Diffusion of Hydrogen through Hot Platinum. 29 
ee 
of molecules of hydrogen per c.c. outside, as, for instance, by 
supposing an action between the hydrogen molecule and the 
palladium, wherein one atom was taken up and the other 
rejected. _ A theory in which the palladium combined witha 
molecule of hydrogen, and then detached the atoms separately, 
would yield results similar to those given by the theory here 
adopted, and therefore probably explain the results. In fact, 
such a theory would be thermodynamically equivalent to the 
one here given. 
On this hypothesis a formula for the number of gramme- 
molecules of gas diffusing per c.c. per second, has previously 
been found by one of the authors. In the case where the 
diffusion of the undissociated gas is very small and the 
dissociated fraction of the external gas is negligible, this 
7 becomes 
d 2 Je y 
where d is the thickness of the walls, yw: the coefficient of 
diffusion of the atoms through the platinum, 4, the dis- 
sociation constant of the hydrogen inside the platinum, A, 
the solubility of molecular hydrogen in platinum, and P the 
external concentration of the gas at 0° C. under the pressure 
in the experiments. i? is shown to be =C@%e~*"", where C is 
a constant, @ the absolute temperature, and gq, the heat of 
dissociation of the hydrogen inside the platinum. 3 
A mean of 15 concordant determinations of the heat of 
dissociation (q,) of hydrogen inside platinum gives 36,500 
calories. For the mean value of the constant w.C/A® we find 
859x107". Putting in these numbers, the mass Q of gas 
diffusing per sec. through each square cm. of a platinum disk 
d ems. thick when the pressure on one side is 7 ems. of 
mercury and on the other zero, the absolute temperature 
being @, is given by 
9°125 
Q=6°60 x 10-°7 Oe Q. 
This formula has been shown to hold with considerable 
accuracy over a range of pressure from 0:1 to 76°0 cms, and 
of temperature from 576° C. to 1176° C. 
Finally the experiments point to the view that the 
ionization produced by hot platinum in an atmosphere of 
hydrogen is due to collisions taking place between the atoms 
of hydrogen dissolved in the platinum. 
In conclusion, the authors wish to record their gratitude 
to Professor Thomson for his advice and suggestions during 
the course of this work, which was carried out in the 
Cavendish Laboratory. 
