Predicting Material Properties Using a 3D Graph Neural Network with Invariant Local Descriptors

TL;DR

This paper introduces an adaptive 3D graph neural network that effectively captures atomic spatial relationships to improve the prediction of material properties like gas adsorption and ion conductivity.

Contribution

The paper presents a novel convolution mechanism in GCNNs that models all atomic interactions in 3D space, enhancing property prediction accuracy in materials science.

Findings

Outperforms existing models on MOF gas adsorption data

Achieves better results on solid-state ion conductivity prediction

Successfully captures 3D geometric information of atomic structures

Abstract

Accurate prediction of physical properties is critical for discovering and designing novel materials. Machine learning technologies have attracted significant attention in the materials science community for their potential for large-scale screening. Graph Convolution Neural Network (GCNN) is one of the most successful machine learning methods because of its flexibility and effectiveness in describing 3D structural data. Most existing GCNN models focus on the topological structure but overly simplify the three-dimensional geometric structure. However, in materials science, the 3D-spatial distribution of atoms is crucial for determining the atomic states and interatomic forces. This paper proposes an adaptive GCNN with a novel convolution mechanism that simultaneously models atomic interactions among all neighbor atoms in three-dimensional space. We apply the proposed model to two distinctly challenging problems on predicting material properties. The first is Henry's constant for gas adsorption in Metal-Organic Frameworks (MOFs), which is notoriously difficult because of its high sensitivity to atomic configurations. The second is the ion conductivity in solid-state crystal materials, which is difficult because of few labeled data available for training. The new model outperforms existing graph-based models on both data sets, suggesting that the critical three-dimensional geometric information is indeed captured.