Integral Test of Evaluated Nuclear Data Files

Criticality Benchmark for Thermal Reactors

    1. Okumura, Y. Nagaya, T. Mori: “Criticality Benchmarks with a Continuous-Energy Monte Carlo Code MVP and JENDL-3.3,” Proc. of the 2004 Symposium on Nuclear Data, Nov. 11-12, 2004, JAERI, Tokai, Japan, pp.56-63, JAERI-Conf 2005-003, (2005).
Criticality benchmark testing of JENDL-3.3 was performed with MVP for many critical experiments in the ICSBEP Handbook: highly enriched uranium (HEU) and low enriched uranium (LEU) fueled solutions, LEU fueled water-moderated lattices, MOX fueled water-moderated lattices and metal fueled small assemblies with hard neutron spectra. The calculations with the preliminary U235 and U238 data for ENDF/B-VII (preVII) are also made for some benchmarks to confirm their performances and to compare with the JENDL-3.3 results. The prediction accuracy of criticality with JENDL-3.3 is satisfactory, within ±0.5%Δk in most cases, except for slightly enriched uranium fueled lattices. The preVII data of U238 with lower thermal capture cross section improves the under prediction of keff for the slightly enriched lattices, while slight overestimations were observed for the LEU lattices whose U235 enrichment is more than 3wt.%. On the other hand, the preVII data gives excellent results for the metal fueled assemblies compared with JENDL-3.3.

PDF file (166KB)

  • K.Okumura, and T.Mori: “Integral Tes of JENDL-3.3 for Thermal Reactors,” Proc. of 2002 Symposium on Nuclear Data (JAERI-Conf 2003-006)
Criticality benchmark testing was carried out for 59 experiments in various thermal reactors using a continues-energy Monte Carlo code MVP and its different libraries generated from JENDL-3.2, JENDL-3.3, JEF-2.2 and ENDF/B-VI (R8). From the benchmark results, we can say JENDL-3.3 generally gives better keff values compared with other nuclear data libraries. However, further modification of JENDL-3.3 is expected to solve the following problems: 1) systematic underestimation of keff depending on U-235 enrichment for the cores with low (less than 3wt.%) enriched uranium fueled cores, 2) dependence of C/E value of keff on neutron spectrum and plutonium composition for MOX fueled cores. These are common problems for all of the nuclear data libraries used in this study.

PDF file (208KB)

    1. Okumura and T. Mori : “Investigation of Dependence of Criticality Evaluation Accuracy on U-235 Enrichment in Light Water Moderated Uranium Fueled Systems”, pp. JAERI-Review 2003-023 , pp.59-61 (2003)

PDF file (245KB)

Criticality Benchmark for Fast Reactors

  • Shigeaki Okajima, Masahiro Fukushima, and Masaki Andoh: “Benchmark Test for JENDL-3.3 Library by Analyses of a Series of Experiments at the Fast Critical Assembly (FCA) of JAERI,” Proc. of International Conf. of Nuclear Data for Science and Technology (2005).
To validate the JENDL-3.3 library, eight different FCA cores were selected and were used for the benchmark test on criticality (keff value). In the test, the Monte Carlo calculation code was used. The results were compared with those of the JENDL-3.2 library. The spectrum dependence of the C/E values with the JENDL-3.3 library was found in the softer neutron spectrum cores. This spectrum dependence was caused by the revised capture cross section of U-235 in the JENDL-3.3 library.

PDF file (1.15MB)


Post Irradiation Experiment Analysis

  • K.Okumura : Post irradiation experiment analysis for UO2 Fuels irradiated in PWR (Takahama-3) with SRAC and recent nuclear data libraries : JENDL-3.2, JENDL-3.3, JEF-2.2, ENDF/B-VI(R5), ENDF/B-VI(R8)

Sample fuels : 4.1wt.% U-235 enriched UO2

Number of samples : 10 (SF95-1, SF95-2, SF95-3, SF95-4, SF95-5, SF97-2, SF97-3, SF97-4, SF97-5, SF97-6)

Burn-up up range of samples : 14GWd/t-47GWd/t


Sample averaged C/E values for atomic number density (Heavy nuclides)


Sample averaged C/E values for atomic number density (Fission products)

Details are published in

Nuclear Data News, Vol.110, No.74, pp.39-55 (2003)

PDF file (408KB, in Japanese)