Abstract

A life model is necessary to predict the lifetime and state of health of a cell under various conditions. Main battery failure mechanisms include Li plating, Solid electrolyte interface (SEI) growth, loss of active material, and increased internal impedance. In this work, a comprehensive physics-based life model was developed to first understand the effects of different mechanisms such as loss of active material on capacity fade, and then to predict capacity fade and DCR of the cell at different operating conditions. A previously developed and validated electrochemical-thermal (ECT) model [1, 2] was adopted to be used in our life model. Li-plating mechanism was added to the ECT model and validated by comparing with charge/discharge voltage, mixed potential during relaxation [3] (for Li plating detection), and temperature. Li was observed to plate at higher rates, higher SOCs, and lower temperatures when Li plating reaction overpotential becomes negative [4]. A small fraction of the plated Li was considered as dead Li that contributes to capacity fade (loss of active Li ions). SEI mechanism was first separately modeled for storage condition (no current) in order to obtain the SEI reaction parameters by comparing a physics-based calendar model [5] with experimental storage data and minimizing the error between capacity retention of those two. Parameters include reaction rate constant at different temperatures and diffusion coefficient (of the solvent through the previously formed SEI layer) at different temperatures. We further consider gas formation on the cathode side at higher storage temperatures and SOCs to predict experimental cell DCR. After SEI parameterization, SEI reaction is added to the life model with the extracted parameters. A small fraction of the plated Li is also considered to turn into SEI. Both SEI and dead Li are assumed to form on the surface of the anode active particles and affect porosity and tortuosity, especially close to the separator. Moreover, only SEI affects film resistance because conductivity of plated Li is very high compared to SEI conductivity. Finally, active material loss in the anode and cathode side is added to the model. The existence of the active material loss is important since losing a fraction of active material causes usable SOC window to shift (usually wider window) resulting in usable capacity reduction. In addition, active material loss can reduce anode OCV at high cell SOCs that results in more negative Li plating and SEI reaction overpotential that accelerates the rates of SEI and Li plating. [1] M.M. Forouzan, S.K. Rahimian, S. Han, Y. Liu, Y. Tang, in: Meeting s, The Electrochemical Society, 2019, pp. 394-394. [2] M.M. Forouzan, S.K. Rahimian, S. Han, Y. Liu, Y. Tang, in: Meeting s, The Electrochemical Society, 2019, pp. 597-597. [3] X.-G. Yang, S. Ge, T. Liu, Y. Leng, C.-Y. Wang, Journal of Power Sources, 395 (2018) 251-261. [4] M.M. Forouzan, B.A. Mazzeo, D.R. Wheeler, J. Electrochem. Soc., 165 (2018) A2127-A2144. [5] N. Kamyab, J.W. Weidner, R.E. White, J. Electrochem. Soc., 166 (2019) A334-A341.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call