Thermal conductivity of h-BN monolayers using machine learning interatomic potential

TL;DR

This paper demonstrates that a machine learning interatomic potential can efficiently and accurately predict the thermal conductivity of h-BN monolayers, significantly reducing computational costs for thermal management material design.

Contribution

The study introduces a Gaussian approximation potential-based MLIP for h-BN monolayers, providing a new approach to evaluate thermal conductivity with fewer configurations and higher reliability.

Findings

MLIP achieves accurate thermal conductivity with 30% configurations

High-order force constants improve MLIP reliability

MLIP results agree well with frozen phonon calculations

Abstract

Thermal management materials are of critical importance for engineering miniaturized electronic devices, where theoretical design of such materials demands the evaluation of thermal conductivities which are numerically expensive. In this work, we applied the recently developed machine learning interatomic potential (MLIP) to evaluate the thermal conductivity of hexagonal boron nitride monolayers. The MLIP is obtained using the Gaussian approximation potential (GAP) method, and the resulting lattice dynamical properties and thermal conductivity are compared with those obtained from explicit frozen phonon calculations. It is observed that accurate thermal conductivity can be obtained based on MLIP constructed with about 30% representative configurations, and the high-order force constants provide a more reliable benchmark on the quality of MLIP than the harmonic approximation.