Data-driven Design of Electrolyte Additives for High-Performance 5 V LiNi0.5Mn1.5O4 Cathodes

Abstract

LiNi0.5Mn1.5O4 (LNMO) is a high-capacity spinel-structured material with an average lithiation/de-lithiation potential at ca. 4.6–4.7 V, far exceeding the stability limits of electrolytes. An efficient way to enable LNMO in lithium-ion batteries is to reformulate an electrolyte composition that stabilizes both graphitic (Gr) anode with solid-electrolyte-interphase (SEI) and LNMO with cathode-electrolyte-interphase (CEI). In this study, we selected and tested a diverse collection of 28 single and dual additives for the LNMO||Gr system. Subsequently, we trained machine learning (ML) models using this dataset and employed these models to identify 6 optimal binary compositions out of 125, based on their predicted final area-specific-impedance, impedance-rise, and final specific-capacity. The additives generated through this ML approach demonstrated superior performance compared to those in the in the initial dataset. This finding not only underscores the efficacy of ML in identifying new materials in a highly complicated application space, but also showcases an accelerated material discovery workflow that directly integrates data-driven methods with battery testing experiments.