The velocity distribution of photoelectrons produced by soft X-rays

In the literature published on the subject of soft X-rays, there are two papers to be found, which deal particularly with the velocity of the photoelectrons produced by such rays. Both authors of these papers state, that they have found a photoelectric emission built up of homogeneous groups of electrons with characteristic velocities, giving rise to very sharp and steep jumps in tire curves obtained by P. Lukirsky with retarding electric fields, and to equally sharp lines on the photographic plates in J. A. Becker's magnetic defection apparatus. There were scarcely any electrons with velocities present. In the autumn, 1925, the writer engaged himself in an investigation on the same problem at King's College, London, as a direct continuation of experiments which Prof. O. W. Richardson had been carrying out, chiefly in collaboration with Mr. A. A. Newbold. The electric stopping potential method was need, and a number of different tubes were designed and tested. In tins way the many disturbing factors which may easily completely mask the true effects in experiments of this kind were gradually reduced, and the remaining sources of error could be allowed for. As a result of these experiments, it was quite clear, that the greater portion of the total emission from metals consists of electrons with energies of a few volts only. In addition, a small discontinuity in the current vs retarding voltage curves was found in the higher voltage region, the position of which at 265 volts agreed very well with the value of the Kα -quantum for carbon, the substance used for the anode, as calculated from the best critical potential determinations and confirmed at about the same time by A. Dauvillier's first measurements using a crystal method. The curves obtained are totally different from those published by Lukirsky. From the experience gained in experiments with several different tubes the writer is inclined to think, that the shielding of the photoelectric target in Lukirsky's apparatus from the electric fields of neighbouring electrodes at high potentials may not have been sufficient to prevent a considerable distortion of the true curve. This might explain the convicting results arrived at. As regards Becker's work, it seems to the present writer a dangerous thing to keep the photoelectric part at the same potential as the anode, for any electrons leaking out through the gauzes from the X-ray tube must inevitably have been recorded as fast photoelectrons. In a note by Becker's collaborator, which appears to have escaped the attention of those who quote this investigator, such an effect is actually stated to have occurred. In the stopping potential method the current between the photoelectric emitter and the surrounding electrode is always composed of two: the electrons emitted from the target against the opposing field, and a current of electrons produced by scattered radiation at the surface of the outer electrode. This current will always be present; it flows in the opposite direction to the first one and causes the total current from the emitter, as measured by the electrometer, to reverse its direction for a low value of the stopping potential. For higher stopping potentials this second current constitutes almost the whole of the measured negative current in which the changes in the small superposed current carried by fast electrons from the target must tend to disappear. Since it does not appear to be possible to get rid of the effects of scattered radiation, the stopping potential method has been abandoned for a magnetic deflection method of a special find, described in the following.