Local crystallographic shear structures in a extended mixed dislocations of SrTiO3 unraveled by atomic-scale imaging using transmission electron microscopy and spectroscopy
Recently, extended mixed dislocations were observed at a /(100) low-angle tilt grain boundary of a SrTiO3 bicrystal because of a slight twist between the two crystal parts. The b = a/(100) mixed dislocations at the grain boundary dissociate into three dislocations with Burgers vector b of a/2, a, and a/2, respectively. A structure model has been proposed in particular for the dislocation cores of the two partials with b = a/2 based on the high-angle annular dark-field (HAADF) images acquired by scanning transmission electron microscopy (STEM). However, the details of the atomic structure and the chemical composition of the dislocation cores remain unexplored, especially for the b = a dislocation that is evidently disassociated into two b = a/2 partial dislocations. In this work, we study the further atomic details of the extended mixed dislocations, in particular the local chemistry, in a SrTiO3 bicrystal using STEM, electron energy loss spectroscopy (EELS), and energy dispersive X-ray (EDX) spectroscopy techniques. By these atomic-scale imaging techniques, we reveal a unique feature for the atomic structure of the b = a/(100) extended mixed dislocation, which we named as local crystallographic shear (LCS) structures. In addition, we identify a rock salt FCC-type TiOx (x = 0.66–1.24) phase at the locations of the extended mixed dislocations. In contrast to the insulating TiO2 phases, the TiOx phase is known to exhibit very low electrical resistivity of only several μΩ cm. In this regard, the extended mixed dislocations of SrTiO3 comprising the FCC TiOx phase may function as the conducting filament in resistive switching processes by completion and disruption of the TiOx phase along the dislocation cores through electrically stimulated redox reactions.
- This article is part of the themed collection: New memory paradigms: memristive phenomena and neuromorphic applications