Parametric stochastic 3D model for the microstructure of anodes in lithium-ion power cells

Abstract

The microstructure of anodes in lithium-ion batteries has a strong influence on their electrochemical performance and degradation effects. Thus, optimizing the morphology with respect to functionality is a main goal in battery research. Doing so experimentally in the laboratory causes high costs with regard to time and resources. One way to overcome this problem is the usage of parametric 3D microstructure models, which allow the realization of virtual morphologies on the computer. The functionality of microstructures generated with such models can be investigated by means of numerical transport simulations. The results of this procedure, which is called virtual materials testing, can be used to design anodes with improved morphologies that lead to a better electrochemical performance. Recently, a particle-based stochastic microstructure model for anodes in lithium-ion energy cells has been proposed. In the present paper, an extension of this model to describe the morphology of anodes in power cells, whose structure strongly differs from energy cell anodes, is introduced. The extensions include techniques to model anisotropic morphologies with a low volume fraction of the particle phase and strongly irregular particle shapes. The model is fitted to 3D image data of a power cell anode and validated using morphological image characteristics. Furthermore, we show examples of modifications of our microstructure model that can be made for generating further virtual morphologies. Finally, we briefly explain how electrochemical characteristics can be estimated using thermodynamically consistent transport theory. To illustrate this, we compute the cell potential over time during lithiation for image data of real microstructures as well as corresponding microstructures simulated by our model.