Combinatorial chemistry has recently burst on the scene as a valuable tool for the discovery of new drug candidates. The ability to synthesize hundreds of compounds for screening is a useful complement to rational drug design. There are many similarities between the design of new therapeutic agents and the development of new asymmetric ligands, the most important of which is the limitation of a rational design strategy. The proposed research aims to develop a combinatorial approach to two classes of ligands that have proven to be useful in the past, bisphosphines and phosphine-oxazoline ligands. The goals of the proposal are direct, the development of ligands for a given set of reactions and indirect, develop methods for the utilization of a parallel approach to the discovery of phosphine and phosphine- oxazoline ligands in general. Amino acid building blocks that will allow for the synthesis of both types of ligands by known combinatorial technology will be discussed. Preliminary results will be presented that show that, libraries of ligands can be synthesized and that a structural motif based on a known beta- turn forming sequence is a selective ligand for asymmetric pi- allyl additions. A general method for the synthesis of libraries of turn derived bisphosphine and phosphine-oxazoline ligands on solid supports will be proposed. A plan for the use of other turn forming sequences will be presented. Work on the continuation of the development of proline based phosphine- oxazoline ligands is included. Initially three reactions will be screened, palladium catalyzed pi-allyl alkylation, the Heck reaction and a rhodium catalyzed [4+2] cycloisomerization reaction. These reactions were chosen because of their differing mechanisms and the fact that all are potentially useful carbon- carbon bond forming reactions.
