Conductive Networks and Their Impact on Uncertainty, Degradation, and Failure of Lithium-Ion Battery Electrodes

Abstract

Manufacturing high-quality lithium-ion batteries requires a detailed understanding of the complex correlations between the electrode microstructure and the kinetic processes. Sophisticated microstructure design is of interest to enable optimal ionic and electrical transport to minimize uncertainties and capacity fade. One important feature of the microstructure is the carbon black-binder matrix (CBM) forming a network-like structure. The CBM is often solely evaluated by its electrical bulk conductivity of the electrode. In this work, a hybrid model approach is established and applied to gain mechanistic understanding of the influence of different electrical network structures on uncertainty, degradation, and failure of battery cells. Artificial network structures are generated based on degree distributions, that is, normal distributions or power-law distributions. Degradation is modeled by edge removal, representing the mechanical decomposition of the CBM during battery utilization. Simulation results show that the electrical bulk conductivity is not sufficient to rate the quality of the electrical network, and also other properties such as the connectivity and the number of transit paths need to be taken into account. It can be seen that normally distributed connectivity in general yields low uncertainties and is robust against edge removal. In contrast, power-law distributions, that is, scale-free-like networks, possess few critical edges that yield significant uncertainty. The presented model approach allows us to study network structures in-depth and to identify beneficial structural network properties.