Dark Matter 2016 Los Angeles, CA 17-19 February 2016 Results from DarkSide-50 with underground argon Alden Fan UCLA for the DarkSide collaboration 1
DarkSide WIMP dark matter search using direct detection Dual-phase Liquid Argon Time Projection Chamber (LArTPC) Ultra low background Deep underground at LNGS Low-background materials, including Ar target Powerful background rejection Pulse Shape Discrimination (PSD) Ionization/Scintillation ratio (S2/S1) Surface rejection using 3D position reconstruction Active neutron and muon vetoes In situ background measurement 2
Why Argon? Relatively dense Ionization and scintillation Easy to purify (chemically) Scales to large mass Transparent to its own scintillation light Exceptional discrimination power PSD S2/S1 Main challenge: 39 Ar contamination Atmospheric argon: high concentration of 39 Ar to 40 Ar cosmogenically activated (1 Bq/kg) β decay (T1/2: 269 yr, Q: 565 kev) Underground argon: significantly reduced 39 Ar activity 3
Multi-stage DarkSide program Gran Sasso National Laboratory, Italy DarkSide- 2011-2013 DarkSide-50 2013-201x DarkSide-20k 2020-202? ARGO 202?-20?? See C.J. Martoff talk 4
Dual-phase LArTPC 6 μs NR S1 time [μs] [arb] [arb] 6 μs ER PSD parameter: F90 = fraction of light in first 90 ns time [μs] 5
Dual-phase LArTPC [arb] 65 μs S2 S1 e - e - e - e - e - e - Eextr Edrift NR time [μs] [arb] 65 μs ER S2 allows for 3D position reconstruction and additional discrimination power time [μs] 6
DarkSide-50 TPC 46 kg active volume 36 cm diameter, 36 cm height 38 3 PMTs Cold pre-amps High reflectivity Teflon walls Fused silica anode and cathode windows Coated with transparent conductor (Indium Tin Oxide) All inner surfaces coated with wavelength shifter (Tetraphenyl Butadiene) 0.2 kv/cm drift, 2.8 kv/cm extraction 7
Vetoes Liquid Scintillator Veto 4 m diameter sphere Boron-loaded: PC + TMB 1 8 PMTs Active neutron veto tag neutrons in TPC in situ measurement of neutron BG Neutron and gamma shielding Water Tank 11 m diameter x m high Existing Borexino CTF tank 80 PMTs Active muon veto tag cosmogenic neutrons Neutron and gamma shielding 8
DarkSide-50 Assembly e-50 Assembly 9
Calibrations Insertion system deployed Sept 2014 Calibrations: 83mKr (injected), 57Co, 133Ba, 137Cs, AmBe, AmC Validate NR band obtained from SCENE Evaluate Light Yield f90 Validate MC 1 20 40 60 80 0 120 P. Agnes et al. / Physics B 743 (2015) 456 466 461 Energy [kevnrletters ] FIG. 2. 160 left: CALIS installation inside the CRH radon-suppressed clean room atop the WCD. right: Photograph taken 140 180after200 a camera looking into the LSV from the WCD. It shows a source deployed next to the cryostat of the LAr-TPC. 7 the first be S2, and the in DarkSide-50 0.9 pulse is assumed to be S1, the second to AmBe 83m + 39Arthe source is brough 95 Source Deployment: In order to deploy a radioactive source next to Kr the LAr-TPC, third to be S3. 0.8 96 the radon-suppressed clean room atop the WCD tank, and mounted in a source holder connected t Several corrections are applied to the S1 and S2 CRH integrals to aclight Yield: 97 deployment device within CALIS. This is lowered into the LSV through a dedicated access port, to a position 0.7 count for geometrical variations of light production and collection. 98 to the cryostat. The source holder is on the end of a 60 cm long articulated arm, which can be moved and rota ± 0.4 PE/keV @ nulla slightly field pressu Due surface,thelight col99 position source in 3-D relative to the cryostat (Fig.7.9 2, right). Throughout the deployment 0.6to total internal reflection at the liquid nitrogenthe atmosphere to avoid exposing the liquid@ scintillator to oxygen or w lection of scintillation pulses varies by 19% 0 between top andhas to be maintained in order7.0 ± 0.3 PE/keV 200 V/cm 0.5 1 which would degrade it. All materials used for CALIS, such as stainless steel, teflon, and viton, that com bottom of the TPC. An empirical z-dependent correction, derived 2 contact with the scintillator are certified for contact with the liquid scintillator. The system includes several s 83m 39 0.4 from Kr and Ar calibration data and normalized totothe center 3 features ensure safe deployment and retrieval of the source without affecting the stability and perform 4 of the LSV. After data taking, the deployment device is retracted and a sequence of N2 purging and evacu of the0.3 TPC, is applied to S1. 5 removes all traces of scintillator from the deployment device. Thereafter the source holder can be extracted fro Electronegative impurities in the LAr capture drifting electrons. 0.2 6 deployment device and the radioactive source returned to storage. The CALIS concept has the built-in flexibil This results in the number of drifting electrons, and hence S2,devices de- next to the cryostat, e.g. the deployment of a (d,d) neutron generator is considered 7 deploy different 0.1 creasing exponentially with the time taken 8to future. drift between the Calibration Campaigns: After the calibration device and the deployment procedures were approved by the 0 point and the liquid-gas interface.9 We interaction fit for this elec50 0 150 200 250 300 350 400 450 1 Side scientific committee, two extensive calibration campaigns were performed during October through Dece tron drift lifetime, then correct S2, normalizing to the top S1 of through the [PE] February 2015. The performance and stability of the LSV and LAr-TPC was not aff 111 2014 and January TPC. Our very high electron mean drift lifetime a maxi- campaigns. 112 byinduces the two calibration Fig. 3. The primary scintillation (S1) spectrum from a zero-field run of the mum 7% correction for the runs acquired through February 2014. Li(p,n) in SCENE 241
UAr Extracted from Doe Canyon CO2 wells Transported to Fermilab for distillation 6 yr effort to obtain 155 kg of UAr Shipped to LNGS by sea (15.6 kg by air) Map data 2015 Google, INEGI 500 mi 11
AAr vs. UAr - null field s] kg Events / [50 PE 1 2 3 4 5 6 7 AAr Data (0 V/cm) UAr Data (0 V/cm) 8 0 2000 4000 6000 8000 000 12000 14000 16000 18000 20000 S1 Late [PE] LY unchanged from AAr to UAr 12
AAr vs. UAr - 200 V/cm s] kg Events / [50 PE 1 2 3 4 5 6 7 AAr (200 V/cm) UAr (200 V/cm) 8 0 2000 4000 6000 8000 000 12000 S1 [PE] 13
AAr vs. UAr - 200 V/cm s] kg Events / [50 PE 1 2 3 4 5 6 7 AAr (200 V/cm) UAr (200 V/cm) AAr (200 V/cm, LSV Anti-coinc.) UAr (200 V/cm, LSV Anti-coinc.) 8 0 2000 4000 6000 8000 000 12000 S1 [PE] Slight excess at 39 Ar endpoint 14
39 Ar depletion s] kg Events / [50 PE 1 2 3 4 5 6 7 8 AAr (200 V/cm) UAr (200 V/cm) AAr (200 V/cm, LSV Anti-coinc.) UAr (200 V/cm, LSV Anti-coinc.) MC total (Global Fit) 85 Kr (Global Fit) 39 Ar (Global Fit) saturation 0 2000 4000 6000 8000 000 12000 S1 [PE] MC fit prefers 85 Kr component to explain excess See P. Agnes talk for MC 15
39 Ar depletion 39 Ar reduction factor: 1400 s] kg Events / [50 PE 1 AAr Data 2 3 4 5 6 7 214 Bi 609 kev (C+P) 60 Co 1.17 MeV (C+F) 60 Co 1.33 MeV (C+F) 40 K 1.46 (P) MeV C: Cryostat P: PMTs F: Fused Silica 214 Bi 1.77 MeV (C+P) UAr Data UAr MC Total MC MC 85 Kr 39 Ar 208 Tl 2.62 (P) MeV 8 0 2000 4000 6000 8000 000 12000 14000 16000 18000 20000 S1 Late Fitted 85 Kr activity in UAr: 2.05 ± 0.13 mbq/kg Fitted 39 Ar activity in UAr: 0.73 ± 0.11 mbq/kg 39 Ar activity in AAr: 00 mbq/kg [PE] 16
85 Kr delayed coincidences 85 Kr: 0.4% BR to 85m Rb (T1/2 = 1 μs, 514 kev γ) Signature: two S1s (β+γ) in delayed coincidence Events / [0.1 µs] 120 0 80 60 Decay time 1.64 ± 0.12 µs 40 20 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 Δt [µs] Rates Observe: 33.1 ± 0.9 events/d From spectral fit: 35.3 ± 2.2 events/d 17
Dark Matter search I Dark Matter search with UAr begins immediately after turning on fields. First WIMP search with UAr AmC calibration 83m Kr calibration Accumulated livetime with UAr ~1.5 Hz trigger rate, predominantly ER events 18
F90 Use analytic model for F90 distributions Fit to high statistics AAr data Scale to UAr data Derive 0.01 ER leakage events / S1 bin S1: [60.00, 70.00] PE S1: [250.00, 260.00] PE 4 3 AAr UAr 4 3 AAr UAr 2 2 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 f 90 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 f 90 19
NR acceptance Cuts: select single scatters (single S1 + single S2) with no signal in veto. Efficiencies evaluated using UAr data + AmBe data + MC Dominant acceptance loss: accidental coincidences in veto NR Acceptance 1 0.9 0.8 0.7 0.6 Energy [kev nr ] 20 40 60 80 0 120 140 160 180 200 0.5 0.4 0.3 0.2 0.1 Total Cut Efficiency f 90 NR Acceptance Overall NR Acceptance 0 0 50 0 150 200 250 300 350 400 450 500 S1 [PE] 20
Veto efficiency Veto neutrons via thermalization or capture in LSV >99.1% efficiency to veto neutrons from capture alone (AmBe + simulation) arxiv:1512.07896 Will increase efficiency using neutron thermalization signal Analysis in progress using new AmC source data (Dec 15 - Jan 16) AmC 21
Dark Matter search II 70.9 live-days, 36.9 kg fiducial volume Expect <0.15 ER leakage events f 90 1 0.9 0.8 Energy [kev nr ] 20 40 60 80 0 120 140 160 180 200 WIMP Search Region 250 0.7 0.6 0.5 0.4 0.3 0.2 0.1 39 Ar + 85 Kr + γ 50% 90% 99% 200 NR acceptance contours from AmBe in DS50 + SCENE 150 0 50 0 50 0 150 200 250 300 350 400 450 S1 [PE] 0 No events in the WIMP search region. 22
Dark Matter search III Combined limit of UAr and AAr exposures in DS50: minimum at 0 GeV/c 2 : 2 x -44 cm 2 [cm 2 ] σ 41 42 43 PandaX-I (2014) PICO (2015) WARP (2007) DarkSide-50 (AAr, 2014) DarkSide-50 (UAr, 2015) DarkSide-50 (combined) 44 45 CDMS (2015) XENON0 (2012) LUX (2015) 46 2 3 M χ 2 [GeV/c ] 4 arxiv:15.00702 23
DS50 3 yr projection [cm 2 ] σ 41 42 43 44 CDMS (2015) PandaX-I (2014) PICO (2015) XENON0 (2012) WARP (2007) DarkSide-50 (AAr, 2014) DarkSide-50 (UAr, 2015) DarkSide-50 (combined) DarkSide-50 (3 yr proj.) 45 LUX (2015) 46 2 3 M χ 2 [GeV/c ] 4 24
Summary DarkSide-50 performed first ever dark matter search using Underground Argon. Measured 39 Ar level in UAr to be factor 1400 smaller than in AAr. DarkSide-50 has the strongest WIMP limit using an Ar target, third best limit. Currently in stable WIMP search mode. 25
Backup 26
50d AAr DM search 1422 ± 67 kg-day exposure f 90 1 0.9 Energy [kev nr ] 20 40 60 80 0 120 140 160 180 200 25000 0.8 0.7 50% 90% 99% 20000 0.6 0.5 15000 0.4 0.3 000 0.2 5000 0.1 0 50 0 150 200 250 300 350 400 450 S1 [PE] 0 27
S2/S1 f 90 1 0.9 0.8 S2/S1 cut calibrated on AmBe data in DS50 50% NR acceptance in S2/S1 Energy [kev nr ] 20 40 60 80 0 120 140 160 180 200 50 0.7 0.6 50% 90% 99% 40 0.5 30 0.4 0.3 20 0.2 0.1 0 50 0 150 200 250 300 350 400 450 S1 [PE] Should we ever see a potential WIMP signal, S2/S1 cut is powerful additional handle. 0 28