The theory of the surface photoelectric effect in metals—II

Abstract

1—It is usual to divide observed photoelectric effects into two general categories known as the selective and normal effects. Die former, of the occurrence of which there can be no doubt, is so named from the form of the spectral distribution curve, the photoelectric emission pacing through a well-defined maximum for a frequency usually at about one and a half times the threshold frequency. The normal effect, on the other hand, shows no such maximum, and was given its rather unhappy name by the early experimenter who observed it when using light polarized with its electric vector perpendicular to the plane of incidence. The use of the term normal for this effect is unfortunate for two reasons. In the first place, the effect is of very rare occurrence, and is not clearly shown by pure metals in a solid state. secondly, the term is really needed to describe the "usual" photoelectric effect, which is theoretically predicted, and which is actually found to occur with metals in a very pure or nearly pure state. This effect has a maximum in the spectral distribution curve, the emission at the maximum being of the order of 1-5 × 10 -4 Coulombs per calorie. Now there would be no harm in calling this effect the selective effect without further qualification, were it not that it needs to be distinguished from a more violent type of selective effect, which is observed only when the surface of the metal is specially sensitized. The maximum is then much sharper, and the order of magnitude as much as 100 times greater.