Understanding X-ray absorption spectra by means of descriptors and machine learning algorithms

Abstract

X-ray absorption near-edge structure (XANES) spectra are the fingerprint of the local atomic and electronic structures around the absorbing atom. However, the quantitative analysis of these spectra is not straightforward. Even with the most recent advances in this area, for a given spectrum, it is not clear a priori which structural parameters can be refined and how uncertainties should be estimated. Here, we present an alternative concept for the analysis of XANES spectra, which is based on machine learning algorithms and establishes the relationship between intuitive descriptors of spectra, such as edge position, intensities, positions, and curvatures of minima and maxima on the one hand, and those related to the local atomic and electronic structure which are the coordination numbers, bond distances and angles and oxidation state on the other hand. This approach overcoms the problem of the systematic difference between theoretical and experimental spectra. Furthermore, the numerical relations can be expressed in analytical formulas providing a simple and fast tool to extract structural parameters based on the spectral shape. The methodology was successfully applied to experimental data for the multicomponent Fe:SiO2 system and reference iron compounds, demonstrating the high prediction quality for both the theoretical validation sets and experimental data.

Synergistically Enhanced Single-Atom Nickel Catalysis for Alkaline Hydrogen Evolution Reaction

“…Specifically, the Ni–N 4 sample is characterized by a significant shoulder peak A 1 and energy splitting of the white-line peaks ( A 2 , A 3 ), indicating a D 4 h symmetry comparable to that of Ni­(II)­Pc. In contrast, the near-absence of the shoulder peak A 1 as well as the prominent white-line peak A 2 of Ni–N 2+2 –S act as fingerprints for the pseudooctahedral structural configuration of Ni atoms in the sample . Furthermore, close inspection reveals noticeable differences regarding the broad feature A 4 , which relates to the scattering processes within the first coordination shell of the photoabsorber Ni atom .…”

“…In contrast, the near-absence of the shoulder peak A 1 as well as the prominent white-line peak A 2 of Ni−N 2+2 −S act as fingerprints for the pseudooctahedral structural configuration of Ni atoms in the sample. 28 Furthermore, close inspection reveals noticeable differences regarding the broad feature A 4 , which relates to the scattering processes within the first coordination shell of the photoabsorber Ni atom. 29 According to the semi-empirical Natoli's rule, a lower shift (ΔE ∼ 8 eV) of the pit-energy position of Ni−N 2+2 −S relative to that of Ni− N 4 most probably arises from its elongated average interatomic distance (R).…”

New challenges in oxygen reduction catalysis: a consortium retrospective to inform future research

“…Additional insights into the relevant features in the XAS spectra have been demonstrated as obtainable using polynomial featurization prior to random forest regression and then analyzing model-assigned feature importance. 113 Featurization using local spectral curvature after this work has now been used outside of our consortium 132 as a descriptor to resolve structural properties like coordination number, Fe–O distance, and Fe oxidation state from experimental spectra of silica-supported single-site Fe catalysts. Part of the goal of interpretable AI tools like TRIXS 133 is to build knowledge that is compatible with past human intuition.…”

Informed Chemical Classification of Organophosphorus Compounds via Unsupervised Machine Learning of X-ray Absorption Spectroscopy and X-ray Emission Spectroscopy

“…Here, we apply the framework of Figure to both X-ray absorption spectroscopy (XAS) and X-ray emission spectroscopy (XES). XAS has seen an explosion of ML applications. − XAS is most commonly used in chemistry, biology, and materials science to investigate the element-specific local coordination environment and electronic structure, with applications including energy storage, , catalysis, and photochemical dynamics . XAS, which includes both the X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS), probes the unoccupied electronic states of the excited state of a chosen atomic species.…”