HYDROGEN AND OXYGEN IN CONTACT WITH HOT SURFACES. 
67 
of the surface ; at the end of the sixth hour the ratio H 2 /0 2 for the gases was 1'823, 
indicating incipient reduction, and for the residual gas at the end of the experiment 
the ratio was 1'469. In this connection it may be pointed out that, so long as there 
was no appreciable permanent reduction of the surface (say up to about the end 
of the fourth or fifth hour), the values “ k x ” remained nearly constant, whereas at 
lower pressures, when reduction did set in, they rapidly increased. The same remark 
applies also to Experiment VIII. with the “ oxidised ” gauze. 
Considering the results obtained with the three different reducible oxides as a whole, 
they seem to be quite incompatible with any purely chemical explanation of the 
catalytic process. In the case of copper oxide, it would appear to involve the 
condensation of a film of “ active ” oxygen on the surface, and that this film actually 
protects the catalysing oxide from the attacks of the hydrogen, which would other¬ 
wise energetically reduce it. At low pressures the film becomes too attenuated to 
ensure complete protection, and, in consequence, the formation of steam is accelerated 
by reason of hydrogen penetrating through on to the oxide and reducing it. In 
conformity with this idea, the rate of steam formation when the gases are not 
present in combining ratios was found to be proportional to the partial pressure 
of the oxygen. 
In the cases of the other reducible oxides examined, the fact that the gases 
combined at moderate temperatures, without permanently altering the surface, and at 
rates always far in excess of the rates at which either the surface was reduced by 
hydrogen, or the reduced surface re-oxidised, does not harmonise with the purely 
chemical theory. It is difficult to resist the conclusion that in these cases also the 
catalytic action is primarily due to a physical condensation of one or other of the 
reacting gases at the surface. And since in all experiments where the gases were 
not present in combining ratios the rate of combination was always proportional to 
the partial pressure of the hydrogen, it would appear to be the hydrogen which is 
condensed and so rendered active. 
The authors desire to express their thanks to the Government Grant Committee 
for grants towards the expenses of the research, and also to Messrs. Johnson, 
Matthey & Co. for the loan of gold and platinum gauze. 
■,s 
‘ 
4uV:. wob 
K 2 
