2023, Issue 3

EDN: NQRHWM

Authors & Affiliations

Zhukov V.D., Listov A.S., Sinitsa V.V.
National Research Centre “Kurchatov Institute”, Moscow, Russia

Zhukov V.D. – Research Engineer. Contacts: 1, Akademika Kurchatova sq., Moscow, Russia, 123182. Tel.: +7 (499) 196-71-00 (add 6423); e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it..
Listov A.S. – Researcher.
Sinitsa V.V. – Leading Researcher, Cand. Sci. (Phys.-Math.).

Abstract

Within the framework of the project on verification of processing programs initiated by the IAEA Nuclear Data Section, data on neutron scattering in the thermal energy range from the ENDF/B-VIII.0 were processed and files in ACE format were prepared for calculations using the MCNP program. The processing was carried out using different algorithms and types of representation of double-differential cross-sections, which led to differences in the ACE files and the results of criticality calculations performed with these files. The paper considers the reasons for discrepancies in differential data, and by comparing the results of calculations of integral experiments, the contributions to the error keff due to the choice of the grid of initial energies and the type of representation of energy-angular distributions are estimated.
Processing of the estimated data was carried out using the processing codes GRUCON-2021 and NJOY2016.62 with the energy cross-section interpolation accuracy parameter eps=0.1 %. Criticality calculations and error estimates used ACE files obtained from each of the programs with data for fast and thermal neutrons for 132 materials, including data for 7 nuclei in bound states, namely: H(H₂O), Be(metallic), Be(BeO), O(BeO), C(graphite), H(ZrH_n), and Zr(ZrH_n). The neutron scattering cross-sections in the thermal region differed in the initial energy grids, since in NJOY it is fixed for all materials, while in GRUCON it is chosen adaptively, according to a given interpolation accuracy.
k_eff calculations were carried out for 87 benchmarks from the ICSBEP collection, which include the above linked nucleus. The scatter of the obtained k_eff values associated with the choice of the grid of initial energies for most of the assemblies did not exceed the expected 0.1 %, except for the assemblies with zirconium hydride (~0.3 %). A possible reason is the interpolation error of the fixed grid of initial energies: as a comparison of detailed cross-sections showed, this error is several percent, with a maximum value of ~4 % in the case of H(ZrH_n).
To estimate the error associated with the presentation of energy-angular distributions specified by the user (discrete, equiprobable values, or continuous, tabular), the results obtained by other project participants for H(ZrH_n) are used. It was found that this error is also 0.3 %, which may be unacceptable for precision calculations.

Keywords
evaluated nuclear data files, thermal neutron scattering, processing error, keff calculation, Monte Carlo method

Article Text (PDF, in Russian)

References

UDC 621.039.5

Problems of Atomic Science and Technology. Series: Nuclear and Reactor Constants, 2023, issue 3, 1:3