Validation of CENDL and JEFF evaluated nuclear data files for TRIGA calculations through the analysis of integral parameters of TRX and BAPL benchmark lattices of thermal reactors

The aim of this paper is to present the validation of evaluated nuclear data files CENDL-2.2 and JEFF-3.1.1 through the analysis of the integral parameters of TRX and BAPL benchmark lattices of thermal reactors for neutronics analysis of TRIGA Mark-II Research Reactor at AERE, Bangladesh. In this process, the 69-group cross-section library for lattice code WIMS was generated using the basic evaluated nuclear data files CENDL-2.2 and JEFF-3.1.1 with the help of nuclear data processing code NJOY99.0. Integral measurements on the thermal reactor lattices TRX-1, TRX-2, BAPL-UO₂-1, BAPL-UO₂-2 and BAPL-UO₂-3 served as standard benchmarks for testing nuclear data files and have also been selected for this analysis. The integral parameters of the said lattices were calculated using the lattice transport code WIMSD-5B based on the generated 69-group cross-section library. The calculated integral parameters were compared to the measured values as well as the results of Monte Carlo Code MCNP. It was found that in most cases, the values of integral parameters show a good agreement with the experiment and MCNP results. Besides, the group constants in WIMS format for the isotopes U-235 and U-238 between two data files have been compared using WIMS library utility code WILLIE and it was found that the group constants are identical with very insignificant difference. Therefore, this analysis reflects the validation of evaluated nuclear data files CENDL-2.2 and JEFF-3.1.1 through benchmarking the integral parameters of TRX and BAPL lattices and can also be essential to implement further neutronic analysis of TRIGA Mark-II research reactor at AERE, Dhaka, Bangladesh.     

Qualification of the neutronic efficiency of erbium at zero burnup

The main asset of erbium as an alternative burnable poison to gadolinium is that it has a much lower thermal absorption efficient cross section that contributes to giving it much slower consumption kinetics than gadolinium, and also helps to generate much lower perturbation in the power distribution. The calculations performed with the APOLLO code and its associated library must be qualified and validated with an experiment in order to obtain a sufficient degree of confidence to envisage an industrial application of this poison. For this purpose the MIRTE UOX and MOX experiments were performed in the critical reactor EOLE at Cadarache within the framework of the EROÏNE programme. These experiments concern the neutronic assessment of a (U,Er)O₂ rod in a representative core of a pressurised water reactor lattice with an enhanced moderation ratio. The purpose of this paper is to show that the APOLLO2 code associated with its APOLLIB CEA 93 library is perfectly qualified at time zero to calculate erbium reactivity worth.     

Monte Carlo Modelling of Increasing Void Fraction in 100% MOX ABWR: Lessons Drawn from the FUBILA Program

The French Atomic Energy Commission CEA and the Japanese Incorporated Administrative Agency JNES (Japan Nuclear Energy Safety Organisation) have undertaken first-of-a-kind full MOX core physics experiments, FUBILA, in the EOLE critical facility of the CEA Cadarache Centre. The experiments have been designed to obtain core physics data under high-burn-up 9 × 9 and 10 × 10 BWR MOX assemblies operating conditions. The experimental program, consisting of eight different core configurations, started in January 2005 and ended on September 1, 2006. The analysis of the void increase part of the experimental data between 0 and 70% void has been carried out using the French TRIPOLI-4.5 continuous-energy Monte Carlo calculation code with the newly released JEFF3.1.1 nuclear data library. The average C/E discrepancies obtained on critical masses, reactivity worth, and pin-by-pin power distributions enable us to estimate all the integral and local parameters with uncertainties largely within the target uncertainties, demonstrating the capability of the code to treat complex geometries with a high degree of accuracy. Additional k_eff calculations performed with the latest ENDF/B-VII evaluation exhibit a clear tendency to overestimate the k_eff by about 500 to 650 pcm and the void worth by more than 4%, showing that the JEFF3.1.1 library is more precise for MOX lattices.     

Analysis of PHWR lattice experiments with organic coolant by the computer code club

An efficient computer code club, based on the combination of a small-scale collision probability and a large-scale interface current method, was developed for the analysis of pressurized heavy-water reactor (PHWR) lattice cells. A large number of experiments with different fuel clusters and D₂O and air coolants were analysed using this code. The results were found to be very encouraging. However, when club was used for analysing experiments with organic coolants, the results were found not to be in good agreement with the experiments. This paper discusses the reasons for this and proposes a remedy. Finally, it gives the results of the analysis of these experiments with the modified computer code club.     

Computational analysis of neutronic parameters for TRIGA Mark-II research reactor using evaluated nuclear data libraries

The aim of this study is to analyze the neutronic parameters of TRIGA Mark-II research reactor using the chain of NJOY-WIMS-CITATION computer codes based on evaluated nuclear data libraries CENDL-2.2 and JEFF-3.1.1. The nuclear data processing code NJOY99.0 has been employed to generate the 69 group WIMS library for the isotopes of TRIGA core. The cell code WIMSD-5B was used to generate the cross sections in CITATION format and then 3-dimensional diffusion code CITTATION was used to calculate the neutronic parameters of the TRIGA Mark-II research reactor. All the analyses were performed using the 7-group macroscopic cross section library. The CITATION test-runs using different cross section sets based on different models applied in WIMS calculations have shown a strong influence of those models on the final integral parameters. Some of the cells were specially treated with PRIZE options available in WIMSD-5B to take into account the fine structure of the flux gradient in the fuel-reflector interface region. It was observed that two basic parameters, the effective multiplication factor, k_eff and the thermal neutron flux, were in good agreement among the calculated results with each other as well as the measured values. The maximum power densities at the hot spot were 1.0446E02 W/cc and 1.0426E02 W/cc for the libraries CENDL-2.2 and JEFF-3.1.1 respectively. The calculated total peaking factors 5.793 and 5.745 were compared to the original SAR value of 5.6325 as well as MCNP result. Consequently, this analysis will be helpful to enhance the neutronic calculations and also be used for the further thermal–hydraulics study of the TRIGA core.     

Measurement of Local Power Peaking Factors in Heavy-Water Moderated Plutonium Lattices

Local power peaking factors (LPF) in the heavy-water moderated plutonium lattices were measured by a new method of γ-scanning of fuel pins using the calculated power correction factor of which accuracy was evaluated with the aid of foil activation method. Accuracy in measurement of the LPF was evaluated to be within 1%.

By this measurement, behaviors of the LPF have been made clear concerning the differences in fuel materials, coolant materials and arrangement in fuel enrichments. Depression of the thermal power in the fuel cluster makes LPF in the plutonium fuel lattice larger than in the uranium lattice. This tendency is more remarkable in air coolant lattice than in H₂O coolant lattice. The value of LPF for the plutonium fuel cluster of different enrichments is smaller than that of a uniformly enriched fuel cluster. The reduction of LPF is smaller in H₂O coolant than in air coolant lattice.

The values of LPF by WIMS-D code based on the transport theory and by METHUSELAH-II code based on the diffusion theory are in agreement with the measured ones, within 1.5 and 2.4% respectively.     

An Improvement of the Transverse Leakage Treatment for the Nodal SN Transport Calculation Method in Hexagonal-Z Geometry

To overcome the divergent behavior of the NSHEX code, a nodal S_N code for hexagonal geometry, for some transport calculations, an improvement has been made in the calculation of the axial leakage. The axial leakage, previously calculated by using the quadratic transverse leakage approximation (QLA), is calculated by a new method of analytically treating the spatial distribution within a node, based on the axial homogeneity of the ordinary core. The verification tests were performed for the KNK-II model geometry of the NEACRP 3-D Neutron Transport Benchmarks and the large assembly-size KNK-II model. It is found that k_ett values obtained by introducing the new method agree with the reference Monte Carlo calculation results within 0.1% Δk/k for the KNK-II model, although the QLA method did not converge for two cases. Furthemore the new method succeeded in calculations for the large assembly-size model, in which the QLA method failed for all cases. Thus the new method has been found accurate and convergence achieved for the cases in which the QLA method failed.     

Comparison between triangular and hexagonal modeling of a hexagonal-structured reactor core using box method

A hexagonal-structured reactor core (e.g. VVER-type) is mostly modeled by structured triangular and hexagonal mesh zones. Although both the triangular and hexagonal models give good approximations over the neutronic calculation of the core, there are some differences between them that seem necessary to be clarified. For this purpose, the neutronic calculations of a hexagonal-structured reactor core have to be performed using the structured triangular and hexagonal meshes based on box method of discretisation and then the results of two models should be benchmarked in different cases.In this paper, the box method of discretisation is derived for triangular and hexagonal meshes. Then, two 2-D 2-group static simulators for triangular and hexagonal geometries (called TRIDIF-2 and HEXDIF-2, respectively) are developed using the box method. The results are benchmarked against the well-known CITATION computer code in case of a VVER-1000 reactor core. Furthermore, the relative powers calculated by the TRIDIF-2 and HEXDIF-2 along with the ones obtained by the CITATION code are compared with the verified results which have been presented in the Final Safety Analysis Report (FSAR) of the aforementioned reactor.Different benchmark cases revealed the reliability of the box method in contrast with the CITATION code. Furthermore, it is shown that the triangular modeling of the core is more acceptable compared with the hexagonal one.     

Qualification of the APOLLO2.8 code package for the calculation of the fuel inventory and reactivity loss of UOx spent fuels in BWRs

This paper describes the neutronics qualification of the APOLLO2.8 code package for the calculation of UO_x fuel inventory and reactivity loss with burn-up in Boiling Water Reactors (BWR).     

Fission Product Model for BWR Lattice Calculation Code

A benchmark calculation of full fission product was performed for thermal reactor application using an isotope transmutation code DCHAIN based on 185 nuclides with revised nuclear data library. The fission product model for BWR lattice calculation was studied and tested with the benchmark results, and a model containing 45 explicit nuclides and one pseudo nuclide was selected as a reasonably best model to predict the burn up reactivity with high precision for practically all types of fuel and reactor operating conditions. The evaluated thermal cross section and resonance integral for the pseudo nuclide are σ_2,200 = 2.6b and.RI = 10.6b, combined with the pseudo fission yield values of 1.3898, 1.3233, 1.3675 and 1.2773 for fissions from ²³⁵U, ²³⁸U, ²³⁹Pu and ²⁴¹Pu, respectively. The present results are believed as equally applicable to PWR lattice calculation.     

基于SCIENCE程序包的IFBA组件模型的可行性研究

表面涂有一薄层硼化锆的一体化燃料可燃吸收体(IFBA)被用作轻水堆UO2燃料组件的反应性控制。法国AREVA公司开发的SCIENCE程序包具有模拟IFBA组件的能力,但其模拟精度需经标定。本文利用APOLLO2-F程序建立IFBA组件模型和不含IFBA组件模型,研究了组件的无限增殖因数k∞及IFBA价值,并与西屋公司结果进行比较。分析了燃料和包壳温度的处理方法以及数据库的差异对结果的影响。利用硼化锆密度修正因子评估IFBA价值偏差对堆芯参数和功率分布等的影响。结果表明:SCIENCE计算的k∞及IFBA价值与西屋公司的结果符合较好,低燃耗区SCIENCE计算的价值偏小2%。装载8个104根IFBA棒组件的堆芯,组件相对功率最大偏差约为1%;硼浓度、功率峰因子FQ和焓升因子FΔH的变化均不到0.1%,可忽略。先导组件采用28根或更少的IFBA棒时,可直接采用SCIENCE程序进行计算。     

Monte Carlo simulation of ion implantation in crystalline SiC

We report first the results of a simulation study of ion implantation in crystalline 6H-SiC. Calculations were performed with a Monte Carlo code modified to account for the hexagonal lattice structure of the material. After an approximate determination of empirical parameters of electronic energy loss, performed by comparison of simulated profiles with experimental data as reported in the literature, a detailed study of the effects of beam-target orientation has been made for a few specific cases. Results have been compared with those of similar simulations made in cubic 3C---SiC, where the same model parameters were used, in order to emphasize differences due to the different crystallographic structure and surface orientation of the two phases. Conditions which originate deep channeling tails in the implanted profiles are identified, as well as conditions suitable to obtain the minimum width profile.     

Calculation of Effective Doses for External Neutrons

Monte Carlo calculations of the effective dose, on the basis of 1CRP Publication 60, were performed for external neutrons from thermal energy to 18.3 MeV for five irradiation geometries: AP, PA, RLAT, ROT and ISO. A unisex anthropomorphic phantom and the MORSE-CG code were used in conjunction with a nuclear data set based on the JENDL-3 library. The effective dose was found to be superior to the effective dose equivalent, the former quantity, for neutrons below about 1 MeV and inferior above this energy for all the geometries. The ambient dose equivalent based on the new Q-L relationship proposed in the Publication was found not necessarily to give a conservative estimate of the effective dose for the AP and PA geometries. The results obtained here were in good agreement with those calculated with a different computer code and a different nuclear data set.     

MCNPX Monte Carlo burnup simulations of the isotope correlation experiments in the NPP Obrigheim

This paper describes the simulation work of the Isotope Correlation Experiment (ICE) using the MCNPX Monte Carlo computer code package. The Monte Carlo simulation results are compared with the ICE-Experimental measurements for burnup up to 30 GWD/t. The comparison shows the good capabilities of the MCNPX computer code package for predicting the depletion of the uranium fuel and the buildup of the plutonium isotopes in a PWR thermal reactor. The Monte Carlo simulation results show also good agreements with the experimental data for calculating several long-lived and stable fission products. However, for the americium and curium actinides, it is difficult to judge the predication capabilities for these actinides due to the large uncertainties in the ICE-Experimental data. In the MCNPX numerical simulations, a pin cell model is utilized to simulate the fuel lattice of the nuclear power reactor. Temperature dependent libraries based on JEFF3.1 nuclear data files are utilized for the calculations. In addition, temperature dependent libraries based ENDF/B-VII nuclear data files are utilized and the obtained results are very close to the JEFF3.1 results, except for ∼10% differences in the prediction of the minor actinide isotopes buildup.     

Lattice Parameter Measurements on Cluster-Type Fuel for Advanced Thermal Reactor

Abstract

The behavior of neutrons in a highly heterogeneous unit cell consisting of D₂O moderator, H₂O coolant and a 28-pin fuel cluster contained in a pressure tube has been studied through lattice parameter measurements covering three different ²³⁵U enrichments, four coolant void fractions and two lattice pitches. A single-region core configuration was adopted, with which measurements were made to determine—in relation to coolant void fraction—the critical D₂O level, as well as various lattice parameters

A strong dependence on coolant void was observed for the critical level and the lattice parameters, in the case of the smaller 22.5 cm pitch lattice, due to the positive effect on core reactivity exerted by the slowing-down faculty of H₂O in the epithermal energy region. With the larger 25.0 cm pitch, however, no meaningful dependence on void fraction was shown by any of the measured values, and this was ascribed to a compensating negative effect due to enhanced thermal neutron self-shielding in the fuel region produced by the H₂O coolant.

The results of cell calculations obtained by means of the METHUSELAH-II code showed generally good agreement with experimentally determined data, for both critical D₂O levels and lattice parameters, in the case of coolant-filled lattices (0, 30 and 70% void fractions). For lattices devoid of coolant (100% void fraction), however, discrepancies in lattice parameters—particularly in p ²⁸—produced corresponding deviations in core reactivity amounting to 1% in excess of those incurred with other void fractions.     

Performance of advanced self-shielding models in DRAGON Version4 on analysis of a high conversion light water reactor lattice

A high conversion light water reactor lattice has been analysed using the code DRAGON Version4. This analysis was performed to test the performance of the advanced self-shielding models incorporated in DRAGON Version4. The self-shielding models are broadly classified into two groups – “equivalence in dilution” and “subgroup approach”. Under the “equivalence in dilution” approach we have analysed the generalized Stamm’ler model with and without Nordheim model and Riemann integration. These models have been analysed also using the Livolant–Jeanpierre normalization. Under the “subgroup approach”, we have analysed Statistical self-shielding model based on physical probability tables and Ribon extended self-shielding model based on mathematical probability tables. This analysis will help in understanding the performance of advanced self-shielding models for a lattice that is tight and has a large fraction of fissions happening in the resonance region. The nuclear data for the analysis was generated in-house. NJOY99.90 was used for generating libraries in DRAGLIB format for analysis using DRAGON and A Compact ENDF libraries for analysis using MCNP5. The evaluated datafiles were chosen based on the recommendations of the IAEA Co-ordinated Research Project on the WIMS Library Update Project. The reference solution for the problem was obtained using Monte Carlo code MCNP5. It was found that the Ribon extended self-shielding model based on mathematical probability tables using correlation model performed better than all other models.