151. DOI: 10.1149/2.022307jes.

2014;4:25957–25962. DOI: 10.1039/C4RA03467H.

148 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Acta. 2009;164:357–362. DOI: 10.1007/s00604-008-0066-6.

fuel cells. Nature. 2004;431:170–173. DOI: 10.1038/nature02863.

DOI: 10.1016/j.actamat.2003.08.004.

10.1021/cg3014305

com.2010.04.199.

10.1016/j.jallcom.2009.12.072.


[176] Werner J, Dubuis G, Walter A, Löper P, Moon S, Nicolay S, Morales-Masis M, Wolf SD, Niesen B, Ballif C. Solar Energy Materials and Solar Cells. 2015;141:407–413. DOI: 10.1016/j.solmat.2015.06.024.

[164] Jin C, Cao X, Lu F, Yang Z, Yang R. Electrochemical study of Ba0.5Sr0.5Co0.8Fe0.2O3 per‐ ovskite as bifunctional catalyst in alkaline media. International Journal of Hydrogen

[165] Malkhandi S, Manohar AK, Yang B, Prakash GKS, Narayanan SR. Properties of calci‐ um-doped lanthanum cobalt oxide perovskite electrocatalysts for oxygen evolution in alkaline medium. The Electrochemical Society. Abstract 292, 220th ECS Meeting.

[166] Kahoul A, Hammouche A, Nâamoune F, Chartier P, Poillerat G, Koenig JF. Solvent effect on synthesis of perovskite-type La1–xCaxCoO3 and their electrochemical proper‐ ties for oxygen reactions. Materials Research Bulletin. 2000;35:1955–1966. DOI:

[167] Bao X, Wang Y, Zhu Q, Wang N, Zhu D, Wang J, Yang A, Yang R. Efficient planar perovskite solar cells with large fill factor and excellent stability. Journal of Power

[168] Bhatt P, Pandey K, Yadav P, Tripathi B, KanthP C, Pandey MK, Kumar M. Investigat‐ ing the charge carrier transport within the hole-transport material free perovskite so‐ lar cell processed in ambient air. Solar Energy Materials and Solar Cells.

[169] Chen Q, Marco ND, Yang YM, Song T, Chen C, Zhao H, Hong Z, Zhou H, Yang Y. Under the spotlight: the organic-inorganic hybrid halide perovskite for optoelectron‐

[170] Green MA, Ho-Baillie A, Snaith HJ. The emergence of perovskite solar cells. Nature

[171] Park N. Perovskite solar cells: an emerging photovoltaic technology. Materials To‐

[172] Schoonman J. Organic-inorganic lead halide perovskite solar cell materials: a possi‐ ble stability problem. Chemical Physics Letters. DOI: 10.1016/j.cplett.2014.11.063.

[173] Sun H, Ruan P, Bao Z, Chen L, Zhou X. Shell-in-Shell TiO2 hollow microspheres and optimized application in light-trapping perovskite solar cells. Solid State Sciences.

[174] Tripathi B, Bhatt P, KanthP C, Yadav P, Desai B, Pandey MK, Kumar M. Tempera‐ ture induced structural, electrical and optical changes in solution processed perov‐ skite material: Application in photovoltaics. Solar Energy Materials and Solar Cells.

[175] Wang X, Li Z, Xu W, Kulkarni SA, Batabyal SK, Zhang S, Cao A, Wong LH. TiO2 nanotube arrays based flexible perovskite solar cells with transparent carbon nano‐

Energy. 2013;38:10389–10393. DOI: 10.1016/j.ijhydene.2013.06.047.

Sources. 2015;297:53–58. DOI: 10.1016/j.jpowsour.2015.07.081.

ic applications. Nano Today. DOI: 10.1016/j.nantod.2015.04.009.

Photonics. 2014;8:506–514. DOI: 10.1038/NPHOTON.2014.134.

2015;40:60–66. DOI: 10.1016/j.solidstatesciences.2014.12.015.

tube electrode. Nano Energy. DOI: 10.1016/j.nanoen.2014.11.042.

2015;132:615–622. DOI: 10.1016/j.solmat.2014.10.017.

2015;140:320–327. DOI: 10.1016/j.solmat.2015.04.028.

day. DOI: 10.1016/j.mattod.2014.07.007.

2011.

10.1016/S0025-5408(00)00395-0.

150 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications


## **Perovskite Oxide Nanocrystals — Synthesis, Characterization, Functionalization, and Novel Applications**

Heng Wu and Xinhua Zhu

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61640

## **Abstract**

Perovskite oxide nanocrystals exhibit a wide spectrum of attractive properties such as ferroelectricity, piezoelectricity, dielectricity, ferromagnetism, magne‐ toresistance, and multiferroics. These properties are indispensable for applica‐ tions in ferroelectric random access memories, multilayer ceramic capacitors, transducers, sensors and actuators, magnetic random access memories, and the potential new types of multiple-state memories and spintronic devices control‐ led by electric and magnetic fields. In the past two decades, much effort has been made to synthesize and characterize the perovskite oxide nanocrystals. Various physical and chemical deposition techniques and growth mechanisms are explored and developed to control the morphology, identical shape, uni‐ form size, perfect crystalline structure, defects, and homogenous stoichiometry of the perovskite oxide nanocrystals. This chapter provides a comprehensive review of the state-of-the-art research activities that focus on the rational syn‐ thesis, structural characterization, functionalization, and unique applications of perovskite oxide nanocrystals in nanoelectronics. It begins with the rational synthesis of perovskite oxide nanocrystals, and then summarizes their struc‐ tural characterizations. Fundamental physical properties of perovskite oxide nanocrystals are also highlighted, and a range of novel applications in nanoe‐ lectronics, information storages, and spintronics are discussed. Finally, we con‐ clude this review with some perspectives/outlook and future researches in these fields.

**Keywords:** Perovskite oxide nanocrystals, synthesis, structural characterization, function‐ alization, applications

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

## **1. Introduction**

Perovskite oxide nanocrystals have important properties in ferroelectricity, piezoelectricity, dielectricity, ferromagnetism, magnetoresistance, and multiferroics. The perovskite structure is named for the prototype CaTiO3 mineral called perovskite, which is generally a metal oxide with the formula ABO3, where B is a small transition metal cation and A is a larger s-, d-, or fblock cation. In a cubic perovskite, the larger cation A resides on the corners of the unit cell, the smaller cation B is in the center of the unit cell, and the oxygen ions (O2 – ) are on the centers of the faces [1]. The perovskite structure can also be built from three-dimensional cornersharing BO6 octahedra that are connected through B–O–B linkages. The A-site cation fits in the large cavity at the center of eight corner-sharing BO6 octahedra, and the B-site cation resides in the interstitial site of an octahedron of oxygen anions [1]. Interestingly, and of technological importance, a variety of crystallize in the perovskite structure. Typical perovskite materials of technological importance are ferroelectric BaTiO3, PbTiO3, dielectric (Ba,Sr)TiO3, piezoelectric Pb(Zr,Ti)O3, electrostrictive Pb(Mg,Nb)O3, magnetoresistant (La,Ca)MnO3, and multiferroic BiFeO3. They have attracted interest for several decades, with tremendous applications including ferroelectric random access memories, multilayer ceramic capacitors, transducers, sensors and actuators, magnetic random access memories, and the potential new types of multiple-state memories and spintronic devices controlled by electric and magnetic fields [1– 8]. The major challenge in manufacturing these materials is the processing of the materials with reliable and reproducible properties [9, 10]. Following a similar trend to the miniaturi‐ zation as the conventional CMOS (complementary metal oxide semiconductor) devices, the down-sized electronic devices based on perovskite electronic ceramic materials have also been developed. Advances toward nanoscale electronics have additionally increased interest in this field of perovskite oxide nanocrystals [11–13]. For example, to develop high volume efficient multilayered ceramic capacitors (MLCCs), the sizes of BaTiO3 nanoparticles with high purity and uniform shape used for fabricating the next generation of MLCCs will be lowered down to tens of nanometers. Therefore, synthesis of high-purity, ultra-fine, and agglomerate-free perovskite nanoparticles with controlled particle size, morphology, and stoichiometry is the critical step in processing of perovskite ceramics with desirable properties. Perovskite oxide nanocrystals with the general formula of ABO3 play a very important role in today's techno‐ logical advances, which can be found in the application fields of passive dielectric materials, piezoelectric actuators and transducers, sensors, and micro/nano electromechanical systems and many more.

The evolution of a method to produce perovskite oxide nanocrystals with precise stoichiom‐ etry and desired properties is much complex. Conventionally, perovskite nanoparicles are prepared by solid-state reactions between the corresponding oxides or oxides and carbonates at temperatures above 1000°C [14, 15]. However, the resulting microstructures of perovskite nanoparticles obtained from this method are not suitable for the miniaturization of electronic devices, due to their significant particle agglomeration, poor chemical homogeneity, and coarse large particle sizes. To resolve the problems arising from the conventional ceramic techniques and to produce homogeneous and stoichiometric perovskite nanocrystals, in recent years, wet chemical routes have been developed [12, 16–18]. They can be better controlled from the molecular precursor to the final material to give highly pure and homogeneous products, allowing for the low reaction temperatures used. The size and morphology of the nanocrystals can be controlled, and metastable phases could be prepared [18].

The objective of this chapter is to provide an overview of the state-of-the-art in perovskite nanocrystals, which covers their synthesis, characterization, functionalization, and novel applications. First, we review the synthetic methods for perovskite nanocrystals, which include the syntheses using solid and liquid precursors. The second section deals with the electron microstructural characterization of perovskite nanocrystals. In the context of functionalization, we discuss the unique properties of perovskite oxide nanocrystals (e.g., ferroelectric and dielectric, electrical, magnetic, and multiferroic properties), and the size effects for these unique properties are also discussed. And then a broad range of novel applications of perov‐ skite oxide nanocrystals is addressed. Finally, we provide a perspective on the future outlook of perovskite nanocrystals.

## **2. Synthesis of perovskite oxide nanocrystals**

Due to the particle size, dimensionality, and composition governing the resultant properties of the nanostructured perovskite materials that are assembled from nanocrystals as building blocks to achieve certain desired properties, the synthesis of high-purity, ultra-fine, and agglomerate-free perovskite oxide nanocrystals with controlled particle size, morphology, and stoichiometry is the first and perhaps the most crucial step in processing of perovskite ceramics with desirable properties. The major issues for the synthesis of perovskite oxide nanocrystals include: (a) the control of particle size and composition and (b) the control of the interfaces and distributions of the nanobuilding blocks within the fully formed nanostructured perov‐ skite compounds. Over the past several decades, various methods have been developed to prepared perovskite oxide nanocrystals and the related nanostructured perovskite com‐ pounds. In the subsequent sections, some important methods for the preparation of perovskite oxide nanocrystals are described.

## **2.1. Solid-state reaction method**

**1. Introduction**

and many more.

Perovskite oxide nanocrystals have important properties in ferroelectricity, piezoelectricity, dielectricity, ferromagnetism, magnetoresistance, and multiferroics. The perovskite structure is named for the prototype CaTiO3 mineral called perovskite, which is generally a metal oxide with the formula ABO3, where B is a small transition metal cation and A is a larger s-, d-, or fblock cation. In a cubic perovskite, the larger cation A resides on the corners of the unit cell,

of the faces [1]. The perovskite structure can also be built from three-dimensional cornersharing BO6 octahedra that are connected through B–O–B linkages. The A-site cation fits in the large cavity at the center of eight corner-sharing BO6 octahedra, and the B-site cation resides in the interstitial site of an octahedron of oxygen anions [1]. Interestingly, and of technological importance, a variety of crystallize in the perovskite structure. Typical perovskite materials of technological importance are ferroelectric BaTiO3, PbTiO3, dielectric (Ba,Sr)TiO3, piezoelectric Pb(Zr,Ti)O3, electrostrictive Pb(Mg,Nb)O3, magnetoresistant (La,Ca)MnO3, and multiferroic BiFeO3. They have attracted interest for several decades, with tremendous applications including ferroelectric random access memories, multilayer ceramic capacitors, transducers, sensors and actuators, magnetic random access memories, and the potential new types of multiple-state memories and spintronic devices controlled by electric and magnetic fields [1– 8]. The major challenge in manufacturing these materials is the processing of the materials with reliable and reproducible properties [9, 10]. Following a similar trend to the miniaturi‐ zation as the conventional CMOS (complementary metal oxide semiconductor) devices, the down-sized electronic devices based on perovskite electronic ceramic materials have also been developed. Advances toward nanoscale electronics have additionally increased interest in this field of perovskite oxide nanocrystals [11–13]. For example, to develop high volume efficient multilayered ceramic capacitors (MLCCs), the sizes of BaTiO3 nanoparticles with high purity and uniform shape used for fabricating the next generation of MLCCs will be lowered down to tens of nanometers. Therefore, synthesis of high-purity, ultra-fine, and agglomerate-free perovskite nanoparticles with controlled particle size, morphology, and stoichiometry is the critical step in processing of perovskite ceramics with desirable properties. Perovskite oxide nanocrystals with the general formula of ABO3 play a very important role in today's techno‐ logical advances, which can be found in the application fields of passive dielectric materials, piezoelectric actuators and transducers, sensors, and micro/nano electromechanical systems

The evolution of a method to produce perovskite oxide nanocrystals with precise stoichiom‐ etry and desired properties is much complex. Conventionally, perovskite nanoparicles are prepared by solid-state reactions between the corresponding oxides or oxides and carbonates at temperatures above 1000°C [14, 15]. However, the resulting microstructures of perovskite nanoparticles obtained from this method are not suitable for the miniaturization of electronic devices, due to their significant particle agglomeration, poor chemical homogeneity, and coarse large particle sizes. To resolve the problems arising from the conventional ceramic techniques and to produce homogeneous and stoichiometric perovskite nanocrystals, in recent years, wet chemical routes have been developed [12, 16–18]. They can be better controlled from

–

) are on the centers

the smaller cation B is in the center of the unit cell, and the oxygen ions (O2

154 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The solid-state reaction method is the most traditional one for preparing perovskite oxide nanocrystals (e.g., BaTiO3, PbTiO3, Pb(Zr,Ti)O3, etc.) [14, 15]. This process includes weighting starting materials (the corresponding oxides or oxides and carbonates), mixing, milling, and calcining them at elevated temperatures to form the perovskite phase. For example, in synthesis of BaTiO3 nanocrystals by solid-state reaction method, the reaction process in air has been proposed to take place in at least three stages and relies on the diffusion of Ba2+ ions into TiO2 [19]. Firstly, BaCO3 reacts with the outer surface region of TiO2 to form a surface layer of BaTiO3 on individual TiO2 grains. Further diffusion of Ba2+ ions into TiO2 necessitates the formation of Ba2TiO4 between the unreacted BaCO3 and the previously formed BaTiO3. After prolonged sintering periods, the intermediate Ba-rich phase Ba2TiO4 reacts with the remaining TiO2 in the core-regions of the TiO2 grains to form BaTiO3. The high-temperature calcination produces an agglomerated powder with a coarse particle size which requires additional milling process. However, contamination and other undesirable features during the milling process can create defects in the manufactured products. Furthermore, the more components in the ceramic powders, the more difficult it may be to achieve the desired homogeneity, stoichiometry, and phases. By using nanocrystalline BaCO3 and TiO2 as starting materials, Buscaglia et al. [20] have recently synthesized the perovskite BaTiO3 nanocrystals with size of ~100 nm and narrow particle size distribution, *via* a solid-state reaction at calcination temper‐ atures as low as 800°C. The average particle size of powders obtained *via* this method is essentially determined by the particle size of the used TiO2 because the reaction rate is controlled by the diffusion rate of barium ions into the TiO2 lattice [21]. Similar reaction mechanism was also found in the synthesized process of BaZrO3 powders [22]. The morphol‐ ogy of BaZrO3 nanoparticles was dependent upon the initial size and shape of the used starting ZrO2 particles. Therefore, fine BaZrO3 powders with particle size of 70–100 nm composing of crystallites of ~ 20–30 nm can be synthesized by using very fine (70–90 nm) starting ZrO2 particles and coarse (~ 1 μm) BaCO3 particles commercially available and calcination at ~1000°C. Higher calcination temperatures accelerate the initial stage of reaction but often lead to coarser and more-agglomerated powders.

## **2.2. Molten-Salt Method (MSS)**

Molten-salt synthesis (MSS) is one of the methods for preparing perovskite oxide nanopow‐ ders, which involves the use of a molten salt as the medium for preparing perovskite oxides from their constituent materials (oxides and carbonates) [23]. This method allows melt-solid reacting much faster due to the small diffusion distances and higher mobility of oxides in the melt. Many molten salt solvents such as alkali chlorides, sulphates, carbonates, and hydroxides are used in the past acting as a medium of reaction for the constituent oxides. MSS attracted much attention due to its advantages such as one-step, rapid, and environmentally friendly for synthesizing perovskite oxide functional materials. The melt increases the reaction rate due to the small diffusion distances and high mobility of oxides in it. In addition, as a medium of the reaction, the choice of the molten salt system is variety, e.g., alkali chlorides, sulphates, carbonates, and hydroxides. The features of this synthetic method are the simplest, versatile, and cost-effective approaches available for obtaining a pure perovskite phase at a relatively low temperature for a shorter soaking time. Generally, the procedure of synthesize perovskite oxide powders by MSS is shown below. First, heating the raw materials consisting of reactants and salt system above the melting temperature of the salt, so that the produced particles are synthesized at the melted salt solvent. The characteristics of the powders can be controlled by tuning the heating temperature and duration time. After cooling down to room temperature naturally, the product is washed by appropriate solvent (typically, deionized water) and dried subsequently, and the complex perovskite oxide powder is obtained. In the past decade, MSS has been widely used to synthesize a range of perovskite oxide nanocrystals. For example, perovskite SrFeO3 nanocrystals were successfully synthesized in molten NaNO3–KNO3 eutectic with Na2O2 from a mixture of strontium nitrate and ferric nitrate [24]. The effects of metal precursors, salt medium, annealing temperature, and oxidizing properties of the melt on the phase compositions, crystallite sizes, and morphology of the resulting metal oxides were systematically investigated. It was found that the formation of the SrFeO3 phase was mainly dependent upon the nature of the metal precursor and salt medium. Metal nitrates were the suitable precursors and NaNO3–KNO3 eutectic with Na2O2 was the suitable salt medium, which resulted in the formation of pure SrFeO3 nanocrystals at a much lower temperature of 400°C. Pure cubic phase SrTiO3 nanocrystals were synthesized in the eutectic NaCl–KCl molten salts at 700°C by heating TiO2 and Sr(NO3)2 powders for 6 h [25]. The sizes of the as-synthesized SrTiO3 nanocrystals were dependent upon the kinds of TiO2 precursors, indicating that the formation process of SrTiO3 in the molten salts was mainly controlled by the template formation mechanism. Single-crystalline perovskite BaZrO3 nanocrystals were also obtained using BaC2O4 and ZrO2 as precursors and NaOH/KOH as the molten salts at 700°C [26]. Rare earth orthoferrites with the general formula of LnFeO3 (Ln = La, Pr, Nd) were synthesized in molten NaOH flux at 400°C [27], and LaMO3 (M = Mn, Fe, Co, Ni) were obtained in molten nitrates or nitrites [28]. Recently, perovskite Pb-based relaxors [29], Ba-based dielectric oxides [30], and perovskite multiferroic bismuth ferrites [31–35] were also synthesized by MSS process.

## **2.3. Wet chemical methods**

produces an agglomerated powder with a coarse particle size which requires additional milling process. However, contamination and other undesirable features during the milling process can create defects in the manufactured products. Furthermore, the more components in the ceramic powders, the more difficult it may be to achieve the desired homogeneity, stoichiometry, and phases. By using nanocrystalline BaCO3 and TiO2 as starting materials, Buscaglia et al. [20] have recently synthesized the perovskite BaTiO3 nanocrystals with size of ~100 nm and narrow particle size distribution, *via* a solid-state reaction at calcination temper‐ atures as low as 800°C. The average particle size of powders obtained *via* this method is essentially determined by the particle size of the used TiO2 because the reaction rate is controlled by the diffusion rate of barium ions into the TiO2 lattice [21]. Similar reaction mechanism was also found in the synthesized process of BaZrO3 powders [22]. The morphol‐ ogy of BaZrO3 nanoparticles was dependent upon the initial size and shape of the used starting ZrO2 particles. Therefore, fine BaZrO3 powders with particle size of 70–100 nm composing of crystallites of ~ 20–30 nm can be synthesized by using very fine (70–90 nm) starting ZrO2 particles and coarse (~ 1 μm) BaCO3 particles commercially available and calcination at ~1000°C. Higher calcination temperatures accelerate the initial stage of reaction but often lead

Molten-salt synthesis (MSS) is one of the methods for preparing perovskite oxide nanopow‐ ders, which involves the use of a molten salt as the medium for preparing perovskite oxides from their constituent materials (oxides and carbonates) [23]. This method allows melt-solid reacting much faster due to the small diffusion distances and higher mobility of oxides in the melt. Many molten salt solvents such as alkali chlorides, sulphates, carbonates, and hydroxides are used in the past acting as a medium of reaction for the constituent oxides. MSS attracted much attention due to its advantages such as one-step, rapid, and environmentally friendly for synthesizing perovskite oxide functional materials. The melt increases the reaction rate due to the small diffusion distances and high mobility of oxides in it. In addition, as a medium of the reaction, the choice of the molten salt system is variety, e.g., alkali chlorides, sulphates, carbonates, and hydroxides. The features of this synthetic method are the simplest, versatile, and cost-effective approaches available for obtaining a pure perovskite phase at a relatively low temperature for a shorter soaking time. Generally, the procedure of synthesize perovskite oxide powders by MSS is shown below. First, heating the raw materials consisting of reactants and salt system above the melting temperature of the salt, so that the produced particles are synthesized at the melted salt solvent. The characteristics of the powders can be controlled by tuning the heating temperature and duration time. After cooling down to room temperature naturally, the product is washed by appropriate solvent (typically, deionized water) and dried subsequently, and the complex perovskite oxide powder is obtained. In the past decade, MSS has been widely used to synthesize a range of perovskite oxide nanocrystals. For example, perovskite SrFeO3 nanocrystals were successfully synthesized in molten NaNO3–KNO3 eutectic with Na2O2 from a mixture of strontium nitrate and ferric nitrate [24]. The effects of metal precursors, salt medium, annealing temperature, and oxidizing properties of the melt on the phase compositions, crystallite sizes, and morphology of the resulting metal oxides were

to coarser and more-agglomerated powders.

156 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**2.2. Molten-Salt Method (MSS)**

Since perovskite oxide nanocrystals synthesized *via* conventional solid-state reactions have uncontrolled and irregular morphologies, the electrical properties of the sintered ceramics are very poor. To solve this problem, various wet chemical methods have been developed to synthesize perovskite nanocrystals. The popular wet chemical methods for the preparation of perovskite nanocrystals, include sol–gel method [36–38], alkoxide–hydroxide sol-precipitation method [39–41], hydrothermal method [42–44], microwave-hydrothermal [45–47], and solvothermal syntheses [48–50]. The most important advantages of the wet chemical methods include easy controlling of the chemical stoichiometry, producing nanocrystals with narrow size distribution, and low crystallization temperature due to the constituents mixed at the quasi-atomic level in a solution system. Due to the wet chemical solution process, a dopant such as paramagnetic ions or rare-earth ions could be readily introduced during the prepara‐ tion of the precursor solution. In the following subsequent sections, various wet chemical methods used for preparation of perovskite oxide nanocrystals are introduced.

## *2.3.1. Sol–gel (colloidal) processing*

Sol–gel process is a popular processing route for the synthesis of perovskite oxide nanocrys‐ tals (e.g., BaTiO3, PbTiO3, BiFeO3) [36–38]. This process involves the formation of a sol by dissolving the metal aloxide, metal-organic, or metal-inorganic salt precursors in a suitable solvent, subsequent drying of the gel followed by calcination and sintering at high temperature to form perovskite nanocrystals. Due to the reacting species homogenized at the atomic level in a sol–gel process, the diffusion distances are considerably reduced compared to a conventional solid-state reaction; therefore, the product can be formed at much lower temperatures. In this process, the selection of starting materials, concentra‐ tion, pH value, and heat treatment schedule play an important role in affecting the properties of perovskite nanocrystals. This has been demonstrated in the case of BaTiO3 perovskite oxide nanocrystals [38, 51, 52]. Barium acetate and titanium isopropoxide are often used as starting materials to synthesize BaTiO3 nanocrystals. However, the different rates in the hydrolysis and condensation of Ba and Ti precursors often give rise to the chemical component segregation in the obtained gels. To solve this problem, acetic acid or acetylaceton was often used to control the hydrolysis rate of the Ti precursor, since these complexing agents acts as chelating agents to coordinate with Ti species [53,54].

For the obtained gels, a heat treatment at high temperature over 600<sup>ο</sup> C is required to remove the unreacted organics and to crystallize the powders. Several steps involve in the transfor‐ mation from the precursor to the crystalline BaTiO3 nanocrystals, including the transformation from the precursor to the amorphous BaTiO3, and then to the three-dimensional nucleation of the crystalline BaTiO3 in the amorphous matrix, and finally to the nanocrystal growth of BaTiO3 *via* a solid-state reaction [55]. To better control the grain size and its distribution, the heat treatment process parameters of the gels (e.g., post-annealing temperature, time and atmosphere, heating rate) must be optimized [56]. Normally, higher annealing temperature or longer annealing time can lead to larger grain size of the powders, while slow heating rate and inert annealing atmosphere can inhabit the aggregated behavior of nanocrystals in comparison to air or oxygen atmosphere. That was demonstrated in the synthesis of Pb(Zr,Ti)O3 nano‐ crystals [56]. By using these techniques, monodispersed perovskite oxide nanocrystals and related nanostructured materials have been successfully fabricated. The particle size can be adjusted from a few nanometers to micrometers *via* controlling the solid-state polymerization and the heat treatment process [36, 55].

### *2.3.2. Hydrothermal process*

Hydrothermal synthesis involves heating an aqueous suspension of insoluble salts in an autoclave at a moderate temperature and pressure where the crystallization of a desired phase is taking place. As a powerful method for synthesis of very fine and homogeneous perovskite powders with a controllable size distribution and morphology, its application to the growth of BaTiO3 powders with the desired size and particle morphology has been widely investigated [57–60]. Based on the high-resolution transmission electron microscopy (HRTEM) observa‐ tions on the incompletely and fully reacted powders, Pinceloup et al. [57] proposed a disso‐ lution–precipitation model for hydrothermal synthesis of BaTiO3 nanocrystals using Ba(OH)2 and TiO2 as precursors. In this model, TiO2 particles are first dissolved to form hydroxyl titanium complexes [Ti(OH)*<sup>n</sup>*– ] and then react with barium ions in the solution to precipitate BaTiO3. On the other hand, Hertl [58] and Hu et al. [61] proposed another in situ heterogeneous transformation model, in which TiO2 particles react initially with the dissolved barium to produce a continuous layer of BaTiO3, and the additional barium must diffuse through this layer and reacts with TiO2 until the supply of TiO2 is exhausted. This model was supported experimentally by the hydrothermal conversion from TiO2 microspheres to nanocrystalline BaTiO3 [59]. Eckert et al. [60] also reported on a mechanism evolution from a dissolution–precipitation process at the early stage of the reaction to an in situ mechanism for the longer reaction times. Recently, Walton et al. [62] investigated the hydrothermal crystalli‐ zation of BaTiO3 by time-resolved powder neutron diffraction methods in situ, using the newly developed Oxford/ISIS hydrothermal cell. They directly observed that the rapid dissolution of the barium source was followed by dissolution of the titanium source before the onset of crystallization of BaTiO3. These qualitative observations strongly suggest that a homogeneous dissolution–precipitation mechanism dominates in the hydrothermal crystallization of BaTiO3 rather than other possible mechanisms proposed in the literatures [81–83]. These contradictive experimental observations reported previously probably result from the different hydrothermal conditions.

## *2.3.3. Microwave-hydrothermal process*

perovskite oxide nanocrystals [38, 51, 52]. Barium acetate and titanium isopropoxide are often used as starting materials to synthesize BaTiO3 nanocrystals. However, the different rates in the hydrolysis and condensation of Ba and Ti precursors often give rise to the chemical component segregation in the obtained gels. To solve this problem, acetic acid or acetylaceton was often used to control the hydrolysis rate of the Ti precursor, since these

the unreacted organics and to crystallize the powders. Several steps involve in the transfor‐ mation from the precursor to the crystalline BaTiO3 nanocrystals, including the transformation from the precursor to the amorphous BaTiO3, and then to the three-dimensional nucleation of the crystalline BaTiO3 in the amorphous matrix, and finally to the nanocrystal growth of BaTiO3 *via* a solid-state reaction [55]. To better control the grain size and its distribution, the heat treatment process parameters of the gels (e.g., post-annealing temperature, time and atmosphere, heating rate) must be optimized [56]. Normally, higher annealing temperature or longer annealing time can lead to larger grain size of the powders, while slow heating rate and inert annealing atmosphere can inhabit the aggregated behavior of nanocrystals in comparison to air or oxygen atmosphere. That was demonstrated in the synthesis of Pb(Zr,Ti)O3 nano‐ crystals [56]. By using these techniques, monodispersed perovskite oxide nanocrystals and related nanostructured materials have been successfully fabricated. The particle size can be adjusted from a few nanometers to micrometers *via* controlling the solid-state polymerization

Hydrothermal synthesis involves heating an aqueous suspension of insoluble salts in an autoclave at a moderate temperature and pressure where the crystallization of a desired phase is taking place. As a powerful method for synthesis of very fine and homogeneous perovskite powders with a controllable size distribution and morphology, its application to the growth of BaTiO3 powders with the desired size and particle morphology has been widely investigated [57–60]. Based on the high-resolution transmission electron microscopy (HRTEM) observa‐ tions on the incompletely and fully reacted powders, Pinceloup et al. [57] proposed a disso‐ lution–precipitation model for hydrothermal synthesis of BaTiO3 nanocrystals using Ba(OH)2 and TiO2 as precursors. In this model, TiO2 particles are first dissolved to form

precipitate BaTiO3. On the other hand, Hertl [58] and Hu et al. [61] proposed another in situ heterogeneous transformation model, in which TiO2 particles react initially with the dissolved barium to produce a continuous layer of BaTiO3, and the additional barium must diffuse through this layer and reacts with TiO2 until the supply of TiO2 is exhausted. This model was supported experimentally by the hydrothermal conversion from TiO2 microspheres to nanocrystalline BaTiO3 [59]. Eckert et al. [60] also reported on a mechanism evolution from a dissolution–precipitation process at the early stage of the reaction to an in situ mechanism for the longer reaction times. Recently, Walton et al. [62] investigated the hydrothermal crystalli‐ zation of BaTiO3 by time-resolved powder neutron diffraction methods in situ, using the newly

] and then react with barium ions in the solution to

C is required to remove

complexing agents acts as chelating agents to coordinate with Ti species [53,54].

For the obtained gels, a heat treatment at high temperature over 600<sup>ο</sup>

158 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

and the heat treatment process [36, 55].

hydroxyl titanium complexes [Ti(OH)*<sup>n</sup>*–

*2.3.2. Hydrothermal process*

The microwave-hydrothermal process is a rapid process, which has the potential to enhance the crystallization kinetics of hydrothermal process. The term microwave-hydrothermal process was named by Komarneni et al. [63] in 1992, and this process has been used for the rapid synthesis of numerous ceramic oxides, hydroxylated phases, porous materials, and hematite powders [64–67]. It offers many distinct advantages over the conventional hydro‐ thermal synthesis, such as cost savings, rapid internal heating, and synthesis of new materials.

Up to date, numerous reports have been published on the synthesis of BaTiO3 nanocrystals by microwave-hydrothermal process below 200°C, and these processes were found to be very rapid but they all yielded cubic phase [45–47, 68]. For example, Khollam et al. [47] obtained submicron-sized BaTiO3 powders (0.1–0.2 μm) at holding time of 30 min. One of the first approaches on the synthesis of the nanosized BaTiO3 powders (about 30 nm) at 30 min was reported by Jhung et al. [69]. Recently, tetragonal BaTiO3 powders are synthesized by micro‐ wave-hydrothermal method at a typical temperature of 240°C from hydrous titanium oxide and barium hydroxide, in the absence of chloride ions and alkali metal ions to avoid contam‐ inations. The effects of synthesis conditions, including reaction temperature and time, and reactant composition, on the formation of tetragonal structure and particle size of BaTiO3 powders, have been systematically investigated [70]. The results have shown that the amount of the tetragonal phase and the particle size increased quickly with reaction time, whereas the content of lattice hydroxyl groups decreased. Tetragonal BaTiO3 powder with nearly full tetragonality (c/a ratio = 1.010) was obtained *via* the microwave-hydrothermal process performed at 240°C for 20 h [70]. As the reaction temperature was lowered down to 220°C, the formation of tetragonal structure and the growth of particles slowed down substantially, showing a critical effect of the reaction temperature on the microwave-hydrothermal process‐ ing of tetragonal BaTiO3. Higher Ba(OH)2/Ti mole ratio enhanced the formation of tetragonal BaTiO3 and so did higher initial concentration of Ti with fixed Ba(OH)2/Ti ratio. Besides the BaTiO3 nanocrystals, Ba1–*x*Sr*x*TiO3 (*x* = 0.1–0.4) nanocrystals with the average size about 20 nm were also prepared at a relatively short period of time (10 min) *via* microwave-hydrother‐ mal synthesis [71]. The structure and the average size of BST were determined to be in the range of 20–50 nm depending on the synthesis time (10–90 min).

Perovskite BiFeO3 nanocrystals exhibit multiferroic properties at room temperature (ferro‐ electric and magnetic order coexisting in the same phase) and is, therefore, a good candidate for potential multiferroic application in information technology. Recently, perovskite BiFeO3 nanocrystals with diameters of 10–50 nm are synthesized by microwave-hydrothermal synthesis [72,73]. The starting reactants are used as Fe(NO3)3∙9H2O and Bi(NO3)3∙5H2O together with KOH as the mineralizer. Figure 2 shows the XRD pattern of the perovskite BiFeO3 nanocrystals synthesized by microwave-hydrothermal process. As shown in Figure 2, all the reflection peaks can be readily indexed as a rhombohedrally distorted perovskite BFO (JCPDS card No. 86-1518) with space group R3c and lattice parameters of *a* = 5.582 Å and *c* = 13.876 Å. No peaks from other phase were detected. In addition, the sharp diffraction peaks indicate that highly crystallized and phase-pure BFO nanocrystals can be obtained under the present synthesis conditions. The particle size and morphology of the as-obtained perovskite BiFeO3 nanocrystals were revealed by TEM images. Figure 3(a) represents a typical lowmagnitude TEM image, in which the BFO nanocrystals exhibit a spherical morphology with particle sizes of 15–55 nm. No agglomerated particles were observed, and nearly monodis‐ persive behavior was observed in these BFO nanocrystals. Their statistic particle size distri‐ bution is shown in Figure 3(b), which clearly demonstrates that the average particle size of the BFO nanoparticle is ~35 nm, this value is much smaller than the previously reported ones [61, 74–77]. A high-magnification TEM image is shown in Figure 3(c), which clearly demonstrates the well-dispersed BFO nanocrystals with a spherical morphology. The selected area electron diffraction (SAED) pattern of the as-obtained BFO nanocrystals is shown in Figure 3(d), which exhibits polycrystalline diffraction rings consisting of discrete diffraction spots. Based on the analysis of the diffraction rings, the first six diffraction rings can be indexed as (012), (104)/ (110), (006)/(202), (024), (116)/(122), and (214), which is in well agreement with the XRD results. The high crystallinity nature of the BFO nanocrystals is also proven by the lattice fringes observed in the HRTEM image. Figure 3(e) shows an HRTEM image taken from a single BFO nanocrystal with a particle size of ~12 nm, and the lattice fringes of (202) and (113) crystal planes are clearly resolved. The Fourier filtered HRTEM image of the single BFO nanocrystal is shown in Figure 3(f), and the inset is a fast Fourier transform (FFT) pattern of the nanocrystal obtained by Gatan Digital Micrography software. The angle between the (202) and (113) crystal planes measured from Figure 3(f) is 58ο , which is very close to the theoretical value of 57.99° (the angle ∠(202):(113) = 57.99° for BFO). Therefore, the single-crystalline nature of BFO nanocrystals is confirmed by the SAED pattern and HRTEM image.

**Figure 1.** XRD pattern of the as-synthesized perovskite BiFeO3 nanocrystals by microwave-hydrothermal reaction. Re‐ produced with permission from [73].

synthesis [72,73]. The starting reactants are used as Fe(NO3)3∙9H2O and Bi(NO3)3∙5H2O together with KOH as the mineralizer. Figure 2 shows the XRD pattern of the perovskite BiFeO3 nanocrystals synthesized by microwave-hydrothermal process. As shown in Figure 2, all the reflection peaks can be readily indexed as a rhombohedrally distorted perovskite BFO (JCPDS card No. 86-1518) with space group R3c and lattice parameters of *a* = 5.582 Å and *c* = 13.876 Å. No peaks from other phase were detected. In addition, the sharp diffraction peaks indicate that highly crystallized and phase-pure BFO nanocrystals can be obtained under the present synthesis conditions. The particle size and morphology of the as-obtained perovskite BiFeO3 nanocrystals were revealed by TEM images. Figure 3(a) represents a typical lowmagnitude TEM image, in which the BFO nanocrystals exhibit a spherical morphology with particle sizes of 15–55 nm. No agglomerated particles were observed, and nearly monodis‐ persive behavior was observed in these BFO nanocrystals. Their statistic particle size distri‐ bution is shown in Figure 3(b), which clearly demonstrates that the average particle size of the BFO nanoparticle is ~35 nm, this value is much smaller than the previously reported ones [61, 74–77]. A high-magnification TEM image is shown in Figure 3(c), which clearly demonstrates the well-dispersed BFO nanocrystals with a spherical morphology. The selected area electron diffraction (SAED) pattern of the as-obtained BFO nanocrystals is shown in Figure 3(d), which exhibits polycrystalline diffraction rings consisting of discrete diffraction spots. Based on the analysis of the diffraction rings, the first six diffraction rings can be indexed as (012), (104)/ (110), (006)/(202), (024), (116)/(122), and (214), which is in well agreement with the XRD results. The high crystallinity nature of the BFO nanocrystals is also proven by the lattice fringes observed in the HRTEM image. Figure 3(e) shows an HRTEM image taken from a single BFO nanocrystal with a particle size of ~12 nm, and the lattice fringes of (202) and (113) crystal planes are clearly resolved. The Fourier filtered HRTEM image of the single BFO nanocrystal is shown in Figure 3(f), and the inset is a fast Fourier transform (FFT) pattern of the nanocrystal obtained by Gatan Digital Micrography software. The angle between the (202) and (113) crystal

160 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

(the angle ∠(202):(113) = 57.99° for BFO). Therefore, the single-crystalline nature of BFO

**Figure 1.** XRD pattern of the as-synthesized perovskite BiFeO3 nanocrystals by microwave-hydrothermal reaction. Re‐

nanocrystals is confirmed by the SAED pattern and HRTEM image.

, which is very close to the theoretical value of 57.99°

planes measured from Figure 3(f) is 58ο

produced with permission from [73].

**Figure 2.** (a) Typical low-magnitude TEM image of the perovskite BiFeO3 nanocrystals, (b) statistic particle size distri‐ bution obtained from low-magnitude TEM image, (c) high-magnitude TEM image, (d) selected area electron diffrac‐ tion pattern, in which the first six diffraction rings are indexed as (012), (104)/(110), (006)/(202), (024), (116)/(122), and (214). (d) HRTEM image of a single perovskite BFO nanocrystal with a diameter of ~12 nm, and (e) Fourier filtered HRTEM image. Inset is a fast Fourier transform (FFT) pattern of the nanocrystal. Reproduced with permission from [73].

## **3. Structural characterization of perovskite oxide nanocrystals**

Up to date, numerous TEM investigations have been carried out to investigate the microstruc‐ tural features of perovskite BaTiO3 nanocrystals [79–83]. For example, the internal pore structures in the hydrothermal BaTiO3 nannopowders were examined by TEM [79]. Figure 3 shows TEM images of (a) as-received BaTiO3 powder (particle size ~60 nm) and (b) BaTiO3 powder annealed at 673 K. It was observed that some particles had internal pores of various sizes in as-received BaTiO3 powders, as indicated by the arrows in Figure 3(a). A large pore with a cubic-shaped was observed in the particle annealed at 673 K in Figure 3(b). Lattice images were observed around the particle, and these indicated that the pore existed inside the particle. A three-dimensional structure of the internal pore was successfully observed and constructed by TEM tomography system. The results showed that no inclusion was found in the pores, and such internal pores were not lattice defects at the atomic level because their sizes were a few tens of nanometers. Large pores were involved in as-received BaTiO3 powders, and their numbers decreased at >1,128 K. Some of the internal pores were released from the particle's surface and/or during the grain growth. The presence of the pores affected the density of the BaTiO3 particle. The behavior of the internal pore was observed in situ with increasing temperature on the thermal stage of a TEM device. The results showed that at >1,128 K, some pores moved out from the particle's surface during TEM observation. This temperature roughly agreed with the temperature at which the density of BaTiO3 powder sharply increases. During observation with increasing temperature, a thin layer appeared on the particle's surface at temperature over 573 K and then disappeared at 1,193 K.

The hydrothermal BaTiO3 powder with a small particle size is stabilized in a cubic phase at room temperature [80–83], which implies that the distortion of the [TiO6] structure resulting in a cubic-to-tetragonal phase transition as cooling the sample through the Curie temperature is not taking place. A plausible reason is that the small size of the BaTiO3 nanocrystals, which are so small that the structural defects in the particles prevent the completion of the structural transition, leading to high strains within the crystals. The high strains inside the nanoparticles introduced by structural defects (e.g., lattice defects) would make the unit cell distortion (c/a ratio) much smaller than that in the standard BaTiO3. To reveal the high strains in the hydro‐ thermal BaTiO3 nanoparticles by TEM images, Zhu et al. [81] recorded both bright- and darkfield TEM images from the hydrothermal BaTiO3 nanoparticles. Figure 4(a) is a bright-field TEM image recorded by using a small objective aperture that selects only the (000) central transmitted beam, which shows narrow-distribution spherical nanoparticles. The dark-field image shown in Figure 4(b) was recorded by using a smaller objective aperture that selects the part of the {100} and {110} reflections, as indicated by a circle in Figure 4(c). The dark-field image displayed in Figure 4(b) clearly shows high strains in some BaTiO3 nanoparticles. By using the bright- and dark-field TEM images, Lu et al. [82] also reported several types of TEM contrast variations in an individual BaTiO3 nanocrystal synthesized *via* hydrothermal method at a temperature of 230<sup>ο</sup> C. It is believed that the different types of variations of TEM contrast indicate the existence of different strains in BaTiO3 nanograins. Therefore, in a TEM image, large strain is indicated by a contrast variation across a particle. If a particle is single crystalline and has no strain, it should be uniform in contrast. However, for a single crystalline particle, if the TEM image shows dark-bright variation in contrast, it is likely to have a high strain within the grain. Strain affects the diffraction behavior of the electrons, resulting in dramatic contrast change. The hydrothermal BaTiO3 nanoparticles exhibit a cubic structure (a high-temperature phase) at room temperature; such an abnormal crystallographic phenomenon is closely related to the existence of high strains in these BaTiO3 nanoparticles. The strains introduced by a high concentration of lattice defects such as OH– ions and barium vacancies can make the unit cell distortion (*c/a* ratio) much smaller compared with that of the standard BaTiO3. As a result, no peak splitting was detected in the XRD patterns of the hydrothermal BaTiO3 powders even though they belong to the tetragonal phase.

**3. Structural characterization of perovskite oxide nanocrystals**

162 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

at temperature over 573 K and then disappeared at 1,193 K.

at a temperature of 230<sup>ο</sup>

Up to date, numerous TEM investigations have been carried out to investigate the microstruc‐ tural features of perovskite BaTiO3 nanocrystals [79–83]. For example, the internal pore structures in the hydrothermal BaTiO3 nannopowders were examined by TEM [79]. Figure 3 shows TEM images of (a) as-received BaTiO3 powder (particle size ~60 nm) and (b) BaTiO3 powder annealed at 673 K. It was observed that some particles had internal pores of various sizes in as-received BaTiO3 powders, as indicated by the arrows in Figure 3(a). A large pore with a cubic-shaped was observed in the particle annealed at 673 K in Figure 3(b). Lattice images were observed around the particle, and these indicated that the pore existed inside the particle. A three-dimensional structure of the internal pore was successfully observed and constructed by TEM tomography system. The results showed that no inclusion was found in the pores, and such internal pores were not lattice defects at the atomic level because their sizes were a few tens of nanometers. Large pores were involved in as-received BaTiO3 powders, and their numbers decreased at >1,128 K. Some of the internal pores were released from the particle's surface and/or during the grain growth. The presence of the pores affected the density of the BaTiO3 particle. The behavior of the internal pore was observed in situ with increasing temperature on the thermal stage of a TEM device. The results showed that at >1,128 K, some pores moved out from the particle's surface during TEM observation. This temperature roughly agreed with the temperature at which the density of BaTiO3 powder sharply increases. During observation with increasing temperature, a thin layer appeared on the particle's surface

The hydrothermal BaTiO3 powder with a small particle size is stabilized in a cubic phase at room temperature [80–83], which implies that the distortion of the [TiO6] structure resulting in a cubic-to-tetragonal phase transition as cooling the sample through the Curie temperature is not taking place. A plausible reason is that the small size of the BaTiO3 nanocrystals, which are so small that the structural defects in the particles prevent the completion of the structural transition, leading to high strains within the crystals. The high strains inside the nanoparticles introduced by structural defects (e.g., lattice defects) would make the unit cell distortion (c/a ratio) much smaller than that in the standard BaTiO3. To reveal the high strains in the hydro‐ thermal BaTiO3 nanoparticles by TEM images, Zhu et al. [81] recorded both bright- and darkfield TEM images from the hydrothermal BaTiO3 nanoparticles. Figure 4(a) is a bright-field TEM image recorded by using a small objective aperture that selects only the (000) central transmitted beam, which shows narrow-distribution spherical nanoparticles. The dark-field image shown in Figure 4(b) was recorded by using a smaller objective aperture that selects the part of the {100} and {110} reflections, as indicated by a circle in Figure 4(c). The dark-field image displayed in Figure 4(b) clearly shows high strains in some BaTiO3 nanoparticles. By using the bright- and dark-field TEM images, Lu et al. [82] also reported several types of TEM contrast variations in an individual BaTiO3 nanocrystal synthesized *via* hydrothermal method

indicate the existence of different strains in BaTiO3 nanograins. Therefore, in a TEM image, large strain is indicated by a contrast variation across a particle. If a particle is single crystalline and has no strain, it should be uniform in contrast. However, for a single crystalline particle, if the TEM image shows dark-bright variation in contrast, it is likely to have a high strain within the grain. Strain affects the diffraction behavior of the electrons, resulting in dramatic contrast

C. It is believed that the different types of variations of TEM contrast

**Figure 3.** Transmission electron microscopy images of (a) as-received BaTiO3 powder (particle size ~60 nm) and (b) Ba‐ TiO3 powder annealed at 673 K. Figures reproduced with permission from [79].

**Figure 4.** (a) Bright-field and (b) dark-field TEM images recorded from the hydrothermal BaTiO3 nanoparticles. (c) A selected area electron diffraction pattern from the BaTiO3 particles showing a perovskite structure. The circle indicates the size and position of the objective aperture used to record the dark-field image displayed in (b). Figures reproduced with permission from [80].

Due to the ability of revealing the local atomic structures, HRTEM image is the most useful and appropriate technique for identifying the local structures at the edges of perovskite nanocrystals. For example, a terrace-ledge-kink (TLK) surface structure is frequently observed at the edges of the hydrothermal perovskite BaTiO3 [84,85] and ZnZrO3 [86] nanocrystals with rough surface morphology, and in most cases the terraces and ledges lie on the {100} planes, as shown in Figures 5 and 6, respectively. Such a TLK surface structure can be explained by the periodic bond chain theory, which was originally developed by Hartmann and Perdok [87]. The rarely seen {110} surface in the perovskite BaTiO3 and ZnZrO3 nanoparticles were found to be reconstructed so that the surface was composed of corners bound by {100} mini-faces like the triangular small islands.

Internal defect textures, such as nanoscale multiple (111) twining and complicated (111) intergrowth defects, were also observed in the BaTiO3 nanocrystals synthesized by sol–gel and stearic acid-gel (SAG) methods. They were identified as hexagonal-type BaTiO3 structure [88, 89]. Complex arrangements of defects lying on the (111) plane were observed in the SAGderived BaTiO3 nanocrystal with particle size of 10 nm. The density of the small defects was estimated to be on the order of 1027/m3 in the SAG-derived BaTiO3 nanocrystals. These high density of defects could result in the cubic phase structure of SAG-derived BaTiO3 powders even with grain size large up to 3.50 μm [88].

**Figure 5.** HRTEM images of the surface structures at the edges of BaTiO3 nanoparticles viewed from the [001] direc‐ tion. (a) Both a terrace-ledge-kink (TLK) surface structure and small nucleated and triangular islands with two to three atomic layer thickness are observed. (b) and (c) TLK surface structure with both terraces and ledges lying on the {100} planes; only a small amount of ledges lie on the (110) plane. The inset in (b) is a Fourier-filtered image of the corre‐ sponding position, which clearly demonstrates two perpendicular sets of (100) and (010) planes. Reproduced with per‐ mission from [84].

Due to the ability of revealing the local atomic structures, HRTEM image is the most useful and appropriate technique for identifying the local structures at the edges of perovskite nanocrystals. For example, a terrace-ledge-kink (TLK) surface structure is frequently observed at the edges of the hydrothermal perovskite BaTiO3 [84,85] and ZnZrO3 [86] nanocrystals with rough surface morphology, and in most cases the terraces and ledges lie on the {100} planes, as shown in Figures 5 and 6, respectively. Such a TLK surface structure can be explained by the periodic bond chain theory, which was originally developed by Hartmann and Perdok [87]. The rarely seen {110} surface in the perovskite BaTiO3 and ZnZrO3 nanoparticles were found to be reconstructed so that the surface was composed of corners bound by {100} mini-faces like

Internal defect textures, such as nanoscale multiple (111) twining and complicated (111) intergrowth defects, were also observed in the BaTiO3 nanocrystals synthesized by sol–gel and stearic acid-gel (SAG) methods. They were identified as hexagonal-type BaTiO3 structure [88, 89]. Complex arrangements of defects lying on the (111) plane were observed in the SAGderived BaTiO3 nanocrystal with particle size of 10 nm. The density of the small defects was

density of defects could result in the cubic phase structure of SAG-derived BaTiO3 powders

**Figure 5.** HRTEM images of the surface structures at the edges of BaTiO3 nanoparticles viewed from the [001] direc‐ tion. (a) Both a terrace-ledge-kink (TLK) surface structure and small nucleated and triangular islands with two to three atomic layer thickness are observed. (b) and (c) TLK surface structure with both terraces and ledges lying on the {100} planes; only a small amount of ledges lie on the (110) plane. The inset in (b) is a Fourier-filtered image of the corre‐ sponding position, which clearly demonstrates two perpendicular sets of (100) and (010) planes. Reproduced with per‐

in the SAG-derived BaTiO3 nanocrystals. These high

the triangular small islands.

mission from [84].

estimated to be on the order of 1027/m3

even with grain size large up to 3.50 μm [88].

164 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 6.** HRTEM images of the surface structures at the edges of perovskite ZnZrO3 nanocrystals synthesized at dif‐ ferent Zn/Zr molar ratios in the precursors. (a) Zn/Zr = 3.0, and (b) Zn/Zr = 4.0. Inset in Figure a is the FFT patterns of the corresponding HRTEM image, and inset in Figure d is the SAED pattern taken from the [010]-zone axis. The {100} and (101) facets are indicated in Figure a, and surface steps lying on the {100} planes are indicated in Figure b. Repro‐ duced with permission from [86].

**Figure 7.** Atomic-scale imaging of the electric dipoles formed by the relative displacements of the Zr/Ti cation columns and the O anion columns in the approximately 10-nm-thick PbZr0.2Ti0.8O3layer sandwiched between two SrTiO3 layers. The image was viewed along the [] direction and recorded under negative spherical-aberration imaging conditions. The atom columns appear bright on a dark background. The horizontal arrows denote the horizontal interfaces be‐ tween the PbZr0.2Ti0.8O3 film and the top and the bottom SrTiO3 film layers. The dotted line traces the 180° domain wall between the domain I and domain II. The arrows denoted by 'PS' show the directions of the polarization in the 180° domains. Two insets show higher magnifications of the dipoles formed by the displacements of ions in the unit cells. Yellow symbols denote PbO atom columns seen end-on, red symbols for Zr/Ti columns, and blue symbols for oxygen. Reproduced with permission from [92].

Traditionally, spherical aberration (Cs) of magnetic lenses limits the resolutions of HRTEMand STEM images. In recent years, spherical aberration correctors (e.g., hexapole type Cs-correc‐ tors proposed by Rose [90]) have been developed to substantially reduce the effective value of Cs of the objective lens. The Cs-corrected HRTEM mode offers a tunable spherical aberration coefficient from negative to positive values. Properly combining a negative Cs with a positive defocus, at no cost to point resolution, an HRTEM image with bright-contrast of atoms on dark background can be obtained, which can be directly interpreted without image simulation, and light elements such as oxygen atoms and even their vacancies can also be imaged [91,92]. For example, by using the Cs-corrected imaging technique,Jia et al.[92]first performed the atomicscale investigations of the electric dipoles near (charged and uncharged) 180<sup>ο</sup> domain walls in thin epitaxial PbZr0.2Ti0.8O3 film sandwiched between two SrTiO3 layers. Figure 7 is an atomicscale image of the electric dipoles formed by the relative displacements of the Zr/Ti cation columns andtheOanioncolumns inPbZr0.2Ti0.8O3 film,viewedfromthedirectionandrecorded under negative spherical-aberration imaging conditions. The local tetragonality c/a and spontaneous polarization inside the domains and across the domain wall were calculated. For the firsttime, a largedifference in atomicdetails between chargedandunchargeddomain walls was reported. Such breakthrough would improve our ability to see and thoroughly explore the properties of perovskite nanocrystals. We can foresee that the new Cs-corrected HRTEM and STEM will benefit perovskite nanopowder materials research in the new era.

## **4. Properties of perovskite nanocrystals**

## **4.1. Ferroelectric and dielectric properties**

Perovskite BaTiO3 nanocrystals can be used as initial building blocks to fabricate thin films, which exhibit highly uniform nanostructured texture and grain sizes. Recently, well-isolated BaTiO3 nanocrystals smaller than 10 nm with control over aggregation and crystal densities have been synthesized and used to construct films with a uniform nanocrystalline grain texture [93]. The ferroelectric behavior was found in these BaTiO3 nanocrystallined films with grain sizes in the range of 10–30 nm. Their relative dielectric constants were in the range of 85–90 over the 1–100 KHz with low dielectric loss of 0.03–0.04, representing a promising application in thin-film capacitance [93]. The nanometer-scale ferroelectric property of tetragonal BaTiO3 particles with sizes of 6–12 nm is also reported by Nuraje et al. [94]. The ferroelectric polari‐ zation of these nanoparticles can be manipulated by electrostatic force microscopy (EFM), as demonstrated in Figure 8. First, the electric polarization of the BaTiO3 nanoparticles was manipulated by applying a voltage, *V*write, to the conductive atomic force microscopy(AFM) tip that gently contacts the nanoparticles (Figure 8a). After the local electric polarization is written onto the nanoparticles, the resulting polarization is probed using EFM with a lower voltage, *V*probe, by measuring the shift in the resonance frequency of the AFM tip. As shown in Figure 8(a), during the probing process the AFM tip is raised at a constant height above the nanoparticles in order to avoid interference between the manipulated polarization and *V*probe. The raised distance of 50 nm enables one to image only the contribution from the surface charges associated with the local electric polarization of the BaTiO3 nanoparticles. After a *V*write of +12 V was applied to BaTiO3 nanoparticles with an average diameter of 12 nm (Figure 8b),

**Figure 8.** (a) Schematic diagram of manipulating and probing the electric polarization of BaTiO3 nanoparticles with EFM. (b) Topological AFM image of BaTiO3 nanoparticles. (c) and (d) EFM images of BaTiO3 nanoparticles with *V*probe = +2 V after *V*write = ± 12 V was applied on the nanoparticles across a conductive AFM tip and a gold substrate. Repro‐ duced with permission from [95].

 the EFM image of those nanoparticles to which a *V*probe of +2 V was applied, shown in Figure 8(c), appears in a brighter contrast compared with the background due to the repulsive electrostatic interaction between the tip and the nanoparticles. After a *V*write of –12 V applied to the same BaTiO3 nanoparticles, the EFM image of those nanoparticles to which a *V*probe of – 2 V was applied, shown in Figure 8(d), appears in a darker contrast compared with the background due to the attractive electrostatic interaction. It should be noted that the experi‐ ments that involved scanning the manipulated nanoparticles with *V*probe = –2 V resulted in reverse EFM images, which confirms that the probe voltage did not interfere significantly with the written polarization. These EFM images indicate that the BaTiO3 nanoparticles synthesized in the peptide nanorings at room temperature possess a ferroelectric property with spontane‐ ous electric polarization, which can be reoriented by an external electric field.

## **4.2. Magnetic properties**

Traditionally, spherical aberration (Cs) of magnetic lenses limits the resolutions of HRTEMand STEM images. In recent years, spherical aberration correctors (e.g., hexapole type Cs-correc‐ tors proposed by Rose [90]) have been developed to substantially reduce the effective value of Cs of the objective lens. The Cs-corrected HRTEM mode offers a tunable spherical aberration coefficient from negative to positive values. Properly combining a negative Cs with a positive defocus, at no cost to point resolution, an HRTEM image with bright-contrast of atoms on dark background can be obtained, which can be directly interpreted without image simulation, and light elements such as oxygen atoms and even their vacancies can also be imaged [91,92]. For example, by using the Cs-corrected imaging technique,Jia et al.[92]first performed the atomic-

thin epitaxial PbZr0.2Ti0.8O3 film sandwiched between two SrTiO3 layers. Figure 7 is an atomicscale image of the electric dipoles formed by the relative displacements of the Zr/Ti cation columns andtheOanioncolumns inPbZr0.2Ti0.8O3 film,viewedfromthedirectionandrecorded under negative spherical-aberration imaging conditions. The local tetragonality c/a and spontaneous polarization inside the domains and across the domain wall were calculated. For the firsttime, a largedifference in atomicdetails between chargedandunchargeddomain walls was reported. Such breakthrough would improve our ability to see and thoroughly explore the properties of perovskite nanocrystals. We can foresee that the new Cs-corrected HRTEM and

Perovskite BaTiO3 nanocrystals can be used as initial building blocks to fabricate thin films, which exhibit highly uniform nanostructured texture and grain sizes. Recently, well-isolated BaTiO3 nanocrystals smaller than 10 nm with control over aggregation and crystal densities have been synthesized and used to construct films with a uniform nanocrystalline grain texture [93]. The ferroelectric behavior was found in these BaTiO3 nanocrystallined films with grain sizes in the range of 10–30 nm. Their relative dielectric constants were in the range of 85–90 over the 1–100 KHz with low dielectric loss of 0.03–0.04, representing a promising application in thin-film capacitance [93]. The nanometer-scale ferroelectric property of tetragonal BaTiO3 particles with sizes of 6–12 nm is also reported by Nuraje et al. [94]. The ferroelectric polari‐ zation of these nanoparticles can be manipulated by electrostatic force microscopy (EFM), as demonstrated in Figure 8. First, the electric polarization of the BaTiO3 nanoparticles was manipulated by applying a voltage, *V*write, to the conductive atomic force microscopy(AFM) tip that gently contacts the nanoparticles (Figure 8a). After the local electric polarization is written onto the nanoparticles, the resulting polarization is probed using EFM with a lower voltage, *V*probe, by measuring the shift in the resonance frequency of the AFM tip. As shown in Figure 8(a), during the probing process the AFM tip is raised at a constant height above the nanoparticles in order to avoid interference between the manipulated polarization and *V*probe. The raised distance of 50 nm enables one to image only the contribution from the surface charges associated with the local electric polarization of the BaTiO3 nanoparticles. After a *V*write of +12 V was applied to BaTiO3 nanoparticles with an average diameter of 12 nm (Figure 8b),

domain walls in

scale investigations of the electric dipoles near (charged and uncharged) 180<sup>ο</sup>

166 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

STEM will benefit perovskite nanopowder materials research in the new era.

**4. Properties of perovskite nanocrystals**

**4.1. Ferroelectric and dielectric properties**

The magnetic properties of perovskite oxide nanocrystals are dependent upon the particle size or large surface-area-to-volume ratio. For most biomedical and magnetofluidic applications, magnetic nanoparticles with fairly uniform size and having a Curie temperature above room temperature are highly required. In this application, a well-controlled, reproducible, and narrow distribution of ferromagnetic nanoparticle sizes is important in addition to having a large magnetic moment per particle. Lipham et al. [99] have synthesized polycrystalline, nanometer-sized powders of La1–*x*SrxMnO3 by a citrate gel technique. The particle sizes of which were in the range of 11–17 nm, their saturated magnetizations at 300 K in the range of 7–26 emu/g, and magnetic transition temperature from 275 to 350K. The saturated magneti‐ zation and magnetic transition temperature increased linearly with increasing the average particle size.

The magnetic properties of single-crystalline multiferroic BiFeO3 nanoparticles were also reported by Park et al. [100] and Selbach et al. [101]. Their results demonstrate that the BiFeO3 nanoparticles exhibit strong size-dependent magnetic properties that correlate with: (a) increased suppression of the known spiral spin structure (period length of ~62 nm) with decreasing nanoparticle size and (b) uncompensated spins and strain anisotropies at the surface. Both zero-field-cooled and field-cooled magnetization curves exhibit spin-glass freezing behavior due to a complex interplay between finite size effects, inter-particle inter‐ actions, and a random distribution of anisotropy axes in the nanoparticle assemblies.

## **4.3. Multiferroic properties**

Multiferroic materials are characterized by the coexistence of spin, orbital, and/or electricdipole orders and cross-term effects between the corresponding different degrees of freedom [102,103]. Magnetoelectric (ME) multiferroics are an example of materials that combine simultaneously ferromagnetic and ferroelectric in the same phase, with coupling between the two orders. That implies they possess spontaneous magnetization which can be reoriented by an applied magnetic field, and spontaneous polarization which can be reoriented by an applied electric field. Up to date, large ME effects have been observed in the form of ferroelectric phase transitions induced by magnetic fields in perovskite manganites [104] and switching local ferromagnetism by an electric field through the coupling between multiferroic BiFeO3 and a ferromagnet [105,106]. Magnetoelectric memory effects and magnetic switching of ferroelectric domains (and the converse process) have been demonstrated in many multiferroic materials. By using an optical second harmonic generation technique ferrotoroidic domains are spatially revealed in LiCoPO4 multiferroics, where independent antiferromagnetic domains exist; with this method, the coupling between magnetic and electric domains is also confirmed [107]. Due to the weak coupling behaviors between the ferroelectric and the magnetic order parameters in the single-phase multiferroics and small values of the electric and magnetic polarizations at room temperature, the applications of single-phase multiferroics are not very attractive in the near future. However, for the single-phase multiferroics, there still exist some basic questions to be answered such as the origin of the ferroelectricity in some unusual multiferroics [104]. Up to date, several coupling mechanisms between the magnetic and ferroelectric ordering in the single-phase multiferroics have been proposed and extensively investigated; it is still unclear how to obtain high-temperature single-phased ferromagnetic–ferroelectric multiferroicity. From a viewpoint of the practical applications, it is very important and is still a major challenge to develop robust room-temperature ferromagnetic ferroelectrics that are sufficiently insulating to sustain a large macroscopic polarization [108].

## **5. Applications of perovskite oxide nanocrystals**

Due to their high dielectric, ferroelectric, piezoelectric, pyroelectric, and electro-optic proper‐ ties, perovskite oxide nanocrystals have wide ranges of applications, such as multilayered ceramic capacitors, ferroelectric memories, voltage tunable capacitors, surface acoustic wave devices, microactuators, and IR detectors. An emerging application seeks to exploit the multiferroic properties of perovskite oxide nanocrystals to develop novel multifunctional devices controlled by magnetic and electric fields. This magnetoelectric coupling enables the manipulation of the ferroelectric polarization by a magnetic field [104] or the control of the antiferromagnetic vector orientation by an electric field [105]. This latter opportunity is very appealing for spintronics as it may allow to control the magnetization of a ferromagnet, exchange coupled to a ferroelectric antiferromagnet, through an electric field. In other words, it may enable electrical writing of magnetic information with a low power consumption, like an magnetic random access memory(MRAM) cell [109]. In the subsequent sections, typical applications of perovskite oxide nanocrystals are introduced.

## **5.1. Nanoelectronics**

The magnetic properties of single-crystalline multiferroic BiFeO3 nanoparticles were also reported by Park et al. [100] and Selbach et al. [101]. Their results demonstrate that the BiFeO3 nanoparticles exhibit strong size-dependent magnetic properties that correlate with: (a) increased suppression of the known spiral spin structure (period length of ~62 nm) with decreasing nanoparticle size and (b) uncompensated spins and strain anisotropies at the surface. Both zero-field-cooled and field-cooled magnetization curves exhibit spin-glass freezing behavior due to a complex interplay between finite size effects, inter-particle inter‐

168 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

actions, and a random distribution of anisotropy axes in the nanoparticle assemblies.

sufficiently insulating to sustain a large macroscopic polarization [108].

Due to their high dielectric, ferroelectric, piezoelectric, pyroelectric, and electro-optic proper‐ ties, perovskite oxide nanocrystals have wide ranges of applications, such as multilayered ceramic capacitors, ferroelectric memories, voltage tunable capacitors, surface acoustic wave devices, microactuators, and IR detectors. An emerging application seeks to exploit the

**5. Applications of perovskite oxide nanocrystals**

Multiferroic materials are characterized by the coexistence of spin, orbital, and/or electricdipole orders and cross-term effects between the corresponding different degrees of freedom [102,103]. Magnetoelectric (ME) multiferroics are an example of materials that combine simultaneously ferromagnetic and ferroelectric in the same phase, with coupling between the two orders. That implies they possess spontaneous magnetization which can be reoriented by an applied magnetic field, and spontaneous polarization which can be reoriented by an applied electric field. Up to date, large ME effects have been observed in the form of ferroelectric phase transitions induced by magnetic fields in perovskite manganites [104] and switching local ferromagnetism by an electric field through the coupling between multiferroic BiFeO3 and a ferromagnet [105,106]. Magnetoelectric memory effects and magnetic switching of ferroelectric domains (and the converse process) have been demonstrated in many multiferroic materials. By using an optical second harmonic generation technique ferrotoroidic domains are spatially revealed in LiCoPO4 multiferroics, where independent antiferromagnetic domains exist; with this method, the coupling between magnetic and electric domains is also confirmed [107]. Due to the weak coupling behaviors between the ferroelectric and the magnetic order parameters in the single-phase multiferroics and small values of the electric and magnetic polarizations at room temperature, the applications of single-phase multiferroics are not very attractive in the near future. However, for the single-phase multiferroics, there still exist some basic questions to be answered such as the origin of the ferroelectricity in some unusual multiferroics [104]. Up to date, several coupling mechanisms between the magnetic and ferroelectric ordering in the single-phase multiferroics have been proposed and extensively investigated; it is still unclear how to obtain high-temperature single-phased ferromagnetic–ferroelectric multiferroicity. From a viewpoint of the practical applications, it is very important and is still a major challenge to develop robust room-temperature ferromagnetic ferroelectrics that are

**4.3. Multiferroic properties**

An important application of perovskite oxide nanocrystals in electronics is the multilayered ceramic capacitors(MLCCs). The MLCCs based on BaTiO3 powders have used in many electronic devices such as video camera, cell phone, laptop computers, and automobiles. Two key features of the MLCCs have attracted high degree of interest in them. One is the discovery of the ferroelectric properties of perovskite BaTiO3 and its high relative dielectric constant. The other is the technical breakthrough for making multilayer ceramic units in a small volume, thus satisfying the need for an economical manufacture of systems, and the requirements for printed circuit boards and for hybrid circuits on ceramic substrates. With increasing the volumetric efficiency of electronic devices, MLCCs are needed to use much thin dielectric layers. Recently, it is expected that the dielectric layers as thin as 1 μm or less will be available in MLCCs for the next generation of electronic components. Besides the thin dielectric layers, other key factors for the development of future highly volume-efficient and high-capacitance MLCCs should be considered, such as the use of high dielectric films, large numbers of active dielectric layers, improvement in the overlap area, and stacking precision of the electrodes. For ideal BaTiO3 nanopowders used for the next generation MLCCs, they should have high purity, homogeneous compositions and cation distributions, uniform sizes and shapes, and weak agglomeration. Generally, solution-based techniques are used to prepare such very fine powders, since these techniques could synthesize homogeneous, phase-pure, and stoichio‐ metric BaTiO3 nanocrystals with finer particle sizes due to the finer scale of mixing and subsequently lower processing temperature.

## **5.2. Information storage devices**

Ferroelectric materials possess an electric dipole moment even in the absence of an external electric field. Such a spontaneous polarization is caused by the positional bistability of constituent ions in the crystal and its direction can be adjusted by an external field. Since the response time of the ion displacement is the order of ns or less, so non-volatile random access memory called FeRAM (ferroelectric random access memory) can be realized using ferroelec‐ tric capacitors, in which two states of "0" and "1" in the binary logic are represented by the direction of the spontaneous polarization. Two different types of FeRAM cells can be used to achieve data storage and read operations, which are named as 1T-1C and 2T-2C configurations [110]. Because the 2T-2C cell occupies a large area, it is only used for FeRAM memories with densities of 256 kbit or less. A significantly smaller cell size can be achieved by eliminating the complementary reference capacitor in each cell and accessing a reference signal from a single reference capacitor placed outside the cell array. The cell size in 1T-1C configuration is reduced because only one transistor and one capacitor are required to form the popular cell. In the FeRAM cell, the polarization direction is set by a positive or negative voltage pulse to the ferroelectric capacitor defining logical "0" or "1". For readout, another voltage pulse is applied and the stored bit configuration determines whether or not the polarization switched direction. The ferroelectric capacitor suffers from the fact that data are destroyed during readout and bit reprogramming is required after each read cycle, i.e., the data read process is destructive. As a result, switching the polarization during both write and read operations would cause reduced endurance, resulting in short lifetime of FeRAM cells because of the fatigue problem of the ferroelectric material. In addition, during the read-processing data storage is volatile and data could be erased if powder supply is lost during the reading process. To overcome the destructive readout scheme of FeRAM, recently considerable efforts have focused on the development of a so-called ferroelectric gate field-effect transistor (FeFET). The principle of a ferroelectric-gate field-effect transistor is based on a conventional Si MOSFET (metal oxide semiconductor ferroelectric-gate field effect transistor) whose gate dielectrics is a ferroelectric materials. This FET-type FeRAMs have such unique features as non-volatile data storage, nondestructive data readout, and the single-transistor-type cell structure. Further advantages are the reduced power consumption and better scaling properties as in the case of usual FeRAM cells. Although the FET-type FeRAMs have been studied since the 1950s [111], however, up to date the commercially available devices have not been fabricated, mainly due to the interfacial structures between the perovskite ferroelectric and a semiconductor. Any imperfections at the interface, such as the formation of undesirable phases or electronic trapping states, will seriously degrade the performance of the device. Essential challenges for the FeFET are the improvement of the retention time and the suppression of serious parasitic effects such as the charge traps at the Si–ferroelectric interface. One possible solution in the case of a FeFET is the incorporation of one insulating buffer layer between the Si and the ferroelectric, which is composed of either a dielectric material (MFIS structure) or a stacked structure of conductive and dielectric materials (MFMIS structure). Several buffer layers such as SiO2, CeOx, and Si3N4 have been investigated. The most recent results by using alternative gate oxides such as HfO2 or HfAlO show encouraging results, and the retention time for the gate layer sequence of Pt/SBT/HfO2/Si is up to 30 days [112]. However, insertion of the buffer layer causes new problems such as current injection that lead to short data retention time. Another considered factor is the leakage current of both the ferroelectric film and the buffer layer. If the charge neutrality at a node between the two capacitors is destroyed by the leakage current, electric charges on the electrodes of the buffer layer capacitor disappear, which means that carriers on the semiconductor surface disappear and the stored data cannot be readout by drain current of the FET, even if the polarization of the ferroelectric film is retained. FeFET based on epitaxial perovskite heterostructures was also demonstrated, in which doped rare-earth manganates such as La0.7Ca0.3MnO3 were used as the semiconductor channel material, and Pb(Zr0.2Ti0.8)O3 as the ferroelectric gate in the prototypical epitaxial field effect device [112]. The carrier concentration of the semiconductor channel can be tuned by varying the manganate stoichi‐ ometry. The enhanced interface characteristics associated with the ferroelectric-manganate interface allows for the fabrication of field effect devices with channel resistance modulation of at least a factor of 3 and retention on the order of hours.

Perovskite oxide nanocrystals also exhibit good dielectric properties, which can be used for dynamic random access memory (DRAM). The DRAM is the primary working medium for information storage in the microelectronic devices that comprise the entire litany of electronic systems. DRAM works by using a submicron-sized capacitor, representing one bit of memo‐ ry, to store a given amount of electrical charge: if the charge is present, it represents a digital "1", if not then the bit is a "0". Each bit is addressed using a complementary metal oxide semiconductor field effect transistor, which acts as a valve for adding to or removing charge from the capacitor upon application of a voltage. Conventional DRAMs employ either SiO2 capacitors, or a combination of SiO2 and Si3N4 nitride, which is termed ONO (oxy-nitride). The next generation of DRAMs will utilize Ta2O5 replacing ONO, but the tantalum oxide relative dielectric constant is only about 25, and it appears that the DRAM evolution will skip this intermediate stage and pass directly to very high dielectric materials (εr = 500 – 1500) that are ferroelectric or nearly ferroelectric. Many of these materials are perovskite oxides with ABO3 perovskite, or closely related variations of perovskites. Several good reviews of this technolo‐ gy are given by Gnade et al.[113], Tasch and Parker[114], and Scott[115].In going from 64 Mbit to 4 Gbit DRAM,the most promising dielectric material is perovskite barium strontium titanate (BST), which can be processed by physical deposition, especially sputtering with good results [116]. A dielectric material should have a low leakage current and high dielectric constant, and it is desirable to use the paraelectric BST for DRAM. Among the Pb(Zr,Ti)O3 (PZT), SrTiO3, and BST ferroelectric materials, (Ba,Sr)TiO3 is the most promising material for DRAM capacitors because it is a better insulator with a higher dielectric constant than PZT and SrTiO3, and it can be controlled to be the paraelectric phase with an appropriate ratio of Ba/Sr composition [117].

## **5.3. Spintronics**

densities of 256 kbit or less. A significantly smaller cell size can be achieved by eliminating the complementary reference capacitor in each cell and accessing a reference signal from a single reference capacitor placed outside the cell array. The cell size in 1T-1C configuration is reduced because only one transistor and one capacitor are required to form the popular cell. In the FeRAM cell, the polarization direction is set by a positive or negative voltage pulse to the ferroelectric capacitor defining logical "0" or "1". For readout, another voltage pulse is applied and the stored bit configuration determines whether or not the polarization switched direction. The ferroelectric capacitor suffers from the fact that data are destroyed during readout and bit reprogramming is required after each read cycle, i.e., the data read process is destructive. As a result, switching the polarization during both write and read operations would cause reduced endurance, resulting in short lifetime of FeRAM cells because of the fatigue problem of the ferroelectric material. In addition, during the read-processing data storage is volatile and data could be erased if powder supply is lost during the reading process. To overcome the destructive readout scheme of FeRAM, recently considerable efforts have focused on the development of a so-called ferroelectric gate field-effect transistor (FeFET). The principle of a ferroelectric-gate field-effect transistor is based on a conventional Si MOSFET (metal oxide semiconductor ferroelectric-gate field effect transistor) whose gate dielectrics is a ferroelectric materials. This FET-type FeRAMs have such unique features as non-volatile data storage, nondestructive data readout, and the single-transistor-type cell structure. Further advantages are the reduced power consumption and better scaling properties as in the case of usual FeRAM cells. Although the FET-type FeRAMs have been studied since the 1950s [111], however, up to date the commercially available devices have not been fabricated, mainly due to the interfacial structures between the perovskite ferroelectric and a semiconductor. Any imperfections at the interface, such as the formation of undesirable phases or electronic trapping states, will seriously degrade the performance of the device. Essential challenges for the FeFET are the improvement of the retention time and the suppression of serious parasitic effects such as the charge traps at the Si–ferroelectric interface. One possible solution in the case of a FeFET is the incorporation of one insulating buffer layer between the Si and the ferroelectric, which is composed of either a dielectric material (MFIS structure) or a stacked structure of conductive and dielectric materials (MFMIS structure). Several buffer layers such as SiO2, CeOx, and Si3N4 have been investigated. The most recent results by using alternative gate oxides such as HfO2 or HfAlO show encouraging results, and the retention time for the gate layer sequence of Pt/SBT/HfO2/Si is up to 30 days [112]. However, insertion of the buffer layer causes new problems such as current injection that lead to short data retention time. Another considered factor is the leakage current of both the ferroelectric film and the buffer layer. If the charge neutrality at a node between the two capacitors is destroyed by the leakage current, electric charges on the electrodes of the buffer layer capacitor disappear, which means that carriers on the semiconductor surface disappear and the stored data cannot be readout by drain current of the FET, even if the polarization of the ferroelectric film is retained. FeFET based on epitaxial perovskite heterostructures was also demonstrated, in which doped rare-earth manganates such as La0.7Ca0.3MnO3 were used as the semiconductor channel material, and Pb(Zr0.2Ti0.8)O3 as the ferroelectric gate in the prototypical epitaxial field effect device [112]. The carrier concentration of the semiconductor channel can be tuned by varying the manganate stoichi‐

170 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Applications of the magnetoelectric coupling and multiferroics in the fields of spintronics are increasing rapidly, and numerous possible device architectures have been proposed and fabricated [118]. For example, in the ferroelectric antiferromagnets, such as the multiferroic perovskites, the magnetic structure could be modulated or controlled by the application of an electric field [119,120]. For example, non-volatile control of the orientation of the antiferro‐ magnetic axis can be achieved by using the coupling between ferroelectricity and antiferro‐ magnetism that can provide permanent [121]. Similarly, the coupling between the antiferromagnetic component and an adjacent ferromagnet also allows to switch the ferro‐ magnetism by the application of an electric field. Therefore, potential new types of multiplestate memories and spintronic devices controlled by electric and magnetic fields can be developed. One specific demonstration is a multiferroic tunnel junction used as a spin filter device with the potential to control both electrically and magnetically [122,123]. In such a spin filter device, the multiferroic tunnel barrier should be as thin as 1–2 nm thickness. Experiments and theoretical calculations show that a critical thickness of a few unit cells is closely related to the screening properties of the electrodes [124,125]. Another potential application of multiferroics in spintronics is the spin wave devices controlled by electric field. Smolenskii and Chupis theoretically analyzed the excitation of spin waves by an alternating electric field, and of ferroelectric oscillations by a magnetic field [126], and concluded that such excitations should be strongest in ferroelectric ferrimagnets, which could be used to produce magneto‐ electric generators or spin wave amplifiers driven by electric field or current. Up to now, few works on the influence of electric fields on spin waves are reported. Recently, Pimenov et al. first demonstrated such hybrid excitations in multiferroic manganites 125]. It is expected that this field should be a hot topic in near future.

## **6. Future outlook of perovskite nanocrystals**

Perovskite nanopowdes (e.g., BaTiO3) are widely used in MLCCs in the electronic industry due to their excellent dielectric properties. In recent years, miniaturization of ferroelectric BaTiO3 powders to the nanometer scale is very desirable for their applications in the next generation of electronics; however, the main challenge lies in the synthesis of barium titanate nanocrystals at room temperature with a tetragonal crystalline structure, which induces the ferroelectric property. Much effort has been concentrated on synthesizing the high-purity, homogeneous, weakly agglomerated nanocrystals with tetragonal structure. However, it is known that the ferroelectricity becomes weaker with a decrease in particle size and disappears below a certain critical size, known as the size effect. Lots of fundamental and experimental studies are needed to understand the size effect of ferroelectricty for nanosized perovskite powders, to develop future high volume-efficient and high-capacitance MLCCs.

Perovskite oxide nanocrystals also exhibit multiferroic behavior, which is the fundamental of giant ME effects and ME phase control. However, these phenomena are now primarily investigated in the viewpoint of basic research rather than the practical applications. The main reason is that the number of multiferroic compounds with perovskite structure is very small, in many cases use of bulk single crystals is necessary and only very few multiferroics exhibit ME behavior at room temperature. Several microscopic physical mechanisms of ME behav‐ iours in multiferroics have been revealed and the precise criteria searching for new multifer‐ roics have been given out. More and more perovskite multiferroics will be developed and the improvement on tuning the ME performance will drive multiferroics much closer to practical applications in the near future. Now, the major challenge is developing the room-temperature perovskite multiferroics insulators and expanding their applications in the fields of micro‐ electronics.

## **7. Conclusions**

In this chapter, we have reviewed various processing routes, characterization, functionaliza‐ tion, and novel application areas of perovskite oxide nanocrystals. Significant progress has been achieved in the development of processing routes for perovskite oxide nanocrystals, which ranges from the solid-state reaction to chemical solution depositions. By introducing the BaTiO3 nanoparticles into the multilayer dielectrics, much advancement has been achieved in reducing dielectric thickness and increasing the volumetric efficiency of BaTiO3-based MLCCs, to meet the miniaturization demand for electronic devices. At the nanoscale, highly accurate microstructural characterizations are usually required for investigating the micro‐ structures of perovskite nanocrystals. The unique properties of perovskite oxide nanocrystals (e.g., ferroelectric and dielectric, electrical, magnetic, and multiferroic properties) are ad‐ dressed based on the selected recent literature. An important conclusion from the extensive review is that there exists size effects for these unique properties. Better understanding of the size effects in perovskite oxide nanocrystals will be helpful in selecting the critical size and dimension for the purpose of implementing perovskite oxide nanocrystals in various devices. Applications of perovskite oxide nanocrystals have been identified with a major focus in areas such as nanoelectronics, information storage devices, and their potential applications in spintronics are also briefly introduced. As research into perovskite oxide nanocrystals spreads its wings, becoming more extensive, a complete review on this subject has become an arduous task. However, in this chapter a modest attempt is made to analyze recent significant devel‐ opments in researches of perovskite oxide nanocrystals and their possible applications in various industries.

## **Acknowledgements**

and Chupis theoretically analyzed the excitation of spin waves by an alternating electric field, and of ferroelectric oscillations by a magnetic field [126], and concluded that such excitations should be strongest in ferroelectric ferrimagnets, which could be used to produce magneto‐ electric generators or spin wave amplifiers driven by electric field or current. Up to now, few works on the influence of electric fields on spin waves are reported. Recently, Pimenov et al. first demonstrated such hybrid excitations in multiferroic manganites 125]. It is expected that

Perovskite nanopowdes (e.g., BaTiO3) are widely used in MLCCs in the electronic industry due to their excellent dielectric properties. In recent years, miniaturization of ferroelectric BaTiO3 powders to the nanometer scale is very desirable for their applications in the next generation of electronics; however, the main challenge lies in the synthesis of barium titanate nanocrystals at room temperature with a tetragonal crystalline structure, which induces the ferroelectric property. Much effort has been concentrated on synthesizing the high-purity, homogeneous, weakly agglomerated nanocrystals with tetragonal structure. However, it is known that the ferroelectricity becomes weaker with a decrease in particle size and disappears below a certain critical size, known as the size effect. Lots of fundamental and experimental studies are needed to understand the size effect of ferroelectricty for nanosized perovskite

Perovskite oxide nanocrystals also exhibit multiferroic behavior, which is the fundamental of giant ME effects and ME phase control. However, these phenomena are now primarily investigated in the viewpoint of basic research rather than the practical applications. The main reason is that the number of multiferroic compounds with perovskite structure is very small, in many cases use of bulk single crystals is necessary and only very few multiferroics exhibit ME behavior at room temperature. Several microscopic physical mechanisms of ME behav‐ iours in multiferroics have been revealed and the precise criteria searching for new multifer‐ roics have been given out. More and more perovskite multiferroics will be developed and the improvement on tuning the ME performance will drive multiferroics much closer to practical applications in the near future. Now, the major challenge is developing the room-temperature perovskite multiferroics insulators and expanding their applications in the fields of micro‐

In this chapter, we have reviewed various processing routes, characterization, functionaliza‐ tion, and novel application areas of perovskite oxide nanocrystals. Significant progress has been achieved in the development of processing routes for perovskite oxide nanocrystals, which ranges from the solid-state reaction to chemical solution depositions. By introducing

powders, to develop future high volume-efficient and high-capacitance MLCCs.

this field should be a hot topic in near future.

electronics.

**7. Conclusions**

**6. Future outlook of perovskite nanocrystals**

172 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

This work is partially supported by National Natural Science Foundation of China (grant nos. 11174122 and 11134004), National Basic Research Program of China (grant no. 2015CB654900), the open project from National Laboratory of Solid State Microstructures, Nanjing University (grant no. M26012), and six big talent peak project from Jiangsu province (grant no. XCL-004).

## **Author details**

Heng Wu and Xinhua Zhu\*

\*Address all correspondence to: xhzhu@nju.edu.cn

National Laboratory of Solid State of Microstructures, School of Physics, Nanjing University, Nanjing, China

## **References**

[1] Schaak RE, Mallouk TE. Perovskites by design: a toolbox of solid-state reactions. Chem Mater 2002;14(4):1455–71. DOI: 10.1021/cm010689m


[17] Joshi UA, Lee JS. Template-free hydrothermal synthesis of single-crystalline barium titanate and strontium titanate nanowires. Small 2005;1(12):1172–6. DOI: 10.1002/ smll.200500055

[2] Bhalla A, Guo R, Roy R. The perovskite structure–a review of its role in ceramic sci‐ ence and technology. Mater Res Innovat 2000;4(1):3–26. DOI: 10.1007/s100190000062

[3] Dawber M, Rabe K, Scott J. Physics of thin-film ferroelectric oxides. Rev Mod Phys

[4] Gruverman A, Kholkin A. Nanoscale ferroelectrics: processing, characterization and future trends. Rep Prog Phys 2006;69(8):2443–74. DOI: 10.1088/0034-4885/69/8/R04 [5] Hill NA. Why are there so few magnetic ferroelectrics. J Phys Chem B 2000;104(29):

[6] Fiebig M, Lottermoser T, Frohlich D, Goltsev AV, Pisarev RV. Observation of cou‐ pled magnetic and electric domains. Nature 2002;419(6909):818–20. DOI: 10.1038/

[7] Wang J, Neaton JB, Zheng H, Nagarajan V, Ogale SB, Liu B, et al. Epitaxial BiFeO3 multiferroic thin film heterostructures. Science 2003;299(5613):1719–22. DOI: 10.1126/

[8] Hur N, Park S, Sharma PA, Ahn JS, Guha S, Cheong SW. Electric polarization rever‐ sal and memory in a multiferroic material induced by magnetic fields. Nature

[9] Haertling GH. Ferroelectric ceramics: history and technology. J Am Ceramic Soc

[10] Moulson AJ, Herbert JM. Electroceramics: Materials, Properties, Applications. 1st ed. New Jersey, USA: John Wiley & Sons; 2003. 557 p. DOI: 10.1002/0470867965

[11] Yoon DH, Lee BI. BaTiO3 properties and powder characteristics for ceramic capaci‐

[12] Yoon D. Tetragonality of barium titanate powder for a ceramic capacitor application.

[13] Pithan C, Hennings D, Waser R. Progress in the synthesis of nanocrystalline BaTiO3 powders for MLCC. Int J Appl Ceramic Technol 2005;2(1):1–14. DOI: 10.1111/j.

[14] Chu MS, Rae AW. Manufacturing dielectric powders. Am Ceramic Soc Bull

[15] Hu YH, Harmer MP, Smyth DM. Solubility of BaO in BaTiO3. J Am Ceramic Soc

[16] Phule P, Risbud S. Low-temperature synthesis and processing of electronic materials in the BaO-TiO2 system. J Mater Sci 1990;25(2):1169–83. DOI: 10.1007/BF00585422

2005;77(4):1083–130. DOI: 10.1103/RevModPhys.77.1083

6694–709. DOI: 10.1021/jp000114x

174 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

2004;429(2990):392–5. DOI: 10.1038/nature02572

tors. J Ceramic Process Res 2002;3(2):41–7.

J Ceramic Process Res 2006;7(4):343.

1744-7402.2005.02008.x

1995;74(6):69–72.

1999;82(4):797–818. DOI: 10.1111/j.1151-2916.1999.tb01840.x

1985;68(7):372–6. DOI: 10.1111/j.1151-2916.1985.tb10145.x

nature01077

science.1080615


[43] Clark I, Takeuchi T, Ohtori N,C, Sinclair D. Hydrothermal synthesis and characteri‐ sation of BaTiO3 fine powders: precursors, polymorphism and properties. J Mater Chem 1999;9(1):83–91. DOI: 10.1039/A805756G

[30] Thirumal M, Jain P, Ganguli AK. Molten salt synthesis of complex perovskite-related dielectric oxides. Mater Chem Phys 2001;70(1):7–11. DOI: 10.1016/

[31] Chen J, Xing X, Watson A, Wang W, Yu R, Deng J, et al. Rapid synthesis of multifer‐ roic BiFeO3 single-crystalline nanostructures. Chem Mater 2007;19(15):3598–600. DOI:

[32] Chen J, Yu R, Li L, Sun C, Zhang T, Chen H, et al. Structure and shape evolution of Bi1–xLaxFeO3 perovskite microcrystals by molten salt synthesis. Eur J Inorg Chem

[33] He X, Gao L. Synthesis of pure phase BiFeO3 powders in molten alkali metal nitrates.

[34] Zheng XH, Chen PJ, Ma N, Ma ZH, Tang DP. Synthesis and dielectric properties of BiFeO3 derived from molten salt method. J Mater Sci: Mater Electronics 2012;23(5):

[35] Zhu X, Zhou J, Jiang M, Xie J, Liang S, Li S, et al. Molten salt synthesis of bismuth ferrite nano- and microcrystals and their structural characterization. J Am Ceramic

[36] Frey MH, Payne DA. Synthesis and processing of barium titanate ceramics from alk‐ oxide solutions and monolithic gels. Chem Mater 1995;7(1):123–9. DOI: 10.1021/

[37] Hwang U-Y, Park H-S, Koo K-K. Low-temperature synthesis of fully crystallized spherical BaTiO3 particles by the gel–sol method. J Am Ceramic Soc 2004;87(12):2168–

[38] Brutchey RL, Morse DE. Template-free, low-temperature synthesis of crystalline. An‐

[39] Kiss K, Magder J, Vukasovich MS, Lockhart RJ. Ferroelectrics of ultrafine particle size: I, synthesis of titanate powders of ultrafine particle size. J Am Ceramic Soc

[40] Yoon S, Baik S, Kim MG, Shin N. Formation mechanisms of tetragonal barium tita‐ nate nanoparticles in alkoxide–hydroxide sol-precipitation synthesis. J Am Ceramic

[41] Yoon S, Baik S, Kim MG, Shin N, Kim I. Synthesis of tetragonal barium titanate nano‐ particles via alkoxide–hydroxide sol-precipitation: effect of water addition. J Am Ce‐

[42] Dutta PK, Gregg JR. Hydrothermal synthesis of tetragonal barium titanate (BaTiO3).

gewandte Chemie 2006;118(39):6714–6. DOI: 10.1002/ange.200602571

1966;49(6):291–5. DOI: 10.1111/j.1151-2916.1966.tb13265.x

Soc 2006;89(6):1816–21. DOI: 10.1111/j.1551-2916.2006.01056.x

ramic Soc 2007;90(1):311–4. DOI: 10.1111/j.1551-2916.2006.01361.x

Chem Mater 1992;4(4):843–6. DOI: 10.1021/cm00022a019

Ceramics Int 2009;35(3):975–8. DOI: 10.1016/j.ceramint.2008.04.013

2008;2008(23):3655–60. DOI: 10.1002/ejic.200800263

176 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

990–4. DOI: 10.1007/s10854-011-0533-4

Soc 2014;97(7):2223–32. DOI: 10.1111/jace.12897

74. DOI: 10.1111/j.1151-2916.2004.tb07486.x

S0254-0584(00)00384-9

10.1021/cm070790c

cm00049a019


[69] Jhung SH, Lee J-H, Yoon JW, Hwang YK, Hwang J-S, Park S-E, et al. Effects of reac‐ tion conditions in microwave synthesis of nanocrystalline barium titanate. Mater Lett 2004;58(25):3161–5. DOI: 10.1016/j.matlet.2004.06.006

[56] Liu C, Zou B, Rondinone AJ, Zhang ZJ. Sol−gel synthesis of free-standing ferroelec‐ tric lead zirconate titanate nanoparticles. J Am Chem Soc 2001;123(18):4344–5. DOI:

[57] Pinceloup P, Courtois C, Vicens J, Leriche A, Thierry B. Evidence of a dissolution– precipitation mechanism in hydrothermal synthesis of barium titanate powders. J

[58] Hertl W. Kinetics of barium titanate synthesis. J Am Ceramic Soc 1988;71(10):879–83.

[59] Hu MZC, Kurian V, Payzant EA, Rawn CJ, Hunt RD. Wet-chemical synthesis of monodispersed barium titanate particles -hydrothermal conversion of TiO2 micro‐ spheres to nanocrystalline BaTiO3. Powder Technol 2000;110(1–2):2–14. DOI: 10.1016/

[60] Eckert JO, Hung-Houston CC, Gersten BL, Lencka MM, Riman RE. Kinetics and mechanisms of hydrothermal synthesis of barium titanate. J Am Ceramic Soc

[61] Gao F, Chen XY, Yin KB, Dong S, Ren ZF, Yuan F, et al. Visible-light photocatalytic properties of weak magnetic BiFeO3 nanoparticles. Adv Mater 2007;19(19):2889–92.

[62] Walton RI, Millange F, Smith RI, Hansen TC, O'Hare D. Real time observation of the hydrothermal crystallization of barium titanate using in situ neutron powder diffrac‐

[63] Komarneni S, Roy R, Li QH. Microwave-hydrothermal synthesis of ceramic powders.

[64] A. Katsuki H, Komarneni S. Microwave-hydrothermal synthesis of monodispersed nanophase α-Fe2O3. J Am Ceramic Soc 2001;84(10):2313–7. DOI: 10.1111/j.

[65] Komarneni S, Li Q, Stefansson KM, Roy R. Microwave-hydrothermal processing for synthesis of electroceramic powders. J Mater Res. 1993;8(12):3176–83. DOI: 10.1557/

[66] Komarneni S, Li QH, Roy R. Microwave-hydrothermal processing for synthesis of layered and network phosphates. J Mater Chem 1994;4(12):1903–6. DOI: 10.1039/

[67] Komarneni S, Li QH, Roy R. Microwave-hydrothermal processing of layered anion

[68] Liu S-F, Abothu IR, Komarneni S. Barium titanate ceramics prepared from conven‐ tional and microwave hydrothermal powders. Mater Lett 1999;38(5):344–350. DOI:

exchangers. J Mater Res 1996;11(8):1866–9. DOI: 10.1557/JMR.1996.0236

tion. J Am Chem Soc 2001;123(50):12547–55. DOI: 10.1021/ja011805p

Mater Res Bull 1992;27(12):1393–405. DOI: 10.1016/0025-5408(92)90004-J

1996;79(11):2929–39. DOI: 10.1111/j.1151-2916.1996.tb08728.x

Eur Ceramic Soc 1999;19(6–7):973–7. DOI: 10.1016/S0955-2219(98)00356-2

10.1021/ja001893y

S0032-5910(99)00262-4

DOI: 10.1002/adma.200602377

1151-2916.2001.tb01007.x

10.1016/S0167-577X(98)00187-6

JMR.1993.3176

JM9940401903

DOI: 10.1111/j.1151-2916.1988.tb07540.x

178 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications


tron microscopy. J Crystal Growth 2008;310(2):434–41. DOI: 10.1016/j.jcrysgro. 2007.10.076


[95] Markovic D, Kusigerski V, Tadic M, Blanusa J, Antisari MV, Spasojevic V. Magnetic properties of nanoparticle La0.7Ca0.3MnO3 prepared by glycine–nitrate method with‐ out additional heat treatment. Scripta Materialia 2008;59(1):35–8. DOI: 10.1016/j.scrip‐ tamat.2008.02.020

tron microscopy. J Crystal Growth 2008;310(2):434–41. DOI: 10.1016/j.jcrysgro.

[82] Wang ZL, Cowley JM. Simulating high-angle annular dark-field stem images includ‐ ing inelastic thermal diffuse scattering. Ultramicroscopy 1989;31(4):437–53. DOI:

[83] Pennycook SJ, Jesson DE. High-resolution incoherent imaging of crystals. Phys Rev

[84] Zhu X, Wang J, Zhang Z, Zhu J, Zhou S, Liu Z, et al. Atomic-scale characterization of barium titanate powders formed by the hydrothermal process. J Am Ceramic Soc

[85] Zhu X, Zhu J, Zhou S, Liu Z, Ming N, Hesse D. BaTiO3 nanocrystals: hydrothermal synthesis and structural characterization. J Crystal Growth 2005;283(3–4):553–62.

[86] Zhu X, Zhou J, Zhu J, Liu Z, Li Y, Al-Kassab T. Structural characterization and opti‐ cal properties of perovskite ZnZrO3 nanoparticles. J Am Ceramic Soc 2014;97(6):

[87] Hartman P, Perdok WG. On the relations between structure and morphology of crys‐ tals. I. Acta Crystallograph 1955;8(1):49–52. DOI: 10.1107/S0365110X55000121

[88] Jiang B, Peng JL, Bursill LA, Ren TL, Zhang PL, Zhong WL. Defect structure and physical properties of barium titanate ultra-fine particles. Phys B: Condens Matt

[89] Jiang B, Peng JL, Bursill LA. Intergrowth defects and dielectric response of BaTiO3 nano-particles. Ferroelectrics 1998;207(1):587–610. DOI: 10.1080/00150199808217271

[90] Rose H. Outline of a spherically corrected semiaplanatic medium-voltage transmis‐

[91] Urban KW. Studying atomic structures by aberration-corrected transmission electron

[92] Jia C-L, Mi S-B, Urban K, Vrejoiu I, Alexe M, Hesse D. Atomic-scale study of electric dipoles near charged and uncharged domain walls in ferroelectric films. Nature Ma‐

[93] Huang L, Chen Z, Wilson JD, Banerjee S, Robinson RD, Herman IP, et al. Barium tita‐ nate nanocrystals and nanocrystal thin films: synthesis, ferroelectricity, and dielectric

[94] Nuraje N, Su K, Haboosheh A, Samson J, Manning EP, Yang Nl, et al. Room temper‐ ature synthesis of ferroelectric barium titanate nanoparticles using peptide nanorings

as templates. Adv Mater 2006;18(6):807–11. DOI: 10.1002/adma.200501340

microscopy. Science 2008;321(5888):506–10. DOI: 10.1126/science.1152800

properties. J Appl Phys 2006;100(3):034316. DOI: 10.1063/1.2218765

Lett 1990;64(8):938–41. DOI: 10.1103/PhysRevLett.64.938

2008;91(3):1002–8. DOI: 10.1111/j.1551-2916.2007.02227.x

2000;291(1–2):203–12. DOI: 10.1016/S0921-4526(99)01874-8

sion electron-microscope. Optik 1990;85(1):19–24.

ter 2008;7(1):57–61. DOI: 10.1038/nmat2080

2007.10.076

10.1016/0304-3991(89)90340-9

180 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

DOI: 10.1016/j.jcrysgro.2005.05.080

1987–92. DOI: 10.1111/jace.12883


[121] Kimura T, Kawamoto S, Yamada I, Azuma M, Takano M, Tokura Y. Magnetocapaci‐ tance effect in multiferroic BiMnO3. Phys Rev B 2003;67(18):180401. DOI: 10.1103/ PhysRevB.67.180401

[108] Yu J, Chu J. Progress and prospect for high temperature single-phased magnetic fer‐ roelectrics. Chin Sci Bull 2008;53(14):2097–112. DOI: 10.1007/s11434-008-0308-3

[109] Setter N, Damjanovic D, Eng L, Fox G, Gevorgian S, Hong S, et al. Ferroelectric thin films: review of materials, properties, and applications. J Appl Phys 2006;100(5):

[110] Kohlstedt H, Ishiwara H. Nanoelectronics and Information Technology - Advanced Electronic Materials and Novel Devices. 1st ed. Wienheim, Germany: WILEY-VCH

[111] Martinelli G, Carotta MC, Traversa E, Ghiotti G. Thick-film gas sensors based on nano-sized semiconducting oxide powders. MRS Bull 1999;24(06):30–6. DOI: 10.1557/

[112] Mathews S, Ramesh R, Venkatesan T, Benedetto J. Ferroelectric field effect transistor based on epitaxial perovskite heterostructures. Science 1997;276(5310):238–40. DOI:

[113] Tasch AF, Jr, Parker LH. Memory cell and technology issues for 64- and 256-Mbit one-transistor cell MOSD DRAMs. Proc IEEE 1989;77(3):374–88. DOI: 10.1109/5.24125

[114] Scott J. High-dielectric constant thin films for dynamic random access memories (DRAM). Annu Rev Mater Sci 1998;28(1):79–100. DOI: 10.1146/annurev.matsci.28.1.79

[115] Hwang CS, Park SO, Cho HJ, Kang CS, Kang HK, Lee SI, et al. Deposition of ex‐ tremely thin (Ba, Sr) TiO3 thin films for ultra‐large‐scale integrated dynamic random access memory application. Appl Phys Lett 1995;67(19):2819–21. DOI:

[116] Horikawa T, Mikami N, Makita T, Tanimura J, Kataoka M, Sato K et al. Dielectric properties of (Ba, Sr)TiO3 thin films deposited by RF sputtering. Jpn J Appl Phys

[117] Yoshino K, Shimoda Y, Kawagishi Y, Nakayama K, Ozaki M. Temperature tuning of the stop band in transmission spectra of liquid-crystal infiltrated synthetic opal as tunable photonic crystal. Appl Phys Lett 1999;75(7):932–4. DOI: 10.1063/1.124558

[118] Martí X, Sánchez F, Hrabovsky D, Fàbrega L, Ruyter A, Fontcuberta J, et al. Exchange biasing and electric polarization with YMnO3. Appl Phys Lett 2006;89(3):032510. DOI:

[119] Béa H, Bibes M, Cherifi S, Nolting F, Warot-Fonrose B, Fusil S, et al. Tunnel magneto‐ resistance and robust room temperature exchange bias with multiferroic BiFeO3 epi‐

[120] Žutić I, Fabian J, Das Sarma S. Spintronics: fundamentals and applications. Rev Mod

taxial thin films. Appl Phys Lett 2006;89(24):242114. DOI: 10.1063/1.2402204

Phys 2004;76(2):323–410. DOI: 10.1103/RevModPhys.76.323

1993;32(9S):4126–30. DOI: 10.1143/JJAP.32.4126

051606. DOI: 10.1063/1.2336999

182 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Verlag; 2003. 1002 p.

S0883769400052489

10.1063/1.114795

10.1063/1.2234285

10.1126/science.276.5310.238


## **Synthesis, Crystal Structure, and Physical Properties of the Perovskite Iridates**

Yunqi Cai, Yan Li and Jinguang Cheng

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61281

## **Abstract**

Perovskite iridates have emerged as a new paradigm for studying the strongly correlated electron physics with strong spin-orbit coupling. The "113" alkaline-earth iridates AIrO3 (A = Ca, Sr, Ba) display a rich variety of crystallographic and electronic states and are now attracting growing research interest. This chapter aims to provide an overview for these "113" iridates, including the materials' synthesis, crystal structure, major physical properties, and other interesting results such as the effects of pressure and chemical sub‐ stitutions, as well as theoretical perspectives.

**Keywords:** Perovskite iridates, Spin-orbit coupling, Post-perovskite, Polytype, Semimetal

## **1. Introduction**

The discoveries of high-transition-temperature superconductivity in cuprates and the colossal magnetoresistance in manganites made the first-row (3d) transition-metal oxides (TMOs) with perovskite-related structures the central topics of condensed matter physics over the past four decades. The strong electron–electron correlations intrinsic for these narrow-band 3d-electron systems are believed to be at the heart of rich physics. Following the general wisdom based on the 3d TMOs, the third-row (5d) counterparts having a spatially much extended 5d orbitals were expected to have much reduced electron–electron correlations, U, and broaden band‐ width, W, i.e. U << W, leading to a Pauli paramagnetic metallic ground state, Figure 1(a). Such an expectation, however, was recently found to be violated in many 5d-electron iridium oxides (iridates), such as Sr2IrO4 [1], in which an antiferromagnetic insulating ground state was instead observed. Recent studies have revealed that such discrepancy originates from the inherently strong spin-orbit coupling (SOC) for these heavy 5d elements, which have a typical

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

value of SOC, ζSO ≈ 0.3–0.5 eV, comparable with the magnitude of U and W, and thus cannot be treated as a negligible perturbation as in the 3d TMOs.

Since an unrealistically large U is required to open a Mott gap in Sr2IrO4, Figure 1(b), Kim et al. [2] proposed that the strong SOC splits the otherwise broad t2g band of the octahedral-site, low-spin Ir4+(5d 5) array into a filled, low-energy Jeff = 3/2 quartet band and a half-filled, highenergy Jeff = 1/2 doublet band, Figure 1(c, e). A moderate Hubbard U can then open a Mott gap, leading to the SOC-driven Jeff = 1/2 Mott insulating state, Figure 1(d). Subsequent experimental [3] and theoretical [4] investigations have confirmed such a novel Jeff = 1/2 state in the strong SOC limit. Since then, the 5d TMOs have emerged as a new paradigm for studying the strongly correlated electron physics with strong SOC. In particular, the iridates have attracted special attention in that the combination of relativistic SOC and electron–electron correlations has been proposed to generate more exotic, unprecedented quantum states of matters, such as the strong topological insulators, Weyl semimetal, quantum spin liquids, and even unconventional superconductors [5].

**Figure 1.** Schematic energy diagrams for the 5d5 (t2g5 ) configuration: (a) without SOC and U, (b) with an unrealistically large U but no SOC, (c) with SOC but no U, and (d) with SOC and U, (e) 5d level splitting by the crystal field and SOC. Adapted from Reference [2].

Since the importance of SOC was first recognized in Sr2IrO4, which is the *n* = 1 member of the Ruddlesden–Popper series Srn+1IrnO3n+1 (*n* = 1, 2, ∞), these perovskite (Pv) iridates are currently archetypal systems for studying the interplay of SOC and electron–electron correlations. During the past few years, numerous studies have been performed on the single-layer Sr2IrO4 (*n* = 1) and bilayer Sr3Ir2O7 (*n* = 2); for a review, see Reference [6]. With increasing the number of Pv layers *n*, the bandwidth of the Jeff = 1/2 band is expected to increase, and a dimensionality controlled insulator-to-metal transition is eventually realized in the ortho‐ rhombic Pv SrIrO3 (*n* = ∞) [7]. Recent advances in this emergent field have turned much attention to the "113" alkaline-earth iridates AIrO3 (A = Ca, Sr, Ba) with the Pv-related structures. However, in-depth investigations on these compounds are hindered to a great extent by the harsh synthesis conditions as well as their complex structural variations. For example, the above-mentioned orthorhombic Pv SrIrO3 can only be stabilized in the bulk form under high-pressure and high-temperature (HPHT) conditions [8], or in the form of thin films by applying epitaxial strain [9]. Recent theoretical investigations proposed an intriguing topological semimetal state for the orthorhombic Pv phase [10]. When synthesized at ambient pressure, on the other hand, SrIrO3 adopts a so-called six-layer (6H) polytype [8], which has been characterized as a non-Fermi-liquid metal approaching a ferromagnetic quantum critical point [11]. As the sister compounds of SrIrO3, both CaIrO3 and BaIrO3 also display multiple structural polymorphs with interesting structural–property relationships. Depending on the synthesis conditions, CaIrO3 can be stabilized in either the Pv or the post-perovskite (pPv) structure, which displays, respectively, a paramagnetic metal and an antiferromagnetic insulator ground states [12, 13]. Although the pPv CaIrO3 was regarded as the Jeff = 1/2 Mott insulator [14], recent studies revealed a clear deviation from the ideal Jeff = 1/2 state due to the pronounced structural distortions [15]. In addition, in the field of geosciences, CaIrO3 has been studied extensively as an analogy of MgSiO3 to elucidate the mechanism of Pv to pPv transition at the boundary of Earth's lowermost mantle, or the D'' layer [16]. In the case of BaIrO3, it also exhibits multiple structural polytypes with interesting structural–property relationship. At ambient pressure, BaIrO3 adopts a nine-layer structure (9R). It is the first known ferromagnetic insulator among the 5d TMOs and exhibits intriguing charge-density-wave formation accompanying the ferromagnetic order [17, 18]. When treating the 9R phase under different pressures, three more polytypes, i.e. 5H, 6H, and 3C, have been identified [19]. Following the sequence of 9R → 5H → 6H → 3C, their ground states change progressively from a weak ferromagnetic insulator with *T*<sup>c</sup> = 180 K for 9R, through a ferromagnetic metal with *T*<sup>c</sup> = 50 K for 5H, and an exchange enhanced non-Fermi-liquid metal for 6H approaching a ferromagnetic quantum critical point, finally to a Pauli paramagnetic Fermi-liquid metal for 3C [20, 21]. These results demonstrate an intimate structure–property relationship that has been well document‐ ed in the 3d TMOs. A brief summary of the structural types and interesting physical properties for these "113" alkaline-earth iridates AIrO3 (A = Ca, Sr, Ba) is given in Table 1.

value of SOC, ζSO ≈ 0.3–0.5 eV, comparable with the magnitude of U and W, and thus cannot

Since an unrealistically large U is required to open a Mott gap in Sr2IrO4, Figure 1(b), Kim et al. [2] proposed that the strong SOC splits the otherwise broad t2g band of the octahedral-site, low-spin Ir4+(5d 5) array into a filled, low-energy Jeff = 3/2 quartet band and a half-filled, highenergy Jeff = 1/2 doublet band, Figure 1(c, e). A moderate Hubbard U can then open a Mott gap, leading to the SOC-driven Jeff = 1/2 Mott insulating state, Figure 1(d). Subsequent experimental [3] and theoretical [4] investigations have confirmed such a novel Jeff = 1/2 state in the strong SOC limit. Since then, the 5d TMOs have emerged as a new paradigm for studying the strongly correlated electron physics with strong SOC. In particular, the iridates have attracted special attention in that the combination of relativistic SOC and electron–electron correlations has been proposed to generate more exotic, unprecedented quantum states of matters, such as the strong topological insulators, Weyl semimetal, quantum spin liquids, and even unconventional

be treated as a negligible perturbation as in the 3d TMOs.

186 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

superconductors [5].

**Figure 1.** Schematic energy diagrams for the 5d5

Adapted from Reference [2].

(t2g5

large U but no SOC, (c) with SOC but no U, and (d) with SOC and U, (e) 5d level splitting by the crystal field and SOC.

) configuration: (a) without SOC and U, (b) with an unrealistically

Although there are many publications dealing with an individual compound, a monograph that provides a comprehensive overview for these "113" alkaline-earth iridates is still lacking to our knowledge. Taking into account the growing research interests on these iridates, it is imminent to summarize the currently available knowledge in a single chapter. Thus, this chapter aims to bring together the available information in literature for these "113" iridates. In the following, we will give a comprehensive literature survey for each AIrO3, covering the materials' synthesis, crystal structure, and major physical properties, as well as other inter‐ esting results such as the effects of chemical substitutions and theoretical investigations. Finally, we will give a brief concluding remark on the current research status and provide an outlook on the future research directions on these iridates.


AF: Antiferromagnetic; PM: Paramagnetic; FM: Ferromagnetic; FL: Fermi liquid;

nFL: non Fermi liquid; CDW: Charge density wave; QCP: Quantum critical point

**Table 1.** A summary of the "113" Alkaline-earth iridates AIrO3 (A=Ca, Sr, Ba)

## **2. CaIrO3**

CaIrO3 has two different orthorhombic polymorphs, i.e. the layered pPv structure with space group *Cmcm* and the GdFeO3-type Pv structure with space group *Pbnm*. These two compounds have been known since 1960s [22, 23] and received significant attention from geologists since 2004 as an analogy of MgSiO3, the main constituent mineral of the Earth's lower mantle [16, 24]. More recently, they have emerged as important correlated 5delectron systems with strong SOC [14, 25]; the strong local distortion in pPv CaIrO3 makes it a model system to investigate the interplay of non-cubic crystal field splitting and SOC [15], while the orthorhombic Pv CaIrO3 might be considered as an intriguing semimetal with symmetry-protected Dirac points [26].

## **2.1. Synthesis**

There are some discrepancies in literature regarding the synthesis of pPv CaIrO3 at ambient pressure. In the earlier studies [12, 22], it was reported that single-phase pPv phase cannot be obtained at ambient pressure through a solid-state reaction from CaCO3 and IrO2 in air. Recently, Harai et al. [27] reported that pure pPv CaIrO3 can be prepared by heating the stoichiometric mixture of CaO and IrO2 powders sealed in an evacuated silica tube at 1000°C over 20 h. On the other hand, since the pPv structure is a high-pressure phase, pPv CaIrO3 can be readily obtained by utilizing HPHT synthesis. For example, Ohgushi et al. [25] reported the synthesis of single-phase pPv CaIrO3 at 4 GPa and 1150°C.

Needle-shaped pPv CaIrO3 single crystals have been reported to grow out of the CaCl2 flux. By adopting a tenfold flux and a relatively low soaking temperature of 836 and 950°C, respectively, Sugahara et al. [28] and Hirai et al. [29] obtained tiny single crystals for the purpose of crystal-structure refinements. On the other hand, Ohgushi et al. [14] seems to grow sizable pPv CaIrO3 single crystals for anisotropic magnetic property measurements by employing a higher flux molar ratio (16:1) and a higher soaking temperature of 1200°C. However, our attempts by using the latter approach ended up with Ca2IrO4 rather than the pPv CaIrO3.

Because Pv CaIrO3 is a metastable phase, it cannot be synthesized via a solid-state reaction route at ambient pressure. Alternatively, Sarkozy et al. [12] reported the preparation of pure Pv phase by thermal decomposition at 650–700°C in air of the hydroxide intermediate CaIr(OH)6, which can be obtained according to the following wet-chemical reaction scheme:

$$\text{K}\_2\text{IrCl}\_6 \xrightarrow{H\_2O} \text{IrCl}\_6^{2-} \xrightarrow{kOH, pH 10-12} \text{Ir}\text{(OH)}\_6^{2-} \xrightarrow{\text{Ca}^{2+}} \text{CaIr}\text{(OH)}\_6\downarrow\text{.}$$

By following this approach, we obtained nearly single-phase Pv CaIrO3 with a trace amount of IrO2 (0.2 wt.%) and Ca2IrO4 (1.3 wt.%) [30]. Recently, Kojitani et al. [31] determined a large positive Clapeyron slope for the pPv/Pv transition of CaIrO3, i.e. Pv structure is the hightemperature phase of pPv. Thus, Pv CaIrO3 can be obtained by transforming pPv phase at higher temperature under given pressures. For example, Ohgushi et al. [13] have reported the synthesis of single-phase Pv CaIrO3 at 1 GPa and 1450°C. In addition, thin films of Pv CaIrO3 have recently been epitaxially stabilized on various substrates [26, 32].

## **2.2. Crystal structure**

In the following, we will give a comprehensive literature survey for each AIrO3, covering the materials' synthesis, crystal structure, and major physical properties, as well as other inter‐ esting results such as the effects of chemical substitutions and theoretical investigations. Finally, we will give a brief concluding remark on the current research status and provide an

> pPv AF insulator with TN = 110 K, stripe-type AF order with spin canting; Pv PM semimetal with possible Dirac node protected by reflection symmetry

6H Exchange enhanced PM metal with nFL behaviors due to proximity to a FM QCP

6H Exchange enhanced PM metal with nFL behaviors due to proximity to a FM QCP

Pv PM semimetal with possible Dirac node protected by reflection symmetry

9R Weak FM insulator with a simultaneous CDW formation below Tc ≈ 180 K

CaIrO3 has two different orthorhombic polymorphs, i.e. the layered pPv structure with space group *Cmcm* and the GdFeO3-type Pv structure with space group *Pbnm*. These two compounds have been known since 1960s [22, 23] and received significant attention from geologists since 2004 as an analogy of MgSiO3, the main constituent mineral of the Earth's lower mantle [16, 24]. More recently, they have emerged as important correlated 5delectron systems with strong SOC [14, 25]; the strong local distortion in pPv CaIrO3 makes it a model system to investigate the interplay of non-cubic crystal field splitting and SOC [15], while the orthorhombic Pv CaIrO3 might be considered as an intriguing semimetal

There are some discrepancies in literature regarding the synthesis of pPv CaIrO3 at ambient pressure. In the earlier studies [12, 22], it was reported that single-phase pPv phase cannot be obtained at ambient pressure through a solid-state reaction from CaCO3 and IrO2 in air. Recently, Harai et al. [27] reported that pure pPv CaIrO3 can be prepared by heating the stoichiometric mixture of CaO and IrO2 powders sealed in an evacuated silica tube at 1000°C

outlook on the future research directions on these iridates.

188 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

CaIrO3

SrIrO3

BaIrO3

**2. CaIrO3**

**2.1. Synthesis**

**Compound Structure type Interesting physical properties**

5H Weak FM metal with Tc ≈ 50 K

3C FL PM metal

AF: Antiferromagnetic; PM: Paramagnetic; FM: Ferromagnetic; FL: Fermi liquid; nFL: non Fermi liquid; CDW: Charge density wave; QCP: Quantum critical point

**Table 1.** A summary of the "113" Alkaline-earth iridates AIrO3 (A=Ca, Sr, Ba)

with symmetry-protected Dirac points [26].

*pPv CaIrO*3 As shown in Fig. 2(a), the crystal structure of pPv CaIrO3 can be described as a pseudo-2D layered structure having IrO6 octahedral sheets stacked alternatively with the Ca2+ ions along the crystallographic *b* axis. Within the octahedral sheets, IrO6 octahedra share edges along the *a* axis to form rutile-type chains; these chains are then interconnected with each other via apical oxygen atoms along the *c* axis. Because of the significant interest in geosciences, the crystal structure of pPv CaIrO3 have been thoroughly studied by several groups [28, 29]. To illustrate the peculiar features of pPv CaIrO3, here we adopted the results given by Sugahara *et al.* [28] who refined the crystal structure based on the single-crystal Xray diffraction (XRD) technique. The crystal structure was refined in an orthorhombic space group *Cmcm* (No. 63) with Ca at *4c* (0, y, 1/4), Ir at *4a* (0, 0, 0), O1 at *4c* (0, y, 1/4), and O2 at *8f* (0, y, z) sites, respectively. The obtained unit cell parameters are *a* = 3.147 Å, *b* = 9.866 Å, *c* = 7.302 Å, and *V* = 226.7 Å3 at room temperature. The refined positional parameters and selected bond lengths and bond angles after Reference [28] are listed in Table 2. Within the buckled octahedral layer, the IrO6 octahedral chains display an alternative rotation about the *a* axis, resulting in Ir-O1-Ir bond angles of 134.3°. As a result, for a given octahedron the local *z* axis that is along the Ir-O1 bond deviates from the crystallographic *c* axis by about 23°. In addition, IrO6 octahedra show a significant tetragonal compression, with two short Ir-O1 (1.978 Å) and four long Ir-O2 (2.066 Å) bonds. Octahedral-site distortions can be generally described by the orthorhombic vibrational modes Q2 = *l*<sup>x</sup> *– l*y and Q3 = (2*l*z*–l*<sup>x</sup> *–l*y)/√3, where *l*x*, l*y*, l*z are the bond lengths for bonding along local *x, y, z* directions. In pPv CaIrO3, the octahedral-site distortion corresponds to a negative mode of Q3 = –0.102. For comparison, the PtO6 octahedra in the pPv CaPtO3 with a filled t2g6 manifold show a negligible Q3 = –0.001. Such a comparison highlights a strong orbital-lattice coupling in pPv CaIrO3 with low-spin t2g5 configuration for Ir4+ ions, for which the single hole would be expected to have a dominant *yz ± izx* orbital character [30], which has been confirmed recently by the resonant inelastic X-ray spectroscopy [15].

**Figure 2.** Crystal structure of CaIrO3 polymorphs: (a) pPv and (b) Pv.

*Pv CaIrO3* As shown in Fig. 2(b), the crystal structure of Pv CaIrO3 is built up from cornershared IrO6 octahedra in three dimensions with Ca cations in the interstitial positions. It has been known over 40 years that Pv CaIrO3 adopts the GdFeO3-type structure; however, structural refinements have not been performed until recently. We present here our Rietveld refinement results [30] based on the high-resolution synchrotron XRD on polycrystalline Pv CaIrO3 prepared with the wet-chemical method mentioned above. The crystal structure was refined in space group *Pbnm* (No. 62) with Ca at *4c* (*x*, *y*, 1/4), Ir at *4b* (0.5, 0, 0), O1 at *4c* (*x*, *y*, 1/4) and O2 at *8d* (*x*, *y*, *z*) sites, respectively. The lattice parameters at room temperature are determined as *a* = 5.35046 Å, *b* = 5.59291 Å, *c* = 7.67694 Å, and *V* = 229.73 Å3 . The obtained positional parameters and selected bond lengths and bond angles after Reference [30] are listed in Table 3. In comparison with the pPv phase, the IrO6 octahedra are less distorted with three sets of Ir-O distances of 2.006 Å, 2.020 Å, and 2.038 Å; the average Ir-O distance of 2.021 Å is


consistent with the ionic radii sum for Ir4+ (0.625 Å) and O2– (1.40 Å). The averaged Ir-O-Ir bond angle is about 145.5° in the Pv phase.

**Table 2.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for pPv CaIrO3 from singlecrystal XRD [28]: space group *Cmcm* (No. 63), *a* = 3.1472 Å, *b* = 9.8655 Å, *c* = 7.3018 Å, *V* = 226.71 Å3 , Z = 4.


**Table 3.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for Pv CaIrO3 from powder XRD[30]: space group *Pbnm* (No. 62), *a* = 5.35046 Å, *b* = 5.59291 Å, *c* = 7.67694 Å, *V* = 229.73 Å3 , Z = 4.

## **2.3. Physical properties**

resulting in Ir-O1-Ir bond angles of 134.3°. As a result, for a given octahedron the local *z* axis that is along the Ir-O1 bond deviates from the crystallographic *c* axis by about 23°. In addition, IrO6 octahedra show a significant tetragonal compression, with two short Ir-O1 (1.978 Å) and four long Ir-O2 (2.066 Å) bonds. Octahedral-site distortions can be generally described by the orthorhombic vibrational modes Q2 = *l*<sup>x</sup> *– l*y and Q3 = (2*l*z*–l*<sup>x</sup> *–l*y)/√3, where *l*x*, l*y*, l*z are the bond lengths for bonding along local *x, y, z* directions. In pPv CaIrO3, the octahedral-site distortion corresponds to a negative mode of Q3 = –0.102. For comparison, the PtO6 octahedra in the pPv CaPtO3 with a filled t2g6 manifold show a negligible Q3 = –0.001. Such a comparison highlights

which the single hole would be expected to have a dominant *yz ± izx* orbital character [30],

*Pv CaIrO3* As shown in Fig. 2(b), the crystal structure of Pv CaIrO3 is built up from cornershared IrO6 octahedra in three dimensions with Ca cations in the interstitial positions. It has been known over 40 years that Pv CaIrO3 adopts the GdFeO3-type structure; however, structural refinements have not been performed until recently. We present here our Rietveld refinement results [30] based on the high-resolution synchrotron XRD on polycrystalline Pv CaIrO3 prepared with the wet-chemical method mentioned above. The crystal structure was refined in space group *Pbnm* (No. 62) with Ca at *4c* (*x*, *y*, 1/4), Ir at *4b* (0.5, 0, 0), O1 at *4c* (*x*, *y*, 1/4) and O2 at *8d* (*x*, *y*, *z*) sites, respectively. The lattice parameters at room temperature are

positional parameters and selected bond lengths and bond angles after Reference [30] are listed in Table 3. In comparison with the pPv phase, the IrO6 octahedra are less distorted with three sets of Ir-O distances of 2.006 Å, 2.020 Å, and 2.038 Å; the average Ir-O distance of 2.021 Å is

determined as *a* = 5.35046 Å, *b* = 5.59291 Å, *c* = 7.67694 Å, and *V* = 229.73 Å3

which has been confirmed recently by the resonant inelastic X-ray spectroscopy [15].

configuration for Ir4+ ions, for

. The obtained

a strong orbital-lattice coupling in pPv CaIrO3 with low-spin t2g5

190 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 2.** Crystal structure of CaIrO3 polymorphs: (a) pPv and (b) Pv.

*pPv CaIrO3* It is an antiferromagnetic insulator with *T*<sup>N</sup> ≈ 110 K. Due to the difficulty in obtaining single-phase samples, the physical properties of pPv CaIrO3 were not characterized until 2006 by Ohgushi et al. [25], who first reported its electrical transport and magnetic properties on polycrystalline samples synthesized under HPHT conditions. As shown in Figure 3, its resistivity ρ(T) increases quickly upon cooling, following the Arrhenius-type behaviour, i.e. ρ(T) = ρ<sup>0</sup> exp(∆/T), with the activation energy ∆ = 0.17 eV; magnetic susceptibility χ = M/H exhibits a sharp transition at *T*<sup>N</sup> = 115 K, below which a weak ferromagnetic moment of ~ 0.04 μB/Ir was observed from the isothermal magnetization curve. In addition, a huge coercive field Hc ≈ 4 T was evidenced at low temperatures. Density functional calculations by Subedi [33] demonstrated that the inclusion of SOC can split the t2g bands into fully filled Jeff = 3/2 bands and half-filled Jeff = 1/2 bands, as shown schematic in Figure 1(c), and that both SOC and moderate U are required to reconcile the experimentally observed Mott insulating behaviour. By performing the resonant X-ray diffraction at the L absorption edges of pPv CaIrO3 single crystals, Ohgushi et al. [14] determined its magnetic structure as a stripe-type antiferromag‐ netic order, i.e. the Ir moments are aligned parallel along the *a* axis and antiparallel along the *c* axis with a canted ferromagnetic component along the *b* axis. Bogdanov et al. [34] carried out *ab initio* quantum chemical calculations and reproduced such a striped antiferromagnetic structure. Moreover, their calculations predicted a strong antiferromagnetic exchange inter‐ action of Jc = 121 meV through the corner-shared path along the *c* axis, and a weak nearestneighbour ferromagnetic coupling of Ja ≈ -7.3 meV within the edge-shared chains along the *a* axis. In this regard, pPv CaIrO3 can be regarded as a Jeff = 1/2 quasi-1D antiferromagnet. Although the above results suggested that a Jeff = 1/2 ground state is realized in pPv CaIrO3, first-principles calculations [33, 34] evidenced significant deviations from the ideal Jeff = 1/2 state with highly uneven admixture of the t2g components due to the pronounced tetragonal distortion. In agreement with these calculations, a very recent resonant inelastic X-ray scattering (RIXS) study by Sala et al. [15] confirmed the departure from the Jeff = 1/2 state. By analyzing the RIXS spectrum, they estimated the effective tetragonal crystal field splitting ∆ = –0.71 eV and the SOC ζSO = 0.52 eV, from which a ground state wave function |**0**, ± = ∓ **0.32**| *xy*, ∓ + **0.67**(| *yz*, ± ∓ *i* | *zx*, ± ) with a dominant yz±izx orbital character was derived.

The Mott insulating nature of quasi-2D pPv CaIrO3 have motivated Ohgushi et al. [25] to metallize it via the carrier doping. They successfully prepared a series of hole-doped Ca1– *<sup>x</sup>*Na*x*IrO3 (0 ≤ *x* ≤ 0.37) with pPv structure under HPHT conditions and realized a filling-control antiferromagnetic insulator to paramagnetic metal transition around *x* = 0.3, near which, however, no superconductivity was observed. Nevertheless, anomalous properties such as the non-Fermi-liquid behaviour and positive magnetoresistance violating Kohler's rule were observed for the metallic samples. In the same report, attempt of electron doping via Y3+ substitutions for Ca2+ was unsuccessful due to the formation of pyrochlore phase. In a recent work by Gunasekera et al. [35], however, Y-substituted Ca1–*x*Y*x*IrO3 (0 ≤ *x* ≤ 0.5) samples with pPv structure were successfully fabricated via a conventional solid-state reaction route at ambient pressure. Similar with the Na+ -doped case, about 30%-Y3+ doping can also drive an insulator-to-metal transition; superconductivity was not observed either in this case. In striking contrast with the gradual suppression of *T*<sup>N</sup> by Na doping, surprisingly, *T*N remains unchanged upon Y substitutions up to 50%, except that the magnitude of ac susceptibility peak decreases about one order. Further experiments are needed to exclude the possibility that the magnetic order arises from the minor CaIrO3 phase. Anyhow, metallization of the quasi-2D pPv CaIrO3 represents an interesting direction to pursue exotic electronic state in the vicinity of metal–insulator transition.

*Pv CaIrO3* In sharp contrast with the antiferromagnetic insulating ground state of pPv CaIrO3, the Pv phase has been reported as a Pauli paramagnetic metal by Sarkozy et al. [12] in 1974.

Synthesis, Crystal Structure, and Physical Properties of the Perovskite Iridates http://dx.doi.org/10.5772/61281 193

Hc ≈ 4 T was evidenced at low temperatures. Density functional calculations by Subedi [33] demonstrated that the inclusion of SOC can split the t2g bands into fully filled Jeff = 3/2 bands and half-filled Jeff = 1/2 bands, as shown schematic in Figure 1(c), and that both SOC and moderate U are required to reconcile the experimentally observed Mott insulating behaviour. By performing the resonant X-ray diffraction at the L absorption edges of pPv CaIrO3 single crystals, Ohgushi et al. [14] determined its magnetic structure as a stripe-type antiferromag‐ netic order, i.e. the Ir moments are aligned parallel along the *a* axis and antiparallel along the *c* axis with a canted ferromagnetic component along the *b* axis. Bogdanov et al. [34] carried out *ab initio* quantum chemical calculations and reproduced such a striped antiferromagnetic structure. Moreover, their calculations predicted a strong antiferromagnetic exchange inter‐ action of Jc = 121 meV through the corner-shared path along the *c* axis, and a weak nearestneighbour ferromagnetic coupling of Ja ≈ -7.3 meV within the edge-shared chains along the *a* axis. In this regard, pPv CaIrO3 can be regarded as a Jeff = 1/2 quasi-1D antiferromagnet. Although the above results suggested that a Jeff = 1/2 ground state is realized in pPv CaIrO3, first-principles calculations [33, 34] evidenced significant deviations from the ideal Jeff = 1/2 state with highly uneven admixture of the t2g components due to the pronounced tetragonal distortion. In agreement with these calculations, a very recent resonant inelastic X-ray scattering (RIXS) study by Sala et al. [15] confirmed the departure from the Jeff = 1/2 state. By analyzing the RIXS spectrum, they estimated the effective tetragonal crystal field splitting ∆ = –0.71 eV and the SOC ζSO = 0.52 eV, from which a ground state wave function |**0**, ± = ∓ **0.32**| *xy*, ∓ + **0.67**(| *yz*, ± ∓ *i* | *zx*, ± ) with a dominant yz±izx orbital character

192 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The Mott insulating nature of quasi-2D pPv CaIrO3 have motivated Ohgushi et al. [25] to metallize it via the carrier doping. They successfully prepared a series of hole-doped Ca1– *<sup>x</sup>*Na*x*IrO3 (0 ≤ *x* ≤ 0.37) with pPv structure under HPHT conditions and realized a filling-control antiferromagnetic insulator to paramagnetic metal transition around *x* = 0.3, near which, however, no superconductivity was observed. Nevertheless, anomalous properties such as the non-Fermi-liquid behaviour and positive magnetoresistance violating Kohler's rule were observed for the metallic samples. In the same report, attempt of electron doping via Y3+ substitutions for Ca2+ was unsuccessful due to the formation of pyrochlore phase. In a recent work by Gunasekera et al. [35], however, Y-substituted Ca1–*x*Y*x*IrO3 (0 ≤ *x* ≤ 0.5) samples with pPv structure were successfully fabricated via a conventional solid-state reaction route at

insulator-to-metal transition; superconductivity was not observed either in this case. In striking contrast with the gradual suppression of *T*<sup>N</sup> by Na doping, surprisingly, *T*N remains unchanged upon Y substitutions up to 50%, except that the magnitude of ac susceptibility peak decreases about one order. Further experiments are needed to exclude the possibility that the magnetic order arises from the minor CaIrO3 phase. Anyhow, metallization of the quasi-2D pPv CaIrO3 represents an interesting direction to pursue exotic electronic state in the vicinity

*Pv CaIrO3* In sharp contrast with the antiferromagnetic insulating ground state of pPv CaIrO3, the Pv phase has been reported as a Pauli paramagnetic metal by Sarkozy et al. [12] in 1974.


was derived.

ambient pressure. Similar with the Na+

of metal–insulator transition.

**Figure 3.** Temperature dependence of (a) resistivity ρ(T) and (b) magnetic susceptibility M/H for the two polymorphs of CaIrO3, pPv for post-perovskite and Pv for perovskite. Adapted from Reference [13].

Recent characterizations by Ohgushi et al. [13] on the dense pellets obtained via HPHT synthesis confirmed the paramagnetic nature, but the temperature dependence of resistivity shows bad metal behaviour with a non-diverge upturn at low temperatures, Figure 3. Such a bad metallic behaviour was reproduced on the epitaxially stabilized thin films grown on various substrates [26, 32], and has been ascribed to a semimetallic ground state with the conduction and valence band touching at Fermi level. The observations of a sign change and a nonlinear magnetic-field dependence of the Hall resistance are consistent with the coexis‐ tence of electron and hole charge carriers [26]. As discussed below for Pv SrIrO3, the semime‐ tallic state might originate from the symmetry-protected Dirac nodes around the Fermi level due to a combined effect of SOC and reflection symmetry of the *Pbnm* orthorhombic lattice. In light of the recent theoretical proposals for the orthorhombic Pv iridates discussed below, further experimental studies on the semimetallic Pv CaIrO3 are highly desirable.

*pPv versus Pv CaMO3 (M = Ir, Rh, Ru)* The distinct ground states of pPv and Pv CaIrO3 reflect the intimate structural–property relationships. In addition to CaIrO3, both CaRuO3 [36] and CaRhO3 [37] have also been reported to possess quenchable Pv and pPv polymorphs. Besides the importance in geosciences as analogy materials of MgSiO3, these compounds with partially filled d-electron shells are important correlated electron systems with intriguing physical properties [37, 38]. In a similar manner as CaIrO3, their ground states differ sharply as the

**Figure 4.** Insulator–metal transition in pPv CaIrO3 induced by hole (Na+ ) and electron (Y3+) doping. (a, b) shows the temperature dependence of resistivity and magnetic susceptibility of Ca1–*x*Na*x*IrO3. Adapted from Reference [25]; (c, d, e) show the temperature dependence of resistivity and ac magnetic susceptibility of Ca1–*x*Y*x*IrO3. Adapted from Refer‐ ence [35].

structure changes: Pv CaRuO3 is a well-known exchanged-enhanced paramagnetic metal on the verge of a ferromagnetic instability, whereas the pPv phase is an antiferromagnetic insulator with *T*N = 270 K [38]; Pv CaRhO3 is a Pauli paramagnetic metal while the pPv phase is insulating and undergoes a canted antiferromagnetic transition below *T*N = 90 K [37]. Current first-principles calculations for the pPv compounds failed to capture the correct ground state; it seems that in addition to electron–electron correlations, SOC also plays an essential role in producing the insulating ground state for these 4d and 5d-electron systems [39].

In addition to the interest in fundamental physics, the CaIrO3 ceramics have also been investigated by Keawprak et al. [40] for the potential thermoelectric applications. They prepared both phases of CaIrO3 with spark plasma sintering technique and evaluated their thermoelectric properties from room temperature up to 1023 K. The highest dimensionless figure of merit (ZT) reaches 0.02 and 0.003 for Pv and pPv phase, respectively.

## **3. SrIrO3**

Depending on the synthesis conditions, SrIrO3 can form in two different structures, i.e. the monoclinically distorted 6H polytype and the orthorhombic GdFeO3-type Pv structure [8]. The former is a rare stoichiometric oxide exhibiting non-Fermi-liquid behaviours near a ferromag‐ netic quantum critical point [11]. The latter was recently found to be an exotic narrow-band semimetal that may harvest many topological and magnetic insulating phases [10, 41, 42].

## **3.1. Synthesis**

The 6H phase can be readily prepared in the polycrystalline form at ambient pressure by sintering the stoichiometric mixture of SrCO3 and IrO2 (or Ir) at 900–1,100°C in air [8]. Single crystals of 6H phase with dimensions ~0.4 × 0.4 × 0.6 mm3 have been grown in Pt crucibles with the SrCl2 self-flux techniques [11]. The Pv phase is a HP form of SrIrO3. Longo et al. [8] performed the first HPHT syntheses and established the temperature–pressure phase diagram for the 6H-Pv transformation of SrIrO3. It was found that the 6H phase transforms to the Pv structure above 1,650°C at 2 GPa and above 700°C at 5 GPa. Recent HPHT syntheses of Pv SrIrO3 were usually performed at 1,000–1,100°C and 5–6 GPa [43, 44]. For these samples, Rietveld refinements on the powder XRD patterns evidenced the presence of ~3–4 wt.% IrO2 impurity. Since the Pv phase is metastable, it remains a challenge to obtain sizable bulk single crystals under HP conditions. However, Pv SrIrO3 films and superlattices have been stabilized at ambient pressure via applying the epitaxial strain with various techniques, including the metalorganic chemical vapour deposition [9], pulsed laser deposition [45], and reactive oxide molecular-beam expitaxy [42]. As discussed below, given the tolerance factor *t* < 1, it is unusual for SrIrO3 to adopt the 6H structure at ambient pressure. It was recently reported [46, 47] that the 6H structure of SrIrO3 can be destabilized by partial substitution of M = Li, Fe, Co, Ni, Zn for Ir in SrIr1–*x*M*x*O3 and converted to the Pv structure within a narrow composition range around *x* = 0.2. In these cases, it was suggested that the presence of eg type orbitals on the M ions contributes to the breakdown of face-sharing octahedral dimmers in the 6H structure.

## **3.2. Crystal structure**

structure changes: Pv CaRuO3 is a well-known exchanged-enhanced paramagnetic metal on the verge of a ferromagnetic instability, whereas the pPv phase is an antiferromagnetic insulator with *T*N = 270 K [38]; Pv CaRhO3 is a Pauli paramagnetic metal while the pPv phase is insulating and undergoes a canted antiferromagnetic transition below *T*N = 90 K [37]. Current first-principles calculations for the pPv compounds failed to capture the correct ground state; it seems that in addition to electron–electron correlations, SOC also plays an essential role in

temperature dependence of resistivity and magnetic susceptibility of Ca1–*x*Na*x*IrO3. Adapted from Reference [25]; (c, d, e) show the temperature dependence of resistivity and ac magnetic susceptibility of Ca1–*x*Y*x*IrO3. Adapted from Refer‐

) and electron (Y3+) doping. (a, b) shows the

In addition to the interest in fundamental physics, the CaIrO3 ceramics have also been investigated by Keawprak et al. [40] for the potential thermoelectric applications. They prepared both phases of CaIrO3 with spark plasma sintering technique and evaluated their thermoelectric properties from room temperature up to 1023 K. The highest dimensionless

Depending on the synthesis conditions, SrIrO3 can form in two different structures, i.e. the monoclinically distorted 6H polytype and the orthorhombic GdFeO3-type Pv structure [8]. The

producing the insulating ground state for these 4d and 5d-electron systems [39].

**Figure 4.** Insulator–metal transition in pPv CaIrO3 induced by hole (Na+

194 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

figure of merit (ZT) reaches 0.02 and 0.003 for Pv and pPv phase, respectively.

**3. SrIrO3**

ence [35].

*6H SrIrO3* The crystal structure of 6H SrIrO3 was first determined by Longo et al. [8] as a monoclinic distortion of the hexagonal BaTiO3 structure. The 6H hexagonal structure can be described as close-packed SrO3 layers stacked perpendicular to the c axis in the sequence hcchcc, where h and c refer to hexagonal (ABAB...) and cubic (ABCABC...) close packing, respectively. The Ir atoms occupy the oxygen octahedra formed by the SrO3 layers, and the IrO6 octahedra share common faces across an h layer and common corners across a c layer. As a result, the above hcchcc stacking sequence results in two independent positions for the Ir atoms. As shown in Figure 5(a), two Ir2O6 octahedra form pairs of face-shared octahedra that are joined by common corners to a plane of corner-sharing Ir1O6 octahedra. Therefore, the 6H structure can be alternatively depicted as a stacking of layers of corner- (C) and face (F)-sharing IrO6 octahedra in the sequence FCCFCC along the c axis.

In the original work by Longo et al., the oxygen positional parameters were not refined due to the low scattering of oxygen relative to Ir and Sr. Based on the neutron diffraction data, Qasim et al. [46] recently provided a full refinement on the crystal structure of 6H SrIrO3 with *a* = 5.6040 Å, *b* = 9.6256 Å, *c* = 14.1834 Å, and *β* = 93.202° in space group *C2/c* (No. 15). The refined positional parameters and selected bond lengths and bond angles after Reference [46] are listed in Table 4. In this structure, the Ir1O6 octahedron has an average Ir-O distance of 2.006 Å typical of Ir4+ with the individual distances in a narrow range 1.987–2.038 Å. In contrast, the Ir2O6 octahedra in the Ir2O9 dimers are not regular with a longer average Ir-O distance of 2.030 Å. The Ir2–Ir2 distance, 2.770 Å, is relatively short and close to the separation of 2.72 Å found in Ir metal, suggesting a strong Ir–Ir bonding across the common faces. Variable temperature XRD measurements on the 6H phase confirmed that the monoclinic structure is stable without any structural transition up to 1,000°C in air.

**Figure 5.** Crystal structure of SrIrO3 polytypes: (a) 6H and (b) Pv.

*Pv SrIrO3* Similar as Pv CaIrO3, it consists of a 3D network of corner-sharing IrO6 octahedra that are cooperatively rotated and tilted about the pseudocubic [110] and [001] axes, Figure 5(b). Blanchard et al. [44] recently performed a thorough study on the crystal structure of Pv SrIrO3 based on the synchrotron and neutron diffraction data. The crystal structure was refined in space group *Pbnm* (No. 62) with Sr at *4c* (*x*, *y*, 1/4), Ir at *4a* (0, 0, 0), O1 at *4c* (*x*, *y*, 1/4), and O2 at *8d* (*x*, *y*, *z*) sites, respectively. The lattice parameters at room temperature are determined as *a* = 5.60075 Å, *b* = 5.57115 Å, and *c* = 7.89601 Å. The obtained positional parameters and selected bond lengths and bond angles after Reference [44] are listed in Table 5. As can be seen, the individual IrO6 octahedron is relatively rigid with three Ir–O distances being approxi‐ mately equal. The average Ir–O distance at 300 K of 2.016 Å is consistent with the ionic radii sum for Ir4+ and O2–. From the refined atomic coordinates, Blanchard et al. also estimated the two independent octahedral tilt angles, i.e. ψ = 11.5° for out-of-phase tilt about the pseudocubic [110] axis, and φ = 8.7° for in-phase tilt about the pseudocubic [001] axis, respectively. These tilting angles were found to be nearly temperature-independent below room temperature. The orthorhombic *Pbnm* structure was shown to persist over the temperature range 3–1,070 K.

*6H-Pv transformation* As pointed out by Longo et al. [8], SrIrO3 and SrMnO3 are the only SrBO3 (B = Ti, Zr, Hf, Cr, Mo, Tc, Fe, Ru, Sn, Pb, Ce, Th) compounds that do not adopt the Pv structure at ambient pressure. Given the tolerance factor *t* ≤ 1, the Pv structure would be stabilized for these compounds. However, SrMnO3 has the 4H polytype structure with a stacking sequence of hchc along the *c* axis, while SrIrO3 crystallizes in the 6H polytype as mentioned above. It would appear that the hexagonal polytypes with their face-shared octahedra and trigonal crystal fields are stabilized by the outer electron configurations that allow for metal–metal bonding along the *c* axis. For example, in the case of low-spin Ir4+, the trigonal crystal field of the hexagonal polytype should split the t2g5 orbitals to egσ<sup>0</sup> egπ<sup>4</sup> a1g1 , which allow for metal–metal bonding along the *c* axis via the half-filled a1g orbitals. Thus, the 6H SrIrO3 and 4H SrMnO3 can be regarded as a compromise between the continuous face-shared chains of the 2H polytype and the geometrically favoured Pv structure. Since high pressure prefers the cubic close packing with a higher density than the hexagonal close packing, 6H SrIrO3 transforms to the Pv structure under high-pressure conditions with a ~3% volume reduction.

in Table 4. In this structure, the Ir1O6 octahedron has an average Ir-O distance of 2.006 Å typical of Ir4+ with the individual distances in a narrow range 1.987–2.038 Å. In contrast, the Ir2O6 octahedra in the Ir2O9 dimers are not regular with a longer average Ir-O distance of 2.030 Å. The Ir2–Ir2 distance, 2.770 Å, is relatively short and close to the separation of 2.72 Å found in Ir metal, suggesting a strong Ir–Ir bonding across the common faces. Variable temperature XRD measurements on the 6H phase confirmed that the monoclinic structure is stable without

*Pv SrIrO3* Similar as Pv CaIrO3, it consists of a 3D network of corner-sharing IrO6 octahedra that are cooperatively rotated and tilted about the pseudocubic [110] and [001] axes, Figure 5(b). Blanchard et al. [44] recently performed a thorough study on the crystal structure of Pv SrIrO3 based on the synchrotron and neutron diffraction data. The crystal structure was refined in space group *Pbnm* (No. 62) with Sr at *4c* (*x*, *y*, 1/4), Ir at *4a* (0, 0, 0), O1 at *4c* (*x*, *y*, 1/4), and O2 at *8d* (*x*, *y*, *z*) sites, respectively. The lattice parameters at room temperature are determined as *a* = 5.60075 Å, *b* = 5.57115 Å, and *c* = 7.89601 Å. The obtained positional parameters and selected bond lengths and bond angles after Reference [44] are listed in Table 5. As can be seen, the individual IrO6 octahedron is relatively rigid with three Ir–O distances being approxi‐ mately equal. The average Ir–O distance at 300 K of 2.016 Å is consistent with the ionic radii sum for Ir4+ and O2–. From the refined atomic coordinates, Blanchard et al. also estimated the two independent octahedral tilt angles, i.e. ψ = 11.5° for out-of-phase tilt about the pseudocubic [110] axis, and φ = 8.7° for in-phase tilt about the pseudocubic [001] axis, respectively. These tilting angles were found to be nearly temperature-independent below room temperature. The orthorhombic *Pbnm* structure was shown to persist over the temperature range 3–1,070 K.

*6H-Pv transformation* As pointed out by Longo et al. [8], SrIrO3 and SrMnO3 are the only SrBO3 (B = Ti, Zr, Hf, Cr, Mo, Tc, Fe, Ru, Sn, Pb, Ce, Th) compounds that do not adopt the Pv structure at ambient pressure. Given the tolerance factor *t* ≤ 1, the Pv structure would be stabilized for these compounds. However, SrMnO3 has the 4H polytype structure with a

any structural transition up to 1,000°C in air.

196 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 5.** Crystal structure of SrIrO3 polytypes: (a) 6H and (b) Pv.


**Table 4.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for 6H SrIrO3 from neutron diffraction [46]: space group *C2/c* (No. 15), *a* = 5.60401 Å, *b* = 9.6256 Å, *c* = 14.1834 Å, *β* = 93.202°, *V* = 763.89 Å3 , Z = 12.


**Table 5.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for Pv SrIrO3 from neutron diffraction [44]: space group *Pbnm* (No. 62), *a* = 5.60075 Å, *b* = 5.57115 Å, *c* = 7.89601 Å, *V* = 246.376 Å3 , Z = 4.

## **3.3. Physical properties**

*6H SrIrO3* Although the 6H SrIrO3 has been synthesized more than 50 years ago, its physical properties were not characterized in detail until 2007 by Cao et al. [11], who reported the magnetic, electrical transport, and calorimetric properties of 6H SrIrO3 single crystals grown out of the SrCl2 flux. The primary results are summarized in Figure 6. Magnetic susceptibility χ(*T*) measurements evidenced no long-range magnetic order down to 1.7 K, but exhibited at low temperatures strong enhancements that diverge as χ∝ *T*<sup>γ</sup> with 1/2 < *γ* < 1, suggesting the proximity to a ferromagnetic instability. The isothermal magnetization *M*(*H*) at 1.7 K indeed displays a saturation behaviour at *H* ~ 3 *T*, yet the saturation moment is very small, being less than 0.03 μB/Ir. The low-temperature specific heat *C*(*T*) exhibits a pronounced –*T*log*T* dependence, which is characteristic of non-Fermi-liquid systems. Such a *C*/*T* ~ –log*T* behaviour can be readily enhanced in low applied fields up to 1.1 *T*, vanishes for *H* > 2 *T*, and eventually changes to a *T*3/2 power law expected for a ferromagnetically ordered state at *H* = 8 *T*. In accordance with the *C*(*T*) results, both the *c*-axis resistivity, ρc, and the *ab*-plane resistivity, ρab(T), follow a non-Fermi-liquid T3/2 dependence over a wide temperature range up to 120 K under zero field, while a Fermi-liquid *T*<sup>2</sup> behaviour is restored upon applying an external field *H* ≥ 5 *T*. Taking into account all these observations, 6H SrIrO3 can be regarded a rare example of stoichiometric oxide that exhibits non-Fermi-liquid behaviours near a ferromagnetic quantum critical point. As will be shown explicitly in BaIrO3, such a quantum critical point can be realized via subtle structural variations.

*Pv SrIrO3* In the original work by Longo et al. [8], Pv SrIrO3 has been described as a Pauli paramagnetic metal. In 2008, Zhao et al. [43] reinvestigated the physical properties of Pv SrIrO3 bulk sample prepared under 5 GPa and 1,000°C. They observed two characteristic temperatures *T*\* ≈ 170 K and *T*MI ≈ 44 K: at *T*\* , the paramagnetic susceptibility χ(*T*) starts to increase with temperature, and the resistivity ρ(*T*) exhibits a slope change, followed below *T*\* by the presence of unusual linear field dependence positive magnetoresistance (MR) that reaches about 12% at 5 K and 7 T; a broad metal–insulator transition was observed at *T*MI. However, these observations are largely disapproved by the very recent work of Blanchard et

Synthesis, Crystal Structure, and Physical Properties of the Perovskite Iridates http://dx.doi.org/10.5772/61281 199

**Atom Site** *x y z* **Biso (Å2**

Sr 4c -0.0068 0.4687 1/4 0.019 Ir 4a 0 0 0 0.017 O1 4c 0.0718 0.0049 1/4 0.019 O2 8d 0.2126 0.2877 -0.0369 0.022

Ir-O1 (×2) 2.015 Ir-O1-Ir 156.92 Ir-O2 (×2) 2.018 Ir-O2-Ir 156.22

**Table 5.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for Pv SrIrO3 from neutron

*6H SrIrO3* Although the 6H SrIrO3 has been synthesized more than 50 years ago, its physical properties were not characterized in detail until 2007 by Cao et al. [11], who reported the magnetic, electrical transport, and calorimetric properties of 6H SrIrO3 single crystals grown out of the SrCl2 flux. The primary results are summarized in Figure 6. Magnetic susceptibility χ(*T*) measurements evidenced no long-range magnetic order down to 1.7 K, but exhibited at low temperatures strong enhancements that diverge as χ∝ *T*<sup>γ</sup> with 1/2 < *γ* < 1, suggesting the proximity to a ferromagnetic instability. The isothermal magnetization *M*(*H*) at 1.7 K indeed displays a saturation behaviour at *H* ~ 3 *T*, yet the saturation moment is very small, being less than 0.03 μB/Ir. The low-temperature specific heat *C*(*T*) exhibits a pronounced –*T*log*T* dependence, which is characteristic of non-Fermi-liquid systems. Such a *C*/*T* ~ –log*T* behaviour can be readily enhanced in low applied fields up to 1.1 *T*, vanishes for *H* > 2 *T*, and eventually changes to a *T*3/2 power law expected for a ferromagnetically ordered state at *H* = 8 *T*. In accordance with the *C*(*T*) results, both the *c*-axis resistivity, ρc, and the *ab*-plane resistivity, ρab(T), follow a non-Fermi-liquid T3/2 dependence over a wide temperature range up to 120 K

*H* ≥ 5 *T*. Taking into account all these observations, 6H SrIrO3 can be regarded a rare example of stoichiometric oxide that exhibits non-Fermi-liquid behaviours near a ferromagnetic quantum critical point. As will be shown explicitly in BaIrO3, such a quantum critical point

*Pv SrIrO3* In the original work by Longo et al. [8], Pv SrIrO3 has been described as a Pauli paramagnetic metal. In 2008, Zhao et al. [43] reinvestigated the physical properties of Pv SrIrO3 bulk sample prepared under 5 GPa and 1,000°C. They observed two characteristic

increase with temperature, and the resistivity ρ(*T*) exhibits a slope change, followed below *T*\* by the presence of unusual linear field dependence positive magnetoresistance (MR) that reaches about 12% at 5 K and 7 T; a broad metal–insulator transition was observed at *T*MI. However, these observations are largely disapproved by the very recent work of Blanchard et

diffraction [44]: space group *Pbnm* (No. 62), *a* = 5.60075 Å, *b* = 5.57115 Å, *c* = 7.89601 Å, *V* = 246.376 Å3

Ir-O2 (×2) 2.018 <Ir-O> 2.017

198 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

under zero field, while a Fermi-liquid *T*<sup>2</sup>

can be realized via subtle structural variations.

temperatures *T*\* ≈ 170 K and *T*MI ≈ 44 K: at *T*\*

**3.3. Physical properties**

**)**

, Z = 4.

behaviour is restored upon applying an external field

, the paramagnetic susceptibility χ(*T*) starts to

**Figure 6.** Physical properties of 6H SrIrO3 single crystal. Adapted from Reference [11]. (a, b) show the temperature de‐ pendence of magnetic susceptibility measured along the *c* axis and *ab* plane, inset of (a) shows the magnetization curve at 1.7 K. (c) shows the specific heat C(T) data illustrating the *C*/*T*~ –log*T* behaviour. (d) shows the resistivity following the *T*3/2 behaviour.

al. [44], who studied the bulk samples prepared under 6 GPa and 1,100°C. As shown in Figure 7, it was found in the latter work that ρ(*T*) exhibits metallic conductivity down to 2 K, following Fermi-liquid *T*<sup>2</sup> dependence between 2 and 30 K, without showing an upturn at low temper‐ ature. In addition, they observed a smaller positive MR up to 2% at 7 *T* and 2–50 K. These discrepancies might originate from the polycrystalline nature of the studied samples, in which the extrinsic effects such as the grain boundary and impurities can largely influence the transport properties. As mentioned above, Pv SrIrO3 films of single-crystal quality can be stabilized via applying epitaxial strain. The resistivity upturn at low temperatures have been frequently observed in these thin films [9, 48], but the upturn temperature displays a broad distribution, in support of an extrinsic property due to weak Anderson localization. Never‐ theless, such sensitivity to defects reflects the bad metal character of semimetallic Pv SrIrO3 approaching the boundary of metal–insulator transition.

As the end member of the Ruddlesden–Popper series Sr*n*+1Ir*n*O3*<sup>n</sup>*+1 (*n* = 1, 2, ∞), Pv SrIrO3 has recently attracted much attention due to the presence of nontrivial features within the Jeff = 1/2 bands. Density-functional theory first-principles calculations by Carter et al. [10] found that in the strong SOC limit the bands near the Fermi energy are mostly composed of Jeff = 1/2 states. Interestingly, they found a node near the U point, Figure 8(a), thus revealing the semimetallic

**Figure 7.** Temperature dependence of (a) magnetic susceptibility and (b) resistivity of Pv SrIrO3. Adapted from Refer‐ ence [44].

nature of Pv SrIrO3. By constructing a tight-binding model, they confirmed the presence of a line node near the U point in the Brilloiun zone, and further shown that the line node originates from the reflection symmetry of the crystal structure at the *z* = 1/4 and 3/4 planes presented in the orthorhombic *Pbnm* space group. Since the line node is protected by the underlying lattice symmetry, it has been further proposed that perturbations breaking the sublayer reflection symmetry can lift the line node and convert the system into an insulating phase; the system may become a strong topological insulator at a certain point. In addition, as shown in Figure 8(b), magnetically ordered metallic and insulating phases have also been proposed to arise in the U versus SOC phase diagram of Pv SrIrO3 [41]. Moreover, Chen et al. [49] further proposed that the presence of reflection symmetry in orthorhombic Pv iridates may realize a novel class of topological crystalline metals with zero-energy surface states at certain planes.

Recent angle-resolved photoemission spectroscopy on Pv SrIrO3 films by Nie et al. [42] has uncovered such an exotic semimetallic state with very narrow bands near the Fermi surface consisting of heavy hole-like pockets around (±π, 0) and (0, 0) and light electron-like pockets at (±π/2, ±π/2). Surprisingly, the bandwidth of Pv SrIrO3 is found to be narrower than that of Sr2IrO4, in contrary to the general expectations of broaden bandwidth with increasing dimen‐ sionality [7]. Since the semimetallic ground state has been confirmed experimentally, it is of particular interest to achieve the proposed topological and/or magnetic states via tuning the SOC, U, and/or lattice symmetry. In this regard, Matsuno et al. [45] have made an important step towards these exotic phases; they tailored a spin-orbit magnetic insulator out of the semimetallic state via controlling the dimensionality of [(SrIrO3)m, SrTiO3] superlattices. By utilizing HPHT synthesis, we prepared a series of Sn-doped SrIr1–*x*Sn*x*O3 orthorhombic perovskites. We found that substitutions of isovalent, nonmagnetic Sn4+ for Ir4+ ions lead to a breakdown of the semimetallic state, and convert the paramagnetic, semimetallic ground state

**Figure 8.** (a) LDA band structure of Pv SrIrO3 with Hubbard U = 2 eV and SOC, demonstrating the presence of the node near the U point of Jeff = 1/2 band near the Fermi level; (b) the phase diagram of Pv SrIrO3 in the U-SOC plane containing three phases: magnetic metal (MM), nonmangetic metal or semimetal (M/SM), and magnetic insulator (MI). Adapted from Reference [41].

of Pv SrIrO3 to an antiferromagnetic insulator with a concomitant metal–insulator transition at *T*<sup>N</sup> [50]. These recent experimental efforts demonstrated that semimetallic Pv SrIrO3 is a promising candidate for realizing distinct topological and magnetic insulating states that deserve further investigations in the near future. On the other hand, the Pv SrIrO3 film has also been regarded as potential electrode material for microelectronic devices [9].

## **4. BaIrO3**

nature of Pv SrIrO3. By constructing a tight-binding model, they confirmed the presence of a line node near the U point in the Brilloiun zone, and further shown that the line node originates from the reflection symmetry of the crystal structure at the *z* = 1/4 and 3/4 planes presented in the orthorhombic *Pbnm* space group. Since the line node is protected by the underlying lattice symmetry, it has been further proposed that perturbations breaking the sublayer reflection symmetry can lift the line node and convert the system into an insulating phase; the system may become a strong topological insulator at a certain point. In addition, as shown in Figure 8(b), magnetically ordered metallic and insulating phases have also been proposed to arise in the U versus SOC phase diagram of Pv SrIrO3 [41]. Moreover, Chen et al. [49] further proposed that the presence of reflection symmetry in orthorhombic Pv iridates may realize a novel class

**Figure 7.** Temperature dependence of (a) magnetic susceptibility and (b) resistivity of Pv SrIrO3. Adapted from Refer‐

200 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

ence [44].

of topological crystalline metals with zero-energy surface states at certain planes.

Recent angle-resolved photoemission spectroscopy on Pv SrIrO3 films by Nie et al. [42] has uncovered such an exotic semimetallic state with very narrow bands near the Fermi surface consisting of heavy hole-like pockets around (±π, 0) and (0, 0) and light electron-like pockets at (±π/2, ±π/2). Surprisingly, the bandwidth of Pv SrIrO3 is found to be narrower than that of Sr2IrO4, in contrary to the general expectations of broaden bandwidth with increasing dimen‐ sionality [7]. Since the semimetallic ground state has been confirmed experimentally, it is of particular interest to achieve the proposed topological and/or magnetic states via tuning the SOC, U, and/or lattice symmetry. In this regard, Matsuno et al. [45] have made an important step towards these exotic phases; they tailored a spin-orbit magnetic insulator out of the semimetallic state via controlling the dimensionality of [(SrIrO3)m, SrTiO3] superlattices. By utilizing HPHT synthesis, we prepared a series of Sn-doped SrIr1–*x*Sn*x*O3 orthorhombic perovskites. We found that substitutions of isovalent, nonmagnetic Sn4+ for Ir4+ ions lead to a breakdown of the semimetallic state, and convert the paramagnetic, semimetallic ground state

At ambient pressure, BaIrO3 crystallizes in the nine-layer (9R) polytype. It is the first known ferromagnetic insulator with *T*c ≈ 180 K among the 5d TMOs [17]. Detailed studies on single crystals revealed a charge-density-wave (CDW) formation below the ferromagnetic order [18, 51]. Recent experimental [52] and theoretical [53] investigations further revealed it as an exotic spin-orbit Mott insulator that is of great current research interest. Following the general trend of perovskite hexagonal polytypes, we have explored the high-pressure sequences of BaIrO3 and found three more polytypes, i.e. 5H, 6H, and 3C [19, 20, 21]. Their ground states exhibit an interesting evolution from a ferromagnetic insulator to a Pauli paramagnetic metal passing through a ferromagnetic quantum critical point tuned by the gradual structural changes as detailed below.

## **4.1. Synthesis**

The ambient-pressure 9R phase can be readily obtained by sintering the stoichiometric mixtures of BaCO3 and Ir at 1,000°C in air. The sample should be cooled down slowly for the last sintering in order to ensure an oxygen stoichiometry [54]. Single crystals have been reported to grow out of the BaCl2 flux at a relatively low temperature of 1,000 K [18]. HPHT synthesis is needed for all the other polytypes [19, 21, 55, 56]. For the HP syntheses around 1,000°C, the 9R polytype is stable up to 3 GPa, the 5H phase exists only in a narrow pressure range around 4 GPa, the 6H phase is stabilized in a wide pressure range from 5 to ~20 GPa, and the 3C phase was finally obtained at 25 GPa. We have employed the two-stage (Walkeror Kawai-type) multianvil systems for the HPHT syntheses. During the HPHT experiments, the sample was first compressed to the desired pressure by eight truncated tungsten carbide anvils, and then the temperature was increased to ~1,000°C and kept for 30 min before quenching to room temperature. The resultant samples were recovered after releasing pressure and then subjected to various characterizations at ambient pressure.

## **4.2. Crystal structure**

*9R BaIrO3* As shown in Figure 9(a), the crystal structure of the 9R phase consists of Ir3O12 trimers of face-sharing octahedra that are linked by their vertices to form columns parallel to the *c*axis, with a stacking of layers of corner-sharing (C) and face-sharing (F) IrO6 octahedra in the order FFCFFCFFC along the *c* axis. Except for the monoclinic distortion, it is isostructural with the 9R BaRuO3. The monoclinic distortion generates twisting and buckling of the Ir3O12 trimers that are tilted ~12° relative to each other. Here, we adopted the crystal structure of 9R BaIrO2.94 obtained by Powell et al. [54] from the NPD data, which were refined in a structural model defined in the *C2/m* space group, with three kinds of Ba atoms at 4*i* (*x*, 0, *z*) positions, four types of unequivalent Ir atoms at 4*i*, 2*a* (0, 0, 0) and 2*d* (0.5, 0, 0.5) sites, and six types of oxygen atoms at 4*i* and 8*j* (*x*, *y*, *z*) positions. The obtained unit-cell parameters are *a* = 9.9992 Å, *b* = 5.7490 Å, *c* = 15.1707 Å, and *β* = 103.27°. The final positional parameters and the selected bond lengths and bong angles after Reference [54] are listed in Table 6. Ir1–Ir2 and Ir3–Ir4 distances, of 2.618 Å and 2.627 Å respectively, are even smaller than the separation of 2.72 Å found in Ir metal, which indicates significant interactions between iridium cations at the centre of face-shared pairs of octahedra. It is important to note that, although this polytype has been compared to the ambient 9R BaRuO3, with rhombohedral (*R-3m*) symmetry and a stacking sequence (*FFC*)3, the monoclinic distortion described for ambient 9R BaIrO3 actually involves a shorter periodicity, with a stacking sequence (*FFC*)2 along the *c* axis, as shown in Figure 9(a).

**Figure 9.** Crystal structure of the BaIrO3 polytypes: (a) 9R, (b) 5H, (c) 6H, and (d) 3C.

crystals revealed a charge-density-wave (CDW) formation below the ferromagnetic order [18, 51]. Recent experimental [52] and theoretical [53] investigations further revealed it as an exotic spin-orbit Mott insulator that is of great current research interest. Following the general trend of perovskite hexagonal polytypes, we have explored the high-pressure sequences of BaIrO3 and found three more polytypes, i.e. 5H, 6H, and 3C [19, 20, 21]. Their ground states exhibit an interesting evolution from a ferromagnetic insulator to a Pauli paramagnetic metal passing through a ferromagnetic quantum critical point tuned by the gradual structural changes as

202 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The ambient-pressure 9R phase can be readily obtained by sintering the stoichiometric mixtures of BaCO3 and Ir at 1,000°C in air. The sample should be cooled down slowly for the last sintering in order to ensure an oxygen stoichiometry [54]. Single crystals have been reported to grow out of the BaCl2 flux at a relatively low temperature of 1,000 K [18]. HPHT synthesis is needed for all the other polytypes [19, 21, 55, 56]. For the HP syntheses around 1,000°C, the 9R polytype is stable up to 3 GPa, the 5H phase exists only in a narrow pressure range around 4 GPa, the 6H phase is stabilized in a wide pressure range from 5 to ~20 GPa, and the 3C phase was finally obtained at 25 GPa. We have employed the two-stage (Walkeror Kawai-type) multianvil systems for the HPHT syntheses. During the HPHT experiments, the sample was first compressed to the desired pressure by eight truncated tungsten carbide anvils, and then the temperature was increased to ~1,000°C and kept for 30 min before quenching to room temperature. The resultant samples were recovered after releasing pressure

*9R BaIrO3* As shown in Figure 9(a), the crystal structure of the 9R phase consists of Ir3O12 trimers of face-sharing octahedra that are linked by their vertices to form columns parallel to the *c*axis, with a stacking of layers of corner-sharing (C) and face-sharing (F) IrO6 octahedra in the order FFCFFCFFC along the *c* axis. Except for the monoclinic distortion, it is isostructural with the 9R BaRuO3. The monoclinic distortion generates twisting and buckling of the Ir3O12 trimers that are tilted ~12° relative to each other. Here, we adopted the crystal structure of 9R BaIrO2.94 obtained by Powell et al. [54] from the NPD data, which were refined in a structural model defined in the *C2/m* space group, with three kinds of Ba atoms at 4*i* (*x*, 0, *z*) positions, four types of unequivalent Ir atoms at 4*i*, 2*a* (0, 0, 0) and 2*d* (0.5, 0, 0.5) sites, and six types of oxygen atoms at 4*i* and 8*j* (*x*, *y*, *z*) positions. The obtained unit-cell parameters are *a* = 9.9992 Å, *b* = 5.7490 Å, *c* = 15.1707 Å, and *β* = 103.27°. The final positional parameters and the selected bond lengths and bong angles after Reference [54] are listed in Table 6. Ir1–Ir2 and Ir3–Ir4 distances, of 2.618 Å and 2.627 Å respectively, are even smaller than the separation of 2.72 Å found in Ir metal, which indicates significant interactions between iridium cations at the centre of face-shared pairs of octahedra. It is important to note that, although this polytype has been compared to the ambient 9R BaRuO3, with rhombohedral (*R-3m*) symmetry and a stacking

and then subjected to various characterizations at ambient pressure.

detailed below.

**4.1. Synthesis**

**4.2. Crystal structure**



**Table 6.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for 9R BaIrO3 from neutron diffraction [54]: space group *C2/m* (No. 12), *a* = 9.9992 Å, *b* = 5.7490 Å, *c* = 15.1707 Å, *β* = 103.27°, *V* = 848.81 Å3 , Z = 12.

*5H BaIrO3* The 5H phase was discovered as a new perovskite polytype [19]. As shown in Figure 9(b), its crystal structure contains chains of double dimer units that are corner-connected via oxygen atoms. These clusters of four octahedra are interleaved with single layers of vertexsharing IrO6 octahedra, forming infinite chains along the *c* axis. Adjacent chains are interlinked along the *a* and *b* directions via Ir–O–Ir vertex-sharing bridges. Alternatively, the structure can be described as stacking of layers of corner-sharing (C) and face-sharing (F) IrO6 octahedra along the sequence FCFCC. The crystal structure was refined in the monoclinic *C2/m* (No. 12) space group, with three kinds of Ba atoms at 2*c* (0.5, 0.5, 0.5) and 4*i* (*x*, 0, *z*), three types of Ir atoms at 2*a* (0, 0, 0) and 4*i* sites, and six unequivalent oxygen atoms at 8*j* (*x*, *y*, *z*), 4*i*, 2*d* (0.5, 0, 0.5), and 4*f* (0.75, 0.75, 0) positions. The lattice parameters at room temperature are determined as *a* = 9.9554 Å, *b* = 5.7434 Å, *c* = 13.8049 Å, and *β* = 119.23°. The final positional parameters and the selected bond lengths and bond angles after Reference [19] are listed in Table 7. As can be seen, Ir–O distances vary in the range 1.90 Å for Ir1–O4 to 2.23 Å for Ir2–O3. The average value, of 2.03 Å, is consistent with the ionic radii sum for Ir4+ and O2-. It is noteworthy that the structure contains three kinds of octahedra with rather distinct average sizes: <Ir–O> are 1.985 Å, 2.072 Å, and 2.017 Å for Ir1, Ir2, and Ir3 octahedra. The two largest octahedra, Ir2 and Ir3, are those forming dimers, where the Ir–O bonds are weakened by the Ir–Ir bonds. According to these bond distances, the bond valences for the three types of octahedra are 4.26(8)+, 3.35(8)+, and 4.06(9)+, indicating that Ir1 and Ir2 are under certain compressive and tensile stresses, respectively.

*6H BaIrO3* Same as the 6H SrIrO3, the crystal structure of 6H BaIrO3 consists of dimers of facesharing octahedra separated by single corner-sharing octahedron, showing the sequence FCCFCC along the *c* axis. Based on the XRD data, we have refined its crystal structure in the monoclinic *C2/c* space group with two kinds of Ba atoms at 4*e* (0, *y*, ¼) and 8*f* (*x*, *y*, *z*) positions, Ir1 at 4*a* (0, 0, 0) and Ir2 at 8*f* sites, and four independent oxygen atoms, O1 at 4*e*, O2, O3, and O4 at 8*f* positions. The obtained unit-cell parameters are *a* = 5.7483 Å, *b* = 9.9390 Å, *c* = 14.3582 Å, and *β* = 91.319°. The final positional parameters and selected bond lengths and bond angles after Reference [19] are listed in Table 8. As can be seen, the Ir2O6 octahedra within the facesharing dimmers are considerably more expanded than the Ir1O6 octahedra, with average Ir– O distances of 2.16 and 1.99 Å, respectively. This is probably a consequence of the metal–metal bond linking the couples of Ir2 atoms in the dimmers, with Ir2–Ir2 distances of 2.710 Å.

**Atom Site** *x y z* **Biso (Å2**

Ir2-O4 (×4) 2.034 Ir1-O4-Ir2 80.1 <Ir2-O> 2.04 Ir3-O5-Ir4 81.4 Ir3-O1 1.978 Ir3-O6-Ir4 80.0

**Table 6.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for 9R BaIrO3 from neutron diffraction [54]: space group *C2/m* (No. 12), *a* = 9.9992 Å, *b* = 5.7490 Å, *c* = 15.1707 Å, *β* = 103.27°, *V* = 848.81 Å3

*5H BaIrO3* The 5H phase was discovered as a new perovskite polytype [19]. As shown in Figure 9(b), its crystal structure contains chains of double dimer units that are corner-connected via oxygen atoms. These clusters of four octahedra are interleaved with single layers of vertexsharing IrO6 octahedra, forming infinite chains along the *c* axis. Adjacent chains are interlinked along the *a* and *b* directions via Ir–O–Ir vertex-sharing bridges. Alternatively, the structure can be described as stacking of layers of corner-sharing (C) and face-sharing (F) IrO6 octahedra along the sequence FCFCC. The crystal structure was refined in the monoclinic *C2/m* (No. 12) space group, with three kinds of Ba atoms at 2*c* (0.5, 0.5, 0.5) and 4*i* (*x*, 0, *z*), three types of Ir atoms at 2*a* (0, 0, 0) and 4*i* sites, and six unequivalent oxygen atoms at 8*j* (*x*, *y*, *z*), 4*i*, 2*d* (0.5, 0, 0.5), and 4*f* (0.75, 0.75, 0) positions. The lattice parameters at room temperature are determined as *a* = 9.9554 Å, *b* = 5.7434 Å, *c* = 13.8049 Å, and *β* = 119.23°. The final positional parameters and the selected bond lengths and bond angles after Reference [19] are listed in Table 7. As can be seen, Ir–O distances vary in the range 1.90 Å for Ir1–O4 to 2.23 Å for Ir2–O3. The average value, of 2.03 Å, is consistent with the ionic radii sum for Ir4+ and O2-. It is noteworthy that the structure contains three kinds of octahedra with rather distinct average sizes: <Ir–O> are 1.985 Å, 2.072 Å, and 2.017 Å for Ir1, Ir2, and Ir3 octahedra. The two largest octahedra, Ir2 and Ir3, are those forming dimers, where the Ir–O bonds are weakened by the Ir–Ir bonds. According to these bond distances, the bond valences for the three types of octahedra are 4.26(8)+, 3.35(8)+, and 4.06(9)+, indicating that Ir1 and Ir2 are under certain compressive and tensile stresses,

*6H BaIrO3* Same as the 6H SrIrO3, the crystal structure of 6H BaIrO3 consists of dimers of facesharing octahedra separated by single corner-sharing octahedron, showing the sequence FCCFCC along the *c* axis. Based on the XRD data, we have refined its crystal structure in the monoclinic *C2/c* space group with two kinds of Ba atoms at 4*e* (0, *y*, ¼) and 8*f* (*x*, *y*, *z*) positions, Ir1 at 4*a* (0, 0, 0) and Ir2 at 8*f* sites, and four independent oxygen atoms, O1 at 4*e*, O2, O3, and O4 at 8*f* positions. The obtained unit-cell parameters are *a* = 5.7483 Å, *b* = 9.9390 Å, *c* = 14.3582

Ir3-O2 (×2) 2.037 Ir3-O5 (×2) 1.955 Ir3-O6 2.057 <Ir3-O> 2.02 Ir4-O6 (×2) 2.030 Ir4-O5 (×4) 2.035 <Ir4-O> 2.03

204 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

respectively.

**)**

, Z = 12.


**Table 7.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for 5H BaIrO3 from neutron diffraction [19]: space group *C2/m* (No. 12), *a* = 9.9554 Å, *b* = 5.7434 Å, *c* = 13.8049 Å, *β* = 119.231°, *V* = 688.8 Å3 , Z = 10.

*3C BaIrO3* A single-phase Pv BaIrO3 was finally obtained at 25 GPa [21]. Instead of the simple cubic phase with space group *Pm-3m*, it was found that the XRD pattern of the Pv phase can be refined excellently in the tetragonal *I4/mcm* (No. 140) space group with the Ba atom at *4b* (0, ½, ¼), the Ir atom at *4c* (0, 0, 0), and two kinds of O atoms at *4a* (0, 0, ¼) and *8h* (*x*, *x*+½, 0) sites. The obtained unit-cell parameters are *a* = *b* = 5.7044 Å and *c* = 8.0926 Å. The final positional parameters and the main bond distances and bond angles after Reference [21] are listed in Table 9. It should be noted that we denoted this phase as "3C" in order to follow the conven‐ tional notations.


**Table 8.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for 6H BaIrO3 from powder XRD [19]: space group *C2/c* (No. 15), *a* = 5.7483 Å, *b* = 9.9390 Å, *c* = 14.3582 Å, *β* = 91.319°, *V* = 820.12 Å3 , Z = 12.

The small tetragonal distortion of the 3C BaIrO3 phase is unexpected; we should have a cubic phase as found for BaRuO3 formed under high pressure. Such a distortion to tetragonal symmetry by cooperative rotations of the IrO6/2 octahedra about the *c* axis is typical of an A2+ B4+O3 perovskite with a tolerance factor *t ≡* (A−O) / 2 (B−O) a little smaller than unity. However, stabilization of BaIrO3 in the 9R polytype at ambient pressure is consistent with *t* > 1 obtained from tabulated equilibrium ionic radii. The larger compressibility of the Ba−O bond makes it possible to stabilize the 3C phase of BaIrO3 under 25 GPa pressure, but compression of the Ba−O bond by cubic symmetry should not reduce the tolerance factor below *t* = 1. Retention of the cubic symmetry of the IrO6 octahedra shows that the threefold degeneracy of the 5d π\* bands is not a factor. In fact, the tolerance factor calculated from the measured <Ba −O> and <Ir−O> bond lengths in Table 9 gives a *t* = 0.998 consistent with the tetragonal structure observed. Therefore, we conclude that at 25 GPa there has been a first-order transition of the Ba−O equilibrium bond length to give a *t* < 1, which indicates that the 5d π\* bands of the IrO6 array may also have transitioned for Jeff =1/2 and 3/2 bands as a result of a reduction of the orbital angular momentum where the bandwidth is broadened. The high-pressure equilibrium (Ba–O) bond length is retained as a metastable bond length on removal of the pressure, and the Ir–O bonds are not under a tensile stress.


**Table 9.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for 3C BaIrO3 from powder XRD [21]: space group *I4/mcm* (No. 140), *a* = *b* = 5.7044 Å, *c* = 8.0916 Å, *V* = 263.30 Å3 , Z = 4.

*Polytype structures* The polytype structures of the ABO3 oxides and the phase transformation under high pressure were established during the 1960–1970s, see the Review [57]. As a general trend, the number of the hexagonal close packing along the *c* axis in a unit cell is reduced as the synthesis pressure increases, which led to a decrease (increase) of face(corner)-sharing octahedra. This is consistent with that fact that pressure stabilizes preferentially the denser phase. The observed crystallographic densities of the 9R, 5H, 6H and 3C phases of BaIrO3 are 8.84, 9.08, 9.17, and 9.36 g/cm3 , respectively. They progressively increase as expected, since these phases have been stabilized at increasing pressures. This sequence corresponds, there‐ fore, to more dense packing of the BaO3 layers along the *c* axis, showing an evolution to structures with more corner (C) sharing and fewer face (F) sharing octahedra, i.e. C:F ratios increase in the order 1:2 (9R), 3:2 (5H), 2:1 (6H), and ∞ (3C). As detailed below, the physical properties exhibit interesting evolution in response to these systematic structural variations.

## **4.3. Physical properties**

be refined excellently in the tetragonal *I4/mcm* (No. 140) space group with the Ba atom at *4b* (0, ½, ¼), the Ir atom at *4c* (0, 0, 0), and two kinds of O atoms at *4a* (0, 0, ¼) and *8h* (*x*, *x*+½, 0) sites. The obtained unit-cell parameters are *a* = *b* = 5.7044 Å and *c* = 8.0926 Å. The final positional parameters and the main bond distances and bond angles after Reference [21] are listed in Table 9. It should be noted that we denoted this phase as "3C" in order to follow the conven‐

206 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Atom Site** *x y z* **Biso (Å2**

Ba1 4e 0 -0.0052 1/4 0.3 Ba2 8f 0.0078 0.3349 0.0912 0.25 Ir1 4a 0 0 0 0.4 Ir2 8f 0.9936 0.3323 0.8442 0.27 O1 4e 0 0.499 1/4 -0.2 O2 8f 0.2180 0.2390 0.2427 -0.2 O3 8f 0.036 0.846 0.0852 -0.2 O4 8f 0.286 0.087 0.049 -0.2 O5 8f 0.809 0.090 0.103 -0.2

Ir1-O3 (×2) 1.93 Ir1-O3-Ir2 164.4 Ir1-O4 (×2) 2.02 Ir1-O4-Ir2 151.4 Ir1-O5 (×2) 2.01 Ir1-O5-Ir2 153.6

Ir2-O1 2.19 Ir2-O1-Ir2 76.4 Ir2-O2 2.22 Ir2-O1-Ir2 75.1

**Table 8.** Refined positional parameters and selected bond lengths (Å) and bond angles (°) for 6H BaIrO3 from powder

The small tetragonal distortion of the 3C BaIrO3 phase is unexpected; we should have a cubic phase as found for BaRuO3 formed under high pressure. Such a distortion to tetragonal symmetry by cooperative rotations of the IrO6/2 octahedra about the *c* axis is typical of an A2+ B4+O3 perovskite with a tolerance factor *t ≡* (A−O) / 2 (B−O) a little smaller than unity. However, stabilization of BaIrO3 in the 9R polytype at ambient pressure is consistent with *t* > 1 obtained from tabulated equilibrium ionic radii. The larger compressibility of the Ba−O bond

XRD [19]: space group *C2/c* (No. 15), *a* = 5.7483 Å, *b* = 9.9390 Å, *c* = 14.3582 Å, *β* = 91.319°, *V* = 820.12 Å3

<Ir1-O> 1.99

Ir2-O2 2.23 Ir2-O3 2.10 Ir2-O4 2.09 Ir2-O5 2.11 <Ir2-O> 2.16 Ir2-Ir2 2.710 **)**

, Z = 12.

tional notations.

*9R BaIrO3* As mentioned above, 9R BaIrO3 is the first known ferromagnet among the 5d TMOs [17]. Cao et al. [18] performed the first detailed experimental study on the single-crystal samples and uncovered a CDW formation accompanying the ferromagnetic order at *T*c ≈ 180 K. The experimental evidences in support of the CDW formation included [18]: (1) a sudden increase of resistivity at *T*c, (2) the presence of non-linear conductivity with negative differen‐ tial resistivity below *T*c, (3) an optical gap formation at ~ 1,200 cm–1 ≈ 9κB*T*c in the electron excitation spectrum and a splitting of a phonon mode at 350 cm–1 for *T* < *T*c, and (4) the emergent X-ray satellite structure below *T*c. Besides the transitions at *T*<sup>c</sup> ≈ 180 K, two additional anomalies have also been observed [18] upon cooling on the c-axis resistivity ρc, which first changes to a metallic behaviour below *T*1 = 80 K and then suddenly enters a Mott-like insulating state below *T*<sup>2</sup> = 26 K, Figure 10. These two additional transitions are absent for the resistivity within the *ab* plane. The simultaneous occurrence of ferromagnetic order and CDW formation is quite unusual, and 9R BaIrO3 has thus been the subject of extensive investigations since then. Later on, Nakano and Terasaki [51] carried out similar current-voltage (I-V) measurements on their single crystals by using a pulsed current in order to exclude the self-heating effects. Their observations of the giant nonlinear conduction only appearing below 30 K, well below *T*<sup>c</sup> ≈ 180K, questioned the above scenario of a simultaneous onset of a CDW and a ferromagnetic transition. Instead of the sliding motion of CDW, they proposed an interplay between two different bands is likely the origin of the nonlinear conduction observed in BaIrO3 [51]. Such discrepancy might arise from the different sample quality. Nevertheless, a clear gap opening is unambiguously evidenced at *T*c by other experimental probes, such as the Seebeck coefficient [51, 58] and the high-resolution photoemission spectroscopy [59]. Currently, it remains elusive whether the gap opening is driven by the magnetic order or the Fermi surface nesting.

The observation of weak ferromagnetism and insulating ground state in the 9R BaIrO3 has attracted renewed interest in recent years in light of the SOC-driven Mott insulating state for iridates. As for the nature of the weak ferromagnetism, there also exist long-standing discrep‐ ancies. Experimentally, a tiny Ir moment of ~0.03 μB/Ir was observed below *T*c. In addition, a modified Curie-Weiss fitting to the inverse susceptibility also evidenced a small effective moment of ~ 0.13 μB [18]. Originally, a spin canting from a localized full-moment antiferro‐ magnetic configuration had been invoked to explain the tiny ordered moment [17]. In contrast, Cao et al. [18] proposed a model of band magnetism with intrinsically small Ir moment due to d–p hybridization and small exchange splitting. Indeed, the muon-spin relaxation meas‐ urements by Brooks et al. [60] provided direct experimental evidences in support of a small Ir moment, i.e. they observed clear oscillations below Tc and found an extremely small internal field at the muon site. Such an itinerant picture of band magnetism, however, is incompatible with the observation of high coercive force and anisotropy in magnetization measurements. By employing the X-ray absorption spectroscopy (XAS) and X-ray magnetic circular dichroism (XMCD) techniques, Laguna-Marco et al. [52] recently elucidated an atomic-like nature of the Ir moment with the orbital moment being ~1.5 times larger than the spin moment, thus highlighting the importance of SOC in addressing the magnetic order of 9R BaIrO3. After taking into account both SOC and moderate on-site coulomb interactions, first-principles calculations by Ju et al. [53] identified 9R BaIrO3 as an exotic spin-orbit Mott insulator with multiple Jeff = 1/2 states associated with the unique face-sharing Ir3O12 octahedral units within the structure.

Although the atomic-like nature of Ir local moment in 9R BaIrO3 was found to be extremely stable against temperature, pressure, and chemical substitutions [52, 61], these external stimuli

K. The experimental evidences in support of the CDW formation included [18]: (1) a sudden increase of resistivity at *T*c, (2) the presence of non-linear conductivity with negative differen‐ tial resistivity below *T*c, (3) an optical gap formation at ~ 1,200 cm–1 ≈ 9κB*T*c in the electron excitation spectrum and a splitting of a phonon mode at 350 cm–1 for *T* < *T*c, and (4) the emergent X-ray satellite structure below *T*c. Besides the transitions at *T*<sup>c</sup> ≈ 180 K, two additional anomalies have also been observed [18] upon cooling on the c-axis resistivity ρc, which first changes to a metallic behaviour below *T*1 = 80 K and then suddenly enters a Mott-like insulating state below *T*<sup>2</sup> = 26 K, Figure 10. These two additional transitions are absent for the resistivity within the *ab* plane. The simultaneous occurrence of ferromagnetic order and CDW formation is quite unusual, and 9R BaIrO3 has thus been the subject of extensive investigations since then. Later on, Nakano and Terasaki [51] carried out similar current-voltage (I-V) measurements on their single crystals by using a pulsed current in order to exclude the self-heating effects. Their observations of the giant nonlinear conduction only appearing below 30 K, well below *T*<sup>c</sup> ≈ 180K, questioned the above scenario of a simultaneous onset of a CDW and a ferromagnetic transition. Instead of the sliding motion of CDW, they proposed an interplay between two different bands is likely the origin of the nonlinear conduction observed in BaIrO3 [51]. Such discrepancy might arise from the different sample quality. Nevertheless, a clear gap opening is unambiguously evidenced at *T*c by other experimental probes, such as the Seebeck coefficient [51, 58] and the high-resolution photoemission spectroscopy [59]. Currently, it remains elusive whether the gap opening is driven by the magnetic order or the Fermi surface nesting.

208 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The observation of weak ferromagnetism and insulating ground state in the 9R BaIrO3 has attracted renewed interest in recent years in light of the SOC-driven Mott insulating state for iridates. As for the nature of the weak ferromagnetism, there also exist long-standing discrep‐ ancies. Experimentally, a tiny Ir moment of ~0.03 μB/Ir was observed below *T*c. In addition, a modified Curie-Weiss fitting to the inverse susceptibility also evidenced a small effective moment of ~ 0.13 μB [18]. Originally, a spin canting from a localized full-moment antiferro‐ magnetic configuration had been invoked to explain the tiny ordered moment [17]. In contrast, Cao et al. [18] proposed a model of band magnetism with intrinsically small Ir moment due to d–p hybridization and small exchange splitting. Indeed, the muon-spin relaxation meas‐ urements by Brooks et al. [60] provided direct experimental evidences in support of a small Ir moment, i.e. they observed clear oscillations below Tc and found an extremely small internal field at the muon site. Such an itinerant picture of band magnetism, however, is incompatible with the observation of high coercive force and anisotropy in magnetization measurements. By employing the X-ray absorption spectroscopy (XAS) and X-ray magnetic circular dichroism (XMCD) techniques, Laguna-Marco et al. [52] recently elucidated an atomic-like nature of the Ir moment with the orbital moment being ~1.5 times larger than the spin moment, thus highlighting the importance of SOC in addressing the magnetic order of 9R BaIrO3. After taking into account both SOC and moderate on-site coulomb interactions, first-principles calculations by Ju et al. [53] identified 9R BaIrO3 as an exotic spin-orbit Mott insulator with multiple Jeff = 1/2 states associated with the unique face-sharing Ir3O12 octahedral units within the structure.

Although the atomic-like nature of Ir local moment in 9R BaIrO3 was found to be extremely stable against temperature, pressure, and chemical substitutions [52, 61], these external stimuli

**Figure 10.** (a) Temperature dependence of resistivity for 9R BaIrO3 for two major crystallographic directions. The first in‐ set shows details of c-axis conductivity and the second the sharp peak in dlnρ/d(1/*T*) denotes the onset of ferromagnet‐ ism. (b) Field-cooled (FC) ad zero field-cooled (ZFC) magnetization showing the ferromagnetic transition at *T*c. The inset shows isothermal magnetization at several temperatures, illustrating a large hysteresis. Adapted from Reference [18].

can easily lead to a breakdown of the weak ferromagnetism and nonmetallic ground state. For example, Cao et al. [62] grown a series of Sr-doped Ba1-xSrxIrO3 single crystals and found that the chemical pressure applied via Sr doping drastically suppresses *T*c and immediately leads to a non-metal to metal transition at high temperatures. On the other hand, although the application of external pressure of ~4.5 GPa can also quench the weak ferromagnetism as Sr doping, BaIrO3 becomes more insulating under pressure [61]. Such a disparate response of transport and magnetic properties to the chemical and physical pressure has been ascribed to the different compression rates of the lattice parameters *a* and *c* upon Sr doping and external pressure. Interestingly, Korneta et al. [63] found that a dilute rare-earth R3+ doping (~ 4%) of BaIrO3 can also suppress the weak ferromagnetism and lead to a metallic state, whereas the application of modest external pressure readily restores the insulating state. Further studies are needed to clarify whether the weak ferromagnetism is also recovered in the pressureinduced insulating state. All these above results demonstrate a delicate interplay between structural and electronic degrees of freedom in 9R BaIrO3.

**Figure 11.** Temperature dependence of magnetic susceptibility χ(*T*) and its inverse χ–1(*T*) for the BaIrO3 polytypes, il‐ lustrating the evolution of the magnetic ground state. Adapted from Reference [21].

*5H BaIrO3* The 5H phase is a weak ferromagnetic metal with *T*c ≈ 50 K, Figure 11. Cheng et al. [20] and Zhao et al. [56] have characterized this compound through measurements of magnetic susceptibility χ(*T*), resistivity ρ(*T*), thermoelectric power S(*T*), and specific heat C(*T*). A weak ferromagnetic transition at *T*c ≈ 50 K was clearly observed in χ(*T*), and well reflected as a kink in the plots of *ρ* vs *T*, *S*/*T* vs ln*T*, and *C*/*T* vs *T*. *In situ* high-pressure resistivity measurements show that *T*c decreases gradually with pressure, and reaches about 40 K under 1.5 GPa.

*6H BaIrO3* The 6H phase has been independently identified and characterized by Zhao et al. [55] and Cheng et al. [20]. Similar with the 6H SrIrO3, it is an exchange-enhanced paramagnetic metal with non-Fermi-liquid behaviours. Zhao et al. [55] reported that its resistivity *ρ*(*T*) follows a linear *T* dependence below 20 K, whereas a *T*5/3 dependence was observed for *T* < 60 K by Cheng et al. [20]. Such a discrepancy should arise from the polycrystalline nature of the studied samples. As mentioned above, a non-Fermi-liquid *ρ*~*T*3/2 behaviour has also been found in the 6H SrIrO3 single crystals due to the proximity to a ferromagnetic quantum critical point. In order to verify similar situation taking place in 6H BaIrO3, we measured the thermo‐ power *S*(*T*) that is insensitive to grain boundaries. We indeed found a linear relationship in the plot of *S*/*T* versus –ln*T* over a wide temperature range, in strong support of the realization of ferromagnetic quantum critical point [20]. Based on the low-temperature specific heat and magnetic susceptibility, the obtained Sommerfeld–Wilson ratio Rw <sup>=</sup> <sup>π</sup><sup>2</sup> <sup>3</sup> ( kB μB ) 2 *χ*<sup>0</sup> <sup>γ</sup> =2.14(3) provides further evidence for strong electron-electron correlations.

*3C BaIrO3* 3C BaIrO3 is characterized as a Pauli paramagnetic metal with a Fermi-liquid behaviour [21]. Its resistivity *ρ*(*T*) displays a metallic behaviour down to at least 1.8 K and follows the Fermi-liquid behaviour, i.e. *ρ*(*T*) = *ρ*0 + A*T*<sup>2</sup> below 6 K with *ρ*<sup>0</sup> = 0.0584(1) Ω cm and *A* = 8.1(1) μΩ cm K–2, respectively. The magnetic susceptibility χ(*T*) exhibits a nearly temper‐ ature-independent Pauli paramagnetism with a shallow minimum around 85 K as observed around *T*\* ≈ 170 K in Pv SrIrO3. Such an upturn with temperature has been ascribed to the higher-order temperature-dependent term in the Pauli paramagnetism. Low-temperature specific heat C(T) analysis yields an electronic specific-heat coefficient γ = 6.84(6) mJ/mol K<sup>2</sup> and a Debye temperature ΘD = 335 K.

*5H BaIrO3* The 5H phase is a weak ferromagnetic metal with *T*c ≈ 50 K, Figure 11. Cheng et al. [20] and Zhao et al. [56] have characterized this compound through measurements of magnetic susceptibility χ(*T*), resistivity ρ(*T*), thermoelectric power S(*T*), and specific heat C(*T*). A weak ferromagnetic transition at *T*c ≈ 50 K was clearly observed in χ(*T*), and well reflected as a kink in the plots of *ρ* vs *T*, *S*/*T* vs ln*T*, and *C*/*T* vs *T*. *In situ* high-pressure resistivity measurements show that *T*c decreases gradually with pressure, and reaches about 40 K under 1.5 GPa.

**Figure 11.** Temperature dependence of magnetic susceptibility χ(*T*) and its inverse χ–1(*T*) for the BaIrO3 polytypes, il‐

lustrating the evolution of the magnetic ground state. Adapted from Reference [21].

210 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

*6H BaIrO3* The 6H phase has been independently identified and characterized by Zhao et al. [55] and Cheng et al. [20]. Similar with the 6H SrIrO3, it is an exchange-enhanced paramagnetic metal with non-Fermi-liquid behaviours. Zhao et al. [55] reported that its resistivity *ρ*(*T*) follows a linear *T* dependence below 20 K, whereas a *T*5/3 dependence was observed for *T* < 60 K by Cheng et al. [20]. Such a discrepancy should arise from the polycrystalline nature of the studied samples. As mentioned above, a non-Fermi-liquid *ρ*~*T*3/2 behaviour has also been found in the 6H SrIrO3 single crystals due to the proximity to a ferromagnetic quantum critical point. In order to verify similar situation taking place in 6H BaIrO3, we measured the thermo‐ power *S*(*T*) that is insensitive to grain boundaries. We indeed found a linear relationship in the plot of *S*/*T* versus –ln*T* over a wide temperature range, in strong support of the realization of ferromagnetic quantum critical point [20]. Based on the low-temperature specific heat and

**Figure 12.** A schematic phase diagram of the BaIrO3 polytypes showing the evolution of magnetic transition temperau‐ tre *T*c (left) and the electronic specific-heat coeffcieint γ (right) as a function of the corner-to-face sharing octahedral C:F ratio. Adapted from Reference [21].

*Structure*–*property evolutions in the BaIrO3 polytypes* As shown in Figure 9, the major change of the crystal structures for these polytypes can be described by the stacking sequence of IrO6 octahedra that evolves from 9R(CFFCFFCFF) → 5H(CFCCF) → 6H(CCFCCF) → 3C(CCC), where C and F stands for corner- and face-sharing, respectively. Figure 11 displays the temperature dependence of magnetic susceptibility χ(*T*) and its inverse χ–1(*T*) for the BaIrO3 polytypes, illustrating the evolution of the magnetic ground state. Figure 12 shows a schematic phase diagram of the BaIrO3 polytypes. With increasing C:F ratio in the sequence 9R(1:2) → 5H(3:2) → 6H(2:1) → 3C(∞), the ground states of BaIrO3 evolve from a ferromagnetic insulator with *T*<sup>c</sup> = 180 K in the 9R phase to a ferromagnetic metal with *T*<sup>c</sup> = 50 K in the 5H phase, then to an exchange-enhanced paramagnetic metal with non-Fermi-liquid behaviour near a ferromagnetic quantum critical point in the 6H phase, and finally to a Fermi-liquid metal in the 3C phase. Such a structure–physical property evolution has been ascribed to a progressive bandwidth broadening in the sense that the corner-shared arrangement of IrO6 octahedron can facilitate the Ir overlap integral mediated via O-2p orbitals. Since the electronic specific-heat coefficient γ is proportional to the density of states at the Fermi energy, the gradual enhance‐ ment of γ from 9R to 6H phase shown in Figure 12 is in agreement with the bandwidth argument. Both the 6H and 3C phases are metallic. Since the 6H phase is close to a ferromag‐ netic quantum critical point, the γ is much enhanced due to critical fluctuations relative to that in the 3C phase with an even broader bandwidth. From this systematic study on BaIrO3 polytypes, we can understand why the 6H SrIrO3 is a non-Fermi-liquid metal near a ferro‐ magnetic quantum critical point [11]. Moreover, the study on BaIrO3 demonstrated that HPHT synthesis of structurally closely related perovskite polytypes represents an effective approach to fine tune the physical properties of interest via modifying the octahedral arrangement.

## **5. Conclusions**

We have summarized in this chapter the current knowledge on the materials' synthesis, crystal structure, and physical properties of the "113" alkaline-earth iridates AIrO3 (A = Ca, Sr, Ba), which display a rich variety of crystallographic and electronic states that are of great current research interest. For CaIrO3, it can form in either the layered pPv or the orthorhombic Pv structure, and thus serves as an important analogue of MgSiO3 to investigate the Pv/pPv transformation in the Earth's lowermost mantle in geosciences. Corresponding to different crystal structures, their electronic ground states differ sharply: the pPv phase is an antiferro‐ magnetic Mott insulator with *T*N = 110 K while the Pv phase is a paramagnetic semimetal with possible Dirac nodes protected by the lattice symmetry. The presence of strong structural distortion in pPv CaIrO3 makes it a model system to investigate the interplay of non-cubic crystal field and SOC in iridates. On the other hand, metallization of the pPv phase via electron or hole doping represents an important approach to realize the exotic electronic states on the verge of insulator–metal transition. For SrIrO3, it crystallizes in the 6H polytype at ambient pressure and transforms to the orthorhombic Pv structure under high-pressure conditions. The 6H phase is an exchange enhanced paramagnetic metal with non-Fermi-liquid behaviour due to the proximity of ferromagnetic quantum critical point, while the Pv phase is revealed as an exotic narrow-band semimetal with symmetry-protected Dirac nodes within the Jeff = 1/2 band near the Fermi level. The presence of nontrivial features in the low energy electronic states makes these "113" orthorhombic Pv iridates AIrO3 (A = Ca, Sr) promising candidates for realizing various topological and magnetic insulating phases via tuning the SOC, Hubbard interactions, and/or lattice symmetry. In this regard, epitaxial growth of superlattices and highpressure synthesis of bulk materials with proper chemical design are currently important approaches to tailor the proposed quantum phases out of the semimetallic state. For BaIrO3, it adopts a nine-layer 9R polytype at ambient pressure, and can be transformed to 5H, 6H, and 3C phases under different high pressure conditions. The 9R phase is a weak ferromagnetic insulator with *T*<sup>c</sup> = 180 K, and can be regarded as an exotic SOC Mott insulator with multiple Jeff = 1/2 states associated with the unique Ir3O12 structural units. An atomic-like nature of the Ir moment driven by strong SOC is rather stable against external perturbations, but the weak ferromagnetism can be easily suppressed by applying chemical and physical pressures or dilute rare-earth substitutions for Ba2+. In contrast, the nonmetallic ground state displays distinct response to the chemical and physical pressure, highlighting the delicate interplay of crystal structure and electronic degrees of freedom for this quasi-1D compound. With increasing the ratio of corner-to-face sharing octahedra in the sequence 9R(1:2) → 5H(3:2) → 6H(2:1) → 3C(∞), the ground states of BaIrO3 evolve from a ferromagnetic insulator with *T*c = 180 K in the 9R phase to a ferromagnetic metal with *T*c = 50 K in the 5H phase, then to an exchange-enhanced paramagnetic metal with non-Fermi-liquid behaviour near a ferromag‐ netic quantum critical point in the 6H phase, and finally to a Fermi-liquid metal in the 3C phase. Such a structure–physical property evolution demonstrated that HPHT synthesis of structur‐ ally closely related perovskite polytypes represents an effective approach to fine tune the physical properties of interest via modifying the octahedral arrangement.

## **Acknowledgements**

ferromagnetic quantum critical point in the 6H phase, and finally to a Fermi-liquid metal in the 3C phase. Such a structure–physical property evolution has been ascribed to a progressive bandwidth broadening in the sense that the corner-shared arrangement of IrO6 octahedron can facilitate the Ir overlap integral mediated via O-2p orbitals. Since the electronic specific-heat coefficient γ is proportional to the density of states at the Fermi energy, the gradual enhance‐ ment of γ from 9R to 6H phase shown in Figure 12 is in agreement with the bandwidth argument. Both the 6H and 3C phases are metallic. Since the 6H phase is close to a ferromag‐ netic quantum critical point, the γ is much enhanced due to critical fluctuations relative to that in the 3C phase with an even broader bandwidth. From this systematic study on BaIrO3 polytypes, we can understand why the 6H SrIrO3 is a non-Fermi-liquid metal near a ferro‐ magnetic quantum critical point [11]. Moreover, the study on BaIrO3 demonstrated that HPHT synthesis of structurally closely related perovskite polytypes represents an effective approach to fine tune the physical properties of interest via modifying the octahedral arrangement.

212 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

We have summarized in this chapter the current knowledge on the materials' synthesis, crystal structure, and physical properties of the "113" alkaline-earth iridates AIrO3 (A = Ca, Sr, Ba), which display a rich variety of crystallographic and electronic states that are of great current research interest. For CaIrO3, it can form in either the layered pPv or the orthorhombic Pv structure, and thus serves as an important analogue of MgSiO3 to investigate the Pv/pPv transformation in the Earth's lowermost mantle in geosciences. Corresponding to different crystal structures, their electronic ground states differ sharply: the pPv phase is an antiferro‐ magnetic Mott insulator with *T*N = 110 K while the Pv phase is a paramagnetic semimetal with possible Dirac nodes protected by the lattice symmetry. The presence of strong structural distortion in pPv CaIrO3 makes it a model system to investigate the interplay of non-cubic crystal field and SOC in iridates. On the other hand, metallization of the pPv phase via electron or hole doping represents an important approach to realize the exotic electronic states on the verge of insulator–metal transition. For SrIrO3, it crystallizes in the 6H polytype at ambient pressure and transforms to the orthorhombic Pv structure under high-pressure conditions. The 6H phase is an exchange enhanced paramagnetic metal with non-Fermi-liquid behaviour due to the proximity of ferromagnetic quantum critical point, while the Pv phase is revealed as an exotic narrow-band semimetal with symmetry-protected Dirac nodes within the Jeff = 1/2 band near the Fermi level. The presence of nontrivial features in the low energy electronic states makes these "113" orthorhombic Pv iridates AIrO3 (A = Ca, Sr) promising candidates for realizing various topological and magnetic insulating phases via tuning the SOC, Hubbard interactions, and/or lattice symmetry. In this regard, epitaxial growth of superlattices and highpressure synthesis of bulk materials with proper chemical design are currently important approaches to tailor the proposed quantum phases out of the semimetallic state. For BaIrO3, it adopts a nine-layer 9R polytype at ambient pressure, and can be transformed to 5H, 6H, and 3C phases under different high pressure conditions. The 9R phase is a weak ferromagnetic insulator with *T*<sup>c</sup> = 180 K, and can be regarded as an exotic SOC Mott insulator with multiple Jeff = 1/2 states associated with the unique Ir3O12 structural units. An atomic-like nature of the

**5. Conclusions**

We are grateful to J.-Z. Zhou, J. B. Goodenough, José Alonso, Y. Uwatoko, and M. Akaogi for collaborations on work related to this review. This work is supported by the National Basic Research Program of China (Grant No. 2014CB921500), the National Science Foundation of China (Grant Nos. 11304371, 51402019), and the strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDB07020100).

## **Author details**

Yunqi Cai1 , Yan Li2 and Jinguang Cheng1\*

\*Address all correspondence to: jgcheng@iphy.ac.cn

1 Beijing National Laboratory for Condensed Matter Physics and Institute of Physics, Chinese Academy of Sciences, Beijing, China

2 College of Materials Science and Engineering, Beijing Institute of Petrochemical Technology, Beijing, China

## **References**

[1] Cao G, Bolivar J, McCall S, Crow JE, Guertin RP. Weak ferromagnetism, metal-tononmetal transition, and negative differential resistivity in single-crystal Sr2IrO4. Phys Rev B 1998;57:R11039.


[17] Lindsay R, Strange W, Chamberland BL, Moyer JR. Weak ferromagnetism in BaIrO3. Solid State Comm 1993;86:759–63.

[2] Kim BJ, Jin H, Moo SJ, Kim JY, Park BG, Leem CS, Yu J, Noh TW, Kim C, Oh SJ, Park JH, Durairaj V, Cao G, Rotenberg E. Novel Jeff=1/2 Mott state induced by relativistic

[3] Kim BJ, Ohsumi H, Komesu T, Sakai F, Morita T, Takagi H, Arima T. Phase sensitive

[4] Jackeli G, Khaliullin G. Mott insulators in the strong spin-orbit coupling limit: from Heisenberg to a quantum compass and Kitaev models. Phys Rev Lett

[5] Witczak-Krempa W, Chen G, Kim YB, Balents L. Correlated quantum phenomena in the strong spin-orbit regime. Annu Rev Condens Matter Phys 2014;5:57–82.

[6] Cao G, Delong LE. Frontiers of 4d- and 5d-Transition Metal Oxides. World Scientific

[7] Moon SJ, Jin H, Kim KW, Choi WS, Lee YS, Yu J, Cao G, Sumi A, Funakubo H, Bern‐ hard C, Noh TW. Dimensionality-controlled insulator-metal transition and correlated metallic state in 5d transition metal oxides Srn+1IrnO3n+1 (n = 1, 2, and infinite). Phys

[8] Longo JM, Kafalas JA, Arnott RJ. Structure and properties of the high and low pres‐

[9] Kim YK, Sumi A, Takahashi K, Yokoyama S, Ito S, Watanabe T, Akiyama K, Kaneko S, Saito K, Funakubo H. Metalorganic chemical vapour deposition of epitaxial perov‐

[10] Carter JM, Vijay Shankar V, Ahsan Zeb M, Kee HY. Semimetal and topological insu‐

[11] Cao G, Durairaj V, Chikara S, DeLong LE, Parkin S, Schlottmann P. Non-Fermi-liq‐ uid behaviour in nearly ferromagnetic SrIrO3 single crystal. Phys Rev B

[12] Sarkozy RF, Moeller CW, Chamberland BL. The characterization of calcium iridium

[13] Ohgushi K, Yagi T, Gotou H, Kiuchi Y, Ueda Y. Metallization of quasi-two-dimen‐ sional Mott insulator CaIrO3 with S = 1/2 spins. Physica B 2009;404:3261–3.

[14] Ohgushi K, Yamaura J, Ohsumi H, Sugimoto K, Takeshita S, Tokuda A, Takagi H, Takata M, Arima T. Resonant X-ray diffraction study of the strongly spin-orbit-cou‐

[15] Moretti Sala M, Ohgushi K, Al-Zein A, Hirata Y, Monaco G, Krisch M. CaIrO3: a spinorbit Mott insulator beyond the jeff=1/2 ground state. Phys Rev Lett 2014;112:176402.

[16] Shim S-H. The postperovskite transition. Annu Rev Earth Planet Sci 2008;36:569.

skite SrIrO3 films on (100) SrTiO3 substrates. J Appl Phys 2006;45:L36–8.

sure forms of SrIrO3. J Solid State Chem 1971;3:174–9.

lator in perovskite iridates. Phys Rev B 2012;85:115105.

pled Mott insulator CaIrO3. Phys Rev Lett 2013;110:217212.

oxides. J Solid State Chem 1974;9:242–6.

observation of a spin-orbital Mott state in Sr2IrO4. Science 2009;323:1329.

spin-orbit coupling in Sr2IrO4. Phys Rev Lett 2008;101:076402.

214 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

2009;102:017205.

Publishing Co. Pte. Ltd.; 2013.

Rev Lett 2008;101:226402.

2007;76:100402(R).


[45] Matsuno J, Ihara K, Yamamura S, Wadati H, Ishii K, Vijay Shankar V, Kee HY, Taka‐ gi H. Engineering spin-orbital magnetic insulator by tailoring superlattice. ArXiv: 1401.1066. 2014.

[31] Kojitani H, Furukawa A, Akaogi M. Thermochemistry and high-pressure equilibria of the post-perovskite phase transition in CaIrO3. Am Miner 2007;92:229–32.

[32] Jang SY, Kim H, Moo SJ, Choi WS, Jeon BC, Yu J, Noh TW. The electronic structure of epitaxially stabilized 5d perovskite Ca1-xSrxIrO3 (x = 0, 0.5, and 1) thin films: the role

[33] Subedi A. First-principles study of the electronic structure and magnetism of CaIrO3.

[34] Bagdanov NA, Katukuri VM, Stoll H, Brink J, Hozoi L. Post-perovskite CaIrO3: A j=1/2 quasi-one-dimensional antiferromagnet. Phys Rev B 2012;85:235147.

[35] Gunasekera J, Chen Y, Kremenak JW, Miceli PF, Singh DK. Mott insulator-to-metal transition in yttrium doped CaIrO3. J Phys: Condens Matter 2015;27:052201.

[36] Kojitani H, Shirako Y, Akaogi M. Post-perovskite phase transition in CaRuO3. Phys

[37] Yamaura K, Shirako Y, Kojitani H, Arai M, Young DP, Akaogi M, Nakashima M, Kat‐ sumata T, Inaguma Y, Takayama-Muromachi E. Synthesis and magnetic and chargetransport properties of the correlated 4d post-perovskite CaRhO3. J Am Chem Soc

[38] Shirako Y, Satsukawa H, Wang XX, Li JJ, Guo YF, Arai M, Yamaura K, Yoshida M, Kojitani H, Katsumata T, Inaguma Y, Hiraki K, Takahashi T, Akaogi M. Integer spinchain antiferromagnetism of the 4d oxide CaRuO3 with post-perovskite structure.

[39] Zhong GH, Li YL, Liu Z, Zeng Z. Ground state properties of perovskite and post-per‐ ovskite CaRuO3: ferromagnetism reduction. Solid State Sci 2010;12:2003–9.

[40] Keawpark N, Tu R, Goto T. Thermoelectricitiy of CaIrO3 ceramics prepared by spark

[41] Ahsan Zeb M, Kee HY. Interplay between spin-orbit coupling and Hubbard interac‐ tions in SrIrO3 and related Pbnn perovskite oxides. Phys Rev B 2012;86:085149. [42] Nie YF, King PDC, Kim CH, Uchida M, Wei HI, Faeth BD, Ruf JP, Ruff JPC, Xie L, Pan X, Fennie CJ, Schlom DG, Shen KM. Interplay of spin-orbit interactions, dimen‐ sionality, and octahedral rotations in semimetallic SrIrO3. Phys Rev Lett

[43] Zhao JG, Yang LX, Yu Y, Li FY, Ru RC, Fang Z, Chen LC, Jin CQ. High-pressure syn‐ thesis of orthorhombic SrIrO3 perovskite and its positive magnetoresistance. J Appl

[44] Blanchard PER, Reynolds E, Kennedy BJ, Kimpton JA, Avdeev M, Belik AA. Anoma‐ lous thermal expansion in orthorhombic perovskite SrIrO3: interplay between spin-

orbit coupling and the crystal lattice. Phys Rev B 2014;89:214106.

of strong spin-orbit coupling. J Phys: Condens Matter 2010;22:485602.

Phys Rev B 2012;85:020408(R).

216 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Earth Planet Inter 2007;165:127–34.

2009;131:2722–6.

2015;114:016401.

Phys 2008;103:103706.

Phys Rev B 2011;83:174411.

plasma sintering. J Cera Soc Jpn 2009;117(4):466–9.


**Perovskite Materials: Properties**

[59] Maiti K, Singh RS, Medicherla VRR, Rayaprol S, Sampathkumaran EV. Origin of charge density wave formation in insulators from high resolution photoemission

[60] Brooks ML, Blundell SJ, Lancaster T, Hayes W, Pratt FL, Frampton PPC, Battle PD. Unconventional magnetic properties of the weakly ferromagnetic metal BaIrO3. Phys

[61] Laguna-Marco MA, Fabbris G, Souza-Neto N, Chikara S, Schilling JS, Cao G, Haskel D. Different response of transport and magnetic properties of BaIrO3 to chemical and

[62] Cao G, Lin XN, Chikara S, Durairaj V, Elhami E. High-temperature weak ferromag‐ netism on the verge of a metallic state: impact of dilute Sr doping on BaIrO3. Phys

[63] Korneta OB, Chikara S, Parkin S, DeLong LE, Schlottmann P, Cao G. Pressure-in‐ duced insulating state in Ba1-xRxIrO3 (R=Gd, Eu) single crystals. Phys Rev B

study of BaIrO3. Phys Rev Lett 2005;95:016404.

218 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

physical pressure. Phys Rev B 2014;90:014419.

Rev B 2005;71:220411(R).

Rev B 2004;69:174418.

2010;81:045101.

## **Metal–Insulator Transitions and Non-Fermi Liquid Behaviors in 5d Perovskite Iridates**

Abhijit Biswas, Ki-Seok Kim and Yoon Hee Jeong

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61285

## **Abstract**

Transition metal oxides, in particular, 3d or 4d perovskites, have provided diverse emer‐ gent physics that originates from the coupling of various degrees of freedom such as spin, lattice, charge, orbital, and also disorder. 5d perovskites form a distinct class be‐ cause they have strong spin-orbit coupling that introduces to the system an additional en‐ ergy scale that is comparable to bandwidth and Coulomb correlation. Consequent new physics includes novel Jeff = 1/2 Mott insulators, metal–insulator transitions, spin liquids, and topological insulators. After highlighting some of the phenomena appearing in the Ruddlesden–Popper iridate series Srn+1IrnO3n+1 (n = 1, 2, and ∞), we focus on the transport properties of perovskite SrIrO3. Using epitaxial thin films on various substrates, we dem‐ onstrate that metal–insulator transitions can be induced in perovskite SrIrO3 by reducing its thickness or by imposing compressive strain. The metal–insulator transition driven by thickness reduction is due to disorder, but the metal–insulator transition driven by com‐ pressive strain is accompanied by peculiar non-Fermi liquid behaviors, possibly due to the delicate interplay between correlation, disorder, and spin-orbit coupling. We examine various theoretical frameworks to understand the non-Fermi liquid physics and metal– insulator transition that occurs in SrIrO3 and offer the Mott–Anderson–Griffiths scenario as a possible solution.

**Keywords:** Transition metal oxides, iridates, metal–insulator transitions, non-Fermi liq‐ uid behaviors, localization

## **1. Introduction**

Transition metal oxides (TMOs), ranging from simple binary oxides to more complex ternary or quaternary compounds, have been a subject of intense activities in condensed matter physics and materials science [1–5]. Of various transition metal oxides, perovskites and their variations are a particularly important class because they display a rich spectrum of various competing

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

phases and physical properties. Structurally, in perovskites with the chemical formula ABO3, large rare-earth or alkali-metal cations occupy the A-sites (tetrahedral oxygen interstices) and transition metal cations occupy the B-sites (octahedral oxygen interstices). Each transition metal ion is thus surrounded by six oxygen anions to form a BO6 octahedron, and the octahedra are connected to each other three-dimensionally. This metal–oxygen network transition mostly determines the electronic properties, while the close-packed A cations and oxygen anions provide structural stability. One characteristic feature of the perovskite family is that numer‐ ous transition metal ions of different size and valence are allowed in the same structure. Along with this elemental diversity, mutually interacting quantum degrees of freedom such as lattice, spin, charge, and orbital lead to numerous emergent physical properties in perovskites. For example, the conducting properties of perovskites range from insulating to semiconducting to metallic and finally to superconducting. The electrical properties of perovskites are equally diverse; they also display various magnetic and multifunctional properties [6–14]. These unparalleled varieties in the electronic properties of perovskites hold tremendous application potential and thus provide a continuous impetus for research in the field.

The network of BO6 octahedra in transition metal perovskites strongly influences their electronic properties because *d* orbitals form bands with characteristic effective masses and different strengths of on-site Coulomb interactions. Indeed, diverse phenomena have been treated in terms of various relevant energy scales such as bandwidth (*W*) and Coulomb repulsion (*U*), and the relevant efforts can be termed *d* orbital physics [15,16]. Disorder (*D),* which is unavoidable in real solids also, has an important influence on the physics of transition metal oxides.

In materials science, numerous studies of 3*d-* and 4*d*-based transition metal perovskites have been conducted to find ways to exploit their functionalities that depend sensitively on structural distortion and crystal chemistry. However, perovskite studies have been limited mainly to those that contain 3*d* or 4*d* transition metal elements. In this regard, 5*d* perovskites form a special class because the strong spin-orbit coupling (SOC, *Λ*) introduces to the system an additional energy scale that is comparable to bandwidth and Coulomb correlation [17]. For example, metal oxide compounds containing 5*d* iridium (e.g., Sr2IrO4, Sr3Ir2O7, SrIrO3, BaIrO3, Na2IrO3, Eu2Ir2O7, Na4Ir3O8, and Sr2GdIrO6) display a variety of emerging properties (Table 1). Of the various exotic properties of 5*d* perovskites, the transport properties and, in particular, the metal–insulator transitions (MIT) in iridate perovskites are the subject of the present chapter.

Simple extrapolation of the characteristics of 3*d* or 4*d* systems fails to predict the characteristics of 5*d* systems. Moving down the periodic table from 3*d* → 4*d* → 5*d*, the orbitals in the solids that contain the corresponding *d* orbitals become increasingly extended and so does the bandwidth (*W*3*<sup>d</sup>* < *W*4*<sup>d</sup>* < *W*5*<sup>d</sup>*). Along with an increase in the bandwidth, the associated on-site Coulomb repulsion would decrease correspondingly (*U*3*<sup>d</sup>* > *U*4*<sup>d</sup>* > *U*5*<sup>d</sup>*). From these considera‐ tions, one may expect higher metallicity and less magnetic instability for the materials with more extended 5*d* orbitals, compared with the systems containing 3*d* or 4*d* elements, because the Stoner criterion in these compounds favors paramagnetic metallic states. Surprisingly, however, the properties of the 5*d*-based transition metal oxides exhibit extremely rich behav‐ iors (Table 1). The Ruddlesden–Popper (RP) series of iridium-based Sr*n*+1Ir*n*O3*n*+1 (*n* = 1, 2, and ∞) as well as other structural families such as pyrochlores, kagome-type lattices, honeycombtype structures, and double perovskites are mostly insulating (few are metallic) but may have exotic physical properties, including novel Mott insulator, lattice-driven magnetoresistance, giant magnetocaloric effect, quantum criticality, charge density wave, geometrically frustrated magnetism, possible topological insulator, and Weyl semimetal [17–19]. These diverse behaviors in iridates require that a new area of physics must be considered; this is SOC (*Λ*3*<sup>d</sup>* < *Λ*4*<sup>d</sup>* < *Λ*5*<sup>d</sup>*). In iridates, the energy scales of *W, U,* and *Λ* are all comparable and thus compete with each other. In addition, *D* is also important, and the emergent properties are the results of strong interplay among Coulomb interaction, SOC, and disorder.

phases and physical properties. Structurally, in perovskites with the chemical formula ABO3, large rare-earth or alkali-metal cations occupy the A-sites (tetrahedral oxygen interstices) and transition metal cations occupy the B-sites (octahedral oxygen interstices). Each transition metal ion is thus surrounded by six oxygen anions to form a BO6 octahedron, and the octahedra are connected to each other three-dimensionally. This metal–oxygen network transition mostly determines the electronic properties, while the close-packed A cations and oxygen anions provide structural stability. One characteristic feature of the perovskite family is that numer‐ ous transition metal ions of different size and valence are allowed in the same structure. Along with this elemental diversity, mutually interacting quantum degrees of freedom such as lattice, spin, charge, and orbital lead to numerous emergent physical properties in perovskites. For example, the conducting properties of perovskites range from insulating to semiconducting to metallic and finally to superconducting. The electrical properties of perovskites are equally diverse; they also display various magnetic and multifunctional properties [6–14]. These unparalleled varieties in the electronic properties of perovskites hold tremendous application

The network of BO6 octahedra in transition metal perovskites strongly influences their electronic properties because *d* orbitals form bands with characteristic effective masses and different strengths of on-site Coulomb interactions. Indeed, diverse phenomena have been treated in terms of various relevant energy scales such as bandwidth (*W*) and Coulomb repulsion (*U*), and the relevant efforts can be termed *d* orbital physics [15,16]. Disorder (*D),* which is unavoidable in real solids also, has an important influence on the physics of transition

In materials science, numerous studies of 3*d-* and 4*d*-based transition metal perovskites have been conducted to find ways to exploit their functionalities that depend sensitively on structural distortion and crystal chemistry. However, perovskite studies have been limited mainly to those that contain 3*d* or 4*d* transition metal elements. In this regard, 5*d* perovskites form a special class because the strong spin-orbit coupling (SOC, *Λ*) introduces to the system an additional energy scale that is comparable to bandwidth and Coulomb correlation [17]. For example, metal oxide compounds containing 5*d* iridium (e.g., Sr2IrO4, Sr3Ir2O7, SrIrO3, BaIrO3, Na2IrO3, Eu2Ir2O7, Na4Ir3O8, and Sr2GdIrO6) display a variety of emerging properties (Table 1). Of the various exotic properties of 5*d* perovskites, the transport properties and, in particular, the metal–insulator transitions (MIT) in iridate perovskites are the subject of the present

Simple extrapolation of the characteristics of 3*d* or 4*d* systems fails to predict the characteristics of 5*d* systems. Moving down the periodic table from 3*d* → 4*d* → 5*d*, the orbitals in the solids that contain the corresponding *d* orbitals become increasingly extended and so does the bandwidth (*W*3*<sup>d</sup>* < *W*4*<sup>d</sup>* < *W*5*<sup>d</sup>*). Along with an increase in the bandwidth, the associated on-site Coulomb repulsion would decrease correspondingly (*U*3*<sup>d</sup>* > *U*4*<sup>d</sup>* > *U*5*<sup>d</sup>*). From these considera‐ tions, one may expect higher metallicity and less magnetic instability for the materials with more extended 5*d* orbitals, compared with the systems containing 3*d* or 4*d* elements, because the Stoner criterion in these compounds favors paramagnetic metallic states. Surprisingly, however, the properties of the 5*d*-based transition metal oxides exhibit extremely rich behav‐

potential and thus provide a continuous impetus for research in the field.

222 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

metal oxides.

chapter.


**Table 1.** 5*d* transition metal iridium (Ir)-based oxides, which are mostly insulators with exotic magnetic states.

Electrical conductivity is one of the properties that used to characterize or to classify solids, so a transition from a metal to insulator or vice versa as a function of a control parameter such as temperature, pressure, or magnetic field has been a topic of interest to the condensed matter physics community for several decades [20–24]. However, despite enormous efforts, under‐ standing of the MIT at the microscopic level is still under debate. In transition metal oxides and strongly correlated systems, the MIT is often accompanied by a change in structural or magnetic symmetry. In contrast, in the presence of a sufficient amount of disorder, MIT is often not associated with any uniform ordering or change in symmetry. Perhaps the most important mechanisms that underlay MITs in correlated systems are interaction-driven or correlationdriven Mott localization, magnetic-order-driven Slater insulator, and disorder-driven Ander‐ son localization. In perovskites that involve heavy 5*d* transition metal elements (e.g., Re, Os, Ir), SOC comes as another parameter and becomes comparable in strength to other relevant energy scales. All of these different parameters in the system compete with each other so their interplay stabilizes new exotic ground states. In this chapter, our foremost aim is to provide a brief description of current research on MITs that occur in Ir-based perovskites. Our emphasis will be on a model system (SrIrO3) to provide experimental evidence that the interplay among correlation, SOC, and disorder is important in achieving different MITs. We also wish to present our understanding about the observed non-Fermi liquid physics and MITs to provide insight and motivation for further activities in this rapidly developing, yet poorly understood, field of strong spin-orbit-coupled 5*d*-based oxide physics.

## **2. Metal–insulator transitions and representative types of insulators**

## **2.1. Metal–insulator transitions**

Electrical resistance *R* (or resistivity *ρ* if intrinsic quantity is used) is a key property that characterizes materials, which are typically classified as metals or insulators. A very good metal (e.g., Cu) can have an electrical resistivity as small as *ρ* ∼10-10 Ω cm, whereas a good electrical insulator (e.g., quartz) has a resistivity as high as *ρ* ∼1020 Ω cm. The causes of this huge difference in resistivity are now well understood in modern solid-state physics. The difference between the resistivities of metals and those of insulators implies that MITs may be accompanied by a large change in *ρ* (up to several orders of magnitude). However, many MITs of current interest are often accompanied not by a large change in magnitude but by a qualitative change in conducting behaviors.

Metals can be distinguished from insulators by their different responses of *R* to temperature *T*. Metals are defined as materials in which *R* decreases as *T* decreases (d*R/*d*T* > 0), whereas insulators are materials in which *R* increases as *T* decreases (d*R/*d*T* < 0) (Figure 1). Strictly, the ultimate distinction between the metal and the insulator can be made at absolute zero; a metal would continue to be conductive, whereas an insulator would lose its conductivity. MITs of fundamental interest change a system from a phase with d*R/*d*T* < 0 to another with d*R/*d*T* > 0, or vice versa.

The conventional theory of electrical transport was first formulated by Drude [25], immedi‐ ately after the discovery of the electron. The semiclassical Drude theory of electronic conduc‐ tivity was built on the idea of the kinetic theory of gases, considering a metal as a gas of electrons. The key concepts in that description are the mean free path *ℓ* (i.e., the average length that an electron travels between successive collisions) and relaxation time *τ* (i.e., the average time between successive collisions). According to the Drude theory, the electronic conductivity *σ* (=<sup>1</sup> *ρ* ) is directly proportional to *τ* as *<sup>σ</sup>* <sup>=</sup> *ne* <sup>2</sup> *τ <sup>m</sup>* , where *n* is the density of electrons, *e* is the electron charge, and *m* is electron mass. For good metals, *ℓ* ∼100 nm and *τ* ~10-14 s. This semiclassical Drude model is still used today as a quick way to estimate a material's property.

driven Mott localization, magnetic-order-driven Slater insulator, and disorder-driven Ander‐ son localization. In perovskites that involve heavy 5*d* transition metal elements (e.g., Re, Os, Ir), SOC comes as another parameter and becomes comparable in strength to other relevant energy scales. All of these different parameters in the system compete with each other so their interplay stabilizes new exotic ground states. In this chapter, our foremost aim is to provide a brief description of current research on MITs that occur in Ir-based perovskites. Our emphasis will be on a model system (SrIrO3) to provide experimental evidence that the interplay among correlation, SOC, and disorder is important in achieving different MITs. We also wish to present our understanding about the observed non-Fermi liquid physics and MITs to provide insight and motivation for further activities in this rapidly developing, yet poorly understood,

**2. Metal–insulator transitions and representative types of insulators**

Electrical resistance *R* (or resistivity *ρ* if intrinsic quantity is used) is a key property that characterizes materials, which are typically classified as metals or insulators. A very good metal (e.g., Cu) can have an electrical resistivity as small as *ρ* ∼10-10 Ω cm, whereas a good electrical insulator (e.g., quartz) has a resistivity as high as *ρ* ∼1020 Ω cm. The causes of this huge difference in resistivity are now well understood in modern solid-state physics. The difference between the resistivities of metals and those of insulators implies that MITs may be accompanied by a large change in *ρ* (up to several orders of magnitude). However, many MITs of current interest are often accompanied not by a large change in magnitude but by a

Metals can be distinguished from insulators by their different responses of *R* to temperature *T*. Metals are defined as materials in which *R* decreases as *T* decreases (d*R/*d*T* > 0), whereas insulators are materials in which *R* increases as *T* decreases (d*R/*d*T* < 0) (Figure 1). Strictly, the ultimate distinction between the metal and the insulator can be made at absolute zero; a metal would continue to be conductive, whereas an insulator would lose its conductivity. MITs of fundamental interest change a system from a phase with d*R/*d*T* < 0 to another with d*R/*d*T* > 0,

The conventional theory of electrical transport was first formulated by Drude [25], immedi‐ ately after the discovery of the electron. The semiclassical Drude theory of electronic conduc‐ tivity was built on the idea of the kinetic theory of gases, considering a metal as a gas of electrons. The key concepts in that description are the mean free path *ℓ* (i.e., the average length that an electron travels between successive collisions) and relaxation time *τ* (i.e., the average time between successive collisions). According to the Drude theory, the electronic conductivity

*τ*

electron charge, and *m* is electron mass. For good metals, *ℓ* ∼100 nm and *τ* ~10-14 s. This semiclassical Drude model is still used today as a quick way to estimate a material's property.

*<sup>m</sup>* , where *n* is the density of electrons, *e* is the

field of strong spin-orbit-coupled 5*d*-based oxide physics.

224 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**2.1. Metal–insulator transitions**

qualitative change in conducting behaviors.

) is directly proportional to *τ* as *<sup>σ</sup>* <sup>=</sup> *ne* <sup>2</sup>

or vice versa.

*σ* (=<sup>1</sup> *ρ*

**Figure 1.** Metals are broadly defined as materials with d*R/*d*T* > 0, whereas insulators are as ones with d*R/*d*T* < 0. *R*: resistance; and *T*: temperature. At absolute zero, the resistance of an insulator would reach infinity with zero conduc‐ tivity while a metal would still possess finite conductivity.

**Figure 2.** Schematic band diagram of metal, semiconductor, and insulator. *E*F, and *E*g are the Fermi energy and band gap. A semiconductor is an insulator with a small energy gap. Upper one: conduction band (CB); lower one: valence band (VB).

Quantum mechanics has been used to clarify the transport properties of solids, and now metals, semiconductors, and insulators are classified according to the band theory of solids (Figure 2). In metals, Fermi energy *E*<sup>F</sup> is within the conduction band, whereas in insulators, *E*<sup>F</sup> separates the electronic band into an empty upper and filled lower band. For example, quartz is an insulator with a band gap *E*g ∼8.9 eV, whereas Cu is a metal. Si is also an insulator but has a small *E*g ∼1.1 eV and thus is called a semiconductor. Although band theory successfully describes the conducting nature of many materials, it fails to account for the behaviors of many transition metal oxides that have relatively narrow separation between conduction and valence band. Many materials with partially filled *d* orbitals and an odd number of electrons per lattice site, which should be metals according to band theory, are actually insulators (e.g., NiO, V2O3, and Fe3O4). The band theory considers single electron and thus may not adequately consider many-body effects such as electron–electron interaction that occur in these complex materials. In addition, *D*, which is unavoidable in real solids, must also be considered. Many theoretical explanations have been proposed to account for the metallic and insulating properties of so-called correlated materials; we briefly summarize these ideas in this section.

## **2.2. Mott insulator**

In 1937, de Boer and Verwey found that many metal–oxides (e.g., NiO, MnO, and FeO) show insulating features despite the fact that these oxides have partially filled *d*-bands [26]. Soon after their discovery, Peierls suggested that strong Coulomb repulsion between electrons could be the origin of the insulating behaviors [27]. He remarked that ''it is quite possible that the electrostatic interaction between the electrons prevents them from moving at all. At low temperatures the majority of the electrons are in their proper places in the ions. The minority which have happened to cross the potential barrier find therefore all the other atoms occupied, and in order to get through the lattice have to spend a long time in ions already occupied by other electrons. This needs a considerable addition of energy and so is extremely improbable at low temperatures'' [20]*.* Peierl's speculation aroused interest in so-called strongly correlated systems. In 1949, Mott offered a theoretical explanation of how electron–electron correlation could yield an insulating state, now called a Mott Insulator [21].

Consider a lattice model with a single electron orbital at each site. If the electron–electron interactions are not considered, a single band would be formed from the overlap of the atomic orbitals: when two electrons, one with spin up (↑ ) and the other with spin down (↓ ), occupy each site, the band becomes full. However, when two electrons occupy the same site, they would feel a large Coulomb repulsion, which can be explained using the Hubbard Hamiltonian (Eq. (1)):

$$H = -t\sum\_{i\mid\sigma} \left( c\_{i\sigma}^{\dagger} c\_{j\sigma} + h.c.\right) + \left. \mathcal{U} \sum\_{i} n\_{i\uparrow} n\_{i\downarrow} - \mu \sum\_{i,\sigma} n\_{i\sigma} \right. \tag{1}$$

where *ci*,*<sup>σ</sup>* † and *<sup>c</sup> <sup>j</sup>*,*<sup>σ</sup>* (*<sup>σ</sup>* <sup>=</sup> <sup>↑</sup>*or* <sup>↓</sup>) are the creation and the annihilation operators, respectively, for electrons on the site *i* and *j* with spin *σ*. Electron hopping between nearest neighbor sites occurs with hopping constant −*t. U* is the amount of energy for each pair of electrons that occupy the same lattice site and represents on-site Coulomb correlation. *ni<sup>σ</sup>* is the number operator and *μ* is the chemical potential.

The ratio between *t* and *U* determines whether the system is a metal or an insulator. When electron–electron correlation is negligible compared to hopping, *t/U* >> 1, the electrons tunnel between the sites without hindrance and the system is metallic. When *t/U* << 1, electron– electron correlation is strong, the electrons are localized due to Coulomb interaction, and the system becomes an insulator. In the Hubbard model, at half filling *ni<sup>σ</sup>* <sup>=</sup> <sup>1</sup> <sup>2</sup> , the Mott–Hubbard insulating phase with one electron per site would appear.

**Figure 3.** (a) Schematic representation of the energy levels for a Mott–Hubbard insulator where on-site Coulomb inter‐ action *U* splits the *d*-band into lower Hubbard band and upper Hubbard band. (b) Evolution of the density of states (DOS) of electrons as a function of *U/W* (*W* = bandwidth) as the system evolves from a metal to an insulator. Adapted with permission from Kotliar and Vollhardt [28]. © 2009, AIP Publishing LLC

In transition metal oxides, often oxygen *p*-bands remain unchanged when interaction (*U)* occurs (Figure 3(a)). In contrast, the *d*-band splits into two subbands (upper Hubbard band and lower Hubbard band), and the half-filled state becomes an insulator with the opening of a charge gap (Figure 3(b)). The Hubbard model allows two types of MIT: the filling-control metal–insulator transition (FC-MIT), which originates from variation of electron concentration or chemical potential (*μ/U*); and a bandwidth-control metal–insulator transition (BC-MIT), which originates from variation of hopping energy or bandwidth (*t/U*).

## **2.3. Slater insulator**

V2O3, and Fe3O4). The band theory considers single electron and thus may not adequately consider many-body effects such as electron–electron interaction that occur in these complex materials. In addition, *D*, which is unavoidable in real solids, must also be considered. Many theoretical explanations have been proposed to account for the metallic and insulating properties of so-called correlated materials; we briefly summarize these ideas in this section.

In 1937, de Boer and Verwey found that many metal–oxides (e.g., NiO, MnO, and FeO) show insulating features despite the fact that these oxides have partially filled *d*-bands [26]. Soon after their discovery, Peierls suggested that strong Coulomb repulsion between electrons could be the origin of the insulating behaviors [27]. He remarked that ''it is quite possible that the electrostatic interaction between the electrons prevents them from moving at all. At low temperatures the majority of the electrons are in their proper places in the ions. The minority which have happened to cross the potential barrier find therefore all the other atoms occupied, and in order to get through the lattice have to spend a long time in ions already occupied by other electrons. This needs a considerable addition of energy and so is extremely improbable at low temperatures'' [20]*.* Peierl's speculation aroused interest in so-called strongly correlated systems. In 1949, Mott offered a theoretical explanation of how electron–electron correlation

Consider a lattice model with a single electron orbital at each site. If the electron–electron interactions are not considered, a single band would be formed from the overlap of the atomic orbitals: when two electrons, one with spin up (↑ ) and the other with spin down (↓ ), occupy each site, the band becomes full. However, when two electrons occupy the same site, they would feel a large Coulomb repulsion, which can be explained using the Hubbard Hamiltonian

> .. *i j i i <sup>i</sup> ij i i H t c c hc U n n n*

† and *<sup>c</sup> <sup>j</sup>*,*<sup>σ</sup>* (*<sup>σ</sup>* <sup>=</sup> <sup>↑</sup>*or* <sup>↓</sup>) are the creation and the annihilation operators, respectively,

for electrons on the site *i* and *j* with spin *σ*. Electron hopping between nearest neighbor sites occurs with hopping constant −*t. U* is the amount of energy for each pair of electrons that occupy the same lattice site and represents on-site Coulomb correlation. *ni<sup>σ</sup>* is the number

The ratio between *t* and *U* determines whether the system is a metal or an insulator. When electron–electron correlation is negligible compared to hopping, *t/U* >> 1, the electrons tunnel between the sites without hindrance and the system is metallic. When *t/U* << 1, electron– electron correlation is strong, the electrons are localized due to Coulomb interaction, and the

,

m¯ =- + + - å åå (1)

> s

s

<sup>2</sup> , the Mott–Hubbard

could yield an insulating state, now called a Mott Insulator [21].

226 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

( ) †

system becomes an insulator. In the Hubbard model, at half filling *ni<sup>σ</sup>* <sup>=</sup> <sup>1</sup>

insulating phase with one electron per site would appear.

s s

s

operator and *μ* is the chemical potential.

**2.2. Mott insulator**

(Eq. (1)):

where *ci*,*<sup>σ</sup>*

Mott insulators in the previous section are electron systems with an odd number of electrons per unit cell; despite the fact that although band theory suggests that these systems would be metallic, they are insulators. Mott insulators belong to the *U >> t* regime in the Hubbard model. However, another aspect must be considered to explain the insulating property even in the Hubbard model. In fact, many insulating systems, especially transition metal oxides (e.g., Na2IrO3 [19]), often have antiferromagnetic ground states. Slater focused on this point. He proposed that formation of spin density waves (long-range magnetic order) due to electron– electron interaction may be an origin of the insulating phase itself [29]. The difference between a Mott insulator and a Slater insulator is that for a Mott insulator, the system remains insulating even above Néel temperature *T*N, whereas a Slater insulator the system should be metallic above *T*N.

The principle behind a magnetic-order-driven Slater insulator may be easily understood from the band picture. Slater insulators belong to the *t >> U* regime, which is opposite to the case for Mott insulators. Suppose for simplicity that the lattice is half filled and thus on average each site holds one electron. A low-energy state for a half-filled system ensues if a periodicity doubling of the lattice occurs (i.e., the Brillouin zone is halved). At the new Brillouin zone boundary, an energy gap for charge excitations occurs and the system becomes insulating. According to Slater, such an insulating behavior is closely connected with the appearance of magnetic order at *T*N, for example, electrons in a bipartite lattice [30], i.e., one that consists of two interpenetrate sublattices A and B in such a way that the nearest neighbors of any site are members of the opposite sublattice, for example, a rock salt arrangement in a simple cubic lattice. Thus, the nearest neighbors of an electron from A are those from B, and vice versa. Because ↑ and ↓ electrons mutually repel due to Coulomb interaction, they become prefer‐ entially arranged on alternating A and B sublattice sites. Hence, spins form a spin-density wave (SDW) whose wave vector is commensurate with the lattice. Because the electrons tend to avoid each other, the potential energy increases, and this gain is balanced by a loss in kinetic energy because of localization of electrons. The spin-density wave provides a necessary periodicitydoubling potential. With an increase in temperature, thermal fluctuations affect the ordering and narrow the energy gap. Eventually, at *T*N, which is typically ~102 K, the ordering is destroyed and the insulating property disappears.

## **2.4. Anderson insulator**

Defects and impurities are unavoidable in real materials, and disorder can never be neglected in reality. Thus, disorder-induced MIT is a subject of continuing interest in condensed matter physics. In 1958, Anderson initiated the field of so-called localization by arguing that suffi‐ ciently strong randomness will localize all the electronic states within a given band and that diffusion may be completely suppressed, thereby leading to an MIT at *T* = 0 K [31]. In the absence of disorder, even a small amount of hopping can delocalize the electrons. However, in the presence of sufficient disorder, the hopping process only spreads an initially localized state over a finite distance; defined as the localization length ξ0. As a result, the states at the band tail become localized and sometimes even a whole band becomes localized. Mott argued that there must exist a critical energy *EC* called mobility edge, which separates the localized states from the extended ones (Figure 4(a)) [21]. If *EF* < *EC*, the system is an insulator; otherwise, it is a metal.

The basic idea behind the localization phenomenon is rather straightforward. Suppose an electron propagates in a disordered medium from point A to point B. To obtain the probability that an electron reaches point B, the probabilities of all possible paths from A to B must be summed. In a disordered medium, the phases of the interference terms may be so random that on average they mutually cancel; the resultant state can be explained by the diffusion model of conductivity, but this simple argument may not be always applicable. Imagine a wave that travels from point A along a random path to point B and then goes back to A. Two possible paths are illustrated (Figure 4(b)): a randomly chosen path and the same path traversed in time reversed sense. The two paths interfere constructively and should be treated coherently as long as time reversal symmetry is preserved. Then the probability of the electron's return to A is twice as large as it would have been if probabilities were added (first squared and then added). The enhanced back-scattering, known as weak localization, reduces the conductance between A and B by increasing the electron's likelihood of returning to its starting point, and eventually leads to localization [32]. At high temperatures, coherence is lost due to thermal vibrations and back-scattering effects diminish. However, at low temperatures, thermal vibrations and inelastic scattering cease and other channels through which electrons can exchange energy become interrupted, so quantum interference immobilizes electrons to induce localization transitions. In 1979, Abrahams et al. developed the phenomenological scaling theory of localization [33, 34]. In metallic systems in two dimensions, resistivity often increases at low temperatures when *T* is decreased. This *σ*2D ln ∝*T* variation is due to weak localization and was first experimentally demonstrated by Dolan and Osheroff [35].

The principle behind a magnetic-order-driven Slater insulator may be easily understood from the band picture. Slater insulators belong to the *t >> U* regime, which is opposite to the case for Mott insulators. Suppose for simplicity that the lattice is half filled and thus on average each site holds one electron. A low-energy state for a half-filled system ensues if a periodicity doubling of the lattice occurs (i.e., the Brillouin zone is halved). At the new Brillouin zone boundary, an energy gap for charge excitations occurs and the system becomes insulating. According to Slater, such an insulating behavior is closely connected with the appearance of magnetic order at *T*N, for example, electrons in a bipartite lattice [30], i.e., one that consists of two interpenetrate sublattices A and B in such a way that the nearest neighbors of any site are members of the opposite sublattice, for example, a rock salt arrangement in a simple cubic lattice. Thus, the nearest neighbors of an electron from A are those from B, and vice versa. Because ↑ and ↓ electrons mutually repel due to Coulomb interaction, they become prefer‐ entially arranged on alternating A and B sublattice sites. Hence, spins form a spin-density wave (SDW) whose wave vector is commensurate with the lattice. Because the electrons tend to avoid each other, the potential energy increases, and this gain is balanced by a loss in kinetic energy because of localization of electrons. The spin-density wave provides a necessary periodicitydoubling potential. With an increase in temperature, thermal fluctuations affect the ordering

and narrow the energy gap. Eventually, at *T*N, which is typically ~102

Defects and impurities are unavoidable in real materials, and disorder can never be neglected in reality. Thus, disorder-induced MIT is a subject of continuing interest in condensed matter physics. In 1958, Anderson initiated the field of so-called localization by arguing that suffi‐ ciently strong randomness will localize all the electronic states within a given band and that diffusion may be completely suppressed, thereby leading to an MIT at *T* = 0 K [31]. In the absence of disorder, even a small amount of hopping can delocalize the electrons. However, in the presence of sufficient disorder, the hopping process only spreads an initially localized state over a finite distance; defined as the localization length ξ0. As a result, the states at the band tail become localized and sometimes even a whole band becomes localized. Mott argued that there must exist a critical energy *EC* called mobility edge, which separates the localized states from the extended ones (Figure 4(a)) [21]. If *EF* < *EC*, the system is an insulator; otherwise,

The basic idea behind the localization phenomenon is rather straightforward. Suppose an electron propagates in a disordered medium from point A to point B. To obtain the probability that an electron reaches point B, the probabilities of all possible paths from A to B must be summed. In a disordered medium, the phases of the interference terms may be so random that on average they mutually cancel; the resultant state can be explained by the diffusion model of conductivity, but this simple argument may not be always applicable. Imagine a wave that travels from point A along a random path to point B and then goes back to A. Two possible paths are illustrated (Figure 4(b)): a randomly chosen path and the same path traversed in time reversed sense. The two paths interfere constructively and should be treated coherently as long

destroyed and the insulating property disappears.

228 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**2.4. Anderson insulator**

it is a metal.

K, the ordering is

**Figure 4.** (a) Concept of the mobility edge *EC*; electronic states > *EC* are extended and those states < *EC* are localized. If the Fermi energy is in the localized state, then the system is insulating; otherwise, it is metallic. DOS: density of states. (b) Propagation of electronic waves in a disordered medium. If two waves follow the same path from A to B and then to A, one in a clockwise direction and the other in a time-reversed direction, then they interfere constructively on re‐ turning to A.

In 1960, Ioffe and Regel realized that the electron mean free path *ℓ* can never be shorter than the lattice spacing *a,* because at *ℓ ≤ a,* coherent quasi-particle motion would vanish and the system would be an insulator [36]. In 1972, Mott proposed a similar argument and formed a criterion *ℓ*min ≈ *a* for minimum metallic conductivity [37]. This criterion was named the Mott– Ioffe–Regel (MIR) limit for resistivity saturation [38, 39]. Unfortunately, the MIR limit is not universally observed and disagrees with the now-widely accepted scaling theory of localiza‐ tion. Nevertheless, the MIR limit is still quoted frequently and is now generalized to more complex media; the criteria ranging from *k*F*ℓ*min ≈ 1 through *ℓ*min ≈ *a* to *k*F*ℓ*min ≈ 2π, where the wave vector *k*F = 2*π/λ*. In some materials MIT occurs at the MIR limit. For two-dimensional materials, *k*F*ℓ* is related to the ratio of quantum resistance *h*/*e*<sup>2</sup> to sheet resistance *R*sheet as *kF <sup>l</sup>* <sup>=</sup> *<sup>h</sup>* <sup>ℓ</sup>*<sup>e</sup>* <sup>2</sup> *<sup>R</sup>*sheet <sup>≈</sup> <sup>26</sup> kΩℓ□ *<sup>R</sup>*sheet [40]*.*

## **3. Recent results on 5***d* **perovskite iridates**

5*d* perovskites, in particular iridates, show many new interesting phenomena; examples include novel insulating states, exotic magnetism, spin-liquid behaviors, Weyl semimetals, and topological insulators. In this regard, RP series Sr*n*+1Ir*n*O3*n*+1 have been widely investigated and are probably the most widely studied iridate system. In this section, we review some of the recent results on 5*d* perovskite iridates.

## **3.1. Perovskites**

The term *perovskite* was named for the discovery of CaTiO3 in honor of Count Lev Alexander Von Perovski, a Russian mineralogist. An ideal perovskite structure has an ABX3 stoichiome‐ try. Although most of the perovskite are oxides (X = Oxygen), other forms like fluorides, halides, and sulfides also exist in literature. Perovskite ABO3 and, more generally, perovskite families including its variations accept a large number of transition metal elements of various size and valence into the B-sites (e.g., A1+ B5+ O3 2− , A2+ B4+ O3 2− , and A3+ B3+ O3 2− ), so the variety of transition metal oxides is unlimited: examples include cuprates, manganites, ruthenates, nickelates, titanates, and recently iridates [41, 42]. Along with this elemental diversity, mutually interacting quantum degrees of freedom such as lattice, spin, charge, and orbital leads to numerous emergent physical properties in perovskite families [1–14]. The perovskite families of current interest, iridates, can be classified as follows:

**Figure 5.** The *n* = 1 (Sr2IrO4), *n* = 2 (Sr3Ir2O7), *n* = 3 (Sr4Ir3O10), and *n* = ∞ (SrIrO3) members of the homologous Ruddles‐ den–Popper series Sr*n*+1Ir*n*O3*n*+1. Structures were drawn using VESTA software.

## **a. Perovskites**

**3. Recent results on 5***d* **perovskite iridates**

230 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

the recent results on 5*d* perovskite iridates.

size and valence into the B-sites (e.g., A1+

families of current interest, iridates, can be classified as follows:

den–Popper series Sr*n*+1Ir*n*O3*n*+1. Structures were drawn using VESTA software.

**3.1. Perovskites**

5*d* perovskites, in particular iridates, show many new interesting phenomena; examples include novel insulating states, exotic magnetism, spin-liquid behaviors, Weyl semimetals, and topological insulators. In this regard, RP series Sr*n*+1Ir*n*O3*n*+1 have been widely investigated and are probably the most widely studied iridate system. In this section, we review some of

The term *perovskite* was named for the discovery of CaTiO3 in honor of Count Lev Alexander Von Perovski, a Russian mineralogist. An ideal perovskite structure has an ABX3 stoichiome‐ try. Although most of the perovskite are oxides (X = Oxygen), other forms like fluorides, halides, and sulfides also exist in literature. Perovskite ABO3 and, more generally, perovskite families including its variations accept a large number of transition metal elements of various

transition metal oxides is unlimited: examples include cuprates, manganites, ruthenates, nickelates, titanates, and recently iridates [41, 42]. Along with this elemental diversity, mutually interacting quantum degrees of freedom such as lattice, spin, charge, and orbital leads to numerous emergent physical properties in perovskite families [1–14]. The perovskite

**Figure 5.** The *n* = 1 (Sr2IrO4), *n* = 2 (Sr3Ir2O7), *n* = 3 (Sr4Ir3O10), and *n* = ∞ (SrIrO3) members of the homologous Ruddles‐

, and A3+

B3+ O3 2−

), so the variety of

B5+ O3 2− , A2+ B4+ O3 2− General formula of ABO3 where A or B are metal cations and O as anions. B cations are surrounded by six oxygen anions forming BO6 octahedra (e.g., SrIrO3) (Figure 5).

## **b. Double perovskites**

General formula of A2BB'O6 (ABO3+AB'O3), where the BO6 and B′O<sup>6</sup> octahedra are alternatively arranged in two sublattices. For double perovskites, alkaline or rare earth ions occupy the Asites, while the B-sites represent transition metal ions (e.g., Sr2RuIrO6 constitutes alternative unit cells of SrRuO3 and SrIrO3).

## **c. Layered perovskites**

Layered perovskites consist of stacked two-dimensional slabs of the ABO3 layer. Three subcategories have been recognized.

## **i. Ruddlesden**–**Popper serie**s

The general formula of Ruddlesden–Popper (RP) series is A*n*+1B*n*O3*n*+1, where *n* represents the number of octahedral layers in the repeating unit and can be visualized as repeated stacking of AO(ABO3)*<sup>n</sup>* [43, 44]. *n* = 1 corresponds to one BO6 octahedron (Sr2IrO4), and *n* = 2 to two BO6 octahedra (Sr3Ir2O7) (Figure 5). The differentiating characteristics for the layered perov‐ skites are the motif (Sr2+) that separates the layers, and the offsetting of the layers from each other. As *n* increases from *n* = 1 to *n* = ∞, the dimensionality of the compounds changes from two to three.

## **ii. Aurivillius phase**

The general formula of the Aurivillius phase is (Bi2O2) (A *n+*1B*n*O3*n*+1) [45]. This structure consists of layers of Bi2O2 separated by *n* layers of perovskites.

## **iii. Dion–Jacobson phases**

The general formula of the Dion**–**Jacobson phase is M(A *n+*1B*n*O3*n*+1), where M is a cation with valence +1, usually an alkali metal [46, 47].

Ideal perovskite ABO3 has symmetric cubic structure with a lattice constant of ∼4 Å. However, most perovskites, deviate from the cubic structure if the Goldschmidt tolerance factor *tf* , given

by *tf* <sup>=</sup> *<sup>r</sup>*<sup>A</sup> <sup>+</sup> *<sup>r</sup>*<sup>O</sup> 2(*r*<sup>B</sup> <sup>+</sup> *<sup>r</sup>*O) (*r*A, *r*B, and *r*O represents the ionic radii of ions A, B, and O, respectively), deviates from 1 [48]. When *tf* <1, (i.e., radius of cation A is small), the O anions move toward the A cation, so BO6 octahedra tilt to shrink the available volume for A cations that is empty. Because B-O-B bonds are highly flexible and BO6 octahedra are flexible in shape and size depending upon the cationic size, valence, and position, the overall deformation reduces the cubic symmetry, and result in structural transitions to orthorhombic, tetragonal, or hexagonal states that have lower symmetry than the cubic state.

## **3.2.** *d* **orbitals**

The critical factor in the orbital physics involving *d* orbitals is their anisotropic charge distri‐ butions that arise from the wave functions, which take different shapes depending on energy when electrons are bound to atomic nuclei by Coulomb force. Electronic properties of perov‐ skites would be severely affected by the chemistry of the transition metals at the center of BO6 octahedra that have corner-sharing oxygen anions [15, 16]. Consider a transition metal element which is surrounded by six O2- in the BO6 octahedron. This configuration gives rise to a crystal field which hinders the free motion of *d*-electrons; consequently, orbital angular momentum is usually quenched, and the *d* orbitals due to the crystal field energy *∆*oct split in energy into (1) *d*x2-y2 and *d*z2 states, which form twofold degenerate *e*g orbitals with higher energy, and (2) threefold degenerate *t*2g orbitals *d*xy, *d*yz, and *d*zx states at lower energy (Figure 6). As the degeneracy of spherical symmetry of an isolated atom is removed, the *d* orbitals begin to fill, starting from a low-energy state and continuing to higher energy states. The actual filling arrangement depends on the competition between the crystal field and on-site exchange interaction described by Hund's rule [49]. For example, for 5*d* perovskite SrIrO3, Ir4+ has five electrons in the *d* orbitals (5*d*<sup>5</sup> ); the electrons distributions from basic viewpoint are illustrated (Figure 7). For transport properties, the band structure would depend sensitively on an overlap between the *d* orbital of the B-site transition metal element and the *p* orbitals of the surrounding oxygen's.

**Figure 6.** Five *d* orbitals in a cubic crystal field which split into two *e*g orbitals and three *t*2g orbitals.

Generally, in transition metal oxides, the electronic properties are further complicated by the interaction of various degrees of freedom surrounding the *d*-electrons, i.e., charge, orbital, spin, and lattice. These degrees of freedom would produce relevant energy scales of similar magnitude such as bandwidth, Coulomb repulsion, and SOC. Because of the large number of transition metals, the number of possible metal-based perovskite transition is enormous. These

**Figure 7.** Simple view of distribution of five *d*-electrons of Ir4+ (*d5* ) in *e*<sup>g</sup> *and t*2g orbitals split by octahedral crystal field energy *∆*oct.

transition metals include those of which the order orbital is 3*d* (e.g., Fe, Co, Ni), 4*d* (e.g., Mo, Ru, Rh), or 5*d* (e.g., W, Re, Ir). 3*d* orbitals are well localized and thus form a narrow band (*W)* with a large on-site Coulomb interaction (*U)*. 4*d* orbitals are spatially more extended than their 3*d* counterparts. As 5*d* orbitals are more spatially extended than 3*d* or 4*d* ones, as a result, nearest-neighbor orbitals overlap significantly, and therefore *W* is wider in 5*d* orbitals than in 3*d* or 4*d* cases, i.e., *W*3*<sup>d</sup>* < *W*4*<sup>d</sup>* < *W*5*<sup>d</sup>*.

Because 5*d* orbitals are extended, on-site Coulomb repulsion or correlation *U* is weaker for 5*d* orbitals than form 3*d* or 4*d* ones, i.e., *U*5*<sup>d</sup>* < *U*4*<sup>d</sup>* < *U*3*<sup>d</sup>*. Thus, naively one would expect 5*d* systems to be more metallic and less magnetic than those based on 3*d* and 4*d* transition metal oxides. Indeed, 5*d* perovskite SrIrO*3* is a correlated paramagnetic metal according to the expectation, but this is an exception. Surprisingly, many other 5*d* perovskites such as Sr2IrO4, and Sr3Ir2O7 are insulators. This unexpected fact can be explained by the high SOC in 5*d* systems. In general, 5*d* systems have larger SOC than do 3*d* or 4*d* systems, i.e., *Λ*3*<sup>d</sup> < Λ*4*<sup>d</sup> < Λ*5*<sup>d</sup>* because SOC is proportional to the fourth power of the atomic number *Z* (i.e., Λsoc<sup>∝</sup> *<sup>Z</sup>* <sup>4</sup> ) [49]. In iridate compounds, *Z* = 77 for Ir, so the SOC strength is very high, even comparable to onsite Coulomb repulsion (∼0.5 eV). This high SOC strength leads to modification of the electronic structure and gives rise to novel emergent phenomena.

## **3.3. Ruddlesden–Popper series Sr***n***+1Ir***n***O3***n***+1**

**3.2.** *d* **orbitals**

electrons in the *d* orbitals (5*d*<sup>5</sup>

oxygen's.

The critical factor in the orbital physics involving *d* orbitals is their anisotropic charge distri‐ butions that arise from the wave functions, which take different shapes depending on energy when electrons are bound to atomic nuclei by Coulomb force. Electronic properties of perov‐ skites would be severely affected by the chemistry of the transition metals at the center of BO6 octahedra that have corner-sharing oxygen anions [15, 16]. Consider a transition metal element which is surrounded by six O2- in the BO6 octahedron. This configuration gives rise to a crystal field which hinders the free motion of *d*-electrons; consequently, orbital angular momentum is usually quenched, and the *d* orbitals due to the crystal field energy *∆*oct split in energy into (1) *d*x2-y2 and *d*z2 states, which form twofold degenerate *e*g orbitals with higher energy, and (2) threefold degenerate *t*2g orbitals *d*xy, *d*yz, and *d*zx states at lower energy (Figure 6). As the degeneracy of spherical symmetry of an isolated atom is removed, the *d* orbitals begin to fill, starting from a low-energy state and continuing to higher energy states. The actual filling arrangement depends on the competition between the crystal field and on-site exchange interaction described by Hund's rule [49]. For example, for 5*d* perovskite SrIrO3, Ir4+ has five

232 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

(Figure 7). For transport properties, the band structure would depend sensitively on an overlap between the *d* orbital of the B-site transition metal element and the *p* orbitals of the surrounding

**Figure 6.** Five *d* orbitals in a cubic crystal field which split into two *e*g orbitals and three *t*2g orbitals.

Generally, in transition metal oxides, the electronic properties are further complicated by the interaction of various degrees of freedom surrounding the *d*-electrons, i.e., charge, orbital, spin, and lattice. These degrees of freedom would produce relevant energy scales of similar magnitude such as bandwidth, Coulomb repulsion, and SOC. Because of the large number of transition metals, the number of possible metal-based perovskite transition is enormous. These

); the electrons distributions from basic viewpoint are illustrated

## *3.3.1. Spin-orbit coupling and band structure evolution*

In this section, we present some unusual phenomena that are observed in RP series Sr*n*+1Ir*n*O3*<sup>n</sup>* +1 (*n* = 1, 2, and ∞). Because the SOC strength is so much larger in Ir than in a typical 3*d* system, SOC contributes to lifting the fivefold degeneracy of the atomic *d*-levels. In crystals, the crystal field and SOC act together to split the *t*2g levels into *J*eff = 3/2 and *J*eff = 1/2 levels (Figure 8). Two equivalent views of the splitting of 5*d* orbitals due to *∆*oct and *Λ*soc are illustrated [17].

Now, considering electron hopping in solids, the *J*eff levels would become bands. Ir4+ has five 5*d* electrons; four of them fill the lower *J*eff = 3/2 band and one partially fills the *J*eff = 1/2 band with the Fermi level in the *J*eff = 1/2 band. The band structure evolves across the RP series Sr*<sup>n</sup>* +1Ir*n*O3*n*+1 (Figure 9). In the *n* = 1 case, Sr2IrO4, SOC splits the *t*2g band into two *J*eff bands, so the

**Figure 8.** Two equivalent views of the splitting of 5*d* orbitals due to crystal field ∆oct and strong SOC *Λ*soc. Horizontal lines in the center column represent the energy levels of the final configuration. The final configuration can be reached from the left side that depicts the crystalline view: crystal field splitting occurs first and then SOC generates *J*eff = 3/2 and 1/2 states. This result can also be viewed from the atomic viewpoint: the atomic fivefold degeneracy is lifted due to SOC, and then the crystal field turns quantum number *J* into *J*eff*.*

bandwidth of the conduction band (*J*eff = 1/2) is effectively reduced and correlation *U* can split the conduction band again into upper Hubbard bands (UHB) and lower Hubbard bands (LHB). Thus, Sr2IrO4 becomes a Mott insulator (Figure 9(a)). As *n* increases in the RP series, the bandwidth of the *J*eff = 1/2 band also increases, possibly due to increase in the coordination number. Sr3Ir2O7, a barely insulator, is the intermediate case (Figure 9(b)). In SrIrO3 with *n* = ∞, *U* cannot split the conduction band with relatively wide bandwidth, so this compound is a correlated metal (Figure 9(c)).

**Figure 9.** Evolution of the band structure across Ruddlesden–Popper series Sr*n*+1Ir*n*O3*n*+1. Schematic diagrams (upper panel) and calculated results (lower panel, LDA+U+SOC). Sr2IrO4 (*n* = 1) is a Mott insulator; SrIrO3 (*n* = ∞) is a correlat‐ ed metal. Barely insulating Sr3Ir2O7 (*n* = 2) is in intermediate. UHB: upper Hubbard; LHB: lower Hubbard band, respec‐ tively. In band structure calculation, the red and dark lines represent the *J*eff =1/2 and *J*eff =3/2 bands. The lower panels are reprinted with permission from Moon et al. [50]. © American Physical Society

## *3.3.2. Sr2IrO4*

bandwidth of the conduction band (*J*eff = 1/2) is effectively reduced and correlation *U* can split the conduction band again into upper Hubbard bands (UHB) and lower Hubbard bands (LHB). Thus, Sr2IrO4 becomes a Mott insulator (Figure 9(a)). As *n* increases in the RP series, the bandwidth of the *J*eff = 1/2 band also increases, possibly due to increase in the coordination number. Sr3Ir2O7, a barely insulator, is the intermediate case (Figure 9(b)). In SrIrO3 with *n* = ∞, *U* cannot split the conduction band with relatively wide bandwidth, so this compound is a

**Figure 8.** Two equivalent views of the splitting of 5*d* orbitals due to crystal field ∆oct and strong SOC *Λ*soc. Horizontal lines in the center column represent the energy levels of the final configuration. The final configuration can be reached from the left side that depicts the crystalline view: crystal field splitting occurs first and then SOC generates *J*eff = 3/2 and 1/2 states. This result can also be viewed from the atomic viewpoint: the atomic fivefold degeneracy is lifted due to

**Figure 9.** Evolution of the band structure across Ruddlesden–Popper series Sr*n*+1Ir*n*O3*n*+1. Schematic diagrams (upper panel) and calculated results (lower panel, LDA+U+SOC). Sr2IrO4 (*n* = 1) is a Mott insulator; SrIrO3 (*n* = ∞) is a correlat‐ ed metal. Barely insulating Sr3Ir2O7 (*n* = 2) is in intermediate. UHB: upper Hubbard; LHB: lower Hubbard band, respec‐ tively. In band structure calculation, the red and dark lines represent the *J*eff =1/2 and *J*eff =3/2 bands. The lower panels

are reprinted with permission from Moon et al. [50]. © American Physical Society

correlated metal (Figure 9(c)).

SOC, and then the crystal field turns quantum number *J* into *J*eff*.*

234 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The first material in the RP series of Sr*n*+1Ir*n*O3*n*+1 is a layered compound Sr2IrO4 (*n* = 1). The crystal structure of Sr2IrO4 consists of alternating layered stacking of SrO-IrO2-SrO perovskite units with ideal K2NiF4-type tetragonal cell (Figure 5). The lattice constants are ∼5.494 Å in the *ab*-plane and ∼25.796 Å along the *c*-axis with space group *I41/acd* [51]. Most importantly, it has a highly insulating nature (Figure 10(a)) [52] in contrast to a naive expectation that the extended nature of 5*d* orbitals would lead to a significant overlap of the nearest-neighbor orbitals and thus a broad electronic bandwidth. This insulating state arises from strong SOC and subse‐ quent Coulomb repulsion (Figure 9(a)) and is termed a "Mott *J*eff = 1/2 insulating state" of the 5*d* electron system [53, 54]. Magnetically, Sr2IrO4 exhibits weak ferromagnetism (canted antiferromagnetism) with *T*N ~ 240 K (Figure 10(b)) with very small ferromagnetic moment (0.023 *μB*/Ir) [52]. The Mott insulating state due to strong SOC was first confirmed by Kim et al. by x-ray absorption (Figure 10(c)), based on the selection rules associated with the 2*p* to 5*d* transitions. Kim et al. also used theoretical calculations to identify the unusual nature of the Ir4+ state in accord with [53]. Angle Resolved Photoemission Spectroscopy (ARPES) (Figure 10(d)) also confirmed the Mott *J*eff = 1/2 state [55].

Considering the origin of the Mott *J*eff = 1/2 insulating state, it should be very sensitive to external perturbations; if so, the relevant energy parameters can be tuned to modify the electronic states. In Sr2IrO4 thin films under tensile (compressive) strain, the correlation energy is affected by in-plane lattice strain with increase (decrease) in bandwidth [56]. Sr2IrO4 remains an insulator even under pressure up to 55 GPa [57]; this stability illustrates the robustness of this insulating state. Also, a pressure-induced, fully reversible, giant piezoresistance was detected at room temperature [58]. The electronic band gap could be tuned electrically, and this characteristic demonstrates potential application in next-generation electronic devices [59]. Alkali-metal doping induces a close resemblance of the electronic state to that of hightemperature cuprates and therefore represents a step toward high-temperature superconduc‐ tivity [60]. Many other exciting properties have been observed in this compound, including magnetic structural change, spin-orbit tuned MITs, lattice-driven magnetoresistivity, electrondoped tuned electronic structure, anisotropic magnetoresistance, and excitonic quasi-particle [61–66]. The interplay of crystal field splitting, SOC, and correlation effects in layered Sr2IrO4 determines the 5*d* electronic structure and leads to realization of a completely new class of materials with a novel quantum state.

## *3.3.3. Sr3Ir2O7*

The crystal structure of Sr3Ir2O7 (*n* = 2) in this RP series of Sr*n*+1Ir*n*O3*n*+1, is of tetragonal cell with *a* = 3.896 Å and *c* = 20.879 Å and space group *I4/mmm*[67]. It consists of strongly coupled bilayers of Ir-O octahedra which are separated by Sr-O interlayer (Figure 5). With the increase of number of octahedral layers (*n*), the electronic bands progressively broaden, and in particular, the bandwidth of the *J*eff = 1/2 band increases from 0.48 eV for *n* = 1 to 0.56 eV for *n* = 2 [68]. Still the transport measurement shows a well-defined, barely insulating *J*eff = 1/2 states (Figure 11(a)) [69]. Theoretical calculations based on LDA + *U* + SOC also provide evidence for the existence of the barely insulating band structure in which Fermi energy is between the *J*eff = 1/2

**Figure 10.** Characteristic properties of Sr2IrO4. (a) Resistivity along *ab-*plane and *c*-axis showing insulating nature with activation energy of *≈*70 meV. (b) Magnetic measurements showing weak ferromagnetism with *T*N ~ 240 K. (c) Black lines: X-ray absorption spectra indicating the presence of Ir *L*<sup>3</sup> (2*p*3/2) and *L*<sup>2</sup> (2*p*1/2) edges around 11.22 and 12.83 keV. Red dots: intensity of the magnetic (1 0 22) peak. (d) Experimental ARPES spectra for Sr2IrO4 (hν = 85 *eV*; *T* = 100 K). *J*eff = 1/2 band (red line) and *J*eff=3/2 band (black line) are shown. Reprinted with permission from Cao et al. [52], Kim et al. [54], and Wojek et al. [55]. © American Physical Society, © AAAS and © IOP Publishing.

bands (Figure 9(b)) [50]. The onset of weak ferromagnetism occurs at *T*<sup>C</sup> ~ 285 K and is closely associated with the rotation of IrO6 octahedra about the *c*-axis. Indeed, the temperature dependence M(*T*) of magnetization closely tracks the rotation of the octahedra, as character‐ ized by Ir-O-Ir bond angle. Sr3Ir2O7 also exhibits an intriguing *M* reversal for in-plane mag‐ netization below 20 K with the onset of a rapid reduction at *T*D ~ 50 K (Figure 11(b)). The barely insulating nature related to splitting of the *J*eff = 1/2 band was again observed with x-ray scattering and absorption (Figure 11(c)) [70] and supported by ARPES measurements (Figure 11(d)) [55].

The ground state of bilayered Sr3Ir2O7 is highly sensitive to small external perturbations such as chemical doping, high pressures, and magnetic field. By replacing Sr2+ with La3+, electrons can be doped into bulk samples and can lead to an insulator-to-metal transition [71]. Also, the application of a high hydrostatic pressure leads to a drastic reduction in the electrical resis‐ tivity; this observation suggests that the system is near an MIT [57]. Although this system has not been fully explored yet, some studies such as resonant inelastic x-ray scattering, scanning tunneling spectroscopy, and optical conductivity have been performed [72–74]. These obser‐ vations indicated that Sr3Ir2O7 is a good model system to explore the mechanism for novel Mott states near an MIT boundary at which competitive interplay between SOC and Coulomb interactions persists.

bands (Figure 9(b)) [50]. The onset of weak ferromagnetism occurs at *T*<sup>C</sup> ~ 285 K and is closely associated with the rotation of IrO6 octahedra about the *c*-axis. Indeed, the temperature dependence M(*T*) of magnetization closely tracks the rotation of the octahedra, as character‐ ized by Ir-O-Ir bond angle. Sr3Ir2O7 also exhibits an intriguing *M* reversal for in-plane mag‐ netization below 20 K with the onset of a rapid reduction at *T*D ~ 50 K (Figure 11(b)). The barely insulating nature related to splitting of the *J*eff = 1/2 band was again observed with x-ray scattering and absorption (Figure 11(c)) [70] and supported by ARPES measurements (Figure

[54], and Wojek et al. [55]. © American Physical Society, © AAAS and © IOP Publishing.

236 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 10.** Characteristic properties of Sr2IrO4. (a) Resistivity along *ab-*plane and *c*-axis showing insulating nature with activation energy of *≈*70 meV. (b) Magnetic measurements showing weak ferromagnetism with *T*N ~ 240 K. (c) Black lines: X-ray absorption spectra indicating the presence of Ir *L*<sup>3</sup> (2*p*3/2) and *L*<sup>2</sup> (2*p*1/2) edges around 11.22 and 12.83 keV. Red dots: intensity of the magnetic (1 0 22) peak. (d) Experimental ARPES spectra for Sr2IrO4 (hν = 85 *eV*; *T* = 100 K). *J*eff = 1/2 band (red line) and *J*eff=3/2 band (black line) are shown. Reprinted with permission from Cao et al. [52], Kim et al.

The ground state of bilayered Sr3Ir2O7 is highly sensitive to small external perturbations such as chemical doping, high pressures, and magnetic field. By replacing Sr2+ with La3+, electrons can be doped into bulk samples and can lead to an insulator-to-metal transition [71]. Also, the application of a high hydrostatic pressure leads to a drastic reduction in the electrical resis‐ tivity; this observation suggests that the system is near an MIT [57]. Although this system has not been fully explored yet, some studies such as resonant inelastic x-ray scattering, scanning tunneling spectroscopy, and optical conductivity have been performed [72–74]. These obser‐

11(d)) [55].

**Figure 11.** Characteristic properties of Sr3Ir2O7. (a) Resistivity as a function of *T* for the basal plane and along the *c-*axis. (b) In-plane magnetization *M* vs. *T* at magnetic field of 100 Oe. (c) Energy scan of (0 1 19) reflection scanned around Ir L3 and L2 resonances. Red dots: scattering intensity; black lines: x-ray absorption spectra. (d) Experimental ARPES spectra (hν = 10.5 *eV*; *T* = 9 K). Calculated band structure: *J*eff = 1/2 (red line) and *J*eff = 3/2 bands (black line). Reprinted with permission from Wokec et al. [55], Cao et al. [69], and Kim et al. [70]. © American Physical Society, and © IOP Publishing.

## *3.3.4. SrIrO3*

The last compound in the RP series (*n* = ∞) is SrIrO3; it is a correlated metal. SrIrO3 is a threedimensional system and has the largest coordination number in the RP series. An increase in the coordination number would lead to the increase of the bandwidth, and thus correlationdriven band splitting would not occur (Figure 9(c)). In fact, in SrIrO3, the bandwidth of the *J*eff = 1/2 band reaches 1.01 eV, so no gap appears within the *J*eff = 1/2 band or between the *J*eff = 1/2 band and *J*eff = 3/2 band [50].

In 1971, Longo et al. first synthesized polycrystalline SrIrO3 [75]. The stable ambient structure of SrIrO3 is a monoclinic distortion of hexagonal BaTiO3 structure [75, 76]. However, the structure transforms to an orthorhombic perovskite (space group *Pnma*) at 40 kbar and *T* = 1000 °C (Table 2). If a perovskite sample obtained at high temperature and high pressure is quenched, it remains a perovskite at room temperature. Because this book is about perovskites, our main focus remains on the properties of perovskite SrIrO3. Zhao et al. [77] and Blanchard et al. [78] again synthesized orthorhombic perovskite samples under high pressure and performed electric and magnetic measurements which showed that perovskite SrIrO3 is truly a paramagnetic metal (Figure 12).


**Table 2.** Bulk SrIrO3 can assume two forms depending on synthesis conditions. Corresponding lattice constants are summarized [75].

In situ ARPES showed that perovskite SrIrO3 is an exotic narrow-band semimetal [79]. The bandwidth was surprisingly narrower than other two-dimensional RP phases, and the semimetallic nature is caused by the unusual coexistence of heavy hole-like and light electronlike bands contrary to the coordination number argument given previously. The observed unusual property may originate from the interplay of strong SOC, dimensionality, and both in- and out-of-plane IrO6 octahedral rotations. Recent theoretical calculations also suggest an extremely interesting possibility that the interplay of the lattice structure and large SOC produces Dirac nodes in the *J*eff = 1/2 band, and engineering topological phases at interfaces and in superlattices would alter the system to be close to a topological crystalline metal [80– 82]. It is obvious that SrIrO3 is an intriguing system in its own right and further studies are warranted.

*3.3.4. SrIrO3*

band and *J*eff = 3/2 band [50].

238 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

a paramagnetic metal (Figure 12).

Atmospheric pressure, *T* = 900 °C

> PO2 = 40 kbar, *T* = 1000 °C

summarized [75].

warranted.

The last compound in the RP series (*n* = ∞) is SrIrO3; it is a correlated metal. SrIrO3 is a threedimensional system and has the largest coordination number in the RP series. An increase in the coordination number would lead to the increase of the bandwidth, and thus correlationdriven band splitting would not occur (Figure 9(c)). In fact, in SrIrO3, the bandwidth of the *J*eff = 1/2 band reaches 1.01 eV, so no gap appears within the *J*eff = 1/2 band or between the *J*eff = 1/2

In 1971, Longo et al. first synthesized polycrystalline SrIrO3 [75]. The stable ambient structure of SrIrO3 is a monoclinic distortion of hexagonal BaTiO3 structure [75, 76]. However, the structure transforms to an orthorhombic perovskite (space group *Pnma*) at 40 kbar and *T* = 1000 °C (Table 2). If a perovskite sample obtained at high temperature and high pressure is quenched, it remains a perovskite at room temperature. Because this book is about perovskites, our main focus remains on the properties of perovskite SrIrO3. Zhao et al. [77] and Blanchard et al. [78] again synthesized orthorhombic perovskite samples under high pressure and performed electric and magnetic measurements which showed that perovskite SrIrO3 is truly

**Synthesis conditions Structure Lattice parameters**

Monoclinic distortion of hexagonal BaTiO3 structure

Orthorhombic

**Table 2.** Bulk SrIrO3 can assume two forms depending on synthesis conditions. Corresponding lattice constants are

In situ ARPES showed that perovskite SrIrO3 is an exotic narrow-band semimetal [79]. The bandwidth was surprisingly narrower than other two-dimensional RP phases, and the semimetallic nature is caused by the unusual coexistence of heavy hole-like and light electronlike bands contrary to the coordination number argument given previously. The observed unusual property may originate from the interplay of strong SOC, dimensionality, and both in- and out-of-plane IrO6 octahedral rotations. Recent theoretical calculations also suggest an extremely interesting possibility that the interplay of the lattice structure and large SOC produces Dirac nodes in the *J*eff = 1/2 band, and engineering topological phases at interfaces and in superlattices would alter the system to be close to a topological crystalline metal [80– 82]. It is obvious that SrIrO3 is an intriguing system in its own right and further studies are

*a* = 5.604 Å *b* = 9.618 Å *c* = 14.17 Å *β* = 93.26°

*a* = 5.60 Å *b* = 5.58 Å *c* = 7.89 Å

**Figure 12.** Characteristic properties of perovskite SrIrO3. (a) Temperature dependence of electrical resistivity measured under different conditions. It remains metallic down to low temperatures. (b) It remains paramagnetic down to low temperatures. Magnetic susceptibility gives no sign of long range ordering. Inset confirms paramagnetism at *T* = 5 K. Reprinted with permission from Blanchard et al. [78]. © American Physical Society.

## **4. Non-Fermi liquid physics and metal–insulator transitions in SrIrO3 films**

Among RP series compounds Sr*n*+1Ir*n*O3*n*+1, the end member SrIrO3 is of particular interest because it is a correlated metal that exhibits unusual electronic transport properties that deviate from the normal Fermi liquid behaviors. The MITs and associated non-Fermi liquid physics that occur in SrIrO3 with strong SOC would provide an opportunity to extend the limit of our current knowledge of the physics of MITs as presented in the previous section. Thus, the present section constitutes the core of this chapter, and we first summarize salient features of the transport properties of SrIrO3, particularly in thin film form. Then we proceed to present the relevant theoretical frameworks to understand the non-Fermi liquid physics that underlay the MITs in the system.

## **4.1. Salient features of the metal–insulator transitions in SrIrO3**

SrIrO3 is believed to be close to an MIT as evidenced by the evolution of the RP series Sr*n*+1Ir*n*O3*<sup>n</sup>* +1 from being an insulator to a correlated metal with increasing *n* = 1 → 2 → ∞. The transport properties of correlated SrIrO3 can be anticipated to be susceptible to external perturbations and MITs, and that the associated unusual properties could be induced in SrIrO3 if, for example, an external stress is applied to the system. As introduced in the earlier section, the two most important mechanisms for MITs in correlated transition metal oxides are correlation-driven Mott localization and disorder-driven Anderson localization. When the system is under variable external stress, one may be able to tune *W, U,* and *D* to some extent and expose the interesting physics that is controlled by the parameters known as effective correlation (*U/W)*  and effective disorder (*D/W)*. For these reasons, we attempted to synthesize perovskite SrIrO3. Perovskite SrIrO3, however, is metastable at room temperature and is obtainable only by applying an elevated pressure (~40 kbar) at high temperature (~1,000 °C) and subsequent quenching. While it is not easy to obtain single crystals of perovskite SrIrO3 due to technical difficulties dealing with high pressures, the crystals can be stabilized by using thin film synthesis to produce them. In this case, the underlying substrates provide compressive strain, which replaces pressure, and epitaxial perovskite SrIrO3 thin films are easily obtained. More importantly, compressive strains can be imposed on SrIrO3 films by choosing substrates with appropriate lattice parameters. By depositing SrIrO3 films on GdScO3, DyScO3, SrTiO3, and NdGaO3, one can impose progressively larger compressive strain in the films (Figure 13). Biswas et al. [83] provides further details on thin-film synthesis.

**Figure 13.** Pseudocubic lattice parameters of SrIrO3 and various substrates available. SrIrO3 and the substrates GdScO3, DyScO3, and NdGaO3 are orthorhombic; SrTiO3 is cubic.

The various scenarios on MITs suggest that an MIT may be obtained by varying the thickness of SrIrO3 films. Indeed in films deposited on GdScO3, which has a lattice constant well matched with that of SrIrO3 (Figure 13), an MIT occurs as the thickness is reduced from 4 nm to 3 nm [83, 84]; a 4-nm film is metallic (with a resistivity upturn at low temperatures), but a 3-nm film is insulating. The resistivity upturn at low temperatures in the 4-nm film is well described by the weak localization in two dimensions, and the insulating behavior of the 3-nm film can be explained by the variable-range hopping. Thus, the thickness-driven MIT for SrIrO3 (Figure 14(a)) is due to disorder and falls in the class of Anderson localization [83]. Reducing the thickness of the film seems to increase the effectiveness of given disorder and of the grain size effect to scatter the charge carriers. The transition from being metallic with a low temperature upturn in resistivity for small disorder to fully insulating over the whole temperature range with the increase of disorder is a realization of the disorder-driven MIT. However, the temperature variation of the resistance of all the metallic films (thickness 4 nm, 10 nm, and 35 nm) follows *ρ* ∝ *T4/5*. This nontrivial exponent indicates that although the MIT itself is driven by disorder, the underlying transport mechanism not so simple. Also, in some materials (e.g., SrVO3), thickness-dependent MIT is caused by a decrease of the coordination number or by an increase in the correlation effect [85].

MIT also occurs in perovskite SrIrO3 thin films when compressive strain is imposed on the films when the thickness is kept constant (Figure 14(b)) [83]. The imposed strain changes the

Metal–Insulator Transitions and Non-Fermi Liquid Behaviors in 5d Perovskite Iridates http://dx.doi.org/10.5772/61285 241

 and effective disorder (*D/W)*. For these reasons, we attempted to synthesize perovskite SrIrO3. Perovskite SrIrO3, however, is metastable at room temperature and is obtainable only by applying an elevated pressure (~40 kbar) at high temperature (~1,000 °C) and subsequent quenching. While it is not easy to obtain single crystals of perovskite SrIrO3 due to technical difficulties dealing with high pressures, the crystals can be stabilized by using thin film synthesis to produce them. In this case, the underlying substrates provide compressive strain, which replaces pressure, and epitaxial perovskite SrIrO3 thin films are easily obtained. More importantly, compressive strains can be imposed on SrIrO3 films by choosing substrates with appropriate lattice parameters. By depositing SrIrO3 films on GdScO3, DyScO3, SrTiO3, and NdGaO3, one can impose progressively larger compressive strain in the films (Figure 13).

**Figure 13.** Pseudocubic lattice parameters of SrIrO3 and various substrates available. SrIrO3 and the substrates GdScO3,

The various scenarios on MITs suggest that an MIT may be obtained by varying the thickness of SrIrO3 films. Indeed in films deposited on GdScO3, which has a lattice constant well matched with that of SrIrO3 (Figure 13), an MIT occurs as the thickness is reduced from 4 nm to 3 nm [83, 84]; a 4-nm film is metallic (with a resistivity upturn at low temperatures), but a 3-nm film is insulating. The resistivity upturn at low temperatures in the 4-nm film is well described by the weak localization in two dimensions, and the insulating behavior of the 3-nm film can be explained by the variable-range hopping. Thus, the thickness-driven MIT for SrIrO3 (Figure 14(a)) is due to disorder and falls in the class of Anderson localization [83]. Reducing the thickness of the film seems to increase the effectiveness of given disorder and of the grain size effect to scatter the charge carriers. The transition from being metallic with a low temperature upturn in resistivity for small disorder to fully insulating over the whole temperature range with the increase of disorder is a realization of the disorder-driven MIT. However, the temperature variation of the resistance of all the metallic films (thickness 4 nm, 10 nm, and 35 nm) follows *ρ* ∝ *T4/5*. This nontrivial exponent indicates that although the MIT itself is driven by disorder, the underlying transport mechanism not so simple. Also, in some materials (e.g., SrVO3), thickness-dependent MIT is caused by a decrease of the coordination number or by

MIT also occurs in perovskite SrIrO3 thin films when compressive strain is imposed on the films when the thickness is kept constant (Figure 14(b)) [83]. The imposed strain changes the

Biswas et al. [83] provides further details on thin-film synthesis.

240 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

DyScO3, and NdGaO3 are orthorhombic; SrTiO3 is cubic.

an increase in the correlation effect [85].

**Figure 14.** Metal–insulator transitions in perovskite SrIrO3. (a) Sheet resistance for films of varying thickness and the thickness-driven metal–insulator transition. Mott–Ioffe–Regel (MIR) limit corresponds to the two-dimensional quan‐ tum sheet resistance (*h* /*e* 2) ∼26 kΩ / □. (b) Resistivity for films of constant thickness 35 nm on various substrates and compressive-strain-driven metal–insulator transitions. The vertical arrow indicates an increase of compressive strain. Small arrows in both figures indicate the temperature below which resistivity upturns show up. Reprinted with permission from Biswas et al. [83]. © 2014, AIP Publishing LLC.

Ir-O bond length *d* and the Ir-O-Ir bond angle *θ*; these changes affect the bandwidth as *<sup>W</sup>* <sup>∝</sup> cos*<sup>ψ</sup> <sup>d</sup>* 3.5 , where *ψ = (π – θ)/2* is the buckling deviation of the *Ir-O-Ir* bond angle *θ* from *π* [83]. *d* is relatively difficult to change, but *θ* can be readily affected by lattice strain. When *θ* is decreased by compressive strain, the electronic hopping integral between Ir 5*d* orbitals is reduced, so the buckling drives the system toward the insulating state. In addition, compres‐ sion would also slightly increase correlation (*U)*. Thus, the overall change in effective corre‐ lation (*U/W)* would seem to induce the MIT; however, as compressive strain increases, the effective disorder (*D/W)* in the system also increase. Disorder was already shown to play a decisive role in the thickness-driven MIT and has a strong influence in the strain-driven MIT, as exposed by optical conductivity measurements [86]. Optical absorption spectroscopy provides insight into electronic band structure and free-carrier dynamics; optical absorption spectroscopy measurements of compressively strained SrIrO3 films showed Drude-like, metallic responses without an optical gap opening. This result indicates that localization has a measurable effect on strain-induced MIT in perovskite SrIrO3 thin films. In fact, extensive transport measurements in the compressively strained SrIrO3 revealed is that the MIT is not simply due to either disorder or correlation [83, 87]. Thus, SrIrO3 seems to provide a rare example in which the interplay of correlation and disorder in the presence of SOC causes the MIT.

The most remarkable feature of the electrical transport of perovskite SrIrO3 under compres‐ sion is that the temperature variation of resistivity deviates from the Fermi liquid behav‐ ior in a peculiar way. For SrIrO3 films of 35-nm thickness on GdScO3, DyScO3, SrTiO3, and NdGaO3 substrates, the electrical resistivity not only shows non-Fermi liquid behaviors (*ρ* ∝ *T<sup>ε</sup>* with *ε* ≠2), but *ε* evolves from 4/5 to 1 to 3/2 as the compressive strain is increased, specifically, *ρ* ∝ *T*4/5 for films on GdScO3, *ρ* ∝ *T* for films on DyScO3, and *ρ* ∝ *T*3/2 for films on SrTiO3 (Figure 15) [83]. Films on NdGaO3 are subjected to the largest strain and become

insulating (Figure 14(b)). The present strain-driven MIT is clearly contrasted to the thicknessdriven MIT, during which *ε* remains constant at 4/5 as the thickness is reduced and the system approaches the MIT [83].

**Figure 15.** Temperature-dependent resistivity of SrIrO3 thin films of thickness 35 nm on various substrates. The films show distinctly non-Fermi liquid behaviors as *ρ* ∝ *T<sup>ε</sup>* with (a) *ε* = 4/5 for films on GdScO3, (b) *ε* = 1 for films on DyScO3, and (c) *ε* = 3/2 for films on SrTiO3. The films on DyScO3 and SrTiO3 show resistivity upturns at low temperatures. Adapted with permission from Biswas et al. [83]. © 2014, AIP Publishing LLC.

The cause of these peculiar non-Fermi liquid behaviors and consequent MIT in the straindriven case is not clear. Arguably, because SrIrO3 is paramagnetic without long-range magnetic ordering, localized states might induce the formation of local magnetic moments without collective magnetic fluctuations because they occur near the MIT. Such localized moments or small magnetic clusters can influence the electronic transport significantly in the presence of disorder. Indeed, the presence of disorder in the strained films is indicated by the increases in resistivity at low temperatures (Figure 14). Also, disorder and the evolution of non-Fermi liquid physics are possibly inter-connected. In a correlated metal, when disorder is sufficiently high, the system can enter a so-called Griffiths phase, which consists of a mixture of islands of Fermi liquid and Mott insulating regions, and that has non-Fermi liquid behaviors. In the following sections, we will try to propose a model that incorporates all these factors and that may explore a new paradigm for non-Fermi liquid physics.

## **4.2. UV physics: Emergence of localized magnetic moments and dynamical mean-field theory**

Essential experimental features for SrIrO3 films of thickness 35-nm grown on GdScO3, DyS‐ cO3, SrTiO3, and NdGaO3 are non-Fermi liquid transport phenomena near MIT, where electrical resistivity *ρ* ∝*T <sup>ε</sup>* shows anomalous temperature dependences with *ε* = 4/5, 1, 3/2, and –1/4 respectively. (The resistivity for the insulating phase can be described with a negative exponent.) The continuous change of the temperature exponent *ε* implies that a particular type of interplay between correlations of electrons and disorders is expected to have an important influence on physics near the MIT and raises fundamental questions about the nature of the non-Fermi liquid. The first question is whether this non-Fermi liquid physics is involved with either UV (ultraviolet) or IR (infrared) physics. Here, UV physics means that localized magnetic moments appear near an MIT and are regarded to be the source of strong inelastic scattering events due to their extensive entropy and to be responsible for non-Fermi liquid transport phenomena. The theoretical framework of dynamical mean-field theory was designed to simulate this local-moment physics quite well [88].

insulating (Figure 14(b)). The present strain-driven MIT is clearly contrasted to the thicknessdriven MIT, during which *ε* remains constant at 4/5 as the thickness is reduced and the

**Figure 15.** Temperature-dependent resistivity of SrIrO3 thin films of thickness 35 nm on various substrates. The films

and (c) *ε* = 3/2 for films on SrTiO3. The films on DyScO3 and SrTiO3 show resistivity upturns at low temperatures.

The cause of these peculiar non-Fermi liquid behaviors and consequent MIT in the straindriven case is not clear. Arguably, because SrIrO3 is paramagnetic without long-range magnetic ordering, localized states might induce the formation of local magnetic moments without collective magnetic fluctuations because they occur near the MIT. Such localized moments or small magnetic clusters can influence the electronic transport significantly in the presence of disorder. Indeed, the presence of disorder in the strained films is indicated by the increases in resistivity at low temperatures (Figure 14). Also, disorder and the evolution of non-Fermi liquid physics are possibly inter-connected. In a correlated metal, when disorder is sufficiently high, the system can enter a so-called Griffiths phase, which consists of a mixture of islands of Fermi liquid and Mott insulating regions, and that has non-Fermi liquid behaviors. In the following sections, we will try to propose a model that incorporates all these factors and

**4.2. UV physics: Emergence of localized magnetic moments and dynamical mean-field**

Essential experimental features for SrIrO3 films of thickness 35-nm grown on GdScO3, DyS‐ cO3, SrTiO3, and NdGaO3 are non-Fermi liquid transport phenomena near MIT, where

and –1/4 respectively. (The resistivity for the insulating phase can be described with a negative exponent.) The continuous change of the temperature exponent *ε* implies that a particular type of interplay between correlations of electrons and disorders is expected to have an important influence on physics near the MIT and raises fundamental questions about the nature of the non-Fermi liquid. The first question is whether this non-Fermi liquid physics is involved with either UV (ultraviolet) or IR (infrared) physics. Here, UV physics means that localized magnetic moments appear near an MIT and are regarded to be the source of strong inelastic scattering

shows anomalous temperature dependences with *ε* = 4/5, 1, 3/2,

with (a) *ε* = 4/5 for films on GdScO3, (b) *ε* = 1 for films on DyScO3,

system approaches the MIT [83].

show distinctly non-Fermi liquid behaviors as *ρ* ∝ *T<sup>ε</sup>*

**theory**

electrical resistivity *ρ* ∝*T <sup>ε</sup>*

Adapted with permission from Biswas et al. [83]. © 2014, AIP Publishing LLC.

242 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

that may explore a new paradigm for non-Fermi liquid physics.

Dynamical mean-field theory describes Mott quantum criticality successfully and suggests that it appears at high temperatures, where not only MITs but also the so-called bad metal physics have been revealed [89, 90]. A noticeable point is that this local-moment UV physics seems to be universal, regardless of IR physics in which such local moments are expected to disappear by forming either singlets or magnetic orders and thereby reducing the entropy dramatically at low temperatures. Because incompletely screened local moments have important effects in non-Fermi liquid physics, the appearance of negative magnetoresistance (MR) (MR = <sup>ρ</sup>(B) <sup>−</sup> <sup>ρ</sup>(0) <sup>ρ</sup>(0) ) can be expected. However, the experiment on 35-nm SrIrO3 films on various substrates confirms positive MR, which is less than 1% up to magnetic field strength of 9 Tesla (Figure 16) [83]. Although this result does not necessarily mean that the local-moment physics may not be important in the MIT of SrIrO3 thin films, the positive MR leads us to focus on IR physics.

**Figure 16.** Magnetoresistance (MR) at *T* = 5 K of SrIrO3 thin films of thickness 35 nm grown on various substrates. MR was positive in all films regardless of compressive strain which determines the magnitude of MR

Another important phenomenon is that spectral weight transfer in optical conductivity from the Drude part to a mid-infrared region as temperature increases [38]. This change is a characteristic feature of bad metals and occurs at finite temperatures in the metallic side near an MIT. Dynamical mean-field theory reveals that local-moment UV physics describes this bad-metal Mott physics nicely. If local-moment physics are not considered, the spectral weight transfer in the optical conductivity cannot be easily reproduced without symmetry breaking. Unfortunately, no experiment has been performed yet in determining the importance of UV local-moment physics on non-Fermi liquid states of SrIrO3 thin films. With the contradictory picture of UV physics and observed positive MR, in the next section, we focus on IR physics, which assumed to be responsible for the observed non-Fermi liquid physics near the MIT of SrIrO3 thin films.

## **4.3. IR physics: Slater quantum criticality vs. Mott–Anderson–Griffiths scenario**

Here, IR physics means that long-wave length and low-energy fluctuations determine the non-Fermi liquid physics near the MIT of SrIrO3 thin films. Then, quantum criticality would be the first choice, where quantum critical fluctuations involved with symmetry breaking, which cause to strong inelastic scattering between low-energy electrons. This scattering is responsible for non-Fermi liquid transport phenomena [91]. Unfortunately, we failed to figure out quantum criticality involved with any kinds of orders, in particular, those associated with magnetism, although we do not exclude more-delicate symmetry breaking near the MIT of SrIrO3 thin films. Furthermore, quantum criticality itself cannot readily explain the continuous change of *ε* in the relationship *ρ* ∝*T <sup>ε</sup>* of electrical resistivity to temperature on the metallic side of the MIT. One way to understand this continuous change is to consider the effect of disorder on quantum criticality. The Harris criterion is on the stability of quantum criticality against weak randomness, in the sense of average, involved with a space dimension and a critical exponent of correlation length [92]. When the Harris criterion is violated, the clean quantum critical point becomes destabilized. As a result, a novel disordered quantum critical point can emerge, respecting the Harris criterion. However, the continuous evolution of the non-Fermi liquid physics is difficult to understand even in this situation. However, the strength of randomness can grow indefinitely, resulting in the so-called infinite randomness fixed point; when this happens, samples become extremely inhomogeneous due to effectively enhanced disorders, with ordered regions coexisting with disordered islands. The statistical distribution of the ordered islands shows a power-law tail, which implies that rare events that correspond to the power-law tail of the distribution function have important influences in non-Fermi liquid physics, particularly thermodynamics [93]. For example, an antiferromagnetic Heisenberg model with random exchange coupling lies at an infinite randomness fixed point, at which the antiferromagnetic quantum critical point disappears and is replaced by a random singlet state [94]. An attractive feature of the infinite randomness fixed point is that non-Fermi liquid physics still survive near it, i.e., away from the quantum critical point, where dynamics of rare regions associated with the long tail part of the distribution function governs singular behaviors of the extremely inhomogeneous state. These phenomena are referred to as quantum Griffith effects [95]. The continuous change of non-Fermi liquid transport exponents reminds us of the quantum Griffith phase.

We suspect that the MIT in 35-nm SrIrO3 films on GdScO3, DyScO3, SrTiO3, and NdGaO3 can be achieved by electron correlations, in which lattice mismatches between substrates and SrIrO3 thin films are expected to control the ratio between interactions and hopping integrals. An essential question is on the role of disorder in this Mott transition, combined with electron correlations. Considering that the films show positive MR at magnetic field strength up to 9 Tesla, we suggest that magnetic correlations may not have important influence in this MIT. In fact, the positive MR suggests that the resulting insulating phase may be paramagnetic.

Furthermore, the residual resistivity is close to the Mott–Ioffe–Regel (MIR) limit; this similarity implies that the concentration of disorders is not low. Therefore, we conjecture that the interplay between electron correlations and disorders can give rise to a Griffith-type phase between Landau's Fermi liquid state and the Mott–Anderson insulating phase; the Griffithtype phase allows the continuous change of transport exponents. We call this physics the Mott– Anderson–Griffith scenario. Until now, the Griffith scenario has been realized near the infinite randomness fixed point [93], at which extreme inhomogeneity and associated rare events are responsible for non-Fermi liquid physics that have varying critical exponents. Although the mechanism by which such an infinite randomness fixed point can appear in the Mott– Anderson transition has not been identified, fluctuations between metallic and insulating islands as rare events are expected to allow development of the Mott–Anderson–Griffith phase.

## **4.4. Model Hamiltonian and tentative Global phase diagram**

which assumed to be responsible for the observed non-Fermi liquid physics near the MIT of

Here, IR physics means that long-wave length and low-energy fluctuations determine the non-Fermi liquid physics near the MIT of SrIrO3 thin films. Then, quantum criticality would be the first choice, where quantum critical fluctuations involved with symmetry breaking, which cause to strong inelastic scattering between low-energy electrons. This scattering is responsible for non-Fermi liquid transport phenomena [91]. Unfortunately, we failed to figure out quantum criticality involved with any kinds of orders, in particular, those associated with magnetism, although we do not exclude more-delicate symmetry breaking near the MIT of SrIrO3 thin films. Furthermore, quantum criticality itself cannot readily explain the continuous

side of the MIT. One way to understand this continuous change is to consider the effect of disorder on quantum criticality. The Harris criterion is on the stability of quantum criticality against weak randomness, in the sense of average, involved with a space dimension and a critical exponent of correlation length [92]. When the Harris criterion is violated, the clean quantum critical point becomes destabilized. As a result, a novel disordered quantum critical point can emerge, respecting the Harris criterion. However, the continuous evolution of the non-Fermi liquid physics is difficult to understand even in this situation. However, the strength of randomness can grow indefinitely, resulting in the so-called infinite randomness fixed point; when this happens, samples become extremely inhomogeneous due to effectively enhanced disorders, with ordered regions coexisting with disordered islands. The statistical distribution of the ordered islands shows a power-law tail, which implies that rare events that correspond to the power-law tail of the distribution function have important influences in non-Fermi liquid physics, particularly thermodynamics [93]. For example, an antiferromagnetic Heisenberg model with random exchange coupling lies at an infinite randomness fixed point, at which the antiferromagnetic quantum critical point disappears and is replaced by a random singlet state [94]. An attractive feature of the infinite randomness fixed point is that non-Fermi liquid physics still survive near it, i.e., away from the quantum critical point, where dynamics of rare regions associated with the long tail part of the distribution function governs singular behaviors of the extremely inhomogeneous state. These phenomena are referred to as quantum Griffith effects [95]. The continuous change of non-Fermi liquid transport exponents reminds

We suspect that the MIT in 35-nm SrIrO3 films on GdScO3, DyScO3, SrTiO3, and NdGaO3 can be achieved by electron correlations, in which lattice mismatches between substrates and SrIrO3 thin films are expected to control the ratio between interactions and hopping integrals. An essential question is on the role of disorder in this Mott transition, combined with electron correlations. Considering that the films show positive MR at magnetic field strength up to 9 Tesla, we suggest that magnetic correlations may not have important influence in this MIT. In fact, the positive MR suggests that the resulting insulating phase may be paramagnetic.

of electrical resistivity to temperature on the metallic

**4.3. IR physics: Slater quantum criticality vs. Mott–Anderson–Griffiths scenario**

244 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

SrIrO3 thin films.

change of *ε* in the relationship *ρ* ∝*T <sup>ε</sup>*

us of the quantum Griffith phase.

Recalling the interplay between the spin-orbit coupling and the Hubbard interaction, we start from an Anderson–Hubbard model with one band (Eq. (2)):

$$H = -t\sum\_{i\mid \sigma} c\_{i\sigma}^{\dagger} c\_{\mid \sigma} + \downarrow \mathcal{U} \sum\_{i} \left(\sum\_{\sigma} c\_{i\sigma}^{\dagger} c\_{i\sigma}\right)^{2} - \sum\_{i} v\_{i} \left(\sum\_{\sigma} c\_{i\sigma}^{\dagger} c\_{i\sigma}\right) \tag{2}$$

where c*<sup>i</sup>*,*<sup>σ</sup>* represents an electron field at site *i*, and *σ* expresses a Kramers doublet state given by total angular momentum, *t* is a hopping integral, *U* is the strength of on-site Coulomb interaction, and *vi* is a random potential introduced by disorder. When electron correlation becomes negligible, the Hamiltonian reduces to the Anderson model, showing a continuous phase transition from a diffusive Fermi liquid state to an Anderson insulating phase in three dimensions. Local density of states (LDOS) occurs (Figure 17), in which localized eigenstates are given by discrete energy levels. This condition occurs below the mobility edge in the diffusive Fermi liquid state, and over the whole range of energy in the insulating phase. When contributions of random impurities can be neglected, the Hamiltonian becomes the Hubbard model, which shows a Mott transition from Landau's Fermi liquid state to a paramagnetic Mott insulating phase. Regarding this Mott transition based on the UV local-moment physics yields first-order MIT [88]. In contrast, using IR physics to explain the Mott transition yields a continuous MIT, which will be discussed below. In the Landau's Fermi liquid state, a coherent peak occurs at the Fermi energy in the electron spectral function besides incoherent humps, and it disappears to transfer into an incoherent background, resulting in upper and lower Hubbard bands in the paramagnetic Mott insulating phase. A further question is whether the Griffith-type phase appears near the Mott–Anderson transition in the middle region of the phase diagram (Figure 17).

As discussed, we try to describe IR physics to explain the Mott–Anderson MIT in SrIrO3 thin films. Incompletely screened local moments can be expected to be screened completely at low temperatures. One mechanism for this screening is the Kondo effect, in which localized magnetic moments form singlets with itinerant electrons, well described by dynamical mean-

**Figure 17.** A schematic phase diagram of Anderson–Hubbard model with one band and local density of states (LDOS) of each phase. Adapted with permission from Byczuk [96]. © 2005 American Physical Society.

field theory. The other mechanism entails localized magnetic moments form singlets with themselves as an RKKY-type (Ruderman, Kittel, Kasuya, Yosida) interactions [97–99]. The resulting paramagnetic Mott insulating phase is called a spin-liquid state, in which chargeneutral spinons may emerge as low-energy elementary excitations and may interact among themselves through abundant singlet fluctuations, referred to as gauge fields [100]. Recalling the positive MR, we discuss the Mott–Anderson transition based on the spin-liquid physics.

## **4.5. Generalization of Finkelstein's nonlinear sigma model approach near the Mott– Anderson transition**

To describe the Mott transition involved with the spin-liquid physics, we take the U(1) slaverotor representation [101] as *Ci<sup>σ</sup>* = *e* <sup>−</sup>*iθ<sup>i</sup> <sup>f</sup> <sup>i</sup>σ*, in which an electron field is expressed as a composite operator of charge and spin degrees of freedom. *θ<sup>i</sup>* accounts for the dynamics of collective charge fluctuations (sound modes), and *f <sup>i</sup>σ* expresses a charge-neutral spinon field for collective dynamics of spin degrees of freedom. Then, an effective mean-field theory for such variables of *θ<sup>i</sup>* and *f <sup>i</sup>σ* can be easily formulated from the Hubbard model without randomness; in this model, the Hubbard Hamiltonian is decomposed into two sectors describing the dynamics of spinons and zero-sound modes, respectively, given by (Eq. (3) and Eq. (4)):

$$S\_F = \int\_0^\theta d\tau \left[ \sum\_{i,\sigma} f\_{i\sigma}^\dagger \left( \partial\_\tau - \mu \right) f\_{i\sigma} - t \,\mathcal{X}\_f \sum\_{i|,\sigma} \left( f\_{i\sigma}^\dagger f\_{i\sigma} + h.c. \right) \right] \tag{3}$$

Metal–Insulator Transitions and Non-Fermi Liquid Behaviors in 5d Perovskite Iridates http://dx.doi.org/10.5772/61285 247

$$\mathcal{S}\_{\mathsf{B}} = \mathop{\rm l\bar{\sigma}}\_{\mathsf{0}} \left[ \frac{1}{2\mathcal{U}} \sum\_{i} \Big( \partial\_{\tau} b\_{i}^{\dagger} \Big) \Big( \partial\_{\tau} b\_{i} \Big) - \mathop{\rm l\bar{\mathcal{K}}}\_{\stackrel{\scriptstyle{\bar{\mathcal{B}}}}} \sum\_{\stackrel{\scriptstyle{\bar{\mathcal{B}}}}} \Big( b\_{i\sigma}^{\dagger} b\_{\downarrow\sigma} + \mathop{\rm H.c.} \Big) + \mathop{\mathcal{A}} \sum\_{i} \Big( \Big[ b\_{i} \Big]^{2} - 1 \Big) + 2 \, \text{L}^{2} \text{zt} \, \mathcal{X}\_{\stackrel{\scriptstyle{\bar{\mathcal{B}}}}} \chi\_{\mathcal{A}} \right] \tag{4}$$

Here, the conventional saddle-point approximation has been performed for a spin-liquid-type Mott insulating phase. *χ<sup>f</sup>* describes band renormalization for electrons, and *χθ* approximately expresses the width of incoherent bands. *λ* is a Lagrange multiplier field to control the spinliquid to Fermi liquid phase transition and is regarded to be the chemical potential of bosons. This results from a nonlinear σ− model description, in which the rotor variable *e* −*iθ<sup>i</sup>* is replaced with *bi* and the unimodular constraint |*bi* | <sup>2</sup> =1 is considered. *z* is the nearest coordinate number of our lattice, and *L* <sup>2</sup> is the size of the system.

field theory. The other mechanism entails localized magnetic moments form singlets with themselves as an RKKY-type (Ruderman, Kittel, Kasuya, Yosida) interactions [97–99]. The resulting paramagnetic Mott insulating phase is called a spin-liquid state, in which chargeneutral spinons may emerge as low-energy elementary excitations and may interact among themselves through abundant singlet fluctuations, referred to as gauge fields [100]. Recalling the positive MR, we discuss the Mott–Anderson transition based on the spin-liquid physics.

**Figure 17.** A schematic phase diagram of Anderson–Hubbard model with one band and local density of states (LDOS)

of each phase. Adapted with permission from Byczuk [96]. © 2005 American Physical Society.

246 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**4.5. Generalization of Finkelstein's nonlinear sigma model approach near the Mott–**

To describe the Mott transition involved with the spin-liquid physics, we take the U(1) slave-

charge fluctuations (sound modes), and *f <sup>i</sup>σ* expresses a charge-neutral spinon field for collective dynamics of spin degrees of freedom. Then, an effective mean-field theory for such

in this model, the Hubbard Hamiltonian is decomposed into two sectors describing the dynamics of spinons and zero-sound modes, respectively, given by (Eq. (3) and Eq. (4)):

( ) ( ) † †

é ù = ¶- - ê ú +

> s

 .. *<sup>F</sup> <sup>i</sup> i f ij i ij S d f f t f f hc*

mc

 s

0 , ,

st

s

and *f <sup>i</sup>σ* can be easily formulated from the Hubbard model without randomness;

s s

ë û <sup>ò</sup> å å (3)

<sup>−</sup>*iθ<sup>i</sup> <sup>f</sup> <sup>i</sup>σ*, in which an electron field is expressed as a composite

accounts for the dynamics of collective

**Anderson transition**

variables of *θ<sup>i</sup>*

rotor representation [101] as *Ci<sup>σ</sup>* = *e*

operator of charge and spin degrees of freedom. *θ<sup>i</sup>*

b

t

**Figure 18.** Evolution of the electron spectral function from Landau's Fermi liquid state to a spin-liquid Mott insulating phase [102].

The Mott transition in this spin-liquid approach is realized by the condensation transition of *bi* , by which zero-sound modes are gapless in the Fermi liquid state, i.e., *bi* ≠0 but become gapped in the spin-liquid phase, i.e., *bi* =0. As a result, the height of the coherent peak in the electron spectral function, proportional to the condensation amplitude |*bi* | 2, decreases gradually to disappear toward the spin-liquid Mott insulating phase; during this process, the spectral weight is transferred to the incoherent background of upper and lower Hubbard bands (Figure 18). Our problem is to introduce a random potential into this spin-liquid Mott transi‐ tion. As long as the strength of the random potential remains smaller than the spinon band‐ width, we are allowed to deal with the role of the random potential perturbatively. This renormalization group analysis has been performed to reveal that the clean spin-liquid Mott critical point becomes unstable as soon as the random potential is turned on. As a result, a disorder critical point appears to be identified with a transition from diffusive spin-liquid glass insulator to diffusive Fermi liquid metal, in which the diffusive spin-liquid glass state consists of a diffusive spin-liquid phase of spinons and a charge glass phase of sound modes (Figure 19) [103].

**Figure 19.** A schematic phase diagram for the spin-liquid Mott–Anderson transition based on weak-coupling renorm‐ alization group analysis. FL: Fermi liquid metal, SLMI: spin liquid Mott insulator, SLBG: spin liquid Bose glass, AI: Anderson insulator. Bose glass means that charge dynamics described by sound modes displays glassy behaviors. *U*: strength of local interactions, *D*: disorder, and *t*: hopping integral of electrons. Adapted with permission from Kim [103]. © 2006 American Physical Society.

In this renormalization group analysis, the spinon conductivity has been used as an input parameter, rather than being self-consistently determined, and the diffusive dynamics of spinons has been assumed. As a result, the renormalization group study reached the conclu‐ sion that the weak-disorder quantum critical point depends on the residual spinon conduc‐ tivity, which means that the universality may not appear around this disorder quantum critical point. Suppose that SrIrO3 thin films on the substrate of GdScO3 are near this disordered MIT. Since SrIrO3 thin films on DyScO3, SrTiO3, and NdGaO3 substrates are also near this quantum critical point and their spinon conductivities differ from each other, we may observe contin‐ uous change of the temperature exponent ε of the electrical resistivity. Unfortunately, however, this previous study does not evaluate the spinon conductivity self-consistently, so the discus‐ sion cannot be beyond our speculation. Therefore, a theoretical framework must be developed to determine both transport coefficients of spinons and sound modes self-consistently.

One problem is that the previous renormalization group analysis does not consider effective interactions between diffusions and Cooperons, i.e., weak-localization corrections [34]. The replica nonlinear σ− model approach serves a natural theoretical framework to introduce such quantum corrections [104, 105]. By introducing the replica trick into the effective action for spinons and sound modes, and taking the diffusive spin-liquid fixed point as a saddle point for the spinon dynamics, one can derive an effective field theory as follows (Eq. (5)) [106]:

$$\begin{split} S\_{\textit{eff}} &= \bigwedge\_{0}^{\theta} \text{tr} \left[ d^{\tau} r \left( \frac{\pi N\_{\text{F}}^{f}}{4} \sqrt{-\hat{c}\_{r}^{2}} \text{tr} \left( \mathbf{K} \mathbf{Q} \right) + \frac{\pi N\_{\text{F}}^{f}}{4} D\_{\text{c}} \text{tr} \left( \nabla \mathbf{Q} - i \text{ar}^{\tau} \left[ \mathbf{Q}\_{r} \pi^{3} \right] \right) \right. \\ &+ \frac{1}{\iota \mathcal{U}} \left( \left( \mathcal{C}\_{r} + i \boldsymbol{\rho} \right) b^{\dagger} \right) \left\{ 1 - \mathcal{U} \frac{\sqrt{-\hat{\mathcal{C}}\_{r}^{2}}}{\sqrt{-\hat{\mathcal{C}}\_{r}^{2} + \mathcal{D}\_{\text{c}} \left( -\nabla^{2} \right)} \right) \left\{ \left( \mathcal{C}\_{r} - i \boldsymbol{\rho} \right) b \right\} + t \eta\_{\theta} \left| \left( \nabla - i \mathbf{a}^{\varsigma} \right) b \right|^{2} + \hat{\lambda}\_{\text{c}} b^{\dagger} b \right. \\ &+ \frac{1}{\beta} \sum\_{\nu} \sum\_{q} a\_{i}^{\tau} \left( q, i \nu \right) \left( \mathcal{V}\_{a} \left| \nu \right| + v\_{a}^{2} \left| q \right|^{2} \right) \mathcal{P}\_{\text{i}}^{T} a\_{i}^{\tau} \left( -q, -i \nu \right) \end{split} \tag{5}$$

critical point becomes unstable as soon as the random potential is turned on. As a result, a disorder critical point appears to be identified with a transition from diffusive spin-liquid glass insulator to diffusive Fermi liquid metal, in which the diffusive spin-liquid glass state consists of a diffusive spin-liquid phase of spinons and a charge glass phase of sound modes (Figure

248 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 19.** A schematic phase diagram for the spin-liquid Mott–Anderson transition based on weak-coupling renorm‐ alization group analysis. FL: Fermi liquid metal, SLMI: spin liquid Mott insulator, SLBG: spin liquid Bose glass, AI: Anderson insulator. Bose glass means that charge dynamics described by sound modes displays glassy behaviors. *U*: strength of local interactions, *D*: disorder, and *t*: hopping integral of electrons. Adapted with permission from Kim

In this renormalization group analysis, the spinon conductivity has been used as an input parameter, rather than being self-consistently determined, and the diffusive dynamics of spinons has been assumed. As a result, the renormalization group study reached the conclu‐ sion that the weak-disorder quantum critical point depends on the residual spinon conduc‐ tivity, which means that the universality may not appear around this disorder quantum critical point. Suppose that SrIrO3 thin films on the substrate of GdScO3 are near this disordered MIT. Since SrIrO3 thin films on DyScO3, SrTiO3, and NdGaO3 substrates are also near this quantum critical point and their spinon conductivities differ from each other, we may observe contin‐ uous change of the temperature exponent ε of the electrical resistivity. Unfortunately, however, this previous study does not evaluate the spinon conductivity self-consistently, so the discus‐ sion cannot be beyond our speculation. Therefore, a theoretical framework must be developed to determine both transport coefficients of spinons and sound modes self-consistently.

One problem is that the previous renormalization group analysis does not consider effective interactions between diffusions and Cooperons, i.e., weak-localization corrections [34]. The replica nonlinear σ− model approach serves a natural theoretical framework to introduce such quantum corrections [104, 105]. By introducing the replica trick into the effective action for spinons and sound modes, and taking the diffusive spin-liquid fixed point as a saddle point for the spinon dynamics, one can derive an effective field theory as follows (Eq. (5))

19) [103].

[106]:

[103]. © 2006 American Physical Society.

Dynamics of diffusions and Cooperons of the spinon sector are described by the replica nonlinear σ− model of the first line. *Q* is a 4*N* ×4*N* matrix field, located on the Grassmannian manifold of *Sp*(4*<sup>N</sup>* ) *Sp*(2*<sup>N</sup>* ) <sup>×</sup> *Sp*(2*<sup>N</sup>* ) , which spans retarded and advanced, Kramers doublet, and *N* replica spaces. *Κ* is a constant diagonal 4*N* ×4*N* matrix, given by *I*2*<sup>N</sup>* ×2*<sup>N</sup>* for the retarded part and − *I*2*<sup>N</sup>* ×2*<sup>N</sup>* for the advanced sector. *NF <sup>f</sup>* is the spinon density of states at the spinon Fermi-energy and *Dc* is the diffusion coefficient. *<sup>a</sup> <sup>c</sup>* a gauge field to describe singlet excitations, more precisely, spin-chirality fluctuations that are neglected in the previous mean-field analysis but intro‐ duced here beyond the saddle-point approximation [100], in which the dynamics of gauge fluctuations is described by the last term, which is given by the polarization function of the diffusive dynamics of spinons. When the effects of gauge fluctuations in the dynamics of diffusions and Cooperons can be neglected, the replica nonlinear *σ* − model reduces to that for the Anderson localization [104, 105]. The boson sector of sound modes is essentially the same as before, but gauge fluctuations are introduced beyond the mean-field analysis and the time sector is modified by the renormalization of the diffusive dynamics of spinons. Nonlocal derivative terms for both space and time should be defined in the momentum and frequency space, where the above expression is just for our formal writing. As discussed before, the boson sector describes the spin-liquid Mott transition, whereas the nonlinear *σ* − model part describes Anderson localization.

We claim that this replica nonlinear *σ* − model approach [107–110] generalizes the Finkelstein's nonlinear *σ* − model approach into the region of Mott transitions. Finkelstein's approach does not incorporate Mott physics associated with strong correlations of electrons. Instead, this approach considers the influences of effective interactions of both singlet and triplet channels in the dynamics of diffusions and Cooperons, considers instabilities of interacting diffusive Fermi liquids, and reveals the nature of the Anderson MIT. In contrast, the existing nonlinear *σ* − model approach considers instabilities of interacting diffusive spin-liquids and the nature of the Mott–Anderson MIT. In diffusive Fermi liquids, effective interactions of the triplet channel have an important influence on the Anderson MIT of two spatial dimensions [107– 110]. In our problem, we do not consider triplet-channel interactions because the positive magnetoelectrical resistivity suggests that the magnetic correlations may not be important. However, we speculate that gauge fluctuations, regarded to be singlet-channel interactions, affect the dynamics of Cooperons seriously, thereby suppressing weak-localization corrections and stabilizing the metallic state of spinons. This would be a novel mechanism to produce metallicity near the Mott–Anderson transition. Renormalization group analysis has not been performed yet for this effective field theory.

## **5. Summary**

Transition metal perovskites have been the platform for numerous emergent physics that originate from the coupling of the fundamental degrees of freedom such as spin, lattice, charge, and orbital as well as of disorder, which is unavoidable in solids. The physics of 5*d* perovskite iridates, in particular, has attracted considerable attention after the discovery of the novel *J*eff = 1/2 Mott insulating state and the evolution of dimensionality controlled MIT in the RP series Sr*n*+1Ir*n*O3*n*+1, originating due to strong SOC of 5*d* element which is comparable to its Coulomb correlation or bandwidth. SrIrO3, the end member of the RP series, shows many unusual phenomena, including different kinds of MITs. We fabricated epitaxial thin films of various thickness on various substrates to induce MIT in perovskite SrIrO3 by thickness reduction or imposed compressed strain. The MIT driven by thickness reduction occurs due to disorder, but the MIT driven by compressive strain in the films on different substrates is accompanied by peculiar non-Fermi liquid behaviors with an evolving temperature exponent in the electrical resistivity relationship. The latter MIT and associated non-Fermi liquid behaviors are probably due to the delicate interplay between correlation, SOC, and disorder, and thus pose a theoret‐ ical challenge to our understanding of non-Fermi liquid physics and MIT.

To reveal the nature of this non-Fermi liquid physics near the MIT of SrIrO3 thin films on various substrates, we first discussed the influence of emergent localized magnetic moments, referred to as UV physics. The observed positive MR in the whole temperature range led us to pursue another direction, referred to as IR physics, in which long-wave length and lowenergy fluctuations would be important. Because quantum criticality itself cannot explain the continuous evolution of the non-Fermi liquid physics, we speculate that the interplay between strong correlation of electrons and not-so-weak disorder may contribute to the continuously varying non-Fermi liquid physics. One possible scenario within IR physics is to consider quantum Griffiths effects with extreme inhomogeneity, in which local fluctuations between metallic and insulating phases, referred to as rare events, and may dominate the non-Fermi liquid physics. To realize this so-called Mott–Anderson–Griffiths scenario, we try to combine spin-liquid Mott physics with Anderson localization that describes a Mott transition from Landau's Fermi liquid state to a spin-liquid Mott insulating phase. Theoretical understanding of this special class of materials is in its early stages, and many new emergent phenomena in iridates are yet to be explored theoretically or experimentally.

## **Acknowledgements**

YHJ acknowledges the support by NRF (grant no. 2011-0009231). KSK was supported by NRF (grant no. 2012R1A1B3000550) and also by TJ Park Science Fellowship of the POSCO TJ Park Foundation. YHJ and KSK were jointly supported by CTM at POSTECH (grant nos. 2011-0030785 and 2011-0030786).

## **Author details**

metallicity near the Mott–Anderson transition. Renormalization group analysis has not been

Transition metal perovskites have been the platform for numerous emergent physics that originate from the coupling of the fundamental degrees of freedom such as spin, lattice, charge, and orbital as well as of disorder, which is unavoidable in solids. The physics of 5*d* perovskite iridates, in particular, has attracted considerable attention after the discovery of the novel *J*eff = 1/2 Mott insulating state and the evolution of dimensionality controlled MIT in the RP series Sr*n*+1Ir*n*O3*n*+1, originating due to strong SOC of 5*d* element which is comparable to its Coulomb correlation or bandwidth. SrIrO3, the end member of the RP series, shows many unusual phenomena, including different kinds of MITs. We fabricated epitaxial thin films of various thickness on various substrates to induce MIT in perovskite SrIrO3 by thickness reduction or imposed compressed strain. The MIT driven by thickness reduction occurs due to disorder, but the MIT driven by compressive strain in the films on different substrates is accompanied by peculiar non-Fermi liquid behaviors with an evolving temperature exponent in the electrical resistivity relationship. The latter MIT and associated non-Fermi liquid behaviors are probably due to the delicate interplay between correlation, SOC, and disorder, and thus pose a theoret‐

To reveal the nature of this non-Fermi liquid physics near the MIT of SrIrO3 thin films on various substrates, we first discussed the influence of emergent localized magnetic moments, referred to as UV physics. The observed positive MR in the whole temperature range led us to pursue another direction, referred to as IR physics, in which long-wave length and lowenergy fluctuations would be important. Because quantum criticality itself cannot explain the continuous evolution of the non-Fermi liquid physics, we speculate that the interplay between strong correlation of electrons and not-so-weak disorder may contribute to the continuously varying non-Fermi liquid physics. One possible scenario within IR physics is to consider quantum Griffiths effects with extreme inhomogeneity, in which local fluctuations between metallic and insulating phases, referred to as rare events, and may dominate the non-Fermi liquid physics. To realize this so-called Mott–Anderson–Griffiths scenario, we try to combine spin-liquid Mott physics with Anderson localization that describes a Mott transition from Landau's Fermi liquid state to a spin-liquid Mott insulating phase. Theoretical understanding of this special class of materials is in its early stages, and many new emergent phenomena in

YHJ acknowledges the support by NRF (grant no. 2011-0009231). KSK was supported by NRF (grant no. 2012R1A1B3000550) and also by TJ Park Science Fellowship of the POSCO TJ Park

ical challenge to our understanding of non-Fermi liquid physics and MIT.

iridates are yet to be explored theoretically or experimentally.

**Acknowledgements**

performed yet for this effective field theory.

250 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**5. Summary**

Abhijit Biswas1 , Ki-Seok Kim1,2 and Yoon Hee Jeong1\*

\*Address all correspondence to: yhj@postech.ac.kr

1 Department of Physics, POSTECH, Pohang, South Korea

2 Institute of Edge of Theoretical Science, POSTECH, Pohang, South Korea

## **References**


[26] Boer J H, Verwey E J W. Semi-conductors with partially and with completely filled 3*d*-lattice bands. Proc. Phys. Soc. 1937; 49: 59–71. (DOI:10.1088/0959-5309/49/4S/307)

[10] Granozio F M, Koster G, Rijnders G. Functional oxide interfaces. MRS Bull. 2013; 38:

[11] Ngai J H, Walker F J, Ahn C H. Correlated oxide physics and electronics. Annu. Rev.

[12] Sulpizio J A, Ilani S, Irvin P, Levy J. Nanoscale Phenomena in oxide Heterostructures. Annu. Rev. Mater. Res. 2014; 44: 117–149. (DOI: 10.1146/annurev-mats‐

[13] Chakhalian J, Freeland J W, Millis A J, Panagopoulos, Rondinelli J M. Colloquium: emergent properties in plane view: strong correlations at oxide interfaces. Rev. Mod. Phys. 2014; 86: 1189–1202. (DOI: http://dx.doi.org/10.1103/RevModPhys.86.1189) [14] Schlom D G, Chen L-Q, Fennie C J, Gopalan V, Muller D A, Pan X, Ramesh R, Uecker R. Elastic strain engineering of ferroic oxides. MRS Bull. 2014; 39: 118–130. (DOI:

[15] Tokura Y, Nagaosa N. Orbital physics in transition-metal oxides. Science. 2000; 288:

[16] Cheong S W. Transition metal oxides: the exciting world of orbitals. Nat. Mater. 2007;

[17] Cao G, De-Long L. Frontiers of 4*d*- and 5*d*-Transition Metal Oxides. 1st edition.

[18] Krempa W W, Chen G, Kim Y B, Balents L. Correlated quantum phenomena in the strong spin-orbit regime. Annu. Rev. Condens. Matter Phys. 2014; 5: 57–82. (DOI:

[19] Rau J G, Lee E K-H, Kee H-Y. Spin-Orbit Physics Giving Rise to Novel Phases in Cor‐

[20] Imada M, Fujimori A, Tokura Y. Metal–insulator transitions. Rev. Mod. Phys. 1998;

[22] Dobrosavljević V, Trivedi N, Valles J M Jr. Conductor insulator quantum phase tran‐

[23] Miranda E, Dobrosavljević V. Disorder-driven non-Fermi liquid behavior of correlat‐ ed electrons. Rep. Prog. Phys. 2005; 68: 2337–2408. (DOI:10.1088/0034-4885/68/10/R02)

[24] Edwards P P, Johnston R L, Rao C N R, Tunstall D P, Hensel. The metal–insulator transition: a perspective. Phil. Trans. R. Soc. Lond. A. 1998; 356: 5–22. (DOI: 10.1098/

[25] Drude P. Zur Elektronentheorie der metalle. Annalen der Physik. 1900; 306: 566–613.

related Systems: Iridates and Related Materials. arXiv:1507.06323

[21] Mott N F. Metal–Insulator Transitions. Taylor & Francis, London; 1990.

Mater. Res. 2014; 44: 1–17. (DOI: 10.1146/annurevmatsci 070813-113248)

1017–1023. (DOI: 10.1557/mrs.2013.282)

252 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

462–468. (DOI: 10.1126/science.288.5465.462)

10.1146/annurev-conmatphys-020911-125138)

70: 1039–1263. (DOI: 10.1103/RevModPhys.70.1039)

6: 927–928. (DOI:10.1038/nmat2069)

World Scientific; 2013. 328 pp.

sitions. Oxford; 2012.

rsta.1998.0146)

(DOI: 10.1002/andp.19003060312)

ci-070813-113437)

10.1557/mrs.2014.1)


[56] Nichols J, Terzic J, Bittle E G, Korneta O B, Long L E D, Brill J W, Cao G, Seo S S A. Tuning electronic structure via epitaxial strain in Sr2IrO4 thin films. Appl. Phys. Lett. 2013; 102: 141908-4. (DOI:http://dx.doi.org/10.1063/1.4801877)

[43] Ruddlesden S N, Popper P. New compounds of the K2NiF4 type. Acta Crystallogr.

[44] Ruddlesden S N, Popper P. The compound Sr3Ti2O7 and its structure. Acta Crystal‐

[46] Dion M, Ganne M, Tournoux M. Nouvelles familles de phases MIMII2Nb3O10 a feuil‐ lets "perovskites." Mater. Res. Bull. 1981; 16: 1429−1435. (DOI:

[47] Jacobson A J, Johnson J W, Lewandowski J T. Interlayer chemistry between thick transition-metal oxide layers: synthesis and intercalation reactions of K[Ca2Nan-3NbnO3n+1] (3 ≤ *n* ≤ 7). Inorg. Chem. 1985; 24: 3727−3729. (DOI: 10.1021/

[48] Goldschmidt V M. Die gesetze der krystallochemie. Naturwissenschaften, 1926; 14:

[50] Moon S J, Jin H, Kim K W, Choi W S, Lee Y S, Yu J, Cao G, Sumi A, Funajubo H, Bernhard C, Noh T W. Dimensionality-controlled insulator-metal transition and cor‐ related metallic state in 5d transition metal oxides Srn+1IrnO3n+1 (*n* = 1, 2, and ∞). Phys.

[51] Crawford M K, Subramanian M A, Harlow R L, Fernandez-Baca, Wang Z R, John‐ ston D C. Structural and magnetic studies of Sr2IrO4. Phys. Rev. B. 1994; 49: 9198–

[52] Cao G, Bolivar J, McCall S, Crow J E, Guertin R P. Weak ferromagnetism, metal-tononmetal transition, and negative differential resistivity in single-crystal Sr2IrO4. Phys. Rev. B (R). 1998; 57: 11039–11042. (DOI:http://dx.doi.org/10.1103/PhysRevB.

[53] Kim B J, Jin H, Moon S J, Kim J-Y, Park B-G, Leem C S, Yu J, Noh T W, Kim C, Oh S-J, Park J-H, Durairaj V, Rotenberg E. Novel *J*eff = 1/2 Mott state induced by relativistic spin-orbit coupling in Sr2IrO4. Phys. Rev. Lett. 2008; 101: 076402-4. (DOI:10.1103/

[54] Kim B J, Ohsumi H, Komesu T, Sakai S, Morita T, Takagi H, Arima T. Phase-sensitive observation of a spin-orbital mott state in Sr2IrO4. Science. 2009; 323: 1329–1332. (DOI:

[55] Wojek B M, Bernsten M H, Boseggia S, Boothroyd A T, Prabhakaran D, Mcmorrow D F, Ronnow H M, Chang J, Tjernberg O. The *J*eff = 1/2 insulator Sr3Ir2O7 studied by means of angle-resolved photoemission spectroscopy. J. Phys. Condens. Matter. 2012;

[49] Cullity B D. Introduction to Magnetic Materials. Addison-Wesley; 1972.

Rev. Lett. 2008; 101: 226402-4. (DOI:10.1103/PhysRevLett.101.226402)

1957; 10: 538−539. (DOI:10.1107/S0365110X57001929)

logr. 1958; 11: 54−55. (DOI:10.1107/S0365110X58000128)

[45] Aurivillius B. Ark. Kemi 1949; 1: 463−480, 499−512.

254 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

10.1016/0025-5408(81)90063-5)

477–485. (DOI: 10.1007/BF01507527)

9201. (DOI:10.1103/PhysRevB.49.9198)

ic00217a006)

57.R11039)

PhysRevLett.101.076402)

10.1126/science.1167106)

24: 415602-4. (DOI:10.1088/0953-8984/24/41/415602)


[79] Nie Y F, King P D C, Kim C H, Uchida M, Wei H I, Faeth B D, Ruf J P, Ruff J P C, Xie L, Pan X, Fennie C J, Schlom D G, Shen K M. Interplay of spin-orbit interactions, di‐ mensionality, and octahedral rotations in semimetallic SrIrO3. Phys. Rev. Lett. 2015; 114: 016401-6. (DOI:10.1103/PhysRevLett.114.016401)

[67] Subramanian M A, Crawford M K, Harlow R L. Single crystal structure determina‐ tion of double layered strontium iridium oxide [Sr3Ir2O7]. Mat. Res. Bull. 1994; 29:

[68] Li L, Kong P P, Qi T F, Jin C Q, Yuan S J, DeLong L E, Schlottmann P, Cao G. Tuning the Jeff=1/2 insulating state via electron doping and pressure in the double-layered iri‐ dates Sr3Ir2O7. Phys. Rev. B. 2013; 87: 235127-6. (DOI:10.1103/PhysRevB.87.235127) [69] Cao G, Xin Y, Alexander C S, Crow J E, Schlottmann P, Crawford M K, Hatlow R L, Marshall W. Anomalous magnetic and transport behavior in the magnetic insulator

Sr3Ir2O7. Phys. Rev. B. 2002; 66: 214412-7. (DOI:10.1103/PhysRevB.66.214412)

Rev. Lett. 2012; 109: 037204-5. (DOI:10.1103/PhysRevLett.109.037204)

[70] Kim J W, Choi Y, Kim J, Mitchell J F, Jackeli G, Daghofer M, van der Brink J, Khaliul‐ lin G, Kim B J. Dimensionality driven spin-flop transition in layered iridates. Phys.

[71] Li L, Kong P P, Qi T F, Jin C Q, Yuan S J, DeLong L E, Schlottmann P, Cao G. Tuning *J*eff=1/2 insulating state via electron doping and pressure in double-layered iridate

[72] Kim J, Said A H, Casa D, Upton M H, Gog T, Daghofer M, Jackeli G, van den Brink J, Khaliullin G, Kim B J. Large spin-wave energy gap in the bilayer iridate Sr3Ir2O7: evi‐ dence for the enhanced dipolar interactions near the Mott metal–insulator transi‐ tions. Phys. Rev. Lett. 2012; 109: 157402-5. (DOI:10.1103/PhysRevLett.109.157402) [73] Okada Y, Walkup D, Lin H, Dhital C, Chang T-R, Khadka S, Zhou W, Jeng H-T, Ban‐ sil A, Wang Z, Wilson S, Madhavan V. Imaging the evolution of metallic states in a

Sr3Ir2O7. Phys. Rev. B. 2013; 87: 235127-6. (DOI: 10.1103/PhysRevB.87.235127)

correlated iridates. Nat. Mater. 2013; 12: 707–713. (DOI:10.1038/nmat3653)

Phys. Rev. B. 2014; 89: 155115-6. (DOI:10.1103/PhysRevB.89.155115)

10.1016/0022-4596(71)90022-3)

10.1103/PhysRevB.89.214106)

76: 100402-4. (DOI: 10.1103/PhysRevB.76.100402)

Appl. Phys. 2008; 103: 103706-5. (DOI:10.1063/1.2908879)

[74] Park H J, Sohn C H, Jeong D W, Cao G, Kim K W, Moon S J, Jin H, Cho D-Y, Noh T W. Phonon-assisted optical excitation in the narrow bandgap Mott insulator Sr3Ir2O7.

[75] Longo J M, Kafalas J A, Arnott R J. Structure and properties of the high and low pres‐ sure forms of SrIrO3. J. Solid State Chem. 1971; 3: 174–179. (DOI:

[76] Cao G, Durairaj V, Chikara S, DeLong L E, Parkin S, Schlottmann P. Non-Fermi liq‐ uid behavior in nearly ferromagnetic SrIrO3 single crystals. Phys. Rev. B. (R), 2007;

[77] Zhao J G,Yang L X, Yu Y, Li F Y, Yu R C, Fang Z, Chen L C, Jin C Q. High-pressure synthesis of orthorhombic SrIrO3 perovskite and its positive magnetoresistance. J.

[78] Blanchard P E R, Reynolds E, Kennedy B J, Kimpton J A, Avdev M, Belik A A. Anomalous thermal expansion in orthorhombic peovskite SrIrO3: interplay between spin-orbit coupling and the crystal lattice. Phys. Rev. B. 2014; 89: 214106-8. (DOI:

645–650. (DOI: 10.1016/0025-5408(94)90120-1)

256 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications


[106] Kim K-M, Kim K-S. in preparation

[91] Löhneysen H, Rosch A, Vojta M, Wölfle P. Fermi-liquid instabilities at magnetic quantum phase transition. Rev. Mod. Phys. 2007; 79: 1015–1075. (DOI:10.1103/

[92] Harris A B. Nature of the "Griffiths" singularity in dilute magnets. Phys. Rev. B.

[93] Vojta T. Phase and phase transitions in disordered quantum systems. AIP Conf. Proc.

[94] Bhatt R N, Lee P A. Scaling studies of highly disordered spin-1/2 antiferromagnetic systems. Phys. Rev. Lett. 1982; 48: 344–347. (DOI:10.1103/PhysRevLett.48.344)

[95] Griffiths R B. Nonanalytic behavior above the critical point in a randomising ferro‐ magnet. Phys. Rev. Lett. 1969; 23: 17–19. (DOI:10.1103/PhysRevLett.23.17)

[96] Byczuk K. Metal–insulator transitions in the Falicov–Kimball model with disorder.

[97] Ruderman M A, Kittel C. Indirect exchange coupling of nuclear magnetic moments by conduction electrons. Phys. Rev. 1954; 96: 99–102. (DOI:http://dx.doi.org/10.1103/

[98] Kasuya T. A theory of metallic ferro-and antiferromagentism on Zener's model.

[99] Yosida K. Magnetic properties of Cu-Mn alloys. Phys. Rev.1957; 106: 893–898. (DOI:

[100] Lee P A, Nagaosa N, Wen X-G. Doping a Mott insulator: physics of high-temperature superconductivity. Rev. Mod. Phys. 2006; 7: 17–85. (DOI: http://dx.doi.org/10.1103/

[101] Florens S, Georges A. Slave-rotor mean-field theories of strongly correlated systems and the Mott transition in finite dimensions. Phys. Rev. B 2004; 70: 035114-15. (DOI:

[102] Cho D, Cheon S, Kim K-S, Lee S-H, Cho Y-H, Cheong S-W, Yeom H W. Nanoscale manipulation of the Mott insulating state coupled to charge order in 1T-TaS2. arXiv:

[103] Kim K-S. Role of disorder in the Mott–Hubbard transition. Phys. Rev. B. 2006; 73:

[104] Wegner F. The mobility edge problem: continuous symmetry and a conjecture. Z.

[105] Schäfer L, Wegner F. Disordered system with n orbitals per site: Lagrange formula‐ tion, hyperbolic symmetry, and goldstone modes. Z. Phys. B. 1980; 3: 113–126. (DOI:

Phys. Rev. B. 2005; 71: 205105-8. (DOI:10.1103/PhysRevB.71.205105)

Prog. Theor. Phys. 1956; 16: 45–57. (DOI: 10.1143/PTP.16.45)

RevModPhys.79.1015)

PhysRev.96.99)

10.1103/PhysRev.106.893)

10.1103/PhysRevB.70.035114)

235115-10. (DOI: 10.1103/PhysRevB.73.235115)

Phys. B. 1979; 35: 207–210. (DOI: 10.1007/BF01319839)

RevModPhys.78.17)

10.1007/BF01598751)

1505.00690

1975; 12: 203–207. (DOI: 10.1103/PhysRevB.12.203)

2013; 1550: 188–247. (DOI: 10.1063/1.4818403)

258 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications


## **Structural, Magnetic and Transport Properties of B-Site Substituted Perovskite La0.7Sr0.3MnO3**

J.B. Yang, M.S. Kim, T. F. Creel, H. Zhao, X.G. Chen, W.B. Yelon and W.J. James

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61770

## **Abstract**

In this chapter, in order to understand the structural related magnetic and transport properties of B site substituted perovskites La0.7Sr0.3MnO3 (LSMO), we have systematical‐ ly investigated the effects of replacing some of the Mn with nonmagnetic elements Ti, Zr, Cu, Al, Zn and magnetic elements Co, Ni, Cr, Fe. The structural, magnetic and electrical phase transitions and transport properties of these compounds were investigated by neu‐ tron diffraction, magnetization and electric resistivity measurements.

The abnormal behaviors relative to the parent manganite perovskite are explained by the competition between the double exchange and super exchange interactions, Mn-O bond distance, Mn-O-Mn bond angle, local Jahn-Teller distortion, the dilution of magnetiza‐ tion, the frustration of spins, and the change of valence states.

**Keywords:** Pervoskite, crystal structure, magnetic structure, neutron diffraction, magnet‐ ic properties

## **1. Introduction**

Perovskite oxides have been an interesting research area for scientists due to their promising physical properties including colossal magnetoresistance (CMR), superconductivity, multi‐ ferroelectricity, metal-insulator transition (MIT), charge/orbital ordering, etc. Among various perovskite oxides, the manganite is a representative one with fascinating physical properties [37]. The manganite materials such as La1-xSrxMnO3, Nd1-xSrxMnO3, and Pr1-xCaxMnO3 exhibit rich phase diagram involving spin-charge-orbital ordering, canted antiferromagnetic, antifer‐ romagnetic/ferromagnetic ordering, and electronic phase separation [14, 17, 27, 39]. In addition to the abundant magnetic behavior, the MIT often occurs coincidently with structural or

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

magnetic transition [17]. Pure LaMnO3 is an A-type antiferromagnetic insulator. La3+ can be partially substituted by a divalent cation such as Sr2+ or Ca2+, and La1-x(Sr,Ca)xMnO3 can become a metallic and ferromagnetic material. The Mn ions are in Mn3+ and Mn4+ states, which both have a local spin (S=3/2) from their *t*2*<sup>g</sup>* <sup>3</sup> orbitals, and Mn3+ ion provides an extra electron from the eg orbital responsible for conduction. The magnetic and electronic properties in these compounds can be explained using Zener's double exchange (DE) interaction [4, 11, 20, 40]. There is ferromagnetic interaction between the spin of the eg 1 electron in Mn3+ and the local spin of *t*2*<sup>g</sup>* <sup>3</sup> . By Sr or Ca doping, the holes were introduced into the eg band near the Fermi energy, which leads to mobile holes and conduction under an electric field. It was revealed that both DE interactions and the strong electron-phonon coupling should be considered to understand the Sr(Ca)-doped systems [10, 19, 30, 31, 35]. Polaron hopping was also proposed as the dominant conduction mechanism below TC [3, 12, 21, 32, 41].

In the past decades, the A-site doped manganites A1-xDxMnO3 have been extensively studied with various attractive properties [14, 17, 27, 29, 33, 39]. In contrast, the B-site doped manganites have not been well studied. The substitution for the Mn (B site) has shown dramatic effect on the magnetic and transport properties of the perovskites [1, 2, 5, 6, 36, 38]. Generally, the B-site doping with 3d ions would destroy the ferromagnetic ordering of the Mn network, leading to the changes in the magnetic and electrical properties of manganites. The reentrant spin glass behavior has been observed in the Cr-doped A-type antiferromagnetic La0.46Sr0.54Mn1-xCrxO3 due to the competing interaction between the FM and the A-type AFM coupling. The charge-orbital ordered Nd0.5Ca0.5Mn1-xCrxO3 is a relaxor ferromagnet [26]. The Fe-doped La1-xCaxMnO3 has gone through the localization-delocaliza‐ tion transition as the increase of the dopant concentration [1, 34]. Two ferromagnetic phases appeared in the LaMn0.5Ni0.5O3 sample, which is critically related to the preparation process [18].

Therefore, in this chapter, the samples of La0.7Sr0.3Mn1-xTxO3 (T= Ti, Zr, Cu, Co and Cr) were prepared, and the effects of substitution on Mn were studied using neutron diffraction(ND), X-ray photoelectron spectra (XPS), magnetic and electric resistivity measurements. The relationship between structure and physical properties are explained by the competition between the DE and super exchange interactions, bandwidth W, bond angle, bond length and the frustration of spins and the change of valence states.

## **2. Experiments**

Samples of La0.7Sr0.3Mn1-xTxO3 (T=Ti, Zr, Cu, Co, Cr) were synthesized using the standard solidstate reaction method, starting with the high purity La2O3, MnO2, TiO2, CuO, Cr2O3, Co3O4 and SrCO3 powders. Appropriate amounts of these powders were weighed and mixed according to the desired stoichiometry for each sample, then sintered in air for one day at 800℃, and cooled naturally to room temperature as the raw material. The raw materials were ground and sintered again in air for one day at 1350℃ with a room-air quench. The reacted powders were ground and cold pressed into disks with the thickness of ~2 mm under a pressure of ~10 MPa. These disks were sintered in air for one more day at 1350℃ and cooled naturally to room temperature. X-ray diffraction of the powders was performed at room temperature using with Cu-Kα radiation. Powder neutron diffraction experiments were performed at the University of Missouri-Columbia Research Reactor (MURR, λ = 1.4875Å) and high resolution powder diffractometer at HZB Germany using neutrons of wavelength (λ = 1.79821Å). The patterns were collected at the temperature range from 5K to 300K. Refinement of the XRD and ND data were carried out using the FULLPROF program. Magnetic measurements were conducted with a SQUID magnetometer (MPMS, Quantum design). The zero-field cooling (ZFC) and field cooling (FC) magnetization curves were measured under applied magnetic field of 50Oe. Magnetoresistance data were collected using a physical properties measurement system (PPMS, Quantum design) with a standard four-point probe method.

## **3. Results and discussion**

magnetic transition [17]. Pure LaMnO3 is an A-type antiferromagnetic insulator. La3+ can be partially substituted by a divalent cation such as Sr2+ or Ca2+, and La1-x(Sr,Ca)xMnO3 can become a metallic and ferromagnetic material. The Mn ions are in Mn3+ and Mn4+ states, which both

the eg orbital responsible for conduction. The magnetic and electronic properties in these compounds can be explained using Zener's double exchange (DE) interaction [4, 11, 20, 40].

energy, which leads to mobile holes and conduction under an electric field. It was revealed that both DE interactions and the strong electron-phonon coupling should be considered to understand the Sr(Ca)-doped systems [10, 19, 30, 31, 35]. Polaron hopping was also proposed

In the past decades, the A-site doped manganites A1-xDxMnO3 have been extensively studied with various attractive properties [14, 17, 27, 29, 33, 39]. In contrast, the B-site doped manganites have not been well studied. The substitution for the Mn (B site) has shown dramatic effect on the magnetic and transport properties of the perovskites [1, 2, 5, 6, 36, 38]. Generally, the B-site doping with 3d ions would destroy the ferromagnetic ordering of the Mn network, leading to the changes in the magnetic and electrical properties of manganites. The reentrant spin glass behavior has been observed in the Cr-doped A-type antiferromagnetic La0.46Sr0.54Mn1-xCrxO3 due to the competing interaction between the FM and the A-type AFM coupling. The charge-orbital ordered Nd0.5Ca0.5Mn1-xCrxO3 is a relaxor ferromagnet [26]. The Fe-doped La1-xCaxMnO3 has gone through the localization-delocaliza‐ tion transition as the increase of the dopant concentration [1, 34]. Two ferromagnetic phases appeared in the LaMn0.5Ni0.5O3 sample, which is critically related to the preparation process

Therefore, in this chapter, the samples of La0.7Sr0.3Mn1-xTxO3 (T= Ti, Zr, Cu, Co and Cr) were prepared, and the effects of substitution on Mn were studied using neutron diffraction(ND), X-ray photoelectron spectra (XPS), magnetic and electric resistivity measurements. The relationship between structure and physical properties are explained by the competition between the DE and super exchange interactions, bandwidth W, bond angle, bond length and

Samples of La0.7Sr0.3Mn1-xTxO3 (T=Ti, Zr, Cu, Co, Cr) were synthesized using the standard solidstate reaction method, starting with the high purity La2O3, MnO2, TiO2, CuO, Cr2O3, Co3O4 and SrCO3 powders. Appropriate amounts of these powders were weighed and mixed according to the desired stoichiometry for each sample, then sintered in air for one day at 800℃, and cooled naturally to room temperature as the raw material. The raw materials were ground and sintered again in air for one day at 1350℃ with a room-air quench. The reacted powders were ground and cold pressed into disks with the thickness of ~2 mm under a pressure of ~10 MPa. These disks were sintered in air for one more day at 1350℃ and cooled naturally to room

<sup>3</sup> . By Sr or Ca doping, the holes were introduced into the eg band near the Fermi

<sup>3</sup> orbitals, and Mn3+ ion provides an extra electron from

electron in Mn3+ and the local

1

have a local spin (S=3/2) from their *t*2*<sup>g</sup>*

spin of *t*2*<sup>g</sup>*

[18].

**2. Experiments**

There is ferromagnetic interaction between the spin of the eg

262 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

as the dominant conduction mechanism below TC [3, 12, 21, 32, 41].

the frustration of spins and the change of valence states.

## **3.1. Ti-substituted perovskites, La0.7Sr0.3Mn1-xTixO3 [22]**

Ti-substituted perovskites La0.7Sr0.3Mn1-xTixO3, with 0 ≤ x ≤ 0.20 were studied using XRD, ND, magnetizatic and magnetoresistance (MR) measurements [22]. Typical ND patterns of the La0.7Sr0.3Mn1-xTixO3 (x=0.05, 0.1 and 0.2) were shown in Fig. 1. All samples show a rhombohedral structure (space group *R*3 ¯ *c*) from RT to 10 K. The crystal structure of the samples can be well fitted with the rhombohedral space group (No. 167), and the atomic positions of La(Sr): 6a (0,0,1/4), Mn(Ti): 6b (0,0,0); O18e (x,0,1/4). The magnetic structure was refined with P1 space group. It was confirmed that the Ti ions replaced the Mn ions at B sites, since the ionic radius of Ti4+ (0.605Å) lies between the ionic radius of Mn4+ (0.530 Å) and Mn3+ (0.645 Å)[38]. Fig. 2 plots the lattice parameter of the samples at different temperatures. It can be seen that the different behaviours are observed at RT and 10K. The lattice parameter *a* and unit cell volume reach a maximum value for x = 0.10, and then decrease for x > 0.10, while lattice parameter *c* increases with increasing x at 10 K. However, at RT, the lattice parameters *a*, *c* and the unit cell volume *V* all show similar increasing trend with increasing Ti content x. This difference should be related to the different magnetic ordering temperatures due to Ti substitution. In this system, as Mn ions are substituted by Ti ions, a lattice distortion may occur. We have calculated the so-called tolerance factor *t* of the La0.7Sr0.3Mn1-xTixO3, which can be used to indicate the geometric measure of size mismatch of perovskites. The calculated values of *t* are about 0.928 and 0.921 for La0.7Sr0.3MnO3 and La0.7Sr0.3Mn0.8Ti0.2O3, respectively. These values are in the stable range of the 0.89 < t < 1.02 for the rhombohedral structure. The decrease of the *t* will lead to the decrease of Mn-O-Mn bond angle from 166.5 º to 165 º for x=0 and x=0.2 without a structure change.

Fig. 3 plots the average Mn-O bond length and Mn-O-Mn bond angle of La0.7Sr0.3Mn1-xTixO<sup>3</sup> obtained from the ND refinement at 10 K and RT. At 10 K, the Mn-O bond length increases up to *x = 0.10* and keeps constant for *x>0.10*, while the Mn-O-Mn bond angle decreases with increasing x. At RT, the bond length of Mn-O increases up to x = 0.15 and keeps constant for *x≥0.15*, while the Mn-O-Mn bond angle decreases and reaches a minimum value for x = 0.15. The changes in the bond length and bond angle of MnO6 octahedral may help to diminish the internal strain induced by Ti substitution. The Ti substitution will change the oxygen positions 0

<sup>10</sup> x=0.05 T=RT

)

5

10

15

x=0.05 T=10 K

positions). Arrows shows some of the major magnetic diffraction peaks.

345.0 347.5 350.0 352.5

5.48 5.50 5.52

Volume (A3

)

a (A)

La0.7Sr0.3Mn1-xTi<sup>x</sup>

O3

3.1. Ti-substituted perovskites, La0.7Sr0.3Mn1-xTixO3 (Kim et al., 2005)

Ti-substituted perovskites La0.7Sr0.3Mn1-xTixO3, with 0 ≤ x ≤ 0.20 were studied using XRD, ND, magnetizatic and

magnetoresistance (MR) measurements (Kim et al., 2005a). Typical ND patterns of the La0.7Sr0.3Mn1-xTixO3 (x=0.05, 0.1 and 0.2) were shown in Fig. 1. All samples show a rhombohedral structure (space group R c3 ) from RT to 10 K. The crystal structure of the samples can be well fitted with the rhombohedral space group (No. 167), and the atomic positions of La(Sr): 6a (0,0,1/4), Mn(Ti): 6b (0,0,0); O18e (x,0,1/4). The magnetic structure was refined with P1 space group. It was confirmed that the Ti ions replaced the Mn ions at B sites, since the ionic radius of Ti4+ (0.605Å) lies between the ionic radius of Mn4+ (0.530 Å) and Mn3+ (0.645 Å)(Shannon 1976). Fig. 2 plots the lattice parameter of the samples at different temperatures. It can be seen that the different behaviours are observed at RT and 10K. The lattice parameter a and unit cell volume reach a maximum value for x = 0.10, and then decrease for x > 0.10, while lattice parameter c increases with increasing x at 10 K. However, at RT, the lattice parameters a, c and the unit cell volume V all show similar increasing trend with increasing Ti content x. This difference should be related to the different magnetic ordering temperatures due to Ti substitution. In this system, as Mn ions are substituted by Ti ions, a lattice distortion may occur. We have calculated

3.1. Ti-substituted perovskites, La0.7Sr0.3Mn1-xTixO3 (Kim et al., 2005)

Ti-substituted perovskites La0.7Sr0.3Mn1-xTixO3, with 0 ≤ x ≤ 0.20 were studied using XRD, ND, magnetizatic and

magnetoresistance (MR) measurements (Kim et al., 2005a). Typical ND patterns of the La0.7Sr0.3Mn1-xTixO3 (x=0.05, 0.1 and

0.2) were shown in Fig. 1. All samples show a rhombohedral structure (space group R c3 ) from RT to 10 K. The crystal

structure of the samples can be well fitted with the rhombohedral space group (No. 167), and the atomic positions of

La(Sr): 6a (0,0,1/4), Mn(Ti): 6b (0,0,0); O18e (x,0,1/4). The magnetic structure was refined with P1 space group. It was

confirmed that the Ti ions replaced the Mn ions at B sites, since the ionic radius of Ti4+ (0.605Å) lies between the ionic

radius of Mn4+ (0.530 Å) and Mn3+ (0.645 Å)(Shannon 1976). Fig. 2 plots the lattice parameter of the samples at different

temperatures. It can be seen that the different behaviours are observed at RT and 10K. The lattice parameter a and unit

cell volume reach a maximum value for x = 0.10, and then decrease for x > 0.10, while lattice parameter c increases with

trend with increasing Ti content x. This difference should be related to the different magnetic ordering temperatures due

to Ti substitution. In this system, as Mn ions are substituted by Ti ions, a lattice distortion may occur. We have calculated

mismatch of perovskites. The calculated values of t are about 0.928 and 0.921 for La0.7Sr0.3MnO3 and La0.7Sr0.3Mn0.8Ti0.2O3,

respectively. These values are in the stable range of the 0.89 < t < 1.02 for the rhombohedral structure. The decrease of the

x=0.20 T=RT

x=0.20 T=10 K

20 30 40 50 60

2θ (degree)

increasing x at 10 K. However, at RT, the lattice parameters a, c and the unit cell volume V all show similar increasing

the so-called tolerance factor t of the La0.7Sr0.3Mn1-xTixO3, which can be used to indicate the geometric measure of size

t will lead to the decrease of Mn-O-Mn bond angle from 166.5 º to 165 º for x=0 and x=0.2 without a structure change.

the so-called tolerance factor t of the La0.7Sr0.3Mn1-xTixO3, which can be used to indicate the geometric measure of size mismatch of perovskites. The calculated values of t are about 0.928 and 0.921 for La0.7Sr0.3MnO3 and La0.7Sr0.3Mn0.8Ti0.2O3, respectively. These values are in the stable range of the 0.89 < t < 1.02 for the rhombohedral structure. The decrease of the

> x=0.10 T=RT

x=0.10 T=10 K

Figure 1. ND patterns of La0.7Sr0.3Mn1-xTixO3 (x = 0.05, 0.10, 0.20) at RT and 10 K. (The bottom curves (red line) are the difference between experimental data and the refinement data. The vertical bars (blue line) represent the magnetic (bottom) and Bragg (top) peak **Figure 1.** ND patterns of La0.7Sr0.3Mn1-xTixO3 (x = 0.05, 0.10, 0.20) at RT and 10 K. (The bottom curves (red line) are the difference between experimental data and the refinement data. The vertical bars (blue line) represent the magnetic (bottom) and Bragg (top) peak positions). Arrows shows some of the major magnetic diffraction peaks. experimental data and the refinement data. The vertical bars (blue line) represent the magnetic (bottom) and Bragg (top) peak positions). Arrows shows some of the major magnetic diffraction peaks.

refinement at 10 K and RT. At 10 K, the Mn-O bond length increases up to x = 0.10 and keeps constant for x>0.10, while **Figure 2.** Lattice parameters a, c, and unit cell volumes of La0.7Sr0.3Mn1-xTixO3 with different Ti content x at 10 K and RT.

Fig. 3 plots the average Mn-O bond length and Mn-O-Mn bond angle of La0.7Sr0.3Mn1-xTixO3 obtained from the ND

and affect the Mn-O bond length and the Mn-O–Mn bond angle. This is consistent with the change of the tolerance factor, indicating the increase of Mn-O bond length and the decrease of Mn-O-Mn bond angle are related. Figure 2. Lattice parameters a, c, and unit cell volumes of La0.7Sr0.3Mn1-xTixO3 with different Ti content x at 10 K and RT. Fig. 3 plots the average Mn-O bond length and Mn-O-Mn bond angle of La0.7Sr0.3Mn1-xTixO3 obtained from the ND

The temperature dependent resistivities for La0.7Sr0.3Mn1-xTixO3 compounds (x = 0.0(a), 0.05(b), 0.10(c), and 0.15(d)) under magnetic fields H = 0, 1, 3, and 5 T were plotted in Fig. 4. refinement at 10 K and RT. At 10 K, the Mn-O bond length increases up to x = 0.10 and keeps constant for x>0.10, while

As a comparison, the Curie temperatures *Tc* of the samples are also shown in the figure 4. It can be seen that the resistivity for the x ≤ 0.05 sample shows a metallic-like behavior below the *TC*. A MIT is observed for all the x ≥ 0.10 samples at low temperature. A maximum peak in the resistivity is observed below TC for all samples, and shifts to a lower temperature when x increases. The one-electron bandwidth *W* is one of the fundamental parameters for controlling the Mn-O-Mn bond angle decreases with increasing x. At RT, the bond length of Mn-O increases up to x = 0.15 and keeps

constant for x≥0.15, while the Mn-O-Mn bond angle decreases and reaches a minimum value for x = 0.15. The changes in

the bond length and bond angle of MnO6 octahedral may help to diminish the internal strain induced by Ti substitution.

The Ti substitution will change the oxygen positions and affect the Mn-O bond length and the Mn-O–Mn bond angle.

Figure 3. Average Mn-O bond lengths (a), Mn-O-Mn bond angles (b), and electronic bandwidth parameter W (c), of La0.7Sr0.3Mn1-xTixO<sup>3</sup> **Figure 3.** Average Mn-O bond lengths (a), Mn-O-Mn bond angles (b), and electronic bandwidth parameter *W* (c), of La0.7Sr0.3Mn1-xTixO3 at room temperature and at 10K.

at room temperature and at 10K.

interaction between Mn-Mn ions.

the magnetic and electric behavior of correlated electrons system such as perovskite [15, 35]. Using the tight binding approximation, the empirical formula of the *W* for ABO3-type perov‐ skites is [29]. The temperature dependent resistivities for La0.7Sr0.3Mn1-xTixO3 compounds (x = 0.0(a), 0.05(b), 0.10(c), and 0.15(d)) under magnetic fields H = 0, 1, 3, and 5 T were plotted in Fig. 4.

$$\mathcal{W} = \frac{\cos \beta}{\left(d\_{\text{Min} \to 0}\right)^{3.5}} \tag{1}$$

resistivity for the x ≤ 0.05 sample shows a metallic-like behavior below the TC. A MIT is observed for all the x ≥ 0.10

(c). It is obvious that W decreases with the increasing Ti content x, which will reduce the 2p-3d hybridization between O

and Mn ions and increases the electron-phonon coupling. Therefore, it leads to a lower magnetic ordering temperature

It is obvious that a field-induced shift of the resistivity maximum occurs for x> 0.05 samples. The MR ratio increases with

the Ti content x, and reaches to about 70% for La0.7Sr0.3Mn0.8Ti0.2O3, which is related to the weaker magnetic interaction

TC and higher resistivity with increasing Ti content. Since the double exchange interaction between Mn-Mn ions is

strongly dependent on both bond angle and bond distance, the substitution of Mn by Ti will decrease the exchange

Figure 2. Lattice parameters a, c, and unit cell volumes of La0.7Sr0.3Mn1-xTixO3 with different Ti content x at 10 K and RT. Fig. 3 plots the average Mn-O bond length and Mn-O-Mn bond angle of La0.7Sr0.3Mn1-xTixO3 obtained from the ND where *β* =(*π* −*θ*<*Mn*−*O*−*Mn*>) / 2, *dMn-O* is the average Mn-O bond length, and *θ*<*Mn*−*O*−*Mn*> is the average Mn-O-Mn bond angle. By using the data obtained from ND refinements, the calculated values of the *W* are plot in Fig. 3 (c). It is obvious that *W* decreases with the increasing Ti content x, which will reduce the 2p-3d hybridization between O and Mn ions and increases the electron-phonon coupling. Therefore, it leads to a lower magnetic ordering temperature TC and higher resistivity with increasing Ti content. Since the double exchange interaction between Mn-Mn ions is strongly dependent on both bond angle and bond distance, the substitution of Mn by Ti will decrease the exchange interaction between Mn-Mn ions. samples at low temperature. A maximum peak in the resistivity is observed below TC for all samples, and shifts to a lower temperature when x increases. The one-electron bandwidth W is one of the fundamental parameters for controlling the magnetic and electric behavior of correlated electrons system such as perovskite (Hwang et al., 1995 ; Radaelli et al., 1997). Using the tight binding approximation, the empirical formula of the W for ABO3-type perovskites is (Medarde et al., 1995).

and affect the Mn-O bond length and the Mn-O–Mn bond angle. This is consistent with the change of the tolerance factor, indicating the increase of Mn-O bond length and the decrease

**Figure 2.** Lattice parameters a, c, and unit cell volumes of La0.7Sr0.3Mn1-xTixO3 with different Ti content x at 10 K and RT.

The temperature dependent resistivities for La0.7Sr0.3Mn1-xTixO3 compounds (x = 0.0(a), 0.05(b),

As a comparison, the Curie temperatures *Tc* of the samples are also shown in the figure 4. It can be seen that the resistivity for the x ≤ 0.05 sample shows a metallic-like behavior below the *TC*. A MIT is observed for all the x ≥ 0.10 samples at low temperature. A maximum peak in the resistivity is observed below TC for all samples, and shifts to a lower temperature when x increases. The one-electron bandwidth *W* is one of the fundamental parameters for controlling

0.10(c), and 0.15(d)) under magnetic fields H = 0, 1, 3, and 5 T were plotted in Fig. 4.

2θ (degree) 2θ (degree)

positions). Arrows shows some of the major magnetic diffraction peaks.

 RT 10K

(b)

0.00 0.05 0.10 0.15 0.20

Ti content X

0.00 0.05 0.10 0.15 0.20

(bottom) and Bragg (top) peak positions). Arrows shows some of the major magnetic diffraction peaks.

Ti content X

x=0.10 T=RT

20 30 40 50 60

x=0.10 T=10 K

La0.7Sr0.3Mn1-xTi<sup>x</sup> O3

264 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

20 30 40 50 60

3.1. Ti-substituted perovskites, La0.7Sr0.3Mn1-xTixO3 (Kim et al., 2005)

Ti-substituted perovskites La0.7Sr0.3Mn1-xTixO3, with 0 ≤ x ≤ 0.20 were studied using XRD, ND, magnetizatic and

magnetoresistance (MR) measurements (Kim et al., 2005a). Typical ND patterns of the La0.7Sr0.3Mn1-xTixO3 (x=0.05, 0.1 and 0.2) were shown in Fig. 1. All samples show a rhombohedral structure (space group R c3 ) from RT to 10 K. The crystal structure of the samples can be well fitted with the rhombohedral space group (No. 167), and the atomic positions of La(Sr): 6a (0,0,1/4), Mn(Ti): 6b (0,0,0); O18e (x,0,1/4). The magnetic structure was refined with P1 space group. It was confirmed that the Ti ions replaced the Mn ions at B sites, since the ionic radius of Ti4+ (0.605Å) lies between the ionic radius of Mn4+ (0.530 Å) and Mn3+ (0.645 Å)(Shannon 1976). Fig. 2 plots the lattice parameter of the samples at different temperatures. It can be seen that the different behaviours are observed at RT and 10K. The lattice parameter a and unit cell volume reach a maximum value for x = 0.10, and then decrease for x > 0.10, while lattice parameter c increases with increasing x at 10 K. However, at RT, the lattice parameters a, c and the unit cell volume V all show similar increasing trend with increasing Ti content x. This difference should be related to the different magnetic ordering temperatures due to Ti substitution. In this system, as Mn ions are substituted by Ti ions, a lattice distortion may occur. We have calculated

3.1. Ti-substituted perovskites, La0.7Sr0.3Mn1-xTixO3 (Kim et al., 2005)

Ti-substituted perovskites La0.7Sr0.3Mn1-xTixO3, with 0 ≤ x ≤ 0.20 were studied using XRD, ND, magnetizatic and

magnetoresistance (MR) measurements (Kim et al., 2005a). Typical ND patterns of the La0.7Sr0.3Mn1-xTixO3 (x=0.05, 0.1 and

0.2) were shown in Fig. 1. All samples show a rhombohedral structure (space group R c3 ) from RT to 10 K. The crystal

structure of the samples can be well fitted with the rhombohedral space group (No. 167), and the atomic positions of

La(Sr): 6a (0,0,1/4), Mn(Ti): 6b (0,0,0); O18e (x,0,1/4). The magnetic structure was refined with P1 space group. It was

confirmed that the Ti ions replaced the Mn ions at B sites, since the ionic radius of Ti4+ (0.605Å) lies between the ionic

radius of Mn4+ (0.530 Å) and Mn3+ (0.645 Å)(Shannon 1976). Fig. 2 plots the lattice parameter of the samples at different

temperatures. It can be seen that the different behaviours are observed at RT and 10K. The lattice parameter a and unit

cell volume reach a maximum value for x = 0.10, and then decrease for x > 0.10, while lattice parameter c increases with

trend with increasing Ti content x. This difference should be related to the different magnetic ordering temperatures due

to Ti substitution. In this system, as Mn ions are substituted by Ti ions, a lattice distortion may occur. We have calculated

mismatch of perovskites. The calculated values of t are about 0.928 and 0.921 for La0.7Sr0.3MnO3 and La0.7Sr0.3Mn0.8Ti0.2O3,

respectively. These values are in the stable range of the 0.89 < t < 1.02 for the rhombohedral structure. The decrease of the

x=0.20 T=RT

x=0.20 T=10 K

increasing x at 10 K. However, at RT, the lattice parameters a, c and the unit cell volume V all show similar increasing

the so-called tolerance factor t of the La0.7Sr0.3Mn1-xTixO3, which can be used to indicate the geometric measure of size

t will lead to the decrease of Mn-O-Mn bond angle from 166.5 º to 165 º for x=0 and x=0.2 without a structure change.

the so-called tolerance factor t of the La0.7Sr0.3Mn1-xTixO3, which can be used to indicate the geometric measure of size mismatch of perovskites. The calculated values of t are about 0.928 and 0.921 for La0.7Sr0.3MnO3 and La0.7Sr0.3Mn0.8Ti0.2O3, respectively. These values are in the stable range of the 0.89 < t < 1.02 for the rhombohedral structure. The decrease of the

t will lead to the decrease of Mn-O-Mn bond angle from 166.5 º to 165 º for x=0 and x=0.2 without a structure change.

x=0.10 T=RT

x=0.10 T=10 K

2θ (degree) 2θ (degree)

20 30 40 50 60

**Figure 1.** ND patterns of La0.7Sr0.3Mn1-xTixO3 (x = 0.05, 0.10, 0.20) at RT and 10 K. (The bottom curves (red line) are the difference between experimental data and the refinement data. The vertical bars (blue line) represent the magnetic

La0.7Sr0.3Mn1-xTi<sup>x</sup>

O3

13.30 13.35 13.40

(b)

c (Α)

Figure 2. Lattice parameters a, c, and unit cell volumes of La0.7Sr0.3Mn1-xTixO3 with different Ti content x at 10 K and RT.

Fig. 3 plots the average Mn-O bond length and Mn-O-Mn bond angle of La0.7Sr0.3Mn1-xTixO3 obtained from the ND

refinement at 10 K and RT. At 10 K, the Mn-O bond length increases up to x = 0.10 and keeps constant for x>0.10, while

x=0.20 T=RT

Figure 1. ND patterns of La0.7Sr0.3Mn1-xTixO3 (x = 0.05, 0.10, 0.20) at RT and 10 K. (The bottom curves (red line) are the difference between

0.00 0.05 0.10 0.15 0.20

Ti content X

experimental data and the refinement data. The vertical bars (blue line) represent the magnetic (bottom) and Bragg (top) peak

positions). Arrows shows some of the major magnetic diffraction peaks.

x=0.20 T=10 K

20 30 40 50 60

20 30 40 50 60

13.30

13.35

c (Α)

13.40

2θ (degree)

of Mn-O-Mn bond angle are related.

0.00 0.05 0.10 0.15 0.20

 RT 10K

Ti content X

0 5 <sup>10</sup> x=0.05 T=RT

Intensity (103

345.0 347.5 350.0 352.5 355.0 (c)

Volume (A3

)

345.0 347.5 350.0 352.5 355.0 (c)

5.48

5.50

a (A)

5.52

(a)

5.48 5.50 5.52 (a)

Volume (A3

)

a (A)

)

x=0.05 T=10 K

Intensity (103

)


<sup>10</sup> x=0.05 T=RT

0

5

0

5

10

15

x=0.05 T=10 K

refinement at 10 K and RT. At 10 K, the Mn-O bond length increases up to x = 0.10 and keeps constant for x>0.10, while It is obvious that a field-induced shift of the resistivity maximum occurs for x> 0.05 samples. The MR ratio increases with the Ti content x, and reaches to about 70% for La0.7Sr0.3Mn0.8Ti0.2O3, which is related to the weaker magnetic interaction between Mn-Mn ions. The separation of TC and the resistivity maximum temperature Tρ,max becomes wider as Ti content increases due to the weak coupling between the magnetic ordering and the resistivity as compared with La0.7Sr0.3MnO3. 3.5 cos <sup>W</sup> ( ) Mn O d β − = (1) Dummy Text where ( )/2 Mn O Mn β πθ< −− > = − , dMn-O is the average Mn-O bond length, and Mn O Mn θ< −− > is the average Mn-O-Mn bond angle. By using the data obtained from ND refinements, the calculated values of the W are plot in Fig. 3

between Mn-Mn ions. The separation of TC and the resistivity maximum temperature Tρ,max becomes wider as Ti

20 40 60 80 100

2θ (degree)

Fig. 6 plots the average Mn-O bond length,Mn-O-Mn bond angle and bandwidth W obtained from the refined ND

patterns at RT and 10K. Similar to those of the Ti substituted samples, the Mn-O bond length increases, while Mn-O-Mn

bond angle and band width W decrease with the increase of Zr content. The decrease of the bandwidth W will reduce

the overlap between the O-2p and the Mn-3d orbitals, which will reduce the exchange interactions between Mn-Mn in

this system. This is confirmed by the results that the reduction in magnetic moments and the Curie temperature with

increased Zr content. A metallic-like behavior was observed for the La0.7Sr0.3Mn1-xZrxO3 at low temperature. The

contribution from the two-magnon scattering in resistivity becomes larger with increasing Zr content.

Figure 5. Neutron diffraction patterns of La0.7Sr0.3Mn0.9Zr0.1O3 and La0.7Sr0.3Mn0.85Zr0.15O3 at room temperature.

La0.7Sr0.3Mn0.85Zr0.15O<sup>3</sup>

O7

content increases due to the weak coupling between the magnetic ordering and the resistivity as compared with

Figure 4. The temperature dependent resistivity for La0.7Sr0.3Mn1-xTixO3 compounds (x = 0.0(a), 0.05(b), 0.10(c), and 0.15(d)) under **Figure 4.** The temperature dependent resistivity for La0.7Sr0.3Mn1-xTixO3 compounds (x = 0.0(a), 0.05(b), 0.10(c), and 0.15(d)) under magnetic field H = 0, 1, 3, and 5T. Arrows indicate the resistivity maximum temperature Tρ,max. The inset in (d) is the plot of resistivity of x = 0.15 compound (with log scale) in H = 0T. 0 50 100 150 200 250 300 350 Temperature (K) 0 50 100 150 200 250 300 350 Temperature (K)

0 100 200 300 <sup>10</sup><sup>0</sup>

0

Figure 4. The temperature dependent resistivity for La0.7Sr0.3Mn1-xTixO3 compounds (x = 0.0(a), 0.05(b), 0.10(c), and 0.15(d)) under

#### magnetic field H = 0, 1, 3, and 5T. Arrows indicate the resistivity maximum temperature Tρ,max. The inset in (d) is the plot of resistivity **3.2. Zr-substituted perovskites La0.7Sr0.3Mn1-xZrxO3[23]** magnetic field H = 0, 1, 3, and 5T. Arrows indicate the resistivity maximum temperature Tρ,max. The inset in (d) is the plot of resistivity of x = 0.15 compound (with log scale) in H = 0T.

that of the Mn ions.

0.0 2.0

of x = 0.15 compound (with log scale) in H = 0T. 3.2. Zr-substituted perovskites La0.7Sr0.3Mn1-xZrxO3(Kim et al., 2007) We have tried to synthesize the La0.7Sr0.3Mn1-xZrxO3 compounds with different Zr contents. However, it was found that solubility limit of Zr is about x ~ 0.10, due to the large size (0.72 Å) of Zr4+. Fig. 5 is the ND patterns of La0.7Sr0.3Mn1-xZrxO3. We have tried to synthesize the La0.7Sr0.3Mn1-xZrxO3 compounds with different Zr contents. However, it was found that solubility limit of Zr is about x ~ 0.10, due to the large size (0.72 Å) of Zr4+. Fig. 5 is the ND patterns of La0.7Sr0.3Mn1-xZrxO3. It reveals that Zr goes only to the Mn-site. A single phase of La0.7Sr0.3Mn1-xZrxO3 was obtained for x≤0.1, which exhibits a rhombohedral structure from 10 K to RT. An impurity La2Zr2O7 phase was found for x>0.1 samples. The refined lattice parameters *a, c* and unit cell volume increase with Zr content due to the large ionic radius of the Zr ions as compared to that of the Mn ions. 3.2. Zr-substituted perovskites La0.7Sr0.3Mn1-xZrxO3(Kim et al., 2007) We have tried to synthesize the La0.7Sr0.3Mn1-xZrxO3 compounds with different Zr contents. However, it was found that solubility limit of Zr is about x ~ 0.10, due to the large size (0.72 Å) of Zr4+. Fig. 5 is the ND patterns of La0.7Sr0.3Mn1-xZrxO3. It reveals that Zr goes only to the Mn-site. A single phase of La0.7Sr0.3Mn1-xZrxO3 was obtained for x≤0.1, which exhibits a rhombohedral structure from 10 K to RT. An impurity La2Zr2O7 phase was found for x>0.1 samples. The refined lattice parameters a, c and unit cell volume increase with Zr content due to the large ionic radius of the Zr ions as compared to

+La<sup>2</sup> Zr<sup>2</sup> 10 Yclac with R-3C Figure 5. Neutron diffraction patterns of La0.7Sr0.3Mn0.9Zr0.1O3 and La0.7Sr0.3Mn0.85Zr0.15O3 at room temperature. **Figure 5.** Neutron diffraction patterns of La0.7Sr0.3Mn0.9Zr0.1O3 and La0.7Sr0.3Mn0.85Zr0.15O3 at room temperature.

2θ (degree)


8

)

0 2 4 6 Bragg-position Intensirty (103 Fig. 6 plots the average Mn-O bond length,Mn-O-Mn bond angle and bandwidth W obtained from the refined ND patterns at RT and 10K. Similar to those of the Ti substituted samples, the Mn-O bond length increases, while Mn-O-Mn bond angle and band width W decrease with the increase of Zr content. The decrease of the bandwidth W will reduce the overlap between the O-2p and the Mn-3d orbitals, which will reduce the exchange interactions between Mn-Mn in this system. This is confirmed by the results that the reduction in magnetic moments and the Curie temperature with Fig. 6 plots the average Mn-O bond length,Mn-O-Mn bond angle and bandwidth W obtained from the refined ND patterns at RT and 10K. Similar to those of the Ti substituted samples, the Mn-O bond length increases, while Mn-O-Mn bond angle and band width W decrease with the increase of Zr content. The decrease of the bandwidth W will reduce the overlap between the O-2p and the Mn-3d orbitals, which will reduce the exchange interactions between Mn-

Yobs-Yclac

increased Zr content. A metallic-like behavior was observed for the La0.7Sr0.3Mn1-xZrxO3 at low temperature. The

contribution from the two-magnon scattering in resistivity becomes larger with increasing Zr content.

20 40 60 80 100

Mn in this system. This is confirmed by the results that the reduction in magnetic moments and the Curie temperature with increased Zr content. A metallic-like behavior was observed for the La0.7Sr0.3Mn1-xZrxO3 at low temperature. The contribution from the two-magnon scattering in resistivity becomes larger with increasing Zr content.

Figure 6. Average Mn-O bond lengths (a), Mn-O-Mn bond angles (b), and electronic bandwidth parameter W, of La0.7Sr0.3Mn1-xZrxO3 (x = **Figure 6.** Average Mn-O bond lengths (a), Mn-O-Mn bond angles (b), and electronic bandwidth parameter *W*, of La0.7Sr0.3Mn1-xZrxO3 (x = 0.0, 0.03, 0.05, 0.10) at room temperature and at 10K.

#### We have tried to synthesize the La0.7Sr0.3Mn1-xZrxO3 compounds with different Zr contents. However, it was found that 0.0, 0.03, 0.05, 0.10) at room temperature and at 10K. **3.3. Cu-substituted perovskites La0.7Sr0.3Mn1-xCuxO3 [24, 25]**

between Mn-Mn ions. The separation of TC and the resistivity maximum temperature Tρ,max becomes wider as Ti

content increases due to the weak coupling between the magnetic ordering and the resistivity as compared with

0 100 200 300 <sup>10</sup><sup>0</sup>

0 1x10<sup>6</sup> 2x10<sup>6</sup> 3x10<sup>6</sup> 4x10<sup>6</sup>

20 40 60 80 100

2θ (degree)

Fig. 6 plots the average Mn-O bond length,Mn-O-Mn bond angle and bandwidth W obtained from the refined ND

patterns at RT and 10K. Similar to those of the Ti substituted samples, the Mn-O bond length increases, while Mn-O-Mn

bond angle and band width W decrease with the increase of Zr content. The decrease of the bandwidth W will reduce

the overlap between the O-2p and the Mn-3d orbitals, which will reduce the exchange interactions between Mn-Mn in

this system. This is confirmed by the results that the reduction in magnetic moments and the Curie temperature with

increased Zr content. A metallic-like behavior was observed for the La0.7Sr0.3Mn1-xZrxO3 at low temperature. The

contribution from the two-magnon scattering in resistivity becomes larger with increasing Zr content.

Figure 5. Neutron diffraction patterns of La0.7Sr0.3Mn0.9Zr0.1O3 and La0.7Sr0.3Mn0.85Zr0.15O3 at room temperature.

La0.7Sr0.3Mn0.85Zr0.15O<sup>3</sup>

+La<sup>2</sup> Zr<sup>2</sup> O7

0.0

0.1

Tc

0.2

10<sup>2</sup> 10<sup>4</sup> 10<sup>6</sup>

3.2. Zr-substituted perovskites La0.7Sr0.3Mn1-xZrxO3(Kim et al., 2007)

La0.7Sr0.3Mn0.85Zr0.15O<sup>3</sup>

+La<sup>2</sup> Zr<sup>2</sup> O7

We have tried to synthesize the La0.7Sr0.3Mn1-xZrxO3 compounds with different Zr contents. However, it was found that solubility limit of Zr is about x ~ 0.10, due to the large size (0.72 Å) of Zr4+. Fig. 5 is the ND patterns of La0.7Sr0.3Mn1-xZrxO3. It reveals that Zr goes only to the Mn-site. A single phase of La0.7Sr0.3Mn1-xZrxO3 was obtained for x≤0.1, which exhibits a rhombohedral structure from 10 K to RT. An impurity La2Zr2O7 phase was found for x>0.1 samples. The refined lattice parameters a, c and unit cell volume increase with Zr content due to the large ionic radius of the Zr ions as compared to

20 40 60 80 100

2θ (degree)

Fig. 6 plots the average Mn-O bond length,Mn-O-Mn bond angle and bandwidth W obtained from the refined ND patterns at RT and 10K. Similar to those of the Ti substituted samples, the Mn-O bond length increases, while Mn-O-Mn bond angle and band width W decrease with the increase of Zr content. The decrease of the bandwidth W will reduce the overlap between the O-2p and the Mn-3d orbitals, which will reduce the exchange interactions between Mn-Mn in this system. This is confirmed by the results that the reduction in magnetic moments and the Curie temperature with increased Zr content. A metallic-like behavior was observed for the La0.7Sr0.3Mn1-xZrxO3 at low temperature. The

Tc =308 K Tρ,max=295 K H=0T

ρ (MΩ cm)

between Mn-Mn ions. The separation of TC and the resistivity maximum temperature Tρ,max becomes wider as Ti content increases due to the weak coupling between the magnetic ordering and the resistivity as compared with

(d) x=0.15

0 100 200 300 <sup>10</sup><sup>0</sup> 10<sup>2</sup> 10<sup>4</sup> 10<sup>6</sup>

Tc

(b) x=0.05

Tc =175 K Tρ, m ax=143 K

Tc =308 K Tρ,max=295 K H=0T

 H=0T H=1T H=3T H=5T

(b) x=0.05

(d) x=0.15

Tc =175 K Tρ, m ax=143 K

Tc

**Figure 4.** The temperature dependent resistivity for La0.7Sr0.3Mn1-xTixO3 compounds (x = 0.0(a), 0.05(b), 0.10(c), and 0.15(d)) under magnetic field H = 0, 1, 3, and 5T. Arrows indicate the resistivity maximum temperature Tρ,max. The

0 50 100 150 200 250 300 350

Temperature (K)

ρ (MΩ cm)

We have tried to synthesize the La0.7Sr0.3Mn1-xZrxO3 compounds with different Zr contents. However, it was found that solubility limit of Zr is about x ~ 0.10, due to the large size (0.72 Å) of Zr4+. Fig. 5 is the ND patterns of La0.7Sr0.3Mn1-xZrxO3. It reveals that Zr goes only to the Mn-site. A single phase of La0.7Sr0.3Mn1-xZrxO3 was obtained for x≤0.1, which exhibits a rhombohedral structure from 10 K to RT. An impurity La2Zr2O7 phase was found for x>0.1 samples. The refined lattice parameters *a, c* and unit cell volume increase with Zr content due

3.2. Zr-substituted perovskites La0.7Sr0.3Mn1-xZrxO3(Kim et al., 2007)

Tc=369 K Tρ, max =365 K

> H=0T H=1T H=3T H=5T

of x = 0.15 compound (with log scale) in H = 0T.

to the large ionic radius of the Zr ions as compared to that of the Mn ions.

2θ (degree)

<sup>12</sup> Yobs

2θ (degree)

20 40 60 80 100

 Yclac Yobs-Yclac Bragg-position

**Figure 5.** Neutron diffraction patterns of La0.7Sr0.3Mn0.9Zr0.1O3 and La0.7Sr0.3Mn0.85Zr0.15O3 at room temperature.

La0.7Sr0.3Mn0.9Zr0.1O<sup>3</sup>

<sup>12</sup> Yobs

La0.7Sr0.3Mn0.9Zr0.1O<sup>3</sup> with R-3C

20 40 60 80 100

 Yclac Yobs-Yclac Bragg-position

Fig. 6 plots the average Mn-O bond length,Mn-O-Mn bond angle and bandwidth W obtained from the refined ND patterns at RT and 10K. Similar to those of the Ti substituted samples, the Mn-O bond length increases, while Mn-O-Mn bond angle and band width W decrease with the increase of Zr content. The decrease of the bandwidth W will reduce the overlap between the O-2p and the Mn-3d orbitals, which will reduce the exchange interactions between Mn-

Figure 5. Neutron diffraction patterns of La0.7Sr0.3Mn0.9Zr0.1O3 and La0.7Sr0.3Mn0.85Zr0.15O3 at room temperature.

contribution from the two-magnon scattering in resistivity becomes larger with increasing Zr content.

0 50 100 150 200 250 300 350

Temperature (K)

Figure 4. The temperature dependent resistivity for La0.7Sr0.3Mn1-xTixO3 compounds (x = 0.0(a), 0.05(b), 0.10(c), and 0.15(d)) under magnetic field H = 0, 1, 3, and 5T. Arrows indicate the resistivity maximum temperature Tρ,max. The inset in (d) is the plot of resistivity

0 1x10<sup>6</sup> 2x10<sup>6</sup> 3x10<sup>6</sup> 4x10<sup>6</sup>

0.0 0.1 0.2

La0.7Sr0.3MnO3.

6.0 (c) x=0.10

0.0 2.0 4.0 6.0 (c) x=0.10

0 1 2 3 La0.7Sr<sup>3</sup> Mn1-xTi<sup>x</sup> O3 (a) x=0.00

ρ (Ω cm)

ρ(10-1

Ω cm)

3 La0.7Sr<sup>3</sup>

(a) x=0.00

Mn1-xTi<sup>x</sup> O3

266 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

La0.7Sr0.3MnO3.

0.0 2.0 4.0

0 1 2

ρ (Ω cm)

ρ(10-1

Ω cm)

0 50 100 150 200 250 300 350

Tc =323 K Tρ, max=208 K

0 50 100 150 200 250 300 350

Temperature (K)

that of the Mn ions.

Intensirty (103

)

that of the Mn ions.

with R-3C

Intensirty (103

)

Tc =323 K Tρ, max=208 K

Temperature (K)

Tc=369 K Tρ, max =365 K Tc

inset in (d) is the plot of resistivity of x = 0.15 compound (with log scale) in H = 0T.

of x = 0.15 compound (with log scale) in H = 0T.

**3.2. Zr-substituted perovskites La0.7Sr0.3Mn1-xZrxO3[23]**

solubility limit of Zr is about x ~ 0.10, due to the large size (0.72 Å) of Zr4+. Fig. 5 is the ND patterns of La0.7Sr0.3Mn1-xZrxO3. It reveals that Zr goes only to the Mn-site. A single phase of La0.7Sr0.3Mn1-xZrxO3 was obtained for x≤0.1, which exhibits a rhombohedral structure from 10 K to RT. An impurity La2Zr2O7 phase was found for x>0.1 samples. The refined lattice parameters a, c and unit cell volume increase with Zr content due to the large ionic radius of the Zr ions as compared to 3.3. Cu-substituted perovskites La0.7Sr0.3Mn1-xCuxO3 (Kim et al., 2005b; 2008) La0.7Sr0.3Mn1-xCuxO3 samples with 0 <sup>p</sup> <sup>n</sup> x i<sup>α</sup> <sup>z</sup> 0.20 were prepared. XRD and ND patterns indicate that Cu goes into the Mn sites and all samples can be refined with the rhombohedral structure with the R 3 c space-group from 10 K to RT(Kim et al., 2005b; 2008). It was noticed that the lattice parameters a,c and the unit cell volumes decrease with increasing Cu content at RT, while they remain nearly constant with increasing Cu content at 10 K. Since Cu2+ has a larger radius (0.73Å) than those of Mn3+ and Mn4+, both Cu2+ and Cu3+ states may appear in these compounds accounting for the abnormal La0.7Sr0.3Mn1-xCuxO3 samples with 0 *np* x *ziα* 0.20 were prepared. XRD and ND patterns indicate that Cu goes into the Mn sites and all samples can be refined with the rhombohedral structure with the R 3 ¯ c space-group from 10 K to RT [24, 25]. It was noticed that the lattice parameters a,c and the unit cell volumes decrease with increasing Cu content at RT, while they remain nearly constant with increasing Cu content at 10 K. Since Cu2+ has a larger radius (0.73Å) than those of Mn3+ and Mn4+, both Cu2+ and Cu3+ states may appear in these compounds accounting for the abnormal changes of the unit cell volume and the lattice parameters with Cu content. The refined magnetic moments of Mn atoms at 10 K and RT decrease with increasing Cu content. These results agree well with the values obtained from magnetic measurements. There is no sign of antiferromagnetic couplings between Cu and Mn ions.

> changes of the unit cell volume and the lattice parameters with Cu content. The refined magnetic moments of Mn atoms at 10 K and RT decrease with increasing Cu content. These results agree well with the values obtained from magnetic measurements. There is no sign of antiferromagnetic couplings between Cu and Mn ions. Fig. 7 plots the Mn-O bond length and Mn-O-Mn bond angle obtained from the ND data. The changes of the Mn-O bond Fig. 7 plots the Mn-O bond length and Mn-O-Mn bond angle obtained from the ND data. The changes of the Mn-O bond length vs. Cu content show similar trend corresponding to the changes of lattice parameters at 10 K and RT. The average bond angle increases and reaches to a maximum at x = 0.10, then slightly decreases for x > 0.1 at 10K an RT. The bandwidth *W* of the Cu substituted samples shows different effects as compared those of Ti and Zr substi‐ tuted samples. At 10 K, an increase of the bandwidth *W* with Cu content *x* was observed, which may be related to the Cu valence states(See later discussion)*.* This will affect the exchange

> > RT 10K

0.00 0.05 0.10 0.15 0.20

Cu content X

0.095

0.096

W (arb. units)

dMn-O (A)

0.097

1.945

1.950

1.955

length vs. Cu content show similar trend corresponding to the changes of lattice parameters at 10 K and RT. The average

bandwidth W of the Cu substituted samples shows different effects as compared those of Ti and Zr substituted samples.

bond angle increases and reaches to a maximum at x = 0.10, then slightly decreases for x > 0.1 at 10K an RT. The

At 10 K, an increase of the bandwidth W with Cu content x was observed, which may be related to the Cu valence

in the Mn3+-O-Mn4+ ferromagnetic double exchange interactions, a direct result of the reduction in Mn3+ ions.

0.00 0.05 0.10 0.15 0.20

Cu content X

states( See later discussion). This will affect the exchange mechanisms and magnetic interactions in this system and is observed through the reduction in ferromagnetic moments with increased Cu content. This indicates a likely reduction

> 165.6 165.9 166.2 166.5 166.8

<Mn-O-Mn> (degree)

0.00 0.02 0.04 0.06 0.08 0.10

Zr content X

0.097 Zr content X (c)

0.0, 0.03, 0.05, 0.10) at room temperature and at 10K.

O3

La0.7Sr0.3Mn1-xZr<sup>x</sup>

(b)

165.0

165.5

<Mn-O-Mn> (degree)

Figure 6. Average Mn-O bond lengths (a), Mn-O-Mn bond angles (b), and electronic bandwidth parameter W, of La0.7Sr0.3Mn1-xZrxO3 (x =

La0.7Sr0.3Mn1-xCuxO3 samples with 0 <sup>p</sup> <sup>n</sup> x i<sup>α</sup> <sup>z</sup> 0.20 were prepared. XRD and ND patterns indicate that Cu goes into the

Mn sites and all samples can be refined with the rhombohedral structure with the R 3 c space-group from 10 K to RT(Kim et al., 2005b; 2008). It was noticed that the lattice parameters a,c and the unit cell volumes decrease with increasing Cu content at RT, while they remain nearly constant with increasing Cu content at 10 K. Since Cu2+ has a larger radius (0.73Å)

than those of Mn3+ and Mn4+, both Cu2+ and Cu3+ states may appear in these compounds accounting for the abnormal changes of the unit cell volume and the lattice parameters with Cu content. The refined magnetic moments of Mn atoms at 10 K and RT decrease with increasing Cu content. These results agree well with the values obtained from magnetic

Fig. 7 plots the Mn-O bond length and Mn-O-Mn bond angle obtained from the ND data. The changes of the Mn-O bond length vs. Cu content show similar trend corresponding to the changes of lattice parameters at 10 K and RT. The average

bandwidth W of the Cu substituted samples shows different effects as compared those of Ti and Zr substituted samples.

increase the magnetic ordering temperature TC at a low Cu-doping ratio. At a high doping ratio, the magnetic dilution effect of Cu is predominant, which gives rise to a sharp drop in the Curie temperature (Tc=250 K) for x=0.15. This is also related to the Cu ionic states. While the signature separation between the zero field cooling(ZFC) and field cooling (FC) curves is present, the complex curves observed in the other 3d-substituted systems is only weakly observable in the x=0.15 sample. This indicates there may be antiferromagnetic ordering up to 50 K that is too small to be observable

3.3. Cu-substituted perovskites La0.7Sr0.3Mn1-xCuxO3 (Kim et al., 2005b; 2008)

measurements. There is no sign of antiferromagnetic couplings between Cu and Mn ions.

166.0

166.5

0.094 0.095 0.096

1.945 1.950 1.955 1.960 1.965 (a)

 RT 10K

W (arb. units)

dMn-O (Α)

mechanisms and magnetic interactions in this system and is observed through the reduction in ferromagnetic moments with increased Cu content. This indicates a likely reduction in the Mn3+-O-Mn4+ ferromagnetic double exchange interactions, a direct result of the reduction in Mn3+ ions. At 10 K, an increase of the bandwidth W with Cu content x was observed, which may be related to the Cu valence states( See later discussion). This will affect the exchange mechanisms and magnetic interactions in this system and is observed through the reduction in ferromagnetic moments with increased Cu content. This indicates a likely reduction in the Mn3+-O-Mn4+ ferromagnetic double exchange interactions, a direct result of the reduction in Mn3+ ions.

**Figure 7.** The average Mn-O bond lengths (a), Mn-O-Mn bond angles (b) and band width (c) of La0.7Sr0.3Mn1-xCuxO3 at RT and at 10 K. Figure 7. The average Mn-O bond lengths (a), Mn-O-Mn bond angles (b) and band width (c) of La0.7Sr0.3Mn1-xCuxO3 at RT and at 10 K.

X-ray photoelectron spectra (XPS) was used to determine the Cu valence states in these compounds. Fig. 8 plots the mole percent of different Cu ions in the Mn sites obtained by fitting the XPS data. The Two kinds of Cu ions (Cu3+ and Cu2+) were observed when x > 0.1, unlike the other metal substituted systems. The binding energies of both Cu2+ and Cu3+ states shift to the lower BE energy region, which suggests a strong hybridization between the Cu-*2p* state and the O-*1s* state. X-ray photoelectron spectra (XPS) was used to determine the Cu valence states in these compounds. Fig. 8 plots the mole percent of different Cu ions in the Mn sites obtained by fitting the XPS data. The Two kinds of Cu ions (Cu3+ and Cu2+ ) were observed when x > 0.1, unlike the other metal substituted systems. The binding energies of both Cu2+ and Cu3+ states shift to the lower BE energy region, which suggests a strong hybridization between the Cu-2p state and the O-1s state.

Figure 8. The mole percent of different Cu ions in the Mn sites obtained from XPS **Figure 8.** The mole percent of different Cu ions in the Mn sites obtained from XPS

Fig. 9 plots the Curie temperatures for the Cu substituted samples. The TC decreases with increased Cu content. It was found that the low Cu-doped samples (x= 0.05, Tc=365 K) show almost no decrease in TC as compared to the parent LSMO. This suggests that the Cu-doping could enhance the exchange coupling of Mn3+-Mn4+ due to the Cu2+, and Fig. 9 plots the Curie temperatures for the Cu substituted samples. The *TC* decreases with increased Cu content. It was found that the low Cu-doped samples (x= 0.05, Tc=365 K) show almost no decrease in TC as compared to the parent LSMO. This suggests that the Cu-doping could enhance the exchange coupling of Mn3+-Mn4+ due to the Cu2+, and increase the magnetic

0.00 0.05 0.10 0.15 0.20 <sup>0</sup>

Cu content x

La0.7Sr0.3Mn1-xCu<sup>x</sup>

O3

Figure 9. The Curie temperature of La0.7Sr0.3Mn1-xCuxO3.

Tc (K)

within the resolution of the neutron diffraction analysis.

Figure 8. The mole percent of different Cu ions in the Mn sites obtained from XPS

0.0 0.1 0.2 0.3

 Cu2+ Cu3+ Mn3+ Mn4+

Cu content x

Figure 7. The average Mn-O bond lengths (a), Mn-O-Mn bond angles (b) and band width (c) of La0.7Sr0.3Mn1-xCuxO3 at RT and at 10 K.

X-ray photoelectron spectra (XPS) was used to determine the Cu valence states in these compounds. Fig. 8 plots the mole

percent of different Cu ions in the Mn sites obtained by fitting the XPS data. The Two kinds of Cu ions (Cu3+ and Cu2+ )

were observed when x > 0.1, unlike the other metal substituted systems. The binding energies of both Cu2+ and Cu3+

states shift to the lower BE energy region, which suggests a strong hybridization between the Cu-2p state and the O-1s

Fig. 9 plots the Curie temperatures for the Cu substituted samples. The TC decreases with increased Cu content. It was

increase the magnetic ordering temperature TC at a low Cu-doping ratio. At a high doping ratio, the magnetic dilution

found that the low Cu-doped samples (x= 0.05, Tc=365 K) show almost no decrease in TC as compared to the parent

At 10 K, an increase of the bandwidth W with Cu content x was observed, which may be related to the Cu valence states( See later discussion). This will affect the exchange mechanisms and magnetic interactions in this system and is observed through the reduction in ferromagnetic moments with increased Cu content. This indicates a likely reduction in the Mn3+-O-Mn4+ ferromagnetic double exchange interactions, a direct result of the reduction in Mn3+ ions. ordering temperature TC at a low Cu-doping ratio. At a high doping ratio, the magnetic dilution effect of Cu is predominant, which gives rise to a sharp drop in the Curie temperature (Tc=250 K) for x=0.15. This is also related to the Cu ionic states. While the signature separation between the zero field cooling(ZFC) and field cooling (FC) curves is present, the complex curves observed in the other 3d-substituted systems is only weakly observable in the x=0.15 sample. This indicates there may be antiferromagnetic ordering up to 50 K that is too small to be observable within the resolution of the neutron diffraction analysis. effect of Cu is predominant, which gives rise to a sharp drop in the Curie temperature (Tc=250 K) for x=0.15. This is also related to the Cu ionic states. While the signature separation between the zero field cooling(ZFC) and field cooling (FC) curves is present, the complex curves observed in the other 3d-substituted systems is only weakly observable in the x=0.15 sample. This indicates there may be antiferromagnetic ordering up to 50 K that is too small to be observable within the resolution of the neutron diffraction analysis.

state.

0.0

0.2

0.4

Mole % of Mn-site

0.6

Figure 7. The average Mn-O bond lengths (a), Mn-O-Mn bond angles (b) and band width (c) of La0.7Sr0.3Mn1-xCuxO3 at RT and at 10 K. Figure 9. The Curie temperature of La0.7Sr0.3Mn1-xCuxO3. **Figure 9.** The Curie temperature of La0.7Sr0.3Mn1-xCuxO3.

mechanisms and magnetic interactions in this system and is observed through the reduction in ferromagnetic moments with increased Cu content. This indicates a likely reduction in the Mn3+-O-Mn4+ ferromagnetic double exchange interactions, a direct result of the reduction in

> RT 10K

**Figure 7.** The average Mn-O bond lengths (a), Mn-O-Mn bond angles (b) and band width (c) of La0.7Sr0.3Mn1-xCuxO3 at

X-ray photoelectron spectra (XPS) was used to determine the Cu valence states in these compounds. Fig. 8 plots the mole percent of different Cu ions in the Mn sites obtained by fitting the XPS data. The Two kinds of Cu ions (Cu3+ and Cu2+) were observed when x > 0.1, unlike the other metal substituted systems. The binding energies of both Cu2+ and Cu3+ states shift to the lower BE energy region, which suggests a strong hybridization between the Cu-*2p* state

0.0 0.1 0.2 0.3

 Cu2+ Cu3+ Mn3+ Mn4+

Cu content x

Fig. 9 plots the Curie temperatures for the Cu substituted samples. The *TC* decreases with increased Cu content. It was found that the low Cu-doped samples (x= 0.05, Tc=365 K) show almost no decrease in TC as compared to the parent LSMO. This suggests that the Cu-doping could enhance the exchange coupling of Mn3+-Mn4+ due to the Cu2+, and increase the magnetic

Figure 8. The mole percent of different Cu ions in the Mn sites obtained from XPS

within the resolution of the neutron diffraction analysis.

0.00 0.05 0.10 0.15 0.20 <sup>0</sup>

Cu content x

La0.7Sr0.3Mn1-xCu<sup>x</sup>

O3

Figure 9. The Curie temperature of La0.7Sr0.3Mn1-xCuxO3.

Tc (K)

0.00 0.05 0.10 0.15 0.20

Cu content X

state.

0.0

**Figure 8.** The mole percent of different Cu ions in the Mn sites obtained from XPS

0.2

0.4

Mole % of Mn-site

0.6

0.095

0.096

W (arb. units)

dMn-O (A)

0.097

1.945 1.950 1.955

268 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

0.00 0.02 0.04 0.06 0.08 0.10

Zr content X

0.097 Zr content X (c)

0.0, 0.03, 0.05, 0.10) at room temperature and at 10K.

O3

La0.7Sr0.3Mn1-xZr<sup>x</sup>

(b)

165.0

165.5

<Mn-O-Mn> (degree)

Figure 6. Average Mn-O bond lengths (a), Mn-O-Mn bond angles (b), and electronic bandwidth parameter W, of La0.7Sr0.3Mn1-xZrxO3 (x =

La0.7Sr0.3Mn1-xCuxO3 samples with 0 <sup>p</sup> <sup>n</sup> x i<sup>α</sup> <sup>z</sup> 0.20 were prepared. XRD and ND patterns indicate that Cu goes into the

Mn sites and all samples can be refined with the rhombohedral structure with the R 3 c space-group from 10 K to RT(Kim et al., 2005b; 2008). It was noticed that the lattice parameters a,c and the unit cell volumes decrease with increasing Cu content at RT, while they remain nearly constant with increasing Cu content at 10 K. Since Cu2+ has a larger radius (0.73Å)

than those of Mn3+ and Mn4+, both Cu2+ and Cu3+ states may appear in these compounds accounting for the abnormal changes of the unit cell volume and the lattice parameters with Cu content. The refined magnetic moments of Mn atoms at 10 K and RT decrease with increasing Cu content. These results agree well with the values obtained from magnetic

Fig. 7 plots the Mn-O bond length and Mn-O-Mn bond angle obtained from the ND data. The changes of the Mn-O bond length vs. Cu content show similar trend corresponding to the changes of lattice parameters at 10 K and RT. The average

bandwidth W of the Cu substituted samples shows different effects as compared those of Ti and Zr substituted samples.

165.6 165.9 166.2 166.5 166.8

<Mn-O-Mn> (degree)

X-ray photoelectron spectra (XPS) was used to determine the Cu valence states in these compounds. Fig. 8 plots the mole

increase the magnetic ordering temperature TC at a low Cu-doping ratio. At a high doping ratio, the magnetic dilution effect of Cu is predominant, which gives rise to a sharp drop in the Curie temperature (Tc=250 K) for x=0.15. This is also related to the Cu ionic states. While the signature separation between the zero field cooling(ZFC) and field cooling (FC) curves is present, the complex curves observed in the other 3d-substituted systems is only weakly observable in the x=0.15 sample. This indicates there may be antiferromagnetic ordering up to 50 K that is too small to be observable

bond angle increases and reaches to a maximum at x = 0.10, then slightly decreases for x > 0.1 at 10K an RT. The

3.3. Cu-substituted perovskites La0.7Sr0.3Mn1-xCuxO3 (Kim et al., 2005b; 2008)

measurements. There is no sign of antiferromagnetic couplings between Cu and Mn ions.

0.00 0.05 0.10 0.15 0.20

Cu content X

166.0

166.5

0.094 0.095 0.096

1.945 1.950 1.955 1.960 1.965 (a)

 RT 10K

W (arb. units)

dMn-O (Α)

Mn3+ ions.

RT and at 10 K.

and the O-*1s* state.

Figure 10. Temperature dependent resistivities of La0.7Sr0.3Mn1-xCuxO3 (x = 0.10(a), 0.15(b), 0.20(c)) under applied magnetic fields of H = 0, 1, and 3T. **Figure 10.** Temperature dependent resistivities of La0.7Sr0.3Mn1-xCuxO3 (x = 0.10(a), 0.15(b), 0.20(c)) under applied mag‐ netic fields of H = 0, 1, and 3T.

Fig. 9 plots the Curie temperatures for the Cu substituted samples. The TC decreases with increased Cu content. It was found that the low Cu-doped samples (x= 0.05, Tc=365 K) show almost no decrease in TC as compared to the parent LSMO. This suggests that the Cu-doping could enhance the exchange coupling of Mn3+-Mn4+ due to the Cu2+, and The temperature dependence of resistivity under various applied fields was measured using PPMS and shown in Fig. 10. With increasing Cu content, the resistivity of the compound increases, while the resistivity decreases with increasing magnetic field. This is ascribed to a reduction of the Mn3+/Mn4+ ratio to account for the DE interaction and a reduction in the number of hopping electrons and hopping sites by Cu substitution. The resistivity shows a metal-like behavior with decreasing temperature when x is less than 0.10 samples. A MIT occurs for the x ≥ 0.15 samples (Fig. 10). A resistivity peak corresponding to the magnetic transition is present. The suppression of the resistivity by the applied magnetic field The temperature dependence of resistivity under various applied fields was measured using PPMS and shown in Fig. 10. With increasing Cu content, the resistivity of the compound increases, while the resistivity decreases with increasing magnetic field. This is ascribed to a reduction of the Mn3+/Mn4+ ratio to account for the DE interaction and a reduction in the number of hopping electrons and hopping sites by Cu substitution. The resistivity shows a metal-like behavior with decreasing temperature when x is less than 0.10 samples. A MIT occurs for the x ≥ 0.15 samples (Fig. 10). A resistivity peak corresponding to the magnetic transition is present.

3.4. Co-substituted La0.7Sr0.3Mn1-xCoxO3 (Chen et al., 2014)

respectively), with the transition from the metal to insulator.

exchange interaction (Kim et al., 2008).

occurs over the entire temperature range for all samples. The highest MR ratio of about 80% was obtained for x = 0.15 sample, which might result from the co-existence of Cu3+/ Cu2+ and the dilution effect of Cu-doping on the double

Typical ND patterns for the La0.7Sr0.3Mn1-xCoxO3 samples at RT are shown in Figure 11 (Chen et al., 2014). It is obvious that the peak intensity of (012) decreases and those of (110) and (104) increase with the increase of the cobalt content x, which is due to the different scattering lengths of Mn and Co ions. Since these peaks are related to magnetic scattering, the changes correspond to the decrease of the magnetic contribution. A Rietveld refinement for all the polycrystalline samples was carried out to understand the detailed crystal and magnetic properties. Figure 12 displays the refined lattice parameters for the La0.7Sr0.3Mn1-xCoxO3 samples at RT. It shows that the lattice parameters and unit cell volume decrease with increasing the Co content due to the fact that the radius of Co ions (0.55 Å for Co3+ and 0.40 Å for Co4+) is smaller than that of Mn ions (0.65 Å for Mn3+ and 0.53 Å for Mn4+). Generally speaking, the DE interaction strength among Mn3+ and Mn4+ can be estimated using the transfer integral, t∝cos (θ/2) and thus strongly depends on the Mn-O-Mn bond angle. The changes in θ value also have strong influence on the effective bandwidth W. For a charge-transfer insulator,

the band gap energy Eg in the insulating phase can be written as <sup>g</sup> E W =∆− , where ∆ is the charge-transfer energy

and W is the O-2p-like bandwidth (Kim et al., 2005). In fact, ∆ changes little in the La1-xSrxMnO3 system thus the bandwidth W becomes a main factor in turning the band gap energy (Harrison 1980). The decrease of bandwidth for Co-substituted compounds reduces the overlap between the O-2p and the Mn-3d orbitals. It also increases the band gap energy Eg, which in turn decreases the fragile double exchange coupling of Mn3+-O-Mn4+, as well as the magnetization and Curie temperature Tc (375 K for La0.7Sr0.3MnO3, 226 K for La0.7Sr0.3CoO3, and 184, 193, 180 K for x=0.4, 0.5 and 0.6,

The suppression of the resistivity by the applied magnetic field occurs over the entire tem‐ perature range for all samples. The highest MR ratio of about 80% was obtained for x = 0.15 sample, which might result from the co-existence of Cu3+/ Cu2+ and the dilution effect of Cudoping on the double exchange interaction [25].

## **3.4. Co-substituted La0.7Sr0.3Mn1-xCoxO3**

Typical ND patterns for the La0.7Sr0.3Mn1-xCoxO3 samples at RT are shown in Figure 11. It is obvious that the peak intensity of (012) decreases and those of (110) and (104) increase with the increase of the cobalt content x, which is due to the different scattering lengths of Mn and Co ions. Since these peaks are related to magnetic scattering, the changes correspond to the decrease of the magnetic contribution. A Rietveld refinement for all the polycrystalline samples was carried out to understand the detailed crystal and magnetic properties. Figure 12 displays the refined lattice parameters for the La0.7Sr0.3Mn1-xCoxO3 samples at RT. It shows that the lattice parameters and unit cell volume decrease with increasing the Co content due to the fact that the radius of Co ions (0.55 Å for Co3+ and 0.40 Å for Co4 +) is smaller than that of Mn ions (0.65 Å for Mn3+ and 0.53 Å for Mn4+). Generally speaking, the DE interaction strength among Mn3+ and Mn4+ can be estimated using the transfer integral, t∝cos (θ/2) and thus strongly depends on the Mn-O-Mn bond angle. The changes in θ value also have strong influence on the effective bandwidth *W*. For a charge-transfer insulator, the band gap energy *Eg* in the insulating phase can be written as *Eg* =Δ−*W* , where Δ is the charge-transfer energy and W is the O-2p-like bandwidth (Kim et al., 2005). In fact, Δ changes little in the La1-xSrxMnO3 system thus the bandwidth *W* becomes a main factor in turning the band gap energy [16]. The decrease of bandwidth for Co-substituted compounds reduces the overlap between the O-2p and the Mn-3d orbitals. It also increases the band gap energy *Eg*, which in turn decreases the fragile double exchange coupling of Mn3 +-O-Mn4+, as well as the magnetization and Curie tempera‐ ture Tc (375 K for La0.7Sr0.3MnO3, 226 K for La0.7Sr0.3CoO3, and 184, 193, 180 K for x=0.4, 0.5 and 0.6, respectively), with the transition from the metal to insulator.

13.10 13.15 13.20 13.25 13.30 13.35

166

168

<Mn-O-Mn> (degree)

Figure 12. The lattice parameters a, c, unit cell volume, bandwidth W, bond length dMn-O, and bond angle <Mn-O-Mn> for

The temperature dependent ND patterns of La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 1.0) samples are shown in Figure 13(a). The representative Bragg reflections of neutron diffraction prior to and with the addition of magnetic phase are shown in Figure 13(b). The misfits indicate the magnetic contributions. It is obvious that the (012) reflection has both nuclear and magnetic intensities and the (104), (110) reflections show little magnetic intensity for x=0.4. However, for x=1.0 sample, the (012) peak has magnetic intensity only and (104), (110) has both nuclear and magnetic intensities. There is almost no change of the intensity for x=0.5 and x=0.6 samples whether to add magnetic phase or not. The intensity of the magnetic peak (012) for LSCO decreases with the increase of temperature until vanishes finally, which is similar to LSMO-Co0.4

170

0.00 0.04 0.08 0.12 W(arb.)

Figure 11. Neutron diffraction patterns of the La0.7Sr0.3Mn1-xCoxO3 sample at RT. **Figure 11.** Neutron diffraction patterns of the La0.7Sr0.3Mn1-xCoxO3 sample at RT.

a

0.0 0.2 0.4 0.6 0.8 1.0 1.92

(i.e. the magnetic peaks (104), (110)).

Co content (x)

c

5.44 5.48 5.52 5.56 5.60

340 345 350

1.93 1.94 1.95 1.96

La0.7Sr0.3Mn1-xCoxO3 at RT.

Bond length (A)

Lattice parameter (A)

Unit cell volume A3

Structural, Magnetic and Transport Properties of B-Site Substituted Perovskite La0.7Sr0.3MnO3 http://dx.doi.org/10.5772/61770 271

Figure 11. Neutron diffraction patterns of the La0.7Sr0.3Mn1-xCoxO3 sample at RT.

(202)

(211)

(202)

(113)

(104) (110)

(012)

Intensity ( A.U.)

(024)

20 40 60 80 100 2θ(degree)

(042)

(208)

(220)

(226)

x=0.1

x=0.2

x=0.5

x=1.0

x=0.8

The suppression of the resistivity by the applied magnetic field occurs over the entire tem‐ perature range for all samples. The highest MR ratio of about 80% was obtained for x = 0.15 sample, which might result from the co-existence of Cu3+/ Cu2+ and the dilution effect of Cu-

Typical ND patterns for the La0.7Sr0.3Mn1-xCoxO3 samples at RT are shown in Figure 11. It is obvious that the peak intensity of (012) decreases and those of (110) and (104) increase with the increase of the cobalt content x, which is due to the different scattering lengths of Mn and Co ions. Since these peaks are related to magnetic scattering, the changes correspond to the decrease of the magnetic contribution. A Rietveld refinement for all the polycrystalline samples was carried out to understand the detailed crystal and magnetic properties. Figure 12 displays the refined lattice parameters for the La0.7Sr0.3Mn1-xCoxO3 samples at RT. It shows that the lattice parameters and unit cell volume decrease with increasing the Co content due to the fact that

Å for Mn3+ and 0.53 Å for Mn4+). Generally speaking, the DE interaction strength among Mn3+ and Mn4+ can be estimated using the transfer integral, t∝cos (θ/2) and thus strongly depends on the Mn-O-Mn bond angle. The changes in θ value also have strong influence on the effective bandwidth *W*. For a charge-transfer insulator, the band gap energy *Eg* in the insulating phase can be written as *Eg* =Δ−*W* , where Δ is the charge-transfer energy and W is the O-2p-like bandwidth (Kim et al., 2005). In fact, Δ changes little in the La1-xSrxMnO3 system thus the bandwidth *W* becomes a main factor in turning the band gap energy [16]. The decrease of bandwidth for Co-substituted compounds reduces the overlap between the O-2p and the Mn-3d orbitals. It also increases the band gap energy *Eg*, which in turn decreases the fragile

ture Tc (375 K for La0.7Sr0.3MnO3, 226 K for La0.7Sr0.3CoO3, and 184, 193, 180 K for x=0.4, 0.5 and

(202)

(211)

(202)

(113)

(104) (110)

(024)

20 40 60 80 100 2θ(degree)

Figure 11. Neutron diffraction patterns of the La0.7Sr0.3Mn1-xCoxO3 sample at RT.

13.10 13.15 13.20 13.25 13.30 13.35

166

168

<Mn-O-Mn> (degree)

Figure 12. The lattice parameters a, c, unit cell volume, bandwidth W, bond length dMn-O, and bond angle <Mn-O-Mn> for

The temperature dependent ND patterns of La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 1.0) samples are shown in Figure 13(a). The representative Bragg reflections of neutron diffraction prior to and with the addition of magnetic phase are shown in Figure 13(b). The misfits indicate the magnetic contributions. It is obvious that the (012) reflection has both nuclear and magnetic intensities and the (104), (110) reflections show little magnetic intensity for x=0.4. However, for x=1.0 sample, the (012) peak has magnetic intensity only and (104), (110) has both nuclear and magnetic intensities. There is almost no change of the intensity for x=0.5 and x=0.6 samples whether to add magnetic phase or not. The intensity of the magnetic peak (012) for LSCO decreases with the increase of temperature until vanishes finally, which is similar to LSMO-Co0.4

170

0.00 0.04 0.08 0.12 W(arb.) +) is smaller than that of Mn ions (0.65

+-O-Mn4+, as well as the magnetization and Curie tempera‐

x=0.1

x=0.2

x=0.5

x=1.0

x=0.8

(042)

(208)

(220)

(226)

doping on the double exchange interaction [25].

270 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

the radius of Co ions (0.55 Å for Co3+ and 0.40 Å for Co4

0.6, respectively), with the transition from the metal to insulator.

(012)

Intensity ( A.U.)

5.44 5.48 5.52 5.56 5.60

340 345 350

1.93 1.94 1.95 1.96

La0.7Sr0.3Mn1-xCoxO3 at RT.

Bond length (A)

a

**Figure 11.** Neutron diffraction patterns of the La0.7Sr0.3Mn1-xCoxO3 sample at RT.

0.0 0.2 0.4 0.6 0.8 1.0 1.92

(i.e. the magnetic peaks (104), (110)).

Co content (x)

c

Lattice parameter (A)

Unit cell volume A3

**3.4. Co-substituted La0.7Sr0.3Mn1-xCoxO3**

double exchange coupling of Mn3

Figure 12. The lattice parameters a, c, unit cell volume, bandwidth W, bond length dMn-O, and bond angle <Mn-O-Mn> for **Figure 12.** The lattice parameters a, c, unit cell volume, bandwidth *W*, bond length *dMn-O,* and bond angle <Mn-O-Mn> for La0.7Sr0.3Mn1-xCoxO3 at RT.

La0.7Sr0.3Mn1-xCoxO3 at RT. The temperature dependent ND patterns of La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 1.0) samples are shown in Figure 13(a). The representative Bragg reflections of neutron diffraction prior to and with the addition of magnetic phase are shown in Figure 13(b). The misfits indicate the magnetic contributions. It is obvious that the (012) reflection has both nuclear and magnetic intensities and the (104), (110) reflections show little magnetic intensity for x=0.4. However, for x=1.0 sample, the (012) peak has magnetic intensity only and (104), (110) has both nuclear and magnetic intensities. There is almost no The temperature dependent ND patterns of La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 1.0) samples are shown in Figure 13(a). The representative Bragg reflections of neutron diffraction prior to and with the addition of magnetic phase are shown in Figure 13(b). The misfits indicate the magnetic contributions. It is obvious that the (012) reflection has both nuclear and magnetic intensities and the (104), (110) reflections show little magnetic intensity for x=0.4. However, for x=1.0 sample, the (012) peak has magnetic intensity only and (104), (110) has both nuclear and magnetic intensities. There is almost no change of the intensity for x=0.5 and x=0.6 samples whether to add magnetic phase or not. The intensity of the magnetic peak (012) for LSCO decreases with the increase of temperature until vanishes finally, which is similar to LSMO-Co0.4 (i.e. the magnetic peaks (104), (110)).

change of the intensity for x=0.5 and x=0.6 samples whether to add magnetic phase or not. The intensity of the magnetic peak (012) for LSCO decreases with the increase of temperature until vanishes finally, which is similar to LSMO-Co0.4 (i.e. the magnetic peaks (104), (110)). The Curie temperature (TC), the coercivity (i HC), the magnetization and the resistivity of La0.7Sr0.3Mn1-xCoxO3 are shown in Figure 1. The critical Co-doping contents for the onset/ disappearance of the glassy behavior are x=0.3 and 0.8. The i HC and resistivity show a maxi‐ mum value, while the TC and magnetization show a minimum value at the critical Co-doping point (x=0.3, 0.8). The ferromagnetic ordered La0.7Sr0.3MnO3 gradually turns into disordered glassy system by the B site Co-doping, which is attributed to the break of the double exchange interaction between Mn-Mn ions and random substitution of the Mn ions. At the intermediate Co-doping region, the ferromagnetic ordered clusters embed in the antiferromagnetic ordering cluster matrix, forming the superparamagnetic-like free spin and reentrant spin glassy states. The resistivity increases due to the break of the double exchange interaction and the phase

with only the nuclear phase.

150 300 450

(a)

(c)

(b)

TC (K)

M (emu/g)

107

HC (kOe)

10-1

0.0 0.2 0.4 0.6 0.8 1.0 <sup>10</sup>-5

Co content x

Co-doping, which is attributed to the break of the double exchange interaction between Mn-Mn ions and random substitution of the Mn ions. At the intermediate Co-doping region, the ferromagnetic ordered clusters embed in the Figure 13. (a) ND patterns for the La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 1.0) samples at different temperature. (b) Representative Bragg reflections of neutron diffraction data for the La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 0.5, 0.6, 1.0) samples. Each data graphic contains two groups of plot (red for the experimental data and black for the refined data), the under one with the nuclear and magnetic phase refined and the upper one **Figure 13.** (a) ND patterns for the La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 1.0) samples at different temperature. (b) Representative Bragg reflections of neutron diffraction data for the La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 0.5, 0.6, 1.0) samples. Each data graphic contains two groups of plot (red for the experimental data and black for the refined data), the under one with the nu‐ clear and magnetic phase refined and the upper one with only the nuclear phase.

point (x=0.3, 0.8). The ferromagnetic ordered La0.7Sr0.3MnO3 gradually turns into disordered glassy system by the B site

antiferromagnetic ordering cluster matrix, forming the superparamagnetic-like free spin and reentrant spin glassy states.

separation. And the destruction of the long range exchange coupling leads to the decrease of the Curie temperature. The sparsely alignment of the cluster makes the decrease of the magnetization The resistivity increases due to the break of the double exchange interaction and the phase separation. And the destruction of the long range exchange coupling leads to the decrease of the Curie temperature. The sparsely alignment of the cluster makes the decrease of the magnetization The Curie temperature (TC), the coercivity (iHC), the magnetization and the resistivity of La0.7Sr0.3Mn1-xCoxO3 are shown in Figure 1. The critical Co-doping contents for the onset/disappearance of the glassy behavior are x=0.3 and 0.8. The iH<sup>C</sup>

and resistivity show a maximum value, while the TC and magnetization show a minimum value at the critical Co-doping

**Figure 14.** (a) Curie temperature (TC), (b) the magnetization at the field of 5 T and the ferromagnetic component ex‐ tracted from the M-H curve, (c)the coercivity at 5 K, (d) the resistivity at 70 K of La0.7Sr0.3Mn1-xCoxO3. 103 (d) ρ (Ω•cm)

Figure 13. (a) ND patterns for the La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 1.0) samples at different temperature. (b) Representative Bragg reflections of neutron diffraction data for the La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 0.5, 0.6, 1.0) samples. Each data graphic contains two groups of plot (red for the experimental data and black for the refined data), the under one with the nuclear and magnetic phase refined and the upper one The Curie temperature (TC), the coercivity (iHC), the magnetization and the resistivity of La0.7Sr0.3Mn1-xCoxO3 are shown in Figure 1. The critical Co-doping contents for the onset/disappearance of the glassy behavior are x=0.3 and 0.8. The iH<sup>C</sup> and resistivity show a maximum value, while the TC and magnetization show a minimum value at the critical Co-doping point (x=0.3, 0.8). The ferromagnetic ordered La0.7Sr0.3MnO3 gradually turns into disordered glassy system by the B site Co-doping, which is attributed to the break of the double exchange interaction between Mn-Mn ions and random substitution of the Mn ions. At the intermediate Co-doping region, the ferromagnetic ordered clusters embed in the The typical hysteresis loops at various temperatures for La0.7Sr0.3Mn0.5Co0.5O3 were plotted in figure 15. This is a representative one for the intermediate Co-doping samples (0.2<x<0.8), which is consistent with the simple cluster model (where the Co3+-Co4+ or Mn3+-Mn4+ ferro‐ magnetic clusters exist in the Co3 +-Co3 +(Mn3+) or Co4+-Co4+(Mn4+) antiferromagnetic cluster). It should be noted that the hysteresis loop at 5K shows a jump at the vicinity of 0 T. The jump disappears when the temperature is just above the Curie temperature. This suggests that the observed phenomenon is related to the competition between the ferromagnetic double exchange interaction and the antiferromagnetic superexchange interaction. The i HC is are significantly enhanced due to the freezing and pinning of the domain wall. The hysteresis loops do not show saturation under a magnetic field of 5T, which is consistent with the cluster glass behaviour. It is proposed that under a high magnetic field, the ferromagnetic clusters and partial antiferromagnetic clusters are forced to align along the direction of magnetic field. But the magnetization could not be saturated on account of the existence of antiferromagnetic clusters, so partial soft ferromagnetic clusters with small coercivity are demagnetized easily around H=0. As a result, a sudden drop of the magnetization has occurred, giving rise to a jump in the hysteresis curve. Figure 14. (a) Curie temperature (TC), (b) the magnetization at the field of 5 T and the ferromagnetic component extracted from the M-H curve, (c)the coercivity at 5 K, (d) the resistivity at 70 K of La0.7Sr0.3Mn1-xCoxO3. The typical hysteresis loops at various temperatures for La0.7Sr0.3Mn0.5Co0.5O3 were plotted in figure 15. This is a representative one for the intermediate Co-doping samples (0.2<x<0.8), which is consistent with the simple cluster model (where the Co3+-Co4+ or Mn3+-Mn4+ ferromagnetic clusters exist in the Co3+-Co3+(Mn3+) or Co4+-Co4+(Mn4+) antiferromagnetic cluster). It should be noted that the hysteresis loop at 5K shows a jump at the vicinity of 0 T. The jump disappears when the temperature is just above the Curie temperature. This suggests that the observed phenomenon is related to the competition between the ferromagnetic double exchange interaction and the antiferromagnetic superexchange interaction. The iHC is are significantly enhanced due to the freezing and pinning of the domain wall. The hysteresis loops do not show saturation under a magnetic field of 5T, which is consistent with the cluster glass behaviour. It is proposed that under a high magnetic field, the ferromagnetic clusters and partial antiferromagnetic clusters are forced to align along the direction of magnetic field. But the magnetization could not be saturated on account of the existence of antiferromagnetic clusters, so partial soft ferromagnetic clusters with small coercivity are demagnetized easily around H=0. As a result, a sudden drop of the magnetization has occurred, giving rise to a jump in

3.5. Cr-doped La0.7Sr0.3Mn1-xCrxO3 (Creel et al., 2010, 2013a, 2013b)

Figure 15. The hysteresis loops of La0.7Sr0.3Mn0.5Co0.5O3 at different temperatures. **Figure 15.** The hysteresis loops of La0.7Sr0.3Mn0.5Co0.5O3 at different temperatures.

the hysteresis curve.

#### The La0.7Sr0.3Mn1-xCrxO3 (0<x<0.6) have been prepared and influence of the Cr3+ substitution for Mn3+ was investigated **3.5. Cr-doped La0.7Sr0.3Mn1-xCrxO3 [7, 8, 9]**

separation. And the destruction of the long range exchange coupling leads to the decrease of the Curie temperature. The sparsely alignment of the cluster makes the decrease of the

**Figure 13.** (a) ND patterns for the La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 1.0) samples at different temperature. (b) Representative Bragg reflections of neutron diffraction data for the La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 0.5, 0.6, 1.0) samples. Each data graphic contains two groups of plot (red for the experimental data and black for the refined data), the under one with the nu‐

0.0 0.2 0.4 0.6 0.8 1.0 <sup>10</sup>-5

**Figure 14.** (a) Curie temperature (TC), (b) the magnetization at the field of 5 T and the ferromagnetic component ex‐

Co content x

of the cluster makes the decrease of the magnetization

Figure 13. (a) ND patterns for the La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 1.0) samples at different temperature. (b) Representative Bragg reflections of neutron diffraction data for the La0.7Sr0.3Mn1-xCoxO3 (x=0.4, 0.5, 0.6, 1.0) samples. Each data graphic contains two groups of plot (red for the experimental data and black for the refined data), the under one with the nuclear and magnetic phase refined and the upper one

The Curie temperature (TC), the coercivity (iHC), the magnetization and the resistivity of La0.7Sr0.3Mn1-xCoxO3 are shown in Figure 1. The critical Co-doping contents for the onset/disappearance of the glassy behavior are x=0.3 and 0.8. The iH<sup>C</sup> and resistivity show a maximum value, while the TC and magnetization show a minimum value at the critical Co-doping point (x=0.3, 0.8). The ferromagnetic ordered La0.7Sr0.3MnO3 gradually turns into disordered glassy system by the B site Co-doping, which is attributed to the break of the double exchange interaction between Mn-Mn ions and random substitution of the Mn ions. At the intermediate Co-doping region, the ferromagnetic ordered clusters embed in the antiferromagnetic ordering cluster matrix, forming the superparamagnetic-like free spin and reentrant spin glassy states.

> M at 5 T MFM

destruction of the long range exchange coupling leads to the decrease of the Curie temperature. The sparsely alignment

The resistivity increases due to the break of the double exchange interaction and the phase separation. And the

antiferromagnetic ordering cluster matrix, forming the superparamagnetic-like free spin and reentrant spin glassy states.

destruction of the long range exchange coupling leads to the decrease of the Curie temperature. The sparsely alignment

The resistivity increases due to the break of the double exchange interaction and the phase separation. And the

with only the nuclear phase.

150

 M at 5 T MFM

10-1 103 107

300

TC (K)

M (emu/g)

of the cluster makes the decrease of the magnetization

HC (kOe)

ρ (Ω•cm)

0.0 0.2 0.4 0.6 0.8 1.0 <sup>10</sup>-5

Co content x

450

clear and magnetic phase refined and the upper one with only the nuclear phase.

272 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

(a)

(c)

(d)

tracted from the M-H curve, (c)the coercivity at 5 K, (d) the resistivity at 70 K of La0.7Sr0.3Mn1-xCoxO3.

(b)

magnetization

with only the nuclear phase.

150 300 450

(a)

(c)

(d)

(b)

TC (K)

M (emu/g)

HC (kOe)

ρ (Ω•cm)

10-1 103 107

(Creel et al., 2010, 2013a, 2013b). Figure 14 is the ND patterns of La0.7Sr0.3Mn1-xCrxO3 (0<x<0.5) at 10 K. The magnetic contributions to the (012) and (110)+(104) peaks are evident. For x<0.2, the samples are simple ferromagnetic with magnetic moments decreasing with increasing Cr content. For 0.2≤x≤0.4, the (104) + (110) reflections become weaker, but two new peaks ((003) + (011)), inconsistent with a simple ferromagnetic solution emerge. The ((003) + (011)) peak is purely magnetic, while the (113) peak has nuclear and magnetic components. For x=0.5, the (104) + (110) reflections are present but weak, while the (003) + (011) reflections are now dominant. A single, homogeneous, long-range magnetically ordered state with compositionally-dependent charge ordering was proposed to fit the ND patterns (Creel et al., 2010, 2013b). The magnetic structures are related to the competition between Mn-Mn, Mn-Cr and Cr-Cr interactions (double-exchange and superexchange). The metal to semi-metal and semi-metal to insulator transitions can be quantitatively described as due to the localization effect of superexchange. The presence of charge ordered states above the M-I transition concentration (x) arises from the favourable energetics of Mn4+-O-Cr3+ superexchange bonds relative to Mn3+-O-Cr3+ bonds. The La0.7Sr0.3Mn1-xCrxO3 (0<x<0.6) have been prepared and influence of the Cr3+ substitution for Mn3+ was investigated [7, 8, 9]. Figure 14 is the ND patterns of La0.7Sr0.3Mn1-xCrxO3 (0<x<0.5) at 10 K. The magnetic contributions to the (012) and (110)+(104) peaks are evident. For x<0.2, the samples are simple ferromagnetic with magnetic moments decreasing with increasing Cr content. For 0.2≤x≤0.4, the (104) + (110) reflections become weaker, but two new peaks ((003) + (011)), inconsistent with a simple ferromagnetic solution emerge. The ((003) + (011)) peak is purely magnetic, while the (113) peak has nuclear and magnetic components. For x=0.5, the (104) + (110) reflections are present but weak, while the (003) + (011) reflections are now dominant. A single, homogeneous, long-range magnetically ordered state with composition‐ ally-dependent charge ordering was proposed to fit the ND patterns [8, 9]. The magnetic structures are related to the competition between Mn-Mn, Mn-Cr and Cr-Cr interactions (double-exchange and superexchange). The metal to semi-metal and semi-metal to insulator

transitions can be quantitatively described as due to the localization effect of superexchange. The presence of charge ordered states above the M-I transition concentration (x) arises from the favourable energetics of Mn4+-O-Cr3 + superexchange bonds relative to Mn3 +-O-Cr3+ bonds.

Figure 16. The neutron diffraction patterns of the La0.7Sr0.3Mn1-xCrxO3 sample at 10 K. **Figure 16.** The neutron diffraction patterns of the La0.7Sr0.3Mn1-xCrxO3 sample at 10 K.

At low temperature with small x, the net ferromagnetic behavior of the system is due to the large quantities of

Figure 17 displays the refined lattice parameters, unit cell volume, bond length dMn-O, bond angle <Mn-O-Mn>, and band

The B-site substituted LSMOs can be divided into following two groups, (1) those made with the replacement of Mn by other 3d transition metal ions and (2) those made with the replacement of Mn by non-magnetic, closed shell, metal ions such as Ti, Zr. The ionic radii of the substituted elements, the Mn-O-Mn bond angles, the Mn-O bond length, the calculated bandwidths W, and the corresponding TC's are given in Table I. It should be pointed out that neutron diffraction scattering lengths of the 3d elements are sufficiently different, and uniquely, the scattering length of Mn is

width W of La0.7Sr0.3Mn1-xCrxO3 samples at RT and 10 K. It can be seen that that the lattice parameters, cell volume and bond distance slightly decrease with increasing the Cr content x, which originates from the fact that the radius of Cr ions Figure 17. The lattice parameters a, c, unit cell volume, bandwidth *W*, bond length *dMn-O,* and **Figure 17.** The lattice parameters a, c, unit cell volume, bandwidth *W*, bond length *dMn-O,* and bond angle <Mn-O-Mn> for La0.7Sr0.3Mn1-xCrxO3 at RT and 10 K.

(0.615 Å for Cr3+) is little bit smaller than that of Mn ions (0.65 Å for Mn3+). The Mn-O-Mn bond angle decreases to a minimum for x=0.2 and then increases to a maximum for x=0.4. A similar trend can be observed for the band width W. This is consistent with the fact that a charge ordering creates a layered structure and the antiferromagnetic Cr3+-O-Mn4+ superexchange continues to drive the system towards an antiferromagnetic state. Figure 17 displays the refined lattice parameters, unit cell volume, bond length *dMn-O*, bond angle <Mn-O-Mn>, and band width *W* of La0.7Sr0.3Mn1-xCrxO3 samples at RT and 10 K. It can At low temperature with small x, the net ferromagnetic behavior of the system is due to the large quantities of Mn3+-O-Mn4+ ferromagnetic double exchanges taking place while the system is being driven towards a layered ferromagnetic structure by the antiferromagnetic Cr3+-O-Mn4+ superexchanges. In the intermediate region (0.2<x<0.4) charge ordering creates a layered

be seen that that the lattice parameters, cell volume and bond distance slightly decrease with increasing the Cr content x, which originates from the fact that the radius of Cr ions (0.615 Å for Cr3+) is little bit smaller than that of Mn ions (0.65 Å for Mn3+). The Mn-O-Mn bond angle decreases to a minimum for x=0.2 and then increases to a maximum for x=0.4. A similar trend can be observed for the band width *W*. This is consistent with the fact that a charge ordering creates a layered structure and the antiferromagnetic Cr3+-O-Mn4+ superexchange

19

The B-site substituted LSMOs can be divided into following two groups, (1) those made with the replacement of Mn by other 3d transition metal ions and (2) those made with the replacement of Mn by non-magnetic, closed shell, metal ions such as Ti, Zr. The ionic radii of the substituted elements, the Mn-O-Mn bond angles, the Mn-O bond length, the

4. Conclusion

continues to drive the system towards an antiferromagnetic state.

**4. Conclusion** 

bond angle <Mn-O-Mn> for La0.7Sr0.3Mn1-xCrxO3 at RT and 10 K.

structure and the antiferromagnetic Cr3+-O-Mn4+ superexchange continues to drive the system towards an antiferromagnetic state. As x>0.4, the antiferromagnetic Cr3+-O-Cr3+ and Cr3+-O-Mn4+ superexchange mechanisms become dominate, with charge order persisting, producing a ferrimagnetic structure in lieu of an antiferromagnetic one.

Figure 17 displays the refined lattice parameters, unit cell volume, bond length *dMn-O*, bond angle <Mn-O-Mn>, and band width *W* of La0.7Sr0.3Mn1-xCrxO3 samples at RT and 10 K. It can be seen that that the lattice parameters, cell volume and bond distance slightly decrease with increasing the Cr content x, which originates from the fact that the radius of Cr ions (0.615 Å for Cr3+) is little bit smaller than that of Mn ions (0.65 Å for Mn3+). The Mn-O-Mn bond angle decreases to a minimum for x=0.2 and then increases to a maximum for x=0.4. A similar trend can be observed for the band width *W*. This is consistent with the fact that a charge ordering creates a layered structure and the antiferromagnetic Cr3+-O-Mn4+ superexchange continues to drive the system towards an antiferromagnetic state.

## **4. Conclusion**

transitions can be quantitatively described as due to the localization effect of superexchange. The presence of charge ordered states above the M-I transition concentration (x) arises from

(208)(202)/(006)

 (024)

(113)

(110)/(104)

(011)/(003) (012)

274 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

antiferromagnetic one.

**Figure 16.** The neutron diffraction patterns of the La0.7Sr0.3Mn1-xCrxO3 sample at 10 K.

bond angle <Mn-O-Mn> for La0.7Sr0.3Mn1-xCrxO3 at RT and 10 K.

345.0 347.5 350.0 (c)

13.25 13.30 13.35 13.40 (b)

5.48 5.50 5.52 (a)

Volume (A3

for La0.7Sr0.3Mn1-xCrxO3 at RT and 10 K.

**4. Conclusion** 

)

c (A)

a (A)

4. Conclusion

continues to drive the system towards an antiferromagnetic state.

Intensity ( a.b)

20 40 60 80 100

Figure 16. The neutron diffraction patterns of the La0.7Sr0.3Mn1-xCrxO3 sample at 10 K.

1.945

166.0 166.5 167.0 167.5

1.950

bond distance (A)

<Mn-O-Mn> ( degree)

Figure 17. The lattice parameters a, c, unit cell volume, bandwidth *W*, bond length *dMn-O,* and

**Figure 17.** The lattice parameters a, c, unit cell volume, bandwidth *W*, bond length *dMn-O,* and bond angle <Mn-O-Mn>

At low temperature with small x, the net ferromagnetic behavior of the system is due to the large quantities of Mn3+-O-Mn4+ ferromagnetic double exchanges taking place while the system is being driven towards a layered ferromagnetic structure by the antiferromagnetic Cr3+-O-Mn4+ superexchanges. In the intermediate region (0.2<x<0.4) charge ordering creates a layered

Figure 17 displays the refined lattice parameters, unit cell volume, bond length *dMn-O*, bond angle <Mn-O-Mn>, and band width *W* of La0.7Sr0.3Mn1-xCrxO3 samples at RT and 10 K. It can be seen that that the lattice parameters, cell volume and bond distance slightly decrease with increasing the Cr content x, which originates from the fact that the radius of Cr ions (0.615 Å for Cr3+) is little bit smaller than that of Mn ions (0.65 Å for Mn3+). The Mn-O-Mn bond angle decreases to a minimum for x=0.2 and then increases to a maximum for x=0.4. A similar trend can be observed for the band width *W*. This is consistent with the fact that a charge ordering creates a layered structure and the antiferromagnetic Cr3+-O-Mn4+ superexchange

1.955

(d)

(e)

2θ(degree)

 RT 10K

+ superexchange bonds relative to Mn3

(042) (226)

La0.7Sr0.3Mn1-xCr<sup>x</sup>

x=0.05 x=0.1 x=0.2 x=0.3 x=0.4

At low temperature with small x, the net ferromagnetic behavior of the system is due to the large quantities of Mn3+-O-Mn4+ ferromagnetic double exchanges taking place while the system is being driven towards a layered ferromagnetic structure by the antiferromagnetic Cr3+-O-Mn4+ superexchanges. In the intermediate region (0.2<x<0.4)

 Mn-O 10K Mn-O RT

mechanisms become dominate, with charge order persisting, producing a ferrimagnetic structure in lieu of an

(0.615 Å for Cr3+) is little bit smaller than that of Mn ions (0.65 Å for Mn3+). The Mn-O-Mn bond angle decreases to a minimum for x=0.2 and then increases to a maximum for x=0.4. A similar trend can be observed for the band width W.

The B-site substituted LSMOs can be divided into following two groups, (1) those made with the replacement of Mn by other 3d transition metal ions and (2) those made with the replacement of Mn by non-magnetic, closed shell, metal ions such as Ti, Zr. The ionic radii of the substituted elements, the Mn-O-Mn bond angles, the Mn-O bond length, the calculated bandwidths W, and the corresponding TC's are given in Table I. It should be pointed out that neutron diffraction scattering lengths of the 3d elements are sufficiently different, and uniquely, the scattering length of Mn is

superexchange continues to drive the system towards an antiferromagnetic state.

19

The B-site substituted LSMOs can be divided into following two groups, (1) those made with the replacement of Mn by other 3d transition metal ions and (2) those made with the replacement of Mn by non-magnetic, closed shell, metal ions such as Ti, Zr. The ionic radii of the substituted elements, the Mn-O-Mn bond angles, the Mn-O bond length, the

x=0.5

(220) +-O-Cr3+ bonds.

the favourable energetics of Mn4+-O-Cr3

charge ordering creates a layered structure and the antiferromagnetic Cr3+-O-Mn4+ superexchange continues to drive the system towards an antiferromagnetic state. As x>0.4, the antiferromagnetic Cr3+-O-Cr3+ and Cr3+-O-Mn4+ superexchange The B-site substituted LSMOs can be divided into following two groups, (1) those made with the replacement of Mn by other 3d transition metal ions and (2) those made with the replace‐ ment of Mn by non-magnetic, closed shell, metal ions such as Ti, Zr. The ionic radii of the substituted elements, the Mn-O-Mn bond angles, the Mn-O bond length, the calculated bandwidths W, and the corresponding TC's are given in Table I. It should be pointed out that neutron diffraction scattering lengths of the 3d elements are sufficiently different, and uniquely, the scattering length of Mn is negative. This allows relatively small amounts of other elements substituted into the manganites to be accurately located in the unit cell structure by employing neutron diffraction.


This is consistent with the fact that a charge ordering creates a layered structure and the antiferromagnetic Cr3+-O-Mn4+ **Table 1.** The electronic configuration, ionic radii, Mn-O bond length and Mn-O-Mn bond angle at room temperature, and TC for La0.7Sr0.3Mn1-xMxO3 (x=0.1, M = Ti, Zr, Zn, Cr, Fe, Co, Ni, and Cu).

## **4.1. Substitution the Mn-site by 3d-transition metals (TM); Cr, Fe, Co, Ni, Cu**

For the TM-substituted LSMOs, they show the same crystal structure with space group *R*3 ¯ *c*. There are small changes in the lattice parameters, and almost the same valence-band structures near the Fermi level are observed. A decrease in TC is observed for all elements and a metal to insulator (MIT) transition occurs with increasing substitution content around x=0.2. The decrease in TC of the parent LSMO with substitution by other 3d transition metals is strongly related to the following two factors; (i) the magnetic moments of the substituting ions and (ii) the ionic radius mismatch between the substituting ion and Mn ions. For the TM-substituted LSMO, the degree of reduction in TC and saturation magnetization is related to the valence states of the substituting ions and the ionic size mismatch. The highest TC is obtained from the Cr- and Co-substituted LSMO. This is mainly due to the large magnetic moments of Cr and Co ions and also to the formation of the additional DE interactions between Cr and Mn (or between Co ions). Of the six transition metal-substituted systems, the Cr-based system presented some of the most interesting magnetic properties, followed by the Ni-based system and, finally, the Cu-based system. All systems exhibit layered magnetic behavior, a reduction in ferromagnetism with increasing transition metal content, complex magnetic interactions well below TC, and a metal to insulator transition around x~0.2 (a value also near the percolation threshold described in many works of around x~0.16). Our data for the Ni [7, 9] and Cr [8] systems strongly suggest the onset of charge ordering occurs coincident with the metal to insulator transition, and not only at specific nodal quantities. We suggest this type of ordering is highly likely in the remaining transition metal-substituted systems, but that additional analysis with neutron data is needed.

## **4.2. Substitution of the Mn-site by closed shell ions: Ti, Zr or Zn**

There are several advantages in using closed shell ions to investigate metal substituted-LSMOs. First, the closed shell ions normally do not affect the magnetic interactions between the Mn ions due to their having no magnetic moment. Second, they have inert gas configurations, and therefore do not contribute to the electron charge density. But there still remains the possibility of secondary effects such as a disturbance of the magnetic ordering and a redistribution of electron charge density by a large ionic size mismatch at the B-site, as with, e.g. Zr4+. In general a decrease in TC and MS with increasing substitution is observed for the closed shell ionsubstituted LSMOs. This is attributed to the dilution of the magnetic ions and the weakening of the ferromagnetic DE interaction between them. Substitution with Ti is selective because Ti4+ substitutes for Mn4+. It would be expected as well that Zr4+ or Zn2+ would substitute for Mn4+. However the severe ionic size mismatch of Zr4+and Zn2+ may not allow the substitution of Mn4+ by Zr4+ or Zn2+. Therefore mixed-valent Mn ions in LSMO are selectively diluted by partial substitution of Mn by these ions. LSMO has the highest TC when the ratio of Mn3+/ Mn4+ has an optimal value of 7/3. Upon substitution of the Mn ions with closed shell ions, the ratio of Mn3+/Mn4+ deviates from the optimal value of the parent compound. For Ti and Zr substitutions, the Mn3+/Mn4+ ratio increases with increasing Ti and Zr content. It should be noted that Zn2+ replaces Mn3+ which has a larger magnetic moment than Mn4+, and the difference of the magnetic moments in the Mn-sites of the Zn-substituted LSMO and Ti- or Zrsubstituted LSMO is only 0.1μB per Mn-site. In this case, the competition between DE and SE interactions is a more important control factor for predicting TC and MS. Therefore, substitution onto the Mn-site with Zn2+ produces more DE couplings and less SE couplings. In turn one would expect the Zn-substituted LSMO to have a larger MS and a higher TC than with Ti or Zr substitution, which produces less DE couplings and more SE couplings.

## **Acknowledgements**

¯ *c*.

**4.1. Substitution the Mn-site by 3d-transition metals (TM); Cr, Fe, Co, Ni, Cu**

276 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

analysis with neutron data is needed.

**4.2. Substitution of the Mn-site by closed shell ions: Ti, Zr or Zn**

For the TM-substituted LSMOs, they show the same crystal structure with space group *R*3

There are small changes in the lattice parameters, and almost the same valence-band structures near the Fermi level are observed. A decrease in TC is observed for all elements and a metal to insulator (MIT) transition occurs with increasing substitution content around x=0.2. The decrease in TC of the parent LSMO with substitution by other 3d transition metals is strongly related to the following two factors; (i) the magnetic moments of the substituting ions and (ii) the ionic radius mismatch between the substituting ion and Mn ions. For the TM-substituted LSMO, the degree of reduction in TC and saturation magnetization is related to the valence states of the substituting ions and the ionic size mismatch. The highest TC is obtained from the Cr- and Co-substituted LSMO. This is mainly due to the large magnetic moments of Cr and Co ions and also to the formation of the additional DE interactions between Cr and Mn (or between Co ions). Of the six transition metal-substituted systems, the Cr-based system presented some of the most interesting magnetic properties, followed by the Ni-based system and, finally, the Cu-based system. All systems exhibit layered magnetic behavior, a reduction in ferromagnetism with increasing transition metal content, complex magnetic interactions well below TC, and a metal to insulator transition around x~0.2 (a value also near the percolation threshold described in many works of around x~0.16). Our data for the Ni [7, 9] and Cr [8] systems strongly suggest the onset of charge ordering occurs coincident with the metal to insulator transition, and not only at specific nodal quantities. We suggest this type of ordering is highly likely in the remaining transition metal-substituted systems, but that additional

There are several advantages in using closed shell ions to investigate metal substituted-LSMOs. First, the closed shell ions normally do not affect the magnetic interactions between the Mn ions due to their having no magnetic moment. Second, they have inert gas configurations, and therefore do not contribute to the electron charge density. But there still remains the possibility of secondary effects such as a disturbance of the magnetic ordering and a redistribution of electron charge density by a large ionic size mismatch at the B-site, as with, e.g. Zr4+. In general a decrease in TC and MS with increasing substitution is observed for the closed shell ionsubstituted LSMOs. This is attributed to the dilution of the magnetic ions and the weakening of the ferromagnetic DE interaction between them. Substitution with Ti is selective because Ti4+ substitutes for Mn4+. It would be expected as well that Zr4+ or Zn2+ would substitute for Mn4+. However the severe ionic size mismatch of Zr4+and Zn2+ may not allow the substitution of Mn4+ by Zr4+ or Zn2+. Therefore mixed-valent Mn ions in LSMO are selectively diluted by partial substitution of Mn by these ions. LSMO has the highest TC when the ratio of Mn3+/ Mn4+ has an optimal value of 7/3. Upon substitution of the Mn ions with closed shell ions, the ratio of Mn3+/Mn4+ deviates from the optimal value of the parent compound. For Ti and Zr substitutions, the Mn3+/Mn4+ ratio increases with increasing Ti and Zr content. It should be noted that Zn2+ replaces Mn3+ which has a larger magnetic moment than Mn4+, and the difference of the magnetic moments in the Mn-sites of the Zn-substituted LSMO and Ti- or ZrThis work was supported by the National Natural Science Foundation of China (Grant Nos 51371009, 50971003 and 51171001), the National Basic Research Program of China (No 2010CB833104, MOST of China), the National High Technology Research and Development Program of China (No 2011AA03A403).

## **Author details**

J.B. Yang1,2\*, M.S. Kim3 , T. F. Creel3 , H. Zhao1 , X.G. Chen1 , W.B. Yelon3 and W.J. James3

\*Address all correspondence to: jbyang@pku.edu.cn

1 State Key Laboratory for Mesoscopic Physics, School of Physics, Peking University, Beijing, P. R. China

2 Collaborative Innovation Center of Quantum Matter, Beijing, P. R. China

3 Materials Research Center, Missouri University of Science and Technology, USA

## **References**


[19] Ju, H. L. ; Sohn, H. C. ; & Krishnan, K. M. (1997). Evidence for O-2p hole-driven con‐ ductivity in La1−xSrxMnO3 (0≤ x≤ 0.7) and La1−xSrxMnOz thin films, Phys. Rev. Lett., 74, 3230 (1997).

[5] Blasco, J. ; Garcia, J. ; de Teresa, J. M. ; Ibarra,M. R. ; Perez, J. ; Algarabel, P. A. ; Mar‐ quina, C. ; & Ritter, C. (1997). Structural, magnetic, and transport properties of the giant magnetoresistive perovskites La 2/3Ca 1/3Mn 1−xAl xO 3−δ Phys. Rev. B, 8905. [6] Cao, D. ; Bridges, F. ; Anderson, M. ; Ramirez, A. P. ; Olapinski, M. ; Subramanian, M. A. ; Booth, C. H. ; and Kwei, G. H. (2001). Local distortions in La0.7Ca0.3Mn1−bAbO3 (A=Ti and Ga) colossal magnetoresistance samples: Correlations with magnetization

[7] Creel, Thomas F.; Yang, Jinbo; Kahveci, Mehmet; Malik, Satish K.; Quezado, S.; Prin‐ gle, O.A.; Yelon, William B.; & James, William J. (2013a). Structural and magnetic properties of La0.7Sr0.3Mn1-xNixO3 (x<=0.4), Journal of Applied Physics, 114(1), 013911.

[8] Creel, Thomas F.; Jinbo Yang, Kahveci, Mehmet; Lamsal, Jagat; Malik, S. K.; Queza‐ do, Sylvio; Pringle, Oran A.; Yelon, William B.; & James, William J. (2010). Structural and Magnetic Properties of La0.7Sr0.3Mn1-xCrxO3 (x<= 0.5), IEEE Transactions on Mag‐

[9] Creel, Thomas F. (2013b). Structural and magnetic properties of Ni and Cr substitut‐

[10] Dai, P. ; et al., (1996). Experimental Evidence for the Dynamic Jahn-Teller Effect in

[11] De, Gennes P. G. (1960). Effects of Double Exchange in Magnetic Crystals, Phys. Rev.,

[12] Dessau, D. S. ; Saitoh, T.; Park, C. H.; Shen, Z. X. ; Villella, P. ; Hamad,a N. ; Morito‐ mo, Y. ; and Tokura, Y. (1998). k-dependent electronic structure, a large "ghost" fermi surface, and a pseudogap in a layered mangnetoresistive oxide, Phys. Rev. Lett., 81,

[13] Dho, J. ; Kim, W. S.; & Hur, N. H. (2002). Reentrant Spin Glass Behavior in Cr-Doped

[14] Geck, J.; et al., (2001). Evidence for canted antiferromagnetism in lightly doped

[15] Hwang, H. Y.; Cheong, S-W.; Radaelli, P. G.; Marezio, M.; & Batlogg, B. (1995). Lat‐ tice Effects on the Magnetoresistance in Doped LaMnO3 , Phys. Rev. Lett. 75, 914. [16] Harrison, W. A. (1980) The electronic structure and properties of solids (Freeman,

[17] Imada, M. ; Fujimori, A.; & Tokura, Y. (1998). Metal-insulator transitions, Rev. Mod.

[18] Joly, V. L. J. ; Joy, P. A. ; Date, S. K. ; & Gopinath, C. S. (2002). Two ferromagnetic phases with different spin states of Mn and Ni in LaMn0.5Ni0 .5O3, Phys. Rev. B 65,

and evidence for cluster formation, Phys. Rev. B. , 64, 184409.

ed La₀.₇Sr₀.₃MnO₃, Doctoral Dissertations. Paper 2031.

Perovskite Manganite, Phys. Rev. Lett. 89, 027202.

La1−xSrxMnO3, Phys. Rev. B 64, 144430.

San Francisco, 1980).

Phys. 70, 1039.

184416.

La0.65Ca0.35MnO3, Phys. Rev. B 54, R3694.

278 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

netics 46(6), 1832-1835.

118, 141.

192.


## **Microwave Dielectrics with Perovskite-Type Structure**

## Hitoshi Ohsato

[33] Okada, K. ; & Yamada, S. (2012). Anisotropic magnetic properties of P 1−xCaxMnO3,

[34] Ogale, S. B. ; Shreekala, R.; Bathe, R. ; Date, S. K. ; Patil, S. I. ; Hannoyer, B. ; Petit, F. ; & Marest, G. (1998). Transport properties, magnetic ordering, and hyperfine interac‐ tions in Fe-doped La0.75Ca0.25MnO3 : Localization-delocalization transition, Phys. Rev.

[35] Radaelli, P. G. ; Iannone, G., Marezio, M. ; Hwang, H. Y. ; Cheong, S-W.; Jorgensen, J. D. ; & Argyriou, D. N. (1997). Structural effects on the magnetic and transport prop‐

[36] Raveau, B. ; Maignan, A. ; & Martin, C. (1997). Insulator–Metal Transition Induced by Cr and Co Doping in Pr0.5Ca0.5MnO3, Journal of Solid State Chemistry, 130, 162–166 [37] Salamon, Myron B. ; & Jaime, M. (2001). The physics of manganites: Structure and

[38] Shannon, R. D. (1976). Revised effective ionic radii and systematic studies of intera‐ tomic distances in halides and chalcogenides, Acta Crystallogr. Sect. A: Cryst. Phys.,

[39] Tokura, Y.; Kuwahara, H.; Moritomo, Y.; Tomioka, Y.; & Asamitsu, A. (1996). Com‐ peting Instabilities and Metastable States in (Nd,Sm)1/2Sr1/2MnO 3, Phys. Rev. Lett. 76,

[40] Zener C. (1951). Interaction between the d -Shells in the Transition Metals. II. Ferro‐ magnetic Compounds of Manganese with Perovskite Structure, Phys. Rev., 82, 403.

[41] Zhao, G. M. ; Smolyaninova, V. ; Prellie,r W. ; & Keller, H. (2000). Electrical transport in the ferromagnetic state of manganites: Small-polaron metallic condition at low

erties of perovskite A1−x A′x MnO 3 (x=0.25, 0.30), Phys. Rev. B 56, 8265 (1997).

Phys. Rev. B 86, 064430

280 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

transport, Rev. Mod. Phys. 73, 583.

Diffr., Theor. Gen. Crystallogr., 751.

temperatures, Phys. Rev. Lett. 84, 6086.

B 57, 7841.

3184.

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61718

## **Abstract**

Most electroceramics are ferroelectrics, but microwave dielectrics are mostly paraelectrics with a center of symmetry *i*. Microwave dielectrics should possess a perfect crystal struc‐ ture with neither defects nor internal strain in order to be microwave friendly. They have been used in resonators and filters in mobile telecommunications devices. Perovskite and related compounds are also mostly ferroelectrics, but paraelectrics with a perovskite structure also exist, and are used in microwave dielectrics. Owing to the flexibility of the perovskite structure, many kinds of microwave dielectrics with a perovskite structure have been designed for microwave dielectrics. In this chapter, simple and complex perov‐ skite, and perovskite related materials such as pseudo-tungsten-bronze solid solutions and homologous compounds are introduced for microwave dielectrics. The microwave dielectric properties are revealed through the crystalline structure of the material. There‐ fore, the relationship between the crystalline structure and properties of the material is presented, and is expected to be of use in the design of novel dielectrics. As many superi‐ or materials for microwave dielectrics have been developed and are expected to be used in new applications such as wireless sensors and wireless power transfer by resonant coupling, wave absorption by interference and transparent ceramics with no birefrin‐ gence, these new applications are also discussed.

**Keywords:** Microwave dielectrics, Complex Perovskite, Ordering, Tungsten- bronze com‐ pounds, Homologous series

## **1. Introduction**

Perovskite and related compounds are the main materials used in microwave dielectrics, as shown in Fig. 1. The data are listed in a database proposed by M. T. Sebastian and published in the book "*Dielectric materials for wireless communication*" [1]. The book cites about 2,300 compounds with about 750 references making it an excellent publication for material scientists and researchers, particularly with respect to microwave technology. The data for dielectric

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

materials is now being updated with about 3,000 compounds and 1,600 references which has now been published [2]. A sizeable amount (about 46%) includes rare-earth (*R*) ions with colors as shown in Fig. 2(a), and these were reviewed in a chapter of the book [3] and various papers [4, 5]. The largest amount of compounds (45%) are of the perovskite-type, known as the 'King' of electroceramics, and the second largest group of compounds, at 21%, are the pseudotungsten-bronze solid solutions, also related to perovskite compounds, as shown in Fig. 1.

**Figure 1.** Ratio of microwave dielectric compounds with different crystal structures.

**Figure 2.** (a) Dielectric resonators. (b) LTCC for LC filter. (NTK/NGK)

Microwave dielectrics have been used as a key constituent of wireless communications [6–9]. Microwave dielectrics are used in resonators, filters and temperature stable capacitors with a near zero temperature coefficient of resonate frequency (*TCf*) / temperature coefficient of

**Figure 3.** Dielectric resonator with resonate coupling.

materials is now being updated with about 3,000 compounds and 1,600 references which has now been published [2]. A sizeable amount (about 46%) includes rare-earth (*R*) ions with colors as shown in Fig. 2(a), and these were reviewed in a chapter of the book [3] and various papers [4, 5]. The largest amount of compounds (45%) are of the perovskite-type, known as the 'King' of electroceramics, and the second largest group of compounds, at 21%, are the pseudotungsten-bronze solid solutions, also related to perovskite compounds, as shown in Fig. 1.

282 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 1.** Ratio of microwave dielectric compounds with different crystal structures.

**Dielectrics** (a)

Microwave dielectrics have been used as a key constituent of wireless communications [6–9]. Microwave dielectrics are used in resonators, filters and temperature stable capacitors with a near zero temperature coefficient of resonate frequency (*TCf*) / temperature coefficient of

**Figure 2.** (a) Dielectric resonators. (b) LTCC for LC filter. (NTK/NGK)

**LTCC LC filter** (b)

dielectric constant (*TCε*r) and so on (Fig. 3). Originally, microwave dielectrics were developed from the temperature stable capacitor which shows a near zero *TCε*<sup>r</sup> as explained in section 2. Resonators and filters are used in mobile communication technology. In the early days of information technology (IT), microwave dielectrics replaced cavity resonators and worked extremely well in reducing the size of the equipment used — from the car telephone to the shoulder phone by 1987 for example. Moreover, microwave dielectrics have been developed for a wide range of telecommunication applications, such as mobile phones, wireless LAN, and intelligent transport system (ITS). Fig. 4 shows three directions of development of microwave dielectrics, which presented *Q⋅f* as a function of the dielectric constant *ε*r [10]. The curve in the figure shows the outline of the upper limit of *Q⋅f* obtained for a given *ε*r. The first direction with a high *ε*r is mainly used in the miniaturization of mobile phone components. The second one, with a high *Q* and a high *ε*r is in demand for use in mobile phone base stations. The third direction, with a high *Q* and a low *ε*r is for devices working in the millimeter-wave range — the new frontiers of microwave dielectrics because the utilizable frequency region is expanding towards the millimeter-wave due to a shortage of conventional radio frequency (RF) regions. The three important microwave dielectric properties of *ε*r, *Q⋅f* and *TCf* are explained in section 2.

These compounds are friendly with electromagnetic waves. When irradiated with an electro‐ magnetic wave, the materials should be resonating owing to dielectric polarization changing under an alternating electromagnetic field as shown in Fig. 5(a). The direction of the dielectric polarization should be easily changeable to the opposite direction depending on the electric field. If the material has spontaneous polarization as in ferroelectrics, then inversion losses become large. As a result most microwave dielectrics are paraelectrics with a center of symmetry *i*. The structure of perovskite is flexible, and as a result perovskite shows many kinds of structure, such as cubic, tetragonal, orthorhombic, trigonal, and monoclinic, depending on the particular *A* and *B* cations in *AB*O3. The author recommends referring to some reports written by the author himself [11-16].

Microwave dielectrics have been studied for more than a half of a century now. Many materials with suitable properties have been identified and should be used in new applications to develop new technologies to aid the survival of humans on the Earth. The next generation of functional advances in microwave dielectrics are presented in a chapter of the "Handbook of Multifunctional Ceramics" [17].

In this chapter, perovskite and related materials used in microwave dielectrics are presented and the relationships between the crystal structure and the properties of the materials are discussed. Moreover, new applications for microwave dielectrics developed up to date are also are presented.

**Figure 4.** Three directions of R&D for microwave dielectrics. The *Q⋅f* of microwave dielectrics is shown as a function of *ε*r.

## **2. Three important microwave dielectric properties**

There are three important properties of microwave dielectrics: the quality factor *Q*, the dielectric constants *ε*r, and the temperature coefficient of resonant frequency *TCf* [18].

## **2.1. Quality factor**

Firstly, dielectric materials placed in an electromagnetic field should resonate easily with the electromagnetic waves. In other words, they should have a high quality factor. The quality factor *Q* is the inverse of the dielectric loss (tan*δ*) and is presented as follows:

$$Q = 1/
\tan\delta$$

However, upon measurement, it is usual to obtain a so-called unloaded quality factor *Q*u. This is the sum of the reciprocals of the other factors and depends on the dielectric loss of the materials *Q*d, conduction loss *Q*c, and radiation loss *Q*r.

$$1/\mathbb{Q}\mathbf{u} \; = 1/\mathbb{Q}\mathbf{d} \; + \, 1/\mathbb{Q}\mathbf{c} \; + \, 1/\mathbb{Q}\mathbf{r}$$

The losses are generated by dielectric polarization in the presence of an electromagnetic wave. Ferroelectric materials with spontaneous polarization have large dielectric losses because of the large movement of cations. So, paraelectric materials with a center of symmetry *i* are suitable for microwave and millimeter-wave dielectrics. Dielectric loss increases with an increase in frequency as shown in Fig. 5 (b). In the case of ultra-high frequencies, the number of polarity changes increases with frequency. Therefore, dielectric materials with a high *Q* value are desirable.

## **2.2. Dielectric constant** *ε***<sup>r</sup>**

In this chapter, perovskite and related materials used in microwave dielectrics are presented and the relationships between the crystal structure and the properties of the materials are discussed. Moreover, new applications for microwave dielectrics developed up to date are also

> **High***<sup>Q</sup>* **High**

**r for millimeter-wave applications**

**0 20 40 60 80 100 120 140 160**

**Dielectric constant**

**Figure 4.** Three directions of R&D for microwave dielectrics. The *Q⋅f* of microwave dielectrics is shown as a function of

There are three important properties of microwave dielectrics: the quality factor *Q*, the

Firstly, dielectric materials placed in an electromagnetic field should resonate easily with the electromagnetic waves. In other words, they should have a high quality factor. The quality

*Q* = 1 tan

However, upon measurement, it is usual to obtain a so-called unloaded quality factor *Q*u. This is the sum of the reciprocals of the other factors and depends on the dielectric loss of the

d

dielectric constants *ε*r, and the temperature coefficient of resonant frequency *TCf* [18].

factor *Q* is the inverse of the dielectric loss (tan*δ*) and is presented as follows:

:**Perovskite**

:**Others**

**2. Three important microwave dielectric properties**

materials *Q*d, conduction loss *Q*c, and radiation loss *Q*r.

**High***Q* **Low** 

284 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**100**

**1000**

**10,000**

**100,000**

**1,000,000**

:**Pseudo-tungsten-bronze**

**r for base station**

**High<sup>r</sup> for miniaturization of mobile phone**

> **r**

are presented.

*ε*r.

**2.1. Quality factor**

The dielectric constant *ε*<sup>r</sup> causes a shortening of wave length *λ* in dielectrics as shown in Fig. 5(c) according to the following equation:

$$
\mathcal{J} = \mathcal{J}\_0 \ne \sqrt{\mathcal{E}\_r}
$$

Here, *λ*<sup>0</sup> is the wave length in a vacuum. In the microwave region, the *ε*r value is expected to be large for the miniaturization of mobile communication equipment. In the millimeter-wave region, the *ε*r value is expected to be small. As the wave length is in the millimeter order, miniaturization is not needed.

**Figure 5.** (a) When irradiated by electromagnetic waves, the materials should resonate due to changing dielectric po‐ larization under alternating electromagnetic fields. (b) Dielectric losses increase with an increase in frequency. (c) *ε*<sup>r</sup> causes a shortening of wavelength *λ* in dielectrics.

There are other more important phenomena such as the time delay *T*PD according to the following equation:

$$T\_{PD} = \sqrt{\varepsilon\_r} \text{ / c}$$

Here, *ε*r is the dielectric constant and c is the velocity of light. The time delay is desirable in order to improve the speed of the signal.

The origin of *ε*r was considered by difference of crystal structure as shown in Fig. 6 [19]. Silicates with a low *ε*<sup>r</sup> are formed by the tetrahedral framework of SiO4, with 45% ionic bonds and 55% covalent bonds. Covalent bonds reduce *ε*r, because the rattling effect of the cations in a polyhedron should be reduced as a result of the high bond strength. On the other hand, titanates (Fig. 6(c)) with a large *ε*<sup>r</sup> such as SrTiO3, are formed by a TiO6 octahedral framework, which is almost of ionic bond and has space for ionic displacement. In the case of aluminates, although Al ions also occupy an octahedral framework, the Al ions located in the paired octahedral on the threefold axis repulse each other as shown in Fig. 6(b). The Al ions are immovable in the octahedron and produce a medium *ε*r. The order of *ε*r is as follows:

> r silicate r aluminate r titanate < <

Silicates with a low *ε*r are good candidates for millimeter-wave dielectrics [19].

Fig. 6. Dielectric constants due to crystal structure: (a) SiO4 tetrahedron, (b) Al2O3 and (c) TiO6 octahedron. **Figure 6.** Dielectric constants due to crystal structure: (a) SiO4 tetrahedron, (b) Al2O3 and (c) TiO6 octahedron.

## 2.3 Temperature coefficient of resonant frequency TCf **2.3. Temperature coefficient of resonant frequency** *TCf*

The TCf is required to be near 0 ppm/<sup>o</sup>C for global usage in different environmental temperatures. The TCf has a relationship with the temperature coefficient of dielectric constant TCεr as follows: The *TCf* is required to be near 0 ppm/°C for global usage in different environmental temper‐ atures. The *TCf* has a relationship with the temperature coefficient of dielectric constant *TCε*<sup>r</sup> as follows:

$$\text{TC}f = \ -\ \left(a + \text{TC}\mathfrak{x}\_r/2\right).$$

Most millimeter-wave compounds with a low εr have a large negative TCf such as alumina: - 65, and forsterite: -70 ppm/˚C. The TCf of these millimeter-wave dielectrics was improved by Where *α* is the thermal expansion coefficient.

small degradation of Q∙f.

3. Specialized study

3.1.1) Simple perovskite

3.1 Perovskite-type compounds

two different methods. The first requires the addition of materials with the opposite charge (i.e. a positive TCf). The addition of rutile with a TCf = +450 ppm/˚C can adjust the TCf of the Most millimeter-wave compounds with a low *ε*r have a large negative *TCf* such as alumina: -65, and forsterite: -70 ppm/˚C. The *TCf* of these millimeter-wave dielectrics was improved by

compound in question [20–23]. The second method is to adjust the TCf to near 0 ppm/˚C by the formation of the solid solution phases [24]. This is the preferred method because of the

Although perovskite compounds commonly used in ferroelectrics shouldn't be used for microwave dielectrics as described in the previous section, perovskite compounds can be flexibly applied to microwave dielectrics. This flexibility is due to the depth of the crystal structure. Table 1 shows the polymorphism of BaTiO3 — a representative perovskite-type two different methods. The first requires the addition of materials with the opposite charge (*i.e*. a positive *TCf*). The addition of rutile with a *TCf* = +450 ppm/˚C can adjust the *TCf* of the compound in question [20–23]. The second method is to adjust the *TCf* to near 0 ppm/˚C by the formation of the solid solution phases [24]. This is the preferred method because of the small degradation of *Q⋅f*.

## **3. Specialized study**

The origin of *ε*r was considered by difference of crystal structure as shown in Fig. 6 [19]. Silicates with a low *ε*<sup>r</sup> are formed by the tetrahedral framework of SiO4, with 45% ionic bonds and 55% covalent bonds. Covalent bonds reduce *ε*r, because the rattling effect of the cations in a polyhedron should be reduced as a result of the high bond strength. On the other hand, titanates (Fig. 6(c)) with a large *ε*<sup>r</sup> such as SrTiO3, are formed by a TiO6 octahedral framework, which is almost of ionic bond and has space for ionic displacement. In the case of aluminates, although Al ions also occupy an octahedral framework, the Al ions located in the paired octahedral on the threefold axis repulse each other as shown in Fig. 6(b). The Al ions are

immovable in the octahedron and produce a medium *ε*r. The order of *ε*r is as follows:

Silicates with a low *ε*r are good candidates for millimeter-wave dielectrics [19].

(a) (b) (c)

repulsion

・Al ion located in octahedron, just on 3-fold axis and

**Figure 6.** Dielectric constants due to crystal structure: (a) SiO4 tetrahedron, (b) Al2O3 and (c) TiO6 octahedron.

was fixed by the repulsion of each Al ion.

Fig. 6. Dielectric constants due to crystal structure: (a) SiO4 tetrahedron, (b) Al2O3 and (c)

The TCf is required to be near 0 ppm/<sup>o</sup>C for global usage in different environmental temperatures. The TCf has a relationship with the temperature coefficient of dielectric

The *TCf* is required to be near 0 ppm/°C for global usage in different environmental temper‐ atures. The *TCf* has a relationship with the temperature coefficient of dielectric constant *TCε*<sup>r</sup>

TCf = - (α + TCεr / 2).

 *<sup>r</sup>* )

*TCf TC* =- + 2. (a

Most millimeter-wave compounds with a low εr have a large negative TCf such as alumina: - 65, and forsterite: -70 ppm/˚C. The TCf of these millimeter-wave dielectrics was improved by two different methods. The first requires the addition of materials with the opposite charge (i.e. a positive TCf). The addition of rutile with a TCf = +450 ppm/˚C can adjust the TCf of the compound in question [20–23]. The second method is to adjust the TCf to near 0 ppm/˚C by the formation of the solid solution phases [24]. This is the preferred method because of the

Most millimeter-wave compounds with a low *ε*r have a large negative *TCf* such as alumina: -65, and forsterite: -70 ppm/˚C. The *TCf* of these millimeter-wave dielectrics was improved by

Although perovskite compounds commonly used in ferroelectrics shouldn't be used for microwave dielectrics as described in the previous section, perovskite compounds can be flexibly applied to microwave dielectrics. This flexibility is due to the depth of the crystal structure. Table 1 shows the polymorphism of BaTiO3 — a representative perovskite-type

2.6Å

si4+

・SiO4 tetrahdron is formed by ionic bond of 45% and covalent bond of 55%.

TiO6 octahedron.

as follows:

constant TCεr as follows:

small degradation of Q∙f.

3. Specialized study

3.1.1) Simple perovskite

3.1 Perovskite-type compounds

1.6Å

286 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

SiO4 tetrahedron

<sup>ε</sup>rsilicate <sup>&</sup>lt; <sup>ε</sup>raluminate <sup>&</sup>lt; <sup>ε</sup>rtitanate

2.3 Temperature coefficient of resonant frequency TCf

**2.3. Temperature coefficient of resonant frequency** *TCf*

Where α is the thermal expansion coefficient.

Where *α* is the thermal expansion coefficient.

r silicate r aluminate r titanate

3-fold axis

octahedron

Rattling factor --> large

Al3+

< <

 

> ・TiO6 octahedron is formed almost by ionic bond.

## **3.1. Perovskite-type compounds**

## *3.1.1. Simple perovskite*

Although perovskite compounds commonly used in ferroelectrics shouldn't be used for microwave dielectrics as described in the previous section, perovskite compounds can be flexibly applied to microwave dielectrics. This flexibility is due to the depth of the crystal structure. Table 1 shows the polymorphism of BaTiO3 — a representative perovskite-type structure. At room temperature, it is stable due to the tetragonal system, but is ferroelectric without a center of symmetry *i*. It transforms to a cubic structure with paraelectricity above a Curie point of 120 °C. In the near feature, if devices capable of operating under temperatures of more than 120 °C appear, then it may be applied to microwave devices. At low temperatures, all structures of BaTiO3 are ferroelectrics without *i*.


**Table 1.** Crystal data for BaTiO<sup>3</sup>

Currently, microwave dielectrics of perovskite-type and related compounds are detailed in the database created by Sebastian [1, 2]. Some simple perovskite-type compounds are *A*TiO3, *A*ZrO3 (*A*2+ = Ba, Sr and Ca) and *RB*O3 (*R*3+ = rare earth, *B*3+ = Al, Ga). MgTiO3 and ZnTiO3 with their small ionic cations of Mg and Zn are not perovskite-type structures, but are of the ilmenite-type similar to the structure of Al2O3 with oxygen closest packing structure. Table 2(a) shows three microwave properties of simple perovskite-type compounds. These have different crystal structures such as cubic, orthorhombic and hexagonal, but qualify as micro‐ wave dielectrics because they have a center of symmetry *i*. SrTiO3 has the crystal structure closest to BaTiO3. It is expected to be a microwave dielectric due to the cubic structure of the paraelectric at room temperature. However, one disadvantageous point is that the temperature coefficient of resonant frequency *TCf* is too large at 1,200 ppm/°C. CaTiO3 with the mineral name "perovskite" is orthorhombic in the space group *Pnma* (No.62) with *i* [25]. The charac‐ teristic structure of CaTiO3 is a tilting octahedral. This compound also has a large *TCf* of over 859 ppm/°C, so it could not be used by itself as a microwave dielectrics. Nonetheless, this compound has been used as a stabilizer of *TCf* against microwave compounds with a negative *TCf*, as most useful microwave dielectrics have a positive *TCf*. MgTiO3 with *TCf* = -45 ppm/°C was improved to a near zero *TCf* by adding CaTiO3. This compound with *ε*<sup>r</sup> = 21 and *Q⋅f* = 8,000 GHz was the first one used in practice in microwave dielectrics. Recently, in an ilmenite system, a Co doped MgTiO3 dielectric with a high *Q⋅f* (864,000 GHz) was found, and its *TCf* was improved to near zero by the addition of CaTiO3 [28].



*A*ZrO3 (*A*2+ = Ba, Sr and Ca) and *RB*O3 (*R*3+ = rare earth, *B*3+ = Al, Ga). MgTiO3 and ZnTiO3 with their small ionic cations of Mg and Zn are not perovskite-type structures, but are of the ilmenite-type similar to the structure of Al2O3 with oxygen closest packing structure. Table 2(a) shows three microwave properties of simple perovskite-type compounds. These have different crystal structures such as cubic, orthorhombic and hexagonal, but qualify as micro‐ wave dielectrics because they have a center of symmetry *i*. SrTiO3 has the crystal structure closest to BaTiO3. It is expected to be a microwave dielectric due to the cubic structure of the paraelectric at room temperature. However, one disadvantageous point is that the temperature coefficient of resonant frequency *TCf* is too large at 1,200 ppm/°C. CaTiO3 with the mineral name "perovskite" is orthorhombic in the space group *Pnma* (No.62) with *i* [25]. The charac‐ teristic structure of CaTiO3 is a tilting octahedral. This compound also has a large *TCf* of over 859 ppm/°C, so it could not be used by itself as a microwave dielectrics. Nonetheless, this compound has been used as a stabilizer of *TCf* against microwave compounds with a negative *TCf*, as most useful microwave dielectrics have a positive *TCf*. MgTiO3 with *TCf* = -45 ppm/°C was improved to a near zero *TCf* by adding CaTiO3. This compound with *ε*<sup>r</sup> = 21 and *Q⋅f* = 8,000 GHz was the first one used in practice in microwave dielectrics. Recently, in an ilmenite system, a Co doped MgTiO3 dielectric with a high *Q⋅f* (864,000 GHz) was found, and its *TCf*

(a) Simple perovskite

(b) Modified simple perovskite

0.2SrTiO3・0.8LaAlO3 26.7 139 -50 0.64CaTiO3・0.34LaGaO3 46.5 48 -2.9 0.7CaTiO3・0.3NdAlO3 43 47 0 0.35CaTiO3・0.65LaAlO3 37 47 -2

SrTiO3 304 3.3 1700 CaTiO3 162 12.96 859 BaZrO3 35 8.8 - SrZrO3 30 13.6 -60 CaZrO3 30 26.4 -27 NdGaO3 22 85 - LaAlO3 23.4 68 -44 SmAlO3 20.4 65 -74 NdAlO3 22.3 58 -33 YAlO3 15.7 58 -59 PrAlO3 23.2 51 -25

\*GHz) *TCf* (ppm/°C)

Compound *ε*<sup>r</sup> *Q∙f* (103

was improved to near zero by the addition of CaTiO3 [28].

288 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications



0.7Ba(Mg1/3Ta2/3)O3·0.3Ba(Co1/3Nb2/3)O3 27 165 -1.3 Ba(ZnTa)O3・Ba(ZnNb)O3 30 164 0 Ba(Mg1/3Ta2/3)O3・Ba(Zn1/3Ta2/3)O3 27 150 0 0.5Ba(MgTa)O3・0.5Ba(ZnTa)O3 27 135 1.95 0.95Ba(Zn1/3Nb2/3)O3·0.05Ba(Ga1/2Ta1/2)O3 38 102.96 19 Ba(Ni1/3Ta2/3)O3・Ba(ZrZnTa)O3 30 100 0 (f) Modified complex perovskite Ba(Mg1/3Ta2/3)O3: BaSnO3, BaWO4, 24 430 5 Ba[(Zn0.6Co0.4)1/3Nb2/3]O3 35.6 351.95 - Ba[(Mg1-xZn*x*)1/3Ta2/3]O3 24-26 200-300 -0.5-1.7 Ba(Zn1/3Ta2/3)O3: Ga, Zr 30 165.4 0 Ba(SnMgTa)O3 24.2 120 - Ba[(Mg0.4Zn0.6)Ta2/3]O3 27.7 109.9 6.3 Ba(Zr0.05Zn0.32Ta0.63)O3 30.4 105 8 Ba(M0.33Ta0.63Ti0.017W0.017)O3 24.5 100.7 12.6 Ba(Mg0.30Ta0.60Ti0.10)O3 26.3 100 14.4

290 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

(Ba1-zSr*z*)[Zn1/3(Ta*p*Nb1-p)2/3](Sr1-xCa*x*) (Ga1/2Ta1/2)O3 32-34 180-80 0-10

Ba4(La1-y-zSm*y*Bi*z*)9.33Ti18O54 (*x*=2/3, *y*=0.7, *z*=0.04) 88.4 6.69 1 Ba4(Sm1-yNd*y*)9.33(Ti9.95Sn0.05)O54 (*x*=2/3, *y*=0.8, *z*=0.05 ) 80 10.6 11

Ba4Sr2Nd8Ti18O54 (*x*= 0) 98.0 6 20 (Ba1-αSr*α*)6Sm8Ti18O54 (*x*= 0, *α* = 0.32) 91.3 8.02 61 (Ba1-αSr*α*)5.7Sm8.2Ti18O54(*x*=0.1,*α*= 0.298) 85.3 8.71 24 (h) (111) type layered perovskite (Ba*n*La4Ti3+nO12+3n-type homologous series)

(g) Pseudo-tungsten-bronze solid solutions (*x:* Ba6-3x *R* 8+2xTi18O54, *R*: Rare earth)

Ba(SnMgTa)O3 24.2 120 -

Ba4Sm9.33Ti18O54 (*x*=2/3) 80 10.7 -15 Ba4Nd9.33Ti18O54 (*x*=2/3) 82.5 10.1 71.1 Ba4Sm8.33EuTi18O54 (*x*=2/3) 78.7 9.56 -10.5 Ba4Nd5.33Eu4Ti18O54 (*x*=2/3) 78 10.46 10.4 Ba4Nd8.33DyTi18O54 (*x*=2/3) 78.6 10.04 33.8 Ba4Sm8.08Li0.25Ti18O54 (*x*=2/3) 82.1 5.62 -2 Ba4.2(Sm0.9La0.1)9.2Ti18O54 (*x*=0.6) 84 9.05 1.6 Ba4.5(Nd0.8Bi0.2)9Ti18O54 (*x*=0.5) 106 4.2 8

**Table 2.** Microwave dielectric properties of perovskite and perovskite related compounds. No. in (i) list are cited from Sebastian's data base (Book) [1].

*R*3+*B*3+O3 compounds containing rare-earth ions (*R*) in the *A*-site of the perovskite structure are one of simplest perovskite-type compounds [29]. As the *R* ion is trivalent, the *B* ion in the *B*site should also be trivalent. Almost all rare earth ions (that is Y to Er) can occupy the *A*-site. In some compounds, the *A*-site can be occupied by two or more *R* ions. Compounds including Sc, Yb and Lu ions have not been reported because of their small radius size. The *B*-site is occupied by a single ion such as Al, Ga and B, and by a pair of ions that are either divalent or tetravalent such as Mg2+Ti4+ [30, 31]. These *RB*O3-type compounds as shown in Table 2(a) are preferred for microwave dielectrics because of their small dielectric losses. The crystal structure changes from trigonal to orthorhombic depending on the tolerance factor, as shown

**Figure 7.** Variation of *ε*r as a function of the tolerance factor *t* in *R*AlO3.

in Fig. 7 [29]. The compounds with a larger size of ion (La to Nd) are trigonal (*R3c* No.167), and those with a smaller size (Sm to Er) are orthorhombic (*Pnma*, No.62).

The single crystals of LaAlO3 which can be grown easily from melts [32] are used as substrates for superconductor materials such as YBa2Cu3O7-*x*, because of their low dielectric losses and their small mismatch for epitaxial growth. It is noticed that strip-line resonators formed by superconductors grown epitaxially on the LaAlO3 single crystal substrate are used in the bandpass filter of base stations in microwave mobile communications. The low dielectric losses come from the low conductivity, based on zero electrical resistivity. Although the *TCf* of LaAlO3 is above -60 ppm/°C [29], this is not an issue whenever it is used as the substrate for a superconductor at a fixed low temperature.

**Figure 8.** *Q⋅f* value (a), *ε*r (b) and *TCf* (c) of (1-*x*)LaAlO3-*x*SrTiO3 as a function of composition *x*.

**Figure 9.** Crystal structure of LaAlO3 (a) with Space Group *R3c* (167), and of SrTiO3 doped LaAlO3 (b) with S.G. *R3* (148).

The *TCf* value is problematic whenever it is used as a resonator or filter at room tempera‐ ture. In such cases, it is proposed that the *TCf* value is suppressed to near zero ppm/°C by the doping of SrTiO3 or CaTiO3 as shown in Fig. 8(c) [33, 34]. As these solid solutions show a high *Q* and a high *ε*r as shown in Fig. 8(a) and (b), the reason for the improved high *Q* value is seen through the study and analysis of a single crystal structure. Inagaki *et al.* [35] showed that the crystal system changed from *R3c* (No. 167) to *R3* (No. 148), thereby creating a new position for the Sr ion, as shown in Fig. 9(b) [35–36], and the observed disappear‐ ance of the polysynthetic twin. These facts suggest the improvement of *Q⋅f*. Moreover, a NdTiO3-CaTiO3 solid solution system is used for microwave dielectrics with a higher *ε*<sup>r</sup> instead of LaAlO3-SrTiO3 solid solutions. The properties are as follows: 0.2SrTiO3-0.8LaA‐ lO3 [34]: *ε*r = 26.7, *Q⋅f* = 139,000 GHz and *TCf* = -50 ppm/˚C ; 0.67CaTiO3-0.33NdAlO3 [37]: *ε*<sup>r</sup> = 41.98, *Q⋅f* = 42,900 GHz and *TCf* = 45 ppm/˚C.

## *3.1.2. Complex perovskite*

in Fig. 7 [29]. The compounds with a larger size of ion (La to Nd) are trigonal (*R3c* No.167),

**Tolerance factor**

**Orthorhombic Trigonal**

The single crystals of LaAlO3 which can be grown easily from melts [32] are used as substrates for superconductor materials such as YBa2Cu3O7-*x*, because of their low dielectric losses and their small mismatch for epitaxial growth. It is noticed that strip-line resonators formed by superconductors grown epitaxially on the LaAlO3 single crystal substrate are used in the bandpass filter of base stations in microwave mobile communications. The low dielectric losses come from the low conductivity, based on zero electrical resistivity. Although the *TCf* of LaAlO3 is above -60 ppm/°C [29], this is not an issue whenever it is used as the substrate for a

and those with a smaller size (Sm to Er) are orthorhombic (*Pnma*, No.62).

**Figure 7.** Variation of *ε*r as a function of the tolerance factor *t* in *R*AlO3.

292 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 8.** *Q⋅f* value (a), *ε*r (b) and *TCf* (c) of (1-*x*)LaAlO3-*x*SrTiO3 as a function of composition *x*.

superconductor at a fixed low temperature.

## *3.1.2.1. Types of complex perovskite and their properties*

Complex perovskite differs from simple perovskite by a single ion in both the *A-* and *B*-site. It is constituted maintaining the charge valance through the differently charged ions in each site, and is distinguished from the substance containing the substituted different plural ions. Complex perovskite compounds reported so far are as follows [38]:

## **1:1 type in B-site**

$$A^{2\*}\left(\mathcal{B}^{3\*}{}\_{1/2}\mathcal{B}^{5\*}{}\_{1/2}\right)\mathcal{O}\_{3'}A^{2\*}\left(\mathcal{B}^{2\*}{}\_{1/2}\mathcal{B}^{6\*}{}\_{1/2}\right)\mathcal{O}\_{3'}A^{2\*}\left(\mathcal{B}^{1\*}{}\_{1/2}\mathcal{B}^{7\*}{}\_{1/2}\right)\mathcal{O}\_{3'}A^{3\*}\left(\mathcal{B}^{2\*}{}\_{1/2}\mathcal{B}^{4\*}{}\_{1/2}\right)\mathcal{O}\_{3'}$$

**1:2 type in B-site**

$$A^{2^{\*}}\left(\mathcal{B}^{2^{\*}}{}\_{1/3}\mathcal{B}^{5^{\*}}{}\_{2/3}\right)\mathcal{O}\_{3^{\*}}A^{2^{\*}}\left(\mathcal{B}^{3^{\*}}{}\_{2/3}\mathcal{B}^{6^{\*}}{}\_{1/3}\right)\mathcal{O}\_{3^{\*}}$$

**1:3 type in B-site**

$$A^{2^\*} \left(B^{1^\*} {}\_{1/4}B^{5^\*} {}\_{3/4} \right) \mathcal{O}\_{3^\*}$$

**1:1 type in A-site**

$$\left(\boldsymbol{A}^{1\*}\boldsymbol{\iota}\_{1/2}\boldsymbol{A}^{3\*}\boldsymbol{\iota}\_{1/2}\right) \text{TiO}\_{3'}\left(\boldsymbol{A}^{2\*}\boldsymbol{\iota}\_{2/5}\boldsymbol{A}^{3\*}\boldsymbol{\iota}\_{2/5}\right) \text{TiO}\_{3'}\boldsymbol{\iota}$$

In the 1:2 type complex perovskite, many compounds exist with suitable properties. Tables 2(c), (d), (e) and (f) show characteristic microwave dielectrics with good properties selected from Sebastian's database, as referred to above [1, 2]. In the data, the microwave dielectric with the highest *Q⋅f* value of 430,000 GHz is Ba(Mg1/3Ta2/3)O3 (BMT) — the 'king' of microwave dielectrics [39]. The *TCf* also has a desirable value, being near to zero at 3.3 ppm/°C. The *Q⋅f* value of Ca(Mg1/3Ta2/3)O3 when Ca was substituted for Ba decreased to 78,000 GHz [40], and when Sr was substituted for Ba, it decreased even more, to 5,600 GHz [41]. Kageyama [42] showed the *Q⋅f* values of 1:2 type complex perovskites as a function of the tolerance factor in the Ba and Sr-system as shown in Fig. 10. It brings a high *Q⋅f* so that the tolerance factors of the Ba-system with large size ions in the *A*-site are large, and the electronic structure of the *B*site ions is a closed shell. In the case of 1:1 type compounds, La(Mg1/2Ti1/2)O3 (LMT) [43] shows the highest *Q⋅f* of 114,000 GHz. The *A*-site of this compound is occupied by the trivalent rare earth La ion, and the valence of the *B*-site is trivalent and composed of the 1:1 ratio of Mg2+ and Ti4+. However, the *TCf* of -81 ppm/°C is not a desirable value. Kageyama [44] systematically studied 1:1-type compounds and clarified that Ca(Ga1/2Ta1/2)O3 (CGT) and Sr(Ga1/2Ta1/2)O3 (SGT) show high a *Q⋅f*. In this system, though the correlation with the tolerance factor is small, Ga with a closed shell electronic structure contributed to the improvement in *Q⋅f* values. Wakino *et al.* [45] reported Ba(Mg1/2W1/2)O3 (BMW) with a high *Q⋅f*, composed of divalent Mg and six valenced W. These compounds also have the disadvantage of a large *TCf*. One of compounds with a near zero *TCf* is Ba(Tb1/2Nb1/2)O3 (BTN) [46] with high *ε*r = 39, *Q⋅f* = 52,400 GHz and *TCf* = -2 ppm/°C.

### *3.1.2.2. Is ordering a necessary condition for a high Q value?*

The origin of a high *Q* value, especially the relationship between a high *Q* value and ordering based on an order-disorder transition, has been under discussion for a long time [47, 51]. The feature of complex perovskite *A*(*B*1/3*B*'2/3)O3 exhibits the phenomenon of the ordering of *B* cations. The ordered phase appeared at low temperature is low symmetry trigonal (rhombo‐

**Figure 10.** *Q⋅f* of *A*(*B*2+1/3*B*5+2/3)O3 as a function of the tolerance factor *t*.

**1:2 type in B-site**

**1:3 type in B-site**

**1:1 type in A-site**

GHz and *TCf* = -2 ppm/°C.

*3.1.2.2. Is ordering a necessary condition for a high Q value?*

( ) ( ) 22 5 23 6 1/3 2/3 3 2/3 1/3 3 *AB B AB B* O, O, ++ + ++ +

294 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

( ) 21 5 *AB B* 1/4 3/4 3 O , ++ +

( ) ( ) 1 3 2 3 1/2 1/2 3 2/5 2/5 3 *AA AA* TiO , TiO , + + + +

In the 1:2 type complex perovskite, many compounds exist with suitable properties. Tables 2(c), (d), (e) and (f) show characteristic microwave dielectrics with good properties selected from Sebastian's database, as referred to above [1, 2]. In the data, the microwave dielectric with the highest *Q⋅f* value of 430,000 GHz is Ba(Mg1/3Ta2/3)O3 (BMT) — the 'king' of microwave dielectrics [39]. The *TCf* also has a desirable value, being near to zero at 3.3 ppm/°C. The *Q⋅f* value of Ca(Mg1/3Ta2/3)O3 when Ca was substituted for Ba decreased to 78,000 GHz [40], and when Sr was substituted for Ba, it decreased even more, to 5,600 GHz [41]. Kageyama [42] showed the *Q⋅f* values of 1:2 type complex perovskites as a function of the tolerance factor in the Ba and Sr-system as shown in Fig. 10. It brings a high *Q⋅f* so that the tolerance factors of the Ba-system with large size ions in the *A*-site are large, and the electronic structure of the *B*site ions is a closed shell. In the case of 1:1 type compounds, La(Mg1/2Ti1/2)O3 (LMT) [43] shows the highest *Q⋅f* of 114,000 GHz. The *A*-site of this compound is occupied by the trivalent rare earth La ion, and the valence of the *B*-site is trivalent and composed of the 1:1 ratio of Mg2+ and Ti4+. However, the *TCf* of -81 ppm/°C is not a desirable value. Kageyama [44] systematically studied 1:1-type compounds and clarified that Ca(Ga1/2Ta1/2)O3 (CGT) and Sr(Ga1/2Ta1/2)O3 (SGT) show high a *Q⋅f*. In this system, though the correlation with the tolerance factor is small, Ga with a closed shell electronic structure contributed to the improvement in *Q⋅f* values. Wakino *et al.* [45] reported Ba(Mg1/2W1/2)O3 (BMW) with a high *Q⋅f*, composed of divalent Mg and six valenced W. These compounds also have the disadvantage of a large *TCf*. One of compounds with a near zero *TCf* is Ba(Tb1/2Nb1/2)O3 (BTN) [46] with high *ε*r = 39, *Q⋅f* = 52,400

The origin of a high *Q* value, especially the relationship between a high *Q* value and ordering based on an order-disorder transition, has been under discussion for a long time [47, 51]. The feature of complex perovskite *A*(*B*1/3*B*'2/3)O3 exhibits the phenomenon of the ordering of *B* cations. The ordered phase appeared at low temperature is low symmetry trigonal (rhombo‐

hedral) structure of space group *P3m*1 (No. 164) and the disordered phase appeared at high temperature is high symmetry cubic structure of *Pm*3*m* (No. 221) as shown in Fig. 11. Kawa‐ shima *et al.* [52] reported that Ba(Zn1/3Ta2/3)O3 (BZT) has a high *Q*. BZT shows ordering of *B* cations, as revealed by the splitting and super structure lines on the X-ray powder diffraction (XRPD) patterns for a long sintering time. When the Zn and Ta ions occupy the same position, the structure is a disordered cubic one. On the other hand, if both ions occupy different independent sites, that is ordering, the structure becomes trigonal. This transition is sluggish and the temperature of transition is not clear in some compounds. The relationship between cubic and trigonal crystal structures is shown in Fig. 11. The *B* cations occupy the octahedra located between the hexagonal closed packing layers composing BaO3. The ordering is apparent by the periodic arrangement of Zn-Ta-Ta along the *c*-axis of the trigonal. Though it is believed that ordering brings a high *Q*, some examples contradicting this have arisen, such as BMT-Ba(Co1/3Ta2/3)O3 [53] and Ba(Mg1/3Ta2/3Sn)O3 [54]. Recently, Koga *et al*. [55–59] present‐ ed the quantification of the ordering ratio using the Rietveld method and the ordering state in the vicinity of BZT. Kugimiya [60] reported that the composition which deviated from BMT has a high *Q* because of the high density composition. More recently Surendran *et al.* [61] showed that Ba and Mg deficient BMT compositions have a high *Q*. In this section, the author presents the primary factors for a high value of *Q* instead of ordering based on Koga's data [55– 59, 61, 62].

compositions have a high Q. In this section, the author presents the primary factors for a

high value of Q instead of ordering based on Koga's data [55–59,61,62].

Fig. 11. Order-disorder transition of perovskite. (a) High temperature and high symmetry phase with cubic, (b) Low temperature and low symmetry phase with trigonal. **Figure 11.** Order-disorder transition of perovskite. (a) High temperature and high symmetry phase with cubic, (b) Low temperature and low symmetry phase with trigonal.

#### **• Ordering ratio and** *Q⋅f* [55] ∙ Ordering ratio and Q∙f [55]

Fig. 12 shows the XRPD patterns (a) and the high angle diffraction peaks (b) of BZT ceramics as a function of sintering time at 1,350 °C. According to sintering time, superlattice lines (asterisked) became clear and the 420 cubic diffraction peak splits gradually into two peaks, namely 226 and 422, in the trigonal system. It is considered that ordered and disordered structures coexist and ordered peaks become intense on sintering of 80 hours or more. These results are consistent with the report by Kawashima *et al* [52]. Fig. 12 shows the XRPD patterns (a) and the high angle diffraction peaks (b) of BZT ceramics as a function of sintering time at 1,350 oC. According to sintering time, superlattice lines (asterisked) became clear and the 420 cubic diffraction peak splits gradually into two peaks, namely 226 and 422, in the trigonal system. It is considered that ordered and disordered structures coexist and ordered peaks become intense on sintering of 80 hours or more. These results are consistent with the report by Kawashima et al [52]. Fig. 13 shows Q∙f as functions of ordering ratio (a) obtained by the Rietveld method [63], density (b) and grain size (c). The ordering ratio saturates at about 80% but the Q∙f varies

Fig. 13 shows *Q⋅f* as functions of ordering ratio (a) obtained by the Rietveld method [63], density (b) and grain size (c). The ordering ratio saturates at about 80 % but the *Q⋅f* varies from 40,000 to 100,000 GHz. However, the *Q⋅f* increases with density and grain size. This indicates that the effect of ordering on the *Q* value is not so important. indicates that the effect of ordering on the Q value is not so important.

from 40,000 to 100,000 GHz. However, the Q∙f increases with density and grain size. This

Fig. 12 (a) XRPD patterns of BZT ceramics with different sintering time at 1,350 oC. Asterisks are superlattice reflections. (b) Magnified XRPD patterns around 2θ = 115<sup>o</sup> in which 420 diffraction peak split to 226 and 422. **Figure 12.** (a) XRPD patterns of BZT ceramics with different sintering time at 1,350 °C. Asterisks are superlattice reflec‐ tions. (b) Magnified XRPD patterns around 2*θ* = 115° in which 420 diffraction peak split to 226 and 422.

(a) (b) (c)

∙ Disordered BZT with a high Q∙f sintered by SPS [57]

size (c).

Fig. 13. The Q∙f of BZT ceramics as functions of ordering ratio (a), density (b), and grain

(a) (b)

2θ (CuKα) 113 114 115 116 117

(b)

120 h

100 h

80 h

20 h

4 h

(a) (b)

120 h

100 h

80 h

20 h

4 h

Fig. 12 (a) XRPD patterns of BZT ceramics with different sintering time at 1,350 oC. Asterisks

 Fig. 13. The Q∙f of BZT ceramics as functions of ordering ratio (a), density (b), and grain size (c). **Figure 13.** The *Q⋅f* of BZT ceramics as functions of ordering ratio (a), density (b), and grain size (c).

## **• Disordered BZT with a high** *Q⋅f* **sintered by SPS** [57]

diffraction peak split to 226 and 422.

\* \*

(a)

\*:superlattice line

2θ (CuKα) 20 25 30 45 50 55 60 65

\* \* \* \*\*

**• Ordering ratio and** *Q⋅f* [55]

High temp. form High symmetry Disordering

∙ Ordering ratio and Q∙f [55]

temperature and low symmetry phase with trigonal.

Fig. 12 shows the XRPD patterns (a) and the high angle diffraction peaks (b) of BZT ceramics as a function of sintering time at 1,350 °C. According to sintering time, superlattice lines (asterisked) became clear and the 420 cubic diffraction peak splits gradually into two peaks, namely 226 and 422, in the trigonal system. It is considered that ordered and disordered structures coexist and ordered peaks become intense on sintering of 80 hours or more. These

Fig. 12 shows the XRPD patterns (a) and the high angle diffraction peaks (b) of BZT ceramics as a function of sintering time at 1,350 oC. According to sintering time, superlattice lines (asterisked) became clear and the 420 cubic diffraction peak splits gradually into two peaks, namely 226 and 422, in the trigonal system. It is considered that ordered and disordered structures coexist and ordered peaks become intense on sintering of 80 hours or more. These

Fig. 11. Order-disorder transition of perovskite. (a) High temperature and high symmetry

**Figure 11.** Order-disorder transition of perovskite. (a) High temperature and high symmetry phase with cubic, (b) Low

composition. More recently Surendran et al. [61] showed that Ba and Mg deficient BMT compositions have a high Q. In this section, the author presents the primary factors for a

> Ba O

> > Mg Ta

high value of Q instead of ordering based on Koga's data [55–59,61,62].

296 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Trigonal <sup>P</sup>3m1 (No.164) - Cubic Pm3<sup>m</sup> (No.221) -

Ta Ba O

Order-disorder transition

(a) (b)

phase with cubic, (b) Low temperature and low symmetry phase with trigonal.

Fig. 13 shows *Q⋅f* as functions of ordering ratio (a) obtained by the Rietveld method [63], density (b) and grain size (c). The ordering ratio saturates at about 80 % but the *Q⋅f* varies from 40,000 to 100,000 GHz. However, the *Q⋅f* increases with density and grain size. This indicates

Fig. 13 shows Q∙f as functions of ordering ratio (a) obtained by the Rietveld method [63], density (b) and grain size (c). The ordering ratio saturates at about 80% but the Q∙f varies from 40,000 to 100,000 GHz. However, the Q∙f increases with density and grain size. This

> (a) (b) Fig. 12 (a) XRPD patterns of BZT ceramics with different sintering time at 1,350 oC. Asterisks are superlattice reflections. (b) Magnified XRPD patterns around 2θ = 115<sup>o</sup> in which 420

**Figure 12.** (a) XRPD patterns of BZT ceramics with different sintering time at 1,350 °C. Asterisks are superlattice reflec‐

120 h

100 h 80 h 20 h 4 h

> 2θ (CuKα) 113 114 115 116 117

(b)

Low temp. form Low symmetry Ordering

Mg Octahedron layer

b a

c

<sup>←</sup>BaO<sup>3</sup> Close packing layer

←TaO<sup>6</sup>

<sup>←</sup>MgO<sup>6</sup> Octahedron layer

120 h 100 h 80 h 20 h 4 h

(a) (b) (c)

∙ Disordered BZT with a high Q∙f sintered by SPS [57]

Fig. 13. The Q∙f of BZT ceramics as functions of ordering ratio (a), density (b), and grain

results are consistent with the report by Kawashima *et al* [52].

results are consistent with the report by Kawashima et al [52].

indicates that the effect of ordering on the Q value is not so important.

that the effect of ordering on the *Q* value is not so important.

\* \*

(a)

\*:superlattice line

diffraction peak split to 226 and 422.

2θ (CuKα) 20 25 30 45 50 55 60 65

tions. (b) Magnified XRPD patterns around 2*θ* = 115° in which 420 diffraction peak split to 226 and 422.

\* \* \* \*\*

size (c).

∙ Disordered BZT with a high Q∙f sintered by SPS [57] As both ordered and disordered BZT — with similar microstructures — can be obtained by various heat treatments using a conventional solid state reaction (SSR) as described in the previous section, the effects of the crystal-structural ordering and ceramic microstructure were discussed independently. In the case of BZT, although the high density and high *Q* ceramics of ordered BZT were synthesized by SSR with a long sintering time of over 80 hours, the dense ceramics of disordered BZT have not been obtained by SSR. Koga *et al.*[57] created high density disordered BZT ceramics with a short sintering time by spark plasma sintering (SPS). In this section, the effects of crystal-structural ordering and ceramic microstructure on the high *Q* are discussed.

Fig. 14 shows the *Q⋅f* as a function of the densities in BZT fabricated using SSR and SPS. The samples obtained by SPS were of the disordered cubic type of perovskite as shown in the XRPD pattern (Fig. 15) with a lone 422 reflection compared with the ordered trigonal type with peak separations of 422 and 226 when sintered using SSR (1400 °C 100 h). The SPS samples with high densities were obtained using an extremely short sintering time of 5 mins between 1150 and 1300 °C under 30 Mpa [57]. The short time sintering when using SPS may result in the disordered BZT with a high density of 7.62 g/cm3 , which is approximately 50% higher than that of low density samples of 5.0-6.0 g/cm3 synthesized by conventional SSR. The full width at half maximum (FWHM) of the 420 peak became narrower with an increase in the temper‐ ature from 1,100 to 1,300 °C (Fig. 15). This indicates that the degree of crystallization of the disordered cubic phase is improved without the need for conversion to the ordered trigonal phase. Regardless of the method of synthesis, *Q⋅f* is strongly dependent on density, and *Q⋅f* values were improved with density as shown in Fig. 14. The highly crystallized dense disordered BZT ceramics synthesized by SPS showed a significantly high *Q⋅f* (= 53,400 GHz) similar to that of the ordered BZT sample with the same density (= ca. 7.5 g/cm3 ) synthesized by SSR. The crystallization with densification of BZT ceramics should play a more important role in the improvement of the *Q* factor in the BZT system than structural ordering and grain size. In the high density region (> 7.5 g/cm3 ), the variation of the *Q⋅f* should be clarified.

microstructure on the high Q are discussed.

the variation of the Q∙f should be clarified.

As both ordered and disordered BZT — with similar microstructures — can be obtained by various heat treatments using a conventional solid state reaction (SSR) as described in the previous section, the effects of the crystal-structural ordering and ceramic microstructure were discussed independently. In the case of BZT, although the high density and high Q ceramics of ordered BZT were synthesized by SSR with a long sintering time of over 80 hours, the dense ceramics of disordered BZT have not been obtained by SSR. Koga et al. [57] created high density disordered BZT ceramics with a short sintering time by spark plasma

Fig. 14 shows the Q∙f as a function of the densities in BZT fabricated using SSR and SPS. The samples obtained by SPS were of the disordered cubic type of perovskite as shown in the

same density (= ca. 7.5 g/cm<sup>3</sup>) synthesized by SSR. The crystallization with densification of

the variation of the Q∙f should be clarified.

microstructure on the high Q are discussed.

As both ordered and disordered BZT — with similar microstructures — can be obtained by various heat treatments using a conventional solid state reaction (SSR) as described in the previous section, the effects of the crystal-structural ordering and ceramic microstructure were discussed independently. In the case of BZT, although the high density and high Q ceramics of ordered BZT were synthesized by SSR with a long sintering time of over 80 hours, the dense ceramics of disordered BZT have not been obtained by SSR. Koga et al. [57] created high density disordered BZT ceramics with a short sintering time by spark plasma sintering (SPS). In this section, the effects of crystal-structural ordering and ceramic

Fig. 14 shows the Q∙f as a function of the densities in BZT fabricated using SSR and SPS. The samples obtained by SPS were of the disordered cubic type of perovskite as shown in the XRPD pattern (Fig. 15) with a lone 422 reflection compared with the ordered trigonal type with peak separations of 422 and 226 when sintered using SSR (1400<sup>o</sup>C 100 h). The SPS samples with high densities were obtained using an extremely short sintering time of 5 mins between 1150 and 1300 <sup>o</sup>C under 30 Mpa [57]. The short time sintering time when using SPS may result in the disordered BZT with a high density of 7.62 g/cm<sup>3</sup>, which is approximately 50% higher than that of low density samples of 5.0-6.0 g/cm<sup>3</sup> synthesized by conventional SSR. The full width at half maximum (FWHM) of the 420 peak became narrower with an increase in the temperature from 1,100 to 1,300 <sup>o</sup>C (Fig. 15). This indicates that the degree of crystallization of the disordered cubic phase is improved without the need for conversion to the ordered trigonal phase. Regardless of the method of synthesis, Q∙f is strongly dependent on density, and Q∙f values were improved with density as shown in Fig. 14. The highly crystallized dense disordered BZT ceramics synthesized by SPS showed a significantly high Q∙f (= 53,400 GHz) similar to that of the ordered BZT sample with the same density (= ca. 7.5 g/cm<sup>3</sup>) synthesized by SSR. The crystallization with densification of

BZT system than structural ordering and grain size. In the high density region (> 7.5 g/cm<sup>3</sup>),

**Figure 14.** *Q⋅f* of BZT by solid state reaction (SSR) and spark plasma sintering (SPS) as a function of density. Order: ordered perovskite, Disorder: disordered perovskite. BZT ceramics should play a more important role in the improvement of the Q factor in the BZT system than structural ordering and grain size. In the high density region (> 7.5 g/cm<sup>3</sup>),

**Figure 15.** XRPD patterns around 420 diffraction of BZT sintering by SPS for 5 min under 30 MPa with different sinter‐ ing temperature.

## **• Ba(Zn1/3Nb2/3)O3 (BZN) with clear order-disorder transition** [58]

Ordering based on the order-disorder transition brings low symmetry, and disordering brings high symmetry as described above. Usually, high symmetry also brings a high *Q*, similar to ordering. We present an example showing that high symmetry is more influential in bringing about a high *Q* than ordering is.

BZN clearly shows an order-disorder transition temperature at 1,350 °C as shown in Fig. 16 (a). The transition temperatures of BMT and BZT are un clear because of the high transition temperature. The ordering was confirmed using X-ray diffraction patterns. Fig. 16 shows *Q⋅f*, grain size and density as a function of the sintering temperature of BZN. The disordered sample sintered at 1,400 °C shows a drastic increase of *Q⋅f*, grain size and density when compared with ordered samples sintered at 1,200 and 1,300 °C. As a result of the postannealing at 1,200 °C over 100h for the disordered sample sintered at 1,400 °C, the structure transformed to order, but the *Q⋅f* did not improve and instead it decreased slightly in an inverse manner. The grain sizes and densities were not changed by the annealing, as shown in Figs. 16(b) and (c).

**Figure 16.** *Q⋅f* (a), grain size (b) and density (c) of BZN with transition temperature at 1350 °C as a function of sintering temperature. Although the disorder phase with a high *Q* sintered at 1400 °C annealed at 1200 °C, the *Q⋅f* did not im‐ prove.

Therefore, we can conclude that the crystal-structural ordering in the ceramic BZN system has no significant effect on the improvement of the *Q* factor. The *Q* factor strongly depends on the density and grain size, but not on the crystal-structure order. The decrease in *Q⋅f* as a result of annealing might be dependent on the low symmetry that accompanies ordering.

*3.1.2.3. Phase relations and Q⋅f in the vicinity of BZT [56, 59]*

## **• Koga's research on BZT**

P26, Fig. 22

As both ordered and disordered BZT — with similar microstructures — can be obtained by various heat treatments using a conventional solid state reaction (SSR) as described in the previous section, the effects of the crystal-structural ordering and ceramic microstructure were discussed independently. In the case of BZT, although the high density and high Q ceramics of ordered BZT were synthesized by SSR with a long sintering time of over 80 hours, the dense ceramics of disordered BZT have not been obtained by SSR. Koga et al. [57] created high density disordered BZT ceramics with a short sintering time by spark plasma sintering (SPS). In this section, the effects of crystal-structural ordering and ceramic

Fig. 14 shows the Q∙f as a function of the densities in BZT fabricated using SSR and SPS. The samples obtained by SPS were of the disordered cubic type of perovskite as shown in the XRPD pattern (Fig. 15) with a lone 422 reflection compared with the ordered trigonal type with peak separations of 422 and 226 when sintered using SSR (1400<sup>o</sup>C 100 h). The SPS samples with high densities were obtained using an extremely short sintering time of 5 mins between 1150 and 1300 <sup>o</sup>C under 30 Mpa [57]. The short time sintering time when using SPS may result in the disordered BZT with a high density of 7.62 g/cm<sup>3</sup>, which is approximately 50% higher than that of low density samples of 5.0-6.0 g/cm<sup>3</sup> synthesized by conventional SSR. The full width at half maximum (FWHM) of the 420 peak became narrower with an increase in the temperature from 1,100 to 1,300 <sup>o</sup>C (Fig. 15). This indicates that the degree of crystallization of the disordered cubic phase is improved without the need for conversion to the ordered trigonal phase. Regardless of the method of synthesis, Q∙f is strongly dependent on density, and Q∙f values were improved with density as shown in Fig. 14. The highly crystallized dense disordered BZT ceramics synthesized by SPS showed a significantly high Q∙f (= 53,400 GHz) similar to that of the ordered BZT sample with the same density (= ca. 7.5 g/cm<sup>3</sup>) synthesized by SSR. The crystallization with densification of BZT ceramics should play a more important role in the improvement of the Q factor in the BZT system than structural ordering and grain size. In the high density region (> 7.5 g/cm<sup>3</sup>),

microstructure on the high Q are discussed.

the variation of the Q∙f should be clarified.

As both ordered and disordered BZT — with similar microstructures — can be obtained by various heat treatments using a conventional solid state reaction (SSR) as described in the previous section, the effects of the crystal-structural ordering and ceramic microstructure were discussed independently. In the case of BZT, although the high density and high Q ceramics of ordered BZT were synthesized by SSR with a long sintering time of over 80 hours, the dense ceramics of disordered BZT have not been obtained by SSR. Koga et al. [57] created high density disordered BZT ceramics with a short sintering time by spark plasma sintering (SPS). In this section, the effects of crystal-structural ordering and ceramic

Fig. 14 shows the Q∙f as a function of the densities in BZT fabricated using SSR and SPS. The samples obtained by SPS were of the disordered cubic type of perovskite as shown in the XRPD pattern (Fig. 15) with a lone 422 reflection compared with the ordered trigonal type with peak separations of 422 and 226 when sintered using SSR (1400<sup>o</sup>C 100 h). The SPS samples with high densities were obtained using an extremely short sintering time of 5 mins between 1150 and 1300 <sup>o</sup>C under 30 Mpa [57]. The short time sintering time when using SPS may result in the disordered BZT with a high density of 7.62 g/cm<sup>3</sup>, which is approximately 50% higher than that of low density samples of 5.0-6.0 g/cm<sup>3</sup> synthesized by conventional SSR. The full width at half maximum (FWHM) of the 420 peak became narrower with an increase in the temperature from 1,100 to 1,300 <sup>o</sup>C (Fig. 15). This indicates that the degree of crystallization of the disordered cubic phase is improved without the need for conversion to the ordered trigonal phase. Regardless of the method of synthesis, Q∙f is strongly dependent on density, and Q∙f values were improved with density as shown in Fig. 14. The highly crystallized dense disordered BZT ceramics synthesized by SPS showed a significantly high Q∙f (= 53,400 GHz) similar to that of the ordered BZT sample with the same density (= ca. 7.5 g/cm<sup>3</sup>) synthesized by SSR. The crystallization with densification of BZT ceramics should play a more important role in the improvement of the Q factor in the BZT system than structural ordering and grain size. In the high density region (> 7.5 g/cm<sup>3</sup>),

298 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 14.** *Q⋅f* of BZT by solid state reaction (SSR) and spark plasma sintering (SPS) as a function of density. Order:

**Figure 15.** XRPD patterns around 420 diffraction of BZT sintering by SPS for 5 min under 30 MPa with different sinter‐

Ordering based on the order-disorder transition brings low symmetry, and disordering brings high symmetry as described above. Usually, high symmetry also brings a high *Q*, similar to ordering. We present an example showing that high symmetry is more influential in bringing

BZN clearly shows an order-disorder transition temperature at 1,350 °C as shown in Fig. 16 (a). The transition temperatures of BMT and BZT are un clear because of the high transition temperature. The ordering was confirmed using X-ray diffraction patterns. Fig. 16 shows *Q⋅f*, grain size and density as a function of the sintering temperature of BZN. The disordered sample sintered at 1,400 °C shows a drastic increase of *Q⋅f*, grain size and density when compared with ordered samples sintered at 1,200 and 1,300 °C. As a result of the postannealing at 1,200 °C over 100h for the disordered sample sintered at 1,400 °C, the structure transformed to order, but the *Q⋅f* did not improve and instead it decreased slightly in an inverse manner. The grain sizes and densities were not changed by the annealing, as shown

**• Ba(Zn1/3Nb2/3)O3 (BZN) with clear order-disorder transition** [58]

ordered perovskite, Disorder: disordered perovskite.

the variation of the Q∙f should be clarified.

about a high *Q* than ordering is.

in Figs. 16(b) and (c).

ing temperature.

microstructure on the high Q are discussed.

Koga *et al*. [56, 59] studied the phase relation in the vicinity of BZT in the BaO-ZnO-TaO5/2 ternary system as shown in Fig. 17. These samples were sintered at 1,400 °C for 100 hours as reported in Koga's paper [56]. These diffraction patterns fit the Rietveld method well [63]. Ordering ratios obtained are shown in Fig. 18(a). Three areas in the vicinity of BZT are presented as shown in Fig. 17.

P20, Fig.17 **Figure 17.** BaO-ZnO-TaO5/2 partial ternary system in the vicinity of BZT. Synthesized compositions are shown by the letters A to S. The A point is pure BZT. Three areas are shown and these are (I) for order/single phase, (II) for order/ secondary phase, and (III) for disorder/single phase.

n=4 4:2:7

n=2 2:2:5 n=1 1:2:4

n=0

Pseudo-tungsten-bronze solid solutions Ba6-3xR8+2xTi18O<sup>54</sup> s. s. 3Ba2+ → 2R3+ + V

/ Table 2: Style of this table changed: Corrected table is located at Correction 4/4. 1. 1st line of the table at each page is set a line with species of perovskite such as (a ) Simple perovskite. In the case of the column continued, put the "cont." at the end of the line.

**Figure 18.** Ordering ratio (a), *Q⋅f* (b) and density (c) as a function of composition deviation from pure BZT.


The first area (I) is composed of a single phase of BZT with an ordered structure and a high *Q⋅f*. The compositions E and K have a *Q⋅f* about 50 % higher than that of the pure BZT composition A. Composition K is located on the boundary area (I) and has a minor secondary phase as revealed by the SEM figure reported in a previous paper [59]. The ordering ratio in composition E is lower than A, and the density of composition E is same as that of A. The second area (II) is an ordered BZT with a secondary phase BaTa2O6 with a specific amount of Zn [59] analyzed by X-ray microanalyzer (XMA). The ordering ratio of compounds located in this area is high at about 70 to 8 0 % as shown in Fig. 18(a). Although the structure is ordered, the *Q⋅f* values decrease according to the deviation from pure BZT as shown in Fig. 18(b). The composition of the ordered BZT compounds is located on Ta2O5 rich side, which is precipitated with secondary phase as a eutectic phase diagram system. The third area (III) is precipitated as a single phase BZT solid solution with a disordered structure. The *Q⋅f* values degrade with a decrease in the ordering ratio and density as shown in Fig. 18(c). The lower density comes from the existence of numerous pores due to hard sintering. The single phase in this area is originated by a solid solution accompanying defects in *B*- and O-sites, which causes degrada‐

**Figure 19.** Partial BaO-MgO-TaO5/2 ternary system in the vicinity of BMT. On the tie line BMT-BaTa4/5O3, Ba(Mg1/3−α/3Ta2/3+2α/15Vα/5)O3 solid solutions are formed with high densities and high *Q* values, in which *A*- and Osites are filled, and the *B*-site has vacancies without charge. Three areas are divided by two lines: α = 5γ /4 and α = γ /2. The first one is *B*- and the O-site is vacant although the *A*-site is filled. The second one is *A*- and the *B*-site is vacant although the O-site is filled. The third one is *A*- and the O-site has vacancies, although the *B*-site is filled.

tion of *Q⋅f*. The pores and defects were examined by SEM [59] and Raman scattering spectra [62] respectively.

## **• Kugimiya's research on BMT/BMT** [60, 61]

**Figure 18.** Ordering ratio (a), *Q⋅f* (b) and density (c) as a function of composition deviation from pure BZT.

The first area (I) is composed of a single phase of BZT with an ordered structure and a high *Q⋅f*. The compositions E and K have a *Q⋅f* about 50 % higher than that of the pure BZT composition A. Composition K is located on the boundary area (I) and has a minor secondary phase as revealed by the SEM figure reported in a previous paper [59]. The ordering ratio in composition E is lower than A, and the density of composition E is same as that of A. The second area (II) is an ordered BZT with a secondary phase BaTa2O6 with a specific amount of Zn [59] analyzed by X-ray microanalyzer (XMA). The ordering ratio of compounds located in this area is high at about 70 to 8 0 % as shown in Fig. 18(a). Although the structure is ordered, the *Q⋅f* values decrease according to the deviation from pure BZT as shown in Fig. 18(b). The composition of the ordered BZT compounds is located on Ta2O5 rich side, which is precipitated with secondary phase as a eutectic phase diagram system. The third area (III) is precipitated as a single phase BZT solid solution with a disordered structure. The *Q⋅f* values degrade with a decrease in the ordering ratio and density as shown in Fig. 18(c). The lower density comes from the existence of numerous pores due to hard sintering. The single phase in this area is originated by a solid solution accompanying defects in *B*- and O-sites, which causes degrada‐

(I): Ordering area with BZT single phase

(II): Ordering area with secondary phase

(III): Disordering area with BZT single phase

300 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Kugimiya [60] presented the highest *Q⋅f* composition at the Ta and Ba rich side in a BMT system as shown in Fig. 19. The synthesized samples are precise compositions formed by master batches which mixed samples from the four master batch method. Here, chemical formulae in the vicinity of BMT are reported as follows: Kugimiya presented three areas divided by drawing two lines as shown in Table 3 and Fig. 19.


**Table 3.** Chemical formula for three areas divided by two lines: *α* = 5*γ*/4 and *α* = *γ*/2, here, *α* and *γ* are in Ba*α*Ta*γ*O*<sup>α</sup>*+5*γ*/2 and vacancies are on the *A-*, *B-* and O*-*sites.

**Figure 20.** Part of the BaO–MgO–Ta2O5 phase diagram in the vicinity of the BMT phase indicating a composition de‐ pendence of the *Q⋅f* for samples sintered at 1650 °C for 20 h. Small black dots indicate the target sample compositions. Green line indicates an approximate boundary of the single-phase BMT.

Here, *α* and *γ* are in Ba*α*Ta*γ*O*<sup>α</sup>*+5*<sup>γ</sup>*/2. In the region *α* > 5*γ*/4, the composition denoted by Ba1+*α*(Mg1/3Ta2/3+*γ*V*<sup>α</sup>*-*<sup>γ</sup>*)O3+*<sup>α</sup>*+5*<sup>γ</sup>*/2V2*<sup>α</sup>*-5*<sup>γ</sup>*/2 has *B*- and O-site vacancies with holes and electrons. In the *α*=5*γ*/4line,thecompositionsdenotedbyBa1+*α*(Mg1/3Ta2/3+4*α*/5V*α*/5)O3+3*<sup>α</sup>* aretheidealoneswithout vacancies in the *A-* and O- sites. The *B*-site vacancy is neutralized without charge. The highest *Q⋅f* composition is located near the line *α* = 5*γ*/4 as shown in Fig. 19. The compositions in the line are ideal for microwave dielectrics because there are no oxygen defects and the density is high due to the substitution of Ta for Mg. In the region 5*γ*/4 > *α* > *γ*/2, the composition denot‐ ed by Ba1+*α*V*5γ*/6-2*<sup>α</sup>*/3 (Mg1/3Ta2/3+*γ*V*<sup>α</sup>*/3-*<sup>γ</sup>*/6)O3+*<sup>α</sup>*+5*<sup>γ</sup>*/2 has a defect in the *A*- and *B*-sites filled with holes and electrons. In the region at *α* = *γ*, the composition denoted by Ba1+*α*V*α*/6(Mg1/3Ta2/3+*α*V*α*/6)O3+7*<sup>α</sup>*/ 2has the sameamountofvacancies inboth*A*-and*B*-sites filledwiththe sameholesandelectrons. In the region *α* = *γ*/2, the composition denoted by Ba1+*α*V*α*(Mg1/3Ta2/3+*γ*)O3+6*α* only has vacancies in the *A*-site with holes and in the *B*-site with excess electrons which introduced instability. In the region *α* < *γ*/2, the composition denoted by Ba1+*α*V*<sup>γ</sup>*-*<sup>α</sup>*(Mg1/3Ta2/3+*γ*)O3+*<sup>α</sup>*+5*<sup>γ</sup>*/2V*<sup>γ</sup>*/2-*<sup>α</sup>* has holes in both the *A*- and O-sites with electrons and excess electrons in the *B*-site, which leads to an unstable crystal structure.

The contour lines in Fig. 19 show *Q* values from 2,000 in the outer area to 30,000 in the center. The highest *Q* value of 50,000 was obtained in the center. The contour is elongated parallel to the *Q* max line as drawn in Fig. 19 and it changes steeply on the perpendicular to the line.

## **• Kolodiazhnyi's research on BMT** [64]

The author presented a part of the BaO-MgO-Ta2O3 phase diagram in the vicinity of the BMT phase as shown in Fig. 20 [64]. Ceramic samples whose chemical composition falls within the A, B and C compositional triangles (CTs) in Fig. 20 reach a relative density of 96–98% after sintering at 1,550–1,580 °C for 20 hours. Samples located in the H and G CTs required tem‐ peratures of 1,630–1,650 °C to reach a relative density of 96-98 %. The specimens located in the D, E and F CTs retained a density of ≦80% after heat treatment at 1,680 °C. The pure BMT sintered at 1,650 °C for 20 h shows a poor dielectric performance with a *Q⋅f* ≈ 20,000-40,000 GHz. A very large variation in the dielectric properties and density of ceramics was found upon a slight deviation from pure BMT composition. The tendency of the variation was similar to Koga's results as shown in Fig. 17 [56]. Significant improvement in the *Q⋅f* vales is seen in samples with a slight Mg deficiency, which are located in the G and H CTs. The highest *Q⋅f* compositions of 330,000–340,000 GHz were found within the H CT close to the BMT-Ba3Ta2O8 tie line. Upon approaching the BMT-Ba5Ta4O15 tie line from the H CT, the *Q⋅f* starts to decrease and then drops sharply after crossing into the A CT. Mg-rich BMT with a high density and a high degree of 1:2 cation order within B and C CTs showed low *Q⋅f* values (e.g. *Q⋅f* < 20,000 GHz). The dominant source of the extrinsic dielectric loss is identified as Mg occupation substituted for Ba in the *A*-site (MgBa) which improves 'rattling' inside the dodec‐ ahedral position. Ta-poor, non-pure BMT samples located in the D, E and F CTs showed a very low density and high dialectric losses after sintering at 1,650 °C for 20 h.

## **• High** *Q* **by high density composition** [60, 61]

Here, *α* and *γ* are in Ba*α*Ta*γ*O*<sup>α</sup>*+5*<sup>γ</sup>*/2. In the region *α* > 5*γ*/4, the composition denoted by Ba1+*α*(Mg1/3Ta2/3+*γ*V*<sup>α</sup>*-*<sup>γ</sup>*)O3+*<sup>α</sup>*+5*<sup>γ</sup>*/2V2*<sup>α</sup>*-5*<sup>γ</sup>*/2 has *B*- and O-site vacancies with holes and electrons. In the *α*=5*γ*/4line,thecompositionsdenotedbyBa1+*α*(Mg1/3Ta2/3+4*α*/5V*α*/5)O3+3*<sup>α</sup>* aretheidealoneswithout vacancies in the *A-* and O- sites. The *B*-site vacancy is neutralized without charge. The highest *Q⋅f* composition is located near the line *α* = 5*γ*/4 as shown in Fig. 19. The compositions in the line are ideal for microwave dielectrics because there are no oxygen defects and the density is high due to the substitution of Ta for Mg. In the region 5*γ*/4 > *α* > *γ*/2, the composition denot‐ ed by Ba1+*α*V*5γ*/6-2*<sup>α</sup>*/3 (Mg1/3Ta2/3+*γ*V*<sup>α</sup>*/3-*<sup>γ</sup>*/6)O3+*<sup>α</sup>*+5*<sup>γ</sup>*/2 has a defect in the *A*- and *B*-sites filled with holes and electrons. In the region at *α* = *γ*, the composition denoted by Ba1+*α*V*α*/6(Mg1/3Ta2/3+*α*V*α*/6)O3+7*<sup>α</sup>*/ 2has the sameamountofvacancies inboth*A*-and*B*-sites filledwiththe sameholesandelectrons. In the region *α* = *γ*/2, the composition denoted by Ba1+*α*V*α*(Mg1/3Ta2/3+*γ*)O3+6*α* only has vacancies in the *A*-site with holes and in the *B*-site with excess electrons which introduced instability. In the region *α* < *γ*/2, the composition denoted by Ba1+*α*V*<sup>γ</sup>*-*<sup>α</sup>*(Mg1/3Ta2/3+*γ*)O3+*<sup>α</sup>*+5*<sup>γ</sup>*/2V*<sup>γ</sup>*/2-*<sup>α</sup>* has holes in both the *A*- and O-sites with electrons and excess electrons in the *B*-site, which leads to an

**Figure 20.** Part of the BaO–MgO–Ta2O5 phase diagram in the vicinity of the BMT phase indicating a composition de‐ pendence of the *Q⋅f* for samples sintered at 1650 °C for 20 h. Small black dots indicate the target sample compositions.

Green line indicates an approximate boundary of the single-phase BMT.

302 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The contour lines in Fig. 19 show *Q* values from 2,000 in the outer area to 30,000 in the center. The highest *Q* value of 50,000 was obtained in the center. The contour is elongated parallel to the *Q* max line as drawn in Fig. 19 and it changes steeply on the perpendicular to the line.

The author presented a part of the BaO-MgO-Ta2O3 phase diagram in the vicinity of the BMT phase as shown in Fig. 20 [64]. Ceramic samples whose chemical composition falls within the

unstable crystal structure.

**• Kolodiazhnyi's research on BMT** [64]

Koga's data [56] and Kolodiazhnyi's [64] data are comparable with Kugimiya's BMT data [60]. The area (I) and the H CT with the highest *Q⋅f* as shown in Figs 17 and 20 respectively, are located on the opposite side of Kugimiya's data against the BMT-Ba5Ta4O15 tie line (Fig. 19). These compositions will be comparable with the completed ideal crystal structure Ba1+*α*(Mg1/3Ta2/3+4*α*/5V*α*/5)O3+3*<sup>α</sup>* reported by Kugimiya [60]. The formula is rewritten as Ba(Mg1/3 *<sup>α</sup>*/3Ta2/3+2*α*/15V*α*/5)O3 solid solutions on the tie-line BMT-Ba5Ta4O3. The crystal structure in the composition region is perfect, without defects, and with a high density. The density of BMT increases with the introduction of the Ba5Ta4O15 phase, because Mg ions are substituted by heavy Ta ions.

Surendran *et al.* [61] also presented compositions with high *Q* values in the two kinds of magnesium and barium deficient nonstoichiometric compositions Ba(Mg1/3-*x*Ta2/3)O3 [*x*=0.015] and Ba1-*x*(Mg1/3Ta2/3)O3 [*x*=0.0075] as shown in Fig. 21(a). The microwave dielectric proper‐ ties of Ba0.9925(Mg0.33Ta0.67)O3 [*ε*r = 24.7, *TCf* = 1.2 ppm/°C, *Q⋅f* = 152,580 GHz] and Ba(Mg0.3183Ta0.67)O3 [*ε*r = 25.1, *TCf* = 3.3 ppm/°C and *Q⋅f* = 120,500 GHz] were found to be better than stoichiometric BMT [*ε*r = 24.2, *TCf* = 8 ppm/°C and *Q⋅f* = 100,500 GHz]. The important difference from Kugimiya's results [60] is standing on the nonstoichiometry with a barium or magnesium deficiency. We consider that Surendran's data [61] is based on Kugimiya's results [60]. In the case of Mg-deficient BMT, as the composition is located near Kugimiya's area with a high *Q⋅f*, the composition of the main compound must be Ba(Mg1/3 *<sup>α</sup>*/3Ta2/3+2*<sup>α</sup>*/15 V*α*/5)O3 solid solutions on the tie-line BMT-Ba5Ta4O3. As shown in Fig. 21(b), in the solid solution area, the Mg deficiencies are filled with Ta and create vacancies in the *B*-site, so that density and the ordering ratio are maintained. On the other hand, the existing area of Ba-deficient BMT is included in Koga's (II) area as shown in Fig. 17, composed of ordered BMT and secondary phase. The ordered BMT will have a similar composition with a high a high Q are usually free of defects.

density and a high *Q⋅f* on the BMT-Ba5Ta4O3 tie-line presented by Kugimiya [60]. The compound by Surendran *et al*. [61] may be located in the eutectic phase diagram region accompanying the secondary phase. However, as the amount of secondary phases is small, detection may be difficult. Though the density and ordering ratio are maintained at a high level as shown in Fig. 21(c), *Q⋅f* values degrade steeply according to the secondary phase. The compound should be stoichiometric and complete, because microwave dielectrics with a high *Q* are usually free of defects. BMT and secondary phase. The ordered BMT will have a similar composition with a high density and a high Q∙f on the BMT-Ba5Ta4O3 tie-line presented by Kugimiya [60]. The compound by Surendran et al. [61] may be located in the eutectic phase diagram region accompanying the secondary phase. However, as the amount of secondary phases is small, detection may be difficult. Though the density and ordering ratio are maintained at a high level as shown in Fig. 21(c), Q∙f values degrade steeply according to the secondary phase. The compound should be stoichiometric and complete, because microwave dielectrics with

Fig. 21.(a) Q∙f for Ba(Mg1/3−<sup>x</sup>Ta2/3)O3 and Ba1−<sup>x</sup>(Mg1/3Ta2/3)O3 as a function of composition deviation (x), (b) Bulk density and ordering parameter for Ba(Mg1/3−<sup>x</sup>Ta2/3)O3 as a function of x, (c) Bulk density and ordering parameter for Ba1−<sup>x</sup>(Mg1/3Ta2/3)O3 as a function of x. **Figure 21.** (a) *Q⋅f* for Ba(Mg1*/*3−*<sup>x</sup>*Ta2*/*3)O3 and Ba1−*<sup>x</sup>*(Mg1*/*3Ta2*/*3)O3 as a function of composition deviation (*x*), (b) Bulk den‐ sity and ordering parameter for Ba(Mg1*/*3−*<sup>x</sup>*Ta2*/*3)O3 as a function of *x*, (c) Bulk density and ordering parameter for Ba1−*<sup>x</sup>*(Mg1*/*3Ta2*/*3)O3 as a function of *x*.

#### 3.1.2.4) Important points concerning complex perovskite *3.1.2.4. Important points concerning complex perovskite*


the ordered phase is a trigonal (rhombohedral) *R3c* (No. 167), and the disorder phase is a cubic *Pm3m* (No. 221) [55, 65].


## **3.2. Perovskite related compounds**

density and a high *Q⋅f* on the BMT-Ba5Ta4O3 tie-line presented by Kugimiya [60]. The compound by Surendran *et al*. [61] may be located in the eutectic phase diagram region accompanying the secondary phase. However, as the amount of secondary phases is small, detection may be difficult. Though the density and ordering ratio are maintained at a high level as shown in Fig. 21(c), *Q⋅f* values degrade steeply according to the secondary phase. The compound should be stoichiometric and complete, because microwave dielectrics with

Fig. 21.(a) Q∙f for Ba(Mg1/3−<sup>x</sup>Ta2/3)O3 and Ba1−<sup>x</sup>(Mg1/3Ta2/3)O3 as a function of composition deviation (x), (b) Bulk density and ordering parameter for Ba(Mg1/3−<sup>x</sup>Ta2/3)O3 as a function of

**Figure 21.** (a) *Q⋅f* for Ba(Mg1*/*3−*<sup>x</sup>*Ta2*/*3)O3 and Ba1−*<sup>x</sup>*(Mg1*/*3Ta2*/*3)O3 as a function of composition deviation (*x*), (b) Bulk den‐ sity and ordering parameter for Ba(Mg1*/*3−*<sup>x</sup>*Ta2*/*3)O3 as a function of *x*, (c) Bulk density and ordering parameter for

∙ A complex perovskite is composed of different ions with different charges such as

**•** A complex perovskite is composed of different ions with different charges such as *A*2+

**•** A complex perovskite usually has an order-disorder phase transition. The order phase is a low temperature phase with low crystallographic symmetry, while the disorder phase is a high temperature phase with high symmetry. In the case of a 1:3 type complex perovskite,

x, (c) Bulk density and ordering parameter for Ba1−<sup>x</sup>(Mg1/3Ta2/3)O3 as a function of x.

3.1.2.4) Important points concerning complex perovskite

*3.1.2.4. Important points concerning complex perovskite*

Ba1−*<sup>x</sup>*(Mg1*/*3Ta2*/*3)O3 as a function of *x*.

A2+(B2+1/3B5+2/3) O3, thereby maintaining the charge valance.

(*B*2+1/3*B*5+2/3) O3, thereby maintaining the charge valance.

BMT and secondary phase. The ordered BMT will have a similar composition with a high density and a high Q∙f on the BMT-Ba5Ta4O3 tie-line presented by Kugimiya [60]. The compound by Surendran et al. [61] may be located in the eutectic phase diagram region accompanying the secondary phase. However, as the amount of secondary phases is small, detection may be difficult. Though the density and ordering ratio are maintained at a high level as shown in Fig. 21(c), Q∙f values degrade steeply according to the secondary phase. The compound should be stoichiometric and complete, because microwave dielectrics with

a high *Q* are usually free of defects.

a high Q are usually free of defects.

304 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

## *3.2.1. Pseudo-tungsten-bronze solid solutions*

## **• Crystal structure of pseudo-tungsten-bronze solid solutions**

The pseudo-tungsten-bronze Ba6-3*xR*8+2*<sup>x</sup>*Ti18O54 (*R* = rare earth) solid solutions [68, 69] are located on the perovskite-type compound tie-line of BaTiO3 and *R*2Ti3O9 compositions on the BaO-*R*2O3-TiO2 ternary phase diagram as shown in Fig. 22. The crystal structure contains perovskite blocks of 2 x 2 unit cells (2x2), and pentagonal (*A*2) sites as shown in Fig. 23, which are named from similar tetragonal tungsten-bronze structure with 1x1 perovskite blocks [70–72]. These compounds contain two ions with different atomic sizes. The larger Ba ions are found mainly in the pentagonal *A*2-site and the smaller rare-earth (*R*) ions in the rhombic *A*1-site. This structure has two more sites, *B* and *C*. The *B*-site is same as perovskite octahedral sites, and the *C*-site is a triangular site which is usually empty. This structure has a close relationship to the structure of perovskite. If the two ions become the same size, the structure changes to perovskite with only cubic *A*1-sites owing to the combination of the *A*2 and *C*-sites as described later at section 4 (Fig. 40). The crystal data are as follows: orthorhombic crystal system of space group *Pbnm* (No.62), point group *mmm*, lattice parameter *a* = 12.13, *b* = 22.27, *c* = 7.64 Å, *Z* = 2, *D*x = 5.91 g/cm3 . This structure has a super lattice along the c-axis of twice [73]. As the space group has a center of symmetry of *i* as do paraelectrics, it qualifies for microwave dielectrics. The chemical formula of all occupied sites is Ba6*R*8*C*4Ti18O54 and the structural formula is [*R*8Ba2]*A*1[Ba4]*A*2[V]*C*[Ti18]BO54, where V is vacancy. As this chemical formula is *x* = 0, the chemical formula of the solid solutions is Ba6-3*xR*8+2*<sup>x</sup>*Ti18O54, and the structural formula is [Ba4]*A*2[Ba2-3*xR*8+2x]*A*1Ti18O54. Here, the amount of Ba in the *A*1-site becomes zero if 2-3*x* = 0, that is, *x* = 2/3. This composition is special due to one factor : the structure formula is [Ba4]*A*2[*R*8+4/3]*A*1Ti18O54 and is occupied separately by Ba in *A*2 and by *R* in *A*1 as shown in Fig. 24. This special composition is called "compositional ordering".

P20, Fig.17

P26, Fig. 22

33 34

● ● ●

<sup>M</sup> <sup>L</sup> <sup>K</sup>

S R Q

ZnO <sup>35</sup> TaO5/2

phase Ba(Zn1/3Ta2/3)O<sup>3</sup>

▲ ▲

B C D

▲

17

line 3: y = 1.0 x = 0.97-1.03 z = 0.97-1.03

18

line 2: x = 1.0 y = 0.97-1.03 z= 0.97-1.03

19

Disorder Single phase

(III)

□ □ □

▲ ▲

▲

G J E

●

(I) Order Single phase

□ □ □

N O P

A

51

line 1: z = 1.0 x = 0.97-1.03 y = 0.97-1.03

50

(II) Order Secondary

● ●

<sup>H</sup> <sup>I</sup> <sup>J</sup>

16 52

BaO

/ Table 2: Style of this table changed: Corrected table is located at Correction 4/4. **Figure 22.** Part of the BaO-*R*2O3-TiO2 ternary phase diagram with pseudo-tungsten-bronze type solid solutions and ho‐ mologous compounds.

1. 1st line of the table at each page is set a line with species of perovskite such as (a ) Simple

**Figure 23.** Crystal structure of the pseudo-tungsten-bronze solid solutions viewed in projection along [001]. Pentago‐ nal sites (*A*2) are located among 2x2 perovskite blocks with rhombic sites (*A*1).

### **• Microwave dielectric properties of pseudo-tungsten-bronze solid solutions**

The quality factor *Q⋅f* of the *x* = 2/3 composition, in which *R* and Ba ions separately occupy the rhombic site *A*1 and the pentagonal site *A*2 respectively, show the highest *Q f* values: 10,549 GHz in the Sm system, 10,010 GHz in the Nd system, and 2,024 GHz in the La system4) as shown in Fig.25 (a) [74]. The highest quality factor is based on the compositional ordering of *R* and Ba ions in the *A*1 and *A*2 sites respectively, as shown in Fig. 24. The ordering distribution of the ions reduces the internal strain and results in the non-linear variation in quality factor.

variation in quality factor.

The quality factor Q∙f of the x = 2/3 composition, in which R and Ba ions separately occupy the rhombic site A1 and the pentagonal site A2 respectively, show the highest Q·f values: 10,549 GHz in the Sm system, 10,010 GHz in the Nd system, and 2,024 GHz in the La system4) as shown in Fig.25(a) [74]. The highest quality factor is based on the compositional ordering of R and Ba ions in the A1 and A2 sites respectively, as shown in Fig. 24. The

P20, Fig.17

P26, Fig. 22

mologous compounds.

/ Table 2: Style of this table changed: Corrected table is located at Correction 4/4. 1. 1st line of the table at each page is set a line with species of perovskite such as (a ) Simple perovskite. In the case of the column continued, put the "cont." at the end of the line.

**Figure 22.** Part of the BaO-*R*2O3-TiO2 ternary phase diagram with pseudo-tungsten-bronze type solid solutions and ho‐

**Figure 23.** Crystal structure of the pseudo-tungsten-bronze solid solutions viewed in projection along [001]. Pentago‐

The quality factor *Q⋅f* of the *x* = 2/3 composition, in which *R* and Ba ions separately occupy the rhombic site *A*1 and the pentagonal site *A*2 respectively, show the highest *Q f* values: 10,549 GHz in the Sm system, 10,010 GHz in the Nd system, and 2,024 GHz in the La system4) as shown in Fig.25 (a) [74]. The highest quality factor is based on the compositional ordering of *R* and Ba ions in the *A*1 and *A*2 sites respectively, as shown in Fig. 24. The ordering distribution of the ions reduces the internal strain and results in the non-linear variation in quality factor.

**• Microwave dielectric properties of pseudo-tungsten-bronze solid solutions**

nal sites (*A*2) are located among 2x2 perovskite blocks with rhombic sites (*A*1).

Disorder Single phase

(III)

□ □ □

▲ ▲

▲

G J E

●

(I) Order Single phase

□ □ □

N O P

A

33 34

n=4 4:2:7

n=2 2:2:5

n=0

n=1 1:2:4 ● ● ●

<sup>M</sup> <sup>L</sup> <sup>K</sup>

S R Q

ZnO <sup>35</sup> TaO5/2

phase Ba(Zn1/3Ta2/3)O<sup>3</sup>

▲ ▲

B C D

▲

17

line 3: y = 1.0 x = 0.97-1.03 z = 0.97-1.03

18

line 2: x = 1.0 y = 0.97-1.03 z= 0.97-1.03

19

306 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Pseudo-tungsten-bronze solid solutions Ba6-3xR8+2xTi18O<sup>54</sup> s. s. 3Ba2+ → 2R3+ + V

51

line 1: z = 1.0 x = 0.97-1.03 y = 0.97-1.03

50

(II) Order Secondary

● ●

<sup>H</sup> <sup>I</sup> <sup>J</sup>

16 52

BaO

Fig. 24. Crystal structure of disordering (a) and compositional ordering (b) on pseudotungsten-bronze solid solutions. **Figure 24.** Crystal structure of disordering (a) and compositional ordering (b) on pseudo-tungsten-bronze solid solu‐ tions.

Fig. 25. Q∙f values (a), εr (b) and TCf (c) of pseudo-tungsten-bronze type solid solutions as a function of x in Ba6-3xR8+2xTi18O54 solid solutions. **Figure 25.** *Q⋅f* values (a), *ε*r (b) and *TCf* (c) of pseudo-tungsten-bronze type solid solutions as a function of *x* in Ba6-3x*R*8+2*<sup>x</sup>*Ti18O54 solid solutions.

Internal strain η values for x = 0.3, 0.5, 2/3 and 0.7 are shown in Fig. 26 [74]. It should be noted that the internal strain for x = 2/3 is the lowest in the series of Ba6-3xSm8+2xTi18O54 solid solutions. This low internal strain comes from the compositional ordering. As the x-values decrease according to the structural formula [R8+2xBa2-3xVx]A1[Ba4]A2Ti18O54 in the range of 0 ≤ x ≤ 2/3, Ba ions with their larger ionic radii will also occupy a part of the rhombic sites with their smaller size. The location of Ba ions in the A1-site leads to internal strain around the ions themselves, lowering the Q·f values. Moreover, the vacancies generated in the A1 sites by the substitution of 3Ba by 2R might be the second reason for the lowering of the internal strain and may lead to the high Q·f values. On the other hand, as the x-values increase according to the structural formula [R9.33+2(x-2/3)V0.66-(x-2/3)]A1[Ba4-3(x-2/3)V3(x-2/3)]A2Ti18O54 in the range of 2/3 ≤ x ≤ 0.7, then Ba ions in pentagonal A2-sites are substituted with vacancies and R ion occupy the vacancies in A1-site. The decrease in Ba ions produces vacancies in A2 sites and may lead to unstable crystal structures. Moreover, the decrease in the number of vacancies in the rhombic A1-sites, accompanied by the decrease of Ba ions in the pentagonal sites might lead to an additional internal strain. These strains are the reason for the lower quality factor at x = 0.7. The internal strain around x = 0.7 might lead to the limit of solid solutions as shown in Fig. 27 [75]. The solid solution area is different based on the R ions: the region is 0.3 ≤ x ≤ 0.7 in the case of Sm and 0.0 ≤ x ≤ 0.7 in Pr, Nd and La with

inflection points at x = 0.2, which may be based on the different substitution sites.

On the other hand, the Q·f values of each R analogue with x = 2/3 in the Ba6-3xR8+2xTi18O<sup>54</sup> solid solutions increase according to a decrease in the rare-earth ion size (lanthanide

contraction) as shown in Fig. 28. The Sm analogue has a better Q·f than the La analogue, at ca. 10,000 GHz. This crystal structure is maintained by the size difference of the cations between the Ba and R ions. It was revealed that the crystal structure with the largest size difference between Ba and Sm ions shows an excellent quality factor as it has low internal strain. On the other hand, the La analogue shows a low Q∙f of ca. 2,000 GHz. Though the Q∙f values of the Pr, Nd and Sm analogues show a linear relationship, that of the La analogue deviates from the linear relationship as shown in Fig. 28 [76]. If the changes in Q∙f

Fig. 26. (a) Internal strain η obtained from the slope of equation β cosθ = λ/t + 2ηsinθ. (b) **Figure 26.** (a) Internal strain *η* obtained from the slope of equation *β* cos*θ* = *λ*/*t* + 2*η*sin*θ*. (b)

Internal strain *η* values for *x* = 0.3, 0.5, 2/3 and 0.7 are shown in Fig. 26 [74]. It should be noted that the internal strain for *x* = 2/3 is the lowest in the series of Ba6-3*x*Sm8+2*x*Ti18O54 solid solutions. This low internal strain comes from the compositional ordering. As the *x*-values decrease according to the structural formula [*R*8+2*<sup>x</sup>*Ba2-3*x*V*x*]*A*1[Ba4]*A*2Ti18O54 in the range of 0 ≤ *x* ≤ 2/3, Ba ions with their larger ionic radii will also occupy a part of the rhombic sites with their smaller size. The location of Ba ions in the *A*1- site leads to internal strain around the ions themselves, lowering the *Q f* values. Moreover, the vacancies generated in the *A*1- sites by the substitution of 3Ba by 2*R* might be the second reason for the lowering of the internal strain and may lead to the high *Q f* values. On the other hand, as the *x*-values increase according to the structural formula [*R*9.33+2(*<sup>x</sup>*-2/3)V0.66-(*x*-2/3)]*A*1[Ba4-3(*x*-2/3)V3(*x*-2/3)]*A*2Ti18O54 in the range of 2/3 ≤ *x* ≤ 0.7, then Ba ions in pentagonal *A*2- sites are substituted with vacancies and *R* ion occupy the vacancies in *A*1 site. The decrease in Ba ions produces vacancies in *A*2-sites and may lead to unstable crystal structures. Moreover, the decrease in the number of vacancies in the rhombic *A*1-sites, accompanied by the decrease of Ba ions in the pentagonal sites might lead to an additional internal strain. These strains are the reason for the lower quality factor at *x* = 0.7. The internal strain around *x* = 0.7 might lead to the limit of solid solutions as shown in Fig. 27 [75]. The solid solution area is different based on the *R* ions: the region is 0.3 ≤ *x* ≤ 0.7 in the case of Sm and 0.0 ≤ *x* ≤ 0.7 in Pr, Nd and La with inflection points at *x* = 0.2, which may be based on the different substitution sites.

On the other hand, the *Q⋅f* values of each *R* analogue with *x* = 2/3 in the Ba6-3x*R*8+2xTi18O54 solid solutions increase according to a decrease in the rare-earth ion size (lanthanide contraction) as shown in Fig. 28. The Sm analogue has a better *Q⋅f* than the La analogue, at ca. 10,000 GHz. This crystal structure is maintained by the size difference of the cations between the Ba and *R* ions. It was revealed that the crystal structure with the largest size difference between Ba and Sm ions shows an excellent quality factor as it has low internal strain. On the other hand, the La analogue shows a low *Q⋅f* of ca. 2,000 GHz. Though the *Q⋅f* values of the Pr, Nd and Sm analogues show a linear relationship, that of the La analogue deviates from the linear rela‐ tionship as shown in Fig. 28 [76]. If the changes in *Q⋅f* are affected only by ionic radius, then the relationship should be linear. The reason for the deviation might be internal stress depending on the stability of the crystal structure. There are two different cation sites: the *A*1 site in the perovskite block and the *A*2-site in the differently sized pentagonal columns as described before, which are occupied by differently sized cations. As the difference in ionic radius between Ba and La is not large in comparison with other *R* ions, the crystal structure is not stable, and shows a tendency of changing toward a perovskite structure which has only a single site for large cations. So, in the case of the La-ion, the internal stress always exists as an intrinsic quality, and the internal stress might cause the deviation of *Q⋅f* from the expected linear relationship. The *ε*r and *TCf* lines against ionic radius of *R* increase according to the increasing size of ionic radius. The parameters are not affected by the crystal structure. The reason why the *ε*r and *TCf* lines are proportional has not yet been clarified [76].

The dielectric constant *ε*r is affected by the following three factors: (I) volume of TiO6 octahedra; (II) tilting of octahedra strings; and (III) polarizabilities of *R* and Ba ions [77]. The dielectric constants *ε*r of the solid solutions are proportional to lattice parameters or cell volumes as shown in Fig. 29. As *x* increased, *ε*r decreased linearly (Fig. 25(b)), and lattice parameters or cell volumes also decreased linearly (Fig. 27). Usually, in the perovskite structure, the polarity of the Ti ion in the octahedra is produced as a result of the large octahedral volume. Thus, as the mean value of the volume decreased from 9.946 Å3 at *x*=0.5 to 9.925 Å3 at *x*=0.7, this volume change is considered to have decreased *ε*r. However, the volume change is very small, thus other effects should be examined such as tilting of the TiO6 octahedra strings as suggested by Valant *et al*. [78]. The tilting angle, which is that between the *c*-axis and the central axis of the octahedra as shown in Fig. 30, is inversely proportional to lattice parameters: the mean tilting angle is 9.99° at *x*=0.5 and 10.63° at *x*=0.7, based on the refined crystal structure of the Sm solid solution series [13]. From Fig. 29, it was also deduced that the polarizabilities of *R* ions affect *ε*r and *TCf*. In the table of polarizabilities derived by Shannon [79], the La ion, which gives the largest *ε*<sup>r</sup> in the series, also has the largest polarizability among these *R* ions: 6.03 for La, 5.31 for Pr, 5.01 for Nd and 4.74 Å3 for Sm. The *ε*r values decrease with the polarizabilities. On the other hand, the *ε*r values also vary linearly as a function of cell volume in each *R*-system as shown in Fig. 29 (a). The variations in *ε*r are also affected by the polarizabilities of *R* and Ba ions. The substitution is performed according to the following equation:

Internal strain *η* values for *x* = 0.3, 0.5, 2/3 and 0.7 are shown in Fig. 26 [74]. It should be noted that the internal strain for *x* = 2/3 is the lowest in the series of Ba6-3*x*Sm8+2*x*Ti18O54 solid solutions. This low internal strain comes from the compositional ordering. As the *x*-values decrease according to the structural formula [*R*8+2*<sup>x</sup>*Ba2-3*x*V*x*]*A*1[Ba4]*A*2Ti18O54 in the range of 0 ≤ *x* ≤ 2/3, Ba ions with their larger ionic radii will also occupy a part of the rhombic sites with their smaller size. The location of Ba ions in the *A*1- site leads to internal strain around the ions themselves, lowering the *Q f* values. Moreover, the vacancies generated in the *A*1- sites by the substitution of 3Ba by 2*R* might be the second reason for the lowering of the internal strain and may lead to the high *Q f* values. On the other hand, as the *x*-values increase according to the structural formula [*R*9.33+2(*<sup>x</sup>*-2/3)V0.66-(*x*-2/3)]*A*1[Ba4-3(*x*-2/3)V3(*x*-2/3)]*A*2Ti18O54 in the range of 2/3 ≤ *x* ≤ 0.7, then Ba ions in pentagonal *A*2- sites are substituted with vacancies and *R* ion occupy the vacancies in *A*1 site. The decrease in Ba ions produces vacancies in *A*2-sites and may lead to unstable crystal structures. Moreover, the decrease in the number of vacancies in the rhombic *A*1-sites, accompanied by the decrease of Ba ions in the pentagonal sites might lead to an additional internal strain. These strains are the reason for the lower quality factor at *x* = 0.7. The internal strain around *x* = 0.7 might lead to the limit of solid solutions as shown in Fig. 27 [75]. The solid solution area is different based on the *R* ions: the region is 0.3 ≤ *x* ≤ 0.7 in the case of Sm and 0.0 ≤ *x* ≤ 0.7 in Pr, Nd and La with inflection points at *x* = 0.2, which may be based on the

308 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

On the other hand, the *Q⋅f* values of each *R* analogue with *x* = 2/3 in the Ba6-3x*R*8+2xTi18O54 solid solutions increase according to a decrease in the rare-earth ion size (lanthanide contraction) as shown in Fig. 28. The Sm analogue has a better *Q⋅f* than the La analogue, at ca. 10,000 GHz. This crystal structure is maintained by the size difference of the cations between the Ba and *R* ions. It was revealed that the crystal structure with the largest size difference between Ba and Sm ions shows an excellent quality factor as it has low internal strain. On the other hand, the La analogue shows a low *Q⋅f* of ca. 2,000 GHz. Though the *Q⋅f* values of the Pr, Nd and Sm analogues show a linear relationship, that of the La analogue deviates from the linear rela‐ tionship as shown in Fig. 28 [76]. If the changes in *Q⋅f* are affected only by ionic radius, then the relationship should be linear. The reason for the deviation might be internal stress depending on the stability of the crystal structure. There are two different cation sites: the *A*1 site in the perovskite block and the *A*2-site in the differently sized pentagonal columns as described before, which are occupied by differently sized cations. As the difference in ionic radius between Ba and La is not large in comparison with other *R* ions, the crystal structure is not stable, and shows a tendency of changing toward a perovskite structure which has only a single site for large cations. So, in the case of the La-ion, the internal stress always exists as an intrinsic quality, and the internal stress might cause the deviation of *Q⋅f* from the expected linear relationship. The *ε*r and *TCf* lines against ionic radius of *R* increase according to the increasing size of ionic radius. The parameters are not affected by the crystal structure. The

reason why the *ε*r and *TCf* lines are proportional has not yet been clarified [76].

The dielectric constant *ε*r is affected by the following three factors: (I) volume of TiO6 octahedra; (II) tilting of octahedra strings; and (III) polarizabilities of *R* and Ba ions [77]. The dielectric constants *ε*r of the solid solutions are proportional to lattice parameters or cell volumes as

different substitution sites.

$$\mathbf{3Ba} \Leftrightarrow \mathbf{2R+V}$$

The total polarizabilities due to the substitution equation are reduced from 3×6.40 to 2×6.03 Å3 for the La system. Here, the value of 6.40 Å3 for the Ba+2 ion is larger than that for the La+3 ion.

The *TCf* is also plotted as a function of cell volume in Fig. 29 (b). Though a similar tendency to *ε*r is observed, the mechanism of *TCf* has not yet been clarified. The *TCf* values of the Sm system are usually negative but close to zero as shown in Fig. 25(c). As *TCf* obeys additional rules, we could easily get a material with *TCf* = 0 ppm/°C. Outstanding materials with *TCf* = 0 ppm/°C have been realized by adding Nd or La to Sm-systems, which are composed of a solid solution with a single phase of *x* = 2/3 [80]. So, *TCf* is improved to near zero ppm/°C without the degradation of the *Q⋅f* value. Usually, as doped materials with different sigh *TCf* located as secondary phase, the *Q⋅f* values are degraded.

**•** Design of outstanding materials based on the crystal structure

In this section, some cases concerning material designs based on the crystal structure are presented. Ohsato *et al.* [72, 81] have researched the crystal structure of microwave materials and clarified the relationship between material properties and crystal structure to aid the design of new outstanding materials.

## **• Case 1: Design by the distribution of cations for the improvement of properties when** *x* **= 0** [82, 83]

Sr ions are introduced in to this system, in which the ionic size is located between Ba and Sm ions. As mentioned above, *Q⋅f* values of Ba6-3*xR*8+2xTi18O56 solid solutions have the maximum value at *x*=2/3. In the region *x* < 2/3, the structural formula of the solid solutions is [*R*8+2*<sup>x</sup>*Ba2-3*<sup>x</sup>*

Fig. 27. Lattice parameters of R6-3xR8+2xTi18O54 (R = La, Pr, Nd and Sm) solid solutions. **Figure 27.** Lattice parameters of *R*6-3*xR*8+2*<sup>x</sup>*Ti18O54 (*R* = La, Pr, Nd and Sm) solid solutions.

 V*x*]*A1*[Ba4]*A*2Ti18O54. In this region, Ba ions located in *A*1-sites result in a deterioration of the quality factor. In the case of *x* = 0, *Q⋅f* values are very low as shown in Fig. 31 (a). When Ba ions are substituted by Sr ions such as in [*R*8Sr2]*A*1[Ba4]*A*2Ti18O54, *Q⋅f* values improved markedly from 206 to 5,880 GHz in the case of *R* = Nd as shown in Fig. 31 (b) [82]. The introduction of Sr ions into *A*1-sites may reduce the internal strain / fluctuation of *d*-spacing, due to the reduction in ionic size in *A*1-sites. Mercurio *et al.* [84] reported that the Sr ions occupy *A*13 special sites (Fig. 23), which have a medium size between that of *A*1- and *A*2-sites. Hence it is expected that *R*, Sr and Ba ions are ordering in *A*1-, *A*13- and *A*2-sites respectively [83]. Fig. 28. Microwave dielectrics properties as a function of ionic radius of R ion. The TCf is also plotted as a function of cell volume in Fig. 29(b). Though a similar tendency to εr is observed, the mechanism of TCf has not yet been clarified. The TCf values of the Sm system are usually negative but close to zero. As TCf obeys additional rules, we could easily get a material with TCf = 0 ppm/oC. Outstanding materials with TCf =0 ppm/oC have been

Fig. 27 Fig. 28

#### **• Case 2: Substituting Sr for Ba in** *A***1-sites when** *x* **= 0.6** [85] realized by adding Nd or La to Sm-systems, which are composed of a solid solution with a

The effects of substituting Sr for Ba in the *A*1-sites of Ba6-3*x*Sm8+2*x*Ti18O54 solid solutions were studied in terms of the lattice parameters and microwave dielectric properties as shown in single phase of x = 2/3 [80]. So, TCf is improved to near zero ppm/oC without the degradation of the Q∙f value. Usually, as doped materials with different sigh TCf located as secondary phase, the Q∙f values are degraded.

∙ Design of outstanding materials based on the crystal structure

**Figure 28.** Microwave dielectrics properties as a function of ionic radius of *R* ion. is expected that R, Sr and Ba ions are ordering in A1-, A13- and A2-sites respectively [83].

Fig. 27 Fig. 28

= 0 [82, 83]

degradation of the Q∙f value. Usually, as doped materials with different sigh TCf located as

Fig. 27. Lattice parameters of R6-3xR8+2xTi18O54 (R = La, Pr, Nd and Sm) solid solutions.

occupy A13 special sites, which have a medium size between that of A1- and A2-sites. Hence it

single phase of x = 2/3 [80]. So, TCf is improved to near zero ppm/oC without the Fig. 29. εr (a) and TCf (b) are shown as a function of unit cell volume. **Figure 29.** *ε*r (a) and *TCf* (b) are shown as a function of unit cell volume.

 V*x*]*A1*[Ba4]*A*2Ti18O54. In this region, Ba ions located in *A*1-sites result in a deterioration of the quality factor. In the case of *x* = 0, *Q⋅f* values are very low as shown in Fig. 31 (a). When Ba ions are substituted by Sr ions such as in [*R*8Sr2]*A*1[Ba4]*A*2Ti18O54, *Q⋅f* values improved markedly from 206 to 5,880 GHz in the case of *R* = Nd as shown in Fig. 31 (b) [82]. The introduction of Sr ions into *A*1-sites may reduce the internal strain / fluctuation of *d*-spacing, due to the reduction in ionic size in *A*1-sites. Mercurio *et al.* [84] reported that the Sr ions occupy *A*13 special sites (Fig. 23), which have a medium size between that of *A*1- and *A*2-sites. Hence it is expected

Fig. 27 Fig. 28

Fig. 27. Lattice parameters of R6-3xR8+2xTi18O54 (R = La, Pr, Nd and Sm) solid solutions.

Fig. 28. Microwave dielectrics properties as a function of ionic radius of R ion.

single phase of x = 2/3 [80]. So, TCf is improved to near zero ppm/oC without the

∙ Design of outstanding materials based on the crystal structure

The effects of substituting Sr for Ba in the *A*1-sites of Ba6-3*x*Sm8+2*x*Ti18O54 solid solutions were studied in terms of the lattice parameters and microwave dielectric properties as shown in

secondary phase, the Q∙f values are degraded.

that *R*, Sr and Ba ions are ordering in *A*1-, *A*13- and *A*2-sites respectively [83].

**Figure 27.** Lattice parameters of *R*6-3*xR*8+2*<sup>x</sup>*Ti18O54 (*R* = La, Pr, Nd and Sm) solid solutions.

310 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**• Case 2: Substituting Sr for Ba in** *A***1-sites when** *x* **= 0.6** [85]

The TCf is also plotted as a function of cell volume in Fig. 29(b). Though a similar tendency to εr is observed, the mechanism of TCf has not yet been clarified. The TCf values of the Sm system are usually negative but close to zero. As TCf obeys additional rules, we could easily get a material with TCf = 0 ppm/oC. Outstanding materials with TCf =0 ppm/oC have been realized by adding Nd or La to Sm-systems, which are composed of a solid solution with a degradation of the Q∙f value. Usually, as doped materials with different sigh TCf located as Figs 32 and 33 respectively [85]. The compositions of the compounds in which Sr is substituted for Ba are as follows: *x* in (Ba1-*α*Sr*α*)6-3*x*Sm8+2*x*Ti18O54 is fixed at 0.6, at which point the Ba6-3*x*Sm8+2*x*Ti18O54 solid solution has excellent properties, and composition *α*, in which Sr is substituted for Ba, ranges from 0.0 to 0.2. The properties are concerned with the strain in the crystal structure due to Sr substitution. We derived a structural formula [Sm8+2*x*Ba2-3*x*]*A*1[Ba4]*A*2Ti18O54, in the range 0 ≤ *x* ≤ 2/3 and another [Sm9.33+2(*x*-2/3)]*A*1[Ba4-3(*x*-2/3)]*A*2Ti18O54 in the range 2/3 ≤ *x* ≤ 1. In the composition *x* = 0.6, the formula is [Sm9.2Ba0.2]*A*1[Ba4]*A*2Ti18O54, which includes four Ba ions in *A*2-sites on the pentagonal columns, secondary phase, the Q∙f values are degraded. ∙ Design of outstanding materials based on the crystal structure

**Figure 30.** Correlationship between dielectric constant and the tilting angle of the octahedral. Fig. 30. Correlationship between dielectric constant and the tilting angle of the octahedral.

Fig. 31(a) Low Q∙f around x = 0 on the Q∙f composition figure. (b) Q∙f values of Nd-system with x = 0 improved from 200 to 6000 GHz by substitution Sr for Ba. **Figure 31.** (a) Low *Q⋅f* around *x* = 0 on the *Q⋅f* composition figure. (b) *Q⋅f* values of Nd-system with *<sup>x</sup>* = 0 improved from 200 to 6000 GHz by substitution Sr for Ba.

and 0.2 Ba ions in *A*1-sites on the perovskite blocks with 9.2 Sm. The 0.2 Ba ions in *A*1-sites produce the internal strain because the size of the Ba ions is fairly large for the *A*1-sites. When 0.2 Ba ions are completely substituted by Sr ions, then the *Q⋅f* values improve to 10,205 GHz, which shows that the strain in the crystal structure has relaxed somewhat. The composition in which Sr is substituted for 0.2 Ba is α = 0.048 in the (Ba1-*α*Sr*α*)6-3*x*Sm8+2*x*Ti18O54 substitutional formula. The dielectric properties depend on the lattice parameters, the values of which change at the composition *α* = 0.048 due to the change in the substitution mode of the Sr ions. The temperature coefficient of the resonate frequency *TCf* was changed in the same manner as the dielectric constant [86].

(a) (b) (c)

**Figure 32.** Lattice parameters of (Ba1-αSr*α*)4.2Sm9.2Ti18O54 solid solutions. Fig. 32. Lattice parameters of (Ba1-<sup>α</sup>Srα)4.2Sm9.2Ti18O54 solid solutions.

Fig. 32. Lattice parameters of (Ba1-<sup>α</sup>Srα)4.2Sm9.2Ti18O54 solid solutions.

∙ Case 2: Substituting Sr for Ba in A1-sites when x = 0.6 [85] solid solutions as a function of composition α. **Figure 33.** Microwave dielectric properties *ε*r (a), *Q*⋅*f* (b) and *TCf* (c) of (Ba1-αSr*α*)4.2Sm9.2Ti18O54 solid solutions as a func‐ tion of composition *α*.

The effects of substituting Sr for Ba in the A1-sites of Ba6-3xSm8+2xTi18O54 solid solutions were

Fig. 33. Microwave dielectric properties εr (a), Q∙f (b) and TCf (c) of (Ba1-<sup>α</sup>Srα)4.2Sm9.2Ti18O<sup>54</sup>

Figs 32 and 33 respectively [85]. The compositions of the compounds in which Sr is

#### *3.2.2. Homologous compounds with perovskite layered structure* studied in terms of the lattice parameters and microwave dielectric properties as shown in The effects of substituting Sr for Ba in the A1-sites of Ba6-3xSm8+2xTi18O54 solid solutions were

∙ Case 2: Substituting Sr for Ba in A1-sites when x = 0.6 [85]

solid solutions as a function of composition α.

There are three kinds of homologous series composed of a perovskite layered structure. The perovskite layers have three different orientations such as (111) the plane series for Ba*n*La4Ti3+*n*O12+3*n*, (100) the series for the Ruddlesden-Popper phase and (110) the series for *AnBn*O3*n*+2 homologous compounds as shown in Tables 2(h) –(j). [87, 88] substituted for Ba are as follows: x in (Ba1-<sup>α</sup>Srα)6-3xSm8+2xTi18O54 is fixed at 0.6, at which point the Ba6-3xSm8+2xTi18O54 solid solution has excellent properties, and composition α, in which Sr is substituted for Ba, ranges from 0.0 to 0.2. The properties are concerned with the strain in the crystal structure due to Sr substitution. We derived a structural formula [Sm8+2xBa2 studied in terms of the lattice parameters and microwave dielectric properties as shown in Figs 32 and 33 respectively [85]. The compositions of the compounds in which Sr is substituted for Ba are as follows: x in (Ba1-<sup>α</sup>Srα)6-3xSm8+2xTi18O54 is fixed at 0.6, at which point the Ba6-3xSm8+2xTi18O54 solid solution has excellent properties, and composition α, in which Sr is substituted for Ba, ranges from 0.0 to 0.2. The properties are concerned with the strain in

#### **• (111) series for Ba***n***La4Ti***3+n***O12+3n homologous compounds** <sup>3</sup>x]A1[Ba4]A2Ti18O54, in the range 0 ≤ x ≤ 2/3 and another [Sm9.33+2(x-2/3)]A1[Ba4-3(x-2/3)]A2Ti18O54 in the range 2/3 ≤ x ≤ 1. In the composition x = 0.6, the formula is [Sm9.2Ba0.2]A1[Ba4]A2Ti18O54, which the crystal structure due to Sr substitution. We derived a structural formula [Sm8+2xBa2- <sup>3</sup>x]A1[Ba4]A2Ti18O54, in the range 0 ≤ x ≤ 2/3 and another [Sm9.33+2(x-2/3)]A1[Ba4-3(x-2/3)]A2Ti18O54 in the

and 0.2 Ba ions in *A*1-sites on the perovskite blocks with 9.2 Sm. The 0.2 Ba ions in *A*1-sites produce the internal strain because the size of the Ba ions is fairly large for the *A*1-sites. When 0.2 Ba ions are completely substituted by Sr ions, then the *Q⋅f* values improve to 10,205 GHz, which shows that the strain in the crystal structure has relaxed somewhat. The composition in which Sr is substituted for 0.2 Ba is α = 0.048 in the (Ba1-*α*Sr*α*)6-3*x*Sm8+2*x*Ti18O54 substitutional formula. The dielectric properties depend on the lattice parameters, the values of which change at the composition *α* = 0.048 due to the change in the substitution mode of the Sr ions. The temperature coefficient of the resonate frequency *TCf* was changed in the same manner as the

Fig. 31(a) Low Q∙f around x = 0 on the Q∙f composition figure. (b) Q∙f values of Nd-system with x = 0 improved from 200 to 6000 GHz by substitution Sr for Ba. **Figure 31.** (a) Low *Q⋅f* around *x* = 0 on the *Q⋅f* composition figure. (b) *Q⋅f* values of Nd-system with *<sup>x</sup>* = 0 improved

(a) (b)

Tilting angle

Tilting angle

Large *R* ion Small *R* ion

Large R ion Small R ion

*c P*

θ

c P

θ

dielectric constant [86].

from 200 to 6000 GHz by substitution Sr for Ba.

*r*∝*P***cos**<sup>θ</sup>

312 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

εr∝Pcos<sup>θ</sup>

**Figure 30.** Correlationship between dielectric constant and the tilting angle of the octahedral.

Fig. 30. Correlationship between dielectric constant and the tilting angle of the octahedral.

The homologous compounds are also perovskite related compounds composed of a layered structure. The compounds are located in a *R*2O3-rich region, compared to pseudo-tungstenbronze solid solutions in the BaO-*R*O3-TiO2 ternary phase diagram as shown in Fig. 22. The chemical formula is shown as Ba*n*La4Ti3+*n*O12+3*n*, and there are four compounds at intervals *n* = 0, 1, 2, and 4 as shown in Fig. 34 [89–92]. As a compound with *n* = 4 is unstable below 1,450 °C, other compounds where *n* = 0, 1 and 2 are studied in this paper [93–97]. In particular, we mainly synthesized two compounds of *n* = 1 and 2. These compounds, *n* = 0: La4Ti3O12 (2:3), *n* = 1: BaLa4Ti4O15 (1:2:4), and *n* = 2: Ba2La4Ti5O18 (2:2:5), show hexagonal layered perovskite struc‐ includes four Ba ions in A2-sites on the pentagonal columns, and 0.2 Ba ions in A1-sites on the perovskite blocks with 9.2 Sm. The 0.2 Ba ions in A1-sites produce the internal strain because the size of the Ba ions is fairly large for the A1-sites. When 0.2 Ba ions are completely substituted by Sr ions, then the Q∙f values improve to 10,205 GHz, which shows that the strain in the crystal structure has relaxed somewhat. The composition in which Sr is substituted for 0.2 Ba is α = 0.048 in the (Ba1-<sup>α</sup>Srα)6-3xSm8+2xTi18O54 substitutional formula. The dielectric properties depend on the lattice parameters, the values of which change at the range 2/3 ≤ x ≤ 1. In the composition x = 0.6, the formula is [Sm9.2Ba0.2]A1[Ba4]A2Ti18O54, which includes four Ba ions in A2-sites on the pentagonal columns, and 0.2 Ba ions in A1-sites on the perovskite blocks with 9.2 Sm. The 0.2 Ba ions in A1-sites produce the internal strain because the size of the Ba ions is fairly large for the A1-sites. When 0.2 Ba ions are completely substituted by Sr ions, then the Q∙f values improve to 10,205 GHz, which shows that the strain in the crystal structure has relaxed somewhat. The composition in which Sr is substituted for 0.2 Ba is α = 0.048 in the (Ba1-<sup>α</sup>Srα)6-3xSm8+2xTi18O54 substitutional formula. The dielectric properties depend on the lattice parameters, the values of which change at the tures as shown in Fig. 35 [89]. Another *R* ion included in this homologous compound is the Nd ion, and the alkali earth ions Ca and Sr, substituted for Ba.

**Figure 34.** Binary phase diagram for Ba*n*La4Ti3+*n*O12+3*<sup>n</sup>* Fig. 34. Binary phase diagram for homologous compounds. homologous compounds.

Fig. 35. Crystal structure of homologous compounds (a) La4Ti3O12 (n = 0), (b) BaLa4Ti4O15 (n = 1), and (c) Ba2La4Ti5O18 (n = 2). **Figure 35.** Crystal structure of Ba*n*La4Ti3+*n*O12+3*n* homologous compounds (a) La4Ti3O12 (*<sup>n</sup>* = 0), (b) BaLa4Ti4O15 (*n* = 1), and (c) Ba2La4Ti5O18 (*n* = 2).

tures as shown in Fig. 35 [89]. Another *R* ion included in this homologous compound is the

Nd ion, and the alkali earth ions Ca and Sr, substituted for Ba.

314 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 34.** Binary phase diagram for Ba*n*La4Ti3+*n*O12+3*<sup>n</sup>* Fig. 34. Binary phase diagram for homologous compounds. homologous compounds.

(a) (b) (c)

and (c) Ba2La4Ti5O18 (*n* = 2).

Fig. 35. Crystal structure of homologous compounds (a) La4Ti3O12 (n = 0), (b) BaLa4Ti4O15 (n = 1), and (c) Ba2La4Ti5O18 (n = 2). **Figure 35.** Crystal structure of Ba*n*La4Ti3+*n*O12+3*n* homologous compounds (a) La4Ti3O12 (*<sup>n</sup>* = 0), (b) BaLa4Ti4O15 (*n* = 1),

**Figure 36.** (a) Closed packing layer composed by oxygen and La/Ba atoms, which is one of specific features, (b) Junc‐ tion slab composed by octahedron. Cleavage easily occurs at the center octahedron without cation.

The crystal structure of homologous BaLa4Ti4O15 (*n* = 1) compounds is illustrated in Fig. 35(b). The crystal data are trigonal, *P3c*1 (No. 165) and *Z* = 2. This compound has a sequence of five layers with La/BaO3 closed packing as shown in Fig. 36 (a), like *hccch* of the Ba5Nb4O15 type. Here, Ba is the cation located in the cuboctahedron of the perovskite structure, where *h* means hexagonal close packing (*hcp*) and *c* is cubic close packing (*ccp*). The perovskite slab with *ccp* is composed of four TiO6 octahedral layers and the junction slab with *hcp* between perovskite slabs is composed of a three-octahedron string shared face with an empty octahedron in the center as shown in Fig. 36 (b).

**Figure 37.** Alkaline earth elements in the *A*1, *A*2, and *A*3-sites in the closed packing layer of BaLa4Ti4O15 (a) and CaLa4Ti4O15 (b) viewed along (210). Ba ions with a large ionic radius are located only in *A*1-sites of Ba analogues, while on the other hand, Ca ions of Ca analogues are distributed across all sites with La ions.

analogues randomly occupy the A-sites [96].

composition n.

are distributed across all sites with La ions.

The compound can include Sr and Ca ions substituted in place of Ba ions. The homologous ALa4Ti4O15 (A = Ba, Sr, or Ca) compounds have a different ordering for A-site cations as shown in Fig. 37. In the case of Ba analogues, the ordering of Ba (r = 1.61 Å) and La ions (r = 1.36 Å) will occur as follows: Ba ions located in A1-sites near the junction slab, and La ions in all A1, A2 and A3-sites as shown in Fig. 37(a) [96]. In the case of Sr and Ca analogues, the Sr ions (r = 1.44 Å) and the Ca ions (r = 1.34 Å) are all located in A-sites including La ions as shown in Fig. 37(b). As the space of A1-sites is larger than those of A2- and A3-sites, Ba2+ ions with their large ionic radii predominantly occupy A1-sites. On the other hand, as the ionic

Fig. 38. Microwave dielectric properties of BanLa4Ti3+nO12+3n ceramics as a function of the **Figure 38.** Microwave dielectric properties of Ba*n*La4Ti3+*n*O12+3*n* ceramics as a function of the composition *n*.


Ca 1550<sup>o</sup> C 24 h 94.8 41.1 50,246 -25.5 **Table 4.** Relative density and microwave dielectric properties of *A*La4Ti4O15 (*A*=Ba, Sr and Ca).

The compound can include Sr and Ca ions substituted in place of Ba ions. The homologous *A*La4Ti4O15 (*A* = Ba, Sr, or Ca) compounds have a different ordering for *A*-site cations as shown in Fig. 37. In the case of Ba analogues, the ordering of Ba (*r* = 1.61 Å) and La ions (*r* = 1.36 Å) will occur as follows: Ba ions located in *A*1-sites near the junction slab, and La ions in all *A*1, *A*<sup>2</sup> and *A*3-sites as shown in Fig. 37 (a) [96]. In the case of Sr and Ca analogues, the Sr ions (*r* = 1.44 Å) and the Ca ions (*r* = 1.34 Å) are all located in *A*-sites including La ions as shown in Fig. 37(b). As the space of *A*1-sites is larger than those of *A*2- and *A*3-sites, Ba2+ ions with their large ionic radii predominantly occupy *A*1-sites. On the other hand, as the ionic radii of Sr and Ca2+ are close to that of La3+ ions, the Sr and the Ca ions of the Sr and Ca analogues randomly occupy the *A*-sites [96]. The microwave dielectric properties of BanLa4Ti3+nO12+3n are shown in Fig. 38 as functions of composition [93]. The sample with the composition n = 1 shows the best properties, such as the highest Q∙f = 46,000 GHz, εr = 46, and TCf = −11 ppm/oC, which can be improved to near zero ppm/oC by means of the substitution of Al for La [94]. The microwave dielectric properties of the Ba, Sr, and Ca analogue ceramics are shown in Table 4 [97]. These samples showed excellent microwave dielectric properties for use in base stations of mobile phones, such as a value of εr greater than 40, a Q∙f greater than 40,000 GHz and a TCf within −30

The microwave dielectric properties of Ba*n*La4Ti3+*n*O12+3*n* are shown in Fig. 38 as functions of composition [93]. The sample with the composition *n* = 1 shows the best properties, such as the highest *Q⋅f* = 46,000 GHz, *ε*<sup>r</sup> = 46, and *TCf* = −11 ppm/°C, which can be improved to near zero ppm/°C by means of the substitution of *A*l for La [94]. The microwave dielectric properties of the Ba, Sr, and Ca analogue ceramics are shown in Table 4 [97]. These samples showed excellent microwave dielectric properties for use in base stations of mobile phones, such as a value of *ε*<sup>r</sup> greater than 40, a *Q⋅f* greater than 40,000 GHz and a *TCf* within ± 30 ppm/°C. The highest *ε*<sup>r</sup> of 44.4 was observed in the case of the Ba analogue and the value decreased to 41.1 for Ca. The highest *Q⋅f* of 50,246 GHz was observed in the case of the Ca analogue, and the value decreased to 46,220 GHz for Sr and to 41,008 GHz for Ba. These values are much higher than those in an earlier report.

We would like to consider the reason for the large *ε*r and the high *Q⋅f* based on the crystal structure. There are three characteristic points of the crystal structure: one is the size of the cation sites, another is the shift of the cation positions, and the third is the different divalent cation distributions. The large volume of cation sites results in the large *ε*r. Divalent cation *A* with a large ionic radius, such as Ba, Ca and Sr, expands the lattice and brings an enlargement of cation sites. In particular, the expansion of the *B*-site volume affects Ti ion movement as a result of the rattling effect. In the case of the Ba analogue with the highest *ε*<sup>r</sup> value, the volume is larger than that of the Ca analogue. The Ba ion with its large ionic radius of 1.61 Å is more effective than the Ca ion with *r* = 1.34 Å. In the second case of the shift of the cation positions, the *ε*r of the Ba analogue with a large shift is larger than that of the Ca analogue. This shift might increase the movability of the La ion with a small ionic radius, so that the *ε*<sup>r</sup> of the Ba analogue with a large shift is greater than that of the Ca analogue. The high *Q⋅f* value might come from the cation distribution and the volume of cation sites. In the case of the Ca analogue, as the shift of cations from the close packed layer of oxygen is smaller than it is for Ba analogue, then widely occupied *A*-sites might be distributed periodically with La ions to bring a high *Q⋅f*. The *ε*<sup>r</sup> values also depend on the ionic polarizations of Ba (*P* = 6.4), Sr (*P* = 4.24), and Ca (*P* = 3.16) [97]. These homologous compounds show characteristic near-zero ppm/°C values of the *TCf*. The *TCf* of the Sr analogue is near zero ppm/°C compared with that of the others, whose might come from the analogue of SrTiO3 having a large positive *TCf* of 1,200 ppm/°C. Moreover, the *TCf* of the Ba analogue was improved to a near zero 1.3 ppm/°C, with a high *ε*<sup>r</sup> of 44 and a *Q⋅f* of 47,000 GHz by substituting Al ions for La ions [97].

## **• (100) series for Ruddlesden-Popper phase**

The compound can include Sr and Ca ions substituted in place of Ba ions. The homologous *A*La4Ti4O15 (*A* = Ba, Sr, or Ca) compounds have a different ordering for *A*-site cations as shown in Fig. 37. In the case of Ba analogues, the ordering of Ba (*r* = 1.61 Å) and La ions (*r* = 1.36 Å) will occur as follows: Ba ions located in *A*1-sites near the junction slab, and La ions in all *A*1, *A*<sup>2</sup> and *A*3-sites as shown in Fig. 37 (a) [96]. In the case of Sr and Ca analogues, the Sr ions (*r* = 1.44 Å) and the Ca ions (*r* = 1.34 Å) are all located in *A*-sites including La ions as shown in Fig. 37(b). As the space of *A*1-sites is larger than those of *A*2- and *A*3-sites, Ba2+ ions with their large ionic radii predominantly occupy *A*1-sites. On the other hand, as the ionic radii of Sr and Ca2+ are close to that of La3+ ions, the Sr and the Ca ions of the Sr and Ca analogues randomly

C 2 h 98.4 44.4 41,008 -26

C 48 h 98.9 43.7 46,220 -8.4

C 24 h 94.8 41.1 50,246 -25.5

The microwave dielectric properties of BanLa4Ti3+nO12+3n are shown in Fig. 38 as functions of composition [93]. The sample with the composition n = 1 shows the best properties, such as the highest Q∙f = 46,000 GHz, εr = 46, and TCf = −11 ppm/oC, which can be improved to near zero ppm/oC by means of the substitution of Al for La [94]. The microwave dielectric properties of the Ba, Sr, and Ca analogue ceramics are shown in Table 4 [97]. These samples showed excellent microwave dielectric properties for use in base stations of mobile phones, such as a value of εr greater than 40, a Q∙f greater than 40,000 GHz and a TCf within −30

(a) (b) (c) Fig. 38. Microwave dielectric properties of BanLa4Ti3+nO12+3n ceramics as a function of the

Table 4. Relative density and microwave dielectric properties of ALa4Ti4O15 (A=Ba, Sr and Ca).

Ba 1600°C 2 h 98.4 44.4 41,008 -26 Sr 1550°C 48 h 98.9 43.7 46,220 -8.4 Ca 1550°C 24 h 94.8 41.1 50,246 -25.5

*A* **Sintering condition** *D***r (%)** *ε***<sup>r</sup>** *Q∙f* **(GHz)** *TCf* **(ppm/**°**C)**

**Figure 38.** Microwave dielectric properties of Ba*n*La4Ti3+*n*O12+3*n* ceramics as a function of the composition *n*.

<sup>A</sup> Sintering condition <sup>D</sup>r (%) <sup>ε</sup><sup>r</sup> Q·f (GHz) TCf

**Table 4.** Relative density and microwave dielectric properties of *A*La4Ti4O15 (*A*=Ba, Sr and Ca).

(ppm/<sup>o</sup> C)

The compound can include Sr and Ca ions substituted in place of Ba ions. The homologous ALa4Ti4O15 (A = Ba, Sr, or Ca) compounds have a different ordering for A-site cations as shown in Fig. 37. In the case of Ba analogues, the ordering of Ba (r = 1.61 Å) and La ions (r = 1.36 Å) will occur as follows: Ba ions located in A1-sites near the junction slab, and La ions in all A1, A2 and A3-sites as shown in Fig. 37(a) [96]. In the case of Sr and Ca analogues, the Sr ions (r = 1.44 Å) and the Ca ions (r = 1.34 Å) are all located in A-sites including La ions as shown in Fig. 37(b). As the space of A1-sites is larger than those of A2- and A3-sites, Ba2+ ions with their large ionic radii predominantly occupy A1-sites. On the other hand, as the ionic radii of Sr and Ca2+ are close to that of La3+ ions, the Sr and the Ca ions of the Sr and Ca

The microwave dielectric properties of Ba*n*La4Ti3+*n*O12+3*n* are shown in Fig. 38 as functions of composition [93]. The sample with the composition *n* = 1 shows the best properties, such as the highest *Q⋅f* = 46,000 GHz, *ε*<sup>r</sup> = 46, and *TCf* = −11 ppm/°C, which can be improved to near zero ppm/°C by means of the substitution of *A*l for La [94]. The microwave dielectric properties of the Ba, Sr, and Ca analogue ceramics are shown in Table 4 [97]. These samples showed excellent microwave dielectric properties for use in base stations of mobile phones, such as a value of *ε*<sup>r</sup> greater than 40, a *Q⋅f* greater than 40,000 GHz and a *TCf* within ± 30 ppm/°C. The highest *ε*<sup>r</sup> of 44.4 was observed in the case of the Ba analogue and the value decreased to 41.1 for Ca. The highest *Q⋅f* of 50,246 GHz was observed in the case of the Ca analogue, and the value decreased to 46,220 GHz for Sr and to 41,008 GHz for Ba. These values are much higher

occupy the *A*-sites [96].

composition n.

Ba 1600<sup>o</sup>

Sr 1550<sup>o</sup>

Ca 1550<sup>o</sup>

are distributed across all sites with La ions.

analogues randomly occupy the A-sites [96].

316 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

than those in an earlier report.

The Ruddlesden-Popper phase [98, 99] is shown with the chemical formula *An*+1*Bn*O3*n*+1. Here, *A* is a cuboctahedral site and B is an octahedral site. These space groups are all *I*4/*mmm* (No. 139) with a center of symmetry *i*. The phase is composed of SrTiO3 layers with (100)cubic plane as shown in Fig. 39(a). The layers are stacked, shifting each other by (1/2*a*, 1/2*b*). Among the layers, a SrO halite type structure is formed. The chemical formula is shown as *A*O⋅*n*(*AB*O3). Here, *n* = 1: Sr2TiO4 [98] is K2NiF4 type stacking of one *AB*O3 and three unit cells of perovskite; *n* = 2: Sr3Ti2O7 is the stacking of two *AB*O3 similar to *n* = 1, *n* = 3: Sr4Ti3O10 is also the stacking of three *AB*O3 with (100). CaSmAlO4 ceramics with a K2NiF4 crystal structure are presented as microwave dielectrics [100]. The microwave dielectric properties are shown in Table 2 (i). Although the Auriviellius phase A*n*+1B*n*O3*<sup>n</sup>*+3 [101, 102] and the Dion-Jacobson phase *M*<sup>+</sup> [Ca2Nan-3Nb*n*O3*n*+1] exist in addition to the Ruddlesden-Popper phase in the (100) series [103,104], the microwave dielectric properties are not presented.

## **• (110) series for** *AnBn***O***3n+2* **homologous compounds**

A (110) planar perovskite layered structure also creates a homologous series of *AnBn*O3*n*+2. A typical example of this series is Sr2Ta2O7 with *n* = 4 as shown in Fig. 39 (b) [105, 106]. The value of *n* is that of the octahedral layer in the single perovskite layer as seen in the figures. The layered structure was formed by the addition of oxygen atoms to non-bridged oxygen atoms of the *B*O6 octahedron cut. The microwave dielectric properties are also shown in Table 2 (j).

Fig. 39. Crystal structure of layered perovskite: (a) (100) series: Ruddlesden-Popper phase of Sr3Ti2O7 (n = 2), and (b) (110) series: AnBnO3n+2 homologous compound of Sr2Ta2O7 (n = 4). **Figure 39.** Crystal structure of layered perovskite: (a) (100) series: Ruddlesden-Popper phase of Sr3Ti2O7 (*n* = 2), and (b) (110) series: *AnBn*O3*n*+2 homologous compound of Sr2Ta2O7 (*n* = 4).

value of n is that of the octahedral layer in the single perovskite layer as seen in the figures.

#### **• Application of perovskite related compounds** A (110) planar perovskite layered structure also creates a homologous series of AnBnO3n+2. A typical example of this series is Sr2Ta2O7 with n = 4 as shown in Fig. 39(b) [105,106]. The

∙ (110) series for AnBnO<sup>3</sup>n+2 homologous compounds

These perovskite related compounds are applied to microwave dielectrics. Dr. Okawa studied pseudo-tungsten-bronze solid solutions and (111) plane homologous series in his doctoral thesis [108], and these materials have been used in microwave applications. Microwave dielectrics based on Ba4(*R*1-*<sup>y</sup>*Bi*y*)9+1/3Ti18O54 (*R*: rare earth) solid solutions with Bi partially substituted for *R* were clarified with a high *ε*r of > 80 [109,110]. They are used widely in the wireless communication systems of fire engines in Japan. An Al doped BaLa4Ti4O15 homolo‐ gous compound, which has the best microwave dielectric properties for use as a resonator was developed for use in the base stations of wireless communication systems [94]. It is currently used in base stations of the mobile communication systems of the Tokyo metro. In addition, at near *n* = 0 in Ba*n*La4Ti3+*n*O12+3*<sup>n</sup>* homologous compounds, a superior material was developed with properties of *ε*r = 42, *Q⋅f* = 86,000 GHz, and *TCf* = -17 ppm/°C [111, 112]. The layered structure was formed by the addition of oxygen atoms to non-bridged oxygen atoms of the BO6 octahedron cut. The microwave dielectric properties are also shown in Table 2 (j). ∙ Application of perovskite related compounds These perovskite related compounds are applied to microwave dielectrics. Dr. Okawa studied pseudo-tungsten-bronze solid solutions and (111) plane homologous series in his doctoral thesis [108], and these materials have been used in microwave applications. Microwave dielectrics based on Ba4(R1-yBiy)9+1/3Ti18O54 (R: rare earth) solid solutions with Bi partially substituted for R were clarified with a high εr of > 80 [109,110]. They are used

## **4. Flexibility of perovskite structure and microwave dielectric properties**

As already described in many parts of this book, the structure of perovskite is flexible, producing many specific phases such as ferroelectrics and paraelectrics. This flexibility is due to the inclusion of many cations in the perovskite structure. There are three important features of the perovskite structure, detailed below :


**•** The third important feature is the large cation site, that is, cuboctahedron with basically 12 coordination.

The first produces two spaces for cations — octahedron and cuboctahedron — described as the third feature above, which many kinds of cations with different ionic radii and electric charges can occupy. This framework will be deformed and tilting. These features produce many kinds of ferroelectric and paraelectric properties. The closed packing of *A*O3 discussed as the second feature of the perovskite structure, produces a high density, as heavy *A* ions are incorporated in the layer instead of oxygen. These high density and heavy materials are found in high pressure environments such as the deeper parts of the earth [113]. The cuboctahedron points are produced by connecting the *A*O3 packing layer with *B* ions. This polyhedron includes special large cations such as Ba, Ca and Sr. If smaller sized ions such as Mg are present, the crystal structure formed will be that of ilmenite, similar to the Al2O3-type.

Furthermore, the flexibility of the perovskite structure is shown by the fact that different sized plural large cations can be also included in the deformed crystal structure. There are some examples as follows:

**•** Example 1: In the case of the order-disorder transition in complex perovskite, when the structure is ordered, two different sites appear under the changing crystal structure form from cubic to trigonal [38] as shown in Fig. 11.

**• Application of perovskite related compounds**

(110) series: *AnBn*O3*n*+2 homologous compound of Sr2Ta2O7 (*n* = 4).

318 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Table 2 (j).

of the perovskite structure, detailed below :

dimensions.

in Fig. 36 (a).

These perovskite related compounds are applied to microwave dielectrics. Dr. Okawa studied pseudo-tungsten-bronze solid solutions and (111) plane homologous series in his doctoral thesis [108], and these materials have been used in microwave applications. Microwave dielectrics based on Ba4(*R*1-*<sup>y</sup>*Bi*y*)9+1/3Ti18O54 (*R*: rare earth) solid solutions with Bi partially substituted for *R* were clarified with a high *ε*r of > 80 [109,110]. They are used widely in the wireless communication systems of fire engines in Japan. An Al doped BaLa4Ti4O15 homolo‐ gous compound, which has the best microwave dielectric properties for use as a resonator was developed for use in the base stations of wireless communication systems [94]. It is currently used in base stations of the mobile communication systems of the Tokyo metro. In addition, at near *n* = 0 in Ba*n*La4Ti3+*n*O12+3*<sup>n</sup>* homologous compounds, a superior material was developed

These perovskite related compounds are applied to microwave dielectrics. Dr. Okawa studied pseudo-tungsten-bronze solid solutions and (111) plane homologous series in his doctoral thesis [108], and these materials have been used in microwave applications. Microwave dielectrics based on Ba4(R1-yBiy)9+1/3Ti18O54 (R: rare earth) solid solutions with Bi partially substituted for R were clarified with a high εr of > 80 [109,110]. They are used

A (110) planar perovskite layered structure also creates a homologous series of AnBnO3n+2. A typical example of this series is Sr2Ta2O7 with n = 4 as shown in Fig. 39(b) [105,106]. The value of n is that of the octahedral layer in the single perovskite layer as seen in the figures. The layered structure was formed by the addition of oxygen atoms to non-bridged oxygen atoms of the BO6 octahedron cut. The microwave dielectric properties are also shown in

**Figure 39.** Crystal structure of layered perovskite: (a) (100) series: Ruddlesden-Popper phase of Sr3Ti2O7 (*n* = 2), and (b)

(a) (b) Fig. 39. Crystal structure of layered perovskite: (a) (100) series: Ruddlesden-Popper phase of Sr3Ti2O7 (n = 2), and (b) (110) series: AnBnO3n+2 homologous compound of Sr2Ta2O7 (n = 4).

**4. Flexibility of perovskite structure and microwave dielectric properties**

As already described in many parts of this book, the structure of perovskite is flexible, producing many specific phases such as ferroelectrics and paraelectrics. This flexibility is due to the inclusion of many cations in the perovskite structure. There are three important features

**•** The first is the framework of octahedra connecting all their apexes with each other in three

**•** The second is the closed packing layer of *A*O3 instead of oxygen closed packing as shown

with properties of *ε*r = 42, *Q⋅f* = 86,000 GHz, and *TCf* = -17 ppm/°C [111, 112].

∙ (110) series for AnBnO<sup>3</sup>n+2 homologous compounds

∙ Application of perovskite related compounds


**Figure 40.** The structure of perovskite changes to a tungsten-bronze structure after the inclusion of two differently sized large ions, producing rhombic (*A*1) and pentagonal (*A*2) sites.

**Figure 41.** Resonant coupling for the temperature measurement of isolated places.

The intrinsic reasons for the ordering and symmetry effects on *Q⋅f* properties are that ordering reduces the internal stress and high symmetry reduces the formation of poles. Which effect is predominant ? As described above, in the absence of phase transition such as in pseudotungsten-bronze solid solutions, compositional ordering is predominant [74]. In the case of complex perovskite with an order-disorder transition, high symmetry is predominant rather than ordering, as described above [58, 65].

## **5. Functional advances in the next generation of microwave dielectrics**

In this section, the future large scale application of microwave dielectrics will utilize some new and novel functions — based on microwave properties — as follows:


As the content has been published in both "*A Handbook of Mutifunctional Ceramics*" [17] and a paper [115], please refer to those publications. Some important functions are presented below.

Fig. 41 shows a type of temperature sensor utilizing resonant coupling, which can measure the temperature on the opposite side of a wall without the need for an electric wire. The materials should have an extremely large *TCf* depending on the temperature. Fig. 42 shows a well-known

**Figure 43.** Transmissivity of Lumicera compared with quartz glass. Transparent region expanded to middle IR region.

principle of an electromagnetic wave absorber in a good design using wave interference, dielectric losses and wave retardation in the materials [116]. As in this example, new functions will be derived from the properties of the materials, and from physical principles. Fig. 43 shows the extreme transparency ceramics of Lumicera produced by Murata Manufacturing Co., Ltd. in Japan [117]. These materials are also microwave dielectrics, such as BMT — the 'king' of microwave dielectrics — as described above. The fabrication technology takes full advantage of our current understanding of the materials, such as controlled to cubic phase without birefringence.

## **6. Conclusion**

The intrinsic reasons for the ordering and symmetry effects on *Q⋅f* properties are that ordering reduces the internal stress and high symmetry reduces the formation of poles. Which effect is predominant ? As described above, in the absence of phase transition such as in pseudotungsten-bronze solid solutions, compositional ordering is predominant [74]. In the case of complex perovskite with an order-disorder transition, high symmetry is predominant rather

**5. Functional advances in the next generation of microwave dielectrics**

and novel functions — based on microwave properties — as follows:

**Figure 41.** Resonant coupling for the temperature measurement of isolated places.

320 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

In this section, the future large scale application of microwave dielectrics will utilize some new

As the content has been published in both "*A Handbook of Mutifunctional Ceramics*" [17] and a paper [115], please refer to those publications. Some important functions are presented below. Fig. 41 shows a type of temperature sensor utilizing resonant coupling, which can measure the temperature on the opposite side of a wall without the need for an electric wire. The materials should have an extremely large *TCf* depending on the temperature. Fig. 42 shows a well-known

than ordering, as described above [58, 65].

(1) electromagnetic resonance,

(3) electromagnetic wave delay,

(2) electromagnetic wave shortening,

(5) electromagnetic wave absorption

(4) temperature variation of resonant frequency,

(6) other functions such as transparency and refractive index.

Many kinds of microwave dielectrics with perovskite and related structure have been pro‐ duced based on research into the relationship between the crystal structure and its properties. In this chapter, the following compounds related perovskite are introduced and discussed: simple perovskite, complex perovskite, pseudo-tungsten-bronze solid solutions and layered perovskite compounds such as Ba*n*La4Ti3+*n*O12+3*<sup>n</sup>* homologous with (111), Ruddlesden-Popper phase with (100) and *AnBn*O3*n*+2 homologous with (110) oriented perovskite layers. Most of these compounds are paraelectrics with a center of symmetry *i*, and include many types of ion with different ionic radii and electric charges. They are also designed using stoichiometric techni‐ ques to develop superior properties. The superior microwave materials developed should be utilized in new and useful applications for the benefit of future generations.

## **Acknowledgements**

I would like to thank Professors and graduate students of NIT, Meijo University and Hoseo University, and Doctors and researchers in the many companies which collaborated with NIT. A part of this work was supported by the following projects: (1) Support industries of Japan by Ministry of Economy, Trade and Industry (METI), Japan. (2) MEXT/JSPS KAKENHI Grant Number 25420721. (3) Adaptable & Seamless Technology Transfer Program (A-step) by MEXT, Japan

## **Author details**

Hitoshi Ohsato1,2\*

Address all correspondence to: ohsato.hitoshi@nitech.ac.jp

1 Nagoya Industrial Science Research Institute, Nagoya, Japan

2 Nagoya Institute of Technology, Nagoya, Japan

## **References**


[6] Wakino K. Recent Development of Dielectric Resonator Materials and Filters in Ja‐ pan. Ferroelectrics. 1989;91:69-86.

compounds are paraelectrics with a center of symmetry *i*, and include many types of ion with different ionic radii and electric charges. They are also designed using stoichiometric techni‐ ques to develop superior properties. The superior microwave materials developed should be

I would like to thank Professors and graduate students of NIT, Meijo University and Hoseo University, and Doctors and researchers in the many companies which collaborated with NIT. A part of this work was supported by the following projects: (1) Support industries of Japan by Ministry of Economy, Trade and Industry (METI), Japan. (2) MEXT/JSPS KAKENHI Grant Number 25420721. (3) Adaptable & Seamless Technology Transfer Program (A-step) by MEXT,

[1] Sebastian MT. Dielectric Materials for Wireless Communication. Amsterdam: Elsevi‐

[2] Sebastian MT, Ubic R, Jantunen H. Low-loss dielectric ceramic materials and their

[3] Ohsato H. High frequency dielectric ceramics. in: Adachi G editor. Materials Tech‐ nology Hand Book for Rare-earth Elements. Tokyo: NTS Inc; 2008. p. 346–358. (Japa‐

[4] Ohsato H, Kagomiya I, Chae KW. Microwave dielectric ceramics with rare-earth ele‐

[5] Ohsato H, Kagomiya I, Kim JS. Microwave Dielectric Ceramics with Rare-Earth (II).

properties. International Materials Review. 2015;60:395-415.

utilized in new and useful applications for the benefit of future generations.

322 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Address all correspondence to: ohsato.hitoshi@nitech.ac.jp

2 Nagoya Institute of Technology, Nagoya, Japan

er. 2008; ISBN-13:978-0-08-045330-9

ments (I). J Korean Phys Soc. 2012;61:971-979.

Integrated Ferroelectrics. 2010;115:95-109.

1 Nagoya Industrial Science Research Institute, Nagoya, Japan

**Acknowledgements**

Japan

**Author details**

Hitoshi Ohsato1,2\*

**References**

nese).


[35] Inagaki Y, Ishizawa N, Ohsato H, Kagomiya I, Kakimoto K, Shimada T. Structure of Sr and Ti codoped LaAlO3 perovskite. In: XXI Congress of the International Union of Crystallography Congress and General Assembly. 2008; P11.11.42(C520).

[21] Tsunooka T, Sugiyama H, Kakimoto K, Ohsato H, Ogawa H. Zero Temperature Co‐

[22] Guo Y, Ohsato H, Kakimoto K. Characterization and dielectric behavior of willemite and TiO2-doped willemite ceramics at millimeter-wave frequency. J Eur Ceram Soc.

[23] Surendran KP, Santha N, Mohanan P, Sebastian M. Temperature stable low loss ce‐ ramic dielectrics in (1-*x*) ZnAl2O4-xTiO2 system for microwave substrate applications.

[24] Ohsato H, Kato K, Mizuta M, Nishigaki S, Okuda T. Microwave Dielectric Properties of the Ba6-3X (Sm1-Y, *R*Y)8+2XTi18O54 (*R* = Nd and La) Solid Solutions with Zero Tempera‐ ture Coefficient of the Resonant Frequency. Jpn J Appl Phys. 1995;34(9B):5413-5417.

[25] Ball CJ, Begg BD, Cookson DJ, Thorogood GJ, Vance ER. Structures in the System

[26] Huang CH, Pana CL, Shim SJ. Liquid phase sintering of MgTiO3-CaTiO3 microwave

[27] Yao GG, Liu P. Low temperature sintering and microwave dielectric properties of (1-

[28] Huang CH, Pana CL, Leeb WC. Microwave dielectric properties of mixtures of glassforming oxides Zn–B–Si and dielectric ceramics MgTiO3–CaTiO3 for LTCC applica‐

[29] Cho SY, Kim IT, Hong KS. Microwave dielectric properties and applications of rare

[30] Cho SY, Kim CH, Kim DW, Hongs KS, Kim JH. Dielectric properties of Ln(Mg1/2Ti1/2)O3 as substrates for high-Tc superconductor thin films. J Mater Res.

[31] Geller S, Bala VB. Crystallographic studies of perovskite-like compounds. II. Rare

[32] Inagaki Y, Suzuki S, Kagomiya I, Kakimoto K, Ohsato H, Sasaki K, Kuroda K, Shima‐ da T. Crystal structure and microwave dielectric properties of SrTiO3 doped LaALO3

[33] Cho SY, Hong KS, Ko KH. Mixture-like behavior in the microwave dielectric proper‐

[34] Shimada T, Kura K, Ohtsuki S. Dielectric properties and far infrared reflectivity of lanthanum aluminate–strontium titanate ceramics. J Euro Ceram Soc. 2006;26:2017–

ties of the (1−x)LaAlO3–*x*SrTiO3 system. Mater Res Bull. 1999;34:511–516.

single crystal grown by FZ. J Euro Ceram Soc. 2007;27:2861-2864.

CaTiO3/SrTiO3. J Solid State Chem. 1998;139:283-247.

dielectric ceramics. Mater Chem Phys. 2003;78:111-115.

tions. J Alloys & Compounds. 2008;462:5-8.

earth aluminates. J Mater Res. 1999;14:114-119.

earth aluminates. Acta Cryst. 1956;9:1019-1025.

*x*) Mg4Nb2O9-*x*CaTiO3 ceramics. Physica. 2010;B 405:547-551.

and Sinterability of Forsterite Ceramics by Rutile Addition. J Ceram Soc

efficient *τ*<sup>f</sup>

2006;26:1827-1830.

1999;14:2484-2487.

2021.

Jpn Suppl. 2004;112:S1637-S1640.

324 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Eur Phys J B. 2004;41:301–306.


[63] Izumi F, Ikeda T. A Rietveld-analysis program RIETAN-98 and its applications to zeolites. Mater Sci Forum. January 2000;321-324:198-203.

[50] Kim ES,Yoon KH. Microwave Dielectric Properties of Complex Perovskite

[51] Lu CH, Tsai CC. Reaction kinetics, sintering characteristics, and ordering behavior of

[52] Kawashima S, Nishida M, Ueda I, Ouchi H. Ba(Zn1/3Ta2/3)O3 Ceramics with Low Die‐

[53] Yokotani Y, Tsuruta T, Okuyama K, Kugimiya K. Low-Dielectric Loss Ceramics for

[54] Matsumoto H, Tamura H, Wakino K. Ba(Mg, Ta)O3-BaSnO3 High-*Q* Dielectric Reso‐

[55] Koga E, Moriwake H. Effects of Superlattice Ordering and Ceramic Microstructure on the Microwave *Q* Factor of Complex Perovskite-Type Oxide Ba(Zn1/3Ta2/3)O3. J Ce‐

[56] Koga E, Moriwake H, Kakimoto K, Ohsato H. Influence of Composition Deviation from Stoichiometric Ba(Zn1/3Ta2/3)O3 on Superlattice Ordering and Microwave Quali‐

[57] Koga E, Mori H, Kakimoto K, Ohsato H. Synthesis of Disordered Ba(Zn1/3Ta2/3)O3 by Spark Plasma Sintering and Its Microwave *Q* Factor. Jpn J Appl Phys. 2006;45(9B):

[58] Koga E, Yamagishi Y, Moriwake H, Kakimoto K, Ohsato H. Order-disorder transi‐ tion and its effect on Microwave quality factor *Q* in Ba(Zn1/3Nb2/3)O3 system. J Elec‐

[59] Koga E, Yamagishi Y, Moriwake H, Kakimoto K, Ohsato H. Large *Q* factor variation within dense, highly ordered Ba(Zn1/3Ta2/3)O3 system. J Euro Ceram Soc.

[60] Kugimiya K. Crystallographic study on the *Q* of Ba(Mg1/3Ta2/3)O3 dielectrics. In:Ab‐ stract for Kansai branch academic meeting. 5 September 2003; Senri-Life Science. B-20. In: Abstract for Meeting of Microwave/Millimeterwave Dielectrics and Related Materials on the Ceramic Soc. 2004; Nagoya Institute of Technology:Nagoya Japan.

[61] Surendran KP, Sebastian MT, Mohanan P, Moreira RL, Dias A. Effect of Nonstoichi‐ ometry on the Structure and Microwave Dielectric Properties of Ba(Mg0.33Ta0.67)O3.

[62] Koga E, Moriwake H, Kakimoto K, Ohsato H. Raman Spectroscopic Evaluation and Microwave Dielectric Property of Order/Disorder and Stoichiometric/Non-Stoichio‐

metric Ba(Zn1/3Ta2/3)O3. Ferroelectrics. 2007;356:146–152.

lectric Loss at Microwave Frequencies. J Am Ceram Soc. 1983;66:421-423.

Microwave Uses. National Technical Report. 1994;40:11-16. (Japanese).

Ba(Mg1/3Ta2/3)O3. Ferroelect. 1992;133:1187.

326 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

microwave. J Mater Res. 1996;11:1219-1227.

nator. Jpn J Appl Phys. 1991;30:2347-2349.

ram Soc Jpn. 2003;111:767-775 (Japanese).

7484-7488.

troceram. 2006;17:375-379.

Chem Mater. 2005;17:142-151.

2006;26:1961-1964.

(Japanese).

ty Factor *Q*. J Ceram Soc Jpn. 2005;113: 172-178 (Japanese).


Morawiec H, Stroz D, editors. Applied Crystallography. World Scientific Publishing; 1998: p. 440-447.


[89] Trolliard G, Harre N, Mercurio D, Frit B. Cation-deficient perovskite-related (Ba, La)*n*Tin-δ O3*<sup>n</sup>* (*n* ≥ 4δ) microphases in the La4Ti3O12-BaTiO3 system: An HRTEM ap‐ proach. J Solid State Chem. 1999;145.678-693.

Morawiec H, Stroz D, editors. Applied Crystallography. World Scientific Publishing;

[77] Ohsato H, Imaeda M, Komura A, Okuda T. Non-Linear Microwave Quality Factor Change Based on the Site Occupancy of Cations on the Tungstenbronze-type Ba6-3x*R*8+2xTi18O54 (*R* = Rare Earth) Solid Solutions. In: K. M. Nair KM, Bhalla AS, edi‐ tors. Dielectric Ceramic Materials. In: Ceramic Transaction. John Wiley & Sons;

[78] Valant M, Sovorov D, Kolar D. X-Ray Investigations and Determination of Dielectric Properties of the Compound Ba4.5Gd9Ti18O54. Jpn J Appl Phys. 1996;35:144-150.

[79] Shannon RD. Dielectric polarizabilities of ions in oxides and fluorides. J Appl Phys.

[80] Ohsato H, Sugino J, Komura A, Nishigaki S, Okuda T. Microwave Dielectric Proper‐ ties of Ba4(Nd28/3-y*R*y)Ti18O54 (*R* = Eu, Dy, Ho, Er and Yb) Solid Solutions. Jpn J Appl

[81] Ohsato H. Crystallography and R&D for Material Science from Our Research: Elec‐

[82] Nagatomo T, Otagiri T, Suzuki M, Ohsato H. Microwave dielectric properties and crystal structure of the tungstenbronze-type like (Ba1 − αSrα)6(Nd1 − βYβ)8Ti18O54 solid

[83] Suzuki M, Ohsato H, Kakimoto K, Nagatomo T, Otagiri T. Crystal structure and mi‐ crowave dielectric properties of (Ba1−αSrα)6−3xSm8+2xTi18O54 solid solutions. J Eur Ceram

[84] Mercurio D, Abou-Salama M, Mercurio PJ, Investigations of the Tungsten-bronze-Type (Ba1-αSrα)6-xLa8+2x/3Ti18O54 (0 = *x* =3) Solid Solutions. J Eur Ceram Soc.

[85] Imaeda M, Ito K, Mizuta M, Ohsato H, Nishigaki S, Okuda T. Microwave dielectric properties of Ba6-3xSm8+2xTi18O54 solid solutions with Sr substituted for Ba. Jpn J Appl

[86] Ohsato H, Suzuki M, Kakimoto K. Ionic Distribution and Microwave Dielectric Prop‐ erties for Tungstenbronze-Type Like Ba6-3xR8+2xTi18O54 (R = Sm, Nd and La) Solid Solu‐

[87] Ishizawa N. Layered Perovskite-Structural Classification. Ceramics. 1996;31:409-413

[88] Magnéli A. Structures of the ReO3-type with recurrent dislocations of atoms: `homol‐ ogous series' of molybdenum and tungsten oxides. Acta Cryst. 1953;6:495–500. DOI:

tions. Ceramic Engineering and Science Proceedings. 2005;26(5):135-145.

troceramics. Advanced Materials Research. 2006;11-12:95-100.

solutions. J Eur Ceram Soc. 2006;26:1895-1898.

1998: p. 440-447.

328 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

1998;100:p41-50.

1993;73:348.

Phys. 1999;38(9B):5625-5628.

Soc. 2006;26:2035-2038.

Phys. 1997;36(9B):6012-6015.

10.1107/S0365110X53001381

2001;21:2713-2716.

(Japanese).


## **ESR and Magnetization Studies of Bi-manganites**

## Rajender Singh and Ramesh Ade

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61936

## **Abstract**

[103] Dion M, Ganne M, Tournoux M. Nouvelles familles de phases MI

[104] Jacobson AJ, Johnson JW, Lewandowski JT. Interlayer chemistry between thick transi‐ tion-metal oxide layers: synthesis and intercalation reactions of K[Ca2Nan-3NbnO3n+1]

[105] Ishizawa N, Marumo F, Kawamura T, Kimura M. Compounds with perovskite-type slabs. II. The crystal structure of Sr2Ta2O7. Acta Cryst.1976;B32:2564-2566.

[106] Ishizawa N, Marumo F, Iwai S. Compounds with perovskite-type slabs. IV. Ferro‐ electric phase transitions in Sr2(Ta1-xNbx)2O7 (x 0.12) and Sr2Ta2O7. Acta Cryst.

[107] Nanot PM, Queyroux F, Gilles JC, Chevalier R. Structure cristalline du composé Nd4Ca2Ti6O20, terme n= 6 de la série (Nd,Ca)n, TinO3n+2. Acta Cryst. 1976; B32:

[108] Okawa T. Research for BaO-*R*2O3-TiO2 (*R* = Rear Earth) microwave dielectric ceram‐

[109] Valant M, Suvorov D, Kolar D. Role of Bi2O3 in optimizing the dielectric properties of

[110] Okawa T, Imaeda M, Ohsato H. Microwave Dielectric Properties of Bi-Added Ba4Nd9+1/3Ti18O54 Solid Solutions. Jpn J Appl Phys. 2000;39(9B):5645-5649.

[111] Yamada H, Okawa T, Tohdo Y, Ohsato H. Microwave dielectric properties of BaxLa4Ti3 + xO12 + 3x (*x* = 0.0–1.0) ceramics. J Eur Ceram Soc. 2006;26:2059-2062.

[112] Ohsato H, Harada A, Okawa T, Okabe H. Microwave dielectric composite composi‐

[113] Murakami M, Hirose K, Kawamura K, Sata N, Ohishi Y. Post-perovskite phase tran‐

[114] Lundberg M, Sundberg M, Magneli A. The "Pentagonal Column" as a Building Unit in Crystal and Defect Structure of Some Groups of Transition Metal Compounds. J

[115] Ohsato H. Functional advances of microwave dielectrics for next generation. Ceram‐

[116] Tenue Publicity Group. Grain 29: TV ghost hunted/effect of natural resonating/elec‐ tromagnetic-wave absorption. In: A guidebook of micro cosmos for ferrite/With fer‐

[117] Tanaka N. High refractive transparency ceramics. FC Report. 2003;21:90–91. (Japa‐

ics [Dr. thesis]. Nagoya Institute of Technology; 2003. (Japanese)

Ba4.5Nd9Ti18O54 based microwave ceramics. J Mat Res. 1996;11:928-931.

lets "perovskites". Mater Res Bull. 1981;16:1429-1435.

330 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

(3.ltoreq. n.ltoreq. 7). Inorg Chem. 1985;24:3727-3729.

1981;B37:26-31.

tion. US 7,046,258 B2, (2006).

Solid State Chem. 1982:44:32-40.

nese).

ics International, 2012;38S:S141-S146.

rite. TDK Corporation; 2007: p. 86–87. (Japanese)

sition in MgSiO3. Science. 2004;304:855-858.

1115-1120.

MII

2Nb3O10 a feuil‐

The electron spin resonance (ESR) and magnetization (M) studies of Bi-manganites were undertaken to understand the coexistence of various magnetic phases and their effect on charge ordering as a function of composition and temperature. The studies on several compositions of Bi-manganites reveal that the electronic phase separation (PS) is an in‐ trinsic phenomenon in this system.

**Keywords:** Bi-manganites, ESR, magnetization, electronic phase separation

## **1. Introduction**

The perovskite structure is represented by general composition ABO3. Manganites belong to this family with A as trivalent (La3+, Pr3+, Bi3+, etc.) and/or divalent (Ba2+, Ca2+, Sr2+, etc.) ion and B as Mn ion. Manganites have a complex phase diagram and display interesting properties like colossal magnetoresistance (CMR) and giant magnetoresistance (GMR) [1‒7]. These properties are the result of strong coupling between the charge, spin, and orbital degrees of freedom. The strength of this coupling depends on hydrostatic pressure [8, 9], magnetic and electric fields [10, 11], grain size [12, 13], and disorder created due to substitution at A-/B-site. The trivalent rare-earth ions doping of different sizes into the perovskite structure causes disorder in the sample due to change in chemical pressure, leading to the evolution of coexistence of various magnetic phases viz: paramagnetic (PM), ferromagnetic (FM), cantedantiferromagnetic (C-AFM), and antiferromagnetic (AFM). The phenomenon of phase separation (PS) has been proposed to explain the properties of manganites in view of inho‐ mogeneities arising due to doping of the material. The PS can be electronic or structural. The electronic PS occurs when cluster formation takes place at nanometric level. The structural PS, which is due to disorder, can induce formation of up to micrometer-size clusters, which assist in percolation leading to first-order transitions [14–16]. The Griffiths Phase (GP) concept [17]

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

has been used to explain the coexistence of FM and AFM domains in different types of manganites.

Bi1-*x*Ca*x*MnO3 (BCMO) and Bi1-*x*Sr*x*MnO3 (BSMO) manganites are insulating over the entire range of compositions with various phases [18–20]. The charge ordering temperature (*TCO*) is 300–500 K for these materials, which is high compared to rare-earth materials. The charge ordering (CO) in these materials does not depend on the one electron bandwidth mechanism like in other manganites as it is due to highly polarizable 6s2 lone pair of Bi3+ions.

In this chapter, we will review the recent work reported on these systems. The effect of substitution of various elements including transition elements at Bi- and Mn-site of these systems is reviewed. The study on various systems is considered to elucidate the intrinsic nature of PS phenomenon in these systems. In order to understand the evolution of various magnetic phases as a function of dopants at different sites and grain size; the following series of samples synthesized by solid-state (SS) or sol–gel (SG) methods are considered.


The properties of the samples were studied using the following characterization techniques.


## **2. Electron spin resonance**

There are two experimental techniques, namely magnetic resonance and inelastic neutron scattering, which can be used to describe the spin dynamics. Since double-exchange (DE) interactionmechanismis anintrinsicdynamicalprocess,the investigationof spindynamics and their static properties is crucial to study the magnetic properties of manganites. Although neutron scattering can provide information about magnetically ordered phases, but it has limitations in probing the spin dynamics of PM phases. The electron spin/paramagnetic resonance (ESR/EPR) is a sensitive technique which can help to understand the spin structure

anditsdynamics in complexmagnetic-orderedphases.This technique has beenusedby several scientists to study the spin dynamics of manganites. In the present work, the ESR technique is used to investigate various dynamical processes over a wide range of temperatures.

## **2.1. Origin of ESR in manganites**

has been used to explain the coexistence of FM and AFM domains in different types of

Bi1-*x*Ca*x*MnO3 (BCMO) and Bi1-*x*Sr*x*MnO3 (BSMO) manganites are insulating over the entire range of compositions with various phases [18–20]. The charge ordering temperature (*TCO*) is 300–500 K for these materials, which is high compared to rare-earth materials. The charge ordering (CO) in these materials does not depend on the one electron bandwidth mechanism

In this chapter, we will review the recent work reported on these systems. The effect of substitution of various elements including transition elements at Bi- and Mn-site of these systems is reviewed. The study on various systems is considered to elucidate the intrinsic nature of PS phenomenon in these systems. In order to understand the evolution of various magnetic phases as a function of dopants at different sites and grain size; the following series

of samples synthesized by solid-state (SS) or sol–gel (SG) methods are considered.

**3.** La-doped Bi-manganites [Bi0.7-*x*La*x*Ca0.3MnO3 (*x* = 0.07–0.70) and Bi0.30La0.37Ca0.33MnO3]

The properties of the samples were studied using the following characterization techniques.

There are two experimental techniques, namely magnetic resonance and inelastic neutron scattering, which can be used to describe the spin dynamics. Since double-exchange (DE) interactionmechanismis anintrinsicdynamicalprocess,the investigationof spindynamics and their static properties is crucial to study the magnetic properties of manganites. Although neutron scattering can provide information about magnetically ordered phases, but it has limitations in probing the spin dynamics of PM phases. The electron spin/paramagnetic resonance (ESR/EPR) is a sensitive technique which can help to understand the spin structure

lone pair of Bi3+ions.

like in other manganites as it is due to highly polarizable 6s2

332 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**1.** Bi1-*x*Ca*x*MnO3 (*x* = 0.30–0.90) (BCMO)

**b.** Scanning electron microscopy (SEM)

**d.** Electron spin resonance (ESR)

**2. Electron spin resonance**

**c.** Energy dispersive x-ray spectroscopy (EDX)

**e.** Magnetization by vibrating sample magnetometer.

**a.** X-ray diffraction (XRD)

**2.** Bi0.5-*x*Pr*x*Ca0.5MnO3 (*x* = 0.0–0.50) (BPCMO)

**4.** Bi0.5Ca0.5Mn0.95TE0.05O3 (TE = V, Cu, and Zn)

**5.** Bi0.5Ca0.5Mn0.95TM0.05O3 (TM = Ni, Fe, Co, and Cr)

**6.** Bi0.55Ca0.45MnO3 and Bi0.5Ca0. 5MnO3 nanoparticles

manganites.

The Mn2+ ion (*S* = 5/2) is the most common ion detected by ESR. The Mn3+ (3d4 with *S* = 2) exhibits a large zero-field splitting and strong spin-lattice relaxation and so unlikely to be ESR active [21]. The Mn4+ ions (3d3 with *S* = 3/2) give an ESR signal usually at low temperatures. The manganese ions, Mn3+ and Mn4+, are commonly seen in doped manganites. Theories suggesting that the two Mn3+ ions transforming into Mn2+ and Mn4+ ion pairs via thermally excited disproportionation were common [22–24]. It is generally accepted that manganites do not have, or have a very small ratio of, Mn2+ ions and hence do not contribute to the ESR signals exhibited by manganites. ESR signals obtained in La-based manganites have been attributed primarily to Mn4+ ions in an octahedral anion crystal electric field in ground state with a weak spin-lattice relaxation [25], which makes this ion ESR active even at higher temperatures [23]. Shengelaya et al. [25] attributes the ESR signal to Mn4+ in a system of three components such as Mn4+ ions, Mn3+ ions, and the lattice, and predicted an FM Curie–Weiss like behavior for the double-integrated line intensity, which is proportional to the static susceptibility arising from the FM couplings of the Mn4+ and Mn3+ magnetic subsystems [26]. But the ESR signals cannot be attributed to isolated Mn4+ ions. The origin of ESR signal observed in manganites has been assigned to some combinations or clusters of Mn3+–Mn4+ ions coupled by a strong short-range FM–DE interaction [4, 27]. All the Mn ions are assumed to contribute to the ESR signal and the normalized double-integrated intensity (*DI*) is therefore proportional to the number of ESR centers and also a measure of ESR susceptibility (χESR) [4, 28].

## **2.2. Theoretical background**

The experimental ESR spectra (Fig. 1) was analyzed by Lorentzian expression given as

$$\frac{dP}{dH} = \frac{d}{dH}A\left(\frac{\Delta H}{4\left(H - H\_0\right)^2 + \Delta H^2} + \frac{\Delta H}{4\left(H + H\_0\right)^2 + \Delta H^2}\right) \tag{1}$$

where *H*0 is the resonance field, Δ*H* is the linewidth, and *A* is a quantity proportional to the area under the curve which is related to the intensity of the signal [21, 29, 30]. The two terms in Eq. 1 describe the contribution from the clockwise and anticlockwise circularly polarized components of microwave radiation.

It was suggested [28] that when the ESR signals are intense near the critical point (Curie temperature, *T*C), even for relatively small amount of samples, the Δ*H* depends strongly on the number of spins in the sample. Whenever large amount of samples are used, the sample size effects arise from overloading the cavity through magnetic losses [28, 31]. The ESR signals

**Figure 1.** Lorentzian fit (solid red line) to the ESR spectra (open circles) of Bi1-*<sup>x</sup>*Ca*x*MnO3 at 273 K [6].

remain Lorentzian as long as the magnetic losses are not large enough to drive the diode detector out of linearity [28]. When the sample is loaded in the cavity, at a fixed microwave frequency, ∆H can be given by the following expression

$$
\Delta H^{\text{obs}} = \Delta H \left(1 + b\right)^{1/2} \tag{2}
$$

with

$$b = \frac{\binom{4\pi}{\mathcal{Z}} \binom{}{\mathcal{Z}} \eta \mathcal{Q}\_L \mathcal{X}\_{\text{ESR}} \alpha}{\eta \Delta H} \tag{3}$$

where *χ*ESR is the static susceptibility corresponding to resonant species, *η* is the fitting factor, and *QL* is the loaded *Q* of the microwave cavity. Similarly, the changes in line intensity (*I*) also occur by size effects due to magnetic losses, and the expression for this can be given [28] as

$$I = \eta \chi\_{ESR} \alpha Q\_L \left(1 + b\right)^{-1/2} \tag{4}$$

These two equations (2) and (4) have certain limitations such as the value of *b* should be « 1. In addition to the sample dependent effects, frequency or applied field dependences of ∆*H* have also been reported [28, 32, 33]. Apart from the above-mentioned parameters, skin depth (*δ*) also affects the parameters *I* and Δ*H* of the ESR signal, and it is given by the expression

$$
\delta = \left(\frac{\rho}{\mu\_\circ \alpha}\right)^{0.5} \tag{5}
$$

where *ρ* is the resistivity of the sample measured at room temperature (RT), *μ*<sup>o</sup> (= 4π × 10–7 Vs/ Am) is the permeability of free space, and *ω* (= 2π × 9 GHz) is the microwave frequency. The estimated values of *δ* are found to be in the range 0.2–1.5 mm, and it is larger than that of the size of samples [4]. In the present work, we have also taken care of cavity over loading problems. This ensures that the observed ESR signal intensity, Δ*H,* and resonance field (*H0*) as a function of temperature are intrinsic in nature.

The study of magnetic properties of the samples with ESR technique is limited in the low temperatures due to its instrumental limitations. In order to support the ESR findings and to study the magnetic properties down to the temperature up to 5 K (from 350 K), the Quantum design Physical Property Measurement System (PPMS) was used. Magnetization as a function of temperature was measured in both zero field cooled (ZFC) and field cooled (FC) by applying a field of 500 Oe.

## **2.3. Studies on Bi1-***x***Ca***x***MnO3 (***x* **= 0.30–0.90)**

remain Lorentzian as long as the magnetic losses are not large enough to drive the diode detector out of linearity [28]. When the sample is loaded in the cavity, at a fixed microwave

> ( ) 1/2

<sup>3</sup> *QL ESR*

hcw

*H*

where *χ*ESR is the static susceptibility corresponding to resonant species, *η* is the fitting factor, and *QL* is the loaded *Q* of the microwave cavity. Similarly, the changes in line intensity (*I*) also occur by size effects due to magnetic losses, and the expression for this can be given [28] as


These two equations (2) and (4) have certain limitations such as the value of *b* should be « 1. In addition to the sample dependent effects, frequency or applied field dependences of ∆*H* have also been reported [28, 32, 33]. Apart from the above-mentioned parameters, skin depth (*δ*) also affects the parameters *I* and Δ*H* of the ESR signal, and it is given by the expression

0.5

*o* r

æ ö = ç ÷ ç ÷ è ø

m w

d

( ) 1/2

g<sup>=</sup> <sup>D</sup>

1 *ESR L I Qb* hc w

( ) <sup>4</sup>

p

**Figure 1.** Lorentzian fit (solid red line) to the ESR spectra (open circles) of Bi1-*<sup>x</sup>*Ca*x*MnO3 at 273 K [6].

*b*

1 *obs* D =D + *H Hb* (2)

= + (4)

(3)

(5)

frequency, ∆H can be given by the following expression

334 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

with

Figure 2 depicts the Rietveld refinement of XRD pattern for the sample with *x* = 0.65. With the refinement, XRD data are well reproduced with *Pnma* space group orthorhombic structure of all the samples except the sample with *x* = 0.30. Using the refined parameters crystal structure is generated for all the samples, as shown in Fig. 3 for Bi0.35Ca0.65MnO3 which is representative of all the samples. With increase in Ca content, decrease in unit cell volume is observed. This is ascribed to lower ionic radii of Ca2+ (1.12 Å) than that of Bi3+ (1.24 Å) ion.

**Figure 2.** Rietveld refinement of the XRD data of Bi1-*x*Ca*x*MnO3.

Figure 4 shows *DI* of ESR signal as a function of temperature. For samples with 0.40 ≤ *x* ≤ 0.80, as the temperature decreases from high temperature (453K), *DI* increases and reaches its maximum, which is assigned as charge ordering temperature (*TCO*) [5]. Below *TCO*, *DI* decreases, evidencing the existence of AFM correlations and reaches a minimum and then it increases with further decrease in temperature for samples with 0.65 ≤ *x* ≤ 0.80. For *x* = 0.30 sample, a weak *TCO* is observed. For *x* = 0.85 and 0.90 samples, no *TCO* is observed. For *x* = 0.85 and 0.90 samples, on cooling from high temperature, *DI* increases gradually. For 0.30 ≤ *x* ≤ 0.80 samples,

**Figure 3.** Crystal structure of Bi0.35Ca0.65MnO3.

sharp decrease in *DI* below the temperature ~126–150 K is assigned as the temperature at which onset of long-range AFM ordering takes place, that is, Neel temperature (*TN*).

**Figure 4.** *DI* vs *T* plots of Bi1-*x*Ca*x*MnO3 samples.

The 1/*DI* vs *T* plots for various samples are shown in Fig. 5. The linear fit in the high-temper‐ ature range is as per Curie–Weiss law [5]. The positive intercept on the x-axis in the temperature range *T* > *TCO* for samples with 0.30 ≤ *x* ≤ 0.80, indicates the existence of FM–DE correlations. The charge ordering transition of 0.65 ≤ *x* ≤ 0.80 samples is smeared compared to what is observed for samples with *x* ≤ 0.60. The CO transition at 303 K is weak for sample with *x* = 0.75. The existence of Bi-rich CO phase, a chemical phase separation in the samples, with *x* = 0.75, 0.70, and 0.65 may be the reason for this. The upward and downward trend in 1/*DI* vs *T* plots

**Figure 5.** 1/*DI* vs *T* plots of Bi1-*x*Ca*x*MnO3 samples. Ref. [5]

sharp decrease in *DI* below the temperature ~126–150 K is assigned as the temperature at which

The 1/*DI* vs *T* plots for various samples are shown in Fig. 5. The linear fit in the high-temper‐ ature range is as per Curie–Weiss law [5]. The positive intercept on the x-axis in the temperature range *T* > *TCO* for samples with 0.30 ≤ *x* ≤ 0.80, indicates the existence of FM–DE correlations. The charge ordering transition of 0.65 ≤ *x* ≤ 0.80 samples is smeared compared to what is observed for samples with *x* ≤ 0.60. The CO transition at 303 K is weak for sample with *x* = 0.75. The existence of Bi-rich CO phase, a chemical phase separation in the samples, with *x* = 0.75, 0.70, and 0.65 may be the reason for this. The upward and downward trend in 1/*DI* vs *T* plots

onset of long-range AFM ordering takes place, that is, Neel temperature (*TN*).

**Figure 4.** *DI* vs *T* plots of Bi1-*x*Ca*x*MnO3 samples.

**Figure 3.** Crystal structure of Bi0.35Ca0.65MnO3.

336 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 6.** Δ*H* vs *T* plots of Bi1-*x*Ca*x*MnO3 samples. Ref. [5].

below *TCO* is an indication whether AFM or FM interactions dominate in the respective temperature range [5]. The minima in these plots indicate the onset of CO or AFM ordering. For samples with *x* = 0.65, 0.75, and 0.80, 1/*DI* vs *T* plots show a broad peak at ~160 K, which is attributed to C-type AFM correlations. The down turn behavior of 1/*DI* in the temperature range below 148 K for the samples with *x* = 0.85 and 0.90 can also be explained in view of shortrange ordering of FM clusters in PM matrix as per Griffiths theory. The negative intercept on the x-axis for samples with *x* = 0.85 and 0.90 indicates dominant AFM-superexchange (SE) interactions above 150 K. Below 150 K, localized FM correlations dominate [4–6].

Figure 6 shows Δ*H* as a function of temperature. For samples with *x* ≥ 0.65, ∆*H* shows promi‐ nent minima at ~150 K. The ESR signal persists only for a few degrees below 150 K for all the samples and then it disappears indicating the onset of long-range AFM state (*TN*). ∆*H* increases linearly with increase in temperature in the temperature range *T*≥*TCO.* ∆*H* saturates with increase in temperature as *TCO* decreases with increase in *x.* This effect is clearly seen for sample with *x* = 0.8. ∆*H* increases with decrease in temperature for the sample with *x* = 0.90, indicating the strengthening of AFM interactions with lowering of temperature.

**Figure 7.** *M* vs *T* plots of Bi1-*x*Ca*x*MnO3 samples.

Figure 7 shows magnetization vs temperature (*M* vs *T)* plots for various samples. As the temperature decreases from 350 K, both ZFC and FC curves show a peak at *TCO* for samples with 0.30 ≤ *x* ≤ 0.80. *M* decreases slightly on further decrease in temperature. The broad maximum at ~150 K observed for samples with *x* ≤ 0.50 represents long-range AFM order [6]. The sharp rise in *M* for samples with *x* ≤ 0.60 at ~50 K is related to the spin-glass (SG) or clusterglass (CG) transitions which arises due to competition between FM interaction in BiMnO3-type clusters and AFM coupling in clusters in which Mn3+ orbits are frozen in random orientations [6]. For these samples, *M* vs *T* plots in ZFC mode show a peak in *M* at spin freezing temperature (*Tf* ). A small peak in *M* at ~120 K observed for sample with *x* = 0.50 is ascribed to field-induced spin canting in the AFM phase aided by the spontaneous canting moment [6]. The trend and the conclusions drawn from 1/*χ* vs *T* (not shown) and 1/*DI* vs *T* plots are similar.

is attributed to C-type AFM correlations. The down turn behavior of 1/*DI* in the temperature range below 148 K for the samples with *x* = 0.85 and 0.90 can also be explained in view of shortrange ordering of FM clusters in PM matrix as per Griffiths theory. The negative intercept on the x-axis for samples with *x* = 0.85 and 0.90 indicates dominant AFM-superexchange (SE)

Figure 6 shows Δ*H* as a function of temperature. For samples with *x* ≥ 0.65, ∆*H* shows promi‐ nent minima at ~150 K. The ESR signal persists only for a few degrees below 150 K for all the samples and then it disappears indicating the onset of long-range AFM state (*TN*). ∆*H* increases linearly with increase in temperature in the temperature range *T*≥*TCO.* ∆*H* saturates with increase in temperature as *TCO* decreases with increase in *x.* This effect is clearly seen for sample with *x* = 0.8. ∆*H* increases with decrease in temperature for the sample with *x* = 0.90, indicating

Figure 7 shows magnetization vs temperature (*M* vs *T)* plots for various samples. As the temperature decreases from 350 K, both ZFC and FC curves show a peak at *TCO* for samples with 0.30 ≤ *x* ≤ 0.80. *M* decreases slightly on further decrease in temperature. The broad maximum at ~150 K observed for samples with *x* ≤ 0.50 represents long-range AFM order [6]. The sharp rise in *M* for samples with *x* ≤ 0.60 at ~50 K is related to the spin-glass (SG) or clusterglass (CG) transitions which arises due to competition between FM interaction in BiMnO3-type clusters and AFM coupling in clusters in which Mn3+ orbits are frozen in random orientations [6]. For these samples, *M* vs *T* plots in ZFC mode show a peak in *M* at spin freezing temperature

). A small peak in *M* at ~120 K observed for sample with *x* = 0.50 is ascribed to field-induced

interactions above 150 K. Below 150 K, localized FM correlations dominate [4–6].

the strengthening of AFM interactions with lowering of temperature.

338 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 7.** *M* vs *T* plots of Bi1-*x*Ca*x*MnO3 samples.

(*Tf*

**Figure 8.** Phase diagram of BCMO system. Open circles (*TCO*) and closed circles (*TN*) are from ESR data and closed stars (*TCO*) and open stars (*TN*) from magnetization data [6]

Figure 8 shows the estimated values of *TCO* and *TN* from the ESR and magnetization data for Bi1-*x*Ca*x*MnO3. The magnetic phase in the temperature range *T* > *TCO* is dominated by FM–DE interactions for samples with 0.30 ≤ *x* ≤ 0.80. The PM–AFM transition at ~150 K coexists with PM to C-AFM transition at ~120 K for samples with 0.50 ≤ *x* ≤ 0.80, whereas for samples with *x* = 0.85 and 0.90, only PM to C-AFM transition is observed.

## **2.4. Studies on Bi0.5-***x***Pr***x***Ca0.5MnO3 (0.0 ≤** *x* **≤ 0.50) manganites**

The XRD study of the Bi0.5-*x*Pr*x*Ca0.5MnO3 (0.0 ≤ *x* ≤ 0.50) samples [34] confirms the single-phase orthorhombicnaturewith*Pnma*spacegroup.Asharpdecreaseintheunitcellvolumeisobserved asPr contentincreases from*x* = 0.0 to 0.05. However,theunit cell volume increases steadily with further increase in Pr content from 0.05 to 0.50. The Bi3+ has effective ionic size of 1.170 or 1.24 Å depending on whether the 6s2 lone pair character is constrained or dominant, respectively [34]. The higher cell volume of undoped sample may be due to the strengthened dominant lone pair character of Bi3+ ion, whereas the sharp drop in the cell volume for sample with *x* = 0.05 indi‐ cates thattheconstrainedeffectofthe6s2 lonepairbecomes strongduetoPrdoping.Theincrease in cell volume for samples with *x* > 0.05 indicates that the constrained effect of the lone pair maintains the Bi3+ ionic size at 1.170 Å, that is, smaller than Pr3+ ionic size (1.179 Å).

The temperature dependence of Δ*H* of ESR signal is shown in Fig. 9. As temperature decreases from high temperature (453 K), ∆*H* decreases and reaches a minimum value at *TCO*. At ~200 K, Δ*H* decreases sharply and reaches a minimum at 130, 150, and 160 K for samples with 0.05 ≤ *x* ≤ 0.15, 0.20 ≤ *x* ≤ 0.40, and *x* = 0.50, respectively. These minima indicate the existence of FM

**Figure 9.** Linewidth versus Temperature plots of Bi0.5-*x*Pr*x*Ca0.5MnO3 samples.

**Figure 10.** *TCO* versus *x* of Bi0.5-*x*Pr*x*Ca0.5MnO3 samples.

magnetic clusters embedded in the AFM matrix. Further, decrease in temperature leads to increase in Δ*H* due to the evolution of canted-AFM (C-AFM) phase.

The value of *TCO,* as a function of Pr content (Fig. 10), estimated from ESR data matches quite well with those estimated from magnetization data analysis. *TCO* decreases as *x* increases from 0.0 to 0.20 and becomes independent of doping content with further increase in *x*. For samples with 0.0 ≤ *x* ≤ 0.15, *TCO* decreases and Curie–Weiss temperature (*θC*) increases with increase in Pr doping.

**Figure 11.** *M* versus *T* plots of Bi0.5-*x*Pr*x*Ca0.5MnO3 samples sizes measured under the applied static field of 500 Oe.

The sharp rise in *M* (Fig. 11) value below 45 K for undoped sample and below 115 K for doped samples is observed, which is characteristic of FM-cluster glass (CG) behavior. The tempera‐ ture *Tf* is called spin-freezing temperature. Below this temperature, *M* in ZFC mode decreases due to the competition between the evolving FM and AFM interactions, whereas *M* in FC mode continuously increases with decreasing temperature.

## **2.5. Studies on La-doped Bi-manganites**

magnetic clusters embedded in the AFM matrix. Further, decrease in temperature leads to

increase in Δ*H* due to the evolution of canted-AFM (C-AFM) phase.

**Figure 10.** *TCO* versus *x* of Bi0.5-*x*Pr*x*Ca0.5MnO3 samples.

**Figure 9.** Linewidth versus Temperature plots of Bi0.5-*x*Pr*x*Ca0.5MnO3 samples.

340 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The base compound, Bi0.7Ca0.3MnO3, has monoclinic structure, and the details are given in section 2.3, whereas the La-doped Bi-manganites, Bi0.7- *x*La*x*Ca0.3MnO3 (*x* = 0.07, 0.14, 0.28, 0.35 and 0.7), have cubic structure. The unit cell volume increases with increasing La content.

Figure 12 shows the ESR spectra for *x* = 0.07 and 0.70 samples at some selected temperatures. For all samples, the ESR spectra above a particular temperature (~260–270 K) show a single resonance line at resonance field of 325 mT, which is characteristic of the existence of PM phase. Below this temperature, FM correlations evolve in the PM matrix as indicated by the appear‐ ance of a shoulder (shown with '↓1') in the low-field region of the ESR signal. This shoulder can be assigned to a low-moment state. With further decrease in temperature, FM correlations grow at the expense of PM interactions, as a result of which a complete PM–FM transition occurs below ~180 K. This is indicated by the shift of the shoulder toward further lower-field region (shown with '↓2'). The data indicate development of a high-moment state as tempera‐ ture decreases.

**Figure 12.** ESR spectra of Bi0.7-*x*La*x*Ca0.3MnO3 (*x* = 0.07, 0.7).

Figure 13 shows *M* vs *T* plots of all the samples. *M* increases with increasing La content. The spin or cluster glass behavior of Bi- and La-containing samples is similar to that of Bi1-*x*Ca*x*MnO<sup>3</sup> (*x* ≤ 0.6) system [6]. The samples show the coexistence of PM–FM phases over a wide temper‐ ature range. The estimated *T*C, *θ*C, and the temperature at which ∆*H* becomes minimum (*T*min) increase with increase in La content.

The critical behavior of Bi0.30La0.37Ca0.33MnO<sup>3</sup> (orthorhombic structure with *Pnma* space group) manganite at the critical point, where the system undergoes phase transition from PM to FM state, is investigated by using modified-Arrott plots (Fig. 14), Kouvel-Fisher method (Fig. 15), and critical isotherm analysis [35].

The sample shows second-order phase transition near the critical point [36]. The estimated critical exponents are close to as per the prediction by mean-field theory (MFT).

**Figure 13.** *M* versus *T* plots of Bi0.7-*x*La*x*Ca0.3MnO3 samples measured under the applied static field of 500 Oe.

**Figure 14.** Modified Arrott plots of Bi0.30La0.37Ca0.33MnO3.

Using scaling equation,

Below this temperature, FM correlations evolve in the PM matrix as indicated by the appear‐ ance of a shoulder (shown with '↓1') in the low-field region of the ESR signal. This shoulder can be assigned to a low-moment state. With further decrease in temperature, FM correlations grow at the expense of PM interactions, as a result of which a complete PM–FM transition occurs below ~180 K. This is indicated by the shift of the shoulder toward further lower-field region (shown with '↓2'). The data indicate development of a high-moment state as tempera‐

342 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Figure 13 shows *M* vs *T* plots of all the samples. *M* increases with increasing La content. The spin or cluster glass behavior of Bi- and La-containing samples is similar to that of Bi1-*x*Ca*x*MnO<sup>3</sup> (*x* ≤ 0.6) system [6]. The samples show the coexistence of PM–FM phases over a wide temper‐ ature range. The estimated *T*C, *θ*C, and the temperature at which ∆*H* becomes minimum (*T*min)

The critical behavior of Bi0.30La0.37Ca0.33MnO<sup>3</sup> (orthorhombic structure with *Pnma* space group) manganite at the critical point, where the system undergoes phase transition from PM to FM state, is investigated by using modified-Arrott plots (Fig. 14), Kouvel-Fisher method (Fig. 15),

The sample shows second-order phase transition near the critical point [36]. The estimated

critical exponents are close to as per the prediction by mean-field theory (MFT).

ture decreases.

**Figure 12.** ESR spectra of Bi0.7-*x*La*x*Ca0.3MnO3 (*x* = 0.07, 0.7).

increase with increase in La content.

and critical isotherm analysis [35].

$$M\left(H,\mathfrak{c}\right) = \mathfrak{s}^{\beta} f\_{\pm}\left(H \mid \mathfrak{s}^{\beta \ast \gamma}\right) \tag{6}$$

**Figure 15.** Kouvel-Fisher plots of Bi0.30La0.37Ca0.33MnO3.

**Figure 16.** Scaling plots below and above TC using the critical exponents determined from Kouvel–Fisher method of Bi0.30La0.37Ca0.33MnO3.

where *f+* for *T* > *TC* and *f* for *T* < *TC* are regular functions [35], we have plotted as shown in Fig. 16. There are two curves, one above the *TC* and the other one below *TC*, in agreement with the scaling theory. This indicates that the calculated critical exponents are reliable.

Bi1-*x*Ca*x*MnO3 (BCMO) exhibits charge ordering property in the composition range 0.30 ≤ *x* ≤ 0.80, which is related to first-order transition [3–6]. At critical point, La1-*x*Ca*x*MnO3 (LCMO) manganite shows first-order transition in the composition range 0.30 ≤ *x* ≤ 0.40 and secondorder transition in the composition range 0.18 ≤ *x* ≤ 0.25 and 0.40 ≤ *x* ≤ 0.48 [37–39], whereas the present Bi0.30La0.37Ca0.33MnO3 sample also exhibits second-order transition which is similar to La0.7Sr0.3MnO3 (LSMO) [40]. In the present system, due to La3+ doping, the average A-site ionic radius increases, as in the case of Sr2+-doped La-manganite. The increased average A-site radius might be responsible for the observed second-order phase transition.

The ESR and magnetization studies on La-doped Bi-manganites reveal that the La-doping increases *M* as well as various transition temperatures describing the evolution of various magnetic phases. These changes are ascribed to the difference in ionic radii of Bi3+ and La3+ ions and the variations in 6s2 lone pair character of Bi3+ ions with La-doping.

## **2.6. Comparative studies on fixed doping level of transition element (TE = V, Cu and Zn) at Mn-site in Bi0.5Ca0.5MnO3 system**

XRD data analysis indicates that all the samples possess orthorhombic structure [41]. The difference between *a* and *b* lattice parameters decreases, and the *c* lattice parameter increases with doping. The change in the lattice parameters with doping confirms that the dopants are substituted at the Mn-site [41].

**Figure 17.** Linewidth versus temperature plots for Bi0.5Ca0.5Mn0.95TE0.05O3 (a) undoped and doped samples with TE (b) V (c) Cu and (d) Zn [41]

where *f+* for *T* > *TC* and *f-*

Bi0.30La0.37Ca0.33MnO3.

**Figure 15.** Kouvel-Fisher plots of Bi0.30La0.37Ca0.33MnO3.

344 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

for *T* < *TC* are regular functions [35], we have plotted as shown in Fig.

16. There are two curves, one above the *TC* and the other one below *TC*, in agreement with the

**Figure 16.** Scaling plots below and above TC using the critical exponents determined from Kouvel–Fisher method of

Bi1-*x*Ca*x*MnO3 (BCMO) exhibits charge ordering property in the composition range 0.30 ≤ *x* ≤ 0.80, which is related to first-order transition [3–6]. At critical point, La1-*x*Ca*x*MnO3 (LCMO) manganite shows first-order transition in the composition range 0.30 ≤ *x* ≤ 0.40 and secondorder transition in the composition range 0.18 ≤ *x* ≤ 0.25 and 0.40 ≤ *x* ≤ 0.48 [37–39], whereas the present Bi0.30La0.37Ca0.33MnO3 sample also exhibits second-order transition which is similar

scaling theory. This indicates that the calculated critical exponents are reliable.

ESR (Fig. 17) studies on Bi0.5Ca0.5Mn0.95TE0.05O3 with TE = V, Cu, and Zn as a function of temperature and composition show a strong interplay between FM and AFM interactions. FM and AFM phases coexistence in the temperature range *TCO* > *T* > *TO* for all the samples. In the temperature range *T* > *TCO*, all samples show FM-dominated interactions in PM state.

**Figure 18.** 1/*M* versus *T* plots for Bi0.5Ca0.5Mn0.95TE0.05O3 (a) undoped and doped samples with TE (b) V (c) Cu (d) Zn [41]

The magnetization data in Fig. 18 supports the ESR observations. The estimated values of *θ<sup>C</sup>* in the temperature range *TCO* > *T* > *TO* (Fig. 18) are 5, ‒7, ‒93, and ‒172 K for undoped, V-, Cu-, and Zn-doped samples, respectively. The changes in magnetic properties of BCMO system with TE doping at Mn-site are due to the electronic nature of the dopants [41].

## **2.7. Comparative studies on fixed doping level of transition metals (TM = Ni, Fe, Co, and Cr) at Mn-site in Bi0.5Ca0.5MnO3 system**

All the samples have orthorhombic structure [42]. The Mn replacement by TM was confirmed by change in the lattice parameters with respect to lattice parameters of undoped one. This effect is weak for Co-doped sample compared to Fe-, Cr-, and Ni-doped samples. The peak widths of Cr-doped XRD pattern are essentially broader than for other TM-doped samples. It can be due to the non-homogeneous Cr-doped phase. It can also be due to smaller crystallite size in Cr-doped sample than other samples [42].

The ESR (Fig. 19) and magnetization (Fig. 20) data on the present BCMO and BCMTMO samples show that the FM correlations dominate in the temperature region *T* > *TCO*. In the temperature range, *TCO* > *T* > *TO*, FM and AFM coexist for the undoped, Ni- and Fe-doped samples. The 3d TM doping melts the charge ordering and AFM ordering. The effect of Fe, Co, and Cr doping is same and stronger compared to Ni. This suggests that the dopants modify the interaction between Mn ions and possibly alter the band structure of the material instead of participating directly in the magnetic interaction mechanisms [42].

**Figure 19.** Linewidth versus Temperature plots for A-Bi0.5Ca0.5MnO3 and Bi0.5Ca0.5Mn0.95TM0.05O3 (TM = B-Ni, C-Fe and E-Cr)

**Figure 20.** 1/*M* versus *T* plots for doped Bi0.5Ca0.5Mn0.95TM0.05O3 (A-undoped, B-TM = Ni)

The magnetization data in Fig. 18 supports the ESR observations. The estimated values of *θ<sup>C</sup>* in the temperature range *TCO* > *T* > *TO* (Fig. 18) are 5, ‒7, ‒93, and ‒172 K for undoped, V-, Cu-, and Zn-doped samples, respectively. The changes in magnetic properties of BCMO system

**Figure 18.** 1/*M* versus *T* plots for Bi0.5Ca0.5Mn0.95TE0.05O3 (a) undoped and doped samples with TE (b) V (c) Cu (d) Zn [41]

**2.7. Comparative studies on fixed doping level of transition metals (TM = Ni, Fe, Co, and**

All the samples have orthorhombic structure [42]. The Mn replacement by TM was confirmed by change in the lattice parameters with respect to lattice parameters of undoped one. This effect is weak for Co-doped sample compared to Fe-, Cr-, and Ni-doped samples. The peak widths of Cr-doped XRD pattern are essentially broader than for other TM-doped samples. It can be due to the non-homogeneous Cr-doped phase. It can also be due to smaller crystallite

The ESR (Fig. 19) and magnetization (Fig. 20) data on the present BCMO and BCMTMO samples show that the FM correlations dominate in the temperature region *T* > *TCO*. In the temperature range, *TCO* > *T* > *TO*, FM and AFM coexist for the undoped, Ni- and Fe-doped samples. The 3d TM doping melts the charge ordering and AFM ordering. The effect of Fe, Co, and Cr doping is same and stronger compared to Ni. This suggests that the dopants modify the interaction between Mn ions and possibly alter the band structure of the material instead

with TE doping at Mn-site are due to the electronic nature of the dopants [41].

**Cr) at Mn-site in Bi0.5Ca0.5MnO3 system**

size in Cr-doped sample than other samples [42].

346 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

of participating directly in the magnetic interaction mechanisms [42].

## **2.8. Studies on nanoparticles**

The comparative ESR studies (Fig. 21) on bulk and nanometer-size (grain size 21 and 15 nm) samples of Bi0.55Ca0.45MnO3 show that *T*CO decreases from 313 to 306 K as sample grain size changes from bulk to 15 nm. The *TN* value remains around 153 K for all the samples. In Fig. 21, the solid lines are as per Curie–Weiss law. Above *T*CO, the positive intercepts on the x-axis indicate the dominant FM correlations. In the temperature range, *T*CO > *T* > *T*N, the domains of FM and AFM spin correlations coexist [43]. The FM phase volume decreases, and the transi‐ tions become sharp with decrease in grain size [43].

**Figure 21.** 1/*DI* versus *T* plots for the samples Bulk sample (filled squares), 21 nm nanoparticles (open circles) and 15 nm nanoparticles (filled stars).

The ESR data have been explained [43] in view of the Griffith's phase theory [9]. Griffith's theory suggests that above *TG* large FM clusters form. Below *TG,* smaller-sized clusters form, reducing the average size of the FM clusters. In Fig. 21, nonlinear decrease of *1/DI* with decrease in temperature indicates existence of FM cluster in a PM matrix. In the temperature range, *TCO* > *T* > *TN*, orbital ordering sets in, bringing about long-range AFM ordering. In this temperature range, the Mn spins do not completely undergo OO and hence there is a coexistence of FM and AFM phases.

The effect of grain size on the properties of Bi0.5Ca0.5MnO3 manganite samples synthesized by SG method is also studied. The samples with grain size (27, 450, and 1080 nm) were obtained by appropriate heat treatment schedule. The magnetic behavior of samples with grain size 450 and 1,080 nm is similar to that of bulk sample described in (section 2.3), whereas for sample with GS 27 nm, the long-range charge ordering and AFM ordering transitions are suppressed (Fig. 22). Magnetization results (Fig. 23) also support the ESR findings.

**2.8. Studies on nanoparticles**

nm nanoparticles (filled stars).

and AFM phases.

tions become sharp with decrease in grain size [43].

348 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The comparative ESR studies (Fig. 21) on bulk and nanometer-size (grain size 21 and 15 nm) samples of Bi0.55Ca0.45MnO3 show that *T*CO decreases from 313 to 306 K as sample grain size changes from bulk to 15 nm. The *TN* value remains around 153 K for all the samples. In Fig. 21, the solid lines are as per Curie–Weiss law. Above *T*CO, the positive intercepts on the x-axis indicate the dominant FM correlations. In the temperature range, *T*CO > *T* > *T*N, the domains of FM and AFM spin correlations coexist [43]. The FM phase volume decreases, and the transi‐

**Figure 21.** 1/*DI* versus *T* plots for the samples Bulk sample (filled squares), 21 nm nanoparticles (open circles) and 15

The ESR data have been explained [43] in view of the Griffith's phase theory [9]. Griffith's theory suggests that above *TG* large FM clusters form. Below *TG,* smaller-sized clusters form, reducing the average size of the FM clusters. In Fig. 21, nonlinear decrease of *1/DI* with decrease in temperature indicates existence of FM cluster in a PM matrix. In the temperature range, *TCO* > *T* > *TN*, orbital ordering sets in, bringing about long-range AFM ordering. In this temperature range, the Mn spins do not completely undergo OO and hence there is a coexistence of FM

The effect of grain size on the properties of Bi0.5Ca0.5MnO3 manganite samples synthesized by SG method is also studied. The samples with grain size (27, 450, and 1080 nm) were obtained

**Figure 22.** Δ*H* versus *T* plots of Bi0.5Ca0.5MnO3 sample with different grain sizes. Inset shows *Hr* versus *T* plot for 27 nm sample.

**Figure 23.** *M* versus *T* plots of Bi0.5Ca0.5MnO3 sample with different grain sizes measured under the applied field of 500 Oe.

For these samples, *M* in FC mode increases with decreasing temperature. However, the ESR and magnetization results reveal that even for sample with GS 27 nm the AFM correlations still exist in the form of short-range order. The shift in FM–CG transition and *Tf* toward higher temperature and the sharp rise in *M* below 50 K for sample with GS 27 nm are observed. The evolving different magnetic correlations with decrease in GS are ascribed to increase in surfaceto-volume ratio of grains.

## **3. Conclusions**

The ESR and M studies of Bi-manganites of various compositions synthesized by SS or SG methods are reviewed. The effect of substitution of divalent (Ca2+) and trivalent (Pr3+ and La3+) cations at Bi-site and partial replacement of Mn by transition elements on the structure and magnetic properties of these materials is reviewed. The effect of grain size on the evolution of various magnetic interactions is also described. The study on various systems is useful in understanding the evolution of various magnetic phases as a function of composition and temperature. The studies on several compositions of Bi-manganites reveal that electronic phase separation is an intrinsic phenomenon in this system.

## **Acknowledgements**

Ramesh Ade would like to thank the award of UGC-(JRF+SRF) from CSIR India.

## **Author details**

Rajender Singh\* and Ramesh Ade

\*Address all correspondence to: rssp@uohyd.ernet.in

School of Physics, University of Hyderabad, Central University, Hyderabad, Telangana, India

## **References**

[1] Jonker G H and Van Santen J H. Ferromagnetic compounds of manganese with per‐ ovskite structure. Physica, 1950; XVI: 337–349.

[2] Wollan E O and Koehler W C. Neutron diffraction study of the magnetic properties of the series of perovskite-type compounds [(1–*x*)La, *x*Ca]MnO3. Physical Review, 1955; 100: 545–563.

For these samples, *M* in FC mode increases with decreasing temperature. However, the ESR and magnetization results reveal that even for sample with GS 27 nm the AFM correlations

temperature and the sharp rise in *M* below 50 K for sample with GS 27 nm are observed. The evolving different magnetic correlations with decrease in GS are ascribed to increase in surface-

The ESR and M studies of Bi-manganites of various compositions synthesized by SS or SG methods are reviewed. The effect of substitution of divalent (Ca2+) and trivalent (Pr3+ and La3+) cations at Bi-site and partial replacement of Mn by transition elements on the structure and magnetic properties of these materials is reviewed. The effect of grain size on the evolution of various magnetic interactions is also described. The study on various systems is useful in understanding the evolution of various magnetic phases as a function of composition and temperature. The studies on several compositions of Bi-manganites reveal that electronic phase

Ramesh Ade would like to thank the award of UGC-(JRF+SRF) from CSIR India.

School of Physics, University of Hyderabad, Central University, Hyderabad, Telangana,

[1] Jonker G H and Van Santen J H. Ferromagnetic compounds of manganese with per‐

toward higher

still exist in the form of short-range order. The shift in FM–CG transition and *Tf*

350 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

to-volume ratio of grains.

**Acknowledgements**

**Author details**

Rajender Singh\*

**References**

India

separation is an intrinsic phenomenon in this system.

and Ramesh Ade

\*Address all correspondence to: rssp@uohyd.ernet.in

ovskite structure. Physica, 1950; XVI: 337–349.

**3. Conclusions**


[31] Seehra M S. New method for measuring the static magnetic susceptibility by para‐ magnetic resonance. Review of Scientific Instruments, 1968; 39: 1044–1047.

[16] Moreo A, Mayr M, Feiguin A, Yunoki S and Dagotto E. Giant cluster coexistence in doped manganites and other compounds. Physical Review Letters, 2000; 84: 5568–

[17] Griffiths R B. Nonanalytic behavior above the critical point in a random Ising ferro‐

[18] Woo H, Tyson T A, Croft M, Cheong S W and Woicik J C. Correlations between the magnetic and structural properties of Ca-doped BiMnO3. Physical Review B, 2001;

[19] Bokov V A, Grigoryan N A and Bryzhina M F. X-Ray diffraction and magnetic stud‐ ies of solid solutions Bi1-xCaxMnO3. Physica Status Solidi, 1967; 20: 745–754.

[20] Bao W, Axe J D, Chen C H and Cheong S W. Impact of charge ordering on magnetic correlations in perovskite (Bi, Ca) MnO3. Physical Review Letters, 1997; 78: 543–546.

[21] Banwell C N and McCash E M. Fundamental of molecular spectroscopy (Tata

[22] Stevenson J W, Nasrallah M M, Anderson H U and Sparlin D M. Defect structure of Y1–yCayMnO3 and La1–yCayMnO3. Journal of Solid State Chemistry, 1993; 102: 175–184.

[23] Muller K A. Electron Paramagnetic Resonance of Manganese IV in SrTiO3. Physical

[24] Muller K A, Berlinger W, Blazey K W and Albers J. Electron paramagnetic resonance

[25] Shengelaya A, Zhao G M, Keller H and Muller K A. EPR Evidence of Jahn-Teller po‐ laron formation in La1-xCaxMnO3+y. Physical Review Letters, 1996; 77: 5296–5299. [26] Gulley J E and Jacarrino V. Impure exchange-coupled paramagnets; Electron-Para‐

[27] Oseroff B, Torikachvili M, Singley J and Ali S. Evidence for collective spin dynamics above the ordering temperature in La1–xCaxMnO3+δ. Physical Review B, 1996; 53:

[28] Causa M T, Tovar M, Caneiro A, Prado F, Ibanez G, Ramos C A, Butera A, Alascio B, Obradors X, Pinol S, Tokura Y and Oseroff S B. High-temperature spin dynamics in CMR manganites: ESR and magnetization. Physical Review B, 1998; 58: 3233–3239.

[29] Kumar A, Ghosh N, Joshi J P, Bhat H L and Bhat S V. Electron paramagnetic reso‐ nance studies of the insulating ferromagnetic manganite Nd0.8Pb0.2MnO3 above the

[30] Anuradha K N, Rao S S and Bhat S V. EPR Evidence for premonitory charge-order‐ ing fluctuations in hydrothermally grown Pr0.57Ca0.41Ba0.02MnO3 Nanowires. Applied

transition temperature. Solid State Communications, 2002; 123: 379–382.

Magnetic Resonance, 2009; 36: 347–356.

of Mn4+ in BaTiO3. Solid State Communications, 1987; 61: 21–25.

magnetic-Resonance Studies. Physical Review B, 1972; 6: 58–73.

magnet. Physical Review Letters, 1969; 23: 17–19.

352 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

5571.

63: 134412(1)–134412(12).

McGraw-Hill, 1997).

6521–6525.

Review Letters, 1959; 2: 341.


## **Charge Carrier Dynamics in Organometal Halide Perovskite Probed by Time-Resolved Electrical Measurements**

Carlito S. Ponseca Jr.

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61631

## **Abstract**

This chapter presents the fate of the charge carriers from the moment of its photogenera‐ tion in the perovskite to injection and transport into electrodes. Time-resolved electrical measurement techniques, terahertz (THz) spectroscopy and microwave (MW) conductiv‐ ity, are primarily used to deconvolute ultrafast processes and to directly access behavior of charged species from the ps to μs timescales. Transient absorption and photolumines‐ cence spectroscopy were also utilized to gain insight on carrier population dynamics and radiatively recombining charges. Photogenerated charged species were converted into highly mobile charges (μe = 12.5 cm2 V-1s-1 and μh = 7.5 cm2 V-1s-1) almost instantaneously (< 0.2 ps), while the remaining loosely bounded excitons dissociate into mobile charges after 2-3 ps. This high mobility is maintained for at least 1 ns as obtained by THz spectroscopy, while its lifetime is at least few tens of μs as measured by the MW conductivity techni‐ que. Lowering the temperature increases carrier mobilities with T-1.6.Dependence and a 75 meV barrier energy is required for temperature-activated recombination. Finally, injec‐ tion of hole from MAPbI3 to Spiro-OMeTAD was found to be ultrafast and the state and population of dark holes dictate its recombination.

**Keywords:** THz spectroscopy, Time-resolved Microwave Conductivity (TRMC), photo‐ conductivity, mobility

## **1. Introduction**

The qualities of highly efficient solar cell material are its ability to absorb light with the widest spectral range possible, high light-to-charge conversion ratio, and transport of these charges to electrodes with least losses. This seems to be the case in organometal halide perovskite (OMHP)-based solar cell as its overall power conversion efficiencies (PCEs) have risen from a

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

meagre 4% [1] to 20.1% [2] to date. Such increase has not been seen in any other solar cell technology since the conception of light harvesting. In fact, the PCE of dye-sensitized solar cell, where the OMHP was first used as a dye substitute, remained modest [2]. In recent years, spectroscopic studies on these materials have started to trickle the much-needed fundamental investigations. Most of its well-known characteristics include electron-hole diffusion lengths longer than one micrometer [3], high mobility, and very slow recombination [4]. Despite this, there is a long list of unexplained early timescale processes, which is fundamental to under‐ stating its solar cell function. For instance, details whether molecular excitons or to highly mobile charges are the initial photoproduct and how is this related to the exceptionally long diffusion lengths, remains unanswered. It is also unclear to what extent generation and recombination of mobile charges are affected by temperature. The use of metal oxide electrodes such as TiO2 and Al2O3, as electron acceptor and isolating nanoparticles (NP), respectively, and their role on its electronic properties has not yet been understood. Many groups have also recently attempted to use organic electrodes, e.g., PCBM and Spiro-OMeTAD and were able to get decent PCE. Questions on its charge transfer mechanism, timescale, and details of injection are yet to yield convincing answers.

In this chapter, an ensemble of spectroscopic techniques, primarily time-resolved terahertz spectroscopy (TRTS) and time-resolved microwave conductivity (TRMC) complemented by photoluminescence and transient absorption, was used to monitor the creation of charged species induced by photoexcitation at the sub-ps timescale and probe its fate up to a hundred of microsecond. Both techniques have been used in an assortment of solar cell technologies including organic [5,6], dye-[7], and quantum dot- sensitized [8], and inorganic [9] systems. Neat methylammonium lead iodide (MAPbI3) was used to study the intrinsic transport properties of perovskite material both at room and low temperatures. To determine how electron accepting metal and the role of NP in general, MAPbI3 were introduced to TiO2 and Al2O3, respectively. Charge transfer mechanism and the corresponding recombination dynamics when attached to organic electrodes, PCBM and Spiro-OMeTAD were also explored. Note that detailed discussion of different material contacts, their heterogeneity, and their morphologies' influence on charge carrier dynamics are presented in Ref. [10].

On light excitation, changes in conductivity from ground state (*σ*) to photo-induced or transient state (*∆σ*) are measured in the solar cell material. This conductivity when normalized with charge density yields mobility (*μ*) per charge carrier. This photoconductivity is measured at high frequencies, i.e., at the THz regime (0.1–1.5 THz) and at the GHz regime (12 GHz). As such, these high-frequency waves are propagating through free space and are able to inter‐ rogate the sample without any electrodes. The TRTS and TRMC signal size can be expressed as product of two quantities, quantum yield, and electron and hole mobilities (*μe + μh*). This product is calculated according to:

$$
\Delta \sigma = \varphi \times \left(\mu\_c + \mu\_h\right) = \frac{\Delta \sigma L}{e I\_0 F\_A}
$$

where Δσ is the measured change in photoconductivity, *L* is the thickness, *e* is the elementary charge, *I0* is the number of photons per unit area per pulse, and *FA* is the fraction of absorbed

light. A rise in the photoconductivity kinetics signals either creation of charged species, since only these photoproducts can induce changes in conductivity or an increase in mobility. On the other hand, decay in the kinetics means disappearance of these species, through recombi‐ nation, injection to lower mobility acceptor molecules, and/or decrease in mobility.

Section 2 of this chapter presents the early timescale generation mechanism of charge carrier in neat perovskite in order to understand the intrinsic property of this material. It also discusses the influence of NPs, TiO2, and Al2O3 on the mobility of the photo-induced carriers. In Section 3, a discussion on the fate of the charges when temperature is lowered will be presented and relate these results to the origin of the very long recombination time. Lastly, in Section 4, the mechanism of electron and hole injection to organic electrodes, PCBM and Spiro-OMeTAD, is reported and the role of dark carriers in the recombination dynamics is discussed.

## **2. Probing the charge carrier dynamics of intrinsic MAPbI33**

meagre 4% [1] to 20.1% [2] to date. Such increase has not been seen in any other solar cell technology since the conception of light harvesting. In fact, the PCE of dye-sensitized solar cell, where the OMHP was first used as a dye substitute, remained modest [2]. In recent years, spectroscopic studies on these materials have started to trickle the much-needed fundamental investigations. Most of its well-known characteristics include electron-hole diffusion lengths longer than one micrometer [3], high mobility, and very slow recombination [4]. Despite this, there is a long list of unexplained early timescale processes, which is fundamental to under‐ stating its solar cell function. For instance, details whether molecular excitons or to highly mobile charges are the initial photoproduct and how is this related to the exceptionally long diffusion lengths, remains unanswered. It is also unclear to what extent generation and recombination of mobile charges are affected by temperature. The use of metal oxide electrodes such as TiO2 and Al2O3, as electron acceptor and isolating nanoparticles (NP), respectively, and their role on its electronic properties has not yet been understood. Many groups have also recently attempted to use organic electrodes, e.g., PCBM and Spiro-OMeTAD and were able to get decent PCE. Questions on its charge transfer mechanism, timescale, and details of

In this chapter, an ensemble of spectroscopic techniques, primarily time-resolved terahertz spectroscopy (TRTS) and time-resolved microwave conductivity (TRMC) complemented by photoluminescence and transient absorption, was used to monitor the creation of charged species induced by photoexcitation at the sub-ps timescale and probe its fate up to a hundred of microsecond. Both techniques have been used in an assortment of solar cell technologies including organic [5,6], dye-[7], and quantum dot- sensitized [8], and inorganic [9] systems. Neat methylammonium lead iodide (MAPbI3) was used to study the intrinsic transport properties of perovskite material both at room and low temperatures. To determine how electron accepting metal and the role of NP in general, MAPbI3 were introduced to TiO2 and Al2O3, respectively. Charge transfer mechanism and the corresponding recombination dynamics when attached to organic electrodes, PCBM and Spiro-OMeTAD were also explored. Note that detailed discussion of different material contacts, their heterogeneity, and their

On light excitation, changes in conductivity from ground state (*σ*) to photo-induced or transient state (*∆σ*) are measured in the solar cell material. This conductivity when normalized with charge density yields mobility (*μ*) per charge carrier. This photoconductivity is measured at high frequencies, i.e., at the THz regime (0.1–1.5 THz) and at the GHz regime (12 GHz). As such, these high-frequency waves are propagating through free space and are able to inter‐ rogate the sample without any electrodes. The TRTS and TRMC signal size can be expressed as product of two quantities, quantum yield, and electron and hole mobilities (*μe + μh*). This

( )

where Δσ is the measured change in photoconductivity, *L* is the thickness, *e* is the elementary charge, *I0* is the number of photons per unit area per pulse, and *FA* is the fraction of absorbed

*e h*

 m m

<sup>D</sup> D=´ + =

s

0

*A L eI F*

morphologies' influence on charge carrier dynamics are presented in Ref. [10].

sj

injection are yet to yield convincing answers.

356 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

product is calculated according to:

The early time THz photoconductivity kinetics of neat MAPbI3, MAPbI3/Al2O3 and MAP‐ bI3/TiO2 is shown in Fig. 1a. For both MAPbI3 and MAPbI3/Al2O3, a two-step increase is observed. An instrument-limited rise with amplitude of about 70% is followed by 2-3 ps increase of about 30%. Tightly bound molecular excitons are, by definition, neutrally charged and hence cannot affect the conductivity of the material. This means that even if these excitons are generated, their contribution to the THz signal is nil. Therefore, the sub-ps increase in the photoconductivity kinetics comes from generation of charged species. It was reported that the binding energy of exciton in these materials may vary between 4 meV [11] to 50 meV [12]. The heterogeneity of the binding energy in a material could cause photogenerated species to dissociate at different rates. As such, the slower rise in the kinetics could be due to separation of loosely bounded charge pair becoming independent charges. Initially, charges bounded by Coulombic force require an energy that could separate them; in this case it seems enough that thermal energy, *kT,* dissociates them and only takes few ps, 2-3 ps rise in the THz kinetics of neat MAPbI3 and MAPbI3/Al2O3. Unlike the previous two samples, for MAPbI3/TiO2 the THz transient rises faster and reaches the maximum signal with a single-step instrument-limited rise time. It was reported before that light absorbers attached to metal oxide of high electron affinity, ultrafast electron injection can be expected. For example, RuN3 dye attached to TiO2 [7] and in CdSe quantum dot attached to ZnO [8] shows ultrafast injection. Therefore, the ultrafast limited rise in THz kinetics of MAPbI3/TiO2 is due to sub-ps injection of electrons from the OMHP to TiO2. Favorable alignment of energy band levels between the perovskite and metal oxide aides the separation of any remaining loosely bound electron–hole pair favoring ultrafast injection. This is the reason why there is no second-step rise seen in the THz kinetics. Shown in Fig. 1b is the THz kinetics per photon absorbed per pulse. It shows that the mobility of MAPbI3/TiO2 is 7.5 cm2 V-1s-1, about 3-4 smaller lower than the mobility in neat MAPbI3 and the MAPbI3/Al2O3, which is 20 cm2 V-1s-1. The difference in the mobility also supports the proposal that electron injection occurs in MAPbI3/TiO2. From the total mobility of neat MAPbI3, electrons disappear due to their injection to TiO2. Because of this, only the holes left in the perovskite are seen by the instrument, resulting in lower mobility.

mobility.

Therefore, the ultrafast limited rise in THz kinetics of MAPbI3/TiO2 is due to sub‐ps injection of electrons from the OMHP to TiO2. Favorable alignment of energy band levels between the perovskite and metal oxide aides the separation of any remaining loosely bound electron– hole pair favoring ultrafast injection. This is the reason why there is no additional rise seen in the THz kinetics. Shown in Fig. 1b is the THz kinetics per photon absorbed per pulse. It shows that the mobility of MAPbI3/TiO2 is 7.5 cm2V‐1s‐1, about 3‐4 smaller lower than the mobility in neat MAPbI3 and the MAPbI3/Al2O3, which is 20 cm2V‐1s‐1. The difference in the mobility also supports the proposal that electron injection occurs in MAPbI3/TiO2. From the total mobility of neat MAPbI3, electrons disappear due to their injection to TiO2. Because of

**Figure 1.** THz photoconductivity kinetics of neat MAPbI3, MAPbI3/Al2O3, and MAPbI3/TiO2. a. Normalized to 1 (�pump = 400 nm, Iexc = 1.7 x 1013 ph/cm2 per pulse), b. Normalized with *nexce*. (Reprinted with permission from [57], Copyright 2014, American Chemical Society) **Figure 1.** THz photoconductivity kinetics of neat MAPbI3, MAPbI3/Al2O3, and MAPbI3/TiO2. a. Normalized to 1 (λpump = 400 nm, Iexc = 1.7 x 1013 ph/cm2 per pulse), b. Normalized with *nexce*. (Reprinted with permission from [57], Copyright 2014, American Chemical Society)

To further confirm electron injection, optical transient absorption (TA) was used to probe the arrival of electrons to TiO2 after photoexciting the perovskite layer. The normalized TA kinetics of neat MAPbI3 and MAPbI3/Al2O3 are shown in Fig. 2a. It has a response‐limited rise (negative), and then an approximately 2 ps of further decrease, very similar to the two‐ component increase in THz photoconductivity kinetics, showing that indeed there are charges that are not created right away. The negative signal, probed 970 nm is a part of the spectral region where there is no ground state absorption, and therefore indicative of stimulated emission. This is consistent with the ground state photoluminescence (PL) emission spectra (Fig. 2b) showing intense emission from these two samples. On the other hand, an ultrafast rise with only one component with positive sign, which means absorption, is obtained in the kinetics of MAPbI3/TiO2. This is in agreement with the timescale of arrival of electrons in TiO2 [10,13] and very similar to the THz kinetics (Fig. 1a). Despite the fact that there is still some emission, the strong quenching of photoluminescence in MAPbI3/TiO2 is an additional proof that there is injection of electrons. From the SEM image of the three samples (Fig. 3), MAPbI3/TiO2 shows no indication of domains of MAPbI3 bigger than 500 nm, meaning that the mesoporous network of MAPbI3 is formed within To further confirm electron injection, optical transient absorption (TA) was used to probe the arrival of electrons to TiO2 after photoexciting the perovskite layer. The normalized TA kinetics of neat MAPbI3 and MAPbI3/Al2O3 are shown in Fig. 2a. It has a response-limited rise (nega‐ tive), and then an approximately 2 ps of further decrease, very similar to the two-component increase in THz photoconductivity kinetics, showing that indeed there are charges that are not created right away. The negative signal, probed 970 nm is a part of the spectral region where there is no ground state absorption, and therefore indicative of stimulated emission. This is consistent with the ground state photoluminescence (PL) emission spectra (Fig. 2b) showing intense emission from these two samples. On the other hand, an ultrafast rise with only one component with positive sign, which means absorption, is obtained in the kinetics of MAPbI3/ TiO2. This is in agreement with the timescale of arrival of electrons in TiO2 [10,13] and very similar to the THz kinetics (Fig. 1a). Despite the fact that there is still some emission, the strong quenching of photoluminescence in MAPbI3/TiO2 is an additional proof that there is injection of electrons. From the SEM image of the three samples (Fig. 3), MAPbI3/TiO2 shows no indication of domains of MAPbI3 bigger than 500 nm, meaning that the mesoporous network of MAPbI3 is formed within TiO2 NPs analogous to that reported in Ref. [13]. Despite this, reduced domains of MAPbI3 may still be formed brought about by voids due TiO2 NPs. However, this could be small, as shown by the strong quenching of emission and the fast rise in THz photoconductivity kinetics.

TiO2 NPs analogous to that reported in Ref. [13]. Despite this, reduced domains of MAPbI3 As shown in Fig. 1b, the first 40 ps of the THz kinectics for the three samples manifests a slow decay. It is usually assumed that at the earliest timescale, quantum yield is 1. At later timescale, the THz photoconductivity becomes a product of charge concentration and mobility since photogenerated charges are either starting to recombine or lose their mobility. In such case, it is not clear to conclude, based only on the photoconducivity kinetics, if the decay is related to depopulation of charges or lowering mobility with time or both. Transient absorption is the appropriate technique to measure the charge population as a function of time. Shown in Fig.

4a are kinetic traces of neat MAPbI3, obtained through TA and TRTS at similar pump levels (~1013 ph/cm2 per pulse). Within the signal-to-noise, the decay of the two plots is identical. This implies that the decay in the THz kinetics should be coming from the depopulation of charge carriers since the TA kinetics has the same decay. Consequently, this means that mobility of charge carrier remains the same at least up to 1 ns, otherwise a faster decay in the THz kinetics would be observed. To determine the reason for the depopulation of charges, the excitation density dependence of THz kinetics was obtained. At the lowest pump excitation (Fig. 4b, 2.0 × 1012 ph/cm2 per pulse), the kinetics remained flat until 200 ps, where mobility obtained is 25 cm2 V-1s-1. These highly mobile charges measured in THz for OMHP material are within the same order reported by Wehrenfennig, et al. [4]. At highest intensity, the decay is fastest and mobility lowest. These are strong indications that the decay is due to non-geminate recombi‐ nation of charges, similar to those reported in bulk heterojunction solar cell [5,6]. function of time. Shown in Fig. 4a are kinetic traces of neat MAPbI3, obtained through TA and TRTS at similar pump levels (~1013 ph/cm2 per pulse). Within the signal‐to‐noise, the decay of the two plots is identical. This implies that the decay in the THz kinetics should be coming from the depopulation of charge carriers since the TA kinetics has the same decay. Consequently, this means that mobility of charge carrier remains the same at least up to 1 ns, otherwise a faster decay in the THz kinetics would be observed. To determine the reason for the depopulation of charges, the excitation density dependence of THz kinetics was obtained. At the lowest pump excitation (Fig. 4b, 2.0 � 1012 ph/cm2 per pulse), the kinetics remained flat until 200 ps, where mobility obtained is 25 cm2V‐1s‐1. These highly mobile charges measured in THz for OMHP material are within the same order reported by Wehrenfennig, et al. [4]. At highest intensity, the decay is fastest and mobility lowest. These are strong indications that the decay is due to non‐geminate recombination of charges,

similar to those reported in bulk heterojunction solar cell [5,6].

timescale, the THz photoconductivity becomes a product of charge concentration and mobility since photogenerated charges are either starting to recombine or lose their mobility. In such case, it is not clear to conclude, based only on the photoconducivity kinetics, if the

Therefore, the ultrafast limited rise in THz kinetics of MAPbI3/TiO2 is due to sub‐ps injection of electrons from the OMHP to TiO2. Favorable alignment of energy band levels between the perovskite and metal oxide aides the separation of any remaining loosely bound electron– hole pair favoring ultrafast injection. This is the reason why there is no additional rise seen in the THz kinetics. Shown in Fig. 1b is the THz kinetics per photon absorbed per pulse. It shows that the mobility of MAPbI3/TiO2 is 7.5 cm2V‐1s‐1, about 3‐4 smaller lower than the mobility in neat MAPbI3 and the MAPbI3/Al2O3, which is 20 cm2V‐1s‐1. The difference in the mobility also supports the proposal that electron injection occurs in MAPbI3/TiO2. From the total mobility of neat MAPbI3, electrons disappear due to their injection to TiO2. Because of this, only the holes left in the perovskite are seen by the instrument, resulting in lower

**Figure 1.** THz photoconductivity kinetics of neat MAPbI3, MAPbI3/Al2O3, and MAPbI3/TiO2. a. Normalized to 1 (�pump = 400 nm, Iexc = 1.7 x 1013 ph/cm2 per pulse), b. Normalized with *nexce*. (Reprinted

**Figure 1.** THz photoconductivity kinetics of neat MAPbI3, MAPbI3/Al2O3, and MAPbI3/TiO2. a. Normalized to 1 (λpump =

per pulse), b. Normalized with *nexce*. (Reprinted with permission from [57], Copyright

To further confirm electron injection, optical transient absorption (TA) was used to probe the arrival of electrons to TiO2 after photoexciting the perovskite layer. The normalized TA kinetics of neat MAPbI3 and MAPbI3/Al2O3 are shown in Fig. 2a. It has a response‐limited rise (negative), and then an approximately 2 ps of further decrease, very similar to the two‐ component increase in THz photoconductivity kinetics, showing that indeed there are charges that are not created right away. The negative signal, probed 970 nm is a part of the spectral region where there is no ground state absorption, and therefore indicative of stimulated emission. This is consistent with the ground state photoluminescence (PL) emission spectra (Fig. 2b) showing intense emission from these two samples. On the other hand, an ultrafast rise with only one component with positive sign, which means absorption, is obtained in the kinetics of MAPbI3/TiO2. This is in agreement with the timescale of arrival of electrons in TiO2 [10,13] and very similar to the THz kinetics (Fig. 1a). Despite the fact that there is still some emission, the strong quenching of photoluminescence in MAPbI3/TiO2 is an additional proof that there is injection of electrons. From the SEM image of the three samples (Fig. 3), MAPbI3/TiO2 shows no indication of domains of MAPbI3 bigger than 500 nm, meaning that the mesoporous network of MAPbI3 is formed within TiO2 NPs analogous to that reported in Ref. [13]. Despite this, reduced domains of MAPbI3

As shown in Fig. 1b, the first 40 ps of the THz kinectics for the three samples manifests a slow decay. It is usually assumed that at the earliest timescale, quantum yield is 1. At later timescale, the THz photoconductivity becomes a product of charge concentration and mobility since photogenerated charges are either starting to recombine or lose their mobility. In such case, it is not clear to conclude, based only on the photoconducivity kinetics, if the decay is related to depopulation of charges or lowering mobility with time or both. Transient absorption is the appropriate technique to measure the charge population as a function of time. Shown in Fig.

To further confirm electron injection, optical transient absorption (TA) was used to probe the arrival of electrons to TiO2 after photoexciting the perovskite layer. The normalized TA kinetics of neat MAPbI3 and MAPbI3/Al2O3 are shown in Fig. 2a. It has a response-limited rise (nega‐ tive), and then an approximately 2 ps of further decrease, very similar to the two-component increase in THz photoconductivity kinetics, showing that indeed there are charges that are not created right away. The negative signal, probed 970 nm is a part of the spectral region where there is no ground state absorption, and therefore indicative of stimulated emission. This is consistent with the ground state photoluminescence (PL) emission spectra (Fig. 2b) showing intense emission from these two samples. On the other hand, an ultrafast rise with only one component with positive sign, which means absorption, is obtained in the kinetics of MAPbI3/ TiO2. This is in agreement with the timescale of arrival of electrons in TiO2 [10,13] and very similar to the THz kinetics (Fig. 1a). Despite the fact that there is still some emission, the strong quenching of photoluminescence in MAPbI3/TiO2 is an additional proof that there is injection of electrons. From the SEM image of the three samples (Fig. 3), MAPbI3/TiO2 shows no indication of domains of MAPbI3 bigger than 500 nm, meaning that the mesoporous network of MAPbI3 is formed within TiO2 NPs analogous to that reported in Ref. [13]. Despite this, reduced domains of MAPbI3 may still be formed brought about by voids due TiO2 NPs. However, this could be small, as shown by the strong quenching of emission and the fast rise

with permission from [57], Copyright 2014, American Chemical Society)

358 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

400 nm, Iexc = 1.7 x 1013 ph/cm2

2014, American Chemical Society)

in THz photoconductivity kinetics.

mobility.

**Figure 2.** a. Transient absorption kinetics (�pump = 603 nm, �probe = 970 nm, Iexc = 6.0 x 1014 ph/cm2 per pulse) b. Photoluminescence spectra (�pump = 550 nm) of the three samples. (Reprinted with permission from [57], Copyright 2014, American Chemical Society) **Figure 2.** a. Transient absorption kinetics (λpump = 603 nm, λprobe = 970 nm, Iexc = 6.0 x 1014 ph/cm2 per pulse) b. Photolu‐ minescence spectra (λpump = 550 nm) of the three samples. (Reprinted with permission from [57], Copyright 2014, American Chemical Society)

With the very high mobility of charges, it is interesting to know the nature of charge carrier transport. The THz photoconductivity spectra is a very useful tool in determining whether charges are behaving like free electron gas or in a confined, hopping‐like manner. For this reason, the THz spectra of the three samples were measured after photoexcitation, i.e., at 10 ps as shown in Fig. 5. Within the signal‐to‐noise, the amplitude and also the shape of THz photoconductivity spectra are identical for neat MAPbI3 and MAPbI3/Al2O3. However, the spectral shape of MAPbI3/TiO2 is qualitatively different as well as its signal size is lower by four times. The real part of the conductivity is positive while the imaginary part is negative. From these observations several conclusions can be drawn. First, the THz photoconductivity With the very high mobility of charges, it is interesting to know the nature of charge carrier transport. The THz photoconductivity spectra is a very useful tool in determining whether charges are behaving like free electron gas or in a confined, hopping-like manner. For this reason, the THz spectra of the three samples were measured after photoexcitation, i.e., at 10 ps as shown in Fig. 5. Within the signal-to-noise, the amplitude and also the shape of THz photoconductivity spectra are identical for neat MAPbI3 and MAPbI3/Al2O3. However, the spectral shape of MAPbI3/TiO2 is qualitatively different as well as its signal size is lower by four times. The real part of the conductivity is positive while the imaginary part is negative. From these observations several conclusions can be drawn. First, the THz photoconductivity kinetics as well as the spectra (Figs. 1a and 1b), and TA kinetics of neat MAPbI3 and MAPbI3/ Al2O3 (Fig. 2a) show that Al2O3 NPs do not change the dynamical properties and mobility of charge carriers in perovskite, at least on the timescale probed by the THz measurements (1 ns). Secondly, the favorable band energy alignment between TiO2 NPs and perovskite causes ultrafast injection, as shown in the reduction of signal size in the THz photoconductivity

kinetics (Fig.1b) and spectra (Fig. 5). As consequence of injection to TiO2, the mobility of electrons becomes the mobility of TiO2 (0.1 cm2 V-1s-1), which leads to unbalanced transport of charges. Third, sign of real and imaginary parts of conductivity is a signature of confined motion of charges [6], which is rather counterintuitive considering the very high mobility obtained here. One could hypothesize that at the spatial vicinity of photogeneration, maybe tens of nm, the morphology of the material is very much favorable for fast motion of the charges. However, at larger spatial scale, charges may start encountering scattering centers, e.g., traps, which can make its motion more restricted. The THz spectra (Fig. 5), which is normalized, decay due to non-geminate recombination as well as with excitation density, at several delay times (100–950 ps), are identical, supporting the interpretation that mobility is constant for 1 ns but with confined mode of transport. leads to unbalanced transport of charges. Third, sign of real and imaginary parts of conductivity is a signature of confined motion of charges [6], which is rather counterintuitive considering the very high mobility obtained here. One could hypothesize that at the spatial vicinity of photogeneration, maybe tens of nm, the morphology of the material is very much favorable for fast motion of the charges. However, at larger spatial scale, charges may start encountering scattering centers, e.g., traps, which can make its motion more restricted. The THz spectra (Fig. 5), which is normalized, decay due to non‐ geminate recombination as well as with excitation density, at several delay times (100–950 ps), are identical, supporting the interpretation that mobility is constant for 1 ns but with confined mode of transport.

**Figure 3.** Scanning electron microscope image of a. neat MAPbI3, b. MAPbI3/Al2O3 and c. MAPbI3/TiO2. (Reprinted with permission from [57], Copyright 2014, American Chemical Society) **Figure 3.** Scanning electron microscope image of a. neat MAPbI3, b. MAPbI3/Al2O3 and c. MAPbI3/TiO2. (Reprinted with permission from [57], Copyright 2014, American Chemical Society)

Differences in the mobilities of electrons and holes, typical with bulk heterojunction organic solar cells, which are sometimes several orders of magnitude, lowers the overall PCE due to built‐in electric field brought about by space‐charge‐limited photocurrents [14]. This is because one of the charges has already reached the electrode while the other is still traversing the active layer of the solar cell. Therefore, it is important to assess not only the total mobility of the materials but to estimate both mobility of electrons and holes. Admittedly, THz measurement on one material alone would not provide this information, but rather by comparing it with known electron or hole mobilities. The reported intrinsic electron mobility of porous TiO2 films as measured by THz was shown to be <<1 cm2V‐1s‐<sup>1</sup> [15]. As shown in Fig. 1b, the mobility obtained for MAPbI3/TiO2 is 7.5 cm2V‐1s‐1. From the discussion above, it was established that there is an ultrafast injection of electrons from perovskite to TiO2, meaning that it should only be the mobility of holes that are left in the perovskite material that is seen by the instrument. As a consequence, the measured THz mobility of ~20 cm2V‐1s‐<sup>1</sup> for MAPbI3 and MAPbI3/Al2O3, where both electrons and holes are in the perovskite material and therefore contributing to the THz signal, 12.5 cm2V‐1s‐<sup>1</sup> should be coming from electrons since 7.5 cm2V‐1s‐<sup>1</sup> is from the holes. The resulting ratio of electron and hole mobilities in the perovskite is therefore about two, in agreement with the recent Differences in the mobilities of electrons and holes, typical with bulk heterojunction organic solar cells, which are sometimes several orders of magnitude, lowers the overall PCE due to built-in electric field brought about by space-charge-limited photocurrents [14]. This is because one of the charges has already reached the electrode while the other is still traversing the active layer of the solar cell. Therefore, it is important to assess not only the total mobility of the materials but to estimate both mobility of electrons and holes. Admittedly, THz measurement on one material alone would not provide this information, but rather by comparing it with known electron or hole mobilities this could be answered. The reported intrinsic electron mobility of porous TiO2 films as measured by THz is <<1 cm2 V-1s-1 [15]. As shown in Fig. 1b, the mobility obtained for MAPbI3/TiO2 is 7.5 cm2 V-1s-1. From the discussion above, it was established that there is an ultrafast injection of electrons from perovskite to TiO2, meaning that it should only be the mobility of holes that are left in the perovskite material that is seen by the instrument. As a consequence, the measured THz mobility of ~20 cm2 V-1s-1 for MAP‐ bI3 and MAPbI3/Al2O3, where both electrons and holes are in the perovskite material and therefore contributing to the THz signal, 12.5 cm2 V-1s-1 should be coming from electrons since 7.5 cm2 V-1s-1 is from the holes. The resulting ratio of electron and hole mobilities in the perovskite is therefore about two, in agreement with the recent theoretical calculations of the relative effective masses of electrons and holes [16]. This also justifies the balanced long

theoretical calculations of the relative effective masses of electrons and holes [16]. This also justifies the balanced long diffusion lengths reported by the group of Stranks et al. [3] and diffusion lengths reported by the group of Stranks et al. [3, 17, 45]. The almost balanced electron and hole mobility is a unique key information rationalizing high PCE in OMHP-only or OMHP/Al2O3 solar cells.

kinetics (Fig.1b) and spectra (Fig. 5). As consequence of injection to TiO2, the mobility of

leads to unbalanced transport of charges. Third, sign of real and imaginary parts of conductivity is a signature of confined motion of charges [6], which is rather counterintuitive considering the very high mobility obtained here. One could hypothesize that at the spatial vicinity of photogeneration, maybe tens of nm, the morphology of the material is very much favorable for fast motion of the charges. However, at larger spatial scale, charges may start encountering scattering centers, e.g., traps, which can make its motion more restricted. The THz spectra (Fig. 5), which is normalized, decay due to non‐ geminate recombination as well as with excitation density, at several delay times (100–950 ps), are identical, supporting the interpretation that mobility is constant for 1 ns but with

charges. Third, sign of real and imaginary parts of conductivity is a signature of confined motion of charges [6], which is rather counterintuitive considering the very high mobility obtained here. One could hypothesize that at the spatial vicinity of photogeneration, maybe tens of nm, the morphology of the material is very much favorable for fast motion of the charges. However, at larger spatial scale, charges may start encountering scattering centers, e.g., traps, which can make its motion more restricted. The THz spectra (Fig. 5), which is normalized, decay due to non-geminate recombination as well as with excitation density, at several delay times (100–950 ps), are identical, supporting the interpretation that mobility is

**Figure 3.** Scanning electron microscope image of a. neat MAPbI3, b. MAPbI3/Al2O3 and c. MAPbI3/TiO2.

**Figure 3.** Scanning electron microscope image of a. neat MAPbI3, b. MAPbI3/Al2O3 and c. MAPbI3/TiO2. (Reprinted

Differences in the mobilities of electrons and holes, typical with bulk heterojunction organic solar cells, which are sometimes several orders of magnitude, lowers the overall PCE due to built‐in electric field brought about by space‐charge‐limited photocurrents [14]. This is because one of the charges has already reached the electrode while the other is still traversing the active layer of the solar cell. Therefore, it is important to assess not only the total mobility of the materials but to estimate both mobility of electrons and holes. Admittedly, THz measurement on one material alone would not provide this information, but rather by comparing it with known electron or hole mobilities. The reported intrinsic electron mobility of porous TiO2 films as measured by THz was shown to be <<1 cm2V‐1s‐<sup>1</sup> [15]. As shown in Fig. 1b, the mobility obtained for MAPbI3/TiO2 is 7.5 cm2V‐1s‐1. From the discussion above, it was established that there is an ultrafast injection of electrons from perovskite to TiO2, meaning that it should only be the mobility of holes that are left in the perovskite material that is seen by the instrument. As a consequence, the measured THz mobility of ~20 cm2V‐1s‐<sup>1</sup> for MAPbI3 and MAPbI3/Al2O3, where both electrons and holes are in the perovskite material and therefore contributing to the THz signal, 12.5 cm2V‐1s‐<sup>1</sup> should be coming from electrons since 7.5 cm2V‐1s‐<sup>1</sup> is from the holes. The resulting ratio of electron and hole mobilities in the perovskite is therefore about two, in agreement with the recent theoretical calculations of the relative effective masses of electrons and holes [16]. This also justifies the balanced long diffusion lengths reported by the group of Stranks et al. [3] and

Differences in the mobilities of electrons and holes, typical with bulk heterojunction organic solar cells, which are sometimes several orders of magnitude, lowers the overall PCE due to built-in electric field brought about by space-charge-limited photocurrents [14]. This is because one of the charges has already reached the electrode while the other is still traversing the active layer of the solar cell. Therefore, it is important to assess not only the total mobility of the materials but to estimate both mobility of electrons and holes. Admittedly, THz measurement on one material alone would not provide this information, but rather by comparing it with known electron or hole mobilities this could be answered. The reported intrinsic electron

established that there is an ultrafast injection of electrons from perovskite to TiO2, meaning that it should only be the mobility of holes that are left in the perovskite material that is seen

bI3 and MAPbI3/Al2O3, where both electrons and holes are in the perovskite material and

perovskite is therefore about two, in agreement with the recent theoretical calculations of the relative effective masses of electrons and holes [16]. This also justifies the balanced long

V-1s-1 is from the holes. The resulting ratio of electron and hole mobilities in the

by the instrument. As a consequence, the measured THz mobility of ~20 cm2

(Reprinted with permission from [57], Copyright 2014, American Chemical Society)

with permission from [57], Copyright 2014, American Chemical Society)

mobility of porous TiO2 films as measured by THz is <<1 cm2

the mobility obtained for MAPbI3/TiO2 is 7.5 cm2

therefore contributing to the THz signal, 12.5 cm2

7.5 cm2

V-1s-1), which leads to unbalanced transport of

V-1s-1 [15]. As shown in Fig. 1b,

V-1s-1 for MAP‐

V-1s-1. From the discussion above, it was

V-1s-1 should be coming from electrons since

electrons becomes the mobility of TiO2 (0.1 cm2

360 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

constant for 1 ns but with confined mode of transport.

confined mode of transport.

**Figure 4**. a. Comparison TA and TRTS kinetics for neat MAPbI3 showing similar decay rates up to 1 ns. b. Intensity dependence THz kinetics of MAPbI3/Al2O3. (Reprinted with permission from [57], Copyright 2014, American Chemical Society) **Figure 4.** a. Comparison TA and TRTS kinetics for neat MAPbI3 showing similar decay rates up to 1 ns. b. Intensity dependence THz kinetics of MAPbI3/Al2O3. (Reprinted with permission from [57], Copyright 2014, American Chemical Society)

Besides the ultrafast generation of highly mobile charges and balanced transport, another very important characteristic required for highly efficient solar cell is the timescale of **Figure 5.** Photoconductivity spectra of MAPbI3/Al2O3 at pump-probe delay of 10 ps normalized with *nexce* (line with symbols). Solid lines are spectra at 100 ps (blue), 200 ps (cyan), 600 ps (magenta), and 950 ps (gray). (Reprinted with permission from [57], Copyright 2014, American Chemical Society)

recombination. It is desirable that charge carriers do not meet since every charge pair that recombines means one charge pair is not extracted. By using the TRMC, the photoconductivity measured by TRTS can be extended from tens of ns to a hundred microseconds. In addition, due to the superior stability of MW emitter, a significant increase in signal‐to‐noise is achieved allowing the use of excitation density two orders lower than TRTS. As a result, excitation‐dependent second‐order recombination is minimized, giving more precise dynamical information on recombination of charges. Figure. 6a is the plot of TRMC photoconductivity kinetics of the three samples measured at 5.9 � 109 ph/cm2 per pulse for 1 �s. For MAPbI3, MAPbI3/Al2O3, the mobility is 3 cm2V‐1s‐1, while for MAPbI3/TiO2 it is 1 cm2V‐1s‐1. The intrinsic mobility of electron in TiO2 as measured by TRMC has been previously reported as <<0.1 cm2V‐1s‐<sup>1</sup> [18,19]. As ultrafast injection in this material has been discussed above, the 1 cm2V‐1s‐<sup>1</sup> of mobility in MAPbI3/TiO2 should be coming from holes only. Ergo, the mobility of electrons in MAPbI3 and MAPbI3/Al2O3 is 2 cm2V‐1s‐1, which is

Besides the ultrafast generation of highly mobile charges and balanced transport, another very important characteristic required for highly efficient solar cell is the timescale of recombination. It is desirable that charge carriers do not meet since every charge pair that recombines means one charge pair is not extracted. By using the TRMC, the photoconduc‐ tivity measured by TRTS can be extended from tens of ns to a hundred microseconds. In addition, due to the superior stability of MW emitter, a significant increase in signal-tonoise is achieved allowing the use of excitation density two orders lower than TRTS. As a result, excitation-dependent second-order recombination is minimized, giving more precise dynamical information on recombination of charges. Figure. 6a is the plot of TRMC photoconductivity kinetics of the three samples measured at 5.9 × 109 ph/cm2 per pulse for 1 μs. For MAPbI3and 9MAPbI3/Al2O3, the mobility is 3 cm2 V-1s-1, while for MAPbI3/TiO2 it is 1 cm2 V-1s-1. The intrinsic mobility of electron in TiO2 as measured by TRMC has been previously reported as <<0.1 cm2 V-1s-1 [18,19]. As ultrafast injection in this material has been discussed above, the 1 cm2 V-1s-1 of mobility in MAPbI3/TiO2 should be coming from holes only. Ergo, the mobility of electrons in MAPbI3 and MAPbI3/Al2O3 is 2 cm2 V-1s-1, which is consistent with the analysis of TRTS data. Such very high mobility is unique to this material especially in this later timescale since organic solar cell P3HT:PCBM, for example, has 0.045 cm2 V-1s-1 only [18]. The signal-to-noise seems worse in Fig. 6a than in Fig. 1b but one should note that there is almost three orders' difference in their excitation intensity. At this very low fluence, the TRMC photoconductivity kinetics of the three materials is rather flat, up to 1 μs. This means that neither the charge population nor the mobility is decaying in this very long timescale. A much longer time window was used to determine the onset of recombination. Figures 6b and 6c are TRMC photoconductivity kinetics of MAPbI3/Al2O3 and MAPbI3/TiO2 obtained from fluences of 5.9 × 109 –6 × 1011 ph/cm2 per pulse, which is 100 times lower than that used in TRTS and at timescale up to 100 μs. The decay is faster at higher intensities and slower with increasing lifetime and amplitude as fluence is lowered, a signature that charges are recombining non-geminately only [20,21]. This also means that there is no first-order recombination even at very low excitation, which suggests charges are diffusing rapidly away from its locus of generation aided by their very high mobility. TRMC signal is similar to photoconductivity measured in TRTS, which means that obtained response is from both charge concentration and mobility. Therefore, a single transient decay trace cannot give information whether it is charge carrier recombination or carrier relaxa‐ tion that is observed. Nevertheless, it is clear in Fig. 6a that the TRMC kinetics is flat for 1-μs time, which means both its charge population and mobility do not change and any recombination or relaxation of mobility must all be considerably slower than 1 μs. Under ambient sunlight conditions, onset of charge recombination extends to tens of μs, which is the most important conclusion from these results.

Dynamical understanding of charge carriers of the perovskite samples can now be painted from the timescale of sub-ps to a hundred of μs at a wide range of excitation fluences. At excitation densities of 1013-1014 ph/cm2 per pulse (Figs. 1, 2, 4,), second-order non-geminate recombination directly leads to decay of photoconductivity signal. At low excitation fluence, the TRTS kinetics at lowest intensity remained flat up to 1 ns, which means that the carrier

or relaxation of mobility must all be considerably slower than 1 �s. Under ambient sunlight conditions, onset of charge recombination extends to tens of �s, which is the most important Charge Carrier Dynamics in Organometal Halide Perovskite Probed by Time-Resolved Electrical Measurements http://dx.doi.org/10.5772/61631 363

conclusion from these results.

Society)

slower with increasing lifetime and amplitude as fluence is lowered, a signature that charges are recombining non‐geminately only [20,21]. This also means that there is no first‐ order recombination even at very low excitation, which suggests charges are diffusing rapidly away from its locus of generation aided by their very high mobility. TRMC signal is similar to photoconductivity measured in TRTS, which means that obtained response is from both charge concentration and mobility. Therefore, a single transient decay trace cannot give information whether it is charge carrier recombination or carrier relaxation that is observed. Nevertheless, it is clear in Fig. 6a that the TRMC kinetics is flat for 1‐�s time, which means both its charge population and mobility do not change and any recombination

Besides the ultrafast generation of highly mobile charges and balanced transport, another very important characteristic required for highly efficient solar cell is the timescale of recombination. It is desirable that charge carriers do not meet since every charge pair that recombines means one charge pair is not extracted. By using the TRMC, the photoconduc‐ tivity measured by TRTS can be extended from tens of ns to a hundred microseconds. In addition, due to the superior stability of MW emitter, a significant increase in signal-tonoise is achieved allowing the use of excitation density two orders lower than TRTS. As a result, excitation-dependent second-order recombination is minimized, giving more precise dynamical information on recombination of charges. Figure. 6a is the plot of TRMC photoconductivity kinetics of the three samples measured at 5.9 × 109 ph/cm2 per pulse for

V-1s-1. The intrinsic mobility of electron in TiO2 as measured by TRMC has been

consistent with the analysis of TRTS data. Such very high mobility is unique to this material especially in this later timescale since organic solar cell P3HT:PCBM, for example, has 0.045

100 times lower than that used in TRTS and at timescale up to 100 μs. The decay is faster at higher intensities and slower with increasing lifetime and amplitude as fluence is lowered, a signature that charges are recombining non-geminately only [20,21]. This also means that there is no first-order recombination even at very low excitation, which suggests charges are diffusing rapidly away from its locus of generation aided by their very high mobility. TRMC signal is similar to photoconductivity measured in TRTS, which means that obtained response is from both charge concentration and mobility. Therefore, a single transient decay trace cannot give information whether it is charge carrier recombination or carrier relaxa‐ tion that is observed. Nevertheless, it is clear in Fig. 6a that the TRMC kinetics is flat for 1-μs time, which means both its charge population and mobility do not change and any recombination or relaxation of mobility must all be considerably slower than 1 μs. Under ambient sunlight conditions, onset of charge recombination extends to tens of μs, which is

Dynamical understanding of charge carriers of the perovskite samples can now be painted from the timescale of sub-ps to a hundred of μs at a wide range of excitation fluences. At

recombination directly leads to decay of photoconductivity signal. At low excitation fluence, the TRTS kinetics at lowest intensity remained flat up to 1 ns, which means that the carrier

V-1s-1 only [18]. The signal-to-noise seems worse in Fig. 6a than in Fig. 1b but one should note that there is almost three orders' difference in their excitation intensity. At this very low fluence, the TRMC photoconductivity kinetics of the three materials is rather flat, up to 1 μs. This means that neither the charge population nor the mobility is decaying in this very long timescale. A much longer time window was used to determine the onset of recombination. Figures 6b and 6c are TRMC photoconductivity kinetics of MAPbI3/Al2O3

V-1s-1 [18,19]. As ultrafast injection in this material has been

V-1s-1 of mobility in MAPbI3/TiO2 should be coming from holes

V-1s-1, while for MAPbI3/TiO2 it

–6 × 1011 ph/cm2 per pulse, which is

per pulse (Figs. 1, 2, 4,), second-order non-geminate

V-1s-1, which is

1 μs. For MAPbI3and 9MAPbI3/Al2O3, the mobility is 3 cm2

362 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

and MAPbI3/TiO2 obtained from fluences of 5.9 × 109

the most important conclusion from these results.

excitation densities of 1013-1014 ph/cm2

only. Ergo, the mobility of electrons in MAPbI3 and MAPbI3/Al2O3 is 2 cm2

is 1 cm2

cm2

previously reported as <<0.1 cm2

discussed above, the 1 cm2

**Figure 6**. Time‐resolved microwave conductivity kinetics of a. MAPbI3, MAPbI3/Al2O3, and MAPbI3/TiO2 measured at 5.9 � 109 ph/cm2 per pulse for 1 �s and b. at different excitation densities measured up to 100 �s. (Reprinted with permission from [57], Copyright 2014, American Chemical **Figure 6.** Time-resolved microwave conductivity kinetics of a. MAPbI3, MAPbI3/Al2O3, and MAPbI3/TiO2 measured at 5.9 × 109 ph/cm2 per pulse for 1 μs and b. at different excitation densities measured up to 100 μs. (Reprinted with per‐ mission from [57], Copyright 2014, American Chemical Society)

mobility and its population occurs on a timescale longer than 1 μs, as shown by TRMC traces. Moreover, this means that charges can move over large distances allowing them to be extracted at the electrodes. Dynamical understanding of charge carriers of the perovskite samples can now be painted from the timescale of sub‐ps to a hundred of �s at a wide range of excitation fluences. At

As a summary of this section, the combination of several time-resolved spectroscopy techni‐ ques, i.e., optical transient absorption, TRTS, and TRMC, characterization of the charge carrier dynamics in neat MAPbI3, and MAPbI3 attached with TiO2 and Al2O3 was presented. Thin film of perovskite and that attached to Al2O3 have properties of an ideal solar cell device: Electron and holes are formed in sub-ps timescale with very high mobilities. Its estimated values suggest that they are almost balanced and carriers do not recombine until after tens of microseconds. These characteristics almost guarantee efficient charge collection, a very desirable property of solar cell device. Electron injection to TiO2 lowers electron mobility resulting to unbalanced charge transport that could lead to built-in electric field. Engineering of electrodes such that balanced transport is still achieved is one possible way of improving its overall power conversion efficiency. excitation densities of 1013‐1014 ph/cm2 per pulse (Figs. 1, 2, 4,), second‐order non‐geminate recombination directly leads to decay of photoconductivity signal. At low excitation fluence, the TRTS kinetics at lowest intensity remained flat up to 1 ns, which means that the carrier

## **3. Influence of temperature on the charge dynamics**

One of the key issues that need to be addressed in studying this class of materials is its dependence on temperature. This is due to the fact that until now there is a disagreement on how the charges are actually generated. As discussed in the previous section, at room tem‐ perature the heterogeneity in the exciton binding energy of the material may lead to different rates of charge dissociation. In such case, lowering the temperature should give an indication if it is indeed by excitonic means that charges are created or is it like bulk crystalline silicon that has band-to-band transition. It has been argued that on optical absorption of a photon with energy exceeding the bandgap, an electron is promoted conduction band leaving a hole in the valence band, thereby forming a correlated charge species or exciton. In the presence of, for example, TiO2 NP electrons into the conduction band is injected thereby freeing exciton. Similar mechanism is found in hole transfer to Spiro-OMeTAD [23-25]. This is in line with the relatively high binding energies (*EB*) estimated to be in the range of 19 and 45 meV for MAPbI3 [26,27]. However, by replacing the TiO2 with insulating Al2O3, where no injection is observed (see discussion above) and therefore exciton remained bound, similar high PCE has been achieved by Snaith et al. In fact, it was shown that it works efficiently well in a flat p-i-n configuration, where perovskite active layer serves as both light absorber, charge generation site and transporter [13,28,29]. Moreover, as presented in Section 1, both the TRTS and TRMC obtained very high mobility in the absence of electron or hole transporting electrodes, i.e., for neat MAPbI3 and MAPbI3/Al2O3. Due to these results, questions arise on the excitonic charac‐ teristics of the initial photoproduct. Crystallographic data show two-phase transitions, orthorhombic to tetragonal at about 160 K and tetragonal to cubic transition at 330 K [30-32]. By using PL and TRMC, exciton dissociation, charge carrier generation, and recombination is explored from 80 K to 300 K to elucidate the role of binding energy in determining the ratio between bound electron hole pairs and mobile charges. PL probes the emission from bound electron hole pairs, while TRMC probes unbound charges.

Shown in Fig. 7 is the plot of the peak emission intensity of MAPbI3 as a function of tempera‐ ture. A decrease in temperature results in a more intense emission which is a strong indication that generation of charges is thermally activated. At room temperature, the maximum number of mobile charges generated is reached where emission is at minimum. Using equation 1 of Ref. [33], an activation energy (EB) of 32 ± 5 meV is obtained. It should be noted though that this EB is specific to this particular sample and could vary depending on the preparation conditions as well as according to the technique used to probe it. In comparison with organic solar cell (EB=0.3 eV-0.4 eV) [34], the binding energy obtained here is at least ten times lower and very similar to that of silicon, 15 meV. If one assumes that photoexcitation results only in emission or in charge carriers, then the resulting fit in Fig. 7 represents the charge carrier yield as a function of temperature as shown by the black trace.

Figure 8 are plots of TRMC kinetics for MAPbI3/Al2O3 normalized with excitation intensity varied over a factor of 50 and measured at 165 K, 240 K, and 300 K. On the one hand, the fastest decay is observed at highest excitation intensity of the traces at 300 K, implying second-order recombination. On the other hand, at low intensities the lifetime of the charge carriers exceeds

**3. Influence of temperature on the charge dynamics**

364 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

electron hole pairs, while TRMC probes unbound charges.

as a function of temperature as shown by the black trace.

One of the key issues that need to be addressed in studying this class of materials is its dependence on temperature. This is due to the fact that until now there is a disagreement on how the charges are actually generated. As discussed in the previous section, at room tem‐ perature the heterogeneity in the exciton binding energy of the material may lead to different rates of charge dissociation. In such case, lowering the temperature should give an indication if it is indeed by excitonic means that charges are created or is it like bulk crystalline silicon that has band-to-band transition. It has been argued that on optical absorption of a photon with energy exceeding the bandgap, an electron is promoted conduction band leaving a hole in the valence band, thereby forming a correlated charge species or exciton. In the presence of, for example, TiO2 NP electrons into the conduction band is injected thereby freeing exciton. Similar mechanism is found in hole transfer to Spiro-OMeTAD [23-25]. This is in line with the relatively high binding energies (*EB*) estimated to be in the range of 19 and 45 meV for MAPbI3 [26,27]. However, by replacing the TiO2 with insulating Al2O3, where no injection is observed (see discussion above) and therefore exciton remained bound, similar high PCE has been achieved by Snaith et al. In fact, it was shown that it works efficiently well in a flat p-i-n configuration, where perovskite active layer serves as both light absorber, charge generation site and transporter [13,28,29]. Moreover, as presented in Section 1, both the TRTS and TRMC obtained very high mobility in the absence of electron or hole transporting electrodes, i.e., for neat MAPbI3 and MAPbI3/Al2O3. Due to these results, questions arise on the excitonic charac‐ teristics of the initial photoproduct. Crystallographic data show two-phase transitions, orthorhombic to tetragonal at about 160 K and tetragonal to cubic transition at 330 K [30-32]. By using PL and TRMC, exciton dissociation, charge carrier generation, and recombination is explored from 80 K to 300 K to elucidate the role of binding energy in determining the ratio between bound electron hole pairs and mobile charges. PL probes the emission from bound

Shown in Fig. 7 is the plot of the peak emission intensity of MAPbI3 as a function of tempera‐ ture. A decrease in temperature results in a more intense emission which is a strong indication that generation of charges is thermally activated. At room temperature, the maximum number of mobile charges generated is reached where emission is at minimum. Using equation 1 of Ref. [33], an activation energy (EB) of 32 ± 5 meV is obtained. It should be noted though that this EB is specific to this particular sample and could vary depending on the preparation conditions as well as according to the technique used to probe it. In comparison with organic solar cell (EB=0.3 eV-0.4 eV) [34], the binding energy obtained here is at least ten times lower and very similar to that of silicon, 15 meV. If one assumes that photoexcitation results only in emission or in charge carriers, then the resulting fit in Fig. 7 represents the charge carrier yield

Figure 8 are plots of TRMC kinetics for MAPbI3/Al2O3 normalized with excitation intensity varied over a factor of 50 and measured at 165 K, 240 K, and 300 K. On the one hand, the fastest decay is observed at highest excitation intensity of the traces at 300 K, implying second-order recombination. On the other hand, at low intensities the lifetime of the charge carriers exceeds

**Figure 7.** Temperature dependence of PL intensity of MAPbI3 (black circles, λexc = 550 nm, λdec = 760 nm). Dashed line is exponential fit, while full line represents the yield of charges on assuming that photoexcitations lead either to radiative decay or to generation of charges. (Reprinted with permission from [33], Copyright 2014, American Chemical Society)

5 μs. Since the decay of TRMC kinetics is a lot slower than corresponding luminescence lifetimes (10 ns) [3,35] at similar intensities, it means that despite the radiative recombination in the ns timescale, there are long-lived charge carriers that survived and remained mobile. More importantly, the mobility of the charges is strongly dependent on the temperature, i.e., from 10.0 cm2 V-1s-1 at 165 K to 7.2 cm2 V-1s-1 at 240 K and 4.7 cm2 V-1s-1 at 300 K. If at these temperatures the changes in the photoconductivity are caused by the change in the yield of charges, then it should have been observed in the PL yield in Fig. 7. Hence, one can assign the changes as due to mobility only. The dependence of mobility on temperature is attributed to quieting of phonon at low temperature with T-1.6 dependence, very similar to that of silicon [36]. An increase in mobility usually entails that the second-order recombination would be more efficient since there is higher probability of charges to meet. However, this is not the case in this material. With the increase of mobility, second-order recombination rate diminishes by a factor of 6, which is more than 100 times slower than the reported Langevin recombination, i.e., diffusion controlled recombination [4]. This is a strong indication that there is a barrier in recombination, which should be a temperature-activated process, i.e., meeting of electrons and holes does not necessarily result in recombination. Calculation results yielded a thermal activation over an energy barrier estimated to be ~75 meV [33]. There are two possible origins of this behavior currently reported in the literature: (1) due to the induced dipole moment brought about by the intermittent rotation of MA ions [30,37] and the (2) preferential spatial localization of charge carriers in different parts/materials of the perovskite unit cell. Density functional theory calculations have revealed that 6s- and 5p-orbitals of lead and iodine, respectively, consist of the maxima of the valence band, while 6p-orbitals of lead is where conduction band minima is mostly incorporated [38].

Shown in Fig. 7 is the plot of the peak emission intensity of MAPbI3 as a function of temperature. A decrease in temperature results in a more intense emission which is a strong indication that generation of charges is thermally activated. At room temperature, the maximum number of mobile charges generated is reached where emission is at minimum. Using equation 1 of Ref. [33], an activation energy (EB) of 32 ��5 meV is obtained. It should be noted though that this EB is specific to this particular sample and could vary depending on the preparation conditions as well as according to the technique used to probe it. In comparison with organic solar cell (EB=0.3 eV‐0.4 eV) [34], the binding energy obtained here

photoexcitation results only in emission or in charge carriers, then the resulting fit in Fig. 7 represents the charge carrier yield as a function of temperature as shown by the black trace.

Figure 8. TRMC kinetics of MAPbI3/Al2O3 at several excitation intensities measured at 165 K, 240 K, and 300 K (λexc = 410 nm). (Reprinted with permission from [33], Copyright 2014, American Chemical **Figure 8.** TRMC kinetics of MAPbI3/Al2O3 at several excitation intensities measured at 165 K, 240 K, and 300 K (λexc = 410 nm). (Reprinted with permission from [33], Copyright 2014, American Chemical Society)

Temperature dependent behavior of MAPbI3 is presented in this section. Results show that this material has very similar characteristics as silicon, i.e. it has low exciton binding energy (32 meV)and its mobility has T-1.6 dependence brought about by quieting of phonons. More importantly, charges carrier do not recombine when it they meet but instead require a ther‐ mal activation over an energy barrier of 75 meV, allowing the charges to be collected at the electrodes efficiently. Figure 8 are plots of TRMC kinetics for MAPbI3/Al2O3 normalized with excitation intensity varied over a factor of 50 and measured at 165 K, 240 K, and 300 K. On the one hand, the fastest decay is observed at highest excitation intensity of the traces at 300 K, implying second‐order recombination. On the other hand, at low intensities the lifetime of the charge carriers exceeds 5 μs. Since the decay of TRMC kinetics is a lot slower than corresponding

luminescence lifetimes (10 ns) [3,35] at similar intensities, it means that despite the radiative

## **4. Role of dark carriers** recombination in the ns timescale, there are long‐lived charge carriers that survived and

Society)

In this section, the intrinsic property of MAPbI3 is examined when organic electrodes, PCBM and Spiro-OMeTAD are deposited on top of the neat perovskite material. From these samples, the timescale of injection of charges and the dynamics of its recombination are analyzed. Plotted in Fig. 9 are the TRTS kinetics of neat MAPbI3, MAPbI3/PCBM, and MAPbI3/Spiro-OMeTAD normalized with excitation density. At the earliest timescale, the mobility is found to be ~15 cm2 V-1s-1 for this particular neat MAPbI3 sample. This mobility slightly decreased

after 1 ns due to second-order recombination, similar to that shown in Fig. 4b. In contrast, the decay of MAPbI3/PCBM is faster, down to almost a third at the same timescale and initial mobility. Such decay is more clearly seen at 7 ns time window in Fig. 9b. For MAPbI3/Spiro-OMeTAD, the initial mobility is 5 cm2 V-1s-1, i.e., three times smaller than the neat but remains flat up to 1 ns. At the later timescale, i.e., hundreds of ns to us (Fig. 10a), rapid decay is still seen in MAPbI3/PCBM, while neat MAPbI3 and MAPbI3/Spiro-OMeTAD have very similar, if not identical, slower decay. In Fig.10b, the TRMC kinetics is normalized with excitation density and the corresponding mobility is plotted versus incident photon flux. For the neat MAPbI3, mobility decreases on increasing fluences (>1011 ph/cm2 ) due to second-order recombination. This behavior is also seen in the bilayer samples, but for MAPbI3/PCBM, the threshold for this process is shifted to higher fluence (>1012 ph/cm2 ). decreased after 1 ns due to second‐order recombination, similar to that shown in Fig. 4b. In contrast, the decay of MAPbI3/PCBM is faster, down to almost a third at the same timescale and initial mobility. Such decay is more clearly seen at 7 ns time window in Fig. 9b. For MAPbI3/Spiro‐OMeTAD, the initial mobility is 5 cm2/Vs, i.e., three times smaller than the neat but remains flat up to 1 ns. At the later timescale, i.e., hundreds of ns to us (Fig. 10a), rapid decay is still seen in MAPbI3/PCBM, while neat MAPbI3 and MAPbI3/Spiro‐OMeTAD have very similar, if not identical, slower decay. In Fig.10b, the TRMC kinetics is normalized with excitation density and the corresponding mobility is plotted versus incident photon flux. For the neat MAPbI3, mobility decreases on increasing fluences (>1011 ph/cm2) due to second‐order recombination. This behavior is also seen in the bilayer samples, but for **Role of dark carriers**  In this section, the intrinsic property of MAPbI3 is examined when organic electrodes, PCBM and Spiro‐OMeTAD are deposited on top of the neat perovskite material. From these samples, the timescale of injection of charges and the dynamics of its recombination are analyzed. Plotted in Fig. 9 are the TRTS kinetics of neat MAPbI3, MAPbI3/PCBM, and MAPbI3/Spiro‐OMeTAD normalized with excitation density. At the earliest timescale, the mobility is found to be ~15 cm2/Vs for this particular neat sample. This mobility slightly decreased after 1 ns due to second‐order recombination, similar to that shown in Fig. 4b. In contrast, the decay of MAPbI3/PCBM is faster, down to almost a third at the same timescale and initial mobility. Such decay is more clearly seen at 7 ns time window in Fig. 9b. For

MAPbI3/PCBM, the threshold for this process is shifted to higher fluence (>1012 ph/cm2).

MAPbI3/Spiro‐OMeTAD, the initial mobility is 5 cm2/Vs, i.e., three times smaller than the

carriers in different parts/materials of the perovskite unit cell. Density functional theory calculations have revealed that 6s‐ and 5p‐orbitals of lead and iodine, respectively, consist of the maxima of the valence band, while 6p‐orbitals of lead is where conduction band minima

In this section, the intrinsic property of MAPbI3 is examined when organic electrodes, PCBM and Spiro‐OMeTAD are deposited on top of the neat perovskite material. From these samples, the timescale of injection of charges and the dynamics of its recombination are analyzed. Plotted in Fig. 9 are the TRTS kinetics of neat MAPbI3, MAPbI3/PCBM, and

carriers in different parts/materials of the perovskite unit cell. Density functional theory calculations have revealed that 6s‐ and 5p‐orbitals of lead and iodine, respectively, consist of the maxima of the valence band, while 6p‐orbitals of lead is where conduction band minima

mobility is found to be ~15 cm2/Vs for this particular neat sample. This mobility slightly

is mostly incorporated [38].

**Role of dark carriers** 

Figure 9. a. TRTS kinetics of neat MAPbI3, MAPbI3/PCBM, and MAPbI3/Spiro‐OMeTAD normalized to the excitation intensity (*�pump* <sup>=</sup> <sup>590</sup> nm). b. TRTS kinetics with <sup>7</sup> ns time window. **Figure 9.** a. TRTS kinetics of neat MAPbI3, MAPbI3/PCBM, and MAPbI3/Spiro-OMeTAD normalized to the excitation intensity (*λpump* = 590 nm). b. TRTS kinetics with 7 ns time window. Figure 9. a. TRTS kinetics of neat MAPbI3, MAPbI3/PCBM, and MAPbI3/Spiro‐OMeTAD normalized to the excitation intensity (*�pump* = 590 nm). b. TRTS kinetics with 7 ns time window.

Temperature dependent behavior of MAPbI3 is presented in this section. Results show that this material has very similar characteristics as silicon, i.e. it has low exciton binding energy (32 meV)and its mobility has T-1.6 dependence brought about by quieting of phonons. More importantly, charges carrier do not recombine when it they meet but instead require a ther‐ mal activation over an energy barrier of 75 meV, allowing the charges to be collected at the

Figure 8 are plots of TRMC kinetics for MAPbI3/Al2O3 normalized with excitation intensity varied over a factor of 50 and measured at 165 K, 240 K, and 300 K. On the one hand, the fastest decay is observed at highest excitation intensity of the traces at 300 K, implying second‐order recombination. On the other hand, at low intensities the lifetime of the charge carriers exceeds 5 μs. Since the decay of TRMC kinetics is a lot slower than corresponding luminescence lifetimes (10 ns) [3,35] at similar intensities, it means that despite the radiative recombination in the ns timescale, there are long‐lived charge carriers that survived and

Figure 8. TRMC kinetics of MAPbI3/Al2O3 at several excitation intensities measured at 165 K, 240 K, and 300 K (λexc = 410 nm). (Reprinted with permission from [33], Copyright 2014, American Chemical

**Figure 8.** TRMC kinetics of MAPbI3/Al2O3 at several excitation intensities measured at 165 K, 240 K, and 300 K (λexc =

410 nm). (Reprinted with permission from [33], Copyright 2014, American Chemical Society)

Shown in Fig. 7 is the plot of the peak emission intensity of MAPbI3 as a function of temperature. A decrease in temperature results in a more intense emission which is a strong indication that generation of charges is thermally activated. At room temperature, the maximum number of mobile charges generated is reached where emission is at minimum. Using equation 1 of Ref. [33], an activation energy (EB) of 32 ��5 meV is obtained. It should be noted though that this EB is specific to this particular sample and could vary depending on the preparation conditions as well as according to the technique used to probe it. In comparison with organic solar cell (EB=0.3 eV‐0.4 eV) [34], the binding energy obtained here is at least ten times lower and very similar to that of silicon, 15 meV. If one assumes that photoexcitation results only in emission or in charge carriers, then the resulting fit in Fig. 7 represents the charge carrier yield as a function of temperature as shown by the black trace.

366 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

In this section, the intrinsic property of MAPbI3 is examined when organic electrodes, PCBM and Spiro-OMeTAD are deposited on top of the neat perovskite material. From these samples, the timescale of injection of charges and the dynamics of its recombination are analyzed. Plotted in Fig. 9 are the TRTS kinetics of neat MAPbI3, MAPbI3/PCBM, and MAPbI3/Spiro-OMeTAD normalized with excitation density. At the earliest timescale, the mobility is found

V-1s-1 for this particular neat MAPbI3 sample. This mobility slightly decreased

electrodes efficiently.

Society)

to be ~15 cm2

**4. Role of dark carriers**

**Figure 10.** a. TRMC traces for the three samples recorded at an excitation intensity of 1.0 × 1011 photons/cm2 per pulse (*λpump* = 600 nm) and normalized to unity. b. Mobility versus incident intensity for neat MAPbI3 and bilayers.

It has been shown that Spiro-OMeTAD is a good hole transporting material (HTM) as it has been extensively used in solar cell devices. One of its favorable characteristics is its 0.5 eV valence band offset with respect to MAPbI3 [23,39,40,41]. However, in order to extract decent PCE, additives like lithium bis(trifluoromethanesulfonyl) imide (LiTFSI) are necessary, which enables this HTM to substantially increase its very low intrinsic hole mobility, i.e., 10-8 S/cm [25,42]. Using TRTS, the mobility of MAPbI3/Spiro-OMeTAD is found to be three times lower (5 cm2 V-1s-1) than in neat MAPbI3 (15 cm2 V-1s-1) (see Fig. 9a). In agreement with this measure‐ ment, the mobility measured by TRMC has similar decrease, i.e., from 9 cm2 V-1s-1 in neat MAPbI3 to 3.5 cm2 V-1s-1 in MAPbI3/Spiro-OMeTAD (Fig. 10b). The decrease in the mobility can be interpreted as disappearance of either electrons or holes in the neat MAPbI3 as a result of charge transfer. On the basis that Spiro-OMeTAD is an HTM, it should therefore be the holes that disappeared and were injected. The hole transfer is confirmed by both techniques and the timescale of injection as seen by TRTS is sub-ps. This suggests that the energy offset at the interface alone is sufficient to allow efficient sub-ps hole injection despite the fact that in this sample no additive was added. This finding consequently means that the mobility of 5 cm2 V-1s-1 in TRTS and 3.5 cm2 V-1s-1 in TRMC should have originated from electrons that are left in MAPbI3. This process is schematically shown in Figure 11. From the initial 15 cm2 V-1s-1 in neat MAPbI3 and 5 cm2 V-1s-1 in MAPbI3/Spiro-OMeTAD (both from TRTS), this leads to a hole mobility of 10 cm2 V-1s-1 in the perovskite, implying that holes should have diffused at least 30 nm within 1 ps, which is consistent with a sub-ps injection time. Piatkowski et al. also recently reported that timescale of hole injection is 0.7 ps as measured by transient absorption spec‐ troscopy [43].

It has been reported that that charge transfer over an interface typically leads to different decay kinetics as compared to dynamics of the carriers generated in a single semiconductor [44]. However, in the case of MAPbI3/Spiro-OMeTAD, although hole injection is confirmed by TRTS, its decay in TRMC is similar to the neat MAPbI3. In fact, even at different excitation intensity their decay is still very similar as shown in Fig. 12a. The identical TRMC kinetics imply that their decay pathways must be identical, if not very similar. It was previously reported that depending on the preparation conditions, perovskite could either be an n-doped or a p-doped semiconductor. For example, in the work of Leijtens et al., an n-type perovskite material was obtained when deposited to Al2O3 NPs [45]. In contrast, calculation of Shi et al. [46] suggests that p-type is usually obtained; while Kim et al. [47] reported that it can be controlled as n- or p-type depending on the defects. It is therefore not unrealistic, at least for a moment, to postulate that perovskite measured here has concentration of holes already in the dark (*p0*), i.e., that MAPbI3 is an unintentionally doped, p-type semiconductor. This would mean that once carriers are photogenerated and holes transferred to Spiro-OMeTAD, electrons left in the conduction band of MAPbI3 will recombine with both, the dark holes in the valence band of MAPbI3 and the photogenerated holes injected into Spiro-OMeTAD. So long as the concentration of photogenerated electrons is smaller than the total concentration of holes (dark and light-induced carriers), the electron hole recombination kinetics in MAPbI3 should be barely dependent on whether there is hole injection or not. This scenario would then result in conductivity decay very similar to neat MAPbI3. Since the obtained kinetics are identical, it can be surmised that the perovskite sample measured here is p-type. It should also be stressed

Charge Carrier Dynamics in Organometal Halide Perovskite Probed by Time-Resolved Electrical Measurements http://dx.doi.org/10.5772/61631 369

It has been shown that Spiro-OMeTAD is a good hole transporting material (HTM) as it has been extensively used in solar cell devices. One of its favorable characteristics is its 0.5 eV valence band offset with respect to MAPbI3 [23,39,40,41]. However, in order to extract decent PCE, additives like lithium bis(trifluoromethanesulfonyl) imide (LiTFSI) are necessary, which enables this HTM to substantially increase its very low intrinsic hole mobility, i.e., 10-8 S/cm [25,42]. Using TRTS, the mobility of MAPbI3/Spiro-OMeTAD is found to be three times lower

be interpreted as disappearance of either electrons or holes in the neat MAPbI3 as a result of charge transfer. On the basis that Spiro-OMeTAD is an HTM, it should therefore be the holes that disappeared and were injected. The hole transfer is confirmed by both techniques and the timescale of injection as seen by TRTS is sub-ps. This suggests that the energy offset at the interface alone is sufficient to allow efficient sub-ps hole injection despite the fact that in this sample no additive was added. This finding consequently means that the mobility of 5

in MAPbI3. This process is schematically shown in Figure 11. From the initial 15 cm2

nm within 1 ps, which is consistent with a sub-ps injection time. Piatkowski et al. also recently reported that timescale of hole injection is 0.7 ps as measured by transient absorption spec‐

It has been reported that that charge transfer over an interface typically leads to different decay kinetics as compared to dynamics of the carriers generated in a single semiconductor [44]. However, in the case of MAPbI3/Spiro-OMeTAD, although hole injection is confirmed by TRTS, its decay in TRMC is similar to the neat MAPbI3. In fact, even at different excitation intensity their decay is still very similar as shown in Fig. 12a. The identical TRMC kinetics imply that their decay pathways must be identical, if not very similar. It was previously reported that depending on the preparation conditions, perovskite could either be an n-doped or a p-doped semiconductor. For example, in the work of Leijtens et al., an n-type perovskite material was obtained when deposited to Al2O3 NPs [45]. In contrast, calculation of Shi et al. [46] suggests that p-type is usually obtained; while Kim et al. [47] reported that it can be controlled as n- or p-type depending on the defects. It is therefore not unrealistic, at least for a moment, to postulate that perovskite measured here has concentration of holes already in the dark (*p0*), i.e., that MAPbI3 is an unintentionally doped, p-type semiconductor. This would mean that once carriers are photogenerated and holes transferred to Spiro-OMeTAD, electrons left in the conduction band of MAPbI3 will recombine with both, the dark holes in the valence band of MAPbI3 and the photogenerated holes injected into Spiro-OMeTAD. So long as the concentration of photogenerated electrons is smaller than the total concentration of holes (dark and light-induced carriers), the electron hole recombination kinetics in MAPbI3 should be barely dependent on whether there is hole injection or not. This scenario would then result in conductivity decay very similar to neat MAPbI3. Since the obtained kinetics are identical, it can be surmised that the perovskite sample measured here is p-type. It should also be stressed

ment, the mobility measured by TRMC has similar decrease, i.e., from 9 cm2

V-1s-1) (see Fig. 9a). In agreement with this measure‐

V-1s-1 in MAPbI3/Spiro-OMeTAD (Fig. 10b). The decrease in the mobility can

V-1s-1 in TRMC should have originated from electrons that are left

V-1s-1 in MAPbI3/Spiro-OMeTAD (both from TRTS), this leads to a hole

V-1s-1 in the perovskite, implying that holes should have diffused at least 30

V-1s-1 in neat

V-1s-1 in

(5 cm2

cm2

MAPbI3 to 3.5 cm2

V-1s-1 in TRTS and 3.5 cm2

neat MAPbI3 and 5 cm2

mobility of 10 cm2

troscopy [43].

V-1s-1) than in neat MAPbI3 (15 cm2

368 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 11.** Schematic diagram of injection and recombination dynamics in from neat MAPbI3 to PCBM and Spiro-OMe‐ TAD.

Figure 12. Excitation intensity depdendence of TRMC kinetics of (a). MAPbI3/Spiro-OMeTAD and (b). MAPbI3/PCBM. **Figure 12.** Excitation intensity dependence of TRMC kinetics of (a) MAPbI3/Spiro-OMeTAD and (b) MAPbI3/PCBM.

that both TRTS and TRMC probe only the change in conductivity due to optical excitation but not the dark conductivity.

For electron acceptors, PCBM has been extensively used for organic solar cells and lately has been also utilized for perovskite-based solar cells as well. Despite its small energy offset with respect to perovskite, the reported power conversion efficiency is over 10% [48,49]. This indicates that electrons are transferred from the perovskite to PCBM, which is schematically represented as process 2 in Fig. 11. The electron mobility in PCBM is quite small, i.e., 10-3 cm2 V-1s-1 [49-53], and therefore contributing insignificantly to the measured mobility in both TRTS and TRMC. Hence, the measured signal in Figs. 9 and 10b, only represent the mobile photogenerated charges in the perovskite. This implies that the decay measured in photocon‐ ductivity kinetics could represent not only charge recombination within the MAPbI3 (process 1), electron injection from MAPbI3 to PCBM (process 2), but the recombination at MAPbI3/ PCBM interface (process 3) as well. To disentangle the contributions of these processes, one can consider this qualitative consideration: that electron injection (process 2) is much faster than interfacial recombination between holes left in the perovskite and electrons injected to For electron acceptors, PCBM has been extensively used for organic solar cells and lately has been also utilized for perovskite-based solar cells as well. Despite its small energy offset with respect to perovskite, the reported power conversion efficiency is over 10% [48,49]. This indicates that electrons are transferred from the perovskite to PCBM, which is schematically represented as process 2 in Fig. 11. The electron mobility in PCBM is quite small, i.e., 10-3 cm<sup>2</sup>/Vs [49-53], and therefore contributing insignificantly to the measured mobility in both TRTS and TRMC. Hence, the measured signal in Figs. 9 and 10b, only represent the mobile photogenerated charges in the perovskite. This implies that the decay measured in photoconductivity kinetics could represent not only charge recombination within the MAPbI3 (process 1), electron injection from MAPbI3 to PCBM (process 2), but the recombination at MAPbI3/PCBM interface (process 3) as well. To disentangle the contributions of these processes, one can consider this qualitative consideration:; that

electron injection (process 2) is much faster than interfacial recombination between holes left in the perovskite and electrons injected to PCBM (process 3). In this scenario, a clear nonexponential, slower decay is expected since mobile electrons in MAPbI3 become immobile in PCBM. However, since there is no such plateau is observed, (Fig. 9b), it can be concluded that the rate constant for the interfacial electron hole recombination (process 3) is similar to, or exceeds, the electron injection rate, which is consistent with the fact that the difference in

At the earliest timescale of the THz kinetics, MAPbI3/PCBM bilayer and neat MAPbI3 have the same mobility (~15 cm<sup>2</sup>/Vs,; Figs. 9a and 9b). This shows that mobile charges are rapidly formed (< 1 ps) and that the photogenerated charges stay in the perovskite for at least a few ps. After 1 ns, the mobility is reduced to approximately a third, suggesting that charges disappearing on this time scale. As discussed above, there are three different processes, either consecutively or simultaneously occurring that may be responsible for this decay. The very small energy offset at the interface between MAPbI3 and PCBM could retard electron injection to the ns timescale. Transient absorption spectroscopy as reported by Xing et al. estimated electron injection from perovskite to PCBM to be within several ns [17]. Therefore, electron injection into PCBM can be one of the processes leading to the THz decay. The

the energy gap in their conduction bands is small.

PCBM (process 3). In this scenario, a clear nonexponential, slower decay is expected since mobile electrons in MAPbI3 become immobile in PCBM. However, since there is no such plateau observed (Fig. 9b), it can be concluded that the rate constant for the interfacial electron hole recombination (process 3) is similar to, or exceeds, the electron injection rate, which is consistent with the fact that the difference in the energy gap in their conduction bands is small.

At the earliest timescale of the THz kinetics, MAPbI3/PCBM bilayer and neat MAPbI3 have the same mobility (~15 cm2 V-1s-1; Figs. 9a and 9b). This shows that mobile charges are rapidly formed (<1 ps) and that the photogenerated charges stay in the perovskite for at least a few ps. After 1 ns, the mobility is reduced to approximately a third, suggesting that charges disappear on this timescale. As discussed above, there are three different processes either consecutively or simultaneously occurring that may be responsible for this decay. The very small energy offset at the interface between MAPbI3 and PCBM could retard electron injection to the ns timescale. Transient absorption spectroscopy as reported by Xing et al. estimated electron injection from perovskite to PCBM to be within several ns [17]. Therefore, electron injection into PCBM can be one of the processes leading to the THz decay. The second plausible origin of the THz decay is the inevitable recombination of electrons injected in the PCBM, which are pinned at the interface due to their low mobility, with dark and photogenerated holes in perovskite. Similar to MAPbI3/Spiro-OMeTAD, concentration of dark holes here is expected to be at least the same order since the preparation method used is similar. Third, the excitationdependent second-order recombination within the perovskite process could also manifest as decay of the conductivity signal. The THz kinetics of neat MAPbI3 (Figs. 9 a and Figs. 9) show that at 2.1 × 1012 ph/cm2 per pulse, this process occurs on the many ns timescale, significantly slower than the decay of the MAPbI3/PCBM conductivity. Hence, this process only weakly contributes. The THz conductivity decay can therefore be assigned to convolution of electron injection to PCBM and electron–hole recombination at the perovskite/PCBM interface wherein both processes are occurring on a similar timescale.

The TRMC measurements of MAPbI3/PCBM in Fig. 10a show extended time window of conductivity measurements and were obtained at lower excitation intensity than the TRTS measurements. Because of this, the second-order recombination could only occur on the microsecond timescale [33] and will only marginally influence the decay. Hence, the origin of the decay should only be from electron injection and interfacial recombination. The excitationdependent mobility in Fig. 10b provides another indication of the timescale of electron injection and interfacial recombination. On the one hand, MAPbI3/PCBM has mobility of 2 cm2 V-1s-1, representing mobile holes only since injection is apparent. On the other hand, MAPbI3/Spiro-OMeTAD has hole mobility of 5.5 cm2 V-1s-1, i.e., 9 cm2 V-1s-1 from the neat minus 3.5 cm2 V-1s-1. There is almost a three-time difference between the hole mobilities measured from the two samples. Assuming 100% electron injection into PCBM after several ns, it means that two-thirds of the photogenerated mobile holes have disappeared, most probably through interfacial recombination, on the timescale similar to injection time. Admittedly, Wojciechowski et al. reported TRMC kinetics decay similar to that observed here. However, the decay was assigned to electron injection under the assumption that the difference electron and hole mobility is at least ten times [54]. This is in contrast with several theoretical papers on the electrons and holes' effective masses stating that they do not differ more than a factor 2 [16,55,56], to experimental works showing balanced electron–hole diffusion lengths [3,17], and to previous THz and TRMC conductivity measurements [33,57]. Moreover, using the results presented here it is possible to understand the TRMC kinetics as ns electron injection into PCBM in convolution with interfacial recombination between immobile electrons in PCBM and photogenerated and dark holes in the perovskite. The resulting picture is an oppositely charged bilayer material wherein recombination dynamics is not influenced by the excitation density. In fact, as shown in Figure 12b, the TRMC kinetic decays of MAPbI3/PCBM do not have any dependency on excitation. This is a strong indication of the first-order character of the recombination.

These results have far-reaching implications in understanding the fundamental photophysical processes in these materials and to the operation of perovskite solar cells. Utmost care should be taken in interpreting photophysical data as these are strongly influenced by the state and population of defects that control their doping, as reported by several papers [45-47]. Recom‐ bination at the MAPbI3/PCBM interface represents a loss mechanism and therefore is detri‐ mental in the operation of solar cells. Similar to strategy in optimizing the performance of perovskite devices by adding dopants to Spiro-OMeTAD in order to increase its conductivity, PCBM should also be doped. Moreover, developing new methods in reducing unintentional doping would certainly be beneficial. In summary of this section, it was found that hole transfer from MAPbI3 into Spiro-OMeTAD occurs on a sub-ps timescale, while its recombination dynamics is identical to neat MAPbI3 and controlled by a high concentration of dark holes. Electron injection in PCBM is slower, few ns, which is convoluted with the interfacial recom‐ bination between the electrons residing in PCBM and the photogenerated and dark holes in MAPbI3. The positively charged majority carriers brought about by unintentional doping, dictate not only the recombination of photoexcited carriers in neat MAPbI3 layer but also control the charge injection dynamics in bilayer samples. Finally, reduction of the hole concentration in the perovskite could help to retard the recombination yielding a higher overall power conversion efficiency.

## **5. Conclusion**

V-1s-1,

V-1s-1.

V-1s-1 from the neat minus 3.5 cm2

PCBM (process 3). In this scenario, a clear nonexponential, slower decay is expected since mobile electrons in MAPbI3 become immobile in PCBM. However, since there is no such plateau observed (Fig. 9b), it can be concluded that the rate constant for the interfacial electron hole recombination (process 3) is similar to, or exceeds, the electron injection rate, which is consistent with the fact that the difference in the energy gap in their conduction bands is small.

At the earliest timescale of the THz kinetics, MAPbI3/PCBM bilayer and neat MAPbI3 have the

formed (<1 ps) and that the photogenerated charges stay in the perovskite for at least a few ps. After 1 ns, the mobility is reduced to approximately a third, suggesting that charges disappear on this timescale. As discussed above, there are three different processes either consecutively or simultaneously occurring that may be responsible for this decay. The very small energy offset at the interface between MAPbI3 and PCBM could retard electron injection to the ns timescale. Transient absorption spectroscopy as reported by Xing et al. estimated electron injection from perovskite to PCBM to be within several ns [17]. Therefore, electron injection into PCBM can be one of the processes leading to the THz decay. The second plausible origin of the THz decay is the inevitable recombination of electrons injected in the PCBM, which are pinned at the interface due to their low mobility, with dark and photogenerated holes in perovskite. Similar to MAPbI3/Spiro-OMeTAD, concentration of dark holes here is expected to be at least the same order since the preparation method used is similar. Third, the excitationdependent second-order recombination within the perovskite process could also manifest as decay of the conductivity signal. The THz kinetics of neat MAPbI3 (Figs. 9 a and Figs. 9) show that at 2.1 × 1012 ph/cm2 per pulse, this process occurs on the many ns timescale, significantly slower than the decay of the MAPbI3/PCBM conductivity. Hence, this process only weakly contributes. The THz conductivity decay can therefore be assigned to convolution of electron injection to PCBM and electron–hole recombination at the perovskite/PCBM interface wherein

The TRMC measurements of MAPbI3/PCBM in Fig. 10a show extended time window of conductivity measurements and were obtained at lower excitation intensity than the TRTS measurements. Because of this, the second-order recombination could only occur on the microsecond timescale [33] and will only marginally influence the decay. Hence, the origin of the decay should only be from electron injection and interfacial recombination. The excitationdependent mobility in Fig. 10b provides another indication of the timescale of electron injection and interfacial recombination. On the one hand, MAPbI3/PCBM has mobility of 2 cm2

representing mobile holes only since injection is apparent. On the other hand, MAPbI3/Spiro-

V-1s-1, i.e., 9 cm2

There is almost a three-time difference between the hole mobilities measured from the two samples. Assuming 100% electron injection into PCBM after several ns, it means that two-thirds of the photogenerated mobile holes have disappeared, most probably through interfacial recombination, on the timescale similar to injection time. Admittedly, Wojciechowski et al. reported TRMC kinetics decay similar to that observed here. However, the decay was assigned to electron injection under the assumption that the difference electron and hole mobility is at least ten times [54]. This is in contrast with several theoretical papers on the electrons and holes'

V-1s-1; Figs. 9a and 9b). This shows that mobile charges are rapidly

same mobility (~15 cm2

both processes are occurring on a similar timescale.

370 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

OMeTAD has hole mobility of 5.5 cm2

Despite the advancement in the understanding of the device properties of perovskite solar cells, research on its fundamental electrical characteristics have remained scarce. This work has shown that using time-resolved THz spectroscopy and microwave conductivity measure‐ ments, complemented by transient absorption and photoluminescence spectroscopy, an indepth understanding may be achieved. Among its nearly ideal solar cell characteristics are ultrafast charge generation (<0.2 ps), high mobility (μe = 12.5 cm2 V-1s-1 and μh = 7.5 cm2 V-1s-1) which remained constant up to at least 1 ns, charge lifetime of tens of μs, and recombination barrier energy of 75 meV. One of the challenges that need to be addressed by the solar cell community is to standardized protocol on growth/preparation methods such that the state and concentration of defects, thereby, dark carriers will be controlled as desired. As presented in this chapter, such dark carriers play a vital role in recombination dynamics and hence could spell success or failure of the device.

## **Acknowledgements**

This work was supported by the Swedish Energy Agency (STEM), the Swedish Research Council, the Knut&Alice Wallenberg Foundation, the European Research Council (Advanced Investigator Grant to VS, 226136-VISCHEM), the Nanometer Consortium at Lund University (nmc@LU), and the Lund Laser Center. The time-resolved THz setup is partly developed by Pascher Instruments (www.pascherinstruments.com). The author would also like to express his gratitude to Prof. Tom J. Savenije of Delft University of Technology, The Netherlands, where all the TRMC measurements were done. Finally, this work is dedicated to Prof. Villy Sundström, a mentor, a colleague, and a friend.

## **Author details**

Carlito S. Ponseca Jr.\*

Address all correspondence to: carlito.ponseca@chemphys.lu.se

Division of Chemical Physics, Lund University, Lund, Sweden

## **References**


[7] Nemec H, Rochford J, Taratula O, et al. Influence of the Electron-Cation Interaction on Electron Mobility in Dye-Sensitized ZnO and TiO2 Nanocrystals: A Study Using Ultrafast Terahertz Spectroscopy. *Phy Rev Lett* 2010; 104:197401.

**Acknowledgements**

**Author details**

**References**

342:341–344.

*Phys Chem Lett* 2012; 3:2442–2446.

Carlito S. Ponseca Jr.\*

Sundström, a mentor, a colleague, and a friend.

372 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Address all correspondence to: carlito.ponseca@chemphys.lu.se

Division of Chemical Physics, Lund University, Lund, Sweden

This work was supported by the Swedish Energy Agency (STEM), the Swedish Research Council, the Knut&Alice Wallenberg Foundation, the European Research Council (Advanced Investigator Grant to VS, 226136-VISCHEM), the Nanometer Consortium at Lund University (nmc@LU), and the Lund Laser Center. The time-resolved THz setup is partly developed by Pascher Instruments (www.pascherinstruments.com). The author would also like to express his gratitude to Prof. Tom J. Savenije of Delft University of Technology, The Netherlands, where all the TRMC measurements were done. Finally, this work is dedicated to Prof. Villy

[1] Kojima A, Teshima K, Shirai Y, Miyasaka T. Organometal Halide Perovskites as Visi‐ ble-Light Sensitizers for Photovoltaic Cells. *J Am Chem Soc* 2009; 48: 6050–6051.

[2] Anon National Renewable Energy Labs (NREL) efficiency chart (2015); http:// www.nrel.gov/ncpv/images/efficiency\_chart.jpg Natl Renew Energy Labs Effic chart

[3] Stranks SD, Eperon GE, Grancini G, et al. Electron-Hole Diffusion Lengths Exceeding 1 micrometer in an Organometal Trihalide Perovskite Absorber. *Science* 2013;

[4] Wehrenfennig C, Eperon GE, Johnston MB, et al. High Charge Carrier Mobilities and Lifetimes in Organolead Trihalide Perovskites *Adv Mater* 2013; 26:1584–1589.

[5] Ponseca CS, Yartsev A, Wang E, Andersson MR, et al. Ultrafast Terahertz Photocon‐ ductivity of Bulk Heterojunction Materials Reveals High Carrier Mobility up to

[6] Ponseca CS, Němec H, Vukmirović N, et al. Electron and Hole Contributions to the Terahertz Photoconductivity of a Conjugated Polymer:Fullerene Blend Identified *J*

(2015); http//www.nrel.gov/ncpv/images/efficiency\_chart.jpg

Nanosecond Time Scale. *J Am Chem Soc* 2012; 134:11836–11839.


[37] Knop O, Wasylishen RE, White MA, et al. Alkylammonium Lead Halides 0.2. CH3NH3PbCl3, CH3NH3PbBr3, CH3NH3PbI3 Perovskites – Cuboctahedral Halide Cages with Isotropic Cation Reorientation. *Can J Chem: Rev Can Chim* 1990; 68:412−422.

[22] Bi D, Yang L, Boschloo G, et al. Effect of Different Hole Transport Materials on Re‐ combination in CH3NH3PbI3 Perovskite-Sensitized Mesoscopic Solar Cells. *J Phys*

[23] Kim HS, Lee C-R, Im J.H, et al. Lead Iodide Perovskite Sensitized All-Solid-State Sub‐ micron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%. *Sci Rep* 2012;

[24] Jeng, JY, Chiang, YF, Lee MH, et al. CH3NH3PbI3 Perovskite/Fullerene Planar- Het‐

[25] Marchioro A, Teuscher J, Friedrich D, et al. Unravelling the Mechanism of Photoin‐ duced Charge Transfer Processes in Lead Iodide Perovskite Solar Cells. *Nat Photonics*

[26] Ishihara T. Optical-Properties of PbI-Based Perovskite Structures. *J Lumin* 1994; 60(1):

[27] Sun S, Salim T, Mathews N, et al. The Origin of High Efficiency in Low-Temperature Solution-Processable Bilayer Organometal Halide Hybrid Solar Cells. *Energy Environ*

[28] Liu M, Johnston MB, Snaith HJ. Efficient Planar Heterojunction Perovskite Solar Cells

[29] Ball JM, Lee MM, Hey A, Snaith HJ. Low-Temperature Processed Meso-Superstruc‐ tured to Thin-Film Perovskite Solar Cells. *Energy Environ Sci* 2013; 6:1739−1743.

[30] Wasylishen RE, Knop O, Macdonald JB. Cation Rotation in Methylammonium Lead

[31] Onodayamamuro N, Yamamuro O, Matsuo T, Suga HP-T. Phase-Relations of CH3NH3PbX3 (X = CL, BR, I) Crystals. *J Phys Chem Solids* 1992; 53:277−281.

[32] Onodayamamuro N, Matsuo T, Suga H. Dielectric Study of CH3NH3PbX3 (X = CL,

[33] Savenije TJ, Ponseca CS, Kunneman L, et al. Thermally Activated Exciton Dissocia‐ tion and Recombination Control the Carrier Dynamics in Organometal Halide Perov‐

[34] Servaites JD, Ratner MA, Marks TJ. Organic Solar Cells: A New Look at Traditional

[35] Stoumpos CC, Malliakas, CD, Kanatzidis MG. Semiconducting Tin and Lead Iodide Perovskites with Organic Cations: Phase Transitions, High Mobilities, and Near-In‐

[36] van Zeghbroeck B. Principles of Semiconductor Devices. University of Colorado:

frared Photoluminescent Properties. *Inorg Chem* 2013; 52:9019−9038.

erojunction Hybrid Solar Cells. *Adv Mater* 2013; 25:3727−3732.

by Vapour Deposition. *Nature* 2013; 6:1739−1743.

Halides. *Solid State Commun* 1985; 56:581−582.

BR, I). *J Phys Chem Solids* 1992; 53:935−939.

Models. *Energy Environ Sci* 2011; 4:4410−4422.

skite *J Phys Chem Lett* 2014; 5:2189–94.

Boulder CO, 2006.

*Chem Lett* 2013; 4:1532−1536.

374 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

2:591-595.

2014; 8:250−255.

*Sci* 2014; 7:399−407.

269−274.


## **Photoexcitations and Emission Processes in Organometal Trihalide Perovskites**

Michele Cadelano, Michele Saba, Nicola Sestu, Valerio Sarritzu, Daniela Marongiu, Feipeng Chen, Roberto Piras, Francesco Quochi, Andrea Mura and Giovanni Bongiovanni

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61282

## **Abstract**

[50] Von Hauff E, Dyakonov V, Parisi J. Study of Field Effect Mobility in PCBM Films and

[51] Mihailetchi VD, Xie H, De Boer B, et al. Charge Transport and Photocurrent Genera‐ tion in Poly(3-hexylthiophene): Methanofullerene Bulk-Heterojunction Solar Cells.

[52] Warman JM, De Haas MP, Anthopoulos TD, De Leeuw DM. The Negative Effect of High-Temperature Annealing on Charge-Carrier Lifetimes in Microcrystalline

[53] Savenije TJ, Ferguson AJ, Kopidakis N, Rumbles G. Revealing the Dynamics of Charge Carriers in Polymer:Fullerene Blends Using Photoinduced Time-Resolved

[54] Wojciechowski K, Stranks, SD, Abate A, et al. Heterojunction Modification for High‐ ly Efficient Organic–Inorganic Perovskite Solar Cells. *ACS Nano* 2014; 8:12701–12709.

[55] Chang Y, Park C, Matsuishi K. First-Principles Study of the Structural and the Elec‐ tronic Properties of the Lead-Halide-Based Inorganic-Organic Perovskites (CH3NH3)PbX3 and CsPbX3 (X = Cl, Br, I). *J Korean Phys Soc* 2004; 44: 889–893. [56] Giorgi G, Fujisawa, JI, Segawa H, Yamashita K. Small Photocarrier Effective Masses Featuring Ambipolar Transport in Methylammonium Lead Iodide Perovskite: A

[57] Ponseca, CS, Savenije, TJ, Abdellah M, et al. Organometal Halide Perovskite Solar Cell Materials Rationalized: Ultrafast Charge Generation, High and Microsecond-Long Balanced Mobilities, and Slow Recombination. J Am Chem Soc 2014; 136:5189–

P3HT:PCBM Blends. *Sol Energy Mater Sol Cells* 2005; 87:149–156.

Microwave Conductivity. *J Phys Chem C* 2013; 117:24085–24103.

Density Functional Analysis. *J Phys Chem Lett* 2013; 4:4213–4216.

*Adv Funct Mater* 2006; 16:699–708.

376 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

PCBM. *Adv Mater* 2006; 18;2294–2298.

5192.

Organometal halide perovskites have recently attracted widespread attention among sci‐ entists, as they combine the advantages of low-cost processability with strong light ab‐ sorption, band-gap tunability from the near-infrared to the visible region of the electromagnetic spectrum, efficient light emission and charge transport. Such combina‐ tion of features is unique among solution-processed materials and makes perovskites ap‐ pealing for several optoelectronic applications, in particular those related to energy sustainability, which could help the advent of a new generation of low-cost but efficient solar cells and large-area light-emitting devices.This chapter reports a critical review of the efforts that scientists have made until now to understand the photophysics of organo‐ metal halide perovskites. We address the ongoing debate on the nature of the photoexcit‐ ed species, namely the role played by free carriers and excitons, the determination of the exciton binding energy as a measure of the Coulomb interaction strength in these materi‐ als, the competition between radiative and non-radiative processes, the role and density of charge carrier traps, and last but not least a critical analysis of those phenomena at the base of laser action, highlighting the most relevant results and possible solutions to issues that still remain open.

**Keywords:** Hybrid perovskites, excitons, free carriers, amplified spontaneous emission, laser

## **1. Introduction**

Organometal halide perovskites are solution-processed semiconductors showing efficient charge transport, favorable emission properties, strong light absorption and optical gap

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

tunability from the visible to the near-infrared spectrum [1–16]. Such interesting properties make these materials very appealing for the realization of solar cells and optical emitters [17– 19]. Despite the widespread research activities on these novel semiconductors, some funda‐ mental aspects of the photophysics underlying perovskites have been elusive, especially for what concerns the excited state dynamics [20–29]. For instance, due to the hybrid organic– inorganic nature of organometal perovskites, in principle it is not clear if the excited state properties are dominated by bound or unbound electron-hole states, that is excitons or free carriers. While initial reports assumed the presence of excitons, much like in organic materials, recent optical spectroscopy evidences have emerged that the majority of band-edge optical excitations at room temperature appears to be free carriers [6,22,30].

The architecture of a solar cell depends on this feature. Indeed, if the majority of photoexcited carriers happens to be bound electron-hole states, a heterojunction is necessary to split charges and then produce a current flow in a photovoltaic device. On the other hand, if photoexcitation results in the formation of free carriers, the solar cell structure is simplified, since charges are already separated and can be easily collected at the electrodes.

A lively debate has then started to understand the physical reasons that underpin the preva‐ lence of free carriers over excitons. Clearly, within such debate, a precise and reliable deter‐ mination of the exciton binding energy is of paramount importance. Reports of exciton binding energies have varied widely, from less than 5 meV to over 50 meV [22,30–37]; it has even been suggested that the exciton binding energy may be temperature-dependent, due to ionic screening effects [33,34,38].

Beyond the potential application in photovoltaics, organometal halide perovskites exhibit also excellent emission properties. In fact, amplified spontaneous emission (ASE) was also dem‐ onstrated for photoexcited carrier densities above a threshold value [17–19], representing the first step towards a laser device. Until now, many experiments probed optical amplification under impulsive excitation, meaning that the pumping pulse duration is much shorter than the typical lifetime of the excited states of the crystals [17–19,39–41]. In these kinds of experi‐ ments, ASE occurs within a sub-nanosecond time window, which is far from the continuous wave (cw) operation of a real laser device. Hence, it is not clear how long these materials can sustain optical amplification and what warming issues and parasitic processes are involved. Understanding these features is crucial in the perspective of the realization of a perovskitebased cw laser [19].

The aim of this chapter is to summarize the progress made in understanding the physics of excitons and free carriers in perovskites, both in the low excitation regime relevant for solar cells and in the high excitation regime needed for optical amplification. Experimental results obtained from optical spectroscopy measurements carried out in the Department of Physics of the University of Cagliari will be analyzed and critically compared with results published by other research groups.

Sections 2–5 of this chapter describe the absorption and the emission properties of pure and mixed methylammonium lead iodide perovskite films (MAPbI3 and MAPbI3-xClx, respective‐ ly), showing that the prevailing excited species in these materials is a correlated electron-hole plasma at room temperature, from injected carrier densities typical of solar illumination to those typical to obtain light amplification [30].

An investigation about amplified spontaneous emission in organolead halide perovskites is provided in Sections 6–10, where both the temporal and the spectral photoluminescence (PL) signals have been studied under short- and long-pulsed excitation at high laser fluence, focusing on the physical parameters that allow or inhibit optical amplification in methylam‐ monium lead iodide and bromide (MAPbBr3) perovskite thin films.

## **2. Exciton binding energy in perovskite films**

tunability from the visible to the near-infrared spectrum [1–16]. Such interesting properties make these materials very appealing for the realization of solar cells and optical emitters [17– 19]. Despite the widespread research activities on these novel semiconductors, some funda‐ mental aspects of the photophysics underlying perovskites have been elusive, especially for what concerns the excited state dynamics [20–29]. For instance, due to the hybrid organic– inorganic nature of organometal perovskites, in principle it is not clear if the excited state properties are dominated by bound or unbound electron-hole states, that is excitons or free carriers. While initial reports assumed the presence of excitons, much like in organic materials, recent optical spectroscopy evidences have emerged that the majority of band-edge optical

The architecture of a solar cell depends on this feature. Indeed, if the majority of photoexcited carriers happens to be bound electron-hole states, a heterojunction is necessary to split charges and then produce a current flow in a photovoltaic device. On the other hand, if photoexcitation results in the formation of free carriers, the solar cell structure is simplified, since charges are

A lively debate has then started to understand the physical reasons that underpin the preva‐ lence of free carriers over excitons. Clearly, within such debate, a precise and reliable deter‐ mination of the exciton binding energy is of paramount importance. Reports of exciton binding energies have varied widely, from less than 5 meV to over 50 meV [22,30–37]; it has even been suggested that the exciton binding energy may be temperature-dependent, due to ionic

Beyond the potential application in photovoltaics, organometal halide perovskites exhibit also excellent emission properties. In fact, amplified spontaneous emission (ASE) was also dem‐ onstrated for photoexcited carrier densities above a threshold value [17–19], representing the first step towards a laser device. Until now, many experiments probed optical amplification under impulsive excitation, meaning that the pumping pulse duration is much shorter than the typical lifetime of the excited states of the crystals [17–19,39–41]. In these kinds of experi‐ ments, ASE occurs within a sub-nanosecond time window, which is far from the continuous wave (cw) operation of a real laser device. Hence, it is not clear how long these materials can sustain optical amplification and what warming issues and parasitic processes are involved. Understanding these features is crucial in the perspective of the realization of a perovskite-

The aim of this chapter is to summarize the progress made in understanding the physics of excitons and free carriers in perovskites, both in the low excitation regime relevant for solar cells and in the high excitation regime needed for optical amplification. Experimental results obtained from optical spectroscopy measurements carried out in the Department of Physics of the University of Cagliari will be analyzed and critically compared with results published

Sections 2–5 of this chapter describe the absorption and the emission properties of pure and mixed methylammonium lead iodide perovskite films (MAPbI3 and MAPbI3-xClx, respective‐ ly), showing that the prevailing excited species in these materials is a correlated electron-hole

excitations at room temperature appears to be free carriers [6,22,30].

378 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

already separated and can be easily collected at the electrodes.

screening effects [33,34,38].

based cw laser [19].

by other research groups.

A large absorption coefficient across the visible spectrum and the consequent efficient solar light harvesting is one of the strong points of perovskites. Initial works rationalized the strong absorption close to the energy-gap and the nature of optical excitations close to the band-gap in terms of excitons, as in organic semiconductors; further investigations have recently converged to state that photoexcitation results in the generation of free carriers in a wide range of pumping intensities, even at ones comparable to solar illumination, as typical of inorganic semiconductors. Such finding is consistent with small values of the exciton binding energy, for which thermal energy is sufficient to ionize excitons in free electrons and holes.

A reliable framework usually adopted in solid state physics to estimate the exciton binding energy is Elliot's theory of Wannier excitons [42], which is valid for bulk semiconductors having exciton binding energies much smaller than the energy gap (Wannier excitons). Such theory accurately describes the optical transitions near the band-gap in inorganic semicon‐ ductors like GaAs and GaP [43,44] and models the shape of the optical absorption coefficient *α*(ℏ*ω*) by the following equation:

$$\alpha \left(\hbar \alpha \right) \approx \frac{\mu\_{\rm ce}^{2}}{\hbar \alpha} \sum\_{j} \left| \phi\_{j} \left(r = 0 \right) \right|^{2} \delta \left(\hbar \alpha - E\_{j} \right) \approx \frac{\mu\_{\rm ce}^{2}}{\hbar \alpha} \left[ \sum\_{j} \frac{4 \pi \sqrt{E\_{\rm b}^{3}}}{j^{3}} \delta \left(\hbar \alpha - E\_{j}^{b} \right) + \frac{2 \pi \sqrt{E\_{\rm b}} \theta \left(\hbar \alpha - E\_{\rm g} \right)}{1 - e^{-2 \pi \sqrt{\frac{E\_{\rm b}}{\hbar \alpha - E\_{\rm g}}}}} \right] \tag{1}$$

The absorption coefficient depends on the weighted density of electron-hole pair states, with the weight provided by the probability for an electron and a hole to be at the same position, that is |*φ<sup>j</sup>* (*r* =0)| 2, while *φ<sup>j</sup>* represents the wave functions of bound and unbound states. *μ*cv represents the transition dipole moment between conduction and valence bands, and ℏ*ω* is the photon energy involved in the transition. In the third expression of Eq. (1), the first term describes the transitions to bound exciton states *Ej* b , while the second term takes into account the band-to-band transitions above the energy gap *E*g. *δ*(*x*) and *θ*(*x*) are the Dirac-delta and the Heaviside step function, respectively, and *E*b is the exciton binding energy.

Eq. (1) can be used to fit the measured absorption spectra of perovskite films, provided that it is convoluted with a bell-shaped function accounting for line broadening. In addition, the introduction of the joint valence-band energy-momentum dispersion leads to a better approx‐ imation between theory and experiment, as it overcomes the limits of the parabolic bands approximation. Upon these improvements, Eq. (1) can be modified as follows:

$$\alpha \left( \hbar o \right) = \frac{a\_{\text{cal}}}{\hbar o} \left| \sum\_{j} \frac{2E\_{\text{b}}}{j^3} \text{sech} \left( \frac{\hbar o - E\_{j}^{b}}{\Gamma} \right) + \int\_{\tilde{E}\_{\text{g}}}^{\tilde{\text{s}}} \text{sech} \left( \frac{\hbar o - E}{\Gamma} \right) \frac{1}{\frac{-2\omega}{1 - E\_{\text{g}}} \sqrt{\frac{E\_{\text{b}}}{E - E\_{\text{g}}}}} \left[ 1 + \frac{10\mu b}{\hbar^{4}} \left( E - E\_{\text{g}} \right) \right] dE \right| \tag{2}$$

The density of states has been developed in series at the first order for small values of *b*, which is a factor that takes into account the non-parabolic dispersion. The hyperbolic secant function of width Γ accounts for thermal and inhomogeneous broadening; *α*scal is a constant that is adjusted in order to obtain the right scaling between the model and the experimental absorp‐ tion spectrum; in the fitting procedure, we also allowed for absorbance offsets often found in experimental curves due to unbalanced reference or zero spectra in spectrophotometers.

In order to establish that the model can describe the absorption of MAPbI3 perovskite films, regardless of the particular morphology and growth conditions of each sample, we fitted Eq. (2) with a least square method to several absorption spectra reported in literature by different research groups. The binding energy is obtained as a fitting parameter, and the fitting range has been extended as much as possible at low and high energy in order to minimize uncer‐ tainties connected to the calculation procedure. Published data were extracted from the pdf files through the CurveSnap software, freely available online.

First, we analyzed the absorption measurements at low temperature, where the broadening is less pronounced and therefore the fitting procedure is more sensitive to the value of the exciton binding energy. Figures 1a,b show a comparison between the absorption spectra reported by D'Innocenzo et al. and our experimental data, recorded at 160 and 170 K, respectively [22,30]; the relative contributions of excitons and free carriers to the absorption spectrum are also shown. While the broadening is different in the two cases, the values of the exciton binding energy are comparable and coincide within the uncertainty. Figures 1c–f show the same analysis carried out at room temperature, applied to absorption spectra reported in references [5,22,45]. Even at room temperature, the fitted values for *E*<sup>b</sup> do not differ more than 20% from the average of 23 meV. Such value is intermediate between the binding energy typical of Frenkel excitons in organic semiconductors (>100 meV) and the one typical of Wannier excitons in inorganics (<10 meV) [46,47].

The overall picture emerging from the analysis of optical absorption data seems therefore converging towards a value of 23 meV for the exciton binding energy. These results are in contrast with what observed with magnetometry, as some recent publications have reported values of the exciton binding energy that depend on temperature, as a consequence of screening effects. Such works report exciton binding energy to be nearly 15 meV at low

Photoexcitations and Emission Processes in Organometal Trihalide Perovskites http://dx.doi.org/10.5772/61282 381

**Figure 1.** Absorption spectra in MAPbI3 perovskite films. The red empty circles represent the theoretical fits to the ex‐ perimental data (continuous black lines). The contributions to the absorption due to both excitonic and band-to-band transitions are modelled in the framework of the Elliott's theory of Wannier excitons. The dotted green lines are rela‐ tive to excitonic transitions, while the continuous blue lines are relative to band-to-band contributions with the inclu‐ sion of Coulomb interactions between electrons and holes. (a,b) Absorption spectrum reported by Saba et al. at 170 K and by D'Innocenzo et al. at 160 K, respectively [22,30]. (c,d,e,f) Absorption spectra reported at room temperature by Saba et al., Sutherland et al., D'Innocenzo et al. and Stranks et al., respectively [5,22,30,45].

temperature and a few meV at 300 K [33,34,38,48], while previous works with similar techniques reported values as large as 50 meV [31,49]. Since the interpretation of magneto‐ metry data requires several analysis steps, at the moment it is not clear what the origin of this discrepancy is.

## **3. Emission properties from free carriers**

Eq. (1) can be used to fit the measured absorption spectra of perovskite films, provided that it is convoluted with a bell-shaped function accounting for line broadening. In addition, the introduction of the joint valence-band energy-momentum dispersion leads to a better approx‐ imation between theory and experiment, as it overcomes the limits of the parabolic bands

<sup>g</sup> <sup>g</sup>

*<sup>E</sup> <sup>E</sup> E b E E dE*

1

<sup>h</sup> <sup>h</sup> (2)

3 4 g

é ù

w

ê ú æ ö - æö é ù - <sup>=</sup> ê ú ç ÷ <sup>+</sup> ç ÷ + - ê ú ç ÷ èø ë û è ø ë û -

The density of states has been developed in series at the first order for small values of *b*, which is a factor that takes into account the non-parabolic dispersion. The hyperbolic secant function of width Γ accounts for thermal and inhomogeneous broadening; *α*scal is a constant that is adjusted in order to obtain the right scaling between the model and the experimental absorp‐ tion spectrum; in the fitting procedure, we also allowed for absorbance offsets often found in experimental curves due to unbalanced reference or zero spectra in spectrophotometers.

In order to establish that the model can describe the absorption of MAPbI3 perovskite films, regardless of the particular morphology and growth conditions of each sample, we fitted Eq. (2) with a least square method to several absorption spectra reported in literature by different research groups. The binding energy is obtained as a fitting parameter, and the fitting range has been extended as much as possible at low and high energy in order to minimize uncer‐ tainties connected to the calculation procedure. Published data were extracted from the pdf

First, we analyzed the absorption measurements at low temperature, where the broadening is less pronounced and therefore the fitting procedure is more sensitive to the value of the exciton binding energy. Figures 1a,b show a comparison between the absorption spectra reported by D'Innocenzo et al. and our experimental data, recorded at 160 and 170 K, respectively [22,30]; the relative contributions of excitons and free carriers to the absorption spectrum are also shown. While the broadening is different in the two cases, the values of the exciton binding energy are comparable and coincide within the uncertainty. Figures 1c–f show the same analysis carried out at room temperature, applied to absorption spectra reported in references [5,22,45]. Even at room temperature, the fitted values for *E*<sup>b</sup> do not differ more than 20% from the average of 23 meV. Such value is intermediate between the binding energy typical of Frenkel excitons in organic semiconductors (>100 meV) and the one typical of Wannier excitons

The overall picture emerging from the analysis of optical absorption data seems therefore converging towards a value of 23 meV for the exciton binding energy. These results are in contrast with what observed with magnetometry, as some recent publications have reported values of the exciton binding energy that depend on temperature, as a consequence of screening effects. Such works report exciton binding energy to be nearly 15 meV at low

b

 m

2


*e* p

approximation. Upon these improvements, Eq. (1) can be modified as follows:

b

*j*

w

380 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

files through the CurveSnap software, freely available online.

<sup>å</sup> <sup>ò</sup> <sup>h</sup> <sup>h</sup> <sup>h</sup>

scal b

a

w

in inorganics (<10 meV) [46,47].

a w

*j*

( ) ( )

*<sup>E</sup> <sup>j</sup> <sup>E</sup> E E*

¥

<sup>2</sup> 1 10 sech sech <sup>1</sup> Γ Γ

> Linear absorption spectroscopy is extremely useful to obtain information about the optical response and quickly highlights benefits and disadvantages of the material in terms of light harvesting, which represents the main ingredient for photovoltaics applications. Further insight into the electronic properties is provided by the study of the excited state emission properties, as the techniques probe how optical excitations populate the available states. Particularly, photoluminescence under pulsed, femtosecond excitation is useful in determin‐ ing relaxation rates and excited state dynamics.

> Emission in MAPbI3 and MAPbI3–*x*Cl*x* perovskites occurs close to the excitonic resonance and the high energy side of the PL peak overlaps with the continuum of band-to-band transitions

in a wide excitation range, as shown in Figure 2a. Short-pulse excitation experiments show that the PL signal rises instantaneously after the laser pulse arrival; the time decay of photo‐ luminescence can be observed from Figures 2b,c. Two different samples are presented, MAPbI3 and MAPbI3-xCLx perovskites; absorption and emission properties are quite similar in both perovskites, since chlorine content has been determined to be a small fraction, not larger than 2%. However, the dynamics of the PL is different in the two samples, with much longer decay in the MAPbI3-xCLx film; several reports in literature state that the role of Cl is to alter the perovskite crystallization dynamics and, as a result, MAPbI3-xCLx perovskites have lower trap densities than what observed in MAPbI3 [50,51], with a corresponding longer excited state lifetime. Inset in Figure 2a shows a clear difference between perovskites with and without chlorine in terms of PL lifetimes at low laser intensity, according to what reported in literature.

**Figure 2.** (a) Transient photoluminescence spectra for mixed MAPbI3–*x*Cl*x* films at 300 K, integrated in 60 ps (temporal resolution of our streak camera), compared with the absorption spectrum at the same temperature (gray shadow). Pho‐ toluminescence was excited by 130-fs-long laser pulses with a repetition rate of 1 kHz and 3.18 eV photon energy. The legend reports the injected carrier density, calculated from laser pulse fluence and film absorbance. Inset: time resolved PL signal from mixed MAPbI3–*x*Cl*x* (blue line) and pure MAPbI3 (red line) films, relative to an excitation density of at 1.7x1017cm–3. The different mean PL lifetime is attributed to different trap densities in the samples. (b,c) Transient PL signal as a function of the injected electron–hole pair density at the film surface in MAPbI3–*x*Cl*<sup>x</sup>* (b) and MAPbI3 films (c). The initial decays of the signal are fitted by an exponential function (black dotted lines). The injected carrier densi‐ ties are relative to the PL spectra shown in (a).

Radiative recombination processes can be investigated studying the intensity of the photolu‐ minescence signal immediately after the laser excitation (PL0), before electronic states are depopulated by slow recombination mechanisms. The electron-hole pair density *n*0 at the sample surface, injected from a short laser pulse, can be estimated multiplying the laser pulse fluence times the absorption coefficient of the sample at the excitation wavelength. The evolution of PL0 with *n*0 shows what type of carrier prevails as outcome of the photoexcitation.

in a wide excitation range, as shown in Figure 2a. Short-pulse excitation experiments show that the PL signal rises instantaneously after the laser pulse arrival; the time decay of photo‐ luminescence can be observed from Figures 2b,c. Two different samples are presented, MAPbI3 and MAPbI3-xCLx perovskites; absorption and emission properties are quite similar in both perovskites, since chlorine content has been determined to be a small fraction, not larger than 2%. However, the dynamics of the PL is different in the two samples, with much longer decay in the MAPbI3-xCLx film; several reports in literature state that the role of Cl is to alter the perovskite crystallization dynamics and, as a result, MAPbI3-xCLx perovskites have lower trap densities than what observed in MAPbI3 [50,51], with a corresponding longer excited state lifetime. Inset in Figure 2a shows a clear difference between perovskites with and without chlorine in terms of PL lifetimes at low laser intensity, according to what reported in literature.

> MAPbI3-xClx Carrier density 1.7 1017 cm-3 5.3 1017 cm-3 5.3 1018 cm-3

**Figure 2.** (a) Transient photoluminescence spectra for mixed MAPbI3–*x*Cl*x* films at 300 K, integrated in 60 ps (temporal resolution of our streak camera), compared with the absorption spectrum at the same temperature (gray shadow). Pho‐ toluminescence was excited by 130-fs-long laser pulses with a repetition rate of 1 kHz and 3.18 eV photon energy. The legend reports the injected carrier density, calculated from laser pulse fluence and film absorbance. Inset: time resolved PL signal from mixed MAPbI3–*x*Cl*x* (blue line) and pure MAPbI3 (red line) films, relative to an excitation density of at 1.7x1017cm–3. The different mean PL lifetime is attributed to different trap densities in the samples. (b,c) Transient PL signal as a function of the injected electron–hole pair density at the film surface in MAPbI3–*x*Cl*<sup>x</sup>* (b) and MAPbI3 films (c). The initial decays of the signal are fitted by an exponential function (black dotted lines). The injected carrier densi‐

Radiative recombination processes can be investigated studying the intensity of the photolu‐ minescence signal immediately after the laser excitation (PL0), before electronic states are depopulated by slow recombination mechanisms. The electron-hole pair density *n*0 at the sample surface, injected from a short laser pulse, can be estimated multiplying the laser pulse

a) b)

Log PL Intensity (a.u.)

Log PL Intensity (a.u.)

c)

0 1 2 3 4 Time (ns)

0 1 2 3 4 Time (ns)

MAPbI3-xClx

MAPbI3

1.3 1.4 1.5 1.6 1.7 1.8 1.9 Energy (eV)

Normalized PL Intensity

PL Intensity (a.u.)

0 5 10 15 20 Time (ns)

382 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

 MAPbI3-xClx MAPbI3

ties are relative to the PL spectra shown in (a).

Figure 3 shows PL0 as a function of *n*0 in MAPbI3 and MAPbI3–*x*Cl*x* perovskites. The PL intensity scales quadratically in the injected carrier densities from less than 1016 to 1018 cm-3, which corresponds to laser pulse fluence values from less than 0.1 to about 10 μJ/cm2 . Such quadratic scaling is the signature of bimolecular recombination and is consistent with the radiative recombination from a gas of unbound electron-hole pairs [52]. On the other hand, isolated excitons should give rise to monomolecular recombination, with the PL0 intensity linear in the exciting laser fluence. PL0 tends to saturate above 2.5x1018 cm–3, as a consequence of the band filling.

**Figure 3.** MAPbI3 and MAPbI3–*x*Cl*<sup>x</sup>* photoluminescence emission intensity estimated at *<sup>t</sup>*=0 (PL0) as a function of inject‐ ed electron-hole density (lower axis) and laser pulse fluence (upper axis). PL0 values relative to laser pulse fluences higher than 100 *μ*J/cm2 could not be extracted from experimental decays, since they were much faster than the tempo‐ ral resolution of our instrument. The black-dashed lines represent the PL0 quadratic dependence on *n*0, and are re‐ ported as a guide for eyes.

Focusing only at time zero, all the recombination processes that occur later after laser excitation are neglected. On the other hand, they can be investigated studying the quantum yield (QY), defined by the ratio between the time-integral of the PL signal (TIPL) and the laser fluence *ϕ*.

Figure 4a shows the QY as a function of the laser fluence, which can be explained in terms of interplay between radiative and non-radiative recombination processes. The rise of the QY below *n*<sup>0</sup> ~1017 cm–3 is due to the fact that carrier trapping is the fastest recombination process in such excitation regime, while radiative recombination becomes faster for growing carrier densities, so that a larger fraction of optical excitations recombine radiatively. The increase in QY reaches a maximum above *n*<sup>0</sup> ~10<sup>17</sup> cm–3, as the radiative recombination rate overcomes the carrier trapping and becomes the fastest recombination channel; under such conditions, a further increase in the radiative recombination rate only makes recombination faster, but it does not significantly increase the fraction of carriers that recombine radiatively, which is already close to unity. For even higher excited densities, above *n*<sup>0</sup> ~10<sup>17</sup> cm–3, non-radiative density-dependent Auger recombination, whose rate is cubic in the density, and therefore increases faster than the radiative rate, becomes dominant, causing the observed QY decrease [30]. The remarkable high values of the QY (30% for MAPbI3 and 70% for MAPbI3–*x*Cl*x*) are reached for injected carrier densities close to the amplified spontaneous emission thresholds reported for these materials [18,19].

**Figure 4.** Photoluminescence quantum yield and photoluminescence decay rates for the two perovskite samples (MAPbI3 red triangles, MAPbI3xClx blue squares). (a) The emission quantum yields are calculated as TIPL/ *ϕ*, where *ϕ* is the laser pulse fluence. The injected carrier density is calculated multiplying the laser pulse photon fluence by the absorption coefficient of the films. As a reference, the laser pulse fluence directly measured in the experiments is also reported on the top axis. Initially, the QY grows with fluence for both films, as the bimolecular recombination becomes faster and a growing fraction of the injected excitations recombine radiatively, instead of being trapped. At excitations 4×1019 cm–3, non-radiative Auger processes dominate. The absolute QY is scaled to match theoretical predictions. The maximum QY values are about 30% for MAPbI3 and 70% for MAPbI3–*x*Cl*x*. The dotted lines represent predictions from a rate equation model accounting for the main relaxation channels for electrons and holes. Simulations take into ac‐ count the exponential spatial profile of the electron-hole density created by laser pulses. The very good agreement be‐ tween model and data indicates that the main photophysical processes are accounted for in the model. (b) Decay rates for the transient PL signal plotted as a function of the injected carrier density on bottom axis and the laser pulse flu‐ ence on top axis. The rates, defined as *k*PL <sup>=</sup> *dPL dt* <sup>⋅</sup> <sup>1</sup> *PL <sup>t</sup>*=0, are extracted from the data in Figures 2b,c. Such rates represent the initial decay and should not be mistaken for the average photoluminescence decay rate obtained by fit‐ ting the entire decay with an exponential function. The error bars represent the standard deviation from a least square fit to an exponential decay and are reported only when they exceed the size of the marker. The rates are very similar

for the two samples (although the average exponential decay rates are significantly different) and grow together for growing injected carrier densities. Such rates measure the intrinsic density-dependent bimolecular and Auger recombi‐ nation processes. The dotted lines represent the results of the rate equation used to model the experimental data: the agreement with the experiment is satisfactory also in this case.

Also the analysis of the initial PL decay rate at time zero, extracted right after laser pulse excitation, confirms our interpretation of the hierarchy of the recombination rates. Figure 4b shows that the initial PL decay rate increases with *n*0, as a consequence of the activation of density-dependent recombination mechanisms. Differently from the average PL decay rate, calculated at low excitation intensity, which is different in various samples (inset in Figure 2a), the initial PL decay rate is quite similar among them. This fact reflects that while the mean PL decay time is sensitive to extrinsic effects like traps and defects, which are significant in all solution-processed semiconductors, the initial PL decay rate is dictated only by intrinsic nonlinear processes [30]. For growing excitation densities, the initial recombination rates increase in both perovskite samples, as radiative and Auger recombination become faster and faster.

## **4. Recombination rates**

densities, so that a larger fraction of optical excitations recombine radiatively. The increase in QY reaches a maximum above *n*<sup>0</sup> ~10<sup>17</sup> cm–3, as the radiative recombination rate overcomes the carrier trapping and becomes the fastest recombination channel; under such conditions, a further increase in the radiative recombination rate only makes recombination faster, but it does not significantly increase the fraction of carriers that recombine radiatively, which is already close to unity. For even higher excited densities, above *n*<sup>0</sup> ~10<sup>17</sup> cm–3, non-radiative density-dependent Auger recombination, whose rate is cubic in the density, and therefore increases faster than the radiative rate, becomes dominant, causing the observed QY decrease [30]. The remarkable high values of the QY (30% for MAPbI3 and 70% for MAPbI3–*x*Cl*x*) are reached for injected carrier densities close to the amplified spontaneous emission thresholds

reported for these materials [18,19].

10-1 100 101 102 103 Laser pulse fluence (µJ/cm<sup>2</sup>

384 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

)

a) b)

1016 1017 1018 1019 1020 Injected carrier density (cm-3

ence on top axis. The rates, defined as *k*PL <sup>=</sup> *dPL*

QY MAPbI3-xClx

)

QY MAPbI3 108

109

1016 1017 1018 1019 Injected carrier density (cm-3

*PL <sup>t</sup>*=0, are extracted from the data in Figures 2b,c. Such rates

0.1 1 10 Laser pulse fluence (µJ/cm<sup>2</sup>

> MAPbI3 MAPbI3-xClx

> > )

)

Photoluminescence decay rate (s-1

**Figure 4.** Photoluminescence quantum yield and photoluminescence decay rates for the two perovskite samples (MAPbI3 red triangles, MAPbI3xClx blue squares). (a) The emission quantum yields are calculated as TIPL/ *ϕ*, where *ϕ* is the laser pulse fluence. The injected carrier density is calculated multiplying the laser pulse photon fluence by the absorption coefficient of the films. As a reference, the laser pulse fluence directly measured in the experiments is also reported on the top axis. Initially, the QY grows with fluence for both films, as the bimolecular recombination becomes faster and a growing fraction of the injected excitations recombine radiatively, instead of being trapped. At excitations 4×1019 cm–3, non-radiative Auger processes dominate. The absolute QY is scaled to match theoretical predictions. The maximum QY values are about 30% for MAPbI3 and 70% for MAPbI3–*x*Cl*x*. The dotted lines represent predictions from a rate equation model accounting for the main relaxation channels for electrons and holes. Simulations take into ac‐ count the exponential spatial profile of the electron-hole density created by laser pulses. The very good agreement be‐ tween model and data indicates that the main photophysical processes are accounted for in the model. (b) Decay rates for the transient PL signal plotted as a function of the injected carrier density on bottom axis and the laser pulse flu‐

*dt* <sup>⋅</sup> <sup>1</sup>

represent the initial decay and should not be mistaken for the average photoluminescence decay rate obtained by fit‐ ting the entire decay with an exponential function. The error bars represent the standard deviation from a least square fit to an exponential decay and are reported only when they exceed the size of the marker. The rates are very similar

1010

)

0.01

0.1

Quantum Yield

1

Fundamental parameters of semiconductors can be estimated theoretically by analyzing absorption and photoluminescence experimental results. As well as the absorption coefficient and the emission wavelength, bimolecular and Auger recombination rates represent two constants that characterize the material and can be used as figures of merit for some optoe‐ lectronic applications. Values of the bimolecular recombination constant from 10-11 to 10-9 cm3 s–1 have been reported in literature for MAPbI3 and MAPbI3–*x*Cl*x* perovskites [20,30,53]. One would expect that for such an intrinsic process similar values are to be obtained in all samples. Such wide range of reported values can be attributed to the different methods used to calculate it. It is common to extract the bimolecular constant from the PL decay curves, fitting them to a multi-power decay. Such method, however, does not provide reliable estimates whenever applied to noisy curves with several competing processes that are simultaneously relevant; in the particular case of perovskites, a monomolecular decay from trapping, a bimolecular one from radiative recombination and trimolecular one for Auger need to be accounted simulta‐ neously, all convoluted with the temporal resolution of the experimental apparatus. On the other hand, extracting the bimolecular constant from the absorption coefficient is a widely used method in solid state physics that makes use of the symmetry between absorption and emission processes and is completely independent on non-radiative recombination processes. In order to extract the bimolecular coefficient, one has to fit the absorption spectrum with the Kubo-Martin-Schwinger relation, as detailed in references [54,55]. Our estimate of the bimo‐ lecular constant in MAPbI3 perovskites, extracted from absorption, is 2.6x10-10 cm3 s–1, a value included between the two extrema reported in literature.

Differently from the bimolecular constant, the Auger recombination rate could significantly change among samples, as it increases with the defect density of a crystal, which can depend on the synthesis method and the surface morphology [56,57]. Recent works report Auger recombination constants from 10-29 to 10-28cm6 s–1 for pure and mixed methylammonium lead iodide perovskites. Our estimates of 2x10-28 cm6 s–1 for MAPbI3–*x*Cl*x* and 4x10-28 cm6 s–1 for MAPbI3 perovskites are consistent with other published results [20,30]. Auger recombination rate is higher in pure iodide perovskites than in those with chlorine, according to the less crystallinity in samples synthesized without Cl.

## **5. Steady-state photoluminescence**

The experimental results obtained under impulsive regime have provided numerical estimates of fundamental parameters in perovskites and a clearer view of the excited state dynamics. Nevertheless, such kinds of experiments do not investigate what occurs when samples are continuously excited and there is interplay between absorption and relaxation processes, as happens in steady-state operation. In perspective of the realization of perovskite-based working devices, it is important to have a deep understanding of these processes, as real devices work under continuous operation. A reliable instrument of investigation of the steadystate properties is represented by experiments carried out exciting samples under cw pumping or with pulses much longer in time than the PL lifetime of the excited states.

**Figure 5.** Steady-state photoluminescence. (a) Photoluminescence signal as a function of the intensity of the exciting laser. The empty markers represent the measurements obtained with a cw Nd:Yag 532 nm laser; the filled markers are instead measured in quasi-cw conditions, exciting the samples with 300-ns-long laser pulses from a Q-switched 527 nm Nd:Ylf laser; the pulse duration is much longer than all the relevant relaxation rates, so that steady-state conditions are expected to be achieved during laser excitation. The PL signal grows as *I* 3/2 for a wide range of excitations. Investiga‐ tions were extended from laser intensities much lower than the solar one to intensities large enough to generate popu‐ lation inversion and optical gain. The unusual 3/2 power law is attributed to intra-gap trap states with only electrons

or only holes, but not both of them. The dashed black line shows the PL dependence in laser intensity as a result of the trap model, under the assumption that electrons are the trapped species. (b) shows a sketch of the relaxation of optical excitations under steady-state conditions (VB is the valence band and CB the conduction band).

Figure 5a shows our experimental results carried out under cw excitation, spanning light intensities from 10-4 to 104 W/cm2 . Below 100 W/cm2 , the PL signal follows a 3/2 power law in the laser intensity for five decades excitation intensity. Such 3/2 power dependence implies that the radiative recombination is not the only process that governs the electron-hole dynam‐ ics under this regime; otherwise a linear dependence of the PL in the laser intensity would be expected, if radiative processes would be dominant. The particular 3/2 power law could be explained accounting the role of intra-gap states that act only on one kind of carrier, meaning traps only for electrons or only for holes (Figure 5b). Above 102 W/cm2 , the deviation from the 3/2 power law to a linear dependence is attributed to the increase of radiative recombination with respect to carrier trapping [30]. Trap density is estimated to be of the order of 1016-1017cm–3, in agreement with what reported by other works [6]. Despite such large trap density, carrier trapping does not prejudice electronic properties, as PL lifetimes exceed several nanoseconds from intensities smaller than solar illumination to those typical to obtain light amplification. Hence, the resulting carrier mobility is sufficiently high to justify the remarkable transport properties and the efficient charge collection in perovskite-based solar cells reported in literature.

recombination constants from 10-29 to 10-28cm6

386 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

crystallinity in samples synthesized without Cl.

**5. Steady-state photoluminescence**

s–1 for pure and mixed methylammonium lead

CB

Trap state

te

bth beh

t~1/2

~3/2

~ 

Excitation

h~1/2

e~

VB

iodide perovskites. Our estimates of 2x10-28 cm6 s–1 for MAPbI3–*x*Cl*x* and 4x10-28 cm6 s–1 for MAPbI3 perovskites are consistent with other published results [20,30]. Auger recombination rate is higher in pure iodide perovskites than in those with chlorine, according to the less

The experimental results obtained under impulsive regime have provided numerical estimates of fundamental parameters in perovskites and a clearer view of the excited state dynamics. Nevertheless, such kinds of experiments do not investigate what occurs when samples are continuously excited and there is interplay between absorption and relaxation processes, as happens in steady-state operation. In perspective of the realization of perovskite-based working devices, it is important to have a deep understanding of these processes, as real devices work under continuous operation. A reliable instrument of investigation of the steadystate properties is represented by experiments carried out exciting samples under cw pumping

**Figure 5.** Steady-state photoluminescence. (a) Photoluminescence signal as a function of the intensity of the exciting laser. The empty markers represent the measurements obtained with a cw Nd:Yag 532 nm laser; the filled markers are instead measured in quasi-cw conditions, exciting the samples with 300-ns-long laser pulses from a Q-switched 527 nm Nd:Ylf laser; the pulse duration is much longer than all the relevant relaxation rates, so that steady-state conditions are expected to be achieved during laser excitation. The PL signal grows as *I* 3/2 for a wide range of excitations. Investiga‐ tions were extended from laser intensities much lower than the solar one to intensities large enough to generate popu‐ lation inversion and optical gain. The unusual 3/2 power law is attributed to intra-gap trap states with only electrons or only holes, but not both of them. The dashed black line shows the PL dependence in laser intensity as a result of the trap model, under the assumption that electrons are the trapped species. (b) shows a sketch of the relaxation of optical

excitations under steady-state conditions (VB is the valence band and CB the conduction band).

or with pulses much longer in time than the PL lifetime of the excited states.

a) b)

## **6. Organolead halide perovskites as active media in laser devices**

The strong absorption at the band-edge, in addition to the ambipolar transport with high carrier mobility and gap-tunability, makes organometal halide perovskites very promising not only for the realization of solar cells but also as optical gain media in laser devices, as high‐ lighted by many reports showing amplified spontaneous emission under intense photoexci‐ tation [17–19,39,45,58]. Deschler et al. and Xing et al. reported optically pumped lasing and amplified spontaneous emission from perovskites arranged in cavity resonators and in cavityfree configuration, respectively [17,18]. Recently, Zhu et al. reported threshold densities of the order of 1016 cm–3 from single-crystal lead halide perovskite nanowires, which is a value about two orders of magnitude smaller than what was reported by the previous works [39]. In the light of these results, it is clear that organometal halide perovskites have a potential application as active media in laser devices. However, demonstrations reported up to date report ASE under impulsive excitation, a condition well away from real laser devices, which work under continuous operation. Probing emission properties with ns-long-pulse excitation simulates a regime more similar to what happens in a real device, and provides further information about the potentialities and the shortcomings of perovskites as optical gain media [19]. The first report about nanosecond excitation is the one published by Sutherland et al., which shows lasing from MAPbI3 perovskite thin layers deposited on spherical glass resonators, exciting samples with 2-ns-long pulses [45]. Nevertheless, this regime is still far from a real cw excitation and other experiments are required to have a deeper understanding of emission properties in perovskites [19]. Results obtained in different excitation regimes, from the impulsive to the quasi-cw, and improvements of our experiments with respect to other works are reported in the following sections and compared with the most recent researches.

## **7. Lasing and ASE under impulsive excitation**

ASE in MAPbI3 and MAPbBr3 perovskite thin films manifests itself through a sharp peak that appears under short-pulse excitation in the low energy side of the PL spectrum of both films, when the injected carrier density reaches a threshold value. Figure 6 shows emission spectra at different excitation intensities from methylammonium iodide and bromide perovskite thin films at room temperature.

**Figure 6.** Time integrated photoluminescence spectra of MAPbI3 (a) and MAPbBr3 (b) perovskites at 300 K. Emission was excited by 130-fs-long laser pulses with a repetition rate of 1 kHz and 3.18 eV photon energy. Spontaneous and amplified spontaneous emissions were detected by a CCD camera. From top to bottom: laser fluence = 90 μ J/cm2 , 70 μ J/cm2 , 40 μ J/cm2 , 30 μ J/cm2 , 20 μ J/cm2 , 10 μ J/cm2 .

ASE threshold densities are calculated by averaging the laser fluence times the absorption coefficient at the excitation wavelength over the film thickness. Even if samples realized with the same procedure are excited with the same laser source, they show slightly different values of the ASE threshold, but for most samples the threshold is in the interval 2−5 ×1018 cm–3 [19]. Such values are comparable to the ones reported by Xing et al. (1.7×1018 cm–3 in MAPbI3 films), who investigated MAPbX3 (where X = Cl, Br, I) thin films under short-pulse excitation [18]. The dispersion of the ASE thresholds among the same set of samples can be attributed to their surface morphology, which is the main responsible for optical losses. Instead, the large difference in ASE values reported in various publications can be associated to the sample architecture, in addition to the issues originated from the surface morphology. In the experi‐ ments reported by Deschler et al., MAPbI3–*x*Cl*x* perovskite layers were placed between dielectric and evaporated gold top mirrors [17]. The same perovskite layers show ASE in cavity-free configuration, but at fluences above 1 mJ/cm2 , about two orders of magnitude higher than what observed in both our results and the ones reported by Xing et al. [18]. The remarkable low lasing thresholds reported by Zhu et al. (1.5 x1016 cm–3) are attributed to both the high crystallinity of nanowires and their morphology, as the linear shape of nanowires acts as waveguide [39].

Even though such investigations are extremely useful in perspective of the realization of a perovskite-based laser, they are not sufficient to state if these materials could be effectively employed as optical gain media. Actually, experiments carried out under impulsive regime do not take into account any interplay between laser excitation rates and relaxation processes that occur in real cw lasers, as the carrier injection is ultrafast and the emission occurs imme‐ diately after, when pumping has already stopped. Real lasers work under cw operation, a more complicated condition than what happens under impulsive regime, as excitation and emission occur at the same time and optical amplification is observed until the population inversion condition is kept. Under cw excitation, issues connected to warming could considerably affect the light emission properties, especially at high-injected carrier densities (above 1018 cm–3), where non-radiative Auger recombination begins to compete with optical emission [19,30].

## **8. Amplified spontaneous emission under ns excitation**

**7. Lasing and ASE under impulsive excitation**

388 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

PL Intensity (a.u.)

, 20 μ J/cm2

cavity-free configuration, but at fluences above 1 mJ/cm2

1.5 1.6 1.7 1.8 Energy (eV)

, 10 μ J/cm2 .

**Figure 6.** Time integrated photoluminescence spectra of MAPbI3 (a) and MAPbBr3 (b) perovskites at 300 K. Emission was excited by 130-fs-long laser pulses with a repetition rate of 1 kHz and 3.18 eV photon energy. Spontaneous and amplified spontaneous emissions were detected by a CCD camera. From top to bottom: laser fluence = 90 μ J/cm2

ASE threshold densities are calculated by averaging the laser fluence times the absorption coefficient at the excitation wavelength over the film thickness. Even if samples realized with the same procedure are excited with the same laser source, they show slightly different values of the ASE threshold, but for most samples the threshold is in the interval 2−5 ×1018 cm–3 [19]. Such values are comparable to the ones reported by Xing et al. (1.7×1018 cm–3 in MAPbI3 films), who investigated MAPbX3 (where X = Cl, Br, I) thin films under short-pulse excitation [18]. The dispersion of the ASE thresholds among the same set of samples can be attributed to their surface morphology, which is the main responsible for optical losses. Instead, the large difference in ASE values reported in various publications can be associated to the sample architecture, in addition to the issues originated from the surface morphology. In the experi‐ ments reported by Deschler et al., MAPbI3–*x*Cl*x* perovskite layers were placed between dielectric and evaporated gold top mirrors [17]. The same perovskite layers show ASE in

higher than what observed in both our results and the ones reported by Xing et al. [18]. The remarkable low lasing thresholds reported by Zhu et al. (1.5 x1016 cm–3) are attributed to both

2.2 2.3 2.4 Energy (eV)

, 70 μ

, about two orders of magnitude

MAPbBr3

films at room temperature.

J/cm2

, 40 μ J/cm2

, 30 μ J/cm2

ASE in MAPbI3 and MAPbBr3 perovskite thin films manifests itself through a sharp peak that appears under short-pulse excitation in the low energy side of the PL spectrum of both films, when the injected carrier density reaches a threshold value. Figure 6 shows emission spectra at different excitation intensities from methylammonium iodide and bromide perovskite thin

MAPbI3

a) b)

Investigation of emission properties under excitation pulses having duration comparable or longer than the PL lifetime supplies the limits of the ultrafast pumping. Under 4-ns long-pulse excitation, ASE is observable at room temperature in both MAPbI3 and MAPbBr3 thin films at threshold densities of the order of 10 - 15 kW/cm2 (~40−60 μ J/cm2 ) [19]. Such values are comparable with what reported by Sutherland et al. (65 ± 8 μ J/cm2 ), who excited perovskite layers of ~75 nm in thickness with pulses of 2 ns [45]. At this excitation intensity, the PL lifetime of our samples, measured under femtosecond laser excitation, is about 1 ns, which is a value lower than the 4-ns laser pulse.

**Figure 7.** Quasi-steady-state stimulated emission in trihalide perovskite films. Photoluminescence was excited at dif‐ ferent temperatures by 300-ns-long laser pulses with a repetition rate of 6 Hz and 2.35 eV photon energy. (a–d) Timeintegrated photoluminescence spectra of MAPbI3 (red lines) and MAPbBr3 (blue lines) perovskites detected by a CCD camera; ASE occurs at cryogenic temperatures (180 K), but disappears around 220 K.

Despite using excitation pulses of 4 ns could be considered a quasi-cw excitation regime, such pumping is not sufficient to guarantee if perovskites can act as active media in a real device. Using 300-ns-long pulses as laser source, no ASE is observed at room temperature in any perovskite film. A very small hint of amplification is observed at 220 K and a clear ASE peak appears at lower temperatures, as can be observed from Figure 7. Such findings suggest that some processes connected to temperature clearly affect the ASE threshold density under cw excitation. ASE inhibition is likely due to the higher amount of energy deposited by the long pulses, which results in more warming than what obtained under short-pulse excitation [19].

## **9. Optical thermometry**

The plasma temperature can be extracted from PL spectra by fitting the high energy tail of the emission peak to a *Ae* −*E*/*k*B*T*<sup>P</sup> Boltzmann function, where *A* is an arbitrary multiplication factor, *E* the photon energy, *k*B the Boltzmann constant and *T*P the plasma temperature [59].

Inset in Figure 8 illustrates the fitting procedure and shows the plasma temperature as a function of the laser fluence. As one may expect, the plasma temperature increases for increasing laser fluence. The main plasma heating sources under photoexcitation are due to the excess energy of the laser photons with respect to the energy gap and to non-radiative processes from both carrier trapping (at low intensities) and Auger recombination (at high intensities) [19]. Sample excitation with different laser wavelengths, from energies higher than the energy gap to about resonant with it, shows direct effects of excess energy on plasma temperature. Experimental results, reported in Figure 8, show that both contributions are significant, as larger excess energy causes larger warming, but some warming occurs even for quasi-resonant excitation, when Auger is the main source of heating. The role of Auger recombination is also evident at high densities by the rapid increase of the plasma temperature with the excitation intensity.

More significant warming effects can be observed using 300-ns-long excitation pulses and measuring the time-resolved PL evolution at different lattice temperatures *T*<sup>L</sup> and, for each *T*L, at different excitation intensities. Here, the thermometry is applied to each of the spectra in the time-resolved spectrogram and therefore yields information about the temporal dynamics of the plasma temperature. At low excitation intensities, the time-resolved PL signal follows instantaneously the laser pulse, while for densities comparable and higher than ASE threshold, a clear temporal reshape of the PL signal is observed. Such process is evident at high laser intensity even when ASE is not noticeable from PL spectra as shown in Figure 9c,d. The temperature dynamics helps providing a physical explanation, showing that the temporal reshape is accompanied by an increase in the plasma temperature.

Both the temporal reshaping and the observation of ASE only at low lattice temperature under long-pulse excitation can be interpreted in terms of decrease of the radiative efficiency with increasing laser intensity. Such process is a consequence of the plasma warming generated during the intense pulse [19]. An interesting point is that every time ASE is observed, it stops when the plasma temperature overcomes about 370 K, independently from the initial tem‐ perature of the lattice (Figures 9e,f).

Photoexcitations and Emission Processes in Organometal Trihalide Perovskites http://dx.doi.org/10.5772/61282 391

Using 300-ns-long pulses as laser source, no ASE is observed at room temperature in any perovskite film. A very small hint of amplification is observed at 220 K and a clear ASE peak appears at lower temperatures, as can be observed from Figure 7. Such findings suggest that some processes connected to temperature clearly affect the ASE threshold density under cw excitation. ASE inhibition is likely due to the higher amount of energy deposited by the long pulses, which results in more warming than what obtained under short-pulse excitation [19].

The plasma temperature can be extracted from PL spectra by fitting the high energy tail of the

Inset in Figure 8 illustrates the fitting procedure and shows the plasma temperature as a function of the laser fluence. As one may expect, the plasma temperature increases for increasing laser fluence. The main plasma heating sources under photoexcitation are due to the excess energy of the laser photons with respect to the energy gap and to non-radiative processes from both carrier trapping (at low intensities) and Auger recombination (at high intensities) [19]. Sample excitation with different laser wavelengths, from energies higher than the energy gap to about resonant with it, shows direct effects of excess energy on plasma temperature. Experimental results, reported in Figure 8, show that both contributions are significant, as larger excess energy causes larger warming, but some warming occurs even for quasi-resonant excitation, when Auger is the main source of heating. The role of Auger recombination is also evident at high densities by the rapid increase of the plasma temperature

More significant warming effects can be observed using 300-ns-long excitation pulses and measuring the time-resolved PL evolution at different lattice temperatures *T*<sup>L</sup> and, for each *T*L, at different excitation intensities. Here, the thermometry is applied to each of the spectra in the time-resolved spectrogram and therefore yields information about the temporal dynamics of the plasma temperature. At low excitation intensities, the time-resolved PL signal follows instantaneously the laser pulse, while for densities comparable and higher than ASE threshold, a clear temporal reshape of the PL signal is observed. Such process is evident at high laser intensity even when ASE is not noticeable from PL spectra as shown in Figure 9c,d. The temperature dynamics helps providing a physical explanation, showing that the temporal

Both the temporal reshaping and the observation of ASE only at low lattice temperature under long-pulse excitation can be interpreted in terms of decrease of the radiative efficiency with increasing laser intensity. Such process is a consequence of the plasma warming generated during the intense pulse [19]. An interesting point is that every time ASE is observed, it stops when the plasma temperature overcomes about 370 K, independently from the initial tem‐

reshape is accompanied by an increase in the plasma temperature.

*E* the photon energy, *k*B the Boltzmann constant and *T*P the plasma temperature [59].

Boltzmann function, where *A* is an arbitrary multiplication factor,

**9. Optical thermometry**

with the excitation intensity.

perature of the lattice (Figures 9e,f).

−*E*/*k*B*T*<sup>P</sup>

390 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

emission peak to a *Ae*

**Figure 8.** Plasma temperatures of a MAPbI3 film as a function of the laser fluence. The empty markers represent the measurements obtained exciting the sample at a lattice temperature of 300 K with 130-fs-long laser pulses at repetition rate of 1 kHz and different photon energies; the filled markers represent the opposite regime, with 5-ns-long laser puls‐ es at a repetition rate of 10 Hz and different photon energies. The arrows represent the ASE thresholds for the two excitation regimes. The corresponding average excitation power density during the ns laser pulses is also reported on the top axis as a reference; such axis does not apply to measurements under fs excitation. The excess energy was calcu‐ lated as the difference between the excitation photon energy and the energy gap (1.55 eV); the wavelength indicated in brackets is the actual central wavelength of the excitation laser. Inset: plasma temperatures are extracted fitting the high energy tail of photoluminescence spectra with a *Ae* −*E*/*k*B*T*<sup>P</sup> exponential function, representing a Boltzmann ther‐ mal distribution, *E* being the photon energy, *T*<sup>P</sup> the temperature, *k*<sup>B</sup> the Boltzmann constant and *A* an arbitrary mul‐ tiplication factor. The blue lines represent the fitting functions.

As a confirmation, one can measure the ASE threshold densities carried out under short-pulse excitation as a function of the lattice temperature, from 300 K to 160 K. Experimental results, reported in Figure 10, show that ASE threshold fluence follows a quadratic dependence in the lattice temperature in both MAPbI3 and MAPbBr3 perovskites [19]. Since absorption does not change significantly from room temperature to 160 K, such dependence cannot be ascribed to a variation of the injected electron-hole pairs density with temperature, but to the influence of temperature on the radiative recombination rate. Indeed, it is known that the radiative rate of an electron-hole plasma is inversely proportional to the plasma temperature [52].

**Figure 9.** Photoluminescence was excited at different lattice temperatures by 300-ns-long laser pulses with a repetition rate of 6 Hz and 2.35 eV photon energy. (a,b) Time-resolved PL spectrograms of a MAPbI3 film, acquired with a gated intensified camera at different lattice temperatures. (c,d) Corresponding PL temporal profiles, which demonstrate a narrowing of the emission profile with respect to the exciting laser pulse duration; we attributed the effect to stimulat‐ ed emission and Auger recombination. (e,f) Plasma temperature extracted from MAPbI3 photoluminescence spectro‐ grams in panels a,b, at different lattice temperatures, with the fitting procedure reported in the inset of Figure 8. Significant warming occurs during the excitation pulse and limits the duration of ASE.

**Figure 10.** ASE threshold fluence as a function of the lattice temperature in MAPbI3 and MAPbBr3 perovskites, under short-pulse excitation. The dashed lines are fits of measured values to a quadratic dependence of the ASE threshold on temperature.

## **10. Comparison with nitride semiconductors**

Table 1 compares the values of parameters relevant to optical amplification obtained for perovskites to those known for nitride semiconductors at room temperature. Such comparison is very instructive in perspective of the realization of a perovskite-based cw laser. In fact, the advancement of nitride-based lasers has overtaken the problems concerning warming with success and may serve as a useful guide to improve perovskite device performances.


**Table 1.** Comparison between perovskites and nitrides ASE threshold density *n*thr, lifetime at ASE threshold *τ*thr, Auger recombination constant *γ*A and thermal resistance *R*th, respectively. Values are relative to room temperature.

It can be observed that *τ*thr, that is, the carrier lifetime at ASE threshold density, is slightly longer in nitrides than in perovskites, thanks to the significantly lower value of the Auger recombination coefficient *γ*A, which has a value almost two orders of magnitude smaller than in perovskites. Auger recombination is the major responsible for the efficiency reduction in nitride-based LEDs at high power and the value of *γ*A is higher in the presence of disorder due to defects [56,57]. It is therefore important to reduce the Auger recombination coefficient in perovskites, for example, acting in order to decrease both the trap density and the ASE threshold, for example, improving the crystallinity and the surface morphology quality. Another crucial parameter is the thermal resistance *γ*A, which is about four to eight times higher in perovskites than in nitrides. Furthermore, the substrate on which perovskites are deposited could seriously contribute to enhance the value of the thermal resistance, thus increasing the ASE threshold, particularly under long-pulse excitation. Such arguments make the outlook for perovskite-based cw lasers very promising, provided that a great effort should be made in terms of reduction of ASE threshold, Auger recombination and thermal resistance.

## **11. Conclusions**

60

Significant warming occurs during the excitation pulse and limits the duration of ASE.

Normalized PL intensity

Normalized PL intensity

1.8 c)

PL Intensity

392 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

1.7

Energy (eV)

1.6

1.5

1.8

1.7

Energy (eV)

1.6

1.5

180 K

b)

240 K

0 100 200 300 Time (ns)

a)

 MAPbI3 MAPbBr3

0 100 200 300 Time (ns)

**Figure 9.** Photoluminescence was excited at different lattice temperatures by 300-ns-long laser pulses with a repetition rate of 6 Hz and 2.35 eV photon energy. (a,b) Time-resolved PL spectrograms of a MAPbI3 film, acquired with a gated intensified camera at different lattice temperatures. (c,d) Corresponding PL temporal profiles, which demonstrate a narrowing of the emission profile with respect to the exciting laser pulse duration; we attributed the effect to stimulat‐ ed emission and Auger recombination. (e,f) Plasma temperature extracted from MAPbI3 photoluminescence spectro‐ grams in panels a,b, at different lattice temperatures, with the fitting procedure reported in the inset of Figure 8.

d) 240 K

50

40

30

ASE threshold (µJ/cm2

temperature.

)

20

10

0

150 200 250 300 Temperature (K)

400

e) 180 K

f) 240 K

0 100 200 300 Time (ns)

300

Temperature (K)

180 K

Laser Intensity kW/cm<sup>2</sup>

> > 200

500

400

Temperature (K)

300

200

**Figure 10.** ASE threshold fluence as a function of the lattice temperature in MAPbI3 and MAPbBr3 perovskites, under short-pulse excitation. The dashed lines are fits of measured values to a quadratic dependence of the ASE threshold on This chapter has reviewed the most recent progresses concerning the investigation of the photophysical properties in organolead trihalide perovskites. The absorption spectrum in MAPbI3 perovskites shows the influence of excitonic states at the band-edge, even if there is evidence that the majority of photogenerated carriers results in free electrons and holes, in a wide excitation range at room temperature. Based on a large body of optical absorption data, the exciton binding energy is estimated to be 23±4 meV by applying the Elliot's theory of Wannier excitons to published absorption spectra; the estimated value stays constant from cryogenic to room temperature. This value is in contrast with what evidenced by magneto‐ metry measurements, which provide temperature-dependent values of the exciton binding energy. Additional investigations should be addressed to understand the origin of the discrepancy between what observed by optical spectroscopy and magnetometry.

Trap density in perovskites is significant, like in most solution-processed semiconductors, and its effects emerge particularly under continuous excitation. Although trapping has a negative impact in optoelectronic devices, its capture cross-section happens to be low, as attested by the typical PL lifetime in perovskites exceeding several nanoseconds. The resulting values of carrier mobility can exceed 10 cm2 V–1 s–1 and justify the efficient charge collection reported for perovskite-based PV devices.

Emission properties can also play an essential role in the development of new devices based on hybrid perovskites. Many reports provide evidence of amplified spontaneous emission from perovskite thin films under short-pulse excitation, with ASE threshold densities compa‐ rable to those of the best state-of-art organic crystals. Under cw pumping, warming suppresses light amplification, in the same way it happens in nitride semiconductors. Similarities between these latter and hybrid perovskites could suggest a way to improve light emission perform‐ ances in perovskites, in perspective of the realization of a real perovskite-laser.

## **Author details**

Michele Cadelano, Michele Saba, Nicola Sestu, Valerio Sarritzu, Daniela Marongiu, Feipeng Chen, Roberto Piras, Francesco Quochi, Andrea Mura and Giovanni Bongiovanni\*

\*Address all correspondence to: giovanni.bongiovanni@dsf.unica.it

Dipartimento di Fisica, Università degli Studi di Cagliari, Monserrato, Italy

## **References**


[5] Stranks SD, Eperon GE, Grancini G, Menelaou C, Alcocer MJP, Leijtens T, Herz LM, Petrozza A, Snaith HJ. Electron-hole diffusion lengths exceeding 1 micrometer in an organometal trihalide perovskite absorber. Science 2013;342(6156):341.

metry measurements, which provide temperature-dependent values of the exciton binding energy. Additional investigations should be addressed to understand the origin of the

Trap density in perovskites is significant, like in most solution-processed semiconductors, and its effects emerge particularly under continuous excitation. Although trapping has a negative impact in optoelectronic devices, its capture cross-section happens to be low, as attested by the typical PL lifetime in perovskites exceeding several nanoseconds. The resulting values of carrier mobility can exceed 10 cm2 V–1 s–1 and justify the efficient charge collection reported for

Emission properties can also play an essential role in the development of new devices based on hybrid perovskites. Many reports provide evidence of amplified spontaneous emission from perovskite thin films under short-pulse excitation, with ASE threshold densities compa‐ rable to those of the best state-of-art organic crystals. Under cw pumping, warming suppresses light amplification, in the same way it happens in nitride semiconductors. Similarities between these latter and hybrid perovskites could suggest a way to improve light emission perform‐

discrepancy between what observed by optical spectroscopy and magnetometry.

394 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

ances in perovskites, in perspective of the realization of a real perovskite-laser.

Michele Cadelano, Michele Saba, Nicola Sestu, Valerio Sarritzu, Daniela Marongiu,

\*Address all correspondence to: giovanni.bongiovanni@dsf.unica.it

cell. Nanoscale 2011;3(10):4088.

cells. Nature 2013;499(7458):316.

Dipartimento di Fisica, Università degli Studi di Cagliari, Monserrato, Italy

Feipeng Chen, Roberto Piras, Francesco Quochi, Andrea Mura and Giovanni Bongiovanni\*

[1] Kojima A, Teshima K, Shirai Y, Miyasaka T. Organometal halide perovskites as visible-

[2] Im J, Lee C, Lee J, Park S, Park N. 6.5% efficient perovskite quantum-dot-sensitized solar

[3] KimHS,LeeCR,ImJH,LeeKB,MoehlT,MarchioroA,MoonSJ,Humphry-BakerR,Yum JH, Moser JE, Grätzel M, Park NG.Lead iodide perovskite sensitized all-solid-state submicron thin film mesoscopic solar cell with efficiency exceeding 9%. Sci Rep

[4] Burschka J, Pellet N, Moon S, Humphry-Baker R, Gao P, Nazeeruddin MK, Grätzel M. Sequential deposition as a route to high-performance perovskite-sensitized solar

light sensitizers for photovoltaic cells. J Am ChemSoc 2009;131(17):6050.

perovskite-based PV devices.

**Author details**

**References**

2012;2:591.


nescence efficiency and optically pumped lasing in solution-processed mixed halide perovskite semiconductors. J PhysChem Lett 2014;5(8):1421.


[29] Kim J, Ho PKH, Murphy CE, Friend RH. Phase separation in polyfluorene-based conjugated polymer blends: lateral and vertical analysis of blend spin-cast thin films. Macromolecules 2004;37(8):2861.

nescence efficiency and optically pumped lasing in solution-processed mixed halide

[18] Xing G, Mathews N, Lim SS, Yantara N, Liu X, Sabba D, Grätzel M, Mhaisalkar S, Sum TC. Low-temperature solution-processed wide wavelength tunableperovskites

[19] Cadelano M, Sarritzu V, Sestu N, Marongiu D, Chen F, Piras R, Corpino R, Carbo‐ naro CM, Quochi F, Saba M, Mura A, Bongiovanni G. Can trihalide lead perovskites support continuous wave lasing? Adv Optic Mater 2015. DOI: 10.1002/adom.

[20] Wehrenfennig C, Eperon GE, Johnston MB, Snaith HJ, Herz LM. High charge carrier mobilities and lifetimes in organolead trihalide perovskites. Adv Mater 2013;26(10):

[21] Marchioro A, Teuscher J, Friedrich D, Kunst M, van de Krol R, Moehl T, Grätzel M, Moser JE. Unravelling the mechanism of photoinduced charge transfer processes in

[22] D'Innocenzo V, Grancini G, Alcocer MJP, Kandada ARS, Stranks SD, Lee MM, Lan‐ zani G, Snaith HJ, Petrozza A. Excitons versus free charges in organo-lead tri-halide

[23] Roiati V, Colella S, Lerario G, De Marco L, Rizzo A, Listorti A, Gigli G. Investigating charge dynamics in halide perovskite-sensitized mesostructured solar cells. Energy

[24] De Wolf S, Holovsky J, Moon S, Löper P, Niesen B, Ledinsky M, Haug FJ, Yum JH, Ballif C. Organometallic halide perovskites: sharp optical absorption edge and its re‐

[25] Zhao Y, Nardes AM, Zhu K. Solid-State Mesostructured perovskite CH3NH3PbI3so‐ lar cells: charge transport, recombination, and diffusion length. J PhysChem Lett

[26] Giorgi G, Fujisawa J, Segawa H, Yamashita K. Small photocarriereffective masses fea‐ turing ambipolar transport in methylammonium lead iodide perovskite: adensity

[27] Filippetti A, Mattoni A. Hybrid perovskites for photovoltaics: insights from first

[28] Mosconi E, Amat A, Nazeeruddin MK, Grätzel M, De Angelis F. First-principles modeling of mixed halide organometal perovskites for photovoltaic applications. J

lation to photovoltaic performance. J PhysChem Lett 2014;5(6):1035.

functional analysis. J PhysChem Lett 2013;4(24):4213.

principles. Phys Rev B 2014;89(12).

PhysChem C 2013;117(27):13902.

perovskite semiconductors. J PhysChem Lett 2014;5(8):1421.

lead iodide perovskite solar cells. Nat Photonics 2014;8(3):250.

for lasing. Nat Mater. 2014;13(5):476.

396 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

perovskites. Nat Commun2014;5:3586.

Environ Sci 2014;7(6):1889.

2014;5(3):490.

201500229.

1584.


[59] Leo K, Ruhle W, Ploog KH. Hot-carrier energy-loss rates in GaAs/AlxGa1-xAs quan‐ tum wells. Phys Rev B 1988;38(3):1947.

[43] Sturge MD. Optical absorption of gallium arsenide between 0.6 and 2.75 eV. Phys

[44] Sell DD, Lawaetz P. New analysis of direct exciton transitions: application to GaP.

[45] Sutherland BR, Hoogland S, Adachi MM, Wong CTO, Sargent EH. Conformal orga‐ nohalideperovskites enable lasing on spherical resonators. ACS Nano 2014;8(10):

[46] Shah J. Ultrafast Spectroscopy of Semiconductors and Semiconductor Nanostruc‐

[48] Miyata A, Mitioglu A, Plochocka P, Portugall O, Wang JT, Stranks SD, Snaith HJ, Nicholas RJ. Direct measurement of the exciton binding energy and effective masses for charge carriers in organic-inorganic tri-halide perovskites. Nat Phys 2015;11(7):

[49] Kitazawa N, Watanabe Y, Nakamura Y. Optical properties of CH3NH3PbX3 (X= hal‐

[50] Buin A, Pietsch P, Xu J, Voznyy O, Ip AH, Comin R, Sargent EH. Materials process‐

[51] Yu H, Wang F, Xie F, Li W, Chen J, Zhao N. The role of chlorine in the formation process of «CH3NH3PbI3-xClx» perovskite. AdvFunct Mater 2014;24:7102.

[52] Bongiovanni G, Staehli JL. Density dependence of electron-hole plasma lifetime in

[53] Manser JS, Kamat PV. Band filling with free charge carriers in organometal halide

[54] Kubo R. Statistical-mechanical theory of irreversible processes. I. J PhysSocJpn.

[55] Martin PC, Schwinger J. Theory of Many-Particle Systems. I. Phys Rev 1959;115(6):

[57] Kioupakis E, Rinke P, Delaney KT, Van de Walle CG. Indirect auger recombination as a cause of efficiency droop in nitride light-emitting diodes. ApplPhys Lett

[58] Xing J, Liu XF, Zhang Q, Ha ST, Yuan YW, Shen C, Sum TC, Xiong Q. Vaporphase synthesis of organometal halide perovskite nanowires for tunableroom-temperature

[56] Smart G. New clues to LEDs' efficiency droop. Phys Today 2013;66(7):12.

ogen) and their mixed-halide crystals. J Mater Sci 2002;37(17):3585.

ing routes to trap-free halide perovskites. Nano Lett 2014;14(11):6281.

semiconductor quantum wells. Phys Rev B 1992;46(15):9861.

perovskites. Nat Photonics 2014;8(9):737.

nanolasers. Nano Lett 2015;15(7):4571.

1957;12(6):570.

2011;98(16):161107.

1342.

[47] Lanzani G. The Photophysics behind Photovoltaics and Photonics. Wiley; 2012.

Rev 1963;129(2835).

tures. Springer; 1999.

10947.

582.

Phys Rev Lett 1971;26(6):331.

398 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications


**Perovskite Materials: Applications**

## **Optical Absorption, Charge Separation and Recombination Dynamics in Pb and Sn/Pb Cocktail Perovskite Solar Cells and Their Relationships to the Photovoltaic Properties**

Shen Qing, Ogomi Yuhei, Toyoda Taro, Yoshino Kenji and Hayase Shuzi

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/62101

## **Abstract**

Due to the unique characteristics such as simple low-temperature preparation method and high efficiency with a record of over 20%, organometal trihalide perovskite (CH3NH3PbI3)-based solid-state hybrid solar cells have attracted an increasing interest since 2012 when it was reported. During the last several years, some of the fundamental photophysical properties of perovskite related to the high photovoltaic performance have been investigated. Optical absorption, charge separation and recombination are very im‐ portant factors influencing the perovskite solar cell performance. In this chapter, our re‐ cent results of optical absorption, charge separation (electron and hole injection) and charge recombination dynamics at each interface in perovskite solar cells, and their rela‐ tionships to photovoltaic properties will be introduced. Our results suggest that charge recombination is a key factor in improving the performance of the perovskite solar cells.

**Keywords:** Urbach energy, charge separation, charge recombination, optical absorption, Sn/Pb cocktail perovskite, perovskite

## **1. Introduction**

Organolead halide perovskites in the format of AMX3 (A=organic molecule, e.g., CH3NH3(MA), B=Pb, X=Cl, Br, and I) can be easily crystallized from solution at relatively low temperature (i.e., ≤100 °C), which makes it possible to use them as light absorbing materials in different kinds of solar cells. Following the recently reported certified high power conver‐ sion efficiency (PCE) of over 20%, [1-12] the interest in organolead halide perovskite-based

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

organic-inorganic hybrid solid-state solar cells has increased dramatically over the past several years. The higher PCEs of organolead halide perovskites (especially MAPbI3) result from their unique properties that are key for achieving high photovoltaic performance, which are (1) a direct band gap and a high optical absorption coefficient; [13, 14] (2) large dielectric coefficient leading to smaller exciton binding energy; [15] (3) long photoexcited carrier lifetimes (>100 ns) and long diffusion lengths (100 – 1000 nm or even longer); [16, 17] (4) no deep state defects and very small Urbach energy. [18]

It is reasonable to expect that further improvements in photovoltaic performance can be achieved by increasing the light harvesting up to NIR wavelengths of 1000 nm, since MAP‐ bI3 perovskite only absorbs light at wavelengths below 800 nm constrained by its optical band gap of 1.5 eV. In addition, practically, Pb-free organometal halide perovskites are preferred due to the potential toxicology issue of Pb. Replacing Pb with Sn or mixing Pb and Sn in organometal halide perovskites can result in increased light harvesting in the NIR region up to 1000 nm [19, 20] and, at the same time, reduce the toxicity issue related to Pb. Several research groups have reported, very recently, Sn-based or Pb/Sn cocktail MASnxPb1-xI3 (0≦x≦1) perovskite solar cells. [21-24] However, it is found that the PCE of Sn/Pb cocktail MASnxPb1 xI3 perovskite solar cells is far inferior to that of MAPbI3 perovskite solar cells.

In order to improve the photovoltaic performance of Pb and Sn/Pb cocktail perovskite solar cells, it is critical to gain a thorough understanding of the optical absorption properties, the photoexcited carrier lifetimes, as well as the charge separation and recombination dynamics at each interface. In this chapter, we will focus on recent studies of the optical absorption, photoexcited carrier lifetime, the charge separation and charge recombination dynamics at each interface in perovskite solar cells, including Pb-based and Sn/Pb cocktail perovskite solar cells. The relationships of each of these physical properties to the photovoltaic performance of the solar cells will be discussed and the methodologies for improving the photovoltaic performance of perovskite solar cells will be proposed.

## **2. Experimental**

## **2.1. Sample preparation**

Samples of Pb-based perovskite hybrid solar cells were prepared by the following method. [25] F-doped SnO2 layered glass (FTO glass, Nippon Sheet Glass Co. Ltd) was patterned using Zn powder and 6 N HCl aqueous solution. Titanium diisopropoxide bis(acetylacetonate) solution in ethanol was sprayed onto this patterned FTO glass at 300 °C to prepare compact TiO2 layers. Porous TiO2 layers were fabricated by spin-coating TiO2 pastes of different nanoparticle sizes (18 nm: PST-18NR or 30 nm: PST-30NRD, JGC Catalysts and Chemicals Ltd.) in ethanol (TiO2 paste : ethanol = 1:2.5 weight ratio for PST-18NR or TiO2 paste : ethanol = 2:7 weight ratio for PST-30NRD), followed by heating the substrates at 550 °C for 30 min. For some TA measurements, glass, instead of FTO was used as the substrate and a porous Y2O3 layer was fabricated on the glass substrates. Next, CH3NH3I and PbCl2 were mixed with a 3:1 molar ratio for preparing a 40 % solution of perovskite in N,N-dimethylformamide and the mixture was spin-coated on the TiO2 and Y2O3 porous substrates. After heating at 100 °C for 45 minutes, the substrates were spin-coated with a mixture of 55 mM of tert-butylpyridine, 9 mM of lithium bis(trifluoromethylsyfonyl)imide salt, and 68 mM of *spiro-*OMeTAD. Finally, Ag and Au electrodes were fabricated by vacuum deposition for the photovoltaic measurements. The photovoltaic performance was evaluated using an AM1.5G 100 mW/cm2 irradiance solar simulator (CEP-2000, Bunkoukeiki Inc) with a 0.4 cm x 0.4 cm mask.

The Sn/Pb cocktail perovskite samples were prepared using the following method. [23] PbI2 (Purity: 99.999 %), SnI2, and regioregular poly(3-hexylthiophene-2,5-diyl) (P3HT) were purchased from Sigma Aldrich and used as received. F-doped SnO2 coated glass ((FTO glass), Nippon Sheet Glass Co. Ltd) was patterned using Zn and a 6 N HCl aqueous solution. A compact TiO2 layer was produced by spraying titanium diisopropoxide bis(acetylacetonate) solution in ethanol onto this patterned glass at 300 °C. The substrate was then dipped in a 40 mM solution of TiCl4 in water for 30 min. Then the substrate was baked at 500 °C for 20 min. Next, a porous TiO2 layer was produced by spin-coating with a TiO2 paste (PST-18NR, JGC Catalysts and Chemicals Ltd.) in ethanol (TiO2 paste : ethanol = 2:7 weight ratio). The substrate was then heated at 550 °C for 30 min. For certain TA measurements, glass substrates other than FTO ones, with a porous Al2O3 layer deposited on them, were used. The substrates were then spin-coated with a mixture of SnI2, PbI2, and CH3NH3I (0.5 : 0.5 : 1.0 (molar ratio)) in dime‐ thylformamide (DMF) (40 wt %), and they were baked at 70 °C for 30 min. Next, P3HT in chlorobenzene solution (15 mg/ml) was spin-coated on the prepared perovskite layer and the substrate was put in nitrogen at ambient temperature for 1 h. For conducting TA measure‐ ments, samples of Sn/Pb cocktail perovskite deposited on both Al2O3 and TiO2 substrates with and without P3HT as a hole transport material (HTM) were employed. Ag and Au electrodes were finally deposited by vacuum deposition for characterizing the photovoltaic performance, which was evaluated using an AM 1.5G 100 mW/cm2 irradiance solar simulator (CEP-2000SRR, Bunkoukeiki Inc) with a 0.4 cm x 0.4 cm mask.

## **2.2. Characterization methods**

organic-inorganic hybrid solid-state solar cells has increased dramatically over the past several years. The higher PCEs of organolead halide perovskites (especially MAPbI3) result from their unique properties that are key for achieving high photovoltaic performance, which are (1) a direct band gap and a high optical absorption coefficient; [13, 14] (2) large dielectric coefficient leading to smaller exciton binding energy; [15] (3) long photoexcited carrier lifetimes (>100 ns) and long diffusion lengths (100 – 1000 nm or even longer); [16, 17] (4) no deep state defects

It is reasonable to expect that further improvements in photovoltaic performance can be achieved by increasing the light harvesting up to NIR wavelengths of 1000 nm, since MAP‐ bI3 perovskite only absorbs light at wavelengths below 800 nm constrained by its optical band gap of 1.5 eV. In addition, practically, Pb-free organometal halide perovskites are preferred due to the potential toxicology issue of Pb. Replacing Pb with Sn or mixing Pb and Sn in organometal halide perovskites can result in increased light harvesting in the NIR region up to 1000 nm [19, 20] and, at the same time, reduce the toxicity issue related to Pb. Several research groups have reported, very recently, Sn-based or Pb/Sn cocktail MASnxPb1-xI3 (0≦x≦1) perovskite solar cells. [21-24] However, it is found that the PCE of Sn/Pb cocktail MASnxPb1-

In order to improve the photovoltaic performance of Pb and Sn/Pb cocktail perovskite solar cells, it is critical to gain a thorough understanding of the optical absorption properties, the photoexcited carrier lifetimes, as well as the charge separation and recombination dynamics at each interface. In this chapter, we will focus on recent studies of the optical absorption, photoexcited carrier lifetime, the charge separation and charge recombination dynamics at each interface in perovskite solar cells, including Pb-based and Sn/Pb cocktail perovskite solar cells. The relationships of each of these physical properties to the photovoltaic performance of the solar cells will be discussed and the methodologies for improving the photovoltaic

Samples of Pb-based perovskite hybrid solar cells were prepared by the following method. [25] F-doped SnO2 layered glass (FTO glass, Nippon Sheet Glass Co. Ltd) was patterned using Zn powder and 6 N HCl aqueous solution. Titanium diisopropoxide bis(acetylacetonate) solution in ethanol was sprayed onto this patterned FTO glass at 300 °C to prepare compact TiO2 layers. Porous TiO2 layers were fabricated by spin-coating TiO2 pastes of different nanoparticle sizes (18 nm: PST-18NR or 30 nm: PST-30NRD, JGC Catalysts and Chemicals Ltd.) in ethanol (TiO2 paste : ethanol = 1:2.5 weight ratio for PST-18NR or TiO2 paste : ethanol = 2:7 weight ratio for PST-30NRD), followed by heating the substrates at 550 °C for 30 min. For some TA measurements, glass, instead of FTO was used as the substrate and a porous Y2O3 layer was fabricated on the glass substrates. Next, CH3NH3I and PbCl2 were mixed with a 3:1 molar ratio for preparing a 40 % solution of perovskite in N,N-dimethylformamide and the mixture was spin-coated on the TiO2 and Y2O3 porous substrates. After heating at 100 °C for 45 minutes,

xI3 perovskite solar cells is far inferior to that of MAPbI3 perovskite solar cells.

performance of perovskite solar cells will be proposed.

**2. Experimental**

**2.1. Sample preparation**

and very small Urbach energy. [18]

404 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

## *2.2.1. Optical absorption measurements*

A gas-microphone photoacoustic (PA) technique [26] was used to study the optical absorption properties of the samples. The light source was a 300 W xenon arc lamp. By passing the light through a monochromator, a monochromatic light beam was obtained. This beam was modulated with a mechanical chopper and focused onto the surface of the sample placed in a sealed PA cell filled with N2 gas. The measurements of PA spectrum were carried out in the wavelength range of 500–1200 nm with a modulation frequency of 33 Hz at room temperature. The PA signal was measured by first passing the microphone output through a preamplifier and then to a lock-in amplifier. The PA spectra were normalized with the PA spectrum obtained from a carbon black sheet.

## *2.2.2. Transient Absorption (TA) Measurements*

Two different TA setups were used: the charge separation (electron injection and hole injection) dynamics was characterized by a femtosecond TA technique (fs-TA) [27, 28] and charge recombination dynamics was characterized by a nanosecond TA technique (ns-TA) [28-31]. In the fs-TA setup for characterizing the charge separation, [27, 28, 31] the laser source used was a titanium/sapphire laser (CPA-2010, Clark-MXR Inc.) with a wavelength of 775 nm, a repetition rate of 1 kHz, and a pulse width of 150 fs. The light was separated into two parts. One part was used as a probe pulse. The other part was used to pump an optical parametric amplifier (OPA) (a TOAPS from Quantronix) to generate light pulses with a wavelength tunable from 290 nm to 3 μm. This was used as a pump light to excite the sample. In this study, a pump light wavelength of 470 nm and a probe beam wavelength of 775 nm were used. Deschler and co-workers [32] reported very recently that the TA response of ground state bleaching (GSB) for CH3NH3PbClI2 on glass peaked at 1.65 eV and was spectrally unchanged up to times beyond 200 ns. Therefore, the probe light wavelength of 775 nm (i.e., the photon energy of 1.64 eV) used in this study is appropriate to monitor the GSB of CH3NH3PbClI2 on Y2O3 and TiO2 substrates with or without *spiro-*OMeTAD, and thus electron transfer and hole transfer processes at each interface can be investigated systematically. On the other hand, in the case of Sn/Pb cocktail perovskite, the probe light of 775 nm was used to detect the photo‐ excited carrier absorption.

In the ns-TA setup for characterizing the charge recombination [28-31], the pump light source was an OPO (Surelite II – 10FP) output excited by a Nd:YAG nanosecond pulse laser (Panther, Continuum, Electro-Optics Inc.), with a pulse width of 5 ns and a repetition rate of 0.5 Hz. A pulse light with a wavelength of 470 nm was used as the pump light to excite the sample. The probe light was produced from a fiber coupled CW semiconductor. Three different probe wavelengths, of 785 nm, 658 nm and 1310 nm, were used. Specifically, the probe beam of 785 nm was employed to measure the TA responses of GSB for MAPbClI2 on Y2O3 substrates, i.e., the recombination of electrons and holes in MAPbClI2. [ 32] The probe beam of 658 nm was used to measure the trapped electrons in TiO2 [33] based on the research of Yoshihara and coworkers, which was used to investigate charge recombination between the electrons in TiO2 and the holes in the perovskite. The probe beam of 1310 nm was used to monitor the holes in *spiro*-OMeTAD [34] and thus measure the charge recombination between holes in *spiro*-OMeTAD and electrons in TiO2 and/or in perovskite. For all measurements, the pump and probe lights were irradiated from the glass side and the TA measurements were carried out in a nitrogen atmosphere.

## **3. Optical Absorption Study: Bandgap and Urbach Energy of Pb and Sn/Pb cocktail perovskite**

Figure 1 shows the optical absorption spectra of the Pb (MAPbI2) and Sn/Pb cocktail (MASn0.5Pb0.5I3) perovskite samples on a porous TiO2 substrate measured using the PA technique (we refer to this as the PA spectrum in the following) at room temperature. From the position of the shoulder in each PA spectrum, [35] the bandgap energies of MAPbI2 and MASn0.5Pb0.5I3 are determined to be 1.52 and 1.21 eV, respectively, which are almost the same as those given in our earlier reports. [23, 25] We find that, below the shoulder, the trend of the absorption coefficient is exponential. The slope of this exponential tail, known as the Urbach irradiated from the glass side and the TA measurements were carried out in a nitrogen atmosphere.

3. Optical Absorption Study: Bandgap and Urbach Energy of Pb and Sn/Pb cocktail perovskite

between holes in spiro-OMeTAD and electrons in TiO2 and/or in perovskite. For all measurements, the pump and probe lights were

Figure 1 shows the optical absorption spectra of the Pb (MAPbI2) and Sn/Pb cocktail (MASn0.5Pb0.5I3) perovskite samples on

a porous TiO2 substrate measured using the PA technique (we refer to this as the PA spectrum in the following) at room temperature. From the position of the shoulder in each PA spectrum,<sup>35</sup> the bandgap energies of MAPbI2 and MASn0.5Pb0.5I3 are

determined to be 1.52 and 1.21 eV, respectively, which are almost the same as those given in our earlier reports.23,25 We find that,

18,31,36

tail, corresponds to the absorption tail states, and is usually quantitatively expressed by the Urbach Energy *E*U. [18, 31, 36] Investigation of these exponential tails can offer important information on the band structure, disorder, defects, impurities, and electron-phonon inter‐ actions in semiconductor materials. The numerical relationship between the optical absorption coefficient *α* in the exponential tail and the photon energy (*hυ*) is shown in eq. (1), through which the values of *E*U of the samples can be calculated. [37, 38] Investigation of these exponential tails can offer important information on the band structure, disorder, defects, impurities, and electron-phonon interactions in semiconductor materials. The numerical relationship between the optical absorption coefficient α in the exponential tail and the photon energy (hυ) is shown in eq. (1), through which the values of EU of the samples can be calculated.37,38

recombination dynamics was characterized by a nanosecond TA technique (ns-TA) [28-31]. In the fs-TA setup for characterizing the charge separation, [27, 28, 31] the laser source used was a titanium/sapphire laser (CPA-2010, Clark-MXR Inc.) with a wavelength of 775 nm, a repetition rate of 1 kHz, and a pulse width of 150 fs. The light was separated into two parts. One part was used as a probe pulse. The other part was used to pump an optical parametric amplifier (OPA) (a TOAPS from Quantronix) to generate light pulses with a wavelength tunable from 290 nm to 3 μm. This was used as a pump light to excite the sample. In this study, a pump light wavelength of 470 nm and a probe beam wavelength of 775 nm were used. Deschler and co-workers [32] reported very recently that the TA response of ground state bleaching (GSB) for CH3NH3PbClI2 on glass peaked at 1.65 eV and was spectrally unchanged up to times beyond 200 ns. Therefore, the probe light wavelength of 775 nm (i.e., the photon energy of 1.64 eV) used in this study is appropriate to monitor the GSB of CH3NH3PbClI2 on Y2O3 and TiO2 substrates with or without *spiro-*OMeTAD, and thus electron transfer and hole transfer processes at each interface can be investigated systematically. On the other hand, in the case of Sn/Pb cocktail perovskite, the probe light of 775 nm was used to detect the photo‐

In the ns-TA setup for characterizing the charge recombination [28-31], the pump light source was an OPO (Surelite II – 10FP) output excited by a Nd:YAG nanosecond pulse laser (Panther, Continuum, Electro-Optics Inc.), with a pulse width of 5 ns and a repetition rate of 0.5 Hz. A pulse light with a wavelength of 470 nm was used as the pump light to excite the sample. The probe light was produced from a fiber coupled CW semiconductor. Three different probe wavelengths, of 785 nm, 658 nm and 1310 nm, were used. Specifically, the probe beam of 785 nm was employed to measure the TA responses of GSB for MAPbClI2 on Y2O3 substrates, i.e.,

to measure the trapped electrons in TiO2 [33] based on the research of Yoshihara and coworkers, which was used to investigate charge recombination between the electrons in TiO2 and the holes in the perovskite. The probe beam of 1310 nm was used to monitor the holes in *spiro*-OMeTAD [34] and thus measure the charge recombination between holes in *spiro*-OMeTAD and electrons in TiO2 and/or in perovskite. For all measurements, the pump and probe lights were irradiated from the glass side and the TA measurements were carried out in

**3. Optical Absorption Study: Bandgap and Urbach Energy of Pb and Sn/Pb**

Figure 1 shows the optical absorption spectra of the Pb (MAPbI2) and Sn/Pb cocktail (MASn0.5Pb0.5I3) perovskite samples on a porous TiO2 substrate measured using the PA technique (we refer to this as the PA spectrum in the following) at room temperature. From the position of the shoulder in each PA spectrum, [35] the bandgap energies of MAPbI2 and MASn0.5Pb0.5I3 are determined to be 1.52 and 1.21 eV, respectively, which are almost the same as those given in our earlier reports. [23, 25] We find that, below the shoulder, the trend of the absorption coefficient is exponential. The slope of this exponential tail, known as the Urbach

32] The probe beam of 658 nm was used

excited carrier absorption.

a nitrogen atmosphere.

**cocktail perovskite**

the recombination of electrons and holes in MAPbClI2. [

406 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Figure 1 Optical absorption spectra of MAPbI3 and Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite measured using a photoacoustic (PA) technique. The bandgap energies were determined to be 1.52 eV and 1.21 eV, respectively. The Urbach energies were determined to be 22 **Figure 1.** Optical absorption spectra of MAPbI3 and Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite measured using a photoa‐ coustic (PA) technique. The bandgap energies were determined to be 1.52 eV and 1.21 eV, respectively. The Urbach energies were determined to be 22 meV and 34 meV, respectively.

meV and 34 meV, respectively.

$$\alpha = \alpha\_0 \exp\left(\frac{hv - hv\_0}{E\_u}\right) \tag{1}$$

4 � = � exp � � (1) where h is Planck's constant, and α0, υ0, Eu are fitting parameters. The values of Eu for Pb and Sn/Pb cocktail perovskites were determined to be 22 meV and 34 meV, respectively. The value of Eu for Sn/Pb cocktail perovskite is larger than that for the Pb perovskite. It can be assumed that the value of Eu is a reflection of the disorder and/or defects in the semiconductor crystal.18,36-38 where *h* is Planck's constant, and *α*0, *υ*0, *Eu* are fitting parameters. The values of *Eu* for Pb and Sn/Pb cocktail perovskites were determined to be 22 meV and 34 meV, respectively. The value of *Eu* for Sn/Pb cocktail perovskite is larger than that for the Pb perovskite. It can be assumed that the value of *Eu* is a reflection of the disorder and/or defects in the semiconductor crystal. [18, 36-38] Thus, the width of the exponential tail increases with increasing density of defects. Therefore, the larger *Eu* determined for Sn/Pb cocktail perovskite (i.e., 34 meV) implies that there could be higher defect states in the sample compared to MAPbI3 (i.e., 22 meV). This will lead to faster recombination of photoexcited carriers in the Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite and lower values of *V*oc and *FF*, which we will discuss in detail below together with the TA measurement results.

Relationships to the Photovoltaic Properties25,39,40

#### **4. Dynamics of Photoexcited Carrier Recombination and Charge Transfer in Pb-based Perovskite Solar Cells and their Relationships to the Photovoltaic Properties [25, 39, 40]** measurements were conducted for MAPbI3 on either Y2O3 or TiO2 substrates, with and without a spiro-OMeTAD layer. For perovskite deposited on Y2O3 substrate, the lifetime of photoexcited charge carriers in MAPbI3 and the charge separation

Thus, the width of the exponential tail increases with increasing density of defects. Therefore, the larger Eu determined for Sn/Pb

cocktail perovskite (i.e., 34 meV) implies that there could be higher defect states in the sample compared to MAPbI3 (i.e., 22

meV). This will lead to faster recombination of photoexcited carriers in the Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite and lower

4. Dynamics of Photoexcited Carrier Recombination and Charge Transfer in Pb-based Perovskite Solar Cells and their

To perform systematic investigations on the charge separation and recombination dynamics in perovskite solar cells, TA

values of Voc and FF, which we will discuss in detail below together with the TA measurement results.

To perform systematic investigations on the charge separation and recombination dynamics in perovskite solar cells, TA measurements were conducted for MAPbI3 on either Y2O3 or TiO2 substrates, with and without a *spiro*-OMeTAD layer. For perovskite deposited on Y2O3 substrate, the lifetime of photoexcited charge carriers in MAPbI3 and the charge separation dynamics at the MAPbI3/*spiro*-OMeTAD interface were studied by the fs-TA technique. The pump light wavelength used to excite MAPbI3 only was 470 nm, and the probe light wave‐ length used was 775 nm, which is just at the optical absorption edge of MAPbI3. Figure 2 shows the normalized TA response of MAPbI3/Y2O3 for a time scale of up to 3 ns with a lower pump light intensity (0.9 μJ/cm2 ) with and without the *spiro*-OMeTAD layer. We can observe a bleaching signal with a very slow decay in the figure. Because electron injection from the perovskite to Y2O3 cannot occur due to the large band gap of Y2O3, it is reasonable to attribute the slow decay to the slow recombination process of photoexcited charge carriers in MAPbI3. Two processes were found in the TA decay process and the TA signal obtained can be fitted to the biexponential function shown below very well: dynamics at the MAPbI3/spiro-OMeTAD interface were studied by the fs-TA technique. The pump light wavelength used to excite MAPbI3 only was 470 nm, and the probe light wavelength used was 775 nm, which is just at the optical absorption edge of MAPbI3. Figure 2 shows the normalized TA response of MAPbI3/Y2O3 for a time scale of up to 3 ns with a lower pump light intensity (0.9 µJ/cm<sup>2</sup> ) with and without the spiro-OMeTAD layer. We can observe a bleaching signal with a very slow decay in the figure. Because electron injection from the perovskite to Y2O3 cannot occur due to the large band gap of Y2O3, it is reasonable to attribute the slow decay to the slow recombination process of photoexcited charge carriers in MAPbI3. Two processes were found in the TA decay process and the TA signal obtained can be fitted to the biexponential function shown below very well:

Figure 2 Normalized TA responses of MAPbClI2/Y2O<sup>3</sup> with and without **Figure 2.** Normalized TA responses of MAPbClI2/Y2O3 with and without *spiro*-OMeTAD as a hole transport material (HTM). The red solid lines represent the fitting results with eq. (2). (reference: 25)

$$Y = A\_1 e^{-q\mathbf{f}\_1} + A\_2 e^{-q\mathbf{f}\_2} \tag{2}$$

5

where t1 and t2 are two time constants, A1 and A2 are the contributions from the corresponding components. The time constants t1 and t2 of

the two charge recombination processes were calculated to be 34 ps (A1/(A1+A2): 9%) and much larger than 10 ns (A2/(A1+A2): 91%),

respectively. The faster decay process could be attributed to nonradiative recombination of electrons and holes through defects or trap

<sup>1</sup> <sup>2</sup> / / tt tt Y eA eA <sup>−</sup> <sup>−</sup> = + where *t*1 and *t*2 are two time constants, *A*1 and *A*2 are the contributions from the corresponding components. The time constants *t*1 and *t*2 of the two charge recombination processes were calculated to be 34 ps (*A*1/(*A*1+*A*2): 9%) and much larger than 10 ns (*A*2/(*A*1+*A*2): 91%), respec‐

(2)

1

2

measurements were conducted for MAPbI3 on either Y2O3 or TiO2 substrates, with and without a spiro-OMeTAD layer. For perovskite deposited on Y2O3 substrate, the lifetime of photoexcited charge carriers in MAPbI3 and the charge separation dynamics at the MAPbI3/spiro-OMeTAD interface were studied by the fs-TA technique. The pump light wavelength used to excite MAPbI3 only was 470 nm, and the probe light wavelength used was 775 nm, which is just at the optical absorption edge of MAPbI3. Figure 2 shows the normalized TA response of MAPbI3/Y2O3 for a time scale of up to 3 ns with a lower pump light ) with and without the spiro-OMeTAD layer. We can observe a bleaching signal with a very slow decay in tively. The faster decay process could be attributed to nonradiative recombination of electrons and holes through defects or trap states in MAPbI3 or at the MAPbI3/Y2O3 interface. The relative contribution of the faster decay is less than 10%, which suggests that the defect or trap state density in MAPbI3 or at the interface is very small, which is consistent with a thermally stimulated current measurement. [39, 40] The slower decay process could be attributed to the recombination of free electrons and holes in MAPbI3. This result suggests that the lifetime of the photoexcited charge carriers was very long, which was confirmed to be as much as the order of microseconds by the ns-TA measurement results shown in Figure 3. The TA decay was fitted with eq. (2), and two decay processes were found with lifetimes of 3.7±0.1 μs (70%) and 60±1 μs (30%), respectively. Combining the fs-TA and ns-TA results for MAPbI3/Y2O3, the lifetime of the photoexcited charge carriers in MAPbI3 can be mostly considered to be as long as the order of microseconds. confirmed to be as long as µs by the ns-TA measurement result as shown in Figure 3. By fitting the TA decay with eq. (1), it was found that two decay processes existed with lifetimes of 3.7±0.1 µs (70%) and 60±1 µs (30%), respectively. Combing the fs-TA and ns-TA results for MAPbI3/Y2O3, the lifetime of photoexcited charge carriers in MAPbI<sup>3</sup> can be mainly considered to be as long as µs.

**4. Dynamics of Photoexcited Carrier Recombination and Charge Transfer**

To perform systematic investigations on the charge separation and recombination dynamics in perovskite solar cells, TA measurements were conducted for MAPbI3 on either Y2O3 or TiO2 substrates, with and without a *spiro*-OMeTAD layer. For perovskite deposited on Y2O3 substrate, the lifetime of photoexcited charge carriers in MAPbI3 and the charge separation dynamics at the MAPbI3/*spiro*-OMeTAD interface were studied by the fs-TA technique. The pump light wavelength used to excite MAPbI3 only was 470 nm, and the probe light wave‐ length used was 775 nm, which is just at the optical absorption edge of MAPbI3. Figure 2 shows the normalized TA response of MAPbI3/Y2O3 for a time scale of up to 3 ns with a lower pump

bleaching signal with a very slow decay in the figure. Because electron injection from the perovskite to Y2O3 cannot occur due to the large band gap of Y2O3, it is reasonable to attribute the slow decay to the slow recombination process of photoexcited charge carriers in MAPbI3. Two processes were found in the TA decay process and the TA signal obtained can be fitted

Figure 2 Normalized TA responses of MAPbClI2/Y2O<sup>3</sup> with and without spiro-OMeTAD as a hole transport material (HTM). The red solid lines represent the

> 1 2 1 2

where *t*1 and *t*2 are two time constants, *A*1 and *A*2 are the contributions from the corresponding components. The time constants *t*1 and *t*2 of the two charge recombination processes were calculated to be 34 ps (*A*1/(*A*1+*A*2): 9%) and much larger than 10 ns (*A*2/(*A*1+*A*2): 91%), respec‐

*t t t t Y Ae Ae* - - = + (2)

**Figure 2.** Normalized TA responses of MAPbClI2/Y2O3 with and without *spiro*-OMeTAD as a hole transport material

) with and without the *spiro*-OMeTAD layer. We can observe a

the figure. Because electron injection from the perovskite to Y2O3 cannot occur due to the large band gap of Y2O3, it is

Thus, the width of the exponential tail increases with increasing density of defects. Therefore, the larger Eu determined for Sn/Pb

cocktail perovskite (i.e., 34 meV) implies that there could be higher defect states in the sample compared to MAPbI3 (i.e., 22

meV). This will lead to faster recombination of photoexcited carriers in the Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite and lower

4. Dynamics of Photoexcited Carrier Recombination and Charge Transfer in Pb-based Perovskite Solar Cells and their

values of Voc and FF, which we will discuss in detail below together with the TA measurement results.

5

where t1 and t2 are two time constants, A1 and A2 are the contributions from the corresponding components. The time constants t1 and t2 of

the two charge recombination processes were calculated to be 34 ps (A1/(A1+A2): 9%) and much larger than 10 ns (A2/(A1+A2): 91%),

respectively. The faster decay process could be attributed to nonradiative recombination of electrons and holes through defects or trap

**in Pb-based Perovskite Solar Cells and their Relationships to the**

**Photovoltaic Properties [25, 39, 40]**

Relationships to the Photovoltaic Properties25,39,40

408 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

light intensity (0.9 μJ/cm2

intensity (0.9 µJ/cm<sup>2</sup>

below very well:

to the biexponential function shown below very well:

fitting results with eq. (2). (reference: 25)

(2)

/ 1 tt tt Y eA eA <sup>−</sup> <sup>−</sup> = +

<sup>1</sup> <sup>2</sup> / 2

(HTM). The red solid lines represent the fitting results with eq. (2). (reference: 25)

Figure 3 TA responses of MAPbI3/Y2O3 without spiro-OMeTAD for a time scale of **Figure 3.** TA responses of MAPbI3/Y2O3 without *spiro*-OMeTAD for a time scale of 200 μs measured with a pump light wavelength of 470 nm and a probe light wavelength of 785 nm. The red solid lines represent the fitting results with eq. (2). (reference: 25)

200 µs measured with a pump light wavelength of 470 nm and a probe light wavelength of 785 nm. The red solid lines represent the fitting results with eq. (1). (reference: 25) The TA response of MAPbI3/Y2O3 with *spiro*-OMeTAD was fitted to eq. (2), and only one exponential decay with a time constant of 16±2 ns was found. Compared with the long lifetime of the TA decay in Y2O3/MAPbI3, the fast decay process in Y2O3/MAPbI3/*spiro*-OMeTAD can be considered to originate from the photoexcited hole injection from MAPbI3 to *spiro*-OMe‐ TAD. This result indicates that charge separation occurred at the interface between MAPbI3 and *spiro*-OMeTAD. [25]

By fitting the TA response of MAPbI3/Y2O3 with spiro-OMeTAD to eq. (1), we found that only one exponential decay with a time constant of 16±2 ns appeared. Comparing with the long lifetime of the TA decay in Y2O3/MAPbI3, the fast decay Figure 4 shows the TA response of MAPbI3/Y2O3 with a layer of *spiro*-OMeTAD measured at a probe wavelength of 1310 nm. This result clearly indicates that charge separation at the MAPbI3/*spiro*-OMeTAD interface truly occurred, which is entirely consistent with the fs-TA results shown in Figure 2. The decrease of the TA signal in the sample of MAPbI3/Y2O3 with a layer of *spiro*-OMeTAD originated from the recombination process between the electrons in MAPbI3 and the holes in *spiro*-OMeTAD, which can be fitted very well with eq. (2). Only one exponential decay process was observed, with a time constant determined to be 0.37±0.07 μs,

process in the Y2O3/ MAPbI3/spiro-OMeTAD could be considered to originate from the

photoexcited hole injection from MAPbI3 to spiro-OMeTAD. This result means that

charge separation occurred at the interfaces between MAPbI3 and spiro-OMeTAD. 25)

which corresponds to the recombination time [ 25]. Figure 5 shows the photoexcited charge carrier (electrons and holes) dynamics in MAPbI3 deposited on a Y2O3 substrate with *spiro*-OMeTAD as a HTM. (2). Only one exponential decay process was observed, with a time constant determined to be 0.37±0.07 µs, which corresponds to the recombination time<sup>25</sup>. Figure 5 shows the photoexcited charge carrier (electrons and holes) dynamics in MAPbI3 deposited on a Y2O<sup>3</sup>

states in MAPbI3 or at the MAPbI3/Y2O3 interface. The relative contribution of the faster decay is less than 10%, which suggests that the

defect or trap state density in MAPbI3 or at the interface is very small, which is consistent with a thermally stimulated current

measurement.39,40 The slower decay process could be attributed to the recombination of free electrons and holes in MAPbI3. This result

suggests that the lifetime of the photoexcited charge carriers was very long, which was confirmed to be as much as the order of microseconds

by the ns-TA measurement results shown in Figure 3. The TA decay was fitted with eq. (2), and two decay processes were found with

lifetimes of 3.7±0.1 µs (70%) and 60±1 µs (30%), respectively. Combining the fs-TA and ns-TA results for MAPbI3/Y2O3, the lifetime of

Figure 3 TA responses of MAPbI3/Y2O3 without spiro-OMeTAD for a time scale of 200 µs measured with a pump light wavelength of

The TA response of MAPbI3/Y2O3 with spiro-OMeTAD was fitted to eq. (2), and only one exponential decay with a time constant of

16±2 ns was found. Compared with the long lifetime of the TA decay in Y2O3/MAPbI3, the fast decay process in

Y2O3/MAPbI3/spiro-OMeTAD can be considered to originate from the photoexcited hole injection from MAPbI3 to spiro-OMeTAD. This

Figure 4 shows the TA response of MAPbI3/Y2O3 with a layer of spiro-OMeTAD measured at a probe wavelength of 1310 nm. This

result clearly indicates that charge separation at the MAPbI3/spiro-OMeTAD interface truly occurred, which is entirely consistent with the

from the recombination process between the electrons in MAPbI3 and the holes in spiro-OMeTAD, which can be fitted very well with eq.

470 nm and a probe light wavelength of 785 nm. The red solid lines represent the fitting results with eq. (2). (reference: 25)

result indicates that charge separation occurred at the interface between MAPbI3 and spiro-OMeTAD.<sup>25</sup>

the photoexcited charge carriers in MAPbI3 can be mostly considered to be as long as the order of microseconds.

6 **Figure 4.** TA responses of MAPbI3/Y2O3 with *spiro*-OMeTAD measured with a pump light wavelength of 470 nm and a probe light wavelength of 1310 nm. The red solid lines represent the fitting results with eq. (2). (reference: 25) Figure 4 TA responses of MAPbI3/Y2O3 with spiro-OMeTAD measured with a pump light wavelength of 470 nm and a probe light wavelength of 1310 nm. The red solid lines represent the fitting results with eq. (2). (reference: 25)

Figure 5 Schematic illustration of photoexcited charge carrier (electrons and holes) dynamics in MAPbI3 deposited on a Y2O3 substrate with spiro-OMeTAD as a HTM. (reference: 25) **Figure 5.** Schematic illustration of photoexcited charge carrier (electrons and holes) dynamics in MAPbI3 deposited on a Y2O3 substrate with *spiro*-OMeTAD as a HTM. (reference: 25)

 The photoexcited electron injection and the recombination dynamics were measured using the TA techniques for MAPbI3 deposited on TiO2 substrates. Figure 6(a) shows the normalized TA responses of MAPbI3/TiO2 for 400 ps measured with different pump light intensities. The decay processes in the normalized TA responses overlapped very well with each other when the pump light intensity was lower. However, when the pump light intensity became larger than 3.75 μJ/cm2 , a faster decay process, resulting from Auger recombination, appeared in the TA response. [25] To determine the electron injection dynamics from MAPbI3 to TiO2, the TA

results of the TA response (pump light intensity: 1.25 µJ/cm<sup>2</sup>

However, when the pump light intensity became larger than 3.75 µJ/cm<sup>2</sup>

measured under a lower pump light intensity (0.9 µJ/cm<sup>2</sup>

7

Figure 6 (a) Dependence of the normalized TA responses of CH3NH3PbClI2/TiO2 on pump light intensity and (b) theoretical fitting

CH3NH3PbClI2/Y2O3 is also shown in (b). The pump light wavelength is 470 nm and the probe light wavelength is 775 nm. (reference: 25)

The photoexcited electron injection and the recombination dynamics were measured using the TA techniques for MAPbI3 deposited on TiO2 substrates. Figure 6(a) shows the normalized TA responses of MAPbI3/TiO2 for 400 ps measured with different pump light intensities. The decay processes in the normalized TA responses overlapped very well with each other when the pump light intensity was lower.

appeared in the TA response.<sup>25</sup> To determine the electron injection dynamics from MAPbI3 to TiO2, the TA response of MAPbI3/TiO<sup>2</sup>

6(b). The TA response of MAPbI3/TiO2 decayed much faster than that of MAPbI3/Y2O3 and can be fitted to a single exponential decay function very well, with a time constant determined to be 1.8±0.1 ns, which is approximately 2 - 3 orders smaller than the photoexcited

) with a biexponential function (eq. (2)). For comparison, the TA response of

, a faster decay process, resulting from Auger recombination,

) was used to make a comparison with that of MAPbI3/Y2O3 as shown in Figure

fs-TA results shown in Figure 2. The decrease of the TA signal in the sample of MAPbI3/Y2O3 with a layer of spiro-OMeTAD originated with spiro-OMeTAD as a HTM. (reference: 25) Optical Absorption, Charge Separation and Recombination Dynamics in Pb and Sn/Pb Cocktail Perovskite... http://dx.doi.org/10.5772/62101 411

16 ns

wavelength of 1310 nm. The red solid lines represent the fitting results with eq. (2). (reference: 25)

E

Figure 4 TA responses of MAPbI3/Y2O3 with spiro-OMeTAD measured with a pump light wavelength of 470 nm and a probe light

Figure 5 Schematic illustration of photoexcited charge carrier (electrons and holes) dynamics in MAPbI3 deposited on a Y2O3 substrate

Spiro OMeTAD

0.37 µs

MAPbI<sup>3</sup>

e-

34 ps (9%) A few µs – 10 µs (91%)

h+

Y2O<sup>3</sup>

which corresponds to the recombination time [

410 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

substrate with spiro-OMeTAD as a HTM.

1.0x10-6

E

with spiro-OMeTAD as a HTM. (reference: 25)

Y2O<sup>3</sup>

results of the TA response (pump light intensity: 1.25 µJ/cm<sup>2</sup>

However, when the pump light intensity became larger than 3.75 µJ/cm<sup>2</sup>

measured under a lower pump light intensity (0.9 µJ/cm<sup>2</sup>


3.75 μJ/cm2

a Y2O3 substrate with *spiro*-OMeTAD as a HTM. (reference: 25)

5.0x10-5

∆A

1.0x10-4

0.0

OMeTAD as a HTM.

carrier (electrons and holes) dynamics in MAPbI3 deposited on a Y2O3 substrate with *spiro*-

Experimental

Fitting

1.5x10


**Figure 4.** TA responses of MAPbI3/Y2O3 with *spiro*-OMeTAD measured with a pump light wavelength of 470 nm and a probe light wavelength of 1310 nm. The red solid lines represent the fitting results with eq. (2). (reference: 25)

MAPbI<sup>3</sup>

**Figure 5.** Schematic illustration of photoexcited charge carrier (electrons and holes) dynamics in MAPbI3 deposited on

The photoexcited electron injection and the recombination dynamics were measured using the TA techniques for MAPbI3 deposited on TiO2 substrates. Figure 6(a) shows the normalized TA responses of MAPbI3/TiO2 for 400 ps measured with different pump light intensities. The decay processes in the normalized TA responses overlapped very well with each other when the pump light intensity was lower. However, when the pump light intensity became larger than

response. [25] To determine the electron injection dynamics from MAPbI3 to TiO2, the TA

e-

34 ps (9%) A few µs – 10 µs (91%)

h+

Time (s)

wavelength of 1310 nm. The red solid lines represent the fitting results with eq. (2). (reference: 25)

16 ns

, a faster decay process, resulting from Auger recombination, appeared in the TA

2.0x10-6

Figure 4 TA responses of MAPbI3/Y2O3 with spiro-OMeTAD measured with a pump light wavelength of 470 nm and a probe light

Figure 5 Schematic illustration of photoexcited charge carrier (electrons and holes) dynamics in MAPbI3 deposited on a Y2O3 substrate

Spiro OMeTAD

0.37 µs

Figure 6 (a) Dependence of the normalized TA responses of CH3NH3PbClI2/TiO2 on pump light intensity and (b) theoretical fitting

CH3NH3PbClI2/Y2O3 is also shown in (b). The pump light wavelength is 470 nm and the probe light wavelength is 775 nm. (reference: 25)

The photoexcited electron injection and the recombination dynamics were measured using the TA techniques for MAPbI3 deposited on TiO2 substrates. Figure 6(a) shows the normalized TA responses of MAPbI3/TiO2 for 400 ps measured with different pump light intensities. The decay processes in the normalized TA responses overlapped very well with each other when the pump light intensity was lower.

appeared in the TA response.<sup>25</sup> To determine the electron injection dynamics from MAPbI3 to TiO2, the TA response of MAPbI3/TiO<sup>2</sup>

6(b). The TA response of MAPbI3/TiO2 decayed much faster than that of MAPbI3/Y2O3 and can be fitted to a single exponential decay function very well, with a time constant determined to be 1.8±0.1 ns, which is approximately 2 - 3 orders smaller than the photoexcited

) with a biexponential function (eq. (2)). For comparison, the TA response of

, a faster decay process, resulting from Auger recombination,

) was used to make a comparison with that of MAPbI3/Y2O3 as shown in Figure

7

25]. Figure 5 shows the photoexcited charge

result indicates that charge separation occurred at the interface between MAPbI3 and spiro-OMeTAD.<sup>25</sup>

states in MAPbI3 or at the MAPbI3/Y2O3 interface. The relative contribution of the faster decay is less than 10%, which suggests that the

defect or trap state density in MAPbI3 or at the interface is very small, which is consistent with a thermally stimulated current

measurement.39,40 The slower decay process could be attributed to the recombination of free electrons and holes in MAPbI3. This result

suggests that the lifetime of the photoexcited charge carriers was very long, which was confirmed to be as much as the order of microseconds

by the ns-TA measurement results shown in Figure 3. The TA decay was fitted with eq. (2), and two decay processes were found with

lifetimes of 3.7±0.1 µs (70%) and 60±1 µs (30%), respectively. Combining the fs-TA and ns-TA results for MAPbI3/Y2O3, the lifetime of

Figure 3 TA responses of MAPbI3/Y2O3 without spiro-OMeTAD for a time scale of 200 µs measured with a pump light wavelength of

The TA response of MAPbI3/Y2O3 with spiro-OMeTAD was fitted to eq. (2), and only one exponential decay with a time constant of

16±2 ns was found. Compared with the long lifetime of the TA decay in Y2O3/MAPbI3, the fast decay process in

Y2O3/MAPbI3/spiro-OMeTAD can be considered to originate from the photoexcited hole injection from MAPbI3 to spiro-OMeTAD. This

Figure 4 shows the TA response of MAPbI3/Y2O3 with a layer of spiro-OMeTAD measured at a probe wavelength of 1310 nm. This

result clearly indicates that charge separation at the MAPbI3/spiro-OMeTAD interface truly occurred, which is entirely consistent with the

from the recombination process between the electrons in MAPbI3 and the holes in spiro-OMeTAD, which can be fitted very well with eq.

470 nm and a probe light wavelength of 785 nm. The red solid lines represent the fitting results with eq. (2). (reference: 25)

the photoexcited charge carriers in MAPbI3 can be mostly considered to be as long as the order of microseconds.

6

Figure 6 (a) Dependence of the normalized TA responses of CH3NH3PbClI2/TiO2 on pump light intensity and (b) theoretical fitting **Figure 6.** (a) Dependence of the normalized TA responses of CH3NH3PbClI2/TiO2 on pump light intensity and (b) theo‐ retical fitting results of the TA response (pump light intensity: 1.25 μJ/cm2 ) with a biexponential function (eq. (2)). For comparison, the TA response of CH3NH3PbClI2/Y2O3 is also shown in (b). The pump light wavelength is 470 nm and the probe light wavelength is 775 nm. (reference: 25)

response of MAPbI3/TiO2 measured under a lower pump light intensity (0.9 μJ/cm2 ) was used to make a comparison with that of MAPbI3/Y2O3 as shown in Figure 6(b). The TA response of MAPbI3/TiO2 decayed much faster than that of MAPbI3/Y2O3 and can be fitted to a single exponential decay function very well, with a time constant determined to be 1.8±0.1 ns, which is approximately 2 - 3 orders smaller than the photoexcited charge carrier lifetimes (∼μs) of MAPbI3 as shown above. Then the electron injection time *t*ET and electron injection rate *k*ET were calculated to be about 1.8 ns and 5.5×108 s-1, respectively. [25] Thus, the electron injection efficiency *η*Einj was calculated to be about 100% for all electrons with lifetimes longer than 100 ns. It is supposed here that the other 9% of charge carriers that recombined with a lifetime of 34 ps were not present in the MAPbI3 prepared on TiO2, because no faster decay component was observed in the TA responses of MAPbI3/TiO2. [25] results of the TA response (pump light intensity: 1.25 µJ/cm<sup>2</sup> ) with a biexponential function (eq. (2)). For comparison, the TA response of CH3NH3PbClI2/Y2O3 is also shown in (b). The pump light wavelength is 470 nm and the probe light wavelength is 775 nm. (reference: 25) The photoexcited electron injection and the recombination dynamics were measured using the TA techniques for MAPbI3 deposited on

Next, the recombination dynamics in MAPbI3/TiO2 without the layer of *spiro*-OMeTAD, i.e., the recombination process between the electrons in TiO2 and the holes in MAPbI3 was measured using ns-TA. Figure 7 shows the TA response of MAPbI3/TiO2 measured with a pump light wavelength of 470 nm and a probe light wavelength of 658 nm. [25] According to the TA signal in nanocrystalline TiO2 films reported by Yoshikawa [33] and our earlier studies, [18-21] it is reasonable to think that the TA response in MAPbI3/TiO2 observed at 658 nm corresponds to the electrons in TiO2 injected from MAPbI3. [25] Then the recombination time between the electrons in TiO2 and the holes in perovskite was determined to be 0.14 μs by fitting Figure 7 to eq. (2). TiO2 substrates. Figure 6(a) shows the normalized TA responses of MAPbI3/TiO2 for 400 ps measured with different pump light intensities. The decay processes in the normalized TA responses overlapped very well with each other when the pump light intensity was lower. However, when the pump light intensity became larger than 3.75 µJ/cm<sup>2</sup> , a faster decay process, resulting from Auger recombination, appeared in the TA response.<sup>25</sup> To determine the electron injection dynamics from MAPbI3 to TiO2, the TA response of MAPbI3/TiO<sup>2</sup> measured under a lower pump light intensity (0.9 µJ/cm<sup>2</sup> ) was used to make a comparison with that of MAPbI3/Y2O3 as shown in Figure

Then, the recombination dynamics in MAPbI3/TiO2 with *spiro*-OMeTAD as a hole transport material, i.e., the recombination between the electrons in TiO2 and the holes in *spiro*-OMe‐ TAD, were measured using ns-TA. Figure 8 shows the TA responses of MAPbI3/TiO2 without and with *spiro*-OMeTAD, which were measured with a pump light wavelength of 470 nm and a probe light wavelength of 1310 nm, which was used to monitor the relaxation dynamics of holes in *spiro*-OMeTAD. From fig. 8 we can see that no TA signal was observed for the 7 6(b). The TA response of MAPbI3/TiO2 decayed much faster than that of MAPbI3/Y2O3 and can be fitted to a single exponential decay function very well, with a time constant determined to be 1.8±0.1 ns, which is approximately 2 - 3 orders smaller than the photoexcited MAPbI3/TiO2.

eq. (2).

25

calculated to be about 1.8 ns and 5.5×10<sup>8</sup>

 Figure 7 TA response of MAPbI3/TiO2 without a HTM measured with a pump light wavelength of 470 nm and a probe light wavelength of 658 nm. The red solid line represents the fitting result with a single exponential decay function with a time constant of 0.14 µs. **Figure 7.** TA response of MAPbI3/TiO2 without a HTM measured with a pump light wavelength of 470 nm and a probe light wavelength of 658 nm. The red solid line represents the fitting result with a single exponential decay function with a time constant of 0.14 μs. (reference: 25)

sample without *spiro*-OMeTAD. However, for the sample with *spiro*-OMeTAD, an absorp‐ tion signal can be clearly observed, which confirms that the TA signal probed at 1310 nm originated from the holes in *spiro*-OMeTAD. The TA decay in Figure 8 can be fitted very well with a single exponential decay function and the time constant was determined to be 60 ±0.5 μs. This is a typical result for cells with 18 nm sized TiO2 nanoparticles. It is important to understand how the carrier dynamics relate to the photovoltaic properties of the cells. Several cells were measured and the energy conversion efficiency was typically 5-7%. These cells are called cell A. (reference: 25) Next, the recombination dynamics in MAPbI3/TiO2 with spiro-OMeTAD as a hole transport material, i.e., the recombination between the electrons in TiO2 and the holes in spiro-OMeTAD, were measured using ns-TA. Figure 8 shows the TA responses of MAPbI3/TiO2 without and with spiro-OMeTAD, which were measured with a pump light wavelength of 470 nm and a probe

 Figure 8 TA responses of MAPbI3/TiO2 without (a) and with (b) spiro-OMeTAD as a HTM measured with a pump light wavelength of **Figure 8.** TA responses of MAPbI3/TiO2 without (a) and with (b) *spiro*-OMeTAD as a HTM measured with a pump light wavelength of 470 nm and a probe light wavelength of 1310 nm. The red solid line in (b) represents the fitting result with a single exponential decay function with a time constant of 60 μs. (reference: 25)

function with a time constant of 60 µs. (reference: 25)

and energy conversion efficiency η of cell A are 11.81 mA/cm<sup>2</sup>

mA/cm<sup>2</sup>

larger than that of cell A, only 70%.

470 nm and a probe light wavelength of 1310 nm. The red solid line in (b) represents the fitting result with a single exponential decay

light wavelength of 1310 nm, which was used to monitor the relaxation dynamics of holes in spiro-OMeTAD. From fig. 8 we

charge carrier lifetimes (~µs) of MAPbI3 as shown above. Then the electron injection time tET and electron injection rate kET were

for all electrons with lifetimes longer than 100 ns. It is supposed here that the other 9% of charge carriers that recombined with a lifetime of

34 ps were not present in the MAPbI3 prepared on TiO2, because no faster decay component was observed in the TA responses of

Next, the recombination dynamics in MAPbI3/TiO2 without the layer of spiro-OMeTAD, i.e., the recombination process

between the electrons in TiO2 and the holes in MAPbI3 was measured using ns-TA. Figure 8 shows the TA response of MAPbI3/TiO2 measured with a pump light wavelength of 470 nm and a probe light wavelength of 658 nm.<sup>25</sup> According to the TA

signal in nanocrystalline TiO2 films reported by Yoshikawa<sup>33</sup> and our earlier studies,18-21\_ENREF\_23 it is reasonable to think that

recombination time between the electrons in TiO2 and the holes in perovskite was determined to be 0.14 µs by fitting Figure 7 to

the TA response in MAPbI3/TiO2 observed at 658 nm corresponds to the electrons in TiO2 injected from MAPbI3.

s-1, respectively.<sup>25</sup> Thus, the electron injection efficiency ηEinj was calculated to be about 100%

<sup>25</sup> Then the

Then another kind of cell with 30 nm sized TiO2 nanoparticles were also studied, which we call cell B. The energy conversion

efficiency of cell B was typically 8-10%. It was found that the recombination dynamics in the Perovskite solar cells has a great

influence on the IPCE, Jsc and the energy conversion efficiency.<sup>25</sup> Two kinds of TiO2/MAPbI3/spiro-OMeTAD solar cells

showing different IPCE spectra and photovoltaic performance were studied, and the relationships between the charge separation

and recombination dynamics and the photovoltaic properties were investigated.<sup>25</sup> Figure 9 shows the IPCE spectra and

current-voltage (I-V) curves of the two kinds of cell. For cells A and B, the TiO2 mesoporous layers were made of TiO<sup>2</sup>

nanoparticles with sizes of 18 nm and 30 nm, respectively. The short circuit current Jsc, open circuit voltage Voc, fill factor FF

(the pump light wavelength used in the TA measurements) is 58%. On the other hand, for cell B, Jsc, Voc, FF and η are 17.60

between the photoexcited charge carrier dynamics and the photovoltaic properties, especially the IPCE, the charge separation and

recombination dynamics of cell B were also characterized<sup>25</sup>. Figure 10 compares the charge carrier dynamics of the two cells A

and B, it can be seen that the recombination time of electrons in TiO2 and holes in the perovskite was almost identical (i.e.,

0.13-0.14 µs), and the electron injection time was a little faster (0.7 ns) in cell B. However, the recombination time of electrons

in TiO2 and holes in spiro-OMeTAD in cell B became as long as 600 µs, which is ten times longer than that in cell A (60 µs). It

is important to see how these changed separation and recombination dynamics influence the IPCE and Jsc of the two kinds of

cell. It is well known that IPCE is proportional to the charge separation efficiency ηCsep and the charge collection efficiency ηCcol,

where ηCsep is the product of the electron injection efficiency ηEinj and the hole injection efficiency ηHinj. ηEinj, ηHinj and ηCcol can

be calculated using the photoexcited carrier lifetimes, electron and hole injection times, and recombination times as shown in

Figures 5 and 10. As discussed in detail in reference 25, the charge separation efficiency of the two kinds of cell were found to be

almost the same, about 90%. However, the charge collection efficiency of cell B was found to be almost 100%, which is much

, 0.78 V, 0.69 and 9.54%, respectively. The IPCE value of cell B at 470 nm is about 85%. To investigate the correlation

, 0.79 V, 0.71 and 6.59%, respectively. The IPCE value at 470 nm

10

 Figure 7 TA response of MAPbI3/TiO2 without a HTM measured with a pump light wavelength of 470 nm and a probe light wavelength of 658 nm. The red solid line represents the fitting result with a single exponential decay function with a time constant of 0.14 µs. Next, the recombination dynamics in MAPbI3/TiO2 with spiro-OMeTAD as a hole transport material, i.e., the recombination between the electrons in TiO2 and the holes in spiro-OMeTAD, were measured using ns-TA. Figure 8 shows the TA responses of MAPbI3/TiO2 without and with spiro-OMeTAD, which were measured with a pump light wavelength of 470 nm and a probe light wavelength of 1310 nm, which was used to monitor the relaxation dynamics of holes in spiro-OMeTAD. From fig. 8 we can see that no TA signal was observed for the sample without spiro-OMeTAD. However, for the sample with spiro-OMeTAD, Then another kind of cell with 30 nm sized TiO2 nanoparticles were also studied, which we call cell B. The energy conversion efficiency of cell B was typically 8-10%. It was found that the recombination dynamics in the Perovskite solar cells has a great influence on the IPCE, Jsc and the energy conversion efficiency. [25] Two kinds of TiO2/MAPbI3/*spiro*-OMeTAD solar cells showing different IPCE spectra and photovoltaic performance were studied, and the relation‐ ships between the charge separation and recombination dynamics and the photovoltaic properties were investigated. [25] Figure 9 shows the IPCE spectra and current-voltage (I-V) curves of the two kinds of cell. For cells A and B, the TiO2 mesoporous layers were made of TiO2 nanoparticles with sizes of 18 nm and 30 nm, respectively. The short circuit current *J*sc, open circuit voltage *V*oc, fill factor *FF* and energy conversion efficiency *η* of cell A are 11.81 mA/ cm2 , 0.79 V, 0.71 and 6.59%, respectively. The IPCE value at 470 nm (the pump light wavelength used in the TA measurements) is 58%. On the other hand, for cell B, *J*sc, *V*oc, *FF* and *η* are 17.60 mA/cm2 , 0.78 V, 0.69 and 9.54%, respectively. The IPCE value of cell B at 470 nm is about 85%. To investigate the correlation between the photoexcited charge carrier dynamics and the photovoltaic properties, especially the IPCE, the charge separation and recombination dynamics of cell B were also characterized [25]. Figure 10 compares the charge carrier dynamics of the two cells A and B, it can be seen that the recombination time of electrons in TiO2 and holes in the perovskite was almost identical (i.e., 0.13-0.14 μs), and the electron injection time was a little faster (0.7 ns) in cell B. However, the recombination time of electrons in TiO2 and holes in *spiro*-OMeTAD in cell B became as long as 600 μs, which is ten times longer than that in cell A (60 μs). It is important to see how these changed separation and recombination dynamics influence the IPCE and *J*sc of the two kinds of cell. It is well known that IPCE is proportional to the charge separation efficiency *η*Csep and the charge collection efficiency *η*Ccol, where *η*Csep is the product of the electron injection efficiency *η*Einj and the hole injection efficiency *η*Hinj. *η*Einj, *η*Hinj and *η*Ccol can be calculated using the photoexcited carrier lifetimes, electron and hole injection times, and recombination times as shown in Figures 5 and 10. As discussed in detail in reference 25, the charge separation efficiency of the two kinds of cell were found to be almost the same, about 90%. However, the charge collection efficiency of cell B was found to be almost 100%, which is much larger than that of cell A, only 70%.

sample without *spiro*-OMeTAD. However, for the sample with *spiro*-OMeTAD, an absorp‐ tion signal can be clearly observed, which confirms that the TA signal probed at 1310 nm originated from the holes in *spiro*-OMeTAD. The TA decay in Figure 8 can be fitted very well with a single exponential decay function and the time constant was determined to be 60 ±0.5 μs. This is a typical result for cells with 18 nm sized TiO2 nanoparticles. It is important to understand how the carrier dynamics relate to the photovoltaic properties of the cells. Several cells were measured and the energy conversion efficiency was typically 5-7%. These cells are

**Figure 7.** TA response of MAPbI3/TiO2 without a HTM measured with a pump light wavelength of 470 nm and a probe light wavelength of 658 nm. The red solid line represents the fitting result with a single exponential decay function

8x10-7 <sup>10</sup>-6 1.2x10-6 1.4x10-6 -2.0x10-4

Time (s)

calculated to be about 1.8 ns and 5.5×10<sup>8</sup>

MAPbI3/TiO2.

eq. (2).

∆A

(reference: 25)

with a time constant of 0.14 μs. (reference: 25)

0.0

2.0x10-4

4.0x10-4

412 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

25

**Figure 8.** TA responses of MAPbI3/TiO2 without (a) and with (b) *spiro*-OMeTAD as a HTM measured with a pump light wavelength of 470 nm and a probe light wavelength of 1310 nm. The red solid line in (b) represents the fitting

10

470 nm and a probe light wavelength of 1310 nm. The red solid line in (b) represents the fitting result with a single exponential decay

8

an absorption signal can be clearly observed, which confirms that the TA signal probed at 1310 nm originated from the holes in

charge carrier lifetimes (~µs) of MAPbI3 as shown above. Then the electron injection time tET and electron injection rate kET were

for all electrons with lifetimes longer than 100 ns. It is supposed here that the other 9% of charge carriers that recombined with a lifetime of

34 ps were not present in the MAPbI3 prepared on TiO2, because no faster decay component was observed in the TA responses of

Next, the recombination dynamics in MAPbI3/TiO2 without the layer of spiro-OMeTAD, i.e., the recombination process

between the electrons in TiO2 and the holes in MAPbI3 was measured using ns-TA. Figure 8 shows the TA response of MAPbI3/TiO2 measured with a pump light wavelength of 470 nm and a probe light wavelength of 658 nm.<sup>25</sup> According to the TA

signal in nanocrystalline TiO2 films reported by Yoshikawa<sup>33</sup> and our earlier studies,18-21\_ENREF\_23 it is reasonable to think that

recombination time between the electrons in TiO2 and the holes in perovskite was determined to be 0.14 µs by fitting Figure 7 to

the TA response in MAPbI3/TiO2 observed at 658 nm corresponds to the electrons in TiO2 injected from MAPbI3.

Experimental Fitting

s-1, respectively.<sup>25</sup> Thus, the electron injection efficiency ηEinj was calculated to be about 100%

<sup>25</sup> Then the

Then another kind of cell with 30 nm sized TiO2 nanoparticles were also studied, which we call cell B. The energy conversion

efficiency of cell B was typically 8-10%. It was found that the recombination dynamics in the Perovskite solar cells has a great

influence on the IPCE, Jsc and the energy conversion efficiency.<sup>25</sup> Two kinds of TiO2/MAPbI3/spiro-OMeTAD solar cells

showing different IPCE spectra and photovoltaic performance were studied, and the relationships between the charge separation

and recombination dynamics and the photovoltaic properties were investigated.<sup>25</sup> Figure 9 shows the IPCE spectra and

current-voltage (I-V) curves of the two kinds of cell. For cells A and B, the TiO2 mesoporous layers were made of TiO<sup>2</sup>

nanoparticles with sizes of 18 nm and 30 nm, respectively. The short circuit current Jsc, open circuit voltage Voc, fill factor FF

(the pump light wavelength used in the TA measurements) is 58%. On the other hand, for cell B, Jsc, Voc, FF and η are 17.60

between the photoexcited charge carrier dynamics and the photovoltaic properties, especially the IPCE, the charge separation and

recombination dynamics of cell B were also characterized<sup>25</sup>. Figure 10 compares the charge carrier dynamics of the two cells A

and B, it can be seen that the recombination time of electrons in TiO2 and holes in the perovskite was almost identical (i.e.,

0.13-0.14 µs), and the electron injection time was a little faster (0.7 ns) in cell B. However, the recombination time of electrons

in TiO2 and holes in spiro-OMeTAD in cell B became as long as 600 µs, which is ten times longer than that in cell A (60 µs). It

is important to see how these changed separation and recombination dynamics influence the IPCE and Jsc of the two kinds of

cell. It is well known that IPCE is proportional to the charge separation efficiency ηCsep and the charge collection efficiency ηCcol,

where ηCsep is the product of the electron injection efficiency ηEinj and the hole injection efficiency ηHinj. ηEinj, ηHinj and ηCcol can

be calculated using the photoexcited carrier lifetimes, electron and hole injection times, and recombination times as shown in

Figures 5 and 10. As discussed in detail in reference 25, the charge separation efficiency of the two kinds of cell were found to be

almost the same, about 90%. However, the charge collection efficiency of cell B was found to be almost 100%, which is much

, 0.78 V, 0.69 and 9.54%, respectively. The IPCE value of cell B at 470 nm is about 85%. To investigate the correlation

, 0.79 V, 0.71 and 6.59%, respectively. The IPCE value at 470 nm

solar cells A (a) and B (b). (reference: 25)

CH3NH3PbClI<sup>2</sup>

h+

e-

60 µs

1.8 ns

0.14 µs

E

TiO<sup>2</sup>

Relationships to the Photovoltaic Properties<sup>31</sup>

called cell A.

mA/cm<sup>2</sup>

larger than that of cell A, only 70%.

function with a time constant of 60 µs. (reference: 25)

result with a single exponential decay function with a time constant of 60 μs. (reference: 25)

and energy conversion efficiency η of cell A are 11.81 mA/cm<sup>2</sup>

 Figure 8 TA responses of MAPbI3/TiO2 without (a) and with (b) spiro-OMeTAD as a HTM measured with a pump light wavelength of **Figure 9.** Incident photon to current conversion efficiency (IPCE) spectra of the solar cells A (red) and B (black) (a) and current voltage (I-V) curves of TiO2/ MAPbI3/*spiro*-OMeTAD solar cells A (red) and B (black) (b). (reference: 25)

Spiro OMeTAD

(a) (b)

16 ns

(I-V) curves of TiO2/ MAPbI3/spiro-OMeTAD solar cells A (red) and B (black) (b). (reference: 25)

E

TiO<sup>2</sup>

Figure 9 Incident photon to current conversion efficiency (IPCE) spectra of the solar cells A (red) and B (black) (a) and current voltage

Figure 10 Schematic illustration of photoexcited electron injection and recombination dynamics in two TiO2/MAPbI3 /spiro-OMeTAD

CH3NH3PbClI<sup>2</sup>

h+

Spiro OMeTAD

16 ns

e-

600 µs

0.7 ns

0.13 µs

5. Dynamics of Photoexcited Carrier Recombination and Charge Transfer in Sn/Pb Cocktail Perovskite Solar Cells and their

To find the mechanism responsible for the low PCE, especially the low Voc and FF in Sn/Pb cocktail perovskite solar cells,

which has been reported<sup>23</sup>, both the fs-TA and ns-TA techniques were used to investigate the charge separation and

recombination dynamics in these solar cells systematically.<sup>31</sup> Sn/Pb cocktail perovskites deposited on either Al2O3 or TiO2 films,

with and without P3HT, were measured by the TA techniques using probe beams with different wavelengths. Similar to the

Pb-based perovskite solar cells discussed above, firstly, photoexcited carrier lifetimes, hole injection to P3HT and recombination

at the Sn/Pb cocktail perovskite /P3HT interface were clarified as shown in Figure 11.<sup>31</sup> The photoexcited carrier lifetimes are

4.2±0.4 ps (A1/(A1+A2+y0): 55%), 650±57 ps (A2/(A1+A2+y0): 25%), and much longer than 3 ns (y0/(A1+A2+y0): 20%),

respectively. Therefore, the recombination dynamics and lifetimes of photoexcited carriers in Sn/Pb cocktail perovskite are much

faster and shorter than those in Pb-based perovskite, for which case they are much greater than 100 ns for over 90% of the

11

photoexcited carriers as mentioned above. It is considered that the faster decay processes are due to some nonradiative

350 450 550 650 750

cell A

Wavelength / nm

cell B

0.0 0.2 0.4 0.6 0.8 1.0

IPCE

solar cells A (a) and B (b). (reference: 25)

(I-V) curves of TiO2/ MAPbI3/spiro-OMeTAD solar cells A (red) and B (black) (b). (reference: 25)

0.0 4.0 8.0 12.0 16.0 20.0

cell B cell A

Current Density / mAcm-2

(a)

Figure 9 Incident photon to current conversion efficiency (IPCE) spectra of the solar cells A (red) and B (black) (a) and current voltage

0.0 0.2 0.4 0.6 0.8

(b)

Volatage / V

Figure 10 Schematic illustration of photoexcited electron injection and recombination dynamics in two TiO2/MAPbI3 /spiro-OMeTAD **Figure 10.** Schematic illustration of photoexcited electron injection and recombination dynamics in two TiO2/MAPbI3 / *spiro*-OMeTAD solar cells A (a) and B (b). (reference: 25)

### 5. Dynamics of Photoexcited Carrier Recombination and Charge Transfer in Sn/Pb Cocktail Perovskite Solar Cells and their Relationships to the Photovoltaic Properties<sup>31</sup> **5. Dynamics of Photoexcited Carrier Recombination and Charge Transfer in Sn/Pb Cocktail Perovskite Solar Cells and their Relationships to the Photovoltaic Properties [31]**

11 To find the mechanism responsible for the low PCE, especially the low Voc and FF in Sn/Pb cocktail perovskite solar cells, which has been reported<sup>23</sup>, both the fs-TA and ns-TA techniques were used to investigate the charge separation and recombination dynamics in these solar cells systematically.<sup>31</sup> Sn/Pb cocktail perovskites deposited on either Al2O3 or TiO2 films, with and without P3HT, were measured by the TA techniques using probe beams with different wavelengths. Similar to the Pb-based perovskite solar cells discussed above, firstly, photoexcited carrier lifetimes, hole injection to P3HT and recombination at the Sn/Pb cocktail perovskite /P3HT interface were clarified as shown in Figure 11.<sup>31</sup> The photoexcited carrier lifetimes are 4.2±0.4 ps (A1/(A1+A2+y0): 55%), 650±57 ps (A2/(A1+A2+y0): 25%), and much longer than 3 ns (y0/(A1+A2+y0): 20%), respectively. Therefore, the recombination dynamics and lifetimes of photoexcited carriers in Sn/Pb cocktail perovskite are much faster and shorter than those in Pb-based perovskite, for which case they are much greater than 100 ns for over 90% of the photoexcited carriers as mentioned above. It is considered that the faster decay processes are due to some nonradiative To find the mechanism responsible for the low PCE, especially the low *V*oc and *FF* in Sn/Pb cocktail perovskite solar cells, which has been reported [23], both the fs-TA and ns-TA techniques were used to investigate the charge separation and recombination dynamics in these solar cells systematically. [31] Sn/Pb cocktail perovskites deposited on either Al2O3 or TiO2 films, with and without P3HT, were measured by the TA techniques using probe beams with different wavelengths. Similar to the Pb-based perovskite solar cells discussed above, firstly, photoexcited carrier lifetimes, hole injection to P3HT and recombination at the Sn/Pb cocktail perovskite /P3HT interface were clarified as shown in Figure 11. [31] The photoexcited carrier lifetimes are 4.2±0.4 ps (*A*1/(*A*1+*A*2+*y*0): 55%), 650±57 ps (*A*2/(*A*1+*A*2+*y*0): 25%), and much longer than 3 ns (*y*0/(*A*1+*A*2+*y*0): 20%), respectively. Therefore, the recombination dynamics and lifetimes of photoexcited carriers in Sn/Pb cocktail perovskite are much faster and shorter than those in Pb-based perovskite, for which case they are much greater than 100 ns for over 90% of the photoexcited carriers as mentioned above. It is considered that the faster decay processes are due to some nonradiative recombination of electrons and holes through defects or trap states in the Sn/Pb cocktail perovskite, which is consistent with the measured larger Urbach energy in the Sn/Pb cocktail perovskite (i.e., 34 meV). It is larger than that in Pb-based perovskite (i.e., 22 mev). On the other hand, the charge separation at the interface between Sn/ Pb cocktail perovskite and P3HT occurred in approximately 1 ps and interfacial charge recombination between the Sn/Pb cocktail perovskite (on an Al2O3 substrate) and P3HT occurred in approximately 16 ps. [31]

Secondly, the photoexcited charge separation and recombination dynamics of Sn/Pb cocktail perovskite deposited on a TiO2 substrate with P3HT as a hole transport material were evalu‐ ated, which is shown in Figure 12. The charge separation at the perovskite/TiO2 interface occurs in as fast as 1 ps. It is worth noting that the recombination time at the Sn/Pb cocktail perovskite/

P3HT occurred in approximately 16 ps.<sup>31</sup> Optical Absorption, Charge Separation and Recombination Dynamics in Pb and Sn/Pb Cocktail Perovskite... http://dx.doi.org/10.5772/62101 415

recombination of electrons and holes through defects or trap states in the Sn/Pb cocktail perovskite, which is consistent with the

measured larger Urbach energy in the Sn/Pb cocktail perovskite (i.e., 34 meV). It is larger than that in Pb-based perovskite (i.e.,

22 mev). On the other hand, the charge separation at the interface between Sn/Pb cocktail perovskite and P3HT occurred in

approximately 1 ps and interfacial charge recombination between the Sn/Pb cocktail perovskite (on an Al2O3 substrate) and

 Figure 11 Schematic illustration of photoexcited charge carrier (electrons and holes) dynamics in Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite deposited on an Al2O3 substrate without (a) and with (b) P3HT as a hole transport material. (reference: **Figure 11.** Schematic illustration of photoexcited charge carrier (electrons and holes) dynamics in Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite deposited on an Al2O3 substrate without (a) and with (b) P3HT as a hole transport material. (reference: 31)

TiO2 interface is 880 μs, which is about two to three orders of magnitude slower compared to that occurring at the MAPbI3/TiO2 interface as shown in Figure 10. On the other hand, the recombination time at the TiO2/P3HT interface is 190 μs, which is much faster compared to that without P3HT. This indicates that the recombination of electrons in the TiO2 becomes faster when P3HT is used, which originates from the recombination at the TiO2/P3HT interface. This recombination is one of the main reasons for the lower *V*oc and *FF* of the Sn/Pb cocktail solar cells. [23] Pinhole-free Sn/Pb cocktail perovskite needs to be prepared in order to reduce direct interaction between the TiO2 and P3HT, thus suppressing potential recombination. [31] 31) Secondly, the photoexcited charge separation and recombination dynamics of Sn/Pb cocktail perovskite deposited on a TiO<sup>2</sup> substrate with P3HT as a hole transport material were evaluated, which is shown in Figure 12. The charge separation at the perovskite/TiO2 interface occurs in as fast as 1 ps. It is worth noting that the recombination time at the Sn/Pb cocktail perovskite/TiO2 interface is 880 µs, which is about two to three orders of magnitude slower compared to that occurring at the

11

Figure 9 Incident photon to current conversion efficiency (IPCE) spectra of the solar cells A (red) and B (black) (a) and current voltage

0.0 0.2 0.4 0.6 0.8

(b)

Volatage / V

Figure 10 Schematic illustration of photoexcited electron injection and recombination dynamics in two TiO2/MAPbI3 /spiro-OMeTAD

CH3NH3PbClI<sup>2</sup>

h+

Spiro OMeTAD

16 ns

e-

600 µs

0.7 ns

0.13 µs

5. Dynamics of Photoexcited Carrier Recombination and Charge Transfer in Sn/Pb Cocktail Perovskite Solar Cells and their

To find the mechanism responsible for the low PCE, especially the low Voc and FF in Sn/Pb cocktail perovskite solar cells, which has been reported<sup>23</sup>, both the fs-TA and ns-TA techniques were used to investigate the charge separation and recombination dynamics in these solar cells systematically.<sup>31</sup> Sn/Pb cocktail perovskites deposited on either Al2O3 or TiO2 films, with and without P3HT, were measured by the TA techniques using probe beams with different wavelengths. Similar to the Pb-based perovskite solar cells discussed above, firstly, photoexcited carrier lifetimes, hole injection to P3HT and recombination at the Sn/Pb cocktail perovskite /P3HT interface were clarified as shown in Figure 11.<sup>31</sup> The photoexcited carrier lifetimes are 4.2±0.4 ps (A1/(A1+A2+y0): 55%), 650±57 ps (A2/(A1+A2+y0): 25%), and much longer than 3 ns (y0/(A1+A2+y0): 20%), respectively. Therefore, the recombination dynamics and lifetimes of photoexcited carriers in Sn/Pb cocktail perovskite are much faster and shorter than those in Pb-based perovskite, for which case they are much greater than 100 ns for over 90% of the photoexcited carriers as mentioned above. It is considered that the faster decay processes are due to some nonradiative

(I-V) curves of TiO2/ MAPbI3/spiro-OMeTAD solar cells A (red) and B (black) (b). (reference: 25)

E

TiO<sup>2</sup>

0.0 4.0 8.0 12.0 16.0 20.0

cell B cell A

Current Density / mAcm-2

(a)

solar cells A (a) and B (b). (reference: 25)

CH3NH3PbClI<sup>2</sup>

h+

Spiro OMeTAD

(a) (b)

**Figure 10.** Schematic illustration of photoexcited electron injection and recombination dynamics in two TiO2/MAPbI3 /

**5. Dynamics of Photoexcited Carrier Recombination and Charge Transfer in Sn/Pb Cocktail Perovskite Solar Cells and their Relationships to the**

To find the mechanism responsible for the low PCE, especially the low *V*oc and *FF* in Sn/Pb cocktail perovskite solar cells, which has been reported [23], both the fs-TA and ns-TA techniques were used to investigate the charge separation and recombination dynamics in these solar cells systematically. [31] Sn/Pb cocktail perovskites deposited on either Al2O3 or TiO2 films, with and without P3HT, were measured by the TA techniques using probe beams with different wavelengths. Similar to the Pb-based perovskite solar cells discussed above, firstly, photoexcited carrier lifetimes, hole injection to P3HT and recombination at the Sn/Pb cocktail perovskite /P3HT interface were clarified as shown in Figure 11. [31] The photoexcited carrier lifetimes are 4.2±0.4 ps (*A*1/(*A*1+*A*2+*y*0): 55%), 650±57 ps (*A*2/(*A*1+*A*2+*y*0): 25%), and much longer than 3 ns (*y*0/(*A*1+*A*2+*y*0): 20%), respectively. Therefore, the recombination dynamics and lifetimes of photoexcited carriers in Sn/Pb cocktail perovskite are much faster and shorter than those in Pb-based perovskite, for which case they are much greater than 100 ns for over 90% of the photoexcited carriers as mentioned above. It is considered that the faster decay processes are due to some nonradiative recombination of electrons and holes through defects or trap states in the Sn/Pb cocktail perovskite, which is consistent with the measured larger Urbach energy in the Sn/Pb cocktail perovskite (i.e., 34 meV). It is larger than that in Pb-based perovskite (i.e., 22 mev). On the other hand, the charge separation at the interface between Sn/ Pb cocktail perovskite and P3HT occurred in approximately 1 ps and interfacial charge recombination between the Sn/Pb cocktail perovskite (on an Al2O3 substrate) and P3HT

Secondly, the photoexcited charge separation and recombination dynamics of Sn/Pb cocktail perovskite deposited on a TiO2 substrate with P3HT as a hole transport material were evalu‐ ated, which is shown in Figure 12. The charge separation at the perovskite/TiO2 interface occurs in as fast as 1 ps. It is worth noting that the recombination time at the Sn/Pb cocktail perovskite/

16 ns

e-

60 µs

1.8 ns

414 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

0.14 µs

0.0 0.2 0.4 0.6 0.8 1.0

E

TiO<sup>2</sup>

*spiro*-OMeTAD solar cells A (a) and B (b). (reference: 25)

**Photovoltaic Properties [31]**

occurred in approximately 16 ps. [31]

IPCE

350 450 550 650 750

cell A

Wavelength / nm

cell B

Relationships to the Photovoltaic Properties<sup>31</sup>

 Figure 12 Schematic illustration of photoexcited charge separation and recombination dynamics in Sn/Pb cocktail MASn0.5Pb0.5I<sup>3</sup> **Figure 12.** Schematic illustration of photoexcited charge separation and recombination dynamics in Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite solar cells with TiO2 as an electron transport material (ETM) and P3HT as a HTM. (refer‐ ence: 31)

perovskite solar cells with TiO2 as an electron transport material (ETM) and P3HT as a HTM. (reference: 31)

12

MAPbI3/TiO2 interface as shown in Figure 10. On the other hand, the recombination time at the TiO2/P3HT interface is 190 µs,

It is important to understand how the optical absorption, charge separation and recombination dynamics relate to the incident

photon to current conversion efficiency (IPCE) spectrum and the current-voltage (I-V) characteristics of the solar cell. Figure 13

shows a typical IPCE spectrum of a Sn/Pb cocktail perovskite solar cell. Based on the photoexcited carrier relaxation dynamics

of Sn/Pb perovskite on an Al2O3 substrate, and the charge transfer dynamics at TiO2/perovskite/P3HT interfaces, which are

summarized in Table 1, it is possible to calculate the charge separation efficiency.<sup>31</sup> First, it is determined that the hole injection

efficiency ηHinj is almost 100% and the electron injection efficiency ηEinj is about 86%. Thus, the charge separation efficiency

ηCsep (ηCsep =ηHinjηEinj) is 86%. The 14% loss in ηCsep results from the fast recombination component, with a lifetime of 4 ps, of

photoexcited carriers in the Sn/Pb cocktail perovskite, which may originate from the nonradiative recombination owing to the

defects in the sample. Therefore, in order to achieve 100% charge separation efficiency, the crystalline quality of the Sn/Pb

cocktail perovskite has to be improved. The optical absorption is estimated to be about 95% in this case, due to the fact that the

perovskite material is much more absorbing over a broader range up to 1000 nm, which could result in complete absorption (such

as at 470 nm) in thin films such as 500 nm.<sup>23</sup> As shown in Figure 13, the IPCE value at 470 nm, corresponding to the excitation

wavelength used for the TA measurements, is 68%. So the charge collection efficiency ηCcol is estimated to be 83%.<sup>31</sup> The 17%

loss in ηCcol mainly originates from the charge recombination occurring at the TiO2/P3HT interface. In order to improve IPCE

and Jsc for the solar cell, direct contact between P3HT and TiO2 should be avoided or at least suppressed by preparing a

pinhole-free Sn/Pb cocktail perovskite layer on TiO2 or by inserting a barrier/blocking layer at the interface.<sup>31</sup>

13

It is important to understand how the optical absorption, charge separation and recombination dynamics relate to the incident photon to current conversion efficiency (IPCE) spectrum and the current-voltage (I-V) characteristics of the solar cell. Figure 13 shows a typical IPCE spectrum of a Sn/Pb cocktail perovskite solar cell. Based on the photoexcited carrier relaxation dynamics of Sn/Pb perovskite on an Al2O3 substrate, and the charge transfer dynamics at TiO2/perovskite/P3HT interfaces, which are summarized in Table 1, it is possible to calculate the charge separation efficiency. [31] First, it is determined that the hole injection efficiency *η*Hinj is almost 100% and the electron injection efficiency *η*Einj is about 86%. Thus, the charge separation efficiency *η*Csep (*η*Csep =*η*Hinj*η*Einj) is 86%. The 14% loss in *η*Csep results from the fast recombination component, with a lifetime of 4 ps, of photoexcited carriers in the Sn/Pb cocktail perovskite, which may originate from the nonradiative recombination owing to the defects in the sample. Therefore, in order to achieve 100% charge separation efficiency, the crystalline quality of the Sn/Pb cocktail perovskite has to be improved. The optical absorption is estimated to be about 95% in this case, due to the fact that the perovskite material is much more absorbing over a broader range up to 1000 nm, which could result in complete absorption (such as at 470 nm) in thin films such as 500 nm. [23] As shown in Figure 13, the IPCE value at 470 nm, corresponding to the excitation wavelength used for the TA measurements, is 68%. So the charge collection efficiency *η*Ccol is estimated to be 83%. [31] The 17% loss in *η*Ccol mainly originates from the charge recombination occurring at the TiO2/P3HT interface. In order to improve *IPCE* and *J*sc for the solar cell, direct contact between P3HT and TiO2 should be avoided or at least suppressed by preparing a pinhole-free Sn/Pb cocktail perovskite layer on TiO2 or by inserting a barrier/blocking layer at the interface. [31]

Figure 13 A typical incident photon to current conversion efficiency (IPCE) spectrum (a) and current-voltage (I-V) curve (b) of a Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite solar cell.(reference: 31) **Figure 13.** A typical incident photon to current conversion efficiency (IPCE) spectrum (a) and current-voltage (I-V) curve (b) of a Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite solar cell.(reference: 31)

Table 1. Photoexcited carrier lifetimes, hole injection and electron injection dynamics as well as charge recombination at each interface in As shown in Figure 13(b), *J*sc, *V*oc, *FF* and PCE of the Sn/Pb cocktail perovskite solar cells are 22.61 mA/cm2 , 0.21 V, 0.37 and 1.77%, respectively. It is obvious that the low PCE is mainly due to the low *V*oc and FF. *J*sc could be improved to as high as 30 mA/cm2 if the charge separation efficiency can be increased to 100% by reducing the fast nonradiative recombination rate and if the charge collection efficiency can be increased to more than 95% through suppression of

> Al<sup>2</sup> O3 /

MASn0.5Pb0.5I

4 ps (55%)

650 ps (25%)

>>3 ns (20%

3

Al<sup>2</sup> O3 /

MASn0.5Pb0.5I

P3HT

Interfacial recombination N/A 16 ps 880 µs 190 µs

and 1.77%, respectively. It is obvious that the low PCE is mainly due to the low Voc and FF. Jsc could be improved to as high as

if the charge collection efficiency can be increased to more than 95% through suppression of the recombination occurring at the

TiO2/P3HT interface. For Voc, the bandgap-voltage offset EG – qVoc is calculated to be as large as 1 eV, where EG is the bandgap

of 1.21 eV and q is the elementary charge. This bandgap-voltage offset is quite similar to that of amorphous silicon, and is

thought to be closely related to the larger Urbach energy as shown in Figure 14. The recombination resistance from the I-V curve

due to the faster recombination of electrons and holes at the TiO2/P3HT interface. Thus, the lower Voc in the Sn/Pb cocktail

perovskite solar cells is due to two reasons: (1) the larger Urbach energy; (2) the smaller recombination resistance. The lower FF

is also due to reason (2). Therefore, these findings imply that the photovoltaic performance of the Sn/Pb cocktail perovskite solar

if the charge separation efficiency can be increased to 100% by reducing the fast nonradiative recombination rate and

, which could also result in a lower Voc. The smaller recombination resistance estimated is mostly

As shown in Figure 13(b), Jsc, Voc, FF and PCE of the Sn/Pb cocktail perovskite solar cells are 22.61 mA/cm<sup>2</sup>

3 / TiO<sup>2</sup> /

MASn0.5Pb0.5I

Hole injection Electron injection Electron/hole injections

1 ps 1 ps 1 ps/1 ps

N/A N/A N/A

3

TiO<sup>2</sup> /

MASn0.5Pb0.5I

P3HT

3 /

, 0.21 V, 0.37

MASn0.5Pb0.5I3 perovskite solar cells.<sup>31</sup>

Charge separation N/A

Samples

Bulk recombination

(photoexcited carrier

lifetimes)

30 mA/cm<sup>2</sup>

was estimated to be 27 Ω·cm<sup>2</sup>

14

cells can be further improved by decreasing the Urbach energy, i.e., reducing the defects in the perovskite, and by reducing the

Optical Absorption, Charge Separation and Recombination Dynamics in Pb and Sn/Pb Cocktail Perovskite... http://dx.doi.org/10.5772/62101 417


It is important to understand how the optical absorption, charge separation and recombination dynamics relate to the incident photon to current conversion efficiency (IPCE) spectrum and the current-voltage (I-V) characteristics of the solar cell. Figure 13 shows a typical IPCE spectrum of a Sn/Pb cocktail perovskite solar cell. Based on the photoexcited carrier relaxation dynamics of Sn/Pb perovskite on an Al2O3 substrate, and the charge transfer dynamics at TiO2/perovskite/P3HT interfaces, which are summarized in Table 1, it is possible to calculate the charge separation efficiency. [31] First, it is determined that the hole injection efficiency *η*Hinj is almost 100% and the electron injection efficiency *η*Einj is about 86%. Thus, the charge separation efficiency *η*Csep (*η*Csep =*η*Hinj*η*Einj) is 86%. The 14% loss in *η*Csep results from the fast recombination component, with a lifetime of 4 ps, of photoexcited carriers in the Sn/Pb cocktail perovskite, which may originate from the nonradiative recombination owing to the defects in the sample. Therefore, in order to achieve 100% charge separation efficiency, the crystalline quality of the Sn/Pb cocktail perovskite has to be improved. The optical absorption is estimated to be about 95% in this case, due to the fact that the perovskite material is much more absorbing over a broader range up to 1000 nm, which could result in complete absorption (such as at 470 nm) in thin films such as 500 nm. [23] As shown in Figure 13, the IPCE value at 470 nm, corresponding to the excitation wavelength used for the TA measurements, is 68%. So the charge collection efficiency *η*Ccol is estimated to be 83%. [31] The 17% loss in *η*Ccol mainly originates from the charge recombination occurring at the TiO2/P3HT interface. In order to improve *IPCE* and *J*sc for the solar cell, direct contact between P3HT and TiO2 should be avoided or at least suppressed by preparing a pinhole-free Sn/Pb cocktail perovskite layer on

14

cells can be further improved by decreasing the Urbach energy, i.e., reducing the defects in the perovskite, and by reducing the

TiO2 or by inserting a barrier/blocking layer at the interface. [31]

416 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

cocktail MASn0.5Pb0.5I3 perovskite solar cell.(reference: 31)

curve (b) of a Sn/Pb cocktail MASn0.5Pb0.5I3 perovskite solar cell.(reference: 31)

Al<sup>2</sup> O3 /

**Figure 13.** A typical incident photon to current conversion efficiency (IPCE) spectrum (a) and current-voltage (I-V)

As shown in Figure 13(b), *J*sc, *V*oc, *FF* and PCE of the Sn/Pb cocktail perovskite solar cells are

due to the low *V*oc and FF. *J*sc could be improved to as high as 30 mA/cm2 if the charge separation efficiency can be increased to 100% by reducing the fast nonradiative recombination rate and if the charge collection efficiency can be increased to more than 95% through suppression of

, 0.21 V, 0.37 and 1.77%, respectively. It is obvious that the low PCE is mainly

MASn0.5Pb0.5I

4 ps (55%)

650 ps (25%)

>>3 ns (20%

3

Al<sup>2</sup> O3 /

MASn0.5Pb0.5I

P3HT

Interfacial recombination N/A 16 ps 880 µs 190 µs

and 1.77%, respectively. It is obvious that the low PCE is mainly due to the low Voc and FF. Jsc could be improved to as high as

if the charge collection efficiency can be increased to more than 95% through suppression of the recombination occurring at the

TiO2/P3HT interface. For Voc, the bandgap-voltage offset EG – qVoc is calculated to be as large as 1 eV, where EG is the bandgap

of 1.21 eV and q is the elementary charge. This bandgap-voltage offset is quite similar to that of amorphous silicon, and is

thought to be closely related to the larger Urbach energy as shown in Figure 14. The recombination resistance from the I-V curve

due to the faster recombination of electrons and holes at the TiO2/P3HT interface. Thus, the lower Voc in the Sn/Pb cocktail

perovskite solar cells is due to two reasons: (1) the larger Urbach energy; (2) the smaller recombination resistance. The lower FF

is also due to reason (2). Therefore, these findings imply that the photovoltaic performance of the Sn/Pb cocktail perovskite solar

if the charge separation efficiency can be increased to 100% by reducing the fast nonradiative recombination rate and

, which could also result in a lower Voc. The smaller recombination resistance estimated is mostly

As shown in Figure 13(b), Jsc, Voc, FF and PCE of the Sn/Pb cocktail perovskite solar cells are 22.61 mA/cm<sup>2</sup>

3 / TiO<sup>2</sup> /

MASn0.5Pb0.5I

Hole injection Electron injection Electron/hole injections

1 ps 1 ps 1 ps/1 ps

N/A N/A N/A

3

TiO<sup>2</sup> /

MASn0.5Pb0.5I

P3HT

3 /

6. Conclusions

, 0.21 V, 0.37

MASn0.5Pb0.5I3 perovskite solar cells.<sup>31</sup>

Charge separation N/A

Samples

22.61 mA/cm2

Bulk recombination

(photoexcited carrier

lifetimes)

30 mA/cm<sup>2</sup>

was estimated to be 27 Ω·cm<sup>2</sup>

**Table 1.** Photoexcited carrier lifetimes, hole injection and electron injection dynamics as well as charge recombination at each interface in MASn0.5Pb0.5I3 perovskite solar cells. [31]

the recombination occurring at the TiO2/P3HT interface. For *V*oc, the bandgap-voltage offset *E*G – q*V*oc is calculated to be as large as 1 eV, where *E*<sup>G</sup> is the bandgap of 1.21 eV and q is the elementary charge. This bandgap-voltage offset is quite similar to that of amorphous silicon, and is thought to be closely related to the larger Urbach energy as shown in Figure 14. The recombination resistance from the I-V curve was estimated to be 27 Ω cm<sup>2</sup> , which could also result in a lower *V*oc. The smaller recombination resistance estimated is mostly due to the faster recombination of electrons and holes at the TiO2/P3HT interface. Thus, the lower *V*oc in the Sn/ Pb cocktail perovskite solar cells is due to two reasons: (1) the larger Urbach energy; (2) the smaller recombination resistance. The lower *FF* is also due to reason (2). Therefore, these findings imply that the photovoltaic performance of the Sn/Pb cocktail perovskite solar cells can be further improved by decreasing the Urbach energy, i.e., reducing the defects in the perovskite, and by reducing the recombination occurring at the TiO2/P3HT interface through appropriate interface engineering such as passivation or inserting a barrier/blocking layer. [31] recombination occurring at the TiO2/P3HT interface through appropriate interface engineering such as passivation or inserting a barrier/blocking layer.<sup>31</sup>

Table 1. Photoexcited carrier lifetimes, hole injection and electron injection dynamics as well as charge recombination at each interface in Figure 14 The bandgap-voltage offset (EG – qVoc) versus Urbach energy for typical photovoltaic absorber materials (from reference 18) and **Figure 14.** The bandgap-voltage offset (*E*G – q*V*oc) versus Urbach energy for typical photovoltaic absorber materials (from reference 18) and for the Pb and Sn/Pb cocktail perovskites used in our study at room temperature.

for the Pb and Sn/Pb cocktail perovskites used in our study at room temperature.

and Sn/Pb cocktail perovskite solar cells can be improved.

hybrid solar cells is crucial for achieving a high efficiency.

15

separation and charge recombination dynamics were explored. We find that both the electron injection into TiO2 and hole

In summary, by conducting a systematic study on the optical absorption, photoexcited charge carrier lifetimes, and charge

separation and charge recombination dynamics, we have explored the ways through which the photovoltaic performance of Pb

 For MAPbI3 solar cells, we find that the great differences in the IPCE and Jsc of the two kinds of cell result from the great difference in the charge recombination between the electrons in TiO2 and holes in spiro-OMeTAD. These results indicate that the key for improving the IPCE and Jsc of the perovskite-based solar cells is charge collection efficiency, instead of charge separation efficiency. Thus suppressing the recombination through proper interfacial engineering for perovskite-based solid

For the Sn/Pb cocktail perovskite solar cells, it is determined that the bandgap is 1.21 eV and light harvesting can be

extended to a wider wavelength over 1000 nm. The Urbach energy is calculated to be 34 meV, which is more than twice that of

the MAPbI3 perovskite. Three recombination processes for the photoexcited carriers were found with lifetimes being 4 ps (55%), 650 ps (25%) and one much larger than 3 ns (20%). The larger Urbach energy and the faster recombination suggest that there are

defects in the prepared Sn/Pb cocktail perovskite. These results are significantly different from those of the MAPbI3 perovskite, in which case the photocarrier lifetimes are larger than 100 ns and almost no defects were observed. Moreover, the charge

## **6. Conclusions**

In summary, by conducting a systematic study on the optical absorption, photoexcited charge carrier lifetimes, and charge separation and charge recombination dynamics, we have explored the ways through which the photovoltaic performance of Pb and Sn/Pb cocktail perovskite solar cells can be improved.

For MAPbI3 solar cells, we find that the great differences in the IPCE and *J*sc of the two kinds of cell result from the great difference in the charge recombination between the electrons in TiO2 and holes in *spiro*-OMeTAD. These results indicate that the key for improving the IPCE and *J*sc of the perovskite-based solar cells is charge collection efficiency, instead of charge separation efficiency. Thus suppressing the recombination through proper interfacial engi‐ neering for perovskite-based solid hybrid solar cells is crucial for achieving a high efficiency.

For the Sn/Pb cocktail perovskite solar cells, it is determined that the bandgap is 1.21 eV and light harvesting can be extended to a wider wavelength over 1000 nm. The Urbach energy is calculated to be 34 meV, which is more than twice that of the MAPbI3 perovskite. Three recombination processes for the photoexcited carriers were found with lifetimes being 4 ps (55%), 650 ps (25%) and one much larger than 3 ns (20%). The larger Urbach energy and the faster recombination suggest that there are defects in the prepared Sn/Pb cocktail perovskite. These results are significantly different from those of the MAPbI3 perovskite, in which case the photocarrier lifetimes are larger than 100 ns and almost no defects were observed. Moreover, the charge separation and charge recombination dynamics were explored. We find that both the electron injection into TiO2 and hole transfer to the P3HT layer occurred on a timescale of 1 ps. It was surprising to find that the charge recombination lifetime at the Sn/Pb cocktail perovskite/TiO2 interface was as long as 880 μs, which is 2-3 orders of magnitude greater than that occurring at the MAPbI3/TiO2 interface. However, the charge recombination lifetime became shorter, i.e., 190 μs, when P3HT was employed as a HTM. This implies that the charges can be more effectively collected without the HTM. On the basis of the above results, we find that the loss in the charge separation efficiency originates from fast photoexcited carrier recombination with a lifetime of 4 ps, and the loss in charge collection efficiency was due to the charge recombination occurring at the TiO2/P3HT interface, thus leading to lower IPCE values. Also, the low Voc and FF were found to result from the large Urbach energy and the recombination occurring at the TiO2/P3HT interface. These findings indicate that the photo‐ voltaic performance of Sn/Pb cocktail perovskite solar cells can be further improved by reducing the defects in the material and the recombination occurring at the TiO2/HTM interface through proper interfacial engineering.

## **Author details**

Shen Qing1\*, Ogomi Yuhei1 , Toyoda Taro1 , Yoshino Kenji3 and Hayase Shuzi2\*

\*Address all correspondence to: shen@pc.uec.ac.jp; hayase @life.kyutech.ac.jp

1 Department of Engineering Science, Faculty of Informatics and Engineering, The Universi‐ ty of Electro-Communications, Chofu, Tokyo, Japan

2 Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology, Wakamatsu, Kitakyushu, Japan

3 Department of Electrical and Electronic Engineering, Miyazaki University, Kibanadainishi, Miyazaki, Japan

This work was supported by the CREST program, Japan Science and Technology Agency (JST).

## **References**

**6. Conclusions**

solar cells can be improved.

418 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

through proper interfacial engineering.

, Toyoda Taro1

\*Address all correspondence to: shen@pc.uec.ac.jp; hayase @life.kyutech.ac.jp

, Yoshino Kenji3

and Hayase Shuzi2\*

**Author details**

Shen Qing1\*, Ogomi Yuhei1

In summary, by conducting a systematic study on the optical absorption, photoexcited charge carrier lifetimes, and charge separation and charge recombination dynamics, we have explored the ways through which the photovoltaic performance of Pb and Sn/Pb cocktail perovskite

For MAPbI3 solar cells, we find that the great differences in the IPCE and *J*sc of the two kinds of cell result from the great difference in the charge recombination between the electrons in TiO2 and holes in *spiro*-OMeTAD. These results indicate that the key for improving the IPCE and *J*sc of the perovskite-based solar cells is charge collection efficiency, instead of charge separation efficiency. Thus suppressing the recombination through proper interfacial engi‐ neering for perovskite-based solid hybrid solar cells is crucial for achieving a high efficiency.

For the Sn/Pb cocktail perovskite solar cells, it is determined that the bandgap is 1.21 eV and light harvesting can be extended to a wider wavelength over 1000 nm. The Urbach energy is calculated to be 34 meV, which is more than twice that of the MAPbI3 perovskite. Three recombination processes for the photoexcited carriers were found with lifetimes being 4 ps (55%), 650 ps (25%) and one much larger than 3 ns (20%). The larger Urbach energy and the faster recombination suggest that there are defects in the prepared Sn/Pb cocktail perovskite. These results are significantly different from those of the MAPbI3 perovskite, in which case the photocarrier lifetimes are larger than 100 ns and almost no defects were observed. Moreover, the charge separation and charge recombination dynamics were explored. We find that both the electron injection into TiO2 and hole transfer to the P3HT layer occurred on a timescale of 1 ps. It was surprising to find that the charge recombination lifetime at the Sn/Pb cocktail perovskite/TiO2 interface was as long as 880 μs, which is 2-3 orders of magnitude greater than that occurring at the MAPbI3/TiO2 interface. However, the charge recombination lifetime became shorter, i.e., 190 μs, when P3HT was employed as a HTM. This implies that the charges can be more effectively collected without the HTM. On the basis of the above results, we find that the loss in the charge separation efficiency originates from fast photoexcited carrier recombination with a lifetime of 4 ps, and the loss in charge collection efficiency was due to the charge recombination occurring at the TiO2/P3HT interface, thus leading to lower IPCE values. Also, the low Voc and FF were found to result from the large Urbach energy and the recombination occurring at the TiO2/P3HT interface. These findings indicate that the photo‐ voltaic performance of Sn/Pb cocktail perovskite solar cells can be further improved by reducing the defects in the material and the recombination occurring at the TiO2/HTM interface


[15] S. Sun, T. Salim, N. Mathews, M. Duchamp, C. Boothroyd, G. Xing, T. C. Sum and Y.

[16] S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou, M. J. P. Alcocer, T. Leijtens, L.

[17] G. C. Xing, N. Mathews, S. Y. Sun, S. S. Lim, Y. M. Lam, M. Gratzel, S. Mhaisalkar

[18] S. De Wolf, J. Holovsky, S.-J. Moon, P. Löper, B. Niesen, M. Ledinsky, F.-J. Haug, J.-

[19] C. C. Stoumpos, C. D. Malliakas and M. G. Kanatzidis, *Inorg. Chem.*, 2013, 52,

[20] Y. Takahashi, R. Obara, Z.-Z. Lin, Y. Takahashi, T. Naito, T. Inabe, S. Ishibashi and K.

[21] F. Hao, C. C. Stoumpos, D. H. Cao, R. P. H. Chang and M. G. Kanatzidis, *Nat. Photon*,

[22] N. K. Noel, S. D. Stranks, A. Abate, C. Wehrenfennig, S. Guarnera, A. Haghighirad, A. Sadhanala, G. E. Eperon, S. K. Pathak, M. B. Johnston, a. petrozza, L. Herz and H.

[23] Y. Ogomi, A. Morita, S. Tsukamoto, T. Saitho, N. Fujikawa, Q. Shen, T. Toyoda, K. Yoshino, S. S. Pandey, T. Ma and S. Hayase, *J. Phys. Chem. Lett.*, 2014, 5, 1004-1011. [24] F. Hao, C. C. Stoumpos, R. P. H. Chang and M. G. Kanatzidis, *J. Am. Chem. Soc.*, 2014,

[25] Q. Shen, Y. Ogomi, J. Chang, S. Tsukamoto, K. Kukihara, O. Takuya, N. Osada, K. Yoshino, K. Katayama, T. Toyoda and S. Hayase, *Phys. Chem. Chem. Phys.*, 2014, 16,

[27] Q. Shen, Y. Ogomi, B.-w. Park, T. Inoue, S. S. Pandey, A. Miyamoto, S. Fujita, K. Ka‐ tayama, T. Toyoda and S. Hayase, *Phys. Chem. Chem. Phys.*, 2012, 14, 4605-4613. [28] Q. Shen, Y. Ogomi, S. K. Das, S. S. Pandey, K. Yoshino, K. Katayama, H. Momose, T.

[29] N. Maeda, H. Hata, N. Osada, Q. Shen, T. Toyoda, S. Kuwahara and K. Katayama,

[30] N. Osada, T. Oshima, S. Kuwahara, T. Toyoda, Q. Shen and K. Katayama, *Phys.*

Toyoda and S. Hayase, *Phys. Chem. Chem. Phys.*, 2013, 15, 14370-14376.

M. Herz, A. Petrozza and H. J. Snaith, *Science*, 2013, 342, 341-344.

H. Yum and C. Ballif, *J. Phys. Chem. Lett.*, 2014, 5, 1035-1039.

M. Lam, *Energy Environ. Sci.*, 2014, 7, 399-407.

420 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

and T. C. Sum, *Science*, 2013, 342, 344-347.

Terakura, *Dalton Trans.*, 2011, 40, 5563-5568.

Snaith, *Energy Environ. Sci.*, 2014, 7, 3061-3068.

[26] A. Rosencwaig and A. Gersho, *J. Appl. Phys.*, 1976, 47, 64-69

*Phys. Chem. Chem. Phys.*, 2013, 15, 11006-11013.

*Chem. Chem. Phys.*, 2014, 16, 5774-5778.

9019-9038.

2014, 8, 489-494.

136, 8094-8099.

19984-19992.


## **Optical, Excitonic, and Electronic Properties of CH3NH3PbI3 Thin Films and Their Application in Photovoltaics**

Sheng Hsiung Chang, Hsin-Ming Cheng, Sheng-Hui Chen and Kuen-Feng Lin

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61278

## **Abstract**

In the past two years, the highest power conversion efficiency of perovskite absorber (PA)–based photovoltaics has been 20.2%. The PA can be fabricated on flat substrates (for example, ZnO, TiO2, and PEDOT:PSS) using solution processes, which have a low-cost advantage in terms of industry production. In this report, the recent advances of PAbased photovoltaics will be mentioned. Then, the optoelectronic properties of PA, materi‐ al fabrication, and photovoltaic performance will be discussed. On the other hand, we used scanning electron microscopy, two-dimensional X-ray diffractometer, and photolu‐ minescence spectroscopy to investigate the fundamental properties of CH3NH3PbI3 thin films fabricated with and without toluene washing treatment, which provides an assess‐ ment of the development potential of PA-based photovoltaics.

**Keywords:** Perovskite photovoltaic, Exciton, Nanoplasmonic structure, 2D X-ray diffrac‐ tion

## **1. Introduction**

The perovskite structure is named after the structure of calcium carbonate (CaTiO3) compound and its molecular formula is ABX3. Perovskite oxide materials (SrTiO3 and LaAlO3) [1] and nonoxide perovskite materials (MgCNi3) [2] have high-temperature superconductivity properties, which indicate that perovskite structures have excellent electrical properties. In the past two years, the most popular light-absorbing materials for photovoltaic cells have been organic–inorganic lead halide perovskite absorbers (CH3NH3PbI3), in which absorption bandgap is approximately 1.57 eV [3] and exciton binding energy is less than 50 meV [4], hence

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

they have excellent power conversion efficiency (PCE). The molecular structure of the earliest application of perovskite materials in an optoelectronic device ((C6H5C2H4NH3)2PbI4) was in 1994, in Kyushu University, Japan, where Tsutsui's research focused on the fluorescence characteristics of materials via electrical excitation [5]. Additionally, the emission wavelength of the luminous element was 520 nm at a low temperature, the full-width at half-maximum (FWHM) of fluorescence was about 10 nm, and the luminance of the perovskite-based device was 10,000 cd/m2 with an injection current density of 2 A/cm2 . The brightness of perovskitebased light-emitting diodes was equivalent to four to five times of the organic light-emitting diode. In April 2009, Miyasaka's research team from the University of Tokyo published the world's first article relating to perovskite-sensitized solar cells. Their study showed that CH3NH3PbI3 and CH3NNH3PbBr3 could convert sunlight into electrical energy as lightabsorbing materials through dye-sensitized solar cell structures, and the obtained PCE was 3.81% [6]. Although the study was published in an important academic journal, it did not receive much attention because of poor efficiency. In 2012, Gratzel from the Swiss Federal Institute of Technology, Lausanne (EPFL), optimized the device parameters of the dyesensitized solar cells to enhance the PCE to 7.28% [7] under 0.1 sunlight intensity. In the same year, Snaith led his research team from the University of Oxford in England to publish their research on organic–inorganic lead halide perovskite photovoltaics using solution process in a scientific journal. They used an insulator Al2O3 nanoparticle film to replace the TiO2 nanoparticle film for the first time, and named the structure meso-superstructured solar cell (MSSC). Additionally, the PCE reached a record of 10.9% [8]. Since then, scientists have launched a series of battles for higher PCE. In July 2013, Bruschka and Pellet adopted a twostep solution process method to produce CH3NH3PbI3 on the porous TiO2 substrate and the efficiency obtained was 12.9% [9]. Also, in September 2013, Snaith produced high-quality perovskite light-absorbing materials using a vapor deposition method and obtained a PCE of 5.4% [10]. In addition, in July 2014, Professor Wu from the National Central University used PC71BM thin film as an electron transport layer and improved the PCE up to 16.3% [11]. In August 2014, Yang's team optimized the energy level of each layer of the solar cells and increased the PCE to 19.3% [12]. Gratzel and Park also optimized the proportion of CH3NH3I and PbI2 in the two-step solution process method and achieved an average PCE of up to 16.3% (±0.35%) [13].

This chapter will describe the optoelectronic properties, molecular structure, and fabrication method of perovskite materials, and calculate the effects of the nanoplasmonic structure on exciton generation and the dissociation of perovskite materials. In addition, it will use scanning electron microscope (SEM), two-dimensional X-ray diffractometer (2D-XRD), UV-vis spec‐ trometer and photoluminescence to explore the morphology, structure, and electronic and excitonic properties of CH3NH3PbI3 thin films formed by a solution process with and without a nonpolar solvent washing treatment [14], and will discuss the development direction and potential of the perovskite photovoltaics.

## **2. The optical and electronic properties of perovskite absorbers**

The perovskite structure is shown in Figure 1. According to the first-principles calculation, each CH3NH3 and Pb provides one and two electrons to I3, and the structure of CH3NH3 and Pb-I can maintain electric neutrality due to van der Waals force, so CH3NH3 cations hardly affect the characteristics of Pb in the conduction band and "I" determines the properties of the valence band [15]. Although the CH3NH3 cation has no significant effect on the electronic structure, the orientation of the CH3NH3 cation can influence the dielectric constant of CH3NH3PbI3 materials. As shown in Table 1, the exciton binding energy can be calculated by *Eb* = *e*<sup>2</sup> /[4πε0ε*dr*], where, *r* is the Bohr radius (2.8 nm) [16] and *ε<sup>d</sup>* is the dielectric constant [17] of CH3NH3PbI3. Exciton binding energy is as low as possible to be more favorable for exciton self-dissociation at room temperature and exciton dissociation at the interface to produce photocurrent, so the orientation of CH3NH3 cation is very essential to excitonic properties. The most direct physical parameter for device efficiency is the exciton diffusion length. Snaith estimated that the exciton diffusion length could be over 1 μm, through measurement of the exciton's lifetime. The long diffusion length in poly-crystalline CH3NH3PbI3 thin films has been directly observed using a spatial and temporal imaging system [18]. Therefore, photons can be effectively converted into photocurrent after being absorbed. Experimental results showed that the exciton lifetime of the perovskite materials could exceed 100 ns [19].

**Figure 1.** Perovskite structure. MA is [CH3NH3] + [15].

they have excellent power conversion efficiency (PCE). The molecular structure of the earliest application of perovskite materials in an optoelectronic device ((C6H5C2H4NH3)2PbI4) was in 1994, in Kyushu University, Japan, where Tsutsui's research focused on the fluorescence characteristics of materials via electrical excitation [5]. Additionally, the emission wavelength of the luminous element was 520 nm at a low temperature, the full-width at half-maximum (FWHM) of fluorescence was about 10 nm, and the luminance of the perovskite-based device

based light-emitting diodes was equivalent to four to five times of the organic light-emitting diode. In April 2009, Miyasaka's research team from the University of Tokyo published the world's first article relating to perovskite-sensitized solar cells. Their study showed that CH3NH3PbI3 and CH3NNH3PbBr3 could convert sunlight into electrical energy as lightabsorbing materials through dye-sensitized solar cell structures, and the obtained PCE was 3.81% [6]. Although the study was published in an important academic journal, it did not receive much attention because of poor efficiency. In 2012, Gratzel from the Swiss Federal Institute of Technology, Lausanne (EPFL), optimized the device parameters of the dyesensitized solar cells to enhance the PCE to 7.28% [7] under 0.1 sunlight intensity. In the same year, Snaith led his research team from the University of Oxford in England to publish their research on organic–inorganic lead halide perovskite photovoltaics using solution process in a scientific journal. They used an insulator Al2O3 nanoparticle film to replace the TiO2 nanoparticle film for the first time, and named the structure meso-superstructured solar cell (MSSC). Additionally, the PCE reached a record of 10.9% [8]. Since then, scientists have launched a series of battles for higher PCE. In July 2013, Bruschka and Pellet adopted a twostep solution process method to produce CH3NH3PbI3 on the porous TiO2 substrate and the efficiency obtained was 12.9% [9]. Also, in September 2013, Snaith produced high-quality perovskite light-absorbing materials using a vapor deposition method and obtained a PCE of 5.4% [10]. In addition, in July 2014, Professor Wu from the National Central University used PC71BM thin film as an electron transport layer and improved the PCE up to 16.3% [11]. In August 2014, Yang's team optimized the energy level of each layer of the solar cells and increased the PCE to 19.3% [12]. Gratzel and Park also optimized the proportion of CH3NH3I and PbI2 in the two-step solution process method and achieved an average PCE of up to 16.3%

This chapter will describe the optoelectronic properties, molecular structure, and fabrication method of perovskite materials, and calculate the effects of the nanoplasmonic structure on exciton generation and the dissociation of perovskite materials. In addition, it will use scanning electron microscope (SEM), two-dimensional X-ray diffractometer (2D-XRD), UV-vis spec‐ trometer and photoluminescence to explore the morphology, structure, and electronic and excitonic properties of CH3NH3PbI3 thin films formed by a solution process with and without a nonpolar solvent washing treatment [14], and will discuss the development direction and

The perovskite structure is shown in Figure 1. According to the first-principles calculation, each CH3NH3 and Pb provides one and two electrons to I3, and the structure of CH3NH3 and

**2. The optical and electronic properties of perovskite absorbers**

. The brightness of perovskite-

with an injection current density of 2 A/cm2

424 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

was 10,000 cd/m2

(±0.35%) [13].

potential of the perovskite photovoltaics.


**Table 1.** Anisotropic dielectric constants and exciton binding energies of CH3NH3PbI3 crystals.

From the perspective of materials design, "I" can be replaced by Br or Cl, and hence the absorption bandgap of CH3NH3PbI3, CH3NH3PbBr3, and CH3NH3PbCl3 are 1.6, 1.95, and 2.46 eV [20], respectively, so the absorption bandgap will increase with the decrease of halogens' atomic number. For application, it can be used with different ratios of halogen to achieve colorful commercial applications. However, the greater the absorption bandgap, the lower the amount of sunlight that can be absorbed. Also, the toxicity of heavy metal Pb is a key factor hindering the development of perovskite solar cells; thus, one of the directions for future development is to use nontoxic Sn to replace Pb. As the conduction band of CH3NH3SnI3 is 0.24 eV lower than that of CH3NH3PbI3, CH3NH3SnI3 has a lower absorption bandgap at about 1.3 eV [21]. So far, the maximum PCE of CH3NH3SnI3-based photovoltaics is about 5.2%, so there is still a lot of room for improvement in the material and device production.

For the part of the organic molecules, based on the theoretical calculation, Walsh proposed in October 2013, that the absorption bandgap of NH4PbI3 could decrease to 1.20 eV by replacing CH3NH3 with NH4, which was expected to absorb more sunlight [17]. That following year, Walsh pointed out that the carrier recombination rate [22] in the perovskite absorbers was strongly associated with the strength of electric dipole moment of the organic cation, and he suggested that the use of larger electric dipole moments of CF3NH3 to replace CH3NH3 could decrease the carrier recombination rate in the perovskite absorbers. In January 2014, Snaith used HC(NH2)2 to replace CH3NH3 and formed HC(NH2)2PbI3 to obtain a lower absorption bandgap at about 1.48 eV, and the PCE of HC(NH2)2PbI3-based photovoltaics reached 14.2% [23]. In August 2014, Park produced HC(NH2)2PbI3 on the substrate of mesoporous TiO2/FTO and obtained device PCE of up to 16% [24]. Although the exciton diffusion length in perovskite materials could reach to micrometers, the thickness of light absorption materials in perovskitebased photovoltaics shall be controlled at around 300 to 400 nm to obtain a better PCE: the optimal short-circuit current density (*J*SC) is about 22 mA/cm2 , depending on the thickness and bandgap of light absorption materials; the maximum open circuit voltage (*V*OC) is about 1.1 V depending on device structure; and the optimal fill factor (FF) is about 75%, which is lower than the FF of gallium arsenide photovoltaics (FF = 86%) [25]. Recently, a high PCE of 20.2% was achieved using HC(NH2)2PbI3 (FAPbI3) as the light absorber in photovoltaics [26].

The FF of photovoltaics indicates the recombination status—the carriers will recombine inside and on the interface of light absorption materials. When the light absorption materials are thicker, the carrier recombination mainly occurs inside the light absorption materials. The decrease of thickness of the light absorption material will reduce the amount of absorbed sunlight, resulting in a decrease in short-circuit current density. For the same light absorption material, the appropriate thickness can be calculated by the product of optimal *J*sc and FF. At present, there are theoretical calculations for optimizing the thickness of perovskite light absorption materials [27], but there are still no experimental results on the relationship between the changes in thickness of perovskite absorbers and the PCE of photovoltaics.

## **3. Exciton and carrier in perovskite materials**

The energy is obtained by absorbing the photon, and thus the electron is excited from the ground state shifting to the excited state. Such an excited state can be regarded as an electron and a hole that is bound together by electrostatic attraction, and such an electron–hole pair is called an exciton.

Figure 2 shows when the electron in the ground state is excited to the excited state in perovskite materials, such an electron in the excited state will be called a hot electron. The hot electron can interact with the crystal lattice to generate coherent and incoherent collisions with two different paths back to the ground state and the conduction band. Raman scattering signal is generated by the coherent collision caused by hot electrons and lattice vibration. However, as the frequency of the hot electron vibration is much higher than that of the lattice vibration, Raman scattering signal strength is very weak. On the other hand, after the incoherent collision produced by the interaction between hot electrons and lattice vibration, the electron which is relaxed in the conduction band is called a cold electron. At this time, the electrons in the conduction band and the holes in the valence band will form excitons. For CH3NH3PbI3 absorbers, the lifetime of hot electrons is about 1 ps [28], and the exciton lifetime associated with the material structure is approximately 10–100 ns.

**Figure 2.** Energy relaxation pathways of photoexcited electrons.

development is to use nontoxic Sn to replace Pb. As the conduction band of CH3NH3SnI3 is 0.24 eV lower than that of CH3NH3PbI3, CH3NH3SnI3 has a lower absorption bandgap at about 1.3 eV [21]. So far, the maximum PCE of CH3NH3SnI3-based photovoltaics is about 5.2%, so

For the part of the organic molecules, based on the theoretical calculation, Walsh proposed in October 2013, that the absorption bandgap of NH4PbI3 could decrease to 1.20 eV by replacing CH3NH3 with NH4, which was expected to absorb more sunlight [17]. That following year, Walsh pointed out that the carrier recombination rate [22] in the perovskite absorbers was strongly associated with the strength of electric dipole moment of the organic cation, and he suggested that the use of larger electric dipole moments of CF3NH3 to replace CH3NH3 could decrease the carrier recombination rate in the perovskite absorbers. In January 2014, Snaith used HC(NH2)2 to replace CH3NH3 and formed HC(NH2)2PbI3 to obtain a lower absorption bandgap at about 1.48 eV, and the PCE of HC(NH2)2PbI3-based photovoltaics reached 14.2% [23]. In August 2014, Park produced HC(NH2)2PbI3 on the substrate of mesoporous TiO2/FTO and obtained device PCE of up to 16% [24]. Although the exciton diffusion length in perovskite materials could reach to micrometers, the thickness of light absorption materials in perovskitebased photovoltaics shall be controlled at around 300 to 400 nm to obtain a better PCE: the

bandgap of light absorption materials; the maximum open circuit voltage (*V*OC) is about 1.1 V depending on device structure; and the optimal fill factor (FF) is about 75%, which is lower than the FF of gallium arsenide photovoltaics (FF = 86%) [25]. Recently, a high PCE of 20.2% was achieved using HC(NH2)2PbI3 (FAPbI3) as the light absorber in photovoltaics [26].

The FF of photovoltaics indicates the recombination status—the carriers will recombine inside and on the interface of light absorption materials. When the light absorption materials are thicker, the carrier recombination mainly occurs inside the light absorption materials. The decrease of thickness of the light absorption material will reduce the amount of absorbed sunlight, resulting in a decrease in short-circuit current density. For the same light absorption material, the appropriate thickness can be calculated by the product of optimal *J*sc and FF. At present, there are theoretical calculations for optimizing the thickness of perovskite light absorption materials [27], but there are still no experimental results on the relationship between

The energy is obtained by absorbing the photon, and thus the electron is excited from the ground state shifting to the excited state. Such an excited state can be regarded as an electron and a hole that is bound together by electrostatic attraction, and such an electron–hole pair is

Figure 2 shows when the electron in the ground state is excited to the excited state in perovskite materials, such an electron in the excited state will be called a hot electron. The hot electron can interact with the crystal lattice to generate coherent and incoherent collisions with two

the changes in thickness of perovskite absorbers and the PCE of photovoltaics.

, depending on the thickness and

there is still a lot of room for improvement in the material and device production.

optimal short-circuit current density (*J*SC) is about 22 mA/cm2

426 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**3. Exciton and carrier in perovskite materials**

called an exciton.

Exciton binding energy is a very important parameter for the light absorption materials of photovoltaics. When the exciton in the heterogeneous interface of different materials is dissociated by the built-in electric field and becomes a free electron and hole after overcoming the binding energy. The exciton binding energy for organic materials is about 150 to 1000 meV, so with the exciton dissociation at the region of charge transfer radius, the potential difference of p–n interface must be larger than the exciton binding energy. Taking P3HT/PCBM as an example, the exciton binding energy of P3HT is about 300 meV, and LUMO potential difference of P3HT-PCBM interface is about 800 meV (3.9–3.1 eV); therefore, the exciton in the P3HT-PCBM interface can be smoothly dissociated with approximately 1.2 ps exciton dissociation time [29]. The exciton binding energy of CH3NH3PbI3 is smaller than 50 meV but higher than the thermal energy of room temperature (KBT = 25.6 meV, *T* = 25°C). Therefore, about half of the excitons will self-dissociate to become free carriers at room temperature [30], and the other half of the excitons will have to diffuse to the interface to dissociate and become free carriers (electrons and holes).

The electrons and holes formed by the self-dissociation of excitons at room temperature must not recombine, and each has to transport to the electrodes to form a photocurrent. In March 2014, Walsh, based on theoretical calculations, predicted that the transporting paths of electrons and holes inside CH3NH3PbI3 are different (see Figure 3) [22], and estimated that the carrier recombination rate in CH3NH3PbI3 was very low. The arrow direction shown in Figure 3 is the orientation of the CH3NH3 cation (electric dipole). Electrons move along the lowest potential (blue arrow) while the holes flow along the maximum potential (red arrow). Therefore, the electric dipole of the organic cation would affect the recombination probability of electrons and holes. Also Walsh pointed out that the CF3NH3 cations with larger electric dipole moments should be able to effectively reduce the carrier recombination rate in perov‐ skite absorbers.

**Figure 3.** The propagation pathways of electrons and holes in CH3NH3PbI3 [22].

## **4. Device architecture and manufacturing process of perovskite-based photovoltaics**

Perovskite-based photovoltaics with more than 17% PCE can be grouped into three structures: the dye-sensitized solar cell structures, regular-type organic photovoltaic structures, and inverted-type organic photovoltaic structures. The dye-sensitized solar cell structures, based on the process order, includes FTO-conductive glass substrate/TiO2 compact layer/mesopo‐ rous TiO2/HC(NH2)2PbI3 perovskite absorber/PTAA hole transporting layer/Au and its currently published maximum PCE is 20.2% [26]. For regular-type organic photovoltaic structures, its process sequence is ITO-conductive glass/PEDOT:PSS/CH3NH3PbI3 perovskite absorber/PCBM electron transport layer/Al and its currently published maximum PCE is 17.8% [29]. For inverted-type organic photovoltaic structures, its process sequence is ITOconductive glass/PEIE/Y:TiO2/CH3NH3PbI3-*x*Cl*x* perovskite absorber/Spiro-OMeTAD hole transport layer/Au, and its currently published maximum PCE is 19.3% [12]. Currently, the manufacturing process of a regular-type organic photovoltaic structure is the easiest.

Much research has been conducted on the energy conversion of perovskite-based photovol‐ taics in the last two years, mainly because perovskite absorbers can be produced through simple solution process methods. The main fabrication methods include one-step and twostep solution processes. The one-step method is first adopted so that the precursor of perov‐ skite materials is dissolved in dimethyl formamide (DMF), followed by soaking of the perovskite materials on the substrate. However, the biggest drawback of this method is the poor coating for the film, which can easily result in short-circuits due to the contact between the upper and lower electrodes. The use of an appropriate upper electrode can reduce the incidence of short-circuits [32]. As shown in Figure 4, using p-type P3HT can fill the pores of CH3NH3PbI3-*x*Cl*<sup>x</sup>* thin films. The addition of a small amount of 1,8-diiodooctane (DIO) inside the perovskite precursors [33] can effectively reduce the generation of voids. In addition, the perovskite film with high coverage can also be obtained on the rough FTO substrate. Thus, the surface properties of the substrate have an essential effect on the continuity of perovskite thin film.

**Figure 4.** Cross-sectional view of device under a scanning electron microscope [30].

3 is the orientation of the CH3NH3 cation (electric dipole). Electrons move along the lowest potential (blue arrow) while the holes flow along the maximum potential (red arrow). Therefore, the electric dipole of the organic cation would affect the recombination probability of electrons and holes. Also Walsh pointed out that the CF3NH3 cations with larger electric dipole moments should be able to effectively reduce the carrier recombination rate in perov‐

**Figure 3.** The propagation pathways of electrons and holes in CH3NH3PbI3 [22].

428 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**4. Device architecture and manufacturing process of perovskite-based**

manufacturing process of a regular-type organic photovoltaic structure is the easiest.

Much research has been conducted on the energy conversion of perovskite-based photovol‐ taics in the last two years, mainly because perovskite absorbers can be produced through simple solution process methods. The main fabrication methods include one-step and twostep solution processes. The one-step method is first adopted so that the precursor of perov‐ skite materials is dissolved in dimethyl formamide (DMF), followed by soaking of the perovskite materials on the substrate. However, the biggest drawback of this method is the

Perovskite-based photovoltaics with more than 17% PCE can be grouped into three structures: the dye-sensitized solar cell structures, regular-type organic photovoltaic structures, and inverted-type organic photovoltaic structures. The dye-sensitized solar cell structures, based on the process order, includes FTO-conductive glass substrate/TiO2 compact layer/mesopo‐ rous TiO2/HC(NH2)2PbI3 perovskite absorber/PTAA hole transporting layer/Au and its currently published maximum PCE is 20.2% [26]. For regular-type organic photovoltaic structures, its process sequence is ITO-conductive glass/PEDOT:PSS/CH3NH3PbI3 perovskite absorber/PCBM electron transport layer/Al and its currently published maximum PCE is 17.8% [29]. For inverted-type organic photovoltaic structures, its process sequence is ITOconductive glass/PEIE/Y:TiO2/CH3NH3PbI3-*x*Cl*x* perovskite absorber/Spiro-OMeTAD hole transport layer/Au, and its currently published maximum PCE is 19.3% [12]. Currently, the

skite absorbers.

**photovoltaics**

In July 2014, Spiccia and Seok spontaneously used organic solvent (chlorobenzene or toluene) [34,35] to inject perovskite precursor thin films during the spin coating process, which can greatly enhance the film continuity of perovskite thin films. In September 2014, Cheng sprayed argon gas positively into the substrate to improve the solvent evaporation speed in the spin coating process of perovskite precursors and achieved up to 16.97% PCE. Figure 5 shows that perovskite thin films with argon gas processing have good film continuity and crystallinity; on the contrary, the surface of perovskite thin films without gas treatment would appear as reticular defect structures [36]. In November 2014, Jen used different solvents (toluene, chlorobenzene, and dichlorobenzene) to fabricate perovskite thin films with high continuity on top of PEDOT:PSS/ITO/glass substrate, and obtained a high PCE of 13.7% [37]. Moreover, in the manufacturing process, when the heating temperature decreases from 100°C down to 70°C, the photovoltaics still have excellent PCE of 12%. Therefore, such low-temperature deposition manufacturing techniques can be integrally applied to soft substrates.

Bruschka and Pellet from EPFL were the first to adopt a two-step coating method [9] to produce perovskite materials in the porous TiO2 substrates. They first used a spin coating process for even distribution of PbI2 on the mesoporous TiO2 thin films, and then soaked the substrate into the IPA solvent with solved CH3NH3I. Hence, the PbI2 on the TiO2 surface would be reacted to CH3NH3PbI3. Using a long-term interdiffusion process [38], the two-step solution process method can also produce high continuity of perovskite thin films on a flat PEDOT:PSS/ITO/ glass substrate, which is shown in Figure 6.

**Figure 5.** Scanning electron microscopy image of the CH3NH3PbI3 surface: (a) and (c) are fabricated without argon gas; (b) and (d) are fabricated with argon gas [36].

**Figure 6.** Cross-sectional view of multilayer film under a scanning electron microscope [11].

**Figure 7.** The J–V curves and external quantum efficiency (EQE) spectrum of CF3NH3PbI3–based photovoltaics [26].

At present, among the perovskite thin film production methods with three matching photo‐ voltaic structures and two fabrication methods, the two-step solution process method can obtain the highest PCE of 20.2% but has the longest manufacturing process, and its photovol‐ taic performances are shown in Figure 7. In addition, the one-step solution process method with argon gas treatment, although with a slightly lower PCE of 16.97%, can save the most time and is suitable for making large area high-quality perovskite thin films.

## **5. Exciton properties in the interface between metal particles and perovskite absorbers**

In order to be able to simultaneously improve the JSC and FF of perovskite-based photovoltaics, we calculate the exciton properties generated by the nanoplasmonic structure embedded in perovskite absorbers to analyze the effects of nanoplasmonic structures on the properties of perovskite-based photovoltaics.

The essential parameters for electromagnetic simulations are optical parameters (refractive index and absorption coefficient), using a transfer-matrix method with Lorentz model to describe the dielectric constant of the perovskite absorber. Figure 8 shows the transmittance spectrum of perovskite/substrate, which are used for calculating the refractive index and extinction coefficient of perovskite absorber, as shown in Figure 9. Moreover, we buried a nanoplasmonic structure inside the perovskite thin film (see Figure 10), and selected appro‐ priate structure parameters (the period = 100 nm and the metal ellipsoid length = 150 nm). Figure 11 shows the absorption spectra of nanoplasmonic structures embedded in perovskite absorber with different ellipsoid gap. When the distance between the ellipsoid is 30 nm (the short axis of the ellipsoid is 70 nm), it has a better absorption enhancement effect; at the same time, when the thickness of the perovskite absorber is decreased to 300 nm, it is expected to have a better FF.

**Figure 8.** Transmittance spectra of CH3NH3PbI3/substrate [39].

**Figure 6.** Cross-sectional view of multilayer film under a scanning electron microscope [11].

(b) and (d) are fabricated with argon gas [36].

430 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 7.** The J–V curves and external quantum efficiency (EQE) spectrum of CF3NH3PbI3–based photovoltaics [26].

**Figure 5.** Scanning electron microscopy image of the CH3NH3PbI3 surface: (a) and (c) are fabricated without argon gas;

**Figure 9.** Refractive index and extinction coefficient of CH3NH3PbI3 thin film [39].

**Figure 10.** Side view and top view of nanoplasmonic structure embedded in the perovskite absorber [39].

**Figure 11.** Absorption spectra of nanoplasmonic structures embedded in perovskite absorber with different structure parameters [39].

The nanoplasmonic structures enhanced absorptions have currently been realized in the organic solar cells [40]. However, the effect of the spatial distribution of the exciton around nanoplasmonic structures is seldom discussed. Figure 12 shows the spatial distribution of electric field (exciton). When the sunlight transmits from the ITO/glass substrate to the nanoplasmonic structure, the light field is localized at the interface between the CH3NH3PbI3 and the Cu ellipsoid, thus most of the excitons are generated at the interface, and the exciton can be dissociated at the interface between CH3NH3PbI3 (valance band = –5.4 eV) and Cu (Fermi energy level = –4.94 eV). After the exciton dissociation, the hole can propagate along the Cu ellipsoid to the ITO electrode, while the electron can propagate along the CH3NH3PbI3, which will help to inhibit the carrier recombination inside the CH3NH3PbI3. Thus, the introduction of the nanoplasmonic structure can improve the FF of CH3NH3PbI3-based photovoltaics while enhancing the light absorption.

**Figure 12.** The distribution of the electric field (exciton) with different ellipsoid gap [39].

## **6. Future development direction**

**Figure 9.** Refractive index and extinction coefficient of CH3NH3PbI3 thin film [39].

432 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 10.** Side view and top view of nanoplasmonic structure embedded in the perovskite absorber [39].

**Figure 11.** Absorption spectra of nanoplasmonic structures embedded in perovskite absorber with different structure

parameters [39].

The PCE of GaAs single-junction photovoltaics is 28.9%, which is the closest to the Shockley– Queisser (SQ) limit of 33.7%. According to SQ theory, if the photon energy is larger than the energy gap (*E*g) of light absorption materials, the energy of the incident photon will be converted from the electron to the excited state. The electron in the excited state (hot electron) and lattice will have incoherent collision and then relax to the conduction band to form exciton with the hole in the valence band, and such excitons must be dissociated to generate photo‐ current. Also, assuming that the remaining energy of the absorbed photon (*E*<sup>P</sup> = 2.26 eV) after thermal relaxation is equal to the energy gap of the material (*E*<sup>g</sup> = 1.43 eV), so if a photon can produce a free electron, the voltage loss of this photoelectric conversion process is 0.83 V (*E*P/ *e*-*E*g/*e*). Additionally, we can also use a simple formula to evaluate the SQ limit:

$$PCE = A \times FF \times (E\_{\mathbb{Z}} \mid E\_P)\_{\prime} \tag{1}$$

where *PCE* is theoretical PCE, *A* is the amount of sunlight which can be absorbed by the material, and *FF* is fill factor. Taking GaAs and CH3NH3PbI3 as examples, their SQ limits are 32.7% (0.6 × 0.86 × 1.43/2.26) and 26.6% (0.5 × 0.75 × 1.6/2.26), respectively, which indicates that the solar cells with CH3NH3PbI3 as the main material still have space to improve.

In the past ten years, low production costs have been an advantage for organic photovoltaics and dye-sensitized solar cells, but the PCE is always less than 15%. Except that the absorption bandgap of materials cannot be effectively extended to the near-infrared region, the interaction between the light and the material is also the reason for the constrained efficiency. Taking P3HT-based photovoltaics for example, they are constrained by the exciton binding energy (~300 meV) and exciton diffusion length (~10 nm), and they must use the p–n interface with larger potential difference (△*V*P3HT-ICBA = 0.64 V) and blended thin film structure (P3HT:ICBA) in order to effectively convert excitons into the photocurrent. If we use formula (1) to estimate the theoretical PCE of P3HT:ICBA, *E*g has to be replaced with *eV*OC to obtain a PCE of 8.16% (0.32 × 0.75 × 0.87/2.26), which is much closer to the experimental value (7.4%) [41]. In addition, the absorption bandgap and the exciton binding energy seem to limit the development of organic photovoltaics. Table 2 indicates different photovoltaic performances and the charac‐ teristics of light absorption materials, and exciton binding energy of organic P3HT is much larger than inorganic materials (*E*<sup>b</sup> < 20 meV). Also noteworthy is the Raman shift of organic materials (1000 to 1700 cm–1), which is larger than that of inorganic materials (<600 cm–1), even much larger than Raman scattering of CH3NH3PbI3 (Pb-I stretching: 91 cm–1). Also, the phonon energy of organic materials is much larger than that of inorganic materials. Because a large part of the energy is converted into thermal energy after the photons are absorbed by organic materials, it thus restricts the development of organic photovoltaics and dye-sensitized solar cells.


**Table 2.** Different photovoltaic performances and the characteristics of light absorption materials

According to an estimation of the SQ limit, there is still space for the development of perov‐ skite-based photovoltaics. From the perspective of the manufacturing process, most of the research over the past two years has focused on how to enhance the continuity of the CH3NH3PbI3 film to avoid contact between the upper and lower electrodes, which can result in a short-circuit situation. At first, using perovskite materials with a mixture of Cl can obtain better film continuity. As the mixed amount of Cl is very small, the contribution of Cl is to decrease the crystallinity of CH3NH3PbI3 and reduce the agglomeration of particles due to the rapid crystallization to further enhance the continuity of the films [42]. Moreover, the different surface roughness of substrate (ITO and FTO) can influence the continuously of CH3NH3PbI3 films, CH3NH3PbI3 fabricated in a more rough substrate will have a better film continuity, which is interpreted as a rough surface that is unfavorable for CH3NH3PbI3 crystallization. Furthermore, organic solvent or inert gas is injected into CH3NH3PbI3 precursors during the spin coating process, which can obtain CH3NH3PbI3 thin films with good continuity and excellent PCE of 17.8%. In order to verify the mechanism of this method, we used a SEM, a 2D-XRD and fluorescence spectroscope to analyze CH3NH3PbI3 thin films fabricated with and without toluene washing treatment [14].

where *PCE* is theoretical PCE, *A* is the amount of sunlight which can be absorbed by the material, and *FF* is fill factor. Taking GaAs and CH3NH3PbI3 as examples, their SQ limits are 32.7% (0.6 × 0.86 × 1.43/2.26) and 26.6% (0.5 × 0.75 × 1.6/2.26), respectively, which indicates that

In the past ten years, low production costs have been an advantage for organic photovoltaics and dye-sensitized solar cells, but the PCE is always less than 15%. Except that the absorption bandgap of materials cannot be effectively extended to the near-infrared region, the interaction between the light and the material is also the reason for the constrained efficiency. Taking P3HT-based photovoltaics for example, they are constrained by the exciton binding energy (~300 meV) and exciton diffusion length (~10 nm), and they must use the p–n interface with larger potential difference (△*V*P3HT-ICBA = 0.64 V) and blended thin film structure (P3HT:ICBA) in order to effectively convert excitons into the photocurrent. If we use formula (1) to estimate the theoretical PCE of P3HT:ICBA, *E*g has to be replaced with *eV*OC to obtain a PCE of 8.16% (0.32 × 0.75 × 0.87/2.26), which is much closer to the experimental value (7.4%) [41]. In addition, the absorption bandgap and the exciton binding energy seem to limit the development of organic photovoltaics. Table 2 indicates different photovoltaic performances and the charac‐ teristics of light absorption materials, and exciton binding energy of organic P3HT is much larger than inorganic materials (*E*<sup>b</sup> < 20 meV). Also noteworthy is the Raman shift of organic materials (1000 to 1700 cm–1), which is larger than that of inorganic materials (<600 cm–1), even much larger than Raman scattering of CH3NH3PbI3 (Pb-I stretching: 91 cm–1). Also, the phonon energy of organic materials is much larger than that of inorganic materials. Because a large part of the energy is converted into thermal energy after the photons are absorbed by organic materials, it thus restricts the development of organic photovoltaics and dye-sensitized solar

**Absorber** *VOC (V)* **PCE (%)** *Eg (eV) Eb (meV)* **Raman shift (cm–1)** GaAs 1.12 28.9 1.43 4 LO 295 Si 0.71 25.0 1.11 18 520 InP 0.88 22.1 1.34 5 LO 348 CIGS 0.72 19.8 1.2 12 Cu-S 472 CdTe 0.86 19.6 1.5 10.6 LO 162 CH3NH3PbI3 1.13 19.3 1.6 50 LO 91 Prophyrin dye 0.91 12.3 1.79 330 1000–1700 P3HT 0.75 7.4 1.9 300 1000–1700

According to an estimation of the SQ limit, there is still space for the development of perov‐ skite-based photovoltaics. From the perspective of the manufacturing process, most of the research over the past two years has focused on how to enhance the continuity of the

**Table 2.** Different photovoltaic performances and the characteristics of light absorption materials

the solar cells with CH3NH3PbI3 as the main material still have space to improve.

434 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

cells.

Figure 13 shows the morphologies of CH3NH3PbI3 thin films fabricated with and without toluene treatment. The CH3NH3PbI3 thin film without toluene treatment appeared as a flowerlike structure on rough surfaces, indicating that CH3NH3PbI3 may have specific crystallization orientation. Additionally, the CH3NH3PbI3 thin film with toluene treatment appeared to be a more even and continuous film.

**Figure 13.** The SEM images of CH3NH3PbI3 thin films; (a) without toluene treatment; (b) with toluene treatment [14].

In order to understand the crystallization orientation of CH3NH3PbI3 materials, we use 2D-XRD as a detection tool. Figure 14 shows 2D-XRD images with and without toluene treatment. For the CH3NH3PbI3 thin film without toluene treatment, the preferred orientation of the crystallization tendency is at (112), while the CH3NH3PbI3 thin film with toluene treatment has no specific crystallization orientation, which indicates that CH3NH3PbI3 precursors with the toluene treatment will interfere with the crystallization of CH3NH3PbI3 and thus obtain more smooth and continuous CH3NH3PbI3 thin films. Analysis of the FWHM of X-ray diffraction peak can also obtain the crystal domain size of CH3NH3PbI3 thin films with and without toluene treatment at 17.6 nm and 29.6 nm, respectively, indicating that the toluene treatment indeed inhibited the crystallization of CH3NH3PbI3.

**Figure 14.** Two-dimensional X-ray diffraction patterns: (a) PEDOT:PSS/ITO/glass; (b) CH3NH3PbI3/PEDOT:PSS/ITO/ glass without toluene treatment; (c) CH3NH3PbI3/PEDOT:PSS/ITO/glass with toluene treatment [14].

Figure 15 displays the absorbance spectra of perovskite absorbers on PEDOT:PSS/ITO/ glass substrate fabricated with and without the toluene treatment. The perovskite films have two absorption peaks at ~480 nm and ~750 nm, which indicate the formation of CH3NH3PbI3. When neglecting the reflectance, the absorbance can be treated as absorption. The absorp‐ tion coefficient can be obtained by *α* = *A/D*, where *A* is the absorbance, and *D* (~260 nm) is the film thickness measured by an *α* step. The absorption coefficient of the CH3NH3PbI3 film fabricated with (without) the toluene treatment is 6.02 × 104 cm–1 (4.63 × 104 cm–1) at a wavelength of 479 nm.

In order to understand the crystallization orientation of CH3NH3PbI3 materials, we use 2D-XRD as a detection tool. Figure 14 shows 2D-XRD images with and without toluene treatment. For the CH3NH3PbI3 thin film without toluene treatment, the preferred orientation of the crystallization tendency is at (112), while the CH3NH3PbI3 thin film with toluene treatment has no specific crystallization orientation, which indicates that CH3NH3PbI3 precursors with the toluene treatment will interfere with the crystallization of CH3NH3PbI3 and thus obtain more smooth and continuous CH3NH3PbI3 thin films. Analysis of the FWHM of X-ray diffraction peak can also obtain the crystal domain size of CH3NH3PbI3 thin films with and without toluene treatment at 17.6 nm and 29.6 nm, respectively, indicating that the toluene treatment indeed

**Figure 14.** Two-dimensional X-ray diffraction patterns: (a) PEDOT:PSS/ITO/glass; (b) CH3NH3PbI3/PEDOT:PSS/ITO/

Figure 15 displays the absorbance spectra of perovskite absorbers on PEDOT:PSS/ITO/ glass substrate fabricated with and without the toluene treatment. The perovskite films have two absorption peaks at ~480 nm and ~750 nm, which indicate the formation of CH3NH3PbI3. When neglecting the reflectance, the absorbance can be treated as absorption. The absorp‐

glass without toluene treatment; (c) CH3NH3PbI3/PEDOT:PSS/ITO/glass with toluene treatment [14].

inhibited the crystallization of CH3NH3PbI3.

436 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 15.** Absorbance spectra of CH3NH3PbI3/PEDOT:PSS/ITO/glass without and with toluene treatment [14].

In order to understand the exciton characteristics of the CH3NH3PbI thin film with toluene treatment, we measured the fluorescence spectra of the sample, as shown in Figure 16. The CH3NH3PbI3 thin film with toluene treatment had weaker fluorescence, showing that the excitons in the CH3NH3PbI3 thin film with toluene treatment had shorter lifetimes. The experimental data showed that the toluene treatment can improve the continuity of CH3NH3PbI3 thin films but inhibit crystallinity, leading to a reduction in the exciton's lifetime. In addition, we also used this method to fabricate photovoltaics with structures starting with the substrate sequentially to glass/ITO/PEDOT:PSS/CH3NH3PbI3/PCBM/Ag, and the obtained highest PCE was 11.7% (*V*OC = 0.92 V, JSC = 18.2 mA/cm2 , FF = 0.7) [14], which proves that toluenetreated CH3NH3PbI can be used as a light absorption material for excellent PCE.

Although the CH3NH3PbI3 absorbers with toluene treatment have good continuity, and the PCE can easily exceed 10%, the morphologies of these perovskite thin films and exciton properties showed that the substrate in the absence of specific crystallization orientation has to sacrifice the crystallinity of perovskite thin films to achieve the good continuity, which could produce high-efficiency photovoltaics through solution process methods. Therefore, we believe that it is necessary to find substrates with crystallinity or specific structure for the production of perovskite-based photovoltaics in order to manufacture perovskite absorbers with both high continuity and crystallization by solution process methods to be able to improve the *V*OC and FF of the photovoltaics.

Since the hot electron lifetime of CH3NH3PbI3 is about 1 ps, high conductive nanostructures can be used to improve the efficiency of hot electron injection from CH3NH3PbI3 to electron

**Figure 16.** The photoluminescence spectra of CH3NH3PbI3 thin films [14].

acceptors. The issue of hot electron injection and multiple exciton generation has been widely discussed in studies of quantum dot–sensitized solar cells. In theoretical estimation, the limit of the PCE of the photovoltaics with hot electron injection is 42% [43], which is higher than the SQ limit of 33.7%. Therefore, we predicted that if we use ZnO with nanorod structures to collect hot electron injection, it should be able to improve the *V*OC of perovskite-based photovoltaics. Because the hot electrons inject into ZnO at a higher energy potential, and if the conductivity of ZnO remains high enough that the hot electrons are able to maintain its energy potential transporting to the electrode, then higher *V*OC should be achieved.

Moreover, in the aging test of perovskite solar cells, Park conducted tests measuring the stability of PCE under different humidity storage conditions [44]. Under 55% relative humid‐ ity, the PCE of CH3NH3PbI3-based photovoltaics decreased from 11% to 3.5% after 20 days of storage. In addition, if using CH3NH3PbI2.4Br0.6 as the light absorption material under the same storage conditions (55% relative humidity), the PCE could be maintained at 9.5% for 20 days. Although "I" of the CH3NH3PbI3 was replaced by Br, which could improve the stability of perovskite-based photovoltaics, an increase of the Br substituted amount would also increase the absorption bandgap of the perovskite materials, which would result in the decrease of PCE from 11% to 9.5%, at the expense of light absorption. The results for aging tests of long-term exposure to sunlight have not yet been published in internationally important journals. As a result, it is still doubtful whether perovskite photovoltaics have long-term stability, which shall be further tested under actual conditions for the properties of the high PCE of perovskite-based photovoltaics.

## **7. Conclusion**

The power conversion efficiency (PCE) of perovskite-based photovoltaics have exceeded 20%, and the efficiency in different types of substrates by solution process methods can reach over 17%, indicating that there is very high possibility for commercializing perovskite-based photovoltaics. However, the light absorption materials contain Pb, which is the biggest obstacle for commercialization of perovskite-based photovoltaics. The absorption bandgap of CH3NH3SnI3 with Pb replaced with Sn is about 1.3 eV, which is a promising light absorption material. So far, the highest PCE of CH3NH3SnI3-based photovoltaics is about 5.2% far away from the Shockley–Queisser limit of 33.7%. Therefore, research on CH3NH3SnI3-based photo‐ voltaics have been slowed due to the benefit issues, and thus it is essential to conduct more basic research to further evaluate the limits of perovskite-based photovoltaics for the arrival of earlier commercialization process.

## **Acknowledgements**

This work was supported by the Material and Chemical Research Laboratories, Industrial Technology Research Center and the National Science Council under Grant NSC 101-2731- M-008-002-MY3.

## **Author details**

acceptors. The issue of hot electron injection and multiple exciton generation has been widely discussed in studies of quantum dot–sensitized solar cells. In theoretical estimation, the limit of the PCE of the photovoltaics with hot electron injection is 42% [43], which is higher than the SQ limit of 33.7%. Therefore, we predicted that if we use ZnO with nanorod structures to collect hot electron injection, it should be able to improve the *V*OC of perovskite-based photovoltaics. Because the hot electrons inject into ZnO at a higher energy potential, and if the conductivity of ZnO remains high enough that the hot electrons are able to maintain its energy potential

Moreover, in the aging test of perovskite solar cells, Park conducted tests measuring the stability of PCE under different humidity storage conditions [44]. Under 55% relative humid‐ ity, the PCE of CH3NH3PbI3-based photovoltaics decreased from 11% to 3.5% after 20 days of storage. In addition, if using CH3NH3PbI2.4Br0.6 as the light absorption material under the same storage conditions (55% relative humidity), the PCE could be maintained at 9.5% for 20 days. Although "I" of the CH3NH3PbI3 was replaced by Br, which could improve the stability of perovskite-based photovoltaics, an increase of the Br substituted amount would also increase the absorption bandgap of the perovskite materials, which would result in the decrease of PCE from 11% to 9.5%, at the expense of light absorption. The results for aging tests of long-term exposure to sunlight have not yet been published in internationally important journals. As a result, it is still doubtful whether perovskite photovoltaics have long-term stability, which shall be further tested under actual conditions for the properties of the high PCE of perovskite-based

The power conversion efficiency (PCE) of perovskite-based photovoltaics have exceeded 20%, and the efficiency in different types of substrates by solution process methods can reach over

transporting to the electrode, then higher *V*OC should be achieved.

**Figure 16.** The photoluminescence spectra of CH3NH3PbI3 thin films [14].

438 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

photovoltaics.

**7. Conclusion**

Sheng Hsiung Chang1\*, Hsin-Ming Cheng2 , Sheng-Hui Chen3 and Kuen-Feng Lin1,3

\*Address all correspondence to: shchang@ncu.edu.tw

1 Research Center for New Generation Photovoltaics, National Central University, Taoyuan, Taiwan, ROC

2 Material and Chemical Research Laboratories, Industrial Technology Research Center, Hsinchu, Taiwan, ROC

3 Department of Optics and Photonics, National Central University, Taoyuan, Taiwan, ROC

## **References**


[16] M. Hirasawa, T. Ishihara, T. Goto, K. Uchida, and N. Miura. Magnetoabsorption of the lowest exciton in perovskite-type compound (CH3CH3)PbI3. *Physica B: Condensed Matter* 201 (1994) 427–430.

[3] G. Xing, N. Mathews, S. Sun, S. S. Lim, Y. M. Lam, M. Gratzel, S. Mhaisalkar, and T. C. Sum. Long-range balanced electron- and hole-transport lengths in organic–inor‐

[4] K. Tanaka, T. Takahashi, T. Ban, T. Kondo, K. Uchida, and N. Miura. *Solid State Com‐*

[5] M. Era, S. Morimoto, T. Tsutsui, and S. Saito. Organic–inorganic heterostructure elec‐ troluminescent device using a layered perovskite semiconductor

[6] A. Kojima, K. Teshima, Y. Shirai, and T. Miyasaka. Organometal halide perovskite as visible-light sensitizers for photovoltaic cells. *Journal of the American Chemical Society*

[7] L. Etgar, P. Gao, Z. Xue, Q. Peng, A. K. Chandiran, B. Liu, Md. K. Nazeeruddin, and M. Gratzel. Mesocopic CH3NH3PbI3/TiO2 heterojunction solar cells. *Journal of The*

[8] Michael M. Lee, Joël Teuscher, Tsutomu Miyasaka, Takurou N. Murakami, Henry J. Snaith. Efficient hybrid solar cells based on meso-superstructured organometal hal‐

[9] J. Burschka, N. Pellet, S-J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin, and M. Gratzel. Sequential deposition as a route to high-performance perovskite-sensi‐

[10] M. Liu, M. B. Johnston, and H. J. Snaith. Efficient planar heterojunction perovskite

[11] C-H. Chiang, Z-L. Tseng, and C-G. Wu. Planar heterojunction perovskite/PC71BM so‐ lar cells with enhanced open-circuit voltage via a (2/1)-step spin-coating process.

[12] H. Zhou, Q. Chen, G. Li, S. Luo, T-B. Song, H-S. Duan, Z. Hong, J. You, Y. Liu, and Y. Yang. Interface engineering of highly efficient perovskite solar cells. *Science* 345

[13] J-H. Im, I-H. Jang, N. Pellet, M. Gratzel, and N-G. Park. Growth of CH3NH3PbI3 cu‐ boids with controlled sized for high-efficiency perovskite solar cells. *Nature Technolo‐*

[14] K-F. Lin, S-H. Chang, K-H. Wang, H-M. Cheng, K. Y. Chiu, K-M. Lee, S-H. Cheng, and C-G. Wu. Unraveling the high performance of tri-iodide perovskite absorber based photovoltaics with a non-polar solvent washing treatment. *Solar Energy Materi‐*

[15] W-J. Yin, T. Shi, and Y. Yan. Unusual defect physics in CH3NH3PbI3 perovskite solar

(C6H5C2H4NH3)2PbI4. *Applied Physics Letters* 65 (1994) 676–678.

ganic CH3NH3PbI3. *Science* 342 (2013) 344–347.

440 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

*American Chemical Society* 134 (2012) 17396–17399.

ide perovskites. *Science* 338 (2014) 643–647.

tized solar cells. *Nature* 499 (2013) 316–319.

*als and Solar Cells*, accepted for publication.

cell absorber. *Applied Physics Letters* 104 (2014) 063903.

solar cells by vapour deposition. *Nature* 501 (2013) 395–398.

*Journal of Materials Chemistry A* 2 (2014) 15897–15903.

*munications* 127 (2003) 619–623.

131 (2009) 6050–6051.

(2014) 542–546.

*gy* 9 (2014) 927–932.


[39] S. H. Chang, K-F. Lin, C-H. Chiang, S-H. Chen, and C-G. Wu. Plasmonic structure enhanced exciton generation at the interface between the perovskite absorber and copper nanoparticles. *Scientific World Journal* 2014 (2014) 128414.

[28] H-Y. Hsu, C-Y. Wang, A. Fathi, J-W. Shiu, C-C. Chung, P-S. Shen, T-F. Guo, P. Chen, Y-P. Lee, and E. W-G. Diau. Femtosecond excitonic relaxation dynamics of perov‐ skite on mesoporous films of Al2O3 and NiO nanoparticles. *Angewandte Chemie* 53

[29] S. H. Chang, C-H. Chiang, H-M. Cheng, C-Y. Tai, and C-G. Wu. Broadband charge transfer dynamics in P3HT:PCBM blended film. *Optics Express* 38 (2013) 5342–5345.

[30] T. C. Sum and N. Mathews. Advancements in perovskite solar cells: photophysics behind the photovoltaics. *Energy Environmental Science* 7 (2014) 2518–2534.

[31] W. Nie, H. Tsai, R. Asadpour, J-C. Blancon, A. J. Neukirch, G. Gupta, J. J. Crochet, M. Chhowalla, S. Tretiak, M. A. Alam, H-L. Wang, and A. D. Mohite. High-efficiency solution-processed perovskite solar cells with millimeter-scale grains. *Science* 347

[32] B. Conings, L. Baeten, C. D. Cobbelaere, J. D'Haen, J. Manca, and H-G. Boyen. Perov‐ skite-based hybrid solar cells exceeding 10% efficiency with high reproducibility us‐

[33] P-W. Liang, C-Y. Laio, C-C. Chueh, F. Zuo, S. T. Williams, X-K. Xin, J. Lin, and A. K-Y. Jen. Additive enhanced crystallization of solution-processed perovskite for highly efficiency planar-heterojunction solar cells. *Advanced Materials* 26 (2014) 3748–3754.

[34] M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A. Gray-Weale, U. Bach, Y-B. Cheng, and L. Spiccia. A fast deposition-crystallization procedure for highly efficient lead iodide perovskite thin-film solar cells. *Angewandte Chemie* 126

[35] N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. Ryu, and S. I. Seok. Solvent engineer‐ ing for high-performance inorganic–organic hybrid perovskite solar cells. *Nature Ma‐*

[36] F. Huang, Y. Dkhissi W. Huang, M. Xiao, I. Benesperi, S. Rubanov, Y. Zhu, X. Lin, L. Jiang, Y. Zhou, A. Gray-Weale, J. Etheridge, C. R. McNeill, R. A. Caruso, U. Bach, L. Spiccia, and Y-B. Cheng. Gas-assisted preparation of lead iodide perovskite films consisting of a monolayer of single crystalline grains for high efficiency planar solar

[37] J. W. Jung, S. T. Williams, and A. K-Y. Jen. Low-temperature processed high-per‐ formance flexible perovskite solar cells via rationally optimized solvent washing

[38] C. Bi, Y. Shao, Y. Yuan, Z. Xiao, C. Wang, Y. Gao, and J. Huang. Understanding the formation and evolution of interdiffusion grown organolead halide perovskite thin

films by thermal annealing. *Journal of Chemistry A* 2 (2014) 18508–18514.

ing a thin film sandwich approach. *Advance Materials* 26 (2014) 2041–2046.

(2014) 9339–9342.

442 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

(2015), 522–525.

(2014) 10056–10061.

*terials* 13 (2014) 897–903.

cells. *Nano Energy* 10 (2014) 10–18.

treatments. *RSC Advances* 4 (2014) 62971–62977.


## **Chapter 15**

## **Numerical Simulations on Perovskite Photovoltaic Devices**

## Bernabé Marí Soucase, Inmaculada Guaita Pradas and Krishna R. Adhikari

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61751

## **Abstract**

Organometal halide perovskites have recently attracted tremendous attention due to their potential for photovoltaic applications, and they are also considered as promising materials in light emitting devices. In particular, in the last years promising photovoltaic devices with efficiencies above 20% have already been prepared using organometal hal‐ ide perovskites as absorbent materials.

A planar heterojunction perovskite-based solar cell is made of three main layers sand‐ wiched between the two conducting electrodes. The standard design for a planar hetero‐ junction perovskite-based solar cell is: Back electrode/ Hole Transport Material (HTM)/ Perovskite absorber / Electron Transport Material (ETM) / Transparent electrode. For pla‐ nar heterojunction-based solar cells, recent efforts have revealed that increasing conduc‐ tivity of the hole transport materials by doping and optimizing charge collection by adjusting the absorber thickness could bring a positive impact on the efficiency. Electron transporting materials are also a crucial component in perovskite-based solar cells. The effect of different electron transporting materials in the final behaviour of the PV device can also be numerically simulated. Several PV parameters such as thicknesses of the ab‐ sorber, HTM and ETM, respectively can be optimized by simulation methods and subse‐ quently implemented by experimentalists. The hole mobility and acceptor concentration of the HTM, interface trap density and work-function of back contact metal have shown significant influence on the device performance. Even with these strong merits, enhance‐ ment of hole mobility and conductivity of HTM, stability of perovskite and TiO2 and re‐ placement of toxic lead are still crucial. Through suitable processing/synthesizing of the perovskite absorbers, best engineering the selective contact, and increasing conductivity of HTM and ETM will boost the stability as well as performance of the device.

This chapter presents a review of the evolution of perovskite materials from their discov‐ ery to their present significance as the main constituent of a new class of photovoltaic de‐ vices. We also evaluate the use of numerical simulation methods for determining the optimal configuration of perovskite-based solar cells and analyzing their optoelectronic behavior. The outcome of a simulation study on organometal halide perovskite focusing on the role of the different components of the solar cell using Solar Cell Capacitance Sim‐ ulator as a simulation tool are discussed. A photoconversion efficiency of 22.7%, VOC =

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1.12 V, JSC = 24.66 mA/cm2 and FF = 82.1% has been found after optimizing the different parameters involved. These results still are slightly higher than the experimental ones but the confluence between both are expected in the short time.

**Keywords:** Organometal halide perovskites, CH3N3PbI3, Hole Transport Materials, Elec‐ tron Transport Materials, photovoltaic solar cell, photocurrent, photoconversion efficien‐ cy

## **1. Introduction**

The world is highly mechanized, competitive, depends almost upon machinery, life style is being improving day by day consequently energy consumption rate is increasing unexpect‐ edly. To fulfill the current energy demand, the world depends, to date, upon fossil fuel based (coal, natural gas, and petroleum products) energy resource. At the mean time, its reservoirs are limited and getting depleted gradually, cost is highly influencing by socio-economic and political situations, and its use creates an adverse effect to the environment as well as to the human health. Thus, everyone should think its best alternative before it comes to an end. So, it is the high time to look at and develop the unexhausted, environment friendly green/clean energy and associated technologies. As a source of energy solar energy is free, unlimited, and available everywhere in this world. The best option, hereby, might be solar energy and solar cell technology which is the main purpose of this paper and discussed briefly below. However, to date the relatively low efficiency to cost ratio of photovoltaic solar cells limiting its use. Thus, increasing efficiency and decreasing cost of solar cell is the major challenge to the researchers, academicians and industrialists.

## **1.1. Solar cells**

In every step and activity in our life, such as: cooking a food, lighting a home and a street, opening a college and a hospital, running a vehicle and a factory etc we need energy. Thus energy is extremely crucial factor for better quality of life for one and all, employment creation and industrial as well as economic development in a country. Global population, their living style/standard, industries and hence the energy demand is increasing day by day where world energy generation capacity is nearly 18 TW [1] and otherwise almost remaining the same. The main resource of energy, worldwide, is fossil fuel based energy. Actually it is untenable and also contributes substantially to climate change and global warning. World nuclear energy report 2014 highlighted the fact that "the nuclear share in the world's power generation declined steadily from a historic peak of 17.6% in 1996 to 10.8% in 2013". In the world, 31 countries with 388 reactors are currently generating 333 GW. 67 reactors are under construction since July 2014 with a total capacity of 64 GW [2]. Nuclear energy data indicates that it would not meet the current energy demand and it is quite hazardous to the mankind too. The renewable resources; hydroelectricity, geothermal, wind, bio-fuels are limited. On the other hand solar energy is inexhaustible, accessible, nonhazardous and environment friendly too which can directly be changed into heat and electricity. In facts feasible capacity of generation of electricity from the sun light is 1000 times higher than the current world energy demand [1]. Moreover, production of photo-electricity on one day is sufficient for one year by using even a less efficient photovoltaic solar cell. So, it would be the best option to resolve the world future energy demand and crisis.

1.12 V, JSC = 24.66 mA/cm2

cy

**1. Introduction**

academicians and industrialists.

**1.1. Solar cells**

and FF = 82.1% has been found after optimizing the different

parameters involved. These results still are slightly higher than the experimental ones but

**Keywords:** Organometal halide perovskites, CH3N3PbI3, Hole Transport Materials, Elec‐ tron Transport Materials, photovoltaic solar cell, photocurrent, photoconversion efficien‐

The world is highly mechanized, competitive, depends almost upon machinery, life style is being improving day by day consequently energy consumption rate is increasing unexpect‐ edly. To fulfill the current energy demand, the world depends, to date, upon fossil fuel based (coal, natural gas, and petroleum products) energy resource. At the mean time, its reservoirs are limited and getting depleted gradually, cost is highly influencing by socio-economic and political situations, and its use creates an adverse effect to the environment as well as to the human health. Thus, everyone should think its best alternative before it comes to an end. So, it is the high time to look at and develop the unexhausted, environment friendly green/clean energy and associated technologies. As a source of energy solar energy is free, unlimited, and available everywhere in this world. The best option, hereby, might be solar energy and solar cell technology which is the main purpose of this paper and discussed briefly below. However, to date the relatively low efficiency to cost ratio of photovoltaic solar cells limiting its use. Thus, increasing efficiency and decreasing cost of solar cell is the major challenge to the researchers,

In every step and activity in our life, such as: cooking a food, lighting a home and a street, opening a college and a hospital, running a vehicle and a factory etc we need energy. Thus energy is extremely crucial factor for better quality of life for one and all, employment creation and industrial as well as economic development in a country. Global population, their living style/standard, industries and hence the energy demand is increasing day by day where world energy generation capacity is nearly 18 TW [1] and otherwise almost remaining the same. The main resource of energy, worldwide, is fossil fuel based energy. Actually it is untenable and also contributes substantially to climate change and global warning. World nuclear energy report 2014 highlighted the fact that "the nuclear share in the world's power generation declined steadily from a historic peak of 17.6% in 1996 to 10.8% in 2013". In the world, 31 countries with 388 reactors are currently generating 333 GW. 67 reactors are under construction since July 2014 with a total capacity of 64 GW [2]. Nuclear energy data indicates that it would not meet the current energy demand and it is quite hazardous to the mankind too. The renewable resources; hydroelectricity, geothermal, wind, bio-fuels are limited. On the other hand solar energy is inexhaustible, accessible, nonhazardous and environment friendly too which can directly be changed into heat and electricity. In facts feasible capacity of generation

the confluence between both are expected in the short time.

446 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Solar cell is usually called photovoltaic which literary means the conversion of light or photon into electricity. Edmund Becquerel discovered photovoltaic effect in 1839 by illuminating silver chloride in an acidic solution and measuring the current whereas this effect was observed in 1876 by W. G. Adams and R. E. Day for the first time in a solid: selenium [3, 4].

In 1883, Charles Fritts demonstrated thin film selenium cell and he had realized enormous potential of photovoltaic devices [3]. Albert Einstein discovered the photoelectric effect in1905, and awarded by the Nobel Prize in Physics 1921 for this theoretical work [5]. The next significant step forward came 34 years later, after Fritts, where the work was carried out with the copper-cuprous oxide by Grondahl in 1927 [6] and he described the outcome of the work as development of a rectifier and a photovoltaic cell.

Almost at the same time, silicon had drawn the attention of researchers as it was emerging for use as point contact rectifier. Russel S. Ohl, Bell laboratory engineer, developed a "diode" in 1939 by doping one side of silicon with electron donor and other side with acceptor material. After one year he had observed an electric voltage across the ends of the so formed "p-n junction diode" when light shone it. It was the first invention of the photovoltaic device based on silicon semiconductor thus; Ohl got a patent for his work and this effect. This fact revealed after their paper published in 1952 [7]. In the same laboratory in 1954 D. M. Chapin, C. S. Fuller, and G. L. Pearson announced the first "modern photovoltaic silicon cell" with the efficiency of 6% [8]. Vanguard I was the first satellite to use solar power launched by USA in 1958. After that Explorer III, Vanguard II and Sputnik-3 were launched with PV-powered system on board.

In order to design a solar cell and realize the role of different components we must understand basic steps or processes they are undertaken in the solar cell layers. Following are the processes that undergo to convert electromagnetic energy into electrical energy in a photovoltaic [9] where this phenomenon is known as photovoltaic action:


#### Source: [10, 11]

**Figure 1.** Global PV market by Technology

#### Source: [10, 11]

**Figure 2.** Worldwide growth of Photovoltaics

Since that time different solar cell technologies have been developed and achieved different efficiencies. Now NREL divided the solar cell technologies into 5 categories they are: Crystal‐ line silicon cells, single junction GaAs, multijunction cells, thin film technologies and emerging PV. Gradual development of solar cells and their best efficiency is illustrated in Figure 3 [12]. Leonid A. Kosyachenko categorized the solar cell technologies as so called first, second and third generation photovoltaics. Based on Kosyachenko classification, all types of silicon wafer based and GaAs solar cells are represented as the first generation photovoltaics. Amorphous silicon (a-Si) and non silicon based thin film such as CIGS, CdTe based solar cells are so-called second generation photovoltaics, relatively younger dye-sensitized solar cells (DSSCs), organic and quantum dots solar cells are the third generation photovoltaics [1], whereas hybrid solar cells are considered as fourth generation solar cells. There is always more space in third and fourth generation photovoltaic technology to identify and develop further new materials and solar cell devices to achieve a low cost and high efficiency.

**Figure 3.** Best research cell efficiencies [12]

Source: [10, 11]

Source: [10, 11]

**Figure 2.** Worldwide growth of Photovoltaics

**Figure 1.** Global PV market by Technology

448 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Silicon is indirect band gap (BG) material even though its BG is 1.1eV thus, silicon based solar cell consumes a huge material and energy, as a consequence, a low productivity, high labor cost and hence a very low efficiency to cost ratio even for mass production with feasible efficiency of 16-17% and 13-15% in case of single- and polycrystalline materials respectively [13]. Even with such demerits, since the date of first comercial manufacture of PV solar cells around in 60's decade of 20th century to date the silicon based solar cells have been extremly dominating the photovoltaic technology. However, in the last decade new generation of materials and devices (eg; dye sensitized, cadmium telluride and copper indium gallium selenide based solar cells) has spread the photovoltaics region.

It is illustrated by Figure 1. Figure 2, 3 and 4, inspiring gestures, show the worldwide orien‐ tation towards the solar cell technologies; hopefully it would be the first in near future. Organic sensitizers in dye sensitized solar cells (DSSCs) have low absorption coefficients and narrow absorption bands which limit the photovoltaic performance [15]. Liquid organic electrolyte in DSSCs is volatile and can easily be degraded by ultravoilet radiation, so stability is the another major issue in this technology [1].

**Figure 4.** Projected global growth of PV. [14] (Source: Growth of photovoltaics)

Cadmium, tellurium, indium and galium are rare elements and almost they are expensive too furthermore, Cd and Te are toxic as well as scattered elements [1, 16]. These facts about existing materials always look for the low cost, easy processability, low consumption of material, reproducible, environment friendly, non toxic and easy abundant absorbing material for the solar cell industries.

## **1.2. Perovskites**

The main concern in designing a photovoltaic solar cell is to maximize the efficiency to cost ratio i.e., to reduce the total cost, increase the efficiency and life time of PV module [17]. There are various factors they affect the efficiency of a solar cell. The first and foremost factor is the percentage of electromagnetic energy reaching the absorber which the solar cell converts into electricity. Secondly, the types of materials since different materials have different absorption coefficients and band gap and hence different maximum theoretical efficiency [18]. Third factor is the structure of the material or semiconductor used in the cell. Higher performance of solar cell is expected and usually observed in perfect crystalline structure with suitable dopants. Fourth is thickness of the absorbing material. Too thin and too thick; both are not suitable for good photovoltaic action. In the former one the thickness may not sufficient for diffusion of charge carriers whereas latter case increases the cost and reduces efficiency. Fifth factor is the amount of light reaching the absorbing material i.e., reflectance, transmittance and absorbance of the material. Sixth factor affecting the efficiency is the temperature since different materials have different response to temperature higher than room temperature.

## *1.2.1. Structure of perovskites*

materials and devices (eg; dye sensitized, cadmium telluride and copper indium gallium

It is illustrated by Figure 1. Figure 2, 3 and 4, inspiring gestures, show the worldwide orien‐ tation towards the solar cell technologies; hopefully it would be the first in near future. Organic sensitizers in dye sensitized solar cells (DSSCs) have low absorption coefficients and narrow absorption bands which limit the photovoltaic performance [15]. Liquid organic electrolyte in DSSCs is volatile and can easily be degraded by ultravoilet radiation, so stability is the another

selenide based solar cells) has spread the photovoltaics region.

450 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 4.** Projected global growth of PV. [14] (Source: Growth of photovoltaics)

Cadmium, tellurium, indium and galium are rare elements and almost they are expensive too furthermore, Cd and Te are toxic as well as scattered elements [1, 16]. These facts about existing materials always look for the low cost, easy processability, low consumption of material, reproducible, environment friendly, non toxic and easy abundant absorbing material for the

The main concern in designing a photovoltaic solar cell is to maximize the efficiency to cost ratio i.e., to reduce the total cost, increase the efficiency and life time of PV module [17]. There are various factors they affect the efficiency of a solar cell. The first and foremost factor is the percentage of electromagnetic energy reaching the absorber which the solar cell converts into electricity. Secondly, the types of materials since different materials have different absorption coefficients and band gap and hence different maximum theoretical efficiency [18]. Third factor

major issue in this technology [1].

solar cell industries.

**1.2. Perovskites**

In 1839, Gustav Rose discovered the mineral calcium titanate (CaTiO3) in Ural Mountain, Russia and gave the name Perovskite to honor the Russian mineralogist L. A. Perovski [19]. Any material that resembles the crystal structure of mineral CaTiO3 is termed as perovskite structure or simply perovskite. It is generally represented by the formula ABX3 where A is a big-sized cation (either organic or inorganic), B is divalent small metallic cation (Cu2+, Mg2+, Ge2+, Sn2+, Pb2+, Eu2+, Yb2+, etc), and X is halide ion (Cl- , Br and I- ) which binds to both cations [20]. The perovskites can be divided into two main categories: alkali halide perovskite and organo-metal halide perovskite.

Perovskites possess different astonishing optoelectronic behaviors which make the perovskite as promising candidate of photovoltaic absorber. Ferroelectric behavior, discovered half a century ago, is one of the characteristics among them.

In 1979, Salau reported the potassium lead iodide as a direct band gap material with the value 1.4 to 2.2 eV, which suits the solar spectrum [21, 22]. Capability of halide perovskite to convert light to electricity was discovered in 1990s and fabricated for LED.

α, β, γ, & δ are four possible phases of perovskite where α is high temperature phase T > 327 K and has cubic structure (eg CsSnI3). This structure allows only one formula unit per unit cell, so CH3NH3 + cannot obtain cubic structure. For a temperature T < 327 K perovskite changes from α to β phase usually found in tetragonal structure with lattice parameters a = 8.855 Å and c = 12.659 Å where exact values depend on molecular orientation [23].

Generally, ABX3 has cubic structure and α phase where B has 6 nearest neighbor X ions (octahedral) and A has twelve fold coordination sites as shown in Figure 5.

One crucial parameter to maintain the cubic structure of perovskite is the tolerance factor, t = (RA+RX)/√2 (RB+RX). It should be close to one to obtain in the cubic structure where, R is radius of the ions and suffixes A, B and X are as defined above. Hendon et al. mentioned that stable perovskite is found in the range 0.7< T < 1 [24], which guides cation A must be larger than B in this regard, CH3NH3 ion is one of the best options. Smaller t could be lead to lower symmetry tetragonal β phase or orthorhombic γ phase. For MAPbI3 perovskite, transition from α to β to γ occurs at 327 and 1600 K respectively. The transition of perovskite is depending on the tilting and rotation of the BX6 polyhedra in the lattice. Whereas the fourth phase δ is non perovskite phase [23, 24, 25, 26, 27].

Source: Perovskite http://www.surrey.ac.uk/chemistry/perovskite/what/history/index.htm

**Figure 5.** Cubic structure of MAPbI3 perovskite

## *1.2.2. Electronic Behavior*

Pb has an occupied 6s orbital, which is below the top of valence bands of the perovskite. This lone pair of s electrons often gives rise to unusual behaviors in perovskite [28]. Unlike GaAs and CdTe, in the first principle study carried out by Walsh et al., 2011 density of states (DOS) and partial charge density plots of halides perovskite showed the coupling between Pb s and I p (anti-bonding state) contribute to VBM but CBM is derived almost from the Pb p state [28, 29]. And perovskite gets ionic and covalent, dual nature in electronic structures. Thus organic part/ion does not play a direct role to determine electronic behaviors but takes part in stabi‐ lizing perovskite structure and changing the lattice constants.

Experimental/research works and DFT-PBE calculations show the different values of band gap for perovskites. Band gap of perovskite depends on synthesizing process and the size of organic/inorganic cation, metallic ion and very less in halide ion. Band gap of CH3NH3PbI3, CH3NH3PbBr3, CH3NH3PbCl3, CH3NH3Pb3-xClx are 1.49-1.61, 1.95 eV, 2.46 eV and 1.59 eV respectively [20, 30] whereas mostly used band gap for CH3NH3PbI3 is 1.5 eV [27].

## *1.2.3. Ambipolar conductivity*

The electronic structure of MAPbI3 perovskite is different compared to conventional semicon‐ ductor. A cation Pb p orbital has a much higher energy level than anion p orbital as in p-s semiconductor and hence CBM of MAPbI3 is more dispersive. At the meantime VBM is also dispersive due to strong s-p coupling. Based upon the formula *<sup>m</sup>* \* <sup>=</sup>*<sup>h</sup>* <sup>2</sup> <sup>∂</sup><sup>2</sup> *<sup>ε</sup>*(*k*) ∂<sup>2</sup> *k* −1 [31] effective

mass of electron is balanced by that of hole in MAPbI3 perovskites which finally results into an ambipolar charge transport behavior in perovskite based solar cells.

The electronic structure of MAPbI3 perovskite is different i. e., inverted compared to conventional semiconductor. A cation Pb p orbital has a much higher energy level than anion p orbital as in p-s semiconductor and hence CBM of MAPbI3 is more dispersive. At the meantime VBM is also dispersive due to strong s-p coupling. Based upon

Experimental/research works and DFT-PBE calculations show the different values of band gap for perovskites. Band gap of perovskite depends on synthesizing process and the size of organic/inorganic cation, metallic ion and very less in halide ion. Band gap of CH3NH3PbI3, CH3NH3PbBr3, CH3NH3PbCl3, CH3NH3Pb3-xClx are 1.49-1.61, 1.95 eV, 2.46 eV and 1.59 eV respectively [20, 30] whereas mostly used band gap for CH3NH3PbI3 is 1.5 eV [27].

#### *1.2.4. Optical properties* perovskites which finally results into an ambipolar charge transport behavior in perovskite based solar cells.

*1.2.2. Electronic Behavior*

**Figure 5.** Cubic structure of MAPbI3 perovskite

*1.2.3. Ambipolar conductivity*

Pb has an occupied 6s orbital, which is below the top of valence bands of the perovskite. This lone pair of s electrons often gives rise to unusual behaviors in perovskite [28]. Unlike GaAs and CdTe, in the first principle study carried out by Walsh et al., 2011 density of states (DOS) and partial charge density plots of halides perovskite showed the coupling between Pb s and I p (anti-bonding state) contribute to VBM but CBM is derived almost from the Pb p state [28, 29]. And perovskite gets ionic and covalent, dual nature in electronic structures. Thus organic part/ion does not play a direct role to determine electronic behaviors but takes part in stabi‐

Experimental/research works and DFT-PBE calculations show the different values of band gap for perovskites. Band gap of perovskite depends on synthesizing process and the size of organic/inorganic cation, metallic ion and very less in halide ion. Band gap of CH3NH3PbI3, CH3NH3PbBr3, CH3NH3PbCl3, CH3NH3Pb3-xClx are 1.49-1.61, 1.95 eV, 2.46 eV and 1.59 eV

The electronic structure of MAPbI3 perovskite is different compared to conventional semicon‐ ductor. A cation Pb p orbital has a much higher energy level than anion p orbital as in p-s semiconductor and hence CBM of MAPbI3 is more dispersive. At the meantime VBM is also

mass of electron is balanced by that of hole in MAPbI3 perovskites which finally results into

∂<sup>2</sup> *k*

−1

[31] effective

respectively [20, 30] whereas mostly used band gap for CH3NH3PbI3 is 1.5 eV [27].

dispersive due to strong s-p coupling. Based upon the formula *<sup>m</sup>* \* <sup>=</sup>*<sup>h</sup>* <sup>2</sup> <sup>∂</sup><sup>2</sup> *<sup>ε</sup>*(*k*)

an ambipolar charge transport behavior in perovskite based solar cells.

lizing perovskite structure and changing the lattice constants.

Source: Perovskite http://www.surrey.ac.uk/chemistry/perovskite/what/history/index.htm

452 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Absorber is the key player in photovoltaic solar cells to achieve good performance. In this context, optical absorption strength and range are the crucial factor of the materials. The edge transition for MAPbI3 perovskite comes from mixed s-p coupling and Pb p orbital so that transition probability from Pb s to Pb p is high [27]. Moreover, perovskite is direct band gap material and hence it has high optical absorption strength and wider range to absorb sufficient solar energy to achieve high value of power conversion efficiency. Figure 6 shows the absorb‐ ance spectra for three MAPbX3 perovskites with different halide components; X=I, Br and Cl. Notice that the band gap changes with the selection of halide whereas MAPBI3 has the best band gap for photovoltaic applications. 1.2.4. Optical properties Absorber is the key player in photovoltaic solar cells to achieve good performance. In this context, optical absorption strength and range are the crucial factor of the materials. The edge transition for MAPbI3 perovskite comes from mixed s-p coupling and Pb p orbital so that transition probability from Pb s to Pb p is high [27]. Moreover, perovskite is direct band gap material and hence it has high optical absorption strength and wider range to absorb sufficient solar energy to achieve high value of power conversion efficiency. Figure 6 shows the absorbance spectra for three MAPbX3 perovskites with different halide components; X=I, Br and Cl. Notice that the band gap changes with the selection of halide whereas MAPBI3 has the best band gap for photovoltaic applications.

1.2.3. Ambipolar conductivity

the formula <sup>1</sup>

<sup>m</sup> <sup>h</sup> <sup>ε</sup>

2

k

2 <sup>2</sup> ] )( \* [ <sup>−</sup> ∂ <sup>∂</sup> <sup>=</sup> <sup>k</sup>

Figure 6. Absorbance for different organometal halide perovskites MAPbI3, MAPbBr3 and MAPbCl3. **Figure 6.** Absorbance for different organometal halide perovskites MAPbI3, MAPbBr3 and MAPbCl3.

The performance of the photovoltaic solar cells mainly depends upon absorption value, and thickness of the absorption layer. Plot in Figure 7 shows the variation of performance of the solar cell with thickness of the perovskite. Usually performance of solar cell increases with increase in thickness of the perovskite. Furthermore, the plot shows that thin layer of MAPbI3 layer yields high fill factor (FF). The performance of the photovoltaic solar cells mainly depends upon absorption value, and thickness of the absorption layer. Plot in Figure 7 shows the variation of performance of the solar cell with thickness of the perovskite. Usually performance of solar cell increases with increase in thickness of the perovskite. Furthermore, the plot shows that thin layer of MAP‐ bI3 layer yields high fill factor (FF).

## *1.2.5. Defects in perovskites: intrinsic/point defect*

Efficiency of a solar cell is highly affected by defects such as point/intrinsic and grain boun‐ daries, crystalline structure and amount of doping of an absorbing material when processed via low-cost method. Electron and hole diffusion length and open circuit voltage (VOC) of a solar cell is greatly influenced by the point (Schottky and Frenkel) defects [27]. The defect densities of perovskite depend on the formation energy and hence chemical potential, it corresponds to precursors, partial pressure and temperature. The defects with low formation

Figure 7. Variation of FF and PCE with thickness of perovskite **Figure 7.** Variation of FF and PCE with thickness of perovskite

energy create only shallow level which results the long electron-hole diffusion length and high VOC. On the other hand defects with deep levels have high formation energy it results into unpleasant effect on electron-hole diffusion length and VOC. Experimental and simulation results of the long electron-hole diffusion length and high VOC conforms the unusual shallow defect levels [27]. 1.2.5. Defects in perovskites: intrinsic/point defect Efficiency of a solar cell is highly affected by defects such as point/intrinsic and grain boundaries, crystalline structure and amount of doping of an absorbing material when processed via low-cost method. Electron and hole diffusion length and open circuit voltage (VOC) of a solar cell is greatly influenced by the point (Schottky and Frenkel) defects [27]. The defect densities of perovskite depend on the formation energy and hence chemical

In CIGS and CdTe, p-type doping is easier in equilibrium and n-type doping is rather difficult due to self-compensation. But in MAPbI3 the formation energy of methylammonium (MA) interstitial defect (donor like) and iodine vacancy defect (acceptor like) have similar values that make both; p-type and n-type doping possible. Unlike other researchers, Agiorgousis et al. and Baumann et al. suggested and pointed out the strong covalent bonds and deep level defects by using first principles calculations [32, 33]. Despite that the presence of defects and traps in perovskite and particulars of their impact are so far under discussion and investigation. potential, it corresponds to precursors, partial pressure and temperature. The defects with low formation energy create only shallow level which results the long electron-hole diffusion length and high VOC. On the other hand defects with deep levels have high formation energy it results into unpleasant effect on electron-hole diffusion length and VOC. Experimental and simulation results of the long electron-hole diffusion length and high VOC conforms the unusual shallow defect levels [27]. In CIGS and CdTe, p-type doping is easier in equilibrium and n-type doping is rather difficult due to selfcompensation. But in MAPbI3 the formation energy of methylammonium (MA) interstitial defect (donor like) and

#### *1.2.6. Progress in perovskite solar cells* iodine vacancy defect (acceptor like) have similar values that make both; p-type and n-type doping possible. Unlike other researchers, Agiorgousis et al. and Baumann et al. suggested and pointed out the strong covalent

In 2009, Miyasaka et al. opened up first perovskite solar cell based on mesoporous TiO2 photoanode and observed the power conversion efficiency (PCE) 3.81% and 3.13 % for MAPbI3 and MAPbBr3 respectively along with their poor cell stability [15]. bonds and deep level defects by using first principles calculations [32, 33]. Despite that the presence of defects and traps in perovskite and particulars of their impact are so far under discussion and investigation. 1.2.6. Progress in perovskite solar cells

In 2012, Kanatzidis and his coworkers synthesized alkali metal perovskite (floride doped CsSnI3) as p-type Hole Transporting Material (HTM) in dye sensitized solar cells where 10.2% PCE was reported. At the mean time, Michael Gratzel & coworkers with N. G. Park used MAPbI3 as light absorber with SpiroMETAD on mesoporous TiO2 where measured efficiency was 9.7%. Snaith in collaboration with Miyasaka boosted PCE and VOC to 10.9 % and 1.13 V respectively by replacing the n type mesoporous titanium oxide by an inert Al2O3 scaffold [34] in the consequence of faster diffusion of electron through the perovskite. In 2009, Miyasaka et al. opened up first perovskite solar cell based on mesoporous TiO2 photo-anode and observed the power conversion efficiency (PCE) 3.81% and 3.13 % for MAPbI3 and MAPbBr3 respectively along with their poor cell stability [15]. In 2012, Kanatzidis and his coworkers synthesized alkali metal perovskite (floride doped CsSnI3) as p-type Hole Transporting Material (HTM) in dye sensitized solar cells where 10.2% PCE was reported. At the mean time, Michael Gratzel & coworkers with N. G. Park used MAPbI3 as light absorber with SpiroMETAD on mesoporous

Liu and Snaith et al. in 2013 further boosted the efficiency to 15.4% via architecturing the vapor deposition heterojunction solar cell without electron conduction scaffold [35]. Since that time TiO2 where measured efficiency was 9.7%. Snaith in collaboration with Miyasaka boosted PCE and VOC to 10.9 % and 1.13 V respectively by replacing the n type mesoporous titanium oxide by an inert Al2O3 scaffold [34] in the consequence of faster diffusion of electron through the perovskite.

> Liu and Snaith et al. in 2013 further boosted the efficiency to 15.4% via architecturing the vapor deposition heterojunction solar cell without electron conduction scaffold [35]. Since that time to about end of 2014, in reference [36 - 40] reported 15.6%, 15.9%, 16.7%, 19.3% and 20.1% PCE respectively. Astonishingly it is a great improvement in the efficiency for perovskite based solar cells, as shown in Figure 3, also makes the perovskite as a premising candidate for immediate future PV solar cells. The main aim of photovoltaic design is to optimize efficiency to cost ratio. In this context, among the perovskite family, MAPbI3 is one who proved itself as a best and champaign material due to its favorable opto-electric behavior, long lifetime, low temperature solution processability, ferroelectricity and hence a superb photovoltaic performance. There are lots of space to further

to about end of 2014, in reference [36 - 40] reported 15.6%, 15.9%, 16.7%, 19.3% and 20.1% PCE respectively. Astonishingly it is a great improvement in the efficiency for perovskite based solar cells, as shown in Figure 3, also makes the perovskite as a premising candidate for immediate future PV solar cells. The main aim of photovoltaic design is to optimize efficiency to cost ratio. In this context, among the perovskite family, MAPbI3 is one who proved itself as a best and champaign material due to its favorable opto-electric behavior, long lifetime, low temperature solution processability, ferroelectricity and hence a superb photovoltaic perform‐ ance. There are lots of space to further improve the efficiency and deep insight into excellent optoelectronic behavior, thermodynamic stability of the perovskite absorber and the formation mechanism of the dominant intrinsic defects.

## **2. Numerical Simulations**

energy create only shallow level which results the long electron-hole diffusion length and high VOC. On the other hand defects with deep levels have high formation energy it results into unpleasant effect on electron-hole diffusion length and VOC. Experimental and simulation results of the long electron-hole diffusion length and high VOC conforms the unusual shallow

1.2.5. Defects in perovskites: intrinsic/point defect

Efficiency of a solar cell is highly affected by defects such as point/intrinsic and grain boundaries, crystalline

diffusion length and open circuit voltage (VOC) of a solar cell is greatly influenced by the point (Schottky and Frenkel) defects [27]. The defect densities of perovskite depend on the formation energy and hence chemical

create only shallow level which results the long electron-hole diffusion length and high VOC. On the other hand defects with deep levels have high formation energy it results into unpleasant effect on electron-hole diffusion length and VOC. Experimental and simulation results of the long electron-hole diffusion length and high VOC

In CIGS and CdTe, p-type doping is easier in equilibrium and n-type doping is rather difficult due to self-

and traps in perovskite and particulars of their impact are so far under discussion and investigation.

In 2009, Miyasaka et al. opened up first perovskite solar cell based on mesoporous TiO2 photo-anode and observed the power conversion efficiency (PCE) 3.81% and 3.13 % for MAPbI3 and MAPbBr3 respectively along

Transporting Material (HTM) in dye sensitized solar cells where 10.2% PCE was reported. At the mean time,

and 1.13 V respectively by replacing the n type mesoporous titanium oxide by an inert Al2O3 scaffold [34] in the

Liu and Snaith et al. in 2013 further boosted the efficiency to 15.4% via architecturing the vapor deposition heterojunction solar cell without electron conduction scaffold [35]. Since that time to about end of 2014, in reference [36 - 40] reported 15.6%, 15.9%, 16.7%, 19.3% and 20.1% PCE respectively. Astonishingly it is a great improvement in the efficiency for perovskite based solar cells, as shown in Figure 3, also makes the perovskite as a premising candidate for immediate future PV solar cells. The main aim of photovoltaic design is to optimize efficiency to cost ratio. In this context, among the perovskite family, MAPbI3 is one who proved itself as a best and champaign material due to its favorable opto-electric behavior, long lifetime, low temperature solution processability, ferroelectricity and hence a superb photovoltaic performance. There are lots of space to further

iodine vacancy defect (acceptor like) have similar values that make both; p-type and n-type doping possible. Unlike other researchers, Agiorgousis et al. and Baumann et al. suggested and pointed out the strong covalent bonds and deep level defects by using first principles calculations [32, 33]. Despite that the presence of defects

Figure 7. Variation of FF and PCE with thickness of perovskite

0 200 400 600 800 1000

Thickness (nm)

In CIGS and CdTe, p-type doping is easier in equilibrium and n-type doping is rather difficult due to self-compensation. But in MAPbI3 the formation energy of methylammonium (MA) interstitial defect (donor like) and iodine vacancy defect (acceptor like) have similar values that make both; p-type and n-type doping possible. Unlike other researchers, Agiorgousis et al. and Baumann et al. suggested and pointed out the strong covalent bonds and deep level defects by using first principles calculations [32, 33]. Despite that the presence of defects and traps in perovskite and particulars of their impact are so far under discussion and investigation.

conforms the unusual shallow defect levels [27].

1.2.6. Progress in perovskite solar cells

In 2009, Miyasaka et al. opened up first perovskite solar cell based on mesoporous TiO2 photoanode and observed the power conversion efficiency (PCE) 3.81% and 3.13 % for MAPbI3 and

In 2012, Kanatzidis and his coworkers synthesized alkali metal perovskite (floride doped CsSnI3) as p-type Hole Transporting Material (HTM) in dye sensitized solar cells where 10.2% PCE was reported. At the mean time, Michael Gratzel & coworkers with N. G. Park used MAPbI3 as light absorber with SpiroMETAD on mesoporous TiO2 where measured efficiency was 9.7%. Snaith in collaboration with Miyasaka boosted PCE and VOC to 10.9 % and 1.13 V respectively by replacing the n type mesoporous titanium oxide by an inert Al2O3 scaffold [34]

Liu and Snaith et al. in 2013 further boosted the efficiency to 15.4% via architecturing the vapor deposition heterojunction solar cell without electron conduction scaffold [35]. Since that time

consequence of faster diffusion of electron through the perovskite.

defect levels [27].

*1.2.6. Progress in perovskite solar cells*

80

**Figure 7.** Variation of FF and PCE with thickness of perovskite

81

FF (%)

PCE (%)

82

454 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

MAPbBr3 respectively along with their poor cell stability [15].

with their poor cell stability [15].

in the consequence of faster diffusion of electron through the perovskite.

structure and amount of doping of an absorbing material when processed via low-cost method. Electron and hole potential, it corresponds to precursors, partial pressure and temperature. The defects with low formation energy compensation. But in MAPbI3 the formation energy of methylammonium (MA) interstitial defect (donor like) and Simulation is a crucial technique to realize deep insight into the physical operation, viability of proposed physical explanation and effect of physical changes on performance of the solar cell devices. There are various simulation models (SCAPS, AMPS, SCAP, etc) for solar cells simulation. SCAPS (Solar Cell Capacitance Simulator) is one-dimensional simulation program with seven semiconductor input layers developed by a goup of solar cell researcher at the department of Electronics and Information System, University of Gent, Belgium [41]. It is impractical as well as wastage of time and money to design a solar cell without simulation works. It minimizes not only the risk, time and money rather analyzes layers properties and role to optimize the solar cell performance. In order to simulation a device all the basic input parameters should be well defined so that it behaves as a real counterpart. The perovskitebased solar cells have employed a similar structure with inorganic semiconductor solar cells, such as CIGS, and Wannier-type exciton in the perovskite is found. Thus SCAPS like 1D simulator can be employed to simulate the perovskite based solar cells [42].

> The main features of the latest version of SCAPS, to address the basic parameters, are as follows [41]:



**Figure 8.** SCAPS-Solar cell definition panel.

**Figure 9.** SCAPS-Solar cell definition panel with simulation



**Figure 10.** SCAPS-Solar cell simulations with sample data and defects.

**Figure 9.** SCAPS-Solar cell definition panel with simulation

**Figure 8.** SCAPS-Solar cell definition panel.

456 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications


In this study, perovskite based solar cell is simulated with three input layers; where two layers are p-type one is Hole Transporting Material and Electron Blocked Layer (say HTM), another low doped organic lead halide perovskite is used as active/absobing layer. N-type transparent conducting oxide can be used as the Electron Transporting Material (ETM) which blocks the holes. Thicknesses of HTM/MAPbI3/ETM layers are 400/450/90 nm, respectively, if not stated there. Furthermore, front and back contact are used to collect and transport the charge carriers effectively to the external circuit. The simulations were carried out under 1 Sun using differnt material to optimize the solar cell devices and to realize the role of individual layer.

## **3. Solar cells physics**

The basic physical principles [9, 41, 43] of a solar cell device are given below:

## **3.1. Absorbance and absorption coefficient**

When an absorbing material of thickness d is being illuminated by an electromagnetic energy (usually light) of wavelength λ and intensity I (λ), some of light is reflected (R), some is absorbed (A) and remaining is transmitted (T) through the material. Now absorbance [Aabs (λ)] and absorption coefficient [α (λ)] can be defined as

$$A\_{\rm abs}(\lambda) = -\log\_{10} \left[ \frac{T(\lambda)}{I(\lambda) - R(\lambda)} \right] \tag{1}$$

$$\alpha(\lambda) = \frac{\mathbf{A}\_{\text{abs}}(\lambda)}{d} \tag{2}$$

### **3.2. Density of States**

The valence band (NV) and conduction band (NC) effective densities of states are temperature dependent material properties which are given by formulae

Numerical Simulations on Perovskite Photovoltaic Devices http://dx.doi.org/10.5772/61751 459

$$N\_V = \frac{(m\_p^\*)^{\frac{1}{2}}\sqrt{2}}{\pi^2 \hbar^3} \sqrt{E - E\_V} \tag{3}$$

$$N\_{\mathbb{C}} = \frac{(m\_u^\*)^{\frac{1}{2}}\sqrt{2}}{\pi^2 \hbar^3} \sqrt{E - E\_{\mathbb{C}}} \tag{4}$$

where *mp* \* and *mn* \* are effective mass of hole and electron, E, EV and EC are energy levels at steady state, valence band edge and conduction band edge respectively.

## **3.3. Carrier concentration**

**•** batch calculations possible; presentation of results and settings as a function of batch

**•** loading and saving of all settings; startup of SCAPS in a personalized configuration; a script

**•** a script language facility to run SCAPS from a "script file"; all internal variables can be

In this study, perovskite based solar cell is simulated with three input layers; where two layers are p-type one is Hole Transporting Material and Electron Blocked Layer (say HTM), another low doped organic lead halide perovskite is used as active/absobing layer. N-type transparent conducting oxide can be used as the Electron Transporting Material (ETM) which blocks the holes. Thicknesses of HTM/MAPbI3/ETM layers are 400/450/90 nm, respectively, if not stated there. Furthermore, front and back contact are used to collect and transport the charge carriers effectively to the external circuit. The simulations were carried out under 1 Sun using differnt

When an absorbing material of thickness d is being illuminated by an electromagnetic energy (usually light) of wavelength λ and intensity I (λ), some of light is reflected (R), some is absorbed (A) and remaining is transmitted (T) through the material. Now absorbance [Aabs

> l

> > l

*I R* (1)

<sup>=</sup> A () abs ( ) *<sup>d</sup>* (2)

l

l

The valence band (NV) and conduction band (NC) effective densities of states are temperature

 = - - <sup>10</sup> ( ) ( ) log [ ] () () *abs <sup>T</sup> <sup>A</sup>*

material to optimize the solar cell devices and to realize the role of individual layer.

The basic physical principles [9, 41, 43] of a solar cell device are given below:

parameters,

language including a free user function,

458 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**•** a panel for the interpretation of admittance measurements.

accessed and plotted via the script,

**•** very intuitive user interface,

**•** a built-in curve fitting facility,

**3. Solar cells physics**

**3.2. Density of States**

**3.1. Absorbance and absorption coefficient**

(λ)] and absorption coefficient [α (λ)] can be defined as

l

a l

dependent material properties which are given by formulae

In thermal equilibrium, number of free electrons and holes charge densities; n and p respec‐ tively are given by

$$m = N\_{\odot} \exp(\frac{E\_{\mathrm{F}} - E\_{\odot}}{kT\_{\mathrm{u}}}) = N\_{\odot} \exp(\frac{V\_{\mathrm{u}}}{kT\_{\mathrm{u}}}) \tag{5}$$

$$p = N\_v \exp(\frac{E\_V - E\_F}{kT\_p}) = N\_c \exp(\frac{V\_p}{kT\_p}) \tag{6}$$

Where EF if Fermi level and V is built in potential, k Boltzmann constant and T is spatially varying electron and hole temperature.

### **3.4. Poisson's equation**

Poisson's equation defines the electric field (E) that is modified as a result of the current flowing and the charge in the delocalized states, traps and recombination centers in the device as given below

$$\frac{\partial}{\partial \mathbf{x}} (\boldsymbol{\varepsilon} \frac{\partial \boldsymbol{\Psi}}{\partial \mathbf{x}}) = -q[\boldsymbol{p} - \boldsymbol{n} + \mathbf{N}\_D^+ - \mathbf{N}\_A^- + \frac{\rho\_{\mathrm{def}}}{q}] \tag{7}$$

where, ψ is electrostatic potential, ε is dielectric constant and q is electronic charge.

The first two terms in right side are free charge carriers per volume, third and fourth are ionized donor and acceptor-like dopants i.e,. localized states and ρdef is defect charge density.

## **3.5. Formula for p-n junctions diode**

Solar cell behaves as a p-n junction diode. The voltage and current in ideal diode are given by formulae as below:

$$V\_0 = \frac{kT}{q} \ln(\frac{N\_A N\_D}{n\_i^2}) \tag{8}$$

where ni is intrinsic carriers concentrations.

$$I = I\_0 \text{[exp(}\frac{qT}{kT}) - 1\text{]} \tag{9}$$

where I0 is saturation current.

## **3.6. Solar Cell Formulae**

$$V\_{\rm OC} = \frac{nkT}{q} \ln(\frac{I\_{\rm SC}}{I\_0} + 1) \tag{10}$$

$$J = J\_0 \left[ \exp(\frac{qT}{nkT}) - 1 \right] - J\_{\text{SC}} \tag{11}$$

where, VOC, JSC and J0 are open circuit voltage, short circuit current and saturation current density respectively.

### **3.7. Electron and hole Diffusion lengths**

Collection length, a distance over which light absorption caused excitaions can be transported, is limited by diffusion, drift, or combination of the two. The electron and hole diffusion length are given by

$$\mathbf{L}\_{\boldsymbol{u}}^{D\boldsymbol{\beta}\boldsymbol{\mathcal{Y}}} = \mathbf{[}\mathbf{D}\_{\boldsymbol{u}}\boldsymbol{\tau}\_{\boldsymbol{u}}\mathbf{I}^{\frac{1}{2}}\tag{12}$$

elecron diffusion length.

$$\mathbf{L}\_p^{Dif} = \mathbf{[}\mathbf{D}\_p \boldsymbol{\tau}\_p\text{]}^{\frac{1}{2}} \tag{13}$$

hole diffusion length.

**3.5. Formula for p-n junctions diode**

where I0 is saturation current.

**3.6. Solar Cell Formulae**

density respectively.

elecron diffusion length.

are given by

**3.7. Electron and hole Diffusion lengths**

is intrinsic carriers concentrations.

460 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

formulae as below:

where ni

Solar cell behaves as a p-n junction diode. The voltage and current in ideal diode are given by

=<sup>0</sup> <sup>2</sup> ln( ) *A D*

<sup>=</sup> - <sup>0</sup>[exp( ) 1] *qT I I*

= +

= -- <sup>0</sup>[exp( ) 1] *SC*

where, VOC, JSC and J0 are open circuit voltage, short circuit current and saturation current

Collection length, a distance over which light absorption caused excitaions can be transported, is limited by diffusion, drift, or combination of the two. The electron and hole diffusion length

t

1

=

<sup>2</sup> [ ] *Diff*

*nkT <sup>I</sup> <sup>V</sup>*

*OC*

0 ln( 1) *SC*

*kT N N <sup>V</sup>*

*i*

*<sup>q</sup> <sup>n</sup>* (8)

*kT* (9)

*q I* (10)

*qT J J <sup>J</sup> nkT* (11)

*n nn L D* (12)

where *Dn = kTnμn* is electron diffusion coefficient, *Dp = kTpμp* hole diffusion coefficient, *τ* s life time and *μ*s mobility of electron and hole.

## **3.8. Transport equation**

Mobility of free electrons, holes, variation of carrier concentrations and hence quasi Fermi level with position result electron and hole currents. Thus, in thermal equilibrium

$$J\_n = -\frac{\mu\_n n}{q} \frac{\partial E\_{Fu}}{\partial \mathbf{x}}\tag{14}$$

$$J\_p = +\frac{\mu\_p p}{q} \frac{\partial E\_{\text{Fp}}}{\partial \mathbf{x}} \tag{15}$$

where μn and μp are electron and hole mobilities, and n and p are free electrons and holes density, EFn and EFp are electron and hole quasi-Fermi level respectively.

## **3.9. Continuity equations**

In the absence of illumination and recombinations, in steady state, the free electron current density at any point in a solar cell device must be equal to that at some other piont. Same is valid for holes too. In a device generation and recombinations both are undertaken. Thus, the conservation of free electrons and free holes in the device is expressed as continuty equations

$$\frac{\partial \mathfrak{N}}{\partial t} = -\frac{\partial \mathfrak{J}\_n}{\partial \mathfrak{x}} + \mathbf{G}^- - \mathfrak{R}\_n \tag{16}$$

$$\frac{\partial p}{\partial t} = -\frac{\partial I\_p}{\partial \mathbf{x}} + \mathbf{G} - \mathfrak{R}\_p \tag{17}$$

where Gs are photo generation rates and ℜ s are recombination rates in the device.

### **3.10. Recombination**

Shockley-Read-Hall (SRH), Radiative and Auger recombinations are the possible recombina‐ tion mechanisms in solar cells, they are given as

$$\mathfrak{R}^L = \frac{\upsilon \sigma\_n \sigma\_p N\_T [\upsilon p - n\_i^2]}{\sigma\_p [p + p\_1] + \sigma\_n [n + n\_1]} \tag{18}$$

SRH recombination

$$\mathfrak{R}^{R} = \mathbb{I}\frac{\mathcal{G}\_{th}^{R}}{n\_{\parallel}^{2}} \| \left(pn - n\_{\parallel}^{2}\right) \tag{19}$$

Radiative recombination

$$\mathfrak{M}\_{\boldsymbol{n}}^{A} = \mathbb{I}\frac{\boldsymbol{n} - \boldsymbol{n}\_{0}}{\boldsymbol{\pi}\_{\boldsymbol{n}}^{A}}\mathbb{I} \tag{20}$$

$$\mathfrak{R}\_p^A = \mathbb{I}\frac{p - p\_0}{\mathfrak{r}\_p^A} \mathbb{I} \tag{21}$$

(Eqs. 20 and 21 are Auger recombinations)

where σs are capture cross-sections for electrons and holes, v electron thermal velocity, NT number of gap states per volume, ni intrinsic number density, gthR gives the number of electrons in conduction band and holes in valence band generated per unit time per unit volume and τn <sup>A</sup> & τ<sup>p</sup> A are electron and hole lifetimes.

### **3.11. Power conversion efficiency**

The fraction of incident power converted into usefull electricity by the solar cells is termed as Power conversion efficiency (η). It is given as below:

$$\eta = \frac{P\_{\text{out}}}{P\_{\text{in}}} = \frac{FF \times V\_{\text{OC}} J\_{\text{SC}}}{P\_{\text{in}}} \tag{22}$$

$$\text{Fill Factor } FF = \frac{V\_{mp} J\_{mp}}{V\_{\infty} J\_{SC}} \tag{23}$$

where Vmp and Jmp are voltage and current at maximum power point respectively.

## **3.12. Work function of contacts**

s s


*L np T i p n*

Â= - <sup>2</sup> <sup>2</sup> [ ]( ) *R R th*

*<sup>g</sup> pn n*

t

t

*A p*

where σs are capture cross-sections for electrons and holes, v electron thermal velocity, NT number of gap states per volume, ni intrinsic number density, gthR gives the number of electrons in conduction band and holes in valence band generated per unit time per unit volume and

The fraction of incident power converted into usefull electricity by the solar cells is termed as

 ´ = = *out OC SC in in P FF V J*

Fill Factor = *mp mp*

where Vmp and Jmp are voltage and current at maximum power point respectively.

*V J FF*

*OC SC*


*A n n n*


*i*

 s

[ ][ ]

1 1

*i*

*<sup>n</sup>* (19)

*p p* (21)

*P P* (22)

*V J* (23)

[ ]

2

(18)

(20)

++ +

*v N np n pp nn*

s

462 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

SRH recombination

Radiative recombination

τn <sup>A</sup> & τ<sup>p</sup>

(Eqs. 20 and 21 are Auger recombinations)

**3.11. Power conversion efficiency**

A are electron and hole lifetimes.

Power conversion efficiency (η). It is given as below:

h

In most of the cases work function of the front and back contacts are well defined. Otherwise, in flat band case the work function of contact (s) is calculated as follows:

$$\Phi\_m = \mathcal{X} + k\_B T \ln(\frac{N\_c}{N\_D - N\_A}) \tag{24}$$

when contact is n-type

$$
\Phi\_m = \mathcal{X} + E\_g - k\_B T \ln(\frac{N\_c}{N\_A - N\_D}) \tag{25}
$$

when contact is p-type

$$
\Phi\_m = \mathcal{X} + k\_B T \ln(\frac{N\_\odot}{n\_i}) \tag{26}
$$

when contact is intrinsic.

where Φ<sup>m</sup> is workfunction of metal/material, χ is electron affinity, kB is Boltzmann constant, T is operating temperature, NC is conduction band effective density of states, and NA and ND are shallow acceptor and donor dopant concentrations respectively.

## **4. Results and Discussions**

Figure 11 is the absorptance plots where the absorption coefficient, α (1/m) of titanium oxide is derived from reference [44], Zinc oxide is from SCAPS abs. file and that of MAPbI3 is based on [45, 46]. The absorption coefficient for MAPbI3 perovskite displayed in the Figure is very similar to that measured for perovskite synthesized in UPV laboratory, Valencia, Spain. The energy band diagram for the solar cell used in the simulation is shown in Figure 12.

From Figure 11, it is clear that the cutoff wavelength of perovskite lays around 825 nm so, the band gap ≈ 1.5 eV, which is the closer value required for ultimate theoretical limit according to W. Shockley and H. J. Queisser [18]. The band gap of perovskite depends on size of organic/ inorganic (larger cation), metal (small cation), and halide (anion) components chosen during the processing and preparation. Increase in size of the cations and anion changes the covalent character between the cation (s) and anion and size of ionic radius which ultimately yields blue shift and red-shift respectively of the band gap [22, 47]. The band gap of the absorbing material is a crucial parameter for photovoltaic action as the absorber is the key material in a solar cell device [9]*.* Thus, suitable combination of the cationic components and anionic component in when contact is p-type

when contact is intrinsic.

dopant concentrations respectively.

4. Results and Discussions

simulation is shown in Figure 12.

m B

ln( ) i

n <sup>N</sup> <sup>Φ</sup> <sup>=</sup> <sup>χ</sup> <sup>+</sup> <sup>k</sup> <sup>T</sup> (26)

C

where Φm is workfunction of metal/material, χ is electron affinity, kB is Boltzmann constant, T is operating temperature, NC is conduction band effective density of states, and NA and ND are shallow acceptor and donor

Figure 11 is the absorptance plots where the absorption coefficient, α (1/m) of titanium oxide is derived from reference [44] Quanrong Deng et al., 2012, Zinc oxide is from SCAPS abs. file and that of MAPbI3 is based on [45, 46]. The absorption coefficient for MAPbI3 perovskite displayed in the Figure is very similar to that measured for perovskite synthesized in UPV laboratory, Valencia, Spain. The energy band diagram for the solar cell used in the

From Figure 11, it is clear that the cutoff wavelength of perovskite lays around 825 nm so, the band gap ≈ 1.5 eV, which is the closer value required for ultimate theoretical limit according to W. Shockley and H. J. Queisser [18]. The band gap of perovskite depends on size of organic/ inorganic (larger cation), metal (small cation), and halide (anion) components chosen during the processing and preparation. Increase in size of the cations and anion

blue shift and red-shift respectively of the band gap [22, 47]. The band gap of the absorbing material is a crucial

The absorber layer, charge transporting layers, front and back contacts, defects and interface/surface property

are the major components and properties they have an effect on the performance of the solar cell systems. Operating temperature of the solar cells plays a vital role in the performance of the PV devices. Considering

these facts the simulation works are undertaken to find out role of individual layer and their associated

As absorber is the key component in the solar cell devices the deep knowledge and understanding about this

matter is crucial to design and fabricate a solar cell. Here our study has focused on realization of role of absorber to make more practical and optimization of the performance of the PV devices. For compensation (NA/ND) ratio and Gaussian energy distribution we have chosen two ETMs; TiO2 and ZnO. Simulation with different aspects of

Performance of the cell depends mainly on optoelectronic characteristics and thichness of the absorber layer. Thus

simulations are carried out to examine the device parameters with thickness of the absorber from 50 nm to 700 nm under 1 Sun (AM1.5G) illumination. First of all simulations were carried out without considering interface trap density of states neither shallow minority carrier concentration. But inputs value of band tail density of states, and Gaussian acceptor/donor states of MAPbI3 were 10×10<sup>14</sup> eV-1cm-3 and10×10<sup>14</sup>cm-3 respectively. The short circuit current and PCE both are found to be increased sharply with increase in thickness up to 500 nm as shown in Figure 13. After this, increment is very slow and reaches to almost optimal efficiency 25.22% , VOC 1.2 V, JSC 25.49 mA/cm<sup>2</sup> and FF 82.56% at 700 nm which is closed to the detailed bance limit [18]. At the 700 nm thickness MAPbI3 absorbs almost incident photons to create the electron-hole pairs and the photo generated almost carriers

are separated and transproted to the HTM and ETM by the built in field with minimum recombination thus, it

perovskite results into an extremely suitable absorber to yield the solar cells of better per‐ formance. parameter for photovoltaic action as the absorber is the key material in a solar cell device [9]. Thus, suitable combination of the cationic components and anionic component in perovskite results into an extremely suitable absorber to yield the solar cells of better performance.

Figure 11. Absorbance of ZnO, TiO2 and MAPbI<sup>3</sup> **Figure 11.** Absorbance of ZnO, TiO2 and MAPbI3

**Figure 12.** Energy band allignment

Figure 12. Energy band allignment

4.1. Role of absorber

properties which are briefly discussed below.

the absorbing material is discussed below.

4.1.1. Variation of thickness of the absorber

can be consider as the length of optimal photovoltaic action.

4.1.1.1. Without considering interface trap density of states

parameter for photovoltaic action as the absorber is the key material in a solar cell device [9]. Thus, suitable combination of the cationic components and anionic component in perovskite results into an extremely suitable The **absorber layer**, **charge transporting layers**, front and back contacts, defects, interface and surface property are the major components and properties they have effect on the performance of the solar cell systems. **Operating temperature** of the solar cells plays a vital role in the performance of the PV devices. Considering these facts the simulation works are undertaken to find out role of individual layer and their associated properties which are briefly discussed below.

## **4.1. Role of absorber**

perovskite results into an extremely suitable absorber to yield the solar cells of better per‐

absorber to yield the solar cells of better performance.

 ZnO TiO<sup>2</sup> MAPbI<sup>3</sup>

200 300 400 500 600 700 800 900

Wavelength (nm)

Conduction Band Mimnimum but Workfunction for Contacts

p +

> Ag Back contact

p+

The absorber layer, charge transporting layers, front and back contacts, defects and interface/surface property

are the major components and properties they have an effect on the performance of the solar cell systems. Operating temperature of the solar cells plays a vital role in the performance of the PV devices. Considering

these facts the simulation works are undertaken to find out role of individual layer and their associated

As absorber is the key component in the solar cell devices the deep knowledge and understanding about this

matter is crucial to design and fabricate a solar cell. Here our study has focused on realization of role of absorber to make more practical and optimization of the performance of the PV devices. For compensation (NA/ND) ratio and Gaussian energy distribution we have chosen two ETMs; TiO2 and ZnO. Simulation with different aspects of

Performance of the cell depends mainly on optoelectronic characteristics and thichness of the absorber layer. Thus

simulations are carried out to examine the device parameters with thickness of the absorber from 50 nm to 700 nm under 1 Sun (AM1.5G) illumination. First of all simulations were carried out without considering interface trap density of states neither shallow minority carrier concentration. But inputs value of band tail density of states, and Gaussian acceptor/donor states of MAPbI3 were 10×10<sup>14</sup> eV-1cm-3 and10×10<sup>14</sup>cm-3 respectively. The short circuit current and PCE both are found to be increased sharply with increase in thickness up to 500 nm as shown in Figure 13. After this, increment is very slow and reaches to almost optimal efficiency 25.22% , VOC 1.2 V, JSC 25.49 mA/cm<sup>2</sup> and FF 82.56% at 700 nm which is closed to the detailed bance limit [18]. At the 700 nm thickness MAPbI3 absorbs almost incident photons to create the electron-hole pairs and the photo generated almost carriers

are separated and transproted to the HTM and ETM by the built in field with minimum recombination thus, it

SpiroMeOTAD

0

Figure 11. Absorbance of ZnO, TiO2 and MAPbI<sup>3</sup> **Figure 11.** Absorbance of ZnO, TiO2 and MAPbI3

Vacuum Level

1x10<sup>7</sup>

2x10<sup>7</sup>

Absorbance (m-1


**Figure 12.** Energy band allignment

Front Contact

Figure 12. Energy band allignment

4.1. Role of absorber




Energy Band (eV)


e-

e-

Valence Band Maximum

e -

> p +

FTO TiO<sup>2</sup> Au

MAPbI<sup>3</sup>

ETM HTM

properties which are briefly discussed below.

the absorbing material is discussed below.

4.1.1. Variation of thickness of the absorber

can be consider as the length of optimal photovoltaic action.

4.1.1.1. Without considering interface trap density of states



0

)

3x10<sup>7</sup>

4x10<sup>7</sup>

5x10<sup>7</sup>

464 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

when contact is p-type

when contact is intrinsic.

dopant concentrations respectively.

4. Results and Discussions

simulation is shown in Figure 12.

m B

ln( ) i

n <sup>N</sup> <sup>Φ</sup> <sup>=</sup> <sup>χ</sup> <sup>+</sup> <sup>k</sup> <sup>T</sup> (26)

C

where Φm is workfunction of metal/material, χ is electron affinity, kB is Boltzmann constant, T is operating temperature, NC is conduction band effective density of states, and NA and ND are shallow acceptor and donor

Figure 11 is the absorptance plots where the absorption coefficient, α (1/m) of titanium oxide is derived from reference [44] Quanrong Deng et al., 2012, Zinc oxide is from SCAPS abs. file and that of MAPbI3 is based on [45, 46]. The absorption coefficient for MAPbI3 perovskite displayed in the Figure is very similar to that measured for perovskite synthesized in UPV laboratory, Valencia, Spain. The energy band diagram for the solar cell used in the

From Figure 11, it is clear that the cutoff wavelength of perovskite lays around 825 nm so, the band gap ≈ 1.5 eV, which is the closer value required for ultimate theoretical limit according to W. Shockley and H. J. Queisser [18]. The band gap of perovskite depends on size of organic/ inorganic (larger cation), metal (small cation), and halide (anion) components chosen during the processing and preparation. Increase in size of the cations and anion changes the covalent character between the cation(s) and anion and size of ionic radius which ultimately yields blue shift and red-shift respectively of the band gap [22, 47]. The band gap of the absorbing material is a crucial

formance.

As absorber is the key component in the solar cell devices the deep knowledge and under‐ standing about this matter is crucial to design and fabricate a solar cell. Here our study has focused on realization of role of absorber to make more practical and optimization of the performance of the PV devices. For compensation (NA/ND) ratio and Gaussian energy distri‐ bution we have chosen two ETMs; TiO2 and ZnO. Simulation with different aspects of the absorbing material is discussed below.

## *4.1.1. Variation of thickness of the absorber*

## *4.1.1.1. Without considering interface trap density of states*

Performance of the cell depends mainly on optoelectronic characteristics and thichness of the absorber layer. Thus simulations are carried out to examine the device parameters with thickness of the absorber from 50 nm to 700 nm under 1 Sun (AM1.5G) illumination. First of all simulations were carried out without considering interface trap density of states neither shallow minority carrier concentration. But inputs value of band tail density of states, and Gaussian acceptor/donor states of MAPbI3 were 10×1014 eV-1cm-3 and 10×1014 cm-3 respectively. The short circuit current and PCE both are found to be increased sharply with increase in thickness up to 500 nm as shown in Figure 13. After this, increment is very slow and reaches to almost optimal efficiency 25.22%, VOC 1.2 V, JSC 25.49 mA/cm2 and FF 82.56% at 700 nm which is closed to the detailed bance limit [18]. At the 700 nm thickness MAPbI3 absorbs almost incident photons to create the electron-hole pairs and the photo generated almost carriers are separated and transproted to the HTM and ETM by the built in field with minimum recom‐ bination thus, it can be consider as the length of optimal photovoltaic action.

## *4.1.1.2. Considering interface trap density of states*

Dark current and current under illumination in PV solar cells are illustrated in Figure 14. In dark, solar cell behaves as a large flat diode and produces a very small current due to the minority charge carriers in the device structure which resembles to J-V characteristics of a pn junction diode. Device without trap defect is not possible so, to make more practical solar cell we introduce the total trap density = 1.0×109 cm-2 in the interfaces and added shallow level donor and acceptor density ND = 10% of NA and NA = 10% of ND in HTM and ETM respectively. Simulation results shown in Figures 7 is quite interesting and suggested to more rational. State of affairs here is somewhat different than the former case. It reveals that: i) the device param‐ eters are found to be decreasing with introduction of interface traps due to change in shunt

Figure 13. JSC and PCE vs thickness **Figure 13.** JSC and PCE vs thickness

**Figure 14.** J-V curve as a function of illuminations

JSC (mA/cm2

)

resistance, ii) parameters are increasing more or less steeply up to 350 nm and very slowly beyound 450 nm, iii) fill factor is observed maximum value of 82.24% for 100 nmthickness where other parameters also have significant values. The results show that 450 nm thichness is also sufficient for good photovoltaic action. Furthermore, 700 nm is sufficient for optimal photovoltaic performance where VOC, JSC, FF and PCE are observed 1.12 V, 25.49 mA/cm2 , 82.29% and 23.48% respectively. Beyond this thickness there might increase in resistance that Figure 14. J-V curve as a function of illuminations 4.1.1.2. Considering interface trap density of states Dark current and current under illumination in PV solar cells are illustrated in Figure 14. In dark, solar cell

0.0 0.4 0.8 1.2

Voltage (V)

improve to move towards the practicality than reported by [45].

behaves as a large flat diode and produces a very small current due to the minority charge carriers in the device structure which resembles to J-V characteristics of a pn junction diode. Device without trap defect is not possible so, to make more practical solar cell we introduce the total trap density = 1.0×10<sup>9</sup> cm-2 in the interfaces and added shallow level donor and acceptor density ND = 10% of NA and NA = 10% of ND in HTM and ETM respectively. Simulation results shown in Figures 7 is quite interesting and suggested to more rational. State of affairs here is somewhat different than the former case. It reveals that: i) the device parameters are found to be decreasing with introduction of interface traps due to change in shunt resistance, ii) parameters are increasing more or less steeply

up to 350 nm and very slowly beyound 500 nm, iii) fill factor is observed maximum value of 82.24% for 100

nmthickness where other parameters also have significant values. The results show that 500 nm thichness is also sufficient for good photovoltaic action. Furthermore, 700 nm is sufficient for optimal photovoltaic performance where VOC, JSC, FF and PCE are observed 1.12 V, 25.49 mA/cm<sup>2</sup>, 82.29% and 23.48% respectively. Beyond this thickness there might increase in resistance that results in insignificant increment in PCE. The rate of the

photogeneration of charge carriers depends on the amount of light reaching, absorption range and thichness of the absorber, here MAPbI3 perovskite since it is the key component of the solar cell. Figure 15 is the quantum efficiency curves as a function of wavelength of incident light for different thickness of the absorber also verifies the above mentioned upshot. Besides, our results for an absorber thickness of 400 nm is similar but expected to

results in insignificant increment in PCE. The rate of the photogeneration of charge carriers depends on the amount of light reaching, absorption range and thichness of the absorber, here MAPbI3 perovskite since it is the key component of the solar cell. Figure 15 is the quantum efficiency curves as a function of wavelength of incident light for different thickness of the absorber also verifies the above mentioned upshot. Besides, our results for an absorber thickness of 400 nm is similar but expected to improve to move towards the practicality than reported by [45].

Figure 15. Variation of quantum efficiency with thickness of absorber. **Figure 15.** Variation of quantum efficiency with thickness of absorber.

### 4.1.2. Band tail density of states of perovskite The input parameters for band tail density has varied from 1.0×10<sup>12</sup> eV-1.cm-3 to 1.0×10<sup>17</sup>eV-1.cm-3. J-V curves as a *4.1.2. Band tail density of states of perovskite*

resistance, ii) parameters are increasing more or less steeply up to 350 nm and very slowly beyound 450 nm, iii) fill factor is observed maximum value of 82.24% for 100 nmthickness where other parameters also have significant values. The results show that 450 nm thichness is also sufficient for good photovoltaic action. Furthermore, 700 nm is sufficient for optimal photovoltaic performance where VOC, JSC, FF and PCE are observed 1.12 V, 25.49 mA/cm2

0.0 0.4 0.8 1.2

0.0 0.4 0.8 1.2

Voltage (V)

Voltage (V)

Figure 14. J-V curve as a function of illuminations

0 200 400 600 800

minority carrier concentration

Without considering interface defects and

Thickness (nm)

JSC

PCE

10

12

14

16

18

PCE (%)

20

22

24

26

10


**Figure 14.** J-V curve as a function of illuminations



JSC

JSC (mA/cm2

)

**Figure 13.** JSC and PCE vs thickness


 (mA/cm2

0

)

6

12

Figure 13. JSC and PCE vs thickness

(1 Sun)

(1 Sun)

VOC = 1.12 V

FF = 82.12% PCE = 22.72%

SC = 24.66 mA/cm2

VOC = 1.12 V JSC = 24.66 mA/cm2 FF = 82.12% PCE = 22.72%

J





0

6

12

Dark Current

Dark Current

Current Under Illumination

Current Under Illumination

Thickness of absorber: 450 nm

Thickness of absorber: 450 nm

12

14

16

18

JSC (mA/cm2

)

20

22

24

26

466 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

82.29% and 23.48% respectively. Beyond this thickness there might increase in resistance that

4.1.1.2. Considering interface trap density of states

improve to move towards the practicality than reported by [45].

,

behaves as a large flat diode and produces a very small current due to the minority charge carriers in the device structure which resembles to J-V characteristics of a pn junction diode. Device without trap defect is not possible so, to make more practical solar cell we introduce the total trap density = 1.0×10<sup>9</sup> cm-2 in the interfaces and added shallow level donor and acceptor density ND = 10% of NA and NA = 10% of ND in HTM and ETM respectively. Simulation results shown in Figures 7 is quite interesting and suggested to more rational. State of affairs here is somewhat different than the former case. It reveals that: i) the device parameters are found to be decreasing with introduction of interface traps due to change in shunt resistance, ii) parameters are increasing more or less steeply

up to 350 nm and very slowly beyound 500 nm, iii) fill factor is observed maximum value of 82.24% for 100

nmthickness where other parameters also have significant values. The results show that 500 nm thichness is also sufficient for good photovoltaic action. Furthermore, 700 nm is sufficient for optimal photovoltaic performance where VOC, JSC, FF and PCE are observed 1.12 V, 25.49 mA/cm<sup>2</sup>, 82.29% and 23.48% respectively. Beyond this thickness there might increase in resistance that results in insignificant increment in PCE. The rate of the

photogeneration of charge carriers depends on the amount of light reaching, absorption range and thichness of the absorber, here MAPbI3 perovskite since it is the key component of the solar cell. Figure 15 is the quantum efficiency curves as a function of wavelength of incident light for different thickness of the absorber also verifies the above mentioned upshot. Besides, our results for an absorber thickness of 400 nm is similar but expected to

Dark current and current under illumination in PV solar cells are illustrated in Figure 14. In dark, solar cell function of band tail defect is shown in Figure 16, which shows the defects in the absorber has greatly influenced the VOC, and finally to the performance of the device. Infact, band tail density of states are the intra-band gap states which may act as recombination centers for charge carriers and are also called defects in layer(s) of the solar cell. Simulation studies have envisaged that similar nature is not observed for HTM and ETM as reported by [45] (Feng Liu et al., 2014) too, since photogeneration of the charge carriers is taken place in the absorber say active layer. The Figure 16 tells that defects do not significantly influence the JSC since amount of incident photon and thickness are remaing unchanged. Relatively higher value of VOC is observed which is similar to [45, 48, 49]. It is also observed that VOC and JSC are almost same from 1.0×10<sup>12</sup> to 1.0×10<sup>14</sup> eV-1·cm-3. These are the consequences of the larger absorption rang, high crystalinity, low pinhole, shallow level defect densities and a prolonged electronhole lifetime in perovskite which are characteritics of materials and also created unintentionally/purposefully during the deposition/coating/processing the layer(s). Similar result is reported by L. M. Herz et. al. [49]. The input parameters for band tail density has varied from 1.0×1012 eV-1.cm-3 to 1.0×1017 eV-1.cm-3. J-V curves as a function of band tail defect is shown in Figure 16, which shows the defects in the absorber has greatly influenced the VOC, and finally to the performance of the device. Infact, band tail density of states are the intra-band gap states which may act as recombination centers for charge carriers and are also called defects in layer (s) of the solar cell. Simulation studies have envisaged that similar nature is not observed for HTM and ETM as reported by [45] too, since photogeneration of the charge carriers is taken place in the absorber say active layer. The Figure 16 tells that defects do not significantly influence the JSC since amount of incident photon and thickness are remaing unchanged*.* Relatively higher value of VOC is observed which is similar to [45, 48, 49]. It is also observed that VOC and JSC are almost same from 1.0×1012 to 1.0×1014 eV-1 cm-3. These are the consequences of the larger absorption rang, high crystalinity, low pinhole, shallow level defect densities and a prolonged

electron-hole lifetime in perovskite which are characteritics of materials and also created unintentionally/purposefully during the deposition/coating/processing the layer (s). Similar result is reported by L. M. Herz et. al. [49].

Figure 16. JV curve as a function of band tail density of states **Figure 16.** JV curve as a function of band tail density of states

4.1.3. Role of interface trap density

#### Research process always endeavor towards the perfect by realizing and optimizing the overall behaviours of the *4.1.3. Role of interface trap density*

and studied the results. Figure 17 shows that VOC and PCE are decreasing with increase in interface trap density upto 1.0×10<sup>13</sup> cm-2 and remaining almost same beyond this. Figure 18 shows that QE is considerably small when trap density is above 1.0×10<sup>11</sup> cm-2. The plots show the crucial role of interface trap density in efficiency of the solar cells since interface traps at high level are also the recombination centers and hence change in shunt resistance. Optimizing doping of the HTM and ETM, and formation of homogeneous, smooth, and flat surface will significantly reduce the interface trap density. 0.9 1.0 1.1 1.2 VOC (V) Research process always endeavor towards the perfect by realizing and optimizing the overall behaviours of the material*.* So, simulations were carried out with setting the interface trap densities from 1.0×108 cm-2 to 1.0×1014 cm-2 and studied the results. Figure 17 shows that VOC and PCE are decreasing with increase in interface trap density upto 1.0×1013 cm-2 and remaining almost same beyond this. Figure 18 shows that QE is considerably small when trap density is above 1.0×1011 cm-2. The plots show the crucial role of interface trap density in efficiency of the solar cells since interface traps at high level are also the recombination centers and hence change in shunt resistance. Optimizing doping of the HTM and ETM, and formation of homogeneous, smooth, and flat surface will significantly reduce the interface trap density.

material. So, simulations were carried out with setting the interface trap densities from 1.0×10<sup>8</sup>cm-2 to 1.0×10<sup>14</sup>cm-2

#### 0.8 22 24 *4.1.4. Compensation ratio (NA/ND) of the absorber*

20

1E8 1E9 1E10 1E11 1E12 1E13 1E14 14 16 18 PCE (%)Interface Trap (cm-2) Figure 17. VOC and PC vs interface traps For this study, simulation works have been carried out with two different ETMs; TiO2 and ZnO. Figure 19 shows the effect of compensation ratio in MAPbI3 perovskite. In this case, simulation works have been carried out with variation of compensation ratio from 0 to 20%. Increase in compensation ratio of perovskite, an absorber, and results into slightly decreased in the performance of the PV devices in both cases suggested to increase in recombination/loss within it. The similar effect which has been observed in both ETMs increases the aptness of substituting TiO2 by ZnO.

Research process always endeavor towards the perfect by realizing and optimizing the overall behaviours of the material. So, simulations were carried out with setting the interface trap densities from 1.0×10<sup>8</sup>cm-2 to 1.0×10<sup>14</sup>cm-2 and studied the results. Figure 17 shows that VOC and PCE are decreasing with increase in interface trap density upto 1.0×10<sup>13</sup> cm-2 and remaining almost same beyond this. Figure 18 shows that QE is considerably small when trap density is above 1.0×10<sup>11</sup> cm-2. The plots show the crucial role of interface trap density in efficiency of the solar cells since interface traps at high level are also the recombination centers and hence change in shunt resistance. Optimizing doping of the HTM and ETM, and formation of homogeneous, smooth, and flat surface

0.0 0.4 0.8 1.2

Voltage (V)

 10<sup>12</sup> 10<sup>13</sup> 10<sup>14</sup> 10<sup>15</sup> 10<sup>16</sup> 10<sup>17</sup>

Band Tail Density of States

0.0 0.4 0.8 1.2





JSC (mA/cm2

)




(eV-1

cm

Figure 16. JV curve as a function of band tail density of states

4.1.3. Role of interface trap density

0

**Figure 17.** VOC and PC vs interface traps Figure 17. VOC and PC vs interface traps

electron-hole lifetime in perovskite which are characteritics of materials and also created unintentionally/purposefully during the deposition/coating/processing the layer (s). Similar

0.0 0.4 0.8 1.2

 10<sup>12</sup> 10<sup>13</sup> 10<sup>14</sup> 10<sup>15</sup> 10<sup>16</sup> 10<sup>17</sup>

Band Tail Density of States

) of MAPbI<sup>3</sup>

0.0 0.4 0.8 1.2

Voltage (V)

Research process always endeavor towards the perfect by realizing and optimizing the overall behaviours of the material*.* So, simulations were carried out with setting the interface trap densities from 1.0×108 cm-2 to 1.0×1014 cm-2 and studied the results. Figure 17 shows that VOC and PCE are decreasing with increase in interface trap density upto 1.0×1013 cm-2 and remaining almost same beyond this. Figure 18 shows that QE is considerably small when trap density is above 1.0×1011 cm-2. The plots show the crucial role of interface trap density in efficiency of the solar cells since interface traps at high level are also the recombination centers and hence change in shunt resistance. Optimizing doping of the HTM and ETM, and formation of homogeneous, smooth, and flat surface will significantly reduce the interface trap density.

Research process always endeavor towards the perfect by realizing and optimizing the overall behaviours of the material. So, simulations were carried out with setting the interface trap densities from 1.0×10<sup>8</sup>cm-2 to 1.0×10<sup>14</sup>cm-2 and studied the results. Figure 17 shows that VOC and PCE are decreasing with increase in interface trap density upto 1.0×10<sup>13</sup> cm-2 and remaining almost same beyond this. Figure 18 shows that QE is considerably small when trap density is above 1.0×10<sup>11</sup> cm-2. The plots show the crucial role of interface trap density in efficiency of the solar cells since interface traps at high level are also the recombination centers and hence change in shunt resistance. Optimizing doping of the HTM and ETM, and formation of homogeneous, smooth, and flat surface

result is reported by L. M. Herz et. al. [49].


**Figure 16.** JV curve as a function of band tail density of states



JSC (mA/cm2

0.8 0.9 1.0 1.1 1.2

*4.1.4. Compensation ratio (NA/ND) of the absorber*

VOC (V)

*4.1.3. Role of interface trap density*

PCE (%)

substituting TiO2 by ZnO.

)




(eV-1 cm-3

Figure 16. JV curve as a function of band tail density of states

will significantly reduce the interface trap density.

1E8 1E9 1E10 1E11 1E12 1E13 1E14

Interface Trap (cm-2

Figure 17. VOC and PC vs interface traps

)

For this study, simulation works have been carried out with two different ETMs; TiO2 and ZnO. Figure 19 shows the effect of compensation ratio in MAPbI3 perovskite. In this case, simulation works have been carried out with variation of compensation ratio from 0 to 20%. Increase in compensation ratio of perovskite, an absorber, and results into slightly decreased in the performance of the PV devices in both cases suggested to increase in recombination/loss within it. The similar effect which has been observed in both ETMs increases the aptness of

4.1.3. Role of interface trap density

0

468 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Figure 18. QE as a function of interface traps **Figure 18.** QE as a function of interface traps

#### *4.1.5. Gaussian Energy Distribution in Perovskite* 4.1.4. Compensation ratio (NA/ND) of the absorber

1.10 1.15 1.20

VOC (V)

JSC (mA/cm2

)

24.6

24.2

81

22.0 22.2 22.4 22.6

PCE (%)

82

FF (%)

83

24.4

Figures 20 and 21 are the JV curves as the function of the Gaussian energy distribution (defect) of MAPbI3 perovskite. Simulations have been carried out with varying the defect values from 1012 to 1019 cm-3*.* It is found that VOC and hence PCE of PV devices have been decreasing with increase in Gaussian defect for both ETMs. Beyond 1017 cm-3 i.e., at high defect levels perform‐ ance of the device goes significantly decreasing due to more recombination, loss and change in resistance. For this study, simulation works have been carried out with two different ETMs; TiO2 and ZnO. Figure 19 shows the effect of compensation ratio in MAPbI3 perovskite. In this case, simulation works have been carried out with variation of compensation ratio from 0 to 20%. Increase in compensation ratio of perovskite, an absorber, and results into slightly decreased in the performance of the PV devices in both cases suggested to increase in recombination/loss within it. The similar effect which has been observed in both ETMs increases the aptness of substituting TiO2 by ZnO.

0.00 0.04 0.08 0.12 0.16 0.20

NA/ND of MAPbI<sup>3</sup>

Figure 19. PVC Parameters vs Compensation ratio (NA/ND) of MAPbI<sup>3</sup>

4.1.5. Gaussian Energy Distribution in Perovskite

recombination, lose and change in resistance.

 TiO<sup>2</sup> ZnO

> TiO<sup>2</sup> ZnO

 TiO2 ZnO

TiO<sup>2</sup>

ZnO

Figures 20 and 21 are the JV curves as the function of the Gaussian energy distribution (defect) of MAPbI3

10<sup>17</sup> cm-3 i.e., at high defect levels performance of the device goes significantly decreasing due to more

perovskite. Simulations have been carried out with varying the defect values from 10<sup>12</sup> to 10<sup>19</sup> cm-3. It is found that VOC and hence PCE of PV devices have been decreasing with increase in Gaussian defect for both ETMs. Beyond

0

20

40

QE (% ) 60

80

100

300 400 500 600 700 800 900

Interface Trap Concentration

Figure 18. QE as a function of interface traps

Wavelength (nm)

4.1.4. Compensation ratio (NA/ND) of the absorber

For this study, simulation works have been carried out with two different ETMs; TiO2 and ZnO. Figure 19 shows the effect of compensation ratio in MAPbI3 perovskite. In this case, simulation works have been carried out with variation of compensation ratio from 0 to 20%. Increase in compensation ratio of perovskite, an absorber, and results into slightly decreased in the performance of the PV devices in both cases suggested to increase in recombination/loss within it. The similar effect which has been observed in both ETMs increases the aptness of

 1E8 1E9 1E11 1E14

Figure 19. PVC Parameters vs Compensation ratio (NA/ND) of MAPbI<sup>3</sup> **Figure 19.** PVC Parameters vs Compensation ratio (NA/ND) of MAPbI3

4.1.5. Gaussian Energy Distribution in Perovskite

Figure 20. Gaussian Defects of Perovskite with TiO<sup>2</sup> **Figure 20.** Gaussian Defects of Perovskite with TiO2

### **4.2. Role of Electron Transporting Materials**


discussed below.

4.2.1. Thickness of the ETMs



JSC (mA/cm2

)



0


> 10<sup>12</sup> 10<sup>13</sup> 10<sup>14</sup> 10<sup>15</sup> 10<sup>16</sup> 10<sup>17</sup> 10<sup>18</sup> 10<sup>19</sup>

0.0 0.2 0.4 0.6 0.8 1.0

4.2. Role of Electron Transporting Materials

Two usual conducting oxides; titanium oxide and zinc oxide have been used to analyze the role of electron transporting layer in the heterojunction thin film solar cells. The band gap and electron affinity of the materials should be such that it conducts the free electrons and creates barrier to the holes. Here simulation works are carried out with varying layer properties of two transparent conducting oxides as ETMs. They are briefly

Figure 22 is the plot of solar cell parameters; VOC, JSC, FF and PCE versus thickness of the ETMs; TiO2 andZnO. In both cases VOC, JSC and PCE are gradually decreasing due to fractional absorption of incident light by the ETMs

Thickness of ETMs has been varied from 50 nm to 450 nm to make the practical devices. Observation showed that

transmittance than ZnO. It is obvious that the increase in thickness of the ETM lessen the performance of the solar

layer, the bulk recombination and surface recombination at the interface and change in series resistance.

TiO2 is more sensitive than that of ZnO due to its high absorption coefficient and reflectance and less

cells due to increase in partial absorption of photons and resistance of the device.

Figure 21. Gaussian Defects of Perovskite with ZnO

Voltage (V)

0.0 0.2 0.4 0.6 0.8 1.0

Voltage (V)

 10<sup>12</sup> 10<sup>13</sup> 10<sup>14</sup> 10<sup>15</sup> 10<sup>16</sup> 10<sup>17</sup> 10<sup>18</sup> 10<sup>19</sup>

Gaussian Defects Donor/Acceptor

State Density (cm-3)\_TiO<sup>2</sup>

Figure 21. Gaussian Defects of Perovskite with ZnO **Figure 21.** Gaussian Defects of Perovskite with ZnO




JSC (mA/cm2

)




0

electron affinity of the materials should be such that it conducts the free electrons and creates barrier to the holes. Here simulation works are carried out with varying layer properties of two transparent conducting oxides as ETMs. They are briefly discussed below. 4.2. Role of Electron Transporting Materials Two usual conducting oxides; titanium oxide and zinc oxide have been used to analyze the role of electron

#### *4.2.1. Thickness of the ETMs* transporting layer in the heterojunction thin film solar cells. The band gap and electron affinity of the materials

0.0 0.2 0.4 0.6 0.8 1.0

0.0 0.2 0.4 0.6 0.8 1.0

4.2. Role of Electron Transporting Materials

Two usual conducting oxides; titanium oxide and zinc oxide have been used to analyze the role of electron transporting layer in the heterojunction thin film solar cells. The band gap and electron affinity of the materials should be such that it conducts the free electrons and creates barrier to the holes. Here simulation works are carried out with varying layer properties of two transparent conducting oxides as ETMs. They are briefly

Figure 22 is the plot of solar cell parameters; VOC, JSC, FF and PCE versus thickness of the ETMs; TiO2 andZnO. In both cases VOC, JSC and PCE are gradually decreasing due to fractional absorption of incident light by the ETMs

Thickness of ETMs has been varied from 50 nm to 450 nm to make the practical devices. Observation showed that

transmittance than ZnO. It is obvious that the increase in thickness of the ETM lessen the performance of the solar

layer, the bulk recombination and surface recombination at the interface and change in series resistance.

TiO2 is more sensitive than that of ZnO due to its high absorption coefficient and reflectance and less

cells due to increase in partial absorption of photons and resistance of the device.

Figure 21. Gaussian Defects of Perovskite with ZnO

Voltage (V)

 10<sup>12</sup> 10<sup>13</sup> 10<sup>14</sup> 10<sup>15</sup> 10<sup>16</sup> 10<sup>17</sup> 10<sup>18</sup> 10<sup>19</sup>

Gaussian Defects Donor/Acceptor

Two usual conducting oxides; titanium oxide and zinc oxide have been used to analyze the role of electron transporting layer in the heterojunction thin film solar cells. The band gap and

State Density (cm-3

Figure 20. Gaussian Defects of Perovskite with TiO<sup>2</sup>

Voltage (V)

)\_ZnO

 10<sup>12</sup> 10<sup>13</sup> 10<sup>14</sup> 10<sup>15</sup> 10<sup>16</sup> 10<sup>17</sup> 10<sup>18</sup> 10<sup>19</sup>

Gaussian Defects Donor/Acceptor

Figure 19. PVC Parameters vs Compensation ratio (NA/ND) of MAPbI<sup>3</sup> 4.1.5. Gaussian Energy Distribution in Perovskite

300 400 500 600 700 800 900

Interface Trap Concentration

Figure 18. QE as a function of interface traps

substituting TiO2 by ZnO.

470 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

1.10 1.15 1.20

VOC (V)

JSC (mA/cm2

)

24.2 24.4 24.6

FF (%)

81 82 83

22.0 22.2 22.4 22.6

PCE (%)

Wavelength (nm)

4.1.4. Compensation ratio (NA/ND) of the absorber

For this study, simulation works have been carried out with two different ETMs; TiO2 and ZnO. Figure 19 shows the effect of compensation ratio in MAPbI3 perovskite. In this case, simulation works have been carried out with variation of compensation ratio from 0 to 20%. Increase in compensation ratio of perovskite, an absorber, and results into slightly decreased in the performance of the PV devices in both cases suggested to increase in recombination/loss within it. The similar effect which has been observed in both ETMs increases the aptness of

> TiO<sup>2</sup> ZnO

> > TiO<sup>2</sup> ZnO

 TiO2 ZnO

TiO<sup>2</sup>

ZnO

Figures 20 and 21 are the JV curves as the function of the Gaussian energy distribution (defect) of MAPbI3 perovskite. Simulations have been carried out with varying the defect values from 10<sup>12</sup> to 10<sup>19</sup> cm-3. It is found that VOC and hence PCE of PV devices have been decreasing with increase in Gaussian defect for both ETMs. Beyond

10<sup>17</sup> cm-3 i.e., at high defect levels performance of the device goes significantly decreasing due to more

 1E8 1E9 1E11 1E14

0

20

40

QE (% ) 60

80

100

)\_TiO<sup>2</sup>

0.00 0.04 0.08 0.12 0.16 0.20

NA/ND of MAPbI<sup>3</sup>

State Density (cm-3

recombination, lose and change in resistance.



discussed below.

4.2.1. Thickness of the ETMs



JSC (mA/cm2

)

**Figure 20.** Gaussian Defects of Perovskite with TiO2

**4.2. Role of Electron Transporting Materials**




0



JSC (mA/cm2

)




0

**Figure 19.** PVC Parameters vs Compensation ratio (NA/ND) of MAPbI3

Figure 22 is the plot of solar cell parameters; VOC, JSC, FF and PCE versus thickness of the ETMs; TiO2 and ZnO. In both cases VOC, JSC and PCE are gradually decreasing due to fractional absorption of incident light by the ETMs layer, the bulk recombination and surface recombi‐ nation at the interface and change in series resistance. Thickness of ETMs has been varied from 50 nm to 450 nm to make the practical devices. Observation showed that TiO2 is more sensitive than that of ZnO due to its high absorption coefficient and reflectance and less transmittance than ZnO. It is obvious that the increase in thickness of the ETM lessen the performance of the solar cells due to increase in partial absorption of photons and resistance of the device. should be such that it conducts the free electrons and creates barrier to the holes. Here simulation works are carried out with varying layer properties of two transparent conducting oxides as ETMs. They are briefly discussed below. 4.2.1. Thickness of the ETMs Figure 22 is the plot of solar cell parameters; VOC, JSC, FF and PCE versus thickness of the ETMs; TiO2 andZnO. In both cases VOC, JSC and PCE are gradually decreasing due to fractional absorption of incident light by the ETMs

Moreover, we have carried out simulation works with practically viable thickness of TiO2 - ETM as shown in Figure 23 which is the plot of FF and PCE as a function of thickness of TiO2. FF and PCE are gradually decreasing due to fractional absorption of incident light by the TiO2 layer, the bulk recombination and surface recombination at the interface [22]. But fill factor in this case is found to be slightly increasing with increase in thickness suggested the higher conductivity of the TiO2 than MAPbI3 and SpiroMeOTAD and partial absorption of the light. layer, the bulk recombination and surface recombination at the interface and change in series resistance. Thickness of ETMs has been varied from 50 nm to 450 nm to make the practical devices. Observation showed that TiO2 is more sensitive than that of ZnO due to its high absorption coefficient and reflectance and less transmittance than ZnO. It is obvious that the increase in thickness of the ETM lessen the performance of the solar cells due to increase in partial absorption of photons and resistance of the device.

## *4.2.2. Band tail density of states of perovskite*

Figure 24 and 25 are the J-V curves as a function of band tail defect varies from 1012 to 1017 eV-1.cm-3. The plots show that the defects in the absorber has influenced the VOC, and finally to the performance of the device due to the change in shunt resistance. Band tail density of states 1.11 1.12 1.13

V

OC (V)

24

 TiO<sup>2</sup> ZnO

 TiO<sup>2</sup> ZnO

TiO<sup>2</sup>

ZnO

TiO<sup>2</sup>

ZnO

Figure 22. Variation of PV cells parameters with thickness of ETMs **Figure 22.** Variation of PV cells parameters with thickness of ETMs Figure 22. Variation of PV cells parameters with thickness of ETMs

82.0

4.2.2. Band tail density of states of perovskite Figure 24 and 25 are the J-V curves as a function of band tail defect varies from 10<sup>12</sup> to 10<sup>17</sup> eV-1.cm-3. The plots Figure 23. FF and PCE vs thickness of TiO2. **Figure 23.** FF and PCE vs thickness of TiO2.

are particularly intra-band gap recombination centers are also called defects. Both the Figures tell that defects do not significantly influence the JSC since amount of incident photon and thickness of system are remaing unchanged*.* Relatively higher value of VOC is observed as [48, 49]. It is also observed that VOC and JSC are almost same from 1.0×1012 to 1.0×1014 eV-1 cm-3. the change in shunt resistance. Band tail density of states are particularly intra-band gap recombination centers are also called defects. Both the Figures tell that defects do not significantly influence the JSC since amount of incident photon and thickness of system are remaing unchanged. Relatively higher value of VOC is observed as [48, 49]. It is also observed that VOC and JSC are almost same from 1.0×10<sup>12</sup> to 1.0×10<sup>14</sup> eV-1·cm-3. Moreover, we have carried out simulation works with practically viable thickness of TiO2 -ETM as shown in Figure 23 which is the plot of FF and PCE as a function of thickness of TiO2. FF and PCE are gradually decreasing due to fractional absorption of incident light by the TiO2 layer, the bulk recombination and surface recombination at the interface [22]. But fill factor in this case is found to be slightly increasing with increase in thickness

show that the defects in the absorber has influenced the VOC, and finally to the performance of the device due to

These are the consequences of the larger absorption rang, high crystalinity, low pinhole, shallow level defect densities and a prolonged electron-hole lifetime in perovskite which are characteritics of materials and also created unporposefully/purposefully during the coating/ processing/fabrication of the layer (s). This is similar result as reported by [49]. Furthermore, TiO2 has found relatively less sensitive than in ZnO due to small electron mobility and higher doping concentration than ZnO. suggested the higher conductivity of the TiO2 than MAPbI3 and SpiroMeOTAD and partial absorption of the light. 4.2.2. Band tail density of states of perovskite Figure 24 and 25 are the J-V curves as a function of band tail defect varies from 10<sup>12</sup> to 10<sup>17</sup> eV-1.cm-3. The plots show that the defects in the absorber has influenced the VOC, and finally to the performance of the device due to

the change in shunt resistance. Band tail density of states are particularly intra-band gap recombination centers are also called defects. Both the Figures tell that defects do not significantly influence the JSC since amount of incident photon and thickness of system are remaing unchanged. Relatively higher value of VOC is observed as

[48, 49]. It is also observed that VOC and JSC are almost same from 1.0×10<sup>12</sup> to 1.0×10<sup>14</sup> eV-1·cm-3.

Figure 24. Band Tail of Perovskite with TiO2 as ETM **Figure 24.** Band Tail of Perovskite with TiO2 as ETM Figure 24. Band Tail of Perovskite with TiO2 as ETM

0

due to fractional absorption of incident light by the TiO2 layer, the bulk recombination and surface recombination Figure 25. Band Tail of Perovskite with ZnO as ETM **Figure 25.** Band Tail of Perovskite with ZnO as ETM

#### suggested the higher conductivity of the TiO2 than MAPbI3 and SpiroMeOTAD and partial absorption of the *4.2.3. Interface trap density of states* Figure 25. Band Tail of Perovskite with ZnO as ETM These are the consequences of the larger absorption rang, high crystalinity, low pinhole, shallow level defect

are particularly intra-band gap recombination centers are also called defects. Both the Figures tell that defects do not significantly influence the JSC since amount of incident photon and thickness of system are remaing unchanged*.* Relatively higher value of VOC is observed as [48, 49]. It is also observed that VOC and JSC are almost same from 1.0×1012 to 1.0×1014 eV-1 cm-3.

[48, 49]. It is also observed that VOC and JSC are almost same from 1.0×10<sup>12</sup> to 1.0×10<sup>14</sup> eV-1·cm-3.

4.2.2. Band tail density of states of perovskite

Figure 24 and 25 are the J-V curves as a function of band tail defect varies from 10<sup>12</sup> to 10<sup>17</sup> eV-1.cm-3. The plots show that the defects in the absorber has influenced the VOC, and finally to the performance of the device due to the change in shunt resistance. Band tail density of states are particularly intra-band gap recombination centers are also called defects. Both the Figures tell that defects do not significantly influence the JSC since amount of incident photon and thickness of system are remaing unchanged. Relatively higher value of VOC is observed as

Moreover, we have carried out simulation works with practically viable thickness of TiO2 -ETM as shown in Figure 23 which is the plot of FF and PCE as a function of thickness of TiO2. FF and PCE are gradually decreasing due to fractional absorption of incident light by the TiO2 layer, the bulk recombination and surface recombination at the interface [22]. But fill factor in this case is found to be slightly increasing with increase in thickness suggested the higher conductivity of the TiO2 than MAPbI3 and SpiroMeOTAD and partial absorption of the

0 100 200 300 400 500

Thickness (nm)

0 100 200 300 400 500

Thickness of ETM (nm)

Figure 22. Variation of PV cells parameters with thickness of ETMs

 TiO<sup>2</sup> ZnO  TiO<sup>2</sup> ZnO

 TiO<sup>2</sup> ZnO

TiO<sup>2</sup>

ZnO

TiO<sup>2</sup>

ZnO

 TiO<sup>2</sup> ZnO

 TiO<sup>2</sup> ZnO

TiO<sup>2</sup>

ZnO

Moreover, we have carried out simulation works with practically viable thickness of TiO2 -ETM as shown in Figure 23 which is the plot of FF and PCE as a function of thickness of TiO2. FF and PCE are gradually decreasing

at the interface [22]. But fill factor in this case is found to be slightly increasing with increase in thickness

the change in shunt resistance. Band tail density of states are particularly intra-band gap recombination centers are also called defects. Both the Figures tell that defects do not significantly influence the JSC since amount of incident photon and thickness of system are remaing unchanged. Relatively higher value of VOC is observed as

[48, 49]. It is also observed that VOC and JSC are almost same from 1.0×10<sup>12</sup> to 1.0×10<sup>14</sup> eV-1·cm-3.

0 100 200 300 400 500

Thickness of ETM (nm)

These are the consequences of the larger absorption rang, high crystalinity, low pinhole, shallow level defect densities and a prolonged electron-hole lifetime in perovskite which are characteritics of materials and also created unporposefully/purposefully during the coating/ processing/fabrication of the layer (s). This is similar result as reported by [49]. Furthermore, TiO2 has found relatively less sensitive than in ZnO due to small electron mobility and higher

doping concentration than ZnO.

light.

1.11 1.12 1.13

FF (%)

J

SC (mA/cm2

)

1.11 1.12 1.13

21

83

81

24

20

22

82

24

472 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

V

OC (V)

V

OC (V)

21 24

J

FF (%)

PCE (%)

81.5

82.0

81.5

Figure 23. FF and PCE vs thickness of TiO2.

**Figure 22.** Variation of PV cells parameters with thickness of ETMs

4.2.2. Band tail density of states of perovskite

Figure 23. FF and PCE vs thickness of TiO2.

PCE (%)

PCE (%)

light.

**Figure 23.** FF and PCE vs thickness of TiO2.

82.0

FF (%)

FF (%)

SC (mA/cm2

)

81 82 83

PCE (%)

20 22 24

Figure 22. Variation of PV cells parameters with thickness of ETMs

0 100 200 300 400 500

Thickness (nm)

Figure 24 and 25 are the J-V curves as a function of band tail defect varies from 10<sup>12</sup> to 10<sup>17</sup> eV-1.cm-3. The plots show that the defects in the absorber has influenced the VOC, and finally to the performance of the device due to Figures 26 and 27 are the JV curves as a function of interface trap density varied from 1.0×108 to 1.0×1015 cm-2. It is quite clear from the plots that VOC and PCE are decreasing with increase in interface trap density upto 1.0×1014 cm-2. It shows that interface trap increases the recombination centers and hence change in shunt resistance. There is decrease in JSC with the interface trap beyond 1.0×1012 cm-2 suggested to increase in series resistance. Relatively more These are the consequences of the larger absorption rang, high crystalinity, low pinhole, shallow level defect densities and a prolonged electron-hole lifetime in perovskite which are characteritics of materials and also created unporposefully/purposefully during the coating/processing/fabrication of the layer(s). This is similar result as reported by [49]. Furthermore, TiO2 has found relatively less sensitive than in ZnO due to small electron mobility and higher doping concentration than ZnO. densities and a prolonged electron-hole lifetime in perovskite which are characteritics of materials and also created unporposefully/purposefully during the coating/processing/fabrication of the layer(s). This is similar result as reported by [49]. Furthermore, TiO2 has found relatively less sensitive than in ZnO due to small electron mobility and higher doping concentration than ZnO.

> Figures 26 and 27 are the JV curves as a function of interface trap density varied from 1.0×10<sup>8</sup>to 1.0×10<sup>15</sup> cm-2. It is quite clear from the plots that VOC and PCE are decreasing with increase in interface trap density upto 1.0×10<sup>14</sup> cm-<sup>2</sup>. It shows that interface trap increases the recombination centers and hence change in shunt resistance. There is decrease in JSC with the interface trap beyond 1.0×10<sup>12</sup> cm-2 suggested to increase in series resistance. Relatively more sensitiveness of ZnO than TiO2 towards interface trap indicates the relatively small defect and small electron mobility of TiO2. Optimizing doping of the ETM, and formation of homogeneous, smooth, and flat

> Figures 26 and 27 are the JV curves as a function of interface trap density varied from 1.0×10<sup>8</sup>to 1.0×10<sup>15</sup> cm-2. It is quite clear from the plots that VOC and PCE are decreasing with increase in interface trap density upto 1.0×10<sup>14</sup> cm-<sup>2</sup>. It shows that interface trap increases the recombination centers and hence change in shunt resistance. There is decrease in JSC with the interface trap beyond 1.0×10<sup>12</sup> cm-2 suggested to increase in series resistance. Relatively more sensitiveness of ZnO than TiO2 towards interface trap indicates the relatively small defect and small electron mobility of TiO2. Optimizing doping of the ETM, and formation of homogeneous, smooth, and flat

surface will significantly reduce the interface trap density and hence increase the performance.

surface will significantly reduce the interface trap density and hence increase the performance.

4.2.3. Interface trap density of states

4.2.3. Interface trap density of states

Voltage (V)

0

sensitiveness of ZnO than TiO2 towards interface trap indicates the relatively small defect and small electron mobility of TiO2. Optimizing doping of the ETM, and formation of homogene‐ ous, smooth, and flat surface will significantly reduce the interface trap density and hence increase the performance.

**Figure 26.** Interface Trap of Perovskite with TiO2 Figure 26. Interface Trap of Perovskite with TiO<sup>2</sup>

Figure 27. Interface Trap of Perovskite with ZnO

4.2.4. Role of Dopant Concentrations of ETMs.

4.2.4. Role of Dopant Concentrations of ETMs.

smaller value of electron mobility and hence small conductivity of TiO2.

smaller value of electron mobility and hence small conductivity of TiO2.

The plot of PV cell parameters as a function of dopant concentration (ND) of both ETMs TiO2 and ZnO is shown in

The plot of PV cell parameters as a function of dopant concentration (ND) of both ETMs TiO2 and ZnO is shown in

Figure 28. The study has been carried out from 10<sup>15</sup> to 10<sup>21</sup> cm-3 values. Both ETMs have exhibited the similar performance here too. Voc, FF and PCE have been increasing with increase in dopant concentration of TiO2 and ZnO up to around 10<sup>18</sup> cm-3 due to the increase in conductivity of ETM. Although dopant concentrations have been increasing beyond this value PV parameters remain unchanged due to Moss-Burstein effect [50].Variation of dopant concentration of TiO2 is found to be little bit more sensitive than that of ZnO up to around 10<sup>18</sup> cm-3 due to

Figure 28. The study has been carried out from 10<sup>15</sup> to 10<sup>21</sup> cm-3 values. Both ETMs have exhibited the similar performance here too. Voc, FF and PCE have been increasing with increase in dopant concentration of TiO2 and ZnO up to around 10<sup>18</sup> cm-3 due to the increase in conductivity of ETM. Although dopant concentrations have been increasing beyond this value PV parameters remain unchanged due to Moss-Burstein effect [50].Variation of dopant concentration of TiO2 is found to be little bit more sensitive than that of ZnO up to around 10<sup>18</sup> cm-3 due to

Figure 26. Interface Trap of Perovskite with TiO<sup>2</sup>

Figure 27. Interface Trap of Perovskite with ZnO **Figure 27.** Interface Trap of Perovskite with ZnO

## *4.2.4. Role of Dopant Concentrations of ETMs.*

sensitiveness of ZnO than TiO2 towards interface trap indicates the relatively small defect and small electron mobility of TiO2. Optimizing doping of the ETM, and formation of homogene‐ ous, smooth, and flat surface will significantly reduce the interface trap density and hence

0.0 0.2 0.4 0.6 0.8 1.0 1.2

0.0 0.2 0.4 0.6 0.8 1.0 1.2

0.0 0.2 0.4 0.6 0.8 1.0 1.2

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Volatage (V)

Volatage (V)

4.2.4. Role of Dopant Concentrations of ETMs.

4.2.4. Role of Dopant Concentrations of ETMs.

smaller value of electron mobility and hence small conductivity of TiO2.

smaller value of electron mobility and hence small conductivity of TiO2.

The plot of PV cell parameters as a function of dopant concentration (ND) of both ETMs TiO2 and ZnO is shown in

The plot of PV cell parameters as a function of dopant concentration (ND) of both ETMs TiO2 and ZnO is shown in

Figure 28. The study has been carried out from 10<sup>15</sup> to 10<sup>21</sup> cm-3 values. Both ETMs have exhibited the similar performance here too. Voc, FF and PCE have been increasing with increase in dopant concentration of TiO2 and ZnO up to around 10<sup>18</sup> cm-3 due to the increase in conductivity of ETM. Although dopant concentrations have been increasing beyond this value PV parameters remain unchanged due to Moss-Burstein effect [50].Variation of dopant concentration of TiO2 is found to be little bit more sensitive than that of ZnO up to around 10<sup>18</sup> cm-3 due to

Figure 28. The study has been carried out from 10<sup>15</sup> to 10<sup>21</sup> cm-3 values. Both ETMs have exhibited the similar performance here too. Voc, FF and PCE have been increasing with increase in dopant concentration of TiO2 and ZnO up to around 10<sup>18</sup> cm-3 due to the increase in conductivity of ETM. Although dopant concentrations have been increasing beyond this value PV parameters remain unchanged due to Moss-Burstein effect [50].Variation of dopant concentration of TiO2 is found to be little bit more sensitive than that of ZnO up to around 10<sup>18</sup> cm-3 due to

Figure 26. Interface Trap of Perovskite with TiO<sup>2</sup>

Figure 26. Interface Trap of Perovskite with TiO<sup>2</sup>

 10<sup>9</sup> 10<sup>10</sup> 10<sup>11</sup> 10<sup>12</sup> 10<sup>13</sup> 10<sup>14</sup> 10<sup>15</sup>

 10<sup>9</sup> 10<sup>10</sup> 10<sup>11</sup> 10<sup>12</sup> 10<sup>13</sup> 10<sup>14</sup> 10<sup>15</sup>

Figure 27. Interface Trap of Perovskite with ZnO

Figure 27. Interface Trap of Perovskite with ZnO

Interface Trap-ZnO

Interface Trap-ZnO

Voltage (V)

Voltage (V)

increase the performance.












JSC (mA/cm2

JSC (mA/cm2

**Figure 27.** Interface Trap of Perovskite with ZnO

)

)

**Figure 26.** Interface Trap of Perovskite with TiO2





10<sup>8</sup>

10<sup>8</sup>



0

0



JSC (mA/cm2

JSC (mA/cm2

)

)



10<sup>8</sup>

10<sup>8</sup>

Interface Trap-TiO<sup>2</sup>

Interface Trap-TiO<sup>2</sup>

 10<sup>9</sup> 10<sup>10</sup> 10<sup>11</sup> 10<sup>12</sup> 10<sup>13</sup> 10<sup>14</sup> 10<sup>15</sup>

 10<sup>9</sup> 10<sup>10</sup> 10<sup>11</sup> 10<sup>12</sup> 10<sup>13</sup> 10<sup>14</sup> 10<sup>15</sup>




0

0

474 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The plot of PV cell parameters as a function of dopant concentration (ND) of both ETMs TiO2 and ZnO is shown in Figure 28. The study has been carried out from 1015 to 1021 cm-3 values. Both ETMs have exhibited the similar performance here too. Voc, FF and PCE have been increasing with increase in dopant concentration of TiO2 and ZnO up to around 1018 cm-3 due to the increase in conductivity of ETM. Although dopant concentrations have been increasing beyond this value PV parameters remain unchanged due to Moss-Burstein effect [50].Variation of dopant concentration of TiO2 is found to be little bit more sensitive than that of ZnO up to around 1018 cm-3 due to smaller value of electron mobility and hence small conductivity of TiO2.

4.2.5. Interface Trap Electron/Hole Capture Cross-Section. **Figure 28.** PVC Parameters vs Dopant concentrations (ND) of TiO2 and ZnO

#### of the device where the value varies. The electron/hole capture cross-sections have been varied from 10-14/10-15 and *4.2.5. Interface Trap Electron/Hole Capture Cross-Section.*

0.8

4.3. Role of HTM

4.3.1. Role of thickness of the HTM

0.9

1.0

Voltage (Volt)

1.1

1.2

10-15/10-14 to 10-20/10-21 and 10-21/10-20 cm<sup>2</sup> for first and second interfaces respectively. Plots show that VOC and PCE both are decreasing with increase in cross-section of TiO2 and ZnO due to increase in recombination/loss and change in shunt resistance. ZnO is observed more sensitive than that of TiO2 toward the higher values of interface trap electron/hole capture cross-sections. It is suggested that ZnO has such more defects than TiO2 and hence less efficient in real practice. 24 Figure 29 shows the effect of interface trap electron/hole capture cross-section for both interfaces on performance of the device where the value varies. The electron/hole capture cross-sections have been varied from 10-14/10-15 and 10-15/10-14 to 10-20/10-21 and 10-21/10-20 cm2 for first and second interfaces respectively. Plots show that VOC and PCE both are decreasing with increase in cross-section of TiO2 and ZnO due to increase in recombination/loss and change in shunt resistance. ZnO is observed more sensitive than that of TiO2 toward the higher values of interface trap electron/hole capture cross-sections. It is suggested that ZnO has such more defects than TiO2 and hence less efficient in real practice.

> PCE of TiO<sup>2</sup> PCE of ZnO

1E-14/1E-15 1E-16/1E-17 1E-18/1E-19 1E-20/1E-21

Interface Trap-Capture Cross-section (cm<sup>2</sup>

Figure 29. VOC and PCE vs Interface Trap Electron/Hole Capture Cross-section.

Figures 30 is the plot of FF and PCE versus practically viable thickness of the

Figure 28. PVC Parameters vs Dopant concentrations (ND) of TiO2 and ZnO

Figure 29 shows the effect of interface trap electron/hole capture cross-section for both interfaces on performance

 VOC of TiO<sup>2</sup> VOC of ZnO

)

14

16

18

PCE (%)

20

22

1.110 1.115 1.120 1.125 1.130

VOC (V)

JSC (mA/cm2

)

24.5 25.0 25.5

PCE (%0

FF (%)

10<sup>15</sup> 10<sup>16</sup> 10<sup>17</sup> 10<sup>18</sup> 10<sup>19</sup> 10<sup>20</sup> 10<sup>21</sup>

Dopant concentration (ND) of ETM (cm-3)

Figure 28. PVC Parameters vs Dopant concentrations (ND) of TiO2 and ZnO

4.2.5. Interface Trap Electron/Hole Capture Cross-Section.

 TiO<sup>2</sup> ZnO

> TiO<sup>2</sup> ZnO

 TiO<sup>2</sup> ZnO

Figure 29 shows the effect of interface trap electron/hole capture cross-section for both interfaces on performance of the device where the value varies. The electron/hole capture cross-sections have been varied from 10-14/10-15 and 10-15/10-14 to 10-20/10-21 and 10-21/10-20 cm<sup>2</sup> for first and second interfaces respectively. Plots show that VOC and PCE both are decreasing with increase in cross-section of TiO2 and ZnO due to increase in recombination/loss and change in shunt resistance. ZnO is observed more sensitive than that of TiO2 toward the higher values of interface trap electron/hole capture cross-sections. It is suggested that ZnO has such more defects than TiO2 and hence less

 TiO<sup>2</sup> ZnO

Figure 29. VOC and PCE vs Interface Trap Electron/Hole Capture Cross-section. **Figure 29.** VOC and PCE vs Interface Trap Electron/Hole Capture Cross-section.

#### **4.3. Role of HTM** 4.3.1. Role of thickness of the HTM

#### *4.3.1. Role of thickness of the HTM* Figures 30 is the plot of FF and PCE versus practically viable thickness of the

4.3. Role of HTM

Figures 30 is the plot of FF and PCE versus practically viable thickness of the

Figure 30. Variation of FF and PCE with thickness of HTM. **Figure 30.** Variation of FF and PCE with thickness of HTM.

recombination and resistance of the device.

highly appreciable for better performance of the device.

0.0 0.2 0.4 0.6 0.8 1.0 1.2

 1E-4 2E-4 5E-4 1E-3 2E-3 5E-3

Hole mobility of SpiroMeOTAD

1E-2

at 3.0×10<sup>16</sup> cm-3 N<sup>A</sup>

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Voltage (V)

Figure 31. J-V curve as a function of hole mobolity.


JSc (mA/cm2

)

suggested the increase in recombination and resistance. Thus, a superb junction diode like contact between the absorbar and HTM is necessary to increase the fill factor and hence to improve the efficiency of the device. It is obvious that the decrease in thickness of the HTM improves the performance of the solar cells due to decrease in HTM (Spiro-MeOTAD). In the case of HTM, fill factor and PCE both are decreasing with increase in thickness suggested the increase in recombination and resistance. Thus, a superb junction diode like contact between the absorbar and HTM is necessary to increase the fill

4.3.2. Role of hole mobility and acceptor concentration of the HTM

HTM (Spiro-MeOTAD). In the case of HTM, fill factor and PCE both are decreasing with increase in thickness

Figures 31 and 32 show the effect of hole mobolity and acceptor contrantration of HTM respectively on the device performance. The Spiro-MeOTAD has relatively low hole mobility value. Note that different researcher [34, 45, 51] have reported different values of hole mobility of SpiroMeOTAD but 2.0×10-4 cm<sup>2</sup>/V·s has been used in this study. Due to small value of hole mobility and acceptor concentration low value of FF and hence the low PCE are observed. The value goes on increasing with increase in hole mobility and acceptor concentration. Properties such as hole mobility and dopant concentration (NA), are responsible for resistance/conductance of the HTM, here Spiro-MeOTAD, they highly influence the performance of the device. Although Spiro-MeOTAD has merits either enhencement of conductivity and dopant concentration of this material or replacement of it by suitable HTM is

suggested the increase in recombination and resistance. Thus, a superb junction diode like contact between the

100 200 300 400 500 600

Thickness (nm)

factor and hence to improve the efficiency of the device. It is obvious that the decrease in thickness of the HTM improves the performance of the solar cells due to decrease in recom‐ bination and resistance of the device. absorbar and HTM is necessary to increase the fill factor and hence to improve the efficiency of the device. It is obvious that the decrease in thickness of the HTM improves the performance of the solar cells due to decrease in recombination and resistance of the device.

Figure 30. Variation of FF and PCE with thickness of HTM.

### *4.3.2. Role of hole mobility and acceptor concentration of the HTM* 4.3.2. Role of hole mobility and acceptor concentration of the HTM

81.5

22.5 22.6 22.7 22.8 22.9

82.0

FF (%)

PCE (%)

82.5

Figures 31 and 32 show the effect of hole mobolity and acceptor contrantration of HTM respectively on the device performance. The Spiro-MeOTAD has relatively low hole mobility value. Note that different researcher [34, 45, 51] have reported different values of hole mobility of SpiroMeOTAD but 2.0×10-4 cm2 /V s has been used in this study. Due to small value of hole mobility and acceptor concentration low value of FF and hence the low PCE are observed. The value goes on increasing with increase in hole mobility and acceptor concentration. Properties such as hole mobility and dopant concentration (NA), are responsible for resistance/conduc‐ tance of the HTM, here Spiro-MeOTAD, they highly influence the performance of the device. Although Spiro-MeOTAD has merits either enhencement of conductivity and dopant concen‐ tration of this material or replacement of it by suitable HTM is highly appreciable for better performance of the device. Figures 31 and 32 show the effect of hole mobolity and acceptor contrantration of HTM respectively on the device performance. The Spiro-MeOTAD has relatively low hole mobility value. Note that different researcher [34, 45, 51] have reported different values of hole mobility of SpiroMeOTAD but 2.0×10-4 cm<sup>2</sup>/V·s has been used in this study. Due to small value of hole mobility and acceptor concentration low value of FF and hence the low PCE are observed. The value goes on increasing with increase in hole mobility and acceptor concentration. Properties such as hole mobility and dopant concentration (NA), are responsible for resistance/conductance of the HTM, here Spiro-MeOTAD, they highly influence the performance of the device. Although Spiro-MeOTAD has merits either enhencement of conductivity and dopant concentration of this material or replacement of it by suitable HTM is highly appreciable for better performance of the device.

Figure 31. J-V curve as a function of hole mobolity. **Figure 31.** J-V curve as a function of hole mobolity.

## **4.4. Role of front and back contacts**

**4.3. Role of HTM**

*4.3.1. Role of thickness of the HTM*

4.3. Role of HTM

0.8

0.9

1.0

Voltage (Volt)

1.1

1.2

1.110 1.115 1.120 1.125 1.130

VOC (V)

JSC (mA/cm2

)

24.5 25.0 25.5

efficient in real practice.

476 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

PCE (%0

FF (%)

FF (%)

PCE (%)

81.5

22.5 22.6 22.7 22.8 22.9

**Figure 30.** Variation of FF and PCE with thickness of HTM.


JSc (mA/cm2

)

82.0

82.5

4.3.1. Role of thickness of the HTM

**Figure 29.** VOC and PCE vs Interface Trap Electron/Hole Capture Cross-section.

Figures 30 is the plot of FF and PCE versus practically viable thickness of the

Figures 30 is the plot of FF and PCE versus practically viable thickness of the

Figure 29. VOC and PCE vs Interface Trap Electron/Hole Capture Cross-section.

1E-14/1E-15 1E-16/1E-17 1E-18/1E-19 1E-20/1E-21

Interface Trap-Capture Cross-section (cm<sup>2</sup>

10<sup>15</sup> 10<sup>16</sup> 10<sup>17</sup> 10<sup>18</sup> 10<sup>19</sup> 10<sup>20</sup> 10<sup>21</sup>

Dopant concentration (ND) of ETM (cm-3

Figure 28. PVC Parameters vs Dopant concentrations (ND) of TiO2 and ZnO

 PCE of TiO<sup>2</sup> PCE of ZnO

4.2.5. Interface Trap Electron/Hole Capture Cross-Section.

 TiO<sup>2</sup> ZnO

> TiO<sup>2</sup> ZnO

 TiO<sup>2</sup> ZnO

 TiO<sup>2</sup> ZnO

)

Figure 29 shows the effect of interface trap electron/hole capture cross-section for both interfaces on performance of the device where the value varies. The electron/hole capture cross-sections have been varied from 10-14/10-15 and 10-15/10-14 to 10-20/10-21 and 10-21/10-20 cm<sup>2</sup> for first and second interfaces respectively. Plots show that VOC and PCE both are decreasing with increase in cross-section of TiO2 and ZnO due to increase in recombination/loss and change in shunt resistance. ZnO is observed more sensitive than that of TiO2 toward the higher values of interface trap electron/hole capture cross-sections. It is suggested that ZnO has such more defects than TiO2 and hence less

> VOC of TiO<sup>2</sup> VOC of ZnO

> > )

14

16

18

PCE (%)

20

22

24

Figure 30. Variation of FF and PCE with thickness of HTM.

highly appreciable for better performance of the device.

0.0 0.2 0.4 0.6 0.8 1.0 1.2

 1E-4 2E-4 5E-4 1E-3 2E-3 5E-3

Hole mobility of SpiroMeOTAD

N<sup>A</sup>

1E-2

at 3.0×10<sup>16</sup> cm-3

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Voltage (V)

Figure 31. J-V curve as a function of hole mobolity.

recombination and resistance of the device.

100 200 300 400 500 600

Thickness (nm)

4.3.2. Role of hole mobility and acceptor concentration of the HTM

HTM (Spiro-MeOTAD). In the case of HTM, fill factor and PCE both are decreasing with increase in thickness suggested the increase in recombination and resistance. Thus, a superb junction diode like contact between the absorbar and HTM is necessary to increase the fill

HTM (Spiro-MeOTAD). In the case of HTM, fill factor and PCE both are decreasing with increase in thickness suggested the increase in recombination and resistance. Thus, a superb junction diode like contact between the absorbar and HTM is necessary to increase the fill factor and hence to improve the efficiency of the device. It is obvious that the decrease in thickness of the HTM improves the performance of the solar cells due to decrease in

Figures 31 and 32 show the effect of hole mobolity and acceptor contrantration of HTM respectively on the device performance. The Spiro-MeOTAD has relatively low hole mobility value. Note that different researcher [34, 45, 51] have reported different values of hole mobility of SpiroMeOTAD but 2.0×10-4 cm<sup>2</sup>/V·s has been used in this study. Due to small value of hole mobility and acceptor concentration low value of FF and hence the low PCE are observed. The value goes on increasing with increase in hole mobility and acceptor concentration. Properties such as hole mobility and dopant concentration (NA), are responsible for resistance/conductance of the HTM, here Spiro-MeOTAD, they highly influence the performance of the device. Although Spiro-MeOTAD has merits either enhencement of conductivity and dopant concentration of this material or replacement of it by suitable HTM is

## *4.4.1. Role of workfunction of the back contact metal*

In this case, simulations were carried out at thicknesses 400/400/90 nm of the device layers so as to compare the outcome with references [45, 46]. Figure 33 is the plot of fill factor and PCE versus workfunction of the counter electrode. It is found that fill factor and PCE both are -20



)



0

N

at 1.0×10-4

hole mobility

<sup>A</sup> of Spiro-MeOTAD

 cm 2 V-1 s -1

3E+16 3E+17

0.0 0.2 0.4 0.6 0.8 1.0 1.2

3E+20 3E+21

electrode. It is found that fill factor and PCE both are decreasing with decrease in workfunction of the back Figure 32. J-V curve as a function of dopant concentrations(NA). **Figure 32.** J-V curve as a function of dopant concentrations (NA).

decreasing with decrease in workfunction of the back contact metal. Back contact is the counter electrode connected to the HTM to collects the holes or to enter the almost relaxed electrons into the device from the external circuit. Figure 12 shows the energy band alignment between different layers of the device. almost relaxed electrons into the device from the external circuit. Figure 12 shows the energy band alignment between different layers of the device. 4.4. Role of front and back contacts 4.4.1. Role of workfunction of the back contact metal

contact metal. Back contact is the counter electrode connected to the HTM to collects the holes or to enter the

In this case, simulations were carried out at thicknesses 400/400/90 nm of the device layers so as to compare the

Figure 33. FF and PCE vs workfunction of back contact.


Work function (eV)

Figure 33. FF and PCE vs workfunction of back contact.

3.6 3.8 4.0 4.2 4.4 4.6

Workfunction (eV)

3.6 3.8 4.0 4.2 4.4 4.6

Workfunction (eV)

Figure 34. PV all parameters vs workfunction of front contact.

Figure 34. PV all parameters vs workfunction of front contact.

WF

WF

15

**Figure 33.** FF and PCE vs workfunction of back contact. 60

64

1.08 1.14 1.20

VOC (V)

)

JSC (mA/cm2

24.4 24.6 24.8

> )

VOC (V)

1.08 1.14 1.20

24.4 24.6 24.8

> 72 76 80

> 20 22

72 76 80

JSC (mA/cm2

FF (%)

PCE (%)

20 22

FF (%)

PCE (%)

15

18

PCE (%)

21

24

In this case, simulations were carried out at thicknesses 400/400/90 nm of the device layers so as to compare the outcome with others [45, 46]. Figure 33 is the plot of fill factor and PCE versus workfunction of the counter electrode. It is found that fill factor and PCE both are decreasing with decrease in workfunction of the back contact metal. Back contact is the counter electrode connected to the HTM to collects the holes or to enter the almost relaxed electrons into the device from the external circuit. Figure 12 shows the energy band alignment


Work function (eV)

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Figure 32. J-V curve as a function of dopant concentrations(NA).

4.4.1. Role of workfunction of the back contact metal

4.4. Role of front and back contacts

between different layers of the device.

Fill Factor

Efficiency (PCE)

Voltage (V)

 3E+16 3E+17 3E+18 3E+19 3E+20 3E+21

0.0 0.2 0.4 0.6 0.8 1.0 1.2

NA of Spiro-MeOTAD

 cm<sup>2</sup> V-1 s -1

at 1.0×10-4

hole mobility




JSC (mA/cm2

)




0

electrode. It is found that fill factor and PCE both are decreasing with decrease in workfunction of the back Figure 34. PV all parameters vs workfunction of front contact. **Figure 34.** PV all parameters vs workfunction of front contact.

60

64

68

FF (%)

72

76

80

84

decreasing with decrease in workfunction of the back contact metal. Back contact is the counter electrode connected to the HTM to collects the holes or to enter the almost relaxed electrons into the device from the external circuit. Figure 12 shows the energy band alignment between

4.4. Role of front and back contacts


15

PCE (%)

15

18

21

24

18

PCE (%)

21

24

Work function (eV)

Figure 33. FF and PCE vs workfunction of back contact.


Work function (eV)

Figure 33. FF and PCE vs workfunction of back contact.

3.6 3.8 4.0 4.2 4.4 4.6

Workfunction (eV)

3.6 3.8 4.0 4.2 4.4 4.6

Workfunction (eV)

Figure 34. PV all parameters vs workfunction of front contact.

Figure 34. PV all parameters vs workfunction of front contact.

WF

WF

between different layers of the device.

Fill Factor

Efficiency (PCE)

between different layers of the device.

Fill Factor

Efficiency (PCE)

0.0 0.2 0.4 0.6 0.8 1.0 1.2

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Figure 32. J-V curve as a function of dopant concentrations(NA).

 3E+16 3E+17 3E+18 3E+19 3E+20 3E+21

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Figure 32. J-V curve as a function of dopant concentrations(NA).

4.4.1. Role of workfunction of the back contact metal

Voltage (V)

4.4.1. Role of workfunction of the back contact metal

outcome with others [45, 46]. Figure 33 is the plot of fill factor and PCE versus workfunction of the counter

4.4. Role of front and back contacts

NA of Spiro-MeOTAD

 cm<sup>2</sup> V-1 s -1

at 1.0×10-4

hole mobility

Voltage (V)

 3E+16 3E+17 3E+18 3E+19 3E+20 3E+21

0.0 0.2 0.4 0.6 0.8 1.0 1.2

different layers of the device.

60

**Figure 33.** FF and PCE vs workfunction of back contact.

FF (%)

60

64

68

72

76

80

84

1.08 1.14 1.20

VOC (V)

)

JSC (mA/cm2

24.4 24.6 24.8

> )

VOC (V)

1.08 1.14 1.20

24.4 24.6 24.8

> 72 76 80

> 20 22

72 76 80

JSC (mA/cm2

FF (%)

PCE (%)

20 22

FF (%)

PCE (%)

64

68

FF (%)

72

76

80

84



**Figure 32.** J-V curve as a function of dopant concentrations (NA).



JSC (mA/cm2

)




0

478 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications



JSC (mA/cm2

)




0

N

at 1.0×10-4

hole mobility

<sup>A</sup> of Spiro-MeOTAD

 cm 2 V-1 s -1

contact metal. Back contact is the counter electrode connected to the HTM to collects the holes or to enter the almost relaxed electrons into the device from the external circuit. Figure 12 shows the energy band alignment In this case, simulations were carried out at thicknesses 400/400/90 nm of the device layers so as to compare the outcome with others [45, 46]. Figure 33 is the plot of fill factor and PCE versus workfunction of the counter electrode. It is found that fill factor and PCE both are decreasing with decrease in workfunction of the back contact metal. Back contact is the counter electrode connected to the HTM to collects the holes or to enter the almost relaxed electrons into the device from the external circuit. Figure 12 shows the energy band alignment An ohmic contact between them is necessary to transport the holes efficiently to the back contact. As workfunction of back contact is 5.0 eV or above than the HTM i.e., workfunction of the back contact is nearly equal to or slightly greater than that of HTM then holes transport effectively and there would be barrier for electrons. Below 5.0 eV a dipole oriented positive in the metal and negative in the HTM results an electrostatic barrier to the holes or Schottky barrier [9] and hence difficultly arise in holes transportation at contact. This indicates the requirement of gold (workfunction = 5.1 eV) like higher workfunction material to develop ohmic contact to conduct the hole to the electrode [44, 45, 47]. Moreover, it is quite clear from the Figure 33 that, at workfunction 4.64 eV i.e., of Ag [52], PCE is decreased to about 18 % with VOC = 1.12 V,JSC = 24.32 mA/cm2 and FF = 69 %. This result has close agreement with the experimentally determined value 15. 4% PCE, 1.07 V VOC, 21.5 mA/cm2 JSC and 67% FF for MAPbI3 perovskite based planner solar cell with the Ag as counter electrode [46]. Thus, our result claims the realism in performance of device than reported by [45]. Consequently, it always opens the pathway to replace the Au-counter electrode if someone is focused on the cost. Besides, there are other options too with favorable high value workfunction material to replace the counter electrode [53].

### *4.4.2. Role of workfunction of the front contact*

For the sake of convenient and to make real solar cells simulations were carried out at thicknesses 400/400/90 nm of the device layers*.* Figure 34 is the the PV all parameters versus workfunction of the electrode/front contact. It is found that fill factor and PCE both are increasing with decrease in workfunction of the front contact material. Front contact is the electrode connected to the ETM to collects the energetic electrons to go round the external circuit to do some useful work over there and at the mean time it allows the light to reach the absorber. Figure 12 shows the energy band alignment between different layers of the device. An ohmic contact between them is necessary to transport the electrons efficiently to the front contact. As workfunction of front contact is slightly smaller than or nearly equal to the workfunction of the ETM, there is ohmic contact for the electrons and barriers for the holes. Thus, a comparatively low workfunction and transparent material is suitable to use as front contact material. value workfunction material to replace the counter electrode [53]. 4.4.2. Role of workfunction of the front contact For the sake of convenient and to make real solar cells simulations were carried out at thicknesses 400/400/90 nm of the device layers. Figure 34 is the the PV all parameters versus workfunction of the electrode/front contact. It is found that fill factor and PCE both are increasing with decrease in workfunction of the front contact material. Front contact is the electrode connected to the ETM to collects the energetic electrons to go round the external circuit to do some useful work over there and at the mean time it allows the light to reach the absorber. Figure 12 shows the energy band alignment between different layers of the device. An ohmic contact between them is

An ohmic contact between them is necessary to transport the holes efficiently to the back contact. As workfunction of back contact is 5.0 eV or above than the HTM i.e., workfunction of the back contact is nearly equal to or slightly greater than that of HTM then holes transport effectively and there would be barrier for electrons. Below 5.0 eV a dipole oriented positive in the metal and negative in the HTM results an electrostatic barrier to the holes [9] and hence difficultly arise in holes transportation at contact. This indicates the requirement of gold (workfunction = 5.1 eV) like higher workfunction material to develop ohmic contact to conduct the hole to the electrode [44, 45, 47]. Moreover, it is quite clear from the Figure 33 that, at workfunction 4.64 eV i.e., of Ag [52], PCE is decreased to about 18 % with VOC = 1.12 V,JSC = 24.32 mA/cm<sup>2</sup> and FF = 69 %. This result has close

realism in performance of device than reported by [45]. Consequently, it always opens the pathway to replace the Au-counter electrode if someone is focused on the cost. Besides, there are other options too with favorable high

necessary to transport the electrons efficiently to the front contact. As workfunction of front contact is slightly

#### **4.5. Role of temperature** smaller than or nearly equal to the workfunction of the ETM, there is ohmic contact for the electrons and barriers for the holes. Thus, a comparatively low workfunction and transparent material is suitable to use as front contact

Usually in simulation studies and fabrications 300K is chosen as the operating temperature where it is regarding as the room temperature. In actual practice the operating temperature varies with latitude, altitude, day of the year and time of the day of the place concerned. Moreover, performance of a solar cell is affected by the operating temperature as well. In this context we have varied the temperature from 277K to 410K to realize actual behavior of the perovskite based solar cells. material. 4.5. Role of temperature Usually in simulation studies and fabrications 300K is chosen as the operating temperature where it is regarding as the room temperature. In actual practice the operating temperature varies with latitude, altitude, day of the year and time of the day of the place concerned. Moreover, performance of a solar cell is affected by the operating temperature as well. In this context we have varied the temperature from 277K to 410K to realize actual behavior of the perovskite based solar cells.

**Figure 35.** Variation of PV parameters with temperature (K)

Figure 35 shows the variation of PV parameters with variation of operating temperature of the solar cells where performance of the solar cells is influenced by operating temperature. At temperature lower than room temperature the solar cells have exhibited higher performance and at room temperature all PV parameters have moderate value. The temperatures above room temperature have detrimental to the overall performance. At higher temperature the carrier concentrations, mobility of the charge carries, resistance and band gap of the materials would be greatly affected that would ultimately alter PV parameters. Figure 36, JV curve as a function of temperature, conforms the variation of VOC, FF and hence resistance in the solar cell devices.

Figure 36. JV curve as a function of temperature **Figure 36.** JV curve as a function of temperature

absorber. Figure 12 shows the energy band alignment between different layers of the device. An ohmic contact between them is necessary to transport the electrons efficiently to the front contact. As workfunction of front contact is slightly smaller than or nearly equal to the workfunction of the ETM, there is ohmic contact for the electrons and barriers for the holes. Thus, a comparatively low workfunction and transparent material is suitable to use as front

For the sake of convenient and to make real solar cells simulations were carried out at thicknesses 400/400/90 nm of the device layers. Figure 34 is the the PV all parameters versus workfunction of the electrode/front contact. It is found that fill factor and PCE both are increasing with decrease in workfunction of the front contact material. Front contact is the electrode connected to the ETM to collects the energetic electrons to go round the external circuit to do some useful work over there and at the mean time it allows the light to reach the absorber. Figure 12 shows the energy band alignment between different layers of the device. An ohmic contact between them is necessary to transport the electrons efficiently to the front contact. As workfunction of front contact is slightly smaller than or nearly equal to the workfunction of the ETM, there is ohmic contact for the electrons and barriers for the holes. Thus, a comparatively low workfunction and transparent material is suitable to use as front contact

Usually in simulation studies and fabrications 300K is chosen as the operating temperature where it is regarding as the room temperature. In actual practice the operating temperature varies with latitude, altitude, day of the year and time of the day of the place concerned. Moreover, performance of a solar cell is affected by the operating temperature as well. In this context we have varied the temperature from 277K to 410K to realize actual behavior

value workfunction material to replace the counter electrode [53].

4.4.2. Role of workfunction of the front contact

480 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

An ohmic contact between them is necessary to transport the holes efficiently to the back contact. As workfunction of back contact is 5.0 eV or above than the HTM i.e., workfunction of the back contact is nearly equal to or slightly greater than that of HTM then holes transport effectively and there would be barrier for electrons. Below 5.0 eV a dipole oriented positive in the metal and negative in the HTM results an electrostatic barrier to the holes [9] and hence difficultly arise in holes transportation at contact. This indicates the requirement of gold (workfunction = 5.1 eV) like higher workfunction material to develop ohmic contact to conduct the hole to the electrode [44, 45, 47]. Moreover, it is quite clear from the Figure 33 that, at workfunction 4.64 eV i.e., of Ag [52], PCE is decreased to about 18 % with VOC = 1.12 V,JSC = 24.32 mA/cm<sup>2</sup> and FF = 69 %. This result has close agreement with the experimentally determined value 15. 4% PCE, 1.07 V VOC, 21.5 mA/cm<sup>2</sup>JSC and 67% FF for MAPbI3 perovskite based planner solar cell with the Ag as counter electrode [46]. Thus, our result claims the realism in performance of device than reported by [45]. Consequently, it always opens the pathway to replace the Au-counter electrode if someone is focused on the cost. Besides, there are other options too with favorable high

Usually in simulation studies and fabrications 300K is chosen as the operating temperature where it is regarding as the room temperature. In actual practice the operating temperature varies with latitude, altitude, day of the year and time of the day of the place concerned. Moreover, performance of a solar cell is affected by the operating temperature as well. In this context we have varied the temperature from 277K to 410K to realize actual behavior of the

280 300 320 340 360 380 400

Figure 35 shows the variation of PV parameters with variation of operating temperature of the solar cells where performance of the solar cells is influenced by operating temperature. At temperature lower than room temperature the solar cells have exhibited higher performance and at room temperature all PV parameters have moderate value. The temperatures above room temperature have detrimental to the overall performance. At higher temperature the carrier concentrations, mobility of the charge carries, resistance and band gap of the materials would be greatly affected that would ultimately alter PV parameters. Figure 36, JV curve as a function of temperature, conforms the variation of VOC, FF and hence resistance in the solar

A

A

Temperature (K)

contact material.

**4.5. Role of temperature**

material.

4.5. Role of temperature

of the perovskite based solar cells.

1.02 1.08 1.14

VOC (V)

)

JSC (mAc/m2

24.31 24.32 24.33 24.34

FF (%)

PCE (%)

cell devices.

19.5 21.0 22.5

**Figure 35.** Variation of PV parameters with temperature (K)

perovskite based solar cells.

#### **4.6. Concentrator solar cells and multi-Sun** Figure 35 shows the variation of PV parameters with variation of operating temperature of the solar cells where performance of the solar cells is influenced by operating temperature. At temperature lower than room

Figure 37 is the solar irradiance at air mass zero (AM0, i.e., extraterrestrial irradiance) and air mass 1.5 global (AM1.5G, terrestrial irradiance). When sunlight enters the Earth's atmosphere then there is absorption, transmission, reflection, and scattering the light by ozone, oxygen, water in different phases, carbon dioxide and atmospheric contents. As a consequence the extraterrestrial radiation greatly modifies into global (direct plus diffuse) radiation on the Earth's surface. Extraterrestrial radiation is concerned mainly to the PV devices in space where as global is significant to the terrestrial activities including PV solar cell devices. temperature the solar cells have exhibited higher performance and at room temperature all PV parameters have moderate value. The temperatures above room temperature have detrimental to the overall performance. At higher temperature the carrier concentrations, mobility of the charge carries, resistance and band gap of the materials would be greatly affected that would ultimately alter PV parameters. Figure 36, JV curve as a function of temperature, conforms the variation of VOC, FF and hence resistance in the solar cell devices. 4.6. Concentrator solar cells and multi-Sun Figure 37 is the solar irradiance at air mass zero (AM0, i.e., extraterrestrial irradiance) and air mass 1.5 global (AM1.5G, terrestrial irradiance). When sunlight enters the Earth's atmosphere then there is absorption,

atmospheric contents. As a consequence the extraterrestrial radiation greatly modifies into global (direct plus

transmission, reflection, and scattering the light by ozone, oxygen, water in different phases, carbon dioxide and

**Figure 37.** Solar irradiance vs wavelength for different air mass. Figure 37. Solar irradiance vs wavelength for different air mass.

Figure 38. PV parameters vs Air Mass. **Figure 38.** PV parameters vs Air Mass.

Figure 38 shows the variation of all PV parameters with solar spectrum in different air mass for perovskite as absorber. Efficiency is maximum at AM1.5G. Fill factor is went regularly decreasing from AM1.5D to AM0. Intense light would be changed the temperature, carrier concentrations, band gap, mobility of the charge carriers and resistance they would ultimately change the performance of the PV devices. Figure 38 shows the variation of all PV parameters with solar spectrum in different air mass for perovskite as absorber. Efficiency is maximum at AM1.5G. Fill factor is went regularly decreasing from AM1.5D to AM0. Intense light would be changed the temperature, carrier concentrations, band gap, mobility of the charge carriers and resistance they would ultimately change the performance of the PV devices.

Figure 39. Solar spectrums at different Suns.

0 10 20 30 40 50 60 70 80 90 100 110

Suns

**Figure 39.** Solar spectrums at different Suns.

1.1 1.2 1.3 1.4

VOC (V)

JSC (mA/cm2

)

0 1000 2000

FF (%)

PCE (%)

Figure 40. PV parameters vs Suns.

Figure 38 shows the variation of all PV parameters with solar spectrum in different air mass for perovskite as absorber. Efficiency is maximum at AM1.5G. Fill factor is went regularly decreasing from AM1.5D to AM0.

Intense light would be changed the temperature, carrier concentrations, band gap, mobility of the charge carriers

Variation of solar irradiance for different 'Sun' is illustrated in figure 39. An array of mirrors or lenses have been employed in order to focus onto a small sized absorber which is thereby subjected to intense sunlight for the

generation of energy in one form to another [54] this is so called 'multi-sun' which enhances efficiency and reduce

Variation of solar irradiance for different 'Sun' is illustrated in figure 39. An array of mirrors or lenses have been employed in order to focus onto a small sized absorber which is thereby subjected to intense sunlight for the generation of energy in one form to another [54] this is so called 'multi-sun' which enhances efficiency and reduce the size of the solar cell devices and the PV device is called concentrator solar cells. Figure 39. Solar spectrums at different Suns.

1.5 Direct 1.5 Global 0

Figure 38. PV parameters vs Air Mass.

Air Mass

and resistance they would ultimately change the performance of the PV devices.

the size of the solar cell devices and the PV device is called concentrator solar cells.

Figure 40. PV parameters vs Suns. **Figure 40.** PV parameters vs Suns.

1.10 1.11 1.12 1.13

VOC (V)

JSC (mA/cm2

FF (%)

)

81.6 82.0 82.4

> 21 22

PCE (%)

1.10 1.11 1.12 1.13

482 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

VOC (V)

JSC (mA/cm2

FF (%)

)

81.6 82.0 82.4

> 21 22

change the performance of the PV devices.

PCE (%)

**Figure 38.** PV parameters vs Air Mass.

1.5 Direct 1.5 Global 0

Figure 38 shows the variation of all PV parameters with solar spectrum in different air mass for perovskite as absorber. Efficiency is maximum at AM1.5G. Fill factor is went regularly decreasing from AM1.5D to AM0. Intense light would be changed the temperature, carrier concentrations, band gap, mobility of the charge carriers and resistance they would ultimately

Figure 38. PV parameters vs Air Mass.

Figure 39. Solar spectrums at different Suns.

0 10 20 30 40 50 60 70 80 90 100 110

Suns

1.1 1.2 1.3 1.4

VOC (V)

**Figure 39.** Solar spectrums at different Suns.

JSC (mA/cm2

)

0 1000 2000

FF (%)

PCE (%)

Figure 40. PV parameters vs Suns.

Air Mass

and resistance they would ultimately change the performance of the PV devices.

the size of the solar cell devices and the PV device is called concentrator solar cells.

Variation of solar irradiance for different 'Sun' is illustrated in figure 39. An array of mirrors or lenses have been employed in order to focus onto a small sized absorber which is thereby subjected to intense sunlight for the generation of energy in one form to another [54] this is so called 'multi-sun' which enhances efficiency and reduce Figure 40 shows the effect of 'multi-sun' on the performance of the solar cell devices. Obser‐ vations found that this is effective from 2 'Sun' to around 30 'Suns' and maximum values are found at around 5 Suns for the perovskite solar cells. After 30 'Suns' they produce detrimental effect to the performance which would be due to change in carrier concentrations, mobility of the charge carries, resistance and band gap of the materials.

## **5. Conclusions**

In this paper we have discussed the outcome of a simulation study on organometal halide perovskite focusing on the role of the different layers of the solar cell using SCAPS as a simulation tool. An interestingly high performance; PCE 22.72%, VOC = 1.12 V, JSC = 24.66 mA/ cm2 , and FF = 82.12% have been observed for 450 nm thickness of absorber. Increase in thickness of the absorber has increased the PCE. In the case of HTM and ETM performance has been decreasing with increase in thickness. The higher values of JSC, VOC, FF, and quantum efficiency plots have proved the low pinholes and shallow defect density, outstanding absorption range and strength. The hole mobility and acceptor/donor concentration of the HTM and ETM, Band tail of perovskite, interface trap density and workfunction of back and front contacts (electrode and counter electrode) have shown significant influence on the device performance. Even with these strong merits of HTM and ETM, enhancement of hole mobility and conductivity of HTM and ETM, stability of perovskite and TiO2 and replacement of toxic lead are still crucial. Through suitable processing/synthesizing of the perovskite absorbers, best engineering the selective contact, and increasing conductivity of HTM and ETM will boost the stability as well as performance of the device.

This paper has also been focused on the study of two Electron Transporting Materials on organometal halide perovskite based solar cells. An interestingly high PCE, in both cases, have been found for 400 nm thickness of absorber. ZnO as ETM has been found slightly more efficient than TiO2 by using similar baseline parameters. In the case of increase in thickness of the ETMs PCE have found significant influence but the effect in TiO2 is more prominent than in ZnO due to low hole mobility and somewhat more absorption as well as less transmission of light through TiO2. Furthermore, dopant concentrations of ETMs and compensation ratio (ND/NA) of perovskite have also changed the performance of the devices due to increase in conductivity and recombination/loss respectively. This study has pointed out two things - first, ZnO is found slightly more defective than TiO2 as practical solar cell with ZnO is less efficient. Although simulations have shown ZnO to be more efficient than TiO2 in real practice it is less efficient since ZnO possessed relatively more band tail and interface traps. Second, ZnO is a good alternative of TiO2 for highly efficient solar cells to reduce the cost and enhance the electron mobility.

Change in temperature and illumination have affected the performance of the solar cells due to change in carrier concentrations, mobility of the charge carries, resistance and band gap of the materials.

## **Acknowledgements**

This work has been supported by the Spanish Government through Ministerio de Economía y Competitividad (ENE2013-46624-C4-4-R) and Generalitat valenciana (Prometeus 2014/044)

## **Author details**

Bernabé Marí Soucase1\*, Inmaculada Guaita Pradas2 and Krishna R. Adhikari3

\*Address all correspondence to: bmari@fis.upv.es;adhikarikrishna@wrc.edu.np

1 Departament de Física Aplicada-IDF, Universitat Politècnica de València, València, Spain

2 Departament d'Economia i Ciències Socials, Universitat Politècnica de València, València, Spain

3 Tribhuvan University, Kathmandu, Nepal

## **References**

and counter electrode) have shown significant influence on the device performance. Even with these strong merits of HTM and ETM, enhancement of hole mobility and conductivity of HTM and ETM, stability of perovskite and TiO2 and replacement of toxic lead are still crucial. Through suitable processing/synthesizing of the perovskite absorbers, best engineering the selective contact, and increasing conductivity of HTM and ETM will boost the stability as well

This paper has also been focused on the study of two Electron Transporting Materials on organometal halide perovskite based solar cells. An interestingly high PCE, in both cases, have been found for 400 nm thickness of absorber. ZnO as ETM has been found slightly more efficient than TiO2 by using similar baseline parameters. In the case of increase in thickness of the ETMs PCE have found significant influence but the effect in TiO2 is more prominent than in ZnO due to low hole mobility and somewhat more absorption as well as less transmission of light through TiO2. Furthermore, dopant concentrations of ETMs and compensation ratio (ND/NA) of perovskite have also changed the performance of the devices due to increase in conductivity and recombination/loss respectively. This study has pointed out two things - first, ZnO is found slightly more defective than TiO2 as practical solar cell with ZnO is less efficient. Although simulations have shown ZnO to be more efficient than TiO2 in real practice it is less efficient since ZnO possessed relatively more band tail and interface traps. Second, ZnO is a good alternative of TiO2 for highly efficient solar cells to reduce the cost and enhance the

Change in temperature and illumination have affected the performance of the solar cells due to change in carrier concentrations, mobility of the charge carries, resistance and band gap of

This work has been supported by the Spanish Government through Ministerio de Economía y Competitividad (ENE2013-46624-C4-4-R) and Generalitat valenciana (Prometeus 2014/044)

1 Departament de Física Aplicada-IDF, Universitat Politècnica de València, València, Spain

2 Departament d'Economia i Ciències Socials, Universitat Politècnica de València, València,

\*Address all correspondence to: bmari@fis.upv.es;adhikarikrishna@wrc.edu.np

and Krishna R. Adhikari3

as performance of the device.

484 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

electron mobility.

the materials.

**Acknowledgements**

**Author details**

Spain

Bernabé Marí Soucase1\*, Inmaculada Guaita Pradas2

3 Tribhuvan University, Kathmandu, Nepal


[31] Giorgi G, Fujisawa JI, Segawa H and Yamashita K. Small Photocarrier Effective Masses Featuring Ambipolar Transport in Methylammonium Lead Iodide Perov‐ skite: A Density Functional Analysis. J. Phys. Chem. Lett. 2013; 4: 4213–4216.

[17] Bottaro D and Moscowitz J. Current photovoltaic technology: current progress and

[18] Shockley W and Queisser HJ. Detailed balance limit of efficiency of pn junction solar

[19] chemistry/perovskite/what/history/index; http://www.chemistry/perovskite/what/

[20] Luo S and Daoud WA. Recent progress in organic-inorganic halide perovskite solar cells: mechanism and material design. J. Mater. Chem. A. 2015. doi:10.1039/

[21] Shalau AM. Fundamental absopion edge in lead (II) iodode:Potassium iodide alloys.

[22] Gao P, Gratzel M, and Nazeeruddin MK. Organohalide lead perovskites for photo‐ voltaic applications. Energy and Environmental Science. 2014; 7: 2448-2463.

[23] Bretschneider SA, Weickert J, Dorman JA, and Schmidt-Mende L. Research update: Physical and electrical characteristics of lead halide perovskites for solar cells appli‐

[24] Frost JM, Butler KT, Brivio F, Hendon CH, Schilfgaarde MV and Walsh A. Atomistic origin of high performance in hybrid halide perovskite solarcells. Nano Lett. 2014;

[25] Baikie T, Fang YN, Kadro JM, Schreyer M, Wei FX, Mhaisalkar SG, Gratzel M and White TJ. Synthesis and crystal chemistry of the hybrid perovskite (CH3NH3)PbI3 for solid-state sensitised solar cell applications. J. Mater. Chem. A. 2013; 1: 5628–5641. [26] Stoumpos C, Malliakas CD and Kanatzidis MG. Semiconducting Tin and Lead Io‐ dide Perovskites with Organic Cations: Phase Transitions, High Mobilities, and

Near-Infrared Photoluminescent Properties. Inorg. Chem. 2013; 52: 9019–9038. [27] Yin WJ, Yang JH, Kang J, Yan Y and Wei SH. Halide perovskite materials for solar

[28] Walsh A, Watson GW. The origin of the stereochemically active Pb (II) lone pair: DFT calculations on PbO and PbS. Journal of Solid State Chemistry. 2005; 178: 1422–1428;

[29] Walsh A, Payne DJ, Egdell RG and Watson GW. Stereochemistry of post-transition metal oxides: revision of the classical lone pair model. Chem.Soc. Rev. 2011; 40: 4455–

[30] Keith T. Butler, Jarvist M. Frost and Aron Walsh,. Band alignment of the hybrid hal‐ ide provskites CH3NH3PbCl3, CH3NH3PbBr3, and CH3NH3PbI3. Materials Horizons.

future prospects. 1977. MIT Energy laboratory Reprot- MIT-EL 77-041W.

cells. Journal of Applied Physics. 1961; 31: 510-519.

486 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Sol. Energy Mater. 1980; 2 (3): 327-332.

cations. APL Materials. 2014; 2: 040701 1-9.

cells: a theretical review. J. Mater. Chem. A. 2014.

history/ index.htm

c4ta04953e

14: 2484-2590.

446.

2014; 2: 228-231.


## **Tantalate-based Perovskite for Solar Energy Applications**

Yiguo Su, Junyu Lang, Chunfang Du and Xiaojing Wang

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61390

## **Abstract**

[45] Liu F, Zhu J, Wei J, Li Y, Lv M, Yang S, Zhang B, Yao J, and Dai S. Numerical simula‐ tion: Toward the design of high efficiency planar perovskite solar cells. Applied

[46] Wolf SD, Holobsky J, Moon SJ, Loper P, Niesen B, Ledinsky M, Haug FJ, and Ballif C. Organometallic halide perovskites: Sharp optical absorption edge and its relation to photovoltaic performance. The Journal of Physical Chemistry Letters. 2014;

[47] Xiang HJ, Gong XG, Lang L, Yang JH, and Liu HR. First principles study on the elec‐ tric and optical properties of cubic ABX3 halide perovskite. Physics Letters A.

[48] Park NG, Kim HS, and Im HS. Organohalide perovskite: New horizons in solar cell

[49] Laura M. Hertz, Christian Wehrenfennig, Giles E. Eperon, Michael B. Johston and Henry J. Snaith. (2014). High charge carrier mobilities and lifetimes in organolead tri‐

[50] Trukhanov VA, Bruevich VV and Paraschuk DY. Effect of doping on performance of

[51] Wang MK, Gratzel C, Moon SJ, Humphry-Baker R, Rossier-Iten N, Zakeeruddin SM, and Gratzel M. Surface design in solid state dye sensitized solar cells: Effect of Zwit‐ terionic co-adsorbants on photovoltaic performance. Advanced fuctional Materials.

[53] Yamben SD, Haldar A, Liao KS, Dillon EP, Barron AR, and Curran SA. Optimization of organic solar cells with thin film Au as anode. Solar Energy Materials and Solar

[54] Middy Gerald W. Solar insolation and concentration by coupled fiber optic, United

[52] Kittel C. Introduction to Solid State Physics. John Wiley and Sons, Inc. (2005)

research. Journal of Physical Chemistry C. 2014;118:5615-5625.

halide perovskite. *Advanced Materials*. 26, 1584-1589.

organic solar cells. Physics Review B. 2011;84: 205318.

Physics Letters. 2014;104: 253508.

488 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

5:1035-1039.

2014;378:290-293.

2009;19:2163-2172.

Cells. 2011;95:2424-24.

States Patents. 1984.

To realize a sustainable society in the near future, the development of clean, renewable, cheap and sustainable resources and the remediation of environmental pollutions using solar energy as the driving force would be important. During the past few decades, plen‐ ty of efforts have been focused on this area to develop solar light active materials to meet the increased energy and environmental crisis. Owning to the unique perovskite-type structure, tantalate-based semiconductors with unable chemical composition show high activities toward the conversion of solar radiation into chemical energy. Moreover, vari‐ ous engineering strategies, including crystal structure engineering, electronic structure engineering, surface/interface engineering, co-catalyst engineering and so on, have been developed in order to modulate the charge separation and transfer efficiency, optical ab‐ sorption, band gap position and photochemical and photophysical stability, which would open a realm of new possibilities for exploring novel materials for solar energy applica‐ tions.

**Keywords:** Tantalate, photocatalysis, solar energy, perovskite

## **1. Introduction**

In view of global energy crisis and environmental pollution, the search for renewable and clean energy resources and the development of eco-friendly systems for environmental remediation have received great attention. Solar energy is the prime renewable source of energy for every life on the earth. The amount of solar energy that strikes the Earth yearly in the form of sunlight is approximately ten thousand times the total energy that is consumed on this planet [1]. However, sunlight is diffuse and intermittent, which impedes its collecting and storage that play critical roles in the full exploitation of its potentials. As one of the most promising solutions for storing and converting solar energy, semiconductor photocatalysis has attracted much

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

attention, since it provides an environmental benign strategy for splitting water into hydrogen and oxygen, reducing carbon dioxide into useful chemicals and fossil fuels, and completely eliminating all kinds of contaminants under the sunlight illumination under ambient condi‐ tions [2–4]. Generally, the fundamental principles of semiconductor photocatalysis have been extensively reported in previous works [5]. A photocatalytic reaction is initiated on the basis of the formation of photogenerated charges (such as electrons and holes) after capture of sunlight by a semiconductor. Consequently, electrons can transit from valence band (VB) to conduction band (CB) leaving behind holes in the VB. If the photogenerated electron–hole pairs' separation is maintained, the photogenerated carriers can move to the semiconductor surfaces to react with the adsorbed small molecules (dioxygen and water), generating the redox foundations of active species which lead to water splitting and/or destruction of organic compounds [6]. It is also noted that the photogenerated electrons in the CB can also recombine with the photogenerated holes in the VB to dissipate the input energy in the form of heat or radiated light (see Figure 1). From the perspective of efficient utilization of solar energy, the recombination between the photogenerated electrons and holes is not desired, which limits the efficiency of a semiconductor photocatalyst. For better photocatalytic performance, the photogenerated electrons and holes must be separated effectively, and charges have to be transferred rapidly at the mean time across the photocatalysis to astrict the recombination.

**Figure 1.** Schematic illustration of the principle of photocatalysis.

To date, several semiconductors, including TiO2, ZnO, SnO2, BiVO4 and so forth, have been extensively investigated [7–10]. Among them, tantalate-based semiconductors with perov‐ skite-type structure have certainly verified to be one of the brilliant photocatalysts for pro‐ ducing hydrogen from water and the oxidative disintegration of much organic containment [11]. For instance, NaTaO3 with a perovskite-type structure showed a quantum yield of 56% under ultraviolet light irradiation after lanthanum doping and NiO co-catalyst loading [12]. Nevertheless, because of their broad band gap, most tantalate-based semiconductors can only react under ultraviolet or near-ultraviolet radiation, which reduces the utilization of ~43% of the solar spectrum To efficiently utilize the sunlight in visible region, the design of visible-light response tantalate-based catalysts is current demanded. Up to now, numerous methodologies have been developed to prepare different visible-light-driven tantalate-based photocatalysts, including doping strategy, heterojunction, facet control and so on [13,14].

This chapter emphasizes certain topical works that concentrate on tantalate-based photocata‐ lysts for solar energy application. The aim is to display that the rational design, fabrication and modifications of tantalate-based semiconductors have tremendous effects onto their final photocatalytic activity, simultaneously, providing some stimulating perspectives on the future applications.

## **2. Alkali tantalate-based perovskite semiconductors**

## **2.1. Synthetic strategies of alkali tantalates**

attention, since it provides an environmental benign strategy for splitting water into hydrogen and oxygen, reducing carbon dioxide into useful chemicals and fossil fuels, and completely eliminating all kinds of contaminants under the sunlight illumination under ambient condi‐ tions [2–4]. Generally, the fundamental principles of semiconductor photocatalysis have been extensively reported in previous works [5]. A photocatalytic reaction is initiated on the basis of the formation of photogenerated charges (such as electrons and holes) after capture of sunlight by a semiconductor. Consequently, electrons can transit from valence band (VB) to conduction band (CB) leaving behind holes in the VB. If the photogenerated electron–hole pairs' separation is maintained, the photogenerated carriers can move to the semiconductor surfaces to react with the adsorbed small molecules (dioxygen and water), generating the redox foundations of active species which lead to water splitting and/or destruction of organic compounds [6]. It is also noted that the photogenerated electrons in the CB can also recombine with the photogenerated holes in the VB to dissipate the input energy in the form of heat or radiated light (see Figure 1). From the perspective of efficient utilization of solar energy, the recombination between the photogenerated electrons and holes is not desired, which limits the efficiency of a semiconductor photocatalyst. For better photocatalytic performance, the photogenerated electrons and holes must be separated effectively, and charges have to be transferred rapidly at the mean time across the photocatalysis to astrict the recombination.

490 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

To date, several semiconductors, including TiO2, ZnO, SnO2, BiVO4 and so forth, have been extensively investigated [7–10]. Among them, tantalate-based semiconductors with perov‐ skite-type structure have certainly verified to be one of the brilliant photocatalysts for pro‐ ducing hydrogen from water and the oxidative disintegration of much organic containment [11]. For instance, NaTaO3 with a perovskite-type structure showed a quantum yield of 56% under ultraviolet light irradiation after lanthanum doping and NiO co-catalyst loading [12]. Nevertheless, because of their broad band gap, most tantalate-based semiconductors can only react under ultraviolet or near-ultraviolet radiation, which reduces the utilization of ~43% of the solar spectrum To efficiently utilize the sunlight in visible region, the design of visible-light

**Figure 1.** Schematic illustration of the principle of photocatalysis.

Alkali tantalate-based perovskite semiconductors (likewise LiTaO3, NaTaO3 and KTaO3) have a general formula of ABO3 and have drawn a lot of attention due to the peculiar supercon‐ ductivity, photocatalytic property, electrochemical reduction and electromagnetic features. There are two kinds of totally different cationic sites in a perovskite photocatalysis, in which A-site is coordinated by twelve O2–, and is usually occupied by relative bigger cations (Li, Na and K). The B-site is taken up by smaller cations (Ta) with a coordination of six, as illustrated in Figure 2. The bond angles of Ta–O–Ta are 143o for LiTaO3, 163o for NaTaO3 and 180o for KTaO3, respectively [15]. Wiegel and coworkers have reported the relationship between crystal structures and energy delocalization for alkali tantalates. When the Ta–O–Ta bond angle is close to 180o , the migration of excitation energy can be accelerated and the band gap decreases [16]. Thereby, the delocalization of excited energy of LiTaO3, NaTaO3 and KTaO3 increases in turn. This result suggests that KTaO3 may be predicted to be with the best photocatalytic activity.

**Figure 2.** Crystal structure of LiTaO3, NaTaO3 and KTaO3, respectively.

Alkali tantalate with different sizes, morphologies and compositions can be prepared via traditional solid-state method, solvothermal, sol–gel, molten salt and other methods. Basically, the traditional solid-state method is quite often used to prepare alkali tantalates, which includes the high temperature processing of the combination of alkali salts and tantalum pentaoxide. Kudo and coworkers successfully prepared ATaO3 (A = Li, Na and K) materials with high crystallinity via solid-state method. It is found that all alkali tantalate showed superior photocatalytic activity toward stoichiometric water splitting under ultraviolet condition [17]. The high photocatalytic activity is chiefly depending on the high CB level consisting of Ta 5d orbitals [15]. Among them, KTaO3 is the most photocatalytic active, which may be ascribed to the fact that KTaO3 can absorb the most of photons and possesses the least distorted perovskite structure, being consistent with the above-mentioned discussion. The evolution rate of H2 and O2 was determined to be 29 and 13 μmolh–1, respectively. To improve more photocatalytic activity of ATaO3, a modified solid-state method was adopted by adding extra amount of alkali to compensate the loss [18]. When preparing the alkali tantalates with the existence of excess alkali, the photocatalytic activity of LiTaO3, NaTaO3 and KTaO3 materials were improved ten to hundred times. LiTaO3 is the naked alkali tantalate photoca‐ talyst which showed the highest activity. This is because LiTaO3 possesses higher conduction band levels than that of NaTaO3 and KTaO3, which may predict higher transfer rate of excited energy and the subsequent higher phtocatalytic activity. This type of phenomena was likewise observed for CaTa2O6, SrTa2O6 and BaTa2O6 photocatalysts with similar crystal structures [19].

**Figure 3.** Ta–O–Ta bond angle in a 2 × 2 × 2 supercell of monoclinic phase NaTaO3. Red ball, green ball and grey ball represent O, Na and Ta atoms, respectively.

One should note that the synthetic strategy also has great influence on the structural features as well as photocatalytic activity. For instance, sol–gel method was also used to prepare NaTaO3 nanoparticles. By using CH3COONa 3H2O and TaCl5 as the raw materials and citric acid as the complexing agent, NaTaO3 nanoparticles with monoclinic phase that shows an indirect band gap, high densities of states near the band edges and a Ta–O–Ta bond angle close to 180o (Figure 3) are obtained. This result is quite different to NaTaO3 that was synthesized via solid-state method, which formed the orthorhombic phase that has a direct band gap and a Ta–O–Ta bond angle of 163°. It is found that monoclinic NaTaO3 has lots of effective states available for the photogenerated charge pairs. Meanwhile, the larger surface area and the advantageous features in the electronic and crystalline structures for the monoclinic NaTaO3 have resulted in a remarkably higher photocatalytic activity for the sol–gel synthesized NaTaO3 than that for the solid-state derived orthorhombic NaTaO3 [20]. Besides sol–gel method, the molten-salt approach is also adopted to prepare alkali tantalate materials [21,22]. By a convenient molten-salt process, a series of NaTaO3 and KTaO3 efficient photocatalysts is successfully synthesized, which are highly crystallized single crystal nanocubes (about 100 nm large). Doping tetravalent Zr4+ and Hf4+ in NaTaO3 and KTaO3 efficiently increases the activity and stability of catalyst at the same time, although the energy levels have no change. Moreover, Zr4+ and Hf4+ doping can also led to particle size reduction and nearly monodispersed feature of NaTaO3 and KTaO3 nanoparticles. In the absence of co-catalyst, the photocatalytic activity can reach 4.65 and 2.31 mmolh–1 toward H2 and O2 production, respectively [22]. A novel kind of strontium-doped NaTaO3 mesocrystals was also prepared by a common molten-salt way. The obtained three-dimensional architectures showed high crystallinity, preferred orientation growth and high surface area. The ability for hydrogen generation of photocatalyst achieves 27.5 and 4.89 mmolh–1 for methanol aqueous solution and pure water splitting under ultra‐ violet light irradiation [23]. However, either solid-state method or molten-salt approach often leads to ultra-low surface areas of alkali tantalates, which limits the photocatalytic activity. Hydrothermal synthesis is advantageous for regular nucleation of nanocrystals with welldefined particles, morphologies, crystallinity and surface areas [24]. For instance, nano-sized Ta2O5 and NaTaO3, KTaO3 and RbTaO3 cubes are prepared by a facile hydrothermal method [25]. It is observed that pH influences much in the process of tantalum compound nanoparticles preparation. The obtained morphologies ranging from agglomerated particles in acidic medium over sticks at neutral pH value to cubes in elementary media can be achieved, which are similar to titanates [26]. A microwave-assisted hydrothermal technique was reported using Ta2O5 and NaOH as starting materials under quite mild conditions with short reaction time. The BET surface area of NaTaO3 nanoparticles prepared by microwave-assisted hydrothermal method is about 1.5 times than that prepared by conventional hydrothermal method [27]. After loading NiO as co-catalyst, this photocatalyst showed photocatalytic activity for overall water splitting more than two times greater than those prepared by conventional hydrothermal process [28]. As an outstanding example, NaTaO3 nanoparticles through hydrothermal treatment highly improved the photocatalytic activity by a factor of 8 toward water splitting in comparison with the photocatalysts obtained by traditional solid-state method, which is attributed to their smaller particle size, larger surface area and higher crystallinity [29,30].

## **2.2. Electronic structure engineering**

## *2.2.1. Doping strategies*

includes the high temperature processing of the combination of alkali salts and tantalum pentaoxide. Kudo and coworkers successfully prepared ATaO3 (A = Li, Na and K) materials with high crystallinity via solid-state method. It is found that all alkali tantalate showed superior photocatalytic activity toward stoichiometric water splitting under ultraviolet condition [17]. The high photocatalytic activity is chiefly depending on the high CB level consisting of Ta 5d orbitals [15]. Among them, KTaO3 is the most photocatalytic active, which may be ascribed to the fact that KTaO3 can absorb the most of photons and possesses the least distorted perovskite structure, being consistent with the above-mentioned discussion. The evolution rate of H2 and O2 was determined to be 29 and 13 μmolh–1, respectively. To improve more photocatalytic activity of ATaO3, a modified solid-state method was adopted by adding extra amount of alkali to compensate the loss [18]. When preparing the alkali tantalates with the existence of excess alkali, the photocatalytic activity of LiTaO3, NaTaO3 and KTaO3 materials were improved ten to hundred times. LiTaO3 is the naked alkali tantalate photoca‐ talyst which showed the highest activity. This is because LiTaO3 possesses higher conduction band levels than that of NaTaO3 and KTaO3, which may predict higher transfer rate of excited energy and the subsequent higher phtocatalytic activity. This type of phenomena was likewise observed for CaTa2O6, SrTa2O6 and BaTa2O6 photocatalysts with similar crystal structures [19].

492 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 3.** Ta–O–Ta bond angle in a 2 × 2 × 2 supercell of monoclinic phase NaTaO3. Red ball, green ball and grey ball

One should note that the synthetic strategy also has great influence on the structural features as well as photocatalytic activity. For instance, sol–gel method was also used to prepare NaTaO3 nanoparticles. By using CH3COONa 3H2O and TaCl5 as the raw materials and citric acid as the complexing agent, NaTaO3 nanoparticles with monoclinic phase that shows an indirect band gap, high densities of states near the band edges and a Ta–O–Ta bond angle close to 180o (Figure 3) are obtained. This result is quite different to NaTaO3 that was synthesized via solid-state method, which formed the orthorhombic phase that has a direct band gap and a Ta–O–Ta bond angle of 163°. It is found that monoclinic NaTaO3 has lots of effective states available for the photogenerated charge pairs. Meanwhile, the larger surface area and the

represent O, Na and Ta atoms, respectively.

Introducing foreign elements, including metal ions or non-metal ions, into semiconductor host matrix is one of the most effective methods to modulate the electronic structure of the host semiconductor and produce enhanced photocatalytic performance. Owing to a big difference in radius of A- and B-site ions in alkali tantalates, the dopants can selectively permeate into the A or B sites, which determine chemical composition, surface features, electronic structure and their photocatalytic properties. To date, studies on the alkali tantalates derived by doping strategy are thoroughly investigated. La-doped NaTaO3 is the most active photocatalyst in photocatalytically splitting water area [12]. In this case, the catalytic activity of NaTaO3 is extremely modulated by doping with La3+. For example, the crystallinity growths and a surface stair structure with nanometer-scale features are constructed, which improve the separation efficiency of the photogenerated electron–hole pairs and the photocatalytically splitting water activity. The surface step structure is also formed in alkaline earth metal ion doped NaTaO3, which showed improvement of photocatalytic water splitting properties [31]. Bi3+ doped NaTaO3 nanoparticles are prepared under different initial stoichiometric ratio by traditional solid-state reaction, which showed visible light absorption and tunable photocatalytic activity. Controlling the original molar ratio of the reactants, the intrusion of bismuth at sodium site and tantalum site in NaTaO3 can be well-modulated and the optimum performance can be easily changed. Occupancy of Bi atom at Na site of NaTaO3 is not contributing to increase visible light absorption while occupancy of Bi at Ta site or at both Na and Ta site induces visible light absorption and the subsequent methyl blue degradation under visible light [32,33]. La, Cr codoping NaTaO3 system is also developed by spray pyrolysis from aqueous and polymeric precursor solution. The hydrogen evolution rate of La, Cr codoped NaTaO3 was enhanced 5.6 times to 1467.5 μmol g–1h–1, and the induction period was shortened to 33%, compared to the identical values achieved by the Cr-doped NaTaO3 photocatalyst prepared from aqueous precursor solution [34]. Besides metal ion doping, several non-metal ions are also incorporated into the host matrix of alkali tantalates for improved visible light absorption and photocatalytic performance [35,36]. A plane-wave-based density functional theory calculation is conducted to predict the doping effects on the variations of the band structure of non-metal ions doped NaTaO3. There were studies about nitrogen, sulfur, carbon and phosphorus monodoping and nitrogen–nitrogen, carbon–sulfur, phosphorus–phosphorus and nitrogen–phosphorus codoping. Nitrogen and sulfur monodoping can improve the valence band edge to higher and keep the ability to split water into H2 and O2 remain unchanged, as is shown in Figure 4. Double hole-mediated codoping can decrease the band gap dramatically. Nitrogen–nitrogen, carbon– sulfur and nitrogen–phosphorus codoping could narrow band gap to 2.19, 1.70 and 1.34 eV, respectively, which could absorb visible light.

**Figure 4.** Band alignment of non-metal ions doped NaTaO3. The position of the valence band edge of pure NaTaO3 is adopted from experiment [37].

## *2.2.2. Defect chemistry engineering*

photocatalytically splitting water area [12]. In this case, the catalytic activity of NaTaO3 is extremely modulated by doping with La3+. For example, the crystallinity growths and a surface stair structure with nanometer-scale features are constructed, which improve the separation efficiency of the photogenerated electron–hole pairs and the photocatalytically splitting water activity. The surface step structure is also formed in alkaline earth metal ion doped NaTaO3, which showed improvement of photocatalytic water splitting properties [31]. Bi3+ doped NaTaO3 nanoparticles are prepared under different initial stoichiometric ratio by traditional solid-state reaction, which showed visible light absorption and tunable photocatalytic activity. Controlling the original molar ratio of the reactants, the intrusion of bismuth at sodium site and tantalum site in NaTaO3 can be well-modulated and the optimum performance can be easily changed. Occupancy of Bi atom at Na site of NaTaO3 is not contributing to increase visible light absorption while occupancy of Bi at Ta site or at both Na and Ta site induces visible light absorption and the subsequent methyl blue degradation under visible light [32,33]. La, Cr codoping NaTaO3 system is also developed by spray pyrolysis from aqueous and polymeric precursor solution. The hydrogen evolution rate of La, Cr codoped NaTaO3 was enhanced 5.6 times to 1467.5 μmol g–1h–1, and the induction period was shortened to 33%, compared to the identical values achieved by the Cr-doped NaTaO3 photocatalyst prepared from aqueous precursor solution [34]. Besides metal ion doping, several non-metal ions are also incorporated into the host matrix of alkali tantalates for improved visible light absorption and photocatalytic performance [35,36]. A plane-wave-based density functional theory calculation is conducted to predict the doping effects on the variations of the band structure of non-metal ions doped NaTaO3. There were studies about nitrogen, sulfur, carbon and phosphorus monodoping and nitrogen–nitrogen, carbon–sulfur, phosphorus–phosphorus and nitrogen–phosphorus codoping. Nitrogen and sulfur monodoping can improve the valence band edge to higher and keep the ability to split water into H2 and O2 remain unchanged, as is shown in Figure 4. Double hole-mediated codoping can decrease the band gap dramatically. Nitrogen–nitrogen, carbon– sulfur and nitrogen–phosphorus codoping could narrow band gap to 2.19, 1.70 and 1.34 eV,

494 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 4.** Band alignment of non-metal ions doped NaTaO3. The position of the valence band edge of pure NaTaO3 is

respectively, which could absorb visible light.

adopted from experiment [37].

Defect chemistry plays an important role in modulating the electronic structure, charge carrier conductivity and photocatalytic performance [38]. Defect chemistry often shows different impacts on the photocatalytic efficiency for most of the semiconductors. Previous literature on NaTaO3 indicated that the accretion of the extra quantity of Na in the synthesis of NaTaO3 blocked construction of sodium ion defects in NaTaO3 crystals, leading to the extreme enhancement of photocatalytic activity [18]. Basically, the native defects, such as oxygen vacancies and sodium ion defects, are often observed in NaTaO3. Oba and coworkers investi‐ gated the formation energies and electronic structure of lattice vacancies, antisite defects and lanthanum impurities in NaTaO3 using first-principles calculations based on density-func‐ tional theory [39]. Under oxygen-poor environments, oxygen vacancy as a double donor is a main defect. In La-NaTaO3, the replacement of La at Ta site is similar to make up as a shallow acceptor under oxygen-rich environments whereas the replacement of La at Na site forms as a double donor under oxygen-poor environments. The location predilection of lanthanum leads to self-compensation in heavily doped cases, which have great impact on the change in carrier concentration and photocatalytic activity [12]. Defective center not only alters the carrier concentration but also induces visible light absorption. In Eu3+ doped NaTaO3, a nonstoichio‐ metric Na/Ta molar ratio led to site-selective occupation of Eu3+ dopant ions, which resulted in a monotonous lattice expansion and local symmetry distortion [11]. The site-selective occupation of Eu3+ gave rise to certain types of defective centers due to the charge difference between Eu3+ ions and Na+ and/or Ta5+ ions, which is crucial to the modification of absorption in visible region and photocatalytic activity.

## **2.3. Surface/interface engineering**

## *2.3.1. Heterojunction of nano-/microarchitectures*

The constructions of heterojunction by combining a semiconductor with other semiconductors have attracted much research attention because of their perfect effectiveness in the separation of the photogenerated charge carriers and boosting the photocatalytic activity. In the past few years, a lot of significant findings have been described on the heterojunction of nano-/ microarchitectures. Nano-Cu2O/NaTaO3 composite for the degradation of organic pollutants have also been successfully developed [13]. Nano-Cu2O/NaTaO3 composite exhibits highly enhanced photocatalytic activity in comparison to their individual counterpart. Furthermore, C3N4/NaTaO3 and C3N4/KTaO3 composite photocatalysts were also developed [40,41]. Loading of C3N4 is a good strategy to achieve the visible light photocatalytic activity (Figure 5). Photogenerated electron jumped from the VB to CB of C3N4 could unswervingly insert into the conduction band of NaTaO3 or KTaO3, making C3N4/NaTaO3 and C3N4/KTaO3 as visible light-driven photocatalyst. Both of the composites showed superior photocatalytic activity toward Rhodamine B degradation under visible light irradiation, being close to commercial P25. Yin and coworkers reported the preparation of novel C–NaTaO3–Cl–TiO2 composites via a facile solvothermal method. When C–NaTaO3 is joined with Cl–TiO2 to construct a core shell configuration, the visible light-induced degradation activity toward NOx of the catalysts under visible light irradiation could be highly improved because of the suppression of the recombi‐ nation of photogenerated charge carriers [42]. Zaleska et al. prepared a series of novel binary and ternary composite photocatalysts based on the combination of KTaO3, CdS and MoS2 semiconductors via hydro/solvothermal precursor route. They found that the highest photo‐ catalytic activity toward phenol degradation under both UV-Vis and visible light irradiation and superior stability in toluene removal was observed for ternary hybrid obtained by calcination of KTaO3, CdS and MoS2 powders at the 10: 5: 1 molar ratio [43].

**Figure 5.** Schematic illustration of the photocatalytic degradation process of RhB by visible light-irradiated C3N4/ NaTaO3 or C3N4/KTaO3.

## *2.3.2. Mesoporous structures construction*

As one of the most important factors, surface area also imposes a big effect on the photocatalytic activity of the semiconductors. The majority of photocatalytic reactions occur at semiconductor surfaces, and therefore the photocatalytic activities of semiconductor oxides are usually greatly improved by the increase in surface area [44]. To further improve the surface area, nanocrys‐ talline NaTaO3 thin films with ordered three-dimensional mesoporous and nanostick-like constructions were successfully produced by PIB-*b*-PEO polymer-based sol–gel method. NaTaO3 prepared at 650<sup>ο</sup> C exhibits a BET surface area of about 270 m2 cm–3, which is much larger than the ever reported values [45]. These nanocrsytalline mesoporous NaTaO3 samples show both enhanced ultraviolet light photocatalytic activity and can keep steady performance. A confined space synthesis process was also used for preparing colloidal array of NaTaO3 by using three-dimensional mesoporous carbon as the hard template. This method brings about the creation of a colloidal collection of mesoporous NaTaO3 particles (20 nm). After NiO loading, the mesoporous NaTaO3 nanoparticles showed photocatalytic activity for overall water splitting more than three times as high as non-structured bulk NaTaO3 particles [46]. A carbon modified NaTaO3 mesocrystal nanoparticle was also successfully synthesized by a onepot solvothermal method by employing TaCl5, NaOH and glucose as the starting materials and distilled H2O/ethylene glycol mixed solution as a reaction solvent. The as-synthesized mesocrystal nanoparticles exhibited a high specific surface area of 90.8 m2 g–1 with large amounts of well-dispersed mesopores in the particles. The carbon-modified NaTaO3 meso‐ crystal demonstrated excellent efficiency for continuous NO gas destruction under visible light irradiation, which is considerably superior to those of the unmodified NaTaO3 specimen and commercial Degussa P25, owning to large specific surface area, high crystallinity and visible light absorption [47].

## **2.4. Co-catalyst engineering**

visible light irradiation could be highly improved because of the suppression of the recombi‐ nation of photogenerated charge carriers [42]. Zaleska et al. prepared a series of novel binary and ternary composite photocatalysts based on the combination of KTaO3, CdS and MoS2 semiconductors via hydro/solvothermal precursor route. They found that the highest photo‐ catalytic activity toward phenol degradation under both UV-Vis and visible light irradiation and superior stability in toluene removal was observed for ternary hybrid obtained by

**Figure 5.** Schematic illustration of the photocatalytic degradation process of RhB by visible light-irradiated C3N4/

As one of the most important factors, surface area also imposes a big effect on the photocatalytic activity of the semiconductors. The majority of photocatalytic reactions occur at semiconductor surfaces, and therefore the photocatalytic activities of semiconductor oxides are usually greatly improved by the increase in surface area [44]. To further improve the surface area, nanocrys‐ talline NaTaO3 thin films with ordered three-dimensional mesoporous and nanostick-like constructions were successfully produced by PIB-*b*-PEO polymer-based sol–gel method.

C exhibits a BET surface area of about 270 m2

larger than the ever reported values [45]. These nanocrsytalline mesoporous NaTaO3 samples show both enhanced ultraviolet light photocatalytic activity and can keep steady performance. A confined space synthesis process was also used for preparing colloidal array of NaTaO3 by using three-dimensional mesoporous carbon as the hard template. This method brings about the creation of a colloidal collection of mesoporous NaTaO3 particles (20 nm). After NiO loading, the mesoporous NaTaO3 nanoparticles showed photocatalytic activity for overall water splitting more than three times as high as non-structured bulk NaTaO3 particles [46]. A carbon modified NaTaO3 mesocrystal nanoparticle was also successfully synthesized by a onepot solvothermal method by employing TaCl5, NaOH and glucose as the starting materials and distilled H2O/ethylene glycol mixed solution as a reaction solvent. The as-synthesized

amounts of well-dispersed mesopores in the particles. The carbon-modified NaTaO3 meso‐ crystal demonstrated excellent efficiency for continuous NO gas destruction under visible light irradiation, which is considerably superior to those of the unmodified NaTaO3 specimen and

mesocrystal nanoparticles exhibited a high specific surface area of 90.8 m2

cm–3, which is much

g–1 with large

calcination of KTaO3, CdS and MoS2 powders at the 10: 5: 1 molar ratio [43].

496 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

NaTaO3 or C3N4/KTaO3.

NaTaO3 prepared at 650<sup>ο</sup>

*2.3.2. Mesoporous structures construction*

## *2.4.1. Noble metal co-catalyst engineering*

As well documented in previous literatures, co-catalyst introduces two positive factors into the photocatalyst, including promotion on the separation of photogenerated charge carriers and construction of active sites for reduction and/or oxidation reaction. Several noble metals have been commonly used as co-catalysts for photocatalytic applications. For example, water splitting activity of NaTaO3:La was improved when Au was loaded either by photodeposition method or by impregnation method. Moreover, Au/NaTaO3:La prepared by impregnation method exhibits much higher and more stable photocatalytic activity toward water splitting due to the fact that O2 reduction on photodeposited Au co-catalyst was more efficient than that of impregnated Au co-catalyst [48]. Besides Au nanoparticles, Pt is also frequently used as cocatalyst for increasing the photocatalytic activity of alkali tantalates. With the deposition of Pt nanoparticles as co-catalyst, rare earth (including Y, La, Ce and Yb) doped NaTaO3 exhibits a clear improvement of the hydrogen evolution, which is due to the fact that Pt nanoparticles act as electron scavengers reducing the photogenerated charge carrier recombination rate and facilitating the electron move to metal sites from the CB of NaTaO3, being as the catalytic center for hydrogen generation [49]. Moreover, Pd nanoparticles are also used as a co-catalyst for H2 production from water containing electron donor species. Su et al. prepared novel Pd/NiO core/shell nanoparticles as co-catalyst, which are placed on the surface of La doped NaTaO3 photocatalyst. It is noted that Pd nanoparticles are more effective for H2 generation from water containing methanol, while Pd/NiO core/shell nanoparticles exhibit a higher H2 generation by splitting pure water. The presence of NiO not only provides hydrogen evolution sites and suppresses the reverse reactions on Pd-based catalysts but also improves the stability of the Pd nanoparticles on the La doped NaTaO3 surfaces [50]. In another case, when RuO2 (1 wt.%) was introduced as co-catalyst, the ability for H2 generation of NaTaO3 prepared by an inno‐ vative solvo-combustion reaction was improved significantly, reaching around 50 mmol of H2 after 5 h, which is the best of other reports in literature [51].

## *2.4.2. Earth abundant elements co-catalyst engineering*

Due to too much scarcity and expense of noble metal co-catalyst to apply for wider scope solar energy applications, the development of high-efficiency and low-cost noble-metal-free cocatalysts is acutely necessary. Lately, co-catalysts composed of earth abundant elements have been explored extensively to replace noble metal co-catalysts for solar energy applications [52]. NiO is a p-type semiconductor with a band gap energy ranging within 3.5–4.0 eV, which is widely used as the co-catalyst of tantalates-based semiconductors for enhancing photocatalytic activity [53]. In the case of NiO/NaTaO3:La photocatalyst with high photocatalytic reactivity, NiO acts as co-catalyst loading as ultrafine NiO particles, which possesses characteristic absorption bands at 580 and 690 nm, The ultrafine NiO particles were highly active for hydrogen evolution as well as Pt of an excellent co-catalyst [12]. A detailed study on the structural features of NiO nanoparticles indicated that the interdiffusion of Na+ and Ni2+ cations created a solid–solution transition zone on the outer sphere of NaTaO3. The high photocatalytic activity resulting from a low NiO loading suggests that the interdiffusion of cations heavily doped the p-type NiO and n-type NaTaO3, reducing the depletion widths and facilitating charge transfers through the interface barrier [54]. Besides NiO, Ni metallic nanoclusters were also used as co-catalyst. For instance, a series of nickel-loaded La*x*Na1–*x*TaO3 photocatalysts was synthesized by a hydrogen peroxide-water based solvent method. Systematical investi‐ gation indicated that the activity of hydrogen generation from pure water is in sequence: Ni/NiO *>* NiO *>* Ni, whereas the activity sequence with respect to aqueous methanol is: Ni *>* Ni/NiO *>* NiO. Ni metallic nanoclusters exhibit the most active sites and facilitate the formation of hydrogen from aqueous methanol. In the case of Ni/NiO core/shell structure, Ni metallic nanoclusters induce the migration of photogenerated electrons from the bulk to catalyst surface, while NiO acts as H2 evolution site and prevents water formation from H2 and O2 [55].

## *2.4.3. Molecular co-catalyst engineering*

Molecular co-catalyst engineering have received much research attention in recent years. In a molecular/semiconductor hybrid system, the noble-metal-free molecular complex as cocatalyst can not only facilitate the charge separation but also help us to understand the mechanisms of hydrogen evolution and carbon dioxide reduction at molecular level [56]. Although the study on molecular sensitized alkali tantalates is limited, an excellent research has been done by Hong and coworkers. In this case, by using a molecular co-catalyst [Mo3S4] 4+, the photocatalytic activity of NaTaO3 was significantly improved. The hydrogen production rate is about 28 times higher than pure NaTaO3 because [Mo3S4] 4+ clusters can provide a large number of effective active sites for hydrogen evolution and the matching of the conduction band of NaTaO3 and the reduction potential of [Mo3S4] 4+ also acts as one of the major deter‐ minants for the enhancement of the photocatalytic activity [57].

## **3. Alkaline earth and transition metal tantalate-based perovskite semiconductors**

## **3.1. Synthetic methodologies of alkaline earth and transition metal tantalates**

Solid-state reaction method and hydrothermal method are used routinely to synthesize alkaline earth and transition metal tantalates. Almost all the alkaline earth and transition metal tantalates can be obtained by high-temperature solid-state method using Ta2O5 and other salts as starting materials. For instance, Sr2Ta2O7 [58], Sr0.8Bi2.2Ta2O9 [59], Bi2SrTa2O9 [60] H1.81Sr0.81Bi0.19Ta2O7 [61], Ba(Zn1/3Ta2/3)O3 [62], Ba(Mg1/3Ta2/3)O3 [63], Ba4Ta2O9 [64] and Ba5Ta4O15 [65] have been synthesized successfully by this method, which show prospects in many application including photocatalytic semiconductor, solar cells and electronic device. The high-temperature treatment of traditional solid-state reaction will increase the size of particles and thus decrease the surface area. Sr2Ta2O7 photocatalysts of layered perovskite structures gotten from the solid state reaction had better activity which is mainly because of their more negative conduction band. H2ATa2O7 (A = Sr or La2/3) [66] and LiCa2Ta3O10 [67] were reported to be obtained by similar way with extra alkali, which can supply the loss at high temperature to suppress defects formation. This makes the crystal structure grow well and has better catalytic efficiency than others synthesized with a theoretical ratio in most cases. This improved solid-state reaction method would efficiently inhibit the recombination of photo‐ carrier to enhance the photocatalytic activity. A new polymerizable complex technique is one of the preparation methods of alkaline earth tantalates, which has a relative moderate condi‐ tion. This method includes the provision of Ta-base compound and then come into being the sticky sol–gel, after the treatment at 600–700 °C. Comparing with solid-state method, the tantalate-based photocatalysts synthesized by a polymerizable complex way often have greater crystallinity and better crystal size, which will lead to remarkably increase the photocatalytic efficiency [68]. Comparing with solid-state method, hydrothermal method has been widely used in synthesizing perovskite tantalates with very lower reaction temperature. Lots of alkaline earth tantalate could be prepared by the hydrothermal method exhibiting higher activity. In 2006, Zhu and coworkers synthesize monomolecular-layer Ba5Ta4O15 nanosheets by hydrothermal method [65], which show enhanced activity ten times better than that of solid-state method-derived Ba5Ta4O15 particles in photodegradation reactions of Rhodamine B solution. Perovskite Ca2Ta2O7 has also been synthesized by hydrothermal process in aqueous KOH solution at 373 K for 120 h, which shows photocatalytic water splitting activity under UV-light irradiation [69]. Moreover, sol–gel route, as a common way to prepare the nanomaterials, also can be used for preparation of some perovskite tantalates. One typical case is that the ferroelectric SrBi2Ta2O9 [70] and SrBi2Ta2O9 nanowires [71] were synthesized using ethylene glycol as solvent, which showed greater dielectric and ferroelectric properties than the ceramics prepared by the solid-state reactions owning to a denser and more homo‐ geneous microstructure with a better distribution of grain orientations. Sol–gel method is also used to prepare metastable phase like Sr0.5TaO3 nanosheets [72] with photocatalytic activities of water splitting under ultraviolet light irradiation. Several transition metal tantalates with perovskite structure can also be synthesized by these methods, including AgTaO3 [73], LaTaO4 [74], H2La2/3Ta2O7 [75] and so forth.

## **3.2. Crystal structure engineering**

hydrogen evolution as well as Pt of an excellent co-catalyst [12]. A detailed study on the

created a solid–solution transition zone on the outer sphere of NaTaO3. The high photocatalytic activity resulting from a low NiO loading suggests that the interdiffusion of cations heavily doped the p-type NiO and n-type NaTaO3, reducing the depletion widths and facilitating charge transfers through the interface barrier [54]. Besides NiO, Ni metallic nanoclusters were also used as co-catalyst. For instance, a series of nickel-loaded La*x*Na1–*x*TaO3 photocatalysts was synthesized by a hydrogen peroxide-water based solvent method. Systematical investi‐ gation indicated that the activity of hydrogen generation from pure water is in sequence: Ni/NiO *>* NiO *>* Ni, whereas the activity sequence with respect to aqueous methanol is: Ni *>* Ni/NiO *>* NiO. Ni metallic nanoclusters exhibit the most active sites and facilitate the formation of hydrogen from aqueous methanol. In the case of Ni/NiO core/shell structure, Ni metallic nanoclusters induce the migration of photogenerated electrons from the bulk to catalyst surface, while NiO acts as H2 evolution site and prevents water formation from H2 and O2 [55].

Molecular co-catalyst engineering have received much research attention in recent years. In a molecular/semiconductor hybrid system, the noble-metal-free molecular complex as cocatalyst can not only facilitate the charge separation but also help us to understand the mechanisms of hydrogen evolution and carbon dioxide reduction at molecular level [56]. Although the study on molecular sensitized alkali tantalates is limited, an excellent research has been done by Hong and coworkers. In this case, by using a molecular co-catalyst [Mo3S4]

the photocatalytic activity of NaTaO3 was significantly improved. The hydrogen production

number of effective active sites for hydrogen evolution and the matching of the conduction

Solid-state reaction method and hydrothermal method are used routinely to synthesize alkaline earth and transition metal tantalates. Almost all the alkaline earth and transition metal tantalates can be obtained by high-temperature solid-state method using Ta2O5 and other salts as starting materials. For instance, Sr2Ta2O7 [58], Sr0.8Bi2.2Ta2O9 [59], Bi2SrTa2O9 [60] H1.81Sr0.81Bi0.19Ta2O7 [61], Ba(Zn1/3Ta2/3)O3 [62], Ba(Mg1/3Ta2/3)O3 [63], Ba4Ta2O9 [64] and Ba5Ta4O15 [65] have been synthesized successfully by this method, which show prospects in many application including photocatalytic semiconductor, solar cells and electronic device. The high-temperature treatment of traditional solid-state reaction will increase the size of particles and thus decrease the surface area. Sr2Ta2O7 photocatalysts of layered perovskite

**3. Alkaline earth and transition metal tantalate-based perovskite**

**3.1. Synthetic methodologies of alkaline earth and transition metal tantalates**

rate is about 28 times higher than pure NaTaO3 because [Mo3S4]

minants for the enhancement of the photocatalytic activity [57].

band of NaTaO3 and the reduction potential of [Mo3S4]

and Ni2+ cations

4+,

4+ clusters can provide a large

4+ also acts as one of the major deter‐

structural features of NiO nanoparticles indicated that the interdiffusion of Na+

498 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

*2.4.3. Molecular co-catalyst engineering*

**semiconductors**

For ideal perovskite alkaline earth and transition metal tantalates, the cubic-symmetry structure has the Ta atom in 6-fold coordination, surrounded by an octahedron of oxygen atoms, and the alkaline earth or transition metal cation in 12-fold cuboctahedral coordination (Figure 6). The relative ion size requirements for stability of the cubic structure are quite stringent, so slight buckling and distortion can produce several lower-symmetry distorted versions, in which the coordination numbers of cations are reduced. [77] But in fact, almost all perovskite alkaline earth and transition metal tantalates have compound perovskite structures covering two different cations at Ta site in TaO6 octahedra or at cation site in 12-fold cubocta‐ hedral coordination. This led to the chance of alternatives between ordered and disordered. Crystal structure is a very important factor manipulating the band gaps of perovskite tantalates containing the following aspects: (1) the bond angle of tantalum and oxygen ions of octahedra units; (2) the interlayer spacing of the perovskite; (3) the interaction between perovskite layers; (4) the polarization ability of cations at the interlayer toward the oxygen ions of octahedra facing the interlayer. Some of alkaline earth and transition metal tantalates with layered perovskite exhibited outstanding photo catalytic activity, including Ba5Ta4O15, H1.81Sr0.81Bi0.19Ta2O7, SrBi2Ta2O9, LaTaO4, H2La2/3Ta2O7 and Sr0.5TaO3 reported by many groups [61,65,70,72,74,75]. This kind of perovskite composites are promising materials with multiple elements, perovskite framework and layer-like structures, which can be classified into three category structures by the different interlayer structure. On the other hand, it has been reported that some perovskite alkaline earth tantalates with double-perovskite structure also show photocatalytic activity under ultraviolet light irradiation [79]. And some transition metal tantalates have simple cubic perovskite-type structure like AgTaO3, Ba3ZnTa2O9, Sr2GaTaO6 [80] and NaCaTiTaO6, NaCaTiNbO6, NaSrTiTaO6 and NaSrTiNbO6 [81].

**Figure 6.** General structure of perovskite and layered perovskite [76].

### **3.3. Metal/non-metal doping strategies for band gap engineering**

Introducing external ions into crystal structure has been generally approved as a positive way to improve the visible-light photocatalytic activity of semiconductors with larger band gaps. For nitrogen-doped layered oxide Sr5Ta4O15–*x*N*x*, the extension of the visible light absorption has been ascribed to the substitution of nitrogen for oxygen atoms as well as the formation of Ta–N bonds. The N 2p states mixed with pre-existing O 2p states shift the valence band maximum upward and result in wide visible light absorption [82]. A slight N dopant led to hinder the recombination of photo-generated charge pairs. N-doped Ba5Ta4O15 also displays a brilliant photocatalytic activity under solar condition. The doping resulted in a significant narrowing of the band gap from 4.06 eV to ca. 1.76 eV, indicating that it can use more visible light [83]. Furthermore, Sun et al. investigated affection of band gap doping with several metal and non-metal by DFT calculation [84]. It is found that, in most perovskite cases, the valence band levels were shifted upwards, in which the maximum contribution to valence band maximum comes from the p orbitals of the dopant anions, which shift. On the other hand, the dopant cations shift the CB level downwards because the CBM is chiefly governed by the d orbitals of foreign cations. This conclusion was applicable to perovskite structure tantalates system (Figure 7) directly by Liu and coworkers [85].

**Figure 7.** The electronic band edge positions with respect to the water reduction and oxidation potential levels for the pure and doped Sr2Ta2O7 systems [85].

## **3.4. Multi-component heterojunction**

units; (2) the interlayer spacing of the perovskite; (3) the interaction between perovskite layers; (4) the polarization ability of cations at the interlayer toward the oxygen ions of octahedra facing the interlayer. Some of alkaline earth and transition metal tantalates with layered perovskite exhibited outstanding photo catalytic activity, including Ba5Ta4O15, H1.81Sr0.81Bi0.19Ta2O7, SrBi2Ta2O9, LaTaO4, H2La2/3Ta2O7 and Sr0.5TaO3 reported by many groups [61,65,70,72,74,75]. This kind of perovskite composites are promising materials with multiple elements, perovskite framework and layer-like structures, which can be classified into three category structures by the different interlayer structure. On the other hand, it has been reported that some perovskite alkaline earth tantalates with double-perovskite structure also show photocatalytic activity under ultraviolet light irradiation [79]. And some transition metal tantalates have simple cubic perovskite-type structure like AgTaO3, Ba3ZnTa2O9, Sr2GaTaO6

[80] and NaCaTiTaO6, NaCaTiNbO6, NaSrTiTaO6 and NaSrTiNbO6 [81].

500 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

**Figure 6.** General structure of perovskite and layered perovskite [76].

**3.3. Metal/non-metal doping strategies for band gap engineering**

Introducing external ions into crystal structure has been generally approved as a positive way to improve the visible-light photocatalytic activity of semiconductors with larger band gaps. For nitrogen-doped layered oxide Sr5Ta4O15–*x*N*x*, the extension of the visible light absorption has been ascribed to the substitution of nitrogen for oxygen atoms as well as the formation of Ta–N bonds. The N 2p states mixed with pre-existing O 2p states shift the valence band maximum upward and result in wide visible light absorption [82]. A slight N dopant led to hinder the recombination of photo-generated charge pairs. N-doped Ba5Ta4O15 also displays a brilliant photocatalytic activity under solar condition. The doping resulted in a significant narrowing of the band gap from 4.06 eV to ca. 1.76 eV, indicating that it can use more visible

Multi-component semiconductor combination tactic shows effectivity to improve photocata‐ lytic activity by separation of the photo-generated charge carriers with a formation of a heterojunction structure. Heterojunction structure is the interface that is located at two areas of different crystalline semiconductors. This kind of material has to consider the following points, including near crystal structure, similar interatomic spacing and close coefficient of thermal expansion. Otherwise, they should have discrepant band gap values, which is exact contrary to a homojunction. It is benefited to regulate the electronic energy bands. To promote the redox ability and photocatalytic activity, composite photocatalysts involving two or more components were extensively studied. One type of such composites is usually constructed by coupling semiconductors with larger band gap for the purpose of the higher redox ability. A charming work is the Ba5Ta4O15/Ba3Ta5O15 composite reported by Roland Marschall et al., which synthesized through the sol–gel method showed brilliant activities in OH radical generation and photocatalytic hydrogen production [86]. The outstanding activity is expected to come from enhanced charge carrier separation. In 2011, Wang and coworkers present Ptloaded graphene-Sr2Ta2O7–*x*N*<sup>x</sup>* (Figure 8) with enlarged visible light absorption region and enhanced photocatalytic hydrogen generation [87].

**Figure 8.** Schematic diagram for Pt-loaded graphene-Sr2Ta2O7–*x*N*<sup>x</sup>* photocatalyst under simulated solar light irradiation [87].

## **3.5. Co-catalyst surface modification**

Transition metals and their oxides are usually used as practical co-catalysts for photocatalysis. The role of the co-catalysts attached on the surface of the semiconductor material is particularly significant. It increases the overall photocatalytic activity by helping to separate charge pairs, which can work for both bulk and surface electron/hole pathway. The chemical reaction that took place at surface is promoted by the co-catalysts. Various metals and oxides loaded on the surface of semiconductor show different effects. In most photocatalytic water splitting systems, several metals like Au and Pt can accelerate the rate of reduction of hydrogen observably [88, 89]. On the other hand, some oxides like NiO, NiO*x* and RuO2 can promote the rates of both hydrogen and oxygen production [90–92]. Among them, NiO*x* exhibited highest activity in photocatalytic process [78]. As a hydrogen evolution site, the co-catalyst has to extract the photogenerated electrons from the CB of host materials. Thus, the conduction band level of co-catalyst should be below that of photocatalyst. In addition, photocatalytic water splitting is sensitive to the deposition methods of co-catalysts. Kudo et al. reported that photocatalysts show diversity in photocatalytic water splitting with different deposition methods [59]. Moreover, transition-metal sulfides like MS (M = Ni, Co, Cu) have also been developed as cocatalysts to improve the photocatalytic activity. These sulfides have the same effects with other co-catalysts in reaction process [93].

## **4. Summary and outlook**

Tantalate-based perovskite semiconductors are well known for their wide spread applications in photocatalysis, ionic conductors, luminescence host materials and ferroelectric ceramics. Drawbacks of wide band gap and low charge separation efficiency inhibit the further devel‐ opment of tantalate-based perovskite semiconductors as superior photocatalysts. The combi‐ nation of various strategies, such as doping, heterojunction and co-catalyst engineering, induces a thrilling beginning for exploring visible light active and highly efficient photocata‐ lysts for solar energy applications. However, the studies on tantalate-based perovskite semiconductors are currently unsystematic. Meanwhile, the as-mentioned strategies and the derived photocatalytic systems with high efficiency and stability still need to be further developed.

## **Acknowledgements**

This work is financially supported by the National Natural Science Foundation of China (Grants 21267014, 21367018, 21563021), the Project of Scientific and Technological Innovation Team of Inner Mongolia University (12110614), Fund of Key Laboratory of Optoelectronic Materials Chemistry and Physics, Chinese Academy of Sciences (2008DP173016-1410).

## **Author details**

**Figure 8.** Schematic diagram for Pt-loaded graphene-Sr2Ta2O7–*x*N*<sup>x</sup>* photocatalyst under simulated solar light irradiation

Transition metals and their oxides are usually used as practical co-catalysts for photocatalysis. The role of the co-catalysts attached on the surface of the semiconductor material is particularly significant. It increases the overall photocatalytic activity by helping to separate charge pairs, which can work for both bulk and surface electron/hole pathway. The chemical reaction that took place at surface is promoted by the co-catalysts. Various metals and oxides loaded on the surface of semiconductor show different effects. In most photocatalytic water splitting systems, several metals like Au and Pt can accelerate the rate of reduction of hydrogen observably [88, 89]. On the other hand, some oxides like NiO, NiO*x* and RuO2 can promote the rates of both hydrogen and oxygen production [90–92]. Among them, NiO*x* exhibited highest activity in photocatalytic process [78]. As a hydrogen evolution site, the co-catalyst has to extract the photogenerated electrons from the CB of host materials. Thus, the conduction band level of co-catalyst should be below that of photocatalyst. In addition, photocatalytic water splitting is sensitive to the deposition methods of co-catalysts. Kudo et al. reported that photocatalysts show diversity in photocatalytic water splitting with different deposition methods [59]. Moreover, transition-metal sulfides like MS (M = Ni, Co, Cu) have also been developed as cocatalysts to improve the photocatalytic activity. These sulfides have the same effects with other

Tantalate-based perovskite semiconductors are well known for their wide spread applications in photocatalysis, ionic conductors, luminescence host materials and ferroelectric ceramics. Drawbacks of wide band gap and low charge separation efficiency inhibit the further devel‐ opment of tantalate-based perovskite semiconductors as superior photocatalysts. The combi‐ nation of various strategies, such as doping, heterojunction and co-catalyst engineering, induces a thrilling beginning for exploring visible light active and highly efficient photocata‐ lysts for solar energy applications. However, the studies on tantalate-based perovskite semiconductors are currently unsystematic. Meanwhile, the as-mentioned strategies and the derived photocatalytic systems with high efficiency and stability still need to be further

[87].

**3.5. Co-catalyst surface modification**

502 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

co-catalysts in reaction process [93].

**4. Summary and outlook**

developed.

Yiguo Su, Junyu Lang, Chunfang Du and Xiaojing Wang\*

\*Address all correspondence to: wang\_xiao\_jing@hotmail.com

College of Chemistry and Chemical Engineering, Inner Mongolia University, Hohhot, Inner Mongolia, P. R. China

## **References**


[20] Hu CC, Teng H. Influence of structural features on the photocatalytic activity of Na‐ TaO3 powders from different synthesis methods. Appl Catal A: Gen 2007;331:44–50. DOI: 10.1016/j.apcata.2007.07.024

[7] Lin H, Li L, Zhao M, Huang X, Chen X, Li G, Yu R. Synthesis of high-quality brookite TiO2 single-crystalline nanosheets with specific facets exposed: tuning catalysts from inert to highly reactive. J Am Chem Soc 2012;134:8328–31. DOI: 10.1021/ja3014049

[8] Schumann J, Eichelbaum M, Lunkenbein T, Thomas N, Galván MCÁ, Schlögl R, Beh‐ rens M. Promoting strong metal support interaction: doping ZnO for enhanced activ‐ ity of Cu/ZnO:M (M = Al, Ga, Mg) catalysts. ACS catalysis. 2015;5:3260–70. DOI:

[9] Long J, Xue W, Xie X, Gu Q, Zhou Y, Chi Y, Chen W, Ding Z, Wang X. Sn2+ dopant induced visible-light activity of SnO2 nanoparticles for H2 production. Catal Com‐

[10] Thalluri SM, Hussain M, Saracco G, Barber J, Russo N. Green-synthesized BiVO4 ori‐ ented along {040} facets for visible-light-driven ethylene degradation. Indust Eng

[11] Su Y, Peng L, Guo J, Huang S, Lv L, Wang X. Tunable optical and photocatalytic per‐ formance promoted by nonstoichiometric control and site-selective codoping of tri‐ valent ions in NaTaO3. J Phys Chem C 2014;118:10728–39. DOI: 10.1021/jp412236u

[12] Kato H, Asakura K, Kudo A. Highly efficient water splitting into H2 and O2 over lan‐ thanum-doped NaTaO3 photocatalysts with high crystallinity and surface nanostruc‐

[13] Su Y, Lang J, Cao N, Wang T, Zhu B, Wang X. Morphological reconstruction and photocatalytic enhancement of NaTaO3 nanocrystals via Cu2O loading. J Nanoparti‐

[14] Wang B, Kanhere PD, Chen Z, Nisar J, Pathak B, Ahuja R. Anion-doped NaTaO3 for visible light photocatalysis. J Phys Chem C 2013;117:22518–24. DOI: 10.1021/

[15] Kato H, Kudo A. Photocatalytic water splitting into H2 and O2 over various tantalate photocatalysts. Catal Today 2003;78:561–9. DOI: 10.1016/S0920-5861(02)00355-3

[16] Wiegel M, Emond MHJ, Stobbe ER, Blasse G. Luminescent of alkali tantalates and niobates. J Phys Chem Solid 1994;55:773–8. DOI: 10.1016/0022-3697(94)90030-2

[17] Kato H, Kudo A. New tantalate photocatalysts for water decomposition into H2 and O2. Chem Phys Lett 1998;295:487–92. DOI: 10.1016/S0009-2614(98)01001-X

[18] Kato H, Kudo A. Water splitting into H2 and O2 on alkali tantalate photocatalysts ATaO3 (A = Li, Na, and K). J Phys Chem B 2001;105:4285–92. DOI: 10.1021/jp004386b

[19] Kato H, Kudo A. Photocatalytic decomposition of pure water into H2 and O2 over SrTa2O6 prepared by a flux method. Chem Lett 1999;28:1207–8. DOI: 10.1246/cl.

mun 2011;16:215–9. DOI: 10.1016/j.catcom.2011.10.002

504 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Chem Res 2014;53:2640–6. DOI: 10.1021/ie403999g

ture. J Am Chem Soc 2003;125:3082–9. DOI: 10.1021/ja027751g

cle Res 2015;17:63–9. DOI: 10.1007/s11051-015-2877-9

10.1021/acscatal.5b00188

jp407025r

1999.1207


[45] Reitz C, Brezesinski K, Haetge J, Perlinch J, Brezesinski T. Nanocrystalline NaTaO3 thin film materials with ordered 3D mesoporous and nanopillar-like structures through PIB-*b*-PEO polymer templating: towards high-performance UV-light photo‐ catalysts. RSC Adv 2012;2:5130–3. DOI: 10.1039/c2ra20203d

[32] Kanhere PD, Zheng J, Chen Z. Site specific optical and photocatalytic properties of bi-doped NaTaO3. J Phys Chem C 2011;115:11846–53. DOI: 10.1021/jp2003936

[33] Kanhere P, Tang Y, Zheng J, Chen Z. Synthesis, photophysical properties, and photo‐ catalytic applications of Bi doped NaTaO3 and Bi doped Na2Ta2O6 nanoparticles. J

[34] Kang HW, Lim SN, Park SB, Park AA. H2 evolution under visible light irradiation on La and Cr co-doped NaTaO3 prepared by spray pyrolysis from polymeric precursor. Int J Hydrogen Energy 2013;38:6323-6334. DOI: http://dx.doi.org/10.1016/j.ijhydene.

[35] Li FF, Liu DR, Gao GM, Xue B, Jiang YS. Improved visible-light photocatalytic activi‐ ty of NaTaO3 with perovskite-like structure via sulfur anion doping. Appl Catal B

[36] Liu DR, Jiang YS, Gao GM. Photocatalytic degradation of an azo dye using N-doped NaTaO3 synthesized by one-step hydrothermal process. Chemosphere. 2011;83:1546–

[37] Wang B, Kanhere PD, Chen Z, Nisar J, Pathak B. Anion-doped NaTaO3 for visible light photocatalysis. J Phys Chem C 2013;117:22518–24. DOI: 10.1021/jp407025

[38] Chen S, Wang LW. Intrinsic defects and electronic conductivity of TaON: First-prin‐

[39] Choi M, Oba F, Tanaka I. First-principles study of native defects and lanthanum im‐ purities in NaTaO3. Phys Rev B 2008;78:014115–8. DOI: 10.1103/PhysRevB.78.014115

[40] Yong Z, Ren J, Hu H, Li P, Ouyang S, Xu H, Wang D. Synthesis, characterization, and photocatalytic of g-C3N4/KTaO3 composites under visible light irradiation. J Nano‐

[41] Kumar S, Kumar B, Surendar T, Shanker V. g-C3N4/NaTaO3 organic-inorganic hybrid nanocomposite: high-performance and recyclable visible light driven photocatalyst.

[42] Wu X, Yin S, Dong Q, Sato T. Preparation and visible light induced photocatalytic ac‐ tivity of C-NaTaO3 and C-NaTaO3-Cl-TiO2 composite. Phys Chem Chem Phys

[43] Bajorowicz B, Cybula A, Winiarski MJ, Klimczuk T, Zaleska A. Surface properties and photocatalytic activity of KTaO3, CdS, MoS2 semiconductors and their binary and ternary semiconductor composites. Molecules 2014;19:15339–60. DOI: 10.3390/

[44] Su Y, Hou L, Du C, Peng L, Guan K, Wang X. Rapid synthesis of Zn2+ doped SnWO4 nanowires with the aim of exploring doping effects on highly enhanced visible pho‐

tocatalytic activities. RSC Adv 2012;2:6266–73. DOI: 10.1039/c2ra20401k

Mater Res Bull 2014;49:310–8. DOI: 10.1016/j.materresbull.2013.09.013

ciples insights. Appl Phys Lett 2011;99:222103–3. DOI: 10.1063/1.3664346

Phys Chem Solid 2013;74:1708–13. DOI: 10.1016/j.jpcs.2013.06.013

Environ 2015;166–167:104–11. DOI: 10.1016/j.apcatb.2014.10.049

52. DOI: 10.1016/j.chemosphere.2011.01.033

506 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

mater 2015;2015:821986–7. DOI: 10.1155/2015/821986

2013;15:20633–40. DOI: 10.1039/c3cp53437e

molecules190915339

2013.03.048


[69] Ikeda S, Fubuki M, Takahara YK, Matsumura M. Photocatalytic activity of hydro‐ thermally synthesized tantalate pyrochlores for overall water splitting. Appl Catal A: Gen 2006;300:186–90. DOI: 10.1016/j.apcata.2005.11.007

[56] WenF,LiC.Hybridartificialphotosyntheticsystemscomprisingsemiconductorsaslight harvesters and biomimetic complexes as molecular cocatalysts. Account Chem Res

[57] Seo SW, Park S, Jeong HY, Kim SH, Sim U, Lee CW, Nam KT, Hong KS. Enhanced

[58] Kudo A, Kato H, Nakagawa S. Water splitting into H2 and O2 on new Sr2M2O7 (M = Nb andTa)photocatalysts withlayeredperovskite structures: factors affecting thephotoca‐

[59] Sugandha, Jha AK. Effect of the sintering temperature on nanocrystalline non-stoichio‐ metric Sr0.8Bi2.2Ta2O9 ferroelectric ceramics. Ferroelectrics 2014;459:160–71. DOI:

[60] Peng Z, Xing X, Chen X. Preparation and structure of a new layered organic-inorganic hybridbetweentheprotonatedformofaperovskiteBi2SrTa2O9 andtetraphenylporphyr‐ in by the intercalation behavior. J Alloys Compounds 2006;425:323–8. DOI: 10.1016/

[61] Li Y, Chen G, Zhou C, Li Z. Photocatalytic water splitting over a protonated layered perovskite tantalate H1.81Sr0.81Bi0.19Ta2O7. Catal Lett 2008;123:80–3. DOI: 10.1007/

[62] Nedelcu L, Toacsan MI, Banciu MG, Loachim A. Microwave properties of Ba(Zn1/3Ta2/3)O3 dielectric resonators. J Alloys Compounds 2011;509:477–81. DOI:

[63] Park TK, Kim NK. Phase development in Ba(Mg1/3Ta2/3)O3 via conventional and B-site precursor routes. Ceramics Int 2008;34:1955–8. DOI: 10.1016/j.ceramint.2007.07.013 [64] Ling CD, Avdeev M, Kharton VV, Yaremchenko AA, Macquart RB, Hoelzel M. Structures, phase transitions, hydration, and ionic conductivity of Ba4Ta2O9. Chem

[65] Xu TG, Zhang C, Shao X, Wu K, Zhu YF. Monomolecular-layer Ba5Ta4O15 nanosheets: Synthesis and investigation of photocatalytic properties. Adv Funct Mater

[66] Shimizu K, Itoh S, Hatamachi T, Kitayama Y, Kodama. Pillaring of Ruddlesden–Pop‐ per perovskite tantalates, H2ATa2O7 (A = Sr or La2/3), with n-alkylamines and oxide

[67] Ikeue K, Mitsuyama T, Arayama K, Tsutsumi A, Machida M. Effect of heat treatment on local structure and photocatalytic water splitting activity of Ni-loaded Li‐ Ca2Ta3O10. J Ceramic Soc Jpn 2009;117:1161–5. DOI: doi.org/10.2109/jcersj2.117.1161 [68] Yoshioka K, Petrykin V, Kakihana M, Kato H, Kudo A. The relationship between photocatalytic activity and crystal structure in strontium tantalates. J Cat‐

nanoparticles. J Mater Chem 2006;16:773–9. DOI: 10.1039/B514066H

4+ for water splitting into H2

2013;46:2355–64. DOI: 10.1021/ar300224u

508 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

10.1080/00150193.2013.849518

j.jallcom.2006.01.051

s10562-008-9397-5

10.1016/j.jallcom.2010.09.069

Mater 2010;22:532–540. DOI: 10.1021/cm903170t

2006;16:1599–607)DOI: 10.1002/adfm.200500849

al2005;232:102–7. DOI: 10.1016/j.jcat.2005.02.021

performance of NaTaO3 using molecular co-catalyst [Mo3S4]

and O2. Chem Commun 2012;48:10452–4. DOI: 10.1039/c2cc36216c

talytic activity. J Phys Chem B 2000;104:571–5. DOI: 10.1021/jp9919056


## **Energy Production, Decontamination, and Hydrogenation Reactions over Perovskite-Type Oxide Catalyst**

Gina Pecchi, Nestor Escalona, I. Tyrone Ghampson and Ruddy Morales

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61522

## **Abstract**

[82] ChenS,YangJ,DingC,LiR,JinS,WangD,HanH,ZhangF,LiC.Nitrogen-dopedlayered oxide Sr5Ta4O15-xNx for water reduction and oxidation under visible light irradiation. J

[83] MukherjiA,SunC,SmithSC,LuGQ,WangL.Photocatalytichydrogenproductionfrom water using N-doped Ba5Ta4O15 under solar irradiation. J Phys Chem 2011;115:15674–

[84] Guo Z, Sa B, Pathak B, Zhou J, Ahuja R, Sun Z. Band gap engineering in huge-gap semiconductor SrZrO3 for visible-light photocatalysis. Int J Hydrogen Energy

[85] Liu P, Nisar J, Ahuja R, Pathak B. Layered perovskite Sr2Ta2O7 for visible light photoca‐ talysis: a firstprinciples study.JPhys ChemC 2013;117:5043–50.DOI: 10.1021/jp310945e

[86] Marschall R, Soldat J, Wark M. Enhanced photocatalytic hydrogen generation from barium tantalate composites. Photochem Photobiol Sci 2013;12:671–7. DOI: 10.1039/

[87] Mukherji A, Seger B, Lu GQ, Wang L. Nitrogen doped Sr2Ta2O7 coupled with graphene sheets as photocatalysts for increased photocatalytic hydrogen production. ACS Nano

[88] Iwase A, Kato H, Kudo A. Nanosized Au particles as an efficient cocatalyst for pho‐ tocatalytic overall water splitting. Catal Lett 2006;108:7–10. DOI: 10.1007/

[89] Hagiwara H, Inoue T, Kaneko K, Ishihara T. Charge-transfer mechanism in Pt/ KTa(Zr)O3 photocatalysts modified with porphyrinoids for water splitting. Chem –

[90] Sreethawong T, Ngamsinlapasathian S, Suzuki Y, Yoshikawa S. Nanocrystalline mes‐ oporous Ta2O5-based photocatalysts prepared by surfactant-assisted templating solgel process for photocatalytic H2 evolution. J Molecul Catal A – Chem 2005;235:1–11.

[91] Chen X, Chen W, Gao H, Yang Y, Shangguan W. In situ photodeposition of NiOx on CdS for hydrogen production under visible light: Enhanced activity by controlling solution environment. Appl Catal B: Environ 2014;152–153:68–72. DOI: 10.1016/

[92] Jorge AB, Martin DJ, Dhanoa MTS, Rahman AS, Makwana N, Tang J, Sella A, Corà F, Firth S, Darr JA, McMillan PF. H2 and O2 evolution from water half-splitting reac‐ tions by graphitic carbon nitride materials. J Phys Chem C 2013;117:7178–85. DOI:

[93] Wang J, Li B, Chen J, Li N, Zheng J, Zhao J, Zhu Z. Enhanced photocatalytic H2-pro‐ duction activity of CdxZn1-xS nanocrystals by surface loading MS (M = Ni, Co, Cu)

species. Appl Surface Sci 2012;259:118–23. DOI: 10.1016/j.apsusc.2012.07.003

Mater Chem A 2013;1:5651–9. DOI: 10.1039/C3TA10446J

510 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

2014;39:2042–8. DOI: 10.1016/j.ijhydene.2013.11.055

2011;5:3483–92. DOI: 10.1021/nn102469e

DOI: 10.1016/j.molcata.2005.03.021

Euro J 2009;15:12862–70. DOI: 10.1002/chem.200901772

8. DOI: 10.1021/jp202783t

C2PP25200G

s10562-006-0030-1

j.apcatb.2014.01.022

10.1021/jp4009338

Heterogeneous catalysis is an important tool in industrial processes because of the re‐ coverability of the catalysts. Transition metal perovskites-type oxides, with the general formula ABO3, offer attractive alternative to noble metal catalysts due to their high ac‐ tivity, high thermal stability, and low cost. Moreover, their physicochemical properties can be tailored to create a family of catalysts by varying the compositions of A and B. Indeed, the partial substitution at the A- and/or B-site with another metal cation stabil‐ izes unusual oxidation states of the B cation with the simultaneous formation of struc‐ tural defects. In particular, lanthanum-based perovskites have been used extensively and can be grouped into: (i) perovskites with oxygen vacancies as catalysts for oxida‐ tion reactions and (ii) perovskites as precursors to prepare nanosized catalysts for hy‐ drogenation reactions. This chapter focuses on the use of pure and doped lanthanum perovskites as active and selective heterogeneous catalysts for catalytic energy produc‐ tion reaction (DME combustion), decontamination reactions (methane, acetyl acetate, toluene, n-hexane, and soot combustion), and hydrogenation reactions (guaiacol, glyc‐ erol, and xylose hydrogenation).

**Keywords:** Lanthanum, substituted, oxidation, hydrogenation, energy

## **1. Introduction**

Perovskite-type oxides have been extensively studied since the pioneering work of Voorhoeve et al. [1] in the early 1970s on their potential as a cheaper alternative to noble metals as an automobile exhaust catalyst. Since then, the literature has been inundated with studies expanding their applications, particularly involving their use in combustion. Perovskites have

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

been demonstrated to be effective in several oxidation reactions. For environmental processes, the main objective is to achieve high conversions with almost no by-products or pollutants. Recently, there has been a particular emphasis on developing catalysts capable of either transforming pollutants into less harmful products or valorizing the by-products.

The tendency of perovskites to form unique structures is responsible for their reactivity. In general, perovskite-type oxides are represented by the general formula ABO3, wherein the larger A-site metals, usually rare earth metals, have a dodecahedral coordination and the smaller B-site elements, first-row transition metals, have a sixfold coordination. The ability to tailor the properties of perovskites due to the different combinations of elements employed and the high degree of flexibility in their composition allow for their use in a wide range of applications.

In heterogeneous catalysis, the creation of oxygen vacancies, generally obtained by substitu‐ tion in the A- and B-site, accounts for their use in catalyzing chemically and environmentally relevant reactions, such as oxidation reactions. For pure perovskites-type oxides (ABO3), the combustion activity is dependent mainly on the B component oxides. In our previous work, we have mainly utilized Mn, Fe, Co, and Ni for the B-site of the perovskite-oxides due to their high catalytic activities; on the other hand, we have exclusively used La cation for the A-site position.

In addition to the research conducted into perovskite-based oxidation reactions, advances in preparation methods have led to perovskites being used as precursors to prepare nanosized catalysts for hydrogenation reactions. This is achieved by reduction or reduction–oxidation cycles of perovskites-type oxides under controlled conditions to leave the metal in the B-site in a high degree of dispersion on a matrix of the Ln2O3 oxide. Studies have demonstrated that during the reduction–oxidation cycles of LnCoO3 (Ln = La, Pr, Nd, Sm, Gd) perovskites, the stability of the perovskite is affected by the nature of the Ln, with the largest lanthanide ion in the series, La, forming the most stable perovskite structure.

This chapter focuses on the use of lanthanum perovskites as active and selective catalysts in heterogeneous reactions for catalytic energy production, decontamination, and hydrogenation reactions.

## **2. Energy production**

## **2.1. Fischer-Tropsch reaction**

The Fischer-Tropsch synthesis (FTS) is the most important catalytic process for the synthesis of gasoline and/or diesel from syngas (a mixture of CO and H2 formed by methane reforming) [2]. Efforts are being made worldwide to shift from conventional feedstock such as coal due to their negative environmental impact, and biomass appears to be an attractive alternative as a feedstock for the production of hydrocarbons. The gasification of biomass stream produces a gas mixture called "biosyngas" consisting mainly of CO2, CO, CH4, H2, and N2. The H2/CO ratio in the resulting biosyngas is close to 1 and can be adjusted using the water–gas shift reaction. The advantages of hydrocarbons produced from FTS using a biosyngas feed include being near CO2-neutral and being free of sulfur and nitrogen compounds. However, there is a dearth of systematic studies of catalytic systems in FTS using biosyngas [3-5]. Perovskites have been shown to be highly active in FTS using a H2/CO mixture, and its activity was related to the properties of the perovskite precursor: stability, crystalline structure, and degree of reduction [6, 7]. These results paved the way for the evaluation of perovskites for FTS of biosyngas. Hence, this study was focused on FTS over LaFe1−xCoxO3 perovskites using biosyngas as the feed.

## **2.2. B-site substitution**

been demonstrated to be effective in several oxidation reactions. For environmental processes, the main objective is to achieve high conversions with almost no by-products or pollutants. Recently, there has been a particular emphasis on developing catalysts capable of either

The tendency of perovskites to form unique structures is responsible for their reactivity. In general, perovskite-type oxides are represented by the general formula ABO3, wherein the larger A-site metals, usually rare earth metals, have a dodecahedral coordination and the smaller B-site elements, first-row transition metals, have a sixfold coordination. The ability to tailor the properties of perovskites due to the different combinations of elements employed and the high degree of flexibility in their composition allow for their use in a wide range of

In heterogeneous catalysis, the creation of oxygen vacancies, generally obtained by substitu‐ tion in the A- and B-site, accounts for their use in catalyzing chemically and environmentally relevant reactions, such as oxidation reactions. For pure perovskites-type oxides (ABO3), the combustion activity is dependent mainly on the B component oxides. In our previous work, we have mainly utilized Mn, Fe, Co, and Ni for the B-site of the perovskite-oxides due to their high catalytic activities; on the other hand, we have exclusively used La cation for the A-site

In addition to the research conducted into perovskite-based oxidation reactions, advances in preparation methods have led to perovskites being used as precursors to prepare nanosized catalysts for hydrogenation reactions. This is achieved by reduction or reduction–oxidation cycles of perovskites-type oxides under controlled conditions to leave the metal in the B-site in a high degree of dispersion on a matrix of the Ln2O3 oxide. Studies have demonstrated that during the reduction–oxidation cycles of LnCoO3 (Ln = La, Pr, Nd, Sm, Gd) perovskites, the stability of the perovskite is affected by the nature of the Ln, with the largest lanthanide ion in

This chapter focuses on the use of lanthanum perovskites as active and selective catalysts in heterogeneous reactions for catalytic energy production, decontamination, and hydrogenation

The Fischer-Tropsch synthesis (FTS) is the most important catalytic process for the synthesis of gasoline and/or diesel from syngas (a mixture of CO and H2 formed by methane reforming) [2]. Efforts are being made worldwide to shift from conventional feedstock such as coal due to their negative environmental impact, and biomass appears to be an attractive alternative as a feedstock for the production of hydrocarbons. The gasification of biomass stream produces a gas mixture called "biosyngas" consisting mainly of CO2, CO, CH4, H2, and N2. The H2/CO ratio in the resulting biosyngas is close to 1 and can be adjusted using the water–gas shift

the series, La, forming the most stable perovskite structure.

transforming pollutants into less harmful products or valorizing the by-products.

512 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

applications.

position.

reactions.

**2. Energy production**

**2.1. Fischer-Tropsch reaction**

The FTS reaction using a simulated gas mixture similar to biosyngas over LaFe1-xCoxO3 perovskites presented a wide distribution of liquid hydrocarbons (C6 to C18+) and a high production of CH4 attributed to Co metal segregation on the surface. The distribution of the liquid products was related to the average size of iron particles [8].

A series of LaFe1-xCoxO3 perovskites (0.0 ≤ x ≤ 1.0) were prepared using the amorphous citrate precursor method and characterized by N2 physisorption, X-ray diffraction (XRD), Fouriertransform IR spectroscopy (FTIR), O2-Temperature programmed desorption (O2-TPD), Thermal-programmed reduction (TPR), and Scanning electron microscopy (SEM). FTS was carried out in a stainless steel fixed-bed reactor at 300°C and 1 MPa. The BET surface areas of the samples (13 m2 g-1) were as expected for these kinds of structures. The reduction behaviors of the catalysts (TPR1-TPR2) are shown in Fig. 1. This analysis provides information on the thermal stability of the catalyst (segregation of Co particles) under reducing conditions, deduced from the reducibility of the oxide precursor. The procedure involved: (i) a first reduction step under H2/Ar flow up to 700°C (TPR1); (ii) cooling under Ar flow to room temperature; (iii) an oxidation step under flowing O2/He mixture up to 700°C; and (iv) a second reduction treatment (TPR2) as in step 1. Figure 1 shows that the largest differences between TPR1 and TPR2 were observed for pure LaFeO3 and LaCoO3 perovskites, while the substituted perovskites showed similar profiles. Pure LaFeO3 is essentially irreducible. On the other hand, the profile for pure LaCoO3 revealed two well-defined peaks: a peak at 350°C is indicative of the formation of an oxygen-deficient perovskite structure, while the second peak above 500°C corresponds to the complete reduction of the perovskite to form Co0 . The substitution of Fe3+ by Co3+ affected the initial reduction stage, showing only one reduction peak, probably by passing an intermediate reduction state. The subsequent reduction of Co3+/Co2+ to Co0 occurred at a higher temperature, then decreased in intensity with increasing Fe, and subsequently disappeared when the level of substitution reached 50%. The slight difference in the TPR2 profiles in comparison with TPR1 is indicative of the reversibility of the redox cycle. The small and broad peaks observed at 410°C for LaFeO3 in the TPR2 profile are associated with Fe3+ segregation during the first reduction step, since this perovskite is a Co-free sample and La is not reducible under the experimental conditions. Thus, during TPR1, metallic iron is expelled from the perovskite structure at 620°C. The quantification of the first reduction peak at temperatures below 500°C is representative of the extent of Co insertion into the perovskite structure: our data revealed a near linear increase in hydrogen consumption with x, an indication of the progressive insertion of Co into the perovskite structure. The XRD patterns of the calcined perovskites displayed two crystal systems, orthorhombic for x<0.5 and rhombohedral for x≥ 0.5. On the other hand, XRD analyses of reduced perovskite showed no significant changes after H2 reduction, indicative of the high thermal stability of these LaFe1 xCoxO3 perovskite solid solutions. The diffraction lines broadened upon substitution of Fe3+ by Co3+, which suggests a decrease in the crystallinity. The highest stability under hydrogen was obtained for the perovskite with x = 0.5, while the thermal stability of x = 0.1 and 0.3 perovskites were similar to that of the pure LaCoO3 and x = 0.4 perovskites, respectively. The observed behavior of the x = 0.2 perovskite indicates a loss in the perovskite structure, suggesting segregation of the B cation. FTIR spectra indicated that for the x ≤ 0.5 perovskites, Co3+ ions were incorporated into the orthorhombic LaFeO3 perovskite structure and that a new rhom‐ bohedral structure appeared for x = 0.5. The pure LaFeO3 tolerated Co insertion with no structural changes up to values of x ≤ 0.3. For x = 0.4 and x = 0.5, the presence of a mixture of orthorhombic and rhombohedral structures were detected. The catalytic activity indicated a high intrinsic CO conversion for the substitution degree of x = 0.1, 0.2, and 1.0, in line with their orthorhombic structures and the presence of segregated Co metal. The x = 0.2 perovskite displayed the highest CO intrinsic conversion attributed to the presence of highly dispersed, segregated Co species. The absence of segregated Co species in the x = 0.3 and 0.4 perovskites explains their lower activities. In fact, these perovskites possessed higher thermal stability under reducing atmospheres due to the presence of a mixture of orthorhombic and rhombo‐ hedral crystalline structures, which implies that they did not segregate Co species. In relation to selectivity, an increase in x led to a decrease in CH4 selectivity, while a decrease in x favored selectivity to longer-chain length condensable hydrocarbons: this was attributed to changes in the average particle size of segregated Co species. The smaller particles of the x = 0.2 and 0.1 perovskites favored the formation of longer-chain hydrocarbons. Pure LaCoO3 perovskite displayed the highest extent of poorly dispersed segregated Co, favoring the formation of short-chain hydrocarbons (C8-C9). Thus, the catalytic activity of LaFe1-xCoxO3 perovskites in FTS can be explained in terms of the crystalline structure, segregated Co content, and particle size.

**Figure 1.** Temperature-programmed reduction profiles of LaFe1-xCoxO3 perovskites: left: TPR1 and right: TPR2

## **3. Dimethyl Ether (DME) combustion**

The easier transportation, lower emission of polluting particles, and higher thermal efficiency of dimethyl ether (DME) make it a useful, low-cost replacement for liquefied petroleum gas (LPG) and diesel fuel. Recently, the catalytic combustion of DME has been studied on Cedoped manganese oxide octahedral molecular sieves (OMS-2) [9]. Mixed oxides with perov‐ skite-type structures, in particular LaMnO3.15, have the potential to be highly active catalysts for the deep oxidation of DME. This catalyst has been previously used for the catalytic oxidation of toluene [10], carbon monoxide [11], and NH3 [12].

## **3.1. B-site substitution**

indication of the progressive insertion of Co into the perovskite structure. The XRD patterns of the calcined perovskites displayed two crystal systems, orthorhombic for x<0.5 and rhombohedral for x≥ 0.5. On the other hand, XRD analyses of reduced perovskite showed no significant changes after H2 reduction, indicative of the high thermal stability of these LaFe1 xCoxO3 perovskite solid solutions. The diffraction lines broadened upon substitution of Fe3+ by Co3+, which suggests a decrease in the crystallinity. The highest stability under hydrogen was obtained for the perovskite with x = 0.5, while the thermal stability of x = 0.1 and 0.3 perovskites were similar to that of the pure LaCoO3 and x = 0.4 perovskites, respectively. The observed behavior of the x = 0.2 perovskite indicates a loss in the perovskite structure, suggesting segregation of the B cation. FTIR spectra indicated that for the x ≤ 0.5 perovskites, Co3+ ions were incorporated into the orthorhombic LaFeO3 perovskite structure and that a new rhom‐ bohedral structure appeared for x = 0.5. The pure LaFeO3 tolerated Co insertion with no structural changes up to values of x ≤ 0.3. For x = 0.4 and x = 0.5, the presence of a mixture of orthorhombic and rhombohedral structures were detected. The catalytic activity indicated a high intrinsic CO conversion for the substitution degree of x = 0.1, 0.2, and 1.0, in line with their orthorhombic structures and the presence of segregated Co metal. The x = 0.2 perovskite displayed the highest CO intrinsic conversion attributed to the presence of highly dispersed, segregated Co species. The absence of segregated Co species in the x = 0.3 and 0.4 perovskites explains their lower activities. In fact, these perovskites possessed higher thermal stability under reducing atmospheres due to the presence of a mixture of orthorhombic and rhombo‐ hedral crystalline structures, which implies that they did not segregate Co species. In relation to selectivity, an increase in x led to a decrease in CH4 selectivity, while a decrease in x favored selectivity to longer-chain length condensable hydrocarbons: this was attributed to changes in the average particle size of segregated Co species. The smaller particles of the x = 0.2 and 0.1 perovskites favored the formation of longer-chain hydrocarbons. Pure LaCoO3 perovskite displayed the highest extent of poorly dispersed segregated Co, favoring the formation of short-chain hydrocarbons (C8-C9). Thus, the catalytic activity of LaFe1-xCoxO3 perovskites in FTS can be explained in terms of the crystalline structure, segregated Co content, and particle

514 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

size.

Signal, a.u.

xCo= 1.0 xCo= 0.5 xCo= 0.4 xCo= 0.3 xCo= 0.2 xCo= 0.1

xCo= 0.0

100 200 300 400 500 600 700

100 200 300 400 500 600 700

Temperature, °C

xCo= 1.0 xCo= 0.5 xCo= 0.4 xCo= 0.3 xCo= 0.2

xCo= 0.1 xCo= 0.0

Signal, a.u.

**Figure 1.** Temperature-programmed reduction profiles of LaFe1-xCoxO3 perovskites: left: TPR1 and right: TPR2

Temperature, °C

The electronic properties and performance of La(Mn0.5B0.5)O3.15 (B = Cr, Fe, Co, Ni, Cu) perovskites in the combustion of dimethyl ether (DME) revealed that an increase in the catalytic performance, expressed as intrinsic activity (μmolm-2h-1), is contingent on B-site substitution in LaMnO3 and a decrease in the required calcination temperature to obtain the perovskite structure. The similar tolerance factor (*t*) of the substituted perovskites, a measure of the geometric effect, plays a secondary role on the surface and catalytic performance. Consequently, the electronic effect explains the relationship between catalytic performance, chemisorbed oxygen species, and reducibility of the substituted perovskites. XRD results indicated that calcined pure LaMnO3.15 and Cr-, Co-, and Fe-substituted perovskites presented well-defined rhombohedral structure, whereas Cu- and Ni-doped perovskites displayed La2CuO4 and La2NiO4 phases. The largest surface area corresponded to the LaMnO3.15 perovskite with 11 m2 g-1, and subsequently decreased upon B-substitution. The TPR profiles obtained for the pure and oxygen-rich LaMnO3.15 perovskite presented a concentration of 32% Mn4+ ions after H2 reduction. An observed reduction peak at 350°C with a shoulder at 420°C corresponds to the reduction of Mn4+ to Mn3+, followed by the destruction of the perovskite structure (Mn3+ to Mn2+) at 800°C [13-15]. Upon B- substitution, the two reductions steps of pure LaMnO3.15 perovskite were maintained when Co and Ni were used, while only one reduction step was detected for the Cr- and Fe-substituted perovskite. This result is closely related to the thermal behavior of nonreducible perovskites such as LaCrO3 and LaFeO3. Thus, Fe- and Cr-substituted perovskites showed lower reducibility compared to the parent LaMnO3.15 but higher reducibility in comparison with the nonreducible LaCrO3 and LaFeO3 perovskites. The observed increase in the reducibility of the Co- and Ni-substituted perovskites is similar to the effect induced by La substitution of divalent alkaline-earth cations. The high stabilities of Co2+ and Ni2+ helped stabilize the Mn4+ ions, with corresponding increase in the reduction peaks. On the other hand, the TPR results for the Cu-substituted perovskite showed a single-step complete reduction of the Cu ions to the metallic state, similar to the reduction of CuO but at lower temperatures. The O2-TPD experiments were used to evaluate the redox properties of the catalysts. The data revealed two different behaviors, consistent with the different patterns observed from TPR results: for the Cr- and Fe-substituted perovskites, a behavior similar to the parent LaMnO3.15 perovskite was observed, with the presence of α- and β-oxygen species beginning at ∼600°C. Conversely, the absence of β-oxygen at temperatures below 700°C for the Co-, Ni-, and Cu-substituted perovskites is related to the lattice oxygendeficient structures. XPS analyses showed that the La3d5/2 spectra presented peaks with BE at 834.3 eV and 838.1 eV that correspond to La3+ in the perovskite. The small difference between the BEs of Mn3+ at 641.3 eV and Mn4+ at 642.4 eV makes it difficult to distinguish them. Regarding the Cr 2p3/2 spectra, two peaks with BEs centered at 576.6 and 579.4 eV were detected: the peaks were assigned to surface Cr3+ and Cr6+ species, respectively. The main peak of the Fe 2p3/2 spectra was located at 710.8 eV, and is attributed to surface Fe3+ species. For the Co 2p3/2 spectra, the two peaks detected at 780.6 and 786.9 eV are associated with Co3+ and Co2+ species, respectively. Because the Ni 2p3/2 signal coincides with the La 3d3/2 signal, the BE of Ni 2p1/2 was measured instead. The displacement of the Ni BE to 873.2 eV suggests the presence of surface Ni2+ species. Considering the low BET surface area of the substituted perovskites, the surface composition was similar to the nominal composition in the bulk, and the surface analysis may be extrapolated to the bulk. The BE of O 1s peak less than 532 eV suggests the absence of surface oxygen vacancies. Therefore, the substitution of Mn3+ with a first-row transition metal in the (III) oxidation state preserved the oxygen-rich perovskite structure for the LaMn0.5Cr0.5O3.15 and LaMn0.5Fe0.5O3.15 perovskites. On the other hand, the substitution by Co2+, Ni2+, and Cu2+ generated a stoichiometric cubic perovskite structure. These differences in behavior are in agreement with interpretations deduced from TPR and O2-TPD results. The catalytic activity of the B-site substituted perovskites were tested in the combustion of DME with excess oxygen in a flow reactor. The light-off curves allow the calculation of the reaction rate of the oxidation reaction at low conversion levels (<10%) at 180°C. The reaction rate was correlated with the H2 consumption calculated from TPR profiles and the α-oxygen derived from O2-TPD (both parameters normalized by the specific surface area), and shown in Fig. 2. It can be inferred from this correlation that the high stability of (III) oxidation state of Cr3+ and Fe3+cations did not increase the redox character of the Mn4+/Mn3+ pair; on the other hand, the higher redox properties and catalytic performances of the LaMn0.5Co0.5O3 and LaMn0.5Ni0.5O3 catalysts stem from the high stability of (II) oxidation state of Co2+ and Ni2+, leading to a progressive decrease of Mn3+ species, while stabilizing the Mn4+ species in the network.

## **4. Decontamination reactions**

## **4.1. VOCs abatement**

Metal oxides with a perovskite structure have been consistently proposed during the last two decades as alternative catalysts for the deep oxidation of hydrocarbons. It is well-known that perovskite-like mixed oxide substitution of the trivalent A-site metal ion with a bivalent or tetravalent metal cation (A') is accompanied by a modification of the oxidation state of the Bsite metal cation, thus modifying its catalytic activity. Moreover, modification of the oxidation state of the B-site metal cation by insertion of A' may be accompanied by the formation of structural defects, thus leading to nonstoichiometry. This usually means the creation of oxygen defects in the Co-containing perovskites, while in the Mn-perovskites it results in excess of oxygen [13]. Fe-based [12] and Ni-based perovskites [14] were shown to exhibit an intermedi‐ ate behavior. Good catalytic performances in combustion reactions were displayed by La- or

Energy Production, Decontamination, and Hydrogenation Reactions over Perovskite-Type Oxide Catalyst http://dx.doi.org/10.5772/61522 517

deficient structures. XPS analyses showed that the La3d5/2 spectra presented peaks with BE at 834.3 eV and 838.1 eV that correspond to La3+ in the perovskite. The small difference between the BEs of Mn3+ at 641.3 eV and Mn4+ at 642.4 eV makes it difficult to distinguish them. Regarding the Cr 2p3/2 spectra, two peaks with BEs centered at 576.6 and 579.4 eV were detected: the peaks were assigned to surface Cr3+ and Cr6+ species, respectively. The main peak of the Fe 2p3/2 spectra was located at 710.8 eV, and is attributed to surface Fe3+ species. For the Co 2p3/2 spectra, the two peaks detected at 780.6 and 786.9 eV are associated with Co3+ and Co2+ species, respectively. Because the Ni 2p3/2 signal coincides with the La 3d3/2 signal, the BE of Ni 2p1/2 was measured instead. The displacement of the Ni BE to 873.2 eV suggests the presence of surface Ni2+ species. Considering the low BET surface area of the substituted perovskites, the surface composition was similar to the nominal composition in the bulk, and the surface analysis may be extrapolated to the bulk. The BE of O 1s peak less than 532 eV suggests the absence of surface oxygen vacancies. Therefore, the substitution of Mn3+ with a first-row transition metal in the (III) oxidation state preserved the oxygen-rich perovskite structure for the LaMn0.5Cr0.5O3.15 and LaMn0.5Fe0.5O3.15 perovskites. On the other hand, the substitution by Co2+, Ni2+, and Cu2+ generated a stoichiometric cubic perovskite structure. These differences in behavior are in agreement with interpretations deduced from TPR and O2-TPD results. The catalytic activity of the B-site substituted perovskites were tested in the combustion of DME with excess oxygen in a flow reactor. The light-off curves allow the calculation of the reaction rate of the oxidation reaction at low conversion levels (<10%) at 180°C. The reaction rate was correlated with the H2 consumption calculated from TPR profiles and the α-oxygen derived from O2-TPD (both parameters normalized by the specific surface area), and shown in Fig. 2. It can be inferred from this correlation that the high stability of (III) oxidation state of Cr3+ and Fe3+cations did not increase the redox character of the Mn4+/Mn3+ pair; on the other hand, the higher redox properties and catalytic performances of the LaMn0.5Co0.5O3 and LaMn0.5Ni0.5O3 catalysts stem from the high stability of (II) oxidation state of Co2+ and Ni2+, leading to a progressive decrease of Mn3+ species, while stabilizing the Mn4+ species in the

516 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Metal oxides with a perovskite structure have been consistently proposed during the last two decades as alternative catalysts for the deep oxidation of hydrocarbons. It is well-known that perovskite-like mixed oxide substitution of the trivalent A-site metal ion with a bivalent or tetravalent metal cation (A') is accompanied by a modification of the oxidation state of the Bsite metal cation, thus modifying its catalytic activity. Moreover, modification of the oxidation state of the B-site metal cation by insertion of A' may be accompanied by the formation of structural defects, thus leading to nonstoichiometry. This usually means the creation of oxygen defects in the Co-containing perovskites, while in the Mn-perovskites it results in excess of oxygen [13]. Fe-based [12] and Ni-based perovskites [14] were shown to exhibit an intermedi‐ ate behavior. Good catalytic performances in combustion reactions were displayed by La- or

network.

**4. Decontamination reactions**

**4.1. VOCs abatement**

**Figure 2.** Intrinsic activity as a function of quantity of hydrogen consumed (above) and quantity of α-oxygen desorbed

LaxSr1-x-based perovskites containing Co, Fe, or Mn as B cation [15, 16]. A major drawback for their application, however, is their low surface area, and hence significant efforts are underway to increase the specific surface area of conventional A1-xA'xBO3-δ perovskites through various preparation methods [17, 18].

The insertion of calcium ions into the LaFeO3 perovskite lattice results in a solid with structural and electronic defects possessing highly active sites for redox reactions. A continuous increase in number of oxygen vacancies in LaFeO3 perovskites upon A-site substitution of La3+ by Ca2+ was previously reported [19]. In this work, LaxCa1-xFeO3 (x = 0.0, 0.1, 0.2, 0.3, 0.4) perov‐ skites prepared by precursor citrate method were characterized and evaluated in catalytic combustion reaction. The effect of La substitution by calcium on the physical, chemical, and catalytic properties for the total combustion of acetyl acetate and methane was studied. Surface area values ranging from 19 to 39 m2 g-1 were obtained as the calcium content of the perovskites varied from 0.0 to 0.4. XRD patterns indicated that for pure LaFeO3 an orthorhombic perovskite structure was observed; on the other hand, a shift toward larger 2θ angles appeared as La3+ was partially substituted by Ca2+, evidence of lattice distortion. Even though the ionic radii are very similar (La3+ = 0.136 nm; Ca2+ = 0.134 nm), the substitution of La3+ with a cation of lower valence like Ca2+ entails either the oxidation of the cation at the B-site (i.e., Fe3+ → Fe4+) or the formation of oxygen vacancies in order to retain electroneutrality. Since the ionic radius of Fe4+ (0.0585 nm) is smaller than that of Fe3+ (0.0645 nm), the distortion can be explained by the presence of Fe4+ in the perovskite structure, in agreement with previous results reported for other iron-based perovskites. The shift of the 2θ angles as a function of the degree of calcium substitution is illustrated in Fig. 3: a linear shift of the 2θ values is observed with increasing Ca content, a clear indication of the insertion of calcium in the studied range.

**Figure 3.** Shift of the peak position as a function of substitution degree 2*θ* (●) 40°; (■) 46°; (▲) 58°

IR spectra of the perovskites show a band at 865 cm-1, attributed to O-C-O bending in the carbonates. This peak appeared as a shoulder for x = 0.1 and increased linearly when x ≥ 0.2, without showing any change in the wavelength. This band is ascribed to the presence of calcium and corresponds to the out-of-plane bending of the carbonate ions supporting the incorporation of calcium in the perovskite structure. O2-TPD profiles showed clear increase of the desorbed oxygen as the degree of Ca substitution increased. In addition, an appreciable desorption peak was observed at temperatures above 500°C for the perovskite with the highest degree of substitution. This desorbed oxygen corresponds to the so-called β oxygen, which is associated with lattice oxygen or with oxygen species occupying the inner vacancies created by substitution of Ca by La. The plot of the amount of O2 desorbed from 150°C up to 700°C, which involves α-oxygen species, as a function of Ca substitution showed a continuous increase. Desorption of large amount of lattice oxygen in the bulk from the samples indicates a large substitution vacancy and/or defects. TPR profiles for the pure LaFeO3 showed no reduction peaks, indicative of the nonreducibility of this perovskite. However, the TPR profile for the substituted perovskites showed a pronounced peak at lower temperatures that shifted to higher temperatures and increased in intensity as the degree of Ca substitution increased. The replacement of part of La3+ by Ca2+ led to a well-defined H2 consumption peak due to the higher amount of Fe4+ ions generated to compensate for the unbalanced charge, consistent with the deduction from XRD results. In situ XRD measurements also revealed no detectable differences between diffraction patterns of pre- and post-TPR measurements, suggesting that the integrity of the structure was retained after hydrogen treatment, although the nature of the phases present in the sample changed. The post-TPR XRD pattern of the pure LaFeO3 perovskite was essentially the same as that obtained for the fresh perovskite: the only differ‐ ence was a minimal increase in the intensity of the peaks and small shifts toward lower angles, indicating increased crystallinity with a great expansion of the cell parameters. The catalytic activity in the oxidation of methane and acetyl acetate in a flow reactor using an excess amount of oxygen showed two contrasting behaviors: a near independency of the ignition temperature with Ca substitution for the methane reaction; and a decrease in ignition temperature with Ca substitution for the acetyl acetate reaction. These results clearly demonstrate the differences between a suprafacial and intrafacial reaction mechanism. The presence of substituted Ca ions enhances the catalytic activity of suprafacial reactions (e.g., acetyl acetate combustion), which typically occurs at lower temperatures. Conversely, the activity of intrafacial reactions (e.g., methane combustion), which requires a much higher temperature, is only slightly modified by Ca2+ insertion into the perovskite.

## **4.2. A-site substitution**

combustion reaction. The effect of La substitution by calcium on the physical, chemical, and catalytic properties for the total combustion of acetyl acetate and methane was studied. Surface

varied from 0.0 to 0.4. XRD patterns indicated that for pure LaFeO3 an orthorhombic perovskite structure was observed; on the other hand, a shift toward larger 2θ angles appeared as La3+ was partially substituted by Ca2+, evidence of lattice distortion. Even though the ionic radii are very similar (La3+ = 0.136 nm; Ca2+ = 0.134 nm), the substitution of La3+ with a cation of lower valence like Ca2+ entails either the oxidation of the cation at the B-site (i.e., Fe3+ → Fe4+) or the formation of oxygen vacancies in order to retain electroneutrality. Since the ionic radius of Fe4+ (0.0585 nm) is smaller than that of Fe3+ (0.0645 nm), the distortion can be explained by the presence of Fe4+ in the perovskite structure, in agreement with previous results reported for other iron-based perovskites. The shift of the 2θ angles as a function of the degree of calcium substitution is illustrated in Fig. 3: a linear shift of the 2θ values is observed with increasing

0,0 0,1 0,2 0,3 0,4

Substitution degree

IR spectra of the perovskites show a band at 865 cm-1, attributed to O-C-O bending in the carbonates. This peak appeared as a shoulder for x = 0.1 and increased linearly when x ≥ 0.2, without showing any change in the wavelength. This band is ascribed to the presence of calcium and corresponds to the out-of-plane bending of the carbonate ions supporting the incorporation of calcium in the perovskite structure. O2-TPD profiles showed clear increase of the desorbed oxygen as the degree of Ca substitution increased. In addition, an appreciable desorption peak was observed at temperatures above 500°C for the perovskite with the highest degree of substitution. This desorbed oxygen corresponds to the so-called β oxygen, which is associated with lattice oxygen or with oxygen species occupying the inner vacancies created

Ca content, a clear indication of the insertion of calcium in the studied range.

**Figure 3.** Shift of the peak position as a function of substitution degree 2*θ* (●) 40°; (■) 46°; (▲) 58°

g-1 were obtained as the calcium content of the perovskites

area values ranging from 19 to 39 m2

518 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

0,0

0,2

0,4

, degrees

0,6

0,8

A different behavior was detected upon the substitution of La2+ by Ca2+ in LaNiO3 and LaFeO3 perovskite-type oxides. In the Fe series (La1-xCaxFeO3), the perovskite structure is maintained within the studied range of substitution. In contrast, a completely different behavior was observed for the Ni series (La1-xCaxNiO3), whereby a Brownmillerite-type structure (La2Ni2O5) with a large degree of segregated phase as highly dispersed mixed oxides was observed [20]. There was no monotonic dependence of the SBET values on the extent of La substitution. The XRD patterns shown in Fig. 4 revealed the presence of orthorhombic structure for the Fe series; however, for the pure LaNiO3 counterpart, the peak at 2θ = 43.3° indicated the presence of segregated NiO. For the Ni series, the most prominent diffraction line at 2θ = 32.7° changed from a singlet for x = 0.0 to a doublet for both x = 0.2 and 0.4, as shown in the inset of Fig. 4. As x increased, the diffraction line at 2θ = 32.7° corresponding to LaNiO<sup>3</sup> structure decreased in intensity and, consequently, a new diffraction line at 2θ = 32.4° corre‐ sponding to La2Ni2O5 brownmillerite-type structure appeared. Additionally, NiO and CaO were also present as segregated phases for the x = 0.2 and 0.4 solids. Thus, it is difficult to substitute La3+ with Ca2+ in LaNiO3 even to a limited extent. TPR profiles showed increase in the extent of reduction in the Fe series, explained by the higher amount of Fe4+ ions generated to compensate for the unbalanced charge. For the Ni series, a large amount of H2 consumption indicated a total reduction of Ni3+ ions in LaNiO3 to yield the La2Ni2O5 phase, followed by an H2 consumption peak at 500°C, ascribed to Ni° deposited on La oxide. The evolution of oxygen during TPD measurements showed the expected presence of physisorbed species at low temperatures in the Fe series, in addition to α-oxygen, which appeared as a shoulder for the pure LaFeO3. There was a noticeable increase in the intensity of the peak attributed to the socalled β-oxygen upon Ca2+ substitution. Since this peak is associated with the lattice oxygen or with oxygen species occupying inner vacancies, their evolution is a measurement of the replacement of La3+ by Ca2+ in the perovskite structure. For the Ni series, the desorption profiles for the pure LaNiO3 and the substituted samples showed a large increase in the desorption peak detected at 350°C, indicative of either the presence of segregated metal oxide phases in a high degree of dispersion or crystalline oxides (such as La2Ni2O5), as detected by XRD. With regard to the XPS spectra of the Fe series, the BE of the most intense peaks of doublets for La3d5/2 at 834.8 eV and Fe2p3/2 at 710.3 eV were almost constant. For the Ni counterpart, on the other hand, the overlap of the La3d3/2 and Ni2p3/2 peaks may mask the accurate measurement of the intensity of the Ni BE; thus, the Ni2p1/2 was measured instead. The least intense Ni2p1/2 peak at 873.4 eV was almost constant, and the detected satellite line at 881.2 eV is indicative of the presence of Ni2+ ions on the surface. The Ca 2p3/2 doublet has two peaks at BE of 346.2 and 346.8 eV, corresponding to surface Ca-O and carbonated surface Ca2+ ions, respectively. O1s peak was deconvoluted into three components: at BE of 529.2 eV assigned to surface O2- species, at 531.0 eV which arises from lattice [La-O-M] bonds and at 532.3 eV due to hydroxyl/carbonate groups. In the Fe series, x = 0.2 showed the largest extent of Ca incor‐ porated into the perovskite structure at 346.2 eV, while a clear trend was not observed for the Ni series. The (Fe/La+Ca) and (Ni/La+Ca) surface ratios were higher than the nominal com‐ positions: i.e., the solid surface became enriched with Ni and Fe. Since the carbonation degree is related to the extent of segregated phases, the (CO3 2-/La+Ca) surface ratio for the Fe and Ni series can provide an insight into the surface structure of the solid. A decrease in the carbo‐ nation degree upon Ca substitution was obtained for the Ca series; conversely, a larger increase with increasing x was detected for the Ni series. This finding is in agreement with the previous characterization results, and demonstrates that upon Ca substitution, Ni3+ ions are more difficult to stabilize than Fe3+ ions in the perovskite lattice. The perovskites were tested in toluene combustion (4000 ppmv) in a flow reactor under an excess of oxygen using a GHSV of 47000 h-1. For the Fe series, the substituted perovskites displayed higher catalytic activity than the pure LaFeO3 perovskite and the maximum catalytic activity was obtained for the perovskite with x = 0.2. The observed trend in activity indicated no simple dependence on the metal ion oxidation state, and appeared to be due to the effect of the pair Fe4+/Fe3+ on the vacancy ordering, in that a disordered distribution of oxygen vacancies led to an increase in the catalytic activity. The lower activation energy displayed by La0.8Ca0.2FeO3 can be attributed to cationic vacancies compensating the amount of Fe4+ involved in the reaction. Conversely, for the Ni series, substitution resulted in solids with lower activity than the pure LaNiO3 perovskite due to the loss of the perovskite structure and the corresponding change in the oxidation state of Ni3+ to Ni2+.

**Figure 4.** X-ray diffraction patterns of (a) La1-xCaxFeO3: (■) orthorhombic; (○) cubic LaFeO3; (b) La1-xCaxNiO3; (○)La‐ NiO3; (■) NiO; (●) La2Ni2O5

## **4.3. B-site substitution**

to compensate for the unbalanced charge. For the Ni series, a large amount of H2 consumption indicated a total reduction of Ni3+ ions in LaNiO3 to yield the La2Ni2O5 phase, followed by an H2 consumption peak at 500°C, ascribed to Ni° deposited on La oxide. The evolution of oxygen during TPD measurements showed the expected presence of physisorbed species at low temperatures in the Fe series, in addition to α-oxygen, which appeared as a shoulder for the pure LaFeO3. There was a noticeable increase in the intensity of the peak attributed to the socalled β-oxygen upon Ca2+ substitution. Since this peak is associated with the lattice oxygen or with oxygen species occupying inner vacancies, their evolution is a measurement of the replacement of La3+ by Ca2+ in the perovskite structure. For the Ni series, the desorption profiles for the pure LaNiO3 and the substituted samples showed a large increase in the desorption peak detected at 350°C, indicative of either the presence of segregated metal oxide phases in a high degree of dispersion or crystalline oxides (such as La2Ni2O5), as detected by XRD. With regard to the XPS spectra of the Fe series, the BE of the most intense peaks of doublets for La3d5/2 at 834.8 eV and Fe2p3/2 at 710.3 eV were almost constant. For the Ni counterpart, on the other hand, the overlap of the La3d3/2 and Ni2p3/2 peaks may mask the accurate measurement of the intensity of the Ni BE; thus, the Ni2p1/2 was measured instead. The least intense Ni2p1/2 peak at 873.4 eV was almost constant, and the detected satellite line at 881.2 eV is indicative of the presence of Ni2+ ions on the surface. The Ca 2p3/2 doublet has two peaks at BE of 346.2 and 346.8 eV, corresponding to surface Ca-O and carbonated surface Ca2+ ions, respectively. O1s peak was deconvoluted into three components: at BE of 529.2 eV assigned to surface O2- species, at 531.0 eV which arises from lattice [La-O-M] bonds and at 532.3 eV due to hydroxyl/carbonate groups. In the Fe series, x = 0.2 showed the largest extent of Ca incor‐ porated into the perovskite structure at 346.2 eV, while a clear trend was not observed for the Ni series. The (Fe/La+Ca) and (Ni/La+Ca) surface ratios were higher than the nominal com‐ positions: i.e., the solid surface became enriched with Ni and Fe. Since the carbonation degree

series can provide an insight into the surface structure of the solid. A decrease in the carbo‐ nation degree upon Ca substitution was obtained for the Ca series; conversely, a larger increase with increasing x was detected for the Ni series. This finding is in agreement with the previous characterization results, and demonstrates that upon Ca substitution, Ni3+ ions are more difficult to stabilize than Fe3+ ions in the perovskite lattice. The perovskites were tested in toluene combustion (4000 ppmv) in a flow reactor under an excess of oxygen using a GHSV of 47000 h-1. For the Fe series, the substituted perovskites displayed higher catalytic activity than the pure LaFeO3 perovskite and the maximum catalytic activity was obtained for the perovskite with x = 0.2. The observed trend in activity indicated no simple dependence on the metal ion oxidation state, and appeared to be due to the effect of the pair Fe4+/Fe3+ on the vacancy ordering, in that a disordered distribution of oxygen vacancies led to an increase in the catalytic activity. The lower activation energy displayed by La0.8Ca0.2FeO3 can be attributed to cationic vacancies compensating the amount of Fe4+ involved in the reaction. Conversely, for the Ni series, substitution resulted in solids with lower activity than the pure LaNiO3 perovskite due to the loss of the perovskite structure and the corresponding change in the

2-/La+Ca) surface ratio for the Fe and Ni

is related to the extent of segregated phases, the (CO3

520 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

oxidation state of Ni3+ to Ni2+.

The catalytic activity expressed as intrinsic activity (mol m-2 h-1) for the total combustion of acetyl acetate and ethanol over substituted LaFe1-yNiyO3 perovskites revealed that the simul‐ taneous presence of Ni<sup>0</sup> and Ni2+ is required to achieve optimum performance for these perovskites-type oxides [21]. The largest BET surface area of 20 m<sup>2</sup> g-1 was obtained for the unadulterated LaFeO3 perovskite, and then decreased progressively upon Ni substitution. XRD patterns indicated an orthorhombic phase for LaFeO3 and a rhombohedral phase for LaNiO3. Upon substitution of Fe3+ ions by Ni3+ ions, a shift toward larger 2θ angles was observed. The shift of the peak reached a maximum for a substitution degree y = 0.1, and then remained constant for higher substitutions. These findings suggest complete insertion of nickel for substitution y = 0.1, whereas a simple mixed NiO<sup>x</sup> phase became segregated for substitu‐ tions y> 0.1. FTIR spectra indicated that upon Ni3+ substitution the band at 560 cm-1 shifted toward higher wavenumbers and increased in intensity compared to that for pure LaFeO3. The dependence of these two parameters on the substitution degree runs in parallel with the shift observed in the main diffraction lines of the XRD patterns, confirming that Ni3+ ions were incorporated within the perovskite structure. The TPR1-TPR<sup>2</sup> cycles of the substituted perov‐ skites did not show significant differences between them, indicating that their structures were restored after intermediate oxidation. No reduction peaks were observed for the nonreducible LaFeO3, while two reduction peaks were observed for LaNiO3, attributed to reduction of Ni3+ ions of LaNiO3 to yield La2Ni2O5. The second reduction peak at 500°C led to Ni0 . Only one less-intense peak was observed for the substituted perovskite with almost no change in intensity (in comparison with TPR2), indicative of the high stability of LaFe1-yNiyO3 (y = 0.1, 0.2, 0.3). The evolution of oxygen during TPD experiments indicated the presence of α-oxygen in the pure LaNiO3 and the two highest substituted perovskites. The intensity of the desorption peaks increased as the degree of substitution increased. There was no evidence of the presence of α-oxygen on substituted perovskite with y = 0.1, and this was related to the absence of segregated metal oxide phases in this sample. The BEs of the XPS spectra for O1s was decon‐ voluted into three components: 529.2 eV attributed to surface O2- species; 531.0 eV, which arose from lattice [La-O-M] bonds; and 532.3 eV, which was due to hydroxyl/carbonate groups. The intensity of the two peaks at 529.2 eV and 531.0 eV changed almost linearly in a perfect opposite way, which was related to the incorporation of Ni into the structure. The increase in intensity of the lattice oxygen component in the sample with increase in Ni substitution was related to the increasing catalytic activity in combustion reactions. The surface atomic ratios Fe/La, Ni/La, and CO3 2-/La was plotted as a function of Ni substitution degree and shown in Fig. 5: upon Ni substitution, the surface of the substituted perovskites became Ni- and Fe-enriched. The Fe/La surface ratio for the nonsubstituted LaFeO3 is consistent with the slight La-enrich‐ ment often reported for La-containing perovskites, whereas the larger Ni/La surface ratio for LaNiO3 and also for the substituted LaFe1-yNiO3 samples indicated that Ni3+ ions are more difficult to be stabilized than Fe3+ ions in the perovskite lattice and results in surface segregation of a separate NiO phase. The extent of carbonation was higher for LaFeO3 due to basic character of the slight La-enrichment observed in this perovskite. The catalytic activity tested in the combustion of acetyl acetate and ethanol in a flow reactor under an excess of oxygen showed differences in catalytic behavior depending on the nature of the substrate to be combusted. Ethanol reacted at lower temperatures and exhibited a higher reaction rate. The catalytic activity of LaFeO3 increased upon nickel substitution but declined for LaNiO3. The approach undertaken to modulate oxygen adsorption and release properties by inserting Ni3+ ions into the LaFe1-yNiyO3 structure while retaining a separate NiO phase is crucial to enhance the performance of substituted perovskites in the combustion of ethanol and acetyl acetate.

## **4.4. B-site substitution**

For substituted LaMn1-yCoyO3 perovskite-type oxides, a crystal phase transformation occurred for y values above 0.5 and the general trend obtained was: for y ≤0.5, Co ions were inserted into the manganite structure; for y >0.5, the Mn ions were inserted into the cobaltite structure. The enhancement of the ferromagnetic properties and the thermal stability against reduction for y = 0.5 was attributed to the optimized Co2+-Mn4+ interactions [22, 23].

The manganite had the highest surface area (38 m2 g-1), and that upon Co substitution, a progressive decrease in surface area was observed until reaching the cobaltite surface value of 16 m2 g-1. XRD analysis showed the existence of cubic and rhombohedral perovskites phases

Energy Production, Decontamination, and Hydrogenation Reactions over Perovskite-Type Oxide Catalyst http://dx.doi.org/10.5772/61522 523

skites did not show significant differences between them, indicating that their structures were restored after intermediate oxidation. No reduction peaks were observed for the nonreducible LaFeO3, while two reduction peaks were observed for LaNiO3, attributed to reduction of

less-intense peak was observed for the substituted perovskite with almost no change in intensity (in comparison with TPR2), indicative of the high stability of LaFe1-yNiyO3 (y = 0.1, 0.2, 0.3). The evolution of oxygen during TPD experiments indicated the presence of α-oxygen in the pure LaNiO3 and the two highest substituted perovskites. The intensity of the desorption peaks increased as the degree of substitution increased. There was no evidence of the presence of α-oxygen on substituted perovskite with y = 0.1, and this was related to the absence of segregated metal oxide phases in this sample. The BEs of the XPS spectra for O1s was decon‐ voluted into three components: 529.2 eV attributed to surface O2- species; 531.0 eV, which arose from lattice [La-O-M] bonds; and 532.3 eV, which was due to hydroxyl/carbonate groups. The intensity of the two peaks at 529.2 eV and 531.0 eV changed almost linearly in a perfect opposite way, which was related to the incorporation of Ni into the structure. The increase in intensity of the lattice oxygen component in the sample with increase in Ni substitution was related to the increasing catalytic activity in combustion reactions. The surface atomic ratios Fe/La,

2-/La was plotted as a function of Ni substitution degree and shown in Fig. 5:

upon Ni substitution, the surface of the substituted perovskites became Ni- and Fe-enriched. The Fe/La surface ratio for the nonsubstituted LaFeO3 is consistent with the slight La-enrich‐ ment often reported for La-containing perovskites, whereas the larger Ni/La surface ratio for LaNiO3 and also for the substituted LaFe1-yNiO3 samples indicated that Ni3+ ions are more difficult to be stabilized than Fe3+ ions in the perovskite lattice and results in surface segregation of a separate NiO phase. The extent of carbonation was higher for LaFeO3 due to basic character of the slight La-enrichment observed in this perovskite. The catalytic activity tested in the combustion of acetyl acetate and ethanol in a flow reactor under an excess of oxygen showed differences in catalytic behavior depending on the nature of the substrate to be combusted. Ethanol reacted at lower temperatures and exhibited a higher reaction rate. The catalytic activity of LaFeO3 increased upon nickel substitution but declined for LaNiO3. The approach undertaken to modulate oxygen adsorption and release properties by inserting Ni3+ ions into the LaFe1-yNiyO3 structure while retaining a separate NiO phase is crucial to enhance the performance of substituted perovskites in the combustion of ethanol and acetyl acetate.

For substituted LaMn1-yCoyO3 perovskite-type oxides, a crystal phase transformation occurred for y values above 0.5 and the general trend obtained was: for y ≤0.5, Co ions were inserted into the manganite structure; for y >0.5, the Mn ions were inserted into the cobaltite structure. The enhancement of the ferromagnetic properties and the thermal stability against reduction

progressive decrease in surface area was observed until reaching the cobaltite surface value

g-1. XRD analysis showed the existence of cubic and rhombohedral perovskites phases

g-1), and that upon Co substitution, a

for y = 0.5 was attributed to the optimized Co2+-Mn4+ interactions [22, 23].

The manganite had the highest surface area (38 m2

. Only one

Ni3+ ions of LaNiO3 to yield La2Ni2O5. The second reduction peak at 500°C led to Ni0

522 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

Ni/La, and CO3

**4.4. B-site substitution**

of 16 m2

**Figure 5.** XPS ratios for the LaFe1-yNiyO3 perovskite: (■) atomic surface ratio, (●) nominal (a) Fe/La, (b) Ni/La, (c) CO3 2-/La

for pure LaMnO3 and LaCoO3, respectively. The general trend obtained for the LaMn1-yCoyO3 perovskites were as follows: (i) the XRD patterns were quite similar to the cobaltite's rhom‐ bohedral structure for the y>0.5 samples; (ii) the structure of the perovskite was modified to form a new orthorhombic structure for y<0.5; (iii) and for the case of y=0.5, the pattern was practically equal to the one obtained for pure LaMnO3. The vibration frequencies in the infrared region allowed us to conclude that the insertion of Mn into the cobaltite resulted in notable distortions of the highly symmetrical rhombohedral structure, and that manganite can withstand Co insertion without large structural changes up to values of y ≤ 0.3. With regard to TPR profiles, LaCoO3 and LaMnO3 displayed two reduction peaks, ascribed to the formation of oxygen-deficient intermediate and the reduction of the perovskite structure at higher temperatures. The substituted perovskites exhibited nearly identical behavior with tempera‐ ture displacements, which was attributed to the previously discussed distortions. The mag‐ netic properties were investigated in the paramagnetic and magnetically ordered regimes: in both cases the data revealed two well-identified regions separated by the equimolar compo‐ sition y = 0.5. For samples with y > 0.5 (Mn substitutes Co in the cobaltite LaCoO3) the Curie-Weiss temperature sharply decreased, reaching negative values at y = 1. This nonmonotonous variation was also observed in the ordering temperature Tc, which reached a maximum value for y = 0.5. The magnetic moment of the paramagnetic regime showed a linear decrease during the first half of the cobalt substitution, which was interpreted as the progressive transformation of Mn3+ into Mn4+ ions. This decrease became more pronounced above y = 0.5 since Mn ions were all converted to Mn4+, and cobalt was being introduced as Co3+. The overall behavior observed through the compositional variation of the Curie-Weiss temperature is explained by the varying character of the magnetic interactions between Co and Mn. During the first half of the series (y ≤ 0.5), strong Co2+-Mn4+ double-exchange ferromagnetic interactions were triggered when Co2+ was introduced into the lattice, while antiferromagnetism was augmented when Co was introduced as Co3+ (y>0.50). To investigate the ordered regime, the thermal dependence (ZFC/FC cycles) and the field dependence (M loops) of the magnetization was carried out. Figure 6 compares the (ZFC/FC cycles) regions separated by the equimolar composition y = 0.5. A strong ferromagnetic behavior was observed for the first half of the series since the magnetization MFC reached a characteristic plateau and the ordering temper‐ ature increased toward a maximum value of 235 K. For y compositions above 0.5, the mag‐ netization decreased by one or two orders of magnitude, meaning that the ferromagnetic Co2+- Mn4+ interactions were weaker, being progressively replaced by the less effective Co2+-Co3+ interactions. Thus, magnetic iso-field studies (ZFC-FC) revealed that, for y ≤ 0.50, the system presented an antiferromagnetic canted-like ordering of the Mn/Co sublattice, in which the presence of divalent Co ion created Mn3+-Mn4+ pairs that interact ferromagnetically through the oxygen orbital. This was confirmed by the magnetization loops in which the magnetic moment increased when substituting Mn for Co. The catalytic activity in total acetyl acetate combustion in a flow reactor under an excess of oxygen (60000 cm3 g-1h-1) indicated that lower ignition temperature (i.e., the higher activity) was displayed by manganite, while higher ignition temperature was exhibited by the cobaltite. The substituted perovskite displayed ignition temperatures between these two values, indicating that changes in y values modified the active sites required for the oxidation of organic compound. The larger intrinsic activity (molm-2h-1) of the substituted perovskites with y= 0.5 was attributed to the magnetic order of Mn4+-Co2+. This ordered state confirmed the presence of two regimes, separated by the equimolar composition LaMn0.5Co0.5O3: strong ferromagnetic interactions for yCo ≤ 0.5 and a superposition of antiferromagnetism and ferromagnetism for yCo > 0.5.

**Figure 6.** Magnetization cycles for LaMn1-yCoyO3 measured under 0.025 T: above: y = 0.0, 0.1, 0.3, 0.5; below: y = 0.5, 0.7, 0.9, 1.0

## **4.5. A- and B-site substitution**

bohedral structure for the y>0.5 samples; (ii) the structure of the perovskite was modified to form a new orthorhombic structure for y<0.5; (iii) and for the case of y=0.5, the pattern was practically equal to the one obtained for pure LaMnO3. The vibration frequencies in the infrared region allowed us to conclude that the insertion of Mn into the cobaltite resulted in notable distortions of the highly symmetrical rhombohedral structure, and that manganite can withstand Co insertion without large structural changes up to values of y ≤ 0.3. With regard to TPR profiles, LaCoO3 and LaMnO3 displayed two reduction peaks, ascribed to the formation of oxygen-deficient intermediate and the reduction of the perovskite structure at higher temperatures. The substituted perovskites exhibited nearly identical behavior with tempera‐ ture displacements, which was attributed to the previously discussed distortions. The mag‐ netic properties were investigated in the paramagnetic and magnetically ordered regimes: in both cases the data revealed two well-identified regions separated by the equimolar compo‐ sition y = 0.5. For samples with y > 0.5 (Mn substitutes Co in the cobaltite LaCoO3) the Curie-Weiss temperature sharply decreased, reaching negative values at y = 1. This nonmonotonous variation was also observed in the ordering temperature Tc, which reached a maximum value for y = 0.5. The magnetic moment of the paramagnetic regime showed a linear decrease during the first half of the cobalt substitution, which was interpreted as the progressive transformation of Mn3+ into Mn4+ ions. This decrease became more pronounced above y = 0.5 since Mn ions were all converted to Mn4+, and cobalt was being introduced as Co3+. The overall behavior observed through the compositional variation of the Curie-Weiss temperature is explained by the varying character of the magnetic interactions between Co and Mn. During the first half of the series (y ≤ 0.5), strong Co2+-Mn4+ double-exchange ferromagnetic interactions were triggered when Co2+ was introduced into the lattice, while antiferromagnetism was augmented when Co was introduced as Co3+ (y>0.50). To investigate the ordered regime, the thermal dependence (ZFC/FC cycles) and the field dependence (M loops) of the magnetization was carried out. Figure 6 compares the (ZFC/FC cycles) regions separated by the equimolar composition y = 0.5. A strong ferromagnetic behavior was observed for the first half of the series since the magnetization MFC reached a characteristic plateau and the ordering temper‐ ature increased toward a maximum value of 235 K. For y compositions above 0.5, the mag‐ netization decreased by one or two orders of magnitude, meaning that the ferromagnetic Co2+- Mn4+ interactions were weaker, being progressively replaced by the less effective Co2+-Co3+ interactions. Thus, magnetic iso-field studies (ZFC-FC) revealed that, for y ≤ 0.50, the system presented an antiferromagnetic canted-like ordering of the Mn/Co sublattice, in which the presence of divalent Co ion created Mn3+-Mn4+ pairs that interact ferromagnetically through the oxygen orbital. This was confirmed by the magnetization loops in which the magnetic moment increased when substituting Mn for Co. The catalytic activity in total acetyl acetate

524 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

combustion in a flow reactor under an excess of oxygen (60000 cm3

ignition temperature (i.e., the higher activity) was displayed by manganite, while higher ignition temperature was exhibited by the cobaltite. The substituted perovskite displayed ignition temperatures between these two values, indicating that changes in y values modified the active sites required for the oxidation of organic compound. The larger intrinsic activity (molm-2h-1) of the substituted perovskites with y= 0.5 was attributed to the magnetic order of Mn4+-Co2+. This ordered state confirmed the presence of two regimes, separated by the

g-1h-1) indicated that lower

Partial substitution of La3+ by Ag+ in LaMn0.9Co0.1O3 substantially increased the catalytic activity in the removal of n-hexane in a flow reactor under an excess of oxygen, attributed to the interaction between O- and the low coordinated metal site. Silver segregated phases further increased the catalytic activity [24].

The BET surface areas of the perovskites varied only slightly with substitution degree. The XRD profiles revealed that the substitution of small amounts of La3+ by Ag+ cations in a Cosubstituted manganite (LaMn0.9Co0.1O3) did not alter its rhombohedral crystallographic structure, and that it only decreased their crystallinity. Segregated silver oxide (Ag2O) phase was also detected upon increasing the Ag content. TPR profiles indicated that upon introduc‐ tion of Ag, two features were seen: the first reduction peak shifted toward lower temperatures and became wider upon increasing Ag content; the second reduction peak also started at a lower temperature but became more intense upon increasing Ag content. Since the first reduction step involves the formation of oxygen-deficient perovskite, and the shoulder includes the reduction of individual segregated oxides, quantification of this peak can be taken as a measurement of the extent of Ag insertion in the perovskite structure. The increase in H2 consumption in the first reduction step upon Ag-substitution is indicative of the appearance of a higher proportion of micro crystals from segregated silver, MnO2 phases, and/or surface carbonates species. The shift of the peak toward lower temperatures indicates a growth in crystal size. These results are in line with the XRD patterns, which showed diminished crystallinity as silver increased. The FTIR spectra showed an increase in the asymmetric and symmetric O-C-O stretching vibration modes of carboxylate structures with the substitution degree associated with the presence of segregated phases in the form of carbonates in the Agsubstituted perovskites. The XPS spectra of La 3d5/2 with BE at 834.8 eV and Co 2p3/2 with BE at 780.6 eV demonstrated the presence of La3+ and Co3+ species on the surface, whereas manganese appeared at slightly higher BE than is expected for Mn3+ ions, suggestive that the surface manganese was present as Mn4+. Differences in the surface O-species upon silver substitution was observed. For the x = 0.0 sample, the two surface oxygen species were the ones expected for a nonsubstituted perovskite. On the other hand, upon Ag substitution, the lower oxidation state of Ag+ inserted into the network of the LaMn0.9Co0.1O3 perovskite induced the appearance of O2- species at BE higher than 532 eV known as oxygen vacancies, which is necessary to maintain the electroneutrality of the solid. A relative proportion of this kind of oxygen species of 18% was exhibited by the perovskite with an Ag substitution of x = 0.1, which suggests that the maximum content of Ag with almost no segregated phases was at x = 0.1, and that this segregated phase disappeared at larger quantity of Ag. The EDAX and surface Mn/La and Co/La atomic ratios were much higher than the nominal compositions, and the differences were even larger upon increasing Ag substitution. These results indicate that the surface of the silver-substituted perovskites became Mn- and Ag-enriched, in line with deductions from oxygen-TPD profiles, which showed that Ag-substituted perovskites displayed a new desorption peak associated with the redox cycle of the B cation. This result further indicated that in these perovskites the Mn4+/Mn3+ redox cycle was almost constant, in agreement with XPS results. Therefore, the charge compensation due to the substitution of La3+ by Ag+ occurred mainly via a nonstoichiometric decrease rather than through a change in the Mn4+/Mn3+ratio. The catalytic activity of the perovskites was evaluated in the total com‐ bustion of n-hexane in a flow reactor under an excess of oxygen and shown in Fig. 7. The typical S-shaped curves represent the total conversion of n-hexane oxidation as a function of the reaction temperature up to complete combustion. There was an appreciable increase in the catalytic activity, expressed in terms of the total conversion, when Ag was substituted into the perovskite. For Ag substitution of x = 0.1, a noticeable shift of the curve toward lower tem‐ peratures is indicative of a higher catalytic activity of the perovskite upon a 10% Ag substitu‐ tion of La. The changes in the activity of the La1-xAgxMn0.9Co0.1O3 systems are mainly dependent on their crystalline structure and surface composition. The highest activity displayed by the x = 0.3 sample can be explained in terms of the large amount of segregated silver on the surface. Sample x = 0.2 was only slightly less active, although the silver surface segregation was much lower than in its x = 0.3 counterpart. This could be due to the less intense diffraction lines of the rhombohedral phase displayed by the x = 0.3 perovskite. For the sample x = 0.1, in which the rhombohedral structure and almost no silver segregation existed, the relation between catalytic activity and the nonstoichiometric oxygen ratio suggests an intrafacial reaction mechanism.

**Figure 7.** Stationary-state conversion over La1-xAgxMnx0.90Co0.1O3 perovskite for (▼) x = 0.0, (●) x = 0.1, (▲) x = 0.2, and (■) x = 0.3.

## **5. Soot combustion**

The BET surface areas of the perovskites varied only slightly with substitution degree. The XRD profiles revealed that the substitution of small amounts of La3+ by Ag+ cations in a Cosubstituted manganite (LaMn0.9Co0.1O3) did not alter its rhombohedral crystallographic structure, and that it only decreased their crystallinity. Segregated silver oxide (Ag2O) phase was also detected upon increasing the Ag content. TPR profiles indicated that upon introduc‐ tion of Ag, two features were seen: the first reduction peak shifted toward lower temperatures and became wider upon increasing Ag content; the second reduction peak also started at a lower temperature but became more intense upon increasing Ag content. Since the first reduction step involves the formation of oxygen-deficient perovskite, and the shoulder includes the reduction of individual segregated oxides, quantification of this peak can be taken as a measurement of the extent of Ag insertion in the perovskite structure. The increase in H2 consumption in the first reduction step upon Ag-substitution is indicative of the appearance of a higher proportion of micro crystals from segregated silver, MnO2 phases, and/or surface carbonates species. The shift of the peak toward lower temperatures indicates a growth in crystal size. These results are in line with the XRD patterns, which showed diminished crystallinity as silver increased. The FTIR spectra showed an increase in the asymmetric and symmetric O-C-O stretching vibration modes of carboxylate structures with the substitution degree associated with the presence of segregated phases in the form of carbonates in the Agsubstituted perovskites. The XPS spectra of La 3d5/2 with BE at 834.8 eV and Co 2p3/2 with BE at 780.6 eV demonstrated the presence of La3+ and Co3+ species on the surface, whereas manganese appeared at slightly higher BE than is expected for Mn3+ ions, suggestive that the surface manganese was present as Mn4+. Differences in the surface O-species upon silver substitution was observed. For the x = 0.0 sample, the two surface oxygen species were the ones expected for a nonsubstituted perovskite. On the other hand, upon Ag substitution, the

526 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

inserted into the network of the LaMn0.9Co0.1O3 perovskite induced

the appearance of O2- species at BE higher than 532 eV known as oxygen vacancies, which is necessary to maintain the electroneutrality of the solid. A relative proportion of this kind of oxygen species of 18% was exhibited by the perovskite with an Ag substitution of x = 0.1, which suggests that the maximum content of Ag with almost no segregated phases was at x = 0.1, and that this segregated phase disappeared at larger quantity of Ag. The EDAX and surface Mn/La and Co/La atomic ratios were much higher than the nominal compositions, and the differences were even larger upon increasing Ag substitution. These results indicate that the surface of the silver-substituted perovskites became Mn- and Ag-enriched, in line with deductions from oxygen-TPD profiles, which showed that Ag-substituted perovskites displayed a new desorption peak associated with the redox cycle of the B cation. This result further indicated that in these perovskites the Mn4+/Mn3+ redox cycle was almost constant, in agreement with XPS results. Therefore, the charge compensation due to the substitution of La3+ by Ag+ occurred mainly via a nonstoichiometric decrease rather than through a change in the Mn4+/Mn3+ratio. The catalytic activity of the perovskites was evaluated in the total com‐ bustion of n-hexane in a flow reactor under an excess of oxygen and shown in Fig. 7. The typical S-shaped curves represent the total conversion of n-hexane oxidation as a function of the reaction temperature up to complete combustion. There was an appreciable increase in the catalytic activity, expressed in terms of the total conversion, when Ag was substituted into the

lower oxidation state of Ag+

Diesel engines play a vital role in modern society because they are widely used for heavy-duty transportation and power generation. However, they are one of the largest contributors to environmental pollution because they emit particulate matter and nitrogen oxide. Particulate matter consists mostly of carbonaceous soot, minor amounts of volatile organic fraction from unburned fuel, as well as inorganic compounds such as ash and sulfur compounds. These materials are detrimental to human health, plants physiology, the environment, and building materials [25, 26]. Thus, stringent emission standards for soot have been set by most countries, leading to continuous efforts in emission control technology. Diesel particulate filter (DPF) is the most effective and widely employed approach to reduce soot in exhaust gases. However, one drawback for using the filter is that it requires regeneration through continuous combus‐ tion to remove accumulated soot. Recently, perovskite-type oxides have received considerable attention as catalytic systems for soot combustion [27-29]. Pure and doped lanthanum chromites, manganites, and cobaltites have been tested with mixed results in the catalytic combustion of soot [30]. Cerium has been usually reported as a good promoter in perovskite lattice. Partial substitution of La by Ce or Sr on cobalt-based perovskite led to an increase in catalytic oxidation [31]. When perovskite-type oxides were reduced, their low thermal behavior under reducing atmosphere allowed the formation of highly dispersed metal nanoparticles [22], which consequently modified the oxygen storage capacity of the catalyst. The substitution of the lanthanum cation La3+ by an oxygen storage cation such as cerium, with the capacity of the redox Ce3+/Ce4+ pair, will have a direct effect in the reducibility of the perovskite structure and the dispersion degree of the pure or mixed La2O3 and CeO2 oxides.

## **5.1. A-site substitution**

The catalytic activity of soot combustion over oxidized and partially reduced substituted La1 xCexCoO3 perovskites is enhanced by two factors: (1) the increase in CeO2 content, closely related to the increase in the material's capacity for oxygen transfer to the carbonaceous surface; and (2) the formation of larger CeO2 crystals in the higher-Ce-content samples after H2 reduction, responsible for the improvement of the exchange of oxygen and number of contact points between catalyst and soot [32].

The BET specific surface areas of the calcined solids ranged from 10 to 19 m2 g-1, and increased upon incorporation of Ce. After reduction, SBET decreased and ranged from 7 to 16 m2 g-1. XRD patterns of the calcined perovskites indicated that Ce was introduced into the lattice of the parent LaCoO3 solid at lower degree of substitution (i.e., x≤0.5), while for larger Ce content (x≥0.7) formation of CeO2 phase was detected. Figure 8 showed the XRD of the (a) calcined and the (b) reduced solids. Larger structural changes after H2 treatment at 700 ºC were observed for the samples with lower Ce content (x≤0.3), while calcined and reduced samples with higher Ce content (x = 0.7 and 0.9) showed nearly identical XRD patterns. For the samples with lower Ce content (x = 0.1, 0.3, 0.5), the most intense diffraction peak of the perovskite-type structure (2θ = 33°) fully disappeared after reduction treatment, and the reduced solids showed the presence of La2O3, La2CoO4, CeO2 as segregated phases as well as Co°, which appeared as two broad peaks at 2θ = 44.3° and 51.5°. TPR analysis showed that the reduction of pure LaCoO<sup>3</sup> solid occurred in two steps: at 365°C forming an intermediate oxygen-deficient perovskite structure (LaCoO3-δ); and, at 530°C which corresponds to the complete reduction of the perovskite, leading to the formation of La2O3, La(OH)3, H2O, and Coº. With the addition of Ce, there was a noticeable shift of the peaks to lower temperatures, which implies an increase in the reducibility of the catalysts. The evolution of oxygen during TPD analysis showed a beneficial effect on the presence of α-oxygen with increasing Ce content. The importance of these surface oxygen species is their close relation with the oxygen vacancies in the perovskite structure that can activate molecular oxygen in soot. The catalytic activity determined by nonisothermal thermogravimetric analysis indicated that the activity of calcined catalysts increased with Ce content. La0.1Ce0.9CoO3, the most active catalyst, was the sample with the highest Ce/La mass ratio and possessed CeO2 and Co3O4 crystalline phases instead of perov‐ skite-type structure (from their XRD patterns). These results suggest that CeO2 played the main role in soot combustion. The higher activity of the higher Ce-substituted catalysts indicated the availability of reactive oxygen species on the surface of the catalyst, attributed to the weaker binding of oxygen. The enhanced activity of the two catalysts with larger Ce content (x = 0.7 and 0.9) pointed out the importance of structural defects and oxygen mobility by lattice oxygen via La3+-doped CeO2 in the CeO2/soot interface for the catalytic oxidation of soot. In order to evaluate the effect of perovskite-type structure on soot combustion, the destruction of the perovskite structure was carried out in hydrogen at 700°C. After reduction, an increase in the catalytic activity was observed in varying degrees: only a slight increase of the activity for the samples with low CeO2 content (x≤0.5), while the activity for the catalysts with higher amount of CeO2 (x = 0.7, 0.9) increased sharply. These results suggest that the activity of these materials for soot combustion is improved by increasing the CeO2 content, closely related to the higher material capacity for oxygen transfer from catalyst to carbonaceous surface and the formation of larger CeO2 crystals in the higher-Ce-content samples after reduction.

**Figure 8.** XRD profiles for La1-xCexCoO3 catalysts: (left) calcined; (right) reduced

## **5.2. A- and B-site substitution**

one drawback for using the filter is that it requires regeneration through continuous combus‐ tion to remove accumulated soot. Recently, perovskite-type oxides have received considerable attention as catalytic systems for soot combustion [27-29]. Pure and doped lanthanum chromites, manganites, and cobaltites have been tested with mixed results in the catalytic combustion of soot [30]. Cerium has been usually reported as a good promoter in perovskite lattice. Partial substitution of La by Ce or Sr on cobalt-based perovskite led to an increase in catalytic oxidation [31]. When perovskite-type oxides were reduced, their low thermal behavior under reducing atmosphere allowed the formation of highly dispersed metal nanoparticles [22], which consequently modified the oxygen storage capacity of the catalyst. The substitution of the lanthanum cation La3+ by an oxygen storage cation such as cerium, with the capacity of the redox Ce3+/Ce4+ pair, will have a direct effect in the reducibility of the perovskite structure and the dispersion degree of the pure or mixed La2O3 and CeO2 oxides.

The catalytic activity of soot combustion over oxidized and partially reduced substituted La1 xCexCoO3 perovskites is enhanced by two factors: (1) the increase in CeO2 content, closely related to the increase in the material's capacity for oxygen transfer to the carbonaceous surface; and (2) the formation of larger CeO2 crystals in the higher-Ce-content samples after H2 reduction, responsible for the improvement of the exchange of oxygen and number of

g-1, and increased

g-1. XRD

The BET specific surface areas of the calcined solids ranged from 10 to 19 m2

upon incorporation of Ce. After reduction, SBET decreased and ranged from 7 to 16 m2

patterns of the calcined perovskites indicated that Ce was introduced into the lattice of the parent LaCoO3 solid at lower degree of substitution (i.e., x≤0.5), while for larger Ce content (x≥0.7) formation of CeO2 phase was detected. Figure 8 showed the XRD of the (a) calcined and the (b) reduced solids. Larger structural changes after H2 treatment at 700 ºC were observed for the samples with lower Ce content (x≤0.3), while calcined and reduced samples with higher Ce content (x = 0.7 and 0.9) showed nearly identical XRD patterns. For the samples with lower Ce content (x = 0.1, 0.3, 0.5), the most intense diffraction peak of the perovskite-type structure (2θ = 33°) fully disappeared after reduction treatment, and the reduced solids showed the presence of La2O3, La2CoO4, CeO2 as segregated phases as well as Co°, which appeared as two broad peaks at 2θ = 44.3° and 51.5°. TPR analysis showed that the reduction of pure LaCoO<sup>3</sup> solid occurred in two steps: at 365°C forming an intermediate oxygen-deficient perovskite structure (LaCoO3-δ); and, at 530°C which corresponds to the complete reduction of the perovskite, leading to the formation of La2O3, La(OH)3, H2O, and Coº. With the addition of Ce, there was a noticeable shift of the peaks to lower temperatures, which implies an increase in the reducibility of the catalysts. The evolution of oxygen during TPD analysis showed a beneficial effect on the presence of α-oxygen with increasing Ce content. The importance of these surface oxygen species is their close relation with the oxygen vacancies in the perovskite structure that can activate molecular oxygen in soot. The catalytic activity determined by nonisothermal thermogravimetric analysis indicated that the activity of calcined catalysts increased with Ce content. La0.1Ce0.9CoO3, the most active catalyst, was the sample with the

**5.1. A-site substitution**

contact points between catalyst and soot [32].

528 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

The increase in the catalytic performance in soot combustion by Ag+ substitution in LaMn0.9Co0.1O3 perovskite structure occurred due to increase in surface Ag species and superoxide ions content. These species have been found to be responsible for the higher capacity for oxygen transfer from the catalyst to the carbonaceous material in the solids with higher-Ag-content [33].

No significant differences between BET specific areas upon Ag substitution were found. XRD profiles revealed that the crystallinity of the rhombohedral structure of the Co-substituted manganite (LaMn0.9Co0.1O3) decreased upon Ag substitution. For the higher Ag-substituted perovskites, the intensities of three diffraction lines- indexed to a segregated Ag2O phaseincreased with increasing Ag substitution: a mean crystal size of 7 nm was estimated by the Debye Scherrer equation at 2θ = 37.96°. The main differences observed in relation to the TPR profiles of Ag-substituted perovskites with respect to Ag-free perovskite were as follows: the first reduction peak became more intense with increasing x, and the two reduction peaks become closer and shift toward lower temperatures. The increase in H2 consumption upon Agsubstitution is indicative of the appearance of a higher proportion of Ag2O segregated crystallites and/or other segregated species. The shift of the peaks toward lower temperatures coincided with the increase in crystallite size; furthermore, it indicates that the redox properties of the solids increased with Ag content. These results were consistent with XRD patterns, which showed a drop in crystallinity of the structure of the perovskites upon increasing silver content, indicating that the Ag+ inserted into the network made the formation of the perovskite structure more difficult. The BEs from XPS spectra revealed the presence of carbonates strongly attached to the surface, as expected for La-containing perovskites. The core-level spectra indicated the presence of La3+ and Co3+ species, whereas Mn appeared at slightly higher BE than that expected for Mn3+ ions, suggesting that surface Mn was present as Mn4+. Other notable observations were: (i) the Ag3d5/2 peaks with BE at 368.2 and 369.3 eV indicated the presence of surface Ag0 and Ag+ species, respectively; (ii) the O1s peak was attributed to surface lattice oxygen species (Olatt2-) at 529.8 eV, hydroxyl and/or carbonate groups at 531.0–531.4 eV, and loosely adsorbed superperoxide species (O2ads- ) at 532.4–532.6 eV. The surface Ag/La+Ag atomic ratio of the silver-substituted perovskites indicated the enrichment of Ag upon increasing substitution; in addition, the nearly unchanged surface O2ads- /Olatt2- atomic ratio of the Ag-free and the Ag-containing perovskites means that the charge compensation due to the substitution of La3+ by Ag+ occurred via nonstoichiometric decrease rather than through a change in the Mn4+/Mn3+ ratio. Since identical trends were obtained for the Ag+ /Ag and the O2ads- /Olatt2- surface ratios for the Ag-containing perovskites, it was proposed that these superoxides ions acted as the most important active species, facilitating oxygen transportation for soot oxidation. In addition, due to the fact that surface Ag and superoxides' ions formation increased with increasing Ag content, surface Ag2O was proposed as an active phase in the reaction. The evolution O2-TPD experiments revealed a continuous increase in total α-oxygen species upon Ag doping, supporting the hypothesis of the formation of Ag+ sites in the lattice of the perovskite. The catalytic activity for soot combustion was related to the temperature corresponding to the maximum from the DTG curve shown in Fig. 9. Higher values for Tm mean lower catalytic activity. Tm value for the uncatalyzed CB combustion was found to be 650°C, while much lower Tm values were observed for all the catalyzed reactions, indicative of the catalytic nature of the materials for soot combustion. Catalytic activity was observed to increase with the silver content, with the best catalyst being the one with Ag content of x = 0.3 (burned off soot at 370°C). This important catalytic effect was attributed to the presence of oxygen vacancies due to the substitution of La3+ by Ag+ in the perovskite lattice. These vacancies enhanced the availability and reactivity of oxygen species on the catalytic surface, because of the weaker oxygen binding. The results show that structural defects and oxygen mobility are important factors that control the catalytic activity of these types of crystalline structures. The presence of surface Ag2O phase improved the O2 exchange and number of contact points between catalyst and soot, with no significant effect on BET surface area and crystal size. These results contrast the proposed mechanism for soot combustion on La-containing perovskites, in which the oxidation of soot particles is attributed to the surface oxygen species, formed on the perovskite structure.

**Figure 9.** Thermogravimetric assays for the catalytic combustion of CB

## **6. Hydrogenation reactions**

profiles of Ag-substituted perovskites with respect to Ag-free perovskite were as follows: the first reduction peak became more intense with increasing x, and the two reduction peaks become closer and shift toward lower temperatures. The increase in H2 consumption upon Agsubstitution is indicative of the appearance of a higher proportion of Ag2O segregated crystallites and/or other segregated species. The shift of the peaks toward lower temperatures coincided with the increase in crystallite size; furthermore, it indicates that the redox properties of the solids increased with Ag content. These results were consistent with XRD patterns, which showed a drop in crystallinity of the structure of the perovskites upon increasing silver

structure more difficult. The BEs from XPS spectra revealed the presence of carbonates strongly attached to the surface, as expected for La-containing perovskites. The core-level spectra indicated the presence of La3+ and Co3+ species, whereas Mn appeared at slightly higher BE than that expected for Mn3+ ions, suggesting that surface Mn was present as Mn4+. Other notable observations were: (i) the Ag3d5/2 peaks with BE at 368.2 and 369.3 eV indicated the presence

oxygen species (Olatt2-) at 529.8 eV, hydroxyl and/or carbonate groups at 531.0–531.4 eV, and

atomic ratio of the silver-substituted perovskites indicated the enrichment of Ag upon

the Ag-free and the Ag-containing perovskites means that the charge compensation due to the

enhanced the availability and reactivity of oxygen species on the catalytic surface, because of the weaker oxygen binding. The results show that structural defects and oxygen mobility are important factors that control the catalytic activity of these types of crystalline structures. The presence of surface Ag2O phase improved the O2 exchange and number of contact points between catalyst and soot, with no significant effect on BET surface area and crystal size. These results contrast the proposed mechanism for soot combustion on La-containing perovskites,

/Olatt2- surface ratios for the Ag-containing perovskites, it was proposed that these superoxides ions acted as the most important active species, facilitating oxygen transportation for soot oxidation. In addition, due to the fact that surface Ag and superoxides' ions formation increased with increasing Ag content, surface Ag2O was proposed as an active phase in the reaction. The evolution O2-TPD experiments revealed a continuous increase in total α-oxygen species upon Ag doping, supporting the hypothesis of the formation of Ag+ sites in the lattice of the perovskite. The catalytic activity for soot combustion was related to the temperature corresponding to the maximum from the DTG curve shown in Fig. 9. Higher values for Tm mean lower catalytic activity. Tm value for the uncatalyzed CB combustion was found to be 650°C, while much lower Tm values were observed for all the catalyzed reactions, indicative of the catalytic nature of the materials for soot combustion. Catalytic activity was observed to increase with the silver content, with the best catalyst being the one with Ag content of x = 0.3 (burned off soot at 370°C). This important catalytic effect was attributed to the presence of

increasing substitution; in addition, the nearly unchanged surface O2ads-

change in the Mn4+/Mn3+ ratio. Since identical trends were obtained for the Ag+

and Ag+ species, respectively; (ii) the O1s peak was attributed to surface lattice

inserted into the network made the formation of the perovskite

occurred via nonstoichiometric decrease rather than through a

) at 532.4–532.6 eV. The surface Ag/La+Ag

in the perovskite lattice. These vacancies

/Olatt2- atomic ratio of

/Ag and the

content, indicating that the Ag+

substitution of La3+ by Ag+

loosely adsorbed superperoxide species (O2ads-

530 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

oxygen vacancies due to the substitution of La3+ by Ag+

of surface Ag0

O2ads-

For these kinds of heterogeneous catalytic reactions, perovskite-type oxides are not used directly as catalysts but rather as precursors to obtain metal nanoparticles catalysts. Therefore, prior to the reaction, perovskites are activated *ex situ* under H2 at 500°C.

## **7. Guaiacol conversion**

Catalytic studies focusing on hydrodeoxygenation (HDO) of guaiacol, the most representative phenolic compound in bio-oil, has provided insights into catalyst optimization for lignin valorization technologies. Different active phases, including sulfides [34-36] and metals [37-39], have been investigated as catalysts for the HDO of guaiacol. Nickel-based catalysts have recently received extensive attention for the HDO of lignin derived compounds, and have consistently been proven to be highly active.

## **8. A-site substitution**

The preparations of Ni catalysts from the reduction of La1-xCexNiO3 perovskite-type oxides were effective in catalyzing the conversion of guaiacol to cyclohexanol, an important feedstock for the chemical industry. The most active catalyst was the solid with Ce content x = 0.5 and 0.7, which was ascribed to the highest thermal stability under reductive atmosphere of the CeO2-solid solution partially substituted with La [40].

The BET specific surface areas of the calcined and reduced solids showed a monotonic dependence of the surface area on Ce content with a slight decrease after reduction treatment. The main difference in the XRD patterns of the as-calcined and as-reduced solids after Ce substitution was the absence of the perovskite structure and the appearance of segregated phases in the latter. The most crystalline perovskites corresponded to the undoped LaNiO3 material and the lowest Ce-substituted perovskite (La0.9Ce0.1NiO3). Upon increasing the Ce content (x≥0.5), there were neither diffraction peaks characteristic of the perovskite structure nor single or mixed La oxides; however, new crystalline phases corresponding to CeO2 and NiO appeared. The XRD patterns of the Ni-reduced solids indicated a complete loss of the LaNiO3 perovskite structure and the formation of Ni0 for the material with Ce content x = 0.0 and 0.1. For higher Ce contents (i.e., x≥0.5), there were no observable changes in the crystalline structure after H2 treatment, and a shift of the main CeO2 peak toward higher 2θ angles upon Ce doping can be taken as a clear indication of the modification of the crystalline structures, suggestive of the insertion of La3+ into the CeO2 structure. These results indicate the formation of the perovskite structure with low thermal stability under H2 for the samples with low Ce content x≤0.3, and the formation of a solid solution of CeO2-La2O3 with higher thermal stability for samples with higher degree of Ce substitution x≥ 0.5. The Ni-reduced catalyst with Ce content x = 0.5 displayed the lowest crystallinity and the highest Ni0 dispersion. The TPR1- TPR2 cycles demonstrated clear differences in the H2 consumption peaks between TPR cycles, an indication of the structural changes under H2 treatment. The TPR1 profiles showed two different behaviors attributed to the reduction of Ni2+ to Ni0 : two reduction peaks for the samples with Ce contents x≤ 0.3 and only one reduction peak for those with Ce contents x≥0.5. The amount of H2 consumed during TPR1 and TPR2 cycles is plotted as a function of Ce content and shown in Fig. 10. The TPR1 profile showed larger H2 consumption for the perovskites with Ce content x≤0.3 and almost constant value for those with Ce content x≥0.5. Considering that the theoretical hydrogen uptake for sample reduction assuming that NiO is stoichiometrically reduced to Ni° is 4.1 mmol g-1, the larger obtained values suggest that other oxides entities are also being reduced. Since XRD results did not show any structural changes under reduction atmosphere, it is possible to calculate the degree of reduction of the perovskite structure considering only the reduction of Ni3+ to Ni0 . The obtained reduction degree of 24% fits quite well with previously reported results corresponding to the total reduction of Ni present in the original perovskites. The TEM micrographs of all of the Ni-reduced catalysts showed a narrower particle size distribution for metallic Ni between 5 and 25 nm. The H2 chemisorption capacity assuming a stoichiometry of H/Nis = 1 was smaller than the expected one considering the particle size obtained from TEM and XRD, indicating that Ni particles were partially inserted and not available for H2 uptake. In the conversion of guaiacol over Ni-reduced catalysts, cyclohexanol and methoxycyclohexanol were principally obtained via the hydroge‐ nation and demethylation routes. In other words, the Ni-reduced catalysts displayed a high hydrogenation activity and a low activity toward hydrogenolysis of C-OH bond. The products' distribution calculated both at 20% conversion of guaiacol and at 4 h of reaction time showed that all the Ni-reduced catalysts displayed similar trends in product distribution. The changes in Ni particle size and the presence of segregated phase influenced by the degree of Ce substitution did not affect the selectivity. This result suggests that the simultaneous presence of highly dispersed nickel partially inserted into a CeO2-solid solution is a prerequisite for improved catalytic performance for the conversion of guaiacol over these catalytic systems.

**Figure 10.** Plot of hydrogen consumption of the La1-xCexNiO3 vs. nominal Ce content (●) TPR1; (■) TPR<sup>2</sup>

## **9. Glycerol reforming**

for the chemical industry. The most active catalyst was the solid with Ce content x = 0.5 and 0.7, which was ascribed to the highest thermal stability under reductive atmosphere of the

The BET specific surface areas of the calcined and reduced solids showed a monotonic dependence of the surface area on Ce content with a slight decrease after reduction treatment. The main difference in the XRD patterns of the as-calcined and as-reduced solids after Ce substitution was the absence of the perovskite structure and the appearance of segregated phases in the latter. The most crystalline perovskites corresponded to the undoped LaNiO3 material and the lowest Ce-substituted perovskite (La0.9Ce0.1NiO3). Upon increasing the Ce content (x≥0.5), there were neither diffraction peaks characteristic of the perovskite structure nor single or mixed La oxides; however, new crystalline phases corresponding to CeO2 and NiO appeared. The XRD patterns of the Ni-reduced solids indicated a complete loss of the LaNiO3 perovskite structure and the formation of Ni0 for the material with Ce content x = 0.0 and 0.1. For higher Ce contents (i.e., x≥0.5), there were no observable changes in the crystalline structure after H2 treatment, and a shift of the main CeO2 peak toward higher 2θ angles upon Ce doping can be taken as a clear indication of the modification of the crystalline structures, suggestive of the insertion of La3+ into the CeO2 structure. These results indicate the formation of the perovskite structure with low thermal stability under H2 for the samples with low Ce content x≤0.3, and the formation of a solid solution of CeO2-La2O3 with higher thermal stability for samples with higher degree of Ce substitution x≥ 0.5. The Ni-reduced catalyst with Ce

TPR2 cycles demonstrated clear differences in the H2 consumption peaks between TPR cycles, an indication of the structural changes under H2 treatment. The TPR1 profiles showed two

samples with Ce contents x≤ 0.3 and only one reduction peak for those with Ce contents x≥0.5. The amount of H2 consumed during TPR1 and TPR2 cycles is plotted as a function of Ce content and shown in Fig. 10. The TPR1 profile showed larger H2 consumption for the perovskites with Ce content x≤0.3 and almost constant value for those with Ce content x≥0.5. Considering that the theoretical hydrogen uptake for sample reduction assuming that NiO is stoichiometrically reduced to Ni° is 4.1 mmol g-1, the larger obtained values suggest that other oxides entities are also being reduced. Since XRD results did not show any structural changes under reduction atmosphere, it is possible to calculate the degree of reduction of the perovskite structure

well with previously reported results corresponding to the total reduction of Ni present in the original perovskites. The TEM micrographs of all of the Ni-reduced catalysts showed a narrower particle size distribution for metallic Ni between 5 and 25 nm. The H2 chemisorption capacity assuming a stoichiometry of H/Nis = 1 was smaller than the expected one considering the particle size obtained from TEM and XRD, indicating that Ni particles were partially inserted and not available for H2 uptake. In the conversion of guaiacol over Ni-reduced catalysts, cyclohexanol and methoxycyclohexanol were principally obtained via the hydroge‐ nation and demethylation routes. In other words, the Ni-reduced catalysts displayed a high hydrogenation activity and a low activity toward hydrogenolysis of C-OH bond. The products' distribution calculated both at 20% conversion of guaiacol and at 4 h of reaction time showed

dispersion. The TPR1-

: two reduction peaks for the

. The obtained reduction degree of 24% fits quite

CeO2-solid solution partially substituted with La [40].

532 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

content x = 0.5 displayed the lowest crystallinity and the highest Ni0

different behaviors attributed to the reduction of Ni2+ to Ni0

considering only the reduction of Ni3+ to Ni0

The use of environmentally friendly and renewable feedstock for clean energy production such as hydrogen has been extensively studied, with particular focus on the catalytic green reforming process of glycerol [41]. Even though the reaction pathway is a complex process, as a result of the occurrence of wide variety of reactions, the type of catalysts used play a key role in determining the product distribution. For supported metal catalysts as well as Ni-based catalysts, high basic character of the catalysts has consistently been observed to lead to higher activity and resistance toward deactivation [42, 43].

## **9.1. A-site substitution**

Mixed oxides with the general formula La1-xCexNiO3 used as precursors for supported Ni catalysts have led to active catalysts for the production of hydrogen-rich gas stream. Substi‐ tution of 50% of La by Ce provided more deactivation-resistant catalyst, allowing only minor amount of deactivation, associated with the formation of CeO2-La2O3 solid solutions, which acted as an oxygen buffer and, hence, facilitated the removal of carbon deposits on the catalysts surface [44].

**Figure 11.** Evolution of global glycerol conversion as a function of time on stream for glycerol steam reforming at 500 °C over all mixed oxide-derived catalysts

The La1-xCexNiO3 perovskites used as precursors for supported Ni catalysts were previously fully characterized and used as catalysts in glycerol steam reforming. Figure 11 illustrates the global performance of the catalysts for glycerol reforming. Regardless of the features of the catalysts or degree of La substitution by Ce, a drop in global conversion versus time on stream was observed. However, the sample formed from the mixed oxide with an equivalent content of La3+ and Ce3+ (x=0.5) was much more stable and promising, which may be associated with the formation of a CeO2-La2O3 solid solution as detected by XRD analysis. The catalysts with Ce substitution other than 50% led to a dramatic decrease in glycerol conversion, revealing a strong deactivation process: indeed x = 0.1 and x = 0.7 samples presented conversion within 5 and 20% right after the first hour. To investigate the causes of deactivation, Fe-SEM analysis revealed that the deactivation of unsubstituted LaNiO3 perovskite-derived catalysts (x = 0.0) was due to the formation of plentiful amounts of filamentous carbon. Even though, the formation of the carbon nanostructures did not occur equally over all Ce-containing catalysts due to varying degrees of Ce substitution, considerable amounts of carbon filaments and similar FE-SEM images were obtained even for unsubstituted LaNiO3 perovskite-derived catalyst (x=0.0). Conversely, after a much more detailed and meticulous analysis, only few filaments could be found on the x = 0.5 solid. Thus, it can be argued that although deactivation occurred irrespective of catalysts composition, the sample with 50% La substituted by Ce was more resistance to deactivation compared to other loadings. TEM micrographs allowed the identification of carbon nanotubular structures with an open end, where it was suggested that Ni nanoparticles were detached. From this analysis, it was suggested that Ni nanoparticles were intimately involved in the formation of those carbon filaments. TPO/TGA analyses combined with mass spectrometry were complementarily performed for the used catalysts. The evolution of CO2 (m/z=44) observed in the analysis originated from the gasification of carbon deposits. The x = 0.5 sample possessed the lowest amount of the more reactive carbon deposits. Thus, the minimization of the deactivation process in glycerol reforming requires an appropriate Ce content, which can lead to the formation of CeO2-La2O3 solid solutions.

## **10. Xylose hydrogenation**

Liquid phase hydrogenation of aldoses is an important process in the synthesis of mono- and disaccharides, which have long been used as natural sweetening agents. Xylitol, an excellent sweet molecule, can also be commercialized because its sweetening capacity exceeds that of sugar by 20-25%, in addition to having no insulin requirements [45]. Several well-known hydrogenation catalysts such as noble metals or nickel-based [46] catalysts have been used for xylose hydrogenation [47]. Due to their lower price, the main advantage of Ni catalysts is their utilization in a typical slurry batch reactor, leading to good activity and selectivity. Neverthe‐ less, the major drawback of Ni catalysts (which has been reported for Raney Ni) is their high deactivation rate as a result of poisoning of the active sites and metal leaching.

## **10.1. A-site substitution**

0 1 2 3 4 5 6

**Figure 11.** Evolution of global glycerol conversion as a function of time on stream for glycerol steam reforming at 500

The La1-xCexNiO3 perovskites used as precursors for supported Ni catalysts were previously fully characterized and used as catalysts in glycerol steam reforming. Figure 11 illustrates the global performance of the catalysts for glycerol reforming. Regardless of the features of the catalysts or degree of La substitution by Ce, a drop in global conversion versus time on stream was observed. However, the sample formed from the mixed oxide with an equivalent content of La3+ and Ce3+ (x=0.5) was much more stable and promising, which may be associated with the formation of a CeO2-La2O3 solid solution as detected by XRD analysis. The catalysts with Ce substitution other than 50% led to a dramatic decrease in glycerol conversion, revealing a strong deactivation process: indeed x = 0.1 and x = 0.7 samples presented conversion within 5 and 20% right after the first hour. To investigate the causes of deactivation, Fe-SEM analysis revealed that the deactivation of unsubstituted LaNiO3 perovskite-derived catalysts (x = 0.0) was due to the formation of plentiful amounts of filamentous carbon. Even though, the formation of the carbon nanostructures did not occur equally over all Ce-containing catalysts due to varying degrees of Ce substitution, considerable amounts of carbon filaments and similar FE-SEM images were obtained even for unsubstituted LaNiO3 perovskite-derived catalyst (x=0.0). Conversely, after a much more detailed and meticulous analysis, only few filaments could be found on the x = 0.5 solid. Thus, it can be argued that although deactivation occurred irrespective of catalysts composition, the sample with 50% La substituted by Ce was more resistance to deactivation compared to other loadings. TEM micrographs allowed the identification of carbon nanotubular structures with an open end, where it was suggested that Ni nanoparticles were detached. From this analysis, it was suggested that Ni nanoparticles were intimately involved in the formation of those carbon filaments. TPO/TGA analyses combined with mass spectrometry were complementarily performed for the used catalysts.

Time on stream, h

xCe=0.7 xCe=0.1

xCe=0.5

xCe=0.0

0

°C over all mixed oxide-derived catalysts

20

40

Global conversion, %

60

80

100

534 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

A series of active and selective partially Ni-substituted La1-xCexAl0.18Ni0.82O3 perovskite catalysts with a constant Ni loading of 20 wt.% were tested in the hydrogenation of xylose to xylitol to gain insight into the importance of the perovskite structure in preventing leaching.

The specific surface areas obtained showed that the solids with lower Ce substitution degree displayed lower values between 9 and 15 m2 g-1, while higher values above 30 m2 g-1 were obtained for solids with higher Ce substitution degrees. The XRD patterns of the reduced solids shown in Figure 12 indicated that the Ce-free solid presented perovskite structure while upon Ce substitution segregated phases appeared. The x = 0.1 solid showed both the perovskite phase and the presence of mixed oxides. For larger Ce substitution degree the perovskite structure was not present and only peaks corresponding to Ce-La oxide solid solution were found. TPR results show only two reduction peaks corresponding to reduction of Ni species. TEM micrographs revealed that Ni was dispersed on the oxide support without forming nanoparticles. Catalytic activity results indicated that all the catalysts were active in the hydrogenation of xylose to xylitol. The catalysts with lower Ce substitution degree (x=0.0, 0.1) displayed higher catalytic activity, demonstrating the importance of crystalline perovskite structure in this reaction. Xylitol selectivity between 40 and 60% was obtained at a conversion of 30%. Hydrogenolysis products were also detected. Atomic absorption spectroscopy (AAS) analysis of the catalyst before and after reaction indicated that for the unsubstituted and lower Ce substituted solids (x = 0.0, 0.1), a constant Ni content close to the nominal loading was obtained, indicative of the high stability under aqueous medium. On the other hand, the solid with the higher Ce substitution degree (x = 0.5, 0.7) showed loss of Ni by leaching. The results from this study demonstrated clear dependence on the reactivity and stability of the catalyst on the presence of crystalline perovskite, modulated by the degree of Ce substitution.

**Figure 12.** XRD pattern of Ni-substituted La1-xCexAl0.18Ni0.82O3 catalyst

## **11. Conclusion**

Perovskite-type oxides with La in the A-site and a d-transition metal in the B-site position are the most widely used materials for catalytic purposes. The activity of La perovskites can be noticeably improved by partial replacement of the A-site and/or B-site metal. The different combinations of partial substitution of A-site by an adequate A' metal with similar ionic radii but lower oxidation state modifies the oxidation state of the B-site metal cation and/or oxygen vacancies formation. This substitution also affects the thermal stability, crystalline structure, and activity of the perovskites. Two clear applications can be identified:


Due to their tunable properties, perovskites show different crystallization structures and high flexibility in their composition, and, hence, it is important to understand how different perovskites behave under different reaction conditions and what would be the most effective metal combination for a given catalytic process.

## **Acknowledgements**

The authors thank CONICYT Fondecyt No 1130005 and Red Doctoral REDOC, and the MINEDUC project UCO1202.

## **Author details**

Gina Pecchi1\*, Nestor Escalona2,3, I. Tyrone Ghampson2 and Ruddy Morales1

\*Address all correspondence to: gpecchi@udec.cl

1 Dpto Físico Química, Facultad Ciencias Químicas, Universidad de Concepción, Concep‐ ción, Chile

2 Dpto Ingeniería Química y Bioprocesos, Escuela de Ingeniería, Pontificia Universidad Ca‐ tólica de Chile, Macul, Santiago, Chile

3 Facultad de Química, Pontificia Universidad Católica de Chile, Santiago, Chile

## **References**

20 30 40 50 60 70 80 90

2

Perovskite-type oxides with La in the A-site and a d-transition metal in the B-site position are the most widely used materials for catalytic purposes. The activity of La perovskites can be noticeably improved by partial replacement of the A-site and/or B-site metal. The different combinations of partial substitution of A-site by an adequate A' metal with similar ionic radii but lower oxidation state modifies the oxidation state of the B-site metal cation and/or oxygen vacancies formation. This substitution also affects the thermal stability, crystalline structure,

**•** Perovskites to be used as precursors to prepare nanosized catalysts to be used in hydroge‐

Due to their tunable properties, perovskites show different crystallization structures and high flexibility in their composition, and, hence, it is important to understand how different perovskites behave under different reaction conditions and what would be the most effective

\*

\*

and activity of the perovskites. Two clear applications can be identified:

**•** Perovskites with oxygen vacancies to be used as catalysts in oxidation reactions

+ # + # #


+ NiO

\* \* \* \*

xCe=0.7 xCe=0.5

xCe=0.1

xCe=0.0

\* \* \* \* \* \*

O3

\*

+ + #

+

536 Perovskite Materials - Synthesis, Characterisation, Properties, and Applications

\*

#

\*

\* \*

**Figure 12.** XRD pattern of Ni-substituted La1-xCexAl0.18Ni0.82O3 catalyst

metal combination for a given catalytic process.

\*

Intensity, a.u.

**11. Conclusion**

nation reactions
