Perovskites are just part of ABX3 materials class, which are considered promising materials for renewable energy and many other applications. In order to perform applicable high-throughput computations for this class of materials, it is fundamentally important to identify the most stable structural phase, which is not necessarily perovskite class. Also, the method of structural optimization has to be adequately efficient and reliable to be employed in screening of vast materials space. In this work, we demonstrate an approach to identify highly stable structures of ABX3 materials applied to CaTiO3 and CsGeCl3 using density functional theory (DFT) incorporated with the Standard Solid-State Pseudopotentials (SSSP) precision library and random sampling. First, supercells of each compound with 40 atoms are constructed in the ideal simple-cubic structure based on their Shannon's ionic radii. After that, we generated 10 supercells in the similar crystal symmetry. Then, these supercells are randomly distorted and are used as initial speculations to identify the stable structure of each compound. It is found that the highly stable structure of CaTiO3 is perovskite in Pnma (#62) spacegroup while it is non-perovskite in P1 (#1) spacegroup for CsGeCl3.
|Original language||English (US)|
|Title of host publication||2021 6th International Conference on Renewable Energy: Generation and Applications (ICREGA)|
|Number of pages||4|
|State||Published - 2021|