Lawrence Berkeley National Laboratory

Triplite-type LiFeSO4F has attracted considerable attention as a promising cathode for next-generation lithium-ion batteries because of its high redox potential based on earth-abundant Fe2+/3+. However, successful extraction/reinsertion of all the lithium ions in triplite host is challenging even at a low current rate, resulting in a low specific capacity. These experimental findings contrast with previous theoretical works that predicted that the triplite structure would be a fast ionic conductor with low activation barriers for lithium-ion hopping. Origin of this discrepancy is elusive to date. Herein, combined first-principles calculations and high-angle annular dark-field scanning transmission electron microscopy analyses reveal that typical triplite structure is composed of nanodomains consisting of corner-shared FeO4F2 octahedra, whereas their domain boundaries are regions of mixed corner/edge-shared FeO4F2 octahedra. More importantly, these locally disordered domain boundaries significantly reduce the overall lithium diffusivity of the materials. Inspired by these findings, this study redesigns triplite structure with sufficiently small sizes to avoid local bottlenecks arising from the domain boundaries, successfully achieving nearly full lithium extraction/reinsertion with high power and energy density. This work represents the first direct observation of the presence of domain boundaries within a crystalline structure playing a critical role in governing the lithium diffusivity in a battery electrode.