Molecular Dynamics Simulations of Solid Electrolytes with NequIP Equivariant Machine Learning Models

Optimizing ion transport kinetics in solid-state energy storage systems is critical for performance. These devices contain interfaces, such as internal grain boundaries, which appear unavoidably from synthesis/processing, and electrolyte/cathode interfaces. Owing to the experimental difficulty in probing the complex chemical reactions and structural rearrangement that happens during operation, the details of how these microstructures impact ion transport remain elusive. A computational model that predicts ionic transport across these interfaces would be invaluable to the design of these solid-state energy storage systems. 

 

In this work, we examine the applicability of machine-learned interatomic potentials to studying ionic diffusion dynamics in bulk and interfaces. We make use of a state of the art equivariant graph neural network (NequIP) model to learn the interatomic interactions for the garnet (Li7La3Zr2O12) LLZO solid electrolyte with internal grain boundaries and the interface between LLZO and LiCoO2 (LCO) cathode. The training data is composed of energies and forces derived from ab-initio molecular dynamics (AIMD).