Abstract
2, 2, 3‐Triphenyl‐2__H__‐azirine (4a) in a matrix of 2, 2‐dimethylbutane/pentane 8:3 (DMBP) at −185° gave rise on irradiation with light of 250–350 nm to a new UV.‐maximum at 350 nm (Fig. 1). We assign the dipole benzonitrildiphenylmethylide (1a) to this new maximum. Irradiation with monochromatic light of 366 nm destroyed this maximum and the initial absorption curve reappeared (Fig. 2). When the azirine 4a was photolysed in DMBP at −185° in the presence of methyl trifluoracetate (TFEM), the maximum at 350 nm was obtained again. This maximum vanished upon increasing the temperature to −160°. Through gas chromatography we were able to show that 5‐methoxy‐5‐trifluormethyl‐2, 2, 4‐triphenyl‐3‐oxazoline (6a) was produced. 6a was also obtained upon irradiation of 4a at room temperature in the presence of TFEM (scheme 1 and table 1). Modification of the previously described experiment, in which the maximum at 350 nm was extinguished in the matrix due to irradiation at 366 nm gave, after warm up, almost no dipole adduct 6a (table 1). From these experiments, an extinction coefficient of 17, 000 for the 350 nm maximum of 1a, was calculated. These experiments have shown that irradiation of triphenylazirine 4a leads to the dipole 1a, which can be reversed photochemically – but not tharmally – into azirine 4a. 1a reacts at less than −160° with TFEM to give adduct 6a. The results which were obtained with triphenylazirine 4a could be correspondingly obtained with 2, 3‐diphenyl‐2__H__‐azirine 4b (Fig. 3, scheme 2 and table 2). The dipole 1b showed two UV.‐maxima at 330 nm (ε = 17, 500) and 343 nm (ε = 21, 000). Later experiments established, that the two maxima belonged to a single dipole species.
The dipole 1c obtained upon irradiation of 2, 2‐dimethyl‐3‐phenyl‐2__H__‐azirine (4c) in DMBP at −190°, appears to absorb in the same region as the azirine 4c. The presence of the dipole 1c was univocally established by low temperature trapping experiments with TFEM. The dipole 1a showed no ESR.‐spectrum characteristic for a triplet state. We assume therefore, that 1a is in a singlet state. Photolysis of oxazolinone 7 at −190° in DMBP led to the dipole 1a with loss of CO~2~. 1a recombines apparently in considerable amount with the CO~2~ trapped in the matrix to give starting oxazolinone 7 because the 350 nm‐maximum of 1a appeared with low extinction. Irradiation with light of 366 nm into this matrix produccd triphenylazirine 4a. Low temperature trapping experiments with TFEM led to small amounts of 5‐methoxy‐5‐trifluormethyl‐triphenyl‐3‐oxazolinc (6a).