cis-2,5-Diaminobicyclo[2.2.2]octane, a New Chiral Scaffold for Asymmetric Catalysis

Catalysis of widely used chemical transformations in which the goal is to obtain the product as a pure enantiomer has become a major preoccupation of synthetic organic chemistry over the past three decades. A large number of chiral entities has been deployed to this end, many with considerable success, but one of the simplest and most effective catalytic systems to have emerged from this effort is that based on a chiral diamine, specifically trans-1,2-diaminocyclohexane. While there have been attempts to improve upon this scaffold in asymmetric synthesis, few have gained the recognition needed to take their place alongside this classic diamine. The challenge is to design a scaffold that retains the assets of trans-1,2-diaminocyclohexane while enhancing its intrinsic chirality and maximizing the scope of its applications. It occurred to us that cis-2,5-diaminobicyclo[2.2.2]octane could be such a scaffold. Synthesis of this diamine in enantiopure form was completed from benzoic acid, and the (1R,2R,4R,5R) enantiomer was used in all subsequent experiments in this laboratory. Condensation of the diamine with various salicyl aldehydes generated imine derivatives which proved to be excellent "salen" ligands for encapsulation of transition and other metals. In total, 12 salen-metal complexes were prepared from this ligand, many of which were crystalline and three of which, along with the ligand itself, yielded to X-ray crystallography. An advantage of this ligand is that it can be tuned sterically or electronically to confer specific catalytic properties on the salen-metal complex, and this feature was used in several applications of our salen-metal complexes in asymmetric synthesis. Thus, replacement of one of the tert-butyl groups in each benzenoid ring of the salen ligand by a methoxy substituent enhanced the catalytic efficiency of a cobalt(II)-salen complex used in asymmetric cyclopropanation of 1,1-disubstituted alkenes; the catalyst was employed in an improved synthesis of the cyclopropane-containing drug candidate Synosutine. Reduction of the pair of imine functions of the ligand to secondary amines permitted formation of a copper(I)-salen complex that catalyzed asymmetric Henry ("nitroaldol") condensation with excellent efficiency; this catalyst was applied in an economical synthesis of three drugs of the "beta-blocker" family including (S)-Propanolol. Chromium(II) and chromium(III) complexes were prepared from our bicyclooctane-salen ligand bearing a pair of tert-butyl groups in each benzenoid ring. These complexes were found to catalyze, respectively, enantioselective formation of homoallylic alcohols from Nozaki-Hiyama-Kishi allylation of aromatic aldehydes and dihydropyranones from hetero-Diels-Alder cycloaddition. Plausible reaction models emerging from knowledge of the absolute configuration of products from each of these reactions place the metal-coordinated substrate in a quadrant beneath the bicyclooctane scaffold so that one face of the substrate is blocked by an aryl ring of the salen ligand while the opposite face is left open to attack. The consistent and predictable stereochemical outcome from reactions catalyzed by salen-metal complexes derived from our diaminobicyclo[2.2.2]octane scaffold adds a valuable new dimension to asymmetric synthesis.