531
Physics-Based Modeling of Aging in NMC/Graphite Cell Under High Voltage Operation at Different Temperatures

Wednesday, 3 October 2018
Universal Ballroom (Expo Center)
J. Y. Ko, M. Varini (KTH Royal Institute of Technology), H. Ekström (KTH Royal Institute of Technology, COMSOL AB), and G. Lindbergh (KTH Royal Institute of Technology)
The high demand on specific capacity and occasionally harsh environment in electric vehicles has detrimental effects on electrochemical, mechanical and thermal properties of NMC/graphite Li-ion batteries. The battery lifetime is reduced at high voltage and high temperature, both in terms of capacity fade and impedance increase. At the graphite electrode, the notion of SEI layer formation and growth leading to battery aging is widely accepted. On the other hand, aging effects on the NMC electrode needs to be investigated more thoroughly. Xiong et al. [1] reported that Li-ion cells could experience rapid failure due to impedance rise at the NMC electrode when cycled at an elevated temperature or a cut-off voltage above 4.2 V. A surface film formation for transition metal oxides, the growth of which accelerates upon high temperature and charge voltages, has been reported [2,3]. Literature also shows that the interaction between the positive and negative electrodes in full lithium-ion cells contribute to declined cell performance [4]. However, these aging mechanisms are not independently distinguishable. Therefore, it is important to be able to quantify each contribution and to decouple different aging mechanisms occurring on each electrode at different temperatures under high voltage condition.

This study focuses on impedance behavior of commercial NMC111 (LiNi1/3Mn1/3Co1/3O2) and graphite electrodes in a small laboratory pouch cells. The NMC111/graphite in full-cell configuration are calendar aged and cycle aged between +3.8 V and +4.6 V vs. Li at 25 °C and 40 °C until 50% capacity fade is reached. Capacity and impedance behavior of the full cells are measured periodically in order to track the performance changes with aging. Each electrode is harvested at specific intervals of cycling stages and electrochemically characterized in symmetric cells by means of electrochemical impedance spectroscopy (EIS).

A physics-based model is employed to parameterize aging parameters by fitting the EIS results from the symmetric cells using a least square fitting tool in COMSOL Multiphysics. The pseudo two-dimensional (P2D) model is an extension to the work of Doyle et al. [5] incorporating intercalation kinetics, mass and charge transport based on concentrated solution porous electrode theory.

From the preliminary EIS results on NMC111 symmetric cells in Fig. 1, impedance increases at a small magnitude at 25 °C at the end of life (EOL) compared to the beginning of life (BOL). At 40 °C, however, large impedance rise at EOL is seen. On the other hand, it had been shown that capacity fade occurs at both temperatures during slow cycling on half-cells. The model optimization is firstly done for cells at BOL to deduce the aging-independent parameters. Towards EOL, the quantification of intrinsic physical aging properties such as the particle sizes, porosity, local contact resistance and surface film resistance which are the causes of impedance rise or capacity fade can be made. The potential effects of calendar aging on cell degradation is separated from the effects of cycle aging. The comparison with experimental capacity loss and impedance rise will further validate the model. The extraction of the parameters from the model enables deduction of different aging-induced mechanisms at each electrode. This quantitative analysis of aging behavior will contribute to improved lifetime predictive model for NMC111/graphite lithium-ion batteries.

[1] D. J. Xiong, R. Petibon, M. Nie, L. Ma, J. Xia, J.R. Dahn, J. Electrochem. Soc. 163 (2016), 3, A546-A551

[2] P. Niehoff, M. Winter, Langmuir 29 (2013), 51, 15813-15821

[3] M. Wohlfahrt-Mehrens, C. Vogler, J. Garche, J. Power Sources 127 (2004) 58-64

[4] K. Xu, Chem. Rev. 114 (2014) 11503-11618

[5] M. Doyle, J. Newman, A. C. Gozdz, C. N. Schmutz, J.-M. Tarascon, J. Electrochem. Soc. 143 (1996) 1890–1903.