Triangle Universities Nuclear Laboratory

TUNL XLVIX

PROGRESS REPORT

1 SEPTEMBER 2009 – 31 AUGUST 2010

Triangle Universities Nuclear Laboratory

Duke University

North Carolina State University

University of North Carolina at Chapel Hill

Box 90308, Durham, North Carolina 27708-0308, USA

Work described in this Progress Report is supported by the United States Department of Energy,Office of High Energy and Nuclear Physics, under:

Grant No. DE-FG02-97ER41033 (Duke University),Grant No. DE-FG02-97ER41042 (North Carolina State University), andGrant No. DE-FG02-97ER41041 (University of North Carolina).

Contents

Introduction vii

Personnel xii

1 Fundamental Symmetries in the Nucleus 11.1 Time-Reversal Violation: The Neutron Electric Dipole Moment . . . . . . . . . . . 2

1.1.1 Search for the Neutron Electric Dipole Moment . . . . . . . . . . . . . . . . . 21.1.2 Search for the Neutron Electric Dipole Moment: Geometric Phase Effects . . 41.1.3 Development of the Cryogenic Design for the nEDM Experiment . . . . . . . 61.1.4 3He Spin Transport Simulation in the nEDM Experiment . . . . . . . . . . . 8

1.2 Fundamental Coupling Constants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101.2.1 The UCNA Experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101.2.2 High-precision Measurement of the 19Ne Lifetime . . . . . . . . . . . . . . . . 12

2 Neutrino Physics 152.1 ββ Decay Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2.1.1 The MAJORANA DEMONSTRATOR Neutrinoless Double-Beta Decay Experiment 162.1.2 A Tonne-Scale Ge Neutrinoless Double-Beta Decay Experiment . . . . . . . . 182.1.3 Project Engineering for the Majorana Demonstrator Project . . . . . . . 202.1.4 Copper Electroforming for the Majorana Demonstrator at the Sanford

Underground Laboratory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222.1.5 Deployment of the Majorana Low-Background Broad-Energy Germanium De-

tector at KURF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242.1.6 Initial Data from the Majorana Low-Background Broad-Energy Germanium

Detector at KURF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262.1.7 Majorana Demonstrator Module Design Activities at TUNL . . . . . . . 282.1.8 A Monte-Carlo Background Model for the MALBEK Detector . . . . . . . . 302.1.9 Creating a Muon Veto for MALBEK . . . . . . . . . . . . . . . . . . . . . . . 322.1.10 The SEGA Detector: A 1.4 kg, Segmented Germanium Ionization Detector

Enriched to 85% 76Ge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342.1.11 Differential Cross-Section Measurements for Elastic and Inelastic Scattering

of Neutrons from Neon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362.1.12 Double–Beta Decay of 150Nd to the First Excited Final State of 150Sm . . . . 382.1.13 Double–Beta Decay of 150Nd to Excited Final States of 150Sm . . . . . . . . 402.1.14 Background Sources Observed in Long-Duration Counting of an Enriched

Sample of 150Nd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422.2 Tritium β Decay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

2.2.1 The KArlsruhe TRItium Neutrino (KATRIN) Experiment . . . . . . . . . . . 442.2.2 Automated Penning Trap Searches in the KATRIN Experiment . . . . . . . . 462.2.3 BEM Techniques for Computing Electric Fields in the KATRIN Experiment 48

2.3 Dark Matter Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502.3.1 DEAP/CLEAN Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502.3.2 Searches for Direct Detection of Dark Matter with a Majorana Prototype

Detector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

iii

iv Contents TUNL XLIX 2009–10

3 Nuclear Astrophysics 553.1 Nucleosynthesis in Hydrostatic and Explosive Environments . . . . . . . . . . . . . . 56

3.1.1 Measurement of 17O(p,γ)18F Between the Narrow Resonances at Elabr = 193

and 519 keV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563.1.2 Resonance Strength in 22Ne(p,γ)23Na from Depth Profiling in Aluminum . . 583.1.3 Target Development for a Measurement of 23Na(p,γ)24Mg at Astrophysical

Energies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 603.1.4 Resonant Proton Capture on 23Na at Stellar Burning Temperatures . . . . . 623.1.5 The Effects of Thermonuclear-Reaction-Rate Variations on Nucleosynthesis in

Classical Novae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 643.1.6 Hydrodynamic Models of Type I X-Ray Bursts: Metallicity Effects . . . . . . 66

3.2 26Al in the Interstellar Medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 683.2.1 Theoretical Evaluation of the Reaction Rates of 26Al(n,p)26Mg and 26Al(n,α)23Na 683.2.2 The Effects of Thermonuclear Reaction-Rate Variations on 26Al Production

in Massive Stars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 703.3 Thermonuclear Reaction Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

3.3.1 STARLIB: A Next-Generation Reaction-Rate Library for Nuclear Astrophysics 72

4 Few-Nucleon Interaction Dynamics 754.1 Sub-Nucleonic Degrees of Freedom . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

4.1.1 Measurement of Single Target-Spin Asymmetry in Semi-Inclusive Pion Elec-troproduction on a Transversely Polarized 3He Target . . . . . . . . . . . . . 76

4.2 The A=2 Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 784.2.1 Final Neutron-Proton Analyzing Power Data at En=7.6 MeV . . . . . . . . . 78

4.3 The A=3 Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 804.3.1 Cross-Section Measurement of 2H(n,np)n at 16 and 19 MeV in Symmetric

Constant-Relative-Energy Configurations . . . . . . . . . . . . . . . . . . . . 804.3.2 Monte-Carlo Simulation of the Neutron-Deuteron Breakup Cross-Section Mea-

surements at 16.0 and 19.0 MeV . . . . . . . . . . . . . . . . . . . . . . . . . 824.4 Reaction Dynamics of Light Nuclei . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

4.4.1 Study of the 11B(p,α) Reaction Below Ep = 3.8 MeV . . . . . . . . . . . . . 844.4.2 The Study of the 11B(α,p) Reaction with an α-Particle Energy from 4.0 MeV

and 7.0 MeV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 864.4.3 11B(α,α) Elastic Cross-Section Measurements . . . . . . . . . . . . . . . . . . 884.4.4 Phase-Shift Analysis of 11B(α,α)11B from 2.0 to 5.4 MeV . . . . . . . . . . . 90

5 The Many-Nucleon Problem 935.1 Preequilibrium Nuclear Reactions and Random Matrix Theory . . . . . . . . . . . . 94

5.1.1 Random Matrices and Chaos in Nuclear Physics . . . . . . . . . . . . . . . . 945.1.2 Preequilbrium Reaction Phenomenology . . . . . . . . . . . . . . . . . . . . . 96

5.2 Neutron-Induced Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 985.2.1 Energy Dependance of the Fission-Product Yields in Fast and Thermal Neutron-

Induced Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 985.2.2 Test Measurements of the 235U(n, 2nγ)234U Reaction with PPAC-Based Fis-

sion Chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1005.2.3 Measurement of the 187Re(n,2nγ)186mRe Destruction Cross Section . . . . . 1025.2.4 Neutron-Induced Reaction Cross-Section Measurements on GaAs . . . . . . . 1045.2.5 Neutron Capture Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . 106

6 Photonuclear Reactions at HIγS 1096.1 Nuclear Atsrophysics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110

6.1.1 Measurements of the angular distributions of the 12C(γ, 3α) reaction at Eγ ≈9.8 MeV with the O-TPC detector. . . . . . . . . . . . . . . . . . . . . . . . . 110

6.1.2 Photoneutron Measurements of Astrophysically Important States in 26Mg . . 1126.1.3 A New Astrophysical Reaction Rate for α+α+n . . . . . . . . . . . . . . . . 114

6.2 γ-3He Interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1166.2.1 Measurements of the Absolute Cross Section of the Three-Body Photodisinte-

gration of 3He Between Eγ = 11.4 MeV and 14.7 MeV at HIγS . . . . . . . . 116

TUNL XLIX 2009–10 Contents v

6.2.2 Three-Body Photodisintegration of 3He: Comparison of Neutron Backgroundsfrom Different Glasses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

6.2.3 Two-Body Photodisintegration of 3He between 7 and 16 MeV . . . . . . . . . 1206.3 γ-4He Interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

6.3.1 The Photodisintegration Reaction 4He(γ, p)3H below 30 MeV . . . . . . . . . 1226.3.2 The Photodisintegration Reaction 4He(γ, n)3He below 30 MeV . . . . . . . . 124

6.4 Study of Many-Body Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1266.4.1 Compton@HIγS on 209Bi from Eγ = 15-30 MeV . . . . . . . . . . . . . . . . 1266.4.2 Fine and Gross Structure of the Pygmy Dipole Resonance . . . . . . . . . . . 1286.4.3 Dipole Excitations of 76Se in the Energy Range 4-9 MeV . . . . . . . . . . . 1306.4.4 Cross-Section Measurements of the 180mTa(γ,n) Reaction and the Photo-

Induced Depopulation of 180mTa Isomer . . . . . . . . . . . . . . . . . . . . . 1326.4.5 Dipole Strength Distribution in 232Th . . . . . . . . . . . . . . . . . . . . . . 1346.4.6 Implementation of Nuclear Resonance Fluorescence Support in GEANT4 . . 1366.4.7 Nuclear Self-Absorption Experiment on 11B . . . . . . . . . . . . . . . . . . . 137

7 Applied Research 1397.1 Homeland and National Nuclear Security . . . . . . . . . . . . . . . . . . . . . . . . 140

7.1.1 Photoneutron Yield Ratios of (�γ,n) Reactions in the GDR Region . . . . . . 1407.1.2 Photofission Neutron Yield Ratios on 238U near Eγ = 6.2 MeV using Linearly

Polarized γ rays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1427.1.3 Measurement of γ-Ray Induced Activation of 235U . . . . . . . . . . . . . . . 1447.1.4 Linearly Polarized Beam Induced Photo-Neutron Yield Ratios in 9Be between

10.5 and 15.5 MeV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1467.1.5 Polarized (γ,n) reaction studies of natCd, natSn, and 181Ta . . . . . . . . . . 1487.1.6 Simulation of Background from Neutron Detector Array and from (γ, n) on

14N in the Target Room . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1507.1.7 New Electric and Magnetic Dipole Transitions in 238U . . . . . . . . . . . . 152

7.2 Pyroelectric Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1547.2.1 Neutron Production with a Pyroelectric Double-Crystal Assembly Without

Nano-Tip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1547.3 Nuclear Data Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

7.3.1 Nuclear Data Evaluation Activities . . . . . . . . . . . . . . . . . . . . . . . . 156

8 Accelerator Physics 1598.1 The High Intensity Gamma Ray Source (HIγS) . . . . . . . . . . . . . . . . . . . . . 160

8.1.1 Developing New and Enhanced Gamma-ray Capabilities for HIGS Research:I : Eγ = 60 MeV Operation and Helicity Switch . . . . . . . . . . . . . . . . 160

8.1.2 Developing New and Enhanced Gamma-ray Capabilities for HIGS Research:II: Mirror Perfromance Monitoring . . . . . . . . . . . . . . . . . . . . . . . . 162

8.1.3 Accelerator Physics Program: I : Free-Electron Laser Research . . . . . . . . 1648.1.4 Accelerator Physics Program: II: Beam Instabilities Research and Fast Feed-

back Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1668.1.5 Gamma Imagers for HIGS User Experiments . . . . . . . . . . . . . . . . . . 1688.1.6 HIGS Accelerator and Gamma-ray Beam Operations . . . . . . . . . . . . . . 170

8.2 The FN Tandem Accelerator and Ion Sources . . . . . . . . . . . . . . . . . . . . . . 1728.2.1 Tandem Accelerator Operation . . . . . . . . . . . . . . . . . . . . . . . . . . 172

8.3 The LENA Accelerator and Ion Sources . . . . . . . . . . . . . . . . . . . . . . . . . 1748.3.1 Status of the LENA ECR Accelerator . . . . . . . . . . . . . . . . . . . . . . 174

9 Nuclear Instrumentation and Methods 1779.1 Targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178

9.1.1 A New 3He-Target Design for Compton Scattering Experiments . . . . . . . . 1789.1.2 Scintillating Target for d(�γ, γ)d at HIγS . . . . . . . . . . . . . . . . . . . . . 1809.1.3 Status of the HIγS Frozen-Spin-Target Transverse Coil at TUNL . . . . . . . 182

9.2 Detector Development and Characterization . . . . . . . . . . . . . . . . . . . . . . . 1849.2.1 A Detector Array for Photoneutron Studies . . . . . . . . . . . . . . . . . . . 1849.2.2 BrilLanCe Detector Energy Resolution Characterization at HIγS . . . . . . . 186

vi Contents TUNL XLIX 2009–10

9.2.3 Characterization of a Neutron Counter for High-Precision (γ,n) Cross-SectionMeasurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188

9.2.4 Determination of the Absolute γ-ray Detection Efficiency of the Molly detec-tor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190

9.2.5 Characterization of a Gd-Loaded Neutron Detector . . . . . . . . . . . . . . . 1929.2.6 Ratio of Ge Detector Peak Efficiencies at Energies of 4.4 and 11.7 MeV . . . 1949.2.7 Initial Studies of HPGe Detectors for Nuclear Astrophysics . . . . . . . . . . 1969.2.8 Upgrade and Assembly of the APEX Trigger Detector . . . . . . . . . . . . . 1989.2.9 Rutherford-Backscattering Analysis of Polymer Membranes Used for Water

Purification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2009.3 Facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202

9.3.1 Development of an Ultracold Neutron Source at the NC State PULSTARReactor Facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202

9.3.2 Systematic Studies for the nEDM Experiment at the PULSTAR UCN Facility 2049.3.3 Development of the Fundamental Neutron Physics Beamline at the SNS . . . 2069.3.4 Status of the KURF Low-Background Counting Facility . . . . . . . . . . . . 2089.3.5 Unfolding Full-Energy Peaks in γ-ray Spectra . . . . . . . . . . . . . . . . . . 210

9.4 Data Acquisition Hardware and Software Development . . . . . . . . . . . . . . . . . 2129.4.1 Status of the orca Data Acquisition System . . . . . . . . . . . . . . . . . . 2129.4.2 iorca, An iPad-based DAQ monitoring Tool . . . . . . . . . . . . . . . . . . 2149.4.3 Validation and Characterization of the KATRIN DAQ Electronics System . . 2169.4.4 Installation and Commissioning of the MALBEK Low-Background BEGe De-

tector Data Acquisition System . . . . . . . . . . . . . . . . . . . . . . . . . . 2189.4.5 Characterization of DAQ Hardware for the HALO Experiment . . . . . . . . 220

A Appendices 223A.1 Graduate Degrees Awarded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224A.2 Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226A.3 Invited Talks, Seminars, and Colloquia . . . . . . . . . . . . . . . . . . . . . . . . . . 238A.4 Professional Service Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245

Index 248

Introduction

The Triangle Universities Nuclear Laboratory (TUNL) is a center for low-energy nuclear physicsthat is staffed by faculty from three consortium universities: Duke University, North Carolina StateUniversity, and the University of North Carolina at Chapel Hill. TUNL faculty supervise about 50graduate students conducting Ph.D. thesis projects on a wide variety of topics that include nuclearastrophysics, fundamental symmetries, weak interactions, few-nucleon and sub-nucleon systems,many-body nuclear systems, neutrino physics, and applications of nuclear physics. This documentis a compilation of status reports of research projects conducted by TUNL research groups duringthe period of September 1, 2009 to August 31, 2010. It covers the second half of year-1 (March1, 2009 - February 28, 2010) and first halve of year-2 (March 1, 2010 - February 28, 2011) of thethree-year grant (March 1, 2009 - February 29, 2012) to the three collaborating universities fromthe Low-Energy Program in the Office of Nuclear Physics at the U.S. Department of Energy.

During this reporting period research groups at TUNL published 47 papers in refereed journalsof which 5 were letter articles. Consortium members delivered 66 invited talks at APS meetings,conferences workshops and department colloquia and seminars. In addition, the consortium groupspublished 18 papers in conference proceedings and gave 40 contributed talks at APS meetings andconferences. About 35% of the journal papers and 30% of the invited talks were based on workdone at the on-site accelerator and research facilities. Highlights of research accomplishments madeduring the reporting period are given below.

• New data from HIγS reveal dominant electric nature of pygmy dipole resonance:A common feature in medium- and heavy-mass nuclei is the prevalence of neutron excess andthe resulting formation of a neutron skin. A consequence of the neutron skin is a collectivevibrational mode of motion at low excitation energies. This excitation mode is dipole in natureand is referred to as the pygmy dipole resonance (PDR) because its small strength relative tothe dominate giant dipole resonance (GDR) which occurs at higher excitation energies. ThePDR is often modeled as the vibration of the neutron skin against the core of the nucleus whichis composed of equal numbers of neutrons and protons. The PDR is of practical importancein nuclear astrophysics because of its impact on the reaction landscape at energies relevantto nuclear synthesis in the stellar medium. For example, the existence of this low-energyexcitation mode influences (n, γ) reaction rates relevant to the rapid neutron capture processin stars.

Using nuclear resonance fluorescence at HIγS we have verified for the first time in N = 82isotones that the observed pygmy dipole strength from 4 MeV to the particle separation energyis predominantly electric in nature. Interpretation of our data with a quasiparticle phononmodel indicates that there are two components to the PDR: the low-lying group of 1− statesare due to the excitation of the neutron-skin and the higher-lying 1+ states that belong toa transitional region on the tail of the GDR. For details see the article by Tonchev et al. inSect. 6.4.2.

• Two-body photodisintegration of 3He between Eγ=7 and 16 MeV:We have completed the most accurate measurements of the total cross-section of the 3He(γ,p)2Hreaction in the energy range from 7 to 16 MeV at HIγS. These data confirm the reported dis-crepancy between three-nucleon calculations and measurements. However, these new datafrom HIγS are closer to theory than almost all of the data obtained during the last half cen-tury for this classical three-nucleon reaction. For details see the article by W. Tornow et al.in Sect. 6.2.3.

viii Introduction TUNL XLIX 2009–10

• Space-star anomaly confirmed in new neutron-deuteron breakup measurements:Our recent cross-section measurements for the 2H(n, np)n reaction at 16 MeV incident neutronenergy confirm earlier results for the 2H(n, nn)p reaction in the Space Star configuration atneutron beam energies below 25 MeV. These new data along with reported results show thattheoretical calculations based on all modern nucleon-nucleon potential models under-predictthis cross section by about 30%. For details see the article by A. H. Couture et al. in 4.3.1.

• First Compton scattering measurements with HINDA: A new method for observ-ing the Isovector Giant Quadrupole Resonance (IVGQR) in nuclei:Six of the eight NaI detectors in the HIγS NaI Detector Array (HINIDA) were used to measurethe Compton scattering cross section from 209Bi in the energy range from 15 to 30 MeV. Thesedata provide the first high precision determination of the energy, width, and strength of theIVGQR. The IVGQR is a general feature of the collective modes of motion of nuclei. However,because of the lack of accurate measurements a detailed theorectical understanding has notbeen developed. The combination of the 100% polarized beams at HIγS and the HINDA arrayalong with the realization that the E1-E2 interference term observed using polarized γ rayschanges sign when going from a forward to a backward angle has lead to an order of magnitudeimprovement in the values of the parameters which describe the IVGQR. The collaborationplans to prepare a proposal to study the IVGQR via Compton scattering as a function of massnumber (A) to submit to the 2011 HIγS PAC. For further details refer to the article in thisreport by S. Henshaw et al. in Sect. 6.4.1.

• First photodisintegration measurements on radioactive targets at HIγS:The tunable monoenergetic γ-ray beam and high resolution detection systems at HIγS enableprecision cross-section measurements of photonuclear reactions on targets having masses of onlya few hundred micrograms. This capability creates opportunities to study nuclear structure ofradioactive nuclei, which are typically available in sparse amounts, and to make measurementsimportant for national security and energy research. The first photodisintegration cross-sectionmeasurements on a radioactive target at HIγS were made this year on 241Am. We performedthe highest accuracy (γ,n) cross-section measurements on 241Am in the energy range from 9 to16 MeV. Unlike the common observation for strongly deformed nuclei, the (γ,n) cross sectionfor 241Am shows only a single Lorentzian peak at around 12 MeV. This feature is due to theopening of a strong second-chance fission decay channel (γ,nf) around 12 MeV that inhibitsthe (γ,n) channel. For more details see A. P. Tonchev et al., Phys. Rev. C, 82, 054620 (2010).

• New evaluation of thermonuclear reaction rates:Evaluated thermonuclear reaction rates, which are the foundation for any successful stellarmodel, were first presented by Willy Fowler and collaborators (until 1988) and later updatedby the large European NACRE collaboration (in 1999). TUNL has led the way in completing anew rate evaluation effort that is based on an entirely new method (a Monte Carlo procedure).Our method provides for the first time realistic rate uncertainties and rate probability densityfunctions. This technique opens a window of opportunity for new kinds of nucleosynthesisstudies that were previously unattainable. This comprehensive work has been published as aseries of four papers in a ”Topical Issue”, see Nucl. Phys. A 841 (2010). Applications of thesenew evaluations are presented in Chapter 3 of this report.

• MAJORANA Demonstrator status updates:

– The temporary cleanroom at UNC passed safety inspection and was put into service inJanuary 2010.

– The low background counting facility (MALBEK) was put into operation at the Kimball-ton Underground Research Facility in Kimballton, VA in January 2010.

– The MAJORANA Demonstrator (MJD) project passed CD-1 review in May 2010.

– Operation of a P-type point contact detector with low-mass cable and custom front-endelectronics was demonstrated in July 2010. The operational characteristics of this detectorillustrate many of the main design features of the concepts in the MJD detector modules.

– An underground copper electroforming facility was put into operation at DUSEL.

TUNL XLIX 2009–10 Introduction ix

• Upstream target room at HIγS:The new upstream target room (UTR) at HIγS was completed and put into service duringthe first quarter of calendar year 2010. The UTR is used to run experiments that require onlymoderate amounts of floor space. Examples of experiments that run in the UTR include nuclearfluorescence resonance measurements and photoneutron reaction measurements with compactneutron detector arrangements. Experiments with large equipment setups are run in the maintarget room, which is referred to as the Gamma Vault (GV). The photofission, Compton-scattering, 16O(γ,α) with the OTPC, nucleon polarizabilities and the GDH experiments willrun in the GV. Having two target rooms greatly enhances the efficiency of running experimentsat HIγS by providing the capability of running an experiment in the UTR while another oneis being set up in the GV.

• New ECR Source at LENA:We have completed construction and commissioning of the new ECR accelerator for LENA.The maximum proton beam current delivered to target is 1.2 mA for Ep = 150 – 200 keV,which exceeds the design goal of 1 mA. The source is currently being used for a cross-sectionmeasurement of the 23Na(p,γ)24Mg reaction. For details see the articles by J.M. Cesaratto etal. in Sections 3.1.3 and 8.3.1.

• 2H(d, n)3He neutrons produced with a pyroelectric two-crystal assembly:For the first time neutrons have been produced during the heating and cooling phase of apyroelectric crystal assembly. Also for the first time, a substantial neutron yield has beenobtained without an electric field enhancing nano-tip that is normally attached to one of thetwo crystals. For details see the article by W. Tornow et al. in Sect. 7.2.1.

Groups at TUNL are conducting research in all three physics frontier areas identified in the mostrecent Long Range Plan of the DOE/NSF Nuclear Science Advisory Committee. In addition, theiractivities include applications of nuclear physics, instrumentation R&D and nuclear data evaluationand dissemination. The work in each area is summarized below.

NUCLEAR PHYSICS RESEARCH

1. Hadron StructureThe main activities in this area are:

• nucleon EM and spin polarizabilities

• the GDH sum rule on the deuteron and 3He

The primary off-site facilities used for this work are Jefferson Laboratory and MAX-lab (Swe-den).

2. Nuclei: From Structure to Exploring StarsThere are three major efforts at TUNL in this area. Each effort is summarized below.

(a) Few-nucleon systemsThe main aim of this work is to test and refine few-nucleon calculations and theoreti-cal description of the nuclear force and its currents used in such calculations. Specificexperiments address:

• the role of three-nucleon forces in the 3N and 4N continuum using hadronic andelectromagnetic probes

• the analyzing power puzzle in 3N and 4N scattering• light-nucleus reaction dynamics

(b) Many-body physicsThe emphasis of this research is on:

• descriptions of nuclear reaction data with random matrix theory• the study of collective response of nuclei using polarized γrays at HIγS

x Introduction TUNL XLIX 2009–10

• measurements of (n, 2n) cross sections on actinide nuclei• the excitation of isomeric states• the study of preequilibrium nuclear reactions

The main off-site facility used by TUNL research groups for this research is the LosAlamos Neutron Science Center (LANSCE).

(c) Nuclear astrophysicsThe focus of this research is on measurements which are important for solar-neutrinophysics, stellar evolution and nucleosynthesis. Specific experiments address:• astrophysical S-factors of (p, γ) reactions• the explosive nucleosynthesis in novae• the evolution of massive stars• reaction rates important for the r, s and p processes

The main off-site facilities used by TUNL research groups are Argonne National Labora-tory (ANL) and Oak Ridge National Laboratory (ORNL).

3. In Search of the New Standard ModelThe two main research activities in this frontier are summarized below.

(a) Fundamental-symmetry studiesThe experiments in this area include:• measurement of time-reversal invariance (neutron EDM)• non-unitarity tests of the CKM matrix• measurement of the neutron lifetime

Off-site facilities used for this research include the Spallation Neutron Source (SNS), theKernfysisch Versneller Instituut (KVI), LANSCE, and the National Institute of Standardsand Technology (NIST).

(b) Neutrino physicsThe main activities in this research program include:• neutrinoless ββ decay search using 76Ge (Majorana Collaboration)• geo-anti-neutrino and solar neutrino studies at KamLAND• measurements of data needed to correct the neutrinoless ββ decay and dark-matter

detector measurements for neutron-induced backgrounds• the study of double-beta decay to excited 0+ states• neutrino mass determination by 3H β decay (KATRIN Collaboration)

Off-site facilities used for this research include ORNL, the Deep Underground Scienceand Engineering Laboratory (DUSEL), the Kimballton Underground Research Facility(KURF), the Waste Isolation Pilot Plant (WIPP), and the Karlsruhe Institute of Tech-nology (KIT).

INSTRUMENTATION R&D

Developments in technology and instrumentation are vital to our research and training program.Our current instrumentation development activities include:

• instrumentation for characterizing the γ-ray beam at HIγS

• polarized target development

• detector and scattering chamber development

• polarimeters for charged particles, neutrons and γ-rays

BROADER IMPACT ACTIVITIES

In addition to the above research in the frontier areas of nuclear physics, we run a nucleardata program and are conducting research important for applications of nuclear physics in nationalnuclear and homeland security, biology and medicine.

TUNL XLIX 2009–10 Introduction xi

1. Nuclear Data Program

• nuclear data evaluation for A = 3 − 20 for which TUNL is the international center

• web dissemination of nuclear structure information for A = 3 − 20

2. ApplicationsThe focus of these activities includes:

• gamma-ray induced reactions for homeland security and national nuclear security

• neutron-induced reactions for national nuclear security

• plant physiology using accelerator produced 11CO2 and 13NO3

• nuclear energy

EDUCATION

These activities include:

1. Research Experience for UndergraduatesTUNL runs a NSF-supported Research Experience for Undergraduates (REU) program innuclear physics. This year ten students participicated in this 10-week summer program; eightof them were supported by the NSF REU grant and the other two by grants to TUNL faculty.

2. TUNL Seminar ProgramThe TUNL seminar program continues with characteristic vigor (22 invited speakers). Thisseries is augmented by the TUNL Informal Lunch Talks (TILT) where graduate students andpostdocs present their research projects, the TUNL Astrophysics Journal Club and a speciallecture series given by local speakers on Advances in Physics as part of the REU programduring the summer. A related talk series, the Triangle Nuclear Theory Colloquia, is alsobeneficial to TUNL faculty and students.

The success of our research program is largely attributable to the talents and enthusiasm of thefaculty members (17 tenured/tenure track), post-doctoral associates and research staff and graduatestudents from the three Triangle universities. Also essential to our research program are long-term collaborations with research groups from: Gettysburg College, North Carolina A&T StateUniversity, North Carolina Central University, North Georgia College and State University, PennState Altoona, Tennessee Technological University, University of Conneticut at Avery Point, All-Russian Research Institute of Technical Physics (Snezhinsk), Istituto Nazionale di Fisica Nucleare(Pisa), Jagiellonian University (Cracow), Joint Institute for Nuclear Research (Dubna), TechnischeHochschule Darmstadt, University of Cologne, and University of Mainz. In our applications andinterdisciplinary research program we colloborate with scientists from several national laboratories,including Lawrence Livermore National Laboratory, Thomas Jefferson National Accelerator Facility,and Los Alamos National Laboratory and with scientists from private companies, such as, theAccelerator Driven Neutron Applications Corporation, the Physical Optics Corporation, and TriAlpha Energy, Inc.

Personnel

Department of Physics, Box 90308, Duke University,Durham, NC 27708-0308Department of Physics, Box 8202, North Carolina State University,Raleigh, NC 27695-8202Department of Physics and Astronomy, University of North Carolina,Chapel Hill, NC 27599-3255

FacultyAhmed, M. W. (Assistant Research Professor) DukeBilpuch, E. G. (Professor Emeritus) DukeChampagne, A. E. (Professor) UNCClegg, T. B. (Professor) UNCGao, H. (Professor) DukeGolub, R. (Professor) NCSUGould, C. R. (Professor) NCSUHaase, D. G. (Professor) NCSUHenning, R. (Assistant Professor) UNCHowell, C. R. (Director, Professor) DukeHuffman, P. R. (Associate Director, Professor) NCSUIliadis, C. (Professor) UNCKarwowski, H. J. (Professor) UNCKelley, J. H. (Research Associate Professor) NCSULudwig, E. J. (Professor Emeritus) UNCMerzbacher, E. (Professor Emeritus) UNCMitchell, G. E. (Professor) NCSURoberson, N. R. (Professor Emeritus) DukeTilley, D. R. (Professor Emeritus) NCSUTonchev, A. P. (Assistant Research Professor) DukeTornow, W. (Professor) DukeWalter, R. L. (Professor Emeritus) DukeWeller, H. R. (Associate Director for Nuclear Physics at HIγS, Professor) DukeWilkerson, J. F. (Associate Director, Professor) UNCWu, Y. K. (Associate Director for Light Sources, Associate Professor) DukeYoung, A. R. (Professor) NCSU

TUNL XLIX 2009–10 Personnel xiii

TUNL Advisory Committee

Balamuth, D. P. University of PennsylvaniaBalantekin, A. B. University of WisconsinBeise, E. J. University of MarylandFriar, J. L. Los Alamos National LaboratoryGarvey, G. T. Los Alamos National LaboratoryVigdor, S. E. Brookhaven National LaboratoryWiescher, M. C. University of Notre Dame

HIγS Program Advisory Committee

Garvey, G. T. (Chair) Los Alamos National LaboratoryBeise, E. J. University of MarylandGriesshammer, H. George Washington UniversityRehm, K. E. Argonne National LaboratorySherrill, B. M. National Superconducting Cyclotron LaboratoryChen, J. P. Thomas Jefferson National Accelerator LaboratoryDonnelly, T. W. Massachusettes Institute of Technology

Associated Faculty

Crawford, B. E. Gettysburg CollegeCrowe, B. J. North Carolina Central UniversityDutta, D. Mississippi State UniversityEngel, J. University of North CarolinaGai, M. University of Connecticut at Avery PointKeeter, K. J. Idaho State UniversityKorobkina, E. North Carolina State UniversityMarkoff, D. M. North Carolina Central UniversityMcLaughlin, G. C. North Carolina State UniversityNorum, B. University of VirginiaPedroni, R. S. North Carolina A&T State UniversityPrior, R. M. North Georgia College and State UniversityPurcell, J.E. Georgia State UniversityShriner, J. F. Tennessee Technological UniversitySpraker, M. C. North Georgia College and State UniversityStephenson, S. L. Gettysburg CollegeWeisel, G. J. Penn State Altoona

xiv Personnel TUNL XLIX 2009–10

Research StaffAdekola, A. S.1 (Research Associate) UNCAkashi-Ronquest, M.1 (Research Associate) UNCBaramsai, B.2 (Research Associate) NCSUBawitlung, L.3 (Research Associate) DukeBrown, N. J. (Research Associate) DukeCrowell, A. S. (Research Scientist) DukeDashdorj, D.2 (Research Associate) NCSUGreen, M.4 (Research Associate) UNCHowe, M. A. (Senior Research Scientist) UNCKalbach Walker, C. (Senior Research Scientist) DukeKwan, E.1,2 (Research Associate) DukeMalace, S. (Research Scientist) DukeMehta, V.5 (Research Associate, part time) NCSUOginni, B. M.6 (Research Associate) UNCPhillips, D. (Research Associate) UNCQiang, Y. (Research Associate) DukeRaut, R. (Research Associate) DukeRusev, G. (Research Associate) DukeSeo, P.-N. (Research Associate) Duke/UVAStave, S. (Research Scientist) Duke/NGC&SUTajima, S. (Visiting Research Scholar) Duke/NCCUYe, Q. (Research Scientist) Duke

Accelerator Physics Staff for Operations

Li, J. Research Scientist (at HIγS)Mikhailov, S. F. Senior Research Scientist (at HIγS)Popov, V. G. Senior Research Scientist (at HIγS)Westerfeldt, C. R. Senior Research Scientist (LE accel.) & Rad. Safety Manager

Technical and Nuclear Data Support Staff

Addison, J. H. Lab Apparatus DesignerBusch, M. Mechanical EngineerCarlin, B. P. Electronics SupervisorCarter, E. P.7 Accelerator TechnicianDunham, J. D. Accelerator TechnicianEmamian, M. Mechanical Engineer (at HIγS)Faircloth, J. Project and Building Maint. Coordinator (at HIγS)Johnson, M. Vacuum Technician (at HIγS)Mulkey, P. H. Senior Electronics TechnicianOakley, O.7 Electronics Technician (at HIγS)O’Quinn, R. M. Accelerator SupervisorPentico, M. Accelerator Technician (at HIγS)Rathbone, V. Accelerator Technician (at HIγS)Sheu, C.G. Project Coordinator, Nuclear DataSwift, G.8 Vacuum Engineer (at HIγS)Wallace, P.8 Electrical Engineer and Safety Manager (at HIγS)Wang, P. RF Engineer (at HIγS)

1Departed 06/102Supported by NNSA/DOE3As of 5/104As of 7/105Supported by NSF/DOE6As of 3/107Part-time8Supported by DHS

TUNL XLIX 2009–10 Personnel xv

Graduate Students

Agbeve, K. A. NCCU Kidd, M.F.1 DukeArnold, C. W. UNC Laskaris, G. DukeBaramsai, B.1,2 NCSU Leviner, L. NCSUBertone, P.1 UNC Lin, L. UNCBroussard, L. J. Duke Longland, R.1 UNCBuckner, M. UNC MacMullin, S.3 UNCCesaratto, J. M. UNC Martel, P. P.4 U. MassChen, W.1 Duke Medlin, G.3 NCSUChyzh, A.1,2 NCSU Meyer, E. UNCCombs, D. NCSU Minges, E. UNCCorona, T. J. UNC Mueller, J. M. DukeCottrell, C.3,5 NCSU Newton, J. R.1 UNCCouture, A. H. UNC O’Shaugnessy, C.1,3 NCSUCumberbatch, L. C. Duke Palmquist, G. NCSUDaigle, S. UNC Pattie, R. W.,Jr.3 NCSUDavis, B. J. NCCU Perdue, B. A. DukeDownen, L. UNC Qian, X.1 DukeEsterline, J. H. Duke Rich, G. UNCFallin, B. A. Duke Schelhammer,K.3 NCSUFinch, S. Duke Strain, J. UNCFinnerty, P. UNC Sun, C.1 DukeGooden, M. E. NCSU Swank, C. M. NCSUGiovanetti, G. K. UNC Swindell, A. G. U. ConnHammond, S. L. UNC Ticehurst, D. R. UNCHarrell, A. NC A&T Tompkins, J. R. UNCHenshaw, S. S. Duke VornDick, B.3 NCSUHolley, A. T.3 NCSU Walker, C.2 NCSUHuang, M. Duke Wu, W. Z. DukeHuffer, C.3 NCSU Ye, Q. J. DukeJia, B. Duke Zheng, W. Z. DukeKelley, J. A. UNC Zimmerman, W. U. ConnKendellen, D. P. NCSU Zong,X.1 Duke

Visiting Scientists

Delaroche, J.-P. 6/10 - 7/10 CEA DAM, Bruyeres-le-Chatel, FranceKriticka, M. 5/10 - 6/10 Charles University, Prague, Czech RepublicWita�la, H. 7/10 - 8/10 Jagiellonian University, Cracow, Poland

Administrative Support Personnel

Pulis, T Temporary StaffStrope, J.6 Staff AssistantWest, B. Staff AssistantKengla, B. Temporary Staff

1Graduated with Ph.D. degree between 9/09 and 8/102Supported by NNSA/DOE3Supported by NSF/DOE4Departed 02/105Graduated with M.S. degree between 9/09 and 8/106Partially supported by the Duke University Physics Department

xvi Personnel TUNL XLIX 2009–10

Undergraduates

Student Institution Faculty Advisor

Adamek, E. NCSU A. R. YoungAttayek, P. UNC T. B. CleggBaird, R. Duke Y. K. WuBancroft, H. Duke Y. K. WuBieff, G.1 George Washington Univ. R. Plesser (Duke)Brown, M.1 Morehead State Univ. R. HenningCasarella, C.1 NGSCU J. H. KelleyCollins, A. F. UNC H. J. KarwowskiCorse, W. Duke W. TornowDoehrmann, J.1 Bethel Univ. H. GaoEvans, L. UNC R. HenningFox, J. Duke W. TornowGrothe, T. Duke Y. K. WuHardin, J. UNC H. J. KarwowskiHill, A. UNC H. J. KarwowskiHolmes, R. UNC R. HenningJones, M. D. UNC H. J. KarwowskiKaleko, D. UNC R. HenningLeandre, V.1 NC A&T C. R. HowellLozier, J.1 Yale K. Scholberg (Duke)Marley, D. NCSU P. R. HuffmanMimms, S. NCSU A. R. YoungNash, L. UNC H. J. KarwowskiPace, G. UNC C. IliadisPalomares, A. NCCU C. R. HowellPedersen, B. UNC H. J. KarwowskiPerkins, W. UNC J. F. WilkersonPogrebnyak, I. UNC C. IliadisPratt, S.1 U. of Rochester G. RusevRyan, B. UNC H. J. KarwowskiSchmookler, B.1 NCSU A. R. Young/P. R. HuffmanSmith, W. C. George Washington Univ. G. FeldmanStevens, A. UNC R. HenningWalker, J. NCCU/UNC D. M. Markoff/H. J. Karwowski

1Supported by the TUNL NSF REU Program

Fundamental Symmetries in theNucleus

Chapter 1

• Time-Reversal Violation: The Neutron Electric DipoleMoment

• Fundamental Coupling Constants

2 Fundamental Symmetries in the Nucleus TUNL XLIX 2009–10

1.1 Time-Reversal Violation: The Neutron Electric DipoleMoment

1.1.1 Search for the Neutron Electric Dipole Moment

M.W. Ahmed, M. Busch, H. Gao, R. Golub, C.R. Gould, D.G. Haase, P.R. Huffman, D.

Kendellen, E. Korobkina, C.M. Swank, Q. Ye, A.R. Young, Y. Zhang, W.Z. Zheng,TUNL; the nEDM collaboration

TUNL plays a major role in an experimental program to develop a new technique to search

for the neutron electric dipole moment (EDM). This technique offers an improvement in

sensitivity of up to two orders of magnitude over existing measurements. A more precise

value for this moment has the potential to challenge calculations that propose extensions

to the Standard Model. During the past year, researchers at TUNL have constructed and

commissioned an apparatus for studying the geometric phase effect; developed a prototype

slow controls EPICS system; finalized the design of the 4He refrigeration system, dilution

refrigeration system, and cryovessel; simulated the spin transport of polarized 3He into the

apparatus; and are in the process of developing an apparatus to study systematic effects using

polarized 3He and ultracold neutrons.

The experimental search for a neutron electricdipole moment has the potential to reveal newsources of time reversal (T) and charge conserva-tion and parity (CP) violation and to challengecalculations that propose extensions to the Stan-dard Model. The goal of the current experimentis to improve the measurement sensitivity of theneutron EDM by one to two orders of magni-tude. Achievement of this objective will impactour understanding of the physics of both weakand strong interactions. The physics goals of thisexperiment are thus timely and of unquestionedimportance.

The experiment is based on the magnetic-resonance technique of rotating a magnetic dipolein a magnetic field. Polarized neutrons and po-larized 3He atoms coexist in a bath of superfluid4He at a temperature of ∼ 450 mK. When placedin an external magnetic field, both the neutronand 3He magnetic dipoles precess in the planeperpendicular to the magnetic field. The mea-surement of the neutron EDM involves measur-ing the difference in the precession frequenciesof the neutrons and the 3He atoms in a strongelectric field parallel or anti-parallel to the mag-netic field. In this comparison measurement, theneutral 3He atom is assumed to have a negligibleelectric dipole moment.

In principle, this new type of nEDM exper-iment in conjunction with the Spallation Neu-

tron Source (SNS) can achieve more than twoorders of magnitude improvement in the exper-imental limit for the neutron EDM. This factorresults from the possibility of an increased elec-tric field (a factor of ∼ 5) due to the excellent di-electric properties of superfluid 4He, an increasein the total number of ultracold neutrons (UCNs)stored (∼ 100 fold improvement that leads to fac-tor of 10 gain in sensitivity) and an increasedstorage time (∼ 5 times, which produces a fac-tor of 2 in sensitivity). The current experimen-tal EDM bound, however, is limited by magneticfield systematics. With the proposed experiment,an EDM limit of 10−28 e·cm is possible; the use of3He as a volume comagnetometer is crucial to theminimization of the magnetic-field systematics.

TUNL plays an active role in all aspects ofthis experiment. Crucial to the initial develop-ment of this project is the research and devel-opment work of key components of the overallexperiment. Researchers at TUNL are presentlyaddressing a number of outstanding issues re-lated to maximizing the sensitivity of the mea-surement. These contributions include measur-ing the lifetime of polarized 3He in a ∼ 450 mKdeuterated TPB/polystyrene-coated acrylic cellfilled with superfluid 4He; testing the efficiencyat which the 3He injector system can infuse po-larized 3He into a bath of ∼ 450 mK liquid he-lium, including spin transport; investigating the

TUNL XLIX 2009–10 Fundamental Symmetries in the Nucleus 3

use of the heat flush technique to transport thepolarized 3He from the collection volume to themeasurement cell and, afterwards, the depolar-ized 3He from the measurement cell to the pu-rifier; determining a method by which one canremove the unpolarized 3He from the liquid 4He;theoretical and experimental validation of the be-havior of the geometric phase effect in our exper-imental geometries; development of a prototypeEPICS slow controls system; and development ofan apparatus that will enable tests using a singlefull-size measurement cell (without an E field) ofkey systematic effects.

As the experiment proceeds toward the CD-2/3A milestone, TUNL researchers are design-ing the 4He refrigeration system, dilution refrig-eration system, and cryovessel. Working in col-laboration with faculty from the NCSU mechan-ical engineering department and a NASA consul-tant, we have also designed the 1,000 l fiberglass-composite vessel that will house the experimental

cells and high-voltage capacitor system. We arealso designing the 3He injection system. In ad-dition, work has begun on the development andcharacterization of a transient waveform digiti-zation system for the light collection system. Anoverview of the apparatus is shown in Fig. 1.1.

In addition to the research outlined above, de-tails of which can be found in subsequent or pre-vious contributions, several TUNL faculty servein leadership roles in the project. Gao, Golub,and Huffman serve on the project’s executivecommittee; Haase and Huffman serve as subsys-tem managers for the construction of the cryoves-sel and for the assembly and commissioning of thesubsystems, respectively; Golub serves as one ofthe principal scientists; and Huffman serves onthe federal project team as the project’s tech-nical coordinator and deputy project manager.Several others serve as work-package managersfor components of the subsystems.

Figure 1.1: (Color online) Cross sectional view of the nEDM apparatus. The approximate dimensionsare 7.5 m long, 5.5 m tall, and 2.2 m wide. The neutron beam enters the central volumefrom the right and is down-scattered in liquid helium to produce ultracold neutrons thatare confined within the measurement cells. The cells are positioned within a strong electricfield and weak magnetic field. The cells are surrounded by a roughly 1,000 l of liquid heliumhoused in a composite vessel. The 3He system and central volume are cooled to below500 mK with a 3He/4He dilution refrigerator.


1.1.2 Search for the Neutron Electric Dipole Moment: Geometric PhaseEffects

R. Golub, H. Gao, P.R. Huffman, C.M. Swank, Q. Ye, TUNL; the nedm collaboration

In order to achieve the targeted precision of the nEDM experiment, it is necessary to deal

with the systematic error associated with the interaction of the well known �v × �E field with

magnetic field gradients. This systematic effect was studied theoretically for a number of op-

erating conditions and is being investigated experimentally in a newly constructed apparatus

at TUNL.

In the neutron electric dipole moment(nEDM) apparatus, the interaction of the field�Beff ∼ �v × �E with magnetic field gradients willproduce a frequency shift linearly proportional tothe electric field. Known as the geometric phaseeffect and mimicking a true EDM, this effect hasrecently emerged as a primary systematic errorlimiting the precision of the next generation ofneutron EDM searches.

In the context of neutron EDM experiments,this effect was first investigated experimentallyand theoretically by Pendelbury et al. [Pen04]

and later by Lamoreaux and Golub [Lam05]. Inour recently published work [Bar06], we intro-duced an analytic form for the correlation func-tion that determines the behavior of the fre-quency shift, and we showed in detail how it de-pends on the operating conditions of the exper-iment. Using this analytic function for the caseof gas collisions, we averaged over a Maxwellianvelocity distribution to calculate the temperaturedependence of the frequency shift for 3He diffus-ing in superfluid 4He.

Figure 1.2: (Color online) Fixed-velocity frequency shift induced by the geometric phase effect for

UCN’s and for 3He [Bar06].


The results indicate that it may be possibleto fine-tune the effect to a high degree by an ap-propriate choice of operating temperature. Thepresent understanding of the effect is summarizedin Fig. 1.2. This is a plot of the normalized (lin-ear in E) frequency shift δω vs. the normalizedLarmor frequency ω0 for various values of the col-lision mean free path λ and wall specularity. Val-ues are calculated for particles moving with fixedvelocity v in a cylindrical measurement cell of ra-dius R, and the horizontal scale is fixed by thefrequency of motion around the cell, v/R. Ingeneral the shift for ultracold neutrons (UCNs)will be given by a value of ω0/(v/R) > 4, andfor 3He by ω0/(v/R) ∼ 1. From an experimentalpoint of view it is very appealing to try to makeuse of the zero crossings apparent in the figure toreduce the systematic effects to zero. Our theo-retical work indicates it is plausible to do this for3He, by tuning the temperature of the apparatusto take advantage of the 1/T 7 dependence of thediffusion constant for 3He in superfluid 4He.

Specifically for the nEDM experiment, wehave used our simulation results in combinationwith 3He transport calculations to set the nomi-nal operating temperature for the nEDM exper-iment, 450 mK.

To study these geometric phase effects exper-imentally, we have constructed a cryogenic appa-ratus consisting of a dilution refrigerator incorpo-rated into a non-magnetic Dewar. The apparatusis shown in Fig. 1.3. The Dewar is surrounded bya series of eight Helmholtz coils to provide a uni-form field.

The initial experiments aimed at verifying thepredicted depolarization rates of polarized 3Hein superfluid 4He at T ≤ 500 mK in a deuter-ated polystyrene rectangular cell have been com-pleted. We verified the depolarization probabilityper bounce of ∼ 1 × 10−7 using an improved setof NMR excitation and detection coils and super-conducting constant-field and gradient coils. Wecarried out a series of measurements of the appro-priate correlation function at different densities.To compensate for the fact that the relaxationtakes place at different frequencies, we rampedthe magnetic field so that measurements can bemade at a constant frequency. It is necessary toramp the gradient on and off to enable measure-ments of the polarization as a function of decaytime. We have recently improved the attainabletemperature, which did not quite reach the nec-essary level, and are working on improving thesignal-to-noise ratio further, so as to push the

measureable 3He density to lower limits.The geometric-phase-effect measurements

rely on theoretical calculations that show thatone can determine the effect by measuring theT1 relaxation rate of the 3He polarization ina magnetic field gradient. Thus it is not re-quired to measure directly the (�v × �B)-inducedfrequency shift. If one applies a uniform mag-netic field gradient ∂Bz/∂z large enough so thatit dominates all other field gradients, the relax-ation time T1 can be determined by traditionalNMR techniques, assuming wall relaxation canbe neglected. This technique does not require anelectric field and thus significantly simplifies theexperiment.

Figure 1.3: (Color online) TUNL dilution-

refrigerator setup for studying 3Hedepolarization in superfluid 4He andthe temperature dependence of thegeometric phase effect.

[Bar06] A. L. Barabanov, R. Golub, and S. K.Lamoreaux, Phys. Rev. A, 74, 052115(2006).

[Lam05] S. K. Lamoreaux and R. Golub, Phys.Rev. A, 71, 032104 (2005).

[Pen04] J. M. Pendlebury et al., Phys. Rev. A,70, 032102 (2004).


1.1.3 Development of the Cryogenic Design for the nEDM Experiment

D.G. Haase, P.R. Huffman, D.P. Kendellen, TUNL; J. Boissevain, California Institute ofTechnology, Pasadena, CA; L. Bartoszek, Bartoszek Engineering, Aurora, IL; W. Yao, Oak RidgeNational Laboratory, Oak Ridge, TN

A multi-institution collaboration is designing an experimental apparatus for a new measure-

ment of the nEDM. The experiment will be mounted on a dedicated beam line at the Spal-

lation Neutron Source (SNS) at Oak Ridge National Laboratory. In the last year there have

been several advancements in the design of the cryogenic systems, which must cool a cubic

meter of liquid helium to 0.25 K to produce ultracold neutrons. The design, procurement,

construction, and implementation of the cryogenic subsystem are led by TUNL faculty.

1.1.3.1 The nEDM Cryogenic System

The nEDM experiment requires cooling a 1 m3

volume of superfluid liquid helium to an operat-ing temperature between 0.25 K and 0.45 K. Thiscentral volume encloses a separate target volumeof superfluid in which 0.89 nm neutrons from theSNS are down-scattered into ultracold neutronsand confined in an acrylic measurement cell. Thecentral volume and target, plus associated ex-perimental services, are cooled by a high-flow3He-4He dilution refrigerator. The vacuum con-tainer enclosing the cryogenic systems is calledthe cryovessel and is cooled using liquid heliumthrough a permanent connection to a helium liq-uefier system. The cryogenics Work Package 1.03for the nEDM construction project includes thecryovessel with its enclosed 80 K and 4 K thermalshields, the helium liquefier system, the 3He-4Hedilution refrigerator, the central volume, and theassociated vacuum hardware, cryogenic sensors,and controls.

Over the last year, the nEDM cryogenic sys-tem has passed two significant project reviews,and refinements have been made to the experi-mental design. The 1.03 subsystem was reviewedat an nEDM collaboration meeting at ORNL inFebruary and the DOE annual project review inMarch 2010. The modifications have accompa-nied improved calculations of the cryogenic cool-downs and thermal loads. Several modificationsare driven by the stringent requirements for lowmagnetic fields and minimal electrical noise.

1.1.3.2 Modifications to the helium lique-fier system

A dedicated system with a helium liquefier andhelium flow to the cryovessel has been adopted.The nEDM project and the SNS are composing aMemorandum of Understanding for the construc-tion and operation of a common helium liquefierthat will serve the SNS users, the nEDM exper-iment, and another proposed SNS user project.In this plan, the nEDM experiment will receiveliquid helium and return cold or warm heliumgas to a liquefier located in a building adja-cent to the Fundamental Neutron Physics Beam(FNPB) external building. This arrangementshould improve the reliability of helium supplyto the nEDM cryovessel and reduce maintenancerequirements. The nEDM experiment will main-tain a large liquid-helium (LHe) storage volumeand piping system to connect to the cryovessel.The design of the flow system is driven by theneed to fill the 1 m3 central volume with liquidhelium at 4 K and to cool the rest of the cry-ovessel as quickly and efficiently as possible. Toreduce vibrations, the 80 K thermal shields willbe cooled by a positive pressure flow of liquid ni-trogen driven by an external cryogenic pump andphase separator.

1.1.3.3 Modification to the dilution re-frigerator

The process for cooling the central volume andtarget cell from 4 K to 0.25 K has been simpli-fied. The design of the 3He-4He dilution refriger-ator is modified so that it can function as a 3Heor 4He evaporative refrigerator above 1 K. Thisis achieved by adding a flow circuit to recircu-


late 4He and a “stopper” below the still that canshort-circuit the low-temperature heat exchang-ers and increase the effective pumping speed ofthe recirculation system.

At the beginning of the cooling process, the 1m3 target volume will be filled with liquid fromthe helium liquefier. At this point, the stopperwill be raised, and the dilution refrigerator willact as a helium evaporator by circulating andpumping LHe from an 80 l LHe entrainment vol-ume inside the cryovessel. The refrigerator cool-ing rate will be controlled by a liquid flow valve.As the target volume is cooled from 4 K to about2.6 K, the density of the LHe increases, so the vol-ume will be continuously filled with liquid fromthe entrainment volume. At 2.6 K, the target-volume fill will be ended, and the target volumewill be thermally insulated from the entrainmentvolume by a vacuum valve. As the target vol-ume cools further to near 1 K, the LHe flow willbe stopped, and 3He from an external gas stor-age tank will be circulated through the dilutionrefrigerator. In the temperature range of 1.5 Kto 0.6 K, 3He evaporation is a powerful coolingprocess. Finally near 0.6 K the dilution refrigera-tor stopper will be lowered to start flow throughthe refrigerator’s heat exchangers, and the gasmixture will be adjusted to start the 3He-4He di-lution process.

1.1.3.4 Re-entrant thermal shields

In the latest design of the nEDM system, theneutron beam guide penetrates the cryovessel toa point only a few inches from the liquid-heliumtarget volume. Therefore the vacuum jacket andthe 80 K and 4 K thermal shields must includere-entrant sections extending as far as 60 inchesfrom the target to the upstream end of the cry-ovessel. The re-entrant components must be elec-trically non-conducting and non-magnetic so thatJohnson noise and eddy currents will not over-whelm the SQUID sensors in the target volume.On the other hand the thermal shields shouldhave thermal conductivities sufficient to cool theinterior end caps to near 80 K and 4 K; other-wise an increased thermal load is passed on tothe sub-4 K refrigerator.

The re-entrant vacuum jacket and the 80 Kre-entrant tube of the cryovessel pressure barrierwill be constructed of G-10 composite. At theupstream end of the 80 K shield, an aluminumflange will be cooled by recirculating liquid ni-trogen. The re-entrant tube will be composedof sintered alumina (Al2O3) rods epoxied onto a

G-10 form. Alumina is a non-magnetic electricalinsulator, but it is also a good thermal conduc-tor at 77 K (λ = 157 W/m-K). Because of con-struction difficulties, the original 4 K re-entrantthermal shield design has been replaced by a 4K shield directly attached to the magnet systemthat surrounds the central volume. This G-10shield will be cooled by the same helium gas flowthat cools the magnet package.

The designs of the helium liquefier, the di-lution refrigerator, and the cryovessel are ap-proaching completion. The next steps includenegotiations with vendors, completion of theSNS/ORNL safety and quality assurance pro-cesses, approval from DOE to proceed, and thebeginning of the procurement process.

Figure 1.4: (Color online) Schematic of the cryo-genic components of the nEDM ex-periment that compose Work Package1.03.

Figure 1.5: (Color online) Schematic of the mod-ified design of the nEDM dilution re-frigerator to allow operation as a 4Heevaporative refrigerator in the range1.5 K < T < 4.2 K.


1.1.4 3He Spin Transport Simulation in the nEDM Experiment

H. Gao, Q. Ye, W.Z. Zheng, TUNL;

The transport of 3He spin through a magnetic field gradient is simulated in order todemonstrate whether nearly 100%-polarzied 3He can maintain its polarization after injectionfrom the ABS into the collection volume in the nEDM experiment. A projection is also madeof the spin transport from the collection volume to the central measurement cells. This yieldsnegligbile polarization loss if only the field-gradient-induced relaxation is considered.

In the nEDM experiment, nearly 100%-polarized 3He will be produced using the atomicbeam source (ABS). The polarized 3He beamwill be injected into the collection volume of thenEDM experiment, which is filled with super-fluid 4He, before being transferred to the centralmeasurement cells. Maintaining a high polariza-tion of 3He both through the injection stage andthrough transport from the collection volume tothe measurement cells is crucial to the success ofthe experiment. In this report, we present studiesof the depolarization of 3He due to magnetic fieldgradients between the exit of the ABS and thenEDM measurement cells, as well as projectionson spin transport from the collection volume tothe central measurement cell.

1.1.4.1 Simulation Methods

The simulation code (from Justin Torgerson,written in C++) is divided into three steps. Thefirst step simulates the polarization of 3He at theexit to the ABS. The second step is the 3He spintransport and calculates how many 3He atomscan successfully pass through the long transporttube without undergoing collisions with the wall.The last step calculates how the spin precessesas the 3He travels from the ABS to the collec-tion volume inside the transport tube. The 4thorder Runge-Kutta method is used to calculatethe Bloch equation of the spin �M precessing in amagnetic field �B. The vector form of the Blochequation is written as:

d �M(t)dt

= γ �M(t) × �B(�r) (1.1)

where �r is the position vector of the spin. Byintegrating the Bloch equation, one can obtainthe magnetization of the 3He spin inside the col-lection volume. Since the magnetic field insidethe collection volume is in the x direction, the x

component of the spin, Mx, gives the polarizationafter normalization.

1.1.4.2 Simulation Results

Figure 1.6: Counts vs. cosθ. The average polar-ization in the positive current config-uration is 76%.

The ABS simulation shows that 59.2% of theatoms in the right spin state can pass through thequadrupole magnets and < 0.01% of the atomsin the wrong state can pass through, resulting ina polarization of > 99.9%. The spin-transportsimulation shows that 82.8% of the atoms in theright spin state can pass through the transporttube and arrive at the collection volume. Thelatest magnetic field design (March 2010 fromSeptimiu Balascuta) has two current configura-tions: one with the current in the cylindricalcos θ coil in the positive direction, so that theB field in the collection volume is in the +x di-rection, and the other with the current in all thecoils reversed, so that the B field is in the −xdirection. Nearly 10,000 events have been sim-ulated for each configuration, giving a statisti-cal uncertainty of ∼ 1%. Figure 1.6 shows thehistogram of cos θ for the positive current con-figuration (The negative current configuration is


similar). Here, θ is the angle between the spinand the holding field inside the collection volume.The mean value of the histogram gives the aver-age polarization of all the events and is shown inthe box in the upper right corner of the figure.

For the positive current configuration, the av-eraged polarization is about 76%, and for thenegative current configuration, it is slightly bet-ter, 79%. We also simulate the average polar-ization of the ensemble as a function of distancemeasured from the center of the collection vol-ume. The same number of events are simulatedfor both current configurations. These results areshown in Figs. 1.7 and 1.8. The origin of the xaxis is the center of the collection volume, and180 cm corresponds to the exit of the ABS. Tworegions leading to polarization loss are identified.One is due to the strong and irregular fringe fieldat the ABS exit, the other is close to the collec-tion volume, where the spins start to rotate fromthe 3He injection direction to the x direction inorder to match the holding field inside the collec-tion volume.

1

0.95

0.9

0.85

0.8

Pol

ariz

atio

n

0.75

0 20 40 60 80 100 120 140 160 180Distance (cm)

Figure 1.7: Polarization vs. distance in the posi-tive current configuration.

1

0.95

0.9

0.85

0.8

0 20 40 60 80 100 120 140 160 180Distance (cm)

Pol

ariz

atio

n

Figure 1.8: Polarization vs. distance in the nega-tive current configuration.

The nEDM experiment requires > 95% polar-ization of 3He in the collection volume. Thereforethe present results are inadequate, and a moreuniform magnetic field is needed to improve the3He polarization in the collection volume. Solu-tions to this problem can involve either increasing

the overall transport field or stretching the field-transition region so that the spins have more timeto reorient themselves to the B-field direction inthe collection volume.

1.1.4.3 Polarization loss between the col-lection volume and the centralmeasurement cells

Once 3He atoms accumulate in the collection vol-ume, they are transferred to the central measure-ment cells through diffusion. The diffusion coef-ficient of 3He in the 4He bath is calculated asD = 1.8/T 7 cm2/sec [Lam02], where T is thetemperature of the 4He bath. At 350 mK, thediffusion coefficient is roughly 3000 cm2/sec. As-suming that the distance between the collectionvolume and the central measurement cells is l = 3m, the transport time for 3He can be estimatedby t ≈ l2/D = 30 s. So, the effective diffusionspeed is roughly Veff = l/t = 0.1 m/s. Thisis very slow compared to ∼100 m/s, the speedof 3He as it travels from the ABS to the collec-tion volume. The adiabatic factor for a spin pre-cessing in a changing magnetic field is defined as[Sli63]

α = tan−1(1

γ∣∣∣ �B∣∣∣2

∣∣∣∣∣d�B

dt

∣∣∣∣∣) = tan−1(1

γ∣∣∣ �B∣∣∣2

∣∣∣∣∇ �Bd�x

dt

∣∣∣∣)(1.2)

where α can be thought of as the angle betweenthe spin and the magnetic field. If α << 1, theadiabatic condition is satisfied, and the polariza-tion loss is negligible. The second equality showsthe importance of the spin velocity in determin-ing the adiabatic factor. In the current field de-sign, because of the roughly 20% polarization lossduring transport from the ABS to the collectionvolume, the adiabatic factor for the first part isabout 0.64 rad. A fair assumption is that thegradient in the transition region between the col-lection volume and the central measurement cellis no worse than that between the ABS and thecollection volume. Since the effective diffusionspeed is only 0.1 m/s, compared with ∼ 100 m/sin the first part, the adiabatic factor for the sec-ond part is projected to be 7.5 × 10−4 rad or0.04 degree. Therefore, in the region from thecollection volume to the measurement cells, thepolarization loss from the field gradient is negli-gible.

[Lam02] S. K. Lamoreaux et al., Europhys. Lett.,5, 718 (2002).

[Sli63] C. P. Slichter, Principles of MagneticResonance, Harper and Row, 1963.


1.2 Fundamental Coupling Constants

1.2.1 The UCNA Experiment

H.O. Back, L.J. Broussard, B.M. VornDick, C.R. Cottrell, J. Hoagland, A.T. Holley,R.W. Pattie, Jr., A.R. Young, TUNL

During the 2009 run period, the UCNA collaboration completed data-taking for a 1.4% mea-

surement of the β-asymmetry in polarized neutron decay. A publication incorporating a

total of 14.7×106 decays from the 2008 and 2009 run periods was prepared, with a value of

-.11966±0.00089(stat)±0.00123±0.0014 (syst) quoted for the β-asymmetry at the 67% confidence level.

A value for the axial form factor in the charged weak interaction of the nucleon, gA/gV , of

−1.27590+0.00409−0.00445 was extracted from the measured asymmetry.

Measurements of neutron decay provide fun-damental information on the parameters char-acterizing the weak interaction of the nucleon.These parameters, including the axial form factorfor the charged weak interaction, impact predic-tions of the solar neutrino flux, big-bang nucle-osynthesis, the spin content of the nucleon, andtests of the Goldberger-Treiman relation. Theyalso place constraints on standard-model exten-sions such as supersymmetry and left-right sym-metric models.

The UCNA (ultracold neutron—the A pa-rameter) experiment is the first attempt to uti-lize ultracold neutrons (UCNs) for an angularcorrelation measurement in neutron decay. Theuse of UCNs for such a measurement provides adifferent and powerful approach to the system-atic errors characteristic of traditional cold- orthermal-neutron-beam experiments. In particu-lar, UCNs provide significant advantages for de-termining the neutron polarization and for con-trolling neutron-generated backgrounds. Histori-cally, the neutron density in cold beams has beenmuch higher than the available UCN densities,making beam experiments the favored methodfor angular correlation measurements, but sys-tematic errors have increasingly driven the uncer-tainties in such measurements, providing impetusto explore the use of UCN. The first experimentalresults, a 4.5% measurement of the β-asymmetry,were published in 2009 [Pat09].

UCNA is located in the Los Alamos Neu-tron Science Center (LANSCE) and utilizes asolid-deuterium ultracold-neutron source devel-oped for this project. The 800 MeV proton beamat LANSCE is directed at a tungsten target sit-

uated within an outer, bucket-shaped shell ofgraphite and an inner shell of Be, both of whichserve to reflect and partially moderate the spal-lation neutrons produced. The neutrons are fur-ther moderated to temperatures under 100 K bya layer of polyethelene held at room tempera-ture and by a cold-moderator at temperaturesbetween 20 and 80 K. A 2 l volume of solid deu-terium, held at 5 K inside of a UCN guide, servesas the UCN converter. The UCNs are coupledto the decay volume through a 15 m guide sys-tem that incorporates a 7 T polarizer magnet, aspin-flipper system, and finally the bore of a 1 Tspectrometer magnet, where decays are observed(see Fig. 1.9).

1.0 T Superconducting solenoidal magnet

Cu decay volume

MWPC Plasticscintillator

Cu guide

DLC-coatedquartz guide

PMT

Be-coatedmylar foil

7T Polarizer/AFPLight guide

UCN from SD source2

Spectrometer

Polarizer/AFP

Switcher

PPMSource

Gatevalve

Figure 1.9: A schematic diagram of the UCNApolarizer and β-spectrometer. The in-set depicts the layout of the sourceand guides coupled to the experiment.

The µ · B interaction of the neutron’s mag-


netic moment, µ with a magnetic field B pro-duces an energy change of ±60 neV/T, depend-ing on the whether the neutron spin is parallel orantiparallel to the applied field. The extremelylow kinetic energy of UCNs (below 180 neV forour experiment) makes possible a method uniqueto UCNs for producing highly polarized samples.The µ · B interaction produces a 420 neV bar-rier to the neutrons with spin parallel to thefield (µ is negative) in our 7 T polarizer mag-net, completely blocking the transmission of thisspin state through the high field system. Thereis therefore an essentially 100% polarized sam-ple of UCNs after traversing this region. UCNsthen pass through a high-power, radio-frequencyadiabatic spin-flipper which permits the prepa-ration of a UCN sample polarized either parallelor antiparallel to the field. The spin-flipper re-gion is then coupled by our guide system to adecay volume within the uniform 1 T field of theβ-spectrometer.

During the 2008/2009 run period, weachieved an average decay rate of about 18 Hzfor most of the run, an improvement of roughly afactor of 3 over the decay rate in 2007. A run cy-cle included 60 min of β-decay measurement, fol-lowed by an approximately 4 min measurement ofthe depolarized UCN fraction that accumulatedin our decay volume during the β-decay measure-ment, and about a 12 min measurement of thebackground. The UCN spin-flipper was set ei-ther “on” or “off” in an eight cycle sequence de-signed to measure decay rates with the neutronspin parallel and antiparallel to the field axis, andto cancel, up to second order, drifts in the back-grounds.

UCNs were confined to a 12.5 cm diame-ter, 3 m long Cu decay trap, coaxial with a 1T uniform field at the center of the spectrom-eter. The ends of the decay trap were sealedwith Be-coated, Mylar windows. β particles wererecorded in detector assemblies situated at ei-ther end of the solenoidal spectrometer, wherethe field has fallen to 0.6 T. Each detector assem-bly is composed of a position-sensitive multi-wireproportional counter (MWPC) with aluminum-coated entrance and exit windows, backed by aplastic scintillator to measure the full energy ofthe β particles. Backgrounds were reduced by re-quiring a coincidence between the scintillator andMWPC, and by using a cosmic ray veto. Over14.7 × 106 decays were observed in our analysiswindow between 275 keV and 625 keV, yieldinga β-spectrum in good agreement with expecta-tions.

The emphasis of the 2008 run period wasto assess the scattering and energy-loss correc-tions. To this end, we made measurements of

the β-asymmetry with three different sets of de-cay trap and detector foils. In 2009, we uti-lized the geometry with the smallest foil thick-nesses (and therefore the smallest scattering andenergy-loss corrections), and focused on improve-ments to the decay trap designed to reduce po-tential spin-dependent systematic effects due tothe wall-coatings in our decay trap and polarizedneutron transport guides, and on improved neu-tron polarimetry.

Rat

e (H

z/25

keV

)R

ate

(Hz/

25 k

eV)

0

0.10.1

0.20.2

0.30.3

0.40.4

0.50.5

Background subtracted Background subtracted

spectrum spectrum MC

Measured backgroundMeasured background

(keV) (keV)reconreconE0 100100 200200 300300 400400 500500 600600 700700 800800 900900 10001000

0A-0.13-0.13

-0.125-0.125

-0.12-0.12

-0.115-0.115

-0.11-0.11

-0.105-0.105 0.0089(stat) 0.0089(stat)±A0 = -0.11966 A0 = -0.11966

Figure 1.10: Upper: background subtracted betaenergy spectra (solid red circles),combining both sides of the detectorsand two spin states, overlaid withthe Monte Carlo spectrum (blue his-togram). The open black circlesrepresent the measured backgroundspectrum. Lower: beta-asymmetryvs. the reconstructed beta energy.

The data taken in 2009 completed the neces-sary data for a convincing measurement of theβ-asymmetry in neutron decay at the 1.4% level[Liu09]. Our experimental results are summa-rized in Fig. 1.10. We also continue to explorethe introduction of Si detectors into the UCNAspectrometer in a geometry permitting the de-tection of β particles and recoil protons in co-incidence. Such a scheme should permit the re-duction of the systematic error in the polariza-tion and the measurement of additional angularcorrelations, some of which are particularly inter-esting in constraining supersymmetric extensionsto the standard model. This project, referred toas UCNB, has already benefited from an inten-sive development effort at Los Alamos producingthree different sets of Si detectors with propertiesappropriate for neutron β-decay measurements.

[Liu09] J. L. Liu et al., ArXiv 1007.3790v1, 2009.

[Pat09] R. W. Pattie et al., Phys. Rev. Lett.,102 (2009).


1.2.2 High-precision Measurement of the 19Ne Lifetime

L.J. Broussard, H.O. Back, M.S. Boswell, A.S. Crowell, M.F. Kidd, C.R. Howell, R.W.

Pattie, Jr., A.R. Young, TUNL; P.G. Dendooven, G.S. Giri, D.J. Van der Hoek, K. Jung-

mann, W.L. Kruithof, C.J.G. Onderwater, P.D. Shidling, M. Sohani, O.O. Versolota,L. Willmann, H.W. Wilschut, Kernfysich Versneller Instituut, Groningen, The Netherlands

The T= 12

mirror transitions, such as the β+ decays of 19Ne, 37K and 21Na, have beenidentified as excellent candidates for high-precision studies of the weak interaction in theStandard Model, complementing the well-studied 0+ → 0+-decay and neutron-decay measure-ments. In 2009 we completed measurements to extract a high-precision lifetime result for thedecay of 19Ne to 19F at the Trapped Radioactive Isotopes Microlaboratories for FundamentalPhysics (TRIμP) facility at the Kernfysich Versneller Instituut, and we hope to achieve a finaluncertainty of 0.03%. Upon completion of a detailed analysis of the systematic uncertaintiesof our experiment and analysis methods, the program can be expanded to high-precisionmeasurements of other mirror transitions.

The lifetime of the decay of 19Ne, with an ad-ditional observable such as the β-asymmetry, canbe used to precisely extract the CKM matrix el-ement Vud and study proposed Standard Modelextensions such as left-right symmetric models.We used the TRIμP isotope separator to createa very pure beam of 19Ne. The isotopes wereimplanted into an aluminum tape, and a customtape-drive system transported the implantationsite into a shielded detector array. The tape wasclamped between plastic blocks in which the de-cay positrons annihilate and release two 511 keVγ rays. The annihilation radiation was detectedby two HPGe clover detectors in coincidence. Ta-ble 1.1 details the accumulated decays during thespring 2009 data-taking.

Table 1.1: Total Data Acquired

Isotope Prev. Meas. (s) Decays19Ne 17.248 ± 0.17% 3 × 107

37K 1.2248 ± 6.0% 4 × 105

21Na 22.487 ± 0.24% 2 × 106

To meet the goal of a high-precision lifetimemeasurement, all systematic uncertainties mustbe very well controlled and understood. We haveidentified several potential sources of error inour experimental method: ambient backgrounds,contaminant populations in the sample, diffusionof the sample, and rate-dependent effects in thedata acquisition. The ambient background levelswere evaluated by blocking the implantation of19Ne into the tape. By requiring a coincidencebetween clovers, we were able to achieve a signalto background of better than 1000:1. A silicondetector installed at the isotope separator exit

allowed us to examine the relative isotope lev-els directly. Most isotopes were present at neg-ligible levels, although in some cases the levelsin certain long-lived isotopes could not be evalu-ated using this method. We performed a separatehigh-precision measurement of the lifetime withlong implanting times, such that the 19Ne pro-duction saturates but longer-lived potential con-taminants continue to accumulate, and we do notyet see evidence of contamination at our currentlevel of precision. The third potential source ofuncertainty comes from possible diffusion of 19Neout of the tape. We performed several lifetimemeasurements with different implantation depthsin order to accentuate the effect, and we saw noevidence of diffusion (see Fig. 1.11).

Rel

ativ

e L

ifet

ime

m)μDegrader Thickness (

0 9 18 230.9990

0.9995

1.0000

1.0005

1.0010

Figure 1.11: Diffusion from the tape is accentu-ated by degrading the energy of theparticle beam before it is implanted.A 23μm degrader corresponds to theparticles being implanted near thesurface of the tape. Rate-dependentcorrections are not yet fully under-stood, so only relative lifetimes areshown.


The rate-dependence of our data acquisitionand analysis methods were studied in great detailboth during and after the experiment. In orderto minimize deadtime due to piled-up pulses ormissed events, we digitized all of the shaped pre-amplifier pulses using the CAEN V1724 wave-form digitizer. A constant-rate pulser used con-tinuously throughout the experiment provided anin-situ measurement of deadtime, and extensivepulser studies were performed after the exper-iment to further explore the rate-dependence ofthe detection system. Event pulses are comparedto a known average pulse shape, which improvestime-resolution of distinct events to less than 1μs.In order to minimize the uncertainty due to acci-dental coincidences, a fixed analysis deadtime ofseveral μs is applied, and the form of the fittingfunction is corrected for deadtime. Finally, useof Gaussian or Poisson statistics for such a high-

precision measurement can introduce large biasesinto the extracted lifetime. We found that wecould minimize bias and uncertainty by insteadperforming a least-squares minimization usingthe binomial distribution, and we performed acareful study of the statistical bias as a functionof rate and fixed deadtime, using simulated data.

We have made significant progress towardsunderstanding the rate-dependent systematics atour desired precision. Obtaining a 0.03% resultfor the lifetime of 19Ne would be a strong in-dication that our system is well-suited for morehigh-precision measurements using other mirrortransitions. We have already performed some de-velopment work in minimizing contamination inthe production of 37K and 21Na, and we hope tocontinue the program with high-precision mea-surements of the lifetimes of these and other mir-ror transitions.

Neutrino and Dark Matter Physics

Chapter 2

• ββ Decay Experiments

• Tritium β Decay

• Dark Matter Search

16 Neutrino and Dark Matter Physics TUNL XLIX 2009–10

2.1 ββ Decay Experiments

2.1.1 The MAJORANA DEMONSTRATOR Neutrinoless Double-Beta DecayExperiment

J.F. Wilkerson. M. Busch, D. Combs, P. Finnerty, G.K. Giovanetti, M. Green, R. Hen-

ning, M.A. Howe, M.F. Kidd, L. Leviner, S. MacMullin, D. Phillips, K. Snavely, J.

Strain, G. Swift, W. Tornow, A.R. Young, TUNL

The Majorana Collaboration is constructing the Majorana Demonstrator aimed at deter-mining the feasibility of a tonne-scale 76Ge-based neutrinoless double-beta-decay experiment.The baseline design includes 15 kg of 86% enriched Ge detectors and 20 kg of natural Gedetectors. The measurement may be able to reveal whether neutrinos are Dirac or Majoranaparticles, while providing insights into both lepton number violation and the absolute massscale of neutrinos. The Demonstrator with its low-energy sensitivity, is also expected to becapable of searching for dark matter.

The primary goal of the Majorana Col-laborationa is to develop a 1-tonne 76Ge-basedneutrinoless-double-beta-decay experiment withthe sensitivity to probe neutrino mass in theinverted hierarchy region (20 to 40 meV).The collaboration is building the Majorana

Demonstrator, an array of high-purity natGeand enriched 76Ge crystals that will be de-ployed in ultra-low-background electroformed Cucryostats. The Demonstrator should be capa-ble of reaching the ultra-low-background levelsrequired for a 1-tonne scale experiment.

The Demonstrator will consist of 15-30 kgof 86% enriched 76Ge detectors and 10-20 kg ofnatGe detectors. The HPGe detectors will be fab-ricated as P-type point-contact detectors, whichhave very low capacitance and excellent single-site to multi-site pulse-shape-discrimination ca-pabilities. The detectors will be assembled intostrings and housed in two ultra-pure electro-formed Cu cryostats. These cryostats will be em-bedded in a passive layered shield of ultra-pureelectroformed Cu, then Pb, all of which will besurrounded by an active cosmic-ray veto detectorsystem and passive neutron shielding.

Significant progress has been made by the col-laboration in moving forward with the construc-

tion of the Demonstrator and with additionalR&D for a 1-tonne scale detector. Below is abrief summary of research activities carried outby TUNL collaborators. More details are avail-able in sections that follow.

• Assay: The group has carried out bothradiometric and neutron-activation-basedmaterials assay at the Kimballton Under-ground Research Facility (KURF) and atthe North Carolina State University reac-tor. Materials counted include Pb, plas-tics, and low-mass cables. An ultra-cleansample-preparation clean-room has beencommissioned at the University of NorthCarolina at Chapel Hill (UNC) This facilityallows materials to be prepared for count-ing or for assembly in prototype systems.

• Detector Development: A customizedbroad-energy germanium (BEGe) detector,mounted in a low-background cryostat hasbeen installed in a TUNL-designed andTUNL-fabricated low-background shield atKURF. This Majorana low-backgroundBEGe detector (MALBEK) is being used toinvestigate BEGe detector optimization, aswell as low-energy (sub-keV) sensitivity. A

aMajorana Collaboration Institutions: Black Hills State Univ., Spearfish, SD; Duke Univ., Durham, NC; Insti-tute for Theoretical and Experimental Physics, Moscow, Russia; Joint Institute for Nuclear Research, Dubna, Russia;Lawrence Berkeley National Laboratory, Berkeley, CA; Los Alamos National Laboratory, Los Alamos, NM; Oak RidgeNational Laboratory, Oak Ridge, TN; Osaka Univ., Osaka, Japan; North Carolina State Univ., Raleigh, NC; PacificNorthwest National Laboratory, Richland, WA; Queen’s Univ., Kingston, Canada; South Dakota School of Mine andTechnology, Rapid City, SD; Triangle Universities Nuclear Laboratory, Durham, NC; Univ. of Alberta, Edmonton,Alberta, Canada; Univ. of California-Berkeley, Berkeley, CA; Univ. of Chicago, Chicago, IL; Univ. of North Carolina,Chapel Hill, NC; Univ. of South Carolina, Columbia, SC; Univ. of South Dakota, Vermillion, SD; Univ. of Tennessee,Knoxville, TN; Univ. of Washington, Seattle, WA.

TUNL XLIX 2009–10 Neutrino and Dark Matter Physics 17

number of additional Ge detectors, includ-ing a second BEGe detector that is similarto MALBEK but not in a low backgroundshield, have been acquired. These detec-tors are being used for a variety of detec-tor characterization and DAQ developmenttests. Work continues on the redeploymentof the existing segmented enriched germa-nium detector.

• Module Design and fabrication: Sig-nificant design, fabrication, and construc-tion of prototype string parts and theirrelated assembly tooling have been com-pleted. The collaboration is close to a finalstring design. Tests have also been com-pleted on finding suitable low-activity ma-terials to be used for the vacuum seals ofthe cryostat.

• Data Acquisition: The object-orientedreal-time control and acquisition system isbeing used for prototype systems at LANL,LBNL, UNC, Univ. of Washington, and forunderground detectors at Soudan, WIPP,and KURF. Evaluation of several differentpotential digitizer systems has been com-pleted.

• Ge refinement: In conjunction with col-laborators at the University of South Car-olina and Oak Ridge National Laboratory,we have supported the development of a Gerefinement system that will produce elec-tronic grade Ge crystals suitable for detec-tor fabrication from enriched Ge materialin the form of GeO2. A commercial facil-ity has been established in Oak Ridge, anda team of consultants working with Maj-

orana collaborators have completed a de-tailed refinement plan.

• Underground Facilities: Working withcollaborators at LANL and the engineeringteam from the Sanford Underground Lab-oratory at Homestake, a design has beencompleted for the main Demonstrator

laboratory, and the cavity has been exca-vated. Space for a temporary electoform-ing facility that will also be located on the

4850-foot level at Sanford has been pre-pared, with beneficial occupancy expectedin the fall of 2010.

• Electroforming Facilities: Equipmentand material have been procured for anunderground electroforming facility as partof a NSF Major Research Instrumentationgrant. The facility is needed to produce theultra-low-background Cu cryostat compo-nents as well as the inner-shield Cu sheets.In addition to the electroforming baths,this facility will host a full machine shopto allow fabrication of custom parts. Oper-ating procedures have been developed andseveral readiness and safety reviews havebeen successfully completed.

• Simulations and Analysis: The TUNLgroup continues to make contributions tothe Majorana-GERDA simulation devel-opment framework known as MaGe. Workis underway to implement the detailed de-sign geometry in the simulation programframework.

• Dark Matter Searches When coupledwith low-noise electronics, the low capac-itance of the point-contact and BEGe de-tectors provides sensitivity to sub-keV en-ergies. At these energies, one can per-form searches for certain dark-matter can-didates, such as WIMPS, with masses inthe range of 1-10 GeV/c2. A search hasbeen carried out with a point-contact de-tector as part of the CoGeNT collabora-tion, and the MALBEK detector is beingoptimized for a similar search.

• Project Management TUNL faculty andstaff are guiding the management and engi-neering efforts of the Demonstrator and1-Tonne Ge R&D activities. This past yearthe collaboration successfully completed aDOE/NSF Technical, Cost, Schedule, andManagement Review and passed its CriticalDesign CD-1 review of the Demonstrat-

or project. A NSF DUSEL panel reviewof its 1 Tonne Ge efforts was also passedsuccessfully.


2.1.2 A Tonne-Scale Ge Neutrinoless Double-Beta Decay Experiment

J.F. Wilkerson, R. Henning, TUNL; D. Steele, S. Elliott, Los Alamos National Laboratory,Los Alamos, NM

The 1TGe Collaboration is carrying out R&D aimed at the development of a tonne-scale 76Ge-based neutrinoless double beta decay experiment that could be sited at the DeepUnderground Science and Engineering Laboratory (DUSEL). We describe what has beenaccomplished and outline our future R&D plans.

The motivation for a tonne-scale (1TGe) neu-trinoless double beta decay (0νββ) experimentis to determine the nature of the neutrino andreach a Majorana-neutrino-mass sensitivity in-dicated by atmospheric neutrino oscillation re-sults. Worldwide there are currently two majorefforts exploring the feasibility of a tonne-scale,Ge-based 0νββ decay experiment: the Majo-

rana Demonstrator and GERDA Phases Iand II. These experiments are exploring two radi-cally different approaches to shield against γ raysoriginating from primordial radioactivity exter-nal to the Ge diodes. GERDA, located at theGran Sasso National Laboratory in Italy, is im-mersing Ge diodes in high purity liquid argon sur-rounded by a large water tank. The Majorana

Demonstrator (See Sect. 2.1.1), to be locatedat the 4850-foot level at the Sanford Laboratoryin South Dakota, utilizes a conventional layereddesign, in which ultra-pure electroformed copperand lead are used to shield a vacuum cryostat.In addition, the collaborations have also been ex-ploring a number of alternative technologies re-lated to HPGe detectors and their readout. Thetwo collaborations have strong cooperative tiesand intend to join efforts to build a single tonne-scale experiment. The eventual 1TGe experimentwill utilize the best aspects of the Demonstra-

tor and GERDA approaches.

The 1TGe Collaboration consists of the en-tire Majorana Collaboration and collabora-tors from the Max-Planck-Institut fur Physik,Munich. In addition, we have had informalparticipation and contributions from a numberof other GERDA collaboration members. The1TGe experimental activities are being managedas a task within the Majorana Collaboration’sexisting management structure. Partial supportis being provided by NSF as part of their DUSELR&D activities.

During the past year our primary focus has

been to:

• define experimental-facilities requirementsfor the 1TGe experiment;

• identify possible major environmental,health, and safety hazards;

• develop preliminary layouts for the experi-mental hall;

• identify and begin to study DUSEL inte-gration issues;

• justify the depth requirement for the 1TGeexperiment at DUSEL;

• develop initial cost and schedule estimates;

• finalize plans for 1TGe simulation and en-gineering R&D;

• initiate the 1TGe experiment’s simulationprogram;

• initiate the Ge recycling R&D activities.

We have developed two sets of preliminaryconceptual designs and experimental layouts,based on the two different shielding approaches,to assist DUSEL facility planning. Our R&Dplan includes simulation and design studies tocompare the background suppression capabilitiesand technical feasibility of shielding approachesbased on these two concepts. From the pre-liminary layouts, we have identified a set of is-sues related to the integration of the 1TGe ex-periment at DUSEL that require further study.The LAr/water shielding configuration may re-quire excavation beyond the current 7400-foot-level laboratory-module baseline dimensions, de-pending on the outcome of our design and simu-lation studies. The transport of potentially largeamounts of cryogen underground presents a lo-gistical challenge that we will also study. A plan


for the underground fabrication and installationof the experiment will be developed, based onrefined conceptual layouts.

Many of the underground facility require-ments for the tonne-scale experiment are inde-pendent of the shielding configuration. In ei-ther configuration, 1TGe requires undergroundspace for a copper electroforming laboratory, ma-chine shop, germanium-detector fabrication facil-ity, and an experimental hall. We have providedDUSEL with estimates for the requirements ofthese facilities, including environmental require-ments, utilities needs such as power and water,lifting capacity, network capacity, and materialtransport. The primary hazards at these facili-ties arise due to the presence of (potentially largeamounts of) cryogens in the experimental halland the use of acids in the electroforming and ger-manium laboratories. Part of our R&D scope in-cludes the development of a Ge acquisition plan,in order to optimize the use of the enriched ma-terial. We are also exploring different scenar-ios for the underground purification, processing,

and fabrication of Ge diodes, in order to reducecosmic-ray induced backgrounds.

The cost of a tonne-scale experiment is domi-nated by the procurement of enriched Ge isotopeand the production of HPGe detectors. Based onour initial estimates, the cost of a tonne-scale ex-periment is therefore largely independent of theshielding configuration to be chosen. Cost andschedule estimates have been based on scaling upthe actual or estimated costs for the Majorana

Demonstrator and GERDA experiments. Thepreliminary schedule calls for installation of thetonne-scale experiment at DUSEL to begin in2017.

As part of our education and outreach ef-forts, we are developing and will maintain a se-ries of three public exhibits related to particleand nuclear astrophysics at the Morehead Plan-etarium and Science Center at the University ofNorth Carolina. The first such exhibit, a cosmic-ray spark-chamber, is nearing completion and isscheduled to be operational by September 2010.


2.1.3 Project Engineering for the Majorana Demonstrator Project

M. Busch, R. Henning, J.F. Wilkerson, TUNL;

Project engineering for the Majorana Demonstrator project (see Sect. 2.1.1) is being man-

aged by engineers at TUNL. The primary responsibility of the project engineer is to provide

design guidance and coordination for all Majorana Demonstrator hardware. This oversight

is provided in three ways: (i) develop an engineering strategy and chair design reviews in

accordance with this strategy, (ii) coordinate all engineering and design activities required to

complete the project, and (iii) communicate and archive all design content to the collabora-

tion.

2.1.3.1 Engineering Strategy and DesignReviews

The project engineer is responsible for developingthe engineering strategy, arranging and chairingdesign reviews, and documenting the review re-ports.

The engineering strategy outlines the designreview process for the Majorana Demonstra-

tor. There are three major review stages: con-ceptual design, preliminary design, and final de-sign. Each design review clearly marks a mile-stone in the design process. The conceptual de-sign review establishes a baseline configurationthat allows engineers and designers to completethe necessary analysis and to specify all of themajor components of the system. The prelimi-nary design review serves as a way to lock in theconfiguration and specifications for a final config-uration. The final design review brings togetherthe final engineering calculations, safety analysis,fabrication drawings, and assembly procedures toensure a quality finished product.

2.1.3.2 Coordination of Design Activities

The Majorana Demonstrator engineeringteam consists of over twenty design contribu-tors at seven institutions around the country.The project engineer is responsible for provid-ing clear design guidance for all design contribu-tors. The Majorana Demonstrator projecthas been divided into eleven major tasks. A taskleader has been assigned to manage each task.The project engineer works with each task leaderand with the project controls office to documentmajor design activities and milestones in accor-dance with the engineering strategy. The im-plementation of the engineering strategy is re-

ferred to as the engineering plan. The engineer-ing plan is documented as a subset of the overallproject work breakdown structure, and is trackedmonthly by the project engineer, task leaders,and the project office. Any discrepancy in theengineering plan and actual progress is addressedin the monthly status updates, and corrective ac-tions are coordinated by the project engineer.

In developing the engineering plan, some as-pects of the design have been identified as outsideof the task leaders’ area of expertise or as incor-porating aspects of several tasks, and the projectengineer takes a more direct role in the design de-velopment of these aspects of the project. An ex-ample of this is the detailed production plan forunderground electro-forming of ultra-pure cop-per. This activity incorporates aspects of the in-frastructure task, the electro-forming task, detec-tor task, module task, and shielding task. In de-veloping the electro-forming production plan, theproject engineer has refined the underground ma-chine shop tool specifications to meet the needs ofthe detector, module, and shielding design, andhas developed a production plan to optimize thedelivery of copper parts in accordance with thebackground goals of the project.

2.1.3.3 Communicating Design Status

A series of weekly conference calls are utilizedas a way of communicating and coordinating thedesign status and near term goals of the Ma-

jorana Demonstrator. The project engi-neer attends most weekly status calls and chairsa weekly engineering call to address engineer-ing issues. Documents and images of the de-sign are posted to a central conference-call-and-collaboration-meeting website in file formats thatare readable by the entire collaboration.


All fabrication drawings, including proto-types and test parts, are assigned unique partnumbers and posted in pdf format to an inter-nal website. The drawing numbers follow a con-vention devised and managed by the project en-gineer. This convention is documented on thewebsite.

All three-dimensional design content for theMajorana Demonstrator is produced usingSolidWorks. Currently, SolidWorks models areexchanged periodically between designers. There

is a plan in place to implement a server-basedarchive and product-development managementsoftware tool called SolidWorks Enterprise PDM.This system will allow designers at any collabo-rating institution to update their model to thecurrent version, check in and checkout designcontent, manage version and revision control, andhave an automated mirrored archiving system.This software implementation and the necessarytraining required for this system will be managedby the project engineer.


2.1.4 Copper Electroforming for the Majorana Demonstrator at theSanford Underground Laboratory

R. Henning, D. Phillips, M. Busch, J. Wilkerson, TUNL; E. Hoppe, Pacific Northwest Na-tional Laboratory, Richland, WA; S. Elliott, Los Alamos National Laboratory, Los Alamos, NM ;K. Jeskie, C.-H. Yu, Oak Ridge National Laboratory, Oak Ridge, TN

The Copper Electroforming in a Temporary Cleanroom facility is a project to produce ultra-

pure copper at the 1480-m level in the Sanford Underground Laboratory. This copper will

be used as a construction material for the future Majorana Demonstrator project. We review

recent progress on this facility and the roles of TUNL researchers.

The Cu Electroforming in a TemporaryCleanroom facility (CETC) is a project to pro-duce ultra-pure copper at the 1480-m level inthe Sanford Underground Laboratory (SUL) inLead, South Dakota. This copper will be usedas a construction material for the future Ma-

jorana Demonstrator project, which will belocated at the same level in the SUL. The cop-per is purified of natural trace radioactive con-taminants using an electroforming process. Cop-per for the Majorana project must be electro-formed underground to minimize cosmogenic iso-topes, specifically 60Co that forms if it is exposedto cosmic rays at the surface of the earth. It willtake a considerable amount of time to electro-form all of the Cu needed for the Majorana

Demonstrator. The CETC allows the neces-sary early start to this critical path process. Ma-

jorana collaborators at TUNL are responsiblefor the overall coordination and commissioningof the CETC, as well as the procurement of com-ponents using an NSF-MRI grant. In additionto the group at TUNL, E. Hoppe at PNNL isresponsible for the design and operation of theelectroforming baths, S. Elliott at LANL is re-sponsible for the assembly and commissioning ofthe clean room building, and K. Jeskie at ORNLis responsible for the environmental, safety, andhealth aspects of the project.

The electroforming process purifies copper viathe well-known process of electroplating. Electriccurrent flows from a sacrificial anode of oxygen-free, high-conductivity copper stock, through aplating bath composed of sulfuric acid and cop-per sulfate, and onto a cathode, completing anelectrical circuit. The current decomposes thesacrificial anode into copper ions that traversethe plating bath and are deposited on the cath-

ode. The cathode consists of a stainless steelmandrel of a size and shape driven by the de-sired final part. For the CETC, the mandrels areright cylinders 54 cm long and 33 cm in diame-ter. Figure 2.1 shows a mandrel that has copperelectroformed on it. Once electroforming is com-plete, the electroformed Cu stock will be rinsed,left on the mandrel, double-bagged, and storedunderground. The final removal from the man-drel, machining, and cleaning of the part will beperformed at the Majorana Lab machine shop,once that facility is operational. This final proce-dure does not fall under the scope of the CETC.Until that time, the part will be stored in thevicinity of the CETC in sealed plastic drums.

The CETC consists of a modular, pre-fabricated cleanroom containing seven to tenelectroforming baths and a small annex area (seeFig. 2.2). The building footprint is 40 ft by 12ft, and the cleanroom area with the baths will beclass 1000-10000. The CETC will become a sig-nificant production facility but is intended to betemporary. The future Majorana Lab itself willcontain an electroforming facility with 16 baths.

The CETC had its initial readiness reviewat SUL in January 2010. The collaborationhas followed up on the recommendations fromthe review and is generating the documentationrequired by the reviewers and SUL. The maindocuments include a project description, waste-stream management plan, oxygen-deficiency-hazard analysis, an emergency response and spillplan, an environmental monitoring plan, a mem-orandum of understanding between the Majo-

rana project and SUL, detailed chemical in-ventories, detailed experimental operating pro-cedures, material safety data sheets, a what-if type hazards analysis, liquid-nitrogen trans-


portation procedures, and a training check listfor all Majorana collaborators who will workat the CETC. The Majorana group at TUNLhas also procured the majority of the electroform-ing hardware for the CETC and has shipped itto SUL and PNNL.

We intend to commission the CETC duringfall of 2010 and start operations at the end ofCY 2010. The operating life of the CETC is in-tended to be at least two years.

Figure 2.1: A copper piece that the collabo-ration electroformed underground atthe Waste Isolation Pilot Plant inCarlsbad, NM in 2007. The mandrelused is also visible. The pieces thatwill be electroformed at the CETCwill be similar, but larger.

Figure 2.2: Functional layout of the tempo-rary cleanroom. The electroformingarea contains the seven electroform-ing baths on the right of the diagram.The annex or ante area will be usedto change from dirty mine gear intocleanroom garb.


2.1.5 Deployment of the Majorana Low-Background Broad-Energy Ger-manium Detector at KURF

P. Finnerty, G.K. Giovanetti, R. Henning, M.A. Howe, S. MacMullin, J. Strain, J.F. Wilk-

erson, TUNL; J.I. Collar, University of Chicago, Chicago, IL;

We have deployed a customized broad-energy germanium (BEGe) detector, the Majorana

low-background BEGe at Kimballton (MALBEK) detector, in a low-background cryostat and

shield at the Kimballton Underground Research Facility in Virginia.

The operation of a new low-background de-tector in KURF required the development of adedicated infrastructure. We obtained, modified,and deployed two connex trailers, one for the dataacquisition system and one for the detector andshield. These trailers were deployed in June 2009(see Fig. 2.3). Each trailer is outfitted with inter-nal air filtration and dehumidifiers. For more in-formation on the MALBEK detector deployment,see Ref. [Fin09]. The details on the infrastruc-ture in place at KURF in the Kimballton mineare given in Ref. [Fin08].

Figure 2.3: (Color online) MALBEK detector andDAQ connex trailers being deployedat KURF in June 2009.

The detector and DAQ were deployed under-ground in January 2010. Figure 2.4 shows thedetector in the shield stand with ancient lead be-low the cryostat. At this time, the lead housewas stacked, all slow controls installed, the DAQbrought online, and the first data/calibrationruns were taken. The implemented MALBEKDAQ is currently operating underground in a

production data-taking mode and is remotely ac-cessible 24 hours per day. See Sect. 9.4.4 for moreinformation regarding the MALBEK DAQ andSect. 2.1.6 for details on the operating perfor-mance of the detector.

Figure 2.4: (Color online) MALBEK’s low-background copper cryostat.

The lead shield is comprised of 2.54 cm of an-cient lead and 20.32 cm of standard lead. Theentire lead shield and internal cavity is contin-uously purged with LN2 boil-off to reduce po-tential radon backgrounds (see Fig. 2.5). Addi-tional neutron shielding is provided by 25.4 cmof polyethylene (see Fig. 2.6). The status of asoon-to-be-installed muon veto can be found inSect. 9.3.4.1. For more information on the MAL-BEK shield, see Ref. [Fin09].


Figure 2.5: (Color online) MALBEK lead shield-ing and radon purge.

Figure 2.6: (Color online) MALBEK polyethyleneshielding.

[Fin08] P. Finnerty et al., TUNL Progress Re-port, XLVII, 160 (2008).

[Fin09] P. Finnerty et al., TUNL Progress Re-port, XLVIII, 156 (2009).


2.1.6 Initial Data from the Majorana Low-Background Broad-EnergyGermanium Detector at KURF

P. Finnerty, G.K. Giovanetti, R. Henning, M.A. Howe, S. MacMullin, J. Strain, J.F. Wilk-

erson, TUNL; J.I. Collar, University of Chicago, Chicago, IL; M.G. Marino, M.L. Miller,A.G. Schubert, University of Washington, Seattle, WA; R.J. Cooper, D.C. Radford, OakRidge National Laboratory, Oak Ridge, TN ; M. Yocum, Canberra Industries, Meriden, CT

We have deployed a customized broad-energy germanium (BEGe) detector, the Majorana

low-background BEGe at Kimballton (MALBEK) detector, at the Kimballton Underground

Research Facility. Currently, MALBEK has an energy threshold of ∼1 keV and a significant

low-energy continuum. We hope to further reduce both the threshold and continuum in the

near future.

As part of the research and development ef-forts for the Majorana Demonstrator we havedeployed MALBEK, a prototype BEGe detectorsimilar to one described in Refs. [Aal08, Aal10],at the Kimballton Underground Research Facil-ity in Ripplemead, VA (see Sect. 2.1.5). MAL-BEK is a 450 g natural Ge crystal housed in alow-background (LB) cryostat and outfitted withcustom made LB components. MALBEK dif-fers from commercial BEGe detectors in threerespects: (1) absence of the typical thin front-entrance window, (2) the passivation ditch radiusof 15 mm is almost a factor of two larger, and (3)the point-contact size of 4.0 mm is nearly a factorof two smaller. The reduction of the point con-tact should decrease the capacitance and hencethe electronic noise. Simulations (see Fig. 2.7)indicate that the larger ditch radius should opti-mize the charge-collection and depletion regionswithin the crystal.

The physical dimensions of the crystal are 60mm × 30 mm (diameter × height), with a 15mm ditch radius. The capacitance of MALBEKhas been measured to be ∼1.55 pF, as shown inFig. 2.8, and the noise corner is located at 6 μs,∼165 eV, as shown in Fig. 2.9.

Pulse shape analysis (PSA) of digitized wave-forms (for information on the MALBEK data ac-quisition system see Sect. 9.4.4) allows the ex-traction of useful waveform characteristics, suchas rise-time, baseline, peak height, extrema val-ues, derivative, integral, etc. This is essential inorder to remove non-physics backgrounds. Off-line event-by-event processing also enables fur-ther digital processing of the waveform, for ex-ample pulse smoothing either by wavelet de-

noising or by performing a moving average (seeFig. 2.10). The MALBEK data analysis frame-work, based on a code developed Marino and oth-ers at the University of Washington, is as follows:(1) baseline removal, (2) smoothing (waveletsor moving average), (3) pole-zero compensation,and (4) extraction of waveform characteristics.

Figure 2.7: Weighting potential (WP) [Sch38] andcharge-carrier (hole) drift paths forMALBEK. The WP quickly falls offnearly isotropically, resulting in longcharge drift-times followed by rapidcharge collection around the pointcontact (centered at the origin in thefigure). This results in distinct cur-rent pulses for each interaction.

Energy calculations are done by performing atrapezoidal energy filter (TF). We are currentlyoptimizing our choice of gap and peaking timesin terms of resolution.

A 20-day spectrum with no cuts (runs inwhich LN2 fills are not used) is shown inFig. 2.11. The detector threshold is ∼1 keV,but we expect significant improvement with theaddition of cuts based on PSA to microphon-ics and other sources of background, such as


“slow” pulses from surface interactions. Theseslow pulses have rise times on the order of 1.5-2.0μs and result from interactions near the surfacedead layer of the crystal.

Bias Voltage [V]2000 2200 2400 2600 2800 3000 3200 3400 3600

Cap

acit

ance

[p

F]

1.56

1.58

1.60

1.62

1.64

1.66

1.68

Figure 2.8: Detector capacitance vs. bias voltage(1.55 pF at depletion).

s]μShaping Time [0 2 4 6 8 10

FW

HM

[eV

]

170

180

190

200

210

Figure 2.9: Electronic noise of MALBEK as mea-sured by a pulser, showing the noisecorner located at 6 μs, ∼165 eV.

Time [ns]0 10000 20000 30000 40000 50000

Vo

ltag

e [a

rb. u

nit

s]

0

200

400

600

800

1000

Figure 2.10: (Color online) Example of waveformprocessing on a 1.73 keV event. Thesolid black line represents the base-line corrected and smoothed wave-form and the dashed black line rep-resents a 100 kHz low-bandpass fil-ter of a trapezoidal filtered waveform(gap time = 1.0 μs, peaking time =11.0 μs). Vertical bars indicate theextracted rise time start/stop of thepulse.

Figure 2.11: A 20-day spectrum for MALBEKwith no cuts. Dominant features in-clude cosmogenic peaks from K-shellelectron capture in 65Zn (8.99 keV)and 68,71Ge (10.36 keV) and from L-shell peaks from 68,71Ge (1.1, 1.29keV). The noise pedestal is locatedat ∼1 keV.

Even with only the LN2 fills removed, thecontinuum is within a factor of two of what is re-ported in Refs. [Aal08, Aal10]. Implementationof cuts to remove microphonic and surface in-teraction events will decrease the contribution tothe continuum. We will also be commissioning a∼4π muon veto and 3.81 cm of borated polyethy-lene. We plan to investigate the performanceof MALBEK at various operating temperatures.The temperature of the field effect transistor isdirectly proportional to the series noise, thereforethe colder the detector is, the smaller the contri-bution from series noise. (Parallel noise also hasa temperature component but is only relevant forresistive feedback preamplifiers.) Further refine-ment of the data analysis is also needed. We arepursuing several digital shaping methods, such asa dynamic TF that selects the optimal gap timebased on pulse rise-time on an event-by-event ba-sis.

[Aal08] C. Aalseth et al., Phys. Rev. Lett., 101,251301 (2008).

[Aal10] C. E. Aalseth et al., Arxiv preprintarXiv:1002.4703v2, (2010).

[Sch38] W. Shockley, J. Appl. Phys., 9, 635(1938).


2.1.7 Majorana Demonstrator Module Design Activities at TUNL

S. MacMullin, G. Swift, M. Busch, G. Giovanetti, W. Perkins, TUNL; J. Fast, PNNL

Many portions of the Majorana Demonstrator modules are being developed at TUNL. The

module task consists of the mechanical design of the cryostat vacuum enclosure, thermal

management system, vacuum system, mechanical supports for the Ge crystals, and wiring.

Also included in this task is designing the fabrication techniques and assembly techniques for

these subsystems.

One of the most technically challenging as-pects of the Majorana Demonstrator is tobuild the experiment using only the most radio-pure construction materials and follow an assem-bly process that leads to unprecedentedly lowbackgrounds.

2.1.7.1 String Design and Process Devel-opment

Each cryostat for the Majorana Demonstra-

tor will be made almost entirely out of ultra-pure copper electro-formed underground andultra-pure low-radioactivity PTFE plastic. Eachcryostat houses up to 35 germanium crystals in7 “strings” of up to 5 detectors in height (seeFig. 2.12. The detailed design of the detectormounts and strings is being completed at TUNLusing Solidworks 3D modeling software.

Figure 2.12: (Color online) Majorana string de-sign

In addition to the string design, the man-ufacture of these parts must be in accordancewith the background goals of the experiment. To

that end, a temporary cleanroom has been pur-chased for developing clean manufacturing tech-niques at UNC. TUNL and the UNC physicsmachine shop are collaborating to develop tech-niques for computer-numerical-controlled (CNC)machining in a cleanroom environment. Thiscleanroom will contain CNC machining equip-ment and other assembly tooling.

Building on techniques developed at PNNL,we will develop detailed procedures for clean-ing the CNC machines and operating them in away that minimizes contamination to the clean-room environment and to the materials beingmachined. Commercially available copper andplastics will be used for the majority of thesetests, but small samples of production materi-als are available to evaluate whether techniqueswill need to be modified for the productionrun. Special tools, work-holding and assem-bly fixtures are being developed to allow theclean handling and assembly of these parts. Allof these tools (including the specially cleanedCNC machines) and detailed procedures will thenbe shipped to Sanford Underground Laboratory(SUL) for the underground production run. ASodick VZ300L wire electric-discharge-machining(EDM) machine will be used in the manufactureof some of the copper string parts. A local man-ufacturer with a similar model Sodick wire EDMhas been identified and has agreed to provide cus-tom machining jigs and prototype parts in collab-oration with TUNL and the UNC physics ma-chine shop. These parts will then be shipped toSUL for installation on the Sodick VZ300L whenit is installed in the underground-cleanroom ma-chine shop.

2.1.7.2 Flange Test

Conventional vacuum seals such as indium wireand Viton O-rings cannot be used to seal the


Majorana cryostats because they can introduceradioactivity into the experiment. Alternativesconsist of clean lead or gold wire, but since bothare considerably harder than indium, it is dif-ficult to attain the large pressures required forthem to flow and seal with an all-copper cryo-stat. We have fabricated two 4-in.-diameter testcryostats made from commercial copper. Severalflanges were designed and machined into eachcryostat design to test vacuum sealing properties.

Figure 2.13: Cryostat flange test design

One design in particular, displayed inFig. 2.13, showed no signs of vacuum leak whentested with a helium leak detector. An inter-face of 2-mil Parylene-C, provided by collabora-tors at the University of Washington, was placed

at the vacuum sealing surface to reinforce theseal. Parylene-C has the advantage of having lowgas permeability and known radio-purity. A full-sized 13-in.-diameter test cryostat will be fabri-cated and tested using this design in the future.

2.1.7.3 Vacuum System Design

The vacuum system for the Majorana Demon-

strator cryostats will be developed and pro-totyped at TUNL. This vacuum system mustdemonstrate high reliability with low interven-tion, due to the high monetary value of the de-tectors and the expense associated with man-ual intervention in a deep underground clean-room. Due to the likelihood of power failures atSUL, this system must also be designed to re-cover gracefully from a short (less than 24 h)power failure, and shutdown gracefully in theevent of an extended power outage. Uniquesystem requirements include the evaluation ofUHV feedthroughs with low radon emanationand low vibration components to reduce micro-phonic noise.

2.1.7.4 Conclusions

The Majorana Demonstrator relies onachieving extremely low backgrounds. This isa difficult task and offers many unique designchallenges. TUNL is leading the effort to designand fabricate the detector strings, using materi-als and techniques that satisfy the strict back-ground criteria. In addition, we are developingthe vacuum system and seal for the Majorana

cryostat. The completion of these tasks will becrucial to the success of the Demonstrator.


2.1.8 A Monte-Carlo Background Model for the MALBEK Detector

A.G. Schubert, M.G. Marino, M.L. Miller, University of Washington, Seattle, WA; P. Finnerty,G.K. Giovanetti, R. Henning, M.A. Howe, S. MacMullin, J. Strain, J.F. Wilkerson,TUNL; P.S. Barbeau, J.I. Collar, University of Chicago, Chicago, IL; R.J. Cooper, D.C.

Radford, Oak Ridge National Laboratory, Oak Ridge, TN ; J. Orrell, Pacific Northwest Na-tional Laboratory, Richland, WA; M. Yocum, Canberra Industries, Meriden, CT

The Majorana low-background BEGe detector at KURF (MALBEK) is a custom low-background

broad-energy germanium (BEGe) detector. MALBEK was modeled in MaGe, a geant4- and

root-based simulation package. Results from MaGe simulations and radiopurity data for

materials near the detector will be combined to create a background model for MALBEK.

The Majorana Collaboration proposes asearch for neutrinoless double-beta decay (0νββ)of 76Ge using an array of enriched Ge detec-tors. The Majorana project will require anextremely low background rate to achieve highsensitivity to this decay. Radioactive decays andcosmogenically-induced radiation create back-grounds to the potential (0νββ) signal.

Figure 2.14: (Color online) The MaGe/geant4model of the MALBEK detector,shown in a cross-sectional view.

Understanding and minimizing these back-

grounds is critical to the sensitivity of theMajorana experiment. The Majorana low-background BEGe at KURF (MALBEK) detec-tor can be used to study these backgrounds.The MALBEK detector was modeled in MaGe,a geant4- and root-based Monte-Carlo sim-ulation package developed by the Majorana

and Gerda collaborations [Hen05, Ago03, Ant09].The MaGe model of MALBEK is shown inFig. 2.14.

MaGe/geant4 will be used to simulateMALBEK’s response to background radiation.The simulated response will be combined withinformation about the radiopurity of materialsin the detector and experimental hall to predictthe MALBEK background energy spectrum. Theprediction can be compared to energy spectra col-lected at KURF in order to validate the simula-tion model and guide our understanding of thebackgrounds for the Majorana experiment.

[Ago03] S. Agostinelli et al., Nucl. Instrum.Methods Phys. Res., A506, 250 (2003).

[Ant09] I. Antcheva et al., Comput. Phys. Com-mun., 180, 2499 (2009).

[Hen05] R. Henning and the Majorana Collab-oration, Nucl. Phys. B - Proc. Suppl.,143, 544 (2005), NEUTRINO 2004.


2.1.9 Creating a Muon Veto for MALBEK

J. Strain, G.K. Giovanetti, R. Henning, A. Long, J.F. Wilkerson, TUNL

MALBEK (the Majorana low-background BEGe detector at KURF) is a broad-energy germa-

nium (BEGe) detector in an ultra-low background cryostat from Canberra. One component

of MALBEK’s shielding is a muon veto made of plastic scintillator. The design and techniques

used to prepare the muon veto for MALBEK are discussed.

In January of 2010 MALBEK was commis-sioned at the Kimballton Underground ResearchFacility (KURF). The goal of MALBEK is toperform pulse-shape-analysis studies and charac-terize the prototype point-contact BEGe detec-tor for the Majorana experiment [Fin09] (seealso Sect. 2.1.6). Background suppression, andthus proper shielding, is important to achieve thisgoal. Working from the innermost to the outer-most, MALBEK’s shield includes 1 in. of Span-ish galleon ancient lead, 8 in. of low-backgroundlead, a 222Rn exclusion box, a muon veto, 1.46 in.of borated polyethylene, and 10 in. of polyethy-lene (see Sect. 2.1.5).

MALBEK’s muon veto will surround the de-tector from all four sides and from above. Figure2.15 shows a model of how the muon veto will bemounted to MALBEK’s stand.

Figure 2.15: (Color online) The scintillator canbe seen as the transparent blue pad-dles surrounding the detector’s leadshielding (green). The black cylin-ders represent the PMTs. Figuredrawn to scale.

The muon veto is composed of five plasticscintillator paddles. The scintillator material is

Saint Gobain’s BC-412 with a polyvinyltoluenebase, and the PMTs were made by Phillips. Thescintillator and PMTs were acquired as surplusfrom KARMEN [Rei05]. The 2-in.-thick scintilla-tor was delivered to UNC in 123.5 in. x 25.75 in.paddles, each with two lightguides attached, oneat each end. From each end a 33 in. x 25.75 in.piece of scintillator was cut (with the attachedlightguide) using a hand-held circular saw (seeFig. 2.16). A constant flow of clean water wasmaintained over the area being cut, to avoid over-heating the scintillator and causing it to melt orcraze.

Figure 2.16: (Color online) A piece of scintilla-tor right before it was cut. Mask-ing tape covers the area being cut tohelp protect the scintillator from be-ing further scratched and so that aline could be drawn to help guide thesaw.

Any scratching or crazing in the scintillatorcan lower the light collection efficiency. There-fore, after each piece of scintillator was cut, itwas polished to remove any scratches and to helpreduce some of the crazing.

Before any polishing was done on the scintil-lator for the muon veto, a detailed procedure wasdeveloped and the procedure was tested on small


scintillator test blocks. The polishing procedureconsisted of dry sanding with progressively finergrits, wet sanding with finer grit Micro-Mesh,and then buffing. All sides of the scintillator weredry sanded down to a 400 grit. Generally, sidescut with the circular saw were sanded with a 36,80, 220, and then 400 grit, while uncut sides weremostly just dry sanded with the 400 grit unlessdeeper scratches were found. The next step waswet sanding with a fine grit Micro-Mesh fromMicro-Surface. All of the scintillator was wetsanded with 1500, 2400, 3600, and then 6000 gritMicro-Mesh. The final steps involved buffing allof the scintillator with a random orbital sanderusing a liquid abrasive, followed by hand buffingwith an antistatic cream (see Fig. 2.17).

Figure 2.17: (Color online) A piece of scintillatorin its final polishing stage, hand buff-ing with an antistatic cream.

After the scintillator was thoroughly polished,the PMTs were glued to it using Saint Gobain’sBC-600 Optical Cement. It was then wrapped inaluminum foil and made light-tight with a blackplastic ground cover (see Fig. 2.18).

Work is now being done to calibrate thescintillator paddle’s spectrum and determine thelight collection efficiency. The scintillator signalis read out by an object-oriented software pro-gram, orca (see Sect. 9.4.4). Figure 2.19 is anoscilloscope image of a scintillator signal after ithas been through a discriminator and logic unit.

Once the scintillator spectrum has been char-acterized, future work will include bringing the

scintillator paddles underground and mountingthem on the stand for MALBEK.

Figure 2.18: (Color online) A fully polished andwrapped piece of scintillator.

Figure 2.19: (Color online) The signal for twoscintillator paddles in coincidencealong with the GATE signal.

[Fin09] P. Finnerty et al., TUNL Progress Re-port, XLVIII, 157 (2009).

[Rei05] J. Reichenbacher, Final KARMEN-Results on Neutrino-Oscillations andNeutrino-Nucleus-Interactions in theEnergy-Regime of Supernovae, Techni-cal Report Wissenschaftliche Berichte,FZKA-7093, Forschungszentrum Karl-ruhe, 2005.


2.1.10 The SEGA Detector: A 1.4 kg, Segmented Germanium IonizationDetector Enriched to 85% 76Ge

M.W. Ahmed, H.O. Back, M. Boswell, C. Coppola, R. Henning, J.D. Kephart, M.F.

Kidd, L. Leviner, W. Tornow, J.F. Wilkerson, A.R. Young, TUNL

We present progress in implementing a segmented, enriched-germanium-assembly detector

(the SEGA detector), in analyzing segmentation and pulse-shape data to reconstruct single-

site events, and in transforming SEGA into a low-background detector.

The SEGA detector is a 1.4 kg, N-type Gedetector, with 6-fold azimuthal and 2-fold axialsegmentation. This detector is unique, in thatit is also enriched to 85% 76Ge, making it thefirst enriched, segmented Ge detector. The low-background SEGA detector will serve as a test-bench for (1) engineering low-background crystalmounts and detector cryostats and (2) determin-ing known and previously unknown backgrounds(as well as methods for reducing them) for theone-tonne Majorana experiment designed toprobe Majorana-neutrino mass scales down toroughly the 20 meV level. For further details,please see http://majorana.npl.washington.edu.

The strategy to eliminate background in theMajorana ββ-decay experiment rests on four pri-mary techniques: the high resolution achiev-able with Ge-diode technology, finely segment-ing the detector, implementing pulse-shape anal-ysis (PSA), and using ultra-low radioactive back-ground materials. The first of these relies on thefact that neutrinoless ββ (0νββ) decay shouldnormally result in the deposition of a monoen-ergetic line at 2.04 MeV in our detector. Thus,the intrinsically high resolution (nominally 0.2%)of Ge detectors allows an excellent rejection ofmost backgrounds present in a typical under-ground counting experiment. Segmentation andPSA rely on the fact that 0νββ decay is a “single-site” event, in that all of the detected energy re-leased in the decay is carried by two β particleswith very small (less than 2 mm) range in the Ge.Many background events, on the other hand, aremulti-site events, and can deposit energy at sev-eral points in the crystal separated by centime-ters or more. Perhaps the most important exam-ple of these backgrounds comes from high-energyγ rays, which typically Compton scatter one ormore times before depositing their full energy inthe crystal.

SEGA’s 6-fold azimuthal and 2-fold axial seg-mentation is created with 6 azimuthally dis-tributed outer electrical contacts (S1 throughS6) and by segmenting the central electricalcontact radially in two parts (C1 and C2),thereby dividing the crystal into 12 physical seg-ments. SEGA’s unique geometry and segmenta-tion serves as a powerful tool for rejecting multi-site events; however, in order to maximize thepotential of the detector it is necessary to char-acterize each of the 12 physical segments thor-oughly. During the past year, analysis of newsource data has taken place, including studies ofelectronic noise, intrinsic peak efficiency, and sys-tem linearity.

In order to characterize the noise performanceof SEGA, we used the methods of Betuccio andPulic [Ber93] to represent the equivalent noisecharge (ENC) in a given physical segment as(ENC)2 = h1

τ + h2 + h3τ , where τ is the peak-ing time of the shaping network, and where h1,h3, and h2 are due to the series white noise, theparallel white noise, and the non-white series andparallel noises, respectively.

In order to quantify the noise sources, wevaried the peaking time of our shaping network.Typically a pulser is used to decouple the addi-tional contributing factors to the measured peakwidth, but since SEGA currently isn’t equippedwith a pulser, we chose to place a 60Co sourceabove the crystal and record data with very long(40-μs) traces. The long traces allowed us toapply an offline digital trapezoidal filter [Tan04](the same filter implemented in the XIA module),while varying the shaping parameters. This elim-inated the need for multiple data sets for whichthe shaping parameters were varied from run torun. FWHM values were measured by fittingthe 1333-keV photopeak in the trapezoidal-filter-output spectra using a Gaussian function with


a linear background. To obtain only the elec-tronic noise contribution to the photopeak width,all other components, including charge creationstatistics and charge collection efficiency, were re-moved using the coefficients obtained from theprevious resolution-versus-energy studies. The(ENC)2 vs. τ curves in Fig. 2.20 provide anunderstanding of how the three sources are con-tributing individually to the peak width.

Rise Time [microsecond]1 10

2E

NC

410

510

610 vs Rise Time2S1C2 ENC

τ/1h

2h

τ3h

Rise Time [microsecond]1 10

2E

NC

410

510

610 vs Rise Time2C2S1 ENC

τ1h

2hτ3h

Figure 2.20: (Color online) Equivalent noise-charge vs. peaking time, along withthe three individual noise compo-nents for the C2S1 segment. C2 andS1 refer to a central and outer con-tact respectively, thus labeling one ofthe twelve physical segments.

In order to assess the active volume within thedetector, the intrinsic peak efficiency was mea-sured using a 3.15 μCi 60Co source. The sourcewas placed approximately 25 cm above the topof the crystal, which was measured by scanningthe detector with a 57Co source, and data wererecorded for approximately 43 h to reduce theerror in the sum-peak to less than 5%. To de-termine the full energy deposited within the de-tector for a given event, the outer S channels(S1-S6) energies were summed together. How-ever, because the channels are electrically cou-pled, a cross-talk correction had to be applied tomulti-segment events. By analyzing the numberof counts in the sum-peak and the two 60Co γ-raypeaks, it was possible to determine the intrinsicpeak efficiencies for the 1173 keV and 1333 keVγ rays. These were 14% and 13% respectively.Sum-coincidence and angular correlation effectswere accounted for within this study, while theeffect of system dead time was avoided by usingthe sum-peak method.

In addition to the electronic noise and effi-ciency studies, the system linearity was quicklyanalyzed using the data-set from the efficiencystudy and fitting the various background andsource lines using a Gaussian function with a lin-ear background. The extent of the linearity canbe seen in the energy response curve in Fig.2.21along with the residuals from the fit.

ADC Channel

10000 20000 30000 40000

Pea

k E

ner

gy

[keV

]

500

1000

1500

2000

2500

C2S1 Energy Response

Peak Energy [keV]500 1000 1500 2000 2500

Res

idu

al [

%]

−0.005

0

0.005

0.01

C2S1 Residuals

Figure 2.21: (Color online) Energy response ofthe C2 channel for C2S1 events froma 60Co source (left) along with theresiduals from the fit (right).

These studies complete the evaluation phaseof SEGA in its current cryostat. Together withour measurements of leakage currents, the reso-lution vs. deposited γ-ray energy, peak energyvs. bias voltage, integral crosstalk, and segmen-tation performance, these studies provide a solidbasis for characterizing the basic performance ofthe SEGA detector.

The next step in the SEGA project is to trans-fer the detector into a low-background cryostatand move it to an underground counting labo-ratory. In the past year, SEGA served to exer-cise the first Majorana-dedicated electroform-ing bath at PNNL in the production of the low-background contact assembly (made entirely ofelectroformed copper and Duponts NXT-85 plas-tic). All components of this contact assemblyand a transfer cryostat have been fabricated, andthe plastic has been screened for radiopurity us-ing neutron activation analysis at NCSU’s PUL-STAR reactor to determine the uranium and tho-rium content. The test/transfer cryostat was de-signed at the University of Washington and builtat NCSU. The preamplifier electronics were de-veloped at LBNL and are being sent to TUNLfor integration into the new cryostat.

At present, we are finalizing details of thewire-routing in the test and transfer cryostatand attempting to schedule the crystal trans-fer, which will take place at AMETEK’s facilityin Oak Ridge, TN. We are also completing de-sign details of the final low-background cryostat,which has been mostly designed, and will be in-stalled in the DOE’s Waste Isolation Pilot Plantin Carlsbad, NM.

[Ber93] G. Bertuccio and A. Pullis, Rev. Sci.Instrum., 64 (1993).

[Tan04] H. Tan et al., IEEE Trans. Nucl. Sci.,51 (2004).


2.1.11 Differential Cross-Section Measurements for Elastic and InelasticScattering of Neutrons from Neon

S. MacMullin, M. Kidd, W. Tornow, R. Henning, M. Brown, C.R. Howell, TUNL

We have measured the absolute partial cross sections for elastic scattering of neutrons fromnatNe at the TUNL NTOF facility. Neutron energies were chosen to correspond to those

produced from (α,n) reactions due to 238U and 232Th decays in borated photo-multiplier-tube

glass. These measurements will help to enrich the nuclear databases, identify backgrounds in

dark-matter experiments, and refine Monte Carlo simulations and background estimates.

2.1.11.1 Introduction

The next generation of low-background physicsexperiments, including direct detection ofweakly-interacting-massive-particle (WIMP)dark matter and searches for neutrinoless double-beta decay (0νββ), may provide significant in-sight into physics beyond the Standard Model.The successful detection of WIMP interactionswill provide valuable information about the mass,cross section, and nature of dark matter. Manyof the next generation large-scale detectors de-signed to search for WIMP dark matter willmake use of large volumes of noble liquids (Ne,Ar, Xe). These experiments will search for thescintillation light generated from WIMP-nuclearscattering. Neutron-induced nuclear recoils area particularly dangerous background for theseexperiments, as it is difficult to distinguish themfrom WIMP events. Experimental data for elas-tic and inelastic cross sections of neutrons withneon were previously unavailable. We have per-formed measurements of these cross sections foruse as inputs to Monte Carlo simulations of suchneutron backgrounds in these low-backgroundexperiments, especially MiniCLEAN [Bou04].These data will also be added to existing simu-lation databases.

2.1.11.2 Experimental Details

The neutron time-of-flight (NTOF) facility atTUNL was used to measure the partial elasticand inelastic scattering cross sections of natNe.The 10 MV Tandem Van de Graaff acceleratoris capable of producing pulsed, mono-energeticneutron beams with energies of about 5-15 MeVusing the 2H(d, n)3He reaction [Hut07]. Neu-tron energies of specific interest are determinedby where the 238U- and 232Th-induced α-neutron

yields are the largest in Ar, Ne and Xe [Mei09].Neutrons are generated by bombarding a smalldeuterium gas cell with a deuteron beam pulsedat 2.5 MHz with a 2 ns width. Typical deuteronbeam current on target is about 1.0-1.5 μA.

The gas target cell is described fully in[Rup09]. It is a 21.0 mm diameter stainless steelsphere with 0.5 mm wall thickness. The cell isfilled using a coupling device connected to a highpressure filling station at TUNL. For the neon ex-periment, the pressure was maintained at around170 atm, corresponding to a steel-to-neon ratio ofabout 4:1. The cell was operated well below itsmaximum pressure rating of 550 atm. This waslimited by the ability to condense the gas in thecell at the filling station. The amount of gas inthe cell is determined by measuring the weightbefore and after filling. The pressure in the cellwas stable, with no measurable difference duringthe course of a run, sometimes up to 48 hours.

The scattered neutrons are detected by twoNE-213 and NE-218 liquid scintillation detec-tors, placed at symmetric angles with respectto the beam axis. The detectors are shieldedfrom external backgrounds by copper, lead andpolyethylene. A time-of-flight spectrum is ac-quired for each angle at each neutron energy.After each “sample in” measurement, an iden-tical run is done with an unfilled gas cell, so thatbackgrounds may be subtracted. Pulse shapediscrimination is utilized to discriminate acci-dental γ-ray backgrounds that may fall into theneutron time-of-flight window. The background-subtracted yields are corrected for dead time andmultiple scattering, and normalized to a neutroncounter which has a direct line of sight to theneutron production cell. To determine the crosssection, these yields are normalized to 1H(n, n)1Hscattering using a polyethylene (C2H4) and car-


bon scatterer. For each neutron energy, a differ-ential elastic scattering and inelastic scatteringcross section may be obtained.

Figure 2.22: (Color online) Ne(n,n) elastic-scattering differential cross sectionat 8.0 MeV. The curve is a fifth-orderLegendre polynomial fit.

[degrees]CMθ0 20 40 60 80 100 120 140 160

[m

b/s

r]Ω

/dσd 10

210

) 8.0 MeVγNe(n,n’Q = -1.634 MeV

Figure 2.23: (Color online) Ne(n,n′γ) inelasticscattering differential cross section at8.0 MeV, with the residual nucleus inits first excited state. The curve is athird-order Legendre polynomial fit.

In September 2009, an experiment was per-formed for 8.0 MeV neutrons scattering fromnatNe, and additional 8.0 MeV data were takenin March 2010. Several data points at 5.0 MeVwere taken in June 2010, and these measure-ments are scheduled to continue in August 2010.The differential cross sections for natNe(n,n) andnatNe(n,n′γ) were measured. Results are shownin Figs. 2.22 and 2.23.

2.1.11.3 Conclusions

The lack of experimental data for some neutroncross sections in Ne, Ar and Xe pose a significantproblem to dark-matter experiments that rely onMonte Carlo simulations to model expected back-grounds. Elastic and inelastic neutron-scatteringcross sections were measured for neon and willbe measured for argon and xenon in the future.These cross sections will help to enrich the crosssection databases in the neutron energy range inwhich the Monte Carlo simulations rely on data.These measurements will make new informationavailable for Monte Carlo simulations and back-ground estimates for current and future experi-ments.

[Bou04] M. Boulay et al., Arxiv preprintarXiv:nucl-ex/0410025, (2004).

[Hut07] A. Hutcheson et al., Nucl. Instrum.Methods, B261, 369 (2007).

[Mei09] D.-M. Mei et al., Nucl. Instrum. Meth-ods, A606, 651 (2009).

[Rup09] G. Rupp et al., Nucl. Instrum. Methods,A608, 152 (2009).


2.1.12 Double–Beta Decay of 150Nd to the First Excited Final State of150Sm

M. F. Kidd, J. H. Esterline, S. Finch, W. Tornow, TUNL;

We report the first measurements from the TUNL double-beta decay effort at the Kimballton

Underground Research Facility (KURF). We investigated the decay of 150Nd to the first

excited 0+ state of 150Sm by detecting the 333.9 keV and 406.5 keV deexcitation γ rays in

coincidence. After an exposure of 21.2 kg-days, we observed 21 net coincidence events, which

corresponds to a half-life of T1/2=0.76+0.33−0.18(stat)±0.04(syst)×1020 y.

Studying two neutrino double-beta (2νββ)decay is an excellent test of theoretical mod-els used to predict the half-lives of neutrino-less double-beta decay (0νββ). Therefore resultsfrom studying 2νββ decay can aid in the searchfor 0νββ decay, which in turn can provide infor-mation on the fundamental properties of the neu-trino and serve as an excellent probe for physicsbeyond the Standard Model. For example, thepresence of 0νββ decay would show that the neu-trino is its own antiparticle, (i.e., that it is aMajorana particle). Detecting 0νββ decay is themost viable experiment known to verify this con-jecture. Evidence of this process would also in-dicate violation of lepton number conservation.

In addition, 0νββ decay is currently a promis-ing method of obtaining the neutrino mass. Thehalf-life of 0νββ decay can be related to the massof the neutrino using the relation

[T 0ν1/2]−1 = G0ν |M0ν

F − (gA/gV )2M0νGT|2|〈mββ〉|2

Here G0ν is the two-body phase-space factor withthe coupling constant; |〈mββ〉| is the effective Ma-jorana electron neutrino mass; gV and gA are, re-spectively, the vector and axial-vector couplingconstants; and M0ν

F and M0νGT are the Fermi and

Gamow-Teller nuclear matrix elements, respec-tively. Extracting the mass from the half-lifemeasurement is not as straightforward as it ap-pears. Knowledge of the 0νββ decay nuclearmatrix elements is currently based on calcula-tions using techniques from quasi-particle ran-dom phase approximations, the nuclear shellmodel, and the interacting boson model. Theparameters in these calculations (in particular,the strength of the particle-particle interactiongpp) in turn are adjusted using measured half-lives from 2νββ decay. Because the behavior ofthe gpp parameter is different for transitions to

the ground state and transitions to excited states,studying 2νββ decay to excited final states is anexcellent test of the theoretical calculations.

Figure 2.24: Level scheme of 150Sm, the daugh-ter nucleus of the double-beta decayprocess of 150Nd.

To date, 2νββ decay to the ground state ofthe daughter nucleus has been observed in 10different nuclei, but 2νββ decay to excited fi-nal states has only been observed in 100Mo and150Nd. 150Nd is an excellent candidate for ββ-decay study for several reasons. Its high Q-value,even for decay to excited final states, makes it aviable isotope for the TUNL-ITEP (Institute forTheoretical and Experimental Physics) ββ-decaysetup, though very little information is currentlyavailable regarding the nuclear matrix elementsof 150Nd. Finally, 150Nd is the nuclide of choicefor the next phase of the Solar Neutrino Obser-vatory, called SNO+, which will attempt to mea-sure 0νββ decay of 150Nd. In this case, the 2νββ


Energy (keV)360 380 400 420 440

Co

un

ts

02468

1012

(a)

Energy (keV)280 300 320 340 360

Co

un

ts

02468

1012

(b)

Energy (keV)360 380 400 420 440

Co

un

ts

02468

1012

(c)

Energy (keV)280 300 320 340 360

Co

un

ts

02468

1012

(d)

Figure 2.25: (Color online) Coincidence data from 529 days of runtime. Spectra (a) and (c) are incoincidence with 333.9 keV, and (b) and (d) are in coincidence with 406.5 keV. Spectra(a) and (b) show an energy coincidence cut of ±2.5 keV while (c) and (d) have an energycoincidence cut of ±3.0 keV.

decay will be a large background factor and mustbe well-understood.

The recent result obtained by Barabash et al.for the 2νββ decay of 150Nd to the first excited 0+

state is T1/2=[1.33+0.36−0.23(stat)

+0.27−0.13(syst)]×1020 y.

This measurement was performed at Modane Un-derground Laboratory by observing single γ-raytransitions. Though an excess of counts was seenat the de-excitation energies of 333.9 keV and406.5 keV, the level of background in these mea-surements is comparatively high [Bar09]. Thebackground lines of greatest concern in this mea-surement are listed in Table 2.1 and coincide withthe γ-ray transitions of interest. The associatedisotopes are present in the natural decay chainsshown in the table. Because of the coincidencetechnique employed in the experiment describedhere, none of these decays can mimic the 2νββsignature, despite the 228Ac contamination froma 232Th impurity in our enriched 150Nd source.

50.00 grams of powdered neodymium oxide(Nd2O3) enriched to 93.60% 150Nd were leasedfrom Oak Ridge National Laboratory. This cor-responds to 42.87 g Nd metal and 40.13 g 150Nd.The Nd2O3 is sealed in an acrylic holder witha cavity measuring 5.72 cm in diameter and 0.72cm deep. Using the technique outlined in [Kid09],we used a pair of HPGe detectors surrounding thesource to detect the two coincident γ rays result-ing from the decay of the 150Sm excited 0+

1 state

via 0+1 →2+

1 →0+gs (see Fig. 2.24). The TUNL-

ITEP ββ-decay setup is located at KURF. Thedepth of the present facility is 518 meters, whichcorresponds to approximately 1450 meters waterequivalent.

Table 2.1: Background Contamination

Isotope Intensity Energy Decay(%) (keV) Chain

214Bi 0.065 333.37 238U214Bi 0.036 334.78 238U214Bi 0.169 405.72 238U228Ac 0.40 332.37 232Th227Th 1.54 334.37 238U211Pb 3.78 404.853 238U

After counting for 529 days, the spec-tra shown in Fig. 2.25 were obtained.The shaded bins are those within theevent gates. An excess of 21 events hasbeen seen, corresponding to a half-life ofT1/2=0.76+0.33

−0.18(stat)±0.04(syst)×1020 y. This islower than the result of Barabash et al. [Bar09]but still in agreement with it within the errorbars.

[Bar09] A. S. Barabash et al., eprintarXiv:0904.1924v1, 2009.

[Kid09] M. F. Kidd, J. H. Esterline, and W.Tornow, Nucl. Phys. A, 821, 251 (2009).


2.1.13 Double–Beta Decay of 150Nd to Excited Final States of 150Sm

M.F. Kidd, J.H. Esterline, S.W. Finch, W. Tornow, TUNL

During the first measurements from the TUNL double-beta (ββ) decay effort at the Kimballton

Underground Research Facility, we investigated the decay of 150Nd to the higher excited states

of 150Sm by detecting the deexcitation γ rays in coincidence. After an exposure of 21.2 kg-

days, we were able to set limits on the double-beta decay to higher excited final states of150Sm.

The ββ decay of 150Nd to higher excitedstates of 150Sm was studied using the techniquedescribed in Sect 2.1.12. The decay scheme inFig. 2.26 shows the coincidence energies for thesethree cascades to be 712.2 keV - 333.9 keV, 859.9keV - 333.9 keV, and 921.2 keV - 333.9 keV. Notethat both the 712.2 keV and 859.9 keV transi-tions are in 2+ → 2+ → 0+ decays, and the effi-ciency must be adjusted accordingly. As seen inFig. 2.27, no events were detected above back-ground for any of these decays, meaning thathalf-life limits may be set for ββ decay to thesestates. The half-life limit is given by

T1/2 >ln2 N0 t fb εtot

γγ

Nd, (2.1)

where N0 is the number of 150Nd nuclei in thesample, εtot

γγ is the coincidence efficiency, t is thecounting time, and fb is the branching ratio forthe particular cascade. The value of Nd is rec-ommended by the Particle Data Group [Par09]and is determined by a statistical estimator de-rived for a process obeying Poisson statistics. Nd

is defined as the desired upper limit on the signalabove background, and it depends on the chosenlevel of certainty. For ββ-decay half-life limits,reporting the 90% confidence level (CL) has be-come standard. For the 90% CL, Nd is the max-imum signal that will be observed in 90% of allrandom repeats of the experiment, so that thereis less than a 10% chance of observing a countrate above Nd.

The first case to be considered is the 2+2 →

2+1 → 0+

gs transition, which proceeds through theemission of a 712.2 keV γ ray in coincidence withthe 333.9 keV γ ray. The εγγ for this transition is(0.275±0.014)%, and the branching ratio is 89%.The accumulated data can be seen in Fig. 2.27(a.) and (b.). The background is 2.5 counts perkeV. The large peak seen at 707.4 keV in spec-

trum (a.) is a coincidence peak in the 228Ac de-cay to 228Th (the associated γ ray in the decayhas an energy of 332.0 keV). The peak labeled at1001.0 keV in spectrum (b.) is a Compton scat-tering peak (i.e. the energy deposited in the de-tector is 1001.0 keV minus 712.2 keV) from thedecay of 234mPa to 234U. The typical Comptonscattering coincidence (CSC) peaks mentioned inSect. 2.1.14 at 328 keV and 338 keV can also beseen. With 16 counts in the region of interest, apositive signal is not detected. The half-life limitis found to be T1/2 > 5.40 × 1019 y.

333.9

333.9

Figure 2.26: Level scheme of 150Nd ββ decay to

higher excited states of 150Sm.

The next case is the 2+3 → 2+

1 → 0+gs tran-

sition, which proceeds through the emission ofa 859.9 keV γ ray in coincidence with the 333.9keV γ ray. The accumulated data are shown inFig. 2.27 (c.) and (d.). The εγγ for this tran-sition is (0.275±0.014)%, and the branching ra-tio is 45%. The background is 0.8 counts perkeV. In spectrum (b.), another coincidence peak


Energy (keV)660 680 700 720 740 760

Co

un

ts02468

101214

(a.)

726.9 keV707.4 keV

Energy (keV)280 300 320 340 360 380

Co

un

ts

2468

101214

(b.)1001.0 keV CS

Energy (keV)800 820 840 860 880 900

Co

un

ts

2

4

6

8

10(c.)

824.9 keV

Energy (keV)280 300 320 340 360 380

Co

un

ts

0

2

4

6

8

10(d.)

Energy (keV)860 880 900 920 940 960

Co

un

ts

2

4

6

8

10(e.)

911.2 keV

Energy (keV)280 300 320 340 360 380

Co

un

ts

0

2

4

6

8

10(f.)

Figure 2.27: Coincidence data for the higher excited states. Spectra (a.) and (b.) are in coincidencewith 333.9 keV and 712.2 keV, respectively, (c.) and (d.) are in coincidence with 333.9keV and 859.9 keV, respectively, and (e.) and (f.) are in coincidence with 333.9 keV and921.2 keV, respectively. The coincidence-energy window is ± 3.0 keV.

associated with the 332.0 keV γ ray in the de-cay of 228Ac can be seen at 824.9 keV. With 5counts in the region of interest, a positive signalis not detected. The half-life limit is found to beT1/2 > 8.28 × 1019 y.

Finally, the transition 0+2 → 2+

1 → 0+gs pro-

ceeds through the emission of a 921.1 keV γ ray incoincidence with the 333.9 keV γ ray. The accu-mulated data are shown in Fig. 2.27 (e.) and (f.).The εγγ for this transition is (0.595±0.030)%,and the branching ratio is 91%. The backgroundis 0.8 counts per keV. The only peak in this spec-trum is at 911.0 keV which is a CSC peak. With7 counts in the region of interest a positive signalis not detected. The half-life limit is found to beT1/2 > 1.48 × 1020 y.

The limits determined here were also mea-

sured by Barabash et al. [Bar09], and their lim-its are about an order of magnitude better thanthose from the present work. Roughly, this canbe accounted for by comparing the exposure time× efficiency for each transition in Ref. [Bar09]and the present work. Their exposure is 1732kg-h, compared to about 509 kg-h in the presentexperiment. In addition, the efficiency in a sin-gles experiment is considerably higher than in acoincidence measurement.

[Bar09] A. S. Barabash et al., eprintarXiv:0904.1924v1, 2009.

[Par09] Particle Data Group, Probability Review,http://pdg.lbl.gov/2009/reviews/rpp2009-rev-probability.pdf, 2009.


2.1.14 Background Sources Observed in Long-Duration Counting of anEnriched Sample of 150Nd

M.F. Kidd, J.H. Esterline, S.W. Finch, W. Tornow, TUNL;

We report the backgrounds observed over 529 days of counting an enriched 150Nd sample

with high-purity germanium detectors in coincidence. Such information is valuable for other

low-background experiments, especially those involving 150Nd.

In the measurement reported in Sect. 2.1.12,a neodymium sample enriched to 93.60% 150Ndwas counted for 529 days in order to observethe two-neutrino double-beta decay. Such a longcounting period also leads to observation of back-ground radiation that is not always seen in highcount rate systems. This radiation, both naturaland due to source enrichment, can be a problemfor other low-background experiments involving150Nd, such as the next phase of the Solar Neu-trino Observatory (SNO), called SNO+.

All materials, unless processed to removethem, naturally contain some amount of potas-sium, thorium, and uranium, all of which haveisotopes that contribute to natural backgroundradiation. Since the TUNL-ITEP (Institute forTheoretical and Experimental Physics) ββ-decaysetup is located in a limestone mine, typical con-centrations of these isotopes in limestone arelisted in Table 2.2.

Radon can also be a worrisome contaminantas it is a gas and can plate out on surfaces inthe detector setup. 220Rn, 222Rn, and 219Rn arebyproducts of natural decay chains and have halflives of 55.6 s, 3.82 d, and 3.92 s, respectively.220Rn is also definitely present in the sample dueto 232Th contamination; any decay-chain con-tamination from 220Rn is in addition to the sam-ple contamination. It is unlikely that there willbe noticeable contamination from 219Rn becausethe concentration of its parent, 235U, is substan-tially smaller than that of the parent of 222Rn.

Evidence of 222Rn is found by detecting γ-ray emissions from its daughters, specifically inthe β− decay of 214Bi to 214Po. In Fig. 2.28,the characteristic 609.320 keV γ-ray emission wasmonitored over time. The decay rate seen in thedetector was corrected for the γ-ray intensity of45.49%. The radon contamination does not leadto specific coincidence events in the regions ofinterest, but it does produce some γ rays uncom-

fortably close to the region of interest (ROI). Italso increases the Compton continuum and thusthe constant background in the ROI.

Table 2.2: Typical concentrations of naturally oc-curring radioactive isotopes in lime-stone [The96].

Nuclide Radiation (Bq/kg)40K 90238U 27232Th 7

In the experiment described in Sect. 2.1.12,detection of a coincidence between 333.9 keV and406.5 keV γ rays indicated an event. Thus, themost important backgrounds are those which canmimic that event. In the 238U decay chain, themetastable state of 234mPa, which β− decays to234U, produces a weak γ ray with energy 742.8keV and intensity 0.11%. This γ-ray transition isnotable because it can Compton scatter directlyinto the ROI.

Time Elapsed (days)0 100 200 300 400 500 600 700 800 900

)-1

Co

un

t R

ate

(s

0.001

0.002

0.003

0.004

0.005

Figure 2.28: The radon contamination measured

in decays of 214Bi starting from April2008. The seasonal dependence is es-pecially noticeable.


In the 234U decay chain, though a multitudeof γ rays is produced, those which interfere withthe ROI include those associated with the β−

decay of 214Bi to 214Po. Their energies andabundances are 333.37 keV (0.065%), 334.78 keV(0.018%), 405.72 keV (0.169%), and 768.4 keV(4.895%). Because neither of the transitions ac-companied by the γ rays near 333.9 keV occur incoincidence with the one at 405.72 keV, they arenot expected to contribute to the regions of inter-est. The 768.4 keV γ ray, however, can Comptonscatter into the ROI. [Nat09].

The largest source of background from a nat-ural decay chain is the 150Nd sample itself. Thissample was leased from Oak Ridge National Lab-oratory (ORNL) where it was enriched in the ca-lutrons. The enrichment process also resultedin a contamination of 232Th decay products.This contamination manifests itself in a multi-tude of γ-ray emissions from the β−-decay of228Ac to 228Th. Three of these characteristic γ-ray emissions are in close proximity to the ROI ofthe 333.9-406.5 keV coincidence. Their energiesand intensities are 328.000 keV (2.95%), 338.320keV (11.27%), and 409.462 keV (1.92%) [Nat09],though none of these γ-ray energies occur in co-incidence with each other. However, the 328.000keV and the 338.320 keV γ rays do appear as ex-cesses of counts to the left and right of the ROIat 333.9 keV.

Further down the 232Th decay chain, the212Bi β− decay to 212Po produces a 727.3 keV γray at 6.67% intensity which can create a Comp-ton scattering peak near the ROI. This decaychain can also contribute to the areas surround-ing the regions of interest via true coincidenceswhich will form peaks in the one-dimensionalsummed event spectrum. Because the statisticsare so low, and the intensities of these coinci-dences often so small, it is difficult to verify thatthey explain any excesses of counts in the areasaround the ROI.

A rare-earth contaminant that has been foundin this 150Nd sample is 152Eu. Europium formsthe same oxide structure as neodymium, Eu2O3,and 152Eu β− decays to 152Gd and emits (amongothers) a γ ray at 344.3 keV (26.6%) in coinci-dence with a γ ray at 411.1 keV (2.24%). The en-ergy cut that defines the 406.5 keV ROI includespart of the tail of this coincidence, which thusappears in the two-dimensional spectrum. Thisseems to account for an excess of counts near 344keV.

The background discussed above cannot beremoved; its contribution in and around the ROI

must be understood. The first natural back-ground lines to be discussed are the 742.8 keVγ ray emitted in the decay of 234mPa, the 768.4keV γ ray emitted in the decay of 214Bi, and the727.3 keV γ ray emitted in the decay of 212Bi.

Examining the singles spectrum near theseenergies, it is apparent that the 742.8 keVfull-energy peak is substantially smaller thanthe surrounding 727.3 keV and 768.4 keV full-energy peaks. If the one-dimensional coinci-dence spectrum is then examined, the “full-energy Compton-scattering peaks” (i.e. peaks inthe coincidence spectrum due to a full-energy γray depositing part of its energy in one detectorin one region of interest and the rest in the otherdetector) attributed to 727.3 keV and 768.4 keVcan be identified. By comparing the intensityof these peaks to the intensity of the full-energysingles peak, it can be seen that the 742.8 keVfull-energy Compton-scattering peak would notcontribute above background in the ROI.

The other background that contributes in theareas surrounding the ROI for 333.9 keV arethe γ rays emitted in the decay of 228Ac to228Th. These have energies of 328.000 keV and338.320 keV. They contribute in the ROI whenthe full energy of a γ-ray transition is depositedin one detector in coincidence with a Compton-scattered γ ray in the other detector that doesnot deposit its full energy. Thus an excess ofcounts at 328.000 and 338.320 keV can occur.

It must then be considered whether a peakcan be expected in the other one-dimensionalspectrum from a Compton-scattering coincidencewith the 409.5 keV peak in the decay of 228Ac to228Th. It can be found that 2.9% of all suchdecays proceed in coincidence with the 328.000keV transition and with energies large enough toCompton scatter into the 406.5 keV ROI. Simi-larly, the percentage of decays in coincidence withthe 338.320 keV transition is 2.3%, and for the409.5 keV transition it is only 0.26%. Thus, itmakes sense that the excesses of counts in the328.000 keV and 338.320 keV areas are of a com-parable size, and it can be deduced that only one-tenth of that number of counts are expected at409.5 keV.

[Nat09] National Nuclear Data Center, Chart ofthe Nuclides, 2009.

[The96] P. Theodorsson, Measurement of WeakRadioactivity, World Scientific, Singa-pore, 1996.


2.2 Tritium β Decay

2.2.1 The KArlsruhe TRItium Neutrino (KATRIN) Experiment

J.F. Wilkerson, T.J. Corona, M.A. Howe, D. Phillips, TUNL; A. Kopmann, T. Bergmann,J. Wolf, Karlsruhe Institute of Technology, Karlsruhe, Germany

The KArlsruhe TRItium Neutrino (KATRIN) experiment will directly probe absolute neu-

trino masses via precise observations of the decay kinematics of nuclear β decay. Construction

of the apparatus is proceeding, with full data-taking expected in 2012. An operational data-

acquisition system has been developed and is being used for focal-plane-detector commission-

ing (Seattle) and pre-spectrometer testing (Karlsruhe). Detailed studies are also underway

of the electron transport properties of the apparatus.

The KArlsruhe TRItium Neutrino (KATRIN)experiment, a next-generation tritium β-decayexperiment with a sensitivity to sub-eV neutrinomasses, is being constructed at the Karlsruhe In-stitute of Technology (KIT), in Karlsruhe, Ger-many [Ott08, Rob08]. KATRIN employs a 10 mdiameter solenoidal retarding electrostatic spec-trometer (ΔE = 0.93 eV) coupled to a win-dowless gaseous molecular tritium source. Af-ter 5 years of data-taking, KATRIN expects toreach an estimated sensitivity of mβ = 0.20 eV(90 % C.L.) and have a discovery potential (5σ)of 350 meV. Here mβ is the neutrino mass deter-mined from β-decay measurements in the quasi-degenerate region. Full data-taking is expectedto commence in 2012.

The TUNL group is responsible for one ofthe major KATRIN tasks, data acquisition (Task50). Figure 2.29 provides an overview of the con-trols and data-acquisition systems for KATRIN.Basic machine control is implemented with pro-grammable logic controllers (PLCs), in order toensure secure operation of the tritium materialand source. The physics monitoring and controlsystem, which is independent of the PLC sys-tem, provides system monitoring, direct controland acquisition of physics data, and an interfaceto the slow control systems. There are two ap-plications: zeus for slow control and orca forfast detector readout. The zeus system, devel-oped by collaborators at KIT, is implementedin LabVIEW and integrates various PC-basedcontrollers. The Object-oriented Real-time Con-trol and Acquisition (orca) data-acquisition sys-tem (DAQ), developed and supported by UNC[How08], is responsible for controlling and read-

ing out the focal-plane detector and veto elec-tronics, while creating a data stream that ismerged with the information from the slow con-trols system.

The DAQ is capable of taking data fromthe 148-pixel detector at rates that vary frommHz (neutrino running) to MHz (calibration),as well as from a separate 32 channel veto sys-tem that runs at hundreds of Hz. All fast signalsare digitized and recorded using an electronicsdesign from the Institut fur Prozessdatenverar-beitung und Elektronik (IPE). The hardware in-corporates extensive field-programmable gate ar-rays (FPGAs) that allow acquisition either in anevent-by-event mode or with fast histogrammingof data at very high rates.

The following is a brief summary of highlightsand accomplishments during the past year. Moredetailed sections follow.

• Data Acquisition: This past year we workedwith collaborators at IPE at KIT to im-plement full support of a new version ofthe electronics system that will be used toread out the segmented PIN-diode focal-plane detector. The new Mark 4 version ofthe hardware, consisting of first level trig-ger cards, a second level trigger card, and acustom 6U crate and power supply, was de-livered from KIT to UNC in the fall of 2009.Initial work concentrated on shaking downthe FPGA firmware, the orca software,and the various acquisition modes. The op-erational system was shipped to Seattle inthe winter of 2010, where it is now beingused for commissioning of the focal-planedetector.


Figure 2.29: (color online) The KATRIN apparatus and schematic view of the experimental controland data acquisition systems.

• IPE Hardware Verification: We are under-taking a careful validation of the operationof the KATRIN Mark 4 hardware system(see Sect. 9.4.3). In addition to the systemnow in Seattle, a small mini-crate Mark 4system is being used at UNC to facilitatethese tests. Thus far the system has beenshown to meet the specifications needed forthe experiment.

• EMD Main Spectrometer Tests: We areworking towards providing a fully inte-grated DAQ system for the electromagneticdesign (EMD) tests of the main spectrom-eter scheduled for 2011. A number of re-quested run-control and scripting capabil-ities have been added and are currentlyin use as part of the focal-plane-detectorcommissioning taking place in Seattle (seeSect. 9.4.1).

• Data Structure and Database: We areworking with collaborators at Munster andIPE to define the database that will beused by KATRIN. This activity involvesa number of major groups, including theDAQ, detector, slow controls, simulations,and analysis groups. Implementation of thedatabase is proceeding, following guidelinesdeveloped and agreed to this past year.

• Focal-Plane-Detector and Veto-DetectorCommissioning: Working in collaborationwith the UW and MIT groups, we areinvolved in the focal-plane-detector com-missioning activities, and assisting whereneeded in the commissioning process. Inaddition to providing off-site DAQ support,we have made multiple trips to UW to fa-cilitate the commissioning process. We arealso working with KIT and MIT to developand implement FPGA support for the Vetodetector.

• Simulations: Detailed electromagnetic fieldcalculations are being developed to fa-cilitate our understanding of potentialbackgrounds to the observed signal (seeSects. 2.2.3 and 2.2.2).

[How08] M. Howe, M. Marino, and J. Wilkerson,In Nuclear Science Symposium Confer-ence Record, NSS ’08. IEEE, pp. 3562–3567, 2008.

[Ott08] E. W. Otten and C. Weinheimer, Rep.Prog. Phys., 71, 086201 (2008).

[Rob08] R. G. H. Robertson and KATRIN Col-laboration, J. Phys. Conf. Ser., 120,052028 (2008).


2.2.2 Automated Penning Trap Searches in the KATRIN Experiment

T.J. Corona, J.F. Wilkerson, TUNL

A possible background in the KArlsruhe TRItium Neutrino (KATRIN) experiment is from

electric breakdown due to Penning discharge, where a charged particle confined within a

Penning trap (a potential well along a magnetic field line) ionizes residual gas molecules.

Using simulation tools developed to locate potential Penning traps within a given magnet and

electrode configuration, it is possible to characterize and eliminate or minimize this source of

background.

The use of high-field (∼ 6 T) magnetics, elec-trodes held at differing potentials, and large re-gions of ultra high (∼ 10−11 mbar) vacuum in theKATRIN experiment introduced the possibilitythat potential wells may exist along the trajec-tory of a charged particle within the experimentalapparatus. These potential wells, known as Pen-ning traps, can trap charged particles, dramati-cally increasing their path length and, therefore,the probability of ionizing residual gas molecules(known as Penning discharge).

If ions emitted during Penning discharge en-ter the fiducial region of the KATRIN beamline,they can produce an aperiodic background that isdifficult to decouple from the true signal. Addi-tionally, arcing due to Penning discharge causes

fluctuations in electrode potentials and can dam-age the experimental apparatus. It is for thesereasons that it is critical to fully characterize andhopefully minimize the presence of Penning trapswithin the KATRIN experiment.

Simulation software designed to computeboth the electrostatic and magnetostatic fieldsof complicated electrode, dielectric, and magnetconfigurations can be used to locate and quantifythe dimensions of existing Penning traps withinthe KATRIN experiment (see Sect. 2.2.3. Thisinformation is useful for understanding back-ground signals due to Penning discharge and canalso help to alter design configurations to mini-mize the presence of such traps.

Tra

p d

epth

(eV

)

0

20

40

60

80

100

120

140

z (m)-2 -1.9 -1.8 -1.7 -1.6 -1.5 -1.4 -1.3 -1.2 -1.1 -1

r (m

)

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

Figure 2.30: (Color online)An example of a graphical representation of a Penning trap in an axiallysymmetric electrode system. Each point is colored according to the depth (in eV) of thePenning trap at that location, and the electrode geometry is overlayed in black.


The task of developing software to aid in Pen-ning trap characterization is ongoing within theKATRIN experiment, and is led by F. Gluck ofKarlsruhe Institute of Technology. As techniquesfor electrostatic and magnetostatic field simu-lation continue to improve, it becomes possibleto apply Penning trap characterization methodsto a broader range of geometrical configurationsand, as a result, to achieve a more accurate rep-resentation of Penning trap characteristics.

Currently, software designed to automate thedetection and characterization of Penning trapsin an arbitrary electrode, dielectric and magnetconfiguration is being developed at UNC (seeFig. 2.30). With this tool in place, it will be pos-sible to perform near-online analysis of potentialPenning traps due to the most current electrodeconfigurations, allowing operators to inspect po-tential Penning traps caused by a given electrodeconfiguration as data are being collected.


2.2.3 BEM Techniques for Computing Electric Fields in the KATRINExperiment

T.J. Corona, J.F. Wilkerson, TUNL; J.A. Formaggio, Massachusetts Institute of Tech-nology, Cambridge, MA; F. Gluck, Karlsruhe Institute of Technology, Karlsruhe, Germany

The measurement of the ν mass in the KArlsruhe TRItium Neutrino experiment requires a

thorough understanding of electrostatic fields resulting from complex electrode configurations.

The boundary element method is employed to simulate these fields for various geometries,

and the recent addition of triangular electrode primitives facilitates the simulation of general

three-dimensional geometries.

Due to the complex interplay of electrostaticand magnetostatic fields in the Karlsruhe TRI-tium Neutrino (KATRIN) experiment, specialcare must be taken to precisely understand theelectric fields resulting from various electrodeconfigurations. The characterization of thesefields is critical for determining the experiment’soverall sensitivity to the ν mass, as well as forminimizing systematic errors due to such back-ground effects as ionization from Penning dis-charge.

A suite of tools for simulating electrostaticand magnetostatic fields has been in develop-ment at Karlsruhe Institute of Technology, andthis code has been ported into C++ and devel-oped further. A critical step in the computationof electrostatic fields is the implementation ofthe boundary element method (BEM), where theLaplace equation is solved over a discretized sur-face containing mixed Neumann-Dirichlet bound-ary conditions. This technique requires an ana-lytic form for the electric potential from a uni-formly charged geometry primitive.

Using mathematical techniques based on thework of A. Birtles et. al(Ref. [Bir73]), an analyticform for the electric potential due to a uniformlycharged triangle has been developed, allowing for

the use of triangle primitives in the discretizationof electrode geometries. The addition of triangu-lar electrodes facilitates the description of a muchbroader set of three-dimensional geometries and,when used in conjunction with surface meshingalgorithms, allows for the computation of elec-trostatic fields from electrode configurations inan automated fashion.

Since both the accuracy and computationtime of the BEM are proportional to the gran-ularity of the boundary discretization, it is help-ful to employ multi-threading and grid comput-ing techniques for high-precision measurementsof electric fields. To this end, both the OpenMPAPI and Message Passing Interface (MPI) libraryspecification are used to increase computationalefficiency. Measurements for the KATRIN ex-periment that exploit these features are currentlyunderway using the UNC Topsail computer clus-ter. Tools for the design and near-online anal-ysis of the KATRIN electrode system based onthese techniques are in continued development.A sample discretized representation of KATRINelectrodes is shown in Fig. 2.31.

[Bir73] A. B. Birtles, B. J. Mayo, and A. W. Ben-nett, Proc. IEEE, 120, 213 (1973).


Figure 2.31: (Color online) A discretized representation of the electrodes in the KATRIN internaldetector region, colored by electric potential. The geometry is constructed using rectan-gular and triangular sub-elements.


2.3 Dark Matter Search

2.3.1 DEAP/CLEAN Activities

M.C. Akashi-Ronquest, L.T. Evans, R. Henning, G. Swift, TUNL

The DEAP/CLEAN project will search for weakly-ionizing-massive-particle (WIMP) dark

matter via the observation of nuclear recoils in noble liquids. TUNL activities include con-

struction of the MiniCLEAN calibration system, work on Monte Carlo simulations, and anal-

ysis of data from the DEAP-1 prototype detector.

The DEAP (Dark Matter Experiment us-ing Argon Pulse-Shape Discrimination)/CLEAN(Cryogenic Low Energy Astrophysics with No-ble Gases) collaboration’s near term goal is todevelop a dark-matter detector with a sensitiv-ity to the spin-independent WIMP-nucleon crosssection of 10−46 cm2 for a WIMP mass MWIMP ≈100 GeV. The DEAP/CLEAN concept is to ob-serve only the scintillation light from the no-ble liquid target, removing the need for time-projection chambers, which can lower the lightyield and make larger-scale experiments a chal-lenge. The ratio of prompt-to-total scintillationlight provides an excellent means to discrimi-nate background electrons from signal nuclearrecoils. Currently the collaboration is runningtwo smaller R&D detectors: MicroCLEAN andDEAP-1. MicroCLEAN, situated at Yale Uni-versity, is used to measure the scintillation char-acteristics of liquid argon (LAr) [Lip08] and liq-uid neon (LNe) [Nik08] and to study a number ofengineering issues. DEAP-1 has been used to fur-ther probe pulse-shape discrimination (PSD) inLAr [Bou09]. Having been moved undergroundto SNOLab, it is about to embark on a prelimi-nary dark-matter search.

Our involvement centers on the next step inthe DEAP/CLEAN project: MiniCLEAN. Mini-CLEAN will have a sensitivity to the WIMP-nucleon cross section of approximately 10−45 cm2

for MWIMP ≈ 100 GeV when running with a tar-get mass of approximately 500 kg of LAr. In theevent of a possible dark matter signal, the tar-get can be exchanged for LNe, which will allowthe expected A2 dependence of the cross sectionto be probed and will produce very different in-trinsic background characteristics. An analyticmodel of PSD based on DEAP-1 data [Bou09]indicates that MiniCLEAN will be able to com-pletely eliminate the dominant 39Ar background

via PSD.In addition to the focus on MiniCLEAN, we

are also involved in DEAP-1 operations, analysisand simulations. DEAP-1 is a 7 kg LAr pro-totype detector developed in order to measurebackground rejection via PSD to the 10−9 level.We also participate in the design and simulationof the DEAP-3600 detector, which will run with3600 kg of LAr. This experiment is expected tobegin taking data in 2011 and should push thesensitivity down to the collaboration’s near-termgoal. The next logical step would then be theconstruction of a 10-100 ton dark-matter detec-tor at the Deep Underground Science and Engi-neering Laboratory. This detector would also becapable of measuring the pp solar-ν flux to the1% level.

2.3.1.1 Calibration System

Our main responsibility is the design and con-struction of the calibration systems for Mini-CLEAN. During the previous year we had pur-sued the development of an articulated sourcemanipulator that would allow the placement ofa source at an arbitrary location within the sen-sitive volume in order to map out the detector re-sponse. After detailed study, we have determinedthat such a system is too complicated within thecurrent budget constraints. We have developeda more modest calibration program that consistsof:

• A transfer-module light-injection systemthat uses LEDs to inject light directly intothe sensitive volume and the PMTs. Thiswill directly determine the PMT responses.

• A dedicated input into the process systemfor injection of radioisotope spikes into the


cryogen. A candidate isotope that has beenstudied is m83Kr.

• Naturally occurring 39Ar present in the ar-gon.

• Four calibration ports that penetrate thecryostat vacuum volume but not the sen-sitive volume. Three of these ports willconsist of penetrating, insulated tubes thatare filled with water from the shield tank.Figure 2.32 shows one of these tubes, andFig. 2.33 shows the location of the tubewithin the MiniCLEAN cryostat. Thefourth tube will consist of a penetratinginsulated tube filled and purged with air,nitrogen, or a similar dry, clean cover gas.Sources will be deployed into these portsvia a mechanical winch system from thedeck through deployment tubes into theOV nozzles. The sources include standardγ-ray sources and a miniature pulsed DDneutron generator.

Figure 2.32: Concept for calibration port tube.Additional thermal insulation will bewrapped around the outside of thetube and will further mitigate thethermal load from the shield waterinside the tube. A VCR fitting and avacuum pump may be used to createand maintain an insulating vacuum.The tube assembly is about 40 cm inlength.

Figure 2.33: The calibration tube (yellow, topright) location.

2.3.1.2 Other Activities

Another thrust of our research program is the fur-ther development of the DEAP/CLEAN MonteCarlo program named rat, which is, in turn,based on geant4 and co-developed by the SNO+collaboration. Another important aspect of oursimulation work is the validation within geant4

of neutron transport. This is important in under-standing MiniCLEAN’s background due to neu-tron interactions within the liquid cryogen. Wehave participated in the ongoing validation effortwithin the collaboration.

[Bou09] M. G. Boulay et al., arXiv:astro-ph,IM/0904.2930 (2009).

[Lip08] W. H. Lippincott et al., Phys. Rev.,C78, 035801 (2008).

[Nik08] J. A. Nikkel et al., Astropart. Phys., 29,161 (2008).


2.3.2 Searches for Direct Detection of Dark Matter with a Majorana

Prototype Detector

M.G. Marino, M.L. Miller University of Washington, Seattle, WA; J.I. Collar University ofChicago, Chicago, IL; J.L. Orrell Pacific Northwest National Laboratory, Richland, WA; J.F.

Wilkerson TUNL

P-type point-contact (P-PC) germanium detectors present an exciting detector technology,yielding sub-keV thresholds and intrinsically-low electronic noise. These characteristics makethem sensitive to low-mass WIMPs and axioelectric interactions. Limits on the strength ofthese processes using data from a P-PC detector deployed at Soudan Underground Laboratoryare presented.

P-type point contact (P-PC) germanium de-tectors will be used in the Majorana Demon-

strator because their low-noise and longcharge-drift times make them well suited forbackground reduction. The low noise of thesedetectors also allows a very low threshold (∼0.5 keV), making them sensitive to low-mass(≤ 10 GeV) weakly interacting massive particles(WIMPs).

WIMP Mass [GeV/c2]

Cro

ss−

sect

ion

[cm

2 ] (n

orm

alis

ed to

nuc

leon

)

100

101

10210

−42

10−41

10−40

10−39

10−38

Figure 2.34: (color online) Constraints on spin-independent scattering cross sectionsfrom various experiments. Limitsare calculated at 90% CL. Blacklines are from these results withrise-time cuts applied with 95%(dashed) and 70% (solid) acceptanceefficiency. Other limits are fromCDMS [Ahm09a] (solid grey line)and CoGeNT, which uses a simi-lar P-PC detector [Aal08] (red dot-ted). The green shaded regions areDAMA/LIBRA acceptance regionsat 3σ and 5σ [Sav09].

A custom P-PC BEGe (broad-energy germa-nium) detector from Canberra Industries, sim-

ilar to those to be used in the Majorana

Demonstrator, was deployed underground atthe Soudan Underground Laboratory in Soudan,MN. Care was taken to reduce backgrounds, andsome critical components internal to the detec-tor cryostat were replaced with clean versions.This 0.44 kg detector was placed undergroundin a shielded environment within a 2-inch innershield of clean Pb from the University of Chicagoand an 8-inch outer shield of less radiopure lead,all inside a Rn-exclusion box. For mitigation ofneutrons from cosmic-ray interactions in the rockwalls of the laboratory, ∼ 8 inches of boratedpolyethylene was placed surrounding the radon-exclusion box. The detector has been takingdata underground since December 2009 and hasdemonstrated a threshold of less than 0.5 keV.

Figure 2.35: (color online) Example fit showingan excluded signal at 90% CL froman 8 GeV WIMP (dashed, decreas-ing blue line) and background com-ponents: L-capture lines from 65Znand 68Ge (red solid-line peaks), andan exponential plus flat background(fairly flat red dotted line). The solidblue curve is the overall fit to thedata.

Analysis techniques using wavelet de-noisingwere developed to remove slow-rise-time pulses


arising from poor charge drift and collection dueto weak electric fields near the n+ Li outer con-tact on the crystal [Aal10]. Because the two-dimensional rise-time vs. energy distribution isa priori unknown for slow-rise-time pulses, sim-ulations were used to calculate cuts with differentacceptance efficiencies for “good” pulses with fastrise-times. Since it removes events near the outersurface, the cut on slow-rise-time pulses providesan effective fiducial volume cut on the crystal,reducing the active mass to 0.33 kg.

Axion Mass (keV) 1 10

eeag

-1310

-1210

-1110

-1010

-910

sνSolar

Globular Clusters

Figure 2.36: (color online) Axioelectric limitsfrom these results (solid black curve)as well as CDMS [Ahm09b] (reddotted) and CoGeNT [Aal08] (reddashed). A corrected interpreta-tion of the DAMA acceptance region[Col09], as well as cosmological lim-its from solar neutrinos [Gon09] andred giant stars in globular clusters[Raf95] are also shown.

Limits on dark matter interactions, includ-ing nuclear recoils from WIMPs and axioelec-tric interactions from keV-mass axions, have beencalculated using data with 150.6 days of live-time. Limits on the spin-independent cross sec-tion of WIMP interactions are shown in Fig. 2.34,where current results are compared to those fromthe CDMS experiment [Ahm09a], past results

from a similar P-PC detector [Aal08], and al-lowed regions from the DAMA/LIBRA experi-ment [Sav09]. An example of an exclusion fitto a WIMP signal is shown in Fig. 2.35. Lim-its on the strength of the axioelectric effect havealso been derived from these data and are shownin Fig. 2.36, where they are compared to con-straints from other direct detection experimentsas well as those from cosmological observations.The DAMA acceptance region is currently dis-favored due to considerations of the velocity de-pendence of the axioelectric interaction [Pos08].

[Aal08] C. E. Aalseth et al., Phys. Rev. Lett.,101, 251301 (2008), Erratum: ibid.102 (2009) 109903.

[Aal10] C. E. Aalseth et al., arXiv:1002.4703,(2010).

[Ahm09a] Z. Ahmed et al., arXiv:0912.3592,(2009).

[Ahm09b] Z. Ahmed et al., Phys. Rev. Lett.,103, 141802 (2009).

[Col09] J. I. Collar and M. G. Marino,arXiv:0903.5068, (2009).

[Gon09] P. Gondolo and G. G. Raffelt, Phys.Rev. D, 79, 107301 (2009).

[Pos08] M. Pospelov, A. Ritz, and M. B.Voloshin, Phys. Rev. D, 78, 115012(2008).

[Raf95] G. Raffelt and A. Weiss, Phys. Rev.D, 51, 1495 (1995).

[Sav09] C. Savage et al., J. Cosm. Astro.Phys., 0904, 010 (2009).

Nuclear Astrophysics

Chapter 3

• Nucleosynthesis in Hydrostatic and Explosive Environments

• 26Al in the Interstellar Medium

• Thermonuclear Reaction Rates

56 Nuclear Astrophysics TUNL XLIX 2009–10

3.1 Nucleosynthesis in Hydrostatic and Explosive Environ-ments

3.1.1 Measurement of 17O(p,γ)18F Between the Narrow Resonances at Elabr

= 193 and 519 keV

J.R. Newton, C. Iliadis, A.E. Champagne, J.M. Cesaratto, S. Daigle, R. Longland,TUNL

The 17O(p,γ)18F reaction is important to nucleosynthesis in several stellar sites, including

classical novae. These are close binary systems which exhibit thermonuclear explosions. The

explosions attain temperatures between 0.1 and 0.4 GK, where the thermonuclear reaction

rate of 17O(p,γ)18F is currently believed to be dominated by direct capture. We measured the

cross section at six energies between the resonances at Elabr = 193 and 519 keV. Our extracted

direct-capture cross section differs significantly from previous, less accurate measurements.

Hydrogen burning of 17O is important for avariety of stellar environments such as red giants,AGB stars, massive stars, and classical novae.The latter occur in a close binary system, wherethe accretion of hydrogen-rich material onto thesurface of a white dwarf results in a thermonu-clear explosion. The temperature range of inter-est in these explosions is T = 0.1-0.4 GK. Be-cause 17O represents a major branch point in theCNO cycles, the relative reaction rates for the17O(p,γ)18F and 17O(p,α)14N reactions are cru-cial to our understanding of the nucleosynthesisof the oxygen isotopes and the production of ra-dioactive 18F in these sites [Fox05].

In recent progress reports, we reported onpreliminary experimental results. Here, webriefly summarize our final findings. The ex-periment was carried out at LENA using the JNVan de Graaff to supply protons ranging in en-ergy from Elab

p = 275 to 500 keV. The targetswere made by anodizing 91% enriched 17O wateronto 0.5 mm thick tantalum backings. 18O tar-gets were also used to calibrate the beam energyand detection efficiencies using the well knownElab

r = 151 keV resonance in 18O(p,γ)19F. Alead shield was designed to almost completelysurround the large volume HPGe detector by atleast 5 cm of lead on all sides. The detector is in-serted into the lead shield at an angle of 55◦ withrespect to the incident beam direction at a dis-tance of 3.5 cm from the target. Data were takenat six different energies with this arrangement.Because of the close running geometry, coinci-dence summing corrections had to be performed

on all spectra.Results are displayed in Fig. 3.1 and show

astrophysical S-factors in the relevant bombard-ing energy range. Measured total S-factors fromRefs. [Rol73, Cha07] and from the present workare shown. The data points from Ref. [Rol73]gave rise to the previous estimate of the direct-capture S-factor (dotted line), also from Ref.[Rol73]. The single data point by Chafa et al.[Cha07], obtained via the activation technique,has a rather large uncertainty bar and is not use-ful for constraining the direct-capture S-factor.The present data clearly show the influence ofthe tails of two broad resonances at Ecm

r = 557and 677 keV. Our measured total S-factors canbe used to extract the direct-capture contribu-tion. To this end, we subtract the calculated S-factor arising from the two well-known broad res-onances from our measured total S factor. Theextracted direct-capture S-factor is nearly energyindependent and is well described by a constantS-factor below Ecm = 500 keV: SDC(E) = 4.6keV b (≈23%).

Our extracted direct-capture S-factor dis-agrees with the result of Rolfs (dotted line) by afactor of ≈2. We improved the uncertainty of thedirect-capture S-factor from an assumed value of±50% in Ref. [Rol73] to an experimental valueof ±23% in the present work. Our results arein good agreement with the predictions of Ref.[Fox05].

This work has now been published. For de-tailed results, see Ref. [New10]. We are now inthe process of performing reaction network simu-

TUNL XLIX 2009–10 Nuclear Astrophysics 57

Figure 3.1: Astrophysical S-factors at low energies in 17O(p,γ)18F. The data from the present work,corrected for efficiencies and coincidence summing, are shown as circles. The two verticallines mark the energy region of primary interest for classical novae. The dotted line isthe prediction of Rolfs [Rol73]. The dashed lines show the direct-capture and resonancecontributions from Fox et al. [Fox05], and the solid line is their sum.

lations for classical novae using our newly derivedreaction rates for the 17O+p reactions.

[Cha07] A. Chafa et al., Phys. Rev. C, 75,035810 (2007).

[Fox05] C. Fox et al., Phys. Rev. C, 71, 055801(2005).

[New10] J. Newton et al., Phys. Rev. C, 81,045801 (2010).

[Rol73] C. Rolfs, Nucl. Phys., A217, 29 (1973).


3.1.2 Resonance Strength in 22Ne(p,γ)23Na from Depth Profiling in Alu-minum

R. Longland, C. Iliadis, J.M. Cesaratto, A.E. Champagne, S. Daigle, J.R. Newton

TUNL

A novel method for extracting absolute resonance strengths has been investigated. By im-

planting 22Ne ions into a thick aluminum backing and simultaneously measuring the 22Ne+p

and 27Al+p reactions, the strength of the Elabr = 479 keV resonance in 22Ne(p,γ)23Na was

determined to be ωγ = 0.524(51) eV. This result has significantly smaller uncertainties than

earlier work. Our results are important for the absolute normalization of resonance strengths

in the 22Ne(p,γ)23Na hydrogen-burning reaction and in the 22Ne+α s-process neutron-source

reactions.

Nuclear reaction rates of many importantstellar-burning processes are dominated by res-onances in the reaction cross section. In or-der to improve the accuracy in these rates, re-cent evaluations [Ili10] have determined reso-nance strengths relative to a backbone of accu-rately measured standard strengths [Ili07]. Thosestrengths have been determined relative to elasticscattering, where many systematic effects, suchas those from detector efficiency, beam charge,and target stoichiometry, cancel out. Conse-quently standard resonance strengths measuredin this way depend only on the measured countrates and Rutherford scattering cross sections.However, as is apparent from Table 1 in Ref.[Ili01], there are no proposed standard resonancestrengths for proton-induced reactions involvingnoble-gas targets.

The 22Ne+p reaction is one example, whichis important for hydrogen-burning, particularlyin the Ne-Na cycle. Many resonance strengthsfor capture reactions involving noble gases havebeen measured previously using implanted tar-gets, where in most cases the target backingwas a thin tantalum sheet. In reactions suchas 22Ne+p, it is notoriously difficult to obtainabsolute strengths because the stoichiometry ofthe implanted region must be known in orderto extract a strength from a thick-target exci-tation function. There is currently one measuredabsolute strength in 22Ne(p,γ)23Na correspond-ing to the Elab

r = 1278 keV resonance [Kei77].For this resonance, the stoichiometry of the im-planted 22Ne targets was obtained with Ruther-ford back-scattering, but the measurement hasnever been independently verified. In addition,

this resonance is beyond the reach of most low-energy accelerators, necessitating the need for astandard resonance strength at low energy.

Here, we report on an absolute resonancestrength measurement of the Elab

r = 479 keV res-onance in 22Ne(p,γ)23Na. The experiment uti-lizes a target composed of 22Ne ions implantedinto an aluminum substrate. By measuringthe well-known Elab

r = 406 keV resonance in27Al(p,γ)28Si and the Elab

r = 479 keV resonancein 22Ne(p,γ)23Na simultaneously, the absolutestrength of the latter resonance can be obtained,independent of the absolute beam current, abso-lute detector efficiency, or target stoichiometry.Obtaining a resonance strength independent ofthe target stoichiometry is a significant advan-tage of this method. Our new method is notspecific to 22Ne(p,γ)23Na but could also be use-ful for other reactions. Note that an accuratelymeasured 22Ne(p,γ)23Na resonance strength willalso improve estimates of α-particle-induced re-actions on 22Ne, since the targets for the lattercases were previously characterized using 22Ne+presonances [Gie93].

In brief, we implanted 22Ne ions into athick aluminum sheet and measured the yieldsof the narrow resonances at Elab

r = 406 keVin 27Al(p,γ)28Si and at Elab

r = 479 keV in22Ne(p,γ)23Na simultaneously. Since the im-planted 22Ne ions are concentrated near the sur-face of the aluminum sheet, we expect a well-defined peak-shape for the 22Ne+p yield curve.On the other hand, the 27Al+p yield curve willreveal an interesting structure. In the pure alu-minum region, beyond the implanted 22Ne depth,the 27Al+p yield will be at a maximum. How-


ever, in the 22Ne implanted region the yield willbe reduced. Therefore, we expect a step in theexcitation function caused by the implanted re-gion near the surface of the target. The generalstrategy was the following: (i) fit the measured27Al+p yield curve, including the step on thefront edge, allowing us to extract the stoichiom-etry; (ii) with the stoichiometry and the abso-lute yield normalization factor determined fromthe previous step, fit the measured 22Ne+p yieldcurve with the resonance strength in 22Ne+p leftas a free parameter.

All previously measured strengths for theElab

r = 479 keV resonance in 22Ne+p were nor-malized relative to higher-lying resonances in thesame reaction. In our work, we have obtained theresonance strength independently of other reso-nances in 22Ne+p, with no dependence on abso-lute detector efficiencies or absolute beam-chargeintegration. The result is ωγ = 0.524(51) eV.We have significantly improved the uncertainty ofthe resonance strength from the previous value ofaround 30% [Kei77] to 10%. Our new techniqueremoves any systematic uncertainty due to thetarget stoichiometry, which is difficult to quantifyusing traditional methods. It was already men-tioned above that our new value for the Elab

r =479 keV resonance strength in 22Ne+p is impor-tant in two respects. First, it will reduce the

rate uncertainties of the 22Ne(p,γ)23Na reaction,since the strengths of the low-energy resonancescan be renormalized relative to our precisely mea-sured strength for Elab

r = 479 keV, thus improv-ing predictions of hydrogen-burning nucleosyn-thesis. Secondly, our precise strength can beused to determine more reliable stoichiometriesfor implanted 22Ne-Ta targets, which have beenemployed in measurements of the important s-process neutron-source reactions involving 22Ne.

This work has now been published. For de-tailed results, see Ref. [Lon10].

[Gie93] U. Giesen et al., Nucl. Phys., A561, 95(1993).

[Ili01] C. Iliadis et al., Astrophys. J. Suppl.,134, 151 (2001).

[Ili07] C. Iliadis, Nuclear Physics of Stars,Wiley-VCH, Weinheim, 2007.

[Ili10] C. Iliadis et al., Nucl. Phys., A841, 31(2010).

[Kei77] J. Keinonen and M. Riihonen, Phys.Rev. C, 15, 579 (1977).

[Lon10] R. Longland et al., Phys. Rev. C, 81,055804 (2010).


3.1.3 Target Development for a Measurement of 23Na(p,γ)24Mg at Astro-physical Energies

J.M. Cesaratto, A.E. Champagne, T.B. Clegg, C. Iliadis, TUNL

In preparation for a measurement of the 23Na(p,γ)24Mg reaction at energies relevant for

novae, significant effort was devoted to target development. Sodium was implanted into four

different metal backings using an ion implanter located on the UNC campus. In this report

we describe the motivation for investigating implanted targets and the characteristics of these

targets. The conclusions is that this type of target is inferior to evaporated targets in terms

of overall yield and structure.

Low-energy charged-particle reactions impor-tant for nuclear astrophysics have exceedinglysmall cross sections. Measuring these crosssections requires using intense beam currents,high-efficiency detection systems, and isotopi-cally pure targets. The former two requirementsare fulfilled at LENA, while the latter can be ac-complished at UNC using an Eaton positive-ionimplanter. This implanter is capable of produc-ing beams from both gaseous and solid materials.More details about the implanter can be found in[Cot06, Ces09].

A measurement of the expected but un-observed resonance at Elab

R = 144 keV in23Na(p,γ)24Mg prompted the need for high yield,uniform, and stable targets. The decision topursue implanted 23Na (hereafter referred to asNa) targets was two-fold. First, previous re-sults [Seu87] for Na implanted into nickel re-ported both uniform structure and high yield.The authors state that a pure Na layer could beachieved with 1.1 × 1018 atoms/cm2 at a bom-barding energy of 30 keV. Second, a previousstudy of 23Na(p,γ)24Mg showed that evaporatedtargets were not stable under proton beam loadsof > 50 μA. Evaporated targets composed ofNaCl, NaBr, and Na2Ti3O7 all proved to be un-stable under such beam loads [Row04]. How-ever, targets evaporated with Na2WO4 provedsufficiently stable for that experiment as well forother experiments [Zij90]. In the previous mea-surement, typical beam currents on target werelimited to 50 μA to minimize target degradationand preserve the life of the target. However, thatmeasurement failed to meet its goal in observingthe Elab

R = 144 keV resonance in 23Na(p,γ)24Mg.It was determined that more beam current inci-

dent on target was needed in order to boost thesignal rate over the background.

It was believed that an increase in beam cur-rent would surely cause these evaporated targetsto degrade, so implanted targets seemed like awiser choice to produce more robust targets withhigh yield. Sodium was implanted into severaldifferent backings: Ta, Ni, Cu, and Ni evaporatedonto Ta (NiTa). Target backing dimensions (inmm) were 38 x 38 x 0.5 thick. The goal for thethickness of the active target material was 15 keVat a proton energy of 144 keV. Stopping powersfrom srim [Zie08] were used to calculate the nec-essary dose and energy required for the implantinto each target backing. The implanted regionhad a diameter of 25 mm, as determined by a col-limator located 0.5 m upstream in the implanterbeam line. Table 3.1 shows a listing of targetbacking and associated implant parameter for 15keV thick targets with assumed stoichiometry of1 to 1 (active to backing). The calculated dosesassumed no losses of implanted particles.

Table 3.1: Backings used for Na implantationswith associated Na bombarding ener-gies for 15 keV thick targets (at Ep =144 keV) and calculated dose assuminga ratio of active to backing atoms of 1:1

Eb implanted doseBacking (keV) (1018 ions/cm2)

Ta 94 0.15Ni 51 0.3Cu 54 0.3

After implantation, the targets were testedfor yield (stoichiometry) and uniformity by prob-ing the 309 keV resonance in 23Na(p,γ)24Mg.Thick-target yield curves (excitation functions)were measured over the thickness of the targets.


Excitation curves of Na implanted into differenthost backings are shown in Fig. 3.2. For eachof these targets, the yield over the implanted re-gion is not uniform. For the Ta and Cu backings,Na seemed to migrate to the surface of the tar-get, whereas for the Ni backing, the migrationseemed to be further into the backing. Becauseof the non-uniform nature of the targets, the sto-ichiometry of the target changed depending onthe depth into the target. At the maximum, thestoichiometry for Na in a Ni backing was Na:Ni= 1:1. In Ta, it is not appropriate to estimate thestoichiometry based on the narrow surface peak.However, the stoichiometry based on the regionpast the surface peak, was Na:Ta = 1:6.

300 310 320 330 3400

0.05

0.1

0.15

0.2

γ-ra

y yi

eld

(arb

. uni

ts) 23

Na in Ta

300 310 320 330 3400

0.02

0.04

0.06

0.08 23Na in Cu

300 310 320 330 340Energy (keV)

00.10.20.30.40.5 23

Na in Ni

300 310 320 330 3400

0.1

0.2

0.3

0.423

Na in NiTa

Figure 3.2: Thick-target yield curves for the reac-

tion 23Na(p,γ)24Mg at Er = 309 keV.Sodium was implanted into Ta, Cu,Ni, and Ni evaporated onto Ta back-ings (NiTa). The lines connecting thedata points are included to guide theeye.

We tried different methods to obtain the mostuniform targets: implanting at different energiesand varying the dose. We learned that it was pos-sible to over-implant the targets, actually end-ing up with less active target material than westarted with. Changing energies did not improve

the quality of the implant.Although the Ni backing produced the best

target, we were not able to reproduce the resultsof Ref. [Seu87]. We were able to reproduce thestructure for Na implanted into Ta, which alsocompared well in structure what is described ina new paper from the University of Washington[Bro09].

Although much effort went into this devel-opment, it was ultimately decided that Na-implanted targets would not be suitable for mea-surement because of the non-uniformity and lowyield exhibited for Na implanted into Ni, Ta, Cu,and Ni evaporated onto Ta. Ultimately, we re-sorted to targets evaporated with Na2WO4 forthe experiment, since these targets were far moreuniform, had higher yield, and showed the bestresiliency to degradation in previous measure-ments.

[Bro09] T. A. D. Brown et al., Nucl. Instrum.Methods, B267, 3302 (2009).

[Ces09] J. M. Cesaratto, T. B. Clegg, and A. E.Champagne, TUNL Progress Report,XLVIII, 135 (2009).

[Cot06] A. Cotrell and C. Ugalde, TUNLProgress Report, XLV, 113 (2006).

[Row04] C. Rowland, Ph.D. thesis, University ofNorth Carolina at Chapel Hill, ChapelHill, NC, 2004.

[Seu87] S. Seuthe et al., Nucl. Instrum. Meth-ods, A260, 33 (1987).

[Zie08] J. F. Ziegler and J. P. Biersack, SRIM:The Stopping Range of Ions in Mat-ter, computer program srim availableat http://www.srim.org, 2008.

[Zij90] F. Zijderhand et al., Nucl. Instrum.Methods, A286, 490 (1990).


3.1.4 Resonant Proton Capture on 23Na at Stellar Burning Temperatures

J.M. Cesaratto, M.Q Buckner, A.E. Champagne, T.B. Clegg, S.M. Daigle, C. Iliadis,B.M. Oginni TUNL

The motivation for and details of a new search for the predicted 144 keV resonance in the23Na(p,γ) reaction at stellar burning temperatures are presented. Measurements were carried

out at LENA using twenty times the proton beam current of previous measurements. The

data are being analyzed.

Globular clusters are simple stellar popula-tions that are assumed to contain stars of thesame age and chemical composition. They havebeen used as a testing ground for stellar evolu-tion and nucleosynthesis. All galactic globularclusters studied have shown star-to-star elemen-tal variations of light elements (C, N, O, Na, Mg,and Al) in their surface compositions, pointing tothe simultaneous operation of the CNO, NeNa,and MgAl cycles of hydrogen burning [Car08].This is contrary to what is seen in halo field stars,which do not exhibit the extreme elemental vari-ations of globular clusters [Ven06]. There existsa pronounced anti-correlation between O and Naamong evolved RGB stars [Kra97] as well as un-evolved stars slightly above and below the mainsequence turn-off in globular clusters [Gra01].The origin of these variations is thought to ariseeither from evolutionary effects within a star—effects involving extra physics beyond standardstellar theory—or from primordial differences inthe material from which a star was created.

The 23Na+p reaction is a branching pointwhere material can be either recycled back intothe NeNa cycle with emission of an α particle,or advanced to the MgAl cycle with emission ofa γ ray. The competing reaction rates betweenthe two exit channels determine the most proba-ble path for material to follow. Figure 3.3 showsthe ratio of the reaction rates of the two com-peting exit channels. The uncertainties of thecurrent reaction rate of 23Na(p,γ)24Mg are dom-inated by an expected but unobserved resonanceat Elab

r = 144 keV. In the temperature regions forthe environments of interest, it is apparent thatthe rate is uncertain by several orders of magni-tude.

An explosive environment where the23Na(p,γ)24Mg reaction influences the yields of

other elements is in classical novae. A classi-cal nova is a binary star system comprised of awhite dwarf and a main sequence star. As thewhite dwarf accretes hydrogen-rich material fromits companion onto its surface, a thermonuclearoutburst occurs. A study completed in Iliadis etal. [Ili02] showed that the 23Na(p,γ)24Mg reac-tion affects many abundances for ONe core novamodels. It has also been thought that classi-cal novae might contribute to the production of26Al, since type II supernovae cannot producethe amount detected in the interstellar medium[Pra04].

RGB

AGB

Classical Novae

Figure 3.3: (Color Online) Current reaction rate

ratio for the 23Na(p,α) and 23Na(p,γ)reactions. The solid line representsthe ratio of recommended rates forthe two exit channels. The upper andlower dashed lines represent the un-correlated uncertainties of the rates.The astrophysical sites relevant forthis reaction and their associated hy-drogen burning temperatures are alsoshown.

The current upper limit of the resonancestrength for this reaction is ωγ ≤ 0.15 μeV


[Row04]. A new search for this resonance wasconducted at LENA utilizing the high-beam-intensity ECR ion source. The ECR ion sourcewas able to produce >1 mA of proton beam ontarget, an increase in beam current by a factor of20 compared to the previous measurement. Re-sults of this measurement are forthcoming.

[Car08] E. Carretta, Memorie della Societa As-tronomica Italiana, 79, 508 (2008).

[Gra01] R. G. Gratton et al., Astron. Astro-phys., 369, 87 (2001).

[Ili02] C. Iliadis et al., Astrophys. J. Suppl.Ser., 142, 105 (2002).

[Kra97] R. P. Kraft et al., Astron. J., 113, 279(1997).

[Pra04] N. Prantzos, Astron. and Astrophys.,420, 1033 (2004).

[Row04] C. Rowland et al., Astrophys. J., 615,L37 (2004).

[Ven06] P. Ventura and F. D’Antona, Astron.Astrophys., 457, 995 (2006).


3.1.5 The Effects of Thermonuclear-Reaction-Rate Variations on Nucle-osynthesis in Classical Novae

L.N. Downen, C. Iliadis, A.E. Champagne, TUNL

The effects of thermonuclear-reaction-rate variations on the production of 20Ne, 28Si, 31P, and32S in classical novae were investigated. To determine the sensitivity of nucleosynthesis to

these variations, post-processing nucleosynthesis calculations were performed on temperature-

density-time profiles obtained from hydrodynamic nova simulations. The goal of this research

is to demonstrate the extent to which nova nucleosynthesis calculations rely on uncertain

thermonuclear reaction rates and to guide future measurements.

A classical nova occurs in a binary star sys-tem consisting of a main sequence star and awhite dwarf. A thermonuclear explosion occursas a result of accretion of hydrogen-rich materialonto the white dwarf. The isotopic abundancesobserved in nova ejecta are of particular impor-tance for constraining models of stellar evolutionand explosion [Sta98, Sta00].

A total of 12,720 nuclear network reactioncalculations were performed. Six temperature-density-time profiles of the hottest hydrogen-burning zone were obtained from hydrodynamicnova simulations and used for our calculations.These profiles pertain to white dwarfs primar-ily composed of oxygen and neon (ONe) withpeak temperatures ranging from 0.231 to 0.567GK and masses from 1.15 to 1.35 solar masses[Ili02]. STARLIB, a new reaction rate librarybased on a recent evaluation of experimentalMonte-Carlo reaction rates, provided most of therecommended reaction rates for 1 MK through10 GK, the temperature range of this study[Ili10, Lon10].

The post-processing approach used in thisresearch is capable of completing a single net-work calculation in less than a minute, whereasdirectly-coupled reaction network and hydro-dynamic calculations require significantly morecomputing time. Decoupling the hydrodynamicsand the reaction network effectively ignores con-vection and its effect on final nova abundances[Ili02]. For this reason, this approach cannotbe used to calculate accurate absolute isotopicabundances, but it can be used to obtain quanti-tative results on abundance changes. Therefore,the recommended rates for 265 reactions were in-dividually multiplied by factors of 0.01, 0.1, 0.2,0.5, 2, 5, 10, and 100 for each profile. Reac-

tion rates were analyzed for their impact on theabundances of 20Ne, 28Si, 31P, and 32S by com-paring the final abundance of an isotope calcu-lated using the recommended reaction rate withthe abundance calculated using the rate multi-plied by a factor. The three reactions with thestrongest impact on the nucleosynthesis of eachof these isotopes are listed in Table 3.2.

Table 3.2: Significant reactions given by isotope.

Isotope Reaction20Ne 20Ne(p,γ)21Na

23Na(p,α)20Ne23Na(p,γ)24Mg

28Si 28Si(p,γ)29P20Ne(p,γ)21Na23Na(p,α)20Ne

31P 31P(p,γ)32S28Si(p,γ)29P30P(p,γ)31S

32S 28Si(p,γ)29P31P(p,γ)32S32S(p,γ)33Cl

A sensitivity study for the effects of reaction-rate variations on the isotopes of interest hasbeen completed for reactions resulting in a fi-nal abundance change of 20% or more. Cur-rently, the relationships between the isotopicabundances are being investigated, as are the ef-fects that reaction rate changes and peak tem-peratures have on these relationships. We planto submit this work for publication by late fall2010.

[Ili02] C. Iliadis et al., Astro. J. Suppl. Ser.,142, 105 (2002).


[Ili10] C. Iliadis et al., unpublished article,2010.

[Lon10] R. Longland et al., Nucl. Phys., A841,1 (2010).

[Sta98] S. Starrfield et al., Mon. Not. R. Astron.Soc., 296, 502 (1998).

[Sta00] S. Starrfield et al., Astro. J. Suppl. Ser.,127, 485 (2000).


3.1.6 Hydrodynamic Models of Type I X-Ray Bursts: Metallicity Effects

J. Jose, F. Moreno, A. Parikh, Escola Universitaria d’Enginyeria Tecnica Industrial de Barcelona,Barcelona, Spain; C. Iliadis, TUNL

Type I x-ray bursts (XRBs) are thermonuclear stellar explosions driven by charged-particle

reactions. In the regime for combined H/He-ignition, the main nuclear flow is dominated by

the rp-process, the 3α-reaction, and the αp-process. We performed a detailed hydrodynamic

study of the nucleosynthesis and nuclear processes powering type I XRBs. In total, 11 bursts

have been computed by means of a spherically symmetric (one-dimensional) Lagrangian hy-

drodynamic code, linked to a nuclear reaction network that contains 325 nuclides (from 1H

to 107Te) and 1392 nuclear processes. These evolutionary sequences, followed from the onset

of accretion up to the explosion and expansion stages, have been performed for two different

metallicities to explore the dependence between the extension of the main nuclear flow and

the initial metal content.

Type I X-ray bursts (XRBs) are cataclysmicstellar events. They are powered by thermonu-clear runaways (TNRs) in the H/He-rich en-velopes accreted onto neutron stars in close bi-nary systems. These events constitute the mostfrequent type of thermonuclear stellar explosionin our galaxy, in part because of their short recur-rence period (hours to days). About 90 galacticlow-mass X-ray binaries exhibiting such burst-ing behavior have been found since the discov-ery of XRBs. Type I XRBs and their associatednucleosynthesis have previously been extensivelymodeled, reflecting the astrophysical interest indetermining the nuclear processes that power theexplosion, determining the light curve, and pro-viding reliable estimates for the chemical compo-sition of the neutron star surface.

With a neutron star hosting the explosion, thetemperatures and densities in the accreted enve-lope reach high values (109 K and ρ = 106 g cm3).As a result, detailed nucleosynthesis studies re-quire the use of hundreds of nuclides, linked bythousands of nuclear interactions, extending allthe way up to the SnSbTe-mass region. Becauseof computational constraints, XRB nucleosynthe-sis studies have traditionally been performed us-ing limited nuclear reaction networks. Until re-cently, it has not been possible to couple hydro-dynamic stellar calculations (in one dimension)and detailed networks. This has prompted a de-tailed analysis of the nuclear activity poweringthe bursts.

Clearly, the identification of the most relevantreactions in the mass region A = 65-100 remains

to be addressed in detail within the framework ofhydrodynamic simulations. This is particularlyrelevant since proton captures on heavy nucleihave a dramatic effect on the shape of XRB lightcurves. To this end, we computed a new set oftype I XRBs using a one-dimensional, sphericallysymmetric, hydrodynamic, implicit, Lagrangiancode(see Jose and Hernanz 1998 [Jos98]). Thecode has been linked to a fully updated nu-clear reaction network containing 324 nuclidesand 1392 nuclear processes, including the mostrelevant charged-particle-induced reactions oc-curring between 1H and 107Te, as well as theircorresponding reverse processes. Experimentalrates are available for a small subset of reactions(adopted from Angulo et al. 1999 [Ang99], Il-iadis et al. 2001 [Ili01], and some recent updatesfor selected reactions). For all other reactions forwhich experimental rates are not available, weused the rates from Hauser-Feshbach codes.

Computations of evolutionary sequences, fol-lowed from the onset of accretion up to the explo-sion and expansion stages, have been performedfor two different metallicities to explore the de-pendence between the extension of the main nu-clear flow and the initial metal content. Wecarefully analyzed the dominant reactions andthe products of nucleosynthesis, together withthe physical parameters that determine the lightcurve (including recurrence times, ratios betweenpersistent and burst luminosities, and the extentof the envelope expansion). Results are in quali-tative agreement with the observed properties ofsome well-studied bursting sources. Production


of 12C (and its implications for the mechanismthat powers superbursts)and light p-process nu-clei, and the amount of H left over after the burst-ing episodes have also been investigated.

This extensive work has now been published.For detailed results, see Ref. [Jos10].

[Ang99] C. Angulo et al., Nucl. Phys., A656, 3

(1999).

[Ili01] C. Iliadis et al., Astrophys. J. Suppl.,134, 151 (2001).

[Jos98] J. Jose and M. Hernanz, Astrophys. J.,494, 680 (1998).

[Jos10] J. Jose et al., Astrophys. J. Suppl., 189,204 (2010).


3.2 26Al in the Interstellar Medium

3.2.1 Theoretical Evaluation of the Reaction Rates of 26Al(n,p)26Mg and26Al(n,α)23Na

B.M. Oginni, A.E. Champagne, C. Iliadis, TUNL;

The reactions that destroy 26Al have significance in a number of astrophysical contexts. The

reaction rates of 26Al(n,p)26Mg and 26Al(n,α)23Na have been evaluated using cross sections

obtained from the nuclear reaction codes empire and talys. These have been compared with

the rates obtained from the non-smoker code, the compilation of Caughlan and Fowler, and

some experimental data. We show that the results obtained from empire and talys are com-

parable and similar to those from non-smoker, whereas the rates tabulated in Caughlan and

Fowler are systematically lower.

The nucleus 26Al β-decays to the first excitedstate of 26Mg, which emits a γ ray of 1809 keV.These γ rays were first detected by the HEAO3satellite [Mlw84], and also by the Solar Maxi-mum Mission Satellite [Skk85], the Gamma-RayImaging Spectrometer [Tbg91], and others. Sub-sequently, an all-sky image of 26Al γ-ray emissionat 1809 keV was derived from the COMPTELinstrument on board the Compton Gamma RayObservatory [Pds01, Ilc07]. This map providesinformation about star formation and nucleosyn-thesis over a time period comparable to the half-life of 26Al (7.17 × 105 y).

The observed overabundance of 26Mg in mete-orite materials is attributed to the decay of 26Al,which implies that the latter was present duringthe formation of the meteorites [Sww07]. Since,meteorites were formed during the early stagesof the solar system, this observation should pro-vide information about the last nucleosynthesisevents that contributed material to the solar neb-ula. The abundance of 26Al depends largely onthe reactions which lead to its production anddestruction. The main production mechanismin a variety of sites is the 25Mg(p,γ)26Al reac-tion, while the main destruction mechanisms athigher temperatures are the 26Al(n,p)26Mg and26Al(n,α)23Na reactions [Wwg01].

In this study, we have used the codes em-

pire [Hcc07] and talys [Khd08] to evaluate thecross sections for the 26Al(n,p) and 26Al(n,α)reactions. The cross sections were then usedto evaluate reaction rates, which have beencompared with the results from non-smoker

[Rat00, Rat01] and Caughlan and Fowler (here-

after, cf88) [Cfo88].

0 2 4 6 8 10

108

ETRCF

0 2 4 6 8 10Temperature, T

9 (GK)

106

107

108

Rea

ctio

n ra

tes

(cm

3 mol

-1 s

-1)

ETRCF

26Al(n,p)

26Al(n,α)

Figure 3.4: (Color online) (upper) 26Al(n,p) reac-tion rates calculations using EMPIRE(E), TALYS (T), NON-SMOKER[Rauscher] (R) and Caughlan (CF);(lower) similar calculations for26Al(n,α).

In Fig. 3.4, we show the reaction rates for26Al(n,p)26Mg and 26Al(n,α)23Na as calculatedby the empire, talys, non-smoker codes andas compiled by Caughlan. In the 26Al(n,p) case,the ratios of the maximum reaction rate to theminimum reaction rate for the empire, talys

and non-smoker results were less than a factorof 1.2 at temperatures from 0.2 GK to 10 GK.When the Caughlan results were included, theratios increased but were still generally less than2. Similar ratios were determined for 26Al(n,α)for the empire, talys and non-smoker results.


The ratios varied from 1.7 to 3 between 0.2 GKand 10 GK, but the ratios increased to between2.5 and 3.4 when the cf88 results were included.

This result indicates that the Caughlan re-sults for these reactions were underestimated,at least when compared with the other reactioncodes used in this study. This is a very impor-tant observation, because the Caughlan compi-lations are often used in nucleosynthesis calcu-lations. Koehler et al. [Kkv97] made a similarobservation. Their measurements of the reactionrates for the 26Al(n,α) reaction were two to threetimes larger than the rates given by Caughlanover much of the relevant range of nucleosynthe-sis calculations. The reaction rates from empire,

talys, and non-smoker are about the same fac-tor larger than rates from Caughlan.

0 0.01 0.02 0.03 0.04 0.05E

n (MeV)

101

102

cros

s se

ctio

n (m

b)

EMPIRETALYSSmet

Figure 3.5: (Color online) A comparison betweenthe Maxwellian averaged cross sec-tion for 26Al(n,α0 + α1)23Na and cal-culations using the empire and talyscodes.

Only a few experimental data are avail-able for the reactions we studied, and most ofthem are partial cross sections, i.e., the pro-duction cross section for a certain energy levelin a nucleus. However, Smet [Sww07] evalu-ated the Maxwellian averaged cross section for26Al(n,α0 +α1)23Na, where 0 and 1 in the α sub-

script indicate the ground state and first excitedstate, respectively, of the residual nucleus. Wecompare those results with our calculations fromempire and talys in Fig. 3.5. Though, the datafall within a small energy range, there was consis-tency between the data and the two theory codesin at least three of the six data points.

[Cfo88] G. R. Caughlan and W. A. Fowler,At. Data Nucl. Data Tables, 40, 283(1988).

[Hcc07] M. Herman et al., Nucl. Data Sheets,108, 2655 (2007).

[Ilc07] C. Iliadis, Nuclear Physics of Stars,Wiley-VCH, 2007.

[Khd08] A. J. Koning, S. Hilaire, and M. C. Dui-jvestijn, In Proceedings of the Interna-tional Conference on Nuclear Data forScience and Technology, p. 211, 2008.

[Kkv97] P. E. Koehler et al., Phys. Rev. C, 56,1138 (1997).

[Mlw84] W. A. Mahoney et al., Astrophys. J.,286, 578 (1984).

[Pds01] S. Pluschke et al., In Space De-bris, Proceedings of the Third EuropeanConference, volume ESA SP 459, p. 55.European Space Agency, 2001.

[Rat00] T. Rauscher and F. Thielemann, At.Data Nucl. Data Tables, 75, 1 (2000).

[Rat01] T. Rauscher and F. Thielemann, At.Data Nucl. Data Tables, 79, 47 (2001).

[Skk85] G. H. Share et al., Astrophys. J., 292,L61 (1985).

[Sww07] D. L. Smet et al., Phys. Rev. C, 76,045804 (2007).

[Tbg91] B. J. Teegarden et al., Astrophys. J.,375, L9 (1991).

[Wwg01] J. Wagemans et al., Nucl. Phys.,A696, 31 (2001).


3.2.2 The Effects of Thermonuclear Reaction-Rate Variations on 26AlProduction in Massive Stars

C. Iliadis, A.E. Champagne TUNL; A. Chieffi Istituto Nazionale di Astrofisica - Istituto di As-trofisica Spaziale e Fisica Cosmica, Roma, Italy; M. Limongi Istituto Nazionale di Astrofisica -Osservatorio Astronomico di Roma, Monteporzio Catone, Italy

We investigate the effects of thermonuclear reaction-rate variations on 26Al production in mas-

sive stars. The dominant production sites in such events were recently investigated by using

hydrodynamic simulations and were found to be explosive neon-carbon burning, convective

shell carbon burning, and convective core hydrogen burning. Post-processing nucleosynthesis

calculations are performed for each of these sites by adopting temperature-density-time pro-

files from recent hydrodynamic models. For each profile, we individually multiplied the rates

of all relevant reactions by factors of 10, 2, 0.5, and 0.1 and analyzed the resulting abundance

changes of 26Al. Our simulations are based on a next-generation nuclear physics library, called

STARLIB, which is partially based on a new evaluation of Monte Carlo reaction rates.

The radioisotope 26Al is of outstanding im-portance for γ-ray astronomy and cosmochem-istry. It has been discovered in three distinctsites: (i) the galactic interstellar medium, viadetection of its decay emission line at 1809 keV[Mah82, Die95]; (ii) meteorites, via observed ex-cesses of its radioactive-decay (daughter) prod-uct 26Mg [Mac95], implying an injection of live26Al into the early solar system nebula; and(iii) presolar dust grains that are uncontami-nated by solar system material and thus are oflikely stellar origin, again via observed 26Mg ex-cesses [Hop94, Hus97]. Identification of the mainsources of 26Al would have far-reaching implica-tions, ranging from questions related to the cir-cumstances and conditions of the solar system’sbirth to imposing strong constraints on the chem-ical evolution of the galaxy. A number of differentsources have been suggested over the years: AGBstars, classical novae, Wolf-Rayet stars, and typeII supernovae.

However, the origin of 26Al remains contro-versial. The observation of galactic γ-rays from26Al is important since it provides unambiguousdirect evidence for the theory of nucleosynthe-sis in stars. The half-life of 26Al amounts to7.17×105 y and is small compared to the timescale of galactic chemical evolution (≈1010 y).Consequently, nucleosynthesis is currently occur-ring in the interstellar medium and, in particular,26Al is synthesized throughout the galaxy. Fromthe observed γ-ray intensity, it is estimated thatthe production rate of 26Al in the galaxy, depend-

ing on the assumed density distribution, amountsto ≈ 1 − 3 M� per 106 y. The observational evi-dence in this case favors massive stars as a source.First, the all-sky map of the 1809 keV γ-ray linedetected by the COMPTEL instrument onboardCGRO showed that 26Al is confined along theGalactic disk and that the measured intensity isclumpy and asymmetric; second, the measure-ment of the 1809-keV-line Doppler shift by theSPI spectrometer onboard INTEGRAL demon-strated that 26Al co-rotates with the galaxy andhence supports a galaxy-wide origin.

Massive stars may produce 26Al during sev-eral different phases of their evolution: (i) dur-ing pre-supernova stages in the C/Ne convectiveshell, where a fraction of the 26Al survives thesubsequent explosion and is ejected into the in-terstellar medium; (ii) during core collapse viaexplosive Ne/C burning; and (iii) in Wolf-Rayetstars, i.e., stars with masses in excess of about 30M�, during core hydrogen burning, where 26Almay appear via convection at the surface andsubsequently be ejected by strong stellar winds.These 26Al production mechanisms were recentlyanalyzed in detail by Limongi and Chieffi [Lim06]by using extensive hydrodynamic simulations ofsolar metallicity massive stars. In that work theyalso emphasized the impact of rate uncertaintiesfor selected reactions on the final 26Al yields.

In the present work we expand this effortby presenting a comprehensive investigation ofthe impact of nuclear reaction-rate uncertaintieson the synthesis of 26Al. The general strategy


consists of varying the rates of many reactionsby different factors (in this work, 10, 2, 0.5,and 0.1) and analyzing the impact of each in-dividual reaction-rate change on the final 26Alyields. At present it is not feasible to performthis computationally intensive procedure with aself-consistent stellar model. Instead, we ex-tract a few representative temperature-density-time evolutions from recent hydrodynamic mod-els of massive stars and execute a large number ofpost-processing reaction-network sensitivity cal-culations using these profiles. Our goal is two-fold. On the one hand, we would like to quantifyto what degree predictions of 26Al yields dependon currently uncertain nuclear-physics input. Onthe other hand, by identifying the most importantnuclear reactions, our results represent a guidefor future measurements.

There are a number of novel aspects about thepresent work. First, we employ a new-generationlibrary, called STARLIB (see Sect. 3.3.1),of nuclear-reaction and weak-interaction rates.This library is partially based on a recent evalu-ation of experimental Monte Carlo reaction rates[Ili10]. In fact, this work represents the first ap-plication of STARLIB. Second, we carefully in-vestigate the equilibration effects of 26Al. Atleast two species of 26Al, the ground state andthe isomeric state, take part in nucleosynthe-sis. In all previous investigations, either a sin-gle species or two species of 26Al were taken intoaccount, depending on whether thermal equilib-rium is achieved or not. Obviously, these are twoextreme assumptions, and in a hot stellar plasmathe ground and isomeric state may communicatevia γ-ray transitions involving higher-lying 26Allevels.

Our study has some obvious limitations.Since we perform post-processing calculations,we necessarily focus our investigation on the ef-fects of nuclear reaction rates. In other words,the important effects of convection, mass loss,rotation, and so on are outside the scope of thepresent work. This also implies that our simu-lations are unsuitable for defining absolute 26Alyields. Instead, we claim that our procedure isuseful for exploring the effects of 26Al abundancechanges that result from reaction rate variations.We only explore a few temperature-density-timeevolutions that are representative of solar metal-licity stars. A more comprehensive study cover-ing a broad range of stellar masses and metallic-ities is also beyond the scope of this work. Thepresent investigation is ongoing, and first resultswill be presented soon.

[Die95] R. Diehl et al., Astron. Astrophys.,298, 445 (1995).

[Hop94] P. Hoppe et al., Astrophys. J., 430, 870(1994).

[Hus97] G. Huss et al., Geochim. Cosmochim.Acta, 61, 5117 (1997).


[Lim06] M. Limongi and A. Chieffi, Astrophys.J., 647, 483 (2006).

[Mac95] G. MacPherson et al., Meteoritics, 30,365 (1995).

[Mah82] W. Mahoney et al., Astrophys. J., 262,742 (1982).


3.3 Thermonuclear Reaction Rates

3.3.1 STARLIB: A Next-Generation Reaction-Rate Library for NuclearAstrophysics

C. Iliadis TUNL

We have recently completed a new charged-particle thermonuclear-reaction-hrate evaluation

for A = 14-40 target nuclei. These results were obtained using an entirely different approach

from the one used in previous evaluations. A Monte-Carlo technique was employed, with

special emphasis on reliability and the presentation of meaningful reaction-rate uncertainties.

This data set represents the backbone of a new kind of reaction-rate library for nuclear

astrophysics. The library is presently under construction.

The reaction-rate library most widely usedfor stellar model simulations is called REACLIB,originally constructed by F.-K. Thielemann. Anumber of updated versions exist. Reaction ratelibraries such as REACLIB typically contain therate for nuclear reactions or the decay constantfor weak decays on a specified temperature grid.In the particular case of the REACLIB, the infor-mation is available as an analytical (fit) expres-sion.

We recently began construction of a new kindof library, called STARLIB. It originated from aprevious version of REACLIB that we modifiedover the years and that was used for all of thereaction network calculations presented in Ref.[Ili07]. At that point in time, several importantchanges occurred. A recent evaluation of reac-tion rates for the A = 14-40 target range wascompleted and is scheduled for publication in Oc-tober of 2010 [Ili10]. These 62 experimental ratesare based on a Monte Carlo technique, describedin Ref. [Lon10], allowing for a rigorous statis-tical definition of recommended reaction ratesand their associated uncertainties. The MonteCarlo procedure provides, for the first time, forany given temperature the (output) reaction-rateprobability-density function that is based on the(input) probability densities of measured nuclear-physics quantities such as S-factors, resonanceenergies, resonance strengths, and upper limitsin spectroscopic factors. From the cumulativedistributions of the rate probability densities, alow rate, median rate, and high rate can thenbe defined as the 0.16, 0.50 and 0.84 quantiles,respectively, assuming a coverage probability of68%. The meaning of these rates in general dif-fers from the commonly reported, but statisti-

cally meaningless, literature expressions “lowerlimit,” “nominal value,” and “upper limit” ofthe total reaction rate. Furthermore, it hasbeen shown [Lon10] that in the majority of casesthe Monte-Carlo-rate probability-density func-tion can be approximated by a lognormal dis-tribution, that is determined by only two param-eters: the lognormal location parameter μ andthe lognormal spread parameter σ. The formerparameter determines the recommended reactionrate via NA 〈σv〉rec = eμ, while the latter param-eter corresponds to the rate factor uncertaintyvia f.u. = eσ.

The information on the rate probability den-sity was not available previously and opens inter-esting windows of opportunity for Monte Carlostudies of nucleosynthesis and energy genera-tion in stars. However, it becomes clear fromthe above discussion that three quantities (T ,NA 〈σv〉rec, lognormal σ) instead of the tradi-tional two (T and NA 〈σv〉rec) need to be re-ported, so that the user can calculate the rateprobability density for each reaction at each tem-perature. Therefore, it was decided to convertour 2007 version of the REACLIB, which lists therecommended reaction rates as analytical func-tions of temperature by employing a number ofrate fitting parameters, to a tabular format. Thenew format consists of three columns and lists foreach reaction the temperature, the recommendedrate, and the rate factor uncertainty on a grid of60 temperatures between 1 MK and 10 GK. Atthis stage, the rate factor uncertainty for eachreaction was set equal to a nominal value of 10.In a subsequent step, the rates and factor un-certainties of 62 reactions in the A = 14-40 re-gion were replaced with their exact Monte Carlo


results. In addition, the rates of the followinginteractions were replaced with more recent in-formation: (i) 10 Big Bang reactions, using therates of Ref. [Des03], which were derived froman R-matrix description of the available data;(ii) 30 reactions in the mass range of A = 1-26 from the NACRE evaluation of experimentalrates [Ang99],; (iii) (n,γ) reactions based on theKADoNiS v0.2 evaluation of experimental rates[Dil06]; (iv) a number of special reactions, suchas 14N(p,γ)15O and 12C(α,γ)16O; and (v) 550experimental rates for β decays and β-delayedparticle decays, including associated uncertain-ties calculated from the half lives and branchingratios compiled in Ref. [Aud03]. All experimen-tal reaction rates were corrected for the effectsof thermal target excitations using the stellarenhancement factors and partition functions ofRauscher and Thielemann [Rau00]. For all otherreactions for which insufficient experimental in-formation is available to compute reliable exper-imental rates, the results of statistical model cal-culations [Rau00] were adopted. The present ver-sion of the new library described above extendsup to antimony. A number of features still needto be implemented. Publication of this library,which is expected to become the foundation of

many stellar models (e.g., for massive stars, AGBstars and classical novae) is anticipated for thenear future.

[Ang99] C. Angulo et al., Nucl. Phys., A656, 3(1999).

[Aud03] G. Audi, A. H. Wapstra, and C.Thibault, Nuclear Physics A, 729, 337(2003).

[Des03] P. Descouvemont et al., At. Data Nucl.Data Tables, 88, 1 (2003).

[Dil06] I. Dillmann et al., In AIP Conf. Proc.,volume 819, p. 123, 2006.

[Ili07] C. Iliadis, Nuclear Physics of Stars,Wiley-VCH, Weinheim, 2007.


[Lon10] R. Longland et al., Nucl. Phys., A841,1 (2010).

[Rau00] T. Rauscher and F.-K. Thielemann, At.Data Nucl. Data Tables, 75, 1 (2000).

Few-Nucleon Interaction Dynamics

Chapter 4

• Sub-Nucleonic Degrees of Freedom

• The A=2 System

• The A=3 System

• Reaction Dynamics of Light Nuclei

76 Few-Nucleon Interaction Dynamics TUNL XLIX 2009–10

4.1 Sub-Nucleonic Degrees of Freedom

4.1.1 Measurement of Single Target-Spin Asymmetry in Semi-InclusivePion Electroproduction on a Transversely Polarized 3He Target

W. Chen, H. Gao, M. Huang, X. Qian, Y. Qiang, X. Zong, TUNL; and others from Isti-tuto Nazionale di Fisica Nucleare (INFN) Roma-I, INFN Roma-III, INFN Bari, and University ofBari, Italy; Thomas Jefferson National Accelerator Facility, Newport News, VA; Temple University,Philadelphia, PA; Rutgers University, Piscataway, NJ ; University of Illinois, Urbana-Champaign,IL; University of Virginia, Charlottesville, VA; College of William and Mary, Williamsburg, VA

A first measurement of single-target spin asymmetry (SSA) from semi-inclusive electro-

production of charged pions from a transversely polarized 3He target in deep-inelastic scat-

tering kinematics was completed in Hall A at Thomas Jefferson National Accelerator Facility

in early 2009. Such SSAs will allow the extraction of much-desired information on the Collins

and Sivers asymmetry from the “neutron” in order to probe the quark transversity distribu-

tions. The Medium Energy Group at TUNL has been playing a leading role in this experi-

ment. Preliminary results on the SSAs from 3He have been reported at various conferences

and workshops this summer. A first paper reporting on these results will be submitted to

Physical Review Letters in the near future.

There are eight transverse-momentum-dependent distribution functions (TMDs) atleading twist [Mul96, Boe98]. TMDs describethe parton distribution beyond the collinear ap-proximation. They depend not only on the lon-gitudinal momentum fraction x, but also on thetransverse momentum kT of the quark. Amongthese eight TMDs, three survive upon integra-tion over the quark transverse momentum. Theyare f1, g1 and h1.

After several decades of experimental andtheoretical efforts, the unpolarized parton distri-bution function (PDF), f1(x, Q2), has been ex-tracted with excellent precision over a large rangeof x from deep inelastic scattering (DIS), Drell-Yan, and other processes. In recent years, thelongitudinal polarized parton distribution func-tion, g1(x, Q2), has been determined with signif-icantly improved precision over a large region ofx and Q2 from polarized DIS experiments. Whatremains elusive among these three leading-twistparton distribution functions is the transversityfunction, h1(x, Q2). In the parton picture, thenucleon transversity distribution describes thenet quark transverse polarization in a trans-versely polarized nucleon. In the non-relativisticlimit, the transversity distribution function is thesame as g1(x, Q2). Therefore, the transversitydistribution function probes the relativistic na-

ture of the quarks inside the nucleon.

Because of its chiral-odd nature, the exper-imental determination of transversity requiresprocesses where an additional chiral-odd func-tion is involved. Double-polarized Drell-Yan pro-cesses, which involve two transversity distribu-tions, and single target-spin azimuthal asym-metries from semi-inclusive DIS pion electro-production, in which transversity is convolutedwith a chiral-odd fragmentation function, theCollins function [Col93], are two methods tomeasure transversity. The Collins fragmentationfunction describes the probability for a trans-versely polarized quark to fragment into an un-polarized hadron.

Considering the differential cross section ina single-spin semi-inclusive DIS (SIDIS) (e, e′h)reaction with a transversely polarized target,the transversity function convoluted with theCollins fragmentation function gives rise to anazimuthal asymmetry in sin(φh + φS), wherethe azimuthal angles of both the hadron (φh)and the target spin (φS) are about the virtualphoton axis and relative to the lepton scatter-ing plane. Two more leading-twist TMDs con-tribute to the single-spin azimuthal asymmetry:the Sivers function [Siv91], which gives rise tothe azimuthal asymmetry of sin(φh −φS) due tothe correlation between the nucleon’s transverse

TUNL XLIX 2009–10 Few-Nucleon Interaction Dynamics 77

spin and the quark’s transverse momentum, andthe pretzelosity, with a sin(3φh − φS) asymme-try in SIDIS, which measures the quark polariza-tion perpendicular to both the transverse polar-ization of the nucleon [Ava08] and the longitu-dinal direction. The orthogonal nature of thesethree azimuthal asymmetries gives them an un-ambiguous separation in SIDIS. The HERMESCollaboration [tHC05] and the COMPASS Col-laboration [tCC05, tCC07, tCC08] reported re-sults on single-spin asymmetries from SIDIS ontransversely polarized proton and deuteron tar-gets. The asymmetry related to the Collins frag-mentation function and the quark transversitydistribution was also measured by the BELLECollaboration [tBC06] in the e+e− → h1h2X re-action. In order to extract the transversity forthe u and d quarks, a measurement on the neu-tron is essential.

The first experiment on the SIDIS neutronsingle-spin asymmetry using transversely polar-ized 3He finished taking data in February 2009 inJefferson Lab Hall A. Details on the experimentalconditions are given in Ref. [Che09]. Xin Qian, agraduate student from Duke, worked on this ex-periment for his Ph.D. thesis. He was in charge ofthe large acceptance BigBite spectrometer. Hiswork included the background study, wire cham-ber preparation, tracking-related software devel-opment, on-line data quality check, wire-chamberalignment, and BigBite optics calibration. TheBigBite wire chamber successfully survived fromthe extreme 60 MHz rate with a 15 μA electronbeam, and the optics calibration has reached thedesign goal. Yi Qiang, a former postdoctoral fel-low in our group, also played a key role by lead-ing the polarized-3He-target effort for this exper-iment as well as for a number of experiments em-ploying the polarized 3He target after the neutron

transversity experiment. Since the experiment,both Yi and Xin have been actively involved inthe data analysis, with Xin leading the overallanalysis effort of the entire collaboration. Re-cently, Xin successfully defended his Ph.D. the-sis based on his work on this experiment. Xin isalso the lead author on the first paper from thisexperiment on SSAs, which will be submitted toPhysical Review Letters this year.

[Ava08] H. Avakian et al., Phys. Rev., D78,114024 (2008).

[Boe98] D. Boer and P. Mulders, Phys. Rev.,D57, 5780 (1998).

[Che09] W. Chen et al., TUNL Progress Report,XLVIII, 53 (2009).

[Col93] J. Collins, Nucl. Phys., B396, 161(1993).

[Mul96] P. Mulders and R. Tangerman, Nucl.Phys., B461, 197 (1996).

[Siv91] D. Sivers, Phys. Rev., D43, 261 (1991).

[tBC06] the Belle Collaboration, Phys. Rev.Lett, 26, 232002 (2006).

[tCC05] the COMPASS Collaboration, Phys.Rev. Lett., 94, 202002 (2005).

[tCC07] the COMPASS Collaboration, Nucl.Phys., B765, 31 (2007).

[tCC08] the COMPASS Collaboration, Proceed-ing for Transversity 2008 Workshop,Ferrara, (2008).

[tHC05] the HERMES Collaboration, Phys.Rev. Lett., 94, 012002 (2005).


4.2 The A=2 Systems

4.2.1 Final Neutron-Proton Analyzing Power Data at En=7.6 MeV

W. Tornow, TUNL; R.T. Braun, Duke University; G.J. Weisel, Penn State Altoona, PA

We present new corrections for the polarization dependent efficiency of previously reported

n-p Ay(θ) data at 7.6 MeV. These data indicate that only ”complete” angular distributions

of Ay(θ) data provide a stringent test of phase-shift analysis and nucleon-nucleon potential

model predictions. They confirm our earlier conclusion that an accurate determination of the

low-energy 3Pj nucleon-nucleon interactions has not been achieved yet.

The motivation for collecting high-precisionneutron-proton (n-p) analyzing power, Ay(θ),data is to test the accuracy of nucleon-nucleon(NN) phase-shift analysis (PSA) predictions forthis observable. PSA results are often thedatabase for fitting NN potential models parame-ters. At low energies, the n-p Ay(θ) is very small,typically well below 1%, and it depends greatlyon the 3P0, 3P1 and 3P2 NN phase shifts. Thehigh-precision data of Holslin et al. [Hol88] atEn=10.03 MeV are in very good agreement withthe Nijmegen PSA results [Sto93]. However, theHolslin et al. data set contains only one datumat angles below 65◦ c.m., while the more accuratedata of Braun et al. [Bra08] at 12.0 MeV extendall the way down to 30◦ c.m., where consider-able discrepancies have been observed betweenthe PSA predictions of the Nijmegen group andthe experimental data. This observation indi-cates that only ”complete” angular distributionsof the n-p Ay(θ), including sufficient forward an-gle data points, provide an accurate comparisonof PSA predictions and experimental data. Be-sides the 12 MeV n-p Ay(θ) angular distributionreferred to above, only the 7.6 MeV data set ofBraun [Bra98] covers a large enough angularrange to be considered ”complete”.

This is the reason why considerable effort hasbeen spent on the polarization dependent effi-ciency (PDE), which is endemic to the organicscintillator based neutron side detectors used inn-p Ay(θ) experiments. In each side detector, po-larized neutrons can scatter first from a carbonnucleus and then from a proton. As is knownfrom the work of Galati et al. [Gal72], the Ay(θ)of 12C(n, n)12C shows large fluctuations with en-ergy in the range between 3 and 7 MeV. Becauseof the kinematic variation of the neutron energies

over the face of the side detectors, the 12C-p dou-ble scattering events can introduce a false asym-metry into the n-p Ay(θ) measurement [Hol89].This false asymmetry does not cancel with theusual spin-flip technique used in Ay(θ) measure-ments. The only way that the PDE effect canbe accounted for is by using Monte-Carlo tech-niques, which employ an accurate n-12C scat-tering library. In order to improve this library,Roper et al. [Rop05] gathered 33 angular distri-butions of 12C(n, n)12C Ay(θ) data from 2.2 to8.5 MeV. Further details about the 12C(n, n)12Clibrary used in the present work are given in[Bra08, Wei10].

0 20 40 60 80 100 120 140 160 180θ

c.m. (deg)

0.000

0.002

0.004

0.006

0.008

0.010

Ay(θ

)

7.6 MeV

Figure 4.1: The n-p Ay(θ) data at En=7.6 MeV.The crosses use the original PDE cor-rections, while the circles use the PDEcorrections of the present work. Thesolid curve is the Nijmegen PSA pre-diction.

Results for the n-p Ay(θ) at En=7.6 MeV af-ter the PDE corrections are displayed in Fig. 4.1.Here the crosses use the original PDE correction[Bra98], while the circles use the PDE correctionsof the present work. The difference between thetwo corrections is very small for scattering an-


0 20 40 60 80 100 120 140 160 180θ

c.m. (deg)

0.000

0.005

0.010

0.015

0.020

Ay(θ

)

12.0 MeV

Figure 4.2: The n-p Ay(θ) data at En=12.0 MeV of [Bra08]. The solid curve is the Nijmegen PSAprediction.

gles larger than 80◦ c.m. However, for the for-ward angular region the new corrections providea smoother angular dependence of Ay(θ) than ob-tained with the original PDE corrections. Thesolid curve shown in Fig. 1 is the Nijmegen PSAprediction. In agreement with our previous find-ings at 12 MeV [Bra08] shown in Fig. 4.2, theNijmegen PSA prediction overestimates the n-pAy(θ), especially at forward angles. The samestatement is true for all high-precision NN po-tential models. This observation may well be oneof the reasons for the well-known three-nucleonanalyzing power puzzle [Wit94].

[Bra98] R. Braun, Ph.D. thesis, Duke Univer-sity, 1998.

[Bra08] R. Braun et al., Phys. Lett. B, 660, 161

(2008).

[Gal72] W. Galati et al., Phys. Rev., 5, 1508(1972).

[Hol88] D. Holslin et al., Phys. Rev. Lett., 61,1561 (1988).

[Hol89] D. Holslin et al., Nucl. Instrum. Meth-ods A, 274, 207 (1989).

[Rop05] C. Roper et al., Phys. Rev. C, 72,024605 (2005).

[Wei10] G. Weisel et al., Phys. Rev. C, 82,027001 (2010).

[Wit94] H. Wita�la et al., Phys. Rev. C, 49, R14(1994).


4.3 The A=3 Systems

4.3.1 Cross-Section Measurement of 2H(n,np)n at 16 and 19 MeV in Sym-metric Constant-Relative-Energy Configurations

A.H. Couture, T.B. Clegg, C.R. Howell, A.S. Crowell, J.H. Esterline, B. Fallin,TUNL; B.J. Crowe, D.M. Markoff, L. Cumberbatch, K. Agbeve, North Carolina CentralUniversity (NCCU), Durham, NC ; R.S. Pedroni, North Carolina A&T State University, Greens-boro, NC ; S. Tajima, TUNL and NCCU ; H. Wita�la, Jagiellonian University, Krakow, Poland

Cross-section measurements of neutron-deuteron (nd) breakup were made at incident neutron

energies of 16 and 19 MeV. The scattered proton was detected in coincidence with one of the neu-

trons. The energies of the scattered particles were determined via time-of-flight. Neutron beam flux

normalization is obtained from nd elastic scattering performed concurrently with the main experi-

ment. Our current measurements are of two special cases of the symmetric constant-relative-energy

(SCRE) configuration: the space-star (SST) and the coplanar star. We have completed analysis

of the 16 MeV data, which confirms the space-star anomaly established by previous measurements.

Analysis of the 19 MeV data is in progress.

The space-star anomaly is a discrepancy be-tween theoretical predictions and experimen-tal measurements for the nd-breakup differen-tial cross sections. The data are systematicallyhigher than theory at all energies where mea-surements have been taken. This anomaly hasbeen established by eight previous measurementstaken at neutron beam energies of 10.3, 13.0,16.0, and 25.0 MeV. Three of these experimentswere performed in Germany at Bochum [Ste89]and Erlangen [Str89, Geb93], one at the ChinaInstitute of Atomic Energy [Zho01], and four atTUNL [Set96, Cro01, Mac04]. Figure 4.3 showsall the SST cross-section measurements, includ-ing the current experiment, compared with theo-retical predictions based on the Bonn-B potential[Glo96]. All previous measurements were takenwith essentially the same experimental setup,the common features being: (1) the scattererwas a deuterated scintillator, (2) two neutronswere detected in coincidence, (3) the target-beamintegrated luminosity was determined throughnd elastic scattering by detection of the scat-tered neutron. To determine if there could be acommon experimental error in previous measure-ments, our experiment utilizes a technique simi-lar to the one developed by Huhn et al. [Huh00]to measure the nn scattering length. The pri-mary distinctions between our technique and pre-vious SST measurements are: (1) the deuteratedtarget was a thin foil, (2) a neutron was detected

in coincidence with the scattered proton, and (3)the integrated target-beam luminosity was deter-mined by nd elastic scattering via detection of thescattered deuteron.

10 15 20 25Neutron Beam Energy (MeV)

0.5

1

1.5

2

Cro

ss S

ecti

on (

mb/

sr2 M

eV) Bonn-B Calculation

Bochum [Ste89]Erlangen [Str89, Geb 93]TUNL [Set96, Cro01, Mac04]CIAE [Zhou01]TUNL (Present Work)

Figure 4.3: (Color online) World data for the ndbreakup cross section in the SST con-figuration. The curve is a three-nucleon calculation based on theBonn-B potential.

The experimental setup was surveyed and dis-assembled last year. Updates of various param-eters given in last year’s progress report [Cou09]are given below. It has been determined that theneutron beam energy spread was ±280 keV forthe 19 MeV beam and ±340 keV for the 16 MeVbeam rather than the ±300 MeV spread reportedfor both last year. The space- and coplanar-star neutron detectors have flight paths from the


CD2 target foil of 70 cm and 100 cm, respec-tively as opposed to 50 cm and 70 cm reportedpreviously. Finally, the CH2 target used in therecoil-proton telescope was not properly alignedwith the CD2 target used in the main experiment.Thus we were not able to use np scattering to de-termine the integrated beam flux as we originallyintended. Instead, we used nd scattering off ofthe CD2 target foil.

Determining the integrated beam-target lu-minosity from nd scattering presented some ma-jor difficulties. The deuterons that scatteredfrom our CD2 foil lost significant energy as theyleft the target foil and then had to traversethe ΔE detector before being detected in thethick plastic scintillator in the charged-particlearm. More than 30% of the scattered deuteronswere stopped either in the target or in the ΔEdetector. Also, even though there was morethan 5 MeV separating the “prompt” neutronsin the beam, produced via 2H(d,n)3He, from the“source breakup” neutrons, produced mainly by2H(d,n)dp, there was significant contaminationin our elastic deuteron spectrum from this pro-cess. We dealt with these difficulties by creatingan nd elastic Monte-Carlo simulation that mod-eled scattered deuteron losses as well as the in-fluence of source-breakup neutrons on the elasticspectrum. We then took the high-energy regionof the scattered deuteron spectrum, where thesource breakup influence was minimal, and usedit in conjunction with the Monte-Carlo simula-tion to determine the total number of deuteronselastically scattered by the prompt portion of ourneutron beam. Using this method, the integratedbeam-target luminosity was determined to ±4%for the 16 MeV data.

The results of the 16 MeV data analysis areshown in Fig. 4.4. The data points represent thenumber of counts measured in 1 MeV bins alongthe neutron energy axis after applying an acci-dental background subtraction and adjusting fornd breakup events arising from incident source-breakup neutrons. The error bars represent onlythe statistical uncertainty, which was ±5% at theSST point of 4.5 MeV. The theoretical curve wasproduced by averaging calculated point-geometrycross sections over the finite geometry of our ex-periment. These smeared cross sections wereused in conjunction with neutron-detector effi-ciencies, solid angles, and the nd elastic normal-ization to predict the number of counts expectedas a function of detected-neutron energy. Thetheoretical band shows the ±5.5% systematic un-certainty of our experiment. Our measurementsconfirm the space-star anomaly at 16 MeV, withabout 35% more counts than predicted by theory.Preliminary analysis of the 19 MeV data suggests

the anomaly is present at that energy as well.

0 2 4 6 8Detected Neutron Energy (MeV)

0

50

100

150

200

250

300

Cou

nts

0 5 10 15Detected Neutron Energy (MeV)

0

50

100

150

200

250

300

Cou

nts

Figure 4.4: (Color Online) Results of the 16 MeVmeasurements. The top(bottom)graph is the space(coplanar)-star.The points show the data, the curvesgive the predicted results, and thevertical dotted lines indicate the en-ergy of the SCRE condition.

[Cou09] A. Couture et al., TUNL Progress Re-port, XLVIII, 56 (2009).

[Cro01] A. Crowell, Ph.D Dissertation, (2001).

[Geb93] K. Gebhardt et al., Nuc. Phys., A561,232 (1993).

[Glo96] W. Glockle et al., Phys. Rep., 274, 107(1996).

[Huh00] V. Huhn et al., Phys. Rev. C, 63,014003 (2000).

[Mac04] R. Macri, Ph.D Dissertation, (2004).

[Set96] H. Setze et al., Phys. Lett., B388, 229(1996).

[Ste89] M. Stephan et al., Phys. Rev. C, 39,2133 (1989).

[Str89] J. Strate et al., Nucl. Phys., A501, 51(1989).

[Zho01] Z. Zhou et al., Nucl. Phys., A684, 545c(2001).


4.3.2 Monte-Carlo Simulation of the Neutron-Deuteron Breakup Cross-Section Measurements at 16.0 and 19.0 MeV

S. Tajima, TUNL and North Carolina Central University (NCCU), Durham, NC ; D.M. Markoff,NCCU

We have upgraded the Monte-Carlo simulation code for the neutron-deuteron (nd) breakup

experiment at 16.0 and 19.0 MeV in order to determine the 2H(n,np)n cross sections from

the data. We describe below the major improvements in the simulation code over last year’s

version. The code has been integrated into the data analysis and interpretation for the 16

MeV data.

Neutron-deuteron breakup (NDBU) cross-section data at incident neutron energies of 16.0and 19.0 MeV were measured at TUNL [Cou09].The data were obtained using the coplanar-starand space-star configurations to study the space-star anomaly in nd breakup. In this experiment,a neutron beam is incident on a deuterium target(CD2 foil). We detect one of the neutrons fromthe breakup event in a neutron detector in coin-cidence with the scattered proton detected in thecharged-particle arm using ΔE and E detectors.The ΔE detector consists of a thin plastic scin-tillator with thin aluminum covers. The chargedparticles from the target lose energy in the tar-get foil and in the cover foils of the ΔE detector.Energies of the detected particles are calculatedfrom TOF measurements. A more detailed de-scription of this experiment and data analysis canbe found in Sect. 4.3.1.

A Monte-Carlo (MC) simulation programserves as a tool to obtain predicted quantitieswhich can be compared with data. We reportbelow an update from last year’s MC simulationcode for this experiment [Taj09].

To compare our measured cross section withtheory, a point-geometry calculation [Glo96] isaveraged over the energy spread of the beam andover the finite geometry of the detectors and tar-get. The theoretical nd cross-section libraries at16 and 19 MeV for this experiment have beenrecreated using the CD-Bonn potential and up-dated kinematics parameters. The S value is oneof the important quantities in our analysis andrepresents the arc-length of a locus for a givenset of beam energy and scattering angles, whereS = 0 corresponds to the point on the neutronenergy axis. In the previous MC code, the inter-polation process to calculate S and cross-section

values from the database was not complete be-cause the final interpolation along the S curvewas not done. We modified the algorithm toinclude the interpolation along the S curve us-ing a projection method, and we verified thatthe current interpolation process correctly repro-duces the S and cross section plots. In addition,we employed a better algorithm to calculate theestimated energy loss of charged particles in thefoils by taking into account the effective foil thick-ness, which depends on the scattering angle of acharged particle. The energy of a proton detectedin the E detector differs by several percent in theworst case compared to that from the calculationassuming a single foil thickness.

Our neutron beam basically consists of themono-energetic prompt neutrons (produced inthe gas cell via the 2H(d,n)3He reaction) andsource breakup neutrons (produced mainly via2H(d,n)dp). The latter has much lower beam en-ergy than the former but can still cause NDBU tohappen. To study any contamination from thisprocess, we extended the NDBU cross-section li-brary to include energies ranging from 5.0 to19.0 MeV (8.0 to 19.0 MeV) for the coplanar-star (space-star) configuration, and we includedthe low-energy neutrons in the simulation. In ad-dition, incident neutron energies were sampled inthe code according to the probability distributionof the measured beam energy. The simulationfor the 16 MeV case showed that the majority ofthe scattered protons generated from the incidentsource breakup neutrons are not detected in theE detector, because they do not have sufficientenergy to penetrate the foils. However, for the19 MeV case, we found that the contaminationfrom this process is more significant. Figure 4.5shows the distribution of simulated NDBU events


Figure 4.5: Distribution of simulated NDBU events. Energy of the detected neutron (proton) is on thex (y) axis. The top (bottom) plot is made with the coplanar-star (space-star) configuration.See text for details.

for the 19 MeV case. Events from the sourcebreakup neutrons are clustered in the low-energyregion (near the origin) as shown in the plots,while the other cluster, which appears as a biglocus band, represents events from the 19 MeVprompt neutrons with a beam energy spread of±0.28 MeV. To select the neutrons of interest, a2-dimensional gate is set around the locus bandregion, and we applied the same gate to the datato extract yields. We also set the gate aroundthe source breakup cluster in order to estimatethe background yields.

In addition to the MC simulation program forgenerating NDBU events, we made an nd elas-tic version of the MC program in which only thedeuterons are detected. We found that the largecontamination from the source breakup neutronsis also seen in the deuteron TOF spectrum. How-ever, in order to minimize such a contamination,

the TOF spectrum of high-energy deuterons wascompared with data and the normalization factorfor the simulation yields were obtained.

The simulation code has been applied to com-plete the 16 MeV data analysis as reported inSect. 4.3.1. The predicted yields were calculatedusing this MC simulation based on theoreticalmodels of nucleon-nucleon interactions and theexperimental configuration. The predicted yieldswere then directly compared to the data in orderto study the space-star anomaly.

[Cou09] A. Couture et al., TUNL Progress Re-port, XLVIII, 56 (2009).

[Glo96] W. Glockle et al., Phys. Rep., 274, 107(1996).

[Taj09] S. Tajima and D. M. Markoff, TUNLProgress Report, XLVIII, 58 (2009).


4.4 Reaction Dynamics of Light Nuclei

4.4.1 Study of the 11B(p,α) Reaction Below Ep = 3.8 MeV

P.-N. Seo, M.W. Ahmed, N. Brown, S.S. Henshaw, B.A. Perdue, S. Stave, H.R. Weller,TUNL; R.M. Prior, M.C. Spraker, North Georgia College and State University, Dahlonega, GA;P.P. Martel, A. Teymurazyan, University of Massachusetts, Amherst, MA

We have measured the angular distributions and total cross section for the 11B(p,α) reaction

over an incident energy range of 1.4 MeV to 3.8 MeV in 100-keV steps. Eight silicon surface-

barrier detectors were used at angles from θlab = 30◦ to 160◦. Sample results are reported.

Detailed knowledge of the angular and energydistribution of the outgoing α particles is impor-tant to model an aneutronic fusion reactor usingthe 11B(p,α) reaction. The state-of-the-art modelof the reaction assumes a sequential process, theso-called “two-step model,” leading to one high-energy α particle and a nearly flat continuum ofenergies for the two secondary α particles result-ing from the breakup of the remaining 8Be* nu-cleus [Bec87]. We have extensively studied the11B(p,α) reaction in several previous experiments[Fra07, Fra08b, Fra08a, Sta09, Seo09, Fra09] atenergies between 0.15 and 2.7 MeV. Our resultsindicated that the two-step model is not appro-priate at the very important 0.675 MeV reso-nance [Sta09].

New measurements of the 11B(p,α) reactionin the proton energy range from 1.4 MeV to 3.8MeV in 100-keV steps were made to expand thedata set through the known resonances at 2.64and 3.5 MeV. Our previous studies showed thatthe two-step model works well at the 2.64 MeVresonance.

Protons were incident on a thin 11B targetand the outgoing α particles were detected by sil-icon surface-barrier detectors positioned at eightangles: 30◦, 45◦, 60◦, and 75◦ of beam left and90◦, 115◦, 135◦, and 160◦ of beam right. Theoutgoing particles from the reaction were colli-mated by 1.5mm × 9.2 mm collimators in frontof each detector.

The α-particle counts were normalized by theintegrated beam luminosity, target thickness, andsolid angle ΔΩ. The measured distributions werethen boosted from the laboratory frame to thecenter-of-mass frame. Finally the data in theoverlapping energy region (1.4 MeV to 2.7 MeV)between the current measurement and the previ-

ous measurement were used to scale the currentdistributions to those measured in 2007.

Figure 4.6 shows the distribution ofCounts/(L ΔΩ) vs. ECM

α and Ep. Here L is theintegrated luminosity, or the number of target nu-clei multiplied by the number of incident protons.The quantity Counts/(L ΔΩ) is analogous to thecross section but makes no assumption about theexpected number of α particles produced in agiven reaction. The figure includes the new datafrom 1.4 MeV to 3.8 MeV as well as the previousdata from 0.5 to 2.7 MeV [Fra07, Fra08b] andfrom 0.15 to 0.4 MeV [Seo09]. Figure 4.7 is theplot of the same distribution of Counts/(L ΔΩ)vs. ECM

α and Ep as Fig. 4.6 but with a rescaledz-axis to emphasize the structures seen at higherproton energies, to show the cross section varia-tion more clearly.

[Bec87] H. Becker, C. Rolfs, and H. Trautvetter,Z. Phys., A327, 341 (1987).

[Fra07] R. France et al., TUNL Progress Re-port, XLVI, 72 (2007).

[Fra08a] R. France et al., TUNL Progress Re-port, XLVII, 68 (2008).

[Fra08b] R. France et al., TUNL Progress Re-port, XLVII, 66 (2008).

[Fra09] R. France et al., TUNL Progress Re-port, XLVIII, 64 (2009).

[Seo09] P.-N. Seo et al., TUNL Progress Re-port, XLVIII, 68 (2009).

[Sta09] S. Stave et al., TUNL Progress Report,XLVIII, 66 (2009).


[MeV]pE

0.5 1 1.5 2 2.5 3 3.5 4 [MeV]

CMαE

01

23

45

67

89

[m

b/s

r]ΩΔ

Co

un

ts/L

02468

1012141618

Figure 4.6: (Color online) Distribution of Counts/(LΔΩ) vs. ECMα and Ep in the proton energy range

from 0.150 MeV to 3.8 MeV for the beam right 90◦ detector.

[MeV]p

E

0.51

1.52 2.5

33.5 4

[MeV]CMαE

0 1 2 3 4 5 6 7 8 9

[m

b/s

r]ΩΔ

Co

un

ts/L

0

1

2

3

4

5

6

Figure 4.7: (Color online) Plot of the same distribution as in Fig. 4.6 but with a rescaled z-axis toemphasize the structures seen at higher proton energies


4.4.2 The Study of the 11B(α,p) Reaction with an α-Particle Energy from4.0 MeV and 7.0 MeV

P.-N. Seo, M.W. Ahmed, N. Brown, S.S. Henshaw, B.A. Perdue, S. Stave, H.R. Weller,TUNL; R.M. Prior, M.C. Spraker, North Georgia College and State University, Dahlonega, GA;P.P. Martel, A. Teymurazyan, University of Massachusetts, Amherst, MA R.H. France III,Georgia College and State University, Milledgeville, GA

We measured angular distributions for the emission of protons from the 11B(α,p)14C reaction

at α particle energies from 4.0 to 7.0 MeV. Eight silicon detectors were used at eight angles

from θlab = 45◦ to 160◦. Measured angular distributions were fitted with Legendre polynomials

to fourth order, and total cross sections were obtained. The uncertainty of the measured total

cross sections is less than 5%. The present result is compared with previous data.

As a part of our ongoing study of the 11B(p,α)reaction for aneutronic reactor design, we mea-sured cross sections and angular distributions forthe 11B(α,p)14C reaction at α-particle energiesfrom 4.0 to 7.0 MeV in 100-keV steps, exceptfor 4.1 and 4.2 MeV. In a 11B-fueled fusion re-actor, α particles are produced mostly by the11B(p,α) reaction, and the possible productionof large quantities of the long-lived isotope 14Cby the secondary reaction 11B(α,p)14C can be-come an important problem [Day76]. This reac-tion may occur when the α particles produced bythe 11B(p,α) reaction slow down within the hot,dense plasma. To estimate the 14C productionrate in a (11B+p)-fueled fusion reactor, cross sec-tions for the 11B(α,p)14C reaction are required.Because the Q-value of this reaction is 0.784 MeVand the 14C first excited state lies at 6.093 MeV,only the (α,p0) reaction is energetically open forthe current energy range. Only a few groups[Man63, Bon97, Gio96, Hou78] have previouslymeasured angular distributions and total crosssections for the outgoing protons; however, theirdata show significant discrepancies. We compareour results with the previous measurements.

This measurement was performed using thesame experimental setup as the elastic scatter-ing experiment [Fra09] 11B(α,α), except that thinaluminum foils were placed in front of each silicondetector to stop elastically scattering α particleswhile allowing the passage of outgoing protons.Details of the experiment are described, anda typical proton spectrum is shown in [Fra09].Counts in the proton peaks were normalized us-ing the total charge on the target, target thick-ness, and the solid angle. Measured angular dis-

tributions in the center-of-mass system were plot-ted, and a three dimensional view of the angu-lar distributions for the full range of incident α-particle energies is shown in Fig. 4.8. The angulardistributions show a strong dependence on Eα.

[MeV]αE4 4.5 5 5.5 6 6.5 7 [degree]

CMθ

6080

100120

140160

Dif

f. X

S [

mb

/sr]

0

2

4

6

8

10

Figure 4.8: (Color online) Differential cross sec-tion as a function of angle and incidentα-particle energy

The resulting angular distributions were fit-ted with Legendre polynomials through the forthorder. Examples of this fit are shown in Fig. 4.9.The angular distribution at the 4.3-MeV reso-nance shows a large fore-aft asymmetry. Thevariation of the reduced Legendre polynomial co-efficients (i.e., divided by the zeroth-order co-efficient) with α-particle energy is shown inFig. 4.10. The error bars in the figure are smallerthan the data points and include a statistical er-ror of less than 3.2%.


Figure 4.9: (Color online) Angular distributions in the CM system from the 11B(α,p)14C reaction at4.0 and 4.3 MeV. The curves show a fourth order Legendre polynomial fit to the data.

Figure 4.10: (Color online) Reduced coefficientsfrom the Legendre polynomial fitsto the proton angular distributions.The lines through the data are onlyto guide the eye.

The total cross section was extracted fromthe zeroth-order coefficient of the fit. The re-sulting excitation function and its uncertaintyare shown in Fig. 4.11 along with other availabledata. The total cross section changes stronglywith Eα, indicating that the reaction is pro-ceeding through the compound nucleus 15N. Thenon-vanishing odd-order coefficients indicate thepresence of interference between levels of oppo-site parity in the compound nucleus. Results of[Bon97] and [Hou78] differ by about a factor offive at 4.3 MeV. Our data are closer to Bonetti’scross sections [Bon97], and the shape of the over-

all cross section is in excellent agreement withtheirs. After normalization, the excitation func-tion of [Man63] in the 4.3-MeV resonance regionis also plotted and is in good agreement with thepresent data.

Figure 4.11: (Color online) Measured total crosssections as a function of incident α-particle energy are compared withprevious data

[Bon97] R. Bonetti, A. Guglielmetti, and G.Poli, Appl. Radiat. Isot., 48, 873(1997).

[Day76] R. Dayras, Z. Switkowski, andT.A.Tombrello, Nucl. Phys., 261,365 (1976).

[Fra09] R. France et al., TUNL Progress Re-port, XLVIII, 64 (2009).

[Gio96] G. Giorginis et al., Nucl. Instrum.Methods, B118, 224 (1996).

[Hou78] W.-S. Hou et al., Nucl. Sci. Eng., 66,188 (1978).

[Man63] G. Mani, P.D.Forsyth, and R. Perry,Nucl. Phys., 44, 625 (1963).


4.4.3 11B(α,α) Elastic Cross-Section Measurements

N. Brown, M.W. Ahmed, S. Stave, S.S. Henshaw, B.A. Perdue, P.-N. Seo, H.R. Weller,TUNL; M. Spraker, R. Prior, North Georgia College and State University, Dahlonega GA; P.P.

Martel, A. Teymurazyan, University of Massachusetts, Amherst, MA;

Measurements of the 11B(α,α) elastic scattering reaction were performed at energies fromEα = 2.0 to 8.0 MeV. The experimental method is described, and results are shown.

11B(α,α) scattering data were collected atTUNL using an incident α-particle beam with en-ergies ranging from 2.0 to 8.0 MeV. With an aver-age of approximately 10 nA of beam on target, weobserved scattered α-particles using eight siliconsurface-barrier detectors. The detector thicknesswas sufficient to stop α-particles at all relevantenergies. Detectors set at laboratory angles of45, 60, 75 and 90 degrees were situated on thebottom plate of the vacuum chamber, and detec-tors at angles of 90, 115, 135 and 160 degreeswere attached to the top plate.

Three targets were employed for this exper-iment: gold, carbon and boron. The gold foiltarget is 105 μg/cm2, and data collected wereused for calibration in comparison to calculatedRutherford scattering data. The carbon targetwas 20 μg/cm2, and data collected were usedto measure the 12C cross section at each angleand energy. The boron target was 76 μg/cm2 ofboron sandwiched between 20 and 46 μg

cm2 of ti-tanium. For each energy step throughout the 1.5to 8.0 MeV region, data for each target was col-lected to ensure a calibration value for each datapoint. Energy calibration of the detector spec-tra was established using the Rutherford datafrom the Au target. A two-point fit between alow energy and a high energy run was sufficientto establish a linear relationship. This linear fitwas used for all data collected using a specificspectroscopy-amplifier setting or gain setting.

Three different gain settings were usedthroughout the energy range to ensure that theexpected α-scattering peak location was visiblein each detector. A linear fit was created foreach of the eight detectors at each of the threegain settings and was used to calibrate each de-tector spectrum. The corrected detector spec-tra were then managed as a collection of his-tograms for the remainder of the analysis pro-cess. The energy-calibrated boron-target spec-tra contain cross section contributions from sev-

eral different nuclei. Easily identifiable are peaksfrom boron, carbon, oxygen and titanium. Whilethe oxygen and titanium peaks are well separatedfrom the boron peak, the carbon peak begins tomerge with the boron peak at low energies andforward angles.

At high energies (7-8 MeV) these peaks areseparated enough in energy to estimate the thick-ness of the carbon layer in the boron target bymeasuring the carbon peak and comparing thatto the carbon yield from the known 20 μg/cm2

carbon target. This provides an estimate of 5μg/cm2 for the carbon layer in the boron tar-get. This allows a bin-by-bin subtraction of thehistograms, ensuring no over subtraction whenremoving the carbon contribution. The remain-ing peak is summed to provide the boron crosssections at each angle and energy.

A transfer from the laboratory referenceframe to the center-of-mass reference frame waspreformed using the computer code rkin. The11B center-of-mass α, α scattering cross sectionmeasurements are summarized at seven differentenergies between 2.0 and 8.0 MeV. Above 5 MeVthe boron cross section is only near Rutherfordat the forward angles, while the back angles re-main well above the Rutherford value. As thealpha energy decreases, the cross section at allangles approaches the Rutherford values. Whenthe alpha energy is decreased to 3.0 MeV and be-low, virtually all angles are within 10-20% of theRutherford value.

Representative results are shown inFigs. 4.12-4.18 at Eα=2.0, 3.0, 4.0, 5.0, 6.0, 7.0,and 8.0 MeV. An investigation into our estimatesof the systematic errors pertinent to the absolutecross-section scale allows reporting to within an≈ 5% error. Additionally, a phase-shift analy-sis is being developed for the extension of thisdata set to intermediate unmeasured angles andenergies. It is described in Sect. 4.4.4.


[deg]cmθ20 40 60 80 100 120 140 160 180

[m

b/s

r]cm

Cro

ss s

ecti

on

210

310

410TUNL July 09

Rutherford

Figure 4.12: (Color online) 11B(α,α) center-of-mass scattering cross section forEα = 2.0 MeV.

[deg]cmθ20 40 60 80 100 120 140 160 180

[m

b/s

r]cm

Cro

ss s

ecti

on

210

310

410TUNL July 09

Rutherford

Figure 4.13: (Color online) Same as Fig. 4.12 forEα = 3.0 MeV.

[deg]cmθ20 40 60 80 100 120 140 160 180

[m

b/s

r]cm

Cro

ss s

ecti

on

210

310

410TUNL July 09

Rutherford


[deg]cmθ20 40 60 80 100 120 140 160 180

[m

b/s

r]cm

Cro

ss s

ecti

on

210

310

410TUNL July 09

Rutherford


[deg]cmθ20 40 60 80 100 120 140 160 180

[m

b/s

r]cm

Cro

ss s

ecti

on

210

310

410TUNL July 09

Rutherford


[deg]cmθ20 40 60 80 100 120 140 160 180

[m

b/s

r]cm

Cro

ss s

ecti

on

10

210

310

410TUNL July 09

Rutherford


[deg]cmθ20 40 60 80 100 120 140 160 180

[m

b/s

r]cm

Cro

ss s

ecti

on

10

210

310

410TUNL July 09

Rutherford



4.4.4 Phase-Shift Analysis of 11B(α,α)11B from 2.0 to 5.4 MeV

R.M. Prior, North Georgia College and State University, Dahlonega, GA; S. Stave, N. Brown,H.R. Weller, TUNL

The 11B(α,α)11B elastic-scattering cross section has been measured as a part of a study ofthe 11B(p,α)αα reaction and its use as a power source in an aneutronic fusion reactor. Aphase-shift analysis has been performed to allow interpolation of the scattering data.

As a part of the consideration of using the11B(p,α)αα reaction in an aneutronic fusion re-actor, it is necessary to know the properties ofother reactions that may occur within the reac-tor. The elastic scattering of alpha particles from11B has been measured at TUNL at laboratoryscattering angles of 45◦, 60◦, 75◦, 90◦, 115◦, 135◦,and 160◦. The experimental details and resultshave been reported in Sect. 4.4.3.

An exact phase-shift analysis of the elasticscattering data would be complicated by the 3/2

spin of 11B, and would require data at a muchlarger number of scattering angles at each en-ergy because of the number of partial waves thatwould have to be included. Physically, the spinof the nucleus is not likely to have much effect onthe scattering, so a simpler formalism was usedto fit the data set. The phase-shift formalism forspin-0 particles scattering from a spin-0 targetwas used to fit the data. The formalism is givenbelow.

[deg]CM

θ

6080

100120

140160

180

[MeV]

αE

22.5

33.5

44.5

5

[m

b/s

r]Ω

/dσd

10

210

310

Figure 4.19: (Color online) The differential cross section evaluated using the fitted phase shifts for

the 11B(α, α) reaction for incident α energies between 2 and 5.4 MeV as a function of theoutgoing α-particle polar angle in the center-of-mass frame.


The differential cross section can be writtenin terms of two complex scattering amplitudes,FC and FN :

dσ

dΩ(θ) =

1k2

|FC + FN |2 (4.1)

where the wave number k = 0.2187 ×1013μ

√E

m1cm for E in MeV and m1 in amu.

The quantity μ is the reduced mass μ = m1m2m1+m2

.The terms m1 and m2 are the incident and targetparticle masses respectively. The Coulomb scat-tering amplitude FC is written in terms of thecenter-of-mass scattering angle θ and the quan-tity η as:

FC = −12η

(1

sin2 θ2

)e

iηln

„1

sin2 θ2

«, (4.2)

where η = Z1Z2hν = 0.1575Z1Z2

√m1E and E is

in MeV. The terms Z1 and Z2 are the incidentand target particle atomic numbers, respectively.The nuclear scattering amplitude FN is:

FN = − i

2

lmax∑l=0

eiαl(2l + 1)(Sl − 1)Pl(cos θ) (4.3)

where the Coulomb phase shift is given by α0 =0, αl = αl−1 + 2 tan−1

(ηl

), and Sl = γle

2iδl . Theparameters that have to be determined to repro-duce the experimental cross-section data are δl,the real part of the phase shift, and γl, the damp-ing parameter that represents the imaginary partof the phase shift for each l-value included in thefit. The real part of the phase shift is usually ex-pressed in degrees or radians. The damping pa-

rameter represents absorption of the correspond-ing partial wave; it is dimensionless and varies be-tween 1.0 (meaning no absorption) and 0 (mean-ing total absorption).

No attempt has been made to interpret theanalysis in terms of the reaction mechanism. Theresults provide a way to parameterize the en-tire data set and allow interpolation over energieswithin the range of the data and to interpolateand extrapolate to any scattering angle.

In this analysis, partial waves for angular-momentum quantum numbers 0 through 3 wereincluded, meaning four complex phase shifts wereadjusted. Since each phase shift is a complexnumber, two parameters were adjusted for eachphase shift giving a maximum of 8 parameters.At the lower energies the higher order phase shiftswere not needed so that fewer parameters hadto be determined. The fitting was done with aFortran code that used the Marquart method forparameter determination.

Angular distribution data at energies from 2.0MeV to 5.4 MeV were used. The angular distri-bution at each energy was fitted. Tables of phaseshifts and damping parameters can then be usedto calculate cross sections at any energy between2.0 and 5.4 MeV by interpolating the table of fit-ted parameters and using the parameters to cal-culate the corresponding cross section. The crosssections can also be calculated at any angle. Fig-ure 4.19 shows the calculated cross section as afunction of angle and energy over the range from2.0 to 5.4 MeV. Over 80% of the fitted differ-ential cross-section values are within 20% of theexperimental input data.

The Many-Nucleon Problem

Chapter 5

• Preequilibrium Nuclear Reactions and Random MatrixTheory

• Neutron-Induced Reactions

94 The Many-Nucleon Problem TUNL XLIX 2009–10

5.1 Preequilibrium Nuclear Reactions and Random MatrixTheory

5.1.1 Random Matrices and Chaos in Nuclear Physics

G.E. Mitchell, TUNL; J.F. Shriner, Jr., Tennessee Technological University, Cookeville, TN ;A. Richter, Technische Universitat Darmstadt, Darmstadt, Germany; H.A. Weidenmuller,Max-Planck-Institut fur Kernphysik, Heidelberg, Germany.

The last comprehensive review of random-matrix theory (RMT)in nuclear physics was givenby Brody et al. over 25 years ago. A two-part review on this topic has been prepared forthe Reviews of Modern Physics. The first part (on spectra) has recently been published; thesecond part (on reactions) has been accepted for publication. One important application ofRMT is the evaluation of the completeness of neutron resonance data. We have prepareda report for the International Atomic Energy Agency that describes a missing-level analysisusing measures that characterize the spacings rather than the more conventional widths.

The last comprehensive review concerning theapplication of random-matrix theory (RMT) tonuclear physics was by Brody et al. in 1981[Bro81]. In response to a request from Reviewsof Modern Physics, we have prepared a two-partseries of reviews. The first, entitled Random Ma-trices and Chaos in Nuclear Physics: NuclearStructure [Wei09], was published in 2009 and thesecond part, Random Matrices and Chaos in Nu-clear Physics: Nuclear Reactions [Mit], has beenaccepted for publication.

Random matrices were introduced into nu-clear physics by Wigner in the late 1950s. Thisstep was preceded and probably motivated byBohr’s insight that nuclei are systems of greatcomplexity. Bohr’s idea that the nucleus isa complex, strongly interacting system is mostclearly demonstrated in the compound nucleus(CN) model. Wigner first applied random-matrixideas to the spacings of neutron resonances. Therandom-matrix approach characterizes spectraby their fluctuation properties. The distributionof spacings of nearest neighbors is the most com-mon measure. To implement this approach oneneeds a statistical theory of spectra, and ran-dom matrices provide this tool. One considersan ensemble of Hamiltonians (in matrix form).The ensemble is defined in terms of the probabil-ity distribution for the matrix elements, thus thename random matrices. The ensemble is chosensuch that the Hamiltonians incorporate genericfeatures. The spectral distribution functions arecalculated as averages over the ensemble and arecompared with the actual fluctuation properties

of nuclear spectra. Since nuclei are invariant un-der time reversal, the nuclear Hamiltonian canbe chosen to be real and symmetric.

In part I of the series, following a brief defini-tion of the Gaussian orthogonal ensemble (GOE)and an interpretation of the results for the ensem-ble (average level density, universality, informa-tion content, etc.), we (HAW and GEM) considerthe standard GOE fluctuation measures (Porter-Thomas distribution, nearest-neighbor-spacingdistribution, Dyson-Mehta statistic, etc.). Wethen provide a comparison of the experimentalresults with the predictions of RMT. We focus onneutron and proton resonances, low-lying levels,high-spin states, low-lying modes of excitation,and eigenvector distributions. We then considerboth the models and the data for isospin symme-try violation and for tests of time-reversal invari-ance. Most of the rest of the review is devoted tothe issue of chaos in nuclear models. In additionto consideration of the shell model and the col-lective model in various forms, we also considerrandom-matrix models that were inspired by nu-clear structure concepts. The evidence stronglysupports the view that chaos is a generic propertyof nuclei.

In part II, we (GEM, AR, and HAW) firstconsider early work in nuclear reactions, such asthe Hauser-Feshbach formula. Later work ex-tended this formula to direct reactions. Anothermajor development was Ericson’s prediction ofstatistical fluctuations in nuclear cross sections.Attempts were made to connect the statisticalmodels of CN scattering with RMT. Following

TUNL XLIX 2009–10 The Many-Nucleon Problem 95

the development of the theory of chaotic mo-tion, CN scattering was recognized as a paradig-matic case of quantum chaotic scattering. Af-ter a summary of some basic facts and conceptsof nuclear-reaction theory, we present the cen-tral ideas and models that were developed beforethe connection to RMT was established (Hauser-Feshbach formula, Weisskopf estimate, Ericsonfluctuations, etc.). In order to avoid repeat-ing the arguments later in a different form, wepresent them in modern terminology. Tests ofthe theory are reviewed, as are applications innuclear physics, with special attention given toviolation of symmetries (isospin, parity) and totime-reversal invariance.

A classic problem in applied nuclear physicsis the evaluation of data quality. A key exampleinvolves the determination of nuclear level densi-ties from neutron resonances. This informationis crucial for such areas as shielding calculationsand advanced-fuel-cycle issues. The key ques-tion is whether some levels are missing, and if so,how many. The standard approach is to assumethat RMT applies to the level statistics of neu-tron resonances and to use the Porter-Thomaswidth distribution in order to estimate the num-ber of missing levels. Unfortunately, the widthdistribution is strongly affected by the presenceof non-statistical effects, which can severely dis-tort the distributions and lead to incorrect valuesfor the missing level correction and thus for thelevel density. The spacing distributions are notaffected by non-statistical effects. Since widthsand spacings are independent in RMT (the eigen-values and eigenfunctions factor in the probabil-

ity distribution), missing-level analysis using thespacings provides an independent check of thesame data set. In principle any of the measuresthat are used to evaluate spacing data can be uti-lized. We (GEM and JFS) have used the nearest-neighbor-spacing, the Δ3 statistic, the thermody-namic energy, and the linear correlation of adja-cent spacings in order to evaluate neutron reso-nance data.

Our initial efforts have been for targets withspin zero. This work is summarized in a reportfor the Nuclear Data section of the InternationalAtomic Energy Agency [Mit09]. The specific fo-cus is on the results for key actinides. The gen-eral effort is to provide best practice procedures.Analysis codes for the five measures that we useare available under the RIPL (Reference InputParameters Libraries) framework. Our presentefforts are to extend these analysis methods totargets with non-zero spin.

[Bro81] T. A. Brody et al., Rev. Mod. Phys., 53,385 (1981).

[Mit] G. E. Mitchell, A. Richter, and H. A.Weidenmuller, to be published in Rev.Mod. Phys.

[Mit09] G. E. Mitchell and J. J. F. Shriner,Missing Level Corrections Using Neu-tron Spacings, Technical ReportINDC(NDS)-0561, International AtomicEnergy Agency, Vienna, 2009.

[Wei09] H. A. Weidenmuller and G. E. Mitchell,Rev. Mod. Phys., 81, 539 (2009).


5.1.2 Preequilbrium Reaction Phenomenology

C. Kalbach Walker, TUNL

Work has continued on a phenomenological model for light-projectile breakup reactions. Pre-viously reported systematics for the angular distributions and the absolute breakup crosssections have been revised, and the revised model is yielding good agreement with experi-ment for d and 3He breakup. Comparisons for α-particle breakup are continuing.

For light-particle (A ≤ 4) induced reactionsat incident energies of 14 to 200 MeV, the semi-classical exciton model of preequilibrium nuclearreactions provides a simple way to describe thecontinuum energy and angular distributions ofthe light particles emitted during energy equili-bration. The TUNL code system preco is basedon the exciton model and has been used (eitheralone or in Hauser-Feshbach codes) in appliedprojects and in basic research. Model develop-ment uses simple physical concepts and relies onavailable data to direct choices between alterna-tive model formulations and to provide values forkey model parameters.

An important strength of preco is its abil-ity to treat reactions with complex particles inthe entrance channel and/or exit channel, but forloosely bound projectiles—d, t and 3He—and forα particles at incident energies above about 50MeV, the lack of a model for projectile breakupmakes the description incomplete. Yet deuteron-induced reactions are of increased importance.The International Atomic Energy Agency has ini-tiated a Coordinated Research Project (CRP)to upgrade the Fusion Energy Nuclear Data Li-brary to version FENDL-3. An important ad-dition will be “data” (mostly model-generated)for deuteron-induced reactions. Therefore a phe-nomenological projectile-breakup model has beendeveloped as part of the FENDL-3 CRP. Oncevalidated, the model will be included in the nextrelease of preco.

Projectile breakup is here defined as the emis-sion of a projectile fragment with a fairly narrowenergy distribution peaked at an emission energycorresponding to the projectile velocity. The in-clusion of data for d, 3He and α-particle projec-tiles in the database for this work leads to a morerobust and global model than one developed fora single projectile type.

Early work [Kal08] concentrated on the cen-troid energies and shapes of the breakup peaks.

Then the angular distribution systematics andthe variation of the breakup cross section withtarget, incident energy, and breakup channelwere described [Kal09]. This year, early compar-isons between model calculations and experimen-tal data indicated that the angular distributionsystematics needed revision.

The angular distributions for projectilebreakup typically follow an exponential falloffwith abuθlab, where θlab is the emission an-gle in the laboratory system and abu is theangular-distribution slope parameter for projec-tile breakup. When the detected fragment is onemass unit lighter than the projectile, the slopeparameter was previously described as increasinglinearly with the energy of the breakup peak. Forall lighter fragments, it was given the value 5.87rad−1. The new description shown in Fig. 5.1 ismore complicated but also more accurate.

Figure 5.1: Empirical values for the breakup-angular-distribution slope parameterabu as a function of the average peakenergy. The lines show the systemat-ics.

For relatively large values of Ca/Einc (the ra-tio of the entrance-channel Coulomb barrier tothe incident energy), this behavior is modified bya forward-angle dip in the experimental breakup


cross section. The dip is described as due to pen-etration through an angular barrier. With thechange in the systematics of abu, the systemat-ics for this barrier also had to be revised. Theforward-angle dip is best characterized for (d,p)breakup at 15 MeV. The experimental angulardistributions for other breakup systems do notinclude small enough angles to show the dip, butthe angular barrier plays a role in determiningthe relative breakup yields for a range of targetswith a fixed breakup channel, incident energy,and measurement angle, thus providing furtherguidance for the systematics.

The changes in the angular distribution sys-tematics resulted in revised estimates of theangle-integrated cross sections, so the normaliza-tion systematics of the overall breakup cross sec-tions as a function of breakup channel and inci-dent energy were revised. The target dependenceof the breakup cross section was unchanged.

The small computer code used to calculatethe breakup cross section using the current modelwas modified to include all of these changes, anda comprehensive comparison of model calcula-tions with measured double-differential cross sec-tions is being carried out. Good agreement isfound for (d,p) breakup at 15 and 56 MeV, asseen in Figs. 5.2 and 5.3. Similar results havebeen obtained for (3He,d) breakup at 70, 90, 110,and 130 MeV (see Fig. 5.4) and for (3He,p) at 70and 90 MeV.

A more complete description of the model andadditional data comparisons are given in a reportto the second Research Coordination Meeting ofthe FENDL-3 project [Kal10].

Figure 5.2: Comparison of experimental and cal-culated breakup peaks for Zr(d,p) at14.8 MeV. The points show the datafrom Ref. [Ham61], the Gaussian-shaped dashed curves show the calcu-lated breakup peaks, the other dashedcurves show a reasonable estimate ofthe underlying continuum cross sec-tion at each angle, and the solid curvesshow the total estimated spectra.

Figure 5.3: Comparison of experimental and cal-

culated breakup peaks for 118Sn(d,p)at 56 MeV. The points and curveshave the same significance as in Fig.5.2, except that the data are from Ref.[Mat80].

Figure 5.4: Comparison of experimental and cal-

culated breakup peaks for 90Zr(3He,d)at 90 MeV. The points and curveshave the same significance as in Fig.5.2, except that the data are from Ref.[Mat78].

[Ham61] E. W. Hamburger, B. L. Cohen, andR. E. Price, Phys. Rev., 121, 1143(1961).

[Kal08] C. Kalbach Walker, TUNL Progress Re-port, XLVII, 78 (2008).

[Kal09] C. Kalbach Walker, TUNL Progress Re-port, XLVIII, 74 (2009).

[Kal10] C. Kalbach Walker, Report to the 2ndmeeting of the FENDL-3 CRP, Vienna,Austria, March 2010, posted at www-nds.iaea.org/fendl3/vardocs.html., 2010.

[Mat78] N. Matsuoka et al., Nucl. Phys., A311,173 (1978).

[Mat80] N. Matsuoka et al., Nucl. Phys., A345,1 (1980).


5.2 Neutron-Induced Reactions

5.2.1 Energy Dependance of the Fission-Product Yields in Fast and Ther-mal Neutron-Induced Reactions

A.P. Tonchev, C.R. Howell, J.H. Kelly, E. Kwan, R. Raut, G. Rusev, W. Tornow,TUNL; J.A. Becker, C.Y. Wu, R. Macri, C. Hagmann, M.A. Stoyer, I.J. Thompson,Lawrence Livermore National Laboratory, Livermore, CA; D.J. Vieira, R.S. Rundberg, J.B.

Wilhelmy, Los Alamos National Laboratory, Los Alamos, NM

The goal of this experimental activity is to gain insight into the possible neutron-energydependence of fission-product yields by measuring their γ-ray yield ratios. The long-standingexperimental problem is whether the yield of specific fission products, such as 140Ba and147Nd, and their ratio varies as a function of incident neutron energy. The existing literatureindicates that the high-yield fission product 140Ba does not have any energy dependence. Weperformed high-precision fission-product-yield measurements on 238U using monoenergetic14.5 MeV neutrons. The results will be compared to theoretical analyses to provide a morefundamental understanding of the fission process.

One of the primary interests of the stockpilestewardship program is to gain knowledge of theenergy dependence of fission product yields incertain mass regions as a function of incident neu-tron energy. The primary interest is 239Pu withthermal neutrons and fast neutrons from En = 1to 16 MeV.

Figure 5.5: (Color online) Partial γ-ray spectra of238U after irradiation with En = 14.5MeV neutrons.

The first proof-of-principle experiment hasbeen performed at TUNL to demonstrate the fea-sibility of the 147Nd/140Ba ratio technique. An1100 mg/cm2 238U target was irradiated with14.5 MeV neutrons produced via the 2H(d,n)3Hereaction. The depleted 238U target was irradi-

ated for 24 h at an average neutron flux of 7x106

n/(s cm2). The uranium target was sandwichedbetween thin gold and aluminum foils, whichserved as flux monitors. Following the irradi-ation, the foils were counted using the GENIEMultiport II high-resolution γ-ray spectrometersystem. No radiochemical separation procedurewas applied to the irradiated target.

Figure 5.6: Cross section data for the 238U(n,2n)reaction in the energy region aroundEn = 14 MeV.

The intensities of the characteristic γ raysfrom the decay of the 12.75 d 140Ba (Eγ = 537keV) and 10.98 d 147Nd (Eγ = 531 keV) isotopeswere measured continuously during the follow-ing three months. A portion of the γ-ray spec-tra recorded in the first week after irradiation isshown in Fig. 5.5. It should be noted that at


short times, the 147Nd peak at Eγ = 531 keV in-terferes with the strong 530 keV transition fromthe decay of the 20.8 h 133I isotope. Due to thelarge difference in the half lives, the 133I decaysaway, leaving a pure 147Nd γ-ray peak after afew days. The ratio of the yield of the 140Ba fis-sion fragment to that of 147Nd as obtained fromthe present measurement is equal to 2.222±0.037.This ratio agrees with the data value for the239Pu(n,f) reaction at En = 14 MeV.

Besides the fission product yield measure-ments, the cross section of the 238U(n,2n)237Ureaction was determined by measuring the emis-sion rate of the 208.0 keV γ-rays from the 237Udecay. The residual 237U nucleus has very a longhalf-life (T1/2 = 6.75 d) and hence the activa-tion method is a very convenient technique to ob-tain certain reaction cross sections [Ton08]. Onedrawback of the activation measurements in thisparticular case is the large attenuation correc-tions which need to be applied for the 208.0 keVγ ray exiting the thick 238U sample.

The result from the present 238U(n,2n) cross-section measurement and the existing literaturedata [Fil99, Gol87, Ves78] around 14 MeV neu-tron energy are shown in Fig. 5.6. As can be seen,the present result and previous measurements are

in good agreement.Our future plans include analyzing the energy

dependent yields of various fission product iso-topes to investigate whether this ratio changeswith increasing neutron energy and whether it issimilar for different fissile actinides.

This work was supported by the National Nu-clear Security Administration under the Stew-ardship Science Academic Alliances Programthrough Department of Energy grant DE-FG52-09NA29465.

[Fil99] A. A. Filatenkov et al., Systematic Mea-surement of Activation Cross Sectionsat Neutron Energies from 13.4 to 14.9MeV, Technical Report RI 252, KhlopinRadium Institute, Leningrad, 1999.

[Gol87] V. Y. Golovnya et al., In First Int. Conf.on Neutron Physics, Kiev, volume 3, p.281, 1987.

[Ton08] A. Tonchev et al., Phys. Rev. C, 77,054610 (2008).

[Ves78] L. R. Veeser et al., In International Con-ference on Neutron Physics and NuclearData, Harwell, p. 1054, 1978.


5.2.2 Test Measurements of the 235U(n, 2nγ)234U Reaction with PPAC-Based Fission Chamber

G. Rusev, E. Kwan, C.R. Howell, R. Raut, W. Tornow, J.H. Kelley, TUNL; J.A. Becker,A. Chyzh, C.Y. Wu, Lawrence Livermore National Laboratory, Livermore, CA; R.O. Nelson,Los Alamos National Laboratory, Los Alamos, NM

Future measurements of neutron-induced reactions on minor actinides at TUNL’s shielded-

neutron-source area require an efficient detector setup in combination with a low background.

First measurements with a PPAC-based fission chamber were carried out to determine the

level of background. Results of these measurements are presented.

Measurements of neutron-induced reactionson 235U and 238U were carried out with mo-noenergetic neutron beams at TUNL with tar-get masses of about 1.5 g [Hut09]. The natu-ral radioactivity of the uranium targets togetherwith the beam-related background make mea-surements of γ-ray transitions from low-lyingstates of the product nuclei very difficult. Anovel approach to overcoming this backgroundproblem was proposed by the experimental nu-clear physics group at LLNL. An experiment witha parallel-plate-avalanche-counter (PPAC)-basedfission chamber, shown in Fig. 5.7, would havethe advantages that (i) a thin target with a massof 20 mg used in the PPAC would provide 75times less natural radioactivity than the 1.5 gtarget used in the past, and (ii) the thinner tar-get will not stop the fission fragments, which willthen give a signal in the PPAC and can be used asa veto for the HPGe detectors to further reducethe background. Note that the induced fission in235U from 10 MeV neutrons has a cross sectionof 1.7 barn. Furthermore, measurement of theγ-rays from the 235U(n, 2nγ) reaction can bene-fit from the reduced background of the 4+ → 2+

transition in 234U at 99.9 keV which collects mostof the γ cascades.

A scan of the neutron beam distribution atTUNL’s shielded-neutron-source area was per-formed prior to the measurement with the PPAC.A liquid scintillator detector with a diameter of5 cm was placed downstream to measure the fluxof the neutron beam. Another small plastic scin-tillator detector with an area of 0.32 × 2.54 cm2

was mounted on a remotely controlled stand po-sitioned at the location where the PPAC chamberwas to be installed. We deduced the diameter ofthe beam to be 4 cm at half intensity. This mea-

surement was used for precision alignment of thePPAC chamber.

Figure 5.7: (Color online) A photograph of thePPAC-based fission chamber, showingthe Al chamber, the Kapton windowsfor the neutron beam and the tubingof the gas-handling system.

We carried out a test measurement with thePPAC chamber containing two 235U foils, eachwith a mass of 10 mg, at En = 10 MeV. Twoaluminized Mylar foils on either sides of the 235Ufoil collected the charge produced by the fissionfragments in isobutane gas. Two thin Kaptonfoils on the entrance and exit windows were usedin order to reduce the scattering of the neutronbeam off the windows. Two 40% HPGe pla-nar detectors, positioned on either side of thePPAC chamber and 90◦ relative to the beam,registered the γ-rays from the neutron-inducedreaction in 235U. The detectors were equippedwith BGO escape-suppression shields. Passive


shielding consisting of two 20 cm thick tungstenwalls installed at the upstream side of each de-tector was built. A copper ring with a length of22 cm was positioned at the end of the collima-tor to serve as a post-collimator and reduce thebeam halo. We installed an evacuated aluminumpipe with thin Kapton windows upstream fromthe collimator and the PPAC chamber in orderto reduce the scattering of neutrons in air. Inprevious measurements with an iron target, wedetermined that the vacuum pipe reduced thetime spread of the neutron beam. The resultsfrom this measurement are shown in Fig. 5.8.The whole setup was placed in a housing made ofparaffin blocks in order to reduce the accidentalbackground from neutrons scattered in the ex-perimental hall. Tests with the paraffin 2π half-housing revealed a 10% reduction of the acciden-tal background, as shown in Fig. 5.8.

0 50 100 150 200 250 300 350 4000

5

10

15

20

25

30

35

40

without vacuum pipe

with vacuum pipe

(ns)t

Co

un

ts /

1 n

s-3

10

0 50 100 150 200 250 300 350 4000

5

10

15

20

25

30

35

40

without paraffin housing

with paraffin housing

(ns)t

Co

un

ts /

1 n

s-3

10

range ofintegration

Figure 5.8: (Color online) Measurements with aniron target with and without a vac-uum pipe (top panel) and with andwithout a paraffin housing (bottompanel) covering 2π of the Ge detectors.

We carried out measurements with the de-scribed setup, without the 235U foils, withoutthe PPAC chamber and without the neutronbeam. These measurements helped us to decom-pose the background contributions from the alu-minum chamber, the accidental background andthe natural activity of the 235U foils in the mea-surement with the PPAC chamber . The resultsof this decompositions are shown in Fig. 5.9 inthe energy range where the 6+ → 4+ transitionsin 234U at 152.7 keV is expected.

100 110 120 130 140 150 160 170 180 190 2000

2

4

6

8

10

12

14

16

(keV)γE C

ou

nts

/ 0.

5 ke

V-3

10

Total

Al Chamber

Accidental

Nat. activity

Figure 5.9: (Color online) Decomposition of thebackground in the γ-ray spectrumfrom the measurement with thePPAC-based fission chamber. Thecurves show the total background andthe contributions of the scattered neu-trons from the Al chamber, the ac-cidental background and the back-ground due to the natural activity ofthe 235U target.

A redesign of the aluminum chamber or re-duction of the diameter of the neutron beamshould be performed, because scattering of thebeam in the chamber makes the greatest contri-bution to the background. Such a reduction ina future experiment will allow us to measure theefficiency of the PPAC and to deduce the detec-tion sensitivity for the transitions at 99.9 keV in234U.

This work was supported by the National Nu-clear Security Agency under the US Departmentof Energy grant DE-FG52-09NA29465.

[Hut09] A. Hutcheson et al., Phys. Rev. C, 80,014603 (2009).


5.2.3 Measurement of the 187Re(n,2nγ)186mRe Destruction Cross Section

C.R. Casarella, North Georgia College and State University, Dahlonega, GA; J.H. Kelley,R. Raut, C. Howell, G. Rusev, A.P. Tonchev, E. Kwan, W. Tornow, H.J. Karwowski,S.L. Hammond, TUNL; F.G. Kondev, S. Zhu Argonne National Laboratory, Argonne, IL

We are continuing a program to measure cross sections for 187Re(n,2nγ) reactions with par-

ticular interest in confirming a transition that has tentatively been identified as a doorway

transition feeding 186mRe at Ex = 149(7) keV. The cross sections are being measured using

pulsed, nearly mono-energetic neutron beams and an array of planar HPGe γ-ray detectors.

At present, the reaction cross sections for the 187Re(n,2nγ) reaction are poorly known, so

measuring the cross sections can have positive implications, for example, on the physics of

reactors, where Re can be produced. Furthermore, refining the cross section measurements

may reduce uncertainties in the Re/Os cosmochronometer.

The level structure of 186Re includes a long-lived isomeric state identified at Ex = 149(7) keVwith a half-life of 2 × 105 y. At present, noγ-ray transitions have been identified that feedinto the isomer [Bag03]. A recent conferenceproceeding reporting on a 186W(d,2n) experi-ment has suggested that the isomer is fed froman Ex = 324 keV level that decays sequen-tially via Eγ = 144.2 and 30 keV transitions[Kon06], but these observations have not beenconfirmed. In addition to uncertainties in thefeeding of the isomeric state, the situation for the187Re(n,2n)186,186mRe cross sections is also openfor great improvement, since the current data onthe cross sections are much lower than the crosssections given in the recent ENDF evaluations.

In general, studies of rhenium nuclei cover arange of topics including medical physics [Sze09],reactor physics [Yam94] and nuclear astrophysics[Mos10]. The latter two areas of study may be in-fluenced by improved cross section measurementsfor the 187Re(n,2n) reactions For example, the187Re isotope is produced as a fission fragment inreactors, where it acts as a neutron multiplier via(n,2n) reactions. Its effect as a catalyst speedingup chain reactions needs to be considered to en-sure optimal running conditions inside the reac-tor.

Improvement in the 187Re(n,2n)186,186mRecross sections may also benefit the Re/Os cos-mochronometer. The chronometer evaluation,which depends on the composition of 187Os(produced by the s-process) and 187Re (pro-duced by the r-process) in a meteorite, canbe interpreted to deduce the age of the speci-

men. While the (n,γ) and (γ,n) reactions areof primary importance for the chronometer, the187Re(n,2n)186,186mRe reactions also act to de-stroy cosmogenic 187Re, and improvements in thecross sections may provide improvements in thechronometer’s accuracy.

In previous years, we have reported on twomeasurements. In 2007 we attempted to mea-sure prompt γ rays emitted during an in-beammeasurement aimed at observing the transitionfeeding the isomer, and in 2008 we attempted anactivation experiment intended to measure thecross sections by standard activation and off-linecounting techniques. Feasible improvements havebeen implemented in the prompt (n,xnγ) mea-surements reported here. On the other hand,the extremely long half-life of 186Re and the highbackground from the t1/2 = 6 month isomer pro-duced in the 185Re(n,2n)184mRe reaction essen-tially rule out the activation technique as a fea-sible approach for measuring the cross sectionswith natRe samples.

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Direction of neutron beam

natRe Target

Tungsten Attenuators

HPGe Planar Detectors

Active Shield−Veto System

Liquid Nitrogen Dewars

Copper Attenuators

Pulsed beam of deuterons

Copper Collimators

H(d,n) gas cell2

Figure 5.10: Diagram of experimental apparatusused in May 2010.


In May 2010, 187Re(n,xnγ) reaction datawere taken using a 5 cm by 5 cm, 0.1 mmthick natRe target. A thin Fe foil matching thesize of the Re target was included in the tar-get package and will be used for neutron beamflux normalization. The target was placed be-tween two HPGe planar detectors surrounded bypassive shielding in the Shielded Source Area atTUNL (see Fig. 5.10). The target was bom-barded by a pulsed, quasi-monoenergetic beamof 12 MeV neutrons, and a neutron detector wasplaced at zero degrees with respect to the beamto act as a neutron-beam-flux monitor. Thebeam flux on the target was approximately 104

neutrons/(cm2s). Data from this measurementare presented here, though a subsequent mea-surement incorporated two planar HPGe detec-tors and an HPGe clover detector.

��

�

��

��

��

��

��

��

��

Figure 5.11: Spectrum illustrating observed γ-raytransitions near the 144 keV region.

The present γ-ray spectra are selected by gat-ing on the prompt neutron time-of-flight peak.The so-called time-of-flight spectra show a flatbackground with peaks corresponding to eventsproduced by 2H(d,n)3He prompt neutrons and2H(d,np)D breakup neutrons. Software gateson the “beam-on” and “beam-off” parts of theTOF spectrum are used to generate the associ-ated spectra, and the final spectrum shown inFig. 5.11 is obtained by subtracting the beam-off background spectrum from the beam-on spec-trum. Numerous transitions in the Re isotopes

are visible, including the suggested 144 keV tran-sition.

Figure 5.12: Level diagram showing presently ob-

served 186Re γ-ray transitions from185Re(n,n′γ)186Re reactions.

A simple analysis of the transitions populatedin the 187Re(n,2nγ) reaction is illustrated in thedecay scheme shown in Fig. 5.12. Unfortunately,the measured yields do not permit a coincidenceanalysis. A survey of the transitions shows ambi-guity in the placement of the 144 keV γ-ray. Fu-ture analysis of these data requires efficiency cor-rections in order to obtain the differential crosssections for the observed transitions and a deter-mination of the total cross sections. An analy-sis of the sequential-decay transitions may shedlight on which levels are involved in producing the144 keV transition yield. A further extension ofthis analysis must involve, for example, a Hauser-Feshbach analysis to determine the total (n,2n)cross section. A second approach in the analysiscenters on the 17.3 ns half-life of the Ex = 324keV level. Evaluation of the 144 keV emission-time characteristic may be the most convincingevidence to confirm the doorway state transitionand to evaluate the 187Re(n,2n)186mRe cross sec-tion.

[Bag03] C. Baglin, Nucl. Data Sheets, 99, 1(2003).

[Kon06] F. Kondev et al., In Int. Symp.on Nuclear Astrophysics, Nuclei in theCosmos - IX, p. 41, CERN, Geneva,Switzerland, 2006.

[Mos10] M. Mosconi et al., Phys. Rev. C, 82, 1(2010).

[Sze09] F. Szelecsenyi et al., J. Radioanal Nucl.Chem., 282, 261 (2009).

[Yam94] N. Yamamuro, Nucl. Sci. Eng., 118,249 (1994).


5.2.4 Neutron-Induced Reaction Cross-Section Measurements on GaAs

R. Raut, A.S. Crowell, B. Fallin, C.R. Howell, C. Huibregtse J.H. Kelly, E. Kwan,G. Rusev, A.P. Tonchev, W. Tornow, TUNL; D.J. Vieira, Los Alamos National Laboratory,Los Alamos, NM

Neutron-induced reaction cross sections on GaAs have been measured at ten different energies

from 7.5 to 15 MeV. Monoenergetic neutron beams were produced via the 2H(d,n)3He reaction

at the TUNL 10 MV tandem accelerator facility. The induced activity in the irradiated

samples was measured using HPGe detectors in a low-background counting area. The present

data can serve as a test for statistical model calculations over a wide range of neutron energies

and for a number of different reaction channels.

Cross section measurements of fast-neutron-induced reactions on GaAs are important for thecharacterization of this widely used semiconduc-tor in high radiation environments. Such studiescan have significant applications in the area ofhomeland security through nuclear forensics andthe Stockpile Stewardship program. From a ba-sic research perspective, these measurements areuseful for testing statistical model codes and re-fining their parameter sets. Some of the mea-surements on GaAs were previously reported[Rau09]. Here we record the complete set of mea-surements carried out over the last year.

Neutron activation of GaAs foils was carriedout at the 10 MV FN Tandem Accelerator inTUNL at ten different neutrons energies: En =7.5(2), 8.0(1), 8.5(2), 9.5(1), 10.2(1), 11.0(1),11.5(1), 12.5(1), 13.25(10), 14.0(1), 14.5(1), and15.0(1) MeV, where the numbers in parenthesesrepresent the energy spread of the incident beam.The monoenergetic neutron beam was producedvia the 2H(d,n)3He reaction, known for its highneutron yield in the energy regime of the mea-surements. The deuterium gas was contained ina 3 cm long cylindrical cell at a pressure of around2 atm and was sealed from the beamline vacuumby a thin Havar foil. The GaAs samples, 99.9%pure and cut into dimensions 1 cm × 1 cm, weremounted normal to the incident beam and 2.6cm downstream from the exit end of the gas cell.Natural Al and Au foils with the same dimensionswere irradiated together with the GaAs samplesin order to estimate the incident neutron flux.

The neutron flux was ∼ 107 s−1 and remainedconstant throughout the run. Following the ir-radiations, the samples were measured in a low-background counting area using HPGe detectors.

The γ-ray spectra acquired from the offlinemeasurements of the activated samples were an-alyzed to identify the reaction products and es-timate the respective γ-ray-transition peak areasin order to calculate the corresponding reactioncross sections. Table 5.1 lists the different re-action channels studied in the present measure-ments along with the half-life of the product andthe γ-ray-transition energy used to identify andcalculate the respective cross sections.

The cross sections for the reactions of interestfrom the present work are plotted in Figure 1 as afunction of the incident neutron energy En = 7.5-15 MeV. Statistical model calculations using thetalys [Kon08] code were carried out for these re-actions, and the results are plotted along with theexperimental numbers. The theoretical excita-tion functions are found to be in good agreementwith the experimental cross sections for most ofthe reactions. Similar calculations with the em-

pire [Her07] and gnash [You92] codes are cur-rently in progress, and the preliminary results forsome of the reactions have been shown in theplot. The existing measurements for the differentreactions and the latest cross-section-data evalu-ations have also been plotted.

This work is supported by the National Nu-clear Security Agency under Grant No. DE-FG52-06NA-26155.


Table 5.1: Neutron induced reactions on GaAs and the monitor foils measured in the present work.

Reaction Product Q-value Gamma IntensityChannel Half-life Energy

(keV) (keV) (%)GaAs Reactions

69Ga(n,2n)68Ga 67.71 m -10312.95 1077.34 3.2269Ga(n,p)69mZn 13.76 h -127.44 438.634 94.7771Ga(n,p)71mZn 3.96 h -2031.0 386.28 0.9375As(n,2n)74As 17.77 d -10243.76 634.78 15.475As(n,p)75Ge 82.78 m -393.63 264.6 11.4

Monitor Reactions197Au(n,2n)196Au 6.1669 d -8072.39 355.73 8727Al(n,α)24Na 14.997 h -3132.14 1368.626 99.9936

��

��

��

��

� � ��

��

��

��

��

� � ��

�

�

��

��

��

��

��

��

��

� ��

�

��

��

��

��

��

��

��

� !"��#��$��% ��&��' ��!()*#+�,,-��!./0�."1/�2%,&�

��3�4��$'�

��2��

��5��$6�

� !"��!��7��% ��."1/�2%,&�

��3�4��$'�

��2��

��5��$6�

� !"��!��7��% ��8�4$��4�4��."1/�2%,&�

��3�4��$'�

��2��

��.��.�

� !"��9��%��:��;��#��$��3< =��4��#4��!()*#+�,,-��2%,&��."1/

��3�4��$'�

��2��

��.��5��

� !"��9��#4��#��< ��>� $ ��=-��?�!("+�-�?�!("+�-��2%,&��."1/

��3�4��$'�

��2��

Figure 5.13: Plot of cross section against the neutron energies for the reactions studied in the presentmeasurements. Each plot also includes the theoretical calculations using the statisticalmodel codes, the latest evaluations and the available literature data.

[Her07] M. Herman et al., Nucl. Data Sheets,108, 2655 (2007).

[Kon08] A. J. Koning, S. Hilaire, and M. C. Due-jvestijin, In Proceedings of the Interna-tional Conference on Nuclear Data forScience and Technology, p. 211, 2008.

[Rau09] R. Raut et al., TUNL Progress Report,XLVIII, 109 (2009).

[You92] P. Young, E. Arthur, and M. Chadwick,Comprehensive Nuclear Model Calcula-tions, Introduction to the Theory andUse of the GNASH Code, TechnicalReport LA-12343-MS, Los Alamos Na-tional Laboratory, 1992.


5.2.5 Neutron Capture Experiments

G.E. Mitchell, B. Baramsai, A. Chyzh, D. Dashdorj, C. Walker, TUNL; OTHERS,Charles University, Prague and Rez Institute of Nuclear Physics, Rez, Czech Republic; North Car-olina State University, Raleigh, NC ; Los Alamos National Laboratory, Los Alamos, NM ; LawrenceLivermore National Laboratory, Livermore, CA

The neutron capture reaction is very important for a wide variety of pure and applied physics.

Our measurements utilize the DANCE array (at LANSCE/Lujan)—a highly efficient calorime-

ter consisting of 160 BaF2 detectors. The high degree of segmentation can be used to perform

neutron resonance spectroscopy. The study of the statistical γ-ray cascade from different res-

onances and for different multiplicities provides unique opportunities to test models of the

photon strength function.

The neutron capture reaction is important fora variety of pure and applied physics, rangingfrom stewardship science (radiochemical applica-tions) to astrophysics to advanced fuel cycle is-sues. We are performing neutron capture exper-iments with the Detector for Advanced NeutronCapture Experiments (DANCE) at the ManuelLujan Center at Los Alamos National Labora-tory. DANCE is an array of barium fluoridecrystals (162 segments with 160 crystals). Thehigh efficiency of DANCE (this is a calorimeterthat identifies capture by the γ-ray total energy)makes possible the study of very small samples,including radioactive samples. The high degreeof segmentation makes possible the separation ofthe statistical γ-ray cascade for different multi-plicities. One limitation of this calorimeter sys-tem involves setting the energy gate on the totalγ-ray energy. In order to accumulate sufficientstatistics, the range of this gate is typically oneMeV.

We first focused on the photon strength func-tion (PSF) in 95Mo. We tested various models bysimulating the statistical cascade with the pro-gram dicebox. The DANCE data provide in-formation on the γ-ray spectra for many differ-ent multiplicities. We label these as multi-step-cascade (MSC) measurements. Another mea-surement is the two-step-cascade (TSC) measure-ment, which is essentially a multiplicity-two mea-surement. The MSC measurements provide in-formation on many multiplicities but with poorresolution, while the TSC measurements haveexcellent resolution but are limited (by defini-tion) to only one multiplicity. To examine fur-ther the issue of how well the statistical model

works in this mass region, we also performed aTSC measurement on 95Mo at the Rez Institute.This measurement complements and extends theDANCE results, so we were able to study theTSC to 11 different final states. We determineda rather standard set of values for the level den-sity and the PSFs that fit all of the TSC datavery well.

We then adopted the parameter set that pro-vided this excellent agreement and applied it inthe simulation of our DANCE data with the sta-tistical model code dicebox. The result wasvery good agreement. Thus, there is strong ev-idence that the extreme statistical model worksvery well in this mass region. These DANCE re-sults have been published [She09].

Measurement of the neutron capturereaction has now been performed on152,154,155,156,157,158,160Gd. We are in the processof analyzing all of these data. The gadoliniumisotopes are very important both for practicalreasons and for pure physics. For example, mea-surement of the scissors-mode resonance for allof the gadolinium isotopes should clarify the de-pendence of this excitation mode on mass anddeformation. Our initial focus was on 155Gd and157Gd, because these isotopes have very largecross sections and important applications. Anal-ysis of these two nuclides formed the basis ofthe dissertations of A. Chyzh (157Gd) and B.Baramsai (155Gd). The scissors mode resonanceis required in order to obtain even a qualitative fitto the DANCE γ-ray spectra. The location andwidth of the resonance seem in reasonable agree-ment with other data, but the strength appearssomewhat low. Analysis of the complete set of


isotopes is needed before reaching any definitiveconclusions.

Another interesting aspect of these data isthe neutron spectroscopy. At DANCE we havemeasured neutron capture on 94,95,97Mo. Theanalysis of the recently measured 97Mo data is inprogress, and additional measurements are sched-uled for the summer of 2010. For s-wave reso-nances on 95Mo (ground state Jπ = 5/2+), thecompound-state s-wave resonances have J = 2or 3. The ground state of 96Mo has J = 0. Ifthe decay is statistical, one expects, on average,a higher multiplicity for the J = 3 states. Inaddition, the energy spectra for resonances withdifferent spin are different, especially for the mul-tiplicities M = 2 and 3. The combination of theaverage multiplicity and the M = 2 and 3 spectraleads to a very clear signature. There is excellentagreement with previous information on 96Mo.

In the study on 95Mo, determining the aver-age multiplicity was sufficient to determine theJ-values for the s-wave resonances. For the oddmass gadolinium targets, the average multiplicityis not sufficient to determine J . For 155Gd, theaverage multiplicities appear to separate into twogroups, but there are many ambiguities. A vari-ety of methods have been developed that enablethe determination of the spin J . One method isdescribed in the thesis of B. Baramsai [Bar10].If there were no experimental errors or Porter-Thomas (PT) fluctuations, then the values of themultiplicity distributions for the two s-wave spins(J = 1 and 2) would be points in a multiplic-ity hyperspace. The errors and PT fluctuationslead to two clusters rather than two points. De-termining which of the two clusters a given res-onance belongs to (and with what probability)is a problem in pattern-recognition theory. Wehave applied pattern-recognition methods to de-termining the spins of the s-wave resonances in

the compound system 156Gd. The s-wave leveldensity agrees well with previously establishedvalues. The separate level densities for J = 1and 2 obey the expected (2J + 1) dependenceextremely well.

In principle one can apply this analysismethod to more complicated cases, such as com-bined s and p sequences. For targets with spin,this leads to two s-wave and four p-wave states.In the Mo isotopes there are a significant numberof p-wave resonances, and the spins of the reso-nances are of astrophysical interest. One crit-ical issue is the determination of the parity ofthe resonance. After separation by parity, thepattern-recognition analysis method works rea-sonably well, even for the p-wave resonances. Forspin-zero targets, where there are only one s-waveand two p-wave possibilities, this analysis methodwork well even without the parity separation.

The DANCE array can be used to determinethe cross section, even in unfavorable cases wherethe scattering cross section is many times greaterthan the capture cross section. This is case for89Y, which is used as a radiochemical detector.The capture cross section for 89Y has been mea-sured with significantly reduced errors. Muchof the reduction in the cross-section errors in-volved the use of nuclides with similar ratios ofscattering-to-capture cross sections in order tobetter determine the backgrounds. The detailsare covered in the thesis of A. Chyzh [Chy09].

Much of this research is supported by an aca-demic alliances grant from the National NuclearSecurity Agency.

[Bar10] B. Baramsai, Ph.D. thesis, NCSU, 2010.

[Chy09] A. Chyzh, Ph.D. thesis, NCSU, 2009.

[She09] S. A. Sheets et al., Phys. Rev. C, 79,024301 (2009).

Photonuclear Reactions at HIγS

Chapter 6

• Nuclear Astrophysics

• γ-3He Interaction

• γ-4He Interaction

• Study of Many-Body Systems

110 Photonuclear Reactions at HIγS TUNL XLIX 2009–10

6.1 Nuclear Atsrophysics

6.1.1 Measurements of the angular distributions of the 12C(γ, 3α) reactionat Eγ ≈ 9.8 MeV with the O-TPC detector.

W.R. Zimmerman, M. Gai, University of Connecticut at Avery Point, Groton, CT ; M.W. Ahmed,S.S. Henshaw, C.R. Howell, P.-N. Seo, S. Stave, H.R. Weller, TUNL;

We have used the optical readout time projection chamber (O-TPC) detector to study the12C(γ, 3α) reaction at Eγ = 9.363, 9.556 and 9.771 MeV in search for the elusive 2+

2 state of12C. We developed a method to separate 12C and 16O dissociation events emanating from the

CO2 target. Pure E2 angular distributions were measured at all energies.

Carbon is formed during stellar helium burn-ing in the triple-alpha process, the 8Be(α, γ)12Creaction, that is mostly governed by the contribu-tion of the 0+ Hoyle state at 7.654 MeV. At hightemperatures (T > 3 GK) higher lying states in12C may contribute. Indeed a broad (Γ = 560keV, Γγ = 0.2 eV) 2+ state at 9.11 MeV in 12Cwas included in the NACRE compilation [Ang99]following a theoretical prediction [Des87] for the2+ member of the rotational band built on topof the 0+ Hoyle state at 7.654 MeV. It increasesthe production of carbon at temperatures in ex-cess of 1 GK by up to a factor of 15. A largerproduction of 12C at high temperatures increasesthe neutron density as required for an r-process,due to the competition between the 8Be(α, γ) re-action and the 8Be(n, γ) reaction [Del71, Pru05].A 2+ member of the rotational band build on topof the Hoyle state is not predicted in the conjec-tured alpha condensate [Fun09] which predictsa spherical Hoyle state. Evidence for a broad2+ state was found in a 12C(α, α′) experiment[Ito04] and at 9.6 MeV in a 12C(p, p′) measure-ment [Fre09] but such a state was not observedin the beta-decay of 12B and 12N [Hyl09].

We used our Optical-Readout Time Projec-tion Chamber (O-TPC) [Gai10] operating withCO2 gas and gamma beams extracted from theHIγS facility [Wel09] to search for such a 2+ statevia the identification of triple alpha events fromthe 12C(γ, 3α) reaction. We have studied this re-action at E = 9.363, 9.556 and 9.771 MeV.Onemajor draw back for using CO2 gas is that thedifference of Q-values for the dissociation of 16Oand 12C (112 keV) is considerably smaller thanthe beam width and comparable to the detec-tor resolution of approximately 80 keV [Gai10].

In addition the larger quenching factor for thelow energy 12C projectiles leads to a smaller gridcharge-signal from the dissociation of 16O withan energy very similar to that of the dissociationof 12C. Hence the total energy deposited in theO-TPC detector (grid charge signal) cannot beused to separate (and thus identify) 12C and 16Odissociation events.

The energy shared by the two body α + A2

decay differs for 16O and 12C events. We definethe ratio: R = E(α)

E(A2) . This ratio is 3 for 16O

events and 2 for 12C events. Due to the atten-uation factor of the outgoing 12C the measuredratio R for 16O events is in fact closer to 4. Thusthe ratio R can be used to distinguish and iden-tify 12C events. This ratio R can be measured bymeasuring the ratio of the emitted light as well asthe ratio of the track lengths measured in time.In addition a similar measurement can be per-formed using the pixel content of the CCD cam-era as long as the CCD camera provides cleansignals; a goal that has not yet been achievedwith our current camera but is anticipated withthe new camera we purchased. Thus the ratio Rcan be measured four times. A multiplication ofthe measured R values yields a separation whichis approximately a factor of 2 better since thefluctuations increase by only approximately thesquare root of 4. We also fit each event with aline shape calculated for an α + 12C event, yield-ing good χ2 for 16O events and bad χ2 for 12Cevents.

In Fig. 1 we show the so obtained surfaceplot of event identification using the ”effectiveχ2” and the multiplication of the ratio R as ob-tained from the PMT signal only; the ratio of theintegrated light (I) times the track length (L). In

TUNL XLIX 2009–10 Photonuclear Reactions at HIγS 111

Fig. 2 we show the same surface plot for eventsthat were clearly identified as 12C events by ob-serving kinks in the PMT signal or in the CCDimage. In both Figs. 1 and 2 we show data binsincluding at least two events. The separation of12C events using the above discussed method isestimated to yield an uncertainty which is smallerthan 10%.

05

1015

2025

02

46

810

121402468

α + A2 Event Identification A(γ,α)A2 at Eγ = 9.771 MeV

A2 = 12C

A2 = 8Be

“χ2”

Lα /L

A2 * Iα /IA2

Figure 6.1: Event identification surface plot (1count suppressed) for all events as dis-cussed in the text.

05

1015

2025

30

0

2

4

6

8

10

12

140

2

4

6

“χ2”Lα /L

A2 * Iα /IA2

Well Identied 12C(γ,α)8Be at Eγ = 9.771 MeV

Figure 6.2: Event identification surface plot (1count suppressed) for well identified12C dissociation events as discussed inthe text.

The in plane angle (α) measured by the trackregistered in the CCD image and the out-of-planeangle (β) measured by the Time projection sig-nal of the PMT allow us to deduce for each eventthe scattering angle (θ) and the azimuthal angle(φ) of the polar coordinate system used in scat-tering theory: Cosθ = Cosβ x Cosα and tanφ= tanβ

Sinα . The so obtained angular distribution is

shown in Fig. 3 together with that predicted fora pure 0+ → 2+ E2 transition. For these datawe used only in plane (β < 20◦) data for whichthe scattering angle (θ) is determined with highaccuracy.

0 20 40 60 80 100 120 140 160 1800

20

40

60

80

100

120

θ (Deg)

Cou

nts

12C(γ,α)8Be Eγ = 9.771 MeV

Jπ = 2+

(β < 20º - In Plane)

Figure 6.3: Measured angular distribution for in

plane (β < 20◦) 12C(γ, 3α) eventscompared to the prediction for a pure0+ → 2+ E2 transition.

[Ang99] C. Angulo et al., Nucl. Phys. A, 656, 3(1999).

[Del71] M. Delano and A. Cameron, Astr SpaceSci., 10, 203 (1971).

[Des87] P. Descouvemont and D. Baye, Phys.Rev. C, 36, 54 (1987).

[Fre09] M. Freer et al., Phys. Rev. C, 80,041303(R) (2009).

[Fun09] Y. Funaki et al., Phys. Rev. C, 80,64326 (2009).

[Gai10] M. Gai et al., submitted to JINST, 2010.

[Hyl09] S. Hyldegaard et al., Phys. Lett. B, 678,459 (2009).

[Ito04] M. Itoh et al., Nucl. Phys. A, 738, 268(2004).

[Pru05] J. Pruet et al., Astr. Jour., 623, 325(2005).

[Wel09] H. R. Weller et al., Prog. Part. Nucl.Phys., 62, 257 (2009).


6.1.2 Photoneutron Measurements of Astrophysically Important Statesin 26Mg

R. J. deBoer, M. Wiescher, J. Gorres, Joint Institute for Nuclear Astrophysics, Department ofPhysics, University of Notre Dame, Notre Dame, Indiana; R. Longland, C. Iliadis, G. Rusev,A. P. Tonchev, TUNL

The photoneutron cross section of 26Mg near the neutron-separation energy, which is of in-

terest for s-process nucleosynthesis, was studied at the High Intensity Gamma-Ray Source of

the Triangle Universities Nuclear Laboratory. A nearly mono-energetic and linearly polar-

ized γ-ray beam was used to scan the excitation energy range from 10.8 to 12.05 MeV as a

preparatory run for a high resolution time of flight measurement.

The s-process is a mechanism for the for-mation of many of the heavy isotopes up tolead. The s-process proceeds by neutron cap-ture onto stable light isotopes who subsequentlyβ-decay, but a mechanism for neutron produc-tion must be available. In stellar environments,22Ne can be produced by the reaction chain14N(α, γ)18F(β+)18O(α, γ)22Ne where 14N is oneof the main products of the CNO cycle. It hasbeen suggested [Ibe75] that a main source ofneutrons for the s-process is the 22Ne(α, n)25Mgreaction rate. Therefore, knowledge of the levelparameters of 26Mg in the critical energy rangefrom 10.9 MeV < Ex < 11.5 MeV, are essential.Photoexciation at the High Intensity Gamma-Ray Source (HIγS) provides an alternative ap-proach to the direct reaction method which ismade hampered by high coulomb penetrability.

A recent successful experiment, also per-formed at HIγS [Lon09], has prompted fur-ther investigation. While the last experimentfocused on photon scattering on 26Mg, the aimof the current experiment is to investigate levelsby 26Mg(γ, n). The main goal is to investigatekey resonances near the neutron-separation en-ergy by the time-of-flight (TOF) method. Theexperiment has been divided into two parts: apreliminary background and count rate measure-ment run, and the TOF measurement. The back-ground run has been completed with encouragingresults prompting the continuation of the secondpart of the experiment later this year. The fol-lowing summarizes the results obtained so far.

The experimental setup consisted of an en-riched 26MgO target placed at the center of a4π 3He neutron counter [Spr93] which was posi-tioned symmetrically along the beam axis. One

of the main goals of this part of the experimentwas to measure and then minimize background.To test the beam induced background several ini-tial runs were performed at 10.8 MeV, well be-low the 26Mg neutron-separation energy at 11.1MeV. Background was tested in five stages: ablank run, target holder only, aluminum scat-terer, a natural MgO sample, and the enrichedtarget. For the first two tests the backgroundwas found to be 20 neutrons/s. The Al scattererand the natural sample showed in ceases of 2 and4 neutrons/s respectively.

The enriched target was 26MgO and had atotal mass of 16.4185 grams and was isotopi-cally enriched to 99.41(6)% 26Mg (0.41(2)% 24Mgand 0.18(4)% 25Mg) with natural oxigen isotopicaboundance. Other impurities were of the orderparts per million. Of main concern was 25Mg,17O, 18O which have neutron-separation energiesof 7.3, 4.1, and 8.0 MeV respectively. Cross sec-tions for all isotopes had been measured in theenergy range of interest previously and are ofthe same order (mbarns) as the 26Mg. Since theamount these isotopes is small compared to the26Mg, minimal background contribution were ex-pected. The total count rate was found to beabout 30 neutrons/s at 10.8 MeV.

The main source of background was foundto be the beam induced background most likelyfrom lead near the beam pipe. The lead (γ, n)cross section is very high at these energies (∼200mb). In fact, the small lead plug used forbeam alignment was observed to produced countrates of ∼50 neutrons/s when it was placed infront and then behind the detector. An effortwas made to decrease the beam induced neutronbackground by shielding the front of the detec-


tor with cadmium sheets and Lithium and Boronloaded polyethylene blocks. Final analysis findsthe neutron background at 24 neutrons/s afterall shielding was in place.

0

2

4

6

8

10

11 11.5 12Eγ [MeV]

0

1

σ avg.

[m

b]

Figure 6.4: (Color online) Integrated cross sec-

tions of 26Mg(γ, n) and natMg(γ, n)(Preliminary). The upper plot showsthe 26Mg(γ, n) while the bottom plotshows the natMg(γ, n) cross section.The data shown in red trianglesare our recent measurements whilethose in black circles are those fromRef. [Ful71].

The 26Mg(γ, n) and natMg(γ, n) cross sec-

tions were measured at 14 energies between 10.8and 12.2 MeV as shown in Fig. 6.4. Count ratesvaried from ∼100 neutrons/s near the neutron-separation energy up to ∼1000 neutrons/s on topof the large resonance at Eγ = 11.7 MeV. The av-erage cross section was found to be in good agree-ment with that of Ref. [Ful71] and smaller un-certainties were obtained. The excitation curvewas mapped up to 12.05 MeV in order to scanthe first resolvable resonance at Eγ = 11.7 MeV.Area analysis (gγΓγ0Γn/Γ = 32(2) eV) of thispeak finds it to be in excellent agreement withthat of Ref. [Bag71].

[Bag71] R. J. Baglan, C. D. Bowman, and B. L.Berman, Phys. Rev. C, 3, 672 (1971).

[Ful71] S. C. Fultz et al., Phys. Rev. C, 4, 149(1971).

[Ibe75] I. Iben, Jr., Astrophys. J., 196, 525(1975).

[Lon09] R. Longland et al., Phys. Rev. C, 80,055803 (2009).

[Spr93] J. Sprinkle et al., Los Alamos NationalLaboratory, Los Alamos, New Mexico87545, la-12496-ms edition, 1993.


6.1.3 A New Astrophysical Reaction Rate for α+α+n

C.W. Arnold, T.B. Clegg, C.R. Howell, C. Iliadis, H.J. Karwowski, G.C. Rich, J.R.

Tompkins TUNL

A new astrophysical reaction rate for the three-body 〈ααn〉 reaction has been calculated using

new, high-precision 9Be(γ,n) cross section data. This rate is important for modeling nucle-

osynthesis in neutrino-driven winds from core-collapse Type II supernovae (νwSII). Reaction

network codes rely on accurate cross-section data to produce meaningful universal-nuclear-

abundance predictions. The production of heavy r-process nuclei under realistic νwSII con-

ditions is extremely sensitive to the 〈ααn〉 rate. For temperatures relevant to the r-process at

this astrophysical site, the present calculated rate is 24 to 30% larger than rates currently in

use. The effects of this new rate in reaction network codes are discussed.

The 4He(αn,γ)9Be reaction and the triple-α reaction are the two 3-body reactions knownto have significant astrophysical relevance fortheir ability to bridge the mass stability gaps atA = 5 and A = 8. The 4He(αn,γ)9Be reactionhas predominance over the 3α reaction in high-temperature, high-neutron-flux, and short-time-scale environments such as the neutrino-windfrom core-collapsed Type II supernovae. The for-malism for interpreting 3-body astrophysical re-action rates described by Angulo et al. [Ang99]was adopted, and is described in brief below.

Calculation of the astrophysical reaction ratefor 〈ααn〉 is performed in two steps. First twoα-particles with center-of-mass (COM) energy Einteract to form a metastable 8Be nucleus (τ ∼10−16 s). A neutron and a 8Be with COM energyE′ subsequently interact, forming stable 9Be. Fora given temperature, α-particles and neutronswill have a Maxwell-Boltzman distribution of en-ergies. To calculate the rate of production of 9Be,a double integral of the form

N2A 〈σv〉ααn = NA

(8πh

μ2αα

)(μαα

2πkbT

)3/2

×∫ ∞

0

σαα(E)Γα(8Be, E)

exp(−E/kbT )

×NA 〈σv〉n 8Be EdE (6.1)

with

NA 〈σv〉n 8Be = NA

(8πh

μ2n 8Be

)(μn8Be

2πkbT

)3/2

×∫ ∞

0

σn 8Be(E′; E)exp(−E′/kbT )E′dE′. (6.2)

was used, where μαα and μn 8Be are reducedmasses, and E and E′ are the formation energiesof the α + α and n+8Be systems respectively.

The α + α → 8Be cross section used in theouter integral over E of Eq. 6.1, has been mea-sured to high precision [Wus92]. High-quality ab-solute total cross section data for the 9Be(γ,n) re-action with an overall uncertainty of ± 5% wereobtained at TUNL’s HIγS facility [Arn09]. Breit-Wigner parameters for Γn and Γγ were deducedby fitting the cross-section data, and the resultsare shown in Table 6.1.

Table 6.1: Resonance parameters from the present work.

Jπ ER(MeV) Γγ(eV) Γn(keV)1/2+ 1.739 ± 0.002 0.730 ± 0.013 221.4 ± 6.75/2− 2.431 ± 0.006 0.097 ± 0.026 —1/2− 2.854 ± 0.019 1.752 ± 0.163 480 ± 255/2+ 3.024 ± 0.001 0.789 ± 0.054 271.0 ± 2.3


The total cross section is then the sum of fourBreit-Wigner resonances. The reciprocity theo-rem relates the 9Be(γ,n)8Be cross section to the8Be(n,γ)9Be cross section yielding

σn 8Be(E′; E) =∑

j= 12 , 52

12

(2j + 1)πh2

2μn 8BeE′

×Γn(9Bej , E′)Γγ(9Bej , E′ + E)(Eγ − Ej

R)2 + 14Γn(9Bej , E′)2

. (6.3)

This neutron capture cross section goes di-rectly into the inner integral over E′ of Eq. 6.2.

The present cross section yields a reactionrate that is ∼30% higher than the rate used byNACRE [Ang99] and ∼24% higher than the rateof Sumiyoshi et al. [Sum02], as shown in Fig. 6.5.These two rates have been widely used in reac-tion network codes over the last several yearsand have typically replaced the rate of Caugh-lan and Fowler [Cau88], which does not includeoff-resonance contributions.

Figure 6.5: (Color Online) Plots of the presentrate and of Ref. [Sum02] relative tothe rate of NACRE. An error band forthe NACRE rate is shown.

The relative increase of the present rate is ex-pected to effect the results from reaction networkcodes. For the same realistic νwSII conditions,the present rate will over-produce seed nuclei andlight r-process nuclei (60 < A < 120) and under-produce heavy r-process nuclei (A > 130) relativeto production under the rates currently in use.The reason for these relative differences in abun-dance is the sensitivity of r-process nucleosyn-thesis to the ratio of the densities of seed nucleiand neutrons. Under realistic νwSII conditions,production of seed nuclei, light r-process nuclei,

and heavy r-process nuclei are in competition forneutrons. The produced abundance of seed nu-clei determines the number of neutrons availablefor production of light and heavy r-process ele-ments. If seed nuclei are produced such that theabundance of neutrons is sufficiently diminished,then heavy r-process nuclei may not be producedat all. As an example of the extreme sensitiv-ity, a scenario exists where a 2.5% change in theneutron-excess parameter causes a truncation ofthe r-process at A =∼ 130 [Sur10].

Recently, a new cross-section and reactionrate calculation was extracted from 9Be(e,e′)data across the 1/2+ threshold resonance of 9Be[Bur10]. The rate of Burda et al. is reported tobe 10 to 60% smaller than the NACRE rate, incontrast to the present rate, which is as muchas 30% larger. The authors of Ref. [Bur10]point out that the intriguing disagreement withprevious (γ,n) data may point to a violation ofSiegert’s theorem, regarding the manner in whichthe (γ,n) cross section is extracted from (e, e′)data. The present cross section provides a newbenchmark for identifying the nature of the ap-parent violation, and the new rate will give im-proved constraints on the space of realistic pa-rameters at the νwSII site. Hopefully it will alsolead to better understanding of the role of TypeII supernovae in r-process nucleosynthesis.

This work was supported, in part, by U.S.Department of Energy Grants No. DE-FG02-97ER41033 and DE-FG02-97ER41041.

[Ang99] C. Angulo et al., Nucl. Phys., A656, 3(1999).

[Arn09] C. W. Arnold et al., TUNL Progess Re-port, XLVIII, 100 (2009).

[Bur10] O. Burda et al., Phys. Rev. C, 82,015808 (2010).

[Cau88] G. R. Caughlan and W. A. Fowler, At.Data Nucl. Data Tables, 40, 283 (1988).

[Sum02] K. Sumiyoshi et al., Nucl. Phys., A709,476 (2002).

[Sur10] R. Surman and G. McLaughlin, privatecomunication, 2010.

[Wus92] S. Wustenbecker et al., Z. Phys. A, 344,205 (1992).


6.2 γ-3He Interaction

6.2.1 Measurements of the Absolute Cross Section of the Three-BodyPhotodisintegration of 3He Between Eγ = 11.4 MeV and 14.7 MeVat HIγS

B.A. Perdue, M.W. Ahmed, S.S. Henshaw, P.-N. Seo, S. Stave, H.R. Weller, TUNL; P.P.

Martel, A. Teymurazyan University of Massachusetts, Amherst, MA

Measurements of the absolute differential cross sections of the 3He(γ,n)pp reaction have been

completed at four γ-ray energies. The measurements were made at the incident γ-ray energies

of 11.4, 12.8, 13.5, and 14.7MeV. It has been found that the shape of the outgoing neutron

energy distributions at a given scattering angle at 12.8, 13.5, and 14.7MeV agrees with

current theoretical predictions [Del09]. The shape of the neutron energy distribution at

Eγ = 11.4MeV was found to agree with both a phase-space-only shape and the shape predicted

by theory.

Two experiments were conducted at the HIγSfacility to provide the data for this study. Oneexperiment (denoted as experiment A) focusedon the collection of the data at 11.4 MeV, andthe other experiment (denoted as experiment B)acquired data at 12.8, 13.5 and 14.7 MeV. Bothexperiments utilized a pulsed beam of γ rays col-limated to a diameter of 25 mm. The neutronsresulting from the photon-induced three-bodybreakup of 3He were detected using the TUNLBicron liquid-scintillator detectors [Gon09]. Inexperiment A, the target was a 39.8 cm longGE180 aluminosilicate glass cylinder filled with≈ 12 atm of 3He gas. This polarized-3He targetwas developed by the Medium Energy Physicsgroup at Duke University [Kra07]. The targetused in experiment B was a 400 mL cylindricalaluminum bottle filled with 168 atm of 3He gas.

A simultaneous measurement of d(γ,n)p wasmade to monitor the γ-ray beam intensity and toprovide an absolute normalization for the 3He-breakup reaction. This experimental setup con-sisted of two TUNL Bicron liquid-scintillator de-tectors placed at 90◦ and a 47.4 mm long cylin-drical heavy-water target. The deuterium pho-todisintegration reaction has been measured tobetter than ± 3% in the energy range of the cur-rent experiments [Jau96].

The analysis of the data from both experi-

ments has been completed, and the results of theanalysis are presented in the Ph.D. dissertation ofPerdue [Per10]. Selected results from the analysisare shown in Fig. 6.6. The experimental data arecompared with the results of two geant4 simula-tions. In each of the figures, the solid black curveis the result of a simulation using a theoreticalcross section [Del09] as input. This cross sectionincludes the contribution of the Coulomb inter-action of the protons in the final state [Del05].The theory predicts a peak in the neutron en-ergy distribution at around 85% of the maximumneutron energy E max

n allowed by energy and mo-mentum conservation. The dashed curve is thesimulated result when a phase-space input crosssection is used. Since phase space does not pro-vide an absolute scale, the simulation results werenormalized to the total cross section predicted byDeltuva, Fonseca, and Sauer [Del09].

All the data shown are absolute cross sectionsand agree well with the magnitude of the pre-dicted cross section. The data at Eγ = 11.4 MeVare described well by both the phase space shapeand the theoretical shape. However, the highneutron energy threshold only allowed a smallportion of the neutron energy distribution to bemeasured. The data at the higher γ-ray energiesshow the peaking near 85% of E max

n , which is inagreement with the theoretical calculations.


(MeV)labnE

0 1 2 3 4 5

b /

sr M

eV)

μ ( n d

En

Ωdσ3 d

0

20

40

60 ° = 90labnθ

= 11.4 MeVlabγE

(a)

(MeV)labnE

0 1 2 3 4 5

b /

sr M

eV)

μ ( n d

En

Ωdσ3 d

0

10

20

30

40

50 ° = 90labnθ

= 12.8 MeVlabγE

(b)

(MeV)labnE

0 1 2 3 4 5

b /

sr M

eV)

μ ( n d

En

Ωdσ3 d

0

10

20

30

40

50° = 90lab

nθ

= 13.5 MeVlabγE

(c)

(MeV)labnE

0 1 2 3 4 5b

/ sr

MeV

)μ ( n

dE

nΩd

σ3 d0

10

2030

40

50 ° = 90labnθ

= 14.7 MeVlabγE

(d)

Figure 6.6: (Color online) Neutron energy spectra at Eγ = 11.4, 12.8, 13.5, and 14.7MeV for a detector

at θ labn = 90◦. Experimental data are plotted against two sets of simulated results, one using

a normalized phase-space cross section (dashed line) and the other using a theoretical crosssection [Del09] (solid line) as the input to the simulation. The neutron energy thresholdis indicated by the vertical line. The error bars represent the statistical uncertaintiesassociated with the data points.

[Del05] A. Deltuva, A. C. Fonseca, and P. U.Sauer, Phys. Rev. C, 72, 054004 (2005).

[Del09] A. Deltuva, Private Communication,2005, 2006, 2007, 2009.

[Gon09] D. E. Gonzalez Trotter et al., Nucl. In-strum. Methods, A599, 234 (2009).

[Jau96] W. Jaus and W. S. Woolcock,Nucl. Phys., A608, 399 (1996).

[Kra07] K. Kramer et al., Nucl. Instrum. Meth-ods, A582, 318 (2007).

[Per10] B. A. Perdue, Ph.D. thesis, Duke Uni-versity, 2010.


6.2.2 Three-Body Photodisintegration of 3He: Comparison of NeutronBackgrounds from Different Glasses

H. Gao, M. Huang, G. Laskaris, S. Stave, X. Qian, Q. Ye, Q.J. Ye, W. Zheng,TUNL

Experiment E-03-09 on three-body photodisintegration of 3−→He with double polarizations using

a circularly polarized photon beam, 3−→He(−→γ ,n)(pp), was approved for 180 hours of beam timeat HIγS. One major concern is the neutron background from the front and back windows ofthe target chamber of the hyper-polarized 3He target. An in-beam test took place in April2010 at the HIγS facility. We present preliminary results from the test at an incident photonenergy of 11.4 MeV using different glasses.

A one-day in-beam test was carried out at theHIγS facility in April 2010 on three-body pho-todisintegration of 3He by detecting neutrons.Eight liquid-organic-scintillator detectors filledwith BC-501A [Cro01] fluid from Bicron Corpo-ration were placed at 30◦, 50◦, 90◦, 130◦, and160◦ at a distance of 1 m from the center ofthe aluminum frame recently constructed by theTUNL capture group. The detectors at 50◦,90◦, and 130◦ were doubled. All the detectorswere configured by the TUNL capture group withlower discrimination thresholds in the MPD-4module [Rub07], allowing the detectors to runwith ∼98% live-time.

A vertical, multi-layer, motor-controlled sup-port was placed at the center of the frame andwas used to place different targets in the γ-ray beam. The targets were: (1) a liquid D2Otarget; (2) a Corning-1720-glass [Dan81] refer-ence cell with ∼100 torr of N2; (3) a thin alu-minum rod for calibration purposes; (4) a GE180-glass [Tho78] reference cell with ∼100 torr of N2;(5) empty; (6) a Corning-1720-glass Rb-only cellnamed PABLUM filled with 9.1±0.2 amg of 3He.PABLUM contains ∼100 torr of N2 (at roomtemperature) as a buffer gas. Targets (2) and(4) were 40 cm long, 20 mm in diameter and∼250 μm thick at the front and back windows.At the end of the test, target (4) was replaced bya pyrex-glass [Chy89] reference cell 40 cm long,30 mm in diameter and with a uniform wall thick-ness of 1 mm, filled with ∼97.5 torr of N2.

A vacuum pipe constructed by the TUNLcapture group was placed upstream in the γ-rayvault to minimize beam scattering from air. Theenergy of the photon beam was 11.4 MeV, andthe photon flux was ∼ 7.0×107 s−1. The beamwas monitored by a five-paddle system located af-

ter the 12-mm beam collimator. The five-paddlesystem gives consistent results for the estimationof the incident number of photons on each target.

Figure 6.7: (Color online) Comparison betweenneutron energy spectra from theCorning-1720-glass 3He PABLUM cell(black line with peak around 2.2 MeV)and the Corning-1720-glass N2 refer-ence cell (red line without this peak)for the left 50◦ detector.

Figure 6.7 shows a comparison between theneutron energy spectra from the PABLUMand Corning-1720-glass N2 reference cells af-ter pulse-shape-discrimination (PSD) and pulse-height (PH) cuts. The spectra were measured inthe left 50◦ detector. The PH cut is set to behigher than 0.2 MeVee for all the detectors. ThePSD cut varies between detectors. The Corning-1720-glass N2 reference cell is used to estimatethe neutron background due to the Corning 1720glass and the N2 inside the 3He cell. The signal-to-noise ratio is defined as the neutron yield of


the 3He cell corrected by the Corning-1720-glassN2 reference-cell yield divided by the Corning-1720-glass N2 reference cell yield.

After the time-of-flight/energy cut, which isfrom 1.1 to 2.6 MeV, the signal-to-noise ratiovaries from ∼3:1 for the 30◦ detector to ∼10:1for the left 90◦ detector. Two detectors—thoseat right 90◦ and 130◦—present higher levels ofelectronic noise. In those detectors the signal-to-noise ratio is ∼3:1. The low neutron backgroundcontribution from Corning 1720 glass was ob-served in online analysis during the experiment,and an extended beam position scan on both theGE180-glass and Corning-1720-glass N2 referencecells then followed.

Figure 6.8 shows the ratio of the normal-ized neutron yields at different laboratory anglesunder the aforementioned cuts for the Corning-1720-glass and GE180-glass N2 reference cells,corrected by the mean yield of the empty-targetruns for each detector. The position of the centerof the beam was 0.51 cm above the center of thecell in both runs. Due to its barium content, theGE180-glass N2 reference cell gives a much higherneutron background than the Corning-1720-glassN2 reference cell.

Figure 6.8: (Color online) The data points, shownwith their statistical uncertainties,give the ratio between the correctedneutron yields from the Corning-1720-glass and GE180-glass N2 refer-ence cells.

In addition, an in-beam test was carried outat HIγS in May 2009 [YeQ10]. The energy ofthe photon beam was 11.4 MeV, and the photonflux was ∼ 1.5 × 107 s−1, monitored by a three-

paddle system right after the 22 mm collimator.One goal of the test was to estimate the neu-tron background of the GE180- and pyrex-glassN2 reference cells. All the neutron events wereselected from the raw data after the multiplic-ity cut, PSD cut, PH cut, and time-of-flight cut.They had energies ranging from 1 to 2.5 MeV.The GE180-glass N2 reference cell was found togive a higher neutron background than the pyrex-glass N2 reference cell.

Although preliminary studies from the April2010 test run indicate that neutron backgroundsfrom the pyrex- and Corning-1720-glass N2 refer-ence cells are comparable, Corning 1720 glass hasa much lower contribution to the wall relaxationof 3−→He [Wag04] and a much slower 3He leakagerate [Smi98, Deu08] than pyrex glass.

Thus, Corning 1720 glass is considered tobe a serious candidate for the construction ofhyper-polarized targets of 3He for future double-polarized two-body or three-body photodisinte-gration experiments. New Corning-1720-glasscells are currently being made and will soon betested.

[Chy89] K. Chyung et al., Providing ReinforcedAlkaline Earth Aluminosilicate Glasses,U.S.Pat. No. 4,846,866, 1989.

[Cro01] A. Crowell, Ph.D. thesis, Duke Univer-sity, 2001.

[Dan81] P. Danielson et al., Glass for seal-ing to Molybdenium Metal, U.S.Pat.No.4,255,198, 1981.

[Deu08] A. Deur, Estimate of the 3He leakagefrom a sol-gel coated cell, Technical re-port, University of Virginia, 2008.

[Rub07] A. Ruben et al., In IEEE NuclearScience Symposium Conference Record,pp. 681–684. IEEE, 2007.

[Smi98] T. Smith, Ph.D. thesis, University ofMassachusetts, Amherst, 1998.

[Tho78] G. Thomas, High Temperature GlassComposition, U.S.Pat. No. 4,105,826,1978.

[Wag04] G. Wang, Ph.D. thesis, Caltech, 2004.

[YeQ10] Q. Ye et al., Eur. Phys. J. A, 44, 55(2010).


6.2.3 Two-Body Photodisintegration of 3He between 7 and 16 MeV

W. Tornow, J.H. Kelley, R. Raut, G. Rusev, S.C. Stave, A.P. Tonchev TUNL

We extended our previously reported data set for the two-body photodisintegration cross

section of 3He from 8.8 MeV down to 7.0 MeV and from 12.8 MeV up to 16.0 MeV, dou-

bling the number of data points in this important giant dipole resonance energy region of the3He(γ,p)2H reaction. Our extended data set puts on more solid ground our earlier conclusion

that our data differ significantly from the majority of the previous data, including the recent

data of Naito et al. obtained with a mono-energetic photon beam and a time-projection cham-

ber. However, our present data are in good agreement with recent theoretical calculations

up to about 10 MeV, but differ at higher energies.

New data were obtained for the cross sectionof the reaction 3He(γ, p)2H at 7.0, 8.0, 12.0, 14.0,15.0 and 16.0 MeV. Details of the experimen-tal setup and measuring technique are given inprevious TUNL Progress Reports [Est04, Est05,Tor08, Tor09]. Therefore, here we present onlyour results for the photodisintegration cross sec-tion of 3He between 7.0 and 16.0 MeV.

Figure 6.9 compares our still preliminary datato previous experimental data taken during thepast half century and calculations by the Lis-bon [Del07] and Krakow [Wit04] groups. Thedata represented by solid symbols were obtainedwith bremsstrahlung beams, except for our dataand the two recent data points by Naito et al.[Nai06] (red dots), which were measured witha mono-energetic photon beam (produced viaCompton backscattering of an external laser fromelectrons in a storage ring). The open sym-bols are from p-d capture experiments using theprinciple of detailed balance. The crosses showdata obtained with incident electrons using thelong wavelength approximation. Our data are inagreement with the very old bremsstrahlung dataof Fetisov et al. [Fet65] obtained with a bub-ble chamber, and with the old p-d capture dataof Belt et al. [Bel70]. They differ significantlyfrom all other bremsstrahlung and p-d capturedata. They are also in disagreement with thetrend of the data of Naito et al. [Nai06] mea-sured with a mono-energetic photon beam, andusing a time-projection chamber as target anddetector. The black curve is the result of a cal-culation by the Krakow group using the NN po-tential Av18 plus the UR-IX three-nucleon force(3NF) and full treatment of meson-exchange cur-rents (MECs). The Coulomb interaction is not

included in this calculation. The other threecurves are recent calculations from the Lisbongroup. Here, the blue curve is based on theNN potential CD-Bonn, while 3NF effects are in-cluded via Delta excitations. MECs are included,but the Coulomb interaction is neglected. Thegreen curve gives the result with the Coulombinteraction included, but without Delta degreesof freedom. Finally, the red curve represents thefull calculation, i.e., CD-Bonn + Delta & MECsand with the Coulomb interaction included. Thedifference between the blue and red curves is ameasure of the importance of Coulombs effects,while the difference between the green and redcurves shows the magnitude of 3NF effects.

[Bel70] B. D. Belt et al., Phys. Rev. Lett., 24,1120 (1970).

[Del07] A. Deltuva, 2007, Private Communica-tion.

[Est04] J. Esterline et al., TUNL Progress Re-port, XLIII, 110 (2004).

[Est05] J. Esterline et al., TUNL Progress Re-port, XLIV, 168 (2005).

[Fet65] V. Fetisov, A. Gorbunov, and A. Var-folomeev, Nucl. Phys., 71, 305 (1965).

[Nai06] S. Naito et al., Phys. Rev. C, 73, 034003(2006).

[Tor08] W. Tornow et al., TUNL Progress Re-port, XLVII, 100 (2008).

[Tor09] W. Tornow et al., TUNL Progress Re-port, XLVIII, 84 (2009).


Figure 6.9: (Color online) Comparison of present data (inverted blue triangles) for the reaction3He(γ, p)2H to previous experimental data and theoretical calculations. See text for details.

[Wit04] H. Wita�la et al., TUNL Progress Re-port, XLIII, 112 (2004).


6.3 γ-4He Interaction

6.3.1 The Photodisintegration Reaction 4He(γ, p)3H below 30 MeV

R. Raut, W. Tornow, M.W. Ahmed, A.S. Crowell, J.H. Kelley, G. Rusev, S.C. Stave,A.P. Tonchev, TUNL

The two-body photodisintegration cross section of 4He into a proton and triton was measured

between 22 and 30 MeV in 0.5 MeV energy steps. High-pressure 4He-Xe gas scintillators

of various 4He/Xe ratios served as targets and detectors. Pure Xe gas scintillators were

used for background measurements. A NaI detector was employed for the determination

of the incident photon flux. Our data are in fairly good agreement with recent theoretical

calculations of the Trento group, but differ significantly from the majority of the previous

data, including the recent data of Shima et al.

For many reasons the 4He nucleus is oftenconsidered as the link between the classical few-body systems, i.e., deuteron, triton and 3He, andmore complex nuclei. For example, one expectsthree-nucleon force (3NF) effects to be more im-portant in the four-nucleon (4N) system than inthe three-nucleon (3N) system. On the otherhand, the theoretical treatment of the photodis-integration of 4He is not as advanced as that of3He and 3H. Only recently it became possibleto calculate the photoabsorption cross section of4He [Qua04] with a realistic nucleon-nucleon po-tential model (Av18) and a 3NF (Urbana IX) us-ing the Lorentz integral transform method of theTrento group [Efr00]. The calculations reproducethe strong giant dipole resonance peak, but theyprovide only a 6% reduction in the cross sectionin this energy regime once the 3NF is turned on.This is a surprising result, because the same cal-culations yield a 35% enhancement of the totalphotoabsorption cross section at pion threshold.

As can be seen from Fig. 6.10, the exper-imental situation for the total cross section ofthe reaction 4He(γ, p)3H is very unsatisfactory.In the energy region between 20 and Eγ=35MeV the available experimental data are scat-tered considerably, and provide hardly any guid-ance for different theoretical approaches. Espe-cially, the recent data of Shima et al. (red dotsin Fig. 6.10) obtained with a quasi-monoenergeticphoton beam and a time-projection chamber areabout a factor of two lower than the predictionsof the Trento group at about 26 MeV.

The Q-value for the reaction 4He(γ, p)3H isQ=-19.81 MeV. The Q-value for the three-body

breakup of 4He is -26.07 MeV. According toShima et al. [Shi05], the three-body cross sec-tion in the energy region of interest is only about3% of the two-body cross section. In addition,events from the three-body breakup reaction areat lower pulse heights than the events of interest.

0.0

0.5

1.0

1.5

2.0

2.5

20 25 30 35

σ (

mb)

Eγ (MeV)

4He(γ,p)3H

Figure 6.10: (Color online) Exisiting data for

the reaction 4He(γ, p)3H in compar-ison to a calculation of the Trentogroup [Qua04]. The data shownby solid symbols were obtainedwith bremsstrahlung or monoener-getic beams. The data indicated byopen symbols were obtained from thecapture reaction 3H(p, γ)4He via de-tailed balance. The data of Shima etal. are given by the red dots.

We used high-pressure 4He-Xe gas scintilla-tors with total pressure of 750 psi as target anddetector. In such detectors the measured pulseheight is a linear function of the deposited en-ergy, independent of the nature of the stronglyionizing particles. For mono-energetic incident γ


rays of 26 MeV, the protons and tritons deposit atotal energy of 5.8 MeV, while the proton alonehas an energy of 5.3 MeV. In pure 4He gas of750 psi pressure, the range of those protons isabout 4.6 cm, resulting in unwanted edge effectsin our 5.1 cm diameter cylindrical stainless steelvessels (see Fig.6.14 in Sect. 6.3.2) of 1 mm wallthickness. Therefore, depending on incident γ-ray energy, we added Xe varying between 7 and47% admixture (and keeping the total pressure at750 psi) to provide the necessary stopping powerto reduce the proton range to less than 1 cm.Of course, the admixture of Xe increases consid-erably the number of electrons due to Comptonscattering of the incident γ rays and, in addition,increases their stopping power as well. In orderto account for photon induced reactions on Xeand the MgO coating on the inner surface of thescintillator vessel, runs were taken with identi-cal and gained-matched scintillator vessels filledwith Xe only.

Figure 6.11: Typical spectra at Eγ = 22 MeV (toppanel) and Eγ = 28 MeV (bottompanel) for 4He(γ, p)3H. Note that the22 MeV spectrum is plotted with alogarithmic scale on the y-axis.

Two 10 mm diameter and 10 cm long colli-mators made of lead defined the size and energyspread of the incident γ-ray beam. The actualγ-ray energy was determined using a calibratedNaI detector. It agreed within 30 keV with theenergy determined from the electron energy andFEL photon wave length. The absolute γ-rayflux was determined in the standard way usingthe combination of a NaI detector and the HIγSscintillator paddle system. In order to eliminatepile-up events in the 4He-Xe gas scintillator, andto operate the HIγS scintillator paddle systemused for relative flux determination in its lin-ear counting-rate range, an 8-in. long attenuatormade of copper was inserted into the γ-ray beam(inside of the concrete shielding wall some 50 m

from the location of the 4He-Xe gas scintillator),reducing the γ-ray flux on target to 106 γ-ray/s.

Figure 6.11 shows typical spectra obtainedwith incident γ-ray energies of 22.0 MeV for a710 psi 4He-40 psi Xe (top panel), and of 28 MeVfor a 500 psi 4He-250 psi Xe (bottom panel) gasscintillator, respectively. The yield at small pulseheights is dominated by electrons. The pulses ofinterest due to the protons and tritons from thetwo-body breakup of 4He produce the enhance-ment seen at higher pulse heights. The back-ground due to photon-induced reactions on Xeand the vessel wall extends to even higher pulseheights. The xenon only spectra show no en-hancement in the region of interest.

Figure 6.12 presents the preliminary data (tri-angles with 10% error bars) for the total crosssection of the reaction 4He(γ, p)3H in compari-son to the theoretical predictions of the Trentogroup (solid curve) and the recent data of Shimaet al. (circles). Our data are in close agreementwith the theoretical calculations, and therefore,in strong disagreement with the data of Shima etal., which have been considered by many as thestate-of-the-art data set. The statistical uncer-tainty of our data is 1% or below. The uncer-tainty in the incident γ-ray flux determination isabout 3 to 4%. In addition, we estimate that ourbackground subtraction procedure contributes anadditional uncertainty of about 3 to 6%.

Figure 6.12: (Color online) Preliminary results(blue triangles) for the total crosssection of the reaction 4He(γ, p)3H incomparison to the prediction of theTrento group (red curve) and the re-cent data of Shima et al. (green cir-cles with error bars).

[Efr00] V. Efros et al., Phys. Lett. B, 484, 223(2000).

[Qua04] S. Quaglion et al., Phys. Rev. C, 69,044002 (2004).

[Shi05] T. Shima et al., Phys. Rev. C, 72,044004 (2005).


6.3.2 The Photodisintegration Reaction 4He(γ, n)3He below 30 MeV

W. Tornow, J.H. Kelley, R. Raut, G. Rusev, S.C. Stave, A.P. Tonchev, TUNL

The two-body total photodisintegration cross section of 4He into a neutron and helion was

measured at incident photon energies between 25 and 30 MeV. High-pressure 4He-Xe gas

scintillators of various 4He/Xe ratios served as targets and detectors. Pure Xe gas scintillators

were used for background measurements. A NaI detector in combination with the standard

HIγS scintillator paddle was employed for absolute photon flux determination. Data analysis

is underway.

The determination of the 4He(γ, p)3H to4He(γ, n)3He cross-section ratio has a long his-tory. By accurately determining this ratio it ap-peared that Shima et al. [Shi05] had resolved thelong-standing problem associated with claims oflarge charge-symmetry breaking (CSB) effects inthese two reactions. Their result for the cross-section ratio agrees well with the recent calcu-lations of the Trento group [Qua04] without in-troducing CSB effects beyond those already in-cluded in the Av18 nucleon-nucleon potentialmodel. However, the individual cross sections forthe two reactions 4He(γ, p)3H and 4He(γ, n)3Hedisagree by about a factor of two with those cal-culations at the peak of the giant dipole reso-nance region at about 26 MeV, while the dis-agreement at 30 MeV is only about 10%. Fur-thermore, the findings of Shima et al. for the4He(γ, n)3He cross section at Eγ=24.3 and 26.5MeV are in disagreement with the very recentdata of Nilsson et al. [Nil07] at Eγ=24.6 and26.7 MeV, although the latter data have fairlylarge uncertainties. While Shima et al. used anearly 4π time-projection chamber to detect thecharged particles from the breakup of 4He, Nils-son et al. detected the emitted neutrons in theangular range between 30◦ and 130◦.

As can be seen from Fig. 6.13, the experimen-tal situation for the total cross section of the re-action 4He(γ, n)3He is very unsatisfactory. Over-all, the picture looks pretty similar to that givenin Sect. 6.3.1 for the reaction 4He(γ, p)3H. Thecurve is the theoretical prediction of the Trentogroup. The red dots are the data of Shima et al.,while the green upright triangles represent therecent data of Nilsson et al.

From an experimental point of view, the ”tar-get equal detector” scheme of Shima et al. is theideal experimental approach, because it allows

for the simultaneous measurement of both the4He(γ, p)3H and 4He(γ, n)3He cross sections, andin addition, it avoids the complications associ-ated with neutron detection. We used the ”targetequal detector” approach described in Sect. 6.3.1for the measurement of the 4He(γ, p)3H totalcross section by employing a high-pressure 4He-Xe gas scintillator.

0.0

0.5

1.0

1.5

2.0

2.5

20 25 30 35

σ (m

b)

Eγ (MeV)

4He(γ,n)3He

Figure 6.13: (Color online) Exisiting data for the

reaction 4He(γ, n)3He in compari-son to a calculation of the Trentogroup [Qua04]. The data shownby solid symbols were obtainedwith bremsstrahlung or monoener-getic beams. The data indicated byopen symbols were obtained from thecapture reaction 3He(n, γ)4He via de-tailed balance. The data of Shima etal. are given by the red dots.

The Q-value for the reaction 4He(γ, n)3He isQ=-20.58 MeV. Incident γ rays of Eγ=26 MeVprovide helions of energies between 1.14 and 2.2MeV, depending on the emission angle of the un-detected neutrons. The pulse height producedby strongly ionizing charged particles in gas scin-tillators is proportional to the energy depositedand independent of the particle type. For exam-


ple, 3 MeV protons, deuterons, tritons, helions,and 4He ions all produce the same pulse height,provided they are completely stopped in the gasvolume. Therefore, in principle the pulse-heightinterval associated with helions is clearly sepa-rated from the summed pulse height of the pro-tons and tritons from the reaction 4He(γ, p)3H,which deposit a total energy of 5.8 MeV for 26MeV incident γ rays. However, in standard 4He-Xe gas scintillators the pulse heights produced byCompton scattered electrons overlap with thosegenerated by some fraction of the low-energy he-lions, unless the Xe/4He ratio is kept very low. Inaddition, at high incident γ-ray flux, the electron-generated pulses tend to pile up on top of eachother, creating pulse heights beyond those ex-pected from the maximum range available withinthe scintillator vessel. Therefore, a quite com-plicated optimization process is required if onewants to measure the 4He(γ, n)3He cross sectionwith 4He-Xe gas scintillators. The required smallXe/4He ratio makes it impossible to measure the4He(γ, p)3H and 4He(γ, n)3He cross sections si-multaneously. Xe/4He ratios well below 5% notonly result in deteriorated energy resolution ofthe gas scintillator, but in addition, the protonsfrom the 4He(γ, p)3H reaction will range out, i.e.,run into the inner wall of the gas scintillator hous-ing without depositing their full energy. As amatter of fact, at very small Xe admixtures thereaction products of the reactions 4He(γ, p)3Hand 4He(γ, n)3He become indistinguishable, be-cause the pulse heights produced by tritons andhelions will start to overlap, due to the lack ofsufficient pulse height produced by the energeticprotons.

We used a high-pressure 4He-Xe gas scintilla-tor filled with a mixture of 710 psi 4He and 40psi Xe as target and detector. For backgroundestimation we used an identical scintillator hous-ing filled with 40 psi of Xe. Data were takenin 0.5 MeV energy steps between 27.0 and 28.5MeV. Experimental details are similar to thosegiven in Sect. 6.3.1. A pulse-height spectrum ob-tained with 28 MeV incident photons is shown

in the top panel of Fig. 6.14. The events dueto electrons are located at the low pulse heights.The broad and symmetric enhancement locatedat larger pulse-heights is due to the helions of in-terest. The symmetric shape is due to the sin2θangular distribution of the associated, but un-detected neutrons. Following this pulse-heightdistribution one notices the broad distributioncaused by tritons and ranged out protons, whichare also responsible for the ”saturated” events atthe very end of the pulse-height distribution.

Figure 6.14: Pulse-height distribution obtained

with a 710 psi 4He and 40 psi Xegas scintillator and 28 MeV incidentγ rays. See text for details.

Data analysis is currently underway. We planto extend our 4He(γ, n)3He total cross-sectionmeasurements to lower energies by using 570 psi4He-30 psi Xe and 380 psi 4He-20 psi Xe gas scin-tillators.

[Nil07] B. Nilsson et al., Phys. Rev. C, 75,014007 (2007).

[Qua04] S. Quaglion et al., Phys. Rev. C, 69,044002 (2004).

[Shi05] T. Shima et al., Phys. Rev. C, 72,044004 (2005).


6.4 Study of Many-Body Systems

6.4.1 Compton@HIγS on 209Bi from Eγ = 15-30 MeV

S.S. Henshaw, M.W. Ahmed, N.J. Brown, B.A. Perdue, S. Stave, H.R. Weller, TUNL;P.P. Martel, R. Miskimen, A. Teymurazyan, University of Massachusetts, Amherst, MA; C.S.

Whisnant, James Madison University, Harrisonburg, VA; R.M. Prior, M.C. Spraker, NorthGeorgia College and State University, Dahlonega, GA; R. Pywell, University of Saskatchewan,Saskatoon, Saskatchewan, Canada; G. Feldman, George Washington University, Washington, DC ;A.M. Nathan, University of Illinois at Urbana-Champaign, IL

As part of the ongoing Compton@HIγS effort, data have been collected to investigate the

isovector-giant-quadrupole-resonance (IVGQR) region in 209Bi at Eγ = 15-30 MeV. Linearly

polarized �γ rays were incident on a 209Bi target, and scattered γ rays were detected in the

HIγS NaI Detector Array (HINDA). Angular distributions, both parallel and perpendicular

to the plane of polarization, are used to study the sensitivity to the IVGQR parameters.

Preliminary extraction of the IVGQR resonance parameters for 209Bi are reported.

In June 2009 the Compton@HIγS group per-formed an experiment to commission the HIγSNaI Detector Array (HINDA) and measure theCompton-scattering cross section from 209Bi inthe giant resonance region [Hen09]. Althoughthese data were of high quality, the quantity wasinsufficient to make an unambiguous statementregarding the parameters of the isovector giantquadrupole resonance in 209Bi. For this reason,in May 2010 six HINDA detectors were placedboth in and out of the HIγS polarization plane atscattering angles of θ = 55◦ and 125◦, as shownin Fig. 6.15, in order to measure the IVGQR pa-rameters.

γ−ray

Figure 6.15: (Color online) May 2010 detectorsetup showing six HINDA detectors.The arrow indicates the γ beam di-rection.

During the run, a linearly polarized γ-raybeam was incident on an isotopically pure (>

99.9%) 209Bi target for approximately 80 hoursover the energy range Eγ = 15-30 MeV. Duringthe running at 15 MeV, a 12C target was placedin the beam to obtain the intrinsic detector re-sponse using the mono-energetic 15.1 MeV NRFγ rays. A typical spectrum from this 12C run isshown in Fig. 6.16 along with a fit from a geant4

simulation.

Energy (MeV)

Co

un

ts

0

20

40

60

80

100

120

140

9 10 11 12 13 14 15 16

NaI Detector Response

GEANT4 Simulation

Figure 6.16: (Color online) The 15.1 MeV NRF γ

ray in 12C The red (thick) line is afit to the data using the geant4 sim-ulation convolved with a gaussian.

The simulation calculates the total energy de-posited by mono-energetic γ rays, taking into ac-count target geometry, detector geometry, energyand intensity loss of the γ rays through differentmaterials, and multiple scattering effects. Thesimulation output is convolved with a Gaussianto account for detector resolution effects not in-cluded in the simulation. Once convolved and


fit, the simulation output can be used to predictthe HINDA response to Compton-scattered γrays from 209Bi by convolving a second Gaussianrepresentative of the γ-ray beam energy width.An exponential background is added to accountfor atomic Compton-scattering events from 209Bi,which is a minimal effect at backward angles butplays a major role at forward angles where theatomic Compton scattering is large.

Energy (MeV)

Co

un

ts

0

10

20

30

40

50

60

70

80

12 12.5 13 13.5 14 14.5 15 15.5 16

NaI Detector ResponseFit ResultFit Result: Photo PeakFit Result: Background

Figure 6.17: (Color online) Typical fitting result

from the 209Bi scattering data.

Figure 6.17 shows the results of a fit to a for-ward angle detector at Eγ = 15 MeV. The solidline is fitted to the spectrum, and the peak andbackground line shapes are extracted. The back-ground is subtracted bin-by-bin from the data,and the counts left over are summed to give theγ-ray yield. This procedure is used for each de-tector at each energy step. Thirteen energy stepswere made for Eγ = 15-25 MeV in 80 hours ofbeam on target. The in-plane detector yields aresummed for each scattering angle, fore (θ = 55◦)and aft(θ = 125◦), and the ratio of perpendicularto parallel yields (σ⊥/σ‖) is formed. Figure 6.18shows the results of this procedure as a functionof γ-ray energy (Eγ). The solid lines correspondto a preliminary fit of a model to the fore and aftratios, respectively. This is the first measurementconfirming the phase difference between the foreand aft ratio curves and exploiting this fact tobetter determine the IVGQR parameters. Thephase difference is due to the cos(θ) dependenceof the interference term, which changes sign whengoing from a forward angle to a backward one.

The model uses a phenomenological calcula-tion of Wright [Wri85], which starts with thephoto-absorption cross sections, the charge andexchange form factors, and the nucleon polar-izabilities and uses the optical theorem anddispersion relations to calculate the Compton-scattering cross section. The giant-dipole-resonance (GDR) parameters come from previ-

ous studies of Dale et al. [Dal88], and theform factors are from (e, e′) studies done by Eu-teneuer et al. [Eut78]. The IVGQR parame-ters are free parameters in the fit. The prelimi-nary parameters extracted from this analysis areEres = 23.0 ± 0.13 MeV, Γres = 3.9 ± 1.4 MeV,and σ

[−2]res = 0.6 ± 0.04 IVEWSR. Here the to-

tal energy-weighted integrated strength σ[−2]res is

given in units of the isovector energy-weightedsum rule (IVEWSR) [Liu76], and the errors arepurely statistical. The scale of σ

[−2]res is deter-

mined from the input GDR cross section given byDale [Dal88]. Results for these parameters in theneighboring 208Pb nucleus are Eres = 20.2 ± 0.5MeV, Γres = 5.5±0.5 MeV, and σ

[−2]res = 1.4±0.3

IVEWSR.Much work is still needed to characterize the

systematic errors, as well as the robustness of thefitting procedure. This will be done in the com-ing months and will result in a PhD thesis.

(MeV)γE14 16 18 20 22 24 26

σ

/

σ

0

1

2

3

4

5

)°=125θHenshaw et. al ()°=55θHenshaw et. al (

)°=125θIVGQR Fit ()°=55θIVGQR Fit (

) °=125θNo IVGQR () °=55θNo IVGQR (

Figure 6.18: (Color online) Preliminary fit to theperpendicular-to-parallel ratios ex-tracted from the May 2010 data. Thesolid curves are fits to the data usingthe model described in the text. Thedashed lines show the ratios when noIVGQR strength is present.

[Dal88] D. S. Dale et al., Phys. Lett., B214, 329(1988).

[Eut78] H. Euteneuer et al., Nucl. Phys., A298,452 (1978).

[Hen09] S. S. Henshaw et al., TUNL ProgressReport, XLVIII, 94 (2009).

[Liu76] K. Liu and G. Brown, Nucl. Phys.,A265, 385 (1976).

[Wri85] D. H. Wright et al., Phys. Rev. C, 32,1174 (1985).


6.4.2 Fine and Gross Structure of the Pygmy Dipole Resonance

A.P. Tonchev, S.L. Hammond, J.H. Kelly, E. Kwan, G. Rusev, W. Tornow, TUNL; H.

Lenske, N. Tsoneva, Institut fur Theoretische Physik, Universitat Gieβen, Gieβen, Germany

High-sensitivity studies of E1 and M1 transitions observed in the reaction 138Ba(�γ,γ′)at energies below the one-neutron separation energy have been performed using the nearlymonoenergetic and 100% linearly polarized photon beams of the HIγS facility. The electricdipole character of the so-called “pygmy” dipole resonance was experimentally verified forexcitations from 4.0 to 8.6 MeV. The fine structure of the M1 “spin-flip” mode was observedfor the first time in N = 82 nuclei.

In stable and weakly bound neutron-rich nu-clei, a resonance-like concentration of dipolestrength has been observed at excitation ener-gies around the neutron separation energy. Thisclustering of strong dipole transitions has beennamed the pygmy dipole resonance (PDR). It wassuggested that the oscillation of a small portionof nuclear matter relative to the rest of the nu-cleus is responsible for the generation of enhanceddipole strength. The theoretical calculations inthe framework of the quasiparticle random-phaseapproximation (QRPA) indicated a correlationbetween the observed total B(E1) strength ofthe PDR and the neutron-to-proton ratio N/Z[Tso08].

The existence of the PDR mode near the neu-tron threshold also has important astrophysicalimplications, while for very neutron-rich exoticnuclei, the PDR will be an important topic ofstudy at the new generation of radioactive-ion-beam facilities. The evolution of the giant dipoleresonance (GDR) strength from stable to moreweakly bound nuclei with extreme neutron-to-proton ratios is an interesting and open question.

In the present work we report first on the ex-perimental determination of the character of thePDR as a predominantly E1 mode of excitation.Furthermore, the fine structure of the dominantE1 and dispersed M1 strength distributions be-low the particle separation energy is determined.Finally, the dynamics of the PDR as an interplaybetween isoscalar and isovector excitation modesis revealed. Our experimental study focuses onthe 138Ba(�γ,γ′) reaction below the neutron sepa-ration energy of about 8.6 MeV using the nearlymonoenergetic and 100% linearly polarized pho-ton beams of the HIγS facility.

The photon flux was measured by Comptonscattering of the beam from a 1.0 mm thick cop-per plate. At energies above 8.1 MeV, the photon

flux was also monitored by the 197Au(γ,n) reac-tion.

Figure 6.19: (Color online) Upper panel: Elas-tic (σγγ), inelastic (σγγ′ ), and to-tal absorption cross sections (σγ) in138Ba below the one-neutron sepa-ration energy (Sn=8.6 MeV) aver-aged over the beam energy spreadof about 3%. Lower panel: Totalabsorption cross section (σγ) fromthe present photon-scattering exper-iment is combined with the (γ,xn)data from the GDR. A Lorentziancurve is applied to the experimentalphotoneutron cross-section data.

The 138Ba target consisted of 4290.5 mg ofBaCO3 powder enriched to 99.68% in 138Ba. Itwas placed into an evacuated plastic tube at thecenter of an array of five large-volume HPGe de-tectors. These detectors were arranged around


the target at (θ, φ) = (90◦, 90◦ and 270◦), (90◦,0◦ and 180◦), and (135◦, 0◦), where θ is the po-lar angle of the outgoing radiation with respectto the horizontally polarized incoming photonbeam, and φ is the azimuthal angle measuredfrom the polarization plane.

The 100% polarized and nearly monoener-getic photon beam allows us not only to deter-mine unambiguously the electromagnetic charac-ter of the transitions, but it also enables a closerexamination of the dynamics of the decay pat-tern in the PDR region. The elastic (σγγ), in-elastic (σγγ′), and photoabsorption (σγ = σγγ +σγγ′) cross sections for the 138Ba(�γ,γ′) reactionbelow the neutron separation energy are shownin the upper panel of Fig. 1 in [Ton10]. The elas-tic cross section, which is due to ground-statetransitions, shows resonance-like structures anddominates the nuclear dipole response up to theneutron separation energy. In contrast, the in-elastic cross section, which is due to statisticaldecays, shows an exponential increase with ex-citation energy. The deduced photoabsorptioncross section (σγ) is shown in the lower panel ofFig. 6.19 along with experimental data for the138Ba(γ,xn) reaction in the GDR region. As canbe seen, for certain energies below the neutronseparation energy Sn, the photoabsorption crosssection has values larger than the extrapolationof the tail of the GDR implies.

Figure 6.20: (Color online) Experimental (toppanel) and QRPA calculations (lowerpanel) for E1 and M1 elastic scat-tering cross sections in 138Ba below9.0 MeV. The cross sections are av-eraged over 0.2 MeV energy bins.

In Fig. 6.20 QRPA results for the electric and

magnetic photoabsorption cross sections are pre-sented in comparison to the present experimen-tal data for the 1± states in 138Ba at E∗ < 8.5MeV. The measured elastic E1 and M1 cross sec-tions, integrated over 200 keV bins, are shown inthe upper part of Fig. 6.20. At excitation ener-gies below the particle emission threshold, the E1transitions in 138Ba are distributed in a broad,resonance-like structure. The observed concen-tration of 1− states is characterized by the highlevel density which increases steeply towards thethreshold. In contrast, the magnetic dipole tran-sitions, which result from the decay of single-particle states, are much more isolated, and theyare concentrated in well specified regions.

The comparison of E1 and M1 strengths overthe whole energy range between 4 and 8.6 MeVshown in Fig. 6.20 leads to the conclusion thatE1 transitions clearly dominate M1 transitionsby intensity and amplitude. The ratio of elasticM1 and E1 cross sections is only 2.7 (0.8)%.

In conclusion, the systematic spin and paritymeasurements on 138Ba at the HI�γS facility haveunambiguously verified that the observed dipolestrength from 4 MeV to the neutron separationenergy is predominantly of electric dipole nature.In addition, the fine structure of the M1 “spin-flip” mode was observed for the first time in N =82 nuclei. The QRPA predictions for the charac-ter and strength of this dipole mode of excitationare confirmed by our measurements. Enhanceddipole strength above the Lorentzian extrapola-tion of the GDR has been directly measured forelastic and inelastic transitions below the neutronseparation energy and found to be related to thedominance of electric transitions. This strengthfalls into two categories: elastic, originated by theskin oscillation of the excess neutrons against theproton-neutron saturated core, and inelastic, dueto states coupled to more complicated configura-tions than 1p-1h states. Combining these twostrengths, they define the gross structure of thePDR resonance, which exhibits a resonance-likeshape distribution with pronounced peak struc-ture above the low-energy extrapolation of theGDR. According to the theoretical model, elec-tric and magnetic transitions are found to be ofsingle-particle character, but related to differentphysical mechanisms.

[Ton10] A. Tonchev et al., Phys. Rev. Lett., 104,072501 (2010).

[Tso08] N. Tsoneva and H. Lenske, Phys. Rev.C, 77, 024321 (2008).


6.4.3 Dipole Excitations of 76Se in the Energy Range 4-9 MeV

V. Werner, P. Goddard, M. Smith, N. Cooper, Yale University, New Haven, CT ; D. Savran,N. Pietralla, Yale University and Technischen Universitat Darmstadt, Hessen, Germany; J.H.

Kelley, R. Raut, G. Rusev, A.P. Tonchev, W. Tornow, TUNL; A. Chakraborty, B.

Crider, E. Peters, S.W. Yates, University of Kentucky, Lexington, KY

The chains of Ge and Se isotopes are interesting from many nuclear structure points of

view. The influence of increasing deformation on the pygmy resonance is investigated in the

isotope 76Se using the 100% polarized and nearly-monoergetic beam at HIγS to complement

measurements made at Darmstadt. Determining and understanding the structure of this

isotope is also important because of its role in the search for neutrinoless double-beta decay.

The chains of Ge and Se isotopes are in-teresting from many nuclear structure points ofview. For the lighter isotopes, coexisting struc-tures of different deformation are known. Theheavier, stable or near-stable isotopes are transi-tional from spherical to deformed nuclei, as onemoves away from the Z = 28 shell closure. Theinfluence of increasing deformation on the pygmyresonance is investigated in the isotope 76Se. De-termining and understanding the structure of thisisotope is also important because of its role in thesearch for neutrinoless double-beta decay.

The pygmy dipole resonance (PDR) is a struc-ture of enhanced dipole strength, which is su-perimposed on the low-energy tail of the giantdipole resonance. The PDR has so far been stud-ied mainly in nuclei near shell closures (see, e.g.,Refs. [Sch96, Adr05, Wie09, Sav08]). The com-monly used intuitive picture of the PDR is thevibration of a proton-neutron core against a neu-tron skin. Therefore the magnitude of the res-onance is expected to become smaller in moreneutron-deficient nuclei. A splitting of the PDRinto isoscalar and isovector parts has been ob-served experimentally [Sav06, End09] and was re-cently treated in theory [Paa09].

The evolution of the PDR with deformationis not yet known. The nucleus 76Se is located 6protons above the Z = 28 shell closure, half-waybetween the closed shell and the mid-shell point.Because deformation is increasing (β2(76Se) =−0.28(1) compared to β2(74Ge) = −0.19(2)), wecan quantitatively investigate its effects on thestrength and location of the PDR in 76Se. Theseresults will be important for a further systematicanalysis and interpretation of the nature of thePDR.

Another important aspect of 76Se is its role asthe daughter nucleus in the neutrino-less double-beta decay of 76Ge. In order to derive a neu-trino mass from searches for this exotic decaymode involving large volume 76Ge detectors, itwill be necessary to know the nuclear matrix el-ements for the decay. However, those matrix ele-ments can only be extracted from models like theshell model or the quasi-particle random-phaseapproximation (QRPA) (see, e.g., Ref. [Suh08]).Because QRPA calculations, in particular, areoften used to describe the PDR (see the re-cent review article [Paa07]), a test of the QRPAagainst the measured dipole strength distributionshould help validate the model’s nuclear matrixelements.

The present experiment is one of a series ofmeasurements being performed at Yale Univer-sity, the Technischen Universitat (TU) Darm-stadt, and the University of Kentucky, probinglow-spin structures in 76Ge, 76As, and 76Se, usinga variety of techniques. A pilot experiment at TUDarmstadt using nuclear resonance fluorescencehad been performed in order to obtain the distri-bution and absolute excitation strengths of dipolestates at energies up to 9 MeV. The experimentat HIγS also involved photon scattering, but us-ing the fully polarized, nearly mono-energeticbeams delivered by the free electron laser. Theaim was to directly measure the parities of dipoleexcited states, in order to be able to separate E1from M1 excitations. In addition, the photon fluxat high energies is larger than that at TU Darm-stadt from bremsstrahlung, allowing the observa-tion of many more excited states.

Analysis of the spectra from our last runon 76Se at HIγS is nearly complete. In March


2010, we had 60 hours of beamtime and scannedthe energy region from 5 to 9 MeV with abouttwenty energy settings. At each setting, stateswithin the energy window of the beam were se-lectively excited. Figure 6.21 contains sample re-sults, showing the simulated photon beam andthe spectra corresponding to detectors in the hor-izontal and vertical planes.

0

50

100

6400 6600 6800Energy (keV)

0

50

100

Vertical

HorizontalCou

nts

Figure 6.21: (Color online) Example spectra from

the March run on 76Se. The simu-lated beam profile at the 6.6 MeVbeam setting is shown along withthe observed peaks from nuclear res-onance fluorescence in that region.Peaks observed only in the verticaldetectors (upper plot) correspond toground state decays of E1 excitedstates.

Due to the rather high background fromCompton-scattered photons, the analysis of tran-sitions from dipole excited states to lower-lyingexcited states is only possible in a few cases, andonly for transitions to the first excited 2+ states.However, these cases can be compared to Darm-stadt data to obtain multipole mixing ratios, es-pecially for transitions from M1 excited states.

The determination of the parities is complete,and the results are shown in Fig. 6.22. Most ob-served states are E1 excitations, but a numberof states with positive parity have also been ob-served. We are currently determining the exci-tation strengths of several newly observed statesrelative to those observed at TU Darmstadt, inorder to obtain a complete picture of the E1 ex-citation strength in 76Se up to 9 MeV. This in-cludes the region of the pygmy resonance. Futureexperiments will be dedicated to dipole states be-low 5 MeV, as well as to analogous experimentson the double-beta decay partner 76Ge.

One peak at about 5.3 MeV energy is ob-

served rather strongly in both the horizontal andvertical Ge detectors of the polarimeter setup.The solid angle of the detectors do not explainthe observed strengths within about a 4-sigmaconfidence level. Therefore, it is likely that theobserved peak is due to a doublet of transitionsonly few keV apart, containing both an M1 andan E1 excited state. This clearly demonstratesthe power of the polarimeter setup.

Energy (MeV)

Asy

mm

etry

8 9 7 6 5 4−2

−1

0

1

2

Figure 6.22: (Color online) Asymmetries for theobserved states, states with valuesnear -1 have negative parities, thosewith values near +1 have positiveparity.

[Adr05] P. Adrich et al., Phy. Rev. Lett., 95,132501 (2005).

[End09] J. Endres et al., Phy. Rev. C, 80, 034302(2009).

[Paa07] N. Paar et al., Rep. Prog. Phys., 70, 691(2007).

[Paa09] N. Paar et al., Phy. Rev. Lett., 103,032502 (2009).

[Sav06] D. Savran et al., Phy. Rev. Lett., 97,172502 (2006).

[Sav08] D. Savran et al., Phy. Rev. Lett., 100,232501 (2008).

[Sch96] R. Schwengner et al., Phys. Rev. C, 78,064314 (1996).

[Suh08] J. Suhonen and M. Kortelainen, Int. J.Mod. Phys. E, 17, 1 (2008).

[Wie09] O. Wieland et al., Phy. Rev. Lett., 102,092502 (2009).


6.4.4 Cross-Section Measurements of the 180mTa(γ,n) Reaction and thePhoto-Induced Depopulation of 180mTa Isomer

R. Raut, J.H. Kelly, E. Kwan, G. Rusev, A.P. Tonchev, W. Tornow, TUNL;

Cross-section measurements were carried out at HIγS for the 180mTa(γ,n) reaction with Eγ

from 6.5 to 7.57 MeV. A 4π, 3He proportional counter was used to detect the emitted neutrons,

and attempts were made to derive the cross section from their yield. Background from the

Compton scattered photon beam was observed obscuring the neutron counts. Simultaneous

measurements on the photo-deexcitation of the 180mTa isomer were also carried out and have

indicated positive results.

One of the key question in nuclear astro-physics pertains to the nucleosynthesis processescontributing to the observed abundances of thechemical elements in the universe. Productionof elements beyond Fe (Z = 26) is primarily at-tributed to s- and r-process neutron capture. Inaddition, there are 35 neutron-deficient nuclei,called the p-process nuclei, that cannot be pro-duced by the neutron capture processes. Theseare often ascribed to photo-dissociation reactionsin hot stellar environments. The 180Ta nucleus isthe rarest isotope in the universe, existing onlydue to its isomer at 75.3 keV (Jπ=9−) with ahalf-life greater than 1.2×1015 years. It is, infact, the only naturally occurring isomer.

However, the puzzle is not in the rarity ofthe isotope but rather in its abundance. The180mTa nucleus sits beside the s-process path,which proceeds through the stable Hf isotopesdirectly to 181Ta. It is also shielded from r-process production by its stable isobar 180Hf.A number of possible production processes for180mTa have been pursued with no conclusiveresults. These include the p-process, neutrinonucleosynthesis, spallation reactions in the in-terstellar medium, photodisintegration reactions,and certain complex paths within the regulars-process [Ton00, Bel99]. The photodisintegra-tion reaction on 180mTa and neutron capture on179Ta leading to the 180mTa are experimentallyunverified aspects of possible s- and p-processexplanations for the origin and abundance ofthis isomer. They thus rely on model calcula-tions [Gok06]. The first steps towards measuringthe neutron-capture cross section on radioactive179Ta have been presented in Ref. [Sch99]. 179Tawas produced by the 180Hf(p,2n) reaction, radio-chemically separated, and used to measure the

thermal-neutron-capture cross section. Extend-ing the measurements to stellar (n,γ), however,remains an experimental challenge.

The proposal for measuring the 180mTa(γ,n)cross section was presented in the HIGS-PAC-2009 and was approved for 50 hours of beam time.In April 2010, around 60 hours of beam time wereused to carry out measurements at photon ener-gies of Eγ = 6.50, 6.70, 6.85, 7.00, 7.15, 7.30, and7.57 MeV. Natural Ta weighing 50.863 g was usedas the target for the 180mTa(γ,n) photodisinte-gration reaction. A 4π, 3He proportional countershown in Fig. 6.23, designed and fabricated atLANL, was used to count the emitted neutronsfrom the photo-neutron reaction. At each en-ergy, a Bi target with an approximately equalnumber of atoms was also measured at the neu-tron counter in order to estimate the backgroundfrom beam photons that Compton scatter off thetarget and are incident on the detector. Bi waschosen because its atomic number (83) is closeto that of Ta (73). Simultaneous to the neutronmeasurements, natural Ta and In foils of thick-ness 0.01 mm were activated at the exit end of thecollimator and counted in the low-backgroundcounting facility to measure photo-induced de-population of the 180mTa isomer. The incidentphoton flux was monitored using Compton scat-tering from an aluminum plate into a 123% HPGedetector positioned downstream at an angle of8.3◦ with respect to the beam.

The 3He proportional counter used for neu-tron counting was also found to be sensitive toγ-rays. From long overnight measurements ofthe average room background in the counter, wefound the rate to be ∼2.3 counts/s. A similar ratewas found in the blank-target measurements withthe beam. However, Ta is a high-Z element with


a large number of electrons acting as scatteringcenters for the incident photon beam. This con-tribution from atomic scattering of the incidentphotons was found to be huge, overshadowing theneutron counts from the photodisintegration re-action of interest.

Figure 6.23: (Color online) The 4π 3He counterused for detecting neutrons from the180mTa(γ,n) reaction.

In the next phase of the approved beam time,

we plan to introduce additional Pb shieldingaround the target, in order to eliminate or re-duce the γ-ray pileup. We have plans to positionthe 180Ta target further downstream, near theaxial end of the counter, so that the high-energyCompton-scattered photons from the target canescape the counter without triggering it.

The activation of the Ta and In foils, carriedout alongside the neutron measurements, havegiven positive results. Figure 6.24 shows a typi-cal spectrum from counting the activated Ta foilin the low background area using a 20% HPGedetector. The photo-induced depopulation of the180mTa isomer is inferred from the observation ofthe Hf Kα and Kβ X-rays that originate fromthe EC decay of the 180Ta ground state with ahalf-life of T1/2 = 8.1 h. The cross section forthe depopulation of 180mTa is expected to be de-rived from the observed Hf Kα peaks, with no in-terference from background. The primary prob-lem in deriving this depopulation cross section isthat the cross section for photo-activation of Inis poorly measured and cannot be reliably usedfor photon flux estimation. We are currently inthe process of estimating the flux from simula-tions using the mcnp code and comparing theseresults with the scattered spectra measured withthe 123% HPGe flux monitor.

Figure 6.24: (Color online) 180Hf X-rays used to identify the photo deexcitation of the 180mTa isomer.

[Bel99] D. Belic et al., Phys. Rev. Lett., 83,5242 (1999).

[Gok06] S. Goko et al., Phys. Rev. Lett., 96,192501 (2006).

[Sch99] M. Schumann and F. Kappeler, Phys.Rev. C, 60, 025802 (1999).

[Ton00] A. P. Tonchev and J. Harmon, Appl.Rad. Isotopes., 52, 873 (2000).


6.4.5 Dipole Strength Distribution in 232Th

A.S. Adekola, S.L. Hammond, C.R. Howell, H.J. Karwowski, J.H. Kelley, E. Kwan,TUNL; C.T. Angell, University of California at Berkeley, Berkeley, CA

Properties of low-energy dipole states in 232Th have been investigated with the nuclear-

resonance-fluorescence technique using mono-energetic γ-ray beams at energies of 2-4 MeV

from the High-Intensity γ-ray Source. Over 40 transitions corresponding to de-excitations to

the ground state and to the first excited state were observed for the first time. Excitation

energies, integrated cross sections, decay widths, branching ratios, and transition strengths

for those states in 232Th were determined and compared with quasi-particle random-phase-

approximation calculations.

There have been numerous experimental andtheoretical efforts to study the orbital magneticdipole response over a wide mass region [Hey10].However, most of the investigations of M1 excita-tions have focused on light nuclei or on deformedrare-earth nuclei [Hey10, End05], with limited ex-perimental information on the actinides. Evenless is known about E1 strength below the giantdipole resonance in the actinides. Until recently,the lack of polarized γ-ray beams made unam-biguous identification of E1 strength very diffi-cult.

This report describes nuclear-resonance-fluorescence (NRF) measurements on a 232Thtarget performed using nearly mono-energeticand 100% linearly-polarized γ-ray beams. Thepresent search for dipole excitations in 232Thwas performed at ten incident γ-ray energies be-tween 2 and 4 MeV, with an average energyspread of FWHM = 5%. No data were takenat beam energies between 2.4 and 2.7 MeV dueto the presence of a very strong background lineat Eγ = 2.614 MeV. The energy distributionsof the incident photons were measured using alarge volume HPGe detector positioned in thebeam. The photon flux was monitored by thesame detector positioned at an angle of 8.3◦ rel-ative to the beam axis. This detector measuredCompton scattering of the γ-ray beam from a1.1 mm thick copper plate positioned 167 cm up-stream and perpendicular to the beam axis (seeFig. 6.25). The absolute flux was determined us-ing the Klein-Nishina formula from the spectrumof the Compton-scattered photons corrected forthe detector response.

A schematic drawing of the experimentalsetup is shown in Fig. 6.25. The NRF γ rays from

the target were measured with a detector arraypositioned about 4.8 m downstream from the col-limator. The array consists of four HPGe detec-tors with 60% efficiency relative to a 7.62 cm ×7.62 cm NaI scintillation detector. These detec-tors were positioned 10 cm away from the centerof the target at 90◦ relative to the beam. Twoof these detectors were located in the horizontalplane and the other two in the vertical plane. Anadditional HPGe detector was placed in the hor-izontal plane at a backward angle of 140◦. Thisdetector configuration allows for the unambigu-ous determination of the multipolarities (E1, M1,and E2) of observed γ-ray transitions. The de-tectors had passive shielding and absorbers com-posed of Pb and Cu. This helped to reduce con-tributions from low-energy background.

A total of 20 states that decay to the groundstate as well as to the 2+ state were identified.The branching transitions to the 2+ state wereidentified from the 49-keV energy differences ofthe observed peaks. Five levels were observedwithout detectable branching to the 2+ state. Anupper limit has been set on the branching ratiofor each of these five levels. The level energiesdeduced for these states, the integrated cross sec-tions, the branching ratios, the decay widths tothe ground state, and the strengths have beendetermined for all of the transitions.

In Fig. 6.26, the strength distributions of theM1 and E1 excitations observed in this workare presented along with previous experimentaldata from Heil et al. [Hei88]. For the previ-ously known dipole excitations, which are below2.5 MeV, M1 strengths of 1.34(13), 0.79(9), and0.39(4) μ2

N were determined for the states ob-served at Ex = 2.043, 2.249 and 2.296 MeV, re-


O8.3

Cu plate

Detectors

Flux monitor

polarized γ−ray beam100% linearly−

#3

#1#5

Pb Collimator

Legend:Concrete wallPb brick wall

Target

Figure 6.25: Schematic diagram of the experimental layout showing two HPGe detectors at 90◦ andone at a backward angle of 140◦ in the horizontal plane. The other two HPGe detectors,located in the vertical plane, are not shown. The flux monitor at 0◦ (8.3◦) is indicatedas a dashed (solid) rectangle. The figure is not drawn to scale

spectively. These values agree reasonably wellwith the M1 strengths of 1.48(9), 0.56(7) and0.31(6) μ2

N , respectively, reported in Ref. [Hei88].All the states above 2.5 MeV are observed forthe first time. No E2 transitions were observedwithin the 2σ detection limit.

0

0.5

1

1.5

B(M

1) (

μ N

2 )

Present dataHeil et al.

2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4E

x (MeV)

0

0.4

0.8

1.2

1.6

B(E

1) (

10-3

e2 fm2 ) x1/3

Figure 6.26: Comparison of experimental and cal-culated B(M1) (upper panel) andB(E1) (lower panel) values in 232Th.The solid line in both panels repre-sents QRPA calculated values. Thecalculated E1 transition at 2.1 MeVhas been scaled down by a factor of3.

Theoretical calculations of low-lying electricand magnetic dipole states in 232Th withinthe quasi-particle random-phase-approximation(QRPA) approach [Kul10] are shown as solidlines in Fig. 6.26. An observed summed E1strength of 3.3(7)×10−3e2fm2, with a mean exci-tation energy of 3.7(8) MeV, was obtained for the

first time, and the observed summed M1 strengthof 4.3(6) μ2

N , with mean excitation energy of2.5(3) MeV, is in good agreement with the otheractinides and with the systematics of the scissorsmode in deformed rare-earth nuclei [Ade10].

The fragmentation of the M1 strengthsspreading from 2.0 to 3.3 MeV is consistent withthe general trend of QRPA calculations. The dif-ference between the data and the details of thecalculations is attributed most likely to the frag-mentation of the strength into transitions thatare below the present detection limit. It is alsopossible that these transitions do not exist.

The high quality beams from HIγS allowedthe observation of more than 20 discrete de-excitations to the ground state, and the branch-ing transitions to the first excited state were iden-tified for most of them. Twelve E1 and eleven M1excitations are observed in this work for the firsttime.

We thank E. B. Norman of UC Berkeley forloaning us the 232Th target, and D. R. Ticehurst,J. R. Tompkins, R. Raut, G. Rusev, W. Tornowand A. P. Tonchev for help in taking data andfor helpful discussions.

[Ade10] A. S. Adekola et al., Submitted, (2010).

[End05] J. Enders et al., Phys. Rev. C, 71,014306 (2005).

[Hei88] R. D. Heil et al., Nucl. Phys., A476, 39(1988).

[Hey10] K. Heyde et al., Rev. Mod. Phys.,(2010), (accepted).

[Kul10] A. A. Kuliev et al., Eur. Phys. J. A, 43,313 (2010).


6.4.6 Implementation of Nuclear Resonance Fluorescence Support in GEANT4

B. Ryan, UNC Chapel Hill

A new physics process implementation has been developed for the GEANT4 Monte-Carlo

simulation toolkit to support (γ,γ′) resonance reactions associated with individual isotopes.

The features of this extension are described.

geant4 is a widely used C++ toolkit for con-ducting Monte-Carlo simulations of the propaga-tion of particles through matter. While it offerselaborate support for most nuclear processes, itlacks the ability to simulate nuclear resonancefluorescence (NRF), the process by which a γ rayincident on a target isotope excites a dipole res-onance which quickly decays, releasing anotherγ ray with comparable energy. This γ-ray emis-sion is isotropic for an unpolarized incident beam,but polarizing the incoming γ rays creates a pro-nounced angular dependence in the plane of po-larization, particularly in even-even nuclei. Littleenergy is lost to recoil, due to the Mossbauer ef-fect in solids.

Physics processes in geant4 work by calcu-lating a table of mean-free-paths for incident par-ticle energies and materials at the beginning ofthe simulation. This table is used to determinethe probability of a physical process occurringwhen a particle passes through a particular ma-terial. Based on the probabilities of the varioususer-registered processes occurring, the geant4

code stochastically determines whether a phys-ical event will occur, and if so, which one. Ifa process is selected to occur, the geant4 codecalls a function unique to that process to carryout the interaction solution.

The NRF removes deletes the incident γ rayand generates a new photon with identical energyand random momentum direction. The nucleusis not considered as a recoil product because, forthe NRF regime with which we are concerned(elastic scattering), this recoil is negligible.

NRF spectral lines typically have widths oftens of meV. Since these widths are small rel-ative to the energies at which they occur, thestock class with which geant4 creates the mean

free path table cannot resolve the resonance lineswithout resorting to a multidimensional array ofa size hopelessly impractical for typical hardware.The geant4 code was therefore modified to cre-ate the mean free path table for the NRF processby only binning the values corresponding to theenergies around the registered resonance lines in-stead of using the same binning resolution forthe entire range of energies. Mean free paths arebinned as nearly rectangular pulses which risefrom zero to full value in roughly 1/1000th thewidth of the resonance line.

Resonance lines are registered individually bycalling a function that inputs the resonance linewidth, its cross section, its energy, and the massnumber and atomic number of the correspondingisotope. These are stored in private data vectorsof the modified cross-section-handler class.

While this relatively simple model is, whenbased on experimental data, a fairly accurate re-construction of NRF, it does not take into ac-count all possibilities. When a nucleus is excitedto an energy state above its ground level, it is pos-sible that there will be more than one decay pathfor re-emission of the absorbed energy. The pos-sibility of decaying to multiple states can allowfor the reliable identification of certain isotopessuch as 238U, which generates two NRF peaksspaced about 50 keV apart originating from thesame excited state. Given the variety of possibledecay paths, it would be very difficult to developa general process to account for all of them; itwould be more practical for individual users tomodify the code to model their particular situ-ations. Additionally, the model currently doesnot account for the angular dependence of pho-ton emission. This feature will be included in thenear future.


6.4.7 Nuclear Self-Absorption Experiment on 11B

S. Pratt, G. Rusev, J.H. Kelley, E. Kwan, R. Raut, A.P. Tonchev TUNL; R. Schwengner,Research Center Dresden-Rossendorf, Germany

The first nuclear self-absorption experiment at HIγS is reported. Measurements were carried

out at beam energies of 7.3 and 8.9 MeV with the goal of reducing the uncertainty of the

widths of the levels at 7.285 and 8.920 MeV in 11B.

Most of the previous photon-scattering exper-iments relevant to nuclear astrophysics and nu-clear structure studies at energies close to theneutron-separation energy were carried out withbremsstrahlung beams. 11B was used as a cal-ibration standard in those experiments, provid-ing a few strong levels up to 9 MeV. There-fore, the accuracy of the results from the photon-scattering experiments strongly depend on theprecision of the level widths in 11B. The widthsof the levels at 7.285 and 8.920 MeV are knownwith a relative uncertainty of not better than7% and 5%, respectively. We report an effortto reduce these uncertainties through the firstexperiment based on the nuclear self-absorptionmethod [Met59] at HIγS. This activity is a con-tinuation of our previous investigation of the mul-tipole mixing ratios of the ground-state transi-tions in levels of 11B [Rus09].

The experiment was performed at the up-stream target room (UTR) with circularly po-larized photon beams with energies of 7.3 and8.9 MeV. The energy distribution of the beammeasured with a large HPGe detector placed inthe beam is shown in Fig. 6.27 (a). For eachbeam energy, we carried out four measurements:three with 11B absorbers positioned behind thecollimator and one measurement without any ab-sorber. The thicknesses of the absorbers were2.495(5), 4.987(7), and 8.802(9) g/cm2. Two de-tector setups consisting of four 60% relative ef-ficiency and two 100% relative efficiency HPGedetectors were used to measure the attenuationof the photon beam at the energy of the lev-els in 11B by means of photon scattering fromthick 11B4C targets. In total the measurementslasted between 1.5 and 3 hours to collect about104 counts from the sum spectrum of all six de-tectors. An example of the spectrum measuredat 8.9 MeV is shown in Fig. 6.27 (b). We also car-ried out a measurement with the thickest boronabsorber but without the 11B4C targets in orderto test the Pb shielding between the collimator

room and the UTR. The collected spectrum ispresented in Fig. 6.27 (c), which clearly showsan absence of a peak at 8.920 MeV. The analysisof the data collected is currently in progress.

8.4 8.6 8.8 9.0

50

100

150

200)-1

keV

-1 (

sγ

Φ-3

10 0

(a)

8.4 8.6 8.8 9.0

200

400

600

800

Co

un

ts /

1 ke

V

0

(b)

8.2 8.4 8.6 8.8 9.0 9.20

20

40

60

80

Co

un

ts /

10 k

eV

(MeV)γE

(c)

Figure 6.27: Spectra from the measurement at8.92 MeV. (a) Spectra of the incidentbeam showing energy distribution ofthe photon flux, Φγ , (b) γ rays scat-tered off of 11B4C targets (the single-escape peak is visible at 8.409 keV),(c) measurement with 11B absorbers,but without any targets, revealinghow effective the Pb shielding was.

[Met59] F. R. Metzger, Prog. Nucl. Phys., 7, 54(1959).

[Rus09] G. Rusev et al., Phys. Rev. C, 79,047601 (2009).

Applications of Nuclear Physicsand Nuclear Data Evaluation

Chapter 7

• Homeland and National Nuclear Security

• Pyroelectric Research

• Nuclear Data Evaluation

140 Applied Research and Nuclear Data Evaluation TUNL XLIX 2009–10

7.1 Homeland and National Nuclear Security

7.1.1 Photoneutron Yield Ratios of (�γ,n) Reactions in the GDR Region

J.R. Tompkins, M.W. Ahmed, N. Brown, B. Davis, S.S. Henshaw, H.J. Karwowski, J.M.

Mueller, B.A. Perdue, S. Stave, H.R. Weller, TUNL; D.M. Markoff, North CarolinaCentral University, Durham, NC ; G. Feldman, C. Smith, George Washington University, Wash-ington, DC ; L. Myers, University of Illinois at Urbana-Champaign, Urbana, IL; J.S. Hauver, W.

Henderson, C.S. Whisnant, James Madison University, Harrisonburg, VA

Photoneutron angular distributions from (�γ,n) reactions measured at HIγS are powerful ob-

servables that yield insight into the nature of this reaction. Studies of eight different materials

have been conducted in two measurement efforts during the past year, and anisotropies have

been observed with respect to rotations about the beam axis. Preliminary analyses of the

actinide data reveal that the largest ratios are produced by neutrons with energies near the

maximum possible from the (γ,n) reaction.

The detection of photoneutrons emitted fol-lowing the absorption of 100% linearly-polarizedγ rays with energies relevant to the GDR is ameans to study the structure of nuclei. The angu-lar distribution of these neutrons has been stud-ied in the past using unpolarized γ-ray beams[Bak61, Hus67, Nai77]. The present work ex-tends previous studies at TUNL that used linearpolarization [Sta09]. Polarization effects in theangular distribution can be studied by measuringthe yields of the neutrons emitted perpendicular,Iperp, and parallel, Ipar, to the plane of polar-ization, which is parallel to the floor. The ratioIpar/Iperp as a function of the emitted neutronenergy En may provide insight into the physicsof the (�γ,n) reaction.

Two experiments have been conducted dur-ing the past year to study the following mate-rials: Pb, 235U, 238U, Cd, Sn, Ta, 239Pu, and232Th. The neutrons were detected using the de-tector array for photoneutrons (see Sect. 9.2.1)with the 1.0 m frame in the first data collectioneffort and the 0.5 m frame in the second. Data foreach of these have been acquired at γ-ray ener-gies between 11.0 and 15.5 MeV with γ-ray fluxeson target between 106 and 107 γ rays per sec-ond. This is the first systematically measuredset of neutron angular distributions using 100%linearly-polarized γ-ray beams.

The measured neutron energy spectra arecontinuous from the detector threshold, En = 2MeV, to ∼10 MeV. They were rebinned into 0.5MeV wide bins, which were used to form the ra-

tio previously mentioned. Differences in the solidangle coverage and responses of the detectorscan lead to systematic errors in the formed ra-tios. Corrections for these were made using dataobtained with circularly-polarized γ-ray beams,since these must produce ratios of 1.0.

(MeV)nE0 2 4 6 8 10 12

per

p/I

par

I

0

0.5

1

1.5

2

2.5

3

3.5

4

239Pu

232Th

238U

235U

Figure 7.1: (Color online) A comparison of theratios from the actinide targets as afunction of En for Eγ = 15.5 MeV atθ = 90◦. (Note that 232Th correspondsto Eγ = 15.0 MeV.)

A comparison of the ratio as a function ofEn at Eγ = 15.5 MeV indicates that for 238U,235U, and 239Pu, the ratio decreases as fissilityincreases (see Fig. 7.1). The fissility parameterZ2/A for these three nuclides has values of 35.6(238U), 36.0 (235U), and 37.0 (239Pu) This ob-servation cannot be extended to 232Th (fissilityparameter of 34.9) because it corresponds to a

TUNL XLIX 2009–10 Applied Research and Nuclear Data Evaluation 141

(MeV)nE2 4 6 8 10 12

per

p/I

par

I

0

0.5

1

1.5

2

2.5

3

= 11.0 MeVγE

= 13.0 MeVγE

= 14.0 MeVγE

= 15.5 MeVγE

Figure 7.2: (Color online) The 235U ratio as a function of En for various Eγ. The data are truncatedat ∼0.3 MeV above the maximum neutron energy produced by a (γ,n) reaction at thecorresponding Eγ

different γ-ray energy. Despite differences in themagnitude of the ratio, the general trends as afunction of En are similar. The ratio spectrumfor each of these materials has one large peak pro-duced by the higher energy neutrons, and a dou-ble hump is present in this peak for 232Th and235U. The neutrons corresponding to the maxi-mum ratio are consistently ∼1-2 MeV below themaximum energy possible from the (�γ,n) reac-tion (see Fig. 7.2). The linearity of the relation-ship between the energy of these neutrons andEγ indicates that they are the result of decays toexcited states in the daughter nucleus during a(�γ,n) reaction rather than during processes suchas (γ,2n) or (γ,f). However, it is unclear whythe location of the peak in the ratio at each Eγ

is consistently below the maximally allowed neu-tron energy.

One further observation is that the neutronswith Eγ ≈ 2 MeV are uniformly emitted into theparallel and perpendicular detectors at a givenθ for all of the targets. For the actinides, theratio decreases at the lower neutron energies, ap-proaching unity below 3.0 MeV. This observationalso holds for Pb.

[Bak61] R. G. Baker and K. G. McNeill, Can. J.Phys., 39, 1158 (1961).

[Hus67] S. M. Hussain and K. G. McNeill, Can.J. Phys., 45, 2851 (1967).

[Nai77] S. Nair et al., J. Phys. G., 3, 965 (1977).



7.1.2 Photofission Neutron Yield Ratios on 238U near Eγ = 6.2 MeV usingLinearly Polarized γ rays

S. Stave, M.W. Ahmed, N.J. Brown, S.S. Henshaw, H.J. Karwowski, J.M. Mueller, B.A.

Perdue, J.R. Tompkins H.R. Weller TUNL; M.S. Johnson Lawrence Livermore NationalLaboratory, Livermore, CA

Neutron yields and the ratios of the yields measured parallel and perpendicular to the plane ofγ-ray polarization (Ipar/Iperp) have been measured for the first time for a 238U target near the(γ,n) threshold of Eγ � 6.2 MeV. Measurements were performed at γ-ray energies of 5.7 MeV(near the photofission threshold), through the (γ,n) threshold, and continuing up to 6.5 MeV.The Ipar/Iperp data taken with the nearly 100% linearly polarized beams at HIγS have valuesof 3 to 4 in the pure fission region below the (γ,n) threshold, decreasing to about 2 at energiesjust above the (γ,n) threshold. In an effort to understand these new data, a model has beendeveloped in which the neutrons are emitted isotropically in the center-of-mass frame of thefission fragments. The fission-fragment angular distributions are taken from previous γ-ray-and neutron-induced fission data and are used to predict the values of Ipar/Iperp for both thefission fragments and the neutrons.

Linearly polarized γ rays with energies at andabove the fission threshold can be a powerful toolfor the interrogation of materials. In addition totheir ability to penetrate shielding, they also in-duce the emission of several-MeV neutrons. Theratio of neutron yields parallel and perpendicularto the plane of γ-ray polarization as a function ofoutgoing neutron energy (Ipar/Iperp) can providea unique signature of isotopes. Reference [Sta08]and Sect. 7.1.1 provide details about studies withγ rays between 10 and 15.5 MeV on various ac-tinide targets and shielding/background materi-als.

Neutron yield ratios Ipar/Iperp have beenmeasured using a 238U target near the (γ, n)threshold of Eγ 6.2 MeV for the first time. Atthese energies, the neutrons from the (γ, n) reac-tion either do not exist or have energies below thedetection threshold for the neutron detectors thatwere used during the experiment. (See Sect. 9.2.1for more details on the detectors and the neutrondetector array). Figure 7.3 shows the total crosssection for 238U(γ, f) as a function of incidentγ-ray energy measured in March 2010. The re-sults using γ-ray beam energies from 5.7 (near thephotofission threshold) through the (γ, n) thresh-old up to 6.5 MeV are compared with cross-section data from Ref. [Dic75] that was convo-luted with the measured energy spread of the γ-ray beam. Twelve neutron detectors filled withBC-501A liquid scintillator were placed both inand out of the plane of γ-ray polarization at az-imuthal angles of φ = 0◦ (in the plane of thepolarization vector), 90◦, 180◦ and 270◦ and at

polar angles of θ = 75◦, 90◦ and 125◦. The rangeof θ angles allowed for a measurement of the un-polarized angular distribution. The differentialcross sections at each angle were determined byfirst averaging over the φ angles and then scal-ing by the appropriate efficiency corrections aswell as solid angle, target thickness, and beam in-tensity. The differential cross sections were thenfit with Legendre polynomials to give the angle-integrated yield at each energy.

[MeV]γE5.6 5.8 6 6.2 6.4 6.6 6.8 7

[m

b]

σ

02468

1012

S/TUNLγHIDickey/Axel

[MeV]γE5.6 5.8 6 6.2 6.4 6.6 6.8 7

[m

b]

σ

02468

1012

Figure 7.3: (Color online) The 238U(γ, f) cross sec-tion from Ref. [Dic75] convolutedwith the γ-ray beam energy spreadcompared with the measured crosssection determined at HIγS.

A typical neutron energy spectrum is shownin Fig. 7.4. The neutrons in Fig. 7.4 were de-tected at θ = 90◦, φ = 0◦ for Eγ = 6.2 MeV.The neutrons were detected at a rate of approx-imately 600 in 10 minutes for 10 grams of 238Uwith a γ-ray flux of 107 γ/s. This type of detectedneutron rate is promising for possible use as anindicator of the presence of fissionable materials.At Eγ = 6.2 MeV on 238U, neutrons can only beproduced by fission. At, or slightly above this en-


ergy, neutrons with energies above 1.5 MeV couldbe used to signal the presence of 238U and otherfissionable materials including 232Th, 233U,235U,237Np, 239Pu, and 241Am.

[MeV]nE2 3 4 5 6 7

Co

un

ts

020406080

100120140160180

Figure 7.4: (Color online) Neutron counts for238U(γ, f) versus neutron energy atEγ = 6.2 MeV, θ = 90◦, φ = 0◦.

As shown in Fig. 7.5, the Ipar/Iperp data takenwith the nearly 100% linearly polarized beams atHIγS have values ranging from 3 to 4 in the purefission region below the (γ, n) threshold to about2 at energies just above the (γ, n) threshold. Fig-ure 7.5 shows this variation as a function of inci-dent γ-ray energy for neutrons with energies be-tween 2 and 5 MeV. This new physics discoverycould also be an indicator of the presence of fis-sionable materials especially if the observed ratiois distinct for different isotopes. Further stud-ies near the fission and (γ, n) thresholds on otheractinides are planned.

[MeV]γE5.8 6.0 6.2 6.4

per

p/I

par

I

0

2

4

024681012

,f) [mb

]γ(

σ

0.5 MeV±=2.0nE

TUNLDickey/Axel

[MeV]γE5.8 6.0 6.2 6.4

per

p/I

par

I

0

2

4

024681012

,f) [mb

]γ(

σ

0.5 MeV±=3.0nE

[MeV]γE5.8 6.0 6.2 6.4

per

p/I

par

I

0

2

4

024681012

,f) [mb

]γ(

σ

0.5 MeV±=4.0nE

[MeV]γE5.8 6.0 6.2 6.4

per

p/I

par

I

0

2

4

024681012

,f) [mb

]γ(

σ

0.5 MeV±=5.0nE

Figure 7.5: (Color online) Ipar/Iperp at θ = 90◦ as a function of γ-ray energy for 1 MeV wide neutronenergy slices. The cross section from Ref. [Dic75] is plotted for reference to show theenergy region at which the (γ, n) channel opens.

In an effort to understand these new fissiondata, a model has been developed where the neu-trons are emitted isotropically in the center-of-mass frame of the fission fragments. The fissionfragment angular distributions are taken fromprevious γ-ray induced fission data [Rud88] and,when combined with the angular momentum for-malism from Ref. [Rat82], are used to predictthe values of Ipar/Iperp for both the fission frag-ments and the neutrons. The neutrons are evap-orated isotropically off of the fission fragmentsin the fission fragment center-of-mass frame andboosted back into the laboratory frame. The neu-tron evaporation spectra are adjusted to give thesame neutron energy distributions as those mea-sured using an “unpolarized” beam. The angu-lar momentum formalism from Ref. [Rat82] isthen employed to properly handle the linearlypolarized γ-ray interaction. This model is still inits early stages, but it has qualitative agreementwith the data. Further work is being performedto refine the model and improve the agreement

with the data.To conclude, the 238U(γ, f) cross section near

threshold has been measured at HIγS and foundto agree well with previous data. The ratioIpar/Iperp was measured for the first time on 238Uand found to deviate significantly from unity forneutrons with energies between 2 and 5 MeV.This finding combined with the presence of neu-trons above 1.5 MeV may provide a way of iden-tifying fissionable materials. The development ofa model of the new findings is underway.

[Dic75] P. Dickey and P. Axel, Phys. Rev. Lett.,35, 501 (1975).

[Rat82] R. Ratzek et al., Z. Phys. A, 308, 63(1982).

[Rud88] V. Rudnikov et al., Sov. J. Nucl. Phys.,48, 414 (1988).



7.1.3 Measurement of γ-Ray Induced Activation of 235U

J.M. Mueller, M.W. Ahmed, N. Brown, S.S. Henshaw, H.J. Karwowski, B.A. Perdue, S.

Stave, J.R. Tompkins, H.R. Weller, Y. Wu, TUNL; B. Davis, D. Markoff, North CarolinaCentral University, Durham, NC ; G. Feldman George Washington University, Washington, DC ;L. Myers, University of Illinois Urbana-Champaign, IL

A photofission experiment on 235U was performed at HIγS using 11, 13, and 15 MeV linearly

polarized γ-ray beams. After exposure to the beam for approximately 12 hours, the 235U

target was placed in front of a HPGe detector, and data were accumulated for a period of

about 10 hours. Fission fragments and subsequent decay-chain isotopes were identified using

characteristic γ-rays and their lifetimes. The fragment mass distributions for photofission

and thermal-neutron-induced fission were compared. Potential applications include active

interrogation of cargo containers.

Photofission experiments at HIγS are used tounderstand the dynamics of the photofission re-action mechanism. A comparison of the fission-fragment mass distributions for photofission andneutron-induced fission may give insight into thephotofission mechanism. For example, thermal-neutron-induced-fission fragments of 235U have adouble humped mass distribution, and relativeyields for a fission fragment of mass number Aand atomic number Z are known [Eng92]. How-ever no such measurements exist for photofissionon this nucleus.

E [keV]1000 1500 2000 2500 3000

Co

un

ts

0

2000

4000

6000

8000

10000

Figure 7.6: (Color online) Comparison of thepre-(black, lower) and post-activation(red, upper) spectra of the 235U tar-get.

In addition to improving the understandingof the photofission process, there is an interestin potential applications of this work to nationalsecurity. Identifying characteristic γ-rays emit-ted only upon photofission of 235U may lead to

a viable method for active interrogation of cargocontainers. Interrogating cargo containers in do-mestic ports would lower the likelihood of thesuccessful smuggling of nuclear materials.

To accomplish these goals, a photofission ex-periment on a 235U target was performed atHIγS. Prior to exposure to the beam, a γ-rayspectrum of the target was obtained using aHPGe detector. The target was then exposedto the beam for approximately 12 hours at γ-rayenergies of 11, 13, and 15 MeV. The γ-ray in-tensities were 107γ/s on target. After exposure,the target was placed in front of the same HPGedetector, and data were taken overnight for ap-proximately 10 hours. This process was repeatedfor two consecutive days.

Time [s]0 5000 10000 15000 20000 25000 30000

En

erg

y [k

eV]

1360

1380

1400

1420

1440

1460

1480

Co

un

ts

1

10

Sr92

Cs138

K40

Figure 7.7: (Color online) Energy of the γ raysversus time since exposure to thebeam for Eγ between 1350 and 1480keV. Lines from 92Sr, 138Cs, and 40Kare indicated.


Table 7.1: Identified Photofission Fragments

Isotope Characteristic Relative RelativeIdentified γ-rays (keV) NDS T1/2 Observed T1/2 Yield γ-ray Yield

88Kr 2392.1 2.84 ± .03 h 2.86 ± .09 h 9.1 3.1834.8 1.21529.8 1.1

92Sr 1383.9 2.71 ± .04 h 2.85 ± .04 h 10.0 9.0133I 529.9 20.8 ± .1 h 31.5 27.4134I 847.0 52.5 ± .2 m .15 .14

884.1 .10135I 1260.4 6.57 ± .03 h 21.5 6.2

1131.5 4.9138Cs 1435.6 33.4 ± .2 m .60 .46

2639.6 .05142La 641.3 1.52 ± .01 h 1.59 ± .13 h 2.6 1.2

2397.8 .352542.7 .26

The pre-exposure spectrum measured γ-raysfrom the decays of various isotopes in the target,as well as background contributions, as shownin Fig. 7.6. Subsequent photofission of the tar-get generated unstable fission fragments, some ofwhich emitted γ-rays when they decayed. Theadditional γ-rays in our post-exposure spectrumare attributed to decays of these photofissionfragments. The energies of these additional γ-rays can be used to identify specific isotopes thatwere generated during the photofission process.

A70 80 90 100 110 120 130 140 150 160

Z

30

35

40

45

50

55

60

65

Kr88Sr92

I135I, 134I, 133Cs138

La142

Figure 7.8: (Color online) Mass distribution ofthe photofission fragments (red ordark squares) and thermal neutroninduced fission fragments (squares inshades of gray).

During the activation measurement, the en-ergy of the detected γ ray and the time sinceexposure were recorded. The energy distributionof the detected γ rays are shown as a function

of time in Fig. 7.7. The half-lives of various fis-sion fragments were measured by fitting the γ-rayintensities for a given line in the spectrum as afunction of time with the function c0 + c1 ·e−t/c2 ,where t is the elapsed time and the c quantitiesare fitting constants, with c2 yielding the half-life. The γ-ray energy and half-life measurementswere used to uniquely identify several photofis-sion fragments.

A comparison between photofission-fragmentmass distributions and thermal-neutron-induced-fission-fragment mass distributions is shown inFig. 7.8. The photofission fragments coincidewith the familiar double-humped mass struc-ture of the thermal-neutron-induced-fission frag-ments. This implies similarities in the reactionmechanisms of photofission and neutron-inducedfission.

A summary of the results is shown in Ta-ble 7.1. The observed T1/2 (or half-life) errorsare from fitting only. Due to its high yieldand low background contributions, the γ-ray at1383.9 keV, emitted in the decay of 92Sr, is themost identifiable peak which is directly linkedto photofission of 235U. Future work will investi-gate the feasibility of interrogating cargo contain-ers for 235U by exposing them to a γ-ray beamand subsequently identifying the 1383.9 keV 92Srpeak.

[Eng92] T. England and B. Rider, Evalua-tion and Compilation of Fission Yields,1992, ENDF-349, LA-UR-94-3106, Ex-tracted from ENDF.


7.1.4 Linearly Polarized Beam Induced Photo-Neutron Yield Ratios in9Be between 10.5 and 15.5 MeV

D.M. Markoff, B. Davis North Carolina Central University, Durham, NC ; M.W. Ahmed, H.J.

Karwowski, J.M. Meuller, S. Stave, J. Tompkins, H.R. Weller TUNL;

The ratio of photo-neutron yield parallel to the yield perpendicular (with respect to the plane

of polarization of the beam) from 9Be has been measured using nearly 100% linearly polarized

photons for 5 γ-ray energies at the HIγS facility. The energies were chosen to study the

asymmetries both in resonance regions (11.3, 11.8 and 13.8 MeV) and far from resonances

(10.5 and 15.5 MeV) in 9Be. Preliminary analysis of the data indicates the presence of large

photo-neutron yield ratios.

In July 2010, we measured photo-neutronyields from 9Be as part of the program to measurethe ratio of neutron yields parallel and perpen-dicular to the plane of polarization as a functionof outgoing neutron energy for actinides, metalsand other materials of interest for cargo interro-gation. A 2.54 cm thick 9Be cylindrical target ofdiameter 1.9 cm was placed axially in the beam.Twelve detectors were located nominally 58 cmfrom the target with 4 detectors each at anglesof 55◦, 90◦ and 125◦. These detectors were lo-cated “up” and “down” or perpendicular to thebeam polarization plane and “right” and “left” orparallel to the beam polarization plane. An ad-ditional 3 pairs of detectors were located 58 cmfrom the target in plane and out of plane at an-gles of 72◦, 107◦ and 142◦. The apparatus used inthese measurements is described in Section 9.2.1.For this initial report, we will discuss the datafrom the 4 detectors at 90◦, labeled “Top”, “Bot-tom”, “Left” and “Right” in the figures below.

The photo-neutron yields from 9Be were mea-sured using both a circularly (Fig. 7.9) and lin-early (Fig. 7.10) polarized beam at 15.5 MeV.The plots show the left, right, top and bottomdetector yields at 90◦. The spectra show struc-ture in the regions of the energies associated withdecays through the ground, first and part of thesecond excited states in 8Be with respective en-ergies of 1.665 MeV, 4.565 MeV, and 13.07 MeVrelative to the ground state in 9Be [Coc68]. Theenergies of the expected neutrons emitted at 90◦

for the decay to the ground state (n0), the firstexcited state (n1) and the second excited state(n2) are indicated on the figures.

[MeV]nE2 4 6 8 10 12 14 16

Co

un

ts

020406080

100120140160180200 Top

Bottom 0n1n2n

[MeV]nE2 4 6 8 10 12 14 16

Co

un

ts

020406080

100120140160180200 Left

Right 0n1n2n

Figure 7.9: [Color online] Photo-neutron yields at

90◦ from 9Be for 15.5 MeV circularlypolarized γ rays.

[MeV]nE2 4 6 8 10 12 14 16

Co

un

ts

0

50

100

150

200

250

300

350 TopBottom

0n1n2n

[MeV]nE2 4 6 8 10 12 14 16

Co

un

ts

0

50

100

150

200

250

300

350 LeftRight

0n1n2n

Figure 7.10: [Color online] Photo-neutron yields

at 90◦ from 9Be for 15.5 MeV linearlypolarized γ rays.


As expected, the yield spectra are nearly thesame in each detector for the circularly polar-ized beam runs. For the case of linearly polar-ized γ rays at 15.5 MeV, it is clear in the spectraof Fig. 7.10 that there are more counts in-planethan out-of-plane.

The greater yield from in-plane detectorscompared to out-of-plane for the linearly polar-ized γ-ray beam is consistent with E1 photo-absorption. However, the resonance levels in9Be include [Til04] 11.28 MeV, 11.81 MeV, and13.79 MeV each with an assigned negative parity.Photo-absorption excitation of these resonanceswould be expected to proceed via M1 and/orE2 followed by the emission of neutrons havingodd l-values. The data suggests that the photo-neutron reaction proceeds through some combi-nation of direct capture, which is predominantlyE1, and resonance reactions which are predomi-nantly M1 and E2.

We measured the ratio of yields parallel toperpendicular ( Ipar

Iperp) for an additional 4 incident

energies both near and far from the resonancestates in 9Be. Figure 7.11 shows the yield ra-tio in the 90◦ detectors for the 5 incident γ-rayenergies. Note that the ratios are very large incomparison with those which we have measuredin other nuclei. (See 7.1.5, and 7.1.1.) A directcapture plus resonance model is being developed

in an attempt to interpret these results.

[MeV] nE0 2 4 6 8 10 12 14

perp

/Ipa

rI

0

2

4

6

8

10

1210.5 MeV

11.3 MeV

11.8 MeV

13.8 MeV

15.5 MeV

Figure 7.11: (Color online) Photo-neutron yield

ratios from 9Be for linearly polarizedγ rays at 5 energies and lab angle of90◦.

[Coc68] C. Cocke and P. Christensen, Nuc. Phys.A 111 623, (1968).

[Til04] D. Tilley et al., Nuc. Phys. A 745 155,(2004).


7.1.5 Polarized (γ,n) reaction studies of natCd, natSn, and 181Ta

W.C. Smith, G. Feldman, GWU ; M.W. Ahmed, S. Stave, S.S. Henshaw, J.R. Tompkins,J. Mueller, H.R. Weller, H.J. Karwowski, TUNL; S. Whisnant, W.R. Henderson, J.S.

Hauver, JMU ; D.M. Markoff, B. Davis, NCCU

HIγS was used to measure asymmetries in neutron emission from linearly polarized γ rays

incident on natCd, natSn, and 181Ta. Neutrons were detected both parallel and perpendicular

to the plane of polarization at six photon energies between 11.0 and 15.5 MeV. The curves

formed by plotting ratios of neutron yields parallel to yields perpendicular as a function of

neutron energy were found to be distinct from those of previously studied targets at HIγS.

Asymmetrically produced neutrons from lin-early polarized γ rays incident on fissile andnonfissile nuclei make it possible to form a ra-tio of counts parallel to counts perpendicular(Ipar/Iperp) to the plane of γ-ray polarization.By plotting this observable as a function of out-going neutron energy, a “signature” curve can beconstructed for each target isotope. These iso-topic signatures can be used in many ways, in-cluding as a means of photon cargo interrogation.

Intense, nearly 100% linearly polarized γ raysbetween 11.0 and 15.5 MeV were produced byHIγS, passed through a 1” collimator, and im-pinged upon the targets. Neutrons producedby processes including (γ, n), (γ, F ), and (γ, 2n)were detected by an array of 18 liquid-scintillator(BC-501A) detectors. These detectors weremounted above, below, to the right, and to theleft of the beam line at angles in the range θlab

= 55◦ - 142◦ (see Sect. 9.2.1). Plates of thick-ness 0.25”, 0.2”, and 0.188” of natCd, natSn, and181Ta, respectively, were positioned in the centerof the array, 0.5 m away from each detector. Tolimit the effects of multiple scattering, the tar-get plates were oriented at θ = 45◦, φ = 45◦,thus minimizing the amount of target materialthe neutrons pass through while exiting the tar-get.

The results for natCd, natSn, and 181Ta wereadded to a preexisting catalog of isotopic sig-natures determined at HIγS since March 2009.Amongst the previously studied targets are thefissile actinides 235U and 238U, as well as the non-fissile 209Bi, natPb, and natFe [Sta09]. Due to thenonfissile nature of natCd, natSn, and 181Ta aswell as their (γ, 2n) thresholds being 16.4, 15.8,and 14.3 MeV, respectively, the measured asym-metries result unambiguously from the (γ, n) re-

action. The targets were chosen due to their com-mercial accessibility, their relatively low (γ, n)thresholds (9.4, 6.5, 7.6 MeV for natCd, natSn,and 181Ta, respectively), and their nonunitaryexpected Ipar/Iperp values. Additionally, 181Tais isotopically pure, meaning physical conclusionscan more easily be drawn from experimental re-sults.

In July 2010, data were collected for natCd,natSn, and 181Ta as summarized in Table 7.1.5.

Table 7.2: Incident γ-ray energies used.

Target 11.0 12.0 13.0 14.0 15.0 15.5natCd - - X X X XnatSn - - X - X X181Ta X X X X - X

natCd was not measured at 11.0 and 12.0 MeVbecause its (γ, n) threshold combined with the∼2 MeV hardware cut do not allow for many neu-trons to be seen at these energies.

Outgoing neutrons were distinguished fromCompton-scattered photons by two-dimensionalcuts on the time-of-flight (TOF) vs. pulse-shapediscrimination plots. These highly nonrelativisticneutrons then had their energies (En) determinedusing TOF techniques. Additional data for eachtarget at 15.5 MeV were taken using circularlypolarized γ rays. These data were then used tocorrect for differences in the neutron detection ef-ficiency in each detector. The same neutron de-tection efficiency corrections were applied to theEn spectra from the linearly polarized runs.


[MeV]nE2 3 4 5 6 7 8

Co

un

ts

2000

4000

6000

8000

10000

12000

14000

16000 LeftTopRightBottom

Figure 7.12: (Color online) Energy spectra fornatCd with 15.5 MeV circularly po-larized γ rays.

[MeV]nE2 3 4 5 6 7 8

Co

un

ts

2000

4000

6000

8000

10000

12000LeftTopRightBottom

Figure 7.13: (Color online) Energy spectra fornatCd with 15.5 MeV linearly polar-ized γ rays.

For each detector angle, the En spectra wereplotted and compared using circular polarization(Fig. 7.1.5) as well as using linear polarization(Fig. 7.1.5). It is clear that the neutron energydistribution is the same for each detector whenusing a circularly polarized beam. However, witha linearly polarized beam the left and right detec-tors received noticeably greater neutron countsthan the top and bottom detectors.

[MeV] nE0 2 4 6 8 10

per

p/I

par

I

0

1

2

3

4

5

6

7

Cd 15.5 MeV (Lin)

Sn 15.5 MeV (Lin)

Ta 15.5 MeV (Lin)

Figure 7.14: (Color online) Ratios plotted as afunction of neutron energy for natCd,natSn, and 181Ta.

The disagreement between the two planes of de-tectors warrants the calculation of Ipar/Iperp val-ues across the outgoing neutron energy range,which, when plotted, form isotopic signaturecurves (Fig. 7.1.5). Interestingly enough, de-spite similar values of Z for natCd and natSn,their curves seem to be distinct. Also, the ratiofor 181Ta peaks at 2.4 around 4.5 MeV, a lowerenergy than previously studied targets. Thesedata do not take into account instrumental asym-metries, but the adjustment is not expected tochange the results by more than ∼20%.

As mentioned in previous reports [Sta09], anexpression for the ratio Ipar/Iperp was derivedfrom Satchler’s work on (γ, n) angular distribu-tions [Sat55]:

Ipar

Iperp=

1 − 2a2

1 + a2. (7.1)

Here, a2 is a dimensionless coefficient from theneutron angular distribution assuming pure elec-tric dipole absorption - a reasonable assumptionfor gamma ray energies between 10 and 20 MeV.The measured Ipar/Iperp ratios are compared tothe calculated ratios from previously measureda2 values [Bak61] in Table 7.1.5.

Table 7.3: Data collected at HIγS using linear po-larization at 15.5 MeV compared toprevious predictions.

Target Measured Ratio Predicted RationatCd 2.15±0.02 2.1±0.1natSn 2.05±0.02181Ta 1.86±0.02 1.9±0.2

The measured ratios at HIγS agree very well withthe ones calculated from a2 measurements de-spite the experimental differences (see [Sta09]).

Signature curves plotting Ipar/Iperp as a func-tion of En were produced from linearly polarizedphoton-induced neutron emission data for natCd,natSn, and 181Ta. Although only a preliminaryanalysis has been performed, the curves for thethree targets and previously studied targets atHIγS appear to be substantially distinct.

[Bak61] R. G. Baker and K. G. McNeill, Can. J.Phys., 39, 1158 (1961).

[Sat55] G. Satchler, Proc. Phys. Soc. A, 68,1041 (1955).



7.1.6 Simulation of Background from Neutron Detector Array and from(γ, n) on 14N in the Target Room

J.A. Walker, TUNL

Neutron backgrounds in multi-detector array from multiple scattering in neighboring detec-

tors and from the beam passing through air were simulated using geant4. Realistic light

output curves and detector construction were used to properly model the detector response.

The effect of multiple scattering between two detectors positioned 18◦ from each other was

studied as well as the effect of (γ, n) on a long 14N target.

The photofission experiment used an array ofscintillator detectors (see Sect. 9.2.1) to measureoutgoing neutron asymmetries from (γ, n) and(γ, fission) on various targets. The detectors inthe array were close enough to each other for mul-tiple scattering in the detectors to be a concern.

The goal of the present work was to evalu-ate the importance of multiple scattering eventsas well as to determine the background inducedby the gamma ray beam passing through the airin the target room. These effects were simulatedusing a simplified detector setup in geant4.

The detectors themselves were realisticallymodeled in geant4 in order to give correct lightresponse curves [Per10]. Each modeled detectorincluded a cylinder of liquid scintillator (5.08 cmlength and 6.34 cm radius) which was enclosed inan Aluminum container and backed by a photo-tube made entirely of air.

7.1.6.1 Multiple Detector Scattering

Detectors were defined at positions (r, θ, φ) of(58 cm, 0, 0) (Det. 1) and (58 cm, 18◦, 0) (Det.2) measured with respect to the neutron source.Neutrons of 2, 4, and 6 MeV were then fireduniformly into a cone covering, and slightly ex-ceeding, Det. 1. The neutron beam did not hitDet. 2. Any particle seen in Det. 2 must havealready interacted in Det. 1.

En Thr 0.129 Det: 2

Entries 53522Integral 5.338e+04

Incident Energy of Detected Neutrons (MeV)0 1 2 3 4 5 6 7

0

2000

4000

6000

8000

10000

12000

En Thr 0.129 Det: 2

Entries 53522Integral 5.338e+04

En Thr 0.129 Det: 2

Figure 7.15: The energy spectrum of neutrons inDet. 2 resulting from scattering of 6MeV neutron beam in Det. 1.

In [Fig. 7.15] we see 2 peaks, one at .85 of theinitial energy which comes from scattering off of12C in the scintillator fluid, and one at .95 of theinitial energy which comes from scattering off of27Al in the detector. The results of the simula-tion for three neutron energies are given in [Table7.4].

The multiple scattering effect for a given de-tector must be summed over the effect due toeach of the other detectors. This can be approx-imated using the known flux due to one detec-tor, and assuming that each detector creates thesame contribution to the multiple scattering persolid angle. Due to solid angle effects, the clos-est detectors contribute the most of the multi-ple scattering counts. Using this result we eval-uated the aggregate effect of all other detectorson a single detector in the array. The detectorlocated at 90 degrees was selected since it hasmost nearly located neighbors. The calculatedyields are given in row 3 of [Table 7.4]. Whilethis estimate assumes isotropic neutron distribu-


tions with respect to theta, the effect of the lackof isotropy should be much smaller than that ofthe distance.

Table 7.4: Counts in each detector at various in-cident neutron energies

2 MeV 4 MeV 6 MeVDet. 1 5.0 · 107 4.4 · 107 3.7 · 107

Det. 2 4.9 · 104 6.0 · 104 5.4 · 104

Sum 1.6 · 105 1.9 · 105 1.7 · 105

Ratio .3% .4% .5%

The ratios of total multiple scattering eventsto total single scattering events for mono-energetic incident neutron energies are given inrow 4 of [Table 7.4]. If each detector recieved, at6 MeV, a signal equal to the number of counts inDet. 1, then the top 90◦ detector would likely seeabout 1/200 of it’s signal due to multiple scatter-ing from the other detectors. This backgroundwould not have energy of 6 MeV, rather, as canbe seen in [Fig. 7.15] most of the backgroundoccurs at some fraction of the initial energy. Thevalue of this fraction depends on the scatte ringangle between the 2 detectors. Since the mostrelevant detectors are near 90 degrees, we canapproximate that all of the multiple scatteringhits take place at .85 of the initial energy.

Figure 7.16: Energy spectrum seen in top 90◦ de-tector from incident circularly polar-ized 15.5 MeV γ-ray beam on 238U

In [Fig. 7.16] we see that there are about

200 counts in the 6 MeV bin. If all detectors sawa similar spectrum then the bin correspondingto 6 · .85 = 5.1 MeV (which also contains about200 counts) would contain an additional 1/215counts or about 1 count as multiple scatteringbackground, this is a signal to background ratioof about 1 in 200. Similar calculations for 4 and2 MeV yield about 2 counts (out of 500) at 3.4MeV and 5 counts (out of 1200) at 1.7 MeV.

7.1.6.2 (γ, n) in Air

We evaluated background originating from thepresence of air in the target room. The 15.5 γ-ray beam was fired at an 8 m long air targetat standard density. For comparison, a 6.6 mmthick 238U target was used in the same condi-tions. Detectors were defined at (58 cm, 165◦, 0)(Det. 1), and (58 cm, 15◦, 0) (Det. 2). It shouldbe noted that these simulated detectors are closerto the beamline than any of the detectors in theactual array. This means that the backgroundsimulated should be higher than the actual back-ground encountered in any detector. The neutroncounts above 2 MeV per 1.6 · 108 γ-rays in eachdetector are shown in [Table 7.5].

Table 7.5: Neutron counts in each detector foreach target

Air 238U RatioDet. 1 0.4 ± 0.1 75 ± 9 0.5% ± 0.1%Det. 2 0.4 ± 0.1 90 ± 9 0.4% ± 0.1%

The background to signal ratio is given in thethird column of [Table 7.5], it is on the order ofabout .5%.

In summary, both the background from mul-tiple detector scattering and from (γ, n) on 14Nare on the order of 0.5%. Further studies can bedone for multiple scattering by including the en-tire array and using realistic energy and angulardistributions.

[Per10] B. A. Perdue, Ph.D. thesis, Duke Uni-versity, 2010.


7.1.7 New Electric and Magnetic Dipole Transitions in 238U

S.L. Hammond, A.S. Adekola, C.T. Angell, C.R. Howell, H.J. Karwowski, J.H. Kelley,E. Kwan, G. Rusev, A.P. Tonchev, W. Tornow TUNL

Precise experimental information on the distribution of M1 and E1 transitions in 238U has been

obtained with the nuclear resonance fluorescence technique for ground-state deexcitations and

for decays to the 2+ state using photon beams with incident energies between 2.0 and 5.5

MeV. These measurements are compared with a modern QRPA calculation.

Much experimental effort has been focusedon measuring the magnetic dipole and electricdipole strengths in heavy-mass nuclei near sta-bility. These dipole states are important be-cause they can characterize the collective natureof the nuclear excitations [Hey10]. Below theneutron separation energy (Sn) in the actinidenuclei, narrow energy regions have been selec-tively chosen in order to measure the orbital M1“scissors” mode. Previous nuclear resonance flu-orescence experiments on 238U, for example, haveprimarily measured the summed strength of thismode between 1.7 and 2.5 MeV [Hei88, Zil95].However, current theoretical calculations usingthe quasi-particle random phase approximation(QRPA) suggest that substantial γ-ray strengthfrom both M1 and E1 excitations should exist be-tween the “scissors” mode and the giant dipoleresonance (GDR) in neutron-rich, deformed nu-clei [Kul10, Sol99]. Above Sn, the GDR has beenstudied extensively and is not of concern for thiswork.

There are also important nuclear forensic in-terests in the low-energy dipole states of 238U.The spectroscopic data obtained can be used forisotope identification of special nuclear materials(SNM) within shipment containers and for situ-ations where tracing the origin of an SNM is im-portant to national security. Currently the exist-ing spectroscopic nuclear data about the scissorsmode states [Hei88] are used within exploratorytechniques for shipment interrogation. Yet it isknown that higher energy γ rays are more pen-etrating than lower energy ones, even though adecrease in strength is associated with states ex-cited by higher energy γ rays. Further identifi-cation of γ rays from 238U can help to improveisotope identification in shipment interrogation.In a similar fashion, nuclear resonance fluores-cence techniques could also prove useful in the

management and control of nuclear-waste assays.

The measurements described in this re-port were performed at the HIγS facility usingnearly monoenergetic, high-intensity, and 100%linearly-polarized photon beams at excitation en-ergies of 2.0 to 5.5 MeV. The energy spread of thephoton beam, shaped by a 1.9 cm lead collima-tor, was ∼2–5%. The beam was polarized in thehorizontal plane. The scattered γ rays were de-tected with six high-purity germanium (HPGe)detectors: two detectors were placed perpendic-ular (vertical) and two were placed parallel (hor-izontal) to the polarization plane, with two addi-tional detectors at the backward polar angles of135◦ in the horizontal scattering plane. All de-tectors were furnished with passive copper andlead shielding. A similar configuration was usedwith highly segmented HPGe detectors. Specificdetails for both detector arrays are described in[Ham09] and references therein. The detectorswere placed about 10 cm away from a 238U targetwith a mass of either 12.9 or 19.2 g, dependingon the intensity of the photon beam. The targetwas positioned inside an evacuated plastic tubein order to reduce the background in the mea-sured spectra. The photon beam irradiated thetarget for 6 to 8 hours per energy.

The photon beam energy was measured at lowflux with an HPGe detector placed in the pathof the beam. The photon fluxes during data col-lection ranged from 3×106 to 1×107 γ/s as mea-sured by the HPGe detector placed at 11.2(1)◦

relative to the beam direction and by observingCompton scattering from a 1.1-mm thick copperplate.

New spectroscopic data—including excita-tion energies, integrated cross sections, decaywidths, parities, branching ratios, and transitionstrengths—were determined in 238U below Sn.Data collected at photon beam energies of 2.8 to


4.2 MeV have been analyzed so far. The dipolestates found in the current work were studied bycomparing spectra measured in the horizontal,vertical, and backward-angle detectors. This en-abled the identification of 27 E1, ten M1, andone E2 states. In addition to the deexcitation tothe ground state, strong transitions to the firstexcited state, located at an excitation energy of45 keV, were also observed for all states. See Ref.[Ham09] for 238U spectra at an excitation energyof 3.1 MeV.

For incident γ-ray energies in the range of 2.84to 4.20 MeV, the summed M1 strength was foundto be 4.7(8) μN

2 with a mean excitation energyωM1 of 3.1(2) MeV, and the summed E1 strengthwas 115(10) × 10−3 e2fm2 with an ωE1 of 4.0(1)MeV.

The strengths of the dipole states found in thecurrent work were of a similar magnitude to thestrengths predicted by QRPA calculations doneon 238U [Kul10]. This calculation by Kuliev etal. was fully renormalized and involved numericalcalculations using a Warsaw deformed Woods-Saxon potential on 232Th, 236U, and 238U to ob-tain the single particle energies and the summedstrengths. For Eγ = 2.0 to 4.0 MeV, the calcula-tion for 238U predicts a summed M1 strength of∼6 μN

2 with an ωM1 of 2.8 MeV and a summedE1 strength of ∼ 33×10−3 e2fm2 with an ωE1 of3.0 MeV.

Although the calculation was able to predictthe existence of both M1 and E1 states between2.7 and 4.0 MeV (see Fig. 7.17), it was not ableto predict well the details, such as the locationof each state, its strength, and the distributionof the strength over the interval studied. It un-derestimated the summed M1 strength by 40%and the summed E1 strength by 72% in the en-ergy interval of 2.7 to 4.0 MeV. Over half of thepredicted M1 strength lies between 2.0 and 2.5MeV and is a part of the scissors-mode strength.Away from this narrow energy region, there isless M1 strength predicted. This is reproducedin the experiment where few states above the de-tector sensitivity were found away from the scis-sors mode, and no discrete M1 states at all weredetected above 3.5 MeV. As for the E1 strength,the number of states should increase as the ex-citation energy moves toward Sn and the regionwhere the GDR is dominant. However, the E1states may be a part of a low-lying pygmy reso-nance which exists separately from the GDR. Acomparison of the strength from the tail of theGDR that would be in this energy region withthe experimental strength needs to be made inorder to determine whether the observed statesare from a pygmy resonance [Ton10].

We plan to finish the analysis of the data for

238U from Eγ = 2.0 to 5.5 MeV. This involves (1)identification of any new states between 2 and 3MeV, (2) comparison of the strengths for knownstates found in this work with those from previ-ous measurements, and (3) investigation of thecontinuum within the entire range. These newdata will help to improve the understanding ofM1 and E1 excitations found in deformed nuclei.This is necessary to constrain modern QRPA cal-culations.

Figure 7.17: (Color online) Experimental (a) M1and (b) E1 strengths (points withstatistical error bars) compared withQRPA calculation (vertical bars).

[Ham09] S. L. Hammond et al., TUNL ProgressReport, XLVIII, 111 (2009).

[Hei88] R. D. Heil et al., Nucl. Phys., A476,39 (1988).

[Hey10] K. Heyde et al., Rev. Mod. Phys.,(2010), (accepted).

[Kul10] A. A. Kuliev et al., Euro. Phys. J., A43(2010).

[Sol99] V. G. Soloviev et al., J. Phys. G, 25,1023 (1999).

[Ton10] A. Tonchev et al., Phys. Rev. Lett.,104, 072501 (2010).

[Zil95] A. Zilges et al., Phys. Rev. C, 52, R468(1995).


7.2 Pyroelectric Research

7.2.1 Neutron Production with a Pyroelectric Double-Crystal AssemblyWithout Nano-Tip

W. Tornow, W. Corse, J. Fox, S. Crimi, TUNL

Two cylindrical LiTaO3 crystals facing each other’s deuterated circular face were exposed

to deuterium gas at an ambient pressure of about 12 mTorr. With a distance of about 4

cm between the z+ and z− crystal faces, neutrons were produced via the 2H(d, n)3He fusion

reaction during the heating and cooling phase of the crystals. The 2.5 MeV neutrons were

detected with organic liquid scintillation detectors. During the cooling phase deuterium ion

beam (D+2 ) energies of up to 400 keV were obtained as deduced from the associated electron

bremsstrahlung endpoint energy. In contrast to earlier studies, an electric-field enhancing

nano-tip was not employed. Neutron yields up to 500 per thermal cycle were observed,

resulting in a total neutron production yield of about 1.6 × 104 neutrons per thermal cycle.

Since the ground-breaking work of the UCLA[Nar05] and RPI [Geu06] groups in 2005, theproduction of neutrons with pyroelectric crystalshas been the subject of considerable interest, butalso of some skepticism, which is based on thelack of reproducibility. In the present work weremoved the two main sources of potential irre-producibility, the nano-tip and any issues relatedto the detection of 2.5 MeV neutrons in the pres-ence of intense and high-energy bremsstrahlungX rays. We adopted the double-crystal scheme[Geu06] of the RPI group, and for the first timeobserved the production of neutrons during boththe heating and cooling phase of the crystal as-sembly.

Figure 7.18: Cartoon of double-crystal arrange-ment. The polarity of the polariza-tion charges shown is for the coolingphase.

Following our recent work [Tor10] where ac-celeration potentials of up to 310 kV have beenachieved with a single pyroelectric crystal, we as-

sembled two 2.5 cm long and 2.5 cm diameterLiTaO3 crystals. Figure 7.18 shows a cartoon ofthe double-crystal assembly. The crystal to theleft was z-cut to provide a positive polarizationcharge during cooling, while at the same time thez-cut crystal shown on the right side provided anegative polarization charge. During the heatingphase of the crystals the two polarization chargesare inverted. Deuterated (99.3 %) polyethylenewith thickness of 0.3 mg/cm2 was evaporated onthe crystal faces. The separation between thetwo crystal faces was 3.9 cm, and the deuteriumgas pressure was typically 10 to 13 mTorr. Theheating phase from room temperature to 130 ◦Clasted 30 minutes, while the optimal cooling timefrom 130 ◦C to 0 ◦C was found to be 20 minutes.Typically, during the heating and cooling phaseswe experienced two and three discharges, respec-tively. Discharges occur once the electrostatic po-tential between the two crystals or between thecrystal and their surroundings exceeds the break-down potential of the ionized gas. Ultimately,discharges were the limiting factor in reachinghigher potentials than 400 kV. A HPGe planardetector positioned outside of the vacuum cham-ber was used to determine the potential energy.Two 5” diameter and 2” thick neutron detectors(filled with Bicron 501A liquid scintillator fluid)were positioned on opposite sides of the vacuumchamber. They were encased within cylindersmade of lead with 1.25 cm thick front face and1.25 cm wall thickness to provide sufficient X-rayattenuation for reliable neutron detection. In ad-


dition, two 2” diameter and 2” thick neutron de-tectors, also surrounded by lead, were placed ontop of the chamber. The neutron detectors haveexcellent neutron-gamma pulse-shape discrimi-nation properties and were connected to a MPD-4 module from Mesytec which provided an am-plified pulse-height spectrum and a pulse-shapespectrum with clear separation between gamma-ray/X-ray and neutron induced events.

Figure 7.19 shows a typical two-dimensionalpulse-shape (horizontal axis) versus pulse-height(vertical axis) spectrum obtained with a dis-criminator threshold set to one-fourth times theCompton edge of 137Cs gamma rays, i.e., 85keV. This electron-energy threshold correspondsto about 450 keV neutron energy for the scintilla-tor type used in the present work. The maximumneutron energy expected from the 2H(d, n)3Hereaction is about 2.5 MeV at the angular po-sition of the neutron detectors. Our four neu-tron detectors were gain-matched and they havewell-known neutron detection efficiencies [6]. Theevents located inside of the rectangular gate ofFig. 7.19 are almost exclusively due to the neu-trons of interest, while the band to the left, whichextends to larger pulse heights, is due to room-background γ rays.

Pul

se H

eigh

t (ar

bitr

ary

units

)

PSD Timing (arbitrary units)

Figure 7.19: Two-dimensional spectra of pulseshape versus pulse height for ourneutron detectors. The horizontalaxis shows the zero-crossing timeof X ray and room background γ-ray induced pulses (strong verticalbands) and neutron induced pulses(surrounded by rectangles) for thetwo large neutron detectors (left andcenter spectrum) and for the twosmall neutron detectors added to-gether (right spectrum). The verti-cal axis displays the pulse height.

During a complete thermal cycle which typ-ically lasts about one hour, neutrons of sub-stantial rate above background were producedonly during a few time intervals each lasting oneto five minutes. These time intervals are sep-arated by discharges and the associated recov-ery time. A maximum net rate of 110 neutronsper minute has been observed. Using the neu-tron detection efficiencies, the geometry of ourexperimental set-up and the neutron attenuationin the lead shield placed around the neutron de-tectors, we find that the maximum yield of 493neutrons per thermal cycle found in the presentwork corresponds to 16250±1620 neutrons gen-erated. This yield is about 60% higher than themost recent yield reported by the RPI group in[Gil09] using an electric field enhancing nano-tip.However, it is about a factor of 3.7 lower thanthe maximal yield reported in [Geu07] by thesame group. Taking into account the consider-ably higher energies achieved in the present workand the fact that neutrons were produced bothduring the heating and cooling phase of the crys-tals, a substantially higher neutron yield couldhave been expected. Obviously, the nano-tipsused in [Nar05, Geu06, Geu07] are providing alarge enhancement factor. This conjecture willbe explored in the near future.

The authors acknowledge the contributionsof B. Carlin, A.S. Crowell, G. Rich, and S.M.Shafroth to this work.

[Geu06] J. Geuther, Y. Danon, and F. Saglime,Phys. Rev. Lett, 96, 054803 (2006).

[Geu07] J. Geuther and Y. Danon, J. Appl.Phys., 90, 174103 (2007).

[Gil09] D. Gillich et al., Nucl. Instrum. Meth-ods A, 602, 306 (2009).

[Nar05] B. Naranjo, J. Gimzewski, and S. Put-terman, Nature, 434, 1115 (2005).

[Tor10] W. Tornow, S. Lynam, and S. Shafroth,J. Appl. Phys., 107, 063302 (2010).


7.3 Nuclear Data Evaluation

7.3.1 Nuclear Data Evaluation Activities

J.H. Kelley, E. Kwan, C.G. Sheu, D.R. Tilley, H.R. Weller, TUNL; J.E. Purcell, GeorgiaState University, Atlanta, GA

The Nuclear Data Evaluation Group at TUNL is part of the United States Nuclear Data Pro-

gram and the International Nuclear Structure and Decay Data network. After the retirement

of Fay Ajzenberg-Selove in 1990, TUNL assumed responsibility for evaluation of nuclides in

the mass range A = 3 to 20. The status of the published evaluations and preliminary reviews

is presented in Table 7.6.

Along with producing evaluations of theA = 3 to 20 nuclei in the “Energy Levels ofLight Nuclei” series that is published in Nu-clear Physics A, the Nuclear Data EvaluationGroup has been charged with providing the cor-responding updates to the Evaluated NuclearStructure Data Files (ENSDF) database that ismaintained at the National Nuclear Data Cen-ter (NNDC) at Brookhaven National Laboratory.We also provide a web-based service for the nu-clear science and applications communities athttp://www.tunl.duke.edu/nucldata/.

7.3.1.1 Publications

Table 7.6 displays the status of our most recentpublished evaluations and preliminary reviews.A first draft of a new evaluation for A = 11 isanticipated for fall 2010.

Table 7.6: Current publication status.

Nuclear Mass Publication Institution

Published:

A = 3 Nucl. Phys. A474 (1987) 1 TUNLA = 4 Nucl. Phys. A541 (1992) 1 TUNL a

A = 5 to 7 Nucl. Phys. A708 (2002) 3 TUNL a,b

A = 8 to 10 Nucl. Phys. A745 (2004) 155 TUNL c

A = 11, 12 Nucl. Phys. A506 (1990) 1 Penn d

A = 13 to 15 Nucl. Phys. A523 (1991) 1 Penn d

A = 16, 17 Nucl. Phys. A564 (1993) 1 TUNLA = 18, 19 Nucl. Phys. A595 (1995) 1 TUNLA = 20 Nucl. Phys. A636 (1998) 247 TUNL e

Reviews in Progress:

A = 3 Submitted to Nucl. Phys. A, TUNLJune 2010

A = 11 Expected late 2010 TUNLA = 12, 13 Expected 2011 TUNL

a Co-authored with G.M. Hale, LANL.b Co-authored with H.M. Hofmann, Universitat Erlangen-Nurnberg, Germany.c Co-authored with D.J. Millener, BNL.d F. Ajzenberg-Selove, University of Pennsylvania.e Co-authored with S. Raman, ORNL.

7.3.1.2 Evaluated Nuclear StructureData Files

The ENSDF files contain concise nuclear struc-ture information such as tables of adopted levelenergies and tables of properties for levels thathave been observed in various nuclear reactionsand decays. The ENSDF files are updated con-currently with the last published reviews in the“Energy Levels of Light Nuclei” series.

Work on the A = 3 and A = 11 to 13 ENSDFfiles is presently underway.

7.3.1.3 Experimental Unevaluated Nu-clear Data List

TUNL has taken responsibility for creating A = 2to 20 data sets for the Experimental UnevaluatedNuclear Data List (XUNDL) beginning in April2009. This activity was developed at McMasterUniversity by Dr. Balraj Singh, who aimed toquickly provide the most current data to users inthe high-spin community. The nuclear structuredata in recent references is compiled in XUNDLwith minimal evaluation effort. Since becominginvolved, TUNL has prepared roughly 100 datasets, a rate of 6 per month on average. The datasets are reviewed by Dr. Singh and then addedto the database at NNDC.

7.3.1.4 World Wide Web Services

Our group continues to develop web-based ser-vices for the nuclear science and applicationscommunities. The website layout and contentsare constantly revised and kept up to date to en-sure high-quality service and accurate informa-tion. Figure 7.20 displays the usage of our web-site from the nuclear science communities since


April 2002.The following items are currently available:

• Ground-State Decay Data to provide eval-uated data from recent work on ground-state β decays and charged-particle decaysas well as compiled data from earlier mea-surements has been completed for nuclideswith A = 3 to 20.

• Thermal Neutron Capture Data on A = 2to 20 nuclei is provided based on compileddata.

• PDF and HTML documents are onlinefor TUNL’s and Fay Ajzenberg-Selove’s re-views from 1959 to the present. The PDFversions include hyperlinks for references,Tables of Recommended Level Energies,Electromagnetic Transitions Tables, Gen-eral Tables, Energy Level Diagrams, andErratum to the Publications. The HTMLdocuments are more comprehensive thanthe PDF documents, as they include hyper-links to tables in the PDF and PS formats,reactions and reaction discussions, TUNLand NNDC references, Energy Level Dia-grams, and General Tables.

We have essentially completed the re-creation of PDF files for our publications,in order to provide the most current NNDCreference keys and to correct all errorsfound since the articles went to press. We

will continue to work on the correspondingHTML and table/PS/PDF files.

• Energy Level Diagrams for publicationyears from 1959 to the present are providedin GIF, PDF and EPS/PS formats.

• Tables of Energy Levels provides a brieflisting of the tables of recommended energylevels (in PDF and PS formats) from themost recent publications for nuclides withA = 4 to 20.

• General Tables that reference theoreticalwork related to TUNL’s most recent re-views are available on our website for themasses A = 5 to 10. The tables includedynamic links to the NSR (Nuclear ScienceReferences) database.

• ENSDF information for A = 3 to 20 nu-clides is available through the NNDC site.

• A link to NuDat (Nuclear Structureand Decay Data), which allows usersto search and plot nuclear-structure andnuclear-decay data interactively, is avail-able through the NNDC site.

• Links to the National Nuclear Data Centerand other useful sites, as well as to the on-line electronic journals that the nuclear sci-ence communities use most often, are pro-vided. There is also a sitemap that containsa complete listing and links to everythingon the website.

Figure 7.20: (Color online) Overview of web usage deduced from Analog Web Analysis Package

Accelerator Physics

Chapter 8

• The High Intensity Gamma Ray Source (HIγS)

• The FN Tandem Accelerator and Ion Source

• The LENA Accelerator and Ion Sources

160 Accelerator Physics TUNL XLVIII 2009–10

8.1 The High Intensity Gamma Ray Source (HIγS)

8.1.1 Developing New and Enhanced Gamma-ray Capabilities for HIGSResearch: I : Eγ = 60 MeV Operation and Helicity Switch

Y.K. Wu, J. Li, S.F. Mikhailov, V.G. Popov, P. Wallace, TUNL

In 2010, the HIGS research and development was focused on producing high-performancegamma-ray beams for user experiments with a high degree of consistency and predictabil-ity, and on developing new gamma-ray beam capabilities. The new capabilities include thegamma-beam helicity switch using the OK-5 FEL, and a high-flux, high-energy gamma-beamoperation at 60 MeV. Using 190 nm mirrors, the operation energy of HIGS gamma-beams willbe extended to about 100 MeV in the near future.

During the last reporting period (Aug. 2009to Jul 2010), we focused on improving thepredictability of producing high-performancegamma-ray beams for user experiments and ondeveloping new capabilities for HIGS. Usingnewly acquired high-reflectivity 240 mirrors, weextended the HIGS user operation from about45 MeV in 2009 to about 60 MeV in 2010. Withcircularly polarized OK-5 FEL, we developed anew capability of switching the helicity of theHIGS gamma-ray beam in a few minutes.

One major direction of the continued HIGSdevelopment is to increase the energy of theCompton gamma-ray beam at a high flux for userexperiments. This requires us to develop reliableFEL mirrors in the deep UV and VUV region.One of our mirror vendors, Laser Zentrum Han-nover (LZH), Germany, is a world leader in devel-oping special radiation-resistive mirror coatingsin a wide range of wavelengths. Over the years,the visible and UV mirrors produced by this com-pany have proven to have a higher degree of radi-ation hardness and a longer mirror lifetime thanmirrors produced by other companies. In 2009,we worked with LZH closely to develop high-reflectivity, multilayer dielectric mirrors at 450,350, 240, and 190 nm. In May and June, 2010,we commissioned HIGS operation with new 350and 240 nm mirrors. Using the low-loss 350 nmmirrors, a very high level of gamma-ray flux canbe realized. For example, operating the storagering at 701 MeV, a 26 MeV gamma-ray beamwas produced with an on-target flux of 1–2× 107

γ/s after a 12 mm diameter collimator. Subse-quently, these mirrors were used to produce 15 –26 MeV gamma-ray beams for user experiments.With 240 nm mirrors, we were able to produce

a 60 MeV gamma-ray beam with good stabilityand a flux level adequate for the planned userexperiments.

As of June, 2010, we have a set of productionFEL mirrors in house to cover a wide range ofgamma-ray energies from 1 MeV to 60 MeV. Us-ing the newly acquired 190 nm mirrors, we planto reach the next gamma-ray energy milestone ofabout 100 MeV in 2011.

To reduce systematic errors in experimentsusing circularly polarized gamma-ray beams,many experiments require the switch of thegamma-ray beam helicity. At HIGS, the helicityof the gamma beam can be changed by switch-ing between left- and right-circular polarizationsof the FEL beam produced by the helical OK-5 wiggler. The helicity of the OK-5 wiggler, anelectromagnet, can be switched by reversing thedirection of the DC current in one of the twosets of wiggler coils. In our system, the currentreversion is realized using a solid-state DC cur-rent switch rated for a maximum current of 3 kA.The wiggler helicity switch is performed in threesteps: (1) the wiggler current is first ramped tozero; (2) the solid-state switch reverses the direc-tion of the current in one of wiggler coils; and (3)the wiggler current is ramped back to the oper-ational value. During this process, the magneticoptics of the storage ring have to undergo a se-ries of transitions accordingly in order to sustaina large circulating beam current in the storagering without causing beam loss.

The wiggler helicity switch system was com-missioned in Fall, 2009. This system can be op-erated with either one, or two OK-5 wigglers, de-pending on the configuration of the FEL. Op-erating the OK-5 FEL with a single wiggler at

TUNL XLIX 2009–10 Accelerator Physics 161

770 nm, we demonstrated the gamma-ray beamhelicity switch (see Fig. 8.1). Reversing the wig-gler current (from 2.3 kA to 0 to −2.3 kA) wascarried in less than 90 seconds. No significantelectron beam loss was observed. Additional timewas spent on optimizing the FEL operation bytuning up the FEL optical resonator, and on re-covering the FEL power by refilling the storagering to bring the electron beam current back toa level before the helicity switch-over. The totaltime needed for the helicity switch and subse-quent FEL tuning is about 5 to 10 minutes. Weplan to speed up the helicity switch process to re-duce the switch time to less than 5 minutes whileminimizing the gamma-ray flux variation beforeand after the helicity switch.

0 500 1000 1500 2000 25000

500

1000

1500

2000

2500

3000

Iw [A

]

Time [s]

Eb = 484 MeV, OK5W lasing, 770 nm, two bunch

N R N R N

0 500 1000 1500 2000 25002

2.5

3

3.5

4

4.5

5

FE

L P

ower

[a.u

]

Figure 8.1: (color online). Helicity switch demon-stration between the normal and re-versed polarity with the OK-5 FELlasing at 770 nm. The correspondinggamma-ray beam energy is 7.1 MeV.The blue curve is the wiggler currentand the green curve is the extractedFEL power.


8.1.2 Developing New and Enhanced Gamma-ray Capabilities for HIGSResearch: II: Mirror Perfromance Monitoring

Y.K. Wu, J. Li, S.F. Mikhailov, V.G. Popov, P. Wallace, TUNL

In 2010, the HIGS research and development was focused on producing high-performancegamma-ray beams for user experiments with a high degree of consistency and predictability,and on developing new gamma-ray beam capabilities. The new capabilities include a mirrorperfromance monitoring program.

The maximum total gamma-flux producedat the HIGS facility is up to 1010 γ/s around10 MeV. Production of this high level of flux re-quires a very large average FEL power inside theresonator, exceeding 1 kW. The high-power FELoperation can cause significant degradation of theFEL mirrors due to very intense wiggler har-monic radiation, especially when operating theFEL in the UV region at a high electron beamenergy. Taking advantage of the fact that higher-order harmonic radiation of the helical wiggler ispeaked off-axis, we have limited the high-energy,high-flux HIGS operation mostly to circular po-larization. To ensure the predictability and sta-bility of the HIGS operation for user researchprograms, we have developed a comprehensiveprogram to monitor the performance of the FELmirrors on a regular basis. This program has en-abled us to use a particular set of FEL mirrorsfor a few hundred hours of high gamma-flux op-eration with predictable performance.

Figure 8.2: (color online). The measured FELcavity losses over time. Upper plot:FEL cavity loss for 780 nm mirrors;Lower plot: FEL cavity loss for 540nm mirrors;

The degradation of the FEL mirrors clearlyshows up in the measured cavity loss curve as afunction of wavelength. The mirror degradationcan be monitored using three measurable effects,namely (1) an increase of the minimum cavityloss; (2) a shift of the wavelength at the minimalcavity loss toward the longer wavelength region;and (3) a reduction of the high-reflectivity band.Fig. 2 shows the measured FEL cavity losses asa function of wavelength over a period of morethan 12 months for two production mirror setsat 780 and 540 nm. For these mirrors, the mini-mal cavity loss doubled after being used for high-flux user experiments for about 380 (780 nm) and520 hours (540 nm), respectively (see Fig. 8.3).These mirrors are expected to be useful for ad-ditional two to three hundred hours of high-fluxoperation for user experiments.

For the present HIGS operation, we have anumber of production mirrors with their centerwavelengths around 1060, 780, 540, 450, 350, and


240 nm. The performance degradation of thesemirrors is measured and evaluated after each pe-riod of extended use. This mirror performancemonitoring program has enabled us to reliablyproject the performance of the gamma-ray beamas mirrors degrade. Based upon the measuredmirror loss curve, we can deploy a set of usedFEL mirrors with confidence for a specific userexperiment, knowing that this mirror set will beable to produce the required gamma-beam per-formance. This practice extends the useful du-ration of the FEL mirrors without compromisingthe performance of the gamma-ray beam. It alsomakes it possible to plan for the procurement ofnew mirrors for future HIGS operation.

Figure 8.3: (color online). The measured increaseof the minimum FEL cavity loss as afunction of hours of operation for 780and 540 nm FEL mirrors.

Currently, we focus on two areas of FEL mir-ror development. The first one is the develop-ment of special long wavelength mirrors from in-frared to UV. Working closely with mirror ven-dors, we will develop new FEL mirrors whichwill allow high-flux HIGS operation in energy re-gions which are poorly covered by the stock ofin-house mirrors. In addition, we are investigat-ing the possibility of developing high-reflectivity,dual-band mirrors which can be used to producehigh-flux gamma-ray beams in two different en-ergy regions. The use of this type of mirrors will

significantly reduce the number of FEL mirrorchanges, therefore reducing the effort and riskassociated with each change of FEL mirrors in-stalled in an ultra-high vacuum system.

Our second focus is to develop durable VUVFEL mirrors for high-energy gamma-ray pro-duction. In Spring 2010, we received our firstVUV FEL mirrors at 190 nm. Using these mir-rors, we expect to increase the energy range ofthe HIGS operation to about 100 MeV. To ex-tend the HIGS operation toward and beyond thepion-threshold energy of about 150 to 160 MeV,radiation-resistive 150 nm FEL mirrors with rea-sonably high reflectivity (> 90%) should be de-veloped. Unlike mirrors above 190 nm whichuse Oxygen based materials for mirror substratesand coating, VUV mirrors below 190 nm re-quires the use of Fluoride based materials. Thischange of materials for substrates and coatingsrequires the vendor to optimize the coating tech-nique to produce high-density Fluoride coatingsand to develop effective protective overcoat lay-ers to enhance the radiation resistance of the mir-ror. Working closely with LHZ, we plan to de-velop prototype Fluoride based mirrors at about175 nm to test their durability. These mirrorswill allow us to produce gamma-ray beams witha maximum energy of 115 MeV. If the develop-ment of prototype 175 nm mirrors is successful,we will move on to the development of 150 nmmirrors which are critical for photo-pion physicsresearch.

The VUV mirrors are significantly lossy com-pared with mirrors in the visible region. A highergain FEL with four OK-5 wigglers will be nec-essary for the HIGS operation with VUV FELmirrors. This requires the installation of two ad-ditional OK-5 wigglers in the FEL straight sec-tion of the Duke storage ring. To retain the lin-ear polarization capability of the HIGS which isenabled by the planar OK-4 wigglers, we are de-veloping a wiggler switchyard system to mechan-ically switch between two planar OK-4 wigglersand two helical OK-5 wiggler in the middle ofthe FEL straight section. This wiggler switch-yard will be installed and commissioned in 2011.


8.1.3 Accelerator Physics Program: I : Free-Electron Laser Research

Y.K. Wu, TUNL

The accelerator physics program at the DFELL is focused on making a direct impact on theHIγS research programs by enhancing accelerator performance and creating new capabilities.The FEL research is aimed at increasing the FEL power and gain using various wiggler con-figurations and low-loss FEL mirrors. This research will lead to a higher gamma-ray flux andwill extend the HIGS operation towards 100 MeV and above.

The accelerator physics research is critical tothe continued development of the High IntensityGamma-ray Source (HIGS) to enhance its per-formance and create new capabilities. At thepresent time, we focus our research effort in twoimportant areas: (1) the FEL physics research toincrease the FEL power and FEL gain, and to re-duce losses in the optical resonator; (2) the beaminstability research to stabilize electron beam op-eration at a higher beam current. The effort inthese areas can make a direct impact to the HIGSdevelopment, by increasing the gamma-ray fluxand extending the gamma beam energy first to100 MeV, and later to above the pion-thresholdenergy (> 150 MeV). We report here on the FELphysics research.

The Duke storage ring is a dedicated accel-erator driver for FELs. Four FEL wigglers arecurrently installed in the 34 m long straight sec-tion, including two planar OK-4 wigglers in themiddle section and two helical OK-5 wigglers onthe sides (see Fig. 8.4). The FEL wigglers canbe configured in several ways to produce coher-ent radiation with different polarizations, and foreither high-power or high-gain operation. Com-monly used FEL configurations include (1) high-power FEL configurations with a single wiggler,either a planar OK-4 wiggler or a helical OK-5wiggler; (2) high-gain FEL configurations withan optical klystron using either two OK-4 wig-glers or two OK-5 wigglers; (3) the highest gainconfiguration with a distributed optical klystronFEL (DOK-1) with all four wigglers [Wu06] . Bycolliding an intense electron beam inside the lasercavity, the FEL beam is used as the photon driveto power the High Intensity Gamma-ray Source(HIGS) [Wel08].

The storage ring based oscillator FELs arethe main coherent light source at the DFELL.The Duke storage ring FEL has been operatedin a wide wavelength range from infrared (λ ∼

1 to 2 microns) to vacuum ultraviolet (VUV)(194 nm). In the recent years, we have developedvarious experimental and simulation techniquesto study the storage ring FEL.

The electron beam energy spread is a very im-portant parameter for understanding the longitu-dinal beam dynamics and beam instabilities. Forthe storage ring FEL, the induced energy spreaddetermines the FEL power. The electron beamenergy spread also determines the minimal en-ergy spread of a collimated HIGS gamma beam.We have recently developed a versatile methodto accurately measure the electron beam relativeenergy spread from 10−4 to 10−2 using the radi-ation from an optical klystron which consists oftwo FEL undalators (or wigglers) sandwiching abuncher magnet. A novel numerical model basedupon the Gauss-Hermite expansion has been de-veloped to treat both spectral broadening andmodulation on an equal footing. A large dy-namic range of the measurement is realized byproperly configuring the optical klystron. In ad-dition, a model based scheme has been developedto compensate the beam emittance induced inho-mogeneous spectral broadening effect. Using thistechnique, we have successfully measured the rel-ative energy spread of the electron beam in theDuke storage ring from 6×10−4 to 6×10−3 witha high degree of accuracy [Jia10].

Using the newly developed energy spreadmeasurement technique, an experimental studyon the FEL power scaling with the electron beamcurrent, energy, and energy spread has been car-ried out. Measurements have been made to di-rectly link the FEL power to the measured en-ergy spread of the electron beam. The exper-imental results (Fig. 8.5) show that the rela-tion between the FEL power and induced energyspread holds well for different optical detuningconditions. This research is expected to enableus to predict the FEL power and FEL induced


e-beam from linac

e-beam

collision point

e-beam

FEL beam

booster

FEL mirror

FEL mirror

gbeam

LTB

BTR

synchrotron

OK-4A

OK-5A

OK-5B

OK-4B

0.18 GeV

0.24−1.2 GeV

52.8 m

0.18−1.2 GeV

UTR

Collimator

GV

Figure 8.4: (color online). The schematic of the Duke storage ring accelerator facility including thelinac pre-injector, full-energy top-off booster injector, and the storage ring.

energy spread for different levels of optical cavityloss as the FEL mirrors degrade. A reliable pre-diction of the FEL beam parameters will enableus to produce detailed projections of the gamma-ray beam performance, therefore allowing furtheroptimization of the user experiments using theHIGS gamma-ray beam.

One focus of our FEL research program is torealize FEL lasing at VUV wavelengths (200 -150 nm) with FEL mirrors which have a lower re-flectivity than visible mirrors. The VUV FEL re-search will require a higher gain FEL with a longwiggler system which can be realized at Dukewith four helical OK-5 wigglers. In addition totwo existing OK-5 wigglers, two more OK-5 wig-glers need to be installed in the middle of theFEL straight section (Fig. 8.4). In order to pre-serve the linear polarization capability of the pla-nar OK-4 FEL, we have devised a wiggler switch-yard system to mechanically move in/out eithertwo planar OK-4 or two helical OK-5 wigglers inthe straight section. The mechanical design ofthe wiggler switchyard system is well under way,which will be followed by the fabrication of re-lated vacuum and electrical components for theupgrade. In the meantime, research is being car-ried out to design magnetic optics compensationschemes to overcome the adverse effects of turn-ing on all four FEL wigglers. These compensa-tion schemes will be used to develop transitionalmagnetic optics to allow switching of the helic-ity of the VUV FEL. To minimize the impact ofthe wiggler switchyard project on the HIGS re-search programs, we have planned to carried outthe project in two steps. First, we will design,fabricate, and prepare on the bench all switch-yard related hardware components while operat-ing HIGS for the user experiments. Second, we

will plan a relatively short shutdown to installwiggler switchyard system in 2011 to reduce theoperation downtime and the impact on the sub-sequent HIGS operation after the upgrade.

430 435 440 445 450 455 4600

0.5

1

λ (nm)S

pect

ra (

arb.

uni

ts)

σE,GHfit = 6.00 × 10−4Measured Spectrum

G−H Method

420 430 440 450 460 470 4800

0.5

1

λ (nm)

Spe

ctra

(ar

b. u

nits

)

σE,GHfit = 6.00 × 10−3

Measured SpectrumG−H Method

Figure 8.5: (color online). Measured opticalklystron spectra with a very smalland very large energy spread. Up-per: The electron beam energy is280 MeV, the single-bunch beam cur-rent is 0.045 mA, and the FEL is off.The measured relative electron beamenergy spread is σE = (6.00 ± 0.015) ×10−4. Lower: The electron beamenergy is 400 MeV, the single-bunchbeam current is 40 mA, and the OK-5FEL is turned on to increase the en-ergy spread of the electron beam. Therelative measured electron beam en-ergy spread is σE = (6.00±0.029)×10−3.

[Jia10] B. Jia et al., Phys. Rev. ST Acccel.Beams, 13, 080702 (2010).

[Wel08] H. Weller et al., Prog. Part. Nucl. Phys.,(2008).

[Wu06] Y. Wu et al., Phys. Rev. Lett., page224801 (2006).


8.1.4 Accelerator Physics Program: II: Beam Instabilities Research andFast Feedback Systems

Y.K. Wu, TUNL

The accelerator physics research program at the DFELL provides new insight into the relationsbetween various parameters of the electron beam and FEL resonator and the performance ofthe FEL optical beam. Using advanced bunch-by-bunch longitudinal and transverse feedbacksystems, the beam instability research is aimed at increasing the electron beam current forstable, high-flux HIGS operation. We will also report other accelerator research activitieswhich are important to the further development of the Duke storage ring and other similaraccelerators.

The accelerator physics research is critical tothe continued development of the High IntensityGamma-ray Source (HIGS) to enhance its perfor-mance and create new capabilities. One of the ar-eas of our effort is in the beam instability researchto stabilize electron beam operation at a higherbeam current. The effort in these areas can makea direct impact to the HIGS development, by in-creasing the gamma-ray flux and extending thegamma beam energy first to 100 MeV, and laterto above the pion-threshold energy (> 150 MeV).

The accelerator research program at theHIGS facility also covers a broader range of top-ics, including the nonlinear dynamics of chargedparticles, polarized electron beam research, ad-vanced accelerator design and optimization, andnovel light source development.

In the Duke storage ring, the maximum beamcurrent in the multi-bunch mode is limited bylongitudinal coupled bunch instabilities. To com-bat these instabilities, a bunch-by-bunch lon-gitudinal feedback (LFB) system has been de-veloped. This system employs an integratedGigasample processor (iGp) and a broadband,waveguide overloaded kicker cavity specially de-signed for the Duke storage ring. The LFB sys-tem has been very effective in suppressing lon-gitudinal instabilities with a number of bunchfill-patterns, including symmetric 2-, 4-, 8-, 16-, 32-bunch modes and the completely filled 64-bunch mode. The LFB system has played animportant role in realizing and stabilizing high-flux HIGS operation with a high electron beamcurrent, especially at a low electron beam energywhere natural radiation damping is lacking, orwith UV mirrors which have a lower reflectivity,but a higher level of thermal loading. Fig. 8.6shows the effect of the LFB system to stabilizea 16-bunch electron beam at 1.15 GeV, a high

energy storage ring operation critical for VUVFEL research. When the high ambient temper-ature in the summer causes significant thermaldrifts in a variety of hardware systems, the LFBsystem has been effective in many cases to sup-press beam stabilities, thus preventing electronbeam loss.

0 50 100 150 200 250 3000

1

2

3

4

5

6

7

8

9

10← f

syn

Frequency (kHz)

Pow

er s

pect

rum

(dB

m)

1 2 30

0.2

0.4

0.6

0.8

Time (ns)

Am

plitu

de (

arb.

uni

ts)

0 50 100 150 200 250 3000

0.5

1

1.5

2

2.5

3

3.5

4

← fsyn

Frequency (kHz)

Pow

er s

pect

rum

(dB

m)

1 2 30

0.2

0.4

0.6

0.8

1

Time (ns)

Am

plitu

de (

arb.

uni

ts)

Figure 8.6: (color online). The LFB systembrings the 1.15 GeV, 16-bunch elec-tron beam to stability. The totalbeam current is about 78 mA andcharge is almost evenly distributed inall bunches. Upper: the LFB is off.The beam is unstable with substan-tial synchrotron oscillation sidebandsand unstable longitudinal distribution(inset); Lower: the LFB is turned onand the beam is stable.

The maximum current in a single electron


bunch is limited by transverse instabilities. Re-cently, we started to develop a bunch-by-bunchtransverse feedback (TFB) system by upgrad-ing an analogy system to an field-programmablegate array (FPGA) based digital system usingthe same integrated Gigasample processor as theLFB system. This upgraded TFB is currentlyoperational but further tuning and adjustmentare being carried out to optimize its performance.The effectiveness of TFB to extend the beam cur-rent per bunch in realistic operational conditionshas yet to be demonstrated.

Both the LFB and TFB systems have beenused as diagnostic tools to study beam instabil-ities in the storage ring. These studies will helpgain new insight into mechanisms of various in-stabilities and help us develop remedies to effec-tive eliminate or suppress these instabilities whileextending the stable electron beam operation toa higher current.

Other research activities at DFELL include(1) accurate measurements the electron beam en-ergy in the storage ring using the HIGS gamma-ray beam; (2) the development of novel end-to-end gamma spectrum reconstruction techniquefor the Compton gamma-ray beam; (3) a study ofelectron beam polarization process at 1.15 GeV;and (4) a research program to explore nonlineardynamics of charged particles in a circular accel-erator.

The HIGS gamma-ray beam has been usedas an advanced diagnostic tool to measure elec-tron beam parameters. Recently, we have devel-oped techniques to accurately measure the cen-troid energy of the electron beam in the storagering with a relative uncertainty of a few times10−5 for low energy beams at a few hundreds ofMeV [Sun09a].

In addition, we have developed a novel end-to-end gamma spectrum reconstruction methodusing a numerical model which simulates theentire process of gamma beam production (viaCompton scattering), collimation, and detection[Sun09b]. This method allows us to constructa detailed energy distribution of a Comptongamma-ray beam at various energies, in particu-lar between 10 and 15 MeV in which the conven-tional spectrum reconstruction techniques havedifficulties in producing a truthful gamma-ray en-ergy distribution due to a complex detector re-sponse.

The electron beam in the storage ring canbuild up polarization due to the Sokolov-Ternoveffect. The polarized electron beam will have areduced Touschek loss rate due to electron colli-sion inside a bunched beam. Beam lifetime mea-

surements have been used to study the radiativepolarization buildup process of a 1.15 GeV elec-tron beam, by measuring the time constant ofthe process and equilibrium degree of the polar-ization [Sun10]. To make such an experimentsuccessful, we worked hard to establish highlyreproducible beam conditions to allow repeat-able beam lifetime measurements as a functionof beam currents. In addition, the accuracy ofthe beam lifetime measurement has to be im-proved to better than about 5%. We found thatthe 1.15 GeV electron beam could become highlypolarized with a degree of polarization of 85%.

The research with polarized electron beamscan have a direct impact to HIGS research pro-grams. For a Compton gamma-ray beam, anaccurate and direct measurement of its energyin the tens to about 130 MeV region remains achallenge. One effective method to determine thegamma-beam energy is to measure the energy ofthe electron beam used in Compton scattering.A polarized electron beam can be used to accu-rately measure its centroid energy using the res-onant spin depolarization technique.

Nonlinear dynamics are a critical issue forthe light source storage rings as well as futurecircular accelerators with strong nonlinear mag-netic elements. Our recent work focuses on thestudy of nonlinear dynamics of charged particlesin a circular accelerator using scaling techniques.For example, we have studied the beam dynam-ics with varying strengths of nonlinear elementsin a circular accelerator with a fixed structure,or by varying the number of repetitive cells ofthe accelerator with each cell retaining the sim-ilar focusing arrangement. We have developed anew scaling law between the particle’s nonlineardynamics and the number of repetitive cells inthe circular accelerator. This so-called “lattice-scaling” allows us to compare beam dynamics ofdifferent storage rings of different sizes and beamemittances. It also allows us to predict the degra-dation of beam dynamics as the chromatic com-pensation is increased in a certain manner. Thescaling methods developed in this research areexpected to become a very useful technique instudying the nonlinear dynamics of charged par-ticles in the circular accelerators.

[Sun09a] C. Sun et al., Phys. Rev. ST Acccel.Beams, 12, 062801 (2009).

[Sun09b] C. Sun et al., Nucl. Instrum. MethodsA, 606, 312 (2009).

[Sun10] C. Sun et al., Nucl. Instrum. MethodsA, 614, 339 (2010).


8.1.5 Gamma Imagers for HIGS User Experiments

Y.K. Wu, C. Sun, M. Eamian, J. Li, W.Z. Wu, TUNL

Two gamma imagers have been developed and deployed in the UTR and GV experimental

areas for routine HIGS user operation. A gamma imager is a critical instrument for aligning

the collimator and experimental apparatus with the gamma-ray beam. A third gamma imager

in the SW optical room is being developed as a diagnostic to stabilize the gamma-beam

pointing and potentially as an integrated flux monitor.

The HIγS beam is produced by colliding ahigh intensity electron beam and a high powerFEL beam inside the FEL cavity. The highestenergy Compton scattered photons are concen-trated in a small solid angle along the directionof the electron beam. A collimator is used toselect a small portion of the gamma-ray beamwith a desirable energy resolution. To minimizethe energy spread and maximize the flux of thegamma beam, the alignment of the collimator iscritical for all nuclear physics experiments, espe-cially for experiments with a long target systemsuch as a gas cell. At the HIγS, we have devel-oped high-resolution gamma imagers to “see” thegamma-ray beam directly in order to align thecollimator with the gamma-ray beam, and alignthe experimental apparatus with the collimatedgamma-ray beam.

8.1.5.1 Gamma Imagers for HIGS Oper-ation

Since 2008, we have successfully developed threedifferent gamma imagers, with spatial resolutionbetter than 0.5 mm, and contrast sensitivity bet-ter than 5%. Two of the imagers, the UTR im-ager and GV imager, are installed in the UTRand GV experimental areas, respectively (see up-per panel of Fig. 8.7). These imagers are usedto align the gamma-ray beam with the collima-tor and experimental apparatus in the two targetrooms (lower panel of Fig. 8.7), by looking up-stream at the shadow of the collimator and exper-iment apparatus made in the gamma-ray beam.The UTR imager is located about 60 m down-stream from the collision point where the Comp-ton gamma-ray beam is produced, and has a 2 indiameter field of view. It is the newest imager,developed with a set of compact optics to mini-mize the use of valuable floorspace in the UTR.The GV imager is located about 70.2 m down-

stream from the collision point, and has a 3 indiameter field of view. The GV imager is a versa-tile system with reconfigurable optics to providedifferent magnifications and fields of view. Bothimagers are mounted on translational platforms,which provide movement of the imagers into andout of the gamma beam by remote control. Theremotely controlled translation stage for the GVimager has been operational since Jan, 2010 andthe remote control mechanism for the UTR im-ager is nearly completed.

Figure 8.7: (color online). Two gamma imagersdeployed in the UTR and GV areas.(a) Upper: the UTR gamma imagerwith a 2 in field of view; (b) Lower:the GV gamma imager with a 3 in fieldof view.


Figure 8.8: Locations of gamma imagers inside the UTR and GV experimental areas.

In Dec. 2009, we established a remote con-trol user interface to control CCD cameras in theUTR and GV imagers from the accelerator con-trol room. Since Jan. 2010, both the UTR andGV imagers have been used for all HIGS user ex-periments. When the misalignment is small, thegamma-ray beam is aligned with the collimatorby changing the direction of the electron beamat the collision point. Fig. 3 shows the gamma-beam images with the collimator before and afterperforming alignment.

GV

Y p

ositi

on (

mm

)

X position (mm)20 30 40 50

10

15

20

25

30

35

GV

Y p

ositi

on (

mm

)

X position (mm)20 30 40 50

10

15

20

25

30

35

UTR

Y p

ositi

on (

mm

)

X position (mm)20 30 40

10

15

20

25

30

35

UTR

Y p

ositi

on (

mm

)

X position (mm)20 30 40

10

15

20

25

30

35

Figure 8.9: (color online). Gamma-imagers areused to align the collimator with thegamma-ray beam. Upper images:gamma beam images taken using theUTR imager – left: misaligned; right:aligned. Lower images: gamma beamimages taken using the GV imager –left: misaligned; right: aligned. Forall images, a 15.5 MeV gamma beamproduced by a 625 MeV electron beamis imaged after a 19 mm (0.75 in) di-ameter lead collimator.

8.1.5.2 Development of SE Gamma Im-ager

A third gamma imager has been installed down-stream from an FEL mirror inside the southeast(SE) optical room. This system has been devel-oped for monitoring the gamma-beam centroidmotion and gamma-beam flux. By direct mon-itoring the gamma-ray beam centroid position,this system will allow us to improve the point-ing stability of the gamma-beam. This is criti-cal for many user experiments which require fre-quent changes of the gamma energy, resulting inchanges of the electron beam orbit around thecollision point. We are also investigating the ca-pability of this gamma-imager as a device formonitoring the integrated gamma-beam flux. Asan integrating flux monitor, it is expected to havea very wide dynamic range, selectable by chang-ing the integration time. However, we have tofind effective ways to deal with issues related tothe photon background, bad CCD pixels, andnonlinear response of the imager with the vary-ing gamma beam energy. We expect to reportour findings on this system in the next reportingperiod.


8.1.6 HIGS Accelerator and Gamma-ray Beam Operations

Y.K. Wu, P. Wang, P. Wallace, S.F. Mikhailov, TUNL

During the period from August 1, 2009 through July 31, 2010, the HIGS operated 2539 hours

for a variety of research programs and activities, and achieved a high level of reliability of

about 96%. A total of 1452 hours of on-target gamma-ray beamtime were delivered to various

HIγS user research programs.

8.1.6.1 HIγS Operation: Aug. 2009 –Jul. 2010

During the period from August 1, 2009 throughJuly 31, 2010, the HIGS accelerators operated204 days, providing 2539 hours for a variety ofresearch programs and activities. The acceler-ator and light source operation was carried outmainly with a two-shift, five-day operation sched-ule. Overall, operation of HIGS accelerators andlight sources was at a very high level of reliability,near 96%, exceeding the typical performance goalset for national synchrotron light sources (95%).This level of reliability was realized in spite of thecomplex and challenging operation of the free-electron lasers and Compton light source. A de-tailed break-down of accelerator operation hoursare summarized in Table 8.1 and Fig. 8.10.

In the last twelve months, we delivered 1452hours to research programs using the HIGSgamma-ray beam, for both basic and appliednuclear physics research. The accelerator re-search and development used 243 hours of beam-time. Some of this beamtime was spent on de-veloping future capabilities, including the devel-opment of new operation modes of the longitu-dinal feedback system to stabilize multi-bunchelectron beams and commissioning of a trans-verse feedback system. A substantial amount oftime (844 hours) was used to perform the setupand tune-up of accelerators, FELs, the Comptongamma source, and other activities. The gamma-ray user programs in this period demanded fre-quent changes of gamma-beam parameters, in-cluding gamma-beam energy, flux, and polariza-tion. These changes required frequent changesof the FEL mirrors. With each mirror change,time is spent on conditioning and aligning FELmirrors, and on setting up accelerators, FELs,and the Compton gamma-ray source. These 844hours also include the beamtime spent on thecommissioning of the OK-5 helicity switchyard

system, on the testing and commissioning of thenew UTR personal protection system (PPS), in-cluding radiation surveys around the UTR andalong the gamma beamline.

Table 8.1: Summary of DFELL accelerator andlight source beamtime from August 1,2009 to July 31, 2010. The commis-sioning time of the UTR experimentalarea has been included in the time foraccelerator setup and tuning.

Activities Beamtime(hrs) %

HIγS user research 1, 452 54.9%Acc. R&D 243 9.2%Acc./FEL/HIγS

setup and tune 844 31.9%Unscheduled downtime 108 4.1%

Total scheduled beamtime 2647 100%

Acc R&D, 9.2%

Acc&FEL&HIGS Setup/Tune, 32%

HIGS User Research, 55%

Unscheduled Downtime, 4.1%

Figure 8.10: (color online). Beamtime distribu-tion for various research programsand accelerator activities from Aug.1, 2009 to Jul. 31, 2010.

A monthly distribution of HIγS user beam-time is shown in Fig. 8.11. Due to the instal-lation of the UTR, there was no HIGS opera-tion for three and half months from the middleof September, 2010 to the early part of January,2010. The delivered monthly gamma-ray beam-


time in the remaining 8.5 months is about 170hours per month. In August 2009 and July 2010,week-long three-shift operations were carried outcontinuously for five days and four nights. Thisincreased the HIGS beamtime on target in thesetwo months to 205 and 245 hours, respectively.During the UTR shutdown, accelerators were runmainly for accelerator and FEL research withlimited staff support. About 71% of acceleratorand FEL research beamtime (172 hr) was pro-duced from Oct. to Dec. 2009 during this shut-down period.

0

50

100

150

200

250

HIγ

S b

eam

time

(hr)

Aug. 2

009

Sep. 2

009

Oct. 2

009

Nov. 2

009

Dec. 2

009

Jan.

201

0

Feb. 2

010

Mar

. 201

0

Apr. 2

010

May

201

0

Jun.

201

0

Jul. 2

010

Figure 8.11: (color online). Monthly distributionof gamma-ray beamtime hours deliv-ered to the HIγS experiments (Aug.,2009 – Jul., 2010).

Among the total scheduled 2647 acceleratoroperation hours, the unscheduled downtime wasabout 4.1%, or 108 hours. In the last twelvemonths, most of the accelerator failures can becategorized in the four groups: (1) the failureof the cathode in the electron gun; (2) problemswith ring kickers; (3) problems with secondaryDC power supplies; and (4) failures of the RFsystem. The longest unscheduled downtime wasdue to the double failure of the electron gun cath-ode in March 2010 after only eight months ofoperation. Two days were used to replace thecathode, condition the electron gun, and recoverthe accelerator operation. Five days later thisnewly installed cathode failed, and three moredays were used to install another and to performa much more conservative program of thermaland RF conditioning. The most frequently en-countered accelerator problems are the stabilityof kicker systems of the storage ring. These area new generation of fast pulse systems, and thekicker drivers need frequent tuning to keep themnear the optimal operation condition. Also, inthis period, high voltage protection componentshave been added to the kicker thyratron tubes toprotect these from over-voltage during the firingof the kickers, and these protection componentshave a finite lifetime, the length of which we areonly coming to have a measure of now. With con-tinued improvements to the kicker systems, the

kicker stability problems are expected to be lessfrequent in the coming year.

Due to unusually high temperatures in Juneand July, 2010, we suffered from a number ofbeam instability triggered beam losses after a RFtrip. This problem seems to occur in the after-noon when the outside ambient temperature isrising fast to above 95 F (32 C). In some cases,the use of the longitudinal bunch-by-bunch feed-back seemed to be able to stabilize electron beaminstabilities, thus allowing continued HIGS op-eration. We are also investigating whether theRF system has a reduced capability dealing withelectron beam induced oscillations due to agingof its amplifier tubes and malfunctions of its feed-back loops. In July 2010, a combination of a highambient temperature and an unusually intensiveHIGS operation schedule resulted in 12% of un-expected downtime while producing the largestamount of beamtime on target of 245 hours.

The overall high reliability can be attributedto a successful preventive maintenance program.Over the years, we have been able to establish amaintenance schedule for critical hardware sys-tems using the projected mean time to failure.In addition, we continuously monitor hardwaresystems for any sign of degraded performance.A checkup will then be scheduled for the systemin question. Ideally, the maintenance activitieswould be performed every two to three weeks.Fewer frequent mirror changes in the later partof the reporting period have reduced our flexibil-ity to schedule maintenance tasks.

In the long term, the reliability of acceleratorsand light sources will depend on the availabilityof spare hardware in house. The timely procure-ment of expensive spare hardware will remain achallenge as we push to operate the acceleratorswith higher energy and beam current, and forlonger durations to deliver more user beamtime.

8.1.6.2 Near-future HIγS Operation

In the next 12- to 18-month period, a major shut-down is expected, to install the wiggler switch-yard system. The total duration of this shutdownis expected to be on the order of 12 weeks. Toincrease the number of beam hours for HIGS re-search programs, we plan to take advantage offixed energy operation for some nuclear physicsexperiments, and to schedule some weeks with24-hour operation. Because of the very limitedstaffing level at the HIGS, we can only carry outa few weeks of 24-hr operations in a year. Thistype of operation is expected to reduce the over-all reliability of the accelerator operation, as wedo not have the staff to perform quick repairs ofminor hardware problems during the second- andthird-shift operations.


8.2 The FN Tandem Accelerator and Ion Sources

8.2.1 Tandem Accelerator Operation

C.R. Westerfeldt, J.H. Addison, E.P. Carter, J.D. Dunham, R. O’Quinn, TUNL

8.2.1.1 Tandem Operation

The TUNL FN tandem accelerator was operated198 days for 2163 hours at terminal potentialsranging from 0.2 MV to 8.7 MV during the pe-riod August 1, 2009 to July 30, 2010. Beamsaccelerated during this period include unpolar-ized protons and deuterons and also 4He, Theterminal operating potential during the report-ing period is shown graphically in Fig. 8.12.

An opening of the tandem was made onMarch 30 for routine maintenance and to repairseveral pressure leaks of tank insulating gas atthe high-energy end of the tank. No bad col-umn resistors were found at this opening. Wediscovered that the high-energy charging chainwas failing however. The chain was extremelyold having been used for about ten years. Whileit could have been repaired, it was decided to in-stall a new chain that was purchased in 2008 asa spare. This new chain was discovered to be tooshort to install despite having the same numberof links. This was due to the worn bushings in thefailing chain. Two additional links were installedto permit the chain to be joined. The chain wasrun and the stretch monitored. We quickly hadto remove one link as the chain stretched rapidly.We consulted with the manufacturer (NEC) andthey advised that what we were seeing was nor-mal. We removed two more links and contin-ued to run the chain. The stretching seemed tostop so the tank was closed on April 16 and op-erations resumed on April 17. On March 3, thechain stretch interlock on the high-energy chaintripped and an emergency entry was made. Oneadditional link was removed and it was discoveredthat the tension on the high-energy chain was settoo high. One weight was removed and the in-terlock was adjusted as it seemed to have trippedprematurely. The tank was closed on May 4 andoperations resumed May 5. No further problemshave occurred with this chain. We will plan onpurchasing a spare chain for the low-energy end

as it is of the same age and starting to show signsof deterioration as well.

During the first opening, we made measure-ments needed to design a mount for the new mid-plane idler pulley that we have constructed. It ishoped that it will be installed during the spring2011 opening. Plans are to test it in air witha spare chain and pulley system that we haveon hand before installing it in the tandem. Thetank was closed on April 16th, 2010. On May3, the high-energy chain break interlock openedand a quick entry was made to shorten the newchain and reduce the chain tension. The tank wasclosed again on May 4 and operations resumed.

8.2.1.2 Laboratory Projects

The Direct Extraction Negative Ion Source wasoverhauled mechanically and electronically dur-ing the year to improve its output for new classesof experiments that require more beam than wewere presently getting. The extraction geome-try was modified by changing some of the inter-electrode gaps. A new extraction aperture wasmanufactured from solid Molybdenum rod to re-place the pressed in insert and stainless steelholder that had been previously used. It wasthought that this was causing problems in thatthe Molybdenum has a much lower thermal ex-pansion coefficient than the stainlesssSteel whichit was pressed into and so became loose whenheated by the arc. We also noticed that the arcoften struck to the outside edge of the insert sowe desired to eliminate that interface. With thesechanges, analyzed negative proton and deuteronbeam with intensities of 30 plus microamperesare now available for acceleration at the low en-ergy araday cup. A time dependent instability re-mains, however it is not significant enough affectexperiments.

The cooling system for the shielded sourceexperimental area was rebuilt following repeatedfailures with this system. Previously it required


Figure 8.12: (Color online) The TUNL FN Tandem Operating Potential.

manual filling with distilled water periodically.The new system is completely automatic requir-ing the researcher only to turn it on and off. Vis-ible flow indication and quick disconnect fittingswere installed near the water cooled neutron pro-duction gas cell to verify flow and to provide tool-free connection and disconnection of the waterlines for rapid exchange of gas cells which maybe activated.

8.2.1.3 Shielding Changes

In order to reduce the neutron radiation exteriorto the building in the vicinity of the NTOF room,additional concrete shielding was installed alongtwo walls inside the NTOF room. This neces-sitated the relocation of the fume hood in thisroom and also the Tritium stack monitor.Initialsurveys confirm the effectiveness of this change.

The entry maze to the 20-70 magnet vaultwas reconfigured in June by repositioning twolarge 6 ton water tanks. Two improvements were

accomplished: increasing the shielding betweenthe tandem high-energy faraday cup and the con-trol room, and blocking a direct line-of-sight be-tween the 20-70 magnet and the low energy bayentrance area. Subsequent surveys and environ-mental dosimetry indicate that the desired re-duction in the radiation in these areas was ac-complished.

8.2.1.4 Pneumatic Transfer System

A contract has been signed to install a 4 pneu-matic transfer system between the NTOF roomin the TUNL building and the growth chamberarea in the adjacent Phytotron building. Thiswill be installed in the fall of 2010 and will beused to transport short half-live isotopes from theproduction cell in TUNL to the plant researchstation in the Phytotron The capsule is expectedto have a transit time of approximately 30 sec-onds which is critical for 13N which has only a9.9 minute half-life.


8.3 The LENA Accelerator and Ion Sources

8.3.1 Status of the LENA ECR Accelerator

J.M. Cesaratto, A.E. Champagne, T.B. Clegg, TUNL

We have begun using a new electron-cyclotron-resonance (ECR) ion source and acceleration

system for experiments at the Laboratory for Experimental Nuclear Astrophysics (LENA).

The LENA ECR ion source is capable of producing >1 mA proton beam on target for energy

ranges 90 to 200 keV. In the past year, steps have been taken to improve the stability of the

acceleration system. Two low-energy resonances of the reactions 18O(p,γ)19F and 27Al(p,γ)28Si

were measured to test the high voltage acceleration system. Recently, a manuscript was

submitted for publication describing LENA’s ECR ion source and accelerators.

The electron-cyclotron-resonance (ECR) ionsource at LENA operates with input rf power of100 to 500 W at 2.45 GHz. A cylindrical ar-ray of NdFeB permanent magnets produces theaxial 87.5 mT field necessary for the electron-cyclotron resonance. The H+ beam is extractedfrom a plasma chamber in this magnetic fieldby an “accel-decel” extraction lens system andis accelerated to the target with energies up to215 keV. A full description of the LENA ECRion source has been submitted for publication[Ces10].

8.3.1.1 New control instrumentation

The ECR source has been run extensively overthe past year, and this experience has revealedseveral shortcomings, which we have addressed.One was in accelerator control and metering. Theexisting set of control and metering instrumentswere used with the new source but were foundto be unsuitable. In particular, very poor sparkprotection, subtle connections to the JN-Van-de-Graaff accelerator control system, and numer-ous ground loops were noted. These issues werebrought to our attention by two separate in-cidents (September 2008 and January 2009) inwhich the high-voltage platform discharged toground. The surge found its way back into thecontrol equipment and damaged eight NationalInstruments FieldPoint modules, numerous oper-ational amplifiers in the terminal-potential stabi-lizer circuit and corona-current metering circuitfor the JN, a pulse-width modulating chip in theupcharge supply of the JN, and a diode in thevacuum-panel-gate-valve logic.

To prevent this from recurring, careful stepswere taken to install an entirely new control andmetering system solely for the ECR accelerator.This task included adding a new fiber optic com-munication line, Compact FieldPoint modules,shielded cables, and conduit. New control equip-ment was housed in a shielded cabinet, whereconduits provided the path for shielded cablesfrom the cabinet to the associated high voltageequipment. All ground cables were tied to onepoint, and care was taken to eliminate groundloops. Homemade surge suppression circuits wereadded to all control and metering signals on thisnew system. As an extra measure, we added asurveillance and digital video recorder system tohelp in locating any source of future discharge. Inthe months since employing the new control sys-tem, we have not had any incidents of dischargescausing further damage.

8.3.1.2 Accelerator characterization

The quality of beam from the ECR acceleratorwas evaluated by measuring two low-energy res-onances. One was the well-known ER = 150.82±0.09 keV resonance [Bec95] in the 18O(p,γ)19F re-action. It corresponds to Γlab ≤ 0.5 keV [Vog90]and has Qpγ = 7994.8 ± 0.6 keV [Aud03]. Thesecond resonance used was at ER = 202.8 ± 0.9keV [End98] in the 27Al(p,γ)28Si reaction and hasQpγ = 11585.11± 0.12 keV [Aud03]).

As a check of the voltages read out from theacceleration supplies and our subsequent knowl-edge of the energy of the beam, excitation func-tions (i.e., γ-ray yield versus bombarding energy)for both resonances were measured in order to de-


termine both resonance energy and energy reso-lution. Values measured for the 18O(p,γ)19F res-onance were ER = 150.6± 0.1 keV, with a corre-sponding energy resolution of 0.84 ± 0.07 keV atfwhm. The resonance energy in the 27Al(p,γ)28Sireaction was ER = 200.9 ± 0.1 keV, with a cor-responding energy resolution of 1.44 ± 0.17 keV.Thus small corrections must be made in orderto normalize the energies determined from thepower supply to the literature values. The mea-sured energy widths are a factor of 2.8 to 3.6greater than what is expected from the ripple andstability specifications of the power supplies (bet-ter than 0.1% and 0.01%, respectively, [Gla07]for the 200 kV supply). However, we cannot ruleout some contribution from contaminants or non-uniformities in the targets.

8.3.1.3 Future improvement

The microwave supply which drives the ECR ionsource plasma is quite unstable; its output powercan vary ±30 W at 250 W running power. This,in turn, changes the driven plasma, which canchange the plasma meniscus at the beam extrac-tion point. Unfortunately, this can modify thebeam focussing and change the size and shape ofthe beam. Optical parameters can be tuned totry to compensate for this, but such problematic

retuning is not reproducible. In the next year,we would like to purchase a new, well-regulatedmicrowave power supply capable of producing 1kW output. This, would improve the reliabilityof the source, and the increased power capacityshould boost beam output to what is needed dur-ing pulsed beam operation.

[Aud03] G. Audi, A. H. Wapstra, and C.Thibault, Nuclear Physics A, 729, 337(2003).

[Bec95] H. W. Becker et al., Z. Phys. A, 351,453 (1995).

[Ces10] J. M. Cesaratto et al., Nucl. Instrum.Methods A, (July 2010), submitted forpublication.

[End98] P. M. Endt, Nucl. Phys. A, 633, 1(1998).

[Gla07] Glassman High Voltage Inc., 124 WestMain Street, PO Box 317 High Bridge,NJ 08829, Instruction Manual PK Se-ries, 2007.

[Vog90] R. B. Vogelaar et al., Phys. Rev. C, 42,753 (1990).

Nuclear Instrumentation andMethods

Chapter 9

• Targets

• Detector Development and Characterization

• Facilities

• Data Acquisition Hardware and Software Development

178 Nuclear Instrumentation and Methods TUNL XLVIII 2009–10

9.1 Targets

9.1.1 A New 3He-Target Design for Compton Scattering Experiments

H. Gao, M. Huang, G. Laskaris, Q. Ye, Q.J. Ye, W. Zheng, Y. Zhang TUNL

Compton scattering from polarized 3He is designed to measure neutron spin polarizabilities

by using a circularly polarized γ-ray beam at HIγS. A uniform magnetic field is required to

maintain the 3He spin polarization in the longitudinal direction. We are planning to use a

single-layer solenoid instead of Helmholtz coils for providing the field. We are also planning

to add a pair of tungsten collimators for shielding against the background generated from the

target windows. A new 3He target system is designed to accomplish these tasks easily.

Nucleon spin polarizabilities describe thestiffness of the nucleon spin to external electricand magnetic fields. Compared to the nucleonelectric and magnetic polarizabilities, very littleis known about nucleon spin polarizabilities. Theonly quantities that have been determined ex-perimentally are the forward and backward spinpolarizabilities, γ0 and γπ. Recently, double-polarized elastic Compton Scattering from po-larized 3He has been investigated and sensitiv-ity to the neutron spin polarizabilities (γ1...γ4)has been demonstrated [Cho07]. The double-polarized Compton experiment requires circu-larly polarized γ rays and a polarized 3He target.We need a uniform magnetic field along the cylin-drical axis of the target chamber for maintainingthe 3He polarization. In previous polarized 3Heexperiments, a pair of Helmholtz coils was used.However, the Helmholtz coils will not be usedin this experiment so that we can position theNaI detectors near the polarized 3He target. In-stead, a single layer solenoid is being developedto provide the holding field for the polarized 3Hetarget.

The solenoid is designed in a single-layer con-figuration and wound around a 106 cm long PVCpipe (outer diameter = 40.64 cm) by using 1.02-mm diameter copper wire (AWG 18). There are1000 turns in the solenoid, and the current ap-plied is 1 A. The coil-winding system is estab-lished in the machine shop of the Duke Physicsdepartment. After the winding, we applied epoxyto glue the wires together. We also used Kap-ton tape to protect the wires. The single-layersolenoid is shown in Fig. 9.1.

For mapping the magnetic field (Bx,By,Bz)inside the solenoid, a three-axis magnetic field

mapper is designed. It uses a motor-system, aGaussmeter, two different probes (one for Bz , theother for Bx and By), and the data-acquisitionsystem. Details are given in Ref. [Ye110]. Theprobes are placed on a non-magnetic arm thatis capable of incremental three-axis motion usinga three-axis motor. With the motor, the probescan scan the different points in the space in stepsof 1.27 cm. Inside the solenoid, the cylindricalaxis of the target chamber is not the same as theaxis of the solenoid; the offset is about 6.12 cm.A cuboid which includes the target chamber andpumping chamber is scanned as the region of in-terest (ROI). The whole system is controlled bya PC that can store the magnetic-field value vs.position (x,y,z) for future analysis.

Figure 9.1: (Color online) A single-layer solenoidfor Compton scattering experiments

We simulated the solenoid magnetic field byusing the code comsol and compared the re-sults with the field-mapping results. Our mea-surements show consistency with the simulationresults (except for some constant offsets). More

TUNL XLIX 2009–10 Nuclear Instrumentation and Methods 179

details are given in Ref. [Ye110]. The measuredmagnetic-field values in the ROI are shown in Fig.9.2; the units are in inches. The results show theoutstanding uniformity of the solenoid magneticfield in the ROI.

-10

-5

0

5

10

-10

-5

0

5

10

-10

-5

0

5

10

y

x

z

y

11.0

11.1

11.1

11.2

11.2

11.3

11.3

11.4

11.4

11.5

Figure 9.2: (Color online) Magnetic field mappingof the ROI. The B-field is given inGauss. Details are in Ref. [Ye110]

During the experiment, a high polarization3He target is important, and an inhomogeneousmagnetic field can contribute to the depolariza-tion of polarized 3He according to the relation[Cat88]:

1T1

= D|∇Bx|2 + |∇By|2

B2z

. (9.1)

Here 1T1

is the relaxation rate and D is the self-diffusion coefficient, here assumed to be D = 0.28cm2/s. The quantities ∇Bx and ∇By are thegradients of the transverse field components, andBz is the magnitude of the holding field. Fromour field-mapping result in the region of interest,|∇Bx|2 + |∇By|2 = 80.41 (mG/cm)2. For a hold-ing field of Bz = 11.3 G, we obtain a value ofT1 = 1493 hr. Compared with other effects, suchas dipole-dipole collisions and interactions withglass wall, depolarization effects from magneticfield gradients is negligible.

A pair of tungsten collimators will be em-ployed in the experiment. The goal of the targetcollimators is to shield against photons generatedfrom the target windows. These are consideredas the major background contribution to elasticCompton-scattering events. Each collimator is a9.5-cm-long tube with an outer diameter of 7.2cm (2.83 in.) and an inner diameter of 3.2 cm(1.26 in.). The inner diameter is 0.2 cm largerthan the diameter of the target chamber, so thatwe can move the collimators without scratchingthe cells.

Considering the limited space inside thesolenoid, a careful and feasible design of the tar-get system is being developed. Figure 9.3 shows

the conceptual design of the 3He target system.The support structure for the target system is notshown here since it depends on the design of thetable for the HIγS NaI Detector Array (HINDA).During the experiment, a 3He target cell and aN2 reference cell (a cell used for background mea-surement, having the same dimensions as the tar-get cell but filled with N2 instead of 3He) canmove up and down together with the solenoid,in and out of the γ-ray beam, using a motor-controlled support.

Figure 9.3: (Color online) A conceptual design of

the 3He target system for Compton-scattering experiments

A combination of rails and ball bearings areused to move each collimator along the target-chamber axis in the horizontal plane by using an-other motor system. In this way, we can switchbetween two different sets of target cell + colli-mators and reference cell + collimators to studybackground effects. The possible movements arelabeled as arrows in Fig. 9.3. In order to makethe alignment simple, we are planning to alignthe target cell, reference cell, and collimators out-side the solenoid with lasers. After the align-ment, the solenoid is moved to the designed po-sition with the target system inside, instead ofmoving the aligned cells and collimators.

Our next step is to build the target system,including the tungsten collimators, the motor-controlled support, and the support for thesolenoid. Once we set up the new target system,we will test it in our laboratory.

[Cat88] G. D. Cates et al., Phys. Rev. A, 37,2877 (1988).

[Cho07] D. Choudhury et al., Phys. Rev. Lett.,98, 232303 (2007).

[Ye110] Q. J. Ye, In www.phy.duke.edu/∼qy4,2010.


9.1.2 Scintillating Target for d(�γ, γ)d at HIγS

P.P. Martel, R. Miskimen, A. Teymurazyan, University of Massachusetts, Amherst, MA; S.S.

Henshaw, M.W. Ahmed, N.J. Brown, B.A. Perdue, S. Stave, H.R. Weller, TUNL

We plan to use the HIγS facility to extract the neutron’s electric and magnetic polarizabilities,

αn and βn respectively, using Compton scattering of 65-80 MeV γ rays from a scintillating

deuterium target. Here we discuss the progress being made on the scintillating target for this

experiment.

The electric and magnetic polarizabilities,αE1 and βM1 respectively, provide informationon the internal structure of the nucleon. The po-larizabilities are given by the effective interaction[Pas07]

H(2)eff = −4π

[12αE1

�E2 +12βM1

�H2

]. (9.2)

For the proton, the Particle Data Group atLawrence Berkeley National Laboratory givescurrent averages for αp and βp determinedthrough Compton-scattering experiments with aproton target:

αp = (12.0 ± 0.6) × 10−4 fm3 (9.3)βp = (1.9 ± 0.5) × 10−4 fm3. (9.4)

The best current fit (using a Baldin sum ruleconstraint) to world data gives neutron polariz-ability values of [Hil05]

αn = (11.6 ± 1.5 ± 0.6) × 10−4 fm3 (9.5)βn = (3.6 ∓ 1.5 ± 0.6) × 10−4 fm3 (9.6)

where the first quoted error is statistical, and thesecond is from the Baldin constraint. Some of thedata from which these were obtained come fromelastic Compton scattering off a deuteron target,γd → γd [Luc94, Lun03, Hor00].

We plan to complement the previous exper-iments by adding a deuterated liquid scintilla-tor (NE-232, C6D12 with 6.58× 1022 D/cm3 and3.36 × 1022 C/cm3) contained in a small glasscell with dimensions of 4.18 cm diameter by 3.7cm long. In a scattering event, the recoil of thedeuteron produces scintillation light in the tar-get, allowing us to tag the γ ray that elasticallyscattered from the deuteron.

Figure 9.4: (Color online) HIγS NaI Detector Ar-ray with scintillating target

The scattered γ rays will be detected in the8-element HIγS NaI Detector Array (HINDA).The eight elements of HINDA can be placed 35◦

apart (between centers). Starting on beam rightat 150◦, three other detectors will be set at an-gles of 115◦, 80◦, and 45◦. The remaining fourdetectors would be placed at the same angles onbeam left, producing the setup shown in Fig. 9.4.

Figure 9.5: (Color online) d(n,n)d setup with thethree neutron detectors at 15◦, 25◦,and 35◦

The response of the scintillating detector wastested by placing it in a neutron beam at TUNL,


Neutron E (MeV)0 2 4 6 8 10 12 14

Co

un

ts

0

100

200

300

400

500

15 deg

Neutron E (MeV)0 2 4 6 8 10 12 14

Co

un

ts

0

200

400

600

800

1000

25 deg

Neutron E (MeV)0 2 4 6 8 10 12 14

Co

un

ts

0

200

400

600

800

1000

1200

35 deg

Pulse Height (MeVee)0 0.05 0.1 0.15 0.2

Co

un

ts

0

200

400

600

800

1000

1200

1400


Co

un

ts

0

200

400

600

800

1000

1200

1400


Co

un

ts

0

200

400

600

800

1000

1200

1400

Figure 9.6: (Color online) Results for d(n,n)d at an incident neutron energy of 8.525 MeV, with thethree neutron detectors at 15◦, 25◦, and 35◦, corresponding to scattered neutron energiesof 8.24, 7.75, and 7.08 MeV respectively (top), and recoil deuteron energies of 0.29, 0.78,and 1.45 MeV respectively (bottom, shown in MeV electron equivalent, or MeVee, toaccount for the scintillator response to deuterons). Each panel displays the spectra uncut(black), cut on PSD (red/light gray), and cut on PSD and TOF (blue/dark gray). Thedashed vertical lines indicate the expected values.

and looking at the reaction d(n,n)d. Three neu-tron detectors were arranged at various angles(as shown in Fig. 9.5), and a coincidence triggerwas established between the scintillating targetdetector and a neutron detector.

By properly choosing the angles of the neu-tron detectors and the energy of the incident neu-tron, the energy of the recoil deuteron is equiva-lent to the energy of a recoil deuteron in the pro-posed Compton-scattering experiment at HIγS.For the lower beam energy requested for theCompton experiment, 65 MeV, the γ rays scat-tered into the most forward, 45◦, HINDA de-tector correspond to a recoil deuteron energy of0.653 MeV. Therefore, the goal for this test wasto observe recoils in this energy region. Sampleresults from this test are shown in Fig. 9.6.

A test was then performed at HIγS usingthe same scintillating detector but in coincidencewith a HINDA detector. Due to the high fluxesavailable, the target experiences a high rate ofatomic Compton scattering and deuteron photo-disintegration. Previous tests had resulted in the

tube experiencing a form of saturation from aninsufficient “relaxation” time. The bias on thetubes was increased to reproduce this same prob-lem, confirming our suspicions of its cause. How-ever, when run at the level needed to observe thedesired events (as determined through the test atTUNL), the system performance was quite satis-factory.

[Hil05] R. Hildebrandt et al., 2005, Arxivpreprint nucl-th/0512063.

[Hor00] D. Hornidge et al., Phy. Rev. Lett., 84,2334 (2000).

[Luc94] M. Lucas, Ph.D. thesis, University ofIllinois, 1994.

[Lun03] M. Lundin et al., Phy. Rev. Lett., 90,192501 (2003).

[Pas07] B. Pasquini et al., Phy. Rev. C, C76,015203 (2007).


9.1.3 Status of the HIγS Frozen-Spin-Target Transverse Coil at TUNL

P.-N. Seo, H.R. Weller, TUNL; D.G. Crabb, University of Virginia, Charlottesville, VA; R.

Miskimen, University of Massachusetts, Amherst, MA; M. Seely Meyer Tool of Manufacturing,Inc., Oak Lawn, IL

The status of the HIγS Frozen-Spin-Target Transverse Coil at TUNL is reported.

We describe the design, construction, andperformance of a set of saddle coils used to main-tain the spin of a polarized proton target in thetransverse direction with respect to the incidentγ-ray beam direction. The transverse coil assem-bly consists of two racetrack shaped coils formedfrom a single strand of thin NbTi superconduct-ing wire. Four layers of NbTi per saddle coilwere wet-wound in a racetrack shape. They werethen installed on a cylindrical support tube andepoxyed to prevent them from moving when the

coils are energized. As expected from our modelcalculation, a set of two saddle coils produced0.36 T in the middle of the coils with a 25-A cur-rent at 4 K, which is 63% of the critical current ofthe wire. The measured homogeneity of the fieldover the target volume has a maximum variationfrom the average value of 0.6% [Seo10].

[Seo10] P.-N. Seo et al., Nucl. Instrum. Methods,A618, 43 (2010).


9.2 Detector Development and Characterization

9.2.1 A Detector Array for Photoneutron Studies

J.R. Tompkins, M.W. Ahmed, N. Brown, B. Davis, M. Emamian, J.M. Mueller, S. Stave,H.R. Weller, TUNL; C. Smith, George Washington University, Washington, DC ;

A new detector array for photoneutron studies consists of 18 TUNL Bicron-type neutron

detectors oriented parallel and perpendicular to the plane of γ-ray polarization. Two frames

have been designed to be the backbone of this array, facilitating detector placements at

distances of 1 m and 0.5 m from the target.

The capability of HIγS to produce 100%linearly-polarized γ-ray beams has made it pos-sible to study photoneutron polarization asym-metries. Recent experiments to study photofis-sion and (γ,n) reactions (see Sect. 7.1.1) have re-quired the ability to place up to 18 TUNL Bicron-type neutron detectors [Tro08] both in and outof the plane of polarization, which is parallel tothe floor. To accommodate the experimental re-quirements, two new frames have been designedfor positioning the detectors at distances ∼0.5and ∼1.0 m from the target.

Figure 9.7: (Color online) The 0.5 m frame.

The array was designed to create a free flightpath for neutrons traveling from a target to de-tectors mounted parallel and perpendicular tothe plane of polarization. Further design princi-ples were portability and the ease and flexibilityof detector placement. These were accomplishedwith a modular design in which detectors are heldin adjustable cradles that clamp to a sturdy alu-minum frame.

Each frame was constructed by attaching tworings with equal diameters at a right angle (seeFig. 9.7). The rings join so that each is offsetfrom the beam by about 30 cm to allow room forthe detectors. Each of the rings is constructedfrom rectangular aluminum tubing. These werebent to half circles of 1.22 m and 0.71 m outerradii and then joined together to form the com-plete rings. The vertical ring on the 1 m framecontains two straight sections to account for thefact that it mounts to either end of a chord acrossthe horizontal ring (see Fig. 9.8). The 0.5 mframe does not contain this feature due to animproved design. Both frames are supported byextruded aluminum legs.

Table 9.1: Technical comparison of the 1 m and0.5 m setups. Nominal values are givenfor the packing angles and capacities.

Specifications Frame1 m 0.5 m

Flight path (m) 1.0 ± 0.1 0.5 ± 0.1Packing angle (◦) 9 17Capacity (# dets) ≈ 36 ≈ 22Footprint (m2) 4.7 1.6

The use of extruded aluminum increases the ver-satility of the detector array because it is com-mercially available and can easily be appended


to without custom machining. Beyond the di-mensions of the frames, there are no differencesbetween the two setups. For technical compar-isons of the two setups, refer to Table 9.1.

Figure 9.8: (Color online) The 1 m frame.

The detector cradles that clamp to the standare designed to provide the adjustment necessaryfor aligning the detectors. There are three adjust-ments, fore-aft, up-down, and yaw. Twenty fourcradles have been assembled for the 18 detectorsalready existent and in anticipation of 6 more.

The only material that intrudes upon the neu-tron flight paths is the target stand. Because ofthis, the stand was designed to be lightweight. It

is composed of three sections: a horizontal tar-get ring, a target arm, and interchangeable tar-get holders that attach to the arm. The 45.7cm diameter target ring shares the same symme-try axis as the horizontal ring of the frame and isheld in place by three vertical extruded aluminumbars that provide vertical adjustment. The tar-get arm is clamped to the target ring. It is avertical 5/16 inch (0.79 cm) threaded rod with alightweight aluminum joint that holds a 0.64 cmdiameter aluminum rod at 45◦. The aluminumrod can be extended, and various adaptors canbe screwed into the end to hold targets. This isessential, because target shapes are non-uniform,and modifications to them are not always permit-ted. The orientation of the target is determinedusing the joint in the arm and the positioning ofthe arm on the ring.

The array is designed to hold the TUNLBicron-type detectors. The cradles fit these de-tectors and their corresponding photomultipliertubes. Six new detectors intended for use inthe array have been constructed and have beencharacterized with a 137Cs and Am-Be neutronsource. The identification numbers 18 through23 have also been assigned to them.

The array has successfully been commissionedin different experiments with both frames. Dur-ing the next year, it is likely that small improve-ments to the array will be made and that moredetectors will be added.

[Tro08] D. Trotter et al., Nucl. Instrum. Meth-ods A, 522, 234 (2008).


9.2.2 BrilLanCe Detector Energy Resolution Characterization at HIγS

N.J. Brown, M.W. Ahmed, S. Stave, S.S. Henshaw, B.A. Perdue, P.-N. Seo, H.R. Weller,C. Sun, Y.K. Wu, TUNL; P.P. Martel, A. Teymurazyan, University of Massachusetts, Amherst,MA; A. Owens, F. Quarati, European Space Research and Technology Centre, Noordwijk, TheNetherlands

Lanthanum halide crystals represent a major step in finding a room temperature photon

detector having high efficiency plus good timing and energy resolution over a wide range of

energies. We have studied the performance of these detectors between 2.5 and 15.5 MeV

using γ-ray beams produced by the HI�γS facility. Gaussian fits to the BrilLanCeTM detector

spectra were corrected for the beam energy spread to obtain the detector resolution. A

10.16 cm (diameter)x 15.24 cm(long) LaCl3(Ce) detector and a 7.62 cm x 7.62 cm LaBr3(Ce)

detector were characterized in the present study. The energy resolution of each detector is

reported as a function of incident γ-ray energy from 2.5 to 15.5 MeV. In addition, the response

of these detectors to thermal neutrons is described.

The recent discovery [vL00, vL01] of lan-thanum halide (LaX3:Ce) crystals represents amajor step towards finding a photon detec-tor that has high efficiency, good energy andtiming resolution, operates at room temper-ature, and can be used over a wide rangeof energies. These detectors are presentlybeing marketed under the trademark namesof BrilLanCe 350TM(LaCl3(Ce)) and BriLanCe380TM(LaBr3(Ce)).

The energy resolution of the γ-ray beam atHIγS [Wel09] depends on the diameter of the col-limator in place during experimental operation.The 12 mm collimator used during this experi-ment yields an energy spread ranging from 50 to250 keV for γ-ray energies from 2.5 to 15.5 MeV.The maximum delivered intensity of the HIγS fa-cility in the energy range of 2-16 MeV is over 107

γ rays per second, so attenuators were used tomoderate the beam for the measurements of thepresent work.

The full-energy peak response of the B350and B380 detectors were fit with a Gaussian line-shape. The fit yields a full width at half maxi-mum for the full-energy peak with contributionsfrom both the detector and the beam. The en-ergy resolution of the beam was determined usinga 120% HPGe detector. The resolution of the de-tectors was then obtained by correcting the mea-sured spectra for the beam energy spread. TheB350 (LaCl3Ce) and B380 (LaBr3Ce) results aresimilar, as seen in Figs. 9.9 and 9.10. Also in-cluded in the figures are dotted lines representing

an A√E

+ B curve. The coefficients for the curvesare A=76.3 and B=1.04 in Fig. 9.10, and A=76.3and B=0.49 in Fig. 9.10.

The presence of thermal neutrons leads to35Cl(n,γ) and 79Br(n,γ) reaction products in thecase of the B350 and B380 detectors, respectively,as well as the 140La(n,γ) products for both detec-tors. We placed a 200 mC Americium-Berylliumfast neutron and gamma-ray source in a 50.80 cmwide paraffin cube, in order to thermalize the fastneutrons. One meter away, at the same heightas the source, a 6Li-Glass detector stood behind10.16 cm of lead shielding. The lead is required tosuppress the γ-rays from the source by a factor ofabout 100. The 6Li-Glass detector, which detects(thermal) neutrons by means of the 6Li(n,α)3Hreaction, has an efficiency of essentially 100% forthermal neutron detection. The measured countrate indicated that there were 8.5±.5 neutronsper second impinging on the 5.08 cm diameter,0.9 cm thick Li-glass detector. This serves todefine our thermal neutron environment, whichwas identical for the 6Li-Glass, the B350 and theB380 detectors.

The 35Cl (n,γ) transition intensities reportedin ENSDF are shown in Fig. 9.11. Both the fullenergy peak and first escape peak for each rel-evant transition are shown. They are normal-ized to the 6.1 MeV peak, the strongest intensitypeak in the reported work [End98], having 77 outof 100 units. Using this scale, all the intensitiesof the included transitions are depicted by theheights of the vertical dashed lines. It can be


seen that the observed peaks are well correlatedwith the γ-ray lines, as expected.

The highest energy γ-ray for the 79Br(n,γ)reaction at 7.9 MeV is also labeled [Sin05]. Thetransitions from the 139La(n,γ) reaction are vis-ible in Fig. 9.11 and are more pronounced inFig. 9.12 because of the smaller background fromthe 79Br(n,γ) reaction as compared to the 35Cl(n,γ) background in Fig. 9.11.

(MeV) γ E0 2 4 6 8 10 12 14 16

Det

ecto

r re

solu

tio

n (

%)

1

1.5

2

2.5

3

3.5

4

4.5

5

Rad. source data

S dataγHI

Figure 9.9: (Color online) The 10.16 x 15.24 cmB350 detector resolution vs. incidentγ-ray energy after correcting for thebeam energy spreads. The additionallow energy points are from radioactivesource data, and the dotted line is de-scribed in the text.

(MeV) γ E0 2 4 6 8 10 12 14 16

Det

ecto

r re

solu

tio

n (

%)

1

1.5

2

2.5

3

3.5

4Rad. source data

S dataγHI

Figure 9.10: (Color online) The 7.62 x 7.62 cmB380 detector resolution vs. inci-dent γ-ray energy after correcting forthe beam energy spreads. The addi-tional low energy points are from ra-dioactive source data, and the dottedline is described in the text.

(MeV) γ E3 4 5 6 7 8 9

Co

un

ts (

arb

itra

ry u

nit

s)

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

ClCl-1st escapeLaLa-1st escape

Figure 9.11: (Color online) The B350 detector re-

sponse from the 35Cl(n,γ) reaction.

(MeV) γ E3 4 5 6 7 8 9

Co

un

ts (

arb

itra

ry u

nit

s)

0

200

400

600

800

1000

1200

1400

Br

Br-1st escape

La

La-1st escape

Figure 9.12: (Color online) The B380 detector re-

sponse from the 79Br(n,γ) reaction.

[End98] P. M. Endt and R. B. Firestone, Nucl.Phys., A633, 1 (1998).

[Sin05] B. Singh, Nucl. Data Sheets, 105, 223(2005).

[vL00] E. V. D. van Loef et al., Appl. Phys.Lett., 77, 1467 (2000).

[vL01] E. V. D. van Loef et al., Appl. Phys.Lett., 79, 1573 (2001).

[Wel09] H. R. Weller et al., Prog. Part. Nucl.Phys., 62, 257 (2009).


9.2.3 Characterization of a Neutron Counter for High-Precision (γ,n)Cross-Section Measurements

C.W. Arnold, T.B. Clegg, C.R. Howell, H.J. Karwowski, G.C. Rich, J.R. Tompkins

TUNL

The detection efficiency of an INVS counter has been characterized using four well-understood

neutron sources. Both monoenergetic and broad spectrum neutron sources were employed.

Measured efficiencies agree with mcnpx simulations and improve the accuracy of the absolute

detection efficiency to ± 3% for neutrons of energy ≤1.0 MeV.

The model IV inventory sample counter(INVS) developed at Los Alamos National Lab-oratory [SJ93] was designed for fast, non-destructive assay of radioactive materials. Sim-ple adaptations make it an invaluable tool formeasuring (γ,n) cross sections using TUNL’sHIγS facility. For such measurements, it is criti-cal that the energy-dependent, absolute neutron-detection efficiency be known. Efficiency here isgenerally defined as

ε ≡ Ndetected

Nemitted. (9.7)

Four different sources, each able to emit a knownnumber of neutrons, were used. First, a NIST-calibrated 252Cf source provided a broad spec-trum of neutron energies with a neutron fluxknown to ±4.4% [Tho09]. Second, a coincidenceexperiment using the 2H(d,n)3He reaction pro-vided a monoenergetic source of 2.26 MeV neu-trons with a flux known to ±10%. A third inves-

tigation using the 7Li(p,n)7Be reaction produced∼1 MeV neutrons with fluxes known to ±6.6%.Finally, the 2H(γ,n)H reaction used the high res-olution HIγS beam to produce tunable sourcesof monoenergetic neutrons (0.1 ≤ En ≤ 1.0MeV) with ±3% accuracy. A comparison ofall experimental data with Monte Carlo modelsdemonstrates agreement to within the ±6% un-certainty imposed by modeling. However, datafrom 2H(γ,n)H improved the accuracy of the ab-solute efficiency to ±3% for neutrons with energy≤ 1.0 MeV.

A detailed report of these results will be pub-lished in the near future.

[SJ93] J. K. Sprinkle Jr. et al., Los AlamosNational Lab Report No. LA-12496-MS,page 11 (1993).

[Tho09] A. K. Thompson, private communica-tion, 2009.


Figure 9.13: Photograph of the INVS counter. A large cylindrical polyethylene body envelops 183He proportional-counting tubes, each containing 6 atm of 3He. Maximum detectionefficiency for this detector is ∼ 64%.

Figure 9.14: (Color Online) Ratio of measured efficiency to modeled efficiency vs. En. The red lineat a ratio of unity with a ±6% error band displays the modeling uncertainty. The blackline fits the majority of the data points, which are systematically lower than the modeledefficiency by approximately 2%.


9.2.4 Determination of the Absolute γ-ray Detection Efficiency of theMolly detector.

C.W. Arnold, A. Adekola, T.B. Clegg, H.J. Karwowski, G.C. Rich, J.R. Tompkins

TUNL

The 340 keV resonance of the 19F(p,αγ) reaction was accessed using the mini-tandem accel-

erator at TUNL. The γ rays from this reaction are known to be emitted isotropically, and

lead shielding was employed to collimate them into a 2.54 cm diameter distribution centered

on the face of TUNL’s large NaI detector known as Molly. The total efficiency of Molly is

consistent with ∼100% for a beam of 6.130 MeV γ-rays. Any loss of efficiency from threshold

settings may be accurately discerned from modeling in geant4.

For many experiments performed at TUNL’sHIγS facility, an absolute measure of the effi-ciency of the large NaI crystal detector known asMolly is needed in order to understand the limitsof precision for experiments requiring knowledgeof the incident γ-ray flux. Models predict thatMolly has a total-efficiency of ∼100% for γ-raybeams incident on its face with energies between1.5 and 15 MeV. This total efficiency may be af-fected by electronic threshold settings, light col-lection efficiency, and variables for which ascrib-ing uncertainties may be difficult. The 19F(p,αγ)reaction, which provides a monoenergetic γ-raycoincident with an associated α particle, was cho-sen to measure Molly’s absolute γ-ray efficiencyexperimentally.

The 19F(p,α) reaction creates an excited stateof 20Ne that then decays to one of four avail-able excited states or the ground state in 16O.Of the four excited states, three may decay tothe ground state of 16O via γ-ray emission. Thethree γ-ray-associated α particles will have dif-ferent energies, corresponding to the different ex-cited states in 16O∗ to which the 20Ne∗ decays.Figure 9.15 shows a typical silicon detector spec-trum, with three clearly defined α-particle peaks,corresponding to the three states in 16O∗. Theresonance for 340 keV incident protons favors the3− state of 16O∗, which decays by emission ofa 6.13 MeV γ ray. These γ rays are emittedisotropically [Hel72]. To produce beam-like con-ditions, more than 20 cm of lead shielding wasarranged to collimate the γ rays down to a 2.54cm diameter circle centered on the front face ofMolly. The solid angle opening defined by thislead shielding determines the fraction of the α-particle-associated γ rays incident on the face of

Molly.Figure 9.16 shows the experimental setup.

A small scattering chamber designed with verythin windows was set on the beam line followingTUNL’s mini-tandem accelerator. Six silicon de-tectors were placed at backward angles to detectoutgoing α particles. Incident protons with ener-gies ≥ 340 keV were collimated by a double-slitcollimator located 6.35 cm away from the tar-get. The targets were thin films of LiF approx-imately 50 μg/cm2 in thickness evaporated onto∼10 μg/cm2 carbon backings.

The presence of a peak in the TAC spectrumconfirmed α-γ coincidences. Software gates iso-lated events that were both in the resonant α-particle peak of each silicon detector and in thepeak of the TAC spectrum, thus generating aspectrum of coincident events in the NaI detec-tor (see Fig. 9.15). Coincident counts that fellwithin the full energy peak were ideal for thepresent analysis for two reasons. First, naturalbackgrounds at 6.13 MeV are low. Second, con-fusion between γ rays that scattered before enter-ing the detector and those that Compton scatterwithin the detector is eliminated by choosing onlyγ rays in the full energy peak.

The measured peak efficiency of Molly is

εpeak =Npeak

γ

dΩ4π × Nα

(9.8)

where Npeakγ is the number of detected γ rays in

the full energy peak, dΩ4π is the fractional solid an-

gle defined by the lead shielding, and Nα is thenumber of α-particles detected for the resonantreaction.


Figure 9.15: (Color Online) (Top) Three α-particle peaks from a typical silicondetector spectrum. Gates have beendrawn around the resonant peak.(Bottom) The summed spectrum ofall coincident γ rays from all runs.A Gaussian fit to the full energypeak at 6.13 MeV (inset) determines

Npeakγ .

Over 6000 total α-γ coincidences resultedfrom 1.3×107 total α particles counted over thecourse of 50 hours of data collection. The frac-tional solid angle defined by the lead shieldingwas calculated to be (4.87±0.12)×10−4. Preci-sion was limited by uncertainties in the measure-ments of the distance between the target and the

Figure 9.16: (Color Online) Photograph of theexperimental setup. Molly (bot-tom) sits behind a collimating wall oflead. Additional lead shielding wasplaced inside the scattering cham-ber(center).

end of the collimator, and of the collimator di-ameter. Simulations of the solid angle were per-formed using the Monte Carlo code mcnpx toaccount for the effects of finite beam size andknown geometrical constraints in the experiment.A fractional solid angle of (4.91±0.12)×10−4 wasadopted, yielding a peak efficiency of 69.5±1.9%.This peak efficiency is consistent with geant4

models of Molly, which predict a peak efficiencyof 69.7% and a corresponding total efficiency of98.9% for an incident beam of 6.13 MeV γ-rays.From these results we infer that experiment-dependent losses of efficiency caused by electronicthreshold settings can be modeled very accu-rately.

We would like to acknowledge M. Ahmed, S.Stave, P. Martel, S. Henshaw, A. Couture, D.Ticehurst and the Duke Machine Shop for theirhelp in making this experiment successful.

[Hel72] R. Hellborg and L. Ask, Physica Scripta,6, 47 (1972).


9.2.5 Characterization of a Gd-Loaded Neutron Detector

D.R. Ticehurst, M.W. Ahmed, C.R. Howell, H.J. Karwowski, J.R. Tompkins, W.Tornow,TUNL

A neutron detector filled with BC-521 Gd-loaded liquid scintillator has been characterized.

The efficiency, energy resolution, and pulse shape discrimination (PSD) properties were mea-

sured. The detector was developed to assess the possibility of using BC-521 for (γ,n) cross

section measurements or measurements of delayed neutrons from fission.

Direct measurements of total cross sectionsfor (γ,n) reactions and delayed neutrons from fis-sion require neutron detectors that cover a largesolid angle. Detectors with fast time resolution(<100 ns) and that measure energy without TOFwill enable a number of new experiments that re-quire detection of low energy neutrons in the exitchannel. A prototype gadolinium-loaded detec-tor (Gadet) was constructed two years ago andcharacterized this past year using both measure-ments and simulations.

Gadet, shown in Fig. 9.17, consists of a canthat is 20 cm long and 13 cm in diameter, filledwith BC-521. It has a borosilicate glass win-dow attached to a 13 cm PMT. Like most or-ganic scintillators, BC-521 exhibits a fast re-sponse (rise/decay times on the order of 10 ns)and PSD capabilities, allowing neutrons to bedistinguished from γ rays.

Figure 9.17: (Color Online) A photo of Gadet.The detector has a total length ofabout 65 cm and a mass of about 12kg.

One way to quantify the PSD capabilities of ascintillator is with a figure of merit, M, calculatedfrom a 1-dimensional PSD histogram:

M =Δ

δγ + δn

where Δ is separation of the peaks’ centroidsand δγ (δn) is the FWHM of the γ-ray (n) peak

[Ban07]. Using a 241Am-Be source, M = 0.59 forGadet (see Fig. 9.18 compared with M = 1.24for a 5 cm long × 13 cm diameter scintillatordetector filled with BC-501A.

Figure 9.18: A PSD spectum from Gadet using a241Am-Be source

Neutron energies are generally measured us-ing TOF. The addition of natGd to liquid scintil-lator permits another method of measuring neu-tron energy, due to gadolinium’s 49 kb thermalneutron capture cross section. When a neutronis captured by Gd, energy is released, typicallyas three γ rays with a total energy of ∼8 MeV[Ban07]. These γ rays may produce a signal inthe detector. However for a neutron to captureon Gd, it must thermalize. The full thermal-ization process takes several μs but produces aprompt signal (ns wide) proportional to the ini-tial neutron energy. Thus, after a prompt sig-nal from a neutron (γ rays excluded by PSD),the next signal will be from either neutron cap-ture or an uncorrelated random event. At count-ing rates of a few kHz, the spectrum from ran-dom events within 40 μs (the TAC range) of theprompt signal is flat and thus only translates theoverall spectrum up. The characteristic neutronthermalization time can be measured with a time


distribution (TD) spectrum: a histogram of thetimes between the prompt signal and the verynext signal. For neutrons, the histogram consistsof a flat background and a peak centered at theaverage moderation time, about 6 μs . Apply-ing both PSD and TD gates (for neutrons andmoderated neutrons, respectively) to the pulseheight (PH) data gives a PH spectrum of the neu-tron energies (a technique called “capture gat-ing” [Kno00]). Thus a detector such as Gadetcan measure neutron energy without TOF.

Capture gating was used to make neutronenergy spectra for monoenergetic neutrons pro-duced by the 2H(d, n)3He reaction (see Fig.9.19). The width of the peaks characterizesGadet’s energy resolution, is poor due to the lowlight output characteristic of neutrons, but im-proves with increasing energy.

Figure 9.19: (Color Online) Three capture-gatedPH spectra(with their energy reso-lutions) from 6, 8, and 10 MeV neu-trons.

The neutron detection efficiency of Gadet wascalculated with geant4 and measured using mo-noenergetic neutrons at several energies. Thegeant4 simulation used the “QGSP BIC HP”physics library and light output functions fromthe Monte-Carlo efficiency code neff7 [Die82].These functions were measured for BC-501A butwere used here since light output functions forBC-521 have not been measured. The light out-put per event was scaled down according to thepercent anthracene light output published in themanufacturer’s data sheets.

The efficiency measurement was conductedwith a pulsed monoenergetic neutron beam at6, 8, and 10 MeV. The pulsed beam permittedTOF measurements and a plastic scintillator pad-dle provided relative flux normalization. Gadetand a 5 cm × 13 cm BC-501A detector with awell-known efficiency were placed in the beamseparately at each energy. Gadet’s absolute ef-ficiency was calculated as

εGd =YGd

Y0× r2

0

r2Gd

× ε0

where Y is the flux-normalized neutron yield andr is the detector radius. The subscripts Gd and 0denote Gadet and the reference detector, respec-tively.

Figure 9.20: (Color Online) Gadet’s neutron effi-ciency for 6, 8, and 10 MeV neutronsat 1 × Cs and 1/2 × Cs thresholds.

The advantage of a liquid scintillator neutrondetector loaded with Gd or some other elementwith a high thermal-neutron-capture cross sec-tion is the ability to measure neutron energywithout TOF while retaining PSD capabilities.The trade off is poor energy resolution and worsePSD than with a standard liquid scintillator.Since neutrons take several μs to moderate, theADC gate must be long, limiting the countingrate. The low count rate combined with the factthat only about 3% of the incident neutrons endup producing a detected capture signal leads tolong counting times. However, such a detectormight have an application where neutron energyinformation and large solid angle-coverage are si-multaneously desirable.

When Gadet was disassembled after comple-tion of the characterization, a breakdown of thegrease junction between the light guide and thePMT tube was noticed. This likely caused a sys-tematic decrease in the amount of light reachingthe PMT, and thus our characterization measure-ments may be considered lower limits.

[Ban07] K. Banerjee et al., Nucl. Instrum. Meth-ods A, 580, 1383 (2007).

[Die82] G. Dietze and H. Klein, TechnicalReport PTB-ND-22, Physikalisch-Technische Bundesanstalt, Braun-schweig, Germany, 1982.

[Kno00] G. F. Knoll, Radiation Detection andMeasurement, Wiley and Sons, 3rd edi-tion, 2000.


9.2.6 Ratio of Ge Detector Peak Efficiencies at Energies of 4.4 and 11.7MeV

S. Carson, Gordon College, Wenham, MA; C. Iliadis, J. Cesaratto, A. Champagne, L. Dow-

nen, J. Kelley, R. Longland, J. Newton, G. Rusev, A. Tonchev, TUNL; M. Ivanovic,Department of Radiology, University of North Carolina, Chapel Hill, NC

Full-energy peak efficiencies of Ge detectors are frequently investigated at energies below 4

MeV using calibrated radioactive sources, while very accurate peak efficiencies for higher

photon energies are essentially nonexistent. Peak efficiencies in the energy range of E = 4− 12

MeV are crucial for a number of applications, including nuclear astrophysics measurements

of fusion reactions and resonance fluorescence experiments. We report on a novel method of

accurately measuring the ratio of full-energy peak efficiencies at 4.44 and 11.66 MeV using

the 163 keV resonance in the 11B(p,γ)12C reaction. Our values have a precision of 1.4− 1.6%.

The determination of full-energy peak effi-ciencies for Ge detectors has been the subjectof numerous studies. Below an energy of 3.5MeV, radioactive sources are conveniently usedfor such calibrations. Computer simulations, us-ing Monte-Carlo procedures, are also routinelyperformed by many groups. Frequently, only rel-ative peak efficiencies are computed and the sim-ulated results are normalized to the data in orderto aid in the interpolation of efficiencies betweenmeasured energies.

For the measurement of peak efficiencies athigher energies—say, between 4 and 12 MeV—anumber of different strategies are employed, de-pending on the technical capabilities of the lab-oratory. For example, the low-energy peak effi-ciency data from the radioactive source measure-ments are used for normalizing simulated peakefficiencies obtained from Monte-Carlo calcula-tions. The normalized simulation results arethen adopted for extrapolating peak efficienciesto high energies. Such an extrapolation is notwithout dangers, since the energy dependence ofthe peak efficiency may differ at low and high en-ergies. Furthermore, it is difficult to assign uncer-tainties to a peak efficiency that has been extrap-olated over many MeV. The present work wasmotivated by the scarce data on accurate full-energy peak efficiencies of Ge detectors at highphoton energies (E = 4 − 12 MeV).

Here we discuss the use of the Elabr = 163 keV

resonance in 11B(p,γ)12C. It gives rise to a strongcascade with photon energies of 4.44 and 11.66MeV. For such a cascade, the ratio of peak effi-ciencies for the first (RprimaryS) transition and

second (RsecondaryS) transition is determinedonly by the corresponding ratio of measured netpeak intensities. One goal of our study was todetermine how accurately the full-energy peak-efficiency ratio at these two energies can be deter-mined when using this particular resonance. Weaddress all of the important effects that may in-fluence this ratio: side-feeding γ-ray branchings,angular correlations and coincidence summing. Asecond goal of our study was to find out how wellthe Monte Carlo codes geant3 and geant4 areable to reproduce the measured peak-efficiencyratio at high γ-ray energy.

In brief, the experiment was carried out atthe Laboratory for Experimental Nuclear Astro-physics (LENA). A 1 MV JN Van de Graaff ac-celerator supplied proton beams of up to 10 μAon target in the bombarding energy range ofElab

p = 150− 180 keV. The target was fabricatedby evaporating boron, enriched in 11B, onto a0.05 mm thick tantalum backing. The beam wasfocused into a profile of ≈10 mm diameter ontarget. The counter consisted of a p-type, closedend, bulletized high-purity germanium (HPGe)detector. It was purchased from EG&G Ortec(Oak Ridge, TN, USA) in 1995 and was originallyspecified as having a 140% relative efficiency. Sig-nals from the detector were processed using stan-dard VME electronics and analyzed with the dataacquisition system jam. The detector was posi-tioned at an angle of 67o with respect to the inci-dent proton beam direction in order to minimizeangular correlation effects. Three distances be-tween the target midpoint and the center of theouter surface of the detector end cap were chosen


in the measurement: 3.05, 11.91 and 26.36 cm.The main results can be summarized as fol-

lows. First, the precision of the peak-efficiencyratio achieved in our experiment amounts to1.4 − 1.6%. A longer measurement time in orderto accumulate more statistics would be of limiteduse since the uncertainty arising from the angu-lar correlation correction alone amounts to 1.0%.Clearly, a significantly improved precision wouldrequire a dedicated remeasurement of the angularcorrelations in the decay of the Elab

r = 163 keVresonance. Second, the peak-efficiency ratios fordifferent distances are consistent within uncer-tainties, despite the fact that the measurementat the shortest distance (3.05 cm) is expected tosuffer from significant coincidence summing ef-fects.

As a next step, we determined the detec-tor crystal geometry independently from the di-mensions supplied by the manufacturer. Imagesfrom both radiographs and computed tomogra-phy were obtained by scanning our detector atUNC hospitals. Based on the derived crystal di-

mensions, computer simulations using geant3

and geant4 were performed. We can drawthe following conclusions by comparing measuredand simulated peak-efficiency ratios at γ-ray en-ergies of E = 4.44 MeV and 11.66 MeV: (i) themeasured and all of the simulated peak-efficiencyratios show the same dependence on distance;(ii) the geant4 simulation results agree with themeasured peak-efficiency ratios within an uncer-tainty of 1.6%; (iii) the code geant3 underpre-dicts the measured peak-efficiency ratios by 2.6%.The results presented here allow for an estima-tion of uncertainties when peak efficiencies areextrapolated from low energy data to high ener-gies based only on simulations. More experimen-tal work is called for in the future in order tocompare measured and simulated absolute full-energy peak efficiencies.

This work has now been published. For de-tailed results, see Ref. [Car10].

[Car10] S. Carson et al., Nucl. Instrum. Meth-ods, A618, 190 (2010).


9.2.7 Initial Studies of HPGe Detectors for Nuclear Astrophysics

G. Rich, A.E. Champagne, C. Iliadis, TUNL

We have begun simulation studies aimed at a conceptual design for an HPGe detector op-

timized for nuclear-astrophysics measurements. This work will be the starting point for a

project for the Deep Underground Scientific and Engineering Laboratory (DUSEL) to design

a detector for a future underground accelerator laboratory.

A significant amount of effort has been de-voted to designing and constructing HPGe de-tectors for nuclear-structure and double-β-decaystudies, and some of these design concepts couldbe adapted to the needs of nuclear astrophysics.For example, although the usual close target-detector geometry limits the usefulness of track-ing, some degree of segmentation could be usedto suppress external backgrounds. Event local-ization provided by point-contact detectors mayalso prove effective in reducing backgrounds. Toexplore some of these ideas, we have begun somesimple simulation studies of HPGe detectors. Tobegin, we looked at the ideal dimensions for acylindrical, point-contact detector. The masswas fixed at 2.7 kg, which is similar to that ofa commercial, large-volume detector. For γ rayswith energies of 1-10 MeV, the ideal dimensions(as predicted by geant4) are approximately 9cm in diameter by 7.9 cm long, and these dimen-sions were fixed for subsequent calculations.

Pulses from a point-contact detector carry in-formation about the location of an event, whichwill be investigated in detail in future studies.For now, we have explored two simple back-ground cuts based on the number of interactionsfor a fully absorbed γ ray and on the longitudi-nal distribution of energy. We used an isotropiccircular plane source of 1 MeV γ rays, 2.54 cmin diameter, parallel to the “front” face of thedetector and at a distance of 1.6 cm to representγ rays produced in a typical accelerator target.Two background sources (40K and 208Tl) wereemulated as isotropic point sources located 6 cmdirectly above the midline of the detector volume.The number of 1 MeV γ rays interacting versusdistance along the detector is shown in Fig. 9.21.

Although the geometry used here was ratherartificial, it does appear that a cut on energy de-posited versus distance along the detector wouldbe helpful, and this will be explored in more de-tail in future studies. We also found that photo-

peak events undergo about 10 interactions, on av-erage, so that requiring some minimum numberof events would have the effect of suppressing aportion of the Compton continuum. In Fig. 9.22,we have applied cuts requiring z > 0 and morethan six interactions to the 208Tl events, thusreducing them by a factor of three to five. How-ever, these cuts have a much smaller effect on 1-MeV-photo-peak events from target γ rays (seeFig. 9.23), which are reduced by only ∼30%. Atthe same time, there is a significant reduction inCompton events, which could be useful for situa-tions where a weak line of interest is obscured byCompton events from a stronger line.

Figure 9.21: (Color online) Counts (or energy de-posited) versus detector depth. Thefront face of the detector is locatedat z = 4 cm and a 208Tl source wassuspended above z = 0

The results described here are intended toguide more comprehensive calculations that areplanned for later this year. This work will be partof a funded Solicitation-4 DUSEL project aimedat defining a conceptual HPGe detector for a fu-ture underground accelerator. However, it shouldbe noted that such a detector would also be usefulabove ground and may be considered for futurework at LENA.


Figure 9.22: (Color online) Spectra of 208Tl events. The lighter color shows the raw 208Tl spectrum,while the darker spectrum results from the requirements that z > 0 and that there bemore than six interactions.

Figure 9.23: (Color online) Spectra of 1 MeV “target” events. The lighter color is again the rawspectrum, while the darker spectrum results from the requirements that z > 0 and thatthere be more than six interactions.


9.2.8 Upgrade and Assembly of the APEX Trigger Detector

S. Daigle, A.E. Champagne, TUNL

The APEX trigger detector was assembled, and its original 48 PMTs were upgraded with 32

Hamamatsu R580 tubes and 16 Photonis XP2012B tubes. A radial collimator was designed

and constructed to facilitate simultaneous position calibrations of the 24 NaI(Tl) segments.

A new CAEN V862 charge-to-digital converter is currently being tested for use in data ac-

quisition, to eliminate the integration of noise in the PMT output signals.

The 24-element NaI(Tl) scintillator arrayoriginally built for the APEX experiment [Kal93]has been upgraded and assembled for use in low-energy nuclear experiments at LENA. The origi-nal 48 PMTs have been replaced with 32 Hama-matsu R580 tubes and 16 Photonis XP2012Btubes in new aluminum mounts. The new PMThousings have eliminated light leaks, while theupgraded tubes provide improved position andenergy resolution [Dai09]. Figure 9.24 is a pho-tograph of the fully assembled NaI(Tl) array lo-cated in the detector setup area of LENA.

Figure 9.24: (Color online) Photograph of thefully constructed APEX detectoroutfitted with 48 new PMTs and alu-minum mounts.

Before assembly, all detector segments werethoroughly cleaned, photographed, and docu-mented to catalog the current condition of theNaI(Tl) crystals. A few crystals showed signsof damage in the form of streaking and foggingaround the quartz windows. These minor defectsmost likely occurred during transportation butalso may be a result of their age. This informa-tion is kept on file for future reference in the eventwe suspect a scintillator bar is behaving poorly.During assembly, a dime-size amount of Saint-Gobain BC-630 silicone optical grease was placedon the PMT glass surface and then methodicallypressed against the quartz window using constantpressure. Positive pressure is maintained by rub-ber o-rings within the PMT housing. Once as-sembled, the optical grease and quartz windowprovide a transparent interface between the PMTand the scintillator crystal, while the o-rings alsoshield the tube from light leaks.

A radial collimator was constructed to pre-cisely control the physical position of a γ-ray disksource along the position-sensitive axis of the de-tector. Figure 9.25 is an Autodesk Inventor im-age of the radial collimator, which is seated onThomson rails to allow for unrestricted move-ment and quick calibration while the detectoris in position on the beam line at LENA. Thecollimator is constructed of a 11.43 cm diame-ter aluminum pipe that houses a delrin containerattached to the end of a 1.07 m aluminum rod.The aluminum rod has been etched in 0.5 cm in-crements to allow for precise positioning of thesource. Inside the delrin container are two cylin-ders of lead (5.08 cm thick and 10.16 cm in di-ameter) that radially collimate the γ-ray source.The dimensions of the lead collimators were cho-sen so that the size of the γ-ray image on thedetectors would be smaller than their intrinsicposition resolution. The delrin container can slip


freely throughout the aluminum pipe, allowingthe collimator base to be locked in position whilethe source is inserted into the detector using therod-and-container assembly.

Figure 9.25: (Color online) Autodesk Inventorgraphic of radial collimator onThomson rails (three-quarter sec-tion view). The collimator allowsfor precise measurements along theposition-sensitive axis of each scintil-lator bar in 0.5 cm increments.

We have simulated the response of the APEXdetector to the collimated source using thegeant4 simulation package [Ago03]. Figure 9.26is the result of a simulation of 1332 keV γ raysemitted from a 60Co disk source placed insidethe radial collimator. The simulated FWHM ofthose events detected is 2.5 cm, which is less thanthe position resolution of the detector (approxi-mately 3.5 cm).

A CAEN SY2527 multichannel HV powersupply was recently purchased to bias all 48PMTs to their operating voltages (approximately-1500 V). Gain-matching each PMT can be easilyperformed by adjusting the voltages on the power

supply that bias the tubes on either end of a de-tector segment. Progress is being made on devel-oping an optimal electronics setup for data acqui-sition. Currently, we are investigating the advan-tages of using an individual gate charge-to-digitalconverter (QDC) (V862) available from CAENelectronics over our existing setup using a CAENADC (V785). There is evidence that the QDCwill greatly reduce the steep slope of low-energycounts in our spectra that arise from integratingnoise in the output signals of the PMTs. In or-der to complete the proposed electronics setup,two V862 QDCs must be purchased to replacethe V785 ADCs.

Position (mm)-300 -200 -100 0 100 200 300

Co

un

ts

0

50

100

150

200

250

300

350

Figure 9.26: geant4 simulation of γ-rays emitted

from a 60Co source inside the radialcollimator. The FWHM of simulated1332 keV γ-rays is 2.5 cm.

[Ago03] S. Agostinelli et al., Nucl. Instrum.Methods, A506, 250 (2003).

[Dai09] S. Daigle and A. E. Champagne, TUNLProgress Report, XLVIII, 166 (2009).

[Kal93] N. I. Kaloskamis et al., Nucl. Instrum.Methods, A330, 447 (1993).


9.2.9 Rutherford-Backscattering Analysis of Polymer Membranes Usedfor Water Purification

P.J. Attayek, O. Coronell, Gillings School of Public Health, University of North Carolina,Chapel Hill, NC ; T.B. Clegg, TUNL;

The concentration of ionizable functional groups in reverse-osmosis and nanofiltration polymer

membranes, as well as the concentration of contaminants partitioned into the membranes, can

be determined by Rutherford backscattering of He ions from virgin and treated membranes,

respectively. In preparation for such experiments, we have constructed a sample holder to

mount inside an existing scattering chamber and hold multiple membranes. The position of

each membrane can be rastered across the beam by stepper motors controlled via LabVIEW.

9.2.9.1 The Experiment

The concentration and distribution of ionizablefunctional groups in the active layer of reverse-osmosis (RO) and nanofiltration (NF) mem-branes will be studied using Rutherford backscat-tering spectrometry (RBS) [COR08, COR09].Such RO and NF membranes are increasingly at-tractive technologies for the drinking-water andwater-reuse industries, because they are able toremove a broad range of water contaminants inone treatment step without the addition of chem-icals to the water. The structure of RO andNF membranes usually consists of an ultra-thin(≈100 nm), fully aromatic polyamide layer ontop of a polysulfone support (≈50 μm) backedby a layer of polyester fibers (≈300 μm), withthe polyamide layer representing the main bar-rier to the permeation of water and contaminants[BMI06]. The polyamide layer, usually referredto as the active layer, contains ionizable func-tional groups that provide charges to the mem-brane and affect the rejection of ionic contami-nants [COR08]. The ionizable functional groupsare broken links in the polymer structure andthus are directly related to the free volume ofthe active layer and its permeability to water andcontaminants. Accordingly, achieving a thoroughunderstanding of the concentration and distri-bution of ionizable functional groups in activelayers is important to both (i) understand themechanisms of transport of water and contami-nants and (ii) identify the improvements neededin commercial membranes in order to developnew membranes with enhanced performance.

The experiments will consist of saturatingionized functional groups with heavy ion probes,

and determining the ion probe concentration byRBS [COR08, COR09]. Negative and positivefunctional groups will be saturated with silver(Ag+) and tungstate (WO4

2−), respectively. Abeam of 2 MeV He+ will be used for analysisby backscattering. Currents of 40-100 nA will beused, with beams having a diameter of 2-3 mm.The He+ ion fluence on membrane samples willbe maintained below threshold values for degrad-ing the integrity of the polymers [BMI07]. Mem-brane samples will be prepared using ion-probesolutions in the pH of interest for membrane anal-ysis (2-11), with emphasis on the pH range of in-terest for water treatment (6-10). Divalent bar-ium (Ba2+) will also be used to probe negativefunctional groups. By comparing the results ob-tained for Ag+ and Ba2+, the average distancebetween ionizable functional groups will be cal-culated [COR09]. The results from this studywill advance the understanding of the physico-chemical structure of the active layer of RO andNF membranes.

9.2.9.2 The Sample Holder

The sample holder accommodates multiple tar-gets, each of which scans through the beam incontiguous arcs. Samples are mounted on an alu-minum wheel which moves vertically and rotatesazimuthally within the scattering chamber. Thewhole assembly also rotates around an axis per-pendicular to the scattering plane to set the in-cident angle of the He ion beam on the samples.The vertical and azimuthal motions of the sam-ple wheel are achieved with small stepper motors,while the rotation of the whole assembly is man-ual. Absolute encoders are mounted onto the rear


shafts of the stepper motors to register their ex-act position. Both stepper motors are controlledby a single driver located outside of the scatteringchamber.

Figure 9.27: Sample Holder Assembly

The sample wheel is detachable from the as-sembly, allowing the user to load the samplewheel easily for RBS analysis. The wheel is elec-trically insulated from the rest of the assemblyand connected externally via a graphite brushand wire to facilitate beam current integration.

National Instruments software LabVIEW isused to locate the samples on the wheel and de-termine the rastering pattern for each sample. Aphotograph taken of the loaded sample wheel isuploaded to LabVIEW, and the picture is cali-

brated with user input to convert Cartesian co-ordinates in pixels into polar coordinates in mil-limeters. The user then inputs the number ofsamples on the wheel and outlines each sample.Finally, the LabVIEW code determines rasterarcs that will fit within each sample for a par-ticular beam spot size.

A separate LabVIEW program is used to con-trol the stepper-motor driver and to read the sig-nal from the absolute encoders mounted on thestepper motors. The design of the sample holderalong with the absolute encoders allows Lab-VIEW to know the exact position of the beam onthe sample wheel. The LabVIEW program con-trols the stepper motors to raster each sample inthe predetermined arcs. LabVIEW also controlsexternal hardware to allow (inhibit) data collec-tion from the detectors in the chamber when thebeam is (is not) hitting a sample.

[BMI06] B. Mi et al., J. Mem. Sci., 282, 71(2006).

[BMI07] B. Mi, D. Cahill, and B. Marinas, J.Mem. Sci., 291, 77 (2007).

[COR08] O. Coronell et al., Environ. Sci. Tech-nol., 42, 5260 (2008).

[COR09] O. Coronell, B. Marinas, and D. Cahill,Environ. Sci. Technol., 43, 5042 (2009).


9.3 Facilities

9.3.1 Development of an Ultracold Neutron Source at the NC State PUL-STAR Reactor Facility

R. Golub, P.R. Huffman, G. Medlin, G. Palmquist, A.R. Young, TUNL; A. Cook, A.

Hawari, E. Korobkina, B. Wehring, North Carolina State University, Nuclear EngineeringDepartment, Raleigh, NC

We are in the final stages of constructing a next-generation ultracold neutron source at the

PULSTAR reactor facility on the campus of NC State University. We envision using the

source for performing neutron β-decay measurements as well as systematic studies for the

neutron electric dipole moment experiment.

Ultracold neutrons (UCNs) play an importantrole in nuclear-physics investigations that seek totest the Standard Model of particle physics andcharacterize the weak interaction. UCN-sourcetechnology is advancing rapidly, and the nu-merous techniques presently being implementedshould permit several orders of magnitude gainin useful UCN densities. Such higher densitiesof UCNs will not only permit improved mea-surements of quantities such as the neutron elec-tric dipole moment (EDM) and neutron β-decaylifetime, where measurements utilizing UCNs al-ready provide the most precise experimental val-ues, but will also play a role in neutron-angular-correlation measurements, neutron-antineutron-oscillation searches, spectroscopic studies of neu-tron decay, fundamental tests of quantum me-chanics, and, potentially, cutting-edge neutron-scattering applications.

The PULSTAR ultracold neutron (UCN)source is designed to be a next generation UCNsource that utilizes solid deuterium (SD2) ata temperature of 5 K to moderate thermalneutrons to UCN energies. The high leakageof fast/thermal neutrons from the reactor coremakes it possible to couple efficiently the neu-tron flux to the D2O thermal moderator and thesolid methane CH4 cold moderator that surroundthe deuterium.

The concept for our facility is to place theUCN source in a tank of D2O positioned inthe former thermal column of the PULSTAR re-actor. Neutrons leaving the face of the reac-tor core are channeled into the D2O tank by a45 cm × 45 cm × 70 cm long void in a graphiteassembly called the nose port. The source con-sists of a UCN converter of solid ortho deuterium,

17 cm diameter × 4.5 cm thick, held at a tem-perature of < 5 K. The converter is surroundedby a 1 cm thick, cup-shaped cold source of solidmethane held at a temperature of ≈ 25 K. Thisconfiguration was optimized for maximum cold-neutron flux using detailed mcnp calculations.Upcoming measurements of the thermal neutronflux inside the D2O volume will be used to bench-mark these calculations and to predict the use-ful experimental UCN population using a MonteCarlo transport code developed at NCSU. Thebasic geometry of the source is shown in Fig. 9.28,with a closer view of the cryogenic componentsshown in Fig. 9.29. An assembled view of thesource and guide housing is shown in Fig. 9.30.

Figure 9.28: (Color online) MCNP geometry plotshowing the general layout of theUCN source and guide, nose port,and reactor core.


D2 container and heat exchanger

CH4 container and

Neutron guide heat exchanger

Quartz Neutron Zircaloy temperature gradient ring

heat exchanger

guide insert

Figure 9.29: (Color online) UCN source and guidesystem.

Figure 9.30: (Color online) Photograph of the as-sembled UCN source. The large Alcontainer houses the D2O thermalmoderator that surrounds the UCNsource assembly.

Through a separate NSF grant, we have fund-ing to complete the commissioning and to lay thefoundation for a research program at PULSTAR.We will submit separate proposals to operate in-dividual experiments at the PULSTAR source.As an example, funding is being sought for workto support the neutron EDM effort based at OakRidge National Laboratory. A brief outline of theproposed research program is given below.

One area of activity focuses on tailoring thecold neutron energy spectrum to optimize UCNproduction. Previously we were involved in mea-surements of UCN production as a function ofcold neutron energy at Paul Scherrer Institutin Switzerland, together with C.-Y. Liu. Sub-sequently there have been very complete mea-

surements of the cold-neutron-scattering law forsolid deuterium by the Munich group. These newscattering kernels, together with our experimen-tal program, should permit us to produce an ex-tremely detailed evaluation of the performanceof the PULSTAR solid-deuterium source, permit-ting us to optimize our UCN production rate.

The program at PULSTAR will provide uswith unprecedented access to measurements ofour thermal, cold, and ultracold neutron fluxesand to measure and control properties of thesolid deuterium, such as the physical propertiesof the crystal and the para-deuterium contami-nation. This will permit us to understand andmaximize the output of both the PULSTAR andLos Alamos solid-deuterium sources.

Because of the complexity and long coolingcycle (∼6 to 8 weeks turn-around time) of theplanned large-scale neutron electric dipole mo-ment (nEDM) apparatus, the collaboration isbaselining into the project plan a smaller testapparatus that will be operated at the PUL-STAR UCN source. The smaller apparatus willenable the study of different techniques neededfor the nEDM measurement. The primary goalsfor these measurements are: (1) Determinationof the UCN storage time, 3He depolarizationtime, and UV-scintillation-detection efficiency ofnEDM measurement cells before installation inthe main apparatus at ORNL. (2) Studies anddevelopment of operating procedures related tooptimizing the techniques for observing double-resonance effects in the 3He-UCN system. Thesewill include studies to optimize the procedures forcritical spin dressing and techniques for minimiz-ing and measuring the “false EDM” systematicerror. (3) Studies of the trajectory correlationfunctions important for the geometric-phase sys-tematic errors and the influence of surface rough-ness. The apparatus will be based on our exist-ing dilution refrigerator and cryostat and will beflexible enough to allow meaningful studies in areasonable time. The basic idea will be to repro-duce a single full size measuring cell containingall the features of the final experiment except forthe electric field.

There is also the exciting possibility of operat-ing the Los Alamos neutron lifetime experimentat PULSTAR. This relatively compact experi-ment utilizes a 45 cm deep, 0.6 m3 trap formedby a Hallbach array of NdFeB permanent mag-nets (Bmax ≈ 1.3 T). Given that the LANSCEfacility will be operating for five or six monthsper year, sharing the experiment will provide agenuine opportunity to refine the approach andrealize the physics potential of this project.


9.3.2 Systematic Studies for the nEDM Experiment at the PULSTARUCN Facility

R. Golub, H. Gao, D.G. Haase, P.R. Huffman, A. Merizalde, C.M. Swank, Q. Ye,TUNL; A. Hawari, E. Korobkina, North Carolina State University, Nuclear Engineering Depart-ment, Raleigh, NC ; S.K. Lamoreaux, Yale University, New Haven, CT ; the nedm collabora-

tion

We are developing an apparatus, based on our existing dilution refrigerator and cryostat,

that will allow tests using a single full-size measurement cell (without an E field) of key

systematic effects for the neutron-EDM experiment. The tests will focus on demonstrating

and quantifying the double-magnetic-resonance technique using polarized 3He and ultracold

neutrons.

A program at TUNL is being developed withthe primary goal of significantly exploring keysystematic effects for the neutron-electric-dipole-moment (nEDM) experiment. In addition, mea-surements performed during the commissioningof the main apparatus will be costly and time-consuming, so this apparatus can be used to per-form characterization tests on a single full-sizemeasuring cell containing all the features of thefinal experiment except for the electric field. Thedesign of the apparatus will be based on our ex-isting dilution refrigerator and cryostat.

The primary objectives of the design of theapparatus are as follows:

• A standing set-up to measure ultracold-neutron (UCN) storage time, 3He depo-larization time, and ultraviolet scintillationdetection efficiency of the nEDM measure-ment cells prior to installation in the mainapparatus at ORNL.

• Studies and development of the operatingprocedures to optimize the observation ofdouble-resonance effects in the 3He-UCNsystem. These will include studies aimed atoptimizing the procedures for critical spindressing and techniques for minimizing andmeasuring the “false edm” systematic er-ror.

• The system will also serve as a stand-by testbed for studies of systematic errors thatmay be proposed as the full nEDM experi-ment is underway.

A general schematic showing the key componentsof the apparatus is shown in Fig. 9.31.

MixingChamber

BufferCell

Purifier

Helium-filled Measurment Cell

MagneticShielding

4KShield

77KShield

CryostatVacuumVessel

UCN fromPULSTAR

Source

UCNValve

B-field Coils

B0

3HeFill

4HeFill

PumpingLine

Figure 9.31: (Color online) Schematic of the pro-posed apparatus showing the keycomponents.

The measurement cell will be suspended fromthe bottom of a buffer cell connected to the mix-ing chamber of the dilution refrigerator. The longaxis of the measurement cell will be horizontaland enclosed in a set of cylinders containing themagnetic field coils and a ferromagnetic (Met-glass) shield. Additional magnetic shielding willbe mounted on the 4 K and 77 K cryostat heatshields and the vacuum vessel itself. These are allcoaxial cylinders with horizontal axes. The cellwill be cooled by a combination of gold platedcopper wires and 4He-filled acrylic tubes. ThenEDM cells have only one opening used to ad-mit polarized 3He. In our setup this opening willbe used for filling the cell with UCNs. The cell


will be filled with 4He and sealed by a thin foilwindow that will transmit UCNs from the PUL-STAR UCN source. The 4He and polarized 3Hewill enter the measurement cell through a smallhole on top of the cell.

The 3He will be polarized using the same ap-paratus previously used in the studies of 3Hewall depolarization and geometric phase effects.As the turn-around time for measurements us-ing 3He is limited by the polarization time, wewill have several hours to remove the depolarized3He. For the more detailed studies of the 3He-UCN spin dynamics, the 3He will enter and leaveby means of a dedicated 1 mm hole. This will notimpact the UCN storage time. There will be asmall evaporating chamber connected to the holeso that the depolarized 3He can be pumped away.The UCNs will enter through a thin window toconfine the 4He liquid and then through a UCNvalve.

Cosθ coil

MeasurementCell

4K77K

Figure 9.32: (Color online) Model of the cryostatfor the systematic effects apparatus.The length of the apparatus is ap-proximately 2 m.

Two quantities set the size of the apparatus:the measurement cell and the magnetic field coils.The measurement cell has been specified by thecollaboration to be rectangular with inner dimen-sions of 7 × 10 × 30 cm3. The measurementsrequire a uniform magnetic field across the cellregion, so the diameter of the magnetic field coilmust be large enough to accommodate the uni-form field region. Our initial estimate, based onfield calculations using a cos θ winding geometry,suggests that the diameter of the coil will need

to be approximately 50 cm and the length about1.5 m. Based on these dimensions, a preliminarydesign for the cryostat has been developed and isshown in Fig. 9.32.

The physics program will begin with the com-missioning of the 3He transport features and themeasurement of wall relaxation rates for the 3He.This work could be done “off-line,” before theapparatus is installed at the PULSTAR source.The work would continue with studies of the 3Hetrajectory correlation functions.

The same program would then be carried outwith UCNs alone. Finally, we would introduceUCNs and 3He into the cell together and start touse the scintillation detection to study the crit-ical dressing and its behavior as a function ofthe operating parameters. Experiments to studythe pseudomagnetic-field interaction and possi-ble new schemes for double resonance will thenbe undertaken.

The primary issues we envision addressingwith this apparatus are as follows:

• UCN storage and depolarization studiesto optimize cell materials and fabricationtechniques.

• New cell designs and trouble-shooting cellsthat don’t work in the main apparatus.

• Measurement of scintillations due to rela-tive 3He-UCN precession.

• Demonstration of the concept of criticaldressing through the observation of a con-stant scintillation rate.

• Detection of the 3He pseudomagnetic field.

• Techniques for reversing the 3He and UCNspins, σ(3He) and σ(UCN), and the mag-netic field B0.

• Measurement of the “trajectory correlationfunction” for systematic error characteriza-tion.

• Techniques for NMR imaging of 3He.

This program of studies will occur in parallelwith the construction and commissioning of thenEDM apparatus. We estimate that the entireprogram will last 4-5 years.


9.3.3 Development of the Fundamental Neutron Physics Beamline at theSNS

P.R. Huffman, C.R. Gould, TUNL; R. Allen, N. Fomin, V. Cianciolo, G.L. Greene, OakRidge National Laboratory, Oak Ridge, TN

The construction of the polychromatic cold and 0.89 nm beamlines at the Fundamental Neu-tron Physics Beamline (FNPB) facility at the Spallation Neutron Source (SNS) at Oak RidgeNational Laboratory is complete. The neutron spectrum for each line has been characterizedand benchmarked with simulations. The project has achieved CD-4 approval.

TUNL played a major supporting role in thedesign and construction of the Fundamental Neu-tron Physics Beamline (FNPB) facility at theSNS. In collaboration with the University of Ten-nessee/ORNL team, led by G. L. Greene, we haveparticipated in all aspects of the design, construc-tion, and commissioning.

The FNPB includes two beamlines: a poly-chromatic cold beam and a 0.89 nm monochro-matic beam used primarily for experiments thatproduce ultracold neutrons (UCNs) using the su-perthermal process in helium. The FNPB will beoperated as a user facility, with experiment selec-tion based on a peer-reviewed proposal process.An initial program of five experiments in neutrondecay, hadronic weak interaction, and time re-versal symmetry violation (nEDM) has been pro-posed. The n p → d γ experiment will be the firstexperiment to collect data on the cold beamline.

The FNPB has come online and completedcommissioning at the SNS facility. Initial com-missioning runs were aimed at measuring theneutron flux in order to demonstrate the DOECD-4 project deliverables. Additional issuesidentified as important to the success of researchcarried out at this facility are being character-ized. These include the optimization of the neu-tron flux and measurement of the phase spaceof the flux (energy, position, and angle), the timedistribution of neutrons of a given energy relativeto the time the proton pulse hits the spallationtarget, and the number and time distribution ofβ-delayed neutrons.

The first neutron fluence measurements weretaken at the FNPB while the accelerator oper-ated at 1 kW. These data are shown in last year’sprogress report [Huf09]. Monte Carlo simulationswere part of an earlier project at TUNL to de-sign, specify, and procure the neutron optics com-ponents for the FNPB. The performance of the

cold beamline is within about 20% of the initialestimates, which were based on the crude physicsmodel for the SNS moderator performance.

Recently we have focused on the installationand characterization of the 0.89 nm monochro-mator and the λ/2- and λ/3-filter systems forthe 0.89 nm beamline. A 0.89 nm monochro-matic beam is reflected out of the polychromaticbeam and directed to an external facility for UCNexperiments. The monochromatic beamline con-sists of a stage-1, potassium-intercalated graphitemonochromator and a second stage-1, rubidium-intercalated graphite monochromator configuredin a double-crystal design, along with two highlyoriented, pyrolytic graphite (HOPG) monochro-mators to remove the higher-order (λ/2 and λ/3)Bragg reflections from the beam. A secondaryshutter for the beamline is also present.

NC State University, Oak Ridge NationalLaboratory, and the Institut Laue Langevin(ILL) in Grenoble, France worked cooperativelyto construct the intercalated monochromatorsused in the FNPB. A recent photograph of therubidium intercalated graphite monochromatoris shown in Fig. 9.33.

The monochromators, filters, and shutterwere assembled externally and installed into themonochromator housing in the beamline. Theassembled components can be seen in Fig. 9.34prior to installation. The beam enters from therear-facing side of the K-intercalated monochro-mator, is reflected to the left towards towardsthe Rb-intercalated crystals and emerges in theforward direction.

Beam optimization and flux measurementswere performed at the end of the 8 m ballistichorn using two methods. In the first method, anaperture of known area masking a fast detectorwith known efficiency was used to tune the anglesof the monochromators so that the approximate


center of the beam was at 0.89 nm. Next, usingthe same aperture, the fast detector was replacedwith a neutron image plate and the identical mea-surement was performed. This step transfers thecalibration of the detector to the image plate.Lastly, the entire beam was covered with imageplate and the flux distribution was determinedacross the beam.

Figure 9.33: (Color online) A photograph ofthe rubidium intercalated graphitemonochromator assembly prior to in-stallation.

RbCrystals

KCrystals

λ/2 & λ/3filters

SecondaryShutter

Figure 9.34: (Color online) A photograph of thecomplete monochromator and filterassembly prior to installation.

The second method utilized a large formatCCD imaging detector borrowed from the NISTfundamental-neutron-physics group. A profile ofthe neutron intensity taken with the image plateis shown in Fig. 9.35.

The FNPB 413.3 Technical Specification re-quired for satisfying the DOE CD-4 requirementsis specified to be 2.1 × 108 n s−1 A

−1(MW)−1,

defined in terms of the maximum of the 0.89 nmflux distribution integrated over the entire out-put of the guide scaled to accelerator power.The initial measurement yielded a value of 6.7×108 n s−1 A

−1(MW)−1. Additional optimiza-

tions are planned, but the result easily satisfiesthe project requirements.

Figure 9.35: (Color online) The 0.89 nm neutronflux distribution at the end of themonochromatic beam guide.

TUNL also continues to play an active role inthe development of two key experiments that areenvisioned to collect data using the FundamentalNeutron Physics Beamline facility at the SNS: ahigh-precision search for the neutron EDM and ameasurement of the neutron lifetime using mag-netically trapped UCNs. These are discussed indetail in other sections of this report. The exec-utive committee for the instrument developmentteam consists of project manager G. L. Greene(Oak Ridge National Laboratory and Universityof Tennessee, Knoxville) and committee mem-bers W. M. Snow, chair, (Indiana University),V. Cianciolo (ORNL), J. D. Bowman (ORNL),J. Doyle (Harvard), C. R. Gould, and P. R. Huff-man.

[Huf09] R. R. Huffman et al., TUNL ProgressReport, XLVIII, 24 (2009).


9.3.4 Status of the KURF Low-Background Counting Facility

J. Strain, H.O. Back, P. Finnerty, R. Henning, S. MacMullin, J.F. Wilkerson, TUNL;R.B. Vogelaar, Virginia Polytechnic Institute and State University, Blacksburg, VA; R. Lind-

strom, National Institute of Standard and Technology, Gaithersburg, MD

The Kimballton Underground Research Facility houses a γ-ray counting facility consisting of

two HPGe detectors specifically designed for low-background assay work. Counting techniques

and sample preparation for γ-ray assay at this facility are discussed.

9.3.4.1 Introduction to the Facility

The Kimballton Underground Research Facility(KURF) is located about 25 miles from the cam-pus of Virginia Polytechnic and State Univer-sity in the Chemical Lime Company’s Kimball-ton Mine. The experimental area is located onthe 14th level at a depth of 520 m, or approx-imately 1450 meters of water equivalent shield-ing. A trailer owned by NRL and operated byUNC/TUNL houses the low-background count-ing facility.

The facility consists of two HPGe detectorsspecifically designed for low-background assaywork. One detector, “Melissa”, is a 1.1 kg, 50%RE (relative efficiency compared to NaI) Can-berra LB (low-background) detector. Melissa isoriented vertically and is cooled using a dipstickcryostat. Melissa’s shield consists of 15 cm ofDoe-run lead and 2.54 cm OFHC (oxygen-freehigh conductivity) copper. The fwhm at 1.33MeV is 1.70 keV, and the threshold is ∼20 keV.The other detector, “VT-1”, is a 0.956 kg, 35%RE ORTEC LLB Series detector in a J-type con-figuration. VT-1’s shield consists of a 10.1 cmORTEC commercial lead shield and 0.3 cm ofOFHC copper. VT-1 has a fwhm of 1.80 keV at1.33 MeV, and the threshold is ∼20 keV.

The inner cavity of both detectors in contin-uously purged with dry liquid nitrogen boil-offfrom a dedicated 30 L dewar, resulting in a re-duction in 222Rn by a factor of approximately 3.5.See Refs. [Mac09] and [Fin10] for a more detailedexplanation of background reduction techniques.

9.3.4.2 Sample Preparation

Low-background assay work requires that sam-ples are clean so that unwanted radionuclides arenot measured. Therefore samples are preparedbeforehand in a cleanroom and properly bagged

for transport to the low-background counting fa-cility in KURF. Depending on the material andthe required detection limit, different preparationprocedures may be used, including leaching plas-tics with acid, etching metals with acid, or clean-ing with ultra-pure solvents. Samples are thenbagged in clean-room nylon that has a low per-meability to radon.

Past samples have been prepared in aclean laboratory environment at TUNL that isequipped with a snorkel hood. A cleanroomat UNC is currently under construction (seeFig. 9.36) and is expected to be class 1000 orbetter. This cleanroom is equipped with a hood,proper air filtration, cleanroom furniture, ultra-sonic cleaners, and a high-purity water sourceproducing 18.2 MΩ·cm water [Str09].

Figure 9.36: (Color online) Cleanroom at UNC.

9.3.4.3 Assay Analysis

In May 2010 an object-oriented software pro-gram, orca, was installed as the data acquisitionsystem for the low-background counting facilityat KURF (see Sect. 9.4.4.

A Monte Carlo simulation for each assayedsample is done using mage, a geant4-based sim-ulation package maintained and developed by ajoint group of the Majorana and Gerda collab-orations [Bau06]. In a simulation, each sample is


uniformly doped with an isotope, particularly iso-topes from the 238U, 232Th, 40K, and 60Co decaychains. The γ-rays from these decays are trackedfrom their emission in the sample to the detec-tor. Figure 9.37 displays a simulation geometryof lead bricks that were assayed.

Figure 9.37: (Color online) Image of simulation oflead bricks in Melissa.

Using a locally developed root analysis pack-age and Monte Carlo simulations, the absolutepeak efficiency for an isotope can be found. Theactivity of an isotope can then be determined us-ing the equation

Aγ =N

εγ m(9.9)

where N is the net peak area, in units of countsper day, determined by subtracting a backgroundspectrum from the sample spectrum; εγ is the ef-ficiency; and m is the mass of the assayed sam-ple. A typical counting time for a sample is twoto four weeks. If no statistically significant peakis present for an isotope, an upper limit can beplaced on the activity of a sample.

To determine the detectors’ sensitivities, aMarinelli beaker, often used for γ-ray assay, wascounted in Melissa and VT-1. A mage/geant4

simulation was done to determine peak efficien-cies, and an upper limit was placed on the peaksas previously described. Table 9.2 displays thebest detection limits for 40K, 238U and 232Th.See Ref. [Fin10] for a complete table of the de-tection limits for 238U and 232Th, as well as fora more detailed explanation of the analysis tech-niques and Monte Carlo validation for KURF’s

low-background counting facility.

Table 9.2: Detector Sensitivies

Isotope Energy Detection Limits [mBq/kg](Chain) [keV] Melissa VT-1

40K (40K) 1461 30 90214Bi (238U) 609 10 15

212Pb (232Th) 238 6 10

9.3.4.4 Conclusions

KURF’s low-background counting facility hasbeen in operation and doing assay work for overtwo years. Assay work done in the last year in-cludes:

• Axon Picocoax R©, a possible candidate formaking low-background signal cables forthe Majorana project and other low back-ground experiments.

• Sullivan Lead Bricks, low-background leadbricks that could be possibly used forshielding for the Majorana project.

• University of Washington Lead Bricks, low-background lead bricks that could possiblybe used for shielding for the Majorana

project.

• PEEK Plastic, low-background plastic thatcould possibly be used in the Majorana

project.

For a complete list of all assay work done andthe results see Ref. [Fin10].

Future work, beyond assaying more samples,includes continuous radon monitoring and mea-suring the total muon, neutron, and γ-ray flux atKURF.

[Bau06] M. Bauer et al., J. Phys. Conf. Ser., 39,362 (2006).

[Fin10] P. Finnerty et al., arXiv:nucl-ex/1007.0015, (2010).

[Mac09] S. MacMullin et al., TUNL Progress Re-port, XLVIII, 155 (2009).

[Str09] J. Strain et al., TUNL Progress Report,XLVIII, 159 (2009).


9.3.5 Unfolding Full-Energy Peaks in γ-ray Spectra

A.D. Hill, TUNL

A method is developed for spectral unfolding for generic γ-ray detectors in order to determine

the energy distribution of 2-15 MeV γ-ray beams, using germanium detectors as an example.

It is assumed that the distribution is approximately Gaussian.

In the analysis of nuclear spectra, it is oftencritical to determine the energy profile of the in-cident beam. For example, when calculating re-action cross sections, one must first know the fluxof incident γ rays at the resonance energy. Sucha beam profile may be generated via spectralunfolding, which involves processing a detectedspectrum to find the incident energy distribution.While more complicated, more accurate methodsexist, we present a very simple, rough method forthe approximation of this incident γ-ray energyprofile. This method relies solely on processingthe full-energy peaks from zero-degree measure-ments of γ-ray beams, without respect to beamparameters.

(a) Energy [keV]0 500 1000 1500 2000 2500 3000

Co

un

ts

50

100

150

200

250

300

(b) Energy [keV]2500 2600 2700 2800 2900 3000 3100

Co

un

ts

50

100

150

200

250

300

Figure 9.38: (a) Zero-degree γ-ray beam profileobserved in the Ge detector, (b) FEPin detail

Due to electronic noise, quantum mechani-cal effects, and the original beam energy distri-bution, the full-energy-peak (FEP) energy spec-trum will have a nonzero width. Typically, the

fwhm of an incident γ-ray-beam energy distribu-tion, like the beams produced at HIγS, is dom-inated by the incident distribution. We maytherefore assume that the half of the beam-induced FEP above the energy centroid is areasonable approximation to the upper half ofthe incident-beam energy distribution; the lowerhalf of the FEP is distorted by the presence ofCompton-scattering events (see Fig. 9.38).

In the present method, first the centroid of theFEP is found and all counts below the centroidare removed. Then the remaining counts are mir-rored across the centroid, which results in an ap-proximately Gaussian beam profile. This sym-metric peak distribution I(E) may then be ad-justed for the detector’s response function ε(E)to approximate the incident beam flux and en-ergy distribution:

Φ(E) =I(E)ε(E)

where Φ(E) is the beam flux as a function of en-ergy (see Fig. 9.39).

(a) (b)

(c)

Figure 9.39: Illustration of the unfolding pro-cess. (a) The FEP with counts be-low the centroid removed. (b) The“chopped” FEP mirrored across thecentroid. (c) The approximate beamenergy distribution after accountingfor detector efficiency.

The detector efficiencies were simulated usinggeant4. For this example, efficiency simulationsassumed a mono-energetic, infinitely thin beam


striking a virtual detector 2 cm off the detectoraxis. The generated table of simulated detectorefficiencies was queried to obtain the needed val-ues for ε(E). Because it is computationally cum-bersome to generate efficiencies for each energybinning scheme one might encounter in a typi-cal data set, energies are rounded to the near-est value given in the data file before querying.In the examples shown, efficiencies were simu-lated in 100 keV intervals, which ensured thatany discrepancies between true efficiencies andtheir rounded approximations were negligible.

This process produces a spectrum qualita-tively similar to a Gaussian distribution, al-though the resulting isolated peak may be fit withany appropriate function. Figure 9.40 presentsa fit composed of two Gaussian distributions,summed and weighted by polynomial factors,

f(E) =ae−( E−E0

c )2

ε(E)+

b(E − E0)2

ε(E)e−(E−E0

d )2

,

where E0 represents the centroid energy, andε(E) represents a fit to the calculated or mea-sured detector efficiency. This function raises thetails of the Gaussian distribution, thus more ac-curately fitting the FEP.

Figure 9.40: Polynomial-weighted Gaussian dis-tributions summed to form a moreaccurate fitting function.

Using the root data-analysis package, we fitthis function to our example data (see Fig. 9.41).

Energy [keV]2750 2800 2850 2900 2950 3000

Co

un

ts

0

1000

2000

3000

4000

5000

6000

Figure 9.41: A fit of the summed-Gaussian func-tion.

Because the upper and lower halves of thegenerated FEP are almost symmetric, the fit wasonly applied to the upper half of the spectrum.The fit shown in Fig. 9.41 has a reduced χ2

red

of 0.96. For comparison, a simple Gaussian fitresults in a reduced goodness-of-fit statistic ofχ2

red = 1.77, indicating that the proposed modi-fication is reasonable for the given data and bin-ning scheme. Major discrepancies between theunfolded spectrum and the fitting function arelikely due to misestimations of the error of theunaltered data. For this example, we found thefit parameters to be a = 1972±12, b = 0.52±0.02,c = 46.2± 0.8 keV, and d = 64.2± 0.3 keV, withE0 = 2876.1 ± 0.2 keV. Note that these errorscorrespond to the fitting parameters alone, andthe centroid energy may have a higher error car-ried over from the initial peak-finding routine.(We estimate that the total centroid uncertaintyis approximately 25 keV for this example.) De-spite this error estimate, there is good agreementbetween the fitting function and the unfolded in-cident beam profile.

A program which employs this method of un-folding is freely available upon request.


9.4 Data Acquisition Hardware and Software Development

9.4.1 Status of the orca Data Acquisition System

M.A. Howe, J.F. Wilkerson, TUNL; M.G. Marino, J. Kasper, University of Washington,Seattle, WA

We describe additions and improvements to orca, an object-oriented real-time control andacquisition system. orca is currently in use in the development of several experiments in theUS, Canada, and Germany and is under continuous development. Support for a number ofnew VME, USB, and serial devices was added, an improved driver for the VME bridge chipon the VME controller single-board computer (SBC) was developed, and the SBC readoutsoftware was rewritten in C++ .

The orca (Object-oriented Real-time Con-trol and Acquisition) software system [How09] isdesigned for dynamically building flexible and ro-bust data acquisition systems. orca was firstused as the data acquisition system for the Sud-bury Neutrino Observatory (SNO) Neutral Cur-rent Detector (NCD) experiment and success-fully took production data continuously for threeyears until Nov 2006. After the NCD experimentwas decommissioned, orca was used in an al-pha counter to study “hot spots” on the removedNCD tubes, and during the last few months of theSNO experiment it was used in the SNO muon-tracking system. Since then the world-wide baseof orca systems has continued to grow.

The University of Washington’s (UW) Cen-ter for Experimental Nuclear Physics and Astro-physics is using orca in several test stands forthe development of the pre-spectrometer for KA-TRIN (the Karlsruhe Tritium Neutrino experi-ment), the commissioning of the KATRIN mainfocal-plane detector, the development of the KA-TRIN veto system, validation of the electronicsfor the Majorana experiment, and some SNO+development work. The UW radiology depart-ment has been using it for several years in thecharacterization of detectors for a small animalPET scanner.

In Germany, the Karlsruhe Institute of Tech-nology (KIT) is running orca in many KATRIN-related hardware development test stands, a tri-tium source development system, and the KA-TRIN pre-spectrometer.

UNC has several orca test stands running,both on campus and at KURF. Recently two sys-tems have been setup at TUNL.

At SNOLab, the SNO+ and HALO exper-iments have chosen orca as their production

DAQ system and are currently running severaldevelopment test stands.

There are also orca systems running atLANL, LBNL, and MIT.

The catalog of objects supported by orca

continues to grow. In the last year a numberof new VME cards were added.

• CAEN V965/965A 16/8-channel charge-to-digital converter.

• CAEN V775/775N 32/16-channel TDC.

• CAEN V260 Scaler with 16 independent24-bit counting channels at the maximuminput frequency of 100 MHz.

• CAEN V785/785N 32/16-channel peaksensing 12-bit ADC.

• SIS 3800 32-channel 32-bit scaler with amaximum counting frequency of 200 MHz(ECL and NIM) or 100 MHz (TTL), re-spectively.

• SIS 3820 32-channel 32-bit scaler with amaximum counting frequency of 250 MHzcounting rate (ECL and NIM) or 100 MHz(TTL). Only the generic scaler functionsare supported at this time.

• SIS 3302 eight-channel digitizer. The ver-sion implemented is the Gamma firmware,specifically version 0x1407, which sup-ports a finite-response filter-based trigger-ing, asynchronous readout of raw and/orcomputed (i.e. energy) digitizer informa-tion. It is of particular interest for detectorstudies with γ-ray tracking and strip detec-tors.


• SIS 3320 eight-channel 12-bit 250 MHz dig-itizer (currently in advanced state of devel-opment).

• SIS 3350 four-channel 12-bit digitizer,which can sample at up to 500 MSam-ples/sec.

• VHQ224LHV two-channel 4 kV high volt-age supply.

• VHQ4030HV four-channel 3 kV high volt-age supply.

The list of newly supported serial devices in-cludes:

• AMRel HV. Both one and two channel ver-sions are supported.

• A Varian turbo-pump system controller.

• iTrans Modbus Oxygen Sensor.

• KJL Ion Gauge Controller.

• CC4189 Temperature/Humidity Sensor.

• MKS PDR2000 controller for reading outtwo Baratron pressure sensors.

• Ami 286 cryogenic level controller. Thisdevice has been supported for more thata year, but a number of improvementsand alarm handling functions were recentlyadded.

• A variant of the MIT custom pulser, witha different range, was added.

For the KATRIN experiment, the version 4first- and second-level-trigger (FLT and SLT)electronics were finished and delivered to UNCfor testing. After orca support for the SLTand FLT cards was completed and verified, theelectronics were shipped to UW to be used inthe final commissioning of the KATRIN detec-tor. Recently an additional four FLTs and oneSLT, along with a mini-crate were sent from KITto UNC for further software testing and verifi-cation. orca support of the KATRIN version 4hardware is now considered essentially complete,pending any new issues discovered at UW.

For the SNO+ experiment, support for atotally new SNO crate controller/readout card

was added to orca. This card, the XL3, re-places the XL2 crate controllers used in the origi-nal SNO experiment. These field-programmable,gate array-based controllers communicate di-rectly with orca over sockets using TPC/IP. TheXL3 cards read out of each of the SNO crates andstream all data back to orca for storage, moni-toring, and re-broadcast to an event builder.

Major experiments such as KATRIN, Ma-

jorana, and HALO need specially developedhigh-level dialogs to ease the burden on the ex-periment’s operators. New dialogs were addedfor HALO and Majorana. These dialogs col-lect and display information, such as channel-by-channel rates and total rate over time, in oneplace so that the state of an experiment can bemonitored from one dialog. They also provide aninterface for mapping detector channels to elec-tronics channels. The existing dialog for the KA-TRIN experiment was also improved to includesystem pre-amp and crate-level views.

orca supports the use of LINUX-based singleboard computers (SBCs) as crate controllers forVME, cPCI, and IPE crates. The code that exe-cutes on these boards is managed by orca, thusproviding the transparent use of SBCs for dataread-out. This read-out code was originally writ-ten completely in C. In 2009 it was completelyrewritten in C++ to be fully object oriented. Itwas also refactored to put each of the device-read-out functions into separate files and thus into sep-arate objects. It is now much easier to extend thecode and to add support for new objects.

The SBCs use a Universe II VME bridge chipto provide access to VME address space. Theoriginal driver was somewhat limited, in that itdid not support DMA access. This driver wascompletely rewritten and now supports DMAand block-level transfers. The original softwareprogramming interface was retained.

A secondary application to orca, called or-

caroot, provides a tool kit to build analysis pro-grams that can decode raw orca data streams.As new objects are added to orca, new decodersare also added to orcaroot to fully supportroot analysis of data from the new objects.

[How09] Howe, M. A. et al., TUNL Progress Re-port, XLVIII, 179 (2009).


9.4.2 iorca, An iPad-based DAQ monitoring Tool

M.A. Howe, J.F. Wilkerson, TUNL

iorca is a mobile data-acquisition monitoring tool hosted on the iPad platform. Informa-tion that is to be monitored is placed into an SQL (structured query language) database byparticipating DAQ applications such as orca. It can then be retrieved using web-based URLqueries. Each query triggers a PHP script that returns the result in XML (extensible markuplanguage) format. iorca stores the result and displays it as needed. Multiple data-acquisitionsystems can be monitored by multiple iorca applications simultaneously.

With the introduction of highly mobile com-puting and display devices, such as Apple’siPhone and iPad, hand-held remote data acquisi-tion monitoring and control tools have suddenlybecome practical. One such application, iorca

(iPad-orca), has been developed at UNC for theiPad. iorca is designed as a mobile link be-tween remote users and orca data-acquisitionsystems (described in Sect. 9.4.1). It is written inObjective-C for the iOS operating system usingthe Cocoa and iTouch software frameworks. Theapplication design is object-oriented and highlymodular, making the addition of new features rel-atively simple. iorca is currently available onlyfor the iPad, although it will probably be pos-sible to provide a stripped-down version for theiPhone. iorca is only available direct from theUNC developers on an iPad-by-iPad basis usingApple’s ad hoc distribution system. For securityreasons, there are no plans for distribution viathe Apple Store.

iorca is designed around web-based accessto a central data repository in the form of amySQL database that is populated by orca sys-tems. Each iPad running iorca can monitormany orcas but each orca must individuallyopt-in to be monitored. This is done by includ-ing a special SQL object into the orca config-uration. If it is present and configured with theproper database location, username, and pass-word, then the monitored parameters are col-lected, organized into groups, and inserted intothe proper database tables. There is one tablefor the list of all running orca systems, one for1D histograms, one for run status, etc. Each ta-ble is predefined when the database is set up,and the format is known to the orca SQL ob-ject. For example, the table holding run statusholds all information associated with the currentrun state. A partial listing of that table is:

+-----------------------+------------+| Field | Type |+-----------------------+------------+| run_id | int(11) || machine_id | int(11) || runNumber | int(11) || state | int(11) || experiment | varchar(64)|| started | varchar(64)|

... ...| timeLimit | int(11) || repeatRun | tinyint(1) |+-----------------------+------------+

The key that links all the tables is a machineidentifier, which is, in turn, linked to the MACethernet address of each computer running orca.By using a particular MAC address, the identi-fier key for that system can be retrieved fromthe main table holding all machines, and an SQLquery can be constructed to extract data fromany table or combination of tables.

Queries to the database are done via aweb-site PHP hypertext-preprocessor script thattakes an SQL select command and a tag as a pa-rameter. For example, a URL to retrieve datafrom the run status table would be somethinglike:

http://.../queryXML ? tag=runstatus &query = select * from runs wheremachine_id = 12

where the ellipsis is replaced with a path tothe web site holding the query XML script. Thequery script can process only one generic ‘select’-type query at a time, and any attempt to executemultiple commands or other types of commandsis rejected, in order to help prevent SQL injectionattacks. Valid queries are sent to the SQL server,and the response is formatted into XML and re-turned to iorca. The final XML packet from the


above query example would look something likethe following:

<?xml version="1.0"?><runstatus><run_id> 8 </run_id><machine_id> 12 </machine_id><run> 2413 </run><subrun> 0 </subrun><state> 1 </state><experiment> Katrin </experiment><started> 72010 11:48:15 </started>.. truncated for brevity ..<timeLimit>0</timeLimit><repeatRun>0</repeatRun>

</runstatus>

iorca does not use the XML directly butconverts it into standard iOS dictionary objects,where each data item is stored locally with a keythat is the same as the tag associated with thedata field. Knowing the tag value and the datakeys, each display in iorca can extract what-ever data are needed for a particular page. Eachdisplay page is refreshed whenever a query is re-turned from the database. The query rate is vari-able but typically on the order of once every twoseconds. The round trip time for a query de-pends on the network but is generally a few tensof milliseconds. The total bandwidth required islimited by having iorca request only the datathat are needed for a particular display.

The user interface in iorca is arrangedaround a series of display pages. The pages thatare currently defined include the following:

• Database Setup: Here the user sets the ac-cess path for the database and the PHPquery script.

• System Statistics: Displays a query-and-response timing histogram and total accesscounts. Useful for debugging.

• System List: A list of all participatingorca systems connected to a database.

Once a system is selected on this page, it isthe working system for all other displays.

• Summary Page: Shows a summary of runstatus and a list of all orca alarms. Alsoprovides links to the data list, controlpages, and the experiment-specific page.

• Data List: Shows a list of all available his-togram data being accumulated. Also pro-vides a link to the plot page.

• Plot Page: Displays a particular histogram.

• Run Control: Provides access to orca’srun control functions. Password protected.

• Command Center: Allows execution of anarbitrary command via orca’s commandcenter. Password protected.

• Experiment Specific: Different for each ex-periment. For example, the KATRIN dis-play shows information associated with themain focal plane detector.

In addition to displaying information, iorca

has some control functions. This is accomplishedvia a direct socket connection into orca’s com-mand center. The socket is opened only whena page that uses this facility is visible; other-wise it is closed. The control pages are protectedwith each iorca’s expert function password. Themost extensive use of the control functionalityis on the run-control page, where users can re-motely operate all of orca’s run control func-tions. There is also a generic command page thatallows the input of Objective-C commands intothe orca command center. A user with knowl-edge of Objective-C and orca’s internals can ex-ecute almost any method in the orca code base.

iorca has so far been a test of what kind offunctionality would be desirable in a mobile mon-itoring and control application. As it is refinedand new functionality is added, it is anticipatedthat it could find a place in monitoring majorexperiments such as KATRIN and Majorana.


9.4.3 Validation and Characterization of the KATRIN DAQ ElectronicsSystem

D.G. Phillips II, T.J. Corona, M.A. Howe, J.F. Wilkerson, TUNL; T. Bergmann, A. Kop-

mann, Karlsruhe Institute of Technology - Institute for Data Processing and Electronics, Karlsruhe,Germany

This report describes characterization and validation studies performed on the hardware and

software data acquisition (DAQ) systems of the KATRIN experiment. It also describes script

development for the characterization and validation studies, which has been made possible

using orca’s scripting language known as orcaScript.

A next generation tritium β-decay experi-ment known as the KArlsruhe TRItium Neu-trino (KATRIN) experiment is currently un-der construction at the Forschungszentrum Karl-sruhe campus of the Karlsruhe Institute of Tech-nology (KIT). Electronics systems developed atKIT’s Institute of Data Processing and Electron-ics (IPE) [Kop08] will serve as readout for the148-pixel detector and the 32-channel veto sys-tem. The data acquisition software utilized bythe KATRIN experiment is a Mac OS X appli-cation called orca (Object-oriented Real-timeControl and Acquisition) [How08]. A block dia-gram of the data flow in the KATRIN experimentis given in Fig. 1.

A wide range of event rates will be expectedin the KATRIN experiment, where the normalbackground rate will be as low as 10 mHz andthe calibration rate will be as high as 1 MHz.One key point of validation is to verify thatthe KATRIN DAQ system can handle this widerange of data rates. With production data tak-ing set to begin in 2012, a thorough characteriza-tion and validation of the KATRIN DAQ systemis needed. This task will be completed by theTUNL group. A brief list of the hardware andsoftware that is undergoing tests includes the fol-lowing:

• DAQ Hardware: The field-programmable-gate-array (FPGA)-based electronics sys-tems are each divided into a second leveltrigger (SLT) card and a group of first leveltrigger (FLT) cards. The SLT card servesas the controller and module for readoutof the FLT cards. In the earlier Mark3 (Mk3) version, the FLT cards had anADC sample rate of 10 MHz with a pre-cision of 12 bits [Wil09]. The FPGAs on

the current Mark 4 (Mk4) FLT card ver-sion that will be used in the experimenthave a larger memory than their Mk3 coun-terparts. This allows for all DAQ modesto be available without reprogramming thecard. Each Mk4 FLT cards has 24 channels(ADCs), which is two more ADCs than theMk3 FLT card. Data can be acquired fromthe Mk4 FLT cards in single-event mode(energy and time stamp), waveform mode(digitized signal) or histogramming mode(energy histograms for each channel). TheADCs in the Mk4 FLT cards can sampleat 20 MHz (14 bits), and the histograms inhistogram mode have increased from 512bins to 2048 bins, where each bin is 32 bits(vs. the Mk3 value of 24 bits). The Mk4SLT cards have been upgraded to containa single-board computer running LINUX(OpenSUSE version 10.3).

Focal Plane Detector

(148 channels)

T2 Source,

Transport Section &

Spectrometer System

KATRIN DAQ Electronics

Slow Controls(Labview)

Data Storage

ORCA ORCARoot Root Tree

UsersZEUSDatabase

D.R.S.C.R.

D.R. = Data RecordsC.R. = Control RequestsS.C.R. = Slow Controls Records

S.C.R.

C.R.

Muon Veto(32 channels)

Figure 9.42: (Color online) KATRIN data flow.

• DAQ Software: orca software is usedto control and acquire data from theIPE’s FPGA-based electronics. orca uses


software-bus concepts to encapsulate hard-ware objects and data-acquisition con-cepts into software objects that have user-friendly graphical interfaces. In addition tothe hardware used in the KATRIN valida-tion and characterization tests, orca pro-vides control to a large number of otherhardware devices. An experimental setupis assembled in orca’s interface and dy-namically configured at runtime, withoutany re-compilation of orca required. Ex-tensive software validation of all KATRIN’srequired objects is on-going.

• DAQ Scripting Languages: orca containstwo C-like scripting languages [How09].The first is a C-like language with Objec-tiveC extensions called orcaScript. TheorcaScript language can access any soft-ware method in the orca code base, thusallowing for full automation of any ob-ject within orca. The other C-like script-ing language, FilterScript, can be used toconstruct data-filtering and event-buildingroutines within orca.

• Mk4 Linearity Testing: A characterizationof the linearity of the Mk4 ADCs has beenperformed to determine whether they meetthe accuracy needed by the KATRIN ex-periment for energy measurements. En-couraging results from linearity testing re-veal a low integral non-linearity rangingfrom 0.01% to 0.04% across a sample of 110Mk4 ADCs. orcaScript was used to auto-mate all of the validation and characteri-zation studies, which in itself constitutes arigorous test of the scripting languages.

• Mk4 Stability Testing: The long-termsingle-channel stability of the Mk4 DAQsystem has been measured over a period ofsix days. By pulsing individual channelswith KATRIN-signal-like pin-diode wave-forms, the Gaussian mean of the resultantenergy spectrum has been shown to varyby only ±2%. This result is well withinKATRIN’s requirements. Long-term testsusing multiple channels under realistic con-ditions are presently being performed.

• Mk4 Efficiency Testing: Characterizationof the dead time in each DAQ measurementmode must be made in order to deduce theevent rate at which counting efficiency ofthe Mk4 electronics drops below nominal.Tests carried out on single Mk4 FLT chan-nels using random pin-diode pulses with adecay time of 8 μs have shown that thecounting efficiency drops below 100% at 375Hz for waveform mode. A similar test per-formed in single-event mode showed thatthe counting efficiency drops below 100%at 68 KHz using random pin-diode pulseswith a decay time of 2 μs. A more thor-ough study using multiple channels andKATRIN-like pin-diode pulses with a de-cay time of 400 μs is currently underway.

• Future Testing: Many tests will be per-formed in the coming months to com-plete the validation and characterizationof the Mk4 DAQ system. Evaluationof correct functionality between the Mk4histogramming-mode software and hard-ware will be completed to ensure that his-togramming is fully functional and avail-able for periods of high data rates. A com-parison of the Mk4 hardware trapezoidalfilter algorithm (TFA) with a software fil-ter will be made to determine whether theTFA is functioning as expected. Cross-talkbetween FLT channels will also be evalu-ated.

[How08] M. Howe, M. Marino, and J. F. Wilker-son, IEEE Nucl. Sci. Symp. Conf. Rec.,NSS ’08, 3562 (2008).

[How09] M. Howe et al., TUNL Progress Report,XLVIII, 178 (2009).

[Kop08] A. Kopmann et al., IEEE Nucl. Sci.Symp. Conf. Rec., NSS ’08, 3186(2008).

[Wil09] J. Wilkerson et al., TUNL Progress Re-port, XLVIII, 38 (2009).


9.4.4 Installation and Commissioning of the MALBEK Low-BackgroundBEGe Detector Data Acquisition System

G.K. Giovanetti, P. Finnerty, R. Henning, M.A. Howe, J.F. Wilkerson, TUNL; M.G.

Marino, M.L. Miller, University of Washington, Seattle, WA

We describe the current status of the data-acquisition system for the Majorana low-background

BEGe detector at the Kimballton Underground Research Facility, as well as the progress we

have made in evaluating waveform digitizers for use with the Majorana Demonstrator.

As part of the research and developmentefforts of the Majorana collaboration (seeSect. 2.1.1), we have deployed a prototype 450gbroad-energy germanium (BEGe) detector atthe Kimballton Underground Research Facility(KURF) in Ripplemeade, VA. The Majorana

low-background broad-energy germanium detec-tor at KURF (MALBEK) will investigate point-contact, modified BEGe detector performance,explore the sub-keV energy spectrum in a searchfor light weakly interacting massive particles(WIMPs), and serve as a test-bed for electron-ics, software, and data analysis techniques for theMajorana Demonstrator (see Sects. 2.1.5and 2.1.6).

After initial detector characterization andtesting, we installed the MALBEK detector anddata acquisition system underground at KURFin January 2010. The MALBEK data acquisitionsystem uses the Object-oriented Real-time Con-trol and Acquisition (orca) system, an object-oriented data-acquisition application that pro-vides an easy-to-use, graphical interface for ma-nipulation of experimental hardware and datastreams (see Sect. 9.4.1). The orca-managedslow control system is used to control the de-tector and to monitor the detector environment.The high voltage bias for MALBEK is providedby a VME based iseg VHQ224L module allow-ing remote voltage and current monitoring andcontrol. Another iseg VME bias supply, the 12channel VHSC020, will be used for the activemuon-veto photomultiplier tubes. Liquid nitro-gen levels in the MALBEK detector dewar andthe radon purge system’s nitrogen boil-off de-war are monitored and controlled remotely usingan American Magnetics 286 multi-channel liq-uid level controller. Seismic monitoring to vetomicrophonic induced signals in the detector isdone using a GLI Interactive motion node. Seis-

mic data from vibrations exceeding a user-definedvalue are inserted into the data stream for of-fline analysis. Oxygen levels in the trailer hous-ing the MALBEK detector and the trailer hous-ing the DAQ electronics and computer are mon-itored by iTrans gas monitors. Excursions fromatmospheric concentrations of oxygen sound a lo-cal audible alarm and an email notification. Allof these systems have been running stably sinceinstallation.

Figure 9.43: Block diagram of the MALBEKDAQ

For fast DAQ, two different VME-based dig-itizers have been used to collect the analog sig-nals from the MALBEK low-noise, pulse-reset,charge-sensitive preamplifier and the active muon


veto. Both are candidates for use with the Ma-

jorana Demonstrator. The Struck Innova-tive Systeme 3302 ADC is a VME-based, 8 chan-nel, 16-bit digitizer capable of digitization ratesup to 100 MHz. The SIS3302 incorporates botha fast finite impulse response (FIR) trapezoidaltriggering filter capable of triggering on low am-plitude signals and a slow FIR trapezoidal fil-ter for determining signal height. The SIS3302supports data readout in parallel with acquisi-tion and signal decimation. The Gretina MarkIV is a 10 channel, 14 bit digitizer and dig-ital signal processing board developed for theGamma Ray Energy Tracking In-Beam NuclearArray (GRETINA) [And07]. The Mark IV usesa leading-edge-discriminator trigger and has abuilt-in trapezoidal energy filter. An analog spec-troscopy amplifier and an Ortec 927 MCA areused to collect energy spectra.

orca provides a catalog of fully encap-sulated objects representing hardware, data-readout tasks, data analysis modules, and controlmodules. The orca data stream can be mon-itored in real time using orca’s data monitor-ing object. In addition, by using orcaScript

(a small interpreted programming language) op-erators can implement data-stream filtering andreal-time event building and data plotting. or-

caScript also allows for the automation of runcontrol and run-time experimental parameters.orca is self-monitoring, sending email notifi-cation and alarms to operators based on user-configurable preferences.

During a run, data readout from the VME busis performed by a Concurrent Technologies sin-gle board computer that interfaces directly withthe data acquisition computer. Digitized wave-forms are filtered in real time using orcaScript

to eliminate noise events related to preamplifierresets, thus reducing the size of data files savedto disk. Another orcaScript continuously pollsthe liquid nitrogen levels of the detector dewarand the status of the other slow control systemsso as to safely unbias the detector in the eventof a prolonged power failure or an increase in de-tector temperature.

In order to periodically test the trigger effi-ciency of the digitizers and the gain stability ofthe DAQ, an Agilent 33220A arbitrary waveformgenerator coupled to a set of step attenuatorsis used to generate test signals of varying am-plitude and shape at the input of the preampli-fier. This is patterned on a system implementedby the University of Washington at the SoudanUnderground Laboratory. The waveform gen-erator interfaces with the data acquisition com-puter via USB-2.0. The step attenuators can becontrolled via an Acromag IP408 I/O module, a32-channel digital input/output module mountedon an Acromag AVME 9630 Carrier Board thattransfers signals between the IP408 and the SBCcontrolled VME bus. An orcaScript auto-mates this process, varying test pulse sizes andrise times to determine the trigger efficiency ofthe digitizers down to sub-keV energies and forevent rise times as fast as 350 ns. These trigger-efficiency studies are used to determine the abil-ity of the digitizers to meet the Demonstrator

low-energy physics goals.The MALBEK DAQ was installed in Jan-

uary 2010 and has been taking data with theSIS3302 since March. After initial data runs withthe card, we recognized several bugs in orca

related to data readout and card triggering, aswell as a limitation in the length of the pre-trigger waveform the card can digitize. This hin-dered the performance of offline energy calcu-lations. We have now fixed these orca prob-lems and are working with engineers at StruckInnovative Systeme to increase the length of theSIS3302 pre-trigger delay. We are also in con-tact with the Gretina engineers, requesting thatthey implement a fast trapezoidal filter to im-prove the Gretina digitizer’s low-energy trigger-ing. The MALBEK DAQ is currently taking dataat KURF where we continue our evaluation of theSIS3302 and the Gretina Mark IV digitizer.

[And07] J. T. Anderson et al., In NuclearScience Symposium Conference Record,NSS ’07. IEEE, p. 1751, 2007.


9.4.5 Characterization of DAQ Hardware for the HALO Experiment

T.J. Corona, M. Howe, N. Mykins, J.F. Wilkerson, TUNL

The Helium and Lead Observatory is a supernova neutrino detector in development at SNO-

LAB, a deep-underground laboratory located near Sudbury, Ontario. Signals will be read out

using 3He detectors previously developed for the SNO neutral-current detectors. The mod-

ules include custom VME shaping-discriminating-ADC boards used in tandem with a custom

VME trigger board. Configuration and characterization of these boards has been completed.

The signal readout system for the HeliumAnd Lead Observatory (HALO) consists ofcustom-built 8-channel shaper/ADC boards andcustom-built trigger boards, designed for use ina standard 6U VME crate. The shaper/ADCboards will read out signals sent from the 3Heproportional detectors. A custom-built triggercard will be used to attach a time-stamp to eventsand to lock and unlock channels for data taking.

The shaper/ADC boards were originally de-signed and built at the University of Washington(UW) for use in the emiT, SNO and KATRINexperiments. Each channel has its own gain andlevel discriminator and has a sample-and-holdADC allowing the energy of input pulses to be de-termined. The boards are designed for data read-out and parameter switching via Object-orientedReal-time Control and Acquisition (orca), adata acquisition program (see Sect. 9.4.1).

The shaper/ADC boards can be configuredto operate in two modes: continuous and multi-board. During continuous mode, each channel

reads in and digitizes a signal independent ofthe other channels on the board. In multi-boardmode, once a single channel reads in a signal, itlocks out the other channels until the data is re-leased from the card by the orca system. Whenused with the custom-built trigger card, multipleshaper/ADC boards can operate in multi-boardmode as though all of the channels were on oneboard; a lock-out signal from a single channelwill prevent all of the channels from all of theboards from acquiring data until the lock-out sig-nal is released by orca. This allows unique timestamping of each event.

Characterization and testing of theshaper/ADC boards in both continuous andmulti-board mode has been completed. Thelinearity and gain testing on each channel of ev-ery card has been performed, the sensitivity ofthe individual cards to pulses with different rise-times has been determined, and general testing ofeach card’s inhibit logic has been accomplished.

Appendices

• Graduate Degrees Awarded

• Publications

• Invited Talks, Seminars, and Colloquia

• Professional Service Activities

224 Appendices TUNL XLIX 2009–10

A.1 Graduate Degrees Awarded

Ph.D. Degrees

1. Changchun Sun, Characterizations and Diagnostics of Compton Light Source,Duke University, November 2009,Supervisor: Y.K. Wu.

2. Andrii Chyzh, Neutron Capture Measurements on 157Gd and 89Y at DANCE,North Carolina State University, December 2009,Supervisor: G.E. Mitchell.

3. Peter F. Bertone, Indirect Approaches to Constraining the Best Estimate of the AstrophysicalS-factor for Proton Radiative Capture on 14N,University of North Carolina at Chapel Hill, December 2009,Supervisor: A.E. Champagne.

4. Joe Newton, Hydrogen Burning of 170,University of North Carolina at Chapel Hill, December 2009,Supervisor: C. Iliadis.

5. Jianfeng Zhang, Study on Beam Polarization and its Applications on Electron Storage Ring,University of Science and Technology of China, December 2009,Supervisors: Y.K. Wu (Duke), H. Xu (USTC), A.W. Chao (SLAC).

6. Michelle L. Leber, Monte Carlo Calculations of the Intrinsic Detector Backgrounds for theKarlsruhe Tritium Neutrino Experiment,Univerisity of Washington, December 2009,Supervisor: J.F. Wilkerson.

7. Wei Chen, A Measurement of the Differential Cross Section for the Reaction γn → π−p fromDeuterium,Duke University, February 2010,Superivsor: H. Gao.

8. Xing Zong, First Study of Three-body Photodisintegration of 3He with Double Polarizations atHIγS,Duke University, February 2010,Supervisor: H. Gao.

9. Mary F. Kidd, Double-Beta Decay of 150Nd to Excited Final States,Duke University, March 2010,Supervisor: W. Tornow.

10. Christopher O’Shaughnessy, Development of a Precision Neutron Lifetime Measurement: Mag-netic Trapping of Ultracold Neutrons,North Carolina State University, April 2010,Supervisor: P. Huffman.

11. Bayarbadrakh Baramsai, Neutron Capture Reactions on Gadolinium Isotopes,North Carolina State University, May 2010,Supervisors: G.E. Mitchell and U. Agvaanluvsan.

12. Richard Longland, Investigation of s-process Neutron Source 22Ne+a,University of North Carolina at Chapel Hill, May 2010,Supervisor: C. Iliadis.

TUNL XLIX 2009–10 Appendices 225

13. Brent A. Perdue, Measurements of the Absolute Cross Section of the Three-Body Photodisin-tegration of 3He Between Eγ=11.4 MeV and 14.7 MeV at HIγS,Duke University, July 2010,Supervisor: H.R. Weller.

14. Xin Qian, Measurement of Single Spin Asymmetry in n↑(e, e′π±)X on Transversely Polarized3He,Duke University, September 2010,Supervisor: H. Gao.

M.S. and M.A. Degrees

1. Sean Patrick MacMullin, Background Reduction and Detector Characterization at the KURFLow-Background Counting Facility,University of North Carolina at Chapel Hill, August 2009,Supervisor: R. Henning.

2. David Ticehurst, Development of a Gd-loaded Neutron Detector,University of North Carolina at Chapel Hill, December 2009,Supervisor: H.J. Karwowski.

3. Chris Cottrell, Deuterium Source Development at the PULSTAR Reactor,North Carolina State University, December 2009,Supervisor: A. R. Young.


A.2 Publications

The publications co-authored by members of TUNL research groups between September 2009 andAugust 2010 are tabulated in Table A.1. The papers in refereed journals by TUNL research groupsare listed below in chronological order.

Type No.Refereed Journal Papers 47Conference Proceeding Papers 18

Table A.1: Summary of TUNL publications from September 2009 through August 2010. Among 47refereed journal papers, 5 were letters.

Journal Articles Published

1. Chiral Dynamics in Photo-Pion Physics: Theory, Experiment, and Future Studies at the HIγSFacility, A. M. Bernstein, M. W. Ahmed, S. Stave, Y. K. Wu, Annu. Rev. Nucl. Part. Sci.,59, 115-144 (2009).

2. Competition Between Excited Core States and 1 hω Single-Particle Excitations at ComparableEnergies in 207Pb from Photon Scattering, N. Pietralla, T. C. Li, M. Fritzsche, M. W. Ahmed,T. Ahn A. Costin, J. Enders, J. Li, S. Muller, P. von Neumann-Cosel, I. V. Pinayev, V. Yu.Ponomarev, D. Savran, A. P. Tonchev, W. Tornow, H. R. Weller, V. Werner, Y. K. Wu, andA. Zilges, Phys. Lett. B, 681, 134-138 (2009).

3. Direct Measurements of (p, γ) Cross-Sections at Astrophysical Energies Using RadioactiveBeams and the Daresbury Recoil Separator, D. W. Bardayan, K. A. Chipps, R. P. Fitzgerald,J. C. Blackmon, K. Y. Chae, A. E. Champagne, U. Greife, R. Hatarik, R. L. Kozub, C. Matei,B. H. Moazen, C. D. Nesaraja, S. D. Pain, W. A. Peters, S. T. Pittman, J. F. Shriner, Jr., andM. S. Smith, Eur. Phys. J. A, 42, 457, (2009).

4. Investigation of Solid D-2, O-2, and CD4 for Ultracold Neutron Production, F. Atchison, B.Blau, K. Bodek, B. van den Brandt, T. Brys, M. Daum, P. Fierlinger, A. Frei, P. Geltenbort,P. Hautle, R. Henneck, S. Heule, A. Holley, M. Kasprzak, K. Kirch, A. Knecht, J. A. Konter,M. Kuzniak, C.-Y. Liu, C. L. Morris, A. Pichlmeaier, C. Plonka, Y. Pokotilovski, A. Saunders,Y. Shin, D. Tortorella, M. Wohlmuther, A. R. Young, J. Zejma, G. Zsigmond, Nucl. Instrum.and Methods in Phys. Res. A, 611, 252-255 (2009).

5. Measurement of the Cosmic Ray and Neutrino-induced Muon Flux at the Sudbury NeutrinoObservatory, B. Aharmim et al. [SNO Collaboration including M.A. Howe and J.F. Wilkerson],Phys. Rev. D, 80, 012001, (2009).

6. Measurement of the Neutrino-spin Correlation Parameter γ in Neutron Decay Using UltracoldNeutrons, W. S. Wilburn, V. Cirigliano, A. Klein, M. F. Makela, P. L. McGaughey, C. L.Morris, J. Ramsey, A. Salas-Bacci, A. Saunders, L. J. Broussard, and A. R. Young, RevistaMexicana de Fısica Supplemento 55, 119 (2009).

7. Measurement of the Solar Neutrino Capture Rate with Gallium Metal. III. Results for the20022007 Data-taking Period, J. N. Abdurashitov et al. [SAGE Collaboration including J.F.Wilkerson], Phys. Rev. C, 80, 015807, (2009).

8. Nab: Measurement Principles, Apparatus and Uncertainties, D. Pocanic, R. Alarcon, L. P.Alonzi, S. Baessler, S. Balascula, J. D. Bowman, M. A. Bychkov, J. Byrne, J. R. Calarco, V.Cianciolo, C. Crawford, E. Friez, M. T. Gericke, G. L. Greene, R. K. Grzywacz, V. Gudkov, F.W. Hersman, A. Klein, J. Martin, S. A. Page, A. Palladino, S. I. Pentilla, K. P. Rykaczewski,W. S. Wilburn, A. R. Young, G. R. Young, Nucl. Instrum. and Meth. in Phys. Res. A, 611,211-215 (2009).


9. New Measurements of the European Muon Collaboration Effect in Very Light Nuclei, J. Seelyet al. (including D. Dutta, H. Gao, X. Qian, S. Tajima, and X. Zheng, Phys. Rev. Lett. 103,202301, (2009).

10. Photoexcitation of Astrophysically Important States in 26Mg, R. Longland, C. Iliadis, G. Rusev,A. P. Tonchev, R. J. deBoer, J. Gorres, and M. Wiescher, Phys. Rev. C, 80, 055803 (2009).

11. Reconstruction of a Radiation Point Sources Radial Location Using Goodness-of-Fit Test onSpectra Obtained from an HPGe Detector, L. T. Evans, K. Andre, R. De, R. Henning, and E.D. Morgan, Nucl. Instrum. Methods B, 267, 3688, (2009).

12. The BLAST Experiment, D. Hasell, et al. (including D. Dutta, and H. Gao), Nucl. Instrum.Methods A, 603, 247-262, (2009).

13. The Extraction of φ−N Total Cross Section from d(γ, pK+K−)n, X. Qian, W. Chen, H. Gao,et al., Phys. Lett. B, 680, 417, (2009).

14. The KamLAND Full-Volume Calibration System, B. E. Berger et al. (KamLAND Collabora-tion including W. Tornow, H. J. Karwowski, and D. M. Markoff), JINST, 4, P04017 (2009).

15. Transmission-based Detection of Nuclides with Nuclear Resonance Fluorescence Using a Quasi-monoenergetic Photon Source, C. A. Hagmann, J. M. Hall, M. S. Johnson, D. P. McNabb, J.H. Kelley, C. Huibregtse, E. Kwan, G. Rusev, and A. P. Tonchev, J. Appl. Phys., 106, 084901(2009).

16. Water Cooled, In-cavity Apertures for High Power Operation of FEL Oscillators, S. Huang, J.Li, Y. K. Wu, Nucl. Instrum. Methods A, 606, 762-769, (2009).

17. Measuring the Neutron Lifetime Using Magnetically Trapped Neutrons, C.M. O’Shaughnessy,R. Golub, K.W. Schelhammer, P.R. Huffman et al., Nucl. Instrum. Methods A, 611, 171-175(2009).

18. Experimental Searches for the Neutron Electric Dipole Moment, S.K. Lamoreaux and R. Golub,J. Phys. G 36, 104002 (2009).

19. A High Pressure Polarized 3He Gas Target for the High Intensity Gamma Source Facility atthe Duke Free Electron Laser Laboratory, Q. Ye, G. Laskaris, W. Chen, H. Gao, W. Zheng, X.Zong, T. Averett, G. D. Cates, W. A. Tobias , Eur. Phys. J. A, 44, 55, (2010).

20. Charged-particle Thermonuclear Reaction Rates: I. Monte Carlo Method and Statistical Dis-tributions, R. Longland, C. Iliadis, A. E. Champagne, J. R. Newton, C. Ugalde, A. Coc, andR. Fitzgerald, Nucl. Phys. A, 841, 1-30, (2010).

21. Charged-particle Thermonuclear Reaction Rates: II. Tables and Graphs of Reaction Rates andProbability Density Functions, C. Iliadis, R. Longland, A. E. Champagne, A. Coc, and R.Fitzgerald, Nucl. Phys. A, 841, 31-250, (2010).

22. Charged-particle Thermonuclear Reaction Rates: III. Nuclear Physics Input, C. Iliadis, R.Longland, A. E. Champagne, and A. Coc, Nucl. Phys. A, 841, 251-322, (2010).

23. Charged-particle Thermonuclear Reaction Rates: IV. Comparison to Previous Work, C. Iliadis,R. Longland, A. E. Champagne, and A. Coc, Nucl. Phys. A, 841, 323-388, (2010).

24. Corrections for the Polarization Dependent Efficiency and New Neutron-proton AnalyzingPower Data at 7.6 MeV, G. J. Weisel, R. T. Braun, and W. Tornow, Phys. Rev. C, 82,027001 (2010).

25. Determination of the E1 Component of the Low-Energy 12C(α, γ)16O Cross Section, X. D.Tang, K. E. Rehm, C. R. Ahamd, C. Brune, A. E. Champagne, J. P. Greene, A. Hecht, D.J. Henderson, R. V. F. Janssens, C. L. Jiang, L. Jisonna, D. Kahl, E. F. Moore, M. Notani,R. C. Pardo, N. Patel, M. Paul, G. Savard, J. P. Schiffer, R. E. Segel, S. Sinha, and A. H.Wuosmaa, Phys. Rev. C, 81, 045809, (2010).


26. Differential Cross Section for Neutron Scattering from 209Bi at 37 MeV and the Weak Particle-core Coupling, Z. Zhou, X. Ruan, Y. Du, B. Qi, H. Tang, H. Xia, R. L. Walter, R. L. Braun,C. R. Howell, W. Tornow, G. J. Weisel, M. Dupuis, J. P. Delaroache, Z. Chen, Z. Chen, Y.Chen, Phys. Rev. C, 82, 024601, (2010).

27. Dipole Strength in 139La Below the Neutron-separation Energy, A. Makinaga, R. Schwengner,G. Rusev, F. Donau, D. Bremmerer, R. Beyer, P. Crespo, M. Erhard, A. R. Junghans, J. Klug,K. Kosev, C. Nair, K. D. Schilling, and A. Wagner, Phys. Rev. C, 82, 024314, (2010).

28. Dipole strength in 144Sm Studied via (γ, n), (γ, p), and (γ, α) Reactions, C. Nair, A. R. Jung-hans, M. Erhard, D. Bemmerer, R. Beyer, E. Grosse, K. Kosev, M. Marta, G. Rusev, K. D.Schilling, R. Schwengner, and A. Wagner, Phys. Rev. C, 81, 055806, (2010).

29. E1 strength in 208Pb Within the Shell Model, R. Schwengner, R. Massarczyk, B. A. Brown, R.Beyer, F. Donau, M. Erhard, E. Grosse, A. R. Junghans, K. Kosev, C. Nair, G. Rusev, K. D.Schilling, and A. Wagner, Phys. Rev. C, 81, 054315, (2010).

30. Experimental Study of the A(e,e’π+) Reaction on 1H, 2H, 12C, 27Al, 63Cu and 197Au, Q. Xin,et al. (including D. Dutta, and H. Gao), Phys. Rev. C, 81, 055209, (2010).

31. g-factor Measurements at RISING: The Cases of 127Sn and 128Sn, L. Atanasova, et al. (in-cluding G. Rusev), Europhys. Lett. 91, 42001, (2010).

32. Improved Results for the 2H(d,n)3He Transverse Vector Polarization-Transfer Coefficient Ky′y (0◦)

at Low Energies, C. D. Roper, T. B. Clegg, J. D. Dunham, A. J. Mendez, W. Tornow, R. L.Walter, Few-Body Syst., 47, 177-184, (2010).

33. Level Densities and γ-ray Strength Functions in Sn Istopes, H. K. Toft, A. C. Larsen, U.Agvaanluvsan, A. Burger, M. Guttormsen, G. E. Mitchell, H. T. Nyhus, A. Schiller, S. Siem,N. U. H. Syed, and A. Voinov, Phys. Rev. C, 81, 064311, (2010).

34. Measurement of 17O(p, γ)18F Between the Narrow Resonances at Er=193 and 519 keV, J.R. Newton, C. Iliadis, A. E. Champagne, J. M. Cesaratto, S. Daigle, and R. Longland, Phys.Rev. C, 81, 045801, (2010).

35. (n,2n) and (n,3n) Cross Sections of Neutron-Induced Reactions on 150Sm for En from Thresh-old to 35 MeV, D. Dashdorj, G. E. Mitchell, T. Kawano, J. A. Becker, M. Devlin, N. Fotiades,R. O. Nelson, and S. Kunieda, Nucl. Instrum. Meth. in Phys. B, 268, 114, (2010).

36. Neutron-deuteron Analyzing Power Data at 19.0 MeV, G. J. Weisel, W. Tornow, B. J. CroweIII, A. S. Crowell, J. H. Esterline, C. R. Howell, J. H. Kelley, R. A. Macri, R. S. Pedroni, R.L. Walter, and H. Wita�la, Phys. Rev. C, 81, 024003, (2010).

37. Photoexcitation of Astrophysically Important States in 26Mg, R. J. deBoer, M. Wiescher, J.Gorres, R. Longland, C. Iliadis, G. Rusev, and A. P. Tonchev, Submitted to Phys. Rev. C.,82, 025802 (2010).

38. Production of Radioactive Isotopes Through Cosmic Muon Spallation in KamLAND, (TheKamLAND Collaboration, including H. J. Karwowski, D. M. Markoff, W. Tornow), Phys.Rev. C, 81, 025807, (2010).

39. Ratio of Germanium Detector Peak Efficiencies at Photon Energies of 4.4 and 11.7 MeV:Experiment Versus Simulation, S. Carson, C. Iliadis, J. Cesaratto, A. Champagne, L. Downen,M. Ivanovic, J. Kelley, R. Longland, J. R. Newton, G. Rusev, and A. P. Tonchev, Nucl.Instrum. Methods A, 618, 190, (2010).

40. Re-analysis of Recent Neutron Diffusion and Transmission Measurements in Nuclear Graphite,C. R. Gould, A. I. Hawari, and E. I. Sharapov, Nucl. Sci. Eng., 165, 200-209, (2010).

41. Resonance Strength in 22Ne(p,g)23Na From Depth Profiling in Aluminum, R. Longland, C.Iliadis, J. M. Cesaratto, A. E. Champagne, S. Daigle, J. R. Newton, R. Fitzgerald, Phys. Rev.C, 81, 055804, (2010).


42. Spectral Structure of the Pygmy Dipole Resonance, A. P. Tonchev, S. Hammond, J. H. Kelley,H. Lenske, W. Tornow, and N. Tsoneva, Phys. Rev. Lett., 104, 072501 (2010).

43. Substantial Increase in Acceleration Potential of Pyroelectric Crystals, W. Tornow, S. M. Ly-nam, and S. M. Shafroth, J. Appl. Phys., 107, 063302, (2010).

44. Search for a Bosonic Component in the Neutrino Wave Function, W. Tornow, Nucl. Phys. A,844, 57C, (2010).

45. A Feasibility Study Using Radiochromic Films for Fast Neutron 2D Passive Dosimetry, S.L.Brady, R. Gunasingha, T.T. Yoshizumi, C.R. Howell, A.S. Crowell, B. Fallin, A.P. Tonchev,and M. Dewhirst, Phys. Med. Biol., 55, 4977 (2010).

46. Development of a Superconducting Transverse Holding Magnet for the HIγS Frozen Spin Tar-get, P.-N. Seo, D.G. Crabb, R. Miskimen, M. Seely, H.R. Weller, Nucl. Instrum. Methods A,618, 43, (2010).

47. Measurement of the 241Am(γ,n)240Am reaction in the giant dipole resonance region, A. P.Tonchev, S. Hammond, C. R. Howell, A. Hutcheson, T. Kawano, J. H. Kelley, E. Kwan, G.Rusev, W. Tornow, D. Vieira, and J. B. Wilhelmy, Phys. Rev. C, 82, 054620 (2010).


Journal Articles Accepted

1. Electron Beam Energy Spread Measurements Using Optical Klystron Radiation, B. Jia, J. Li,S. Huang, Scott C. Schmidler, and Y. K. Wu, accepted to be published in Phys. Rev. STAccel. Beams (2010).

2. GEMStar: The Alternative Reactor Technology Comprising Graphite, C. D. Bowman, E. G.Bilpuch, D. C. Bowman, A. S. Crowell, C. R. Howell, K. McCabe, G. A. Smith, A. P. Tonchev,W. Tornow, V. Violet, R. B. Vogelaar, R. L. Walter, and J. Yingling, Chapter prepared forthe new International Handbook of Nuclear Engineering, Ed. Dan Gabriel Cacuci.

3. Hydrodynamic Models of Type I x-ray Bursts: Metallicity Effects, J. Jose, F. Moreno, A.Parikh, and C. Iliadis, Astrophys. J., submitted (2010).

4. Neutron Production With a Pyroelectric Double-crystal Assembly Without Nano-tip, W. Tornow,W. Corse, S. Crimi, and J. Fox, submitted to Nucl. Instrum. Methods A, (2010).

5. Random Matrices and Chaos in Nuclear Physics: Nuclear Reactions, G. E. Mitchell, A. Richter,and H. A. Weidenmuller, to be published in Rev. Mod. Phys.

6. Energy Levels of Light Nuclei, A=3, J.E. Purcell, J.H. Kelley, E. Kwan, C.G. Sheu, and H.R.Weller, accepted for publication in Nucl. Phys. A.

7. Production and Characterization of Intercalated Graphite Crystals for Cold Neutron Monochro-mators, P. Courtois et al. (including P.R. Huffman), Nucl. Instrum. Methods A, in press(2010).


Journal Articles Submitted

1. A General Solution to the Diffusion Induced Transverse and Longitudinal Relaxation in a Non-uniform Magnetic Field with Constant Field Gradients, W. Zheng, H. Gao et al., submittedto Phys. Rev. A, (2010).

2. BrilLanCeTM Detector Characterization Below 15.5 MeV, N. Brown, M. W. Ahmed, S. S.Henshaw, B. A. Perdue, P. -N. Seo, S. Stave, H. R. Weller, C. Sun, Y. K. Wu, P. P. Martel,A. Teymurazyan, A. Owens, and F. Quarati, submitted to Nucl. Instr. Methods A, (2010).

3. Determination of the Proton and Alpha-particle Light-response Function for the KamLAND,BC-501A nd BC-517H Liquid Scintillators, B. Braizinha, J. H. Esterline, H. J. Karwowski,and W. Tornow, submitted to Nucl. Inst. Methods A, (2010).

4. Low-background Gamma Counting at the Kimballton Underground Research Facility, P. Finnerty,S. MacMullin, H. O. Back, R. Henning, K. T. Macon, J. Strain, R. M. Lindstrom, and R. B.Vogelaar, submitted to Nucl. Instrum. Methods A, (2010).

5. Nuclear Astrophysics Studies at LENA: The Accelerators, J. M. Cesaratto, A. E. Champagne,T. B. Clegg, M. Q. Buckner, R. C. Runkle, and A. Stefan, to be published in Nuclear Instru-ments and Methods in Physics Research, A, (2010).

6. Spin-Correlation Coefficients and Phase-Shift Analysis for p+3He Elastic Scattering, T. V.Daniels, C. W. Arnold, J. M. Cesaratto, T. B. Clegg, A. H. Couture, H. J. Karwowski, and T.Katabuchi, submitted to Phys. Rev. C, March 30, 2010.

7. Understanding the 11B(p,α)αα Reaction at the 0.675 MeV Resonance, S.S. Stave, M.W.Ahmed, R.H France III, S.S. Henshaw, B. Muller, B.A. Perdue, R.M. Prior, M.C. Spraker,and H.R. Weller, submitted to Phys. Lett., (2010).

8. An Optical Readout TPC (O-TPC) for Studies in Nuclear Astrophysics with Gamma-rayBeams at HIγS, M. Gai et al., submitted to JINST, (2010).

9. Dipole States in 235U, E. Kwan, G. Rusev, A. S. Adecola, S. L. Hammond, C. R. Howell, H. J.Karwowski, J. H. Kelley, R. S. Pedroni, R. Raut, A. P. Tonchev, and W. Tornow, submittedto Phys. Rev. C, (2010).


Special Reports and Books

1. Missing Level Corrections using Neutron Spacings, G. E. Mitchell and J. F. Shriner, Jr.,INDC(NDS)-0561 International Atomic Energy Agency, Vienna, (2009).

2. Proceedings of the First International Conference Ulaanbaatar Conference on Nuclear Physicsand Applications, ed. D. Dashdorj, U. Agvaanluvsan, and G. E. Mitchell, AIP Volume 1109,Melville, New York, (2009).

3. Polarized Particle Sources: Protons and Heavy Ions, T. B. Clegg and W. Haeberli, in Handbookof Accelerator Physics and Engineering, 2nd Edition, eds., A. Chao and M. Tigner, (to bepublished by World Scientific, Singapore).

4. Neutron EDM Experiments, Steve K. Lamoreaux and Robert Golub, in Lepton Dipole Mo-ments, ed. B. Lee Roberts and William J. Marciano World Scientific Publishing Company,Singapore, 583-634, (2009).


Conference Reports and Articles in Conference Proceedings

1. Complete Phenomenological Model for Projectile-breakup Reactions, C. K. Walker, Report tothe Second Research Coordination Meeting of the Fusion Energy Nuclear Data Library Coordi-nated Research Project (FENDL-3 CRP) of the International Atomic Energy Agency, Vienna,March (2010).

2. Introduction to Nuclear Astrophysics, C. Iliadis, Proceedings of the 5th European SummerSchool on Experimental Nuclear Astrophysics, Santa Tecla, Italy, AIP Conf. Proc., AIP, NewYork, in print (2009).

3. New Results from the Bates Large Acceptance Spectrometer Toroid (BLAST), H. Gao, Confer-ence Proceedings paper presented at the 7th International Conference on Nuclear Physics atStorage Rings STORI ’08, Lanzhou, China, September 14 - 18, 2008, International Journal ofModern Physics E, 18, No. 2, 209 - 219, (2009).

4. Recent Few-Body Studies at TUNL - Experimental Results and Challenges, T. B. Clegg, Invitedcontribution at 6th International Workshop on Chiral Dynamics, Univ. of Bern, July 6 - 10,(2009).

5. Spin and Parity Assignments to Dipole Excitations of the Odd-Mass Nucleus 207Pb from Nu-clear Resonance Fluorescence Experiments with Linearly-Polarized-Ray Beams, N. Pietralla,T. C. Li, M. Fritzsche, M. W. Ahmed, D. Savran, A. P. Tonchev, W. Tornow, H. R. Weller,and V. Werner, Varna Summer School, Varna, Bulgaria (2009).

6. Study of Semi-inclusive Deep-inelastic (e,e’ π±) Production at 11 GeV in JLab With a Polar-ized 3He Target, X. Qian, H. Gao, J. -P. Chen, and E. Chudakov, AIP Conf. Proc. 1149, 457,(2009).

7. The First Measurement of Neutron Transversity on a Transversely Polarized 3He Target, Y.Qiang, proceedings of CIPANP 2009, (2009).

8. The Majorana Project, S. R. Elliott et al. (MAJORANA Collaboration), Carolina InternationalSymposium on Neutrino Physics, Journal of Physics: Conference Series 173, 012007, (2009).

9. A Pulsed ECR Source: Environmental and Beam Induced Background Suppression, M. Q.Buckner, J. M. Cesaratto, T. B. Clegg, and C. Iliadis, Poster presentation at DOE StewardshipScience Annual Conference, Washington, DC, June 21-22, (2010).

10. Cross Section Measurement on 139La (γ, γ′) Below Neutron Separation Energy, A. Makinaga,G. Rusev, R. Schwengner, F. Donau, R. Beyer, D. Bremmerer, P. Crespo, A. R. Junghans,J. Klug, C. Nair, K. D. Schilling, and A. Wagner, Tours Symposium on Nuclear Physics andAstrophysics - VII, Kobe, Japan, 2009, AIP Conf. Proc. 1238, 228, (2010).

11. Hydrodynamic Models of Classical Novae and Type I X-ray Bursts, J. Jose, J. Casanova, F.Moreno, E. Garcia-Berro, A. Parikh, and C. Iliadis, Proceedings of the 7th Tours Symposiumon Nuclear Physics and Astrophysics, Kobe, Japan, 2009, AIP Conf. Proc., AIP, New York,in print (2010).

12. Monte Carlo Reaction Rate Evaluation for Astrophysics, A. Coc, C. Iliadis, R. Longland, A. E.Champagne, and R. Fitzgerald, Proceedings of the 7th Tours Symposium on Nuclear Physicsand Astrophysics, Kobe, Japan, 2009, AIP Conf. Proc., AIP, New York, in print (2010).

13. New Developments in Experimental Thermonuclear Reactions, C. Iliadis, Proceedings of the7th Tours Symposium on Nuclear Physics and Astrophysics, Kobe, Japan, 2009, AIP Conf.Proc., AIP, New York, in print (2010).


14. Photoactivation of the p-nucleus 92Mo with Bremsstrahlung at ELBE, M. Erhard, E. Grosse,A. R. Junghans, J. Klug, C. Nair, G. Rusev, K. D. Schilling, R. Schwengner, and A. Wagner,Nuclear Physics in Astrophysics IV, Frascati, Italy, 2009, J. Phys. Conf. Ser. 202, 012014,(2010).

15. Photon Scattering Experiment on 139La below Neutron Separation Energy at ELBE, A. Mak-inaga, G. Rusev, R. Schwengner, A. R. Junghans, and A. Wagner, Perspectives in NuclearPhysics, Tokai, Japan, 2008, AIP Conf. Proc. 1120, 289, (2009).

16. Properties of Warm Nuclei in the Quasicontinuum, M. Guttormsen, U. Agvaanluvsan, A. C.Larsen, R. Chankova, G. E. Mitchell, A. Schiller, S. Siem, and A. Voinov, Proceedings of theSecond International Workshopon Compound Nuclear Reactions and Related Topics, editorsL. Bonneau, N. Dubray, F. Gunsing, B. Jurado, EDP Web of Conferences 2, 04001, (2010).

17. Soft Structures of γ-ray Strength Functions Studied with the Oslo Method, A. C. Larsen, U.Agvaanluvsan, S. Goriely, M. Guttormsen, E. Algin, L. A. Bernstein, R. Chankova, T. Lon-nroth, G. E. Mitchell, J. Rekstad, H. K. Toft, A. Schiller, S. Siem, N. U. H. Syed, and A.Voinov, Proceedings of the Second International Workshopon Compound Nuclear Reactionsand Related Topics, editors L. Bonneau, N. Dubray, F. Gunsing, B. Jurado, EDP Web ofConferences 2, 03001, (2010).

18. The Majorana Project, C. E. Aalseth et al. (MAJORANA Collaboration), Topics in As-troparticle and Underground Physics (TAUP 2009), Journal of Physics: Conference Series203, 012057, (2010).


Abstracts to Meetings and Conferences

1. A New Measurement of the Total Cross Section for the Photodisintegration of 9Be Near Thresh-old, C. W. Arnold, T. B. Clegg, H. J. Karwowski, G. C. Rich, J. R. Tompkins, C. R. Howell,http://meetings.aps.org/link/BAPS.2010.APR.K7.7

2. An Overview of the Data Acquisition System in the KATRIN Experiment, D. Phillips II,http://meetings.aps.org/link/BAPS.2010.APR.G8.9

3. Astroparticle and Nuclear Physics with a Customized Low-Background Broad Energy Germa-nium Detector, P. Finnerty,http://meetings.aps.org/link/BAPS.2010.APR.Y10.4

4. Development of a Data Acquisition System for a Low-Background BEGe Detector, G. K. Gio-vanetti,http://meetings.aps.org/link/BAPS.2010.APR.G8.10

5. Calibration of the MiniCLEAN detector, M. Akashi-Ronquest,http://meetings.aps.org/link/BAPS.2010.APR.G9.3

6. Measurements of Dipole Excitations in 48Ca Between Eγ = 9.5 and 15.3 MeV, J. R. Tompkins,C. W. Arnold, H. J. Karwowski, G. C. Rich, C. R. Howell, L. G. Sobotka,http://meetings.aps.org/link/BAPS.2010.APR.B7.2

7. Differential Cross Section Measurements of Elastic and Inelastic Scattering of Neutrons fromNeon, S. MacMullin et al.,http://meetings.aps.org/link/BAPS.2010.APR.H7.3

8. Characterizing Electron Optics in the KATRIN Experiment, T. Corona,http://meetings.aps.org/link/BAPS.2010.APR.A8.1

9. Enabling Environmental Background Reduction for Nuclear Astrophysics Measurements withPulsed Proton Beams, M. Q. Buckner, J. Cesaratto, and T. B. Clegg,http://meetings.aps.org/link/BAPS.2010.APR.H7.8

10. First Study of Three-body Photo-disintegration of 3He:−−→3He(�γ, n)pp with Double Polarizations,

W. Chen et al.,http://meetings.aps.org/link/BAPS.2010.APR.K7.5

11. Sample Preparation and Gamma-Assay Measurements at the Low-Background Counting Fa-cility at the Kimballton Underground Research Facility, J. Strain et al.,http://meetings.aps.org/link/BAPS.2010.APR.Y10.3

12. Astroparticle Physics with a Customized Low-Background Germanium Detector, P. Finnerty,2010 Symposium on Radiation Measurements and their Applications (SORMA XII),http://rma-symposium.engin.umich.edu/index.php

13. Background Model for the MAJORANA DEMONSTRATOR, R. Henning on behalf of theMAJORANA Collaboration, Poster Presentation at Neutrino 2010, Athens, Greece, 2010

14. Fine Structure of the M1 Resonance in 90Zr, G. Rusev, A. S. Adekola, F. Donau, S. Frauendorf,S. L. Hammond, C. Huibregtse, J. H. Kelley, E. Kwan, R. Schwengner, A. P. Tonchev, W.Tornow, and A. Wagner,http://inpc2010.triumf.ca/

15. Asymmetry of Photo-fission Neutrons from 238U, J. R. Tompkins et al., DNDO meeting,Alexandria, VA, April, (2010)

16. Development of Gd-based Neutron Detector, D. Ticehurst et al., DNDO meeting, Alexandria,VA, April, (2010)


17. Dipole Excitations in 235U (γ, γ′) Scattering, E. Kwan et al., DNDO meeting, Alexandria, VA,April, (2010)

18. Dipole States in 238U, S. L. Hammond et al., DNDO meeting, Alexandria, VA, April, (2010)

19. 235U Photofission Activation Measurement, J.M. Mueller et al., DNDO Meeting, Alexandria,VA, April, (2010)

20. Determination of the 1S0 Neutron-neutron Scattering Length Using nd Breakup in Recoil Ge-ometry at 19 MeV, S. Tajima, A. S. Crowell, J. Deng, J. Esterline, C. R. Howell, M. R. Kiser,R. A. Macri, and W. Tornow, B. J. Crowe III, R. S. Pedroni, W. Von Witsch, H. Wita�la,http://meetings.aps.org/link/BAPS.2009.SES.HC.2

21. Simulation Comparisons between Geant4 and MCNPX, J. R. Tompkins, G. Rich and C. W.Arnold,http://meetings.aps.org/link/BAPS.2009.SES.HC.3

22. Nuclear Resonance Fluorescence from Uranium above 2 MeV, E. Kwan, C. R. Howell, R. Raut,G. Rusev, A. P. Tonchev, W. Tornow, A. Adekola, S. L. Hammond, H. J. Karwowski, J. R.Tompkins, C. Huibregtse, J. H. Kelley, B. Johnson,http://meetings.aps.org/link/BAPS.2009.HAW.BK.6

23. Precision photo-induced cross-section measurements using the monoenergetic and polarizedphoton beams at HIγS, A.P. Tonchev et al.,http://meetings.aps.org/link/BAPS.2009.HAW.LC.3

24. Compton Scattering on 209Bi at HIγS from Eγ=11-30 MeV, S.S. Henshaw et al.,http://meetings.aps.org/link/BAPS.2009.HAW.BH.11

25. Performance of the UConn-TUNL O-TPC with the Upgraded Optical Readout System, W.R.Zimmerman et al.,http://meetings.aps.org/link/BAPS.2009.HAW.LG.10

26. BrilLanCe detector energy resolution characterization at HIγS, N. Brown et al.,http://meetings.aps.org/link/BAPS.2009.HAW.ED.1

27. A New Method for Identifying Nuclear Isotopes Based Upon Polarized (γ, n) Asymmetries, S.Stave et al.,http://meetings.aps.org/link/BAPS.2009.HAW.LC.1

28. The 11B(p,α)αα Reaction at Low Energies, R.M. Prior et al.,http://meetings.aps.org/link/BAPS.2009.HAW.LG.8

29. Neutron-Neutron Scattering Length Determinations Using nd Breakup in Different NucleonDetection Geometries, C.R. Howell et al.,http://meetings.aps.org/link/BAPS.2009.HAW.CH.7

30. High Precision Measurement of the 19Ne Lifetime, L. Broussard et al.,http://meetings.aps.org/link/BAPS.2009.HAW.KJ.11

31. Development of a Conversion Electron Source for Timing Measurements and the Determina-tion of Angle Dependent Detector Response in the UCNA Experiment, A.J. Cetnar, L. Brous-sard, R.W. Pattie, and A.R. Young,http://meetings.aps.org/link/BAPS.2009.HAW.GB.24

32. Progress on the Construction of the PULSTAR Solid Deuterium Ultracold Neutron Source, G.Palmquist et al.,http://meetings.aps.org/link/BAPS.2009.HAW.KK.3

33. Precision UCN Polarimetry and the UCNA Experiment, A.T. Holley,http://meetings.aps.org/link/BAPS.2009.HAW.KK.8

34. Systematics of the UCNA Experiment, R. Pattie,http://meetings.aps.org/link/BAPS.2009.HAW.CK.7


35. Neutron Capture Measurements Using DANCE, A. Chyzh et al.,http://meetings.aps.org/link/BAPS.2009.HAW.DG.11

36. Updates on neutron induced measurements on Gadolinium isotopes at the DANCE array, D.Dashdorj et al.,http://meetings.aps.org/link/BAPS.2009.HAW.KC.11

37. Level Density and Radiative Strength of 116,117Sn, A. Schiller, U. Agvaanluvsan, A. C. Larsen,R. Chankova, M. Guttormsen, G. E. Mitchell, S. Siem, A. Voinov,http://meetings.aps.org/link/BAPS.2009.HAW.DG.10

38. Ambient Neutron Flux Measurements at Kimballton Underground Research Facility (KURF),D. Kaleko, R. Henning, and W. Tornow,http://meetings.aps.org/link/BAPS.2009.HAW.GB.61

39. Generation of Unprecedented high Electric Fields with Pyroelectric Crystals, S. Crimi, W.Tornow, and Z. Corse,http://meetings.aps.org/link/BAPS.2009.HAW.GB.29

40. Double-Beta Decay of 150Nd to Excited Final States, M. Kidd, J.H. Esterline, and W. Tornow,http://meetings.aps.org/link/BAPS.2009.HAW.DJ.6


A.3 Invited Talks, Seminars, and Colloquia

Invited Talks, Seminars, and Colloquia

1. Globular Clusters, Nuclear Physics and the Age of the Galaxy, A. E. Champagne, PhysicsColloquium, College of Charleston Charleston, SC, April, (2010).

2. Stellar Reactions, Update, Status and Plans, A.E. Champagne and C. Iliadis, Invited talk atthe Workshop on Cosmic Chemical Evolution, St. Michaels, Md, June 2-4, (2010).

3. Prospects in Neutron Transverse Spin Study with Polarized 3He at 12 GeV Jefferson Labora-tory, H. Gao, Invited talk at the 3rd Joint Meeting of the APS Division of Nuclear Physicsand the Physical Society of Japan, Waikoloa, HI, Oct. 13 - 17, (2009).

4. Prospects in Neutron Transverse Spin Study with Polarized 3He at 12 GeV Jefferson Labora-tory, H. Gao, Invited talk at the 76th Annual Meeting of the Southeastern Section of APS,Atlanta, GA, Nov. 11 - 14, (2009).

5. Φ Meson Photoproduction on Deuteron and Future Studies Using Heavier Nuclear Targets,H. Gao, Invited talk at the High Energy Nuclear Physics and QCD, Florida InternationalUniversity, Miami, February 3 - 6, (2010).

6. A New Search on the Neutron Electric Dipole Moment, H. Gao, Colloquium, University ofMinnesota, March 3, (2010).

7. Workshop on Partonic Transverse Momentum in Hadrons (summary), H. Gao, Invited talk atthe JLab Users Group annual meeting, Newport News, VA, June 7 - 9, (2010).

8. Prospects/plans for TMD studies at the EIC, H. Gao, Invited talk at the workshop on Transverse-Momentum-Dependent Distributions, Trento, Italy, June 21 - 25, (2010).

9. Frontiers in Hadron Structure, H. Gao, Special Lecture, 2010 US National Nuclear PhysicsSummer School (NNPSS) jointly with the TRIUMF Summer Institute (TSI), Vancouver, BC,Canada, June 21 - July 2, (2010).

10. Recent Results on Structure Functions, H. Gao, Plenary talk at the 35th International Confer-ence on High Energy Physics, Paris, France, July 22 - 28, (2010).

11. 12 GeV Neutron 3He Transversity TMDs with SoLID, H.Gao, Invited talk at the 2nd WorkshopHadron Physics in China and Opportunities with 12 GeV JLab, Beijing, China, July 28 - 31,(2010).

12. Direct Detection of Dark Matter, R. Henning, Southeastern Section of the APS 2009, Atlanta,GA, (2009).

13. Overview of Current and Proposed Searches for Double-beta Decay, R. Henning, Neutrinos andDark Matter 2009, Madison, WI, (2009).

14. Progresss Towards a beta-asymmetry Measurement with Ultracold neutrons, A. T. Holley, 6thInternational Workshop on Cold and Ultracold Neutrons, St. Petersburg, Russia, June (2009).

15. Nuclear Data Measurements on Actinides Using the High Intensity Gamma-ray Source, C.R.Howell, ARI Grantees Conference, Alexandria, VA, April 2010.

16. Simulations of TSSA from SIDIS at EIC, M. Huang, Invited talk at the Workshop on Par-tonic Transverse Momentum in Hadrons: Quark Spin-Orbit Correlations and Quark-GluonInteractions, Duke University, Durham, NC, March 13, (2010).

17. Fundamental Neutron Physics Using in situ UCN Sources., P. R. Huffman, Research Oppor-tunities with Ultracold Neutrons in the US, Santa Fe, NM, November, (2009).


18. The Fundamental Neutron Physics Program at the Spallation Neutron Source, P. R. Huffman,CAARI-2010, Fort Worth, TX, August (2010).

19. New Developments in Experimental Thermonuclear Reactions, C. Iliadis, Invited plenary talkat the 7th Tours Symposium on Nuclear Physics and Astrophysics, Kobe, Japan, November,(2009).

20. Recent Measurements in Nuclear Astrophysics, C. Iliadis, Talk at the Johann Wolfgang Goethe-Universitat, Frankfurt am Main, Germany, June, (2010).

21. Entfernungsmessung im Universum, C. Iliadis, Talk at the Johann Wolfgang Goethe-UniversitatFrankfurt am Main, Germany, June, (2010).

22. Introduction to Nuclear Astrophysics, C. Iliadis, Lecture (2 hours) given at the Fifth EuropeanSummer School on Experimental Nuclear Astrophysics, Santa Tecla, Italy, September, (2009).

23. Cross Sections and Reaction Rates in Nuclear Astrophysics, C. Iliadis, One-week long course(5 hours) given at the ASA/ANITA Summer School in Stellar Nucleosynthesis, Monash Uni-versity, Melbourne, Australia, January, (2010).

24. Topics in Nuclear Astrophysics, C. Iliadis, Lecture (3 hours) given at the Carpathian SummerSchool of Physics 2010 (Exotic Nuclei and Nuclear/Particle Astrophysics III), Sinaia, Romania,June, (2010).

25. Methods to Detect SNM with Nuclear Resonance Flourescence, M. Johnson et al., presented atInternational Conference Actinides 2009, Berkely, July, (2009).

26. Nuclear Resonance Fluorescence on 235,238U, H.J. Karwowski, talk given at Workshop onTopics in Homeland Nuclear Security, Berkeley, July, (2009).

27. Dipole States in 232Th and 238U, H.J. Karwowski, presented at the International NuclearPhysics Conf. Vancouver, July, (2010).

28. Neutron Capture Reactions on Gadolinium Isotopes, B. Baramsai, U. Agvaanluvsan, F. Becvar,T. Bredeweg, A. Chyzh, A. Couture, D. Dashdorj, R. Haight, M. Jandel, A. L. Keksis, M.Krticka, G. E. Mitchell, J. O’Donnell, R. Rundberg, J. Ullmann, D. Vieira, C. Walker, J.Wouters, LANSCE Users Group Meeting, Santa Fe, NM (2009).

29. Neutron Capture Experiments at DANCE, G. E. Mitchell, Forschungzentrum Dresden, Ger-many, November, (2009).

30. Properties of Warm Nuclei in the Quasicontinuum, M. Guttormsen, U. Agvaanluvsan, A. C.Larsen, R. Chankova, G. E. Mitchell, A. Schiller, S. Siem, and A. Voinov, Second InternationalWorkshop on Compound Nuclear Reactions and Related Topics, Bordeaux, France (2009).

31. Soft Structures of γ-ray Strength Functions Studied with the Oslo Method, A. C. Larsen, U.Agvaanluvsan, S. Goriely, M. Guttormsen, E. Algin, L. A. Bernstein, R. Chankova, T. Lon-nroth, G. E. Mitchell, J. Rekstad, H. K. Toft, A. Schiller, S. Siem, N. U. H. Syed, and A.Voinov, Second International Workshop on Compound Nuclear Reactions and Related Topics,Bordeaux, France (2009).

32. Parity Violation in Compound Nuclear Resonances, G. E. Mitchell, From Femtoscience toNanoscience: Nuclei, Quantum Dots, and Nanostructures, Seattle, August, (2009).

33. Cross Sections, Level Densities and Strength Functions, G. E. Mitchell, Stockpile StewardshipAcademic Alliance Symposium, Washington. D.C, January, (2010).

34. Applications of Random Matrix Theory: Parity Violation and Missing Level Corrections, G.E. Mitchell, Oak Ridge Natrional Laboratory, Oak Ridge, May, (2010).

35. Single-spin Asymmetry Measurement at Jefferson Lab Hall A with a Polarized 3He Target, X.Qian, Nuclear Physics Seminar, Caltech, Oct. 9, (2009).


36. Transverse Momentum-dependent Structure of the Nucleon, Y. Qiang, Seminar at RutgersUniversity, New Brunswick, Oct. 28, (2009).

37. Transverse Momentum-dependent Structure of the Nucleon, Y. Qiang, Seminar at JeffersonLab, Newport News, Oct. 30, (2009).

38. Transverse Momentum-dependent Structure of the Nucleon, Y. Qiang, Seminar at JeffersonLab, Newport News, Nov. 25, (2009).

39. Transverse Spin Physics at Jefferson Lab with Polarized 3He, Y. Qiang, Invited talk at theWorkshop on Partonic Transverse Momentum in Hadrons: Quark Spin-Orbit Correlations andQuark-Gluon Interactions, Duke University, Durham, NC, Mar.12 - 13, (2010).

40. Study of the Pygmy Dipole Resonance, A. P. Tonchev, Duke, Aug. 24, (2009).

41. Fine Structure of the Nuclear Dipole Response, A. P. Tonchev, Ohio University, Athens, OH,Sep. 29, (2009).

42. Fine Structure of the Pygmy Dipole Resonance, A. P. Tonchev, National SuperconductingCyclotron Laboratory, Michigan State University, November 11, (2009).

43. Precision Photo-Induced Cross-Section Measurements using the Monoenergetic and PolarizedPhoton Beams at HIγS, A. P. Tonchev, SSAA Symposium, Washington DC, January 20,(2010).

44. Use of Monoenergetic and Polarized Photon Beams for Basic and Applied Nuclear Physics, A.P. Tonchev, Department of Physics and Astronomy, James Madison University, February 4,(2010).

45. Fine and Gross Structure of the Pygmy Dipole Resonance, A. P. Tonchev, The 24th Interna-tional Nuclear Physics Conference (INPC), Vancouver, Canada.

46. Astro-nuclear Physics with Medium Energy Photons from HIGS, W. Tornow, Advanced Pho-tons and Science Evolution 2010 (APSE2010), Osaka, Japan, June 14 - 18, (2010).

47. Research Opportunities at HIγS, H.R. Weller, Los Alamos National Laboratory Colloqium,Los Alamos, NM, March (2010).

48. Development of an Aneutronic Fusion Reactor Using the 11B(p,α)αα Reaction, H.R. Weller,Massachusetts Institute of Technology Colloqium, March 30, (2010).

49. A New Method for Identifying Nuclear Materials Based Upon Polarized (γ,n) Asymmetries,H.R. Weller, ARI Grantees Conference, Alexandria, VA, April (2010).

50. Probing Chiral Dynamics with Photopion Experiments Near Threshold, H.R. Weller, Meson-Nucleon Physics and the Structure of the Nucleon (MENU 2010), College of William and Mary,Williamsburg, VA, May 31, (2010).

51. Neutrinoless Double Beta Decay At the Intersection of Particle and Nuclear Physics, J.F.Wilkerson, 10th Conference on the Intersections of Particle and Nuclear Physics, La Jolla,CA, May, (2009).

52. Experimental Approaches to Neutrinoless Double Beta Decay DM Detection, and CoherentNeutrino Scattering, J.F. Wilkerson, International Neutrino Summer School, Fermilab, Batavia,IL, July, (2009).

53. Beta Decay Probes of the Neutrino Mass Scale, J.F. Wilkerson, International Neutrino SummerSchool, Fermilab, Batavia, IL, July, (2009).

54. The Reticent, Yet Remarkable Neutrino, J.F. Wilkerson, Physics Colloquium, Duke University,Durham, NC, September, (2010).

55. A 1-tonne 76Ge Neutrinoless Double Beta Decay Experiment, J.F. Wilkerson, DUSEL ScienceWorkshop, Lead, SD, October, (2009).


56. Neutrino Mass from Double Beta Decay, J.F. Wilkerson, Institute for Nuclear Physics Work-shop on the Future of Neutrino Mass Measurements, Seattle WA, February, (2010).

57. Status of the MAJORANA DEMONSTRATOR Project, J.F. Wilkerson, GERDA Collabora-tion Meeting, Gran Sasso, Italy, March, (2010).

58. The Reticent, Yet Remarkable Neutrino, J.F. Wilkerson, Physics Colloquium, Va. Tech.,Blacksburg, VA, March, (2010).

59. The Reticent, Yet Remarkable Neutrino, J.F. Wilkerson, Physics Division Seminar, ArgonneNational Laboratory, Argonne, IL, March, (2010).

60. A New Search for the Neutron Electric Dipole Moment, Q. Ye, Seminar, Indiana UniversityCyclotron Facility, Blooming-ton, IN, January 15, (2010).

61. A New Search for the Neutron Electric Dipole Moment, Q. Ye, Seminar, University of Kentucky,Lexington, KY, January 21, (2010).

62. A New Search for the Neutron Electric Dipole Moment, Q. Ye, Seminar, Argonne NationalLab, Argonne, IL, January 28, (2010).

63. A New Search for the Neutron Electric Dipole Moment, Q. Ye, Seminar, Princeton University,Princeton, NJ, February 5, (2010).

64. Fundamental Physics with Ultracold Neutrons: A New Approach to a Challenging Problem, A.R. Young, McGill College Colloquium, November (2009).

65. Fundamental Physics with Ultracold Neutrons: A New Approach to a Challenging Problem, A.R. Young, Indiana University Colloquium, September (2009).

66. Fundamental Physics with Ultracold Neutrons: A New Approach to an Old Problem, A. R.Young, University of Vermont Colloquium, September (2009).


Seminars at TUNL

1. Lt. Col. Nick Prins, Domestic Nuclear Detection Office of the Department of HomelandSecurity, (09/17/2009)Progress in Detection of Nuclear Materials

2. Seth Hoedl, University of Washington, (09/24/2009)A Torsion Pendulum Based Axion Search.

3. Gail McLaughlin, North Carolina State University, (10/01/2009)Neutrinos from Supernovae and Compact Object Mergers

4. Luke Myers, University of Illinois, (10/08/2009)Neutron Polarizabilities Via Deuteron Compton Scattering At MAX-lab

5. Jay Davis, LLNL, (10/20/2009)Twenty Years of Tricks with Tandems

6. Adriana Banu, Texas A & M University, (10/29/2009)Reaction Rates for H-burning in Novae and X-ray Bursts from Breakup at Intermediate Ener-gies

7. Florian Fraenkle, Karlsruher Institut fur Technologie (KIT), (11/12/2009)Measurements with the KATRIN Pre-spectrometer

8. Changchun Sun, Duke University , (11/19/2009)Characterizations and Diagnostics of Compton Light Source

9. Oliver Jahn, Institut fuer Kernphysik, Mainz, (11/23/2009)The GDH Experiment at MAMI

10. Babatunde Oginni, Ohio University, (12/03/2009)Test of Level Density Models from Nuclear Reactions

11. John Watson, Kent State University, (01/21/2010)From the Nuclear Shell Model to Short-Range Correlations: The Adventure of a Lifetime

12. Chris Howard, University of Alberta, (02/25/2010)The Sudbury Neutrino Observatory: Searching for hep Neutrinos

13. Kevin Veal, NNSA, (03/04/2010)Personal Experiences as a Nuclear Nonproliferation Practitioner

14. Simona Malace, University of South Carolina, (03/18/2010)Pinning Down the Nucleon’s Quark Distributions at Large Bjorken-x

15. Richard Firestone, Lawrence Berkeley National Laboratory, (03/30/2010)The IAEA/LBNL Evaluated Gamma-Ray Activation File (EGAF)

16. Jason Detwiler, Lawrence Berkeley National Laboratory, (04/01/2010)1 count / tonne-year? Battling Backgrounds in the MAJORANA DEMONSTRATOR

17. Bernd Surrow, Massachusetts Institute of Technology, (04/08/2010)First W Results from RHIC STAR

18. Larry Nittler, Carnegie Institution of Washington, (04/22/2010)Stellar Fossils: Presolar Stardust in the Solar System

19. Dr. Vladimir Litvinenko, Brookhaven National Laboratory, (04/27/2010)eRHIC high luminosity electron-ion collider at BNL


20. Jocelyn Monroe, MIT, (04/29/2010)Directional Dark Matter Detection and Recent Progress with the DMTPC Detector

21. Robert Charity, Washington University of St. Louis, (04/29/2010)Two and Four-Proton Decay modes of Proton-Rich Nuclei

22. Diana Parno, Carnegie Mellon University, (05/06/2010)High-Precision Compton Polarimetry in Hall A of Jefferson Lab


Advances in Physics Lectures and Seminars

1. Richard Prior, North Georgia College and State University, (6/4/2010)Introduction to Nuclear Reactions

2. Mohammad Ahmed, Duke University, (6/7/2010)Introduction to Nuclear Structure

3. Mohammad Ahmed, Duke University, (6/7/2010)Introduction to Nuclear Electronics

4. Gencho Rusev, Duke University, (6/10/2010)Investigation of Nuclear Structure with Real Photons

5. Qiang Ye, Duke University, (6/24/2010)3He Relaxation Time Measurements for the Neutron Electric Dipole Moment (nEDM) Exper-iment

6. David Haase, North Carolina State University, (7/1/2010)Nuclear Physics at Low Temperatures

7. Sean Stave, Duke University, (7/15/2010)Interrogation of Special Nuclear Materials Based Upon Polarized (γ, n) Asymmetries

8. Reyco Henning, University of North Carolina at Chapel Hill, (7/22/2010)Searching for the Rarest Events in the Universe


A.4 Professional Service Activities

Advisory/Fellowship/Review Committees

1. Member, Editorial Board of the Physical Review C HRIBF Users Executive Committee, 2009- 2010, A. Champagne

2. Member, Southeastern Universities Research Association/Jefferson Science Associates Jeffer-son Laboratory Advisory Committee, 2009 - 2010, T.B. Clegg

3. Member, Fellowship Committee, Division of Nuclear Physics of APS, 2009 - 2010, H. Gao

4. Member, International Advisory Committee, the 12th International Symposium on Meson-Nucleon Physics and the Structure of the Nucleon, College of William and Mary, May 31,2010 - June 4, 2010, H. Gao

5. Member, International Advisory Committee, Second International Conference on NuclearPhysics and Applications, Ulaanbaatar, Mongolia, July 2010, C. Gould

6. Member, Nuclear Physics and Applications, Ulaanbaatar, Mongolia, July 2010, C. Gould

7. Member, National Research Council: Panel on Neutron Research at NIST, 2010, C. Gould

8. Member, Executive Committee of the Division of Nuclear Physics of the APS, C.R. Howell

9. Member, Majorana Senior Advisory Committee, C.R. Howell

10. Member, 2010 Dissertation Grant Selection Panel for the Mellow Mays Undergraduate Fellow-ship Program, C.R. Howell

11. Member, 2010 Stockpile Stewardship Graduate Fellowship program Selection Committee, C.R.Howell

12. Member, DOE Review Panel for the nuclear physics program at Yale University, 2009, C.R.Howell

13. Member, DOE Review Panel for the nuclear physics program at the Massachusetts Instituteof Technology, 2010, C.R. Howell

14. Member, Organizing Committee for the 10th Torino Workshop, New Zealand, 2009, C. Iliadis

15. Member, Organizer for the Nuclear Astrophysics Workshop, Chapel Hill, NC, 2009, C. Iliadis

16. Member, Program Committee, Division of Nuclear Physics of APS, 2009 - present, C. Iliadis

17. Member, the EuroGENESIS (European Science Foundation), 2009-present, C. Iliadis

18. Member, International Advisory Committee, Second International Ulaan Baatar Conferenceon Nuclear Physics and Applications, Ulaan Baatar, Mongolia, 2007 - 2010, G.E. Mitchell

19. Member, International Advisory Committee, International Workshop on Gamma Strength andLevel Densityin Nuclear Physics and Nuclear Technology, Dresden, Germany, 2008 - 2010, G.E.Mitchell

20. Member, Program Advisory Committee for Fundamental Physics for Los Alamos NeutronScience Center (LANSCE), 2004 - 2010, G.E. Mitchell

21. Member, Program Advisory Committee for Fundamental Physics for Los Alamos NeutronScience Center (LANSCE), A.P. Tonchev

22. Member, the Scientific Program Committee, 19th International Conference on Few-Body Prob-lems in Physics, Bonn, Germany, 2009, W. Tornow


23. Member, International Advisory Committee of International Symposium on Advanced Photonsand Science Evolution, Osaka, Japan, 2010, W. Tornow

24. Member, International Advisory Committee of 21st European Conference on Few-Body Prob-lems in Physics, Salamanca, Spain, 2010, W. Tornow

25. Member, Review Committee for the NRF Program at PNNL, 2009 - 2010, H.R. Weller

26. Invitee, DOE SBIR-STTR Exchange Meeting, Gaithersburg, MD, 2010, H.R. Weller

27. Member, DOE/NSF Nuclear Science Advisory Committee, 2007 - 2009, J.F. Wilkerson

28. Member, the Editorial Board Physical Review C, 2007 - 2009, J.F. Wilkerson

29. Member, the Program Committee for the Division of Nuclear Physics, American Physics So-ciety, 2008 - 2009, J.F. Wilkerson

30. Member, the Organizing Committee for the International Neutrino Summer School, 2008 -2009, J.F. Wilkerson

31. Member, the International Organizing Committee, Conference on the Intersections of Particleand Nuclear Physics, 2008 - 2009, J.F. Wilkerson

32. Member, the International Advisory Committee, XIIth International Conference on Topics inAstroparticle and Underground Physics, 2010 - present, J.F. Wilkerson

33. Member, the International Scientific Program Committee, the 31st International FEL Confer-ence, Liverpool, UK, August 23 - 28, 2009, Y.K. Wu

34. Member, Member, the Scientific Advisory Board (SAB) of the First International ParticleAccelerator Conference (IPAC ’10), Kyoto, Japan May, 24 - 28, 2010, Y.K. Wu

35. Member, Panel for the DOE Office of Nuclear Physics, Early Career Award Program Review,DOE Headquarters, Germantown, Maryland, December 3 - 4, 2009, Y.K. Wu

36. Member, the International Scientific Program Committee, the 32nd International FEL Con-ference, Malmo, Sweden, August 23 - 27, 2010, Y.K. Wu

37. Member, the Scientific Advisory Board (SAB) of the Second International Particle AcceleratorConference (IPAC ’11), San Sebastian, Spain, Sepetember 5 - 9, 2011, Y.K. Wu

38. Member, the Scientific Program Committee, 2011 Particle Accelerator Conference (PAC ’11),New York, March 28 - April 1, 2011, Y.K. Wu

39. Member, Organizer for the Workshop ”Research Opportunities with Ultracold Neutrons in theUS”, Nov. 2009 (with S. Wilburn, of Los Alamos National Laboratory), A.R. Young

40. Member, Review Committee for NSERC, NSF and DOE, A.R. Young

41. Member, Journal review for Phys. Rev. A, C and Letters, and Nucl. Instrum. and Methods,A.R. Young


Conferences, APS Meetings and Workshops

1. Chair, the Workshop on Partonic Transverse Momentum in Hadrons: Quark Spin-Orbit Cor-relations and Quark-Gluon Interactions, Duke University, March 12 - 13, 2010, H. Gao

2. Member, National Nuclear Physics Summer School (NNPSS) Committee, 2009 - 2010, H. Gao

3. Vice President, Overseas Chinese Physicist Association, January 2009 - December 2010, H.Gao

4. Member, APS/AIP/AAPT National Task Force on Teacher Education in Physics, D.G. Haase

5. Past Chair, the Southeastern Section of the American Physical Society, D.G. Haase

6. Chair, the Nuclear Astrophysics Workshop, Chapel Hill, NC, 2009, C. Iliadis

Other Service

1. Member, Program Advisory Committee, Jefferson Lab, 2009 - Current, H. Gao

2. Panel Member, DOE Early Career Award Research Application Review, December 3 - 4, 2009,H. Gao

3. Associated Member, the EuroGENESIS (European Science Foundation), 2009 - present, C.Iliadis

4. Field Editor, Experimental Physics of Few-Body Systems, Springer Verlag Journal, W. Tornow

5. Member of Editorial Board for Physical Review C, H.R. Weller

6. Panel Member, the Deutsche Forschungsgemeinschaft (DFG), Sept. 2009, A.R. Young

7. Spokesman, UCNA project, A.R. Young
