Gamma-spectroscopic determination of ²³⁵U in environmental samples using Low Energy Germanium detectors (LEGe)

The determination of ²³⁵U in environmental samples from its 185.72 keV photons requires the deconvolution of the multiplet photopeak at 186 keV, which is the result of the co-existence of the 186.25 photons of ²²⁶Ra in the spectrum. This deconvolution depends highly on the ²³⁵U and ²²⁶Ra content of the sample, as well as the continuous background of the spectrum. Furthermore, in the case of soil samples contaminated with ¹³⁷Cs, the analysis is further complicated in as much as the multiplet photopeak lies in the vicinity of the ¹³⁷Cs 661.62 keV photon backscatter region. This work aimed to investigate the potential and the limitations of using a high-resolution LEGe detector for the deconvolution of the multiplet photopeak at 186 keV and the detection of the 185.72 keV photons. The investigation was based on the analysis of two series of specially composed LEGe detector test spectra. The first series, which consists of 140 spectra, simulated spectra of environmental samples containing only natural radioactivity, with various levels of ²²⁶Ra and ²³⁵U. The second series consists of 280 spectra simulating spectra of environmental samples containing various levels of ²²⁶Ra and ²³⁵U and contaminated with various levels of ¹³⁷Cs. The analysis was performed using the in-house developed SPUNAL computer code. In both series test spectra, the analysis of the 186 keV photopeak was considered acceptable when it was deconvoluted into its two components and both ²²⁶Ra and ²³⁵U activities were accurately determined.

The analysis of the test spectra allowed for very useful conclusions regarding the deconvolution of the 186 keV photopeak to be drawn, the most important being that the presence of ¹³⁷Cs in the sample has a negative effect on the analysis of the 186 keV photopeak. The results of this work show that the 186 keV photopeak may be analyzed in most cases into its two components; however, thorough investigation of the effect of the various parameters affecting this analysis should be performed. For the given detector and settings, the erroneous analysis of the 186 keV photopeak is mainly attributed to poor statistics. The dependence of the deconvolution results on the computer code used and should be further investigated by analyzing the test spectra with other gamma spectroscopy analysis computer codes.

The Project was co-funded by the European Social Fund (75%) and National Resources (25%)-(EPEAEK II)-PYTHAGORAS.