Less is more: optimizing 9Be carrier addition for AMS analysis of 10Be at CAMS
Ms Alianora Walker1,2, Alan J. Hidy1, Susan R. Zimmerman, Scott J. Tumey1, Tom Brown1, Sandra M. Braumann3, Roseanne Schwartz3, Joerg M. Schaefer3
1Lawrence Livermore National Laboratory, Livermore, United States, 2San Jose State University, San Jose, United States, 3Columbia University, Lamont-Doherty Earth Observatory, Palisades, United States
We conducted a series of experiments on the CAMS-LLNL (Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory) 10-MV model FN tandem Van de Graaf accelerator and modified high-intensity cesium sputter source to determine the ideal 9Be carrier mass to use for 10Be targets analyzed with our system. CRONUS-A quartz was dissolved then split to prepare targets identical in 10Be content, but with variable 9Be (10-250 µg). For handling purposes, and to control density of 10Be atoms in targets, samples were bulked with Fe or Al carrier to replace the reduced 9Be. Packing experiments were performed with oxide-Nb mixtures to determine volume-equivalent masses needed to keep each target packed to a consistent depth. Our experiments show a maximum ~40% increase in 10Be counts for samples prepared with 75 µg vs. 250 µg 9Be (typical) after 25 minutes of sputtering. This result is substantiated by increased total 10Be counts observed on targets analyzed to exhaustion, implying significant gains in total system efficiency with the lower carrier mass. Multiple low-level (~104 atoms 10Be/g) quartz samples were prepared using both the typical and lower 9Be carrier amounts and analyzed under normal AMS operational conditions. In each case, significantly better counting statistics were achieved with the lower 9Be mass while reproducing the 10Be concentration obtained from the typical-mass duplicate. Similar improvements were also observed in targets prepared with lower carrier from an external sample prep lab (Columbia University).
Collectively, our results indicate that optimal measurement efficiency for 10Be AMS targets analyzed at CAMS is obtained using 75 µg 9Be carrier and replacing the reduced mass with ~230 µg Fe. We recommend adding Fe carrier during the final hydroxide precipitation step prior to calcination. For this work, we sourced our Fe from an ICP standard purchased from Alpha Aesar that was tested to be sufficiently 10Be-free. However, an implementation where Fe is harvested from the digested quartz sample itself (anion column reject) and later reintroduced to the sample is being explored and may be an elegant alternative to using commercial carrier since most quartz has sufficient Fe content for this purpose. This reduction in per-sample 9Be carrier mass—by a factor of 3.3x—offers substantial benefits to both analysis quality and cost (higher analyzed 10Be/9Be ratios, better counting statistics, shorter run times, longer-lasting expensive carrier bottles) without modification to established AMS operations for 10Be analysis at CAMS. Based on these results, 10Be targets prepared with anywhere between 75 and ~250 µg 9Be is acceptable for submission to the CAMS user facility for analysis.
Prepared by LLNL under Contract DE-AC52-07NA27344. This is LLNL-ABS-826042.
Alianora Walker is a student at the Center for Accelerator Mass Spectrometry at Laurence Livermore National Laboratories and a Master’s student studying tectonic geomorphology at San Jose State University. She received an AB in geology from Smith College in 2011. Today, she hopes to convince you that, in 10Be sample processing, less really can be more.