First published as an Advance Article on the web 1st February 20014,5-Diphenyl-1,2,3-selenadiazole and 3,5-diphenyl-1,2,4-selenadiazole were alkylated with trimethylsilylmethyl trifluoromethanesulfonate. Quaternisations occurred at N-3 and N-2 respectively. The salts were desilylated to generate transient selenadiazoliumylmethanide (ylide) intermediates. 4,5-Diphenyl-1,2,3-selenadiazol-3-ium-3-ylmethanide, 3, was trapped with dimethyl acetylenedicarboxylate and methyl propiolate in a cycloadditionrearrangement reaction which gave the pyrazolylvinyl vinyl selenides 6 and 7. 3,5-Diphenyl-1,2,4-selenadiazol-2-ium-2-ylmethanide, 10, ring-expanded in situ to 4,6-diphenyl-2H-1,3,5-selenadiazine, 13.Interest in organoselenium systems has increased with the recognition of the physiological role of selenium in a secondline anti-oxidant defence against lipid autoxidation, 1 as well as the bioactive nature of some fused bicyclic 1,2,3-selenadiazole systems 2 and organoselenium compounds in general. 3 Transformation of selenium heterocycles has been a major strategy in the synthesis of new organoselenium compounds but the difficulty attached to the instability of selenium azoles has limited the scope of this approach. 4 These difficulties have been overcome in recent years and reliable routes to a range of selenadiazoles have been developed, 4-10 although these substrates are still difficult to work with. Herein we explore the synthetic use of both the 1,2,3-and 1,2,4-selenadiazole systems by generating the first selenadiazolium methanide (ylide) species in situ. This work complements our recent reports 11,12 on thiadiazolium ylide systems and earlier work on oxadiazolium systems 13 and extends this chemistry deeper into Group VI. The selenadiazolium systems were the most unstable and difficult members of the group but they provided useful new additions to organoselenium synthetic routes.
Results and discussion (a) 1,2,3-Selenadiazol-3-ium-3-ylmethanidesWhen the selenadiazole 1 was heated at 80 ЊC with trimethylsilylmethyl trifluoromethanesulfonate alkylation occurred at N-3 giving the salt 2 as a dark sticky gum on cooling. 1 H and 13 C NMR spectra supported the structure and the quaternisation site was confirmed by 15 N NMR spectra which showed a large shielding shift of 177.6 ppm at the quaternised 3-N-atom and a smaller shift of 29.2 ppm at the adjacent N-2 site. In general quaternisation of higher azole N-atoms causes a large upfield shift (>100 ppm) in the 15 N NMR signal of the quaternised azolium nitrogen and a smaller shift (~25 ppm) on other ring N-atoms. 14,15 Similar shifts have been noted for quaternisation of substituted 1,2,3-thiadiazoles by L'abbé et al. 16 and us. 11 The subsequent chemistry also confirmed quaternisation at N-3. Attempted purification of the sticky salt resulted in loss of Se with decomposition. Hence the crude salt was used immediately to generate the unstable 1,3-dipole 3.The species 3 was generated at ambient temperatures in dichloromethane by treatment of 2 with CsF (following a litera...