Kinetic Monte Carlo Study of Li Intercalation in LiFePO<sub>4</sub>.
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 29272576.
- Also identified by DOI 10.1021/acsnano.7b08278.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
Even as a commercial cathode material, LiFePO<sub>4</sub> remains of tremendous research interest for understanding Li intercalation dynamics. The partially lithiated material spontaneously separates into Li-poor and Li-rich phases at equilibrium. Phase segregation is a surprising property of LiFePO<sub>4</sub> given its high measured rate capability. Previous theoretical studies, aiming to describe Li intercalation in LiFePO<sub>4</sub>, include both atomic-scale density functional theory (DFT) calculations of static Li distributions and entire-particle-scale phase field models, based upon empirical parameters, studying the dynamics of the phase separation. Little effort has been made to bridge the gap between these two scales. In this work, DFT calculations are used to fit a cluster expansion for the basis of kinetic Monte Carlo calculations, which enables long time scale simulations with accurate atomic interactions. This atomistic model shows how the phases evolve in Li<sub>x</sub>FePO<sub>4</sub> without parameters from experiments. Our simulations reveal that an ordered Li<sub>0.5</sub>FePO4 phase with alternating Li-rich and Li-poor planes along the ac direction forms between the LiFePO<sub>4</sub> and FePO<sub>4</sub> phases, which is consistent with recent X-ray diffraction experiments showing peaks associated with an intermediate-Li phase. The calculations also help to explain a recent puzzling experiment showing that LiFePO<sub>4</sub> particles with high aspect ratios that are narrower along the [100] direction, perpendicular to the [010] Li diffusion channels, actually have better rate capabilities. Our calculations show that lateral surfaces parallel to the Li diffusion channels, as well as other preexisting sites that bind Li weakly, are important for phase nucleation and rapid cycling performance.