Development of high potential rechargeable batteries (i.e., combining higher energy densities, higher safety, lower cost, etc) is strongly demanded for large-scale energy storage technologies, such as electronic vehicles and stationary electrical energy storage systems. Although lithium ion batteries (LIBs) are currently used for these purposes, their energy densities are approaching the theoretical limit. If Li metal could be used as an anode material instead of carbonaceous materials currently used, LIBs would have significantly high energy densities. However, the fatal problem that Li-metal dendrites are unfortunately formed in charge, which is responsible for short circuit, hinders construction of Li-metal-anode batteries.

Mg metal can be an alternative to Li in that it can deposit non-dendritic formation. Furthermore, Mg metal is very attractive to realize high energy density; it has high theoretical capacity (2205 mAh g^-1), which is approximately six times larger than that of carbonaceous materials, and relatively low electrode potential (-2.38 V vs. SHE) among other candidates for anode materials. Especially, its high abundance and inexpensiveness are advantageous to practical application and elements strategy. Whereas magnesium rechargeable batteries (MRBs) have been seen as a potential candidate for next-generation battery, there are many difficulties to realize MRBs actually. One of the serious problem is that there are few cathode materials which can accommodate large amount of Mg ions at high potentials. Recently, we have substantiated several spinel oxides containing Mg and transition metals such as MgCo₂O₄, MgFe₂O₄, MgMn₂O₄, etc., for new cathode active materials[1]. For example, in the case of MgCo₂O₄, its discharge capacity approximately amounts to 200 mAh g^-1 and the discharge potential is 2.0 - 3.0 V vs. Mg²⁺/Mg. These values are extremely high compared to that of Chevrel compounds[2]. In the previous work[1], by using CsTFSA based ionic liquid (melting temperature: c.a. 120 °C) for the electrolyte, we demonstrated that Mg cation was possible to be a carrier when the operated temperature was set at 150 °C. Furthermore, it was revealed that the spinel structure coherently transformed into rocksalt structure accompanied by the Mg²⁺ insertion, where the Mg²⁺ intercalation into vacant 16c sites (octahedral sites) of a spinel structure causes “push-out” of the cations located at the 8a sites (tetragonal sites) to other 16c sites, which eventually forms a rocksalt structure.

Although these spinel oxides are attractive candidates for cathode materials of MRBs, there remains some problems in cycle properties. In the case of MgCo₂O₄, when the battery tests were conducted at 150 °C in the ionic liquid, the discharge capacity at high potential (~ 2.5 V vs. Mg²⁺ / Mg) decreases with an increase in charge/discharge cycles. This would be due to a difficulty of reverting to ideal initial spinel structure completely after cycling, which would depend on the fact that Mg has a nature of forming rocksalt-structure oxides, that is, Mg intrinsically tends to favor octahedral site. Furthermore, it is expected that such degradation would also be accelerated by its relatively large volume changes in charge/discharge cycles. Based on these assumptions, we examined magnesium-free spinel oxides such as X₁(Co or Fe)₂O₄ (X = Ni, Zn). In ZnFe₂O₄, for example, its cycle property is found to be much improved in comparison with that of MgCo₂O₄, this would be because its lattice size based on the interatomic distance of oxygen anions is very similar to that in MgO and Zn is expected to favor tetrahedral site (in that ZnO forms wurtzite structure). Although its discharge potential is around 1.8 V vs. Mg²⁺ / Mg, being lower than that of MgCo₂O₄, an advantageous point is the fact that both Zn and Fe are inexpensive abundant elements. Thus, we can offer toiler-made spinel oxide materials for MRB cathodes.

References

[1] S. Okamoto, T. Ichitsubo, T. Kawaguchi, Y. Kumagai, F. Oba, S. Yagi, K. Shimokawa, N. Goto, T. Doi, E. Matsubara, Adv. Sci., 2, 1500072. (2015)

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