Effect of donor doping in B sites on the electrocaloric effect of BaTi_1−xNb_xO₃ ceramics

Abstract

This paper demonstrates the effect of donor doping in B sites on the electrocaloric effect (ECE) in BaTi_1−xNb_xO₃ (x = 0–0.01) ferroelectric ceramics. The Nb substitution for Ti does not affect the formation of the perovskite structure, while it obviously inhibits the grain growth in sintering process. The donor doping of Nb controls the ferroelectric phase transition more efficiently than equivalent substitutions in either the A or B sites. It lowers the phase transition temperature about one order of magnitude faster than an equivalent substitution, so the ECE ΔT_max shifts accordingly. Nb doping also diffuses the phase transition, so that the ECE ΔT_max reduces and the peak becomes much wider. When compared with BaTiO₃, the ΔT_max of BaTi_0.994Nb_0.006O₃ drops one third, while the full width at half maximum increases two fold, indicating a better refrigeration capacity.

---

1. Introduction

High performance and small size are the trends of electronic equipment and these features have improved greatly in recent years. However, a large amount of high power devices working in a small space will induce a local high temperature, which degrades the performance of the electronic devices and may even lead to complete failure. Recently, ferroelectric cooling based on the electrocaloric effect (ECE) has been regarded as the best solution, by the solid-state refrigeration of miniaturized electronic products, and has been attracting a lot of attention due to the advantages of easy miniaturization, high efficiency and low cost. Since a giant ECE of ΔT_max = 12 K was obtained in a PbZr_0.95Ti_0.05O₃ thin film in 2006,¹ the amount of research on the ECE in various ferroelectric ceramics boomed.^2–21

The ECE is a basic feature of ferroelectric materials and refers to a reversible change in entropy and temperature caused by the electric field-induced variation of polarization states. Similar to the magnetocaloric effect, the ECE is only remarkable in the local vicinity of a phase transition, so it is thought to be closely related to the features of the ferroelectric phase transition.⁴ BaTiO₃ with a typical first order phase transition (FOPT) has a sharp ECE peak near the Curie temperature with giant ECE strength |ΔT/ΔE|.^5–8 A lot of research has been devoted to material design by various substitutions to widen the temperature window. For example, the substitution of Ba²⁺ with Sr²⁺ increasingly diffuses the phase transition to reduce ΔT_max and widen the peak.^9–11 The substitution of Ti⁴⁺ with Zr⁴⁺ also diffuses the phase transition and results in a high ΔT in a broad temperature range.^12,13 The substitution of Ti⁴⁺ with Sn⁴⁺ leads to a high ΔT at the morphotropic phase boundary.^14,15 In addition, the co-substitution of Ba²⁺ with Ca²⁺ and Ti⁴⁺ with Zr⁴⁺ was also studied.^16–21 Up to now, all reports focused on the effect of various equivalent dopings but inequivalent doping has not been researched for the ECE, although inequivalent dopings were reported to efficiently improve the dielectric and piezoelectric properties.^22,23 The donor doping of Nb⁵⁺ for Ti⁴⁺ in the B sites, a typical soft doping, has a notable effect on modifying the ferroelectric phase transition. This paper reports its effects on the phase composition, microstructure, dielectric, ferroelectric and electrocaloric effects of the BaTiO₃ ceramics, and a better refrigeration capacity is obtained.

2. Experimental procedure

2.1 Sample preparation

The BaTi_1−xNb_xO₃ (x = 0–0.01) ceramics were prepared by the conventional solid-state reaction method. Analytical reagent grade BaCO₃, TiO₂ and Nb₂O₅ were used as raw materials. After the mixed powders were calcined at 1000 °C, they were grinded by a planetary ball mill. The resultant powders were dry-pressed in a stainless-steel die under a pressure of 3 MPa and the pressed pellets were sintered at 1350 °C for 4 hours in air.

2.2 Characterization

The phase composition of calcined powders were characterized by X-ray diffraction (XRD) using Cu Kα radiation (λ = 0.15418 nm) with a scanning rate of 2° min⁻¹. The microstructure of the sintered samples was observed by scanning electron microscope (SEM, JSM-6510A). The densities of the sintered samples were measured by the Archimedes’ method. The heat flow curve was measured using a differential scanning calorimeter (DSC, TA Instruments Q2000) with a heating rate of 10 °C min⁻¹. The permittivity was measured between 50 and 150 °C by an HP4192 impedance analyzer with a temperature chamber. The ferroelectric hysteresis measurements were carried out at 10 Hz using a TF2000 analyzer in the temperature range of 25–150 °C.

3. Results

3.1 Phase composition

Fig. 1 shows the XRD pattern of the BaTi_1−xNb_xO₃ (x = 0.001, 0.004, 0.01) powders calcined at 1000 °C. The results are carefully indexed with the standard XRD pattern (PDF 05-0626). All of the samples show a well-defined perovskite phase and there is no impurity phase in each sample. The lattice parameters of the calcined powders are about a = 4.000–4.004 Å, and this change in the samples with different Nb doping is as tiny as it is comparable to instrumental error. This is because the doping amount is very small and the radius difference between Nb⁵⁺ (0.064 nm) and Ti⁴⁺ (0.061 nm) is very small. In addition, the XRD spectra do not show an obvious tetragonal distortion, i.e. c/a ≈ 1, due to the small particle size of the powder specimens.

Fig. 1 The XRD spectra of BaTi_1−xNb_xO₃ (x = 0.001–0.01) calcined at 1000 °C.

3.2 Microstructure

After all of the samples were sintered at 1350 °C for 4 hours, they exhibited a similar dense microstructure. Fig. 2(a)–(d) show the SEM photos of the microstructure with the x = 0.00–0.008 samples as examples. When compared with the pure BaTiO₃ ceramics (Fig. 2(a)), the Nb doped samples exhibit an obviously smaller grain size, less than 1 μm, which further decreases with an increasing amount of Nb (Fig. 2(b)–(d)). It is because the donor doping of Nb accumulates at the grain boundaries, which hinders the motion of the grain boundaries during the sintering process. The densities of all of the sintered samples are larger than 5.92 g cm⁻³, i.e. 98% of the theoretical density, which agrees well with the SEM observations.

Fig. 2 The microstructures of the samples (x = 0.0, 0.004, 0.006 & 0.008) sintered at 1350 °C.

3.3 Thermal analysis

The thermal characteristics of the tetragonal–cubic (T–C) phase transition are shown in the heat flow curves in Fig. 3. With the rise of the amount of Nb, the endothermic peak gradually moves to a lower temperature, indicating a linear shift of the phase transition with a fitting equation of T₁ = −3140x + 122 (°C), as shown in the upper inset of Fig. 3. The slope is much higher than that in previous reports for equivalent substitutions, such as Sr, Zr and Sn. It implies that the donor doping of Nb is more efficient for the phase transition shift. In addition, the endothermic peak becomes flatter and the latent heat reduces with the rise of the amount of Nb due to the diffusion of the first order phase transition.

Fig. 3 Heat flow curves of the BaTi_1−xNb_xO₃ ceramics in a heating process. The inset shows the variation of the Curie temperature.

3.4 Temperature dependence of permittivity

Fig. 4 shows the temperature dependence of permittivity for the samples with different amounts of Nb. The phase transition shifts to a lower temperature with the rise of the amount of Nb and the peak value of permittivity gradually drops, which agrees with the DSC results.

Fig. 4 Temperature dependence of permittivity of the BaTi_1−xNb_xO₃ ceramics. The inset shows the composition dependence of α.

The reduction of the permittivity maximum at the Curie temperature is related to the diffused ferroelectric phase transition. Based on Smolensky’s composition fluctuation theory and the Curie–Weiss law, the diffused phase transition was characterized by the diffusion exponent α using the equation of

As shown in the inset of Fig. 4, α rises with an increasing Nb amount, i.e. the phase transition is diffused gradually.

3.5 Ferroelectric and electrocaloric properties

The sintered BaTi_1−xNb_xO₃ samples exhibit good ferroelectric hysteresis loops with a high polarization value and the ferroelectric hysteresis loops of all of the samples have a similar variation trend with a rise in the temperature. The P–E loop shrinks gradually and the polarization value decreases, which exhibits a sudden drop at the Curie temperature. Fig. 5(a)–(c) show the typical ferroelectric hysteresis loops of the x = 0.002, 0.006 & 0.01 samples at different temperatures, respectively.

Fig. 5 Ferroelectric hysteresis loops of the BaTi_1−xNb_xO₃ ceramics at different temperatures. (a) x = 0.002; (b) x = 0.006; (c) x = 0.01.

Based on the Maxwell relation of (∂P/∂T)_E = (∂S/∂E)_T, the ECE ΔT and ΔS are calculated as following

Fig. 6 shows the ΔT–T curves of the x = 0.006 sample as an example. There is a remarkable ECE peak in the vicinity of the phase transition. The ECE ΔT_max always occurs above the Curie temperature, when compared with the results of the DSC and permittivity. When the electric field increases, the ECE ΔT is enhanced and the ΔT_max shifts to a higher temperature.

Fig. 6 Temperature dependence of the ECE ΔT of BaTi_1−xNb_xO₃ (x = 0.006) under different applied electric fields.

Fig. 7 shows the ΔT_max as a function of the Nb amount. With the rise of the amount of Nb, the ECE ΔT_max moves to lower temperature, which is associated with the electric field-induced shift of the ferroelectric phase transition. At the same time, the ECE ΔT_max drops gradually and the peak becomes wider, which is determined by the diffusion of the phase transition. This phenomenon is similar to previous reports on Sr, Zr or Sn doped BaTiO₃.^9–14 Under an E = 20 kV cm⁻¹ field, ΔT_max = 0.98 K (ΔS = 1.24 J kg⁻¹ K⁻¹) for x = 0, ΔT_max = 0.89 K (ΔS = 1.14 J kg⁻¹ K⁻¹) for x = 0.002 and ΔT_max = 0.66 K (ΔS = 0.86 J kg⁻¹ K⁻¹) for x = 0.006, while the full width at half maximum (FWHM) of the ECE peaks is 8 °C, 16 °C and 24 °C, respectively. The variations of ΔT_max and the FWHM are also shown in Fig. 7.

Fig. 7 The ECE ΔT_max and the FWHM as a function of Nb amount (x = 0–0.01).

4. Discussions

Donor doping, such as Nb⁵⁺, is soft doping for ferroelectric materials and can improve the dielectric and piezoelectric properties.^22,23 It increases the dielectric permittivity and reduces the coercive force because of easy polarization under external fields. The change of polarization under the coupled effect of the external electric field and temperature plays a key role for the ECE, including ferroelectric phase transition and domain switching.

Our work shows that the donor doping of Nb is an efficient way to modify the ferroelectric phase transition, including shifting the transition point and diffusing the transition, so does the ECE. When a Nb⁵⁺ ion with a bit larger radius (0.064 nm) than a Ti⁴⁺ ion (0.061 nm) substitutes it in the B site, the Nb⁵⁺ ion locates itself at the center site of the oxygen octahedrons which is the same as Ti⁴⁺ but the Nb–O bond is weakened. On the other hand, the valence state of Nb⁵⁺ is higher than that of Ti⁴⁺, which further weakens the crystalline field. In addition, some cation vacancies form at the same time which can ease domain switching. Because the doping amount is too small (x ≤ 0.01) to induce an obvious lattice distortion, which is confirmed by the XRD results, the different valence state affects the crystalline field in a dominant way. That is different from the effect of equivalent dopings, such as Sr²⁺ in the A site and Zr⁴⁺ in B the site, where lattice distortion is the key. Our results show that the phase transition changes more rapidly with the donor doping of Nb than the equivalent substitutions. For example, the shift of the Curie temperature is about one order of magnitude faster than that of Sr substitution.⁹ Each 1% increase in Nb doping lowers the T_c about 31 °C, while each 1% increase in Sr doping only lowers the T_c 3 °C. In addition, the composition fluctuation and the local inhomogeneous internal stress destroy the phase transition consistency, so the phase transition obviously becomes more diffused with an increasing amount of Nb.

5. Conclusions

This paper investigated the ECE of BaTi_1−xNb_xO₃ ceramics prepared by the solid state reaction method. The sintered Nb doped ceramics have fine-grained microstructures. Nb doping modifies the ferroelectric phase transition more efficiently. With the rise of the amount of Nb, the phase transition shifts to a lower temperature more rapidly than that for equivalent substitutions, and the transition is diffused at the same time. As a result, the ECE ΔT_max reduces and shifts to lower temperature with increasing amount of Nb, and the ECE peak becomes much wider. BaTi_0.994Nb_0.006O₃ has ΔT_max = 0.66 K (@E = 20 kV cm⁻¹) and the FWHM = 24 °C, where ΔT_max drops to one third of that in BaTiO₃ and the FWHM increases two fold, i.e. a better refrigeration capacity. This work implies that inequivalent substitution is a more efficient method for adjusting the ECE in ferroelectric ceramics, which can greatly enrich the material design for ferroelectric refrigeration.

Acknowledgements

This work was supported by grants from the National Science Foundation of China (51172020 and 51372018), the National Program for Support of Top-Notch Young Professionals, the Program for New Century Excellent Talents in University (NCET-12-0780), the Beijing Higher Education Young Elite Teacher Project (YETP0414), and the Program for Changjiang Scholars and Innovative Research Team in University (IRT1207).

References

1. S. Mischenko, Q. Zhang, J. F. Scott, R. W. Whatmore and N. D. Mathur, Science, 2006, 311, 1270–1271.
2. B. Neese, B. Chu, S. G. Lu, Y. Wang, E. Furman and Q. M. Zhang, Science, 2008, 321, 821–823.
3. Y. Bai, G. P. Zheng and S. Q. Shi, Appl. Phys. Lett., 2010, 96, 192902.
4. Y. Bai, K. Ding, G. P. Zheng, S. Q. Shi and L. Qiao, Phys. Status Solidi A, 2012, 209, 941–944.
5. Y. Bai, X. Han, X. C. Zheng and L. J. Qiao, Sci. Rep., 2013, 3, 02895.
6. X. Moya, E. Stern-Taulats, S. Crossley, D. González-Alonso, S. Kar-Narayan, A. Planes, L. Mañosa and N. D. Mathur, Adv. Mater., 2013, 25, 1360–1365.
7. Y. Bai, K. Ding, G. P. Zheng, S. Q. Shi, J. L. Cao and L. J. Qiao, AIP Adv., 2012, 2, 022162.
8. H. H. Wu, J. Zhu and T. Y. Zhang, RSC Adv., 2015, 5, 37476–37484.
9. Y. Bai, X. Han, K. Ding and L. J. Qiao, Appl. Phys. Lett., 2013, 103, 162902.
10. X. Q. Liu, T. T. Chen, Y. J. Wu and X. M. Chen, J. Am. Ceram. Soc., 2013, 96, 1021–1023.
11. X. Q. Liu, T. T. Chen, M. S. Fu, Y. J. Wu and X. M. Chen, Ceram. Int., 2014, 40, 11269–11276.
12. X. S. Qian, H. J. Ye, Y. T. Zhang, H. Gu, X. Li, C. A. Randall and Q. M. Zhang, Adv. Funct. Mater., 2014, 24, 1300–1305.
13. H. J. Ye, X. S. Qian, D. Y. Jeong, S. J. Zhang, Y. Zhou, W. Z. Shao, L. Zhen and Q. M. Zhang, Appl. Phys. Lett., 2014, 105, 152908.
14. Z. Luo, D. Zhang, Y. Liu, D. Zhou, Y. Yao, C. Liu, B. Dkhil, X. Ren and X. Lou, Appl. Phys. Lett., 2014, 105, 102904.
15. T. Shi, L. Xie, L. Gu and J. Zhu, Sci. Rep., 2015, 5, 08606.
16. Y. Bai, X. Han and L. J. Qiao, Appl. Phys. Lett., 2013, 102, 252904.
17. M. Sanlialp, V. V. Shvartsman, M. Acosta, B. Dkhil and D. C. Lupascu, Appl. Phys. Lett., 2015, 106, 062901.
18. G. Singh, I. Bhaumik, S. Ganesamoorthy, R. Bhatt, A. K. Karnal, V. S. Tiwari and P. K. Gupta, Appl. Phys. Lett., 2013, 102, 082902.
19. G. Singh, V. S. Tiwari and P. K. Gupta, Appl. Phys. Lett., 2013, 103, 202903.
20. B. Asbani, J. L. Dellis, A. Lahmar, M. Courty, M. Amjoud, Y. Gagou, K. Djellab, D. Mezzane, Z. Kutnjak and M. El Marssi, Appl. Phys. Lett., 2015, 106, 042902.
21. J. F. Wang, T. Q. Yang, S. C. Chen, G. Li, Q. F. Zhang and X. Yao, J. Alloys Compd., 2013, 550, 561.
22. D. Guo, K. Cai, L. Li and Z. Gui, J. Appl. Phys., 2009, 106, 054104.
23. D. Guo, L. Li, C. Nan, J. Xia and Z. Gui, J. Eur. Ceram. Soc., 2003, 23, 2177.

---