avatar University of Washington Inc Services


  • Page 1

    INTRODUCTION The Nuclear Physics Laboratory at the University of Washington pursues a broad program of research in nuclear physics 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 and superconducting linac accelerators, 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. The filling of the SNO detector with heavy water was completed in April, 1999, and the complete detector was put into operation. During the past year the SNO data acquisition system developed by UW has been performing extremely well and reliably. The detector has been acquiring good quality production neutrino data since November 1999. These results suggest that the extreme care taken to maintain cleanliness during construction has paid off. The radioactive backgrounds are equal to, or better than, design goals. We are now organizing our resources to focus on keeping pace with the incoming data, running and maintaining the detector. In addition we are preparing for upgrading the SNO detector with the 3He neutral current detector array. The neutral current detectors for SNO are nearing completion, with about 95% of the detectors built and 75% underground in Sudbury. SNO is presently taking data with pure heavy water; during this time the NCD electronics will be commissioned and the alpha backgrounds of the detectors will be measured in preparation for their installation. The electronics for the NCD array are built around a pair of multiplexed 4-channel high-speed digitizing oscilloscopes and 96 channels of spectroscopy ADCs. The Shaper-ADC boards have been successfully completed, and the multiplexers and controller boards are under construction. The emiT experiment has set a limit on the "D" time reversal sensitive component in beta decay that is slightly better than the current world average. A paper describing this result will soon be submitted. An upgrade to the apparatus is underway, and a new measurement is planned in the coming year. Our initial studies of lead perchlorate solution as a Cerenkov medium for neutrino detection have aroused an enthusiastic response among the neutrino community. This detector has promise for long baseline oscillation studies, supernova and other astrophysical neutrino studies and possibly proton decay. We are focusing our effort on understanding the optical properties of the liquid and the Cerenkov response of such a detector. Preliminary 7Be(p,g)8B data has been taken, and most systematic errors have been reduced to a level compatible with our goal of a 5% determination of the astrophysical S-factor. In addition, we are initiating the first-ever direct measurement of the correction for loss of 8B from the target due to backscattering. In our cluster studies a triple coincidence experiment has provided direct evidence for multifragmentation of C60 into three complex fragments. From a separate experiment we have tentative evidence for production of a stable dianion of Si2O5 by fragmentation of the NaSi2O5 monoanion. UW URHI activities in NA49 at CERN and STAR at RHIC continue at a rapid pace. Recent developments in NA49 event-by-event analysis are providing quantitative information on dynamical and flavor fluctuations, resonance effects in multiplicity correlations and possible evidence for correlated particle emission coupled to radial flow to produce recently-observed large-scale two-particle momentum correlations. The STAR

  • Page 2

    event-by-event program has passed a milestone with the recent completion of Mock Data Challenge 3, a collaboration-wide effort in which several EbyE analysis techniques were applied to up to 25,000 simulated Au-Au events in a pilot study which tested many elements of STAR EbyE analysis infrastructure. Preparations for using the STAR detector at RHIC for Hanbury-Brown Twiss interferometry are progressing well. The HBT simulation and analysis software is being tested extensively in preparation for the initial physics run at RHIC this Summer. The STAR HBT Physics Working Group has developed a detailed plan for the analysis of Year-One STAR data that addresses many physics issues in the new and unexplored regime of energy density that will open with the operation of RHIC. The E\"ot-Wash group's recent work addressing the question "what is the weight of gravity itself?" (i.e. testing the equivalence principle for gravitational self-energy) was featured in Physics Today, Scientific American, Science, and Sky and Telescope. We removed an ambiguity in the classic lunar laser-ranging test by comparing, in effect, the accelerations of miniature earths and moons toward the sun. Our results were sufficiently precise that we could unambiguously determine that gravity has the weight predicted by Einstein to about 1 part in 1000. We have recently begun a new round of experiments that test the gravitational inverse-square law at sub-millimeter separations. This work is motiviated by recent theoretical speculations of "large" extra dimensions, which predict fundamentally new behavior in this regime. We have completed our "Big G" apparatus and conducted our first precision measurement of the gravitation constant. The measurement has much smaller systematic and statistical uncertainties than previous determinations. As always, we encourage outside applications for the use of our facilities. As a convenient reference for potential users, the table on the following page lists the vital statistics of our accelerators. For further information, please write or telephone Professor Derek, W. Storm, Executive Director, Nuclear Physics Laboratory, University of Washington, Seattle, Washington 98195; (206) 543-4080, (e-mail: storm@npl.washington.edu) or can also refer to our web page: http://www.npl.washington.edu. We close this introduction with a reminder that the articles in this report describe work in progress and are not to be regarded as publications or to be quoted without permission of the authors. In each article the names of the investigators have been listed alphabetically, with the primary author to whom inquiries should be addressed underlined. Derek Storm, Editor storm@npl.washington.edu, (206) 543-9522 Barbara Fulton, Assistant Editor

  • Page 3

    2000 TANDEM VAN DE GRAAFF ACCELERATOR SPECIFICATIONS TANDEM VAN DE GRAAFF ACCELERATOR A High Voltage Engineering Corporation Model FN purchased in 1966 with NSF funds; operation funded primarily by the U.S. Department of Energy. See W.G. Weitkamp and F.H. Schmidt, "The University of Washington Three Stage Van de Graaff Accelerator," Nucl. Instrum. Meth. 122 , 65 (1974). Some Available Energy Analyzed Beams Ion Max. Curr ent Max. Ener gy Ion Sour ce (part icl e m A) (MeV) 1H or 2H 50 18 DEIS or 860 3 He or 4He 2 27 Double Charge-Exchange Source 3 He or 4He 30 7.5 Tandem Terminal Source 6Li or 7 Li 1 36 860 11B 5 54 860 12C or 13C 10 63 860 * 14 N 1 63 DEIS or 860 16O or 18O 10 72 DEIS or 860 F 10 72 DEIS or 860 * Ca 0.5 99 860 Ni 0.2 99 860 I 0.01 108 860 * Negative ion is the hydride, dihydride, or trihydride. Additional ion species available include the following: Mg, Al, Si, P, S, Cl, Fe, Cu, Ge, Se, Br and Ag. Less common isotopes are generated from enriched material. BOOSTER ACCELERATOR We give in the following table maximum beam energies and expected intensities for several representative ions. "Status of and Operating Experience with the University of Washington Superconducting Booster Linac," D.W. Storm et al ., Nucl. Instrum. Meth. A 287 , 247 (1990). Available Energy Analyzed Beams Ion Max. Cur r ent Max. Pr act ical (p m A) Ener gy MeV p >1 35 d >1 37 He 0.5 65 Li 0.3 94 C 0.6 170 N 0.03 198 O 0.1 220 Si 0.1 300 35Cl 0.02 358 40Ca 0.001 310 Ni 0.001 395

  • Page 4

    UW NPL Annual Report 1999-2000 v Contents 1 Fundamental Interactions 1 1.1 A precise measurement of the 32 Ar superallowed branch . . . . . . . . . . . . . . . . 1 1.2 4 He(α, γ)8 Be . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Time reversal in β decay - the emiT experiment . . . . . . . . . . . . . . . . . . . . 3 1.4 Upgrades to the emiT detector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.5 PNC spin rotation of cold neutrons in a liquid helium target . . . . . . . . . . . . . . 5 1.6 Charge exchange reactions and hadronic probes of weak strength . . . . . . . . . . . 6 1.7 A precision measurement of the Newtonian constant . . . . . . . . . . . . . . . . . . 7 1.8 Calculation of the source strength for the Eöt-wash III torsion balance . . . . . . . . 9 1.9 Millimeter-scale test of the gravitational inverse square law . . . . . . . . . . . . . . 11 1.10 Final results from the Rot-Wash test of the equivalence principle . . . . . . . . . . . 12 1.11 Progress on the Eöt-Wash III rotating torsion balance . . . . . . . . . . . . . . . . . 13 1.12 Gravity’s gravity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 1.13 A sensitive test for CPT-violation in the electron sector . . . . . . . . . . . . . . . . 15 2 Neutrino Physics 17 2.1 The Sudbury Neutrino Observatory . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.2 Initial operation and performance of SNO . . . . . . . . . . . . . . . . . . . . . . . . 18 2.3 Data acquisition in SNO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.4 Data analysis in the first phase of SNO operation . . . . . . . . . . . . . . . . . . . . 20 2.5 The neutral current detector project at SNO . . . . . . . . . . . . . . . . . . . . . . 22 2.6 The data path for the SNO neutral current detectors . . . . . . . . . . . . . . . . . . 23 2.7 In situ determination of backgrounds from neutral current detectors in the Sudbury Neutrino Observatory: CHIME . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.8 Deployment of neutral current detectors in the SNO . . . . . . . . . . . . . . . . . . 26 2.9 SAGE: The Russian American Gallium Experiment . . . . . . . . . . . . . . . . . . . 28 2.10 Neutrino detection using lead perchlorate . . . . . . . . . . . . . . . . . . . . . . . . 29

  • Page 5

    vi 2.11 Spectroscopy of double-beta and inverse-beta decays from 100 Mo for neutrinos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.12 Nuclear spin isospin responses for low energy neutrinos . . . . . . . . . . . . . . . . 33 3 Nucleus-Nucleus Reactions 34 3.1 40 Ca + 208 Pb fusion barrier distributions: status report . . . . . . . . . . . . . . . . 34 3.2 19 F + 181 Ta evaporation residue cross sections as a probe of fission dynamics . . . . 35 4 Nuclear and Particle Astrophysics 39 4.1 Progress in the 7 Be(p,γ)8 B measurement . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.2 Measurement of 8 B backscatter for the 7 Be(p,γ)8 B experiment . . . . . . . . . . . . 43 4.3 WALTA: The Washington Large-area Time-coincidence Array . . . . . . . . . . . . . 44 5 Ultra-Relativistic Heavy Ions 46 5.1 STAR event-by-event program status . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 5.2 STAR global-variables analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 5.3 STAR two-point correlation analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 5.4 STAR primary track definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 5.5 NA49 two-point correlation analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 5.6 Correlated error structure in two-point correlation analysis . . . . . . . . . . . . . . 51 5.7 Autocorrelations and Hubble flow in heavy-ion collisions . . . . . . . . . . . . . . . . 52 5.8 Small-scale structure of the hadronic freezeout surface in Pb-Pb collisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 5.9 Event-by-event analysis and the central limit theorem . . . . . . . . . . . . . . . . . 55 5.10 Scale-local measures and jet correlations - where is the pQCD? . . . . . . . . . . . . 57 5.11 Jet correlations model analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 5.12 Data set viewer (DSV): status and plans . . . . . . . . . . . . . . . . . . . . . . . . . 61 5.13 DSV: applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 5.14 HBT physics at STAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 5.15 Energy loss in thin layers of argon . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

  • Page 6

    UW NPL Annual Report 1999-2000 vii 6 Atomic and Molecular Clusters 67 6.1 High energy fragmentation of C60 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 6.2 Evidence for gas phase Si2 O5 dianion . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 7 Electronics, Computing and Detector Infrastructure 69 7.1 Offline analysis and support computer systems . . . . . . . . . . . . . . . . . . . . . 69 7.2 VAX-based data acquisition computer systems . . . . . . . . . . . . . . . . . . . . . 70 7.3 Electronic equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 7.4 Electronics upgrade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 7.5 The data acquisition electronics for the emiT and SNO NCD experiments . . . . . . 73 7.6 Status of advanced object oriented real-time data acquisition system . . . . . . . . . 74 7.7 Performance overview of a 7-channel silicon microstrip detector . . . . . . . . . . . . 75 8 Van de Graaff, Superconducting Booster and Ion Sources 78 8.1 Van de Graaff accelerator operations and development . . . . . . . . . . . . . . . . . 78 8.2 Booster operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 8.3 Tandem terminal ion source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 8.4 Cryogenic operating experience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 9 Outside Users 83 9.1 Degradation of solar cells in space radiation environments . . . . . . . . . . . . . . . 83 9.2 Scientific Imaging Technologies (SITe) CCDs in the space radiation environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 10 Nuclear Physics Laboratory Personnel 86 11 Degree Granted, Academic Year, 1999-2000 89 12 List of Publications from 1999-2000 90

  • Page 7

    UW NPL Annual Report 1999-2000 1 1 Fundamental Interactions 32 1.1 A precise measurement of the Ar superallowed branch E. G. Adelberger, M. Bhattacharya, A. Garcia,∗ T. Glasmacher,† V. Guimares,∗ A. K. Komives,∗ P. F. Mantica,† H. E. Swanson and S. L. Tabor‡ A recent measurement1 of the e − ν angular correlation coefficient in the superallowed (SA) decay of 32 Ar at ISOLDE (deduced from the Doppler broadening of the SA proton peak) placed a stringent limit on the contribution of scalar currents to the standard model. The main source of systematic uncertainty in this measurement was the uncertainty in the mass of 32 Ar (−2180 ± 50 keV). In order to reduce this systematic uncertainty below the statistical uncertainty, the mass of 32 Ar had to be known with an accuracy of <4 keV. The authors in Ref [1] circumvented this problem by invoking the Isobaric Multiplet Mass Equation for A=32, T=2 quintet, and used the precisely known masses of the other 4 members of the multiplet to predict the mass of 32 Ar with an uncertainty of 3 keV. A more direct determination of the mass of 32 Ar would be desirable. We recently performed an experiment at the National Superconducting Cyclotron Laboratory with a goal of measuring the SA decay branch of 32 Ar with an accuracy of better than 1%. Using this branching ratio and the precisely known half life2 of 32 Ar one can compute the partial half-life for the 32 Ar SA decay. The partial half-life in conjunction with the expected f t value (twice the precisely measured value for T=1→T=1 pure Fermi decays) can then be used to infer the mass of 32 Ar with an accuracy of ∼10 keV. In addition the results from this experiment can also be used to check isospin-mixing calculations for 0+ → 0+ SA decays needed for the determination of Vud and in turn for unitarity tests of the CKM matrix. In our experiment 32 Ar was produced by the fragmentation of a ∼53 MeV/u 40 Ar beam. The fragments were mass-separated using the A1200 fragment mass analyzer and implanted at the center of a 500 µm thick PIPS detector which also served as our delayed proton counter. This detector was sandwiched between two detectors of same dimensions used for detecting the betas as well as for rejecting fast particles from the beam. Another 500 µm PIPS detector located upstream of the telescope provided energy loss and time of flight information needed to identify the incoming fragments. The telescope was surrounded by 5 large HPGe detectors (3 of them segmented clover detectors) to look for β-delayed γ decays of 32 Ar. Analysis of the data is in progress. Careful attention is being paid to determine the number of 32 Ar ions implanted in the implantation detector and to the absolute proton detection efficiency of this detector. ∗ University of Notre Dame, Notre Dame, IN 46556. † NSCL, Michigan State University, East Lansing, MI 48824. ‡ Florida State University, Tallahassee, FL 32306. 1 E.G. Adelberger et al., Phys. Rev. Lett. 83, 1299 (1999). 2 Unpublished data from the ISOLDE experiment of Ref [1].

  • Page 8

    2 4 1.2 He(α, γ)8Be R. Hazama, K. A. Snover, D. W. Storm and J. P. S. Van Schagen∗ We are in the process of completing the analysis of the high-precision measurement1 of the 4 He(α, γ)8 Bereaction. The goal of this experiment is the determination of the isovector M1 and E2 decay widths for a precision test of CVC and second-class currents in the mass-8 system. Last year we reported2 how we had used the 10 B(3 He,pγ)12 C, 13 C(3 He,pγ)15 N, and 15 N(p,γ)16 O reactions to produce photons of 15.1, 8.3, and 13.3 MeV, respectively. These measurements give us detector response line shapes at those various energies, and the coincidence measurements enable us to determine the photon detector efficiency. Additional measurements of the 12 C(p,γ)13 N reactions at the Ep = 14.23 MeV resonance also provide an efficiency measurement, using the results of Marrs, et al.3 We have expanded the Sandorfi and Collins4 lineshape parameterization by adding a second low-energy exponential tail, and, for two of the three detectors, an additional high energy tail. Using the three energies, 15.1, 13.3, and 8.3 MeV, along with spectral shapes calculated in 1 MeV increments with GEANT, we have determined the energy dependence of the parameters, in order to apply them to the broad γ-ray energy spectrum which is due to the width of the final state (first excited state) in 8 Be. Then the yield in any 4 He(α, γ)8 Be spectrum can be determined by fitting the data to a predicted shape resulting from the convolution of the parameterized detector response with the spectral shape predicted by the R-matrix calculations that we described in last year’s Annual Report.5 This scheme uses estimated R-matrix parameters to obtain the yields, and then by fitting the excitation functions, precise values of these parameters will be obtained. Thus iteration may be necessary. Once the data have been fit and all the relative R-matrix parameters have been determined, the absolute isovector M1 width must be determined from absolute cross sections. For this deter- mination, we need absolute efficiencies, which are obtained from the photon yields (in coincidence with protons) and the singles proton yields, in the (3 He,pγ) measurements. We have found that the efficiencies for the spectra obtained without using the plastic-annulus veto are reproducible over time. However, for some detectors the threshold setting on the plastic veto seems to shift, so the efficiencies obtained using the plastic veto vary by a few percent from one measurement to the next. Furthermore, the absolute efficiencies obtained with the 10 B(3 He,pγ)12 C reaction differ by about 5% from those obtained from the 12 C(p,γ)13 N measurement. We are presently investigating possible sources of these discrepancies. ∗ Presently at WRQ, Seattle, 98109. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1999) p. 5. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1999) p. 4. 3 R.E. Marrs, E.G. Adelberger, and K.A. Snover, Phys. Rev. C 16, 61 (1977). 4 A.M. Sandorfi and M.T. Collins, Nucl. Instrum. Methods 222, 479 (1984). 5 Nuclear Physics Laboratory Annual Report, University of Washington (1999) p. 6.

  • Page 9

    UW NPL Annual Report 1999-2000 3 1.3 Time reversal in β decay - the emiT experiment M. C. Browne,∗ H. P. Mumm, A. W. Myers, R. G. H. Robertson, T. D. Steiger, T. D. Van Wechel, D. I. Will, J. F. Wilkerson and emiT Collaboration† The emiT experiment is a search for time-reversal (T) invariance violation in the beta decay of free neutrons. Both CP (charge conjugation - parity) and explicit T violation have been observed in the neutral K meson system. (Theoretically one expects that a combined CPT symmetry exist in all Lorentz-invariant field theories, thus CP and T symmetries must be intimately related.) However, some 35 years since the discovery of CP violation, neither CP nor T violation has been seen in any other system and possible origins are still not well understood. Although CP(T) violation can be accommodated within the standard model of nuclear and particle physics, it may also be an indication of physics beyond the standard model. The standard model predicts T-violating observable in beta decay to be extremely small (Second order in the weak coupling constant) and hence 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 exotic physics. The emiT experiment probes the T-odd P-even triple correlation between the neutron spin and the momenta of the neutrino and electron, Dσn · Pe × Pν , in neutron beta decay. The coefficient of this correlation, D, is measured by detecting decay electrons in coincidence with recoil protons from a polarized neutron beam. This test is complementary to the more sensitive electric dipole moment (EDM) searches. EDMs violate both T and P and thus result from different physical processes. emiT uses a beam of cold (2.7 meV), polarized neutrons from the Cold Neutron Research Fa- cility at NIST in Gaithersburg, MD. The detector was installed on the NG-6 beamline at NIST from November of 1996 until September of 1997. A total of roughly 14 million coincidence events were recorded with a maximum sustained coincidence rate of ∼ 7 Hz. Analysis of these data has been completed.4 The result, D = −0.1 ± 1.3 × 10−3 , represents a small improvement over the current world average. It is limited by statistical uncertainty; however, systematic effects are not insignificant. The largest, DAT P = 4 × 10−4 , results from an asymmetric neutron beam combined with a slightly misaligned (Transverse) neutron polarization. The primary reason for the unex- pectedly large magnitude of this effect was the lack of detector symmetry due to disabled channels in the proton detection segments (Normally the highly symmetric detector is fairly insensitive to such effects). The root cause of this problem was excessive energy loss in the proton detectors and the associated dead time, noise, and electronic failures due to high voltage sparks. Work on an upgrade to the apparatus (see Sec. 1.4) which will solve the problems experienced during the first run is currently in progress at the NPL. A new measurement is planned at NIST in 2001. ∗ Los Alamos National Laboratory, Los Alamos, NM 87545. † Los Alamos National Laboratory, the National Institute of Standards and Technology (NIST), the University of California at Berkeley/Lawrence Berkeley National Laboratory, the University of Michigan, the University of Notre Dame, and NPL. 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 (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, to be submitted.

  • Page 10

    4 1.4 Upgrades to the emiT detector M. C. Browne,∗ H. P. Mumm, A. W. Myers, R. G. H. Robertson, T. D. Steiger, T. D. Van Wechel, D. I. Will and J. F. Wilkerson Past experiments searching for T violation in the β decay of polarized neutrons have reached a sensitivity to D of 1.4 × 10−3 . Technological advances in neutron polarization and an optimized detector geometry should allow emiT to attain a sensitivity to D of 3 × 10−4 , given the current capture flux available at the NG-6 beamline at NIST (1.4 × 109 n/cm2 s). This level of sensitivity represents a factor of five improvement over previous neutron T tests, and may permit restrictions to be placed on several extensions to the Standard Model that allow values of D near 10−3 . The emiT detector consists of four electron detectors (plastic scintillators) and four proton detectors (large-area PIN diode arrays) arranged in an alternating octagonal array concentric with the neutron beam. The average angle between any given proton detector and it’s opposing electron detector is 135◦ . This configuration was chosen to take advantage of the electron–proton angular distribution which is strongly peaked around 160◦ due to the disparate masses of the decay products. When compared to the 90◦ geometry used in previous experiments, this choice results in an increased signal rate equivalent to roughly a factor of three increase in neutron bean flux.1 The protons produced in the decay of free neutrons have a relatively low energy. (The Q-value for the decay is 782 keV, producing protons with energies ≤ 751 keV.) While this allows for a delayed coincidence trigger between the proton and electron (eliminating much of the background due to cosmic rays) it makes detection difficult. The PIN diode array and associated electronics are therefore held at a nominal voltage of −30 kV which accelerates the protons to detectable energies and focuses them onto the PIN diode detectors. During the first run, high voltage related problems stemming from higher than expected energy loss in the PIN diodes led to damaged electronic components and a non-symmetric detector. Sys- tematic effects were less effectively canceled due to the lack of full detector symmetry and a more complex data analysis scheme was required.2 To assure that the second run is not affected by these problems, a number of detector upgrades are currently in progress at NPL. Sensitive electronics have been isolated through the use of analog fiber-optic links. The DAQ electronics have been extensively reworked. Lower power consumption will decrease the cooling requirements, and sharper ADC thresholds will allow better background rejection. This should allow operation at lower voltages thus increasing the stability of the detector. The fiber-optic links and new ADC cards have been constructed and are performing extremely well. In addition, upgrades to the DAQ software will allow closer monitoring of the detector status and will improve the capability for real time data analysis. A tentative decision to replace the PIN diodes with Surface Barrier detectors has been made based on comparative studies of PIPS, PIN and Surface Barrier detectors. Investigations into the source of the observed high voltage instability are continuing. Finally, an upgrade to the NIST reactor will result in a factor of approximately two higher cold neutron flux. It is expected that emiT will resume collecting data in the fall of 2000, likely reaching the design goal of D < 3 × 10−4 early in 2001. ∗ Los Alamos National Laboratory, Los Alamos, NM 87545. 1 E.G. Wasserman, Time Reversal Invariance in Polarized Neutron Decay, Ph.D. thesis, Harvard University (1994). 2 L.J. Lising et al., Phys. Rev. C, to be submitted.

  • Page 11

    UW NPL Annual Report 1999-2000 5 1.5 PNC spin rotation of cold neutrons in a liquid helium target E. G. Adelberger, J. H. Gundlach, D. Haase,∗ G. Hansen,† P. Huffman,‡ B. R. Heckel, D. Markoff,∗ U. Schmidt, H. E. Swanson, M. Snow† and F. Wietfeldt‡ When a neutron beam traverses a material target, the weak interaction between the target and beam causes the neutron spin vector to rotate about the beam momentum, giving rise to the PNC neutron spin rotation observable. When the target is liquid helium, the PNC spin rotation angle is a measure of the neutron-alpha weak coupling constant, which in turn is sensitive to the weak pion- nucleon coupling constant, fπ . We are building an experiment to measure the PNC neutron spin rotation in a liquid helium target using a cold neutron beam at the NIST reactor. This experiment receives support from the NSF as well as the Nuclear Physics Laboratory. The first data runs, completed in 1997, revealed a need for a higher neutron counting rate and a more reliable cryostat. The experimental apparatus is being rebuilt to allow a measurement of the PNC spin rotation angle at the level of 10−7 , six times smaller than the first data runs. There are three components of the apparatus that must be completed before new data can be taken. 1) The cryostat has been rebuilt with reliable vacuum and cryogenic seals. It has been shipped to NIST and has been successfully cooled to 4K at NIST. 2) A long wavelength neutron filter is required to improve the neutron beam polarization. Such a filter, based upon super-mirror reflections, has been demonstrated by Dubbers’ group at the Institut Laue-Langevin. U. Schmidt has designed a similar filter for use by the spin rotation experiment. This filter is being made in Heidelberg and will be tested at NIST as soon as possible. 3) The target insert inside the cryostat must be modified to create and hold superfluid helium. P. Huffman. at NIST, is designing the new insert system. The final design will be completed in the spring of 2000. While the above three major components are being completed, additional improvements are being made to the apparatus. G. Hansen is working on ways to improve the magnetic shielding. His conclusion is that an additional shield of ‘cryoperm’ be added inside the cryostat. Such a shield will be built and tested. Hansen is also working on a system of fluxgate magnetometers inside of the target region to monitor the magnetic field. H.E. Swanson is creating a more flexible data acquisition system for the experiment. The goal of the collaboration is to take new PNC spin rotation data in 2001. ∗ TUNL, North Carolina State University, Durham, NC 27708. † IUCF, Indiana University, Bloomington, IN 47408. ‡ NIST, Gaithersburg, MD 20899.

  • Page 12

    6 1.6 Charge exchange reactions and hadronic probes of weak strength E. G. Adelberger, M. Bhattacharya, C. D. Goodman∗ and Colleagues in IUCF experiments E356 and E406 Matrix elements for the Gamow-Teller (GT) operator between various nuclear states provide particularly interesting nuclear structure information because these matrix elements simply and directly show the relationships between the quantum states of neutrons and of protons in nuclei. Beyond the inherent nuclear structure interest in measuring GT matrix elements there are applica- tions of GT measurements in neutrino physics and in astrophysics. Neutrino detection by absorption of neutrinos on nuclei occurs through Fermi and GT transitions, and some detection schemes rely exclusively on GT transitions. Some steps in astrophysical nucleosynthesis occur through electron absorption and emission in GT transitions. The density of hot electrons in supernova explosions is controlled by electron capture into GT states. Beta decay is without question the most reliable way to measure GT matrix elements. However, because of the very limited energy window that it can explore, it is incapable of measuring GT giant resonances or GT transitions involved in some neutrino detectors. Charge-exchange reactions offer an alternative for measuring GT matrix elements without the energy limitations of beta decay. However, the specific Fermi and GT cross sections, needed to convert reaction cross sections to Fermi and GT strengths, do not have a smooth mass number dependence that can be accurately predicted by reaction dynamics theory. It was discovered at IUCF that the ratio of specific Fermi to GT cross sections on the other hand, has a strong incident proton energy dependence that is universal (i.e. no mass dependence). One can exploit this energy dependence of the ratio of specific cross sections to extract the number of Fermi counts in a spectrum (the Fermi peak is usually not resolved from nearby and underlying GT transitions) and normalize the spectra to GT strength by normalizing to the Fermi transition, bypassing reaction dynamics uncertainties. The best way to test the accuracy of charge exchange reactions in predicting GT strengths is to compare GT strengths obtained from charge exchange reactions to their β-decay values in nuclei where multiple GT transitions, spanning a wide range of excitation energies, are allowed. Two such nuclei are the 37 Cl and 40 Ar. For both of these nuclei a large number of GT transitions have been observed in the β decay of the isospin mirror and the GT strengths obtained from the β decay studies should be identical to those deduced from 0◦ (p, n) reactions to the extent that isospin symmetry is exact. 37 Cl(p,n)37 Ar studies were carried out at IUCF several years ago but the data were not fully analyzed. We are reanalyzing this data with refined data handling techniques. Also, we recently performed 40 Ar(p,n)40 K measurements at IUCF and are in the process of analyzing the data and preliminary results indicate a fair agreement between β-decay and charge-exchange reactions in this case. ∗ IUCF, 2401 Milo B. Samson Lane, Bloomington, IN 47408.

  • Page 13

    UW NPL Annual Report 1999-2000 7 1.7 A precision measurement of the Newtonian constant J. H. Gundlach and S. Merkowitz The gravitational constant is one of the fundamental constants in nature and can only be determined by direct measurement. In the last ten years several measurements were carried out but the uncertainty of each was larger than the value of Luther and Towler1 published in 1982. In addition some of the measurements scattered far outside of the ±128ppm uncertainty of the 1986 CODATA2 value, putting the accuracy of our knowledge of G into question. We developed a new torsion balance technique that offers several significant advantages over previous torsion balance methods.3 A torsion balance apparatus was placed on a turntable located between a set of attractor masses. First, the turntable was rotated at a constant rate so that the pendulum experiences a sinusoidal torque due to the gravitational interaction with the attractor masses. A feedback was then turned on to accelerate the torsion balance apparatus precisely so that the torsion pendulum did not twist with respect to the turntable. The angular acceleration of the turntable, which contains the gravitational angular acceleration, was determined from the second time derivative of the turntable angle. Since the torsion fiber did not experience any appreciable deflection our measurement is independent of many torsion fiber properties and in particular avoids in- and anelasticity problems that may have led to a bias in previous measurements. The largest quoted systematic uncertainty in the best torsion balance measurement to date is due to the pendulum mass distribution. We used a thin vertical plate as the pendulum in conjuction with selecting the purely quadrupole angular acceleration. This trick frees us from having to know the pendulum mass distribution exactly. The attractor masses were located on a separate but coaxial turntable. This turntable rotated with a constant angular velocity difference to the pendulum turntable so that the signal occured at a constant and higher frequency that was selected by the operator. The rotation of the attrac- tor masses allowed us to effectively remove gravitational interactions due to mass changes in the environment and reduce 1/f-noise. We recorded four ≈100 hour long data sets. After each data set the spheres were moved to positions 90◦ away on the attractor turntable shelves. The orientation of the spheres was also changed to average out density fluctuations and non-spherocity. The coordinates of the spheres were measured before and after each data set. We used specially fabricated measuring tools mostly made of invar. Before and after each distance measurement the tools were compared to invar ball bar standards that were calibrated at NIST to within 0.3µm. We ran tests in which the turntable and signal velocities were varied, the ambient magnetic field was exaggerated, and large rotating temperature gradients in phase with the attractor spheres were induced and found the gravitational acceleration was independent of these effects. A second data set with four of the eight attractor spheres removed was taken to verify the linearity of our system and agreement within errors was found. Numerical simulations were conducted to check the accuracy of our data analysis. 1 G.G. Luther and W.R. Towler, Phys. Rev. Lett. 48, 121 (1982). 2 E.R. Cohen and B.N. Taylor, Rev. Mod. Phys. 59, 1121 (1987). 3 J.H. Gundlach et al., Phys. Rev. D 54, R1256 (1996).

  • Page 14

    8 Our preliminary value for the gravitational constant is G = 6.6742 ± 0.0001 × 10−11 m3 kg−1 s−2 . At about ±15ppm this value represents a substantial increase in precision and is not subject to the most significant sources of systematic error or potential bias associated with other torsion balance measurements. Our value is higher than the 1986 CODATA value and is therefore only in marginal agreement. It may however be noted that the weighted average of all other measurements published after 1995 is also higher, by 248±63ppm, than the accepted value. Figure 1.7-1. Cut-away view of the apparatus. The torsion balance consisting of a flat vertical plate hung from a thin tungsten fiber is mounted on an air bearing turntable. The apparatus rotates continuously and smoothly changes its angular velocity so that the torsion fiber is not twisted. The gravitational angular acceleration is transfered to the turntable and a high resolution shaft encoder is used to read out the angular acceleration. Eight stainless steel spheres produced an almost pure gravitational quadrupole field. To discriminate gravitational accelerations from other nearby masses, the attractor spheres are rotated on a second turntable.

  • Page 15

    UW NPL Annual Report 1999-2000 9 1.8 Calculation of the source strength for the Eöt-wash III torsion balance E. G. Adelberger, K. Choi and B. R. Heckel  We calculated the source strength (I)for the Eöt-Wash III torsion balance which is designed to search for violations of the equivalence principle (EP) from new fundamental forces with Yukawa range > 1m. Since our experiment is sensitive only to the horizontal differential acceleration, we only need to calculate the horizontal component of the source strength,  q5 r e−r/λ I⊥ = G d3 rρs (r)( )s (1 + ) 2 (ı̂ sin θ cos φ + ̂ sin θ cos φ), (1) µ λ r where r denotes the position of the source elements, G is Newton’s constant, q5 and λ are the “charge” and range of the EP-violating interaction, µ is the mass in AMU and ρ is the density. The calculation of I⊥ involves two steps: first modeling, second integration of the effective source masses. We used the MULTI and M2CAD14 program, written by Eric Adelberger and Nathan Collins, and AutoCad to model the source near the pendulum, such as a sand box, magnet and coil, pits under the magnet, cranes, concrete blocks and beams in the cyclotron room, the pond above the cyclotron room, and the Nuclear Physics Laboratory (NPL) building (Fig. 1.8-1). We also used a commercial program called “Surface Display System” (SDS) to model the sources such as the hillside excavations for the NPL, the topographic terrain out to a radius 40km, including Lake Washington, part of Puget Sound, and the bedrock underground down to 20km. Figure 1.8-1. Entire NPL Building. We used rectangular coordinates with x̂ = East, ŷ = North, ẑ = up to integrate over the effective sources. Except for sources in the NPL Building which were calculated by the MULTI program, we integrated each part of the source separately on a 3-dimensional rectangular grid using the Richardson method. The results of calculation,1 expressed in the magnitude |I⊥ | and the direction θ in the EW-NS frame for q5 = B, L, (B−L) √ 2 , and (B−2L) √ 5 , are shown in Fig. 1.8-2, where B and L are the baryon and lepton numbers. 1 E.G. Adelberger et al., Phys Rev. D 42, 3283 (1990). 2 Nuclear Physics Laboratory Annual Report, University of Washington (1999) p. 10. 3 Germund Dahlquist, Numerical Methods, Prentice-Hall, Inc. (1974).

  • Page 16

    10 Figure 1.8-2. The source strength (I⊥ ) and the direction θ.

  • Page 17

    UW NPL Annual Report 1999-2000 11 1.9 Millimeter-scale test of the gravitational inverse square law E. G. Adelberger, N. A. Collins, C. Deufel,∗ J. H. Gundlach, B. R. Heckel, C. D. Hoyle, D. J. Kapner, U. Schmidt and H. E. Swanson Motivated by higher-dimensional theories which predict deviations from the gravitational inverse- square law at the millimeter scale,1 we have constructed a torsion balance experiment sensitive to such anomalous effects. We adapted many components from the Rot-Wash apparatus2 for this short-range test. The apparatus consists of a specially designed torsion pendulum which hangs as close as possible above a rotating attractor of similar geometry (see Fig. 1.9-1). A thin electrostatic screen separates the pieces. We chose the geometry in order to minimize gravitational torques while accentuating any torque due to a short-ranged Yukawa or power law interaction. In the last year, much design and construction effort went into this project. Using a Fourier- Bessel expansion of the gravitational potential, we created a program to calculate gravitational and Yukawa torques on the pendulum. This allowed us to choose the best possible geometry. We constructed a 3-axis rotation and translation platform for precise pendulum positioning. Also, mechanisms for attractor alignment as well as an improved optical system were made. We presently await the completion of the tantalum and copper pendulum and attractor pieces. In the meantime, we constructed a simpler attractor and pendulum from copper and aluminum respectively, each of which has 10 identical holes spaced evenly about the azimuth. Preliminary measurements have been performed with this design. We leveled the pendulum to within 200 µrad of local vertical by rotating it above a differential capacitor. We then made the electrostatic screen and attractor parallel to the pendulum using optical and mechanical methods so that the total angle between them is less than 1 mrad. The pendulum and attractor were made coaxial to within 60 µm with gravitational techniques described in G.L. Smith, et al.2 We are presently finishing our initial data set which includes pendulum/attractor separations between 0.15 mm and 8 mm. We expect to be able to reach separations of 0.10 mm. We should have significant results from this experiment in the coming months. Figure 1.9-1. The pendulum and rotating attractor. The shaded sectors are tantalum, the un- shaded ones copper. The mirrors used for the optical readout system and cutouts for gravitational alignment are also shown. The electrostatic screen has been omitted for clarity. ∗ REU summer student from Wabash College, Crawfordsville, IN 47933. 1 See, for example, N. Arkani-Hamed, et al., Phys. Lett. 429B, 263 (1998). 2 G.L.Smith, et al., Phys. Rev. D 61, 022001 (1999).

  • Page 18

    12 1.10 Final results from the Rot-Wash test of the equivalence principle E. G. Adelberger, J. H. Gundlach, B. R. Heckel, C. D. Hoyle, G. L. Smith∗ and H. E. Swanson We have completed our short-range test of the equivalence principle using the Rot-Wash ap- paratus. The experiment involved rotating a 3 ton 238 U attractor around a torsion pendulum containing a copper/lead composition dipole. We analyzed new data taken in 1997–98 and re-analyzed previously reported 1996 data.1 Com- bining the data sets, we found a differential acceleration of ∆aCu−Be = (1.0 ± 2.8) × 10−13 cm/s2 . Our results set new constraints on equivalence-principle violating interactions with Yukawa ranges down to 1 cm, and improve significantly existing limits between 10 km and 1000 km (see Fig. 1.10-1). For a full account of experimental design, protocol, and results, see G.L. Smith et al.2 Figure 1.10-1. 95% Confidence limits on coupling strengths of hypothetical Yukawa interactions with range λ and vector charges of baryon number, neutron number minus proton number, and baryon number minus lepton number (top to bottom). The top axis shows exchange boson mass and on the right we see dimensionless “fine structure constants” for such interactions. The heavy curve is from this work; the shaded areas represent previously excluded regions. (For references, see G.L. Smith et al.)2 ∗ Presently at Skagit Community College, Mt. Vernon, WA 98273. 1 J.H. Gundlach, et al., Phys. Rev. Lett. 78, 2523 (1997). 2 G.L. Smith, et al., Phys. Rev. D 61, 022001 (1999).

  • Page 19

    UW NPL Annual Report 1999-2000 13 1.11 Progress on the Eöt-Wash III rotating torsion balance E. G. Adelberger, K. Choi, J. H. Gundlach, B. R. Heckel, C. D. Hoyle, D. J. Kapner, S. M. Merkowitz, U. Schmidt and H. E. Swanson We are building a new rotating torsion balance to search for violations of the equivalence principle due to new fundamental forces with Yukawa ranges >1 m. As described in previous reports, the balance will consist of a composition-dipole pendulum containing titanium, beryllium, or aluminum test bodies. The pendulum is suspended inside a vacuum chamber and hangs from a constantly rotating turntable. We completed a new data acquisition and control system written in C++ that runs under Windows 98. The data collection includes averaging filters that can reduce the amount of 60Hz and other high-frequency noise in the data. The operator can now control most of the apparatus from a single program, such as set the turntable speed, rotate the pendulum relative to the turntable, change the settings on the lock-in amplifiers, and lock the turntable. The user can create multiple windows with real-time plots of the data with a variety of settings. The program can perform several special tasks unattended, such as: change the turntable speed without exciting the pendulum, rotate the pendulum with respect to the autocollimator, and execute a calibration run. Future enhancements will include catching and cooling the pendulum. We will also implement a feedback loop that accelerates the turntable to minimize the torsion fiber twist that was successfully used on the apparatus to measure the gravitational constant (see Sec. 1.7). We built an insulated shed around the experiment and use a water chiller to maintain a constant temperature in the enclosure. Another chiller is used to keep the room temperature constant to within 0.2K (at a single location). The vacuum chamber is enclosed in a insulated jacket that is also kept at constant temperature by circulating chilled water. We re-wired the temperature sensors (AD-590s) and built a new controller box with high and low gain options (set through the data acquisition program). With these changes, the noise on the sensors was reduced to about 1mK. We are currently investigating ways to further improve their sensitivity. We installed new legs and linear displacement LVDTs (linear variable differential transformers) on the turntable that will allow us to remotely level the turntable. The leg height can be coarsely adjusted with a stepper motor that drives a leveling screw. Fine adjustments are made with a peltier element inside lead feet that change the temperature of the feet allowing us to use their thermal expansion to change their height (see Sec. 1.12). We installed Q21, Q22, Q31 gravity gradient compensators and have completed first-round measurements of their strength. A second measurement of the gradients is in progress, after which the compensators will be further “tuned” to fully cancel these components of the local mass distri- bution. A second measurement is necessary because several changes in the room mass distribution were made since the initial measurements. We designed and built several new components of the apparatus: vacuum chamber, autocol- limator, corner-mirror and holder, phi-top stage that includes a stepper motor for changing the height of the pendulum, pendulum parking stand, and two shielded tilt sensor holders. The design for the final pendulum1 is finished and is being manufactured commercially. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1999) p. 10.

  • Page 20

    14 1.12 Gravity’s gravity E. G. Adelberger, J. H. Gundlach, B. R. Heckel, S. M. Merkowitz, U. Schmidt and H. E. Swanson We measure the composition dependent acceleration of the four test bodies of our rotating torsion balance towards the sun. Two test bodies have the composition of the earth and the other two the composition of the moon. Our result removes the ambiguity of the lunar-ranging tests of the equivalence principle for gravitational self-energy. Since our last reports1,2,3,4 we upgraded the Eöt-Wash II torsion balance in various ways: - Active tilt compensation: The laboratory and therefore our torsion balance change its tilt typically by 1 µrad per day. This tilt causes an additional twist of the fiber of the torsion balance. With a feed through coefficient of typically 10−2 , measured by purposely tilting the apparatus, the additional twist of the fiber leads to an false effect of 10 nrad, which is 20 times bigger than our instrumental sensitivity. The uncertainty of the correction for this tilt effect was the major contribution to our systematic error budget. We now reduce this effect by a factor of 50. Our active elements (feet that support the apparatus) consist of a lead tube coupled through a peltier element to a temperature stabilized water reservoir. The peltier element acts as a heat pump and allows us to change the temperature of the lead tube and therefore by thermal expansion its length. With a sophisticated computer controlled feed back loop we managed to control the length of our feet within 10 nm. By replacing two of the three feet with our active ones, we are able to keep our apparatus leveled within 20 nrad over weeks. - Vibration isolation of the fiber: We insert a small bellows between the prehanger of the fiber and it upper mount. This bellows decouples the vibration of the rest of apparatus from the fiber above 10 Hz and therefore reduces the noise in our signal. - Improvements of the data acquisition: We changed our data acquisition program from a solely DOS-based to a real-time task switching one. This allow us to introduce a 288 times over sampling together with digital filtering. The maximum error of the computer controlled timing of the ADC conversion is 2µs. Therefore the feed through of the 60 Hz power line and the 120 Hz lock-in synchronization frequencies is reduced by a factor of 2×10−4 due to digital filtering. Further the digitizing noise of our data is reduced by a factor of 20. In addition we reduced the electronic channel crosstalk from 2.3 × 10−5 to 7.5 × 10−6 . In combination with some smaller changes in the temperature-measurement electronics the systematic error of our signal due to temperature effects is now smaller by a factor of 10. We finished a second run of data taking and are working on the data evaluation. For the data taken between June 1999 and February 2000 our preliminary result for a composite-dependent component of the earth-moon differential acceleration toward the sun is ∆aCD /as = 1.0 × 10−13 with a statistical error of 1.4 × 10−13 . We estimate the systematic error to be 0.2 × 10−13 from the current state of the data analysis. 1 S. Baeßler et al., Phys. Rev. Let. 83, 3585 (1999). 2 Nuclear Physics Laboratory Annual Report, University of Washington (1999) p. 13. 3 Physics Today, Nov. 1999, pp 19-21. 4 Scientific American, Feb. 2000, p. 20.

  • Page 21

    UW NPL Annual Report 1999-2000 15 1.13 A sensitive test for CPT-violation in the electron sector E. G. Adelberger and B. R. Heckel Because any quantum field theory is necessarily invariant under the combined symmetries of C, P and T, it is difficult to assess the relative sensitivities of different experimental tests for CPT-violation. Kostelecky and co-workers1 have developed a general theoretical framework which admits CPT invariance at the cost of breaking Lorentz invariance in a controlled way. Their model posits spontaneously generated vector and axial vector fields, fixed in the universe, with which the usual standard model particles interact in Lorentz covariant form. “Observer Lorentz invariance” is maintained (invariance under arbitrary boosts, translations and rotations of the observer) while “particle Lorentz invariance” (invariance under arbitrary boosts, etc. of a test particle) is broken. Recently Bluhm and Kostelecky2 showed that the results from a 1998 Eöt-Wash experiment3 using a torsion pendulum containing nearly a mole [(7.8± 0.6)× 1022 ] fully polarized electrons spins could be used to set powerful constraints for CPT and Lorentz non-invariance in the electron sector. The Kostelecky CPT-and-Lorentz-violating Lagrangian for electrons is L = −aµ ψγ µ ψ − bµ ψγ5 γ µ ψ (2) where aµ and bµ are vector and axial vector fields fixed in the cosmos. The second term in Eq. 2 (plus CPT-conserving but Lorentz-violating terms in Kostelecky’s Lagrangian) lead to an interaction of a polarized electron with the cosmic field H = −b̃ · σ (3) where b̃ = b+ (CPT-conserving but Lorentz-noninvariant terms). This would have cause our spin pendulum to experience a torque around b̃. Torsion balances are sensitive only to the vertical component of the torque, but as the earth rotates the vertical axis sweeps out a cone in fixed space. By monitoring the torque on the pendulum as a function of the orientation of the spin relative to axes fixed in space we can detect interactions of the form given in Eq. 2 and by exploiting the phase of the torque signal in our rotating balance we can probe all three components of b̃. The North-South component of b̃ produces a torque that is constant as the earth rotates, while the remaining two components produce torques that vary with a period of a sidereal day. We have reanalyzed M.G. Harris’s thesis data (which he had analyzed for signals fixed in the lab frame) and obtained the following results for the x̂, ŷ, and ẑ components of b̃ (ẑ points from the center of the earth to its North rotational pole, and x̂ points from the earth to the sun at the vernal equinox) b̃x = (0.1 ± 2.1 ± 0.8) × 10−20 eV (4) b̃y = (1.7 ± 2.3 ± 0.8) × 10 −20 eV (5) b̃z = (−10.3 ± 3.9 ± 7.6) × 10 −20 eV (6) 1 See for example, D. Colladay and V.A. Kostelecky, Phys. Rev. D 55, 6760 (1997). 2 R. Bluhm and V.A. Kostelecky, Phys. Rev. Lett. 84, 1381 (2000). 3 M.G. Harris, Ph.D. thesis, University of Washington (1998).

  • Page 22

    16 where the first error is statistical and the second systematic. The systematic errors for b̃x and b̃y are smaller than for b̃z because only the components of temperature, magnetic field, gravity gradients and turntable wobble that vary over the course of a sidereal day contribute to b̃x and b̃y , while their average values contribute to b̃z . The above results provide the most sensitive test, to date, for Lorentz and CPT violation in the electron sector.

  • Page 23

    UW NPL Annual Report 1999-2000 17 2 Neutrino Physics 2.1 The Sudbury Neutrino Observatory Q. R. Ahmad, J. F. Amsbaugh, M. C. Browne,∗ T. V. Bullard, T. H. Burritt, P. J. Doe, C. A. Duba, S. R. Elliott, R. Fardon, J. E. Franklin, A. A. Hamian, R. Hazama, G. C. Harper, K. M. Heeger, M. A. Howe, A. W. Myers, J. Orrell, A. W. P. Poon,† R. G. H. Robertson, K. Schaffer, M. W. E. Smith, T. D. Steiger, T. D. Van Wechel and J. F. Wilkerson The Sudbury Neutrino Observatory (SNO), is a joint Canadian/US/UK effort to measure the spectral distribution and flavor composition of the flux of the higher-energy, 8 B neutrinos from the Sun by using a 1000 tonne detector of heavy water. The SNO detector relies on three different neutrino interaction processes. 2H + νe → p + p + e− Charged Current (CC) 2H + ν x → p + n + νx Neutral Current (NC) e− + νx → e− + νx Elastic Scattering (ES) Because of this unique sensitivity to the known flavors of neutrinos, SNO should be able to yield a definitive answer to the question of whether neutrino flavor oscillations are occurring in the Sun in a way that is independent of the standard solar models (SSM) of how the Sun works. As a second generation solar neutrino detector, SNO will also have a significant increase in statistical sensitivity compared to present detectors. Assuming an energy threshold of 5 MeV, the SNO detector is expected to observe 12.7 CC events/day (SSM/2), 5.5 NC events/day (SSM), and 1.2 ES events/day (SSM/2). This past year has been very productive and exciting. In November 1999 the detector made the transition from commissioning to taking production neutrino data. The SNO data acquisition system and tools developed by UW have proven very reliable and user friendly. Fabrication of the neutral current detection (NCD) proportional counter array is nearing com- pletion. Approximately 92% of the detectors in the array have been constructed and 75% are underground at SNO where their performance is being monitored. The final data acquisition system nears completion. It is anticipated that the complete system will be underground and func- tioning in Summer 2000. The hardware required to deploy the array has been designed at UW and fabrication by the NPL shop is almost complete. With the successful operation of the detector and the near completion of the NCD array we are focusing our attention on analyzing the PMT data from the detector. To this end we have formed a collaboration, the “West Coast Alliance” (WCA), which consists of University of Washington, Lawrence Berkeley National Laboratory, Los Alamos National Laboratory and the University of British Columbia. ∗ Los Alamos National Laboratory, Los Alamos, NM 87545. † Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720.

  • Page 24

    18 2.2 Initial operation and performance of SNO K. M. Heeger and the SNO collaboration Using heavy water as a target, the Sudbury Neutrino Observatory (SNO) has the unique capa- bility to measure both the flux of electron neutrinos as well as the flux of other neutrino flavors. In the first phase of the experiment SNO will run with pure D2 O and make a measurement of the charged-current interaction of neutrinos with deuterium. SNO is on-line and taking data. Since November of last year SNO has been taking produc- tion data while calibrating the detector and measuring background rates at the same time. The electronics and data acquisition are operating smoothly, with typical trigger rates of about 15-20 Hz. The hardware trigger threshold is set to roughly 2 MeV which corresponds to 18 PMTs above pedestal threshold. Currently the average channel threshold is about 0.25 photoelectrons. More than 98.5% of all 9598 PMT channels are fully working. The PMT death rate from all causes is currently 0.5% per year. In late 1998 serious problems were observed associated with breakdown in the HV connectors leading into the PMT tube bases. This problem was significantly reduced by regassing the light water that surrounds the PMTs with nitrogen. Since then the detector has operated with minimal problems from high voltage breakdown or tube loss. Flashers, which are light-emitting breakdowns inside the PMT, are a potential background to neutrino data. They occur at a rate of about 1/minute and are mostly correlated with seismic events. The signature of these events has been well characterized so that these events can be removed during data cleaning. A number of calibrations have been performed in SNO to measure the optical response of the detector, its energy resolution, and background rates. The optical calibration, performed using a laser source and a diffuser ball suspended in the D2 O, shows that the timing resolution of the phototubes is near the expected design goal of 1.7 ns. The attenuation length of the heavy water is measured to be greater than 100 m at 550 nm. A triggered 16 N source which produces a 6.1 MeV γ-ray is used for the energy calibration near the analysis threshold. The 16 N calibration data is in good agreement with the Monte Carlo prediction of roughly 8-9 hits/MeV, electron equivalent. The analysis threshold for neutrino data is determined by the backgrounds in SNO. Water assays indicate that background levels in the H2 O and D2 O are near the design goals. With full D2 O recirculation ongoing the water cleanliness continues to improve. Current in H2 O Target in H2 O Current in D2 O Target in D2 O Uranium 5.0x10−15 g/g 4.5x10−13 g/g 1.0x10−14 g/g 4.5x10−14 g/g Thorium 1.0x10−13 g/g 3.7x10−14 g/g ≤ 4.0x10−15 g/g ≤ 3.7x10−15 g/g Radon 1.0x10−13 g/g 4.5x10−13 g/g 1.0x10−14 g/g 4.5x10−14 g/g During the initial phase of data taking the SNO detector has performed very well. New calibra- tions sources, such as U and Th sources, and a 8 Li source will come on-line in the next few months to measure the characteristics of principal backgrounds and understand better the detector’s energy response. SNO is expected to report its first results later this year.

  • Page 25

    UW NPL Annual Report 1999-2000 19 2.3 Data acquisition in SNO C. A. Duba, A. A. Hamian, P. Harvey,∗ M. A. Howe, J. Roberts,† P. M. Thornewell‡ and J. F. Wilkerson The Sudbury Neutrino Observatory data acquisition (SNO DAQ) system is designed to provide continuous readout of the detector’s 9547 photomultiplier tubes (PMTs) with a minimum of dead time. It is made up of four main parts: system initialization and control, hardware readout, event building/recording, and monitoring. Since the SNO DAQ system has been described extensively in past Annual Reports,1 only a brief overview with the most recent updates is provided here. The main system interface is the SNO Hardware Acquisition Real-time Control program (SHaRC). A major enhancement to SHaRC was the introduction of a number of pre-defined standard run types, which allow the operator to set up the detector with a single click. New monitoring ca- pabilities were added, including the ability to monitor the battery backup system and safely shut down the detector in the event of a sustained power outage. In addition, SHaRC now maintains and updates a private web page every 15 minutes so that the DAQ group can remotely monitor critical system parameters and system behavior. Two serious bugs, which were causing SHaRC to crash, were discovered and fixed. One was an ftp transfer problem that was causing random crashes during the transfer of log files and the other was a problem writing the log files to disk. With these fixes, SHaRC crashes have been practically eliminated. Hardware readout is provided through a MVME167 68040 embedded processor running tightly optimized code that places the data into dual port memory for event building. There have been no significant changes to the event readout software in the last year. It has proven to be remarkably robust and trouble-free for the last two years. The event building and recording software, which runs on a SUN Ultra Sparc 1 workstation, was recently combined into a single program. The new version is simpler for operators as it eliminates the need to run a separate recording program. It is easier to maintain, uses 30 MB less memory, and has more potential for speed optimization. The data throughput rates of the new builder are the same or slightly better than the old builder, with sustained rates of up to 400 kB/s demonstrated. The new builder fixes a number of record/data header ordering problems and ensures that all dispatched records are in the proper sequence in the final ZDAB files. In addition, audible warnings have been added to alert the operators to critical error situations. Most data monitoring is provided by XSNOED, which can display either the near real time dispatched data stream or offline ZDAB files, and also by SNOSTREAM, which displays running summary plots of detector performance. The DAQ group maintained an almost continuous onsite presence in 1999, and will continue to provide full offsite support in the upcoming year. With the exception of the new builder/recorder program, there have been only minimal changes to the SNO DAQ software since the start of production running in November 1999. The system has been performing reliably, and there are no major upgrades foreseen in the coming year. ∗ Queens’s University, Kingston, Ontario, Canada. † Sudbury Neutrino Observatory, Sudbury, Ontario, Canada. ‡ F5 Networks Inc., Seattle, WA. 1 Nuclear Physics Laboratory Annual Report, University of Washington 1997 p.20-23, 1998 p.18-20, 1999 p.16-18.

  • Page 26

    20 2.4 Data analysis in the first phase of SNO operation Q. R. Ahmad, C. A. Duba, A. A. Hamian, R. Hazama, K. M. Heeger, J. Orrell, R. G. H. Robertson, T. D. Steiger and J. F. Wilkerson SNO production running began on November 2, 1999. Since the data acquisition system is complete and the NCD array is nearly complete, many members of the University of Washington SNO group have turned their focus towards analysis of the phototube data. This analysis is underway on a number of fronts, including work on support analyses which will be used by the entire collaboration, work on a complete solar neutrino analysis in close collaboration with a subset of SNO scientists, and special-interest projects undertaken by individual graduate students. A crucial first step in the SNO analysis is the data cleaning process, which is designed to remove as much of the non-physics events from the data as possible, leaving a data sample with a vastly improved physics to non-physics event ratio. Identifying physics events like neutrino and muon interactions from this reduced data set will be a significantly easier task than starting with all of the original data. Each of the more than twenty data cleaning cuts has been, or is currently being, studied and characterized in terms of how many physics and non-physics events it flags. Two of these cuts are being studied at UW. One is a cut which flags events with a high charge sum, but with few PMTs hit. The second cut flags “orphans”, PMT data bundles which have not been successfully built into a complete event. Once all of the cuts have been fully characterized, the next step will be to identify an optimal set of these cuts which maximizes the non-physics event removal while minimizing the number of “good” events which are rejected. There are two other main areas of support analysis in which UW has been involved; both are considered “detector support” analyses whose results feed directly back into improving the detector performance on an ongoing basis. One is the study of time-variation of the electronics pedestals. These pedestals are measured on a weekly basis, and keeping track of their stability over time is a good way of identifying problem electronics channels. The second area of detector support analysis is summarizing the detector livetime, both the total time (including calibration and maintenance runs), and the neutrino mode livetime. These results are provided to the collaboration on a weekly basis, and are being used as an aid to help maximize the neutrino mode livetime. Fig. 2.4-1 shows the total SNO livetime from the start of production running until March 14, 2000. The UW solar neutrino analysis is being performed in conjunction with collaborators from several other SNO institutions, and is progressing rapidly on all fronts. A recent area of focus for UW has been in producing a candidate sample of solar neutrinos. This has included formulating and implementing a run selection scheme, as well as hand-scanning events which survived a series of non-physics removal cuts. The UW group has also taken on responsibilities in fitter studies for vertex reconstruction, and detector acceptance studies. Muons interacting in the SNO detector are a special interest of a senior UW graduate student (Q.R. Ahmad), and will form the basis of his Ph.D. thesis. In particular, the use of detected neutrons from spallation interactions of muons in the detector as an additional energy calibration is being pursued. In addition, the possibility of using muon events themselves as a way of performing or checking the optical calibration is being investigated. With the SNO detector running smoothly in production mode, the coming year promises to be

  • Page 27

    UW NPL Annual Report 1999-2000 21 a time of significant progress in terms of analysis. The UW group has already made important contributions both to support analyses and towards obtaining solar neutrino results, and expects to be fully integrated in these efforts in the future. SNO Livetime (1999/11/2 to 2000/3/14) %livetime 100 80 60 40 20 Total livetime 0 | | | | Nov. 1999 Dec. 1999 Jan. 2000 Feb. 2000 number of days 140 120 Integrated livetime 100 integrated time 80 integrated SNO livetime 60 40 20 0 | | | | Nov. 1999 Dec. 1999 Jan. 2000 Feb. 2000 Figure 2.4-1. Top figure: fraction of SNO livetime versus days since start of production running. Bottom figure: integrated SNO livetime since start of production running.

  • Page 28

    22 2.5 The neutral current detector project at SNO J. F. Amsbaugh, M. C. Browne,∗ T. V. Bullard, T. H. Burritt, P. J. Doe, C. A. Duba, S. R. Elliott, R. M. Fardon, J. E. Franklin, R. Hazama, G. C Harper, K. M. Heeger, A. W. Myers, R. G. H. Robertson, K. K. Schaffer, M. W. E. Smith, T. D. Steiger, T. D. Van Wechel, S. L. Veatch and J. F. Wilkerson SNO is unique for its neutral current detection capability, with all neutrino flavors interacting with deuterium to produce free neutrons. These neutrons are readily detected, thus measuring the total active neutrino flux from the sun, independent of neutrino oscillations. One technique for neutron detection is to insert an array of 3 He proportional counters into SNO. These Neutral Current Detector (NCDs) observe neutrons via the following capture reaction: n + 3 He → p + 3 H + 764 keV The charged proton and triton ionize the gas, depositing their combined 764 keV of kinetic energy to produce a detectable signal. The NCDs are made from chemical vapor deposition (CVD) of Ni onto a mandrel to form tubing and endcap components. The CVD process results in ultra low U and Th contamination (1-2 ppt Th). Quartz tubing forms the high voltage and signal feedthrough to a Cu anode wire. The ultra low radioactivity of all components is required to minimize backgrounds from d(γ,n)p reactions in the heavy water and also from alpha tracks in the NCDs which can mimic neutrons. The array is designed to contain 300 detectors, construction of which is nearing completion. Following is the status as of 3/17/00. NCDs being stored underground at SNO: 226 = 75% Built or partially built (at UW): 51 = 17% Still needing construction: 23 = 8% The endcaps were previously being built in conjunction with IJ Research in Santa Ana, Cali- fornia. By mutual agreement, the contract with this company was terminated and remaining work was moved to UW. Endcap production has now been successfully implemented at UW for the past 6 months using the RF sputtering machine in the department of physics. The remaining 46 endcaps needed to build the array should be completed in the next few weeks. We are awaiting shipment of the final Ni tubes from CVD Manufacturing, Ontario. NCDs are stored underground prior to their deployment in SNO. The purpose is to allow the cosmogenic 56 Co to decay away with its 79 day half life. This period is also providing an opportunity to measure the intrinsic U and Th levels in the walls of the NCDs. The NCDs are expected to be installed into SNO within the next one to two years. ∗ Los Alamos National Laboratory, Los Alamos, NM 87545.

  • Page 29

    UW NPL Annual Report 1999-2000 23 2.6 The data path for the SNO neutral current detectors M. C. Browne∗ C. A. Duba, A. W. Myers, R. G. H. Robertson, T. D. Van Wechel and J. F. Wilkerson The data path of the SNO NCD system is in its final stages of refinement. The computer data acquisition system will continuously acquire data from two paths, VME and GPIB. The fast signal path is activated by all signals above a low trigger threshold (ADC threshold). The NCD shaper board then integrates the pulse through a shaper and converts it to a digital value. All of the fast signal path information flows through VME and PCI bus. The slower path digitizes signals with a higher threshold, but below that of the lowest expected neutron pulse. The slower path also incorporates GPIB protocol in addition to VME and PCI to carry its digitized pulse information. Figure 2.6-1. The path of data through the NCD electronics. The fast signal path converts each integrated pulse above the low level ADC threshold into an eleven bit digital signal in the shaper boards. The NCD shaper boards are run from a linear VME crate, which itself is controlled by an SBS 618 controller. The 618 communicates to the data acquisition computer through a pair of optical fibers, which significantly reduces the signal-induced noise over electrical controllers. When the system is complete, the triggering of the shaper board will send a signal to increment the SNO Master Trigger Card, which will assign a ‘GTID’ to the event, and return a signal to the NCD GTID counter. The NCD GTID counter will also relay the GTID information to the NCD data acquisition through the VME bus. If no additional digitization data arrives, the computer ∗ Los Alamos National Laboratory, Los Alamos, NM 87545.

  • Page 30

    24 places its data into the main SNO data stream through the builder, and sends a copy into the NCD independent data stream. The slow signal path carries more exacting information about the pulse seen by one of the proportional counter strings. The digitization is triggered by the higher-level digitization threshold within each multiplexer. The multiplexer then sends a signal to the NCD Trigger Controller Card (NCD TCC), which determines the appropriate scope for digitization (see Sec. 7.5). The data acquisition computer communicates with the controller card through a 32 bit VME I/O module riding on the VME bus. The controller card is able to set the digitization thresholds, offset voltages, and read back the analog thresholds through 13 custom DAC boards. The controller card also relays to the NCD DAQ computer information about possible multiple hits during signal digitization. The digitized pulse gets to the NCD DAQ computer through GPIB, and the computer immedi- ately sends off copies of the pulse. First, the pulse splits off to the main SNO data stream, with the appropriate GTID number assigned to it. Second, another copy of the pulse runs through LANL’s Analyst program, which will compile pertinent characterization numbers. Finally, the raw NCD data stream sends a record of the data down the NCD independent data stream, where it will be saved for cases in which the main SNO data stream might be inactive or unavailable.

  • Page 31

    UW NPL Annual Report 1999-2000 25 2.7 In situ determination of backgrounds from neutral current detectors in the Sudbury Neutrino Observatory: CHIME J. F. Amsbaugh, P. J. Doe, S. R. Elliott, G. C. Harper, K. M. Heeger, G. Miller,∗ A. W. P. Poon† and R. G. H. Robertson The use of ultra-low background 3 He proportional counters as neutron detectors in the SNO detector will provide a means of measuring the neutral-current interaction rate of solar neutrinos. Since the binding energy of the deuteron is only 2.2 MeV, gamma rays from natural decay chains in the proportional counters photodisintegrate the deuteron and thus simulate the neutral current signal. This poses stringent limits on the radiopurity requirements for the counters. An in-situ measurement of the background from seven neutral current detectors will determine the activity of 232 Th in the neutral current detectors and thus determine an upper limit on the photodisintegration background generated by the radioactivity in the NCD array. The Neutral Current Detector (NCD) array (see Sec. 2.5) consists of 775 m of 3 He proportional counters arranged in 96 strings with 300 counters. The counter bodies are made of about 450 kg of ultra-pure chemical-vapor deposited nickel which contains natural Uranium and Thorium in equilibrium. The photodisintegration gammas are produced in the decays of the 232 Th (208 Tl → 208 Pb, E =2.615 MeV) and 238 U (214 Bi → 214 Po, E =2.445 MeV) chains and in the decay of the γ γ relatively long-lived 56 Co (56 Co → 56 Fe, Eγ >2.224 MeV). A Construction Hardware In Situ Monitoring Experiment (CHIME) has been designed to mea- sure the NCD originated background in the presence of the D2 O contribution prior to the deploy- ment of the entire NCD array. It is an in-situ measurement of the construction materials used in the NCD array. Seven individual counters arranged in a close-packed configuration with a total mass of about 5000 g and an overall length of 112 cm will be deployed as a background test source in the SNO detector. The construction materials and procedures for the CHIME counters are essentially identical to those in the real NCD array. To assure its cleanliness before deployment the CHIME unit was checked for leachable radon. The table below lists the experimental results expressed as microgram equivalents of 232 Th and 238 U. The poor tracer recovery associated with the CHIME sample is due to physical loss of the sample in processing. Sample 232 Th (212 Po) 238 U (214 Po) 238 U (226 Ra) tracer recovery Background ≤ 0.057 ≤ 0.0044 ≤ 0.0032 82.5% CHIME ≤ 0.26 ≤ 0.035 ≤ 0.037 8.1% The CHIME has been stored underground since December 20, 1998 in order to allow the cosmo- genically produced 56 Co to decay. The deployment of CHIME is expected to be similar to that of a standard calibration source and will have minimum impact on the SNO operating schedule. The CHIME is negative buoyant and can be deployed along the central axis of SNO using a specially designed deployment mechanism. We plan to perform the CHIME background test experiment later this year. ∗ Los Alamos National Laboratory, Los Alamos, NM 87545. † Lawrence Berkeley National Laboratory, Berkeley, CA 94720.

  • Page 32

    26 2.8 Deployment of neutral current detectors in the SNO J. F. Amsbaugh, M. Anaya,∗ P. J. Doe, G. C. Harper, M. W. E. Smith, W. A. Teasdale∗ and J. Wouters∗ The deployment of the neutral current detectors (NCDs) into the heavy water acrylic vessel of the Sudbury Neutrino Detector (SNO) requires specialized equipment and carefully designed procedures. Care must be exercised to prevent damage to and contamination of the acrylic vessel and heavy water. Each NCD counter is 2–3 meters in length and is stored underground at SNO after manufacture. To minimize SNO down time, the predeployment welding will join counters into segments just small enough for deployment in the deck clean room over the vessel. At deployment, the segments are inserted into the vessel and welded together over the neck. The completed NCD string is lowered to the bottom where a remotely operated vehicle (ROV) can take it to the correct attachment point. Finally, the cable is manipulated and connected to the electronics feedthroughs. We have been testing the deployment equipment,1 which replaces the calibration glove box, and developing the procedures at both the UW and a 20 ft deep pool at Los Alamos National Laboratory (LANL) (see Fig. 2.8-1). The global view camera pole and boathook, which manipulate the NCD cables at the bottom of the vessel neck, were assembled over a 39 ft deep tank at UW. The boathook’s float provides neutral buoyancy in heavy water and its position counteracts torque when the pole is not vertical. The two worked very well together. Two sessions of development and testing were done at the LANL pool with what final equipment was finished and prototypes for the rest, except the laser welding fixture. This culminated with a NCD deployment design and plan review on 18-Jun-1999 at LANL. A review panel, two members from the SNO collaboration and two members from outside SNO, were to find things overlooked, comment on the design and procedures developed, paying particular attention to cleanliness and failure mode recovery. Several observers also attended the review which included presentations, a tour of the equipment installed at the pool, and a demonstration deployment of one NCD string with the ROV. Participants got hands-on experience the next day if they so desired. The review in general was positive and several suggestions were made. A summary of the remaining work includes implementing suggestions and improvements learned from the tests with prototypes. The laser welding fixture and a revised NCD lowering mechanism are designed and are being made. The following have had their design thought out but need to be detailed and finalized, after which they will be made as shop time is available: a lifting device to help remove the glove box, lower the ROV into the acrylic vessel, and place the deployment plate; a neck view camera system; revised glove mountings; a predeployment welding station, using the new welding fixture; and various devices to aid in handling, tilting, and inserting the long NCD segments. The expected welding equipment completion should enable weld team training to occur in time for predeployment welding to begin in late summer, assuming NCD counter production has finished. Before and during this welding, final NCD deployment procedures can be developed and reviewed, followed by training the deployment teams. Thus the earliest the NCD deployment could begin is 3–6 months after the underground predeployment welding has started. ∗ Los Alamos National Laboratory, Los Alamos, NM 87545. 1 Nuclear Physics Laboratory Annual Report, University of Washington (1999) p. 22.

  • Page 33

    UW NPL Annual Report 1999-2000 27 Figure 2.8-1. The NCD deployment equipment.

  • Page 34

    28 2.9 SAGE: The Russian American Gallium Experiment S. R. Elliott and J. F. Wilkerson The Russian-American Gallium Experiment (SAGE) is a radiochemical solar neutrino flux mea- surement based on the inverse beta decay reaction, 71 Ga(ν,e− )71 Ge. The threshold for this reaction is 233 keV which permits sensitivity to the p-p neutrinos which comprise the dominant contribution of the solar neutrino flux. The target for the reaction is in the form of 55 tonnes of liquid gallium metal stored deep underground at the Baksan Neutrino Observatory in the Caucuses Mountains in Russia. About once a month, the neutrino induced Ge is extracted from the Ga. 71 Ge is unstable with respect to electron capture (t1/2 = 11.43 days) and, therefore, the amount of extracted Ge can be determined from its activity as measured in small proportional counters. The experiment has measured the solar neutrino flux extractions between January 1990 and December 1997 with the +7.2,+3.5 result; 67.2−7.0,−3.0 SNU which has been published.1 (The former set of uncertainties are statistical and the later set are systematic.) This is well below the standard solar model expectation of 138 SNU. Fig 2.9-1 shows a plot of the extraction data. Data now exists through the end of 1999 and it is being prepared for publication. The collaboration has used a 517-kCi 51 Cr neutrino source to test the experimental operation. The energy of these neutrinos is similar to the solar 7 Be neutrinos and thus makes an ideal check on the experimental procedure. We have published this result in 1996. The result,2 expressed in terms of a ratio of the measured production rate to the expected production rate, is 0.95 ± 0.11. This indicates that the discrepancy between the solar model predictions and the SAGE flux measurement cannot be an experimental artifact. We expect to receive DOE funding for FY00 that will directly support SAGE. A CRDF proposal has been submitted and is in the review process. SAGE is a mature experiment whose operation has become routine. The University of Washing- ton plays a role in the analysis of the data. This past year we assisted in the design and construction of new proportional counters for the experiment. With the publication of the archive papers, the focus is now on the long term measurement of the solar neutrino flux to reduce the statistical uncertainty to a level comparable to the systematic uncertainty. 500 K+L Results SAGE I 400 SAGE II and III Ga theft period 300 SNU 200 100 0 1990 1992 1994 1996 1998 Extraction Date Figure 2.9-1. The individual measurements of the solar neutrino production rate. The uncertain- ties are statistical only. 1 J.N. Abdurashitov et al., Phys. Rev. C 56, 055801 (1999). 2 J.N. Abdurashitov et al., Phys. Rev. Lett. 77, 4708 (1996).

  • Page 35

    UW NPL Annual Report 1999-2000 29 2.10 Neutrino detection using lead perchlorate P. J. Doe, S. R. Elliott, C. Paul,∗ R. G. H. Robertson and J. F. Wilkerson Due to its large cross section and relative cheapness, lead is an attractive neutrino detection medium. Both charged and neutral current reactions are available to distinguish electron neutrinos from other neutrino flavors. νe +208 P b ⇒ 208 Bi∗ + e− ⇓ 207 Bi + xγ + n ′ νx +208 P b ⇒ 208 P b∗ + νx ⇓ 208−y P b + xγ + yn. Unfortunately, detector schemes utilizing lead tend to be complicated and do not lend themselves to the massive scales required for neutrino studies. One of us (S.R.E.) noted that Lead Perchlorate (Pb(ClO4 )2 ) is highly soluble in water; 500g of Pb(ClO4 )2 can be dissolved in 100g H2 O with a resultant density of 2.7 g/cc. This solution appears quite transparent to the eye and raises the question as to whether one could realize massive Pb detectors using the water Čerenkov technique. The presence of 35 Cl enables the free neutron to be detected with high efficiency by means of the 8.4 MeV capture gamma rays. This is the same technique employed by the SNO detector. Finally, at a cost of $10k/tonne in 100 tonne quantities, detectors on the order of kilo-tonnes are conceivable. There are a number of physics possibilities with such a detector. Using Pb as a target would make a powerful supernova detector. The average energies of neutrinos emitted by a supernova are expected to follow a hierarchy: Eνe < Eν̄e < Eνµ,τ The observation of high energy νe would be an indication of oscillations to µ, τ flavors. The large cross section and delayed coincidence νe signature of Pb could provide a high statistics oscillation experiment at a beam stop where a short duration beam spill allows the temporal separation of any monoenergetic νe which result from νµ oscillation. The hydrogen content of Pb(ClO4 )2 solution also makes the detector sensitive to ν̄µ → ν̄e oscillations. Measuring the cross section for neutrino interactions in Pb is also of importance to supernova modelers investigating the explosion mechanism and transmutation of nuclei. To determine if a Pb(ClO4 )2 Čerenkov detector can be built we have investigated the optical properties of the solution. The spectral transmission is given in Fig 2.10-1. This is very encouraging since there are no obvious absorption lines. We constructed a special apparatus in order to measure ∗ Presently at Ratheon, Redondo Beach, CA.

  • Page 36

    30 12 11 10 Attenuation Length =43cm PMT Signal (mA) 9 8 7 6 5 4 3 15 25 35 45 55 65 Path Length (cm) Figure 2.10-1. Spectral transmission of light though a 1 cm long cell containing an 80% solution of Pb(ClO4 )2 . The transmission is referenced to deionized water. 100 90 80 Transmission (%) 70 60 50 40 30 20 10 0 200 300 400 500 600 700 Wavelength (nm) Figure 2.10-2. Attenuation of 430 nm light passing through an 80% solution of Pb(ClO4 )2 . the attenuation of light. The results are given in Fig 2.10-2. An attenuation length of ≈0.5m is not sufficient to build a large Čerenkov detector. Diluting the solution only further reduced the attenuation length. This suggests that the loss of light is due to scattering, perhaps due to the formation of Pb salts or polymeric molecules such as Pb4 (OH)4 , possibly as a result of reactions with dissolved O2 and CO2 . We are not aware of any physical reason why large attenuation lengths are not achievable in Pb(ClO4 )2 , (as they are in salt water), therefore we plan to determine what steps are necessary to achieve this.

  • Page 37

    UW NPL Annual Report 1999-2000 31 100 2.11 Spectroscopy of double-beta and inverse-beta decays from Mo for neutrinos H. Ejiri, J. Engel,∗ R. Hazama, P. Krastev,† N. Kudomi‡ and R. G. H. Robertson Spectroscopic studies of two β-rays from 100 Mo are shown to be of potential interest for inves- tigating both the Majorana ν mass by neutrinoless double β-decay(0νββ) and low energy solar ν’s by inverse β-decay with a multi-ton 100 Mo detector. The unique features of the present approach with 100 Mo are as follows. 1) The β1 and β2 with the large energy sum of E1 + E2 are measured in coincidence for the 0νββ studies, while the inverse β-decay induced by the solar ν and the successive β-decay are measured sequentially in an adequate time window (about 30 s) for the low energy solar-ν studies. The isotope 100 Mo is just the one that satisfies the conditions for the ββ-ν and solar-ν studies, as shown in Fig. 2.11-1. 2) The large Q value of Qββ =3.034 MeV gives a large phase-space factor G0ν to enhance the 0νββ rate and a large energy sum of E1 + E2 = Qββ to place the 0νββ energy signal well above most background (BG) except 208 Tl and 214 Bi. The energy and angular correlations for the two β-rays can be used to identify the ν-mass term. 3) The low threshold energy of 0.168 MeV for the solar-ν absorption allows observation of low energy sources such as pp and 7 Be. The GT strength to the 1+ ground state of 100 Tc is measured to be large: (gA /gV )2 B(GT )=0.52±0.06 by both the (3 He,t) reaction and electron capture. Then 100 Mo is found to have large capture rates even for low energy solar ν’s. The solar-ν capture rates are 639 SNU for pp ν, 206 SNU for 7 Be ν and 27 SNU for 8 B ν. The solar-ν sources are identified by measuring the inverse-β energies. Since only the 100 Tc ground state can absorb 7 Be ν and pp ν, the ratio of 7 Be to pp is independent of the uncertainty(∼ 15%) of the B(GT ) value. 4) The measurement of two β-rays (charged particles) enables one to localize in space and in time the decay-vertex points for both the 0νββ and solar-ν studies. β, γ associated with BG are also measured. The tightly localized event in space and time, together with relevant βγ measurements, are key points for selecting 0νββ and solar-ν signals and for reducing correlated and accidental BG by factors ∼ 10−5 . 5) Possible detectors for the present objective are under study. One realistic possibility is an ensemble of plastic scintillator modules read out by wavelength- shifter fibers to get adequate energy and spatial resolutions. ∗ Dept. Physics and Astronomy, University of North Carolina, Chapel Hill, NC 27599. † Dept. Physics, University of Wisconsin, Madison, WI 53706. ‡ Research Center for Nuclear Physics, Osaka University, Ibaraki, Osaka 567-0047, Japan.

  • Page 38

    32 8 B 8 GR ENERGY(MeV) 3 1+ 2 pep Solar ν 1+ 7 1 γ Be + β + pp 0 0 1 100 100 Mo Tc Qec=0.168 β1 , β Qββ=3.034 β2 0+ 100 Ru Figure 2.11-1. Level and transition schemes of 100 Mo for double beta decays (β1 β2 ) and two beta ′ decays (ββ ) induced by solar ν.

  • Page 39

    UW NPL Annual Report 1999-2000 33 2.12 Nuclear spin isospin responses for low energy neutrinos H. Ejiri Nuclear spin isospin responses for low energy neutrinos of current astroparticle physics interests have been studied. Neutrinos are key particles for new particle physics beyond the standard minimal electro-weak theory, and sensitive probes for studying stellar evolution and astronuclear processes in the sun and stars. Low energy neutrinos with energies of Eν ≃ 0.1 ∼50 MeV have been studied extensively by using nuclei as micro-laboratories for investigating elementary particles and fundamental interactions. Nuclear weak processes involved are vector and axial-vector weak interactions. Accordingly, nuclear isospin and spin-isospin responses for neutrinos are crucial in studying the low energy neutrinos through nuclear weak processes. Nuclei show spin isospin responses characteristic of nuclear spin isospin structures such as nuclear spin isospin core-polarizations and nuclear spin isospin giant resonances. Nuclear spin isospin responses have recently been investigated by means of hadronic charge-exchange reactions. Nuclear spin isospin responses are investigated also by relevant electromagnetic processes. Discussions on nuclear spin isospin responses relevant to low energy neutrino studies in nuclei have been made with emphasis on current subjects as follows: 1. Neutrino studies in nuclear micro-laboratories and nuclear responses for neutrinos. 2. Nuclear spin isospin responses, nuclear spin isospin giant resonances, and nuclear medium effects on nuclear spin isospin responses. 3. Nuclear spin isospin responses studied by hadronic charge-exchange spin-flip and non spin-flip reactions. 4. Nuclear spin isospin responses for neutrinos associated with double beta decays. 5. Nuclear spin isospin responses for solar neutrinos studied by nuclear inverse beta decays and nuclear detectors for solar neutrinos. 6. Nuclear spin isospin responses for supernova and accelerator-based neutrinos.

  • Page 40

    34 3 Nucleus-Nucleus Reactions 40 208 3.1 Ca + Pb fusion barrier distributions: status report A. L. Caraley, A. R. Junghans, T. L. McGonagle and R. Vandenbosch During this past year we began an experiment to measure fission fragment cross sections for 40 Ca + 208 Pb, at laboratory energies from 195 to 232 MeV, with the goal of determining the distribution of fusion barriers for this system. The experimental configuration was similar to that used by Bierman et al.1 A silicon surface barrier E-∆E telescope of ∼7 msr was used to detect primary fission fragments. Coincident frag- ments were detected in a 7-strip silicon detector2 which was positioned on the opposite side of the beam such that the entire angular distribution of the complementary fragments was subtended. Two additional silicon surface barrier detectors were placed at ± 30◦ to monitor the beam en- ergy and quality and to provide for normalization to Rutherford scattering. At the beginning of each period of beamtime the monitor detectors were calibrated in energy using elastically scattered 40 Ca, provided by the laboratory’s tandem Van de Graaff accelerator. To obtain the beam energies needed for measurement of the experimental excitation function the booster linear accelerator was used in conjunction with the tandem. Data of sufficient statistics were first collected during a two-week beamtime in December 1999. (Earlier attempts failed due to beam quality/quantity difficulties caused by the damaged low-energy tube in the tandem. See Sec. 8.1 for details.) During the December run, fission data sets were collected at 9 beam energies, of the nearly 25 needed to complete the excitation function. Addi- tional beamtime was scheduled for January 2000 in order to complete the measurement. However, a preliminary determination of the fusion barrier distribution revealed inconsistencies within the December data. A more detailed off-line analysis was conducted. It was determined that an unde- tected and unexplained failure in the strip detector had resulted in the loss of fragment-fragment coincidence counts. Unfortunately, it was not possible to perform any systematic corrections to the measured data in order to determine a true coincidence yield. As a result of these difficulties, it was decided that the beamtime scheduled for January 2000 be used to investigate the failure of the strip detector. Tests of the overall strip detector collection efficiency, as well as that of each individual strip, were made under a variety of operating conditions. An evaluation of those tests is still in progress and will comprise much of the Undergraduate Independent Study material of T.L. McGonagle. (Preliminary results are presented in Sec. 7.7.) 1 J. D. Bierman, P. Chan, J. F. Liang, M. P. Kelly, A. A. Sonzogni, and R. Vandenbosch, Phys. Rev. Lett. 76, 1587 (1996). 2 Micron Semiconductor Limited, Lancing, Sussex, England.

  • Page 41

    UW NPL Annual Report 1999-2000 35 19 3.2 F + 181 Ta evaporation residue cross sections as a probe of fission dynamics A. L. Caraley As reported earlier,1 fission fragment and evaporation residue cross sections from the 19 F + 181 Ta system were measured at several beam energies from E = 121 to 195 MeV. This experiment lab was part of a series of experiments conducted to study both the fusion-fission and fusion-evaporation channels of this system. The primary goal of these experiments was to investigate the energy depen- dence of the statistical model level density parameter using experimental α-particle multiplicities and spectral shapes. This investigation has been completed2 and the final results prepared for publication.3 Our efforts have now turned to a closer examination of the fission and evaporation residue cross section results themselves. In part (a) of Fig. 3.2-1 statistical model calculations using a Monte Carlo version of cascade4,5 (solid line) and joanne6 (dotted line) are compared to our experimental results, as well as to those of Hinde et al.7 (Parameters used: an =A/11 MeV−1 , af /an =1.04 and kf =1.00.) The measured cross sections remain at ∼300 mb from 121 to 195 MeV, while the calculated values decrease by approximately a factor 2 over the same energy range. Of particular interest is the observation that at beam energies above ∼120 MeV the measured residue cross sections are in excess of standard statistical model predictions. This corresponds to a compound nucleus excitation energy threshold for non-statistical fission of approximately 75 MeV. Similar thresholds have been reported for this system, based on measurements of pre-scission neutrons8,9 and giant-dipole-resonance γ-rays.10,11 Excess residue cross sections have been observed previously in a few other systems;12,13,14 analyses of these results have focused on the role of nuclear viscosity in slowing down the fission decay of the compound nuclei formed in these reactions.13,15,16 A similar analysis of our 19 F+181 Ta results is described here, although alternative explanations are not ruled out.17 As a preliminary examination of the role of fission hindrance in producing the observed excess residue cross sections, a mc cascade calculation was performed with the maximum excitation energy allowed for fission set to 80 MeV. The results of this calculation are depicted by the dashed line in Fig. 3.2-1(a) and indicate that some degree of fission hindrance is consistent with our data. The influence of several different aspects of fission hindrance are compared in Fig. 3.2-1(b). 1 Nuclear Physics Laboratory Annual Report, University of Washington (1998) p.31. 2 Nuclear Physics Laboratory Annual Report, University of Washington (1999) p.31. 3 A.L. Caraley, B.P. Henry, J.P. Lestone, R. Vandenbosch, Phys. Rev. C, to be submitted. 4 F. Pühlhofer, Nucl. Phys A280, 267 (1977). 5 M.G. Herman, U. of Rochester Nuclear Structure Laboratory Report UR-NSRL-318 (1987), unpublished. 6 J.P. Lestone et al., Nucl. Phys. A559, 277 (1993). 7 D.J. Hinde et al., Nucl. Phys. A385, 109 (1982). 8 J.O. Newton et al., Nucl. Phys. A483, 126 (1988). 9 D.J. Hinde et al., Nucl. Phys. A452, 550 (1986). 10 D. Fabris et al., Phys. Rev. Lett. 73, 2676 (1994). 11 R. Butsch et al., Phys. Rev. C 41, 1530 (1990). 12 K.-T. Brinkmann et al., Phys. Rev. C 50, 309 (1994). 13 B. B. Back et al., Phys. Rev. C 60, 044602 (1999). 14 A. L. Caraley, SUNY Stony Brook Ph. D. thesis (1997). 15 I. Diószegi et al., Phys. Rev. C 61, 024613 (2000). 16 D. J. Hofman et al., Phys. Rev. C 51, 2597 (1995). 17 J.P. Lestone, Phys. Rev. C 59, 1540 (1999).

  • Page 42

    36 mc cascade and joanne calculations with fission delay times of 30×10−21 s are given by the short-dashed line and dotted line, respectively. (A total fission delay time, τdelay , of (20–40)×10−21 s has been derived18 from the work of Thoennessen and Bertsch.19 ) The short-long-dashed line represents the results of calculations including both the Kramers factor,20 with γ=1, and the τdelay =30×10−21 s. A calculation with a transient time,21 τtrans , of 30×10−21 s, along with γ=1, is illustrated by the solid line. Figure 3.2-1. mc cascade and joanne calculations using several forms of fission hindrance. Also, results of viscosity-based analyses using an excitation energy independent nuclear dissipation coefficient, γ. For the full viscosity analysis, dissipation was included in mc cascade in an approach similar to that of Butsch et al.22 The fission width was modified by both the Kramers factor20 and a simplified time-dependent expression.9 Transient times used in the time-dependent expression were calculated as a function of γ.21 Elapsed times were determined from time distributions governed by the total decay width, including fission. No other additions/changes to the code were necessary. In our initial analyses, γ was held constant at its input value for all excitation energies in order to allow for direct comparisons with earlier works.13,16 Evaporation residue excitation functions calculated for γ=0, 0.5, 1, 2, 5 and 10 are shown in Fig. 3.2-1(c). Values of γ needed to reproduce the measured residue cross sections at each laboratory energy, γf it , were determined through interpolation. The results for γf it are illustrated in Fig. 3.2-1(d), along with the results of our own similar analyses 18 R. Vandenbosch, Phys. Rev. C 50, 2618 (1994). 19 M. Thoennessen and G.F. Bertsch, Phys. Rev. Lett. 71, 4303 (1993). 20 H. A. Kramers, Physics VII, 284 (1940). 21 K. H. Bhatt, P. Grangé, B. Hiller, Phys. Rev. C 33, 954 (1986). 22 R. Butsch et al., Phys. Rev. C 44, 1515 (1991).

  • Page 43

    UW NPL Annual Report 1999-2000 37 of the 16 O+208 Pb and 32 S+184 W residue cross sections. The present results for γ are comparable, qualitatively, with the work of Back et al.13 and Hofman et al.16 Back et al.13 have suggested that the splitting between the 224 Th and 216 Th systems is due to shell effects. Our results do not contradict that conclusion. Figure 3.2-2. mc cascade calculations, including viscosity, using an excitation energy dependent nuclear dissipation coefficient, γ. Presently, we are conducting a more rigorous analysis that includes an excitation energy de- pendence for γ within the calculations. The same statistical model parameters, an =A/11 MeV−1 , af /an =1.04 and kf =1.00, were maintained in order to fit the near-barrier results without viscos- ity. For simplicity, a threshold for dissipation was estimated at Uj =60 MeV. Dependencies on the square root of Uj and on Uj are both being investigated, to approximate linear and quadratic dependencies on temperature, respectively. Some of our preliminary findings are illustrated in Fig. 3.2-2. The smallest amount of viscosity needed to achieve the maximum hindrance possible, i.e.: comparable to turning fission off completely, is depicted by the solid lines in Figs. 3.2-2(a) (T) and Fig. 3.2-2(b) (T2 ). (The functional forms of the energy dependencies are noted in the figure legends.) Excitations functions with a 50% reduction in γ still result in reasonable reproductions of the data, qualitatively, and are illustrated by the dotted lines. The calculated cross sections are quite sensitive to further reductions in viscosity, as illustrated by the dashed lines which correspond to 75% reduction. The upper and lower limits for the resulting energy dependencies for γ, as determined from our measured 19 F+181 Ta evaporation residue cross sections, are depicted in Fig. 3.2-2(c). Fig. 3.2-2(d) illustrates the corresponding limits for the dependence of viscosity on temperature (approximate: A=200, a=A/11 MeV−1 ) of the emitting system. The lower and upper limits for γ(T) are given

  • Page 44

    38 by the solid and dashed lines, respectively. The results show γ to be increasing moderately with temperature, consistent with the general expectations of one-body dissipation.23 Within the context of fission hindrance, it can be concluded that some dissipation is needed to reproduce our experimental residue cross sections. However, the exact energy dependence of γ is beyond the sensitivity of our results — and dependent also on the choice of statistical model parameters used in the calculations. In addition, although large values of γ are not required to reproduce the residue results, the possibility is not ruled out. The upper limits depicted in Fig. 3.2- 2(c) and (d) should be viewed with caution. In the future, a simultaneous fit to the residue cross sections and as well as other observable would determine more precisely any dependence of γ on excitation energy (temperature). 23 J. Blocki et al., Ann. Phys. 113, 330 (1978).

  • Page 45

    UW NPL Annual Report 1999-2000 39 4 Nuclear and Particle Astrophysics 4.1 Progress in the 7 Be(p,γ)8 B measurement E. G. Adelberger, A. R. Junghans, E. C. Mohrmann, K. A. Snover, T. D. Steiger, H. E. Swanson and TRIUMF Collaborators∗ The cross section of the radiative proton capture of 7 Be at solar energies is of utmost importance for understanding the flux of high-energy solar neutrinos, which can be detected by large neutrino experiments e.g. SNO and Super-K. Our measurement of the cross section of 7 Be(p,γ)8 B is based on detecting the α particles following the β-decay of 8 B. For a precise determination of the cross section we use a nearly homogeneous proton-beam flux over the whole target area and a precision measurement of the total number of 7 Be target atoms. This avoids the large uncertainties associated with an inhomogeneous areal density which enters in a conventional measurement. The target, mounted on one end of a rotating arm, is irradiated in the proton beam and then rotated 180 degrees in front of a Si detector to count the α particles from the 8 B decay. During the counting time the integrated beam flux through an aperture of 3mm diameter is measured with a Faraday cup. We have carried out several test experiments using radioactive 7 Be targets between 10 mCi and 27 mCi. The idea is to prove that every stage of the experiment has the required precision to reach our proposed final error limit on the cross section of ± 5%. We have improved our experimental setup by also integrating the beam current which strikes the target, thus compensating for differences between the beam that strikes the target and the beam that strikes the aperture. To accomplish this, the target arm has been electrically isolated from the chamber. The arm is connected to a precision current integrator (BIC) and is biased to +300 V by means of a battery to reduce secondary electron losses. To eliminate leakage currents, the water cooling for the target arm is operated by an electrically isolated closed-loop chiller (NESLAB) using deionized water with a deionizer in the circuit which is kept under nitrogen atmosphere. The whole cooling apparatus is biased to the same potential as the arm, resulting in essentially zero leakage current. The typical discrepancy between the beam integration on the Faraday cup and on the target arm is smaller than 1%. Of central importance for the principle of the measurement is the production of a homogeneous beam flux over an area larger than the target area (diameter 3mm). The beam is focused to approximately 1mm diameter on the target and then rastered by magnet coils which are driven by triangular waves of incommensurate frequencies (19.03 Hz in x and 41.00 Hz in y direction). The amplitudes determine the size of the swept area. To test the homogeneity of the beam flux we measure the beam current through apertures of 2mm, 3mm, and 4mm diameter. The ratios of the beam fluxes measured through these apertures should approach unity for swept areas larger than the apertures. With a deuteron beam of 770-keV energy we have demonstrated this homogeneity with a precision of ± 1%, shown in Fig. 4.1-1. The homogeneity of the beam flux depends on the stability of the beam current delivered by the accelerator. Random fluctuations in the beam current are reduced by averaging over the time ∗ L. Buchmann, S. Park, J. Vincent and A. Zyuzin, TRIUMF, Vancouver, BC, Canada.

  • Page 46

    40 Figure 4.1-1. Ratio of beam fluxes through different size apertures, measured with a deuteron beam (Ed = 770 keV). Figure 4.1-2. Yield per integrated beam of the reaction 7 Li(d,p)8 Li in a 7 Be target of 13 mCi. For the data marked 1 BIC the beam flux is measured only with the faraday cup, while for the data marked 2 BIC the beam intensity is measured on target. of the measurement. It is impractical to run with a completely homogeneous beam flux since this corresponds to very large sweep amplitude and hence very small beam on target. Thus one must have some knowledge of the target uniformity. The necessary information was obtained by measuring the ratio of the reaction yield to the integrated beam passing through the 3mm aperture for the 7 Li(d,p)8 Li reaction, which was done with the 7 Be target in the same manner as the 7 Be(p,γ)8 B measurement. This ratio should become constant once the beam sweep is large enough to cover both target and aperture. An example of such a measurement is shown in Fig. 4.1-2. The similarity of Fig. 4.1-2 and Fig. 4.1-1 in the plateau region demonstrates that the 7 Li (and hence

  • Page 47

    UW NPL Annual Report 1999-2000 41 Figure 4.1-3. Scan of a 7 Be target (13 mCi) integrated over the y-direction. presumably also the 7 Be) density distribution is reasonably uniform and concentrated in the central 3mm diameter region. In addition, we measure directly a scan of the density distribution. The data shown in Fig. 4.1-3 were obtained by sweeping the beam in the x direction and using the arm rotation to step through the target position in the y direction. These results demonstrate good alignment and localization of the target material. Similar scans with other 7 Be targets have shown a problem with 7 Be density tails extending to large radius, which is currently being worked on. The total number of 7 Be target atoms has been determined by measuring the absolute activity of the target mounted on the target arm using a precisely aligned, collimated Ge detector. This detector has been calibrated to ± 1% using calibration sources 54 Mn, 125 Sb, 133 Ba, 134 Cs, and 137 Cs from Isotope Products Corp. For 400-keV and 500-keV proton energy (c.m.) the α spectrum from 7 Be(p,γ)8 B has been measured with sufficient statistics to deduce cross sections. The lowest proton energy usable with our present setup of terminal ion source TIS and tandem accelerator is 300 keV. The α spectra show a cutoff due to noise at about 600-700 keV, as shown in Fig. 4.1-4. Noise from the motor controller and other sources has been shielded from the detectors or eliminated at the source. The small fraction of α particles below the cutoff will be determined by comparison of the measured spectra and the theoretical shape. The absolute energy of the proton beam and its reproducibility have been determined to ± 1 keV by repeated measurements of sharp, well-known resonances in 19 F(p,αγ)16 O at Ep = 484 keV and 872 keV, detecting the γ-rays with our large NaI spectrometer. The energy width of two 7 Be targets (13 mCi, 27 mCi) has been measured using a 7 Be(α, γ)11 C resonance at Eα =1.378 MeV again using the large NaI spectrometer, as shown in Fig. 4.1-5. The energy width of the resonance was found to be 6 keV and 10 keV respectively. The energy width of a pure 7 Be target with the given activity corresponds to about 20 percent of the measured values. This shows a consistent high purity of the metallic 7 Be targets.

  • Page 48

    42 Figure 4.1-4. Measured α-spectrum with a 13mCi 7 Be target. Figure 4.1-5. Measurement of the energy width for a 7 Be target (13 mCi). For the experimental setup we have constructed a more rigid Ti target arm and an optional target mount directly in the line of the beam to measure the amount of backscattered 8 B nuclei using a catcher plate mounted on the arm (see Sec. 4.2). When these modifications are in place we will be ready to make the final production runs for the absolute cross section determination.

  • Page 49

    UW NPL Annual Report 1999-2000 43 4.2 Measurement of 8 B backscatter for the 7 Be(p,γ)8 B experiment E. G. Adelberger, A. R. Junghans, E. C. Mohrmann, K. A. Snover, T. D. Steiger, H. E. Swanson and TRIUMF Collaborators∗ The 7 Be(p,γ)8 B cross section is being measured by bombarding a metallic 7 Be target with a proton beam and then moving the target to a counting position and detecting the α particles from the decay of 8 B. One important systematic concern in this measurement is the backscatter of 8 B atoms from the target during the bombardment phase. Any backscattered 8 B atoms will not be counted by the alpha detectors, resulting in a systematic underestimate of the 8 B yield. Calculations with the computer code SRIM show that the effect of backscattering increases as the target thickness goes down, the atomic weight (Z) of the target backing material goes up, or the proton energy goes down. The dependences on the target thickness and Z have been verified in measurements by Weissman et al.1 and Strieder et al.2 for the backscattering of 8 Li produced in LiF targets by the 7 Li(d,p)8 Li reaction. For 7 Be targets, which typically have substantial chemical impurities, the backscattering also depends on the Z of the impurities. Previous 7 Be(p,γ)8 B measurements have been potentially susceptible to backscatter because they employed high-Z Pt backings and had unknown chemical impurities in their targets.3,4 How- ever, these experiments did not address the issue of backscatter. A recent measurement (with two data points near Ep = 1 MeV) claims immunity from backscatter due to the nature of the implanted target used.5 Our experiment is expected to be moderately susceptible to backscatter. Though our Mo target backing has a significantly lower Z than the Pt backings used previously, our targets are thin due to their purity (∼ 20% 7 Be) and we are attempting to go to low proton energies. Hence we are undertaking the first ever direct measurement of 8 B backscattering. Because the backscattering depends on target composition and thickness, it is important to measure the backscattering for the same target(s) used in the cross section measurement. The backscatter measurement will be performed in the same chamber as the 7 Be(p,γ)8 B mea- surements. In this case, however, the 7 Be target will be mounted in a fixed position in place of the Faraday cup, and catcher plates will be mounted on both ends of the flipper arm to transport backscattered 8 B atoms to the α-counting position. During bombardment the proton beam will pass through the catcher plate (either through a thin foil or simply a small hole) and strike the fixed target. Both the target and the catcher plate will be electrically isolated to allow for beam integration. A special large-area alpha detector will be used, since the 8 B atoms will be spread over a sizeable area on the catcher plate. The apparatus required for this measurement is currently under construction. It is expected that preliminary tests of the procedure will be conducted before summer ’00, with complete 7 Be(p,γ)8 B and 8 B backscatter measurements to follow soon after. ∗ L. Buchmann, S. Park, J. Vincent and A. Zyuzin, TRIUMF, Vancouver, BC, Canada. 1 L. Weissman et al., Nuc. Phys. A 630, 678 (1998). 2 F. Strieder et al., Euro. Phys. J. A 3, 1 (1998). 3 B.W. Filippone et al., Phys. Rev. C 28, 2222 (1983). 4 F. Hammache et al., Phys. Rev. Lett. 80, 928 (1998). 5 M. Hass et al., Phys. Lett. B 462, 237 (1999).

  • Page 50

    44 4.3 WALTA: The Washington Large-area Time-coincidence Array J. G. Cramer, S. R. Elliott, B. Moorthy, C. Ouch, M. Roddy,∗ P. Schaffer,† J. F. Wilkerson, J. Wilkes† and E. Zager† Ultra-high energy cosmic ray particles that are protons with energies greater than ∼ 1020 eV cannot traverse great distances through the cosmos without losing energy to pion photoproduction in collisions with photons of the cosmic microwave background. The mean free path for this process is about 50 Mpc. Cosmic rays of similar energies that are electrons or gamma rays have similar energy loss processes with even shorter mean free paths. All known cosmic ray particles traveling cosmological distances should be limited to energies of less than about ∼ 1019 eV by such processes, yet a number of cosmic ray events with primary energies well above 1020 eV have been observed. This leads to the conclusion that the primary particles must have originated in our galactic neighborhood. However, the arrival direction of the few events seen so far does not indicate a nearby source. Furthermore there are few if any nearby astrophysical objects which could potentially accelerate a particle to these energies. This paradox is one of the major current mysteries in astrophysics. We have formed a collaboration for the development of a distributed detector network. This system will support measurements of air showers from ultra-high energy cosmic rays and can also support a broad class of other physical measurements. We call the overall network NNODE (North- west Network for Operation of Distributed Experiments), and we call the cosmic ray measurement component WALTA (WAshington Large-area Time-coincidence Array). The project is to be a direct physical science outreach program between faculty and students of the University of Wash- ington and the science teachers and students of Washington-area middle schools and high schools (grades 7-12). The WALTA1 part of the project is modeled on the ALTA initiative pioneered by the University of Alberta and currently being implemented in the Alberta Provincial School System. The local collaboration includes members of the Nuclear Physics Laboratory, physics department personnel in the cosmic ray and physics education groups, and a department of education faculty member from Seattle University. The project was described at the 1999 cosmic ray conference.2 Each WALTA/NNODE measurement module is envisioned to consist of a computer with an Internet connection, a GPS timing system, and measuring equipment. The measuring devices will be of several types. The WALTA modules will consist of scintillation paddles to be placed at the school to detect distributed particle showers produced at the top of the atmosphere by ultra-high energy (∼ 1020 eV) cosmic rays. Groups from three other scientific disciplines have expressed some interesting possibilities of adding additional measurement units to the NNODE network. This past year, we began the assembly of a prototype site for the cosmic ray detector. This was made possible by a DOE capital equipment grant we received to obtain the needed hardware for this development. We purchased but are awaiting the arrival of large scintillator modules and so have been testing detector configurations with smaller scintillator pieces available at NPL. We have bought GPS antennae and decoders and are implementing them. We have identified inexpensive, ∗ Education Department, Seattle University, Seattle, WA. † Physics Department, University of Washington, Seattle, WA 98195. 1 The WALTA web page can be found at http://www.phys.washington.edu/∼walta/. 2 E. Zager et al., Proceedings of the 26th International Cosmic Ray Conference, Salt Lake City, Utah, August 17-25, 1999, to be published.

  • View More

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