On the effect of temperature on the anomalous reflection of silver

Abstract

It has long been known that the reflection coefficient of silver, besides minishing steadily in the ultra-violet, shows a marked anomaly in the ighbourhood of wave-length 3,200 Å, sinking to a value of nearly zero, while either side of this point it amounts to about 50 per cent. The cause of is phenomenon is unknown, though from the classical electro-magnetic int of view it would naturally be regarded as some kind of "resonance" ect; that is to say, that the “free” electrons cannot, for some reason, brate with frequencies corresponding to this region; for the bound ctrons, though they could, if they were in resonance with frequency, increase e total reflection, could hardly decrease it, as actually occurs. If this is so, and if the restoring force on the vibrating electrons varies with e distance between them and the neighbouring atoms and electrons, it may expected that the limiting frequency on either side of this band will vary the metal contracts or expands. A change in the temperature of the silver ould then produce a corresponding shift in the position of the band of weak flection, and measurement of the amount of this shift may afford some ue to the nature and magnitude of the forces which act upon the “free” ectrons.