The first spark spectrum of platinum

Abstract

In 1929, Dr J. J. Livingood and myself attempted unsuccessfully to analyse the first spark spectrum of platinum. That failure was chiefly due to the lack of adequate observations in the Schumann region. The spectrum should, of course, be similar to Ni II and Pd II, both of which were previously analysed in the laboratory, but in the third long period the excitation energies have increased to such a point that the chief lines of the transition MATHS FORMULA are mainly below λ2000. Their observation requires a vacuum spectrograph, which we did not have at that time. Observations Down to the limit of transmission of air, the spectra of the arc and spark were photographed with the 21 ft. grating, and the wave-lengths were measured against iron standards. It was found necessary to use graphite electrodes for the arc, because of the fact that it is very difficult, if not impossible, to run an arc between pure platinum rods. One is much more apt to get a glow discharge in which the cathode is hot, the anode cold and the spectrum that of air. In the Schumann region also, arc-spark observations were the chief source of the data. A modification of the method described by Selwyn (1929) was used, and excellent spectra were obtained down to the fluorite limit at λ1240. The modification ensures an image of the source at the slit and makes the elimination of oxygen from the flowing nitrogen much more complete. The arc in pure nitrogen runs at 10-15 V higher potential than it does in air, and produces the second spark spectrum of some elements with considerable intensity. A diagram of the attachment is shown in fig. 1. The spectrograph is of conventional type employing a glass grating of 2 m. radius and 30,000 lines per inch. The spectrum can be photographed from λ500 to λ 2250. Hilger Schumann plates and Ilford Q, plates were used, the latter being practically perfect for the longer wave-lengths.