Arizona State University Logo College of Liberal Arts and Sciences Logo
6f image #16f image #26f image #3

About the Cameca IMS 3f, continued ...

To support geochemical research in particular, Hervig was hired in 1984 as a Research Scientist, and an increasing amount of instrument time (initially 10-20%, currently ~85%) has been devoted to geochemical work. We have operated as an open laboratory, welcoming collaborators from ASU and elsewhere, with a nominal instrument charge ($15/hour) which covers routine maintenance and consumables but not capital equipment or salaries. We have particularly enjoyed pushing the capabilities of the instrument beyond that defined by the manufacturer and have developed innovative new approaches to quantitative SIMS analysis. In particular, (a) we discovered that the analysis of minerals and glasses for H is simplified when those H+ ions generated as a result of electron-stimulated desorption are removed from the mass spectrum (Hervig and Williams, 1988), (b) we have shown that extreme energy filtering (EEF), in which the analytical signal consists of ions ejected with > 300 eV initial kinetic energy, allows rejection of cluster ion interferences without the need for high mass resolution (Schauer and Williams, 1990), (c) we have developed a high energy electron gun for charge neutralization and have shown that, when used with EEF, it is possible to obtain 17O/16O and 18O/16O ratio measurements with precision of 1-2 per mil (Hervig et al., 1992), (d) we have demonstrated the use of the primary ion column of the ion microprobe as a low-energy ion implanter to generate solid-phase standards and have built a hollow cathode ion source to generate ion beams of virtually any isotope (Streit and Williams, 1988), (e) we have demonstrated the ability to perform sputtered neutral mass spectrometry in the ion microscope using primary ion impact ionization of a fraction of the sputtered neutral flux; this capability allows quantitative analysis of major element levels using a single standard because working curves are linear (Williams and Streit, 1986).