Conformational analysis and nucleophilic addition transition state modeling of bicyclo(m.1.0)alkan-2-ones and bicyclo(m.1.0)alk-3-en-2-ones.

Abstract

Medium and large ring bicyclo (m.1.0) alkan-2-ones and bicyclo (m.1.0) alk-3-en-2-ones are ideal starting materials for stereoselective organic synthesis. Readily available in many ring sizes and in enantiomerically pure form, such carbocyclic skeleta provide entry into numerous natural products. Reactions such as 1,2-addition, α-alkylation, and 1,4-addition have been shown to proceed with high diastereoselectivity due to local conformational anchoring of the cyclopropyl ketone function. In an effort to elucidate the mechanisms of diastereoselectivity and to augment the synthetic utility, computer modeling studies have been performed. The present work began with development of molecular mechanics parameters for the cyclopropyl ketone torsion potential. Cyclopropanes are uniquely composed of sigma bonds containing high p orbital character that are capable of conjugation with α,β pi bonds. Appropriate treatment of cyclopropyl ketone conjugation was derived from ab initio torsion driving studies. The updated force field was then coupled with Monte Carlo conformation searches to explore the three dimensional shapes available to medium and large ring cyclopropyl ketones. Rationale for the observed diastereoselectivity was developed from the starting material conformational preferences, but a more direct probe of the diastereoselectivity of 1,2-asymmetric induction was desired. An empirical force field based on ab initio transition state studies has been developed to describe kinetically controlled nucleophilic additions of methyllithium to cyclopropyl ketones. The developed molecular mechanics model was then applied to transition states constructed from starting material conformations and transition state geometries identified in the ab initio calculations. Computer modeling studies of starting material and transition state conformations are presented, and diastereoselectivity predictions based on the empirical model are compared to experimental results.