**Cu2ZnSnS4 Thin Film Solar Cells: Present Status and Future Prospects**

Minlin Jiang and Xingzhong Yan

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/50702

## **1. Introduction**

[45] Redinger, A., Berg, D. M., Dale, P. J., & Siebentritt, S. (2011). *Journal of the American*

[46] Hibberd, C. J., Chassaing, E., Liu, W., Mitzi, D. B., Lincot, D., & Tiwari, A. N. (2010).

[47] Friedlmeier, T. M., Wieser, N., Walter, T., Dittrich, H., & Schock, H. W. (1997). *Proc. 14th EuropeanConference of Photovoltaic Science and Engineering and Exhibition*, *Bedford,*

[48] Katagiri, H., Saitoh, K., Washio, T., Shinohara, H., Kurumadani, T., & Miyajima., S.

[49] Katagiri, H., Jimbo, K., Moriya, K., & Tsuchida, K. (2003). *Proc. World Conf. on Photo‐*

[50] Katagiri, H., Jimbo, K., Yamada, S., Kamimura, T., Maw, W. S., Fukano, T., Ito, T., &

[52] Mitzi, D., Gunawan, O., Todorov, T., Wang, K., & Guha, S. (2011). *Solar Energy Mate‐*

[53] Jimbo, K., Kimura, R., Kamimura, T., Yamada, S., Maw, W. S., Araki, H., Oishi, K., &

[54] Tanaka, T., Nagatomo, T., Kawasaki, D., Nishio, M., Guo, Q., Wakahara, A., Yoshida, A., & Ogawa, H. (2005). *Journal of Physics and Chemistry of Solids*, 66, 1978-1981.

[55] Wang, K., Gunawan, O., Todorov, T., Shin, B., Chey, S. J., Bojarczuk, N. A., Mitzi, D.,

[56] Araki, H., Mikaduki, A., Kubo, Y., Sato, T., Jimbo, K., Maw, W. S., Katagiri, H., Ya‐ mazaki, M., Oishi, K., & Takeuchi, A. (2008). *Thin Solid Films*, 517, 1457-1460.

[57] Kumar, Y., Kumar, B. K., Babu, G. S., Bhaskar, P. U., & Raja, V. S. (2009). *Physica Sta‐*

[58] Prabhakar, T., & Nagaraju, J. (2010). *Proc. of the 35th IEEE Photovoltaic Specialists Con‐*

[59] Pawar, B. S., Pawar, S. M., Shin, S. W., Choi, D. S., Park, C. J., Kolekar, S. S., & Kim, J.

[60] Fischereder, A., Rath, T., Haas, W., Amenitsch, H., Albering, J., Meischler, D., Laris‐ segger, S., Edler, M., Saf, R., Hofer, F., & Trimmel, G. (2010). *Chemistry of Materials*,

[51] Todorov, K., Reuter, K. B., & Mitzi, D. B. (2010). *Advanced Materials*, 22, E156-E159.

(2001). *Solar Energy Materials and Solar Cells*, 65, 141-148.

Motohiro, T. (2008). *Applied Physics Express*, 1, 041201-2.

Katagiri, H. (2007). *Thin Solid Films*, 515, 5997-5999.

& Guha, S. (2010). *Applied Physics Letters*, 97, 143508-3.

H. (2010). *Applied Surface Science*, 257, 1786-1791.

*Chemical Society*, 133, 3320-3323.

106 Solar Cells - Research and Application Perspectives

*UK, 1242.*

*Progress in Photovoltaics*, 18, 434-452.

*voltaic Energy Conversion*, *Osaka,2874*.

*rials & Solar Cells*, 95, 1421-1436.

*tus Solidi A*, 206, 1525-1530.

*ference*, 1964-1969.

22, 399-3406.

Pollution of the earth and shortage of energy sources have been the bottle-neck of survival and development for human beings since the start of the 21st century. Therefore, lowering energy consumption and protecting the environment have gradually gained attention from countries all over the world. In order to keep sustainable development, governments, re‐ search institutes, and industries have been working on the problems caused by the shortage of available energy sources. It is well known that the best way is to exploit renewable energy resources. Solar energy is considered to be the most economic and effective among all avail‐ able renewable energy resources. Solar energy is inexhaustible and it has already been theo‐ retically and experimentally proved that the earth would not be polluted at all if solar energy was utilized effectively.

To encourage and to promote the direct utilization of solar energy, developed countries have been legislating and deploying solar initiatives [1-3]. Joint Research Centre (Europe) predicted that energy directly harvested from sunlight would be 20% of total energy con‐ sumption in 2050, and this value could be over 50% in 2100 [4]. Solar energy will be widely utilized in industry, agriculture and daily life. Photovoltaic (PV) systems have recently at‐ tracted much attention due to their inherent advantages. Firstly, PV systems are capable of directly translating sunlight into electrical energy. The theoretical conversion efficiency of PV systems is relatively higher than other power generators. Secondly, PV systems do not necessarily contain movable parts. System wear induced by mechanical movement is avoid‐ ed. Therefore, PV systems can work continuously free from maintenance longer than other power generation technologies.

© 2013 Jiang and Yan; licensee InTech. This is an open access article 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. © 2013 Jiang and Yan; licensee InTech. This is a paper 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.

**Figure 1.** Market share of different PV technology [8].

**Figure 2.** Normalized abundance of elements for conventional PV absorbers.

It was reported by Solarbuzz that 16.3 GW PV modules had been shipped to customers in 2010 with the lion's share going to crystalline silicon (c-Si) technology (71%) [5]. Due to high cost and energy consumption input in manufacturing c-Si PV of modules, market share of the c-Si technology has been dropping while thin film PV technologies have been increasing rapidly [6, 7]. There are three main thin film PV technologies, CdTe, CuInxGa1-xS(Se)2 (CIGS), and thin film Si, which has gained 14%, 9%, and 6% of PV market share in 2010, re‐ spectively (Fig.1) [8]. Nevertheless, Si thin film solar cell (TFSC) has been relatively underde‐ veloped due to low efficiency and instability from the Staebler–Wronski effect. For the other two thin film technologies, there are restriction on the usage of heavy metals such as cadmi‐ um, the limitation in supplies for indium and tellurium, and the wide fluctuation in prices of indium and tellurium. These render the combined production capacity of the existing CdTe and CIGS technologies at a small scale lower than 100 GW per year. This is only a small fraction of energy consumption in 2050 which is expected to be 27 TW [4, 9].

Recently, quaternary compound Cu2ZnSnS4 (CZTS) has been intensively examined as an al‐ ternative PV material due to its similarity in material properties with CIGS and the relative abundance of raw materials (Fig. 2). CZTS is a compound semiconductor of (I)2(II)(IV)(VI)4. With a high absorption coefficient (> 104 cm-1) and a desirable bandgap (~1.45 eV), CZTS thin film has been considered an excellent PV material. Theoretical calculations have shown that conversion efficiency as high as 32% was possible for CZTS TFSCs with a CZTS layer of sev‐ eral micrometers. Wadia et al. also calculated the minimum cost of raw materials for the ex‐ isting PV technologies and the emerging PV technologies [10]. Part of the results is shown in Fig. 3. The cost of raw material for CZTS PV technology is much lower than that of the three existing thin film PV technologies.

**Figure 3.** Minimum cost of raw materials for different PV technology [10].

**Figure 1.** Market share of different PV technology [8].

108 Solar Cells - Research and Application Perspectives

**Figure 2.** Normalized abundance of elements for conventional PV absorbers.

It was reported by Solarbuzz that 16.3 GW PV modules had been shipped to customers in 2010 with the lion's share going to crystalline silicon (c-Si) technology (71%) [5]. Due to high cost and energy consumption input in manufacturing c-Si PV of modules, market share of the c-Si technology has been dropping while thin film PV technologies have been increasing rapidly [6, 7]. There are three main thin film PV technologies, CdTe, CuInxGa1-xS(Se)2 (CIGS), and thin film Si, which has gained 14%, 9%, and 6% of PV market share in 2010, re‐ spectively (Fig.1) [8]. Nevertheless, Si thin film solar cell (TFSC) has been relatively underde‐ veloped due to low efficiency and instability from the Staebler–Wronski effect. For the other two thin film technologies, there are restriction on the usage of heavy metals such as cadmi‐ um, the limitation in supplies for indium and tellurium, and the wide fluctuation in prices of indium and tellurium. These render the combined production capacity of the existing CdTe and CIGS technologies at a small scale lower than 100 GW per year. This is only a

small fraction of energy consumption in 2050 which is expected to be 27 TW [4, 9].

Recently, quaternary compound Cu2ZnSnS4 (CZTS) has been intensively examined as an al‐ ternative PV material due to its similarity in material properties with CIGS and the relative abundance of raw materials (Fig. 2). CZTS is a compound semiconductor of (I)2(II)(IV)(VI)4.

Significant progress on this relatively new research area has been achieved in last five years. Champion efficiency of CZTS thin film solar cell (TFSC) has reached 8.4 % and an efficiency of 6.21 % has been demonstrated for CZTS sub-module with an area of 22.6 cm2 . However, these efficiencies are still much lower than those of CIGS PV devices. This chapter reviews the present status of various CZTS TFSC technologies with special emphasis on properties of CZTS thin films deposited by different methods. New results generated by solution-based processing have been reported, and the methodologies to make CZTS photovoltaic technolo‐ gy more marketable are also proposed and discussed. Based on the information reported and our experiences gained on the research and development of CIGS and CZTS solar cells, the challenges and perspectives of CZTS TFSCs have been addressed.

## **2. General properties of CZTS thin film**

In 1967, CZTS single crystal was synthesized and analyzed [11]. However, it had not gained intensive interest from academies and industries until 2007 when solar power was heavily subsidized by governments and Si-based PV technologies encountered a skyrocketing price of highly pure polycrystalline silicon. Thus far, structural, optical, and electrical properties of CZTS thin film have been intensively investigated.

## **2.1. Crystal structure**

CZTS thin films are usually in a polycrystalline form consisting of kesterite crystal struc‐ tures. Kesterite CZTS single crystal was first synthesized by Nitsche et al. using the chemical vapor transport method [11]. X-ray diffraction (XRD) results showed that this synthesized CZTS had sphalerite-like crystal structure with c/a being close to 2 (a=5.43 Å, c=10.83 Å). In 1974, detailed lattice data of a CZTS single crystal were reported by Schäfer and Nitsche (Table 1) [12]. Thereafter, this data was frequently referenced to determine CZTS phase in literatures. In 2011, Lu et al. claimed that wurtzite CZTS nanocrystals were synthesized through a hot injection method [13]. The experimental XRD patterns were indexed to a si‐ mulated crystal structure with a wurtzite phase (Table 2).


**Table 1.** Lattice data of the kesterite CZTS single crystal [12].

Notes: *d* indicates the distance between two neighbor parallel planes, *I*/*I <sup>0</sup>* is the relative peak intensity, (*hkl*) are Miller indexes, and 2*θ* is the twice of Bragg diffraction angle.

Kesterite CZTS has highly similar crystal structure with chalcopyrite CIGS where half of indi‐ um and (or) gallium is replaced by zinc and the other half by tin (Fig. 4(a)). Similar to ZnO or ZnS, the anions and cations in kesterite CZTS crystal are located in a tetrahedral bonding envi‐ ronment with a stacking model which is similar to zincblende (Fig. 4(c)) [14]. The other impor‐ tant structure for CZTS crystal is a stannite structure (Fig. 4(d)) [14]. The difference between kesterite CZTS crystal and stannite CZTS crystal lies in a different order in the cation sub-lat‐ tice. In kesterite CZTS, cation layers of CuSn, CuZn, CuSn, and CuZn alternate at z = 0, 1/2, 1/2, and 3/4, respectively; while in stannite structures, ZnSn layers alternate with Cu2 layers. The wurtzite structure can be formed by replacing Zn (II) with Cu (I), Zn (II) and Sn (IV) in wurtzite ZnS with each sulfur atom equally coordinating with two Cu (I), one Zn (II), and one Sn (IV) (Fig. 4(b)) [13]. First-principle calculations by Chen et al. indicated that the kesterite structure had a lower energy and should be more stable than the stannite structure [15]. Most of CZTS samples crystallized in the kesterite structure as predicted theoretically. XRD results showed that diffraction peaks of (112), (200), (220/204), and (312/116) with a preferred orientation along (112) were commonly observed [16-19]. Peaks of (002), (008), (101), (103), (105), (110), (211), (213), (224), and (332) were also demonstrated in the XRD spectra [20- 24].


**Table 2.** Lattice data of the wurtzite CZTS crystal [13].

### **2.2. Optical properties**

**2.1. Crystal structure**

110 Solar Cells - Research and Application Perspectives

CZTS thin films are usually in a polycrystalline form consisting of kesterite crystal struc‐ tures. Kesterite CZTS single crystal was first synthesized by Nitsche et al. using the chemical vapor transport method [11]. X-ray diffraction (XRD) results showed that this synthesized CZTS had sphalerite-like crystal structure with c/a being close to 2 (a=5.43 Å, c=10.83 Å). In 1974, detailed lattice data of a CZTS single crystal were reported by Schäfer and Nitsche (Table 1) [12]. Thereafter, this data was frequently referenced to determine CZTS phase in literatures. In 2011, Lu et al. claimed that wurtzite CZTS nanocrystals were synthesized through a hot injection method [13]. The experimental XRD patterns were indexed to a si‐

> *d* (Å) *I*/*I0* (%) (*hkl*) 2θ (degree) 5.421 1 002 16.338 4.869 6 101 18.205 3.847 2 110 23.101 3.126 100 112 28.530 3.008 2 103 29.675 2.713 9 200 32.989 2.426 1 202 37.025 2.368 3 211 37.966 2.212 1 114 40.758 2.013 2 105 44.996 1.919 90 220 47.331 1.636 25 312 56.177 1.618 3 303 56.858 1.565 10 224 58.969 1.45 1 314 64.177 1.356 2 008 69.229 1.245 10 332 76.442

Notes: *d* indicates the distance between two neighbor parallel planes, *I*/*I <sup>0</sup>* is the relative peak

Kesterite CZTS has highly similar crystal structure with chalcopyrite CIGS where half of indi‐ um and (or) gallium is replaced by zinc and the other half by tin (Fig. 4(a)). Similar to ZnO or ZnS, the anions and cations in kesterite CZTS crystal are located in a tetrahedral bonding envi‐

intensity, (*hkl*) are Miller indexes, and 2*θ* is the twice of Bragg diffraction angle.

mulated crystal structure with a wurtzite phase (Table 2).

**Table 1.** Lattice data of the kesterite CZTS single crystal [12].

The optical bandgap of stoichiometric kesterite-CZTS was theoretically determined to be 1.50 eV [15]. Experimental results demonstrated that bandgap of CZTS thin film deposited using different method varied from 1.4 eV to 1.5 eV [16, 19, 20, 25, 26]. It is commonly recog‐ nized that CZTS thin film has an absorption coefficient as high as 104 cm-1. Sol-gel derived CZTS thin film from our group confirmed that the absorption coefficient is higher than 104 cm-1 in the photon energy range greater than 1.2 eV (Fig. 5) [27].

Raman spectrum is a powerful characterization method to reveal Raman shift peaks in CZTS especially for the peaks associated with secondary phases such as CuxS, ZnS, SnxS, Cu2SnS3, and Cu3SnS4. The universally acknowledged peak is 338 cm-1 [28]. However, peaks at 96 cm−1, 166 cm−1, 250 cm−1, 251 cm−1, 287 cm−1, 288 cm−1, 289 cm−1, 337 cm−1, 338 cm−1, 352 cm−1, 370 cm−1, 372 cm−1 have also been observed and assigned to CZTS [16, 29-32].

To study the recombination mechanisms, low temperature photoluminescence spectra were recorded from CZTS thin films. Several groups observed a broad peak centering at around 1.24 eV, which was attributed to the typical donor-acceptor pair transition involving tail states created by potentials fluctuations [33-35]. It is claimed that the presence of potential fluctuations indicates CZTS is strongly compensated [33]. Time-resolved PL data illustrated that lifetime of free carriers in CZTS thin film was lower than 1 ns [36].

**Figure 4.** Crystal structures of CIGS and CZTS [13, 14].

### **2.3. Electrical properties**

In contrast with silicon, where either atoms of phosphorus or atoms of boron are intentional‐ ly introduced for producing n-type and p-type semiconductors, respectively, CZTS is selfdoped through a formation of intrinsic defects including vacancies (VCu, VZn, VSn, and VS), antisite defects (CuZn, ZnCu, CuSn, SnCu, ZnSn, and SnZn), and interstitial defects (Cui , Zni , and Sni ). These defects could form during growth of CZTS thin film. Chen et al. systematically studied the defect properties of CZTS using first-principle calculations [37]. It was found that the formation energy of acceptor defects was lower than that of donor defects, which makes n-type doping very difficult in CZTS [37]. The commonly observed p-type conductiv‐ ity of CZTS thin films comes mainly from the CuZn antisite defect, partly explaining why CZTS thin films must be Cu-poor and Zinc-rich to successfully fabricate CZTS solar cells.

fluctuations indicates CZTS is strongly compensated [33]. Time-resolved PL data illustrated

In contrast with silicon, where either atoms of phosphorus or atoms of boron are intentional‐ ly introduced for producing n-type and p-type semiconductors, respectively, CZTS is selfdoped through a formation of intrinsic defects including vacancies (VCu, VZn, VSn, and VS),

). These defects could form during growth of CZTS thin film. Chen et al. systematically studied the defect properties of CZTS using first-principle calculations [37]. It was found

, Zni

, and

antisite defects (CuZn, ZnCu, CuSn, SnCu, ZnSn, and SnZn), and interstitial defects (Cui

that lifetime of free carriers in CZTS thin film was lower than 1 ns [36].

112 Solar Cells - Research and Application Perspectives

**Figure 4.** Crystal structures of CIGS and CZTS [13, 14].

**2.3. Electrical properties**

Sni

**Figure 5.** a) Typical bandgap and (b) absorption coefficients of a CZTS thin film [27].

Reported resistivity of CZTS thin films was significantly different [38-40]. The most suitable value for CZTS thin film should range from 10-3Ω∙cm to 10-1Ω∙cm according to published data for CZTS solar cells [25]. The hole concentration was reported to vary from 1016cm-3 to 1018cm-3 [41-43], although extremely high and extremely low concentration were also report‐ ed [44, 45]. Hall effect measurement results showed that hole mobility of CZTS changed from lower than 0.1 to as high as 30 cm2 ∙V−1∙s−1, while most published values were in the range of 1 to 10 cm2 ∙V−1∙s−1 [40, 44-47]. The lower mobility indicates that the optimized thickness of absorber layer in CZTS TFSCs cannot be as large as that for CIGS TFSCs.

### **3. CZTS thin film solar cell**

We would like to first introduce the basic definitions which are very important parameters when evaluating a solar cell. The solar cell can be basically taken as a battery in a simple electri‐ cal circuit (Fig. 6 (a)). Without sunlight shining on solar cell, it can do nothing. However, the so‐ lar cell will work as a battery if it is activated by light (Fig. 6 (b)). Electrical potential difference will be developed between its two ends and electrical current can flow through the solar cell. The potential difference derived when the resistance of the load is infinite is defined as open circuit voltage, *V OC*. Correspondingly, the electrical current flowing in the circuit when the re‐ sistance of the load is zero is defined as short circuit current, *J SC*. Shown in Fig. 6 (c) is a typical *I-V* relationship curve of the illuminated solar cell when the load resistance changes from zero to infinite. The delivered power by the solar cell, *P*, is given by

$$P = I \times V \tag{1}$$

The typical *P-V* relationship curve is also shown in Fig. 6 (c). *P* reaches a maximum value at certain condition under which the solar cell will deliver the highest power to the external load. This condition is defined as maximum power point and the maximum power is denot‐ ed as *P <sup>m</sup>*. The corresponding voltage and current is denoted as *V <sup>m</sup>* and *I <sup>m</sup>*, respectively. An‐ other important parameter to evaluate a solar cell is fill factor, *FF*, which is defined as

$$FF = \frac{I\_{\rm m} \times V\_{\rm m}}{I\_{\rm SC} \times V\_{\rm OC}} \tag{2}$$

The most important parameter for a solar cell is conversion efficiency, *η*, which describes the solar cell's ability to translate solar energy into electrical energy. The conversion efficiency is given by

$$
\eta = \frac{P\_{\rm m}}{P\_{\rm L}} \tag{3}
$$

where *P <sup>L</sup>* is the power of the simulated light.

Parameters such as *J SC*, *V OC*, *FF*, and *η* are key performance characteristics of a solar cell. These parameters are light-dependent and environment-dependent, which means that the values of these parameters of a specific solar cell will change if the solar cell is illuminated with differ‐ ent light intensity. Worldwide recognized characterization condition for solar cells is the Standard Test Condition (STC) which stipulates that a solar cell should be tested at 25℃ under Air Mass 1.5 spectrum illumination with an incident power density of 100 mW/cm2 .

**Figure 6.** a) Schematic basic electrical circuit, (b) schematic basic operating circuit of solar cell, (c) typical I-V curve of a solar cell.

### **3.1. Basic structure and fabrication procedures**

ed [44, 45]. Hall effect measurement results showed that hole mobility of CZTS changed

We would like to first introduce the basic definitions which are very important parameters when evaluating a solar cell. The solar cell can be basically taken as a battery in a simple electri‐ cal circuit (Fig. 6 (a)). Without sunlight shining on solar cell, it can do nothing. However, the so‐ lar cell will work as a battery if it is activated by light (Fig. 6 (b)). Electrical potential difference will be developed between its two ends and electrical current can flow through the solar cell. The potential difference derived when the resistance of the load is infinite is defined as open circuit voltage, *V OC*. Correspondingly, the electrical current flowing in the circuit when the re‐ sistance of the load is zero is defined as short circuit current, *J SC*. Shown in Fig. 6 (c) is a typical *I-V* relationship curve of the illuminated solar cell when the load resistance changes from zero

The typical *P-V* relationship curve is also shown in Fig. 6 (c). *P* reaches a maximum value at certain condition under which the solar cell will deliver the highest power to the external load. This condition is defined as maximum power point and the maximum power is denot‐ ed as *P <sup>m</sup>*. The corresponding voltage and current is denoted as *V <sup>m</sup>* and *I <sup>m</sup>*, respectively. An‐ other important parameter to evaluate a solar cell is fill factor, *FF*, which is defined as

The most important parameter for a solar cell is conversion efficiency, *η*, which describes the solar cell's ability to translate solar energy into electrical energy. The conversion efficiency is

Parameters such as *J SC*, *V OC*, *FF*, and *η* are key performance characteristics of a solar cell. These parameters are light-dependent and environment-dependent, which means that the values of these parameters of a specific solar cell will change if the solar cell is illuminated with differ‐ ent light intensity. Worldwide recognized characterization condition for solar cells is the

*FF* <sup>=</sup> *<sup>I</sup>*<sup>m</sup> <sup>×</sup>*V*<sup>m</sup> *I*SC ×*V*OC

> *<sup>η</sup>* <sup>=</sup> *<sup>P</sup>*<sup>m</sup> *P*L

thickness of absorber layer in CZTS TFSCs cannot be as large as that for CIGS TFSCs.

∙V−1∙s−1, while most published values were in the

*P* = *I* ×*V* (1)

(2)

(3)

∙V−1∙s−1 [40, 44-47]. The lower mobility indicates that the optimized

from lower than 0.1 to as high as 30 cm2

to infinite. The delivered power by the solar cell, *P*, is given by

where *P <sup>L</sup>* is the power of the simulated light.

**3. CZTS thin film solar cell**

114 Solar Cells - Research and Application Perspectives

range of 1 to 10 cm2

given by

The schematic structure of CZTS solar cell is shown in Fig.7. Molybdenum thin film with thickness of 500~700 nm is sputtering-deposited on glass substrate as back contact because Mo is stable in harsh reactive conditions such as sulfur-containing vapor and high tempera‐ ture. The absorber layer, p-type CZTS thin film with thickness ranging from 1.0 to 2.0 μm is then coated on Mo thin film. To form p-n junction with the p-type CZTS, 50~100 nm n-type CdS thin film is deposited on the absorber layer usually by chemical bath deposition. The surface of CZTS thin film is too rough to be fully covered by CdS thin film, leading to short‐ age between front contact and back contact. To prevent leakage, 50~90 nm intrinsic ZnO (i-ZnO) thin film is usually sputtering-coated on CdS before 500~1000 nm transparent conducting oxide (TCO) thin film is deposited by sputtering as the front contact layer of the cell. Finally, to electrically measure the *I-V* property of CZTS solar cell, Ni/Al grid is sepa‐ rately deposited on both TCO and Mo layer.

## **3.2. Deposition techniques of CZTS thin films**

The first (I)2(II)(IV)(VI)4 solar cell was developed in 1977 by Wagner and Bridenbaugh. A ntype CdS thin film was evaporation-coated on vapor transportation-grown Cu2CdSnS4 sin‐ gle crystal substrate to form the p-n junction [48]. This device showed a short-circuit current density of 7.9 mA/cm2 , an open-circuit voltage of 0.5 V, and a conversion efficiency of 1.6%. The authors pointed out that a large series resistance limited the performance. In 1988, a het‐ erojunction solar cell with an open circuit voltage of 165 mV was achieved by depositing cadmium tin oxide on CZTS thin film [40]. In 1997, the first CZTS TFSC with efficiency of 0.66% was realized by Katagiri using electron beam deposition followed by sulfurization [49]. The highest efficiency of 8.4% for CZTS TFSC and 6.21% for sub-module has been ach‐ ieved by IBM and by Solar Frontier, respectively [36, 50].

**Figure 7.** Schematic structure of typical CZTS solar cell.

Shown in Fig.8 are conversion efficiencies for CZTS solar cells obtained by different meth‐ ods. Evaporation and sputtering have been intensively employed for the deposition of CZTS thin film both because the properties of CZTS thin film are more readily controlled using these two methods and because great success has been achieved experimentally and com‐ mercially for CIGS solar cells manufactured using evaporation and sputtering.

**Figure 8.** Conversion efficiencies obtained for CZTS solar cells by different methods.

The highest efficiencies achieved by the specific methods shown in Fig. 8 are listed in Table 3. Many technologies have been explored for fabricating CZTS TFSCs, as discussed in detail in the following paragraphs. Similar to the CIGS solar cells, whose highest efficiency was obtained by evaporation deposition [75], the highest efficiency of CZTS solar cell was also attained by evaporation deposition [36]. Conversion efficiency of 10.1% by IBM and 7.23% by Guo et al. have been realized for Cu2ZnSn(S,Se)4 (CZTSSe) solar cells made using solu‐ tion-based method and nanoparticle-based method, respectively [76,77]. However, both cas‐


es modified the composition of CZTS thin films through introduction of selenium, which is a rare element in the earth crust.

**Table 3.** Highest efficiency achieved for CZTS solar cell by different method.

### *3.3.1. Evaporation*

**Figure 7.** Schematic structure of typical CZTS solar cell.

116 Solar Cells - Research and Application Perspectives

Shown in Fig.8 are conversion efficiencies for CZTS solar cells obtained by different meth‐ ods. Evaporation and sputtering have been intensively employed for the deposition of CZTS thin film both because the properties of CZTS thin film are more readily controlled using these two methods and because great success has been achieved experimentally and com‐

The highest efficiencies achieved by the specific methods shown in Fig. 8 are listed in Table 3. Many technologies have been explored for fabricating CZTS TFSCs, as discussed in detail in the following paragraphs. Similar to the CIGS solar cells, whose highest efficiency was obtained by evaporation deposition [75], the highest efficiency of CZTS solar cell was also attained by evaporation deposition [36]. Conversion efficiency of 10.1% by IBM and 7.23% by Guo et al. have been realized for Cu2ZnSn(S,Se)4 (CZTSSe) solar cells made using solu‐ tion-based method and nanoparticle-based method, respectively [76,77]. However, both cas‐

mercially for CIGS solar cells manufactured using evaporation and sputtering.

**Figure 8.** Conversion efficiencies obtained for CZTS solar cells by different methods.

Evaporation is a well-known technique in the development of thin film solar cells. In 1997, Katagiri et al. reported electron beam evaporation-deposited CZTS precursor films followed by sulfurization [49]. Solar cell with an efficiency of 0.66% was obtained. In this work, Zn, Sn and Cu layers were sequentially deposited on Mo-coated soda lime glass substrates which were heated up to 150 ℃. The targeted composition ratio was decided by the thickness of metallic layers. Annealing at 500 ℃ in the atmosphere of N<sup>2</sup> + H2S (5%) was then employed to transform Cu/Sn/Zn stacked layers into a CZTS thin film. Finally, chemical bath deposi‐ tion was employed to deposit n-type CdS thin film on the p-type CZTS to form a p-n junc‐ tion. As a result, the open-circuit voltage was significantly enhanced in comparison to the previously reported value [40].

Similar deposition procedures were performed by the same group in 2001 with a replace‐ ment of Zn metal source by ZnS [25]. Also, the annealing temperature was increased to 550 ℃. CZTS thin films with thickness of 0.95 μm, 1.34 μm, and 1.63 μm were deposited on Mocoated SLG substrates, respectively. The *J-V* results demonstrated that the short-circuit cur‐ rent density and the fill factor of these cells drastically decreased with the increase of thickness of CZTS thin film (Table 4). The authors concluded that the extremely high series resistance of CZTS absorber layer was attributed to the significant degradation encountered in the CZTS solar cells. Similar dependence of performance on the thickness of CZTS ab‐ sorber layer was demonstrated in 2010 by Wang et al. (Table 4) [56]. Capacitance-voltage measurement results showed that the density of uncompensated charge in the CZTS layer was to be 5×1016 cm-3 at 27 ℃, which indicates that part of the high series resistance comes from the bulk CZTS absorber layer. Evaluation of the dependence of *R <sup>s</sup>* on temperature from dark *J-V* curve provides insight on another potential source for the series resistance (Fig. 9). The strong dependence indicates a back-contact blocking (Schottky) barrier exists at the interface between CZTS and Mo, leading to the suppression of holes transporting across the interface to Mo.


**Table 4.** Comparison of *I-V* properties of CZTS TFSCs with different thickness of absorber layer [25, 56].

**Figure 9.** The dependence of *Rs* on temperature [56].

The large dependence of physical properties of CIGS thin film on the ratio of Cu/(In+Ga) suggests that it is necessary to investigate the effects of Cu/(Zn+Sn) ratio on the properties of CZTS films to further the understanding of CZTS solar cells [78]. In 2010, Tanaka et at. em‐ ployed evaporation method to fabricate CZTS samples with constant Zn/Sn and S/metal ra‐ tios of 1.1 and 0.93 and with Cu/(Zn+Sn) ratio varying from 0.82 to 1.06 [79]. All samples were determined to be kesterite structure. XRD data showed that FWHM of the diffraction peak of (112) plane became narrower, and the I(112)/I0 increased with increasing Cu/(Zn+Sn) ratio. This indicated that the increasing of Cu/(Zn+Sn) ratio helped improve the crystallinity of CZTS films (Fig. 10 (a)). Surface SEM images of the CZTS films demonstrated that the grain size also increased with increasing Cu/(Zn+Sn) ratio (Fig. 10 (b)), suggesting deposi‐ tion process containing Cu-rich condition could be developed for growing high-quality CZTS thin films. The champion CIGS solar cell was fabricated using three-stage co-evapora‐ tion method where CIGS thin film was changed to Cu-rich in the second stage from Cu-poor in the first stage. In the last stage, the Cu source was blocked and Ga, In, and Se were simul‐ taneously deposited to restore the Cu-poor state. Similar procedures have yet to be proved effective for CZTS solar cells.

sorber layer was demonstrated in 2010 by Wang et al. (Table 4) [56]. Capacitance-voltage measurement results showed that the density of uncompensated charge in the CZTS layer was to be 5×1016 cm-3 at 27 ℃, which indicates that part of the high series resistance comes from the bulk CZTS absorber layer. Evaluation of the dependence of *R <sup>s</sup>* on temperature from dark *J-V* curve provides insight on another potential source for the series resistance (Fig. 9). The strong dependence indicates a back-contact blocking (Schottky) barrier exists at the interface between CZTS and Mo, leading to the suppression of holes transporting across

R & D group Thickness (nm) *J*SC (mA/cm2) *V*OC (mV) *FF* (%) η (%) Katagiri et al. 950 7.01 415 50.3 1.46

Wang et al. 650 17.8 587 65 6.81

**Table 4.** Comparison of *I-V* properties of CZTS TFSCs with different thickness of absorber layer [25, 56].

**Figure 9.** The dependence of *Rs* on temperature [56].

1340 3.41 425 26.5 0.384 1630 1.53 525 26.6 0.214

660 20.4 620 52 6.63 900 18.3 640 38 4.40 1200 14.4 608 28 2.44

The large dependence of physical properties of CIGS thin film on the ratio of Cu/(In+Ga) suggests that it is necessary to investigate the effects of Cu/(Zn+Sn) ratio on the properties of CZTS films to further the understanding of CZTS solar cells [78]. In 2010, Tanaka et at. em‐

the interface to Mo.

118 Solar Cells - Research and Application Perspectives

Na incorporation into the CIGS polycrystalline thin film is a necessary process to fabricate CIGS modules with high efficiency. The enhancement of efficiency largely comes from high‐ er open-circuit voltage, improvement of fill factor, increasing of p-type conductivity as well as improvement of crystallinity of (112)-oriented CIGS films [80-82]. Na incorporation was performed for evaporated-CZTS solar cell by Katagiri et al. using Na2S as Na source [54]. It was found that the efficiency was enhanced from 4.25% to 5.45%. This enhancement is main‐ ly due to the increase of short-circuit current density which increased significantly from 10.3 mA/cm2 to 15.5 mA/cm2 . The open-circuit voltage and fill factor were slightly lower than the CZTS solar cells without Na incorporation. While why the efficiency was improved was not mentioned in the paper, it can be assumed that Na incorporation can improve the quality of CZTS thin film at different mechanism compared with CIGS thin film. The effects of Na in‐ corporation for improving the quality of CZTS thin film has been addressed by Jampana et al. [83]. CZTS thin films were deposited on soda-lime glass (SLG) and low-alkaline glass (LAG). An increase in grain size and an improvement of morphology were obviously dem‐ onstrated in CZTS thin films deposited on SLG (Fig. 11 (a) and Fig. 11 (b)). However, our experiment results indicated that no significant difference could be detected from CZTS thin films deposited on SLG and low-alkaline glass (Fig. 11 (c) and Fig. 11 (d)). The wide varia‐ tion among effects of sodium incorporation on CZTS thin films possibly partly arises from the difference of deposition methods. The other explanation could be tracked to the lack of precision by only changing substrate type. More experiments have to be carried out on so‐ dium-free substrate by exactly controlling the doping amount of sodium before sodium in‐ corporation can be effectively employed to improve the performance of CZTS PV devices.

In 2011, IBM reported a champion efficiency of 8.4% was achieved for CZTS solar cell which was grown on Mo-coated SLG substrates by thermal evaporation using elemental Cu, Zn, Sn, and S as sources [36]. A cracker, which can increase the reactivity of S, was applied to S vapor. The employment of cracker probably helped improve the efficiency.

**Figure 10.** a) FWHM of the (112) diffraction peak and normalized (112) diffraction intensity of CZTS films (I112/d) as a function of Cu/(Zn+Sn) ratio, (b) SEM surface images of CZTS films with different Cu/(Zn+Sn) ratios [79].

All the reports showed that the conventional evaporation method is efficient for the devel‐ opment of CZTS TFSCs. However, non-uniformity caused by splash encountered in the evaporation of copper will significantly deteriorate the performance of CZTS module as does in the fabrication of CIGS module (Unpublished data). The fact that the commercial production capacity of CIGS solar modules by evaporation is far lower than that by sputter‐ ing is a perfect suggestion for CZTS solar industries [84]. More stable and controllable sput‐ tering method is more readily applied to mass production of CZTS solar modules.

### *3.3.2. Sputtering*

In 1988, Ito analyzed the electrical and optical properties of CZTS thin film which was de‐ posited on slide glass substrate by atom beam sputtering [40]. The deposited CZTS thin film was (112)-oriented and polycrystalline. The grain size increased when CZTS thin film was deposited at higher temperature because the mobility of sputtered particles was higher on the substrate surface (Fig. 12(a)). Its resistivity decreased from 4x104 Ω∙cm to 1.3 Ω∙cm with the increase of deposition temperature (Fig. 12(b)). Hall-effect measurement estimated that CZTS thin film had mobility lower than 0.1 cm2 ∙V-1∙s-1 and the carrier concentration was higher than 5×1019 cm-3. This CZTS thin film was considered optically desirable for the ab‐ sorber layer of solar cell because the deposited CZTS thin film had an absorption coefficient larger than 1.2×104 cm-1 and a direct bandgap of 1.45 eV.

**Figure 10.** a) FWHM of the (112) diffraction peak and normalized (112) diffraction intensity of CZTS films (I112/d) as a

**Figure 11.** SEM surface images of CZTS thin films deposited on different substrates: (a) LAG, (b) SLG, (c) LAG, (d) SLG.

All the reports showed that the conventional evaporation method is efficient for the devel‐ opment of CZTS TFSCs. However, non-uniformity caused by splash encountered in the evaporation of copper will significantly deteriorate the performance of CZTS module as does in the fabrication of CIGS module (Unpublished data). The fact that the commercial production capacity of CIGS solar modules by evaporation is far lower than that by sputter‐ ing is a perfect suggestion for CZTS solar industries [84]. More stable and controllable sput‐

In 1988, Ito analyzed the electrical and optical properties of CZTS thin film which was de‐ posited on slide glass substrate by atom beam sputtering [40]. The deposited CZTS thin film

tering method is more readily applied to mass production of CZTS solar modules.

[(a) and (b) were reported by Jampana et al. [83] (c) and (d) are results from our group].

*3.3.2. Sputtering*

function of Cu/(Zn+Sn) ratio, (b) SEM surface images of CZTS films with different Cu/(Zn+Sn) ratios [79].

120 Solar Cells - Research and Application Perspectives

**Figure 12.** a) (112) plane spacing (■) and FWHM of the (112) peak (○) of CZTS, (b) the correlation between resistivity (ρ) of CZTS and substrate temperature [40].

In 2003, Seol et al. deposited CZTS films using RF-magnetron sputtering system and quaterna‐ ry CZTS target (composed of finely mixed Cu2S, ZnS and SnS2 at ratio of 2:1.5:1) followed by annealing in the atmosphere of Ar+S (g) [85]. The effects of sputtering power and annealing temperature on the properties of CZTS thin films were checked. It was found that the atomic ratio of the thin films obtained between 50 W and 100 W was appropriate. However, the Cu content of CZTS thin films was significantly decreased while the Sn content was rapidly in‐ creased with a power above 100 W. The authors suggested that the plasma density caused the abrupt changes of the Cu and Sn contents. CZTS thin films annealed at above 250 ℃ were (112) oriented and other major diffraction peaks were assigned to (200), (220), and (312) planes. As annealing temperature increased, the intensity of the (112) peak was stronger.

In 2005, hybrid sputtering was employed by Tanaka et al. to prepare CZTS thin films [44]. The hybrid sputtering system was constructed in a deposition chamber with two effusion cells for Zn and S and two sputtering sources for Cu and Sn. CZTS thin films were fabricat‐ ed by sequential deposition of Sn, Zn, and Cu followed by annealing in S vapor. The sub‐ strate temperature was varied between 300 ℃ and 500 ℃. The film thickness decreased with increasing of substrate temperature. This was probably caused by the decrease of the stick‐ ing coefficient and/or by the increase of density due to crystallization at high temperature. CZTS thin films remained stoichiometric when the substrate temperature was elevated to up to 400 ℃. However, the composition of the thin films became Zn-poor at the substrate tem‐ perature above 450 ℃. At higher temperature, the vapor pressure of Zn is higher, leading to loss of Zn. It was proposed that Zn loss at higher substrate temperature could be prevented by using binary compound ZnS instead of Zn or by introducing S vapor during the deposi‐ tion of Zn to form zinc sulfide on the surface of precursor.

CZTS precursor films are generally taken out of the deposition chamber and exposed to the atmosphere before sulfurization is performed to grow CZTS polycrystalline thin films. Moisture can be adsorbed on the surface of CZTS precursor films which is thereof oxidized during annealing at high temperature. Thus, in-line sulfurization should be capable of avoiding the problem. Jimbo et al. carried out this process with sputtered CZTS precursor films to curb the oxidization [57]. Targets of Cu, ZnS and SnS were simultaneously sput‐ tered by RF sources. The finished precursor was automatically transferred to the annealing chamber without being exposed to atmosphere and annealed at 580 ℃ for 3 h in an atmos‐ phere of N2+H2S (20%). Measurement data showed that annealed CZTS thin film had the thickness of 2.5 μm and bandgap of 1.45 eV. The film was copper-poor and slightly zinc-rich and sulfur-rich (Cu/Zn+Sn: 0.87, Zn/Sn: 1.15, S/metals: 1.17). The sample had an open circuit voltage of 662 mV, a short circuit current of 15.7 mA/cm<sup>2</sup> , a fill factor of 55%, a conversion efficiency of 5.74%. The improved efficiency was attributed to the in-line annealing process and better CZTS morphology achieved.


**Table 5.** Comparison of *J-V* properties of CZTS solar cells treated with and without deionized-water soaking [57, 59].

In 2008, the champion efficiency for CZTS solar cell was achieved by Katagiri's group through preferential etching technique where the CZTS absorber layer on the Mo-coated SLG substrate was soaked in deionized water (DI-water) for 10 min before the CdS buffer layer was grown on the CZTS absorber layer using chemical bath deposition method [59]. The comparison of *J-V* properties was listed in Table 5. As we can see, DI-water soaking treatment CZTS absorber layer was very effective to improve the efficiency. The effect of the DI-water soaking on the oxygen distribution in CZTS thin film was studied by electron probe X-ray micro analysis (EPMA). Areas with higher concentration of oxygen are scat‐ tered in the CZTS layer before the soaking treatment (Fig. 13(a)). In contrast, the concentra‐ tion of oxygen in the CZTS layer after the soaking treatment is lower than the measurement limit of the EPMA instrument (Fig. 13(b)). The authors suggested that oxygen removed was in the form of metal oxide because metal oxide nanoparticles are easy to dissolve in water. The removal of these metal oxide nanoparticles by DI-water is beneficial to improve the per‐ formance of CZTS solar cell. On the one hand, more sunlight will be absorbed by CZTS, re‐ sulting in higher *J SC* because free carriers will not be generated in the metal oxides as effectively as CZTS. On the other hand, the contact area between CdS buffer layer and CZTS absorber layer will be increased after the removal of metal oxide nanoparticles, leading to lower series resistance of CZTS PV device.

**Figure 13.** Distributions of oxygen in the CZTS layer before (a) and after (b) DI-water soaking for 4 h (bright areas are higher concentration areas of oxygen) [58].

As introduced above, the material properties of CZTS thin films and the performance of CZTS TFSCs are not only highly dependent on the deposition techniques but also dependent on how the sputtering target is made and what it is made of. The cooperation between PV device re‐ searchers and sputtering target vendors has to be intensified to make full use of their respec‐ tive experiences and therefore to expedite the development of CZTS PV technology.

## *3.3.3. Pulsed laser deposition (PLD)*

increasing of substrate temperature. This was probably caused by the decrease of the stick‐ ing coefficient and/or by the increase of density due to crystallization at high temperature. CZTS thin films remained stoichiometric when the substrate temperature was elevated to up to 400 ℃. However, the composition of the thin films became Zn-poor at the substrate tem‐ perature above 450 ℃. At higher temperature, the vapor pressure of Zn is higher, leading to loss of Zn. It was proposed that Zn loss at higher substrate temperature could be prevented by using binary compound ZnS instead of Zn or by introducing S vapor during the deposi‐

CZTS precursor films are generally taken out of the deposition chamber and exposed to the atmosphere before sulfurization is performed to grow CZTS polycrystalline thin films. Moisture can be adsorbed on the surface of CZTS precursor films which is thereof oxidized during annealing at high temperature. Thus, in-line sulfurization should be capable of avoiding the problem. Jimbo et al. carried out this process with sputtered CZTS precursor films to curb the oxidization [57]. Targets of Cu, ZnS and SnS were simultaneously sput‐ tered by RF sources. The finished precursor was automatically transferred to the annealing chamber without being exposed to atmosphere and annealed at 580 ℃ for 3 h in an atmos‐ phere of N2+H2S (20%). Measurement data showed that annealed CZTS thin film had the thickness of 2.5 μm and bandgap of 1.45 eV. The film was copper-poor and slightly zinc-rich and sulfur-rich (Cu/Zn+Sn: 0.87, Zn/Sn: 1.15, S/metals: 1.17). The sample had an open circuit

efficiency of 5.74%. The improved efficiency was attributed to the in-line annealing process

Before DI-water soaking 15.7 662 55 9.04 612 5.74 After DI-water soaking 17.9 610 62 4.25 370 6.77

**Table 5.** Comparison of *J-V* properties of CZTS solar cells treated with and without deionized-water soaking [57, 59].

In 2008, the champion efficiency for CZTS solar cell was achieved by Katagiri's group through preferential etching technique where the CZTS absorber layer on the Mo-coated SLG substrate was soaked in deionized water (DI-water) for 10 min before the CdS buffer layer was grown on the CZTS absorber layer using chemical bath deposition method [59]. The comparison of *J-V* properties was listed in Table 5. As we can see, DI-water soaking treatment CZTS absorber layer was very effective to improve the efficiency. The effect of the DI-water soaking on the oxygen distribution in CZTS thin film was studied by electron probe X-ray micro analysis (EPMA). Areas with higher concentration of oxygen are scat‐ tered in the CZTS layer before the soaking treatment (Fig. 13(a)). In contrast, the concentra‐ tion of oxygen in the CZTS layer after the soaking treatment is lower than the measurement limit of the EPMA instrument (Fig. 13(b)). The authors suggested that oxygen removed was in the form of metal oxide because metal oxide nanoparticles are easy to dissolve in water. The removal of these metal oxide nanoparticles by DI-water is beneficial to improve the per‐

*J*SC (mA/cm2) *V*OC (mV) *FF* (%) *R*S (Ω∙cm2) *R*Sh (Ω∙cm2) η (%)

, a fill factor of 55%, a conversion

tion of Zn to form zinc sulfide on the surface of precursor.

voltage of 662 mV, a short circuit current of 15.7 mA/cm<sup>2</sup>

and better CZTS morphology achieved.

122 Solar Cells - Research and Application Perspectives

So far, laser has only successfully been applied to formation of interconnection paths be‐ tween individual cells in series-connected solar modules such as a-Si, CdTe, and CIGS. In 2007, Moriya et al. deposited CZTS thin films on Mo-coated soda lime glass (SLG) substrate at room temperature using KrF excimer laser for ablating sintered CZTS pellets [68]. Anneal‐ ing at 500 ℃ in N2 was carried out for growing CZTS crystals. The CZTS TFSCs showed an open-circuit voltage of 546 mV, a short-circuit current of 6.78 mA/cm2 , a fill factor of 48% and a conversion efficiency of 1.74%.

In 2010, Pawar et al. investigated the effect of incident energy density of laser on the structur‐ al, morphological and optical properties of CZTS thin films using similar PLD method as re‐ ported previously [68, 86]. Laser incident energy density was changed from 1.0 J/cm2 to 3.0 J/cm2 . XRD results indicated that the crystallinity of the as-deposited CZTS thin films was im‐ proved with the increase of laser incident density up to 2.5 J/cm2 . However, the film was slight‐ ly degraded when the laser energy density was further increased to 3.0 J/cm2 due to the large plasma density and high kinetic energy induced by too intense laser. SEM surface images of the annealed CZTS thin films showed the average grain size increased, and these films be‐ came relatively more uniform as the laser incident energy was increased from 1.0 to 2.5 J/cm2 . The bandgap decreased with the increase of laser incident energy up to 2.5 J/cm2 as well.

The efficiency of CZTS solar cell fabricated by PLD was further improved to 3.14% in 2011 [69]. CZTS pellets were made from Cu2S, ZnS, and SnS2 mixed powders (molar ration of 1:1:1) which were synthesized by solid state reaction method. CZTS thin films were then de‐ posited by PLD method in high vacuum using the CZTS pellets as source. These films were further annealed under N2 (95%) + H2S (5%) atmosphere at 400 ℃ for 1 h. The best CZTS solar cell reported in this work had a *V OC* of 651 mV, an *I SC* of 8.76 mA/cm2 , and a *FF* of 55%.

### *3.3.4. Non-vacuum processes*

Cost-effectiveness is the core of the development of any new technology. Total abandon‐ ment of vacuum facilities in the manufacture of PV systems is the best way to further lower the cost of PV modules. Several non-vacuum methods have been successfully employed in the development of CZTS TFSCs.

### *3.3.4.1. Electrodeposition*

The first CZTS solar cell deposited using electrodeposition was achieved by Scragg et al. [41]. In this method, copper chloride, tin chloride and zinc chloride were separately dis‐ solved in a mixture solution containing NaOH and sorbitol. Metal layers were potentiostati‐ cally deposited at room temperature in the order Cu, Sn, Zn using a conventional 3 electrode electrochemical cell with a platinum counter electrode and Ag/AgCl reference electrode. The electroplated metallic films and sulfur powder were loaded into a graphite container, which was inserted into a furnace tube. CZTS thin films were then synthesized at 550 ℃ by the sulfurization of the electroplated metallic films. The fabricated solar cell dem‐ onstrated an efficiency of 0.8% with an open circuit voltage of 295 mV, a short circuit current density of 8.7 mA/cm2 , and a fill factor of 32%.

The crystallization and sulfurization processes of the elecrodeposited CZTS precursor films were investigated by Schurr et al. using angle-dispersive time-resolved XRD measurements [87]. Two different types of precursor films with copper-rich and copper-poor ratios in the as-deposited films were checked. It was found that the kesterite crystallization was complet‐ ed by the solid state reaction of Cu2SnS3 and ZnS in both cases. However, in-situ XRD data showed reaction path for the formation of Cu2SnS3 depended on the metal ratios in the asdeposited films. The reaction schemes were derived from time-resolved XRD results and shown below. The reactions can be described below.

For copper-rich samples,

$$2\text{Cu}\_3\text{Sn} + 7\text{S} \rightarrow 3\text{Cu}\_2\text{-xS} + 2\text{SnS}\_2;\tag{4}$$

$$2\text{CuZn} + 3\text{S} \rightarrow \text{Cu}\_{2\text{-x}}\text{S} + 2\text{ZnS};\tag{5}$$

Cu2ZnSnS4 Thin Film Solar Cells: Present Status and Future Prospects http://dx.doi.org/10.5772/50702 125

$$\text{Cu}\_{2-x}\text{S} + \text{SnS}\_2 \rightarrow \text{Cu}\_2\text{SnS}\_3\text{:}\tag{6}$$

$$\text{Cu}\_2\text{SnS}\_3 + \text{ZnS} \rightarrow \text{Cu}\_2\text{ZnSnS}\_{\tilde{\Psi}} \tag{7}$$

For copper-poor samples,

.

as well.

, and a *FF* of 55%.

came relatively more uniform as the laser incident energy was increased from 1.0 to 2.5 J/cm2

The efficiency of CZTS solar cell fabricated by PLD was further improved to 3.14% in 2011 [69]. CZTS pellets were made from Cu2S, ZnS, and SnS2 mixed powders (molar ration of 1:1:1) which were synthesized by solid state reaction method. CZTS thin films were then de‐ posited by PLD method in high vacuum using the CZTS pellets as source. These films were further annealed under N2 (95%) + H2S (5%) atmosphere at 400 ℃ for 1 h. The best CZTS

Cost-effectiveness is the core of the development of any new technology. Total abandon‐ ment of vacuum facilities in the manufacture of PV systems is the best way to further lower the cost of PV modules. Several non-vacuum methods have been successfully employed in

The first CZTS solar cell deposited using electrodeposition was achieved by Scragg et al. [41]. In this method, copper chloride, tin chloride and zinc chloride were separately dis‐ solved in a mixture solution containing NaOH and sorbitol. Metal layers were potentiostati‐ cally deposited at room temperature in the order Cu, Sn, Zn using a conventional 3 electrode electrochemical cell with a platinum counter electrode and Ag/AgCl reference electrode. The electroplated metallic films and sulfur powder were loaded into a graphite container, which was inserted into a furnace tube. CZTS thin films were then synthesized at 550 ℃ by the sulfurization of the electroplated metallic films. The fabricated solar cell dem‐ onstrated an efficiency of 0.8% with an open circuit voltage of 295 mV, a short circuit current

The crystallization and sulfurization processes of the elecrodeposited CZTS precursor films were investigated by Schurr et al. using angle-dispersive time-resolved XRD measurements [87]. Two different types of precursor films with copper-rich and copper-poor ratios in the as-deposited films were checked. It was found that the kesterite crystallization was complet‐ ed by the solid state reaction of Cu2SnS3 and ZnS in both cases. However, in-situ XRD data showed reaction path for the formation of Cu2SnS3 depended on the metal ratios in the asdeposited films. The reaction schemes were derived from time-resolved XRD results and

2Cu3Sn + 7S→3Cu2 −xS + 2SnS2; (4)

2CuZn + 3S→Cu2−xS + 2ZnS; (5)

The bandgap decreased with the increase of laser incident energy up to 2.5 J/cm2

solar cell reported in this work had a *V OC* of 651 mV, an *I SC* of 8.76 mA/cm2

, and a fill factor of 32%.

shown below. The reactions can be described below.

*3.3.4. Non-vacuum processes*

*3.3.4.1. Electrodeposition*

density of 8.7 mA/cm2

For copper-rich samples,

the development of CZTS TFSCs.

124 Solar Cells - Research and Application Perspectives

$$\rm Cu\_6Sn\_5 + S \rightarrow \rm SnS\_2 + Cu\_{2-x}S + Cu\_3Sn\gamma \tag{8}$$

$$\text{\textbullet } 2\text{Cu}\_2\text{S} + \text{SnS}\_2 + 2\text{S} \rightarrow \text{Cu}\_4\text{SnS}\_6\text{\textbullet} \tag{9}$$

$$\text{Cu}\_4\text{SnS}\_6 \rightarrow 2\text{Cu}\_{2-x}\text{S} + \text{SnS}\_2\text{(melted)} + 2\text{S};\tag{10}$$

$$\text{Cu}\_{2-x}\text{S} + \text{SnS}\_2\text{(melted)} \rightarrow \text{Cu}\_2\text{SnS}\_3\text{:}\tag{11}$$

$$\text{Cu}\_2\text{SnS}\_3 + \text{ZnS} \rightarrow \text{Cu}\_2\text{ZnSnS}\_{\tilde{\Psi}} \tag{12}$$

These reactions also indicate how the sulfurization and crystallization are completed in CZTS precursor films deposited using vacuum-based technologies since these methods usu‐ ally involve similar precursors.

In 2009, Ennaoui et al. achieved an efficiency of 3.4% using electrodeposited copper-poor CZTS thin film as absorber layer [61]. The formation of the binary sulfides and the thereof resulted liquid phase could contribute to the enhancement of kesterite crystal growth.

In 2012, electrodeposited CZTS solar cell with an efficiency of 7.3% was fabricated by Deli‐ gianni et al. using a three-step method [74]. Firstly, metal stacks of either Cu/Zn/Sn or Cu/Sn/Zn were electrodeposited. Secondly, low temperature annealing at 210–350 °C in N2 was employed to produce homogeneous (Cu, Zn) and (Cu, Sn) alloys. Lastly, these wellmixed CuZn and CuSn alloys were annealed at 550 –590 °C in sulfur vapor for 5 to 15 min. A single highly crystalline CZTS phase was achieved. This is so far the record efficiency for electrodeposited CZTS solar devices.

#### *3.3.4.2. Sol-gel method*

CZTS precursor sol-gel was made by dissolving copper (II) acetate monohydrate, zinc (II) acetate dehydrate and tin (II) chloride dehydrate in mixture solution of 2-methoxyethanol (2-metho), deionized water and binder and then spin-coated on Mo-coated soda lime glass substrates followed by drying at 300 ℃ on a hot plate [67]. The coating and drying process were repeated several times. Lastly, the precursors were annealed at 500 ℃ in an atmos‐ phere of N2 +H2S (5%). The CdS layer was grown on CZTS thin film by the chemical bath deposition (CBD) method. The CdS thickness was optimized by changing deposition time from 5 to 25 minutes. It was found that sample with CdS thin film deposited for 23 minutes showed the best conversion efficiency (*J SC*=6.70 mA/cm2 , *V OC*=554 mV, *FF*=43.4%, *η*=1.61%).

Tanaka et al. fabricated a CZTS solar cell with all semiconductor layers being coated by nonvacuum deposition techniques [66]. The ZnO:Al window layer and the CZTS absorber layer were deposited using sol-gel method. CdS buffer layer was coated by chemical bath deposi‐ tion method. CZTS precursor thin films were coated using sol–gel solution of Cu, Zn, and Sn ions. Sulfurization was employed at 500 ℃ under a mixture atmosphere of H2S and N2 with H2S concentration changing from 3% to 20%. CZTS thin film prepared with a H2S con‐ centration of 3% had grains with size of 1 μm. The best solar cell, which was obtained from a sample sulfurized at a H2S concentration of 3%, demonstrated a conversion efficiency of 2.23%. This is so far the highest efficiency for sol-gel method deposited CZTS solar cells.

**Figure 14.** a) SEM surface image of CZTS thin film, (b) typical J-V curve of CZTS TFSC deposited by sol-gel method [27].

We have deposited CZTS thin films by employing sol-gel method [26]. The deposited CZTS thin film consisted of large densely packed grains with size of more than 400 nm (Fig. 14(a)) and had an optical bandgap of 1.51 eV. The compositional ratios could be optimized through modification of the composition of CZTS sol-gel precursor and annealing processes. An efficiency of 0.63% has been demonstrated (Fig. 14(b)). The relatively low efficiency re‐ sulted from the low *V OC* and *FF* which was caused by the cracks and penetrating pores gen‐ erated during the annealing process. With this ongoing project, we have already improved FF by 19% using ~ 450 nm thick CZTS film by controlling the deposition process, leading to significant improvement of morphology (Fig. 15). A multiple layer deposition process suc‐ cessfully blocked the penetrating pores.

**Figure 15.** a) SEM surface image and (b) J-V property of CZTS TFSC with improved morphology.

### *3.3.4.3. Nanoparticle-based method*

Tanaka et al. fabricated a CZTS solar cell with all semiconductor layers being coated by nonvacuum deposition techniques [66]. The ZnO:Al window layer and the CZTS absorber layer were deposited using sol-gel method. CdS buffer layer was coated by chemical bath deposi‐ tion method. CZTS precursor thin films were coated using sol–gel solution of Cu, Zn, and Sn ions. Sulfurization was employed at 500 ℃ under a mixture atmosphere of H2S and N2 with H2S concentration changing from 3% to 20%. CZTS thin film prepared with a H2S con‐ centration of 3% had grains with size of 1 μm. The best solar cell, which was obtained from a sample sulfurized at a H2S concentration of 3%, demonstrated a conversion efficiency of 2.23%. This is so far the highest efficiency for sol-gel method deposited CZTS solar cells.

**Figure 14.** a) SEM surface image of CZTS thin film, (b) typical J-V curve of CZTS TFSC deposited by sol-gel method [27].

We have deposited CZTS thin films by employing sol-gel method [26]. The deposited CZTS thin film consisted of large densely packed grains with size of more than 400 nm (Fig. 14(a)) and had an optical bandgap of 1.51 eV. The compositional ratios could be optimized through modification of the composition of CZTS sol-gel precursor and annealing processes. An efficiency of 0.63% has been demonstrated (Fig. 14(b)). The relatively low efficiency re‐ sulted from the low *V OC* and *FF* which was caused by the cracks and penetrating pores gen‐ erated during the annealing process. With this ongoing project, we have already improved FF by 19% using ~ 450 nm thick CZTS film by controlling the deposition process, leading to significant improvement of morphology (Fig. 15). A multiple layer deposition process suc‐

cessfully blocked the penetrating pores.

126 Solar Cells - Research and Application Perspectives

Hot-injection method is usually employed to synthesize CZTS nanoparticles [88, 89]. In a typical synthesis, copper salt, zinc salt, and tin salt are dissolved in oleylamine. The mixture solution is heated to 130 ℃ under inert atmosphere. The temperature is then raised to 225 ℃ where mixture solution of sulfur and oleylamine is injected. The mixture is then cooled to 80 ℃. Organic solvents such as toluene and isopropanol are added into the reaction mixture where CZTS nanoparticles are collected using centrifuge. Steinhagen et al. fabricated a CZTS PV device by dispersing CZTS nanoparticles in toluene (20 mg/mL) and spray coating the CZTS layer on CdS/ZnO-coated indium tin oxide (ITO) glass [71]. A typical CZTS solar cell showed an open-circuit voltage of 321 mV, a short-circuit current density of 1.95 mA/cm2 , a fill factor of 37%, and a conversion efficiency of 0.23%.

High quality CZTS nanocrystals with well controlled size, shape, and chemical composition have been successfully synthesized [90-93]. Nevertheless, to what extent these properties will affect CZTS TFSCs has rarely been addressed and needs to be further explored. Anneal‐ ing procedure has to be applied to CZTS nanocrystals to grow polycrystalline CZTS thin films. Cracking and material loss encountered in other methods must be prevented to ach‐ ieve CZTS polycrystalline thin film with high semiconductor quality. For more details, we recommend the recently published review paper by Lin et al. about nanoparticle-based method for CZTS TFSCs [94].

## *3.3.4.4. Screen-printing*

Zhou et al. produced CZTS ink by dispersing CZTS microparticles in mixture solution of isopropanol and ethyl cellulose [72]. CZTS thin film with thickness of about 3 μm was screen printed on Mo-coated glass substrate and then was dried naturally. Organic materials were removed using a hot roll at 195 ℃. I-V measurement of a typical CZTS solar cell showed the feasibility of this method to be used to make CZTS solar cell (*J SC*=4.76 mA/cm2 , *V OC*=386 mV, *FF*=0.27, *η*=0.49%). The authors speculated that the comparatively low efficien‐ cy could attribute to internal deficiencies in the screen-printed CZTS solar cells, such as high contact resistance between CZTS paste and Mo conductive layer and small amount of resid‐ ual oxide in the precipitate.

Screen-printing technology has been successfully applied on Si wafer-based PV technolo‐ gies, which partly contributed to rapid decrease of the price of Si wafer-based PV modules. However, screen-printing process is widely employed to prepare the front and back metal contacts. The p-n junction formed using screen-printing method has yet to be proved to be successful due to non-uniformity generated in the screen-printed film and carbon-based sol‐ vents employed in the preparation of paste. Considering the high sensitivity of CZTS thin film to composition variation, it will be more difficult to achieve desirable CZTS absorber layer using screen-printing technology.

The demonstrated performance of CZTS TFSCs by these methods is lower than those by vacuum-based method. However, non-vacuum-based techniques such as nanoparticlebased method and sol-gel method are promising because of simplicity and versatility associ‐ ated with these methods.

## **4. Prospects**

The efficiency of CZTS solar cell has been significantly improved since 2000. Due to knowhow gained from the research on CIGS solar module, an efficiency of 6.21% has been realiz‐ ed for CZTS solar module with an aperture area of 22.6 cm2 [50]. Collaboration to market CZTS PV technology has been laid ground among semiconductor industry and photovoltaic industry giants [95, 96]. However, a number of technical issues must be addressed and cor‐ responding solutions are provided before CZTS PV technology becomes marketable.

### **4.1. Defect engineering**

### *4.1.1. Defect control*

Defect states in quaternary compounds such as CIGS and CZTS thin films are very compli‐ cated. As introduced above, vacancies such as VCu, VZn, VSn, and VS, antiesite defects such as CuZn, ZnCu, CuSn, SnCu, ZnSn, and SnZn, intrinsic defects such as Cui , Zni , and Sni are possible to form during deposition of CZTS thin film. The formation energy of acceptor defects was lower than that of donor defects, which makes p-type self-doping comparably easy in CZTS. The commonly observed p-type conductivity of CZTS thin films mainly arises from the CuZn antisite defect. Successfully fabricated CZTS solar cells are usually Cu-poor and Zinc-rich.

On the one hand, the conductivity of CZTS thin film derived from intrinsic defects helps to minimize extrinsic defects which have a high density in highly phosphor-doped Si to form emitter which is usually called as dead region because most of photon-excited free carriers recombine at defect states. On the other hand, the electrical properties of CZTS thin film are extremely difficult to be precisely controlled. As far as composition concerned, the transition region from being highly efficient to being dead is very narrow as is CIGS thin film solar cell. Defect engineering such as Na incorporation and Sb doping have been successfully em‐ ployed for CIGS thin film to extend the region and henceforth improve the efficiency. Simi‐ lar experiments could benefit CZTS solar cell as well because CZTS has a lot in common with CIGS. Most researchers have been dedicating their efforts in developing novel deposi‐ tion methods for CZTS thin film. The research focus will soon be turned to defect engineer‐ ing once the newly developed deposition methods gain maturity.

### *4.1.2. Pure CZTS phase generation and secondary phase detection*

*3.3.4.4. Screen-printing*

128 Solar Cells - Research and Application Perspectives

ual oxide in the precipitate.

ated with these methods.

**4.1. Defect engineering**

*4.1.1. Defect control*

**4. Prospects**

layer using screen-printing technology.

Zhou et al. produced CZTS ink by dispersing CZTS microparticles in mixture solution of isopropanol and ethyl cellulose [72]. CZTS thin film with thickness of about 3 μm was screen printed on Mo-coated glass substrate and then was dried naturally. Organic materials were removed using a hot roll at 195 ℃. I-V measurement of a typical CZTS solar cell showed the feasibility of this method to be used to make CZTS solar cell (*J SC*=4.76 mA/cm2

*V OC*=386 mV, *FF*=0.27, *η*=0.49%). The authors speculated that the comparatively low efficien‐ cy could attribute to internal deficiencies in the screen-printed CZTS solar cells, such as high contact resistance between CZTS paste and Mo conductive layer and small amount of resid‐

Screen-printing technology has been successfully applied on Si wafer-based PV technolo‐ gies, which partly contributed to rapid decrease of the price of Si wafer-based PV modules. However, screen-printing process is widely employed to prepare the front and back metal contacts. The p-n junction formed using screen-printing method has yet to be proved to be successful due to non-uniformity generated in the screen-printed film and carbon-based sol‐ vents employed in the preparation of paste. Considering the high sensitivity of CZTS thin film to composition variation, it will be more difficult to achieve desirable CZTS absorber

The demonstrated performance of CZTS TFSCs by these methods is lower than those by vacuum-based method. However, non-vacuum-based techniques such as nanoparticlebased method and sol-gel method are promising because of simplicity and versatility associ‐

The efficiency of CZTS solar cell has been significantly improved since 2000. Due to knowhow gained from the research on CIGS solar module, an efficiency of 6.21% has been realiz‐ ed for CZTS solar module with an aperture area of 22.6 cm2 [50]. Collaboration to market CZTS PV technology has been laid ground among semiconductor industry and photovoltaic industry giants [95, 96]. However, a number of technical issues must be addressed and cor‐

Defect states in quaternary compounds such as CIGS and CZTS thin films are very compli‐ cated. As introduced above, vacancies such as VCu, VZn, VSn, and VS, antiesite defects such as

to form during deposition of CZTS thin film. The formation energy of acceptor defects was lower than that of donor defects, which makes p-type self-doping comparably easy in CZTS. The commonly observed p-type conductivity of CZTS thin films mainly arises from the CuZn antisite defect. Successfully fabricated CZTS solar cells are usually Cu-poor and Zinc-rich.

, Zni

, and Sni

are possible

responding solutions are provided before CZTS PV technology becomes marketable.

CuZn, ZnCu, CuSn, SnCu, ZnSn, and SnZn, intrinsic defects such as Cui

,

The investigation on the phase equilibrium in the Cu2S-ZnS-SnS2 system showed that singlephase CZTS crystals can only be grown in a very narrow region (Fig. 16) [97]. To form pure CZTS phase is a challenge. Secondary phases such as ternary and quaternary compounds are much easier to form than CZTS. Therefore, it is quite challenging to deposit CZTS thin film without significant presence of secondary phases. Time-resolved XRD measurements clearly illustrated that the crystallization of kesterite CZTS was completed by the solid state reaction of Cu2SnS3 and ZnS whatever the precursor was [85]. The formation of binary and ternary secondary phases including ZnxS, CuxS, SnxS, and CuxSnSy are often observed dur‐ ing and after the growth of CZTS crystals.

Highly performed CZTS solar cells are slightly Zn-rich and Cu-poor. However, secondary phases such as ZnS and Cu2SnS3 are readily to be formed during thin-film growth in a Znrich regime. Inhomogeneity due to the presence of these secondary phases was assumed to contribute to the comparatively low efficiencies. Detection of secondary phases will guide how to improve the growth method for CZTS thin film. Nevertheless, it is commonly recog‐ nized that detecting secondary phases using only XRD in CZTS is not as easy as in CIGS because kesterite CZTS shares multiple peaks with cubic ZnS and Cu2SnS3 (Fig. 17). Raman spectroscopy is often combined with XRD results to characterize CZTS thin films [30, 32].

Hartman et al. developed a technique defined as extended X-ray absorption fine structure (EXAFS) which is sensitive to local chemical environment and able to determine the quanti‐ ty of ZnS phase in CZTS films by detecting differences in the second-nearest neighbor shell of the Zn atoms [98]. The results so far are promising. Significant differences in EXAFS spec‐ tra with varying amounts of Zn and Cu in the CZTS films have been observed. Further work is required to quantify the amount of secondary ZnS phase in EXAFS spectra and to enable EXAFS technique to be confidently employed in characterizing CZTS thin film.

**Figure 16.** Phase diagram of SnS2-Cu2S-ZnS system [97].

**Figure 17.** Comparison of XRD peaks of CZTS, ZnS, and Cu2SnS3.

#### **4.2. Bandgap engineering**

Bandgap tuning has been widely and successfully employed in fabrication of CIGS TFSCs. Techniques such as substitution of In by Ga and replacement of Se by S can precisely control the bandgap of CIGS thin film [99-102]. The performance of CIGS TFSCs was significantly improved not only because gradient bandgap was introduced into the absorber layer but al‐ so because the conduction band offset (CBO) between buffer layer and CIGS absorber layer was optimized through tuning the conduction band of CIGS thin film [103,105]. This pro‐ vides CZTS TFSC researchers with another powerful technology to improve the efficiency of CZTS TFSC.

Theoretical calculation and experimental results demonstrated that the bandgap of CZTS nanocrystal could also be controlled through incorporation of Se [105]. Zhang et al. analyzed experimentally and theoretically the effects of Se incorporation on the bandgap of CZTS nanocrystal. It was found that band gap of CZTS nanocrystal demonstrated a parabolic na‐ ture. The bandgap firstly decreased with the increase of Se/(S + Se) ratio and then increased with the increase of Se/(S + Se) ratio when the ratio was higher than 0.55 (Fig 18). The varia‐ tion range of optical band gap for CZTS nanocrystal is from 1.28 eV to 1.5 eV.

**Figure 16.** Phase diagram of SnS2-Cu2S-ZnS system [97].

130 Solar Cells - Research and Application Perspectives

**Figure 17.** Comparison of XRD peaks of CZTS, ZnS, and Cu2SnS3.

Bandgap tuning has been widely and successfully employed in fabrication of CIGS TFSCs. Techniques such as substitution of In by Ga and replacement of Se by S can precisely control the bandgap of CIGS thin film [99-102]. The performance of CIGS TFSCs was significantly improved not only because gradient bandgap was introduced into the absorber layer but al‐ so because the conduction band offset (CBO) between buffer layer and CIGS absorber layer was optimized through tuning the conduction band of CIGS thin film [103,105]. This pro‐

**4.2. Bandgap engineering**

**Figure 18.** The dependence of bandgap of Cu2ZnSnS4xSe4(1-x) nanocrystals on the ratio of Se/(Se+S): (a) theoretical re‐ sults, (b) experimental results [105].

The control of bandgap of CZTS nanocrystal was also realized by Agrawal et al. [106]. GeCl4 was added into the reaction solution to partly replace tin (IV) acetylacetonate dichloride. TEM images and XRD data indicated that the Cu2ZnSn1-xGexS4 (CZTGS) nanocrystals vary‐ ing in size from 5 to 30 nm were successfully produced. UV-Vis results for CZTGS nanocrys‐ tals indicated that the bandgap of CZTGS nanocrystals increased with the increase of Ge/(Sn +Ge) ratio (Fig. 19). CZTGS TFSCs with comparatively high efficiency have been fabricated after selenization. The highest efficiency for CZTGS TFSCs was achieved from the cell with a Ge/(Sn+Ge) ratio of 6.8% (Table 6). This efficiency is slightly lower but comparable to that of CZTGS TFSC without Ge incorporation.

**Figure 19.** Absorbance of CZTGS nanocrystals with different ratio of Ge/(Ge+Sn) [106].


**Table 6.** J-V properties of CZTGS TFSC with different ratio of Ge/(Ge+Sn) [106].

The widening effect of bandgap through Ge incorporation and the narrowing effect of bandgap through Se incorporation facilitate the design of high efficiency CZTS TFSC based on multi-junction (Fig. 20). Interfacial recombination caused by crystal mismatch will be minimized due to the high similarity of crystal structures among these materials.

### **4.3. Toxic element-free**

Toxic chemicals are widely and heavily consumed in the manufacturing process of PV in‐ dustry. The environmental damage could be minimized if the wastes were well treated be‐ cause they are usually confined in a certain area. Nevertheless, toxic elements contained in PV modules have potential to polluting the earth because a large amount of PV modules are required to be installed in the desert and on residential roofs to power households. Lives could be lost due to the emission of toxins during fire. Toxic element-free PV technologies will be more preferred when solar electricity is selected to be main power source.

**Figure 20.** Proposed multi-junction CZTS solar cell.

### *4.3.1. Selenium-free*

after selenization. The highest efficiency for CZTGS TFSCs was achieved from the cell with a Ge/(Sn+Ge) ratio of 6.8% (Table 6). This efficiency is slightly lower but comparable to that of

Ge/(Ge+Sn)=0 Ge/(Ge+Sn)=0.7 Ge/(Ge+Sn)=1.0

The widening effect of bandgap through Ge incorporation and the narrowing effect of bandgap through Se incorporation facilitate the design of high efficiency CZTS TFSC based on multi-junction (Fig. 20). Interfacial recombination caused by crystal mismatch will be

Toxic chemicals are widely and heavily consumed in the manufacturing process of PV in‐ dustry. The environmental damage could be minimized if the wastes were well treated be‐ cause they are usually confined in a certain area. Nevertheless, toxic elements contained in PV modules have potential to polluting the earth because a large amount of PV modules are required to be installed in the desert and on residential roofs to power households. Lives

minimized due to the high similarity of crystal structures among these materials.

**Figure 19.** Absorbance of CZTGS nanocrystals with different ratio of Ge/(Ge+Sn) [106].

**Table 6.** J-V properties of CZTGS TFSC with different ratio of Ge/(Ge+Sn) [106].

**4.3. Toxic element-free**

*J*SC (mA/cm2) 31.2 21.5 4.7 *V*OC (mV) 430 640 320 *FF* (%) 54 49 33.7 *R*S (Ω) 4.9 9.1 30.5 *R*Sh (Ω) 850 460 269 η (%) 7.2 6.8 0.51

CZTGS TFSC without Ge incorporation.

132 Solar Cells - Research and Application Perspectives

Selenium compounds such as hydrogen selenide are extremely toxic while selenium itself is not highly toxic. Selenization and sulfurization are often employed to CZTS precursors to grow high quality CZTS thin films. The efficiency of CZTS solar cell is lower than that of CZTSSe solar cell which is incorporated with selenization (Table 7) [36]. The authors con‐ cluded that the comparatively low efficiency associated with CZTS and CZTSSe solar cells were caused by the extremely low lifetime because a typical lifetime of a high quality CIGS device is beyond 50 ns. The differences of *J SC* and *V OC* were mainly due to the difference of optical bandgap. The difference between the efficiencies of CZTS and CZTSSe PV devices was probably from the difference of bandgap which attributed to unfavourable band align‐ ments with the CdS emitter layer, leading to higher series resistance.


**Table 7.** Comparison of properties of CZTS and CZTSSe TFSCs [36].

For solution-based CZTS solar cells, selenium incorporation plays an even greater role in improving the efficiency of CZTS solar cell. Firstly, Se has higher reactivity than S. Metals in CZTS precursors will more readily react with Se to produce metal selenides and consequent‐ ly to grow CZTSSe thin films. Phase separation in sulfurized CZTS thin film is more com‐ mon than that in selenized CZTS thin film. Secondly, CZTS thin film is usually deposited at so high a temperature that cracks are highly possible to be generated due to volume contrac‐ tion caused by evaporation of metal sulfides and excess sulphur [107]. Displacement of S with Se during selenization can prevent crack forming because the aomic radius of selenium is larger than that of sulfur

However, knowledge achieved from fabrication of CIGS solar cell can shed light on how to make selenium-free CZTS solar cell with high efficiency. It was reported that CIGS solar cell showed higher efficiency when deposited under heat-cracked selenium vapor because cracked selenium vapor is more reactive than selenium vapor without cracking treatment which mainly consists of Se8 [108]. Cracking treatment for sulfur vapor has been first em‐ ployed to fabricate CZTS solar cell by IBM [36]. Treatment details were not available. It would be a feasible method to improve the quality of CZTS thin film in that sulfur vapor is mainly composed of S8 which has lower reactivity than S as does Se8 than Se. Moreover, CIGS solar cell with high efficiency has been successfully fabricated by ISET who synthe‐ sized copper oxide, indium oxide, and gallium oxide nanoparticles as precursors [109]. Simi‐ lar technology can be developed to deposit CZTS thin film. Synthesis of nanoparticles of copper oxide, zinc oxide, and tin oxide has been widely reported in literatures. Volume con‐ traction will be minimized if sulfurization is employed to CZTS precursors containing these metal oxide nanoparticles because the atomic radius of S is larger than that of O as is Se larg‐ er than S. It was found that zinc oxide was easily turned into zinc sulfide after being an‐ nealed in a mixture atmosphere of hydrogen sulfide, hydrogen, and nitrogen [110]. Similar methods have yet proved to be possible to produce copper sulfide and tin sulfide. It would be a novel technology to fabricate selenium-free CZTS solar cell with high efficiency once the methods are available.

### *4.3.2. Cadmium-free*

CdS thin film is commonly incorporated in CZTS PV device as buffer layer to form p-n junc‐ tion with p-type CZTS absorber layer. However, the environmental risk brought by the im‐ plementation of CdS is nontrivial when CZTS PV technology is widely employed, not to mention the marketing problem caused by legal regulations of Cd in electrical or electronic equipment in different countries [111, 112]. Free carriers excited by photons with energy ranging from 2.3 eV to 3.6 eV are lost in CdS thin film. The elimination or replacement of CdS thin film has potential of increasing photocurrent generated in this energy region, and therefore improving the cell efficiency.

So far, Cd-free CZTS solar cells have been reported by just a few groups [50, 70, 113]. The buffer layers employed and *J-V* properties are summarized in Table 8. Solar Frontier has successfully fabricated a Cd-free CZTS sub-module [50]. The Zn-based buffer layer was de‐ posited by chemical bath deposition on CZTS thin film coated by sulfurization of an evapo‐ rated stacking precursor. It was interesting that the efficiency of the Zn-based buffer submodule was higher than that of the Cd-based buffer sub-module even if the Zn-based buffer cell fabricated with same batch of CZTS had lower efficiency than the Cd-based buffer cell. The authors attributed the enhancement of efficiency to the higher transparency and im‐ provement of shunt resistance, resulting in higher external quantum efficiency, *EQE*, higher *J SC* and higher *V OC*.


**Table 8.** *J-V* properties of Cd-free CZTS PV devices.

CZTS precursors will more readily react with Se to produce metal selenides and consequent‐ ly to grow CZTSSe thin films. Phase separation in sulfurized CZTS thin film is more com‐ mon than that in selenized CZTS thin film. Secondly, CZTS thin film is usually deposited at so high a temperature that cracks are highly possible to be generated due to volume contrac‐ tion caused by evaporation of metal sulfides and excess sulphur [107]. Displacement of S with Se during selenization can prevent crack forming because the aomic radius of selenium

However, knowledge achieved from fabrication of CIGS solar cell can shed light on how to make selenium-free CZTS solar cell with high efficiency. It was reported that CIGS solar cell showed higher efficiency when deposited under heat-cracked selenium vapor because cracked selenium vapor is more reactive than selenium vapor without cracking treatment which mainly consists of Se8 [108]. Cracking treatment for sulfur vapor has been first em‐ ployed to fabricate CZTS solar cell by IBM [36]. Treatment details were not available. It would be a feasible method to improve the quality of CZTS thin film in that sulfur vapor is mainly composed of S8 which has lower reactivity than S as does Se8 than Se. Moreover, CIGS solar cell with high efficiency has been successfully fabricated by ISET who synthe‐ sized copper oxide, indium oxide, and gallium oxide nanoparticles as precursors [109]. Simi‐ lar technology can be developed to deposit CZTS thin film. Synthesis of nanoparticles of copper oxide, zinc oxide, and tin oxide has been widely reported in literatures. Volume con‐ traction will be minimized if sulfurization is employed to CZTS precursors containing these metal oxide nanoparticles because the atomic radius of S is larger than that of O as is Se larg‐ er than S. It was found that zinc oxide was easily turned into zinc sulfide after being an‐ nealed in a mixture atmosphere of hydrogen sulfide, hydrogen, and nitrogen [110]. Similar methods have yet proved to be possible to produce copper sulfide and tin sulfide. It would be a novel technology to fabricate selenium-free CZTS solar cell with high efficiency once

CdS thin film is commonly incorporated in CZTS PV device as buffer layer to form p-n junc‐ tion with p-type CZTS absorber layer. However, the environmental risk brought by the im‐ plementation of CdS is nontrivial when CZTS PV technology is widely employed, not to mention the marketing problem caused by legal regulations of Cd in electrical or electronic equipment in different countries [111, 112]. Free carriers excited by photons with energy ranging from 2.3 eV to 3.6 eV are lost in CdS thin film. The elimination or replacement of CdS thin film has potential of increasing photocurrent generated in this energy region, and

So far, Cd-free CZTS solar cells have been reported by just a few groups [50, 70, 113]. The buffer layers employed and *J-V* properties are summarized in Table 8. Solar Frontier has successfully fabricated a Cd-free CZTS sub-module [50]. The Zn-based buffer layer was de‐ posited by chemical bath deposition on CZTS thin film coated by sulfurization of an evapo‐ rated stacking precursor. It was interesting that the efficiency of the Zn-based buffer submodule was higher than that of the Cd-based buffer sub-module even if the Zn-based buffer

is larger than that of sulfur

134 Solar Cells - Research and Application Perspectives

the methods are available.

therefore improving the cell efficiency.

*4.3.2. Cadmium-free*

Excitingly, our tests demonstrated that the efficiency of Cd-free CIGS solar cell had been sig‐ nificantly improved to be over that of Cd-based CIGS solar cell through modifying the com‐ position of CIGS absorber (unpublished data). Similar results (Table 9) have also been observed for CZTS solar cell [50]. The efficiency decreased with the decrease of the ration of Zn to Sn for Cd-based CZTS solar cell. However, the contrary was true for Cd-free CZTS solar cell. The champion efficiency of 5.82% for Cd-free CZTS solar cell was obtained at a Zn/Sn ratio of 1.02. Efforts have to be focused on not only the successful deposition of Cdfree buffer layer but also the optimization of the whole PV device [114].


**Table 9.** Efficiencies of Cd-based and Cd-free CZTS PV devices depend on the ratio of Zn to Sn [50]

### **4.4. Nanostructured CZTS solar cell**

Nanostructured PV devices have gained tremendous interest since the advent of dye-sensi‐ tized solar cell [115-117]. Organic semiconductors have absorption coefficients as high as 105 cm-1 but exciton diffusion length as low as tens of nanometers. Nanostructure provides exci‐ tons with high possibility to be dissociated before they recombine [118]. Furthermore, effi‐ cient light trapping associated with nanostructure reduces the amount of deposited absorber materials [119]. Also, nanostructures are compatible with printing PV technology based on nanoparticles, leading to reduction of processing costs and energy pay-back time of solar cells.

Nanostructured CZTS solar cell has not been reported although ZnO nanorod coated on ITO has been successfully applied to CIGS solar cells [120]. The efficiency was improved for nanostructured CIGS solar cell compared to planar one. However, the efficiencies are much lower than those of conventional CIGS solar cells fabricated on Mo-coated glass. It was as‐ sumed that organic materials contained in the CIGS precursors sprayed on ZnO nanorods could not be totally eliminated, leading to degradation of CIGS thin film. Moreover, the an‐ nealing temperature cannot be as high as that employed in the fabrication of conventional CIGS solar cells because ITO thin film will be damaged under S or Se vapor, leading to poor‐ er crystal qualities of CIGS thin film.

Therefore, we propose a nanostructured CZTS PV device based on Mo nanorods and sol-gel derived CZTS thin film (Fig. 21). Mo thin film and Mo nanorods are sequentially deposited on glass [Fig. 21(a)]. CZTS sol-gel precursor is then spin-coated on Mo nanorods [Fig. 21(b)]. Annealing at high temperature is employed to grow CZTS polycrystalline thin film. Lastly, CdS, i-ZnO, and TCO are sequentially coated on CZTS polycrystalline thin film [Fig. 21(c) and Fig. 21(d)]. The advantages of nanostructure and conventional CZTS will be integrated in a single PV device.

**Figure 21.** Proposed nanostructured CZTS PV device.

### **5. Remarks and conclusions**

The tremendous progresses recently achieved on CZTS have demonstrated the potential of fabricating high-performance and cost-effective PV devices with low environmental pollu‐ tion. Both vacuum-based and non-vacuum-based methods have been successfully explored to fabricate CZTS solar cells. Among vacuum-based methods, evaporation and sputtering are appropriate deposition techniques in terms of efficiencies. Non-vacuum-based techni‐ ques including nanoparticle-based and sol-gel methods are also promising because of sim‐ plicity. To further improve the performance of CZTS solar cells, intensive efforts should be committed to the development of the approaches for forming pure CZTS phase and the de‐ tection techniques of secondary phases formed during the deposition of the CZTS thin film. Also, research efforts need to be focused on cadmium-free and selenium-free CZTS PV tech‐ nologies. Further improvement can be expected in that know-how of CIGS PV technology and related nanotechnology is readily transferred to the research of CZTS PV technology due to great similarity between these two materials. A maturated CZTS PV technology is ex‐ pected for the TFSC family in a near future.

## **Acknowledgements**

We acknowledge the financial support from the NASA through Contract NNX09AU83A.

## **Author details**

nanostructured CIGS solar cell compared to planar one. However, the efficiencies are much lower than those of conventional CIGS solar cells fabricated on Mo-coated glass. It was as‐ sumed that organic materials contained in the CIGS precursors sprayed on ZnO nanorods could not be totally eliminated, leading to degradation of CIGS thin film. Moreover, the an‐ nealing temperature cannot be as high as that employed in the fabrication of conventional CIGS solar cells because ITO thin film will be damaged under S or Se vapor, leading to poor‐

Therefore, we propose a nanostructured CZTS PV device based on Mo nanorods and sol-gel derived CZTS thin film (Fig. 21). Mo thin film and Mo nanorods are sequentially deposited on glass [Fig. 21(a)]. CZTS sol-gel precursor is then spin-coated on Mo nanorods [Fig. 21(b)]. Annealing at high temperature is employed to grow CZTS polycrystalline thin film. Lastly, CdS, i-ZnO, and TCO are sequentially coated on CZTS polycrystalline thin film [Fig. 21(c) and Fig. 21(d)]. The advantages of nanostructure and conventional CZTS will be integrated

The tremendous progresses recently achieved on CZTS have demonstrated the potential of fabricating high-performance and cost-effective PV devices with low environmental pollu‐ tion. Both vacuum-based and non-vacuum-based methods have been successfully explored to fabricate CZTS solar cells. Among vacuum-based methods, evaporation and sputtering are appropriate deposition techniques in terms of efficiencies. Non-vacuum-based techni‐

er crystal qualities of CIGS thin film.

136 Solar Cells - Research and Application Perspectives

**Figure 21.** Proposed nanostructured CZTS PV device.

**5. Remarks and conclusions**

in a single PV device.

Minlin Jiang and Xingzhong Yan

\*Address all correspondence to: xingzhong.yan@sdstate.edu

Department of Electrical Engineering and Computer Science, South Dakota State University, SD, 57007, USA

## **References**


[10] Wadia, C., Alivisatos, A. P., & Kammen, D. *Environ. Sci. Technol. 2009*, 43, 2072-2077.

[16] Wangperawong, A., King, J. S., Herron, S. M., Tran, B. P., Pangan-Okimoto, K., &

[18] Pawar, S. M., Pawar, B. S., Moholkar, A. V., Choi, D. S., Yun, J. H., Moon, J. H., Kole‐

[20] Araki, H., Kubo, Y., Jimbo, K., Maw, W. S., Katagiri, H., Yamazaki, M., Oishi, K., &

[21] Araki, H., Mikaduki, A., Kubo, Y., Sato, T., Jimbo, K., Maw, W. S., Katagiri, H., Ya‐ mazaki, M., Oishi, K., & Takeuchi, A. *Thin Solid Films 2008*, 517, 1457-1460.

[22] Araki, H., Kubo, Y., Mikaduki, A., Jimbo, K., Maw, W. S., & Katagiri, H. *Sol. Energy.*

[23] Katagiri, H., Ishigaki, N., Ishida, T., Saito, K., & Jpn, J. *Appl. Phys. 2001*, 40, 500-504.

[25] Katagiri, H., Saitoh, K., Washio, T., Shinohara, H., Kurumadani, T., & Miyajima, S.

[26] Jiang, M., Dhakal, R., Li, Y., Thapaliya, P., & Yan, X. *Proceedings 37 37th IEEE*

[27] Jiang, M., Li, Y., Dhakal, R., Thapaliya, P., Mastro, M., Caldwell, J. D., Kub, F., & Yan,

[28] Wang, X., Sun, Z., Shao, C., Boye, D. M., & Zhao, J. *Nanotechnology 2011*, 22, 245605.

[29] Fontané, X., Calvo-Barrio, L., Izquierdo-Roca, V., Saucedo, E., Pérez-Rodriguez, A., Morante, J. R., Berg, D. M., Dale, P. J., & Siebentritt, S. *Appl. Phys. Lett. 2011.*, 98,

[31] Wang, K., Shin, B., Reuter, K. B., Todorov, T., & Mitzi, D. B. *Appl. Phys. Lett. 2011.*, 98,

[11] Nische, R., Sargent, D. F., & Wild, P. J. *Cryst. Growth 1967*, 1, 52-53.

[14] Paier, J., Asahi, R., Nagoya, A., & Kresse, G. *Phys. Rev. B. 2009*, 115-126.

[15] Chen, S., Gong, X. G., Walsh, A., & Wei, S. (2009). *Appl. Phys.Lett*, 041-903.

[17] Zhang, X., Shi, X., Ye, W., Ma, C., & Wang, C. *Appl. Phys. A 2009*, 94-381.

[19] Zhou, Y., Zhou, W., Du, Y., Li, M., Wu, S., & Mater, Lett. (20116). 65-1535.

[12] Schäfer, W., & Nitsche, R. *Mat. Res. Bull. 1974*, 9, 645-654.

138 Solar Cells - Research and Application Perspectives

Bent, S. F. *Thin Solid Films 2011*, 519, 2488-2492.

kar, S. S., & Kim, J. H. *Electrochim Acta.*, 55-4057.

Takeuchi, A. *Phys. Status. Solidi 2009*, 6, 1266-1268.

[24] Cao, M., & Shen, Y. J. *Cryst. Growth 2011.*, 318, 1117-1120.

*Sol. Energy. Mater Sol. Cells 2001.*, 65, 141-148.

*Mater Sol. Cells 2009*, 93, 996-999.

*PVSC2011 (Seattle, USA)*.

181905.

051912.

X. J. *Photon Energy 2011.*, 1, 019501.

[30] Yoo, H., & Kim, J. *Thin Solid Films*, 518-6567.

[13] Lu, X., Zhuang, Z., Peng, Q., & Li, Y. *Chem. Comm.*, 47-3141.


[71] Steinhagen, C., Panthani, M. G., Akhavan, V., Goodfellow, B., Koo, B., & Korgel, B. A. J. *Am. Chem. 2009.*, 12554-12555.

[52] Schubert, B., Marsen, B., Cinque, S., Unold, T., Klenk, R., Schorr, S., & Schock, H.

[53] Weber, A., Krauth, H., Perlt, S., Schubert, B., Kötschau, I., Schorr, S., & Schock, H. W.

[54] Katagiri, H., Jimbo, K., Moriya, K., & Tsuchida, K. *Proceedings 3rd World Conference on*

[55] Redinger, A., Berg, D. M., Dale, P. J., & Siebentritt, S. J. *Am. Chem. Soc. 2011.*, 133,

[56] Wang, K., Gunawan, O., Todorov, T., Shin, B., Chey, S. J., Bojarczuk, N. A., Mitzi, D.,

[57] Jimbo, K., Kimura, R., Kamimura, T., Yamada, S., Maw, W. S., Araki, H., Oishi, K., &

[59] Katagiri, H., Jimbo, K., Yamada, S., Kamimura, T., Maw, W. S., Fukano, T., Ito, T., &

[61] Ennaoui, A., Lux-Steiner, M., Weber, A., Abou-Ras, D., Kötschau, I., Schock, H., , W., Schurr, R., Hölzing, A., Jost, S., Hock, R., Vo, T., Schulze, J., & Kirbs, A. *Thin Solid*

[62] Araki, H., Kubo, Y., Jimbo, K., Maw, W. S., Katagiri, H., Yamazaki, M., Oishi, K., &

[63] Araki, H., Kubo, Y., Mikaduki, A., Jimbo, K., Maw, W. S., Katagiri, H., Yamazaki, M.,

[64] Tanaka, K., Fukui, Y., Moritake, N., & Uchiki, H. *Sol. Energy. Mater. Sol. Cells 2011*, 95,

[65] Tanaka, K., Oonuki, M., Moritake, N., & Uchiki, H. *Sol. Energy. Mater Sol. Cells 2009*,

[66] Maeda, K., Tanaka, K., Fukui, Y., & Uchiki, H. *Sol. Energy. Mater Sol. Cells 2011.*, 95,

[67] Moritake, N., Fukui, Y., Oonuki, M., Tanaka, K., & Uchiki, H. *Phys. Status. Solidi 2009*,

[69] Moholkar, A. V., Shinde, S. S., Babar, A. R., Sim, K., Lee, H., Rajpure, K. Y., Patil, P.

[68] Moriya, K., Tanaka, K., & Uchiki, H. *Jpn. J. Appl. Phys. 2007*, 46, 5780-5781.

S., Bhosale, C. H., & Kim, J. H. J. *Alloys Compd. 2011.*, 509, 7439-7446.

[70] Prabhakar, T., & Nagaraju, J. *Proceedings 37th IEEE PVSC 2011 (Seattle, USA)*.

[58] Katagiri, H., & Jimbo, K. *Proceedings 37th IEEE PVSC 2011 (Seattle, USA)*.

[60] Scragg, J. J., Berg, D. M., & Dale, P. J. J. *Electroanal Chem. 2010.*, 646, 52-59.

Oishi, K., & Takeuchi, A. *Sol. Energy. Mater Sol. Cells 2009.*, 996-999.

*Prog. Photovolt. Res. Appl. 2011.*, 19, 93-96.

& Guha, S. *Appl. Phys. Lett. 2010.*, 143508.

*Photovoltaic Energy Conversion 2003.*, 3, 2874-2879.

Katagiri, H. *Thin Solid Films 2007.*, 515, 5997-5999.

Motohiro, T. *Appl. Phys. Express 2008.*, 041201.

Takeuchi, A. *Phys. Status. Solidi 2009*, 96, 1266-1268.

*Thin Solid Films 2009.*, 2524-2526.

140 Solar Cells - Research and Application Perspectives

3320-3323.

*Films*, 517, 2511-2514.

838-842.

93, 583-587.

2855-2860.

1233-1236.


[90] Kameyama, T., Osaki, K. T., Okazaki, Shibayama. T., Kudo, A., Kuwabatade, S., &

[91] Zhou, Y., Zhou, W., Li, M., Du, Y., & Wu, S. J. *Phys. Chem. C. 2011.*, 115, 19632-19639.

[92] Ching, J. Y., Gillorin, A., Zaberca, O., Balocchi, A., & Marie, X. *Chem. Commun. 2011.*,

[95] PV magazine. http://www.pv-magazine.com/news/details/beitrag/solar-frontier-and-

[96] Business insider. http://www.businessinsider.com/delsolar-and-ibm-to-jointly-devel‐

[97] Oleksyuk, I. D., Dudchar, I. V., & Piskach, L. V. J. *Alloys Compd. 2004.*, 368, 135-143.

[99] Lundberg, O., Edoff, M., & Stolt, L. Thin Solid Films 2005. , 480-481, 520-525.

[98] Hartman, K., Newman, B. K., Johnson, J. L., Du, H., Fernandes, P. A., Chawla, V., Bo‐ lin, T., Clemens, B. M., da, Cunha. A. F., Teeter, G., Scarpulla, M. A., & Buonassisi, T.

[100] Sakurai, K., Scheer, R., Nakamura, S., Kimura, Y., Baba, T., Kaufmann, C. A., Neisser, A., Ishizuka, S., Yamada, A., Matsubara, K., Iwata, K., Fons, P., Nakanishi, H., & Ni‐

[101] Jung, S., Ahn, S., Yun, J., Gwak, J., Kim, D., & Yoon, K. *Current Appl. Phys. 2010*, 10,

[104] Minemoto, T., Hashimoto, Y., Kolahi, W., Satoh, T., Negami, T., Takakura, H., & Ha‐

[105] Wei, H., Ye, Z., Li, M., Su, Y., Yang, Z., & Zhang, Y. *Cryst. Eng. Comm. 2011.*, 13, 2222.

[106] Ford, G. M., Guo, Q., Agrawal, R., & Hillhouse, H. W. *Chem. Mater. 2011.*, 23,

[108] Kawamura, M., Fujita, T., Yamada, A., & Konagai, M. J. *Cryst. Growth. 2009.*, 311,

[109] International Solar Electric Technology. http://isetinc.com/technology-overview.php.

[103] Nakada, T., Hongo, M., & Hayashi, E. Thin Solid Films 2003. , 431-432, 242-248.

[93] Shi, L., Pei, C., Xu, Y., & Li, Q. J. *Am. Chem. Soc. 2011.*, 133, 10328-10331.

Torimoto, T. J. *Mater Chem. 2010.*, 20, 5319-5324.

[94] Wang, J., Xin, X., & Lin, Z. Nanoscale, 2011. , 3, 3040.

*Proceedings 37th IEEE PVSC 2011 (Seattle, USA).*

ki, S. *Sol. Energy. Mater Sol. Cells 2006.*, 90, 3377-3384.

[102] Li, W., Sun, Y., Liu, W., & Zhou, L. *Sol. Energy. 2006*, 80, 191-195.

makawa, Y. *Sol. Energy. Mater Sol. Cells 2003.*, 75, 121-126.

[107] Todorov, T., Mitzi, D. B., & Eur, J. *Inorg. Chem. 2010.*, 17, 17-28.

[110] Zhang, R., Wang, B., Wan, D., & Wei, L. *Opt. Mater. 2004.*, 27, 423-419.

ibm-to-develop-czts-cell-100001390/.

op-czts-solar-cell-technology-20109.

47, 5229-5231.

142 Solar Cells - Research and Application Perspectives

990-996.

2626-2629.

753-756.


## **Thin Film Solar Cells Using Earth-Abundant Materials**

Parag S. Vasekar and Tara P. Dhakal

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/51734

## **1. Introduction**

In a p-n junction photovoltaic (PV) cell, a photon of light produces an electron hole pair if the energy of the photon is at least equal to the band gap of the material constituting the p-n junction. The electrons and holes first diffuse toward the respective edge of the depletion re‐ gion, and then drift across the junction due to the built-in potential and are collected at the electrodes. Thus, materials with long minority carrier life times and high carrier mobilities are desired for high efficiency. Because electrons have higher mobility than the holes, a ptype semiconductor is used as light absorber in a p-n junction solar cell.

The theoretical efficiency limit for an ideal homo-junction solar cell as calculated by Loferski [1] is 23%, and this maximum efficiency falls in the vicinity of the absorbers with band gap energy of 1.5 eV. A more justifiable theoretical efficiency limit that used atomic processes was put forward by Shockley and Queisser [2] later in the sixties. According to this theoreti‐ cal efficiency limit, also known as the Shockley-Queisser limit, a maximum efficiency of 30% for a band gap of 1.1 eV is possible (assuming only radiative recombination) if exposed to the sunlight of global air mass 1.5.

The most recent data published in *Progress in Photovoltaics* [3] depict an efficiency of 25% for a crystalline silicon (Si) solar cell measured under the global AM 1.5 spectrum at 25°C. This val‐ ue of efficiency approaches the theoretical value set by Shockley-Queisser limit. However, cou‐ pled with its low absorption cross section and high synthesis and processing cost, this first generation crystalline Si- solar cell doesn't show promise for a low cost, thin film PV device.

For thin film technologies, amorphous silicon-hydrogen alloy (a-Si) solar cells have exhibited efficiencies in the 10-12% range [4] and are fabricated with low cost technology. However, deg‐ radation of the performance over time is a major issue associated with the a-Si solar cells. Re‐ cently, metal induced crystallization of a-Si is used to make thin film polycrystalline Si-solar cell at relatively low temperature in an effort to reduce the processing cost [5], but fabrication

of these devices is combined with high temperature chemical vapor deposition, which makes it incompatible with roll-to-roll processing. Other emerging second generation solar cells are thin film chalcogenides like CIGS (CuInGaSe2) and CdTe, which show terrestrial cell efficiency of 20.3% and 16.5% respectively [6,7]. The scarcity, cost and toxicity associated with In, Ga, and Cd elements present in these cells limit their sustainability in the future.

Thus wide spread applications of solar cells will require dramatic decrease in cost through the use of non-toxic, inexpensive, and earth-abundant materials. The drawbacks in the present PV materials motivate us to look for alternatives. Due to the low absorption cross section, crystalline silicon requires thick layers, which will increase costs. Amorphous silicon (a-Si) has a higher absorption cross section and low processing cost compared to its crystal‐ line counterpart, but it has stability problems. GaAs based solar cells have high efficiency, but the arsenic toxicity and the substrate cost limit their use. CIGS thin film solar cells show promise if the scarce and costly indium and gallium could be replaced by other elements. The quaternary compound semiconductor CZTS (Cu2ZnSnS4) [8] possesses promising char‐ acteristic optical properties; band-gap energy of about 1.5 eV and large absorption coeffi‐ cient in the order of 104 cm−1. The highly efficient CdTe solar cell is promising, but has obstacles such as cadmium toxicity, tellurium scarcity, and cost. While the results of re‐ search and commercialization of crystalline Si, GaAs, CIGS, CdTe, etc. cells are commenda‐ ble, a search for alternative materials is indispensable and necessary to achieve low cost, light weight, and low toxicity arrays.


**Table 1.** Figure of Merits of CZTS, Zn3P2 and FeS2 toward a PV cell compiled from the literature. †Although the exact value of the electron diffusion length of CZTS is not available, the quantum efficiency (QE) spectra suggest a relatively lower carrier diffusion length for this material.

In a recent study [9], a number of promising solar cell materials including CZTS, Zn3P2, and FeS2 etc. were identified as materials with low extraction cost. It is estimated that the cost of material extraction for CZTS, Zn3P2 and FeS2 is around 0.005 cents/W, 0.0007 cents/W and 0.000002 cents/W respectively. More encouraging is the fact that the constituent materials of these absorbers are abundant in the earth's crust. The light absorption cross sections for CZTS, Zn3P2 and FeS2 are all greater than 104 cm-1 (Table 1), thus making thin film devices practical. The band gaps are also in the optimum range for efficient photo-energy conver‐ sion. These three contenders will be discussed mainly in this review with some discussion on other earth-abundant candidates such as tin sulfide (SnS).

## **2. Copper zinc tin sulfide**

### **2.1. Introduction**

of these devices is combined with high temperature chemical vapor deposition, which makes it incompatible with roll-to-roll processing. Other emerging second generation solar cells are thin film chalcogenides like CIGS (CuInGaSe2) and CdTe, which show terrestrial cell efficiency of 20.3% and 16.5% respectively [6,7]. The scarcity, cost and toxicity associated with In, Ga, and

Thus wide spread applications of solar cells will require dramatic decrease in cost through the use of non-toxic, inexpensive, and earth-abundant materials. The drawbacks in the present PV materials motivate us to look for alternatives. Due to the low absorption cross section, crystalline silicon requires thick layers, which will increase costs. Amorphous silicon (a-Si) has a higher absorption cross section and low processing cost compared to its crystal‐ line counterpart, but it has stability problems. GaAs based solar cells have high efficiency, but the arsenic toxicity and the substrate cost limit their use. CIGS thin film solar cells show promise if the scarce and costly indium and gallium could be replaced by other elements. The quaternary compound semiconductor CZTS (Cu2ZnSnS4) [8] possesses promising char‐ acteristic optical properties; band-gap energy of about 1.5 eV and large absorption coeffi‐ cient in the order of 104 cm−1. The highly efficient CdTe solar cell is promising, but has obstacles such as cadmium toxicity, tellurium scarcity, and cost. While the results of re‐ search and commercialization of crystalline Si, GaAs, CIGS, CdTe, etc. cells are commenda‐ ble, a search for alternative materials is indispensable and necessary to achieve low cost,

**Property CZTS Zn3P2 FeS2**

Band gap 1.5 eV [10] 1.5 eV [11] 0.95 eV [12] Absorption Coeff. ˃104 cm-1 [10] ˃104 cm-1 [13] 3.3 x 105 cm-1 [12] Electron diff. length n/a† 4 to 10 µm [11] 0.13 to 1 µm [12] Dark resistivity (ρ) 1 to 39 x103 Ω.cm [14] 2.5 x 103 Ω.cm [15] 1.43 Ω.cm [16]

Carrier mobility (µ) 30 cm2/Vs [10] 450 cm2/Vs [17] 200 to 300 cm2/Vs [12]

**Table 1.** Figure of Merits of CZTS, Zn3P2 and FeS2 toward a PV cell compiled from the literature. †Although the exact value of the electron diffusion length of CZTS is not available, the quantum efficiency (QE) spectra suggest a relatively

In a recent study [9], a number of promising solar cell materials including CZTS, Zn3P2, and FeS2 etc. were identified as materials with low extraction cost. It is estimated that the cost of material extraction for CZTS, Zn3P2 and FeS2 is around 0.005 cents/W, 0.0007 cents/W and 0.000002 cents/W respectively. More encouraging is the fact that the constituent materials of these absorbers are abundant in the earth's crust. The light absorption cross sections for CZTS, Zn3P2 and FeS2 are all greater than 104 cm-1 (Table 1), thus making thin film devices practical. The band gaps are also in the optimum range for efficient photo-energy conver‐ sion. These three contenders will be discussed mainly in this review with some discussion

Cd elements present in these cells limit their sustainability in the future.

light weight, and low toxicity arrays.

146 Solar Cells - Research and Application Perspectives

lower carrier diffusion length for this material.

on other earth-abundant candidates such as tin sulfide (SnS).

Thin film solar cells based on Cu(In,Ga)(S,Se)2 and CdTe have demonstrated significant improvement in the last few years, and they are also being transferred to production lev‐ els [18,19]. Out of these two technologies, CIGS based solar cells are the most efficient ones at the laboratory level and have demonstrated efficiencies in the range of 20% [6]. However, both CIGS and CdTe based thin film solar cells are hindered by potential envi‐ ronmental hazard issues [20] and scarcity issues associated with the constituent elements: mainly Te, In, Ga, and to some extent, Se [20,21]. Recent research trends are moving to‐ wards finding alternatives based on earth-abundant and non-toxic elements. An alterna‐ tive material Cu(Zn,Sn)(S,Se)2 is being explored these days by the thin film photovoltaics community which contains earth-abundant materials like Zn and Sn. CZTS structure can be derived from CuInS2 chalcopyrite structure by replacing one-half of the constituent in‐ dium atoms by zinc and other half by tin. The resulting bandgap varies in the range of 0.8 eV for a selenide structure to 1.5 eV for a sulfide structure [22]. Copper-Zinc-Tin Sul‐ fide or Cu2ZnSnS4 (CZTS) has a nearly ideal direct bandgap (1.5 eV) to absorb the most of the visible solar spectrum as well as a high absorption coefficient (104 cm-1). CZTS film contains neither rare metals nor toxic materials, and combined with the cadmium-free buffer layer, we can expect solar cells with complete non-toxicity.

Several preparation methods have been reported in literature for preparation of CZTS solar cells [23]. These are physical vapor methods such as sulfurization of e-beam evaporated metallic layers and sputtered layers [24,25], RF magnetron sputtering [26], co-evaporation [27], hybrid sputtering [28]. Chemical methods involve sulfurization of electrochemically deposited metallic precursors [29], photochemical deposition [30], sol-gel sulfurization methods [31], sol-gel spin-coated deposition [32] and spray pyrolysis [33,34]. In addition, there are syntheses based on solution methods [35,36].

Ever since Nakazawa reported the photovoltaic effect in a heterodiode of CZTS in 1988 [37], research in CZTS field has come a long way. The current highest efficiency for solutionprocessed CZTSeS based solar cells is 10.1% [38]. The efficiency for pure CZTS based solar cells that use the vacuum based approach is reported to be 8.4% [39]. The Schematic of a CZTSeS solar cell is as shown in Fig.1.

## **2.2. Crystal Structure**

CIGSeS based films are mostly crystallized with a chalcopyrite structure [40], while CZTSeS based films exhibit either a Kesterite type of crystal structure or Stannite structure [41]. Be‐ cause XRD peaks for Kesterite structure are similar to Stannite structure, it is difficult to dis‐ tinguish between them unless other measurement techniques, such as neutron diffraction or Raman spectroscopy, are also employed. All three crystal structures are shown in Fig. 2 in order to understand the similarity and difference between them. It is also quite likely that both Kesterite and Stannite structures co-exist simultaneously because there is not much of energy difference for these structures to achieve/attain a stable state [41].

**Figure 2.** Chalcopyrite, Kesterite and Stannite crystal structures.

### **2.3. Electronic properties**

The band-gap is the most important property of a photovoltaic material and near-optimum band-gap value around 1.5 eV is supposed to be ideal for effective photon absorption and photo-generation. CZTSe phase has a band-gap of 0.8 eV, while CZTS phase has a band-gap value of 1.5 eV [22]. The band-gap can be engineered between these two values by adding sulfur to CZTSe phase. The doping behavior in Kesterite is also controlled by intrinsic dop‐ ing similar to their chalcopyrite counterparts. The p-type behavior of CZTSeS phase is con‐ trolled by CuZn or CuSn antisite donor defects, or this may also result due to direct creation of copper vacancy and compensation for ZnCu antisite and the S vacancy [41].

### **2.4. Phase stability**

methods [31], sol-gel spin-coated deposition [32] and spray pyrolysis [33,34]. In addition,

Ever since Nakazawa reported the photovoltaic effect in a heterodiode of CZTS in 1988 [37], research in CZTS field has come a long way. The current highest efficiency for solutionprocessed CZTSeS based solar cells is 10.1% [38]. The efficiency for pure CZTS based solar cells that use the vacuum based approach is reported to be 8.4% [39]. The Schematic of a

CIGSeS based films are mostly crystallized with a chalcopyrite structure [40], while CZTSeS based films exhibit either a Kesterite type of crystal structure or Stannite structure [41]. Be‐ cause XRD peaks for Kesterite structure are similar to Stannite structure, it is difficult to dis‐ tinguish between them unless other measurement techniques, such as neutron diffraction or Raman spectroscopy, are also employed. All three crystal structures are shown in Fig. 2 in order to understand the similarity and difference between them. It is also quite likely that both Kesterite and Stannite structures co-exist simultaneously because there is not much of

The band-gap is the most important property of a photovoltaic material and near-optimum band-gap value around 1.5 eV is supposed to be ideal for effective photon absorption and photo-generation. CZTSe phase has a band-gap of 0.8 eV, while CZTS phase has a band-gap value of 1.5 eV [22]. The band-gap can be engineered between these two values by adding sulfur to CZTSe phase. The doping behavior in Kesterite is also controlled by intrinsic dop‐ ing similar to their chalcopyrite counterparts. The p-type behavior of CZTSeS phase is con‐ trolled by CuZn or CuSn antisite donor defects, or this may also result due to direct creation of

copper vacancy and compensation for ZnCu antisite and the S vacancy [41].

energy difference for these structures to achieve/attain a stable state [41].

**Figure 2.** Chalcopyrite, Kesterite and Stannite crystal structures.

**2.3. Electronic properties**

there are syntheses based on solution methods [35,36].

CZTSeS solar cell is as shown in Fig.1.

148 Solar Cells - Research and Application Perspectives

**2.2. Crystal Structure**

Similar to CIGSeS thin films, CZTSeS phase also has a narrow region of phase stability in which the device quality phase can be synthesized without any adverse effect of secondary phases. In fact, the phase stability region is even narrower as compared to chalcopyrites, thus making single phase device quality CZTSeS synthesis more difficult [42]. Solid state CZTS can also be synthesized through reactions between ZnS, Cu2S,and SnS2. As can be seen in the phase diagram in Fig. 3, there is very narrow region where CZTS is stable and secon‐ dary phases are very easy to form. Copper-poor and zinc-rich composition has been found to be ideal so far to obtain efficient devices from CZTS films [8].

**Figure 3.** Ternary phase diagram of the Cu2S-ZnS-SnS2 system. (Reprinted from [42] with permission from Elsevier).

### **2.5. Synthesis techniques**

### *2.5.1. Vacuum based approaches ( sputtering/evaporation)*

The synthesis techniques are both vacuum-based processes and non-vacuum based proc‐ esses. Dr. Katagiri's group has pioneered vacuum-based approaches [8,22,24,43]. Their first report on E-B evaporated precursors followed by sulfurization yielded an efficiency of 0.66% [24]. With modification of fabrication process, the conversion efficiency was gradually increased in their successive reports. It was observed that longer time and tem‐ perature cycles were detrimental for the performance of the CZTS films: hence, by adjust‐ ing the time and temperature cycles, Katagiri et al reported an efficiency of 2.62% in their next report [14]. After overcoming the problem of residual gases in the sulfurization chamber, the efficiency number rose to 5.45% [44].

This group has also contributed towards sputtered CZTS films followed by sulfurization [43] and reported an efficiency of 6.7%. They observed that soaking the CZTS layer on the Mo coated sodalime glass in deionized water leads to significant improvement in the device performance. Recently, they also reported 6.48% efficiency using CZTS compound target and a simple single sputtering step [45].

In 2010, IBM group reported an efficiency of 6.8% by using co-evaporated film of Cu, Zn, Sn and S sources followed by reactive annealing [46]. The best efficiency so far for pure sulfide phase CZTS is 8.4% [39], as noted in the IBM group report for evaporated CZTS films. Also, they recently reported an efficiency for pure selenide phase using co-evaporation is 8.9% [47]. An NREL group reports the best efficiency for pure selenide phase CZTSe as 9.15% [48]. The highest reported efficiency using vacuum based techniques is 9.3% [49], which is for mixed CZTSeS phase using co-sputtering of compound targets followed by annealing at a higher temperature. This report is by an industrial research group ( AQT Solar ).

Some other notable recent reports on vacuum based approaches are as follows:


### *2.5.2. Electrodeposition*

Electro-deposition is of particular interest for the thin film photovoltaics community due to the advantages it offers: low cost, environment friendly, large area deposition, room temper‐ ature growth, and less or almost no wastage of materials. CIGS technology based on electrodeposition has already been commercialized [55]. There are numerous reports on electrodeposition of CZTS as well [29,56,57]. Most of the reported efficiencies are in the range of 3% to 4% for electro-deposited CZTS films [57-59].

Some other notable recent reports on electro-deposited CZTS films are as follows:


## *2.5.3. Spray pyrolysis*

gradually increased in their successive reports. It was observed that longer time and tem‐ perature cycles were detrimental for the performance of the CZTS films: hence, by adjust‐ ing the time and temperature cycles, Katagiri et al reported an efficiency of 2.62% in their next report [14]. After overcoming the problem of residual gases in the sulfurization

This group has also contributed towards sputtered CZTS films followed by sulfurization [43] and reported an efficiency of 6.7%. They observed that soaking the CZTS layer on the Mo coated sodalime glass in deionized water leads to significant improvement in the device performance. Recently, they also reported 6.48% efficiency using CZTS compound target

In 2010, IBM group reported an efficiency of 6.8% by using co-evaporated film of Cu, Zn, Sn and S sources followed by reactive annealing [46]. The best efficiency so far for pure sulfide phase CZTS is 8.4% [39], as noted in the IBM group report for evaporated CZTS films. Also, they recently reported an efficiency for pure selenide phase using co-evaporation is 8.9% [47]. An NREL group reports the best efficiency for pure selenide phase CZTSe as 9.15% [48]. The highest reported efficiency using vacuum based techniques is 9.3% [49], which is for mixed CZTSeS phase using co-sputtering of compound targets followed by annealing at

**1.** In the quest to obtain Cd-free devices, the researchers at Solar Frontier, Japan recently demonstrated an efficiency of 6.3% for In-based hetero-junction partners and 5.8% for

**2.** The same group in Japan reports > 8% efficiency on CZTS sub-module [51]. Another in‐ teresting and note-worthy contribution here is that the absorber thickness is reported to

**3.** Salomé et al demonstrate that incorporation of H2 is beneficial for preventing Zn loss

**4.** Shin et al mention that stacking of precursors has a role in getting single-phase final

**5.** Chalapathy and co-workers demonstrate an efficiency of 4.59% using ZnSn (60:40at%)

Electro-deposition is of particular interest for the thin film photovoltaics community due to the advantages it offers: low cost, environment friendly, large area deposition, room temper‐ ature growth, and less or almost no wastage of materials. CIGS technology based on electrodeposition has already been commercialized [55]. There are numerous reports on electrodeposition of CZTS as well [29,56,57]. Most of the reported efficiencies are in the range of 3%

a higher temperature. This report is by an industrial research group ( AQT Solar ).

Some other notable recent reports on vacuum based approaches are as follows:

CZTS compound and reducing the secondary phases [53].

chamber, the efficiency number rose to 5.45% [44].

and a simple single sputtering step [45].

150 Solar Cells - Research and Application Perspectives

Zn-based hetero-junction partners [50].

be just 600 nm.

alloy target [54].

*2.5.2. Electrodeposition*

during the sulfurization [52].

to 4% for electro-deposited CZTS films [57-59].

Spray pyrolysis is a versatile as well as a low-cost technique that has been used to deposit semiconductor films. In this process, a thin film is deposited by spraying a solution on a hot surface. The constituents react to form a chemical compound. The chemical reactants are se‐ lected such that the products other than the desired compound are volatile at the tempera‐ ture of deposition [62]. There have been early reports of sprayed CZTS films [33].

Some other notable recent reports on sprayed CZTS films are as follows:


### *2.5.4. Solution based methods*

Direct liquid deposition or 'ink' based approaches are attractive due to their compatibility with high volume manufacturing techniques such as printing. It is a low-cost scalable route for the thin film solar technology. One of the popular approaches is to prepare precursors using sol-gel method, then sulfurize it, and deposit on substrate using spin-coating [65]. The micro-particles of CZTS have also been prepared using ball milling, sintering type of proc‐ esses and later screen printed on flexible substrates [66]. There are various other reports us‐ ing colloidal nanoparticles [67], Photochemical deposition [30] etc.

The researchers at IBM in their recent publication have reported the highest record effi‐ ciency by any process so far for CZTSeS as 10.1% [38]. The J-V curve of this record sam‐ ple and cross-sectin of the CZTSeS film is as shown in Fig. 4. Here metal chalcogenides were dissolved in hydrazine solution and later spin-coated on glass-substrates followed by a heat-treatment at 540°C.

Some other notable recent reports on solution/ink based CZTS films are as follows:


**Figure 4.** J–V characteristics of the record 10.1% sample at AM1.5G simulated illumination along-with CZTSSe film cross-section (Reprinted from[38] with permission from John Wiley and Sons).

### **2.6. Current status**

The laboratory level efficiencies are more than 10% [38] and there are some recent reports of sub-module efficiencies of more than 8% [51]. However, the technology still needs to go a long way before it can be commercialized. The complex phase diagram and material properties impose most of the challenges. Also, the narrow stoichiometry window for a stable CZTS phase requires a robust control on composition. The processing window is even narrower compared to their CIGS counterparts. The volatile nature of some compo‐ nents, such as tin sulfide and zinc, impedes further challenges for traditional vacuumbased processing approaches which require high sulfurization/selenization temperatures after precursor deposition. The secondary phases formed are difficult to detect with con‐ ventional techniques such as XRD due to the overlapping peaks with pure CZTS phase. The interface properties with hetero-junction partners need to be studied well. However, the initial results in last few years look quite promising.

## **3. Zinc phosphide (Zn3P2)**

### **3.1. Introduction**

With rise in the prices and non-abundance of the materials such as indium and gallium cur‐ rent research trends in thin film solar cells have been moving toward development of earthabundant solar cell materials that can be synthesized using low-cost processes. Also, zincbased hetero-junction partners are preferred over toxic cadmium based compounds such as Cadmium sulfide [70]. Zinc phosphide (Zn3P2) is an important optoelectronic material, which also has applications in lithium ion batteries [71]. It is an important semiconductor from the II-V group, and is used for optoelectronic applications [13,72-74]. Zinc phosphide exhibits favorable optoelectronic properties, such as direct bandgap of 1.5 eV, which corre‐ sponds to the optimum solar energy conversion range [74-77]. Zinc phosphide has a large optical absorption coefficient of >104 cm-1, hence it can be positively used as a p-type absorb‐ er [78]. Also, due to its long minority diffusion length of ~10 μm, high current collection effi‐ ciency can be yielded. Zinc, as well as phosphorous, is abundant in the earth's crust. This makes their cost-effective development quite feasible when it comes to large scale produc‐ tion [73,79]. Zinc phosphide is a tetragonal p-type low cost material with lattice constants of a = b =8.097 Å and c=11.45 Å [17] and has all the right characteristics for photo conversion. Zn3P2 is synthesized most of times with a p-type conductivity [80].

## **3.2. Early efforts**

**1.** Using band-gap engineering with Germanium, researchers at Purdue University have achieved an efficiency of 8.4%. They have utilized nanoparticles of CZTSeS on Moly-

**2.** The researchers at DuPont have demonstrated an efficiency of 8.5% [69]. They use bina‐ ry and ternary chalcogenide nanocrystals produced by simple colloidal syntheses.

**Figure 4.** J–V characteristics of the record 10.1% sample at AM1.5G simulated illumination along-with CZTSSe film

The laboratory level efficiencies are more than 10% [38] and there are some recent reports of sub-module efficiencies of more than 8% [51]. However, the technology still needs to go a long way before it can be commercialized. The complex phase diagram and material properties impose most of the challenges. Also, the narrow stoichiometry window for a stable CZTS phase requires a robust control on composition. The processing window is even narrower compared to their CIGS counterparts. The volatile nature of some compo‐ nents, such as tin sulfide and zinc, impedes further challenges for traditional vacuumbased processing approaches which require high sulfurization/selenization temperatures after precursor deposition. The secondary phases formed are difficult to detect with con‐ ventional techniques such as XRD due to the overlapping peaks with pure CZTS phase. The interface properties with hetero-junction partners need to be studied well. However,

With rise in the prices and non-abundance of the materials such as indium and gallium cur‐ rent research trends in thin film solar cells have been moving toward development of earthabundant solar cell materials that can be synthesized using low-cost processes. Also, zincbased hetero-junction partners are preferred over toxic cadmium based compounds such as Cadmium sulfide [70]. Zinc phosphide (Zn3P2) is an important optoelectronic material,

cross-section (Reprinted from[38] with permission from John Wiley and Sons).

the initial results in last few years look quite promising.

**3. Zinc phosphide (Zn3P2)**

**3.1. Introduction**

coated glass substrates [68].

152 Solar Cells - Research and Application Perspectives

**2.6. Current status**

Thin film solar cell devices using Zn3P2 have been fabricated using Schottky contacts, p-n semiconductor hetero-junctions [81] or liquid contacts [82]. Zinc phosphide was explored extensively in the early eighties and nineties [11,78,80,83,84]. With a Schottky diode, an effi‐ ciency as high as 6% was demonstrated [78]. Zinc phosphide homo-junctions have been dif‐ ficult to make. However, even a zinc phosphide hetero-junction solar cell has not been successful with an efficiency range around 2% [11,85,86].

The researchers at the Institute of Energy Conversion were leaders in the early eighties in exploring Zn3P2 [78,80,83,87,88]. They demonstrated the effect of Mg doping and also fabri‐ cated Schottky barrier solar cells [78,88]. They also successfully made hetero-junction solar cells using ZnO as a hetero-junction partner [11].

Another interesting study is by Misiewicz et al [89]. They made various metal contacts (Au, Ag, Sb, Al, Mg) with Zn3P2 and measured current-voltage characteristics. It was observed that Ag and Sb made the best ohmic contacts, while Au, Mg and Al exhibited rectifying properties. Mg was found to be the most useful for making Schottky barriers.

## **3.3. Synthesis technqiues**

Some of the synthesis techniques to grow Zn3P2 are summarized below :

## *3.3.1. Vapor transport*

Catalano et al [90] report synthesis of Zn3P2 by vapor transport using perforated capsule technique. Crystals were grown in silica tubes placed mid-way in a capsule immediately ahead of Zn3P2 charge. When heated at a high temperature (900°C), effusion of zinc and phospherous out of the inner growth capsule resulted in the condensation of Zn3P2 on the walls of outer tube's cooler portion.A similar vapor transport technique has been reported by Wang et al [91] for growing single crystal Zn3P2.

### *3.3.2. Evaporation*

Deiss et al report synthesis of Zn3P2 thin film by direct evaporation of Zn3P2 on glass sub‐ strates in a vacuum chamber at 750°C [92]. The films deposited on a cold substrate were re‐ ported as amorphous, while the films deposited on a heated sunstrate (>300°C) were reported as crystalline. In a similar technique, Murali et al report that films grown with source-substrate distance < 3 cm were crystalline in nature [93].

## *3.3.3. Metal organic chemical vapor deposition (MOCVD)*

Kakishita et al report a low-pressure MOCVD system with dimethylzinc (DMZ) and diluted phosphine (PH3) as reactant gas [72]. Phosphine was cracked at 800°C and Zn3P2 films were grown on ZnSe single crystal substrates at a growth temperature of 290-410°C. Hermann et al report use of diethlzinc (DEZ) as a zinc precursor [15].

## *3.3.4. Electrodeposition*

Soliman et al [94] report electrodepositon of Zn3P2 with SnO2 coated glass as working elec‐ trode and the aqueous solution containing different concentrations of both ZnCl2, ZnSO4 and Na2PO3. The pH was adjusted at 2.5 by adding H2SO4. It was observed that Zn3P2was obtained, and the degree of crystallinity increased with higher Zn/P molar ratio. There is a recent report similar to that of Soliman et al, reporting electrodeposition of Zn3P2 [95].

### *3.3.5. Chemical reflux technique*

Recently our group has successfully synthesized Zn3P2 using chemical reflux technique [96,97]. The zinc-coated glass substrate is placed on the top of a fritted glass, which is sor‐ rounded by trioctylphosphine (TOP) liquid. The entire vessel is heated while vapors of TOP are refluxed back to the bottom of the glass vessel by a condenser connected at the top of the vessel. The glass vessel is heated to 350°C for two to four hours, and then slowly cooled to room temperature. It was observed that zinc phosphide can be successfully synthesized us‐ ing this simple chemical route in both nano-wire and thin film forms depending on the ex‐ posure mechanism of the phosphorous precursor with the zinc-deposited substrates.

### **3.4. Current status**

Recently Kimball et al have studied the effect of magnesium doping on Zn3P2 thin films [98,99]. Our group has recently demonstrated that zinc phosphide can be successfully syn‐ thesized in both nano-wire and thin film forms [96,97]. Efforts are ongoing to improve the photovoltaic properties using various hetero-junction partner options.

## **4. Iron disulfide ( FeS2)**

### **4.1. Introduction**

Pyrite (FeS2) is a non-toxic material and is abundant on earth, and has a potential to be an inexpensive and sustainable alternative for achieving low-cost and high-efficiency solar cells due both to its environmental compatibility and its optimal optical properties for efficient energy conversion. Its energy band gap of 0.95 eV and optical absorption coefficient of the order of 105 cm-1 are in the optimal range for efficient conversion of sunlight into electricity [12,100]. A saturation current of 40 to 50 mA/cm2 for a thickness between 0.2 μm to 1 μm is possible for pyrite [101]. In addition, a carrier mobility of 200 cm2 /Vs and sufficiently high minority carrier lifetimes makes pyrite highly favorable for photovoltaic application [102]. It has been calculated that theoretical conversion efficiency close to 30% is possible because its band gap is quite close to silicon.

Iron pyrite, like other metal chalcogenides, received considerable attention after the photo‐ electrochemical solar cells using FeS2 single crystal electrodes in contact with iodide/tri-io‐ dide (I<sup>−</sup> /I3 <sup>−</sup> ) redox electrolyte exhibited a quantum efficiency close to 100% across a n-FeS2/(I <sup>−</sup> /I3 <sup>−</sup> ) interface [103-105]. However, a solar conversion efficiency of this photochemical cell was only 2.8% when illuminated under AM 1.5. The open circuit voltage (Voc) was 187 mV. This value of Voc is lower than the theoretical value of about 500 mV [106]. The low photopotential is attributed to the strong pinning of the Fermi level caused by surface states [105] as well as bulk defects caused by the sulphur deficiency [107].

It should be noted that numerous iron sulphides exist in nature with unique properties de‐ pending on different iron (Fe) and sulphur (S) stoichiometric ratios and different crystal structures. Thus creating a phase pure iron pyrite is a challenge. Orthorhombic marcasite FeS2 and hexagonal troilite FeS are both common iron sulfur phases, but because they have much smaller band gaps (0.34 eV for marcasite and 0.04 eV for troilite), even trace amounts would significantly diminish the photovoltaic properties of the pyrite [108]. So, examining the preparation of pyrite by using various techniques is important.

### **4.2. Synthesis techniques**

ported as amorphous, while the films deposited on a heated sunstrate (>300°C) were reported as crystalline. In a similar technique, Murali et al report that films grown with

Kakishita et al report a low-pressure MOCVD system with dimethylzinc (DMZ) and diluted phosphine (PH3) as reactant gas [72]. Phosphine was cracked at 800°C and Zn3P2 films were grown on ZnSe single crystal substrates at a growth temperature of 290-410°C. Hermann et

Soliman et al [94] report electrodepositon of Zn3P2 with SnO2 coated glass as working elec‐ trode and the aqueous solution containing different concentrations of both ZnCl2, ZnSO4 and Na2PO3. The pH was adjusted at 2.5 by adding H2SO4. It was observed that Zn3P2was obtained, and the degree of crystallinity increased with higher Zn/P molar ratio. There is a recent report similar to that of Soliman et al, reporting electrodeposition of Zn3P2 [95].

Recently our group has successfully synthesized Zn3P2 using chemical reflux technique [96,97]. The zinc-coated glass substrate is placed on the top of a fritted glass, which is sor‐ rounded by trioctylphosphine (TOP) liquid. The entire vessel is heated while vapors of TOP are refluxed back to the bottom of the glass vessel by a condenser connected at the top of the vessel. The glass vessel is heated to 350°C for two to four hours, and then slowly cooled to room temperature. It was observed that zinc phosphide can be successfully synthesized us‐ ing this simple chemical route in both nano-wire and thin film forms depending on the ex‐

posure mechanism of the phosphorous precursor with the zinc-deposited substrates.

photovoltaic properties using various hetero-junction partner options.

Recently Kimball et al have studied the effect of magnesium doping on Zn3P2 thin films [98,99]. Our group has recently demonstrated that zinc phosphide can be successfully syn‐ thesized in both nano-wire and thin film forms [96,97]. Efforts are ongoing to improve the

Pyrite (FeS2) is a non-toxic material and is abundant on earth, and has a potential to be an inexpensive and sustainable alternative for achieving low-cost and high-efficiency solar cells due both to its environmental compatibility and its optimal optical properties for efficient energy conversion. Its energy band gap of 0.95 eV and optical absorption coefficient of the

source-substrate distance < 3 cm were crystalline in nature [93].

*3.3.3. Metal organic chemical vapor deposition (MOCVD)*

154 Solar Cells - Research and Application Perspectives

al report use of diethlzinc (DEZ) as a zinc precursor [15].

*3.3.4. Electrodeposition*

*3.3.5. Chemical reflux technique*

**3.4. Current status**

**4.1. Introduction**

**4. Iron disulfide ( FeS2)**

Some of the experimental techniques used to grow pyrite are summarized below:

## *4.2.1. Chemical vapor transport (CVT)*

This technique involves high crystal growth with halogens, such as Br2 and Cl2, and poly‐ crystalline FeS2 sealed in an ampoule [109]. This process produced cubic crystals with (111) and (100) faces up to 5 mm edge length. However, the growth process was too long (10 days) and was performed at high temperature (~1000°C). At elevated temperatures, segrega‐ tion of sulphur and iron species is unavoidable.

### *4.2.2. Metal organic chemical vapor depositon (MOCVD)*

Iron source, such as iron penta-carbonyl [Fe (CO)5], and sulphur sources such as sulphur and H2S gas, are used in this technique [102]. The advantage of this technique is that the growth temperature is low (150°C) and the growth process is fast.

### *4.2.3. Spray pyrolysis*

Chemical spray pyrolysis is used to grow FeS2 using thiourea and FeCl2 [110]. Chlorine con‐ tamination, slow growth, non-uniformity, reduced repeatability are some issues associated with this technique.

## *4.2.4. Chemical methods*

A recent synthesis approach to grow a single phase FeS2 nanoparticle is carried out by dif‐ ferent groups using hydrothermal process [108,111]. Wadia and his group show single phase growth of FeS2 nanoparticles. On the other hand, colloidal FeS2 nano-crystal ink by using a hot-injection method has been synthesized by many groups [112-116]. These proc‐ esses are encouraging in that these are low temperature processes and variation in particle sizes allows the manipulation of band gap, which will help to absorb a broader spectrum of sunlight. To translate these nanocrystals into a form of a compact thin film layer is a chal‐ lenge [113,116,117].

## *4.2.5. Sulfurization of iron oxides*

Sulphurization of iron oxides (Fe3O4 or Fe2O3) is predicted by Gibbs free energy phase dia‐ gram to give FeS2 films [118]. Smestad et. al, sulphurized Fe3O4 or Fe2O3 using gaseous sul‐ phur at 350°C. However, they didn't observe a good photovoltaic behavior from the photochemical cells made by using the FeS2 films grown this way. One of the main reasons attributed to the poor photocurrent and voltage is the presence of micro-pinholes, which led to a short circuit in the photogenerated current and effectively shunted the pyrite cell. Re‐ cently, defect-free pyrite thin films were synthesized from iron oxide by using non-toxic and a more controllable organic precursor such as Di-tert butyl disulfide (TBDS) [119].

## *4.2.6. Sulfurization of iron*

It is also claimed by another group that iron oxide route is not necessary to grow pyrite [12]. They used iron film and sulphurized it under nitrogen flux. But their films had a presence of an amorphous phase, which caused an indirect transion at 1.31 eV incompatible with the strong absorption of FeS2

## **4.3. Issues to be addressed for an optimal FeS2 cell**

Low open circuit (Voc) in pyrite solar cell is attributed to sulphur deficiency among other reasons [120]. A correlation between S deficiency and the transport parameters of pyrite was found, indicating that high quality intrinsic material can be prepared when the S/Fe ratio is close to the stoichiometric value of 2 [121]. In addition to the intrinsic material property, a device design would also affect Voc.

## *4.3.1. Influence of cell design on Voc*

The electrical and optical properties of FeS2 are promising for an efficient photovoltaic ac‐ tion, but the conversion efficiency of the FeS2 cell is not impressive: 1% with Schottky type solar cells [101] and 2.8% with photochemical cells. Cells fabricated with Schottky contacts exhibited short circuit currents (Jsc) below 10 mA/cm2 and open circuit voltages (Voc) below 20 mV. However, quantum efficiencies of 90% and above and high photo-cur‐ rent density observed in the latter type of the cell indicate the potential of an efficient pyrite based solar cell. The reason for the low efficiency was due to low open circuit volt‐ age of about 200 mV. The interfacial chemistry of the p-n junction plays a role in dictat‐ ing the open circuit voltage. A solar cell structure with a wide band gap window layer, such as ZnS, may improve the open circuit voltage.

## *4.3.2. Stoichiometry of FeS2*

*4.2.4. Chemical methods*

156 Solar Cells - Research and Application Perspectives

lenge [113,116,117].

*4.2.5. Sulfurization of iron oxides*

*4.2.6. Sulfurization of iron*

strong absorption of FeS2

device design would also affect Voc.

*4.3.1. Influence of cell design on Voc*

**4.3. Issues to be addressed for an optimal FeS2 cell**

A recent synthesis approach to grow a single phase FeS2 nanoparticle is carried out by dif‐ ferent groups using hydrothermal process [108,111]. Wadia and his group show single phase growth of FeS2 nanoparticles. On the other hand, colloidal FeS2 nano-crystal ink by using a hot-injection method has been synthesized by many groups [112-116]. These proc‐ esses are encouraging in that these are low temperature processes and variation in particle sizes allows the manipulation of band gap, which will help to absorb a broader spectrum of sunlight. To translate these nanocrystals into a form of a compact thin film layer is a chal‐

Sulphurization of iron oxides (Fe3O4 or Fe2O3) is predicted by Gibbs free energy phase dia‐ gram to give FeS2 films [118]. Smestad et. al, sulphurized Fe3O4 or Fe2O3 using gaseous sul‐ phur at 350°C. However, they didn't observe a good photovoltaic behavior from the photochemical cells made by using the FeS2 films grown this way. One of the main reasons attributed to the poor photocurrent and voltage is the presence of micro-pinholes, which led to a short circuit in the photogenerated current and effectively shunted the pyrite cell. Re‐ cently, defect-free pyrite thin films were synthesized from iron oxide by using non-toxic and

It is also claimed by another group that iron oxide route is not necessary to grow pyrite [12]. They used iron film and sulphurized it under nitrogen flux. But their films had a presence of an amorphous phase, which caused an indirect transion at 1.31 eV incompatible with the

Low open circuit (Voc) in pyrite solar cell is attributed to sulphur deficiency among other reasons [120]. A correlation between S deficiency and the transport parameters of pyrite was found, indicating that high quality intrinsic material can be prepared when the S/Fe ratio is close to the stoichiometric value of 2 [121]. In addition to the intrinsic material property, a

The electrical and optical properties of FeS2 are promising for an efficient photovoltaic ac‐ tion, but the conversion efficiency of the FeS2 cell is not impressive: 1% with Schottky type solar cells [101] and 2.8% with photochemical cells. Cells fabricated with Schottky contacts exhibited short circuit currents (Jsc) below 10 mA/cm2 and open circuit voltages (Voc) below 20 mV. However, quantum efficiencies of 90% and above and high photo-cur‐ rent density observed in the latter type of the cell indicate the potential of an efficient pyrite based solar cell. The reason for the low efficiency was due to low open circuit volt‐

a more controllable organic precursor such as Di-tert butyl disulfide (TBDS) [119].

The ratio of Fe and S is extremely important to obtain a defect-free FeS2 phase. It is neces‐ sary that the homogeneity range should be within 1%. It was shown in pyrite grown by MOCVD that whenever a stoichiometry is such that the composition of S is less than 2, a phase mixture of pyrite and pyrrhotite (FeS) existed. The band gap of pyrrhotite phase is 0.04 eV, which greatly diminishes the photovoltaic action. Thus reaffirmation of the pyrite stoichiometry is important.

## *4.3.3. Influence of doping on Voc*

A recent work on simulation of p-n type FeS2 homo-junction diffusion limited solar cells showed that efficiencies around 20% are possible [101] by assuming that at low doping den‐ sity Shockley–Read–Hall (SRH) recombination [122,123] limits the carrier life time, or in oth‐ er words, the cell efficiency. Altermatt et. al [101] simulated a case where the carrier life time (t) limits only Voc, e.g. a cell with d = 1 μm and t = 100 ns. The optimally obtainable photocurrent in pyrite under AM 1.5G illumination is about 50 mA/cm2 . The achievable efficiency level then was 18.5% (13.5%) in a highly (lowly) doped cell. There is trade-off between the carrier life time and open circuit voltage. For a high density of carriers, the voltage across the junction increases, which may create a higher Voc provided the recombination of the gen‐ erated carriers is prevented. This is possible with the low carrier density (thus low dopant induced defects) case where the carrier life time is longer. Thus a subtle balance between Voc and excess carrier life time should be sought by carefully choosing the right dopant density. It was observed in Cobalt doped FeS2 that by manipulating the carrier concentration, the grain barrier height can be changed, which in turn will change the photo-voltage [124].

### *4.3.4. Fermi level pinning at the surface*

Low photo-voltage in a pyrite solar cell is dependent upon its surface structure. Ultravio‐ let photoemission spectroscopy (UPS) measurements carried out in Cobalt doped FeS2 showed that the Fermi level at the surface is pinned [125], implying high density of sur‐ face. Thus controlling the dopant density with a high level of precision is critical. In a sit‐ uation where surface defects are unavoidable, application of a resistive thin layer, such as n-type i-ZnO, could be helpful.

### **4.4. Recent progress in FeS2 solar cell device**

Recently Ganta et. al fabricated solid state based FeS2 thin film solar cell using CdS as an ntype layer [116]. The super-strate type cell structure was glass/FTO/CdS/FeS2/Ag/Cr. FeS2 layer was formed by drop casting the FeS2 nanoparticle ink directly onto the CdS layer at room temperature. The efficiency for CdS/FeS2 cell was 0.03%. The low efficiency was due to the low short-circuit current. The open-circuit voltage was high, which is large compared to a previously reported value of 200 mV [106].

Although the circuit current was low at 0.01 mA/cm2 , an open-circuit voltage of 565 mV was observed. Residual organic components present in the nanoparticle-film could be a reason for low short-circuit current. A compact film with the organic residues removed can be ex‐ pected to improve the short-circuit current and the cell efficiency.

**Figure 5.** Dark and light I-Vcurves of the FeS2 super-strate solar cell.

## **5. Other earth-abundant candidates**

In a recent study, a number of materials were screened based on their abundance and nontoxicity, and semiconducting compounds composed of these identified materials were listed [126]. The authors found that even if binary compounds are focused, there are a good num‐ ber of alternatives. Once semiconducting properties such as ideal band-gap and minority carrier diffusion length are considered, the next step is the processing capability. Obtaining the right phase without any defects is crucial. The next step is to identify ideal hetero-juction partner with right band-alignment as well as chemical and mechanical compatibility. Sever‐ al other earth-abundant candidates are being researched these days.

Tin Sulfide (SnS) is metal chalcogenide semiconductor material from the IV-VI group. It has a near optimum direct bandgap of ~ 1.3 eV [127] and also a good absorption co-efficient (105 cm −1) [128]. Elements such as tin and sulfur are abundantly available. Tin sulfide has been synthe‐ sized using a variety of techniques such as sputtering [127], evaporation [129], electrodeposi‐ tion [130] as well as chemical routes [131]. The highest reported efficiency to date is 1.3% [132].

Cu2FeSnS4 (CFTS) with Stannite structure can also be a viable alternative and there are some recent reports of growing CFTS nanocrystals [133,134]. The bandgap of CFTS was found to be ~ 1.28 eV with p-type conductivity [134].

In a recent study, Fernandes et al identified ternary sulfides Cu2SnS3 and Cu3SnS4 as good alternatives [135]. The estimated bandgap values are between 1 to 1.6 eV and ab‐ sorption coefficient 104 cm−1.

Cu2S is another interesting material that was extensively researched in the eighties. Howev‐ er, performance degradation due to copper diffusion led this technology to be abandoned. Cu2S has a bandgap of 1.2 eV [136]. The highest reported efficiency was 9.1% [137]. Recently this material has again received attention and innovative approaches, such as nanocrystals, are being implemented to overcome the issues encountered in the past [138].

## **6. Conclusion**

the low short-circuit current. The open-circuit voltage was high, which is large compared to

observed. Residual organic components present in the nanoparticle-film could be a reason for low short-circuit current. A compact film with the organic residues removed can be ex‐

In a recent study, a number of materials were screened based on their abundance and nontoxicity, and semiconducting compounds composed of these identified materials were listed [126]. The authors found that even if binary compounds are focused, there are a good num‐ ber of alternatives. Once semiconducting properties such as ideal band-gap and minority carrier diffusion length are considered, the next step is the processing capability. Obtaining the right phase without any defects is crucial. The next step is to identify ideal hetero-juction partner with right band-alignment as well as chemical and mechanical compatibility. Sever‐

Tin Sulfide (SnS) is metal chalcogenide semiconductor material from the IV-VI group. It has a near optimum direct bandgap of ~ 1.3 eV [127] and also a good absorption co-efficient (105

−1) [128]. Elements such as tin and sulfur are abundantly available. Tin sulfide has been synthe‐ sized using a variety of techniques such as sputtering [127], evaporation [129], electrodeposi‐ tion [130] as well as chemical routes [131]. The highest reported efficiency to date is 1.3% [132].

Cu2FeSnS4 (CFTS) with Stannite structure can also be a viable alternative and there are some recent reports of growing CFTS nanocrystals [133,134]. The bandgap of CFTS was found to

In a recent study, Fernandes et al identified ternary sulfides Cu2SnS3 and Cu3SnS4 as good alternatives [135]. The estimated bandgap values are between 1 to 1.6 eV and ab‐

, an open-circuit voltage of 565 mV was

cm

a previously reported value of 200 mV [106].

158 Solar Cells - Research and Application Perspectives

Although the circuit current was low at 0.01 mA/cm2

**Figure 5.** Dark and light I-Vcurves of the FeS2 super-strate solar cell.

al other earth-abundant candidates are being researched these days.

**5. Other earth-abundant candidates**

be ~ 1.28 eV with p-type conductivity [134].

sorption coefficient 104 cm−1.

pected to improve the short-circuit current and the cell efficiency.

One of the important p-type absorber CZTSe(S) has reached a laboratory level efficiency of more than 10% and has matured to a stage where it can be seriously considered for commer‐ cialization. Complex phase diagram and narrow stoichilometry window for CZTS remains a challenge; in addition, interface properties with suitable hetero-junction partners need to be understood better. However, despite these challenges, CZTS is the most promising material synthesized using earth-abundant constituents. Other materials discussed in detail in this review are Zn3P2 and FeS2. These two materials also show considerable promise, despite their current low lab-level efficiency values. Apart from these, there are several other pro‐ mosing materials synthesized using earth-abundant constutuents, such as SnS, CFTS and Cu2SnS3, and these potentially can be used in solar cells due to their photovoltaic properties.

## **Author details**

Parag S. Vasekar\* and Tara P. Dhakal

\*Address all correspondence to: psvasekar@yahoo.com, tpdhakal@gmail.com

Center for Autonomous Solar Power, Binghamton University, Binghamton, NY, USA

## **References**


[5] Gordon, I., Carnel, L., Van Gestel, D., Beaucarne, G., & Poortmans, J. (2007). Efficient thin-film polycrystalline-silicon solar cells based on aluminum- induced crystalliza‐

[6] Jackson, P., Hariskos, D., Lotter, E., Paetel, S., Wuerz, R., Menner, R., et al. (2011). New world record efficiency for Cu(In,Ga)Se2 thin-film solar cells beyond 20%. *Prog*

[7] Wu, X. X. (2004). High-efficiency polycrystalline CdTe thin-film solar cells. *Solar En‐*

[8] Katagiri, H., Jimbo, K., Maw, W. S., Oishi, K., Yamazaki, M., Araki, H., et al. (2009). Development of CZTS-based thin film solar cells. *Thin Solid Films*, 517, 2455-60. [9] Wadia, C. C. (2009). Materials Availability Expands the Opportunity for Large-Scale

[10] Liu, F., Li, Y., Zhang, K., Wang, B., Yan, C., Lai, Y., et al. (2010). In situ growth of Cu2ZnSnS4 thin films by reactive magnetron co-sputtering. *Solar Energy Mater Solar*

[11] Nayar, P. P. S. (1981). Zinc phosphide-zinc oxide heterojunction solar cells. *Appl Phys*

[12] Rezig, B., Dahman, H., & Kenzari, M. (1992). Iron pyrite FeS2 for flexible solar cells.

[13] Babu, V. S., Vaya, P. R., & Sobhanadri, J. (1988). Optical absorption and photocon‐ ductivity studies on Zn3P2 thin films grown by hot wall deposition. *Solar Energy Ma‐*

[14] Katagiri, H. H. (2001). Development of thin film solar cell based on Cu2ZnSnS4 thin

[15] Hermann, A. M., Madan, A., Wanlass, M. W., Badri, V., Ahrenkiel, R., Morrison, S., et al. (2004). MOCVD growth and properties of Zn3 P2 and Cd3P2 films for thermal pho‐

[16] Morsli, M. M. (1995). Electrical Properties of a Synthetic Pyrite FeS Non Stoichiomet‐

[17] Sudhakar, S.S. (2008). Influence of cooling rate on the liquid-phase epitaxial growth

[18] Dhere, N. G. (2007). Toward GW/year of CIGS production within the next decade.

[19] Dhere, N. G. (2011). Scale-up issues of CIGS thin film PV modules. *Solar Energy Mater*

[20] Green, M. M. A. (2006). Consolidation of thin-film photovoltaic technology: the com‐

tovoltaic applications. *Solar Energy Mater Solar Cells*, 82, 241-52.

ing decade of opportunity. *Progress in photovoltaics*, 14, 383-92.

tion and thermal CVD. *Prog Photovoltaics Res Appl*, 15, 575-86.

Photovoltaics Deployment. *Environ Sci Technol*, 43, 2072-7.

*Photovoltaics Res Appl*, 19, 894-7.

*ergy*, 77, 803-14.

160 Solar Cells - Research and Application Perspectives

*Cells*, 94, 2431-4.

*terials*, 18, 65-73.

*Renewable Energy*, 2, 125-8.

films. *Solar Energy Mater Solar Cells*, 65, 141-8.

ric Crystal. *Journal de physique.I : JP*, 5, 699-705.

*Solar Energy Materials and Solar Cells*, 91, 1376-82.

of Zn3 P2 . *J Cryst Growth*, 310, 2707-11.

*Solar Cells*, 95, 277-80.

*Lett*, 39, 105.


[52] Salomé, P. M. P., Malaquias, J., Fernandes, P., Ferreira, M., Leitao, J., Cunha, A., et al. (2011). The influence of hydrogen in the incorporation of Zn during the growth of Cu2ZnSnS4 thin films. *Solar Energy Mater Solar Cells*, 95(12), 3482-9.

[37] K. Ito, T.N. (1988). Electrical and optical properties of stannite-type quaternary semi‐

[38] Barkhouse, D. A. R., Gunawan, O., Gokmen, T., Todorov, T., & Mitzi, D. (2012). De‐ vice characteristics of a 10.1% hydrazine-processed Cu2ZnSn(Se,S)4 solar cell Charac‐ teristics of a 10.1% efficient kesterite solar cell. *Progress in photovoltaics*, 20, 6-11. [39] Shin, B., Gunawan, O., Zhu, Y., Bojarczuk, N. A., Chey, S. J., & Guha, S. (2011). Thin film solar cell with 8.4% power conversion efficiency using an earth-abundant

[40] Stanbery, B. B. J. (2002). Copper Indium Selenides and Related Materials for Photo‐ voltaic Devices. *Critical reviews in solid state and materials sciences*, 27, 73-117.

[41] Siebentritt, S. S. (2012). Kesterites-a challenging material for solar cells Kesterites-a

[42] Olekseyuk, I. D. (2004). Phase equilibria in the Cu2S-ZnS-SnS2 system. *J Alloys Com‐*

[43] Katagiri, H. (2008). Enhanced conversion efficiencies of Cu2ZnSnSP4based thin film solar cells by using preferential etching technique. *Applied physics express*, 1.

[44] Katagiri, H., Jimbo, K., Moriya, K., & Tsuchida, K. (2003). Solar cell without environ‐ mental pollution by using CZTS thin film. *Photovoltaic Energy Conversion, Proceedings*

[45] Katagiri, H., & Jimbo, K. (2011). Development of rare metal-free CZTS-based thin film solar cells. *Photovoltaic Specialists Conference (PVSC), 37th IEEE*, 003516-21. [46] Wang, K., Gunawan, O., Todorov, T., Shin, B., Chey, S. J., Bojarczuk, N. A., et al. (2010). Thermally evaporated Cu2ZnSnS4 solar cells. *Appl Phys Lett*, 97, 143508. [47] Shin, B., Zhu, Y., Bojarczuk, N. A., Chey, S. J., & Guha, S. (2012). High efficiency Cu2ZnSnSe4 solar cells with a TiN diffusion barrier on the molybdenum bottom con‐

[48] Repins, I., Beall, C., Vora, N., De Hart, C., Kuciauskas, D., Dippo, P., et al. (2012). Coevaporated Cu2ZnSnSe4 films and devices. *Solar Energy Mater Solar Cells*, 101, 154-9. [49] Chawla, V., & Clemens, B. (2012). Effect of Composition on High Efficiency CZTSSe Devices Fabricated Using Co-sputtering of Compound Targets. *38th IEEE Photovoltaic*

[50] Hiroi, H., Sakai, N., Muraoka, S., Katou, T., & Sugimoto, H. (2012). Development of High Efficiency Cu2ZnSnS4 Submodule with Cd-Free Buffer Layer. *38th IEEE Photo‐*

[51] Sugimoto, H., Hiroi, H., Sakai, N., Muraoka, S., & Katou, T. (2012). Over 8% Efficien‐ cy Cu2ZnSnS4 Submodules with Ultra-Thin Absorber. *38th IEEE Photovoltaic Special‐*

challenging material for solar cells. *Progress in photovoltaics:n,a-n/a*.

conductor thin films. *Japanese journal of Applied Physics*.

Cu2ZnSnS4 absorber. *Prog Photovoltaics Res Appl:n/a,n/a*.

*pounds*, 368, 135-43.

162 Solar Cells - Research and Application Perspectives

*of 3rd World Conference on*, 3, 2874-2879.

*Specialists Conference (PVSC)*.

*ists Conference (PVSC)*.

*voltaic Specialists Conference (PVSC)*.

tact. *38th IEEE Photovoltaic Specialists Conference (PVSC)*.


[82] Bhushan, M., Turner, J. A., & Parkinson, B. A. (1986). Photoelectrochemical Investiga‐ tion of Zn3P2 . *J Electrochem Soc*, 133, 536-9.

[66] Zhou, Z., Wang, Y., Xu, D., & Zhang, Y. (2010). Fabrication of Cu2ZnSnS4 screen

[67] Chory, C., Zutz, F., Witt, F., Borchert, H., & Parisi, J. (2010). Synthesis and characteri‐

[68] Guo, Q., Ford, G. M., Yang, W., Hages, C. J., Hillhouse, H. W., & Agrawal, R. (2012). Enhancing the performance of CZTSSe solar cells with Ge alloying. *Solar Energy Ma‐*

[69] Guo, Q., Cao, Y., Caspar, J., Farneth, W., Ionkin, A., Johnson, L., et al. (2012). A Sim‐ ple Solution-based Route to High-Efficiency CZTSSe Thin-film Solar Cells. *38th IEEE*

[70] Kumar, B., Vasekar, P. S., Pethe, S., Dhere, N. G., & Galymzhan, T. (2009). ZnxCd1-xS as a heterojunction partner for CuIn1-xGaxS2 thin film solar cells. *Thin Solid Films*,

[71] Shen, G., Chen, P., Bando, Y., Golberg, D., & Zhou, C. (2008). Single-Crystalline and Twinned Zn3P2 Nanowires: Synthesis, Characterization, and Electronic Properties.

[72] Kakishita, K., Aihara, K., & Suda, T. (1994). Zn3P2 photovoltaic film growth for Zn3P2/

[73] Misiewicz, J., Bryja, L., Jezierski, K., Szatkowski, J., Mirowska, N., Gumienny, Z., et al. (1994). Zn3P2 new material for optoelectronic devices. *Microelectron J*, 25, 23-28.

[74] Lousa, A., Bertrán, E., Varela, M., & Morenza, J. L. (1985). Deposition of Zn3P2 thin

[75] Sathyamoorthy, R., Sharmila, C., Natarajan, K., & Velumani, S. Influence of anneal‐ ing on structural and optical properties of Zn3P2 thin films. *Mater Charact*, 58, 745-9.

[76] Murali, K. R., & Rao, D. R. (1981). Optical band gap of [alpha]- Zn3P2 . *Thin Solid*

[77] Long, J. (1983). The Growth of Zn3P2 by Metalorganic Chemical Vapor Deposition. *J*

[78] Bhushan, M. M., & Catalano, A. (1981). Polycrystalline Zn3P2 Schottky barrier solar

[79] Kakishita, K. (1994). Zinc phosphide epitaxial growth by photo-MOCVD. *Appl Surf*

[80] Catalano, A. (1980). Defect dominated conductivity in Zn3P2 . *The Journal of physics*

[81] Casey, M. S., Fahrenbruch, A. L., & Bube, R. H. (1987). Properties of zinc‐phosphide

junctions and interfaces. *Journal of Applied Physics*, 61, 2941-6.

printed layers for solar cells. *Solar Energy Mater Solar Cells*, 94, 2042-5.

zation of Cu2ZnSnS4 . *physica status solidi (c)*, 7, 1486-8.

*ter Solar Cells*, 105, 132-6.

164 Solar Cells - Research and Application Perspectives

517, 2295-9.

*Films*, 86, 283-6.

*Sci*, 79, 281.

*Electrochem Soc*, 130, 725-8.

cells. *Appl Phys Lett*, 38, 39-41.

*and chemistry of solids*, 41, 635.

*Photovoltaic Specialists Conference (PVSC)*.

*The Journal of Physical Chemistry C*, 112, 16405-10.

ZnSe solar cell. *Solar Energy Mater Solar Cells*, 35, 333-40.

films by coevaporation. *Solar Energy Materials*, 12, 51-6.


[114] Bi, Y. Y. (2011). Air Stable, Photosensitive, Phase Pure Iron Pyrite Nanocrystal Thin Films for Photovoltaic Application. *Nano letters*, 11, 4953-7.

[98] Kimball, G. M., Müller, A., Lewis, N. S., & Atwater, H. A. (2009). Photoluminescencebased measurements of the energy gap and diffusion length of Zn3P2. *Appl Phys Lett*,

[99] Kimball, G. M., Lewis, N. S., & Atwater, H. A. (2010). Mg doping and alloying in Zn3P2 heterojunction solar cells. *Photovoltaic Specialists Conference (PVSC), 35th IEEE*,

[100] Wilcoxon, J. P., Newcomer, P. P., & Samara, G. A. (1996). Strong quantum confine‐ ment effects in semiconductors: FeS2 nanoclusters. *Solid State Commun*, 98, 581-5. [101] Altermatt, P. P., Kiesewetter, T., Ellmer, K., & Tributsch, H. (2002). Specifying targets of future research in photovoltaic devices containing pyrite (FeS2) by numerical mod‐

[102] Chatzitheodorou, G. G. (1986). Thin photoactive FeS2 (pyrite) films. *Mater Res Bull*,

[103] Ennaoui, A. A. (1985). Photoactive Synthetic Polycrystalline Pyrite (FeS2). *J Electro‐*

[104] Ennaoui, A. (1993). Iron disulfide for solar energy conversion. *Solar Energy Mater So‐*

[105] Ennaoui, A., Fiechter, S., Smestad, G., & Tributsch, H. (1990). Preparation of iron di‐ sulphide and its use for solar energy conversion. *Proceedings of 1st World Renewable*

[106] Dahman, H. H. (1996). Iron pyrite films prepared by sulfur vapor transport. *Thin Sol‐*

[108] Wadia, C. C. (2009). Surfactant-Assisted Hydrothermal Synthesis of Single phase

[109] Kunst, M., & Tributsch, H. (1984). Surface microwave conductivity spectroscopy. A

[110] Smestad, G., Da Silva, A., Tributsch, H., Fiechter, S., Kunst, M., Meziani, N., et al. (1989). Formation of semiconducting iron pyrite by spray pyrolysis. *Solar Energy Ma‐*

[111] Wu, R., Zheng, Y. F., Zhang, X. G., Sun, Y. F., Xu, J. B., & Jian, J. K. (2004). Hydrother‐

[112] Joo, J. J. (2003). Generalized and Facile Synthesis of Semiconducting Metal Sulfide

[113] Lin, Y. Y. Y. (2009). Extended red light harvesting in a poly(P3-hexylthiophene)/iron

mal synthesis and crystal structure of pyrite. *J Cryst Growth*, 266, 523-7.

disulfide nanocrystal hybrid solar cell. *Nanotechnology*, 20, 405207.

[107] Fiechter, S. (1992). The microstructure and stoichiometry of pyrite. *J Mater Res*, 7.

Pyrite FeS Nanocrystals. *Chemistry of materials*, 21, 2568-70.

new experimental tool. *Chemical Physics Letters*, 105, 123-6.

Nanocrystals. *J Am Chem Soc*, 125, 11100-5.

elling. *Solar Energy Mater Solar Cells*, 71, 181-95.

95, 112103.

166 Solar Cells - Research and Application Perspectives

001039-43.

21, 1481-7.

*chem Soc*, 132, 1579.

*lar Cells*, 29, 289.

*Energy Congress*, 458-64.

*id Films*, 280, 56-60.

*terials*, 18, 299-313.


## **Enhancing the Light Harvesting Capacity of the Photoanode Films in Dye-Sensitized Solar Cells**

Xiang-Dong Gao, Xiao-Min Li and Xiao-Yan Gan

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/51633

## **1. Introduction**

[128] Devika, M., Reddy, N., & Gunasekhar, K. (2011). Structural, electrical, and optical properties of as-grown and heat treated ultra-thin SnS films. *Thin Solid Films*, 520(1),

[129] Devika, M., Reddy, N., Ramesh, K., Gunasekhar, K., Gopal, E., & Reddy, K. T. R. (2006). Influence of annealing on physical properties of evaporated SnS films. *Semi‐*

[130] Gunasekaran, M. M. (2007). Photovoltaic cells based on pulsed electrochemically de‐ posited SnS and photochemically deposited CdS and Cd1−xZnxS. *Solar Energy Mater*

[131] Avellaneda, D., Nair, M. T. S., & Nair, P. K. (2009). Photovoltaic structures using

[132] Ramakrishna Reddy, K. T., Koteswara Reddy, N., & Miles, R. W. (2006). Photovoltaic properties of SnS based solar cells. *Solar Energy Mater Solar Cells*, 90, 3041-6.

[133] Li, L., Liu, X., Huang, J., Cao, M., Chen, S., Shen, Y., et al. (2012). Solution-based syn‐ thesis and characterization of Cu2FeSnS4 nanocrystals. *Mater Chem Phys*, 133, 688-91.

[134] Yan, C., Huang, C., Yang, J., Liu, F., Liu, J., Lai, Y., et al. (2012). Synthesis and charac‐ terizations of quaternary Cu2FeSnS4 nanocrystals. *Chem Commun*, 48, 2603-5.

[135] Fernandes, P. P. A. (2010). A study of ternary Cu2SnS3 and Cu3SnS4 thin films pre‐ pared by sulfurizing stacked metal precursors. *Journal of physics.D, Applied physics*, 43,

[136] Liu, G. G., Schulmeyer, T., Brötz, J., Klein, A., & Jaegermann, W. (2003). Interface properties and band alignment of Cu2S/CdS thin film solar cells. *Thin Solid Films*,

[137] Bragagnolo, J. A., Barnett, A., Phillips, J., Hall, R., & Rothwarf, A. (1980). The design and fabrication of thin-film CdS/Cu2S cells of 9.15-percent conversion efficiency. *IEEE*

[138] Wu, Y. Y., Wadia, C., Ma, W., Sadtler, B., & Alivisatos, A. P. (2008). Synthesis and Photovoltaic Application of Copper(I) Sulfide Nanocrystals. *Nano letters*, 8, 2551-5.

chemically deposited tin sulfide thin films. *Thin Solid Films*, 517, 2500-2.

*conductor science and technology*, 21, 1125-31.

628-32.

215403.

431-432, 477-82.

*Trans Electron Devices*, 27, 645-51.

*Solar Cells*, 91, 774-8.

168 Solar Cells - Research and Application Perspectives

Dye-sensitized solar cell (DSC) has been receiving continuous academic and industrial at‐ tention as a potential low-cost, clean, and renewable energy source, since its inception in 1985 (Desilvestro et al., 1988; Regan et al., 1991; Gao et al., 2008; Hagfeldt et al., 2010; Yella et al., 2011). The DSC is the only photovoltaic device that uses molecules to absorb photons and converts them to electric charges without the need of intermolecular transport of elec‐ tronic excitation. It is also the only photovoltaic device that separates two functions of light harvesting and charge-carrier transport, mimicking the photosynthesis found in green leaves. The primary photon-to-electron conversion process in DSC occurs at the oxide/dye/ electrolyte interface, functioning at a molecular and nano scale. There exist many complicat‐ ed optical, electrical, and chemical processes during the light-to-electric conversion process in DSC, including the light absorption of dye molecules (producing electrons and leaving the dyes in their oxidized states), the electron injection from dyes to the conduction band of semiconductor (e.g., TiO2), the percolation of electrons through the mesoporous semicon‐ ductor network toward the bottom electrode, the recombination of photo-excited electrons in the porous electrode with either oxidized dyes or acceptors in the electrolyte, and the re‐ generation of dye molecules by iodides in the electrolyte etc. For an efficient DSC device, while above processes should be kept in a complicated and delicate balance, the high lightharvesting, the rapid electron transport, and the minimum electron-hole recombination are essential. In special, the light-harvesting efficiency of the photoanode is the most important and indispensable factor for the high-efficiency DSC, which is mainly related to the molar extinction coefficient of the sensitizer, the dye-loading capacity of the porous electrode, and the optical path of the incident light in the electrode.

© 2013 Gao et al.; licensee InTech. This is an open access article 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. © 2013 Gao et al.; licensee InTech. This is a paper 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.

At present, the record efficiency of DSC has reached 12.03% (Yella et al., 2011) after the first breakthrough of M. Grätzel in 1991 and through the continuous efforts of the scientists for more than 20 years. With an aim to further improve the conversion efficiency of DSC to 15% or higher, recently, intensive efforts have been directed to enhance the functions of the ma‐ jor components in DSC, focusing on mainly three aspects, i.e., the ordered nanoporous elec‐ trode, the novel organic or quantum-dot sensitizer, and the new redox electrolyte system (Hagfeldt et al., 2010; Tetreault et al., 2012). During this journey toward high-efficiency DSC, the light-harvesting capacity of the photoanode is expected to play a critical and indispensa‐ ble role by spanning the mesoporous electrodes utilizing varied nano/meso materials and the sensitizers with high molar extinction coefficient or quantum dots, and by connecting the optical functions of the photoanode with its electrical response together.

There are mainly two components concerning the light-harvesting property of DSC, the dye and the mesoporous electrode which supports the sensitizer. While the development of the novel sensitizers with higher molar extinction coefficient and better response to near-infra‐ red wavelength has been long studied since the first breakthrough of DSC in 1991 (Nazeer‐ uddin et al., 2001; Grtäzel et al., 2009; Yum et al., 2011), the significance of the light scattering and/or reflection of the mesoporous electrode on the light-harvesting properties is only recognized in recent years. Accompanying the rapid advance of the ordered or hybrid photoanode materials, novel semiconductor nanostructures or new concepts are widely used to construct new type photoanodes, including the plenary optical waveguide nanowire enhancing the multiple internal reflections (Wei et al., 2010), the plasmonic core-shell metalinsulator nanoparticles enhancing the near-infrared absorption (Brown et al., 2011), the du‐ al-function mesoporous TiO2 structures increasing both the dye-loading and the optical scattering effects (Koo et al, 2008; Huang et al., 2010), and so on. In view of the wide cover‐ age of the optical and photoelectrical nanostructures involved in these studies and their rap‐ id progresses, it is great valuable to make a comprehensive and in-depth review on the current status of the light-harvesting capacities of the photoanode in DSC.

The chapter will start with a brief description of the basic concept of the light-harvesting ef‐ ficiency (LHE), and then give a review on five typical branches representing the significant advances in this area, including


The basic principles of these novel nanostructures/ methods enhancing the light-harvesting capacity of DSC, together with their mutual effects on the electrical and photoelectrochemi‐ cal properties of the nanoporous electrode, will be discussed in detail. Based on the in-depth analysis of literature and the authors' experience, a perspective will be presented, shedding a light on the research road in near future.

## **2. The Light Harvesting Efficiency (LHE)**

At present, the record efficiency of DSC has reached 12.03% (Yella et al., 2011) after the first breakthrough of M. Grätzel in 1991 and through the continuous efforts of the scientists for more than 20 years. With an aim to further improve the conversion efficiency of DSC to 15% or higher, recently, intensive efforts have been directed to enhance the functions of the ma‐ jor components in DSC, focusing on mainly three aspects, i.e., the ordered nanoporous elec‐ trode, the novel organic or quantum-dot sensitizer, and the new redox electrolyte system (Hagfeldt et al., 2010; Tetreault et al., 2012). During this journey toward high-efficiency DSC, the light-harvesting capacity of the photoanode is expected to play a critical and indispensa‐ ble role by spanning the mesoporous electrodes utilizing varied nano/meso materials and the sensitizers with high molar extinction coefficient or quantum dots, and by connecting

There are mainly two components concerning the light-harvesting property of DSC, the dye and the mesoporous electrode which supports the sensitizer. While the development of the novel sensitizers with higher molar extinction coefficient and better response to near-infra‐ red wavelength has been long studied since the first breakthrough of DSC in 1991 (Nazeer‐ uddin et al., 2001; Grtäzel et al., 2009; Yum et al., 2011), the significance of the light scattering and/or reflection of the mesoporous electrode on the light-harvesting properties is only recognized in recent years. Accompanying the rapid advance of the ordered or hybrid photoanode materials, novel semiconductor nanostructures or new concepts are widely used to construct new type photoanodes, including the plenary optical waveguide nanowire enhancing the multiple internal reflections (Wei et al., 2010), the plasmonic core-shell metalinsulator nanoparticles enhancing the near-infrared absorption (Brown et al., 2011), the du‐ al-function mesoporous TiO2 structures increasing both the dye-loading and the optical scattering effects (Koo et al, 2008; Huang et al., 2010), and so on. In view of the wide cover‐ age of the optical and photoelectrical nanostructures involved in these studies and their rap‐ id progresses, it is great valuable to make a comprehensive and in-depth review on the

The chapter will start with a brief description of the basic concept of the light-harvesting ef‐ ficiency (LHE), and then give a review on five typical branches representing the significant

The basic principles of these novel nanostructures/ methods enhancing the light-harvesting capacity of DSC, together with their mutual effects on the electrical and photoelectrochemi‐ cal properties of the nanoporous electrode, will be discussed in detail. Based on the in-depth analysis of literature and the authors' experience, a perspective will be presented, shedding

the optical functions of the photoanode with its electrical response together.

current status of the light-harvesting capacities of the photoanode in DSC.

**3.** the dual-function scattering layer on the top of nanocrystalline (nc) electrode,

**1.** the mesoporous photoanodes with high surface area,

**2.** the hierarchically nanostructured photoanodes,

**5.** the photonic crystal photoanode and others.

a light on the research road in near future.

advances in this area, including

170 Solar Cells - Research and Application Perspectives

**4.** the plasmonic photoanodes, and

The typical DSC device is a sandwich type electrochemical cell, as seen from Figure 1, with the dye molecules absorbing the incident sunlight, and the framework of TiO2 mesoporous electrode transporting the photo-excited electrons generated from dye molecules to the bot‐ tom electrode. Therefore, the semiconductor porous electrode, or the photoanode film, is the kernel component in DSC, undertaking two major functions: supporting dye molecules, and transporting photo-excited electrons. To harvest the sunlight as far as possible, the photoa‐ node film has to possess a high internal surface area, to guarantee the massive uptake of dye molecules. The cell performance of DSC, that is, the solar-to-electric conversion efficiency, is determined by the short-circuit photocurrent (*J* sc), open-circuit photovoltage (*V* oc), and fill factor (*FF*) under a definite intensity of light such as the AM 1.5 solar spectrum. *J* sc, the criti‐ cal parameter affecting the power conversion efficiency (PCE) of DSC most, can be increased by raising the light-harvesting efficiency (LHE), which is defined by Equation 1,

$$\text{LHE } \langle \lambda \rangle = 1 - 10^{-\Gamma\_{\sigma}(\lambda)} \tag{1}$$

where Г is the number of moles of sensitizer per square centimeter of projected surface area of the film, and σ is the absorption cross section in units of cm2 /mol obtained from the deca‐ dic extinction coefficient (units of M-1 cm-l) by multiplication with 1000 cm3 /L (Nazeeruddin et al., 1993). The LHE is directly determined by the surface concentration of the dyes in the film, and the molar extinction coefficient of dye. The LHE, together with the quantum yield of charge injection (φinj) and the efficiency of collecting the injected charge at the back con‐ tact (*η* c), determines the incident monochromatic photon-to-current conversion efficiency (IPCE), defined as the number of electrons generated by light in the external circuit divided by the number of incident photons (Equation 2).

$$\text{IPCE } \begin{pmatrix} \lambda \end{pmatrix} = \text{LHE } \begin{pmatrix} \lambda \end{pmatrix} \rho\_{\text{inj}} \eta\_{\text{c}} \tag{2}$$

Previous studies proved that φinj is a wavelength independent parameter, and almost all of photons absorbed by the sensitizer are quantitatively converted to the conduction band elec‐ trons (Nazeeruddin et al., 1993). Meanwhile, the time-resolved laser photolysis experiments showed that the injected electrons can percolate without significant loss through the net‐ work of interconnected particles present in the nc-TiO2 film, and the back reaction is rela‐ tively slow, indicating that the majority of the injected charges is able to reach the back contact, and the influence of *η* c on IPCE is minor (Regan et al., 1990). So the wavelengthdepedent LHE has predominant effects on the IPCE and the PCE of DSC.

In essence, the LHE of DSC is the electrical response of the photovoltaic device to the solar spectrum projected on earth. It is predominantly related to there factors, the wavelength de‐ pendent light-absorbing capacity of the dye molecules, the quantity of the dye molecules ad‐ sorbed on the porous electrode, and the optical path of the incident light traveled in the electrode (determining the collision numbers of the light with the dye molecules). As a wavelength-dependent parameter, the LHE is usually evaluated by the IPCE spectrum. Fig‐ ure 2 gives the IPCE spectrum of a typical DSC with nc-TiO2 electrode, N719 dye and I2/I3 electrolyte system, and its comparison with the standard solar irradiation spectrum (ASTM G173). To characterize the response of DSC to the solar irradiation, the IPCE spectrum can be divided into four wavelength zones, i.e., the strong absorption zone between 500-550 nm, two moderate absorption zones between 400-500 nm and 550-700 nm, and the weak absorp‐ tion zone above 700 nm. While the broadband or near-infrared absorption sensitizers are de‐ vised to harvest the sun light beyond 700 nm, the modulation of the optical path of the incident light by introducing scattering or reflection is mostly used to improve the light har‐ vesting in two moderate absorption zones. From the viewpoint of the photoanode film (i.e., the porous electrode), there are several measures to improve the LHE, including


**Figure 1.** a) Schematic illustration of dye-sensitized solar cells (Hagfeldt, et al., 2010). b) Working principle of a typical DSC employing an iodide/triiodide-based redox electrolyte and N719 as a sensitizer (Grätzel, 2009).

In current days, the nanocrystallite electrode is confronting the maximum photocurrent theo‐ retically achievable with the Ru-based sensitizers (e.g. N719 dye), and the quest for the novel nanostructured photoanode has becoming prosperous (Zhang et al., 2011; Tetreault et al., 2012). Among numerous innovative nanostructured photoanodes, a considerable amount of studies are directed to improve the LHE of the photoanode, in view of the moderate surface area, the moderate electron transporting property of nc-TiO2 electrode, and the intrinsically low-surface-area nature of the traditional submicron scattering layer. In the following sections, we will introduce five typical research branches in this area, to provide readers with a compre‐ hensive and in-depth overview on the development of the LHE studies in DSC.

electrode (determining the collision numbers of the light with the dye molecules). As a wavelength-dependent parameter, the LHE is usually evaluated by the IPCE spectrum. Fig‐ ure 2 gives the IPCE spectrum of a typical DSC with nc-TiO2 electrode, N719 dye and I2/I3 electrolyte system, and its comparison with the standard solar irradiation spectrum (ASTM G173). To characterize the response of DSC to the solar irradiation, the IPCE spectrum can be divided into four wavelength zones, i.e., the strong absorption zone between 500-550 nm, two moderate absorption zones between 400-500 nm and 550-700 nm, and the weak absorp‐ tion zone above 700 nm. While the broadband or near-infrared absorption sensitizers are de‐ vised to harvest the sun light beyond 700 nm, the modulation of the optical path of the incident light by introducing scattering or reflection is mostly used to improve the light har‐ vesting in two moderate absorption zones. From the viewpoint of the photoanode film (i.e.,

the porous electrode), there are several measures to improve the LHE, including

scattering effects; and

172 Solar Cells - Research and Application Perspectives

ductor structures.

**1.** increasing the internal surface area of the electrode, which has the potential to improve the dye-loading capacity of the photoanode over a specific film thickness and area; **2.** increasing the optical path of the incident light in the electrode, by introducing scatter‐ ing centers in the bulk film, by introducing the scattering layer on the top of nanocrys‐ talline electrode, and by constructing hierarchical structures possessing strong light

**3.** enhancing the absorption of dye molecules by introducing plasmonic metal-semicon‐

**Figure 1.** a) Schematic illustration of dye-sensitized solar cells (Hagfeldt, et al., 2010). b) Working principle of a typical

In current days, the nanocrystallite electrode is confronting the maximum photocurrent theo‐ retically achievable with the Ru-based sensitizers (e.g. N719 dye), and the quest for the novel nanostructured photoanode has becoming prosperous (Zhang et al., 2011; Tetreault et al., 2012). Among numerous innovative nanostructured photoanodes, a considerable amount of studies are directed to improve the LHE of the photoanode, in view of the moderate surface

DSC employing an iodide/triiodide-based redox electrolyte and N719 as a sensitizer (Grätzel, 2009).

**Figure 2.** The light-harvesting capacity of N719 sensitized nc-TiO2 photoanode in DSC and its comparison with AM 1.5 Global solar spectrum.

## **3. Mesoporous photoanodes with high surface area**

It is essential for the photoanode in DSC to possess a high internal surface area, to adsorb sensi‐ tizer as much as possible. Typically, the hydrothermal nc-TiO2 with the particle size of 15-30 nm is used to fabricate the porous electrode, which usually exhibits the BET surface area of 50-100 m2 /g. While these nanocrystallite photoanodes perform fairly well and occupy the high‐ land of DSC with the best efficiency, their moderate dye-loading capacity motivate researchers to seek other materials with higher surface area. In fact, there are several materials which do possess much higher surface area than nc-TiO2 electrode, such as the mesoporous materials, aerogel, and metal-organic framework (MOF), whose surface area usually range from several hundred to three thousand m2 /g. It is greatly intriguing to construct photoanodes using these high-surface-area materials, which is expected to remarkably improve the dye-loading capaci‐ ty of the photoanode and the PCE of DSC in case a good charge transport can be maintained in these mesoporous structures. Up to date, mesoporous TiO2 in the form of powder, sphere, bead, or film, and aerogel have been used to synthesize the photoanode, which usually exhibit‐ ed the surface area of 100-1100 m2 /g. Here we present some typical mesoporous photoanodes prepared by the high-surface-area TiO2 mesostructures and TiO2 aerogel.

Mesoporous powders possessing a large volume of mesopores and irregular micron/ submi‐ cron secondary particles are the mostly-used materials in constructing the mesoporous pho‐ toanodes. The template methods are the common strategy to prepare these mesoporous powders. For example, T. K. Yun et al. prepared two different mesoporous TiO2 powders, using a soft template of tri-block copolymer and a hard template of mesoporous ZnO/ Zn(OH)2-composite, respectively (Yun et al., 2011). Both mesoporous TiO2 possessed the same high surface area (BET ~210 m2 /g) but with different pore sizes of 6.8 (for soft tem‐ plate) and 3.0 nm (for hard template). Different photovoltaic performances were observed. While the photoanode using the mesoporous TiO2 having larger pores showed the PCE of 6.71%, higher than P25 TiO2 nanopowders based photoanode (*η*= 5.62%), only half the per‐ formance (3.05%) was observed for the mesoporous TiO2 having small pores. The work indi‐ cated that, a suitable selection of the pore sizes was important for mesoporous materials used in DSC, and a too small pore size would inhibit the diffusion of dye molecules through the pores, thus greatly reducing the uptake of the dye molecules.

Beside the template method, the external fields such as the microwave irradiation can also be very useful in synthesizing mesoporous TiO2 powders. C. H. Huang et al. reported the rapid synthesis of mesoporous TiO2 via a microwave-assisted hydrothermal route using wa‐ ter-soluble titanium citrate complexes as the precursor, and obtained TiO2 powders showed the surface area of 217–323 m2 /g, and the pore diameter of 5.8–6.9 nm (Huang et al., 2011). The photoanode based on this mesoporous TiO2 exhibited the highest conversion efficiency of 7.1% (active area: 0.25 cm2 ). Also, the cell exhibited good long-term stability, and the PCE of a 1 cm2 device decreased merely from 4.8% to 4.3% after being stored for more than 490 h.

For most photoanodes using the high-surface-area mesoporous materials, it is required to prepare the mesoporous powders first, and the final electrode is obtained via the doctorblade method using the mesoporous paste prepared from the mesoporous powders. This is a relatively complicated process. In contrast, W. Chen et al. developed a facile one-step preparation of crack-free thick ordered mesoporous TiO2 films, via the combination of ''doc‐ tor blade'' technique and the evaporation induced self-assembly method (Chen et al., 2007). By employing the as-synthesized mesoporous film (7 μm in thickness) as the photoanode, a PCE of 6.53% was obtained at 30 mW/cm2 light intensity, illustrating the potential of this simple strategy in constructing powerful DSC devices.

While the mesoporous materials of pure TiO2 give a promising potential to improve the LHE, the introduction of other oxides (e.g., Al2O3, SiO2) in the mesoporous structure of TiO2 opens an alternative route to enhance the LHE of DSC. As an example, J. Y. Kim et al. pre‐ pared highly ordered mesoporous Al2O3/TiO2 via a sol-gel reaction and evaporation-in‐ duced self-assembly route using Pluronic P123 as the structure-modifying reagent (Kim et al., 2011). Obtained mesoporous powders possessed an average pore size of 6.33-6.58 nm, and BET surface area of 181-212 m2 /g, and the content of Al2O3 had significant effects on the BET surface area. The thin Al2O3 layer was located mostly on the surface of mesopore (as seen in X-ray photoelectron spectroscopy (XPS) spectra in Figure 4c), blocking the charge re‐ combination during the photon-to-electron conversion process. The photoanode based on the mesoporous Al2O3/TiO2 (1 mol % Al2O3) exhibited ~10% efficiency improvement com‐ pared to the pure mesoporous TiO2 electrode (*η* increasing from 5.88% to 6.50%). Though the authors didn't compare their results with traditional nanocrystalline counterpart, the much higher surface area of the mesoporous electrode than P25 TiO2 may possibly lead to higher dye-loading quantity, and corresponding higher light-harvesting capacity. Also, the introduction of Al2O3 barrier layer in mesoporous TiO2 illustrated a viable and versatile route to further improve the functions of the mesoporous photoanodes.

Mesoporous powders possessing a large volume of mesopores and irregular micron/ submi‐ cron secondary particles are the mostly-used materials in constructing the mesoporous pho‐ toanodes. The template methods are the common strategy to prepare these mesoporous powders. For example, T. K. Yun et al. prepared two different mesoporous TiO2 powders, using a soft template of tri-block copolymer and a hard template of mesoporous ZnO/ Zn(OH)2-composite, respectively (Yun et al., 2011). Both mesoporous TiO2 possessed the

plate) and 3.0 nm (for hard template). Different photovoltaic performances were observed. While the photoanode using the mesoporous TiO2 having larger pores showed the PCE of 6.71%, higher than P25 TiO2 nanopowders based photoanode (*η*= 5.62%), only half the per‐ formance (3.05%) was observed for the mesoporous TiO2 having small pores. The work indi‐ cated that, a suitable selection of the pore sizes was important for mesoporous materials used in DSC, and a too small pore size would inhibit the diffusion of dye molecules through

Beside the template method, the external fields such as the microwave irradiation can also be very useful in synthesizing mesoporous TiO2 powders. C. H. Huang et al. reported the rapid synthesis of mesoporous TiO2 via a microwave-assisted hydrothermal route using wa‐ ter-soluble titanium citrate complexes as the precursor, and obtained TiO2 powders showed

The photoanode based on this mesoporous TiO2 exhibited the highest conversion efficiency

For most photoanodes using the high-surface-area mesoporous materials, it is required to prepare the mesoporous powders first, and the final electrode is obtained via the doctorblade method using the mesoporous paste prepared from the mesoporous powders. This is a relatively complicated process. In contrast, W. Chen et al. developed a facile one-step preparation of crack-free thick ordered mesoporous TiO2 films, via the combination of ''doc‐ tor blade'' technique and the evaporation induced self-assembly method (Chen et al., 2007). By employing the as-synthesized mesoporous film (7 μm in thickness) as the photoanode, a

While the mesoporous materials of pure TiO2 give a promising potential to improve the LHE, the introduction of other oxides (e.g., Al2O3, SiO2) in the mesoporous structure of TiO2 opens an alternative route to enhance the LHE of DSC. As an example, J. Y. Kim et al. pre‐ pared highly ordered mesoporous Al2O3/TiO2 via a sol-gel reaction and evaporation-in‐ duced self-assembly route using Pluronic P123 as the structure-modifying reagent (Kim et al., 2011). Obtained mesoporous powders possessed an average pore size of 6.33-6.58 nm,

BET surface area. The thin Al2O3 layer was located mostly on the surface of mesopore (as seen in X-ray photoelectron spectroscopy (XPS) spectra in Figure 4c), blocking the charge re‐ combination during the photon-to-electron conversion process. The photoanode based on the mesoporous Al2O3/TiO2 (1 mol % Al2O3) exhibited ~10% efficiency improvement com‐

device decreased merely from 4.8% to 4.3% after being stored for more than 490 h.

/g) but with different pore sizes of 6.8 (for soft tem‐

/g, and the pore diameter of 5.8–6.9 nm (Huang et al., 2011).

). Also, the cell exhibited good long-term stability, and the PCE

light intensity, illustrating the potential of this

/g, and the content of Al2O3 had significant effects on the

same high surface area (BET ~210 m2

174 Solar Cells - Research and Application Perspectives

the surface area of 217–323 m2

of 7.1% (active area: 0.25 cm2

PCE of 6.53% was obtained at 30 mW/cm2

and BET surface area of 181-212 m2

simple strategy in constructing powerful DSC devices.

of a 1 cm2

the pores, thus greatly reducing the uptake of the dye molecules.

**Figure 3.** TEM image of mesoporous TiO2 with a) low and b) high magnifications. c) *J*-V curves and d) IPCE spectra of DSCs prepared from P25 and mesoporous TiO2 with TiCl4 post-treatment. (Yun et al., 2011).

Though above mesoporous powders with irregular secondary particles are promising in im‐ proving the dye-loading capacity of the photoanode, it is usually difficult for them to obtain high quality electrodes with good transparency, due to the poor workability of the paste produced from the irregular micron/submicron particles. Therefore, it is intriguing to pre‐ pare mesoporous TiO2 with regular secondary particles, e.g., TiO2 mesoporous spheres.

**Figure 4.** a-b) TEM images of mesoporous TiO2, illustrating the ordered mesopores. c) XPS spectra of mesoporous Al2O3/ TiO2 with various levels of Al calcined at 525oC. d) *J*-V curves in the illuminated and dark states of pure mesoporous TiO2, 1 and 2 mol% Al2O3/TiO2 (light intensity: 100 mW/cm2; AM 1.5 filter; illumination area: 0.25 cm2). (Kim et al., 2010).

D. Chen et al. did a good job on this area, and synthesized uniform, crystalline, and mesopo‐ rous TiO2 beads in size of ~800 nm through a combination of sol-gel and solvothermal process‐ es, which exhibited the surface area of 108.0 m2 /g and tunable pore sizes (pore diameter: 14.0 ~ 22.6 nm) (Chen et al., 2009). Due to the submicrometer-sized particle diameter and the high specific surface area, the mesoporous TiO2 beads enhanced the LHE without sacrificing the ac‐ cessible surface for dye loading, thereby increasing the PCE compared to P25 nanoparticles. The photoanode based on these mesoporous TiO2 beads exhibited an overall light conversion efficiency of 7.20% (*V* oc = 0.777 V, *J* sc = 12.79 mA/cm2 , *FF* = 0.72), significantly higher than that derived from standard Degussa P25 TiO2 electrode with similar thickness (5.66%).

In a further study, the group selected Ru(II)-based dye C101 and C106 to match with the mesoporous TiO2 beads, and realized an overall PCE of greater than 10% (*η*=10.7%) using a single screen-printed TiO2 layer cell construction (without an additional scattering layer). Hence, a delicate modulation of the micron/submicron size/ shape of mesoporous materials, and a careful selection of the dye to match the dye's photon-absorption characteristics with the light scattering properties of the mesoporous materials, are important to improve the LHE and PCE of DSC (Sauvage et al., 2010).

**Figure 4.** a-b) TEM images of mesoporous TiO2, illustrating the ordered mesopores. c) XPS spectra of mesoporous Al2O3/ TiO2 with various levels of Al calcined at 525oC. d) *J*-V curves in the illuminated and dark states of pure mesoporous TiO2, 1 and 2 mol% Al2O3/TiO2 (light intensity: 100 mW/cm2; AM 1.5 filter; illumination area: 0.25 cm2). (Kim et al., 2010).

D. Chen et al. did a good job on this area, and synthesized uniform, crystalline, and mesopo‐ rous TiO2 beads in size of ~800 nm through a combination of sol-gel and solvothermal process‐

22.6 nm) (Chen et al., 2009). Due to the submicrometer-sized particle diameter and the high specific surface area, the mesoporous TiO2 beads enhanced the LHE without sacrificing the ac‐ cessible surface for dye loading, thereby increasing the PCE compared to P25 nanoparticles. The photoanode based on these mesoporous TiO2 beads exhibited an overall light conversion

In a further study, the group selected Ru(II)-based dye C101 and C106 to match with the mesoporous TiO2 beads, and realized an overall PCE of greater than 10% (*η*=10.7%) using a single screen-printed TiO2 layer cell construction (without an additional scattering layer). Hence, a delicate modulation of the micron/submicron size/ shape of mesoporous materials, and a careful selection of the dye to match the dye's photon-absorption characteristics with

derived from standard Degussa P25 TiO2 electrode with similar thickness (5.66%).

/g and tunable pore sizes (pore diameter: 14.0 ~

, *FF* = 0.72), significantly higher than that

es, which exhibited the surface area of 108.0 m2

176 Solar Cells - Research and Application Perspectives

efficiency of 7.20% (*V* oc = 0.777 V, *J* sc = 12.79 mA/cm2

**Figure 5.** a-b) SEM images of the calcined mesoporous TiO2 beads obtained after a solvothermal process with no ammonia for a) and 1.0 mL ammonia for b). c) IPCE curves and d) *J*-V curves of the TiO2 electrodes prepared from P25 NPs and mesoporous TiO2 beads with two film thicknesses. (Chen et al., 2009) e) SEM image of the screenprinted film composed of TiO2 porous beads; f) *J*-V curves recorded at 100 mW/cm2 of the 12 um thick film com‐ posed of P25 particles (dashed) or TiO2 beads (solid) sensitized with the C101 dye without (in black color) and with TiCl4 post-treatment (in red color). (Sauvage et al., 2010).

**Figure 6.** a-b) SEM images of aerogel frameworks coated with a) 4.4 and b) 8.4 nm ZnO. c-d) Plots of c) *J*sc, *FF* and d) efficiency (η) as a function of thickness of ZnO deposited on aerogel frameworks. (Hamann et al., 2008a) e) SEM image of a ∼30 μm thick aerogel film coated with 9.6 nm TiO2. f) *J*-V curves for aerogel electrodes coated with 12 nm TiO2 (Red) and TiO2 nanoparticle (orange) in the dark (dasked lines) and under AM 1.5 illumination (solid). (Hamann et al., 2008b).

Among various mesoporous materials, aerogel is distinguished from others by its high surface area (100-1600 m2 /g) and light weight (0.003-0.5 g/cm3 ) (Hüsing et al., 1998). The application of aerogel in the photoanode of DSC was pioneered by J. J. Pietron and J. T. Hamann in 2007-2008. The former work prepared the aerogel photoanode by the direct doctor-blading of the crushed TiO2 aerogel particles, and demonstrated the IPCE of ~85% in the 500-600 nm range and ~52% at 700 nm for N719 sensitized photoanode (Pietron et al., 2007). Hamann et al. prepared SiO2 aerogel film first via the supercritical drying proc‐ ess, and then coated ZnO or TiO2 thin layer with controllable thickness via the atomic lay‐ er deposition (ALD) method (Hamann et al., 2008a, 2008b). They realized the aerogel based DSC devices with the efficiency of 2.4% (for ZnO cell) and 4.3% (for TiO2 cell). Al‐ so, the authors investigated systematically the effects of ZnO thickness on SiO2 aerogel template on the photovoltaic performance, and found a nearly linear relationship between the key photovoltaic index and the thickness of ZnO layer (Figure 6c,d), with the optimal ZnO thickness of 10.5 nm. These results indicated that, on one hand, the aerogel film in combination of ALD or other physical or chemical deposition processes was a powerful candidate to prepare mesoporous photoanode with controllable microstructures. On the other, the use of the dangerous supercritical drying process in the synthesis of aerogel, to‐ gether with the expensive and complicated ALD process involved in the preparation of ZnO/TiO2 layer, limited the popularization of this method greatly. There were very few studies on the aerogel photoanodes in recent five years.

**Figure 7.** a) Schematic of the preparation of SiO2-TiO2 hybrid aerogel and aerogel based photoanode. SEM image of b) SiO2 and c) SiO2-TiO2 hybrid aerogel (packing density: 0.037 and 0.125 g/cm3). d) TEM image and e) SAED pattern of SiO2-TiO2 hybrid aerogel. f) *J*-V curves of DSC based on pure TiO2 photoanode with and without scattering layer and aerogel modified photoanode. (Gao et al., 2012).

**Figure 6.** a-b) SEM images of aerogel frameworks coated with a) 4.4 and b) 8.4 nm ZnO. c-d) Plots of c) *J*sc, *FF* and d) efficiency (η) as a function of thickness of ZnO deposited on aerogel frameworks. (Hamann et al., 2008a) e) SEM image of a ∼30 μm thick aerogel film coated with 9.6 nm TiO2. f) *J*-V curves for aerogel electrodes coated with 12 nm TiO2 (Red) and TiO2 nanoparticle (orange) in the dark (dasked lines) and under AM 1.5 illumination (solid). (Hamann et al., 2008b). Among various mesoporous materials, aerogel is distinguished from others by its high

/g) and light weight (0.003-0.5 g/cm3

application of aerogel in the photoanode of DSC was pioneered by J. J. Pietron and J. T. Hamann in 2007-2008. The former work prepared the aerogel photoanode by the direct doctor-blading of the crushed TiO2 aerogel particles, and demonstrated the IPCE of ~85% in the 500-600 nm range and ~52% at 700 nm for N719 sensitized photoanode (Pietron et al., 2007). Hamann et al. prepared SiO2 aerogel film first via the supercritical drying proc‐ ess, and then coated ZnO or TiO2 thin layer with controllable thickness via the atomic lay‐ er deposition (ALD) method (Hamann et al., 2008a, 2008b). They realized the aerogel based DSC devices with the efficiency of 2.4% (for ZnO cell) and 4.3% (for TiO2 cell). Al‐

) (Hüsing et al., 1998). The

surface area (100-1600 m2

178 Solar Cells - Research and Application Perspectives

To crack above problems, X. D. Gao et al. developed a low-cost sol-gel and ambient-dry‐ ing route to fabricate SiO2-TiO2 hybrid aerogels, which possessed a very high surface area of 500-1170 m2 /g, and incorporated them into nc-TiO2 electrode, obtaining the aerogel hy‐ brid photoanodes (Gao, et al., 2012). These hybrid photoanodes yielded significantly high‐ er photocurrent densities than nc-TiO2 photoanodes due to the increased dye-loading capacity and the enhanced visible scattering effect, showing the highest efficiency of 9.41% at optimal condition, 16% higher than the TiCl4-treated TiO2 photoanode modified with a conventional scattering layer. This work indicated that, the hybrid photoanodes in‐ tegrating both the high-surface-area mesoporous materials and nc-TiO2 represented a powerful and viable route toward high-efficiency DSCs.

In brief, current studies on the mesoporous photoanodes built up by the high-surface-area mesoporous materials have unambiguously proved that, the application of high-surfacearea powders/beads/aerogel in the photoanode can effectively increase the dye-loaing ca‐ pacity and the light scattering of the electrode, and correspondingly improve the LHE of DSC in most cases. The preparation of mesoporous materials with regular and uniform size/ shape in micron/submicron scale, and their integration with nc-TiO2 or other nanostruc‐ tures, seem feasible to achieve high conversion efficiency. A remaining problem in these high-surface-area mesoporous materials is the inferior electron-transporting property, in‐ trinsically originating from their lower particle coordination number than the common nc-TiO2 porous electrode. A possible solution may be the development of hybrid electrodes by incorporating metal, carbon nanotube, or graphene etc into the mesoporous structures. A mesoporous material possessing both high surface area and good conductivity is by all means necessary for highly efficient DSCs.

## **4. Hierarchical photoanodes**

Hierarchical nanostructures, those with different morphology at different length scale and usurally characterized by the high surface area, are intresting candidates for efficienct photoanodes in DSC. This special class of nanostrcutures is spurred at the low surface area and consequently the low PCE of nanowire based photoandes, and have receivd wide-spread and intensive studies in recent years. Hierarchical ZnO and TiO2 in the form of grass, forest, popborn ball, and tube etc. have been developed, and a wide range of chemical and physical methods have been used to prepare the hierarchical structures, in‐ cluding the hydrothermal synthesis, sol-gel, chemical bath deposition, pulsed laser depo‐ sition, photolithography, and so on. Due to the huge literature in this field, here we merely present some representive results in recent three years.

ZnO nanowire array may be the piorneer in the nanostructured photoanode since the out‐ standing work of M. Law et al. (Law et al., 2005). However, the PCE of nanowire based pho‐ toanode was low (mostly 1%-3% for a long time) mainly due to its low surface area. So the development of hierarchical ZnO and their application in DSC are wide-spread recently. As an example, Xu et al. reported a two-step synthesis process to produce hierarchical ZnO nanoarchitectures (Figure 8a,b), in which ZnO nanosheet arrays were first prepared by the pyrolysis of the precursor Zn5(OH)8Cl2 electrodeposited on conductive glass substrates, and then dense ZnO nanowires were grown on the surfaces of the primary ZnO nanosheets by the chemical bath deposition method (Xu et al., 2010). Due to the better dye loading through the increased internal surface area and the higher light scattering behavior through extend‐ ing the optical path length within the photoanode, the DSC based on the hierarchical ZnO nanowire-nanosheet architectures showed a PCE of 4.8%, nearly twice as high as that of DSC constructed using a photoanode of bare ZnO nanosheet array.

capacity and the enhanced visible scattering effect, showing the highest efficiency of 9.41% at optimal condition, 16% higher than the TiCl4-treated TiO2 photoanode modified with a conventional scattering layer. This work indicated that, the hybrid photoanodes in‐ tegrating both the high-surface-area mesoporous materials and nc-TiO2 represented a

In brief, current studies on the mesoporous photoanodes built up by the high-surface-area mesoporous materials have unambiguously proved that, the application of high-surfacearea powders/beads/aerogel in the photoanode can effectively increase the dye-loaing ca‐ pacity and the light scattering of the electrode, and correspondingly improve the LHE of DSC in most cases. The preparation of mesoporous materials with regular and uniform size/ shape in micron/submicron scale, and their integration with nc-TiO2 or other nanostruc‐ tures, seem feasible to achieve high conversion efficiency. A remaining problem in these high-surface-area mesoporous materials is the inferior electron-transporting property, in‐ trinsically originating from their lower particle coordination number than the common nc-TiO2 porous electrode. A possible solution may be the development of hybrid electrodes by incorporating metal, carbon nanotube, or graphene etc into the mesoporous structures. A mesoporous material possessing both high surface area and good conductivity is by all

Hierarchical nanostructures, those with different morphology at different length scale and usurally characterized by the high surface area, are intresting candidates for efficienct photoanodes in DSC. This special class of nanostrcutures is spurred at the low surface area and consequently the low PCE of nanowire based photoandes, and have receivd wide-spread and intensive studies in recent years. Hierarchical ZnO and TiO2 in the form of grass, forest, popborn ball, and tube etc. have been developed, and a wide range of chemical and physical methods have been used to prepare the hierarchical structures, in‐ cluding the hydrothermal synthesis, sol-gel, chemical bath deposition, pulsed laser depo‐ sition, photolithography, and so on. Due to the huge literature in this field, here we

ZnO nanowire array may be the piorneer in the nanostructured photoanode since the out‐ standing work of M. Law et al. (Law et al., 2005). However, the PCE of nanowire based pho‐ toanode was low (mostly 1%-3% for a long time) mainly due to its low surface area. So the development of hierarchical ZnO and their application in DSC are wide-spread recently. As an example, Xu et al. reported a two-step synthesis process to produce hierarchical ZnO nanoarchitectures (Figure 8a,b), in which ZnO nanosheet arrays were first prepared by the pyrolysis of the precursor Zn5(OH)8Cl2 electrodeposited on conductive glass substrates, and then dense ZnO nanowires were grown on the surfaces of the primary ZnO nanosheets by the chemical bath deposition method (Xu et al., 2010). Due to the better dye loading through the increased internal surface area and the higher light scattering behavior through extend‐

merely present some representive results in recent three years.

powerful and viable route toward high-efficiency DSCs.

means necessary for highly efficient DSCs.

**4. Hierarchical photoanodes**

180 Solar Cells - Research and Application Perspectives

Apart from these hierarchical structures with relatively simple structures, several groups de‐ veloped ZnO or TiO2 forest (Figure 8c,d), more complicated hierarchical structures. F. Sauv‐ age et al. prepared the hierarchical assemblies of nanocrystalline particles of anatase TiO2 (Sauvage et al., 2010), by a pulsed laser deposition method, via the fine modulation of the plasma expansion dynamics by means of a reactive atmosphere during the ablation process. In combination with the high molar extinction coefficient heteroleptic C101 dye, the authors achieved 3.1% power conversion efficiency for 2 um thick films and 5.0% for films 7 um thick, demonstrating the great potential of the physical vapor deposition method in growing hierarchical nanostructured photoanodes with high conversion efficiency.

**Figure 8.** a) Schematic of ZnO hierarchical nanostructures derived from ZnO nanosheet arrays and b) top-view SEM image of ZnO nanowire-nanosheet architectures obtained by aqueous chemical growth after 4 h. The inset of b) cor‐ responds to the magnified image. (Xu et al., 2010) c) TiO2 forest-like films grown by pulsed laser deposition: (left) over‐ view of a film deposited at 40 Pa; (right) cross sections of films deposited at 20 and 40 Pa. (Sauvage et al., 2010) d) SEM image of ZnO nanowire nanoforest (tilted view). (Ko et al., 2011).

Meanwhile, using the traditional hydrothermal method integrating with the simple selective hierarchical growth sequence, S. Ko et al. demonstrated ZnO nanoforest photoanode with high density, long branched treelike multigeneration hierarchical crystalline ZnO nanowires (Figure 8d) (Ko et al., 2011). The DSC exhibited almost 5 times higher than the efficiency of DSCs constructed by upstanding ZnO nanowires (2.63% v.s. 0.45%). The enhanced surface area for higher dye loading and light harvesting, and the reduced charge recombination by providing direct conduction pathways along the crystalline ZnO "nanotree" multi genera‐ tion branches, were responsible for the efficiency improvement.

**Figure 9.** a) Schematic of the preparation of hierarchical TiO2 NTA: beginning with a ZnO nanowire and followed by multiple procedures of TiO2 coating, ZnO core removal, subsequent TiO2 coating, and hierarchical derivation to a dou‐ ble-shell architecture. b) cross-section SEM image of TiO2 nanotube array with 41 um thickness and c) magnified SEM image of hierarchical TiO2 nanotube. d) *J*–*V* curves of the nanotube DSCs prepared with 20-μm thick nanotube arrays with the original and hierarchical structure. (Zhuge et al., 2011) e) TEM images of triple-shelled TiO2 NTs with different structural features. f) *J*–V curves of DSCs based on multi-sheel TiO2 nanotube electrodes. (Qiu et al., 2012).

Nanotube array (NTA) is another type of hierarchical structures receiving great attention, which is developed on the basis of nanowire in view of its much higher surface area than nanowire. Currently, the hierarchical TiO2 NTA has becoming an intensive research branch. Different from the traditional synthesis of TiO2 nanotube via the anodic oxidation, F. Zhuge et al. reported the synthesis of vertically aligned TiO2 NTAs (up to 40 um) on FTO by using ZnO nanowire array as the hard template, and demonstrated the hierarchical derivation for effective surface area enhancement in DSCs (Figure 9 a-c) (Zhuge et a., 2011). The distinctive double-shell structure of the nanotube provided both the high surface area and the high electron-transport path, showing the overall conversion efficiency of 5.7%, 50% higher than the original nanotube (Figure 8d). In a further work, the group updated their multi-shell fab‐ rication technology via ZnO nanowire template, and successfully obtained coaxial multishelled TiO2 NTA with controllable shell numbers and shell thickness (Figure 9e) (Qiu et al., 2012). Exhibiting the BET surface area of 119-331 m2 /g, DSC devices based on these multishelled NTA exhibited the highest efficiency of 6.2%, higher than that based on the singlelayer TiO2 nanotube array (3.35%) at similar experimental parameters (Figure 9f).

DSCs constructed by upstanding ZnO nanowires (2.63% v.s. 0.45%). The enhanced surface area for higher dye loading and light harvesting, and the reduced charge recombination by providing direct conduction pathways along the crystalline ZnO "nanotree" multi genera‐

**Figure 9.** a) Schematic of the preparation of hierarchical TiO2 NTA: beginning with a ZnO nanowire and followed by multiple procedures of TiO2 coating, ZnO core removal, subsequent TiO2 coating, and hierarchical derivation to a dou‐ ble-shell architecture. b) cross-section SEM image of TiO2 nanotube array with 41 um thickness and c) magnified SEM image of hierarchical TiO2 nanotube. d) *J*–*V* curves of the nanotube DSCs prepared with 20-μm thick nanotube arrays with the original and hierarchical structure. (Zhuge et al., 2011) e) TEM images of triple-shelled TiO2 NTs with different

Nanotube array (NTA) is another type of hierarchical structures receiving great attention, which is developed on the basis of nanowire in view of its much higher surface area than nanowire. Currently, the hierarchical TiO2 NTA has becoming an intensive research branch. Different from the traditional synthesis of TiO2 nanotube via the anodic oxidation, F. Zhuge et al. reported the synthesis of vertically aligned TiO2 NTAs (up to 40 um) on FTO by using ZnO nanowire array as the hard template, and demonstrated the hierarchical derivation for effective surface area enhancement in DSCs (Figure 9 a-c) (Zhuge et a., 2011). The distinctive double-shell structure of the nanotube provided both the high surface area and the high electron-transport path, showing the overall conversion efficiency of 5.7%, 50% higher than the original nanotube (Figure 8d). In a further work, the group updated their multi-shell fab‐ rication technology via ZnO nanowire template, and successfully obtained coaxial multishelled TiO2 NTA with controllable shell numbers and shell thickness (Figure 9e) (Qiu et al.,

structural features. f) *J*–V curves of DSCs based on multi-sheel TiO2 nanotube electrodes. (Qiu et al., 2012).

tion branches, were responsible for the efficiency improvement.

182 Solar Cells - Research and Application Perspectives

Apart from the hierarchical nanowire/nanotube array, there are also a few studies which mod‐ ulate the pores in the photoanode to realize the hierarchical structures. C. Y. Cho et al. reported an interesting method to generate hierarchical electrodes consisting of meso- and macroscale pores (Cho et al., 2011). Mesoscale colloidal particles and lithographically patterned macro‐ pores were used as the dual templates, with the colloidal particles assembled within the mac‐ ropores. An infiltration of TiO2 into the template and subsequent removal of the template produced hierarchical TiO2 electrodes for DSC (Figure 10a). While the holographic lithogra‐ phy defined the macroporous networks, the colloidal crystal templates provided three-dimen‐ sionally organized mesoscale pores with a uniform size distribution (Figure 10 b-e). Owing to the strong scattering and the suppression of charge recombination in the hierarchical TiO2 elec‐ trodes, the photovoltaic performance of the cell was comparable with nc-TiO2 electrodes, showing a maximum efficiency of 5.0% with 50 nm pores and 6 μm thickness.

**Figure 10.** a) Schematic of the preparation of hierarchical TiO2 electrode via double template method. (1) Forma‐ tion of four-beam interference and the fabrication of the macroporous SU-8 structures. (2) Filling of the holo‐ graphic patterns with mesoscale colloidal particles, and (3) Coating of precursors and removal of dual templates. SEM image of b) the surface and c) the cross-section of the macroporous SU-8 surface. SEM image of the hierarch‐ ical porous TiO2 electrode from dual templates of SU-8 filled with colloidal particles with diameters of d) 60 nm and e) 110 nm. Scale bar: 1 μ m. (Cho et al., 2011).

In general, the hierarchical photoanodes, though possessing obviously higher internal sur‐ face area than the nanowire/nanotube/nanoplate counterparts and higher conductivity than nc-TiO2 electrode, haven't show a predominant advantage over other materials up to now. The PCE of most photoanodes based on the hierarchical structures is still much lower than the state-of-the-art performance of nc-TiO2 electrode. A possible vent for the hierarchical photoanodes may be the further increase of its internal surface area while maintaining its high electron-transporting nature. Also the combination of the hierarchical structures with the mesoporous material is a meaningful and feasible strategy.

## **5. Scattering layers on nc-TiO2 electrode**

The photoanode of DSC is traditionally composed of nc-TiO2 film, which is basically trans‐ parent for visible light (Rayleigh scattering, mainly), resulting in a considerable part of light shone on the DSC transmitting through the TiO2 layer without interacting with the sensitiz‐ er. To overcome this problem, scattering layers with different structures, including scatter‐ ing centers and upper scattering layers, have been employed. The theoretical studies since 1998, and subsequent substantial experimental studies on scattering layers based on the sub‐ micron particles of TiO2, ZrO2, ZnO, nanowire, hierarchical structures, mesoporous spheres, photonic crystals, and upper-conversion materials, comprise a vivid picture about the re‐ search activities in this area. By briefing some important advances, we describe here some novel scattering layers developed in recent years.

**Figure 11.** a-b) Diffused reflection spectra of a) undyed and b) sensitized TiO2 layer without scattering layer (■), with the scattering layers of TiO2-Rutile (▲), ZrO2 (●) and 25:75 mixture of TiO2-Rutile and ZrO2 (◆). c) IPCE curves for devi‐ ces with TiO2 electrode with (■) and without (○) a scattering layer, the scattering layer being 25:75 mixture of TiO2- Rutile and ZrO2. d) Current densities obtained from devices with 4 mm thin layer of TiO<sup>2</sup> with and without scattering layers. The thickness of the scattering layers is ~6 um for all samples. (Hore et al., 2006).

In 1998, J. Ferber et al. started from Mie theory, and calculated the multiple scattering in TiO2 electrodes using a numerical solution of the radiative transfer equation (Ferber et al., 1998). These calculations predicted that, a suitable mixture of small particles (20 nm), which resulted in a large effective surface, and of large particles (250-300 nm), which were effective light scatterers, had the potential to enhance the solar absorption significantly.

**5. Scattering layers on nc-TiO2 electrode**

184 Solar Cells - Research and Application Perspectives

novel scattering layers developed in recent years.

The photoanode of DSC is traditionally composed of nc-TiO2 film, which is basically trans‐ parent for visible light (Rayleigh scattering, mainly), resulting in a considerable part of light shone on the DSC transmitting through the TiO2 layer without interacting with the sensitiz‐ er. To overcome this problem, scattering layers with different structures, including scatter‐ ing centers and upper scattering layers, have been employed. The theoretical studies since 1998, and subsequent substantial experimental studies on scattering layers based on the sub‐ micron particles of TiO2, ZrO2, ZnO, nanowire, hierarchical structures, mesoporous spheres, photonic crystals, and upper-conversion materials, comprise a vivid picture about the re‐ search activities in this area. By briefing some important advances, we describe here some

**Figure 11.** a-b) Diffused reflection spectra of a) undyed and b) sensitized TiO2 layer without scattering layer (■), with the scattering layers of TiO2-Rutile (▲), ZrO2 (●) and 25:75 mixture of TiO2-Rutile and ZrO2 (◆). c) IPCE curves for devi‐ ces with TiO2 electrode with (■) and without (○) a scattering layer, the scattering layer being 25:75 mixture of TiO2- Rutile and ZrO2. d) Current densities obtained from devices with 4 mm thin layer of TiO<sup>2</sup> with and without scattering

In 1998, J. Ferber et al. started from Mie theory, and calculated the multiple scattering in TiO2 electrodes using a numerical solution of the radiative transfer equation (Ferber et al., 1998). These calculations predicted that, a suitable mixture of small particles (20 nm), which

layers. The thickness of the scattering layers is ~6 um for all samples. (Hore et al., 2006).

In 2006, S. Hore et al. reported systematically the effects of TiO2/ZrO2 scattering layer on the photovoltaic performances (Hore et al., 2006). Different scattering layers composed of com‐ mercial TiO2-Rutile (Bayer Germany) and ZrO2 (TOSOH Corporation, Tokyo, Japan) with varied proportion were examined, providing an in-depth understanding on the functions of the top scattering layer. DSC device with an area of 2.5 cm2 exhibited the PCE of 6.8% using an only 4 μm layer of TiO2 electrode together with an optimal scattering layer, illustrating the potential of a suitable scattering layer on both the improvement of the efficiency and the reduction of the cost by using less sensitizer.

Very recently, F. E. Gálvez et al. reported an integral optical and electrical theoretical analysis on the effects of the different design of the diffuse light scattering on the performance of DSCs (Gálvez et al., 2012). Based on a Monte Carlo approach, they introduced the light harvesting ef‐ ficiency and the electron generation function extracted from optical numerical calculations in a standard electron diffusion model, to obtain the steady-state characteristics of the different configurations considered (Figure 12). They proposed that the diffuse scattering layers acting as the back reflector provided the largest achievable light harvesting efficiencies, which deter‐ mined an optimum overall performance, for electron diffusion lengths longer than the elec‐ trode thickness. When the electron diffusion length was shorter than the electrode thickness, the embedding of the diffuse scattering particles in the nanocrystalline paste yielded a better output even when the light harvesting was not optimal. These results can provide helpful guidance for us to design and prepare the scattering structure of the photoanode.

**Figure 12.** Simulated trajectories of a photon absorbed by a) standard semi-transparent cell, b) cell with an electrode embedding diffuse scattering particles, and c) cell made of a semi-transparent electrode coupled to a diffuse scatter‐ ing layer, illustrating the different series of scattering events that yield longer optical paths and thus enhance the probability of absorption in each type of modified cell under consideration. (Ga˃lvez et al., 2012).

In recent years, with the rapid development of nanostructured photoanodes, various nano‐ structures have been used as the top scattering layer of nc-TiO2 electrode or the novel photo‐ anodes based on nanowire/ nanotube arrays etc. For example, K Fan et al. applied TiO2 fusiform nanorods (diameter: 20-80 nm, length: 200-400 nm) as the scattering layer of P25 based electrode (Figure 13), and observed 66.5% improvement in the efficiency with lower resistance and longer electron lifetimes as compared to the bare P25-based solar cell (Fan et al., 2011). The reduced charge recombination and the sufficient scattering effect of the nano‐ rods in the film electrode were responsible for this improvement.

Koo H. J. et al. prepared nano-embossed hollow spherical (NeHS) TiO2 particles and investi‐ gated their photovoltaic property in DSC as the bifunctional scattering layer (Koo et al., 2008). The walls of the obtained hollow spheres were composed of nanocrystalline anatase TiO2 particles with mesoporous structure (Figure 14 a-c), which endowed them 5 times high‐ er dye-loading capacity (N719 dye: up to 5.0×10–5 mol/g) than that of the normally used 400 nm-diameter scattering particles with flat surfaces. The NeHS particles scattered incoming light effectively as confirmed by the reflectance spectroscopy (Figure 14 d,e). When the NeHS TiO2 particles were used as the secondary scattering layer in DSC, a substantial im‐ provement in the efficiency has been achieved, with the highest conversion efficiency of 10.34%. Huang F et al. used the submicrometer-sized mesoporous TiO2 beads as the dual function scattering layer for the nanocrystalline electrode, and observed similar results, demonstrating that the dual-function scattering layer represented a powerful alternative for the traditional large-particle scattering layer for high-efficiency DSCs (Huang et al., 2010).

**Figure 13.** a) Schematic of the photoanodes with nanorod scattering layer. b) *J*-V curves of DSCs, and c) UV-VIS spec‐ tra of bare P25 film, and with different nanorod scattering layer. (Fan et al., 2011).

resistance and longer electron lifetimes as compared to the bare P25-based solar cell (Fan et al., 2011). The reduced charge recombination and the sufficient scattering effect of the nano‐

Koo H. J. et al. prepared nano-embossed hollow spherical (NeHS) TiO2 particles and investi‐ gated their photovoltaic property in DSC as the bifunctional scattering layer (Koo et al., 2008). The walls of the obtained hollow spheres were composed of nanocrystalline anatase TiO2 particles with mesoporous structure (Figure 14 a-c), which endowed them 5 times high‐ er dye-loading capacity (N719 dye: up to 5.0×10–5 mol/g) than that of the normally used 400 nm-diameter scattering particles with flat surfaces. The NeHS particles scattered incoming light effectively as confirmed by the reflectance spectroscopy (Figure 14 d,e). When the NeHS TiO2 particles were used as the secondary scattering layer in DSC, a substantial im‐ provement in the efficiency has been achieved, with the highest conversion efficiency of 10.34%. Huang F et al. used the submicrometer-sized mesoporous TiO2 beads as the dual function scattering layer for the nanocrystalline electrode, and observed similar results, demonstrating that the dual-function scattering layer represented a powerful alternative for the traditional large-particle scattering layer for high-efficiency DSCs (Huang et al., 2010).

**Figure 13.** a) Schematic of the photoanodes with nanorod scattering layer. b) *J*-V curves of DSCs, and c) UV-VIS spec‐

tra of bare P25 film, and with different nanorod scattering layer. (Fan et al., 2011).

rods in the film electrode were responsible for this improvement.

186 Solar Cells - Research and Application Perspectives

**Figure 14.** SEM image of a) the as-synthesized nano-emboded hollow sphere (NeHS) TiO2 particles, b) high-magnifica‐ tion image for NeHS TiO2 calcined at 450 °C. c) TEM image of a sliced NeHS TiO2 particle. d-e) Diffused reflectance spectra of the nanocrystalline, CCIC, and NeHS TiO2 particulate films d) without and e) with adsorbed N-719 dye. f) *J*-V curve of a DSC based on NeSH TiO2 particulate film as an overlayer on a nanocrystalline TiO2 film under AM 1.5G-one sun light intensity. (Koo et al., 2008).

Below we gave two examples which used the nanostructured scattering layer on the electro‐ des based on nanowire array or nanofibers. Shao et al. demonstrated an interesting double layer photoanode, with the bottom layer of TiO2 nanorod array providing direct conduction pathway for photo-generated electrons, and the upper layer of micro-flowers built up by TiO2 nanobelt, increasing the light harvesting ability as the scattering part (Shao et al., 2011). The cell based on this hierarchical anatase TiO2 exhibited a conversion efficiency of 5.53%, superior than commercial TiO2 (P25). The much higher optical reflectance of the hierarchical structures than the nanorod array and nanoparticle films (Figure 15) proved that this mor‐ phology was beneficial to the light-scattering capacity of the photoanode.

Yang et al. reported an innovative bilayer TiO2 nanofiber photoanode, which combined both smaller (60 nm) and larger (100 nm) diameter TiO2 nanofibers fabricated by electrospinning (Yang et al., 2011). The smaller diameter nanofiber (SNF) layer with a high surface-to-vol‐ ume ratio was used to adsorb sufficiently dye molecules and directly transport electrons re‐ leased from excited dyes. The bigger-diameter nanofi ber (BNF) layer worked as light scattering, adsorbed sufficient dye molecules for energy harvest, and provided higher pore volume in BNF to facilitate electrolyte diffusion for regenerating sensitized dye molecules in the photoanode. Therefore, this bilayer composite nanostructure photoanode offered excel‐ lent dye-loading, lightharvesting, and electron-transport properties. The PCE of DSC was improved from 7.14% for the single-layer to 8.40% for the bilayer of TiO2 nanofiber photoa‐ node, representing an increase of 17%.

**Figure 15.** Schematic of TiO2 nanorod array-micro flower hierarchical photoanode, SEM images of the photoa‐ node at surface and cross section, and reflectance spectra of three kinds of TiO2 photoanode on FTO substrate. (Shao et al., 2011).

**Figure 16.** a) Schematic of the photoanode based on bi-layer electronspinned TiO2 nanofibers. b) TEM image of smaller nanofiber and corresponding SAED pattern. c-d) SEM images of c) smaller and d) bigger nanofibers. e) UV-VIS transmission spectra of smaller diameter and bigger-diameter nanofiber photoanode with the same thickness (Yang et al., 2011).

As the conclusion of this section, we introduce a special scattering layer using up-converting materials. G. B. Shan et al. utilized the hexagonal nanoplatelets of β-NaYF4:Er3+/Yb3+, a typi‐ cal up-conversion phosphor owing to Er3+ doping with Yb3+ codopant assisting in the energy transfer process, as the second layer on nc-TiO2 electrode (Figure 17) (Shan et al., 2011). Ap‐ proximately 10% enhancements of photocurrent and overall DSC efficiency were achieved by the addition of the external layer, which exhibits two functions of light reflecting and near-infrared (NIR) light harvesting. Though the overall PCE of the cell and the improve‐ ment were moderate, the work opened a new opportunity to use some advanced functional materials to enhance the light-harvesting functions of DSC.

In summary, a wide variety of nanostructures including submicron particles, nanowires, nanofibers, mesoporous beads, spheres, nanopalletes etc have been used as the top scatter‐ ing layer of DSC, which are proved to be very effective in improving the LHE and PCE of DSC, especially for the dual-function materials possessing both high surface area and largeparticle size. At the same time, the application of multiple function materials such as the upconverting nanopalletes in the scattering layer sheds a new light for further development.

**Figure 17.** a) Schematic of the DSC device consisting of one internal TiO2 transparent layer plus an external rear layer of β-NaYF4:Er3+/Yb3+ nanoplatelets. b) SEM image and c) up-conversion fluorescence spectrum of the β-NaYF4:Er3+/ Yb3+ nanoplatelets. (Shan et al., 2011).

## **6. Plasmonic DSCs**

**Figure 15.** Schematic of TiO2 nanorod array-micro flower hierarchical photoanode, SEM images of the photoa‐ node at surface and cross section, and reflectance spectra of three kinds of TiO2 photoanode on FTO substrate.

**Figure 16.** a) Schematic of the photoanode based on bi-layer electronspinned TiO2 nanofibers. b) TEM image of smaller nanofiber and corresponding SAED pattern. c-d) SEM images of c) smaller and d) bigger nanofibers. e) UV-VIS transmission spectra of smaller diameter and bigger-diameter nanofiber photoanode with the same thickness

As the conclusion of this section, we introduce a special scattering layer using up-converting materials. G. B. Shan et al. utilized the hexagonal nanoplatelets of β-NaYF4:Er3+/Yb3+, a typi‐ cal up-conversion phosphor owing to Er3+ doping with Yb3+ codopant assisting in the energy transfer process, as the second layer on nc-TiO2 electrode (Figure 17) (Shan et al., 2011). Ap‐ proximately 10% enhancements of photocurrent and overall DSC efficiency were achieved by the addition of the external layer, which exhibits two functions of light reflecting and near-infrared (NIR) light harvesting. Though the overall PCE of the cell and the improve‐ ment were moderate, the work opened a new opportunity to use some advanced functional

In summary, a wide variety of nanostructures including submicron particles, nanowires, nanofibers, mesoporous beads, spheres, nanopalletes etc have been used as the top scatter‐ ing layer of DSC, which are proved to be very effective in improving the LHE and PCE of DSC, especially for the dual-function materials possessing both high surface area and large-

materials to enhance the light-harvesting functions of DSC.

(Shao et al., 2011).

188 Solar Cells - Research and Application Perspectives

(Yang et al., 2011).

Localized surface plasmon resonance (LSPR) behavior is an intriguing characteristic of metal nanoparticles (NPs), which is generated by the resonance between electric fields of electro‐ magnetic waves and free electrons in metal NPs. The use of plasmonic effects has been pro‐ posed as a promising pathway to increase the light absorption in active layers of solar cells, and has been demonstrated on several thin-film solar-cell materials such as amorphous sili‐ con, gallium arsenide, polymers, and DSC. Previous studies on the plasmonic solar cells with thin active layers have proved the beneficial functions of the LSPR on the photocurrent density through such following channels:

**1.** excitation of localized surface plasmon resonances of metallic NPs;


This has established a solid foundation for the rapid development of plasmonic DSCs in recent years.

The mostly-used metal NPs in the plasmonic DSCs are Ag and Au. While earlier studies have found the corrosion of metal particle in the electrolyte or the undesired introduction of electron-hole recombination by the plasmonic particles, it has become a common prac‐ tice to protect the Ag or Au NPs by a thin TiO2 or SiO2 layer. Coupled with the standard thick nc-TiO2 electrode, these plasmonic photoanodes performed fairly well compared with the control samples.

**Figure 18.** a) Schematic of Ag NPs deposited on TiO2 NPs. b) Plasmonic absorption spectra of Ag NPs in Ag/TiO2 (blue solid curve) and TIP-Ag/TiO2 (red solid curve). Each spectrum is decomposed into two absorption peaks depending upon the geometry of Ag NPs. The inset shows a TEM image of the side view of Ag NP on TiO2. Ag┴ and Ag= indicate the direction in the geometry of Ag NP. c) UV-VIS absorption spectra of D-TiO2, D-TIP-Ag/TiO2, and CAg-D-TIP-Ag/TiO2. d) *J*-V and e) IPCE curves of D-TiO2 and D-TIP-Ag/TiO2 photoanodes. TIP-Ag: TiO2 coated Ag NPs; D-TiO2: N719 dye sen‐ sitized TiO2 electrode; CAg-D-TIP-Ag/TiO2: the extinction spectrum after removing the contributions of Ag NPs from the spectrum of D-TIP-Ag/TiO2. (Jeong et al., 2011).

N. C. Jeong et al. provided a detail examination on the effects of the protection of Ag NPs on the LSPR and the cell performance (Jeong et al., 2011). They deposited Ag NPs on TiO2 framework via the photo-reduction of Ag+ from dissolved AgNO3, and protected them by a thin TiO2 layer by refluxing Ag-TiO2 electrode in isopropyl alcohol and subsequent an‐ nealing at 370o C. The electrode incorporating the protected silver NPs (Ag-TiO2) showed enhanced extinction of a subsequently adsorbed dye (the ruthenium-containing molecule, N719), realizing an overall conversion efficiency of 8.9% and 25% improvement over the performance of otherwise identical solar cells incorporating silver NPs lacking protection (Figure 18). Roughly half the improvement could be traced to the increased dye loading by the photoanodes following silver incorporation, with the remaining improvement com‐ ing from the localized surface plasmon resonance (LSPR) enhancement of the effective ab‐ sorption of N719 dye molecules.

**2.** scattering of light by metallic NPs into dielectric-like waveguide modes of the solar cell;

This has established a solid foundation for the rapid development of plasmonic DSCs in

The mostly-used metal NPs in the plasmonic DSCs are Ag and Au. While earlier studies have found the corrosion of metal particle in the electrolyte or the undesired introduction of electron-hole recombination by the plasmonic particles, it has become a common prac‐ tice to protect the Ag or Au NPs by a thin TiO2 or SiO2 layer. Coupled with the standard thick nc-TiO2 electrode, these plasmonic photoanodes performed fairly well compared

**Figure 18.** a) Schematic of Ag NPs deposited on TiO2 NPs. b) Plasmonic absorption spectra of Ag NPs in Ag/TiO2 (blue solid curve) and TIP-Ag/TiO2 (red solid curve). Each spectrum is decomposed into two absorption peaks depending upon the geometry of Ag NPs. The inset shows a TEM image of the side view of Ag NP on TiO2. Ag┴ and Ag= indicate the direction in the geometry of Ag NP. c) UV-VIS absorption spectra of D-TiO2, D-TIP-Ag/TiO2, and CAg-D-TIP-Ag/TiO2. d) *J*-V and e) IPCE curves of D-TiO2 and D-TIP-Ag/TiO2 photoanodes. TIP-Ag: TiO2 coated Ag NPs; D-TiO2: N719 dye sen‐ sitized TiO2 electrode; CAg-D-TIP-Ag/TiO2: the extinction spectrum after removing the contributions of Ag NPs from the

N. C. Jeong et al. provided a detail examination on the effects of the protection of Ag NPs on the LSPR and the cell performance (Jeong et al., 2011). They deposited Ag NPs on TiO2 framework via the photo-reduction of Ag+ from dissolved AgNO3, and protected them by

**3.** coupling to propagating surface plasmon polariton (SPP) modes (Ding et al., 2011).

recent years.

with the control samples.

190 Solar Cells - Research and Application Perspectives

spectrum of D-TIP-Ag/TiO2. (Jeong et al., 2011).

J. Qi et al. coated TiO2 thin layer (~2 nm) on Ag NPs and incorporated this Ag@TiO2 nano‐ structures into the TiO2 photoanode (Qi et al, 2011), which can easily transferred the carriers to surrounding TiO2 NPs in contact with the shell, while providing a good protection for the metal NPs. By utilizing Ag@TiO2 core-shell nanostructures, the optical absorption of dye molecules in solution and in thin film was enhanced by the strong localized electric field generated by LSPs (Figire 19 a,b). By incorporating Ag@TiO2 NPs, the PCE of DSCs with very thin photoanodes (1.5 μm) was increased from 3.1%to 4.4%, and a small amount of Ag@TiO2 NPs (0.1 wt%) improved efficiency from 7.8% to 9.0% while decreasing the photoa‐ node thickness by 25% for improved electron collection (Figure 19c,d).

**Figure 19.** a) Absorption spectra of Ag NPs solutions stabilized by PVP, TiO2 NPs, and Ag@TiO2 NPs. b) Absorption spectra of Ag@TiO2 NPs, dye, and their mixtures in TiO2 film. c) *J*-V curves of plasmonic DSC (Ag/TiO2 = 0.1 wt %, η = 9.0%, *FF* = 67%, 15 μm) and TiO2-only DSC (η = 7.8%, *FF* = 66%, 20 μm). d) IPCE spectra of the DSCs with and without Ag@TiO2. (Qi et al, 2011).

M. D. Brown et al. integrated Au NPs, which exhibited intense absorption due to the surface plasmon resonance in the visible band, into both the liquid and solid state DSCs (Brown et al., 2011). They successfully overcame the detrimental effects of the incorporating "bare" metal NPs into the bulk DSCs, by coating the Au NPs with a thin shell of silica, which could sustain the sintering process of the TiO2 with negligible influence to the optical properties, resist the corrosion from the iodide/triiodie electrolyte, and enhance the photocurrent generation and solar cell performance as a result of light harvesting by the metal NPs. With a significant in‐ crease in the shortcircuit photocurren, the best solid state plasmonic cell sensitized with Z709 exhibited a PCE of 4.0% (Figure 20). The readily tunable optical properties of metallic NPs, through the use of structures such as nanobars, nanostars, or other complex shapes, provided extensive scope for further work to fully optimize the technology.

**Figure 20.** a) Schematic of plasmonic Au-TiO2 structures. b) Absorption spectra of thin films (∼1 μm) processed with nc-TiO2 paste at different fabrication stages, including directly after doctor blade coating (pre-sinter), after two sinter‐ ing cycles at 500<sup>o</sup>C and a TiCl4 treatment (post-sinter), with final sensitization of Z907 (dye), TiO2 paste incorporating bare Au NPs, Au-SiO2 core shell NPs, and nc-TiO2 film sensitized with Z907. c) *J*-V curves of liquid DSC (N719 sensitiza‐ tion, 1.1 μm thickness), with and without Au-SiO2 core (15 nm)-shell (3 nm) nanoparticles. d) Normalized absorption (as 1-transmission-reflection) and spectral response of liquid DSCs. (Brown et al., 2011).

H. Choi et al. investigated the effects of the oxide capping layer of Au nanoparticles on the performance of plasmonic DSCs (Choi et al., 2012). By employing SiO2- and TiO2-capped Au NPs, they improved the cell efficiency from 9.3% for an N719 sensitized device to 10.2% upon incorporation of 0.7% Au@SiO2 and to 9.8% upon loading of 0.7% Au@TiO2 NPs. The plasmonic effect was observed in Au@SiO2 incorporated DSC, which produced higher pho‐ tocurrent (Figure 21). However, in Au@TiO2 incorporated DSC, Au NPs underwent charge equilibration with TiO2 NPs and shifted the apparent Fermi level of the composite to more negative potentials, which resulted in a higher photovoltage. These observations opened up new opportunities to introduce both these paradigms and to synergetically enhance the pho‐ tocurrent and photovoltage of DSC.

M. D. Brown et al. integrated Au NPs, which exhibited intense absorption due to the surface plasmon resonance in the visible band, into both the liquid and solid state DSCs (Brown et al., 2011). They successfully overcame the detrimental effects of the incorporating "bare" metal NPs into the bulk DSCs, by coating the Au NPs with a thin shell of silica, which could sustain the sintering process of the TiO2 with negligible influence to the optical properties, resist the corrosion from the iodide/triiodie electrolyte, and enhance the photocurrent generation and solar cell performance as a result of light harvesting by the metal NPs. With a significant in‐ crease in the shortcircuit photocurren, the best solid state plasmonic cell sensitized with Z709 exhibited a PCE of 4.0% (Figure 20). The readily tunable optical properties of metallic NPs, through the use of structures such as nanobars, nanostars, or other complex shapes, provided

**Figure 20.** a) Schematic of plasmonic Au-TiO2 structures. b) Absorption spectra of thin films (∼1 μm) processed with nc-TiO2 paste at different fabrication stages, including directly after doctor blade coating (pre-sinter), after two sinter‐ ing cycles at 500<sup>o</sup>C and a TiCl4 treatment (post-sinter), with final sensitization of Z907 (dye), TiO2 paste incorporating bare Au NPs, Au-SiO2 core shell NPs, and nc-TiO2 film sensitized with Z907. c) *J*-V curves of liquid DSC (N719 sensitiza‐ tion, 1.1 μm thickness), with and without Au-SiO2 core (15 nm)-shell (3 nm) nanoparticles. d) Normalized absorption

H. Choi et al. investigated the effects of the oxide capping layer of Au nanoparticles on the performance of plasmonic DSCs (Choi et al., 2012). By employing SiO2- and TiO2-capped Au NPs, they improved the cell efficiency from 9.3% for an N719 sensitized device to 10.2% upon incorporation of 0.7% Au@SiO2 and to 9.8% upon loading of 0.7% Au@TiO2 NPs. The

(as 1-transmission-reflection) and spectral response of liquid DSCs. (Brown et al., 2011).

extensive scope for further work to fully optimize the technology.

192 Solar Cells - Research and Application Perspectives

In brief, the plasmonic photoanodes are still a newly-emerged research area in DSC, span‐ ning a wide range of materials (metal, semiconductor) and optical and electrochemical phe‐ nomena (localized surface plasmon resonance, scattering, electron-hole recombination). It is currently the main task for scientists to seek a suitable plasmonic structures (metal-oxide core-shell structures) and effective strategies to incorporate them into the nanocrystalline or nanostructured electrodes. While at present the nanoparticles of Au and Ag have been at‐ tempted to produce plasmonic DSC, further efforts may be extended to other metals (Ti, Zn, etc.) or morphologies with complex structures of metals (cubic, hexagonal, star-form etc.). Also, in-depth explorations on the mechanism of the plasmonic resonance embedded in the nanoporous electrode, and their detail influences on the photo-to-electric conversion proc‐ ess, are very useful to promote the development of this intriguing area.

**Figure 21.** a) IPCE spectra of DSC employing N719 adsorbed onto TiO2, TiO2/Au@TiO2, and TiO2/Au@SiO2 as photo‐ anodes. The loading of core-shell particles is maintained at 0.7%. b) Photovoltaic performance of DSCs. c) Electron equilibration and its influence on the apparent Fermi level (EF): (1) dye TiO2, (2) dye TiO2/Au@-SiO2, and (3) dye TiO2/ Au@TiO2. LSP influence is seen in both (2) and (3), and shift in Fermi level as a result of electron accumulation in the metal core is seen in only (3). (Choi et al., 2012).

## **7. Photonic crystals based photoanodes and more**

Photonic crystals (PCs) are materials that exhibit periodicities in their refractive index on the order of the wavelength of light, and thus provide many interesting possibilities for "photon management" (John, 1987). PC layers can be coupled to the charge-generating layer of DSC, and increase the LHE via such following mechanisms,


Compared with the traditional scattering techniques using geometrical optic-based elements (incoherent scattering layers, or high-reflectivity metallic mirrors), the application of PCs can obtain a more significant improvement on the LHE via the delicate control of the reflect‐ ed, diffracted, or refracted light passing through the cell, apart from the apparent advantage in obtaining the transparent and colorful DSC rather than the opaque ones. The major limi‐ tation for the incorporation of PCs (especially the self-assembled 3D PCs) into DSC is the in‐ compatibility between the fabrication routes for the photonic structures and the photoanode. So in the current stage, the preparation technology of PCs and their coupling to nc-TiO2 layer represent the emphasis of many studies. In this section, we give several exam‐ ples of PC layer in DSC with distinctive features in their structure.

**Figure 22.** a) Cross-sectional SEM images of the photoelectrodes of nc-TiO2 and with different scattering layers. b) IPCE spectra and c) *J–V* curves of the DSCs employing different scattering layers. (Han et al., 2011).

S. Han et al. fabricated quasi-inverse opal (QIO) layers with imperfect periodicity of hollows and applied them as the new scattering layer in DSC (Han et al., 2011). The porous QIO lay‐ ers composed of highly crystalline anatase TiO2 nanocrystals exhibited good dye-adsorptive properties and effective light-scattering properties over the wavelength range of 600–750 nm. The photocurrent of DSC based on the 420 nm-QIO layer was higher than the commer‐ cial scattering layer-based DSSC, achieving an efficiency of 5.7% (Figure 22).

**7. Photonic crystals based photoanodes and more**

**2.** light reflection within the photonic bandgap at various angles, and

ples of PC layer in DSC with distinctive features in their structure.

and increase the LHE via such following mechanisms,

bandgap,

194 Solar Cells - Research and Application Perspectives

Photonic crystals (PCs) are materials that exhibit periodicities in their refractive index on the order of the wavelength of light, and thus provide many interesting possibilities for "photon management" (John, 1987). PC layers can be coupled to the charge-generating layer of DSC,

**1.** photon localization and enhanced red light absorption near the edges of the photonic

**3.** formation of photon resonance modes within the dye-sensitized layer (Yip et al., 2010). Compared with the traditional scattering techniques using geometrical optic-based elements (incoherent scattering layers, or high-reflectivity metallic mirrors), the application of PCs can obtain a more significant improvement on the LHE via the delicate control of the reflect‐ ed, diffracted, or refracted light passing through the cell, apart from the apparent advantage in obtaining the transparent and colorful DSC rather than the opaque ones. The major limi‐ tation for the incorporation of PCs (especially the self-assembled 3D PCs) into DSC is the in‐ compatibility between the fabrication routes for the photonic structures and the photoanode. So in the current stage, the preparation technology of PCs and their coupling to nc-TiO2 layer represent the emphasis of many studies. In this section, we give several exam‐

**Figure 22.** a) Cross-sectional SEM images of the photoelectrodes of nc-TiO2 and with different scattering layers. b)

S. Han et al. fabricated quasi-inverse opal (QIO) layers with imperfect periodicity of hollows and applied them as the new scattering layer in DSC (Han et al., 2011). The porous QIO lay‐

IPCE spectra and c) *J–V* curves of the DSCs employing different scattering layers. (Han et al., 2011).

**Figure 23.** a) Schematic of the cell fabrication process. 1) Formation of PC layer. 2) Formation of NT layer. 3) De‐ tachment of bi-layer from substrate. 4) Gluing to FTO by TiO2 NPs. 5) Assembly of DSC. b) Cross-section SEM im‐ age of the PC layer. Scale bar:1 μ m. c) Reflectance of PC layers with 20 and 40 periods (lattice constant ∼ 150 nm). Insets: photographs of the samples (with 20 periods) in air and infiltrated with ethanol. The green curves are simulated results with lattice constants of 150 and 190 nm for purple and green samples. d) *J*-V curves under AM 1.5 solar light illumination. The axial lattice parameters in the PC layer of the purple and green cells are ∼ 150 and ∼ 190 nm, and the numbers of periods are 30 and 20, respectively. Purple 1 and purple 2 are similar cells with a slight difference in dye loading. (Yip et al., 2012).

To solve a series of problems related to the PC coupled DSCs, including the poor physi‐ cal contact between the PC layer and the TiO2 absorbing layer, and the poor charge trans‐ port due to the nonconductive nature of PC, etc., C. T. Yip et al. developed a seamless PC-TiO2 nanotube assembly (Yip et al., 2012), where the TiO2 nanotube (NT) layer was fabricated by normal electrochemical anodization and the TiO2 PC layer was obtained by a periodic current pulse anodization (Figure 23). Corresponding DSCs showed a 50% effi‐ ciency improvement compared with the cells without a PC layer, due to the enhanced light harvesting of the DSCs in the spectral range corresponding to the photonic bandgap of the PC and a longer wavelength range.

**Figure 24.** a) Scheme of DSC based on the 1D PC exposed to frontal illumination. b) Cross-section SEM image of nc-TiO2 electrode, onto which a TiO2–SiO2 NP multilayer is deposited. c) Magnified view of the SiO2 and TiO2 layers com‐ posing the 1D PC. d) *J*–V curves of nc-TiO2 electrode (7.5 mm thick) coupled to different 1D PCs under one-sun illumination. The lattice parameters are 120±10 nm (open green triangles), and 160±10nm (open red diamonds). Ref‐ erence cell with the same nc-TiO2 electrode is plotted as the open black circle line. Inset: reflectance spectra of the PCbased cell together with the absorption spectrum of the ruthenium dye. e) IPCE spectra for a DSC containing a 7.5 mm thick dye-sensitized TiO2 electrode (solid blue triangles) and for the same electrode coupled to a PC of parameters 120±10nm (solid red circles), measured under frontal illumination and rear illumination (open blue triangles and open red circles, respectively). (Colodrero et al., 2009).

S. Colodrero, et al. couped a porous and highly reflecting 1D photonic crystal into a nc-TiO2 electrode, by the deposition of alternate layers of NPs of different compositions by spincoating (Colodrero et al., 2009). While the porous mesostructure allowed the electrolyte to flow through it and soak the electrode without interfering with the charge transport through the cell, the PC layer with a thickness of just half a micrometer efficiently localized the inci‐ dent light within the nc-dyed TiO2 electrode in a targeted wavelength range (Figure 24). Consequently, the average PCEs were improved to between 15 and 30% of the reference val‐ ue attained for standard electrodes, while the open-circuit voltage and the transparency of the cell remained intact, contrary to what happened when the scattering layers were em‐ ployed to improve the light harvesting.

Apart from the intensive studies on the PCs based DSC, other optical elements such as planar waveguide were also explored to enhance the LHE of DSC. Typically, Z. L. Wang's group demonstrated a novel three-dimensional DSC by integrating planar optical waveguide and nanowires (NWs) (Wei et al., 2010). The ZnO NWs were grown normally to the quartz slide, and the 3D cell was constructed by alternatively stacking a slide and a planar electrode. While the slide served as a planar waveguide for light propagation, the 3D structure effectively increased the light absorbing surface area due to internal multiple reflections without increasing the electron path length to the collecting electrode. Ob‐ tained 3D DSCs exhibited a significant improvement in energy conversion efficiency by a factor of 5.8 compared to the planar illumination case.

**Figure 24.** a) Scheme of DSC based on the 1D PC exposed to frontal illumination. b) Cross-section SEM image of nc-TiO2 electrode, onto which a TiO2–SiO2 NP multilayer is deposited. c) Magnified view of the SiO2 and TiO2 layers com‐ posing the 1D PC. d) *J*–V curves of nc-TiO2 electrode (7.5 mm thick) coupled to different 1D PCs under one-sun illumination. The lattice parameters are 120±10 nm (open green triangles), and 160±10nm (open red diamonds). Ref‐ erence cell with the same nc-TiO2 electrode is plotted as the open black circle line. Inset: reflectance spectra of the PCbased cell together with the absorption spectrum of the ruthenium dye. e) IPCE spectra for a DSC containing a 7.5 mm thick dye-sensitized TiO2 electrode (solid blue triangles) and for the same electrode coupled to a PC of parameters 120±10nm (solid red circles), measured under frontal illumination and rear illumination (open blue triangles and open

S. Colodrero, et al. couped a porous and highly reflecting 1D photonic crystal into a nc-TiO2 electrode, by the deposition of alternate layers of NPs of different compositions by spincoating (Colodrero et al., 2009). While the porous mesostructure allowed the electrolyte to flow through it and soak the electrode without interfering with the charge transport through the cell, the PC layer with a thickness of just half a micrometer efficiently localized the inci‐ dent light within the nc-dyed TiO2 electrode in a targeted wavelength range (Figure 24). Consequently, the average PCEs were improved to between 15 and 30% of the reference val‐ ue attained for standard electrodes, while the open-circuit voltage and the transparency of the cell remained intact, contrary to what happened when the scattering layers were em‐

red circles, respectively). (Colodrero et al., 2009).

196 Solar Cells - Research and Application Perspectives

ployed to improve the light harvesting.

**Figure 25.** a) Schematic architecture of large scale 3D DSSC. The waveguide-NW 3D unit cells are plugged into the counter electrode housing and then sealed and fully packaged. b) *J*-V curves of DSCs under one full sun illumi‐ nation for the parallell to the waveguide surface (PS) configuration. Inset: typical IPCE curves measured for singleside (SS) and double-side (DS) coated DSCs in the PS case. c) Low-magnification SEM image of a quartz slide with uniformly grown ZnO NWs on DS surfaces. d, f) High-magnification SEM images showing the densely packed ZnO NWs on top and bottom surfaces of the slide, respectively. e) Image of a slide coated with grown ZnO nanowire arrays. (Wei et al., 2010).

In summary, the optical elements such as the photonic crystals and the planar waveguide can provide exciting perspective to deliberately modulate the optical path of the incident lights and to significantly enhance the LHE of DSC. However, the ultimate PCE of corre‐ sponding DSCs is still far lower than that of nc-TiO2 counterpart. How to tackle with the preparation problems related to the PC layers, and to avoid the possible adverse effects of the PC layers on the dye-loading and charge-transporting properties of the original elec‐ trode, are the kernel issues which should be solved in near future.

## **8. Conclusion**

In conclusion, the light-harvesting efficiency (LHE) of the photoanode film has determina‐ tive effects on the power conversion efficiency of DSC. The deliberate modulations of the in‐ ternal surface area of the nanoporous electrode and the optical path of the incident light are currently the main pathway to enhance the LHE of DSC. A wide range of novel materials or techniques have been utilized to improve the LHE of the electrode, including the high-sur‐ face area mesoporous nanostructures or aerogel, scattering-enhanced hierarchical nano‐ structures, up-conversion materials, plasmonic core-shell structures, and photonic crystals etc. However, while most of reported work realized obvious enhancement on one or more specific capacities of DSC, such as the dye-loading properties, optical scattering, or im‐ proved harvesting of near-infrared light, very few study can demonstrate high device per‐ formance comparable with the state-of-the-art nc-TiO2 cell. The intrinsically different particle size, microstructures, preparation strategy of these novel materials from the tradi‐ tional nc-TiO2 electrode will inevitably result in significant changes in the microstructure or the optical/ electrical properties of the photoanode, which may greatly impair the final per‐ formance of the device. How to balance the advantageous and disadvantageous factors in‐ volved in these new-type photoanodes, and realize the solid improvement of the overall performance of DSC, are the emphasis of the scientists in near future. After all, the photoan‐ odes based these novel materials or structures are still in an infant stage, containing infinite possibilities to improve or even revolutionize the basic principle and performance of the tra‐ ditional DSC. We believe, via the intensive and extensive efforts of the scientist all over the world and through the collaborative studies among different areas spanning from material, chemistry to physics, the new-type photoanodes will certainly usher a brilliant future.

## **Acknowledgements**

This work is supported by the 973-project (Grant no. 2009CB623304) of Ministry of Science and Technology of China and the Basic Research Program (Grant no. 51072214, 51002174) of National Natural Science Foundation of China.

## **Author details**

Xiang-Dong Gao\* , Xiao-Min Li and Xiao-Yan Gan

\*Address all correspondence to: xdgao@mail.sic.ac.cn

State Key Lab of High Performance Ceramics and Superfine Microstructures, Shanghai Insti‐ tute of Ceramics, Chinese Academy of Sciences, Shanghai, P. R. China

## **References**

**8. Conclusion**

198 Solar Cells - Research and Application Perspectives

**Acknowledgements**

**Author details**

Xiang-Dong Gao\*

National Natural Science Foundation of China.

, Xiao-Min Li and Xiao-Yan Gan

tute of Ceramics, Chinese Academy of Sciences, Shanghai, P. R. China

\*Address all correspondence to: xdgao@mail.sic.ac.cn

In conclusion, the light-harvesting efficiency (LHE) of the photoanode film has determina‐ tive effects on the power conversion efficiency of DSC. The deliberate modulations of the in‐ ternal surface area of the nanoporous electrode and the optical path of the incident light are currently the main pathway to enhance the LHE of DSC. A wide range of novel materials or techniques have been utilized to improve the LHE of the electrode, including the high-sur‐ face area mesoporous nanostructures or aerogel, scattering-enhanced hierarchical nano‐ structures, up-conversion materials, plasmonic core-shell structures, and photonic crystals etc. However, while most of reported work realized obvious enhancement on one or more specific capacities of DSC, such as the dye-loading properties, optical scattering, or im‐ proved harvesting of near-infrared light, very few study can demonstrate high device per‐ formance comparable with the state-of-the-art nc-TiO2 cell. The intrinsically different particle size, microstructures, preparation strategy of these novel materials from the tradi‐ tional nc-TiO2 electrode will inevitably result in significant changes in the microstructure or the optical/ electrical properties of the photoanode, which may greatly impair the final per‐ formance of the device. How to balance the advantageous and disadvantageous factors in‐ volved in these new-type photoanodes, and realize the solid improvement of the overall performance of DSC, are the emphasis of the scientists in near future. After all, the photoan‐ odes based these novel materials or structures are still in an infant stage, containing infinite possibilities to improve or even revolutionize the basic principle and performance of the tra‐ ditional DSC. We believe, via the intensive and extensive efforts of the scientist all over the world and through the collaborative studies among different areas spanning from material, chemistry to physics, the new-type photoanodes will certainly usher a brilliant future.

This work is supported by the 973-project (Grant no. 2009CB623304) of Ministry of Science and Technology of China and the Basic Research Program (Grant no. 51072214, 51002174) of

State Key Lab of High Performance Ceramics and Superfine Microstructures, Shanghai Insti‐


of Nanocrystalline TiO2 Film with High Molar Extinction Coefficient Ruthenium Sen‐ sitizers for High Performance Dye-Sensitized Solar Cells. *J. Am. Chem. Soc.*, 130(32), 10720-10728.


[27] Law, M., Greene, L. E., Johnson, J. C., Saykally, R., & Yang, P. (2005). Nanowire dyesensitized solar cells. *Nature Materials*, 4(6), 455-459.

of Nanocrystalline TiO2 Film with High Molar Extinction Coefficient Ruthenium Sen‐ sitizers for High Performance Dye-Sensitized Solar Cells. *J. Am. Chem. Soc.*, 130(32),

[14] Gratzel, M. (2009). Recent Advances in Sensitized Mesoscopic Solar Cells. *Acc. Chem.*

[15] Hagfeldt, A., Boschloo, G., Sun, L., Kloo, L., & Pettersson, H. (2010). Dye-Sensitized

[16] Hamann, T. W., Martinson, A. B. F., Elam, J. W., Pellin, M. J., & Hupp, J. T. (2008a). Aerogel Templated ZnO Dye-Sensitized Solar Cells. *Adv. Mater.*, 20(8), 1560-1564. [17] Hamann, T. W., Martinson, A. B. F., Elam, J. W., Pellin, M. J., & Hupp, J. T. (2008b). Atomic Layer Deposition of TiO2 on Aerogel Templates: New Photoanodes for Dye-

[18] Han, S. H., Lee, S., Shin, H., & Jung, H. S. (2011). Quasi-Inverse Opal Layer Based on Highly Crystalline TiO2 Nanoparticles: A New Light-Scattering Layer in Dye-Sensi‐

[19] Hore, S., Vetter, C., Kern, R., Smit, H., & Hinsch, A. (2006). Influence of Scattering Layers on Efficiency of Dye-Sensitized Solar Cells. *Sol. Energy Mater. Sol. Cells*, 90(9),

[20] Huang, F., Chen, D., Zhang, X. L., Caruso, R. A., & Cheng, Y. B. (2010). Dual-Func‐ tion Scattering Layer of Submicrometer-Sized Mesoporous TiO2 Beads for High-Effi‐

[21] Huang, C. H., Yang, Y. T., & Doong, R. A. (2011). Microwave-Assisted Hydrothermal Synthesis of Mesoporous Anatase TiO2 via Sol-Gel Process for Dye-Sensitized Solar

[22] Hüsing, N., & Schubert, U. (1998). Aerogels-Airy Materials: Chemistry, Structure,

[23] Jeong, N. C., Prasittichai, C., & Hupp, J. T. (2011). Photocurrent Enhancement by Sur‐ face Plasmon Resonance of Silver Nanoparticles in Highly Porous Dye-Sensitized So‐

[24] Kim, J. Y., Kang, S. H., Kim, H. S., & Sung, Y. E. (2010). Preparation of Highly Or‐ dered Mesoporous Al2O3/TiO2 and Its Application in Dye-Sensitized Solar Cells.

[25] Ko, S. H., Lee, D., Kang, H. W., Nam, K. H., Yeo, J. Y., Hong, S. J., Grigoropoulos, C. P., & Sung, H. J. (2011). Nanoforest of Hydrothermally Grown Hierarchical ZnO Nanowires for a High Efficiency Dye-Sensitized Solar Cell. *Nano Lett.*, 11(2), 666-671.

[26] Koo, H. J., Kim, Y. J., Lee, Y. H., Lee, W. I., Kim, K., & Park, N. G. (2008). Nano-em‐ bossed Hollow Spherical TiO2 as Bifunctional Material for High-Efficiency Dye-Sen‐

ciency Dye-Sensitized Solar Cells. *Adv. Funct. Mater.*, 20(8), 1301-1305.

10720-10728.

1176-1188.

*Res.*, 42(11), 1788-1798.

200 Solar Cells - Research and Application Perspectives

Solar Cells. *Chem. Rev.*, 110(11), 6595-6663.

Sensitized Solar Cells. *J. Phys. Chem. C*, 112(27), 10303-10307.

tized Solar Cells. *Adv. Energy Mater.*, 1(4), 546-550.

Cells. *Microporous Meseporous Mater.*, 142(2-3), 473-480.

and Properties. *Angew. Chem. Int. Ed.*, 37(1-2), 22-45.

lar Cells. *Langmuir*, 27(23), 14609-14614.

sitized Solar Cells. *Adv. Mater.*, 20(1), 195-199.

*Langmuir*, 26(4), 2864-2870.


## **Metal Oxide Nanomaterials, Conducting Polymers and Their Nanocomposites for Solar Energy**

Sadia Ameen, M. Shaheer Akhtar, Minwu Song and Hyung Shik Shin

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/51432

## **1. Introduction**

[39] Shao, F, Sun, J, Gao, L, Yang, S, & Luo, J. (2011). Template-Free Synthesis of Hier‐ archical TiO2 Structures and Their Application in Dye-Sensitized Solar Cells. *ACS*

[40] Tetreault, N., & Grätzel, M. (2012). Novel Nanostructures for Next Generation Dye-

[41] Wei, Y, Xu, C, Xu, S, Li, C, Wu, W, & Wang, Z. L. (2010). Planar Waveguide−Nano‐ wire Integrated Three-Dimensional Dye-Sensitized Solar Cells. *Nano Lett.*, 10(6),

[42] Xu, F., Dai, M., Lu, Y., & Sun, L. (2010). Hierarchical ZnO Nanowire-Nanosheet Ar‐ chitectures for High Power Conversion Efficiency in Dye-Sensitized Solar Cells. *J.*

[43] Yella, A., Lee, H. W., Tsao, H. N., Yi, C., Chandiran, A. K., Nazeeruddin, M. K., Diau, E. W., Diau, G., Yeh, C. Y., Zakeeruddin, S. M., & Grätzel, M. (2011). Porphyrin-Sen‐ sitized Solar Cells with Cobalt (II/III)-Based Redox Electrolyte Exceed 12 Percent Effi‐

[44] Yang, L., & Leung, W. W. F. (2011). Application of a Bilayer TiO2 Nanofiber Photoa‐ node for Optimization of Dye-Sensitized Solar Cells. *Adv. Mater.*, 23(39), 4559-4562.

[45] Yip, C. T., Huang, H., Zhou, L., Xie, K., Wang, Y., Feng, T., Li, J., & Tam, W. Y. (2012). Direct and Seamless Coupling of TiO2 Nanotube Photonic Crystal to Dye-Sensitized

[46] Yum, J. H., Baranoff, E., Wenger, S., Nazeeruddin, M. K., & Grätzel, M. (2011). Pan‐ chromatic Engineering for Dye-Sensitized Solar Cells. *Energy Environ. Sci.*, 4(3),

[47] Yun, T. K., Park, S. S., Kim, D., Hwang, Y. K., Huh, S., Bae, J. Y., & Won, Y. S. (2011). Pore-Size Effect on Photovoltaic Performance of Dye-Sensitized Solar Cells Com‐

[48] Zhang, Q., & Cao, G. Z. (2011). Nanostructured photoelectrodes for dye-sensitized

[49] Zhuge, F., Qiu, J., Li, X., Gao, X., Gan, X., & Yu, W. (2011). Toward Hierarchical TiO2 nanotube Arrays for Efficient Dye-Sensitized Solar Cells. *Adv. Mater.*, 23(11),

posed of Mesoporous Anatase-Titania. *J. Power Sources*, 196(7), 3678-3682.

Solar Cell: A Single-Step Approach. *Adv. Mater.*, 23(47), 5624-5628.

Sensitized Solar Cells. *Energy Environ. Sci.*, 9(5), 8506-8516.

*Appl. Mater. Interfaces*, 3(6), 2148-2153.

*Phys. Chem. C*, 114(6), 2776-2782.

ciency. *Science*, 334(6056), 629-634.

solar cells. *Nano Today*, 6(1), 91-109.

2092-2096.

202 Solar Cells - Research and Application Perspectives

842-857.

1330-1334.

The increasing concern on energy and the global warming due to the depletion of fossil fuel demands to search the alternative renewable energy resources for covering the energy crisis in the coming decade. A very popular renewable source called photovoltaic device is antici‐ pated to solve energy problem, which converts directly the solar energy from sun to the electricity energy. Recently, dye sensitized solar cells (DSSCs) are widely used as promising photovoltaic device owing to its important properties like high solar to electricity energy conversion efficiency, low production cost, ease of fabrication and vast varieties of various semiconducting materials. DSSC is composed of few micrometer-thick nanocrystalline semiconducting oxides thin film combined with monolayer of charge-transfer dye as a photoanode, a redox electrolyte and a platinum metal electrode as counter electrode. In principle, upon illumination, the electron injection to conduction band of semiconductor takes place by the absorption of photons from dye molecules and the redox electrolyte regenerates the oxidized dye by the transportation of charges or ions. These days, the photovoltaic devices are facing inherent drawbacks such as leakage and evaporation problem that limits its practical application. In this regards, efforts are being done to overcome the leakage and evaporation of liquid electrolyte with solid or gel electrolytes such as room temperature molten salts (RTMSs), p-type semiconductor, conducting organic polymers and polymer gel electrolytes. Furthermore, the choice of catalytic in counter component of DSSCs is crucial to improve the reduction rate of I3 to I in the redox electrolyte. In general, the conducting glass electrode without any catalytic materials such as metals, conducting polymers etc shows very low electrocatalytic activity towards the iodide couple electrolyte due to overvoltage and high energy loss. It has been realized that the low resistance and high electrocatalytic materials

© 2013 Ameen et al.; licensee InTech. This is an open access article 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. © 2013 Ameen et al.; licensee InTech. This is a paper 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.

might deliver the better catalyst to avoid the overvoltage and energy loss for the high reduction of I3 in redox electrolyte.

The semiconducting nanomaterials based thin film electrodes with high surface area are of great significance for acquiring the high amount dye adsorption which leads to the higher light harvesting efficiency and photocurrent density. Various metal oxides semiconductors such as TiO2, ZnO and SnO2 have shown good optical and electronic properties and are accepted as the effective photoanode materials for DSSCs. Additionally, the morphology and sizes of metal oxides semiconducting materials, particularly one dimensional (1D) nanostructures like nanorods (NRs), nanowires (NWs) and nanotubes (NTs) based electrodes have shown increased electron transfer during the operation of DSSCs owing to their high surface to volume ratio and arrangements. Until now, the photoanodes with TiO2, SnO2/ZnO, Nb2O5 and ZnO nanomaterials have presented the maximum solar to electricity energy conversion efficiencies of ~11.2%, ~8%, ~2% and ~5% respectively.

On the other hand, the conducting polymers are regarded as promising semiconducting materials due to distinguishable electrical properties, mechanical flexibility, and the relative ease of processing. The conductive polymers such as polypyrrole, poly (3, 4-ethylenedioxy‐ thiophene) and polyaniline (PANI) are frequently used in DSSC as hole transport materials, electron acceptor and electrocatalytic materials for tri-iodide reduction in redox electrolyte. Among them, PANI is an excellent host for the trapping of semiconducting nanomaterials and conducts the electric charges through the polymeric chain due to extended π-electron conju‐ gation. The conducting polymers and dye sensitized metal oxides are good electron donors upon the photo-excitation during the operation of DSSCs.

In this chapter, we have briefly discussed the different conducting polymers, metal oxides and their application for the performance of DSSCs. The chapter includes the brief literature surveys, properties and photovoltaic properties of various metal oxides nanomaterials, nanofillers in polymer electrolytes and the conducting polymers. Additionally, the latest research advancements are surveyed for the development of efficient conducting polymers as p-type semiconducting nanomaterials for counter electrode materials and efficient nanofillers in the solid polymers of DSSCs. Moreover, the doping and the utilization of TiO2 and ZnO nanomaterials for the performance of DSSCs have been discussed in details. It has been seen that the preparation methods, doping, morphologies, and the sizes of conducting polymers and metal oxides have shown the considerable impact on the electrical properties of the nanomaterials and performances of DSSCs. The study also demonstrates the enhanced properties of inorganic metal oxides like ZnO and TiO2 with different sizes and morphologies for achieving the efficient photovoltaic properties of DSSCs such as JSC, VOC, FF and conversion efficiency.

## **2. Types of conducting polymers**

The conducting polymers are composed of π-conjugated polymeric chain and are known as "synthetic metals" [1-2]. These extended π-conjugated systems of conducting polymers have

alternating single and double bonds along the polymeric chain [3]. The conducting polymers display the overlapping of molecular orbital to allow the formation of delocalized molecular wave functions and secondly these molecular orbital must be partially filled so that there is a free movement of electrons in the polymeric structure. The presence of unusual electronic properties such as electrical conductivity, low ionization potential and high electron affinity are associated with the π-electron backbone of the conjugated polymers. These are promising candidates for electronics applications, and offer possible replacements of the conventional metals and inorganic semiconductors [4-5]. The electrical conductivity of the conducting polymers could be altered upon partial oxidation or reduction by a commonly referred process called 'doping'. The electrical conductivity of these polymers could be changed from insulating to metallic by chemical or electrochemical doping and they could be used to produce electronic devices. These polymers have the electrical properties like that of metals, and have attractive characteristics of organic polymers such as light weight, resistance to corrosion, flexibility and lower cost. Additionally, these polymers could be tailor-made to the requirements of the application through modifications in the polymer structure by varying the functional groups in the organic moiety. The commercial applications of conducting polymers are in thin film transistor, batteries, antistatic coatings, electromagnetic shielding, artificial muscles, lightemitting diodes, gas and bio-sensors [6], fuel and solar cells, fillers [7] and corrosion protective coatings [8]. The conducting polymers are easy to synthesize through chemical or electro‐ chemical processes, and their molecular chain structure could be modified conveniently by the copolymerization or structural derivations. Typically, the conducting polymers are of several types as listed below:

## **2.1. Polypyrrole (PPy)**

might deliver the better catalyst to avoid the overvoltage and energy loss for the high reduction

The semiconducting nanomaterials based thin film electrodes with high surface area are of great significance for acquiring the high amount dye adsorption which leads to the higher light harvesting efficiency and photocurrent density. Various metal oxides semiconductors such as TiO2, ZnO and SnO2 have shown good optical and electronic properties and are accepted as the effective photoanode materials for DSSCs. Additionally, the morphology and sizes of metal oxides semiconducting materials, particularly one dimensional (1D) nanostructures like nanorods (NRs), nanowires (NWs) and nanotubes (NTs) based electrodes have shown increased electron transfer during the operation of DSSCs owing to their high surface to volume ratio and arrangements. Until now, the photoanodes with TiO2, SnO2/ZnO, Nb2O5 and ZnO nanomaterials have presented the maximum solar to electricity energy conversion

On the other hand, the conducting polymers are regarded as promising semiconducting materials due to distinguishable electrical properties, mechanical flexibility, and the relative ease of processing. The conductive polymers such as polypyrrole, poly (3, 4-ethylenedioxy‐ thiophene) and polyaniline (PANI) are frequently used in DSSC as hole transport materials, electron acceptor and electrocatalytic materials for tri-iodide reduction in redox electrolyte. Among them, PANI is an excellent host for the trapping of semiconducting nanomaterials and conducts the electric charges through the polymeric chain due to extended π-electron conju‐ gation. The conducting polymers and dye sensitized metal oxides are good electron donors

In this chapter, we have briefly discussed the different conducting polymers, metal oxides and their application for the performance of DSSCs. The chapter includes the brief literature surveys, properties and photovoltaic properties of various metal oxides nanomaterials, nanofillers in polymer electrolytes and the conducting polymers. Additionally, the latest research advancements are surveyed for the development of efficient conducting polymers as p-type semiconducting nanomaterials for counter electrode materials and efficient nanofillers in the solid polymers of DSSCs. Moreover, the doping and the utilization of TiO2 and ZnO nanomaterials for the performance of DSSCs have been discussed in details. It has been seen that the preparation methods, doping, morphologies, and the sizes of conducting polymers and metal oxides have shown the considerable impact on the electrical properties of the nanomaterials and performances of DSSCs. The study also demonstrates the enhanced properties of inorganic metal oxides like ZnO and TiO2 with different sizes and morphologies for achieving the efficient photovoltaic properties of DSSCs such as JSC, VOC, FF and conversion

The conducting polymers are composed of π-conjugated polymeric chain and are known as "synthetic metals" [1-2]. These extended π-conjugated systems of conducting polymers have

of I3 -

efficiency.

**2. Types of conducting polymers**

in redox electrolyte.

204 Solar Cells - Research and Application Perspectives

efficiencies of ~11.2%, ~8%, ~2% and ~5% respectively.

upon the photo-excitation during the operation of DSSCs.

Polypyrrole (PPy) is a versatile polymer of significant properties like redox activity [9] ionexchange, ion discrimination capacities [10], electrochromic effects, charge/discharge process‐ es [11] and exhibits strong absorptive properties towards gases [12], catalytic activity [13] and corrosion protection properties [14]. It is one of the important conducting polymers due to its good electrochemical reversibility between its conducting and insulating states and the ease of preparation through chemical or electrochemical routes [15].

## **2.2. Poly phenylenes) (PP)**

Poly phenylene (PP) is one of the most unusual electro conducting polymers due to the extended planar conjugated π-system, along with high strength and high heat resistance [16]. The most widely used method of PP production is benzene oxidation with a Friedel-Crafts catalyst (the Kovacic method) [17], which yield a polycrystalline powder. Besides, electro‐ chemical polymerization is also a method for PP synthesis, but the molecular weight of the polymer is limited due to its insolubility and chemical defect [18].

### **2.3. Polyacetylene (PA)**

Polyacetylene (PA) is the polymer of highest conductivity as compared to those of conventional metals. PA has the simplest structure of the linear chains of C-H units with alternating single and double bonds [19]. Moreover, the existence of the two hydrogen atoms in its repeat unit offers ample opportunity to decorate the backbone with pendants which perturbs the elec‐ tronic conjugation and influences the molecular alignment of the polymeric chain. Signifi‐ cantly, the proper structural design might tune the backbone-pendant interplay into harmony and synergy, generating new substituted PAs with novel functionalities [20].

## **2.4. Polyazule (PAz)**

The electron-donor and electron-acceptor character of polyazulene (PAz) has been explained by the electron-donor effect of the seven-membered ring toward the five-membered ring. The five-membered ring carries a partial negative charge and the seven-membered ring of azulene carries a partial positive charge. The polymers and its derivatives show high electrical conductivity almost similar as polythiophene, polypyrrole and polyaniline [21].

## **2.5. Polyindole (PIN)**

Polyindol (PIN) is an electroactive polymer which could be obtained by the anodic oxidation of indole in various electrolytes. It is reported that the conductivity of PIN is lower than that of PPy and PANI but its thermal stability is better with respect to PANI and PPy. PIN, a macromolecular compound, is a good candidate for applications in various areas, such as electronics, electrocatalysis, and active materials for anodes of batteries, anticorrosion coatings and pharmacology.

## **2.6. Polycarbazole (PCz)**

Polycarbazole (PCz) among conducting polymers, is attributed with good electroactivity, and useful thermal, electrical and photophysical properties [22]. However, π-π electron system along its backbone imparts rigidity to the polymer and therefore, makes it infusible and poorly processable. The increasing interest in PCz is towards its role as a hole-transport material and an efficient photoluminescence unit [23]. Derivatives of carbazole are easily prepared by the substitution at -N atom and thus, the solubility and functionality of the resulting polymers could be improved. More importantly, the substituted groups might influence the effective conjugation length which is promising materials in making the emitting light in devices.

## **2.7. Polyaniline (PANI)**

Polyaniline (PANI) exhibits the high stability, conductivity and low cost [24-25]. PANI basically undergoes oxidative polymerization in the presence of a protonic acid. Protonation induces an insulator-to-conductor transition, while the number of π-electrons in the chain remains constant. The oxidation and reduction takes place on this –NH– group, and various forms are obtained due to the number of imine and amine segments on the PANI chain. Other substituted aniline like *N*-benzenesulfaniline [26], o-, p- and m-toluidine, o-chloroaniline [27], o-, m- and p-halogenated anilines [28] and 1-Napthylamine are also the subject of current studies and could be used for the semiconducting polymers based electronic applications.

Outof several conductingpolymers,the interestofresearchers inPANI couldpossiblybe linked to the numerous applications that exist for the electronic conducting polymers and also aniline is cheap product and also a very stable material. On the other hand, the nanocomposites of conducting polymers with inorganic semiconducting nanomaterials show the improved mechanical, electrical and thermal properties due to the combined effects of both the semicon‐ ducting nanomaterials and conducting polymers. In particular, PANI nanocomposites display applications on a large scale for various electrochemical, electrorheological and in the electron‐ ic fields such as batteries, sensors, controlling systems and organic displays [29]. The nanocom‐ posites of PANI with cadmium sulphide has been discussed in the next section of the chapter

## **3. Nanocomposites of conducting polymers**

and double bonds [19]. Moreover, the existence of the two hydrogen atoms in its repeat unit offers ample opportunity to decorate the backbone with pendants which perturbs the elec‐ tronic conjugation and influences the molecular alignment of the polymeric chain. Signifi‐ cantly, the proper structural design might tune the backbone-pendant interplay into harmony

The electron-donor and electron-acceptor character of polyazulene (PAz) has been explained by the electron-donor effect of the seven-membered ring toward the five-membered ring. The five-membered ring carries a partial negative charge and the seven-membered ring of azulene carries a partial positive charge. The polymers and its derivatives show high electrical

Polyindol (PIN) is an electroactive polymer which could be obtained by the anodic oxidation of indole in various electrolytes. It is reported that the conductivity of PIN is lower than that of PPy and PANI but its thermal stability is better with respect to PANI and PPy. PIN, a macromolecular compound, is a good candidate for applications in various areas, such as electronics, electrocatalysis, and active materials for anodes of batteries, anticorrosion coatings

Polycarbazole (PCz) among conducting polymers, is attributed with good electroactivity, and useful thermal, electrical and photophysical properties [22]. However, π-π electron system along its backbone imparts rigidity to the polymer and therefore, makes it infusible and poorly processable. The increasing interest in PCz is towards its role as a hole-transport material and an efficient photoluminescence unit [23]. Derivatives of carbazole are easily prepared by the substitution at -N atom and thus, the solubility and functionality of the resulting polymers could be improved. More importantly, the substituted groups might influence the effective conjugation length which is promising materials in making the emitting light in devices.

Polyaniline (PANI) exhibits the high stability, conductivity and low cost [24-25]. PANI basically undergoes oxidative polymerization in the presence of a protonic acid. Protonation induces an insulator-to-conductor transition, while the number of π-electrons in the chain remains constant. The oxidation and reduction takes place on this –NH– group, and various forms are obtained due to the number of imine and amine segments on the PANI chain. Other substituted aniline like *N*-benzenesulfaniline [26], o-, p- and m-toluidine, o-chloroaniline [27], o-, m- and p-halogenated anilines [28] and 1-Napthylamine are also the subject of current studies and could be used for the semiconducting polymers based electronic applications.

and synergy, generating new substituted PAs with novel functionalities [20].

conductivity almost similar as polythiophene, polypyrrole and polyaniline [21].

**2.4. Polyazule (PAz)**

206 Solar Cells - Research and Application Perspectives

**2.5. Polyindole (PIN)**

and pharmacology.

**2.6. Polycarbazole (PCz)**

**2.7. Polyaniline (PANI)**

## **3.1. Nanocomposites of PANI and cadmium sulphide**

Cadmium Sulphide (CdS)is a semiconductor with a direct band gap of ~2.42 eV which displays superior optical,photophysical andphotochemicalproperties [30].Thenanocomposites of CdS andPANIhavepresentedthe effective electrodematerials formanyelectrochemical,photoelec‐ trochemical, sensing and electrochromic devices [31]. The nanocomposites are anticipated as effective and promising electrode materials in many electrochemical devices. Xi et al studied the influence of optical and absorption properties of CdS by the incorporation of CdS into PANI matrix [32]. R. Seoudi et al studied the dependence of structural, vibrational spectroscopy and opticalpropertiesontheparticlesizesofPANI/CdSnanocomposites [33].B.T.Rautetalreported the novel method of fabrication of PANI/CdS nanocomposites and studied the structural, morphological and optoelectronic properties [34]. In this regards, Ameen et al has reported a simple solution method to synthesize the CdS decorated PANI nanorods (NRs) and studied the electrochemical impedance properties of the nanocomposites [35].

The synthesized PANI NRs exhibit the entangled network with diameter of ~40-50 nm and length of several hundred nanometers, as shown in Fig. 1 (a). The uniform decoration and the thicknesses of CdS-PANI NRs increase gradually with the increase of CdCl2 concentration. After sensitization with the highest concentration of CdCl2 (0.1 M), the surface of PANI NRs (Fig. 1(b)) is completely decorated by CdS nanoparticles which results in the enhanced diameter of ∼60-70 nm. The TEM characterization (Fig. 1 (c and d)) clearly justifies the decoration of CdS nanomaterials on the surface of PANI NRs and shows the increased thickness of PANI NRs with the average diameter of ~60-70 nm due to the decoration of CdS nanoparticles with highest concentration of CdCl2 (0.1 M). From the EDS studies, the overall CdS contents have been estimated as 0.34, 0.53 and 1.08 wt% in the synthesized CdS-PANI NRs with CdCl2 concentrations of 0.01 M, 0.05 M and 0.1 M respectively.

The Raman bands at ∼1175 cm−1, ∼1507 cm−1, ∼1595 cm−1 are observed in all the samples of CdS-PANI NRs (Fig. 2), corresponding to the C-H bending vibration of the semi quinonoid rings (cation-radical segments), N-H deformation vibration associated with the semiquinonoid structures and C=C stretching vibration in the quinonoid ring respectively [36]. PANI NRs show a relatively high band at ∼1368 cm−1 in the spectrum which provides the information of

**Figure 1.** FESEM (a, b) and TEM (c, d) images of synthesized PANI NRs and CdS-PANI NRs. Reprinted with permission from [Ameen S., 2012], Chem. Eng. J. 181 (2012) 806 ©2012, Elsevier Ltd.

the C-N+• vibration of delocalized polaronic structures [37]. The absence of this band in CdS-PANI NRs might due to the efficient interaction of imine (-NH) group of PANI with CdS nanomaterials.

The UV-Vis absorption spectra of the PANI NRs and CdS-PANI NRs are depicted in Fig. 3(a). In PANI NRs, the broad absorption bands at ~617 nm is related to n-π\* transition, and the absorption peak at ~268 nm and ~327 nm arises due to π-π\* transition within the benzenoid segment which is associated to the extent of conjugation between adjacent phenyl rings in the PANI [38]. On comparison to PANI NRs, the red shifting are seen and absorption bands move to higher wavelength of ~279 nm, ~338 nm and ~630 nm respectively in CdS-PANI NRs due to the sensitization of CdS nanomaterials with PANI NRs. The red shift of absorption bands with high intensities reveals that PANI NRs might form a partial bond with CdS nanoparticles. The room temperature photoluminescence (PL) spectra (Fig. 3(b)) of PANI NRs and CdS-PANI NRs exhibit a single large amplitude band in the blue green region which originated due to the π–π\* transition of the benzenoid unit of PANI [39]. The sensitization of PANI NRs with the highest concentration of CdCl2 (0.1 M) causes a significant red shift from ~421 nm to ~438 nm as compared to the PANI NRs which might occur by the chemical interaction between -

Metal Oxide Nanomaterials, Conducting Polymers and Their Nanocomposites for Solar Energy http://dx.doi.org/10.5772/51432 209

**Figure 2.** Raman Spectra of PANI NRs and CdS-PANI NRs. Reprinted with permission from [Ameen S., 2012], Chem. Eng. J.181 (2012) 806 ©2012, Elsevier Ltd.

the C-N+• vibration of delocalized polaronic structures [37]. The absence of this band in CdS-PANI NRs might due to the efficient interaction of imine (-NH) group of PANI with CdS

**Figure 1.** FESEM (a, b) and TEM (c, d) images of synthesized PANI NRs and CdS-PANI NRs. Reprinted with permission

from [Ameen S., 2012], Chem. Eng. J. 181 (2012) 806 ©2012, Elsevier Ltd.

208 Solar Cells - Research and Application Perspectives

The UV-Vis absorption spectra of the PANI NRs and CdS-PANI NRs are depicted in Fig. 3(a). In PANI NRs, the broad absorption bands at ~617 nm is related to n-π\* transition, and the absorption peak at ~268 nm and ~327 nm arises due to π-π\* transition within the benzenoid segment which is associated to the extent of conjugation between adjacent phenyl rings in the PANI [38]. On comparison to PANI NRs, the red shifting are seen and absorption bands move to higher wavelength of ~279 nm, ~338 nm and ~630 nm respectively in CdS-PANI NRs due to the sensitization of CdS nanomaterials with PANI NRs. The red shift of absorption bands with high intensities reveals that PANI NRs might form a partial bond with CdS nanoparticles. The room temperature photoluminescence (PL) spectra (Fig. 3(b)) of PANI NRs and CdS-PANI NRs exhibit a single large amplitude band in the blue green region which originated due to the π–π\* transition of the benzenoid unit of PANI [39]. The sensitization of PANI NRs with the highest concentration of CdCl2 (0.1 M) causes a significant red shift from ~421 nm to ~438 nm as compared to the PANI NRs which might occur by the chemical interaction between -

nanomaterials.

**Figure 3.** UV-vis absorption spectra (a) and photoluminescence spectra (b) of PANI NRs and CdS-PANI NRs. Reprinted with permission from [Ameen S., 2012], Chem. Eng. J.181 (2012) 806 ©2012, Elsevier Ltd.

NH groups of the PANI chains and surface of CdS. The CdS-PANI NRs sensitized with 0.1 M CdCl2, shows the lowest PL intensity and the highest peak shift, suggesting the large π–π\* transition of the benzenoid unit and the strong chemical interaction between -NH groups of the PANI chains and surface of CdS.

The X-rays Photoelectron Spectroscopy (XPS) has studied to examine the interaction between CdS nanoparticles and the PANI NRs, as shown in Fig. 4. The C 1s XPS spectrum (Fig. 4 (a)) of CdS-PANI NRs shows the center peak at ∼284.0 eV with five resolved peaks at the binding energies spanning the range from ∼288 to ∼283 eV. The strong peak at ∼ 283.4 eV represents the carbon (C) of benzonoid ring in which a combination of protonation of imine and amine sites are formed via shake-up processes [40]. The next three resolved peaks at ∼284.8 eV, ∼285.7 eV and ∼286.8 eV confirm the origin of the neutral C-C/C-H bond-PANI backbone, C-N+ /C=N <sup>+</sup> bond and C=O/C-O bond (might occur due to the absorption of moisture on the CdS-PANI), respectively [41]. The resolved peaks at ∼287.8 eV assigns to the π-π\* bonding in a long-range order with a polymer chain shake-up satellite structure and coincides with the doped states. On comparison with typical PANI peak [42], C 1s peak has shifted backwardly, suggesting that C toms of PANI is interacted with other materials (CdS, TiO2 etc.) or impurities [43]. The O 1s XPS spectrum (Fig. 4(c)) exhibits the center peak at ∼530.1 eV with three resolved peaks

**Figure 4.** (a) C 1s, (b) N 1s, (c) O 1s, (d) Cd 3d and (e) S 2p XPS spectra of CdS-PANI NRs. Reprinted with permission from [Ameen S., 2012], Chem. Eng. J.181 (2012) 806© 2012, Elsevier Ltd.

at ∼530.7, ∼531.8 and ∼532.7 eV, suggesting the absorption of moisture and some oxygen impurities on the surface of CdS-PANI NRs during the synthesis. In N 1s XPS spectra (Fig. 4(b)), the main center peak with binding energy at ∼400.5 eV and a resolved peak with lower binding energy at ∼399.4 eV attribute to nitrogen atom originated from benzenoid diamine and quinoid di-imine nitrogen of PANI respectively. The other binding energies of ∼401.7 and ∼402.6 eV are ascribed to the positively charged nitrogens i.e. oxidized amine ( N+ ) and protonated imine ( N+ ) respectively [44]. The positive shifting is seen in the binding energies at ∼401.7 eV and ∼402.6 eV as compared with N 1s spectrum of pristine PANI indicating the involvement of positively charged nitrogen or protonated nitrogen for the partial bonding between PANI and CdS. Furthermore, the singlet peak at ∼404.01 eV observes in Cd 3d XPS (Fig. 4(d)) spectrum, corresponding to Cd 3d5/2 and the typical peak of Cd+2 atoms in CdS [45]. Fig. 4(e) presents the S 2p XPS spectrum of CdS-PANI NRs and observed one distinct peak of S2p3/2 at ∼161.9 eV, corresponds to S−2 of CdS nanoparticles. This suggests the interaction and bonding between CdS nanomaterials and PANI molecules. Thus, it is concluded that the PANI and CdS nanomaterials are partially interacted and bonded by two charged nitrogen species (N+ and N+ ) of PANI with CdS nanomaterials.

Fig. 5 shows the Nyquist plot of EIS measurement for PANI NRs and CdS-PANI NRs electrodes in the electrolyte (LiI, I2 and LiClO4 in ethanol) at a frequency range from 100 kHz to 1 Hz. The almost same RS (electrolyte resistance) with a depressed semi circle in the high frequency region is observed for the all the samples. The presence of depressed semi circle plot is ascribed to the parallel combination of the charge transfer resistance (RCT) of the electrochemical reaction and the double layer capacitance (Cdl) of the PANI film/electrolyte interface [46] has been removed. As shown in Fig. 5, the PANI NRs electrode displays large RCT value of ~17 kΩ. The RCT values drastically decreases by the incremental addition of CdCl2 in PANI NRs, and the order of RCT values are measured as 8.4 kΩ (CdCl2 (0.01 M)-PANI NRs) < 5.7 kΩ (CdCl2 (0.05 M)-PANI NRs) < 4.08 kΩ (CdCl2 (0.10 M)-PANI NRs). Usually, the electrode with large RCT leads the slow charge transfer rate of the electrochemical system [46]. It could be seen that the highest RCT value of PANI NRs electrode might result the low charge transfer rate to the electrochemical system. Moreover, the CdS decorated PANI NRs electrode with 0.01 M CdCl2 presents lowest RCT value which delivers the higher charge transfer rate. The variation in charge transfer resistance (RCT) after the CdS sensitization by different concentrations of CdCl2 might attribute to the changes in the NRs structure. Considerably, the direct band gap of CdS nanoparticles might also affect which improve the electronic state like polarons and bipolarons of PANI for the high charge carriers and enhance the charge transfer. Therefore, the CdS-PANI NRs is potential and cost effective electrode materials for the fabrication of efficient electrochemical (sensor, field emission transistor), photoelectrochemical and photo‐ voltaic devices.

## **4. Basic structure and kinetics of DSSCs**

NH groups of the PANI chains and surface of CdS. The CdS-PANI NRs sensitized with 0.1 M CdCl2, shows the lowest PL intensity and the highest peak shift, suggesting the large π–π\* transition of the benzenoid unit and the strong chemical interaction between -NH groups of

The X-rays Photoelectron Spectroscopy (XPS) has studied to examine the interaction between CdS nanoparticles and the PANI NRs, as shown in Fig. 4. The C 1s XPS spectrum (Fig. 4 (a)) of CdS-PANI NRs shows the center peak at ∼284.0 eV with five resolved peaks at the binding energies spanning the range from ∼288 to ∼283 eV. The strong peak at ∼ 283.4 eV represents the carbon (C) of benzonoid ring in which a combination of protonation of imine and amine sites are formed via shake-up processes [40]. The next three resolved peaks at ∼284.8 eV, ∼285.7 eV and ∼286.8 eV confirm the origin of the neutral C-C/C-H bond-PANI backbone, C-N+

<sup>+</sup> bond and C=O/C-O bond (might occur due to the absorption of moisture on the CdS-PANI), respectively [41]. The resolved peaks at ∼287.8 eV assigns to the π-π\* bonding in a long-range order with a polymer chain shake-up satellite structure and coincides with the doped states. On comparison with typical PANI peak [42], C 1s peak has shifted backwardly, suggesting that C toms of PANI is interacted with other materials (CdS, TiO2 etc.) or impurities [43]. The O 1s XPS spectrum (Fig. 4(c)) exhibits the center peak at ∼530.1 eV with three resolved peaks

**Figure 4.** (a) C 1s, (b) N 1s, (c) O 1s, (d) Cd 3d and (e) S 2p XPS spectra of CdS-PANI NRs. Reprinted with permission

from [Ameen S., 2012], Chem. Eng. J.181 (2012) 806© 2012, Elsevier Ltd.

/C=N

the PANI chains and surface of CdS.

210 Solar Cells - Research and Application Perspectives

Last few decades, considerable researches on DSSCs have extensively explored in terms of both fundamental and applied viewpoints. The basic components of DSSCs involves the

**Figure 5.** Nyquist plots of PANI NRs and CdS–PANI NRs at a frequency range from 100 kHz to 1 Hz. Reprinted with permission from [Ameen S., 2012], Chem. Eng. J.181 (2012) 806 ©2012, Elsevier Ltd.

conducting fluorine doped tin oxide (FTO) glass, sensitized dye, titania nanoparticles, electrolyte and Pt deposited FTO glass. The working principle of DSSC, involves the adsorp‐ tion of photons by dye molecules upon light illumination and the injection of electrons from their excited states into the conduction band of the TiO2 nanoparticles. During the entire process, the oxidized dye molecules are recharged by a redox electrolyte, which transports the positive charges by diffusion to a Pt counter electrode. The low absorption coefficient of a dye monolayer is compensated by the mesoporous structure of the TiO2 film, which leads to a strong increase in the number of TiO2/dye/electrolyte interfaces through which photons pass, thus increases the absorption probability. The following steps are in photoelectric chemical mechanism process of DSSC:

**i.** TiO2|D + hν → TiO2|D\* Excitation

**ii.** TiO2|D\* → TiO2|D+ + ē (cb) Injection

**iii.** TiO2|2D+ + 3I– → TiO2|2S + I<sup>3</sup> – Regeneration

$$\text{i.v.}\qquad \text{I}\_3\text{--}+2 \text{ è (Pt)}\rightarrow 3\text{I}^\cdot \text{Reduction}$$

**v.** I<sup>3</sup> – + 2 ē (cb) → 3I– Recaption (dark reaction)

**vi.** TiO2|S+ + ē (cb) → TiO2|S Recombination (dark reaction)

For the efficient working of DSSCs, the rate of re-reduction of the oxidized dye must be higher than the rate of back reaction of the injected electrons with the dye as well as the rate of reaction of injected electrons with the electron acceptor in the electrolyte. The kinetics of the reaction at the counter electrode and mesoscopic semiconductor materials with an enormous internal surface area to absorb more incident light via dye as sensitizers determines the fast regenera‐ tion of charge mediator performance [47]. Apart from this, the other significant parameters which influence the performance of DSSCs are the mesoporous morphology with high surface area of semiconducting electrode to allow absorption of a larger amount of dye and the efficient charge carriers transport at the interface of photoanode and counter electrode by the semi‐ conducting electrode. Moreover, the oxidized dye should be reduced to its ground state rapidly, after the injection of photoexcited electron from dye into the conduction band of semiconducting electrode. Furthermore, the semiconducting electrode must be able to permit the fast diffusion of charge carriers (higher conductivity) and produces good interfacial contact with the porous nanocrystalline layer and the counter electrode, the long-term stability, including chemical, thermal, optical, electrochemical, and interfacial stability of the electrolyte, which does not cause the desorption and degradation of the dye from the oxide surface and lastly, the optimized concentration of I– /I3 – which could reduce the visible light absorption by the dye, and the efficient reaction of I3 – ions with the injected electrons to increase the dark current.

## **5. DSSCs based on conducting polymers**

### **5.1. PANI as hole transport materials for DSSCs**

conducting fluorine doped tin oxide (FTO) glass, sensitized dye, titania nanoparticles, electrolyte and Pt deposited FTO glass. The working principle of DSSC, involves the adsorp‐ tion of photons by dye molecules upon light illumination and the injection of electrons from their excited states into the conduction band of the TiO2 nanoparticles. During the entire process, the oxidized dye molecules are recharged by a redox electrolyte, which transports the positive charges by diffusion to a Pt counter electrode. The low absorption coefficient of a dye monolayer is compensated by the mesoporous structure of the TiO2 film, which leads to a strong increase in the number of TiO2/dye/electrolyte interfaces through which photons pass, thus increases the absorption probability. The following steps are in photoelectric chemical

**Figure 5.** Nyquist plots of PANI NRs and CdS–PANI NRs at a frequency range from 100 kHz to 1 Hz. Reprinted with

+ ē (cb) Injection

–

Recaption (dark reaction)

+ ē (cb) → TiO2|S Recombination (dark reaction)

Regeneration

For the efficient working of DSSCs, the rate of re-reduction of the oxidized dye must be higher than the rate of back reaction of the injected electrons with the dye as well as the rate of reaction of injected electrons with the electron acceptor in the electrolyte. The kinetics of the reaction

mechanism process of DSSC:

212 Solar Cells - Research and Application Perspectives

**ii.** TiO2|D\* → TiO2|D+

+ 2 ē (Pt) → 3I–

+ 2 ē (cb) → 3I–

**iii.** TiO2|2D+

**vi.** TiO2|S+

–

–

**iv.** I<sup>3</sup>

**v.** I<sup>3</sup>

**i.** TiO2|D + hν → TiO2|D\* Excitation

+ 3I– → TiO2|2S + I<sup>3</sup>

Reduction

permission from [Ameen S., 2012], Chem. Eng. J.181 (2012) 806 ©2012, Elsevier Ltd.

The DSSCs and polymer solar cell have been exploring the new approaches in the design of both active materials and device architectures [48]. J. Wagner et al reported the new carbazolebased polymers for DSSCs with hole-conducting polymer [49]. N. Kudo et al fabricated the organic-inorganic hybrid solar cells based on conducting polymer and SnO2 nanoparticles which were chemically modified with a fullerene derivative [50]. S. Woo et al reported the hybrid solar cells with conducting polymers and vertically aligned silicon nanowire arrays and studied the effects of silicon conductivity [51]. F. Tan et al synthesized PANI films by electrodepostied methods and applied as anode buffer layers in solar cells [52]. M. Y. Chang et al fabricated the polymer solar cells by incorporating one-dimensional polyaniline nano‐ tubes [53]. T. H. Lim et al utilized PANI for a flexible organic solar cells anode [54]. H. Bejbouji et al reported PANI as a hole injection layer on organic photovoltaic cells [55]. S. Zhu et al synthesized the hybrid structure of PANI/ZnO nanograss for the application in dye-sensitized solar cell1 [56]. PANI is also known as large band gap hole transporting material (HMTs) which could easily deposit as thin film on several substrates. In this regards, Ameen et al has reported the application of PANI as photoelectrode using N710 and Z907 as sensitizers for the per‐ formance of DSSCs [57].

The morphologies of ZnO nanoparticles, PANI/N719/ZnO and PANI/Z907/ZnO thin films exhibit well crystalline ZnO nanoparticles of size ~30-40 nm. The size of ZnO nanoparticles increases by ~10-15 nm from their original particle sizes after the plasma enhanced chemical vapor deposition (PECVD) polymerization of PANI molecule on dye (N719 and Z907) sensitized ZnO nanoparticulate thin film. PANI/Z907/ZnO thin film displays uniform covering or coating of PANI, indicating the well penetration of PANI on the surface of Z907/ZnO nanoparticulate thin film. The arrangement of ZnO/PANI nanoparticles are more pronounced for PANI/Z907/ZnO thin films might due to the substantive interaction and the incorporation of Z907 into ZnO nanoparticulate thin film which might allow the uniform deposition and well penetration of PANI through PECVD process. Likewise, the TEM images of ZnO nanoparticles and PANI/Z907/ZnO thin film electrode, again confirm the enhancement in the size of ZnO nanoparticles after PECVD polymerization of PANI. Significantly, after PECVD deposition of PANI, the aggregation of nanoparticles enhances the size of ZnO nanoparticles [57].

UV-Vis spectroscopy is investigated to describe the optical properties of PANI and PANI/ZnO thin film. The PECVD polymerized PANI exhibits the characteristic absorption bands at ~273 nm and ~345 nm which are ascribed to π–π\* transitions. However, the broad band at ~611 nm is referred to *n*-π\* transitions which provides the information of the polarons formation into the conducting PANI. The UV-visible spectrum of PANI/ZnO thin film shows the clear red shifts with the absorption bands at ~299 nm and ~628 nm from ~273 and ~611 nm, indicating the interference in the absorption bands of PANI by ZnO nanoparticles. These shifts in the peaks are usually associated to the interactions between ZnO and PANI in PANI-ZnO thin film which might due to the existence of partial hydrogen bonding between NH (PANI)..O– Zn (metal oxide) [58].

The room temperature PL spectra of PECVD polymerized PANI shows two absorption bands in blue-green region at ~438 nm and ~642 nm. The significant higher absorption wavelength shift at ~452 nm with slightly decreased peak intensity is observed in the PANI/ZnO thin film. The considerable changes in the PL emission peak might arise due to the effective interaction between imine (-NH) of PANI and hydroxyl (-OH) group of ZnO nanoparicles in PECVD polymerized PANI-ZnO thin film [57].

The current density-voltage (J-V) characteristics at one sun light illumination (100 mW/cm2 , 1.5AM) have been carried out to evaluate the performances of solar cell fabricated with PANI/ N719/ZnO and PANI/Z907/ZnO thin film electrode. The measurements of VOC, JSC, FF and overall solar-to-electrical energy conversion efficiency are obtained from the J-V characteristics of both the DSSCs. The PANI/N719/ZnO electrode based DSSC exhibits low solar-to-electricity conversion efficiency of ~0.6%, with JSC of ~2.80 mA/cm2 , VOC ~0.432 V and FF ~0.51, whereas, PANI/Z907/ZnO electrode based DSSC executes the greater overall solar-to-electricity conversion efficiency of ~1.31% with VOC of JSC ~4.56 mA/cm2 , ~0.521 V, and FF ~0.55. On comparison with PANI/N719/ZnO electrode, DSSC with PANI/Z907/ZnO electrode attains considerably improved the solar-to-electricity conversion efficiency by ~53% along with other parameters of J-V characteristics. The enhanced performances and JSC might attribute to the fast movements of photon generated electrons at the interface of the PANI/ZnO and the nature of Ru dye (Z907) with long chain alkyl group at pyridine rings. Moreover, as seen in FESEM results, the high penetration of the hole conductor (PANI) into the pores of Z907 sensitized ZnO thin film might execute reasonably fast charge injection and electron transfer at the interface of PANI and ZnO layer to Pt layer of electrode. Thus, the choice of dye is crucial to obtain the high performance DSSC with PECVD polymerized PANI/ZnO electrodes. In other previous report, Ameen et al showed the effects of PANI on TiO2 as an effective photoelectrode for the performance of DSSCs [59].

or coating of PANI, indicating the well penetration of PANI on the surface of Z907/ZnO nanoparticulate thin film. The arrangement of ZnO/PANI nanoparticles are more pronounced for PANI/Z907/ZnO thin films might due to the substantive interaction and the incorporation of Z907 into ZnO nanoparticulate thin film which might allow the uniform deposition and well penetration of PANI through PECVD process. Likewise, the TEM images of ZnO nanoparticles and PANI/Z907/ZnO thin film electrode, again confirm the enhancement in the size of ZnO nanoparticles after PECVD polymerization of PANI. Significantly, after PECVD deposition of

UV-Vis spectroscopy is investigated to describe the optical properties of PANI and PANI/ZnO thin film. The PECVD polymerized PANI exhibits the characteristic absorption bands at ~273 nm and ~345 nm which are ascribed to π–π\* transitions. However, the broad band at ~611 nm is referred to *n*-π\* transitions which provides the information of the polarons formation into the conducting PANI. The UV-visible spectrum of PANI/ZnO thin film shows the clear red shifts with the absorption bands at ~299 nm and ~628 nm from ~273 and ~611 nm, indicating the interference in the absorption bands of PANI by ZnO nanoparticles. These shifts in the peaks are usually associated to the interactions between ZnO and PANI in PANI-ZnO thin film which might due to the existence of partial hydrogen bonding between NH (PANI)..O–

The room temperature PL spectra of PECVD polymerized PANI shows two absorption bands in blue-green region at ~438 nm and ~642 nm. The significant higher absorption wavelength shift at ~452 nm with slightly decreased peak intensity is observed in the PANI/ZnO thin film. The considerable changes in the PL emission peak might arise due to the effective interaction between imine (-NH) of PANI and hydroxyl (-OH) group of ZnO nanoparicles in PECVD

The current density-voltage (J-V) characteristics at one sun light illumination (100 mW/cm2

1.5AM) have been carried out to evaluate the performances of solar cell fabricated with PANI/ N719/ZnO and PANI/Z907/ZnO thin film electrode. The measurements of VOC, JSC, FF and overall solar-to-electrical energy conversion efficiency are obtained from the J-V characteristics of both the DSSCs. The PANI/N719/ZnO electrode based DSSC exhibits low solar-to-electricity

PANI/Z907/ZnO electrode based DSSC executes the greater overall solar-to-electricity

comparison with PANI/N719/ZnO electrode, DSSC with PANI/Z907/ZnO electrode attains considerably improved the solar-to-electricity conversion efficiency by ~53% along with other parameters of J-V characteristics. The enhanced performances and JSC might attribute to the fast movements of photon generated electrons at the interface of the PANI/ZnO and the nature of Ru dye (Z907) with long chain alkyl group at pyridine rings. Moreover, as seen in FESEM results, the high penetration of the hole conductor (PANI) into the pores of Z907 sensitized ZnO thin film might execute reasonably fast charge injection and electron transfer at the interface of PANI and ZnO layer to Pt layer of electrode. Thus, the choice of dye is crucial to obtain the high performance DSSC with PECVD polymerized PANI/ZnO electrodes. In other

,

, VOC ~0.432 V and FF ~0.51, whereas,

, ~0.521 V, and FF ~0.55. On

PANI, the aggregation of nanoparticles enhances the size of ZnO nanoparticles [57].

Zn (metal oxide) [58].

214 Solar Cells - Research and Application Perspectives

polymerized PANI-ZnO thin film [57].

conversion efficiency of ~0.6%, with JSC of ~2.80 mA/cm2

conversion efficiency of ~1.31% with VOC of JSC ~4.56 mA/cm2

A schematic energy level diagram for the device FTO/TiO2/Dye/PANI/Pt is shown in Fig. 6 (a). The diagram, in accordance with step (1) indicates that the electrons from the dye, upon illumination, jumps from the HOMO level to the LUMO level and thus, as per step (2), these electrons are transferred from the conduction band (C.B) to the valence band (V.B) of TiO2. The step (3) shows the transfer of electrons from V.B to the HOMO of PANI. As indicated by step (4), further transfer of electrons could be preceded through two different possible ways. Firstly, the electron could move either through the LUMO of PANI, followed by step (5) or then may jump to the LUMO of the dye and finally move onwards by repeating the step (2). Secondly, electrons, from step (3), could also jump to the HOMO of dye and would move ahead by following the same step (1), leading to the transfer of electrons to the entire cell. During the entire cycle, the recovery of the holes is accomplished at the counter electrode. Additionally, Fig. 6(b) depicts that PANI participates in the light absorption through the effective injection of electrons from its LUMO to the C.B of TiO2. Therefore, the proposed mechanism presents that FTO/TiO2/Dye/PANI/Pt system might deliver the high transportation of charge carriers during the operation of device under the illumination.

**Figure 6.** An overview of the energy level diagram of the fabricated devices (a) FTO/TiO2/Dye/PANI/Pt (b) FTO/TiO2// PANI/Pt. Reprinted with permission from [Ameen S., 2009], J. Alloys comp. 487 (2009) 382.© 2011, Elsevier Ltd.

To elucidate the charge transfer properties of TiO2/PANI electrodes, an electrical impedance spectroscopy (EIS) measurement is used. According to the diffusion recombination model proposed by Bisquert [60], an equivalent circuit representing device is illustrated [(Inset of Fig. 7 (a)]. Equivalent circuit is composed of the series resistance (RS), the charge transfer resistance at the junction of TiO2 and PANI layer in TiO2/PANI or TiO2/Dye/PANI electrodes (RCT), the charge transfer resistance at the interface of TiO2/PANI or TiO2/Dye/PANI and TCO (RP/TCO), the capacitance of accumulation (of e<sup>−</sup> ) layer of the TiO2 (Cacc), and space charge capacitance (CSC) [61]. The values of real impedance (Zre) are used to estimate the values of RP/ TCO and RCT at different frequencies. Fig. 7 (a, b) exhibits the Nyquist curve of cell fabricated with TiO2/PANI and TiO2/Dye/PANI electrodes. A very high RP/TCO of 52.4Ω and RCT of 3700 Ω observed for TiO2/PANI electrodes based cells, which are estimated from Fig. 7(a). Compa‐ ratively, TiO2/Dye/PANI based device (Fig.7 (b)) shows the low RP/TCO (35.8 Ω) and RCT (81.9 Ω) due to the influence of dye layer which is placed between the TiO2 and PANI layer of the electrode. It is reported that a small RCT of the device suggests the fast electron transfer, while a large RCT indicates the slow charge transfer at the junction of inorganic and organic layer [62]. In our case, it is found that the value of RCT in TiO2/Dye/PANI based device is very low as compared to the RCT of TiO2/PANI based device. Therefore, it explains the high electron transfer at the junction of TiO2 and PANI in TiO2/Dye/PANI based device, resulting in the high photocurrent density and overall conversion efficiency, which are in the excellent agreement with the J-V curve results of the devices. The impedance results clearly indicate that the high photocurrent density, high overall conversion efficiency and low RCT are resulted from the uniform distribution of PANI molecules on the mesoporous surface of TiO2 electrode. Therefore, the lower RCT and RP/TCO in TiO2/Dye/PANI based device reveals that the dye and PANI layers on the surface of TiO2 electrode play an important role in the charge transfer at hole conductor (PANI)-dye absorbed TiO2 region, which results the high JSC, FF, and conver‐ sion efficiency than that of TiO2/PANI electrode based cells.

**Figure 7.** AC impedance of (a) FTO/TiO2/PANI/Pt and (b) TiO2/Dye/PANI thin film electrode based DSSCs at the fre‐ quency range from 10 Hz to 100 kHz. Inset shows the equivalent circuit model of the device. Reprinted with permis‐ sion from [Ameen S., 2009], J. Alloys comp. 487 (2009) 382. ©2011, Elsevier Ltd.

The J-V performance of solar cell FTO/TiO2/Dye/PANI/Pt and FTO/TiO2/PANI/Pt are shown in Fig. 8 (a, b) under 100 mW/cm2 light intensity. On comparison with TiO2/PANI, the solar cell based on TiO2/Dye/PANI electrode executes great improvement in the overall conversion efficiency with the incorporation of dye layer on TiO2/PANI electrode. The conversion efficiency of the solar cell drastically increases from ~0.005% to ~0.68%. It is noteworthy that the photovoltaic properties such as VOC, JSC and FF of the DSSCs enhance dramatically as compared to TiO2/PANI electrode based DSSC. The high JSC is imputed to the high electrical conductivity of PANI/TiO2 thin film. The enhancement in JV parameters are resulted from the formation of TiO2/PANI thin film, where the photon generated electrons could freely travel at the interface of PANI and TiO2 without decay, and dissociate into free charge carriers effec‐ tively. Moreover, the pore filling extent of the hole conductor into the dye-sensitized TiO2 film, and the electric contact of the hole conductor are the two important factors to determine the photovoltaic behaviors of device. The advanced TiO2/Dye/PANI electrode executes reasonably fast charge injection and transfer of electron at the interface of hole conductor (PANI) and Pt layer of electrode.

**Figure 8.** J–V curve of fabricated solar cell (a) FTO/TiO2/Dye/PANI/Pt (b) FTO/TiO2//PANI/Pt. Reprinted with permis‐ sion from [Ameen S., 2009], J. Alloys comp. 487 (2009) 382 ©2011, Elsevier Ltd.

### **5.2. PANI as counter electrodes for DSSCs**

capacitance (CSC) [61]. The values of real impedance (Zre) are used to estimate the values of RP/ TCO and RCT at different frequencies. Fig. 7 (a, b) exhibits the Nyquist curve of cell fabricated with TiO2/PANI and TiO2/Dye/PANI electrodes. A very high RP/TCO of 52.4Ω and RCT of 3700 Ω observed for TiO2/PANI electrodes based cells, which are estimated from Fig. 7(a). Compa‐ ratively, TiO2/Dye/PANI based device (Fig.7 (b)) shows the low RP/TCO (35.8 Ω) and RCT (81.9 Ω) due to the influence of dye layer which is placed between the TiO2 and PANI layer of the electrode. It is reported that a small RCT of the device suggests the fast electron transfer, while a large RCT indicates the slow charge transfer at the junction of inorganic and organic layer [62]. In our case, it is found that the value of RCT in TiO2/Dye/PANI based device is very low as compared to the RCT of TiO2/PANI based device. Therefore, it explains the high electron transfer at the junction of TiO2 and PANI in TiO2/Dye/PANI based device, resulting in the high photocurrent density and overall conversion efficiency, which are in the excellent agreement with the J-V curve results of the devices. The impedance results clearly indicate that the high photocurrent density, high overall conversion efficiency and low RCT are resulted from the uniform distribution of PANI molecules on the mesoporous surface of TiO2 electrode. Therefore, the lower RCT and RP/TCO in TiO2/Dye/PANI based device reveals that the dye and PANI layers on the surface of TiO2 electrode play an important role in the charge transfer at hole conductor (PANI)-dye absorbed TiO2 region, which results the high JSC, FF, and conver‐

**Figure 7.** AC impedance of (a) FTO/TiO2/PANI/Pt and (b) TiO2/Dye/PANI thin film electrode based DSSCs at the fre‐ quency range from 10 Hz to 100 kHz. Inset shows the equivalent circuit model of the device. Reprinted with permis‐

The J-V performance of solar cell FTO/TiO2/Dye/PANI/Pt and FTO/TiO2/PANI/Pt are shown

cell based on TiO2/Dye/PANI electrode executes great improvement in the overall conversion efficiency with the incorporation of dye layer on TiO2/PANI electrode. The conversion

light intensity. On comparison with TiO2/PANI, the solar

sion efficiency than that of TiO2/PANI electrode based cells.

216 Solar Cells - Research and Application Perspectives

sion from [Ameen S., 2009], J. Alloys comp. 487 (2009) 382. ©2011, Elsevier Ltd.

in Fig. 8 (a, b) under 100 mW/cm2

The counter electrode in DSSCs is responsible for the electrocatalytic reduction of I3 <sup>−</sup> ions. Until now, Pt counter electrode shows the high electrocatalytic activity for I3 <sup>−</sup> ions reduction, high conductivity, and stability. Pt is one of the most expensive components in DSSCs. Therefore, the development of counter electrodes with alternative materials is expected to reduce production costs of DSSCs. Several varieties of materials such as carbon nanotubes, activated carbon, graphite, and conducting polymers are employed as active catalysts for counter electrodes. In this regards, M. H. Yeh et al reported the conducting polymer-based counter electrode for a quantum-dot-sensitized solar cell (QDSSC) with a polysulfide electrolyte [63]. K. M. Lee et al fabricated the DSSC based on poly (3, 4-alkylenedioxythiophene) as counter electrode [64]. In another report, K. M. Lee et al exhibited the effects of mesoscopic poly (3, 4 ethylenedioxythiophene) films as counter electrodes for DSSCs [65]. W. Maiaugree et al optimized the TiO2 nanoparticle mixed PEDOT-PSS counter electrodes for high efficiency of DSSCs [66]. J. Chen et al reported polyaniline nanofiber-carbon film as flexible counter electrodes in platinum-free dye-sensitized solar cells [67]. Q. Li et al fabricated the microporous polyaniline thin film as counter electrode for DSSCs [68]. J. Zhang et al applied the nanostruc‐ tured PANI thin film as counter electrode for DSSCs and investigated the electrochemical formation mechanism [69]. Furthermore, G. Wang et al synthesized PANI-graphene hybrids thin film and utilized as a counter electrode in DSSCs [70]. Tai et al prepared the highly uniform and transparent PANI counter electrodes by a facile in situ polymerization method for the DSSCs [71]. In this regards, Ameen et al performed the doping of PANI with sulfamic acid (SFA) and applied as counter electrode for the efficient performance of DSSCs [72].

Fig. 9(a) shows the J-V curve of the DSSCs fabricated with the counter electrodes made of PANI NFs and SFA-doped PANI NFs under dark and light intensity of 100 mW/cm2 (1.5AM). DSSCs fabricated with SFA-doped PANI NFs counter electrode achieve high conversion efficiency (η) of ~5.5% with JSC of ~13.6 mA/cm2 , VOC of ~0.74 V, and FF of ~0.53. Significantly, the conversion efficiency increases from ~4.0% to ~5.5% after SFA doping into the PANI NFs. DSSC fabricated with SFA-doped PANI NFs counter electrode has appreciably improved the conversion efficiency and JSC by ∼27% and ∼20% than that of DSSC fabricated with PANI NFs counter electrode. These improvements are resulted from the higher electrocatalytic activity of SFA-

**Figure 9.** J-V curve (a) and IPCE curves (b) of a fabricated solar cell of PANI NFs and SFA doped PANI NFs as counter electrodes. Reprinted with permission from [Ameen S., 2010], J. Phys. Chem. C 114 (2010) 4760. ©2010, American Chemical Society.

doped PANI NFs, which serves a good path for the charge transport of I- /I3 – redox. Therefore, the superior photovoltaic properties such as η, JSC, and VOC of the cell are attributed to the sufficiently high conductivity and electrocatalytic activity of doped PANI NFs, which allevi‐ ates the reduction of I3 at the thin SFA-doped PANI NFs layers. Fig. 9(b) presents the IPCE curves of DSSCs fabricated with PANI NFs and SFA-doped PANI NFs counter electrodes. DSSCs fabricated with PANI NFs counter electrode exhibits the low IPCE of ~54% in the absorption range of 400-650 nm. The IPCE value is prominently increased by ~70% with the SFA doped PANI NFs counter electrode-based DSSCs. It is noteworthy that the IPCE of the device is considerably enhanced by ~24% upon SFA doping on PANI NFs-based counter electrodes. The enhanced IPCE results are consistent with high electrical conductivity [73] and the electrocatalytic activity of the SFA-doped PANI NFs electrode. Thus, SFA doping signifi‐ cantly enhances the electrical conductivity and increases the higher reduction of I3 to I in the electrolyte at the interface of PANI NFs layer and electrolyte.

## **5.3. Other ions doped PANI counter electrode based DSSCs**

polyaniline thin film as counter electrode for DSSCs [68]. J. Zhang et al applied the nanostruc‐ tured PANI thin film as counter electrode for DSSCs and investigated the electrochemical formation mechanism [69]. Furthermore, G. Wang et al synthesized PANI-graphene hybrids thin film and utilized as a counter electrode in DSSCs [70]. Tai et al prepared the highly uniform and transparent PANI counter electrodes by a facile in situ polymerization method for the DSSCs [71]. In this regards, Ameen et al performed the doping of PANI with sulfamic acid

Fig. 9(a) shows the J-V curve of the DSSCs fabricated with the counter electrodes made of PANI

fabricated with SFA-doped PANI NFs counter electrode achieve high conversion efficiency (η)

efficiency increases from ~4.0% to ~5.5% after SFA doping into the PANI NFs. DSSC fabricated with SFA-doped PANI NFs counter electrode has appreciably improved the conversion efficiency and JSC by ∼27% and ∼20% than that of DSSC fabricated with PANI NFs counter electrode. These improvements are resulted from the higher electrocatalytic activity of SFA-

**Figure 9.** J-V curve (a) and IPCE curves (b) of a fabricated solar cell of PANI NFs and SFA doped PANI NFs as counter electrodes. Reprinted with permission from [Ameen S., 2010], J. Phys. Chem. C 114 (2010) 4760. ©2010, American

, VOC of ~0.74 V, and FF of ~0.53. Significantly, the conversion

(1.5AM). DSSCs

(SFA) and applied as counter electrode for the efficient performance of DSSCs [72].

NFs and SFA-doped PANI NFs under dark and light intensity of 100 mW/cm2

of ~5.5% with JSC of ~13.6 mA/cm2

218 Solar Cells - Research and Application Perspectives

Chemical Society.

Z. Li et al recently studied on the in situ electropolymerized-PANI thin film of thickness ~5– 20 μm, deposited on FTO glass. The PANI thin films were doped by various counter ions like SO2 -4, ClO-4, BF-4, Cl- , p-toluenesulfonate (TsO- ), etc. Different doping counter ions showed different impact on the morphology, electrochemical activity of the electropolymerized-PANI thin film. The electropolymerized-PANI doped by SO2 -4 anion (PANI-SO4) film was much porous morphology with pore size diameter of several micrometers and possessed the higher reduction current for the reduction of I3 and a low charge transfer resistance of 1.3 Ωcm<sup>2</sup> as compared with Pt as counter electrode (CE). Dye-sensitized solar cell (DSSC) with PANI-SO4 as CE was assembled, and the device under full sunlight illumination (100mWcm-2, AM 1.5 G) showed ~5.6% photovoltaic conversion efficiency, which was comparable to ~6.0% of that with Pt CE under the same experimental condition. The electropolymerized-PANI doped with SO2 -4 ion with a porous and homogeneously structure was a promising candidate which showed the high performance of DSSC. On the other hand, PANI-BF4 and PANI-Cl was porous and fibrillar thin film which exhibited the modest efficiency of ~3.9% and ~2.6%. On the other hand, PANI-ClO4 and PANI-TsO showed the very low performance of DSSC ca. <1%, and the RCT was greatly increased accordingly to over 100 Ωcm<sup>2</sup> [74].

## **6. DSSCs based on metal oxide semiconductors**

In DSSCs, the choice of semiconductor is governed by the conduction band energy and density of states, which facilitate the charge separation and minimize the recombination. Secondly, the high surface area and morphology of semiconductors is important to maximize the light absorption by the dye molecules while maintaining the good electrical connectivity with the substrate [75]. The semiconducting metal oxides such as titania (TiO2), zinc oxide (ZnO), and tin oxide (SnO2) have shown good optical and electronic properties and are accepted as the effective photoelectrode materials for DSSCs. To improve the light harvesting efficiency, the metal oxide nanostructures must possess a high surface-to-volume ratio for high absorption of dye molecules. Particularly in DSSCs, the porous nature of nanocrystalline TiO2 films provides the large surface for dye-molecule adsorption and therefore, the suitable energy levels at the semiconductor-dye interface (the position of the conduction band of TiO2 being lower than the excited-state energy level of the dye) allow for the effective injection of electrons from the dye molecules to the semiconductor. Compared with other photovoltaic materials, anatase phase TiO2 is outstanding for its stability and wide band gap and, thus, widely used in the devices [76]. On the other hand, ZnO nanomaterials are chosen as an alternative material to TiO2 photoanodes due to their wide band gap with higher electronic mobility, which would be favorable for the efficient electron transport, with reduced recombination loss in DSSCs. Studies have already been reported on the use of ZnO material photoanodes for the application in DSSCs. Although the conversion efficiencies of ZnO (0.4-5.8%) is comparably lower than TiO2 (~11%), ZnO is still a distinguished alternative to TiO2 due to its ease of crystallization and anisotropic growth. In this part of the chapter, the TiO2 and ZnO have been briefly summarized for the application for DSSCs.

## **6.1. DSSCs Based on TiO2 Photoanode**

Due to versatile and the exotic properties, TiO2 nanomaterials are so far used in many technological applications as a photocatalyst, photovoltaic material, gas sensor, optical coating, structural ceramic, electrical circuit varistor, biocompatible material for bone im‐ plants, and spacer material for magnetic spin valve systems etc [77]. The dimensionality of TiO2 at the nanoscale level is the crucial characteristic for determining the physiological and electrical properties. In recent years, one dimensional (1D) TiO2 nanomaterials like NRs, NWs and NTs display significantly larger surface areas as compared to bulk materials, which deliver unique chemical and the physical properties [78] and contribute towards the electrical and photoelectrochemical applications [79]. The 1D TiO2 such as NRs and NTs have shown the reduced recombination rate for the excited electron-hole pair and display unique optical and the electric properties [80]. Particularly, the vertically grown TiO2 NRs allow shorter and the uninterrupted electrical pathways for the photogenerated carriers and improves the charge separation and charge transport properties in many photoelectrochemical devices like dye sensitized solar cells (DSSCs) [81]. As compared to TiO2 nanoparticles, it is expected that the highly oriented TiO2 NRs could be the potential electrode and photocatalyst material for several photoelectrochemical applications. S. Ameen et al reported the TiO2 nanorods (NRs) based photoanode for the fabrication of DSSCs [82].

The morphology of the TiO2 NR thin films deposited on FTO substrates by the hydrothermal process with variations of the ethanol/DI water precursor solution is shown in FESEM images (Fig. 10). With an ethanol/DI water ratio of 0:100 v/v as the precursor solution, the distorted hexagonal TiO2 NRs (Fig. 10 (a, b)) of average diameter ~100-200 nm and length of 3.0 mm are obtained. However, the round headed TiO2 NRs with ethanol/DI water (50: 50 v/v) as the precursor solution (Fig. 10 (c, d)) is formed. The highly ordered tetragonal TiO2 NRs have grown on the FTO substrate with the precursor solution of ethanol/DI water (80: 20 v/v) as seen in Fig. 10 (e, f). The grown TiO2 NRs possess the average lengths of 2-4 mm and diameters of ~50–70 nm respectively. The high amount of ethanol in the precursor solution is crucial to achieve the highly ordered nanorods.

provides the large surface for dye-molecule adsorption and therefore, the suitable energy levels at the semiconductor-dye interface (the position of the conduction band of TiO2 being lower than the excited-state energy level of the dye) allow for the effective injection of electrons from the dye molecules to the semiconductor. Compared with other photovoltaic materials, anatase phase TiO2 is outstanding for its stability and wide band gap and, thus, widely used in the devices [76]. On the other hand, ZnO nanomaterials are chosen as an alternative material to TiO2 photoanodes due to their wide band gap with higher electronic mobility, which would be favorable for the efficient electron transport, with reduced recombination loss in DSSCs. Studies have already been reported on the use of ZnO material photoanodes for the application in DSSCs. Although the conversion efficiencies of ZnO (0.4-5.8%) is comparably lower than TiO2 (~11%), ZnO is still a distinguished alternative to TiO2 due to its ease of crystallization and anisotropic growth. In this part of the chapter, the TiO2 and ZnO have been briefly

Due to versatile and the exotic properties, TiO2 nanomaterials are so far used in many technological applications as a photocatalyst, photovoltaic material, gas sensor, optical coating, structural ceramic, electrical circuit varistor, biocompatible material for bone im‐ plants, and spacer material for magnetic spin valve systems etc [77]. The dimensionality of TiO2 at the nanoscale level is the crucial characteristic for determining the physiological and electrical properties. In recent years, one dimensional (1D) TiO2 nanomaterials like NRs, NWs and NTs display significantly larger surface areas as compared to bulk materials, which deliver unique chemical and the physical properties [78] and contribute towards the electrical and photoelectrochemical applications [79]. The 1D TiO2 such as NRs and NTs have shown the reduced recombination rate for the excited electron-hole pair and display unique optical and the electric properties [80]. Particularly, the vertically grown TiO2 NRs allow shorter and the uninterrupted electrical pathways for the photogenerated carriers and improves the charge separation and charge transport properties in many photoelectrochemical devices like dye sensitized solar cells (DSSCs) [81]. As compared to TiO2 nanoparticles, it is expected that the highly oriented TiO2 NRs could be the potential electrode and photocatalyst material for several photoelectrochemical applications. S. Ameen et al reported the TiO2 nanorods (NRs)

The morphology of the TiO2 NR thin films deposited on FTO substrates by the hydrothermal process with variations of the ethanol/DI water precursor solution is shown in FESEM images (Fig. 10). With an ethanol/DI water ratio of 0:100 v/v as the precursor solution, the distorted hexagonal TiO2 NRs (Fig. 10 (a, b)) of average diameter ~100-200 nm and length of 3.0 mm are obtained. However, the round headed TiO2 NRs with ethanol/DI water (50: 50 v/v) as the precursor solution (Fig. 10 (c, d)) is formed. The highly ordered tetragonal TiO2 NRs have grown on the FTO substrate with the precursor solution of ethanol/DI water (80: 20 v/v) as seen in Fig. 10 (e, f). The grown TiO2 NRs possess the average lengths of 2-4 mm and diameters of ~50–70 nm respectively. The high amount of ethanol in the precursor solution is crucial to

summarized for the application for DSSCs.

based photoanode for the fabrication of DSSCs [82].

achieve the highly ordered nanorods.

**6.1. DSSCs Based on TiO2 Photoanode**

220 Solar Cells - Research and Application Perspectives

**Figure 10.** Low and high resolution FESEM images of the TiO2 NR thin films obtained with the precursor solutions of ethanol/DI water with ratios of (a, b) 0 : 100 v/v, (c, d) 50 : 50 v/v and (e, f) 80 : 20 v/v. Reprinted with permission from [Ameen S., 2012], RSC Adv. 2 (2012) 4807 ©2012, RSC Publications Ltd.

Fig. 11 shows the transmission electron microscopy (TEM), high resolution (HR) TEM and the selected area electron patterns (SAED) of the grown TiO2 NR coated FTO substrate. Similar to the FESEM results, the highly ordered tetrgonal TiO2 NRs from the precursor solution of ethanol/DI water (80:20v/v) solvent comprises the average length of 2-4 mm and the diameter of 50-70 nm, as shown in Fig. 11 (a). Each NR is made of a bundle of the densely packed nanofibers (NFs) with an average fibril's diameter of ~5 nm. The corresponding SAED pattern (Fig. 11 (b)) displays the clear phases, suggesting the high crystal quality with the single crystalline fibrils derived from TiO2 NRs along the [001] direction. However, the HRTEM image (Fig. 11 (c)) shows the well-resolved lattice fringes of the grown TiO2 NRs and estimates an average interplanar distance of 0.35 nm between the two fringes, which reveals the typical interplanar distance of anatase TiO2 [83]. On the other side, the width and length of distorted hexagonal TiO2 NRs are respectively observed as ~200 nm and ~3.2 mm, as seen in Fig. 11(d).

The XRD patterns (Fig. 12) of grown TiO2 NRs from both precursor solutions exhibit the anatase phase with the peaks at 25.1o , 37.9o , 48.1o , 53.8o and 55.1o , which correspond to typical anatase TiO2 materials and indexes at JCPDS no. 89-4203. However, the diffraction peaks of the FTO substrate are also observedat 33.8o , 35.7o and52.8o (JCPDSno. 88-0287).Oncomparison withthe distorted hexagonal TiO2 NRs, the intensities of XRD diffraction peaks have slightly changed, which might indicate the high crystalline nature of the highly ordered tetragonal TiO2 NRs.

**Figure 11.** TEM image of (a) highly ordered tetragonal TiO2 NRs, (b) SAED patterns, (c) HRTEM image and (d) TEM image of grown hexagonal distorted TiO2 NRs. Reprinted with permission from [Ameen S., 2012], ], RSC Adv. 2 (2012) 4807 ©2012, RSC Publications Ltd.

**Figure 12.** XRD patterns of (a) the distorted hexagonal TiO2 NRs and (b) the highly ordered tetragonal TiO2 NR thin film. Reprinted with permission from [Ameen S., 2012], RSC Adv. 2 (2012) 4807 ©2012, RSC Publications Ltd.

The J-V characteristics (Fig. 13 (A)) have been performed to elucidate the performance of the DSSCs fabricated with the photoanodes of grown TiO2 NRs and are measured under a light intensity of 100 mW cm-2 (1.5 AM). DSSC fabricated with the distorted hexagonal TiO2 NRs photoanode shows a relatively low solar efficiency of ~1.08%, with a low JSC of ~4.48 mA cm-2, VOC ~0.571 V and FF of ~0.42. However, DSSCs fabricated with the highly ordered tetragonal TiO2 NRs photoanode achieves an appreciably improved overall conversion efficiency of ~3.2% with a high JSC of ~8.7 mA cm-2, VOC of ~0.67 V, and FF of ~0.54. As compared to the distorted hexagonal TiO2 NRs photoanode based DSSC, the photovoltaic performance, JSC, VOC and FF are significantly enhanced by ~ 67%, ~48%, ~15% and ~22% respectively. It is seen that the size of the NRs also plays an important role for achieving the high photocurrent density and performance of the device. It is known that the high photovoltaic performance and photocur‐ rent density are related to high light harvesting through the highly uniform and high surface to volume ratio of the photoanode materials [84]. In general, the TiO2 thin film electrodes with larger particles have the smaller surface area and produce moderate contact points between nanoparticles at the interface of the sintered nanoparticles and the conducting substrate, leading to the lower availability of the active surface for dye adsorption, which perhaps decreases the amount of light absorbed and generates the large number of electrons and holes. Whereas, the TiO2 thin film with smaller particles acquires the larger surface area and higher number of contact points of the sintered colloidal particles present at the interface of the nanoparticles and the conducting substrate, which gives rise to larger dye adsorption and higher light harvesting efficiency [85]. In this case, the distorted hexagonal TiO2 NRs consist of larger diameters and lengths as compared to the highly ordered TiO2 NRs, as shown in the FESEM images. It is believed that the smaller diameters of the NRs might generate the high light harvesting efficiency, which might lead to the high photocurrent density and the conversion efficiency. From the UV-Vis spectra (Fig. 13 (B)) of the dye desorption from dye absorbed TiO2 NRs photoanodes in NaOH solution, the photoanode of highly ordered tetragonal TiO2 NRs attains the higher dye loading than the photoanode of the distorted hexagonal TiO2 NRs. Herein, the enhanced photovoltaic performance and JSC are related to the highly ordered NRs morphology, high dye loading and improved light harvesting efficiency through the high surface area of the film. Besides these, the unique ordered morphology of the NRs might retard the recombination rate and contribute to longer electron lifetimes [86], resulting in the increased VOC and FF of device.

The IPCE (Fig. 14) of DSSCs fabricated with highly ordered tetragonal TiO2 NRs and distorted hexagonal TiO2 NRs photoanodes have shown the broad the absorption edge of visible spectrum from 400-800 nm. The highly ordered tetragonal TiO2 NRs photoanode based DSSC exhibits the maximum IPCE of ~31.5% at the highest absorption edge of ~528 nm, whereas ~17.9% IPCE at ~528 nm is achieved by the distorted hexagonal TiO2 NRs photoanode based DSSC. The highly ordered tetragonal TiO2 NRs photoanode based DSSC considerably improves IPCE by approximately two times to DSSC with distorted hexagonal TiO2 NRs photoanode, which is attributed to the high dye loading of the photoanode, resulting in the high light harvesting efficiency and the electron injection from dye to CB of TiO2. Thus, the highly ordered tetragonal TiO2 NRs photoanode with enhanced dye loading, light harvesting and IPCE, have resulted to increased JSC, VOC and the photovoltaic performance for DSSC.

**Figure 12.** XRD patterns of (a) the distorted hexagonal TiO2 NRs and (b) the highly ordered tetragonal TiO2 NR thin film. Reprinted with permission from [Ameen S., 2012], RSC Adv. 2 (2012) 4807 ©2012, RSC Publications Ltd.

**Figure 11.** TEM image of (a) highly ordered tetragonal TiO2 NRs, (b) SAED patterns, (c) HRTEM image and (d) TEM image of grown hexagonal distorted TiO2 NRs. Reprinted with permission from [Ameen S., 2012], ], RSC Adv. 2 (2012)

4807 ©2012, RSC Publications Ltd.

222 Solar Cells - Research and Application Perspectives

**Figure 13.** (A) J-V curve of the DSSC fabricated with (a) distorted hexagonal TiO2 NRs and (b) highly ordered tetrago‐ nal TiO2 NRs. (B) UV-Vis spectroscopy of desorbed dye from (a) distorted hexagonal TiO2 NRs and (b) highly ordered tetragonal TiO2 NRs. Reprinted with permission from [Ameen S., 2012] ], RSC Adv. 2 (2012) 4807©2012, RSC Publica‐ tions Ltd.

### **6.2. DSSCs based ZnO photoanode**

Another metal oxide nanomaterials such as ZnO nanomaterials, are recently dealing with the versatile applications in various fields such as field-effect transistors, lasers, photodiodes, sensors and photovoltaics owing to their unique photoelectric properties, optical transparency, electric conductivity and piezoelectricity properties [87]. Importantly, ZnO nanomaterials possess similar wide band gap (~3.37 eV) with large exciton binding energy (~60 meV) and higher electron mobility [88]. Moreover, ZnO nanomaterials with different nanostructures have presented the versatile properties like higher surface-to-volume ratio, chemical stability, high exciton binding energy, and moderate charge transport capability [89] and therefore, it becomes one of the promising alternatives of nanocrystalline TiO2 photoanode in DSSCs and hybrid solar devices. To control the parameters like morphology, physical and the crystalline properties of ZnO nanomaterials, the performance of DSSCs could be accelerated [90]. So far, the different morphologies of ZnO nanostructures such as nanorods [91], nanotetrapods,

**Figure 14.** IPCE curve of the DSSC fabricated with (a) the distorted hexagonal TiO2 NR photoanode and (b) the highly ordered tetragonal TiO2 NR photoanode. Reprinted with permission from [Ameen S., 2012], RSC Adv. 2 (2012) 4807©2012, RSC Publications Ltd.

nanosheet [92] and nanobelts based photoanodes have been studied for the fabrication of efficient DSSCs [93] and achieved encouraging results. Law et al. firstly designed ZnO NW arrays to increase the electron diffusion length, and applied as photoelectrodes for the fabrication of DSSCs [94]. The grown ZnO NW array films exhibited relatively good resistivity values that ranged from 0.3 to 2.0 *Ω*cm for individual NWs and a mobility of 1-5 cm2 V−1s−1. The overall conversion efficiencies of ~1.2-1.5% were obtained by DSSCs fabricated with ZnO NW arrays with JSC of ~5.3-5.85 mA/cm2 , VOC of ~0.610-0.710 V, and FF of ~0.36-0.38. Another

**6.2. DSSCs based ZnO photoanode**

224 Solar Cells - Research and Application Perspectives

tions Ltd.

Another metal oxide nanomaterials such as ZnO nanomaterials, are recently dealing with the versatile applications in various fields such as field-effect transistors, lasers, photodiodes, sensors and photovoltaics owing to their unique photoelectric properties, optical transparency, electric conductivity and piezoelectricity properties [87]. Importantly, ZnO nanomaterials possess similar wide band gap (~3.37 eV) with large exciton binding energy (~60 meV) and higher electron mobility [88]. Moreover, ZnO nanomaterials with different nanostructures have presented the versatile properties like higher surface-to-volume ratio, chemical stability, high exciton binding energy, and moderate charge transport capability [89] and therefore, it becomes one of the promising alternatives of nanocrystalline TiO2 photoanode in DSSCs and hybrid solar devices. To control the parameters like morphology, physical and the crystalline properties of ZnO nanomaterials, the performance of DSSCs could be accelerated [90]. So far, the different morphologies of ZnO nanostructures such as nanorods [91], nanotetrapods,

**Figure 13.** (A) J-V curve of the DSSC fabricated with (a) distorted hexagonal TiO2 NRs and (b) highly ordered tetrago‐ nal TiO2 NRs. (B) UV-Vis spectroscopy of desorbed dye from (a) distorted hexagonal TiO2 NRs and (b) highly ordered tetragonal TiO2 NRs. Reprinted with permission from [Ameen S., 2012] ], RSC Adv. 2 (2012) 4807©2012, RSC Publica‐

**Figure 15.** FE-SEM images of (a) ZnO–NH3, (b) ZnO-citric, and (c) ZnO-oxalic materials prepared by hydrothermal method. Inset shows the high magnification FE-SEM images of (a), (b), and (c). Reprinted with permission from [Akh‐ tar., 2008], Electrochim. Acta 53(2008)7869.© 2008, Elsevier Ltd.

group synthesized ZnO NWs by the use of ammonium hydroxide for changing the super saturation degree of Zn precursors in solution process [95]. The length-to-diameter aspect ratio of the individual NWs was easily controlled by changing the concentration of ammonium hydroxide. The fabricated DSSCs exhibited remarkably high conversion efficiency of ~1.7%, which was much higher than DSSC with ZnO NR arrays [96]. Jiang et al. fabricated the flexible DSSCs with a highly bendable ZnO NWs film on PET/ITO substrate which was prepared by a low-temperature hydrothermal growth at 85o C [97]. The fabricated composite films obtained by immersing the ZnO NPs powder in a methanolic solution of 2% titanium isopropoxide and 0.02 M acetic acid was treated with heat, which favored a good attachment of NPs onto the NW surfaces [97]. Here, the achieved conversion efficiency of the fabricated DSSCs was less as compared to DSSCs based on NPs. Recently, Akhtar et al. demonstrated that the perform‐ ance of DSSCs effectively altered by varying the morphologies of ZnO nanomaterials. They reported the morphology of ZnO nanomaterials through a hydrothermal process using Zinc acetate, NaOH and different capping agents, as shown in Fig 15. The photoanode was prepared by spreading the ZnO paste on an FTO substrate by a doctor blade technique for the fabrication of DSSCs [98], and they obtained non-uniform surface of film. Unfortunately, the DSSCs with ZnO NRs photoanode presented a very low conversion efficiency of ~0.3% with a high FF of ~0.54 (Fig 16), which might attribute to the low dye absorption on the surface of ZnO NRs due to the less uniformity of the thin film with low surface-to-volume ratio (Fig 17). Furthermore, a flower-like structure consisted of NRs/NWs could deliver a larger surface area and a direct pathway for electron transport with the channels arisen from the branched to NR/NW backbone. Recently, hydrothermally grown ZnO nanoflower films accomplished an improved overall conversion efficiency of ~1.9%, with a high JSC of ~5.5 mA/cm2 and an FF of ~0.53 [99]. These parameters are higher than NR arrays film-based DSSCs of the conversion efficiency ~1.0%, JSC ~4.5 mA/cm2 , and FF ~0.36. Recently, S. Ameen et al reported the nanospikes decorated ZnO sheets thin film as photoanode for the performance of DSSCs [100].

The FESEM image (Fig. 18) shows dense and uniform deposition of the nanospikes decorated ZnO sheets morphology on the FTO substrate. Each nanospikes decorated ZnO sheets is comprised of a sheet with the average thickness of ~50-60 nm and the aligned nanospikes with the average diameter of ~80-100 nm and length of ~150-200 nm. Interestingly, the nanospikes are consisted of the bundles of small nanorods. The nanospikes are aligned either on one side or other side of ZnO sheet, but in some cases, these nanospikes are found on the both sides of ZnO sheets.

Similarly, TEM images (Fig. 19(a)) present the nanospikes with nanosheets in which nano‐ spikes decorated on both the sides of ZnO sheet. The average thickness of the sheet is ~50-60 nm and the decorated nanospikes possess the average diameter of ~30 nm (single rods) and the length of ~150-200 nm. From the HRTEM image (Fig. 19 (b)), the well-resolved lattice reveals that the grown ZnO nanomaterials exhibit the good crystallinity. The inter-planar spacing of ~0.52 nm is observed which is consistent to the lattice constant in the reference (JCPDS No. 36-1451) for ZnO nanomaterials. This inter-planar spacing value of the lattice fringes correspond to the [0001] crystal plane of the wurtzite ZnO confirms that the grown ZnO nanomaterials are almost defect free [100]. Moreover, the corresponding selected area electron diffraction (SAED) also indicates the typical wurtzite single crystalline structure and the ZnO nanomaterials are grown along caxis direction [0001].

Metal Oxide Nanomaterials, Conducting Polymers and Their Nanocomposites for Solar Energy http://dx.doi.org/10.5772/51432 227

group synthesized ZnO NWs by the use of ammonium hydroxide for changing the super saturation degree of Zn precursors in solution process [95]. The length-to-diameter aspect ratio of the individual NWs was easily controlled by changing the concentration of ammonium hydroxide. The fabricated DSSCs exhibited remarkably high conversion efficiency of ~1.7%, which was much higher than DSSC with ZnO NR arrays [96]. Jiang et al. fabricated the flexible DSSCs with a highly bendable ZnO NWs film on PET/ITO substrate which was prepared by

by immersing the ZnO NPs powder in a methanolic solution of 2% titanium isopropoxide and 0.02 M acetic acid was treated with heat, which favored a good attachment of NPs onto the NW surfaces [97]. Here, the achieved conversion efficiency of the fabricated DSSCs was less as compared to DSSCs based on NPs. Recently, Akhtar et al. demonstrated that the perform‐ ance of DSSCs effectively altered by varying the morphologies of ZnO nanomaterials. They reported the morphology of ZnO nanomaterials through a hydrothermal process using Zinc acetate, NaOH and different capping agents, as shown in Fig 15. The photoanode was prepared by spreading the ZnO paste on an FTO substrate by a doctor blade technique for the fabrication of DSSCs [98], and they obtained non-uniform surface of film. Unfortunately, the DSSCs with ZnO NRs photoanode presented a very low conversion efficiency of ~0.3% with a high FF of ~0.54 (Fig 16), which might attribute to the low dye absorption on the surface of ZnO NRs due to the less uniformity of the thin film with low surface-to-volume ratio (Fig 17). Furthermore, a flower-like structure consisted of NRs/NWs could deliver a larger surface area and a direct pathway for electron transport with the channels arisen from the branched to NR/NW backbone. Recently, hydrothermally grown ZnO nanoflower films accomplished an improved

These parameters are higher than NR arrays film-based DSSCs of the conversion efficiency

The FESEM image (Fig. 18) shows dense and uniform deposition of the nanospikes decorated ZnO sheets morphology on the FTO substrate. Each nanospikes decorated ZnO sheets is comprised of a sheet with the average thickness of ~50-60 nm and the aligned nanospikes with the average diameter of ~80-100 nm and length of ~150-200 nm. Interestingly, the nanospikes are consisted of the bundles of small nanorods. The nanospikes are aligned either on one side or other side of ZnO sheet, but in some cases, these nanospikes are found on the both sides of

Similarly, TEM images (Fig. 19(a)) present the nanospikes with nanosheets in which nano‐ spikes decorated on both the sides of ZnO sheet. The average thickness of the sheet is ~50-60 nm and the decorated nanospikes possess the average diameter of ~30 nm (single rods) and the length of ~150-200 nm. From the HRTEM image (Fig. 19 (b)), the well-resolved lattice reveals that the grown ZnO nanomaterials exhibit the good crystallinity. The inter-planar spacing of ~0.52 nm is observed which is consistent to the lattice constant in the reference (JCPDS No. 36-1451) for ZnO nanomaterials. This inter-planar spacing value of the lattice fringes correspond to the [0001] crystal plane of the wurtzite ZnO confirms that the grown ZnO nanomaterials are almost defect free [100]. Moreover, the corresponding selected area electron diffraction (SAED) also indicates the typical wurtzite single crystalline structure and

decorated ZnO sheets thin film as photoanode for the performance of DSSCs [100].

, and FF ~0.36. Recently, S. Ameen et al reported the nanospikes

overall conversion efficiency of ~1.9%, with a high JSC of ~5.5 mA/cm2

the ZnO nanomaterials are grown along caxis direction [0001].

C [97]. The fabricated composite films obtained

and an FF of ~0.53 [99].

a low-temperature hydrothermal growth at 85o

226 Solar Cells - Research and Application Perspectives

~1.0%, JSC ~4.5 mA/cm2

ZnO sheets.

**Figure 16.** Current–voltage curve of DSSC fabricated with (a) ZnO–NH3, (b) ZnO-citric, and (c) ZnO-oxalic at 1.5 AM. Reprinted with permission from [Akhtar., 2008], Electrochim. Acta 53, (2008) 7869.© 2008, Elsevier Ltd.

**Figure 17.** UV-vis absorption of dye (N-719) extracted with 2 M NaOH from the electrodes of (a) ZnO–NH3, (b) ZnOcitric, and (c) ZnO-oxalic. Reprinted with permission from [Akhtar., 2008], Electrochim. Acta 53 (2008) 7869.© 2008, Elsevier Ltd.

**Figure 18.** FESEM images of nanospikes decorated ZnO sheets (a) at low magnification and (b) at high magnification. [Ameen S., 2012], Chem. Eng. J. 195 (2012) 307©2012. Elsevier Ltd.

**Figure 19.** TEM image (a) and HR-TEM image (b) of grown nanospikes decorated ZnO sheets. Inset shows the corre‐ sponding SAED patterns of grown nanospikes decorated ZnO sheets. Reprinted with permission from [Ameen S., 2012], Chem. Eng. J. 195 (2012) 307 ©2012, Elsevier Ltd.

XRD patterns (Fig. 20 (a)) of the nanospikes decorated ZnO sheets obtain all the diffraction peaks appeared at 32.3o (100), 35.2o (002), 36.8o (101), 48.2o (102), 57.2o (110), 63.5o (103) and 66.2o (200) which are well matched with the JCPDS card No.36-1451. It confirms that the ZnO nanomaterials possess the hexagonal wurtzite phase with the lattice parameters: a-3.246 and c-5.206 Å. The intensity of (101) diffraction peak is much higher compared to other peaks, indicating the preferential growth direction due to the instability of polar (101) plane [101]. A single narrow absorption peak is observed near the UV region at ~376 nm in the UV-Vis absorbance spectrum of nanospikes decorated ZnO sheets structures (Fig. 20 (b)), corresponds to the characteristic band of the wurtzite hexagonal structure in bulk ZnO [102]. Moreover, the single peak suggests purity of the grown nanospikes decorated sheets morphology.

**Figure 20.** XRD patterns (a) and UV-Vis spectra (b) of nanospikes decorated ZnO sheets. Reprinted with permission from [Ameen S., 2012], Chem. Eng. J. 195 (2012) 307©2012, Elsevier Ltd.

**Figure 19.** TEM image (a) and HR-TEM image (b) of grown nanospikes decorated ZnO sheets. Inset shows the corre‐ sponding SAED patterns of grown nanospikes decorated ZnO sheets. Reprinted with permission from [Ameen S.,

**Figure 18.** FESEM images of nanospikes decorated ZnO sheets (a) at low magnification and (b) at high magnification.

2012], Chem. Eng. J. 195 (2012) 307 ©2012, Elsevier Ltd.

[Ameen S., 2012], Chem. Eng. J. 195 (2012) 307©2012. Elsevier Ltd.

228 Solar Cells - Research and Application Perspectives

The survey XPS spectrum (Fig. 21 (a)) of grown nanospikes decorated ZnO sheets shows the three strong binding energies of Zn 2p3/2, Zn 2p1/2 and O 1s along with small C 1s binding energy. The other binding energies peaks are not detected, indicating the presence of Zn and O with‐ out other impurities. However, the C1s binding energy at ~ 284.6 eV is usually used as calibra‐ tion for other binding energies in the spectrum to avoid the specimen charging [103]. The Zn 2p spectrum of the doublet peaks with the binding energies of ~1021 eV and ~1045 eV are shown Fig 21(b), corresponding to Zn 2p3/2 and Zn 2p1/2 in better symmetry, respectively. These binding energies and the difference between two binding energies to ~24 eV are attributed to the typical lattice Zinc in ZnO [104]. The peak at ~1021 eV is associated with the Zn2+ in ZnO wurzite structure [105]. Moreover, Zn 2p binding energy and the binding energy difference values confirm that Zn atoms are in +2 oxidation state in ZnO. The deconvolution of O 1s XPS spectrum (Fig. 21(c)) exhibits the main peak at ~528.3 eV along with three resolved peaks at ~529.2 eV, ~530.1 eV and ~531.1 eV. The higher and lower binding energy component at ~528.3 eV and ~529.2 eV are attributed to O2 ions on the wurtzite structure of the hexagonal Zn2+ ions [106]. Every O2 ions are surrounded by Zn atoms with the full appreciation of nearest neighbor O2 ions. The other binding energies at ~530.1 eV and ~531.1 eV are ascribed to few oxygen deficiency or oxygen vacancies within the ZnO materials. Therefore, highest binding energy of Zn 2p and O 1s spectra are associated with Zn2+ and O2 ions which form Zn-O bonds in ZnO crystals.

**Figure 21.** Survey (a), Zn 2p (b) and O 1s (c) XPS spectra of nanospikes decorated ZnO sheets. Reprinted with permis‐ sion from [Ameen S., 2012], Chem. Eng. J. 195 (2012) 307 ©2012, Elsevier Ltd.

The J-V curve (Fig. 22(a)) of DSSC fabricated with nanospikes decorated ZnO sheets photo‐ anodes is demonstrated under the light intensity of 100 mW/cm2 (1.5 AM). The fabricated DSSC with the photoanode of nanospikes decorated ZnO sheets has achieved the overall conversion efficiency of ~2.51% with the reasonably high JSC of ~6.07 mA/cm2 , VOC of ~0.68 V and FF of ~0.60. The relatively high JSC is associated to high dye absorption through nanospikes decorated ZnO sheets morphology, resulting from the high amount of dye absorption (2.05 x10-7 mol/ cm2 ) which is calculated by area integration of the maximum absorbance in the UV-Vis spectrum of desorbed dye from the photoanode (as shown in inset of Fig. 22(a)). The unique morphology of the prepared nanospikes decorated ZnO sheets might pronounce the charge collection and transfer properties of electrode due to the presence of standing spikes on the ZnO sheets [107]. The improved VOC and FF of DSSC might attribute to the reduced charge recombination and the series resistance by the photoanode of nanospikes decorated ZnO sheets. As compared to the reported DSSCs based on ZnO nanostructures photoanodes, the nanospikes decorated ZnO sheets photoanode based DSSC shows the significantly higher conversion efficiency and JSC [108]. It has been estimated that the conversion efficiency and JSC are enhanced by ~40% and ~25% as compared to reported values. In this case, the sheets morphology of ZnO display highly uniform and the standing nanospikes might considerably

The survey XPS spectrum (Fig. 21 (a)) of grown nanospikes decorated ZnO sheets shows the three strong binding energies of Zn 2p3/2, Zn 2p1/2 and O 1s along with small C 1s binding energy. The other binding energies peaks are not detected, indicating the presence of Zn and O with‐ out other impurities. However, the C1s binding energy at ~ 284.6 eV is usually used as calibra‐ tion for other binding energies in the spectrum to avoid the specimen charging [103]. The Zn 2p spectrum of the doublet peaks with the binding energies of ~1021 eV and ~1045 eV are shown Fig 21(b), corresponding to Zn 2p3/2 and Zn 2p1/2 in better symmetry, respectively. These binding energies and the difference between two binding energies to ~24 eV are attributed to the typical lattice Zinc in ZnO [104]. The peak at ~1021 eV is associated with the Zn2+ in ZnO wurzite structure [105]. Moreover, Zn 2p binding energy and the binding energy difference values confirm that Zn atoms are in +2 oxidation state in ZnO. The deconvolution of O 1s XPS spectrum (Fig. 21(c)) exhibits the main peak at ~528.3 eV along with three resolved peaks at ~529.2 eV, ~530.1 eV and ~531.1 eV. The higher and lower binding energy component at ~528.3 eV and ~529.2 eV are

ions on the wurtzite structure of the hexagonal Zn2+ ions [106]. Every O2

ions which form Zn-O bonds in ZnO crystals.

binding energies at ~530.1 eV and ~531.1 eV are ascribed to few oxygen deficiency or oxygen vacancies within the ZnO materials. Therefore, highest binding energy of Zn 2p and O 1s spectra

**Figure 21.** Survey (a), Zn 2p (b) and O 1s (c) XPS spectra of nanospikes decorated ZnO sheets. Reprinted with permis‐

sion from [Ameen S., 2012], Chem. Eng. J. 195 (2012) 307 ©2012, Elsevier Ltd.

are surrounded by Zn atoms with the full appreciation of nearest neighbor O2


 ions

ions. The other


attributed to O2


230 Solar Cells - Research and Application Perspectives

are associated with Zn2+ and O2

**Figure 22.** J-V curve (a) and IPCE (b) curve of the DSSC fabricated with nanospikes decorated ZnO sheets photoanode. Inset (a) shows the UV-Vis spectrum of desorbed dye from the nanospikes decorated ZnO sheets photoanode. Reprint‐ ed with permission from [Ameen S., 2012], Chem. Eng. J. 195 (2012) 307 ©2012, Elsevier Ltd.

facilitates the electrons transfer at the interface of the conduction and the electrolyte layer. The fabricated DSSC with photoanode of nanospikes decorated ZnO sheets attains the moderate IPCE of ~31.8%, as shown in Fig. 22 (b), which is probably originated from the larger amount of dye-loading through large surface area of sheet and the standing spikes of photoanode. Moreover, the presence of nanospikes on ZnO sheets might efficiently enhance the electron transport and reduces the recombination rate to high IPCE and JSC value 109].

## **7. Doping of ZnO for improved electrical and photovoltaic properties**

One of the modifications is still in the developing stage called doping of ZnO nanomaterials by metals like F, Cu, Ag, Ga, Al, In, Sn and Sb which usually tailors the chemical, conductive and the electrical properties of ZnO nanomaterials [110]. The metal doping is an effective procedure to modify the grain size, orientation and the conductivity and could greatly influence the crystalline, optical and the electrical properties of the ZnO nanostructures. Among various metal doping, Sn-ion is recently known as promising dopant to ZnO nano‐ materials for enhancing the electrical and optical properties [111]. Tsay et al. [112] prepared the Sn doped ZnO thin films coated glass substrates and investigated the effects of Sn doping on the crystallinity, microstructures and the optical properties of ZnO thin film. Several reports are available on the preparation of the Sn doped ZnO thin films and the effects of Sn doping on grain size, vibrational structure, optical and the structural properties of ZnO thin film substrates [113]. Ameen et al recently reported the doping of ZnO nanostructures with Sn ion by simple hydrothermal process for the fabrication of DSSC [114].

### **7.1. Sn doped ZnO nanostructures for solar cell performance**

The synthesized ZnO and Sn-ZnO nanostructures are morphologically characterized by the FESEM images, as shown in Fig. 23. The ZnO nanostructure shows the irregular, non-uniform and highly aggregated nanoparticles with the average size of range ~150–200 nm. After Sn-ion doping, the ZnO nanostructures have dramatically arranged into the spindle shaped mor‐ phology and each Sn-ZnO spindle comprises of small aggregated nanoparticles with average size of 350 ±50 nm.

Fig. 24 (a, b) shows the high resolution TEM images of Sn-ZnO nanostructures which are completely consistent with FESEM observations. The aggregated ZnO nanoparticles arranged spindle shaped morphology is observed with some black spots or particles, clearly indicating the presence of the Sn-ions. The HR-TEM image of Sn-ZnO displays dark spots on the fringes which are expressed by the circles in Fig. 24(c). The morphological changes in Sn-ZnO nanostructures might due to the substantive influence of Sn-ion into ZnO nanostructures. The EDS spectrum (Fig. 24 (d)) obtains two high intense peaks (Zn & O) and single small peak (C) along with Sn peaks, again confirming the Sn-ion doping into the ZnO nanostructures.

From the UV-DRS spectra of ZnO and Sn-ZnO nanostructures (Fig. 25), the broad intense absorption edge from ~400 nm to lower wavelengths region is assigned to the charge-transfer process from the valence band to conduction band of ZnO [115]. After Sn-ion doping, the Metal Oxide Nanomaterials, Conducting Polymers and Their Nanocomposites for Solar Energy http://dx.doi.org/10.5772/51432 233

facilitates the electrons transfer at the interface of the conduction and the electrolyte layer. The fabricated DSSC with photoanode of nanospikes decorated ZnO sheets attains the moderate IPCE of ~31.8%, as shown in Fig. 22 (b), which is probably originated from the larger amount of dye-loading through large surface area of sheet and the standing spikes of photoanode. Moreover, the presence of nanospikes on ZnO sheets might efficiently enhance the electron

transport and reduces the recombination rate to high IPCE and JSC value 109].

232 Solar Cells - Research and Application Perspectives

by simple hydrothermal process for the fabrication of DSSC [114].

**7.1. Sn doped ZnO nanostructures for solar cell performance**

size of 350 ±50 nm.

**7. Doping of ZnO for improved electrical and photovoltaic properties**

One of the modifications is still in the developing stage called doping of ZnO nanomaterials by metals like F, Cu, Ag, Ga, Al, In, Sn and Sb which usually tailors the chemical, conductive and the electrical properties of ZnO nanomaterials [110]. The metal doping is an effective procedure to modify the grain size, orientation and the conductivity and could greatly influence the crystalline, optical and the electrical properties of the ZnO nanostructures. Among various metal doping, Sn-ion is recently known as promising dopant to ZnO nano‐ materials for enhancing the electrical and optical properties [111]. Tsay et al. [112] prepared the Sn doped ZnO thin films coated glass substrates and investigated the effects of Sn doping on the crystallinity, microstructures and the optical properties of ZnO thin film. Several reports are available on the preparation of the Sn doped ZnO thin films and the effects of Sn doping on grain size, vibrational structure, optical and the structural properties of ZnO thin film substrates [113]. Ameen et al recently reported the doping of ZnO nanostructures with Sn ion

The synthesized ZnO and Sn-ZnO nanostructures are morphologically characterized by the FESEM images, as shown in Fig. 23. The ZnO nanostructure shows the irregular, non-uniform and highly aggregated nanoparticles with the average size of range ~150–200 nm. After Sn-ion doping, the ZnO nanostructures have dramatically arranged into the spindle shaped mor‐ phology and each Sn-ZnO spindle comprises of small aggregated nanoparticles with average

Fig. 24 (a, b) shows the high resolution TEM images of Sn-ZnO nanostructures which are completely consistent with FESEM observations. The aggregated ZnO nanoparticles arranged spindle shaped morphology is observed with some black spots or particles, clearly indicating the presence of the Sn-ions. The HR-TEM image of Sn-ZnO displays dark spots on the fringes which are expressed by the circles in Fig. 24(c). The morphological changes in Sn-ZnO nanostructures might due to the substantive influence of Sn-ion into ZnO nanostructures. The EDS spectrum (Fig. 24 (d)) obtains two high intense peaks (Zn & O) and single small peak (C) along with Sn peaks, again confirming the Sn-ion doping into the ZnO nanostructures.

From the UV-DRS spectra of ZnO and Sn-ZnO nanostructures (Fig. 25), the broad intense absorption edge from ~400 nm to lower wavelengths region is assigned to the charge-transfer process from the valence band to conduction band of ZnO [115]. After Sn-ion doping, the

**Figure 23.** FESEM images of ZnO (a) and Sn-ZnO (b) nanostructures at low resolution, and FESEM images of Sn-ZnO nanostructures at high resolution (c and d). Reprinted with permission from [Ameen S., 2012], Chem. Eng. J. 187 (2012) 351 ©2012, Elsevier Ltd.

**Figure 24.** TEM images of Sn-ZnO nanostructures at low resolution (a and b), HRTEM image (c), and EDS spectrum of Sn-ZnO nanostructures (d). Reprinted with permission from [Ameen S., 2012], Chem. Eng. J. 187 (2012) 351 ©2012, Elsevier Ltd.

absorption wavelength of ZnO has significantly shifted from ~389 nm to ~406 nm and its band gap has changed from ~3.18 eV to ~3.05 eV. It has arisen due to the presence of interstitially embedded Sn-ion into ZnO nanomaterials. This small variation in band gaps again confirms the Sn-ion doping into ZnO nanomaterials.

**Figure 25.** UV-DRS spectra of ZnO and Sn-ZnO nanostructures. Reprinted with permission from [Ameen S., 2012], Chem. Eng. J. 187 (2012) 351 ©2012, Elsevier Ltd.

The XPS survey (Fig. 26 (a)) spectrum displays all Zn 2p, O 1s and Sn 3d binding energy peaks with very small C 1s binding energy. Sn-ZnO nanomaterials show the doublet bind‐ ing energies at ~1022.1 eV and ~1046.1 eV in Zn 2p XPS spectra (Fig. 26 (b)) which corre‐ spond to Zn 2p3/2 and Zn 2p1/2 respectively. The energy difference between doublet binding energies is calculated to ~24 eV, which is in excellent agreement with the standard value of ~22.97 eV. The deconvoluted O 1s XPS presents the four binding energies peaks at ~533.6 eV, ~532.4 eV, ~530.8 eV and ~529.1 eV, as shown in Fig. 26 (c). The highest binding energy at ~533.6 eV is originated from the oxygen atoms chemisorbed at the surface of synthesized materials [116]. The binding energy at ~532.4 e V is ascribed to O2 <sup>−</sup> ions (surface hydroxyl (OH) group) on the synthesized Sn-ZnO (in the oxygen deficient region) and the lowest binding energy at ~529.1 eV is attributed to O2 − ions in the Zn-O structures The binding en‐ ergy at ~530.8 eV is attributed to oxidized metal ions in the synthesized Sn-ZnO such as, O-Sn and O-Zn in the ZnO lattice. Sn 3d spectra (Fig. 26 (d)) presents the doublet binding energies at ~487.2 eV and ~496.7 eV, correspond to Sn 3d5/2 and Sn 3d3/2 respectively. The ap‐ pearance of these peaks indicates the incorporation of Sn dopant in the form of O-Sn in the ZnO lattice [117], as deduced by O 1s XPS results. Moreover, the energy gap of ~9.5 eV is observed between these two peaks which resembles to the reported value [118]. It is ob‐ served that since no diffraction peaks corresponding to the SnO and SnO2 are observed in the XRD spectra therefore, the O-Sn bonding could be considered as the substitutional dop‐ ing of Sn-ions into the ZnO lattice.

absorption wavelength of ZnO has significantly shifted from ~389 nm to ~406 nm and its band gap has changed from ~3.18 eV to ~3.05 eV. It has arisen due to the presence of interstitially embedded Sn-ion into ZnO nanomaterials. This small variation in band gaps again confirms

**Figure 25.** UV-DRS spectra of ZnO and Sn-ZnO nanostructures. Reprinted with permission from [Ameen S., 2012],

The XPS survey (Fig. 26 (a)) spectrum displays all Zn 2p, O 1s and Sn 3d binding energy peaks with very small C 1s binding energy. Sn-ZnO nanomaterials show the doublet bind‐ ing energies at ~1022.1 eV and ~1046.1 eV in Zn 2p XPS spectra (Fig. 26 (b)) which corre‐ spond to Zn 2p3/2 and Zn 2p1/2 respectively. The energy difference between doublet binding energies is calculated to ~24 eV, which is in excellent agreement with the standard value of ~22.97 eV. The deconvoluted O 1s XPS presents the four binding energies peaks at ~533.6 eV, ~532.4 eV, ~530.8 eV and ~529.1 eV, as shown in Fig. 26 (c). The highest binding energy at ~533.6 eV is originated from the oxygen atoms chemisorbed at the surface of synthesized

(OH) group) on the synthesized Sn-ZnO (in the oxygen deficient region) and the lowest

ergy at ~530.8 eV is attributed to oxidized metal ions in the synthesized Sn-ZnO such as, O-Sn and O-Zn in the ZnO lattice. Sn 3d spectra (Fig. 26 (d)) presents the doublet binding energies at ~487.2 eV and ~496.7 eV, correspond to Sn 3d5/2 and Sn 3d3/2 respectively. The ap‐ pearance of these peaks indicates the incorporation of Sn dopant in the form of O-Sn in the ZnO lattice [117], as deduced by O 1s XPS results. Moreover, the energy gap of ~9.5 eV is

−

<sup>−</sup> ions (surface hydroxyl

ions in the Zn-O structures The binding en‐

materials [116]. The binding energy at ~532.4 e V is ascribed to O2

the Sn-ion doping into ZnO nanomaterials.

234 Solar Cells - Research and Application Perspectives

Chem. Eng. J. 187 (2012) 351 ©2012, Elsevier Ltd.

binding energy at ~529.1 eV is attributed to O2

**Figure 26.** Survey (a), Zn 2p (b), O 1s (c), and Sn 3d (d) XPS spectra of Sn-ZnO nanostructures. Reprinted with permis‐ sion from [Ameen S., 2012], Chem. Eng. J. 187 (2012) 351. ©2012, Elsevier Ltd.

DSSC fabricated (Fig. 27) with Sn-ZnO photoanode depicts reasonably high solar-to -electricity conversion efficiency of ~1.82% with JSC of 5.1 mA/cm2 , VOC of 0.786 V and FF of 0.45. While, DSSC with ZnO photoanode shows relatively low conversion efficiency of ~1.49% with JSC of ~4.05 mA/cm2 , VOC (~0.761 V) and FF of ~0.48. Noticeably, the conversion efficiency and JSC are considerably enhanced by ~20% and ~21% respectively after Sn-ion doping into ZnO nano‐ structures. Moreover, it could be explained that Sn-ZnO nanostructures might due to the increase of high charge collection and the transfer of electrons at the interface of Sn-ZnO and the electrolyte layer. The dopants like Sn, is known to enhance the electrons transport capacity and electron mobility of ZnO nanomaterials [119]. Moreover, the Sn-ion doping into ZnO nanostructures might increase the specific surface area by lowering the particle size and arranging into spindle shaped morphology, which might contribute to high dye absorption. The increased photocurrent density and the improved photovoltaic performance might also result from high dye absorption and the improved electron transport by Sn-ZnO nanostruc‐ tures, leading the enhancement in light harvesting efficiency and photo-excited electron transportation under sun light. Therefore, the arrangement of ZnO nanoparticles into Sn-ZnO spindle shaped, and good optical properties of Sn-ZnO are crucial to improve the conversion efficiency and the photocurrent density of the fabricated DSSCs.

**Figure 27.** J-V curve of the DSSC fabricated with ZnO and Sn-ZnO nanostructures based photoanodes. Reprinted with permission from [Ameen S., 2012], Chem. Eng. J. 187 (2012) 351. ©2012, Elsevier Ltd.

## **7.2. Ga doped ZnO nanostructures with improved electrical properties**

ZnO nanomaterials along with conjugated polymers like PANI, PPy and poly (3-alkylthio‐ phene) comports the high quality organic/inorganic Schottky diodes [120]. Recently, ZnO-PANI films sandwiched between indium tin oxide (ITO) and a Pt electrode have displayed the linear I-V behavior [121]. The effects of Ga ion doping on ZnO NPs have been studied by Ameen et al on the basis of optical and electrical properties of the fabricated heterostructure devices [122]. The morphology of the synthesized ZnO and Ga-ZnO NPs were studied by FESEM and TEM analysis, as shown in Fig. 28 (a-d). The synthesized ZnO NPs obtain an average diameter of ~20–25 nm. After Ga ion doping, the average diameter increases to ~30–35 nm by the agglom‐ eration of NPs due to entrapping and the substantive influence of Ga ion with ZnO NPs.

The optical properties of ZnO and Ga-doped ZnO NPs were studied by the UV-Vis spectra, as shown in Fig. 29 (a, b). ZnO and Ga-doped ZnO NPs present the absorption in the UV region with strong absorption peak at ~370 nm and ~378 nm respectively, corresponding to the

Metal Oxide Nanomaterials, Conducting Polymers and Their Nanocomposites for Solar Energy http://dx.doi.org/10.5772/51432 237

arranging into spindle shaped morphology, which might contribute to high dye absorption. The increased photocurrent density and the improved photovoltaic performance might also result from high dye absorption and the improved electron transport by Sn-ZnO nanostruc‐ tures, leading the enhancement in light harvesting efficiency and photo-excited electron transportation under sun light. Therefore, the arrangement of ZnO nanoparticles into Sn-ZnO spindle shaped, and good optical properties of Sn-ZnO are crucial to improve the conversion

**Figure 27.** J-V curve of the DSSC fabricated with ZnO and Sn-ZnO nanostructures based photoanodes. Reprinted with

ZnO nanomaterials along with conjugated polymers like PANI, PPy and poly (3-alkylthio‐ phene) comports the high quality organic/inorganic Schottky diodes [120]. Recently, ZnO-PANI films sandwiched between indium tin oxide (ITO) and a Pt electrode have displayed the linear I-V behavior [121]. The effects of Ga ion doping on ZnO NPs have been studied by Ameen et al on the basis of optical and electrical properties of the fabricated heterostructure devices [122]. The morphology of the synthesized ZnO and Ga-ZnO NPs were studied by FESEM and TEM analysis, as shown in Fig. 28 (a-d). The synthesized ZnO NPs obtain an average diameter of ~20–25 nm. After Ga ion doping, the average diameter increases to ~30–35 nm by the agglom‐ eration of NPs due to entrapping and the substantive influence of Ga ion with ZnO NPs.

The optical properties of ZnO and Ga-doped ZnO NPs were studied by the UV-Vis spectra, as shown in Fig. 29 (a, b). ZnO and Ga-doped ZnO NPs present the absorption in the UV region with strong absorption peak at ~370 nm and ~378 nm respectively, corresponding to the

permission from [Ameen S., 2012], Chem. Eng. J. 187 (2012) 351. ©2012, Elsevier Ltd.

**7.2. Ga doped ZnO nanostructures with improved electrical properties**

efficiency and the photocurrent density of the fabricated DSSCs.

236 Solar Cells - Research and Application Perspectives

**Figure 28.** FESEM (a, b) and TEM (c, d) images of ZnO NPs and Ga-ZnO NPs. Reprinted with permission from [Ameen S., 2011], Microchim. Acta 172 (2011) 471. © 2012, Springerlink Ltd.

characteristic band of wurtzite hexagonal ZnO nanomaterials [123]. A considerable red shift from ~370 nm to ~378 nm after Ga ion doping is seen in the absorption peak of Ga-ZnO NPs and results that the band gap of ZnO NPs has changed from ~3.4 eV to ~3.26 eV due to the presence of interstitially embedded Ga ion on the surface of ZnO NPs. Thus, the small variation in band gaps again confirms the Ga doping on the surface of ZnO NPs.

Fig. 30 shows the XPS of PANI/Ga-ZnO thin film electrodes. The Carbon (C 1s), oxygen (O 1s), nitrogen (N 1s) and Zinc (Zn 2p) peaks at ~284.4 eV, ~529.8 eV, ~400.9 eV and ~1019.4/1042.5 eV are taken to investigate the interaction between PANI and Ga-ZnO NPs. The deconvoluted C 1s peak at ~284.4 eV presents four resolved peaks at ~289.1, ~286.8, ~285.7 eV and ~284.9 eV (Fig. 30 (a)) and are ascribed to C = O/C–O bond, C–N+ /C = N+ bond, neutral C–C/C–H bond of PANI backbone and C of the benzonoid ring showing a combination of protonation of imine and amine sites via shake-up processes [124]. Figure 30(b) shows the four O 1s XPS resolved peaks of PANI/Ga-ZnO thin film. The main peak at ~529.8 eV confirms the nature of oxygen atom originated from metal oxide [125] i.e. the oxide of ZnO NPs. Zn 2p XPS of PANI/Ga-ZnO thin film typically exhibits the doublet peaks at ~1019.4 eV/1042.5 eV (Fig. 30 d), suggests that Zn atoms are linked with oxide bond in Ga doped ZnO NPs. Moreover, Fig. 30(e) shows the Ga 2p peak at ~1116.2 eV which confirms the doping of ZnO with the Ga+2 oxidation state [126].

**Figure 29.** UV-Vis spectra of ZnO (a) and Ga-ZnO NPs (b). Reprinted with permission from [Ameen S., 2011], Micro‐ chim Acta 172 (2011) 471© 2012, Springerlink Ltd

N 1s XPS spectrum of PANI/Ga-ZnO thin film (Fig. 30(c)) exhibits the bonding between imine group of PANI and Ga-ZnO. The centered peak at ~400.9 eV ascribes quinoid di-imine nitrogen of PANI [127]. The main peak at ~400.9 eV is resolved into four hide peaks at ~400.5, ~401.3, ~401.9 and ~402.8 eV which correspond to benzenoid di-amine nitrogen, quinoid di-imine nitrogen, positively charged nitrogen (−N<sup>+</sup> ) and the protonated imine (=N+ ) respectively. It is known that the protonation of PANI produces electronic defects such as polarons or bipolarons which might form by the addition of protons to the neutral polymer chain. In this case, positively shifted binding energy at ~401.9 and ~402.8 eV might exhibit the participation of protonated N atom for the bond formation between PANI and Ga-ZnO. These two charged nitrogen species (−N<sup>+</sup> and =N+ ) are originated from these defect states [128] and are observed in N 1s results of PANI/Ga-ZnO thin film. In conclusion, the PANI and Ga-ZnO are interacted to each other by the formation of partial hydrogen bonding between two charged nitrogen species (−N<sup>+</sup> and =N+ ) of PANI and surface hydroxyl of Ga-ZnO.

The I-V characteristics of Pt/PANI/ZnO and Pt/PANI/Ga-ZnO heterostructure devices are obtained at 298 K with applied voltage from −1 V to +1 V, shown in Fig. 31. Both the devices display non-linear and rectifying behavior of I–V curves due to the existence of Schottky barrier via a Schottky contact at the interfaces of Pt layer and PANI-ZnO thin film layer. Pt/PANI/ZnO device shows almost Ohmic or very weak rectifying behavior (Fig. 31(a)) that attains very low turn-on voltage (~0.0005 V) with least current (~0.002 mA). Similarly, in forward bias, a breakdown voltage (~0.05 V) and high leakage current (~0.015 mA) indicate poor I-V charac‐ teristics of Pt/PANI/ZnO device. Whereas, Pt/PANI/Ga-ZnO device (Fig. 31 (b)) presents

Metal Oxide Nanomaterials, Conducting Polymers and Their Nanocomposites for Solar Energy http://dx.doi.org/10.5772/51432 239

N 1s XPS spectrum of PANI/Ga-ZnO thin film (Fig. 30(c)) exhibits the bonding between imine group of PANI and Ga-ZnO. The centered peak at ~400.9 eV ascribes quinoid di-imine nitrogen of PANI [127]. The main peak at ~400.9 eV is resolved into four hide peaks at ~400.5, ~401.3, ~401.9 and ~402.8 eV which correspond to benzenoid di-amine nitrogen, quinoid di-imine

**Figure 29.** UV-Vis spectra of ZnO (a) and Ga-ZnO NPs (b). Reprinted with permission from [Ameen S., 2011], Micro‐

known that the protonation of PANI produces electronic defects such as polarons or bipolarons which might form by the addition of protons to the neutral polymer chain. In this case, positively shifted binding energy at ~401.9 and ~402.8 eV might exhibit the participation of protonated N atom for the bond formation between PANI and Ga-ZnO. These two charged

in N 1s results of PANI/Ga-ZnO thin film. In conclusion, the PANI and Ga-ZnO are interacted to each other by the formation of partial hydrogen bonding between two charged nitrogen

The I-V characteristics of Pt/PANI/ZnO and Pt/PANI/Ga-ZnO heterostructure devices are obtained at 298 K with applied voltage from −1 V to +1 V, shown in Fig. 31. Both the devices display non-linear and rectifying behavior of I–V curves due to the existence of Schottky barrier via a Schottky contact at the interfaces of Pt layer and PANI-ZnO thin film layer. Pt/PANI/ZnO device shows almost Ohmic or very weak rectifying behavior (Fig. 31(a)) that attains very low turn-on voltage (~0.0005 V) with least current (~0.002 mA). Similarly, in forward bias, a breakdown voltage (~0.05 V) and high leakage current (~0.015 mA) indicate poor I-V charac‐ teristics of Pt/PANI/ZnO device. Whereas, Pt/PANI/Ga-ZnO device (Fig. 31 (b)) presents

) of PANI and surface hydroxyl of Ga-ZnO.

) and the protonated imine (=N+

) are originated from these defect states [128] and are observed

) respectively. It is

nitrogen, positively charged nitrogen (−N<sup>+</sup>

chim Acta 172 (2011) 471© 2012, Springerlink Ltd

238 Solar Cells - Research and Application Perspectives

and =N+

nitrogen species (−N<sup>+</sup>

and =N+

species (−N<sup>+</sup>

**Figure 30.** (a) C 1 s, (b) O 1 s, (c) N 1 s, (d) Zn 2p and (e) Ga 2p XPS spectra of PANI/Ga-ZnO thin film electrode. Reprint‐ ed with permission from [Ameen S., 2011], Microchim. Acta 172 (2011) 471©2012, Springerlink Ltd.

rectifying behavior of lower turn on voltage (~0.4 V) with least current (~0.09 mA) and breakdown voltage (~0.56 V) with high leakage current (~0.5 mA). The I-V properties of Pt/ PANI/Ga-ZnO device are considerably better than the data reported elsewhere on PANI/ZnO and PANI based heterostructure devices [129]. Herein, the Ga ion doping to ZnO NPs might generate high density of minority charge carriers and the efficient charge movement at the junction of PANI and Ga-ZnO interfaces, resulting in the high leakage current with moderate turn on and breakdown voltage [130]. Additionally, the improved I-V properties might result from the molecular geometry of PANI chains and the increased electronic conduction by Ga ions in ZnO NPs which likely generate the hopping effect.

**Figure 31.** I-V characteristics of (a) Pt/PANI/ZnO (b) Pt/PANI/Ga- ZnO heterostructure device. ln(I) versus (V) plots of (c) Pt/PANI/ZnO and (d) Pt/PANI/Ga-ZnO heterostructure devices. (e) the schematic of Pt/PANI/Ga-ZnO heterostruc‐ ture device. Reprinted with permission from [Ameen S., 2011], Microchim. Acta 172 (2011) 471©2012, Springerlink Ltd.

### **8. Metal oxides as nanofillers in polymer electrolytes**

The inorganic semiconductor especially metal oxides nanomaterials as nanofillers are con‐ ceived to improve the mechanical, thermal, interfacial, and ionic conductivity properties of the polymer electrolytes, which could effectively utilize for high performance solid-state DSSCs. In general, the introduction of inorganic NPs in the polymer alters the conduction mechanisms, which assigns to the ions conduction. In 1998, Croce et al. [131] studied the enhancement of the ionic conductivity of polymer electrolytes by the addition of TiO2 and other NPs. Later, a ternary component polymer-gel electrolyte with TiO2 NPs was prepared by Kang et al and explicated that these NPs led to a light-scattering effect [132]. The fabricated DSSC with the ternary component polymer electrolyte showed a high overall conversion efficiency of ~7.2% under 100 mW/cm2 . Falaras et al [133] developed the polymer composite electrolyte by the addition of commercial TiO2 nanoparticles (NPs, P25, Degussa) consisted of polyethylene oxide (PEO), LiI, and I2. The addition of TiO2 NPs considerably prevented the re-crystallization and decreased the degree of crystallinity of PEO due to their large surface area. The differential scanning calorimetry (DSC) studies revealed that the introduction of TiO2 NPs caused a significant increase in the glass transition temperature of PEO, which indicated the incorpo‐ ration of the polymer to the inorganic oxide fillers. The fabricated DSSCs with TiO2-PEO nanocomposite electrolyte achieved a reasonably high overall conversion efficiency of ~4.2% with a JSC of ~7.2 mA/cm2 , VOC ~0.664 V, and FF ~0.58 at ~65.6 mW/cm2 [134]. Additionally, other research groups have also used TiO2 NPs as nanofillers and explained the effects of nanofillers on different polymer electrolytes. Recently, Akhtar et al [135] reported a composite electrolyte of polyethylene glycol methyl ether (PEGME) and TiO2 NPs and demonstrated the heat treatment effects on the properties of PEGME-TiO2 composite electrolyte. It was found that the heat treatment was an essential step to improve morphology, amorphicity, and ionic conductivity of PEGME-TiO2 composite electrolytes. From AFM images (Fig. 32), TiO2 particles with ~20-30 nm size are well distributed on the PEGME matrix in the case of PEGME-TiO2 composite film (Fig. 32(a). However, TiO2 particles are aggregated and become a bigger size (40-60 nm) on the polymer matrix in PEGME-TiO2/80o C (Fig. 32 (b)). From the film roughness analysis (Fig. 32 (c, d)), it is observed that the surface roughness of the PEGME-TiO2/80o C and PEGME-TiO2 composites are estimated to be about ~23.1 nm and ~18.5 nm for the root mean square roughness (*R*rms), and ~12 nm and 8~.8 nm for the average surface roughness (*Ra*), respectively. Generally, it has been well known that the low surface roughness of the polymer composite film is ascribed to the high-crystallized surface of thr composite materials. There‐ fore, the crystallinity of PEGME-TiO2/80o C might lower than PEGME-TiO2 because the former exhibits the higher *R*rms and *Ra* value than later composite film. Consistently, the 3D AFM images (Fig. 32 (d)) of PEGME-TiO2/80o C exhibit a highly rough surface morphology with nonuniformly distributed TiO2 particles into the PEGME matrix, while a highly uniform and less rough surface is observed in PEGME-TiO2 composite film before heating (Fig. 32 (c)). This rough morphology of PEGME-TiO2/80o C might create free spaces and voids in which the I<sup>−</sup> /I3 − ions could easily migrate, which suggest the PEGME-TiO2/80o C as excellent electrolyte materials. With the improved morphology of PEGME-TiO2 composite, the electrolyte shows the high ionic conductivity of ~1.9 mS/cm as compared to the PEGME acid (1.2 mS/cm) and PEGME-TiO2 (0.92 mS/cm) which results from the enhanced morphological properties in terms of high roughness and amorphicity after heating of PEGME-TiO2. The Raman spectra (Fig. 33) of the PEGME-acid, PEGME-TiO2, and PEGME-TiO2/80o C composite electrolytes exhibit a significant peak at the range of 110-115 cm−1, which is ascribed to the symmetric stretch of I3 − species in redox electrolytes [136]. The heat treatment on PEGME-TiO2 drastically increased the strong Raman peak, indicating a significant increase in the amount of I3 − species in redox

**Figure 31.** I-V characteristics of (a) Pt/PANI/ZnO (b) Pt/PANI/Ga- ZnO heterostructure device. ln(I) versus (V) plots of (c) Pt/PANI/ZnO and (d) Pt/PANI/Ga-ZnO heterostructure devices. (e) the schematic of Pt/PANI/Ga-ZnO heterostruc‐ ture device. Reprinted with permission from [Ameen S., 2011], Microchim. Acta 172 (2011) 471©2012, Springerlink

The inorganic semiconductor especially metal oxides nanomaterials as nanofillers are con‐ ceived to improve the mechanical, thermal, interfacial, and ionic conductivity properties of the polymer electrolytes, which could effectively utilize for high performance solid-state DSSCs. In general, the introduction of inorganic NPs in the polymer alters the conduction mechanisms,

**8. Metal oxides as nanofillers in polymer electrolytes**

Ltd.

240 Solar Cells - Research and Application Perspectives

electrolytes. It might attribute to the increased bond strength between PEGME and TiO2 and high roughness of the composite materials, which might help to absorb a large amount of the iodide couple as compared to PEGME-acid and PEGME-TiO2 composite electrolytes. The increased intensity of peak suggests that a large amount of I3 − species is formed in the PEGME-TiO2/80o C composite electrolyte upon heat treatment. In general, the diffusional I<sup>−</sup> /I3 <sup>−</sup> ions migration in the redox electrolyte is responsible for the ionic conductivity of electrolyte, which causes electron exchange between ions by electronic conduction process [137]. The electronic conduction in redox electrolyte depends on the formation of I3 − ions. Raman results show the proportional relation between the ionic conductivity and concentration of I3 <sup>−</sup> species of the composite electrolytes, which is directly related to relative intensity of Raman peak. The enhanced ionic conductivity of PEGME-TiO2/80o C composite electrolyte might associate with the formation of high I3 − species in redox electrolyte. However, low ionic conductivity in PEGME-acid and PEGME-TiO2 composite electrolyte results from the low relative intensity of the Raman peak and less formation of I3 − species in redox electrolyte. Therefore, a heat treatment step plays an essential role to prepare the improved composite electrolyte with enhanced ionic conductivity. DSSC fabricated with PEGME-TiO2/80o C composite electrolyte shows the maximum overall conversion efficiency of ~3.1% with a JSC of ~8.9 mA/cm2 , VOC of ~0.625 volt, and FF of ~56.2%. The conversion efficiency and JSC of DSSCs with PEGME-TiO2/80o C composite electrolytes is higher than those of fabricated with PEGME-acid (~1.3%) and PEGME-TiO2 (~2.4%) electrolytes. This could be expected from the enhanced ionic conductivity and enlargement of the amorphous phase of the polymer upon heat treatment. The heat treatment on PEGME-TiO2 composites enhances the ionic conductivity and crosslinking properties of composite electrolyte, which are essential factors to achieve the high current density and high PV performance. Furthermore, Akhtar et al [138] investigated the effect of titania nanotubes (NTs) as nanofillers on the properties of PEG-based electrolytes and fabricated solid-state DSSCs. PEG-TiNT electrolytes with 10% of TiNTs exhibit the high penetration and complete filling into the pores of the TiO2 film, as shown in Fig. 34. The XPS studies (Fig. 35) were carried out to elucidate the strong interaction between PEG and TiNTs. PEG-TiNT10 electrolyte shows the highest interaction between the titanium atoms of the NTs and the polymer network as compared to those of other PEG-TiNTs electrolytes. This results to the decrease in the crystallinity degree of the polymer after introduction of the NTs which achieves the highest ionic conductivity of ~2.4×10−3 S/cm. DSSC fabricated with PEG-TiNT composite electrolyte (Fig. 36) exhibits the maximum overall conversion efficiency of ~4.4% with JSC of ~9.4 mA/cm2 , VOC of ~0.73 V, and FF of ~0.65 under 100 mW/cm2 irradiation. No significant decrease of the conversion efficiency for 30 days was observed in DSSCs fabricated with PEG-TiNT10 (inset of Fig. 36), indicating the high stability of the composite electrolytes. The lower current density in PEG-TiNT20 is due to its lower ion conductivity, lower penetra‐ tion, and weak interaction between PEG to TiNTs. It is proved that the better penetration into the pores of the TiO2 layer was obtained at a ratio of TiNT and PEG in the composite electrolyte (PEG-TiNT10). Thus, due to the better interfacial contact between the electrolyte and TiO2 layer, high ion conductivity is obtained, which enhances the photocurrent density. Moreover, the PEG-TiNT composite electrolytes might facilitate the movement of electrons in the redox (I− /I3 − ) couple due to the fast electron transfer characteristics of NTs with less grain-boundary, in comparison to NPs. The other metal oxide nanomaterials, such as ZnO, SiO2 and Al2O3, have been introduced as nanofillers to different polymer electrolytes [139]. Caruso et al. prepared the polymer composite electrolyte with the composition of PEO-poly (vinylidenefluoride) (PVDF) and SiO2 NPs and applied as a solid electrolyte for solid-state DSSCs [140]. These kinds of solid state electrolytes presented high viscosity to the solid state electrolyte with TiO2-based polymer composites. They used a vacuum technique for introducing the composite polymer electrolytes into the dye-sensitized TiO2 electrode which showed that the vacuum method exhibited a better performance than those prepared via the conventional drop casting method. This approach improved the fulfilling of the photoelectrode with a solid electrolyte by vacuum technique, but the optimization of the electrolyte composition is still an important issue. In this regards, Xia et al utilized ZnO NPs as nanofillers for preparing the composite polymeric electrolyte of poly (ethylene glycol methyl ether) (PEGME) [141]. The PEGME was first grafted onto the surface of ZnO NPs through covalent bond formation by a chemical process. The solid composite electrolyte consisted of KI and I2 dissolved in PEGME and ~24 wt% of the polymergrafted ZnO NPs. The obtained prepared electrolyte showed that the ionic conductivity increased as the salt concentration increased and reached a maximum value of ~3.3×10−4 S/cm and then decreased, acting as a classical polymer electrolyte system. DSSC fabricated with polymer-grafted NPs electrolyte presented the lower conversion efficiency of ~3.1% compared to that of DSSC with a liquid electrolyte (~4.0%). After the addition of polymer-grafted ZnO NPs in liquid electrolyte, the VOC of DSSC increased by ~0.13 V while the JSC decreased, this was probably due to the high viscosity of the gel electrolyte. Another report addressed the new polymer electrolyte system of Al2O3 NPs with different sizes and a PVDF derivative and polyacrylonitrile in an ionic-liquid-based electrolyte [142]. The diffusion coefficient of I3 − ions altered by the addition Al2O3 NPs. The variation in sizes of Al2O3 NPs greatly influenced the charge transfer rate at the electrolyte and semiconducting layer interfaces. In this report, the imidazolium cations might adsorb on the NP surface, which might help in the charge transfer at counter and anions I<sup>−</sup> /I3 − gather around them. Some researchers have recently used clay-like NPs as nanofillers in the polymer electrolytes and applied to the DSSCs [143]. Nogueira et al. [144] examined the incorporation of a montmorillonite (MMT) derivative to a polymer electrolyte based on a poly-(ethylene oxide) copolymer, the plasticizer GBL, and Li I/I2. The improved ionic conductivity of the composite electrolyte attributed to the large number of charge carriers introduced into the complex after the addition of the clay. The addition of 5 wt % MMT promoted the increase in the mechanical stability of the nanocomposite polymer electrolyte film, resulted in the lower deformation as compared to the film without any clay. From their observations, it was found that the addition of MMT clay to the plasticized polymer electrolyte not only increased the ionic conductivity but also improved the solidification of the electrolyte. These improvements led to the mechanical stability of the polymer composite films and the stability of DSSCs as well. DSSCs fabricated with the nanocomposites polymer electrolyte showed reasonable conversion efficiencies of ~1.6% and ~3.2% at 100 mW/cm2 and 10 mW/cm2 , respectively. The device presented very poor FF values of ~0.40 at 100 mW/cm2 , which was attributed to the low penetration of the composite electrolyte into the pores of the TiO2 film. The MMT clay as nanofillers was also used by Lin et al. They prepared the nano‐ composites of poly (nisopropylacrylamide) with MMT clay to a liquid electrolyte system as a

electrolytes. It might attribute to the increased bond strength between PEGME and TiO2 and high roughness of the composite materials, which might help to absorb a large amount of the iodide couple as compared to PEGME-acid and PEGME-TiO2 composite electrolytes. The

C composite electrolyte upon heat treatment. In general, the diffusional I<sup>−</sup>

migration in the redox electrolyte is responsible for the ionic conductivity of electrolyte, which causes electron exchange between ions by electronic conduction process [137]. The electronic

composite electrolytes, which is directly related to relative intensity of Raman peak. The

PEGME-acid and PEGME-TiO2 composite electrolyte results from the low relative intensity of

treatment step plays an essential role to prepare the improved composite electrolyte with

~0.625 volt, and FF of ~56.2%. The conversion efficiency and JSC of DSSCs with PEGME-

and PEGME-TiO2 (~2.4%) electrolytes. This could be expected from the enhanced ionic conductivity and enlargement of the amorphous phase of the polymer upon heat treatment. The heat treatment on PEGME-TiO2 composites enhances the ionic conductivity and crosslinking properties of composite electrolyte, which are essential factors to achieve the high current density and high PV performance. Furthermore, Akhtar et al [138] investigated the effect of titania nanotubes (NTs) as nanofillers on the properties of PEG-based electrolytes and fabricated solid-state DSSCs. PEG-TiNT electrolytes with 10% of TiNTs exhibit the high penetration and complete filling into the pores of the TiO2 film, as shown in Fig. 34. The XPS studies (Fig. 35) were carried out to elucidate the strong interaction between PEG and TiNTs. PEG-TiNT10 electrolyte shows the highest interaction between the titanium atoms of the NTs and the polymer network as compared to those of other PEG-TiNTs electrolytes. This results to the decrease in the crystallinity degree of the polymer after introduction of the NTs which achieves the highest ionic conductivity of ~2.4×10−3 S/cm. DSSC fabricated with PEG-TiNT composite electrolyte (Fig. 36) exhibits the maximum overall conversion efficiency of ~4.4%

C composite electrolytes is higher than those of fabricated with PEGME-acid (~1.3%)

, VOC of ~0.73 V, and FF of ~0.65 under 100 mW/cm2

) couple due to the fast electron transfer characteristics of NTs with less grain-boundary,

significant decrease of the conversion efficiency for 30 days was observed in DSSCs fabricated with PEG-TiNT10 (inset of Fig. 36), indicating the high stability of the composite electrolytes. The lower current density in PEG-TiNT20 is due to its lower ion conductivity, lower penetra‐ tion, and weak interaction between PEG to TiNTs. It is proved that the better penetration into the pores of the TiO2 layer was obtained at a ratio of TiNT and PEG in the composite electrolyte (PEG-TiNT10). Thus, due to the better interfacial contact between the electrolyte and TiO2 layer, high ion conductivity is obtained, which enhances the photocurrent density. Moreover, the PEG-TiNT composite electrolytes might facilitate the movement of electrons in the redox

−

shows the maximum overall conversion efficiency of ~3.1% with a JSC of ~8.9 mA/cm2

−

species in redox electrolyte. However, low ionic conductivity in

−

species is formed in the PEGME-

ions. Raman results show the

C composite electrolyte

, VOC of

irradiation. No

C composite electrolyte might associate with

species in redox electrolyte. Therefore, a heat

/I3 <sup>−</sup> ions

<sup>−</sup> species of the

increased intensity of peak suggests that a large amount of I3

conduction in redox electrolyte depends on the formation of I3

enhanced ionic conductivity. DSSC fabricated with PEGME-TiO2/80o

enhanced ionic conductivity of PEGME-TiO2/80o

the Raman peak and less formation of I3

−

the formation of high I3

242 Solar Cells - Research and Application Perspectives

with JSC of ~9.4 mA/cm2

(I− /I3 −

proportional relation between the ionic conductivity and concentration of I3

TiO2/80o

TiO2/80o

gelator and applied as solid polymer electrolyte. The poly (nisopropylacrylamide)-MMT electrolyte-based DSSC achieved a relatively high conversion efficiency of ~5.4% with a JSC of ~12.6 mA/cm2 , VOC of ~0.73 V, and FF of ~0.59, whereas the DSSC prepared with the electrolyte gelled with the pure polymer presented lower photovoltaic parameters of JSC (~7.28 mA/cm2 ), VOC (~0.72 V), FF (0.60), and conversion efficiency (~3.2%) at 100 mW/cm2 . From the electro‐ chemical impedance spectroscopy, a considerable decrease in impedance values was observed by DSSC fabricated with nanocomposite-gelled electrolyte. The impedance at the electrolyte/ dye-coated TiO2 interface, and the Nernstian diffusion within the electrolytes were decreased, resulted in the high photocurrent density leading to the high performance of DSSCs. They also investigated the molar conductivity of the nanocomposite-gelled electrolytes to explain the high ionic conductivity and improved electrochemical behavior of electrolyte.

**Figure 32.** Topographic and three-dimensional AFM images of (a), (c) PEGME-TiO2 and (b), (d) PEGMETiO2/80C. Re‐ printed with permission from [Akhtar, 2011], Mater. Chem. Phys. 127 (2011) 479.© 2011, Elsevier Ltd.

Metal Oxide Nanomaterials, Conducting Polymers and Their Nanocomposites for Solar Energy http://dx.doi.org/10.5772/51432 245

gelator and applied as solid polymer electrolyte. The poly (nisopropylacrylamide)-MMT electrolyte-based DSSC achieved a relatively high conversion efficiency of ~5.4% with a JSC of

gelled with the pure polymer presented lower photovoltaic parameters of JSC (~7.28 mA/cm2

chemical impedance spectroscopy, a considerable decrease in impedance values was observed by DSSC fabricated with nanocomposite-gelled electrolyte. The impedance at the electrolyte/ dye-coated TiO2 interface, and the Nernstian diffusion within the electrolytes were decreased, resulted in the high photocurrent density leading to the high performance of DSSCs. They also investigated the molar conductivity of the nanocomposite-gelled electrolytes to explain the

**Figure 32.** Topographic and three-dimensional AFM images of (a), (c) PEGME-TiO2 and (b), (d) PEGMETiO2/80C. Re‐

printed with permission from [Akhtar, 2011], Mater. Chem. Phys. 127 (2011) 479.© 2011, Elsevier Ltd.

VOC (~0.72 V), FF (0.60), and conversion efficiency (~3.2%) at 100 mW/cm2

high ionic conductivity and improved electrochemical behavior of electrolyte.

, VOC of ~0.73 V, and FF of ~0.59, whereas the DSSC prepared with the electrolyte

),

. From the electro‐

~12.6 mA/cm2

244 Solar Cells - Research and Application Perspectives

**Figure 33.** Raman spectra of PEGME-acid, PEGME–TiO2 and PEGME–TiO2/80 ◦C composite electrolytes. Reprinted with permission from [Akhtar., 2011], Mater. Chem. Phys. 127 (2011) 479. © 2011, Elsevier Ltd.

**Figure 34.** Cross-section and top view (inset) FE-SEM images of the TiO2 thin film (a) before electrolyte filling and af‐ ter introducing composite electrolytes of (b) PEG-TiNT5, (c) PEG-TiNT10, and (d) PEG-TiNT20. Reprinted with permis‐ sion from [Akhtar., 2007], Electrochem. Commun. 9 (2007) 2833.© 2007, Elsevier Ltd.

**Figure 35.** XPS spectra of the composite electrolytes. (a) PEG-TiNT5, (b) PEG-TiNT10, (c) PEG-TiNT20, and (d) O 1 s of PEG (—), PEG-TiNT5 (- - - -) PEG-TiNT10 (…..) and PEG-TiNT20 (-.-.-). Reprinted with permission from [Akhtar., 2007], Electrochem. Commun. 9 (2007) 2833.© 2007, Elsevier Ltd.

**Figure 36.** Current–voltage characteristics of DSSC fabricated with composite electrolytes of (a) PEG-TiNT5, (b) PEG-TiNT10, and (c) PEG-TiNT20. Inset shows the stability test of DSSC fabricated with composite electrolyte of PEG-TiNT10. Reprinted with permission from [Akhtar, 2007], Electrochem. Commun*.* 9 (2007) 2833.© 2007, Elsevier Ltd.

## **9. Conclusions**

**Figure 35.** XPS spectra of the composite electrolytes. (a) PEG-TiNT5, (b) PEG-TiNT10, (c) PEG-TiNT20, and (d) O 1 s of PEG (—), PEG-TiNT5 (- - - -) PEG-TiNT10 (…..) and PEG-TiNT20 (-.-.-). Reprinted with permission from [Akhtar., 2007],

Electrochem. Commun. 9 (2007) 2833.© 2007, Elsevier Ltd.

246 Solar Cells - Research and Application Perspectives

In summary, the morphological, structural, crystalline, optical, electrical and photovoltaic properties of conducting polymers, nanocomposites of conducting polymer/inorganic nanomaterials and semiconducting metal oxides have been discussed. The PANI nanocom‐ posites with semiconducting materials have shown the improved penetration and optoelec‐ tronic properties, and applied for the electrical and electronic application such as diodes and solar cells. Here, the uniform distribution of CdS nanomaterials effectively improves the electronic state of PANI like polarons and bipolarons for the high charge carriers and enhances the charge transfer. The unique conducting polymers, particularly PANI nanomaterials have been used as hole transporting material and as counter electrodes for the applications of DSSCs. The metal oxide semiconducting nanomaterials, particularly TiO2 and ZnO nanomaterials, in terms of morphology, surface properties, dye absorption and application in DSSCs are extensively summarized. Various morphologies of metal oxides nanostructures greatly affect the performances of dye absorption, electrical, electrochemical, and photovoltaic devices. The metal oxides semiconducting nanomaterials with different morphologies and sizes enhance the surface-to-volume ratio and produce the highly advanced photoanodes for the efficient DSSCs. The morphologies of metal oxides semiconducting considerably influence the dye absorption, light harvesting and results in increased electron transfer and reduce the recom‐ bination rate during the operation of DSSCs. The photovoltaic properties such as JSC, VOC, FF, and conversion efficiency have significantly improved by altering the sizes and shapes of the metal oxides semiconductors. The chapter also summarizes the use of various metal oxide semiconducting nanomaterials as nanofillers in polymer electrolytes and describes their effect on the properties of polymer electrolytes and the performances of DSSCs. The introduction of metal oxide semiconducting nanomaterials into the polymer matrix has significantly improved the amorphicity, mechanical, thermal and ionic conductivity of polymer electrolytes. The chapter includes some of the polymer composite electrolytes and their photovoltaic properties for DSSCs.

## **Author details**

Sadia Ameen1 , M. Shaheer Akhtar2 , Minwu Song1 and Hyung Shik Shin1

1 Energy Materials & Surface Science Laboratory, Solar Energy Research Center, School of Chemical Engineering, Chonbuk National University, Jeonju, Republic of Korea

2 New and Renewable Energy Material Development Center (NewREC), Chonbuk National University, Jeonbuk, Republic of Korea

## **References**


[8] Plesu N, Ilia G, Pascariu A, Vlase G. Preparation, degradation of polyaniline doped with organic phosphorus acids and corrosion essays of poly aniline-acrylic blends. Synthetic Metals 2006; 156:230-238.

bination rate during the operation of DSSCs. The photovoltaic properties such as JSC, VOC, FF, and conversion efficiency have significantly improved by altering the sizes and shapes of the metal oxides semiconductors. The chapter also summarizes the use of various metal oxide semiconducting nanomaterials as nanofillers in polymer electrolytes and describes their effect on the properties of polymer electrolytes and the performances of DSSCs. The introduction of metal oxide semiconducting nanomaterials into the polymer matrix has significantly improved the amorphicity, mechanical, thermal and ionic conductivity of polymer electrolytes. The chapter includes some of the polymer composite electrolytes and their photovoltaic properties

, Minwu Song1

Chemical Engineering, Chonbuk National University, Jeonju, Republic of Korea

1 Energy Materials & Surface Science Laboratory, Solar Energy Research Center, School of

2 New and Renewable Energy Material Development Center (NewREC), Chonbuk National

[1] Shirakawa H. The discovery of polyacetylene film: the dawning of an era of conducting

[2] MacDiarmid AG. Synthetic metals: a novel role for organic polymers. Angewandte

[3] Gerard M, Chaubey A, Malhotra BD. Biosens. Application of conducting polymers to

[4] Friend RH, Gymer RW, Holmes AB, Burroughes JH, Marks RN, Taliani C, Bradley DD C, Santos DA, Brédas JL, Lögdlund M, Salaneck WR. Electroluminescence in conjugat‐

[5] Sirringhaus H, Tessler N, Friend RH. Integrated Optoelectronic Devices based on.

[6] Morrin A, Wilbeer F, Ngamna O, Moulton SE, Killard AJ, Wallace GG, Novel biosensor fabrication methodology based on processable conducting polyaniline nanoparticles.

[7] Jang J, Bae J, Lee K. Synthesis and characterization of polyaniline nanorods as curing agent and nanofiller for epoxy matrix composite, Polymer 2005;46:3677-3684.

polymers. Angewandte Chemie International Edition 2001;40:2575–2580.

Chemie International Edition 2001; 40:2581–2590.

Conjugated Polymers. Science 1998; 280:1741-1744.

Electrochemistry Communication 2005; 7:317-322.

ed polymers. Nature 1999; 397:121-128.

biosensors. Biosensors and Bioelectronics 2002;17:345-359.

and Hyung Shik Shin1

for DSSCs.

**Author details**

, M. Shaheer Akhtar2

University, Jeonbuk, Republic of Korea

248 Solar Cells - Research and Application Perspectives

Sadia Ameen1

**References**


[37] Mazeikiene R, Statino A, Kuodis Z, Niaura G, Malinauskas A. In situ Raman spectroe‐ lectrochemical study of self-doped polyaniline degradation kinetics. Electrochemistry Communications 2006;8:1082–1086.

[23] Baba A, Onishi K, Knoll W, Advincula RC. Investigating Work Function Tunable Hole-Injection/Transport Layers of Electrodeposited Polycarbazole Network Thin Films.

[24] Pud AA. Stability and degradation of conducting polymers in electrochemical systems.

[25] Min G. Conducting polymers and their applications in the film industry: polyaniline

[26] Davies JE, Less RJ, May I, Rawson JM. Isolation of the first diselenadiazolyl complex,

[27] Hahiwara T, Demura T, Iwata K. Synthesis and properties of electrically conducting

[28] Snauwaert P, Lazzaroni R, Riga J, Verbist JJ. Electronic structure of polyaniline and

[29] Cho MS, Cho YH, Choi HJ, Jhon MS. Synthesis and electrorheological characteristics of polyaniline-coated poly(methyl methacrylate) microsphere: Size effect. Langmuir

[30] Jindal Z, Verma NK. Electrochemical template–assisted fabrication of CdS micro/

[31] Soudi R, Kamal M, Shabaka AA, Abdelrazek EM, Eisa W. Synthesis, characterization and spectroscopic studies of CdS/polyaniline core/shell nanocomposites. Synthetic

[32] Xi Y, Zhou J, Guo H, Cai C, Lin Z. Enhanced photoluminescence in core-sheath CdS– PANI coaxial nanocables: a charge transfer mechanism. Chemical Physics Letters

[33] Seoudi R, Shabaka AA., Kamal M, Abdelrazek EM, Eisa H. Dependence of structural, vibrational spectroscopy and optical properties on the particle sizes of CdS/polyaniline core/shell nanocomposites. Journal of Molecular Structure 2012;10(13):156-162.

[34] Raut BT, Chougule MA, Sen S, Pawar RC, Lee CS, Patil VB. Novel method of fabrication of polyaniline–CdS nanocomposites: Structural, morphological and optoelectronic

[35] Ameen S, Akhtar MS, Kim YS, Shin HS. Synthesis and electrochemical impedance properties of CdS nanoparticles decorated polyaniline nanorods. Chemical Engineer‐

[36] Boyer MI, Quillard S, Louarn G, Froyer G, Lefrant S. Vibrational study of the FeCl3 doped dimer of polyaniline; a good model compound of emeraldine salt. Journal of

polyimide blended films. Synthetic Metals 1999;102(1–3):1163–1166.

Pd3[PhCNSeSeN]2[PPh3]4 2PhMe New J. Chem. 1998;22:763-765

polymers from aromatic amines. Synthetic Metals 1987;18:317-322.

substituted derivatives. Synthetic Metals 1987; 18:335-340.

nanostructures. Physica E 2009;41:1752–1756.

properties. Ceramics International 2012;38:3999-4007.

ing Journal 2012;181–182:806–812.

Physical Chemistry B 2000;104:8952–896.

Journal of Physical Chemistry B 2004;108:18949-18955.

Synthetic Metals 1994;66(1):1–18.

250 Solar Cells - Research and Application Perspectives

2003;19:5875-5881.

Metals 2010;160:479–484.

2005;412:60-63.


[64] ] Lee KM, Chen PY, Hsu CY, Huang JH, Ho WH, Chen HC, Ho KC. A high-performance counter electrode based on poly(3,4-alkylenedioxythiophene) for dye-sensitized solar cells. Journal of Power Sources. 2009;188:313-318.

[50] Kudo N, Shimazaki Y, Ohkita H, Ohoka M, Ito S. Organicinorganic hybrid solar cells based on conducting polymer and SnO2 nanoparticles chemically modified with a

[51] Woo S, Jeong JH, Lyu HK, Jeong S, Sim JH, Kim WH, Han YS, Kim Y. Hybrid solar cells with conducting polymers and vertically aligned silicon nanowire arrays: The effect of

[52] Tan F, Qu S, Wu J, Wang Z, Jin L, Bi Y, Cao J, Liu K, Zhang J. Wang Z., Electrodepostied polyaniline films decorated with nano-islands: Characterization and application as anode buffer layers in solar cells. Solar Energy Materials Solar Cells 2011;95:440-445.

[53] Chang MY, Wu CS, Chen YF, Hsieh BZ, Huang WY, Ho KS, Hsieh TH, Han YK. Polymer solar cells incorporating one-dimensional polyaniline nanotubes. Organic Electronics

[54] Lim TH, Oh KW, Kim SH. Self-assembly supramolecules to enhance electrical conduc‐ tivity of polyaniline for a flexible organic solar cells anode. Solar Energy Materials Solar

[55] Bejbouji H, Vignau LL, Miane J, Dang MT, Oualim EM, Harmouchi M, Mouhsen A. Polyaniline as a hole injection layer on organic photovoltaic cells. Solar Energy

[56] Zhu S., Wei W., Chen X., Jiang M., Zhou Z. Hybrid structure of polyaniline/ZnO nanograss and its application in dye-sensitized solar cell with performance improve‐

[57] Ameen S, Akhtar MS, Song M, Kim YS, Shin HS. Plasma Deposited Polyaniline on Ruthenium Dye Sensitized Zno Thin Film for Dye Sensitized Solar Cell. Advanced

[58] Malherbe R, Martinez J, Reguera E, Navarro E. Journal of Materials Science 1992;28:274.

[59] Ameen S, Akhtar MS, Kim G-S, Kim YS, Yang O-B, Shin HS. Plasma-enhanced polymerized aniline/TiO2 dye-sensitized solar cells. Journal of Alloys and Compounds

[60] Bisquert J. Theory of the Impedance of Electron Diffusion and Recombination in a Thin

[62] Brad AJ, Faulkner LR. Electrochemical Methods: Fundamentals and Applications, John

[63] Yeh MH, Lee CP, Chou CY, Lin LY, Wei HY, Chu CW, Vittal R, Ho KC. Conducting polymer-based counter electrode for a quantum-dot-sensitized solar cell (QDSSC) with

[61] J.R. Macdonald, Impedance Spectroscopy, John Wiley & Sons, New York, 1987.

fullerene derivative. Solar Energy Materials Solar Cells 2007;91:1243-1247.

silicon conductivity. Physica B: Condensed Matter 2012;407:3059-3062.

2008; 9:1136-1139.

252 Solar Cells - Research and Application Perspectives

2009;487:382–386.

Cells 2012;101:232-240.

Materials Solar Cells. 2010;94:176-181.

ment. Journal of Solid State Chemistry 2012;190:174-179.

Layer. Journal of Physical Chemistry B 2002;106:325-333.

a polysulfide electrolyte. Electrochimica Acta, 2011;57:277-284.

Science Letters 2012: doi:10.1166/asl.2012.3305.

Wiley & Sons, New York, 1980, p. 350.


Insulator Substrates. Science of Advanced Materials 2010;2:64-68. (b) Zhang H, Du N, Chen B, Li D, Yang D. Carbon Nanotube-ZnO Nanosphere Heterostructures: Low-Temperature Chemical Reaction Synthesis, Photoluminescence, and Their Application for Room Temperature NH3 Gas Sensor. Science of Advanced Materials 2009;1:13-17.

[90] Akhtar MS, Khan MA, Jeon MS, Yang OB. Controlled synthesis of various ZnO nanostructured materials by capping agents-assisted hydrothermal method for dyesensitized solar cells. Electrochimica Acta 2008;53:7869–7874.

[76] Kawai T, Takahashi H, Matsushima Y, Ogata T, Unuma H. Evaluation of Photocatalytic Activity of TiO2 Thin Films by Spin-Trap ESR Spectroscopy. Science of Advanced

[77] Thomas J, Kumar KP, Mathew S. Enhancement of Sunlight Photocatalysis of Nano TiO2 by Ag Nanoparticles Stabilized with D-Glucosamine. Science of Advanced

[78] Lee SM, Cho SN, Cheon S. Anisotropic Shape Control of Colloidal Inorganic Nano‐

[79] Hyam RS, Bhosale RK, Lee W, Han SH, Hannoyer B, Ogale SB. Room Temperature Synthesis of Rutile TiO2 Hierarchical Nanoneedle Flower Morphology for Dye Sensi‐ tized Solar Cell. Journal of Nanoscience and Nanotechnology 2010;10:5894-5898.

[80] Du G, Wan B, Guo Z, Shen J, Li Y, Liu H. Effect of Annealing on Electrochemical Performance of Anodized TiO2 Nanotubes for Lithium Ion Batteries. Advanced Science

[81] Mor GK, Shankar K, Paulose M, Varghese PK, Grimes CA. Use of Highly-Ordered TiO2 Nanotube Arrays in Dye-Sensitized Solar Cells. Nano Letters, 2006;6:215-218.

[82] Ameen S, Akhtar MS, Kim YS, Shin HS. Controlled synthesis and photoelectrochemical properties of highly ordered TiO2 nanorods. RSC Advances, 2012;2:4807–4813

[83] Shin K, Seok SI, Im SH, Park JH. CdS or CdSe decorated TiO2 nanotube arrays from spray pyrolysis deposition: use in photoelectrochemical cells. Chemical Communica‐

[84] Du PA, Chen HH, Lu YC. Dye sensitized solar cells using well-aligned zinc oxide

[85] Gregg BA. Interfacial processes in the dye-sensitized solar cell. Coordination Chemistry

[86] Lee S, Cho IS, Lee JH, Kim HD, Kim DW, Kim JY, Shin H, Lee JK, Jung HS, Park NG, Kim K, Ko MJ, Hong KS. Two-Step Sol−Gel Method-Based TiO2 Nanoparticles with Uniform Morphology and Size for Efficient Photo-Energy Conversion Devices.

[87] Chakravarty R, Periasamym C. Effect of aluminium doping on structural and optoe‐ lectronic properties of sol-gel derived nanocrystalline ZnO thin film. Science of

[88] Chao CH, Chan CH, Huang JJ, Chang LS, Shih HC. Manipulated the band gap of 1D ZnO nano-rods array with controlled solution concentration and its application for

[89] Mohanta SK, Kim DC, Kong BH, Cho HK, Liu W, Tripathy S. Optical Properties of ZnO Nanorods and Hybrid Structures Grown on p-type GaN/Sapphire and Silicon-on-

nanotip arrays. Applied Physics Letters 2006;89:253513-253615.

Materials, 2010; 2:74-78.

254 Solar Cells - Research and Application Perspectives

Materials 2011;3:59-65.

Letters, 2011;4:469-473.

tion, 2010;46:2385-2387.

Reviews 2004;248:1215-1224.

Chemistry of Materials, 2010;22:1958-1965.

DSSCs. Current Applied Physics 2011;11:S136–S139

Advanced Materials 2011;3:276–283.

crystals. Advanced Materials 2003;15:441-444.


[116] Yang J, Lee J, Im K, Lim S, Influence of Sn-doping in hydrothermal methods on the optical property of the ZnO nanorods. Physica E 2009;42:51–56.

[102] Laudise RA, Kolb ED, Caporason AJ. Hydrothermal growth of large sound crystals of

[103] Ameen S, Akhtar MS., Kim YS, Yang OB, Shin HS. Influence of seed layer treatment on low temperature grown ZnO nanotubes: performances in dye sensitized solar cells,

[104] Wagner CD, Riggs WM, Davis LE, Moulder JF, Muilenberg GE. Handbook of X-ray

[105] Leunga CY, Djurisic AB, Leung YH, Ding L, Yang CL, Ge WK. Influence of the carrier gas on the luminescence of ZnO tetrapod nanowires, Journal of Crystal Growth

[106] Fan JCC, Goodenough JB. X-ray photoemission spectroscopy studies of Sn doped

[107] Martinson ABF, Elam JW, Hupp JT, Pellin MJ. ZnO nanotube based dyesensitized solar

[108] Umar A, Singh P, Al-Ghamdi AA, Al-Heniti S. Direct growth of ZnO nanosheets on FTO substrate for dye-sensitized solar cells applications, Journal of Nanoscience and

[109] Al-Hajry A, Umar A, Hahn YB, Kim DH. Growth, properties and dyesensitized solar cells–applications of ZnO nanorods grown by lowtemperature solution process.

[110] Ilican S, Caglar Y, Caglar M, Yakuphanoglu F. Structural, optical and electrical properties of F-doped ZnO nanorod semiconductor thin films deposited by solgel

[111] Peiteado M, Iglesias Y, Fernandez JF, De Frutos J, Caballero AC. Microstructural development of tin-doped ZnO bulk ceramics. Materials Chemistry and Physics

[112] Tsay CY, Cheng HC, Tung YT, Tuan WH, Lin CK. Effect of Sn-doped on microstructural and optical properties of ZnO thin films deposited by sol-gel method. Thin Solid Films

[113] Yung KC, Liem H, Choy HS, 185002 enhanced redshift of the optical band gap in Sndoped ZnO free standing films using the sol–gel method. Journal of Physics D: Applied

[114] Ameen S, Akhtar MS, Seo HK, Kim YS, Shin HS. Influence of Sn doping on ZnO nanostructures from nanoparticles to spindle shape and their photoelectrochemical properties for dye sensitized solar cells. Chemical Engineering Journal 2012;187: 351–

[115] Exarhos GJ, Sharma SK. Influence of processing variables on the structure and prop‐

zinc oxide. Journal of the American Ceramic Society 1964;47:9–12.

Photoelectron Spectroscopy, Perkin Elmer, Eden Prairie, 1979.

indium-oxide films, Journal of Applied Physics 1977;48:3524–3531.

Electrochimica Acta 2011;56:1111–1116.

cells Nano. Letters 2007;7:2183–2187.

Nanotechnology 2010;10:6666–6671.

Superlattice Microstructures 2009;45:529–534.

process. Applied Surface Science 2008;255:2353–2359.

erties of ZnO films. Thin Solid Films 1995;270:27–32.

2006;290:131–136.

256 Solar Cells - Research and Application Perspectives

2007;101:1–6.

356

2008;517:1032–1036.

Physics 2009;42:15002-15006.


Effect of the electrolyte components variation on the triiodide ion diffusion properties and charge-transfer resistance at platinum electrode. Solid State Ionics 2006;177:3141-3146.

[143] Tu CW, Liu KY, Chien AT, Yen MH, Weng TH, Ho KC, Lin KF. Enhancement of photocurrent of polymer-gelled dye-sensitized solar cell by incorporation of exfoliated montmorillonite nanoplatelets. Journal of Polymer Science A: Polymer Chemistry 2008;46:47- 53.

[130] Ameen S, Akhtar MS, Ansari SG, Yang OB, Shin HS. Electrophoretically deposited polyaniline/ZnO nanoparticles for p–n heterostructure diodes. Superlattice and

[131] Croce F, Appetechi GB, Persi L, Scrosati B. Nanocomposite polymer electrolytes for

[132] Kang MS, Ahnand KS, Lee JW. Quasi-solid-state dye-sensitized solar cells employing ternary component polymer-gel electrolytes. Journal of Power Sources

[133] Katsaros G, Stergiopoulos T, Arabatzis IM, Papadokostaki KG, Falaras P. A solventfree composite polymer/inorganic oxide electrolyte for high efficiency solid-state dyesensitized solar cells. Journal of Photochemistry and Photobiology A 2002;149:191-198.

[134] Stergiopoulos T, Arabatzis IM, Katsaros G, Falaras P. Binary Polyethylene Oxide/ Titania Solid-State Redox Electrolyte for Highly Efficient Nanocrystalline TiO2

[135] Akhtar MS, Park JG, Kim UY, Yang OB. Composite electrolytes of polyethylene glycol methyl ether and TiO2 for dye-sensitized solar cells—Effect of heat treatment. Materials

[136] Tadayyoni MA, Gao P, Weaver MJ. Application of surface-enhanced Raman spectro‐ scopy to mechanistic electrochemistry: Oxidation of iodide at gold electrodes. Journal

[137] Kubo W, Murakoshi K, Kitamura T, Yoshida S, Haruki M, Hanabusa K, Shirai H, Wada Y, Yanagida S. Quasi-Solid-State Dye-Sensitized TiO2 Solar Cells: Effective Charge Transport in Mesoporous Space Filled with Gel Electrolytes Containing Iodide and

[138] Akhtar MS, Chun J M, Yang OB. Advanced composite gel electrolytes prepared with titania nanotube fillers in polyethylene glycol for the solid-state dye-sensitized solar

[139] Zhang J, Han H, Wu S, Xu S, Yang Y, Zhou C, Zhao X. Conductive carbon nanoparticles hybrid PEO/P(VDF-HFP)/SiO2 nanocomposite polymer electrolyte type dye sensitized

[140] Han H, Bach U, Cheng YB. Caruso R A. Increased nanopore filling: Effect on monolithic all-solid-state dye-sensitized solar cells. Applied Physics Letters

[141] Zhang X, Yang H, Xiong HM, Li FY, Xia YY. A quasi-solid-state dye-sensitized solar cell based on the stable polymer-grafted nanoparticle composite electrolyte. Journal of

[142] Nazmutdinova G, Sensfuss S, Schrödner M, Hinsch A, Sastrawan R, Gerhard., Himmler S, Wasserscheid P. Quasi-solid state polymer electrolytes for dye-sensitized solar cells:

Microstructures 2009;46:872–880.

2008;180:896-901.

258 Solar Cells - Research and Application Perspectives

lithium batteries. Nature (London) 1998;394: 456-458.

Photoelectrochemical Cells. Nano Letters 2002;2:1259-1261.

Iodine. Journal of Physical Chemistry B 2001;105:12809- 12815.

cell. Electrochemistry Communications 2007;9:2833-2837.

solar cells. Solid State Ionics 2007;178:1595-1601.

2007;90:213510-213412.

Power Sources 2006;160:1451-1455.

Chemistry and Physics 2011;127:479-483.

of Electroanalytical Chemistry 1986;198:125-136.

[144] Ito BI, de Freitas JN, De Paoli MA, Nogueira AF. Application of a Composite Polymer Electrolyte Based on Montmorillonite in Dye Sensitized Solar Cells. Journal of Brazilian Chemical Society 2008:19:688-696.

## **Investigation of Organic Bulk Heterojunction Solar Cells from Optical Aspect**

Chunfu Zhang, Yue Hao, Dazheng Chen, Zhizhe Wang and Zhenhua Lin

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/52819

## **1. Introduction**

Low in cost, light in weight and flexible in mechanics, the solution-processed organic solar cells have aroused worldwide interest and have been the promising alternative to the tradi‐ tional silicon-based solar cells [1-4]. However, they are still not available for the commercial‐ ization due to their low power conversion efficiency (PCE). Therefore, many research works have focused on the employing of new materials and device structures to improve the de‐ vice performance. The milestone is the introduction and application of the bulk heterojunc‐ tion structure consisting of an interpenetrating network of electron donor and acceptor materials [5]. By using this structure, the conventional organic solar cell (OSC) with poly(3 hexylthiophene)/[6,6]-phenyl C61-butyric acid methyl ester (P3HT:PCBM) blend shows a su‐ perior performance. Recently, the inverted organic solar cell (IOSC, in which the polarities of the two electrodes are exchanged) has also been introduced as the possible candidate for OSC to remedy the low air stability of OSC [6]. Both OSC and IOSC are now attracting the research interest. However, most of the previous works are mainly done for OSC or IOSC separately, and almost no researches are reported about the systemic comparison between OSC and IOSC for their different performances besides the air stability. Since the reported PCE of IOSC is relatively lower than that of OSC in many researches, one may doubt that which structure is better, the conventional one or the inverted one? As a result, one section of this chapter aims to investigate the performance differences of OSC and IOSC.

Although PCE of the standalone organic solar cell (including OSC and IOSC) is improved continuously with the research development, some bottlenecks still seem to appear because of the drawbacks coming from the molecular and macromolecular materials: First, the or‐ ganic solar cell is dominated by the excitonic effect, the relatively short lifetime and the low

© 2013 Zhang et al.; licensee InTech. This is an open access article 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. © 2013 Zhang et al.; licensee InTech. This is a paper 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.

charge mobility, and these factors limit the maximum thickness of the active layer for light absorption. Second, most organic semiconducting materials show discrete absorption be‐ haviour and cover only a fraction of the solar spectrum, leading to inefficient light harvest. To overcome these drawbacks, the realization of the tandem structure based on complemen‐ tary thin absorber materials provides a reasonable solution to the above obstacles. As a promising concept to achieve high PCE, the tandem solar cell can reduce the loss via sub‐ bandgap transmission of photons, the major loss mechanism in solar cells [7]. For an ideal tandem solar cell, it requires current matching of the subcells, a lossless recombination con‐ tact and a complementary absorption of the subcells. Among them, current matching of the subcells is a leading design criterion for improving the tandem device performance. Then, the organic tandem solar cell optimization by considering the current matching is also in‐ volved in this chapter.

As is well known, the working principle of organic photovoltaic devices can be simply de‐ scribed as a process of "light in- current out". This process consists of seven parts:


The first two parts are the optical mechanisms of the device and the other parts consti‐ tute the electrical aspect. The optical aspect plays a significant role because more inci‐ dent photons and absorbed photons are the baseline for the better device performance. It has been reported that the internal quantum efficiency (IQE) of organic bulk heterojunc‐ tion solar cells can reach 100% [8]. Thus the external quantum efficiency (EQE) can be approximately described as the product of IQE and the ratio of the number of absorbed photons in active layer to the number of incoming photons. As a result, the optimization of organic solar cells from the optical aspect is seriously important. This is why we inves‐ tigate the device performance of standalone and tandem organic solar cells mainly from the optical aspect in this work.

The contents of this chapter are arranged as the following: Section 2 investigates the per‐ formance of the standalone conventional and inverted organic solar cells, especially the per‐ formance differences between the two types of devices. Section 3 discusses the optimization of the organic tandem solar cell from the optical aspect by considering the current matching. At last, a short conclusion is given in Section 4.

## **2. Invetigation of Standalone Organic Solar Cells**

## **2.1. Methology**

charge mobility, and these factors limit the maximum thickness of the active layer for light absorption. Second, most organic semiconducting materials show discrete absorption be‐ haviour and cover only a fraction of the solar spectrum, leading to inefficient light harvest. To overcome these drawbacks, the realization of the tandem structure based on complemen‐ tary thin absorber materials provides a reasonable solution to the above obstacles. As a promising concept to achieve high PCE, the tandem solar cell can reduce the loss via sub‐ bandgap transmission of photons, the major loss mechanism in solar cells [7]. For an ideal tandem solar cell, it requires current matching of the subcells, a lossless recombination con‐ tact and a complementary absorption of the subcells. Among them, current matching of the subcells is a leading design criterion for improving the tandem device performance. Then, the organic tandem solar cell optimization by considering the current matching is also in‐

As is well known, the working principle of organic photovoltaic devices can be simply de‐

The first two parts are the optical mechanisms of the device and the other parts consti‐ tute the electrical aspect. The optical aspect plays a significant role because more inci‐ dent photons and absorbed photons are the baseline for the better device performance. It has been reported that the internal quantum efficiency (IQE) of organic bulk heterojunc‐ tion solar cells can reach 100% [8]. Thus the external quantum efficiency (EQE) can be approximately described as the product of IQE and the ratio of the number of absorbed photons in active layer to the number of incoming photons. As a result, the optimization of organic solar cells from the optical aspect is seriously important. This is why we inves‐ tigate the device performance of standalone and tandem organic solar cells mainly from

The contents of this chapter are arranged as the following: Section 2 investigates the per‐ formance of the standalone conventional and inverted organic solar cells, especially the per‐ formance differences between the two types of devices. Section 3 discusses the optimization of the organic tandem solar cell from the optical aspect by considering the current matching.

scribed as a process of "light in- current out". This process consists of seven parts:

volved in this chapter.

262 Solar Cells - Research and Application Perspectives

**1.** in-coupling of photon,

**2.** photon absorption,

**3.** exciton formation,

**4.** exciton migration,

**5.** exciton dissociation,

**6.** charge transport, and

**7.** charge collection at the electrodes.

the optical aspect in this work.

At last, a short conclusion is given in Section 4.

In order to investigate the standalone organic solar cell, we have performed optical simula‐ tions based on the Transfer Matrix Formalism with two subsets of 2×2 matrices (layer matrix and interface matrix), which was firstly introduced into organic solar cells by Pettersson et al [9] and now has been used widely. In this method, the cell is treated as a one-dimensional stack of homogeneous and isotropic layers with flat interfaces, and the number of photons absorbed in the active layer is obtained by calculating the time average of the energy dissi‐ pated per second in it. The detailed calculation processes are not presented here since the transfer matrix method is widely applied [10].

In the calculation, we assume that one absorbed photon produces one exciton in the active layer and one exciton divides into two free charges (one electron and one hole), and one electron (or hole) is collected by cathode (or anode). As a result, the number of photons ab‐ sorbed in the active layer can be used as the substitute for the maximum possible short cir‐ cuit current density and the EQE can be simplified as the ratio of the number of photons absorbed in the active layer to the number of incident photons.

**Figure 1.** Schematic illustration of the conventional organic solar cell (OSC) with a structure of ITO/PEDOT:PSS/ P3HT:PCBM/Al (a), and the inverted organic solar cell (IOSC) with a structure of ITO/TiOx/P3HT:PCBM/MoO3/Al (b). For OSC, the layers of ITO/PEDOT:PSS act as anode and Al acts as cathode, however for ISOC, the layers of TiOx/ITO and Al play the roles of cathode and anode, respectively. Meanwhile, the layers of TiOx and MoO3 are chosen for electron and hole transport layers. The P3HT:PCBM (1:1) is the active layer and the incident light enters devices from glass in both structures.

In this optical model, the layer of P3TH:PCBM is chosen as the active layer. OSC has the structure of ITO (150 nm)/PEDOT:PSS(50 nm)/P3HT:PCBM(x nm)/Al(100 nm) and ISOC has the structure of ITO(150 nm)/TiOx (10 nm)/P3HT:PCBM(x nm)/MoO3 (10 nm)/Al (100 nm), as shown in Fig.1. The optical constants of P3HT:PCBM, PEDOT:PSS, ITO, TiOx, MoO3, ZnO and Al are obtained from literatures [11-13].

It should be noted that, the reflection of the glass substrate is taken into account to revise the initial intensity of optical electric field at glass/ITO interface and can be described as

$$|\left|E\_{o\_{\mathcal{S}}}\right|^2 = \frac{1 - \boldsymbol{R}^\*}{n\_{\mathcal{S}}(1 - \boldsymbol{R}\boldsymbol{R}^\*)} \left|\left|E\_{o\_{\mathcal{S}}}\right|\right|^2 \tag{1}$$

where R\* represents the reflectance of air/glass interface, R the reflectance for the stack structure, ng the refraction coefficient of glass and |E0|2 (modulus squared of the opti‐ cal electric filed) the initial intensity of optical electric field when light arrives at air/ glass interface.

To calculate the number of absorbed photons (or excitons) in the active layer, the energy flow dissipation per second for single wavelength in active layer, Q is given as

$$\mathcal{Q} = \frac{1}{2} \mathfrak{c} \mathfrak{c}\_0 \mathfrak{a} n \left| E \right|^2 \tag{2}$$

where c is the speed of light, ε0 the permittivity of vacuum, α the absorption coefficient, n the real index of refraction, and |E|2 the total optical electric field intensity in the multilayer stack at single wavelength. Then the number of photons absorbed in the active layer can be expressed as

$$N = \int\_{\lambda = 300\text{nm}} Q(\lambda) \frac{\lambda}{\hbar c} d\lambda \tag{3}$$

where N represents the number of photons absorbed in the active layer, hc/λ the photon en‐ ergy at a specified wavelength λ, h the Planck constant and c the speed of light. The devices are illuminated with AM 1.5G solar spectra.

### **2.2 Results and discussion**

The number of photons absorbed in the active layer as a function of the active layer thick‐ ness is obtained for OSC and IOSC, as well the EQE as a function of the wavelength or the active layer thickness. The optical modulation effect is investigated by inserting a ZnO layer between P3HT: PCBM and Al for OSC, and changing the thickness of MoO3 layer for ISOC, respectively.

### *Comparison of photons absorbed in the active layer*

the structure of ITO(150 nm)/TiOx (10 nm)/P3HT:PCBM(x nm)/MoO3 (10 nm)/Al (100 nm), as shown in Fig.1. The optical constants of P3HT:PCBM, PEDOT:PSS, ITO, TiOx, MoO3, ZnO

It should be noted that, the reflection of the glass substrate is taken into account to revise the

\* 2 2 0 0 \* <sup>1</sup> | | || (1 ) *<sup>g</sup> g <sup>R</sup> E E n RR*

where R\* represents the reflectance of air/glass interface, R the reflectance for the stack structure, ng the refraction coefficient of glass and |E0|2 (modulus squared of the opti‐ cal electric filed) the initial intensity of optical electric field when light arrives at air/

To calculate the number of absorbed photons (or excitons) in the active layer, the energy

where c is the speed of light, ε0 the permittivity of vacuum, α the absorption coefficient, n the real index of refraction, and |E|2 the total optical electric field intensity in the multilayer stack at single wavelength. Then the number of photons absorbed in the active layer can be

> *Q*(*λ*) *λ hc*

where N represents the number of photons absorbed in the active layer, hc/λ the photon en‐ ergy at a specified wavelength λ, h the Planck constant and c the speed of light. The devices

The number of photons absorbed in the active layer as a function of the active layer thick‐ ness is obtained for OSC and IOSC, as well the EQE as a function of the wavelength or the active layer thickness. The optical modulation effect is investigated by inserting a ZnO layer between P3HT: PCBM and Al for OSC, and changing the thickness of MoO3 layer for ISOC,

2

flow dissipation per second for single wavelength in active layer, Q is given as

*N* = *∫ λ*=300*nm*

are illuminated with AM 1.5G solar spectra.

**2.2 Results and discussion**

800*nm*

0 <sup>1</sup> | | <sup>2</sup> *Q c nE* <sup>=</sup> e a


(2)

*dλ* (3)

initial intensity of optical electric field at glass/ITO interface and can be described as

and Al are obtained from literatures [11-13].

264 Solar Cells - Research and Application Perspectives

glass interface.

expressed as

respectively.

The results of the number of photons absorbed in the active layer as a function of the active layer thickness for OSC are shown in Fig. 2(a). It is clear that the number of absorbed pho‐ tons increases with the active layer thickness and one can see a notable behavior of oscilla‐ tion which is due to the optical interference effect induced by the incident light and the reflected light from the mirror metal electrode. The inserting of a ZnO layer leads to the shift of interference maxima to lower thicknesses and the remarkable increase of absorbed pho‐ tons especially for the relatively thin active layer. However, the insertion of a ZnO layer makes no improvement near the active layer thicknesses where the interference maxima are obtained. This variation of the number of absorbed photons with the active layer thickness is the same as that of maximum possible short circuit current density in other researches [14].

For IOSC, Fig. 2(b) shows the same tendency as OSC. It is obvious that the influence of opti‐ cal modulation effect is more slightly when the active layer is relatively thick (about 150 nm). Comparing OSC with IOSC, as shown in Fig. 2 (c), it is clear that the number of absor‐ bed photons in IOSC is larger at any active layer thicknesses except for the thicknesses around which the interference maxima of OSC are obtained. In other words, for most active layer thicknesses, the light absorption in IOSC is more effective, hence the larger contribu‐ tion to photocurrent. It has been reported that the refractive index of TiOx is very similar to that of PEDOT:PSS, therefore the light loss induced by the reflection at the ITO/TiOx inter‐ face and that at the ITO/PEDOT:PSS interface is nearly equivalent. However, light absorp‐ tion loss in the TiOx layer is smaller than that in the PEDOT:PSS layer [15].In addition, the layer of MoO3 in IOSC can act as an optical spacer layer. As a result, one can say the better light absorption in IOSC is attributed to the nearly equal amount of entering light and the smaller absorption loss in TiOx layer, as well as the optical modulation effect caused by MoO3 layer. Although IOSC shows a better performance than OSC at most of the active lay‐ er thicknesses, it is noted that the performance of IOSC is slightly lower than that of OSC around the interference peaks as shown in Fig. 2(c). One possible reason is the parasitic ab‐ sorption in the MoO3 layer, because not only the the optical modulation effect but also the absorption loss could be caused by the MoO3 layer, which produces a tradeoff.

To well understand the difference between OSC and IOSC, the distribution of optical elec‐ tric field is investigated at different active layer thicknesses. According to the absorption coefficient of P3HT:PCBM calculated by α=4πk/λ, the maximum absorption coefficient is obtained at the incident light wavelength of around 512 nm, which agrees to the range of maximum absorption coefficient from 500 nm to 550 nm for P3HT:PCBM active material. As a result, the incident light of 512 nm is chosen to calculate the optical electric field distribu‐ tion. The distributions of normalized modulus squared optical electric field at three differ‐ ent active layer thicknesses (45, 85, and 150 nm) for OSC and IOSC are shown in Fig. 3. The thickness of TiOx is specified as 50 nm to make the active layer region at same position in Fig. 3 for a clear comparison since its thickness has no significant influence on the photons absor‐ bed in the active layer. Fig. 3(a) and (c) illustrate that the area below the curve of IOSC is larger than that below the curve of OSC for both thin and thick devices, which is consistent with the more absorbed photons in IOSC. And from Fig. 3(b), the opposite result can be seen when the active layer is 85 nm where the interference maximum of OSC is obtained, which agrees to the situation in Fig. 2(c). The similar results will also be shown in next part where the different performances of EQE for OSC and IOSC are discussed.

**Figure 2.** Number of photons absorbed in the active layer versus the active layer thickness with various thicknesses of the optical spacer layer. (a) OSC with the ZnO layer thickness ranging from 0 to 30 nm. (b) IOSC with the MoO3 layer

thickness ranging from 10 to 40 nm. (c) Comparison of the number of photons absorbed in active layer of OSC and ISOC as a function of the active layer thickness. In this case, no ZnO layer is inserted in OSC and the thickness of MoO3 is specified as 10 nm for ISOC.

**Figure 3.** Normalized optical electric field distribution varies with the distance from glass/ITO interface for OSC (black) and IOSC (red) devices. The incident light wavelength is specified as 512 nm. The active layer thicknesses are 45 nm (a), 85 nm (b), and 150 nm (c), respectively.

### *Comparison of EQE*

active layer is 85 nm where the interference maximum of OSC is obtained, which agrees to the situation in Fig. 2(c). The similar results will also be shown in next part where the different

**Figure 2.** Number of photons absorbed in the active layer versus the active layer thickness with various thicknesses of the optical spacer layer. (a) OSC with the ZnO layer thickness ranging from 0 to 30 nm. (b) IOSC with the MoO3 layer

performances of EQE for OSC and IOSC are discussed.

266 Solar Cells - Research and Application Perspectives

The EQE as a function of incident light wavelength for OSC and IOSC is shown in Fig. 4, and the thicknesses of the active layer are specified as 40, 80, 160 and 220 nm. For OSC in Fig. 4(a), the EQE trends to increase with the thickness of the P3HT:PCBM layer in the range of wavelength from 450 to 650 nm. It should be noted that, the lower EQE at 160 nm than 80 nm can be explained by the fact that 80 nm is closed to the thickness where the first interfer‐ ence maximum of OSC is obtained (Fig. 2(c)). At the same time, the wavelength range pos‐ sessing higher EQE is also expanded especially at the wavelength near 600 nm, where the shoulders are generally changed into peaks with the increasing of the active layer thickness. Of course, for the light over 650 nm, it prefers thicker active layer to achieve better absorp‐ tion of light, and for the light near 400 nm, the variation of EQE with the active layer thick‐ ness displays a behavior of increase and decrease in turn. The similar results can be obtained from Fig.4 (b) for IOSC. For the thicker active layer beyond 220 nm, the same simulation re‐ sults can be obtained (not presented here) and the results are agreed to the measured EQE of OSC and IOSC with 250 nm thick active layer. In short, the thick active layer brings higher EQE for both OSC and IOSC, and the main absorption range from 400 to 650 nm can be ob‐ served from Fig. 4.

**Figure 4.** EQE vs. wavelength for OSC (a) and IOSC (b) with four different thicknesses of active layer: 40 nm(black), 80 nm(red), 160 nm(blue), and 220 nm(green), respectively.

The comparison of EQE varying with the wavelength for OSC and IOSC is presented in Fig. 5. It is evident that IOSC with 40 and 160 nm active layer has better performance in EQE in the main absorption range. However, for the 80 and 220 nm active layer, the EQE of OSC is closed to or even higher than that of IOSC, which is due to the fact that these two thickness‐ es nearly equal the thicknesses where the interference maxima of OSC are obtained. In other words, IOSC is super to OSC in EQE in the main absorption range for both thin and thick active layers except for thicknesses around which the interference maxima of OSC are ob‐ tained, which agrees to the analysis of the number of photons absorbed in active layer in Fig. 2. Hence the explanation is the same for this result.

**Figure 5.** Comparison of EQE as a function of wavelength for OSC (solid line) and IOSC (dash line) with four different active layer thicknesses: 40 nm, 80 nm, 160 nm, and 220 nm.

**Figure 4.** EQE vs. wavelength for OSC (a) and IOSC (b) with four different thicknesses of active layer: 40 nm(black), 80

The comparison of EQE varying with the wavelength for OSC and IOSC is presented in Fig. 5. It is evident that IOSC with 40 and 160 nm active layer has better performance in EQE in the main absorption range. However, for the 80 and 220 nm active layer, the EQE of OSC is closed to or even higher than that of IOSC, which is due to the fact that these two thickness‐ es nearly equal the thicknesses where the interference maxima of OSC are obtained. In other words, IOSC is super to OSC in EQE in the main absorption range for both thin and thick active layers except for thicknesses around which the interference maxima of OSC are ob‐ tained, which agrees to the analysis of the number of photons absorbed in active layer in

nm(red), 160 nm(blue), and 220 nm(green), respectively.

268 Solar Cells - Research and Application Perspectives

Fig. 2. Hence the explanation is the same for this result.

**Figure 6.** Comparison of EQE as a function of the active layer thickness for OSC (solid line) and IOSC (dash line) with six wavelengths of incident light: 400 nm, 450 nm, 500 nm, 550 nm, 600 nm, and 650 nm.


**Table 1.** The positions and amplitudes of the first maximum EQE of OSC and IOSC at different wavelengths. The positions correspond to the active layer thickness and OSC behaves slighter thickness oscillation behavior than IOSC.

To verify the above conclusion, the EQE as a function of the active layer thickness at differ‐ ent wavelengths of incident light is investigated. For the EQE of OSC and IOSC depicted in Fig. 6, a remarkable increase and oscillation behavior with the increase of the active layer thickness can be observed. From Fig. 6, a conclusion may be obtained that IOSC performs better than OSC at any thickness of active layer for the light ranging from 400 to 650 nm. However, at the active layer thicknesses around which the interference maxima of OSC are obtained, the EQE of OSC is close to or even higher than that of IOSC. To well investigate the difference between OSC and IOSC, the positions and amplitudes of the first maximum EQE at different wavelengths are shown in Table 1. The clear oscillation behavior of the po‐ sitions and amplitudes of the maximum EQE can be seen for both OSC and IOSC, and OSC exhibits more slight thickness oscillation behavior than IOSC. It is evident that the slight os‐ cillation of maxima at single wavelength is beneficial to the final maximum in the range in‐ cluding all wavelengths. Therefore, the results from Table 1 can also be used to explain the results shown in Fig. 2(c) and Fig. 5.

In summary, from pure optical aspect, OSC and IOSC have the same tendency in the num‐ ber of photons absorbed in the active layer, EQE, and the optical electric field distribution as well as the similar influence of optical modulation effect. However, IOSC performs better than OSC except for the case wherein the interference maxima of OSC are obtained, which is due to the better light absorption of ISOC possible absorption loss caused by MoO3 layer.

## **3. Investigation of the Organic Tandem Solar Cell**

Although PCE of the organic solar cell is improved continuously, there are still some draw‐ backs for standalone devices: First, the organic solar cell is dominated by the excitonic effect, the relatively short lifetime and the low charge mobility. All of these limit the maximum thickness of the active layer for light absorption. Second, most organic semiconducting ma‐ terials show discrete absorption behavior and cover only a fraction of the solar spectrum, leading to inefficient light harvest. The realization of the organic tandem solar cell based on complementary thin absorber materials provides a reasonable solution to improve device performance further.

In the existing views, matching the photocurrents of the subcells leads to the maximum PCE in the corresponding tandem cell, making it a crucial design criterion for optimum perform‐ ance. Then to achieve a better device performance, the tandem cell should be optimized by considering current matching.

**Light wavelength (nm)** 400 450 500 550 600 650 **OSC** Position (nm) 65 95 85 75 75 90

**IOSC** Position (nm) 85 115 85 65 75 90

**Table 1.** The positions and amplitudes of the first maximum EQE of OSC and IOSC at different wavelengths. The positions correspond to the active layer thickness and OSC behaves slighter thickness oscillation behavior than IOSC.

results shown in Fig. 2(c) and Fig. 5.

270 Solar Cells - Research and Application Perspectives

performance further.

**3. Investigation of the Organic Tandem Solar Cell**

To verify the above conclusion, the EQE as a function of the active layer thickness at differ‐ ent wavelengths of incident light is investigated. For the EQE of OSC and IOSC depicted in Fig. 6, a remarkable increase and oscillation behavior with the increase of the active layer thickness can be observed. From Fig. 6, a conclusion may be obtained that IOSC performs better than OSC at any thickness of active layer for the light ranging from 400 to 650 nm. However, at the active layer thicknesses around which the interference maxima of OSC are obtained, the EQE of OSC is close to or even higher than that of IOSC. To well investigate the difference between OSC and IOSC, the positions and amplitudes of the first maximum EQE at different wavelengths are shown in Table 1. The clear oscillation behavior of the po‐ sitions and amplitudes of the maximum EQE can be seen for both OSC and IOSC, and OSC exhibits more slight thickness oscillation behavior than IOSC. It is evident that the slight os‐ cillation of maxima at single wavelength is beneficial to the final maximum in the range in‐ cluding all wavelengths. Therefore, the results from Table 1 can also be used to explain the

In summary, from pure optical aspect, OSC and IOSC have the same tendency in the num‐ ber of photons absorbed in the active layer, EQE, and the optical electric field distribution as well as the similar influence of optical modulation effect. However, IOSC performs better than OSC except for the case wherein the interference maxima of OSC are obtained, which is due to the better light absorption of ISOC possible absorption loss caused by MoO3 layer.

Although PCE of the organic solar cell is improved continuously, there are still some draw‐ backs for standalone devices: First, the organic solar cell is dominated by the excitonic effect, the relatively short lifetime and the low charge mobility. All of these limit the maximum thickness of the active layer for light absorption. Second, most organic semiconducting ma‐ terials show discrete absorption behavior and cover only a fraction of the solar spectrum, leading to inefficient light harvest. The realization of the organic tandem solar cell based on complementary thin absorber materials provides a reasonable solution to improve device

In the existing views, matching the photocurrents of the subcells leads to the maximum PCE in the corresponding tandem cell, making it a crucial design criterion for optimum perform‐

EQE 0.7100 0.8252 0.8363 0.8060 0.7437 0.1975

EQE 0.7306 0.8150 0.7987 0.8109 0.7574 0.1952

**Figure 7.** a) Absorption coefficients of both individual active layers. b) Normal and c) Reversed device structure.

As discussed in previous section, P3HT has been widely used as the donor conjugated poly‐ mer in the quest for high-efficiency bulk heterojunction organic solar cells. Its combination with PCBM as the acceptor is a standard active layer of organic solar cells based on poly‐ mer, whereas the low-energy onset of the absorption of this combination at about 650 nm limits the number of photons absorbed in the active layer. Recently, a very promising organic material has been reported and applied [16], namely poly[3,6-bis-(40-dodecyl-[2,20] bithiophenyl-5 yl)-2,5-bis-(2-ethyl-hexyl)-2,5-dihydropyrrolo[3,4-]pyrrole-1,4-dione] (pBBTDPP2). This new polymer combines electron-rich quaterthiophene (BBT) segments with electron-poor diketo pyrrolo-pyrrole (DPP) units to lower the optical bandgap to 1.4 eV in thin films. The onset of the absorption of the blend of pBBTDPP2 and PCBM is significantly shifted to 860 nm with odichlorobenzene as solvent. Thus, a series connected tandem solar cell based on P3HT:PCBM and pBBTDPP2:PCBM almost covers the whole UV and visible parts of the solar spectrum, making it attractive (see Fig. 7 (a)). It is necessary to carefully optimize the thicknesses of the front and back cells for different layer sequences to deeply exploit the opportunity provided by this tandem solar cell. Therefore, detailed optical simulations of tandem cells based on P3HT:PCBM and pBBTDPP2:PCBM have been carried out in this chapter.

## **3.1. Methodology**

As in previous section, the calculation is still based on the Transfer Matrix Formalism. The basic structure of the tandem solar cell is shown in Fig. 7(b). On top of the glass, a 100 nm indium-tin-oxide (ITO) layer is used as the anode, followed by a 50 nm layer of PEDOT:PSS and a front active layer with variable thicknesses. The recombination contact consists of a 30 nm ZnO layer and a 15 nm PEDOT:PSS layer, followed by a back active layer with variable thicknesses. Finally, 100 nm Al is deposited to realize the cathode. In conventional tandem solar cells, materials mainly absorbing light of shorter wavelengths act as the front active layer to provide a window for the back cell while materials mostly absorbing light of longer wavelengths work as the back active layer. Thus the device with P3HT:PCBM in the front cell and pBBTDPP2:PCBM in the back cell is defined as"Normal Tandem Solar Cell" or "NTSC", the device with pBBTDPP2:PCBM in the front cell and P3HT:PCBM in the back cell as "Reverse Tandem Solar Cell" or "RTSC".

Before starting our work, three assumptions should be stated. First, an electron-hole pair is generated in the solar cell with every photon absorbed. Second, Ohmic losses in the recom‐ bination contact and the spacer are negligible. In consequence, the open circuit voltage of the tandem solar cell is a summation of those of both subcells. Thus, the performance of the tandem solar cell is mainly determined by its short circuit current density (JSC). Three, the glass substrate is thicker than the coherence length of light, so optical interference in it can be neglected.

The optical parameters (n and k) of P3HT:PCBM (1:1 in weight), pBBTDPP2:PCBM (1:2 in weight), ITO, ZnO, PEDOT:PSS and Al used in this work are obtained from literatures [11, 14, 17, 18].

Considering current matching, we optimize the thicknesses of the front and back cells for NTSC and RTSC respectively. All the calculations are carried out under AM 1.5G radiation.

### **3.2. Results and discussion**

According to the existing views, the optimized PCE can be obtained when the photocur‐ rents of the subcells are matched. Thus, we vary the thicknesses of both the front (dfront between 10 and 250 nm) and back (dback between 10 and 200 nm) active layers to investi‐ gate JSC of both subcells.

Investigation of Organic Bulk Heterojunction Solar Cells from Optical Aspect http://dx.doi.org/10.5772/52819 273

the absorption of the blend of pBBTDPP2 and PCBM is significantly shifted to 860 nm with odichlorobenzene as solvent. Thus, a series connected tandem solar cell based on P3HT:PCBM and pBBTDPP2:PCBM almost covers the whole UV and visible parts of the solar spectrum, making it attractive (see Fig. 7 (a)). It is necessary to carefully optimize the thicknesses of the front and back cells for different layer sequences to deeply exploit the opportunity provided by this tandem solar cell. Therefore, detailed optical simulations of tandem cells based on

As in previous section, the calculation is still based on the Transfer Matrix Formalism. The basic structure of the tandem solar cell is shown in Fig. 7(b). On top of the glass, a 100 nm indium-tin-oxide (ITO) layer is used as the anode, followed by a 50 nm layer of PEDOT:PSS and a front active layer with variable thicknesses. The recombination contact consists of a 30 nm ZnO layer and a 15 nm PEDOT:PSS layer, followed by a back active layer with variable thicknesses. Finally, 100 nm Al is deposited to realize the cathode. In conventional tandem solar cells, materials mainly absorbing light of shorter wavelengths act as the front active layer to provide a window for the back cell while materials mostly absorbing light of longer wavelengths work as the back active layer. Thus the device with P3HT:PCBM in the front cell and pBBTDPP2:PCBM in the back cell is defined as"Normal Tandem Solar Cell" or "NTSC", the device with pBBTDPP2:PCBM in the front cell and P3HT:PCBM in the back cell

Before starting our work, three assumptions should be stated. First, an electron-hole pair is generated in the solar cell with every photon absorbed. Second, Ohmic losses in the recom‐ bination contact and the spacer are negligible. In consequence, the open circuit voltage of the tandem solar cell is a summation of those of both subcells. Thus, the performance of the tandem solar cell is mainly determined by its short circuit current density (JSC). Three, the glass substrate is thicker than the coherence length of light, so optical interference in it can

The optical parameters (n and k) of P3HT:PCBM (1:1 in weight), pBBTDPP2:PCBM (1:2 in weight), ITO, ZnO, PEDOT:PSS and Al used in this work are obtained from literatures [11,

Considering current matching, we optimize the thicknesses of the front and back cells for NTSC and RTSC respectively. All the calculations are carried out under AM 1.5G radiation.

According to the existing views, the optimized PCE can be obtained when the photocur‐ rents of the subcells are matched. Thus, we vary the thicknesses of both the front (dfront between 10 and 250 nm) and back (dback between 10 and 200 nm) active layers to investi‐

P3HT:PCBM and pBBTDPP2:PCBM have been carried out in this chapter.

**3.1. Methodology**

272 Solar Cells - Research and Application Perspectives

be neglected.

14, 17, 18].

**3.2. Results and discussion**

gate JSC of both subcells.

as "Reverse Tandem Solar Cell" or "RTSC".

**Figure 8.** a) and b) 3D plots of JSC(front) (magenta) and JSC(back) (green) versus dfront and dback for NTSC and RTSC.

Such a result of the calculation is plotted in Figs. 8 (a) and (b) for NTSC and RTSC separate‐ ly. JSC for the front (JSC(front)) and back (JSC(back)) cells are shown in the three-dimensional (3D) space as surfaces, magenta and green, respectively. Figures 9 (a) and (b) display the same results as Figs. 8 (a) and (b) respectively, but in a two-dimensional format. In NTSC, the front cell may provide up to a JSC of 11.22 mA/cm2 when dback=10 nm. Whereas when dback increases up to 200 nm, JSC(front) can decrease down to 10.30 mA/cm2 . Because the front cell is much far from the mirror Al electrode, JSC(front) scarcely shows any interference oscillation in the variation range of dback. Thus, this 8% loss is JSC(front) not induced by the interference effect but the reduction of the amount of light reflected from the Al surface (and arriving at the front cell a second time) as the thickness of the back cell increases. The variation of JSC(back) is more strongly affected by dfront than vice versa. While for dfront=10 nm, JSC(back) goes up to 15.76 mA/cm2 . However, when dfront=250 nm, JSC(back) reduces to 10.29 mA/cm2 . This is obvi‐ ously caused by the absorption spectra overlap of the two active layer materials (see Fig. 7(a)), which reduces photons arriving at the back cell. Because the back cell is much nearer to the Al electrode than the front cell, the interference behavior of JSC(back) is very obvious as shown in Fig. 9(a). Because of the same reason, the same tendency can be observed in RTSC (see Figs. 8(b) and 9(b)). In RTSC, while for dback=10 and 200 nm, JSC(front) goes up to 18.93 and 16.04 mA/cm2 , respectively. With the increase of dfront from 10 to 250 nm, JSC(back) decreases from 10.97 to 5.53 mA/cm2 . A obvious increase of JSC of pBBTDPP2:PCBM layer and a strong decrease of JSC of P3HT:PCBM layer can be observed in RTSC, compared with NTSC. This is caused by the absorption difference of two blends. As shown in Fig. 7(a), the spectral range of pBBTDPP2:PCBM is much wider than that of P3HT:PCBM (the former absorbs photons in the nearly entire wavelength range discussed while the latter hardly absorbs photons of wavelengths beyond 650 nm). Therefore, when acting as the front active layer, pBBTDPP2:PCBM hinders the harvest of photons in P3HT:PCBM more strongly than that in P3HT:PCBM with pBBTDPP2:PCBM as the back active layer.


**Table 2.** Current matching points for NTSC and RTSC, corresponding to the black dots of the bold lines in Figs. 9 (a) and (b) respectively. Corresponding JSC(front), JSC(back), dback and dfront are all displayed here. The differences between matching JSC of NTSC and RTSC displayed in the same row are shown in the last column (Δ). d is in nm and JSC is in mA/cm2 here.

The bold lines in Figs. 9 (a) and (b) represent the intersection between both surfaces in the corresponding 3D plots, namely, current matching points of the subcells, which correspond to the optimized thicknesses of the active layers. Seen from Figs. 9(a) and (b), it is obvious that a single value for dfront can have more than one counterpart dback along the bold line for RTSC while there is only one counterpart dback for NTSC. This interesting thing leads us to list the optimized active layer thicknesses in Table 2 for NTSC and RTSC, respectively. It can be observed from Table 2 that RTSC shows its superiority in matching JSC when the active layers of both subcells are relatively thin. We note that RTSC can provides a larger matching JSC with a smaller dfront when the device is relatively thin (dback is usually less than 100 nm). But NTSC is better as the thicknesses of both active layers increase, in agreement with the general view presented.

shown in Fig. 9(a). Because of the same reason, the same tendency can be observed in RTSC (see Figs. 8(b) and 9(b)). In RTSC, while for dback=10 and 200 nm, JSC(front) goes up to 18.93 and

decrease of JSC of P3HT:PCBM layer can be observed in RTSC, compared with NTSC. This is caused by the absorption difference of two blends. As shown in Fig. 7(a), the spectral range of pBBTDPP2:PCBM is much wider than that of P3HT:PCBM (the former absorbs photons in the nearly entire wavelength range discussed while the latter hardly absorbs photons of wavelengths beyond 650 nm). Therefore, when acting as the front active layer, pBBTDPP2:PCBM hinders the harvest of photons in P3HT:PCBM more strongly than that in

dfront dback Jsc(front) Jsc(back) dfront dback Jsc(front) Jsc(back) **Δ** 10 1.00 0.89 7 10 1.06 1.09 -0.20 20 1.55 1.55 11 20 2.17 2.09 -0.53 30 2.64 2.57 15 30 3.18 3.29 -0.71 40 3.28 3.26 21 40 4.26 4.35 -1.09 50 4.20 4.14 32 50 5.25 5.17 -1.04 60 5.21 5.22 66 60 6.05 6.06 -0.84 70 6.32 6.35 102 70 6.75 6.73 -0.37 80 7.49 7.46 109 80 7.12 7.12 0.34 90 8.76 8.77 110 90 7.42 7.38 1.39 100 9.28 9.29 107 100 7.54 7.53 1.76 110 9.55 9.57 102 110 7.59 7.59 1.98 120 9.72 9.70 96 120 7.62 7.58 2.12 130 9.81 9.80 89 130 7.58 7.57 2.24 140 9.93 9.92 82 140 7.54 7.60 2.32 150 10.09 10.09 77 150 7.73 7.70 2.39 160 10.24 10.23 72 160 7.86 7.92 2.30 170 10.29 10.29 70 170 8.26 8.21 2.08 180 10.28 10.28 68 180 8.51 8.57 1.71 190 10.25 10.25 69 190 8.90 8.90 1.34 200 10.21 10.20 72 200 9.18 9.16 1.04

**Table 2.** Current matching points for NTSC and RTSC, corresponding to the black dots of the bold lines in Figs. 9 (a) and (b) respectively. Corresponding JSC(front), JSC(back), dback and dfront are all displayed here. The differences between matching JSC of NTSC and RTSC displayed in the same row are shown in the last column (Δ). d is in nm and JSC is in

The bold lines in Figs. 9 (a) and (b) represent the intersection between both surfaces in the corresponding 3D plots, namely, current matching points of the subcells, which correspond

P3HT:PCBM with pBBTDPP2:PCBM as the back active layer.

, respectively. With the increase of dfront from 10 to 250 nm, JSC(back) decreases

**NTSC RTSC difference**

. A obvious increase of JSC of pBBTDPP2:PCBM layer and a strong

16.04 mA/cm2

mA/cm2 here.

from 10.97 to 5.53 mA/cm2

274 Solar Cells - Research and Application Perspectives

**Figure 9.** a) and b) Different viewing angle of the 3D plots shown in Figs. 8 (a) and (b), respectively. Every magenta or green line indicates JSC(front) or JSC(back) versus dfront with the fixed dback. dback is ranging from 10 to 200 nm in a uniform step of 10 nm (see the arrow).

In order to well understand this amazing phenomenon, optical electric field distributions of tandem solar cells should be taken into account. Four current matching points listed in Table 2 are used here: a 58-40-nm NTSC, a 21-40-nm RTSC, a 238-150-nm NTSC and a 77-150-nm RTSC. The reason why we choose these points is that the matching JSC differences between the corresponding NTSC and RTSC reach the maxima in positive and negative, respectively (see Δ, the last column in Table 2). The distributions of normalized modulus squared of opti‐ cal electric filed |E|2 for the above tandem solar cells are calculated and shown in Fig. 10 (the shadow area indicates the active layers of the front (left) and back (right) cells). As shown in Fig. 7(a), 512 and 600 nm are around the peak and shoulder of the absorption spec‐ trum of P3HT:PCBM respectively while 727 and 809 nm are around two absorption maxima of pBBTDPP2:PCBM. Thus, the cases for wavelengths of 512, 600, 727 and 809 nm are dis‐ cussed here. By observing optical electric filed distributions of a 58-40 nm NTSC and a 21-40-nm RTSC (both active layers are relatively thin) as shown in Figs. 10 (a) and (b), it is very clear that RTSC has a better optical electric filed distribution in both active layers al‐ though it has a smaller dfront, in accordance with the values of JSC shown in Table 2. It can be explained by the properties of the materials and device structures. As shown in Figs. 10 (a) and (b), no matter with which structure, the peaks of optical electric field for light of wave‐ lengths of 727 and 809 nm are usually near or in the front active layer. Since pBBTDPP2:PCBM mainly absorbs light of longer wavelengths and P3HT:PCBM absorbs light of shorter wavelengths, RTSC (place pBBTDPP2:PCBM in the front and P3HT:PCBM in the back) has a better performance when the active layers are thin. However, things become different when the device becomes thicker and thicker. For a very thick RTSC, the first inter‐ ference peaks for wavelengths of 727 and 809 nm begin to leave the front active layer and then the pBBTDPP2:PCBM subcell is no longer an effective device. At the same time, owing to the much wider spectral range of pBBTDPP2:PCBM, it hinders the harvest of photons in P3HT:PCBM with pBBTDPP2:PCBM as the front active layer. Thus, when the active layers become thicker (dback is usually over 100 nm), RTSC only allows a smaller matching JSC with a smaller dfront compared with NTSC and NTSC shows its superiority. This is in good agree‐ ment with the results in Table 2.

In conclusion, it is observed that RTSC takes over the lead in matching JSC when the active layers are relatively thin; but NTSC allows a larger matching JSC as the active layers are rela‐ tively thick. The similar results are also found in the tandem solar cell based on P3HT:PC70BM and PCPDTBT:PC60BM [19]. The results are very interesting since we can choose the thinner RTSC to achieve matching JSC, which can alleviate the carrier transport problem and save the material cost.

Investigation of Organic Bulk Heterojunction Solar Cells from Optical Aspect http://dx.doi.org/10.5772/52819 277

**Figure 10.** Calculated distributions of normalized modulus squared of optical electric field |E|2 inside a a) 58-40-nm NTSC b) 21-40-nm RTSC c) 238-150-nm NTSC d) 77-150-nm RTSC for four wavelengths of 512, 600, 727 and 809 nm.

## **4. Conclusion**

In order to well understand this amazing phenomenon, optical electric field distributions of tandem solar cells should be taken into account. Four current matching points listed in Table 2 are used here: a 58-40-nm NTSC, a 21-40-nm RTSC, a 238-150-nm NTSC and a 77-150-nm RTSC. The reason why we choose these points is that the matching JSC differences between the corresponding NTSC and RTSC reach the maxima in positive and negative, respectively (see Δ, the last column in Table 2). The distributions of normalized modulus squared of opti‐

(the shadow area indicates the active layers of the front (left) and back (right) cells). As shown in Fig. 7(a), 512 and 600 nm are around the peak and shoulder of the absorption spec‐ trum of P3HT:PCBM respectively while 727 and 809 nm are around two absorption maxima of pBBTDPP2:PCBM. Thus, the cases for wavelengths of 512, 600, 727 and 809 nm are dis‐ cussed here. By observing optical electric filed distributions of a 58-40 nm NTSC and a 21-40-nm RTSC (both active layers are relatively thin) as shown in Figs. 10 (a) and (b), it is very clear that RTSC has a better optical electric filed distribution in both active layers al‐ though it has a smaller dfront, in accordance with the values of JSC shown in Table 2. It can be explained by the properties of the materials and device structures. As shown in Figs. 10 (a) and (b), no matter with which structure, the peaks of optical electric field for light of wave‐ lengths of 727 and 809 nm are usually near or in the front active layer. Since pBBTDPP2:PCBM mainly absorbs light of longer wavelengths and P3HT:PCBM absorbs light of shorter wavelengths, RTSC (place pBBTDPP2:PCBM in the front and P3HT:PCBM in the back) has a better performance when the active layers are thin. However, things become different when the device becomes thicker and thicker. For a very thick RTSC, the first inter‐ ference peaks for wavelengths of 727 and 809 nm begin to leave the front active layer and then the pBBTDPP2:PCBM subcell is no longer an effective device. At the same time, owing to the much wider spectral range of pBBTDPP2:PCBM, it hinders the harvest of photons in P3HT:PCBM with pBBTDPP2:PCBM as the front active layer. Thus, when the active layers become thicker (dback is usually over 100 nm), RTSC only allows a smaller matching JSC with a smaller dfront compared with NTSC and NTSC shows its superiority. This is in good agree‐

In conclusion, it is observed that RTSC takes over the lead in matching JSC when the active layers are relatively thin; but NTSC allows a larger matching JSC as the active layers are rela‐ tively thick. The similar results are also found in the tandem solar cell based on P3HT:PC70BM and PCPDTBT:PC60BM [19]. The results are very interesting since we can choose the thinner RTSC to achieve matching JSC, which can alleviate the carrier transport

for the above tandem solar cells are calculated and shown in Fig. 10

cal electric filed |E|2

276 Solar Cells - Research and Application Perspectives

ment with the results in Table 2.

problem and save the material cost.

The application of the bulk heterojunction structure consisting of an interpenetrating net‐ work of electron donor and acceptor materials greatly improves the standalone organic solar cells. Now there are mainly two types of standalone organic solar cells: OSC and IOSC. IOSC is introduced as the possible candidate for OSC to remedy the low air stability of OSC. However, most of the previous works reported that PCE of IOSC is relatively lower than that of OSC. As a result, the performance differences of OSC and IOSC are discussed in this work. It is concluded that in the optical aspect, OSC and IOSC have the same tendency in the number of photons absorbed in the active layer, EQE, and the optical electric field distri‐ bution as well as the similar influence of optical modulation effect. Normally, IOSC per‐ forms better due to the better light absorption of IOSC because the absorption loss in TiOx layer for IOSC is smaller than that in PEDOT:PSS layer for OSC and the hole transporting layer (MoO3 layer in this work) can play the role of optical layer. However, around the inter‐ ference maxima of OSC, OSC shows better performance because the optical electric field has been optimized under this condition and there is no room for further improvement.

Although PCE of the standalone organic solar cell is improved continuously, there are still some drawbacks for this type of devices: First, the organic solar cell is dominated by the ex‐ citonic effect, the relatively short lifetime and the low carrier mobility. And these limit the maximum thickness of the active layer for light absorption. Second, most organic semicon‐ ducting materials show discrete absorption behavior and cover only a fraction of the solar spectrum, leading to inefficient light harvest. The realization of organic tandem solar cells based on complementary thin absorber materials provides a reasonable solution to improve PCE further. In a tandem cell, to achieve the maximum PCE, it is necessary to ensure current matching of both subcells, which leads to detailed optical simulations in this work. At first, active layer thicknesses of the tandem cell are optimized by considering current matching for normal and reverse structures (see Fig. 7), respectively. Owing to the different spectral ranges of two blend materials (P3HT:PCBM and pBBTDPP2:PCBM) and device structures, it is noted that the reverse tandem cell allows a larger matching JSC when the total device is relatively thin. When the thicknesses of the active layers increase, the normal tandem solar cell begins to present its superiority in the performance. This makes senses in the aspect of application that we can choose a thinner reverse tandem cell to achieve JSC needed in some cases, which saves cost and increases the profit to an extent.

## **Acknowledgements**

The work is supported by National Natural Science Foundation of China (61106063) and the Fundamental Research Funds for the Central Universities (K50511250003).

## **Author details**

Chunfu Zhang1\*, Yue Hao1\*, Dazheng Chen1 , Zhizhe Wang1\* and Zhenhua Lin2


## **References**


[4] Zhang, C. F., Hao, Y., Tong, S. W., Lin, Z. H., Feng, Q., Kang, E. T., & Zhu, C. X. (2011). Effects of cathode confinement on the performance of polymer/fullerene pho‐ tovoltaic cells in the thermal treatment. *IEEE Trans. Electron Devices*, 58(3), 835-842.

spectrum, leading to inefficient light harvest. The realization of organic tandem solar cells based on complementary thin absorber materials provides a reasonable solution to improve PCE further. In a tandem cell, to achieve the maximum PCE, it is necessary to ensure current matching of both subcells, which leads to detailed optical simulations in this work. At first, active layer thicknesses of the tandem cell are optimized by considering current matching for normal and reverse structures (see Fig. 7), respectively. Owing to the different spectral ranges of two blend materials (P3HT:PCBM and pBBTDPP2:PCBM) and device structures, it is noted that the reverse tandem cell allows a larger matching JSC when the total device is relatively thin. When the thicknesses of the active layers increase, the normal tandem solar cell begins to present its superiority in the performance. This makes senses in the aspect of application that we can choose a thinner reverse tandem cell to achieve JSC needed in some

The work is supported by National Natural Science Foundation of China (61106063) and the

[1] Brabec, C. J., Sariciftci, N. S., & Hummelen, J. C. (2001). Plastic solar cells. *Adv. Funct.*

[2] Huynh, W. U., Dittmer, J. J., & Alivisatos, A. P. (2002). Hybrid nanorod-polymer so‐

[3] Kim, Y., Lee, K., Coates, N. E., Moses, D., Nguyen, T. Q., Dante, M., & Heeger, A. J. (2007). Efficient tandem polymer solar cells fabricated by all-solution processing. *Sci‐*

, Zhizhe Wang1\* and Zhenhua Lin2

Fundamental Research Funds for the Central Universities (K50511250003).

cases, which saves cost and increases the profit to an extent.

**Acknowledgements**

278 Solar Cells - Research and Application Perspectives

**Author details**

**References**

*Mater*, 11(1), 15-26.

*ence*, 317(5835), 222-225.

Chunfu Zhang1\*, Yue Hao1\*, Dazheng Chen1

\*Address all correspondence to: cfzhang@xidian.edu.cn

1 School Of Microelectronics, Xidian University, China

2 ECE, National University of Singapore, Singapore

lar cells. *Science*, 295(5564), 425-427.


## **GaAsN Grown by Chemical Beam Epitaxy for Solar Cell Application**

Kazuma Ikeda, Han Xiuxun, Bouzazi Boussairi and Yoshio Ohshita

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/51885

## **1. Introduction**

[17] Ng, A., Li, C. H., Fung, M. K., Djurišić, A. B., Zapien, J. A., Chan, W. K., Cheung, K. Y., Wong, W., & , Y. (2010). Accurate Determination of the Index of Refraction of Pol‐ ymer Blend Films by Spectroscopic Ellipsometry. *J. Phys. Chem. C*, 114(35),

[18] Rakić, A. D. (1995). Algorithm for the determination of intrinsic optical constants of

[19] Dennler, G., Forberich, K., Ameri, T., Waldauf, C., Denk, P., Brabec, C. J., Hingerl, K., & Heeger, A. J. (2007). Design of efficient organic tandem cells: On the interplay be‐ tween molecular absorption and layer sequence. *J. Appl.Phys*, 102(12),

metal films: application to aluminum. *Applied Optics*, 34(22), 4755-4767.

15094-15101.

280 Solar Cells - Research and Application Perspectives

123109-1-123109-6.

InGaAsN is a candidate material to realize the ultrahigh efficiency lattice-matched multijunction solar cell. This material has the 1eV band gap energy and same lattice constant as GaAs or Ge substrate by controlling the In and N compositions to be 9% and 3%, respectively [1, 2]. So far, the highest conversion efficiency of the lattice-matched multi-junction solar cell is 43.5% at 418-suns which was achieved by the 3-junction solar cell, GaInP/GaAs/GaInNAs [3]. By realizing the 4-junction device, InGaP/InGaAs/InGaAsN/Ge, the efficiency is expected to be 41% at 1-sun under the AM1.5G spectrum and 51% at 500-suns under the AM1.5D spectrum [4]. In order to achieve the expected super high efficiency, the short circuit current, Jsc, under a GaAs filter has to be 17 mA/cm2 at 1-sun under AM0 conditions. However, the highest Jsc under that condition has been 10.9mA/cm2 . This corresponds to the 6.2% conversion efficiency and Jsc=26.0 mA/cm2 without GaAs filters at 1-sun under AM1.5 [5]. The diffusion length of the minority carrier in this case is supposed to be much shorter than 1 μm which is required to achieve the ultrahigh efficiency. Therefore, the mobility and lifetime of the minority carrier should be improved.

The large miscibility gap is a particular characteristic of (In)GaAsN. It means that the phase separation easily occurs in the near equilibrium condition. It is caused by the large difference between the atomic radii of As and N [6]. Because of this property, homogeneous alloys are difficult to be obtained. Various growth techniques have been tried to avoid the phase separation and obtain good alloys: the plasma-assisted molecular beam epitaxy (MBE) [7], solid source MBE with a radio frequency (RF) nitrogen plasma source [8], metal organic vapor phase epitaxy (MOVPE) [9], metalorganic MBE [10-13], and chemical beam epitaxy (CBE) [14].

© 2013 Ikeda et al.; licensee InTech. This is an open access article 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. © 2013 Ikeda et al.; licensee InTech. This is a paper 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.

The chemical reactions at the surface of the substrates are expected to occur in the far-fromequilibrium states in those methods.

The electrical properties of (In)GaAsN has been reported to be drastically deteriorated by increasing the N composition [15]. The Hall hole mobility of p-GaAsN as a function of the N composition is shown in Figure 1 [15, 16]. It seems that the hole mobility is largely determined by the amount of N atoms and independent of the growth methods. This indicates that the N atoms contribute to the formation of dominant scattering centers. The electrical deterioration has been considered to be caused by the inhomogeneity of the N distribution. Therefore, it may be possible to improve the electrical property by controlling the growth process. In this work, CBE technique was used for the growth of GaAsN. This method has been developed in our group and shown the improvement in the electrical properties [18, 19]. 2 Book Title

**Fig.1.**TheN composition dependence of the hall mobility ofp-GaAsN. The dashed line is a **Figure 1.** The N composition dependence of the hall mobility of p-GaAsN. The dashed line is a guide for the eye.

2 In this chapter, the improvement of the carrier mobility and minority carrier lifetime in

7 The CBE techniquehasboth natures of themetal organic chemical vapor deposition 8 (MOCVD) and molecular beam epitaxy (MBE):metal organic gas sourcesand low

Substrate Semi-insulating GaAs(001)with 2°offcut toward [010]

Ga TEGa [Ga(C2H5)3] 0.02 ‐ 0.1

As TDMAAs [As(N(CH3)2)3] 1.0

N MMHy [(CH3)N2H3] 9.0

guide for the eye. In this chapter, the improvement of the carrier mobility and minority carrier lifetime in GaAsN grown by CBE is described. The effects of the growth rate and substrate orientation on these electrical properties are discussed.

#### 3 GaAsN grown by CBE isdescribed. The effects of the growth rate and substrate orientation **2. Chemical beam epitaxy**

6 **2.1. Growth procedure ofchemical beam epitaxy** 

**Table 1.**The growth condition of the CBE technique.

Growth pressure 2×10-2 Pa

Post annealing 5 min. at 500 �

Growth temperature (�) 340 ∼480

Growth time (min.) 30

4 on these electrical properties are discussed.

Flow rate

(sccm)

1

Source

### **2.1. Growth procedure of chemical beam epitaxy**

5 **2. Chemical Beam Epitaxy**  The CBE technique has both natures of the metal organic chemical vapor deposition (MOCVD) and molecular beam epitaxy (MBE): metal organic gas sources and low pressure growth [19]. In addition, the growth temperature is lower than the other methods. The low temperature is expected to be advantageous to obtain a non-equilibrium condition and homogeneous distribution of N in As sites. The growth condition of the CBE technique is shown in Table 1. The semi-insulating GaAs(001) oriented in 2° towards [010] was used as the substrate. The growth pressure is around 10-2 Pa or below, therefore the mean free path of the source gas molecules is in the order of 10-1 m. Triethylgallium [TEGa, Ga(C2H5)3], and monomethylhy‐ drazine [MMHy, (CH3)N2H3], were used as the sources of Ga, and N. Trisdimethylaminoar‐ senic [TDMAAs, As(N(CH3)2)3] was used as the As source instead of As2 obtained by cracking As2H4. Typical flow rates of TEGa, TDMAAs, and MMHy were 0.1, 1.0, and 9.0 sccm, respec‐ tively. The substrate was annealed in the chamber at a temperature of 500 °C for 5 minutes with supplying TDMAAs to remove the oxidized layer on the surface. Then, GaAsN thin film was grown by supplying TEGa and MMHy. The growth temperature was 340 to 460 °C. The growth time was 30 minutes.


**Table 1.** The growth condition of the CBE technique.

The chemical reactions at the surface of the substrates are expected to occur in the far-from-

The electrical properties of (In)GaAsN has been reported to be drastically deteriorated by increasing the N composition [15]. The Hall hole mobility of p-GaAsN as a function of the N composition is shown in Figure 1 [15, 16]. It seems that the hole mobility is largely determined by the amount of N atoms and independent of the growth methods. This indicates that the N atoms contribute to the formation of dominant scattering centers. The electrical deterioration has been considered to be caused by the inhomogeneity of the N distribution. Therefore, it may be possible to improve the electrical property by controlling the growth process. In this work, CBE technique was used for the growth of GaAsN. This method has been developed in

2 Book Title

our group and shown the improvement in the electrical properties [18, 19].

**0**

**100**

**Hall mobility [cm2/Vs]**

**200**

**300**

2 In this chapter, the improvement of the carrier mobility and minority carrier lifetime in 3 GaAsN grown by CBE isdescribed. The effects of the growth rate and substrate orientation

**Fig.1.**TheN composition dependence of the hall mobility ofp-GaAsN. The dashed line is a

**Figure 1.** The N composition dependence of the hall mobility of p-GaAsN. The dashed line is a guide for the eye.

In this chapter, the improvement of the carrier mobility and minority carrier lifetime in GaAsN grown by CBE is described. The effects of the growth rate and substrate orientation on these

The CBE technique has both natures of the metal organic chemical vapor deposition (MOCVD) and molecular beam epitaxy (MBE): metal organic gas sources and low pressure growth [19].

**0.0 0.5 1.0**

**N composition [%]**

[15][16]

[15]

CBE Tg = 420 o

MBE

MOCVD

C

7 The CBE techniquehasboth natures of themetal organic chemical vapor deposition 8 (MOCVD) and molecular beam epitaxy (MBE):metal organic gas sourcesand low

Substrate Semi-insulating GaAs(001)with 2°offcut toward [010]

Ga TEGa [Ga(C2H5)3] 0.02 ‐ 0.1

As TDMAAs [As(N(CH3)2)3] 1.0

N MMHy [(CH3)N2H3] 9.0

equilibrium states in those methods.

282 Solar Cells - Research and Application Perspectives

4 on these electrical properties are discussed.

Flow rate

(sccm)

6 **2.1. Growth procedure ofchemical beam epitaxy** 

electrical properties are discussed.

**2. Chemical beam epitaxy**

**2.1. Growth procedure of chemical beam epitaxy**

**Table 1.**The growth condition of the CBE technique.

Growth pressure 2×10-2 Pa

Post annealing 5 min. at 500 �

Growth temperature (�) 340 ∼480

Growth time (min.) 30

5 **2. Chemical Beam Epitaxy** 

Source

guide for the eye.

1

The N composition was estimated from the lattice constant of GaAsN(004) determined by Xray diffraction (XRD). Here, elastic distortion and Vegard's law were assumed [20]. The growth rate was determined by the film thickness estimated from the interference fringe peaks in an XRD curve and growth time, or the Dektak surface profilometer in the case that no fringe peaks appeared. The crystal quality of the films was evaluated by the full width at half maximum (FWHM) of the rocking curve of a GaAsN(004) peak.

### **2.2. Growth of GaAsN thin films**

The growth temperature dependences of the growth rates are shown in Figure 2. The growth temperature can be divided into three regions. In the lower temperature region (340 – 390 °C), the growth rate increases with increasing temperature. In the middle temperature region (390 – 445 °C), the growth rate decreases with increasing temperature. In the higher temperature region (445 – 480 °C), the growth rate slightly changes. The relationship between the TEGa flow rate and growth rate in each growth temperature region is shown in Figure 3. The growth temperatures of 360, 420, and 480 °C in Figure 3 belong to the low, middle, and high growth temperature regions, respectively. Here, the TDMAAs flow rate was 1.0 sccm and MMHy flow rate was 9.0 sccm. In the case that the growth temperature was 360 °C (in the low temperature

**Figure 2.** The dependence of the growth rate on the growth temperature of GaAs and GaAsN thin films.

**Figure 3.** The dependence of TEGa flow rate on the growth rate of GaAsN thin films.

region), the growth rate saturated above the TEGa flow rate of 0.1 sccm. It indicates that the factor that determines the growth rate changed from the TEGa supply to another process such as the decomposition rate of TEGa on the surface. At higher growth temperatures (in the middle and high temperature regions), the growth rate linearly increased as increasing TEGa flow rate, which indicates that the TEGa supply was dominant to determine the growth rate.

**Figure 4.** The XRD rocking curves of the GaAsN thin films grown by CBE. The corresponding growth temperature and N composition of each sample is also shown

The XRD rocking curves of GaAsN(004) are shown in Figure 4. The GaAsN films were grown at the temperatures of 340, 360, 380, and 480 °C. There are interference fringes in the rocking curves of the last three films, which suggest the uniformity of the lattice constant with flat interface. On the other hand, there is no fringe in the rocking curve of the film grown at 340 °C, indicating a poor quality of the crystal. The N compositions estimated from the XRD measurements monotonically decreased by increasing the growth temperatures as shown in Figure 5. This is the same tendency as those in the temperature dependence of the growth rate. In the middle region of the growth temperature (390 – 445 °C), the decrement of the N composition is relatively small compared to those in the lower (340 – 390 °C) and higher (445 – 480 °C) regions.

region), the growth rate saturated above the TEGa flow rate of 0.1 sccm. It indicates that the factor that determines the growth rate changed from the TEGa supply to another process such

**0.00 0.05 0.10 0.15 0.20 0.0**

**TEGa flow rate [sccm]**

**<sup>360</sup> <sup>400</sup> <sup>440</sup> <sup>480</sup> 10-1**

**Figure 2.** The dependence of the growth rate on the growth temperature of GaAs and GaAsN thin films.

**360o C 420o C 480o C**

**Growth temperature [o**

**C]**

**GaAsN GaAs**

**100**

**1.0**

**Figure 3.** The dependence of TEGa flow rate on the growth rate of GaAsN thin films.

**2.0**

**Growth rate [m/h]**

**3.0**

**4.0**

**Growth rate [m/h]**

284 Solar Cells - Research and Application Perspectives

**101**

**Figure 5.** The dependence of growth temperature on the N composition in GaAsN thin films.

### **2.3. Crystal quality of GaAsN grown by CBE**

The FWHM's of measured samples divided by the theoretical calculations of GaAsN(004) peaks are shown in Figure 6. To remove the contribution of the film thickness, those were normalized by using the Schereer's equation. When the film was grown above 360 °C, the ratios are almost 1.0, which indicates good crystal quality. On the other hand, the large ratio of the film grown at 340 °C means poor quality of the crystal, corresponding to no fringe in XRD spectra. This deterioration of the crystal quality possibly occurred due to the incorporation of a large amount of impurities, such as hydrogen and carbon.

The dependence of the N composition on the growth temperature as shown in Figure 5 can be explained qualitatively as follows. A rapid decrease in the N composition with increasing the growth temperature in the low growth temperature region is due to both the increases in the growth rate and desorption rate of N species at the growing surface. Here, we assume that the amount of N species supplied to the growing surface is independent of growth temperatures. Then, the higher growth rate caused the decrement of the N source supply per unit layer growth, since the N source flow rate was constant in every growth conditions. In the middle growth temperature region, the decrement of the growth rate as increasing the growth temperature would enhance the N source supply per unit layer growth. However, the N composition did not increase. It indicates that the desorption rate of N species prevailed in the middle growth temperature region to reduce the N composition as the growth temperature increased. In the high growth temperature region, the N composition again markedly de‐ creases with increasing the growth temperature. The fact that the growth rate remains almost

**Figure 6.** The ratio of FWHM obtained from the measured samples and theoretical calculation at each growth tem‐ perature.

constant in that region as shown in Figure 2 indicates that the increased desorption rate of N species are dominant to determine the N composition. Although these three dependences were also observed in the CBE-grown GaAsN with the atomic N source [14], the slope of XRD rocking curves in each region is steeper in our result. This is considered to mainly come from the difference in desorption rate between the N sources on the growing surface. To further understand the N incorporation mechanism in CBE with compound N sources, such as MMHy, it is necessary to consider the desorption rate of N species.

## **3. Hole mobilities**

**320 360 400 440 480**

**Growth temperature [o**

The FWHM's of measured samples divided by the theoretical calculations of GaAsN(004) peaks are shown in Figure 6. To remove the contribution of the film thickness, those were normalized by using the Schereer's equation. When the film was grown above 360 °C, the ratios are almost 1.0, which indicates good crystal quality. On the other hand, the large ratio of the film grown at 340 °C means poor quality of the crystal, corresponding to no fringe in XRD spectra. This deterioration of the crystal quality possibly occurred due to the incorporation of

The dependence of the N composition on the growth temperature as shown in Figure 5 can be explained qualitatively as follows. A rapid decrease in the N composition with increasing the growth temperature in the low growth temperature region is due to both the increases in the growth rate and desorption rate of N species at the growing surface. Here, we assume that the amount of N species supplied to the growing surface is independent of growth temperatures. Then, the higher growth rate caused the decrement of the N source supply per unit layer growth, since the N source flow rate was constant in every growth conditions. In the middle growth temperature region, the decrement of the growth rate as increasing the growth temperature would enhance the N source supply per unit layer growth. However, the N composition did not increase. It indicates that the desorption rate of N species prevailed in the middle growth temperature region to reduce the N composition as the growth temperature increased. In the high growth temperature region, the N composition again markedly de‐ creases with increasing the growth temperature. The fact that the growth rate remains almost

**C]**

**0.1**

**2.3. Crystal quality of GaAsN grown by CBE**

a large amount of impurities, such as hydrogen and carbon.

**Figure 5.** The dependence of growth temperature on the N composition in GaAsN thin films.

**1.0**

**N composition [%]**

286 Solar Cells - Research and Application Perspectives

**5.0**

### **3.1. Hall hole mobility**

The hole mobility in p-GaAsN films grown by CBE, MBE, and MOCVD is shown in Figure 7 as a function of N composition [16, 21]. Those grown by the CBE in our group were obtained by Van der Pauw method. The dashed line represents the mobility obtained theoretically by considering the alloy and phonon scatterings. The hole mobility monotonically decreases as increasing the N composition and they are smaller than the theoretical values. The deviation of the mobility from the ideal value is at the same level despite the different impurity concen‐ trations [16, 21]. This indicates that some N-related defects are generated due to the N incorporation, which determine the hole mobility at the room temperature. To evaluate the number of N-related defects, the temperature dependence (77-400K) of the hole mobility was 1 **3. Hole mobilities** 

20 follows:

21

guide for the eye.

2 **3.1. Hall hole mobility** 

obtained as shown in Figure 8. The result suggests that there are several different scattering mechanisms. To determine the amount of the N-related scattering centers, we analyzed this temperature dependence. Here, ionized impurity scattering, alloy scattering, phonon scatter‐ ing, and N-related carrier scattering were considered [22, 23]. The inverse of the mobility limited by the N-related carrier scattering 1/μN, which is proportional to the density of Nrelated scattering centers, was determined by excluding the contributions of the mobility limited by ionized impurity (μII), alloy scattering (μAL), and phonon scattering (μPN) from the measured mobility (μexp), as follows: 7 monotonically decreases as increasing the N composition and they are smaller than the 8 theoretical values. The deviation of the mobility from the ideal value is at the same level 9 despite the different impurity concentrations[17], [16].This indicates that some N-related 10 defects are generated due to the N incorporation, which determine the hole mobility at the 11 room temperature. To evaluate the number of N-related defects, the temperature 12 dependence (77-400K) of the hole mobility was obtained as shown in Figure 8. The result 13 suggests that there are several different scattering mechanisms. To determine the amount of

8 Book Title

3 The holemobility in p-GaAsN films grown by CBE, MBE, and MOCVD is shown in Figure7 4 as a function of N composition[17], [16]. Those grown by the CBE in our group were

6 obtainedtheoretically by considering the alloy and phonon scatterings. The hole mobility

$$\frac{1}{\mu\_{\rm N}} = \frac{1}{\mu\_{\rm exp}} - \frac{1}{\mu\_{\rm II}} - \frac{1}{\mu\_{\rm AL}} - \frac{1}{\mu\_{\rm PN}}.\tag{1}$$

Here, 1/μPN is the sum of the contributions from the acoustical phonon (μAC), polar optical phonon (μPO), and non-polar optical phonon scattering (μNPO) as 17 scattering 1/μN, which is proportional to the density of N-related scattering centers, was 18 determined by excluding the contributions of the mobility limited by ionized impurity (μII),

19 alloy scattering (μAL), and phonon scattering (μPN) from the measured mobility (μexp), as

$$\frac{1}{\mu\_{\rm PN}} = \frac{1}{\mu\_{\rm AC}} + \frac{1}{\mu\_{\rm PO}} + \frac{1}{\mu\_{\rm NPO}}.\tag{2}$$

**Fig. 7.**Hole mobility at room temperature in p-GaAsN films grown by CBE, MBE, and MOCVD as a function of N composition. The mobility is the value at room temperature. The **Figure 7.** Hole mobility at room temperature in p-GaAsN films grown by CBE, MBE, and MOCVD as a function of N composition. The mobility is the value at room temperature. The dashed line represents the estimated mobility limited by the alloy and phonon scatterings. In the case of CBE, the growth rate (GR) is shown in the unit of μm/h. The dotted lines are a guide for the eye.

dashed line represents the estimated mobility limited by the alloy and phonon scatterings. In the case of CBE, the growth rate (GR) is shown in the unit of μm/h. The dotted lines are a

9

**Fig. 8.** Temperature dependence of the hole mobility of p-GaAsN films with N composition of approximately 0.34% (μexp.) and its components denoted in eq. (1). The contribution from **Figure 8.** Temperature dependence of the hole mobility of p-GaAsN films with N composition of approximately 0.34% (μexp.) and its components denoted in eq. (1). The contribution from N-related scattering centers (μN) is ob‐ tained by using eq. (1). The mobility with N composition of approximately 0.79% (μ\*exp.) and contribution from the Nrelated scattering centers (μ\*N) are also shown.

N-related scattering centers (μN) is obtained by using eq. (1). The mobility with N

. <sup>1</sup> <sup>1</sup> <sup>1</sup> <sup>1</sup> <sup>1</sup>

*AL*

**N composition [%]**

*PN*

*II*

**0.0 0.5 1.0 0 Figure 9.** 1/μN at each N composition in p-GaAsN films.

*N*

**Fig. 9.**1/μNat each N composition in p-GaAsN films.

exp

1 (1)

obtained as shown in Figure 8. The result suggests that there are several different scattering mechanisms. To determine the amount of the N-related scattering centers, we analyzed this temperature dependence. Here, ionized impurity scattering, alloy scattering, phonon scatter‐ ing, and N-related carrier scattering were considered [22, 23]. The inverse of the mobility limited by the N-related carrier scattering 1/μN, which is proportional to the density of Nrelated scattering centers, was determined by excluding the contributions of the mobility limited by ionized impurity (μII), alloy scattering (μAL), and phonon scattering (μPN) from the

3 The holemobility in p-GaAsN films grown by CBE, MBE, and MOCVD is shown in Figure7 4 as a function of N composition[17], [16]. Those grown by the CBE in our group were 5 obtained by Van der Pauw method. The dashed line represents the mobility 6 obtainedtheoretically by considering the alloy and phonon scatterings. The hole mobility 7 monotonically decreases as increasing the N composition and they are smaller than the 8 theoretical values. The deviation of the mobility from the ideal value is at the same level 9 despite the different impurity concentrations[17], [16].This indicates that some N-related 10 defects are generated due to the N incorporation, which determine the hole mobility at the 11 room temperature. To evaluate the number of N-related defects, the temperature 12 dependence (77-400K) of the hole mobility was obtained as shown in Figure 8. The result 13 suggests that there are several different scattering mechanisms. To determine the amount of 14 the N-related scattering centers, we analyzed this temperature dependence. Here, ionized 15 impurity scattering, alloy scattering, phonon scattering, and N-related carrier scattering 16 were considered[22], [23].The inverse of the mobility limited by the N-related carrier 17 scattering 1/μN, which is proportional to the density of N-related scattering centers, was 18 determined by excluding the contributions of the mobility limited by ionized impurity (μII), 19 alloy scattering (μAL), and phonon scattering (μPN) from the measured mobility (μexp), as

8 Book Title

exp

 

phonon (μPO), and non-polar optical phonon scattering (μNPO) as

**50**

**100**

**150**

**Mobility [cm2**

lines are a guide for the eye.

**V-1**

**-1**

**s**

**]**

**200**

**250**

1 111 1 . *N II AL PN*

111 1 . *PN AC PO NPO* 

Here, 1/μPN is the sum of the contributions from the acoustical phonon (μAC), polar optical

 

**0.4 0.6 0.8 1.0 1.2**

0.46

Lower GR

0.52

**N compositon [%]**

1.60

**Fig. 7.**Hole mobility at room temperature in p-GaAsN films grown by CBE, MBE, and MOCVD as a function of N composition. The mobility is the value at room temperature. The dashed line represents the estimated mobility limited by the alloy and phonon scatterings. In the case of CBE, the growth rate (GR) is shown in the unit of μm/h. The dotted lines are a

**Figure 7.** Hole mobility at room temperature in p-GaAsN films grown by CBE, MBE, and MOCVD as a function of N composition. The mobility is the value at room temperature. The dashed line represents the estimated mobility limited by the alloy and phonon scatterings. In the case of CBE, the growth rate (GR) is shown in the unit of μm/h. The dotted

1.06

1.41

Alloy & Phonon Scatterings

= -- - (1)

<sup>288</sup> Solar Cells - Research and Application Perspectives Running Title

=++ (2)

 CBE MBE [1,2] MOCVD [1]

0.23

 

measured mobility (μexp), as follows:

1 **3. Hole mobilities** 

20 follows:

21

guide for the eye.

2 **3.1. Hall hole mobility** 

The expressions of each component of the mobility are described, for example, in ref. [16].

The relationship between 1/μN and N composition is shown in Figure 9. The inverse of the mobility limited by the N-related carrier scattering center (1/μN) was determined, which is proportional to the amount of N-related scattering centers. Independent of growth technique, the number of N related scattering centers (1/μ∝ ) increases as N composition increases.

### **3.2. Improvement of mobility by controlling the growth rate**

The above result suggested that the number of N-related defects may be determined by the N concentration of the GaAsN independent of the growth technique. However, here we show the increase of the mobility by controlling the growth rate in CBE. The hole mobility at room temperature and the N composition in p-GaAsN films as a function of growth rate are shown in Figure 10. The higher mobilities are obtained by decreasing the growth rate. Generally, the hole mobility is decreased by the increase in the N composition owing to the alloy scattering. To discuss the reason for the increase in the hole mobility, the relationship between the 1/μN and N composition are shown in Figure 11. When the growth rate is higher than 1.41 μm/h, there is no improvement of the mobility. By decreasing the growth rate lower than 1.04 μm/h, the 1/μN (the number of the N-related defects) remains almost constant despite the increase in the N composition. (μN [N])-1 as a function of the growth rate is shown in Figure 12. This is a relative amount of the N-related scattering centers to the total amount of N atoms. (μN [N])-1 is decreased by decreasing the growth rate monotonically. Then, the controlling growth rate in CBE is effective to suppress the formation of N-related scattering centers and to improve the mobility, especially for a higher N composition.

**Figure 10.** Hole mobility at room temperature and N composition in p-GaAsN films as a function of growth rate.

The expressions of each component of the mobility are described, for example, in ref. [16].

**3.2. Improvement of mobility by controlling the growth rate**

290 Solar Cells - Research and Application Perspectives

the mobility, especially for a higher N composition.

**250**

**100**

**150**

**Hole mobility [cm**

**2**

**V -1**

**s-1**

**]**

**200**

The relationship between 1/μN and N composition is shown in Figure 9. The inverse of the mobility limited by the N-related carrier scattering center (1/μN) was determined, which is proportional to the amount of N-related scattering centers. Independent of growth technique, the number of N related scattering centers (1/μ∝ ) increases as N composition increases.

The above result suggested that the number of N-related defects may be determined by the N concentration of the GaAsN independent of the growth technique. However, here we show the increase of the mobility by controlling the growth rate in CBE. The hole mobility at room temperature and the N composition in p-GaAsN films as a function of growth rate are shown in Figure 10. The higher mobilities are obtained by decreasing the growth rate. Generally, the hole mobility is decreased by the increase in the N composition owing to the alloy scattering. To discuss the reason for the increase in the hole mobility, the relationship between the 1/μN and N composition are shown in Figure 11. When the growth rate is higher than 1.41 μm/h, there is no improvement of the mobility. By decreasing the growth rate lower than 1.04 μm/h, the 1/μN (the number of the N-related defects) remains almost constant despite the increase in the N composition. (μN [N])-1 as a function of the growth rate is shown in Figure 12. This is a relative amount of the N-related scattering centers to the total amount of N atoms. (μN [N])-1 is decreased by decreasing the growth rate monotonically. Then, the controlling growth rate in CBE is effective to suppress the formation of N-related scattering centers and to improve

**0.0 0.5 1.0 1.5**

**Figure 10.** Hole mobility at room temperature and N composition in p-GaAsN films as a function of growth rate.

**Growth Rate [m/h]**

**0.0**

**0.5**

**N composition [%]**

**1.0**

**1.5**

 Mobility N composition

**Figure 11.** μN-1 at room temperature as a function of N composition in p-GaAsN films. μ<sup>N</sup> -1 is proportional to the amount of N-ralated scattering centers. In the case of CBE, the growth rate (GR) is shown in the unit of μm/h. The dotted lines are a guide for the eye.

**Figure 12.** (μ<sup>N</sup> [N])-1 as a function of growth rate of p-GaAsN films. (μ<sup>N</sup> [N])-1 is the relative amount of the N-related scattering centers to the total amount of N atoms. The dotted line is a guide for the eye.

We inferred two reasons for the reduction of the amount of N-related scattering centers by decreasing the growth rate: i) The amount of N adsorbed species which are desorbed from a terrace increases by decreasing the growth rate. We expect that the N adsorbed species adsorbed from a terrace is a cause of N-related scattering centers. The desorption from a terrace is expected to be more frequent than that from a step, since the number of connected bonds of N adsorbed species to the growing surface is smaller at terrace than at step; ii) The decrease in the growth rate increases the diffusion length of N adsorbed species on the growing surface. A large diffusion length is expected to increase the probability of a N adsorbed species to come over a step. Therefore, the probability of a N adsorbed species adsorbed at a step increases. Then, i) and ii) are expected to be advantageous for the improvement of the carrier mobility and lifetime.

## **4. Mobility contributed from N-related scattering center**

### **4.1. Minority-carrier lifetime**

#### *4.1.1. Improvement of minority carrier lifetime by controlling growth rate*

As the control of the growth rate improved the mobility, here we show that the decreasing the TEGa flow rate improves the minority carrier lifetime.. The minority-carrier lifetime at room temperature was obtained by time-resolved photoluminescence (TR-PL). A titanium sapphire pulse laser was used for the carrier excitation. The excitation power, wavelength, pulse width, and recurrence frequency were 900 mW, 800 nm, ∼100 fs, and 80 MHz, respectively. To evaluate the minority carrier lifetime of p-GaAsN, the structure of GaAs buffer layer (500 nm)/ p-GaAsN layer (140∼950 nm)/GaAs cap layer (30 nm) was adopted. The cap layer reduced the surface recombination. The PL lifetime (τPL) was obtained by PL decay curves. τPL consists of minority-carrier lifetimes in the bulk of the p-GaAsN layer (τB) and at GaAs/GaAsN interfaces (τS). The contribution of τS to τPL changes as a function of the p-GaAsN layer thickness (d)16) as follows:

$$\frac{1}{\tau\_{PL}} = \frac{1}{\tau\_B} + \frac{1}{\tau\_S} = \frac{1}{\tau\_B} + \frac{2S}{d}.\tag{3}$$

Here, S is the GaAs/GaAsN interface recombination velocity. PL lifetime as a function of GaAsN layer thickness was obtained and τB was determined by eq. (3). The amount of nonradiative recombination centers were estimated qualitatively based on eq. (4),

$$N\_{NR} \propto \frac{1}{\tau\_{NR}} = \frac{1}{\tau\_B} - \frac{1}{\tau\_R} \, \tag{4}$$


We inferred two reasons for the reduction of the amount of N-related scattering centers by decreasing the growth rate: i) The amount of N adsorbed species which are desorbed from a terrace increases by decreasing the growth rate. We expect that the N adsorbed species adsorbed from a terrace is a cause of N-related scattering centers. The desorption from a terrace is expected to be more frequent than that from a step, since the number of connected bonds of N adsorbed species to the growing surface is smaller at terrace than at step; ii) The decrease in the growth rate increases the diffusion length of N adsorbed species on the growing surface. A large diffusion length is expected to increase the probability of a N adsorbed species to come over a step. Therefore, the probability of a N adsorbed species adsorbed at a step increases. Then, i) and ii) are expected to be advantageous for the improvement of the carrier mobility

As the control of the growth rate improved the mobility, here we show that the decreasing the TEGa flow rate improves the minority carrier lifetime.. The minority-carrier lifetime at room temperature was obtained by time-resolved photoluminescence (TR-PL). A titanium sapphire pulse laser was used for the carrier excitation. The excitation power, wavelength, pulse width, and recurrence frequency were 900 mW, 800 nm, ∼100 fs, and 80 MHz, respectively. To evaluate the minority carrier lifetime of p-GaAsN, the structure of GaAs buffer layer (500 nm)/ p-GaAsN layer (140∼950 nm)/GaAs cap layer (30 nm) was adopted. The cap layer reduced the surface recombination. The PL lifetime (τPL) was obtained by PL decay curves. τPL consists of minority-carrier lifetimes in the bulk of the p-GaAsN layer (τB) and at GaAs/GaAsN interfaces (τS). The contribution of τS to τPL changes as a function of the p-GaAsN layer thickness (d)16)

> 1 1 1 12 . *PL B S B*

ttt

nonradiative recombination centers were estimated qualitatively based on eq. (4),

t

1 11 , *NR NR B R*

 tt *S*

=+= + (3)

µ =- (4)

*d*

Here, S is the GaAs/GaAsN interface recombination velocity. PL lifetime as a function of GaAsN layer thickness was obtained and τB was determined by eq. (3). The amount of

**4. Mobility contributed from N-related scattering center**

*4.1.1. Improvement of minority carrier lifetime by controlling growth rate*

t

*N*

and lifetime.

as follows:

**4.1. Minority-carrier lifetime**

292 Solar Cells - Research and Application Perspectives

**Table 2.** Summary of growth rate (GR), N composition ([N]), minority-carrier lifetime in the bulk of p-GaAsN (τB), and radiative, nonradiative recombination lifetime (τR, τNR). τB was estimated by the PL lifetime dependence on the thickness of the p-GaAsN layer. τR was determined by using the hole concentration (p). τNR is obtained by using eq. (3).

**Figure 13.** Fitting results of PL lifetime (τPL) dependence on GaAsN layer thickness (d). The inserted figure shows a typi‐ cal PL decay curve at room temperature, which is for the sample with GaAsN layer thickness of approximately 500 nm grown at a growth rate of 0.4 μm/h.

where NNR is the density of the nonradiative recombination centers, and τNR and τR are the nonradiative and radiative recombination lifetimes, respectively. τR was determined by the hole concentration (p) theoretically.

The decay of PL intensity was shown in Figure 13. The following equation

$$I\_{PL}(t) = I\_0 \exp\left(-\frac{t}{\tau}\right). \tag{5}$$

explains well about the PL decay curve. Here, IPL(t) is the PL intensity at the time t, and I0 is the PL intensity at t = 0, just after the carriers are excited. The lifetime τ obtained by eq. (5) is the PL lifetime (τPL) in eq. (3). To estimate the minority-carrier lifetime in the bulk of the pGaAsN layer (τB), τPL dependence on GaAsN layer thickness (d) was obtained (Figure 13). When the growth rate was 2 μm/h, τB was 3.2 x 10-1 ns ([N] = 0.6%). By decreasing the growth rate to 0.4 μm/h, τB was increased to 9.0 x 10-1 ns ([N] = 0.8%) despite the increase in the N composition. Therefore, decreasing the growth rate in CBE is effective to improve τB. Further, the improvement is more effective when the N composition is higher.

The density of the nonradiative recombination centers determines the nonradiative recombi‐ nation lifetime (τNR). To discuss the amount of nonradiative recombination centers, τNR is estimated by eq. (4). The recombination lifetime (τR) depends on the hole concentration (p), as follows:

$$
\tau\_R = \frac{1}{Bp}.\tag{6}
$$

Here, B is the radiative recombination probability,

$$\begin{aligned} B &= 0.58 \times 10^{-12} \sqrt{\varepsilon} \left( \frac{1}{m\_p^\* + m\_n^\*} \right)^{1.5} \\ \times \left( 1 + \frac{1}{m\_p^\*} + \frac{1}{m\_n^\*} \right) \left( \frac{300}{T} \right)^{1.5} E\_g^{-2} \end{aligned} \tag{7}$$

where ε is the dielectric constant, mp\* and mn\* are the hole and electron effective masses, respectively, in units of the free electron mass, T is the temperature, and Eg is the band gap. For the p-GaAsN layers discussed here, B calculated by eq. (7) was 1.2 x 10-10 cm3/s. When the growth rate was 2 μm/h, the carrier concentration was 1 x 1017 cm-3. By decreasing the growth rate to 0.4 μm/h, it was decreased to 5 x 1016 cm-3. From eqs. (4) and (6), τNR and τ<sup>R</sup> were obtained to be τNR = 3.2 x 10-1 ns and τ<sup>R</sup> = 8.0 x 101 ns for the growth rate of 2 μm/h, and τNR = 9.0 x 10-1 ns and τR = 1.7 x 102 ns for the growth rate of 0.4 μm/h. These results are summarized in Table 2. Since τNR is much smaller than τR, τB is mainly determined by τNR. Therefore, the improvement of τB is mainly due to the increase in τNR. These results suggest that decreasing the growth rate in CBE is effective to suppress the formation of nonradiative recombination centers and to improve the carrier lifetime. Then, CBE is highly expected to realize the minority-carrier lifetime of more than 1 ns and the diffusion length of more than 1 μm with the target value of the N composition (3%).

#### **4.2. Improvement of minority carrier lifetime by controlling substrate orientation**

The effect of the surface orientation on the minority carrier lifetime is studied. GaAsN films were grown on GaAs(311)A, (311)B, and (100) substrates. The (311) surface consists of (111) and (100) components. The surface configurations of the (311)A and (311)B surfaces are shown in Figure 14. On the (311)A surface, a Ga atom connects to three As atoms, while an As atom connects to two Ga atoms. The same bond configuration holds for the (311)B surface by exchanging Ga and As atoms.

GaAsN layer (τB), τPL dependence on GaAsN layer thickness (d) was obtained (Figure 13). When the growth rate was 2 μm/h, τB was 3.2 x 10-1 ns ([N] = 0.6%). By decreasing the growth rate to 0.4 μm/h, τB was increased to 9.0 x 10-1 ns ([N] = 0.8%) despite the increase in the N composition. Therefore, decreasing the growth rate in CBE is effective to improve τB. Further,

The density of the nonradiative recombination centers determines the nonradiative recombi‐ nation lifetime (τNR). To discuss the amount of nonradiative recombination centers, τNR is estimated by eq. (4). The recombination lifetime (τR) depends on the hole concentration (p), as

> <sup>1</sup> . *<sup>R</sup> Bp* t

> > 12

<sup>1</sup> 0.58 10

= ´ ç ÷

e- æ ö

1 1 300 1 ,

**4.2. Improvement of minority carrier lifetime by controlling substrate orientation**

The effect of the surface orientation on the minority carrier lifetime is studied. GaAsN films were grown on GaAs(311)A, (311)B, and (100) substrates. The (311) surface consists of (111) and (100) components. The surface configurations of the (311)A and (311)B surfaces are shown in Figure 14. On the (311)A surface, a Ga atom connects to three As atoms, while an As atom

where ε is the dielectric constant, mp\* and mn\* are the hole and electron effective masses, respectively, in units of the free electron mass, T is the temperature, and Eg is the band gap. For the p-GaAsN layers discussed here, B calculated by eq. (7) was 1.2 x 10-10 cm3/s. When the growth rate was 2 μm/h, the carrier concentration was 1 x 1017 cm-3. By decreasing the growth rate to 0.4 μm/h, it was decreased to 5 x 1016 cm-3. From eqs. (4) and (6), τNR and τ<sup>R</sup> were obtained to be τNR = 3.2 x 10-1 ns and τ<sup>R</sup> = 8.0 x 101 ns for the growth rate of 2 μm/h, and τNR = 9.0 x 10-1 ns and τR = 1.7 x 102 ns for the growth rate of 0.4 μm/h. These results are summarized in Table 2. Since τNR is much smaller than τR, τB is mainly determined by τNR. Therefore, the improvement of τB is mainly due to the increase in τNR. These results suggest that decreasing the growth rate in CBE is effective to suppress the formation of nonradiative recombination centers and to improve the carrier lifetime. Then, CBE is highly expected to realize the minority-carrier lifetime of more than 1 ns and the diffusion length of more than 1

\* \*

æ öæ ö ´+ + ç ÷ç ÷ ç ÷è ø è ø

*p n*

*m m T*

= (6)

(7)

1.5

\* \*

*p n*

*g*

*E*

1.5 2

*m m*

ç ÷ <sup>+</sup> è ø

the improvement is more effective when the N composition is higher.

Here, B is the radiative recombination probability,

294 Solar Cells - Research and Application Perspectives

*B*

μm with the target value of the N composition (3%).

follows:

**Figure 14.** The surface structures of GaAs(311)A (left figure) and (311)B (right figure) substrates. Green balls are Ga atoms, blue balls are As atoms, and yellow and red sticks are Ga-As bonds.

**Figure 15.** Typical PL decay curves at 4.5K of GaAsN with different growth orientations. The growth temperature was 440 ℃.

The results of the TR-PL measurements at 4.5K of GaAsN films with each growth orientation are shown in Figure 15. The PL lifetime is estimated by approximating the decay curve of the PL intensity as a straight line. The film with (311)B orientation has a longer PL lifetime, which indicates the longer minority carrier lifetime. The PL spectra at 4.5K of as-grown films with various substrate orientations are shown in Figure 16(a). The growth temperature was 440 ℃. There are two GaAsN-related peaks denoted as BE (near-band edge) and DL (deep levels) and one GaAs-related peaks in the spectra. The GaAsN-related peaks are in the range of 1.25 – 1.40 eV and 1.1 – 1.2 eV which are caused by BE and DL emissions, respectively. To understand the orientation-dependent photoluminescence of GaAsN films, the ratio of the integrated inten‐ sities of DL to BE (PDL/PBE) were calculated. The PDL/PBE ratio of the film grown on (311)A had the lowest value at any growth temperatures as shown in Figure 16 (b). The film grown on (311)A has almost the same N composition as that of (100). Therefore, N composition is not related to the difference in the integrated intensity ratios. This result indicates that the highindex substrate surfaces contribute to reduce defects and composition fluctuations in the epilayers [24]. On the other hand, in contrast to the continuous increase in total emission intensities of (3 1 1)B sample, a decreasing tendency with a rise in growth temperature was recorded for (3 1 1)A.

**Figure 16.** (a) PL spectra at 4.5K of as-grown films with different substrate orientations. The growth temperature was 440 ℃. (b) Growth-temperature dependences of N composition in GaAsN layers on (3 1 1)A/B and (1 0 0) substrates.

### **5. Lattice defects in GaAsN grown by CBE**

GaInAsN suffers from three main N-related problems: (i) the shortening lifetime of minority carriers, (ii) the lowering of their mobility, and (iii) the increase of carrier concentration with increasing N concentration in the alloy [25-27]. Indeed, the incorporation of a small atomic fraction of N in the lattice of GaAs gives rise to the formation of high density acceptor states, recombination centers, and scattering levels [28]. It is; therefore, clear that the degradation of electrical properties of GaAsN may be attributed only to the formation of lattice defects. These defects were expected to be formed during growth owing to the smaller atomic size of N than of As, as well as, to the large gap miscibility between GaAs and GaN. For that, extensive works on the distribution of lattice defects in the forbidden gap of Ga(In)AsN and their effects on their electrical properties were published during the last decade. Indeed, the summary of experimental results reported by Geisz and Friedman provides a basic knowledge of these defects [4]. Furthermore, undoped Ga(In)AsN alloys grown by chemical compound sources present a high density of residual impurities [29]. This property is a hindrance to a flexible design of Ga(In)AsN based solar cell and reduces the lifetime of minority carriers. Despite the very interesting results about the distribution of lattice defects in GaAsN, the main causes of short diffusion length of minority carriers and high background doping are still elusive. In this section of the chapter, we summarize the distribution of energy states in the forbidden gap of GaAsN grown by CBE, which was obtained using deep level transient spectroscopy and some related methods. Furthermore, we report the relationship of some defects and the poor electrical properties of the alloy, as well as, our recent results to decrease their impacts. All GaAsN samples were grown by CBE on GaAs (100) 2° off toward (110) substrate, under a growth temperature and a pressure of 420-460 °C and 2 x 10-2 Pa, respectively. TEGa, TDMAAs, and MMHy were used as Ga, As, and N chemical compound sources, respectively. SiH4 was used as n-type doping source, whereas undoped films are p-type. The detail of growth conditions using CBE can be found elsewhere [30-33]. The N concentration in each sample were evaluated from the Bragg angles of the GaAs and GaAsN reflection obtained by high resolution X-ray diffraction (HRXRD) method. The Schottky and ohmic contacts were evapo‐ rated through metal masks at a vacuum pressure of 10-4 Pa. The free donor and acceptor concentrations were calculated by the fitting of the Mott-Schottky plot using the capacitancevoltage method. DLTS spectra were collected using a BIO-RAD digital DLTS system (DL8000). The activation energies Ea [Ev + ET (eV)] for hole traps and Ea [EC - ET (eV)] for electron trap were determined from the slope of the Arrhenius plot of the DLTS signals, whereas the capture cross sections n/p (cm2) were evaluated from the intercept values [34]. Illustrated in Figure 17 (a) is the distribution of energy states in the forbidden gap of GaAsN0.005 grown by CBE. The main electron and hole traps in GaAsN were recorded by DLTS method and shown in Figure 17 (b) and (c). For electrons traps, two peaks (E1) and (E2) were observed in as grown film around 150 and 300 K, respectively. After rapid thermal annealing, E2 disappears completely and a new electron trap (E3) appears at 200 K. Compared with the electron traps in N-free GaAs, only E1 appears after the introduction of a small amount of N. Therefore, we have strongly suggested that E2 and E3 could be considered as native defects in GaAs. Based on the classification of electron traps in GaAs by Maunitton et al., E2 and E3 were considered to be EL2 (+/0) and EL5, respectively [35]. On the other hand, three hole traps, labeled H0, H2, and H5 were commonly observed in the DLTS spectrum of Figure 17 (c). Compared with the hole traps in N-free GaAs, we could only attribute H5 to the double donor state of EL2 with the charge state (+/++) [36], however, H2 and H0 appear to be newly observed. The hole trap H0 was confirmed to be a radiative recombination center using single and double carrier pulse DLTS method [30]. Since, our interest is only on the N-related defects which could be correlated with the poor electrical properties and the high carrier density in the alloy, we will focus only on the electron trap E1 and the hole trap H2, which were confirmed to be the main cause of the shortening of minority carrier lifetime and the high background doping in GaAsN films,

one GaAs-related peaks in the spectra. The GaAsN-related peaks are in the range of 1.25 – 1.40 eV and 1.1 – 1.2 eV which are caused by BE and DL emissions, respectively. To understand the orientation-dependent photoluminescence of GaAsN films, the ratio of the integrated inten‐ sities of DL to BE (PDL/PBE) were calculated. The PDL/PBE ratio of the film grown on (311)A had the lowest value at any growth temperatures as shown in Figure 16 (b). The film grown on (311)A has almost the same N composition as that of (100). Therefore, N composition is not related to the difference in the integrated intensity ratios. This result indicates that the highindex substrate surfaces contribute to reduce defects and composition fluctuations in the epilayers [24]. On the other hand, in contrast to the continuous increase in total emission intensities of (3 1 1)B sample, a decreasing tendency with a rise in growth temperature was

(a) (b)

**Figure 16.** (a) PL spectra at 4.5K of as-grown films with different substrate orientations. The growth temperature was 440 ℃. (b) Growth-temperature dependences of N composition in GaAsN layers on (3 1 1)A/B and (1 0 0) substrates.

GaInAsN suffers from three main N-related problems: (i) the shortening lifetime of minority carriers, (ii) the lowering of their mobility, and (iii) the increase of carrier concentration with increasing N concentration in the alloy [25-27]. Indeed, the incorporation of a small atomic fraction of N in the lattice of GaAs gives rise to the formation of high density acceptor states, recombination centers, and scattering levels [28]. It is; therefore, clear that the degradation of electrical properties of GaAsN may be attributed only to the formation of lattice defects. These defects were expected to be formed during growth owing to the smaller atomic size of N than of As, as well as, to the large gap miscibility between GaAs and GaN. For that, extensive works

**0.6 0.8 1.0 1.2 1.4 1.6**

**Energy (eV)**

**5. Lattice defects in GaAsN grown by CBE**

**80**

**120**

**PL lifetime [ps]**

**BE**

**DL GaAs**

**160**

**200**

**0.1 1 10 100**

**Intensity ratio of DL/BE**

**(100)**

**420 o C 440 o C 460 o C**

**(311)A**

**(311)B**

recorded for (3 1 1)A.

**PL intensity (arb.unit)**

**(311)A**

**(311)B**

**(100)**

**4.5K**

296 Solar Cells - Research and Application Perspectives

Running Title

respectively. The properties of these two defects are summarized respectively in the two next sub-sections.

19

**Fig. 17.**(left)Distribution of lattice defects in GaAsN grown by CBE for a N concentration of 0.5%, DLTS spectra for (a) as grown and annealed GaAs, (b) as grown GaAsN, (c) annealed **Figure 17.** (left) Distribution of lattice defects in GaAsN grown by CBE for a N concentration of 0.5%, DLTS spectra for (a) as grown and annealed GaAs, (b) as grown GaAsN, (c) annealed GaAsN, (d) Arrhenius plot for electron traps, (e) DLTS spectrum for hole traps in GaAsN, and (f) Arrhenius plot for hole traps.

GaAsN, (d) Arrhenius plot for electron traps, (e) DLTS spectrum for hole traps in GaAsN,

#### **5.1. N-related non radiative recombination center**

and (f) Arrhenius plot for hole traps.

1

The electron trap E1 was confirmed to be related to the N atom, since its density increas‐ es with increasing the N concentration in GaAsN films [31]. Furthermore, it was not observed in N-free GaAs [31]. The activation energy of E1 was found to vary between 0.3 and 0.4 eV below the conduction band minimum (CBM) [31-33]. This variation in energy level was explained by the effect of the electrical field on the thermal emission of carriers from E1, since the carrier concentration in GaAsN was not identical in all studied sam‐ ples [4]. In addition, E1 was not only observed in CBE grown GaAsN but also in the DLTS spectra of MOCVD and MBE grown GaAsN and InGaAsN [37, 38]. Moreover, E1 approx‐ imately exhibited similar density and tendency with varying the N concentration, despite the quite difference in residual carrier densities between the three growth methods [31, 38, 39]. This result implies that the possible structure of E1 is independent of impurities. The thermal capture cross section of E1 is too large compared with that of native defects in GaAs. For that, it was found that the DLTS peak height of E1 saturates promptly with increasing the time of injection of electrons through the filling pulse parameter [33]. In view of the effective role of E1 in the degradation of electrical properties of GaAsN, we were able to provide evidence that E1 trap plays a major role in the degradation of minority carrier lifetime and acts as a recombination center [33]. For this, we carried out an experiment based on minority carrier capture at majority carrier trap by double carrier pulse DLTS method [30]. In this experiment, E1 was filled with majority carriers during the first pulse, so that E1 centers are occupied with electrons. Then, a forward bias minority carrier injection of short duration was applied in order to introduce capture of minority carriers. When the junction returned of quiescent reverse bias, E1 centers that remained occupied with majority carriers were emptied by thermal emission. The amplitude of this thermal emission provides a measure of the number of carriers that were trapped by E1 and not filled with minority carriers at the end of the injection pulse. A decrease of E1 peak height was observed and confirmed with varying the rate of injection through the duration and the voltage of the second pulse [33]. This result, indeed, indicates that E1 is a recombina‐ tion center [33]. After this step, it was essential to identify the nature of this recombina‐ tion mechanism. For that, the temperature dependence of the capture cross section of E1 (σE1( was investigated. As known in DLTS measurements, varying the emission rate window shifts the peak of the defect and the capture cross section may change. From the Arrhe‐ nius plots of each emission rate, we established the relationship between σE1 and the reciprocal temperature [33]. Absolutely, the recombination process through E1 was confirmed to be nonradiative, since σE1 presents a thermal activation energy of 0.13 eV [33]. Furthermore, σE1 was evaluated at room temperature to around 10-13 cm2 [33]. This result proves that E1 is an active recombination center. In addition, this recombination activity may rises with increasing N in the film, since the activation energy of E1, comes closer to the midgap, following the decrease of the conduction band minimum. The recombination mechanism through E1 was quantified by evaluating the lifetime of electrons from the conduction band the energy level of the electron trap in the forbidden gap of the alloy. Using the SRH model for generationrecombination, the lifetime of electrons from the CBM to E1 was calculated to 0.2 ns [33-39]. This value was experimentally verified by timeresolved photoluminescence measurements. We, therefore, suggest that E1 is the main cause of short minority carrier lifetime in GaAsN alloys.

respectively. The properties of these two defects are summarized respectively in the two next

**(a)**

19

**(b)**

**(c)**

sub-sections.

1

Running Title

1.29

298 Solar Cells - Research and Application Perspectives

*Average Thermal Activation Energy***Ea**

0.15

0.48

0.59

0.84

**H5**  *EL2 (+/++)*

**E2** *EL2 (+/0)*

0.99 **E1((N‐As)As)** 

*Main cause of short minority carriers' lifetime: NR recombination* t

*CBM: GaAsN0.00 5, Eg =1.29 eV,* 

**E3** EL5 *type defect*

0.05 **H1**

and (f) Arrhenius plot for hole traps.

**5.1. N-related non radiative recombination center**

*Radiative recombination. center*

*VBM*

DLTS spectrum for hole traps in GaAsN, and (f) Arrhenius plot for hole traps.

**Fig. 17.**(left)Distribution of lattice defects in GaAsN grown by CBE for a N concentration of 0.5%, DLTS spectra for (a) as grown and annealed GaAs, (b) as grown GaAsN, (c) annealed GaAsN, (d) Arrhenius plot for electron traps, (e) DLTS spectrum for hole traps in GaAsN,

**Figure 17.** (left) Distribution of lattice defects in GaAsN grown by CBE for a N concentration of 0.5%, DLTS spectra for (a) as grown and annealed GaAs, (b) as grown GaAsN, (c) annealed GaAsN, (d) Arrhenius plot for electron traps, (e)

The electron trap E1 was confirmed to be related to the N atom, since its density increas‐ es with increasing the N concentration in GaAsN films [31]. Furthermore, it was not

*Principal cause of high background doping*

**H2 (N‐H‐VGa)**

 **(eV)**

**Figure 18.** The split intestinal (N-As)As (left) and N-H-VGa (right) as a possible origins of E1 and H2, respectively.

To understand the formation process of E1 and limit its effect, it was necessary to investigate its origin. First, it was mentioned above that E1 is independent of impurities, such as H, O, and C. Furthermore, E1 was observed in undoped p-type (minority carrier trap), n-type, and Si-doped GaAsN films, which exclude the doping atom from its origin. Therefore, E1 was tentatively suggested to be sensitive only to the host atoms (N, As, and Ga). In addition, using the isothermal capacitance transient method, the density profiling of E1 was found to be quasiuniform distributed GaAsN [31]. This result indicates that the formation of E1 starts and goes on with the growth process. An obvious reason of this behavior is the compensation for the tensile strain, which could be caused by the small atomic size of N compared with that of As. Theoretically, the split interstitials (N-As)As and (N-N)As were expected to form two electron traps at 0.4 and 0.6 eV below the CBM of GaAsN, respectively [40]. These two defects were qualified to be energetically favorable on the growth surface [40]. As the N-dependence of the density of E1 did not show a straight line and since the total N concentration in the film could not be obtained using HRXRD measurements, we have proceed to investigate the relationship between the density of E1 and the flux of the As source. Indeed, we found that the density of E1 shows a peaking behavior with increasing TDMAAs [41]. This result implies that the formation of E1 is sensitive to both N and As atoms. However, this relationship could not be quantified since the atomic fractions between As and N in GaAsN films are quite different [41]. As shown in Figure 18, the split interstitial (N-As)As is, therefore, expected to be the possible origin for E1. However, more experiments are required to clarify the formation process of this recombination center and to limit its density.

### **5.2. N-related acceptor like state in GaAsN**

The high background doping in undoped GaAsN films is a serious problem for solar cell design and fabrication. A small atomic fraction of N changes the conductivity in the film from n-type in GaAs to p-type. Obviously, the formation of N-related acceptors with high density is the main reason of donor compensation. Indeed, two acceptor levels, A1 and A2, were confirmed using Hall Effect measurements [42]. Their thermal activation energies were evaluated to EA1 = 130 ± 20 meV and EA2 = 55 ± 10 meV, respectively. On another hand, the ionized acceptor density NA, obtained using capacitance voltage (C-V) method, showed a linear dependence with N concen‐ tration in undoped films grown under a Ga flow rate of TEGa = 0.1 sccm [32]. This indicates that NA depends strongly on a N-related acceptor [32]. However, this linear dependence tends to saturate when TEGa is reduced, even the N concentration is enhanced [32]. The reason of this saturation was clarified using SIMS measurements, where the concentration of H impurity decreasesslightlywithdecreasingTEGa[32].Hence,thisresultimpliesthatNA dependsstrongly to N and H atomic concentrations in the film. To identify this N-H related acceptor level, we carried out C-V measurements in the temperature range 20 - 330 K in order to get the tempera‐ ture dependence of NA. A sigmoid increase of NA was observed between 70 and 100 K [32]. Its amplitude showed a strong dependence with N concentration [32]. It approximately reflects the relationshipbetweenNA andNconcentrationatroomtemperature[39].Thisresultindicatesthat the sigmoid increase of NA between 70 and 100 K is owing to the thermal ionization of a Nrelated acceptor-like state, which thermal ionization energy can be estimated to be between 0.1 and 0.2 eV. Furthermore, as given in Figure 17 (c), the hole trap H2 was observed in the temper‐ ature range of sigmoid increase of NA. The activation of H2 is within the expected range of the Nrelatedacceptor stateanditis inconformitywiththeoretical calculation[32,44].Furthermorethe density of H2 was found to be in linear dependence with N concentration in the samples grown under TEGa = 0.1 sccm. Hence, H2 is suggested to be the main acceptor state which governs the tendency of NA and it is the cause of the increase of junction capacitance in the temperature range 70 to 100 K. The origin of H2 was investigated using the results of carrier concentration and the density of residual impurities in undoped GaAsN grown by CBE under various Ga flow rates, obtained using Hall Effect, FT-IR, and SIMS [32, 42]. From SIMS measurements, which were carried out on GaAsN grown under different TEGa flow rates, it was found that NA depends on the atomic concentrations of N and H [32, 42]. In addition, using FTIR measurements, a linear relationship was established between the concentration of N-H complexes and N concentra‐ tion [29]. Therefore, H2 was strongly suggested to be an acceptor state and its origin is related to the N-H bond. However, the slope of the linear relationship between NA and the concentration of N-H complexes showed an increase with increasing the growth temperature (TG ∈ [400, 430] C)[29].This indeedindicates thattheN-Hcomplexescouldbeaccompaniedwithanotherdefect, which binding energy was evaluated to 2.5 eV [29]. On another hand, using theoretical calcula‐ tion, the formation energy of (N-H-VGa) -2 was found to be lower than (N-VGa) -3, (H-VGa) -2, and isolated VGa-3 [43]. These results may expect VGa as a possible origin of the unknown intrinsic defect. These predictions were experimentally supported using positron annihilation spectro‐ scopyresults [44].Therefore,H2mayhaveN-H-VGa asapossiblestructureas showninFigure18.

## **6. Conclusion**

**Figure 18.** The split intestinal (N-As)As (left) and N-H-VGa (right) as a possible origins of E1 and H2, respectively.

recombination center and to limit its density.

300 Solar Cells - Research and Application Perspectives

**5.2. N-related acceptor like state in GaAsN**

To understand the formation process of E1 and limit its effect, it was necessary to investigate its origin. First, it was mentioned above that E1 is independent of impurities, such as H, O, and C. Furthermore, E1 was observed in undoped p-type (minority carrier trap), n-type, and Si-doped GaAsN films, which exclude the doping atom from its origin. Therefore, E1 was tentatively suggested to be sensitive only to the host atoms (N, As, and Ga). In addition, using the isothermal capacitance transient method, the density profiling of E1 was found to be quasiuniform distributed GaAsN [31]. This result indicates that the formation of E1 starts and goes on with the growth process. An obvious reason of this behavior is the compensation for the tensile strain, which could be caused by the small atomic size of N compared with that of As. Theoretically, the split interstitials (N-As)As and (N-N)As were expected to form two electron traps at 0.4 and 0.6 eV below the CBM of GaAsN, respectively [40]. These two defects were qualified to be energetically favorable on the growth surface [40]. As the N-dependence of the density of E1 did not show a straight line and since the total N concentration in the film could not be obtained using HRXRD measurements, we have proceed to investigate the relationship between the density of E1 and the flux of the As source. Indeed, we found that the density of E1 shows a peaking behavior with increasing TDMAAs [41]. This result implies that the formation of E1 is sensitive to both N and As atoms. However, this relationship could not be quantified since the atomic fractions between As and N in GaAsN films are quite different [41]. As shown in Figure 18, the split interstitial (N-As)As is, therefore, expected to be the possible origin for E1. However, more experiments are required to clarify the formation process of this

The high background doping in undoped GaAsN films is a serious problem for solar cell design and fabrication. A small atomic fraction of N changes the conductivity in the film from n-type in GaAs to p-type. Obviously, the formation of N-related acceptors with high density is the main reason of donor compensation. Indeed, two acceptor levels, A1 and A2, were confirmed using Hall Effect measurements [42]. Their thermal activation energies were evaluated to EA1 = 130 ± 20 meV and EA2 = 55 ± 10 meV, respectively. On another hand, the ionized acceptor density NA,

InGaAsN is a candidate material to realize the ultrahigh efficiency multi-junction solar cell. The efficiency of the 4-junction solar cell, InGaP/InGaAs/InGaAsN/Ge, is expected to be 41% under 1-sun (AM0) and 52% under 500-sun (AM1.5). In order to achieve the expected super high efficiency, the electrical property of the InGaAsN should be improved. In this paper, we have shown the following results:


## **Author details**

Kazuma Ikeda, Han Xiuxun, Bouzazi Boussairi and Yoshio Ohshita

Toyota Technological Institute, Japan

## **References**

**1.** The CBE method was developed for the growth of GaAsN films. The growth temperature region was divided into three parts. In the middle growth temperature region (390 – 445 °C), the higher N composition without broadening of the XRD rocking curve width was obtained. In this temperature region, a typical characteristics in the growth process is that the amount of the N incorporation into crystal at the surface is not only dominated by the

**2.** Decreasing the growth rate was effective to improve the hole mobility and minoritycarrier lifetime in p-GaAsN films grown by the chemical beam epitaxy (CBE). The mobility

temperature dependence of the mobility, μN, which was inversely proportional to the amount of N-related scattering centers, was estimated. By decreasing the growth rate, μN

was reduced especially when the N composition was higher. The result indicates the improved mobility by decreasing the N-related scattering centers. The minority-carrier lifetime in the bulk of p-GaAsN films, τB, estimated by PL decay curves was improved from 3.2×10-1 ([N] = 0.6%) to 9.0×10-1 ns ([N] = 0.8%) despite the higher N composition. The bulk minority-carrier lifetime τB was mainly determined by the nonradiative recombina‐ tion lifetime (τNR). This means that the nonradiative recombination centers were reduced

V-1s-1 for the N composition of 0.6%. Based on the


growth rate but also the desorption rate of N species at the surface.

by decreasing the growth rate, which resulted in the improvement of τB.

to be the potential substrate to achieve efficient N incorporation.

Kazuma Ikeda, Han Xiuxun, Bouzazi Boussairi and Yoshio Ohshita

**3.** The electron lifetime of p-GaAsN was also improved by controlling the GaAs substrate orientations. The investigation of the effects from GaAs substrate surface orientation and polarity indicates that N incorporation can be enhanced on the (3 1 1)B plane but reduced on (3 1 1)A compared with (1 0 0), where the three dangling bond sites on the (3 1 1) growing surface are considered to play the dominant role. The examination of intensity ratio between emissions from band edge and deep levels evidenced the improved luminescence efficiency of GaAsN layers on (3 1 1) substrates with both polarity. In response to variations in the growth temperature and the postgrowth annealing, optical behavior shows strong dependence on the epitaxial orientation, as illustrated by the relative intensity and low-energy tail of the band-edge emission. (3 1 1)B is distinguished

**4.** The electron trap E1 located at 0.99 eV above VBM is a dominant non radiative recombi‐ nation center. The formation of E1 is sensitive to both the amounts of N and As atoms. A possible structure of this defect is (N-As)As. The hole trap H2 located at 0.15 eV above VBM is suggested to be the main acceptor state. The origin of H2 is related to the N-H bond. A

was increased from 120 to 150 cm2

302 Solar Cells - Research and Application Perspectives

possible structure is N-H-VGa.

Toyota Technological Institute, Japan

**Author details**


[27] Shan W, Walukiewicz W, Ager J, Haller EE, Geisz JF, Friedman DJ, Olson JM, Kurtz SR. Physical Review Letters 1999;82 1221-1224.

[13] Uesugi K, Suemune I. Metalorganic molecular beam epitaxy of GaNAs alloys on (0 0 1)GaAs. Journal of Crystal Growth 1998;189–190 490-495. DOI: 10.1016/

[14] Sun Y, Yamamori M, Egawa T, Ishikawa H. Incorporation of N into GaAsN under N overpressure and underpressure conditions. Japanese Journal of Applied Physics

[15] Ptak AJ, Johnston SW, Kurtz S, Friedman DJ, Metzger WK. A comparison of MBEand MOCVD-grown GaInNAs. Journal of Crystal Growth 2003;251 392-398. DOI:

[16] Matsuura T, Miyamoto T, Makino S, Ohta M, Matsui Y, Koyama F. p-Type Doping Characteristics of GaInNAs:Be Grown by Solid Source Molecular Beam Epitaxy. Jap‐ anese Journal of Applied Physics 2004;43 L433-L435. DOI: 10.1143/JJAP.43.L433.

[17] Fahy S, O'Reilly EP. Intrinsic limits on electron mobility in dilute nitride semiconduc‐

[18] Yamaguchi M, Warabisako T, Sugiura H. Chemical beam epitaxy as a breakthrough technology for photovoltaic solar energy applications. Journal of Crystal Growth

[19] Lee HS, Nishimura K, Yagi Y, Tachibana M, Ekins-Daukes NJ, Ohshita Y, Kojima N, Yamaguchi M. Chemical beam epitaxy of InGaAsN films for multi-junction tandem solar cells. Journal of Crystal Growth 2005; 275 e1127-e1130. DOI: 10.1016/j.jcrysgro.

[20] Tsang WT, From Chemical Vapor Epitaxy to Chemical Beam Epitaxy. Journal of

[21] Uesugi K, Morooka N, Suemune I. Reexamination of N composition dependence of coherently grown GaNAs band gap energy with high-resolution x-ray diffraction mapping measurements. Applied Physics Letters 1999;74 1254-1256. DOI:

[22] Wiley JD., Mobility of Holes in III-V Compounds. In: Willardson RK, Beer AC. (ed.) Semiconductors and Semimetals. New York: Academic Press; 1975;10 p. 91-174.

[23] Nag BR. Electron Transport in Compound Semiconductors. Berlin Heidelberg:

[24] Moto A, Takahashi M, Takahashi S. Effect of Substrate Orientation on Photolumines‐

[25] Buyanova IA., Chens WM. Physics and Applications of Dilute Nitrides, New York:

cence of GaNAs. Japanese Journal of Applied Physics 2000 39;1267-1269.

Crystal Growth 1989;95 121-131. DOI: 10.1016/0022-0248(89)90364-3.

tors. Applied Physics Letters 2003;83 3731-3733. DOI: 10.1063/1.1622444.

S0022-0248(98)00337-6.

304 Solar Cells - Research and Application Perspectives

10.1016/S0022-0248(02)02201-7.

2004.11.200.

10.1063/1.123516.

Springer-Verlag; 1980.

Taylor & Francis Books Inc.; 2004.

[26] Henini M. Dilute Nitrides, Oxford: Elsevier; 2005.

2004;43 2409-2413. DOI: 10.1143/JJAP.43.2409.

1994;136 29-36. DOI: 10.1016/0022-0248(94)90379-4.


## **Solar Cell Efficiency vs. Module Power Output: Simulation of a Solar Cell in a CPV Module**

Egbert Rodríguez Messmer

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/52707

## **1. Introduction**

[40] Zhang SB, Wei SH. Nitrogen Solubility and Induced Defect Complexes in Epitaxial

[41] Bouzazi B, Lee JH, Suzuki H, Kojima N, Ohshita Y, Yamaguchi M. Origin Investiga‐ tion of a Nitrogen-Related Recombination Center in GaAsN Grown by Chemical

[42] Saito K, Nishimura K, Suzuki H, Kojima N, Ohshita Y, Yamaguchi M. Hydrogen re‐ duction in GaAsN thin films by flow rate modulated chemical beam epitaxy. Thin

[43] Janotti A, Wei SH, Zhang SB, S. Kurtz, Van de Walle CG. Interactions between nitro‐ gen, hydrogen, and gallium vacancies in GaAs1-xNx alloys. Physical Review B 2003;67

[44] Toivonen J, Hakkarainen T, Sopanen M, Lipsanen H, Oila J, Saarinen K. Observation of defect complexes containing Ga vacancies in GaAsN. Applied Physics Letters

Beam Epitaxy. Japanese Journal of Applied Physics 2011;50 051001.

Solid Films 2008;516 3517-3520. DOI: 10.1016/j.tsf.2007.08.022.

GaAs:N. Physical Review Letters 2001;86 1789.

306 Solar Cells - Research and Application Perspectives

161201. DOI: 10.1103/PhysRevB.67.161201.

2003;82 40-42. DOI: 10.1063/1.1533843.

In the past few years Concentrating Photovoltaics (CPV) has moved from R&D and pilot projects (typically installations below 500 kilowatts) to multi-megawatt power plants. The starting point of the commercial deployment of this technology was in 2006 when the Insti‐ tute for Concentration Photovoltaic Systems (ISFOC) purchased several CPV power plants of different technologies and provider that were installed in Puertollano (Spain). Each sup‐ plier provided a power plant that had a size between 200kW and 500kW. These power plants are operating since 2008 and the results are very promising [1]. After these first instal‐ lations several MW-size power plants have been installed ([2], [3] and [4]), and the first mul‐ ti-megawatt CPV power plants are under construction, like a 150MW power plant that Soitec is developing for the San Diego Gas & Electric in California [5], demonstrating here with that CPV technology can be a cost efficient alternative to conventional silicon-based flat-plate photovoltaic (PV) plants in areas of high direct solar irradiation.

A CPV module consists typically of a high-efficient solar cell and a concentrator that concen‐ trates light and that can be made out of a mirror, a parabolic dish or lenses. These modules are then mounted on a 2-axis tracking system to make sure that the module is always per‐ pendicular to the sun, so that the light spot reaches the active area of the solar cell. A CPV system is therefore more complex than a conventional PV system, and, in order to be com‐ mercially competitive with standard systems, it is important to control its cost figure. When making a cost analysis of a CPV system, from manufacturing of solar cells to a finished in‐ stallation [6], the cost figure is given in terms of a monetary unit per Watt (€/W or \$/W). This cost figure should be kept as low as possible, and can be done either for a complete installa‐ tion including all the costs relative to the deployment of the system or it can be done partly, considering only the module or the module and tracking system.There are two possibilities

© 2013 Rodríguez Messmer; licensee InTech. This is an open access article 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. © 2013 Rodríguez Messmer; licensee InTech. This is a paper 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.

to reduce the value of this cost figure, which are either reducing the cost of the system, which is typically done reducing the cost of the raw materials or optimizing production processes, or by increasing the output power of the CPV module, which can be achieved by reducing possible sources of losses inside a module (these can be optical, electrical o ther‐ mal).The advantage of increasing the output power of a module is that this has an important impact to other related costs, since also the manufacturing and installation costs are reduced due to the need of fewer modules or even trackers for a CPV power plant of a given size.

The output power of a CPV module can be optimized by reducing the internal losses that appear in the module design. Therefore a good match of the materials from which a module is made should be aimed. The need of a good match is especially true for the interaction be‐ tween the solar cell and the optical system, where the solar cell can be adapted in size, light spectrum, concentration ratio and interface to the optical system. In practice it is very diffi‐ cult to achieve good matching of different materials, since the different parts of a CPV mod‐ ule are made by different manufacturers. These might have different interests as compared to a system developer, so that a compromise is needed. In order to keep the manufacturing cost of the CPV module low, all elements of the module should be standardized, but, on the other hand, CPV module manufacturer want to have components that are customized to their own module design to maximise their output power. Some CPV module manufacturer use parabolic dishes to concentrate light, while others use lenses, and also the concentration ratio at which the modules work differ from manufacturer to manufacturer. This means that from a technological point of view, a CPV module manufacturer desires to have components that are customized to its own system.

To go more into detail in this issue of matching of materials inside a CPV module, and to analyse its effect on module performance, the interface between solar cell and optical system of the module is analysed in this chapter. A solar cell can be designed to have either a maxi‐ mum efficiency when it is measured as a stand-alone device (having air as the surrounding medium) or to have maximum efficiency when it is surrounded in any other optical medi‐ um that is used inside the CPV module (e.g. glass or an optical encapsulant). This fact has an important impact technically and commercially. The solar cell efficiency is usually defined when it is measured in air, and should for commercial purpose be designed to have maxi‐ mum efficiency under these conditions. On the other hand, if the solar cell is going to be op‐ erated embedded in an encapsulant and a lens, the CPV manufacturer should choose a solar cell that is designed for this operating condition, even though the solar cell will have less efficiency when measured at air. In summary, it is important to consider a CPV module as one system during its development, and not composed of independent components that have to be developed independently.

In order to explain better how the embedding medium affects the solar cell performance and to quantify this effect, a series of simulations has been done with a simulation program that has been developed by Isofotón in collaboration with the University of Granada (Spain). This program is called ISOSIM and is able to simulate the performance of a multijunction solar cell, including its anti-reflection coating (ARC) and taking into consideration the con‐ centration and the medium in which the solar cell is used (e.g. air or an optical gel to couple the light from the lens to the solar cell). It is also possible to add optical layers on top of the solar cell structure and simulating thereby a CPV module.

For the sake of simplicity, in this chapter it has been considered a double junction solar cell that is operating at a temperature of 320K (50ºC). The obtained results give a qualitative in‐ dication on the effect of different parameters, but quantitatively the effects that are descri‐ bed in this chapter will exceeded at real operating conditions. On one hand the solar cells that are used are triple-junction cells and also the operating temperature is usually higher. It is estimated that when adapting the solar cell to an optical system instead of using a stand‐ ard solar cell, an increase of output power of up to 10% can be achieved.

## **2. Outline of experimental work**

to reduce the value of this cost figure, which are either reducing the cost of the system, which is typically done reducing the cost of the raw materials or optimizing production processes, or by increasing the output power of the CPV module, which can be achieved by reducing possible sources of losses inside a module (these can be optical, electrical o ther‐ mal).The advantage of increasing the output power of a module is that this has an important impact to other related costs, since also the manufacturing and installation costs are reduced due to the need of fewer modules or even trackers for a CPV power plant of a given size.

The output power of a CPV module can be optimized by reducing the internal losses that appear in the module design. Therefore a good match of the materials from which a module is made should be aimed. The need of a good match is especially true for the interaction be‐ tween the solar cell and the optical system, where the solar cell can be adapted in size, light spectrum, concentration ratio and interface to the optical system. In practice it is very diffi‐ cult to achieve good matching of different materials, since the different parts of a CPV mod‐ ule are made by different manufacturers. These might have different interests as compared to a system developer, so that a compromise is needed. In order to keep the manufacturing cost of the CPV module low, all elements of the module should be standardized, but, on the other hand, CPV module manufacturer want to have components that are customized to their own module design to maximise their output power. Some CPV module manufacturer use parabolic dishes to concentrate light, while others use lenses, and also the concentration ratio at which the modules work differ from manufacturer to manufacturer. This means that from a technological point of view, a CPV module manufacturer desires to have components

To go more into detail in this issue of matching of materials inside a CPV module, and to analyse its effect on module performance, the interface between solar cell and optical system of the module is analysed in this chapter. A solar cell can be designed to have either a maxi‐ mum efficiency when it is measured as a stand-alone device (having air as the surrounding medium) or to have maximum efficiency when it is surrounded in any other optical medi‐ um that is used inside the CPV module (e.g. glass or an optical encapsulant). This fact has an important impact technically and commercially. The solar cell efficiency is usually defined when it is measured in air, and should for commercial purpose be designed to have maxi‐ mum efficiency under these conditions. On the other hand, if the solar cell is going to be op‐ erated embedded in an encapsulant and a lens, the CPV manufacturer should choose a solar cell that is designed for this operating condition, even though the solar cell will have less efficiency when measured at air. In summary, it is important to consider a CPV module as one system during its development, and not composed of independent components that

In order to explain better how the embedding medium affects the solar cell performance and to quantify this effect, a series of simulations has been done with a simulation program that has been developed by Isofotón in collaboration with the University of Granada (Spain). This program is called ISOSIM and is able to simulate the performance of a multijunction solar cell, including its anti-reflection coating (ARC) and taking into consideration the con‐ centration and the medium in which the solar cell is used (e.g. air or an optical gel to couple

that are customized to its own system.

308 Solar Cells - Research and Application Perspectives

have to be developed independently.

In this Chapter, first, the capabilities and performance of the simulation program are ex‐ plained, followed by several studies that show which factors affect the performance of a so‐ lar cell outside and inside of a CPV module.

The first study shows that the performance of a solar cell does not depend only on the mate‐ rial and thickness of the AR- coating layer(s), but also on the refractive index of the sur‐ rounding medium in which this solar cell is measured or operated. A solar cell does not have the same efficiency when it is operated in air (refractive index of 1) or when it is oper‐ ated in a medium of refractive index of e.g. 1.5, (if the solar cell is covered by glass).

Afterwards the performance of the solar cell in air and when it is assembled in a CPV mod‐ ule will be compared. It has been analyzed the current matching of the double-junction solar cell, varying the thickness of the base of the top cell, and it has been identified the limiting subcell (either top-cell or bottom-cell), which is attributed to be due to different spectral losses (absorption or reflection) in the materials and interfaces of the elements of the system.

In the last part a practical example is given, in which the power output of a CPV module is quantified when it is assembled either with a double-junction solar cell that has been opti‐ mized having air as the interface or a solar cell that is optimized to the CPV module. It shows that even if the variation in efficiency of the solar cell is little, the difference in output power can be significant. This chapter summarizes previous work that has been done with the ISOSIM package ([7], [8] and [9]).

## **3. ISOSIM software and experimental parameters**

There are currently several approaches and solutions proposed from module manufacturer and system integrator that develop CPV modules. On the market are currently several tech‐ niques to concentrate the light, since many companies have developed their own propriet‐ ary technological solution. The type of concentrator can be either based on mirrors, dishes, or lenses, and using only primary optics or also secondary and even tertiary optics. In order to maximize the output power of a module, solar cells have to be optimized for each optical system. The problem that arises is that a solar cell that has been optimized for a given opti‐ cal system does not necessarily have an optimum performance also in another optical sys‐ tem. It is also difficult for a solar cell manufacturer to predict how a solar cell will perform inside a given module type. To take into consideration various types of concentrator tech‐ nology, and to be able to analyze also a stand-alone solar cell, the simulation software is or‐ ganized in layers, in which each layer represents a material with its own material properties and function. In this manner it is possible to simulate any type of CPV module and at any operating condition (temperature, concentration ratio).

The software used for the simulations has been specifically designed for the analysis of mul‐ tijunction solar cells in order to get a tool to aid the design of solar cells and concentrator PV systems. This software, that is called *ISOSIM*, is capable of modelling the performance from stand-alone single junction solar cells up to the performance of multijunction solar cells in‐ side a CPV module under real operating conditions. The simulation program solves the Poisson and continuity equations by using a procedure optimized for multilayer structures. It includes the radiative interband, Shockley-Read-Hall and Auger recombination mecha‐ nisms, and computes the generation function of electron-hole pairs from the optical parame‐ ters of the cell materials. The dependence of these optical parameters on the photon energy has been included, taking into account the doping level and its effect on bandgap narrow‐ ing. The software uses the Rakic model for the calculation of the complex dielectric function, absorption coefficient, extinction and refraction index calculation, with the Gaussian broad‐ ening proposed by Kim. Additionally, several effects are included in the software, such as indirect transition contributions, the shift in the optical band gap due to doping, and freecarrier absorption, among others. The material parameters of several anti-reflection coatings were obtained either empirically by ellipsometry measurements or extracting them from the SOPRA database of refractive indexes [10].

The program also takes into consideration the optical medium that surrounds the solar cell, which can be either air which represents the case of a stand-alone solar cell that is measured, or any other medium, representing the case of a solar cell that is mounted in a CPV module (e.g. epoxies, solar glasses, optical gel). Apart of that, also the illumination spectra (space, terrestrial or any customized spectra), the concentration of the light source and the tempera‐ ture of the solar cell can be modified for simulations. It is therefore possible to simulate the whole system, considering solar cell, lenses, optical gels that are used to couple optical com‐ ponents, possible air gaps between lenses inside the module, concentration ratio and operat‐ ing temperature. It is therefore possible to estimate the total loss of a system if e.g. a wrong choice of materials or of its process parameters has been chosen for the manufacturing of a CPV module.

For the simulations in this chapter, I-V curves and spectral response of a stand-alone solar cell are obtained by simulating a typical solar cell structure. If not otherwise specified, the structure used for the studies in this chapter (Figure 1.) see is a dual-junction GaInP/GaAs solar cell with the two photovoltaic junctions connected by a GaAs tunnel junction, and the antireflection coating consisting of a double layer of TiO2 and Al2O3 [11]. Light concentration and operating temperature were fixed at 1000 suns and 320 K, respectively. It has been used the AM1.5D spectrum and has also been assumed a shadowing loss of the solar cell of 2% due to the area below the grid. Some of the parameters of the materials that are used for simulations show high dispersion. For instance, the band-gap of GaInP depends on the or‐ dering level, and could vary from 1.66 eV for a fully ordered lattice to 2.01 eV for a totally disordered one. For the simulations presented in this article, we assumed that the band-gap of GaInP is 1.9 eV at 300K. In a previous paper in which we presented this software, it has been shown that the results of the simulations obtained by the ISOSIM software match rea‐ sonably well with experimentally observed results [7].

to maximize the output power of a module, solar cells have to be optimized for each optical system. The problem that arises is that a solar cell that has been optimized for a given opti‐ cal system does not necessarily have an optimum performance also in another optical sys‐ tem. It is also difficult for a solar cell manufacturer to predict how a solar cell will perform inside a given module type. To take into consideration various types of concentrator tech‐ nology, and to be able to analyze also a stand-alone solar cell, the simulation software is or‐ ganized in layers, in which each layer represents a material with its own material properties and function. In this manner it is possible to simulate any type of CPV module and at any

The software used for the simulations has been specifically designed for the analysis of mul‐ tijunction solar cells in order to get a tool to aid the design of solar cells and concentrator PV systems. This software, that is called *ISOSIM*, is capable of modelling the performance from stand-alone single junction solar cells up to the performance of multijunction solar cells in‐ side a CPV module under real operating conditions. The simulation program solves the Poisson and continuity equations by using a procedure optimized for multilayer structures. It includes the radiative interband, Shockley-Read-Hall and Auger recombination mecha‐ nisms, and computes the generation function of electron-hole pairs from the optical parame‐ ters of the cell materials. The dependence of these optical parameters on the photon energy has been included, taking into account the doping level and its effect on bandgap narrow‐ ing. The software uses the Rakic model for the calculation of the complex dielectric function, absorption coefficient, extinction and refraction index calculation, with the Gaussian broad‐ ening proposed by Kim. Additionally, several effects are included in the software, such as indirect transition contributions, the shift in the optical band gap due to doping, and freecarrier absorption, among others. The material parameters of several anti-reflection coatings were obtained either empirically by ellipsometry measurements or extracting them from the

The program also takes into consideration the optical medium that surrounds the solar cell, which can be either air which represents the case of a stand-alone solar cell that is measured, or any other medium, representing the case of a solar cell that is mounted in a CPV module (e.g. epoxies, solar glasses, optical gel). Apart of that, also the illumination spectra (space, terrestrial or any customized spectra), the concentration of the light source and the tempera‐ ture of the solar cell can be modified for simulations. It is therefore possible to simulate the whole system, considering solar cell, lenses, optical gels that are used to couple optical com‐ ponents, possible air gaps between lenses inside the module, concentration ratio and operat‐ ing temperature. It is therefore possible to estimate the total loss of a system if e.g. a wrong choice of materials or of its process parameters has been chosen for the manufacturing of a

For the simulations in this chapter, I-V curves and spectral response of a stand-alone solar cell are obtained by simulating a typical solar cell structure. If not otherwise specified, the structure used for the studies in this chapter (Figure 1.) see is a dual-junction GaInP/GaAs solar cell with the two photovoltaic junctions connected by a GaAs tunnel junction, and the antireflection coating consisting of a double layer of TiO2 and Al2O3 [11]. Light concentration

operating condition (temperature, concentration ratio).

310 Solar Cells - Research and Application Perspectives

SOPRA database of refractive indexes [10].

CPV module.


**Figure 1.** Structure of the double junction solar cell used for the analysis in this chapter. The layers in light blue are the ones corresponding to the top-cell, the grey ones to the tunnel junction and the green ones to the bottom cell [12].


**Figure 2.** Schematic representation of the layer sequence that light has to pass through until it reaches the solar cell. It represents a CPV module that is made out of a primary and secondary lens.

The structure of the ISOSIM simulation program makes it possible to add several layers and their parameters (thickness, doping level and gradient, doping type, composition, recombi‐ nation velocity, recombination time constant, refractive index). It is also needed to specify the function of each layer, like e.g. *optical layer* for the window layer of the solar cell or the layers that represent the optical system (e.g. lenses, encapsulating material, air gaps between materials, cover glass). For running a simulation, the parameters that are specified are con‐ centration ratio (in suns), solar cell operating temperature, shadowing loss, series resistance of the contact layer and refractive index of the medium that surrounds the solar (e.g. n=1 if the simulation should be done considering that the solar cell is measured in air). This high degree in flexibility makes it possible to simulate almost any type of multijuncion solar cells and any type of CPV module.

For the simulations presented in this chapter, three different type of simulations were made, a solar cell that is in air (refractive index is 1), a solar cell that is surrounded by a medium with refractive index equal to 1.5 (e.g. glass) and a solar cell that is mounted in a CPV mod‐ ule that is made of a cover glass, primary and secondary lenses, encapsulating materials, and an air gap between primary and secondary lens. A schematic description is shown in Figure 2.

In order to show the usefulness of the simulation software for the prediction of the behavior of solar cells under real operating conditions (i.e. under light concentration and at tempera‐ tures higher than room temperature), several figures are shown (Figure 3. to Figure 7.). These simulations can be of interest to extrapolate the measured parameter at room temper‐ ature to real operating conditions. The two parameters that affect mostly the solar cell per‐ formance when it is mounted in a CPV module are the operating temperature and the optical coupling of the sunlight to the solar cell. Even if the short-circuit current Isc increases slightly with increasing temperature (Figure 3.), the predominant effect is the decreases of the open-circuit voltage Voc (Figure 4.) and the fill-factor FF (Figure 5.), reducing the solar cell efficiency (Figure 6.) and thereby also the output power of a module. Figure 7 shows the efficiency of the solar cell when the concentration is increased. The temperature is set to 320K. It shows the typical shape of this type of curves, although the maximum in efficiency is in our simulation at a concentration of around 2000 suns, which is higher than typical val‐ ues observed experimentally [13]. More details about the performance of the ISOSIM soft‐ ware, as well as the discussion of the figures shown here can be found elsewhere [7].

**Figure 2.** Schematic representation of the layer sequence that light has to pass through until it reaches the solar cell. It

The structure of the ISOSIM simulation program makes it possible to add several layers and their parameters (thickness, doping level and gradient, doping type, composition, recombi‐ nation velocity, recombination time constant, refractive index). It is also needed to specify the function of each layer, like e.g. *optical layer* for the window layer of the solar cell or the layers that represent the optical system (e.g. lenses, encapsulating material, air gaps between materials, cover glass). For running a simulation, the parameters that are specified are con‐ centration ratio (in suns), solar cell operating temperature, shadowing loss, series resistance of the contact layer and refractive index of the medium that surrounds the solar (e.g. n=1 if the simulation should be done considering that the solar cell is measured in air). This high degree in flexibility makes it possible to simulate almost any type of multijuncion solar cells

For the simulations presented in this chapter, three different type of simulations were made, a solar cell that is in air (refractive index is 1), a solar cell that is surrounded by a medium with refractive index equal to 1.5 (e.g. glass) and a solar cell that is mounted in a CPV mod‐ ule that is made of a cover glass, primary and secondary lenses, encapsulating materials, and an air gap between primary and secondary lens. A schematic description is shown in

In order to show the usefulness of the simulation software for the prediction of the behavior of solar cells under real operating conditions (i.e. under light concentration and at tempera‐ tures higher than room temperature), several figures are shown (Figure 3. to Figure 7.). These simulations can be of interest to extrapolate the measured parameter at room temper‐ ature to real operating conditions. The two parameters that affect mostly the solar cell per‐ formance when it is mounted in a CPV module are the operating temperature and the optical coupling of the sunlight to the solar cell. Even if the short-circuit current Isc increases slightly with increasing temperature (Figure 3.), the predominant effect is the decreases of

represents a CPV module that is made out of a primary and secondary lens.

and any type of CPV module.

312 Solar Cells - Research and Application Perspectives

Figure 2.

**Figure 3.** Simulation of the short-circuit current density (Isc) of a double-junction solar cell as a function of temperature at a concentration of 1000 suns.

**Figure 4.** Simulation of the open-circuit Voltage (Voc) of a double-junction solar cell as a function of temperature at a concentration of 1000 suns.

**Figure 5.** Simulation of the fill factor (FF) of a double-junction solar cell as a function of temperature at a concentra‐ tion of 1000 suns.

Solar Cell Efficiency vs. Module Power Output: Simulation of a Solar Cell in a CPV Module http://dx.doi.org/10.5772/52707 315

**Figure 6.** Simulation of the efficiency of a double-junction solar cell as a function of temperature at a concentration of 1000 suns.

**Figure 4.** Simulation of the open-circuit Voltage (Voc) of a double-junction solar cell as a function of temperature at a

**Figure 5.** Simulation of the fill factor (FF) of a double-junction solar cell as a function of temperature at a concentra‐

concentration of 1000 suns.

314 Solar Cells - Research and Application Perspectives

tion of 1000 suns.

**Figure 7.** Simulation of the solar cell efficiency of a double-junction solar cell as a function of light concentration at a temperature of 320 K.

## **4. Effect of the surrounding medium on solar cell efficiency**

There are currently several techniques used to concentrate light inside a CPV module. They can be made of primary and secondary lenses, as in the case of the examples given in this chapter, of only a primary lens or a (parabolic) mirror. To avoid losses due to changes in the refraction in‐ dex, encapsulating materials might be used when using a secondary lens. The refraction index of this type of gels can be chosen to achieve the highest efficiency in the module. In all cases sun‐ light has to be properly guided and matched from air through several materials of different re‐ fractive index to the solar cell until if finally reaches each single junction of a multijunction solar cell, where the light is recombined. This leads to the question on how much is the output power of the CPV module affected due to the optical system. To analyze this issue, several simulations are made on a double-junction solar cell as the one described earlier.

The obtained results can be extrapolated to a triple-junction solar cell, at least qualitatively, since the third junction (typically made out of a germanium p-n junction) generates much more current than either the 1st or 2nd junction, and it does therefore not limit the multijunc‐ tion solar cell. Figure 8 shows the simulated I-V curves of a triple-junction solar cell and all of its subcells, the bottom cell (typically a Ge cell, red curve), the middle cell (typically an InGaAs cell, green curve), the top cell (typical an InGaP cell, blue curve) and the resulting I-V curve of the triple junction solar cell (grey curve). This figure shows a triple-junction cell in which the top cell is the limiting subcell and also that the Ge junction generates much more current than any of the other two.

**Figure 8.** I-V curves of a triple-junction solar cell. There are given the three I-V curves of the subcells (top-cell, middlecell and bottom cell) and the resulting I-V curve of the resulting 3J cell.

The efficiency of a double-junction solar cell with two different ARC layers, either a double layer of TiO2/Al2O3 or a single layer of Si3N4 is analysed. The refractive index of the sur‐ rounding medium has been increased from 1 (air) to 1.5 (e.g. glass). The result of the simu‐ lated efficiency is shown in Figure 9 for a given thickness of the ARC layer. If the solar cell would be operated in air (n=1) the best choice would be to use a TiO2/Al2O3 double layer and the solar cell would have an efficiency of 31.14%. But, if the solar cell would be operated in a medium with a refractive index higher than 1, e.g. 1.5, a Si3N4 single layer would be a better choice, yielding an efficiency of 31.87%. It can also be observed that if the solar cell has been optimized for being measured in air (ARC double layer of TiO2/Al2O3), the efficiency of the system is reduced by around 1% when it is being operated in a medium with a refractive index of 1.5. On the other hand, a solar cell that has an ARC layer that is optimized to a me‐ dium with n=1.5, has a lower efficiency when measured at air. In the case that the solar cell is operated in a medium of n=1.5 and a wrong antireflection coating has been chosen (in this example TiO2/Al2O3 instead of Si3N4), the solar cell would have an efficiency of 30.16% in‐ stead of 31.87%, having a loss in efficiency of 1.71%. It should be remembered that these val‐ ues depend very much on the material and thickness of the ARC layer, and each system configuration has to be analysed separately.

**4. Effect of the surrounding medium on solar cell efficiency**

are made on a double-junction solar cell as the one described earlier.

more current than any of the other two.

316 Solar Cells - Research and Application Perspectives

There are currently several techniques used to concentrate light inside a CPV module. They can be made of primary and secondary lenses, as in the case of the examples given in this chapter, of only a primary lens or a (parabolic) mirror. To avoid losses due to changes in the refraction in‐ dex, encapsulating materials might be used when using a secondary lens. The refraction index of this type of gels can be chosen to achieve the highest efficiency in the module. In all cases sun‐ light has to be properly guided and matched from air through several materials of different re‐ fractive index to the solar cell until if finally reaches each single junction of a multijunction solar cell, where the light is recombined. This leads to the question on how much is the output power of the CPV module affected due to the optical system. To analyze this issue, several simulations

The obtained results can be extrapolated to a triple-junction solar cell, at least qualitatively, since the third junction (typically made out of a germanium p-n junction) generates much more current than either the 1st or 2nd junction, and it does therefore not limit the multijunc‐ tion solar cell. Figure 8 shows the simulated I-V curves of a triple-junction solar cell and all of its subcells, the bottom cell (typically a Ge cell, red curve), the middle cell (typically an InGaAs cell, green curve), the top cell (typical an InGaP cell, blue curve) and the resulting I-V curve of the triple junction solar cell (grey curve). This figure shows a triple-junction cell in which the top cell is the limiting subcell and also that the Ge junction generates much

**Figure 8.** I-V curves of a triple-junction solar cell. There are given the three I-V curves of the subcells (top-cell, middle-

cell and bottom cell) and the resulting I-V curve of the resulting 3J cell.

**Figure 9.** Solar cell efficiency as a function of the refractive index of the surrounding medium, for two alternatives of anti-reflection coatings.

This is a simplified example of how important the condition in which the solar cell operates is. If the solar cell would be encapsulated in a medium with a refractive index of 1.5, the sit‐ uation would be as described above. In reality CPV modules are in general much more com‐ plex, and even more layers might interfere with the sun rays on its way to the solar cell. This means, on one hand, that if the solar cell is optimized to be operating in air, the expected decrease in efficiency when operating in the CPV module might be even higher than in the previous simulations. On the other hand, if the complexity of the module increases, it is also much more difficult to optimize a solar cell to be operating in that specific CPV module. In any case, according to these simulations, it is strongly recommended not to develop sepa‐ rately the solar cell and the optics of the CPV module, rather than optimizing the system as one unit.

## **5. Effect of AR coating material on solar cell efficiency**

Anti-reflection coatings are needed to match the refractive index of the solar cell window layer to the surrounding medium, increasing thereby the amount of light that penetrates in‐ to the photovoltaic structure [11]. These must be carefully designed to have the highest transmission and the broadest bandwidth. The tuning of these could yield by multijunction solar cells to different limiting cell configurations, since different spectral parts of the sun‐ light are absorbed or reflected differently during operation. For this reason, our software is capable to handle spectral responses of complete systems and with different irradiation spectra.

In this section the design of the AR coating is analysed. With a correct design of the AR coat‐ ing, the solar cell can be adapted to the surrounding medium, reducing losses due to un‐ wanted reflections. It is possible to increase the final efficiency of the cell, increasing it up to around 30% of its initial value in a cell/air interface. As this medium is fully dependent on the system design (epoxy, air, lens, potting, etc.), the AR coating should be designed taking into consideration the design of the CPV module. We performed some simulations with sev‐ eral materials that are typically used as AR coating in optoelectronic devices. These are alu‐ minum oxide (Al2O3), titanium dioxide (TiO2), silicon nitride (Si3N4), and magnesium difluoride (MgF2). The result for a double-junction solar cell is shown in Figure 10.

This figure shows that there is a strong dependence of the efficiency of a solar cell on the material of the antireflection coating and on its thickness. The efficiency can therefore be maximised e.g. when it is measured in air. Whenever we change the medium from air to a different higher refractive index medium (glass, epoxy, lens, potting, etc.), results vary. Fig‐ ure 11 shows simulations for the same solar cell as in Figure 10 but considering that the solar cell is mounted in a CPV module. This means that the optical system has a big influence in modifying the spectral performance of the CPV system.

Comparing Figure 10 and Figure 11 we see that the behaviour of the efficiency against AR coating thickness differ if the solar cell is either measured on air or mounted inside a CPV module. It is observed e.g. that an efficiency of 29% of a standalone solar cell with an ap‐ proximately 80 nm thick MgF2 AR coating layer, has only 24% efficiency when this solar cell is mounted in the CPV module.

Solar Cell Efficiency vs. Module Power Output: Simulation of a Solar Cell in a CPV Module http://dx.doi.org/10.5772/52707 319

**Figure 10.** Stand-alone solar cell efficiency vs. AR coating thickness for several materials.

plex, and even more layers might interfere with the sun rays on its way to the solar cell. This means, on one hand, that if the solar cell is optimized to be operating in air, the expected decrease in efficiency when operating in the CPV module might be even higher than in the previous simulations. On the other hand, if the complexity of the module increases, it is also much more difficult to optimize a solar cell to be operating in that specific CPV module. In any case, according to these simulations, it is strongly recommended not to develop sepa‐ rately the solar cell and the optics of the CPV module, rather than optimizing the system as

Anti-reflection coatings are needed to match the refractive index of the solar cell window layer to the surrounding medium, increasing thereby the amount of light that penetrates in‐ to the photovoltaic structure [11]. These must be carefully designed to have the highest transmission and the broadest bandwidth. The tuning of these could yield by multijunction solar cells to different limiting cell configurations, since different spectral parts of the sun‐ light are absorbed or reflected differently during operation. For this reason, our software is capable to handle spectral responses of complete systems and with different irradiation

In this section the design of the AR coating is analysed. With a correct design of the AR coat‐ ing, the solar cell can be adapted to the surrounding medium, reducing losses due to un‐ wanted reflections. It is possible to increase the final efficiency of the cell, increasing it up to around 30% of its initial value in a cell/air interface. As this medium is fully dependent on the system design (epoxy, air, lens, potting, etc.), the AR coating should be designed taking into consideration the design of the CPV module. We performed some simulations with sev‐ eral materials that are typically used as AR coating in optoelectronic devices. These are alu‐ minum oxide (Al2O3), titanium dioxide (TiO2), silicon nitride (Si3N4), and magnesium

This figure shows that there is a strong dependence of the efficiency of a solar cell on the material of the antireflection coating and on its thickness. The efficiency can therefore be maximised e.g. when it is measured in air. Whenever we change the medium from air to a different higher refractive index medium (glass, epoxy, lens, potting, etc.), results vary. Fig‐ ure 11 shows simulations for the same solar cell as in Figure 10 but considering that the solar cell is mounted in a CPV module. This means that the optical system has a big influence in

Comparing Figure 10 and Figure 11 we see that the behaviour of the efficiency against AR coating thickness differ if the solar cell is either measured on air or mounted inside a CPV module. It is observed e.g. that an efficiency of 29% of a standalone solar cell with an ap‐ proximately 80 nm thick MgF2 AR coating layer, has only 24% efficiency when this solar cell

difluoride (MgF2). The result for a double-junction solar cell is shown in Figure 10.

modifying the spectral performance of the CPV system.

is mounted in the CPV module.

**5. Effect of AR coating material on solar cell efficiency**

one unit.

318 Solar Cells - Research and Application Perspectives

spectra.

**Figure 11.** CPV system efficiency vs. AR coating thickness for several materials, using the same cell structure as for the plots in Figure 10.

## **6. Current matching: Stand-alone solar cell vs. performance in CPV module**

In this section the spectral loss of the optical system is analysed. Several simulations were made starting from a double-junction solar cell in which the thickness of the base of the top cell is increased from 500 nm to 950 nm for the case of a stand-alone solar cell, and from 500nm to 1150nm for a solar cell that is assembled in a module. The obtained results are again specific to each configuration of solar cell and type and materials of a CPV module. Nevertheless, the general tendency of the results that is observed (i.e. the switching from a top-cell limiting configuration to a bottom-cell limiting configuration with increasing topcell base thickness) is generally valid.

Figure 12 shows the behavior of the efficiency when increasing the thickness of the base of the top-cell of a double-junction solar cell. It can be seen that when increasing the thickness from 500nm until approximately 700nm, also the efficiency increases. This shows (and is confirmed by the plots of the I-V curves not shown here) that the top-cell of the double-junc‐ tion solar cell is limiting. Increasing the thickness of the base further, from approximately 800nm to 950nm the efficiency decreases, and the bottom-cell is the limiting one. The reason for the drop in efficiency is on one hand that the series resistance of the top-cell increases, and that there is more light absorbed in the top-cell and therefore less light available for cur‐ rent generation in the bottom-cell.

**Figure 12.** Simulation of the efficiency of a stand-alone double-junction solar cell as a function of top cell base thick‐ ness.

**6. Current matching: Stand-alone solar cell vs. performance in CPV**

In this section the spectral loss of the optical system is analysed. Several simulations were made starting from a double-junction solar cell in which the thickness of the base of the top cell is increased from 500 nm to 950 nm for the case of a stand-alone solar cell, and from 500nm to 1150nm for a solar cell that is assembled in a module. The obtained results are again specific to each configuration of solar cell and type and materials of a CPV module. Nevertheless, the general tendency of the results that is observed (i.e. the switching from a top-cell limiting configuration to a bottom-cell limiting configuration with increasing top-

Figure 12 shows the behavior of the efficiency when increasing the thickness of the base of the top-cell of a double-junction solar cell. It can be seen that when increasing the thickness from 500nm until approximately 700nm, also the efficiency increases. This shows (and is confirmed by the plots of the I-V curves not shown here) that the top-cell of the double-junc‐ tion solar cell is limiting. Increasing the thickness of the base further, from approximately 800nm to 950nm the efficiency decreases, and the bottom-cell is the limiting one. The reason for the drop in efficiency is on one hand that the series resistance of the top-cell increases, and that there is more light absorbed in the top-cell and therefore less light available for cur‐

**Figure 12.** Simulation of the efficiency of a stand-alone double-junction solar cell as a function of top cell base thick‐

**module**

ness.

cell base thickness) is generally valid.

320 Solar Cells - Research and Application Perspectives

rent generation in the bottom-cell.

**Figure 13.** Simulation of the efficiency of a double-junction solar cell that is mounted inside a CPV module as a func‐ tion of top cell base thickness.

The same series of simulations have been done for a solar cell mounted in a double lens CPV system like the one shown in Figure 2 and the result is plotted in Figure 13. Even though the general tendency of switching from a top-cell limiting configuration to a bottom-cell limiting configuration is true in both examples, the thickness at which this switching occurs is differ‐ ent, and occurs at higher thickness when the solar cell is mounted inside a module. This fig‐ ure shows that when embedding the cell in the CPV system the current matching for the individual solar cell in a dual junction solar cell moves towards around a 1000 nm-thick topcell base. This can be explained by the fact that part of the sun light gets either absorbed by the materials of the system or get reflected on the interfaces of them, so that less light is available for generating current in the solar cell. It also follows that the efficiency of the solar cell and system is lower than that of only the solar cell.

Continuing with the analysis of the system as a whole, it can be seen that the efficiency of this system is in all the range lower than that of a stand-alone solar cell. As an example two solar cells with different width of the top-cell base will be compared, one of 750 nm and the other one of 950 nm. If the final design of a solar cell is with a 750 nm thick top cell base, its efficiency is 31.21%, and if the final design is with a 950 nm thick top cell base, its efficiency will be 30.82%. If these two solar cells are mounted in the system, the first solar cell will have an efficiency of 27.09%, and the latter one an efficiency of 27.55%. This means that the solar cell with a top-cell base thickness of 750 nm performs better when it is operated in air (i.e. 0.39% higher), whereas the solar cell with a top cell base thickness of 950 nm performs better when it is mounted in the module. According to these simulations, if the thickness of the base of the top-cell is increased from 650 nm to 950 nm in a solar cell that is going to be operated in the optical system described earlier, the overall system efficiency would be in‐ creased by 0.46%. The very important conclusion of this analysis is, once more, that the pri‐ ority should be to optimize the module efficiency and not only the solar cell efficiency.

The previous simulations can be complemented with simulations of the spectral response of a solar cell. In Figure 14 is plotted the spectral response of the solar cell for a stand-alone solar cell of a 650 nm and a 950 nm thick top-cell base. This figure is similar to the ones measured on real devices, like e.g. in [14]. Figure 15 shows the same plot for a solar cell that is mounted in a CPV system. When increasing the width of the top-cell base, the generated current in the top cell increases (blue curve). This behaviour is also true when the solar cell is inserted in a module, even though it is less enhanced.

**Figure 14.** Simulation of the spectral response of a double-junction solar cell for two different TC base thicknesses (650 nm and 950 nm).

the base of the top-cell is increased from 650 nm to 950 nm in a solar cell that is going to be operated in the optical system described earlier, the overall system efficiency would be in‐ creased by 0.46%. The very important conclusion of this analysis is, once more, that the pri‐ ority should be to optimize the module efficiency and not only the solar cell efficiency.

The previous simulations can be complemented with simulations of the spectral response of a solar cell. In Figure 14 is plotted the spectral response of the solar cell for a stand-alone solar cell of a 650 nm and a 950 nm thick top-cell base. This figure is similar to the ones measured on real devices, like e.g. in [14]. Figure 15 shows the same plot for a solar cell that is mounted in a CPV system. When increasing the width of the top-cell base, the generated current in the top cell increases (blue curve). This behaviour is also true when the solar cell

**Figure 14.** Simulation of the spectral response of a double-junction solar cell for two different TC base thicknesses

(650 nm and 950 nm).

is inserted in a module, even though it is less enhanced.

322 Solar Cells - Research and Application Perspectives

**Figure 15.** Simulation of the spectral response of a double-junction solar cell mounted inside a CPV module. The simu‐ lations are made for two different top cell base thicknesses (650 nm and 950 nm).

## **7. Power loss in a CPV module due to a non-matched solar cell**

It has been shown in this chapter with the help of a simulation program that the performance of a solar cell depends on the medium in which it operates. A solar cell that is measured in air as a standalone device does not perform in the same way as in a system that is composed of solar cell and optical concentrator system. The system efficiency is typically lower than the efficien‐ cy of a standalone solar cell, due to losses that are attributed to (spectral) absorption of the ma‐ terials and reflections on the interfaces of different type of materials. These losses between a stand-alone solar cell and the efficiency of the system can be minimized, if the right materials are chosen and the solar cell and the optical system are adapted to each other.

In order to get a feeling on how big the impact of the efficiency loss is on the module power out‐ put, simple calculations are given. It has been shown in previous sections that due to a wrong choice of anti-reflection coating the loss in efficiency of the CPV system can be 1,71%, and that if the thickness of the base of the top cell is not adapted to the system, it is possible to lose another 0,46% in efficiency. This means that by adapting the ARC and the thickness of the base of the so‐ lar cell, it is possible to gain (or not to lose) more than 2% in system efficiency, according to the simulations of the system and parameters described earlier. The simulations were made for a double-junction solar cell, and it can be expected that the loss of 2% that is observed for a nonoptimized solar cell will be even higher for a system with a triple-junction solar cell.

As an example, it is assumed that the optimized system efficiency of a CPV module is 30%, whereas the efficiency of the not optimized CPV module is 28%. Assuming that a given CPV module has 10 solar cells of 1 cm2 of active area each and works at a concentration of 500 suns, then the output power would be 127,5 W for 30% system efficiency and 119 W for 28% system efficiency, assuming an irradiation of 850 W/m2 . With this simple example it is in‐ tended to visualize that a seemingly small difference of 2% in system efficiency has an effect of approx. 6,7% on module level and therefore also on the installation cost of a CPV power plant. The effect that a 2% efficiency difference makes on module level depends on its de‐ sign (number of solar cells, concentration ratio, etc.), and should be therefore only consid‐ ered qualitatively.

## **8. Conclusion**

In this chapter simulations of double-junction solar cells were discussed. The ISOSIM simu‐ lation software is a powerful tool to obtain characteristic curves of solar cells. With this soft‐ ware package it is also possible to understand and predict experimental behaviour of solar cells under real operating conditions. The results obtained in this chapter can be extrapolat‐ ed to triple-junction solar cells, since typically the third junction is made out of Germanium and is far from limiting the multijunciton solar cell. It has been analysed the efficiency of a solar cell and a system composed of solar cell and additional layers that should represent the optical system of a CPV module. It has been analysed the effect of the anti-reflection coating material and its thickness on the efficiency of the solar cell, and has also been shown that when increasing the width of the base to the top cell, the double-junction solar cell switches from being top cell limiting to bottom cell limiting. These two parameters, the type and thickness of the ARC layer and the thickness of the top-cell base thickness play an im‐ portant role on the efficiency of the solar cell. When the solar cell is assembled in a CPV module the optimum values of ARC coating material and its thickness and the optimum thickness of the base layer of the top cell changes. This means that in order to obtain maxi‐ mum module power output, a solar cell and optical system should match each other well, in a way that the design of the solar cell should take into account the optical system of the CPV module or the other way around, the design of the optical system should be adapted to a given solar cell. It is also shown that a small variation in efficiency of a solar cell has a big impact on CPV module power output and therefore also on the installation cost of a CPV power plant.

## **Author details**

Egbert Rodríguez Messmer

Isofotón S.A, Spain

## **References**

As an example, it is assumed that the optimized system efficiency of a CPV module is 30%, whereas the efficiency of the not optimized CPV module is 28%. Assuming that a given CPV

suns, then the output power would be 127,5 W for 30% system efficiency and 119 W for 28%

tended to visualize that a seemingly small difference of 2% in system efficiency has an effect of approx. 6,7% on module level and therefore also on the installation cost of a CPV power plant. The effect that a 2% efficiency difference makes on module level depends on its de‐ sign (number of solar cells, concentration ratio, etc.), and should be therefore only consid‐

In this chapter simulations of double-junction solar cells were discussed. The ISOSIM simu‐ lation software is a powerful tool to obtain characteristic curves of solar cells. With this soft‐ ware package it is also possible to understand and predict experimental behaviour of solar cells under real operating conditions. The results obtained in this chapter can be extrapolat‐ ed to triple-junction solar cells, since typically the third junction is made out of Germanium and is far from limiting the multijunciton solar cell. It has been analysed the efficiency of a solar cell and a system composed of solar cell and additional layers that should represent the optical system of a CPV module. It has been analysed the effect of the anti-reflection coating material and its thickness on the efficiency of the solar cell, and has also been shown that when increasing the width of the base to the top cell, the double-junction solar cell switches from being top cell limiting to bottom cell limiting. These two parameters, the type and thickness of the ARC layer and the thickness of the top-cell base thickness play an im‐ portant role on the efficiency of the solar cell. When the solar cell is assembled in a CPV module the optimum values of ARC coating material and its thickness and the optimum thickness of the base layer of the top cell changes. This means that in order to obtain maxi‐ mum module power output, a solar cell and optical system should match each other well, in a way that the design of the solar cell should take into account the optical system of the CPV module or the other way around, the design of the optical system should be adapted to a given solar cell. It is also shown that a small variation in efficiency of a solar cell has a big impact on CPV module power output and therefore also on the installation cost of a CPV

of active area each and works at a concentration of 500

. With this simple example it is in‐

module has 10 solar cells of 1 cm2

324 Solar Cells - Research and Application Perspectives

ered qualitatively.

**8. Conclusion**

power plant.

**Author details**

Isofotón S.A, Spain

Egbert Rodríguez Messmer

system efficiency, assuming an irradiation of 850 W/m2


**Chapter 12**

## **Electric Energy Management and Engineering in Solar Cell System**

Purnomo Sidi Priambodo, Didik Sukoco, Wahyudi Purnomo, Harry Sudibyo and Djoko Hartanto

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/52572

## **1. Introduction**

[12] Olson, J. (2003). Growth and characterization of high efficiency III-V multijunction solar cells for terrestial and space applications, Proceedings of the 10th European

[13] R.R. King, D.C. Law, K.M. Edmondson, C.M. Fetzer, G.S. KinseyH. Yoon, D.D. Krut, J.H. Ermer, R.A. Sherif, and N.H. Karam; Advances in high-efficiency III-V multi‐ junction solar cells, In: Advances in OptoElectronics, Volume 2007, 29523, doi:

[14] I. García, I. Rey-Stolle, B. Galiana, and C. Algora; A 32.6% efficient lattice-matched dual-junction solar cell working at 1000 suns, Applied Physics Letters 94, 053509

Workshop on MOVPE, Lecce (Italy) 2003

10.1155/2007/29523(2007).

326 Solar Cells - Research and Application Perspectives

(2009).

Solar cell system has many competitive advantages in comparisson to other renewable ener‐ gy resources. For instance, wind-turbin is very dependable to geographical location and has very high noise pollution if applied in residential area. Other example is micro-hydro, which depends on altitude and available in very limited locations. Furthermore, nuclear energy should be forgotten since its high radioactive risk. On the other side, solar cell system has characteristics of zero pollution, no radioactive risk, compact, portable and can be installed in any residential areas and has relatively high energy availability in any location on the earth surface in a year round. In general, solar cell array, which cover a residential roof house can supply the basic electrical energy needs of the residences who live in the house, almost a year round. These competitive advantages of solar cell system over other renewa‐ ble energy resources, make solar cell system the most favorite renewable energy resource.

There are 2 main topics and will be discussed in accordance to energy management of re‐ newable energy resources, based on solar cell system. **The first topic** is how to keep the sys‐ tem sustainable to supply the applied electrical load. As we have already known that solar energy is not available continuosly in a day and year round. For instance, at noon, the avail‐ ability of solar energy is abundant, however, at night there is not available at all. In the rain or winter season, the availability of solar energy is less than in the dry or summer season. On the other hand, the needs of electrical energy may be in opposite situation of the availa‐ bility of solar energy. In order to keep the solar cell system able to serve the total electrical load, it is necessary to design the system, which has sufficient number of solar cells and bat‐

© 2013 Priambodo et al.; licensee InTech. This is an open access article 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. © 2013 Priambodo et al.; licensee InTech. This is a paper 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.

teries to get and store the electrical energy from solar energy at the most available energy time (noon), and delivering to the consumers at the non-available solar energy time (night). In this first topic, in order to keep sustainable, it is conducted by designing the sufficient number of solar cells and batteries to supply the predicted electrical load.

In the first topic, the sustainability perspective in an energy network is emphasized in the form of designing the sufficient number of solar cells and batteries to supply the predicted electrical load. **The second topic** is sustainability to deliver energy perspective. It is focused on how conducting collaboration between several autonomy units of renewable energy sys‐ tem to build a renewable energy resource grid. Even, if possible, to do integration between autonomy units of solar system and the conventional electrical state own company. Here, we emphasize that the key problem of electric energy management of renewable energy re‐ souces such as solar cell system is sustainability.

## **2. Electric energy management in an autonomy unit of solar cell system: A perspective**

There are at least 2 strategic ways to implement renewable energy resources especially solar cell systems to fulfill the national electrical energy needs. The first strategy is to encourage the people to fulfill their own basic residential electric energy need by building private solar cell system. The second strategy is to let the Government as the regulator to drive a consorti‐ um of companies to build large plants of solar cell system to fulfill the regional or national electric energy need. Of course, the consortium will require a large amount of financial in‐ vestment at the starting point, however along with the time, the long term electric cost will be getting lower and more cost effective, since solar cell system requires very minimum maintenance cost and free of solar energy.

A solar cell system as an autonomy energy resource unit must have an energy manage‐ ment control unit, which embeded in the system. In general, there are at least 5 parts should exist in a electrical renewable energy resource system, as shown on Fig 1 below, i.e.: (1) solar cell array; (2) management energy control system; (3) energy storage (s); (4) DC to AC and AC to DC converters and (5) delivery bus. Thus 5 parts should be de‐ signed such that the system becomes more efficient to manage the gathered electrical en‐ ergy and to reach higher sustainability with low investment cost. We need 100% sustainability to fulfill the electrical energy need.

In order to reach 100% sustainability, first of all, the designer has to know how much en‐ ergy need in average in every single day (prediction). Then, the designer must consider the region, where the solar cell system is located and it relates to the earth latitude. Fur‐ ther, it counts to the statistical condition of how long the total time the sun shines on thus region in a year round. Another important info is the statistical info of the longest dura‐ tion of NO sunshine days, which relates to the seasons and weather. Thus are the most critical information to determine the requirement of the total number of solar cells and batteries in the system.

**Figure 1.** Big picture of Solar Cell Based Renewable Energy Unit System.

teries to get and store the electrical energy from solar energy at the most available energy time (noon), and delivering to the consumers at the non-available solar energy time (night). In this first topic, in order to keep sustainable, it is conducted by designing the sufficient

In the first topic, the sustainability perspective in an energy network is emphasized in the form of designing the sufficient number of solar cells and batteries to supply the predicted electrical load. **The second topic** is sustainability to deliver energy perspective. It is focused on how conducting collaboration between several autonomy units of renewable energy sys‐ tem to build a renewable energy resource grid. Even, if possible, to do integration between autonomy units of solar system and the conventional electrical state own company. Here, we emphasize that the key problem of electric energy management of renewable energy re‐

**2. Electric energy management in an autonomy unit of solar cell system:**

There are at least 2 strategic ways to implement renewable energy resources especially solar cell systems to fulfill the national electrical energy needs. The first strategy is to encourage the people to fulfill their own basic residential electric energy need by building private solar cell system. The second strategy is to let the Government as the regulator to drive a consorti‐ um of companies to build large plants of solar cell system to fulfill the regional or national electric energy need. Of course, the consortium will require a large amount of financial in‐ vestment at the starting point, however along with the time, the long term electric cost will be getting lower and more cost effective, since solar cell system requires very minimum

A solar cell system as an autonomy energy resource unit must have an energy manage‐ ment control unit, which embeded in the system. In general, there are at least 5 parts should exist in a electrical renewable energy resource system, as shown on Fig 1 below, i.e.: (1) solar cell array; (2) management energy control system; (3) energy storage (s); (4) DC to AC and AC to DC converters and (5) delivery bus. Thus 5 parts should be de‐ signed such that the system becomes more efficient to manage the gathered electrical en‐ ergy and to reach higher sustainability with low investment cost. We need 100%

In order to reach 100% sustainability, first of all, the designer has to know how much en‐ ergy need in average in every single day (prediction). Then, the designer must consider the region, where the solar cell system is located and it relates to the earth latitude. Fur‐ ther, it counts to the statistical condition of how long the total time the sun shines on thus region in a year round. Another important info is the statistical info of the longest dura‐ tion of NO sunshine days, which relates to the seasons and weather. Thus are the most critical information to determine the requirement of the total number of solar cells and

number of solar cells and batteries to supply the predicted electrical load.

souces such as solar cell system is sustainability.

328 Solar Cells - Research and Application Perspectives

maintenance cost and free of solar energy.

sustainability to fulfill the electrical energy need.

batteries in the system.

**A perspective**

Solar cell produces DC electrical energy, which fits to be storage in batteries. In designing solar cell system, as explained above, it must be determined the assumption of average need of energy per day, for example A Watt-hour/day. Further, the estimate of statistical condition of how long the total time of NO sunshine days on thus region in a year round must be determined, for example N days. The amount of solar cells and batteries needed by the system is written in the following equation:

$$\begin{pmatrix} 1+c \\ \end{pmatrix} \cdot N \cdot A \quad \text{Watt hour} \tag{1}$$

where *c* is a leak energy coefficient of the battery. In general, it has been known that the bat‐ tery is not perfect to store DC electrical energy, it is always a part of stored energy in battery leaks. This is a inefficiency factor of battery and presented as "*c"* coefficient.

Solar cell system performance fully depends on the performance of thus 5 parts in building the system; which has been listed above (Fig 1). The following is explanation of every part.

## **3. Solar cell and eficiency**

The main characteristic of solar cell is I-V curve. It has several derivative parameters such as*Isc* (short circuit current), *Voc* (open circuit voltage) and the maximum possible delivered energy *Pmp* =*Vmp* ⋅ *Imp*, as shown on the following Fig 2.

**Figure 2.** The Graph of the I-V characteristics of an ideal diode solar cell when non-illuminated (dark) and illuminated [1].

The main parameter that determines the solar cell efficiency is the maximum square area (power) as form of multiplication I-V (*Pmp* =*Vmp* ⋅ *Imp*), which is a maximum square formed inside I-V curve as shown on Fig 2 above. The next derivative parameter is fill factor *FF* that represents the ratio PMP to the product *VOC* and *ISC*. This parameter gives an insight abou‐ thow "square" is the output characteristic.

$$FF = \frac{P\_{MP}}{V\_{OC} \cdot I\_{SC}} = \frac{V\_{MP} \cdot I\_{MP}}{V\_{OC} \cdot I\_{SC}} \tag{2}$$

In the case of solar cell with sufficient efficiency, in general, it has FF between 0.7 and 0.85. The energy–conversion efficiency, η can be written as [2]

$$\ln \eta = \frac{V\_{MP} \cdot I\_{MP}}{P\_{in}} = \frac{V\_{OC} \cdot I\_{SC} \cdot FF}{P\_{in}} \tag{3}$$

where *Pin* is the total power of light illumination on the cell. Energy-conversion efficiency of commercial solar cells typically lies in between 12 and 14 % [2]. In designing a good solar cell, we have to consider and put any effort to make those four parameters *ISC*, *VOC*, FF and η as optimum as possible [1]. We like to use term optimum than maximum, since the effort to obtain one parameter to be maximum ,in designing solar cell, will degrade other parame‐ ters. Hence the best is considering the optimum efficiency of solar cell.

## **4. Buss system**

Solar cell produces DC electric energy. For solar cell system, where the solar cell array has radius not more than 100-m to batteries and electrical loads, it is effective and cost efficient to be connected by DC buss system. By using DC buss system, in order to transfer electrical energy from solar cells to batteries and loads, the parameter needs to be considered is volt‐ age. The DC buss is the most efficient and cost effective, since it does not require electrical conversion from DC to AC. DC buss can be extended for more than 100-m, even can be more than several kms. In order to lower the DC electric power loss in the transmission from solar cells to batteries and loads, the DC transmission voltage should be increased, hence, to deliver the electrical power it requires only a very low current. This method is very popular used in AC electric transmission for long distance by using high voltage AC. The DC electric power transmission loss is written by the following equation:

$$P\_L = I\_{DC}^2 R\_\nu \tag{4}$$

*PL* is power loss in the DC transmission line, *Rtr* is the total transmission line resistance and *IDC* is the DC current on DC buss transmission line. Hence, to lower power loss and *Rtr* kept same, then *IDC* must be decrease much lower with consequence of *VDC* must be increase pro‐ portionality to the decrement of *IDC*. To increased voltage *VDC*, in general conducted by using boost converter method.

### **4.1. Boost converter for DC buss system**

Boost converter is an electronic circuit for *DC to DC converting*. It functions to increase volt‐ age VDC higher, i.e. by controlling the signal driver duty cycle. *Boost Converter* base circuit requires only 4 fundamental components, which are: inductor, electronic switch, diode and output capacitor, shown on Fig 3. The converter circuit can be operated in 2 modes, which depends on the energy storage capacities and the relative length of the switching period [3]. Those 2 methods are CCM (*Continuous Conduction Mode*) and DCM (*Discontinuous Coduction Mode*), where CCM is for efficient power conversion and DCM is for low power conver‐ sion[3].

### *4.1.1. Continuous Conduction Mode (CCM)*

## **Mode 1 (0 < t ≤ ton),**

**Figure 2.** The Graph of the I-V characteristics of an ideal diode solar cell when non-illuminated (dark) and illuminated [1].

The main parameter that determines the solar cell efficiency is the maximum square area (power) as form of multiplication I-V (*Pmp* =*Vmp* ⋅ *Imp*), which is a maximum square formed inside I-V curve as shown on Fig 2 above. The next derivative parameter is fill factor *FF* that represents the ratio PMP to the product *VOC* and *ISC*. This parameter gives an insight abou‐

> *MP MP MP OC SC OC SC P VI FF VI VI*

In the case of solar cell with sufficient efficiency, in general, it has FF between 0.7 and 0.85.

where *Pin* is the total power of light illumination on the cell. Energy-conversion efficiency of commercial solar cells typically lies in between 12 and 14 % [2]. In designing a good solar cell, we have to consider and put any effort to make those four parameters *ISC*, *VOC*, FF and η as optimum as possible [1]. We like to use term optimum than maximum, since the effort to obtain one parameter to be maximum ,in designing solar cell, will degrade other parame‐

*MP MP OC SC in in V I V I FF P P*

<sup>×</sup> = = × × (2)

× ×× = = (3)

thow "square" is the output characteristic.

330 Solar Cells - Research and Application Perspectives

The energy–conversion efficiency, η can be written as [2]

h

ters. Hence the best is considering the optimum efficiency of solar cell.

Mode 1 starts, when switch S (MOSFET) switched on at *t = 0* until *t = ton*. The equivalent cir‐ cuit for Mode 1 is shown on the following Fig 4a. By assuming that the serial resistance val‐ ue DC voltage source is relatively low, there will be an inductor current transient *iL(t)* larger than zero and increase linearly at the beginning of transient. Inductor voltage is *VL= Vi* .

#### **Mode 2 (ton< t ≤ T<sup>s</sup> ),**

Mode 2 starts, when switch S (MOSFET) switched off at *t = ton* until *t = Ts*. The equivalent circuit for Mode 2 is shown on Fig 4b. Inductor voltage, *VL* in this period is *Vi – Vo*. In this case *Vi < Vo*, it means in Mode 2, *VL*is in opposite direction to *VL* in Mode 1.

**Figure 3.** Basic DC Voltage Boost Converter Circuit [3]

**Figure 4.** *Continuous Coduction Mode*:(a) Close switch, (b) Open switch

In steady state operation, the signal formed due to switching is repeated over all the time. The integral of inductor voltage *vL* in one period must be equal to zero, where *Ts = ton+toff*. Therefore, the total summation of inductor voltage at open switch and close switch must be equal to zero.

$$(V\_i t\_{ow} + (V\_i - V\_o) t\_{off} = 0) \tag{5}$$

*+*

*\_*

Where:

*Vi* : input voltage

*Vo* : the average of output voltage

*ton* : time *on*

*Vi*

*+*

*\_*

*Vi Ii (t)*

*+*

*\_*

equal to zero.

*I i (t)*

332 Solar Cells - Research and Application Perspectives

*I <sup>L</sup>(t)*

> *t on t off Ts*

*i L vL*

*t*

*vL(t) vL <sup>L</sup> (t) <sup>b</sup> Lb*

*IC(t)*

*Cb*

*IO(t)IL(t)*

**Figure 4.** *Continuous Coduction Mode*:(a) Close switch, (b) Open switch

*on <sup>t</sup>*

*Ts*

*VO*

*LOAD*

*\_*

In steady state operation, the signal formed due to switching is repeated over all the time. The integral of inductor voltage *vL* in one period must be equal to zero, where *Ts = ton+toff*. Therefore, the total summation of inductor voltage at open switch and close switch must be

*+*

*off*

*Vi Ii (t)*

*+*

*\_*

*Vi - Vo*

*i L*

**Figure 3.** Basic DC Voltage Boost Converter Circuit [3]

*Lb*

*IC(t)*

*Lb*

*ID(t)*

(a) (b)

*Cb MOSFETb*

*I <sup>O</sup>(t)*

*VO*

*LOAD*

*IC(t)*

*Cb*

*VO*

*LOAD*

*\_*

*+*

*IO(t)IL(t)*

– 0 ( ) *Vt V V t i on i o off* + = (5)

*\_*

*+*

toff : time off

*Ts*: switching period

By arranging and separating *Vi* and *Vo*, and then dividing the both sides by*Ts*, it results in

$$\frac{Vo}{\frac{Vi}{Vi}} = \frac{Ts}{\frac{1}{1} \cdot D} = \frac{1}{1 \cdot D} \tag{6}$$

where *D* is duty cycle.

By assuming that the circuit has 100% efficiency, i.e *Pi = Po*

$$I\_i V\_i = \, I\_o V\_o \tag{7}$$

$$\frac{I\_0}{\Pi} = 1 \, \text{--} D\tag{8}$$

where

*Io* : the average output current

*Ii* : the average input current

When the switch is close,

 $V\_{L} = V\_{j}; L \stackrel{di\_{L}}{dt} = V\_{j}; \frac{di\_{L}}{dt} = \frac{V\_{i}}{L}$ 
$$\frac{di\_{L}}{dt} = \frac{\Delta i\_{L}}{\Delta t} = \frac{\Delta i\_{L}}{DT} \quad \rightarrow \quad \frac{di\_{L}}{dt} = \frac{V\_{i}}{L}$$

$$\Delta i\_{L \text{ (dose)}} = \frac{V\_{i}DT}{L} \tag{9}$$

When the switch is open,

$$\begin{aligned} \left| V \right\rangle\_L &= V\_i \cdot \left| V \right\rangle; \frac{di\_L}{dt} = V\_i \cdot \left| V \right\rangle; \frac{di\_L}{dt} = \frac{V\_i \cdot \left| V \right\rangle}{L} \\\\ \frac{di\_L}{dt} &= \frac{\Delta i\_L}{\Delta t} = \frac{\Delta i\_L}{(1 \cdot D)T} \quad \rightarrow \quad \frac{di\_L}{dt} = \frac{V\_i \cdot \left| V \right\rangle}{L} \end{aligned}$$

$$\Delta \dot{I}\_L \text{ (open)} = \frac{(V\_i \cdot Vol)(1 \cdot D)T}{L} \tag{10}$$

At the transient time, where *Vo* is going to steady state condition, ∆*i <sup>L</sup>* (*open*) or *iL* slope at Mode 2 also experience transient following the gradient of *Vo* transient. At the time *Vo* ach‐ ieves steady state condition, then ∆*i <sup>L</sup>* (*open*) achieves steady state as well.

### *4.1.2. Discontinuous Conduction Mode (DCM)*

At this mode, the inductor current will drop to zero before finishing one switching period, as shown on Fig 5. As the CCM analysis, the voltage inductor integral during one period is zero.

$$\text{ViDTs} + \text{(Vi - Vol)}D\_1 \text{Ts} = 0 \tag{11}$$

Then

$$\frac{V\_O}{\overline{V\_1}} = \frac{D\_1 + D\_2}{D\_1} \tag{12}$$

and

$$\frac{I\_o}{I\_i} = \frac{D\_1}{D\_1 + D} \tag{13}$$

*I C(t)*

*\_*

*\_*

*IC(t)*

*Cb*

*I*

*VO*

*L*

*OA*

*D*

*\_*

*+*

**Figure 5.** Equivalent circuit for DCM mode.

*I C(t)*

*\_*

*\_*

**a.** Mode 1(0 < t ≤ ton);

*\_*


From Fig 5.c, the average input current is equal to the inductor current,

$$I\_i = \frac{V\_i}{2L\_{\phantom{a}}} DT\_s \{D + D\_1\} \tag{14}$$

By using equation (13),

At the transient time, where *Vo* is going to steady state condition, ∆*i <sup>L</sup>* (*open*) or *iL* slope at Mode 2 also experience transient following the gradient of *Vo* transient. At the time *Vo* ach‐

At this mode, the inductor current will drop to zero before finishing one switching period, as shown on Fig 5. As the CCM analysis, the voltage inductor integral during one period is

*ViDTs* + (*Vi* - *Vo*)*D*1*Ts* =0 (11)

*<sup>D</sup>*<sup>1</sup> <sup>+</sup> *<sup>D</sup>* (13)

(12)

ieves steady state condition, then ∆*i <sup>L</sup>* (*open*) achieves steady state as well.

*Vo Vi* <sup>=</sup> *<sup>D</sup>*<sup>1</sup> <sup>+</sup> *<sup>D</sup> D*1

*I o Ii* <sup>=</sup> *<sup>D</sup>*<sup>1</sup>

*t*

*DTs*

*VO*

*L*

*OA*

*D*

*\_*

*+*

*i L vL*

*I C(t)*

*Cb*

*IO(t)IL(t)*

*on t off Ts*

> *Vi I i (t)*

*\_*

*+*

*i L*

*vL(t)Lb*

*D1Ts*

*Vi - Vo*

> *I C(t)*

*VO Cb*

*L*

*O*

*A*

*D*

*+*

*Vi I i (t)* *vL <sup>L</sup> (t) <sup>b</sup>*

*+*

*\_*

*IC(t)*

*Cb*

*<sup>O</sup>(t)IL(t)*

*I*

*VO*

*L*

*OA*

*D*

*\_*

*+*

*\_*

*IO(t)IL(t)*

*4.1.2. Discontinuous Conduction Mode (DCM)*

334 Solar Cells - Research and Application Perspectives

zero.

Then

and

*Vi I i (t)*

*\_*

**a.** Mode 1(0 < t ≤ ton);

*+*

*vL(t) Lb*

**Figure 5.** Equivalent circuit for DCM mode.

**b.** Mode 2(ton < t ≤ (D+D1)Ts);

**c.** Mode 3(D + D1)Ts< t ≤ T<sup>s</sup>

$$I\_o = \left(\frac{V\_i T\_s}{2L\_{\
u}}\right) D D\_1 \tag{15}$$

In practice, *duty cycle* D should change to respone the *Vi* change, such that obtaining con‐ stant *Vo*. It requires an electronic feedback control system, as a function of loading curent for the change of *Vi /Vd.* This functions to control *duty cycle*. By using equations (12) and (15), we obtain:

$$D = \left[\frac{4V\_o}{2\overline{\tau}V\_i} \left(\frac{V\_o}{V\_i} - 1\right) \frac{I\_o}{I\_{s,avg,max}}\right]^{0.5} \tag{16}$$

Where *Io,avg,max* is the average of maximum output current, which is obtained via the follow‐ ing equation:

$$I\_{o,avg} = \frac{T\_s V\_0}{2L} D (1 \text{ - } D)^2 \tag{17}$$

The average output current will be maximum, when D = 1/3

$$I\_{o,avg,max} = \frac{2}{2\mathcal{T}} \frac{T\_\* V\_o}{L\_{-b}} \tag{18}$$

The critical inductance Lbc, is defined as inductance at the region of boundary, between *con‐ tinuous* and *discontinuous modes,* and defined as:

$$L\_{\
u} = \frac{RD(1 \cdot D)^2}{2F\_s} \tag{19}$$

where:

*R* : equivalent load, Ohm

*Fs* : switching frequency, Hz

### **4.2. DC to AC inverter for AC buss system**

When the distance between renewable energy clusters of solar cells, batteries and electrical loads is relatively far (can be hundred kms) and the electrical loads mostly are AC electrical loads, then it is necessary to consider to use AC buss system, which supported by DC to AC inverters. By utilizing AC buss system, for long distance electrical transmission, the increase of AC voltage can be conducted by using passive transformator, which is common to be used. However, the integration process of several renewable energy autonomy systems is relatively more complex than integration in DC buss system. The problems in AC buss inte‐ gration are due to more parameters that must be synchronized, such as voltage, frequency and phase. While in DC buss integration, it is only facing voltage synchronization.

In general, AC electrical power transmission is delivered in 3 (three) phases, especially for 3 phase electrical-mechanical motor loads, in order to be more smooth and more efficient in operation. 3-phase system is inherent in electrical generator based on mechanical generator, by arranging the three generator coils in three different locations by 1200 phase angle in the generator. It also happens in the DC to AC inverter 3 phase. The sine generator generates 3 equal sine wave with different phase each of 1200 . DC-to-AC 1-phase inverter system is the fundamental to develop 3-phase system. The knowledge of working mechanism of 1-phase system is very helpful to understand 3-phase system.

In this chapter, we will discuss the working mechanism of DC to AC inverter 1-phase sys‐ tem in general. Then it is continued by discussion of the methods to synchronize thus three parameters of AC buss system, i.e: voltage, frequency and phase. In general, the use of DC or AC buss depends on the distance between sources, batteries and loads, also the variances of the loads. The ultimate consideration is energy efficiency and cost effective of the solar cell sytem.

## *4.2.1. Full bridge inverter*

DC-AC inverter is a vital component in the solar cell system in order to support AC buss system for AC load. DC to AC inverter technology has been developed since the beginning of electronics technology era. At the beginning, DC-to-AC inverter was developed based on sinusoidal oscillator, which is amplified by push-pull amplifier of B class that has maximum efficiency of 50%. The 50% power loss is due to instantaneous drop-voltage at the final tran‐ sistors on the push-pull amplifier. The fact of the 50% power loss is due to the sinusoidal form of the current and voltage running through the final transistor in DC to AC inverter circuit.

By realizing that push-pull final has maximum 50% efficiency, then full-bridge inverter tech‐ nology was developed to increase the efficiency of DC to AC inverter. The work mechanism of bridge inverter is based on switching methods, as shown on the circuits of Fig 6 and 7, where at switch "on" the load current (*IL*) goes through maximum, however the voltage drop (*VDR*) across switch is very minimum. While at switch "off" the load current (*IL*) goes through minimum, however the voltage drop (*VDR*) across switch is maximum. Hence, it can be expected that power loss in the final transistors in the bridge inverter method is very small, which can be represented in equation:

$$P\_L = V\_{DR} I\_L \quad \text{always minimum} \tag{20}$$

*PL* or power loss is always minimum, either when switch "on" or "off".

In order that bridge inverter idea realized, then the power input to the final transistors must be a constant voltage *Vcc* and on-off discrete signal to control bride-inverter switches. The on-off signal is in the form of discrete signal. The advantage of using bridge inverter (either half or full) is improving the electrical power conversion DC to AC efficiency, where the ide‐ al is close to 100 %. The high output efficiency makes the bridge converter technique repla‐ ces push-pull B class amplifier for DC to AC inverter. There are at least 2 (two) basic fundamental of bride inverter configurations exist, i.e half bride inverter (Fig 6) and full bridge inverter (Fig 7).

## *4.2.1.1. Half-bridge inverter*

inverters. By utilizing AC buss system, for long distance electrical transmission, the increase of AC voltage can be conducted by using passive transformator, which is common to be used. However, the integration process of several renewable energy autonomy systems is relatively more complex than integration in DC buss system. The problems in AC buss inte‐ gration are due to more parameters that must be synchronized, such as voltage, frequency

In general, AC electrical power transmission is delivered in 3 (three) phases, especially for 3 phase electrical-mechanical motor loads, in order to be more smooth and more efficient in operation. 3-phase system is inherent in electrical generator based on mechanical generator, by arranging the three generator coils in three different locations by 1200 phase angle in the generator. It also happens in the DC to AC inverter 3 phase. The sine generator generates 3

fundamental to develop 3-phase system. The knowledge of working mechanism of 1-phase

In this chapter, we will discuss the working mechanism of DC to AC inverter 1-phase sys‐ tem in general. Then it is continued by discussion of the methods to synchronize thus three parameters of AC buss system, i.e: voltage, frequency and phase. In general, the use of DC or AC buss depends on the distance between sources, batteries and loads, also the variances of the loads. The ultimate consideration is energy efficiency and cost effective of the solar

DC-AC inverter is a vital component in the solar cell system in order to support AC buss system for AC load. DC to AC inverter technology has been developed since the beginning of electronics technology era. At the beginning, DC-to-AC inverter was developed based on sinusoidal oscillator, which is amplified by push-pull amplifier of B class that has maximum efficiency of 50%. The 50% power loss is due to instantaneous drop-voltage at the final tran‐ sistors on the push-pull amplifier. The fact of the 50% power loss is due to the sinusoidal form of the current and voltage running through the final transistor in DC to AC inverter

By realizing that push-pull final has maximum 50% efficiency, then full-bridge inverter tech‐ nology was developed to increase the efficiency of DC to AC inverter. The work mechanism of bridge inverter is based on switching methods, as shown on the circuits of Fig 6 and 7, where at switch "on" the load current (*IL*) goes through maximum, however the voltage drop (*VDR*) across switch is very minimum. While at switch "off" the load current (*IL*) goes through minimum, however the voltage drop (*VDR*) across switch is maximum. Hence, it can be expected that power loss in the final transistors in the bridge inverter method is very

*P VI L DR L* = always minimum (20)

. DC-to-AC 1-phase inverter system is the

and phase. While in DC buss integration, it is only facing voltage synchronization.

equal sine wave with different phase each of 1200

336 Solar Cells - Research and Application Perspectives

small, which can be represented in equation:

cell sytem.

circuit.

*4.2.1. Full bridge inverter*

system is very helpful to understand 3-phase system.

The following Fig 6 shows the circuit configuration of half bridge inverter. The circuit con‐ sists of 2 switching elements, S1 dan S2. Each element has one anti parallel diode. The switch‐ ing element can be transistor, MOSFET, or IGBT.

**Figure 6.** The circuit configuration of half-bridge inverterand example of output signal

The basic operation of half-bridge inverter circuit consists of 2 conditions:


The switching process for S1and S2 must be designed such that both are not in "ON" condi‐ tion at the same time. If this happens, it will happen short connection input *Vdc*, which will cause damage on the switching elements.

### *4.2.1.2. Full-bridge Inverter*

Fig 7 shows the circuit configuration of full-bridge inverter 1-phase. The circuit consists of 4 switching elements: S1, S2, S3, and S4. The circuit operation consists of 2 conditions:


As explained in *half-bridge inverter*, to avoid short condition on VDC, the switching process should be designed such that at S1and S4*ON*, S2and S3must be *OFF* and vice versa. For the sake of this purpose, *gate* driver should use *dead time* mechanism.

**Figure 7.** The circuit configuration of full-bridge inverterand example of output signal.

From Fig 6 and Fig 7, it can be concluded that the peak-to-peak output voltage of half-bridge configuration is a half of full-bridge one. The square wave output voltage has spectrum as shown on the following Fig 8.

**Figure 8.** Square-Wave Harmonics Analysis of Unfiltered Output[4]

*4.2.1.2. Full-bridge Inverter*

338 Solar Cells - Research and Application Perspectives

*VDC*

drop on the load with value of Vdc.

age will drop on the load with value of -Vdc.

sake of this purpose, *gate* driver should use *dead time* mechanism.

*RL*

**Figure 7.** The circuit configuration of full-bridge inverterand example of output signal.

From Fig 6 and Fig 7, it can be concluded that the peak-to-peak output voltage of half-bridge configuration is a half of full-bridge one. The square wave output voltage has spectrum as

<sup>O</sup> <sup>A</sup>

*VAO*

S1 S2

S3 S4

shown on the following Fig 8.

Fig 7 shows the circuit configuration of full-bridge inverter 1-phase. The circuit consists of 4

**1.** At S1 and S4 ON, S2 and S3 OFF, in the first half period, then the output voltage will

**2.** While at S2 and S3 ON, S1 and S4 OFF, in the second half periode, then the output volt‐

As explained in *half-bridge inverter*, to avoid short condition on VDC, the switching process should be designed such that at S1and S4*ON*, S2and S3must be *OFF* and vice versa. For the

*VDC*

*VAO*

*t*

*-VDC*

switching elements: S1, S2, S3, and S4. The circuit operation consists of 2 conditions:

If all spectral components of power exist on the spectrum, are added together and assumed to be the output of the bridge-inverter, then the output efficiency can reach close to 100%. However, the higher order harmonics in the spectrum are not useful, even possible have ru‐ ining effects on the electro- mechanical loads. If higher order harmonics are substracted from the total output (by filtering), hence only the fundamental signal left, and the output efficiency is just 70%. To improve the output efficiency (in bridge inverter context), the dis‐ crete switch input control can be modified on to: (1) modified sine wave or (2) pulse width modulation (PWM). The following Fig 9 illustrates comparisson between pure sine wave, modified sine wave and square wave.

**Figure 9.** Comparisson between pure sine wave, modified sine wave and square wave.

As for instance, after going through filtering process, the square waveoutput signal is still consisting of higher order harmonics as shown on Fig 10, as following.

**Figure 10.** Output signal square wave as shown on Fig 6 and 7 (after filtered).

Eventhough modified-sine wave improves the output by suppressing the harmonics more than the square wave, however technically, it is too expensive, because conducting controls on two inputs: switching and Vcc. Furthermore, we will leave this modified-sine wave method.

The method that is proven good to improve output efficiency by suppressing higher order harmonics is pulse width modulation (PWM). In general PWM is the squential of discrete pulse, which is based on pulse width modulation. PWM technique becomes famous to gen‐ erate pure sinusoidal wave, which is applied for DC to AC inverter and for controlling elec‐ trical motor.

### *4.2.1.3. Sinusoidal pulse wave modulation*

The basic principal in forming PWM sine wave is by comparing two waves i.e. sinus wave as the reference and triangle wave as the carrier in real time. The sine wave has frequency *fr*, which will be the inverter output frequency, i.e. 50 Hz or 60 Hz. The carrier signal has fre‐ quency *fc*, which becomes switching frequency in inverter circuit. The ratio between *fc* and *fr*is called frequency modulation ratio, *mf* , which is defined as:

$$
\Delta m\_f = \frac{fc}{fr} = \frac{ftri}{fsin} \tag{21}
$$

The typical switching frequency is between 2 kHz – 15 kHz, and sufficient for power system applications. The higher carrier frequency, the easier conducting filtering that is separating fundamental frequency output from the carrier frequency and its higher harmonics. Howev‐ er, the higher the switching or triangle frequency, it will increase interference effect to other electronics instruments (electromagnetic compatibility – EMC). Fig 11 below, shows the ba‐ sic concept of signal comparisson between reference and carrier signals, or sometimes called as 2-level PWM.

**Figure 11.** Basic concept formation of 2-level sine wave PWM

The comparisson of two input signals (sinusoidal reference and triangle carrier) results in discrete signal that is called as pulse width modulation (PWM), as shown on Fig 11 above. Moreover, the ratio between reference signal and carrier amplitudes, is called as amplitude modulation ratio *ma*, which is defined as:

$$\text{JCM}\_{a} = \frac{Vm\_{r}}{Vm\_{r}} \frac{\text{sinusoidal reference}}{\text{triangle carrier}} \tag{22}$$

Where :

As for instance, after going through filtering process, the square waveoutput signal is still

Eventhough modified-sine wave improves the output by suppressing the harmonics more than the square wave, however technically, it is too expensive, because conducting controls on two inputs: switching and Vcc. Furthermore, we will leave this modified-sine wave

The method that is proven good to improve output efficiency by suppressing higher order harmonics is pulse width modulation (PWM). In general PWM is the squential of discrete pulse, which is based on pulse width modulation. PWM technique becomes famous to gen‐ erate pure sinusoidal wave, which is applied for DC to AC inverter and for controlling elec‐

The basic principal in forming PWM sine wave is by comparing two waves i.e. sinus wave as the reference and triangle wave as the carrier in real time. The sine wave has frequency *fr*, which will be the inverter output frequency, i.e. 50 Hz or 60 Hz. The carrier signal has fre‐ quency *fc*, which becomes switching frequency in inverter circuit. The ratio between *fc* and

*fr* <sup>=</sup> *ftri*

The typical switching frequency is between 2 kHz – 15 kHz, and sufficient for power system applications. The higher carrier frequency, the easier conducting filtering that is separating fundamental frequency output from the carrier frequency and its higher harmonics. Howev‐

*mf* <sup>=</sup> *fc*

, which is defined as:

*fsin* (21)

consisting of higher order harmonics as shown on Fig 10, as following.

340 Solar Cells - Research and Application Perspectives

**Figure 10.** Output signal square wave as shown on Fig 6 and 7 (after filtered).

method.

trical motor.

*4.2.1.3. Sinusoidal pulse wave modulation*

*fr*is called frequency modulation ratio, *mf*

*Vm, sinusoidal reference* : peak amplitude of reference wave

*Vm, triangle carrrier* : peak amplitude of carrier wave

The PWM output, as result of comparisson between sinusoidal reference signal and triangle carrier signal, can be represented in the form of *transcendental equation*. Later on *ma* can be used to control the output amplitude of the PWM fundamental frequency. Moreover, the value of *ma* can be used to compensate the variation of DC input voltage, such that resulting in constant output AC voltage.

### *4.2.2. 3-Level PWM realization*

2-level PWM, which is illustrated on Fig 11, is successful to suppress the higher order har‐ monics, such that improving inverter efficiency close to 80%. In order to suppress more on higher order harmonics, it is proposed to use 3-level PWM. To realize the 3-level PWM con‐ cept, it is required a circuit that can control the synchronization of switch pairs: S1 with S4 and S2 with S3, which are represented by H-bridge on Fig 7. The circuit that realizes 3-level PWM is shown on Fig 12. It is the development of basic concept of 2-level as shown on Fig 11. The 3-level PWM requires 2 equal inputs of sine wave references with 1800 phase differ‐ ent. The resulting 3-level PWM signal is shown on Fig. 12 below.

**Figure 12.** Illustration of 3-level PWMGenerator dan process generation of 3-level PWM [4, 5]


Fig 13 shows the output spectrum of 3-level PWM, which has output efficiency close to 85%, even one reports achieving 90%. If it is compared to the square wave spectrum, which is shown in Fig 8 and has output efficiency of 65-75%, then it can be concluded that 3-level PWM gives us a significant efficiency improvement.

**Figure 13.** level PWM Harmonics Analysis of Unfiltered Output

#### **4.3. Filter**

value of *ma* can be used to compensate the variation of DC input voltage, such that resulting

2-level PWM, which is illustrated on Fig 11, is successful to suppress the higher order har‐ monics, such that improving inverter efficiency close to 80%. In order to suppress more on higher order harmonics, it is proposed to use 3-level PWM. To realize the 3-level PWM con‐ cept, it is required a circuit that can control the synchronization of switch pairs: S1 with S4 and S2 with S3, which are represented by H-bridge on Fig 7. The circuit that realizes 3-level PWM is shown on Fig 12. It is the development of basic concept of 2-level as shown on Fig

phase differ‐

11. The 3-level PWM requires 2 equal inputs of sine wave references with 1800

+ -

+ -

**Figure 12.** Illustration of 3-level PWMGenerator dan process generation of 3-level PWM [4,

**a.** Comparisson between 2 sine reference signals dan triangle carrier signal

**COMPARATOR 1**

**COMPARATOR 2**

S1 S3

S2 S4

5]

ent. The resulting 3-level PWM signal is shown on Fig. 12 below.

Vr

Vc


in constant output AC voltage.

342 Solar Cells - Research and Application Perspectives

*4.2.2. 3-Level PWM realization*

**b.** Pulse for S1 and S4

**c.** Pulse for S2 san S3

**d.** Output wave

Fig 13 shows that 3-level PWM suppresses higher order harmonics much better than square wave. However, the existing higher order harmonics are still able to annoy the performance of electro-mechanical systems. By this reason, those higher order harmonics must be sop‐ press more such that the harmonics becomes very low and not significant. We need a lowpass filter, which has a cut-off frequency *fc< f2* (where: *f2* is second harmonic frequency). The filter is realized in the form of passive filter that consists of passive components L and C, as shown on Fig 14 below.

**Figure 14.** A simple low-pass L-C filter to filter out higher order harmonics of PWM.

The inductor value is designed such that the drop voltage on the inductor should be <3% of the inverter output voltage,

$$\mathbf{I}\_{\text{load},\text{max}}\mathbf{2}.\pi.\mathbf{f}.\mathbf{L}\le\mathbf{0}.\mathbf{0}\mathbf{3}\mathbf{V}\_{\text{ac}}\tag{23}$$

Where:

*Iload,max* : maximum RMS load current

*Vac* : RMS output voltage

*f* : output frequency

For a simple low-pass L-C filter, the cut-off frequency is[6]:

$$f\_c = \frac{1}{2\pi\sqrt{LC}}\tag{24}$$

$$\mathcal{C} = \frac{\left(\frac{1}{2 \times \pi \times f}\right)^2}{L} \tag{25}$$

However, sometimes, as a recommendation, the cut-off frequency should be set up on 1 or 2 octave above the fundamental frequency, i.e. 150 Hz, for 50 Hz system.

The big picture of 3-level PWM inverter 1-phase full-bridge is illustrated on Fig 15 below. Four op-amps that exist on Fig 15 function as combination of 2 NANDS logic circuits and 2 op-amps as shown

**Figure 15.** Comprehenssive 3-level PWM schematic blocks [4].

on Fig 12. For the sake of integration with external electrical AC buss, the system conducting synchronization of 3 main parameters: frequency (50 or 60 Hz), phase and voltage. To sim‐ plify synchronization process, the sensing (monitoring) and controlling those 3 parameters must be conducted by microprocessor system.

## **5. Energy management: Energy flow-in, flow-out and monitoring energy in the baterrays**

The keyword for energy management in solar cell system is sustainability. The main rule, if the total electrical energy supply from solar cells to the batteries is less than the total used energy, then to maintain the sustainabilty, there must be energy supply from the external energy resource(s). Moreover, if the total electrical energy supply from solar cells to the bat‐ teries is more than the total used energy, then the collected energy from the solar cells must be stored to the batteries, if full, then must be delivered to the external AC buss. This pur‐ pose is that the excess of supply energy can be utilized by other outside consumers that con‐ nected to common external AC buss. If there exist many autonomy renewable energy system, then they can be coordinated to build an energy grid system, which makes possible to collaborate each other to maintain sustainability altogether. By the existence of a grid sys‐ tem, then a business concept of buying-selling electrical energy supply between autonomy renewable energy systems can be realized.

In order such that Solar cell system can function as autonomy energy system, then it must has an energy management system in the form of an electronic control system, supported by microprocessor circuit or computer system, which monitors the total IN and OUT of electric energy on battery storage. Moreover, at the same time, it monitors the number of available energy in the storage and controlling the flow and contingency electric energy in the system. By monitoring IN and OUT energy and the availability energy in the storage, it is expected that the total energy available in the storage can always be monitored real time accurately. The core components of energy management in solar cell system are batterries and process‐ or system. The following is brief explanation regarding to both components.

### **5.1. Batteray analysis**

I*load,max*.2.π.f.L < 0,03V*ac* (23)

*LC* <sup>=</sup> (24)

ç ÷ ´ ´ è ø <sup>=</sup> (25)

Where:

*Iload,max* : maximum RMS load current

344 Solar Cells - Research and Application Perspectives

For a simple low-pass L-C filter, the cut-off frequency is[6]:

1 2 *<sup>c</sup> f*

2 1 <sup>2</sup> *<sup>f</sup> <sup>C</sup> L* p

However, sometimes, as a recommendation, the cut-off frequency should be set up on 1 or 2

The big picture of 3-level PWM inverter 1-phase full-bridge is illustrated on Fig 15 below. Four op-amps that exist on Fig 15 function as combination of 2 NANDS logic circuits and 2

on Fig 12. For the sake of integration with external electrical AC buss, the system conducting synchronization of 3 main parameters: frequency (50 or 60 Hz), phase and voltage. To sim‐ plify synchronization process, the sensing (monitoring) and controlling those 3 parameters

octave above the fundamental frequency, i.e. 150 Hz, for 50 Hz system.

**Figure 15.** Comprehenssive 3-level PWM schematic blocks [4].

must be conducted by microprocessor system.

p

æ ö

*Vac* : RMS output voltage

*f* : output frequency

op-amps as shown

Electric energy storage is used to store the received energy from solar cells (at noon), in or‐ der to be utilized at the time when there is no available electrical energy supply from the solar cell (at night). In general, electric energy storage can be realized in the form of wet and dry batteries, super capacitors and even carrier energy storage in the form of hydrogen gas storage. Priambodo et al [7], have shown that electrical energy received from solar cell array can be stored by converting it into energy carrier in the form of H2 gas by using electrolysis method. Furthermore, the stored H2 gas can be used when there is no available supply elec‐ tric energy from solar cells, by using fuel cell system. In this chapter, we will limit discus‐ sion to battery storage only.

Solar cell system needs battery bank to store the electric energy that collected in at noon. In general, there are two kinds of batteries, wet and dry. The wet battery uses electrolysis method, where it requires a sealed box to keep two plates anode and cathode, which are connected by wet electrolyte which can be base or acid. During charging time, there exists ionization process in electrolyte liquid, while during discharging process, there exists deion‐ ization process. The dry battery, actually is not really dry. The dry battery working concept is still the same with the wet one, however, the dry battery uses electrolyte gel, such that looks more dry. The components of battery are illustrated in Fig 16, as follows.

**Figure 16.** Component parts of a batteray

The illustration of chemical process in wet battery is shown in the following equations:

$$\text{Positive plates: }\ \text{PbO}\_2 + 4\text{H}^+ + \text{SO}\_4^{2-} + 2e \xrightarrow[\text{Crowage}]{\text{Diechange}} \text{PbSO}\_4 + 2\text{H}\_2\text{O}$$

$$\text{Positive plates: }\ \text{Pb} \cdot \text{SO}\_4^{2-} \xleftarrow{\text{Diechmpe}} \text{PbSO}\_4 + 2e^- \tag{26}$$

$$\text{Overall reaction:}\ \text{PbO}\_2 \cdot \text{Pb} \cdot 2\text{H}\_2\text{SO}\_4 \xrightarrow[\text{Crowage}]{\text{Diechmpe}} 2\text{PbSO}\_4 \cdot 2\text{H}\_2\text{O}$$

The battery perfomance is determined by at least 5 parameters: (1) the speed to store electric energy in charging process; (2) the battery capacity; (3) battery leak; (4) energy dissipation; and (5) the speed of capability for discharging to the loads. The following is discussion ac‐ cording the analysis methods for those 5 parameters. There are 2 (two) conditions, i.e. charg‐ ing and discharging, which requires different modelling. The reason to have different models for charging and discharging is to have a simplification, because the whole analysis components are related each other, in complex relation.

### *5.1.1. Batteray charging model*

Charging process can be approximated by capacitance circuit model that is illustrated on Fig 17 below. Capacitance C models and illustrates the capacity of the battery, while Resistance R(V,i) models and illustrates the speed of battery charging and charging dissipation. For the sake of simplification, we assume that R is always constant. The lower R, then the faster bat‐ tery charging process, at the same time it has lower dissipation. The main point, the good battery should has lower R.

**Figure 17.** Batteray charging model [8].

is still the same with the wet one, however, the dry battery uses electrolyte gel, such that

The illustration of chemical process in wet battery is shown in the following equations:

The battery perfomance is determined by at least 5 parameters: (1) the speed to store electric energy in charging process; (2) the battery capacity; (3) battery leak; (4) energy dissipation; and (5) the speed of capability for discharging to the loads. The following is discussion ac‐ cording the analysis methods for those 5 parameters. There are 2 (two) conditions, i.e. charg‐ ing and discharging, which requires different modelling. The reason to have different models for charging and discharging is to have a simplification, because the whole analysis

Charging process can be approximated by capacitance circuit model that is illustrated on Fig 17 below. Capacitance C models and illustrates the capacity of the battery, while Resistance R(V,i) models and illustrates the speed of battery charging and charging dissipation. For the sake of simplification, we assume that R is always constant. The lower R, then the faster bat‐ tery charging process, at the same time it has lower dissipation. The main point, the good

(26)

looks more dry. The components of battery are illustrated in Fig 16, as follows.

**Figure 16.** Component parts of a batteray

346 Solar Cells - Research and Application Perspectives

*5.1.1. Batteray charging model*

battery should has lower R.

components are related each other, in complex relation.

Current equation for charging model based on the capacitance circuit is:

$$\dot{\mathbf{q}} = \mathbf{C} \frac{dv}{dt} + \mathbf{G} \begin{Bmatrix} V \ \ \ \ \end{Bmatrix} \begin{Bmatrix} V \ \ \ \end{Bmatrix} \tag{27}$$

Battery charging cycle on this model, has characteristic graph, which is shown on Fig 18, below.

**Figure 18.** Charging Process [8]

When the battery is not full yet, if there is no charging current (*i = 0*), then *G(V,T,i=0)* illus‐ trates leak conductance (coefficient c), which exist on battery system in passive condition (recall Eq-1). When at *i ≠ 0*, then *i 2 R* is dissipation in charging process and *G(V,T,i ≠ 0)* is leak conductance at charging time. If the battery has already been full, then the battery voltage will be constant and the energy charging excess in charging process will change to heat dis‐ sipation, which shown by the leak conductance of *G(V,T,i=0)* function. R and G values show the degree of dis-efficiency of the battery. For the sake of electric management, it is requires accurate information according to R, C and G(V,T,i) function of baterry. The information of those 3 values can be obtained by careful measurements.

### *5.1.2. Batteray discharging model*

Discharging process can approximated by using a Constant Voltage Source Circuit Model as illustrated on Fig 19 below. The reason we use Constant Voltage Source circuit model, be‐ cause there is a fact that at condition of near to empty, the battery (without load) still has voltage that close to peak voltage when battery at the full condition. Based on this fact, it is difficult to use capacitance model for discharging process. At Constant Voltage Source Cir‐ cuit Model, VDC models Constant Voltage Source of the battery, while Resistance Rs models and represents the battery energy content. Gp represents the battery leak, which is quitely equal to G(V,T,i) function on battery charging model.

**Figure 19.** Baterry discharging model, Rs represents the battery energy content and Gp represents the battery leak.

By assumption that VDC is constant, then the battery energy *Ebatt* can be represented by the value of *RS*, which represents internal series resistance of the battery at discharging process. *RS* can be calculated by the following simple formula:

$$R\_s = \frac{V\_{Batt} - V\_{out}}{I\_{load}} \tag{28}$$

It is requires an algorithm to calculate *RS* as representation of battery stored energy at real time condition. It is for reader exercise to develop thus *Rs* as a function of battery stored en‐ ergy. The illustration of stored energy vs *Rs* is shown on Fig 20, below:

**Figure 20.** Illustration of stored energy vs Rs [9]

When the battery is not full yet, if there is no charging current (*i = 0*), then *G(V,T,i=0)* illus‐ trates leak conductance (coefficient c), which exist on battery system in passive condition

conductance at charging time. If the battery has already been full, then the battery voltage will be constant and the energy charging excess in charging process will change to heat dis‐ sipation, which shown by the leak conductance of *G(V,T,i=0)* function. R and G values show the degree of dis-efficiency of the battery. For the sake of electric management, it is requires accurate information according to R, C and G(V,T,i) function of baterry. The information of

Discharging process can approximated by using a Constant Voltage Source Circuit Model as illustrated on Fig 19 below. The reason we use Constant Voltage Source circuit model, be‐ cause there is a fact that at condition of near to empty, the battery (without load) still has voltage that close to peak voltage when battery at the full condition. Based on this fact, it is difficult to use capacitance model for discharging process. At Constant Voltage Source Cir‐ cuit Model, VDC models Constant Voltage Source of the battery, while Resistance Rs models and represents the battery energy content. Gp represents the battery leak, which is quitely

**Figure 19.** Baterry discharging model, Rs represents the battery energy content and Gp represents the battery leak.

*Batt out*


*load V V*

*I*

*s*

*R*

By assumption that VDC is constant, then the battery energy *Ebatt* can be represented by the value of *RS*, which represents internal series resistance of the battery at discharging process.

*R* is dissipation in charging process and *G(V,T,i ≠ 0)* is leak

*2*

those 3 values can be obtained by careful measurements.

equal to G(V,T,i) function on battery charging model.

*RS* can be calculated by the following simple formula:

(recall Eq-1). When at *i ≠ 0*, then *i*

348 Solar Cells - Research and Application Perspectives

*5.1.2. Batteray discharging model*

By understanding the information in this Chapter, the available information can be used to develop energy management concept for solar cell system. The main point is monitoring flow-in current to and flow-out current from batteries, which then combined with the infor‐ mation according to available energy in batteries, will give accurate information regarding to the energy in the system. The accurate information is required by the system when decid‐ ing integration process to keep maintain sustainability and also to conduct the business of selling and buying electric energy in the grid system.

### **5.2. Electronics energy management system**

There are many tasks, which have to be done, such that energy management can be con‐ ducted accurately. The following, we list the mandatory tasks required in energy manage‐ ment for solar cell system. First of all is current monitoring flow-in to and flow-out from the battery bank. The second one is measuring electrical energy content inside batteray bank by using algorithm of stored energy vs Rs. The third one is an evaluation of the internal energy condition and based on the sustanainability criteria, conducting decission process to do inte‐ gration with external system (grid). The fourth one, when integration is decided, then syn‐ chronization of frequency, phase and voltage must be conducted soon. Those four tasks require algorithms or procedures, which can be very complex for electronic analog circuits. Hence, the only way is by utilizing microprocessor circuit or even computer system to con‐ duct energy management tasks. The comprehenssive block diagram of Solar Cell Based Re‐ newable Energy Unit System has been already shown on Fig 1. The detail of the circuitries are for the reader exercise to give contributions to realize an electrical energy management for solar cell system.

## **6. Conclusion**

Electrical energy management and engineering for solar cell system is started by designing electrical energy requirements, technical specifications of solar cells and batteries, also infor‐ mation of zone latitude and statistical weather of the location. The characterizations to ob‐ tain information of solar cell and batteries efficiency are very important to support in designing the system. Furthermore, electrical energy management and engineering for solar cell system, must deal with 4 tasks listed in sub-chapter 5.2. To cover those 4 tasks, then it has to be developed a processing system based on microprocessor or even computer system. If there exist several units of autonomy solar cell systems, then they can be coordinated to build a grid energy system, which can support each other to keep maintain sustainability service. Ultimately, it can be established independent electric energy collaboration and total‐ ly eco-friendly.

## **Author details**

Purnomo Sidi Priambodo, Didik Sukoco, Wahyudi Purnomo, Harry Sudibyo and Djoko Hartanto

Universitas Indonesia, Indonesia

## **References**


[5] A. Rusdiyanto and B. Susanto, "Perancangan Inverter Sinusoida 1 Fasa dengan Apli‐ kasi Pemograman Rumus Parabola dan Segitiga Sebagai Pembangkit PWM," P2Teli‐ mek, LIPI, 2008

Hence, the only way is by utilizing microprocessor circuit or even computer system to con‐ duct energy management tasks. The comprehenssive block diagram of Solar Cell Based Re‐ newable Energy Unit System has been already shown on Fig 1. The detail of the circuitries are for the reader exercise to give contributions to realize an electrical energy management

Electrical energy management and engineering for solar cell system is started by designing electrical energy requirements, technical specifications of solar cells and batteries, also infor‐ mation of zone latitude and statistical weather of the location. The characterizations to ob‐ tain information of solar cell and batteries efficiency are very important to support in designing the system. Furthermore, electrical energy management and engineering for solar cell system, must deal with 4 tasks listed in sub-chapter 5.2. To cover those 4 tasks, then it has to be developed a processing system based on microprocessor or even computer system. If there exist several units of autonomy solar cell systems, then they can be coordinated to build a grid energy system, which can support each other to keep maintain sustainability service. Ultimately, it can be established independent electric energy collaboration and total‐

Purnomo Sidi Priambodo, Didik Sukoco, Wahyudi Purnomo, Harry Sudibyo and

[1] P.S.Priambodo, N.R. Poespawati, D. Hartanto, "Chapter book: Solar Cell Technolo‐ gy", INTECH Open Access Publisher, ISBN 978-953-307-316-3,www.intechweb.org

[2] M.A. Green, "Solar Cells, Operating Principles, Technology and System Applica‐

[3] B.M. Hasaneen and A.A.E. Mohammed, " Design and Simulation of DC/DC Boost

[4] I.F. Crowley, H.F. Leung and S.J. Bitar, "PWM Techniques: A Pure Sine Wave Inver‐

for solar cell system.

350 Solar Cells - Research and Application Perspectives

**6. Conclusion**

ly eco-friendly.

**Author details**

Djoko Hartanto

**References**

Universitas Indonesia, Indonesia

tions," Prentice Hall, ISBN 0-13-82270, 1982

Converter," 978-1-4244-1993-3/08, ©2008 IEEE

ter," Worcester Polytechnic institute, 2010-2011


## **Chapter 13**

## **Effect of Source Impedance on Hybrid Wind and Solar Power System**

Mu-Kuen Chen and Chao-Yuan Cheng

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/54495

## **1. Introduction**

Large wind turbines use mechanical systems, such as geared or gearless devices to increase the speed of the generator. In addition, an inverter is employed to adjust the output voltage to exceed the grid value. With its phase leading the bus phase, wind power can be integrated into the grid bus. The integration can be easily realized owing to negligible impedance of the utility bus. The main issues for wind-power generating systems include fluctuations in output voltage and quality of power supplied to the utility power system. In small renewable energy systems, wind power and solar energy are integrated to improve the reliability of the indi‐ vidual power system. Conventionally, the AC output voltage of the wind turbine is rectified, and then combined with the output voltage of the solar cell to charge the battery and provide power supply to the load. To characterize the battery, the Thevenin battery model considering the nonlinear effect of source impedances was proposed. A battery evaluation test system was employed to validate this model. The curve of test results follows entirely the theoretical calculation [1]-[3]. For photovoltaic application, the inner resistance of solar panel was also included in the theoretical analysis. The maximum power point tracking (MPPT) of solar panel for different insolation levels verified the proposed solar cell model. The MPPT technique adjusted continuously the battery-charging rate and obtained shorter charge time [4]. It was reported that a dual battery configuration with deep-cycle batteries can increase the available capacity. Moreover, the system may achieve optimum utilization of the PV array and proper maintenance of the storage battery [5]. In another research, a microcontroller was employed to adjust the maximum charging current according to the PV power production and battery voltage level. Using this method, better exploitation of the power produced by the PV power source can be achieved. Moreover, battery lifetime can be increased by restoring high state of charge (SOC) in short charging time [6]. Another study compared the performance of equal

© 2013 Chen and Cheng; licensee InTech. This is an open access article 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. © 2013 Chen and Cheng; licensee InTech. This is a paper 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.

rate charging, proportional rate and pulse current charging in charging multiple batteries. The total charging time is shortest when using pulse current charging strategy. All the batteries become fully charged almost simultaneously when they are charged with proportional rate or pulse current method [7]. The optimum size of the PV module for a specific wind turbine to meet the load requirement for the hybrid wind/PV system was investigated in order to minimize the overall cost of the system [8] - [9]. Nevertheless, the effect of source impedance in a small hybrid wind/PV system has not yet been investigated. In this study, theoretical analysis shows that it is difficult to obtain both wind power and solar energy at the same time by traditional methods, which is verified by field test. To overcome such problem, a micro‐ processor-based controller design for detecting instantaneous voltage variations of both energy sources is proposed, and a charge controller is employed to optimize the charging operation.

## **2. Theory and analysis**

Owing to the large variation in the wind and solar energy, the converter is employed to provide the stable power for normal application. When only one energy source supplies the load, as shown in Fig.1(a), the voltage and frequency of the converter output is adjusted to meet the load specification. In Fig. 1(b), both wind and solar energy supply the same load simultane‐ ously. In addition to the load requirement, the voltage and frequency of both converter outputs are adjusted such as the two energy sources can supply the load at the same time. In case of the DC-DC converter, only the output voltage of both converters should be adjusted to charge the same load.

In a small hybrid power system, a battery is usually utilized to store the renewable energy to improve the reliability of the system. Moreover, to simplify the power system, the power source charges directly the battery. Figure 2 shows the conventional charging system, in which the rectified DC voltage charges two batteries. In addition to source voltages *Ew*, *E1* and *E2*, charging currents *I1* and *I2* are also determined by source resistances *rw*, *r1* and *r2* for the wind power and the two batteries, respectively. The output voltage *V0* is the summation of *V0w*, *V01* and *V02* from wind power and batteries *E1* and *E2*, respectively. According to the circuit theory, the equations for *V0w*, *V01* and *V02* are as follows:

$$V\_{0w} = E\_w(\frac{r\_1 r\_2}{r\_w r\_1 + r\_1 r\_2 + r\_w r\_2}) \tag{1}$$

$$V\_{01} = E\_1(\frac{r\_2}{r\_1 + r\_2}) \tag{2}$$

$$V\_{02} = E\_2(\frac{r\_1}{r\_1 + r\_2}) \tag{3}$$

$$V\_0 = V\_{0w} + V\_{01} + V\_{02} \tag{4}$$

As mentioned, *Vow, Vo1* and *Vo2* are voltages from wind power and the two batteries, respec‐ tively, which contribute to the output voltage *Vo* independently. As mentioned, *Vow, Vo1* and *Vo2* are voltages from wind power and the two batteries, respectively, which contribute to the output

1

1 2

*VV V V* 0 0 01 02 *<sup>w</sup>* (4)

voltage *Vo* independently.

02 2

( ) *<sup>r</sup> V E r r*

Figure 1. (a) Wind or solar power supplies the load through a converter. (b) The wind and solar power use two converters to supply the load at the same time. The voltage and frequency of the converters are adjusted to meet the load requirement. **Figure 1.** (a) Wind or solar power supplies the load through a converter. (b) The wind and solar power use two con‐ verters to supply the load at the same time. The voltage and frequency of the converters are adjusted to meet the load requirement.

The following are possible charging situations. The following are possible charging situations.

> Wind Power

E <sup>w</sup>

r w

Iw

rate charging, proportional rate and pulse current charging in charging multiple batteries. The total charging time is shortest when using pulse current charging strategy. All the batteries become fully charged almost simultaneously when they are charged with proportional rate or pulse current method [7]. The optimum size of the PV module for a specific wind turbine to meet the load requirement for the hybrid wind/PV system was investigated in order to minimize the overall cost of the system [8] - [9]. Nevertheless, the effect of source impedance in a small hybrid wind/PV system has not yet been investigated. In this study, theoretical analysis shows that it is difficult to obtain both wind power and solar energy at the same time by traditional methods, which is verified by field test. To overcome such problem, a micro‐ processor-based controller design for detecting instantaneous voltage variations of both energy sources is proposed, and a charge controller is employed to optimize the charging

Owing to the large variation in the wind and solar energy, the converter is employed to provide the stable power for normal application. When only one energy source supplies the load, as shown in Fig.1(a), the voltage and frequency of the converter output is adjusted to meet the load specification. In Fig. 1(b), both wind and solar energy supply the same load simultane‐ ously. In addition to the load requirement, the voltage and frequency of both converter outputs are adjusted such as the two energy sources can supply the load at the same time. In case of the DC-DC converter, only the output voltage of both converters should be adjusted to charge

In a small hybrid power system, a battery is usually utilized to store the renewable energy to improve the reliability of the system. Moreover, to simplify the power system, the power source charges directly the battery. Figure 2 shows the conventional charging system, in which the rectified DC voltage charges two batteries. In addition to source voltages *Ew*, *E1* and *E2*, charging currents *I1* and *I2* are also determined by source resistances *rw*, *r1* and *r2* for the wind power and the two batteries, respectively. The output voltage *V0* is the summation of *V0w*, *V01* and *V02* from wind power and batteries *E1* and *E2*, respectively. According to the circuit theory,

1 2

2

1 2

( ) *w w w w*

*r r V E*

01 1

( ) *<sup>r</sup> V E*

1 12 2

*r r rr r r* <sup>=</sup> + + (1)

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

operation.

the same load.

**2. Theory and analysis**

354 Solar Cells - Research and Application Perspectives

the equations for *V0w*, *V01* and *V02* are as follows:

0


I

1 2

Vo

r I <sup>1</sup> r2

generator, solar panel and battery, respectively. The related equations are listed below.

E1 E2

e. When *V02> V0w >V01*, similar to case (d), wind power has no effect on the circuit, only *E2* charges battery *E1*.

behavior during the charging process. Cases (d) and (e) are not normal charging conditions.

As seen in above analysis, it is only in case (a) that wind power can charge both batteries at the same time. However, in case (b), when battery *E1* also charges battery *E2*, the voltage drop of *I1 r1* and *I2 r2* cause increase in *V02* and decrease in *V01* respectively. Finally, when *V01* = *V02*, the charging condition returns to case (a). Case (c) is similar to case (b), so it exhibits self-regulating

Figure 2. Wind power charges both batteries. During the charging process, the increase in charging current *I1* leads to increase in internal voltage

Figure 3(a) shows a hybrid wind and PV power generating system. *Ew*, *Ep*, *Eb*, *rw*, *rp* and *rb* are factors that determine the charging current. Similar to the above conventional charging conditions, the output voltage *V0* is made up of *Vow*, *Vop* and *Vob* from the wind

drop *I1 r1*, which raises *Vo*, and in turn increases *I2*. Therefore, this charging configuration comprises a self-regulation mechanism.

1

(3)

As mentioned, *Vow, Vo1* and *Vo2* are voltages from wind power and the two batteries, respectively, which contribute to the output

Figure 1. (a) Wind or solar power supplies the load through a converter. (b) The wind and solar power use two converters to supply the load at the

1 2

*VV V V* 0 0 01 02 *<sup>w</sup>* (4)

voltage *Vo* independently.

02 2

( ) *<sup>r</sup> V E r r*

Power

Wind Power

> Solar Array

**c.** When *V0w > V02 >V01*, similar to case (b), the charging currents from both wind power and battery E2 charge battery E1. The following are possible charging situations.

(b)

(a)

Converter Ep

Converter <sup>E</sup> <sup>w</sup>

Converter <sup>E</sup> <sup>w</sup>

Wind or Solar Load

same time. The voltage and frequency of the converters are adjusted to meet the load requirement.

Load


As seen in above analysis, it is only in case (a) that wind power can charge both batteries at the same time. However, in case (b), when battery *E1* also charges battery *E2*, the voltage drop of *I1 r1* and *I2 r2* cause increase in *V02* and decrease in *V01* respectively. Finally, when *V01* = *V02*, the charging condition returns to case (a). Case (c) is similar to case (b), so it exhibits selfregulating behavior during the charging process. Cases (d) and (e) are not normal charging conditions. d. When *V01 > V0w >V02*, wind power has no effect on the circuit, only battery *E1* charges battery *E2*. e. When *V02> V0w >V01*, similar to case (d), wind power has no effect on the circuit, only *E2* charges battery *E1*. As seen in above analysis, it is only in case (a) that wind power can charge both batteries at the same time. However, in case (b), when battery *E1* also charges battery *E2*, the voltage drop of *I1 r1* and *I2 r2* cause increase in *V02* and decrease in *V01* respectively. Finally, when *V01* = *V02*, the charging condition returns to case (a). Case (c) is similar to case (b), so it exhibits self-regulating behavior during the charging process. Cases (d) and (e) are not normal charging conditions.

Figure 2. Wind power charges both batteries. During the charging process, the increase in charging current *I1* leads to increase in internal voltage **Figure 2.** Wind power charges both batteries. During the charging process, the increase in charging current *I1* leads to increase in internal voltage drop *I1 r1*, which raises *Vo*, and in turn increases *I2*. Therefore, this charging configuration comprises a self-regulation mechanism.

Figure 3(a) shows a hybrid wind and PV power generating system. *Ew*, *Ep*, *Eb*, *rw*, *rp* and *rb* are factors that determine the charging current. Similar to the above conventional charging conditions, the output voltage *V0* is made up of *Vow*, *Vop* and *Vob* from the wind Figure 3(a) shows a hybrid wind and PV power generating system. *Ew*, *Ep*, *Eb*, *rw*, *rp* and *rb* are factors that determine the charging current. Similar to the above conventional charging conditions, the output voltage *V0* is made up of *Vow*, *Vop* and *Vob* from the wind generator, solar panel and battery, respectively. The related equations are listed below.

$$V\_{ow} = E\_w(\frac{r\_b}{r\_w + r\_b}) \tag{5}$$

generator, solar panel and battery, respectively. The related equations are listed below.

drop *I1 r1*, which raises *Vo*, and in turn increases *I2*. Therefore, this charging configuration comprises a self-regulation mechanism.

$$V\_{op} = E\_p(\frac{r\_b}{r\_p + r\_b}) \tag{6}$$

$$V\_o = E\_b + V\_{ow} + V\_{op} \tag{7}$$

Possible charging conditions are as follows:

**c.** When *V0w > V02 >V01*, similar to case (b), the charging currents from both wind power and

(b)

(a)

Converter Ep

Converter <sup>E</sup> <sup>w</sup>

Converter <sup>E</sup> <sup>w</sup>

Wind or Solar Load

The following are possible charging situations.

same time. The voltage and frequency of the converters are adjusted to meet the load requirement.

Load

**d.** When *V01 > V0w >V02*, wind power has no effect on the circuit, only battery *E1* charges battery

**e.** When *V02> V0w >V01*, similar to case (d), wind power has no effect on the circuit, only *E2*

As seen in above analysis, it is only in case (a) that wind power can charge both batteries at the same time. However, in case (b), when battery *E1* also charges battery *E2*, the voltage drop of *I1 r1* and *I2 r2* cause increase in *V02* and decrease in *V01* respectively. Finally, when *V01* = *V02*, the charging condition returns to case (a). Case (c) is similar to case (b), so it exhibits selfregulating behavior during the charging process. Cases (d) and (e) are not normal charging

r w

Iw

**Figure 2.** Wind power charges both batteries. During the charging process, the increase in charging current *I1* leads to increase in internal voltage drop *I1 r1*, which raises *Vo*, and in turn increases *I2*. Therefore, this charging configuration

Figure 3(a) shows a hybrid wind and PV power generating system. *Ew*, *Ep*, *Eb*, *rw*, *rp* and *rb* are factors that determine the charging current. Similar to the above conventional charging conditions, the output voltage *V0* is made up of *Vow*, *Vop* and *Vob* from the wind generator, solar

( ) *<sup>b</sup>*

*w b*

( ) *<sup>b</sup>*

*p b*

1 2

Vo

r I <sup>1</sup> r2

*r r* <sup>=</sup> <sup>+</sup> (5)

*r r* <sup>=</sup> <sup>+</sup> (6)

*V EV V o b ow op* =+ + (7)

I

E1 E2

battery E2 charge battery E1.

356 Solar Cells - Research and Application Perspectives

Wind Power

E <sup>w</sup>

panel and battery, respectively. The related equations are listed below.

*ow w*

*op p*

*<sup>r</sup> V E*

*<sup>r</sup> V E*

1

(3)

As mentioned, *Vow, Vo1* and *Vo2* are voltages from wind power and the two batteries, respectively, which contribute to the output

1 2

*VV V V* 0 0 01 02 *<sup>w</sup>* (4)

voltage *Vo* independently.

02 2

( ) *<sup>r</sup> V E r r*

Power

Wind Power

> Solar Array

charges battery *E1*.

comprises a self-regulation mechanism.

*E2*.

conditions.


d. When *V01 > V0w >V02*, wind power has no effect on the circuit, only battery *E1* charges battery *E2*. e. When *V02> V0w >V01*, similar to case (d), wind power has no effect on the circuit, only *E2* charges battery *E1*. As seen in above analysis, it is only in case (a) that wind power can charge both batteries at the same time. However, in case (b), when battery *E1* also charges battery *E2*, the voltage drop of *I1 r1* and *I2 r2* cause increase in *V02* and decrease in *V01* respectively. Finally, when *V01* = *V02*, the charging condition returns to case (a). Case (c) is similar to case (b), so it exhibits self-regulating behavior during the charging process. Cases (d) and (e) are not normal charging conditions. From the above analysis, in cases (a) and (b), there are two energy sources charging a battery at the same time. However, in case (a), the larger current *Iw* from wind energy may result in a larger internal voltage drop *Ib rb* of the battery. Therefore, when *Vow > Vop*, the charging condition becomes case (c), and solar energy cannot be utilized to charge the battery. Case (b) shows the same behavior. Contrary to the conventional charging system, the hybrid charging system exhibits a competition effect, meaning that only the larger power source can dominate the charging system.

> Figure 3(b) shows the I-V curves of the charging system. Because the source resistance of the wind generator is much smaller than that of the solar panel, the wind I-V curve reduces slowly with increase in charging current. The source resistance of the battery is also much smaller than that of the wind generator. Therefore, the terminal voltage of the battery Vo increases only slightly with increase in charging current. When the solar I-V curve drops to P, which is equal to the terminal voltage of the battery, the solar energy stops charging the battery.

> Figure 3(c) shows the V-T curves of the charging system. Before time To, the battery is in undercharge condition, both power sources behave as the current sources with their ratio of output currents proportional to that of generated power levels. For time To to Tf, the source resistance of the solar panel lowers gradually the solar charging current and the battery terminal voltage Vo increases slowly. Finally, at time Tf , only wind power can charge the battery.

Figure 2. Wind power charges both batteries. During the charging process, the increase in charging current *I1* leads to increase in internal voltage drop *I1 r1*, which raises *Vo*, and in turn increases *I2*. Therefore, this charging configuration comprises a self-regulation mechanism. Figure 3(a) shows a hybrid wind and PV power generating system. *Ew*, *Ep*, *Eb*, *rw*, *rp* and *rb* are factors that determine the charging current. Similar to the above conventional charging conditions, the output voltage *V0* is made up of *Vow*, *Vop* and *Vob* from the wind generator, solar panel and battery, respectively. The related equations are listed below. To improve the performance of the hybrid power generating system shown above, a switch control is employed. It is connected to the battery circuit as shown in Fig. 4. In this operation mode, both wind and solar energy can be utilized, although only one energy source can charge the battery at any time. Owing to the different characteristics of wind and solar energy, as shown in Eq. (10), we can adjust the charging duty cycle ratio *k* of the two energy sources to obtain maximum energy in the battery. The equations are listed below.

$$V\_{ov} = E\_b + E\_w \left(\frac{r\_b}{r\_b + r\_w}\right) \tag{8}$$

$$V\_{op} = E\_b + E\_p(\frac{r\_b}{r\_b + r\_p}) \tag{9}$$

$$\mathcal{W} = k\mathcal{W}\_w + (1 - k)\mathcal{W}\_p \tag{10}$$

Figure 3. (a) Small hybrid wind and PV energy charging system. Two power sources charge a battery. Owing to the internal voltage drop caused by impedance of wind, solar and battery power source, only one power source can contribute to the charging process. (b) I-V characteristics of the charging system. When the solar I-V curve drops to point P, solar power cannot charge the battery. (c) Simplified charging curve of the system. Two current sources charge the battery before time To. Then, charging speed reduces because of increase in resistance of the solar power circuit. After time Tf, only wind power charges the battery. **Figure 3.** (a) Small hybrid wind and PV energy charging system. Two power sources charge a battery. Owing to the internal voltage drop caused by impedance of wind, solar and battery power source, only one power source can con‐ tribute to the charging process. (b) I-V characteristics of the charging system. When the solar I-V curve drops to point P, solar power cannot charge the battery. (c) Simplified charging curve of the system. Two current sources charge the battery before time To. Then, charging speed reduces because of increase in resistance of the solar power circuit. After time Tf, only wind power charges the battery.

<sup>E</sup> <sup>w</sup> <sup>r</sup> <sup>w</sup> <sup>V</sup>

Iw

Switch

ow op V r <sup>p</sup>

Ip

Ep

Solar Array

energy charging. In view of this, we can adjust the wind power charging duration to obtain the maximum energy.

Figure 4. Switch-controlled wind and PV energy charging system. The wind and solar power charge a battery alternately. Both power sources can charge the same battery at different times. During solar energy charging, mechanical energy generated by inertia of the wind turbine will be stored and employed to charge the battery during wind energy charging. On the other hand, solar energy cannot be stored but will be lost during wind

b b r

I

Eb

Wind Power Figure 3. (a) Small hybrid wind and PV energy charging system. Two power sources charge a battery. Owing to the internal voltage drop caused by impedance of wind, solar and battery power source, only one power source can contribute to the charging process. (b) I-V characteristics of the

To Tf T

(c)

(a)

Ep - Ip rp

P

Switch

ow op V r <sup>p</sup>

Ip

Ep

Solar Array

I

b b r

I

<sup>E</sup> <sup>w</sup> <sup>r</sup> <sup>w</sup> <sup>V</sup>

Vo

Iw

Eb

Ew - Iw rw

Wind Power

V

Vo

V

(b)

Ep - Ip rp

Ew - Iw rw

time Tf, only wind power charges the battery.

Figure 4. Switch-controlled wind and PV energy charging system. The wind and solar power charge a battery alternately. Both power sources can charge the same battery at different times. During solar energy charging, mechanical energy generated by inertia of the wind turbine will be stored and employed to charge the battery during wind energy charging. On the other hand, solar energy cannot be stored but will be lost during wind energy charging. In view of this, we can adjust the wind power charging duration to obtain the maximum energy. **Figure 4.** Switch-controlled wind and PV energy charging system. The wind and solar power charge a battery alter‐ nately. Both power sources can charge the same battery at different times. During solar energy charging, mechanical energy generated by inertia of the wind turbine will be stored and employed to charge the battery during wind ener‐ gy charging. On the other hand, solar energy cannot be stored but will be lost during wind energy charging. In view of this, we can adjust the wind power charging duration to obtain the maximum energy.

To overcome the drawbacks of the hybrid wind and PV charging system shown above, a microprocessor-controlled power generating system, as shown in Fig.5, is proposed. The different charging modes, which vary with the weather conditions to obtain the maximum energy, are shown in Table 1. With both energy sources, the system operates in the independent charging mode. The wind and solar energy charge batteries *Ebw* and *Ebp*, respectively. If there is only one energy source, the system runs in the hybrid charging mode. Either energy source can charge batteries *Ebw* and *Ebp* simultaneously. Owing to the instability of wind energy, if both energy sources co-exist, wind energy exceeds the threshold value, the charging system runs in the wind-enhanced mode. In this case, not only can both energy sources be employed to improve the reliability of power system, the fluctuations in the small wind power generating system can also be reduced. To overcome the drawbacks of the hybrid wind and PV charging system shown above, a microprocessor-controlled power generating system, as shown in Fig.5, is proposed. The different charging modes, which vary with the weather conditions to obtain the maximum energy, are shown in Table 1. With both energy sources, the system operates in the independent charging mode. The wind and solar energy charge batteries *Ebw* and *Ebp*, respectively. If there is only one energy source, the system runs in the hybrid charging mode. Either energy source can charge batteries *Ebw* and *Ebp* simultaneously. Owing to the instability of wind energy, if both energy sources co-exist, wind energy exceeds the threshold value, the charging system runs in the wind-enhanced mode. In this case, not only can both energy sources be employed to improve the reliability of power system, the fluctuations in the small wind power generating system can also be reduced.

charging system. When the solar I-V curve drops to point P, solar power cannot charge the battery. (c) Simplified charging curve of the system. Two current sources charge the battery before time To. Then, charging speed reduces because of increase in resistance of the solar power circuit. After Figure 5. Microprocessor-controlled wind and PV energy charging system. Both power sources charge the two batteries. According to the wind and solar energy conditions, the controller regulates the charging conditions for *Ebw* and *Ebp*. There is no power loss in this charging system. When there is only one power source, it can charge both batteries. With both power sources and high wind energy, the excess wind energy charges battery *Ebp*. The controller improves greatly the reliability of this charging system. **Figure 5.** Microprocessor-controlled wind and PV energy charging system. Both power sources charge the two batter‐ ies. According to the wind and solar energy conditions, the controller regulates the charging conditions for *Ebw* and *Ebp*. There is no power loss in this charging system. When there is only one power source, it can charge both batteries. With both power sources and high wind energy, the excess wind energy charges battery *Ebp*. The controller improves greatly the reliability of this charging system.

**3. Results and discussion** 

can be determined.

Ib

A

V 0

(a)

Eb

rb

**3.1. Source resistance measurements** 

current *Ip*, the solar internal resistance can be determined.

Ep

rp Ip

R

L

Ebp : Solar Battery

(b)

A

Wind Energy Solar Energy Energy Source

● : Independent Charging

Solar and Wind Energy (high wind speed) Solar and Wind Energy (low wind speed)

respectively the corresponding battery. Under hybrid charging condition, both energy sources charge the two batteries simultaneously.

Ebw : Wind Battery

Table 1. Microprocessor – controlled charging modes, Under independent charging condition, the solar energy and wind energy charges

To measure battery source resistance, as shown in Fig. 6(a), a 12-V/75-AH battery supplied the load through a switch. We adjusted the load resistance *RL* to change the battery discharging current *Ib*. From the voltage difference *Eb – Vo* and *Ib*, the source resistance *rb*

Because the solar energy is much smaller than the battery energy and the solar cell internal resistance *rp* is much larger than the battery source resistance *rb* , as shown in Fig. 6(b), the output terminal of the solar panel is directly grounded to measure the solar charging current *Ip*. A 75-W solar cell panel was used in the experiment performed outdoors. From the solar voltage *Ep* and the

▲ : Hybrid Charging

▲ ●

▲ ▲

Ebw

▲ ●

▲ ▲

Ebp

Figure 3. (a) Small hybrid wind and PV energy charging system. Two power sources charge a battery. Owing to the internal voltage drop caused by impedance of wind, solar and battery power source, only one power source can contribute to the charging process. (b) I-V characteristics of the

Figure 4. Switch-controlled wind and PV energy charging system. The wind and solar power charge a battery alternately. Both power sources can charge the same battery at different times. During solar energy charging, mechanical energy generated by inertia of the wind turbine will be stored and employed to charge the battery during wind energy charging. On the other hand, solar energy cannot be stored but will be lost during wind

time Tf, only wind power charges the battery.

<sup>E</sup> <sup>w</sup> <sup>r</sup> <sup>w</sup> <sup>V</sup>

Iw

Switch

ow op V r <sup>p</sup>

To Tf T

(c)

**Figure 3.** (a) Small hybrid wind and PV energy charging system. Two power sources charge a battery. Owing to the internal voltage drop caused by impedance of wind, solar and battery power source, only one power source can con‐ tribute to the charging process. (b) I-V characteristics of the charging system. When the solar I-V curve drops to point P, solar power cannot charge the battery. (c) Simplified charging curve of the system. Two current sources charge the battery before time To. Then, charging speed reduces because of increase in resistance of the solar power circuit. After

Ip

Ep

Solar Array

energy charging. In view of this, we can adjust the wind power charging duration to obtain the maximum energy.

b b r

I

(a)

Ep - Ip rp

P

Switch

ow op V r <sup>p</sup>

Ip

Ep

Solar Array

I

b b r

I

<sup>E</sup> <sup>w</sup> <sup>r</sup> <sup>w</sup> <sup>V</sup>

Vo

Iw

Eb

Ew - Iw rw

Wind Power

V

358 Solar Cells - Research and Application Perspectives

Vo

time Tf, only wind power charges the battery.

V

(b)

Ep - Ip rp

Ew - Iw rw

Eb

Wind Power Ebw

Ibw

<sup>E</sup> <sup>w</sup> <sup>r</sup> <sup>w</sup> <sup>V</sup> Iw

wind power generating system can also be reduced.

Controller

Ebp

*Ebp*. The controller improves greatly the reliability of this charging system.

rI r bw bp bp

ow Vop r <sup>p</sup>

Ip

Solar Array

Ep

Wind Power


and solar energy conditions, the controller regulates the charging conditions for *Ebw* and *Ebp*. There is no power loss in this charging system. When there is only one power source, it can charge both batteries. With both power sources and high wind energy, the excess wind energy charges battery

ow Vop r <sup>p</sup>

Ip

Solar Array

Ep

rI r bw bp bp

wind power generating system can also be reduced.

Ibw

<sup>E</sup> <sup>w</sup> <sup>r</sup> <sup>w</sup> <sup>V</sup> Iw

To overcome the drawbacks of the hybrid wind and PV charging system shown above, a microprocessor-controlled power generating system, as shown in Fig.5, is proposed. The different charging modes, which vary with the weather conditions to obtain the maximum energy, are shown in Table 1. With both energy sources, the system operates in the independent charging mode. The wind and solar energy charge batteries *Ebw* and *Ebp*, respectively. If there is only one energy source, the system runs in the hybrid charging mode. Either energy source can charge batteries *Ebw* and *Ebp* simultaneously. Owing to the instability of wind energy, if both energy sources co-exist, wind energy exceeds the threshold value, the charging system runs in the wind-enhanced mode. In this case, not only can both energy sources be employed to improve the reliability of power system, the fluctuations in the small

> To overcome the drawbacks of the hybrid wind and PV charging system shown above, a microprocessor-controlled power generating system, as shown in Fig.5, is proposed. The different charging modes, which vary with the weather conditions to obtain the maximum energy, are shown in Table 1. With both energy sources, the system operates in the independent charging mode. The wind and solar energy charge batteries *Ebw* and *Ebp*, respectively. If there is only one energy source, the system runs in the hybrid charging mode. Either energy source can charge batteries *Ebw* and *Ebp* simultaneously. Owing to the instability of wind energy, if both energy sources co-exist, wind energy exceeds the threshold value, the charging system runs in the wind-enhanced mode. In this case, not only can both energy sources be employed to improve the reliability of power system, the fluctuations in the small

respectively the corresponding battery. Under hybrid charging condition, both energy sources charge the two batteries simultaneously. **3. Results and discussion Table 1.** Microprocessor – controlled charging modes, Under independent charging condition, the solar energy and wind energy charges respectively the corresponding battery. Under hybrid charging condition, both energy sources charge the two batteries simultaneously. Solar and Wind Energy (high wind speed) Solar and Wind Energy (low wind speed) Wind Energy ▲ ● ▲ ● ▲ ▲

Table 1. Microprocessor – controlled charging modes, Under independent charging condition, the solar energy and wind energy charges

#### **3.1. Source resistance measurements 3. Results and discussion** Table 1. Microprocessor – controlled charging modes, Under independent charging condition, the solar energy and wind energy charges

Eb

rb

(a)

#### the load resistance *RL* to change the battery discharging current *Ib*. From the voltage difference *Eb – Vo* and *Ib*, the source resistance *rb* can be determined. **3.1. Source resistance measurements**

Because the solar energy is much smaller than the battery energy and the solar cell internal resistance *rp* is much larger than the battery source resistance *rb* , as shown in Fig. 6(b), the output terminal of the solar panel is directly grounded to measure the solar charging current *Ip*. A 75-W solar cell panel was used in the experiment performed outdoors. From the solar voltage *Ep* and the current *Ip*, the solar internal resistance can be determined. To measure battery source resistance, as shown in Fig. 6(a), a 12-V/75-AH battery supplied the load through a switch. We adjusted the load resistance *RL* to change the battery discharging current *Ib*. From the voltage difference *Eb – Vo* and *Ib*, the source resistance *rb* can be determined. **3. Results and discussion 3.1. Source resistance measurements** 

To measure battery source resistance, as shown in Fig. 6(a), a 12-V/75-AH battery supplied the load through a switch. We adjusted

Ebp : Solar Battery ● : Independent Charging

respectively the corresponding battery. Under hybrid charging condition, both energy sources charge the two batteries simultaneously.

Ebw : Wind Battery

▲ : Hybrid Charging

R Ib V 0 A Ep L rp Ip A Because the solar energy is much smaller than the battery energy and the solar cell internal resistance *rp* is much larger than the battery source resistance *rb*, as shown in Fig. 6(b), the output terminal of the solar panel is directly grounded to measure the solar charging current *Ip*. A 75- W solar cell panel was used in the experiment performed outdoors. From the solar voltage *Ep* and the current *Ip*, the solar internal resistance can be determined. To measure battery source resistance, as shown in Fig. 6(a), a 12-V/75-AH battery supplied the load through a switch. We adjusted the load resistance *RL* to change the battery discharging current *Ib*. From the voltage difference *Eb – Vo* and *Ib*, the source resistance *rb* can be determined. Because the solar energy is much smaller than the battery energy and the solar cell internal resistance *rp* is much larger than the battery source resistance *rb* , as shown in Fig. 6(b), the output terminal of the solar panel is directly grounded to measure the solar charging current *Ip*. A 75-W solar cell panel was used in the experiment performed outdoors. From the solar voltage *Ep* and the

current *Ip*, the solar internal resistance can be determined.

**Figure 6.** (a) Circuit for measuring battery source resistance. Because much energy is stored in the battery *Eb* and the battery source resistance *rb* is small, *RL* is employed to limit the discharging current *Ib*. (b) Circuit for measuring solar cell internal resistance. Solar current *Ip* is varied by adjusting the orientation of the solar cell panel relative to the sun.

As shown in Fig. 7(a), the source resistance of battery decreases with increasing discharge current. The battery source resistance is about 0.02-0.12 Ω for discharging current 1-13 A. The power loss of the source resistance results in temperature rise of the battery. Hence, chemical reaction proceeds more easily with increasing charging current and the resistance to battery charging is reduced.

\$%&?>temperature<\$%&?>rise<\$%&?>of<\$%&?>the<\$%&?>battery.<\$%&?>Hence,<\$%&?>chemical<\$%&?>reaction<\$%&?>procee ds<\$%&?>more<\$%&?>easily<\$%&?>with<\$%&?>increasing<\$%&?>charging<\$%&?>current<\$%&?>and<\$%&?>the<\$%&?>resistan

circuit<\$%&?>current<\$%&?>However;<\$%&?>the<\$%&?>source<\$%&?>resistance<\$%&?>of<\$%&?>the<\$%&?>solar<\$%&?>cell<\$

Figure 5. Microprocessor-controlled wind and PV energy charging system. Both power sources charge the two batteries. According to the wind and solar energy conditions, the controller regulates the charging conditions for *Ebw* and *Ebp*. There is no power loss in this charging system. When there is only one power source, it can charge both batteries. With both power sources and high wind energy, the excess wind energy charges battery The source resistance of solar cell panel, as shown in Fig. 7(b), also decreases with increasing short-circuit current However; the source resistance of the solar cell panel is much larger than that of battery. Even though the area of the solar cell panel is large, the thickness of the solar cell structure is too small to increase the efficiency of optical absorption. Conventionally, the thickness of the solar cell active layer is in the micrometer range. Moreover, the resistivity of solar cell is large, which results in high source resistance. %&?>panel<\$%&?>is<\$%&?>much<\$%&?>larger<\$%&?>than<\$%&?>that<\$%&?>of<\$%&?>battery.<\$%&?>Even<\$%&?>though<\$ %&?>the<\$%&?>area<\$%&?>of<\$%&?>the<\$%&?>solar<\$%&?>cell<\$%&?>panel<\$%&?>is<\$%&?>large,<\$%&?>the<\$%&?>thicknes s<\$%&?>of<\$%&?>the<\$%&?>solar<\$%&?>cell<\$%&?>structure<\$%&?>is<\$%&?>too<\$%&?>small<\$%&?>to<\$%&?>increase<\$%&? >the<\$%&?>efficiency<\$%&?>of<\$%&?>optical<\$%&?>absorption.<\$%&?>Conventionally,<\$%&?>the<\$%&?>thickness<\$%&?>of<\$ %&?>the<\$%&?>solar<\$%&?>cell<\$%&?>active<\$%&?>layer<\$%&?>is<\$%&?>in<\$%&?>the<\$%&?>micrometer<\$%&?>range.<\$%& ?>Moreover,<\$%&?>the<\$%&?>resistivity<\$%&?>of<\$%&?>solar<\$%&?>cell<\$%&?>is<\$%&?>large,<\$%&?>which<\$%&?>results<\$ %&?>in<\$%&?>high<\$%&?>source<\$%&?>resistance. The<\$%&?>stator<\$%&?>of<\$%&?>PMG<\$%&?>has<\$%&?>36<\$%&?>slots<\$%&?>wound<\$%&?>with<\$%&?>40<\$%&?>turns<\$%

ce<\$%&?>to<\$%&?>battery<\$%&?>charging<\$%&?>is<\$%&?>reduced.

To overcome the drawbacks of the hybrid wind and PV charging system shown above, a microprocessor-controlled power generating system, as shown in Fig.5, is proposed. The different charging modes, which vary with the weather conditions to obtain the maximum energy, are shown in Table 1. With both energy sources, the system operates in the independent charging mode. The wind and solar energy charge batteries *Ebw* and *Ebp*, respectively. If there is only one energy source, the system runs in the hybrid charging mode. Either energy source can charge batteries *Ebw* and *Ebp* simultaneously. Owing to the instability of wind energy, if both energy sources co-exist, wind energy exceeds the threshold value, the charging system runs in the wind-enhanced mode. In this case, not only can both energy sources be employed to improve the reliability of power system, the fluctuations in the small

Figure 5. Microprocessor-controlled wind and PV energy charging system. Both power sources charge the two batteries. According to the wind and solar energy conditions, the controller regulates the charging conditions for *Ebw* and *Ebp*. There is no power loss in this charging system. When there is only one power source, it can charge both batteries. With both power sources and high wind energy, the excess wind energy charges battery

Ip

ow Vop r <sup>p</sup>

rI r bw bp bp

Ebp

Controller

wind power generating system can also be reduced.

Table 1. Microprocessor – controlled charging modes, Under independent charging condition, the solar energy and wind energy charges

*Ebp*. The controller improves greatly the reliability of this charging system.

**Table 1.** Microprocessor – controlled charging modes, Under independent charging condition, the solar energy and wind energy charges respectively the corresponding battery. Under hybrid charging condition, both energy sources

To measure battery source resistance, as shown in Fig. 6(a), a 12-V/75-AH battery supplied the load through a switch. We adjusted the load resistance *RL* to change the battery discharging current *Ib*. From the voltage difference *Eb – Vo* and *Ib*, the source resistance *rb*

Because the solar energy is much smaller than the battery energy and the solar cell internal resistance *rp* is much larger than the battery source resistance *rb* , as shown in Fig. 6(b), the output terminal of the solar panel is directly grounded to measure the solar charging current *Ip*. A 75-W solar cell panel was used in the experiment performed outdoors. From the solar voltage *Ep* and the

To measure battery source resistance, as shown in Fig. 6(a), a 12-V/75-AH battery supplied the load through a switch. We adjusted the load resistance *RL* to change the battery discharging current *Ib*. From the voltage difference *Eb – Vo* and *Ib*, the source resistance *rb* can be determined. Because the solar energy is much smaller than the battery energy and the solar cell internal resistance *rp* is much larger than the battery source resistance *rb*, as shown in Fig. 6(b), the output terminal of the solar panel is directly grounded to measure the solar charging current *Ip*. A 75- W solar cell panel was used in the experiment performed outdoors. From the solar voltage *Ep*

Ebw : Wind Battery

▲ : Hybrid Charging

▲ ●

Wind Energy Solar Energy Energy Source

Ebp : Solar Battery ● : Independent Charging

(b)

A

▲ ▲

Ebw

Solar Array

Ep

▲ ●

Solar and Wind Energy (high wind speed) Solar and Wind Energy (low wind speed)

respectively the corresponding battery. Under hybrid charging condition, both energy sources charge the two batteries simultaneously.

Ebw : Wind Battery

Table 1. Microprocessor – controlled charging modes, Under independent charging condition, the solar energy and wind energy charges

To measure battery source resistance, as shown in Fig. 6(a), a 12-V/75-AH battery supplied the load through a switch. We adjusted the load resistance *RL* to change the battery discharging current *Ib*. From the voltage difference *Eb – Vo* and *Ib*, the source resistance *rb*

Because the solar energy is much smaller than the battery energy and the solar cell internal resistance *rp* is much larger than the battery source resistance *rb* , as shown in Fig. 6(b), the output terminal of the solar panel is directly grounded to measure the solar charging current *Ip*. A 75-W solar cell panel was used in the experiment performed outdoors. From the solar voltage *Ep* and the

▲ : Hybrid Charging

▲ ●

▲ ▲

Ebw

▲ ●

▲ ▲

Ebp

▲ ▲

Ebp

To overcome the drawbacks of the hybrid wind and PV charging system shown above, a microprocessor-controlled power generating system, as shown in Fig.5, is proposed. The different charging modes, which vary with the weather conditions to obtain the maximum energy, are shown in Table 1. With both energy sources, the system operates in the independent charging mode. The wind and solar energy charge batteries *Ebw* and *Ebp*, respectively. If there is only one energy source, the system runs in the hybrid charging mode. Either energy source can charge batteries *Ebw* and *Ebp* simultaneously. Owing to the instability of wind energy, if both energy sources co-exist, wind energy exceeds the threshold value, the charging system runs in the wind-enhanced mode. In this case, not only can both energy sources be employed to improve the reliability of power system, the fluctuations in the small

wind power generating system can also be reduced.

Controller

Ebp

360 Solar Cells - Research and Application Perspectives

*Ebp*. The controller improves greatly the reliability of this charging system.

rI r bw bp bp

ow Vop r <sup>p</sup>

Ip

Solar Array

Wind Power

> Ebp : Solar Battery ● : Independent Charging

> > (b)

A

(a)

the solar internal resistance can be determined.

R

L

current *Ip*, the solar internal resistance can be determined.

Ep

**Figure 6.** (a) Circuit for measuring battery source resistance. Because much energy is stored in the battery *Eb* and the battery source resistance *rb* is small, *RL* is employed to limit the discharging current *Ib*. (b) Circuit for measuring solar cell internal resistance. Solar current *Ip* is varied by adjusting the orientation of the solar cell panel relative to the sun.

As shown in Fig. 7(a), the source resistance of battery decreases with increasing discharge current. The battery source resistance is about 0.02-0.12 Ω for discharging current 1-13 A. The power loss of the source resistance results in temperature rise of the battery. Hence, chemical reaction proceeds more easily with increasing charging current and the resistance to battery

rp Ip

**3.1. Source resistance measurements** 

A

can be determined.

Ib

V 0

**3. Results and discussion** 

Wind Energy Solar Energy Energy Source

Solar and Wind Energy (high wind speed) Solar and Wind Energy (low wind speed)

Ebw

Ibw

<sup>E</sup> <sup>w</sup> <sup>r</sup> <sup>w</sup> <sup>V</sup> Iw

respectively the corresponding battery. Under hybrid charging condition, both energy sources charge the two batteries simultaneously.

Ep

Ebw

**3. Results and discussion** 

can be determined.

Ib

A

V 0

(a)

charging is reduced.

and the current *Ip*,

Eb

rb

**3.1. Source resistance measurements** 

**3. Results and discussion**

charge the two batteries simultaneously.

current *Ip*, the solar internal resistance can be determined.

**3.1. Source resistance measurements**

Ep

Eb

rb

rp Ip

R

L

Ibw

<sup>E</sup> <sup>w</sup> <sup>r</sup> <sup>w</sup> <sup>V</sup> Iw

Wind Power

> The stator of PMG has 36 slots wound with 40 turns of 0.8-Φ copper wires. The source resistance (one phase) of PMG is found to be 0.5 Ω. It is much smaller than the source resistance of the solar cell panel (about 6-18 Ω). In general, the difference in voltage between wind energy sourc*e Ew* and solar energy source *Ep* is not large. According to Eqs. (5) and (6), when the wind turbine is in operation, the wind output *Vow* is much larger than the solar output *Vop.* Therefore, the wind turbine dominates the battery charging behavior. &?>of<\$%&?>0.8- Φ<\$%&?>copper<\$%&?>wires.<\$%&?>The<\$%&?>source<\$%&?>resistance<\$%&?>(one<\$%&?>phase)<\$%&?>of<\$%&?>PMG<\$%& ?>is<\$%&?>found<\$%&?>to<\$%&?>be<\$%&?>0.5<\$%&?>Ω.<\$%&?>It<\$%&?>is<\$%&?>much<\$%&?>smaller<\$%&?>than<\$%&?>th e<\$%&?>source<\$%&?>resistance<\$%&?>of<\$%&?>the<\$%&?>solar<\$%&?>cell<\$%&?>panel<\$%&?>(about<\$%&?>6- 18<\$%&?>Ω).<\$%&?>In<\$%&?>general,<\$%&?>the<\$%&?>difference<\$%&?>in<\$%&?>voltage<\$%&?>between<\$%&?>wind<\$%&? >energy<\$%&?>source*<\$%&?>Ew*<\$%&?>and<\$%&?>solar<\$%&?>energy<\$%&?>source<\$%&?>*Ep*<\$%&?>is<\$%&?>not<\$%&?>larg e.<\$%&?>According<\$%&?>to<\$%&?>Eqs.<\$%&?>(5)<\$%&?>and<\$%&?>(6),<\$%&?>when<\$%&?>the<\$%&?>wind<\$%&?>turbine< \$%&?>is<\$%&?>in<\$%&?>operation,<\$%&?>the<\$%&?>wind<\$%&?>output<\$%&?>*Vow*<\$%&?>is<\$%&?>much<\$%&?>larger<\$%& ?>than<\$%&?>the<\$%&?>solar<\$%&?>output<\$%&?>*Vop.*<\$%&?>Therefore,<\$%&?>the<\$%&?>wind<\$%&?>turbine<\$%&?>domina

> > tes<\$%&?>the<\$%&?>battery<\$%&?>charging<\$%&?>behavior.

Figure 7. Under<\$%&?>conventional<\$%&?>wind<\$%&?>speed,<\$%&?>the<\$%&?>source<\$%&?>impedance<\$%&?>of<\$%&?>the<\$%&?>PMG<\$ %&?>comes<\$%&?>from<\$%&?>the<\$%&?>resistance<\$%&?>of<\$%&?>copper<\$%&?>winding<\$%&?>of<\$%&?>the<\$%&?>stator.<\$%&?>Copper< \$%&?>is<\$%&?>a<\$%&?>good<\$%&?>conductor.<\$%&?>Therefore,<\$%&?>there<\$%&?>is<\$%&?>only<\$%&?>a<\$%&?>small<\$%&?>variation<\$% &?>in<\$%&?>resistance<\$%&?>when<\$%&?>the<\$%&?>generator<\$%&?>current<\$%&?>increases.<\$%&?>In<\$%&?>(a),<\$%&?>the<\$%&?>battery <\$%&?>stores<\$%&?>the<\$%&?>chemical<\$%&?>energy.<\$%&?>Under<\$%&?>loaded<\$%&?>condition,<\$%&?>thermal<\$%&?>effect<\$%&?>incre ases<\$%&?>with<\$%&?>the<\$%&?>current<\$%&?>in<\$%&?>the<\$%&?>battery,<\$%&?>which<\$%&?>speeds<\$%&?>up<\$%&?>the<\$%&?>chemica **Figure 7.** Under conventional wind speed, the source impedance of the PMG comes from the resistance of copper winding of the stator. Copper is a good conductor. Therefore, there is only a small variation in resistance when the generator current increases. In (a), the battery stores the chemical energy. Under loaded condition, thermal effect in‐ creases with the current in the battery, which speeds up the chemical reaction. Hence, the source resistance of the battery decreases with the current. However, the solar cell is made up of semiconductors. In addition to the high re‐ sistance of semiconductors, there is also a large variation in resistance of the solar cell when the current increases, as shown in (b).

### **3.2. Conventional hybrid wind and PV power generating system**

In this study, a 250-W permanent magnet alternator driven by wind turbine and a 12- V/75-W solar cell panel were used as the wind and solar energy source, respectively. Both energy sources were output to 12-V/75-AH lead batteries, which were kept in undercharged condition before test. In the experiment, a 100- MHz scope was employed to measure the charging current and battery voltage. A current probe set at 100 mV/A was utilized to sense the charging current. The alternator outputs were converted into DC output by a rectifier module to charge the batteries. As shown in Fig. 8, there was a large variation in current and voltage because of the unstable wind speed. The fluctuations in amplitude of the charging current were attributed to the conventional AC-DC rectification effect. Increase in wind speed also led to increase in charging current. The variation in battery voltage as a result of internal impedance is above 1V.

internal voltage drop increases with the charging current *Ib*. This leads to the increase in the battery voltage *Eb*. A battery was charged by wind and solar energy sources using the conventional wind and PV charging system, as shown in Fig. 3. **Figure 8.** Charging current (Ch1: A/div) and battery voltage (Ch2: 5 V/div) of wind power. Owing to the source resist‐ ance of the battery, the internal voltage drop increases with the charging current *Ib*. This leads to the increase in the battery voltage *Eb*.

Figure 8. Charging current (Ch1: A/div) and battery voltage (Ch2: 5 V/div) of wind power. Owing to the source resistance of the battery, the

In the first 30 minutes, the battery remained in an under-charged condition. As shown in Fig. 9(a), the stable 0.8-A PV charging current is not affected by the large variation in wind charging current. After 1 hour, as seen in Fig. 9(b), the charging curve of solar energy shows an opposite trend as that of wind power. The solar charging current decreases as the wind charging current increases. After 2 hours, as shown in Fig. 9(c), either solar energy or wind power dominates the charging behavior. When the wind charging current exceeds 3 A, the solar energy source does not output any power. Therefore, the system cannot get both wind and solar energy at the same time. Iw A battery was charged by wind and solar energy sources using the conventional wind and PV charging system, as shown in Fig. 3. In the first 30 minutes, the battery remained in an undercharged condition. As shown in Fig. 9(a), the stable 0.8-A PV charging current is not affected by the large variation in wind charging current. After 1 hour, as seen in Fig. 9(b), the charging curve of solar energy shows an opposite trend as that of wind power. The solar charging current decreases as the wind charging current increases. After 2 hours, as shown in Fig. 9(c), either solar energy or wind power dominates the charging behavior. When the wind charging current exceeds 3 A, the solar energy source does not output any power. Therefore, the system cannot get both wind and solar energy at the same time.

> 1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

(a)

(b)

(c)

Iw

Ip

Iw

Ip

Ip

internal voltage drop increases with the charging current *Ib*. This leads to the increase in the battery voltage *Eb*.

Figure 8. Charging current (Ch1: A/div) and battery voltage (Ch2: 5 V/div) of wind power. Owing to the source resistance of the battery, the

A battery was charged by wind and solar energy sources using the conventional wind and PV charging system, as shown in Fig. 3. In the first 30 minutes, the battery remained in an under-charged condition. As shown in Fig. 9(a), the stable 0.8-A PV charging current is not affected by the large variation in wind charging current. After 1 hour, as seen in Fig. 9(b), the charging curve of solar energy shows an opposite trend as that of wind power. The solar charging current decreases as the wind charging current

solar energy at the same time.

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 5 Volt 1 s

Ib

Eb

**3.2. Conventional hybrid wind and PV power generating system**

362 Solar Cells - Research and Application Perspectives

voltage as a result of internal impedance is above 1V.

battery voltage *Eb*.

Iw

Ip

Iw

Ip

Ib

Eb

solar energy at the same time.

the system cannot get both wind and solar energy at the same time.

Iw

Ip

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 5 Volt 1 s

**Figure 8.** Charging current (Ch1: A/div) and battery voltage (Ch2: 5 V/div) of wind power. Owing to the source resist‐ ance of the battery, the internal voltage drop increases with the charging current *Ib*. This leads to the increase in the

A battery was charged by wind and solar energy sources using the conventional wind and PV charging system, as shown in Fig. 3. In the first 30 minutes, the battery remained in an undercharged condition. As shown in Fig. 9(a), the stable 0.8-A PV charging current is not affected by the large variation in wind charging current. After 1 hour, as seen in Fig. 9(b), the charging curve of solar energy shows an opposite trend as that of wind power. The solar charging current decreases as the wind charging current increases. After 2 hours, as shown in Fig. 9(c), either solar energy or wind power dominates the charging behavior. When the wind charging current exceeds 3 A, the solar energy source does not output any power. Therefore,

> 1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

(a)

(b)

(c)

In this study, a 250-W permanent magnet alternator driven by wind turbine and a 12- V/75-W solar cell panel were used as the wind and solar energy source, respectively. Both energy sources were output to 12-V/75-AH lead batteries, which were kept in undercharged condition before test. In the experiment, a 100- MHz scope was employed to measure the charging current and battery voltage. A current probe set at 100 mV/A was utilized to sense the charging current. The alternator outputs were converted into DC output by a rectifier module to charge the batteries. As shown in Fig. 8, there was a large variation in current and voltage because of the unstable wind speed. The fluctuations in amplitude of the charging current were attributed to the conventional AC-DC rectification effect. Increase in wind speed also led to increase in charging current. The variation in battery

internal voltage drop increases with the charging current *Ib*. This leads to the increase in the battery voltage *Eb*.

energy shows an opposite trend as that of wind power. The solar charging current decreases as the wind charging current

increases. After 2 hours, as shown in Fig. 9(c), either solar energy or wind power dominates the charging behavior. When the wind charging current exceeds 3 A, the solar energy source does not output any power. Therefore, the system cannot get both wind and **Figure 9.** Both wind and solar sources charge a single battery. Ch1: wind charging current (A/div), Ch2: solar charging current (A/div). (a) During the first 30 minutes, despite large variations in the wind charging current *Iw*, both energy sources charge the battery simultaneously, indicating that the open-circuit battery voltage *Eb* is small. (b) After 1-hour charging, competitive behavior occurs. When the wind charging current *Iw* exceeds 6.5A, the solar charging current *Ip* decreases. The large wind charging current increases the internal voltage drop of the battery, which leads to decrease in solar charging current. (c) After 2-hour charging, a wind charging current of only 0.5 A can reduce the solar charg‐ ing current from 1.8 A. This points out that the battery voltage builds up gradually and the small internal voltage drop is enough to exclude the solar charging current. Therefore, the small source impedance of the wind generator domi‐ nates the charging operation. A wind charging current of only 3 A can stop the solar charging current.

### **3.3. Switch- control hybrid wind and PV power generating system**

In order to acquire both wind and solar energy at the same time, the system is config‐ ured as in Fig. 4. As seen in Fig. 10(a)-(c), the wind to solar charging duty cycle ratio is changed to examine the charging behavior at three wind speeds. Owing to fluctuations in wind speed at 3 m/s, the system sometimes stops outputting the wind charging current, as seen in Fig. 10(a), while the charging of battery by solar energy remains very stable. When wind speed increases to 4 m/s, as shown in Fig. 10(b), the wind charging current contin‐ ues to charge the battery during its duty cycle, but the current decreases during charg‐ ing. When the solar charging duration is increased to 3.2 seconds, the wind charging current drops to 2 A, as seen in Fig. 10(c) prior to solar charging. Upon completion of solar charging, i.e. after 3.2 seconds, the wind charging current increases to 4 A and falls gradually back to 2 A This phenomenon can be explained as follows. During solar energy charging, the wind turbine is in a no-load condition. Wind energy is thus converted into mechanical energy, which speeds up the alternator. In other words, owing to the inertial momentum, the wind turbine can store mechanical energy, and solar energy is not best utilized or lost during wind charging. Therefore, the charging duty cycle ratio between wind and solar energy can be adjusted to obtain the maximum energy source.

### **3.4. Microprocessor- controlled hybrid wind and PV power generating system**

In the above two cases, there is always some energy loss during the power generating system operation. To obtain both wind and solar energy at the same time, a microproces‐ sor and two batteries as shown in Fig. 5 are employed to control the charging operation from both energy sources. Figure 11 displays the circuit in detail, in which controller IC1 and comparator IC2 control the system operation. Depending on the weather condition, wind energy can charge the wind battery *Ebw* directly or the solar battery *Ebp* indirectly through *Qws*. Both wind and solar energy are sensed by two comparators of IC2. One senses the sunlight to control the load while the other monitors the charging condition of the wind battery. If there has been no wind or sun for some days and both wind and solar ener‐ gies remain insufficient, the two batteries will be in under-charged condition and utility power supply will be used instead. In view of large variations in wind energy, a current transformer CT is employed to detect the charging current of the wind battery. If the charging current exceeds the specification of the battery, *Qw* runs in PWM mode to protect the battery. All functions are controlled by the software of controller IC1.

As shown in Fig. 12, when wind speed is low, about 3 m/s, the system runs in the independent charging mode and the solar battery charging current remains constant at 2A, although there are slight variations in the charging current of the wind battery.

If there is only wind power, as shown in Fig. 14(a), the system runs in the wind- hybrid charging mode, so the unstable wind charging current charges both wind and solar batteries. When the charging current is less than 2 A, as shown in Fig. 14(b), the system remains in the wind- hybrid charging mode and only the wind battery is being charged.

When there is solar energy and a high wind speed at the same time, as shown in Fig. 15, the system runs in the wind-enhanced charging mode. In order to benefit from both energy sources and reduce fluctuations from the wind source, the system runs in the wind-enhanced charging mode when the wind charging current is above 3A. The wind charging current charges the solar battery in addition to the original wind battery, leading to variations in the solar charging current. When the wind charging current is below 3A, the system runs in the independent charging mode. That is, the stable solar charging current of 2A and the fluctuating wind charging current charge the solar battery and the wind battery, respectively. view of large variations in wind energy, a current transformer CT is employed to detect the charging current of the wind battery. If the charging current exceeds the specification of the battery, *Qw* runs in PWM mode to protect the battery. All functions are controlled by the software of controller IC1. As shown in Fig. 12, when wind speed is low, about 3 m/s, the system runs in the independent charging mode and the solar battery charging current remains constant at 2A, although there are slight variations in the charging current of the wind battery.

**3.3. Switch- control hybrid wind and PV power generating system**

364 Solar Cells - Research and Application Perspectives

energy can be adjusted to obtain the maximum energy source.

**3.4. Microprocessor- controlled hybrid wind and PV power generating system**

the battery. All functions are controlled by the software of controller IC1.

are slight variations in the charging current of the wind battery.

charging mode and only the wind battery is being charged.

In the above two cases, there is always some energy loss during the power generating system operation. To obtain both wind and solar energy at the same time, a microproces‐ sor and two batteries as shown in Fig. 5 are employed to control the charging operation from both energy sources. Figure 11 displays the circuit in detail, in which controller IC1 and comparator IC2 control the system operation. Depending on the weather condition, wind energy can charge the wind battery *Ebw* directly or the solar battery *Ebp* indirectly through *Qws*. Both wind and solar energy are sensed by two comparators of IC2. One senses the sunlight to control the load while the other monitors the charging condition of the wind battery. If there has been no wind or sun for some days and both wind and solar ener‐ gies remain insufficient, the two batteries will be in under-charged condition and utility power supply will be used instead. In view of large variations in wind energy, a current transformer CT is employed to detect the charging current of the wind battery. If the charging current exceeds the specification of the battery, *Qw* runs in PWM mode to protect

As shown in Fig. 12, when wind speed is low, about 3 m/s, the system runs in the independent charging mode and the solar battery charging current remains constant at 2A, although there

If there is only wind power, as shown in Fig. 14(a), the system runs in the wind- hybrid charging mode, so the unstable wind charging current charges both wind and solar batteries. When the charging current is less than 2 A, as shown in Fig. 14(b), the system remains in the wind- hybrid

In order to acquire both wind and solar energy at the same time, the system is config‐ ured as in Fig. 4. As seen in Fig. 10(a)-(c), the wind to solar charging duty cycle ratio is changed to examine the charging behavior at three wind speeds. Owing to fluctuations in wind speed at 3 m/s, the system sometimes stops outputting the wind charging current, as seen in Fig. 10(a), while the charging of battery by solar energy remains very stable. When wind speed increases to 4 m/s, as shown in Fig. 10(b), the wind charging current contin‐ ues to charge the battery during its duty cycle, but the current decreases during charg‐ ing. When the solar charging duration is increased to 3.2 seconds, the wind charging current drops to 2 A, as seen in Fig. 10(c) prior to solar charging. Upon completion of solar charging, i.e. after 3.2 seconds, the wind charging current increases to 4 A and falls gradually back to 2 A This phenomenon can be explained as follows. During solar energy charging, the wind turbine is in a no-load condition. Wind energy is thus converted into mechanical energy, which speeds up the alternator. In other words, owing to the inertial momentum, the wind turbine can store mechanical energy, and solar energy is not best utilized or lost during wind charging. Therefore, the charging duty cycle ratio between wind and solar

 <sup>(</sup>b) Fig. 10. Switch-controlled wind and solar sources charge a battery. Ch1: battery charging

current (A/div), Ch2: solar charging current (a),(b), wind charging current (c) (A/div). (a) Wind speed is 3 m/s, but it is unstable. The charging duty cycle ratio between solar and wind power is 2:1. In the absence of wind power, only the solar charging current (0.6 A) charges the battery. In the presence of wind power, the wind charging current gradually decays. Because the wind charging duration is only 0.8 second, steady state cannot be attained. (b) Wind speed is increased to about 4 m/s and solar charging current is decreased to 0.4 A. Wind energy exceeds solar energy. (c) Wind speed is increased to 5 m/s. The charging duty cycle ratio between solar and wind power is changed to 1:2. Moreover, the solar charging duration is extended from 1.8 to 3.2 seconds. In this situation, mechanical energy accumulated by **Figure 10.** Switch-controlled wind and solar sources charge a battery. Ch1: battery charging current (A/div), Ch2: solar charging current (a),(b), wind charging current (c) (A/div). (a) Wind speed is 3 m/s, but it is unstable. The charging duty cycle ratio between solar and wind power is 2:1. In the absence of wind power, only the solar charging current (0.6 A) charges the battery. In the presence of wind power, the wind charging current gradually decays. Because the wind charging duration is only 0.8 second, steady state cannot be attained. (b) Wind speed is increased to about 4 m/s and solar charging current is decreased to 0.4 A. Wind energy exceeds solar energy. (c) Wind speed is increased to 5 m/s. The charging duty cycle ratio between solar and wind power is changed to 1:2. Moreover, the solar charging duration is extended from 1.8 to 3.2 seconds. In this situation, mechanical energy accumulated by inertia of the wind turbine during solar charging can also be utilized during wind charging. Hence, this configuration can ensure full uti‐ lization of wind energy with no loss at all.

inertia of the wind turbine during solar charging can also be utilized during wind charging.

Hence, this configuration can ensure full utilization of wind energy with no loss at all.

current *Iw* and the solar charging current *Ip* that fuel the wind (E*bw*) and solar (E*bp*) battery. According to the weather condition, the wind and solar charging currents may charge either the wind or solar battery. Current transformer CT senses the wind charging current to prevent damage to the battery by excess charging current. If the energy of the batteries cannot meet the power required, utility power supply is used instead. **Figure 11.** Microprocessor-controlled hybrid wind and solar charging circuit. Microprocessor IC1 and Comparator IC2 controls the wind charging current *Iw* and the solar charging current *Ip* that fuel the wind (E*bw*) and solar (E*bp*) battery. According to the weather condition, the wind and solar charging currents may charge either the wind or solar battery. Current transformer CT senses the wind charging current to prevent damage to the battery by excess charging cur‐ rent. If the energy of the batteries cannot meet the power required, utility power supply is used instead. Figure 11.Microprocessor-controlled hybrid wind and solar charging circuit. Microprocessor IC1 and Comparator IC2 controls the wind charging current *Iw* and the solar charging current *Ip* that fuel the wind (E*bw*) and solar (E*bp*) battery. According to the weather condition, the wind and solar

Figure 11.Microprocessor-controlled hybrid wind and solar charging circuit. Microprocessor IC1 and Comparator IC2 controls the wind charging

charging currents may charge either the wind or solar battery. Current transformer CT senses the wind charging current to prevent damage to the

AC <sup>1</sup> V+ <sup>2</sup>

Qc

AC <sup>3</sup> V- <sup>4</sup> BRIDGE

0 110 220

**UTILITY POWER**

0 12

Dc

Dw Dp

LOAD

Ql

Figure 13.The charging system operates in two stages. In Stage 1, only solar energy exists. The solar charging current not only charges the solar battery but also the wind battery. In Stage 2, when wind power also exists, the wind charging current charges the wind battery. Part of the solar

battery by excess charging current. If the energy of the batteries cannot meet the power required, utility power supply is used instead.

Ibp Figure 12.Both solar and wind power co-exist. The controller operates in the independent charging mode. The wind charging current charges the wind battery (Ch1: A/div) while the solar charging current charges the solar battery (Ch2: A/div). **Figure 12.** Both solar and wind power co-exist. The controller operates in the independent charging mode. The wind charging current charges the wind battery (Ch1: A/div) while the solar charging current charges the solar battery (Ch2: A/div).

charging current is switched back to the solar battery. Ch1, Ch2: A/div.

Ibp

Ibw

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

Ibw

Figure 13.The charging system operates in two stages. In Stage 1, only solar energy exists. The solar charging current not only charges the solar battery but also the wind battery. In Stage 2, when wind power also exists, the wind charging current charges the wind battery. Part of the solar

charging current is switched back to the solar battery. Ch1, Ch2: A/div.

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s wind battery (Ch1: A/div) while the solar charging current charges the solar battery (Ch2: A/div).

Figure 11.Microprocessor-controlled hybrid wind and solar charging circuit. Microprocessor IC1 and Comparator IC2 controls the wind charging current *Iw* and the solar charging current *Ip* that fuel the wind (E*bw*) and solar (E*bp*) battery. According to the weather condition, the wind and solar charging currents may charge either the wind or solar battery. Current transformer CT senses the wind charging current to prevent damage to the

**DAY OR NIGHT**

Qps

**WIND PV** Iw Ip

AC <sup>1</sup> V+ <sup>2</sup>

Qc

AC <sup>3</sup> V- <sup>4</sup> BRIDGE

0 110 220

**UTILITY POWER**

0 12

Dc

Dw Dp

LOAD

Ql

battery by excess charging current. If the energy of the batteries cannot meet the power required, utility power supply is used instead.

Ebw Ebp

Ibp

 **SOLAR BAT**

Vout <sup>3</sup> IC3

CT

Vin <sup>1</sup> GND 2

Ibw

Qw

**Vr WIND BAT**

Qws

SD

IC2

1) Ch 1: 100 mVolt 1 s

Figure 13.The charging system operates in two stages. In Stage 1, only solar energy exists. The solar charging current not only charges the solar battery but also the wind battery. In Stage 2, when wind power also exists, the wind charging current charges the wind battery. Part of the solar charging current is switched back to the solar battery. Ch1, Ch2: A/div. **Figure 13.** The charging system operates in two stages. In Stage 1, only solar energy exists. The solar charging current not only charges the solar battery but also the wind battery. In Stage 2, when wind power also exists, the wind charg‐ ing current charges the wind battery. Part of the solar charging current is switched back to the solar battery. Ch1, Ch2: A/div. Fig. 13. The charging system operates in two stages. In Stage 1, only solar energy exists. The solar charging current not only charges the solar battery but also the wind battery. In Stage 2, when wind power also exists, the wind charging current charges the wind battery. Part of the

solar charging current is switched back to the solar battery. Ch1, Ch2: A/div.

Figure 11.Microprocessor-controlled hybrid wind and solar charging circuit. Microprocessor IC1 and Comparator IC2 controls the wind charging current *Iw* and the solar charging current *Ip* that fuel the wind (E*bw*) and solar (E*bp*) battery. According to the weather condition, the wind and solar charging currents may charge either the wind or solar battery. Current transformer CT senses the wind charging current to prevent damage to the

**DAY OR NIGHT**

CT

Ebw Ebp

Ibp

 **SOLAR BAT**

Vout <sup>3</sup> IC3

Qps

**Vr WIND BAT**

Vin <sup>1</sup> GND 2

**WIND PV** Iw Ip

AC <sup>1</sup> V+ <sup>2</sup>

Qc

**WIND PV** Iw Ip

Qps

AC <sup>3</sup> V- <sup>4</sup> BRIDGE

0 110 220

**UTILITY POWER**

0 12

Dc

Dw Dp

LOAD

**DAY OR NIGHT**

Figure 11.Microprocessor-controlled hybrid wind and solar charging circuit. Microprocessor IC1 and Comparator IC2 controls the wind charging current *Iw* and the solar charging current *Ip* that fuel the wind (E*bw*) and solar (E*bp*) battery. According to the weather condition, the wind and solar charging currents may charge either the wind or solar battery. Current transformer CT senses the wind charging current to prevent damage to the

Figure 12.Both solar and wind power co-exist. The controller operates in the independent charging mode. The wind charging current charges the

Figure 13.The charging system operates in two stages. In Stage 1, only solar energy exists. The solar charging current not only charges the solar battery but also the wind battery. In Stage 2, when wind power also exists, the wind charging current charges the wind battery. Part of the solar

battery by excess charging current. If the energy of the batteries cannot meet the power required, utility power supply is used instead.

Ql

AC <sup>1</sup> V+ <sup>2</sup>

Qc

AC <sup>3</sup> V- <sup>4</sup> BRIDGE

0 110 220

**UTILITY POWER**

0 12

Dc

Dw Dp

LOAD

Ql

Figure 12.Both solar and wind power co-exist. The controller operates in the independent charging mode. The wind charging current charges the

wind battery (Ch1: A/div) while the solar charging current charges the solar battery (Ch2: A/div).

Figure 13.The charging system operates in two stages. In Stage 1, only solar energy exists. The solar charging current not only charges the solar battery but also the wind battery. In Stage 2, when wind power also exists, the wind charging current charges the wind battery. Part of the solar

charging current is switched back to the solar battery. Ch1, Ch2: A/div.

wind battery (Ch1: A/div) while the solar charging current charges the solar battery (Ch2: A/div).

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

**Figure 12.** Both solar and wind power co-exist. The controller operates in the independent charging mode. The wind charging current charges the wind battery (Ch1: A/div) while the solar charging current charges the solar battery

> 1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

charging current is switched back to the solar battery. Ch1, Ch2: A/div.

Ibp

Ibw

battery by excess charging current. If the energy of the batteries cannot meet the power required, utility power supply is used instead.

**Figure 11.** Microprocessor-controlled hybrid wind and solar charging circuit. Microprocessor IC1 and Comparator IC2 controls the wind charging current *Iw* and the solar charging current *Ip* that fuel the wind (E*bw*) and solar (E*bp*) battery. According to the weather condition, the wind and solar charging currents may charge either the wind or solar battery. Current transformer CT senses the wind charging current to prevent damage to the battery by excess charging cur‐

rent. If the energy of the batteries cannot meet the power required, utility power supply is used instead.

Ebw Ebp

IC2

Ibp

Qw

 **SOLAR BAT**

Ibw

Vout <sup>3</sup> IC3

CT

Qws

Vin <sup>1</sup> GND 2

Ibw

366 Solar Cells - Research and Application Perspectives

Qw

**Vr WIND BAT**

Qws

SD

(Ch2: A/div).

Ibp

Ibw

IC2

SD

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

Fig. 14. The charging system operates when there is only wind power. (a) When wind speed exceeds 5 m/s, the wind charging current charges both wind and solar batteries. (b) When wind speed drops gradually below 3 m/s, the wind charging current in the solar battery **Figure 14.** The charging system operates when there is only wind power. (a) When wind speed exceeds 5 m/s, the wind charging current charges both wind and solar batteries. (b) When wind speed drops gradually below 3 m/s, the wind charging current in the solar battery decreases to zero. Ch1, Ch2: A/div

If there is only wind power, as shown in Fig. 14(a), the system runs in the wind- hybrid charging mode, so the unstable wind charging current charges both wind and solar batteries. When the charging current is less than 2 A, as shown in Fig. 14(b), the system remains in the

When there is solar energy and a high wind speed at the same time, as shown in Fig. 15, the system runs in the wind-enhanced charging mode. In order to benefit from both energy sources and reduce fluctuations from the wind source, the system runs in the wind-enhanced charging mode when the wind charging current is above 3A. The wind charging current charges the solar battery in addition to the original wind battery, leading to variations in the solar charging current. When the wind charging current is below 3A, the system runs in the independent charging mode. That is, the stable solar charging current of 2A and the fluctuating wind charging current charge the solar battery and the wind battery, respectively.

wind- hybrid charging mode and only the wind battery is being charged**.**

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

Ibw

Ibp

decreases to zero. Ch1, Ch2: A/div

Ch1, Ch2: A/div

Ibp

Ibw

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

Ibw

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

(b)

(a)

Ibp

Figure 14.The charging system operates when there is only wind power. (a) When wind speed exceeds 5 m/s, the wind charging current charges both wind and solar batteries. (b) When wind speed drops gradually below 3 m/s, the wind charging current in the solar battery decreases to zero.

If there is only wind power, as shown in Fig. 14(a), the system runs in the wind- hybrid charging mode, so the unstable wind charging current charges both wind and solar batteries. When the charging current is less than 2 A, as shown in Fig. 14(b), the

When there is solar energy and a high wind speed at the same time, as shown in Fig. 15, the system runs in the wind-enhanced

in addition to the original wind battery, leading to variations in the solar charging current. When the wind charging current is below 3A, the system runs in the independent charging mode. That is, the stable solar charging current of 2A and the fluctuating

system remains in the wind- hybrid charging mode and only the wind battery is being charged.

wind charging current charge the solar battery and the wind battery, respectively.

Figure 15.Both wind and solar power sources co-exist, but wind speed sometimes exceeds 8 m/s. The charging system operates in two stages. In Stage 1, when wind speed is below 5 m/s, same as that in Fig. 12, the solar and wind charging currents charge the solar and wind battery, respectively. In Stage 2, when wind speed is increased, the wind charging current not only charges the wind battery, but also the solar battery. That is, there are two currents charging the solar battery. Ch1, Ch2: A/div **Figure 15.** Both wind and solar power sources co-exist, but wind speed sometimes exceeds 8 m/s. The charging sys‐ tem operates in two stages. In Stage 1, when wind speed is below 5 m/s, same as that in Fig. 12, the solar and wind charging currents charge the solar and wind battery, respectively. In Stage 2, when wind speed is increased, the wind charging current not only charges the wind battery, but also the solar battery. That is, there are two currents charging the solar battery. Ch1, Ch2: A/div

## **4. Conclusion**

In this study, theoretical investigations are performed to examine the effect of source impe‐ dance on a small hybrid wind and PV power system. Because of voltage drop in power sources, both energy sources cannot charge a battery simultaneously after initial charging. This study proposed using a switch circuit to increase the utilization of both energy sources. There is only slight solar energy loss when wind power is in operation. To increase energy efficiency by gaining both wind and solar energy, a microprocessor-based hybrid charging system is proposed. Results show that besides increasing the reliability of the power system, the fluctuations in wind energy source are also reduced.

## **Author details**

Mu-Kuen Chen and Chao-Yuan Cheng

Department of Electrical Engineering, St. John's University, Taiwan

## **References**

[1] Salameh, Z. M., Casacca, M. A. and Lynch, W. A. "A mathematical model for leadacid batteries," *IEEE Trans. On Energy Conversion*, vol.7, no.1, pp 93-98, Mar. 1992.

below 3A, the system runs in the independent charging mode. That is, the stable solar charging current of 2A and the fluctuating [2] H. G. Zimmerman and R. G. Peterson, "An electrochemical cell equivalent circuit for storage battery/power system calculations by digital computer", Proceedings 13th In‐ tersociety Energy Conversion Engineering Conference, pp. 33-38, 1978.

Figure 14.The charging system operates when there is only wind power. (a) When wind speed exceeds 5 m/s, the wind charging current charges both wind and solar batteries. (b) When wind speed drops gradually below 3 m/s, the wind charging current in the solar battery decreases to zero.

If there is only wind power, as shown in Fig. 14(a), the system runs in the wind- hybrid charging mode, so the unstable wind charging current charges both wind and solar batteries. When the charging current is less than 2 A, as shown in Fig. 14(b), the

When there is solar energy and a high wind speed at the same time, as shown in Fig. 15, the system runs in the wind-enhanced charging mode. In order to benefit from both energy sources and reduce fluctuations from the wind source, the system runs in the wind-enhanced charging mode when the wind charging current is above 3A. The wind charging current charges the solar battery in addition to the original wind battery, leading to variations in the solar charging current. When the wind charging current is

Figure 15.Both wind and solar power sources co-exist, but wind speed sometimes exceeds 8 m/s. The charging system operates in two stages. In

system remains in the wind- hybrid charging mode and only the wind battery is being charged.

wind charging current charge the solar battery and the wind battery, respectively.

is, there are two currents charging the solar battery. Ch1, Ch2: A/div

In this study, theoretical investigations are performed to examine the effect of source impe‐ dance on a small hybrid wind and PV power system. Because of voltage drop in power sources, both energy sources cannot charge a battery simultaneously after initial charging. This study proposed using a switch circuit to increase the utilization of both energy sources. There is only slight solar energy loss when wind power is in operation. To increase energy efficiency by gaining both wind and solar energy, a microprocessor-based hybrid charging system is proposed. Results show that besides increasing the reliability of the power system, the

[1] Salameh, Z. M., Casacca, M. A. and Lynch, W. A. "A mathematical model for leadacid batteries," *IEEE Trans. On Energy Conversion*, vol.7, no.1, pp 93-98, Mar. 1992.

**Figure 15.** Both wind and solar power sources co-exist, but wind speed sometimes exceeds 8 m/s. The charging sys‐ tem operates in two stages. In Stage 1, when wind speed is below 5 m/s, same as that in Fig. 12, the solar and wind charging currents charge the solar and wind battery, respectively. In Stage 2, when wind speed is increased, the wind charging current not only charges the wind battery, but also the solar battery. That is, there are two currents charging

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

Ch1, Ch2: A/div

fluctuations in wind energy source are also reduced.

Department of Electrical Engineering, St. John's University, Taiwan

Mu-Kuen Chen and Chao-Yuan Cheng

the solar battery. Ch1, Ch2: A/div

**4. Conclusion**

**Author details**

**References**

368 Solar Cells - Research and Application Perspectives

Ibw

Ibp

Ibp

Ibw

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

Ibw

1) Ch 1: 100 mVolt 1 s 2) Ch 2: 100 mVolt 1 s

(b)

(a)

Ibp

	- [7] Jiang Z. and Dougal, R. A. "Control strategies for active power sharing in a fuel-cellpowered battery-charging station," *IEEE Trans. On Industry Applications*, vol. 40, no. 3, pp 917-924, May/Jun, 2004
	- [8] Borowy, B. S. and Salameh, Z. M., "Optimum photovoltaic array size for a hybrid wind/PV system," *IEEE Trans. On Energy Conversion*, vol.9, no.3, pp 482-488, Sep. 1994.
	- [9] Chedid, R. and Rahman, S., "Unit sizing and control of hybrid wind-solar power sys‐ tems," *IEEE Trans. On Energy Conversion*, vol.2, no.1, pp 79-85, Mar. 1997.

## *Edited by Arturo Morales-Acevedo*

Over the last decade, photovoltaic (PV) technology has shown the potential to become a major source of power generation for the world - with robust and continuous growth even during times of financial and economic crisis. That growth is expected to continue in the years ahead as worldwide awareness of the advantages of PV increases. However, cost remains as the greatest barrier to further expansion of PVgenerated power, and therefore cost reduction is the prime goal of the PV and solar cell investigation. This book intends to contribute to such a purpose by covering a wide range of modern research topics in the solar cell physics and technology fields. The already established -1st generation- silicon solar cell technology, the 2nd generation thin film and the 3rd generation dye sensitized solar cells, including new technologies with very high perspectives for reducing the cost of solar electricity such as CZTS, organic polymer and tandem solar cells based on III-V compounds -under concentrated sunlight- are studied in this book by experts in the field from around the world. At the end, two chapters are also dedicated to the systems engineering, providing a complete PV energy research and application perspectives panorama

Solar Cells - Research and Application Perspectives

Solar Cells

Research and Application Perspectives

*Edited by Arturo Morales-Acevedo*

Photo by stevanovicigor / iStock