Tiglic acid allylation

This oxidation is applicable to a wide variety of both electron-rich and electron-poor alkenes for example a number of tiglic acid derivatives undergo this reaction in moderate to excellent yield, as in equation (5). In cases where there are several nonequivalent allylic sites the course of this reaction is highly substrate dependent, and the yields and selectivity vary from excellent to mediocre. In tiisub-stituted alkenes (except fw most 1-al cyclohexenes), a reactivity pattern has emm ged which has been termed a preference for syn ene attack (or PSEA ) This means that O2 will preferentially attack one of the two allylic carbons which ate cu to each othw. In practical terms, this still translates into relatively low product selectivity in most cases such as in equation (6), although some notable excqrtions are known (equations 7 and 8). 

While y-alkylations did not occur with saturated and benzylic halides, y-selectivities in the 62-99% range were observed in reactions of dicopper(I) dianions of a variety of a, -unsaturated acids with allylic halides. y-Unsubstituted allylic halides reacted by an Sn2 mechanism, y-disubstituted compounds underwent direct 5n2 displacement and y-monosubstituted systems reacted by both 5n2 and 5n2 pathways. Scheme 72 provides an example of the dramatic reversal in regioselectivity that was observed in the allylation of the dianion (147) of tiglic acid when the metal cation was changed from lithium to copper(I). The y-alkylation product from the latter species was exclusively the ( )-isomer. 

---