Abstract
The unique reactivity profile of the dinuclear Pd-I complex [Pd-I(mu-Br)(Bu3P)-Bu-t](2) as an isomerization cocatalyst has enabled orthogonal tandem processes ranging from styrene syntheses to biodiesel refining. We have now elucidated the mechanistic basis of its distinct catalytic profile by density functional theory calculations and experimental studies. Activation of the catalyst proceeds intramolecularly, giving rise to a dinuclear complex composed of a reactive palladium hydride and an inert palladacycle. This complex mediates double bond migrations with an energy span of 9.5 kcal/mol, which is well below those calculated for known catalysts. Its dissociation leads to an even more active monophosphinopalladium hydride catalyst and an inert dinuclear bispalladacycle. In the main deactivation pathway, two mononuclear Pd species react with each other, liberating a hydrogenation product and regenerating the catalyst precursor [Pd](I)(mu-Br)(Bu3P)-Bu-t](2). The experimentally observed buildup of dinuclear palladacycles during the catalysis is, thus, the result of conversion of a binuclear into mononuclear Pd-H catalyst. Phosphines, which would deactivate metathesis cocatalysts, are not liberated at any stage. This explains the unique suitability of [Pd](I)(mu-Br)(Bu3P)-Bu-t](2) for isomerizing metatheses. The mechanistic insights were used for the in silico casting of a catalyst generation, targeting complexes with a reduced barrier toward the formation of dinuclear Pd-H species, a low energy span of the catalytic cycles, and increased barriers either toward deactivation or, alternatively, toward dissociation to short-lived mononuclear complexes. Complexes with bisadamantyl-n-butylphosphine ligands were identified as lead structures. Experimental studies with model catalysts confirmed the validity of the predicted structure-activity relationship.