Slow Combustion of Gaseous Hydrocarbons

The fact that some substances unite with oxygen more readily than others paved the way for what may be termed the preferential theory of combustion, which was widely accepted during the greater part of last century. According to this theory, when a mixture of combustible substances is ignited, there is competition between the different elements for the oxygen. The same applies even in the case of a pure compound, such as a hydrocarbon, for the two constituent elements will compete for the oxygen. If the supply of air is limited the "most favoured" element will burn first, the remainder oxidising as opportunity serves.

This theory afforded a very plausible explanation for the luminosity of hydrocarbon flames. The hydrogen is to be regarded as the favoured element, and thus becomes preferentially oxidised, whilst the less favoured carbon is precipitated out into the flame in the white-hot condition, and either burns in excess of air at the outward fringe or escapes as smoke or soot in the uncombined condition.

Although, as has already been mentioned, preferential combustion may take place in the presence of certain catalysts, the theory as applied promiscuously to all cases of combustion leads to many difficulties. For example, when methane is exploded with its own volume of oxygen - that is, a volume insufficient for complete combustion - hydrogen and carbon monoxide are produced as well as water- vapour, in accordance with the equation

CH₄ + O₂ = CO + H₂O + H₂.

Similarly, ethylene yields carbon monoxide and hydrogen:

C₂H₄ + O₂ = CO + 2H₂

These facts were known to Dalton, and received support from the work of Kersten in 1861, who assumed that when once the hydrocarbon has been decomposed by the heat of the flame into hydrogen and carbon, the latter is preferentially oxidised to carbon monoxide, after which any excess of oxygen distributes itself between this gas and the hydrogen. A similar view was apparently held by Misteli.

The question therefore arises as to what factors decide whether or not an element shall be more favoured than another. If in the ordinary rapid combustion of ethylene, for example, hydrogen is the more favoured element, why should carbon be preferentially oxidised in an explosion? To this, a satisfactory reply has not as yet been forthcoming.

Although the preferential theory has not been disproved, modern opinion inclines towards the association theory of Bone and his collaborators. According to this, the oxygen of the air first combines with the hydrocarbons, forming more or less unstable hydroxylated products which ultimately, in a sufficiency of air or oxygen, decompose to carbon dioxide and water.

This theory was arrived at as the result of a series of classical researches into the mode of slow combustion of gaseous hydrocarbons in oxygen in contact with suitable catalysts.

In the preliminary series of experiments the gases were sealed in glass bulbs and maintained at definite temperatures ranging from 325° to 400° C. for several weeks. As only small volumes of gases (about 70 c.c.) could be dealt with in this manner, and as, moreover, the detection of transient intermediate products was very difficult, the final experiments were conducted by continuously circulating the gases (some 1200 c.c.) through a combustion tube packed with porcelain, and maintained at the desired temperature in a furnace.

A. It was first of all established that the following three reactions are incapable of taking place below 400° C.:

(i) C + H₂O = CO + H₂;
(ii)CO₂ + C = 2CO;
(iii)2C + O₂ = 2CO.

The reversible reaction

(iv) CO + H₂O ⇔ CO₂ + H₂

gave no evidence of proceeding at 325° C. in the direction left to right even after a fortnight. At 350° C. some 1.7 per cent, of carbon dioxide was produced after ten days, and at 400° C. about the same quantity resulted after a week. On the other hand, mixtures of hydrogen and carbon dioxide gave no indication of change at 350° C. for a fortnight, or at 400° C. for a week.

Mixtures of fairly dry carbon monoxide and oxygen underwent no change between 300° and 400° C. although the moist gases very slowly interacted at 325° C. and upwards in the course of a week, yielding carbon dioxide. The reaction

(v) 2CO + O₂ = 2CO₂

could, therefore, like the preceding ones, have but little effect upon the course of the experiments.

(vi) 2H₂ + O₂ = 2H₂O.

This reaction has already been discussed. A temperature of 400° C. is on the border-line where the formation of water may be detected in the course of a week. The fact was also established that the following pairs of gases have no appreciable mutual action at or below 400° C., namely:

CH₄ + CO₂, CH₄ + H₂O, and CO + H₂.

The elimination of these secondary reactions from the slow combustion of methane greatly simplifies the problem.

B. Bone and Wheeler next ascertained that methane combines with oxygen between 300° and 400° C. with an enormously greater velocity than does hydrogen itself, and in no case were they ever able to detect the liberation of free hydrogen or free carbon as an intermediate or final product.

Since, if once formed, it would be impossible for them to be oxidised away in accordance with equations (ii), (iii), or (vi), their detection and isolation would be an easy matter, and hence it may be postulated that under normal conditions of slow combustion the methane is not first dissociated into its constituent elements. It is equally clear that the carbon monoxide and water which were always found when the supply of oxygen was insufficient to completely oxidise the methane, are two of the primary disintegration products of the partial oxidation of the methane molecule at these temperatures, for these are too low for reaction (vi) to take place.

C. An unexpectedly large proportion of carbon dioxide was invariably found in the gases at each stage, but especially towards the end of the oxidation. This cannot be explained on the supposition that the carbon monoxide first formed is oxidised to the dioxide, for reaction (v) takes place with extreme slowness at temperatures below 400° C. It would therefore result from the decomposition of some more complex oxygenated molecule.

D. The presence of formaldehyde could be detected as an unstable intermediate product during the slow combustion of methane at 450° to 500° C., and must be regarded as an oxidation product of methane, since it was not produced when mixtures of hydrogen and carbon q monoxide were circulated through the apparatus under analogous conditions. Formaldehyde readily decomposes when heated in the absence of air or oxygen yielding carbon monoxide and hydrogen; whilst in the presence of air, carbon dioxide and water result. Formaldehyde thus came to be regarded as the initial product of oxidation of methane, but upon the suggestion of Armstrong the view was finally adopted that methyl alcohol is the first product; this, however, cannot be detected on account of the rapidity with which it oxidised, presumably by hydroxylation to the hypothetical dihydroxymethane, which immediately decomposes to formaldehyde and water. This is followed by the formation of formic acid, which is readily detected amongst the intermediate products. Further oxidation yields carbonic acid, which dissociates into water and carbon dioxide. Thus:



One of the most interesting features of this scheme is the assumption that the first act of the oxygen atom is to associate itself with the methane; indeed, the oxygen atom appears to penetrate within the methane molecule, in much the same way as it is believed to enter the carbon network in the combustion of solid charcoal.

In the slow combustion of ethane, on the other hand, ethyl alcohol has actually been detected amongst the oxidation products, and an analogous scheme is suggested to that for methane. Thus:



Hydrogen, methane, and ethylene are also at times to be found amongst the products of oxidation, without, however, any carbon being liberated. Their appearance is believed to be due to the purely thermal decomposition of ethane, formaldehyde, and acetaldehyde. Thus:

C₂H₆ = C₂H₄ + H₂
H.CHO = H₂ + CO
CH₃.CHO = CH₄ + CO

In the case of ethylene, C₂H₄, it was not to be expected that vinyl alcohol could be detected as a transient intermediate product; but force of analogy leads to the assumption that the mechanism of the slow combustion is probably as follows:



Vinyl alcohol could not, of course, be experimentally detected among the products.

For acetylene, the following scheme is suggested:



It is by no means improbable that the rapid combustion of the hydrocarbon gases follows similar lines. Evidence on this point, however, is difficult to obtain.