Dehydrochlorination of α,β-epoxy sulfonyl chlorides 3 leads to 1-alkene-1,3-sultones 1 most probably via epoxy sulfenes 2.1-Alkene-1,3-sultones (1,2-oxathiole-2,2-dioxides) 1 are interesting compounds for synthetic exploration. In fact three places of nucleophilic attack can be envisaged for these α,β−unsaturated γ-sultones,viz. at sulfur, at C 4 (Michael addition) and at C 5 (sulfonate displacement). Surprisingly, the chemistry of unsaturated γ-sultones received little attention in the literature. 1 Compounds 1 are the sulfonyl analogs of butenolides [2(5H)-furanones], which were prepared by us through a ring expansion of α,β-epoxy ketenes readily generated by a photoinduced reaction of α,β-epoxy diazomethyl ketones. 2 On this base, we propose an analogous approach for the synthesis of 1. Thus, retrosynthetically ring expansion of an epoxy sulfene 2 might be expected to give α,β-unsaturated sultones 1. The most appropriate route to epoxy sulfenes would be dehydrochloration of suitable sulfonyl chlorides, 3 i.e. β,γ-epoxy sulfonyl chlorides 3 (Scheme 1).
Scheme 1The required starting materials 3 were obtained by reaction of allylic bromides 4 with sodium sulfite in aqueous medium, 4,5 followed by treatment of the salts 5 with phosphorus oxychloride 4,5 and subsequent epoxidation of alkenesulfonyl chlorides 6. The epoxidation step proved to be much more difficult than expected for a β,γ-olefinic bond. The conventional reagents, such as m-chloroperbenzoic acid, trifluoroacetic acid/H 2 O 2 (35%), tert-butyl hydroperoxide/KF on alumina, dimethyl dioxirane, all failed. Pertrifluoroacetic acid, prepared from trifluoroacetic anhydride (TFAA) and 100% H 2 O 2 gave epoxidation of 6a in 95% yield. 6 However, anhydrous hydrogen peroxide is extremely dangerous to handle and no longer commercially available. Therefore, this was not anymore the reagent of choice. We found however, that methyl trifluoromethyl dioxirane 7 is an appropriate reagent for the smooth epoxidation of 6a-c (Scheme 2). 8
Scheme 2Isolation of the epoxy sulfonyl chlorides 3a-c 9 was easily accomplished by evaporation of dichloromethane and trifluoroacetone. Treatment of 3a,b with triethylamine in ether as the solvent at ambient temperature 10 resulted in the desired 1-alkene-1,3-sultones 1a and 1b 11 in ca 50% yield (Table). Substrate 3c gave only an inseparable mixture of products. The physical and spectral data of product 1a were in full agreement with those reported previously. 12 The yields of the various products prepared are listed in the Table. This result strongly suggests that the expected ring expansion of epoxy sulfene 2 indeed has taken place (Path A). In fact, it is well documented that dehydrochlorination of appropriate alkanesulfonyl chlorides is a general method for the in situ generation of sulfenes. However, an alternative pathway (B) can be envisaged, namely one involving initial deprotonation of the methylene group adjacent to the sulfonyl group, followed by an epoxide opening to give species 7 which then ring closes to yield product ...
