|About this Abstract
||2022 TMS Annual Meeting & Exhibition
||Advanced Materials for Energy Conversion and Storage 2022
||Machine Learning-driven Analytics for Solid-state Batteries
||Debanjali Chatterjee, Bairav S. Vishnugopi, Partha P. Mukherjee
|On-Site Speaker (Planned)
Solid-state batteries (SSBs) hold the potential to improve the energy density and power density of conventional lithium-ion batteries while offering enhanced safety attributes. Microstructural arrangement of both the porous cathode and solid electrolyte play a critical role in determining percolation pathways and effective mechanical/transport properties. Coupled kinetic and transport processes in these electrodes, the spatial distribution of constituent phases and the physiochemical interactions between them dictate the electrochemical performance of the SSB. However, microstructure of the porous electrode and solid electrolyte is intrinsically stochastic, heterogeneous, and anisotropic, which makes the characterization of their effective properties complex. In this regard, machine learning can be used as a fast and effective tool to characterize key electrode microstructural attributes. In this work, we use a combination of physics-based modeling and machine learning-driven analytics to characterize key microstructural features like transport percolation, pore connectivity and active interfacial area in SSB systems.
||Energy Conversion and Storage, Machine Learning, Modeling and Simulation