Reproducibility and accuracy of actinide AMS – lessons learned from precision studies for nuclear data
Prof. Anton Wallner1,2, Marcus Christl3, Michael A.C. Hotchkis4, Joerg Lippold5, Michaela Froehlich2, L. Keith Fifield2, Peter Steier6, Stephen Tims2, Stephan Winkler1
1 Helmholtz-Zentrum Dresden-Rossendorf (HZDR), Dresden, Germany, 2The Australian National University, Canberra, Australia, 3Laboratory of Ion Beam Physics, ETH Zurich, Zurich, Switzerland, 4Australian Nuclear Science and Technology Organisation (ANSTO), Sydney, Australia, 5Ruprecht-Karls-Universität Heidelberg, Institut für Umweltphysik, Heidelberg, Germany, 6VERA Laboratory, Faculty of Physics, University of Vienna, Vienna, Austria
Actinide detection has grown into an important discipline for environmental and geological sciences, for oceanography, e.g. as monitors of anthropogenic activities, but also in nuclear (astro)physics. Consequently, AMS measurements of actinides have become routine at many facilities. In particular, applications in nuclear (astro)physics continue to challenge the present limits in accuracy and abundance sensitivity of actinide detection.
Presently, there is a major ongoing effort in experiment and theory to better understand cross sections at thermal and higher neutron energies. These activities are motivated by the urgent need for improved and highly accurate nuclear data for optimised designs of advanced reactor concepts, nuclear fusion reactors, or next generation nuclear power plants (Gen IV) and accelerator driven systems (ADS). One example is the cross-section value for 235U neutron-capture at thermal energies: serving as a so-called thermal constant, this quantity is believed to be known to better than 1%. Despite its importance, direct measurements are rare (only two older data exist for thermal energies) and exhibit large uncertainties, thus its knowledge is based on indirect information.
For these applications, accurate actinide data are required, e.g. with uncertainties better than 2-3% for capture reactions. The combination of activation and subsequent AMS detection offers a powerful and complementary tool to measure these cross sections. However, this method had been applied only very recently for measurements on actinides. Importantly, adding an independent technique to established methods helps also to identify unrecognized systematic uncertainties in the existing nuclear database.
Several uranium and thorium samples had been irradiated with neutrons of energies between sub-thermal and 22 MeV at seven different neutron-producing facilities. These samples were then analysed at different AMS facilities: at the Vienna Environmental Research Accelerator (VERA), at ANSTO’s ANTARES, at ETH’s TANDY and at HIAF (ANU). These facilities cover terminal voltages for actinide AMS between 0.3 and 4 MV.
We present systematic investigations of nuclear data from a series of neutron-irradiated samples that were obtained by AMS. Long-lived reaction products that were measured include Th-229, Pa-231,233, U-233,236 and various Pu isotopes. Some irradiated samples were directly pressed into sample holders. Some samples were dissolved and spiked with well-known amounts of one or more reference isotopes, relative to which the radionuclides were quantified.
To achieve the highest accuracy, we compared the results from repeated measurements at the different facilities. We also had to take into account the measurement reproducibility of the individual facilities; an uncertainty component that represents unknown uncertainties beyond counting statistics and other known systematic uncertainties. A comparison of these data provides the present limits in the measurement accuracy of heavy-ion AMS.
Biography to come