Inertia of Gas-Filled Photocells1

Studies of inertia in gas-filled photocells have usually been carried out in simple photocells. This has made it almost completely impossible to work out, on the basis of experimental data, the mechanism of the phenomenon, owing to the difficulty of separating the effects of the various factors. As a result, different authors have put forward various explanations of this phenomenon, such as: 1) thermal effects at the cathode2, 2) accumulation of positive ions near the cathode3, 3) knocking-out of electrons by positive ions4, 4) the existence of metastable atoms5, etc.; moreover, in most cases it was assumed that the transit time of the positive ions is so short that, at audio frequencies, the effect of lag in the ionic current may be neglected.

In the paper under review, the task was set of clarifying the role of the motion of positive ions in the phenomena of inertia. From simple theoretical considerations it follows that, if this circumstance plays a role, the curve of the dependence of the photocell output on the frequency of modulation of the light flux should not decrease monotonically, but should exhibit maxima and minima at frequencies for which the periods are respectively equal to integral multiples and half-integral multiples of the transit times. Since such a course of the curve has not been observed up to the present, the author took special measures to eliminate in the experimental apparatus such interfering phenomena as destruction of the cathode by ionic bombardment and the ejection of secondary electrons. For this purpose the experimental apparatus was made in the form of a spherical photocell 63 mm in diameter, with a central anode in the form of a ring 6.3 mm in diameter. The photosensitive part of the cathode surface constituted only a small part of the entire inner surface of the bulb (its diameter was about 12 mm), which, with the exception of the window, was all covered with a conducting, non-photosensitive (carbon) layer. This layer was maintained at a potential 1.5 V higher than the potential of the emitting part of the cathode (this latter consisted of hydrided potassium). Such a device made it possible very successfully to separate the photocurrent from the gas-amplification current even at very large amplification factors (up to 30) and at high voltages (≈ 270 V).

The investigations were carried out with argon filling (pressure 0.2 mm Hg).

As a result it was found that the frequency characteristic does indeed contain maxima and minima situated at the appropriate values of the frequencies. This shows that the lag of the ionic current plays a large and probably even decisive role, since in these experiments no phenomena were observed that would confirm the assumptions listed above under numbers 1 to 4.

The author considers the reason for the mistaken view regarding the role of ions in inertial phenomena to be that, in earlier investigations, the field distribution in the photocell was not taken into account (in a central-anode construction there are large gradients near the anode and a weak field at the cathode), which, as is easy to understand, taking into account collisions of the moving ions with gas atoms, determines the transit time of the positive ion.

N. Khlebnikov, Moscow

Literature

1. A. M. Skellett, Journ. Appl. Phys., 9, 631, 1938.
2. Kingdon and Thomson, Physics, 1, 343, 1931.
3. E. L. E. Wheatcroft, Phil. Mag., 7, 162, 1931.
4. Fourmarier, C. R., 194, 86, 1932.
5. Schroeter u. Libszynski, Physik. Z., 31, 897, 1930.