Diagrammatic multiplet sum method (MSM) density functional theory (DFT): Investigation of the transferability of integrals in "simple" DFT-based approaches to multideterminantal problems.

Abstract

Kohn-Sham density functional theory (DFT) typically works well for describing dynamic correlation. Two other types of correlation, arising in the cases of degenerate (static) or quasidegenerate (nondynamic) zero-order states, represent a difficult problem for DFT. When symmetry is present, multiplet sum method (MSM) DFT [Ziegler et al., Theor. Chim. Acta 4, 877 (1977)] provides one of the earliest and simplest ways to include static correlation in DFT. MSM-DFT assumes that DFT provides a good description of single-determinant energies and uses symmetry and simple ansätze to include the effects of static correlation. This is equivalent to determining the off-diagonal matrix elements in a small configuration interaction (CI) eigenvalue problem. Our ultimate goal, however, is nondynamic correlation in cases where symmetry is inadequate for fixing the dynamic-correlation limitation of DFT. To this end, we have developed a diagrammatic approach to MSM-DFT, which does not, by itself, solve the nondynamic correlation problem in DFT but which facilitates comparison with wave function CI and so allows educated guesses of off-diagonal CI matrix elements even in the absence of symmetry. In every case, an additional exchange-only ansatz (EXAN) allows the MSM-DFT formulas to be transformed into wave function formulas. This EXAN also works for transforming time-dependent DFT into time-dependent Hartree-Fock. Although not enough to uniquely guess DFT formulas from wave function formulas, the diagrammatic approach and the EXAN provide important constraints on any guesses that might be used. We illustrate how diagrammatic MSM-DFT may be used to guess a nondynamic correlation correction for the dissociation of H2 and how diagrammatic MSM-DFT may be used to guess a nonsymmetry-based coupling element in the O2 multiplet problem, which is reasonably close to a previous symmetry-derived result.