Onishi, Seiki*; Kondo, Keitaro*; Azuma, Tetsushi*; Sato, Satoshi; Ochiai, Kentaro; Takakura, Kosuke; Murata, Isao*; Konno, Chikara
Fusion Engineering and Design, 87(5-6), p.695 - 699, 2012/08
A new integral experiment with a deuteron-triton fusion (DT) neutron beam started in order to validate scattering cross section data. First the DT neutron beam was constructed with a collimator. The characteristics of the DT neutron beam were examined experimentally. Second a new integral experiment for type 316 stainless steel (SS316) was carried out with this DT neutron beam. Reaction rates of the Nb(n,2n)Nb reaction on the center of the beam axis and at 15 cm and 30 cm apart from the axis in the assembly were measured with the activation foil method and were calculated with the Monte Carlo transport calculation code MCNP and nuclear data libraries, JENDL-4.0, JENDL-3.3 and ENDF/B-VI.8. The ratios of calculation to experiment became smaller than 1 with the distance from the beam axis for all the nuclear libraries. It was pointed out that the diagonally forward cross section data had some problems.
Kondo, Keitaro; Yagi, Takahiro*; Ochiai, Kentaro; Sato, Satoshi; Takakura, Kosuke; Onishi, Seiki; Konno, Chikara
Fusion Engineering and Design, 86(9-11), p.2184 - 2187, 2011/10
In the neutronics experiment for the ITER test blanket module with a Li-enriched LiTiO layer and a beryllium layer conducted at the FNS facility of Japan Atomic Energy Agency, the calculated tritium production rate (TPR) was by approximately 10% larger than the measured one only when a neutron source reflector composed of SS316 was attached. On the other hand, the influence of the reflector on the TPR prediction accuracy was not seen in the recent blanket experiment with a natural LiTiO layer, beryllium layers and the reflector. We investigated the former experiment in detail, and found an unphysical tendency in the measured TPR distribution. In order to clarify whether the deterioration of the TPR prediction accuracy originates from the reflector or not, we have conducted the same experiment as the previous experiment again. In the present experiment, the measured TPR distribution inside the Li-enriched LiTiO layer well agreed with the calculated one within an estimated experimental error of 6%. We conclude that the overestimation of TPR observed in the previous experiment would be due to some experimental errors and that the TPR prediction accuracy is good even in the case with the reflector.
Konno, Chikara; Wada, Masayuki*; Kondo, Keitaro; Onishi, Seiki; Takakura, Kosuke; Ochiai, Kentaro; Sato, Satoshi
Fusion Engineering and Design, 86(9-11), p.2682 - 2685, 2011/10
JENDL-4, the major revised version of Japanese Evaluated Nuclear Data Library (JENDL), was released in spring, 2010. We analyzed the fusion neutronics benchmark experiments on iron at JAEA/FNS with JENDL-4.0 and MCNP4C as the detail benchmark test of JENDL-4.0 iron data. As a result, it is found out that the problems of iron data in JENDL-3.3 are adequately revised in JENDL-4.0 iron data; e.g. the first inelastic scattering cross section data of Fe and angular distribution of elastic scattering of Fe. The iron data in JENDL-4.0 are comparable to and are partly better than those in ENDF/B-VII.0 and JEFF-3.1.
Konno, Chikara; Takakura, Kosuke; Kondo, Keitaro; Onishi, Seiki*; Ochiai, Kentaro; Sato, Satoshi
Progress in Nuclear Science and Technology (Internet), 2, p.341 - 345, 2011/10
We have already pointed out that the background cross sections and weighting flux are not adequate in multigroup libraries VITAMIN-B6 and JSSTDL-300. This time we examined if the latest multigroup libraries, MATXS-J33, MATJEFF3.1.BOLIB, VITJEFF3.1. BOLIB, VITENEA-J, HILO2k and AMPX file of ENDF/B-VII.0 in SCALE6 and MTXS file in ADS-2.0 have the same problems. The followings are found out from our simple calculations. (1) MATXS-J33 has no problem. (2) VITJEFF3.1.BOLIB, VITENEA-J, HILO2k and AMPX file of ENDF/B-VII.0 in SCALE6 are produced by using the inadequate weighting flux. (3) VITJEFF3.1.BOLIB, MATJEFF3.1.BOLIB and MTXS file in ADS-2.0 have inadequate background cross sections. Note that the self-shielding correction in calculations with VITJEFF3.1.BOLIB, VITENEA-J, HILO2k and AMPX file of ENDF/B-VII.0 in SCALE6, MATJEFF3.1.BOLIB and MTXS file in ADS-2.0 is not always adequate.
Konno, Chikara; Takakura, Kosuke; Wada, Masayuki*; Kondo, Keitaro; Onishi, Seiki*; Ochiai, Kentaro; Sato, Satoshi
Progress in Nuclear Science and Technology (Internet), 2, p.346 - 357, 2011/10
The major revised version of Japanese Evaluated Nuclear Data Library (JENDL), JENDL-4, was released in 2010 spring. As the benchmark test of JENDL-4.0 in the shielding and fusion neutronics fields, we analyzed many integral benchmark experiments (in-situ and Time-of-Flight (TOF) experiments) with DT neutrons at JAEA/FNS with the MCNP code and JENDL-4.0. The experiments with assemblies including beryllium, carbon, silicon, vanadium, copper, tungsten and lead, nuclear data of which were revised in JENDL-4.0, were selected for this benchmark test. As a result, it is found that JENDL-4 improved some problems pointed out in JENDL-3.3 and that it is comparable to ENDF/B-VII.0 and JEFF-3.1.
Onishi, Seiki; Kondo, Keitaro; Sato, Satoshi; Ochiai, Kentaro; Takakura, Kosuke; Konno, Chikara; Murata, Isao*
Journal of the Korean Physical Society, 59(2), p.1949 - 1952, 2011/08
So far we carried out many integral benchmark experiments (in-situ experiments and Time-Of-Flight experiments) for nuclear data with DT neutrons at the Fusion Neutronics Source facility in Japan Atomic Energy Agency. In addition to those, we have a plan to perform new integral benchmark experiments for nuclear data with a DT neutron beam, which can investigate nuclear data for almost the whole angle and the whole energy. Because the large-size tritium target in FNS is difficult to procure, we have started to build a new DT neutron beam with the small tritium target, which is easy to obtain, at the first target room of FNS. We already designed a collimator system for the DT neutron beam based on calculations. In this work, under that design, the collimator was constructed. Then the characteristics of the neutron field were measured in order to confirm the DT neutron beam performance. It was demonstrated that the DT neutron beam was realized as calculated.
Ochiai, Kentaro; Kondo, Keitaro; Onishi, Seiki; Takakura, Kosuke; Sato, Satoshi; Abe, Yuichi; Konno, Chikara; Suzuki, Chihiro*; Yagi, Takahiro*
Journal of the Korean Physical Society, 59(2), p.1953 - 1956, 2011/08
Lead is an important candidate material as multiplier of nuclear fusion reactor. Few DT neutron integral benchmark experiments were performed for lead so far. Therefore, we have carried out an integral benchmark experiment on lead at the DT neutron source facility of JAEA, FNS. A cubic lead assembly on a side of 45.3 cm was set up and was irradiated with the DT neutron source. Reaction rates of the Al(n,)Na, Nb(n,2n)Nb, Zr(n,2n)Zr and In(n,n')In reactions were measured as fast neutron spectrum indices in the assembly. A small NE213 spectrometer was also used for measurement of neutron spectra in the assembly. A Monte Carlo calculation code, MCNP5, was adopted to calculate the above neutron spectra and activation reaction rates. Nuclear data libraries, JENDL-3.3, ENDF/B-VII.0, JEFF-3.1 and FENDL-2.1, were used in the calculation. The calculation results of the three libraries except for JENDL-3.3 agreed with the measuring ones. In case of JENDL-3.3, some remarkable disagreements were found. From our investigations, it was pointed out that the inappropriate evaluation of the (n,2n) and inelastic cross sections of lead in JENDL-3.3 caused such disagreement.
Konno, Chikara; Ochiai, Kentaro; Onishi, Seiki
Journal of the Korean Physical Society, 59(2), p.1092 - 1095, 2011/08
JSSTDL-300 is a multigroup library for shielding applications produced from JENDL-3.2. The self-shielding correction in JSSTDL-300 is probably inadequate due to the following two reasons, (1) the weighting function of Legendre order = 0 is applied for all Legendre orders, (2) the f-table of the scattering matrix is the same as that of the elastic scattering. Thus we examined the effects of these problems through a simple benchmark test, the model of which consisted of an aluminum, iron, nickel or copper sphere of 1 m in radius with a 20 MeV neutron source in the center. Neutron spectra in the sphere were calculated with ANISN and were compared with those obtained with MCNP4C. It was found out that the effects were different dependent to materials and were the largest for copper. Adequate f-table and weighting function should be adopted in generation of multigroup libraries.
Sukegawa, Atsuhiko; Anayama, Yoshimasa*; Onishi, Seiki; Sakurai, Shinji; Kaminaga, Atsushi; Okuno, Koichi*
Journal of Nuclear Science and Technology, 48(4), p.585 - 590, 2011/04
Soft-type neutron shielding resin has been developed by improving an existing hard-type neutron shielding material by the epoxy-based resin as the additional shielding material. A flexible heat resistant neutron shielding material has been developed, which consists of newly polymer-based resin with boron. The neutron shielding performance of the developed flexible heat resistant resin by the Cf neutron source is almost the same as that of the polyethylene. The outgas of H, H, NH, HO, CO, O, CH and CO from the developed resin have been measured at high temperature environment (250C) by thermal desorption spectroscopy methods. The soft-type resin and the newly developed heat resistant resin will be applied to prevent the effects of the neutron streaming and to control the movement of vibrated pipe as the seal material around the plumbing in the future fast reactor and the innovative fission reactor.
Kondo, Keitaro; Ochiai, Kentaro; Tatebe, Yosuke; Yagi, Takahiro; Onishi, Seiki; Takakura, Kosuke; Sato, Satoshi; Konno, Chikara
Progress in Nuclear Science and Technology (Internet), 1, p.61 - 64, 2011/02
At the Fusion Neutronics Source (FNS) facility of JAEA we have conducted various integral experiments with DT neutrons for fusion reactor materials and have made a significant progress in the verification of their nuclear data. Recently we started a new series of integral experiments with DD neutrons at FNS in order to verify nuclear data relating to DD neutrons effectively. An integral experiment on beryllium with the DD neutron source will be presented in this conference. A beryllium pseudo-cylinder assembly of 45 cm in thickness and 63 cm in the diameter was built at the distance of 20 cm from the DD neutron source, and reaction rates of the In(n,n')In, Au(n,)Au and Li(n,)T reactions and a U fission rate were measured. The measured values were compared with calculations with the MCNP5 code and the latest nuclear data libraries; JENDL-3.3, ENDF/B-VII.0 and JEFF-3.1. A slight disagreement between the measurement and the calculation was found in the reaction rate of In, which is sensitive to neutrons above 0.3 MeV. We identified that the disagreement originated from the angular differential cross section data of the elastic scattering around 3 MeV and from the (n,2n) reaction cross section data near its threshold energy. The calculated reaction rates of Au, Li and U, which are sensitive to low energy neutrons, showed a large overestimation, which also appeared at the beryllium integral experiment with DT neutrons previously carried out at FNS. This problem has not been solved yet.
Ochiai, Kentaro; Tatebe, Yosuke; Kondo, Keitaro; Onishi, Seiki; Sato, Satoshi; Takakura, Kosuke; Konno, Chikara
Progress in Nuclear Science and Technology (Internet), 1, p.142 - 145, 2011/02
Nuclear performances of the ITER Test Blanket Module (TBM) can be calculated with a neutron transport code and nuclear data library. Neutron flux spectra in the TBM should be measured in order to validate the calculated nuclear performances of the TBM. The multi-foil activation method (MFAM) is considered to be one of the most prospective candidates for the neutron flux spectrum measurement. We have examined to measure neutron flux spectra in TBM simulating assemblies with a DT neutron source by using MFAM. We deduced neutron flux spectra in the simulated assemblies with a Monte Carlo code MCNP4C, some nuclear data and unfolding code NEUPAC. The results indicated that the adjusted neutron flux was reasonable for fast neutrons and that measured reaction rate data of more (n,) reactions were necessary for more adequate adjustment for slow neutrons.
Onishi, Seiki; Sato, Satoshi; Ochiai, Kentaro; Takakura, Kosuke; Kondo, Keitaro; Konno, Chikara
Progress in Nuclear Science and Technology (Internet), 1, p.73 - 76, 2011/02
There are two target rooms at the Fusion Neutronics Source (FNS) facility in Japan Atomic Energy Agency (JAEA). Recently it becomes increasingly difficult to meet needs for experiments with a DT neutron beam such as instrument development for ITER, because of shortage of the large target. Therefore we plan to construct a neutron beam inside TR1 by using the small tritium target of TR1, which is easier to obtain. We designed the collimator system based on the cylindrical assembly used in the previous ITER shielding experiments at JAEA/FNS. Neutron spectra at the exit of the collimator and at the offset position by 20 cm from the collimator axis were calculated with the two dimensional Sn code DORT and FENDL/MG-2.1 multi-group library in order to investigate the effect of the collimator system.
Konno, Chikara; Ochiai, Kentaro; Onishi, Seiki
Progress in Nuclear Science and Technology (Internet), 1, p.32 - 35, 2011/02
We carried out a simple benchmark calculation test with the multigroup cross-section library VITAMIN-B6 generated from ENDF/B-VI. The model of this test consisted of an iron sphere of 1 m in radius with an isotropic 20 MeV neutron source in the center. Neutron spectra in the sphere were calculated with the Sn code ANISN and VITAMIN-B6 or FENDL/MG-1.1. A calculation with MCNP and FENDL/MC-1.1 was carried out as a reference. The neutron spectra with ANISN and FENDL/MG-1.1 agreed with those with MCNP, while those with ANISN and VITAMIN-B6 were by at most 50% larger than those with MCNP. We examined VITAMIN-B6 in detail in order to investigate causes for the discrepancy. As the result, it was founded out that reasons of the discrepancy were the followings; (1) The smallest background cross section of Fe in VITAMIN-B6 is 1. (2) The weighting function used in generating VITAMIN-B6 is not adequate. VITAMIN-B6 should be revised for adequate self-shielding correction.
Kondo, Keitaro; Tatebe, Yosuke; Ochiai, Kentaro; Sato, Satoshi; Takakura, Kosuke; Onishi, Seiki; Konno, Chikara
Fusion Engineering and Design, 85(7-9), p.1229 - 1233, 2010/12
In the previous blanket neutronics experiments conducted at the FNS facility of Japan Atomic Energy Agency, the following disagreements between experiments and analyses have been pointed out: (1) In the experiment with a Li-enriched LiTiO layer and a beryllium layer, approximately 10% overestimation was found for the tritium production rate (TPR) when a neutron reflector composed of SS316 was attached. (2) In the experiment with natural LiO pebbles sandwiched by beryllium layers, TPR was overestimated near the rear beryllium layer by up to 10%. In order to confirm the above problems clearly, a new blanket neutronics experiment using a natural LiTiO layer and beryllium layers with DT neutrons was conducted at FNS. TPR distributions inside the LiTiO layer were measured with LiCO pellets with and without the source reflector. The measured TPR well agreed with the calculation within an estimated experimental error of 6% in the both experiments. The influence of the reflector was not remarkable in the present experiment. Contrary to our expectation, no remarkable difference was observed in the TPR distribution around the rear beryllium layer.
Konno, Chikara; Ochiai, Kentaro; Takakura, Kosuke; Onishi, Seiki; Kondo, Keitaro; Wada, Masayuki*; Sato, Satoshi
Fusion Engineering and Design, 85(10-12), p.2054 - 2058, 2010/12
In the last ISFNT, we presented re-analyses of fusion neutronics benchmark experiments on beryllium at JAEA/FNS and reported that all the calculations with JENDL-3.3, FENDL-2.1, JEFF-3.1 and ENDF/B-VII.0 overestimated experimental data on low energy neutrons and that the calculation with JEFF-3.1 had a strange peak around 12 MeV. Here we investigate reasons for these problems. As a result, It was found out that the official ACE file MCJEFF3.1 of JEFF-3.1 had an inconsistency with the original JEFF-3.1, which caused the strange larger neutron peak around 12 MeV. We also find out that the calculated thermal neutron peak is probably too large. It is indicated that the coherent elastic scattering cross section data in the thermal neutron flux law data of beryllium metal are too large.
Onishi, Seiki; Maebara, Sunao; Sakaki, Hironao; Sato, Satoshi; Ochiai, Kentaro; Konno, Chikara
Journal of Plasma and Fusion Research SERIES, Vol.9, p.190 - 192, 2010/08
A new deuteron accelerator is planned to be build at Rokkasho-site in IFMIF/EVEDA and its shielding design is required urgently. Therefore, a shielding analysis has done with the prototype model of IFMIF/EVEDA accelerator vault by the Monte-Carlo transport calculation code, MCNP5, and cross section library, FENDL/MC-2.1. The neutron dose rates become 0.5 Sv/h at the side of the beam dump and 0.05 Sv/h at the center of the beam axis. Those are smaller than the limitation dose rate of the regularly accessible controlled area, 25 Sv/h.
Konno, Chikara; Sato, Satoshi; Ochiai, Kentaro; Wada, Masayuki*; Onishi, Seiki; Takakura, Kosuke; Iida, Hiromasa
Nuclear Technology, 168(3), p.743 - 746, 2009/12
The three-dimensional Sn code Attila of Transpire, Inc. can use CAD data as a geometrical input directly and deal with assemblies of complicated geometry without much effort. ITER organization has a plan to adopt this code as one of the standard codes for nuclear analyses. However validation of calculations with this code is not carried out in detail so far. Thus we validate this code through analyses of some bulk experiments and streaming experiments with DT neutrons at JAEA/FNS. Analyses with the Sn code system DOORS and Monte Carlo code MCNP4C were also carried out for comparison. Agreement between Attila and DOORS calculations is very good for the bulk experiments. For streaming experiments Attila requires special treatments (biased angular quadrature sets or last collided source calculation) as well as DOORS in order to obtain similar results as those with MCNP, though Attila consumes much more time and memory than DOORS.
Sato, Satoshi; Takakura, Kosuke; Ochiai, Kentaro; Kondo, Keitaro; Tatebe, Yosuke; Onishi, Seiki; Wada, Masayuki*; Kutsukake, Chuzo; Tanaka, Shigeru; Abe, Yuichi; et al.
Fusion Science and Technology, 56(1), p.227 - 231, 2009/07
Previously DT neutronics experiments were performed by using partial blanket mockups for Japanese ITER test blanket module at JAEA FNS, and tritium production rates (TPR) inside blanket mockups were measured in details. The calculation with the nuclear data library FENDL-2.1 and Monte Carlo code MCNP4C agreed well with most of the measured TPRs within uncertainty of 10%. On the other hand, overestimations were found for the TPR in the experiment with a reflector and the TPR around the boundary between the rear part of the breeder layer and the beryllium layer by more than 10%. In order to confirm this concern, we measured reaction rate distribution in the partial blanket mockups with DT neutrons with two solid breeder blanket partial mockups, (Be/LiTiO/Be, SS316/LiTiO/SS316). Experiments were performed with and without a neutron source reflector. In order to measure reaction rate distributions, the activation foil method was applied using Nb and Au foils in this study. Experimental analyses were performed by MCNP4C with FENDL-2.1. Calculation results to experimental ones (C/Es) on the Au reaction rate with a reflector were larger than those without one. Detailed results are presented in this conference.
Onishi, Seiki; Maebara, Sunao; Iida, Hiromasa; Yamauchi, Michinori*; Sakaki, Hironao; Sato, Satoshi; Ochiai, Kentaro; Konno, Chikara
no journal, ,
no abstracts in English