Mechanism of gaseous reactions : the oxidation of the lower aliphatic esters and propane

The thesis describes an investigation of the gas phase oxidation of the lower aliphatic esters and of propane. The kinetics of the oxidation of methyl, ethyl, and propyl formate, methyl acetate, methyl propionate and methyl butyrate were determined at various temperatures from 280° to 450°C. The experiments were made in a silica reaction vessel and the pressure changes accompanying oxidation were followed on a mercury manometer. The pressure-time curves usually showed an induction period after which the rate of change of pressure in the reaction vessel increased up to a maximum. Two temperature ranges of reactivity were observed with ethyl and propyl formate, methyl butyrate and propane. The rate of oxidation increased smoothly with increasing temperature up to about 350°C. Between 350° and 400°C. the rate decreased rapidly with an increase in temperature. Above 400°C. the rate increased continuously until ignition occurred. At all temperatures, an increase in the initial concentration of the combustible material increased the rate of oxidation and decreased the induction period. In the high temperature region an increase in the oxygen concentration similarly increased the rate of oxidation. In the low temperature region the rate of oxidation was independent of the oxygen concentration. Except at very low partial pressures of oxygen the induction period was independent of ths oxygan concentration in both the low and high temperature regions. A mass spectrometer was used to analyse the combustion gases for the unchanged ester (or propane), for residual oxygen, and for the amount of carbon monoxide and carbon dioxide formed. At all temperatures the concentrations of the major constituents were directly proportional to the pressure changes measured in the reaction vessel. Sufficient experiments were made to establish the pressure measurements as a true measure of the oxidation process. Chemical and polarographic analyses were made for formaldehyde, aoetaldehyde, total acid and peroxide substances. In each case the concentration of the intermediate species passed through a maximum as the reaction went to completion. No significant oxidation was detected during the induction period. The outstanding feature of the general kinetics of the esters was the marked dependence of the rate of oxidation, in the low temperature region, on the molecular structure of the combustible material. An increase in the carbon chain length increased the rate of oxidation much more than would have been expected from tue normal organic chemistry of a homologous series. The effect was most pronounced between the esters of higher molecular weight. An increase in the carbon chain length on the ethereal side of the ester configuration had a much greater effect than a corresponding increase on the carbonyl side. Where only methyl groups, and no metnylene groups occurred, the ease of oxidation wae greatly reduced. The kinetic results are interpreted in terms of a series of reaction steps analogous to those proposed by Cullis and Hinshelwood for paraffin oxidation. The scheme postulates the formation of peroxide substances capable, in the low temperature region, of decomposition and the propogation of a branching chain reaction. The rate expression for the oxidation reaction is of the form f([O2], [RH]) l - 2kF The remarkable effects of the substituent groups are attributed largely to the effect of the structural changes on the rate of decomposition, k , of the peroxide species. The theories of Walsh and Norrish are discussed in relation to the experimental evidence of the present investigation. The analytical investigation of the cool flame mixtures of propane indicated that the products of cool flame combustion are very similar to those detected in the low temperature region of oxidation. The changes in the concentrations of the initial reagents and final products followed the same pattern as that observed at lower temperatures. The polarograme of the combustion samples collected immediately before, and after, the occurrence of a cool flame, showed a marked decrease in the concentration of peroxide material with the explosion. The analyses support the theory of Hardwell and Hinshelwood, that oool flakes result from the self-heating of the gas mixture by the thermal decomposition of the peroxide intermediates. The general scheme proposed is R-+ 02=R00- R00-+ RH=R00H + R- R00H=R0- + -0H where the temperature coefficient of the rate of decomposition of the peroxide is greater than that of the rate of formation. If the decomposition reaction is highly exothermic, under appropriate conditions of concentration and reaction temperature, self-heating would be expected, and the rate would increase rapidly until the peroxide substances are largely consumed.