378 TRANSACTIONS OF SECTION B. 



The principal experimental facts which this, or indeed any alternative, 

 scheme must explain are as follows : 



(1) That whenever mixtures of composition between 2CH^ + 0j and 

 CHj + Oj are exploded under pressure a considerable proportion of the original 

 oxygen appears as steam in the products ; 



(2) That there is a marked minimum in the proportion of such oxygen as the 

 composition of the original mixture approximates to 3CH^+202; and 



(3) That there is a total cessation of anj' separation of carbon (which is very 

 marked with mixtures ^CK^ + O^) after the proportion of oxygen in the original 

 mixture attains or exceeds the limit 3CHj+202. 



Now, if, as I believe, the initial interaction of methane and oxygen is at all 

 temperatures essentially a ' hydroxylation ' process, accompanied by the decom- 

 position (more or less rapid according to the temperature) of the primarily 

 formed hydroxylated molecules, a consideration of the chemical and thermal 

 aspects of the process will point to certain probabilities which are indeed 

 actually realised in fact. 



In the first place, monohydroxymethane (methyl alcohol) CH3.OH is known 

 to decompose at high temperatures, yielding carbon monoxide and oxygen, 

 without any separation of carbon or formation of steam, CH3.0H = CO-f2H2. 

 Also, the very unstable dihydroxy methane, H, : C : (0H)„ would yield formal- 

 dehyde and steam, Hj : C : (0H)2 = H, : (J :0 + H„0, and the formaldehyde would 

 in turn decompose into carbon monoxide and hydrogen, Hj : C : : = C0-^H2, 

 without any deposition of carbon whatever. 



If now the thermal consequences of such facts be considered, it would 

 appear (1) that if the oxidat:on of methane to H, :C.OH (A to B in the 

 scheme) were accompanied by thermal decomposition at this stage (B to B'), the 

 net heat evolution would be (30 — 22-8) = 7-2 kilogram Centigrade units; whereas 

 (2) if the same amount of oxygen reacted in such a way that there was a ' non- 

 stop ' run through the monoliydroxy- to the dihydroxy- stage, with decomposi- 

 tion at the point (A to C and C to C), the corresponding net heat evolution 

 would be ^(30-t-59 — 13'4)=37"8 units, or about five times as much as in (1). 

 Hence there would always be a strong tendency for such a non-stop run from 

 A to C through B, without any decomposition occurring at B, and such would 

 always occur whenever the oxygen present in the original mixture attained the 

 equimolecular proportion CH^-fO,. 



Again, if the original mixture contained only half such proportion of oxygen 

 (2CH^-|-0,,), there would still be a decided preference for an oxidation of one- 

 half of the methane by a non-stop run A to C through B, rather than an 

 oxidation of the whole of the methane to B only, the other half of the methane 

 remaining unchanged, or undergoing thermal decomposition into its elements, 

 CH4 = C-t-2H2 (A to A'). Also, the latter process would use up no more of the, 

 energy developed from the oxidation than would be required for the decomposi- 

 tion of a corresponding quantity of H, : C.OH at stage B (B to B'). Hence 

 when such a mixture, 2CH,,-f02, is exploded under pressure, the formation of 

 carbon and its oxides, hydrogen, and considerable quantities of steam may be 

 expected, which is what actually occurs. 



When, however, the proportion of oxygen in the original mixture reaches 

 the limit 3CH.,-f202, whilst it is still insufficient to oxidise the whole of the 

 hydrocarbon ib the dihydroxy- stage, there is enough of it to prevent any 

 methane remaining unoxidised to (at least) the monohydroxy- stage, and, there- 

 fore, seeing that the affinity of methane for oxygen far exceeds those of either 

 hydrogen or carbon monoxide, it is to be expected that no substantial pro- 

 portion of the original methane would escape oxidation to either the mono- or 

 the dihydroxy- stage. But inasmuch as not more than about one-third of the 

 original methane could, in the circumstances, be transformed into the 

 di-hj'droxy- stage, it follows that a considerable amount of thermal decomposi- 

 tion at the mono-hydroxy- stage would occur. 



If this view is correct, it follows that there should be an entire suppression 

 of carbon deposition at or about the 3CH., + 20„ ratio, and, also, that with this 



