Fast and Accurate Prediction of Material Properties with Three-Body Tight-Binding Model for the Periodic Table

TL;DR

This paper introduces a universal, efficient tight-binding model with three-body interactions, trained on a large DFT database, capable of accurately predicting properties across a wide range of materials and bonding types.

Contribution

The authors develop a universal tight-binding model with three-body terms, trained on a large DFT database, enabling accurate predictions for diverse materials and bonds.

Findings

Model achieves high accuracy across metallic, covalent, and ionic materials.

Inclusion of three-body interactions improves prediction accuracy.

Iterative learning enhances model predictive power.

Abstract

Parameterized tight-binding models fit to first principles calculations can provide an efficient and accurate quantum mechanical method for predicting properties of molecules and solids. However, well-tested parameter sets are generally only available for a limited number of atom combinations, making routine use of this method difficult. Furthermore, most previous models consider only simple two-body interactions, which limits accuracy. To tackle these challenges, we develop a density functional theory database of nearly one million materials, which we use to fit a universal set of tight-binding parameters for 65 elements and their binary combinations. We include both two-body and three-body effective interaction terms in our model, plus self-consistent charge transfer, enabling our model to work for metallic, covalent, and ionic bonds with the same parameter set. To ensure predictive power, we adopt a learning framework where we repeatedly test the model on new low energy crystal structures and then add them to the fitting dataset, iterating until predictions improve. We distribute the materials database and tools developed in this work publicly.