Is the ground state of [RuO4]2– exceptional?

Abstract

Optimised geometries of [RuO₄]^n–(n= 0, 1 or 2) have been obtained. While the local density approximation (LDA) of density functional theory (DFT) gives good Ru–O distances for RuO₄ and [RuO₄]^–, the computed bond length for [RuO₄]^2– is at least 0.07 Å too long. The latter result assumes a conventional ³A₁ electronic structure with the two notional d electrons in the essentially antibonding e molecular orbitals. However, a ³E state, derived from the e¹a₁¹ configuration, is very close in energy and gives a far more satisfactory Ru–O distance although theory does not categorically establish ³E as the outright ground state even after the inclusion of spin-polarisation, lattice, relativistic, Jahn–Teller and non-local functional effects. If the real ground state is ³E then the DFT state energies for [RuO₄]^2– would appear to be in error by about 10 kcal mol^–1. It seems more plausible to accept this error in total energy than the large bond length error associated with the ³A₁ state and therefore to conclude that the ground state for tetrahedral [RuO₄]^2– is ³E.