Mike Jercinovic - Sensitivity in EPMA

Coauthors:
M. L. Williams, University of Massachusettes, Amherst
D. R. Snoeyenbos, Cameca Instruments

Abstract
The application of the electron microprobe to trace element analysis and geochronology is becoming increasingly important, requiring reconsideration of both hardware and methodology. The Ultrachron project was initiated in 2001 as a collaborative effort between the University of Massachusetts and Cameca with the goal of improving EPMA hardware, software, and analytical protocols relevant to trace element microanalysis. The project was inspired by potential benefits in monazite geochronology, emphasizing the difficulties inherent in high spatial resolution, high precision analysis of beam-sensitive phases.

Monte Carlo modeling of electron scattering and X-ray emission for Pb-Mα in monazite (REE+actinide-phosphate) can be used to estimate expected analytical spatial resolution (quadrature sum of the error functions associated with beam diameter and scattering volume, and X-ray emission volume). Such high-Z phases allow analytical resolution in the sub-micron realm if beam brightness is sufficient, and if the accelerating potential is kept below 15kV. At 10kV, the analytical resolution will remain below 500nm for currents up through 200nA if a beam diameter of 200nm or less is realized.

Improving the counting efficiency allows an increase in precision at lower voltage and/or beam currents for the more beam sensitive materials. To this end, new “VLPET” spectrometers were developed which are approximately 5× the collection area of the standard PET, and are equipped with accordingly large detectors. The improved collection efficiency substantially improves count rates above previously existing monochromators. Software integration of multiple spectrometers counting the same analytical line and statistically-based sampling allow further improvement in precision. The application of large and very large crystals, along with count integration results in sensitivity (detection limit) improvement by nearly 4× for Pb-Mα in monazite for single point analysis. For 6-point analysis, the age precision (comparing standard error of the mean) is improved approximately 3× using VLPET and 2-spectrometer integration relative to standard PET on the SX50. When beam damage and/or contamination issues are critical, lengthy count times are not preferable, as in speleothem analysis. The Ultrachron has been applied to Sr analysis in CaCO₃ speleothems using 5-spectrometer integration involving PET, LPET, and VLPET spectrometers. Using 15kV and 100nA, 100 sec analyses were performed giving a precision of 29 ppm (2σ) at a concentration of 200 ppm (single point detection limit = 50 ppm). Similar 5-spectrometer integration for analysis of Zr in rutile at 20kV and 200nA (600 s per point) results in a single point detection limit of 14 ppm, and 3 ppm for a 15 point acquisition.

Alternative presentations:
In-situ Trace Element Analysis in
           Complex, Multiphase Materials by EPMA
Hans Bethe and Nanogeochronology

Biography
Mike Jercinovic is an Associate Professor in the Department of Geosciences at the University of Massachusetts, and supervises the Electron Microprobe/SEM Facility in the department. He received his PhD from the University of New Mexico (1988), where he was introduced to microprobe analysis by Klaus Keil. Research at UNM concentrated on alteration processes in natural basaltic glasses and the application of natural analogues to modeling the corrosion of nuclear waste form glass. Mike then supervised the MIT electron microprobe facility for several years, and worked in the private sector in electron microscopy (microelectronic evaluation) before moving on to UMass in 1997. Research at UMass has primarily involved microanalysis applied to tectonic studies, and also includes other diverse geochemical, mineralogical and materials applications. Most recently, Mike’s research focus has largely related to development of new techniques in electron microprobe trace element analysis. This research primarily applies to the analysis of high-Z accessory phases (monazite, xenotime, thorite, etc.) for geochronology, but is also finding relevance in paleoclimate research, and various petrologic applications. This research has led to collaborative development of the new Cameca SX-Ultrachron electron microprobe. This project has included significant new hardware and software dedicated to improving spatial resolution and count precision in electron probe microanalysis.