Generalizability of Graph Neural Network Force Fields for Predicting Solid-State Properties

TL;DR

This paper evaluates the ability of a graph neural network-based machine-learned force field to predict solid-state properties of materials beyond its training data, demonstrating promising generalization for complex crystalline systems.

Contribution

It introduces benchmark tests and workflows for assessing the generalizability of MLFFs in solid-state simulations, including unseen defect configurations.

Findings

MLFF accurately predicts phonon density of states for FCC crystals.

MLFF successfully estimates vacancy migration rates and energy barriers.

Proposes strategies to improve MLFF generalizability in solid-state modeling.

Abstract

Machine-learned force fields (MLFFs) promise to offer a computationally efficient alternative to ab initio simulations for complex molecular systems. However, ensuring their generalizability beyond training data is crucial for their wide application in studying solid materials. This work investigates the ability of a graph neural network (GNN)-based MLFF, trained on Lennard-Jones Argon, to describe solid-state phenomena not explicitly included during training. We assess the MLFF's performance in predicting phonon density of states (PDOS) for a perfect face-centered cubic (FCC) crystal structure at both zero and finite temperatures. Additionally, we evaluate vacancy migration rates and energy barriers in an imperfect crystal using direct molecular dynamics (MD) simulations and the string method. Notably, vacancy configurations were absent from the training data. Our results demonstrate the MLFF's capability to capture essential solid-state properties with good agreement to reference data, even for unseen configurations. We further discuss data engineering strategies to enhance the generalizability of MLFFs. The proposed set of benchmark tests and workflow for evaluating MLFF performance in describing perfect and imperfect crystals pave the way for reliable application of MLFFs in studying complex solid-state materials.