Learning Thermoelectric Transport from Crystal Structures via Multiscale Graph Neural Network

TL;DR

This paper introduces a multiscale graph neural network model that accurately predicts electronic transport properties of thermoelectric crystals, aiding materials discovery and providing interpretability insights.

Contribution

A novel multiscale GNN architecture tailored for crystal structures that achieves state-of-the-art transport coefficient predictions and offers physical interpretability.

Findings

Achieves state-of-the-art performance on benchmark datasets.

Successfully identifies compounds with exceptional transport properties.

Provides interpretability analyses revealing physical patterns in transport behavior.

Abstract

Graph neural networks (GNNs) are designed to extract latent patterns from graph-structured data, making them particularly well suited for crystal representation learning. Here, we propose a GNN model tailored for estimating electronic transport coefficients in inorganic thermoelectric crystals. The model encodes crystal structures and physicochemical properties in a multiscale manner, encompassing global, atomic, bond, and angular levels. It achieves state-of-the-art performance on benchmark datasets with remarkable extrapolative capability. By combining the proposed GNN with \textit{ab initio} calculations, we successfully identify compounds exhibiting outstanding electronic transport properties and further perform interpretability analyses from both global and atomic perspectives, tracing the origins of their distinct transport behaviors. Interestingly, the decision process of the model naturally reveals underlying physical patterns, offering new insights into computer-assisted materials design.