Wednesday, May 23, 2018 by Dr. Ilona Glatt, Math2Market GmbH.
Edited by Dr. Barbara Planas and Franziska Arnold, Math2Market GmbH.
Improving battery materials is a challenge. The desired properties are not independent and sometimes, even conflicting. For example, a material well suited for fast charging does not necessarily provide a high capacity. Consequently, it is essential to design the battery materials to fit the desired application.
The performance of a battery is determined by the existing transport routes in the electrode material, just as the interaction of the traffic infrastructure determines the efficiency of the city - roads for cars, sidewalks for pedestrians, rails for trains. Individual particles, electrons, dissolved lithium ions, and bound lithium in an electrode material only move along transport routes intended for them. Just as cars that only move on roads and pedestrians only on sidewalks, lithium ions can only move in the pore space filled with electrolyte, electrons can only move in the active material and in the (even better conductive) binder material. Clearly, then, the analysis of geometry of the microstructure must play a decisive role in the development of new battery materials.
The core competence of our GeoDict software has always been the analysis of the microstructure’s geometry and its modification. However, the inference as to how the geometry affects the performance of a battery has not yet been drawn.
To close this gap, the new GeoDict module BatteryDict was introduced with the release of GeoDict 2018 in the fall of 2017. BatteryDict allows testing the digitally developed battery material directly in a charging simulation, using the integrated BEST solver. The Battery and Electrochemistry Simulation Tool (BEST) is developed and already successfully used at the Fraunhofer Institute for Industrial Mathematics (ITWM). Compared to the established BestMicro-solver, BatteryDict is backed by the BESTMicroFFT-solver, based on fast Fourier transformations, that allows the analysis of larger structures.
The result of analysis with BatteryDict is a charging curve with the cell potential displayed as a percentage of the charge state. Also detected are the unconnected parts of the structure that cannot contribute to the capacity and are undesirable. The three-dimensional lithium concentration, that is given for each charge state, provides valuable information to detect bottlenecks in the structure’s transport routes – the so-called overpotential. Similar to a traffic jam on the highway, lithium particles accumulate here and prevent the battery from working effectively.
For the GeoDict 2019 release, in the fall of 2018, a number of new features have been incorporated in BatteryDict. Important is the inclusion of binders, or other active materials in electrodes, as additional Material IDs now handled by BatteryDict. The charging behavior of the battery is now realistically simulated because of the high electron-conductivity of binder that makes its distribution play a major role in the analysis of transport properties.
A particularly interesting aspect in the development of batteries such as the determination of the working lifespan is addressed with the inclusion of further Material IDs. That is, by handling several material IDs and taking into account the binder, a link to the mechanical properties of an electrode can be later established. The mechanical properties, in turn, play an important role in the damage caused by cracks in the particles of the active material. Handling several material IDs is also necessary to simulate the formation of the SEI (solid electrolyte interphase) layer and its repeated breaking open and re-forming.