Understanding Deviations between Spatially Resolved and Homogenized Cathode Models of Lithium‐Ion Batteries

Abstract

Porous electrode models are essential for inexpensively predicting the performance and lifetime of lithium‐ion batteries. Physics‐based models range from microscopic 3D models, which spatially resolve the microstructural characteristics of all phases in porous electrodes, to reduced and computationally effective models, which do not resolve the microstructure. The homogenized Newman model, also known as the pseudo‐2D (P2D) model, is well established and widely used. However, the necessary simplification shows its weaknesses, especially for high charge and discharge rates, and these lead to significant differences in comparison with the microscopic 3D model. Herein, the validity of the homogenized Newman model is investigated with respect to variations of the microstructural characteristics of a porous cathode. The effects of 1) a homogenized conductive additive; 2) non‐spherical particle geometries; and 3) overlapping particles on charge/discharge curves are analyzed. The result is a better understanding of the validity limits of P2D models. These new insights about the individual influences of the simplifications will be used to improve the homogenized model. The simulation of complex cathode structures, where several homogenization assumptions are violated, shows that the improved homogenized model reaches a very high accuracy, and, thus, overcomes the existing limitations of the P2D model approach.