Neutron-capture cross sections: experimental activation and nuclear-data analysis

Accurate neutron-capture cross sections are essential for activation analysis, reactor physics, fusion technology, dosimetry, and nuclear fuel-cycle applications. However, their determination is affected by experimental uncertainties and by the nuclear models adopted in theoretical calculations. This thesis investigates these aspects through an experimental study of gold activation and a broader analysis of selected neutron radiative-capture reactions. The overall aim of this work is to compare experimental information with nuclear-model predictions and to identify the main sources of model spread. The experimental study focused on the ¹⁹⁷Au(n,γ)¹⁹⁸Au reaction. Two gold foils were irradiated using Am–Be neutron sources and measured with NaI(Tl) and LaBr₃(Ce) scintillation detectors. The induced activity was determined from the 411.8 keV gamma emission of ¹⁹⁸Au, leading to an effective cross-section estimate of 8.12 ± 1.71 mb at a representative neutron energy of about 4.2 MeV. The analysis was then extended to the ⁵¹V(n,γ)⁵²V, ¹⁰³Rh(n,γ)¹⁰⁴Rh, ²⁰⁹Bi(n,γ)²¹⁰Bi, and ²³²Th(n,γ)²³³Th reactions. EXFOR measurements were compared with evaluated nuclear data libraries and TALYS calculations obtained using different Level Density and Photon Strength Function models. Restricted chi-square, empirical covariance, and sensitivity analyses were used to quantify agreement and identify the origin of the model spread. The experimental value obtained for ¹⁹⁷Au is close to the mean TALYS prediction at the representative neutron energy. For several reactions, the largest model spreads were mainly associated with a few TALYS combinations clearly separated from the others. However, the reliability of neutron capture predictions is strongly dependent on both isotope and energy, showing that experimental results, evaluated libraries, and model calculations should be treated as complementary sources of information.