Core-hole delocalization for modeling x-ray spectroscopies: A cautionary tale

Area of Science:

• Computational Chemistry
• Quantum Mechanics
• Spectroscopy

Background:

• Core-hole delocalization is a critical factor in accurately calculating x-ray photoelectron, x-ray absorption, and x-ray emission spectra.
• This phenomenon is particularly relevant for systems with symmetrically equivalent atoms or near-degeneracies, where core orbitals can spread across multiple atoms.

Purpose of the Study:

• To investigate the influence of core-hole delocalization on various x-ray spectroscopy calculations.
• To compare results obtained using localized versus delocalized core-hole models across different theoretical approaches.

Main Methods:

• Utilized response theory, transition-potential methods, and ground state schemes.
• Examined calculations involving explicit core-hole states, including core-excited/ionized reference states and fractional occupation numbers.
• Employed coupled-cluster single-double and perturbative triple (CC(2,3)) methods and density functional theory (DFT).

Main Results:

• Discrepancies of up to 0.5 eV in core-ionization energies were observed between localized and delocalized core-hole models, even at the CC(2,3) level.
• In DFT, the discrepancy correlates with exchange interactions involving core orbitals of the same spin symmetry as the delocalized core-hole.
• Localized core-hole models allow for reasonable relaxation effects at lower computational costs, while delocalized models necessitate higher-level theories for accurate correlation effects.

Conclusions:

• Core-hole delocalization introduces notable errors in x-ray spectroscopy calculations, necessitating careful consideration of theoretical approaches.
• Accurate modeling of delocalized core-holes requires advanced methods to capture electron correlation and orbital relaxation effects.
• When using linear response methods for x-ray absorption spectra, ensuring no core orbital delocalization is crucial to avoid shifts in spectral features.