Speaker
Description
This work assesses the impact of nuclear data processing parameters on processed cross sections and criticality safety benchmark results obtained using the nuclear data processing code NJOY. The study focuses on the RECONR and BROADR modules, which govern resonance reconstruction and Doppler broadening, respectively. The main parameters of these modules, including reconstruction tolerance, integral criterion, thinning tolerance, and thinning energy cutoff, are systematically varied around their default values.
For each parameter variation, selected isotopes from the JEFF-4.0 library, including U-235, U-238, Pu-239, Pu-241, Fe-56, and Fe-57, are reprocessed, and the resulting cross sections are compared with a default baseline. Energy-dependent relative differences are analyzed for isotopes of interest in criticality safety through four key reaction channels: total (MT=1), elastic (MT=2), fission (MT=18), and radiative capture (MT=102). Particular attention is given to the resonance region, where numerical processing choices may significantly affect the reconstructed energy grid and the shape of the cross sections. These differences are then propagated to selected ICSBEP benchmarks sensitive to the isotopes under study, covering thermal, epithermal, and fast neutron spectra.
The objective is to identify which processing parameters produce measurable changes in processed cross sections and, more importantly, whether these differences lead to significant effects on keff in criticality safety applications. Beyond the present scope, this work also prepares the ground for a future extension to the THERMR and PURR modules, in order to examine the influence of thermal scattering treatment and unresolved resonance probability tables. Overall, the study aims to provide a clear and practical contribution to the understanding of numerical uncertainties associated with nuclear data processing.
| Session | Processing and Benchmarking |
|---|