Microstructure, electrical properties and temperature stability in Bi_0.5Na_0.5Zr_0.95Ce_0.05O₃ modified R–T phase boundary of potassium-sodium niobium lead-free ceramics

Abstract

(1 − x)K_0.48Na_0.52Nb_0.95Sb_0.05O₃–xBi_0.5Na_0.5Zr_0.95Ce_0.05O₃ [(1 − x)KNNS–xBNZC] lead-free piezoelectric ceramics, with doping ratio of x ranging from 0 to 0.05, were synthesized by the conventional solid state sintering method. The phase transition behavior, microstructure and piezoelectric properties of (1 − x)KNNS–xBNZC ceramics were systematically investigated using XRD, SEM, and other devices with different doping amounts of BNZC. It was found that the piezoelectric properties of (1 − x)KNNS–xBNZC ceramics were improved obviously by adding the proper doping amount, 0.03 < x < 0.04, due to the coexistence of rhombohedral and tetragonal phases in the ceramics near room temperature. The piezoelectric constant d₃₃ of the ceramics first increased and then decreased when increasing the doping amount. A remarkably strong piezoelectricity was obtained in ceramics with a ∼441 pC N⁻¹ peak d₃₃ value. The excellent piezoelectric properties of (1 − x)KNNS–xBNZC ceramics with x = 0.034 were obtained: d₃₃ = 441 pC N⁻¹, k_p = 0.44, Q_m = 31, ε_r = 2447, tan δ = 0.037, T_C = 215 °C, P_r = 15.7 μC cm⁻² and E_C = 8.2 kV cm⁻¹. With the annealing temperature reaching 250 °C, the d₃₃ values of the ceramics were still greater than 330 pC N⁻¹, which represents good temperature stability for the piezoelectric property. It is believed that such a material system is a very promising candidate for lead-free piezoelectric ceramics.

---

1. Introduction

Lead-based piezoelectric ceramics, such as PZT and PMN-PT, have been widely used in various fields as actuators, sensors and transducers due to their excellent piezoelectric properties.^1,2 However, high lead content of lead-based piezoelectric ceramics, which is more than 60%, causes environmental issues during the whole product life period, including the ceramics preparation, processing, and even disposal.^3–6 Therefore, considerable attention has been paid to the replacement of lead-based materials by the lead-free piezoelectric materials in electronic products. Intensive research has focused on several lead free candidate materials such as BaTiO₃, Bi_0.5Na_0.5TiO₃ and K_0.5Na_0.5NbO₃ (abbreviated as BT, BNT and KNN). In the last decade, for KNN-based ceramics, there were special concerns that led to interest in developing lead-free materials with a large piezoelectric response, strong ferroelectric behaviour and high Curie temperature.^3–5,7

In past research, constructing boundaries of coexisting phases was one of the most effective ways to enhance the piezoelectric properties of KNN-based ceramics such as orthorhombic–tetragonal (O–T) phases and rhombohedral–orthorhombic (R–O) phases coexistences.^4,5,8–13 Two phase boundaries are constructed in KNN-based ceramics at room temperature by enhancing the R–O phase transition point to room temperature or reducing the O–T phase transition point. Researchers have made great achievements with KNN-based ceramics with phase boundaries, which present large piezoelectric properties with more polarization orientation on the phase boundary.⁵ Furthermore, R. Zuo and W. Liang suggested that the rhombohedral–tetragonal (R–T) boundary is constructed in KNN-based ceramics by shifting T_R–O and T_O–T to room temperature.^14,15 An excellent piezoelectric constant (d₃₃ ∼ 380 pC N⁻¹ and 344 pC N⁻¹, respectively) proved that the constructed R–T boundary is a feasible method for enhancing piezoelectric properties in KNN-based ceramics.^14,15

In a previous report, Sb⁵⁺ and Bi_0.5Na_0.5ZrO₃ were used to increase T_R–O and decrease T_O–T, respectively.^5,16–19 On the other hand, CeO₂ is often used as a donor dopant or additive for lead-based piezoelectric ceramics to improve their electrical properties.²⁰ It has also been reported that Ce-doping could improve dielectric and piezoelectric properties in KNN-based ceramics.²¹ In this study, Zr of Bi_0.5Na_0.5ZrO₃ was replaced by a small amount of Ce to form a new doping component Bi_0.5Na_0.5Zr_0.95Ce_0.05O₃. The ceramics with a series of compositions (1 − x)K_0.48Na_0.52Nb_0.95Sb_0.05O₃–xBi_0.5Na_0.5Zr_0.95Ce_0.05O₃ [(1 − x)KNNS–xBNZC] were synthesized by a conventional solid-state sintering method. It was found that the R–T boundary had been constructed in (1 − x)KNNS–xBNZC ceramics at room temperature when 0.03 < x < 0.04. A large d₃₃ of 441 pC N⁻¹ was obtained for ceramics with x = 0.034, it is larger than what Saito et al. had previously reported (d₃₃ = 416 pC N⁻¹).⁷ Finally, the ceramics phase structures, microstructures, piezoelectric, ferroelectric and dielectric properties and thermal stability of the (1 − x)KNNS–xBNZC ceramics were investigated and discussed.

2. Experimental procedure

In this study, (1 − x)K_0.48Na_0.52Nb_0.95Sb_0.05O₃–xBi_0.5Na_0.5Zr_0.95Ce_0.05O₃ [(1 − x)KNNS–xBNZC] ceramics were prepared by a conventional solid state sintering method. Na₂CO₃ (99.8%), Nb₂O₅ (99.5%), Bi₂O₃ (99.999%), K₂O₃ (99%), Sb₂O₃ (99.99%), ZrO₂ (99%), and CeO₂ (99.99%) were used as raw powders and were weighed according to stoichiometric ratio of the ceramics. The powders were mixed by ball-milling in ethanol for 12 h followed by drying and calcination at 850 °C for 6 h. The dried powders were mixed with poly vinyl alcohol (PVA, 8%) and pressed into discs with a 10 mm diameter of and 1 mm thickness with 10 MPa pressure. After burning the PVA, the pellets were sintered at a temperature range of 1070–1140 °C for 3 h in air. Silver paste electrodes were coated on both sides of the pellets followed by frying at 700 °C for 10 min. These ceramics were poled while in silicon oil under a 2–4 kV mm⁻¹ DC electric field for 10 to 30 min from room temperature to 60 °C, and all the piezoelectric properties of the samples were measured after the poling for 24 h.

The structural properties of the samples were investigated by X-ray diffraction (XRD) (DX 2700, Dandong, China). The surface morphologies were characterized by field emission scanning electron microscopy (FE-SEM, JSM7500, Japan). The curves of dielectric constant (ε_r) against temperatures of the sintered samples were measured using an LCR analyzer (HP 4980, Agilent, U.S.A. and TH2816A). Polarization versus electric field (P–E) hysteresis loops of the ceramics were conducted at 10 Hz using a Radiant Precision Work-station (USA). Their piezoelectric constant d₃₃ and electromechanical coupling factor k_p was characterized by a Belincourt meter (ZJ-3A, Institute of Acoustics, Chinese Academy of Sciences, China) and an impedance analyzer (HP 4294A, Agilent, U.S.A.) according to IEEE standards.

3. Results and discussion

Fig. 1(a) shows XRD patterns fit the (1 − x)KNNS–xBNZC ceramics. From Fig. 1(a), it was found that all the ceramics form a pure perovskite phase, which presents the formation of a stable solid solution. Fig. 1(b) gives the enlarged XRD patterns of the ceramics in the 44–47° measuring range. The ceramics presented an orthorhombic phase when x was less than 0.01. When the doping content of BNZC was increased, a broad and split peak was observed at the x = 0.2 composition in Fig. 1(b). The coexistence of the rhombohedral phase, orthorhombic phase and tetragonal phase was considered in the ceramics according to the two close dielectric peaks in Fig. 2(b), which implied the coexistence of three phases. The XRD pattern of the ceramics was simulated using a Lorentz function, as shown in Fig. 1(c). The result was consistent with our prediction. For 0.03 ≤ x ≤ 0.04, the coexistence of the rhombohedral phase and tetragonal phase appeared, which was confirmed by Fig. 1(d) and 2(c) and (e). At the same time, the ceramics with x = 0.05 showed a single peak in Fig. 2(b). The ceramics are considered as the R phase because of the P–E loop and non-zero value of d₃₃ [Fig. 5(a) and 6(a)]. We also calculated their crystal parameters, showing a = b = c = 3.9859 Å and α = β = γ = 89.9878°. The results further indicated that the ceramics with x = 0.05 were R phase.

Fig. 1 XRD patterns of the (1 − x)K_0.48Na_0.52Nb_0.95Sb_0.05O₃–xBi_0.5Na_0.5Zr_0.95Ce_0.05O₃ ceramics.

Fig. 2 Temperature-dependent dielectric constant of the ceramics with (a) x = 0, (b) x = 0.02, (c) x = 0.03, (d) x = 0.034, (e) x = 0.04, (f) x = 0.05, measured at 100 kHz in the −120–200 °C temperature range.

The temperature-dependence dielectric constant (ε_r) of the ceramics from −120 to 200 °C was characterized to investigate the influence of doping BNZC on the phase transition of KNN-based ceramics, as shown in Fig. 2. Two phase transition peaks at T_R–O and T_O–T could be found in the ceramics with x = 0 and 0.02. The phase transition peaks of T_O-T gradually decreases and T_R-O gradually increases with x from 0 to 0.04. Eventually, as shown in Fig. 2(c) to (e), the two peaks merged together and became one peak. Considering both XRD patterns and the ε_r versus T curves in Fig. 1 and 2, the R–T phase boundary had been constructed successfully in the ceramics at room temperature when x was in the 0.03 to 0.04 range of. However, the R–T phase boundary was restrained with further increasing the BNZC content (x = 0.05), as shown in Fig. 2(f). At the same time, Fig. 2(f) shows a diffused Curie peak (Curie temperature-T_C) because of the dramatic decrease in grain size [see Fig. 4(d)].

For further study of the BNZC content on T_C and the phase evolution, the temperature dependent dielectric constant (ε_r) of the (1 − x)KNNS–xBNZC ceramics was measured at 100 kHz in the 30–450 °C, temperature range, as shown in Fig. 3(a). The T_C presents the tendency of decreasing gradually with increasing x. The curves were comparatively flat when x = 0.05, which indicated that the T_C was diffused. According to the results of Fig. 2(a)–(f) and 3(a), the phase diagram was drawn to show T_C, T_O–T and T_R–O of the (1 − x)KNNS–xBNZC ceramics as function of BNZC [Fig. 3(b)]. The figure clearly shows that T_C, T_O–T and T_R–O of the (1 − x)KNNS–xBNZC ceramics changed with x. It was found that R–T phase boundary was formed at room temperature in the (1 − x)KNNS–xBNZC ceramics with 0.03 ≤ x ≤ 0.04.

Fig. 3 (a) Temperature dependent dielectric constant of the ceramics as a function of x. (b) Phase diagram of (1 − x)KNNS–xBNZC ceramics.

To identify the evolution of microstructures of the ceramics, their micrographs were investigated. Fig. 4(a)–(d) show the SEM surface micrographs of the ceramics with x = 0, 0.02, 0.034, and 0.05. It was found that the grain size gradually increased with increasing BNZC content when x was less than 0.034 and then dropped dramatically with further increasing the BNZC content (x = 0.05). In addition, the ceramics with x ≤ 0.034 showed nonuniform grain sizes. Small grains exist in grain boundaries and were distributed around the large ones. This is probably caused by the following reasons: (I) for low BNZC content, the liquid phase formed due to Bi having a lower melting point in the sintering process, which could promote the grain growth of the ceramics giving rise to the increase in grain size.²² (II) For higher BNZC content, excess BNZC gathers at grain boundaries, and Zr and Ce prohibited the grain growth during the sintering process, resulting in significantly decreased grain size to less than 1 μm, as shown in Fig. 4(d).^21,23

Fig. 4 Surface SEM micrographs of the ceramics with (a) x = 0, (b) x = 0.02, (c) x = 0.034, (d) x = 0.05.

The ferroelectric switching behavior of the ceramics was studied in terms of their P–E loops. Fig. 5(a) shows the P–E curves of the ceramics as x was increased from 0 to 0.05, and they were measured at room temperature and 10 Hz. All the ceramics had a typical P–E loop except for the one with x = 0.05, because the ceramics with x = 0.05 possesses the rhombohedral phase.¹⁴ For the study of ferroelectric properties in detail as a function of x, the remanent polarization (P_r) and coercive field (E_C) were derived from P–E loops of Fig. 5(a), as shown in Fig. 5(b). From Fig. 5, it was found that the P_r of the ceramics presents the tendency of first increasing and then decreasing with increasing x. The P_r maximum was at x = 0.01 and this value gradually dropped with further increasing BNZC content. E_C presents the downtrend on the whole as BNZC content was increased, but it showed abnormal fluctuations in the x = 0.03–0.04 composition range, which was just at the R–T phase boundary.

Fig. 5 (a) P–E curves and (b) P_r and E_C versus x of the ceramics.

Fig. 6(a) shows the variations of d₃₃, planar coupling factor (k_p) and mechanical quality factor (Q_m) versus x of the ceramics, measured at room temperature after 24 hours of polarization. From Fig. 6(a), it was found that the d₃₃ of the ceramics clearly increased as x changed from 0 to 0.034 and then dropped quickly as x was further increased. The k_p of the ceramics showed the trend of first increasing and then gradually decreasing with increasing x, whereas the Q_m of the ceramics first decreased and then increased as x changed from 0 to 0.04, with a minimum of 31 at x = 0.034.

Fig. 6 (a) d₃₃, k_p and Q_m versus x of the ceramics. (b) ε_r and tan δ versus x of the ceramics.

The dielectric constant (ε_r) and dielectric loss (tan δ) of the ceramics with different BNZC contents are shown in Fig. 6(b). The ε_r increased as x was increased, reaching a maximum at x = 0.036, and then decreased with further increasing of x. In addition to x = 0.05, the trend of tan δ was just the opposite compared with ε_r, and it has a lower value (tan δ = 0.036–0.038) when x = 0.03–0.038. In this study, d₃₃ reaches a maximum value of 441 pC N⁻¹ for x = 0.034 (k_p = 0.44, Q_m = 31, ε_r = 2447, tan δ = 0.037, T_C = 215 °C, P_r = 15.7 μC cm⁻², E_C = 0.82 kV mm⁻¹), which was larger than those of most others reported, as shown in Table 1; the piezoelectric properties and Curie temperature T_C of some typical KNN-based lead-free ceramics with large d₃₃ were also listed. Moreover, the large piezoelectric constant of the ceramics in this study is almost comparable to lead-based piezoelectric ceramics such as PZT4.²⁸

Table 1 Piezoelectric properties and T_C of typical KNN-based lead-free ceramics

Material system	Phase	d33 (pC N^−1)	kp	TC (°C)	Ref.
(K0.44Na0.52Li0.04)(Nb0.86Ta0.10Sb0.04)O3	O–T	416	0.61	∼253	7
(K0.52Na0.40)(Nb0.83Sb0.09)O3–0.08LiTaO3	O–T	400	0.54	∼230	35
(Na0.52K0.40)(Nb0.84Sb0.08)O3–LiTaO3–BaZrO3	R–T	365	0.45	∼170	14
KNN–BaZrO3–LiSbO3	R–T	344	0.324	∼176	15
(1 − x)K0.5Na0.5Nb1−xSbxO3–xBi0.5Na0.5TiO3	R–T	380	0.35	∼276	36
(1 − x)(K0.42Na0.58)(Nb0.96Sb0.04)O3–x(Bi0.5Na0.5)0.90Mg0.10ZrO3	R–T	434	0.47	∼244	24
(1 − x)K0.48Na0.52Nb0.95Sb0.05O3–xBi0.5Na0.5Zr0.95Ce0.05O3	R–T	441	0.44	∼215	This work

---

The ceramics with large d₃₃ have three important factors, (I) doping BNZC makes T_O–T gradually decrease and T_R–O gradually increase, and then the ceramics form R–T phases that coexisted at room temperature when x was in the 0.03 to 0.04 range. The formation of the R–T phase coexistence was responsible for such a large d₃₃ value of this study.^{14,15,24–26,33} (II) It was reported that d₃₃ is proportional to ε_rP_r in piezoelectric ceramics. Larger ε_r were obtained in the ceramics with 0.03 < x < 0.04 [Fig. 6(b)], which led to enhanced piezoelectric properties.^26–28 (III) It is well known that the increased grain sizes result in the enhance piezoelectric properties, because the smaller grain sizes are in favor of the reduction because of the number of domain variants.^29–32 Fig. 4(c) shows giant grain sizes when x = 0.034. In addition, the poor d₃₃ of the ceramics with x = 0.05 was partly attributed to the low and diffuse Curie peak.

Fig. 7(a) shows the thermal stability of the ceramics as a function of x, measured in the 30–350 °C annealing temperature range (the annealing time was 30 min). The d₃₃ of all the ceramics decreased with the increased annealing temperature (T_a), and most of it dropped dramatically when the T_a was close to T_C. However, the thermal stability of d₃₃ of the ceramics with x = 0.03 and 0.034 was extraordinary. Compared with the T_C (215 °C) of x = 0.034, its d₃₃ value was gently reduced until the T_a was over 250 °C. Fig. 7(b) shows the Δd₃₃/d₃₃ vs. T_a curves for all the ceramics. The ceramics with x = 0 and 0.02 exhibited a better thermal curve of d₃₃ due to the orthorhombic structure.^34,37 The value of d₃₃ of the ceramics with x = 0.04 quickly decreased as T_a was increased. That is partly because of the lower and diffused Curie dielectric peaks. In conclusion, the ceramics with x = 0.04 still exhibited a larger d₃₃ value (d₃₃ > 330 pC N⁻¹, Δd₃₃/d₃₃ > 75%) even if the annealing temperature reached 250 °C and had surpassed the T_C (215 °C). In this case, the annealing time was insufficient and the Curie dielectric peak was in the 190–290 °C range. We think these two reasons make a larger d₃₃ value when T_a reached 250 °C.

Fig. 7 (a) Thermal stability of d₃₃ of (1 − x)KNNS–xBNZC ceramics, and the insets are the thermal stability of the d₃₃ of the ceramics with x = 0.034. (b) Δd₃₃/d₃₃ vs. T_a of the (1 − x)KNNS–xBNZC ceramics.

4. Conclusion

(1 − x)K_0.48Na_0.52Nb_0.95Sb_0.05O₃–xBi_0.5Na_0.5Zr_0.95Ce_0.05O₃ ceramics were synthesized using a conventional solid-state method. The ceramics with 0.03 < x < 0.04 exhibited coexistence of both tetragonal and rhombohedral phases at room temperature. Enhanced piezoelectric and dielectric properties were obtained in the ceramics with compositions located in the phase boundary zone. In this study, the ceramics with x = 0.034 showed the maximum value of d₃₃ (∼441 pC N⁻¹), and good temperature stability for the piezoelectricity property of these ceramics was obtained (d₃₃ was still more than 330 pC N⁻¹ when T_a reached 250 °C). As a result, such a material system is a promising candidate for lead-free piezoceramics.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (No. 51332003 and 61201064). The authors also thank Ms Hui Wang for measuring the FE-SEM images.

References

1. B. Jaffe, R. S. Roth and S. Marzullo, J. Res. Natl. Bur. Stand., 1955, 55, 239–254.
2. B. Jaffe, Piezoelectric Ceramics, Academic Press, India, 1971, ch. 7.
3. T. R. Shrout and S. J. Zhang, J. Electroceram., 2007, 19, 113–126.
4. J. Rödel, W. Jo, K. T. P. Seifert and E. M. Anton, J. Am. Ceram. Soc., 2009, 92, 1153–1177.
5. J. Wu, D. Xiao and J. Zhu, Chem. Rev., 2015, 115, 2559–2595.
6. N. Ledermann, P. Muralt, J. Baborowski, S. Gentil, K. Mukati, M. Cantoni, A. Seifert and N. Setter, Sens. Actuators, A, 2003, 105, 162–170.
7. Y. Saito, H. Takao, T. Tani, T. Nonoyama, K. Takatori, T. Homma, T. Nagaya and M. Nakamura, Nature, 2004, 432, 84–87.
8. J. Wu, D. Xiao, Y. Wang, J. Zhu, L. Wu and Y. Jiang, Appl. Phys. Lett., 2007, 91, 252907.
9. K. Wang and J. Li, Adv. Funct. Mater., 2010, 20, 1923–1929.
10. R. Zuo, Y. Chun and X. Fang, Jpn. J. Appl. Phys., 2007, 46, 6733–6736.
11. E. K. Akdogan, K. Kerman, M. Abazari and A. Safari, Appl. Phys. Lett., 2008, 92, 112908.
12. R. P. Wang, H. Bando, T. Katsumata, Y. Inaguma, H. Taniguchi and M. Itoh, Phys. Status Solidi RRL, 2009, 5, 142–144.
13. R. Zuo, J. Fu, D. Lv and Y. Liu, J. Am. Ceram. Soc., 2010, 93(9), 2783.
14. R. Zuo and J. Fu, J. Am. Ceram. Soc., 2011, 94(5), 1467–1470.
15. W. Liang, W. Wu, D. Xiao, J. Zhu and J. Wu, J. Mater. Sci., 2011, 46(21), 6871–6876.
16. Q. Gou, D. Xiao, B. Wu, M. Xiao, S. Feng, D. Mazhao, J. Wu and J. Zhu, RSC Adv., 2015, 5, 30660–30666.
17. J. Wu, H. Tao, Y. Yuan, X. Lv, X. Wang and X. Lou, RSC Adv., 2015, 5, 14575–14583.
18. D. Lin, K. W. Kwok, H. Tian and H. W. L. Chan, J. Am. Ceram. Soc., 2007, 90(5), 1458–1462.
19. Y. Yuan, J. Wu, H. Tao, X. Lv, X. Wang and X. Lou, J. Appl. Phys., 2015, 117, 084103.
20. B. Sahoo and P. K. Panda, J. Mater. Sci., 2007, 42, 9684–9688.
21. D. Gao, K. W. Kwok, D. Lin and H. L. W. Chan, J. Mater. Sci., 2009, 44, 2466–2470.
22. H. Du, F. Luo, S. Qu, Z. Pei, D. Zhu and W. Zhou, J. Appl. Phys., 2007, 102, 054102.
23. B. Malic, J. Bernard, A. Bencan and M. Kosec, J. Eur. Ceram. Soc., 2008, 28, 1191.
24. J. Wu, Y. Yang, X. Wang, D. Xiao and J. Zhu, J. Mater. Sci.: Mater. Electron., 2014, 25, 4650–4656.
25. X. Cheng, J. Wu, X. Wang, B. Zhang, J. Zhu, D. Xiao, X. Wang and X. Lou, Appl. Phys. Lett., 2013, 103, 052906.
26. X. Wang, J. Wu, D. Xiao, J. Zhu, X. Cheng, T. Zheng, B. Zhang, X. Lou and X. Wang, J. Am. Chem. Soc., 2014, 136(7), 2905–2910.
27. C. A. Randall, N. Kim, J. P. Kucera, W. Cao and T. R. Shrout, J. Am. Ceram. Soc., 1998, 81(3), 677–688.
28. S. E. Park and T. R. Shrout, IEEE Trans. Sonics Ultrason., 1997, 44, 1140.
29. Z. Zhao, V. Buscaglia, M. Viviani, M. T. Buscaglia, L. Mitoseriu, A. Testino, M. Nygren, M. Johnsson and P. Nanni, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 70, 024107.
30. W. Cao and C. A. Randall, J. Phys. Chem. Solids, 1996, 57(10), 1499.
31. G. Arlt and N. A. Pertsev, J. Appl. Phys., 1991, 70(4), 2283.
32. J. Wu, X. Wang, X. Cheng, T. Zheng, B. Zhang, D. Xiao, J. Zhu and X. Lou, J. Appl. Phys., 2014, 115, 114104.
33. J. Wu, Y. Wang and H. Wang, RSC Adv., 2014, 4, 64835–64842.
34. Y. Gao, J. L. Zhang, X. J. Zong, C. L. Wang and J. C. Li, J. Appl. Phys., 2010, 107, 074101.
35. R. Zuo, J. Fu and D. Lv, J. Am. Ceram. Soc., 2009, 92(1), 283–285.
36. W. Liang, Z. Wang, D. Xiao, J. Wu, W. Wu, T. Huang and J. Zhu, Integr. Ferroelectr., 2012, 139, 63–74.
37. D. Lin, Z. Li, S. Zhang, Z. Xu and X. Yao, Solid State Commun., 2009, 149, 1646–1649.

---