avatar University of Washington Inc Services

Pages

  • Page 1

    INTRODUCTION The Nuclear Physics Laboratory at the University of Washington in Seattle pursues a broad program of nuclear physics research. Research activities are conducted locally and at remote sites. The current program includes "in-house" research on nuclear collisions using the local tandem Van de Graaff and superconducting linac accelerators as well as local and remote non-accelerator research on fundamental symmetries and weak interactions and user-mode research on relativistic heavy ions at large accelerator facilities around the world. Our results on hot Giant Dipole Resonance decay reverse longstanding conventional wisdom that the GDR width saturates at moderate excitation energy. We show, based on new preequilibrium and GDR decay measurements, together with a reanalysis of previous experiments, that the GDR width does not saturate, but appears to continue growing with excitation energy (temperature) in accord with theoretical expectations. These experiments are fundamental to understanding hot nuclear matter, in particular how normal modes of nuclear oscillation survive at high temperatures. In the past year, we completed an extensive series of 4He(a, g) reaction measurements of 8Be analog-state g - decay, together with measurements of various g efficiency calibration reactions. Currently the data analysis is in progress to extract values for the isovector decay observables to be used in a precision A = 8 CVC/SCC test. When our results are combined with other experiments, we expect a sensitivity in the A = 8 system within a factor of 2 of that achieved in the A = 12 system by the Osaka group, who recently reported an exciting, small positive result. The excitation energy dependence of the nuclear level density parameter has been studied by measuring evaporative proton and alpha spectra as a function of bombarding energy. The energy dependence observed is quite weak, but consistent with a recent theoretical prediction. We have achieved a major milestone in our project to measure the S-factor for the 7Be(p,g)8B reaction. This S-factor is essential to our understanding of the solar production of energetic neutrinos. We have our experimental apparatus constructed and mostly debugged, and we have taken first data on the Ecm = 630 keV resonance and at 400 and 500 keV with a 13 mCi target. We have also measured the 7Be(a,g) resonance reaction which we use as a target quality diagnostic. The Russian-American Gallium Experiment (SAGE) has submitted an archive paper on the 51Cr neutrino source experiment which is in press. The extraction data through December 1997 have been analyzed with the result that the solar neutrino flux times the Ga cross section is measured to be , (statistical) (systematic) SNU. These data are being prepared for publication. The completion of construction of the Sudbury Neutrino Observatory was marked by an opening celebration April 29, 1998. Water fill began shortly before that, and is essentially complete as of this writing a year later. The water quality has met or exceeded expectations with respect to radioactivity, with the exception of Th in the heavy water, and it is likely that this too will fall within specifications once recirculation begins. Problems with high-voltage breakdown in connectors underwater have been substantially mitigated by re-gasifying the water with nitrogen. During the fill, many detector subsystems were brought to completion. The UW SNODAQ group has created a highly versatile realtime data-acquisition software system that has functioned superbly in the commissioning of the detector. The neutral-current detector (NCD) construction project centered at UW is nearing completion following a difficult, but successful, battle with residual 210Po activity from radon daughters. All detectors, the NCD electronics, and the deployment system are expected to be


  • Page 2

    finished this summer. The emiT detector is in Seattle undergoing major upgrades to address problems encountered in the first run. It is expected that these upgrades will greatly enhance the high-voltage stability of the system, and enable the detector to meet its design goals. The second run of emiT is expected to commence in late 1999. Analysis of the data from the first run is essentially complete and is being prepared for publication. We are preparing for the initial operation of RHIC later this year by preparing for HBT analysis of the data which will be produced by the STAR detector. This work includes an improved numerical procedure for dealing with HBT Coulomb corrections. We have continued our investigations of the phase space density of pions produced in ultrarelativistic heavy ion collisions and have found evidence of a universal scaling at CERN SPS energies. In addition, during the past year the URHI group has made major improvements to NA49 global-variables EbyE analysis and STAR offline computing EbyE analysis software production. We have made several advances in fluctuation analysis theory (finite systems effects and the extension of linear response theory near a phase boundary and scale-dependent differential variance measures). We have made upgrades to STAR TPC tracking software and a generalized data browser and visualizer for URHI-related data (TPC tracking data, event generator output and general data bases). Our equivalence-principle tests are becoming ever more powerful as we continue improving our torsion balances. This work has broad implications for cosmology and for Planck-scale physics-- for example Carroll has estimated that the slowly rolling scalar field scenario leads to "equivalence-principle" violations roughly 104 times larger than our 1994 limit. Our recent work on the strong equivalence principle has attained a differential acceleration sensitivity of 2 x 10-13 cm/s2. Our high-precision measurement of the positron-neutrino correlation in the superallowed decay of 32Ar rules out scalar weak interactions at mass scales up to 4 times the W boson mass. A similar measurement in 33Ar resolves a discrepancy regarding the Gamow-Teller contribution to the superallowed transition. The measurements also demonstrated the power of the technique for making precise measurements of the masses of nuclei far from stability. As always, we encourage outside applications 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 accelerators. For further information, please write or telephone Professor Derek, W. Storm, Executive Director, Nuclear Physics Laboratory, University of Washington, Seattle, Washington 98195; (206) 543-4080, (e-mail: storm@npl.washington.edu) or can also refer to our web page: http://www.npl.washington.edu. 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 to whom inquiries should be addressed underlined. We thank Richard J. Seymour and Karin M. Hendrickson for their help in producing this report. Steve Elliott, Editor Sre@u.washington.edu, (206) 543-9522


  • Page 3

    Barbara Fulton, Assistant Editor


  • Page 4

    TANDEM VAN DE GRAAFF ACCELERATOR A High Voltage Engineering Corporation Model FN purchased in 1966 with NSF funds; operation funded primarily by the U.S. Department of Energy. See W.G. Weitkamp and F.H. Schmidt, "The University of Washington Three Stage Van de Graaff Accelerator," Nucl. Instrum. Meth. 122, 65 (1974). Some Available Energy Analyzed Beams Ion Max. Current Max. Energy Ion Source (particle m A) (MeV) 1H or 2H 50 18 DEIS or 860 3He or 4He 2 27 Double Charge-Exchange Source 3He or 4He 30 7.5 Tandem Terminal Source 6Li or 7Li 1 36 860 11B 5 54 860 12C or 13C 10 63 860 * 14N 1 63 DEIS or 860 16O or 18O 10 72 DEIS or 860 F 10 72 DEIS or 860 * Ca 0.5 99 860 Ni 0.2 99 860 I 0.01 108 860 * Negative ion is the hydride, dihydride, or trihydride. Additional ion species available include the following: Mg, Al, Si, P, S, Cl, Fe, Cu, Ge, Se, Br and Ag. Less common isotopes are generated from enriched material. BOOSTER ACCELERATOR We give in the following table maximum beam energies and expected intensities for several representative ions. "Status of and Operating Experience with the University of Washington Superconducting Booster Linac," D.W. Storm et al., Nucl. Instrum. Meth. A 287, 247 (1990). Available Energy Analyzed Beams Ion Max. Current Max. Practical


  • Page 5

    (pm A) Energy MeV p >1 35 d >1 37 He 0.5 65 Li 0.3 94 C 0.6 170 N 0.03 198 O 0.1 220 Si 0.1 300 35Cl 0.02 358 40Ca 0.001 310 Ni 0.001 395


  • Page 6

    TABLE OF CONTENTS 1.0 FUNDAMENTAL SYMMETRIES AND WEAK INTERACTIONS ...................................................... 1 1.1 Beta delayed alpha spectrum from 8B decay and the neutrino spectrum in 8B decay................................... 1 1.2 Electron-neutrino correlations in 32Ar and 33Ar decays ................................................................................ 2 1.3 Determination of the response function and efficiency for the 3 large NaI spectrometers........................... 4 1.4 Precision measurements of the angular distribution and excitation function of the 4He(α,γ)8B reaction..... 5 1.5 Gamma ray spectrum from the 4He(α,γ)8Be reaction ................................................................................... 6 1.6 Time reversal in neutron beta decay – the emiT experiment........................................................................ 7 1.7 Improvements to the emiT detector .............................................................................................................. 8 1.8 PNC spin-rotation of cold neutrons in a liquid helium target ....................................................................... 9 1.9 Construction and tests of a new rotating equivalence principle test apparatus........................................... 10 1.10 Construction of an apparatus to measure the gravitational constant........................................................... 11 1.11 New result of the Rot-Wash torsion balance .............................................................................................. 12 1.12 An unambiguous test of the Equivalence Principle for gravitational self-energy ...................................... 13 2.0 NEUTRINO PHYSICS ............................................................................................................................... 14 2.1 The Sudbury Neutrino Observatory............................................................................................................ 14 2.2 SNO commissioning activities.................................................................................................................... 15 2.3 The SNO data acquisition system............................................................................................................... 16 2.4 Overview and status of the SNO DAQ SHaRC control software............................................................. 17 2.5 SNO data stream monitoring ...................................................................................................................... 18 2.6 The Neutral Current Detector Project at SNO ............................................................................................ 19 2.7 Electronics for the NCD array .................................................................................................................... 20 2.8 In situ determination of backgrounds from neutral current detectors in the Sudbury Neutrino Observatory................................................................................................................................................. 21 2.9 Using a remotely operated vehicle to deploy neutral current detectors in the Sudbury Neutrino Observatory................................................................................................................................................. 22 2.10 Initial results from the cool-down phase of the Neutral Current Detector (NCD) Program for the Sudbury Neutrino Observatory ................................................................................................................................. 23 2.11 Sensitivity of NCDS to Supernovae during cool-down .............................................................................. 24 2.12 SAGE: The Russian American Gallium Experiment.................................................................................. 25 3.0 NUCLEUS-NUCLEUS REACTIONS ....................................................................................................... 26 3.1 The GDR width in highly excited nuclei .................................................................................................... 26 3.2 High-energy γ-ray emission in 12C + 58, 64Ni reactions at 6--11 MeV/u...................................................... 28 3.3 Scaling properties of the GDR width in hot rotating nuclei ....................................................................... 29 3.4 17O inelastic scattering study of the GDR width in decays of excited 120Sn nuclei .................................... 30 3.5 Investigation of the temperature dependence of the level density parameter: results from the 19 F + 181Ta → 200 Pb system ...................................................................................................................... 31 4.0 NUCLEAR AND PARTICLE ASTROPHYSICS .................................................................................... 33 4.1 The 7Be(p,γ)8B cross section at astrophysically interesting energies.......................................................... 33 4.2 Installation of chamber and beamline for the 7Be(p,γ)8B experiment......................................................... 34 4.3 WALTA: The Washington Large-area Time-coincidence Array ............................................................... 35


  • Page 7

    5.0 ULTRA-RELATIVISTIC HEAVY IONS .................................................................................................36 5.1 Event-by-event analysis overview...............................................................................................................36 5.2 NA49 SCA event-by-event analysis status .................................................................................................37 5.3 NA49 pileup detection and EbyE analysis ..................................................................................................38 5.4 Event by event global variables analysis.....................................................................................................39 5.5 Scaled correlation analysis code update ......................................................................................................40 5.6 STAR mock data challenge: general DST design and the EbyE analysis chain .........................................41 5.7 Finite-size effects and fluctuations near a phase boundary .........................................................................42 5.8 Linear near response coefficients and fluctuations: beyond the gaussian model .......................................43 5.9 Symmetry and the Central Limit Theorem..................................................................................................44 5.10 HBT Physics at STAR.................................................................................................................................45 5.11 Numerical procedures for Coulomb size effects in STAR HBT analysis ...................................................46 5.12 An universal pion phase space density........................................................................................................47 5.13 Data set viewer: A powerful data visualizer for relativistic heavy ion collisions ......................................48 5.14 Shell corrections in stopping power ............................................................................................................49 6.0 ATOMIC AND MOLECULAR CLUSTERS............................................................................................50 6.1 Electron detachment cross sections for carbon atom and cluster anions.....................................................50 6.2 Search for smallest molecular species predicted to form a stable dianion ..................................................51 7.0 ELECTRONICS, COMPUTING AND DETECTOR INFRASTRUCTURE ........................................52 7.1 Electronic equipment...................................................................................................................................52 7.2 VAX-based acquisition systems..................................................................................................................53 7.3 Analysis and support system developments ................................................................................................54 7.4 A custom VME-based data acquisition system for the emiT and SNO NCD experiments ........................55 7.5 Development of an advanced object oriented real-time data acquisition system........................................56 8.0 VAN DE GRAAFF, SUPERCONDUCTING BOOSTER AND ION SOURCES..................................57 8.1 Van de Graaff accelerator operations and development..............................................................................57 8.2 Booster operations.......................................................................................................................................58 8.3 Tandem terminal ion source ........................................................................................................................59 8.4 Cryogenic operating experience..................................................................................................................60 9.0 OUTSIDE USERS........................................................................................................................................61 9.1 10 MeV proton induced cascade solar cell degradation..............................................................................61 9.2 Scientific Imaging Technologies (SITe) CCDs in the space radiation environment..................................62 10.0 NUCLEAR PHYSICS LABORATORY PERSONNEL...........................................................................64 11.0 DEGREES GRANTED, ACADEMIC YEAR, 1998-1999........................................................................66 12.0 LIST OF PUBLICATIONS FROM 1998-1999 .........................................................................................67


  • Page 8

    1.0 FUNDAMENTAL SYMMETRIES AND WEAK INTERACTIONS 1.1 Beta delayed alpha spectrum from 8B decay and the neutrino spectrum in 8B decay E.G. Adelberger, H.E. Swanson and K.B. Swartz* SNO will measure the energy spectrum of 8B solar neutrinos reaching the earth. If neutrino oscillations occur, the spectrum will be distorted from its original shape and the deviation will contain information on the neutrino mixing parameters. Bahcall et al.1 have pointed out that our ability to predict the undistorted shape of the 8B neutrinos is limited by our knowledge of the final state continuum fed in 8B decay. Two kinds of data are useful here, the beta and the beta-delayed alpha spectra in 8B decay. Bahcall et al. showed that existing delayed-alpha data were inconsistent and chose to use a single measurement of the beta spectrum in obtaining a ‘standard’ 8B spectrum. Because of the difficulty of making accurate beta spectrum shape measurements, we have chosen to make a careful remeasurement of the delayed alpha spectra in 8Li and 8B decays, paying special attention to the absolute calibration of the energy scale and to understanding the energy response of detectors. The 8B data were taken during the period covered in the previous annual report.2 Analysis of these data are still in progress. Careful attention is being paid to the absolute energy calibration which is based on the 148Gd, 241 Au and 232Th sources plus a relative calibration using a precision pulser. Source thicknesses were determined from data taken with the normal to the sources at various angles to the detector. Detector dead layer thicknesses were extracted from data when sources were moved in arcs centered on the Si detectors. Stopper foil thicknesses were determined by energy loss of 148Gd α‘s in the foils. Results for the 8B α-spectra will be presented as histograms as a function of Eα with corrections applied for all energy loss processes. * Physics Department, Yale University, New Haven, CT. 1 J.N. Bahcall et al., Phys Rev. C 54, 411 (1996). 2 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 1. 1


  • Page 9

    1.2 Electron-neutrino correlations in 32Ar and 33Ar decays E.G. Adelberger, M. Beck,* H. Bichsel, M.J.G. Borge,† A. García,# I. Martel-Bravo,% C. Ortiz,# H.E. Swanson, O. Tengblad† and the ISOLDE collaboration% Our 1997 ISOLDE measurement of the electron-neutrino correlations in the superallowed decays of 32 Ar and 33Ar was undertaken to search for scalar weak interactions which leave a “gold-plated” signature on the e-ν correlation coefficient a. If scalar couplings vanish, the 32Ar decay will have a=+1, while scalar couplings, whatever their parity or time-reversal properties, lead to a <1. We measured a by analyzing the shapes of the delayed proton peaks following the superallowed decays. (Lepton recoil broadens the 32Ar peak by ~30 keV, while the intrinsic width of the daughter state is only 20+10 eV.) We have continued to refine the analysis of our high-resolution proton spectra.1 Our results are now in their final form and a paper on this work is being submitted. During the last year the Monte-Carlo simulation of the experiment was improved to include the effects of a weak electron capture branch, the order-α radiative corrections2 to the velocity distribution of the recoiling 32 Cl daughter nuclei, and the bending of the protons in the 3.5 T B-field. We fitted our data by folding the results of the Monte-Carlo simulation (which accounted for the mean energy losses of the individual protons from each decay event) with an analytic lineshape consisting of two low-energy exponential tails folded with a ~ ~ Gaussian. The 32Ar and 33Ar data were fitted by varying the weak interaction parameters C s and C'S (defined in Ref. 1) as well as the 6 free parameters of the lineshape (position, magnitude, plus 4 parameters describing the shape). The resulting lineshape was then compared to a “first principles” calculation that included the following effects: 1) the gaussian electronic resolution as measured by a pulser (3.0 keV and 3.3 keV, respectively), 2) a gaussian contribution from electron-hole statistics (Fano factor = 0.1), 3) energy-loss straggling (from both electronic and nuclear collisions) in the 22.7 µg/cm2 catcher foil, 4) energy-loss straggling in the 22.6 µg /cm2 dead layers of the PIN detectors (taking into account the fact that a significant fraction of the energy lost to delta electrons was deposited in the sensitive volume of the detectors), 5) energy lost to phonons rather than ionization of the Si detector, and 6) escape of the Si X-rays (the single escape probability is about 19%). The good agreement of the extracted and calculated lineshapes gave us additional confidence in the analysis. Our 1σ experimental 32Ar constraint (which for the case of vanishing Fierz interference is a=0.9989+0.0052(stat.) +0.0036(syst.) corresponds to a lower limit on the mass of a scalar particle with gauge ~ ~ coupling strength of MS > 4.1MW. Our constraints on C s and C'S are shown in Fig. 1.2-1, along with those from previous studies of Fierz interference3 and the R-coefficient in 19Ne beta decay.4 The electron-neutrino correlation in the superallowed decay of 33Ar probes the Fermi-GT ratio of this 1/2 →1/2+ transition as well as weak-interaction parameters. We assume the Standard Model weak interaction + and use a to measure B(F)/B(GT). Our preliminary 33Ar result, a=0.944+0.002+0.003, disagrees with a previous determination,5 a=1.02+0.02, but is in reasonable accord with a shell-model prediction,6 a=0.93. * Katholieke Universiteit, Leuven, Belgium. † Institute de Estructura de al Materia, SCIC, Madrid, Spain. # Department of Physics, University of Notre Dame, Notre Dame, IN. % PPE Division, CERN, Geneva, Switzerland. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 2. 2 F. Glück, Nucl. Phys. A 628, 493 (1998). 3 W.E. Ormand, B.A. Brown and B.R. Holstein, Phys. Rev. C 40, 2914 (1989). 4 M.B. Schneider et al., Phys. Rev. Lett. 51, 1239 (1983). 5 D. Schardt and K. Riisager, Z. Physik A 345, 265 (1993). 6 B.A. Brown and B.H. Wildenthal, At. Data Nucl. Data Tables 33, 347 (1985). 2


  • Page 10

    Our results also give good values for the widths of the analog states in 32Ar and 33Ar. We obtain Γ(32Ar)=(20+10) eV and Γ(33Ar)= (180+15) eV. This latter result disagrees with the value Γ = 115+15 eV from the 32S( p ,p) reaction.7 We plan to resolve this discrepancy by repeating the 32S( p ,p) measurement at the University of Wisconsin. ~ ~ Fig. 1.2-1. 95% confidence limits on C S and C'S . Upper panel: time-reversal-even couplings. The annulus is from our work; the narrow diagonal band is the Fierz-interference result from Ref. 4. The broad diagonal band shows result from A, B, a and t1/2 in n decay;8 the sausage-shaped area includes, in addition, constraints from positron helicities in 14O and 10C (Ref. 9),9 Fierz interference in 22Na (Ref. 10)10 and a in 6He decays. Lower ~ ~ panel: time-reversal-odd couplings. The circles are from our work and correspond to C S and C'S phases +90º, +45º and -45º. The shaded oval is the constraint with no assumptions as to this phase. The diagonal band is from the R-coefficent in 19Ne decay (Ref. 4). 7 P.M. Endt, Nucl Phys. A 521, 1 (1990). 8 Particle Data Group, Eur. Phys. J. C 3, 622 (1998). 9 A.S. Carnoy et al., Phys. Rev. C 43, 2825 (1991). 10 H. Wenninger, J. Stiewe and H. Leutz, Nucl. Phys. A 109, 561 (1968). 3


  • Page 11

    1.3 Determination of the response function and efficiency for the 3 large NaI spectrometers J.F. Amsbaugh, M.P. Kelly, J.P.S. van Schagen, K.A. Snover and D.W. Storm In our measurement of the radiative isovector M1 and E2 decay widths (see Section 1.4) a set-up consisting of 3 large NaI spectrometer1 is used. Since the γ-rays of interest in the 4He(α,γ)8Be radiative capture reaction have an energy of ≈13 MeV, it is important to know the efficiency η∆Ω / 4 π and response for each NaI spectrometer for γ-ray energies in this range. Last year we reported on an elegant way to measure these quantities simultaneously using the 10 B( He, pγ )12 C reaction2 at E(3He) = 4.1 MeV. The set-up has been improved by using a ∆E − E Si-telescope 3 at 0°, consisting of a 200 µm thick transmission Si-detector and a 3000 µ thick Si(Li)-detector. This allows a better separation of the contribution of the (3He,p) channel in the singles spectrum from the contributions due to (3He,d) and (3He,α). Furthermore, the 13 mg/cm2 Ni stopper foil was moved closer to the target. Both improvements lead to a much better separation of the proton group populating the 15.11 MeV level in 12C in the proton singles spectrum from the proton group from the 12C(3He,p)14Ng.s. reaction. The measured absolute efficiencies η∆Ω / 4 π with a cut on the γ-ray energy of 12 MeV, are 1.49×10-3 (UW), 1.58×10-3 (Illinois) and 2.04×10-3 (OSU) with an error of ±2.4%. In addition, during the He+He runs the relative efficiencies at 15.1 MeV between the different detectors were also measured using the 12C(p,γ)13N reaction at Ep = 14.26 MeV. In the Fig. 1.3-1, a typical γ-ray spectrum gated on protons populating the 15.11 MeV level in 12C is shown. The curve is the result of a response function fit to the data. The response function used in the fit is given by Sandorfi and Collins3 and consists of a gaussian part plus an exponential tail on the low energy side. Using the same set-up, the efficiency and response function have been measured at Eγ = 8.31 MeV using the 13C(3He,pγ)15N reaction. The response for γ rays of 13.3 MeV has been measured by studying the 15 N(p,γ)16O reaction at Ep = 1.26 MeV. The analysis of these reactions is underway, with the goal of determining the dependence of the detector response on gamma energies. Illinois Line, fitted Exp B 15.11 3 10 2 Fit above 12 MeV 10 Counts 1 10 0 10 Delta Y / Error 4 ✟ 2 0 −2 −4 ✁ 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 ✂ ✄ ☎ ✆ ✝ Energy (MeV) ✞ Fig. 1.3-1. Gamma-ray spectrum for Illinois detector gated on protons populating the 15.11 MeV level in 12C. The curve is the result of a fit to the response in the region above Eγ = 12 MeV. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1997) pp. 57-58. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 6. 3 A.M. Sandorfi and M.T. Collins, NIM 222, 479 (1984). 4


  • Page 12

    1.4 Precision measurements of the angular distribution and excitation function of the 4He(α α,γγ)8B reaction J.F. Amsbaugh, M.P. Kelly, J.P.S. van Schagen, K.A. Snover and D.W. Storm A high-precision measurement of the 4He(α,γ)8Be radiative capture reaction has been performed with the purpose of determining the isovector M1 and E2 decay widths for a precision test of CVC and second class currents in the mass-8 system.1 Data have been collected with our 3 large NaI spectrometer set-up2 during two data taking periods. In the measurement a short gas-cell (3.75” diameter) was filled with He gas at 750 Torr. At each beam energy, the background was measured by replacing the He-gas with H2-gas at 675 Torr. This ensures the same energy loss of the beam over the gas-cell as when it is filled with He-gas and therefore a similar background due to the Kapton windows. The beam energy was monitored in between runs using two Si-detectors at ±25° and a thin C scattering foil in a separate scattering chamber upstream. In Fig. 1.4-1, the data collected in the first run are shown. The data points were obtained by integrating the background subtracted spectra over a sliding window with limits which shift proportional to the beam energy. As reference values, limits of 12 MeV and 15.5 MeV at E(3He) = 34.1 MeV were employed. The curves through the data are the result of a simultaneous fit of the angular distribution and the excitation function using a formalism given by Barker3 and specifically Eq. (1) in De Braeckeleer4 for the analytic form. The result for the isoscalar to isovector M1 ratio is ε = 0.00±0.02. For the isovector E2 to isovector M1 ratio δ1 we obtain δ1= 0.03±0.03 while the isoscalar E2 to isovector M1 ratio δ0 = 0.15±0.04, in good agreement with our results from the Mark I experiment (see Ref. 3). Fig. 1.4-2 shows the data collected during the second data taking run, for which the statistics are twice that of the first run. Analysis of the second run as well as a determination of the absolute isovector M1 decay width are in progress. Excitation function ✓ 4 8 He(α,γ) Be ✔ ✒ o ✕ o o 1200 ✑ OSU 40 UW 90 Illinois 140 1000 ✑ Yield [arb. units] 800 ✑ 600 ✑ 400 ✑ 200 ✑ 0 ✑ −200 ✑ 33.0 ✌ 33.5 ✌ 34.0 ✍34.5 ✍ 35.0 ✎ 33.5 ✌ 34.0 ✍34.5 ✍ 35.0 ✎ 33.5✌ 34.0 ✍34.5 ✍ 35.0 ✎ Eα [MeV]✏ Eα [MeV] ✏ Eα [MeV]✏ Fig. 1.4-1. Excitation curves for the 4He(α,γ)8Be reaction measured with the OSU, UW and Illinois detectors. The curves through the date are discussed in the text. Excitation function 4 8 He(α,γ) Be 1600 o o o 1400 OSU 40 UW 90 Illinois 140 1200 Yield [arb. units] 1000 800 ☞ 600 400 200 0 −200 33.0 ✠ 33.5✠ 34.0 34.5 ✡ ✡ 35.0☛ 33.5 ✠ 34.0 ✡ 34.5 ✡ 35.0 ☛ 33.5 ✠ 34.0 ✡ 34.5 ✡ 35.0 ☛ Eα [MeV] Eα [MeV] Eα [MeV] Fig. 1.4-2. Excitation curves for the 4He(α,γ)8Be reaction measured with the OSU, UW and Illinois detectors in the second data taking run. The magnitude of the yields differ due to the detector efficiencies and the reaction angular distribution. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 7. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1997) pp. 57-58. 3 L.D. De Braeckeleer et al., Phys. Rev. C 51, 2778 (1995). 4 L.D. De Braeckeleer et al., Phys. Rev. C 52, 3509 (1995). 5


  • Page 13

    1.5 Gamma ray spectrum from the 4He(α α,γ)8Be reaction J.F. Amsbaugh, J.P.S. van Schagen, K.A. Snover and D.W. Storm In order to determine the isovector M1 width in the transition from the analog of the ground states of 8 Li and 8B to the 3 MeV state in 8Be, it is necessary to remove the effects of transitions to the tails of the resonances. This can be done using the R-matrix formalism in conjunction with a measurement of the gamma ray spectrum at a single angle and beam energy, provided the spectrum can be determined at low gamma ray energy where contributions from the tail become significant. In a previous measurement1 the quantity Rγ, the ratio of the isovector transition matrix elements to the tail and to the state at 3 MeV was defined. Results of that measurement were Rγ = 1.6 ± 1.8 These results disagreed with shell-model predictions,1 which suggested values between −3 and −6 . The poor accuracy resulted from the poor quality of the gamma ray data for energies below about 11 MeV and from limited information on the response function. In order to obtain better spectra, we built the large gas cell described previously,2 and we have fit the data obtained with it using the R-matrix formalism, in conjunction with the improved line shapes described in Section 1.3. The beam energy was integrated across the large gas cell, with photon production weighted by the geometrical acceptance calculated for the shielding geometry. Data above 9.75 MeV photon energy are fit. The fit results and the data are shown in Fig. 1.5-1. The χ2 value is 100.4 for 100 degrees of freedom. The preliminary value of Rγ that we now obtain is −12 ± 1.4. ✜ Long Gas Cell results fitted from 9.8 MeV 200 Counts 100 0 ✖ 7 8 ✗ 9 ✘ 10 11 12 13 ✙ 14 15 ✚ 16 Gamma Energy (MeV) ✛ Fig. 1.5-1. Photon spectrum obtained with the large gas cell, with an incident beam energy of 34.1 MeV. The fit, described in the text, is to the data above 9.75 MeV. The normalization, location of the 3 MeV state, and the ratio Rγ, described in the text, are varied. 1 L. D. DeBraeckeleer et al., Phys Rev C 51 2778 (1995). 2 Nuclear Physics Laboratory Annual Report, University of Washington (1997) p. 7. 6


  • Page 14

    1.6 Time reversal in neutron beta decay – the emiT experiment M.C. Browne, H.P. Mumm, A.W. Myers, R.G.H. Robertson, T.D. Steiger, T.D. Van Wechel, D.I. Will and J.F. Wilkerson The emiT experiment is a search for time-reversal (T) invariance violation in the beta decay of free neutrons. The observation of CP (charge conjugation - parity) violation in the neutral kaon system, coupled with the theoretical expectation that the combined CPT symmetry must hold, indicates that T-violation must occur at some level in weak decays. However, today some 35 years since its discovery, the origin of CP violation is still not well understood. Although CP violation can be accommodated within the standard model of nuclear and particle physics, it may also be an indication of physics beyond the current standard model. The T-violating observables in beta decay are predicted to be extremely small in the standard model and hence beyond the reach of any current measurements.1 But potentially-measurable T-violating effects are predicted to occur in some non-standard models, such as left-right symmetric models, exotic fermion models, and lepto-quark models.2,3 Hence a precision search for T-violation in neutron beta decay allows one to test for physics beyond the current standard model. emiT probes the T-odd triple correlation (between the neutron spin and the momenta of the neutrino and electron decay products) in the neutron beta-decay distribution. The coefficient of this correlation, D, is measured by detecting decay electrons in coincidence with recoil protons while controlling the neutron polarization. The emiT experiment utilizes a beam of cold (<10 meV), polarized neutrons from the Cold Neutron Research Facility at the National Institute of Standards and Technology (NIST) in Gaithersburg, MD. The collaboration includes scientists from Los Alamos National Laboratory, NIST, the University of California at Berkeley/Lawrence Berkeley National Laboratory, the University of Michigan, the University of Notre Dame, and the University of Washington's Nuclear Physics Laboratory. The emiT experiment was installed on the NG-6 beamline at NIST from November of 1996 until September of 1997. A total of roughly 14 million coincidence events were recorded and the maximum sustained coincidence rate observed was ~7 Hz. Analysis of these data is nearly complete and we expect to achieve a sensitivity to D at least comparable to the current combined world average. The sensitivity of the initial run will almost certainly be limited by systematic errors which arose from disabled channels in the proton detector segments. The root cause of this problem was excessive energy loss in the proton detectors and associated dead time, noise, and electronic failures due to high voltage sparks. Work is underway on an upgrade to the apparatus (see the next section), which should solve the problems experienced in the first run. A new measurement is planned at NIST in 2000. 1 M. Kobayashi and T. Maskawa, Prog. Theor. 49, 652 (1973). 2 P. Herczeg, in Progress in Nuclear Physics, W.-Y.P. Hwang, ed., Elsevier Sciences Publishing Co. Inc. (1991) p. 171. 3 E.G. Wasserman, Time Reversal Invariance in Polarized Neutron Decay, Ph.D. thesis, Harvard University, (1994). 7


  • Page 15

    1.7 Improvements to the emiT detector M.C. Browne, H.P. Mumm, A.W. Myers, R.G.H. Robertson, T.D. Steiger, T.D. Van Wechel, D.I. Will and J.F. Wilkerson Technological advances in neutron polarization and an optimized detector geometry should allow emiT to attain a sensitivity to D of 3 10-4, given the current neutron capture flux available at the NG-6 beamline at ✢ NIST (1.4 109 n/cm2s). This level of sensitivity represents a factor of five improvements over previous ✢ neutron T tests, and may permit restrictions to be placed on several extensions to the Standard Model that allow values of D near 10-3. The emiT detector consists of four plastic scintillator paddles for electron detection and four arrays of large-area PIN diodes to detect protons. The eight detector segments are arranged in an alternating octagonal array about the neutron beam. The angle between any given proton detector and it's opposing electron detectors is 135 . This takes advantage of the fact that the electron--proton angular distribution is strongly ✣ peaked around 160 due to the disparate masses of the decay products. When compared to the 90 geometry ✣ ✣ used in previous experiments, an octagonal geometry results in an increased signal rate equivalent to roughly a factor of three increase in neutron beam flux.1 The protons produced in the decay of free neutrons have a relatively low energy. (The Q-value for n- decay is 782 keV, producing protons with energies <751 keV.) While this allows for a delayed coincidence trigger between the proton and electron (eliminating much of the background due to cosmic rays) it makes detection difficult. The PIN diode array and associated electronics are therefore held at -30 to -36 kV to accelerate the protons to a detectable energy. Throughout the first run the PIN diodes exhibited a higher than expected energy loss, effectively pushing the proton signal into the background. Using an energy window which included a reasonable fraction of the decay protons resulted in a dramatically increased coincidence rate. To counter this problem the detector was run at around 36 kV, higher than the nominal design value. Operating at this higher voltage led to more frequent instances of high voltage related breakdown and the associated damage to sensitive DAQ electronics. As a consequence, emiT's forty-eight populated proton detector channels were often not operating simultaneously. As the detector was not fully symmetric, systematic effects were less effectively canceled. In order to assure that the second run is not affected by these problems, a number of detector upgrades are currently in progress at NPL. Upgrades include isolation of the VME shaper ADC cards through the use of analog fiber-optic links, improvements to the DAQ electronics, and considerable changes in the data acquisition software. In addition, replacement of the PIN diode detectors with either PIPS or Surface Barrier detectors is under consideration, and attempts are being made to isolate the locations of high voltage breakdown. It is possible that minor design changes will alleviate this problem. The fiber-optic links are modular in design, allowing for easy replacement of failed parts. The links will isolate much of the DAQ electronics. Damage due to breakdowns is therefore expected to be much less. Sharper thresholds in the proton DAQ shaper ADC boards will reduce background rates, allowing operation at lower voltages. Replacement detectors will have thinner dead layers, also allowing operation at lower voltages. In addition, changes in the DAQ software will allow closer monitoring of the state of the detector. With these changes it is likely that the second run will meet the design goal of 3 10-4. ✤ 1 E.G. Wasserman, Time Reversal Invariance in Polarized Neutron Decay, Ph.D. thesis, Harvard University (1994). 8


  • Page 16

    1.8 PNC spin-rotation of cold neutrons in a liquid helium target E.G. Adelberger, S. Baessler, J.H. Gundlach, B.R. Heckel, D.M. Markoff,* U. Schmidt and H.E. Swanson The major accomplishments of the experiment to measure the parity violating rotation of the polarization vector of a cold neutron beam as it traverses a target of liquid helium are listed below. The goal of the research is to measure the PNC neutron-alpha coupling constant, from which the weak pion-nucleon coupling constant can be extracted. This experiment receives support from the National Science Foundation, as well as from the Department of Energy through the Nuclear Physics Laboratory. The NSF grant was renewed for three years in June, 1998. The renewal was to support the second round of experimental measurements at the NIST reactor. The first measurements were completed in 1997 and proved to be of sufficient quality to warrant further effort. The Spin Rotation collaboration has been enlarged significantly. For the second phase of the experiment, investigators from TUNL and Indiana University have joined the group. New collaborators include M. Snow and G. Hansen from Indiana University, D. Haase and C. Gould from TUNL, P. Huffman from NIST, and U. Schmidt and S. Baessler from the University of Washington. They join the original collaboration of B. Heckel, E. Adelberger, D. Markoff, F. Weitfeldt, S. Penn, J. Gundlach, S. Dewey, and H.E. Swanson. The collaboration has held two meetings during the past year to design the new experiment. The major breakthrough on the experimental side has been the decision to rebuild the cryostat to make the helium targets superfluid. The superfluid density is 20% larger than that of normal liquid helium, making the PNC signal 20% larger, and more importantly, superfluid helium does not support temperature gradients or the formation of bubbles that could lead to systematic errors. During the past year, problematic components of the original cryostat (mostly arising from unreliable commercial glue joints) have been replaced by reliable indium seals. The major task that remains before new data can be taken is to rebuild the helium targets to support the use of superfluid helium. The design of this new target chamber has been agreed upon and drawings are being prepared for our shops. * TUNL, North Carolina State University, Durham, NC. 9


  • Page 17

    1.9 Construction and tests of a new rotating equivalence principle test apparatus E.G. Adelberger, M. Benz, J.H. Gundlach, B.R. Heckel, B.P. Henry, C.D. Hoyle, S. Merkowitz and H.E. Swanson We are building a new rotating torsion balance to search for violations of the equivalence principle due to new fundamental forces with Yukawa ranges >1 m. As described in previous reports, the balance will consist of a composition-dipole pendulum containing titanium, beryllium, or aluminum testbodies. The pendulum is suspended inside a vacuum chamber which hangs from a gimbal attached to a constantly rotating turntable. We are currently operating a test torsion balance that is primarily intended for measuring and nulling the ambient low-order gravity gradients and for debugging the turntable. The balance is surround by a constant-temperature shield inside a hermetically sealed enclosure. The temperature of this enclosure will be actively controlled. The Q21, Q22, Q31 gravity gradients, have been measured by mounting special testbodies on the pendulum. We have installed ~800 kg of machined Pb masses which compensate the Q21 gravity gradient to about 101%. A more precise cancellation will be made once the final local mass distribution is established. The Q21 compensation then allowed us to measure the Q31 gradient and design appropriate compensators. A set of Q22 compensators has been installed. The design for the final pendulum has been finished and is being manufactured by a commercial company.1 The pendulum body will be made from beryllium and carry 8 barrel-shaped testbodies bolted into conical seats. When outfitted as a composition dipole, the pendulum's first non-vanishing m=1 multipole moment nominally occurs at l=7. The m=0 moments, which could turn into a m=1 moment due to an unwanted tilt of the pendulum, vanish up to l=5. The pendulum will have four flat mirrors mounted to its side, one of which will be used for two reflections of the autocollimator light beam using a stationary edge reflector. Our final pendulum design was based on aluminum prototypes manufactured in our shops that were used to test the reproducibility of testbody positioning in the conical seats. The beryllium, aluminum and titanium testbodies have been designed and various prototypes have been built. Special gradiometer bodies for this pendulum have also been designed. We have tested several heat-treating schemes using radiative heating with adjacent hot filaments in an attempt to quickly reduce the fiber drift relaxation rate and the associated torsional noise. Even though several hundred °C were reached in the fiber, a bake of the entire system for one day at ~80° C produced the smallest fiber drift rates. 1 Spreedring Inc., Cullman, Alabama. 10


  • Page 18

    1.10 Construction of an apparatus to measure the gravitational constant E.G. Adelberger, J.H. Gundlach, B.R. Heckel and H.E. Swanson We have built the main components of a rotating torsion balance apparatus dedicated to measure the gravitational constant. The apparatus consists of a flat vertical pendulum suspended from a thin torsion fiber inside a vacuum can located between a set of attractor masses. The vacuum can is mounted on a turntable and initially rotates at a constant speed. A feedback is then turned on that accelerates the turntable to minimize the torsion fiber twist. Therefore the gravitational angular acceleration is directly transferred to the turntable and we have demonstrated1 that this acceleration can be read out with an commercial angle encoder. Since the fiber twist is minimal in our method uncertainties due to in- and anelasticity2,3,4 do not enter directly. The flat vertical pendulum is used to reduce uncertainties due to the mass distribution of the pendulum: the pendulum's mass quadrupole moment, q22, divided by the moment of inertia to which the dominant torque is proportional will become a constant. A certain aspect ratio of a rectangular pendulum and an attractor mass distribution as shown in the figure below and described in more detail in Ref. 1 will eliminate any significant coupling due to other multipole moments. The attractor masses are mounted on a second coaxial turntable. They can be rotated at a different speed and direction and will effectively discriminate against gravitational angular accelerations due to objects in the vicinity. We have built all the major components of the apparatus. The pendulum is made from a 1.5 mm thick, 76 mm wide gold coated glass plate hung from a 17 µm diameter W- fiber. A parallel laser beam from an autocollimator is reflected by a series of mirrors twice from the front and twice from the back side of the plate. The system amplifies the twist angle optically by eight while it is insensitive to rotations about any other direction. The vacuum chamber is pumped by an 8l/s ion pump to the 10-7 torr range and sits on an airbearing turntable. A 36,000 line/rev optical shaft encoder that is read with two readheads was mounted directly to the airbearing. The system is driven by an eddy current motor consisting of a stator from a 400 hz motor and a copper drag cup. The motor is driven with three phases powered with three op-amps. A digital signal processor (DSP) will compute the feedback function directly from the autocollimator signal. Fig. 1.10-1. Schematic of the apparatus. The attractor turntable is finished. It is loaded with eight 125 mm diameter 316-stainless steel spheres weighing ~8 kg each and placed at a radius of 165 mm. Each sphere sits on three stainless steel seats that are mounted in a cast aluminum plate which in turn is supported from a 5cm thick aluminum disk bolted to the turntable. The turntable itself is made from a high precision steel double angular contact trust bearing on which a 18,000 line/rev angle encoder is mounted. A small DC motor is driving the system with a friction drive. The DSP will be used to hold the rotation rate constant. The DSP was interfaced with 8 DAC's and 16 ADC's to read both angle encoders and several other sensors. It will upload the data directly to its PC host computer. After placing the apparatus in its final location in the center of the cyclotron cave, we expect to begin debugging and data taking in the next year. The ultimate precision on G we hope to achieve with this instrument is in the 10-5 range. 1 J.H. Gundlach et al., Phys Rev. D 54, R1256 (1996). 2 K. Kuroda, Phys. Rev. Lett. 75, 2796 (1995). 3 C.H. Bagley and G.G. Luther, Phys. Rev. Lett. 78, 3047 (1997). 4 S. Matsumura et al., Phys. Lett A 244, 4 (1998). 11


  • Page 19

    1.11 New result of the Rot-Wash torsion balance E.G. Adelberger, J.H. Gundlach, B.R. Heckel, C.D. Hoyle, G.L. Smith and H.E. Swanson Since our last report,1 we have reduced the statistical uncertainty of our equivalence principle data and improved our understanding of gravitational systematic effects. Because of the close proximity of the attractor to the pendulum, gravitational effects have always been the major systematic concern with this apparatus. We measure the imperfections of the pendulum by rotating sections of the attractor on lazy susans; we measure imperfections of the attractor by using special gradiometer test bodies. With this information, we may correct our signal for the three lowest order torques in the gravity gradient expansion. To test our ability to make such corrections, we produced a pendulum with large values of the two lowest order gravitational moments and placed it in a source with similarly large gradients. We then proceeded with our normal data-taking protocol. We expected the gravitational corrections to account for the observed torques in this case, since any other perturbations will be tiny compared to the exaggerated gravitational signal. Indeed, after applying our standard analysis, we found the corrected signal to be within 1.1 standard deviations of zero (Fig. 1.11-1). The uncertainty in these corrections is ∼0.7% of the observed signal. Our preliminary equivalence principle results yield a differential acceleration, aCu-aPb, of (-0.7±2.9) × 10-13 cm/s2 toward 238U. For comparison, it would take 11 days at this acceleration for an object’s speed to match that of the continental drift in the east pacific (8.8 cm/yr). We expect to publish a complete description of this experiment with new results in the coming months. Fig.1.11-1. The vectors describing the observed torque on the pendulum and the calculated gravitational corrections under the exaggerated circumstances are shown. The shaded region is the 1-σ uncertainty in the difference. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 15. 12


  • Page 20

    1.12 An unambiguous test of the Equivalence Principle for gravitational self-energy E.G. Adelberger, S. Baeβler, J.H. Gundlach, B.R. Heckel, S.M. Merkowitz, U. Schmidt and H.E. Swanson Einstein's Equivalence Principle (EP) requires the exact identity of gravitational and inertial masses. The Strong Equivalence Principle (SEP) specializes this to the component of mass due to gravitational self- energy. The SEP is an important test of theories of gravity because it probes a nonlinear aspect of the interaction, and is violated by metric theories that contain more than one field. Nordtvedt1 noted that the SEP could be tested by using Laser Lunar ranging (LLR) to compare the accelerations of the earth and moon towards the sun. (Tests of the SEP require astronomical bodies -- gravitational self-energy reduces the earth's and moon's mass by only 4.6 and 0.2 parts in 10-10 respectively.) One analysis2 of the LLR data yield ∆a LLR / a s = (3.2±4.6)×10-13 and another analysis3 independent of the first yields ∆a LLR / a s = (-3.6±4.0)×10-13. (We define ∆a / a s = (ae - am) / [(ae + am)/2], ae is the acceleration of the earth towards the sun, and am is the acceleration of the moon towards the sun.) However, the LLR result is ambiguous because the earth and moon “test bodies” differ in two important ways: the earth is more massive than the moon (which probes the EP for gravitational self-energy) and the earth has an Fe-Ni core while the moon does not (which probes the usual composition-dependence test of the EP). We remove the ambiguity of this important test by using a torsion balance to compare the accelerations of “moon” and “earth core” test bodies towards the sun, thereby probing the composition dependent effect alone. In the last year we upgraded the Eöt-Wash II torsion balance with a new turntable controller which enabled us to increase our signal frequency, and we decreased the fiber noise by running at a lower temperature. Our current result for a composition-dependent component of the earth-moon differential acceleration is ∆a CD / a s = [-2+3(stat.)+2(syst.)]×10-13. Table 1.12-1. Preliminary error budget. Source of error: δ( ∆a CD / a s ) Diurnal tilt variations 2.1 × 10-13 Gravity gradients 0.9 × 10-13 Temperature effects 0.6 × 10-13 Magnetic effects 0.05 × 10-13 Statistical error 3.0 × 10-13 The LLR results from Refs. 2 and 3 and our data imply ∆a SEP / a s = ∆a LLR / a s - ∆a CD / a s = (5±6) × -13 10 and (-2±5)×10-13, respectively, providing a test of the SEP at the 1.4 × 10-3 level. The uncertainty of LLR analysis is expected to drop by a factor of 2-3 in the near future.4 Further upgrades of our apparatus are underway to decrease the statistical error and the tilt sensitivity. These upgrades should yield a sufficiently good result that our data will not limit the precision of the SEP test. Preliminary results are reported in proceedings.5,6 One of us (S.B.) thanks the Alexander v. Humboldt-Stiftung for their financial support. 1 K. Nordvedt, Phys. Rev. D 37, 1070 (1988). 2 J.G. Williams et al., Phys. Rev. D 53, 6730 (1996). 3 J. Müller et al., Proceedings of the 8th Marcel Grossman Meeting, Jerusalem,Israel, 1997. 4 K. Nordvedt, private communication. 5 E.G. Adelberger, Proceedings of the WEIN Conference, Sante Fe, NM 1998. 6 B. Heckel et al., Proceedings of the 32nd COSPAR Scientific Assembly, Nagoya, 1998. 13


  • Page 21

    2.0 NEUTRINO PHYSICS 2.1 The Sudbury Neutrino Observatory Q.R. Ahmad, J.F. Amsbaugh, M.C. Browne, T.V. Bullard, T.H. Burritt, P.J. Doe, C.A. Duba, S.R. Elliott, J.E. Franklin, A.A. Hamian, G.C. Harper, K.M. Heeger, M.A. Howe, R. Meijer Drees,* A.W. Myers, A.W.P. Poon,† R.G.H. Robertson, H. Seifert,# M.W.E. Smith, T.D. Steiger, T.D. Van Wechel and J.F. Wilkerson The Sudbury Neutrino Observatory (SNO), is a joint Canadian/US/UK effort to measure the spectral distribution and flavor composition of the flux of the higher-energy, 8B neutrinos from the Sun by using a 1000 metric ton detector of heavy water. The SNO detector relies on three different neutrino interaction processes. 2 H + ν e → p + p + e− Charged Current (CC); 2 H +νx → p + n +νx Neutral Current (NC); e− + ν x → e− + ν x Elastic Scattering (ES). Because of this unique sensitivity to the known flavors of neutrinos, SNO should be able to yield a definitive answer to the question of whether neutrino flavor oscillations are occurring in the Sun. As a second generation solar neutrino detector, SNO will also have a significant increase in statistical sensitivity compared to present detectors assuming an energy threshold of 5 MeV, the SNO detector, is expected to observe 12.7 CC events/day (Standard Solar Model/2), 5.5 NC events/day (SSM), and 1.2 ES events/day (SSM/2). This past year SNO and the University of Washington SNO group have achieved a number of major milestones. Construction of the detector was completed and the filling and commissioning phase of the experiment was started. By the end of 1998, the detector had been filled past the halfway point with heavy water. The acrylic vessel has proven to be sound and has behaved as expected during the fill process. In April 1998, the underground commissioning of the electronics and data acquisition systems was initiated. The SNO data acquisition (DAQ) system and monitoring tools developed by the UW group are currently in routine use supporting readout of the full set of PMT electronics. The object oriented data acquisition system has proven to be quite robust during recent data collection runs. These tools are essentially ready for the start of production solar neutrino runs. A large amount of commissioning data and source calibration data has now been acquired with the DAQ system. We have been involved in various aspects of the analyses of these data. Our major effort to develop a discrete neutral current sensitive detector (NCD) array has also progressed. We have now entered full production and are busy constructing the 3He-filled Ni proportional counters. By the end of 1998 we had delivered 15% of the neutral current array to site along with a preliminary NCD data acquisition system. This system is installed and operating underground and we are currently acquiring cool-down data from these counters. The completion of the heavy water fill to the full 1000 ton capacity of SNO was finished April 26, 1999; acquisition of solar neutrino data should commence shortly thereafter. * 8140 Lakefield Drive, Burnaby, British Columbia, Canada. † Lawrence Berkeley Laboratory, Berkeley, CA. # Sudbury Neutrino Observatory, Sudbury, Ontario, Canada. 14


  • Page 22

    2.2 SNO commissioning activities Q.R. Ahmad, M.C. Browne, C.A. Duba, A.A. Hamian, M.A. Howe, K.M. Heeger, A.W. Myers, R. Meijer Drees,* P.M. Thornewell† and J.F. Wilkerson During 1998 all major construction activities on the SNO detector reached completion, and commissioning activities took precedence. During the year, personnel from all SNO institutions participated in the commissioning tasks at the SNO site in Sudbury, Ontario. The commissioning of the detector involved, among other things, the successful integration and interfacing of the following detector components: • Photo multiplier tubes (PMTs); • Electronics; • Data Acquisition; • Calibration; • Analysis. The SNO group at the University of Washington along with other members of the collaboration was directly involved in the systematic task of integrating these components into the detector and subsequently verifying that the overall system was functioning within allowed parameters. As a result of the concerted efforts of the collaboration, we currently have in place all the primary and critical components of the detector and the commissioning phase of the experiment is nearing completion. An integral part of the detector commissioning is the demonstration of a robust and dependable Electronics and Data Acquisition System (EDAS). The on-site deployment and successful integration of an optimally functional EDAS required the critical characterization and shakedown of all hardware and software components of this system. Members of the UW DAQ group, working in close concert with the Electronics group, spent a considerable amount of time on-site to facilitate and oversee the commissioning of the EDAS. The UW SNO group has also been involved in building comprehensive data analysis and data monitoring tools which have been utilized to study the data acquired during the commissioning phase. SNO started taking air-fill (no water) data with partial electronics installed underground during the winter of 1998. In March the collaboration reached a major milestone by recording the first physics event at SNO, which at the time had around 1500 active photo multiplier tubes. The event comprised a muon traversing the edge of the acrylic vessel, producing light and illuminating the active PMTs. By that time we had managed to implement the DAQ system to the extent that we were able to routinely take data with a minimal amount of expert intervention. Part of the DAQ commissioning process has been both to optimize the software performance and to tune the software to allow easy and clear operator interactions with the complex detector electronics. The resulting intuitive and user friendly DAQ design has proven quite successful and enabled us to quickly train SNO personnel to become detector operators. * 8140 Lakefield Drive, Burnaby, British Columbia, Canada. † F5 Networks Inc., Seattle, WA. 15


  • Page 23

    2.3 The SNO data acquisition system Q.R. Ahmad, C.A. Duba, A.A. Hamian, P. Harvey,* K.M. Heeger, M.A. Howe, R. Meijer Drees,† A.W. Myers, J. Roberts,# P.M. Thornewell,% T.D. Van Wechel and J.F. Wilkerson The SNO data acquisition system (DAQ) is designed to provide continuous readout of the 9547 photomultiplier tubes (PMTs) that comprise the detector. The system has been described in some detail in the 1997 Annual Report,1 only an overview is provided here. The DAQ is divided into four main processes: system initialization and control; hardware readout; event building and data stream recording; and monitoring. The first of these is accomplished by the SNO Hardware Acquisition Realtime Control program (SHaRC), a C++ object oriented code running on a 250 MHz PPC computer. Hardware readout is provided through a VME based Motorola MVME167 68040 single board computer running an optimized C hardware readout program. Event building and subsequent recording of the data stream are handled by processes running on a SUN Ultra Sparc 1 workstation. Online data monitoring is supported through a client/server approach; the data are shipped from the DAQ SUN to a server process (“Dispatcher”) that allows multiple monitoring clients to subscribe and view the data stream in near time. The past year has marked some very important milestones for the SNO DAQ group. In April 1998, the complete system was used to acquire data with approximately half of the PMTs fully instrumented. In August 1998, with all of the PMTs operational, SNO DAQ was used to acquire the first commissioning data sets. In Fall 1998, sustained data throughput rates up to 400 kB/s were demonstrated, with burst rates up to 600 kB/s. The ability to handle these rates represents a significant upgrade to the event building software, which was implemented in the past year. The SHaRC program has undergone an intensive period of upgrades, optimization, and testing in order to provide the many, sometimes complicated, levels of initialization and control for the detector, while retaining its straightforward operator interface. For the first time in 1998, non-expert detector operators were trained to use SHaRC to perform standard tasks, and the operator feedback was useful for keeping the user interface easy to use. In addition to providing control of essentially every aspect of the SNO hardware, SHaRC also provides the ability to insert and manipulate various calibration sources in the detector. This feature was tested with several different sources, and found to perform reliably. In late 1998, the ability to continuously run the DAQ with no stops or pauses between runs was added; this continuous running mode is a very important feature of the system, particularly in the event of a supernova. As part of this mode, the DAQ software supports restarting the various processes with no loss of detector data. In 1998, the DAQ group provided full support of the DAQ system including the continuous presence in Sudbury of one or more group members. The on-site personnel were actively involved in solving day-to-day DAQ questions, and also in the ongoing code verification and implementation process. The overall response to the SNO DAQ by the detector operators, and by the collaboration as a whole, has been very positive. Although the goal of providing a fully operational system by the time the detector is ready to begin taking solar neutrino data has been met, there are always enhancements to be added. The DAQ group continues to add to and modify various aspects of the system, incorporating feedback from the electronics group and operators. * Queen’s University, Kinston, Ontario, Canada. † 8140 Lakefield Drive, Burnaby, British Columbia, Canada. # Sudbury Neutrino Observatory, Sudbury, Ontario, Canada. % F5 Networks Inc., Seattle, WA. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1997) pp. 20-23. 16


  • Page 24

    2.4 Overview and status of the SNO DAQ SHaRC control software Q.R. Ahmad, Y.D. Chan,* P. Harvey,† M.A. Howe, F. McGirt,# R. Meijer Drees,% J. Roberts,+ P.M. Thornewell and J.F. Wilkerson The SNO Hardware Acquisition Real time Control (SHaRC) code provides the control interface for the Sudbury Neutrino Observatory DAQ system. It is compiled with the MetroWerks Code Warrior compiler and runs on Macintosh PowerPC computers. SHaRC exploits the object-oriented nature of C++ to fully encapsulate a complete easy–to–use graphical user interface to the SNO data acquisition hardware. The interface provides for initialization of the entire hardware system and can probe the system to configure automatically the dialogs and the databases with the location and identifications of the hardware that is actually present and working. As the figure shows, there is a hierarchy of dialogs that allows one to display information at any level of the detector hardware. The system- level dialog depicts the actual layout of the electronic crates on the deck. Clicking on a crate brings up a dialog showing crate- level information. Clicking on a front-end card in a crate will bring up information about that card and so on down to a level that displays information about a particular photomultiplier tube. A high-voltage control and safety system is built into SHaRC. HV supplies can be ramped at custom rates. Noise rates can be monitored to have the HV supplies automatically shut down in the event of certain types of breakdown. Also, the operator can panic ramp the high voltage Fig. 2.4-1. A screen dump showing SHaRC in action. back to zero at any time for a particular supply, or for the entire detector. The state of the high voltage system is clearly shown on all of the hardware-display dialogs. There is an alarm system to alert the operators to unusual events. For example, if a hardware problem causes the data buffers to begin filling on a particular card an alarm is posted. There is even an option to take such cards offline automatically and to start a new run in those cases. In addition, there are dialogs to start/stop runs, display run and data-flow information, and automatically schedule certain calibration and monitoring tasks. A “Hardware Wizard” allows one to program a highly selectable set of hardware parameters. In addition to the SNO hardware objects, database objects provide for storage of electronics constants and a run history. At the start of every run, the databases that have changed are copied into a run-history folder. There is an interface to allow the configuration of the detector to be restored to the state it was in during any previous run. Hardware constants can be loaded manually, via automatic calibration routines, or from databases kept by other groups, such as the electronics group. SHaRC is now in routine and continuous use in the SNO experiment. * Lawrence Berkeley National Laboratory, Berkeley, CA. † Queen’s University, Kingston, Ontario, Canada. # Los Alamos National Laboratory, Los Alamos, NM. % 8140 Lakefield Drive, Burnaby, BC, Canada. + Sudbury Neutrino Observatory, Sudbury, Ontario, Canada. 17


  • Page 25

    2.5 SNO data stream monitoring Q.R. Ahmad, C.A. Duba, A.A. Hamian, P. Harvey,* M.A. Howe, R. Meijer Drees,† J. Roberts,# P.M. Thornewell% and J.F. Wilkerson Monitoring of the SNO data stream is accomplished via several independent programs which provide complementary information. Near time monitoring uses the client/server approach described in Section 2.3, while offline analysis often uses one or more of these programs, reading data from SNO-defined, ZEBRA- formatted1 “ZDAB” output data files. Developed by C.A. Duba, SNOStream is a SNO customized front end to the “Histo” software package.2 SNOStream supports a broad array of data-stream monitoring tools, including front-end-card and timing-card acquisition rates, trigger information, event information, and data-stream integrity checks. SNOStream receives the data stream from the dispatcher and passes selected portions to Histo for online graphical display. SNOStream can bin and histogram multiple events, and can also plot the data as functions of time, and of electronics crate and slot. Recent capabilities allow SNOStream also to view data from standard SNO ZDAB data files. SNOStream is proving to be a versatile monitoring tool and is currently being used to develop supernova-burst monitoring capabilities. XSNOED is an X-windows event viewer, written by P. Harvey, that runs on site SUN workstations for near-time monitoring, but can also be run on other UNIX platforms, including DEC alphas and LINUX, for offline analysis. Events can be viewed in a spherical or flat map of the PMT support structure (PSUP) and/or in a map of the electronics crates. Either the charge data or time distribution data can be selected for viewing, with a color map representing the scale. In addition to the event maps, data for each event can be viewed as a histogram. When running in continuous mode, XSNOED displays events which satisfy a user-specified trigger, at a rate which is also user-selected. If an event of particular interest is observed, it is trivial to halt the continuous updating and “backup” to the desired event. A more recent addition to the monitoring tools is a stand-alone PSUP Snapshot Viewer developed by J. Roberts, which is based on the PSUP and Crate View objects developed for the SHaRC program (see Section 2.4). In SHaRC, these views provide a wide variety of information, such as the discriminator threshold settings, the temperature of the electronics cards, and which PMTs are offline or online. Using the Snapshot Viewer, the saved data and parameters from previous runs can quickly be recalled and reviewed. A set of SNO analysis tools has been developed at Queen's University for use with ROOT, a comprehensive object oriented framework for data analysis applications developed at CERN.3 This package has recently been installed on computers at NPL, and it is envisioned that ROOT will be used as a major tool for offline data analysis of SNO data at UW. In addition to offline applications, the SNO/ROOT analysis package has recently been expanded to include some online event monitoring. This extended capability is currently being tested, and will be used during future data taking activities. * Queen’s University, Kingston, Ontario, Canada. † 8140 Lakefield Drive, Burnaby, British Columbia, Canada. # Sudbury Neutrino Observatory, Sudbury, Ontario, Canada. % F5 Networks Inc., Seattle, WA. 1 A Fortran-based memory management system. 2 www-pat.fnal.gov/nirvana/histo.html. 3 http://root.cern.ch. 18


  • Page 26

    2.6 The Neutral Current Detector Project at SNO J.F. Amsbaugh, M.C. Browne, T.V. Bullard, T.H. Burritt, P.J. Doe, C.A. Duba, S.R. Elliott, J.E. Franklin, G.C. Harper, K. M. Heeger, A.W. Myers, R.G.H. Robertson, M.W.E. Smith, T.D. Steiger, T.D. Van Wechel and J.F. Wilkerson SNO will detect Cerenkov light emitted from electrons or positrons produced by charged-current neutrino interactions. This reaction will provide a measure of the flux of electron neutrinos from the sun. Neutrinos of any active flavor can produce free neutrons in the heavy water by neutral-current interactions. Thus the measurement of the neutron production is a measurement of the total flux of neutrinos from the sun. Since solar burning produces only electron neutrinos, a comparison of the total neutrino flux to the electron neutrino flux could provide strong evidence for neutrino oscillations and therefore neutrino mass. The neutral current detectors (NCD's) are an array of 300 3He filled proportional counters designed to detect such neutrons. These NCDs are made by chemical vapor deposition (CVD) on a mandrel to form Ni tubing and endcap components. The CVD process results in very low U and Th contamination (1-2 ppt Th). A quartz tube forms the high voltage and signal feedthrough to a Cu anode wire. Most of the tubes were stored in an underground location at Index, WA to reduce cosmogenic activation. Unfortunately the air contained a very high level of Rn (900 pCi/l). As a result, counters constructed from these tubes had very high 210Po-alpha decay rates (~8000 alphas/m2/day). The noble metal Po adheres very strongly to Ni and plates out on Ni in acidic solutions. Hence electropolishing is the only technique known to us to remove the Po. Electropolishing, which keeps the radioactivity in solution, had reduced the tubes to this level from an initial rate of 105 alphas/m2/day. But as a comparison, counters built early on before storage underground had an alpha rate of <100 alphas/m2/day. Systematic studies of this problem were difficult because testing a new electropolishing procedure required counter assembly. To address this issue, we built an anode wire support structure that could be installed into a tube without laser welding the endcap-feedthrough assembly in place. As a result of these studies, the electropolishing procedures were improved and typical initial alpha rates are again near 100 alphas/m2/day. The various parts needed to assemble endcaps are being fabricated and most are completed. As they are produced, they are being shipped to IJ Research in Santa Ana, California and being assembled. There was also a delay in the delivery of endcaps due to difficulties with the company’s RF sputterer. This problem has been addressed. We should start receiving endcaps again imminently. The NCD's must have very little radioactivity as they will reside in the sensitive inner region of the SNO detector. All parts which comprise the NCD's are being radioassayed to verify their cleanliness. We have assayed samples of almost all materials to be used in the array and all small fabricated parts. This radioassay program is nearing completion. The results indicate that the added photodisintegration background in SNO due to the NCD's will be less than that due to the impurity of the heavy water. Tests of the remotely operated vehicle to be used in the installation of the counters into SNO have been successful. The initial assembly of the deployment hardware is beginning as most of the engineering is nearing completion. The design, prototyping and test of the anchor assemblies for the counter strings are complete and fabrication of the parts is beginning. We have delivered 15% of the array underground at SNO. These counters are being studied to understand their contamination level. We will compare this measured number to our radioassay predictions. Although we have had delays due to the Po contamination and endcap production, we have successfully built and operated 103 detectors of which 80 are part of the array. These have included spare counters for the array, counters to verify the radiopurity, and counters for underwater pressure testing. Our present production schedule will have all detectors underground in Sudbury by late 1999, ready for deployment after a period of cooldown during which cosmogenic 56Co decays away. 19


  • Page 27

    2.7 Electronics for the NCD array M.C. Browne, K.M. Heeger, A.W. Myers, R.G.H. Robertson, T.D. Van Wechel and J.F. Wilkerson The SNO neutral-current detector (NCD) array will be used to detect neutrons liberated by the neutrino disintegration of deuterium. There are 96 `strings' of 3 He-filled detectors to be placed in the heavy water, each connected to a separate external 91- Ω current preamplifier via a cable. Above an input signal of 200 nA, the event rate is dominated by neutrons and alpha particles. Neutrons from muon interactions and NC events are expected to be detected at a rate of 50 per day, and alphas 1000 - 10000 per day. The longest duration of the signal (apart from the ion tail) is about 3 µ s, corresponding to the drift time across a detector. With integration to match this time, smaller signals due to betas and gammas can be detected. The signal current contains a great deal of information that can be used to distinguish between neutrons and other types of event, and to locate events along the length of the detector. High-speed digitization is necessary to recover the position information in particular. To reduce costs one may take advantage of the low rates and multiplex (MUX) the data. But to recover very small signals from the noise, and to deal with high burst rates such as might occur during a supernova, 96 channels of spectroscopy-quality shapers and ADCs are also provided. Signals from preamplifiers enter 2 parallel buffers, one driving 20-m long cables to the shaper-ADCs that reside in VME, and the other driving eight 12-to-1 multiplexers (MUX) via discriminators and 320-ns delay lines. The shaper-ADCs are the same basic units used in the emiT experiment (q.v.), but with time constants chosen to match the proportional counters. Pre-production shaper-ADCs are under construction. The MUX units are AD8180 high-speed video switches. To reduce level shifts, a dummy unit is always connected to the common output except when a signal is present. The MUX output goes to an AD8307 logarithmic amplifier. Logarithmic amplifiers make it possible to digitize signals having a wide dynamic range with a fixed precision of about 1% of voltage (for an 8-bit digitizer). The offset for a log amp functions as a gain control and is set by a DAC level supplied from a controller. The offset is also necessary to prevent rectification of the quiescent input noise, as the log amp is indifferent to the sign of the input voltage. The output is a current that is buffered by two op-amps before being sent to a digitizer. The 2 digitizers are Tektronix 754A digitizing 4-channel oscilloscopes with 50k of memory. Each scope services 48 inputs (4 MUX boxes) autonomously. The scopes are set up by GPIB or front-panel input. The GPIB bus links the 2 scopes and the calibration pulser to VME. At the time of the initial trigger, and again at the time of the end of readout, the MUX hit pattern (48 bits) can be recorded by strobing latches. By reading out the latches at some point during scope readout and then just after scope readout, it can be determined if additional inputs were received while the scope was dead. As the scopes are dead during readout time (up to 1s), no digitization of events arriving then will be obtained, but the shaper-ADC system will give the total energy of each event. A 2-channel prototype MUX system has been tested, and MUX boards have been designed. 20


  • Page 28

    2.8 In situ determination of backgrounds from neutral current detectors in the Sudbury Neutrino Observatory M.C. Browne, T.V. Bullard, P.J. Doe, C.A. Duba, S.R. Elliott, K.M. Heeger, A.W.P. Poon* and R.G.H. Robertson The use of ultra-low background 3 He proportional counters in the SNO detector will provide a means of measuring the neutral-current interaction rate of solar neutrinos. Since the binding energy of the deuteron is only 2.2 MeV gamma rays from natural decay chains in the proportional counters photodisintegrate the deuteron and thus simulate the neutral-current signal. This poses stringent radiopurity requirements for the counters. An in situ measurement of the background from seven neutral-current detectors will determine the activity of 232 Th in the neutral current detectors and thus determine an upper limit on the photodisintegration background generated by the radioactivity in the NCD array. The Neutral Current Detector (NCD) array consists of 775 m of 3 He proportional counters arranged in 96 strings with 300 counters. The counter bodies are made of about 450 kg of ultra-pure CVD nickel which contains natural U and Th. The photodisintegration gammas are produced in the decays of the 232 Th( 208 Tl → 208 Pb, Eγ = 2.615 MeV) and 238 U( 214 Bi → 214 Po, Eγ = 2.445 MeV) chains and in the decay of the relatively long-lived cosmogenically produced 56 Co. A Construction Hardware In Situ Monitoring Experiment (CHIME) has been designed to measure the NCD originated background in the presence of the D 2 O contribution prior to the deployment of the entire NCD array. It is an in situ measurement of the construction materials used in the NCD array. Seven individual counters arranged in a close-packed configuration with a total mass of about 5000 g and a length of 45" will be deployed as a background test source in the SNO detector. The construction materials and procedures for the CHIME counters are essentially identical to those in the real NCD array with the difference being that the CHIME counters are not active. In order to allow the 56 Co to decay, the CHIME must be stored underground for at least 3 months. It was placed on the 6,800' level on December 20, 1998, and will not be ready for deployment until after April 1999. A radon emanation test will be conducted on the CHIME while it is underground to assure its cleanliness before deployment. The detection limit for 228 Th is about 5 disintegrations per day and for 226 Ra about 2.5 disintegrations per day. These are the amounts that would be in secular steady state with 0.6 microgram 232 Th and 2 micrograms 238 U respectively. The radon emanation test hardware is currently being tested at LANL and will be shipped to Sudbury by April. It is anticipated that one month will be required to establish the equipment underground and conduct the emanation tests. The deployment of CHIME is expected to be no different than that of a standard calibration source and will have minimum impact on the SNO operating schedule. The CHIME is negative buoyant and can be deployed along the central axis of SNO using the existing calibration source deployment hardware and manipulator system. Monte Carlo studies have shown that the presence of CHIME will have no measurable impact on the physics goal of SNO. The complete geometry of CHIME, the SNO detector response, and all important decay schemes can be simulated using the SNO Monte Carlo code. Preliminary studies of the reconstruction distribution and the energy spectra have shown that an analysis approach based on fiducial cuts and the shape of the 56 Co energy distribution will allow us to discriminate the 56 Co contribution to the photodisintegration background with high enough accuracy. In addition, further techniques will be investigated to distinguish between reconstructed events in the fiducial volume due to 208 Tl decays in the CHIME and the D 2 O . It is expected that the Construction Hardware In Situ Monitoring Experiment will allow us to determine the photodisintegration background from the NCD array at the level of 10% of solar module after two weeks of in situ counting. * Lawrence Berkeley Laboratory, Berkeley, CA. 21


  • Page 29

    2.9 Using a remotely operated vehicle to deploy neutral current detectors in the Sudbury Neutrino Observatory J.F. Amsbaugh, T.H. Burritt, P.J. Doe, G.C. Harper and M.W.E. Smith The deployment of the NCDs into SNO must be done with a minimum of down time for the observatory, yet must be carried out with care, so not to contaminate the heavy water or damage the acrylic vessel it is contained in. The primary tool for this task is a remotely operated vehicle (ROV) designed by Deep Ocean Engineering. In addition, there are some essential items of deployment hardware that are being designed and fabricated in house at NPL. The procedures for deployment must be carefully thought out and rigorously tested. This process has begun, with the ROV being tested in a 20 ft deep pool at Los Alamos National Laboratory, a collaborator in this project. Members of NPL have traveled to Los Alamos for these tests and have become proficient at operating the ROV. A series of dummy detectors were captured by the ROV, taken to the bottom of the pool and deployed into anchors on an acrylic panel. The environment in the pool was set up to simulate that of SNO, with similar lighting and background appearance. The tests were successful, demonstrating that the ROV provides a sound technique for NCD deployment. The water depth at SNO complicates the deployment program, with the ROV operating at 60 ft below the deck. In addition, the NCD cables, which run inside the upper hemisphere of the acrylic vessel, must hook into a cable attachment ring at the bottom of a 24 ft acrylic chimney. To overcome these difficulties, along with the constraints of overhead clearance, the following hardware has been designed and is being fabricated at NPL. • Global View Camera mechanism, to assist in viewing both the ROV and the NCD cables. • Haul Down, for moving detectors to the bottom of the vessel for capture by the ROV. • Manipulating Pole, for rearranging NCD cables in the chimney of the acrylic vessel. • Welding Station, for laser welding the individual NCDs together as they are deployed. • Deployment Platform, to integrate the hardware and cover the open vessel. Construction of these items is now 80% complete and it is expected that all hardware will be completed by the end of March. Testing of individual items will continue at the University of Washington through March, using a tank in the Engineering Department. During April, at the Los Alamos pool, we will be testing the hardware and ROV as an integrated system. Several members of NPL will be visiting Los Alamos during this period to design and participate in these tests. A program is being developed to train deployment personnel. Approximately 12 people will need to become proficient in laser welding, ROV operation and general deployment duties. It is expected that more than half of these people will come from NPL and will be trained sometime during the next six months. 22


  • Page 30

    2.10 Initial results from the cool-down phase of the Neutral Current Detector (NCD) Program for the Sudbury Neutrino Observatory M.C. Browne, S.R. Elliott, R.G.H. Robertson, T.D. Steiger and J.F. Wilkerson The NCD group has developed material preparation and construction techniques in order to produce counters with intrinsically low levels of 238 U and 232 Th. These isotopes give rise to background neutrons through photodisintegration of deuterium. An additional problematic isotope is 56 Co , which is a spallation product of cosmic rays with nickel. In order to allow this cosmogenic activation to decay away prior to NCD deployment in SNO, a storage period known as the “cool-down phase” has been implemented. The goals of the cool-down phase are to quantify potential backgrounds to the NCDs while safely delivering and characterizing the NCDs at the underground SNO facility. These backgrounds can be grouped into two categories: backgrounds due to the environment (thermal and fast neutrons, fission gammas), and intrinsic backgrounds to the counters (alpha particles from U, Th and Po). There are currently about 115 m of counter underground in the cool-down phase (~15% of the NCD array). Subsets of these counters have been running nearly continuously for six months. The data taken during the past year have provided an initial estimate of the potential NCD backgrounds. The primary measurement during the cool-down phase in 1998 was an attempt to assess the bulk U and Th contamination in the NCD construction materials. In a plot of projected track length (charge-pulse risetime) against total energy, events from bulk alpha emitters occupy regions different from surface activity. Initial results indicate bulk alpha activity from low-energy and high-energy regions of this plot between 2.28±0.78 and 1.20±1.1 alphas/m2/day. Monte Carlo studies predict 2.0 alphas/m2/day for U and 0.5 for Th contamination at the 10 ppt level. Fast neutron interactions in the counters also contribute some events, thus indicating that these values are an upper limit on the U and Th contamination. The production of NCDs in 1998 was significantly affected by contamination of 210Po on the nickel tubes. Additional construction techniques were implemented to remove the majority of this contaminant. Reduction of Po by >1000 was demonstrated at the University of Washington but was limited by hadronic interactions in the counters from cosmic rays. The cool-down phase was thus used to test the surface preparation techniques. The average number of counts recorded on the NCDs in the cool-down phase in the 210 Po peak window was 8.28 ± 1.17 alphas/m2/day, corresponding to a reduction of about 103. Additional long- term tests demonstrated that the activity was decaying with the anticipated 138-day half-life – confirming the success of the surface preparation techniques. In 1999, long-term alpha analysis is planned to obtain better statistics on the bulk U and Th contamination levels. Additional counters are scheduled for shipment to SNO late Spring 1999. 23


  • Page 31

    2.11 Sensitivity of NCDS to Supernovae during cool-down M.C. Browne, C.A. Duba, S.R. Elliott and R.G.H. Robertson During the aforementioned cool-down phase of the NCDs, all 96 strings of proportional counters will monitor neutrons inside SNO’s control room for no less than six months. These neutrons will mostly be created in (α,n) reactions within the norite walls of the Creighton mine. However, elevated neutrino fluxes, such as would be present during a supernova core-collapse, will create elevated neutron liberation rates within the norite. The higher neutron liberation rate will create a larger neutron flux within the control room, which may be detected by any operating NCD strings. Although the probability of a detectable supernova event during the relatively short-duration cool- down phase is low, this test should provide valuable data as to the feasibility of using proportional counters for supernova detection. These data could be used to assess proposals such as OMNIS1 and LAND.2 We will test supernova event backgrounds, counter geometric efficiencies, neutrino target materials, data acquisition systems, and supernova alert protocol. The results of these tests could then be used to optimize supernova detection strategies for permanent supernova detectors. Norite rock is an igneous silicate, made of mostly silicate, a rather poor material in terms of neutrino- neutron cross section. However, the norite contains non-negligible amounts of aluminum, calcium, iron, sodium, and magnesium. Convoluting the energy-dependant neutrino-neutron cross sections of the major constituents of norite with the expected neutrino energy spectra in core-collapse events yields a net neutron production rate of roughly 8×10-6 n/g for a distance of 1 kpc. The total mass of norite, to which the cool-down array will be effectively sensitive, is: r( L × ε L ) 7000d −1 ( 775m ×.40) Ms = (l × ε )R αn (16.5m ×.60)2.8g −1 yr −1 Where r is the rate of neutron detection with the current cool-down array, l is the length of the current array, ε is the geometric efficiency of the NCD setup, L is the length of the complete array, and Rαn is the measured rate of (α,n) creation within the norite.3 Ms works out to be 3×107 g, which implies that only ~4 neutrons will be detected for a galactic-center supernova at 8 kpc. Unfortunately, this is not much higher than the background rate due to norite (α,n). The predicted signal-to-background depends on the supernova model that is used; a one-second burst-model supernova would provide a ratio 7:3, while the data from supernova 1987a seems to favor a ten-second burst, and thus a ratio of 34:30.3 In either case, the chance of triggering on most galactic supernova events is negligible without additional signal-to-background improvements. One possible improvement would involve the insertion of a moderator, such as paraffin or water, into the NCD array, and would amplify both signal and background, effectively increasing the statistical significance of an event. A second improvement could be made by repositioning the counters in the NCD array, thereby increasing the geometric efficiency by minimizing detector “shadowing” within the control room. Finally, a substantial amount of 208Pb would dramatically increase the sensitivity of the NCDs to supernovae, since Pb has a neutrino-neutron cross-section by mass almost 20 times higher than norite. When this is coupled with Pb’s long neutron absorption length and very low neutron background, even 10 tons of strategically placed Pb with moderator should more than double the expected neutron flux from supernovae without substantially altering the background. 1 OMNIS-an improved low-cost detector to measure mass and mixing of mu/tau neutrinos from a galactic supernova, P.F. Smith-, Astroparticle-Physics.vol.8, no.1-2; Dec. 1997; pp. 27-42. 2 LAND lead astronomical neutrino detector: LAND, C.K. Hargrove, I. Batkin, M.K. Sundaresan and J. Dubeau, Astroparticle-Physics.vol.5, no.2; Aug. 1996; pp. 183-96. 3 Annex 9 Report, SNO proposal, 1987, E.D. Earle et al. 24


  • Page 32

    2.12 SAGE: The Russian American Gallium Experiment S.R. Elliott and J.F. Wilkerson The Russian-American Gallium Experiment (SAGE) is a radiochemical solar neutrino flux measurement based on the inverse beta decay reaction, 71Ga(ν,e-)71Ge. The threshold for this reaction is 233 keV, which permits sensitivity to the p-p neutrinos that dominate the solar neutrino flux. The target for the reaction is 55 tonnes of liquid gallium metal stored deep underground at the Baksan Neutrino Observatory in the Caucasus Mountains in Russia. About once a month, the neutrino induced Ge is extracted from the Ga. 71 Ge is unstable with respect to electron capture (τ1/2 = 11.43 days) and, therefore, the amount of extracted Ge can be determined from its activity as measured in small proportional counters. The experiment has measured the solar neutrino flux extractions between January 1990 and December 1997 with the result 67.2 +−77..20 (statistical) +−33..50 (systematic) SNU, which was reported at the Neutrino 98 Conference in Japan in June 1998. This is well below the standard solar model expectation of 129 SNU. Additional extractions are being analyzed. The Figure below shows a plot of the extraction data. The collaboration has used a 517-kCi 51Cr neutrino source to test the experimental operation. The energy of these neutrinos is similar to the solar 7Be neutrinos and thus makes an ideal check on the experimental procedure. We have published this result in 1996. The result, expressed in terms of a ratio of the measured production rate to the expected production rate,1 is 0.95+0.12. This indicates that the discrepancy between the solar model predictions and the SAGE flux measurement cannot be an experimental artifact. The work has also been described in a long archive paper that is currently in press.2 In collaboration with the Institute for Nuclear Research, we submitted a grant request to CRDF.3 This two year grant request was funded in 1997 and is now complete. The monies were used to support Russian scientists employed to continue solar neutrino observations. SAGE is a mature experiment whose operation has become routine. The University of Washington plays a role in the analysis of the data. We are assisting in the design and construction of new proportional counters for the experiment. With the publication of the Cr data, the focus is now on the writing of archive papers summarizing the experimental procedure and its solar neutrino results. ✁ ✁ ✫ ✬ ✭ ✮ ✯ ✮ ✭ ✰ ✱ ✲ ✳ ✴ ✵ ✶ ✱ ✷ ✵ ✮ ✸ ✬ ✹ ✺ ✻ ✱ ✬ ✼ ✱ ✲ ✰ ✻ ✂ ✁ ✁ ✖ ✔✕ ✒✓ ✄ ✁ ✁ ☞ ✑ ✎ ✏ ✟✍ ☞✌ ☎ ✁ ✁ ☛ ✠ ✡ ✟ ✝ ✞ ✆ ✁ ✁ ✁ ✆ ✗ ✗ ✁ ✆ ✗ ✗ ☎ ✆ ✗ ✗ ✂ ✆ ✗ ✗ ✙ ✆ ✗ ✗ ✘ ✚ ✛ ✜ ✢ ✣ ✤ ✥ ✦ ✜ ✧ ✥ ★ ✩ ✢ ✪ ✜ ✥ ✛ Fig. 2.12-1. The individual measurements of the solar neutrino production rate. 1 J.N. Abdurashitov et al., Phys. Rev. Lett. 77, 4708 ( 1996). 2 J.N. Abdurashitov et al., Phys. Rev. C, in press. 3 Civilian Research and Development Foundation for the Independent States of the Former Soviet Union, Award # RP2-159, Proposal # 3126. 25


  • Page 33

    3.0 NUCLEUS-NUCLEUS REACTIONS 3.1 The GDR width in highly excited nuclei M.P. Kelly, M. Kicinska-Habior,* J.P. Lestone,† J.F. Liang,# K.A. Snover, A.A. Sonzogni,% Z. Trnadel* and J.P.S. van Schagen We have completed new measurements of 18O + 100Mo reactions from E(18O) = 122 to 214 MeV in order to understand better the width evolution of the hot GDR versus excitation energy in near-Sn compound nuclei. Our approach combines results from separate measurements of light-charged particles, γ rays and evaporation residues. First, we have measured light charged particle emission and deduced the effect of preequilibrium energy and mass loss prior to compound nucleus decay. Large preequilibrium losses of approximately 20% of the full fusion excitation energy and several mass units are observed for bombarding energies as low as 11 MeV/nucleon. Second, using a new array of three large NaI spectrometers along with a γ-ray multiplicity array, we have measured the γ-ray emission cross sections and angular distributions for five bombarding energies. These data permit a direct separation of the statistical GDR component from the underlying bremsstrahlung. Last, the measured evaporation residue excitation function were used to extract the GDR strengths with good accuracy. Our analysis of GDR spectra includes a simultaneous fit of statistical emission plus bremsstrahlung to both the measured γ-ray strength function and the a1(Eγ) coefficient determined from the angular distributions. A careful account of the important dynamical effects of preequilibrium energy loss and bremsstrahlung emission is necessary for a reliable determination of the GDR parameters. Further insight into the excitation energy dependence of the GDR is gained by examining the dependence of the fitted parameters on the average thermal energy following giant-dipole emission, or equivalently the temperatures, rather than on the initial compound nucleus excitation energy. Results for the energy dependences of the GDR strength, width and centroid energy are determined and the measured widths are compared with the results of thermal fluctuation calculations of the width of the GDR. An additional contribution to the width of the GDR equal to twice the compound nucleus evaporation width may also become important at the higher energies. The result of this study is a new and qualitatively different understanding of the temperature evolution of the GDR in hot nuclei; namely, the width of GDR does not saturate but continues to increase up to at least T~2.4 MeV, which represents the highest temperatures for data measured here. Previously measured widths, reanalyzed for this study, give additional support for an increasing width up to T~3.2 MeV. Width results are summarized in the figure below.1 * Warsaw University, Warsaw, Poland. † Los Alamos National Laboratory, Los Alamos, NM. # Oak Ridge National Laboratory, Oak Ridge, TN. % Argonne National Laboratory, Argonne, IL 1 M.P. Kelly, K.A. Snover, J.P.S. van Schagen, M. Kicinska-Habior and Z. Trznadel, Phys. Rev. Lett., in press. 26


  • Page 34

    40 ✁ Jf (h) 30 ✁ 20 ✂ ✁ 10 ✁ 15 10 Γ (MeV) 5 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 T (MeV) Fig. 3.1-1. Decay of Sn and nearby mass compound nuclei. Top panel: average angular momentum for GDR decay in 18O + 100Mo. Bottom panel: points - measured GDR widths. Filled squares are the result of the present study. Solid line - thermal fluctuation calculation of Kusnezov and Alhassid (private communcation). Dashed line – includes the additional additive contribution 2Γevap. 27


  • Page 35

    3.2 High-energy γ-ray emission in 12C + 58, 64Ni reactions at 6--11 MeV/u M.P. Kelly, M. Kicinska-Habior,* J.P.S. van Schagen, K.A. Snover and Z. Trznadel* Recently, the 12C + 24,26Mg and the 18O + 100Mo mass-asymmetric heavy-ion reactions at 6-11 MeV/u have been used to disentangle statistical high-energy γ-ray emission and bremsstrahlung emission by measuring both γ-ray spectra and angular distributions.1,2 In this manner, reliable Giant Dipole Resonance (GDR) parameters and bremsstrahlung parameters have been extracted. In order to obtain similar information concerning the bremsstrahlung process and the GDR built in compound nuclei formed in medium-mass heavy- ion collisions the 12C + 58, 64Ni reactions were studied at the University of Washington Nuclear Physics Laboratory using the FN Tandem Van de Graaff together with the Superconducting Linear Accelerator. High- energy γ-ray spectra at angles 40°, 55°, 90°, 125° and 140° were measured with a new triple NaI-spectrometer set-up3 at beam energies of 5.5, 8 and 11 MeV/u. The angular coefficients A0, a1 and a2 in the nucleus-nucleus CM frame have been extracted from Legendre Polynomial fits to the singles data and to the fold >2 data for both reactions. The results for 12C + 64Ni are shown in Fig. 3.2-1. A large bremsstrahlung component which increases with projectile energy is clearly visible at γ-ray energies above 20 MeV in A0 as well as in a1. The anisotropy observed in the a2 coefficient suggests a small deformation of the nuclei formed. A similar behavior was found for the 12C + 58Ni reaction. Simultaneous fits of theoretical calculations to both A0 and a1 coefficients are in progress. 12 64 ✆ 12 64 12 64 ✆ C+ Ni @ 5.5 MeV/u C+ Ni @ 8 MeV/u C+ Ni @ 11 MeV/u ☎ ☎ 10 A0 [mb] ✄ -1 10 ✄ -3 10 ✄ -5 10 a1 0.5 0 0.4 a2 0.2 ✞ 0.0 -0.2 -0.4 10 20 30 40 10 20 30 40 10 20 30 40 Energy [MeV] Fig. 3.2-1. Measured high-energy γ-ray spectra A0 (top row), and angular distribution a1 (middle row) and a2 (bottom row) coefficients for 12C + 64Ni at 5.5, 8 and 11 MeV/u. * Institute of Experimental Physics, Warsaw, Poland. 1 M.P. Kelly, K.A. Snover, J.P.S. van Schagen, M. Kicinska-Habior and Z. Trznadel, Proceedings of the Topical Conference on Giant Resonances, May, 1998, Varenna, Italy, Nucl. Phys. A, in press. 2 M. Kicinska-Habior, Z. Trznadel, M.P. Kelly, K.A. Snover and J.P.S. van Schagen, Proceedings of the Topical Conference on Giant Resonances, May, 1998, Varenna, Italy, Nucl. Phys. A, in press. 3 Nuclear Physics Laboratory Annual Report, University of Washington, (1997) p. 57. 28


  • Page 36

    3.3 Scaling properties of the GDR width in hot rotating nuclei Y. Alhassid,* D. Kusnezov ∗ and K.A. Snover We have examined1 the experimental and theoretical systematics of the giant dipole resonance width in hot rotating nuclei as a function of temperature T, spin J and mass A from A = 45 to 208. The calculations are based on the theory of adiabatic thermal shape fluctuations in the liquid drop model, which is known to be generally very successful in describing experimentally measured GDR widths. Calculations for a number of different nuclei were examined and an empirical scaling was deduced, resulting in a simple phenomenological function Γ( A, T, J ) which approximates the global behavior of the calculated GDR width. The deduced function is 4 /[( T / T0 )+3] ☞ ✌✍ ✎✏ ✑ J Γ (T, J, A ) = Γ( T, J = 0, A) L ✟✠ ✡☛ A 5/ 6 where Γ (T, J = 0, A ) = Γ0 ( A) + c(A) ln(1 + T / T0 ) and c(A) = 6.45 − A / 100. Γ0 ( A) is usually extracted from the measured ground-state GDR width in neighboring spherical nuclei, ✒ ✓ and typically lies in the range 3.8 - 5 MeV. T0 = 1 MeV is a reference temperature. L ξ represents the scaling function for the width at constant T, as a function of ξ = J / A 5 / 6 , where L(ξ ) ≅ 1 + 1.8{1 + exp[(1.3 − ξ ) / 0.2]}−1 . The angular momentum scaling as J / A 5/ 6 may be understood as due to the dominance of the rotation energy J 2 / 2 I at high spin, where the moment of inertia I ∝ A 5/ 6 . A comparison with available data shows agreement within 20% for most measurements. ✔ In a separate project, the dependence of the GDR width at low spin and low to moderate temperature was reexamined in two cases of recent experimental and theoretical interest, the decays of 120 Sn * and 208 Pb* populated by inelastic alpha scattering. Our calculated widths are significantly larger than previous ones,2 and disagree with published data.3 A revised computation of the proper temperature scale brings the data into fair agreement with theory in the case of 208 Pb but not 120 Sn . * Yale University, New Haven, CT. 1 D. Kusnezov, Y Alhassid and K.A. Snover, Phys. Rev. Lett. 81, 542 (1998). 2 W.E. Ormand, P.F. Bortignon, and R.A. Broglia, Phys. Rev. Lett. 77, 607 (1996); W.E. Ormand et al., Nucl. Phys. A 614, 217 (1997). 3 E. Ramakrishnan et al., Phys. Rev. Lett. 76, 2025 (1996); Phys. Lett. B 383, 252 (1996). 29


  • Page 37

    17 3.4 O inelastic scattering study of the GDR width in decays of excited 120Sn nuclei J.R. Beene,* Y. Blumenfeld,† M. Halbert,* F. Liang,* E. Mohrmann, T. Nakamura, † K.A. Snover, M. Thoennessen,† E. Tryggestad† and R. Varner* In November 1998 we had a 1-week run at the National Superconducting Cyclotron Laboratory in which we attempted to measure the GDR decay of excited 120Sn nuclei populated by the inelastic scattering of 17 O nuclei at a bombarding energy of 80 MeV/nucleon. The measurement was patterned after an earlier MSU experiment which looked at 120Sn decays populated by inelastic alpha particle scattering.1 In contrast to the usual fusion-evaporation studies, inelastic scattering/decay experiments populate much lower spins and thus probe GDR properties in a different region of spin and temperature parameter space. The motivation for an 17O experiment is 2-fold: 1) GDR widths deduced in the 120Sn(α,α’γ) experiment are not in good agreement with theory, and they are also systematically higher than fusion-evaporation results at low excitation energies where the spins and temperatures are similar in the 2 types of reactions;2 2) a recent experiment suggests significant preequilibrium emission may occur in the (α,α’γ) experiment prior to the formation of an equilibrated excited nucleus.3 In contrast to alpha particle inelastic scattering, 17O inelastic scattering is known to have much smaller nucleon knockout cross sections and thus there is the hope that preequilibrium processes might be weaker in the 17O case. A good 17O inelastic scattering experiment might also offer the opportunity to explore better the low-spin low-temperature region of GDR decay of Sn compound nuclei where measured GDR widths from fusion-evaporation reactions lie below theory.4 We used the S800 superconducting spectrometer to detect the inelastically scattered 17O particles, near O°, approximately 150 detectors of the ORNL-TAMU-MSU BaF2 array to detect the gamma rays, and an array of CsI detectors to measure preequilibrium charged particle emission. One difficulty we experienced was a substantial gamma background in the BaF2 detectors and 17O background in the S800 focal plane detectors due to stopped and scattered beam, respectively. Because the software for optimized angle and position cuts in the focal plane detection was not available on line, the efficacy of such cuts will be determined in the off-line analysis currently in progress. * Oak Ridge National Laboratory, Oak Ridge, TN. † Michigan State University, East Lansing, MI. 1 E. Ramakrishnan et al., Phys. Rev. Lett. 76, 2025 (1996). 2 D. Kusnezov, Y. Alhassid and K.A. Snover, Phys. Rev. Lett. 81, 542 (1998). 3 D. Fabris et al., J. Phys. G 23, 1377 (1997), Phys. Rev. C 58, R624 (1998). 4 M.P. Kelly et al., Phys. Rev. Lett., in press. 30


  • Page 38

    3.5 Investigation of the temperature dependence of the level density parameter: results from the 19 181 F + Ta → 200 Pb system A.L. Caraley, B.P. Henry, J.P. Lestone* and R. Vandenbosch During the last few years, we have been studying both the fusion-evaporation and fusion-fission channels of the 19 F + 181 Ta → 200 Pb reaction with the goal of determining the Fermi-gas level density parameter and its possible dependence on the temperature of the emitting system. Fabris et al.1 have reported, based on α -particle results from the same 19 F + 181 Ta system, that the level density parameter decreases dramatically from A/8.3 MeV-1 at a thermal excitation energy of U=20 MeV to A/12 MeV-1 at U=100 MeV. Last year we presented preliminary results from an experiment that measured light-charged particles (LCP) in coincidence with evaporation residues (ER) at Elab=121, 154 and 195 MeV.2 This year we have performed several additional experiments. Another LCP-ER coincidence data set was collected at Elab=179 MeV. In addition, to guide the simulations necessary to determine particle multiplicities, thorough measurements of the relative ER singles efficiency as a function of the deflector voltage were performed at all four beam energies. Also, another LCP-ER coincidence experiment at Elab=154 MeV was conducted. For this experiment, the light- charged particles were detected at θ lab = +160 and at θ lab = −160 in order to determine any influence of the ✕ ✕ ER coincidence requirement on the LCP spectral shapes. The resulting spectral shapes at the two angles were determined to be essentially identical. These findings, as well as the results of extensive statistical-model- based simulations, have confirmed that the spectra collected previously at θ lab = +160 are unaffected by ✕ kinematic bias. The coincident light-charged particles have been analyzed to determine both apparent temperatures and multiplicities. The multiplicity results for α -particles at θ lab = +160 and at θ lab = −160 , shown in Fig. 3.5-1a ✕ ✕ are consistent with those of Hinde et al.3 and Fabris et al. The error bars on the present results reflect the systematic uncertainties associated with the simulations. The solid and dashed lines indicate multiplicities calculated using Monte Carlo CASCADE4,5 with an equal to A/11 MeV-1 and A/13 MeV-1, respectively. Apparent temperatures were extracted from the center-of-mass energy spectra by fitting a generalized Maxwellian distribution to the high-energy slopes of the spectra. The α -particle results are shown as the solid points in Fig. 3.5 1b. Results of identical analyses of spectra calculated using Monte Carlo CASCADE, with an equal to A/11 MeV-1, A/12 MeV-1 and A/13 MeV-1, are indicated by the solid, dashed and dotted lines, respectively. Although not shown here, calculations with JOANNE6 yield nearly identical results for both the multiplicities and the spectral shapes. As illustrated in Figs. 3.5-1a and 3.5-1b, our experimental results are consistent with standard statistical model predictions using a constant level density parameter of ~A/13 MeV-1. However, as suggested by several theoretical discussions,7,8,9 it is possible that an equally good description of the particle spectra can be made using a level density parameter that varies smoothly with excitation energy. In Fig. 3.5-2 a comparison is made between the experimental α -particle apparent temperatures and several calculations using excitation energy dependent level density parameters. The solid line is a calculation using the strong energy dependence as determined by Fabris et al. The three dot-dashed curves are the results of calculations with the parameterization of Fineman et al.,10 an = A/(8.2 MeV + κ ·U/A) with κ =4.3, 3.0 and 2.0, based on the work of Shlomo and Natowitz.8 The result of a calculation based on the recent work of De et al.,9 where an = A/(9.5 MeV + 1.3·T) for A=208, is illustrated by the dashed curve. It is evident that our data do not support a strong energy dependence of the level density parameter. Instead, the data indicate that calculations with a value of κ between 3.0 and 4.3 could possibly reproduce the experimental * Los Alamos National Laboratory, Los Alamos, NM. 1 D. Fabris et al., Phys. Rev. C 50, R1261 (1994). 2 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 30. 3 D.J. Hinde et al., Nucl. Phys. A 385, 109 (1982). 4 F. Pühlhofer, Nucl. Phys. A 280, 267 (1977). 5 M.G. Herman, U. of Rochester Nuclear Structure Laboratory Report UR-NSRL-318 (1987), unpublished. 6 J.P. Lestone, Nucl. Phys. A 559, 277 (1993). 7 J.P. Lestone, Phys. Rev. C 52, 1118 (1995). 8 S. Shlomo and J.B. Natowitz, Phys. Rev. C 44, 2878 (1991). 9 J.N. De et al., Phys. Rev. C 57, 1398 (1998). 10 B.J. Fineman et al., Phys. Rev. C 50, 1991 (1994). 31


  • Page 39

    results. This value of κ would be comparable to those determined by Fineman et al. for the 193Tl and 213Fr systems. Furthermore, although the offset needs to be reduced, the prediction by De et al. follows the same trend as the experimental data. In either case, it appears that a quite modest excitation energy dependence is adequate to describe our results. Fig. 3.5-1. α -Particle Multiplicities (a) and Apparent Temperatures (b). The solid points are results from the present experiment. The lines are the results of statistical model calculations described in the text. Fig. 3.5-2. Comparisons of calculations with temperature-dependent level densities. The solid points are the experimental α -particle apparent temperatures. The solid and dashed lines illustrate results of calculations with the excitation energy prescriptions of Fabris et al. and De et al., respectively. The three dot-dashed curves illustrate results of calculations with the parameterization of Fineman et al. and κ equal to 4.3, 3.0 and 2.0 (top to bottom). 32


  • Page 40

    4.0 NUCLEAR AND PARTICLE ASTROPHYSICS 4.1 The 7Be(p,γγ)8B cross section at astrophysically interesting energies E.G. Adelberger, J.-M. Casandjian, A. Junghans, E. Mohrmann, K.A. Snover, T.D. Steiger, H.E. Swanson and the TRIUMF Collaborators* The apparatus and target development for our 7Be(p,γ) experiment has been mostly completed and debugged.1 In December 1998 we saw our first alpha particles from the decay of 8B produced by 7Be(p,γ). At that time we also completed the design and installation of our proton shields to reduce scattered proton background in the alpha particle detectors. This background was produced by beam hitting the aperture plate on the opposite end of the rotating arm from the target, during the beam flux monitoring (target counting) phase of the arm rotation. At this point, several different preliminary measurements have been carried out: 1) Yields from 7Be(p,γ) have been measured at 5 proton energies spanning the Ep(cm) = 630 keV resonance as well as points at 400 and 500 keV, all with statistical uncertainties ~10% or smaller. The relative yields are in good agreement with Filippone2 except that the resonance appears somewhat broader. 2) A thick target yield curve for the Eα = 953 keV resonance in 7Li(α,γ) has been measured. 3) Several 7Be(α,γ) yield curves have been measured over the narrow resonance at Eα = 1376 keV. Sharp resonance curves have been observed with a rise of 2 keV or so, indicating good accelerator beam energy resolution. Measurements 2) and 3) were carried out using the large UW NaI spectrometer, which was moved for this purpose to the 7Be target chamber in Cave 1. Measurements 1) and 3) were carried out with a 13 mCi 7Be target doped with 9Be. The 9Be doping was to permit a 9Be(p,γ) resonance diagnostic measurement in the eventuality that the 7Be(α,γ) measurement proved too difficult, or that an alpha beam of the required intensity was not available. However, the 7Be(α,γ) resonance measurement succeeded very well. The observed FWHM is about a factor of 2 larger than the width expected based on only Li+Be in the target, and is thus very encouraging in terms of target purity, although there is a long high energy tail on the resonance. A similar measurement with an undoped 7Be target showed a resonance profile a factor of 2 narrower. We also plan to measure the 7Be(p,γ) resonance yield with good statistics on top of the 630 keV resonance for various sweeping amplitudes and hence various beam flux nonuniformities. This will determine the target uniformity and hence the required beam flux uniformity. Several improvements are being worked on, including electronic noise reduction in the NaI and Silicon detector signals due to pickup/ground loops as well as noise from the arm rotation motor/computer. Our plans for the near future are for further 7Be(p,γ) measurements with more active targets. * N Bateman, L Buchmann, A Zyuzin, J Vincent et al., TRIUMF, Vancouver, Canada. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 10. 2 B.W. Filippone et al, Phys Rev C 28, 2222 (1983). 33


  • Page 41

    4.2 Installation of chamber and beamline for the 7Be(p,γγ)8B experiment J.F. Amsbaugh, J.-M. Casandjian,* C.E. Linder and T.L. McGonagle This experiment rotates an arm-mounted target to positions for beam irradiation and for detector counting of the target. Other positions place apertures on the opposite end of the arm into the beam. The radioactive 7Be targets used in the cross-section measurements will produce 10 to 40 mCi of γ-rays with an energy of 478 keV. Radiation safety dictated many design choices for the experiment chamber. The chamber uses welded aluminum construction and is electrically isolated from the beamline, supports and vacuum equipment. A ferro-fluidic rotary feedthru is provided for the target arm. Ultra high vacuum is provided by a CryoTORR81 cryopump. Three 80 liter sorption pumps provide the chamber roughing vacuum. These sorption pumps will be used for cryopump roughing once high activity targets have been used in the chamber. The cryopump overpressure valve is replaced with a sealed check valve whose exhaust is plumbed to a HEPA type N filter that exhausts into the room. The three sorption pump outlets vent through another HEPA filter. The chamber is vented to overpressure, measured by a thin film pressure sensor, with dry nitrogen. The overpressure is relieved though a valve and a 6 by 6 inch HEPA filter exhausting to the room. While the chamber is open, negative pressure is maintained by a blower through the filter. All three filters are surveyed for 7Be contamination. The upstream beamline has been upgraded with a second cryopump whose overpressure is plumbed to the same HEPA filter, but very little contamination is expected here. Vacuum pressure measurement is done with hot cathode ion, pirani, and thermocouple gauges in both areas. Typical operating vacua are 1.0-3.0 × 10-7 Torr. The major source of 7Be contamination is sputtering by the beam. A liquid nitrogen cooled aperture and annular trap as close as possible to the target intercepts this sputtered material. A second liquid nitrogen trap is in the beamline just upstream of the chamber gate valve. Silicon diodes on the traps and sorption pumps provide liquid present, trap cold, and trap warm indicators. Two channels were modified to provide cryopump temperatures. Contamination from target evaporation due to beam heating is eliminated by water cooling the target and interrupting the beam for insufficient water flow. A 75 pound tungsten alloy cup can be raised to surround the target with about 6 cm of shielding, allowing personnel to work near the chamber. A pneumatic cylinder moves the shield and a mechanical latch keeps the shield up in case of loss of air or electrical power. Both are required to unlatch and lower the shield. A 80486 Personal Computer-based programmable controller2 provides flexible system control and graphic programming interface. It also runs a pushbutton and status light panel for normal user operation. This controller is dynamically programmed with the safety interlocks, vacuum interlocks, sorption pump pumping sequences, vacuum system controls, etc. * GANIL, BP 5027, 14076 Caen, Cedex 5, France. 1 CTI Cryogenics, Waltham, MA. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1993) p. 89. 34


  • Page 42

    4.3 WALTA: The Washington Large-area Time-coincidence Array J.G. Cramer, P.J. Doe, S.R. Elliott, D.J. Prindle, J.F. Wilkerson and the WALTA collaboration* We are forming a collaboration for the development of a distributed detector network. This system will support measurements of air showers from ultra-high energy cosmic rays and can also support a broad class of other physical measurements, for example a network of seismographs for geophysics studies. We call the overall network NNODE (Northwest Network for Operation of Distributed Experiments), and we call the cosmic ray measurement component WALTA (WAshington Large-area Time-coincidence Array). The project is to be a direct physical science outreach program between faculty and students of the University of Washington and the science teachers and students of Washington-area middle schools and high schools (grades 7-12). The WALTA1 part of the project is modeled on the ALTA initiative pioneered by the University of Alberta and currently being implemented in the Alberta Provincial School System. The local collaboration includes members of the Nuclear Physics Laboratory, physics department personnel in the cosmic ray and physics education groups, and a department of education faculty member from Seattle University. Each WALTA/NNODE measurement module is envisioned to consist of a computer with an Internet connection, a GPS timing system, and measuring equipment. The measuring devices will be of several types. The WALTA modules will consist of scintillation paddles to be placed at the school to detect distributed particle showers produced at the top of the atmosphere by ultra-high energy (~1020 eV) cosmic rays. The UW Geophysics Department group plans to install seismographs to the NNODE network that will become part of the Pacific Northwest Seismograph Network. Three other groups have expressed some interesting possibilities of adding additional measurement units to the NNODE network WALTA/NNODE aims to provide students the opportunity to become active participants in forefront scientific projects. A cornerstone of the program will be to install a WALTA/NNODE measurement module at each participating school. This will allow direct hands-on participation by the teachers and students in active experiments as collaborators with UW faculty and students. Each module will be supplied with special display and analysis software so that students and teachers can monitor both the local and the collective measurements made by WALTA/NNODE as well as use the data for class projects. In addition we will develop an educational program to help link aspects of the individual experiments with elements of the middle and high school science curriculum. We envision direct faculty visitation and involvement with the teachers and students as well as UW based workshops aimed at the teachers and students. It may even be possible to enlist the help of the students in the construction of the WALTA measurement modules. We plan for both UW graduate students and undergraduate students to actively participate in the project. The WALTA/NNODE modules involve state of the art hardware and software technology. Learning details about the technology utilized in the WALTA/NNODE modules (Field Programmable Gate Array electronic chips, custom CMOS electronic circuits, GPS receivers for determining absolute timing to 20 nanoseconds, real-time object oriented programming, distributed programming) may also be worth incorporating into the outreach program. The project offers a unique win-win aspect that does not depend on participant altruism for success. The participating scientist will be able to address one of the major unsolved problems in contemporary astrophysics with a unique measurement tool. The students and teachers will be able to learn about scientific techniques, mathematical tools, and the latest measurement technology. The teachers will be part of a rich environment of scientific research from which they can draw materials that give immediacy and emphasis to their teaching. * The non-NPL members of the WALTA collaboration include: Paula Heron and Peter Schaffer from the Physics Education Group; Jeffry Wilkes and Eric Zager from the Cosmic Ray Group; and Mark Roddy from the department of education at Seattle University. 1 The WALTA web page can be found at http://www.phys.washington.edu/~walta. 35


  • Page 43

    5.0 ULTRA-RELATIVISTIC HEAVY IONS 5.1 Event-by-event analysis overview L.D. Carr, D.J. Prindle, J.G. Reid, T.A. Trainor and D.D. Weerasundara The URHI program has as its goal the discovery and study of the quark-gluon plasma. We study full QCD in the neighborhood of the phase boundary between hadronic and partonic matter. Well-established study techniques already applied to this problem depend on inclusive analysis of multiparticle distributions. Event- by-event (EbyE) physics is a more recent effort to study event-wise dynamical fluctuations and correlations as a means to probe the detailed shape of the phase boundary. Some of these techniques have well-established analogues in the study of normal-matter phase transitions in bulk matter. But there are important differences having to do with finite systems, incomplete equilibration and relativistic kinematics. The development of EbyE techniques in heavy ion collisions has been pioneered by the UW group within the NA49 collaboration at SPS/CERN, and is now continuing with the STAR collaboration at RHIC/BNL. NA49 The NA49 EbyE program at the CERN SPS has had two major aspects: 1) application of scaled topological measures (SCA) to true event-wise characterization of event topology and comparison with references to search for special event classes having significant dynamical fluctuations in the final-state momentum distribution and 2) analysis of global momentum variables, some with thermodynamic analogs, to search for dynamical fluctuations on an inclusive basis. The SCA technique has proven to be a highly successful model-independent correlation analysis system. Current work focuses on an upgrade of the SCA analysis package to increase speed and flexibility, on establishing filters to eliminate instrumental effects and multi-collision pileup from candidate anomalous events and on identifying or establishing an upper limit on dynamical fluctuations. Global variables analysis is presently enjoying intense interest from the theoretical community and rapid technical development. Reexamination of fundamental statistical principles and fluctuation theory accompanies an expanding data analysis program. STAR Techniques pioneered by the UW group within the NA49 collaboration are now being imported to STAR as we anticipate the first RHIC beam this Summer. This program has required an extensive software development program within the RHIC Computing Facility (RCF) tailored to the needs of STAR EbyE analysis. EbyE software infrastructure provided by UW has been tested in a sequence of two RCF Mock Data Challenges (MDC1,MDC2) in the last six months. The STAR EbyE program is coordinated within the structure of the EbyE Physics Working Group. Activities in this group, in addition to global variables and SCA analyses which are UW specialties, include flow analysis, low-pt (e.g., DCC) analysis and high-pt (e.g., minijet) analysis. Because of the more complex dynamical picture expected to emerge at RHIC energies we anticipate strong overlap with other Physics Working Groups specializing in strangeness, high-pt and spectra physics as we attempt to unravel the various aspects of full QCD. EbyE Theory Application of scaling analysis and global-variables analysis to NA49 data has resulted in significant conceptual progress in understanding the basic correlation and statistical measures used. Progress includes a better understanding of finite-system (bounded scale interval) effects, the limitations of classical fluctuation theory and possible extensions, the basis of the Central Limit Theorem and possible extensions and the connection between system symmetries, scale dependence, fluctuations and correlations in a multiparticle final state and what these can tell us about the structure of the QCD phase boundary. Some of these results are quite general and may be broadly applicable to finite-system many-body problems. 36


  • Page 44

    5.2 NA49 SCA event-by-event analysis status L.D. Carr,* D.J. Prindle, J.C. Prosser,† J.G. Reid, T.A. Trainor, D.D. Weerasundara and the NA49 Collaboration# In 1998, we completed an event-by-event analysis of 300,000 central Pb+Pb collisions using Scaled Correlation Analysis (SCA)1 which proved to be a very powerful method to search for rare processes occurring in Pb+Pb collisions. As a result of this analysis, we identified a class of anomalous events having excess yield of charged particles in the bend plane φ = 0º, φ = 180º in the main TPC. Analysis and interpretation of anomalous event classes required a thorough search for instrumental effects and conventional hadronic physics as trivial origins of anomalous behavior, and simulation studies with standard and special Monte Carlo (MC) event generators to investigate anomalous behavior that may be nontrivial. A subsequent detailed analysis2 of these anomalous events revealed beam-gas event pile-up to be the principal source of enhanced secondary particle production in a majority of the anomalous events. In this case event pile-up means that a secondary nucleus or nucleon in the beam interacts with the primary target or a gas nucleus while information from a triggered central Pb+Pb collision is being read out by the detector. By comparing the distribution of charged particles associated with the primary vertex for several collision systems (e.g., pp, pA and AA), as well as from the pile-up vertex (x,y) coordinate distributions, we determined that the pile-up events resulted from a second Pb ion interacting with TPC gas downstream of the primary vertex within 20 µsec of the triggering collision. Pile-up events contribute a significant background to the search for true rare processes (perhaps due to new physics) for which the SCA was designed. Therefore, we need to develop methods to understand and remove the known backgrounds to extract true physics signals. We have demonstrated in the past3 that the slopes and intercepts of charged tracks from a common point of origin are correlated. An algorithm which exploits slope-intercept and impact-parameter-slope correlations has been developed to remove the pile-up events and to study in detail the remaining anomalous events. Work is in progress to study the efficiency of this algorithm in identifying pile-up events, especially those occurring in or near the primary Pb target. NA49 has just completed DST production for 400,000 central Pb+Pb collisions from the 1996 experimental run. A program is underway to analyze these events using SCA. * Physics Department, University of Washington, Seattle, WA. † University Computing, University of Washington, Seattle, WA. # CERN, Geneva, Switzerland. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 39. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 40. 3 Nuclear Physics Laboratory Annual Report, University of Washington (1996) p. 43. 37


  • Page 45

    5.3 NA49 pileup detection and EbyE analysis D.J. Prindle, J.G. Reid, T.A. Trainor and D.D. Weerasundara We have used a model-independent method to search NA49 data for dynamical fluctuations1 (see Section 5.2). In this analysis we found a class of anomalous events characterized by excess tracks in the Main TPCs near φ = 0º and φ = 180º. Previous analysis2 showed that conversions of γ-rays or decays of neutral particles could not account for the characteristics of these anomalous events. We were led to examine beam pileup as the probable cause for these events. After a beam-particle interaction in the target triggers the start of a readout it takes approximately 50 µs for all the ionization from the tracks passing through the TPCs to drift to the readout chambers. During this time a few beam particles will pass through the detector. A small fraction of these will have minimum bias interactions with the target or gas and create some number of particles, typically 20 or more, originating at a well defined vertex. These pileup interactions occur at random places and times. Some of them will create tracks that are indistinguishable from those tracks due to the primary interaction. In NA49 the magnetic fields are in the y direction, thus there is little curvature of the track in the y direction. When plotting y versus dy/dz at a reference z position the locus of tracks originating at a common vertex is a straight line. The slope of this line is related to the z position of the vertex. Plotting y versus dy/dz for all tracks we observe most of them to be along a line corresponding to the target position. Tracks from a pileup interaction are along another line, the slope of this line being determined by the z position of the pileup interaction. If the pileup interaction happens later in time, the ionization in the TPCs will reach the readout chambers late (since the clock is started by the trigger interaction) and the inferred y position will be lower, displacing the straight line locus. Thus even if the pileup is in the target it may result in a y versus dy/dz line parallel and distinguishable from the line due to the primary interaction. To search for pileup interactions we scan the y versus dy/dz plot after removing tracks along the target locus. We shear this plot parallel to y and project on y. When we shear by the right amount the pileup tracks all contribute to create a peak in a 1-dimensional histogram which is easy to find algorithmically. The amount of shear required to maximize the height of the peak determines the z position of the pileup interaction. Depending on where the pileup interaction occurs we find its tracks in different TPCs. If the pileup occurs in the target then all TPCs should be able to observe it. We have scanned all the anomalous events and found that the majority of them do have tracks from pileup interactions. Some of them show more than one pileup interaction. The majority of the pileup vertices in the anomalous event sample happened downstream of VT1, but were observable with VT2 and the MTPCs. We compare the z position of the pileup vertex as determined by VT2 and the MTPCs and find that they are consistent. Since the drift velocities in VT2 and the MTPCs were different the inferred y positions of the pileup interaction are different, but they are linearly related. The observation of this slope confirms that the pileup interactions are happening at random times and thus are not associated with the triggered beam particle. We still need to quantify the efficiency with which we can find pileup interactions as a function of z and also the number of particles coming from the interaction. We need to do this in order to see if pileup can explain the entire anomalous event sample. It is also important for other types of anomalous event searches to know what influence pileup can have. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 39. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 40. 38


  • Page 46

    5.4 Event by event global variables analysis J.G. Reid and T.A. Trainor Our recent work in event-by-event physics has focused on characterizing each event by a set of global variables (e.g., multiplicity, event-wise mean transverse momentum) and then analyzing the distributions of these event characteristics. We have carried out this analysis both on NA49 central Pb+Pb data and simulated data in the context of NA49 and STAR. For the purpose of characterizing the events we have focused on two separate analysis methods. First, we have examined the linear correlation coefficient between event multiplicity and eventwise mean transverse momentum. In proton-proton collisions there is an anti-correlation observed between these variables (in part due to simple 4-momentum conservation), so it is natural that we should investigate this correlation in the Pb+Pb data and N+N simulations. We are also interested in the multi-particle correlations which are present in the eventwise final-state transverse momentum distributions. This is an intractably difficult problem because of the large eventwise multiplicity, so we have limited ourselves to addressing 2-particle correlations, bearing in mind that this can be generalized to an n-particle approach using SCA. The original motivation for using these two quantities to characterize the data is their relationship to the Φ measure of Gazdzicki and Mrowczynski.1 Rewriting Φ in a more transparent form it contains a linear correlation coefficient and a 2-particle correlation measure: Z2 (P− Ne⋅p)2 Φ pt = N − σ pt = N − σ pt ✝ ✄ ~ 1 − ✞ ✟ ∑ pi p j − N e (N e − 1) ⋅p2 2 Nσ p t ✠✡ ✁✂ ✂ i≠ j ☎ ✆ ✆ ✑ ✔ ☛☞✌ ✍✎ ✏ − 2 p ⋅ N e < p >e − N e 2 2 ⋅p ✒✓ ✕ ✖ ✓ ✖ 2 2 = A +B 2 Nσ p t After a detailed study of Φ we have found it impossible to interpret the measure Φ unambiguously, so we have turned our focus to the more elementary measures which contribute. Rather than interpret a single number we look at the results of the individual analyses and interpret them separately. In applying these methods to NA49 central Pb+Pb data we have encountered some interesting results. There is a significant but small (lcc = -0.0307±0.003) anticorrelation between eventwise multiplicity and mean pt which is a factor of three smaller then the value observed for N+N collisions, but still nonzero. Also, looking at the 2-particle correlation space we have found the expected Bose-Einstein correlation peak, the Coulomb correlations, and correlations from resonance decays. But we have also identified significant trends in the data which are absent in the simulations and have yet to be fully understood. 1 M. Gazdzicki and St. Mrowczynski, Z. Phys. C 26, 127 (1992). 39


  • Page 47

    5.5 Scaled correlation analysis code update J.G. Reid After last year's successful effort1 to convert our main scaled correlation analysis code into an object- oriented framework using C++ the coding tasks this year have been minimal. There were several rounds of bug fixes and modifications to make the code more stable and results from default parameter analyses less opaque. Significant work has been put into better documentation, commenting and adding command-line options which make the code more distributable. A great deal of input from novice users was used in these developments. The only major change made was a different method for calculating the minimally correlated reference bin occupancies in the 'prior' SCA formalism. Previously we were weighting each bin in the support of the event equally at the scale of the minibinning. As one goes to large scales this can cause problems because a uniform distribution at the scale of the minibinning can be significantly non-uniform at larger scales. This led to uncontrolled behavior of the reference at large scale, and introduced a scale bias into the code. Dynamically weighting each bin of the minimally correlated reference in the support of the event at each scale point removed this problem. However, this may not be the ideal solution. We do not fully understand the implications of using a dynamically changing reference, although it has not presented any major problems yet. Of course, if the user has a specific reference distribution in mind this change makes no difference. The change will only effect model calculations and SCA theory development. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 43. 40


  • Page 48

    5.6 STAR mock data challenge: general DST design and the EbyE analysis chain L.D. Carr,* J.C. Prosser,† J.G. Reid, T.A. Trainor, D.D. Weerasundara and the STAR Collaboration# The RHIC Computing Facility (RCF) and four RHIC experiments successfully completed the Mock Data Challenge (MDC), in two phases, during 1998 Fall and 1999 early Spring.1 The RCF configuration during each phase of the MDC consisted of a Managed Data Server (MDS), a High Performance Storage System, a Central Reconstruction Server farm (CRS), and a Central Analysis Server farm (CAS). During both phases of the MDC, the RCF and the RHIC experiments exercised key aspects involved in recording and off-line analysis of experimental data. Among the goals achieved during the MDC1, for each of the four experiments, were: recording raw data, reading the raw data into the reconstruction farm and writing the Data Summary Tapes (DSTs) back out. An aspect common to both the CRS and the CAS operation is the definition of a DST. The DSTs are produced by the CRS operation while analyses running on CAS use DSTs as their input. A DST is defined to be a collection of event information obtained during event reconstruction, for all the events of a given set. The DST event information represents our best knowledge of the individual collision. This information should contain all the necessary information needed for further physics analyses which will lead to the extraction of physics signals. The UW group made a substantial contribution to the development of the DST objects specific to the MCD1 operation. This DST event model included objects that fully define the state of the hardware, state of the software, state of the collision, momentum space of the event, summary of the information content for each event and the summary of the information content for a given set of events. This event object model was successfully employed during the MCD1 to generate DSTs for about 190,000 reconstructed events. Additional to RCF goals for MDC2, STAR goals for MDC2 were: to exercise off-line software in ROOT environment, fully integrate conditional databases for calibrations and Tag Database for data mining, develop raw & off-line data formats and data models, establish well-defined CAS operations including post- DST analyses & micro-DST generation, and integrate Grand Challenge apparatus for data mining. The STAR collaboration has made significant progress toward achieving its MDC2 goals over the past several months. As part of the MDC2 program STAR generated 0.5 million events run through Geant Detector simulations for several detector configurations and has so far reconstructed about 20% of the total Geant events during the two- week period of MDC2 allocated to STAR. The UW group have taken a lead role in establishing a post-DST analysis program on the CAS farm. Central to the CAS operation is event-by-event (EbyE) physics analysis, a major emphasis of the UW group within STAR.2 The EbyE group intend to examine each of 107 events per year to extract information content, examine events for anomalous behavior, and sort the event population on the basis of any such behavior. This process involves interfacing with the MDS, the Tag Database, CAS batch job submission mechanism (LSF), etc. We have developed a preliminary analysis chain to be run on the CAS farm to perform an EbyE analysis. * Physics Department, University of Washington, Seattle, WA. † University Computing, University of Washington, Seattle, WA. # Brookhaven National Laboratory, Upton, NY. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 36. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p. 37. 41


  • Page 49

    5.7 Finite-size effects and fluctuations near a phase boundary T.A. Trainor General arguments based on universality have recently been put forward in support of a localized structure on the QCD phase boundary: a tricritical point or critical endpoint located at some intermediate position in ( T, µ B ). Detection and study of such a structure would be an important contribution to our understanding of full QCD. From an experimental viewpoint the question immediately arises what effect finite system size has on the observability of such a structure? To explore the consequences of finite system size on critical phenomena the Ising magnet serves as a paradigm. For an infinite lattice (bulk matter) below the critical point the magnetic susceptibility is singular and the magnetization is discontinuous. How does finite system size affect this picture? Finite scale interval, bounded by the characteristic size a of a spin (or more generally a typical momentum-carrying object) and the characteristic size L of a ‘causally connected’ or ‘equilibrated’ region, determines the maximum sharpness of structures near critical points. We can express the total free-energy density in terms of the individual free-energy densities of the two spin populations φ + and φ − . The overall free-energy density can be written: ✗ ✘ d d Φ( H ) ≈ − log e − φ + ( H )⋅( L/a ) + e − φ − ( H )⋅( L/a ) ⋅ (a / L )d 1 If L / a → ∞ and given the nearly linear dependence of φ ± on H near the origin we have Φ( H ) ≈ H for H ≈ 0 at and below the critical temperature, leading to an asymptotic (in scale interval L/a) singularity in the magnetic susceptibility. The functional form of M(H) depends parametrically on the scale interval (L/a)d as illustrated in Fig. 5.7-1. If the QCD phase boundary is ‘observed’ with a finite collision system dynamical fluctuations may be considerably reduced from bulk-matter expectations because of finite-scale-interval effects. Observability may depend critically on a number of experimental issues such as statistical power of the data sample, measure sensitivity and reduction of systematic effects. Quantitative predictions of fluctuation phenomena should include finite-system effects. 0.2 -Φ(H)/V ✣ 0.15 ✙ ✙ -φ+ ✚ -φ- 0.1 0.05 d 0 L increases -0.05 -0.1 -Φ/V ✛ -0.15 -0.2 ✣ ✣ ✣ -0.3 -0.2 -0.1 0 0.1 0.2 0.3 H 0.5 M(H) ✢ 0.4 0.3 0.2 0.1 0 -0.1 -0.2 -0.3 -0.4 -0.5 ✣ ✣ ✣ -0.3 -0.2 -0.1 0 0.1 0.2 ✜ 0.3 H Fig. 5.7-1. Sketch of free-energy Φ and magnetization M dependence on external field for the Ising lattice at T < Tc and various system sizes L. In the asymptotic limit L → ∞ the magnetization becomes discontinuous and the susceptibility becomes singular. Fluctuations near a phase boundary are expected to be large. For finite systems both quantities vary smoothly, and fluctuations may be greatly reduced and difficult to distinguish from finite-number statistics. 42


  • Page 50

    5.8 Linear near response coefficients and fluctuations: beyond the gaussian model T.A. Trainor Critical-point phenomena are accompanied by fluctuations with characteristics determined by the dependence of the free energy on relevant state variables. The standard treatment of fluctuations is the Einstein/gaussian theory or linear response theory. This standard treatment cannot predict fluctuation properties in the neighborhood of a critical point. An equilibrated system can be characterized by a free energy φ = ( TS, A,G...) . A local extensive state variable can then be expressed in terms of its conjugate intensive variable and a free energy as X α = ∂Φ / ∂α . Linear response coefficients χ α (CV ,κ T ,...) can be defined by χ α = ∂X α / ∂α . The gaussian fluctuation model (free energy assumed quadratic in the neighborhood of its minimum) implies that a fluctuation variance is directly proportional to a corresponding linear response coefficient: σ 2Xα ∝ χ α . These relationships are represented in Fig. 5.8-1. The E(T) dependence is piece-wise linear, corresponding to two different DoF (degrees of freedom) density regions separated by a phase boundary. Near the critical temperature fluctuations increase in amplitude but are bounded, whereas the linear response coefficient (slope of E(T)) may be arbitrarily large. Paradoxically, gaussian fluctuation theory would predict σ 2χ α → ∞ when fluctuations are in fact bounded. The paradox is resolved by noting that linear response coefficients measure the curvature of the free energy surface near its minimum. If the free energy surface is approximately quadratic in the neighborhood of its minimum then σ 2χ α ∝ χ α is a good approximation. However, near a critical point the free energy curvature and the width may have no well-defined relationship. The guassian treatment breaks down and linear response coefficients cannot predict or represent fluctuation amplitudes. A separate determination must be made of the local variance density and the linear response coefficients. Both linear response coefficients and fluctuation amplitudes are in principle measurable. Instead of being redundant (as in gaussian theory) they actually provide complementary information about the underlying system Hamiltonian. ✬ ✭ 300 10 E S(E,T) - E(T)/T same curvature 9 250 ρ2 ✤ ✱ 8 200 7 ✰ 6 150 5 ✯ 4 100 3 ✤ ρ1 ✥ 2 ✮ 50 ✭ 1 0 ✬ 0 ✲ ✭ ✬ 0 100 200 300 0 100 200 300 T E(T) 10 σN2/V, β *σE/V, ρ *σV/V ✦ ✪ 9 2 8 2 ✩ 7 variance upper limit ✫ ✦ ★ 6 2 5 ρ2 4 2 ✧ 3 variance lower limit ✦ 2 ρ1 ✥ 1 αc 0 ✲ ✲ -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 α Fig. 5.8-1. Sketch of E(T) dependence near a phase transition (top left), fluctuation distribution for various temperatures (top right) and plot of fluctuation amplitude vs temperature estimated by linear response coefficients (upper limit) and DoF density (lower limit) (bottom). 43

  • View More

Get the full picture and Receive alerts on lawsuits, news articles, publications and more!