Predicting Lattice Phonon Vibrational Frequencies Using Deep Graph Neural Networks

TL;DR

This paper introduces a deep graph neural network model that accurately predicts lattice phonon vibrational frequencies from crystal structures, offering a computationally efficient alternative to DFT calculations for materials screening.

Contribution

The work presents a novel deep graph neural network approach that handles variable spectrum dimensions using zero padding, enabling accurate phonon frequency predictions from crystal structures.

Findings

Achieved R^2 scores of 0.554 and 0.724 on two datasets.

Demonstrated capability to predict phonon spectra beyond density of states.

Addresses variable spectrum dimension with zero padding scheme.

Abstract

Lattice vibration frequencies are related to many important materials properties such as thermal and electrical conductivity as well as superconductivity. However, computational calculation of vibration frequencies using density functional theory (DFT) methods is too computationally demanding for a large number of samples in materials screening. Here we propose a deep graph neural network-based algorithm for predicting crystal vibration frequencies from crystal structures with high accuracy. Our algorithm addresses the variable dimension of vibration frequency spectrum using the zero padding scheme. Benchmark studies on two data sets with 15,000 and 35,552 samples show that the aggregated $R^2$ scores of the prediction reaches 0.554 and 0.724 respectively. Our work demonstrates the capability of deep graph neural networks to learn to predict phonon spectrum properties of crystal structures in addition to phonon density of states (DOS) and electronic DOS in which the output dimension is constant.