avatar University of Washington Inc Services

Pages

  • Page 1

    ANNUAL REPORT Center for Experimental Nuclear Physics and Astrophysics University of Washington May, 2003 Sponsored in part by the United States Department of Energy under Grant #DE-FG03- 97ER41020/A000.


  • Page 2

    This report was prepared as an account of work sponsored in part by the United States Government. Neither the United States nor the United States Department of Energy, nor any of their employees, make any warranty, expressed or implied or assumes any legal liability or responsibility for accuracy, completeness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe on privately-owned rights. Cover photos, clockwise from upper left: One of the SNO NCD welding teams, Sean McGee, Tom Burritt, Ian Lawson (Guelph) and Peter Doe, preparing a pre-deployment-welded NCD segment for storage in the mine at Sudbury. Kareem Kazkaz and Jeremy Kephart(NC State) verifying the performance of the SEGA crystal after taking possession of it from Ortec. Smarajit Triambak and Cristina Bordeanu, adjusting the target position for a Tandem ex- periment. The fixture for welding the NCDs.


  • Page 3

    UW CENPA Annual Report 2002-2003 May 2003 i INTRODUCTION CENPA pursues a broad program of research in nuclear physics, astrophysics and related fields. Research activities are conducted locally and at remote sites. The current program includes “in-house” research on nuclear collisions and fundamental interactions using the local tandem Van de Graaff, as well as local and remote non-accelerator research on fundamental interactions and user-mode research on relativistic heavy ions at large accelerator facilities in the U.S. and Europe. We are pleased to welcome Professor Alejandro Garcı́a, who joined our faculty in September 2002, and Professor Askel Hallin, on sabbatical from Queen’s University. We have completed our “Phase II” 7 Be(p,γ)8 B measurements, extending our earlier work to lower energy and reducing our systematic errors. Our new S17 (0) determination is in excellent agreement with our Phase I value published in 2002. Our measurements remain the most precise determination to date of this important reaction rate. We have developed a 8 B radioactive beam at the Tandem with a flux of half a dozen 8B per second, and have accumulated about 3 × 105 coincidence-tagged 8 B decays, in an experiment searching for a branch to the ground state of 8 Be. We are presently analyzing these data, searching for a signal from the 92-keV alpha particles from the ground state of 8 Be. Event structure analysis of RHIC Au-Au collisions has revealed a number of new corre- lation phenomena related to 1) initial-state multiple scattering and its variation with event centrality and 2) local measure conservation at hadronization and its variation with changes in the prehadronic medium, both indicating the development of a collective medium with dissipative properties. Fluctuation scaling results have been successfully inverted by a nu- merical process to construct autocorrelation distributions which are directly interpretable in terms of physical models. HBT interferometry with STAR data at RHIC have in the past three years provided significant challenges to conventional theoretical models of RHIC physics. The latest such challenges, reported here, show rising pion phase space density and decreasing entropy per particle with centrality, both results suggesting the onset of some unknown low entropy process in central RHIC collisions. The SNO detector has been in operation since June 2001 with 0.2% NaCl added to the heavy water to enhance detection of neutrons released by neutral current interactions. Analysis of 254 live days of that data set is nearly complete and a paper is being prepared for publication in the summer of 2003. As expected, a strong neutron peak is seen, but because a “blind” analysis is being performed the implied NC rate is not yet known. Quite good statistical separation of NC events from charged-current ones based on the differences in anisotropy of Čerenkov light is being realized. The neutral current detector array is complete, and the detectors have been welded to- gether in preparation for deployment. Deployment hardware is in hand and deployment crews are trained in readiness for installation of the array in late 2003. The MOON 100 Mo double beta decay experiment research at CENPA is now focused on development of a bolometric method that could provide high energy resolution and the


  • Page 4

    ii ability to tag double-beta transitions to excited states in 100 Ru. In collaboration with UW condensed-matter physicists a dilution refrigerator is being set up to make preliminary mea- surements of the specific heat and thermal diffusivity of superconducting and normal Mo at millikelvin temperatures. The emiT experiment, a search for time violation in neutron beta decay, has been collect- ing data at the NIST Center for Neutron Research since the fall of 2002. To date, it has been a highly successful run, collecting over 150 million coincidences. The collaboration plans to continue acquiring data and studying potential systematics through the end of 2003. Development of the world’s most sensitive direct kinematical neutrino-mass measurement, via tritium beta-decay, has been proceeding rapidly. The University of Washington intends to provide the detector system for this international experiment that will be constructed at the Forschungszentrum Karlsruhe in Germany. The Majorana collaboration hopes to construct a 76 Ge-based next generation 0-neutrino double-beta-decay detector. Our efforts have concentrated on collaborating with scientists at Pacific Northwest National Laboratories to construct a prototype Multiple-Element Germa- nium Assay (MEGA) detector that consists of an array of 18 high purity Ge detectors. Progress on the 199 Hg EDM experiment includes increased reliability of the 254-nm laser system and a reduction in discharges occurring in the high-voltage cables. Data are currently being taken towards a new measurement. We are developing an experiment to measure the beta asymmetry from neutron beta decay with high accuracy to elucidate the apparent non-unitarity of the CKM matrix. We have started experiments at CENPA to reduce uncertainties regarding high-precision measurements to determine the positron-neutrino correlation and the log(ft) value for the 0+ → 0+ decay of 32 Ar. Torsion-balance experiments have demonstrated that, for distances down to 90 µm, there is no force that violates Newton’s inverse square law coupling to mass with equal or larger strength than gravity. New developments include 1) a pendulum for testing coupling to spin, 2) a new equivalence principle test, and 3) an ‘anapole’ pendulum for an axion search. As always, we encourage outside applications for the use of our facilities. As a conve- nient reference for potential users, the table on the following page lists the capabilities of our accelerators. For further information, please contact Prof. Derek W. Storm, Executive Director, CENPA, Box 354290, University of Washington, Seattle, WA 98195; (206) 543- 4080, or storm@npl.washington.edu. Further information is also available on 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 are listed alphabetically, with the primary author underlined, to whom inquiries should be addressed. Derek Storm, Editor Barbara Fulton, Assistant Editor


  • Page 5

    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). Recently adapted to an (optional) terminal ion source and a non-inclined tube #3, which enables the accelerator to produce high intensity beams of helium and hydrogen isotopes at energies from 100 keV to 5.5 MeV. Some Available Energy Analyzed Beams Ion Max. Current Max. Energy Ion Source (particle µA) (MeV) 1H or 2 H 50 18 DEIS or 860 3 He or 4 He 2 27 Double Charge-Exchange Source 3 He or 4 He 30 7.5 Tandem Terminal Source 6 Li or 7 Li 1 36 860 11 B 5 54 860 12 C or 13 C 10 63 860 ∗14 N 1 63 DEIS or 860 16 O or 18 O 10 72 DEIS or 860 F 10 72 DEIS or 860 ∗ Ca 0.5 99 860 Ni 0.2 99 860 I 0.001 108 860 *Negative ion is the hydride, dihydride, or trihydride. Additional ion species available including the following: Mg, Al, Si, P, S, Cl, Fe, Cu, Ge, Se, Br and Ag. Less common isotopes are generated from enriched material. In addition, we are now producing a separated beam of 15-MeV 8 B at 6 particles/second. BOOSTER ACCELERATOR See “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). The Booster is presently in a “mothballed” state.


  • Page 6


  • Page 7

    UW CENPA Annual Report 2002-2003 May 2003 v Contents INTRODUCTION i 1 Fundamental Symmetries and Weak Interactions 1 Weak Interactions 1 1.1 A second run of the emiT experiment . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Status and updates to emiT DAQ . . . . . . . . . . . . . . . . . . . . . . . . 4 1.3 Parity non-conserving neutron spin rotation in liquid helium . . . . . . . . . . 5 1.4 Beta asymmetry from ultra-cold neutrons . . . . . . . . . . . . . . . . . . . . 7 1.5 Limits on scalar currents from the decay of 32 Ar . . . . . . . . . . . . . . . . 8 1.6 Search for a permanent electric dipole moment of 199 Hg . . . . . . . . . . . . 9 Torsion Balance Experiments 10 1.7 Sub-mm test of Newton’s Inverse-Square Law . . . . . . . . . . . . . . . . . . 10 1.8 Spin pendulum update . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.9 Eötwash data acquisition system development . . . . . . . . . . . . . . . . . . 12 1.10 A new equivalence principle test . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.11 Development of an ‘anapole pendulum’ . . . . . . . . . . . . . . . . . . . . . . 14 1.12 Small force measurements for LISA . . . . . . . . . . . . . . . . . . . . . . . . 15 2 Neutrino Research 16 SNO 16 2.1 The Sudbury Neutrino Observatory . . . . . . . . . . . . . . . . . . . . . . . . 16 2.2 Status and updates to the SNO data acquisition system . . . . . . . . . . . . 17 2.3 Energy and optical calibration of the Sudbury Neutrino Observatory . . . . . 18 2.4 Selection of the neutrino analysis data set for the Salt Phase of SNO . . . . 19 2.5 Distinguishing muon spallation types in SNO . . . . . . . . . . . . . . . . . . 20


  • Page 8

    vi 2.6 Electron antineutrino detection at the Sudbury Neutrino Observatory . . . . 21 2.7 Reactor antineutrinos at the Sudbury Neutrino Observatory . . . . . . . . . 22 2.8 The day-night asymmetry measurement in the salt phase of SNO . . . . . . 23 2.9 SNO signal extraction in the salt phase . . . . . . . . . . . . . . . . . . . . . 24 SNO NCDs 25 2.10 NCD data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.11 NCD array optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 2.12 Determination of Z-position for hits in the NCD array . . . . . . . . . . . . . 27 2.13 Data acquisition for SNO’s neutral current detectors . . . . . . . . . . . . . . 28 2.14 NCD underground status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.15 NCD deployment equipment progress and training . . . . . . . . . . . . . . . 30 2.16 Progress of the underground NCD welding prior to deployment . . . . . . . . 31 Neutrino Detectors 32 Double Beta Decay 32 2.17 Majorana update: construction, evaluation, simulation . . . . . . . . . . . . . 32 2.18 Electron-capture branch of 100 Mo and the efficiency of MOON . . . . . . . . 33 KATRIN 34 2.19 Characterizing silicon detectors for KATRIN . . . . . . . . . . . . . . . . . . 34 3 Nuclear Astrophysics 35 3.1 7 Be(p,γ)8 B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.2 Target composition analysis for 7 Be(p,γ)8 Be S-factor measurement . . . . . . 36 3.3 Search for the 8 B(2+ ) → 8 Be(0+ ) ground state transition . . . . . . . . . . . 37 3.4 Is e+ e− pair emission important in the determination of the 3 He + 4 He S- factor? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38


  • Page 9

    UW CENPA Annual Report 2002-2003 May 2003 vii 4 Nuclear Structure 39 4.1 Testing the isospin multiplet mass equation and its implications . . . . . . . . 39 4.2 Low-temperature measurement of the giant dipole resonance width . . . . . 40 5 Relativistic Heavy Ions 41 5.1 Introduction to event-structure analysis . . . . . . . . . . . . . . . . . . . . . 41 5.2 Mean-pt fluctuations and minijet production in Hijing-1.37 . . . . . . . . . . . 42 √ 5.3 Mean-pt fluctuation scaling on (η, φ) in Au-Au collisions at sN N = 200 GeV 43 5.4 Scale dependence of net-charge fluctuations . . . . . . . . . . . . . . . . . . . 44 5.5 Relating fluctuations and correlations in heavy-ion collisions . . . . . . . . . . 45 5.6 Determination of auto-correlation using net-charge fluctuations . . . . . . . . 46 5.7 Centrality dependence of pt fluctuations in Au-Au collisions . . . . . . . . . 47 5.8 Charge-independent pt ⊗ pt correlations in Au-Au collisions at 130 GeV . . . 48 5.9 Charge-dependent correlations in axial-momentum space . . . . . . . . . . . . 49 5.10 Hard and soft scattering in p-p collisions . . . . . . . . . . . . . . . . . . . . . 50 5.11 Dimensionality of a strange attractor as a function of scale . . . . . . . . . . 51 5.12 Dimensionality of a strange attractor as a function of position . . . . . . . . . 52 5.13 Overview of HBT physics at STAR . . . . . . . . . . . . . . . . . . . . . . . . 53 5.14 Pion phase space density and “Bump Volume” . . . . . . . . . . . . . . . . . 55 5.15 Entropy at freeze-out in RHIC collisions . . . . . . . . . . . . . . . . . . . . . 57 5.16 Opacity effects in Bose-Einstein correlation radii at RHIC . . . . . . . . . . . 59 5.17 Testing the Bowler-Sinyukov-CERES Coulomb-correction procedure with same- vs. opposite-charge pion correlations . . . . . . . . . . . . . . . . . . . . . . . 60 5.18 Particle identification in STAR TPC . . . . . . . . . . . . . . . . . . . . . . . 63 6 Molecular Clusters 64 6.1 Attempt to produce dianions of Mg2 S3 . . . . . . . . . . . . . . . . . . . . . 64 7 Electronics, Computing, and Detector Infrastructure 65


  • Page 10

    viii 7.1 Electron gun for profiling silicon detectors for KATRIN . . . . . . . . . . . . 65 7.2 Nanopore DNA sequencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 7.3 Electronic equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 7.4 PC based data acquisition system using JAM . . . . . . . . . . . . . . . . . . 69 7.5 Status of an advanced object oriented real-time data acquistion system . . . 70 7.6 Laboratory computer systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 8 Accelerator and Ion Sources 72 8.1 Van de Graaff accelerator operations and development . . . . . . . . . . . . . 72 8.2 Injector deck and ion sources . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 8.3 A 8 B beam at the Tandem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 9 The Career Development Organization: A Student Organization 76 10 CENPA Personnel 77 10.1 Faculty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 10.2 Postdoctoral Research Associates . . . . . . . . . . . . . . . . . . . . . . . . 77 10.3 Predoctoral Research Associates . . . . . . . . . . . . . . . . . . . . . . . . . 78 10.4 Research Experience for Undergraduates participants . . . . . . . . . . . . . . 78 10.5 Professional staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 10.6 Technical staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 10.7 Administrative staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 10.8 Part Time Staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 11 Publications 81 11.1 Degrees Granted, Academic Year, 2002-2003 . . . . . . . . . . . . . . . . . . . 90


  • Page 11

    UW CENPA Annual Report 2002-2003 May 2003 1 1 Fundamental Symmetries and Weak Interactions Weak Interactions 1.1 A second run of the emiT experiment J. F. Amsbaugh, L. Grout,∗ H. P. Mumm, A. W. Myers, P. Parazzoli,† R. G. H. Robertson, K. Sundqvist,‡ 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. Current observations of CP(T) violation in the Kaon and B-meson systems can be accommodated within the standard model of particle physics. However, baryogensis as well as attempts to develop unified theories, indicate that additional sources are required. The standard model predicts T-violating observables in beta decay to be extremely small (second order in the weak coupling constant) and hence to be beyond the reach of modern experiments.1 However, potentially measurable T-violating effects are predicted to occur in some non-standard models such as those with left-right symmetry, exotic fermions, or lepto- quarks.2,3 Thus a precision search for T-violation in neutron beta decay provides an excellent test of physics beyond the Standard Model. The emiT experiment in sensitive to the T-odd P-even triple correlation between the neutron spin and the momenta of the neutrino and electron, Dσn · Pe × Pν , in the neutron beta-decay distribution. The coefficient of this correlation, D, is measured by detecting decay electrons in coincidence with recoil protons from a polarized beam of cold (2.7 meV) neutrons. Four electron detectors (plastic scintillators) and four proton detectors (large-area diode arrays) are arranged in an alternating octagonal array concentric with the neutron beam. During the first run, high voltage related problems damaged electronic components, led to high background rates and ultimately produced a non-symmetric detector. Systematic effects were less effectively canceled due to the lack of full detector symmetry and a more complex data analysis scheme was required. The result, D = −0.1 ± 1.3 × 10−3 , represents a small improvement over the current world average.4 To solve these problems, a major overhaul of the emiT detector was started in 1999. A full redesign of the proton focusing assembly maintained focusing efficiency, while reducing high field regions and minimizing the associated field emission, the dominant background during the first run. To reduce dead- layer-proton-energy loss, surface barrier detectors having a Au layer of 20 µg/cm2 , a depletion ∗ Presently at MIT Lincoln Labs, Lexington, MA 02420. † Presently at Los Alamos National Laboratory, Los Alamos, NM 87545. ‡ Presently at Department of Physics, University of California, Berkeley, CA 94720. 1 M. Kobayashi and T. Maskawa, Prog. Theor. Phys. 49, 652 (1973). 2 P. Herczeg, 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). 4 L. J. Lising et al, Phys. Rev. C 62, 055501 (2000).


  • Page 12

    2 (a) 4 10 (b) (linear scale) 250 Proton singles @ 28 kV 100 Beam shutter closed (four hour runs) 200 3 10 80 150 Proton energy (keV) Counts 60 100 2 10 50 40 1 0 10 20 0 10 20 40 60 80 100 120 140 0 1000 2000 3000 4000 Energy (keV) Proton-electron time difference (ns) Figure 1.1-1. Raw proton data for surface barrier A16. (a) Pulse height singles spectrum of decay protons for a typical four hour run. The FWHM is 4.78 keV. The peak near 25 keV in the ‘beam shutter closed’ data, which is the smaller amplitude graph, is from protons and is below 0.6 Hz in all channels. The low energy peak is the minimum ionizing peak in silicon. (b) Coincidence plot of the same data, proton energy vs delay time. Note very low backgrounds. region of 300 microns and an active area of 300 mm2 are being used for the second run. A new liquid nitrogen based system was installed and was critical to achieving acceptable energy resolution. Major upgrades to the preamps, shaper/ADC boards, timing electronics and data acquisition code have been completed. A low-energy < 800 eV, low-intensity proton source, was constructed to facilitate in situ characterization of our detectors.5 Finally, measurements indicate that NIST’s Center for Neutron Research reactor upgrade has yielded the expected factor of 1.8 increase in flux. In June of 2002, the emiT detector was moved to NIST’s Center for Neutron Research. A test stand was set up to allow testing of the individual proton and electron detectors off-line. While final beamline development and magnetic field alignment were being carried out a number of independent detector validation tasks were carried out. While the high voltage behavior was much improved since emiT’s last run, see Fig. 1.1-1, tests lead to a repolishing of the electrodes and minor modifications of the electrode hardware. Cooling tests indicated that surface barrier leakage current was higher than expected. To mitigate the negative effects on resolution, preamp supply voltage was lowered, reducing heat load. Some alterations were also made to the cooling hardware. By September, the detector was installed on the neutron beamline and was fully func- tional. The proton and electron detectors were each calibrated with a selection of gamma sources. At this point it became clear that high background rates were leading to an unac- ceptably high dead time. This problem was attacked from a number of angles. First, timing and gate widths in both the proton and electron electronics chains were optimized. Second, by moving the neutron beam stop half a meter downstream, using lithium plastic in place 5 F. Naab et al., Nucl. Instrum. Methods B, 197, 278 (2002).


  • Page 13

    UW CENPA Annual Report 2002-2003 May 2003 3 of boroflex, and by careful shielding of the spin flipper, background rates in the beta detec- tors was reduced by fifty percent. Attempts were also made to improve the data acquisition software performance. EmiT is now collecting data at a coincidence rate of 28 Hz. Approximately 15 million events have been collected, giving a statistical sensitivity superior to all previous measure- ments. A preliminary test of the dominant systematic effect has been conducted and in- dicates that systematics are below the expected sensitivity of the apparatus. EmiT will continue to take data through August of 2003 and expects to reach the design sensitivity of D ≈ 2 × 10−4 .


  • Page 14

    4 1.2 Status and updates to emiT DAQ M. A. Howe and J. F. Wilkerson The emiT data acquisition (emiTDAQ) software that is being used to collect data for the emiT experiment was installed at NIST during the summer of 2002. Since the emiTDAQ software was described in last year’s annual report,1 only a brief overview of the most significant updates and operational status is provided here. One of the first major upgrades to emiTDAQ after installation was to move the event readout off the Mac and into a VME-based embedded CPU (eCPU). The eCPU runs a small stand-alone application that monitors the CENPA-built 100-MHz latched clock for event signals, reads out hardware as required, bundles the data together with the timing information from the latched clock, and finally places the data into dual memory. The main system reads the data from dual port memory, builds events that fall within a slow coincidence-timing window by using the timing information, and stores the built data to disk. A user specified fraction of non-coincidence data are recorded to disk as well. Data quality and monitoring tools were developed to allow the quality of the data stream to be monitored online in real-time. Histograms of all the shaper ADC, TDC, and QDC data are kept as well as 2D histograms of the coincidence data. Data rates are also calculated and displayed so that the operator can quickly visualize the overall state of the detector, find dead channels, diagnose gain/threshold problems, etc. Integrated control of various system parameters is done using an Acromag IP220 DAC module. One of the parameters controlled is the neutron beam polarization spin direction which is flipped every ten seconds. Analog parameters such as detector temperatures are readout using an Acromag IP320 I/O module and an 1151N scaler module. For monitoring the health of the detector, all the analog data can be displayed, both in table form as well as in strip-chart form. High/Low alarm limits can be set on all parameters. In addition to sounding audible alarms, notifications can be e-mailed to the cell phones or pagers of interested parties. Constraints can be specified for which alarms are sent and acceptable time windows during which to send them. Since the emiT detector is not manned over-night, the email alarm system has helped alert on call operators to serious incipient problems with the detector which would have resulted in detector downtime and/or actual harm to the detector. A completely separate system was added to read out the magnetometer data using Lab- View. The magnetometer data are transferred to the emiTDAQ approximately once per minute and are stored in the data stream. The emiT DAQ system has been running almost continuously since installation and con- tinues to perform reliably. No further upgrades are expected. 1 CENPA Annual Report, University of Washington (2002) p. 1.


  • Page 15

    UW CENPA Annual Report 2002-2003 May 2003 5 1.3 Parity non-conserving neutron spin rotation in liquid helium E. G. Adelberger, A. Garcı́a, B. R. Heckel and H. E. Swanson We have undertaken a project to measure the PNC neutron spin rotation in a 50 cm long liquid helium target to a precision of 10−7 rad and to explore the possibility of measuring the PNC interaction between bare neutrons and protons. This project has been supported by the NSF in addition to the support received from the DOE through CENPA. The collaboration for this project includes E. Adelberger, A. Garcia, B. Heckel and E. Swanson from CENPA, M. Snow and C. Bass from Indiana University, and D. Haase and D. Markoff from TUNL. Despite a considerable body of data of PNC measurements in nuclei, it has not been possible to extract a consistent picture of the weak force between nucleons. Uncertainties about the nuclear wave functions make it difficult to compare experiment to theory unam- biguously. The weak isovector coupling constant, fπ , is of particular interest because it is strongly influenced by the neutral current Z 0 exchange, yet it has proven to be very difficult to measure. We have proposed to measure fπ in PNC n− alpha scattering, a few-body nucleon system that may be calculated with relative confidence. The PNC rotation of the spin vector of a neutron beam as it traverses a liquid helium target, φpnc , provides a measurable observable: φpnc (n − α) ∝ +fπ + .33h0ρ + .23h0ω − .11h1ρ − .23h1ω − .02hρ1 where the coupling constants are the standard DDH amplitudes.1 If fπ vanishes, then PNC p-α measurements, the isospin conjugate system to n-alpha, predict a value for φpnc of ap- proximately 5 × 10−7 rad in a 50-cm long liquid-helium target, five times larger than the experimental limit we propose to achieve. Cold neutrons, having de Broglie wavelengths longer than 0.4 nm, propagate through material targets as neutron waves. The target averaged interaction with the neutrons creates a neutron index of refraction. PNC neutron spin rotation arises from a term G σ · p in the index of refraction that is generated by the weak interaction (where σ is the Pauli spin and p is the momentum of the neutron, and G is proportional to the universal weak coupling constant.) The spins in a neutron beam whose polarization vector is transverse to p will experience a torque about p, resulting in a total rotation angle of: φpnc = −4πN LRe(G ) independent of neutron momentum, where N is the number density of target nuclei and L the target length. The experimental challenge is to distinguish the small PNC rotations from much larger neutron spin rotations due to residual magnetic fields. We have constructed a neutron spin polarimeter that includes a cryostat surrounding a liquid helium target region and have performed the first measurement of the pnc neutron spin rotation in a liquid helium target. 1 V. F. Dmitriev, V. V. Flambaum, O. P. Sushkov and V. B. Telitsin, Phys. Lett. 125, 1 (1983).


  • Page 16

    6 The first data runs with the cryogenic polarimeter, taken at the NIST reactor, achieved a result of φpnc = (3.8 ± 6.5) × 10−7 rad, limited by count rate shot noise.2 To achieve an additional factor of five in experimental sensitivity, we are in the process of rebuilding the polarimeter. There are three essential improvements. The first is to redesign the cryostat so that superfluid helium can be used as a target. Superfluid helium has a density 20% higher than normal liquid helium; it will scatter 30% less neutrons out of the beam; and it provides better thermal conductivity to reduce systematic errors. The second improvement is to employ a long wavelength neutron filter in the beam to remove the longest wavelength neutrons. This is expected to increase the detected beam polarization by 50%. Finally, the beam collimation will be opened to allow a larger count rate. In the past year, we have completed the design for a superfluid-helium target chamber and the construction of the modified cryostat is underway. We have purchased the associated vacuum hardware and the components for a new data acquisition system. After the emiT experiment completes its run at the NIST reactor, we expect to begin mounting the apparatus for the spin rotation measurement. In 18 weeks of beam time, we should have the statistics to measure φpnc at the level of 10−7 rad. 2 D. Markoff, Ph.D. thesis, University of Washington, (1997).


  • Page 17

    UW CENPA Annual Report 2002-2003 May 2003 7 1.4 Beta asymmetry from ultra-cold neutrons A. Garcı́a, A. Sallaska and E. Tatar∗ Neutron β decay presents a unique opportunity to test the present evidence for the non- unitarity of the CKM matrix:1 Here isospin-breaking corrections are negligible and radiative corrections are well understood. However, because neutron decay is a mixed (Gamow-Teller and Fermi) β decay, one needs to measure, in addition to the half life, some other quantity, like the β asymmetry. Several previous measurements of the β asymmetry have disagree- ments that go beyond their claimed uncertainties, indicating that some of those experiments have unknown systematic errors. One main source of problems in neutron β-asymmetry mea- surements is that the polarization of neutrons needs to be known with very high accuracy. We have joined a collaboration2 to produce a measurement of the β asymmetry in neutron decay using ultra-cold neutrons (UCN). Neutrons are produced at LANL by spallation and then moderated by scattering on graphite and solid deuterium. Once the velocities of neutrons are on the order of a few m/s they reflect off the walls of guides and are sent through a 7-Tesla solenoidal field that fully polarizes them. UCN production has been successfully tested at LANL, and an improved version of the moderator is under construction. The UCNs will be sent to a spectrometer consisting of a ≈ 1 Tesla holding field (with non-uniformities of less than 10−3 ). There the UCNs will be contained by a neutron guide made of quartz with diamond coating (to avoid de-polarization and losses) in a volume of ≈ 3 meters of length and 10 cm in diameter where they will be left to decay and observed by two detectors on both ends. The detectors will be a combination of a position-sensitive gas ∆E plus a plastic scintillator or a position-sensitive Si detector. A proposal to NSF and DOE has been reviewed and the project has been funded. Presently we are working on performing Monte Carlo calculations to understand how to minimize potential backgrounds, designing a system to monitor the intensities of UCNs in- side the spectrometer, and putting together a system to perform measurements potential backgrounds in the area of the experiment. ∗ Idaho State University, Department of Physics, Pocatello, ID 83209. 1 I. S. Towner and J. C. Hardy, Phys. Rev. C 66, 035501 (2002) and references therein. 2 The collaboration is composed mainly of people from Caltech, LANL, NCSU, Virginia Tech, and UW.


  • Page 18

    8 32 1.5 Limits on scalar currents from the decay of Ar E. G. Adelberger, A. Garcı́a and H. E. Swanson The determination of the e − ν correlation in the 0+ → 0+ β decay of 32 Ar allows for placing limits on scalar contributions to the weak interactions, which appear naturally in extensions of the Standard Model, such as Lepto-quarks, and Supersymmetries.1 In our experiment1 for determining the e − ν correlation by measuring the beta-delayed protons from 32 Ar,we used as a calibration protons emitted from the T=3/2 state in 33 Cl. In Section 4.1, we briefly describe recent developments with respect to the mass of this state. We are presently re-analyzing the data from the 32 Ar experiment taking into account these developments. Briefly, the mass of the calibration point from 33 Cl has shifted up by about 5 keV.2 Because this was the only calibration point available at proton energies above 2 MeV and the other calibration points remained at their former values, this implies a change in the energy calibration of the spectrum. The latter, by itself, would change the conclusions from our data drastically. However, this change also implies a change in the mass of the T=2 daughter state in 32 Cl which modifies the Q-value, and changes the proton’s energy, both of which modify the expected shape of the proton line. These factors end up canceling each other and the answer remains in agreement with the predictions of the Standard Model. However, the correct determination of the uncertainties and limits on the scalar couplings require careful consideration of many factors, and this is in the works. In addition, we are working on a careful analysis of the data to determine evidence for isospin breaking. 1 E. G. Adelberger et al., Phys. Rev. Lett. 83, 1299 (1999). 2 Pyle et al., Phys. Rev. Lett. 88, 122501-1 (2002).


  • Page 19

    UW CENPA Annual Report 2002-2003 May 2003 9 199 1.6 Search for a permanent electric dipole moment of Hg W. C. Griffith, M. D. Swallows,∗ M. V. Romalis† and E. N. Fortson∗ We are working on a four vapor cell measurement to improve the limit on the permanent electric dipole moment (EDM) of 199 Hg. The measurement of a finite EDM would reveal a new source of CP violation beyond the standard model. Currently, the best limit on an atomic EDM is given by our previous measurement1 using a two vapor cell setup, which gave a limit of |d(199 Hg)| < 2.1 × 10−28 e cm. Upgrades implemented in our current measurement, described in a previous report,2 have enhanced our statistical sensitivity per unit time by a factor of 3. We began taking data towards the new measurement during the past year. Data is taken under a variety of conditions, including the magnetic field direction, high voltage ramp rate, and the direction the probe wavelength is detuned off resonance. These conditions are changed according to a schedule based on factorial design principles, which ensures that the effects of these parameters are sampled evenly throughout the entire data set. We have occasionally observed magnetic fields correlated with the electric field direction that create EDM-like signals. This leads to long delays as the source of the problem must be sought out. One problem found was that discharges were occurring between the high voltage (HV) cables and the cell-holding vessel, as the HV was ramped between ±10 kV, creating magnetic fields large enough to orient possible residual ferromagnetic contaminants. The HV finishes ramping before the spin precession is probed, but magnetic fields from the ferromagnetic impurities would persist into the probe phase, leading to a signal that mimics an EDM. We found that switching to a different type of HV cable has led to the elimina- tion of discharges to the vessel during the HV ramp, which should reduce the possibility of ferromagnetic materials leading to false signals. During the past year we also made several improvements to our UV laser system. We use a semiconductor master oscillator-power amplifier laser operating at 1015 nm, which is then frequency quadrupled by two nonlinear crystal doubling stages, to access the 253.7 nm 61 S0 →63 P1 transition of Hg. Power stability of the system had become an increasing problem requiring frequent realignment of the doubling cavity optics. The crystal for the first doubling stage (KNbO3 ), which had been “bought off the shelf” for use at 1064 nm, was replaced with a crystal custom cut for 1015 nm. We had the second doubling stage crystal (BBO) repolished by the manufacturer to repair hygroscopic surface damage. We installed a HEPA filtration system over the entire laser system which has reduced the frequency of the need for optics cleaning from daily to monthly. Since these changes were made, the laser has required much less maintenance, and the power output from the semiconductor laser required to produce the level of UV light needed for the experiment has been reduced by 40%. ∗ Department of Physics, University of Washington, Seattle, WA 98195. † Department of Physics, Princeton University, Princeton, NJ 08544. 1 M. V. Romalis, W. C. Griffith, J. P. Jacobs and E. N. Fortson, Phys. Rev. Lett. 86, 2505 (2001). 2 CENPA Annual Report, University of Washington (2002) p. 6.


  • Page 20

    10 Torsion Balance Experiments 1.7 Sub-mm test of Newton’s Inverse-Square Law E. G. Adelberger, T. S. Cook, J. H. Gundlach, B. R. Heckel, C. D. Hoyle,∗ D. J. Kapner and H. E. Swanson We have completed the analysis of our third successful measurement of Newton’s Inverse Square Law (ISL). We use a torsion pendulum with 22-fold azimuthal symmetry which we operate as a driven harmonic oscillator. The drive for this oscillator is a similar mass distrib- ution which rotates below the pendulum. We measure the pendulum twist angle in amplitude and phase at the 22nd, 44th and 66th harmonics of the frequency of rotation as a function of vertical and horizontal separation between the pendulum and drive mass. We compare these measurements to a detailed calculation of the expected Newtonian signal.1 We see no deviation from the ISL and can set a limit of 90µm for the scale at which there is no new force or interaction with a strength comparable to gravity’s. We have improved over our own previous result2 by up to a factor of 100. Figure 1.7-1. 95% confidence limits on the parameters α and λ, which characterize an −r addition to the Newtonian potential, VN = − Gm r , as V = VN (1 + αe λ ). The curve labeled “Eöt-Wash” is our new result, which runs parallel to our previous result. ∗ Presently at University of Trento, Trento, Italy. 1 CENPA Annual Report, University of Washington (2002) p. 11. 2 C. D. Hoyle et al., Phys. Rev. Lett. 86, 1418 (2001).


  • Page 21

    UW CENPA Annual Report 2002-2003 May 2003 11 1.8 Spin pendulum update T. S. Cook, E. G. Adelberger, B. R. Heckel, H. E. Swanson and M. White The next generation spin pendulum has been designed and is currently in the final stages of assembly. The spin pendulum is a torsion balance that utilizes the magnetic properties of Alnico and SmCo to produce a spin-polarized test body.1 By accurately measuring the affinity of the test body for a preferred direction in space (or lack thereof), the pendulum puts limits on possible simultaneous Lorentz and CPT symmetry violation as described by Colladay and Kostelecky.2 Over the past year, a number of design improvements have been implemented to increase the overall sensitivity to such a symmetry violation. Smaller - Lighter: The diameter of the pendulum has been reduced while maintaining the same active mass as previous versions. In the past, pendula were built with an aluminum harness to place rectangular magnets into an octagonal ring. For the new pendulum, the magnets are precision machined into trapezoidal wedges to form the octagonal ring. The maximum diameter of a ring has been reduced from 1.67” to just 1.04”. This reduced geometry requires less inert mass to support the ring structures, lowers the torsion moment of the pendulum and, due to the decreased mass, allows the inclusion of a magnetic shield on the pendulum while still using a 30-µm diameter suspension fiber. New Materials: We are now using Sm2 Co17 instead of SmCo5 as the offset material to Alnico. The magnetization in Alnico comes entirely from electron spin alignment and is therefore the material responsible for providing the net polarization. The magnetization in SmCo comes largely from orbital contributions and serves to close the magnetic circuit without canceling the spin vector provided by the Alnico. Sm2 Co17 is able to sustain a larger magnetic field than SmCo5 - 11,600 and 8,600 gauss respectively3 - so the net polarization in the Alnico should be proportionately increased. Improved Optics: The angular position of the pendulum is monitored with an auto- collimator that reflects a laser off a mirror attached to the pendulum. The Spin Pendulum experiments run in the old Eöt-Wash apparatus which is designed for a single reflection off the pendulum mirror. However, one can double the angular resolution by double reflection: reflecting the light to an auxiliary mirror and then back to the pendulum mirror before returning it to the auto-collimator. We are in the process of re-fitting the old Eöt-Wash apparatus to accommodate a double reflection system. As may be expected, the attenuation of all magnetic “leakage” fields from the pendulum is essential to performing a precise measurement. A magnetic coupling between the pendulum and the Earth’s field or any ambient laboratory fields could overwhelm a potential signal or even produce a false signal. To accurately ascertain the strength of our leakage fields, we have constructed a stepper motor driven turntable and data acquisition system to read data from a 3-axis hall probe as the pendulum is rotated. With this system we have been able to measure the fields at any given radius from the pendulum to less than 50 µG. 1 M. G. Harris, Ph.D. thesis, University of Washington (1998). 2 D. Colladay and V. A. Kostelecky, Phys. Rev. D 55, 6760 (1997); ibid. 58, 116002 (1998). 3 www.magnetsales.com.


  • Page 22

    12 1.9 Eötwash data acquisition system development H. E. Swanson A critical element in all our systems has been a 16-bit ISA-bus multifunction data-acquisition board. Recently the manufacturer informed us that they would no longer be able to repair them. Readily available motherboards no longer even have sockets for these ISA boards. We had developed hardware device drivers to run under the Windows 95 operating system but they use an older VxD style architecture that is not supported in Windows 2000 or XP. Modifying this code for newer data acquisition boards would only help in the short term as we would not be able to upgrade the operating system beyond Windows 98. It will be increasingly difficult to maintain these systems. The core software for our existing systems is a windows application developed in Visual C ++ with a graphical user interface built using Lab Windows CVI. Each experiment has its own electronics for multiplexing up to 16 temperature sensors into one ADC channel and reading the angle of the source field. They each require a unique hardware driver to read in a data event. We are currently developing a prototype acquisition system with a modern PCI-bus multifunction board. It retains most of the existing code but requires changes in the way the hardware interface is accessed. We have two versions of this board: One with 64 analog input channels and the other with 16 channels. This is a more generic system because the additional input channels make the external multiplexer unnecessary, and the angle encoder is now monitored with an on-board pulse counter. The supplied hardware driver is also compatible with Windows XP. With the systems currently taking data, the maximum sustained data rate is 20Hz. The new system can achieve kilohertz rates when only ADC channels are read. If we require reading the angle for each event, the rate is considerably lower. Improving the performance in this later mode will probably require writing our own device driver. Two systems are now running with the new multifunction boards and prototype code. One is used in measuring residual external fields of the new spin pendulum (see Section 1.8). The other will be the data-acquisition computer for the LISA project (see Section 1.12). Synchronous detectors and filters are implemented in software in this system so lock- in amplifiers are not needed to demodulate the autocollimator signals.


  • Page 23

    UW CENPA Annual Report 2002-2003 May 2003 13 1.10 A new equivalence principle test E. G. Adelberger, K. Choi, J. H. Gundlach, B. R. Heckel, D. J. Kapner and H. E. Swanson We have constructed1 a new rotating torsion balance to search for violations of the equivalence principle due to fundamentally new forces with Yukawa ranges greater than 1 m. The torsion balance consists of a composition-dipole containing titanium, beryllium, or aluminum test bodies. The pendulum is suspended inside a vacuum chamber and hanging from a constantly rotating turntable. During the past year we have begun taking long data runs with the instrument. The statistical uncertainty is as expected at about 2 nrad in one day of data taking. We found a slowly varying systematic effect which is likely related to small rotation rate variations of the main turntable that rotates the instrument. We are now changing our data taking protocol to modulate the composition dipole orientation w.r.t. the turntable at a frequency (1 day−1 ), which is high compared to the variation of the systematic effect. In addition we have equipped the turntable with extra temperature sensors and an additional set of tilt sensors. We have reduced the clearly identifiable systematic sensitivities of our apparatus: Gravity gradient effect: we measured the strength of the Q21 , Q31 gravity-gradient fields around the pendulum with special test bodies that have exaggerated gravity-gradient mo- ments. The field were compensated with Q21 , Q31 gravity-gradient compensators masses so that their strength was about 0.13% and 6% the uncompensated strength, respectively. We noticed a small change in the Q21 field which is attributable to water (rain) in the nearby soil. Tilt coupling: we eliminated the tilt of the turntable rotational axis by using an active lev- eling system.2 With this system turned on, the tilt angle of the turntable is nearly unresolved in the sensor that is used by the feedback system. Since this sensor cannot be at the exact location of the pendulum a small effective tilt remains. From the local mass distribution we estimate this tilt to be 36 nrad, which causes a 0.6 nrad pendulum rotation. Thermal coupling: the thermal shield around the vacuum chamber has constant-temperature water circulating through it. We modulated the temperature of this water at our nor- mal signal frequency to measure the thermal feedthrough. The thermal feedthrough is 0.099 ± 0.008 nrad/mK, corresponding to 0.028 nrad the signal. Magnetic coupling: We reduced the susceptibility of the apparatus to magnetic fields by installing a second µ-metal shield inside the vacuum chamber. The response of the pendulum to an externally applied field of 870 mG is 2.08 ± 2.15 nrad. The systematic uncertainty due to magnetic coupling is less than 0.05 nrad. 1 CENPA Annual Report, University of Washington (2002) p. 10. 2 CENPA Annual Report, University of Washington (2001) p. 4.


  • Page 24

    14 1.11 Development of an ‘anapole pendulum’ E. G. Adelberger, T. S. Cook, B. R. Heckel, F. V. Marcoline and H. E. Swanson We are currently developing an ‘anapole pendulum’ which will be used in a rotating torsion balance to make a photon mass measurement. If the photon field, Aµ , has a Dirac mass, mγ , then there is an additional term in the electromagnetic energy density equal to 21 m2γ Aµ Aµ . Just as an external magnetic field will exert a torque on a magnetic dipole, an external photon field will exert a torque on a dipole magnetic vector potential (an anapole) towards the minimum photon field energy density configuration. A toroidal current distribution produces a magnetic field confined within the toroid, and an anapole field external to the toroid. Previous experiments1 , 2 have used wire-wrapped iron cores which require constant current supplied to the pendulum. Our active mass will consist of a single toroidal soft ferromagnet with a high permeability and magnetic remanence, giving a large anapole at low mass without needing to supply current to the pendulum. We have constructed a simple setup to determine the hysteresis curve of our magnet. Early tests indicate that the magnetic properties are sufficient to allow a significant improvement on the current upper limit of the photon mass. Initial design of the pendulum is complete, and we have begun manufacturing components. The pendulum is designed to maximize our sensitivity to a photon mass, while minimizing torques due to local gravity gradients and stray magnetic fields. The 70-gram pendulum will have magnetic shielding surrounding the magnet, and will be suspended from a 20-µm tungsten fiber. Annealing and final machining of the magnet is under way. Further tests on the completed magnet are needed to finalize the design. In particular, measurement of leakage fields (see Section 1.8) is needed to determine how much magnetic shielding we need on the pendulum body. 1 R. Lakes, Phys. Rev. Lett. 80, 1826 (1998). 2 J. Luo, L. - C. Tu, Z. - K. Hu and E. -J. Luan, Phys. Rev. Lett. 90, 081801 (2003).


  • Page 25

    UW CENPA Annual Report 2002-2003 May 2003 15 1.12 Small force measurements for LISA E. G. Adelberger, J. H. Gundlach, B. R. Heckel, C. Kurz, P. Searing and H. E. Swanson To date gravitational waves have only been observed indirectly. NASA and ESA are planning jointly to build a space-based gravitational wave antenna, LISA,1 which will have enough sensitivity to observe gravitational waves from various sources such as massive black-hole mergers or even from the early phases of the universe. The instrument will consist of three spacecraft placed at the corners of an equilateral triangle of 5,000,000-km arm length. This constellation will trail Earth on its orbit about the Sun by 20 days. A gravitational wave causes small variations in the arm lengths. Each arm length is measured to a precision of 20 pm using laser beams. The end of each laser beam path is defined by a cubical mass, freely floating inside the spacecraft called a gravitational reference sensor or proof mass. LISA’s sensitivity is optimized in the frequency range from 10−4 to 1 Hz with a minimum detectable fractional length change of 10−23 . To achieve the proposed sensitivity of LISA any time- varying acceleration of the proof masses along the laser beam direction must be avoided. In particular spurious coupling of the proof masses to the spacecraft, which is servoed to follow the proof masses, must be avoided.2 The nearest distance from the proof mass to its housing will be 2 to 4 mm. This close spacing raises concerns about forces acting between nearby surfaces. A known but little studied interaction is due to patch-effect potentials. We are currently building a specialized torsion-balance system to measure patch-effect forces between closely spaced surfaces. Our first setup will consist of a stepped torsion-pendulum body (Fig. 1.12-1) suspended parallel to a larger surface. The separation to the pendulum surface will be varied by modulating the position of the larger surface. To approach the sensitivity requirements for LISA the closest, we will use separations much closer than those in LISA. Our setup will also allow us to vary the electrostatic potential between the pendulum and the proof mass to simulate the effect of charging and of the electrostatic control of the proof masses in LISA. In addition to patch-effect forces our setup will be sensitive to other forces which may arise, for example due to residual gas. We have assembled a stainless steel vacuum chamber and rebuilt an autocollimator system. A prototype translation device for the large plate was built. We chose a pneumatic actuator to avoid electrical cur- rents near the pendulum plate. The dynamic range of this actuator is about 2 mm. We poured a 30 cm thick concrete base plate in a space that was freed by dismantling the cyclotron. A thermal enclosure was constructed. We have built a data acquisition sys- tem, which we are in the process of connecting to the instrument. In the next few months we will install a test pendulum to acquire first test data. The research Figure 1.12-1. Schematic top view of step is funded with a contract from NASA/GSFC. pendulum setup. 1 http://lisa.jpl.nasa.gov/. 2 B. L. Schumaker, Class. Quantum Grav. 20, S239 (2003).


  • Page 26

    16 2 Neutrino Research SNO 2.1 The Sudbury Neutrino Observatory J. F. Amsbaugh, T. H. Burritt, G. A. Cox, P. J. Doe, C. A. Duba, J. A. Formaggio, G. C. Harper, K. M. Heeger,∗ M. A. Howe, K. K. S. Miknaitis, S. McGee, A. W. Myers, N. S. Oblath, J. L. Orrell, R. G. H. Robertson, M. W. E. Smith† and L. C. Stonehill The Sudbury Neutrino Observatory (SNO) is a heavy water Čerenkov detector that was designed to study the flux, spectrum, and oscillation characteristics of 8 B neutrinos from the Sun. Previous solar neutrino experiments, able to detect primarily electron type neutrinos, have observed a deficit in the flux of solar neutrinos relative to the predictions of solar models. This discrepancy between the observed flux of electron type neutrinos and the expected flux is known as the Solar Neutrino Problem. The use of heavy water as a target medium in SNO allows sensitivity not only to electron neutrinos, but also to mu and tau neutrinos, through the neutral current (NC) reaction of neutrinos on deuterium. The three primary reactions through which we detect solar neutrinos are shown below, νx + 2H −→ νx + p + n Neutral Current (NC) νe + 2H −→ p + p + e− Charged Current (CC) νx + e− −→ νx + e− Elastic Scattering (ES) (1) where x denotes any of the three active neutrino flavors. A comparison of SNO’s measured CC and NC rates has shown that solar electron neutrinos are oscillating into mu and tau neutrinos prior to reaching our detector. The flux of mu and tau neutrinos explains the previously observed deficit in solar electron type neutrinos. SNO’s results over the past two years have thus provided strong evidence for the oscillation of solar neutrinos and solved the long-standing puzzle of the missing solar neutrinos.1 The SNO experiment has been designed to maximize our confidence in the NC to CC comparison, and we have therefore incorporated several unique mechanisms for confirming and improving our measurement of the NC reaction rate. Since June 2001 we have been running SNO with ultra-pure salt (NaCl) added to the heavy water. Neutron capture on the salt provides an even more distinct signal than neutron capture in pure heavy water. We will soon deploy an array of discrete Neutral Current Detectors (NCDs), to provide an independent means of detecting the NC neutrons. As we continue to refine our measurement of the NC and CC rates, we will also be able to provide more stringent limits on the fundamental physics parameters that govern solar neutrino oscillations. The University of Washington has played an active role in the analysis of the data from the first and second phases of the experiment, as well as in studying the day-night effect and solar anti-neutrinos. We are the primary institution responsible for the hardware, installation, and analysis for the NCD phase of the experiment, which will begin Summer 2003. ∗ Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720. † DASI, University of Chicago, South Pole. 1 Q. R. Ahmad et al., Phys. Rev. Lett. 89, 011301 (2002).


  • Page 27

    UW CENPA Annual Report 2002-2003 May 2003 17 2.2 Status and updates to the SNO data acquisition system P. Harvey,∗ M. A. Howe, B. L. Wall and J. F. Wilkerson Since the SNO DAQ system has been described extensively in past annual reports,1 only a brief review of the most significant updates is provided here. Last year a new 733-MHz G4 Macintosh was installed underground as the primary SNO DAQ computer. In addition, a new 622 PCI card was installed that uses an optical-fiber cable to communicate from the Mac to the VME Bit 3 controller. After the installation we began to see occasional VME exceptions occurring when shipping records to the event builder through the Sun dual port memory (dpm). These errors commonly stopped the execution of time slope calibration runs with an ‘unable to create/access the Sun dpm’ message printed to the SHaRC status log window. The errors also occurred occasionally at the start of a run during the shipment of the run record. This became known as the ‘Sun dpm error problem’ and was a serious issue because it was adversely affecting detector live time. Because of the sporadic nature of the error, a special software test module was added to SHaRC to emulate an electronic calibration task running at a very high rate. All testing was done on the on-site above-ground test stand. Since it was unclear whether the problem was limited solely to the 622 PCI card, a 617 PCI card was also installed into the test- stand Mac. The test results were very sporadic. Sometimes a test would run for hundreds of thousands of cycles without a single Sun dpm error. At other times, 1% of the records shipped caused an error. The lowest rate was an over-night run that shipped 4.5 million records with only 16 errors. There was some speculation that the effect was temperature related, but experimentation with a heat gun was inconclusive. It was noted that although errors occurred using either PCI card, the error rate was in general 25 times higher when using the 622 PCI card. Once it became possible to reproduce the error on a regular basis, it was possible to pinpoint the cause as either a VME adapter card bus grant starvation error or a PCI bus timeout during a read of the Sun dpm circular buffer control block. The error was never seen while pushing records into the circular buffer. It appears that the Sun dpm errors are entirely hardware related, but the actual root cause is not known at this time. The code that is currently installed underground has a work-around installed that re-reads the circular buffer control block three times before letting an exception propagate into a full-blown Sun dpm error. This solution was successful and the Sun dpm error has not reccurred. There were also a number of minor software updates to streamline detector operations and to transmit detector parameters to remote sites for tracking detector operation over time. ∗ Queens’s University, Kingston, Ontario, Canada K7L 3N6. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1997) pp. 20-23; (1998) pp. 18-20; (1999) pp. 16-18; (2000) p. 19; CENPA Annual Report (2001) p. 25; (2002) p. 26.


  • Page 28

    18 2.3 Energy and optical calibration of the Sudbury Neutrino Observatory J. A. Formaggio and the SNO Collaboration Part of the process of extracting the solar-neutrino flux and separating charged-current from neutral-current events involves knowing to great precision the energy and optical response of the SNO detector. The optical calibration of the SNO detector makes use of a dye laser with a fiber-optic cable connected to a diffuser ball. The laser provides pulsed radiation at 337 nm with a 600-ps pulse width and peak power of 150 kW. The laser pulses up to 45 Hz and can be used directly or as a pump for several dye lasers that provide wavelengths in the range of 360-700 nm. The optical calibration process results in the determination of attenuation and scattering coefficients for the heavy water, light water, and acrylic. In combination with a detailed Monte Carlo that simulates the Čerenkov photon production from electrons, SNO is able to extract the energy response of the detector. The energy response is verified via the employment of a series of calibration sources. The main source that tests the energy response is 16 N. The 16 N is produced via (n,p) on the oxygen in CO2 gas and decays by a beta delayed 6.13-MeV (66%) or 7.12-MeV (4.8%) gamma ray with 7.13-s half-life. The 16 N provides a check of the energy response of the detector as a function of both time and position. In addition to 16 N, SNO also uses a 8 Li β source and a tagged 252 Cf neutron source to further test the energy response of the detector. The 8 Li β source has an end-point energy of 13 MeV, allowing one to test the energy scale above what is probed by 16 N. The tagged neutron source, on the other hand, tests SNO’s sensitivity to neutrons directly. Fig. 2.3-1 shows how well the Monte Carlo tracks the energy response of the detector for 8 Li calibration sources. The combination of laser, radioactive, and tagged sources allow one to limit the energy uncertainty of the SNO detector to approximately ±1.25% for the duration of the salt phase. This systematic error is similar to what has been measured in the D2 O phase of the SNO experiment. Events / MeV 3000 Data (8Li) Monte Carlo 2500 2000 1500 1000 500 0 4 6 8 10 12 14 Energy (MeV) Figure 2.3-1. Energy profile for data (black points) and Monte Carlo (red) for 8 Li tagged β decays.


  • Page 29

    UW CENPA Annual Report 2002-2003 May 2003 19 2.4 Selection of the neutrino analysis data set for the Salt Phase of SNO S. McGee and J. L. Orrell The Sudbury Neutrino Observatory (SNO) collaboration has recently decided upon the period of data taking to be used in the first results from the salt phase of the experiment. During the previous year Orrell was chairperson of the run selection committee composed of graduate students and post-doctoral fellows from SNO member institutions, individuals who have spent a significant amount of time on location at the experiment learning the details and peculiarities of the day-to-day data taking. This on-site experience makes the members of this committee valuable resources as adjudicators of exactly when the SNO detector is in an optimum solar neutrino data taking configuration. Data selection during the salt phase of the SNO experiment consists of three main tasks. First, a simple near-line computer analysis package analyzes all runs and identifies which runs are of the neutrino type, that have the correct number of photomultiplier tubes on-line, and are over 30 minutes long. Second, a data integrity computer analysis package analyzes those runs identified by the first step, and tests that each run has the correct data structureand the expected average data rate, and that no electronics failures occur during the run. Third, for each and every run, two members of the run selection committee independently read the operations reports to verify that there were no unusual circumstances during the run and that no neutrino runs were missed by the computer analysis packages due to operator error in setting the run type. The results of over a year of bi-weekly meetings of the run selection committee produced the a selected data set with characteristics given in Table 2.4-1. Salt phase data taking has continued past 10/10/2002 and the run selection committee continues to add to the salt phase data set. Expected Salt Phase Data Set Data taking period 7/26/2001 → 10/10/2002 Total calendar days 441 Number of selected runs 697 Total days of selected data 258.8 Total livetime fraction 58.7 % Table 2.4-1. The characteristics of the data set expected to be used for the first reported results from the salt phase of the SNO experiment.


  • Page 30

    20 2.5 Distinguishing muon spallation types in SNO J. A. Formaggio, K. K. S. Miknaitis, J. L. Orrell and J. F. Wilkerson As reported in last year’s report,1 the Sudbury Neutrino Observatory (SNO) is sensitive to the spallation products from high energy cosmic ray muons passing through the detector. It was demonstrated that the dominant spallation products were neutrons. These initial studies also showed there was a broad range of muon-follower multiplicities. Muon-follower multiplicity is defined as the number of candidate physics events that follow within 0.5 seconds of a muon. The observed high-multiplicity muons suggested sub-dividing the muons into two classes: (1) real and virtual photo-induced spallation and (2) “other processes” including deep inelastic scattering, muon decay, muon capture, or pion production.2 Inspection of the 10 highest multiplicity sets of events revealed high energy Čerenkov rings following within 5 microseconds after the initiating muon. During this short period of time following the passage of a high-energy muon, the SNO detector continuously retriggers due to both residual light in the detector and the electronics relaxation response to the large charge deposition in the photomultiplier tubes from the muon’s initial Čerenkov signal. An algorithm was developed to search the events following a muon for Gaussian-distributed photomultiplier-tube trigger times above a constant background of photomultiplier-tube “noise” triggers. The goal of this algorithm was to distinguish high-multiplicity muons with- out relying upon the high-multiplicity signature. The results of this work are shown in Fig. 2.5-1. The “all muons” histogram is scaled by a factor of 100 for clarity and shows both high- and low-multiplicity muons in the data sample. The “clean muons” histogram demon- strates that the algorithm successfully distinguishes and removes high-multiplicity muons. It is expected that this separation will allow us to study and characterize a distributed sample of photo-induced spallation neutrons. Muon Follower Multiplicity 6 Pure D2 O Phase 10 Followers from all muons (x100) 5 10 4 10 Followers from "clean" muons 3 10 2 10 10 1 -1 2 10 1 10 10 Number of Recorded Muon Followers Figure 2.5-1. The number of candidate events recorded within 0.5 seconds of a muon. 1 CENPA Annual Report, University of Washington (2002) p. 18. 2 Y.-F. Wang et al., Phys. Rev. D 64, 013012 (2001).


  • Page 31

    UW CENPA Annual Report 2002-2003 May 2003 21 2.6 Electron antineutrino detection at the Sudbury Neutrino Observatory J. L. Orrell and J. F. Wilkerson The charged-current weak interaction of electron antineutrinos on deuterons, ν̄e + d = e+ + n + n Q = −4.03 (CC) produces a positron, e+ , and two neutrons, n. Each of these three products can potentially produce a signal that will trigger the Sudbury Neutrino Observatory (SNO) detector. This coincidence of detection events provides a unique signature of electron-antineutrino interac- tions in the SNO detector. Observing or limiting the flux of electron antineutrinos addresses several hypothetical antineutrino production mechanisms. Specifically, if neutrinos are Ma- jorana particles, then a fraction of the electron neutrinos produced in the Sun’s core may experience resonant spin-flavor transitions in the Sun’s magnetic field, producing electron antineutrinos. There is also a speculative hypothesis of a Geo-nuclear reactor located at the center of the Earth. We are pursuing a line of research that demonstrates a robust, low en- ergy, and large fiducial volume analysis of the SNO data set for the measurement or limiting of the flux of electron antineutrinos. The demonstration of the low energy and large fiducial volume analysis relies on spallation neutrons produced by muons in the SNO detector (see Section 2.5). The left hand of Fig. 2.6- 1 clearly shows the spallation neutron signal that follows within a half second of the muon. For comparison, the right hand plot shows the half second preceding the muon. The plots are energy-measure versus the reconstruction radius relative to the entire detector volume. RAW RAW Scaled Neff Scaled Neff 140 Followers 140 Predecessors 120 120 100 100 80 80 60 60 40 40 20 20 0 0.5 1 1.5 2 2.5 0 0.5 1 1.5 2 2.5 R 3fit / R3AV R 3fit / R3AV Figure 2.6-1. Neutron-like events following and preceding within half a second of a muon. This neutron signal mimics the expected electron antineutrino coincidence. The x-axis is volume with 1 equal to the entire D2 O volume. The y-axis is an energy measure. The small rectangle represents the neutron measurement window reported in SNO’s solar neutrino analysis.1,2 The larger region is the proposed analysis region for an electron anti- neutrino analysis using the SNO data set. We are currently determining the backgrounds expected inside this analysis window. 1 Q. R. Ahmad et al., Phys. Rev. Lett. 89, 011301 (2002). 2 Q. R. Ahmad et al., Phys. Rev. Lett. 89, 011302 (2002).


  • Page 32

    22 2.7 Reactor antineutrinos at the Sudbury Neutrino Observatory J. L. Orrell The Sudbury Neutrino Observatory (SNO) is capable of detecting electron antineutrinos, ν̄e , through weak interactions with deuterons contained in the heavy water of the SNO experi- ment: ν̄e + d = e+ + n + n Q = −4.03 (CC) ν̄e + d = ν̄e + p + n Q = −2.23 (N C) Commercial, electric power generating, nuclear reactors generate approximately 2 × 1017 ν̄e /s per MW of thermal power, with energies from 1 - 10 MeV. The reactor ν̄e flux from distant North American nuclear reactors induces several CC and N C interactions per year. These interactions are a small background to the solar neutrino interactions SNO is designed to measure (see Section 2.1). Furthermore, measurements that attempt to place limits on the total ν̄e flux at SNO are restricted by the reactor induced ν̄e signal (see Section 2.6). A new and more comprehensive calculation of the CC and N C interaction rate at SNO has taken into account (on a monthly basis1 ) the true reactor power levels convolved with SNO’s actual data taking periods. The reactor ν̄e spectrum is determined by a weighted sum of the individual ν̄e spectra from the fission products of 235 U, 238 U, 239 Pu, and 241 Pu. The total flux is determined from the monthly power output of all nuclear reactors within 700 km of the SNO experiment. The total CC and N C interaction rates for each month are calculated including the suppression effect due to neutrino oscillations. The neutrino oscillation portion of the calculation used world data,2 best-fit values for the neutrino mixing parameters sin2 2θ = 0.83 and ∆m2 = 5.5 × 10−5 eV2 . The total number of CC and N C interactions was further weighted by the fraction of time that SNO was recording experimental data during each month. Table 2.7-1 shows the results of this calculation. Phase of Total Detector live experiment # CC # N C # CC # N C Pure D2 O 1.65 7.24 0.89 3.90 Salt 1.41 6.22 0.73 3.24 Table 2.7-1. Calculated number of CC and N C interactions in the SNO detector during each phase of the SNO experiment. Neutrino oscillations are included with sin2 2θ = 0.83 and ∆m2 = 5.5 × 10−5 eV2 . 1 Thank you to the US Nuclear Regulatory Commission and Chalk River Laboratory for providing the historical monthly power output for all North American nuclear reactors. 2 G. L. Fogli et al., Phys. Rev. D 66, 053010 (2002).


  • Page 33

    UW CENPA Annual Report 2002-2003 May 2003 23 2.8 The day-night asymmetry measurement in the salt phase of SNO K. K. S. Miknaitis and the SNO Collaboration Solar-neutrino oscillations are governed by two parameters, ∆m2 , which quantifies the mass difference between the neutrino states that participate in solar-neutrino oscillations, and tan2 θ, which quantifies the strength of the mixing between states. A combined fit of SNO’s solar neutrino results along with other neutrino experiments, notably the first results from KamLAND, has narrowed down the allowed values of these parameters to a single region of the formerly large parameter space. This region, known as the Large Mixing Angle (LMA) solution, includes values in the vicinity of ∆m2 ∼ 10−4 and tan2 θ ∼ 0.4. Solar neutrino oscillations governed by LMA parameters involve matter enhanced oscil- lations in the sun (the MSW effect). Beyond the flavor-changing effects of the oscillation phenomenon, the MSW effect has two additional possible signatures. First, it can result in a distortion of the energy spectrum measured in an experiment like SNO relative to the original solar-neutrino spectrum. Second, the same matter effects that enhance oscillation in the sun can potentially alter the flavor content of the solar neutrino “beam” as it passes through the earth. This latter effect would produce a difference between the rate of electron-type neutrinos detected by SNO during the day and during the night, when solar neutrinos have to pass through the earth to reach our detector. For our first measurement of a potential day-night asymmetry in the electron-type neu- trinos detected by SNO, we constructed an asymmetry parameter ΦN − ΦD A=2 (1) ΦN + ΦD where ΦN and ΦD are the night and day neutrino fluxes, respectively. Neutrino oscillation parameters in the LMA region predict asymmetry values between 0% and around 8%. The day-night asymmetry measurement, if non-zero, could provide direct evidence for the MSW effect. However, even if the day-night asymmetry is too small to be able to make a significant non-zero measurement, this parameter can still limit the LMA parameters to a more restrictive region of the parameter space. In the LMA region, the day-night effect is primarily sensitive to differences in ∆m2 . In the first phase of SNO we measured A = 7.0 ± 4.9(stat.) ± 1.3(syst.)%. This measure- ment was statistically limited, so future measurements will combine the data from multiple phases to improve the statistical sensitivity. We have undertaken a thorough study of the covariances between systematics between the pure D2 O phase and the salt phase in order to be able to perform this joint measurement. We are also laying the groundwork for possible improvements to the measurement, including studying the day-night effect as a function of the zenith angle, or the amount of the earth’s matter that the neutrinos traverse en route to SNO. Precise measurements of the detector “live” time for day and night have been com- pleted for the salt phase, and studies of the variation of systematic effects day and night are in progress.


  • Page 34

    24 2.9 SNO signal extraction in the salt phase K. K. S. Miknaitis and the SNO Collaboration The exceptional opportunity afforded to SNO to make solar-neutrino measurements using heavy water as a neutrino target is so unique that the experiment was designed to include several complementary detection techniques, to ensure robust results that will be worthy of an honored place in the history of nuclear and particle physics. During the first phase of the experiment, our measurements of the charged current (CC) and neutral current (NC) reactions between neutrinos and deuterium provided an answer to the long-standing solar neutrino problem. We now move on to the next two phases of the experiment, designed especially to fine tune, check, and improve our measurement of the NC reaction, which will be SNO’s most significant lasting contribution to physics. Since May 2001 we have been in the second phase of the experiment, in which ultra-pure salt (NaCl) has been added to the D2 O in the detector. This phase will be followed by a third phase involving the deployment of discrete neutral current detectors into the heavy water volume. Both of these additions to the detector are meant to increase sensitivity to the specific signature of the NC reaction, neutrons: NC Reaction: ν + d −→ ν + p + n (1) In pure heavy water, the neutrons capture on deuterium nuclei, producing a 6.25-MeV gamma. The neutron-capture event distribution in energy was characterized by a broad peak centered on the gamma energy and a radial distribution that fell off near the boundary of the detector, as the neutrons would scatter outside the heavy water volume and be captured on other materials. With the addition of salt to the detector, neutrons can also capture on 35 Cl. This reaction has a higher cross section, and the gamma cascade emitted has a higher energy (8.6 MeV). The higher cross section means that neutrons will capture more quickly after they are produced, flattening the distinct radial profile that was characteristic of neutron events in the first phase. In addition, while electron events yield a clear Čerenkov pattern of light, the gamma cascade creates a more diffuse pattern. We can parameterize the different degree of isotropy of the light produced in physics events. Although the different types of event (CC, ES, and NC) all have different distributions in our measurable parameters, we cannot tell the difference between them on an event-by-event basis. Instead, we use detailed models of the physics and of our detector to create probability distribution functions (PDFs) describing the event types. We then do a statistical separation of the signals by performing a maximum likelihood fit of our entire data set to these PDFs, to extract the most likely amplitudes of the different signal contributions. To incorporate the differences between salt and the pure heavy-water phase, we have developed signal extraction software in C++ that is capable of performing such statistical fits to any number of input PDFs with adjustable acceptance regions. This allows us not only to include isotropy distributions in the fits, but to study the features of the signal extraction depending on how many input PDFs are used and what acceptance windows in energy and volume are defined.


  • Page 35

    UW CENPA Annual Report 2002-2003 May 2003 25 SNO Neutral Current Detectors (NCDs) 2.10 NCD data analysis G. A. Cox, P. J. Doe, A. L. Hallin, S. McGee, R. G. H. Robertson, L. C. Stonehill and J. F. Wilkerson The NCD’s data acquisition system, Section 2.13, has limited analysis and detector moni- toring capabilities. However, the NCDDAQ does produce custom formatted data files for off—line analysis. In order to analyze our data we have employed the ROOT analysis frame- work developed at CERN.1 A set of NCD event structures, which contain all information about an event in our system, have been created and our data files have been converted into ROOT readable files based upon these event structures. The first major analysis completed with the ROOT-based tools was determining which NCD’s should be deployed in the reduced array (see Section 2.11). The analysis was done on measurements made by our Shaper/ADC cards,2 which produce an energy spectrum of the events in our detectors. The critical selection criteria were the existence of a neutron peak with good resolution, and low count rate above the neutron peak, most of which are attributable to alphas from intrinsic radioactivity in the counter. We also selected counters that had similar gas gains. The goal was to select the working NCD’s with the lowest levels of intrinsic radioactivity. NCDDAQ also records the pulse shape of events with two digitizing oscilloscopes. Due to a change in our data structure, our previous set of pulse shape analysis tools,3 known as Analyst, have become obsolete. The source code from Analyst is being ported into our ROOT—based analysis tools. Once constructed, work can resume on analysis projects such as the “Water Wall” study to measure the thermal and fast neutron fluxes from the rock walls at SNO.4 Pulse shape analysis begins with the extraction of the real pulse shape from our logarithmically amplified signal. The logarithmic amplifiers are characterized by Vout = a · log(1 − Vin /b) + c, where a, b, and c must be determined for each amplifier. Extraction of these parameters is done with a χ2 —squared minimization routine on the output signal generated from an known input. This has been done on a preliminary level, and automated calibration routines are under construction. The typical pulse shape analysis calculates the energy of the pulse and the amount of time it takes the pulse to reach 90% of its peak value. In the risetime vs. energy space, ∼50% of the captured neutrons fall in a region known as the “background free region” A pulse shape fitting routine is being developed to distinguish neutron events from alphas in the NCD’s. It is based on semi—analytic functions that describe the expected NCD signals from alphas and neutrons. The goal is to reject alphas without sacrificing ∼50% of the neutrons, as in the risetime vs. energy method. 1 http://root.cern.ch. 2 CENPA Annual Report 2001, University of Washington, p. 32. 3 CENPA Annual Report 2001, University of Washington, p. 31. 4 CENPA Annual Report 2002, University of Washington, p. 27.


  • Page 36

    26 2.11 NCD array optimization G. A. Cox, P. J. Doe, J. Formaggio, A. L. Hallin, J. Manor, J. L. Orrell, R. G. H. Robertson, L. C. Stonehill and J. F. Wilkerson The neutral-current detector (NCD) phase at SNO was designed to provide an independent measurement of the neutral-current neutrino flux from the sun. The 3 He filled detectors used for the NCD phase are ideal for this measurement, as they are uniquely sensitive to neutrons produced from the neutral-current reaction νx + d → n + p. The NCD Phase at SNO was initially proposed and constructed prior to knowledge of the exact mechanism responsible for the suppressed neutrino flux from the sun. This assumption considered that solar neutrino experiments prior to SNO were measuring the full non—sterile neutrino flux, rather than just ∼1/3 of the flux predicted by the standard solar model (SSM). In order to have enough statistics in a reasonable amount of time, 96 NCD strings, totaling 770 m in length, were built. However, since the time construction began, SSM—independent measurements have verified that the SSM is an accurate model and that neutrino oscillation is seen as the most likely mechanism for the suppressed flux. A reassessment of the NCD array and the number of detectors to be deployed was made. Installation of the NCDs will impact SNO mainly in two ways. First, it will provide an independent measurement of the neutral current (NC) interaction flux. Second, it will ob- scure light produced by the charged current (CC) and electron scattering (ES) interactions that take place in the D2 O. This effectively increases the energy threshold of the CC interac- tions that SNO’s PMTs can measure, increasing the uncertainty in the CC flux. In general, measurements that depend mostly upon the PMTs at SNO will favor fewer NCDs installed, and those that depend mostly upon the NC flux will favor more NCDs. However, the NCDs will absorb neutrons that would otherwise produce a background to the CC measurement. Thus, for measurements that depend upon the CC interaction, installation of some of the NCDs is desirable. Monte Carlo simulations were done to quantify the effect of installing various percentages of the full NCD array. Overall, there were 11 physics topics that were considered: CC flux, NC flux, CC Day/Night sensitivity, NC Day/Night sensitivity, seasonal flux variations, supernovae, anti- neutrinos, MSW parameters, “hep” neutrinos, sterile neutrinos, and CC spectral distortions. Most of the science addressed by the NCDs optimized in one of two ways: 1. Minimize the amount of light occultation induced by the presence of counters in the array. Light occultation decreases the resolution of charged and neutral current signals, while increasing the amount of background that falls in the fiducial volume. 2. Maximize the number of neutrons capture in the array. Measurements that are sensitive to neutron statistics (such as anti-neutrinos and supernovas) fall in this latter category. The final recommendation after the Monte Carlo study was to install 45 ± 5% of the array. After considering different configurations of installing the NCD array into SNO, it was decided that, using the centrally located strings, 50% of the array was to be installed.


  • Page 37

    UW CENPA Annual Report 2002-2003 May 2003 27 2.12 Determination of Z-position for hits in the NCD array J. V. Germani,∗ S. McGee, R. G. H. Robertson, L. C. Stonehill and J. F. Wilkerson Determination of the z-position of an event in the NCD array will assist in the characterization of backgrounds in the acrylic vessel (AV) and in the NCD’s as well. Other background exclusions could be made by correlating reconstructed PMT events with hits in the NCDs. Also, a determination of the neutral current signal efficiency per fiducial volume would be possible. Knowledge of the z-position of hits could also be used to monitor the health of the seg- mented NCD strings before and after deployment. The success of NCD connections made over the neck of the AV during deployment could be easily checked with a neutron source posi- tioned in close proximity to upper and lower sections of the string. Also, knowing the position of high voltage discharges allows for prompt diagnoses of potentially damaging breakdowns. A z-position determination is possible because the open-ended configuration of the NCD allows for pulse-reflection timing. This is enhanced by the addition of a 30 ns delay line at the bottom of each string. Upon arrival of the charge at the anode, the pulse is split in two. One proceeds directly to the DAQ while its mirror image heads to the bottom of the string before being reflected back. Taking the propagation speed of the pulse in the NCD to be c, the range of separation of the initial and reflected pulse in a 10-m long string is 60 ns for an event at the bottom of the string and 126 ns for an event near the top. The thermalization distance for neutrons in the D2 O is roughly 1.5 m. Therefore, the desired resolution for a z-position measurement would be on the order of 1 m. This is also equivalent to the resolution of the x and y position since the NCD strings will be placed in a 2D square lattice with 1 m spacing. Previous studies of z-position algorithms showed good resolution, σz = 0.75 m, for about a third of the data. The shape of the NCD pulse varies with the orientation of the track of the charged particles with respect to the wire. Previous method’s efficiencies dropped considerably as the track moved from the optimal parallel orientation.1 Currently under study is the use of cepstrum coefficients. The cepstrum (an anagram of spectrum) is the inverse Fourier transform of the log of the Fourier transform. Similar to the standard Fourier transform whose coefficients represent the strengths of wavelengths present in a waveform, this rather convoluted-sounding procedure results in coefficients that represent the strengths of phase differences between like wavelengths. This is precisely what is needed in discerning the separation of two similar waveforms. Using generic waveforms to simulate pulses in the NCDs, preliminary results indicate that the desired resolution is possible over a range of pulse shapes. ∗ Presently at Philips Ultrasound, Philips Medical Systems 22100 Bothell Everett Highway Bothell, WA 98021-8431. 1 J. V. Germani, internal document.


  • Page 38

    28 2.13 Data acquisition for SNO’s neutral current detectors G. A. Cox, M. A. Howe, B. L. Wall and J. F. Wilkerson Within the last year, the data acquisition system for the neutral current detectors (NCD- DAQ) at SNO has been updated to a version closer to our final goal. The description of the NCDDAQ from previous CENPA Annual Reports still sufficiently describes our system,1,2,3 although implementation of the embedded CPU (eCPU) and Global Trigger Identification (GTID) components has yet to be completed. The major updates were the data acquisition from hardware routines, event packaging, definition of our data output stream,4 and the im- plementation of hardware mapping that eliminates the recording of unnecessary data. Other updates include the development of a pulse generator controller for electronics calibration, as well as the usual minor bug fixes. Most major enhancements were completed by the summer of 2002 and data acquisition has been nearly continuous throughout the year (Fig. 2.13-1). 1 0.8 0.6 0.4 0.2 0 May Jun Jul Aug Sep Oct Nov Dec Jan Figure 2.13-1. NCDDAQ livetime between May 2002 and Feb 2003. The major causes for the periods of down—time in our system have been the installation of DAQ related hardware, the shutdown of the INCO mine, and two shipments of NCDs. We also experienced data storage issues that have been resolved. The data is transferred to CENPA via FTP and stored on local disks and onto DVD as backup. Our normal data acquisition rate (excluding calibration data where a neutron source was present) has ranged from 1.5 MB to 6 MB per hour. This was dependent upon the number of detectors connected to the system and the implementation of some software. The data rate will be reduced significantly in the near future by using a packed data structure. 1 CENPA Annual Report, University of Washington, (2002) p. 29. 2 CENPA Annual Report, University of Washington, (2001) p. 32. 3 CENPA Annual Report, University of Washington, (2001) p. 34. 4 G. A. Cox, “A Description of the NCD Data Packets Dispatched from NCDDAQ,” unpublished.


  • Page 39

    UW CENPA Annual Report 2002-2003 May 2003 29 2.14 NCD underground status T. H. Burritt, G. A. Cox, P. J. Doe, C. A. Duba, S. McGee, A. W. Myers, R. G. H. Robertson, L. C. Stonehill, T. D. Van Wechel and J. F. Wilkerson The NCD array and associated electronics are nearly complete. Gas fill was completed in April, 2002 and the last two shipments of NCD’s were delivered to SNO in May and November of 2002. Also delivered in the November shipment was enough hardware to allow data acquisition on all 96 channels. In February, 2003 the readout cables that will connect the deployed NCD’s to the preamplifiers were delivered to SNO, after having passed rigorous microdischarge testing.1 Also in that shipment was the pulser distribution system, which allows remote calibration of the electronics. The only piece of NCD data acquisition hardware that remains to be installed at SNO is the global trigger identification(GTID) board. A period of data acquisition from the entire NCD array between November and February exercised the complete data acquisition system. There were 95 NCD strings hooked into 12 shaper/ADC cards, 8 multiplexer boxes, and two digitizing scopes. There were problems with the ribbon cables that connect the multiplexers to their controller boards, so for about the first month of this period only shaper data was being acquired. Not only did this period of intensive data acquisition help with debugging the electronics hardware, it also produced data that was vital for selecting which NCD’s should be deployed in the reduced array. The NCD pre-deployment welding began in February and, despite setbacks, was com- pleted in early April, 2003. The 40 NCD strings that will be deployed are currently producing data that will be analyzed to detect any problems with the welded NCD’s before deployment. In addition, data is being taken from some spare NCDs and unwelded cables, so replacement with well-characterized spares is possible in case problems are detected with the welded strings. Preparations for deployment are well underway and on schedule for deployment in September, 2003. The cables which will be used for the final system have been identified, labeled, and cleaned, and will be installed in the final cable tray soon. Final locations for NCD electronics have been identified and necessary SNO reviews have been completed. 1 CENPA Annual Report, University of Washington (2002) p. 31.


  • Page 40

    30 2.15 NCD deployment equipment progress and training J. F. Amsbaugh, M. Anaya,∗ J. Banar,∗ T. A. Burritt, P. J. Doe, G. C. Harper, J. Heise,∗ G. Sillman,† J. Wilhelmy,∗ M. Williams† and J. Wouters∗ The development of the equipment needed to deploy the neutral current detectors (NCDs) into the heavy water acrylic vessel (AV) of the Sudbury Neutrino Detector (SNO) is complete. The deployment is expected to occur in the fall of 2003. A previous progress report1 mentioned completion of the gantry crane, the neckview camera system and equipment leaching tests. The next effort is to train the personnel who will install the deployment equipment at SNO and to train the personnel who will deploy the NCDs. The equipment has been installed at the Los Alamos National Laboratory (LANL) test pool as a training exercise by the installation team. The NCD deployment teams will have their training at LANL in May 2003 on this equipment. At some time before the deployment a final activity review will be done and just before deployment a failsafe review will have to be done. The equipment installed at the LANL test pool is what will be used at SNO for deploy- ment, configured for the shallower depth of the test pool. It consist of a Remote Operated Vehicle (ROV),2 mounting plate, global and neck view cameras, NCD hauldown system, laser welding fixture, and various gloves, viewports, and manipulators. Many of the components can by moved around as needed during deployment. The NCD laser welding fixture is the only major component that is mocked up at the test pool as it is currently being used for the pre-deployment welding of NCDs at SNO. The deployment training uses mock NCDs of the correct diameter and shorter in length. Also available are larger diameter mockups with buoyancies closer to the real NCDs. NCD laser welding experience was obtained by the team participation in the Pre-deployment weld- ing. The trainees will hauldown NCDs, mockup NCD welds, deploy the NCD with the ROV (a small submersible) to an attachment point, and manipulate the readout cables. The ROV pilots will practice in small tight places just like the situation at SNO. Many of the differ- ences between the setup at LANL and SNO will be emphasized. The equipment will be disassembled and shipped to SNO at the end of this training. ∗ Los Alamos National Laboratory, Los Alamos, NM 87545. † Oxford University, Oxford, England. 1 CENPA Annual Report, University of Washington (2002) p. 35. 2 Deep Ocean Engineering, San Leandro, CA 94577.


  • Page 41

    UW CENPA Annual Report 2002-2003 May 2003 31 2.16 Progress of the underground NCD welding prior to deployment J. F. Amsbaugh, J. Banar,∗ T. A. Burritt, G. A. Cox, P. J. Doe, D. Earle,† J. A. Formaggio, J. Heise,∗ A. Krimins,‡ I. T. Lawson,§ B. Morissette,† R. W. Ollerhead,§ S. McGee, L. C. Stonehill and B. L. Wall We have completed construction and testing of the laser welding equipment needed for neutral current detector (NCD) deployment in the Sudbury Neutrino Observatory (SNO) detector. The NCD welding occurs in two stages. First the so called pre—deployment welding in which individual NCD detectors are joined into the largest segments that will fit into the room above the SNO detector. The NCD anchors and readout cable ends are also welded to the segments during pre—deployment. The goal is to minimize the time the SNO detector is off for the second stage of welding, NCD deployment. During the NCD deployment these segments are welded together over the SNO detector into complete strings as they are inserted. The NCDs are scheduled to be deployed in fall 2003. In last year’s report,1 an activity review was outlined. The review panel has finished its work. Equipment documentation, safety reports, assembly procedures, welding procedures, electrical inspection, and manpower schedules were finalized and approved. A test concerning the electrical interference (EMI) of the laser welder on the SNO detector and the water group’s monitoring system was done, with no effects seen. The pre—deployment welding is currently underway. It was estimated that pre—deployment welding would take about three months for the full 96 string array. Since then the array was optimized given the now known solar flux. Only 40 strings will be deployed requiring a total of 150 pre—deployment welds plus a few spares. As of 4 April 2003, 88 welds have been finished. Three were bad and one was possibly bad due to a procedural mishap. One of these four has been cut apart, reflared, and successfully rewelded. The second is being cut apart at this writing with the rest to follow soon. The predeployment welding has had to overcome a series of mishaps, failures and inter- ruptions. Shifts have been curtailed or canceled due to seismic activity, blasting, cage safety concerns and weather. Two gripping cuffs were damaged by improper use, the rotation motor shaft was sheared off, its drive shorted out, and the laser focus depth follower bearing seized. The previously reliable welding laser’s control module failed and the replacement module was bad or incompatible requiring a field service technician. The laser was recalibrated and it was discovered the laser had been delivering 1.5 the rated power. This required developing new laser weld settings. All these have been finished and we have continued production the first two weeks of April. The lessons we have learned by these experiences will be applied to the NCD deployment. ∗ Los Alamos National Laboratory, Los Alamos, NM 87545. † Sudbury Neutrino Observatory, Lively, Ontario, Canada P3Y 1M3. ‡ Queens University, Kingston, Ontario, Canada K7L 3N6. § University of Guelph, Guelph, Ontario, Canada N1G 2W1. 1 CENPA Annual Report, University of Washington (2002) p. 34.


  • Page 42

    32 Neutrino Detectors Double Beta Decay 2.17 Majorana update: construction, evaluation, simulation P. J. Doe, V. M. Gehman, K. Kazkaz, R. G. H. Robertson and J. F. Wilkerson The University of Washington continues its involvement initiated last year1 in the Majorana collaboration.2 The goal of the experiment is to determine the Majorana nature of the electron neutrino (i.e., if the electron neutrino is its own antiparticle) and to measure the effective Majorana mass of the electron neutrino, mν . We propose to accomplish this goal by measuring the rate of neutrinoless double-beta (0νββ) decay of 76 Ge. The full Majorana experiment will utilize a total of 500 kg of segmented germanium crys- tals isotopically enriched to 85% 76 Ge, which will act as both source mass and detector mass. Each crystal will weigh ∼2 kg and the crystals will be situated in a close-pack configuration. The segmentation geometry and final packing configuration are currently under evaluation. Majorana will have a sensitivity to the 0νββ decay halflife of 4.2 × 1027 years, corresponding3 to a mν of 20-70 meV. The Majorana experiment proposal is in the last editing stages. Majorana has two testbed experiments, the Segmented Enriched Germanium Assembly (SEGA) and the Multi-Element Germanium Array (MEGA). SEGA utilizes a single seg- mented enriched crystal, and MEGA is an array of unsegmented, unenriched crystals in a close-pack configuration. These experiments will provide direction for the construction and analysis techniques used in the full Majorana experiment, but also have physics goals in- dependent of Majorana. This past January the University of Washington was involved in evaluation of the SEGA crystal at the Triangle Universities Nuclear Laboratory, and has also been participating over the past year in ongoing construction of MEGA at Pacific Northwest National Laboratory. SEGA and MEGA will both be installed in the Waste Isolation Pilot Plant in Carlsbad, NM in the fall of 2003. There are two major projects underway at CENPA that contribute to SEGA, MEGA, and Majorana. One project is the determination of which preamps will best suit the needs of the experiments. Our criteria for the preamp evaluations include restrictions on response time, resolution, noise, exponential decay of the collected charge, and heat dissipation. The other project is the construction of Geant4 models of SEGA and MEGA to provide Monte Carlo data. The simulations will provide a cross-check on the reliability of the Geant4 physics processes as well as provide a measure on the effect of radioactive contamination of the surrounding environment and construction materials. 1 CENPA Annual Report, University of Washington (2002) p.43. 2 http://majorana.pnl.gov. 3 The spread in mass for a single halflife value comes from uncertainties in the nuclear matrix elements.


  • Page 43

    UW CENPA Annual Report 2002-2003 May 2003 33 100 2.18 Electron-capture branch of Mo and the efficiency of MOON I. Ahmad,∗ J. Aystö,† P. Dendooven,† A. Garcı́a, J. Huikari,† A. Jokinen,† I. Leikola,† F. Naab,‡ H. Pentilä,† S. Rinta-Antila,† J. Szerypo† and S. Triambak Recently Ejiri et al.1 proposed to use 100 Mo as a solar neutrino detector. Neutrinos would undergo the reaction: ν + 100 Mo → e− + 100 Tc, and 100 Tc would decay with a half-life of 15.8 s emitting another e− . The signature for a neutrino absorption would be two electrons, providing a way to obtain clean signals. We have performed a first run on an experiment to measure the matrix element for the EC decay of 100 Tc, which determines the neutrino absorption cross section on 100 Mo. 40000 Tc-K a 35000 10000 30000 1000 Counts per channel 25000 100 20000 10 15 16 17 18 19 20 21 22 23 15000 10000 Tc-K Mo-K a 5000 Mo-K b 0 16 17 18 19 20 21 22 23 Energy [keV] Figure 2.18-1. X-ray spectrum from our preliminary data taken at Jyväskylä. A previous experiment2 measured the 100 Tc EC branch to be (1.8±0.9)×10−5 from which one obtains B(GT;100 Mo →100 Tc) = 0.66 ± 0.33, nominally quite large, but still consistent with zero at 5% c.l. The present experiment was undertaken to produce a more significant result by using a separated beam of 100 Tc. This would reduce backgrounds that hindered a more accurate measurement in the previous experiment. The present experiment was performed using the IGISOL3 facility at the University of Jyväskylä. As shown in Fig. 2.18-1, the production of separated 100 Tc was very successful. shows our x-ray spectrum. The Mo x rays are the signature for the EC capture. We are presently analyzing the data and waiting to get more time at Jyväskylä to finish the data taking. ∗ Argonne National Laboratory, 9700 S. Cass Ave. Argonne, IL 60439. † University of Jyväskylä, Jyväskylä, Finland. ‡ Department of Physics, Notre Dame University, Notre Dame, IN 46556. 1 H. Ejiri et al., Phys. Rev. Lett. 85, 2917 (2000). 2 A. Garcı́a et al., Phys. Rev. C 51, 439 (1995). 3 J. Aystö, Nucl. Phys. A 693, 477 (2001).


  • Page 44

    34 KATRIN 2.19 Characterizing silicon detectors for KATRIN P. J. Doe, T. Gadfort, J. A. Formaggio, G. C. Harper, M. A. Howe, S. McGee, R. G. H. Robertson, J. F. Wilkerson and the KATRIN collaborators The goal of KATRIN1 (Karlsruhe TRItium Neutrino Experiment) is to measure electron en- ergies from tritium beta decay with a mass sensitivity of 0.35 eV. A deviation at the endpoint of the electron energy spectrum will be an indication of a neutrino mass. The experiment consists mainly of a gaseous T2 source, a 10-m diameter retarding-field magnetic-electrostatic analyzer, and a large (approximately 10-cm diameter) silicon detector. A schematic of the project is shown in Fig. 2.19-1. Figure 2.19-1. KATRIN Layout. The UW group has taken up the task of characterizing and providing detectors, electronics and data acquisition for the KATRIN experiment. As part of this task, we have constructed an electron gun that produces 20 keV electrons incident on a particular detector surface (see Section 7.1). We also want to simulate the conditions of the KATRIN detector region as closely as possible. Thus, our detectors should reside in a high vacuum as well as be cooled to increase energy resolution. To meet these requirements, our detectors will be in contact with a Peltier device that will lower the surface temperature by 30 ◦ C. Also, the electron gun will be attached to an oil free vacuum pump that will lower the internal pressure to 10−6 mbar. We plan to measure the uniformity of the detection layer, the effective detection region, the dead layer, and the back-scattering properties of various detectors. Proper characterizing of detectors is crucial to the KATRIN experiment since a neutrino mass measurement depends on stable, high-resolution, low-background recording of the energies of electrons arriving at the focal plane. 1 KATRIN Letter of Intent, A. Osipowicz et al., arXiv:hep-ex/0109033.


  • Page 45

    UW CENPA Annual Report 2002-2003 May 2003 35 3 Nuclear Astrophysics 7 3.1 Be(p,γ)8 B E. G. Adelberger, A. R. Junghans∗ E. C. Mohrmann, P. N. Peplowski, K. A. Snover, H. E. Swanson and TRIUMF Collaborators† Our Phase II measurement1 of S17 (0), the astrophysical S-factor for the 7 Be(p,γ)8 B reaction, is nearly completed. The measurement was made with a metallic 7 Be target2 of 340 mCi initial activity and was carried out in a manner similar to the Phase I measurement.3 Several improvements over the previous measurement were incorporated to further reduce systematic uncertainties. Our Phase I solid angle measurements utilized the 7 Li(d,p)8 B reaction to determine the ratio of the counting rates in a “near” α-detector (used to make our 7 Be(p,γ)8 B measure- ments) and a “far” α-detector, whose solid angle we precisely determined from geometrical measurements. This method required a sizeable calculated correction for the portion of the continuous α-spectra from 8 B decay that lies below the experimental threshold. We eliminated this use of the 7 Li(d,p)8 B reaction by making a 148 Gd α-source on a backing of identical design to that used for the 7 Be target, and using it to determine the solid angle ratio of the α-detectors. As before, this was done using the ratio of counting rates in the “near” and “far” detectors. In addition, the near detector was mounted on a translation stage which aided in the solid angle determination, as well as allowing optimization of the target to detector distance. A small area “thin” Si α-detector (∼150 mm2 , 20 µm thick) was used in addition to the large area (∼450 mm2 , ∼30 µm thick) “thick” detector used previously. The thin detector had a lower background from gamma rays from the target radioactivity, enabling the use of a lower threshold for the detected α-particles. This resulted in a smaller correction for the fraction of the α-spectrum from 8 B decay below the experimental threshold. Measurements of the 7 Be(p,γ)8 B cross section were made over the range of energies from Ēcm = 116 to 1754 keV. Backscattering rate calculations were made (see Section 3.2) and compared to measured data in order to determine the target composition. Knowledge of the target composition was important for determining the energy loss of the proton beam in the target, which becomes especially important at low Ep where the cross section varies rapidly with energy. Our Phase II determination of S17 (0) is in good agreement with our earlier result.3 Our new results are currently being finalized for publication. ∗ Presently at F. Z. Rossendorf Institut fuer Kern-und-Hadronenphysik, Dresden, Germany. † L. Buchmann, S. Park and A. Zyuzin, TRIUMF, Vancouver, BC, Canada. 1 CENPA Annual Report, University of Washington (2002) p. 47. 2 A. Zyuzin et al., Nucl. Instrum. Methods B 187, 264 (2002). 3 A. R. Junghans et al., Phys. Rev. Lett. 88, 041101-1 (2002); see also CENPA Annual Report, University of Washington (2001) p. 4.


  • Page 46

    36 3.2 Target composition analysis for 7 Be(p,γ)8 Be S-factor measurement A. R. Junghans,∗ E. C. Mohrmann, P. N. Peplowski and K. A. Snover In order to properly interpret the experimentally measured cross-sections for the 7 Be(p,γ)8 B reaction, it is necessary to understand the beam energy loss within the target. Uncertainty in the deduced S-Factors S17 (E) due to uncertain target composition is largest at low energies where the cross-section is highly energy dependent. This region is important for the extrap- olation to S17 (0). In order to determine the beam energy loss, the target composition must be known. The energy thickness of the target (∆Etarget ) and the energy loss due to the 7 Be in the target (∆E7Be ) must be measured and used to determine the target composition. ∆Etarget was determined by measuring the narrow 1378-keV resonance in the 7 Be(α, γ)11 C reaction. The width of this resonance can be directly attributed to the energy thickness of the target. ∆E7Be was determined from target activity measurements. The energy loss due to contaminants is ∆Econt = ∆Etarget - ∆E7Be . TRIM1 calculations were made with several possible target stoichiometries that account for ∆Econt . The 8 B backscattering probability for the various target compositions was calcu- lated and compared to the experimentally measured values. It is displayed in Fig. 3.2-1. Figure 3.2-1. 8 B backscattering probability for different target compositions compared to the experimentally measured backscattering probability. Target compositions are, from top to bottom, 7 Be:C:Mo = 63:0:37 (solid curve), 58:8:34 (dashed curve), 57:13:30 (dashed curve), and 35:65:0 (solid curve). The best fit target composition of 7 Be:C:Mo=58:8:34 is within 1σ of pure Mo contami- nation, which was used in the final data analysis. ∗ Presently at F. Z. Rossendorf Institut fuer Kern- und Hadronenphysik, Dresden, Germany. 1 TRIM Version 2000.40, J. F. Ziegler, www.srim.org.


  • Page 47

    UW CENPA Annual Report 2002-2003 May 2003 37 3.3 Search for the 8 B(2+ ) → 8 Be(0+ ) ground state transition M. K. Bacrania, D. W. Storm, R. G. H. Robertson, M. W. Gohl∗ and S. Uehara We are searching for the 8 B(2+ ) → 8 Be(0+ ) ground state transition. The decay of 8 B takes place primarily through the allowed 2+ → 2+ transition to the 3 MeV broad excited state in 8 Be, with an endpoint energy of approximately 14 MeV. It is also possible for 8 B to decay to the ground state of 8 Be, through a second forbidden transition (2+ → 0+ ) with an endpoint of approximately 17 MeV. This second forbidden decay is predicted to have an extremely small branching ratio, and it has never been experimentally measured. The astrophysical significance of this transition is discussed elsewhere.1 We are able to produce beams of 15 MeV 8 B using the CENPA Tandem van de Graaff accelerator (see Section 8.3). The 8 B is implanted into a 500µm Si PiN detector. The resulting 8 B decay (t1/2 = 770ms) produces a β + , and the prompt decay of 8 Be produces a pair of α particles. The β + particle is detected via a scintillation counter, and is used as a coincidence trigger which virtually eliminates beam and electronics backgrounds. For 8 Be excitation energies below 7 MeV, both α particles come to rest inside the Si detector. A decay via the 2+ → 2+ branch results in a broad α energy spectrum centered around 3 MeV, and a decay via the 2+ → 0+ branch will result in a spectral line centered at 92 keV. In 2002, we built and tested the 3 He gas cell system, and conducted beam tests in order to determine backgrounds due to scattered and degraded 6 Li and 3 He from the 8 B production region. A potentially large background source is scattering from the beam regulation slits in the image region beamline. To eliminate this background, we operate the Tandem in GVM regulation mode, with the regulating slits fully retracted out of the beamline. We have reduced our singles background rate to approximately 1 kHz. Our 8 B implantation rate is approximately 6 Hz, and our detection efficiency is approximately 40%. We have had three data-taking runs to date, and are presently analyzing the data from these runs. We have accumulated approximately 3 × 105 coincidence-tagged 8 B decays. In the first run we had relatively poor energy resolution (FWHM = 30 keV for 88 keV photons), and also observed a significant beam-independent background in the region of the 2+ → 0+ transition. In the second and third runs, our resolution has been improved by the addition of a Peltier cooling system for the Si detector (FWHM = 20 keV for 81 keV photons), and by more extensive radiation and RF shielding. ∗ Presently at Justus-Liebig-Universität Gießen, Germany. 1 CENPA Annual Report, University of Washington (2002) p. 49.


  • Page 48

    38 3.4 Is e+ e− pair emission important in the determination of the 3 He + 4 He S-factor? A. E. Hurd and K. A. Snover The astrophysical S-factor S34 (0) for 3 He+4 He fusion is very important for determining the production rate of neutrinos from the decay of 7 Be and 8 B in the sun. Experimental deter- minations of S34 (0) have been made by two methods - detection of the capture γ-rays, and detection of the residual 7 Be activity. The relatively large uncertainty in the recommended value of S34 (0) stems from an apparent difference in the results of these two methods, with the activation experiments yielding a somewhat larger mean value than the capture γ-ray experiments.1 While the statistics in these comparisons are suggestive, but not compelling, this apparent difference has led to the question of whether there might be some other capture reaction mechanism that could explain it, such as E0 pair emission. At low bombarding energies, the capture reaction takes place at large radial distances, and hence processes such as E0 pair emission should be enhanced. The dominant reaction mechanism for 3 He+4 He fusion at low energies is direct E1 photon  emission. Since the operators for E0 and E2 transitions, OE0 = i (ei /e)ri2 and OE2 =  2 i (ei /e)ri Y2m (Ωi ), have the same radial dependence (in the long wavelength limit), E0 direct capture amplitudes for specific partial wave transitions may be numerically related to the corresponding E2 direct capture amplitudes. At low bombarding energies, this leads to a simple relation between the cross sections for E0 pair emission and E2 photon emission. For a transition energy of several MeV in 3 He+4 He →7 Be, the result is a cross section ratio σE0 /σE2 ∼ 2 × 10−4 . The magnitude of this ratio is easy to understand: E2 (real) and E0 (virtual) photons have the same degree of forbiddenness, while pair emission brings an extra power of α, the fine structure constant, as well as a smaller phase space factor. Direct capture calculations show σtotal ∼ σE1 ∼ 400σE2 at these energies. Hence σE0 ∼ 10−6 σtotal and is completely negligible. Sum rule arguments show that E0 pair emission by any other mechanism, such as resonance emission, must also be negligible. In general, all electromagnetic multipole transitions may occur by pair emission. An E0 transition is distinguished from other multipole transitions by the absence of single photon emission rather than the presence of pair emission. Pair emission is a weak function of multipole order, hence E1 pair emission is strongest in 3 He+4 He. Published pair emission coefficients show that E1 pair emission is of order 10−3 of E1 single photon emission at several MeV, so that e+ e− pair emission of any multipolarity is negligible in 3 He+4 He radiative capture. Internal conversion is also negligible. Thus there cannot be significant contributions to the 3 He+4 He →7 Be capture cross section at low energies from electromagnetic emission processes other than single photon emission.2 1 E. G. Adelberger et al., Rev. Mod. Phys. 70, 1265, (1998). 2 K. A. Snover and A. E. Hurd, Phys. Rev. C, in press.


  • Page 49

    UW CENPA Annual Report 2002-2003 May 2003 39 4 Nuclear Structure 4.1 Testing the isospin multiplet mass equation and its implications P. Cheung, A. Garcı́a, G. Hodges, H. Iwamoto, A. Sallaska and S. Triambak The isospin multiplet mass equation (IMME) relates the masses of the members of an isospin multiplet: M (Tz ) = a + bTz + cTz 2 . Recently, a precise determination of the mass of 33 Ar led to the conclusion that an unexpectedly large cubic term was needed to fit the members of the lowest T=3/2 state in the A=33 system.1 Later we found out that the problem originated in an incorrect determination of the mass of the lowest T=3/2 state in 33 Cl. Using the more recent measurements of both 33 Ar and 33 Cl we found excellent agreement with the parabola.2 We are presently setting up an experiment to determine the mass of the lowest T=2 state in 32 S whose uncertainty is claimed to be ±3 keV.3 There is a paper presented at a conference claiming an uncertainty of ±0.4 keV with no published details on how the small uncertainty was achieved.4 We aim to determine the energy of this state with ±0.1 keV uncertainty. As a result, the T=2 multiplet in the A=32 system would constitute the most accurately known quintuplet and it would be interesting to show that the IMME holds to this level of accuracy. On the other hand, we want to pursue measurements in the beta decay of 32 Ar (both of the electron-neutrino correlation and of the log f t for the 0+ → 0+ decay) for which the IMME may help determining the Q value for the decay. 1e+06 1e+05 10000 Counts 3923 keV 8126 keV 1000 100 10 0 2000 4000 6000 8000 Energy (keV) Figure 4.1-1. Gamma ray spectrum from 31 P(p, γ) at the T=2 resonance. The arrows indicate gammas from the decay of the T=2 state. Using the sputter ion source, we produced an implanted 31 P target with a thickness of approximately 4 keV, measured using the 31 P(p, γ) reaction to produce the narrow T=2 state. Fig. 4.1-1 shows a γ spectrum taken at the T=2 resonance. We have performed Monte Carlo simulations that calculate Doppler effects on the gammas keeping in mind the energy losses, detector resolution, detector angle with respect to the beam and the gamma- ray angular correlations to help us understand potential problems. To reduce our systematic uncertainties we shall check for Doppler effects in two oppositely placed Ge detectors. 1 F. Herfurth et al., Phys. Rev. Lett. 87, 142501-1 (2001). 2 M. C. Pyle et al., Phys. Rev. Lett. 88, 122501-1 (2002). 3 G. Audi and A. H. Wapstra, Nucl. Phys. A 595, 409 (1995). 4 M. S. Antony, A. Huck, G. Klotz, A. Knipper, C. Miehé and G. Walter, in Proceedings of the International Conference on Nuclear Physics, Lawrence Berkeley Laboratory, Berkeley, CA, Vol. 1 (1980).


  • Page 50

    40 4.2 Low-temperature measurement of the giant dipole resonance width D. Bazin,∗ J. R. Beene,∗ Y. Blumenthal,∗ M. J. Chromik,∗ M. Halbert,∗ P. Heckman,∗ J. F. Liang,∗ E. C. Mohrmann, T. Nakamura,∗ A. Navin,∗ B. M. Sherrill,∗ K. A. Snover, M. Thoennessen,∗ E. Tryggestad∗ and R. L. Varner∗ We have made a new determination of the width of the Giant Dipole Resonance in the inelastic scattering of 17 O particles from 120 Sn at 80 MeV/u.1 The experiment was carried out at the National Superconducting Cyclotron Laboratory at Michigan State University. The inelastically scattered 17 O particles were detected at forward angles in the S800 spectrometer, and the γ rays were detected in the ORNL - Texas A&M - MSU BaF2 array in coincidence with the S800. γ-ray spectra were recorded for different inelasticities ranging from 20 to 90 MeV. The spectrum of inelastically scattered 17 O particles in coincidence with all γ rays exhibits strong peaks corresponding to the opening of the 1n, 2n and 3n channels, up to 30 MeV of 17 O energy loss, suggesting a close correspondence between inelasticity and residual excitation energy over this range. This conclusion is also consistent with a recent inelastic α-scattering study. Accordingly, we performed a CASCADE analysis of the spectral shape measured for energy losses in the range 20 - 30 MeV, assuming equality between energy loss and residual (initial) excitation energy. The spectrum is fitted well by including a bremsstrahlung component and allowing the GDR strength, width and energy to vary. The result is a width Γ = 4 ± 1 MeV for decays with a mean excitation energy following GDR decay of 9.7 MeV, corresponding to a mean (final-state) temperature of 1.0 MeV. This result for the GDR width is comparable to the width of the GDR built on the ground state of similar mass nuclei, and is much narrower than the value calculated in the adiabatic shape fluctuation model. This new data confirms the trend suggested by other experiments in this mass region, in which the GDR width was found to be narrower than the model predictions for temperatures in the range ∼ 1.3 - 1.7 MeV, and shows clearly that the GDR width increases much more slowly with temperature than predicted by the model. Since pairing corrections should not be important at these temperatures, and shell effects are small for 117 Sn and nearby nuclides, this seems to be a clear failure of the adiabatic shape fluctuation model. Deviations from the adiabatic approximation, which assumes that the time scale for shape fluctuations is long compared to the inverse frequency spread associated with the deformation broadening of the GDR, would result in a smaller GDR width; however, it would be surprising if such deviations occured only at low temperature. Attempts to explain the observed GDR widths by mechanisms other than deformation broadening also do not agree with the data. Hence we conclude that narrow GDR widths observed in Sn and nearby nuclides at low temperature are not understood. ∗ For current address, see reference 1. 1 P. Heckman et al., Phys. Lett. B 555, 43 (2003).

  • View More

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