The Probability Law Governing Ionization by Electron Impact in Mercury Vapor

Publisher Website
Google Scholar
Abstract
Despite several attempts, the unusual type of probability law found by Lawrence to govern inelastic impacts in mercury vapor just above the ionization potential, has not been verified by other observers. His experiment, making use of a magnetic separation of the electron beam and a Faraday cylinder type of ionization chamber, has been criticized on the basis that electron reflection from the walls and other spurious wall effects might account for the results. Since no theory has predicted such a probability law it seemed important that a new investigation be launched in an attempt to prove or disprove its validity. An apparatus providing an electron beam approaching in homogeneity that of the Maxwellian distribution should make such a test readily possible. An electron-gun type of ionization tube was devised, capable of high resolving power, and eliminating the objectionable features of the Lawrence method. Using a dull emitter cathode it was found possible to approach very closely the theoretical Maxwellian velocity distribution. With this arrangement the new critical potentials of Lawrence were checked and his probability law verified within the limits of experimental error. The values of these critical potentials are: 10.6; 11.3; 11.7; and 12.1 volts. Additional ionization potentials at 12.3; 12.45; 12.85; and 13.2 volts were also found. A search for critical potentials above 13.2 volts yielded negative results. The results indicate that the failure of other experimenters to detect this probability law was due to their wide distribution of electron velocities. Since this is true, the values which they have given for the probability of ionization in this region can be considered as correct only in order of magnitude. Evidence is presented to show that these critical potentials are a result of the ionization of metastable mercury atoms, the probability law indicating either a high concentration of such excited atoms, or a large collision cross section for a metastable atom toward an ionizing collision. Data concerned with the variation of ionization current with electron current density, seem to favor such an hypothesis but is not decisive in character. When mercury vapor is admitted to the tube, the widening of the velocity distribution indicates the existence of a new type of atomic collision process. In this process the electrons lose energies of only a small fraction of a volt. These low energy losses are now being investigated together with additional evidence concerning the metastable atom hypothesis.