Detection of low levels of plutonium in natural environments from gamma-ray spectra with advanced methods in robust fitting

A newly developed spectral analysis methodology, RobWin, is experimentally tested for its ability to identify plutonium in highly background-dominated gamma-ray spectra. The method, which builds upon nonlinear robust fitting techniques from RobFit spectral analysis software, emphasizes gradually adding structure to the entire spectrum background shape function after accounting for re-optimized combinations of entire-spectrum photopeak response functions from a user-defined set of nuclides at each iteration. This new feature makes it uniquely suited to the identification of weak-strength nuclides in high-background environments. The method is briefly described and applied to a series of background-dominated laboratory gamma-ray spectra from a 98 nCi 239Pu source with logarithmically varying collection times taken with a 100% n-type HPGe detector. The laboratory experiment, the data, and the analysis results are presented. As a measure of confidence in the effectiveness of the several entire-spectrum features of the new approach, comparisons of confidence levels, and detectable total isotopic counts are made with analysis of the same data by fitting only peaks with these new features disabled. It is concluded that for similar applications the new method can improve detectable strength and confidence of detection significantly.