**2. Experimental**

fluid control, and precision machining due to their high generative force, accurate displace‐

To date, most of the commonly used piezoelectric materials are lead-based, such as lead zirconate titanate (PZT) and its solid solutions. Lead-containing materials display high piezoelectric coefficients, especially near the morphotrophic phase boundary (MPB) and have dominated the market of piezoelectric industry [1, 4, 5]. However, the hazardous lead content present within these materials raise serious environmental problems. Therefore, environmen‐ tal issues such as regulations and policies against lead-based materials have been increasingly enacted throughout the world [6–8]. In order to circumvent the drawback of lead toxicity, extensive research is focused on the quest for alternate piezoelectric materials. Numerous research efforts have been devoted to the candidate lead-free piezoelectric materials such as BaTiO3 (BT), K0.5Na0.5NbO3 (KNN), and Bi0.5Na0.5TiO3 (BNT) because of their interesting electromechanical properties. The BT-based ceramics are interesting from the view point of their good ferroelectricity, chemical, and mechanical stability along with easy processing in polycrystalline form [9, 10]; however, they are inadequate for device applications due to their low Curie temperature (*T*<sup>c</sup> ≈ 120°C) [11, 12]. Alternatively, The KNbO3–NaNbO3 solid solution or the (K,Na)NbO3 (KNN)-based material exhibits good piezoelectric properties as well as high *T*<sup>c</sup> ≈ 420°C and have been studied as substitutes for PZT-based ceramics [13–16]. Nevertheless, alkali metal elements present in KNN-based ceramics easily evaporate at high temperatures. Moreover, KNN ceramics are not only hygroscopic but also make transition between ortho‐ rhombic and tetragonal structure (TO–T) around 200°C, and this transition temperature shifts toward room temperature (RT) with change in composition through doping or substituting other element, which hinders KNN-based ceramics from practical applications [17, 18]. As one of the most feasible lead-free candidates, BNT-based solid solutions have good piezoelectric properties, excellent reproducibility, and high maximum dielectric constant temperature (*T*<sup>m</sup> 300°C) [19–22]. However, poling of pure BNT ceramics is difficult due to their large coercive field (*E*c ~ 73 kV/cm), which is overcome by certain amount through compositional modifica‐

In the recent years, extraordinarily large strain has been reported in compositionally designed BNT-based ceramics [20, 21, 26, 27], which seems to be alternative for PZT in specific actuator applications. Beside, compositional design, texture engineering of polycrystalline ceramics is inimitable and vibrant approach to enhance the piezoelectric properties of ceramics without any major change in the base compositions. Several texture development techniques, such as templated grain growth (TGG) [28], reactive TGG (RTGG) [18], oriented consolidation of anisometric particles [29], screen printing [30], multilayer grain growth [31], and directional solidification technology [32], have been employed to improve the electromechanical proper‐ ties of piezoelectric ceramics. Among all these techniques, RTGG is more suitable for the texture development of perovskite-type materials [33, 34]. In this process, the plate-like template particles that have specific microstructure and crystallographic characteristics are oriented in a matrix ceramic powder through a tape-casting process, and then the consequent heat treatment results in the nucleation and growth of desired crystals on aligned template particles to bring into being textured ceramics [35]. Because of the grain orientation effect

ment, and rapid response [2, 3].

26 Piezoelectric Materials

tion and texture development [22–25].

Reagent-grade metal oxide powders of Bi2O3, TiO2, and Na2CO3 (purity >99.9%) were used as starting materials to produce plate-like Bi4.5Na0.5Ti4O15 (BNT4) precursors by molten salt synthesis (MSS) [38]. The stoichiometric amount of the raw BNT4 powder mixture was first mixed with NaCl (99.95%) in a weight ratio of 1:1.5 and then ball milled in polyethylene jar for 24 h. Consequently, the balls were removed and the slurry was dried and then brought to a firmly covered Al2O3 crucible for heat treatment at 1100°C for 4 h. The reaction was assumed complete in accordance with the chemical Eq. (1); the NaCl salt was washed away from the assynthesized product thorough hot de-ionized water. BNT4 platelets, Na2CO3, and TiO2 were then further weighed to provide a total BNT composition in accordance with the chemical Eq. (2). NaCl salt was again added to the powder mixture with 1:1.5 weight ratios, and then milling was carried out in the presence of ethanol through a magnetic stirrer for 5 h. Subsequently, the slurry was dried and a heat treatment at 950°C for 4 h was performed in a firmly covered Al2O3 crucible. Finally, NaCl salt was removed through hot de-ionized water from the product and HCl was utilized to remove the bismuth oxide (Bi2O3) by-products. For comparison, BNT particles were also produced by CMO route [39].

$$2.2\text{SBi}\_2\text{O}\_3 + 0.2\text{SNa}\_2\text{CO}\_3 + 4\text{TiO}\_2 \rightarrow \text{Bi}\_{4.5}\text{Na}\_{0.5}\text{Ti}\_4\text{O}\_{15} + 0.5\text{CO}\_2\tag{1}$$

$$\text{Bi}\_{4.5}\text{Na}\_{0.5}\text{Ti}\_4\text{O}\_{15} + 2\text{Na}\_2\text{CO}\_3 + \text{SiO}\_2 \rightarrow \text{ 9Bi}\_{0.5}\text{Na}\_{0.5}\text{TiO}\_3 + 2\text{CO}\_2\tag{2}$$

Grain-oriented ceramics with a composition of 0.994Bi0.5Na0.5TiO3–0.006BaZrO3 (BNT–BZ) were fabricated through RTGG process utilizing the as-synthesized BNT templates [38]. Commercially accessible carbonate powders such as: Na2CO3 and BaCO3 (99.95%, Sigma Aldrich) along with metal oxide powders such as: Bi2O3, TiO2, and ZrO2 (99.9% Junsei Co., Limited) were first weighed according to the stoichiometric formula of BNT–BZ and then mixed by ball milling for 24 h at 250 rpm. The slurry was dried and then calcined at 850°C for 2 h to form a perovskite phase. The as-prepared calcined powders of BNT–BZ were mixed thoroughly with a solvent (60 vol.% ethanol and 40 vol.% methyl-ethyl-ketone, MEK) and triethyl phosphate (dispersant) in a ball mill for 24 h. Subsequently, polyvinyl butyral (binder) and polyethylene glycol/diethyl-ophthalate (plasticizer) were added to the mixtures and the milling was continued again for another 24 h. BNT templates of 15 wt% were then added to the mixture and ball milled with a slow rotation for another 12 h to form a slurry for tape casting. The viscous slurry was tape cast to form a green sheet with a thickness of ~100 *μ*m on a SiO2-coated polyethylene film by a doctor blade apparatus. Afterward drying, a single layer sheet was cut, laminated, and hot-pressed at a temperature of 45°C and a pressure of 50 MPa for 2 min to form a 2-mm-thick green compact. The compacts were further cut into small samples of about 1 × 1 cm2 and then heated at 600°C for 12 h with intermediated steps of 250 and 350°C for 6 and 8 h and to remove organic substances prior to sintering. The samples were sintered at 1150°C for 15 h in air atmosphere and were then brought to room temperature at cooling rate of 5°C/min. For comparison, non-textured BNT–BZ ceramics were also prepared through conventional solid-state reaction [38].

Crystalline phase and purity information of the as-synthesized BNT particles and BNT–BZ ceramics were checked by X-ray diffraction machine (XRD, RAD III, Rigaku, Japan) using CuK*α* radiation (*λ* = 1.541 Å) at room temperature. The XRD patterns were collected in the Braggs–Brentano configuration operated at 10 mA and 20 kV with a step size of 0.02°. The particle size and shape was observed through field emission scanning electron microscope (FE-SEM, JP/JSM 5200, Japan). At room temperature, Raman scattering investigation was performed using a dispersive Raman spectrometer (ALMEGA, Nicelot, USA). Selected area electron diffraction (SAED) pattern and high-resolution transmission electron microscopy (HRTEM) images were obtained by transmission electron microscope (TEM) using a FE-TEM (JEOL/JEM-2100F version) operated at 200 kV. Both surfaces of the samples were polished and coated with a silver–palladium paste to form electrodes for electrical properties measurements. The dielectric constant and loss response were measured through an impedance analyzer (HP4194A, Agilent Technologies, Palo Alto, CA). The ferroelectric hysteresis loops were measured using a Precision Premier II device (Radiant Technology, Inc.) at 50 Hz. Fieldinduced strain response was measured using a contact-type displacement sensor (Model 1240; Mahr GmbH, Göttingen, Germany) at 50 mHz.
