Ageing Rate of Li-Ion Battery Cells As a Function of Temperature, C-Rate, and State of Health

Abstract

The ageing rate and consequentially service life of mass-produced lithium-ion batteries depends on many external and internal factors. Some of the more important ones are the charge-discharge rate, ambient temperature, electrode thickness and state of health (SoH). This makes the estimation of the lifetime of battery cells very challenging and time-consuming. This motivates the development of models capable of determining the ageing behavior of a battery cell with relatively little experimental data.In this study1 we build on our previous work2,3 and use Arrhenius plots of the ageing rate to analyze the service life of battery cells in a relatively simple and straightforward way. We cycle aged two types of cells – commercial 5 Ah high-energy 21700 cells with Si/graphite anode (≈3 % Si) and LixNi0.90Co0.05Al0.05O2 + LiyNiO2 (NCA + LNO) cathode and lab-made 0.1 Ah pouch cells with LixNi0.33Mn0.33Co0.33O2 (NMC111) cathode and graphite anode. The ageing is ensured by systematically cycling the cells in a temperature range of -15 °C to 60 °C at various C-rates (0.2 C – 1 C).The capacity fade data is analyzed to obtain the ageing rate (% of SoH per cycle) as a function of temperature, which forms the V-shaped Arrhenius plots shown in Figure 1. Each set of data clearly shows two slopes that cross at intermediate temperatures. As shown previously experimentally2,3 and theoretically,4 the slopes correspond to two ageing mechanisms – lithium plating in the low temperature region and growth of solid-electrolyte interface (SEI) at high temperature region. The intersection of the two Arrhenius slopes denotes a transition between the dominating ageing mechanism and also denotes the optimum temperature at which the battery cell has the highest service life.By analyzing the data, we show that the crossover temperature shifts towards higher temperature values with increasing C-rate and increasing thickness of the anode, as both increase the electrode area-normalized current. The crossover temperature depends on the SoH of the battery. The precise trend depends on the cell chemistry, as we observe increase of the crossover temperature for one type of cells and decrease of crossover temperature for the other type of cells. The shift of the crossover temperature to lower temperatures observed in pouch cells could be explained by a more rapid loss of cyclable lithium, whereas the shift towards higher C-rates observed in commercial 21700 cells can be observed due to increasing practical C-rates as the electrodes degrade and loose some of their initial capacity. By using relatively little input data, we are able to predict the temperature- and C-rate-dependent end of life of a battery cell with good precision.Evaluating the Arrhenius-type dependences of the ageing rate of the battery cells provides valuable insights for use in battery management systems and extending the service life of battery cells. Acknowledgements Gints Kucinskis acknowledges Latvian Council of Science project “Cycle life prediction of lithium-ion battery electrodes and cells, utilizing current-voltage response measurements”, (no. LZP-2020/1-0425). The part of the research performed at ZSW was carried out in the framework of the industrial collective research program (IGF no. 20884 N/2). It was supported by the Federal Ministry for Economic Affairs and Energy (BMWi) through the AiF (German Federation of Industrial Research Associations eV) based on a decision taken by the German Bundestag. References G. Kucinskis, M. Bozorgchenani, M. Feinauer, M. Kasper, M. Wohlfahrt-Mehrens, T. Waldmann, submited to J. Power Sources.T. Waldmann, M. Wilka, M. Kasper, M. Fleischhammer, and M. Wohlfahrt-Mehrens, J. Power Sources, 262, 129–135 (2014) https://linkinghub.elsevier.com/retrieve/pii/S0378775314004352.M. Bozorgchenani, G. Kucinskis, M. Wohlfahrt-Mehrens, and T. Waldmann, J. Electrochem. Soc., 169, 030509 (2022) https://iopscience.iop.org/article/10.1149/1945-7111/ac580d.X.-G. Yang and C.-Y. Wang, J. Power Sources, 402, 489–498 (2018) https://linkinghub.elsevier.com/retrieve/pii/S0378775318310462. Figure 1. Arrhenius plots based on ageing rate for (a-e) commercial 21700 cells at 100 %, 95 %, 90 %, 85 % and 80 % of SoH and (f,g) lab-made pouch cells at 100 % and 95 % SoH. Figure 1