direct detection

Alvaro E Chavarria KICP at The

1 Overview

• Dark matter direct detection. • DAMA and new NaI experiments. • Recent results from PICO. • Recent results from DarkSide. • DAMIC experiment. • The future of direct detection.

2 Dark matter • is needed for CMB cosmology... • ... and structure formation. • ... and to explain galaxy rotation curves. • ... and evident from gravitational lensing. • Overall 5.5 times more dark than baryonic matter. • Local density may be non-zero. • Could be made of particles... • ... that could interact with SM particles. • Numerology: “WIMP Miracle,” asymmetric DM, etc.

3 Direct detection Xe Ge Ar Si Ne χ Nucleus From on Earth arXiv:1310.8327v2 [hep-ex] galactic Recoil Mass: 100 GeV halo

B.Loer Thesis

M c2 E keV Mass: 1 TeV ⇠ GeV

4 Direct detection • Large target mass M T many targets. • Count for long time T . • Lowest possible threshold E th , to increase fraction of recoil spectrum probed f . • Lowest possible background <1 event kg-1 y-1: low radioactivity environment, nuclear/electron recoil discrimination. • Large atomic mass A increases total rate (coherent scattering) but increases minimum WIMP mass probed M min . 5 Technologies Noble liquid TPCs Ionization LUX/LZ DAMA/LIBRA CoGeNT XENON ANAIS CDMS-HV DarkSide SABRE CDEX ArDM DM-Ice DAMIC Panda-X KIMS DEAP/CLEAN Superheated Phonon + Ionization XMASS PICO CDMS-II Phonon + Scintillation SuperCDMS CRESST EDELWEISS Nuclear recoil discrimination 6 4%New%NaI%Experiments% (slide%from%Walter%LocationsPe'us,%spoke%here)%% Soudan:% Boulby:% •%CDMS% •%DRIFT% •% CoGeNT% Modane:% Canfranc:% YangYang:% •%EDELWEISS% •%KIMS% KIMS •%ANAIS% ANAIS •% ArDM% •%Rosebud% Kamioka:% Homestake:% SNOLAB:% Gran%Sasso:% •%XMASS% •%LUX/LZ% •%DEAP/CLEAN% •%CRESST% JinDPing:% •%PICASSO% •%DAMA/LIBRA% •%PandaLX% •%PICO% •% DarkSide% •%CDEX% •%DAMIC% •%XENON% •% SuperCDMS% ANDES:% (planned)% Stawell:% SABRE•% SABRE%

South%Pole:% •DM-ICE% DMLICE% 5/21/15% CIPANP%2015%/%Harry%Nelson% 9%

7 Direct detection XENON 10 S2 (2013) 10 39 CDMS-II Ge Low Threshold (2011) PICASSO 2 CoGeNT (2012) 40 PICO-2L (2012) cm 10 (2015) CDMS Si (2013) 10 ρσ 41 SuperCDMS DAMA SIMPLE (2012) section (2014) COUPP (2012) 42 CRESST WARP (2008) 10 ZEPLIN-III (2012) cross while background free: DS-50 (2014) 43 10 EDELWEISS (2011) M Xenon100 (2012) CDMS II Ge (2009) ⇢ nucleon 10 44

/ MT TA f(Eth,M) PANDAX (2014-2015) LUX (2013) d⇢ 45 10

WIMP- = const. dM

10 46 1 10 100 1000 104 WIMP Mass GeV c2 1 EthA f(E ,M ) Mmin where th is maximum. / vesc 2 r M 8 3. Annual Modulation

2-4 keV Signal Modulation DAMA/LIBRA ≈ 250 kg (0.87 ton×yr) Baryons - Baryons travelorbit together ‘together’ in roughly circular orbitsroughly with smallcircular velocity orbits dispersion DM Stars small velocity dispersion

Residuals (cpd/kg/keV) - Dark matter particles travel individually with no circularHalo dependence DM and large velocity dispersion Time (day) 2-5 keV orbit ‘individually’ 0 km/s 220 km/s DAMA/LIBRA ≈ 250 kg (0.87 tonno×yr) circular preference large velocity dispersion Vθ (at out galactic radius)

- As a result, the flux of WIMPs DAMA/LIBRA passing through Earth modulate over the course of a year as Earth Residuals (cpd/kg/keV) NaI scintillating rotates around the sun. Time (day) crystals in Gran Sasso 2-6 keV

DAMA/LIBRA ≈ 250 kg (0.87 ton×yr) 9/10/2013 - TAUP 2013 - Jodi Cooley 30 Residuals (cpd/kg/keV)

Time (day) Observe a highly significant (9 σ) annual modulation, Figure 1: Experimentalconsistent model-independent with the “model residual independent rate of theDMsingle-hit signal”scintillation events, measuredT = by0.999 DAMA/LIBRA,1,2,3,4,5,6 ± 0.002 y and maximum in the ~ (2 June – 4), 2nd (2 ± – 7 5) d and (2 – 6) keV energy intervals as a function of the time. The zero of the time scale is January 1st of the first year of data taking of the9 former DAMA/NaI experiment [15]. The experimental points present the errors as vertical bars and the associated time bin width as horizontal bars. The superimposed curves are the cosinusoidal functions 2π behaviors A cos ω(t − t0) with a period T = ω = 1 yr, with a phase t0 =152.5day (June 2nd) and with modulation amplitudes, A, equal to the central values obtained by best fit over the whole data including also the exposure previously collected by the former DAMA/NaI experiment: cumulative exposure is 1.17 ton × yr (see also ref. [15] and refs. therein). The dashed vertical lines correspond to the maximum expected for the DM signal (June 2nd), while the dotted vertical lines correspond to the minimum. See text.

5 DMDICE%Experiments%

DM-ICE17 DM-ICE37 DM-ICE250 (January 2011 – present) (April 2014 – present) (future)

First dark matter R&D testbed for NaI Science result experiment in South detectors - Definitive test of DAMA Pole ice - Crystal background dark matter claim - Demonstrated viability - Light yield and advantage of - PMT/lightguide environment configurations 5/21/15% CIPANP%2015%/%Harry%Nelson% 10%

10 172 Emily Shields et al. / Physics Procedia 61 ( 2015 ) 169 – 178

Fig. 4. Design for the SABRE experimental setup for one small (1-2 kg) crystal. Opposite faces of the crystal will be coupled to two 3” PMTs operated in coincidence. The PMT pair and crystal will be enclosed in an air-tight enclosure, which will be suspended in a liquid veto detector currently under construction.SABRE%concept% This detector will be outfitted with 10 8” R5912 Hamamatsu PMTs to detect the scintillation light. The scintillation detector will be surrounded in turn with lead and steel passive shielding.

Fig. 5. Left: Design of the SABRE experimental setup. The liquid scintillator vessel is a φ1.5 m 1.5 m stainless steel cylinder × containing 2 tons ofNaI liquid%in% scintillatorScint (in blue)..%Veto% The NaI(Tl) detectors (in brown) are installed in the center of the veto vessel, and 10 ∼ Hamamatsu 8” PMTs are used to collect the veto scintillation light. The whole setup is shielded from external backgrounds by 20- ∼ 25 cm of passive shielding (in dark gray). Right: An illustration of the DarkSide-50 experiment. A φ4 m liquid scintillator detector (the sphere in the center) is contained inside a φ11 m 10 m water tank, which hosts the DarkSide-50 experiment. The SABRE NaI(Tl) 5/21/15% × CIPANP%2015%/%Harry%Nelson% 11% crystal detectors can be installed between the DarkSide-50 TPC and the walls of the veto sphere. DarkSide-50 also has a number of facilities already in place, such as scintillator handling and purification systems, which can be shared with SABRE. 11 crystals. The second phase will be to conduct a dark matter measurement with 50-60 kg of target material in an underground setting, such as LNGS or SNOLab. PICO-60

! Fill of 37 kg CF3I at SNOLAB completed April 2013 ! Results presented here are preliminary

PICO - Jeter Hall - CIPANP 2015 May 20, 2015 10 12 PICO-60 ! Large number of background events ! Significant number of events with AP~1, but inconsistent with neutron calibration distributions ! Similar to COUPP4 backgrounds ! Not spatially uniform 1

PICO - Jeter Hall - CIPANP 2015 May 20, 2015 11 13 PICO-60 Model independent demonstration that implications for DAMADAMA signal cannot be Iodine recoils PRELIMINARY* ! Using DAMA spectrum between 2 and 6 keV ! Applying DAMA iodine quenching factor (0.09) results in expectation of 49 recoils above 22 keV ! PICO-60 observes <4.1 events at 90% C.L.

PICO - Jeter Hall - CIPANP 2015 May 20, 2015 16 14 PICO-2L

! Filled with 2 liters C3F8 in September 2013 ! Stable operations at SNOLAB from October 2013 to May 2014 resulting in over 250 kg day exposure with thresholds of 3, 6, and 8 keV ! Reincarnation of COUPP4 chamber with substantial improvements and new target ! arXiv 1204.3094; PRD 86, 052001 (2012)

PICO - Jeter Hall - CIPANP 2015 May 20, 2015 6 15 PICO-2L results arXiv:1503.00008, accepted in PRL

−36 ! Candidate events are 10

inconsistent with WIMP −37

] 10 ) 2 b PICASSO 2012 b χ ! KS p-value of 0.04 for timing (χ → SIMPLE 2014 −K (soft) PICO 2L −38 distribution of events 10 Super PICO 2L ! Limits are derived 0 bkg −39 −K (hard) 10 − ) Super + W CMS (A−V) W χ (χ → −40 IceCube

proton cross section [cm 10 −

−41 10 PICO 250L , 3 keV SD WIMP F 8 C 3 −42 10

1 2 3 4 10 10 10 10 WIMP mass [GeV/c2]

PICO - Jeter Hall - CIPANP 2015 May 20, 2015 9 16 DarkSide 50 Radon-free clean room

Water Cerenkov Detector

Liquid scintillator Veto

Inner detector TPC

17 TPC for WIMPs: DS50 Nuclear Recoil excites and ionizes the noble liquid, producing scintillation light (S1)that is detected by the photomultipliers

χ Ar S1

Δt ~ 7 us

Scintillation light proportional to recoil energy 18 TPC for WIMPs: DS50

The ionized electrons that survive e- recombination are drifted towards the liquid-gas interface by the electric field

Electron Drift Velocity ~ 0.94 mm/us Max Drift Time ~ 373 us

19 TPC for WIMPs: DS50

The electrons are extracted into the gas region, where they induce electroluminescence (S2)

S1 S2 Δt ~ 30 us

Drift Time

The time between the S1 and S2 signals gives the vertical position 20 −40 ] 10 2

[cm 41 σ 10−

WARP (2007)

DS-50 AAr −42 10

DS-50 (2014) 10−43 PandaX-I (2014) Result XENON-100 (2012) CDMS (2010) 10−44 LUX (2013)

1.42 ton-day −45 10 arXiv:1410.0653

exposure 10−46 1 10 102 103 104 2 M χ [GeV/c ]

1 90 f 35000 0.9 50% Acceptance, < 0.1 DM Search Box 0.8 30000 ER Leakage at 102 PE, 50% 0.7 47 keVr 90% 25000 0.6 39 20000 0.5 Ar 0.4 15000 < 0.01 ER / 5 PE bin 0.3 10000 0.2 5000 S1 > 80 PE 0.1 < 0.01 ER / 5 PE bin 0 0 100 150 200 250 300 350 400 450 90% NR Acceptance S1 [PE] 21 Entries 171011

35

140 30 z [cm] 120 25 Underground Ar 100 20 80 15 60

10 Underground Core (4 kg) 40 Hint of 39Ar spectra visible 5 20 0 0 0 50 100 150 200 250 300 39Ar < 3.3 mBq/kg r2 [cm2]

10-1 AAr (200 V/cm, 44 kg) UAr (200 V/cm, 44 kg) 10-2 UAr (200 V/cm, 4 kg core) Previous AAr

-3 exposure 10 > 300x Reduction Events/50 PE/kg/sec equivalent to 1.2 -4 10 ton-years of UAr 39 10-5 Ar Beta Spectrum ? 10-6 0 1000 2000 3000 4000 5000 6000 7000 S1 [PE] 22 DAMIC Charge-coupled devices (CCDs) as low threshold, low background particle detectors.

WIMP 90% exclusion limits 1

-1 SUPERCDMS(2014) -2 10 Most sensitive < 3 GeV c for Mχ 10-2 0.3 kg d DAMIC(2012) DAMIC(2014) Test setup at SNOLAB 10-3 CDMSII-Si(2013) already shows great 6 kg d 10-4 CDMSLite(2013) potential. DAMIC100(2016) 10-5 Complementary to 30 kg d Will directly probe Xenon searches 10-6 CRESST(2014) WIMP-nucleon cross-section / pb LUX(2013) the possible signal in 10-7 1 10 102 CDMS II-Si. WIMP mass / GeV c-2 23 SNOLAB installation

2 km of rock CCD

Poly- Si support VIB Polyethyleneethylene Kapton signal cable Lead Pb

Copper Lead block V bar Kapton e signal cable s Cu box s with CCDs 42 21 e cm cm l J. Zhou

Cu vacuum vessel

24 y Detector x

Charged particles produce ionization in CCD bulk. z 3.62 eV for

e-h pair. x Charge Charge drifted up and held at gates. collected by y each pixel on z CCD plane is

pixel read out. σxy x~2 e- RMS read-out noise. σxy 25 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 1840 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000

image 1820

2080 10000

1140 90005

8000 1800

1120 70004 2060 image 6000

1100 3

5000 1780

4000

1080 2 2040 3000

Diffusion Diffusion limited 2000 1760

1060 1 image 6 keV back 6 keV front 1000 measuredEnergy by pixel / keV 10 5 30 25 198020 2000 15 2020 2040 2060 2080

image 2020 30 1740 1330 30 960 940 920 900 880 980 4220 26 Back 25 25 1320 2000

μ 20 4210 20

1310 1980 limited 1060 1120 1100 1080 1140 Diffusion

1515 4200 image 1300 10 10

e Energy measuredEnergy by pixel / keV 4190

1290 Front

5 5

α X-ray? 4180

1280 n, WIMP?

4180 4190 4200pixels 50 4210 4220

Particle tracks Particle

1280

1290

1300

1310

1320 1330 CCD Performance CCDs are manufactured with very high resistivity silicon: Low radioactive backgrounds. Low dark current (0.01 e- / pix / day). Very few (if any) defects in the silicon lattice.

Distribution of pixel values in image

105 30 ks exposure

104 blank

103 6.7 eVee ±6% 102 RMS noise!

10

1 0 50 100 150 200 Energy measured in pixel / eVee >95% of the image 10794 images acquired over good quality. 126 days. All good. 27 55Fe source spectrum in Chicago chamber

105 X-rays Mn K! Mn K" 104 Noise Calibration data to X-ray lines Mn K Cr K! escape lines 55 60 keV 103 Si K! Fe Cl K!

102 10 Si Kα 241 Am 10 O Kα 1 Reconstructed energy / keV 1 0 1 2 3 4 5 6 7 Ca Kα Energy / keV

Mn Kα from front and back Al Kα 7 200 1 10 6.8 Energy / keV 180 6.6 160 C Kα (0.28 keV) 6.4 Energy / keV 140 6.2 120 6 E resolution: 100 5.8 80 53 eV at 5.9 keV from front! 5.6 60 Fano = 0.13. 5.4 40 5.2 Front Back 20 5 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 / pixels Depth reconstruction. σxy 28 σxy = 1.4 z = 675 μm Spectrum from 500 keV γ-rays 14 Binding energies neutrons, γs 12 2 K-shell e- 10 - Spectrum from 124Sb source 2 L-shell s e

700 8 Counts 600 - BeO target 6 6 L-shell p e Silicon atom 500 Al target 4 Free electron 400 - 2 4 M-shell e 300 3.2 keVr 0 200 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deposited energy / keVee

100 Sb-Al compared to expected Compton background 600 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Energy / keVee 500 Sb-Al data E =135 keV sim 24 keV neutrons from γ 400 9Be(γ,n) reaction. 300

γ background reveals 200

Compton features near 100 Drop at 150 eVee

0 threshold. 0 1 2 3 4 5 6 7 8 9 10 Energy / keVee 29 Detection efficiency Detection1x1, efficiency Likelihood vs. depth m

µ 1.2 600 Acceptances for different acquisition modes

1 500 Depth / 1 400 0.8 Acceptance 0.8 300 0.6 Simulated 0.6 1x100 200 0.4 1x1 0.4 100 1x1 0.2 0.2 0 0 0 0.05 Acceptance0.1 0.15 0.2 for0.25 1x100,0.3 0.35 3σ0.4-seed0.45 0.5 Energy / keV 140 eVee m 0 µ 600 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 1.2 Simulated energy / keV

500 1 80 eVee 400 0.8 Significant improvement 300

Simulated depth / 0.6

200 0.4 in threshold! 100 1x100 0.2 Already acquired some 0 0 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 data in SNOLAB. Simulated energy / keVee 30 Radioactivity in CCDs

64 keV 1.2 MeV β 210 210 210 Sequence of s Pb Bi Po starting in the same τ = 5 d 1/2 pixel of the CCD in 0.22 MeV 1.7 MeV 32Si 32P 32S different images. τ1/2 = 14 d

32 32 30000 25000 20000 15000 10000 5000 0 Si - P candidate 4890 4888 4886 E1 = 114.5 keV 4884

Δt = 35 days 4882 Cluster #79 4880 (xo, y o) 4878 22000 20000 18000 16000 14000 12000 10000 8000 6000 4000 2000 0 4876 Decay 4876 4874 1360 1355 1350 1345 1340 1335 1330 4872

point 4870 Cluster #51

4868 E2 = 328.0 keV 4866 1364 1362 1361 1360 1358 1357 1365 1363 1359 1356 1355 31 β-β coincidences Search in 57 days of data in 1 CCD: 210Pb < 33 kg-1d-1 (95% C.L.) arXiv: 32 +110 -1 -1 Si = 80 -65 kg d (95% C.L.) 1506.02562

3 5 pairs pairs N Data(13 events) N Data(13 events) Expected accidentals 2.5 Expected accidentals (6.5 events) 4 (6.5 events) 2 32 32 3 32 32 1.5 Si– P Si– P 2 1

1 0.5

0 0 0 2 4 6 8 10 12 14 16 18 20 0 10 20 30 40 50 60 70 Cluster distance / pix Δtpair / day 100 kg-1 d-1 of 32Si ~1 dru at low energies. Spatial correlations will allow DAMIC to veto these decays. (limitation for other silicon technologies?) 32 DAMIC100

We have 24, 16 Mpixel, 675 New Cu box fits μm CCDs. Each is 5.5 g. 18 CCDs inside current SNOLAB infrastructure.

Example in DESI package.

33 Status at SNOLAB

We have installed DAMIC100 copper box with four 8 MPixel, 675 μm thick CCDs at SNOLAB to study backgrounds.

34 Future WIMP searchesLZ%Design%Overview% SI “High”-massBeyond WIMPs G2 Dark Matter LZ Better Low Mass WIMP Detectors Increase targetSingle mass Carrier as much Resolution ! Achieving good resolution towards single charge carrier pairs enables a separation as possible: multi-tonbetween the dominant scale electron backgrounds and the nuclear recoil signals ! Proving this technological advance will enable the next generation scientific experiment Xenon and Argon! See detectors. Pyle, Figueroa-Feliciano, Sadoulet arXiv:1503.01200 for technology description

SI “Low”-mass WIMPsG2%Baseline%Technology% 5/21/15% Next%Genera3on%CIPANP%2015%/%Harry%Nelson% 30%

Lowest (~10 eVee) threshold ionization detectors: DAMIC, CDMS-HV Ge.

SD WIMPs: SuperCDMS - Jeter Hall - CIPANP 2015 May 22, 2015 10 Large C3F8 bubble chamber (PICO 60 /PICO 250). 35 Scaling up CCDs CCD detector technology is quite scalable. Current semiconductor companies can produce 10 cm x 10 cm x 0.1 cm, 20 g CCDs. This box can hold 400 g of Si. A CCD is only read out <5% of the time. Multiplexer allows to read out 20 CCDs with a single controller. This box can be read out with one controller channel.

36 Scaling up CCDs

6.4 kg of CCDs can be controlled with 16 channels. Easy cryogenics (100 K). Man power for building + testing at the scale of DECam. Main challenge is background. 37 Conclusions • Much progress in dark matter direct detection over past year. • Ongoing efforts at have world-best limits for SD interactions and low-mass WIMPs. • DAMIC aims to lead low-mass WIMP searches for current (DAMIC100) + next generation detectors.

• PICO-60 with C3F8 will lead SD searches. • SuperCDMS will move to SNOLAB. Complementary to other technologies.

38 Thank you!

.. and to R. Saldanha, J. Hall and H. Nelson for many of the slides!

39