Excitation energies from G{\"o}rling-Levy perturbation theory along the range-separated adiabatic connection
Elisa Rebolini (ILL), Andrew M. Teale, Trygve Helgaker, Andreas Savin, (LCT), Julien Toulouse (LCT)

TL;DR
This paper evaluates a G{"o}rling-Levy perturbation theory along the range-separated adiabatic connection for calculating electronic excitation energies, showing improved accuracy for Rydberg states over previous methods.
Contribution
It introduces and assesses a G{"o}rling-Levy based perturbation approach that maintains constant ground-state density, improving excitation energy predictions for Rydberg states.
Findings
More accurate Rydberg state excitation energies with GL-based perturbation.
Similar valence state excitation energies compared to RS-based perturbation.
No density-functional approximations used in calculations.
Abstract
A G{\"o}rling-Levy (GL)-based perturbation theory along the range-separated adiabatic connection is assessed for the calculation of electronic excitation energies. In comparison with the Rayleigh-Schr{\"o}dinger (RS)-based perturbation theory introduced in a previous work [E. Rebolini, J. Toulouse, A. M. Teale, T. Helgaker, A. Savin, Mol. Phys. 113, 1740 (2015)], this GL-based perturbation theory keeps the ground-state density constant at each order and thus gives the correct ionization energy at each order. Excitation energies up to first order in the perturbation have been calculated numerically for the helium and beryllium atoms and the hydrogen molecule without introducing any density-functional approximations. In comparison with the RS-based perturbation theory, the present GL-based perturbation theory gives much more accurate excitation energies for Rydberg states but similar…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
