Treatment of dienylpropargyl alcohol 9 with phenylsulfenyl chloride and base produces dienylallene sulfoxide 10, which undergoes competitive pericyclic processes: six-electron electrocyclization leads to drimatriene sulfoxides 11 and 12 while [1,7]-sigmatropic hydrogen shift leads to tetraene sulfoxides 13 and 14. In the absence of a very sterically demanding substituent at the allene terminus bearing the phenylsulfinyl group, the electrocyclization process was found to be favored. Furthermore, the sulfoxide moiety was observed to influence the stereochemical mode of electrocyclic ring closure as well as the [l,7]-sigmatropic hydrogen shift. Of the two possible modes of disrotatory electrocyclization (inward-inward versus outward-outward), the stereochemistry of the sulfoxide plays a decided role in the control of one of these two modes. Of the two possible antarafacial modes of the [ 1,7]-sigmatropic shift involving 10d, the phenylsulfinyl group exerts a small syn directive effect.cis-Dienylpropargyl alcohol 1 is transformed to drimatriene 4 (Scheme I) in a spontaneous, multistep one-pot process under mild conditions (PhSC1/Et3N, -78 "C to room temperature).' Through stereochemical studies,lb insight into the mechanistic pathway of this process has been discerned. This reaction is believed to proceed through initial formation of sulfenate ester 2, which undergoes a [ 2,3]-sigmatropic shift to afford allenyldiene sulfoxide 3 in a manner stereospecific with respect to allene but with little selectivity at sulfur. The allene sulfoxide 3, a species not directly observed, is then considered to undergo a stereospecific, six-electron electrocyclization in only one disrotatory sense to afford the observed product 4 as a pair of sulfoxide diastereomers. Had 3 undergone the alternative, allowed disrotatory electrocyclic ring closure, the opposite exocyclic doublebond geometric isomer of 4 would have resulted. It was originally anticipated (vide infra) in connection with vitamin A related studies3 that the [1,7]-sigmatropic shifted products 5a and 5b rather than electrocyclization product 4 would be observed. Nevertheless, the serendipitous finding that cyclization product 4 is produced preferentially is interesting and the process has synthetic potential. For example, it has been demonstrated with an optically active derivative of 1 (vide infra) that this process occurs with complete stereospecificity. Thus, this process provides a means for the enantiospecific synthesis of chiral building blocks.An investigation of the competitive [ 1,7]-sigmatropic hydrogen shift of the kind represented by the isomerization of 3 to geometric isomers 5a plus 5b, besides being of possible use in our vitamin A studies, would have been of considerable interest in connection with our attempts to evaluate the effect of substituents on 4n versus 4n + 2 pericyclic proce~ses.~ Such a process is closely related to the [1,5]-sigmatropic hydrogen shift of the allene sulfoxide 7 (prepared in a similar way from propargyl alcohol 6) to 8a plus 8b (Sch...