Charles A. Cardot | Searcharxiv

arXiv2026

Machine Learning to Predict Spectral Anisotropy in Valence-to-Core X-ray Emission Spectroscopy

Polarization analysis in x-ray spectroscopy provides an orientation dependent sensitivity to local bonding environments. For a cluster of atoms, polarization sensitivity is most often discussed through the lens of point group symmetries. However, this is a discrete, qualitative method of classifying clusters, and it does little to indicate the degree of spectral anisotropy. Here we adopt a random forest model for quantitative prediction of spectral anisotropy. Its input relies on simplified local geometric and chemical information that can be obtained from any crystal structure file. The model is trained on over 10,000 experimentally realized transition metal structures from the Materials Project, with the target being VtC-XES calculated using the real space Green's function code FEFF. We find that the model can strongly predict the degree of spectral anisotropy, with the primary factors being derived from the spatial moments of ligands in a cluster.

cond-mat.str-el↗

arXiv2025

X-ray Emission Spectropolarimetry of Strongly Anisotropic Single Crystal Systems using a Rowland Circle Geometry

Polarization dependence has historically seen extensive use in x-ray spectroscopy to determine magnetic and local geometric properties, but more broadly as a way to gain extra sensitivity to electronic structure at the level of individual magnetic orbitals. This is often done in the context of x-ray absorption through techniques like x-ray magnetic circular dichroism or x-ray linear dichroism, but it has seen little application to x-ray emission. Here we explore the information contained in the polarized emission of two 3d transition metal systems across both core-to-core (CtC) and valence-to-core emission (VtC) lines. We demonstrate how the Rowland circle geometry can be used as a spectropolarimeter, and apply it to the x-ray emission spectroscopy of spin-1/2 Cu(II) and spin-0 Ni(II) ions in LiVCuO4 and DyNiC2, respectively. From this we explore how the polarized XES interrogates of the occupied density of states at the valence level, either as a second order effect through Coulomb exchange (CtC x-ray emission) or by direct transitions (VtC x-ray emission). We find that the polarized x-ray emission can provide insights into the valence electron orbital occupation, in much the same way that is achievable with polarized absorption or angle-resolved photoemission spectroscopy techniques. Finally, we highlight how the individually polarized dipole emission spectra can be extracted from a linearly independent suite of directed emission spectra, allowing for polarized measurements at high Bragg angle with lower experimental broadening.

cond-mat.str-el↗