80MEASPECTS 0FNU~DATAEVALUATION AT C.A. PHILI8 and .0 B]~SILIDN Service de Physique et Techniques Nucl@aires Centre d'Etudes de Bruy@res-le-Ch~tel B.P. N" ,21 9~680BR~~Z~ATEL Abstract Transport calculations made in different fields of nuclear applications need know- ledge of microscopic data, both complete and coherent, which for some nuclei exist as an evaluated nuclear data file. The increasing number of available experimental results are however, insufficient toa ssurteh e required qualities of a data set. So we have to use the theory of nuclear reactions. The application varies from the interpolation between ener~ ranges where there is a large number of experimental results to data prediction for unmeasurable nuclei. The general method followed and the application of nuclear models to obtain evaluated data are reported, and some examples given. .I INTRODUCTION The wide field of nuclear applications requires particle transport calculations - mainly neutron transport calculations. The computer codes describing these processes demand a large amount of nuclear data to simulate, as well as possible, nuclear interactions in materials. For these data, known as "evaluated nuclear data", the definition could be the following: a set of nuclear data, complete, coherent and recommended, under a computer file format which meets the requirements of the users. Complete towards their needs, coherent as regards energy and cross-section balances, and recommended as the best known. set The purpose of this talk si to give you the outlines of the generalm ethod applied at BR~-LE-CHATEL (BRC) to obtain these evaluated data [PR83], followed by some examples. Although the term "evaluated nuclear data" covers the reactions induced by different incident particles, we will focus onn eutron nuclear data evaluation. EXISTING N[N/LFAR NE[IT~ON DATA LIBRARIES. To meet the requirements of the users of an evaluated nuclear data set, the first step is to look for a convenient evaluation in the existing evaluatedn uclear data libraries for the application being considered. The main libraries consulted at BRC are: - Evaluated Nuclear Data File (ENDF), _Evaluated - Nuclear Data Library (ENDL), - esenapa_J _Evaluated Nuclear ataD_ Library (JENDL), -Kerndatenbibliotek, Karlsruhe (KEDAK), - Joint Evaluated File (JEF), in their latest released versions. These data are obtained from the Data Basic (Nuclear Energy Agency), one of the four interconnected centers, compiling and dea- ling out through, out the world evaluated ~nd experimental data. Before recommending such an evaluated data set to a user, we usu~lly check if pos- sible the quality of the data~ comparing them between different libraries or with a recent set of experimental results. As a matter of fact, it happens that some ewaluations are performed for a given type of application cor- responding to a field (for ex~nple an ener~y r~ige) on which the evaluators have focused their attention. For other applications, the data will not be as good. Two examples of inconsistencies which can exist in these files are given in Figure 1-a. The comparison between the 5U(n,2n) cross-section data given in FATDF/B-V and JF~DI~2 shows a difference of 50 % at the maximum of the peak. )~igure 1-b shows the differential neutron elastic scattering cross-section on deuterium calculated in the laboratory fr~ne from both center of mass cross-section and trans- fer matrix as given in ENDF/B-IV. The negative part of the cross-section between 90 ° and 125" is obviously unphysical. EVALUATION 0FNUCLF~ DATA. If the data set requested is not precise enough or does not exist, a new evaluation has to be performed. Our basic approach to evaluation :si -to use experimental data directly where possible, - to systematically analyse experimental results with nuclear models, - and to use the resulting theoretically-b~sed par~netrizations to clarify experi- mental discrepancies and to extend evaluations to unmeasured energy regions or nuclei. With this philosophy, the outline of our alternative approaches for obtaining nuclear data is shown in Pig~re ,2 and the nuclear model calculation part in more detail in Figure .3 In order to handle, process, and display all the necessary data-experimental, evaluated and calculated-, we have developed the program system "Synopsis". The basic concept is given in Figure 4. EXAF[PI~ OF ~VALUATIONS. This general method is also applied in other laboratories, including some of the U.S. laboratories with which we collaborate in the nuclear data evaluation field. The main evaluations performed under this collaboration are sun~narized in Figure .5 To illustrate the complete evaluation of the four major tungsten )W( isotopes (182, 183, 184, 186) [AR81 ,] we have chosen on the one hand the 182W total cross-section between I and 20 MeV compared with ENDF/B-V and experimental data sets as shown in Figure ,6 and on the other hand the results of integral data testing for natural W at 14 MeV incident neutron energy. These data are represented in Figure 7 where we n a c compare the pulsed sphere experimental results versus Monte-Carlo calculations from ENDF/B-V )a( and the new evaluation )b( I~T~F[ .] The improvement in the 5-13 MeV energy range is clear, due mainly to our harder inelastic continuum spectrum which includes preequilibrium effects. Among the complete evaluations entirely performed at BRC, the bismuth (Bi) evaluated data set [BE82] is illustrated in Figure 8. In Figure 8-a are compared the experi- mental data results with the doppler broadened total cross-section reconstructed from resonance parameters. The total gamma-spectrum following 14.5 MeV neutron interaction is shown in Figure 8-b. If the agreement between calculated and experi- mental [BESO] spectra is good for gamma-energies lower than 4 MeV, the same does not hold for higher energies, where the cross-section is small and experimental uncer- tainties are large. Before concluding, let us come back to the two examples mentioned at the beginning. The ENDF/B-V and JENDL-2 evaluations of the 23~U(n,2n) cross-section are quite different, especially around 12 MeV, where ENDF/B-IV gives a cross-section of 0.46 b whereas J]~NDL-2 gives 0.7 b. - that is a 50 % difference - 0nly new measurements could remove this ambiguity. The recent measurements [FRSO] in the 6-13 MeV energy range give near ~ MeV a value of 0.87 b, which is even larger than the two other values mentioned (figure 9-a). These new results have been taken into consideration in JEF-I. In ENDF/B-V the neutron elastic angular scattering by deuterium is described by Legendre coefficients in the center-of-mass frame. These coefficients are directly transformed to the laboratory system by using the transformation matrix also given in the file. Use of this matrix leads to a laboratory angular distribution which can become negative, as illustrated in Figure 8-b. This undershoot arises for two rea- sons: the matrix elements are incorrect and the matrix size is not large enough. Use of a correct transformation matrix [BE83] gives a quite different behaviour for the angular distribution, with no negative part. The experimental results are reproduced well. Among the evaluations in progress, the complete evaluation of deuterium is under revision for the high energy range, paying special attention to the absolute double-differential cross-section for the (n,2n) reaction. ~ION. As far as nuclear data evaluation at BR~ES-LE-CHATEL is concerned, we have to continue our effort to give the users the best evaluated data set available, to enable more confident results in transport calculations. Thus, we must continue to improve and develop: I - nuclear models to refine data predictions, 2 - data handling systems as regards the increasing amount and varieties of data to handle, 3 - the checking codes. We must also be shrewd: to question data which may seen well-established, and to request new, indispensable experimental results. We are fort~late that a large part of this effort is made in collaboration with Los Alamos and we hope that it will continue for the benefit of everyone. REFER~CES AR81 : .D.,F ARTHUR, P.G. YOUNG, A.B. SMITH, C.A. PHILIS, "New Tungsten Isotope Evaluation for Neutron Energies between 1.0 and 20 MEW', Trans. Am. Nucl. o S c. 39 (1981) 793-794. B~BO : V.M. BEZOTOSNYI, V.M. GORBACHEV, M.S. ShUYESTSOV, L.M. SUROV, "Group and Total Cross Sections of Formation of y-Ray Quanta upon The Interaction of 14-MeV Neutron with Various Nuclei", Soy. Atomic Energy 49 (1980) 690-694. BE82 : .O BERSILLON, B. CAPUT, C.A. PHILIS "A New Evaluation of Neutron Data for the 209~i between 10-5 eV and 20 MeV", Proceedings of the Internatio~m_l Conference : Nuclear Data for Science and Technology, 6-10 september 1982, Antwerp, pp. 665-668. BE83 : 0.B}~KSILIDN, .A SH~TT, .B CAPUT, "On the transformation of Angular Scat- tering Probabilities between Reference Systems: Survey and Numerical Ana- lysis", NEANDC )E( 230 "L",INDC (FR) 59/L, October 1983. ES81 : G.P. ESTF~, R.C. LITTLE, R.E. SF~ON, ~.D. ARTHUR, P.D. SORAN, "Calculations Using New Tungsten Isotope Evaluations", Trans. Am. Nucl. Soc. 39 (1981) 794-796. FR80 : J. A. FRP~IAUT, BF~TIN, .R BOIS, "Measurement of the 235U(n,2n) Cross Section between threshold and 13 MeV", Nucl. Sci. .gn~] 74 (1980) 29-33. F071 : D.G. FOSTER Jr. and D. GLASGOW, "Neutron Total Cross Sections, 2.5 - 15 MeV", Phys. Roy. C3 (1971) 576. GUT9 : K.GUL, .A WAH~D, M. AHMAD, M. SALE~, NAF~AKHAN, "Inelastic Scattering of 14.8 MeV Neutrons from Deuterons", .J Phys. G 5 (1979) 1107-1116. GU82 : P.T. GUENTHER, A.B. SMITH, J.F. WHALEN, "Fast Neutron Total and Scattering Cross Section for 18~, W 184 and 186W", Phys. Rev. C 26 (1982) 2433-2416. MA67 : R.C MARTIN, P.F. YERGIN, R.II. AUGUSTON, N.N. KAWSHAL, H.A. MEDICUS, E.J. WI~OLD, "MeV Neutron Total Cross Sections of Ta and W Isotopes", Bull. Am. Phys. Soc. 12 (1967) 106. MA72 : D.S. MATHER, P.F. BAMPTON, R.E. COLF~, G. JAMES, P.J. NIND, AWRE Report ° N 072/72 I( 972). ~66 : .S F~SSELT, "Neutron Spectra from Breakup of deute- tons by 14.7 MeV Neutrons", Nucl. Phys. 48 (1963) 512-516. PR83 : Compte rendud'activit@ du Service de Physique Neutronique et Nucl@aire pour l'ann@e 1983", Note CEA N-2396, NEANDC )E( 234 "L", INDC (FR) 62/L, 0ctobre 984. S176 : U.NSINGH, J. RAINWATER, H. .I LIOU, G. HACKEN, J.B. GARG, "Neutron Resonance Spectroscopy: 209Bi", Phys. Rev. C 13 (1976) 124-127. 7 N,D~ O[TrL=Uel LANOTATUPleCC Efft~P DRA TAMROF AYA owi FDNE TAMROF YNA OEZGORADNATS TAMROF FIGURE 4 Synopsis System. YR0~AROBAL ~dCk~ t)l SEPOTOSI( ,L~A LL~ B R)SEPOTOSI( B~ ,L~L U~L uL )sEPoms~( ,L~L tNLL W )SEPOTOSI( ,~NL ~L up142 ~AL FIGURE 5 Main evaluations performed in collaboration with U.S. laboratories. TCERID NOIIISIUOCA RAELCUN GNILEDOM H C A D R P P A ~_ TCERID I ETELPMOCNI (~I ATAD TCERIDNI ] S T N E M E R U S A E H 1 ATAD T N E M S S E S S A S T N E M E R U S A E M 1 ATAD J LEDOM I I YRAILIXUA t RAELCUN LEDOM STNEMENIFER SRETEMARAP" TNEMSSESSA SNOITALUCLAC I I NI OITACIFIREV NOSIRAPMOC F OLEDOM HTI US T N ETSMEETR U S A E M • STMEMIREPXE R A L E C U N AID A T E S A B l FIGURE 2 Outline of our alternative approaches for obtaining nuclear data -i S.MN.OOITALUCLAC SEITILIBARTENEP NOITCES-SSORC TNEMETSUJDA( )SRE TFEOFARAP L- ATOT GNIRETT- ACS NOITCUD~ ORP RAELCUN NOITCAER - CISAB ATAD SLED~F FIGURE 3 Outline of nuclear data calculations. .O.O.N LAN]:ITAffL):WPOC FORI'~T - - FONE T.A.HROF ANY DEZ(OR~NATS T~'IROF FIGURE 4 Synopsis System. YROT~R0~5 lliF/F ~T )SE~OTOSI( ~LHAt ,t~t UAK tu )SEPOTOSi( ,L~L t~t ( SO'I'OPES) I P U~ FIGURE 5 Main evaluations performed in collaboration with U.S. laboratories, O OC o FO 71 t"-- o 8U 82 ×HA J-6 O b- cO O b O T I i i I i i i '"I i 10 ° nE (HeY) ~01 FIGURE 6 Evaluated (solid curve} [AR 81] and experimental values for the 182W total cross section. The dashed curve is ENDF/B.V, Rev.O. 1 If'If I T NT '~''+'' '"'°"-i HeY~source ' / '"'""+ .......... counts ' ' ' I 'i CP, 4~ :3: <~ Z~ <~ I'0 -=j ++,=j ++~+I +~+,+ D" t'l:, P~ "~m t:::l ,-' "~r't I Q X ~" 0 o