Go Chiba

Development of a Fine and Ultra-Fine Group Cell Calculation Code SLAROM-UF for Fast Reactor Analyses

A cell calculation code SLAROM-UF has been developed for fast reactor analyses to produce effective cross sections with high accuracy in practical computing time, taking full advantage of fine and ultra-fine group calculation schemes. The fine group calculation covers the whole energy range in a maximum of 900-group structure. The structure is finer above 52.5 keV with a minimum lethargy width of 0.008. The ultra-fine group calculation solves the slowing down equation below 52.5 keV to treat resonance structures directly and precisely including resonance interference effects. Effective cross sections obtained in the two calculations are combined to produce effective cross sections over the entire energy range. Calculation accuracy and improvements from conventional 70-group cell calculation results were investigated through comparisons with reference values obtained with continuous energy Monte Carlo calculations. It was confirmed that SLAROM-UF reduces the difference in k-infinity from 0.15 to 0.01% for a JOYO MK-I fuel subassembly lattice cell calculation, and from −0.21 % to less than a statistical uncertainty of the reference calculation of 0.03% for a ZPPR-10A core criticality calculation.

JENDL-4.0 Benchmarking for Fission Reactor Applications

Benchmark testing for the newly developed Japanese evaluated nuclear data library JENDL-4.0 is carried out by using a huge amount of integral data. Benchmark calculations are performed with a continuous-energy Monte Carlo code and with the deterministic procedure, which has been developed for fast reactor analyses in Japan. Through the present benchmark testing using a wide range of benchmark data, significant improvement in the performance of JENDL-4.0 for fission reactor applications is clearly demonstrated in comparison with the former library JENDL-3.3. Much more accurate and reliable prediction for neutronic parameters for both thermal and fast reactors becomes possible by using the library JENDL-4.0.

JENDL Actinoid File 2008

JENDL Actinoid File 2008 (JENDL/AC-2008) was released in March 2008. It includes nuclear data for neutron-induced reactions for 79 nuclides from Ac (Z = 89) to Fm (Z = 100). The neutron energy range is 10−5 eV to 20 MeV. Almost alldata for 62 actinoids in JENDL-3.3 were revised. New evaluations were performed for 17 nuclides, which have half-lives longer than one day. A new comprehensive theoretical model code CCONE was widely used for the evaluation of cross sections and neutron emission spectra. Thermal cross sections for many nuclides were revised based on experimental data. Resonance parameters were readjusted to reproduce them. Simultaneous evaluations of fission cross sections were performed for six important nuclei. The least-squares fitting code GMA was used for the evaluation of fission cross sections for minor actinoids. In this paper, we present the evaluation methods and results of the JENDL/AC-2008.

Sensitivity Analysis of Fission Product Concentrations for Light Water Reactor Burned Fuel

The accurate prediction of fission product concentrations (FPCs) is necessary for application of the burnup credit to nuclear facilities. In order to specify important nuclear data for the accurate prediction of FPC, we extensively evaluate the sensitivities of FPC to nuclear data with the depletion perturbation theory. The target fission products are twelve important ones for the burnup credit, Mo-95, Tc-99, Rh-103, Nd-143, Nd-145, Sm-147, Sm-149, Sm-150, Sm-152, Cs-133, Eu-153, and Gd-155. The present study successfully specifies the important nuclear data both in a UO2 cell and in a MOX cell. While the obtained sensitivities are mostly similar to each other between the UO2 and MOX cells, large differences are observed in some cases, such as the Gd-155 concentration. It is clearly shown that such differences between the UO2 and MOX cells come from differences in cumulative fission yields between U-235 and Pu-239 and differences in neutron flux energy spectra.

Evaluation of Neutron Nuclear Data on Arsenic-75 for JENDL-4

Neutron nuclear data on 75As have been evaluated for the evaluated nuclear data library the energy region from 10−5 eV to 20 MeV. The thermal capture cross section was updated by considering recent measurements. The statistical model was applied to calculate the cross sections above the resolved resonance region. In the calculation, coupled-channel optical model parameters were used for neutrons. Pre-equilibrium and direct-reaction processes were taken into account in addition to the compound process. The present calculations are almost consistent with available experimental data. The measured leakage neutron spectrum is well reproduced by the presentlyevaluated data at 14 MeV.

Sodium Void Reactivity Worth Calculations for Fast Critical Assemblies without Whole-Lattice Homogenization

In the present paper, we calculate the sodium void reactivity worth of fast critical assemblies without whole-lattice homogenization in order to reduce errors associated with lattice homogenization. Firstly, we solve a neutron transport benchmark problem simulating fast critical assemblies composed of thin material plates with a discrete ordinates transport solver. The discrete ordinates transport solutions agree well with the Monte Carlo reference solutions; hence, we confirm the validity of the deterministic transport calculations for the sodium void reactivity worth of lattice-heterogeneous critical assemblies. Thereafter, the existing experimental data are calculated without whole-lattice homogenization. The result suggests that the lattice homogenization results in the overestimation of the leakage component of sodium void reactivity worth when the leakage component parallel to plate boundaries is dominant. Utilizing the numerical method without whole-lattice homogenization and the nuclear data JENDL-3.3, numerical solutions agree with the experimental data within 3σ of the experimental uncertainties.

Overestimation in Parallel Component of Neutron Leakage Observed in Sodium Void Reactivity Worth Calculation for Fast Critical Assemblies

Overestimation in parallel components of neutron leakage is observed in sodium void reactivity worth calculations for fast critical assemblies. The sodium void reactivity worth (SVRW) is one of the key parameters in sodium-cooled fast reactors, and it is difficult to be predicted accurately by numerical calculations because of its complicated mechanism. By virtue of an index which tells magnitude of cancellation between the nonleakage and leakage components of SVRW, it becomes possible to estimate prediction accuracy for each component separately. In the present study, it is found that prediction accuracy for the non-leakage component is good. On the other hand, overestimation in the leakage component is observed in results of experimental analyses for several data. A comparison in various critical configurations indicates that the overestimation in the leakage component is brought by overestimation in a ‘parallel’ component of neutron leakage. Through a numerical study, it is shown that lattice homogenizations cause the overestimation.

Covariance Analyses of Self-Shielding Factor and Its Temperature Gradient for Uranium-238 Neutron Capture Reaction

Covariances of the self-shielding factor and its temperature gradient for the uranium-238 neutron capture reaction have been evaluated from the resonance parameter covariance matrix and the sensitivity of the self-shielding factor and its temperature gradient to the resonance parameters. The resonance parameters and their covariance matrix for uranium-238 were taken from JENDL-3.3, while the sensitivity coefficients were calculated by varying resonance parameters and temperature. A set of computer code modules has been developed for the calculation of the sensitivity coefficients at numerous resonance levels. The present result shows that the correlation among resonance parameters yields a substantial contribution to the standard deviations of the self-shielding factor and its temperature gradient. In addition to the standard deviations of these quantities, their correlation matrices in the JFS-3 70 group structure are also obtained.

Verification of Homogenization in Fast Critical Assembly Analyses

In the present paper, homogenization procedures for fast critical assembly analyses are investigated. Errors caused by homogenizations are evaluated by the exact perturbation theory. In order to obtain reference solutions, three-dimensional plate-wise transport calculations are performed. It is found that the angular neutron flux along plate boundaries has a significant peak in the fission source energy range. To treat this angular dependence accurately, the double-Gaussian Chebyshev angular quadrature set with S24 is applied. It is shown that the difference between the heterogeneous leakage theory and the homogeneous theory is negligible, and that transport cross sections homogenized with neutron flux significantly underestimate neutron leakage. The error in criticality caused by a homogenization is estimated at about 0.1%Δ/k/kk in a small fast critical assembly. In addition, the neutron leakage is overestimated by both leakage theories when sodium plates in fuel lattices are voided.

On Effective Delayed Neutron Fraction Calculations with Iterated Fission Probability

The iterated fission probability (IFP) is a quantity proportional to the asymptotic power level originated by a neutron introduced to a reactor. The effective delayed neutron fraction βeff can be accurately calculated by the continuous-energy Monte Carlo method using IFP if a sufficiently large number of generations is considered to obtain the asymptotic state. In order to deterministically quantify the required number of generations in the IFP-based βeff calculations, the concept of the generation-dependent importance functions is introduced to βeff calculations. Furthermore, the most appropriate reactor property used in the IFP calculations, which reduces the required number of generations, is theoretically derived. Through numerical calculations, it is shown that several generations are required in the IFP-based βeff calculations and that the use of the appropriate reactor property can reduce the required number of generations. An efficient procedure for the IFP-based βeff calculations by the Monte Carlo method is also proposed.