Training machine-learning potentials for crystal structure prediction using disordered structures

TL;DR

This paper introduces a method to train neural network potentials for crystal structure prediction using disordered structures, enabling efficient exploration of multinary compounds without prior structural information.

Contribution

The authors propose a novel approach to generate training data from liquid and amorphous phases, improving machine-learning potential accuracy for crystal prediction without needing initial structural data.

Findings

NNPs accurately rank low-energy structures compared to DFT

Evolutionary search with NNPs finds metastable phases more efficiently

Strong correlation between NNP and DFT energies across tested materials

Abstract

Prediction of the stable crystal structure for multinary (ternary or higher) compounds with unexplored compositions demands fast and accurate evaluation of free energies in exploring the vast configurational space. The machine-learning potential such as the neural network potential (NNP) is poised to meet this requirement but a dearth of information on the crystal structure poses a challenge in choosing training sets. Herein we propose constructing the training set from densityfunctional-theory (DFT) based dynamical trajectories of liquid and quenched amorphous phases, which does not require any preceding information on material structures except for the chemical composition. To demonstrate suitability of the trained NNP in the crystal structure prediction, we compare NNP and DFT energies for Ba2AgSi3, Mg2SiO4, LiAlCl4, and InTe2O5F over experimental phases as well as low-energy crystal structures that are generated theoretically. For every material, we find strong correlations between DFT and NNP energies, ensuring that the NNPs can properly rank energies among low-energy crystalline structures. We also find that the evolutionary search using the NNPs can identify low-energy metastable phases more efficiently than the DFTbased approach. By proposing a way to developing reliable machine-learning potentials for the crystal structure prediction, this work will pave the way to identifying unexplored multinary phases efficiently.