Machine-Learned Bond-Order Potential for Exploring the Configuration Space of Carbon

TL;DR

This paper introduces a physics-informed machine-learning interatomic potential for carbon that efficiently explores its configuration space, enabling accurate modeling across diverse structures and aiding in the discovery of new carbon materials.

Contribution

A novel bond-order potential based on machine learning that is highly transferable and requires few parameters, improving exploration of carbon's configuration space.

Findings

Accurately models phonon dispersion and phase diagrams.

Effectively searches for stable cluster structures.

Maps enthalpy-volume relationships of local minima.

Abstract

Construction of transferable machine-learning interatomic potentials with a minimal number of parameters is important for their general applicability. Here, we present a machine-learning interatomic potential with the functional form of the bond-order potential for comprehensive exploration over the configuration space of carbon. The physics-based design of this potential enables robust and accurate description over a wide range of the potential energy surface with a small number of parameters. We demonstrate the versatility of this potential through validations across various tasks, including phonon dispersion calculations, global structure searches for clusters, phase diagram calculations, and enthalpy-volume mappings of local minima structures. We expect that this potential can contribute to the discovery of novel carbon materials.