ASAP 2002
Astrophysics, Symmetries, and Applied Physics at Spallation Neutron Sources Editors
Paul E. Koehler
Christopher R. Gould
Robert C. Haight
Timothy E. Valentine
Symmetry Experiments
Proposed ASAP Beam Line at SNS
Astrophysics, Symmetries, and Applied Physics at Spallation Neutron Sources
This page is intentionally left blank
ASAP 2002 Astrophysics, Symmetries, and Applied Physics at Spallation Neutron Sources Editors
Paul E. Koehler Oak Ridge National Laboratory, USA
Christopher R. Gould North Carolina State University, USA
Robert Haight Los Alamos National Laboratory, USA
Timothy E. Valentine Oak Ridge National Laboratory, USA
V f e World Scientific wll
New Jersey • London Singapore • Si • Hong Kong
Published by World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, Singapore 596224 USA office: Suite 202, 1060 Main Street, River Edge, NJ 07661 UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE
British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library.
ASTROPHYSICS, SYMMETRIES, AND APPLIED PHYSICS AT SPALLATION NEUTRON SOURCES Copyright © 2002 by World Scientific Publishing Co. Pte. Ltd. All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the Publisher.
For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy is not required from the publisher.
ISBN 981-238-249-6
Printed by Fulsland Offset Printing (S) Pte Ltd, Singapore
FOREWORD AND ACKNOWLEDGMENTS Intense beams of epithermal neutrons from spallation sources are enabling the exploration of new vistas of research in nuclear astrophysics, the study of violation of fundamental symmetries, and applied nuclear physics. There has been, and continues to be an active and productive world-wide community engaged in these areas of research at existing facilities such as electron linacs (e.g. at the Oak Ridge Electron Linear Accelerator, ORELA) and van de Graaff laboratories (e.g. in Karlsruhe, Germany and at the Tokyo Institute of Technology in Japan) as well as new efforts at spallation sources (e.g. at the Los Alamos Neutron Science Center, LANSCE and the CERN n_TOF facility). Building on, the long and distinguished history of research at these facilities, the much higher flux available at spallation sources makes it possible to work with much smaller samples as well as to make measurements with much higher precision. These new capabilities of spallation sources are enabling the exploration of new, exciting areas of research such as: • The nucleosynthesis of the elements in dynamic stellar environments such as pulsing red giant stars and supernovae. • The study of fundamental symmetries such as the nature of parity violation in complex nuclei and the search for violations of time-reversal invariance. • The feasibility and efficiency of using accelerator-driven sub-critical assemblies to transmute dangerous, long-lived radioactive waste to more benign materials as well as several other topics in applied nuclear physics such as criticality safety and nuclear medicine. A unifying feature of all of these fields is the need for the highest intensity source of pulsed epithermal neutrons. The new Spallation Neutron Source (SNS) being built at Oak Ridge National Laboratory (ORNL) will be by far the highest flux pulsed source of epithermal neutrons in the world when it comes on line in 2006. Although the main thrust of the science program at the SNS will be materials science, the facility could provide outstanding opportunities for research in nuclear astrophysics, fundamental symmetries, and applied nuclear physics. To review the current status of these fields and to begin to assemble the scientific case and the community of researchers for future experiments at the SNS, a workshop on "Astrophysics, Symmetries, and Applied Physics" was held during March 11-13 at ORNL. Over 60 scientists, representing 11 different US and 4 different foreign universities as well as many national laboratories around the world participated in the workshop. We thank the speakers for their excellent presentations and everyone for participating in the discussions and making the workshop a success. The scientific organizing committee would like to thank Fred E. Bertrand, Jr. (ORNL Physics Division), John C. Nemeth (Oak Ridge Associated Universities, ORAU), and R. Gil Gilliland (ORNL) for financial support to make the workshop
v
vi possible. We would also like to thank Doro Wiarda (ORNL, Physics Division) for setting up the web site for the workshop. Finally, we would like to thank Ken Carter and staff of ORAU for providing technical and administrative support to the collaboration and to Carlene Stewart (ORAU) for her excellent help in organizing and running the workshop. Paul Koehler Chris Gould Bob Haight Tim Valentine
Schedule for the Workshop on Astrophysics, Symmetries, and Applied Physics at Spallation Neutron Sources
Sunday March 10
19:00-21:00
Reception at Pollard Auditorium in Oak Ridge
Monday March 11
Welcome (Chair: G. Young) 9:00-9:10 9:10-9:15 9:15-9:30
Welcome/G. Young/ORNL Welcome/R. Townsend/ORAU Welcome/J. Roberto/ORNL
Nuclear Astrophysics (Chair: G. Young) 9:30-10:15 10:15-10:45
Laboratory Experiments for Neutron Capture Nucleosynthesis! F. Kappeler, FZK Karlsruhe, Germany DANCE at LANSCE J. Ullmann, LANL
10:45-11:00
Break
11:00-11:30 11:30-12:00
The Astrophysics Program at the CERN n_TOF Facility A. Mengoni, ENEA Bologna, Italy Recent Astrophysics Results from ORELA and Possible Future Experiments at ORELA and SNS P. Koehler, ORNL
12:00-13:30
Lunch
Nuclear Astrophysics (continued) (Chair: Art Champagne) 13:30-13:50
13:50-14:10
Sensitivity of Isotope Yields to Reaction Rates in the Alpha-Rich Freezeout C. Jordan, Clemson University Neutron Reactions of Light Nuclei from Astrophysics and Nuclear Physics Interest
VIII
14:10-14:30 14:30-14:50
14:50-15:10
15:10-15:30
Y. Nagai, Osaka University, Japan "Offline" Radioactive Targets R. Rundberg, LANL Measurement ofthep(n, y)d Cross Section for Big Bang Nucleosynthesis at the Spallation Source LANSCE E. Esch, LANL Phonon Properties of Materials from Resonance Doppler Broadening J. E. Lynn, LANL Break
Applied Nuclear Physics (Chair: R. Haight) 15:30-16:15 16:15-16:45
Applied Physics at Spallation Neutron Sources P. Oblozinsky, BNL Applied Physics Measurements at the CERN n_TOF Facility E. Gonzalez-Romero, CIEMAT Madrid, Spain
Tuesday March 12
Applied Nuclear Physics (continued) (Chair: P. Oblozinsky) 9:00-9:30
Activities of the DOE Nuclear Criticality Safety Program (NCSP) at the Oak Ridge Electron Linear Accelerator (ORELA) T. Valentine, ORNL
9:30-9:50
Parameters for Nuclear Reaction Calculations -Needs for Improvement M. Herman, IAEA Vienna, Austria New Al(n, tf Measurements and Criticality Safety L. Leal, ORNL
9:50-10:10
10:10-10:30
Break
10:30-10:50
Time-of Flight Spectrometer GNEIS with Spallation Neutron Source A. Laptev, PNPI, Gatchina, Russia Measurements of Neutron Capture Cross Sections of Long-Lived Fission Products H. Harada, JNC, Japan
10:50-11:10
IX
11:10-11:30
11:30-11:50
11:50-13:30
Temperature Measurements in Dynamically-Loaded Systems Using Neutron Resonance Spectroscopy at LANSCE V. Yuan, LANL Radioactive Targets from RIA J. Blackmon, ORNL Lunch
Symmetries (Chair: C. Gould) 13:30-14:00 14:00 -14:30
14:30-14:50 14:50-15:10
15:10-15:30
Parity Violation in Neutron Resonances G. Mitchell, North Carolina State University New Experimental Capabilities for Parity Non-Conservation and Time Reversal InvarianceViolation S. Penttila, LANL Symmetry Tests at the Japanese Spallation Neutron Source Y. Masuda, KEK, Japan Violation of Fundamental Symmetries in Resonance Neutron Induced Fission W. Furman, JINR Dubna, Russia Physics of the Fission Process and Parity Violation in Neutron Induced Reactions V. Gudkov, University of South Carolina
15:30-15:50
Break
15:50-16:20
New Possibilities for Parity Violation Studies A. Hayes, LANL Time Reversal Tests and Sign Correlations in Heavy Nuclei C. Gould, North Carolina State University A Five-Fold Correlation Experiment to Measure Time Reversal Invariance Violation Using Neutron Resonances in Holmium P. Huffman, NIST
16:20-16:40 16:40-17:00
X
Wednesday March 13
SNS and Discussions (Chair: Paul Koehler) 9:00-9:30 9:30-10:15
SNS Letter of Intent and Instrument Development Team Processes T. Mason, ORNL SNS Technical Overview P. Ferguson, ORNL
10:15-10:30
Break
10:30-12:00
Discussion of SNS Beam Line Requirements, Possible Experimental Program, and LoI/IDT Issues Discussion leader: P. Koehler, ORNL
ASAP 2002 PARTICIPANT LIST Name
Affiliation
Email
Chuck Alexander
Oak Ridge National Laboratory
[email protected] Dan Bardayan
Oak Ridge National Laboratory
[email protected] Jon Batchelder
Oak Ridge Associated Universities
[email protected] Jeff Blackmon
Oak Ridge National Laboratory
[email protected] Carl Brune
Ohio University
[email protected] Ken Carter
Oak Ridge Associated Universities
[email protected] Art Champagne
University of North Carolina
[email protected] Walter Furman
JINR, Dubna, Russia
[email protected] inr.ru
Kazuyoshi Furutaka
Japan Nuclear Cycle Development Inst.
[email protected] T. Vincent Cianciolo
Oak Ridge National Laboratory
[email protected] Aaron Couture
University of Notre Dame
[email protected] Yaron Danon
Rensselaer Polytechnic Institute
[email protected] Felix Difilippo
Oak Ridge National Laboratory
[email protected] Ernst Esch
Los Alamos National Laboratory
[email protected] Phillip D. Ferguson
Oak Ridge National Laboratory
[email protected] Enrique Gonzalez
CIEMAT
[email protected] Christopher Gould
North Carolina State University
[email protected] Geoffrey Greene
Los Alamos National Laboratory
[email protected] Colorado School of Mines
[email protected] Uwe Greife Vladimir Gudkov Gueorgui Gueorguiev Robert Haight Hideo Harada Jack Harvey Anna Hayes Michal Herman Paul R. Huffman Masayuki Igashira Cal Jordan Franz Kaeppeler Guinyun Kim Paul Koehler
University of South Carolina
[email protected] University of Florida
george@serverl .nuceng.ufi.edu
Los Alamos National Laboratory
[email protected] Japan Nuclear Cycle Development Inst.
[email protected] Oak Ridge National Laboratory
[email protected] Los Alamos National Laboratory
[email protected] International Atomic Energy Agency
herman@ndsalpha. iaea.org
Natl. Institute of Standards and Tech.
[email protected] Tokyo Institute of Technology
[email protected] Clemson University
[email protected] FZK, Karlsruhe
[email protected] Kyungpook National University
[email protected] Oak Ridge National Laboratory
[email protected] XII
Raymond Kozub
Tennessee Technological University
[email protected] Alexandre Laptev
Petersburg Nuclear Physics Institute
[email protected] Luiz Leal
Oak Ridge National Laboratory
[email protected] Eric Lynn
Los Alamos National Laboratory
eric,
[email protected] Thorn Mason
Oak Ridge National Laboratory
[email protected] Yasuhiro Masuda
High Energy Accelerator Research Org.
[email protected] Alberto Mengoni
ENEA-Applied Physics Division
[email protected] Gary Mitchell
North Carolina State University
[email protected] Paul Mueller
Oak Ridge National Laboratory
[email protected] Yasuki Nagai
Osaka University
[email protected] John Neal
Oak Ridge National Laboratory
[email protected] Pavel Oblozinsky
Brookhaven National Laboratory
[email protected] Toshiro Osaki
Tokyo Institute of Technology
[email protected] Peter Parker
Yale University
[email protected] Seppo Penttila
Los Alamos National Laboratory
[email protected] S. Raman
Oak Ridge National Laboratory
[email protected] Wolfgang Rapp
University of Karlsruhe
[email protected] John-Paul Renier
Oak Ridge National Laboratory
[email protected] James B. Roberto
Oak Ridge National Laboratory
[email protected] Bob Rundberg
Los Alamos National Laboratory
[email protected] Royce Sayer
Oak Ridge National Laboratory
[email protected] Kenneth Toth
Oak Ridge National Laboratory
[email protected]\
Ronald Townsend
Oak Ridge Associated Universities
to wnsenr@orau. gov
John Ullmann
Los Alamos National Laboratory
[email protected] Timothy Valentine
Oak Ridge National Laboratory
[email protected] Steve Wender
Los Alamos National Laboratory
[email protected] Jerry Wilhelmy
Los Alamos National Laboratory
[email protected] Glenn Young
Oak Ridge National Laboratory
[email protected] Vincent Yuan
Los Alamos National Laboratory
[email protected] xiii
CONTENTS Preface Schedule for the Workshop on ASAP Participant List Neutron Capture Nucleosynthesis: Astrophysical Processes and Laboratory Approaches F. Kappeler The Detector for Advanced Neutron Capture Experiments at LANSCE J.L. Ullmann, R.C. Haight, L. Hunt, E. Seabury, R.S. Rundberg, J.B. Wilhelmy, MM. Fowler, D.D. Strottman, F. Kaeppeler, R. Reifarth, M. Heil and E.P. Chanberlin Astrophysics Program at the CERN n_TOF Facility A. Mengoni
v vii xi
1
16
25
Recent Astrophysics Results from ORELA and Possible Future Experiments at ORELA and SNS P.E. Koehler
32
Sensitivity of Isotope Yields to Reaction Rates in the Alpha Rich Freezeout G.C. Jordan IV and B.S. Meyer
42
Neutron Reactions of Light Nuclei from Astrophysics & Nuclear Physics Interest Y. Nagai, T. Shima, A. Tomyo, H. Makii, K. Mishima, M. Segawa, M. Igashira and T. Ohsaki
52
XIV
Measurement of the n+p—>d+y Cross Section for Big Bang Nucleosynthesis with the Spallation Neutron Source at the Los Alamos Neutron Science Center E.-I. Eschm, J.M. O'Donnell, S.A. Wender, D. Bowman, G. Morgan and J. Matthews
58
Phonon Properties of Materials from Neutron Resonance Doppler Broadening Measurements J. Eric Lynn
65
Applied Nuclear Physics at Spallation Neutron Sources Pavel Oblozinsky
73
Applied Physics Measurements at the CERN n_TOF(1) Facility E. Gonzalez
83
Activities of the DOE Nuclear Criticality Safety Program (NCSP) at the Oak Ridge Electron Linear Accelerator (ORELA) Timothy E. Valentine, Luiz C. Leal and Klaus H. Guber Parameters for Nuclear Reaction Calculations - Needs for Improvements M. Herman Aluminum Data Measurements and Evaluation for Criticality Safety Applications L.C. Leal, K.H. Guber, R.R. Spencer, H. Derrien and R.Q. Wright Nuclear Physics Investigations at the Time-of-Flight Spectrometer GNEIS with Spallation Neutron Source O.A. Shcherbakov, A.B. Laptev andA.S. Vorobyev
97
107
115
123
XV
Measurement of Neutron Capture Cross Sections of Long-lived Fission Products H. Harada, S. Nakamura, K. Furutaka, T. Katoh, M.M.H. Miah, O. Shcherbakov, H. Yamana, T. Fujii and K. Kobayashi Temperature Measurements in Dynamically-loaded Systems Using Neutron Resonance Spectroscopy (NRS) atLANSCE V.W. Yuan
131
138
Radioactive Target Production at RIA J.C. Blackmon
146
Parity Violation inEpithermal Neutron Resonances G.E. Mitchell, J.D. Bowman, S.I. Penttila and E.I. Sharapov
155
New Experimental Capabilities for Parity Non-conservation and the Time Reversal Invariance Violation in Neutron Transmission S.I. Penttila T-Violating Three-fold Correlation in Neutron Transmission Y. Masuda
164
175
Violation of Fundamental Symmetries in Resonance Neutron Induced Fission A. Barabanov, W. Furman and A. Popov
184
Physics of the Fission Process and Parity Violation in Neutron Induced Reactions Vladimir Gudkov
194
XVI
Possibilities for Studies of Parity Violation at the SNS Using the Capture Gamma Reaction A. C. Hayes and Luca Zanini Time Reversal Tests with Epithermal Neutrons C.R. Gould An Experiment to Search for Parity-conserving Time Reversal Invariance Using Epithermal Neutrons from the Spallation Neutron Source P.R. Huffman Neutronic Characteristics of the Spallation Neutron Source P.D. Ferguson, E.B. lverson and F.X. Gallmeier Workshop Summary: Opportunities in Astrophysics, Symmetries, and Applied Physics at Spallation Neutron Sources Paul Koehler, Christoper Gould, Robert Haight and Timothy Valentine
202
209
217
225
233
Letter of Intent to the Spallation Neutron Source
234
Author Index
245
N E U T R O N C A P T U R E NUCLEOSYNTHESIS: ASTROPHYSICAL PROCESSES A N D LABORATORY APPROACHES F. K A P P E L E R Forschungszentrum
Karlsruhe,
Institut fur Kernphysik, Postfach Karlsruhe, Germany E-mail:
[email protected] 3640,
D-76021
Neutron reactions are responsible for the formation of the elements heavier than iron. The corresponding scenarios relate to helium burning in Red Giant stars (s process) and to supernova explosions (r and p process). The status of the relevant neutron data for the various scenarios are briefly summarized, followed by an outline of the essential experimental techniques. The direct impact of laboratory results on the interpretation of the observed abundance patterns and their role as crucial tests for astrophysical models are illustrated by representative examples. The very high flux at spallation neutron sources provide a unique possibility for investigating numerous difficult and hitherto inaccessible cases, in particular cross sections of the important radioactive nuclei. In combination with advanced detector concepts these facilities provide a promising step towards a quantitative picture of galactic chemical evolution.
1
Introduction
A first clue for the origin of the chemical elements was obtained in the 1930ies by the analysis of carbonaceous chondrites, a class of primitive meteorites, which preserved the original composition of the protosolar nebula *. At about the same time, nuclear burning was identified as the stellar energy source 2 3 4 ' ' . However, it was not before 1952, when Merrill 5 discovered Tc lines in the spectra of red giant stars - an unstable element with isotopic half-lives much shorter than the stellar evolution time - that stellar nucleosynthesis was accepted as the origin of the chemical elements. The various aspects of this new field of Nuclear Astrophysics, i.e. the elemental composition of astronomical objects, the standard abundance distribution, the nucleosynthesis mechanisms, and the related nuclear physics, were eventually combined in the fundamental and seminal paper by Burbidge, Burbidge, Fowler, and Hoyle 6 . A comprehensive summary of the 40 years of progress in nucleosynthesis since B 2 FH was published recently by Wallerstein et al. 7 . Reviews on more specific topics can be found elsewhere 8>9,10,11,12,13,14,15
1
2
100 MASS NUMBER
150
Figure 1. The isotopic abundance distribution in the solar system (from Ref. 1 7 ).
2 2.1
The Observed Abundances The Solar System
Any nucleosynthesis model must be checked against observations. Originally, the composition of the solar system was considered a standard which can be reliably derived by spectroscopy of the photosphere and by meteorite analyses 16,17 p r o m this distribution (Fig. 1) the signatures of the dominant scenarios can be inferred, starting with the very large primordial H and He abundances from the Big Bang. The abundances of the rare elements Li, Be, and B, which are difficult to produce because of the stability gaps at A=5 and 8, but are easily burnt in stars, were mostly formed by spallation reactions induced by galactic cosmic rays. Stellar nucleosynthesis starts with the ashes of He burning, 12 C and 1 6 0 , which are partly converted to 14 N by the CNO cycle in later stellar generations. In subsequent stages of stellar evolution, the light elements up to the mass 40 to 50 region are produced by charged particle reactions during C, Ne, and O burning 7 . The corresponding yields show a strong preference for the most stable nuclei built from a-particles. This part of the distribution is strongly influenced by the Coulomb barrier, resulting in an exponential decrease with increasing atomic number Z. Ultimately, Si burning leads to such high temperatures and densities that nuclear statistical equilibrium is reached. Under these conditions matter is transformed into the most stable
3 nuclei around Fe, giving rise to the dominant maximum at A = 5 6 . Due to the increasing Coulomb barriers the abundances of all heavier nuclei up to the actinides are essentially shaped by neutron capture nucleosynthesis, leading to a fairly flat distribution characterized by the pronounced r and s maxima. These twin peaks are the signatures of the slow (s) and rapid (r) neutron capture processes discussed below. 2.2
Galactic
Evolution
While the solar abundance distribution is characteristic for most stars it represents just the average enrichment of the Galaxy 4.55 Gyr ago. The chemical evolution prior to this point has become an intense field of investigation. Spectral analysis of stellar atmospheres has become an ever refined source of information. W i t h the astonshing sensitivity of ground and satellite based telescopes extremely faint a n d / o r metal poor stars can be observed from the UV to the far IR, providing an almost complete element p a t t e r n of these objects 1 8 ' 1 9 . Likewise, chemically peculiar stars, which witness ongoing sprocess nucleosynthesis in their deep interiors, or the expanding supernova ejecta can be accessed in great detail as well. For more t h a n three decades, direct spectroscopic observations have been complemented by analyses of circumstellar dust grains from AGB stars or supernovae, which survived the homogenization in the protosolar cloud and are preserved as minute inclusions in meteorites n > 1 4 . T h e isotopic composition of these presolar grains clearly exhibit enrichments, which can be a t t r i b u t e d t o particular nucleosynthetic scenarios such as the s or r process. T h e wealth of new and exciting information on the chemical evolution of the Galaxy calls for an expanded and improved nuclear physics d a t a base, which is indispensable for the quantitative interpretation of these observations, and hence for understanding the history of the universe. 3
N e u t r o n Capture Scenarios
W h e n the concept of neutron capture nucleosynthesis was first formulated 6 the s and r processes were already identified as the mechanisms responsible for the sharp maxima in the abundance distribution. These mechanisms are illustrated in Fig. 2, which shows the respective reaction p a t h s in t h e chart of nuclides. T h e s process being characterized by relatively low neutron densities implies neutron capture times much longer t h a n typical /?-decay half-lives. Therefore, the s-process reaction p a t h follows t h e stability valley as indicated
Seed for s-Process
s-Process Reaction Path
s-Branchings ( M N i 79Se 8 5 Kr,...)
Figure 2. An illustration of the neutron capture processes responsible for the formation of the nuclei between iron and the actinides. The observed abundance distribution in the inset shows characteristic twin peaks, which refer to the points where the s- and r-reaction paths encounter magic neutron numbers. Note that a p process has to be invoked for producing the proton rich nuclei that are not reached by neutron capture reactions. (For details see discussion in text.)
by the solid line in Fig. 2. The s abundances are determined by the respective (n,7) cross sections averaged over the stellar neutron spectrum, such that isotopes with small cross sections are building up large abundances. This holds for nuclei with closed neutron shells giving rise to the sharp s-process maxima in the abundance distribution at A=88, 140, and 208. This represents an illustrative example for the intimate correlation between the relevant nuclear properties and the resulting abundances, a phenomenon that can be used for probing the physical conditions during nucleosynthesis. The r-process counterparts of these maxima are caused by the effect of neutron shell closure on the /3-decay half-lives. Since the r process occurs in regions of extremely high neutron density (presumably during stellar explosions in supernovae) neutron captures are much faster than /3-decays. Therefore, the r-process path is driven off the stability valley until nuclei with neutron separation energies of « 2 MeV are reached. At these points, (n,7) and (7,n) reactions are in equilibrium, and the reaction flow has to wait for /3-decay
5
to the next higher element. Accordingly, the r abundances are proportional to the half-lives of these waiting point nuclei. This means that r-abundance peaks accumulate also at magic neutron numbers, but at significantly lower A compared to the related s-process maxima, resulting in the typical twin peaks of the abundance distribution. While the observed abundances are dominated by the s and r components, which both account for approximately 50% of the abundances in the mass region A>60, the rare proton-rich nuclei can not be produced by neutron capture reactions. This minor part of the abundance distribution had to be ascribed to the p process that is assumed to occur in explosively burning outer shells of supernovae 20>12. Among these processes, the s process is best accessible to laboratory experiments as well as to stellar models and astronomical observations 9 . Attempts to describe the r and p processes are hampered by the large uncertainties in the nuclear physics data far from stability 8 ' 2 1 , but also - and perhaps more severely - by the problems related to a detailed modelling of the stellar explosion 20>22>23. Obviously most isotopes received abundance contributions from the s and r processes. But as indicated in Fig. 2 there are neutron-rich stable isotopes (marked r) that are not reached by the s process because of their short-lived neighbors. Consequently, this species is of pure r process origin. In turn, these r-only nuclei terminate the /?-decay chains from the r-process region, making their stable isobars an ensemble of s-only isotopes. The existence of these two subgroups is of vital importance for nucleosynthesis, since their abundances represent important tests for stellar models.
4
The Case of the s Process
The diret impact of neutron reactions for the processes sketched before is illustrated at the example of the s process. The main nuclear physics input for s-process studies are the (n,7) cross sections averaged over the thermal neutron spectra characteristic for the stellar sites of the s process, typically between T 8 ~ 1 and 3 (in units of 108 K). This information is required for all nuclei along the reaction path from Fe to Bi. In addition, the /?-decay rates for unstable isotopes, which act as branching points in the reaction chain, have to be evaluated 2 4 .
6
4-1
Laboratory Neutron Sources
Neutrons in the energy range between 0.3 and 300 keV required for such measurements are produced in several ways: (i) At low-energy particle accelerators, nuclear reactions, such as 7 Li(p,n) 7 Be offer the possibility of tailoring the neutron spectrum exactly to the energy range of interest. This has the advantage of low backgrounds, allowing for comparably short neutron flight paths to compensate limitations in the neutron source strength 9 ' 25 . (ii) Much higher intensities can be achieved at linear accelerators via (7,n) reactions by bombarding heavy metal targets with electron beams of typically 50 MeV. The resulting spectrum contains all energies from thermal to near the initial electron energy. Since the astrophysically relevant energy range corresponds only to a small window in the entire spectrum, background conditions are more complicated and measurements need to be carried out at larger neutron flight paths. In turn, the longer flight paths are advantageous for high resolution measurements which are important in the resonance region. Refs. 26 27 ' are recent examples of astrophysical measurements at such facilities. (iii) Spallation reactions induced by energetic particle beams provide the most prolific sources of fast neutrons. An advanced spallation source suited for neutron time-of-flight (TOF) work is the LANSCE facility at Los Alamos, allowing for measurements on very small samples as well as on radioactive targets 28 - 29 . While the situation at LANSCE is characterized by a comparably short flight of 20 m and a time resolution of 250 ns (similar to what is planned at the SNS in Oak Ridge) the new n_TOF facility at CERN represents a complementary approach aiming at higher resolution (185 m flight path, 7 ns pulse width) 30 ' 31 - 32 . 4-2
Measurement of Neutron Capture Rates
The experimental methods for measuring (11,7) cross sections fall into two groups, TOF techniques and activations. In principle TOF techniques can be applied to all stable nuclei and require a pulsed neutron source for determining the neutron energy via the flight time between neutron production target and capture sample. Capture events are identified by the prompt 7-ray cascade in the product nucleus. The best signature for the identification of neutron capture events is the total energy of the emitted 7-cascade. To use this feature for accurate (11,7) cross section measurements requires a detector that operates as a calorimeter with good energy resolution such as the Karlsruhe in BaF 2 detector 33 . In the 7-spectrum of a perfect calorimeter, all capture events would fall in a line at the neutron binding energy (typically between 5 and 10 MeV), well separated
7
from backgrounds, which are inevitable in neutron experiments. In this way, an efficiency for capture events of 96 to 98% can be obtained, allowing for cross section uncertainties of ±1 %. Similar calorimeters are presently under construction at Los Alamos and at CERN. Activation in a quasi-stellar neutron spectrum provides a completely different approach for the determination of stellar (n,7) rates, but is restricted to those cases, where neutron capture produces an unstable nucleus. This method has superior sensitivity, allowing to use sub-/ig samples, and is highly selective, which means that isotopically enriched samples are not required. Quasi-stellar neutron spectra can be produced via the 7 Li(p,n) 7 Be 34 ' 35 by bombarding thick metallic lithium targets with protons of 1912 keV, only 31 keV above the reaction threshold. The resulting neutrons exhibit a continuous energy distribution very similar to a Maxwell-Boltzmann distribution for kT = 25 keV. The possibility to use minute samples makes the activation technique an attractive tool for investigating unstable nuclei of relevance for s-process branchings 36 . For example, a measurement of the 155 Eu cross section (ti/2=4-96 yr) could be performed with a sample of only 88 ng corresponding to 3.4xl0 1 4 atoms. This aspect is essential for minimizing the sample activity and, hence the radiation hazard, to a manageable level 3T . 4-3
Theoretical Calculations
In spite of the experimental progress, cross section calculations remain indispensable for determining the (n,7) rates of unstable nuclei with high specific 7-activity as well as the (possible) differences between the laboratory values and the actual stellar cross sections, which can be affected by thermally populated nuclear states with low excitation energies. Theoretical reaction rates are particularly important for explosive scenarios, where nuclei far from stability are involved and where experimental data are completely missing 38,39 Another essential issue are weak interaction rates under astrophysical conditions, both for He burning 24 and explosive scenarios 8 . 5 5.1
s Process Models The Canonical s Process
This phenomenological model 9 ' 40 was suggested by the empirical assumptions that temperature and neutron density are constant during the s-process and that a certain fraction G of the observed 56 Fe abundance was irradiated by an exponential distribution of neutron exposures. Then, an analytical expression
8
can be derived to calculate for all involved isotopes the characteristic s-process quantity, i.e. the product of the stellar cross section and the respective s abundance. Apart from the two parameters G and TQ, which are adjusted to fit the abundances of the s-only nuclei, the stellar (n,7) cross sections (a) are the only input data required for determining the overall abundance distribution. This approach includes also the treatment of the particular sprocess branchings. Given the very schematic nature of the classical approach, it was surprising to see that it provides an excellent description of the s-process abundances. Fig. 3 shows the calculated (cr)Ns values compared to the corresponding empirical products of the s-only nuclei (symbols) in the mass region between A = 56 and 209. The error bars of the empirical points reflect the uncertainties of the abundances and of the respective cross sections. One finds that equilibrium in the neutron capture flow was reached between magic neutron numbers, where the (cr)A?s-curve is almost constant. The small cross sections of the neutron magic nuclei around A~88, 140, and 208 act as bottlenecks for the capture flow, resulting in the distinct steps of the crN-curve. 5.2
Stellar Models
In terms of stellar sites, the main component can be attributed to helium shell burning in low mass stars, where neutron production and concordant s-processing occur in two steps by the 1 3 C(a,n) 1 6 0 reaction at relatively low temperatures around T g ~ l and by the 22 Ne(a,n) 25 Mg reaction at Tg~3 (see Refs. 10 ' 41 for details). The weak component can be ascribed to core He burning in massive stars 42 . 6
s-Process Branchings
Branchings in the reaction chain of the s process occur at unstable nuclei with sufficiently long half-lives that neutron capture can compete with /?decay. The resulting abundance pattern provide direct clues with respect to stellar neutron density, temperature, and pressure and allow to characterize the He-burning zones, where the s process actually takes place. Fig. 4 shows the s-process branchings at 147 Nd and 1 4 7 ' 1 4 8 Pm, which are defined by the sonly nuclei 148 Sm and 150 Sm. Since 148 Sm is partly bypassed by the reaction flow, its (<J)NS value will be smaller than that of 150 Sm, the ratio providing a measure for the combined strength of the branchings. Quantitative branching analyses require (i) the cross sections of the involved s-only nuclei with uncertainties of « 1%, and (ii) the corresponding
9
g 1000
z < Q
z n < z o p o PJ o ft! U
100
10 r
0.1
100
150
200
MASS NUMBER Figure 3. The characteristic product of cross section times s-process abundance plotted as a function of mass number. The solid line was obtained via the classical model, and the symbols denote the empirical products for the s-only nuclei. A complete representation of the empirical values requires at least two different mechanisms, the m a i n and weak (thick and thin solid lines, respectively). The important branchings of the neutron capture chain are indicated as well.
cross sections of the radioactive branch point isotopes with uncertainties of « 5 to 10%,. At present, the lack of experimental information on unstable isotopes is the limiting problem for reliable branching, because statistical model calculations are bound to uncertainties of 20% to 30%. Therefore, future experimental efforts have to be directed to determine cross sections of unstable nuclei 36 ' 43 ' 44 . Since the /3-decay rates of the branch points at A= 147-149 in Fig. 4 are not significantly affected by temperature 24 , these branchings are suited for determining the neutron density. A measurement with the 47r BaF2 detector yields fg = 0.870 ± 0.009 45 leading to an effective neutron density of (4.1 ± 0.6) • 108 cm" 3 4 6 . 7
N e u t r o n D a t a for A s t r o p h y s i c s : S t a t u s a n d N e e d s
The present status of (11,7) data for the s process is summarized in the compilation of Bao et al. 47 . In short, it can be stated that experimental techniques
10 p process
M7
H8
Sra
Sm
\ m
150
""Sm
148pm
Pm
Sm
k
\ 149
Pm
146Nd
\
\
\ —
—
147
Nd
148
Nd
150 N( J
5 process
r process Figure 4. The s-process reaction path in the N d / P m / S m region with the branchings at A=147, 148, and 149. Note that 1 4 8 Sm and 1 5 0 Sm are shielded against the r process. These two isotopes define the strength of the branching.
have reached a stage where the 1% accuracy level required for meaningful analyses of particular abundance patterns can been met, but that this has been achieved so far only for a minority of the relevant isotopes. Apart from the remaining key isotopes, also a large number of cross sections with uncertainties in excess of 10% await improvement. In contrast to the comparably stable situation of the s process, the complex explosive nucleosynthesis scenarios imply huge reaction networks including several thousand reactions. Since the majority of the reaction rates has to be obtained by statistical model calculations, experimental data for stable and as many unstable isotopes as possible are, therefore, required to test the necessary extrapolation to the unstable nuclei of relevance for these networks. In the following, the principal data needs for quantitative nucleosynthesis studies in the heavy element region are schematically summarized. For s-process analyses the requests concentrate on (n/y) measurements in the following areas: • The s-only nuclei are the key isotopes for all s-process investigations including the analyses of s-process branchings. These cross sections should, therefore, be determined with uncertainties of RJ 1%. So far, this has been reached only for half of the 33 s-only nuclei between T0 Ge and 2 0 4 Pb.
11 • Meaningful analyses of the characteristic signatures preserved in presolar grains require also accurate cross sections with uncertainties of ft* 1%. However, the present status is far from being adequate, particularly for the lighter elements oxygen, neon, magnesium, silicon, calcium, titanium, and zirconium. In this group, d a t a for about 70 isotopes have to be determined. • Nuclei at or near magic neutron numbers N = 5 0 , 82, and 126, which act as bottlenecks for the reaction flow in the main s-process region between Fe and Bi. For the majority of these d a t a the necessary uncertainties of < 3 % have not been reached. • T h e cross sections of a b u n d a n t light isotopes below Fe, which may constitute crucial neutron poisons for the s-process, need to be improved. Of particular importance are 1 6 0 , 1 8 0 , and 2 2 Ne. • Cases where Direct C a p t u r e (DC) contributes significantly to the astrophysical reaction rate are of particular interest, because this effect plays an important role in neutron-rich nuclei. Interesting examples are 2 0 8 P b , 14 C , 1 6 0 , 8 8 Sr, and 1 3 8 B a . • Nuclei, which still constitute white spots in the s-process chain or which exhibit very uncertain cross section, are found in the mass region below Fe, around A = 1 0 0 , and near the end of the s-process region. These gaps in the experimental d a t a should be determined at the 5% level. • Last, but not least, enhanced efforts should be directed to measurements on unstable nuclei. In addition to the activation technique, the very high neutron fluxes available at spallation neutron sources appear to be promising options for such studies 48 > 49 . From a list of possible measurements, priority should be given t o the important branch points 7 9 Se, 147 P m , 1 5 1 Sm, 1 8 3 H o , 1 7 0 T m , 1 7 1 T m , 1 7 9 Ta, 2 0 4 T 1 , and 2 0 5 P b . These cases are of immediate relevance to s-process analyses and should not present unexpected experimental problems. In addition to this list, partial cross sections leading to long-lived isomers are important for several branchings. A well-known example is the 10.8 yr isomer in 8 5 K r , which determines the s abundance of the neutron magic isotope 8 6 K r . While such studies were previously limited to activation measurements at very few energies, recently T O F measurements of partial cross section with a total absorption calorimeter have been reported 5 0 , yielding the energy-dependence of the partial cross sections, which is necessary to follow
12 the evolving abundance patterns during the complex He burning scenarios. Similarly, elastic and inelastic scattering data are definitely needed for establishing a quantitative set of stellar enhancement factors, in analogy to the treatment of the Os isotopes 51 ' 52 . Finally, the neutron producing (a,n) reactions on 13 C and 22 Ne exhibit large uncertainties and are not yet directly measured in the stellar energy range. The extrapolation of existing data to stellar energies requires a comprehensive R-matrix analysis, which combines all relevant reaction channels. Accordingly, measurements of the (n,a)-cross sections of 1 6 0 and 25 Mg would provide a significant contribution. Since neutron data for explosive nucleosynthesis are completely missing, any effort in this area provides a most useful support for testing and amending the theoretically calculated rates, which are used in the network calculations. In the r-process, neutron cross sections have a direct impact for scenarios with comparably low neutron densities as well as during freeze-out, where they contribute to smooth the pronounced odd-even effects predicted for the primary yields. In principle, several unstable nuclei on the neutron-rich side of the stability valley could be studied experimentally, e.g. 90 Sr, 123 ' 126 Sn, 182 Hf, 228 Ra, and a number of higher actinides. Such data would also improve the description of freeze-out effects in the p process, where neutrons are liberated by (7,11) reactions during the explosive burning of the Ne/O shell 53 . Furthermore, (n/y) cross sections of protonrich nuclei would be most useful in determining the inverse rates by detailed balance. Experimentally feasible cases include about 25 unstable isotopes between 53 Mn and 2 0 2 Pb. Apart from measurements on unstable nuclei, even data for stable isotopes are urgently required for improving the reaction rates used in explosive nucleosynthesis. In particular, complete data sets for long isotope chains are important for this purpose. This means that stellar (11,7) cross sections should be determined also for all r- and p-only nuclei. At this point it should be mentioned that there are only very few experimental (p,7) and (0,7) cross sections at astrophysically relevant energies in the mass region of the p process, even for stable isotopes. Especially, the (a/y) and (a,p) rates lead to significant uncertainties in the final p-process abundances. Since direct measurements are difficult and time-consuming, the calculated rates are poorly constrained by experimental data. In particular, the a-nucleus potential used in statistical model calculations seems to be rather uncertain. A series of (n,a) measurements at astrophysically meaningful energies could help to solve this persisting problem.
13 8
Summary
Neutron capture nucleosynthesis of the elements heavier than iron operates during the He burning stages of stellar evolution and (presumably) in the final supernova explosion of massive stars. The various scenarios are identified by their typical abundance distributions as well as by increasingly detailed astronomical observations. This information combined with quantitative model calculations allow to probe stellar and galactic evolution. In this context, the strong demand for reliable nuclear physics data represents a continuing challenge, requiring the implementation of more powerful neutron sources and new experimental techniques. References 1. V.M. Goldschmidt, Norske Vidensk. Akad. Skr., Mat.-Naturv. Kl. IV, 1937. 2. H.A. Bethe and C. Critchfield, Phys. Rev. 54, 248 (1938). 3. C.F. von Weizsacker, Physik. Zeitschrift 39, 639 (1938). 4. H.A. Bethe, Phys. Rev. 55, 103 (1939). 5. P.W. Merrill, Science 115, 484 (1952). 6. E.M. Burbidge, G.R. Burbidge, W.A. Fowler, and F. Hoyle, Rev. Mod. Phys. 29, 547 (1957). 7. G. Wallerstein et al., Rev. Mod. Phys. 69, 995 (1997). 8. F. Kappeler, F.-K.. Thielemann, and M. Wiescher, Ann. Rev. Nucl. Part. Sci. 48, 175 (1998). 9. F. Kappeler, Prog. Nucl. Part. Phys. 43, 419 (1999). 10. M. Busso, R. Gallino, and G.J. Wasserburg, Ann. Rev. Astron. Astrophys. 37, 239 (1999). 11. T.J. Bernatowitz and E. Zinner, eds., Astrophysical Implications of the Laboratory Study of Presolar Material, (AIP, New York, 1997). 12. D.L. Lambert, Astron. Astrophys. Rev. 3, 201 (1992). 13. J.J. Cowan, F.-K. Thielemann, and J.W. Truran, Phys. Rep. 208, 267 (1991). 14. E. Zinner, Ann. Rev. Earth Planet. Sci. 26, 147 (1998). 15. H. Schatz et al., Physics Reports 294, 167 (1998). 16. E. Anders and N. Grevesse, Geochim. Cosmochim. Acta 53, 197 (1989). 17. H. Palme and H. Beer, in Landolt-Bornstein New Series, Group VI, Vol. VI/3a, Astronomy and Astrophysics, ed. O. Madelung (Springer, Berlin, 1993), page 196. 18. C. Sneden, J.J. Cowan, D.L. Burris, and J.W. Truran, Ap. J. 496, 235
14
(1998). 19. C. Sneden, Nature 409, 673 (2001). 20. M. Rayet et al., Astron. Astrophys. 298, 517 (1995). 21. F.-K. Thielemann et aJ., in Nuclear and Particle Astrophysics, eds. J.G. Hirsch and D. Page (Cambridge University Press, Cambridge, 1998), page 27. 22. W. Hillebrandt and P. Hoflich, Rep. Prog. Phys. 52, 1421 (1989). 23. E. Miiller, in Nuclear Astrophysics, eds. M. Buballa, W. Norenberg, A. Wambach, and J. Wirzba (GSI, Darmstadt, 1998), page 153. 24. K. Takahashi and K. Yokoi, Atomic Data Nucl. Data Tables 36, 375 (1987). 25. Y. Nagai et al, Ap. J. 381, 444 (1991). 26. P.E. Koehler et al, Phys. Rev. C 54, 1463 (1996). 27. H. Beer, F. Corvi, and P. Mutti, Ap. J. 474, 843 (1997). 28. P.E. Koehler and F. Kappeler, in Nuclear Data for Science and Technology, ed. J.K. Dickens (ANS, La Grange Park, Illinois, 1994), p. 179. 29. R.S. Rundberg et al, Technical report, Los Alamos National Laboratory, Los Alamos, USA (1999). 30. C. Rubbia, et al. Technical report, CERN, Geneva, Switzerland (1998). 31. S. Andriamonje et al, Report CERN/INTC 2000-004, CERN, Geneva, Switzerland (2000). 32. U. Abbondanno et al, Report CERN/INTC 2001-021, CERN, Geneva, Switzerland (2001). 33. K. Wisshak et al, Nucl. Instr. Meth. A 292, 595 (1990). 34. H. Beer and F. Kappeler, Phys. Rev. C 2 1 , 534 (1980. 35. W. Ratynski and F. Kappeler, Phys. Rev. C37, 595 (1988). 36. F. Kappeler, M. Wiescher, and P.E. Koehler, in The Production and Use of Intense Radioactive Beams at the Isospin Laboratory, ed. J.D. Garrett (Joint Institute for Heavy Ion Research, Oak Ridge, 1992), p.163. 37. S. Jaag and F. Kappeler, Phys. Rev. C 5 1 , 3465 (1995). 38. T. Rauscher and F.-K. Thielemann, Atomic Data Nucl. Data Tables 75, 1 (2000). 39. J. Goriely, Long Term Needs for Nuclear Data Development, ed. M. Herman (International Atomic Energy Agency, Vienna, 2001), p.83. 40. P.A. Seeger, W.A. Fowler, and D.D. Clayton, Ap. J. Suppl. 97, 121 (1965). 41. R. Gallino et al, Ap. J. 497, 388 (1998). 42. C M . Raiteri et al, Ap. J. 419, 207 (1993). 43. S. Jaag, F. Kappeler, and P.E. Koehler, Nucl. Phys. A 621, 247c (1997). 44. J.B. Wilhelmy et al, in International Conference on Nuclear Data for
15
Science and Technology, Tsukuba, Japan, October 7 to 12 (2001). 45. K. Wisshak et al., Phys. Rev. C 4 8 , 1401 (1993). 46. F. Kappeler, K.A. Toukan, M. Schumann, and A. Mengoni, Phys. Rev. C 5 3 , 1397(1996). 47. Z.Y. Bao et al., Atomic Data Nucl. Data Tables 76, 70 (2000). 48. A.F. Michaudon and S.A. Wender, Report LA-UR-90-4355, Los Alamos National Laboratory, Los Alamos, USA (1990). 49. S. Abramovich et al, Report CERN/SPSC 99-8; SPSC/P 310, CERN, Geneva, Switzerland (1999). 50. K. Wisshak, F. Voss, C. Arlandini, F. Kappeler, and L. Kazakov, Phys. Rev. C 6 1 , 065801 (2000). 51. R.R. Winters, R.F. Carlton, J.A. Harvey, and N.W. Hill, Phys. Rev. C 34, 840 (1986). 52. R.R. Winters, R.L. Macklin, and R.L. Hershberger, Astron. Astrophys. 171, 9 (1987). 53. M. Rayet, N. Prantzos, and M. Arnould, Astron. Astrophys. 227, 271 (1990).
THE DETECTOR FOR ADVANCED NEUTRON CAPTURE EXPERIMENTS AT LANSCE J.L. ULLMANN, R.C. HAIGHT, L. HUNT, E. SEABURY, R.S. RUNDBERG, J.B. WILHELMY, M.M. FOWLER. D.D. STROTTMAN Los Alamos National Laboratory, Los Alamos NM87544,
USA
F. KAEPPELER, R. RE1FARTH, M. HEIL Forschungszentrum
Karlsruhe, Karlsruhe,
Germany
E.P. CHAMBERLIN Chamberlin Associates, Los Alamos, NM87544
USA
The Detector for Advanced Neutron Capture Experiments (DANCE) is a 159-element 4tt barium fluoride array designed to study neutron capture on small quantities of radioactive material. It is being built on a 20m neutron flight path which views the "upper tier" water moderator at the Manuel J. Lujan Jr. Neutron Scattering Center at the Los Alamos Neutron Science Center. Monte Carlo calculations have suggested ways to minimize backgrounds due to neutron scattering events. Preliminary data on an 8 mg sample of 234U and a 0.5 mg sample of l!l Sm have been taken using dUf, detectors.
1
Introduction
The precise measurement of neutron capture cross sections in the electron-volt and kilo-electron-volt regions on radioactive isotope targets is needed for several applications, including stockpile stewardship and nuclear astrophysics. Capture cross sections are difficult to calculate accurately because they depend on fine details of nuclear structure and level densities at 5 to 10 MeV in excitation. A recent compilation of "Maxwell-averaged" capture cross sections using the "NonSmoker" statistical model code [1] showed that the calculated cross sections differed from measured cross sections by +/- a factor of 2 for masses between 25 and 210. While there are capture measurements on most stable nuclides, there are very few measurements on unstable nuclides One of the main applications for capture cross sections is in understanding sprocess nucleosynthesis [2]. The s process occurs by sequential neutron capture along the line of beta stability. When the capture sequence produces an unstable nuclide, the process can branch. The competition between beta decay and neutron capture at the branch nuclide depends on it's stellar beta-decay half life, the stellar neutron density, and the capture cross section. Combined with observed nuclear abundances, the capture cross section and beta-decay rate can be used to infer the temperature and neutron density at the stellar s-process site. The interesting energy range (Fig. 1) for these cross sections is over a Maxwell distribution centered at 25
16
17 keV, for neutrons from the 22Ne(a,n) reaction, and at about 10 keV, for the 13C(oc,n) reaction. Maxwell Distribution 12.0 10.0 8.0
30keV
,-\
- - - -10keV 1 /
6.0 4.0
!/
2.0 0.0
, ' • • -
•.• —
100
50
200
150
Energy (keV)
Figure 1. Maxwell energy distribution
A second application is in Stewardship Science, where capture cross sections on unstable nuclides are needed to interpret "rad-chem" diagnostics. Stable isotopes were placed in past nuclear explosion tests as diagnostic aids which integrated the neutron exposure over the entire explosion history. The high neutron densities caused multiple reactions creating many nuclei (See Fig 2.) Neutron cross sections are needed for neutron energies up to about 1 MeV.
(n,2n) !68Tm
4
4
"'"Tm •
93 d
(n,2n)
(11,7)
• Stable (11,7)
(n,2n) ,70
4
(n,2n) 171
Tm
4
Tm
• 129 d
(11,7) 1.9 y
172
• OVy)
2J
Tm d
Figure 2. An example of a reaction sequence where cross sections needed for "rad-chem" diagnostics
Three separate experimental components are needed to make these measurements: An intense neutron source, facilities to fabricate and handle radioactive targets, and an efficient, well characterized gamma detector. The source and detector are discussed further below; target preparation is the subject of a separate talk at this conference [3].
18 2
Neutron Source
The DANCE is being constructed on Flight Path 14 at the Manuel J. Lujan, Jr. Neutron Scattering Center at LANSCE. Flight Path 14 views the upper-tier "backscatter" water moderator, from which there are significantly fewer neutrons above 1 MeV. The sample is positioned 20 m from the moderator and the beam stop is at 30 m. A box for remotely inserting various filters is at 7 m. The bulk shielding surrounding the spallation target is 4.72 m in radius. Four discrete collimators are located outside the bulk shield, each constructed of about a meter of copper, brass, and 5% borated polyethylene. The collimation was designed to produce a uniform 1 cm diameter beam spot at the target location with minimal penumbra outside the central beam. The last collimator has a r = 0.3 cm opening with the downstream edge at 18.88 m. This tight collimation limits the beam intensity, which depends on the area of the moderator that is viewed. To reduce gamma backgrounds, the beam pipes and flanges were constructed of aluminum and the use of iron in components outside of the bulk shield was quite to a minimum. The flight path shielding was designed to limit the total gamma plus neutron dose to less than 1.0 mrem/hr along the first 10 m of the flight path and 0.5 mrem/hr beyond 10 m. Magnetite-loaded concrete blocks were used to shield the beam pipes and target area. Only polyethylene and borated polyethylene were used for shielding the roof of the target area. The Monte Carlo shielding calculations predicted significant high-energy gamma-ray production from neutron capture in the polyethylene and concrete, and the interior walls of the target area were faced with 2.54 cm thick 5% borated polyethylene which yields lower energy gamma rays following neutron absorption.. During the 2001 run cycle, the spallation source was operated at 55 uA. The neutron flux on FP14 was measured with three different techniques. First, a standard 3He tube was used [4]. Next, a fission chamber with 286 u,g/cm2 of 235U was used. Lastly, a neutron monitor consisting of a 546 u,g/cm2, 1 cm diameter deposit of 6LiF on an Al foil backing and viewed by a Si surface barrier detector, was employed to detect neutrons via the 6Li(n,at) reaction. The measured flux is shown in Fig. 3. The three measurements were each made at a different location, and were converted to moderator surface current for comparison. The three measurements were not consistent, and were considerably below the anticipated flux. This discrepancy is not understood and is still under study. It may possibly be due to misalignment. The 3He and 6Li measurements can be fit to a surface current of the form / = A/E with E in eV and A = 1.10 x 10 N/cm2/sr/eV/sec at 55 uA. At 20 m, this yields a flux of O = (3.70 x 103 N/cm2/eV/sec)/E.
19 FP-14 Flux 1.E+10 •
1.E+09
u
1.E+08 1.E+07
"
\
1 * n — •
r.
:
_
!
He-3 Geo Li6Sum265 55 uA U235Run128
-
: ^^D»
1.E+06
: 0111
1.E+05 1.E+04
«£#*> %
:
i%
•
;
1.E+03 1.E-01
in
i 1.E+00
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05 1.E+06
Energy (eV) Figure 3. Measured surface currents compared to expected value.
3
Preliminary Data with C6D6 Scintillators
Preliminary data was taken using two C6D6 detectors, each 12.5 cm diameter by 7.5 cm thick, mounted adjacent to a 5 cm diameter target holder made from 0.16 cm Al tubing. The targets studied were Au, ""•"°'^u powder, each contained in a small quartz tube, l50,l52Sm powder in quartz, and 151Sm deposited on a thin Ti foil. These data are still being analyzed. Figure 4 shows the time-of-flight spectrum from 234U target, which was 8 mg of 99.8% pure material prepared by the Chemistry Division. The time scale is 100 ns/chan, and the large resonance near channel 6500 is at 5.16 eV. Figure 5 shows the time-of-flight spectrum for the 0.5 mg 'Sm target. The isotopic composition of this target is 71% 151Sm, 13% 150Sm, 7% 152Sm, 6% l47Sm, and 1% ,49Sm. The large resonance near channel 14000 is the 1.088 eV resonance in l5, Sm. Note that 1 keV corresponds to channel 500. The gray line is the spectrum from a Ti foil blank normalized to the same neutron fluence. 4
DANCE Design
The advanced gamma detector is being built to provide an increased and betterdetermined detection efficiency and also to provide better background rejection. Backgrounds are due to capture in material surrounding the target, but also due to
20
Figure 4. Time of flight spectrum for 234U(n,y). The horizontal scale is 100 ns/channel.
I Sum GEOATI
1000
etn aoa 400
200-
Qt—• • • I 2000
4000
8000
6000
10000
12O0Q 14Q0Q
19000
Figure 5. Time of flight spectrum for lslSm(n,Y). The contribution from the blank Ti foil is also shown.
capture reactions in the detector from neutrons scattered in the target. The neutron scattering cross section is greater than the capture cross section for many materials in the kilovolt region. Three criteria were established for the detector: •
Calorimetric to measure the total gamma ray energy emitted
21
• •
Insensitive to neutrons Segmented and fast to handle radioactive targets (one Curie is 37 decays/ns)
Extensive Monte Carlo calculations were made using GEANT-3 to design the detector [5][6]. Of the commonly available scintillator materials, BaF2 was chosen because it had the smallest neutron capture cross section and a very fast (0.6 ns) component of light. It suffers from an internal alpha particle background due to decay of Ra and its decay chain products, but pulse-shape discrimination can be used if needed. The crystal array should completely cover 4% sr with no gaps, and each crystal should have equal area and volume. The analysis of Habs [7] showed that 162 elements with 4 different shapes will meet this requirement. The array is shown schematically in Fig. 6. The inner radius is 17 cm and each crystal is 15 cm deep, 734 cm3 in volume, and has an inside area of 22.9 cm2. Each crystal is coupled to an Electron Tubes 9921 7.5 cm phototube with quartz window using Dow-Corning Sylgaard 184 for maximum UV transmission.
Figure 6. Schematic design of the DANCE crystal ball. Four different shaped crystals are needed, indicated by different shadings in the figure.
22
The Monte Carlo calculations indicated that scattered neutrons could still contribute a significant background, especially in the 10 to 100 keV range of interest. Several additional methods will be employed to decrease this background. First, the measured reaction Q value can be used in many cases to discriminate between true capture events and events induced by scattered neutrons [5,6]. Next, the Monte Carlo calculations predict that an 8 cm thick 6LiH shell inside the array will reduce the scattered neutrons to 42% of the unattenuated number, in the 10 to 100 keV energy range[6]. However, because of space limitations we will use a 6 cm thick 6LiH sphere surrounding the target. Finally, a "hit pattern" analysis of the event will also be tried. This is illustrated in Fig 7. which shows the calculated number of crystal clusters for events due to neutron scatter and true capture in the target [6]. A cluster is a set of adjacent crystals that give a signal above threshold. True capture events produce several clusters, each due to an individual gamma ray from the decay cascade, while events due to neutron capture in the BaF2 tend to be grouped primarily into one cluster. Neutron Energy = 10 to 100 keV
- - - - Scattered
3500
Capture
3000 2500 2000 1500
1
1000 500 n 0
1
2
3
4
5
6
7
Cluster Multiplicity
Figure 7. Cluster multiplicity (see text) for events due to true capture in the target and capture of neutrons scattered into the BaF2 array.
The data acquisition system will consist of two Acqiris DC-265 8-bit waveform digitizers on the anode signal of each phototube. The digitizers sample at 500 MHz. Each digitizer has a different voltage gain to match the dynamic range of the fast and slow components of the scintillation light. The slow component contains about 85% of the light, and has a decay time of 630 ns. The fast component, while providing only 15% of the light, has a decay time of 0.6 ns and is the dominant feature of the waveform. The front end software will initially return for each event only a time and pulse height number, which the analyzer routine will histogram and log.
23
Figure 8 shows a typical pulse from a completed crystal assembly, acquired with an Acqiris DC-270 1 GHz digitizer. Fig 9. shows a 60Co spectrum obtained by simply adding the counts in a waveform from 10 ns before the trigger time to 1790 ns after. The resolution of the 1173 keV peak is 9.0% fwhm. Waveform
^
^
t n tries Mean RMS
^
D~ 508.1 57S.6
'100 80 60 40 20
-500
500
1000
1500
2000 Channel
Figure 8. Typical waveform from a completed BaF2 crystal assembly using a DC-720 1 GHz digitizer. The fast and slow components are easily recognized. Fast + Slow |
sum Entries 329807 Mean 2.173e+04 RMS 79SG
1600 1400
fl
1200
I
~r
/V
800
r
/ 1/L^
600
r
1000
400
rl
I
VI
-
\ r\A
200
M 10000
,
A 20000
,
i 30000
^""V*. 40000
50000
Figure 9.60Co spectrum obtained by simply summing waveforms from a 1GHz digitizer. The peaks from the alpha-decay background in the crystal are readily seen.
24
5
Summary and Future Plans
The DANCE array is under construction, and all crystals are scheduled to be delivered by Sept, 2002. A multi-year program of targets to be measured for stockpile stewardship and s-process branch point studies has been mapped out. Initially, targets that can be chemically purified have been chosen, and 146Nd, 154Sm, and 170Er will be irradiated at the ILL in spring, 2002, to produce 4 to 10 mg of 17 Pm, 155Eu, and m Tm. It is expected that these isotopes will be studied using the DANCE array in 2002 , along with a new measurement of lsl Sm. References 1. T. Rauscher and F.-K. Thieleman. in Atomic and Nuclear Astrophysics, A. Mezzacappa, ed, IOP, Bristol, (1998), p. 519. 2. F. Kaeppeler. Prog. Part. Nucl. Phys. 43, 419-483 (1999). 3. R.S. Rundberg et al., these proceedings. 4. L. Daemen, private communication. 5. M. Heil, et al., Nucl. Instr. Meth. A459, 229-246 (2001). M. Heil, et al, Los Alamos National Laboratory report LA-UR-99-4046, (1999). 6. R. Reifarth et al., Los Alamos National Laboratory Report LA-UR-014185,(2002). 7. D. Habs, F.S. Stephens, and R.M. Diamond, Lawrence Berkeley Laboratory Report LBL-8945, (1979).
A S T R O P H Y S I C S P R O G R A M AT T H E C E R N n_TOF FACILITY A. MENGONI CERN, EP Division, CH-1211 Genava 23, Switzerland e-mail: alberto.mengoniQcem.ch and The n-TOF Collaboration The set of measurements of neutron capture cross sections for nuclear astrophysics at the CERN neutron time-of-flight facility, n_TOF, is presented. A brief description of each of the planned measurements is given. 1
Introduction
The CERN neutron time-of-flight facility1, n_TOF, is a spallation neutron source based on the high-intensity 20 GeV proton beam of the CERN PS accelerator complex. At n_TOF, a solid 80x80x40 cm 3 lead target coupled to a 5cm-thick water moderator generates a white neutron spectrum of extremely high intensity. The proton beam intensity is typically 7 x 1012 protons/pulse, with a pulse-width of 7 ns. Each proton generates 360 neutrons in the spallation process, a fraction of which can leave the target-moderator assembly to enter the 200m neutron flight-path. The first experimental area (EAR-1) is located at 187.5 meters downstream from the lead target. The neutron energies effectively considered for cross section measurements is in the range 1 eV up to 250 MeV, although neutrons from thermal up to the GeV region are generated by the target-moderator assembly. The neutron flux in EAR-1 can reach 4 x 105 neutrons/cm 2 /pulse. The long flight-path, combined with the time-synchronization characteristics of the lead slowingdown process in the spallation target results in an excellent energy resolution, which reaches 3 x 10~ 4 at En « 3 eV and 1.5 x 10" 3 at En « 30 keV. A peculiar feature of n_TOF is the very low repetition frequency of the PS beam. On average, 1 pulse/2.4 s is extracted and delivered to the n_TOF experimental area. A very low ambient background has been recently achieved in EAR-1 after the implementation of meter-thick concrete and iron walls in the tof tunnel. The prompt flash in EAR-1 due to minimum ionizing particles has been drastically reduced by this shielding. At the same time, the fluence of negative muons, responsible for neutron capture background generation in the experimental area, has been largely reduced 2 , such that sample-induced
25
26 backgrounds are the dominant component in E A R - 1 , thus defining optimal conditions for capture measurements. n . T O F has been constructed with t h e basic motivations of measuring: • cross section relevant to nuclear waste transmutation and related nuclear technologies, • neutron cross sections relevant for nuclear astrophysics, and • cross sections for nuclear structure studies. Here we will describe the experimental program for the measurements of interest in nuclear astrophysics. 2
P r i o r i t y m e a s u r e m e n t s for n u c l e a r a s t r o p h y s i c s
n_TOF delivers neutrons in a wide energy range (from eV t o MeV), covering entirely the maxwellian velocity distributions corresponding to a wide range of stellar temperatures, typical for neutron capture nucleosynthesis. In particular, the extended range of t e m p e r a t u r e s recently considered in stellar models, kT « 5 keV up to kT « 30 keV, can be all explored in neutron c a p t u r e measurements for s-process nucleosynthesis. This, combined with the high neutron flux, led to consider t h e possibility of measuring of small-mass samples, either rare o r / a n d radioactive. In addition to t h e high-flux, another favorable condition for measurements of capture cross sections of radioactive samples has been taken into consideration: the low repetition frequency of t h e neutron pulses. To give a quantitative estimate of this condition, one can compare the repetition frequency, / , and flight-path, L, at which two facilities, n_TOF and GELINA, have approximately the same flux at 25 keV (respectively 187m for n J T O F and 50m for GELINA). One finds /i_ /2
Lx X
L2~
0.28 s " 1 800s-
1
187.5 m
1
50m
762'
Obviously, factors of this magnitude will introduce a considerable advantage in t h e performance for c a p t u r e measurements on radioactive samples, often required in t h e studies of s-process branchings. Another characteristics which represents an advantage in capture measurements at neutron spallation sources such as n_TOF over electron-based installations, is t h e large suppression of t h e 7-fiash due to a reduced 7-ray production in the spallation process.
27 Table 1. Priority measurements for nuclear astrophysics at n_TOF.
Reaction
Notes
151
s-process branching(s) at A w 150. tive (t 1 / 2 = 93 yr)
Sm(n, 7 ) 1 5 2 Sm
186 187 188
'
-
Os(n, 7 )
151
Sm is radioac-
nuclear cosmochronology (Re/Os clock)
24 25 26
isotopic abundance ratios in interstellar grains. Relative importance of the 22 Ne(a, n) 25 Mg as neutron source for the s-process. Light nuclei, small cross sections
204 206 208
termination of the s-process. Small cr7/o"e;
- < Mg(n, 7 )
209
- - Pb(n,7), Bi(n, 7 )
Given these preliminary considerations, the priority list shown in Table 1 for the first period of measurements has been defined. We will give here a brief description of the motivations and of the technique to be used in each of the neutron capture cross section measurements listed in Table 1. 2.1
151
Sm(n,7) 1 5 2 Sm
The s-process reaction flow in the Sm-Eu-Gd region exhibit several branching points. These are due to a sufficiently long /3-decay half-life of nuclei which can compete with the long half-life for neutron capture in s-process-like stellar conditions. One specific and crucial example is 151 Sm. Its neutron capture cross section is expected to be quite large, but experimental values in the keV energy region are completely missing. We plan to measure the 151 Sm(n, 7) 152 Sm cross section at n_TOF with a 180 mg sample provided by Oak Ridge National Laboratory. The experimental setup for this measurement includes a set of two CeD 6 -based liquid scintillator detectors, specifically designed with carbon-fiber container, to reduce their sensitivity to scattered neutrons. The high instantaneous n_TOF neutron flux requires the use of a specifically designed data acquisition system, entirely based on high-frequency flash-ADC, operating at 1 GHz sampling rate with 16 MB buffer memory per channel.
28
Given the expected high o-y/aei for the neutron induced reactions on Sm, favorable background conditions are expected for this capture measurement. The data taking is expected to start in June 2002. 151
2.2
186 187 188
'
'
Os(n ) 7 )
The enhancement of the observed 187 Os solar abundance, as compared with what can be expected from s-process synthesis in the A « 185 mass region, is to be attributed to the 187Re(/3~) —• 187 Os decay. 187 Re has a half-life of 42.3 ± 1.3 x 109 yr and it is an r-only isotope. On the other hand, 1 8 6 0s and 187 Os are shielded against r-process production by 1 8 6 W and 187 Re: they are s-only isotopes. A simple calculation based on the canonical s-process, shows immediately the necessity for the radiogenic production of 187 Os. Prom the s-process condition Cyi[A] = const, it follows that a 186 [ 186 Os] = <x187[187Os] From solar-system abundances, rl87 Os] [186 Os]
0.79.
On the other hand, p
= ( £i56 I CT
187/
=
0.48
exp.
using the experimental capture cross sections of Winters et al.3. From the amount of radiogenic 1 8 7 0s, it is possible to derive, within this simple picture, the galactic age. It is also possible to show that the uncertainty in the capture cross section ratios propagates into an age (to) uncertainty of the order of ARa « 10%
==>
Ai 0 « 3 Gyr
The present estimated uncertainty in the capture cross section ratios, Ra, is of the order of 5%. However, the need for maxwellian averages at low kT (kT ss 8 keV) calls for an extension of the measured data to a wider neutron energy range, in particular on the low energy side (see Figure 1). Clarification of other nuclear physics aspects related to this clock will also benefit from measurements of capture cross sections of the other Os isotopes. A comprehensive analysis of the s-processing based on low-mass AGB stars, together with up-to-date galactic chemical evolution models are part of the n_TOF activities on this subject.
29
Os(n,y)
Os
Cross Section and Maxwellian Average (MACS) I
I I I
" " 1
'
'•
1
i
i i 11111
i
i
— HF calculation • Exp.
=
10
^X5£-
1 --
0.1 0.0001 1 —
-— —
O 0.6 0.4
-
- ^*\»
1
1
1
I 0.01
1
0.001 1
1
1
i
r
0.1
1 1 1 1 1 1
kT = 30KeV kT - 8 KeV
-
-
-
-
-
-
0.2
0 0.0001 Wed Oct 16 17:50:43 2000
0.001
0.01
0.1
Neutron energy [MeV]
Figure 1. The neutron capture cross section of 1 8 6 Os (upper panel). A comparison between the experimental values of Winters et al. 3 and a calculation based on Huser-Feshbach statistical model theory, is shown. In the lower panel, the cumulative maxwellian average, vs neutron energy, assumed as upper-limit in the averaging integral, is shown for two values of the stellar temperature kT.
2.3
24 25 26
> - Mg(n, 7 )
The absence of a 25 Mg anomaly in interstellar grains and the role of the 22 Ne(a, n) 25 Mg in AGB stars represent the basic motivations for the 24 25,26 ' Mg(n, 7) measurements. In addition, nuclear structure properties as well as reaction mechanisms aspects, such as the contribution of a direct capture process in the keV neutron energy region, can be derived from these measurements. From the experimental point of view, this set of measurements is challenging, as for these low-mass targets, the ay/aei is very unfavorable. Here, the problem is represented by the scattered neutrons. These generate additional background signals in the experimental area. To partly cure this problem, a
30
set of specifically designed, low neutron-sensitive CeD6 detectors have been constructed and will be used in these measurement.
2.4
204
>206-208Pb(n,7)
and209Bi(n,7)
T h e termination of the s-process at 2 0 9 B i , due t o the a-decay of 2 1 0 > 2 1 1 p o a n d of 2 1 1 B i , drives t h e basic interest in t h e capture cross sections for 2 0 9 Bi a n d for t h e P b isotopes. T h e capture cross section for 2 0 4 P b , on t h e other h a n d , is essential for the analysis of the s-process branching at 2 0 4 T 1 . Overall, t h e c a p t u r e cross sections for P b isotopes are essential for t h e understanding of the nucleosynthesis of P b , in particular for the evaluation of its non-radiogenic component. From t h e experimental point of view, this set of measurements suffers from the same drawback mentioned in t h e description of the Mg isotopes: t h e extremely small a1/aei rates. It is planned t o detect are the strengths of some of t h e keV-range neutron resonances of P b and Bi. In addition to the nuclear astrophysics motivation, it is perhaps useful t o mention here t h a t the capture cross sections of P b and Bi, in a wide energy region, are requested for the development of accelerator-driven systems (ADS) for nuclear waste transmutation. Some of these nuclear devices are designed to use a Pb-Bi eutectic as neutron spallation source as well as coolant. T h e experimental determination of capture strengths for P b and Bi are clearly basic requirements for the ADS design.
3
Conclusion and perspectives
W i t h the description of the cross section measurement campaign of activities at C E R N n_TOF, we have shown how t h e characteristics of a neutron spallation source can be favorably used for neutron cross section measurements of interest in nuclear astrophysics. As a concluding remark, we mention here t h a t the construction of a 47r, high-efficiency 7-ray detector, based on B a F 2 crystals is under way at n _ T O F . This detector will allow for a considerably larger number of measurements, on even smaller samples. This represents a challenge of considerable importance with respect t o the need for more accurate measurements of neutron c a p t u r e cross sections for t h e quantitative understanding of stellar nucleosynthesis a n d advanced stellar modelling.
31 Acknowledgments This work is the result of the activity of the n_TOF Collaboration. This Collaboration is composed of approximately 130 scientists belonging to about 30 different research Institutions from Europe and the USA. The interest and enthusiasm of all these colleagues for the project and for the activities related to nuclear astrophysics is acknowledged. References 1. C. Rubbia et al., A high resolution spallation driven facility at the CERNPS to measure neutron cross sections in the interval from 1 e V and 250 MeV, CERN/LHC/98-02(EET), Geneva, May 30, 1998. 2. U. Abbondanno et al. Study of background in the measuring station at the n.TOF facility at CERN, CERN/SL-Note 046, 2001. 3. R. R. Winters, R. L. Macklin, Phys. Rev. C 25, 208 (1982).
RECENT ASTROPHYSICS RESULTS FROM ORELA AND POSSIBLE FUTURE EXPERIMENTS AT ORELA AND SNS P. E. KOEHLER Physics Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 E-mail:
[email protected] I present some recent results from experiments at the Oak Ridge Electron Linear Accelerator (ORELA) and discuss their impact in nuclear astrophysics. I then describe some possible future nuclear astrophysics experiments at ORELA and at the Spallation Neutron Source (SNS) being built in Oak Ridge. The SNS and ORELA are complementary, world-class facilities and both will be needed for important future experiments in nuclear astrophysics.
1
Introduction
The ORELA facility [1] has a long and distinguished history of experiments in nuclear astrophysics. Most of the neutron capture reaction rates used in nuclear astrophysics calculations were determined in experiments at ORELA by Dick Macklin and collaborators. More recently, an improved apparatus [2] has made it possible to measure these rates much more accurately and to lower energies than before. These new data are needed to make use of new high precision isotopic anomaly data from meteorites [3] and to test the latest stellar models [4]. Also, it was recently realized [5] that («,a) experiments at ORELA could provide perhaps the best constraints on the many (y,a) rates needed for explosive nucleosynthesis calculations. In addition, ORELA would be an excellent facility for several other types of important nuclear astrophysics experiments, such as inelastic neutron scattering, neutron capture on long-lived radioactive samples, or total cross section measurements on shorter lived radioactive samples. Most of these topics are discussed in an ORELA "White Paper" that is available at http://www.phv.ornl.gov/astrophvsics/nuc/neutrons/whitepaper.pdf. The intense neutron flux at the SNS [6] should allow measurements on much smaller samples than is possible at ORELA. A recent study [7] indicates that the flux at the SNS should be over 10000 times higher that at the ORELA "benchmark" facility where most previous neutron capture measurements for nuclear astrophysics have been made, and over 40 times higher than at the new DANCE instrument (see John Ullmann's paper in these proceedings) at the Los Alamos Neutron Science Center (LANSCE) [8]. Therefore, the SNS should make possible measurements on widest range of radioactive and very small stable samples of interest to nuclear astrophysics.
32
33
2
Recent ORELA Results and the Need for New Measurements
If you are not familiar with this field, I urge you to read the excellent overview paper by Franz Kappeler at the beginning of these proceedings to acquaint yourself with the basic concepts and jargon. 2.1
The cool, new s-process models
The latest, most realistic, and most successful models [4] of the s process indicate that roughly half of the abundances of nuclides heavier than A=100 were made in low-mass Asymptotic Giant Branch (AGB) stars. A major difference between these and previous models is that most of the neutron exposure driving the nucleosynthesis occurs at much lower temperatures (kT=6-8 keV) than previously thought (kT=30 keV). This could be a problem because most of the old measurements, and even some of the new high-precision measurements, were not made to low enough energies to obtain the reaction rates at these new lower temperatures without resorting to extrapolations. At ORELA, we can routinely measure the neutron capture cross sections across the entire range of energies needed. In all [9,10] but one [11] of the cases studied so far, new ORELA measurements indicate that extrapolations from previous data to obtain reaction rates at the low temperatures needed by new stellar models are in error by two to three times the estimated uncertainties. Therefore, extrapolated rates are not sufficiently accurate for meaningful tests of new stellar models. More low-energy measurements are needed, especially for the s-only isotopes that serve as the most important calibration points for the models. 2.2
Cracks in the Classical s-process Model
The so-called classical model of the s process has been used for many years because, by making some simplifying assumptions (constant temperature, neutron density, and matter density), it is possible to find an analytical solution to the large network of time-dependent, coupled differential equations describing the reaction flow during the s process. As a result, the classical model has been very useful for ascertaining the mean conditions of the s-process environment. Although stellar models indicate that the classical model assumptions are too severe, it has been amazingly successful in reproducing the observed s-process abundances. The competition between neutron capture and beta decay at several relatively long-lived radioactive nuclides along the s-process path can yield a very direct handle with which to estimate the average neutron density, temperature and matter density in the stellar plasma during the s process. Because we know assumptions of the classical s-process model are too simplistic for real stars, if (n,y) cross-section
34
data of sufficient accuracy exist, classical analyses of different branchings should eventually yield inconsistent results. Thanks to precise new data from ORELA, cracks in classical model are beginning to show. Previous classical analyses of branchings in the ^-process path had led to a temperature of kT= 29±5 keV. In contrast, recent precise l34'l36Ba(«,y) reaction rate measurements from ORELA [8] were used in a classical analysis of a different branching to deduce a mean .y-process temperature of kT= 15±5 keV. This was the first time that clearly inconsistent temperatures were obtained from different .^-process branchings. Even more recent ORELA measurements on isotopes of Pt [12] have yielded a second example of an inconsistency, this time in the neutron density. Because stellar models are complicated, more precision measurements near other branching points (e.g., 85Kr, 95Zr, 151Sm, 152Eu, ,53Gd, 163Ho, 170'17,Tm,..) will be needed to understand the crucial ingredients in the new stellar models. Additional cracks in the classical s-process model appeared when new ORELA data [12,13] demonstrated for the first time that the classical model of the s process fails to predict the correct abundances of the ,?-only isotopes Nd and Pt. 2.3
Red Giant Stardust
Microscopic grains of silicon carbide and other refractory materials recovered from primitive meteorites represent a new class of observational data with which to constrain astrophysical models. Most of these grains appear to be actual Stardust from red giant (AGB) stars inside of which the s process had occurred. Trapped within these grains are trace amounts of several intermediate- to heavy-mass elements having isotopic patterns that are very non-solar. Qualitatively, these patterns agree with the expectations of nucleosynthesis from the s process - they are relatively enriched in .^-process isotopes and depleted in isotopes thought to come from the r and p processes. Also, the precision to which these isotope ratios can be measured is much higher than the precision of measured element-to-element abundances for the solar system. Therefore, these new meteorite data extend both the number and precision of the calibration points for .y-process models. New highprecision neutron capture data are needed to see if this beautiful, qualitative redgiant Stardust model can be made quantitative. The first precise test of the red-giant Stardust model recently was made possible when new ' Nd(n,y) cross sections were measured [13] with good precision at ORELA. Stellar ^-process model calculations made with previously accepted cross sections were in serious disagreement with the Stardust data. The new ORELA measurements, which were made with an improved apparatus and over a wider energy range, showed that the old data were in error. With the new ORELA data, the agreement between the stellar model and the Stardust data was excellent. Subsequent ORELA measurements for isotopes of barium [9,11] have revealed problems in the red giant Stardust model. Stardust data for other elements exist (e.g.,
35
Sr, Mo, and Dy), but because many of the existing (n,y) data are too imprecise or do not cover the entire energy range needed by the models, new measurements are needed to make use of these data to test and improve the red-giant Stardust model. 2.4
Improving Reaction Rates for Explosive Nucleosynthesis Models
The neutron-deficient isotopes (the so-called p isotopes) of intermediate- to heavymass elements cannot be made by neutron capture reactions starting from stable "seed" nuclides. It is thought that they were synthesized when seeds built up by a previous s process were photo-eroded in an explosive, high-temperature environment during the p process. The site of the p process is unknown, but the leading candidates appear to be the late stages in the lives of massive stars or supernova explosions. The largest nuclear physics uncertainties in these models are the rates for (y,a) reactions. Determining these rates through direct laboratory measurements is very difficult if not impossible because the cross sections are extremely small due to high Coulomb barriers. Because the level densities are high at the excitation energies and masses of interest, the rates can, in principle, be calculated to sufficient accuracy using the statistical model. However, the a+nucleus potential needed for this model is very poorly constrained, so calculated rates are very uncertain. Traditional methods for constraining potentials are problematical because they involve a largely unconstrained extrapolation. A series of low-energy (n,a) measurements, across a wide range of masses, appears to be the best means of constraining the a+nucleus potential and thus improving the calculation of these rates. The Q values for these (w,a) reactions are such that the relative energies between the a particle and the residual nucleus are in the astrophysically interesting energy range, so no extrapolation is necessary At ORELA, the first application of this idea was to measure the 143Nd and 147 Sm(n,a) [5] cross sections across the range of energies needed for astrophysics applications. Previous measurements of this type were limited to energies below a few keV (which is too small of an energy range to be useful for comparison to statistical models) due to overload problems in the detectors and associated electronics resulting from the y flash at the start of each neutron pulse. In the new experiments, this problem was overcome by employing a compensated ion chamber (CIC) [14] as the detector. The CIC reduced the y-flash background to the point where measurements are possible to much higher energies (500 keV in the cases of 143 Nd and 147Sm). Results from the ORELA measurements are shown in Fig. 1. The older calculations of Holmes et al. [15] are much closer to the data than the newer NONSMOKER [16] or MOST [17] calculations, which differ from the data by about a factor of 3 in opposite directions. The better agreement of the older model may be
36
due to a fortuitous cancellation of effects. The newer models employ a neutron potential that is known to be more reliable in this mass region. In addition, the authors of the newer models have attempted to reduce the reliance on empirical fine tuning and to take advantage of the latest physics knowledge in an effort to increase the reliability of the models away from the valley of stability. In the case of the a potential, several parameters are needed to account for the mass, energy, and nuclear structure effects. At present, the values of these parameters in the astrophysically relevant range are poorly constrained by experiment. 10J
'Nd(n,cx)
X •°jo
!
O
ORELA E x p e r i m e n t
^
MOST'3.7 NON-SMOKER/2.7 H o l m e s e l a/.'1.02
147
if
O
K
**k
+
he4 + ru57 ra57+hE4-:—>ai61
fa
99067951
0 99153874
fo 99605758
099635454 |0 99707891
,11156 + 111 h l + c u 6 1
|o 98797416
[0 99736149
f
co57 + h l «
•>ni58
jo 77309935
[0 99876414
ini58 + nl
ni59
jo 96352522
jo 99894000
lco56+hl
t»57
fa 99891188 h Qf.Mnt'ri
0 99909458
->5Q4
W<x»~
0091)0^7 !$lrtemtt
Fig. 1 General layout of the interactive web based application used to display the alpha rich freezeout sensitivity data. Each numbered item corresponds to a subheading in the next section. The URL of the web site is http://photon.phys.clemson.edu/gjordan/Ion/Ti44 .
3
WEB-BASED DATA DISPLAY
As mentioned above, the results of the alpha rich freezeout survey can be viewed on the World Wide Web. In this section we introduce the interactive data display and give instructions for its use. The URL of the interactive data display is http://photon.phys.clemson.edu/gjordan/Ion/Ti44/index.html.
45
To operate the web site, first select the data set you wish to view (specified by the Rate Factor and Neutron Excess options). Next, enter the isotope of interest in the text fields and click the appropriate button to generate the desired data lists. The website also includes a Beta Decay Option which takes the abundances from the alpha rich freezeout calculation and allows them to beta decay for a specified amount of time. This is useful since the astrophysical sites of the alpha-rich freezeout (type II supernovae) are often observed many years after the explosive nucleosynthesis has taken place. The Beta Decay Option allows the abundances to beta decay and thus be compared to experimental data at the time of observation. In the subsequent subsections of this section, we present an explanation for the several user inputs and possible outputs from the web-based data display.
HATE FACTOR
KEUTRON EXCESS r0 r 0.002
