Hyperfine structure in the arc spectrum of bromine

Abstract

In an earlier work Tolansky (1932) reported measurements for the hyperfine structures of 18 lines of the arc spectrum of bromine. Although resolution was incomplete in every line, it was possible to prove that both the bromine isotopes 79 and 81 have the same nuclear spin, 3/2, and have practically identical nuclear magnetic moments. In carrying out the hyperfine structure term analysis at that time, use was made of the interesting graphical method proposed by Fisher and Goudsmit (1931) and thus approximate hyperfine structure interval factors could be derived. At the time this work was done the existence of deviations from the interval rule was not suspected. Any such deviations destroy the validity of the Fisher-Goudsmit method of analysis. The frequent occurrence of nuclear electrical quadrupole moment leads one to suspect that there may be deviations from the interval rule in many spectra; hence we have undertaken a new and more complete examination of the hyperfine structures of the lines of Br I. The measurements reported here are more extensive and more accurate than those previously given. We have measured the structures of 76 lines extending from the infra-red at 8700 A down to 4390 A. All the principal lines in this region have been measured and, with very few exceptions, the structures have been completely accounted for. The results obtained here are much more accurate than those given earlier for the following reasons: (1) The modified source is much more intense. (2) It is cooled, giving sharper lines. (3) In place of the single Fabry-Perot previously available (with one thickness of silvering) we have used various silvered Fabry-Perot interferometers, similar interferometers coated with aluminium, and a high-quality quartz Lummer plate. (4) The camera focal length used now is 125 cm., that of the earlier instrument being only 50 cm. (5) Much more rapid plates are now available, particularly in the infra-red region.