Data-driven learning of total and local energies in elemental boron

TL;DR

This paper develops a machine learning approach using Gaussian approximation potentials and random structure searching to accurately model total and local energies in boron allotropes, aiding materials discovery.

Contribution

It introduces a systematic method combining machine learning and structure searching to generate interatomic potentials for boron, including local energy insights.

Findings

Accurate total energies for boron allotropes achieved

Atom-resolved local energies provided, revealing structural insights

Method enables automated potential generation for materials discovery

Abstract

The allotropes of boron continue to challenge structural elucidation and solid-state theory. Here we use machine learning combined with random structure searching (RSS) algorithms to systematically construct an interatomic potential for boron. Starting from ensembles of randomized atomic configurations, we use alternating single-point quantum-mechanical energy and force computations, Gaussian approximation potential (GAP) fitting, and GAP-driven RSS to iteratively generate a representation of the element's potential-energy surface. Beyond the total energies of the very different boron allotropes, our model readily provides atom-resolved, local energies and thus deepened insight into the frustrated $\beta$-rhombohedral boron structure. Our results open the door for the efficient and automated generation of GAPs and other machine-learning-based interatomic potentials, and suggest their usefulness as a tool for materials discovery.