Kinetics and Passivation in Alkali Metal-O2 Cells: Limiting Mechanisms

Wednesday, 4 October 2017: 09:00
Maryland A (Gaylord National Resort and Convention Center)
K. B. Knudsen (University of California, Berkeley, Lawrence Berkeley National Laboratory) and B. D. McCloskey (Dept. of Chemical Engineering, UC Berkeley, University of California, Berkeley)
Non-aqueous alkali metal-O2 cells (in particular Li-O2 and Na-O2 cells) have received a great deal of attention since the aprotic Li-O2 cell first was reported in 19961. Typical metal-O2 battery configurations include a pure alkali-metal anode and inexpensive carbon acting as the cathode material upon which metal oxide discharge products are deposited during O2 reduction. The interest in alkali metal-O2 batteries is a result of their relatively high theoretical energy density of 1105 Wh/kg and 3456 Wh/kg (normalized to the weight of the discharge product) for Na-O2 and Li-O2, respectively2. The discharge product forms mainly as Li2O2 and NaO2 in Li-O2 and Na-O2 cells, respectively, and the morphology of these deposits depends on current density3,4, type of electrolytes5, and additives6,7. Deposits can generally take the shape of films growing on the cathode surface6, cubic crystals8, and/or larger irregular deposits as micrometer-sized particles6. As both Li2O2 and NaO2 have low electronic conductivities,9,10 once the cathode surface is sufficiently covered by the respective discharge product, charge transfer resistance substantially increases in most cases and a sudden drop in potential is observed in galvanostatic operation. Controlling the morphology is therefore a method being explored to increase the discharge capacity, with the interplay between the different morphologies largely governed by the solvation and solubility of O2, Li+, Na+, and their reduced states as LiO2 (the intermediate to Li2O2 formation) and NaO2. The solubility of these species has been shown to be correlated to a good first approximation with the Lewis basicity and acidity of the electrolyte solvent as described by the Gutmann donor and acceptor numbers (DN and AN). As an example, solvents with a relatively higher DN, such as DMSO compared to DME, lead to a dramatic increase in discharge capacity because cations are more solvated and shielded in the high DN electrolyte, allowing LiO2 to diffuse away from the cathode surface before it disproportionates to insoluble Li2O2 to form larger particles5,6.

This presentation will outline the use electrochemical impedance spectroscopy as an in-situ characterization tool to characterize charge transport in alkali metal-O2 cells while operating. While the impedance of Li+ intercalation materials has been extensively studied, the relatively larger overpotentials (100-200 mV2,11) for non-aqueous O2 reduction in aprotic electrolytes give a number of contributions that require careful evaluation. A particularly useful method is to characterize the electrochemical processes in terms of a physical representative equivalent circuit that includes the flooded porosity of the carbon cathode4,12. The characterization of the cathode impedance allows for the separation of contributions that originates from the flooded porous structure and from the surface/electrolyte interface. In turn, this allows extraction of a parameter that qualitatively describes the amount of discharge product formed in the porous structure of the cathode. Our impedance analysis further gives a more specific deconvolution of individual losses to more precisely evaluate faradaic and non-faradic contributions to the performance of the cathode during operation.


1. K. M. Abraham and Z. Jiang, J. Electrochem. Soc., 143, 1–5 (1996).

2. P. Hartmann et al., Phys. Chem. Chem. Phys., 15, 11661–11672 (2013).

3. B. D. Adams et al., Energy Environ. Sci., 6, 1772–1778 (2013).

4. K. B. Knudsen et al., J. Phys. Chem. C, 120, 10799–10805 (2016).

5. L. Johnson et al., Nat. Chem., 6, 1091–1099 (2014).

6. N. B. Aetukuri et al., Nat. Chem., 7, 50–56 (2015).

7. C. M. Burke, V. Pande, A. Khetan, V. Viswanathan, and B. D. McCloskey, Proc. Natl. Acad. Sci., 201505728 (2015).

8. P. Hartmann et al., Nat. Mater., 12, 228–232 (2012).

9. V. Viswanathan et al., J. Chem. Phys., 135, 214704-214704–10 (2011).

10. S. Yang and D. J. Siegel, Chem. Mater., 27, 3852–3860 (2015).

11. J. S. Hummelshøj, A. C. Luntz, and J. K. Nørskov, J. Chem. Phys., 138, 34703 (2013).

12. K. B. Knudsen, T. Vegge, B. D. McCloskey, and J. Hjelm, J. Electrochem. Soc., 163, A2065–A2071 (2016).