Mechanistic calendar aging model for lithium-ion batteries

In this work we present a novel mechanistic calendar aging model for a commercial lithium-ion cell with NCA cathode and silicon-graphite anode. The mechanistic calendar aging model is a semi-empirical aging model that is parameterized on component states of health, instead of capacity. Three component states of health are derived from the degradation modes, which are calculated by fitting the electrode potential curves at every check-up measurement. The aging data used for model parameterization spans 672 days of storage at 27 different combinations of ambient temperature (Tamb) and state of charge (SOC). To compensate for the influence of the check-up measurements on cell degradation, the aging data is pre-processed in two steps, considering immediate degradation caused by the check-up cycles and accelerated degradation during subsequent storage. The loss of active anode material is negligible during check-up-compensated calendar aging. For loss of lithium inventory and loss of active cathode material, Tafel and Arrhenius terms are successfully used to model Tamb and SOC dependence. The mechanistic calendar aging model predicts the capacity with < 1% mean deviation for 7 different storage conditions after 672 days without check-ups. The check-up compensation increases predicted lifetime by > 150% for exemplary storage at Tamb = 60 & DEG;C and SOC = 50%.

Automated summary

In this study, a novel mechanistic calendar aging model is proposed for a commercial lithium-ion cell with NCA cathode and silicon-graphite anode. The model parameterization is based on component states of health and accurately predicts capacity with < 1% mean deviation for different storage conditions. The check-up compensation significantly increases the predicted lifetime of the cell.