Speaker
Description
Iron is an important structural and shielding material that appears in many applications. Nuclear reaction data of iron have been addressed within the Subgroup-40 (CIELO) of the OECD/NEA Data Bank [1]. The resulting evaluated data files from the BNL/IAEA collaboration for the iron isotopes [2] were included in the ENDF/B-VIII.0 library [3] and performed well in criticality benchmarks. Unfortunately, inadequate performance of the $^{56}$Fe evaluation was discovered for shielding benchmarks just before the release of the ENDF/B-VIII.0 library: the leakage spectra from very thick iron shells with a $^{252}$Cf source in the centre were significantly under-predicted in the energy range 1-8 MeV.
Additional analysis was performed which showed that the root cause of the problem was the non-elastic cross section, which seemed to be too high. This hypothesis was in contradiction with the new measurement of the inelastic cross section by Beyer et al [4] but was partially confirmed by a new measurement of the elastic cross section [5]. A compromise between the measured elastic and the inelastic cross section was sought such that unitarity with the total cross section was preserved. In addition, the elastic cross section was increased in the resonance interference minima (being the largest increase around 300 keV). The increase was guided by the measured leakage spectra from thick iron shells with a $^{252}$Cf source in the centre to eliminate the observed overestimation by 30% of the measured neutron leakage around 300 keV. Some of the improvements were already presented orally at various meetings, but no formal report or publication has been released.
There was also a question of the thermal capture cross section. R. Firestone [6] published the results from prompt-gamma activation analysis, which suggested lowering of the thermal capture cross section by 7%. Validation measurements at CEA within the MAESTRO project [7] did not support such a decrease [8]. The same validation showed that JEFF evaluation was essentially correct, so the recommended value of the thermal capture cross section for $^{56}$Fe of 2.577 barns was adopted. This value is also in agreement with latest Firestone evaluation [8].
The resulting evaluated data files were extensively tested. The previously observed deficiency of under-predicting the fast-neutron leakage spectra from thick iron shells with a $^{252}$Cf source was avoided. The 30% overestimation of the neutron leakage around 300 keV was also eliminated. Similar benchmark experiments with a D-T source also showed an improvement. In a benchmark involving 40 MeV neutrons the new evaluation performed much better than the ENDF/B-VII.1 library. At the same time, performance in criticality benchmarks was not compromised.
The benchmarking exercise illustrates how integral benchmarks can be used to discriminate between discrepant differential data to produce evaluated data files that perform well in situations that are critical for the safe operation of nuclear installations. Details of the changes to the differential data will be given and selected representative benchmark results will be presented.
References
[1] M.B. Chadwick et al.: “CIELO Collaboration Summary Results: International Evaluations of Neutron Reactions on Uranium, Plutonium, Iron, Oxygen and Hydrogen”, Nuclear Data Sheets 148 (2018) 189–213.
[2] M.W. Herman et al: “Evaluation of Neutron Reactions on Iron Isotopes for CIELO and ENDF/B-VIII.0”, Nuclear Data Sheets 148 (2018) 214–253.
[3] D.A. Brown et al.: “ENDF/B-VIII.0: The 8th Major Release of the Nuclear Reaction Data Library with CIELO-project Cross Sections, New Standards and Thermal Scattering Data”, Nuclear Data Sheets 148 (2018) 1–142.
[4] R. Beyer et al.: “Inelastic scattering of fast neutrons from 56Fe”, EPJ Web of Conferences 146, 02017 (2017).
[5] E. Pirovano et al.: “Cross section and neutron angular distribution measurements of neutron scattering on natural iron”, Physical Review C 99, 024601 (2019)
[6] R.B. Firestone et al.: “Thermal neutron capture cross section for 56Fe(n,γ )”, Physical Review C 95, 014328 (2017).
[7] P. Leconte, “Nuclear data feedback on structural, moderating and absorbing materials through the MAESTRO experimental programme in MINERVE,” Tech. Rep. JEF/DOC-1849, Nuclear Energy Agency, OECD, Paris, France (2017).
[8] R.B. Firestone, private communication, to be published (2020).