Predicting Electrolyte Redox Potential Trends through Long-Range Electrostatics and Configuration Ensembles: Insights from MD and QM/MM Simulations

In the quest for advanced energy storage solutions, the stability and performance of electrolyte materials are of paramount importance. In particular, the redox potentials of electrolytes are of great interest, as these values dictate the electrochemical window that can be applied during battery cycling. Although redox trends can be estimated from computed frontier orbital energies and ionization potentials (IPs) and electron affinities (EAs), accurately representing long-range electrostatics is challenging. While solution-phase embedding schemes (e.g., continuum solvation, cluster-continuum, and polarizable/frozen-density embedding) are well established, their direct application and validation in electrolyte systems remain limited. In the current study, we adopt an integrated approach, combining classical molecular dynamics (MD) simulations and quantum mechanics-molecular mechanics (QM/MM) calculations to estimate the electrochemical stability of electrolytes. MD simulations of the electrolytes are first used to generate representative solution configuration ensembles, which are then subjected to QM/MM calculations to determine the highest occupied molecular orbital (HOMO), lowest unoccupied molecular orbital (LUMO), and the vertical IP and EA. To account for long-range electrostatics and configurational sampling, we consider increasingly larger solvation shells surrounding electrolyte molecules and perform ensemble averaging. Using this framework, we specifically study two widely used electrolytes for lithium- and sodium-ion batteries and investigate the effect of anion additives. The computed results are compared with experimental results from linear sweep voltammetry. We conclude that it is important to include several solvation shells to accurately compute the frontier orbitals of electrolyte molecules and to average the orbitals over configuration ensembles. The current approach for the first time computes the oxidative and reductive stability trends of electrolyte species while capturing the long-range effects and configuration ensembles of the environment. We also observe that, while vertical IP and EA values strongly correlate with HOMO and LUMO energies, structural relaxation of oxidized and reduced species weakens this correlation─particularly for organic carbonates upon reduction─highlighting the limitations of the simple HOMO-LUMO picture and the need to account for geometric relaxation in realistic redox predictions.