438 ^Professor Sir James Dewar [June 8, 



LaiV8 of AbsorpUon at Loiv I'eDqmratures. 



The general laws of the dependence of the absorptive power of 

 charcoal upon the pressure, temperature, and volume of the gas 

 absorbed are apparently complicated, but in average circumstances 

 they can be stated approximately. For a supply of gas at constant 

 pressure the volume and temperature are related by a curve of hyper- 

 bolic form. There must clearly be a lower limit of temperature for 

 absorption, for the gas will eventually liquefy ; on the other hand, 

 however small the absorption may be, as the temperature rises it can 

 never entirely vanish. If the temperature be kept constant we 

 should expect an increased absorption with increasing pressure up 

 to a certain limit. Experiment indicates that this is the case, and 

 we have again a curve of hyperbolic form connecting pressure and 

 volume, but differing from the previous case in having the convexity 

 turned the opposite way. When the pressure is small the absorption 

 will be a minimum, but any slight increase of pressure will be 

 associated with the molecular attraction of the charcoal, and the 

 absorption may be expected to increase much more rapidly in pro- 

 portion to the pressure. Finally, when the volume absorbed is kept 

 constant the relation between the pressure and the temperature is'of a 

 logarithmic form, like those of a saturated vapour or a dissociating 

 body. 



For the absorption of hydrogen in charcoal at the temperature 

 of liquid air the expression \ogp = a + hN - cN- holds, where 'p is 

 pressure, V the volume of gas occluded, and a, b, and c constants. 

 For small concentrations of gas the pressure grows in a linear relation 

 to the volume. 



Hypothetical Densities of Gases Occluded in Charcoal. 



Mitscherlich's measurements made on the charcoals used by him 

 showed that in a piece of charcoal the space occupied by charcoal- 

 substance is to the pore-space nearly in the ratio of 13 to 20. The 

 Professor's observations on the real and apparent densities of cocoanut 

 charcoal lead to the conclusion that in 100 grammes the pore-space 

 may be taken on the average as about 15 c.c. This enables us to 

 compare the average hypothetical density of any absorbed gas with 

 the density of the same gas when liquefied. The following table 

 shows some such comparisons, and it may be noted that the densities 

 of the absorbed gases at their boiling-points are equal to or a little 

 greater than the corresponding liquefied gases at the same tempera- 

 tures, so far as they are known. Whether the density of the 

 monadic gases will turn out as satisfactory is a matter for future 

 inquiry. 



