On the absorption of light by electrically luminescent mercury vapour

Abstract

Some of the earliest experiments on the absorption of light by electrically luminous gases were carried out by Pflüger, who, in 1907, investigated the absorption and reversal of the hydrogen lines by luminous hydrogen. He used a condensed discharge in a three-electrode tube, in which a short constriction provided the source of radiation and the wider and longer part the absorbing column. He succeeded in reversing H α . This work was followed up by Landenburg and Loria, who reversed H α and H β . In the same year Kuch and Retschinsky made experiments on selective absorption in mercury vapour lamps. They found that the ratio of the intensities of the spectral lines in the light from the mercury vapour depends on the thickness of the radiating layer of vapour, the intensities of neighbouring lines tending to equalise as the layer increases in thickness, a result which would follow from a relatively greater absorption of the stronger lines. They also made photometric measurements of the illumination from two mercury lamps, one of which was placed behind the other so that the light from the first had to traverse the second. They discovered that the radiation from the combination of lamps, arranged thus, was less than the sum of the radiations from each separately. Pflüger followed with photometric observations on the absorption of the lines 5461 Å. U., 4358 Å. U., 4047 Å. U., 5791 Å. U., and 5770 Å. U. Similar work was done by L. Grebe, on 5461 Å. U. and 4358 Å. U. In all these experiments with luminous mercury vapour, the current densities (of the order of 2 ampères per square centimetre) and the power developed in the absorbing arcs were considerable. Preliminary Experiments . It occurred to the present writers that very faintly luminous mercury vapour might possibly exhibit more marked selective absorption than had been observed by previous experimenters. It was, therefore, thought worth while to look for a means of maintaining a mercury arc of very low current density. This was found possible in a three-electrode tube in which two arcs were formed, having a common cathode. This arrangement, which is somewhat similar to that used in “self-starting ” vapour lamps, is shown in fig. 1. A strong arc is started between mercury pools K and A in the lower part of the vacuous tube; the third electrode B being a thick iron wire cemented in with sealing wax. Connections are made as shown. The apparatus having been pumped out to a very low pressure, an arc is struck between A and K, B being connected to the supply through a rheostat. When the ionised vapour from the arc AK reaches B, the second arc BK starts, taking a current whose strength depends on the resistance R 2 . This current strength can be indefinitely diminished by increasing R 2 . The low power arc BK can only be maintained in the presence of the strong ionising arc AK. If the pressure in the tube is sufficiently low and the current in BK is of the order of 0·1 ampére, the second arc fills the tube with a characteristic faint luminosity. In this state the vapour is found to exhibit marked selective absorption.