The Photo-Electric Threshold for Mercury

Abstract

In order to insure a perfectly clean surface of mercury for test, the apparatus was arranged so that carefully purified mercury flowed intermittently into a shallow iron cup and out through an overflow opening at the surface back into an electrically heated still. Direct observation showed that with such a flow the surface cleans itself of any added contamination very quickly. Also, the results for the wave-length limit were found to be independent of the rate of flow, to a rate far below that normally used, and also unaffected by the presence of gases (air, ${\mathrm{H}}_2$, ${\mathrm{O}}_2$, He and C${\mathrm{O}}_2$) and even of traces of water vapor in the photochamber. The photo-currents to a copper oxide receiver were measured by means of a quadrant electrometer to within about 1 percent for various Hg lines from a Cooper-Hewitt quartz mercury arc, correction being made for stray light by means of intermediate readings. This arc operated at 92-93 volts with 2.1 amp., was constant to 1 percent. Intensity measurements accurate to 1 percent were made by means of a vacuum thermopile connected to a d'Arsonval galvanometer (sensitivity ${10}^{-9\mathrm{}}$ amp.) using a Hilger monochromator to disperse the light and making correction for stray light by taking readings between the lines. Plots of the ratio of corrected photo-current to corrected light intensity as a function of wave-length give the threshold at 2735A. The corresponding photo-electric work function is 4.52 volts. This result is free from the objections raised to those of previous observers and seems clearly to be the evaluation of an intrinsic property of mercury uncontaminated by surface impurities.