This project involves the use of novel techniques in quantum chemistry to study electronic structure problems at the active sites of iron-sulfur proteins, and of related mixed-metal clusters containing Mo, V, Co or Zn, in addition to iron. The principal goal is to understand the nature of spin-coupling in these systems, and to explore how these couplings influence their oxidation-reduction and catalytic properties. This will be accomplished through systematic variation of geometries in calculations on Heisenberg coupling constants, Mossbauer and electron spin resonance parameters (including g and hyperfine tensors) and electric fields and charge densities. All of the calculations will use the X alpha or local spin density effective potentials. Systems to be studied include: (a) three-iron clusters in both linear and pseudo-cubane geometries that are characteristic of active sites in several ferredoxins and in aconitase; (b) four-iron clusters that can exist in multiple oxidation states, with special attention paid to geometries that might allow intermediate or high-spin ground states; (c) models for four-iron clusters linked to a siroheme (as a model for sulfite reductase) or to citrate (as a model for an intermediate step in aconitase); (d) mixed metal clusters with XFe3 stoichiometries (X=Mo, V, Co, Zn), which are implicated in nitrogenase activity or which form useful test systems for our understanding of the active sites of iron-sulfur proteins in general.