Inhomogeneous Degradation in Lithium-Ion Batteries: The Effect of Thermal Gradients

Abstract

Understanding lithium ion battery degradation is a consequence of multiple, tightly coupled and non-linear degradation mechanisms. The interplay between these degradation mechanisms, and how the cell is used, gives rise to path dependent behaviour, and hence data driven approaches to model behaviour and predict lifetime rapidly become prohibitively expensive. Therefore, a physics based approach to model degradation is desirable.In addition, all of the degradation mechanisms are a function of temperature, which adds another variable to the path dependency. To make matters worse, all cells experience internal thermal gradients during operation, which means the path dependency for different regions within a battery will be different. This leads to inhomogeneous degradation. After a few cycles, the degradation pathway of each region will then be affected not just by the thermal gradients and local current density but also by the state-of-health of that region, leading to complicated emergent behaviour.Solving a physics based model with multiple degradation mechanisms, in a detailed 3-dimensional thermally coupled discretised model is also prohibitively expensive. Therefore, in this paper we present a thermally coupled distributed equivalent circuit model, with degradation functions built in, so each node can degrade at different rates, as a function of the operating conditions at that node. We have reproduced experimental data previously published, showing the effect of thermal gradients on degradation for two different cooling strategies for a tab and surface cooled pouch cell. For the first time reproducing the positive feedback mechanism responsible for accelerated degradation for surface cooling.The results of this work show that thermal gradients cannot and should not be ignored, and degradation models that do not take them into account have a significant risk of overfitting and leading to incorrect conclusions.