Thermal Runaway of Lithium-Ion Batteries: Mechanism, Modeling and Aging Effects

Abstract

With the lage-scale application of electric vehicles, safety problems due to the high energy density lithium-ion batteries, such as thermal runaway(TR), happen occasionally and result in fire and explosion, threatening the human lives and properties. Therefore, experimental and modeling effort should be paid to reveal battery TR mechanism, and model-based battery safety design and management algorithms should be developed to improve the safety of battery system. Recent years, researchers in the Battery Safety Laboratory of Tsinghua University have conducted researches on TR issues of lithium-ion batteries, including battery TR mechanism and modeling, TR propagation within battery module, and internal short circuit (ISC) detection algorithm. In this poster, we summarize our recent progress in investigating the battery TR mechanisms and the effects of aging on battery TR features. We hope these findings could benefit our understanding on battery TR problems and provide guidance for battery safety management. Accelerating rate calorimetry (ARC) and differential scanning calorimetry (DSC) are applied to investigate the TR of lithium-on batteries. Characteristic temperatures, i.e. onset temperature of self-heating T 1, TR temperature T 2, and maximum temperature T 3, during the TR process are identified to quantitatively evaluate TR performance of lithium-ion batteries. A novel method named “Time Sequence Map” that tracks the exothermic reactions and heat sources during TR process is then proposed to interpret the TR mechanism. TR of lithium-ion batteries is found to be induced by several possible critical reactions: 1) massive ISC; 2) the drastic oxidation-reduction reaction between the cathode and anode; 3) the reaction between electrolyte and the deposited lithium at anode. TR model based on kinetics analysis of the exothermic reactions is established to predict battery TR behaviors. The TR model shows excellent predictions of adiabatic TR test results and the oven tests results of a 24 Ah lithium-ion battery, presenting its practicability in determining the battery safety performance from kinetics analysis of cell components, without producing test batches of batteries. The effects of aging on battery TR performance have also been investigated in our laboratory. The evolution of battery TR features {T 1, T 2, T 3} under four different aging conditions, i.e. high temperature cycling, high temperature storage, high C-rates cycling and low temperature cycling, is characterized through a series of TR tests. The correlations between aging mechanisms and the changes of battery TR features are then revealed. Lithium plating under low temperature charging and fast charging conditions is observed to be the key factor of the full-life cycle safety of lithium-ion battery, as batteries with lithium metal plated at anode exhibit a much lower onset temperature (less than 50oC) of exothermic reactions and larger heat generation between 50oC and 100oC compared to fresh batteries. Therefore, to ensure safe and reliable full-life cycle operation of lithium-ion batteries, lithium plating should be effectively prevented or detected through safety management algorithms, such as non-destructive fast charging algorithms and lithium plating detection methods.