Effect of an electric field on the electrocaloric response of ferroelectrics
Liu Hongbo
School of Materials Engineering, Shanghai University of Engineering Science, Shanghai 201620, China

 

† Corresponding author. E-mail: bohongliu@gmail.com

Project supported by the National Natural Science Foundation of China (Grant No. 11704242) and the Natural Science Foundation of Shanghai, China (Grant No. 17ZR1447200).

Abstract

The electrocaloric effect of the model ferroelectric BaTiO3 was investigated using phenomenological theory. The results indicate that the applied electric field strength is a key factor for the induced electrocaloric response and there are two distinguishing electrocaloric responses. When a moderate electric field is applied, the electrocaloric temperature variation is small but the electrocaloric strength is high. In contrast, the electrocaloric temperature variation is large but electrocaloric strength is low when a very high electric field is applied. These results are consistent with the experimental observations on BaTiO3 based bulk and thin film ferroelectric materials.

PACS: 77.70.+a
1. Introduction

The electrocaloric (EC) effect refers to the isothermal entropy change ΔS or adiabatic temperature change ΔT of a dielectric material when an electric field is applied or removed. It has attracted great attention after the findings of 12-K temperature change in PbZr0.95Ti0.05O3[1] and more than 12-K temperature variation in P(VDF-TrFE) (55 mol%/45 mol%).[2] It has been particularly used for applications in cooling.[36] Lead-based electrocaloric materials show a large EC effect. However, the toxicity of lead has limited their applications. Therefore, lead-free ferroelectrics have been considered as better candidates for EC refrigeration, such as Bi0.5Na0.5TiO3-based ceramics,[79] and BaTiO3-based materials.[10]

The EC ΔT is closely related to the applied electric field. The electrocaloric strength ΔTE is also evaluated in the studies of EC effect. For instance, the large EC effects in PbZr0.95Ti0.05O3 were induced with a field of 77.6 MV/m,[1] which is huge in comparison with the dielectric strengths of common inorganic ferroelectric materials.

To gain an insight into the effect of the applied electric field, we conducted a theoretical study of the EC effect of the model ferroelectric BaTiO3 (BTO). A variety of experimental results has been reported on the EC effect of this material. In single crystal, the reported ΔT varies from 0.9 K@1.2 MV/m,[11] ∼1.2 K@1.2 MV/m,[12] to 1.6 K@1 MV/m.[13] In multilayer films, the ΔT varies from 1.8 K@17.6 MV/m,[14] to 7.1 K@80 MV/m.[15] These results indicate that the electric field strength is crucial for the EC response.

2. Theory

Based on the Maxwell equation, the EC adiabatic temperature change ΔT can be evaluated by the following equation[16,17] in which ρ and Cp are density and specific heat capacity, respectively, here ρ = 6020 kg · m−3,[16] and Cp = 407 J · kg−1 · K−1.[11] For a monodomain BTO single crystal with both the polarization P and the applied electric field along [001] axis, the Landau–Devonshire potential including up to eighth-power term can be written as

where α1, α11, α111, and α1111 are the Landau coefficients, and g is the Gibbs free energy density without considering the contribution of polarization. The polarization in the (T, E) field can be obtained by finding the minimum of the potential. By combining Eq. (1) and Eq. (2) the EC ΔT can be calculated. In calculations, t he Landau coefficients were adopted from those published by Wang et al.[18] The potential proposed by Wang et al. gives the following characteristic temperatures: the phase transition temperature TC = 403 K and the critical point temperature TCP = 418 K at which the phase transition becomes of the second order under electric field.[12] The electric field at the critical point ECP is 1.5 MV/m.

3. Results and discussion

The EC ΔT of BTO at the different electric field strengths are plotted in Fig. 1. For the case of an applied electric field less than ECP, there exists a sharp peak of EC ΔT versus T; as shown in Fig. 1(a).

Fig. 1. (color online) Calculated EC ΔT of BTO at different electric field strengths by the phenomenological method.

This profile of ΔTT is consistent with the experimental observations.[1113] Both the intensity and the width of the peak increase slightly with the increase of the applied electric field strength. The starting temperature of the ΔT peak is TC, and the stopping temperature is TCP. The predicted ΔT at 1 MV/m is 1.2 K, and the EC strength ΔTE is 1.2 × 10−6 K · m · V−1. These values agree with the experimental observations in the crystals of BTO as listed in Table 1.

Table 1.

Data of EC properties of BTO and doped BTO in the literature.

.

In contrast, the EC ΔTT profiles become broad and flat at the electric field strengths approaching and higher than ECP; as shown in Fig. 1(b). These profiles are similar to the experimental results in the thin films of BTO,[15] PbZr0.95Ti0.05O3,[1] and P(VDF-TrFE) (55mol%/45 mol%).[2] At 100 MV/m, ΔT is 3 K and ΔTE is 0.03 × 10−6 K · m · V−1, which are comparable with those obtained with BTO multilayer film (see Table 1).

To disclose these two types of EC responses, we calculated PE loops at selected temperatures and P at selected electric field strengths. The typical PE loops are shown in Fig. 2(a). These temperatures satisfy the following relations: 400 K < TC, TC < 41 K < TCP, and TCP < 420 K. Thus, the BTO shows ferroelectric PE loop at T < TC. However, the coercive field EC decreases with the increasing temperature. It becomes zero at TC. The double loop presents at TC < T < TCP. Finally, linear PE relation is observed at TCP < T. Since a small electric field can induce a large variation of polarization at TC < T < TCP, a sharp EC ΔT peak with large ΔTE is observed; as shown in Fig. 2(a).

Fig. 2. (color online) The PE loops at selected temperatures (a) and the polarization as a function of temperature at selected electric field strengths (b).

The electric field effect on the evolution of polarization is depicted in Fig. 2(b) at selected electric field strengths. Note that the critical point ECP is 1.5 MV/m for BTO. The discontinuity of polarization versus temperature becomes faint after 1.2 MV/m. When the electric field strength is higher than 10 MV/m, the polarization is a weak function of temperature and the polarization variation decreases with the increasing electric field. Thus, a giant electric field can induce large ΔT but the ΔTE is small, and a flat ΔTT profile is observed; as shown in Fig. 2(b).

Finally, we suggest that the finding in BTO is universal for ferroelectrics with first-order phase transition. At an electric field approaching ECP, ferroelectric materials have small ΔT about 1 K but high EC strength, and a sharp EC ΔTT peak. Under a strong electric field, they produce giant ΔT but low EC strength, and a flat ΔTT profile is observed. Since ferroelectric ceramics in bulk have low dielectric strengths around ECP, a sharp ΔTT peak and tiny ΔT are normally observed. Because the dielectric strength is significantly enhanced in the thin films of ferroelectrics, a broad ΔTT profile and large ΔT are observed. Table 1 lists some recent experimental observations on EC effect of BTO-based materials, which agree quantitatively with the theoretical prediction.

4. Conclusion and perspectives

In conclusion, the present study indicates that there are two distinguishing EC responses for ferroelectrics with first-order phase transition, which are mainly decided by the applied electric field. When the applied electric field strength is comparable with ECP, there exists a sharp ΔTT peak with large ΔTE between [TC, TCP]. When the applied electric field strength is much higher than ECP, because of the vanishing of the first-order phase transition, the polarization is a weak function of temperature. The large EC ΔT can be induced by a large electric field but the EC ΔTE is quite small. A flat EC ΔTT profile is observed commonly. This result explains the experimental observations on BTO-based materials from bulks to thin films. It also inspires researchers to find alternative ways to achieve both large EC ΔT and ΔTE simultaneously.

Reference
[1] Mischenko A Zhang Q Scott J Whatmore R Mathur N 2006 Science 311 1270
[2] Neese B Chu B Lu S G Wang Y Furman E Zhang Q 2008 Science 321 821
[3] Jia Y Sungtaek Ju Y 2012 Appl. Phys. Lett. 100 242901
[4] Gu H Qian X Li X Craven B Zhu W Cheng A Yao S C Zhang Q M 2013 Appl. Phys. Lett. 102 122904
[5] Plaznik U Kitanovski A Rožič B Malič B Uršič H Drnovšek S Cilenšek J Vrabelj M Poredoš A Kutnjak Z 2015 Appl. Phys. Lett. 106 043903
[6] Sette D Asseman A Géard M Strozyk H Faye R Defay E 2016 APL Mater. 4 091101
[7] Weyland F Acosta M Koruza J Breckner P Rödel J Novak N 2016 Adv. Funct. Mater. 26 7326
[8] Li F Chen G Liu X Zhai J Shen B Li S Li P Yang K Zeng H Yan H 2017 Appl. Phys. Lett. 110 182904
[9] Li F Chen G Liu X Zhai J Shen B Zeng H Li S Li P Yang K Yan H 2017 J. Eur. Ceram. Soc. 37 4732
[10] Qian X S Ye H J Zhang Y T Gu H Li X Randall C Zhang Q 2014 Adv. Funct. Mater. 24 1300
[11] Moya X Stern-Taulats E Crossley S González-Alonso D Kar-Narayan S Planes A Mañosa L Mathur N D 2013 Adv. Mater. 25 1360
[12] Novak N Kutnjak Z Pirc R 2013 Europhys. Lett. 103 47001
[13] Bai Y Ding K Zheng G P Shi S Q Qiao L 2012 Phys. Status Solidi 209 941
[14] Bai Y Zheng G Shi S 2010 Appl. Phys. Lett. 96 192902
[15] Bai Y Zheng G P Ding K Qiao L Shi S Q Guo D 2011 J. Appl. Phys. 110 094103
[16] Liu H Yang X AIP Adv. 5 117134
[17] Liu H B 2017 Chin. Phys. 26 117701
[18] Wang Y L Tagantsev A K Damjanovic D Setter N Yarmarkin V K Sokolov A I Lukyanchuk I A 2007 J. Appl. Phys. 101 104115
[19] Jian X D Lu B Li D D Yao Y B Tao T Liang B Guo J H Zeng Y J Chen J L Lu S G 2018 ACS Appl. Mater. Inter. 10 4801
[20] Li M D Tang X G Zeng S M Liu Q X Jiang Y P Zhang T F Li W H 2018 ACS Sustain. Chem. Eng. 6 8920
[21] Qian J Hu P Liu C Jiang J Dan Z Ma J Lin Y Nan C W Shen Y 2018 Sci. Bull. 63 356