In theory, lithium-manganese phosphate could last for a similar number of cycles, because it has a similar crystalline structure. But it has the added advantage of potentially being able to store 20 percent more energy than lithium-iron phosphate, since it operates at a higher voltage. However, it has been particularly hard to modify lithium-manganese phosphate to overcome the fact that it's an electrical insulator.

Previous attempts have required processing precursor materials in a liquid solution before creating solid battery materials--a process that's too expensive for commercial production. The new method developed at PNNL eliminates this separate liquid-processing step, simplifying the process and making it compatible with existing manufacturing techniques.

To prepare the material, the researchers mix chemical precursors with paraffin wax and oleic acid. The wax and acid work together to cause the precursor materials to form crystals of a well-controlled size and shape without clumping up. The wax liquefies at the high temperatures used to process the material and acts as a solvent that replaces the separate liquid processing step used in earlier research.

So far, the material can only be charged at low rates (although it delivers power fast enough for many applications). Choi says one of the next steps is to develop a better process for coating the nanoplates with carbon, which should improve conductivity.

Although lithium-manganese phosphate is attractive because it stores more energy than lithium-iron phosphate, both take up a relatively large amount of volume compared to other types of electrodes for lithium ion batteries. Jeff Dahn, professor of physics and chemistry at Dalhousie University, says this could ultimately make them more attractive for stationary applications--such as storing power on the electricity grid to help smooth out variability from renewable sources--than for electric vehicles.