Gamma spectrometer measurements of microgram quantities of plutonium using uranium L X-rays in the 13–21 keV region

Nineteen groups of Marple series cascade impact filters were used to collect aerosol particles associated with various operations in plutonium glove boxes. The cascade impactor separated particles based upon size, while the total mass of plutonium deposited on the filters ranged from 10−4 to 10−8 g. Gamma spectrometer measurements with an n-type high-purity germanium detector (HPGe) were performed with the goal of accurately quantifying the 239Pu mass on each filter using counts from the uranium L X-rays in the 13–21 keV region. However, the mass was great enough in some of the measurements to also quantify using the 239Pu gamma ray at 129.29 keV. One-half portions of each filter were also analyzed using thermal ionization mass spectroscopy (TIMS) and comparisons between the two methods were performed. A significant challenge was to find a calibration source with very low self-absorption characteristics so that the detector response to the low-energy X-rays could be properly characterized. This was accomplished by doping an actual Marple filter with a small NIST traceable quantity of 239Pu (7.3 μg). The comparison of gamma spectrometry to TIMS results yielded largely favorable results. Some notable sources of measurement bias beyond counting statistics were uncovered in the process. The particle distributions on the filters were not perfectly symmetrical, and since the TIMS analyses were performed on just one-half of a filter, a significant difference could result. Replicate filter gamma measurements revealed potential biases in the reproducibility of sample geometry, and high concentrations of 241Am appeared to cause L-shell X-ray fluorescence effects in two of the filter sets.