1918-19.] The Adsorption Isotherm at Low Concentrations. 49 
If, however, the (log p, log a) curve for any individual temperature be 
examined it will be found to be distinctly concave to the log p axis, the 
gradient increasing as a decreases. This is shown in the second table, 
Table II. 
2 
n 
t 
p = 1 cm. 
10 
20 
- 78°C. 
0-22 
0T2 
O'll 
0° 
'39 
'37 
'31 
which is calculated from Travers’ observations. This change in the value 
of i- is also clearly seen in the values tabulated by Titoff* and by 
Richardson,]* and is found in practically every published case of gaseous 
adsorption. 
Hydrogen is of the gases ordinarily examined the least adsorbed at any 
given temperature, and hence the gas most likely to furnish evidence as to 
the nature of the adsorption isotherm in the region of low values of a. As 
has already been mentioned, with low pressures the value of in this case 
appears to be unity at ordinary temperatures, or the adsorption isotherm is 
represented by the simple formula 
a = a Q p, 
which is the same as Henry’s Law. Even with hydrogen, as the pressure 
increases there is a distinct fall in the value of ^ • With other gases a is 
small in the region of ordinarily measured pressures only when the 
temperature is high, and here again we find that with low pressures 
the value of the exponent appears to be unity. Since the value of — or 
0 1 
^ t ° at different temperatures varies much less with a constant than with 
oiogp 1 
p, and at lower temperatures all values rise towards unity as « falls, it 
seems highly probable that at all temperatures the adsorption isotherm 
assumes the simple form 
a = «o P 
for small values of a. The fact that the value of the exponent appears in 
* Zeits. f. physik. Chem. (1910), lxxiv, p. 641. 
f Journ. Airier. Chem. Soc. (1917), xxxix, p. 1828. 
VOL. XXXIX. 
4 
