A search for low-mass WIMPS with MALBEK
Transcription
A search for low-mass WIMPS with MALBEK
A search for low-mass WIMPS with MALBEK (MAJORANA Low-Background BEGe @ KURF) J.F. Wilkerson Univ. of North Carolina Oak Ridge National Laboratory On behalf of the MAJORANA Collaboration 28 Feb. 2014 Dark Matter 2014, UCLA 1 The MAJORANA Collaboration Black Hills State University, Spearfish, SD Kara Keeter Pacific Northwest National Laboratory, Richland, Washington Jim Fast, Eric Hoppe, Richard T. Kouzes, Brian LaFerriere, John Orrell, Nicole Overman Duke University, Durham, North Carolina , and TUNL Matthew Busch, James Esterline, Gary Swift, Werner Tornow Shanghai Jiaotong University, Shanghai, China James Loach Institute for Theoretical and Experimental Physics, Moscow, Russia Alexander Barabash, Sergey Konovalov, Vladimir Yumatov South Dakota School of Mines and Technology, Rapid City, South Dakota Adam Caldwell, Cabot-Ann Christofferson, Stanley Howard, Anne-Marie Suriano, Jared Thompson Joint Institute for Nuclear Research, Dubna, Russia Viktor Brudanin, Slava Egorov, K. Gusev, Oleg Kochetov, M. Shirchenko, V. Timkin, E. Yakushev Lawrence Berkeley National Laboratory, Berkeley, California and the University of California - Berkeley Nicolas Abgrall, Mark Amman, Paul Barton, Yuen-Dat Chan, Paul Luke, Susanne Mertens, Alan Poon, Kai Vetter, Harold Yaver Los Alamos National Laboratory, Los Alamos, New Mexico Melissa Boswell, Steven Elliott,, Johnny Goett, Keith Rielage, Larry Rodriguez, Harry Salazar, Wenqin Xu North Carolina State University, Raleigh, North Carolina and TUNL Dustin Combs, Lance Leviner, David G. Phillips II, Albert Young Oak Ridge National Laboratory, Oak Ridge, Tennessee Fred Bertrand, Kathy Carney, Alfredo Galindo-Uribarri, Matthew P. Green, Monty Middlebrook, David Radford, Elisa Romero-Romero, Robert Varner, Brandon White, Timothy Williams, Chang-Hong Yu Osaka University, Osaka, Japan Hiroyasu Ejiri, Ryuta Hazama, Masaharu Nomachi, Shima Tatsuji 28 Feb. 2014 Tennessee Tech University, Cookeville, Tennessee Mary Kidd University of Alberta, Edmonton, Alberta Aksel Hallin University of North Carolina, Chapel Hill, North Carolina and TUNL Graham K. Giovanetti, Reyco Henning, Mark Howe, Jacqueline MacMullin, Sam Meijer, Benjamin Shanks, Christopher O’Shaughnessy, Jamin Rager, Jim Trimble, Kris Vorren, John F. Wilkerson University of South Carolina, Columbia, South Carolina Frank Avignone, Vince Guiseppe, David Tedeschi, Clint Wiseman University of South Dakota, Vermillion, South Dakota Dana Byram, Ben Jasinski, Ryan Martin, Nathan Snyder University of Tennessee, Knoxville, Tennessee Yuri Efremenko, Sergey Vasilyev University of Washington, Seattle, Washington Tom Burritt, Micah Buuck, Clara Cuesta, Jason Detwiler, Peter J. Doe, Julieta Gruszko, Ian Guinn, Greg Harper, Jonathan Leon, David Peterson, R. G. Hamish Robertson, Tim Van Wechel MALBEK - MAJORANA, Dark Matter 2014, UCLA 2 The MAJORANA DEMONSTRATOR (0νββ) Funded by DOE Office of Nuclear Physics and NSF Particle Astrophysics, with additional contributions from international collaborators. Goals: - Demonstrate backgrounds low enough to justify building a tonne scale experiment. - Establish feasibility to construct & field modular arrays of Ge detectors. - Searches for additional physics beyond the standard model. • Located underground at 4850’ Sanford Underground Research Facility • Background Goal in the 0νββ peak region of interest (4 keV at 2039 keV) 3 counts/ROI/t/y (after analysis cuts) scales to 1 count/ROI/t/y for a tonne experiment • 40-kg of Ge detectors – 30 kg of 86% enriched 76Ge crystals – 10 kg of natGe – Detector Technology: P-type, point-contact. • 2 independent cryostats – ultra-clean, electroformed Cu – 20 kg of detectors per cryostat – naturally scalable • Compact Shield – low-background passive Cu and Pb shield with active muon veto 28 Feb. 2014 MALBEK - MAJORANA, Dark Matter 2014, UCLA 3 MAJORANA Detector R&D - MALBEK MAJORANA LowBackground BEGe @ † KURF Customized CANBERRA Broad Energy Germanium Detector 455g Ge geometry optimized for charge collection and noise. Low-background Cu cryostat (J.I. Collar -- UC) nat R&D Goals Study optimized point-contact geometry Test MJD DAQ Understand backgrounds over broad energy range from sub-keV to 3 MeV (surface events, noise, slowfast, validate MJD background model) Explore low-E sensitivity of MJD (68Ge 0νββ background, Dark Matter) † P. Finnerty et al., Nucl. Instr. and Meth. A 652, (2011) 692-695. P. Finnerty et al. IEEE NSS-MIC, (2010) 671-673. 1450 m.w.e. at the Kimballton Underground Research Facility in Ripplemead, VA 28 Feb. 2014 MALBEK - MAJORANA, Dark Matter 2014, UCLA 4 Point Contact Detectors (PPC) Luke et al., IEEE trans. Nucl. Sci. 36 , 926(1989); P. S. Barbeau, J. I. Collar, and O. Tench, J. Cosm. Astro. Phys. 0709 (2007). • Ultra-low background requires PSA rejection of multi-site gamma events • Initially considered coaxial n-type detectors with modest segmentation • Chose P-type Point-Contact (PPC) detectors – – – – – No deep hole; small point-like central contact Length is shorter than standard coaxial detector Simple, cost-effective, low background Localized weighting potential gives excellent multi-site rejection Low capacitance (~ 1 pF) gives superb resolution at low energies n+ p p+ p n+ p+ P-type Coaxial 28 Feb. 2014 P-type Point Contact MALBEK - MAJORANA, Dark Matter 2014, UCLA X Point Contact Detectors (PPC) Luke et al., IEEE trans. Nucl. Sci. 36 , 926(1989); P. S. Barbeau, J. I. Collar, and O. Tench, J. Cosm. Astro. Phys. 0709 (2007). • Ultra-low background requires PSA rejection of multi-site gamma events • Initially considered coaxial n-type detectors with modest segmentation • Chose P-type Point-Contact (PPC) detectors No deep hole; small point-like central contact Length is shorter than standard coaxial detector Simple, cost-effective, low background Localized weighting potential gives excellent multi-site rejection Low capacitance (~ 1 pF) gives superb resolution at low energies b) Multi-site 300 200 200 100 100 0 0 80 60 60 40 40 20 20 0 0 200 28 Feb. 2014 400 300 a) Single-site Current Signal Charge Signal – – – – – 600 1000 1400 Time (ns) 200 600 1000 1400 Time (ns) MALBEK - MAJORANA, Dark Matter 2014, UCLA X 0νββ Backgrounds in Ge Crystals 68Ge(EC, T½ = 270 d) → 68Ga(90% 1.9 MeV β+, T½ = 68 min): Data from MJD R&D detector - MALBEK 28 Feb. 2014 Simulations for the Demonstrator MALBEK - MAJORANA, Dark Matter 2014, UCLA 5 0νββ Backgrounds in Ge Crystals 68Ge(EC, T½ = 270 d) → 68Ga(90% 1.9 MeV β+, T½ = 68 min): Data from MJD R&D detector - MALBEK electron equiv. energy (keVee) 28 Feb. 2014 MALBEK - MAJORANA, Dark Matter 2014, UCLA 5 Using Ge to Search for light WIMPS Fit for a 8 GeV WIMP ( 90% CL exclusion) After cuts CoGeNT (circa 2011) 45 Results from a Search for Light-Mass Dark Matter with a P-type Point Contact Germanium Detector", C. E. Aalseth et al. [CoGeNT Collaboration], Phys. Rev. Lett. 106, 131301 (2011) (point contact) detector. •49.7 kg-days (150 days) •Analysis - ML-based exclusion calculation 30 25 20 15 10 5 0 0.5 •Background uncertainty handled as nuisance pars: Rolke, et al., NIM A 551 (2005) 493 28 Feb. 2014 35 Counts/keV/kg/d •.44 kg modified BEGe 40 1 1.5 2 2.5 3 3.5 Energy (keV) analysis from M.G. Marino, PhD Thesis Univ. of Washington (2010) MALBEK - MAJORANA, Dark Matter 2014, UCLA 6 Using Ge to Search for light WIMPS Fit for a 8 GeV WIMP ( 90% CL exclusion) After cuts CoGeNT (circa 2011) 45 Results from a Search for Light-Mass Dark Matter with a P-type Point Contact Germanium Detector", C. E. Aalseth et al. [CoGeNT Collaboration], Phys. Rev. Lett. 106, 131301 (2011) (point contact) detector. •49.7 kg-days (150 days) •Analysis - ML-based exclusion calculation 30 Do simulations based on direct assays agree with the observed spectum? 25 20 15 10 5 0 0.5 •Background uncertainty handled as nuisance pars: Rolke, et al., NIM A 551 (2005) 493 28 Feb. 2014 35 Counts/keV/kg/d •.44 kg modified BEGe Do we understand all possible backgrounds processes? 40 1 1.5 2 2.5 3 3.5 Energy (keV) analysis from M.G. Marino, PhD Thesis Univ. of Washington (2010) MALBEK - MAJORANA, Dark Matter 2014, UCLA 6 MALBEK Measurement ≠ Simulations 3 Counts / 5.0 keV / 55.2 days 10 102 spectrum measured in shielding at Kimballton model prediction 10 1 10-1 10-2 0 28 Feb. 2014 A. Schubert, Univ. Washington, Thesis 2012 500 1000 1500 2000 MALBEK - MAJORANA, Dark Matter 2014, UCLA 2500 3000 Energy [keV] 7 Background model for MALBEK MAGE / GEANT4 simulations to determine efficiencies for contamination to deposit energy in our detectors. ORCA/Struck data Fit Result 232-Th Lower Chain (224-Ra -- 208-Pb) in GermaniumNat 50k CPU hours 232-Th Lower Chain (224-Ra -- 208-Pb) in StainlessSteel304 8k+ runs, 40+ contaminants, 56 components, 21 materials 232-Th Lower Chain (224-Ra -- 208-Pb) in Teflon Based on assays and estimates of backgrounds 232-Th Upper Chain (232-Th -- 224-Ra) in Zeolite 232-Th Upper Chain (232-Th -- 224-Ra) in StainlessSteel304 238-U Lower Chain I (226-Ra -- 210-Pb) in MoxtekFET 238-U Lower Chain I (226-Ra -- 210-Pb) in StainlessSteel304 238-U Lower Chain II (210-Pb -- 206-Pb) in Brass 3 10 238-U Lower Chain II (210-Pb -- 206-Pb) in LeadAin 238-U Lower Chain II (210-Pb -- 206-Pb) in LeadMod 238-U Lower Chain II (210-Pb -- 206-Pb) in RnExposureOutsideCryostat Counts / 5.0 keV / 55.2 days 238-U Upper Chain (238-U -- 226-Ra) in Brass 238-U Upper Chain (238-U -- 226-Ra) in CopperOFHC 102 238-U Upper Chain (238-U -- 226-Ra) in GermaniumNat 3-H in GermaniumNat 40-K in Brass 56-Co in CopperOFHC 10 60-Co in GermaniumNat 60-Co in NickelSilver 65-Zn in GermaniumNat 68-Ge in GermaniumNat 76-Ge 2vBB in GermaniumNat 1 Cosmogenic muons in KURFExperimentalHall A. Schubert, Univ. Washington, Thesis 2012 10-1 10-20 28 Feb. 2014 500 1000 1500 2000 2500 3000 Energy [keV] MALBEK - MAJORANA, Dark Matter 2014, UCLA 8 Excess of 210 210Pb Pb 210Bi 210Po Counts / 0.5 keV / day T1/2 = 22 years γ 46.5 keV T1/2 = 5 days β 1162 keV end point 206Pb T1/2 = 138 days α 5.3 MeV 10 1 A. Schubert, Univ. Washington, Thesis 2012 -1 10 0 28 Feb. 2014 20 40 60 80 100 120 140 160 180 200 Energy [keV] MALBEK - MAJORANA, Dark Matter 2014, UCLA 9 Possible sources of 210 Pb contamination ancient lead shims (< 1.3 x 10-2 Bq/kg) ultra-low-background tin solder brass pins and connectors (< 4% lead) inner shielding: ancient lead bricks (< 1.3 x 10-2 Bq/kg) 28 Feb. 2014 A. Schubert, Univ. Washington, Thesis 2012 MALBEK - MAJORANA, Dark Matter 2014, UCLA 10 A. Schubert, Univ. Washington, Thesis 2012 Brass connectors 102 10 Counts / 2.0 keV / day Counts / 2.0 keV / day Predicted spectral shapes Inner lead shield 102 10 1 1 10-1 10-1 100 200 300 400 500 600 700 800 Energy [keV] Solder 102 0 Counts / 2.0 keV / day Counts / 2.0 keV / day 0 10 10-1 10-1 28 Feb. 2014 200 300 400 500 600 700 800 Energy [keV] 300 400 500 600 700 800 Energy [keV] 600 700 800 Energy [keV] 10 1 100 200 Lead shims 102 1 0 100 0 100 MALBEK - MAJORANA, Dark Matter 2014, UCLA 200 300 400 500 11 Background model fit of spectrum Results of fit in which normalizations of ~90 PDFs were allowed to float ORCA/Struck data Fit Result 232-Th Lower Chain (224-Ra -- 208-Pb) in Zeolite Counts / 10 keV 238-U Lower Chain I (226-Ra -- 210-Pb) in GermaniumNat 238-U Lower Chain I (226-Ra -- 210-Pb) in MoxtekFET 238-U Lower Chain I (226-Ra -- 210-Pb) in Zeolite 238-U Lower Chain II (210-Pb -- 206-Pb) in GermaniumNat 238-U Lower Chain II (210-Pb -- 206-Pb) in LeadPatch lead shims 3 10 100 Bq/kg instead of 10-2 Bq/kg 238-U Lower Chain II (210-Pb -- 206-Pb) in MoxtekFET 238-U Lower Chain II (210-Pb -- 206-Pb) in Zeolite 238-U Upper Chain (238-U -- 226-Ra) in GermaniumNat 238-U Upper Chain (238-U -- 226-Ra) in MoxtekFET 3-H in GermaniumNat 40-K in Resistor 2 10 46-Sc in CopperOFHC 60-Co in Brass 65-Zn in GermaniumNat 68-Ge in GermaniumNat 76-Ge 2vBB in GermaniumNat 10 χ2 / DOF = 132.8 / 115 P-value = 0.12 1 0 28 Feb. 2014 500 1000 1500 2000 2500 Energy [keV] MALBEK - MAJORANA, Dark Matter 2014, UCLA A. Schubert, Univ. Washington, Thesis 2012 X MALBEK Spectra Before/After with lead shims without lead shims P. Finnerty, UNC, Thesis 2013 28 Feb. 2014 MALBEK - MAJORANA, Dark Matter 2014, UCLA 12 Background model for MALBEK MAGE / GEANT4 simulations to determine efficiencies for contamination to deposit energy in our detectors ORCA/Struck data 50k CPU hours 232-Th Lower Chain (224-Ra -- 208-Pb) in StainlessSteel304 8k+ runs, 40+ contaminants, 56 components, 21 materials 232-Th Upper Chain (232-Th -- 224-Ra) in StainlessSteel304 Fit Result 232-Th Lower Chain (224-Ra -- 208-Pb) in GermaniumNat 232-Th Lower Chain (224-Ra -- 208-Pb) in Teflon 232-Th Upper Chain (232-Th -- 224-Ra) in Zeolite 238-U Lower Chain I (226-Ra -- 210-Pb) in MoxtekFET Counts / 10 keV 238-U Lower Chain I (226-Ra -- 210-Pb) in StainlessSteel304 238-U Lower Chain II (210-Pb -- 206-Pb) in Brass 238-U Lower Chain II (210-Pb -- 206-Pb) in LeadAin 238-U Lower Chain II (210-Pb -- 206-Pb) in LeadMod χ2 / DOF = 97.2 / 114 P-value = 0.87 103 238-U Lower Chain II (210-Pb -- 206-Pb) in RnExposureOutsideCryostat 238-U Upper Chain (238-U -- 226-Ra) in Brass 238-U Upper Chain (238-U -- 226-Ra) in CopperOFHC 238-U Upper Chain (238-U -- 226-Ra) in GermaniumNat 3-H in GermaniumNat 40-K in Brass 56-Co in CopperOFHC 102 60-Co in GermaniumNat 60-Co in NickelSilver 65-Zn in GermaniumNat 68-Ge in GermaniumNat 76-Ge 2vBB in GermaniumNat Cosmogenic muons in KURFExperimentalHall 10 1 0 28 Feb. 2014 500 1000 1500 2000 2500 Energy [keV] MALBEK - MAJORANA, Dark Matter 2014, UCLA A. Schubert, Univ. Washington, Thesis 2012 13 Serendipity - 210 Pb shim “source” with lead shims 68 65 28 Feb. 2014 Zn without lead shims (221 day data set) Ge 68 210 Pb 65 Ge Zn MALBEK - MAJORANA, Dark Matter 2014, UCLA 210 Pb 14 Serendipity - 210 Pb shim “source” with lead shims 68 65 without lead shims (221 day data set) Ge 68 Zn 210 65 Zn 68 Pb 65 Ge Zn Ge 210 65 Zn 68 Pb Ge fast slow 28 Feb. 2014 MALBEK - MAJORANA, Dark Matter 2014, UCLA 14 slow surface events 1000 Voltage (a.u.) 800 600 400 200 ~ 20 keV 0 active volume 7000 8000 9000 10000 11000 12000 13000 Time (ns) n+ dead layer transition region – partial charge collection “Characteristics of signals originating near the lithium-diffused N+ contact of high purity germanium p-type point contact detectors”, E. Aguayo et al. (MAJORANA Collaboration), Nucl. Instr. and Methods A 701 176 (2013). 28 Feb. 2014 MALBEK - MAJORANA, Dark Matter 2014, UCLA 15 a qualitative slow pulse diffusion model Counts D.C. Radford and P. Finnerty 104 103 102 0 10 20 30 40 50 60 Energy (keV) Big first step towards understanding physical mechanism responsible for slow-signals in PPC detectors. P. Finnerty, UNC, Thesis 2013 28 Feb. 2014 MALBEK - MAJORANA, Dark Matter 2014, UCLA 16 • Low-energy 80 keV γ that interacts near the surface of the Ge detector. • Measure drift time in Ge detector using the 356 γ as start time. Rise Time t10-90 (ns) A test using γ-γ coincidence 2000 1800 103 1600 1400 1200 102 1000 800 600 10 400 200 0 0 1 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 Drift Time t10_NaI-10_Ge (ns) Drift Time t10_NaI-10_Ge (ns) 1400 1200 102 1000 800 600 10 400 200 0 0 28 Feb. 2014 10 MALBEK - MAJORANA, Dark Matter 2014, UCLA 20 30 40 50 60 70 80 90 100 Energy (keV) 1 17 Lessons learned from MALBEK • Events interacting in the transition region near the surface can result in slow rise time, energy degraded signals in P-type point contact (PPC) Ge detectors. - Have identified a physical process that can contribute background signals into the WIMP energy region. •A - diffusion model qualitatively describes the observed behavior. Find agreement (within about a factor of 2) for the rise times of the degraded signals, using only measured Li thickness & hole mobilities. • Searching for a WIMP signal with PPC detectors requires: a sub-keV energy threshold and good energy resolution • low backgrounds • efficient ID & rejection of slow pulses (contamination vs. acceptance) • calibration & characterization of the detector (in energy & rise time) • • Detailed predictive simulations based on measured assays have proven powerful tools for understanding detector response. • Work continues to further understand potential systematic effects in PPC detectors. 28 Feb. 2014 MALBEK - MAJORANA, Dark Matter 2014, UCLA 18 Risetime Cuts for MALBEK Flat cut- sacrifice some signal, while optimizing rejection of slow backgrounds. See G. Giovanetti, TAUP 2013 68 Ge Prelimi nary this cut results in a reduced slow pulse contribution in the signal region 55 Fe 65 Zn 68 Ge fast slow slow event cut that is constant in wpar 28 Feb. 2014 MALBEK - MAJORANA, Dark Matter 2014, UCLA 19 MALBEK Spectrum - Best Fit Example best fit MALBEK spectrum with 10 GeV WIMP mass DataExclusion WIMP Mass: 10 GeV (Initial fit) Prelimi nary 12 Total fit L lines (65Zn, 68Ge ) Flat background WIMP signal Counts/keV/kg/d 10 8 6 4 2 0 28 Feb. 2014 1 1.5 2 2.5 3 MALBEK - MAJORANA, Dark Matter 2014, UCLA 3.5 4 4.5 Energy (keV) 20 MALBEK Spectrum - 90% CL WIMP Mass Example MALBEK spectrum with fit to a 10 GeV WIMP at 90% CL DataExclusion WIMP Mass: 10 GeV (Final fit, 0.9 CL, σ : 1.2717e-41 cm ) 2 Prelimi nary 12 Total fit L lines (65Zn, 68Ge ) Flat background WIMP signal Counts/keV/kg/d 10 8 6 4 2 0 28 Feb. 2014 1 1.5 2 2.5 3 MALBEK - MAJORANA, Dark Matter 2014, UCLA 3.5 4 4.5 Energy (keV) 21 Sensitivity to light WIMPS • With WIMP-nucleon σSI (cm2) 221 days livetime, the MALBEK R&D detector (0.45 kg) sees no evidence of a WIMP signal. - See G. Giovanetti, TAUP 2013 10-39 CoGeNT 2013 CoGeNT 2014 ML MALBEK constant cut LUX 85 d MALBEK 99% cut 10-40 MALBE K Prelim inary 10-41 10-42 6 28 Feb. 2014 8 MALBEK - MAJORANA, Dark Matter 2014, UCLA 10 12 WIMP mass (GeV) 14 16 18 22 Sensitivity to light WIMPS -38 221 days livetime, the MALBEK R&D detector (0.45 kg) sees no evidence of a WIMP signal. - See G. Giovanetti, TAUP 2013 • The MAJORANA DEMONSTRATOR (40 kg) may have sensitivity to WIMP nuclear recoils and other non standard model physics if - meets background goals - achieves low energy threshold specs. 28 Feb. 2014 DAMA (no ch.) CDMSlite CDMS II Si SuperCDMS SNOLAB DarkSide-50 MJD 100 kg-yr, 300 eV -39 10 CoGeNT 2014 ML MALBEK constant cut SuperCDMS Soudan LUX 85 d MJD 100 kg-yr, 500 eV MALBE K Prelim inary -40 10 WIMP-nucleon σSI (cm2) • With 10 10-41 10-42 -43 10 10-44 -45 10 -46 10 10 WIMP mass (GeV) MJD simulation assumes 0.001 counts/keV/kg/day continuum and 15 days of 3H activation time ( M.G. Marino, PhD Diss., Univ. of Washington (2010)) MALBEK - MAJORANA, Dark Matter 2014, UCLA 22
Similar documents
The MAJORANA DEMONSTRATOR
University of North Carolina, Chapel Hill, North Carolina and TUNL
More informationAutorenverzeichnis
Branchini, Paolo . . . . . . . . . . . . -Koll 5 Brandes, Jürgen . . . . . . . . . . . •T 99.3 Braun, Vladimir . . . . . . . . . . . . . T 29.3 Brayeur, L. . . . . . . . . . . . . . . . . -Koll 20...
More information