University of Washington Inc Services
• Location: Washington
• Founded:
• Website:

#### Pages

• Page 1

• Page 2

ANNUAL REPORT Nuclear Physics Laboratory University of Washington April, 1993 Supported in part by the United States Department of Energy under grant DE-FG06-90ER40537

• Page 3

This report was prepared as an account of work sponsored in part by the United State Gov- ernment. Neither the United States nor the United States Department of Energy, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibil- ity for the accuracy, completeness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately-owned rights.

• Page 4

INTRODUCTION The Nuclear Physics Laboratory of the University of Washington has for over 40 years sup- ported a broad program of experimental physics research. The current program includes \in- house" research using the local tandem Van de Graa and superconducting linac accelerators and non-accelerator research in gravitation as well as \user-mode" research at large accelerator facilities around the world. Our local studies of nuclear structure and reactions include nuclear astrophysics, giant resonance (GDR) studies, studies of fusion and high angular momentum phenomena in nu- cleus nucleus collisions, investigations of fundamental symmetries, AMS studies using the tandem as an ultra-precise mass spectrometer, and a variety of applied activities by outside users. There is an active program at the laboratory to develop instrumentation in support of both the in-house and the user-mode physics activities. The user-mode research includes work at the Argonne ATLAS facility, the CERN SPS, the Indiana Cyclotron, the Michigan State NSCL facility, the Saskatoon electron facility, the Tandar facility in Argentina, and the SLAC facility. Below we present some of the highlights of the UW NPL research program. In our studies of GDR decay in highly excited nuclei, we've extended our search for highly deformed, very rapidly rotating light nuclei to the A 60 mass region. Again, as in the A 45 mass region, strongly broadened GDR strength distributions are observed, indicating highly deformed compound nuclei. Construction of a simple multiplicity lter is now complete, it will be used in coincidence experiments with high energy -rays to provide future information on the nature of the reaction processes. A re-investigation of the spin distributions in near-barrier fusion of 16 O +154 Sm utilizing rota- tional state populations as a probe has been completed. The mean spins at lower energies are in better agreement with theoretical expectations than results from previous investigations. A study of the entrance-channel mass-asymmetry dependence of the mean spin in sub-barrier fusion is near- ing completion. The results are in remarkable agreement with expectations in contrast to results from another study elsewhere for a similar set of reactions. A study of the impact parameter dependence of pre-equilibrium proton and light cluster emission at 13.5 MeV/A has been completed. The proton impact parameter dependence is in good agreement with an extension of the nucleon exchange transport model. The ratio of light cluster to proton multiplicities is found to increase with increasing impact parameter, contrary to the expectation from a simple coalescence model. This study is being extended to much higher bombarding energies. The Eot-Wash group continued to expand the scope of their investigations. During this last year they published a test of the equivalence principle for ordinary matter falling toward galactic dark matter. They brought into operations a new torsion balance that is surrounded by a rotating three tonne uranium source. The research in fundamental symmetries focuses on tests of symmetries in atomic, gravitational, and neutron physics. Two neutron experiments are being pursued: the search for an electric dipole moment of the neutron (time reversal symmetry violation) and a measurement of the parity violating interaction between neutrons and alpha particles. The neutron electric dipole moment experiment has been rebuilt to measure simultaneously in the same vessel the spin precession

• Page 5

frequencies of trapped neutrons and mercury atoms. The experiment is scheduled to begin data collection in 1994. An apparatus to measure the parity violating rotation of the polarization vector of a neutron beam as it traverses a target of liquid helium is being constructed. Neutron beam time is scheduled at the NIST reactor in 1994. An experiment to measure the electric dipole moment of mercury atoms has been completed recently, in collaboration with Professors Fortson and Lamoreaux, and has yielded the highest precision ever achieved for an electric dipole moment search. A publication is being prepared. In the accelerator mass spectrometry (AMS) program, work continues under the two-year NSF grant awarded last year as part of the PALE program (Paleoclimates of Arctic Lakes and Estuaries). The rst objective, validation of our method of dating pollen from lake sediment and peat bog cores (extraction and purication of pollen, AMS measurement of 14 C/13C), has been achieved we are now beginning work toward the second objective, tracing the migration of alder and spruce in Alaska following the latest period of widespread glaciation. The accuracy and precision of our measurements have been further enhanced by additional improvements in the high-energy beam transport and detector systems, and by changes in the low-energy beam transport system and in our method of tuning the ion beam through it. The data from the Saskatoon pion photoproduction experiment have been analysed. The cross sections have a dependence on atomic mass which exhibits eects of Pauli blocking and of absorption of the pions as they exit the nucleus. The model we have used to explain our inelastic pion scattering measurements has been extended to photoproduction and it is able to reproduce the spectral shapes and angular distributions using the same parameters for the pion-nucleus interaction that were used in the scattering calculations. The NE-18 collaboration at SLAC completed the analysis of the elastic electron scattering on the neutron and proton. The results show a signicant deviation from the dipole form factor for elastic magnetic scattering on the proton and establish that the Dirac form factors of the neutron and proton are of comparable magnitude. We have developed a new proposal for elastic photon scattering on the proton at high momentum transfer (SLAC E147). This second order electromagnetic interaction complements the elastic electron scattering. The new ultra-relativistic heavy ion (URHI) physics research program of several members of this laboratory, initiated in mid-1990, is now well established, and URHI activities are well represented in this report. In 1991-92 the URHI group participated in two nal runs of experiment NA35 at CERN. The approximately 200 gigabytes of time projection chamber (TPC) data collected during these measurements are now being analyzed here. The UW group is participating in development of new equipment and software for experiment NA49, to run at CERN in late 1994. Their activities in the past year have focused on the vertex TPCs and the data acquisition software for that experiment. The UW group is a founding institutional member of the STAR collaboration, which is constructing a solenoidal TPC tracking detector for use at the RHIC collider now under construction at Brookhaven National Laboratory and recently re-scheduled to come into operation in early 1999. The Laboratory provides beams for a range of uses outside of conventional nuclear physics. This year researchers from the Ball Aerospace Systems Group and the Boeing Defense and Space Groups

• Page 6

used beams from the tandem and booster to investigate radiation damage in various electronic devices. We have continued to study cross sections of interest in the production of isotopes for positron emission tomography. The tandem accelerator has run for 27 years without a major upgrade. This year we requested funding for the installation of a pelletron charging system and spiral beam tubes to improve the quality and quantity of beams from the accelerator. We recently heard that this request will be funded in the near future. The linac continues to serve us well. No major breakdowns have occurred this year, nor have we made any signicant modications. As always, we welcome applications from outsiders for the use of our facilities. As a convenient reference for potential users, the table on the following page lists the vital statistics of our accelera- tors. For further information, please write or telephone Prof. W.G. Weitkamp, Technical Director, Nuclear Physics Laboratory, University of Washington, Seattle, WA 98195 (206) 543-4080. For the rst time this year we plan to \publish" an electronic version of this 1993 Annual Report, using the World Wide Web (WWW) system developed at CERN. We have established one of our computers as a WWW server and will maintain an index connecting to PostScript les of the individual articles (including gures in most cases) of this report. The WWW address of the UW NPL entry point is \http://128.95.100.71:80/home npl.html" and is accessible using a WWW browser such as xmosaic or midaswww. We encourage interested parties to access our Annual Report and other selected publications and documents using this system. We close this introduction with a reminder that the articles in this report describe work in progress and are not to be regarded as publications or to be quoted without permission of the authors. In each article, the names of the investigators have been listed alphabetically, with the primary author underlined. John G. Cramer Editor Mara Ramrez Assistant Editor

• Page 7

Contents 1 Astrophysics 1 1.1 The excitation energy of the 11 level in O and the hot CNO cycle 14 : : : : : : : : : 1 1.2 Gamma-ray branching ratios of proton-unbound levels in 37K : : : : : : : : : : : : : 2 2 Giant Resonances and Photonuclear Reactions 3 2.1 Giant dipole resonance decays of Cu nuclei formed at high spins and temperatures : 3 2.2 High energy gamma rays from Ni + Zr reactions : : : : : : : : : : : : : : : : : : : : 5 2.3 Isospin mixing in highly excited medium mass nuclei : : : : : : : : : : : : : : : : : : 6 3 Nucleus-Nucleus Reactions 7 3.1 Re nements of the nucleon-exchange transport model for the emission of hard pho- tons and nucleons : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 7 3.2 Rotational state populations in near-barrier fusion : : : : : : : : : : : : : : : : : : : 8 3.3 Impact parameter dependence of pre-equilibrium light charged particle emission at 13.5 MeV/A : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 9 3.4 Impact parameter tagged light charged particle emission at 25, 35 and 100 MeV/u : 10 3.4.1 Introduction : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 10 3.4.2 Particle identi cation and energy calibration : : : : : : : : : : : : : : : : : : 11 3.4.3 Evaporation residue and ssion fragment angular distributions : : : : : : : : 12 3.4.4 Fission fragment angular correlations : : : : : : : : : : : : : : : : : : : : : : : 13 3.5 Fusion cross sections for three systems that produce 170Hf at near and sub-barrier energies : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 14 3.6 A search for entrance channel e ects in near and sub-barrier fusion : : : : : : : : : : 16 3.7 Scattering of 87 Mev 6 7 Li on 12C : : : : : : : : : : : : : : : : : : : : : : : : : : : : : ; 18 3.8 The APEX experiment : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 19 3.9 APEX silicon array cooling system : : : : : : : : : : : : : : : : : : : : : : : : : : : : 20 3.10 APEX monitor detector system : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 22 3.11 Initial operating experience with APEX : : : : : : : : : : : : : : : : : : : : : : : : : 23 4 Fundamental Symmetries 24 4.1 New equivalence principle results : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 24 4.2 A test of the equivalence principle for ordinary matter falling toward dark matter : : 25 4.3 Improvements to the rotating source experiment : : : : : : : : : : : : : : : : : : : : 26 4.4 New constraints on composition-dependent interactions with ranges down to 1 cm : 27 4.5 Development of a spin-polarized torsion pendulum : : : : : : : : : : : : : : : : : : : 28 4.6 Design of an apparatus to measure the PNC spin rotation of transmitted cold neu- trons in a liquid helium target : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 29 4.7 Isovector radiative decays of the 16.6 and 16.9 MeV doublet in 8 Be : : : : : : : : : : 30 4.8 Improved limits on scalar currents in weak interactions : : : : : : : : : : : : : : : : : 31 5 Accelerator Mass Spectrometry 32 5.1 Scienti c program : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 32 5.1.1 AMS 14 C dating of pollen from lake sediments and peat deposits : : : : : : : 32 5.1.2 Atmospheric methane : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 33 i

• Page 8

5.2 Technological program : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 33 5.2.1 Performance of the HE beam transport system and the wide-aperture detector 33 5.2.2 Alteration in the tuning of the LE beam transport system : : : : : : : : : : 34 5.2.3 Current performance of the measurement system : : : : : : : : : : : : : : : : 34 6 Medium Energy 36 6.1 Inclusive pion photoproduction on several nuclei : : : : : : : : : : : : : : : : : : : : 36 6.2 Fermi gas calculations of inclusive pion photoproduction : : : : : : : : : : : : : : : 38 6.3 Nucleon form factors at high momentum transfer : : : : : : : : : : : : : : : : : : : : 40 6.4 Scaling analysis of Compton scattering on the proton : : : : : : : : : : : : : : : : : : 41 6.5 Compton scattering and exclusive  0 photo-production on the proton : : : : : : : : 42 6.6 Shielding studies for a high energy photon calorimeter : : : : : : : : : : : : : : : : : 44 7 Ultra-Relativistic Heavy Ion Collisions 45 7.1 Ultra-relativistic heavy ion physics: an introduction : : : : : : : : : : : : : : : : : : 45 7.2 Comparison of \candidate" 3-body Coulomb corrections to HBT : : : : : : : : : : : 46 7.3 Maximum likelihood analysis in HBT interferometry : : : : : : : : : : : : : : : : : : 47 7.4 HBT source size dependence on (pt ,y) and TPC resolution : : : : : : : : : : : : : : : 48 7.5 NA35 TPC track variance analysis : : : : : : : : : : : : : : : : : : : : : : : : : : : : 50 7.6 Cluster moment analysis for NA35 TPC data : : : : : : : : : : : : : : : : : : : : : : 51 7.7 Alien clusters and calibration anomaly : : : : : : : : : : : : : : : : : : : : : : : : : : 52 7.8 NA35 TPC Landau uctuation and di usion analysis : : : : : : : : : : : : : : : : : : 54 7.9 Energy loss spectra for argon gas : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 56 7.10 Calculation of energy deposition in argon gas for NA35 and NA49 TPCs : : : : : : : 57 7.11 NA35 TPC data analysis program (TRAC) : : : : : : : : : : : : : : : : : : : : : : : 58 7.12 NA35 TPC phase space representations : : : : : : : : : : : : : : : : : : : : : : : : : 59 7.13 Decoding the position-angle ambiguity of the NA35 TPC : : : : : : : : : : : : : : : 61 7.14 Monte Carlo simulations of NA49 TPC performance : : : : : : : : : : : : : : : : : : 62 7.15 NA49 beam hodoscope simulations : : : : : : : : : : : : : : : : : : : : : : : : : : : : 63 7.16 Vertex determination using the STAR SVT : : : : : : : : : : : : : : : : : : : : : : : 64 7.17 Pattern recognition in the STAR SVT : : : : : : : : : : : : : : : : : : : : : : : : : : 66 7.18 Tiling the STAR SVT for increased detection eciency : : : : : : : : : : : : : : : : 68 7.19 Estimating dE=dx in the STAR SVT with maximum likelihood analysis : : : : : : : 69 7.20 STAR SVT: support apparatus and trigger system for 1993 TRIUMF test run : : : : 70 7.21 Autocorrelation multiplicity trigger simulations : : : : : : : : : : : : : : : : : : : : : 71 7.22 STAR conceptual design report: trigger speci cation : : : : : : : : : : : : : : : : : : 72 7.23 Parallel plate avalanche detector generic research and development : : : : : : : : : : 74 8 Cluster Impact Phenomena 75 8.1 Odd-even structure in atomic carbon cluster size distributions : : : : : : : : : : : : : 75 8.2 Cluster size distributions for Cs impact of C, Al, Si, and Cu : : : : : : : : : : : : : : 77 9 External Users 78 9.1 Radiation e ects on opto-electronic devices : : : : : : : : : : : : : : : : : : : : : : : 78 9.2 Summary of single event e ects testing by BPSRC : : : : : : : : : : : : : : : : : : : 78 9.3 Production of radionuclides by 3 He irradiation of carbon and oxygen : : : : : : : : : 80 ii

• Page 9

10 Instrumentation 81 10.1 Background study for detection of atomic electrons in coincidence with nuclear scat- tering. : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 81 10.2 Gamma-ray Multiplicity Detection Array : : : : : : : : : : : : : : : : : : : : : : : : 83 11 Van de Graa , Superconducting Booster and Ion Sources 84 11.1 Van de Graa accelerator operations and development : : : : : : : : : : : : : : : : : 84 11.2 A proposal to upgrade the tandem accelerator : : : : : : : : : : : : : : : : : : : : : : 86 11.3 Booster operations : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 87 11.4 Cryogenic operations : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 88 11.5 An object-oriented programmable controller : : : : : : : : : : : : : : : : : : : : : : : 89 11.6 Improvements to the linac control system : : : : : : : : : : : : : : : : : : : : : : : : 90 11.7 Tandem terminal ion source : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 91 11.8 Tandem belt charge current monitor : : : : : : : : : : : : : : : : : : : : : : : : : : : 92 11.9 Injector Deck and Ion Sources : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 93 12 Computer Systems 94 12.1 Acquisition system developments : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 94 12.2 Analysis and support system developments : : : : : : : : : : : : : : : : : : : : : : : 94 12.3 Migration of CERN and CEBAF software to DEC3100 : : : : : : : : : : : : : : : : : 95 13 Nuclear Physics Laboratory personnel 97 13.1 Degrees granted, academic year 1992{1993 : : : : : : : : : : : : : : : : : : : : : : : : 99 13.2 List of publications from 1993 : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 100 iii

• Page 10

1 Astrophysics 1.1 The excitation energy of the 11 level in 14 O and the hot CNO cycle E.G. Adelberger, A. Garca and P.V. Magnus The hot CNO cycle becomes an important source of stellar energy generation when the 13N(p; ) reaction rate becomes faster than the 13N -decay rate. The 13N(p; ) rate is dominated by a low- energy s-wave E 1 resonance that corresponds to the 5.17 MeV rst excited state of 14O. For a given stellar temperature and pressure, the reaction rate depends sensitively on the parameters of the low-energy resonance, especially its radiative width, , and its resonance energy, ER . Several direct determinations1 ,2 of the -ray branching ratio and a direct measurement3 of the resonance strength using a radioactive 13 N beam give concordant results that can be combined to yield = 3:64  0:85 eV. This best value' is in excellent agreement with the value = 3:1  0:6 eV determined indirectly4 from the breakup of 14O projectiles in the Coulomb eld of 208Pb. On the other hand, a disagreement has arisen concerning the value of ER. The accepted value5 for the excitation energy of the 14 O rst excited state, Ex = (5173  10) keV, implies ERcm = (545  10) keV, while an analysis of 13 N+p scattering3 yielded ERcm = (526  1) keV where only the statistical error was quoted, and an unpublished6 14 N(3He,t) measurement gives Ex = (5168:5  1:8) keV, corresponding to ERcm = (540:5  1:8) keV. This discrepancy is signi cant; a 19 keV reduction in the 13N(p; ) ERcm would increase the reaction rate by 40% at T = 4  106 K. We have therefore remeasured the excitation energy of the 14 O rst excited state, and as a byproduct the mass of the 18Ne ground state which was known with a relatively large uncertainly of 5 keV. We used the University of Washington pulsed-beam time-of- ight (TOF) spectrometer in a di erential mode, by comparing the TOF of a neutron group of interest to an essentially equal TOF of a well-known calibration group produced in a di erent target at the same bombarding energy. First, by comparing TOF spectra for the 16 O(3He,n)18 Ne(1.88) and 11B(3 He,n)13N(T = 3=2) reactions at a beam energy of 7.30 MeV, the mass of the 18Ne 1.88 MeV rst excited state was de- termined. (The Q value of the 11B(3 He,n)13N(T = 3=2) reaction is known to 0.4 keV.) Subtracting the well-known excitation energy of the 1.88 MeV level, the mass excess of the 18 Ne ground state is found to be (5315  2) keV. Having established this secondary standard, we then measured the mass of 14 O(5.17) by comparing its TOF in the 12 C(3He,n) reaction to that of the 18Ne(3.38) (the excitation energy is again well known) group populated in 16 O(3He,n) at a bombarding energy of 9.00 MeV (see Fig. 1.1). This comparison yielded a 14O(5.17) mass excess of (13168  2) keV, which corresponds to an excitation energy of (5161  2) keV. Our excitation energy is in agreement with the previously accepted value (5173  10) keV because of its large error bar, but not with  Lawrence Berkeley Laboratory, 1 Cyclotron Road, Bldg 88, Berkeley, CA 94720. 1 P.B. Fernandez et al., Phys. Rev. C 40, 1887 (1989). 2 P. Aguer et al. Proc. Int. Symp. Heavy-Ion Phys. and Nucl. Astrophys. Prob., eds. S. Kubono, M. Ishihara and T. Nomura, World Scienti c, Singapore (1989), p. 107. 3 Decrock et al., Phys. Rev. Lett. 67, 808 (1991) and Nucl. Phys. A 542, 263 (1992). 4 T. Motobayashi et al. Phys. Lett. B264, 259 (1991). 5 F. Ajzenberg-Selove Nucl. Phys. A523, 1 (1991). 6 T.F. Wang, PhD Thesis, Yale University, 1986. 1

• Page 11

either the (5154  1) keV value of Decrock et al.3 or the (5168:5  1:8) keV value of Wang.6 Work on this problem is continuing. 2

• Page 12

1.2 Gamma-ray branching ratios of proton-unbound levels in 37K E.G. Adelberger, N. Cabot, P.V. Magnus and H.E. Swanson The isospin-analog GT strength distributions measured in 37Ca( + e ) and 37 Cl(p; n) show sev- eral large discrepancies.1 This has raised questions about the reliability of GT strength distributions deduced from intermediate-energy, zero-degree (p,n) cross sections. It should be noted that the the two largest low-energy discrepancies (to the rst 1=2+ and the second 5=2+ states) could be resolved if the Ex = 3.24 MeV 5=2+ state in 37 K, which is unbound to proton decay by 1.382 MeV decays primarily by emission with a 2% proton branch corresponding to a proton spectroscopic factor of only Sp = 1  10 6 . In this case the -delayed proton yield would not properly re ect the + branching ratios. We have studied the decays of low-lying levels in 37 K using the 40Ca(p; p) and 40 Ca(p; X) reactions at a proton energy of 18 MeV. Here X is either a 36Ar or a 37K. The 37 K(p; ) reaction was rst examined at 30, 37 and 45 degrees in order to determine an optimum angle for the detector in the coincidence measurement. The 40 Ca(p; p) reaction was then measured with a thin alpha detector at 37 which tagged events populating various excited states in 37 K. Measurement of coincident protons in detectors at +141 , -140 and -90 then allowed an estimate of the fraction of the time the levels proton decay. The 40Ca(p; X) reaction was measured with the counter at 120 deg and a coincident counter with an opening angle of 5 degrees placed either at 34:19 or 36:71 to detect the recoiling heavy ion (either 37 K or 36 Ar). Because the 36Ar's from the proton decay of 37K* have a larger opening angle and a larger energy spread than the 37 K's from the gamma decay of 37 K*, the measured coincidence eciency and energy distribution of the heavy ions can be used to determine the decay branching ratios. These data are being compared to a Monte Carlo simulation that includes multiple scattering in the target foil, momentum kick from the decay, and geometry of the target and detectors. The data analysis is not complete. However, some things can already be inferred. We observe all the levels listed in the compilation2 up to Ex =3.5 MeV and see two additional levels at Ex =2.97 MeV and Ex =3.27 MeV. The level at Ex =2.97 MeV was observed at lab angles of 30, 37 and 45 degrees, and its kinematics con rm that it is in mass 37. In those data the Ex =3.27 MeV level was in an unresolved triplet. In addition, the sum of the alpha and outgoing proton energies from the 40Ca(p; p) data are consistent with both new levels being in 37K. Our preliminary results on the proton branching ratios of the levels in 37K indicate that the 2.17, 2.29, 2.97 and 3.24 MeV levels predominantly decay while the 2.75, 3.08, 3.27 and 3.31 MeV levels predominantly proton decay. This is in agreement with previous information about the 2.17, 2.29, and 2.75 MeV levels where 36 Ar+p data are available. After the analysis is complete quantitative results will be available. It should be noted that even if the low-energy discrepancies between 37 Ca -decay and 37 Cl(p; n) are resolved, problems apparently remain at higher excitation energies. 1 A. Garca et al. Phys. Rev. Lett. 67, 3654 (1991). 2 P.M. Endt, Nucl. Phys. A521, 1 (1991). 3

• Page 13

2 Giant Resonances and Photonuclear Reactions 2.1 Giant dipole resonance decays of Cu nuclei formed at high spins and tem- peratures A.W. Charlop, Z.M. Drebi, M.S. Kaplan, M. Kicinska-Habior,y K.A. Snover, D.P. Wellsz and D.Ye We continued our studies of the giant dipole resonance decays of nuclei around mass 60. The motivation is to investigate nuclear structure at extreme conditions, particularly the shape evolution as a function of spin and temperature. At nuclear temperatures above 1{2 MeV and spins just below the ssion limit, the rotating liquid drop model (RLDM)1 predicts a shape transition from oblate noncollective to triaxial collective-prolate-like with very large deformation. Our experiments are designed to search for evidence for such a shape transition. We measured the spectral shapes and the angular distributions of the high energy rays emitted in the decays of Cu nuclei formed in 18O + 45Sc and 32S + 27Al reactions. Some of these measurements results were presented earlier.2 Here we will discuss the 32S + 27Al measurements. S beams at projectile energies 90{215 MeV (in the center of target) were used with self- 32 supported rolled 27 Al targets to form 59 Cu compound nuclei at excitation energies ranging from 54 to 111 MeV and with spins up to 43 h in the highest bombarding energy case. Our large NaI spectrometer was used to detect high energy rays at ve lab angles in the range 40{140 degrees with respect to the beam direction. The angular distributions in the C.M. frame were tted with a second order Legendre polynomial expansion. The measured energy spectra in the GDR energy range 13{30 MeV were tted with the statis- tical model CASCADE. The Reisdorf level densities formulation was used, and a RLDM moment of inertia with the oblate to triaxial shape change was used to calculate the statistical yrast line. In all calculations complete fusion was assumed, and measured fusion cross sections were used. From these ts, the average cross sections abs (E ) for the inverse process of photoabsorption can be extracted.3 These cross sections together with the extracted angular distribution coecients a1 and a2 are plotted in Fig. 2.1-1. A One-Lorentzian GDR strength function was sucient to t the lowest three energy cases. In the highest two energy cases a suggestion of a second peak or shoulder at E  25 MeV was observed, and a two-Lorentzian strength function was necessary to t the data. The deduced GDR width (FWHM) increases from 9.2 MeV to 15.8 MeV in the Eproj =90 to 215 MeV range. This broadening of the strength function is due to two e ects: the spin-induced equilibrium deformation of the nucleus, and the temperature-induced uctuations around these equilibrium deformation shapes. Experimentally we can not separate these two e ects. Full thermal Now at: University of Washington Medical Center, Department of Radiology, Seattle, WA 98195. yPresent address: Institute of Experimental Physics, University of Warsaw, Poland. z Now at: Environmental Radiation Section, Department of Health, Radiation Protection Division, Olympia, WA 98504. 1 S. Cohen, F. Plasil, and W.J. Swiatecki, Ann. Phys. 82, (1974), 557. 2 Nucear Physics Laboratory Annual Report, University of Washington (1992) p. 14 and Nuclear Physics Labora- tory Annual Report, University of Washington (1991) p. 3{6. 3 fit :measured =cascade. abs = abs 4

• Page 14

shape and orientation uctuation calculations are needed to interpret the data. These calculations are in progress. The inferred a1(E ) coecients are consistent with zero in the GDR region of interest. However large negative values along with a yield are observed at E 12 MeV. This excess -yield is due presumably to emission from excited binary or ssion reaction fragments. The angular distribution a2 (E ) coecients show, as expected, a negative dip in the low side of the GDR peak. This dip should increase with increasing nuclear deformation, which is not apparent in our data. This can be due to thermal orientation uctuations which tend to reduce the size of the a2(E ) coecients. Furthermore, the a2 's should turn positive at E > EGDR , for prolate-collective or oblate noncollective shapes, which is not apparent in our data due to the poor statistics. Fig. 2.1-1. First row: The measured cross section, and the CASCADE ts (solid lines). Second and Third rows: The corresponding angular distribution coecients a2 , and a1 respectively. 5

• Page 15

2.2 High energy gamma rays from Ni + Zr reactions Z. Drebi, M. Kaplan, K.A. Snover, D. Wellsy and D. Ye We have completed our analysis of the high energy gamma ray spectra measured1 in the reactions 58Ni +92Zr at E lab = 241 Mev. The purpose of our experiment is to address the possibility of the persistence of large deformation, associated with the near mass-symmetric entrance channel, for times comparable to the compound nuclear lifetime. Our technique is based on an analysis of the decay of the Giant Dipole Resonance. This technique should be particularly sensitive, since GDR decay tends to occur early in the evaporation cascade, and the GDR strength function has a well-known sensitivity to deformation. Corrections were applied to the measured spectra for the contributions from light target contam- inants, and for decay following deep inelastic scattering. The light target contaminant corrections were based on spectra measured at Elab = 196 MeV, below the Coulomb barrier for 58Ni + 92 Zr, using 92 Zr, C and O targets. These corrections amounted to 15{20% in the high energy region. Gamma decay contributions from deep inelastic scattering were estimated from Cascade calcu- lations for the decay of product nuclei produced with cross sections estimated from a published study2 of deep inelastic scattering of 58Ni + 112Sn. Deep inelastic corrections were important only for E  8 MeV. We attribute the resulting corrected spectra to decay of the compound nucleus 150Er (E= 57 MeV). A single-Lorentzian GDR strength function was varied in a t of the Cascade statistical model calculation to the data, with the result S=1.4 0.2, EGDR = 13.5  0.3 MeV and = 8.1  0.6 MeV. With the possible exception of EGDR , which is somewhat low (typical values are 14{14.4 MeV in this mass region), the tted GDR parameters are in agreement with systematics. In order to examine the possibility that anomalous shape persistence e ects may depend on the compound nuclear isotope, we have carried out similar measurements for the reaction 64 Ni + 92Zr at the same bombarding energy. Preliminary analysis shows that these data, like the 58Ni + 92Zr data describe above, show no evidence for an excess yield of gamma rays at high energies, above the GDR peak, in contrast to previously published results.3  Now at: University of Washington Medical Center, Department of Radiology, Seattle, WA 98195. y Now at: Environmental Radiation Section, Department of Health, Radiation Protection Division, Olympia, WA 98504. 1 Nucear Physics Laboratory Annual Report, University of Washington (1992) p. 10. 2 F.L.H. Wolfs, Phys. Rev. C 36 1379 (1987). 3 M. Thoennessen et al., Inst. Phys. Conf. Series # 109, 135 (1990). 6

• Page 16

2.3 Isospin mixing in highly excited medium mass nuclei J.A. Behr, Z. Drebi, M. Kaplan,y K.A. Snover, D. Wells z and D. Ye We have studied isospin mixing in A  60 compound nuclei by comparing the -ray production cross sections for GDR decay of 60Zn formed in the fusion of 28 Si and 32S at E  47, 63 and 80 MeV with the cross sections for forming the neighboring N 6= Z compound nuclei 59Cu and 58Ni at similar excitation energies in the 31P + 28Si, 32 S + 27Al and 31P + 27Al reactions. Results from the present data indicated 2>  0.1 at all 3 excitation energies. When the present results are combined with the results of an earlier study of ( ; ); (p; ) and (p; p) reactions which indicated 2>  0.4 at E  = 20 MeV, the combined results provide the clearest indication to date of the restoration of compound nuclear isospin symmetry at high excitation energy.1 Calculations using the Reisdorf prescription for the level density show that the constant value #> = 20 keV provides a good description of the experimental data. Calculations with the Puhlhofer level density imply a larger value #> 40 keV. Current e orts are focussed on a deter- mination of a best value and an uncertainty for #> .  Now at: Department of Physics, State University of New York, Stony Brook, Stony Brook, NY 11794. y Now at: University of Washington Medical Center, Department of Radiology, Seattle, WA 98195. z Now at: Environmental Radiation Section, Department of Health, Radiation Protection Division, Olympia, WA 98504. 1 See also J.A. Behr et al. Phys. Rev. Lett., in press. 7

• Page 17

3 Nucleus-Nucleus Reactions 3.1 Re nements of the nucleon-exchange transport model for the emission of hard photons and nucleons S. J. Luke, J. Randrupy and R. Vandenbosch We have extended an earlier1 2 nucleon exchange transport model to include proton emission in ; addition to neutron emission. This extension is relatively straightforward, requiring only additional e ects of the Coulomb barrier for the escaping protons and the e ect of the Coulomb potentials of both donor and receptor nuclei on the outgoing proton. The multiplicity of the pre-equilibrium protons is lower than that of the neutrons and the proton spectra are harder than the neutron spectra as expected. We have also incorporated the e ect of the known di use momentum distributions of the nu- cleons in the nuclear ground state. This has been done in a somewhat ad hoc manner since it goes beyond the basic one-body nature of the model. The di useness used in our calculations is based on simulating the experimental ground state nucleon momentum distribution rather than by tting pre-equilibrium particle emission spectra. It is therefore gratifying to nd that the high- energy slopes of the calculated emission spectra are in good agreement with experiment. This extension is most important for low bombarding energy and for more asymmetric systems at a given bombarding energy, as it increases the multiplicity and hardens the spectra most in these circumstances. For higher bombarding energies and more symmetric systems energy dissipation at the early stages of the collision leads to hot nuclei with di use momentum distributions which dominate the contribution from the di use corrections to the ground state momentum distribution. We have also investigated changes in the expected pn nucleon-nucleon bremsstrahlung due to both the aforementioned di use momentum distributions and quantum-mechanical e ects on the elementary production mechanism. The latter had been treated classically in the earlier work.2 Nakayama3 and Schafer et al.4 have shown that quantum-mechanical e ects are most important at higher nucleon-nucleon energies and for photons near the kinematic limit. The enhancements are typically less than a factor of two. When incorporated in our transport model they are most important for proton-induced reactions where it is possible to measure the photon emission near the kinematic limit, and for heavy-ion reactions at higher bombarding energies. We have pursued a suggestion of Stuart Gazes that deceleration of the partners in the heavy ion reaction during the transit time of an exchanged nucleon may modify the emission probability more for jets originating from the heavier reaction partner than for those originating from the lighter partner. The results obtained are in the direction expected, but their e ect on residue velocity is modest. Predictions of mean residue velocities exhibit the correct trend with mass asymmetry of the entrance channel but underestimate considerably the magnitude of the dependence. This underestimation is attributed to the fact that considerable momentum is carried away by composite  Present address: Lawrence Livermore National Laboratory, L-397, Livermore, CA 94551. y Lawrence Berkeley Laboratory, University of California, Berkeley, CA 94720. 1 J. Randrup and R. Vandenbosch, Nucl. Phys. A 474, 219 (1987). 2 J. Randrup and R. Vandenbosch, Nucl. Phys. A 490, 418 (1988). 3 K. Nakayama,, Phys. Rev. C 39, 1475 (1989). 4 M. Schafer, T.S. Biro, W. Cassing, U. Mosel, H. Nifenecker and J.A. Pinston, Z. Phys. A 339, 391 (1991). 8

• Page 18

particles, particularly particles. These composite particles may arise from massive transfer or breakup- ssion mechanisms, as well as from coalescence mechanisms. 9

• Page 19

3.2 Rotational state populations in near-barrier fusion J.D. Bierman, A.W. Charlop, D.J. Prindle, R. Vandenbosch and D. Ye We have completed our study of the populations of the ground state rotational band of the 4n channel decay of 170Yb produced using two entrance channels: 16 O + 154Sm and 4He + 166Er. We have used three large Compton-suppressed intrinsic germanium detectors and coincidence tech- niques to measure the rotational state transitions. The populations were determined more accu- rately and more completely than a previous detector system1 allowed. We were able to clean up the spectra enough so that three transitions, which before were unobservable, could be studied. At last report2 we had obtained data at two oxygen bombarding energies, 68 and 65 MeV, and had also repeated the experiment using the 4 He + 166Er reaction at 48 MeV. The latter reaction matches the higher energy oxygen system in excitation energy but is well above the barrier and is thus used to calibrate the parameters in the statistical model code PACE. We have now also obtained data for both systems at lower energies which match in excitation energy, 63 MeV for the oxygen system and 43 MeV for the alpha system. We were then able to con rm that the PACE parameters, which we determined from the higher alpha energy, were valid over our entire energy span. We have used the PACE code to convert our rotational state population results into values of the average angular momentum for the 16 O + 154Sm system at all three near-barrier energies. Figure 3.2 shows our results as well as results from experiments using discrete gamma tagging tech- niques and evaporation residue tagging techniques.3 Also shown is a coupled channels calculation using the code CCDEF. Our results compare well with the calculation and are slightly lower than the results from the previous multiplicity measurements. This suggests that the rotational state population method is a better probe of mean angular momentum values in the near-barrier energy regime than the multiplicity method because of the problems with converting multiplicities to mean angular momentum at near-barrier energies. We have also been able to determine a more detailed idea of what the actual spin distribution of the compound nucleus is. We found that the oxygen system at near-barrier energies had a much broader distribution in the compound nucleus than the sharp-cuto like distribution which can be predicted for the asymmetric, well above barrier, 4 He + 166Er system. While this experimental method is limited in the number of systems which are practical for study, we have found that if a proper calibration system is also studied that useful information about the spin distribution of compound nuclei resulting from near-barrier fusion can 1 R. Vandenbosch, B.B. Back, S. Gil, A. Lazzarini, and A. Ray, Phys. Rev. C 28 1161 (1983). 2 Nuclear Physics Laboratory Annual report, University of Washington, (1992) p. 22. 3 S. Gil, A.W. Charlop, A. Garcia, D.D. Leach, S.J. Luke, S. Kailas, and R. Vandenbosch, Phys. Rev. C 43 701, (1991). 10

• Page 20

be extracted. Fig 3.2 Mean l for 16O + 154Sm de- termined from conversion of rotational band populations using the evapora- tion model code PACE. Also shown are results from gamma multiplicities from discrete gamma (Ge) tags and evaporation residue (De ector) tags. The full curve is from a coupled chan- nels calculation. 11

• Page 21

3.3 Impact parameter dependence of pre-equilibrium light charged particle emission at 13.5 MeV/A A.W. Charlop, C.E. Hyde-Wright, S. Kailas, D.J. Prindle, and R. Vandenbosch The analysis of our previously reported experiment1 on light charged particles produced in 16 O bombardment of Tb, Ta, and Au has been completed. For the rst two targets we have extracted LCP multiplicities for both evaporation residue (ER) and ssion fragment (FF) tags. For the Au target we only have multiplicities for FF tags. From these results we have been able to deduce the dependence of pre-equilibrium multiplicities on impact parameter. The multiplicity of pre- equilibrium protons falls o with increasing impact parameter at a rate that is in good agreement with expectations from a nucleon exchange transport model.2 The ratio of any complex particle (d; t,He) multiplicity to proton multiplicity however is found to increase with increasing impact parameter, as illustrated in Fig. 3.3-1. This dependence continues a trend rst observed by Awes et al. who compared pre-equilibrium multiplicities of peripheral collisions with central collisions. 2 The present experiment has enabled us to examine the impact parameter dependence within the single \central" collision bin associated with nearly full momentum transfer.2 Our observations are inconsistent with a simple coalescence model. They suggest a dynamical dependence on where the cluster formation occurs. A particular model based on an extension of the nucleon exchange transport model has been suggested in a previous report.3 Fig. 3.3-1. Ratio of pre-equilibrium complex particle multiplicity to pro- ton multiplicity as a function of im- pact parameter.  Present address: Nuclear Physics Division, BhaBha Atomic Research Center, Bombay, 400 085 India. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1990), p. 11. 2 T.W. Awes et al., Phys. Rev. C 25 (1982) 2361. 3 Nuclear Physics Laboratory Annual Report, University of Washington (1991) p. 19. 12

• Page 22

3.4 Impact parameter tagged light charged particle emission at 25, 35 and 100 MeV/u D. Bowman, G. Cren, M. Chartier, R. Desouza,y J. Dinius, A. Elmaani, D. Fox,y K. Gelbke, W. Hsi, C.E. Hyde-Wright, W. Jiang, W. Lynch, T. Moore,y G. Peaslee, D.J. Prindle, C. Schwarz, M.-Y. Tsang, R. Vandenbosch and C. Williams 3.4.1 Introduction The mechanism for producing pre-equilibrium complex fragments in heavy ion collisions is not well understood. This is particularly true for more central collisions where prompt and sequential decays of projectile-like particles do not contribute. At low bombarding energies, complex particles are primarily limited to particles with masses 2{4. At higher bombarding energies considerably larger complex particles are produced in appreciable yield. An important aspect of any reaction process is the impact parameter dependence. Characteri- zation of impact parameter on an event-by-event basis is dicult, particularly for central collisions where the energy deposition for a wide range of impact parameters is similar. We have shown that it is possible to use the angular momentum dependence of evaporation- ssion competition to generate tags for the most and less central impact parameters within the general class of fusion-like impact parameters.1 Using this technique we have shown that at 13.5 MeV/u the pre-equilibrium proton multiplicity falls o with increasing impact parameter while the ratio of pre-equilibrium complex particle to proton multiplicity increases with increasing impact parameter. In the present experiment we extend this study to higher bombarding energies and to coincidence detection of two or more light charged particles. We used the MSU Miniball detector2 to measure charged particle yields in 25 MeV/u 16O and 35 MeV/u & 100 MeV/u 14N on Au, Sm, Ta, and Tb targets. The forward Miniball ring was replaced with an array of Si (E,veto) telescopes to measure coincident evaporation residues. In each of the remaining rings, one Miniball element was replaced with an Ion Chamber-Si-CsI (IC) telescope. We used the ion chambers primarily to measure the coincident ssion fragments. From the information obtained in this experiment we will also be able to address some other issues. One of the current topics for these intermediate energy heavy-ion induced reactions is the probe of nuclear matter properties via particle interferometry, or small-angle particle-particle cor- relations. The major driving forces for these correlation measurements are the emission space-time extent of the particle pair used to construct these correlations.3 We intend to use relative mo- mentum and reduced velocity correlation functions, respectively, to characterize the mean emission lifetime for light charged particles (LCP) and intermediate mass fragments (IMF). Our evaporation residue and ssion tags will allow us to compare results for the most and less central collisions as described above.  National Superconducting Cyclotron Laboratory, Michigan State University, E. Lansing MI. y Indiana University Cyclotron Facility, Indiana University, Bloomington IN. 1 See section 3.3 of this report. 2 R.T. Desouza et al., Nucl. Inst. Meth. A 295,109(1990). 3 A. Elmaani et al., Phys Rev C43 R2474 (1991), and A. Elmaani et al., Nucl. Inst. Meth. A313 401 (1992). 13

• Page 23

The study of the evolution of the emission timescales over a wide range of excitation energies (25{100 MeV/u) will be used to look for a change in the emission mechanism. We are particularly interested in the transition from fusion-like reactions, which occur on a timescale of hundreds of fm/c, to nuclear breakup, occurring on timescales of a few fm/c. In addition to small angle particle-particle correlations, we will compare the yields of complex particles with the coincidence yields of their (approximate) constituents, e.g. d vs. pp, 4 He vs pt, dd, etc, as a further guide to the emission mechanism. 3.4.2 Particle identi cation and energy calibration The signal from each miniball detector was fanned out into 4 signals: 1 timing signal and 3 analog signals. The analog signals were integrated over 3 separate gates, de ned relative to the constant fraction discriminator output from the timing signal. The Fast gate has a 5 ns delay and 35 ns width. The Slow gate has a 200 ns delay and a 400 ns width. The Tail gate has a 2 s delay and a 2 s width. Fig. 3.4.1. (a) Energy calibration points and the resulting quadratic ts for 1 2H and 4 He measured ; in element 19 of the Miniball. (b) The corresponding energy spectra for 1 2H, 4He produced in the ; reaction of 25 M eV =u 16 O + 181Ta and detected at   20. lab 14

• Page 24

We identi ed 1 H, 2 H, 3H, 3He, and 4 He by drawing 2-d gates in a plot of Tail vs. Slow, for each detector. This separation is possible because of the variable intensity and lifetime of the di erent light components in CsI as a function of ionization density.4 We have energy calibration runs with a 12C beam on a polyethylene target. This produces two proton points from p(12C; p)12C and 12 C(4.4 MeV). Also we have a mixed deuteron and alpha beam on a polyethylene target which produce 2 proton points from p( ; p) and p(d; p)d. Deuteron points and calibration points were produced by passing a mixed deuteron and alpha beam through a variable degrader and then scattering on a gold target. We separately calibrate the Tail vs. energy and Slow vs. energy. We use a common quadratic t for energy vs light output for H isotopes and a separate quadratic t for He isotopes. Sample calibration data and energy spectra from one element of the Miniball are shown in Fig 3.4-1 3.4.3 Evaporation residue and ssion fragment angular distributions The forward array of Si telescopes was used to measure evaporation residues. The Miniball array of ten Ion-Chamber, Si, CsI telescopes from 19 to 150 was used to measure both evaporation residues and ssion fragments. In the forward array, the rst Si counter of each telescope stopped the residues, and the second counter vetoed quasi-elastic events. Residues were also identi ed by their correlated pulse height vs. time-of- ight distributions. Two of the Si telescopes were on a moveable arm, reaching from 3.5 to 15.5. In the IC telescope, evaporation residues are identi ed by a ion-chamber signal in anticoincidence with the Si counter. In these same elements, the ssion fragments were identi ed by coincidence between the Si and a large pulse height in the ion-chamber in anticoincidence with the CsI. Angular distributions of evaporation residues and ssion fragments are shown in Fig. 3.4-2. Fig. 3.4.2. Evaporation residue and ssion fragment angular distributions. 4 R.S. Storey, W. Jack and A. Ward. Proc. Phys. Soc. 75 (1958) 72. 15

• Page 25

3.4.4 Fission fragment angular correlations In Figs. 3.4-3 and 3.4-4 we present preliminary result for the reaction 25 M eV =u 16O + 181Ta ! F F 1+ F F 2. F F 1 is the trigger ssion fragment measured in one of the ion chambers and F F 2 is the complementary fragment observed in coincidence in one of the regular Miniball plastic-CsI elements. In Fig. 3.4-3, we show the relative azimuthal angular correlation. The 180 preference clearly indicate the ssion fragment co-planarity. And in Fig. 3.4-4 we show the polar angle correlation of these same coincident ssion fragment. We will be able to extract the mean momentum transfer to the ssioning nucleus from the fragment-fragment polar angle correlation. Fig 3.4.3. Azimuthal angular correlation for ssion fragments detected in coinci- dent events between an Ion Chamber and a Miniball Phoswich detector. Fig. 3.4.4. Surface plot of the polar angular distribution (1 ; 2) of both ssion fragments. The coincidence yield as a function 1 (FF1 detected as a trigger fragment in one of the ion chambers) and 2 (FF2 detected in one of the Phoswich detectors of the Miniball Array). 16

• Page 26

3.5 Fusion cross sections for three systems that produce 170Hf at near and sub- barrier energies D. Abriola, J.D. Bierman, A.W. Charlop, Z. Drebi, A. Etchegoyen, M. Etchegoyen, S. Gil, F. Hassenbale, D.J. Prindle, D. Rodriguez, J. Testoni R. Vandenbosch and D. Ye We have completed our analysis of the fusion cross sections for several systems (28Si + 142Ce, 32S + 138Ba, and 48Ti + 122Sn) that lead to the same compound nucleus, 170Hf.1 This work was done as part of a program to search for entrance channel e ects in sub-barrier fusion.2 These fusion cross sections were measured in a collaborative e ort between the Nuclear Physics Laboratory and the Tandar Laboratory in Buenos Aires, Argentina. The evaporation residue (ER) cross section was measured at the Tandar Laboratory using a delayed X-ray technique that they perfected.3 This technique is able to determine the absolute ER cross section as well as the partial cross sections for the individual ER channels from the X-rays emitted by the radioactive decay of the ER's. The extracted cross sections are sensitive to the decay properties of the ER's and their daughters. Considerable e ort has been devoted to estimating the decay properties for the nuclei produced in these reactions. For the systems 32S + 138Ba and 48 Ti + 122Sn, it was determined that there was a signi cant ssion contribution to the total fusion cross section at energies just above the barrier. The fusion- ssion cross sections for these two systems were measured with beams of 32S and 48 Ti from our superconducting linac for lab energies in the range of 150{165 MeV and 192{220 MeV respectively. The forward fragment was detected at 55 in the center of mass with a E{E telescope. The thickness of the E detector was chosen such that the ssion fragments were stopped in the detector but not the elastic particles. The complementary ssion fragment was detected in a silicon strip detector centered at approximately 125 in the center of mass. The strip detector consisted of seven independent strips. The position of the backward detector was chosen so that all of the complementary fragments were within the active area of the strip detector. The total fusion cross section, ER + ssion, is shown in Fig. 3.5-1 for all three systems studied in the our search. The ssion cross sections are shown as the shaded areas of the bottom two panels of the gure. For the 28Si + 142Ce system the ssion cross section is estimated from PACE calculations that reproduce the ER yields for all three systems and the experimental ssion yields measured here. The lines in Fig. 3.5-1 are coupled channels ts to the fusion cross sections using the code CCDEF.4 The only adjustable parameters in the coupled channel ts was the depth of the nuclear potential and for the 32S and 48 Ti systems the coupling strength of a 1-neutron transfer channel. For each system, the quadrupole and octupole deformations are obtained from tabulated B(E2) and B(E3) values respectively. The experimental fusion cross sections are well accounted for by our coupled channels calculations.  Present address: TANDAR, Departamento de Fisica, CNEA, Buenos Aires, Argentina. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1991) p. 18. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1992) p. 21. 3 D. DiGregorio, Phys. Rev. C 42 2108 (1990). 4 J. Fernandez-Niello, Comp. Phys. Comm. 54, (1989) 409{412. 17

• Page 27

Fig. 3.5.1. Fusion excitation functions for 28Si + 142Ce (top), 32 S + 138Ba (middle), and 48Ti + 122Sn (bottom). Solid curve is coupled channels t to the experimental data using literature values for the deformation parameters. Dashed curve is the same coupled channels calculation with the couplings turned o . 18

• Page 28

3.6 A search for entrance channel e ects in near and sub-barrier fusion J.D. Bierman, A.W. Charlop, Z. Drebi, S. Gil, D.J. Prindle, R. Vandenbosch and D. Ye We have completed our e orts to search for a possible correlation between the mass asymmetry of the entrance channel and the broadening of the compound nuclear spin distribution at energies near and below the fusion barrier. We have measured the fusion cross section and -multiplicities for three systems (28 Si + 142Ce, 32S + 138Ba, and 48Ti + 122Sn) that lead to the same compound nucleus (170Hf) within the same excitation energy range. In a previous report1 we discussed the - multiplicity measurements. The fusion cross sections have been determined for all three systems.2 The mean spin of the compound nucleus was determined using a formula based on the procedure of Halbert.3 The statistical model code PACE was used to estimate the multiplicity of the statistical -rays and the average spin carried o by each evaporated particle and emitted statistical -ray, and the average spin of the ssion branch. The parameters for the statistical model were determined by tting the experimental relative yields of the evaporation and ssion channels over the entire excitation energy range for all three systems simultaneously. Fig. 3.6-1 shows the mean spins as a function of excitation energy for each of the three systems. The solid curves are the predictions of the coupled channels ts to the excitation functions.2 The agreement between the model and the experimental results is excellent. There is no evidence for signi cant changes in the spin distribution of the compound nucleus formed by di erent entrance channels that are not accounted for in the coupled channels model using literature values of the deformation parameters.  Present address: TANDAR, Departamento de Fisica, CNEA, Buenos Aires, Argentina. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1992) p. 21. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1993), sec. 3.5. 3 M.L. Halbert et al., Phys. Rev. C 40, 2558 (1989). 19

• Page 29

Fig. 3.6.1. Mean spin of the compound nucleus as a function of excitation energy for 28Si + 142Ce (top), 32S + 138Ba (middle), and 48Ti + 122Sn (bottom). Solid curve is from a coupled channels t to the experimental data using literature values for the deformation parameters. Dashed curve is the same coupled channels calculation with the couplings turned o . 20

• Page 30

3.7 Scattering of 87 Mev 6;7Li on 12C W.J. Braithwaite, J.G. Cramer, S.J. Luke,y B.T. McLain, D.J. Prindle, and D.P. Rosenzweig We are in the last stages of our study of the scattering of 6 7Li at 10{15 MeV/nucleon. We have ; measured elastic and inelastic cross sections for 87 MeV 6 Li + 12C from 4 to 100 in the center of mass and for 87 MeV 7 Li + 12 C from 4 to 82. We were able to resolve the rst excited state of 7 Li at every angle. Our data extend out far enough in angle to determine unique optical model potentials from the few discrete potentials that t the inner angle data. Fig. 3.7-1 shows our 7 Li cross sections and optical model ts using a shallow, deep, and intermediate depth Woods-Saxon real potential. The shallow potential does not t the large angle data very well and the deep potential provides the best overall t. The situation is reversed in our 6 Li data where the shallow potential provides the best t. Near-far decompositions of the cross sections show that the large angle data is dominated by the far side amplitude and exhibits the airy minima and maxima of a nuclear rainbow. The deep potential that ts the 7Li data gives two or three airy minima while the shallow 6 Li potential gives only one. The transmission coecients from our ts indicate that 7 Li is much more transparent than 6 Li at small impact parameter so the trajectories from the deeper potentials can more easily penetrate the nucleus and interfere with the larger impact parameter trajectories to produce the minima and maxima of the rainbow. The inner angle data, which is sensitive mainly to the potential at or outside the nuclear surface, also exhibit di erences between the 6 Li and 7 Li cases. As shown in gure 3.7-2, at larger radii both the shallow and deep potentials are greater in magnitude for 7 Li than for 6 Li. Also, the forward angle 6 Li t is fairly good and does not improve noticeably with modi cations of the potential, but the 7Li t at forward angles is not as good and can be improved considerably by a slight modi cation of the potential at about 10 fm. We plan to use folding model potential ts to see how these di erences can be related to di erences in the density distributions of 6 Li and 7 Li and compare the results to electron scattering data which show a large tail in the charge distribution of 6Li comparaed to 7 Li. Figure 3.7-1. Elastic scattering angu- Figure 3.7-2. Woods-Saxon real po- lar distribution for 87.0 MeV 7Li + tentials for 87.0 MeV 6 Li + 12 C and 12C with optical model ts. 87.0 MeV 7 Li + 12 C.  Department of Physics and Astronomy, University of Arkansas at Little Rock, Little Rock, AK 72204. y Lawrence Livermore National Laboratory, L-397, Livermore, CA 94551 21

• Page 31

3.8 The APEX experiment T. A. Trainor and APEX collaboration The APEX double-arm beta spectrometer is nearing completion at the ATLAS facility at Ar- gonne National Laboratory. It is intended to investigate the production mechanism for electron- positron pairs observed in connection with elastic nucleus-nucleus collisions for masses near uranium and beam energies near 6 MeV/A. The mechanism for producing these pairs, which appear to be decay products of neutral objects with discrete mass spectra, has remained elusive after ten years of study at GSI by three research groups. APEX is designed to be a kinematically complete spectrometer with high eciency. The heavy ion collision partners are detected in a nearly 4 position sensitive avalanche counter array which determines the kinematics of the collision and serves as a trigger condition. Electrons and positrons produced at the target spiral in a 300G eld to symmetrically placed 36 cm long axial silicon arrays, each with 216 1 mm thick silicon diode detectors. Positrons annihilating on a silicon array produce back-to-back 511 keV gamma rays which are detected with high eciency by NaI barrels surrounding the two silicon arrays. Each barrel is azimuthally segmented into 24 bars, and each bar is a position-sensitive gamma detector with resolution of about 2 cm. Thus, detection of a 511 keV pair in two nearly opposite bars serves to identify and locate a positron hit on the corresponding silicon array. This is the primary trigger condition for a positron event. The hit position on the segmented silicon array combined with the particle energy and time of ight determined by the individual silicon detector serve to characterize completely the kinematics of the detected positron or electron. For each positron identi ed by a NaI barrel-silicon array combination an invariant mass spectrum is formed in combination with each coincident electron on either arm of the spectrometer. The GSI observations imply that a discrete mass spectrum superposed on a combinatoric background should result. Questions to be addressed with APEX include the apparent motion of the hypothetical neutral \source" object, the relative kinematics of the pair (e.g., opening angle distribution), the variation of the mass spectrum with heavy ion collision system and kinematics, and the heavy-ion CM energy dependence of the pair production rate. The APEX-ATLAS combination represents a nearly 20-fold increase in pair detection rate and complete determination of invariant mass spectra. One arm of the spectrometer has now been fully instrumented, and initial beam tests are underway. The accompanying articles describe the progress in further detail.  Argonne National Laboratory, Argonne, IL 60439; Michigan State University, East Lansing, MI 48824; Princeton University, Princeton, NJ 08543; Yale University, New Haven, CT 06520; Florida State University, Tallahassee, FL 32306; and University of Rochester, Rochester, NY 14627. 22

• Page 32

3.9 APEX silicon array cooling system G.C. Harper, D. Henderson, E Roa,y and T.A. Trainor The silicon detector arrays for APEX located in the arms of a symmetric two-arm beta spec- trometer each contain 216 1 mm thick silicon diode detectors. These detectors are required to operate with resolutions of <5 keV in energy and <2 ns in time in order to best meet the physics goals of APEX. The time resolution in particular is sensitive to the detector temperature, and temperatures below 60 C are required to optimize timing performance. Therefore, a cooling scheme was required to achieve these temperatures with reasonable cycle time and minimal additional mass in the region of the detectors. We decided to use liquid nitrogen boilo as the coolant. Liquid nitrogen in a double-walled stainless steel dewar is heated with a 400 W heater. The duty cycle of the heater is servo controlled to maintain a pressure of 5 psig above the liquid, independent of gas ow out of the dewar. The cold gas passes through a cryogenic needle valve and a 2.5 m coaxial transmission line to an end ange on the APEX apparatus. The gas passes through the end ange and into a 4 mm ID G-10 tube which serves as the axis for the 36 cm long G-10 assembly on which the silicon detectors are mounted. The detector array is surrounded by a kapton shroud sealed to the end ange. Gas emerging from the far end of the G-10 center tube returns inside the kapton shroud, passing over the detectors and cooling them. After exiting the shroud the cold gas passes over the signal transmission strip lines from the detectors to reduce the heat load into the array. The gas then leaves the end ange and travels along a second 2.5 m coaxial exit line. The gas leaving this line is warmed to room temperature, passes through a Baratron pressure gauge and motorized Baratron valve assembly and nally into a Leybold DR-25B vacuum pump. The Baratron assembly regulates the pressure above the valve, and hence within the shroud region, to the selected operating pressure (150 or 250 Torr) independent of gas ow. The shrouds are made of 8 micron or 12 micron aluminized kapton. A shroud is constructed on a series of jigs. First a 5 cm diam. tube is formed. The seam is a 1 mm overlap and is 20{ 30 microns thick. The seam and other joints are formed with Armstrong A-12 epoxy mixed in a cryogenic proportion of resin to hardener. The 40 cm long kapton tube is then epoxied on a second jig to an acrylic disk on one end and an acrylic ange with o-ring seal on the other. The jigs insure that the resulting shroud assemblies have very good geometric properties. The shrouds have burst strength exactly corresponding to the kapton sheet tensile strength. For the two thicknesses the burst pressures are 500 Torr and 750 Torr. The shrouds are operated at one third of the failure pressure. The aluminum layer is connected to the APEX vacuum vessel by small strips of aluminized kapton and silver epoxy to provide rf shielding for the detectors. The coaxial lines, a double-walled region at the end ange, the double walls of the nitrogen dewar, and a cold valve box surrounding the cryogenic needle valve and solenoid-driven ASCO  Physics Division, Argonne National Laboratory, Argonne, IL. y Physics Department, Florida State University, Tallahassee, FL. 23

• Page 33

shuto valves are all evacuated to provide cryogenic insulation. These regions are also wrapped with several layers of superinsulation to further reduce heat transfer. The major heat load on the gas occurs in the cold valve box, and is dependent on the quality of the insulating vacuum. The performance of the system has improved sharply as small leaks from the shroud region through the end ange assembly into the insulating vacuum have been eliminated. With a nitrogen mass ow of 1 g/s (about 1/3 capacity) the detector array cools down to the desired temperature range ( 80 C) in about 1/2 hour. There is provision to switch from boilo nitrogen to room temperature dry nitrogen to complete a warmup cycle to room temperature in about the same time interval. 24

• Page 34

3.10 APEX monitor detector system T. A. Trainor and S.P. Van Verst Incorporated in the APEX double-arm beta spectrometer is a system of monitor detectors. This system includes two high-resolution ion chambers, two CsI detectors, a parallel-plate avalanche counter (PPAC) and a Ge detector. This system is intended to monitor various beam and target properties The ion chambers are mounted in the vertical plane through the beam and target at lab angles of +/{ 11 degrees. They are housed in 15 cm diam. and 35 cm long stainless steel cylinders. The entrance windows are 220 g/cm2 aluminized mylar. The stopping medium is 25 cm of isobutane at 150 Torr. The entrance apertures are 1.5 mm by 25 mm slots oriented horizontally to minimize kinematic energy spread. The apertures are located at radii of 2 m. The cathode and Frisch grid are separated by 50 mm and are parallel to and symmetrically placed about the aperture slot. The anode is 10 mm below the Frisch grid. The aluminized entrance window is placed at the cathode potential and tilted at 10 degrees from perpendicular to the particle tracks toward the anode to insure complete collection of created charge. The large spacing and symmetry of the cathode and Frisch grid were found to be important for achieving sub 1% resolution because of the problem of completely collecting secondary charge deposited by nite range delta rays from the heavy ion track. The cathode and entrance window assemblies are designed to operate at up to 10 kV in order to provide high drift eld intensities and reduce recombination in the track plasma. During recent tests with 6 MeV/A uranium beam the ion chambers have operated with 0.5% resolution and negligible background. The dominant source of energy spread at this level is still kinematic broadening, due to the 2{3 mm beam spot on target. The remaining sources of spread in the detection process are estimated to make a 0.1{0.2% contribution to the overall resolution. The motivation for achieving high energy resolution is the need to monitor carefully the e ective beam energy on target and the state of the target when the beam intensity is limited by target sputtering and melting. The GSI experimental results indicate that the electron pair production depends sensitively on bombarding energy, but target instabilities made detailed studies of this aspect problematic. The two ion chambers in the vertical plane combined with low resolution CsI detectors in the horizontal plane (also at 11 degrees) provide a monitor of the beam position on target through ratios of the counting rates. Precise information on the beam position is essential in order to make inferences on particle kinematics and angular correlations of electron pairs. A single PPAC at 11 degrees serves as a beam time structure monitor. The PPAC is a 1.25 mm gap operating with 5 Torr of isobutane at 430 volts. The gap area is 20 mm diam. and the electrodes are 220 g/cm2 aluminized mylar with additional 2.5  mylar pressure windows. The rise time after one stage of Philips 776 preampli er is 1.5 ns, with a S/N ratio greater then 25.  University of Virginia, Department of Physics, Charlottesville, VA 22901. 25

• Page 35

The detector-rf timing resolution is estimated to be better than 100 ps, and the typical optimized ATLAS beam time spread is observed to be 500 ps FWHM at APEX. 26

• Page 36

3.11 Initial operating experience with APEX T. A. Trainor and the APEX collaboration In February of this year APEX was operated with about 1 pnA of 6 MeV/A uranium beam from ATLAS and with representative parts of all major components in position. This marked the transition from the construction phase to the data acquisition phase for this spectrometer and also signaled a major accomplishment for the ATLAS uranium beam program. Tests in late Summer and Fall, 1992 centered around beam optics, using intermediate mass test beams, and nally the rst low-intensity uranium beams. The beam transport system was studied and optimized. Beam spot sizes and beam halos were measured. Background gamma-ray levels from the beam dump, especially into the NaI arrays, were also investigated. Monitor detector energy and time spectra were obtained with uranium beam, and the silicon array cooling system was successfully tested with representative silicon detectors in place on an array. Representative elements of the heavy ion counter array were operated with beams. Successful completion of these various diagnostics opened the way to commencement of the experimental program. For the February, 1993 run APEX was operated single-ended, with one silicon array partially lled with detectors. The degree of instrumentation was limited essentially by the delivery schedule of commercial electronic components. In addition, the full heavy ion detector array and a complete NaI barrel were operational. This was also a rst test of the back-to-back 511 keV gamma topology trigger in combination with the heavy ion array and silicon array. Much of this initial run period was devoted to a step-by-step check of each detector system for cabling errors, gas leaks, software glitches and defective elements. The trigger system was analyzed in terms of expected vs observed accidental coincidence rates. Toward the end of the period all systems were run conjointly to examine the positron detection process. With the short time remaining this was not an attempt to observe details of positron spectra but rather a rst investigation, with real hardware and beam, of background processes and eciencies. Given the complexity of the APEX spectrometer, the results achieved during this rst run period are very satisfying. The various systems have performed as expected or better, and the stage is set for an aggressive data acquisition schedule in the future.  Argonne National Laboratory, Argonne, IL 60439; Michigan State University, East Lansing, MI 48824; Princeton University, Princeton, NJ 08543; Yale University, New Haven, CT 06520; Florida State University, Tallahassee, FL 32306; and University of Rochester, Rochester, NY 14627. 27

• Page 37

• Page 38

4.2 A test of the equivalence principle for ordinary matter falling toward dark matter E.G. Adelberger, B.R. Heckel, G.L. Smith and Y. Su The observed centripetal accelerations of stars in spiral galaxies are much greater than can be accounted for by the gravitational attraction of the visible matter in the galaxy. This has led to the conclusion that  90 % of the mass in galaxies is non-luminous dark matter, apparently in some exotic form. This conclusion rests on the very reasonable assumption that gravity is the only long- range force acting between ordinary matter and dark matter. We have checked this assumption using the Eot-Wash rotating torsion balance. The results have recently been published.1 We measured the di erential acceleration of two di erent test-body pairs toward the center of our galaxy. By comparing this di erential acceleration to the acceleration attributed to the Galactic dark matter, aDM gal  5  10 9 cm/s2, we nd the Equivalence Principle parameters: DM (Be; Cu) = a(Be; Cu)=aDM gal = (0:0  1:2)10 3 DM (Be; Al) = a(Be; Al)=aDM gal = (+0:7  1:4)10 : 3 If we separate the acceleration of ordinary matter due to dark matter into its gravitational and non-gravitational components, aDM DM DM gal = ag + ang , then our result probes the di erential contribution of aDM ng for two materials (Be and Al, or Be and Cu). This can be related to the total non-gravitational acceleration of ordinary matter toward dark matter if we assume the non- gravitational interaction couples to a charge q5 , that is carried by both ordinary and dark matter. Then aDM ng DM hq5 =mi DM =  (q =m) ; agal 5 where hq5 =mi is the average of the two test body q5 -to-mass ratios and (q5 =m) is their di erence. (Note that this relation is independent of the q5 -to-mass ratio of dark matter.) To proceed farther one must evaluate the q5 =m ratios for our two test-body pairs. In ref. 1 we consider various plausible tree-level values of hq5 =mi=(q5=m) and show that except for a small region around q5 / mf , where mf is the mass of a fermion constituent of our test bodies (e.g. e, p, or n), an EP-violating acceleration toward dark matter greater than 1/10 that of gravity is ruled out by our results. Even in this region where our sensitivity is poorest we nd aDM ng =aDM gal = (+0:11  0:39), DM which rejects by 2 the hypothesis that agal is predominantly non-gravitational. In summary, we have provided laboratory evidence for the usual assumption that gravitation is the only signi cant long-range interaction between dark matter and ordinary matter. 1 G. Smith et al., Phys. Rev. Lett.70, 123 (1993). 29

• Page 39

4.3 Improvements to the rotating source experiment E.G. Adelberger, T. Bast, J.H. Gundlach, J. Haeuser,y M.G. Harris, B.R. Heckel, G.L. Smith, H.E. Swanson, H. Vijaz Construction of the rotating-source torsion-balance has been completed and we are continuing to improve its performance. The leading cause of systematic e ects is the gravitational coupling of the pendulum to the 3 tonne source mass located just 10 cm from the pendulum. The source contains 120 depleted Uranium blocks which were rolled and sheared to trapezoidal shapes. The tolerances on these pieces exceeded speci cations making it dicult to achieve a homogeneous mass density. The blocks were removed and individually weighed, sized, and pressed at. The source was then rebuilt using jigs to obtain a considerably more uniform mass distribution. The Q21 gradient was reduced by a factor of ten. The remaining Q21 and Q44 gradients were nearly canceled using compensator masses. The L = 3; jM j = 1 moments of the source and pendulum were measured with special test sources and test masses. The L = 5 gravitational coupling (Q55q55 ) was utilized to align the turntable center of rotation with the torsion ber to better than 0:001". (The pendulum q55 moment arises from the q44 moment displaced from the center.) A magnetic damper, consisting of a copper disk in a cylindrically symmetric magnetic eld, was added to the top of the ber. The swing and wobble modes of the pendulum were greatly attenuated with little e ect on the damping of the torsional mode. Damping is required to run at high vacuum or when the isolation table is oating. Compensator masses on the pendulum were originally designed to cancel the q20 moment of the entire pendulum, but not the q40 moment. We experimented with compensators which also canceled the q40 moment, as a tilt of the pendulum would produce a spurious signal (because q40 rotates into q41). These compensators resulted in data with 3 times the statistical uctuations of the original ones and were not used. We developed a set of computer programs to investigate the gravitational torques felt by the pendulum. These run on a 33MHz 486 PC and were written in Turbo Pascal and Fortran. MULTI computes spherical multipole moments and gradients for both gravitational and Yukawa interac- tions. Its input is a descripter le which gives the mass con guration (pendulum or source) in terms of basis shapes that can be translated and rotated about the origin. Torques are computed on a multipole by multipole basis. MOVE calculates the e ective multipole distribution resulting from a rotation or a displacement of the pendulum's origin relative to the source's. The descriptor le can generate an AUTOCAD drawing for visual checks of the object. These programs have proven useful in designing test bodies with speci c multipole moments and for estimating torques that arise from imperfections in the pendulum or the source. We have locked the angular velocity of our source mass to a crystal reference oscillator. This gives us the capability of rotating the source mass in resonance with the torsion pendulum.  Johannes Gutenberg Universitat, Mainz, Germany. y Justus Liebig Universitat, Giessen, Germany. 30

• Page 40

4.4 New constraints on composition-dependent interactions with ranges down to 1 cm E.G. Adelberger, J.H. Gundlach, M.G. Harris, B.R. Heckel, G.L. Smith and H.E. Swan- son We have used our new rotating-source torsion-balance to set limits on fundamental, composition- dependent interactions with ranges, , down to 1 cm (mb = 20 eV). The results are particularly sensitive to interactions coupled to the third component of isospin, I3 , but establish new limits for couplings to B and L as well. The apparatus is described in previous Annual Reports; an update on recent developments is given in this volume. We have collected 24 days of data with a Pb/Cu detector dipole, during which we made two reversals of the dipole on the pendulum and one reversal of the Uranium source on the turntable. To limit the potentially largest systematic error, from gravitational quadrupole gradients, we measured the pendulum mass multipole moment q21 and the source eld, Q21, whenever the con gurations were changed. This was done using a known Q21 arrangement of the source or special q21 testbodies respectively. We found no signi cant systematic e ect at this multipole order. We tested for other potential systematic e ects by exaggerating the assumed source (thermal, magnetic, electric, tilt). Linear extrapolations to the measured level at which these quantities varied at the rotation frequency yielded negligible systematic e ects. Combining all data, we obtain a preliminary value of 6:1  8:0 nrad for the pendulum de ec- tion amplitude properly correlated with the position of the Uranium source. The error is purely statistical, we infer that systematic errors are negligible at this scale. We do observe an  76 nrad signal that is independent of the dipole orientation on the pendulum tray. Data taken with the two dipole con gurations, the two source orientations on the turntable, and with two orientations (180 apart) of the pendulum tray in the vacuum can, show that the common-mode signal depends on the relative orientations of the pendulum tray (not the dipole) and the Uranium source. This is consistent with a small deformation of the pendulum tray coupled to a residual gravity gradient of the source, and is being investigated further. The common-mode signal does not contribute substantially to our error as it is subtracted away by our various con guration ips; furthermore it does not point toward the center of the source. p Our data limit the strength of an interaction coupled to a charge q5 = (B 2L)= 5 to 5 = ( 7  9)  10 6 for  > 1 m and 5 = 0:7  0:9 for  = 1 cm.1 The short- limits are particularly interesting because astrophysical constraints on exotic interactions are not restrictive in this region (the Turner window2 on axions is 10 3 10 6 eV). Long-range 1=r3 QCD forces as suggested by Feinberg3 are also signi cantly constrained by our results. In addition, we set constraints in a regime where experiments using the Earth as a source are relatively insensitive ( = 10 1000 km). We are currently running near the Brownian noise limit of the pendulum and plan to operate at a higher vacuum to reduce this noise. 1 q5 and5 are de ned in Phys. Rev. D 42 (1990) 3267. 2 M.S. Turner, Physics Reports 197 (1990) 67. 3 G. Feinberg, Comments Nucl. Part. Phys. 19 (1989) 51. 31

• Page 41

4.5 Development of a spin-polarized torsion pendulum E.G. Adelberger, M.G. Harris and B.R. Heckel We are developing a new torsion pendulum to test for a spin-dependent potential of the form V (r) = ^  r^[m=r + 1=r2]e mr , which would arise from the exchange of a low-mass particle (mass m) that was a CP-violating scalar/pseudoscalar mixture.1 Such a potential could arise from a CP- violating axion exchange and would be intriguing because it violates CP symmetry on a macroscopic scale. To obtain a pendulum with a spin-dipole moment while minimizing the systematic errors that result from the accompanying magnetic dipole elds we have fabricated several magnetized tori. Each is composed of a semi-torus of SmCo joined to a similar semi-torus of AlNiCo which is magnetized in situ to produce a uniform internal magnetic eld, and a minimum external leakage' eld. Since the relative contributions to the eld from spin- and orbital-angular momentum of the two materials are di erent, we obtain a net electron spin polarization. In an e ort to further reduce the external magnetic elds, four such rings will be stacked in the pendulum with successive rings ipped about their spin axes in an A-B-B-A pattern. This operation preserves the spin direction but reverses the internal magnetic elds, thereby raising the multipolarity of the external elds. With an internal eld of 6000 Gauss we have achieved leakage elds of less than 1 Gauss, 2 cm from the 4 cm diameter ring. This corresponds to a magnetic dipole moment (to which our apparatus is sensitive) of 0.06 erg/Gauss. We plan to place the pendulum in our rotating-source apparatus.2 The paramagnetic susceptibility of the three tonnes of uranium rotating in the Earth's magnetic eld will cause a rotating eld of at most 10 6 Gauss. With magnetic shielding this will result in de ections in the micro-radian region, larger than the signal-to-noise limitations of our detection system. This corresponds to measuring an interaction potential of some 10 22 eV/spin and would constitute the most sensitive test for an electron spin coupled force, comparing favorably with Wineland and Ramsey's result3 of 2  10 19 eV/spin for nuclear spins. Improvements will come from further reduction of the magnetic dipole moment, cancellation of the Earth's magnetic eld, and actual corrections for the magnetic torque. This will be done by exaggerating in turn the external eld and the dipole moment of the pendulum, which will permit us to measure accurately the magnetic properties of the detector and source. 1 J.E. Moody and F. Wilczek, Phys. Rev. D 30, 130 (1984). 2 See J.H. Gundlach and G. Smith, Improvement to the rotating source apparatus in this Annual Report. 3 D.J. Wineland and N.F. Ramsey, Phys. Rev. A 5, 821 (1972). 32

• Page 42

4.6 Design of an apparatus to measure the PNC spin rotation of transmitted cold neutrons in a liquid helium target E.G. Adelberger, B.R. Heckel, S.K. Lamoreaux, D.M. Marko , H.E. Swanson, and Z. Zhao Detailed design work has continued and construction has begun on the apparatus to measure the parity non-conserving (PNC) spin-rotation of transversely polarized neutrons through a liquid helium target. The motivation for this experiment|to improve the experimental constraints on the isovector pion-exchange amplitude in the meson-exchange potential that describes the weak inter- action between hadrons|and its overall design have been discussed in previous Annual Reports.1 Oxford Instruments, Inc. is designing to our speci cations a cryostat that contains a central cold bore on which we will mount an insert containing the liquid helium targets, the target-gas transfer system, and the single central  -coil. This removable insert will separate the liquid helium target medium from the liquid helium cryogenic coolant, thus reducing the frequency of lling the target helium reservoir. The insert system is suciently exible that it can be used in future spin- rotation experiments with other cold targets (such as para-hydrogen). The target chambers have been built; construction of the remaining parts of the insert awaits our review of the nal cryostat drawings to assure compatibility. Our design for the  -coil has changed since last year's Annual Report. The coil now consists of two solenoid coils placed side-by-side with their symmetry axes, the eld axis, in the vertical direction so that the elds are in opposite directions in each coil. The return consists of three radially concentric toroidal coils that guide the eld from one solenoid coil to its neighbor. The windings and sizes were determined using a computer calculation. Leakage elds and deviations from the eld in the vertical direction should be reduced to less than one part in 10 3 while maintaining reasonably practical dimensions for construction. The input and output coils will use the previously developed three-solenoid design (a central solenoid coil with two half-width coils on both sides to serve as the return). Test coils measurements show that adding -metal at the ends of the coils to guide the return eld gives a ve-fold reduction in the leakage elds. This feature will be incorporated into the design. The detector is a segmented ionization chamber run in integration mode to collect the 108 neutrons per second expected from the NIST reactor in Gaithersburg, Maryland. With a segmented detector that separates neutrons of di erent velocity groups, we take advantage of the fact that the neutron spin rotation due to magnetic elds is velocity dependent, while the desired PNC e ect is independent of velocity. We will minimize the magnetic eld in the target region by adjusting external eld coils (previously described)2 until each velocity group sees the same rotation. The apparatus is designed to reduce false signals to less than 10 8radians. To check for system- atic e ects, the collecting plates of the detector will be divided into four sections so that neutron ux and apparent spin rotation can be studied as a function of detector region. By comparing re- sults from the di erent detector regions for the front and back target positions, we hope to identify  Department of Physics, University of Washington, Seattle, WA 98195. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1987) p. 27; (1989) p. 18 and (1990) p. 33. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1992) p. 25. 33

• Page 43

any target-dependent signals that mimic the PNC e ect. 34

• Page 44

4.7 Isovector radiative decays of the 16.6 and 16.9 MeV doublet in 8 Be E.G. Adelberger, L. DeBraeckeleer, J.H. Gundlach, M.S. Kaplan, D.M. Marko , A.M. Nathan, W.R. Schief, K.A. Snover and D.W. Storm It is presumed that in the limit of good isospin both the conservation of the vector current and the absence of second class currents are exact symmetries1 2 , but they are broken by quark mass ; di erences. Thus there is much interest in testing these symmetries with precision enough to probe the level of expected isospin violating e ects3 . Our goal is to make a test of CVC in mass 8 nuclei with improved precision. To make an improved test in mass 8, the width of the analog isovector M1 transition and the E2/M1 ratio must be remeasured with better accuracy. These quantities are determined from measurements of the angular distributions of the photons emitted in the reaction 4 He( , ) at the energies of the 16.6 and 16.9 MeV states and from the measurement of the excitation function across both resonances. Previous measurements4 5 of the angular distributions were done with a long gas cell in order ; to shield the detector from the background generated by the cell windows. However, several disad- vantages are incurred by this approach. Among them is the fact that the -ray collimators must be di erent for each angle, thus making the calculation of the correct angle-dependent solid angle a possible source of error. We have taken a di erent approach, using a short gas cell. To eliminate the background from the windows, we measure photon spectra with the cell lled alternately with helium and hydrogen, to match the dE/dx of the He gas, and subtract the hydrogen spectra from the helium. The problems of the long cell are thus avoided. One systematic correction we must make to the data is for incomplete beam bunching by our LINAC. Without an analyzing magnet, we monitored the beam energy at frequent BIC intervals by measuring the energy of 's elastically scattered from a thin gold foil. We have measured the angular distributions at both resonances and the excitation function across both resonances. Preliminary analysis shows a2 = 0.75  .04 on the upper (16.9 MeV) resonance, and a2 = 0.25  .06 on the lower (16.6 MeV) resonance, in disagreement with previous measurements4 5 . Since the predominant decay resonance is isovector M1, and pure M1 decay would ; result in a2 = 0:5, our angular distribution results indicate that the dominant E2 component has opposite sign on the two resonances and hence is predominantly isoscalar. Final analysis of the data, including the excitation curve, is presently under way.  Department of Radiology, University of Washington Medical Center, Seattle, WA 98195. y Physics Department, University of Illinois at Urbana-Champaign, Urbana, Illinois 61820. 1 E. Commins and P. Bucksbaum, Weak interactions of Leptons and Quarks (Cambridge University Press, Cam- bridge, England, 1983). 2 E. Henley and L. Wolfenstein, Phys. Lett. 36B, 28 (1976). 3 A Long Range Plan for Nuclear Science, a report by the DOE-NSF Nuclear Science Advisory Committee, 1983, pp. 25{27. 4 A.M. Nathan et al, Phys. Rev. Lett. 35, 1137 (1975). 5 T.J. Bowles and G.T. Garvey, Phys. Rev. C 18, 1447 (1978). 35

• Page 45

4.8 Improved limits on scalar currents in weak interactions E.G. Adelberger Although there is much evidence for the V A form of the charged weak current, the constraints on scalar couplings that would arise if a massive charged scalar boson were exchanged instead of the W  are surprisingly poor.1 This occurs because the scalar couplings must be inferred from observables (particularly the e- correlation) in which they enter quadratically, unless one makes restrictive assumptions about their parity or time-reversal properties. Furthermore, the scalar exchange amplitudes participate only in Fermi decays, and there have been no measurements of the e- correlation in a pure Fermi transition. The e- correlation is usually measured by the lepton-recoil e ect on the energy of the daughter nucleus. This is a formidable task because of the low recoil energy (of order 100's of eV) and possible distortion of the recoil-energy distribution by energy loss and chemical e ects. It has not been appreciated that -delayed proton emitters provide a powerful tool for determining the e- correlation in pure Fermi decays. I have extracted new limits2 on scalar weak coupling from recent data of Schardt and Riisager3 on the shapes of the -delayed proton peaks corresponding to the superallowed decays of 32 Ar and 33Ar. Because of the CM-to-lab transformation, the -delayed protons are given an energy spread that exceeds the energy of the recoiling nucleus by a factor of roughly 50. Thus the dicult problem of measuring the 600 eV energy of a recoiling 32 Cl ion has been transferred into the much easier problem of measuring the 30 keV spread in energies of a 3 MeV proton group. Because their proton decays are isospin-forbidden, the natural widths of the superallowed proton groups are negligible (100 eV). On the other hand, the proton decay occurs very rapidly compared to the slowing-down time of the recoil ions, so that the proton energy spread is not a ected by energy loss of the recoil ions. Schardt and Riisager's data yields a value aF = 1:016  0:036 (1) for the e- correlation coecient in a Fermi transition, where the error is 2 . (In ref. 2, I show that the Gamow-Teller component of the 33 Ar superallowed decay has no signi cant e ect on the extracted value of aF .) This establishes a 2 limit on the scalar exchange amplitudes of jCS j2 + jCS0 j2  1:0  10 2 (2) jCV j2 + jCV0 j2 where I employ the conventional notation4 for the scalar and vector amplitudes. This result repre- sents a substantial improvement over previous constraints on both time-reversal even (ref. 1) and time-reversal odd5 scalar weak amplitudes. Details may be found in ref. 2. Together with D. Schardt, J. Sromicki and K. Riisager, I am planning a new experiment at the ISOLDE on-line isotope separator at CERN to obtain substantially improved data on the e- correlation in the superallowed decays of 32Ar and 33Ar. 1 A.I. Boothroyd et al., Phys. Rev. C 29, 603 (1984). 2 E.G. Adelberger, submitted to Phys. Rev. Lett., January 1993. 3 D. Schardt and K. Riisager, Zeitschrift fur Physik A, to be published. 4 J.D. Jackson, S.M. Treiman and H.W. Wyld Jr., Nucl. Phys. 4, 206 (1957). 5 M.B. Schneider et al., Phys. Rev. Lett. 51, 1239 (1983). 36

• Page 46

• Page 47

Arctic regions, concentrating initially on the migrations of spruce and alder across Alaska in the early to mid-Holocene. This research is funded in part by a NSF grant to P.M. Grootes and G.W. Farwell under the Paleoclimate of Arctic Lakes and Estuaries Initiative of the ARCSS Program (Grant No. ATM 91-23963). 5.1.2 Atmospheric methane We have continued our time-series measurements of the 14C concentration of atmospheric methane in collaboration with Paul Quay, School of Oceanography, University of Washington. This research is supported in part by a NASA grant to Dr. Quay (Grant No. NAGW-844). Preliminary analysis of our most recent measurements on samples from the clean air site at Cheeka Peak on the coast of the Olympic Peninsula, Washington, gives an average value for 1991 of about 125 percent Modern Carbon (pMC) and shows some evidence of a small increase in the 14C concentration of atmospheric methane during the period from 1989 to 1992 (it is not clear at this stage if this increase is statistically signi cant given the scatter of the measurements over that period). Signi cantly lower pMC values have been reported previously for atmospheric methane in the southern hemisphere4 and there has been a suggestion of a signi cant increase (on the order of 1 pMC/year) in the 14C concentration of atmospheric methane in the southern hemisphere since 1990. Our time-series data coupled with the results from the southern hemisphere and new data from other groups in the northern hemisphere should add signi cantly to our understanding of the sources, sinks and atmospheric transport of methane. As a part of this project we measured the 14 C concentration of an interlaboratory compari- son methane sample; this sample is also being measured by three other groups who are actively involved in developing time-series records of the 14 C concentration of atmospheric methane. We obtained a value of 121.1  0.8 pMC for the intercomparison sample. Unfortunately, the other three laboratories have not reported their results at this time. 5.2 Technological program 5.2.1 Performance of the HE beam transport system and the wide-aperture detector The wide-aperture 14C detector and associated modi cations to the high energy (HE) beam trans- port system (implemented in late 19915) have signi cantly reduced the sensitivity of our measure- ment system to uctuations in the accelerator terminal voltage. Modi ed tuning of the HE beam transport system to take full advantage of the large acceptance window of the detector, together with improved uniformity in the magnetic eld of the \Wien lter" velocity selector (achieved by the addition of steel shims outside the beam tube), has resulted in a wide and at terminal voltage transmission plateau at the detector (typically 24{28 kV). The reduced sensitivity to ter- minal voltage uctuations that results at both the detector position and the image position of the 90 analyzing magnet has signi cantly decreased the diculties encountered in tuning the beam through the HE transport system to the detector and has improved the sample wheel-to-sample wheel reproducibility of measurements obtained with our AMS system. 4 D.C. Lowe, C.A.M. Brenninkmeuer, S.C. Tyler and E.J. Dlugkencky, J. Geophys. Research 96, 455 (1991). 5 Nuclear Physics Laboratory Annual Report, University of Washington (1992) p. 34. 38

• Page 48

During the year we found that the slits and slit box at the object position of the 90 analyzing magnet were all out of alignment; the most signi cant misalignment was that the top slit was 0.15 inches below the center line when its position indicator showed it as being centered. This particular misalignment seems to have been the cause of considerable diculties in trying to tune the ion beam through the HE transport system with a minimum of vertical steering during previous AMS runs. We have now aligned all of the elements in the object slit box properly. Reorganization of the detector electronics to optimize the signal-to-noise ratios of our E and E detectors and eliminate parasitic ground-loop noise has resulted in a signi cant improvement in the performance of our detector telescope. Under current operating conditions, the FWHM of the E peak is 0.28 MeV (4% of the 7 MeV deposited in the E detector), the FWHM of the E peak is 0.40 MeV (1.5% of the 28 MeV deposited in the E detector), and the FWHM of the Etotal peak is 0.28 MeV (0.8% of the 35 MeV total energy of the 14 C4+ ions). The improved detector resolution increases our ability to discriminate against scattered non-14 C4+ ions which may, by chance, produce E and E signals close to those detected from 14 C4+ ions. The variations in the energy deposited in the E detector (0.28 MeV FWHM) can be largely attributed to the outward bulge of the mylar window of the detector under the typical counting-gas pressure of 200 Torr. The variations in the energy lost by the ions in the E detector gas then appear as a contribution to the width of the E peak (0.28 MeV out of the total width of 0.40 MeV; the other 0.28 MeV component of the E peak width is probably due to the resolution of the E detector). Hence, the 0.28 MeV FWHM value for the Etotal (the sum of E and E) peak shows that the E and E values are correlated (extra energy detected in the E detector due to increased path length in the gas results in decreased energy being detected in the E detector for that ion) and that the variations in path length in the E detector do not contribute signi cantly to the observed peak width of our Etotal measurements. 5.2.2 Alteration in the tuning of the LE beam transport system During the last year we tested and implemented two changes to the low energy (LE) beam transport system. Firstly, we removed the 5/32 inch diameter LE aperture that we had previously inserted at the image position of the in ection magnet and now use the ion-source einzel lens to maximize the C beam current at the LE Faraday cup. This change has increased the C ion transmission from the source to the LE Faraday cup by about 35% and altered the beam pro le so that transmission is insensitive to small di erences in beam trajectory between samples. Secondly, we operated our measurement system without inserting the grid lens at the entrance to the accelerator beam tubes. This has eliminated the 10% beam loss on the grid of the entrance lens and also reduced the sensitivity of ion beam transmission to small di erences in beam trajectory. Together these changes have increased the beam current transmitted from the ion source through the tandem by about 50% and reduced the sensitivity of our system to potential di erences between samples. 5.2.3 Current performance of the measurement system The current performance of our 14C AMS measurement system can be summarized as follows: 1. The ion source is operated at low to medium output levels (20{30 uA 12C ) to ensure contin- uous long term operation; 14 C4+ count rates for approximately modern samples are typically 39

• Page 49

45{60 cps; our typical measurement uncertainty for a routine determination of the 14C/13C ratio of an approximately modern sample is 0.7%, 2. The 14 C/13C ratios obtained for our Chinese Sucrose laboratory standard over the course of a sample wheel (about 20 measurements of the standard for each wheel) have a 1 scatter of about 0.69% while the 1 counting statistics errors for these measurements average about 0.64%; this implies that only 0.26% of the 1 scatter observed in our 14C/13C determinations cannot be accounted for by counting statistics, 3. In recent measurements we obtained percent Modern Carbon (pMC) values of 22.730.15 and 151.10.8 for the IAEA intercomparison materials C-5 and C-6, respectively; these values are in acceptable agreement with the consensus values for these samples of 23.050.02 and 150.610.11, respectively,6 4. In recent measurements we obtained a 14C age of 612050 BP for the approximately one half-life old sample QL11658, which agrees with the 612030 BP 14C age obtained previously by high precision -counting (Minze Stuiver, pers. com.). 5. As part of our pollen study we prepared and measured sputter source target pairs from 27 graphitized CO2 subsamples of our pollen extracts; within statistics, we found no evidence of signi cant di erences between the measured 14C/13C ratios of target pairs; ie., we found no evidence of signi cant target-dependent scatter in the di erences between the 14C/13C ratios for target pairs. 6 K. Rozanski et al. Radiocarbon 34, 506 (1992). 40

• Page 50

6 Medium Energy 6.1 Inclusive pion photoproduction on several nuclei K.G. Fissum, M. Frodyma,y K. Garrow, I. Halpern, H. Kaplan, D.P. Rosenzweig, D.W. Storm, and J. Vogt We measured photoproduction of positive pions on nuclei in order to obtain information about the mean free path of pions in nuclear material. The photons illuminate the nucleus uniformly. Then if we assume the photoproduction proceeds as it would on free (but moving) nucleons, limited only by Pauli blocking, we can calculate spectra. In fact, guided by our scattering studies, we include the possibility that the Delta-like object, produced when a proton absorbs a photon, may de-excite by interacting with a neighboring nucleon rather than by emitting a pion. The pions emerging from the photon interaction site are attenuated, causing a reduction in the magnitude of the cross section. By comparing measured with calculated cross sections, one hopes to determine the e ective cross section for attenuation of pions which can be combined with nuclear density to obtain a mean free path. Since the pions have a broad range of energies (due to the nucleon Fermi motion), these quantities can be obtained as a function of energy. The angular distributions calculated for the photoproduced pions are obtained from the angular distributions for production on free nucleons after accounting for Fermi motion and Pauli blocking. The semi-classical calculation that we have developed for pion scattering is well suited for photoproduction as well. Results of preliminary data analysis were reported last year.1 We have now completed the data analysis and have obtained spectra of  + produced at four angles by tagged photons in the energy range 179 to 217 MeV. The targets were C, Ca, Sn, and Pb, as well as CH2 used for calibration. The detectors were plastic scintillator telescopes. Positive pions were identi ed by detecting the muon from the pion decay. Since we are using the free photoproduction cross section in our calculations and we have mea- sured this cross section, we can compare our results on complex nuclei with the results on hydrogen. We do this by normalizing our measured cross sections by a single common factor, obtained by matching our measured cross sections for photoproduction on hydrogen to the Blomqvist-Laget2 formulation. Then we use that parameterization in the calculation, where cross sections are needed over a range of photon energies, because of the nucleon Fermi motion. It should be possible to determine the absolute cross section without normalizing to hydro- gen measurement, using the measured photon ux and corrections described in last year's annual report.1 (Preliminary results presented last year had an error in the detection eciency, resulting from an incorrect value for the time interval after the pion pulse during which the muon can not be detected.) After applying these corrections we nd that our measured di erential cross sections for photoproduction on hydrogen are about 1/2 those predicted by the parameterization of Blomqvist and Laget. The disagreement between the Blomqvist-Laget formulation and various other mea- surements is around the 10 to 20% level, rather than a factor of two. Thus, it seems that there is  University of Saskatchewan, Saskatoon, S7N 0W0, Canada. y Now at: Stanford Linear Accelerator Center, Stanford, CA 94309. 1 Fissum et al., Nuclear Physics Laboratory Annual Report, University of Washington, p. 39 (1992). 2 I. Blomqvist and J.M. Laget, Nucl. Phys. A280, 405 (1977). 41

• View More