Dielectric and piezoelectric properties of (1−x)Ba0.7Sr0.3TiO3−xBa0.7Ca0.3TiO3 perovskites

Journal of Physics and Chemistry of Solids 73 (2012) 957–960

Contents lists available at SciVerse ScienceDirect

Journal of Physics and Chemistry of Solids journal homepage: www.elsevier.com/locate/jpcs

Dielectric and piezoelectric properties of (1  x)Ba0.7Sr0.3TiO3  xBa0.7Ca0.3TiO3 perovskites Jiangying Wang a,n, Xiaoying Zhang a, Jingji Zhang a, Huiling Li b, Zhengfa Li a a b

College of Materials Science and Engineering, China Jiliang University, Hangzhou 310018, China College of Modern Science and Technology, China Jiliang University, Hangzhou 310018, China

a r t i c l e i n f o

abstract

Article history: Received 24 October 2011 Received in revised form 20 February 2012 Accepted 6 March 2012 Available online 14 March 2012

Dielectric and piezoelectric properties of (1  x)Ba0.7Sr0.3TiO3  xBa0.7Ca0.3TiO3 (BST  xBCT) (x ¼ 0.2–0.9) perovskite ceramics have been investigated. BCT has fully incorporated into BST lattice, forming a complete perovskite solid solution, whose lattice constant w decreases almost linearly with increase in x from 0.2 to 0.4, while showing an anomalous expansion at 0.4 ox r0.6. This, together with the deviation of tetragonal–orthorhombic phase transition temperature (TO–T) from the linear relation TO–T (K) ¼  103.7x þ 239.3 at x ¼ 0.5, suggests that a small amount of Ca2 þ has substituted for Ti4 þ . Curie temperature TC increases linearly with increase in x from 0.2 to 0.9, which is mainly contributed to the increase of the Ba/Sr ratio. The calculated degree of relaxation (g) is in the range of 1.41–1.53, indicating that the BST–xBCT ceramics are ferroelectric materials with diffuse phase transition. Strain and piezoelectric constant (d33) decrease with increasing x, whereas planar electromechanical coefficient (kp) reaches a maximum (17.0%) at x ¼ 0.6. & 2012 Elsevier Ltd. All rights reserved.

Keywords: A. Ceramics D. Phase transitions D. Dielectric properties D. Piezoelectricity

1. Introduction Barium titanate (BaTiO3) was the first perovskite-type ferroelectric and piezoelectric material developed and studied extensively since its discovery in about the early 1940s [1]. It goes through a paraelectric–ferroelectric phase transition from cubic to tetragonal at 130 1C and two ferroelectric polymorphic transitions: tetragonal to orthorhombic at 0 1C and orthorhombic to rhombohedral at  90 1C [1]. Sr2 þ and Ca2 þ ions substitutions for Ba2 þ ions can form solid solutions, such as Ba1  xSrxTiO3 and Ba1 xCaxTiO3. These solid solutions have been, and continue to be, of interest for investigation, not only because of their wide applications in the fields of capacitor [2], memory storage [3] and microwave devices [4], but also for their interesting dielectric and ferroelectric behaviors [5–8]. Ba1  xSrxTiO3 is a solid solution throughout the substitution concentrations: range from 0 to 1, whose TC decreases linearly with respect to the substitution concentrations [7,9,10]. According to Lemanov et al. [6], the variation of TC is usually ascribed to the A-site cation size effect, in which the smaller Sr2 þ ions cause a decrease in the average radius of the A-cation. There is a different case for Ba1  xCaxTiO3. Early reports [5,8] revealed that Ca2 þ replaced Ba2 þ ions to form Ba1  xCaxTiO3 solid solutions with x up to 0.23, which had a slight effect on TC. This effect seems

n

Corresponding author. Tel.: þ86 571 86875609; fax: þ 86 571 86875613. E-mail address: [email protected] (J. Wang).

0022-3697/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jpcs.2012.03.004

to be inconsistent with the A-site cation size effect. For the Ba0.4Sr0.6  xCaxTiO3 system [11], it was found that TC increased linearly with increase in x from 0 to 0.15, while deviating from linear behavior for x 40.15. Up to x¼0.25, TC increased persistently and thereafter decreased. However, the variation of TO–T with Ca concentration has not been reported till now. In addition, Ba0.7Ca0.3TiO3 is usually added into Bi0.5Na0.5TiO3 [12] and BaZr0.2Ti0.8O3 [13] to modify their piezoelectric properties. Thus, this study was focused on the dielectric and the piezoelectric properties of the BST–xBCT (x¼0.2–0.9) ceramics. The relationship between property and structure was also discussed.

2. Experimental procedure All the reagents were of analytical grade (Sinopharm Chemical Reagent Co., Ltd., China) and were used as received without further purification. (1  x)Ba0.7Sr0.3TiO3 xBa0.7Ca0.3TiO3 (x ¼0.2– 0.9) ceramics were prepared by using a solid-state reaction method. BaCO3 (99.0%), SrCO3 (99.0%), CaCO3 (99.0%) and TiO2 (98.0%) powders were used as starting materials for the synthesis of Ba0.7Sr0.3TiO3 and Ba0.7Ca0.3TiO3 powders at 1100 1C. Thereafter, the powders were mixed according to the stoichiometric ratio of BST–xBCT (where x¼0.2, 0.3, 0.4, 0.5, 0.6, 0.7, and 0.9) and milled in polypropylene bottles with zirconia grinding media for 24 h. The dried powders were granulated using 8 wt% polyvinyl alcohol as a binder and then pressed to yield the diameter and thickness of 10 mm and 1mm, respectively. The green pellets

3. Results and discussion

0.3985

1.0045

0.3980

1.0040 a c X c/a

0.3970

(103) (310)

(202)

(212)

(211)

(201)

(002) (200)

(111)

(101) (100)

x=0.9

x=0.7 x=0.6 x=0.5 x=0.4 x=0.3 x=0.2

20

30

40 50 60 2 Theta (degrees)

70

80

Fig. 1. XRD patterns of the BST–xBCT (x ¼0.2, 0.3, 0.4, 0.5, 0.6, 0.7 and 0.9) ceramics.

1.0030

0.3965 1.0025 0.3960 0.2

0.3

0.4

0.5

0.6

0.7

0.8

1.0020 0.9

x Fig. 2. Lattice constants a, c and w of the BST–xBCT ceramics as functions of x.

10000

x=0.2 x=0.3 x=0.4 x=0.5 x=0.6 x=0.7 x=0.9

360 340

8000 6000 4000 0.2

0.4

0.6 x

0.8

320 220 210 200 190 180 1.0

2000

Loss tangent

XRD patterns of the BST–xBCT (x¼0.2, 0.3, 0.4, 0.5, 0.6, 0.7 and 0.9) ceramics are given in Fig. 1. It can be seen that all samples have a perovskite tetragonal structure without impurity phases, which indicates that BCT has been fully incorporated into BST lattice forming a complete perovskite solid solution. Lattice constants a, c and wffi (for tetragonal symmetry, the ffiffiffiffiffiffiffi p 3 average lattice constant w ¼ a2 c is taken, where a and c are the lattice parameters of tetragonal cell) as functions of x are shown in Fig. 2. It is found that w decreases almost linearly with increase in x from 0.2 to 0.4 and then increases with a further increase up to x¼0.6 and thereafter decreases, which is similar to the results of the Ba0.4Sr0.6  xCaxTiO3 system reported by Zheng et al. [11]. ˚ in 12-fold coordination is The ionic radius of Ca2 þ (1.340 A) 2þ ˚ ˚ [14], which smaller than that of Ba (1.610 A) and Sr2 þ (1.440 A) led to the lattice shrinkage in the range of 0.2 rx r0.4. The anomalous expansion at 0.4 oxr0.6 is due to a small amount of ˚ which is confirmed by Zheng Ca2 þ substitution for Ti4 þ (0.605 A), et al. [11]. When x 40.6, besides those that have been substituted for Ti4 þ , the remaining Ca2 þ substitute for Sr2 þ . The variation of

1.0035 c/a

0.3975

Temperature (K)

were kept at 550 1C for 6 h to remove the solvent and the binder. The pellets were sintered at 1350 1C for 4 h in air. All samples had densities higher than 90% of theoretical limits. Phase compositions of the ceramics were investigated by means of X-ray diffraction (XRD, Bruker D8 Advanced, Germany) using CuKa radiation. Temperature dependences of permittivity (e) and loss tangent were measured using an HP 4284A precision LCR meter (Agilent, Palo Alto, CA) at 10 kHz in the temperature range of 125–450 K. Polarization versus electric field (P–E) hysteresis loops were measured in a silicon oil bath by applying an electric field of triangular waveform at 10 Hz by means of a ferroelectric testing system (Radiant Precision Premier II Technology). MTI 2100 photonic sensor was used for strain measurement. Samples were poled in a silicon oil bath at room temperature with a DC electric field of 20 kV/cm for about 30 min, which was used for d33 and kp measurements. d33 was measured using a quasistatic d33 meter (ZJ-6A, Institute of Acoustics, Chinese Academy of Sciences, Beijing, China). The accuracy of d33 measurement is 70.1 pC/N. kp was determined by means of a resonance– antiresonance method using a precision impedance analyzer (HP4294A, Agilent, Palo Alto, CA).

Lattice constant (nm)

J. Wang et al. / Journal of Physics and Chemistry of Solids 73 (2012) 957–960

Permittivity

958

0 0.03 0.02 0.01 0.00 150

200

250 300 Temperature (K)

350

400

Fig. 3. Temperature dependences of permittivity and loss tangent of the BST–xBCT ceramics measured at 10 kHz. Inset shows TC and TO–T of the BST–xBCT ceramics as functions of x.

c/a with x shows a ‘‘V’’ shaped curve in the range of 0.2 rx r0.6, indicating that the tetragonality initially becomes weak with increasing x and then becomes strong. Thereafter, the tetragonality remains almost unchanged. Temperature dependences of permittivity and loss tangent for all samples measured at 10 kHz are displayed in Fig. 3. The dielectric peaks of both cubic–tetragonal phase transition and tetragonal–orthorhombic phase transition for the BST–xBCT ceramics are suppressed and broadened with increasing x. CaTiO3 is known as the typical depressor in BaTiO3 ceramics, which led to significant suppression of dielectric permittivity and loss [15]. Li et al. [16] reported that addition of Ca did not strongly affect the TC of the (Ba1  xCax)(Ti0.96Sn0.04)O3 system, but pushed TO–T toward a lower temperature. Interestingly, in present system, TC increases with increasing x, whereas TO–T decreases. TC and TO–T of the BST–xBCT ceramics as functions of x are shown in the inset of Fig. 3. The relationship between TC and x for 0.2 rxr0.9 is linear, which can be described by TC (K)¼66.0x þ306.4. Amazedly, the deviation of TC from the linear behavior observed in the Ba0.4Sr0.6  xCaxTiO3 system [11] is not found in the present system. The increase of TC with the increase of Ca concentration in the Ba0.4Sr0.6  xCaxTiO3 system is in distinct contrast to the decrease of TC with respect to Sr concentration in the Ba1  xSrxTiO3 system, which cannot be understood within the framework of the A-site cation size effect [11]. According to a slight variation of TC in Ba1  xCaxTiO3 series [5,8] and a linear decrease of TC in

J. Wang et al. / Journal of Physics and Chemistry of Solids 73 (2012) 957–960

ð1Þ

4

10 /ε

4

10 8 6 4 2

330 360 390 420 Temperature (K)

345K

400K

4

12 10 8 6 4 2 300

339K

390K

6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0

x =0.2@γ =1.41,δ =6.15 x =0.3@γ =1.46,δ =7.53 x =0.4@γ =1.49,δ =8.76 x =0.5@γ =1.46,δ =8.83 x =0.6@γ =1.42,δ =8.71 x =0.7@γ =1.49,δ =9.74 x =0.9@γ =1.53,δ =12.62 Fitting

320 330 340 350 360 370 380 390 400 410 Temperature (K)

300

330 360 390 420 Temperature (K)

10 8 6 4 2

353K

300

330 360 390 420 Temperature (K)

10 /ε

300

376K

10 8 6 327K 4 370K 2 0 300 330 360 390 420 Temperature (K)

Fig. 5. Temperature dependences of 1/e for the BST–xBCT ceramics measured at 10 Hz. The symbols indicate the experimental data and the solid lines represent fitting to the power relation (3).

403K

330 360 390 420 Temperature (K)

365K 406K

330 360 390 420 Temperature (K)

Fig. 4. 1/e versus temperature at 10 kHz for the BST–xBCT ceramics: (a) x¼ 0.2, (b) x¼ 0.3, (c) x¼ 0.4, (d) x ¼ 0.5, (e) x¼ 0.6, (f) x ¼0.7 and (g) x¼ 0.9. The symbols indicate the experimental data and the solid lines represent fitting to the Curie– Weiss law (1).

12 2

10 8 6 4 2

333K

4

10 8 6 4 2 0 300

10 /ε

10 8 6 319K 368K 4 2 0 300 330 360 390 420 Temperature (K)

10 /ε

4

10 /ε

4

10 /ε

4

10 /ε

where C is the Curie constant. Fig. 4 shows the variation of 1/e at 10 kHz of all samples with temperature and the fitting of the experimental data to the Curie–Weiss law. For all samples, a deviation from the Curie–Weiss law is clearly visible below a certain temperature TB, the Burns temperature [17]. The deviation from the Curie–Weiss law is commonly observed in relaxor ferroelectrics [18], due to the existence of polar clusters which appear below TB [19]. There are mainly two methods, which have been widely used in research on relaxor ferroelectrics, to characterize the temperature

ð3Þ

where Tmax is the temperature corresponding to emax, d characterizes the degree of diffuseness of dielectric peak, and g characterizes the degree of relaxation. It is found that the data taken from the high-temperature slopes of the permittivity peaks of all samples can be collapsed onto a single scaling line derived from Eq. (2) (not shown here). We fitted our data to Eq. (3) as indicated by the solid line in Fig. 5. The obtained g value is in the range of 1.41–1.53, indicating that the BST–xBCT ceramics is a ferroelectric material with diffuse phase transition. The d value slightly increases besides at x¼0.6 with increasing x. The increased d value is ascribed to compositional heterogeneity and A-site cation size-mismatch induced local disorder. At 0.4oxr0.6, Ca2 þ substitution for Ti4 þ causes compressive stress and then impedes the lattice distortion, therefore decreasing d. On the other hand, Ca2 þ substitution for Ti4 þ also causes charge disorder, which results in the increase of d. The low d value at x¼0.6 is possibly due to the larger contribution of compressive stress than that of charge disorder.

4

ðT 4 TCÞ,

whereas the second one is a power relation [22] described by   1 1 ðTT max Þg ¼ 1þ , 2 e emax 2d

Polarization (μC/cm )

1=e ¼ ðTT C Þ=C,

dependence of permittivity above TC [20]. The first one is a quadratic relation [21] described by " # 1 1 ðTT max Þ2 ¼ 1þ , ð2Þ 2 e emax 2d

10 /ε

Ba1  xSrxTiO3 series [7,9,10] with increasing x, one may conclude that the variation of TC for the BST–xBCT ceramics is mainly dominated by the concentration of Sr. The linear increase of TC of the BST–xBCT ceramics with x is attributed to the increase of Ba/ Sr ratio. It should be mentioned that TO–T reduces linearly with increase in x from 0.2 to 0.4, which can be defined approximately by the relation TO–T (K)¼  103.7xþ239.3, while deviating from linear behavior at x ¼0.5. The deviation of TO–T from linear behavior is possibly related to the lattice expansion (see Fig. 2). this is because larger Ca2 þ ions substitution for Ti4 þ ions causes the expansion of unit cell which exerts compressive stress on the nearest-neighbor unit cells. The compressive stress impedes the ferroelectric distortion of the octahedral of the neighbor unit cell, thus decreasing TO–T. It is well known that the variation of 1/e with temperature of a normal ferroelectric, above TC, follows the Curie–Weiss law described by

959

8 4

8 7 6 5 4 3 2 0.2

4.8 4.4 4.0 3.6 0.4

0.6 x

0.8

0 x =0.2 x =0.3 x =0.4 x =0.5 x =0.6 x =0.7 x =0.9

-4 -8 -12 -30

-20

-10 0 10 Electric field (kV/cm)

20

30

Fig. 6. P–E hysteresis loops of the BST–xBCT ceramics measured at 10 Hz and room temperature. Inset shows Ec and Pr as functions of x.

960

J. Wang et al. / Journal of Physics and Chemistry of Solids 73 (2012) 957–960

0.10 0.08 0.06

Strain (%)

0.04

x=0.2 x=0.3 x=0.4 x=0.5

0.02 0.00

Fig. 8 shows the variations of d33 and kp of the BST–xBCT ceramics with x. d33 decreases from 106 pC/N to 38 pC/N with increase in x from 0.2 to 0.6 and then slightly increases with a further increase in x. However, as x increases, kp initially decreases and then increases up to the maximum (17.0%) at x¼0.6 and thereafter decreases. According to the phenomenological theory, d33 can be described by the equation d33 ¼2e0eQ11Ps [24], where e is the vacuum permittivity, Q11 the electrostrictive coefficient, and Ps the spontaneous polarization. As x increases, both e and Ps decrease (Figs. 3 and 6). This shows that d33 of this present material system is strongly dependent on e and Ps.

0.06

4. Conclusions 0.04

x=0.6 x=0.7 x=0.9

0.02

0.00 -30

-20

-10 0 10 Electric field (kV/cm)

20

30

Fig. 7. S–E curves of the BST–xBCT ceramics measured at 30 kV/cm and room temperature.

Only tetragonal phase is detected for all the compositions, indicating that BCT has fully incorporated into BST lattice forming a complete solid solution. w decreases almost linearly with increase in x from 0.2 to 0.4, while shows an anomalous expansion at 0.4 oxo0.6 and thereafter shows a linear decrease. TC increases linearly with increase in x from 0.2 to 0.9. TO–T decreases linearly with increase in x from 0.2 to 0.4, while deviates from linear behavior at x¼0.5, which can be ascribed to a small amount of Ca2 þ substitution for Ti4 þ . The obtained g value is in the range of 1.41–1.53, indicating a ferroelectric material with diffuse phase transition. Strain and d33 decrease with increasing x, while kp initially decreases and then increases up to the maximum (17.0%) at x¼0.6 and thereafter decreases.

110 17 100 16

Acknowledgments

90 15 80 14 70 13 60

This research was supported by the Zhejiang Provincial Science Foundation (no. Y6110475), the Research Fund for higher Education of Zhejiang Province (no. 01101158) and the undergraduate innovative project of Zhejiang Province (no. 2010R409016).

12 50 11

References

40 0.1

0.2

0.3

0.4

0.5

x

0.6

0.7

0.8

0.9

10 1.0

Fig. 8. d33 and kp of the BST–xBCT ceramics as functions of x.

P–E ferroelectric hysteresis loops of the BST–xBCT ceramics are shown in Fig. 6. All the samples exhibit a typical ferroelectric P–E. As x increases, the P–E loop becomes slanted. Remnant polarization Pr and coercive field Ec as functions of x are given in the inset of Fig. 6. Pr decreases slightly from 4.03 mC/cm2 to 3.48 mC/cm2 with increase in x from 0.2 to 0.5 and then rapidly increases up to the maximum (4.87 mC/cm2) at x¼0.6 and thereafter decreases with a further increase in x, whereas Ec initially increases slowly and then rapidly increases. The increase of Ec means that the domain mobility gets smaller for the sample with higher BCT level. Electric field-induced strain behavior has been studied and strain–electric field (S–E) butterfly curves for various x are exhibited in Fig. 7. At E r30 kV/cm, hysteresis in S–E behavior is present for all samples, accompanying the residual strain value of 0.04–0.10%. The hysteretic strain is associated with domain switching in the samples [23]. The strain decreases from 0.10% to 0.07% with increase in x from 0.2 to 0.3, then remains almost unchanged at 0.3 rxr0.6 and thereafter decreases to 0.04% with further a increase to x¼0.9. The decreased strain is attributed to the decreased domain mobility, as mentioned above.

[1] W.R.C.B. Jaffe, H. Jaffe, Piezoelectric ceramics, Academic Press, London, 1971. [2] C.K. Campbell, J.D. Van Wyk, M.F.K. Holm, J.J.R. Prinsloo, J.J. Schoeman, IEEE Trans. Component Packag. Technol. 16 (1993) 418–423. [3] J.F. Scott, Integr. Ferroelectr. 20 (1998) 15–23. [4] Z. Zhao, X. Wang, K. Choi, C. Lugo, A. Hunt, IEEE Trans. Microwave Theory Technol. 55 (2007) 430–437. [5] T. Mitsui, W.B. Westphal, Phys. Rev. 124 (1961) 1354. [6] V.V. Lemanov, E.P. Smirnova, P.P. Syrnikov, E.A. Tarakanov, Phys. Rev. B 54 (1996) 3151. [7] J. Zhang, J. Zhai, X. Chou, J. Shao, X. Lu, X. Yao, Acta Mater. 57 (2009) 4491–4499. [8] X. Wang, H. Yamada, C.N. Xu, Appl. Phys. Lett. 86 (2005) 022905. [9] V.B. Shirokov, V.I. Torgashev, A.A. Bakirov, Phys. Rev. B 73 (2006) 104116. [10] H. Wu, W.Z. Shen, Phys. Rev. B 73 (2006) 094115. [11] R.K. Zheng, J. Wang, X.G. Tang, Y. Wang, H.L. Chan, C.L. Choy, X.G. Li, J. Appl. Phys. 98 (2005) 084108. [12] J. Yang, P. Liu, X. Bian, H. Jing, Y. Wang, Y. Zhang, Y. Wu, W. Song, Mater. Sci. Eng. B 176 (2011) 260–265. [13] S. Su, R. Zuo, S. Lu, Z. Xu, X. Wang, L. Li, Curr. Appl. Phys. 11 (2011) S120–S123. [14] R.D. Shannon, Acta Crystallogr. A: Found. Crystallogr. 32 (1976) 751–767. [15] X. Wang, L. Zhang, H. Liu, J. Zhai, X. Yao, Mater. Chem. Phys. 112 (2008) 675–678. [16] W. Li, Z. Xu, R. Chu, P. Fu, G. Zang, J. Am. Ceram. Soc. 94 (2011) 4131–4133. [17] G. Burns, F.H. Dacol, Phys. Rev. B 28 (1983) 2527. [18] D. Viehland, S.J. Jang, L.E. Cross, M. Wuttig, Phys. Rev. B 46 (1992) 8003–8006. [19] H. Yu, Z.G. Ye, J. Appl. Phys. 103 (2008) 034114. [20] Z.Y. Cheng, R.S. Katiyar, X. Yao, A. Guo, Phys. Rev. B 55 (1997) 8165–8174. [21] G. Smolensky, J. Phys. Soc. Jpn. 28 (1970) 26. [22] H. Martirena, J. Burfoot, Ferroelectrics 7 (1974) 151–152. [23] L.X. Zhang, W. Chen, X. Ren, Appl. Phys. Lett. 85 (2004) 5658–5660. [24] D. Damjanovic, Rep. Prog. Phys. 61 (1998) 1267–1324.