Crystal structure prediction via combining graph network and Bayesian optimization

TL;DR

This paper introduces GN-BOSS, a novel method combining graph networks and Bayesian optimization to predict crystal structures efficiently without relying on density functional theory, demonstrated on binary compounds.

Contribution

The paper presents a new data-driven approach for crystal structure prediction that bypasses expensive DFT calculations by integrating graph networks with Bayesian optimization.

Findings

Successfully predicted structures of 24 binary compounds

Achieved average prediction time of ~30 minutes per compound

Operated with only one CPU core for each prediction

Abstract

We developed a density functional theory-free approach for crystal structure prediction via combing graph network (GN) and Bayesian optimization (BO). GN is adopted to establish the correlation model between crystal structure and formation enthalpies. BO is to accelerate searching crystal structure with optimal formation enthalpy. The approach of combining GN and BO for crystal Structure Searching (GN-BOSS), in principle, can predict crystal structure at given chemical compositions without additional constraints on cell shapes and lattice symmetries. The applicability and efficiency of GN-BOSS approach is then verified via solving the classical Ph-vV challenge. It can correctly predict the crystal structures of 24 binary compounds from scratch with averaged computational cost ~ 30 minutes each by only one CPU core. GN-BOSS approach may open a new avenue to data-driven crystal structural prediction without using the expensive DFT calculations.