A Novel Approach to Continuum FE Multiphysics Simulation of Batteries via Heterogeneous Homogenization

Abstract

Advancements in modern technology necessitate batteries that can be safely charged and discharged at high rates with minimal aging. A promising strategy to achieve these ends is through optimization of porous microstructures of existing electrode materials to promote better electrical and ionic transport without compromising energy density, as has been shown experimentally by introducing laser ablation channels to the electrodes. The present work proposes a microstructure up-scaling methodology that efficiently accounts for material heterogeneity by assigning a porosity field to a continuum scale finite element mesh, effectively homogenizing materials while maintaining the heterogeneous porosity distribution inherent to the original microstructure. Constitutive relationships relating solid electric conductivity and pore tortuosity to local porosity are directly computed from fully resolved anode, cathode, and separator microstructure 3D models and assigned in the continuum as material properties of a finite element mesh. This allows for utilization of the computationally efficient, homogenized Newman model while maintaining dependency on the heterogeneous local porosity. Then, coupled electrochemical-thermal-mechanical-pore pressure finite element simulations are performed on the battery cell model with this heterogeneous porosity field, and associated local variations on transport properties are prescribed as material properties by the precomputed constitutive relationships. Performance metrics and lithium plating potential are compared to the results of simulations of the fully homogenized cell with the same constitutive relationships and to the base case of a fully homogenous cell with typical modelling assumptions applied to porosity dependent constitutive relationships. These comparisons are made both as a means of validating the methodology and to demonstrate the vast differences that arise when simplifying assumptions are used instead of the constitutive relationships of the actual material microstructure. Finally, the methodology is used in its intended optimization application to compare the performance of a typical battery microstructure to a similar porous microstructure with micro holes added via laser ablation. The results indicate that differences in the microstructure of materials with identical average porosities and similar constitutive relationships can have significant differences in both cell performance and aging that would typically be impossible to detect using fully homogeneous continuum modeling. Figure 1