HYDROGEN AND OXYGEN IN CONTACT WITH HOT SURFACES. 
5 
In 1903 Bobenstehsp published a second paper dealing with the combination of 
hydrogen and oxygen in contact with a catalysing surface of platinum at the ordinary 
temperature. The experiments described are divided into two series. In the first 
series, the platinum surface was kept at the same temperature as the surrounding 
gas, so that a film of water formed on the surface. Under these conditions, the 
velocity of the reaction was found to be approximately proportional to the pressure of 
the electrolytic gas present, any excess of either hydrogen or oxygen merely acting 
as a diluent. This was explained on the “ diffusion ” theory by assuming an infinitely 
great rate of combination at the surface, and that therefore the observed rate depends 
on the rate of solution of oxygen, or on that of hydrogen when the gas contains an 
excess of oxygen, in the water film. But whilst the results of the published experi¬ 
ments support this theory, the author admits that others (unpublished) did not.'!' 
In the second series of experiments, the surface of the metal was kept dry by 
maintaining it at a slightly higher temperature than the surrounding gas. Here the 
absolute rate of combination was much greater than in the first series of experiments, 
and it no longer, even approximately, conformed to an equation of the first order. In 
one experiment, for instance, where “normal’ electrolytic gas was employed, the 
“velocity constant,” (£ = -log^j, increased from 0*038 to 0*117, or by more than 
200 per cent., as the pressure in the apparatus fell from 750 to 24*5 millims. in 14 
minutes. In other experiments the values of “ Jc” similarly increased by 100 per cent. 
In connection with their researches on the mechanism of hydrocarbon combustion, 
the authors devised a circulation apparatus which is well adapted for measuring the 
velocities of catalytic gas reactions, where the product can be quickly removed from 
the sphere of action either by condensation, or by solution in water or other absorbing 
liquid. It was therefore decided to make a systematic study of the case of hydrogen 
and oxygen, in the hope of obtaining sufficiently reliable data to discriminate between 
the various possible explanations of the action of surface in inducing gaseous reactions. 
And, in order to avoid the danger of errors arising from a too restricted view of the 
phenomena, it was decided to examine in detail the action of a great variety of 
surfaces, including porous porcelain, magnesia, platinum, gold, silver, nickel, copper, 
and certain easily reducible metallic oxides, such as copper oxide, nickel oxide, and a 
mixture of ferric and manganous oxides obtained by calcining spathic iron ore. 
Bartya was also examined, but with less satisfactory results, owing to constant 
variations in its catalysing power, which seriously affected the velocity measurements. 
In the case of an oxidisable metal, or a reducible oxide, it seems necessary to 
distinguish in principle between the purely “catalytic” combination of the gases 
* ‘ Zeit. Phys. Chem.,’ 1903, vol. 46, p. 725. 
t In this connection he remarks, “ bei alien spateren (Versuchen) war tier Anschluss an die zu Grunde 
gelegte Gleichung ein merklich schlechterer, und eine gauze Anzahl Beobachtungen wurden gemacht, die 
zu der Annahme der unendlich grossen Verbrennungsgeschwindigkeit nicht stimmen . . . .” (p. 737). 
