[en] Full text: The range of isotopes that can be measured in Accelerator Mass Spectrometry (AMS) is in large part determined by the ability of the technique to eliminate interfering stable isobars either in the ion source or by dE/dx separation in foils or gases, using a gas ionization detector or a gas-filled magnet. In the case of the detection of chlorine-36 by small AMS systems, attempts at selectively suppressing sulfur via ion-molecule reactions have been made using the electron transfer reaction S^- + NO₂^- > S⁰ + NO₂^-, exothermic by 0.2 eV. The equivalent reaction involving a chlorine anion is endothermic by 1.34 eV [Dunkin et al, Chem. Phys. Letters 15 (1972) 257-259]. Unfortunately, energy conditions found in classical AMS systems do not allow this suppression reaction to occur with appropriate efficiency. We discuss the various approaches that were tried to enable this specific suppression reaction in the low energy line of an AMS system, where it would be most useful. We review results obtained initially with a charge exchange canal, then an early Radio Frequency Quadrupole (RFQ) system, the Isobar Separator for Anions (ISA), both tested at the IsoTrace laboratory in Toronto, and a more advanced ISA currently operated at the A. E. Lalonde AMS Laboratory, University of Ottawa. Comparing these three systems, which operate over an energy range from ~1 eV to 1 KeV, illustrates the different phenomena governing the suppression reaction of sulfur, as well as the transmission of chlorine anions towards the tandem accelerator. A comparison is also made with conditions found in Liquid Chromatography Mass Spectrometry (LC-MS), another RFQ-based technique [Douglas, J. Am. Soc. Mass Spectrom. 9 (1998) 101–113]. These results are reinterpreted considering the most recent results obtained at A. E. Lalonde. Notably, the degree of thermalization of anions on their passage through the device is extensively discussed. The transition from non-thermal to near-thermal conditions appears to be essential to accomplish the sulfur suppression reaction to the degree required for AMS determination but entails a cost in the transmission of chlorine anions unless appropriate precautions are taken. Lessons learned over the multi-decade development of LC-MS systems, in which ions are transferred efficiently between thermal and non-thermal zones (albeit at generally lower energy levels), can be applied to improve transmission of chlorine anions through the ISA. The approach also improves markedly the transmission of molecular anions such as SrF₃ -, of great interest in the determination of other rare isotopes, by reducing their fragmentation during their passage through the ISA.

[en] Full text: One of the most important tasks in the design and construction of ultra-low background experiments is the radioassay of the materials used. This requires the selection of the materials and enables the calculation of expected detector background. The ASTREA project (Accelerator mass spectrometry Survey of Trace Radionuclides for Experiments in Astroparticle physics) addresses AMS radioassay challenges for a few rare event experiments. Some examples are nEXO, which is searching for neutrinoless double beta decay; and NEWS-G and DarkSide, which are attempting to directly detect dark matter. This project, led by the André E. Lalonde AMS Laboratory (AEL-AMS) at the University of Ottawa, is performed in collaboration with Carleton University, Queens University and University of Alberta. The main focus of the project is screening Pb-210 in various detector construction materials, with emphasis on low background copper and high-performance polymers. We have studied the possibility of using 2 different materials for the AMS measurements: lead fluoride (PbF₂) and lead oxide (PbO) targets, producing respectively (PbF₃)- and (PbO₂)- ions on the LE side. In both cases, the ²¹⁰Pb/²⁰⁶Pb blank ratio is in the 1e-14–1e-13 range. Measurements on 1-2 g Kapton films have established upper limits in the range 850-2500 mBq/kg at 90% C.L. Future ASTREA activities will focus on the Pb-210 assay in acrylic, which is considered for future low background dark matter detectors. Previous best results, obtained in 2014 by γ-counting 2 kg of acrylic, have established an upper limit for the Pb-210 concentration of 0.3 mBq/kg. Our proposed method, using AMS, should provide a limit of detection in the 0.01-0.1 mBq/kg range. Other important study looks at the Pb-210 contamination in the electroformation process of the copper for the NEWS-G and nEXO detectors. For the Pb-210 concentration in the copper, we estimate a limit of detection in the 0.3-1.0 mBq/kg range.