Novel nanostructured cathodes for Mg-ion batteries

The project started in the fall of 2013. Post doc started in October and PhD candidate in December that same year. SINTEF initiated the project by performing theoretical calculations of different compositions of MgMSiO4 where M is a transition metal. The goal was to find a few select compositions which would be stable and could be used as potential cathodes in Mg-ion batteries. Based on the calculations it was decided to continue the experimental work with the compositions MgFeSiO4, MgMnSiO4 and MgCoO4. The PhD candidate and the post doc worked intensively on modification of synthesis parameters in order to produce phase pure materials. And eventually it was possible to produce phase pure materials of all three compositions by various techniques (sol-gel, molten salt, flame spray pyrolysis). When time came to test these materials in a battery cell, it turned out that they didn't work very well. A considerable amount of work was therefore done to find the source of the problem. All parts of the battery cell were investigated, including electrolyte, anode, separator, and current collector on the cathode side. By investigating a number of different electrolytes using materials that are known to work well (i.e. Mo6S8) we confirmed that there were no problems related to the cell setup and system, as these materials provided results comparable to published data. After a thorough literature study and re-examination of the published data for MgMSiO4, we suspected that the published data is incorrect. This suspicion is supported by the fact that all the published results related to these materials (MgMSiO4) come from one research group. No other research groups have managed to reproduce similar results. In order to strengthen our results even further, SINTEF has performed theoretical calculations on activation energy for transport of Mg in the various compositions which we have tested experimentally. These calculations show that transport of Mg in these materials is only possible in one dimension. This strongly suggests that Mg diffusion in a polycrystalline material, which is the case for all practical purposes, will be extremely slow if possible at all under standard test conditions. These results were published in Journal of Power Sources, and we hope that others will not make the same mistakes later. After we established that MgMSiO4 is not a possible cathode material, we started looking at other materials systems, including a variety of Mn oxides (MnO, Mn2O3 and Mn3O4) and carbon composites of these materials. It turned out that Mn3O4 was a material that worked surprisingly well as cathode in Mg batteries. We have also found that various types of graphite and graphene based materials are very promising as electrode materials in Mg batteries. In addition to the electrode materials, we have worked a lot with different types of electrolytes. The same Mg salt was used for all of them, but we investigated 4 different solvents. We found that there were large variations in both capacity and the underlying mechanisms, depending on the type of solvent used. And rather than intercalation of cations into the host lattice (which is common in Li batteries), we found that the capacity in these cells were mostly dependent on surface reactions and pseudo-capacitance. These are very new and exciting discoveries which may open up for a completely new type of Mg battery with considerably higher capacity and faster kinetics compared to standard Mg-ion batteries. SINTEF has in addition to the modelling work, synthesized and characterized V2O5-based materials as potential cathodes in Mg-ion batteries. This material has been widely studied as Li-ion battery cathode, but is relatively new as potential Mg-battery cathode. V2O5 has been synthesized by flame spray pyrolysis, which has given materials with very high surface area. These materials have shown very good initial capacity. However, the reversible capacity seems to be rather low. On the other hand, after the first irreversible capacity loss, the materials seem to be quite stable upon extended cycling.

Rechargeable Li-ion batteries are already widely used, and the dominating solution for automotive as well as portable, high-power applications due to the superior power density. However, several challenges of Li-ion battery technology are yet to be overco me, e.g. with respect to cost, resource availability, safety issues related to risk of thermal runaway of batteries and the limited temperature window of Li-ion batteries due to electrolyte deterioration at low temperatures. Although expected to dominate the portable battery market in the foreseeable future, the above mentioned drawbacks of Li-based batteries mean they have limited suitability in several markets, including stationary energy storage systems and auxiliary power applications where low cost a nd high reliability under a wide temperature range are more important than low weight and high power density. There is therefore a need for development of new battery types, specifically aimed at low cost markets. Batteries based on magnesium and other divalent metal ions, with the option of transferring two electrons, compared to the one electron transferred in Li-ion batteries, were introduced around year 2000 and are a promising solution for future low cost, high energy density batteries. Still, fund amental research is needed in order to produce high-performance batteries which are also stable and recyclable. In this aspect, an interdisciplinary approach covering modelling, innovative material synthesis routes and advanced experimental characterisati on techniques is a powerful combination which is herein proposed for the development of low cost novel Mg-ion batteries. Such developments will be beneficial for a wide range of applications, mainly stationary, that require use of batteries, such as for s torage of renewable energy from intermittent energy sources (solar, wind and wave), stand-alone power systems, telecommunication base stations, auxiliary power, etc.