Microstructure resolved modeling and simulations of batteries as tool for understanding structure-function relations

Abstract

I will give an overview of our ongoing developments of structure resolved models for the description of transport and electrochemistry in batteries. The underlying principle of our research is to derive physics and chemistry based models as rigorous as possible with the help of concepts from non-equilibrium thermodynamics. The ultimate goal is to identify theories, which are based on quantities, which can either be directly and unambiguously measured or which can be inferred from lower scale theories as DFT in combination with Monte Carlo methods or molecular dynamics simulations. The identified models are implemented in our software tool BEST (Battery and Electrochemistry Simulation Tool), which is a highly object oriented Finite Volume Code. Input for the microstructure resolved simulations are either virtually created geometries or real tomographies of battery electrodes. Result of the simulations are on the micrometer scale locally resolved fields of electrical potentials, concentrations, heat sources, electrical currents and ionic fluxes. For the simulations of full battery cells on centimeter scale, the identified models are homogenized to obtain porous electrode models, which are also implemented in our software tool. The effective parameter of the porous electrode models can directly be obtained by performing numerical averages of the microstructure resolved simulations. With this procedure it is guaranteed that microstructure and cell scale simulations are based on exactly the same models. Incompatibilities between simulation results and experiments can therefore be investigated on several scales and the imposed constraints between microstructure and scale scale help to avoid overfitting of data.