Transition-Potential Coupled Cluster II: Optimization of the Core Orbital Occupation Number

TL;DR

This paper improves core-hole spectroscopy calculations by optimizing fractional core occupation numbers in transition-potential coupled cluster methods, leading to better spectral predictions for biological molecules.

Contribution

It introduces an element-specific optimization of the fractional occupation parameter in TP-CCSD, enhancing core-hole spectral accuracy.

Findings

TP-CCSD reproduces valence spectra well

Requires smaller energy shifts than EOM-CCSD

Improves peak positions and intensities in absorption spectra

Abstract

The issue of orbital relaxation in computational core-hole spectroscopy, specifically x-ray absorption, has been a major problem for methods such as equation-of-motion coupled cluster with singles and doubles (EOM-CCSD). The transition-potential coupled cluster (TP-CC) method is utilized to address this problem by including an explicit treatment of orbital relaxation via the use of reference orbitals with a fractional core occupation number. The value of the fractional occupation parameter $\lambda$ was optimized for both TP-CCSD and XTP-CCSD methods in an element-specific manner due to the differences in atomic charge and energy scale. Additionally, TP-CCSD calculations using the optimized parameters were performed for the K-edge absorption spectra of gas-phase adenine and thymine. TP-CCSD reproduces the valence region well and requires smaller overall energy shifts in comparison to EOM-CCSD, while also improving on the relative position and intensities of several absorption peaks.