Many companies are required to keep lithium-ion battery cells in inventory as reserve parts, e.g. for warranty purposes. This can result in long periods of shelf storage, during which cells remain in a resting state at a specific state of charge and ambient temperature for several years without being used. Due to these extended timeframes, data on shelf storage of lithium-ion cells is limited. It is unclear how batteries will perform after such long storage duration and if they are still able to fulfill specific power requirements for their application.
In this study, we aim to bridge this research gap by investigating the influence of 10 years of uninterrupted shelf storage on 100 commercial 26650 LFP/C lithium-ion cells. Specifically, we aim to answer the following questions: 1) Which signs of degradation do battery cells exhibit after 10 years of shelf storage during initial characterization measurements? 2) How do cells stored for 10 years perform under various long-term operating conditions compared to newly manufactured cells?
To investigate immediate signs of degradation, we compare data collected today, after 10 years of shelf storage at 50% state of charge and 8°C, with measurements taken on the identical cells in 2014 after shipping. By performing a check-up procedure at 25°C, we quantify the influence of shelf storage on cell capacity and internal resistance. To extend the analysis, eight cells exhibiting lowest, highest, and average capacity loss are subjected to further characterization tests. Open-circuit voltage measurements at 0.01C and rate tests at multiple current rates between 0.02C and 3C allow for further investigation into signs of different degradation mechanisms.
To assess the effect of shelf storage on performance under long-term operation, 18 cells with similar capacity values were selected for subsequent cyclic operation tests. Six different cycling protocols at various temperatures, current rates, and discharge depths were applied to groups of three cells each. The design of these cycling protocols was based on an aging study conducted several years ago using cells of the same production batch approximately one year after shipping. These measurements allow for a direct comparison of capacity fade and resistance increase curves of shelf stored cells with newly manufactured cells, under the same conditions.
Both the characterization measurements carried out and the ongoing cycling measurements show preliminary results that offer interesting implications for the usability of such cells after long-term storage.