Heterocyclic Synthesis

Abstract

The cation triphenylmethylium and the cation radical tris-(p-bromophenyl)ammoniumyl are effective catalysts for oxygenation of ergosteryl acetate (1 ; R = Ac) to the peroxide (2; R = Ac) ; the triphenylmethylium-catalysed reaction is a photo-oxygenation, whereas the cation radical reaction is thermal. Several other ' forbidden ' additions of triplet oxygen to cisoid dienes have been demonstrated. A number of cation radicals have been examined for catalytic oxygenation ability. Ergosteryl acetate can also be converted into its peroxide by triplet oxygen in the presence of various Lewis acids. Two mechanisms for overcoming the spin barrier for triplet oxygen addition have been defined by experimental work. A mechanistic interpretation of the facts is proposed. Chemistry Department, Imperial College, South Kensington, London SW7 2AY DURING studies related to the use of trityl cation for the deprotection of masked steroidal alcohols,l we treated ergosterol 3p-methoxymethyl ether (1 ; R = MeO-CH,) with trityl tetrafluoroborate (0.3 equiv.) in dichlorome-thane without precautions to exclude oxygen. At-78 "C the peroxide (2; R = MeOCH,) was formed in high yield. Ergosterol 3P-acetate (1; R = Ac) similarly afforded the 5a,Sa-peroxide (2; R = A c). ~ Further investigation showed that the acetate (1; R = Ac) in dry dichloromethane containing a catalytic amount of tri-tyl tetrafluoroborate a t-78 "C, on exposure to air and laboratory lighting, gave the peroxide (2; R = Ac) When trityl cation in dichloromethane at-78 "C, in the presence of oxygen, was irradiated in the absence of ergosteryl acetate, no oxygen was consumed (cf. ref. To compare conventional singlet oxygen photo-oxygena-tion and the trityl cation system the experiments sum-marised in Table 1 were carried out. The rate ratio k (1 ; R = Ac)/k(3; R = Ac) was ca. 6 000 for the trityl system whereas the ratio for the eosin system was ca. 3.3. In our opinion this dramatic difference in reactivity discounts the trityl cation acting merely as a triplet-to-singlet oxygen sensitiser. Azotriphenylmethane (5) in dichloromethane 3). TABLE 1 Comparative photo-oxygenations with the trityl cation and with eosin Time of 2 min 30 min 100 min Substrate Conditions oxygenation Product (% yield) (1; R = Ac) (3; R = Ac) Substrate (100 mg), Ph,C+BF,-(10 mg), CH2C1,;-78'; 0, 10 h 5% Conversion into (4; R = Ac) (1; R = Ac) (3; R = Xc) Substrate (100 mg), Ph3C+BF,-(10 mg), CH,Cl,;-78'; 0, Substrate (50 mg), eosin (10.5 mg). PhCN-CH,Cl, (2 : 1);-15'; 0, Substrate (50 mg), eosin (10.5 mg), PhCN-CH2C1, (2 : 1);-15' 0, (2; R = Ac) (100) (2; R = Ac) (>95%) (4; R = Ac) (~ 9 5 %) quantitatively in 2.75 h. In the dark there was no reaction. Irradiation of this system with a tungsten lamp (500 W) gave the peroxide (2; R = Ac) in 30 min. Under an atmosphere of pure oxygen, peroxide (2; R = Ac) formation was complete (1.03 mol. equiv. uptake) in 10 min at-78 "C even on a preparative scale (>1 g). The trityl cation was isolated as triphenylmethanol (86%) from aqueous work-up. Photo-oxygenation a t-15 "C, 0 "C, or room temperature did not give clean reactions. Only a t less than-15 "C was the peroxide (2; R = Ac) formed in good yield. "C with ergosteryl acetate (1; R = Ac) was oxy-genated under irradiation. Warming the mixture to-15 "C gave no peroxide (2; R = Ac), only triphenyl-methyl p e r ~ x i d e. ~ Trityl radicals are not, therefore, the catalytic oxygenation species. Trityl hexachloro-antimonate,6 perchlorate, and hexafluorophosphate were all effective oxygenation catalysts, and a comparative study indicated no appreciable difference in the rate of formation of the peroxide (2; R = Ac) with each cata-Ergosteryl acetate in dichloromethane (-78