Hydroxy-group directivity, allylic alcohol epoxidation

Hydroxy group directivity in the epoxidation of chiral allylic alcohols 99ACR703. 

W. Adam, T. Wirth, Hydroxy group directivity in the epoxidation of chiral allylic alcohols Control of diastereoselectivity through allylic strain and hydrogen bonding, Acc. Chem. Res. 32 (1999) 703. 

Allylic alcohols are interesting substrates for epoxidation because they produce epoxides with a hydroxyl group as additional functional group that is able to play an important role in the subsequent synthesis of complex molecules [105]. This synthesis aspect certainly benefits from the hydroxy-group directed selectivity of oxygen delivery. 

The most important hydrogen bond donating group in directed epoxidations is the hydroxy group. For allylic or homoallylic alcohols, peracids or tert-butyl hydroperoxide/vanadylbis[2,4-pentanedionate] (see Houben-Weyl, Vol. IV/la, p 231) are generally the most efficient reagent systems less common catalysts are tri-te/ f-butoxyaluminum, dibutyltin oxide, and molybdenum- and titanium-based systems (see Houben-Weyl, Vol. IV/la, p 227, Vol. E13/2, p 1176). The two classes of reactions show distinct differences in their stereoselectivity patterns. 

W. A. Adam, H.-G. Degen, C. R. Saha-Moller, Regio- and diastereoselective catalytic epoxidation of chiral allylic alcohols with hexafluoroacetone perhydrate. Hydroxy-group directivity through hydrogen bonding, J. Org. Chem. 64 (1999) 1274. 

This changed completely with the hydroxy-group-directed epoxidation of allylic alcohols developed by Sharpless and Katsuki [1]. 

In the epoxidation of acyclic allylic alcohols (Scheme 6), the diastereoselectivity depends significantly on the substitution pattern of the substrate. The control of the threo selectivity is subject to the hydroxyl-group directivity, in which conformational preference on account of the steric interactions and the hydrogen bonding between the dioxirane oxygen atoms and the hydroxy functionality of the allylic substrate steer the favored 7r-facial... 

Interestingly, MCPBA epoxidation of cis alcohol 16 affords a mixture of diastereomeric epoxides (55 45 mixture). Furthermore, protection of the allylic alcohol as TBS ether (17) and subsequent epoxidation results as well in hardly any stereochemical selectivity (53 47 mixture). With regard to these results it is suggested that the trans-allylic hydroxy group is effectively involved in directing the MCPBA epoxidation event. 

In the peracid epoxidation of E or Z medium-size ring allylic alcohols, often in both diastereomeric transition states a directing effect by the hydroxy group is feasible. The strong preference for trans epoxidation by peracids [99.8 0.2 for (Z)-2-cyclooctenol, see Houben-Weyl, Vol. E13/2, pp 1189,1190] is mainly a consequence of the transannular ring strain in the conformer with pseudoaxial hydroxy group. 

Since epoxidation at the vinyl double bond is unproductive, it is desirable to direct reaction on the al-lene moiety. This can be accomplished by taking advantage of the hydroxy-directed epoxidation of allylic alcohols using the t-butyl hydroperoxide/vanadium(V) system.The directing effects of both allylic and homoallylic type hydroxy groups have been examined at both positions of the vinylallene unit. " At the 1-position (64), primary, secondary and tertiary allylic tdcohols are effective, while only primary homoallylic alcohols have bran examined (equation 35). Presumably the directing effect of the hydroxy groups favors formation of the intermediate allene oxide (65). A sample of the compounds prepared by this route is shown in Scheme 32. ... 

Directing effects of various functional groups on the stereoselectivities of peroxy acid epoxidations have been studied in detail. In the case of allylic alcohols, the weak directing effect of the hydroxy... 

The cis and trans allyl alcohols, 29 and 30, were epoxidized and acetylated to afford the acetates 31,32,35, and 25. The stereochemistries were revealed at a later stage. The epoxide 31 was treated with Me2CuLi to give a diol 33 in good yield (Scheme 7a). The other epoxides, 32 and 35 were treated similarly to give diols 34 and 36, respectively, however the epoxide 25 did not react at all. The reason for this is not yet clear, but one reason may be that the epoxides tend to open in the direction to yield diaxial alcohols. In order to obtain an a-hydroxy aldehyde, each diol, 33, 34, and 36, was treated with NCS/Me S to yield an aldehyde with a methylthiomethyloxy group (for example 37). However these... 

Another method for the conversion of an alkene into an allylic alcohol, but with a shift in the position of the double bond, proceeds from the corresponding p-hydroxyselenide. The p-hydroxyselenide can be obtained from the epoxide by reaction with phenylselenide anion or directly from the alkene by addition of phenylselenenic acid, phenylselenenyl chloride in aqueous MeCN, or by acid-catalysed reaction with A-phenylseleno-phthalimide. The hydroxyselenide does not need to be isolated, but can be oxidized directly with tert-BuOOH to the unstable selenoxide, which spontaneously eliminates phenylselenenic acid to form the E-allylic alcohol. For example, 4-octene gave 5-octen -ol (6.15). Elimination takes place away from the hydroxy group to give the allylic alcohol no more than traces... 

---