Theoretical modeling of metal- and enzyme catalyzed transformations

Abstract: This thesis is focused on describing and predicting catalytic reactions. The major part of the work is based on density functional theory (DFT). In some cases where the size of the investigated system precluded the use of more accurate methods molecular dynamics was employed. In several cases the proposed mechanism was later tested in the laboratory. A few examples where the predictions were confirmed are:The formation of an acyl intermediate in the activation of a ruthenium catalyst used for racemizing alcohols. This intermediate was observed by both NMR and in situ FT-IR.The improvement of the substrate specificity and catalytic activity of Candida antarctica lipase A by modifying amino acids close to the active site. The improved specificity of Candida antarctica lipase B toward ?-substituted secondary alcohols by an enzyme variant where the alanine in position 281 was exchanged for a serine.In other cases experimental results were complemented with a theoretical investigation, for example:The observed second order rate constant for a ruthenium based catalyst used for water oxidation was explained and a novel intramolecular mechanism based on a high valent ruthenium dimer was suggested.The effects of electron withdrawing/donating axial ligands on the performance of ruthenium catalyzed water oxidation were addressed.Mechanisms of H2 activation by Lewis acid/Lewis base adducts were rationalized. One example of the predictive power of computational chemistry is the mechanism of hydrogen uptake by phosphanylboranes; the potential energy barrier for the transition state could be predicted within a few kcal/mol based on the orbital energies of the starting material.