Direct Evidence of Metal-Ligand Redox in Li-ion Battery Positive Electrodes

Abstract

Describing Li-ion battery positive electrodes in terms of distinct transition metal or oxygen redox regimes can lead to confusion in understanding metal-ligand hybridisation, oxygen dimerisation, and degradation. There is a pressing need to study the electronic structure of these materials and determine the role each cation and anion plays in charge compensation. Here, we employ transition metal L-edge X-ray Resonance Photoemission Spectroscopy in conjunction with Single Impurity Anderson models, Self-consistent Real Space Multiple Scattering spectral simulations, and Dynamical Mean-Field theory calculations to directly evaluate the redox mechanisms in (de-)lithiated battery electrodes. This approach reconciles the redox description of two canonical cathodes -- LiMn$_{0.6}$Fe$_{0.4}$PO$_{4}$ and LiNiO$_{2}$ -- in terms of varying degrees of charge transfer using the established Zaanen-Sawatzky-Allen framework, common to condensed matter physics. In LiMn$_{0.6}$Fe$_{0.4}$PO$_{4}$, the absence of charge transfer means capacity arises due to the depopulation of metal $\textit{3d}$ states, i.e. conventional metal redox. Whereas, in LiNiO$_{2}$, charge transfer dominates and redox occurs through the formation and elimination of ligand hole states. This work clarifies the role of oxygen in Ni-rich system and provides a framework to explain how capacity can be extracted from oxygen-dominated states in highly covalent systems without needing to invoke dimerisation.