Carbonisation Reactions 1

Abstract

IT is difficult to obtain clear information about the reactions in carbonisation by the direct distillation of coal in the laboratory, especially about minor, though important, products such as toluene, benzene, etc., of which the quantities available become excessively small. Therefore the author, in conjunction with Dr. S. F. Dufton, chose to investigate the stability of individual compounds in different atmospheres when passed over heated coke, with a time of contact of the same order as met with in car bonising practice. The behaviour of the compounds was dependent on conditions of temperature and concentration, apparently in accordance with the laws of chemical dynamics. Benzene, at varying partial pres sures in an atmosphere of nitrogen, which was assumed to be inert, showed signs of incipient decomposition at 550° C., leading to the production of diphenyl. At higher temperatures this was more extensive, and p-diphenylbenzene also appeared. The former condensation at least is reversible. Benzene in an atmo sphere of hydrogen yielded much less diphenyl under otherwise similar conditions, and at 800° C. is scarcely decomposed at a concentration of 5 per cent, by volume, while diphenyl in an atmosphere of hydrogen is reduced to benzene. This doubtless accounts for the practical non-occurrence of diphenyl in coal-tar, although this substance is readily formed from benzene in the absence of hydrogen. Benzene did not yield naphthalene or unsaturated compounds. At 900° C., even in hydrogen, benzene was extensively broken down with the formation of carbon. Toluene in nitrogen shows signs of decomposition at 550°. C., but there are now two possibilities—reactions through the nucleus and also through the side chain. The product is more complex, a solid being formed, which was identified as stilbene (CH.C6H5)2, and also an oil. At 750° C. decomposition was more extensive, naphthalene and anthracene being identified among a number of other products. Oh substituting hydrogen for nitrogen the decomposition was much accelerated, but with the production of benzene and methane and smaller quantities of solids. The formation of stilbene, which occurs with the liberation of hydrogen, is inhibited. Methane and benzene react in the reverse direction to form toluene, but only very slowly. Thus hydrogen pro tects benzene from decomposition, whereas it decom poses toluene, but in the sense of breaking off the side chain while hindering molecular condensations with elimination of hydrogen, which are a characteristic effect of heat on the lower aromatic hydrocarbons. The xylenes resemble toluene in behaviour, while cresol is reduced by hydrogen at 750° C. to toluene, and necessarily to benzene also. The importance of atmosphere in carbonisation is therefore great, and can influence the course of the process. This is seen in practice in the results of steaming charges in vertical retorts. The hydrogen of the water-gas formed doubtless operates in the sense shown above, as well as physically in sweeping out the yapours before decom position can have gone too far. The lightness of the hydrogen molecules and consequent high diffusivity doubtless promote this physical effect, but the pre dominant influence is chemical. Whether it is better to introduce the water-gas from the outside or to generate it in situ by steaming remains to be shown. In any case, there is an advantage to be gained by the admixture of water-gas in that the proportion of the heat of the coal in the gas is increased. When steam is introduced into the retort, reactions which lead to the formation of ammonia, as in the Mond gas producer, are called into operation. It is steam which promotes the formation of ammonia, and not hydrogen, on which undue stress has been laid since the time of Tervet. Experiments made in conjunction with C. A. King show that steam alone among a number of gases tried had any marked effect in the production of ammonia from coke at 800° C. Characteristic of the results of steaming charges in recent experience is the much increased yield of ammonia, and also of tar, which contains less of the heavy complex condensation products than ordinary horizontal-retort tars. This is important in connection with the carbonisation of coal as a possible source of fuel-oil. In considering the effect of heat on hydrocarbons, there is evident a ten dency for molecules to coalesce with the production of more and more complex six-atom carbon-ring compounds, with the elimination of hydrogen, which operates in the reverse sense. The stability of the six-atom ring structure seems characteristic of carbon even under the most drastic treatment, and it is interesting to note that the investigations of the Braggs show that the same orientation can still be detected in the diamond and grarjhite.