Degradation Diagnosis of Aged Li4Ti5O12/LiFePO4 Batteries

Lithium-ion batteries longevity is of major concern for electric transportation applications. It is therefore necessary to understand the performance decay of a battery subject to aging conditions. Degradation mechanisms of the electrodes are generally divided into three families: (i) irreversible loss of active material, (ii) parasitic reactions leading to a loss of cyclable lithium or electrons, (iii) increase in resistance due to passive films formation and loss of contact [1, 2].

Lab-scale Li₄Ti₅O₁₂/LiFePO₄cells were submitted to accelerated aging during 4 to 5 months. All the cells exhibit capacity fade whose extent is not correlated with the aging condition. In order to understand aging phenomena, cells were disassembled at the end of cycle life and the recovered electrodes were analyzed using electrochemistry, electron microscopy, XRD and MAS-NMR. Positive and negative electrodes show no loss in active material and no change in electrochemical activity, active material structure and composite electrode structure. This rules out any irreversible electrode degradation. Lithium stoichiometry estimated by both XRD and electrochemistry is unexpectedly low in the positive electrode when the aging is stopped at full discharge. That indicates a loss of cyclable lithium or electrons leading to cell balancing evolution [3]. That loss has been caused by parasitic reactions occurring at both electrodes, in accordance with their rich surface chemistry evidenced by MAS-NMR.

This surface chemistry is fully characterized by MAS-NMR both qualitatively and quantitatively, using an original calibration method (see Fig. 1) [4]. Correlations are drawn between the quantity of diamagnetic lithiated species deposited on the surface of both electrodes (mainly LiF) and the loss of cyclable lithium or electrons undergone by the cell. This study attempts to develop a parasitic reaction pathway that explains the loss of cyclable lithium or electrons encountered under these aging conditions.

References:

[1] J. Vetter, P. Novák, M. Wagner, C. Veit, K.-C. Möller, J. Besenhard, M. Winter, M. Wohlfahrt-Mehrens, C. Vogler, A. Hammouche, Journal of Power Sources 147 (2005) 269.
[2] A. Barré, B. Deguilhem, S. Grolleau, M. Gérard, F. Suard, D. Riu, Journal of Power Sources 241 (2013) 680.
[3] M. Kassem, C. Delacourt, Journal of Power Sources 235 (2013) 159.
[4] M. Cuisinier, J. F. Martin, P. Moreau, T. Epicier, R. Kanno, D. Guyomard, N. Dupré, Solid State Nuclear Magnetic Resonance 42 (2012) 51–61.

Figure 1: Calibration of NMR-MAS ¹⁹F signal with LiF-electrode mixtures.