Initial Assessment of an Empirical Potential as a Portable Tool for Rapid Investigation of Li+ Diffusion in Li+-Battery Cathode Materials

Abstract

Substantial research activity is currently invested in the pursuit of next generation cathode materials for rechargeable Li-ion batteries. We carry out an initial assessment of the suitability of a recently described empirical potential [J. Phys. Chem. B 2006, 110, 11780] as a rapid, portable, and at least qualitatively accurate computational tool for screening large numbers of potential cathode materials for favorable Li-ion transport capabilities. Selected materials can then be examined more elaborately with more accurate but computationally more expensive first-principles approaches. As test systems for our initial assessment, we chose the group of phosphate olivines LiMPO4 (M = Mn, Fe, Co, Ni), promising candidates for next generation cathode materials and subject of numerous experimental and computational studies. To conduct the assessment, we determined the ground state structures of LiMPO4 from geometry optimizations with this empirical potential and with density functional theory (DFT) and computed activation barriers of Li-ion diffusion in LiMPO4 from molecular dynamics simulations based on the empirical potential and from minimum-energy-path DFT calculations. We show that structural results generated by the empirical potential are in good agreement with the DFT and experimental results and that barrier results produced by this potential are in good agreement with the DFT results and often in better agreement than values generated by custom parametrized empirical potentials.