Comparison of the TRIPOLI-4®, DIANE, and MCNP6 Monte Carlo Codes on the Barber & George Benchmark for Photonuclear Reactions

Abstract

In view of their key role in radiation shielding and nuclear instrumentation applications, photonuclear reactions are receiving growing attention. In this work, we compare the results of the Monte Carlo codes TRIPOLI4®, DIANE, and MCNP® with respect to the Barber and George (B&G) benchmark, with the aim of assessing the accuracy of both nuclear data and particle transport codes for the simulation of photonuclear reactions. We compute the photoneutron yield resulting from the Bremsstrahlung radiation induced by a monoenergetic electron beam (10.5 to 35.5 MeV) impinging on C, Al, Cu, Pb, Ta, and U material targets. The simulation specifications closely follow those of the B&G experiment. For all codes, the reference nuclear data libraries are ENDF/B-VII.1 for neutron transport and photonuclear reactions and EPDL97/EEDL97 for photon/electron transport. Comparisons of the simulation results show an overall agreement between the codes and experimental data and in-between codes, despite some discrepancies.

In order to investigate these effects, we performed a sensitivity analysis by tallying the photon production in addition to neutron production by replacing the electron source with a pure photon source (to single out the impact of electron transport) and by replacing the ENDF/B-VII.1 library with the IAEA/PD-2019. The major contribution to the observed discrepancies is found to be related to the electromagnetic shower models used for coupled electron-photon transport in Monte Carlo codes.

Keywords:

• Photonuclear
• Monte Carlo
• electromagnetic shower
• neutrons
• photons

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

^a See also the recent work at https://arxiv.org/abs/2209.13947.

^b In addition to having rather large overall statistical uncertainties, it has been pointed out that one of the average values of the B&G benchmark is possibly questionable (see case Ta-I for an incident electron energy of 10.3 MeV, where the experimental value for the integrated neutron current is around 80, whereas all codes yield an average tally of about 8) (Ref. 16). For an overview of the experimental results, see Table III.

^c For these calculations, we activated the forced photonuclear reactions in MCNP in order to improve the statistical convergence (see also numerical values in Table III).

Additional information

Funding

We acknowledge the financial support of the Cross-Disciplinary Program on Numerical Simulation of the CEA, the French Alternative Energies and Atomic Energy Commission. TRIPOLI-4 is a registered trademark of CEA. The authors thank EDF for partial financial support.