The gas flux and the lamp pressure cited above have been purposely selected to minimize self-absorption with adequate line broadening and light levels. All the spectral lines have been registered in the first and second diffraction orders. Measurements in first order, the one with the best efficiency, have been used for temperature determination, while those at second order have been employed to obtain Stark parameters due to its highest dispersion.
For every spectral line, 11 different instants of the plasma life have been recorded, between µs and µs, when SiIII emission decreases substantially. For each instant considered ten runs have been made, five with mirror M3 and five without it, the exposure times usually being 5 µs. A sufficient time gap between two consecutive discharges has been selected to clean the waste in the lamp by means of the vacuum system. Nevertheless, a maximum of about 250 discharges has been performed in each single measurement session. Therefore, the evaporation of waste deposited on the lamp walls do not appreciably contribute again to the silicon plasma concentration. A total of 15 experiments with the same prefixed physical magnitudes have been made to complete all measurements. A real time check of plasma electron density and temperature repetitiveness between sessions has been performed during each experiment by periodical monitoring of width and peak intensity of the HeI 667.8 nm line. Exhaustive cleaning and vacuum of the gas installation, window and tube replacement and lamp realignment have been achieved after each measurement session.
Interferometric registers of 500 µs long have been simultaneously performed with spectroscopic records to determine electron density evolution.
© European Southern Observatory (ESO) 2000
Online publication: December 5, 2000