Solubility and Stability of NaO2 in Diglyme-Based Na-O2 Batteries

Abstract

Aprotic Na-O2 batteries have attracted extensive attention because of their relatively large theoretical energy density and low charge overpotential. Several discharge products have been reported, including NaO2,[1] Na2O2,[2] Na2O2·2H2O,[3] Na2CO3,[4] which adds in complexity to specify the chemistry and reaction mechanism of Na-O2 battery.[5] In ether-based electrolyte, NaO2 is recognized as the major discharge products and often appeared as NaO2 cubes with size of 1-50 μm. These NaO2 cubes is believed to form through a dissolution-precipitation process as previously reported by operando transmission electron microscope (operando TEM),[6] rotating ring-disk electrode (RRDE),[7] atomic force microscopy (AFM)[8] and two-working electrode electrochemical technique.[9] The solubility and stability of NaO2 in electrolyte are important factors which influence the dissolution-precipitation path and cycling performance respectively. However, direct measurement of NaO2 solubility in ether-based Na-O2 batteries is still not available[9] and the lack of investigation on NaO2 stability makes such system even more mysterious. The two-working electrode electrochemical technique shows great promise in measuring the solubility of NaO2 in cell conditions, because it can quantitatively detect dissolved NaO2. Herein, modified two-working electrode Na-O2 batteries (Figure a) are fabricated and galvanostatic method is applied to investigate the solubility of NaO2.[9] NaO2 is generated on one air electrode (WE1), dissolved in electrolyte and oxidized on the other air electrode (WE2). Based on the assumptions that all dissolved NaO2 can be oxidized and the oxidation rate of NaO2 is significantly larger than dissolution rate, the solubility of NaO2 in electrolyte can be estimated. As discharge time of WE1 increases, the length of charge plateau of dissolved NaO2 at ~2.3 V vs. Na/Na+ on WE2 firstly increases and gradually approach a maximum value (Figure b), implying saturation of NaO2 in electrolyte is reached. Substantial solubility of NaO2 (~1.4 mM) in diglyme-based electrolyte is obtained. Conventional two-electrode Na-O2 batteries consisting of a Na anode and a carbon cathode are also fabricated to investigate NaO2 stability. These Na-O2 batteries are discharge-(rest)-charge tested at 25 μA/cm2 for varied time (Figure c-e). The percentage of recovered NaO2 (ratio of charge capacity due to NaO2 to total discharge capacity) against discharge time is ploted in Figure f, the longest discharge time shows the highest recovery percentage of NaO2. The recovery results depend on the concentration of NaO2 in electrolyte, which is in turn governed by supersaturation and precipitation processes. The average degradation rate of NaO2 during rest for varied discharge time can also be calculated to evaluate cycling performance. Our research therefore provides additional insights to dissolution-precipitation mechanism and NaO2 stability. The authors acknowledge the financial support from National Natural Science Foundation of China (NSFC)/Research Grant council (RGC) of Hong Kong Joint Research Scheme N_HKU728/17. [1] P. Hartmann, C. L. Bender, M. Vračar, A. K. Dürr, A. Garsuch, J. Janek, P. Adelhelm, Nat. Mater. 2013, 12, 228. [2] Y. Hu, X. Han, Q. Zhao, J. Du, F. Cheng, J. Chen, J. Mater. Chem. A 2015, 3, 3320. [3] X. Bi, R. Wang, L. Ma, D. Zhang, K. Amine, J. Lu, Small Methods 2017, 1, 1700102. [4] J. Kim, H. D. Lim, H. Gwon, K. Kang, Phys. Chem. Chem. Phys. 2013, 15, 3623. [5] S. Zhao, B. Qin, K. Y. Chan, C. Y. V. Li, F. Li, Recent development of aprotic Na-O2 batteries, Batteries & Supercaps 2019, Under review. [6] L. Lutz, W. Dachraoui, A. Demortière, L. R. Johnson, P. G. Bruce, A. Grimaud, J. M. Tarascon, Nano Lett. 2018, 18, 1280. [7] C. Xia, R. Black, R. Fernandes, B. Adams, L. F. Nazar, Nature Chemistry 2015, 7, 496. [8] I. M. Aldous, L. J. Hardwick, Chem. Commun. 2018, 54, 3444. [9] P. Hartmann, M. Heinemann, C. L. Bender, K. Graf, R.-P. Baumann, P. Adelhelm, C. Heiliger, J. Janek, The Journal of Physical Chemistry C 2015, 119, 22778. Figure 1