Seung-Ho PARK, Seyit URAL, Cheol-Woo AHN 1, Sahn NAHM 1 and Kenji UCHINO

Transcription

1 Japanese Journal of Applied Physics Vol. 45, No. 4A, 6, pp #6 The Japan Society of Applied Physics Piezoelectric Properties of Sb-, Li-, and Mn-substituted Pb(Zr x Ti 1 x )O 3 Pb(Zn 1=3 Nb 2=3 )O 3 Pb(Ni 1=3 Nb 2=3 )O 3 Ceramics for High-Power Applications Seung-Ho PARK, Seyit URAL, Cheol-Woo AHN 1, Sahn NAHM 1 and Kenji UCHINO International Center for Actuators and Transducers, Materials Research Institute, The Pennsylvania State University, University Park, PA 16802, U.S.A. 1 Department of Materials Science and Engineering, Korea University, 1-5 Ka, Anam-dong, Sungbuk-Ku, Seoul , Korea (Received October 5, 5; accepted December 26, 5; published online April 7, 6) We developed piezoelectric ceramics suitable for high-power piezoelectric transformer applications. Sb, Li, and Mn were substituted in the 0.8Pb(Zr 0:5 Ti 0:5 )O Pb(Zn 1=3 Nb 2=3 )O Pb(Ni 1=3 Nb 2=3 )O 3 ceramics. All these substituents changed the structure of a tetragonal composition to cubical phase. The Sb and Li substitutions resulted in increases in and d 33, whereas Mn substitution increased Q m. The composition 0.8Pb(Zr x Ti 1 x ) 0.2Pb[(1 c){(1 b)(zn 0:8 Ni 0:2 ) 1=3 (Nb 1 a Sb a ) 2=3 b(li 1=4 (Nb 1 a Sb a ) 3=4 )} c(mn 1=3 (Nb 1 a Sb a ) 2=3 )]O 3 (a ¼ 0:1, b ¼ 0:3, c ¼ 0:3, and x ¼ 0:5) showed ¼ 0:56, Q m ¼ 1951 (planar mode), d 33 ¼ 239 pc/n, " T 3 =" 0 ¼ 739 and a maximum vibration velocity = 0.6 m/s in the 31 mode. By adjusting Zr/ Ti ratio, the structure of the ceramics reverted to the morphotropic phase boundary and showed ¼ 0:57, Q m ¼ 1502 (planar mode), d 33 ¼ 330 pc/n, " T 3 =" 0 ¼ 1653 and a maximum 31 mode vibration velocity = 0.58 m/s when Zr/Ti = 0.48/0.52. [DOI: /JJAP ] KEYWORDS: PZT PZNN ceramics, piezoelectricity, high power, maximum vibration velocity, substitution 1. Introduction Recently, high power piezoelectric devices such as piezoelectric actuators, ultrasonic motors, and piezoelectric transformers have been extensively developed. 1,2) These highpower devices are usually driven at a high electric field and a high vibration level, associated with significant heat generation. The maximum mechanical energy density generated in a piezoelectric can be evaluated by the square of the maximum vibration velocity, above which the temperature rise in this piezoelectric exceeds 20 C above room temperature. Note that the maximum vibration velocity is proportional to the mechanical quality factor Q m (inverse value of the intensive elastic loss tan 0, or imaginary elastic constant), and irrelevant to the elastic constant real part. There are three different kinds of figure of merit (FOM) for evaluating high power characteristics. (1) For piezoelectric actuators used under DC or slow AC that is, the off-resonance state, the piezoelectric strain constant d or electromechanical coupling factor k is the primary FOM. (2) For ultrasonic motors, the maximum vibration velocity v 0 is the FOM. (3) For piezoelectric transformers, the product of v 0 and k is the FOM, because the converted mechanical energy (at the resonance) must be reconverted to electric energy at a rate of k 2. 3,4) In the 1980 s, many efforts were made to develop soft lead zirconate titanate (PZT) ceramics with a high d and a high k (enhancement of the real part of the physical parameters). One of the most important materials is the Pb(Zr,Ti)O 3 Pb(Ni 1=3 Nb 2=3 )O 3 system developed by NEC, Japan and widely used for cofired multilayer actuators. 5,6) To the contrary, in the 1990 s, such efforts were shifted to very hard PZT with a high Q m, aiming at ultrasonic motor (USM) and transformer applications. In collaboration with NEC, Japan, The Penn State University developed the rareearth (e.g., Yb and Eu) doped Pb(Zr,Ti)O 3 Pb(Mn 1=3 Sb 2=3 )- O 3 system, which exhibits a maximum vibration velocity of more than 1 m/s, more than 3 times higher than those of commercially available hard PZTs. 7) However, because these PZT PMnS ceramics do not possess a high d or k, their 2667 transducer performance (coupling effect of converse and direct piezoelectric effects, such as in piezoelectric transformers) was not as superb as that expected initially. Therefore, in the 21st century, we have been focusing on compromised piezoelectric ceramics with a relatively high d and high k, by keeping the maximum vibration velocity at approximately 0.6 m/s (two fold higher than that of conventional hard PZTs). The Pb(Zr,Ti)O 3 Pb(Zn 1=3 Nb 2=3 )O 3 Pb(Ni 1=3 Nb 2=3 )O 3 system near the morphotropic phase boundary (MPB) are currently our target compositions, which can suppress heat generation significantly during a high-vibration-level drive with reasonable electromechanical coupling factors. In particular, 0.8Pb(Zr,Ti)O Pb- (Zn 1=3 Nb 2=3 )O Pb(Ni 1=3 Nb 2=3 )O wt % MnO 2 was reported for its good piezoelectric properties of ¼ 0:554, Q m ¼ 1502 and d 33 ¼ ) Furthermore, this composition is utilized as a base composition for low sintering temperature piezo ceramics. 9) The purpose of this study is to enhance the piezoelectric properties of the Pb(Zr,Ti)O 3 Pb(Zn 1=3 Nb 2=3 )O 3 Pb(Ni 1=3 - Nb 2=3 )O 3 system to provide improved high-power piezoelectric ceramics which have attractive base compositions for low temperature sintering, aiming at multilayer piezoelectric transformer applications. We investigated the effects of Sb, Li, and Mn substitutions, because Sb 2 O 5 and PbO provide a eutectic point with a melting point of approximately 820 C. 10) This low melting temperature can help the densification of the ceramic via liquid phase formation. Furthermore, a Pb(Mn 1=3 Sb 2=3 ) solid solution with PZT is a well known high-power material. 7) Li 2 CO 3 is one of the sintering aids used at low temperatures because of its low melting point. 11) In addition, Pb(Li 1=4 Nb 3=4 ) including Pb(Mg,Nb) Pb(Zr,Ti) + MnO 2 ceramics demonstrated significant characteristic improvement such as ¼ 0:54 and Q m ¼ 2132, when they were milled for a long period ( h). 12) Regarding the Mn doping effect, because Mn ions can have various valencies from Mn 4þ to Mn 2þ, and Mn 4þ occupying the B-sites of the perovskite structure can be reduced to Mn 2þ, consequently Mn 3þ (acceptor in the B

2 Jpn. J. Appl. Phys., Vol. 45, No. 4A (6) Fig. 1. XRD patterns of samples sintered at 1 C for 2 h in 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb(Zn 0:8 Ni 0:2 ) 1=3 (Nb 1 a Sb a ) 2=3 O 3 ceramics: a ¼ 0, a ¼ 0:1, a ¼ 0:2, and a ¼ 0:3. sites) can ameliorate the elastic loss of the piezo ceramics, i.e., an increase in Q m value. 13) 2. Experimental The specimens studied in this research were fabricated according to the formula 0.8Pb(Zr x Ti 1 x ) 0.2Pb[(1 c)- {(1 b)(zn 0:8 Ni 0:2 ) 1=3 (Nb 1 a Sb a ) 2=3 b(li 1=4 (Nb 1 a Sb a ) 3=4 )} c(mn 1=3 (Nb 1 a Sb a ) 2=3 )]O 3, where a ¼ 0{0:3; b ¼ 0{0:4; c ¼ 0{ 0:4; x ¼ 0:46{ 0:5, respectively. Raw materials of PbO, ZrO 2, TiO 2, ZnO, NiO, Nb 2 O 5,Sb 2 O 5,Li 2 CO 3, and MnO 2 with >99% purity were used to prepare samples by conventional ceramic sintering. The obtained mixture was ball-milled using zirconia ball media in a polyethylene jar for 24 h. The mixed slurry was dried and calcined at 750 C for 2 h. The calcined powders were ball-milled again and consolidated into disks of 12.5 mm diameter and rectangular plates using cold isostatic pressing (CIP) at 30,000 psi. The PbO-rich atmosphere sintering of the ceramics was performed in a high-purity alumina crucible at 1 C. The sintered rectangular samples were cut to mm 3 to measure k 31 and the corresponding Q m for this k 31 mode. The crystal structure and symmetry of the sintered bodies were examined by X-ray diffraction (XRD) analysis and the sintered bodies density was measured by the Archimedes method. A Ag Pt electrode was printed on the lapped surfaces to serve as electrode. The electroded specimens were poled in silicone oil at 150 C by applying a DC field of 3 kv/mm for 30 min. The electromechanical properties were measured using an Institute of Acoustics Academida Sinica piezo d 33 meter (ZJ-2) and Agilent precision impedance analyzer (4294A) based on IEEE standards. Dielectric properties as a function of temperature were obtained by measuring capacitance and loss using Stanford Research Systems LCR meter (SR715) in an automated temperaturecontrolled furnace interfaced with a computer for data acquisition. The observation and analysis of the vibration velocity were performed using an HP function/arbitrary waveform generator (33120A), an NF Electronic Instruments High Speed amplifier (4025), an HP oscilloscope (54645A), a Polytec vibrometer (OFV 3001) and a Polytec fiber interferometer (OFV 511). Under the center portion (nodal line) support, the vibration velocity at the edge was 2668 measured, with simultaneous temperature monitoring at the nodal line using an infrared camera. 3. Results and Discussion 3.1 Sb substitution Figure 1 shows the XRD patterns for the samples sintered at 1 C in 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb(Zn 0:8 Ni 0:2 ) 1=3 - (Nb 1 a Sb a ) 2=3 O 3 (a ¼ 0{0:3) ceramics. It can be seen that all the samples exhibit a perovskite structure with no secondary phase detected in the XRD analysis. The base composition (a ¼ 0) had a slight tetragonal symmetry since both () and () peaks show splitting. The substitution of Sb suppresses the () peak splitting by shifting the composition and thereby reducing tetragonality as shown in Figs Scanning electron microscopy (SEM) surface images of the 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb(Zn 0:8 Ni 0:2 ) 1=3 (Nb 1 a Sb a ) 2=3 - O 3 (a ¼ 0{0:3) ceramics sintered at 1 C 2 h are shown in Fig. 2. The SEM images of the sintered body did not exhibit any second phase, either, as indicated in the XRD result. The average grain size was 1.8 mm at a ¼ 0, but the number of small grains was decreased and the average grain size increased to approximately 2.5 mm, when the amount of Sb substitution is increased. Density, dielectric permittivity (" T 3 =" 0), electromechanical coupling factor ( ), mechanical quality factor and piezoelectric constant (d 33 ) are plotted as a function of the amount of Sb substitution in Fig. 3 in the 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb(Zn 0:8 Ni 0:2 ) 1=3 (Nb 1 a Sb a ) 2=3 O 3 composition, sintered at 1 C for 2 h. The density slightly increased with the amount of Sb substitution. This might be because of grain growth which was observed in the SEM images. The properties at a ¼ 0:1 of the 0.8Pb(Zr 0:5 Ti 0:5 ) 0.2Pb- (Zn 0:8 Ni 0:2 ) 1=3 (Nb 1 a Sb a ) 2=3 O 3 ceramics are summarized as follows: d 33 ¼ 458 pc/n, ¼ 0:60, " T 3 =" 0 ¼ 1822 and Q m ¼ 70 in the planar mode. The soft piezoelectric characteristics slightly increased then decreased as the amount of Sb substitution increased. The tendencies of, " T 3 =" 0 and d 33 were similar to that of density. Thus, Sb substitution aided in the solid state sintering of the ceramics and this resulted in a dense body and improved piezoelectric properties at a ¼ 0:1. Further Sb substitution, however, reduced the tetragonality of the ceramics which is shown in Fig. 1. This structural change and decreases in, " T 3 =" 0 and d 33 when a > 0:2 show the possibility that the system moves away from the MPB region with Sb substitution. 3.2 Li substitution Figure 4 shows the XRD patterns of the 0.8Pb(Zr 0:5 Ti 0:5 )- O 3 0.2Pb[(1 b)(zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 bli 1=4 (Nb 0:9 - Sb 0:1 ) 3=4 ]O 3 (b ¼ 0{0:4) ceramics. With the Li substitution, the () and () peak splits did not change much while keeping the () peak sharp. This means that the structure of ceramics was maintained to be in the tetragonal phase. Also, no secondary phase was detected in the XRD analysis. The microstructural variation upon Li substitution is shown in Fig. 5. The average grain size did not show a obvious change with the Li substitution. However, density increased with a small amount of Li substitution, as shown in Fig. 6, which also enhanced the dielectric permittivity " T 3 =" 0, electromechanical coupling factor

3 Jpn. J. Appl. Phys., Vol. 45, No. 4A (6) 5 µ m 2 µ m 5 µ m 2 µ m Fig. 2. SEM images of specimens sintered at 1 C for 2 h in 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb(Zn 0:8 Ni 0:2 ) 1=3 (Nb 1 a Sb a ) 2=3 O 3 ceramics: a ¼ 0, a ¼ 0:1, a ¼ 0:2, and a ¼ 0:3. Density (g/cm 3 ) Sb substitution 'a' Fig. 3. Densities, dielectric permittivities (" T 3 =" 0), electromechanical coupling factors ( ), mechanical quality factors (Q m ) and piezoelectric constant (d 33 ) of 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb(Zn 0:8 Ni 0:2 ) 1=3 (Nb 1 a Sb a ) 2=3 - O 3 ceramics sintered at 1 C for 2 h Qm d 33 (pc/n) Fig. 4. XRD patterns of specimens in 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 b)- (Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 bli 1=4 (Nb 0:9 Sb 0:1 ) 3=4 ]O 3 ceramics: b ¼ 0, b ¼ 0:1, b ¼ 0:2, b ¼ 0:3, and (e) b ¼ 0:4 sintered at 1 C for 2 h. and piezoelectric constant d 33 of Li substituted ceramics. The composition 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 b)(zn 0:8 - Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 bli 1=4 (Nb 0:9 Sb 0:1 ) 3=4 ]O 3 (b ¼ 0:3) exhibited ¼ 0:66, d 33 ¼ 548, " T 3 =" 0 ¼ 2468 and Q m ¼ 59 in the planar mode. The increases in soft piezoelectric characteristics may be due to the following: the role of Li as a sintering aid that can be assumed from the increase in density, and the increase in Nb and Sb amounts with Li substitution. This is because the ratio of Li to Nb 0:9 Sb 0:1 is 1:3, whereas that of Zn 0:8 Ni 0:2 to Nb 0:9 Sb 0:1 is 1:2. Consequently, with the increase in the amount of Li 1=4 - (e)

4 Jpn. J. Appl. Phys., Vol. 45, No. 4A (6) 2 µm 2 µm 2 µm 2 µm Fig. 5. SEM images of specimens in 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 b)(zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 bli 1=4 (Nb 0:9 Sb 0:1 ) 3=4 ]O 3 ceramics: b ¼ 0:1, b ¼ 0:2, b ¼ 0:3, and b ¼ 0:4 sintered at 1 C for 2 h. Density (g/cm 3 ) Li substitution 'b' Fig. 6. Density, dielectric permittivity (" T 3 =" 0), electromechanical coupling factor ( ), mechanical quality factor (Q m ) and piezoelectric constant (d 33 ) of specimens sintered at 1 C for 2 h in 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 b)(zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 bli 1=4 (Nb 0:9 Sb 0:1 ) 3=4 ]O 3 ceramics: b ¼ 0, b ¼ 0:1, b ¼ 0:2, b ¼ 0:3, and (e) b ¼ 0: Q m d 33 (pc/n) (Nb 0:9 Sb 0:1 ) substitution, Nb and Sb amounts also increase. This additional Nb and Sb can act as donors. Thus,, d 33, and " T 3 =" 0 could increase even though low valence Li was substituted for Zn and Ni. 3.3 Mn substitution Manganese is a well-known additive or substituent for enhancing both soft and hard characteristics because of its multi valence effect. On the other hand, Pb(Mn 1=3 Nb 2=3 )- O 3 and Pb(Mn 1=3 Sb 2=3 )O 3 are often used for relaxor modification in PZT ceramics. 14,15) In this study, Mn was substituted in 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[0.7(Zn 0:8 Ni 0:2 ) 1=3 - (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 ]O 3 ceramics. Figure 7 shows the XRD patterns of the samples sintered at 1 C for 2 h in 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 c){0.7(zn 0:8 - Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } cmn 1=3 (Nb 0:9 - Sb 0:1 ) 2=3 ]O 3 (c ¼ 0{0:4) ceramics. According to these patterns, Mn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 substitution increases tetragonality, as in the Sb substitution case. The microstructure of the 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 c){0.7(zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 - Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } cmn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 (c ¼ 0{0:4) ceramics in Fig. 8 exhibits almost the same average grain size without a second phase. Figure 9 shows the density, dielectric and piezoelectric properties of the 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 c)- {0.7(Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } cmn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 (c ¼ 0{0:4) ceramics. According to the formula, Mn was substituted for Zn, Ni, and Li which have the same or lower valence than manganese thus, the hardening effect could be predicted. The optimum 2670

5 Jpn. J. Appl. Phys., Vol. 45, No. 4A (6) Fig. 7. XRD patterns of specimens sintered at 1 C for 2 h in 0.8Pb- (Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 c){0.7(zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 - (Nb 0:9 Sb 0:1 ) 3=4 } cmn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 ceramics: c ¼ 0, c ¼ 0:1, c ¼ 0:2, c ¼ 0:3, and (e) c ¼ 0:4. characteristics for high power applications were obtained for the composition of 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 c)- {0.7(Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } cmn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3, when c ¼ 0:3. This composition yielded the coefficients ¼ 0:56, Q m ¼ 1951 (planar mode), d 33 ¼ 239 pc/n and " T 3 =" 0 ¼ Zr/Ti ratio change The soft characteristics of the 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb- [(1 c){0.7(zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 - Sb 0:1 ) 3=4 } cmn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 ceramics (c ¼ 0:3) (e) were, however, a little bit low. To improve this, Zr/Ti ratio was readjusted to obtain optimum properties. Figure 10 shows the XRD patterns of the 0.8Pb- (Zr x Ti 1 x )O 3 0.2Pb[0.7{0.7(Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } 0.3Mn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 ceramics (x ¼ 0:46{0:5) sintered at 1 C for 2 h. The split of the peak was not large when x ¼ 0:5 however, as Zr concentration is decreased, the split became significant for (), (), (), and () peaks, suggesting the crystal symmetry change from the MPB to the tetragonal. The density, dielectric and piezoelectric properties of 0.8Pb(Zr x Ti 1 x )O 3 0.2Pb[0.7{0.7(Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 - Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } 0.3Mn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]- O 3 ceramics (x ¼ 0:46{0:5) sintered at 1 C for 2 h are shown in Fig. 11., d 33 and dielectric constant increased until Zr amount decreased to x ¼ 0:48. Beyond this amount, the soft characteristics decreased again, and Q m was minimum when and d 33 were maximum. The optimum properties of ¼ 0:573, Q m ¼ 1502 (planar mode), d 33 ¼ 330 pc/n, and " T 3 =" 0 ¼ 1653 were obtained for the 0.8Pb- (Zr x Ti 1 x )O 3 0.2Pb[0.7{0.7(Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } 0.3Mn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 ceramics (x ¼ 0:48). In addition, k 31 and Q m in this mode were measured with rectangular samples of x ¼ 0:48 and 0.5. For x ¼ 0:5, k 31 ¼ 0:3 and Q m ¼ 1815, for x ¼ 0:48, k 31 ¼ 0:33 and Q m ¼ Q m in the 31 mode is smaller than that in the planar mode and the relationship between and k 31 is given as pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ k 31 2=ð1 Þ; ð1þ 2 µ m 2 µ m 2 µ m 2 µ m Fig. 8. SEM images of specimens sintered at 1 C for 2 h in 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 c){0.7(zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } cmn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 ceramics: c ¼ 0:1, c ¼ 0:2, c ¼ 0:3, and c ¼ 0:

6 Jpn. J. Appl. Phys., Vol. 45, No. 4A (6) Density (g/cm 3 ) Mn substitution 'c' Fig. 9. Density, dielectric permittivity (" T 3 =" 0), electromechanical coupling factor ( ), mechanical quality factor (Q m ) and piezoelectric constant (d 33 ) of specimens sintered at 1 C for 2 h in 0.8Pb- (Zr 0:5 Ti 0:5 )O 3 0.2Pb[(1 c){0.7(zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 - (Nb 0:9 Sb 0:1 ) 3=4 } cmn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 ceramics Qm d 33 (pc/n) Density (g/cm 3 ) Zr amount 'd' Fig. 11. Density, dielectric permittivity (" T 3 =" 0), electromechanical coupling factor ( ), mechanical quality factor (Q m ) and piezoelectric constant (d 33 ) of specimens sintered at 1 C for 2 h in 0.8Pb- (Zr d Ti 1 d )O 3 0.2Pb[0.7{0.7(Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 - Sb 0:1 ) 3=4 } 0.3Mn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 ceramics (d ¼ 0:46 { 0:5) Qm d 33 (h) (g) (f) (e) (i) Temperature ( o C) x=0.5 x=0.48 Fig. 10. XRD patterns of specimens sintered at 1 C for 2 h in 0.8Pb(Zr d Ti 1 d )O 3 0.2Pb[0.7{0.7(Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 - (Nb 0:9 Sb 0:1 ) 3=4 } 0.3Mn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 ceramics: d ¼ 0:5, d ¼ 0:495, d ¼ 0:49, d ¼ 0:485, (e) d ¼ 0:48, (f) d ¼ 0:475, (g) d ¼ 0:47, (h) d ¼ 0:465, and (i) d ¼ 0:46. Fig. 12. Temperature vs dielectric permittivity (" T 3 =" 0) of specimens sintered at 1 C for 2 h in 0.8Pb(Zr d Ti 1 d )O 3 0.2Pb[0.7{0.7- (Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } 0.3Mn 1=3 (Nb 0:9 - Sb 0:1 ) 2=3 ]O 3 ceramics (d ¼ 0:48, 0.5). where is Poisson s ratio. 3) Thus, values for x ¼ 0:5 and 0.48 can be obtained as 0.43 and 0.34, respectively. Figure 12 shows plots of the temperature dependence of dielectric permittivity for the two samples, x ¼ 0:48 and 0.5 of the 0.8Pb(Zr x Ti 1 x )O 3 0.2Pb[0.7{0.7(Zn 0:8 Ni 0:2 ) 1=3 - (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } 0.3Mn 1=3 (Nb 0:9 - Sb 0:1 ) 2=3 ]O 3 ceramics. When Zr/Ti ratio was changed from 0.5/0.5 to 0.48/0.52 the Curie temperature increased from 285 to 289 C. Furthermore, the slight hump at approximately 160 C in the composition with x ¼ 0:50 at the exact MPB composition, which may have originated from the 2672 rhombohedral to tetragonal phase transition, disappeared for x ¼ 0:48, suggesting the pure tetragonal phase and more stable temperature dependence of dielectric properties for this composition. To confirm the applicability of the compositions under high-voltage/power drive condition, vibration velocity was measured for the 0.8Pb(Zr x Ti 1 x )O 3 0.2Pb[0.7{0.7- (Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } 0.3Mn 1=3 (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 ceramics (x ¼ 0:48, 0.5) as a function of applied electric field (rms value). The k 31 fundamental mode was used for this measurement (sample size: mm 3 ), and the measurement was conducted

7 Jpn. J. Appl. Phys., Vol. 45, No. 4A (6) Vibration Velocity (m/s) x=0.5 x= Applied Field (V/mm, rms value) 0:33, Q m ¼ 1502 (planar mode), Q m ¼ 1404 (31 mode), d 33 ¼ 330 pc/n, " T 3 =" 0 ¼ 1653 and 31-mode maximum vibration velocity = 0.58 m/s were obtained at Zr/Ti = 0.48/0.52. even though these results show a 40% lower vibration velocity than that in our previous work, this study did yield a 43% enhancement in the k 31, which was the objective of this work. 3,4,7) Acknowledgement This work was supported by the Office of Naval Research, U.S.A. through the Contract N Fig. 13. Vibration velocity variation of 0.8Pb(Zr a Ti 1 a )O 3 0.2Pb[0.7{0.7- (Zn 0:8 Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } 0.3Mn 1=3 (Nb 0:9 - Sb 0:1 ) 2=3 ]O 3 ceramics (a ¼ 0:48 and 0.5). until the temperature at the sample center portion (i.e., the nodal line) was 20 C in excess of room temperature. In Fig. 13, both x ¼ 0:5 and 0.48 specimens showed high maximum vibration velocities of 0.6 and 0.58 m/s, respectively. The right-hand side and points correspond to a 20 C temperature rise from room temperature. 4. Conclusions High power compositions with a reasonably high electromechanical coupling factor k were reported in this paper. The specimen 0.8Pb(Zr 0:5 Ti 0:5 )O 3 0.2Pb[0.7{0.7(Zn 0:8 - Ni 0:2 ) 1=3 (Nb 0:9 Sb 0:1 ) 2=3 0.3Li 1=4 (Nb 0:9 Sb 0:1 ) 3=4 } 0.3Mn 1=3 - (Nb 0:9 Sb 0:1 ) 2=3 ]O 3 ceramics sintered at 1 C for 2 h, yielded the coefficients ¼ 0:56, k 31 ¼ 0:3, Q m ¼ 1951 (planar mode), Q m ¼ 1815 (31 mode), d 33 ¼ 239 pc/n, " T 3 =" 0 ¼ 739 and a maximum vibration velocity in the 31 mode (20 C temperature rise) = 0.6 m/s. By adjusting Zr/ Ti ratio, much improved properties of ¼ 0:573, k 31 ¼ 1) K. Uchino: Piezoelectric Actuators and Ultrasonic Motors (Kluwer Academic, Dordrecht, 1996). 2) K. Uchino: Ferroelectric Devices (Marcel Dekker, New York, 0). 3) K. Uchino and J. R. Giniewicz: Micromechatronics (Marcel Dekker, New York, 3). 4) K. Uchino, J. Zheng, A. Joshi, Y. H. Chen, S. Yoshikawa, S. Hirose, S. Takahashi and J. W. C. De Vries: J. Electroceram. 2 (1998) 33. 5) M. Kondo, M. Hida, M. Tsukada, K. Kurihara and N. Kamehara: J. Ceram. Soc. Jpn. 105 (1997) ) D. Luff, R. Lane, K. R. Brown and H. J. Marshallsay: Trans. Br. Ceram. Soc. 73 (1974) ) Y. Gao, Y. Chen, J. Ryu, K. Uchino and D. Viehland: Jpn J. Appl. Phys. 40 (1) ) S. Priya, K. Uchino, J. Ryu, C. Ahn and S. Nahm: Appl. Phys. Lett. 83 (3) ) C. Ahn, H. Song, S. Park, S. Nahm, K. Uchino, S. Priya, H. Lee and N. Kang: Jpn. J. Appl. Phys. 44 (5) ) Z. Zhu, B. Li, G. Li, W. Zhang and Q. Yin: Mater. Sci. Eng. B 117 (5) ) J. Yoo, C. Lee, Y. Jeong, K. Chung, D. Lee and D. Paik: Mater. Chem. Phys. 90 (5) ) C. Galassi, E. Roncari, C. Capiani and F. Craciun: J. Eur. Ceram. Soc. 19 (1999) ) J. Yoo, J. Hong and S. Suh: Sens. Actuators 78 (1999) ) C. Chen and H. Lin: Ceram. Int. 30 (4) ) Y. Hou, M. Zhu, H. Wang, B. Wang, C. Tian and H. Yan: J. Eur. Ceram. Soc. 24 (4)