Translational Symmetry Breaking at Charged Domain Walls in a Layered Perovskite Ferroelectric

Le résumé fourni par la source

Ferroelectrics are known as important functional materials that play various roles in electronics, various sensors, and energy harvesting applications. The functional properties depend largely on the polarization reversal caused by the movement of the boundary between domains with different polarization directions within the solid. Therefore, it is important to clarify how domain walls exist and how they move microscopically in ferroelectrics [1]. In particular, charged domain walls of ferroelectrics, in which ferroelectric polarizations face each other, have the property that the electrical conductivity changes significantly at the domain wall, even though the width is only a few nanometers. Although such unique charged domain walls have been found in some ferroelectrics, their structures are energetically unstable and thus it is an important issue to elucidate the stabilization mechanism. In the layered perovskite ferroelectric oxide Ca3-xSrxTi2O7 (x = 0.54), many charged domain walls have been observed in the crystal [2], but their microscopic structure has not been clarified. In this study, we visualized the charged domain structure in Ca3-xSrxTi2O7 (x = 0.54) on an atomic scale using scanning transmission electron microscopy (STEM) and clarified its structure [3, 4, 5]. Single crystals of Ca3-xSrxTi2O7 (x = 0.54) were grown via a floating zone method. The specimen was thinned by a focused ion beam. STEM observation was performed using a TEM instrument (JEM-ARM200F) equipped with a spherical-aberration corrector and dual silicon drift elemental detectors. The acceleration voltage was 200 kV. An electron-probe size, a convergence semi-angle, and a current of the incident electron beam were 1.2 Å, 22 mrad, and 60 pA, respectively. Fig. 1(a) shows a HAADF-STEM (High-angle annular dark-field STEM) image of a charged domain wall at [1–10] incidence. Noticeably, the crystalline domains on both sides of the charged domain wall are translating. The domain wall structure is shown in Fig. 1(b). The translational distance of the crystal domain corresponds to 0.1889 times the c-axis length, and such a boundary is called an out-of-phase boundary. In addition, the atoms are brighter at the interface than in the interior of the crystal, suggesting elemental segregation at the interface. Fig. 2(a) shows a dark-field image of the observed region. In this region, linear charged domain walls are formed, indicating the existence of domain walls on a macroscopic scale. The results of elemental analysis [Fig. 2(b)-(f)] show that the Sr and Ca atomic rows face each other across the charged domain wall, which confirms the structure shown in Fig. 1(b). The distribution of Ti also indicates that the crystal domain is translationally shifted. Furthermore, corresponding to the strong intensity in the HAADF-STEM image, Sr elements are segregated at the interface as indicated by the white circles. In summary, we observed the charged domain structure by a combination of TEM dark-field method and high-resolution HAADF-STEM. The charged domain wall is energetically unstable due to the divergence of the electric field caused by the facing electric polarizations. As a stabilization mechanism, an out-of-phase boundary is formed in the charged domain wall. In such a structure, the electric polarization of each atom is partially parallel as indicated by the dotted lines [Fig. 1(b)]. Therefore, such a structure is considered to make the charged domain wall more stable. Structure of a charged domain wall and its schematic. (a) HAADF-STEM image. (b) Crystal structure. Ps is the direction of macroscopic electric polarization. Arrows indicate the direction of displacement of each atom. Elemental analysis of charged domain wall. (a) Dark-field image by 113 reflection. The yellow box indicates the area observed by STEM. (b)-(f). EDS-STEM images at various absorption edge energies.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.