IONICS Since J. J. Thomson, in 1896, applied the ionic theory to the conductivity produced in gases by X-rays, it has been used, more or less successfully, for the explanation of every kind of gaseous conduction.
The conductivity of flames has been studied by Arrhenius, A. Smithells, H. M. Dawson, H. A. Wilson, E. Marx, Lenard, Andrade, Moreau and other physicists. H. A. Wilson in 1899 showed that the negative ions in flames have a much greater mobility than the positive ions. The velocity of the positive ions is only a few centimetres per second due to one volt per centimetre, as first shown by Andrade, while that of the negative ions is about 4,000cm. per second. The negative ions in flames are therefore supposed to be free electrons and the positive ions charged atoms. The conductivity of alkali salt vapours in flames was shown by A. A. Noyes and H. A. Wilson, in 1922, to agree with that calculated by assuming that the salts are disso ciated into the metal vapour which in turn partially dissociates into electrons and positively charged metal atoms. The theory of the dissociation of metal vapours into electrons and positively charged atoms was worked out in 192o by Saha, who used it suc cessfully to explain many results on the spectra of the sun and other stars. (See ELECTRICITY, CONDUCTION OF, In Gases.) Ionization by Collision.—The motion of ions in gases, due to diffusion and to electric fields, has been very thoroughly studied by J. S. Townsend and his co-workers, and results of great im portance obtained. It is found that the velocity of the ions, due to a field X, in a gas at pressure p, depends on X/p. In the case of the positive ions it is proportional to X/p, but in the case of the negative ions it increases more rapidly than X/p, and becomes very large when X/p is large. It appears that, when X/p is large, the negative ions are free electrons. When the velocity of these electrons is great enough, they ionize the gas by collisions with the molecules. The positive ions also appear to ionize the gas molecules in intense electric fields, and Townsend has shown that the production of sparks and continuous electric discharges through gases can be explained by ionization by collision. (See ' ELECTRICITY, CONDUCTION OF, In Gases.) Critical Potentials of Atoms.—A new method of investi gating the ionization of gases by rapidly moving electrons, intro duced by Lenard in 1902, has led to results of fundamental im portance. In Lenard's experiment, electrons, emitted by a hot wire or a plate illuminated by ultra violet light, in a gas at a low pressure, were accelerated by an electric field towards a wire grat ing. Some of the electrons passed through the grating and, if their velocity was great enough, ionized the gas on the other side of the grating. This ionization was detected by means of an electrode, which was charged negatively and so repelled the electrons coming through the grating, so that they could not reach it. If the gas between the grating and the electrode was ionized by the electrons, the positive ions were attracted to the electrode, and so could be detected. The potential difference, P, between the source of the electrons and the grating was gradually increased, and it was found that there was no ionization until it was equal to a definite value depending on the nature of the gas. The value of P at which the electrons begin to ionize the gas is called the ionization potential of the gas molecules. It is found that in many cases the gas is caused to emit radiation when P is less than the ionization poten tial. The potential difference required to cause the gas to emit radiation of a definite wave length is called the radiation potential of the gas molecules for that wave length. The radiation and ion ization potentials for many different elements and compounds have been determined by Franck and Hertz, A. Ll. Hughes, F. Horton, K. T. Compton, Foote and Mohler, and others. These potentials are called the critical potentials of the atoms or mole cules.
According to the quantum theory an atom can only exist in cer tain definite states having definite energies, E,,
E. . . . It is found that the critical potentials of the atom are related to its possible energies, so that Pe=E„—E,,, where e is the charge on the electron, so that Pe is the energy of the electrons. The meas urements of the critical potentials have thus provided a remark able confirmation of the quantum theory. (See ELECTRICITY, CONDUCTION OF, In Gases; and ATOM.) Thermionics.—The escape of electricity from bodies at high temperatures has been investigated very thoroughly since 1900. J. J. Thomson, in 1899, measured e/m for the negative ions emitted by a carbon filament in hydrogen at a very low pressure, and found it equal to I
thus showing the ions emitted to be electrons. It is found that any metal and many compounds emit electrons when heated in a vacuum. In 1902, O. W. Richardson measured the current carried by the electrons emitted by several metals in a vacuum at different temperatures, and found it in creased rapidly as the temperature was raised. He explained the variation of the current with the temperature by supposing that the metal contains free electrons having the same energy as gas molecules, and that those electrons which strike the surface of the metal from inside with a normal velocity greater than a definite value are able to escape. The subject has since been in vestigated by O. W. Richardson, H. A. Wilson, F. Horton, Weh neldt, C. J. Davisson and many others. The idea that the free elec trons in a metal have the same energy as gas molecules has been abandoned, and the theory of the emission of electrons from hot bodies is now based on the analogy with evaporation. The ther modynamical theory of evaporation was applied to the emission of electrons from hot bodies by H. A. Wilson in 1903, and has since been developed by O. W. Richardson, H. A. Wilson, P. Dushman and others. The emission of electricity by hot bodies was called thermionics by O. W. Richardson, to whom a large part of our knowledge of this important branch of physics is due. (See
