PHYSICAL REVIEW C 81, 044616 (2010) 197Au(n,γ ) cross section in the resonance region C. Massimi,1 C. Domingo-Pardo,2,* G. Vannini,1 L. Audouin,3 C. Guerrero,4 U. Abbondanno,5 G. Aerts,6 H. Álvarez,7 F. Álvarez-Velarde,4 S. Andriamonje,6 J. Andrzejewski,8 P. Assimakopoulos,9,† G. Badurek,10 P. Baumann,11 F. Bečvář,12 F. Belloni,5 E. Berthoumieux,6 F. Calviño,13 M. Calviani,14 D. Cano-Ott,4 R. Capote,15,16 C. Carrapiço,6,17 P. Cennini,14 V. Chepel,18 E. Chiaveri,14 N. Colonna,19 G. Cortes,20 A. Couture,21 J. Cox,21 M. Dahlfors,14 S. David,3 I. Dillmann,22 W. Dridi,6 I. Duran,7 C. Eleftheriadis,23 L. Ferrant,3,† A. Ferrari,14 R. Ferreira-Marques,18 K. Fujii,5 W. Furman,24 S. Galanopoulos,25 I. F. Gonçalves,17 E. González-Romero,4 F. Gramegna,26 F. Gunsing,6 B. Haas,27 R. Haight,28 M. Heil,22 A. Herrera-Martinez,14 M. Igashira,29 E. Jericha,10 F. Käppeler,22 Y. Kadi,14 D. Karadimos,9 D. Karamanis,9 M. Kerveno,11 P. Koehler,30 E. Kossionides,31 M. Krtička,12 C. Lampoudis,32 C. Lederer,33 H. Leeb,10 A. Lindote,18 I. Lopes,18 M. Lozano,16 S. Lukic,11 J. Marganiec,8 S. Marrone,19 T. Martı́nez,4 P. Mastinu,26 E. Mendoza,4 A. Mengoni,14,15 P. M. Milazzo,5 C. Moreau,5 M. Mosconi,22 F. Neves,18 H. Oberhummer,10 S. O’Brien,21 J. Pancin,6 C. Papadopoulos,25 C. Paradela,7 A. Pavlik,33 P. Pavlopoulos,34 G. Perdikakis,25 L. Perrot,6 M. T. Pigni,10 R. Plag,22 A. Plompen,32 A. Plukis,6 A. Poch,20 J. Praena,26 C. Pretel,20 J. Quesada,16 T. Rauscher,35 R. Reifarth,28 M. Rosetti,36 C. Rubbia,14 G. Rudolf,11 P. Rullhusen,32 L. Sarchiapone,14 R. Sarmento,17 I. Savvidis,23 C. Stephan,3 G. Tagliente,19 J. L. Tain,37 L. Tassan-Got,3 L. Tavora,17 R. Terlizzi,19 P. Vaz,17 A. Ventura,36 D. Villamarin,4 V. Vlachoudis,14 R. Vlastou,25 F. Voss,22 S. Walter,22 M. Wiescher,21 and K. Wisshak22 (n TOF Collaboration) 1Dipartimento di Fisica, Università di Bologna, and Sezione INFN di Bologna, Italy 2GSI Helmholtzzentrum für Schwerionenforschung GmbH, Darmstadt, Germany 3Centre National de la Recherche Scientifique/IN2P3-IPN, Orsay, France 4Centro de Investigaciones Energeticas Medioambientales y Tecnologicas, Madrid, Spain 5Istituto Nazionale di Fisica Nucleare, Trieste, Italy 6CEA/Saclay-IRFU, Gif-sur-Yvette, France 7Universidade de Santiago de Compostela, Spain 8University of Lodz, Lodz, Poland 9University of Ioannina, Greece 10Atominstitut der Österreichischen Universitäten, Technische Universität Wien, Austria 11Centre National de la Recherche Scientifique/IN2P3-IReS, Strasbourg, France 12Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic 13Universidad Politecnica de Madrid, Spain 14CERN, Geneva, Switzerland 15International Atomic Energy Agency (IAEA), Nuclear Data Section, Vienna, Austria 16Universidad de Sevilla, Spain 17Instituto Tecnológico e Nuclear-ITN, Lisbon, Portugal 18LIP-Coimbra & Departamento de Fisica da Universidade de Coimbra, Portugal 19Istituto Nazionale di Fisica Nucleare, Bari, Italy 20Universitat Politecnica de Catalunya, Barcelona, Spain 21University of Notre Dame, Notre Dame, Indiana, USA 22Karlsruhe Institute of Technology, Institut für Kernphysik, Germany 23Aristotle University of Thessaloniki, Greece 24Joint Institute for Nuclear Research, Frank Laboratory of Neutron Physics, Dubna, Russia 25National Technical University of Athens, Greece 26Istituto Nazionale di Fisica Nucleare, Laboratori Nazionali di Legnaro, Italy 27Centre National de la Recherche Scientifique/IN2P3-CENBG, Bordeaux, France 28Los Alamos National Laboratory, New Mexico, USA 29Tokyo Institute of Technology, Tokyo, Japan 30Oak Ridge National Laboratory, Physics Division, Oak Ridge, Tennessee, USA 31NCSR, Athens, Greece 32EC-JRC-IRMM, Geel, Belgium 33Faculty of Physics, University of Vienna, Austria 34Pôle Universitaire Léonard de Vinci, Paris La Défense, France 35Department of Physics-University of Basel, Switzerland 36ENEA, Bologna, Italy 37Instituto de Fı́sica Corpuscular, CSIC-Universidad de Valencia, Spain (Received 29 January 2010; published 27 April 2010) 0556-2813/2010/81(4)/044616(22) 044616-1 ©2010 The American Physical Society C. MASSIMI et al. PHYSICAL REVIEW C 81, 044616 (2010) The (n,γ ) cross section of 197Au has been measured at n TOF in the resolved resonance region, up to 5 keV, with the aim of improving the accuracy in an energy range where it is not yet considered standard. The measurements were performed with two different experimental setup and detection techniques, the total energy method based on C6D6 detectors, and the total absorption calorimetry based on a 4π BaF2 array. By comparing the data collected with the two techniques, two accurate sets of neutron-capture yields have been obtained, which could be the basis for a new evaluation leading to an extended cross-section standard. Overall good agreement is found between the n TOF results and evaluated cross sections, with some significant exceptions for small resonances. A few resonances not included in the existing databases have also been observed. DOI: 10.1103/PhysRevC.81.044616 PACS number(s): 25.40.Ny, 25.40.Lw, 29.30.Hs, 27.80.+w I. INTRODUCTION The main objectives of the experimental activity of the neutron time-of-flight facility, n TOF, at CERN, are accurate measurements of neutron cross sections related to nuclear astrophysics [1,2] and the collection of nuclear data related to emerging nuclear technologies for energy production and nuclear-waste transmutation [3–5]. Most neutron cross sections are measured relative to cross- section standards [6] for normalization to absolute values. So far, the 197Au(n,γ ) reaction at thermal energy and between 0.2 and 2.5 MeV is the only capture standard and most neutron- capture cross-section measurements refer to one or both energy regions. An alternative to the use of cross-section standards is the saturated resonance technique [7] using a low-energy saturated resonance, like, for example, the 4.9-eV resonance in a 197Au(n,γ ) reaction. Owing to its high capture cross-section value, this resonance is saturated for a sample thickness greater than 30 µm. Because of the convenient neutron-induced radioactivity, chemical and isotopic purity, and large thermal neutron capture and resonance capture integral, the Au capture cross section is of great importance, for example, for flux measurements in nuclear reactors, in accelerator mass spectrometry, and in neutron-activation analysis. The 197Au(n,γ ) cross section is not very accurately known in the resolved resonance region (RRR). The few previous measurements were carried out with liquid scintillation detec- tors containing H or F and did not cover the full RRR up to ≈5 keV. Resonance parameters up to 1 keV were determined in Refs. [8–10] by combining the results of different types of neu- tron cross-section measurements (i.e., transmission through thick and thin samples, capture, self-indication, and elastic scattering) and using the so-called area analysis [11]. From the resonance shape analysis of a transmission measurement Alves et al. [12] determined resonance parameters from 1 to 2.5 keV. In the energy region 2.5–5 keV, capture data from Macklin et al. [13] were combined with differential elastic scattering data from Hoffman et al. [14]. The evaluated cross sections in the neutron-reaction li- braries ENDF/B-VI [15] and ENDF/B-VII [16] (the latter based on the compilation of Ref. [17]) show small discrep- ancies. In addition, a few resonances reported by Desjardins et al. [8] and Julien et al. [10] are only partly included. *Corresponding author: c.domingopardo@gsi.de †Deceased. This situation motivated a new measurement of the capture cross section of 197Au at the n TOF facility with the aim of establishing the Au capture standard also in the energy range below 200 keV. To reduce systematic uncertainties as far as possible, the measurement was carried out with different gold samples and by using two independent detection techniques based on a total absorption calorimeter (TAC) and a pair of C6D6 detectors (Sec. II). This article presents the results of a resonance shape analysis with the R-matrix code SAMMY [18] for the resolved resonances in the energy region between 1 eV and 5 keV. The analysis procedure for the TAC and C6D6 data is illustrated in Secs. III and IV, respectively. The comparison of the two data sets with each other and with evaluated cross section data is given in Sec. V. The unresolved resonance region between 5 keV and 1 MeV is being analyzed in parallel and will be presented separately [19]. II. MEASUREMENTS A. The n TOF facility During Phase I of the n TOF facility (2001–2004) the neu- tron beam was produced by spallation induced by a 20 GeV/c proton beam, with up to 7 × 1012 particles per pulse, impinging on a 80 × 80 × 60 cm3 lead target with a repetition rate of 0.4 Hz. These characteristic features of n TOF allow one: (i) to cover the neutron-energy interval from 1 eV to 250 MeV in a single run, (ii) to achieve an extremely high instantaneous neutron flux, and (iii) to prevent pulse overlap even for subthermal neutrons. A 5.8-cm-thick water layer surrounding the lead target serves as coolant and as a moderator of the initially fast neutron spectrum, providing a wide neutron energy spectrum with a nearly 1/En isolethargic flux dependence in the neutron energy region from 1 eV to 1 MeV. An evacuated beam line leads to the experimental area at a distance of 185 m from the lead target. The neutron beam line is extended for an additional 12 m beyond the experimental area to minimize the background from backscattered neutrons. A full description of the characteristics and performance of the facility can be found in Refs. [20–22]. The neutron beam is shaped by two collimators at 135 and 175 m from the spallation target. For capture measurements, the second collimator is used with an inner diameter of 1.8 cm, 044616-2 http://dx.doi.org/10.1103/PhysRevC.81.044616 197Au(n,γ ) CROSS SECTION IN THE . . . PHYSICAL REVIEW C 81, 044616 (2010) resulting in a nearly symmetric Gaussian-shaped beam profile at the sample position with a standard deviation of about 0.77 cm at low neutron energies [23]. The neutron energy is determined via time of flight (TOF), using the γ flash from the impact of the proton pulse on the spallation target as the time reference. The relationship for converting TOF into neutron energy was accurately verified in the energy range from 1 eV to ∼1 MeV by means of specific capture resonances in 32S, 193Ir, and 238U, which are accepted energy standards [24]. For each detector signal, the corresponding TOF is determined on an event-by-event basis with an accuracy of about 2 ns. The data acquisition system (DAQ) [25] with 54 channels consists of high-frequency flash analog-to-digital converters (FADCs) [26]. Each channel has an 8-Mbyte memory buffer and is operated at a rate of 500 Msamples/s. In combination with the low duty cycle, the DAQ allows one to record the full sequence of signals in each detector in a TOF interval from relativistic neutron energies down to approximately 1 eV. This operation mode corresponds to a zero dead-time data acquisition that is important for avoiding large dead-time corrections at low neutron energies, where the (n,γ ) cross section of Au is rather large. After zero suppression, the data are reduced and stored in the CERN central data recording system. Specially designed pulse-shape-analysis routines are used in the data-reduction stage to extract amplitude, integrated slow and fast component, and TOF from the digitized detector signals. This information, together with the corresponding detector number and the number of protons in the respective pulse, are then used for further data analysis. For more details, see Ref. [25]. B. Neutron-capture detectors Neutron-capture events are characterized by γ -ray cascades leading from the excited state to the ground state of the compound nucleus formed in the reaction. In the n TOF measurements, a total-energy-detection system with two C6D6 liquid scintillation detectors, as well as a total γ -ray-absorption calorimeter (TAC), have been used for measurements of capture cross sections. These two techniques are briefly described in the following. A first set of measurements was carried out using two C6D6 detectors, which have been specially designed [27] with the aim of reducing the γ -ray background induced by neutrons scattered in the sample and captured in or near the detectors. As illustrated in Refs. [28,29], this background has been recognized as a relevant source of error in previous measurements. Recorded events in the C6D6 detectors need to be treated by the pulse height weighting technique [30] to achieve the proper energy-dependence of the γ -ray efficiency, as described in more detail in Sec. IV. The n TOF TAC [31–35] is a 4π detector with nearly 100% detection efficiency for capture γ -ray cascades and an energy resolution of 15% at 662 keV and 6% at 6.1 MeV. It consists of 40 BaF2 crystals contained in 10B-loaded carbon-fiber capsules forming a spherical shell 15 cm in thickness and with an inner diameter of 20 cm. Neutrons scattered from the sample in the center of the TAC are moderated and partly absorbed in a TABLE I. Gold samples for the two capture measurements. TAC C6D6 Diameter (cm) 1.0 2.205 Mass (g) 0.1854 1.871 Thickness (cm) 1.22 × 10−2 2.5 × 10−2 Areal density (at/b) 7.3 × 10−4 1.498 × 10−3 5-cm-thick spherical shell made of C12H20O4(6Li)2 surround- ing the sample. The TAC is ideal for capture measurements of low mass samples, as well as of radioactive and fissile isotopes, owing to its very high total efficiency and because it allows one to select capture reactions via the total energy of the γ -ray cascade and to reject events attributable to other processes, in particular in-beam γ rays from neutron captures in the water moderator of the spallation target. A certain drawback of the device is the relatively high neutron sensitivity, mostly attributable to the capture of scattered neutrons in the Ba isotopes of the scintillator (Sec. III B2). To some extent this problem has been reduced by means of the absorber shell around the sample and the 10B-loaded carbon-fiber capsules. Contrary to the TAC, the C6D6 setup is optimized for cases where the total cross section is strongly dominated by the elastic channel. These detectors are, in fact, characterized by a very low neutron sensitivity of about 10−4, two orders of magnitude smaller than that of the TAC, thus providing reliable results even for very small �γ /�n ratios. The setup for the capture measurements is complemented by the silicon flux monitor (SiMon) [36], which consists of a thin 6Li deposit on a thin Mylar foil surrounded by a set of four silicon detectors outside the neutron beam for recording the tritons and α particles from the 6Li(n,α)3H reaction. C. Samples Gold samples, which differed in size and thickness, were used in the measurements to control sample-related systematic effects. The characteristics of the samples are listed in Table I. In addition to gold, samples of natC and natPb of the same diameter as the Au samples have been used to evaluate the background owing to sample scattered neutrons and in-beam γ rays. III. ANALYSIS OF THE TAC DATA The energy calibration of each individual BaF2 crystal was obtained by means of standard γ -ray sources, that is, 137Cs (662 keV), 88Y (898 and 1836 keV), and Pu/C (6131 keV from 16O). The energy resolution of each BaF2 module and of the entire array were obtained from these measurements as well. A. From measured count rate to capture yield The processed information from the 40 BaF2 crystals is combined off line in a so-called calorimetric routine with the aim of identifying capture events. Although the time resolution of each crystal is less than 2 ns owing to the very fast decay time of the BaF2 scintillators, the overall time resolution of the TAC 044616-3 C. MASSIMI et al. PHYSICAL REVIEW C 81, 044616 (2010) is larger (i.e., 26 ns) owing to the uncertainty in the calibration and synchronization of the different FADCs. Therefore, the condition that defines an event in the TAC is the recording of signals in two or more crystals within a coincidence window of 26 ns. A 150-keV threshold is used for the individual signals to reject electronic noise and to minimize pileup effects (see Sec. III B1). For each processed event, the calorimetric routine returns the total energy deposited in the TAC (ETAC), the incoming neutron energy (En), and Mγ , the number of BaF2 crystals in which a γ ray is detected above threshold. The segmentation of the TAC is enough to ensure a close correlation between the multiplicity of the detected event and the number of γ rays emitted in the capture cascade. The probability that a capture reaction occurs in the sample is the capture yield, that is, that fraction of neutron beam that undergoes a capture reaction in the sample. Experimentally, it is obtained from the ratio of the total counts detected by the TAC, CAu(En), and the incoming neutron fluence �(En) integrated over the beam profile, Yexp(En) = CAu(En) − Cempty(En) ε · f · �(En) , (1) where Cempty(En) are the counts measured without the sample and represent the sample-independent background (other sources of background are discussed later), ε is the TAC efficiency for detecting a capture event, and f is the fraction of the neutron beam intercepting the sample. The correction factors ε and f are independent of neutron energy in the range considered here. The efficiency ε depends on the conditions of the analysis, that is, on the multiplicity window and the energy cuts chosen for the TAC response. The fact that the neutron beam profile varies very slightly with neutron energy was also properly taken into account. Because the absolute normalization in our analysis is obtained via the 4.9-eV saturated resonance, it is not necessary to know the absolute value of the flux, but only the relative energy dependence of the neutron flux up to few keV, which has been measured with a 235U parallel plate fission ionization chamber of PTB Braunschweig, Germany [37]. At neutron energies below 1 keV, we used the flux from the SiMon, which was normalized to the former data in the overlapping energy region. After the normalization factor N was determined via the 4.9 eV resonance, the experimental capture yield is Yexp(En) = N · CAu(En) − Cempty(En) �(En) . (2) Figure 1 shows the total energy deposited in the TAC for the samples used in the measurement. The peak at 6.5 MeV corresponding to the excitation energy of the compound nucleus formed after a neutron capture on 197Au is clearly visible. Moreover, background components are also present. A delicate part of the data analysis consists of the choice of the optimal thresholds for the deposited energy ETAC to maximize the capture-to-background ratio. The selection criteria in the present analysis are illustrated in Figs. 1 and 2. The adopted conditions are 3.5 < ETAC < 7.5 MeV on the total deposited energy and Mγ � 2. As shown 10 -3 10 -2 10 -1 1 10 2 4 6 8 10 12 14 ETAC (MeV) co u n ts /p ro to n b u n ch H(n,γ) Au(n,γ) Ba(n,γ) B(n,αγ) n + Au n + empty sample n + C FIG. 1. (Color online) The spectra of the energy deposited in the TAC measured with the Au and C samples compared to the case without sample (empty) in the neutron energy range 1 < En < 5000 eV. The adopted thresholds for the deposited energy are indicated by dashed vertical lines. later, the choice for ETAC minimizes the neutron sensitivity, because it allows us to reject the 2.2-MeV γ rays produced by hydrogen capture in the absorber around the sample com- pletely and neutron captures by the Ba isotopes in the crystals partly (in particular from the odd nuclei 135,137Ba, which are characterized by capture energies above 7 MeV). As a further advantage, pileup of two consecutive capture cascades, which mimics events with large total energy deposition, is reduced (Sec. III B1). Although the overall efficiency decreased to 60% by these conditions, the resulting signal-to-background ratio is drastically improved, as shown in Fig. 2. B. Corrections and background evaluation The capture yield measured with the TAC must be corrected for systematic effects, before performing a resonance analysis 10 -4 10 -3 10 -2 10 -1 1 1 10 10 2 10 3 10 4 10 5 10 6 En (eV) C ap tu re y ie ld No conditions 3.51 µs−1 with a corresponding dead time correction of 6%, the difference between the yields extracted for the two proton beam modes was found to be less than 1%, 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 58 60 62 En (eV) C ap tu re y ie ld Parasitic mode Dedicated mode 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 240 250 260 En (eV) Parasitic mode Dedicated mode FIG. 6. (Color online) Capture yields from data obtained in parasitic and dedicated beam mode for resonances with the largest dead-time corrections. as illustrated in Fig. 6, thus providing confidence in the validity of the dead-time correction. 2. Background The background components, which result from (i) in-beam γ rays, (ii) ambient background, and (iii) α radioactivity of Ra contaminants in the scintillator [40] were studied by means of dedicated measurements and have been reduced by the conditions on ETAC discussed previously and by pulse-shape discrimination of the BaF2 signals. Another background component is attributable to sample- scattered neutrons. Scattered neutrons can be captured inside the TAC, mainly by the Ba isotopes of the scintillator, and may contaminate the capture yield from the Au sample. This background depends on the neutron sensitivity of the detector [41,42], which can be defined as the ratio between the efficiencies for detecting scattered neutrons, εn, and capture events, εγ . This background causes an artificial increase of the resonance area, particularly in resonances where the neutron width exceeds the radiative width. In Fig. 7 the capture yield of the 197Au sample is compared with the yields from background runs with natC and natPb and without sample, which were all obtained with the analysis conditions described previously. The Pb measurement provides a good estimate of the sample-dependent component of the backgound in the Au measurement, being the areal density, the atomic number and the nonresonant elastic cross section of the two samples comparable. Because of the very similar shape of Pb and C yields, one can conclude that the background owing to in-beam γ rays is small. Moreover, the Pb and C yields are close to the yield obtained without sample, demonstrating the low level of the residual background, which is attributable mainly to the neutron sensitivity of the detector. Several methods have been proposed to determine this background component [43,44]. In the present analysis the neutron sensitivity was determined from a measurement with a thick graphite sample. In this case, the measured count rates are 044616-6 197Au(n,γ ) CROSS SECTION IN THE . . . PHYSICAL REVIEW C 81, 044616 (2010) 10 -3 10 -2 10 -1 1 1 10 10 2 10 3 10 4 En (eV) C ap tu re y ie ld n + Au n + empty sample FIG. 7. (Color online) (Left) Experimental capture yield measured with Au and without sample. (Right) Experimental yields measured without sample and with the natC and natPb samples, all analyzed with the same conditions for energy deposition and multiplicity. Resonances in the natPb yield are due to a small Sb contamination. weighted by the ratio of the capture and elastic cross sections of Au and C, respectively, εn εγ = CC CAu Y Au γ Y C n , (4) where CC and CAu are the background corrected number of counts as a function of neutron energy. The capture yield for 197Au and the one for elastic scattering for 12C are calculated from the evaluated cross sections in the ENDF/B-VI library [45]. As shown in Fig. 8 the neutron sensitivity of the 4π BaF2 detector is about 0.1, three orders of magnitude higher than that of the n TOF C6D6 setup [27]. However, the background attributable to scattered neutrons can be reduced in an efficient way by suitable conditions on the total deposited energy ETAC. The optimal condition can be derived from the distributions of the total deposited energy shown in Fig. 1 for Au, C, 10 -2 10 -1 1 10 2 10 3 10 4 10 5 En (eV) N eu tr o n s en si ti vi ty ( ε n /ε γ) No conditions (i. e. Mγ≥2) Mγ≥3 ETAC>2.5 MeV, Mγ≥2 3.5