Extending neutron activation analysis to materials with high concentrations of neutron absorbing elements

NASA Astrophysics Data System (ADS)

The purpose of this study was to investigate epithermal neutron self-shielding for all nuclides used in Neutron Activation Analysis, NAA. The study started with testing the theory and measuring the nuclear factors characterizing thermal and epithermal self-shielding for 1 mL cylindrical samples containing the halogens Cl, Br and I irradiated in a mixed thermal and epithermal neutron spectrum. For mono-element samples, both thermal and epithermal experimental self-shielding factors were well fitted by sigmoid functions. As a result, to correct thermal neutron self-shielding, the sigmoid uses a single parameter, mth, which can be directly calculated for any element from the sample size, the weighted sum of the thermal absorption cross-sections, sigmaabs, of the elements in the sample and a constant kth characteristic of the irradiation site. However, to correct epithermal self-shielding, the parameter mep, a function of sample geometry and composition, irradiation conditions and nuclear characteristics, needs to be measured for each activated nuclide. Since the preliminary tests were positive and showed that self-shielding, as high as 30%, could be corrected with an accuracy of about 1%, except in cases with significant epithermal shielding of one element by another, we pursued the study with the verification of two additional aspects. First, the dependency of the self-shielding parameters mth, and mep, on the properties of the irradiation site was evaluated using three different irradiation sites of a SLOWPOKE reactor, and it was concluded that the amount of both thermal and epithermal self-shielding varied by less than 10% from one site to another. Second, the variation of the self-shielding parameters, mth, and mep, with the size of the cylinder, as r( r+h), was tested for h/r ratios from 0.02 to 6.0, and this geometry dependence was confirmed even in slightly non-isotropic neutron fields. These results allowed separating from the mep parameter the amount of chemical element and the sample geometrical factor. Therefore, the remaining nuclear factor, considered as a product of nuclide composite nuclear characteristics and irradiation site characteristics, led to the introduction of a so-called epithermal neutron absorption cross-sections, sigmaabs,ep. This new nuclear parameter will allow the calculation of the epithermal self-shielding for all cylindrical samples activated in all types of irradiation sites. For the 13 cases studied, the epithermal self-shielding factor, Gep, was obtained from the experimental effective self-shielding factor, Geff, by extracting the thermal neutron self-shielding factor, calculated with the sigmoid formulation. A least-squares fit of the experimental Gep values as a function of the mass of element yielded sigmaabs,ep for each activated nuclide. In addition, for all nuclides commonly used in neutron activation analysis, sigmaabs,ep was calculated with the Martinho, Salgado and Goncalves sigmoid formulation, which uses the total cross-section values at the peaks of the resonances. A comparison of the calculated sigmaabs,ep with the 13 measured values reveals that the calculated values are accurate to about 20%. Finally, for all 76 nuclides commonly used in NAA, a spreadsheet program was written to use experimental or calculated sigmaabs,ep nuclear parameters to perform iterative self-shielding corrections of concentrations measured by neutron activation analysis. The user provides the parameters f and alpha of the neutron spectrum, the sample mass and dimensions, and the measured concentrations. In a typical case with 10% thermal self-shielding and 30% epithermal self-shielding, the corrected concentrations had uncertainties varying from 2% to 3%. Keywords. Instrumental Neutron Activation Analysis, epithermal, thermal, self-shielding factors. (Abstract shortened by UMI.)