The radionuclide Ar-39 is produced in the atmosphere by cosmic rays and has an isotopic abundance of 8.1 x 10(-16). Because its half life (T-1/2 = 269 years) is well matched to the time periods involved in the oceanic currents around the Earth, the measurement of the Ar-39 isotopic ratio is an ideal tool to date ocean water from different depths. It would complement the information gained by the C-14 measurements (T-1/2 = 5730 years). However, the measurement of the isotopic ratio Ar-39/Ar-40 is a technical challenge: 1 L of modern ocean water contains similar to6500 atoms of Ar-39, and produces similar to17 decays per year. Although it has been possible to detect the Ar-39 decays in large volumes of sea water by using the low level counting technique, the possibility of measuring the number of Ar-39 atoms faster and in smaller samples using the accelerator mass spectrometry (AMS) technique would be a major breakthrough for this type of measurement. The development of a viable AMS method for Ar-39 has been underway for several years at Argonne National Laboratory, and is presently hampered by the presence of stable K-39 ions coming from the ion source. Although the intensity of this isobaric contaminant is low (similar topA extracted from the source), it has to be compared with the Ar-39 beam intensity (atoms per minutes). In order to separate, these two beams (whose mass difference is only 1.6 x 10(-5)), the intensity of the K-39 beam coming from the ion source has to be reduced by several orders of magnitude. This reduction has been investigated both at Argonne National Laboratory and at Louvain-la-Neuve. Two techniques have been tried out. In the first, a quartz liner is used to provide a clean surface, while in the second these impurities are buried in a SiO2 layer formed in situ by running the source with a mixture of silane and oxygen. The K-39 background has been reduced by a factor of 100 with these treatments. These techniques and their results obtained both at Argonne and Louvain-la-Neuve will be presented. The ion source specific requirements for this type of application will also be discussed. (C) 2004 American Institute of Physics.
Part 2827BFTimes Cited:2Cited References Count:4