Development of Multi-Physical Simulation Models for Quasi-Static Load Analysis to Predict the Behavior of Li-Ion and Na-Ion Batteries Under Different Mechanical Abuse Loadings

Abstract

There are lot of studies on understanding the short-circuit phenomena at cell component level and how to address the thermal management at module and pack level to avoid overheating. However, it is also very important, in terms of safety, to evaluate the response and failure of cells during abuse tests as this can promote the thermal runaway event. Safety of metal ion batteries under mechanical loadings is currently one of the most challenging and urgent issues face in the Electric Vehicle (EV) industry. In the battery manufacturing community, the property of strength and resistance of batteries to external loading has never been a design consideration. However, under local mechanical loading, batteries are prone to developing a internal short circuit (ISC), which may lead to the generation of smoke, fire, and possible explosion.This work is focused on the metal ion batteries safety studies (lithium and sodium) including the investigation of predictive models to determine the critical parameters that would lead to potential failure and provide critical insights to understand the mechanical and ISC behaviors of cells under mechanical abuse. Having a numerical model that correctly represents the cells and its response under different abuse tests could allow identifying main issues and helping on the cell design and chemistries to be used. To develop these models using LS-Dyna, it has been necessary to carry out studies of mechanical deformation on cell components and the complete cells. These studies have offered a better knowledge of the deformation of the inner components of the battery, being useful to identify the mechanism that initiate short circuits under mechanical misuse conditions. Building numerical models for batteries requires experimental work that provides not only the data for mechanical behavior of individual components (anode, cathode, separator, etc.), but also validation data for simulations of internal short circuit induced by mechanical abuse. Figure 1