Investigation of Capacity and Homogeneity Recovery of Commercial Cells After Cycle Life Tests

摘要

In this poster, two strategies are introduced to assess irreversible capacity loss during shallow cycling of commercial Li-Ion cells at different average SOCs. The results show that due to superposed reversible capacity effects, a simple evaluation of capacity trend is not sufficient (10.1016/j.est.2018.06.003). Those reversible effects are related to contributions of the anode overhang (geometrical oversized anode) and of the homogeneity of lithium distribution (HLD) in the electrodes. The anode overhang leads to lithium ion migration from or to the active anode leading to a superposed decrease or increase in extractable capacity respectively. The HLD is a concept to measure the uniformity of SOC distribution of the particles within the electrodes during a discharge process. The HLD is qualified by evaluating peak heights in differential voltage analysis (10.1016/j.jpowsour.2017.09.059). Thus, for a high SOC spread within the electrodes, a low HLD is obtained. The cycling test is additionally followed by a rest period to recover capacity, which is further referred to as 'relaxation method'. During this rest period, a certain additional calendar aging must be considered. In the following two corner cases are investigated by storing the cells at a fixed low and the same SOC as during cycling. While the pure contribution of HLD is assessed by storing the cells at the same average SOC as during cycling, the contribution of the anode overhang is evaluated for a defined low SOC. In this work, both SOC strategies are tested and compared. During the storage phase in all cases, the extractable capacity and the HLD rise supporting the reversible capacity theory based on an inhomogeneous SOC distribution over the electrodes. The findings are extremely important regarding accelerated aging tests for applications like electric vehicles (EV) with long idle periods. Cycling of a test cell, neglecting pauses, leads to higher inhomogeneities within the cell and therefore to lower the amount of extractable capacities and potentially faster aging. The here presented method improves the accuracy of lifetime prediction as only the irreversible losses are included in aging models and reversible effects are added in a separate step.