LIMITATIONS OF DIRECT ELECTRON COUNTING IN SOFT-X-RAY SPECTROSCOPY

Recent developments in electron counting detectors have shown that these types of detectors offer the potential to provide significant improvement in the energy resolution for soft X-rays. To evaluate the performance of any electron counting detector, however, requires an accurate knowledge of parameters such as electron mobility, characteristic electron energy, mean electron lifetime against capture to gas impurities, signal-to-noise ratio and resolving time of the associated electronics. In our previous paper [T.J. Harris, E. Mathieson, J.A. Mir and G.C. Smith, Nucl. Instr. and Meth. A 299 (1990) 195] we reported experimental values of these parameters using A-CH4 (10%), A-CH4 (50%) and A-CO2 (20%) at atmospheric pressure. Although examples of the ''best'' electron counting spectra at 1.49 and 5.89 keV were included in that paper, no attempt was made to show the various trade-offs which were needed for optimization. The purpose of this paper is to show the performance of the present direct electron counting system under different signal-to-noise ratios and drift fields. The useful X-ray energy and drift field range of the present electron counting system have been established. Through an understanding of loss processes, modifications have been suggested to improve its performance in any future investigations. Electron counting at atmospheric pressure can also prove to be a powerful tool to provide accurate measurements of basic ionization parameters such as the Fano factor and the average energy required to create an ion-pair (W) in counter gases for X-rays (0.2-10.0 keV) through a knowledge of electron counting efficiencies only.