Electrocaloric effect in ferroelectric ceramics with point defects

The electrocaloric effect has drawn much attention due to its potential application in cooling devices. A negative electrocaloric effect is predicted to be induced in defect-doped ferroelectrics by computational results [A. Grunebohm and T. Nishimatsu, Phys. Rev. B 93, 134101 (2016) and Ma et al., Phys. Rev. B 94, 094113 (2016)], but it need to be confirmed by experimental results. In this work, we prepared a 1 mol. % Mn-doped Pb(Zr0.2,Ti0.8)O3 ceramics (Pb((Zr0.2,Ti0.8)0.99,Mn0.01)O3), and the electrocaloric effect of the defect-containing ferroelectric ceramics has been investigated by both direct and indirect methods. The indirect method shows a similar negative electrocaloric effect signal as the computational results predicted, while the direct method gives a positive electrocaloric effect. The absence of the negative electrocaloric effect obtained by the direct method may originate from: (a) the unavailability and the improper prediction of the Maxwell relation, (b) an improper assumption of fixed defects in the computational models, and (c) the offset of heat loss due to the application of a large electric field. In addition, we find a giant positive electrocaloric effect of 0.55 K at room temperature in the aged ceramics where no phase transition takes place. We attribute this abnormal electrocaloric effect to the restoration force of the defect dipoles. Our results not only provide insights into the origin of the negative electrocaloric effect, but also offer opportunities for the design of electrocaloric materials.