Performance Evaluation of Spectroscopic Detectors for LEU Hold-up Measurements

The hold-up measurement of low-enriched uranium materials may require use of alternate detector types relative to the measurement of highly enriched uranium. This is in part due to the difference in process scale (i.e., the components are generally larger for low-enriched uranium systems), but also because the characteristic gamma-ray lines from ²³⁵U used for assay of highly enriched uranium will be present at a much reduced intensity (on a per gram of uranium basis) at lower enrichments. Researchers at Oak Ridge National Laboratory examined the performance of several standard detector types, e.g., NaI(Tl), LaBr3(Ce), and HPGe, to select a suitable candidate for measuring and quantifying low-enriched uranium hold-up in process pipes and equipment at the Portsmouth gaseous diffusion plant. Detector characteristics, such as energy resolution (full width at half maximum) and net peak count rates at gamma ray energies spanning a range of 60-1332 keV, were measured for the above-mentioned detector types using the same sources and in the same geometry. Uranium enrichment standards (Certified Reference Material no. 969 and Certified Reference Material no. 146) were measured using each of the detector candidates in the same geometry. The net count rates recorded by each detector at 186 keV and 1,001 keV were plotted as a function of enrichment (atom percentage). Background measurements were made in unshielded and shielded configurations under both ambient and elevated conditions of ²³⁸U activity. The highly enriched uranium hold-up measurement campaign at the Portsmouth plant was performed on process equipment that had been cleaned out. Therefore, in most cases, the thickness of the uranium deposits was less than the ''infinite thickness'' for the 186 keV gamma rays to be completely self-attenuated. Because of this, in addition to measuring the 186 keV gamma, the 1,001 keV gamma ray from ^234mPa-a daughter of ²³⁸U in secular equilibrium with its parent-will also need to be measured. Based on the performance criteria of detection efficiency, energy resolution, peak-to-continuum ratios, minimum detectable limits, and the weight of the shielded probe, a shielded (0.5 in. thick lead shield) 2 x 2 in. NaI(Tl) detector is recommended for use. The recommended approach is to carry out analysis using data from both 186 keV and 1,001 keV gamma rays, and select a best result based on propagated uncertainty estimates. It is also highly recommended that a two-point gain stabilization scheme based on an ²⁴¹Am seed embedded in the probe be implemented. Shielding configurations to reduce the impact of background interference on the measurement of 1,001 keV gamma-ray are discussed.