Piezoelectricity and structural defects from elastic measurements on unpoled ferroelectrics

Reference Speaker Authors
(Institution)
Abstract
402-004
Francesco Cordero Cordero, F.(Istituto dei Sistemi Complessi, Area della Ricerca di Roma-Tor Vergata); The elastic properties of ferroelectric (FE) materials are usually measured by piezoelectric excitation of poled samples, in order to obtain the complete set of materials constants necessary for the material characterization and for predicting the behavior of devices, but they can actually provide more valuable information. In the first part of the talk it will be shown that it is possible to quantitatively asses the piezoelectric response of ferroelectric materials from the temperature dependence of the elastic moduli and dielectric constant of unpoled samples [1]. Avoiding the poling process is attractive for several reasons: poling often causes severe stress of the samples, with possible cracking, and the level of poling may be incomplete, due to the high coercive field, microstructure, excessive electrical conductivity or thermally induced domain switching. The values obtained from unpoled samples are therefore closer to the intrinsic ones, even in conditions where proper poling is impossible, which is useful, for example, when testing new materials. In addition, it is also possible to probe the enhancement of the piezoelectric coefficients near morphotropic phase boundaries or other transitions between different FE phases [2]. Examples are shown of ceramic samples excited on their flexural resonances, where only an orientational average of the piezoelectric coefficients is obtained, but it is shown that the measured piezoelectric softening in the tetragonal phase of ceramic BaTiO3 agrees within few percent with the orientational average calculated with the literature single crystal piezoelectric and dielectric constants. Resonant Ultrasound Spectroscopy experiments on unpoled ceramics can in principle provide the complete set of elastic and piezoelectric constants, which has already been done with poled samples [3]. Next will be considered the structural defects introduced by repeated transitions between FE and antiferroelectric (AFE) states and aging with coexisting FE and AFE domains. These defects would affect the performance of devices based on the transitions between FE and AFE states, such as high displacement actuators and high energy storage capacitors. The Young’s modulus of Zr-rich (PbZr1-xTixO3) PZT is measured during several thermal cycles through the FE/AFE transition and during aging at room temperature at compositions x ~ 0.05, where the FE phase may coexist for long time together with the AFE ground state, due to the slow kinetics of the transition. Under these conditions the samples undergo softenings as large as four times the initial compliance, which however are completely recovered by fast annealing at 700-850 K [4]. Since these annealings are made in high vacuum, filling of O vacancies is excluded, and also healing of possible microcracks generated by the mismatch between AFE and FE domains is unlikely at such low temperatures. It is therefore discussed how the native O vacancies formed during sintering with PbO loss might form planar aggregation at the AFE/FE walls, which may be assimilated to flat inclusions with much reduced elastic moduli. According to the available models of the effect of pores on the effective elastic moduli, a small volume concentration of planar defects would be very effective in softening the material. [1] F. Cordero, F. Craciun, F. Trequattrini and C. Galassi, Phys. Rev. B 93, 174111 (2016) [2] F. Cordero, H. T. Langhammer, T. Müller, V. Buscaglia and P. Nanni, Phys. Rev. B 93, 064111 (2016) [3] L. Tang and W. Cao, Appl. Phys. Lett. 106, 052902 (2015) [4] F. Cordero, F. Craciun, F. Trequattrini, P. Galizia, and C. Galassi, J. Appl. Phys. 120, 064104 (2016).
<< Back