CEA-Leti, a leader in the development and manufacturing of integrated solid-state batteries, is collaborating with InjectPower, a cutting-edge start-up, to develop an innovative power solution for miniaturized implantable medical devices. Thin-film all-solid-state battery technology currently stands out as the leading choice for delivering high energy density and customizable form-factor power sources. However, despite this advantage, capacity retention during cycling remains insufficient, with the goal of 1,000 cycles and less than 10% capacity loss still unmet. Additionally, a comprehensive understanding of the physical mechanisms driving performance degradation in microbatteries is lacking.

During this PhD, you will contribute to the development and refinement of our physical model, focusing on accurately describing microbattery behavior during cycling and fast charging. You will also apply our physically informed Bayesian machine learning model to identify key factors that influence battery performance, including charge-discharge protocols, storage conditions, and device architecture. Model training and validation will be based on data collected from automatic probers on silicon wafers containing thousands of microbatteries.