Vol. 6, 1920 
PHYSICS: L. B. LOEB 
109 
ture. In this case the deflection was not instantaneous but the tempera- 
ture rose more gradually, reaching a maximum in about 45 seconds. After 
this it slowly fell as the alcohol and air were consumed in the immediate 
neighborhood of the platinum. 
When air alone was admitted to the apparatus no appreciable rise in 
temperature was detected. 
It is obvious that the heat liberated on the surface of platinum black 
due to adsorption alone either of air or of alcohol vapor is small compared 
to that liberated in the process of oxidation of the alcohol at the surface 
of the platinum. In fact, adsorption gives at most only 1% of the heat 
produced in these experiments. 
It will be of interest to know whether this conclusion is also supported 
by other facts. 
On the basis of a recent theory of Langmuir^ that the layer of molecules 
of a gas adsorbed on a solid surface is never more than one molecule thick, 
it is possible with certain assumptions to calculate the rise in temperature 
of the platinum black. If one considers that the heat of adsorption is 
equal to the heat of condensation, if one neglects heat losses due to radia- 
tion, conduction, and convection, and if one assumes a definite radius 
for the particles of the platinum black the calculation is quite simple. 
Let us assume that the platinum black is made up of aggregates of spheres 
of platinum whose diameter is 100 ju ju, or 1 X 10 ~^ cm. This is about 
the size of the colloidal gold particles of a gold suspension investigated by 
Zsigmondy.2 Such an assumption is justifiable as the particles of platinum 
black used are certainly much larger than this. 
Using 0.41 gram of platinum black we get 
The number particles of diameter 1 X 10~^cm. in 0.41 gram platinum 
black = 3.75 X lO^o. 
The total surface of all the particles = 1.18 X 10^ sq. cm. 
On Langmuir's theory there should be = 0.85 X 10^^ molecules of CH3OH 
per square cm. surface in an adsorbed layer one molecule thick. 
Total number of molecules of CH3OH condensed on the surface of the 
platinum black = 1 X 10^^ molecules. 
Heat generated by these molecules on condensation equals heat of 
evaporation of the 10^^ molecules = 1.53 X 10 ~^ calories. 
The 0.41 gram of platinum will be raised 1.16° C. at 0° C. by this ad- 
sorption. 
If the amount adsorbed were oxidized the temperature would be raised 
19.8° C. 
It is obvious that the relative values calculated on these assumptions 
are in approximate agreement with the values experimentally obtained. 
The surface of the platinum black is in reality much smaller than was 
assumed here. Furthermore the heat liberated by oxidation is continu- 
ously liberated as long as alcohol and oxygen are present, while in the case 
