A Numerical Framework for Concentrated-Solution Theory

Abstract

Effective and stable electrolytes will be crucial to the implementation of next-generation battery chemistries. In particular, the engineering of novel electrolytes that suppress dendrites and facilitate stable SEI formation, whilst maintaining high conductivity and other desirable transport properties, will be necessary to enable lithium-metal anodes. Beyond this, electrolyte models that account for multiple dissolved components would be useful to analyze metal-oxygen batteries and lithium-sulfur batteries, and could shed light on the numerous transport processes potentially associated with the known side reactions that degrade today's lithium-ion cells.The Onsager--Stefan--Maxwell (OSM) equations from irreversible thermodynamics, which underpin concentrated solution theory, provide a natural framework for multicomponent-electrolyte multiphysics. OSM flux laws are force-explicit, however, which has led to difficulties in numerical implementation, and in turn posed a barrier for electrolyte designers to apply the transport equations in their most natural form. Leaving the implementation issues aside, the OSM formalism can be useful because it more readily includes additional physics such as elasticity and pressure, which could be important for developing accurate models of intercalation materials and solid electrolytes.This presentation will focus on numerical methods applicable to complex geometries. We will develop a general methodology to compute solutions of OSM transport problems, with a particular focus on how robust, rapidly converging results can be achieved when simulating materials that span multiple spatial dimensions. This methodology can be applied in both ideal and non-ideal settings, and will be illustrated with a variety of three-dimensional simulations. Particular emphasis is placed on devising a numerical algorithm which preserves and exploits the thermodynamic structure of the problem. The role of physical effects such as pressure and momentum, which tend to be neglected in most continuum simulations, will be highlighted. Insight into electrolyte design from these simulations will be discussed, and comparisons will be made to experimental data.