vapour-density of ammoniura chloride to be 26.75, then the compound having the composition NH4Cl would have the normal molecular volume, that is, its molecule would occupy two unit volumes, and the conclusion would be that the vapour consisted of single molecules of the composition represented by the formula NH4Cl. But ammonium chloride at ordinary temperatures is a solid, and when heated to the temperature necessary to convert it into vapour its molecules break up into separated molecules of the two original gases-ammonia, NH3, and hydrochloric acid, HCl.+ So that we are unable to gain any information in this direction as to the mode in which the atoms are disposed in the compound. When the two gases are brought together under ordinary conditions, they combine with the evolution of considerable heat, owing to loss of energy; this is taken as evidence that true chemical action, in the sense of atomic rearrangement, has resulted, hence it is believed that in this compound the nitrogen is united with the five monovalent atoms, and consequently is pentavalent.

In the case of carbonyl chloride, COCl2, the vapour-density can be ascertained, this compound existing in the gaseous condition at the ordinary temperature. Its vapour-density, determined by experiment, is found to be 50.6. This number, divided into the molecular weight of the compound having the composition COCI gives practically the number 2 as the molecular volume of the compound. Hence we conclude that these four atoms constitute a single molecule.

There are a number of combinations, however, in which molecules of different compounds unite, that do not so readily admit of explanation, because in neither of the molecules is there any atom functioning in a lower state of valency than that which it is known to be capable of. For example, the monovalent elements fluorine and hydrogen form the compound hydrofluoric acid, HF; fluorine also combines with the monovalent element potassium, forming potassium fluoride, KF. Both of these compounds come under the head of saturated compounds, in the sense that neither of them contains an atom which is known to be capable of exercising a higher valency than it exhibits in these compounds. Nevertheless these two molecules unite together and form a definite chemical compound, known as hydrogen-potassium fluoride.

Again, the divalent element zinc combines with two atoms of † See Dissociation, p. 89.

* See p. 43.

E

the monad element chlorine, forming zinc chloride, ZnCl2; the two monovalent elements sodium and chlorine also combine, giving the compound sodium chloride, NaCl. Both of these substances must be regarded as saturated compounds, and yet they unite with each other, forming a distinct chemical compound, known as sodium zinc chloride. Such compounds as these are known as double salts, and examples might be multiplied almost indefinitely. A similar union of molecules, where the recognised valency of the atoms is all satisfied, is seen in a large number of compounds containing water of crystallisation ;* for example, the divalent element copper, in combination with two atoms of chlorine, forms cupric chloride, CuCl. The divalent element oxygen, in combination with two hydrogen atoms, forms water, H2O. When cupric chloride crystallises from aqueous solution, each molecule of the chloride unites to itself two molecules of water, which is therefore termed water of crystallisation.

In chemical notation, it is usual to represent compounds of this order by placing the formulæ of the different molecules that have entered into union in juxtaposition, with a comma between ; accordingly, the examples here quoted would be indicated thus

Hydrogen potassium fluoride

Sodium zinc chloride

Crystallised cupric chloride

HF,KF.
ZnCl2, NaCl.
CuCl, 2H,O.

Combinations of this order are by no means confined to the union of two kinds of molecules, as the following examples will serve to show:

Platinum sodium chloride

Mercuric potassium chloride

PtC14,2NaCl,6H,O.
2HgCl2, KCl, 2H,O.

At the present time our knowledge of the nature of the union between these various molecules is too imperfect to admit of any precise explanation; such compounds are frequently distinguished as molecular combinations.

It must be remembered that our ideas of valency are based mainly upon the consideration of matter in the gaseous state; at present we have little certain knowledge as to the valency of elements in liquid and solid compounds. Most of the compounds belonging to the class we are now discussing are solid, and

* See p. 216.

it is quite conceivable that the valency of an element might increase as the compounds in which it functions pass from the gaseous to the liquid and solid

state.

There is also another consideration that must not be overlooked. The unit of measure that has been adopted for estimating valency, namely, I monovalent atom, is probably only an extremely rough and crude measure, which is incapable of appreciating smaller differences of combining capacity that may, and most probably do, exist. Its use may be compared to the adoption of a single unit, say 1 gramme, for the estimation of mass or weight; when, if a given quantity of matter has a weight equal to 1 gramme, but less than 2 grammes, its weight would be 1; if greater than 2 grammes, but less than 3, then its weight would be 2-a method of estimating which tacitly assumes that no intermediate weights of matter between the various multiples of the selected unit are possible. There is no evidence to show that the combining capacity of an element is exactly expressed by simple multiples of a monovalent atom.

For example, in the simplest form of combination, such as that between hydrogen and chlorine-where the molecule contains 1 atom of each element – 1 hydrogen atom unites with 1 chlorine atom, that is to say, with a mass of chlorine weighing 35.5 times its own weight; and we say that the mutual affinities of these atoms are satisfied. But for anything we know to the contrary, an atom of hydrogen may have an affinity for chlorine which would enable it to unite with a mass of chlorine weighing 40 or 45 or 50 times its own weight, but not a mass weighing 71 (35.5 × 2) times its own. But since a mass of chlorine 35.5 times the weight of a hydrogen atom is the smallest quantity that is ever known to take part in a chemical change, is the chemically indivisible mass we call an atom, it follows that as the hydrogen atom has not sufficient combining capacity to unite with 2 atoms, it is compelled to be satisfied with 1. It might still, however, retain a residual combining capacity. Or the residual combining capacity may be lodged in the chlorine atom, which may be conceived as being able to unite with a greater weight of hydrogen than is represented by 1 atom, but not so much as that of 2 atoms.

Each of the elements fluorine, chlorine, bromine, and iodine unites with I atom of hydrogen, and we represent their compounds in a similar manner, thus

but we make an enormous assumption if we suppose that in each of these compounds the mutual affinities of the atoms is equally satisfied.

For example, the fluorine compound exhibits a tendency to unite itself to other compounds of fluorire (and to a much more marked degree than is seen in the case of hydrochloric acid), resulting in the formation of such double fluorides as the following:

Hydrogen potassium fluoride

Hydrogen bismuth fluoride.

KF, HF.
BiF3,3HF.

Hydrogen silicon fluoride (Hydro-fluo-silicic acid) SiF4,2HF.

And there is reason for believing that the molecules of hydrofluoric acid

are capable of uniting amongst themselves, forming the more complex molecules HF2 or H3F3 or HnFn.*

Assuming the residual combining power to reside in the fluorine atom, and representing this by means of dotted lines, we may express the composition of these compounds thus

CHAPTER IX

GENERAL PROPERTIES OF GASES

UNDER the head of the general properties of gases it will be convenient to consider the following subjects : *—

1. The relation of gases to heat.

2. The relation of gases to pressure.

3. The liquefaction of gases.

4. Diffusion of gases.

5. The kinetic theory of gases.

The Relation of Gases to Heat.-The fact that substances expand when heated, and again contract upon being cooled, was observed in very early times. The fact also that all substances do not undergo the same alterations in volume when subjected to the same changes of temperature has been long known; but it was not until the beginning of the nineteenth century that it was proved by Charles and Gay-Lussac that all gases expanded and contracted equally when exposed to the same alterations of temperature. This law is generally known as the Law of Charles, and may be thus stated: When a gas is heated, the pressure being constant, it increases in volume to the same extent whatever the gas may be.

The increase in bulk suffered by 1 volume of a gas in being heated from o° to 1° is termed the coefficient of expansion, and if the law of Charles is true all gases will have the same coefficient.

Modern research has shown that the law of Charles is not absolutely true, and the extent to which gases deviate from the strict expression will be seen from the coefficients of expansion given in the following table :—

* The study of these subjects belongs more especially to the science of physics or chemico-physics. For fuller information on these points than can be included within the scope of this book students are referred to special treatises on physics.