Hartree—fock—slater-LCAO calculations on [Fe 4S 4(SH) 4] 0, 2−,3−: a model for the 4-Fe active site in high potential iron protein and ferredoxin

Abstract

The electronic structure of the complexes [Fe 4S 4(SH) 4] 0, 2−, 3− which model the 4-Fe active state site in high potential iron protein and ferredoxin, has been calculated with the Hartree—Fock—Slater-LCAO method (in its frozen core and core pseudopotential versions). Results are in agreement with the measured electronic absorption spectrum and magnetic behaviour. The electric field gradient on the Fe nuclei is larger than expected from the observed Mössbauer quadrupole splitting, but the (small) change in this quantity in going the dianion to the trianion is well described. The FeS bonding is mainly covalent and direct FeFe bonding is weak; these conclusions also follow from extended Hückel calculations which we have made in parallel. The self-consistent HFS-LCAO calculations show that in redox reactions the Fe atoms act as charge redistributors. The core pseudopotential version of the method yields results which generally agree with those of the frozen core calculations.