Spectroscopy of core-level electrons is a useful tool for characterization of local electronic structure in chemical systems. However, linear response excited state methods (like TDDFT or EOM-CCSD) are challenged by such excitations, due to the lack of explicit core-hole relaxation, leading to large errors in computed spectra vs experiment. On the other hand, orbital optimized (OO) methods can be much more accurate, but are susceptible to "variational collapse" to lower energy states due to difficulty of converging to saddle points of energy in orbital space. We present an orbital optimization scheme based on square gradient minimization (SGM) that reliably converges to excited state solutions. The computational cost of SGM is only 2-3 times the cost of ground state orbital optimization (per iteration). We also present routes to recouple spin-contaminated single determinant Kohn-Sham energies to obtain approximately spin-pure energies corresponding to multireference excited states. These techniques are applied to study the core excitation energies of both closed and open-shell systems. We demonstrate that OO-DFT approaches reliably predict core excitation energies to &lt; 0.5 eV RMS error for both classes of systems, including cases where TDDFT/EOM-CCSD are qualitatively challenged.

Accurate computation of core-level spectra with orbital-optimized density functional theory

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.