A study of the mechanism of the thermal decomposition of 1-bromo-2-(1-naphthyl)naphthyl-carbinol and related diarylcarbinols

The synthesis of a variety of polynuclear aromatic hydrocarbons by means of the condensation of grignard reagents with aromatic aldehydes did not always lead to the isolation of the desired diarylcarbinols. The delicate relationship between molecular structure and physiological activity demanded that only pure compounds of known structure be used in studying the carcinogenic and/ or carcinolytic effects of these aromatic hydrocarbons. Therefore, the above method of synthesis was eventually by-passed. But the apparently "abnormal" condensation of grignard reagents with aromatic aldehydes and the thermal decomposition of diarylcarbinols at high temperatures generated a great deal of interest from the mechanistic point of view, and these reactions were studied in detail in this work.

The previously unreported ketones, 2-fluorophenyl 1-naphthyl ketone (52) and 2-fluorophenyl 2-naphthyl ketone (53), were prepared by the condensation of the cadmium reagents of 1- and 2-bromonaphthalene with 2-fluorobenzoyl chloride (51). Reduction of these ketones, 52 and 53, with a mixture of lithium aluminum hydride and aluminum chloride afforded the corresponding diarylmethanes, 2-(1-naphthylmethyl)fluorobenzene (56) and 2-(2-naphthylmethyl)fluorobenzene (57), respectively. Reduction of these same ketones, 52 and 53, with sodium hydroborate afforded the corresponding diarylcarbinols, 2-fluorophenyl-1-naphthyl carbinol (54) and 2-fluorophenyl-2-naphthyl carbinol (55), in good yield. The previously unreported 1-bromo-2-(1-naphthyl)naphthyl carbinol (45) was prepared by reduction of 1-bromo-2-(1-naphthyl) naphthyl ketone (44) with sodium hydroborate. This diarylcarbinol 45 was also prepared via the reaction of the Grignard reagent of 1-bromonaphthalene with 1-bromo-2-naphthaldehyde (41). The proof of structure of 45 was accomplished by studying the nuclear magnetic resonance and infrared spectra of 45 in conjunction with the nuclear magnetic resonance and infrared spectra of 1-bromo-2-(1-naphthyl)naphthyl carbinol-OD₁ (46) and 1-bromo-2-(1-naphthyl)naphthyl carbinol-CD₁ (47).

The reported "abnormal" reaction of the Grignard reagents was investigated. The reaction of one equivalent of the Grignard reagent of 2-bromonaphthalene with one equivalent of 2-chlorobenzaldehyde gave the expected 2-chlorophenyl-2-naphthyl carbinol (37). The reaction of two equivalents of 2-chlorobenzaldehyde with one equivalent of the Grignard reagent of 2-bromonaphthalene gave 2-chlorophenyl 2-naphthyl ketone (36) by oxidation of the alkoxide complex formed by the initial condensation of the Grignard reagent with the aromatic aldehyde.

The high temperature decomposition of diarylcarbinol 45 was found to lead to the formation of sym. -1-bromo-2-(1-naphthyl)naphthylmethyl ether (58). The structure of 58 was confirmed by molecular weight studies, infrared and nuclear magnetic resonance spectra and elemental analyses.

The decomposition of ether at 58 240° was found to proceed slowly with the liberation of hydrogen bromide gas. Free radical initiators and hydrogen bromide gas were found to accelerate the decomposition of 58 at 240°. A reasonable mechanism has been postulated for the decomposition of 58 at 240°.

The cleavage of ether 58 in the presence of hydrogen bromide gas would give rise to the formation of a secondary bromide, 1-bromo-2-(1-naphthyl)naphthylmethyl bromide (62). Homolysis of 62 would then give a bromine atom and a secondary free radical, 1-bromo-2- (1-naphthyl)naphthylmethyl free radical (63). The attack of free radical 63 on the benzylic hydrogen atoms of ether 58 is thought to be responsible for the formation of the diarylmethane, 1-bromo-2-(1-naphthylmethyl) naphthalene (48) and 1-brome-2-(1-naphthyl)naphthylmethyl ether free radical (64). Homolysis of 64 would then regenerate the secondary free radical 63 and the ketone 44.

The coupling of two secondary free radicals 63, with subsequent attack of bromine atoms on the system, might give rise to the formation of di-[1-bromo-2-(1-naphthyl)naphthyl] ethylene (66). The attack of hydrogen atoms on 63 might lead to the formation of diarylmethane 48. Oxidation of ethylene 66 might give rise to the isolation of ketone 44. Evidence for these two reactions has been furnished by the vacuum distillation of secondary bromide 62.

The attack of bromine atoms on the benzylic hydrogen atoms of ether 58 would give ether free radical 64 and hydrogen bromide gas. Homolysis of 64 would then give ketone 44 and secondary free radical 63. The decomposition of 58 would then appear to be cyclic in nature.

The decomposition of diarylcarbinols 54 and 55 was judged to be free radical in nature, involving the formation of the two ethers, sym. -2-fluorophenyl-l-naphthylmethyl ether (67) and 2-fluorophenyl-2-naphthylmethyl ether (68), which decompose in the same temperature range as 54 and 55. Ethers 67 and 68 were not isolated. The decomposition of 67 and 68 was postulated to be a series of homolytic cleavages, giving rise to the formation of the diarylmethanes, 56 and 57, and the ketones, 52 and 53.