534 



contrast witli Fkeündi.ich's experience that the absorption ufc/m-o/wrf 

 .tuhntnnces, as iodine, djestuffs, and organic acids, is an equilihrinm, 

 which is readily established independent of the condition from which 

 one starts, and within a few minutes ; this observation of P^kecnot.ich's 

 was confirmed for dissolved substances by many investigators. 



In order to obviate the influence of hysteresis, the equilibiium 

 had to be determined from two sides; then the approximative value 

 of the state ctf equilil)rium was calculated by taking the mean of 

 the two values found in this way. Accordingly twice thirteen samples 

 of airdry carbon, each having a weight of about one gramme, 

 were weighed off in crystallisation dishes. One half of these dishes 

 were dried for one or two weeks in a vacuum exsiccator over 

 sulphuric acid ; they then contained no more than 1 or 2 parts of 

 water to 100 parts by weight of dry carbon. The other half was 

 placed over water in a vacuum exsiccator for the same length of 

 time; they then contained about 90 parts of water to 100 parts of 

 dry carbon. Then thirteen small exsiccators were arranged with 

 sulphuiic acid-water mixtures of known vapctur tension ; in every 

 exsiccator there was placed a dry and a moistened carbon. These 

 acids were refreshed a few times. After 40 — 90 days, when the 

 dishes had become almost quite constant of weight long before, it 

 was assumed that ihey had reached their onesided equilibrium. 

 All the experiments took place at a temperature of 16 — 20° C. in 

 a room in which the variations of temperature were particularly 

 small (a room built specially for thermochemistry). 



The vapour tension h was expressed as fraction of the maximum 



tension of water at the same temperature; the sorbed quantity i. as 



grammes of water per one gramme of dry carbon. The free energy 



at the sorption of one gramme of liquid water is found from the 



- 1252 , 

 relation A = —-— loa^' h. 

 18 ^ 



Fig. 2 shows the isotherm. The curve begins as a real adsorption- 

 curve (or as the isotherm of a concentrated solution), but with a 

 very short horizontal initial portion '), at half its height, {h = 0,40 

 to 0,65) it gets, however, an almost horizontal part; at A = 0,65 

 and / = 0,57 there begins a new part of the curve (which, however, 

 issues from the preceding part without any abrupt transition), which 

 again has an S-shape. It is remarkable how great the quantity 



') This has probably been drawn too long ; has the weight of the carbon not 

 been somewhat diminished by drying at 200° G. through oxidation? The horizontal 

 beginning, if it exists, is probably only little pronounced. 



I 



