Beneficial Vs. Inhibiting Passivation By the Native Li SEI Revealed By Electrochemical Li+ Exchange

Abstract

Li anodes represent a theoretical 10-fold upgrade over the capacity of graphite at comparable potential, and as such are key anode candidates for next-generation high energy density (> 1000 Wh/L) batteries. However, Li still falls below the 99.95-99.97% CE required for long-life cycling(1) and displays rate capability an order-of-magnitude lower than necessary for fast-charging (>2 C).(2) Interestingly, intrinsic Li0/Li+ redox has been reported to be more facile (j 0 > 10 mA/cm2)(3) than graphite intercalation (j 0 = 1-3 mA/cm2)(4). In spite of this, the chemical exchange of Li+ through the native solid electrolyte interphase (SEI) on Li is typically slow (0.5-3 mA/cm2),(5) and as such it can bottleneck Li+/Li0 redox(3) and increase charge-transfer resistance.(6) The SEI thus manifests itself by substantially decreasing the exchange current j 0 measured on Li down from its intrinsic kinetic value.(7) While multiple recent studies see j 0 as a relevant property in determining Li reversibility,(1) measuring j 0 in the presence of an SEI is not straightforward. Consequently, current literature presents widely varying numerical values of j 0 in the presence of an SEI, making it challenging to discern the relationship between j 0 and CE.To bridge this gap, Li+ exchange at the Li anode is here systematically quantified using cyclic voltammetry (CV) at slow scan rates and electrochemical impedance spectroscopy (EIS), both of which allow an SEI to develop natively. To avoid ambiguity with the intrinsic Li0/Li+ redox exchange current j 0, exchange rates are here interpreted in the framework of a “pseudo”-exchange current, j 0 p, that represents the total rate of Li+ exchange on the electrode. j 0 p was measured across a selection of historically-relevant and modern electrolytes, spanning low (78.0%) to high (99.3%) CE. In both methodologies, a strong dependence of j 0 p on electrolyte chemistry was identified. These differences reflect a strong correlation between CE and j 0 p, with electrolytes that display higher j 0 p typically also displaying higher CE. Upon cycling, a dynamic behavior of Li+ exchange on both Cu and Li were observed, with j 0 p typically increasing through cycling, attributed to morphological changes induced by non-uniform plating/stripping inherent to Li electrochemistry.(8) We will discuss the implications of this dynamic behavior on both the formation cycle on Cu, as well as how j 0 p changes report on SEI evolution during cycling. Finally, it was found that cycling Li with current densities j beyond j 0 p leads to substantial capacity loss and low CE, whereas electrolytes that can sustain high j 0 p are insensitive to j. Altogether, our results indicate that Li+ exchange plays a dominant role in determining the rate capability and CE of Li anodes, with high-j 0 p electrolytes displaying higher CE and better rate capability than their low-j 0 p counterparts. G. M. Hobold, J. Lopez, R. Guo, N. Minafra, A. Banerjee, Y. Shirley Meng, Y. Shao-Horn and B. M. Gallant, Nature Energy, 6, 951 (2021).P. Albertus, S. Babinec, S. Litzelman and A. Newman, Nature Energy, 3, 16 (2018).D. T. Boyle, X. Kong, A. Pei, P. E. Rudnicki, F. Shi, W. Huang, Z. Bao, J. Qin and Y. Cui, ACS Energy Letters, 5, 701 (2020).Y.-C. Chang, J.-H. Jong and G. T.-K. Fey, Journal of The Electrochemical Society, 147, 2033 (2000).A. B. Gunnarsdóttir, S. Vema, S. Menkin, L. E. Marbella and C. P. Grey, Journal of Materials Chemistry A, 8, 14975 (2020).A. Zaban, E. Zinigrad and D. Aurbach, The Journal of Physical Chemistry, 100, 3089 (1996).J. N. Butler, D. R. Cogley and J. C. Synnott, Journal of Physical Chemistry, 73, 4026 (1969).J. Z. Lee, T. A. Wynn, M. A. Schroeder, J. Alvarado, X. Wang, K. Xu and Y. S. Meng, ACS Energy Letters, 4, 489 (2019).