Mg Desolvation and Intercalation Mechanism at the Mo6S8 Chevrel Phase Surface

Thursday, October 15, 2015: 15:20
101-B (Phoenix Convention Center)
L. Wan (Lawrence Berkeley National Laboratory), B. R. Perdue (Sandia National Laboratories), C. A. Apblett (University of New Mexico), and D. Prendergast (Lawrence Berkeley National Laboratory)
In recent years, there are strong interests to develop new divalent electrochemical systems that can potentially provide higher energy density compared with current Li-ion batteries, but with lowered cost and improved cycle stability. Mg-ion battery technology, first established in 2000 [Aurbach et al., Nature 407, 724 (2000)], proves the cyclability of such divalent battery cell. In this manuscript, we examine Mg-ion desolvation and intercalation process at the Chevrel phase Mo6S8 cathode surface using first-principles based methods. It is known that in chloride based electrolytes, Mg2+ is strongly coordinated by the counterion Cl- with a binding energy on the order of 3 eV. During cell discharge, this Mg-Cl bond is broken and Mg is intercalated into the cathode material. Our simulation indicates the broken of such strong ionic bond can only be possible if another electrophilic agent, such as Mo, exists on the cathode surface to attract Cl- away from Mg2+. Once Mg is intercalated, the Chevrel phase surface becomes chlorinated, i.e. all open Mo sites are attached with Cl-. Further simulations show that this chlorinated surface can continue to interact with incoming MgCl+ species and form various MgCly adsorbates. These surface adsorbates can then release neutral MgCl2 back into the electrolyte and reopen Mo sites to the electrolyte to permit continuous Mg intercalation. The newly release MgCl2 species will likely increase salt concentration near the cathode surface and initiate MgxCly oligomerization and precipitation, which are characterized by full-cell experiments.

Our results suggest that when designing new divalent battery systems, one should consider the compatibility issue between the electrolyte and the cathode material that has often been overlooked. For example, with the potential of increasing cell voltage, transition metal oxides, such as V2O5 and MoO3 are proposed as the Mg intercalation cathode. However, these layered oxide structures usually do not present open metal sites, but rather oxygen at their surfaces, which is less likely to attract any strongly coordinated counterions away from Mg2+ to assist its desolvation and intercalation. Therefore, such oxide cathodes usually require weakly coordinated anions to be used in the electrolyte, which may no longer be compatible with Mg metal as the anode material.

This work is supported as part of the Joint Center for Energy Storage Research (JCESR), an Energy Innovation Hub funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences.