Abstract
This Ph.D. project is within the scope of the ERC-ARTISTIC Project, which aims to correlate the fabrication process to the performance of lithium-ion batteries. To do so, the mesostructure of the electrode is either simulated with a 3-D physical model in accordance with the different fabrication parameters used, either stochastically generated, or reconstructed from tomography data. The mesostructure is imported and the partial differential equations are solved in Comsol Multiphysics, which relies on the finite element method. This Ph.D. investigations aim to unravel the link between structural observables and the performance of an electrode. The novelty of this study relies on the explicit consideration of the inactive phase of the electrode, as well as the investigations on the interphase that forms at the interface between the active material and the electrolyte.Several algorithms were developed during the course of this Ph.D. to address the obstacles to include the carbon binder phase explicitly in a 3-D electrochemical model. In particular, a stochastic generation functionality allowing a fine tuning over the electrode through the particle size distribution and shape, and the inactive phase morphology and a meshing algorithm enabling the meshing of an indefinite number of phase and saving into a format suitable to be imported into Comsol Multiphysics. All these tools were embedded into a single graphical user interface to increase their user-friendliness. Fine studies on the role of the inactive phase and its location were carried out, highlighting the importance of the explicit consideration of this phase. A particular focus was put on the calendering step and its impact on the mesostructure and the electrochemical performance to unravel the link between the properties of the mesostructure and the electrochemical behavior of a cell. Moreover, we investigated the impact of a heterogeneous layer arising from electrolyte decomposition at both the positive and negative electrodes, i.e. the cathode and solid electrolyte interphase. Finally, a full-cell model demonstrates the relevance of all the tools developed and presented in this work to achieve state-of-the-art studies. Indeed, a discharge of a full-cell with each particle individually identified in both electrodes illustrates the capability of the algorithms to allow fine investigations with the observables of each particles individually accessible