Efficient chemical state imaging of Ni90 cathodes by in situ TXM-XANES white-line peak fitting.

Abstract

Transmission X-ray microscopy combined with X-ray absorption near-edge structure (TXM-XANES) spectroscopy enables nanoscale chemical state imaging of transition metal oxides, yet the low signal-to-noise ratio (SNR) of single-pixel spectra undermines the reliability of conventional edge-energy (E₀) quantification. Herein, we present an analytical framework based on white-line peak fitting that integrates skewed Gaussian fitting for precise peak extraction, area normalization for spectral standardization, and average spectrum analysis for chemical state referencing. Compared with the E₀ method, our approach reduces the full width at half maximum (FWHM) of the energy distribution to one-third (0.28 eV versus 0.83 eV) and achieves an average coefficient of determination (R²) of 0.997 across the entire particle, confirming superior accuracy under low-SNR conditions. Using a high-nickel Ni90 cathode as a model system, we establish a linear correlation between white-line peak energy and cell voltage. In battery cycling, 45 nm chemical-state mapping reveals interparticle reaction heterogeneity, improved uniformity after the first activation cycle, and a non-radial reaction pattern with spatially random heterogeneity originating from intragranular defects rather than Li⁺ diffusion limitations. These observations are consistent with a kinetically trapped state model in which some regions remain inactive during the first charge, gradually overcome the activation barrier, and homogenize, though the activation is not fully reversible. This strategy enhances the quantitative fidelity and throughput of nanoscale chemical-state imaging and offers an efficient analytical route for operando studies of battery materials and heterogeneous systems.