OCCLUSION OF OXYGEN AND HYDROGEN BY PLATINUM BLACK. 
131 
determining the heat of occlusion is in connexion with a statement of Berthelot’s,^'' 
that two definite compounds of platinum and hydrogen exist, viz., PtsoHj and PlggHg, 
corresponding to the amount of hydrogen which can be extracted from platinum black 
in vacuo at two different temperatures and to two different heats of occlusion. 
Favre,! also, found on admitting hydrogen fractionally, in small portions at a time, 
that the heat evolved per gram of hydrogen occluded became less and less. As our 
results will show, the heat evolved per gram of hydrogen occluded for the fraction of 
hydrogen which can be removed at the ordinary temperature, by means of the pump, 
from platinum fully charged with hydrogen at atmospheric pressure, is the same as 
for the jiortion which can be extracted in vacuo at 184° C., and consequently we 
believe that the results obtained by Berthelot and by Favre were due to the fact 
that the samples of platinum black examined by them contained oxygen. 
Having stated in general terms the method adopted by us for the determination of 
the heat of occlusion of hj^drogen by platinum black, we will now describe the 
apparatus employed. 
During the first part of this investigation we found that although the bulk of the 
hydrogen or oxygen occluded by platinum black was absorbed almost immediately, 
a slow absorption went on for hours. This being the case, we found it better to make 
use of a Bunsen’s ice calorimeter instead of an ordinary water calorimeter, which 
gives the best results when the reaction is nearly instantaneous. We are of opinion 
that the ice calorimeter has been too often neglected on account of the difficulty of 
obtaining pure snow ; but, by adopting the device suggested by Professor C. V. 
Boys,| we have been able to obtain very satisfactory results. The calorimeter itself, 
L, which was made sufficiently large to accommodate the experimental tube D, was 
surrounded by an air-jacket, J, as shown in fig. 1, and the whole suspended inside 
three concentric cylinders. The cylinders themselves, with their drain-pipes, were 
sunk in a large cubical wooden box, the space between the outer cylinder and the 
box being packed with cotton wool. Pounded Norwegian ice was placed inside the 
cylinders and heaped up outside, so that the whole of the projecting part of the 
experimental tube was covered with melting ice. The function of the air-jacket is to 
diminish the too rapid transference of heat between the calorimeter, which should be 
exactly at 0° C., and the melting ice, which, since it is more or less impure, has 
always a lower melting-point than pure ice. The calorimeter was filled in the usual 
way with boiled-out distilled water, and, after some mercury had been introduced 
into the lower part, the thistle funnel, tap, and side capillary tube were sealed on. 
In order to produce the sheath of ice on the outside of the inner tube of the calori¬ 
meter, the whole was cooled down to 0° C., and some solid carbon dioxide introduced. 
This produced intense local undercooling, and consequently some ice crystals 
* Berthelot, loc. cit. 
t Fatre, loc. cit. 
+ C. V. Boys, ‘Phil. Mag.,’ 1887, vol. 24, p. 214. 
