Measurements of Low-Enriched Uranium Holdup

A recent effort determined uranium holdup at a large fuel fabrication facility abroad where low enriched (∼ 3%) uranium (LEU) oxide feeds the pellet manufacturing process. Measurements taken with both high- and low-resolution gamma-ray spectrometry systems include extensive data for the ventilation and vacuum systems. Equipment dimensions and the corresponding holdup deposit masses are large for LEU. Because deposits are infinitely thick to the 186 keV gamma ray in many locations in an LEU environment, measurements of both the 186 and 1001 keV gamma-rays were required, and self-attenuation was significant at 1001 keV in many cases. These wide-dynamic-range measruements used short count times, portable scintillator detectors, and portable MCAs. Because equipment is elevated above floor levels, most measurements were made with detectors mounted on extended telescoping poles. One of the main goals of this effort was to demonstrate and validate methods for measurement and quantitative analysis of LEU holdup using low-resolution detectors and the Generalized Geometry Holdup (GGH) techniques. The current GGH approach is applied elsewhere for holdup measurements of plutonium and high-enriched uranium. The recent experience is directly applicable to holdup measruements at LEU facilities such as the Paducah and Portmouth gaseous diffusion enrichment plants and elsewhere, including LEU sites where D and D is active. This report discusses the measurement methodology, calibration of the measurement equipment, measurement control, analysis of the data, and the global and local assay results including random and systematic uncertainties. It includes field-validation exercises (multiple calibrated systems that perform measruements on the same extended equipment) as well as quantitative validation results obtained on reference materials assembled to emulate the deposits in an extended vacuum line that was also measured by these techniques. The paper examines the differences in assay results between the low-resolution system using the GGH method and the high-resolution system utilizing the commercially available ISOCS analysis method.