ADVANTG hybrid code application for the analysis of neutron field depression in uranium foils activated in the MARK III converter of the BR1 reactor

Research outputpeer-review

Abstract

This paper deals with the self-shielding correction calculation for natural and 40% enriched uranium foils activated in a fast-epithermal neutron field in the MARK III converter of the BR1 research reactor. Although the irradiated uranium foils are thin (~70 μm to 90μm), neutron self-shielding is not negligible. The calculation of the neutron field depression in the foils was challenged by unreliable results and long computing times using the MCNP6.2 Monte Carlo code. These simulations require the use of the variance reduction techniques to compute a detector response with low uncertainties within an acceptable computing time. An additional problem was the convergence of the spectrum averaged cross section of the 238U(n,γ) reaction, even with a high number of particle histories (e.g. 1×1010), even when the average neutron flux has low statistical uncertainty. To speed up the calculations and to test the reliability of variance reduction techniques, a comparison was made with the tally results and statistical uncertainties of the MCNP6.2 weight window technique, ADVANTG-accelerated MCNP6.2 and default MCNP6.2 calculations. To generate variance reduction parameters with the ADVANTG code for the final calculations, the consistent adjoint driven importance sampling (CADIS) method is used. The neutron self- shielding has been found considerable in the 238U(n,γ) reaction due to three strong resonances (at 7 keV, 21 keV and 37 keV) leading to a correction factor of 1.57 for the natural uranium foil. The sensitivity study has shown a negligible impact from the selected nuclear data libraries (JEFF and ENDF/B) on the 238U(n,γ) and 235U(n,f) spectrum averaged cross sections in the MARK III converter
Original languageEnglish
Article number110384
Number of pages11
JournalNuclear Engineering and Design
Volume357
DOIs
StatePublished - Feb 2020

Cite this