Atomic-level theoretical and experimental study of lattice dislocations in perovskites
Final Report Abstract
In the first part of this project, we focused our efforts on the understanding of the mechanical behavior of cubic strontium titanate (STO). The dislocations were observed by high resolution transmission electron microscopy (HRTEM), which gave valuable quantitative information about their Burgers vector and basic insights into their core structures. The atomistic simulations enabled detailed studies of dislocation core structures and their behavior under applied loads. Despite apparent preliminary discrepancies, a very good agreement was found between HRTEM experimental observations and atomistic simulations. The latter allowed us to study in more detail the atomic structure of dislocation cores and to model and understand the atomic-scale mechanisms of dislocation motion. The results of atomistic simulations as well as HRTEM experiments were used to formulate and parameterize phenomenological models that were able to describe correctly the macroscopic plastic behavior of STO over a broad temperature range. In the second part of this project, we extended the study to a non-cubic ferroelectric potassium niobate (KNO) in order to examine whether the macroscopic as well as microscopic mechanical behavior of perovskite oxides exhibits common, general features. In addition, these studies were expected to shed light on the role of dislocations in ferroelectricity, namely whether dislocations act as sources for nucleation of ferroelectric domains and/or as pinning sites for domain walls. Both theoretical and experimental studies revealed similarities as well as differences between the two investigated perovskites. Some of the differences are likely related to the more complex crystal structure of KNO, but some, such as the broad core and high mobility of the edge dislocation, may be related to different chemical nature and elastic properties of KNO. Unfortunately, due to the complexity of the problem, more detailed investigations could not be completed within the funding period. However, both Dr. Alison Mark (MPI for Intelligent Systems, Stuttgart), who carried out most of the experimental work on KNO, and Dr. Pierre Hirel (currently University of Lille, France), who performed most of the atomistic simulations, continue to work on this topic in their current positions. We attempted several large-scale atomistic simulations of interactions between dislocations and domain walls in KNO, but they turned out to be too complicated to be carried out in a reliable manner. Instead, we therefore modelled tetragonal ferroelectric barium titanate (BaTiO3, BTO) whose polarization along the <100> directions significantly simplifies the construction of simulation blocks containing the domain walls and dislocations. These simulations have revealed a complex interplay between the two classes of defects including ferroelectric switching and nucleation of domain walls at dislocation cores.
Publications
- Acta Mater. 58 (2010) 6072
P. Hirel, M. Mrovec, C. Elsässer
- Scripta Mater. 62 (2010) 270
M. Castillo-Rodriguez, W. Sigle
- Scripta Mater. 64 (2011) 241
M. Castillo-Rodriguez, W. Sigle
- Acta Mater. 60 (2012) 329
P. Hirel, M. Mrovec, C. Elsässer
