Advances in x-ray computed tomography (XCT) with sub-micron resolution and millimeters of field-of-view have enabled imaging of battery electrodes in sufficient detail to elucidate the mesostructure of the percolated particle network. Most often the secondary conductive binder phase is not visible using XCT and its structure can only be observed using electron-based techniques (e.g. scanning electron microscopy), which typically have a field-of-view too small for meaningful reconstructions. This secondary phase, however, is thought to be critically important for understanding electrode performance at the mesoscale.

We have developed a computational approach to creating computable three-dimensional (3D) mesoscale representations of the active material (particle), conductive binder, and electrolyte/void phases. This approach is built upon the Conformal Decomposition Finite Element Method (CDFEM) and combines 3D XCT image data for the active material phase and a novel “binder bridge” placement algorithm for the conductive binder. This approach enables the calculation of effective transport properties, such as electrical and ionic conductivity and mechanical moduli, that can be upscaled to macroscale battery simulations. Coupled electrochemical and mechanical simulations can also be performed to predict (dis)charge performance and degradation mechanisms of the electrode.

In this work, we discuss the computational requirements, in terms of mesh resolution and domain size, to reasonably minimize computational errors. We also compare our “binder bridge” approach to other approaches in the literature, including a popular stochastic representation. The effect of nano-porosity within the conductive binder phase is explored. As part of this comparison, we explore the differences between smooth and voxelated particle representations along with differences between finite element and finite volume methods. Finally, we discuss upscaling of some of these effective properties into functional forms that can be implemented in macroscale battery simulation platforms.

Sandia National Laboratories is a multi-mission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.