Energy Environ. Sci. http://doi.org/bqwc (2016)

Lithium–oxygen batteries can potentially provide much higher energy densities than state-of-the-art battery technologies. However, the use of pure lithium metal as the anode is problematic as undesired electrolyte decomposition often takes place at the anode surface — especially when gaseous oxygen is present — which severely reduces the battery cycling stability. Besides, lithium deposition from the electrolyte can lead to dendrite formation on the lithium metal, which poses a short-circuit hazard. Silicon, on the other hand, offers an attractive anode candidate because of its high storage capacity and dendrite-free properties, but it suffers from its own problems such as huge volume expansion during charge and discharge, which also diminishes the cycling performance. Now, Haoshen Zhou and colleagues in Japan and China demonstrate how a lithium–oxygen battery with a silicon anode can achieve long-term operation.

Starting with commercial silicon particles, the researchers lithiated the material in a carefully designed lithium-ion half-cell, in which the silicon and lithium metal served as cathode and anode respectively and a lithium bis(trifluoromethylsulfonyl)imide salt with a tetraglyme solvent and a fluoroethylene carbonate additive was used as the electrolyte. The lithium–oxygen battery was then assembled with the pre-lithiated silicon as the anode. Because of the formation of a robust solid electrolyte interphase film on the silicon surface, both the volume change of the anode and undesired side reactions were effectively suppressed during battery cycling.