Asymmetric cyclopentannulation reactions : scope and limitation/

The first part of this dissertation is devoted to the study of an asymmetric [3+2] cycloaddition sequence developed in our laboratory. The cycloaddition sequence used a sulfonamide-based homoenolate equivalent which was cyclocondensed with a cyclic enone. The stereochemistry of the final product was fixed during the first step, the Michael addition to the enone. Our study focused thus on the Michael addition of sulfonamides to enones. We have synthesized a series of chiral and achiral sulfonamides. We then studied the regiochemistry and the stereochemistry of the addition of the anions derived from these sulfonamides to cyclohexenone. The presence of a heteroatom at the (gamma)-carbon of the sulfonamide was crucial for the regiochemical outcome of the reaction. The substituent on the sulfonamide also influenced the facial selectivity of the reaction with chiral sulfonamides, but had no influence on the diastereoselectivity with achiral sulfonamides. The sequence had been applied to various cyclenones. We were also able to apply the method to an acyclic enone with good enantioselectivities. However the method could not be applied to other Michael acceptors. Another part of the present work was devoted to search a new catalytic asymmetric cyclopentannulation sequence. Two different approaches were studied. Phase-transfer catalysis seemed the most appropriate strategy for our objective. We soon realized that sulfur-stabilized nucleophiles could not be used under these conditions. Ketals derived from 3-nitropropanal were thus chosen as potential annulation agents. The racemic version was quite efficient and could be applied to less active acceptors such as unsaturated lactams and lactones. Unfortunately we were not able to realize the reaction with good enantioselectivities. However two new catalysts, obtained by the reaction of cinchonine with 1- and 2-methylnaphthyl chloride, emerged as interesting candidates for the phase-transfer reactions. Organocatalysis was our second approach. The use of rubidium prolinate or the use of proline in the presence of a base proved to be very efficient and the Michael adducts could be obtained with good enantioselectivities. The Michael adducts were easily cyclized and the nitro group could be removed under mild conditions.(CHIM 3)--UCL, 200