An atomistic interpretation of the oxygen K-edge X-ray absorption spectra of layered Li-ion battery cathode materials

Core loss spectroscopies can provide powerful element-specific insight into the redox processes occurring in Li-ion battery cathodes, but this requires accurate interpretation of the spectral features. Here, we systematically interpret oxygen K-edge core loss spectra of layered lithium transition-metal (TM) oxides (LiMO2 where M=Co, Ni ,Mn) from first principles using density-functional theory (DFT). Spectra are simulated using three exchange-correlation functionals, comprising the GGA functional PBE, the DFT-PBE + Hubbard U method and the meta-GGA functional rSCAN. In general, rSCAN provides a better match to experimentally observed excitation energies of spectral features compared to both PBE and PBE+U, especially at energies close to the main edge. Projected density of states of core-hole calculations show that the O orbitals are better described by rSCAN. Hybridisation, structural distortions, chemical composition, and magnetism significantly influence the spectra. The O K-edge spectrum of LiNiO2 obtained using rSCAN shows a closer match to the experimental XAS when derived from a simulation cell which includes a Jahn-Teller distortion, showing that the DFT-calculated pre-edge feature contains information about not only chemical species but also geometric distortion. Core loss spectra derived from DFT can also differentiate between materials with the same structure and magnetic configuration but comprising different TMs; these differences are comparable to those observed in experimental XAS from the same materials. This foundational work helps establish the extent to which DFT can be used to bridge the interpretation gap between experimental spectroscopic signatures and ab initio methods describing complex battery materials, such as lithium nickel manganese cobalt oxides.