​Understanding and modelling lithium-ion battery aging are crucial to improving battery lifespans, whether it is for first-life batteries (for electric vehicles) or second-life batteries (for stationary applications). Until now, the available aging models could only predict battery states from performance that could be measured at cell level. Liten, a CEA Tech institute, developed a model that enables more detailed characterization.

The advance is based on the fact that the voltage signature of each electrode changes over time depending on the state of the electrode. The researchers used a method capable of assessing the available capacity of each electrode and, by extension, the useful capacity of the cell. Ante mortem observations were compared with data gathered during battery operation in a variety of test conditions (calendar testing and cycling) to further refine the model.

This differential voltage model does not yet deliver an understanding of the degradation mechanisms at work. However, it does provide insights into the causes of aging and so that sudden accelerations in the performance degradation curve can be better predicted. In addition, the model will also speed up the development of aging models for specific types of batteries with the goal of reducing the number of tests and the overall duration of testing.