In epoxidation of the conformationally biassed 5-t-butylcyclohex-2-enols by peroxybenzoic acid, higher cisstereoselectivity and faster rate are found for the pseudo-equatorial than for the pseudo-axial alcohol. Cyclo-oct-2-en-1 -01 is epoxidised with stereospecific introduction of the epoxy-group trans to hydroxy, and appropriate acyclic allylic alcohols show a preference for formation of the threo-epoxy-alcohol. A consistent interpretation of these results is advanced, based on the postulate that the preferred conformation of an allylic alcohol in the transition state for epoxidation is as in ( X x ) .IN 1957 Henbest and Wilson 1 reported their work on the epoxidation of cyclohexene derivatives possessing various allylic substituents. Although 3-methoxy-and 3-acetoxy-cyclohexene gave predominantly the trans-epoxide, cyclohex-2-en-1-01 was converted mainly into the cisepoxide, a t a significantly faster rate. The observation was reasonably explained in terms of a transition state involving hydrogen bonding between the hydroxygroup of the allylic alcohol and the peroxy-acid. Since then, numerous examples of the intramolecular facilitation of epoxidation by hydroxy-groups have been described by different workers, particularly in the steroid field,2 but little has been done to put the phenomenon on a more quantitative basis, and to define more closely the preferred geometry of the transition state.An apparently anomalous result has been obtained with medium ring allylic alcohols ; cis-cyclo-oct-2-en-1-01 is reported 3 to give the trans-epoxy-alcohol, and cyclohept-2-en-P-o14 to give a mixture of the cis-and tramepoxy-alcohols on epoxidation. Studies on the stereochemistry of the reaction of peroxy-acids with acyclic allylic alcohols have been published,596 some of which will be discussed later; however, the results were discussed in terms of the preferred conformation of the starting alcohol in violation of the Curtin-Hammett principle .7As a result of another investigation,* 5-t-butylcyclohex-Zen-l-one became readily available. We decided, therefore, to examine the kinetics and stereochemistry of epoxidation of the epimeric 5-t-butylcyclohex-2-enols by peroxy-acids to try to define more closely the optimum requirements for intramolecular catalysis by the hydroxy-group. In addition we planned to re-investigate cyclo-oct-2-enol to see if the apparent anomalies could be resolved. A study of acyclic allylic alcohols was also envisaged but this was curtailed when the work of The n.m.r. data agree with the observation of Jensen and Bushweller l1 that pseudo-axial and pseudo-equatorial allylic hydrogens have similar chemical shifts.

The addition of less than 2 mol equiv. of N-bromosuccinimide (NBS) to a solution of diphenylacetylene in anhydrous dimethylsulfoxide (DMSO) leads to the formation of benzil in near-quantitative yield at room temperature. Under the same conditions stilbene gives the dibromo adduct. The conditions for this novel oxidation of an acetylene have been examined in some detail and it has been established that anhydrous DMSO must be employed as the solvent and that NBS is uniquely able to induce the oxidation. Preliminary studies indicate that alkyl aryl, dialkyl, and terminal acetylenes are converted to the corresponding a-dicarbonyl compounds, and that diphenylbutadiyne is oxidized to diphenyltetraketone. Optimum conditions for these latter oxidations have not yet been established. L'addition a la temperature ordinaire de N-bromosuccinimide (NBS), moins de 2 Cquivalents, a une solution de diphCnylacttylene dans du dimethylsulfoxyde (DMSO) conduit a la formation de benzil avec un rendement presque quantitatif. Dans les mimes conditions, le stilbene conduit au dCrivC dibromC. The combination of an N-haloamide or imide and an aqueous medium is often employed to convert an alcohol into a ketone or an olefin into a halohydrin (1). Hypohalous acid, formed in situ by hydrolysis -of the halogen source, is thought to be the oxidizing agent in most of these reactions (2). However, recent work by van Tamelen and Sharpless (3) and by Dalton and co-workers (4) indicates that the intervention of HOX is not obligatory. In aqueous glyme, N-bromosuccinimide (NBS) appears to effect direct transfer of positive bromine to a double bond (3); and, in the reaction of NBS with an olefin in moist dimethyl sulfoxide (DMSO), attack on the brominated cation or bromocarbonium ion is at least 95% by the DMSO (4). The bromohydrin is then produced by hydrolysis of the oxysulfonium intermediate 1.bromoketones are produced (5) and, with Nchlorosuccinimide or NBS in alcoholic solvents, the products are dichloro (6) or dibromoketals (7). In aqueous glyme, NBS effects direct transfer of bromine to a triple bond (7). This similar behavior of olefins and of acetylenes suggested that, in DMSO, a brominated vinyl cation might, like its dihydro analog, be trapped by the solvent to give 2, an unsaturated oxysulfonium cation. In the case of 2, however, competition may now occur between hydrolysis and either unimolecular (2 + 3 ; eq. 1) or bimolecular (2 + DMSO + 4; eq. 2) elimination of dimethyl sulfide. Loss of Br@ and dimethyl sulfide from 4, formed by the reactions of 2 or 3 with DMSO, would then lead, as indicated in eq. 3, to 5, an a-diketone, in what might be described as NBS-induced DMSO oxidation.Acetylenes resemble olefins in a number of respects in their behavior towards these halogenating agents. With hypobromous acid, di- For personal use only.

No abstract

No abstract

No abstract