Advanced Physico-Chemical Modeling of Lithium Plating in Lithium-Ion Batteries

Abstract

The transport sector is responsible for about 14 % of the global greenhouse gas emissions. To make the transition to sustainable transportation it is important to find sustainable alternatives that can compete with today’s petrol and diesel vehicles. The most promising alternative is electricity stored in lithium-ion batteries (LIBs) in hybrid electric vehicles (HEVs) and electric vehicles (EVs). Two challenges that need to be solved to fully compete with petrol and diesel cars are the charging time of LIBs and the service life of EVs and HEVs, where LIBs typically are the limitation due to ageing phenomena. One ageing phenomenon that is problematic in both of these challenges is lithium metal deposition (or lithium plating). To predict lithium plating accurately complex physics-based models are needed. If accurate enough such models could be used to predict optimal charging conditions to enable for faster battery charging, at low temperatures without damaging the battery. In this project a physics-based model to predict lithium plating in LIBs and fast charging at low temperature has been developed in Simulink. The model has been validated by comparison with results from models found in literature with very good agreement. A sensitivity analysis of the parameters pointed out the most important parameters for the simulations to be parameters in the negative electrode as well as the diffusion coefficient and the ionic conductivity in the electrolyte. To make accurate simulations at low temperatures an Arrhenius dependence is introduced on relevant material parameters. Even though the results are promising the model validation is mostly done versus existing models. In order to use this model to improve charging routines for LIBs in HEVs and EVs it is necessary to collect data on lithium plating and determine material parameters experimentally. Keywords: