XPS on Li-Battery-Related Compounds: Analysis of Inorganic SEI Phases and a Methodology for Charge Correction

Accurate identification of chemical phases associated with the electrode and solid–electrolyte interphase (SEI) is critical for understanding and controlling interfacial degradation mechanisms in lithium-containing battery systems. To study these critical battery materials and interfaces X-ray photoelectron spectroscopy (XPS) is a widely used technique that provides quantitative chemical insights. However, due to the fact that a majority of chemical phases relevant to battery interfaces are poor electronic conductors, phase identification that relies primarily on absolute XPS core-level binding-energies (BEs) can be problematic. Charging during XPS measurements leads to BE shifts that can be difficult to correct. These difficulties are often exacerbated by the coexistence of multiple Li-containing phases in the SEI with overlapping XPS core levels. To facilitate accurate phase identification of battery-relevant phases (and electronically insulating phases in general), we propose a straightforward approach for removing charging effects from XPS data sets. We apply this approach to XPS data sets acquired from six battery-relevant inorganic phases including lithium metal (Li0), lithium oxide (Li2O), lithium peroxide (Li2O2), lithium hydroxide (LiOH), lithium carbonate (Li2CO3), and lithium nitride (Li3N). Specifically, we demonstrate that BE separations between core levels present in a particular phase (e.g., BE separation between the O 1s and Li 1s core levels in Li2O) provide an additional constraint that can significantly improve reliability of phase identification. For phases like Li2O2 and LiOH where the Li-to-O ratios and BE separations are nearly identical, X-ray excited valence-band spectra can provide additional clues that facilitate accurate phase identification. To accurately identity the phases in this study, we show that in situ growth of Li2O on Li0 provides a means for determining absolute core-level positions, where all charging effects can be accurately removed. Finally, as an exemplary case we apply the charge-correction methodology to XPS data acquired from a symmetric cell based on a Li2S–P2S5 solid electrolyte. This analysis demonstrates that accurately accounting for XPS BE shifts as a function of current-bias conditions can provide a direct probe of ionic conductivities associated with battery materials.