Page:EB1922 - Volume 31.djvu/213

Rh

family are connected by a remarkably simple and interesting relation, which was discovered by Mr. Cuthbertson (Phil. Trans. A. 207, p. 135). It is shown in the following table, where the numbers under the symbols denoting the elements are the values

off(K-i)Xio:

He 144 Xi

N

297

P

1197

=299X4

As

1550

= 258X6

O

270

s

IIOI

=275X4 Se

1565 = 261X6

Te

2495 = 249X10

F ' 192 Cl

768

= 192X4 Br

1125 = 187X6

I

1920 = 192X10

Ne

137

Ar.

568 = 142X4

Kr.

850 = 142X6

X

1378 = 138X10

Thus the values of K i for successive elements of the same family (N.P.As): (O,S,Se,Te): (F.Cl.Br.I): (Ne,Ar,Kr,X) are in all cases very nearly in the proportion i, 4, 6, 10. In the simple theory, where the molecules are regarded as conductors, this would indicate that the volumes of the molecules of the successive elements in the same family are in the proportion i, 4, 6, 10, for each of these types of elements. On the theory which regards the atom as built up of electrons arranged round positive centres, the configuration of the outer layer of electrons for different members of the same family would be similar, and it is easy to show that for similar configurations of electrons the value of K i would be proportional to the cube of the linear dimensions, i.e. to the volume enclosed by the outer layer of electrons; so that again on this theory Cuthbertson's result shows that volumes of successive elements in the same family are in the same ratio whether the family be that of the inert gases, the halogens, or the oxygen or nitrogen groups.

Another example of the information as to the nature of the molecule afforded by determinations of the specific inductive capacity is that, while the specific inductive capacity of many gases, e.g. H 2, N2, Oz, CO, COi, C1 2 , is equal (as Maxwell's Electromagnetic Theory of Light suggests) to the square of the refractive index, there are, as Badeker (Zeitschrift Physik. Chem. 36, p. 305) has shown, others, such as NH 3, HC1, SO2, the va- pours of water and the alcohols, whose specific inductive ca- pacity is far in excess of the value given by this rule, and moreover the specific inductive capacity of these gases diminishes much more rapidly as the temperature increases than that of gases of the first type. The difference can be accounted for by supposing that the molecules of gases of the first type have no electrical moment when they are free from the action of an exter- nal electrical force, while those of the second type have an intrin- sic electrical moment apart from that which may be produced by the external force. When there is no electrical field, the collisions between the molecules will cause the axes of electrical moments of the different molecules to be uniformly distributed, so that the average effect will be zero. An electric force will tend to drag the axes of the different molecules into alignment, and the assem- blage of molecules will have a finite electrical moment which will be a measure of the specific inductive capacity. Inasmuch as the collisions between the molecules tend to knock their axes out of line and diminish the specific inductive capacity, the latter will diminish as the temperature and with it the vigour of the en- counters increases. The substances which have an intrinsic elec- trical moment have exceptionally active chemical properties and are good solvents, dissociating the salts dissolved in them.

If the distribution of electrons in a molecule were not sym- metrical about three axes at right angles to each other, the specific inductive capacity of a single molecule would vary with the direction of the electric force, but as the molecules in a gas are orientated in equal numbers in all directions we should not detect this by direct measurements of the specific inductive capacity. We can however detect this effect in another way; for if the molecules have different specific inductive capacities in different directions the light scattered by the molecules at right

angles to the incident unpolarized light will not be plane polarized as it would be if the molecule were symmetrical (J. J. Thomson, Phil. Mag. 40, p. 393), and if the incident light is plane polarized the scattered light will not vanish in any direction. Strutt (Proc. Roy. Soc. gSA. 57) has measured the departure from plane polarization for different gases with the result shown in the fol- lowing table:

Argon 0-46%

Hydrogen 3'83%

Nitrogen 4-06%

Air S'00%

Oxygen 9-40%

Carbon dioxide 11-70%

Nitrous oxide 15-40%

This shows that the molecule of argon is very symmetrical, while the nitrogen molecule is more symmetrical than the oxygen, and this again more symmetrical than that of COj.

Ionized Gases. Gases may in various ways be' put into a state in which they conduct electricity on an altogether different scale from the normal gas. They acquire this conductivity when Rontgen rays or the rays from radioactive substances pass through them, or when they are traversed by cathode or positive rays. Ultra-violet light of very short wave length can impart this property to a gas, while gases recently driven from flames or from near arcs or sparks or bubbled through certain liquids or passed slowly over phosphorus also possess this property.

The conductivity of gases possesses interesting characteristics. In the first place it persists for some time after the agent which made the gas a conductor has ceased to act; it always however diminishes after the agent is removed, in some cases very rapidly, and finally disappears. The conducting gas loses its conductivity if it is sucked through glass-wool, or made to bubble through water. The conductivity may also be removed by making the gas traverse a strong electric field so that a current of electricity passes through it. The removal of the conductivity by filtering the gas through glass-wool or water shows that the conductivity is due to something mixed with the gas which can be removed by filtration, while the removal of the conductivity by the electrical field shows that this something is charged with electricity and moves under the action of the electric force. Since the gas when in the conducting state shows as a whole no charge of electricity, the charges mixed with the gas must be both positive and nega- tive. We conclude that the conductivity of the gas is due to the presence of electrified particles; some of these particles are positively, others negatively, electrified. These electrified parti- cles are called ions, and the process ionization.

The passage of electricity through a conducting gas does not follow the same laws as the flow through metals and liquid elec- trolytes; in these the current is proportional to the electromotive force, while for gases the relation is represented by a graph like fig. i, where the ordinates ar,e proportional to the current and

Scale Divisions

0. 100 200 300400500600700.800 9001000 1100.12001300.1400 1500

Volts RG.1

the abscissae to the electromotive forces. We see that when the electromotive force is small, the current is proportional to the electromotive force, as in the case of metallic conduction; as the electromotive force increases, the current after a time does not increase nearly so rapidly, and a stage is reached where the cur- rent remains constant in spite of the increase in the electromotive force. There is a further stage, which we shall consider later, where the current again increases with the electromotive force, and does so much more rapidly than at any previous stage. The