Page:Outlines of Physical Chemistry - 1899.djvu/36

 16 OUTLINES OF PHYSICAL CHEMISTRY

the vapour density experimentally to be 0*89, which corre- sponds to the molecular weight 25*69.

In order to bring these facts into harmony with Avogadro's hypothesis and with the admitted atomic weights, the supposition has been made that during volatilisation ammonium chloride is split up and the vapour consists of a mixture of molecules of hydrochloric acid and ammonia. With this supposition all difficulties disappear.

We know quite a number of substances which experi- mentally give too low vapour densities. Amongst these are: ammonium carbamate, phosphorus pentachloride, antimony pentachloride and pentabromide, sulphuric acid, nitrogen peroxide, chloral hydrate, merourous chloride, &c.

The vapour density of all these substances is irregular because their vaporisation is accompanied by a more or less complete decomposition, which advances as the temperature rises.

Pebal has demonstrated (by his classical experiment) that ammonium chloride is actually decomposed on vapori- sation, and by means of their different velocities of diffusion through an asbestos plug he was able to prove the presence of the dissociation products.

In the case of phosphorus pentachloride the greenish colour of the vapour indicates the presence of free chlorine. By volatilising this substance at as low a temperature as possible in an atmosphere of phosphorus trichloride, Cahours found a normal vapour density, corresponding to the formula PC1 5. This formula is further confirmed by the existence of phosphorus pentafluoride PF 5, which is volatile without decomposition.

Besides, in almost every case, the experimental vapour density of a dissociable substance does not agree either with the simple formula or with the halved formula, but has an intermediate value, demonstrating the presence of non-dissociated molecules. These have, therefore, a real existence, and if they do not constitute the whole or even

�� �