Robert B. Gagosian

Mechanistic photochemistry of 1-adamantylacetone

Irradiation of I-adamantylacetone (1) results in essentially a quantitative yield of two cyclobutanols, 2 and 3. Quantum yields for the formation of 2 and 3 from the SI and TI states of 1 in benzene and methanol are reported. Formation of cyclobutanols from the excited triplet state of 1 is much more efficient in methanol than in benzene. The cyclobutanol product ratio, 2:3, is 1.0-1.8 for the T I reaction and approximately 5 for the SI reaction. The observed stereoselectivity from the excited singlet state reaction is consistent with the intermediacy of a short-lived 1,4 biradical which undergoes rehybridization, resulting in a preference for rotation and closure to yield 2 rather than 3. The reactivity of the SI and T I states of 1 toward intramolecular y-hydrogen abstraction is determined by fluorescence measurements and Stern-Volmer treatments of 1.3-pentadiene quenching of cyclobutanol formation, respectively. We2 and others3 have recently reported that irradiation of 1 -adamantylacetone (1) results in nearly quantitative formation of two cyclobutanols, 2 and 3 (eq 1). Neither of

Favorskii rearrangement of some α-bromo-ketones

THE base-induced rearrangement of a-halogenoketones to give carboxylic acid derivatives (Favorskii rearrangements1) occurs by a t least two mechanisms. The stereochemi~try29~ and structure* of many Favorskii products can be explained in terms of a cyclopropanone intermediate. However, a “benzylic-like” rearrangement appears to operate when the substrate ketone bears no a’-hydrogen,5 when the =’-hydrogen is relatively non-acidic,6 or when steric or strain factors inhibit cyclopropanone formation.’ In addition, the formation of a dipolar intermediate which may precede (or follow) cyclopropanone formation is supported by theoretical considerations,s loss of stereospecificity upon rearrangement in polar solvents,B the formation of a-substitution products* as a side reaction, and the trapping of a possible Favorskii intermediate .lo “dehydrohalogenate” to form a common intermediate leading to the same product.lsu We report here (Table) our results with the isomeric pairs of a-bromobutan-Z-ones and a-bromo-3-methylbutan-2-ones which provide strong evidence against (a) a “benzylic-like” rearrangement, (b) a bromine atom exchange from the aand a’-position, and (c) the occurrence of a dipolar ion prior to rearrangement. Comparison of these results with the cleavage of cyclopropanones (or the corresponding hemiacetals) provides strong evidence (a) for the intervention of cyclopropanones [e.g. (I)] in the Favorskii rearrangement of the acyclic a-halogenoketones studied and (b) for the occurrence of side products from base attack on the a-halogenoketone carbonyl group, not from attack on a dipolar species such as (111).

The Mechanism of Addition of Diazo-alkanes to Cyclopropanones

The addition of diazo-alkanes to cyclopropanones yields cyclobutanones whose structures may be predicted on the basis of conformational regulation of the migrating carbon–carbon bond.