Lithium-ion batteries provide laptops, smart phones, and tablet computers with reliable energy. However, electric vehicles using conventional lithium-ion batteries do not appear to be competitive with other technologies. This is largely due to currently utilized electrode materials such as graphite only being able to stably adsorb a limited number of lithium ions, restricting the capacity of these batteries.

Semiconductor materials like silicon are therefore receiving attention as alternative electrodes for lithium batteries. Bulk silicon is able to absorb enormous quantities of lithium. However, the migration of the lithium ions can swell the volume by a factor of three, which leads to major mechanical stresses destroying the crystal structure of the silicon.

Now a team from the HZB Institute for Soft Matter and Functional Materials has discovered two layers on the silicon electrode. The first roughly 20-nm layer formed with extremely high lithium content: specifically, 25 lithium atoms lodged among ten silicon atoms. A second adjacent layer contained only one lithium atom for ten silicon atoms. Both layers together are less than 100 nm thick after the second charging cycle. After discharge, about one lithium ion per silicon node in the electrode remained in the silicon boundary layer exposed to the electrolytes.

The team calculates from this observation that the theoretical maximum capacity of silicon-lithium batteries using these 100-nm layers to form electrodes, lies at about 2300 mAh/g (milliamp-hours per gram). This is more than six times the theoretical maximum attainable capacity for today's lithium-ion batteries constructed with graphite (372 mAh/g).

If such thin-layer silicon electrodes can be commercialized, lithium-ion batteries could see a 600 percent leap in energy-density, making them much more practical for real world applications like automobiles and drones.