Multiphysics-informed thermal runaway model for estimating the pressure evolution induced by the gas formation in a lithium-ion battery

Abstract

This study proposes a multiphysics-informed thermal runaway (TR) model for a lithium-ion battery to replicate TR phenomena by considering the evolution of the SOC and SOH, thermodynamics, chemical reactions, pressure evolution, vent, and ejection phenomena. This versatile model contributes to estimate both temperature and pressure evolutions, and thereby understand the underlying multiphysics of TR better by deploying three key features. First, the proposed model accounts for TR dependency on the SOC and SOH of a cell by controlling kinetic parameters of positive and negative active materials. Second, simple yet accurate governing equations of vent and ejection phenomena are coupled to complex and nonlinear thermodynamics, elaborating the TR characteristics predicted by the proposed model. Third, the proposed model directly couples the chemical reactions of the five main components to predict gas formation (CO, C2H4, CO2, H2, and CH4) and evolution because the pressure evolution is initiated by different reaction ratios of these components. Quantitative experimental validation with both nickel-manganese-cobalt and lithium–iron-phosphate cells shows that the maximum error for prediction of temperature and pressure evolution is 2.2 and 14.5 % respectively, confirming the accuracy and robustness of the proposed model. This study also includes in-depth discussions for the applications of the proposed model, including the early detection of TR and gas monitoring. These insightful discussions not only reveal the effectiveness of the proposed model but also provide guidelines for optimal battery thermal management strategies.