Prototyping of an automated cyclic Neutron Activation Analysis setup

Sammanfattning: A prototype Neutron Activation Analysis (NAA) setup based on a Deuterium-Tritium (DT) neutron generator is under development at Lund University. It serves as proof-of-principle for NAA based on a Compact Accelerator-driven Neutron Source. One purpose of the setup is to test the possibility of providing environmental monitoring of pure alpha-emitters for the European Spallation Source currently under construction outside of Lund. As availability of reactor-based NAA is in decline within the European Union, alternative sources of neutrons are of great value. The DT-generator used in the current setup has a neutron yield of $\sim10^8$ n/s. The setup uses a fast pneumatic sample-transport system, which provides capabilities for measuring activation product half-lives in the sub-second range, and two HPGe detectors for gamma-ray spectroscopy providing the option of gamma-gamma coincidence measurements. In recent work, a Programmable Logic Controller-operated pneumatic- and data acquisition system, as well as new sample holders were implemented. This report provides an overview of the implemented improvements, as well as demonstration measurements for the different analysis types that can be performed with this setup. Samples measured include In, NaCl, Gd$_2$O$_3$ and soil samples from the surroundings of the ESS building grounds. The most important results feature a working automated cyclic-NAA prototype, with the capability of measuring samples with half-lives down to the sub-second region (transport time from irradiation to detectors of 350~ms). For pure samples the detection limits can currently reach down to the order of sub-mg while using a neutron generator with an output of only 4.7'10$^8$ n/s. The setup also features the capabilities to investigate the half-life of the desired lines, as well as coincidence measurements on samples where this technique is applicable. In order to get to the state of monitoring environmental samples, more work is needed to significantly push down detection limits in the future. This could include coincidence measurements, improved background subtraction and measuring samples with known mixed composition.