Sample matrix influence on the efficiency function modeling for uranium isotopes determination by gamma spectrometry

Abstract

The aim of the presented study was to explore the influence of precise determination of sample matrix in order to obtain reliable gamma spectrometry results. The test was done using two different approaches which have the same goal: quantifying the activity concentration of uranium isotopes in different samples found in the former uranium ore deposit Mazdreja. The first approach is based on iterative methodology, which utilizes the characteristic of a sample that can be quantified before gamma spectrometric measurement. Density of the sample and calculated mass fractions of its constituents (which vary in each iteration) can help to obtain the efficiency function from which activity concentrations of 235U and 238U can be derived and corresponding masses of uranium for the next iteration. In this approach LabSOCS software is used in order to obtain appropriate efficiency. The second approach that was tested represents semiempirical method which applies Monte-Carlo simulations through Geant4 and EFFTRAN software codes for the purposes of generating the detection efficiency functions needed for further calculations.The idea was to compare different methods for uranium activity concentration determination in sample matrices with high uranium mass content and also establish the need for precise calculation (described within the iterative method) by quantifying the relative deviation in activities, calculated by using different methods.From the obtained results, it can be seen that it is very important to adjust the sample matrix to correct composition if we need a precise gamma spectrometry result. But if there is a need to get results in a short period of time, then for the estimation of the activity concentration of the analyzed sample an approximate sample composition can be used. All three tested analyzed methods gave comparable results inside 15% of difference. The major difference was obtained by EFFTRAN software, but on the other side this software is user-friendly and the results can be obtained faster than by using other analysis methods presented in this study.