PNNL

Abstract

Non-lithium metals such as sodium have attracted wide attention as a potential charge carrying ion for rechargeable batteries, performing the same role as lithium in lithium- ion batteries. As sodium and lithium have the same +1 charge, it is assumed that what has been learnt about the operation of lithium ion batteries can be transferred directly to sodium batteries. Using in-situ TEM, in combination with DFT calculations, we probed the structural and chemical evolution of SnO2 nanowire anodes in Na-ion batteries and compared them quantitatively with results from Li-ion batteries [Science 330 (2010) 1515]. Upon Na insertion into SnO2, a displacement reaction occurs, leading to the formation of amorphous NaxSn nanoparticles covered by crystalline Na2O shell. With further Na insertion, the NaxSn core crystallized into Na15Sn4 (x=3.75). Upon extraction of Na (desodiation), the NaxSn core transforms to Sn nanoparticles. Associated with a volume shrinkage, nanopores appear and metallic Sn particles are confined in hollow shells of Na2O, mimicking a peapod structure. These pores greatly increase electrical impedance, therefore naturally accounting for the poor cyclability of SnO2. DFT calculations indicate that Na+ diffuses 30 times slower than Li+ in SnO2, in agreement with in-situ TEM measurement. Insertion of Na can chemo-mechanically soften the reaction product to greater extent than in lithiation. Therefore, in contrast to the lithiation of SnO2, no dislocation plasticity was seen ahead of the sodiation front. This direct comparison of the results from Na and Li highlights the critical role of ionic size and electronic structure of different ionic species on the charge/discharge rate and failure mechanisms in these batteries.