Graph Neural Networks Based Deep Learning for Predicting Structural and Electronic Properties

TL;DR

This paper develops a Graph Neural Network-based deep learning model to accurately predict multiple structural and electronic properties of materials from crystal structure data, enabling rapid materials screening.

Contribution

It introduces a GNN approach trained on the Materials Project database for simultaneous prediction of various material properties, demonstrating high accuracy.

Findings

Achieved high R^2 scores for multiple properties (e.g., 0.96 for density)

Predicted properties enable efficient materials screening

Validated GNN effectiveness on large crystal dataset

Abstract

This study presents a deep learning approach to predicting structural and electronic properties of materials using Graph Neural Networks (GNNs). Leveraging data from the Materials Project database, we construct graph representations of crystal structures and employ GNNs to predict multiple properties simultaneously. All crystal structures are from the Materials Project database, with a total of 158,874 structures used. Our model achieves high predictive accuracy across various properties, as indicated by \( R^2 \) values: 0.96 for density, 0.97 for formation energy, 0.54 for energy above hull, 0.47 for structural stability (is\_S), 0.76 for band gap, 0.86 for valence band maximum, 0.78 for conduction band minimum, and 0.82 for Fermi energy. These results demonstrate the potential of GNNs in materials science, offering a powerful tool for rapid screening and discovery of materials with desired properties.