ABSTRACT
Good spectroscopic data of Mercury (Hg) are imperative for the diagnostics of florescent light plasmas. The spectrum of neutral Mercury, Hg I, was studied many years ago by, for example, Burns et al. [I] and Fowles [2]. Singly-ionised mercury, Hg II, has been the subject of a very recent comprehensive spectroscopic study by Sansonetti and Reader [3]. Each spectral line has two parameters of interest, wavelength and intensity. The wavelength is of course fixed and independent of the light source and recording conditions. However, the intensity depends on both these experimental conditions and hence can lead to some misinterpretation. The reported intensity data for Hg I should be carefully considered before use as there are a number of separate data and maybe incompatible sources involved and photographic recording was used in all cases. The Hg II data are much better in this respect in that intensities are given for the lines, however it is very likely that the relative intensities depend very much on the mean of excitation in the light source. In studies of the spectra of florescent light tubes we have noticed a number of problems concerning differences in relative line intensities between lamp spectra and the reported data, in for example the N1ST data tables [4]. One particular problem we have noticed concerns the abundance of Hg II in typical florescent tube plasmas. We have observed the line from the 5d96s2 2D512 — 5d1°6p transition at 398.39 nm in a florescent tube plasma and although this line is fairy weak it is a weak component of the three decays from the 5d106p 2P3,2 upper level, see figure 1. In fact, the calculated branching ratio for the 398.39 nm transition is only around 0.3 % of the stronger decay to the 5d1°6s 2S1 2 line at 164.99 nm. If the calculated branching ratio is correct, this would indicate that Hg II should be an important constituent of fluorescent light tube plasmas. In an effort to understand this and other such problems we have initiated a beam-foil spectroscopy study of Hg I and Hg II. The experiments are carried out at the University of Toledo heavy ion accelerator facility, [5]. In these experiments a beam of 220 keV Hg-ions from the accelerator is directed towards a thin (2 µg/cm) carbon foil. Collisions within the foil leave the beam with a charge state distribution containing both Hg I and Hg II in excited states. The beam, after the exciter foil, is viewed by a 1-meter normal incidence vacuum spectrometer. Depending on the choice of gratings (ranging from 2400 to 600 lines/mm) and detectors we can cover the wavelength region from 40 to 600 nm in overlapping wavelength regions. In this way we can relate intensities for the different grating-detector combinations to each other. This is important, as one of the goals of this program is to establish an internally constant set of relative intensities for Hg I and Hg II spectra, at least under beam-foil excitation conditions. The spectra will be calibrated for relative intensities using a beam of Pb I. The spectrum of Pb I has been calibrated for relative intensities by Lotrian et al. [6]. We will also use predicted relative intensities based on intermediate coupling amplitudes deduced from energy level data [7]. By moving the foil upstream of the spectrometer entrance slit we can record the decay curve of the upper level giving rise to the transition under observation, see figure 2 for a decay curve for the 5d1°6p 2P312 level. From such data, after a cascade analysis, we can obtain the 2P312 lifetime.
