Sampling polymorphs of ionic solids using random superlattices

TL;DR

This paper introduces a novel method using random superlattices to explore and predict polymorphs in ionic solids, effectively identifying metastable structures and their likelihood of realization.

Contribution

The study presents a new RSL-based approach for structure prediction in ionic solids, combining random superlattice sampling with DFT relaxations to assess polymorph realizability.

Findings

Successfully identified known metastable polymorphs in MgO, ZnO, and SnO2.

Probability of structure occurrence correlates with experimental stability.

Method provides a quantitative measure of polymorph realizability.

Abstract

Polymorphism offers rich and virtually unexplored space for discovering novel functional materials. To harness this potential approaches capable of both exploring the space of polymorphs and assessing their realizability are needed. One such approach devised for partially ionic solids is presented. The structure prediction part is carried out by performing local DFT relaxations on a large set of random supperlattices (RSLs) with atoms distributed randomly over different planes in a way that favors cation-anion coordination. Applying the RSL sampling on MgO, ZnO and SnO2 reveals that the resulting probability of occurrence of a given structure offers a measure of its realizability explaining fully the experimentally observed, metastable polymorphs in these three systems.