PubMed2003
Photooxygenations of PhSMe and Bu2S sensitized by N-methylquinolinium (NMQ+) and 9,10-dicyanoanthracene (DCA) in O2-saturated MeCN have been investigated by laser and steady-state photolysis. Laser photolysis experiments showed that excited NMQ+ promotes the efficient formation of sulfide radical cations with both substrates either in the presence or in absence of a cosensitizer (toluene). In contrast, excited DCA promotes the formation of radical ions with PhSMe, but not with Bu2S. To observe radical ions with the latter substrate, the presence of a cosensitizer (biphenyl) was necessary. With Bu2S, only the dimeric form of the radical cation, (Bu2S)2+*, was observed, while the absorptions of both PhSMe+* and (PhSMe)2+* were present in the PhSMe time-resolved spectra. The decay of the radical cations followed second-order kinetics, which in the presence of O2, was attributed to the reaction of the radical cation (presumably in the monomeric form) with O2-* generated in the reaction between NMQ* or DCA-* and O2. The fluorescence quenching of both NMQ+ and DCA was also investigated, and it was found that the fluorescence of the two sensitizers is efficiently quenched by both sulfides (rates controlled by diffusion) as well by O2 (kq = 5.9 x 10(9) M(-1) s(-1) with NMQ+ and 6.8 x 10(9) M(-1) s(-1) with DCA). It was also found that quenching of 1NMQ* by O2 led to the production of 1O2 in significant yield (PhiDelta = 0.86 in O2-saturated solutions) as already observed for 1DCA*. The steady-state photolysis experiments showed that the NMQ+- and DCA-sensitized photooxygenation of PhSMe afford exclusively the corresponding sulfoxide. A different situation holds for Bu2S: with NMQ+, the formation of Bu2SO was accompanied by that of small amounts of Bu2S2; with DCA, the formation of Bu2SO2 was also observed. It was conclusively shown that with both sensitizers, the photooxygenations of PhSMe occur by an electron transfer (ET) mechanism, as no sulfoxidation was observed in the presence of benzoquinone (BQ), which is a trap for O2-*, NMQ*, and DCA-*. BQ also suppressed the NMQ+-sensitized photooxygenation of Bu2S, but not that sensitized by DCA, indicating that the former is an ET process, whereas the second proceeds via singlet oxygen. In agreement with the latter conclusion, it was also found that the relative rate of the DCA-induced photooxygenation of Bu2S decreases by increasing the initial concentration of the substrate and is slowed by DABCO (an efficient singlet oxygen quencher). To shed light on the actual role of a persulfoxide intermediate also in ET photooxygenations, experiments in the presence of Ph2SO (a trap for the persulfoxide) were carried out. Cooxidation of Ph2SO to form Ph2SO2 was, however, observed only in the DCA-induced photooxygenation of Bu2S, in line with the singlet oxygen mechanism suggested for this reaction. No detectable amounts of Ph2SO2 were formed in the ET photooxygenations of PhSMe with both DCA and NMQ+ and of Bu2S with NMQ+. This finding, coupled with the observation that 1O2 and ET photooxygenations lead to different product distributions, makes it unlikely that, as currently believed, the two processes involve the same intermediate, i.e., a nucleophilic persulfoxide. Furthermore, the cooxidation of Ph2SO observed in the DCA-induced photooxygenation of Bu2S was drastically reduced when the reaction was performed in the presence of 0.5 M biphenyl as a cosensitizer, that is, under conditions where an (indirect) ET mechanism should operate. This observation confirms that a persulfoxide is formed in singlet oxygen but not in ET photosulfoxidations. The latter conclusion was further supported by the observation that also the intermediate formed in the reaction of thianthrene radical cation with KO2, a reaction which mimics step d (Scheme 2) in the ET mechanism of photooxygenation, is an electrophilic species, being able to oxidize Ph2S but not Ph2SO. It is thus proposed that the intermediate involved in ET sulfoxidations is a thiadioxirane, whose properties (it is an electrophilic species) seem more in line with the observed chemistry. Theoretical calculations concerning the reaction of a sulfide radical cation with O2-* provide a rationale for this proposal.