Abstract
The low temperature irradiation of 2,3-diphenyl-2H-azirine (1) in DMBP-glass at -196° has been reinvestigated. It was possible to convert 1 nearly quantitatively into the dipolar species benzonitrile-benzylide (3, psi3 = 0,78), which exhibits UV.-absorptions at 344 (epsilon = 48000) and 244 nm (epsilon = 28500) (Fig. 1,Tab. 1). Irradiation of 3 with 345 nm light at -196° resulted in aImost complete reconversion to the azirine 1 (psi = 0,15; Fig. 2). When the solution of 3 in the DMBP-glass was warmed up to about -160° a quantitative dimerization to 1,3,4,6-tetraphenyl-2,5-diaza-1,3,5-hexatriene (8) occurred. This proves that 8 is not only formed by the indirect route 3 + 1 to 7 to 11to 8 known before (Scheme 1), but also by dimerization of 3 either by direct head to head coupling or via the intermediate e (p. 2675), followed by a fast thermal hydrogen transfer reaction. The occurrence of the dipolar intermediate 11 in the photochemical conversion of the bicyclic compound 7 to 8 could also be demonstrated by low temperature experiments: On irradiation at -196° 7 gave the cherry red dipolar intermediate 11 (lambda max = 520 nm), which at -120° isomerizes to 8. It should be noted, that neither 7 nor 11 are formed by dimerization reactions of 3. Experiments carried out at room temperature demonstrate, that both processes for the formation of 8 may compete: Irradiation of a solution of 1 (DMBP, c = 8x10-4 to 5x10-3 M) with 350 nm light of high intensity (which does not excite the bicyclic compound 7) leads to a relative high photostationary concentration of the dipolar species 3. Under these conditions the formation of 8 is due to dimerization of 3 (psi8 = 0,19). With low light intensity only a very low stationary concentration of 3 can be obtained. Therefore the reaction of 3 with 1, Ieading to the bicyclic intermediate 7, becomes now predominant (psi-1 = 1,55, which corresponds with the expected value of 2 x 0,8). Irradiation of 1 at - 130° with 350 nm light of high intensity gives 8 with a quantum yield of 0,44. This is in agreement with the theoretical vaIue psi8 = 0,4 for an exclusive formation of 8 by dimerization of 3. The lower quantum yield for the formation of 8 at room temperature makes probable that under these conditions 3 not only dimerizes to 8, but also to another, so far unidentified dimer, e.g. 2,3,5,6-tetraphenyl-2,5-dihydropyrazine. By flash photolysis of a solution of 1 (cyclohcxane, c = 10-4 M, 25°) the disappearance of 3 could directly be measured by UV.-spectroscopy: At relative high concentrations (c larger than 10-7 M) 3 disappeared according to a second order reaction with the rate constant k = 5 x 10-7 M-1 s-1. At lower concentrations (c < 10-7 M) the rate of disappearance of 3 follows first order kinetics. The rate constant of this pseudo first order reaction (3+1 to 7) has been determined to be 1 x 104 M-1 s-1. Using Padwa's table of relative rates for the cycloaddition of the dipolar species 3 to various dipolarophiles, including the azirine 1, an absolute rate constant of k ca. 8 x 108 M-1s-1 for the addition of 3 to the most active dipolarophile fumaronitrile could be estimated. In cyclohexane at room temperature, the diffusion controlled rate constant equals 6,6 x 109 M-1 s-1. In Table 1 the UV.-maxima of several nitrile-ylides, among them a purely aliphatic one, are given.
Orahovats, Alexander