No abstract

The thermal decomposition of nitroglycerin has been investigated in the vapor and liquid phases over a temperature range of 115-160'. Above 140' the decomposition in the vapor phase is first order but in the liquid phase the order is only approximately first. At all temperatures the rates were found to vary markedly with the ratio of the mass of nitroglycerin to the volume of the reactor. Below 140' the decomposition was autocatalytic. -4 mechanism is proposed for the vapor-phase reaction that is in agreement with the experimental data. IntroductionThe kinetics of the thermal decomposition of nitroglycerin have been the subject of considerable study. For the most part, the rate constants and order of reaction have been determined by manometric methods. Because of the complexity of the decomposition, this has resulted in irregularities in values of the kinetic constants. Roginskii,l for example, reports the reaction to be first order between 150 and 190'. At lower temperatures, the decomposition was found to be autocatalytic. Over three different temperature ranges extraordinarily high values for the frequency factor in the Arrhenius equation were obtained. This was explained as being due to the existence of chains. However, Roginskii's extensive study of the effects of known chain inhibitors on the rate of decomposition crf nitroglycerin failed to substantiate their presence.The autocatalytic decomposition of nitroglycerin has been studied also by Lukin2 and more recently by Andreev and Glaakova3 and Gorbunov and S~e t l o v .~ These investigators found that certain oxides of nitrogen, water, and various acids increased the rate of decomposition.Robertson,6 in a brief study on the thermal decomposition of nitroglycerin in the liquid phase, confirmed Roginskii'sl results and concurred with the latter's explanation of the high-frequency factors and temperature dependency of the activation energy. Phillips,e employing a manometric technique, obtained somewhat lower values for the frequency factor and activation energy for the liquid-phase decomposition than did Roginskiil and Robertson.5 Svetlov' also investigated the thermal decomposition of nitroglycerin in the liquid phase between 80 and 140' using manometric methods. He noted that, a t constant temperatures, different rates were obtained depending on the ratio of the mass of nitroglycerin to the volume of the reaction chamber.From a review of the literature it is apparent that the thermal decomposition of nitroglycerin is a complex process and one that is sensitive to the conditions of the experiment. It is also evident that manometric studies alone are not in themselves adequate for deter-

The thermal decomposition of the chemiluminescent compound tetrakis(dimethylamino)ethylene (TMAE) has been investigated in a static system between 283 and 323 "C over initial pressures from 6.0 to 45 mmHg. The order of the reaction was first, as was the order for the formation of methane and dimethylamine, the major decomposition products. The Arrhenius parameters for the TMAE decomposition are given as k¡ = 9.0(±1.0) X 10ue-39 900(±2ooo)/flr g_i wjjjje those for CH4 and DMA are &ch4 = 4.3(±0.03) X 10ne-39 300(±300)/rt s-i an¿ feDMA = 2.5(±0.01) X 10loe-36 5oo(±2O0)/fiT g-i¡ respectively. The effect of certain foreign gases indicated that the decomposition was free radical in nature. A mechanism is suggested which accounts for the predominant gaseous products and calculations from experimental data are offered in support of the proposed mechanism.

The reaction of 2'-substituted biphenyl-2-carboxylic acids (where the 2'-substituent is H, CO,H, NO,, CI, OMe. or C0,Me) with lead tetra-acetate in refluxing benzene solution, under a nitrogen atmosphere, affords 3.4-benzocoumarin as a major organic product. Only a trace is formed when the 2'-substituent is Me. o-Terphenyl-2carboxylic acid gives no 3.4-benzocoumarin, but triphenylene is formed in good yield. It is suggested that these cyclizations involve the intermediacy of free radicals, particularly as the products of the thermal decomposition, in benzene solution, of the diacyl peroxides derived from 2'-substituted biphenyl-2-carboxylic acids, are similar to the products formed on reaction of the acids with lead tetra-acetate.THE most recent studies on the reaction of lead tetraacetate with carboxylic acids 1-5 (acids belonging to the norbornane6 and the bornane' series may be exceptions) indicate that the organic products are derived from free-radical intermediates. A study of the reaction of the 2'-substituted biphenyl-2-carboxylic acids (1a)-(Ig) with lead tetra-acetate in boiling benzene, under a nitrogen atmosphere, has shown that with the exception of the methyl-acid (Ig), 3,4-benzocoumarin (V) is a major product in each case (see Table ). The most likely mechanism for the formation of 3,4-benzocoumarin (V) is therefore via the intermediates ( 11), (111), and (IV), and involves a free-radical displacement of the substituent group in the final steps (111) -w (IV) -+ (V). I n the oxidation of 2'-chlorobiphenyl-2-carboxylic acid (Id), 2-chloro-o-terphenyl and o-chlorobiphenyl are additional reaction products. They provide evidence for the intermediacy of free radicals in that their formation is most satisfactorily explained via 2'-chlorobiphenyl-2-y1 radicals, which either substitute in the solvent benzene or abstract hydrogen from a suitable donor. Competition between the intramolecular substitution reaction (IIId) _t (IVd) __t (V) and decarboxylation to 2'-chlorobiphenyl-2-y1 radicals is readily suggested. 2-Methoxycarbonyl-o-terphenyl is likewise an additional product produced from the methoxycarbonyl-acid (If), but similar products are not formed from the other acids, which indicates that the ease of the displacement of the group X relative to decarboxylation is dependent on the nature of the group X. The thermal decomposition, in benzene solution, of diacyl peroxides,8 even if highly a~ymmetrical,~ invariably proceeds by a free-radical pathway. Further support for the importance of free-radical intermediates in product formation from the lead tetra-acetate oxidt We are grateful to the referees for this suggestion.

No abstract