Evaluated Nuclear Data Covariances: The Journey From ENDF/B-VII.0 to ENDF/B-VII.1

Recent interest from data users on applications that utilize the uncertainties of evaluated nuclear reaction data has stimulated the data evaluation community to focus on producing covariance data to a far greater extent than ever before. Although some uncertainty information has been available in the ENDF/B libraries since the 1970?s, this content has been fairly limited in scope, the quality quite variable, and the use of covariance data confined to only a few application areas. Today, covariance data are more widely and extensively utilized than ever before in neutron dosimetry, in advanced fission reactor design studies, in nuclear criticality safety assessments, in national security applications, and even in certain fusion energy applications. The main problem that now faces the ENDF/B evaluator community is that of providing covariances that are adequate both in quantity and quality to meet the requirements of contemporary nuclear data users in a timely manner. In broad terms, the approach pursued during the past several years has been to purge any legacy covariance information contained in ENDF/B-VI.8 that was judged to be subpar, to include in ENDF/B-VII.0 (released in 2006) only those covariance data deemed then to be of reasonable quality for contemporary applications, and to subsequently devote as much effort as the available time and resources allowed to producing additional covariance data of suitable scope and quality for inclusion in ENDF/B-VII.1. Considerable attention has also been devoted during the five years since the release of ENDF/B-VII.0 to examining and improving the methods used to produce covariance data from thermal energies up to the highest energies addressed in the ENDF/B library, to processing these data in a robust fashion so that they can be utilized readily in contemporary nuclear applications, and to developing convenient covariance data visualization capabilities. Other papers included in this issue discuss in considerable detail various aspects of the data producer community?s efforts to improve the evaluation methods and to add covariance content to the ENDF/B library. The present paper offers just a brief glimpse of these activities by drawing material from covariance papers presented at meetings, workshops and international conferences during the past five years. Highlighted are: advances in methods for producing and processing covariance data, recently developed covariance visualization capabilities, and the development and implementation of quality assurance (QA) requirements that should be satisfied for covariance data to be included in ENDF/B-VII.1.     

VVER-1000 neutronics calculation with ENDF/B-VII data

The initial release of the ENDF/B-VII nuclear data library is verified for VVER-1000 reactors. For neutronics calculation, the MCNP code based on the Monte-Carlo method is applied. Continuous-energy cross-sections for use with MCNP are calculated with the NJOY code. Isotopics for burned fuel is calculated with the WIMSD code. Calculated criticality, pin-to-pin power distribution, time-dependent critical concentration of soluble boron, worth of the control rods, average fuel assembly powers and time-dependent axial power distribution are compared to the corresponding experimental values for both zero-power VVER-1000 model, created at the LR-0 experimental facility, and the first fuel cycle of a real VVER-1000 reactor. For all of these parameters, neutronics calculation with ENDF/B-VII is in good agreement with the measurement. Moreover, for VVER-1000 neutronics calculation, ENDF/B-VII provides better results than ENDF/B-VI.     

Actinide ENDF/B-VII cross-section evaluations and validation testing

We describe advances made recently at Los Alamos for ENDF/B-VII, and for future ENDF releases, of actinide cross-section evaluations. Using americium as an illustrative example, we describe recent experiments that have largely confirmed the nuclear theory predictions that were the basis for the ENDF/B-VII.0 data for these americium reactions. The goal of this paper is to highlight some of the open issues in our understanding of the actinide nuclear data – especially for fast reactor applications – and show examples of how experiment, theory, and integral data validation has advanced our understanding. We will also describe the usage of these data in MCNP and S_N radiation transport simulations of various integral critical systems, for both criticality and for transmutation reaction rates.     

On the performance of ENDF/B-VII.0 data for fast neutron fluence calculations

The newest release of the Evaluated Nuclear Data File, ENDF/B-VII.0, has recently become available, along with the announcement of its significant advances over previous versions. The older ENDF/B releases, however, have been assessed for numerous applications in the past and, as such, currently find wide application in the nuclear industry. An analysis of the influence of the recent upgrade in the data file is thus needed with respect to specific practical applications. The present technical note addresses the effects of modifications of the neutron cross-sections in the ENDF/B-VII.0 library, relative to the preceding ENDF/B-VI.8 release, on calculations of the fast neutron fluence accumulated in a PWR reactor pressure vessel, which is of direct relevance for the licensing and safe operation of nuclear power plants.     

Fission Product Yields from Fission Spectrum n+Pu for ENDF/B-VII.1

We describe a new cumulated fission product yield (FPY) evaluation for fission spectrum neutrons on plutonium that updates the ENDF/B-VI evaluation by England and Rider, for the forthcoming ENDF/B-VII.1 database release.¹ We focus on FPs that are needed for high accuracy burnup assessments; that is, for inferring the number of fissions in a neutron environment. Los Alamos conducted an experiment in the 1970s in the Bigten fast critical assembly to determine fission product yields as part of the Interlaboratory Reaction Rate (ILRR) collaboration, and this has defined the Laboratory's fission standard to this day. Our evaluation includes use of the LANL-ILRR measurements (not previously available to evaluators) as well as other Laboratory FPY measurements published in the literature, especially the high-accuracy mass spectrometry data from Maeck and others. Because the measurement database for some of the FPs is small — especially for ⁹⁹Mo — we use a meta-analysis that incorporates insights from other accurately-measured benchmark FP data, using R-value ratio measurements. The meta-analysis supports the FP measurements from the LANL-ILRR experiment. Differences between our new evaluations and ENDF/B-VI are small for some FPs (less than 1-2%-relative for ⁹⁵Zr, ¹⁴⁰Ba, ¹⁴⁴Ce), but are larger for ⁹⁹Mo (4%-relative) and ¹⁴⁷Nd (5%-relative, at 1.5 MeV) respectively. We present evidence for an incident neutron energy dependence to the ¹⁴⁷Nd fission product yield that accounts for observed differences in the FPY at a few-hundred keV average energy in fast reactors versus measurements made at higher average neutron energies in Los Alamos' fast critical assemblies. Accounting for such FPY neutron energy dependencies is important if one wants to reach a goal of determining the number of fissions to accuracies of 1-2%. An evaluation of the energy-dependence of fission product yields is given for all A values based on systematical trends in the measured data, with a focus on the energy dependence over the fast neutron energy range from 0.2-2 MeV. Based on these trends, we present an evaluation of the FPY data at 0.5 and 2.0 MeV average incident neutron energies. This new set of ENDF/B-VII data will enable users to linearly interpolate between the pooled FPY data at ∼0.5 MeV and our new data at 2 MeV to obtain FPYs at other energies.     

ENDF/B-VII.1 Neutron Cross Section Data Testing with Critical Assembly Benchmarks and Reactor Experiments

The ENDF/B-VII.1 library is the latest revision to the United States? Evaluated Nuclear Data File (ENDF). The ENDF library is currently in its seventh generation, with ENDF/B-VII.0 being released in 2006. This revision expands upon that library, including the addition of new evaluated files (was 393 neutron files previously, now 423 including replacement of elemental vanadium and zinc evaluations with isotopic evaluations) and extension or updating of many existing neutron data files. Complete details are provided in the companion paper [M. B. Chadwick et al., “ENDF/B-VII.1 Nuclear Data for Science and Technology: Cross Sections, Covariances, Fission Product Yields and Decay Data,” Nuclear Data Sheets, 112, 2887 (2011)]. This paper focuses on how accurately application libraries may be expected to perform in criticality calculations with these data. Continuous energy cross section libraries, suitable for use with the MCNP Monte Carlo transport code, have been generated and applied to a suite of nearly one thousand critical benchmark assemblies defined in the International Criticality Safety Benchmark Evaluation Project?s International Handbook of Evaluated Criticality Safety Benchmark Experiments. This suite covers uranium and plutonium fuel systems in a variety of forms such as metallic, oxide or solution, and under a variety of spectral conditions, including unmoderated (i.e., bare), metal reflected and water or other light element reflected. Assembly eigenvalues that were accurately predicted with ENDF/B-VII.0 cross sections such as unmoderated and uranium reflected ²³⁵U and ²³⁹Pu assemblies, HEU solution systems and LEU oxide lattice systems that mimic commercial PWR configurations continue to be accurately calculated with ENDF/B-VII.1 cross sections, and deficiencies in predicted eigenvalues for assemblies containing selected materials, including titanium, manganese, cadmium and tungsten are greatly reduced. Improvements are also confirmed for selected actinide reaction rates such as ²³⁶U, ^238,242Pu and ^241,243Am capture in fast systems. Other deficiencies, such as the overprediction of Pu solution system critical eigenvalues and a decreasing trend in calculated eigenvalue for ²³³U fueled systems as a function of Above-Thermal Fission Fraction remain. The comprehensive nature of this critical benchmark suite and the generally accurate calculated eigenvalues obtained with ENDF/B-VII.1 neutron cross sections support the conclusion that this is the most accurate general purpose ENDF/B cross section library yet released to the technical community.     

基于ENDF/B-VII.0评价库的多群参数库MUSE1.0的开发与初步验证

采用NJOY程序研制了基于ENDF/B-VII.0评价库的172群中子-42群光子多群截面库(MUSE1.0),该库的权重谱采用Vitanim-e谱,角分布采用勒让德P6近似;热散射数据由自由气体模型产生,共振自屏修正选择了10组背景截面。该库含有293、600、800、900 K等温度下的截面数据;采用GENDF、MATXS和ACE多群3种格式存储。采用MCNP程序,从临界计算和屏蔽计算两个方面对该库进行较全面检验。结果表明,MUSE1.0在临界计算以及屏蔽计算方面具有较强的通用性,对于热散射效应以及共振自屏效应具有较好地描述能力,可以满足超临界水堆概念设计研究方面的应用要求。     

Validation of MVP-II with JENDL-4.0, ENDF/B-VII.1, and JEFF-3.2 through HTC critical experiments

In order to validate MVP-II, Haut Taux de Combustion (HTC) experiments were analyzed using a code with relatively new nuclear data libraries, JENDL-4.0, ENDF/B-VII.1, and JEFF-3.2. The effective neutron multiplication factor k_eff values were obtained through analyses of all phases of the HTC experiments. Consequently, the k_eff biases evaluated for each nuclear data library were within 300 pcm. Additionally, microscopic production and capture reaction rates of major actinide isotopes were analyzed to substantiate differences among the libraries for a representative case of Phase 1 of the HTC experiments. The analysis showed that microscopic cross sections of ²³⁸Pu and ²⁴¹Am in JEFF-3.2 were somewhat large compared to those of ENDF/B-VII.1 and JENDL-4.0 for the representative case of Phase 1.     

Impact of spread in minor actinide data from ENDF/B-VII.0, ENDF/B-VI.8, JENDL-3.3 and JEFF-3.0 on an IAEA-CRP FBR benchmark for MA incineration

Under an IAEA Coordinated Research Project (CRP), an FBR Model has been designed to study feasibility of incineration of long-lived minor actinides (MA). The predictions depend on the accuracy of the nuclear data used. There are several evaluations for actinide nuclear data, produced based on state-of-the-art procedures, but substantial deviations persist among them. The effect of spread in the MA nuclear data over a few recent evaluations, on the predicted material and Doppler worths of the FBR model, has been estimated and presented in this paper.     

Analysis of the SPERT-III E-Core experiment using CASMO5/TRACE/PARCS based on JENDL-4.0 and ENDF/B-VII.1

An analysis of the Special Power Excursion Test III E-Core experiment was performed in order to confirm the calculation accuracy of the light-water-reactor core analysis code system, constructed by the authors, of the CASMO5 and the TRACE/PARCS for the reactivity-initiated accident (RIA) analysis. The influence of the resonance up-scatter model (RUM) and the effective Doppler temperature model (EDTM) in the CASMO5 on the Doppler reactivity feedback effect was also discussed through the comparison with the conventional calculation without those models and the perturbation calculation of the Doppler reactivity coefficient. The calculation results by the CASMO5/TRACE/PARCS mostly showed good agreement with the experimental data within the range of experimental uncertainty, which confirmed the calculation accuracy of the analysis code system for the RIA analysis. In the calculation, the JENDL-4.0 and the ENDF/B-VII.1 were used, however, obvious difference was not seen in the calculation results between the two nuclear data libraries. The influence of the RUM on the core parameters was less than that of the 10% increment of Doppler reactivity coefficient on the conventional calculation. The influence of the EDTM became large on the cold condition, which was the same tendency on the Doppler reactivity coefficient discussed in the previous studies.     

ENDF/B-VII.1 Nuclear Data for Science and Technology: Cross Sections, Covariances, Fission Product Yields and Decay Data

The ENDF/B-VII.1 library is our latest recommended evaluated nuclear data file for use in nuclear science and technology applications, and incorporates advances made in the five years since the release of ENDF/B-VII.0. These advances focus on neutron cross sections, covariances, fission product yields and decay data, and represent work by the US Cross Section Evaluation Working Group (CSEWG) in nuclear data evaluation that utilizes developments in nuclear theory, modeling, simulation, and experiment.The principal advances in the new library are: (1) An increase in the breadth of neutron reaction cross section coverage, extending from 393 nuclides to 423 nuclides; (2) Covariance uncertainty data for 190 of the most important nuclides, as documented in companion papers in this edition; (3) R-matrix analyses of neutron reactions on light nuclei, including isotopes of He, Li, and Be; (4) Resonance parameter analyses at lower energies and statistical high energy reactions for isotopes of Cl, K, Ti, V, Mn, Cr, Ni, Zr and W; (5) Modifications to thermal neutron reactions on fission products (isotopes of Mo, Tc, Rh, Ag, Cs, Nd, Sm, Eu) and neutron absorber materials (Cd, Gd); (6) Improved minor actinide evaluations for isotopes of U, Np, Pu, and Am (we are not making changes to the major actinides ^235,238U and ²³⁹Pu at this point, except for delayed neutron data and covariances, and instead we intend to update them after a further period of research in experiment and theory), and our adoption of JENDL-4.0 evaluations for isotopes of Cm, Bk, Cf, Es, Fm, and some other minor actinides; (7) Fission energy release evaluations; (8) Fission product yield advances for fission-spectrum neutrons and 14 MeV neutrons incident on ²³⁹Pu; and (9) A new decay data sublibrary.Integral validation testing of the ENDF/B-VII.1 library is provided for a variety of quantities: For nuclear criticality, the VII.1 library maintains the generally-good performance seen for VII.0 for a wide range of MCNP simulations of criticality benchmarks, with improved performance coming from new structural material evaluations, especially for Ti, Mn, Cr, Zr and W. For Be we see some improvements although the fast assembly data appear to be mutually inconsistent. Actinide cross section updates are also assessed through comparisons of fission and capture reaction rate measurements in critical assemblies and fast reactors, and improvements are evident. Maxwellian-averaged capture cross sections at 30 keV are also provided for astrophysics applications.We describe the cross section evaluations that have been updated for ENDF/B-VII.1 and the measured data and calculations that motivated the changes, and therefore this paper augments the ENDF/B-VII.0 publication [M. B. Chadwick, P. Oblo?inský, M. Herman, N. M. Greene, R. D. McKnight, D. L. Smith, P. G. Young, R. E. MacFarlane, G. M. Hale, S. C. Frankle, A. C. Kahler, T. Kawano, R. C. Little, D. G. Madland, P. Moller, R. D. Mosteller, P. R. Page, P. Talou, H. Trellue, M. C. White, W. B. Wilson, R. Arcilla, C. L. Dunford, S. F. Mughabghab, B. Pritychenko, D. Rochman, A. A. Sonzogni, C. R. Lubitz, T. H. Trumbull, J. P. Weinman, D. A. Br, D. E. Cullen, D. P. Heinrichs, D. P. McNabb, H. Derrien, M. E. Dunn, N. M. Larson, L. C. Leal, A. D. Carlson, R. C. Block, J. B. Briggs, E. T. Cheng, H. C. Huria, M. L. Zerkle, K. S. Kozier, A. Courcelle, V. Pronyaev, and S. C. van der Marck, “ENDF/B-VII.0: Next Generation Evaluated Nuclear Data Library for Nuclear Science and Technology,” Nuclear Data Sheets 107, 2931 (2006)].     

Calculations of Maxwellian-averaged cross sections and astrophysical reaction rates using the ENDF/B-VII.0, JEFF-3.1, JENDL-3.3, and ENDF/B-VI.8 evaluated nuclear reaction data libraries

We have calculated the Maxwellian-averaged cross sections and astrophysical reaction rates of the stellar nucleosynthesis reactions (n, γ), (n, fission), (n, p), (n, α), and (n, 2n) using the ENDF/B-VII.0, JEFF-3.1, JENDL-3.3, and ENDF/B-VI.8 evaluated nuclear reaction data libraries. These four major nuclear reaction libraries were processed under the same conditions for Maxwellian temperatures (kT) ranging from 1 keV to 1 MeV. We compare our current calculations of the s-process nucleosynthesis nuclei with previous data sets and discuss the differences between them and the implications for nuclear astrophysics.     

Evaluating the JEFF 3.1, ENDF/B-VII.0, JENDL 3.3, and JENDL 4.0 nuclear data libraries for a small 100 MWe molten salt reactor with plutonium fuel

This study evaluated the nuclear data libraries for a small 100 Mega Watt electric(MWe) Molten Salt Reactor with plutonium fuel. The reactor has a power output of 100 MWe, which meets the demand for electricity generation in several regions or provinces outside Java Island. Several nuclear data libraries, such as JEFF 3.1, ENDF/B-VII.0, JENDL 3.3, and JENDL 4.0, were used for a more comprehensive evaluation. LiF–BeF2–ThF4–PuF4 was used as the initial fuel composition. The thorium and plutonium c...     

Validation of an MCNP4B whole-reactor model for LWR-PROTEUS using ENDF/B-V,ENDF/B-VI and JEF-2.2 cross-section libraries

A detailed three-dimensional, continuous-energy MCNP4B model of the LWR-PROTEUS critical facility has been developed for the analysis of whole-reactor characteristics using ENDF/B-V, ENDF/B-VI and JEF-2.2 cross-section sets. The model has been applied to the determination of the critical loading, as well as the evaluation of reactivity worths for safety/shutdown rods, control rods, and individual driver-region fuel rods. The initially obtained results for the first configuration investigated (Core 1B) indicated that, for the same geometrical and materials specifications, the ENDF/B-V data library yields the closest critical prediction (discrepancy of 640±40 pcm), followed by ENDF/B-VI (980±40 pcm) and JEF-2.2 (1340±40 pcm). The differences in results between the three data libraries were studied by considering the contributions of individual materials to the neutron balance. ²³⁵U and ²³⁸U cross-sections from JEF-2.2, for example, explain an effect of ∼400 pcm. Refinement of the materials specifications in the MCNP4B whole-reactor model, in particular the impurities assumed for the graphite driver of the driver and reflector regions, has been shown to reduce the final discrepancy of the ENDF/B-V based k_eff result to ∼0.2%. The MCNP4B results for relative reactivity effects, such as control rod worths, are found to agree within experimental errors with the measured values.     

Atomic displacement cross sections for carbon from ENDF/B-III data

Atomic displacement cross sections have been derived from the data in materials file 1165 (Carbon-12) of the third version of the Evaluated Nuclear Data File (ENDF/B-III), Using the Lindhard theoretical treatment for the stopping cross sections of carbon ions in carbon. A comparison of the differences between previous estimates of the carbon atom displacement functions and the results of these calculations indicate that the use of the atomic displacement cross sections presented herein should improve the correlations of damage production in different irradiation facilities.     

Information retrieval system for ENDF format libraries

This report presents a data information retrieval system for ENDF (Evaluated Nuclear Data File) format libraries, which can be run on PC computers under the Windows™ environment. The input is the filename of an ENDF. The system will process this file and generate two others. One contains a list of materials with the corresponding nuclides, laboratory, author and date of evaluation; the other provides information about the MF and MT numbers for each material, expressed by the number of records. This interactive and easy-to-handle system is of interest to nuclear and reactor physics researchers.     

Delayed neutron study using ENDF/B-VI basic nuclear data

The basic nuclear data of the latest releases of ENDF/B-VI were used in preliminary calculations with the CINDER'90 nuclide inventory code to simulate the activity of fission delayed-neutron precursors. Total delayed-neutron production was obtained at times during and following pulse (0.1-ms) and equilibrium (4-hr) fission histories for each of the sixty fission systems having fission-product yields in ENDF/B-VI. The equilibrium studies — at unit fission rate for constant fission periods sufficiently long that all precursors reached saturation inventories — yielded the value for each system. Delayed-neutron production rates at 54 decay times t, extending to 500 s following a fission pulse, were fit using the STEPIT code to the pulse function R(t) = ∑a_iλ_ie^−λ_it. Results following equilibrium irradiations were fit to the equilibrium function R(∞, t) = ∑a_iλ_ie^−λ_it. It was observed that functions from fits to pulse results did not well represent equilibrium results at long cooling times. Similarly, functions fit to the equilibrium results did not well represent pulse results at short cooling times.

A comprehensive series of CINDER'90 calculations was then made for irradiation times T of 0.1 ms, 1 s, 10 s, 100 s, 1000 s, and 4 hours; results were obtained at 60 decay times t extending to 800 s following irradiation. Comprehensive calculations were made using both the 1989 Pn data of England and Brady and the new Pn data of Pfeiffer, Kratz and Möller described elsewhere in this issue. The body of results for each system was included in fits to obtain the neutron production rate R(T, t) = ∑a_ie^−λ_it(1 − e^−λ_iT) for each system. Fits were made for the traditional sum of six exponentials, with all variables free to vary; additional fits were made for a sum of eight exponentials with decay constants set to values suggested by Piksaikin. The resulting pulse functions R(t), defined by the a_i and λ_i thus obtained, accurately represent calculated delayed-neutron production when integrated with any irradiation history.

The pulse functions thus produced and other published pulse functions fit to past measurements and calculations are compared numerically at several times after fission. Reactivity effects of all functions from measurements and calculations for each of the sixty systems are indicated by the asymptotic periods following positive 10¢– 50¢ reactivity steps simulated in point-reactor kinetics calculations using the AIREK-10 code.     

Thermal and Fast Reactor Benchmark Testing of ENDF/B-6.4

In 1995 and 1996, CNDC made homogeneous fast and thermal reactor benchmark testing of CENDL-2 and B-6.2(ENDF/B-6 version 2), respectively. It proved that ~(238)U data of CENDL-2 are better than those of B-6.2. For fast reactor benchmarks, B-6.2 shows 1～2％ larger k_(eff) than CENDL-2 for the cores with ~(238)U fuel and reflector. The difference is mainly caused by ~(238)U data, especially its inelastic data. The inelastic scattering cross sections of ~(238)U from B-6.2 makes the fast neutron spectrum hardened and increases of neutron production rate. In thermal reactor benchmark testing, the k_(eff) values calculated using B-6.2 for both lattice assembly TRX-1 and TRX-2 with metal uranium fuel rod are underestimated from 0.6％ to 1％. And for BAPL-UO_2-1, -2 and -3 with uranium oxide fuel rod lattice, the k_(eff) values calculated using B-6.2 are underestimated by about 0.2％ to 0.5％.     

移植到CYBER-170/825计算机上的核数据处理系统NJOY

一、引言评价核数据的处理是核数据工作的重要环节,是连接核数据的来源(评价核数据库)和核数据的用户(各类核工程、核技术应用单位)的桥梁。NJOY是当今世界上最先进,并在各国广泛使用的核数据处理系统,它是一个综合性的计算机程序包,可将ENDF/B-IV和B-V的评价核数据制作成点状或多群形式的中子和光子截面。本文重点介绍核数据处理系统NJOY的主要用途和功能,简述在CYBER-170/825机上移植它的情况。NJOY在CYBER计算机上移植成功,使核数据中心为核电和核工程设计提供多种格式中子和γ群常数成为可能。今后,NTOY将在我国其它有关单位推广使用。     

MCNP温度相关中子截面库制作方法

在截面库研制过程中,着重考虑了在反应堆设计与运行温度范围的温度点;使用NJOY软件将ENDF格式的中子截面文件处理为ACE (A Compact ENDF) 格式的点截面文件,供MCNP程序使用.验证过程应用了3种不同类型的临界基准题:简单的球形几何基准题、板式燃料元件实验装置和带有可燃毒物的堆芯.结果表明,3种临界基准题所得到的验证结果均较为理想,在精确度方面也达到了要求.证明了使用NJOY制作截面库方法的正确性.