Degradation Mechanisms of Magnesium Metal Anodes in a (CF3SO2)2N- (TFSI)-Based Mg-Ion Electrolyte

Abstract

Efficiently using anodes based on the deposition of light metals would be a breakthrough that would produce leaps in energy density of rechargeable batteries. Magnesium is one of such metal electrodes that are not susceptible to dendrite formation. This unique property stems from (1) the thermodynamic stability in Grignard-based electrolytes and (2) the inherent preference to form three-dimensional deposits. However, the use of Grignard-based electrolytes results in severe chemical reactions between the nucleophilic electrolytes and electrophilic cathodes such as oxide materials. Accordingly, a lot of effort has been paid to find non-nucleophilic electrolyte solutions that deposit and dissolve Mg metal reversibly. Among non-nucleophilic anions, bis(trifluoromethylsulfonyl)imide (CF3SO2)2N–, TFSI) anion has shown exceptional stability on Mg metal anode. And the ethereal solution of the magnesium salt (MgTFSI2) has shown quite reversible Mg metal deposition and dissolution with coulombic efficiency of ~90% and reasonable voltage stability window of about 3 V. The most important difference from the Grignard-based electrolytes is that the TFSI-based electrolyte solution is non-nucleophilic and possibly compatible with oxide cathodes, which can increase the voltage and energy density of Mg rechargeable batteries. Herein, we report the mechanisms of degradation in the Mg metal anode and electrolyte after hundreds of cycles at high current density of 2 mA/cm2. Although the Mg metal anode operated without a significant problem at current densities lower than 0.1 mA/cm2, degradation was evident after long-term cycling at high current density. Overall, the Mg metal became mechanically brittle, with microscopy showing extremely porous structures (Fig. 1a). In addition, large hemispherical deposits up to several hundred micrometers were scattered or aggregated at the Mg surface (Fig. 1b). In turn, the color of the electrolyte turned to dark brown upon cycling, indicating a change in the chemical structure of the solvent. The macroscopic consequence was the short circuiting of the cells after several cycles at a high current density. These results suggest that the lemma of dendrite-free metal deposition is possible when the system meets several conditions such as absolute chemical stability of the metal in the electrolyte, adequate level of current density and charge density, and other engineering factors such as organic or inorganic additives that complexes with the metal ion and lead to more uniform metal deposition. Those understandings will serve as step stones to get to non-nucleophilic electrolytes, in which metal anodes operate without dendrite formation. Figure 1