Physics of Neutron Interactions with 238 U: New Developments and Challenges

Similar documents
Physics of Neutron Interactions with 238 U: New Developments and Challenges

EVALUATION OF NEUTRON INDUCED REACTIONS ON 238 U NUCLEUS

Challenges in nuclear data evaluation of actinide nuclei

Elastic and inelastic scattering of neutrons on 238 U nucleus

Review of nuclear data of major actinides and 56 Fe in JENDL-4.0

Nuclear Data Evaluation of 55 Mn by the EMPIRE code with Emphasis on the Capture Cross Section

Statistical Model Calculations for Neutron Radiative Capture Process

Present Status and Plans of JENDL FP Data Evaluation Project

Nuclear Data Activities in the IAEA-NDS

Invited. The latest BROND-3 developments. 1 Introduction. 2 Cross section evaluations for actinides

Status of Subgroup 24

CIELO Project. M. Herman 1) D. Brown 1), R. Capote 2), G. Nobre 1), A. Trkov 2) for the CIELO Collaboration 3)

Needs for Nuclear Reactions on Actinides

The Updated Version of Chinese Evaluated Nuclear Data Library (CENDL-3.1)

Available online at ScienceDirect. Physics Procedia 64 (2015 ) 48 54

The updated version of the Chinese Evaluated Nuclear Data Library (CENDL-3.1) and China nuclear data evaluation activities

Monte Carlo Methods for Uncertainly Analysis Using the Bayesian R-Matrix Code SAMMY

Introduction to Nuclear Data

Monte Carlo Simulation for Statistical Decay of Compound Nucleus

PIA and REWIND: Two New Methodologies for Cross Section Adjustment. G. Palmiotti and M. Salvatores

AccApp 11 Preprint Williams et al. 1

arxiv: v1 [nucl-th] 17 Jan 2019

RIPL Objective (1993)

Status of ENDF/B-VII.1 library

Statistical-Model and Direct-Semidirect-Model Calculations of Neutron Radiative Capture Process

ORNL Nuclear Data Evaluation Accomplishments for FY 2013

Subgroup A: Nuclear Model Codes Report to the WPEC Meeting, San Diego, May 13-15, 2003

WPEC Sub group 34 Coordinated evaluation of 239 Pu in the resonance region

VERIFICATION OF MONTE CARLO CALCULATIONS OF THE NEUTRON FLUX IN THE CAROUSEL CHANNELS OF THE TRIGA MARK II REACTOR, LJUBLJANA

Neutron-induced reactions on U and Th a new approach via AMS in collaboration with: KIT (Karlsruhe): F. Käppeler, I. Dillmann

VERIFICATION OFENDF/B-VII.0, ENDF/B-VII.1 AND JENDL-4.0 NUCLEAR DATA LIBRARIES FOR CRITICALITY CALCULATIONS USING NEA/NSC BENCHMARKS

Annals of Nuclear Energy

Analysis of the TRIGA Reactor Benchmarks with TRIPOLI 4.4

Integral of--nuclear plus interference components. of the elastic scattering cross section. Sum of binary (p,n ) and (p,x) reactions

Joint ICTP-IAEA Workshop on Nuclear Reaction Data for Advanced Reactor Technologies May 2010

High Energy Neutron Scattering Benchmark of Monte Carlo Computations

Nuclear Cross-Section Measurements at the Manuel Lujan Jr. Neutron Scattering Center

(NUCLEAR) DATA EVALUATION METHODOLOGY INCLUDING ESTIMATES OF COVARIANCE

Excitation functions and isotopic effects in (n,p) reactions for stable iron isotopes from. reaction threshold to 20 MeV.

Application and Validation of Event Generator in the PHITS Code for the Low-Energy Neutron-Induced Reactions

Vladimir Sobes 2, Luiz Leal 3, Andrej Trkov 4 and Matt Falk 5

Sensitivity and Uncertainty Analysis of the k eff and b eff for the ICSBEP and IRPhE Benchmarks

Nuclear Data Section Department of Nuclear Sciences and Applications

NUCLEAR DATA VERIFICATION USING GALILEE AND TRIPOLI-4

Microscopic cross sections : an utopia? S. Hilaire 1, A.J. Koning 2 & S. Goriely 3.

A dispersive optical model potential for nucleon induced reactions on 238 U and 232 Th nuclei with full coupling

Recent Advances in the ENDF/B Library

Status Report of WPEC Subgroup 7 April Nuclear Data Standards Coordinator: A. D. Carlson

Measurement of (n, xn γ) reaction cross sections in W isotopes

in Cross-Section Data

Invited. Status and future work of the NEA Working Party on international nuclear data Evaluation Cooperation. 1 Introduction. 2 Recent achievements

Unified Monte Carlo evaluation method

Invited. Recent advances in the JENDL project. 1 Introduction. 2 Activities for general-purpose file

Nuclear data evaluation of 206 Pb for proton- and neutron-induced reaction in energy region from 20 to 200 MeV

Complete activation data libraries for all incident particles, all energies and including covariance data

FRENDY: A new nuclear data processing system being developed at JAEA

Resonance Evaluations of 235 U for the CIELO Project

Covariance Generation using CONRAD and SAMMY Computer Codes

Photonuclear Reaction Cross Sections for Gallium Isotopes. Serkan Akkoyun 1, Tuncay Bayram 2

Measurements of the Neutron Scattering Spectrum from 238 U and Comparison of the Results with a Calculation at the 36.

Presentation for the CIELO Meeting of the NEA 9-11 May 2016 Paris, France. The Chinese work on 56 Fe

INDC International Nuclear Data Committee

New Neutron-Induced Cross-Section Measurements for Weak s-process Studies

Fission-yield data. Karl-Heinz Schmidt

Statistical Theory for the Beta-Delayed Neutron and Gamma-Ray Emission

Crawling Towards a (more) complete description of deuteron-induced reactions

Introduction to Nuclear Engineering

Testing of Nuclear Data Libraries for Fission Products

Preliminary Uncertainty Analysis at ANL

Assessment of the Unresolved Resonance Treatment for Cross-section and Covariance Representation

IAEA CIELO Evaluation of Neutron-induced Reactions on 235 Uand 238 U Targets

Measurements of Neutron Capture Cross Sections for 237, 238 Np

TENDL-TMC for dpa and pka

Nuclear Reaction Rates for Astrophysical Nucleosynthesis

Optimization studies of photo-neutron production in high-z metallic targets using high energy electron beam for ADS and transmutation

Office of Nonproliferation and Verification Research and Development University and Industry Technical Interchange (UITI2011) Review Meeting

Analytical Validation of Uncertainty in Reactor Physics Parameters for Nuclear Transmutation Systems

Nuclear Data for Innovative Fast Reactors: Impact of Uncertainties and New Requirements

Arjan Plompen. Measurements of sodium inelastic scattering and deuterium elastic scattering

Nuclear Data Activities at the IAEA-NDS

TENDL 2017: better cross sections, better covariances

The surrogate-reaction method: status and perspectives. Beatriz Jurado, CENBG, France

PIA: Progressive Incremental Adjustment

Reactor Physics: General I. Thermal Total Cross Sections of Europium from Neutron Capture and Transmission Measurements

Present Status of the JENDL Project. Osamu IWAMOTO and Kenji YOKOYAMA Japan Atomic Energy Agency

Available online at ScienceDirect. Physics Procedia 64 (2015 ) Neutron-induced fission cross section of 240,242 Pu

Considerations for Measurements in Support of Thermal Scattering Data Evaluations. Ayman I. Hawari

U.S. Experimental Activities Report. WPEC Meeting, May Donald L. Smith, Chair. on behalf of the

Report of Workshop on EXFOR Compilation 2014

Correlated Prompt Fission Data

Coordinated Research Project on Photonuclear Data and Photon Strength Functions Approved in July 2015; Code F41032; Duration 2016 t 2020.

TENDL-2011 processing and criticality benchmarking

A Monte Carlo Simulation of Prompt Gamma Emission from Fission Fragments

Lecture 3 Nuclear Data Neutron Interactions and Applications Spring 2010

Annals of Nuclear Energy

Neutron capture and fission reactions on. sections, -ratios and prompt -ray emission from fission. 1 Introduction and Motivation

International projects for Nuclear Data evaluation and their coordination. Dr Robert Mills, NNL Research Fellow for Nuclear Data

Nuclear transmutation strategies for management of long-lived fission products

Measurement of prompt fission γ-ray spectra in fast neutroninduced

ENDF-6 Formats Manual

Transcription:

Available online at www.sciencedirect.com Nuclear Data Sheets 118 (2014) 26 31 www.elsevier.com/locate/nds Physics of Neutron Interactions with 238 U: New Developments and Challenges R. Capote, 1, A. Trkov, 2 M. Sin, 3 M. Herman, 4 A. Daskalakis, 5 and Y. Danon 5 1 NAPC Nuclear Data Section, International Atomic Energy Agency, PO Box 100, Vienna A 1400, Austria 2 Jožef Stefan Institute, Jamova cesta 39, Ljubljana SI 1000, Slovenia 3 Nuclear Physics Department, Bucharest University, Bucharest-Magurele RO 077125, Romania 4 National Nuclear Data Center, Brookhaven National Laboratory, Upton, NY 11973 5000, USA 5 Gaerttner LINAC Center, Rensselaer Polytechnic Institute, Troy, NY 12180, USA The latest release of the EMPIRE-3.1 system (codename Rivoli) is being used in the advanced modeling of neutron induced reactions on the 238 U nucleus with the aim of improving our knowledge of neutron scattering. The reaction model includes: (i) a new rotational-vibrational dispersive optical model potential coupling the low-lying collective bands of vibrational character observed in even-even actinides, (ii) the Engelbrecht-Weidenmüller transformation allowing for inclusion of compound-direct interference effects enhanced by a dispersive treatment of the optical model potential, (iii) a multi-humped fission barrier with absorption in the secondary well as described within the optical model for fission, and (iv) a modified Lorentzian model (MLO) of the radiative strength function. Impact of the advanced modeling on elastic and inelastic scattering cross section is being assessed by both comparison with selected microscopic experimental data and integral criticality benchmarks (e.g. FLATTOP, JEMIMA and BIGTEN assemblies). Benchmark calculations provide feedback to improve the reaction modeling and reduce both model and model-parameters uncertainties. Additionally, neutron scattering yields on 238 U measured accurately at RPI by the time-of-flight technique at 29, 60, 112 and 153 degrees have been used as a further constraint on the incident energy dependence of elastic and inelastically scattered neutrons. Improvement of scattering cross sections in existing libraries is discussed. I. INTRODUCTION 238 U constitutes more than 90% of nuclear fuel in power reactors, being one of the most important isotopes for neutron transport calculations in the active zone of a reactor. Recently an IAEA technical meeting on Inelastic Scattering and Capture Cross-section Data of Major Actinides in the Fast Neutron Region was held at IAEA Headquarters, Vienna, Austria to review the status of nuclear data libraries for these cross sections, the status of the experimental results by which these data can be tested, and to evaluate what advances in nuclear modeling and measurement technique may improve our knowledgeofthesecrosssections[1]. Recentlyevaluated nuclear data files of the major actinides (e.g. JEFF- 3.1 [2], CENDL-3.1 [3], JENDL-4.0 [4], and ENDF/B- VII.1 [5]) show significant differences in the inelastic scattering cross sections over the fast neutron energy range. Attendees at this IAEA meeting concluded that advances in modeling are substantial,... and that significant improvement can be made in reducing modeling uncertainties for capture and inelastic scattering [1]. Such reduction of the uncertainty of neutron scattering data on 238 U Corresponding author: r.capotenoy@iaea.org in the fast neutron region has also been requested by WPEC-SG26 [6], motivating new evaluation efforts. Fast neutron induced reactions on 238 U can be an ideal target to perform a compound-nucleus cross-section intercomparison, as the fission cross section below 1 MeV is almost negligible and above 1 MeV is a neutron cross section standard [7]), the capture cross section is known to very high accuracy (from the neutron standard fit [7]), and the total cross section is expected to be very well defined by the experimental data. Therefore, the differences among the model calculations concern the partitioning of the compound-nucleus formation cross section over the elastic and inelastic scattering channels [1]. The present work addresses these issues, and a report is provided of on-going evaluation of neutron induced reactions on 238 U in the fast neutron region. II. NUCLEAR REACTION MODELING Given the limited accuracy and availability of experimental inelastic scattering data on 238 U, nuclear reaction modeling plays a central role in the evaluation of inelastic scattering cross sections. Therefore improvement and benchmarks of the modeling and the model parameters are crucial for better prediction of the unknown inelas- http://dx.doi.org/10.1016/j.nds.2014.04.003 0090-3752/2014 Published by Elsevier B.V.

MPIRE (El= kev) MPIRE (El=827 kev) Cross Section (barns) 4 3 2 1 EMPIRE (n,nonel) EMPIRE (n,nonel)* 2011 Dietrich* 1968 Voignier 1965 Batchelor 1965 Batchelor 1965 Batchelor 1965 Degtjarev 1963 Mac Gregor 1963 Mac Gregor 1963 Mac Gregor 1963 Mc Taggart 1963 Didier 1961 Andreev 1961 Andreev 1961 Andreev 1961 Cohen 1961 Cohen 1961 Degtjarev 1961 Degtjarev 1961 Degtjarev 1958 Flerov 1958 Lebedev 1957 Allen 1957 Allen 1957 Allen 1956 Poze 1956 Beyster 1956 Beyster 0 0.5 1 5 10 Incident Energy (MeV) FIG. 1. Excitation function for 1 and 5 coupled levels (left); non-elastic cross section for n + 238 U (right). Experimental data have been taken from EXFOR [16]. tic scattering cross sections. All model calculations in the fast energy range employed the EMPIRE-3.1 system (Rivoli) [8]. Starting values for nuclear model parameters were taken from the RIPL recommendations [9]. Preequilibrium emission was considered in terms of a onecomponent exciton model, while the Hauser-Feshbach [10] and Hofmann-Richert-Tepel-Weidenmüller [11] versions of the statistical model were used for the compound nucleus cross section calculations. Both approaches account for the multiple-particle emission, the optical model for fission [12, 13] and the full gamma-cascade. A modified Lorenzian radiative-strength function (MLO) recommended by Plujko [9] was adopted, and resulted in very good agreement with the experimental neutron capture database in the fast neutron range. Level densities were described by the EMPIRE-specific formalism (EGSM) in which the superfluid model was used below the critical excitation energy and the Fermi gas model above [8, 9]. The fast energy range is of particular interest to many nuclear applications, and under such circumstances we have identified that the calculated inelastic scattering cross sections depend on the compound nucleus decay (especially on the level density of the target nucleus) and the optical model potential. Significant advances have been made in the formulation and parameterization of coupledchannel optical model potentials based on dispersion relations capable of describing existing total cross section measurements within quoted experimental uncertainties in the whole energy range of interest (see Refs. [9, 14] and references therein). A new rotational-vibrational dispersive optical model potential has been derived that couples the low-lying collective bands of vibrational character observed in 238 Uand 232 Th nuclei while preserving the good quality of description of other observables [15]. Special attention is paid to the coupling of almost all excited levels below 1 MeV of the excitation energy, because neutron scattering on these low-lying levels plays a significant role in the energy region from 0.5 to 1 MeV of maximum interest for fast reactor systems. The excitation functions of two members of the low-lying octupole band in 238 U are shown in Fig. 1 (left); while the non-elastic cross sections as calculated with mentioned optical model potential are shown in Fig. 1 (right). Unfortunately, experimental non-elastic cross section data below 2 MeV are discrepant which underlines the need to rely on theoretical modeling. Description of data above 3 MeV is good, once the contribution from the first two excited states is subtracted (σ2 + and σ4 + in Fig. 1 (right)). This substraction highlights the fact that all measurements of the elastic cross section of the actinides above 3.4 MeV are actually quasi-elastic measurements, as the first two excited states in 238 U (at 44.5 and 148 kev) can not be resolved from the elastic peak. Thus, all non-elastic data above 3.4 MeV should be decreased accordingly as shown in Fig. 1 (right). III. DIFFERENTIAL DATA Adequate validation of the results of modeling is critically dependent on the experimental data retrieved from the EXFOR database [16]. One problem is that the information compiled in EXFOR constitutes raw data, and errors do occur due to confusion over units, misreporting by the original authors, miscopying/misinterpretation by compilers, etc. Also, some of these data are simply illdefined or erroneous due to defects in the experiments. Finally, since most neutron reaction experiments make use of decay and standards data in their interpretation, changes in the results due to changes in these reference data are to be expected [17]. Furthermore, corrections as implemented from previous evaluations are not compiled into usable databases. The total non-elastic cross sections shown in Fig. 1 (right) in the energy region below 2 MeV are a good example of the type of issue that can arise. Experimental 27

FIG. 2. Neutron induced reaction cross sections on 238 U (left) and effect of the Engelbrecht-Weidenmüller transformation on elastic and inelastic scattering on the first two excited levels of 238 U (right). Experimental data have been taken from EXFOR [16]. data are discrepant and modeling results also disagree with most of the existing data. Modern evaluations are based on the assumption that the selected experimental data will be combined with modeling results (usually employing least-square techniques) to produce a final evaluation. However, the concept of weeding out bad data and performing necessary adjustments to provide statistically consistent experimental results for an evaluation is of critical importance, even though such a process may be somewhat subjective in some respects. Evaluation methods are very sensitive to the use of discrepant data, which may distort the derived results [17]. However, we are still in an early stage of evaluation in which experimental data are only used for model validation. A significant challenge remains to undertake a comprehensive analysis of the experimental database to produce statistically FIG. 3. Calculated neutron inelastic scattering cross sections on 238 U compared with experimental and evaluated data files. Experimental data have been taken from EXFOR [16]. consistent data. Calculated cross sections for all major channels below 15 MeV incident neutron energy are shown in Fig. 2 (left). Overall agreement with the experimental data is good, and the component contributions show that the elastic cross section is very important over the whole energy range. Of the non-elastic reaction channels, neutron capture is the dominant process below 45 kev incident energy; above which inelastic scattering increases rapidly and becomes dominant up to 5 MeV with competition from neutron induced fission above 1 MeV; finally, above 6 MeV, the fission process and multiple neutron emission dominate. An interesting physical effect is demonstrated in Fig. 2 (right). As discussed by Moldauer [18], interference between two channels (e.g. neutron scattering on the ground state and first inelastic level) causes an enhancement of the fluctuating compound-nucleus (CN) cross sections, especially when there are just two channels strongly coupled by direct reactions. This is precisely the case for neutron incident energies below 500 kev, where the contribution of scattering on the second excited level is very small as seen in Fig. 2 (right). This CN-direct interference effect is probably enhanced by the strong energy dependence of the dispersive optical model at low energies due to the surface dispersive contribution. CN calculations that consider interference effects were carried out by adopting the Engelbrecht Weidenmüller transformation [19] as implemented within the ECIS code [20]. Theinterferenceeffectresultsinanetincreaseoftheinelastic scattering cross section on the first excited level compensated by a reduction in the elastic channel (due to the reduction in the CN elastic cross section). A comparison of calculated total inelastic cross sections (denoted by Empire) with data contained in evaluated application files is shown in Fig. 3. The interference effect is the main culprit in the observed increase of calculated 28

1.0 0.9 0.8 2 EMPIRE (Ei=650 kev) EMPIRE isotropic (Ei=650 kev) U-238(n,el) mu-bar 0.7 0.6 Empire 1990 Litvinskiy+ 0.5 1987 Murzin 1987 Murzin+ 1987 Murzin 0.4 1987 Murzin+ 1986 Goswami+ 0.3 0.2 1982 Haouat 1978 Tsang+ 0.1 1972 Cox 1972 Cox+ 1963 Smith 0 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Incident Energy (MeV) /d (b/sr) 1 0.5 0.2 0 50 100 150 Scattering Angle (deg) FIG. 4. Average cosine of neutron elastic scattering μ on 238 U (left). Angular distribution of neutron elastic scattering at 650 kev incident energy (right) on 238 U. Experimental data have been taken from EXFOR [16]. inelastic cross sections compared with the JEFF-3.1 [2] and JENDL-4 [4] evaluations below 500 kev. However, our calculations are still lower than ENDF/B-VII.1 data in the same energy region [5]. The calculations produce the highest inelastic cross sections from 700 kev up to approximately 1 MeV, as a direct result of the strong coupling between all coupled levels extending up to an excitation energy of 1 MeV. A further point of note is that most of the experimental points shown in Fig. 3for energies higher than 1 MeV are model dependent data (denoted by *). Discrepancies in the 3-6 MeV energy region are being studied on the basis of quasi-differential measurements as discussed in Sect. VI. Differential data such as angle-dependent cross sections, particle-emission spectra and double-differential cross sections defining angle-dependent particle-emission spectra are complex and usually much more difficult to measure accurately. As a rule, evaluations depend significantly on nuclear model calculations, but the choice of models is crucial. Assessments of the models can be made in cases where consistent experimental data are available. The average cosine of scattering μ is a commonly employed measure of the anisotropy of elastic scattering. Experimental data derived from measured angular distributions of resolved elastic scattering for energies lower than 3.4 MeV are in excellent agreement with the smooth curve predicted by the nuclear model calculation in Fig. 4 (left). Such agreement demonstrates the good quality of the optical model potential employed. Unfortunately, as a more integral characteristic, the average cosine of scattering μ, is not sensitive to the fine detail observed in the differential angular distribution. Fig. 4 (right) shows the angle-differential cross sections of 238 Uatanincident neutronenergyof650kev.theredcurveusesamodel in which the CN elastic scattering component is isotropic in the centre-of-mass coordinate system. The green curve is an EMPIRE/ECIS calculation in which the CN elastic scattering is symmetric with respect to 90 degrees, but is not isotropic. An excellent description of the measured angular distribution is achieved by proper consideration of CN elastic anisotropy, which turns out to be highly important in some integral benchmarks with natural or depleted uranium as a reflector due to the enhancement of the backward scattering contribution as shown in Fig. 4 (right). Such an effect is not seen in the comparison of average cosine of scattering μ. IV. ENDF FILE ASSEMBLY ENDF-6 formatted data from the EMPIRE calculations were merged consistently with the resonance data and fission neutrons from the ENDF/B-VII.1 file [5]. For all benchmarking studies presented here, the calculated fission and capture cross sections were also replaced by corresponding cross-section data (MF=3, MT=18, 102) from the ENDF/B-VII.1 file [5]. ENDF/B-VII.1 adopts the IAEA-WPEC-CSEWG standards as described by Carlson et al. [7]. The calculated fission and capture cross sections are very close to standards, but the replacement reduces the variability of the results in highly-sensitive criticality benchmark calculations and allows one to focus on the impact of elastic and inelastic scattering. V. INTEGRAL EXPERIMENTS Benchmark experiments are measurements of integral parameters that can be accurately modeled by computation with very few assumptions and approximations. Typical compilations of this kind are contained in the Handbook of International Criticality Safety Benchmark Experiments [21]. However, we should not overlook additional information from other measurements in well characterized conditions, e.g. thermal cross-section ratio measurements in well-characterised thermal reactor irradiation facilities and cold-neutron beams, cadmium ratio measurements, average cross section measurements in 29

ENDF/B-VII.1 IAEA ib25 FLAPTOP-25 BIGTEN FLAPTOP-Pu JEMIMA-1 JEMIMA-2 JEMIMA-3 JEMIMA-4 FIG. 5. k eff calculations for selected ICSBEP criticality benchmarks sensitive to inelastic neutron scattering on 238 U. Calculation symbols are slightly shifted for clarity. 252 Cf spontaneous fission neutron spectrum and measurements in simulated stellar spectra (30 kev Maxwellian spectrum). Acommonpracticeinthepastwastocompleteanevaluation, and then undertake the benchmarking exercises. This approach has a disadvantage that the feedback to the evaluators is very slow, and therefore an increased tendency has arisen to make benchmarking an integral part of the evaluation process (e.g. see Ref. [22]). Discrepancies between measurements and calculations can guide the evaluator where he/she has the flexibility to adjust certain model parameters that impact on calculated cross sections where no data exist, or discriminate between discrepant measurements of quantities that are being evaluated. An example of improvements in the criticality benchmarks of evaluated nuclear data of 238 U (labelled ib25 ) compared with the data in the ENDF/B- VII.1 library [5] (labelled e71 ) is shown in Fig. 5. Differences are shown between the calculated and reference values for selected benchmarks that are very sensitive to the scattering data of 238 U. Improvements achieved in that trial ib25 evaluation are particularly impressive for the JEMIMA benchmarks. VI. QUASI-DIFFERENTIAL EXPERIMENTS An interesting experiment has been performed at RPI by Danon and co-workers [23]. Neutron yield was measured at different angles by a time-of-flight technique in a white incident neutron spectra extending from 500 kev up to approximately 20 MeV. An important point is that all emitted neutrons were detected including those from elastic and inelastic scattering, fission neutrons and multiple scattering contributions. Comprehensive neutron detection avoids typical uncertainties introduced by splitting of the elastic, inelastic and fission neutron contributions [1]. The disadvantage is that the cross sections can not be inferred from the measurements directly, but the major advantage is the high sensitivity of the measured data to the neutron production cross sections. Neutron production is usually swamped by the elastic component as shown in Fig. 2 (left). However, the elastic contribution in this experiment becomes very low at certain an- Counts 20 10 5000 4000 3000 2000 1000 5 Energy [MeV] 2 1 153 deg Uranium Data ENDF/B-VII.0 JEFF-3.1 IAEA-OUT0067 IAEA-ib25 0 500 1000 1500 2000 2500 3000 Time of Flight [ns] FIG. 6. Neutron yield at 153 degrees backward angle on 238 U target measured at the RPI (dots), and calculated using different nuclear data libraries as described in the text. 0.5 30

gles, raising the importance of the other minor channels, including inelastic scattering. This study could be seen as a quasi-differential benchmark for validating nuclear data, in which the experimental setup is sufficiently simple and very well described to permit an accurate Monte Carlo simulation directly in the time domain. Although a detailed analysis of the experimental data is still in progress, preliminary results are shown in Fig. 6, as a comparison of the measured neutron yield at a backward scattering angle of 153 degrees with results calculated from ENDF/B-VII.1 data and the ib25 evaluation for 238 U. A clear improvement in reproducing the experimental data around an incident neutron energy of 1 MeV is seen for the trial ib25 evaluation when compared with the data taken from the JEFF-3.1 [2] and ENDF/B-VII.1 [5] files. Additional work to improve the description of the measured data is warranted. VII. CONCLUSIONS Advanced modeling of neutron induced reactions on 238 U nucleus has been undertaken to improve our knowledge of the neutron scattering on this extremely important material for nuclear power applications. Anisotropic compound-elastic scattering is shown to be very important to achieve a good description of backward neutron scattering both in differential and quasi-differential experiments for incident neutron energies in the fast neutron range. Interference between the direct and compound reactions predicted by Moldauer is shown to have an impact on calculated inelastic cross sections [18]. This interference effect is particularly significant below 500 kev, where the contribution of scattering on the second excited level to the total inelastic cross section is still very small. Theimpactofadvancedmodelingontheelasticand inelastic scattering cross sections is also being assessed by comparison with integral criticality benchmarks (e.g. FLATTOP, JEMIMA and BIGTEN assemblies). Performance of the trial ib25 evaluation in the selected benchmarks is shown to be better than the ENDF/B-VII.1 evaluation. Additionally, neutron scattering yields on 238 U measured accurately at RPI [23] by the time-of-flight technique at 29, 60, 112 and 153 degrees have been used as an additional constraint on the incident energy dependence of elastic and inelastically scattered neutrons. Our trial evaluation performs better than JEFF-3.1 and ENDF/B-VII.1 evaluations at backward and forward angles. Additional research is still on-going. [1] A.J. Plompen, T. Kawano, R. Capote (Eds), Technical Meeting on Inelastic Scattering and Capture Crosssection Data of Major Actinides in the Fast Neutron Region, report IAEA(NDS)-0597, International Atomic Energy Agency, Vienna, Austria, 2011. [2] A. Koning et al., The JEFF-3.1 Nuclear Data Library, JEFF Report 21, NEA No. 6190, OECD 2006. [3] Chinese Evaluated Nuclear Data Library CENDL-3.1, China Nuclear Data Center, CD-ROM released on 24 December 2009. [4] K. Shibata et al., J. Nucl. Sci. Technol. 48, 1 (2011). [5] M.B. Chadwick et al., Nucl. Data Sheets 112, 2887 (2012). [6] Uncertainty and target accuracy assessment for innovative systems using recent covariance data evaluations, Final report of the NEA/WPEC-26, Coordinator M. Salvatores, WPEC, Nuclear Energy Agency, OECD 2008, Paris, France, ISBN 978-92-64-99053-1. [7] S. Badikov et al., IAEA Scientific and Technical report STI/PUB/1291 (IAEA, Vienna, 2008); A.D. Carlson et al., Nucl. Data Sheets 110, 3215 (2009). [8] M. Herman et al., Nucl. Data Sheets 108, 2655 (2007). [9] R. Capote et al., Nucl. Data Sheets 110, 3107 (2009) (see http://www-nds.iaea.org/ripl-3/). [10] W. Hauser, H. Feshbach, Phys. Rev. 87, 366 (1952). [11] H.M. Hofmann, J. Richert, J.W. Tepel, H.A. Weidenmüller, Ann. Phys. 90, 403 (1975). [12] M. Sin, R. Capote, A. Ventura, M. Herman, P. Obložinský, Phys. Rev. C 74, 014608 (2006). [13] M. Sin, R. Capote, Phys. Rev. C 77, 054601 (2008). [14] E.Sh. Soukhovitskiĩ, R. Capote, J.M. Quesada and S. Chiba, Phys. Rev. C 72, 024604 (2005). [15] J.M. Quesada, R. Capote, E.Sh. Soukhovitskiĩ, S. Chiba, Nucl. Data Sheets 118, 270 (2014). [16] EXchange FORmat database (EXFOR) is maintained by the Network of Nuclear Reaction Data Centers (see http://www-nds.iaea.org/nrdc/); data available online (e.g. at http://www-nds.iaea.org/exfor/). [17] R. Capote, D.L. Smith, A. Trkov, Nuclear Data Evaluation Methodology Including Estimates of Covariances, EPJ Web of Conferences 8, 04001 (2010). [18] P.A. Moldauer, Phys. Rev. C12, 744 (1975). [19] C.A. Engelbrecht, H.A. Weidenmüller, Phys. Rev. C8, 859 (1973). [20] J. Raynal, Optical model and coupled-channel calculations in nuclear physics, p.281, Computing as a language of physics, ICTP International Seminar Course, Trieste, Italy, 1971, International Atomic Energy Agency, Vienna, Austria, 1972. [21] ICSBEP 2006, International Handbook of Evaluated Criticality Safety Benchmark Experiments, Technical Report NEA/NSC/DOC(95)03, NEA Nuclear Science Committee, Nuclear Energy Agency, OECD 2006, Paris, France. [22] A. Trkov et al., Nucl. Data Sheets 112, 3098 (2011). [23] Y. Danon et al., presentationatint.workshoponelastic and Inelastic Scattering, WINS-2012, Boston, 17-19 September 2012. 31