Experimental decay-heat simulation-benchmark for 14 MeV neutrons & complex inventory analysis with FISPACT-II

Abstract

Nuclear inventory simulations have a vital role to play in the planning and execution of future fusion experiments and power plants. They are able to predict the transmutation (burn-up) response of material compositions under neutron irradiation, thus providing information about the build-up of impurities that could impact on material performance. The inventory evolution also quantifies the radiological response of a material by tracing the production (and decay) rates of radioactive nuclides. This information can be used to plan maintenance schedules at nuclear facilities, satisfy nuclear regulators during reactor planning, and quantify the waste disposal needs during reactor decommissioning. However, the validity of inventory simulations must be verified to give confidence in the predictions. This paper describes a validation and verification (V&V) benchmark exercise that tests the quality of nuclear code predictions. Such benchmarks are an important part of the development and release of the FISPACT-II inventory code and its associated input nuclear data libraries. This paper describes the latest V&V based on the fusion decay-heat measurements performed at the Japanese FNS facility. Rigorous and detailed assessment techniques, focussed on the complex breakdown of decay-heat contributions from individual radionuclides, have been employed to interpret the simulated results, benchmark the data against the experimental measurements, and to compare results from different international nuclear data libraries. Example results are presented and discussed for nickel, iron, niobium, tungsten, chromium, and osmium, using FISPACT-II simulations performed with TENDL-2017, JEFF-3.3, ENDF/B-VIII.0, EAF2010, and IRDFF-1.05 nuclear cross section libraries.

The joint evaluated fission and fusion nuclear data library, JEFF-3.3

“…The latter feature is shared with all other libraries except JEFF-3.1.2, but ENDF/B-VIII.0 and FENDL/A-3.0 are much closer to the data during the first 15 min due to the correct prediction of production of the 185m W metastable radionuclide via (n,2n) reactions on 186 W. For lead, JENDL-4 differs significantly from the data while the other libraries are a good match except during the first 10 min of cooling (controlled by 208 Pb(n,p) 208 Tl). Agreement between the simulations and the experiments for iron and chromium is less good due to experimental uncertainties, particularly beyond 40' min of cooling in Cr, where sample impurities (of aluminium and iron) are thought to play a role [400,404]. For copper the agreement with all libraries is excellent.…”

“…In tungsten, the simulations overestimate the decay heat during 400 days of cooling. As noted in [400], this overestimate could be caused by incorrect (n,2n) cross sections on W isotopes. In lead, there is excellent agreement between simulation and experiment during the first 20 days of cooling where 204 Pb(n,2n) 203 Pb dominates, but sig- .…”

“…80) during 1-h of cooling after a 5 min irradiation and for between 50 and 450 days of cooling following 7 h irradiations ( Fig. 81) [400,403]. These elements were chosen as examples here, as they are important constituents of steel (Cr, Fe, Ni), of a fusion device's divertor and inner wall (W), and of the electric and power, and cooling infrastructure (Cu) of fusion reactors.…”

“…Comparison between calculated and experimental cross sections for three versions of the JEFF library as well as JENDL-4.0 and ENDF/B-VIII.0, taken from the FNS integral decay-heat benchmark[400]. Calculation/experiment (C/E) ratios for reactions are plotted against the baryon number of the daughter nuclide.…”

Nuclear data activities of the EUROfusion consortium

“…Comprehensive verification and validation (V&V) analyses were performed on both the activation data library in (groupwise) EAF format and the underlying TENDL-2017 data library. CCFE's validation data base was applied to this end including a wide range of differential and integral measurements, resonance integrals, thermal cross-sections, astrophysical data, and decay heat data [29], [30]. The V&V analyses were further complemented with statistical checks.…”