Simulating lattice thermal conductivity in semiconducting materials using high-dimensional neural network potential

Abstract

We demonstrate that a high-dimensional neural network potential (HDNNP) can predict the lattice thermal conductivity of semiconducting materials with an accuracy that is comparable to that of density functional theory (DFT) calculation. After a training procedure based on force, the root mean square error between the forces predicted by HDNNP and DFT is less than 40 meV Å-1. As typical examples, we present the results of Si and GaN bulk crystals. The deviation from the thermal conductivity calculated using DFT is within 1% at 200 to 500 K for Si and within 5.4% at 200 to 1000 K for GaN.