My research programs have been in four areas:

1. Metal-thiolate and metal-chalcogenide cluster chemistry.
2. Gas phase inorganic chemistry.
3. Crystal supramolecular chemistry and inorganic supramolecular chemistry.
4. The mechanism of biological nitrogen fixation.

Since formally retiring from the University of New South Wales in 2002 my research has focussed on the enzyme nitrogenase, and its mechanism.

Nitrogenase, which supports all life, breaks one of the strongest bonds in chemistry, N-N, reducing it to NH₃.  The stoichiometry of the catalytic cycle under normal conditions is N₂ + 8e + 8H⁺ → 2NH₃ + H₂.  The cycle requires many steps at the active site: binding of N₂, eight introductions of an electron, eight introductions of a proton, breaking the N-N bond, six N-H bond formations, two dissociations of NH₃, and one formation of H₂, a total of 27!  The catalytic site (FeMo-co) is an Fe₇CMoS₉ cluster, with chelation of Mo by obligatory homocitrate, connected to surrounding protein through one cysteine ligand and one histidine ligand.   

Experimental investigation of the chemical mechanism of nitrogenase has been frustrated by inherent difficulties in trapping and structurally characterising intermediates under physiological conditions.  Computational investigations, unrestricted by these difficulties, have explored mechanistic hypotheses.  In 2008 I published a detailed 21 step chemical mechanism for the enzymatic conversion of N₂ to NH₃, together with predictions of the importance of H-atom tunneling in nitrogenase reactions.  My review ‘Computational Investigations of the Chemical Mechanism of the Enzyme Nitrogenase’ [ChemBioChem, 21, 1671-1709 (2020) doi.org/10.1002/cbic.201900636] published in 2020 is a critical account of the current state of computational research on this enzyme.

My computational research uses the density functional software DMol, implemented at the Australian National Computational facility on the Gadi supercomputer (the 24th most powerful computer in the world).  This allows me to routinely used metalloprotein models with up to 500 atoms included in the quantum mechanical calculations.  I have developed methodologies for incorporating the complex electronic structure of FeMo-co in simulations of reaction profiles.

The principal objective is to develop an understanding of the fundamental coordination chemistry of the unprecedented (and still synthetically unavailable) metal sulfide cluster that is the catalytic site.  This underlies the goal to develop and evaluate mechanistic hypotheses for the catalytic reduction of N₂, including explanations for the substantial body of biochemical reactivity data that have been accumulated over many years for wild-type and mutant nitrogenase proteins.  Another component of this research is evaluation and explanation of some recent experimental observations of unusual chemical reactivity at the active site.