1341 



take place in vacuum. Hence the appearance of this gas-ahsorption, 

 and the pressure of dissociation which is not negligible at — 80° C. 

 are the cause that the compound cannot be prepared in pure stale 

 in the waj- described above. These two phenomena have, however, no 

 disturbing intluence on the determination of tlie three-phase pressures, 

 at least when Ihe gas adsorption does not attain to too hirge amounts. 



It is namely clear from the described phenomena that the gas- 

 adsorption is no phenomenon of equilibrium ; at higher temperatures, 

 where the equilibria set in more rapidly, this adsorption soon stops; 

 and the liberated sulphuretted iiydi-ogen cannot influence the three- 

 phase pressure (the three-phase equilibria are monovariant). Nor can 

 a partial decomposition of tlie hydrate have any influence on the 

 pressure of equilibrium. 



Nevertheless a ditlficulty presents itself in these determinations. 

 When the compound iiad absorbed no gas, a gas pressure could 

 only appear on heating through decomposition of the compound into 

 ice and gas; the observation of a gas pressure would then be a 

 sufKicient criterion that a tliree-|)hase pressure existed. Inconsequence 

 of the said adsorption it is, howevei-, possible that with increased 

 temperature through liberation of gas a gas pressure occurs without 

 any decoinposition of the compound taking place. As the transfor- 

 mation of part of the compound into ice and gas cannot be directly 

 observed, the possibility exists that pressures of two-phase equilibria 

 compound-gas are measured instead of three-phase equilibria. It now 

 appeared in the determinations that at lower temperatures no cor- 

 I'esponding values were found for the pressure (the two-phase coexis- 

 tences are divariant); at higher temperature the correspondence 

 became, however, very good. This can, evidently, be accounted for in 

 this way that the solid substance had absorbed diffei'enl quantities 

 of sulphuretted hydrogen, which were liberated on increase of tem- 

 perature. When this released quantity of gas yields a pressure higher 

 than the three-phase pressure, the decomposition of the compound 

 cannot begin. Not until througli increase of temperature the three- 

 phase pressure rises so much that this becomes greater than the 

 pressure caused by the liberated adsorbed gas, thiee-phase pres- 

 sures occur. On account of the different adsorbed quantities this 

 takes place for the different mixtures at different tenjperatures. In 

 none of the experiments did the adsorption prove to be so great 

 that the determinations of the three-phase pressures were rendered 

 impossible. As was natural only the values in the neighbourhood 

 of the quadi'uple point, where the correspondence was good, were 

 used for the calculations. (See H ^ ^'^<^' ^)- 



87 



Proceedings Royal Acad. Amsterdam. Vol. XXI. 



