Assessing battery kinetics with machine learning

Abstract

The ionic diffusion characteristics of electrode materials critically influence the performance of batteries. Over the last decade, disordered rocksalt materials have emerged as promising next-gen battery cathodes. The higher degree of disorder in these materials results in increased computational complexity when assessing ionic diffusion profiles. Recently in <i>Electrochimica Acta</i>, Chang, Jorgensøn and co-authors reported a machine-learning-accelerated method to rapidly evaluate local ionic diffusion barriers in electrode materials with high accuracy.