Revisiting Chevrel Compounds Toward High-Voltage Mg Rechargeable Battery

From the viewpoint of energy and environmental problem, high energy-density battery that can store a surplus electric power, so-called rechargeable battery, is strongly demanded. Currently, lithium ion batteries (LIBs) are indispensable for the modern society. The energy density of LIBs has been enlarged year by year, but recently its growing rate tends to be saturated. Since lithium metal has a high gravimetric/volumetric capacity and the lowest electrode potential (-3.05 V vs SHE), if lithium metal is available as a negative electrode instead of carbonaceous materials currently used, the lithium battery would have shown a significantly high energy density. However, actually lithium metal cannot be adopted for LIBs due to the well-known fatal problem on the dendritic growth of lithium during charge, which leads to a dangerous short circuit.

To overcome this fatal problem, in late years, non-noble polyvalent metals (such as Ca, Mg, Al) are tried to be used as a negative electrode for polyvalent-metal rechargeable batteries. This is because polyvalent metals can be expected to work as a potential negative electrode, whose capacity is much higher (e.g., for Mg, about 2200 mAh/g) than that of carbonaceous material (about 370 mAh/g) used in the current LIBs. Especially since it was reported in 2000 that Chevrel-type sulfide compounds can be used as the positive electrode of a magnesium rechargeable battery (MRB),[1] MRB studies have attracted growing attention not only due to the low standard electrode potential of magnesium (-2.36 V vs SHE), but also due to its abundance, inexpensiveness, and relatively safe handling. As demonstrated by the previous work, the magnesium deposition can be done with the Grignard regent (R-MgX, R: Alkyl or aryl group, X: halogen) and AlR_xCl_3-xsalt in the tetrahydrofuran (THF) solvent, and the coulombic efficiency associated with cathodic electrodeposition and anodic dissolution of magnesium in the Grignard/THF electrolyte is high. More importantly, unlike Li metal, magnesium can be electrodeposited rather smoothly without the dendritic growth that causes a fatal concern about a short circuit in the battery. 

However, comparing to the research field in LIBs, where we can opt many combinations of positive/negative electrodes and electrolytes, the MRB study is quite limited due to few choices in combination of the positive electrode materials and electrolytes. Because, despite that several candidates for the positive electrode materials of MRBs have been reported, currently only Chevrel compounds are known to show superior intercalation/deintercalation characteristics with an excellent cyclability. In the Chevrel structure, the diffusion of Mg cations can occur but, unfortunately, even when the Chevrel compounds are used as the positive electrode, it can deliver less voltage (about 1.2 V) in comparison with the cell voltage of LIBs (about 4 V). However, from our electrochemical experiments, we have noticed that there are several crucial mysteries in the Chevrel compound in using as the positive electrode, as seen in Fig. 1; the various values of the redox potentials can be observed for the Chevrel-compound positive electrode. With the prospect that a realistic solution toward the practical application of MRBs is to clarify these mysteries in the Chevrel compound, here we try to reveal why the Chevrel compound shows such a low redox potential as about 1.2 V vs Mg/Mg²⁺in the conventional battery system of Aurbach et al.[1] and demonstrate how to exploit the intrinsic high redox potential of the Chevrel phase.

[1] D. Aurbach, Z. Lu, A. Schechter, Y. Gofer, H. Gizbar, R. Turgeman, Y. Cohen, M. Moshkovich, and E. Levi, Nature 407, 724-727 (2000).