Answers are grounded in this paper and cite their sources.
An analog of Koopmans’ theorem is formulated for the energies, ε a , of virtual Kohn–Sham (KS) molecular orbitals (MOs) from the requirement that the KS theory provides, in principle, not only the exact electron density, but also its exact response. The starting point is the Kohn–Sham analog of Koopmans’ theorem, relating the vertical ionization energies, I i , to the energies, ε i , of the occupied MOs ( Chong, D.P.; Gritsenko, O.V.; Baerends, E.J. J. Chem. Phys. 2002, 116, 1760 ). Combining this with the coupled-perturbed equations of time-dependent density functional theory (TDDFT), exact relations between the energies, ε a , of virtual KS MOs and the excitation energies, ω ia , and vertical ionization energies (VIPs), I i , are obtained. In the small matrix approximation for the coupling matrix K of TDDFT, two limiting cases of these relations are considered. In the limit of a negligible matrix element, K ia,ia , the energy, ε a , can be interpreted as (minus) the energy of ionization from the ? i → ? a excited state, ε a ≈ –I a , where –I a is defined from the relation I i = ω ia + I a . This relation breaks down in special cases, such as charge-transfer transitions and the HOMO–LUMO (highest occupied molecular orbital – lowest unoccupied molecular orbital) transition of a dissociating electron-pair bond (also of charge-transfer character). The present results highlight the important difference between virtual orbital energies in the Kohn–Sham model (ε a ≈ –I a ) and in the Hartree–Fock model (ε a ≈ –A a ). Kohn–Sham differences ε a – ε i approximate the excitation energy, ω ia , while Hartree–Fock differences [Formula: see text] do not approximate excitation energies but approximate the difference of an ionization energy and an electron affinity, I i – A a .
No takes yet. Share an insight, caveat, or question.
Gritsenko et al. (2009) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: