UC Irvine

Allylic amides, amines, and esters are key synthetic building blocks. Their enantioselective syntheses under mild conditions is a continuing pursuit of organic synthesis methods development. One opportunity for the synthesis of these building blocks is by functionalization of prochiral double bonds using palladium(II) catalysis. In these reactions, nucleopalladation mediated by a chiral palladium(II) catalyst generates a new heteroatom-substituted chiral center. However, reactions where nucleopalladation occurs with antarafacial stereoselectivity are difficult to render enantioselective because of the challenge of transferring chiral ligand information across the square-planar palladium complex to the incoming nucleophile. In this Account, we describe the development and use of enantiopure palladium(II) catalysts of the COP (chiral cobalt oxazoline palladacyclic) family for the synthesis of enantioenriched products from starting materials derived from prochiral allylic alcohols. We begin with initial studies aimed at rendering catalyzed [3,3]-sigmatropic rearrangements of allylic imidates enantioselective, which ultimately led to the identification of the significant utility of the COP family of Pd(II) catalysts. The first use of an enantioselective COP catalyst was reported by Richards' and our laboratories in 2003 for the enantioselective rearrangement of allylic N-arylimidates. Shortly thereafter, we discovered that the chloride-bridged COP dimer, [COP-Cl]₂, was an excellent enantioselective catalyst for the rearrangement of (E)-allylic trichloroacetimidates to enantioenriched allylic trichloroacetamides, this conversion being the most widely used of the allylic imidate rearrangements. We then turn to discuss S_N2' reactions catalyzed by the acetate-bridged COP dimer, [COP-OAc]₂, which proceed by a unique mechanism to provide branched allylic esters and allylic phenyl ethers in high enantioselectivity. Furthermore, because of the unique nucleopalladation/deoxypalladation mechanism of these S_N2' reactions, they provide exclusively the branched allylic product. Importantly, both enantiomers of the [COP-Cl]₂ and [COP-OAc]₂ catalysts are commercially available. We also briefly consider several other enantioselective reactions catalyzed by COP complexes. The mechanism of enantioselective COP-catalyzed allylic rearrangements and allylic substitutions is discussed in some detail. In both reactions, nucleopalladation is found to be the enantiodetermining step. The cyclobutadienyl "floor" of the COP catalyst is critical for transmitting chiral information across the palladium square plane in these reactions. This structural feature enables high enantioselection to be realized in spite of the nearly 180° angle between the catalyst, electrophile and nucleophile in the enantiodetermining step. Our discussion concludes by considering several uses of the COP family of catalysts by other researchers for the enantioselective synthesis of biologically active chiral molecules. We anticipate that additional uses for COP catalysts will emerge in the future. In addition, the structural features of these catalysts that we have identified as important for achieving high enantioselection should be useful in the future development of improved enantioselective Pd(II) catalysts.