Ligand and substrate effects on the mechanism of rhodium-catalyzed hydrogenation of enamides.

The rhodium-catalyzed hydrogenation reaction of enamides is studied computationally using the B3LYP/LACVP** level of theory for a range of ligands and substrates. Two model bidentate phosphine ligands, 1,2-bis(dimethylphosphino)ethane (DMPE) and (Z)-1,2-bis(dimethylphosphino) ethene (ZDMP), and two chiral bidentate phosphine ligands, (R,R)-MeDuPHOS and (R,R)-tetramethylbisoxaphospinane (TMBOP), are investigated in the hydrogenation of alpha-formamidoacrylonitrile as a model substrate. The ZDMP ligand is then studied for three additional substrates: N-(2-propenyl)formamide, (Z)-3-formamido-2-butenenitrile, and (E)-3-formamido-2-butenenitrile. The potential-energy surfaces calculated for the four ligands and alpha-formamidoacrylonitrile are in general agreement with previous computational studies using QM/MM (ONIOM) methods but show consistently higher relative barriers rather than lower. The calculated potential-energy surfaces of hydrogenations of various substrates with a common ligand indicate a mechanistic change based on substrate. The sequence of hydrogen transfer to the two olefinic carbons is calculated to change based on substrate electronics. This has a significant impact on the origins of enantioselectivity for such varied substrates as the first hydride transfer to the substrate is calculated to be irreversible for all substrates, independent of whether it occurs at the alpha or beta carbon of the olefin.