Can Multiconfigurational Self-Consistent Field Theory and Density Functional Theory Correctly Predict the Ground State of Metal-Metal-Bonded Complexes?

J Chem Theory Comput. 2015 Sep 8;11(9):4093-101. doi: 10.1021/acs.jctc.5b00412. Epub 2015 Aug 10.

Abstract

The electronic structure of a diiron (FeFe) complex with strong metal-metal interaction and those of analogous complexes (CoCo, CoMn, CoFe, and FeMn) with much weaker metal-metal bonding are investigated with wave function-based methods and density functional theory. The delocalization and bonding between the metal centers in the diiron complex is only fully captured after inclusion of the complete set of 3d and 4d orbitals in the active space, a situation best suited for restricted active space (RAS) approaches. Truncation of the included set of 4d orbitals results in inappropriate localization of some 3d orbitals, incorrect description of the ground spin state as well as wrong spin state energetics, as compared to experiment. Using density functional theory, some local functionals are able to predict the correct ground spin states, and describe the chemical bonding and structural properties of all the metal-metal complexes considered in this work. In contrast, the introduction of some exact exchange results in increased localization of 3d orbitals and wrong spin state energetics, a situation that is particularly troublesome for the diiron complex.