Dark Matter Direct Detection: a Status Review

Dark Matter Direct Detection: a status review

Kaixuan Ni University of California San Diego

July 2019, DPF 2019 We used to think WIMPs are the perfect candidate for dark matter and detecting them is just a matter of target mass and time (exposure)

incoming WIMPs WIMPs detector

nuclear recoils

H. Baer, et al., arXiv:1407.0017 Phys. Rep. (2015)

2 14

Dark matter direct detection provides a very clean way to probe the nature! of DM. Figure 11. Background sources and shielding in a typical direct detection experiment 2 2 Mχ = 50 GeV/c , σχ-n = 1e-46 cm 3 10 • DM local density: 0.3~0.4 GeV/cm Xe (A=131) Ge (A=73) /Ton/Year] • WIMP dark matter: GeV~TeVneutrinos form a fundamentalnr lower bound on theAr (A=40) cross section for background-free WIMP detection [43].

Next generation experiments1 will15 have evt/ton/year sensitivity > 10 keV within an order of magnitude of the neutrino signal for • Velocity: O(100) km/s most of the mass range, and will actually detect the 8B solar neutrino signal. Rate [Events/keV • Simplest spin-(in)dependentFinally, models another method to deal with backgrounds is to exploit the fact that the Earth is moving through the dark matter that surrounds10−1 our galaxy, yielding a “WIMP wind” that appears to come from the constellation • Produce low energy nuclearCygnus. recoils This should, in principle, create a small “annual modulation” in the detected WIMP rates, as well as a somewhat larger daily modulation, as shown in Fig. 12. However, if such e↵ects were detected in an Annual and diurnal modulation • −2 10 0 10 20 30 40 50 60 70 80 Nuclear Recoil Energy [keV ] nr

June 2

dR ⇢ vesc d = 0 N (v, E )vf(v)dv dE m m dE R R N Zvmin R

Dec 2 3 Figure 12. Schematic of the possible sources of annual modulation (left) and daily modulation (right) e↵ects if WIMPs are detected in direct detection experiments

experiment, there would still have to be a convincing demonstration that there are no such modulations in background sources.

Community Planning Study: Snowmass 2013 Dark Matter “Gold Rush”

Heat

nuclear recoils

Charge Light

4 Dark Matter “Gold Rush”

Target Materials ๏Xe, Ar, Ne, He ๏Ge, Si, He ๏C3F8, CF3I ๏Ge, Si, CaWO4, NaI ๏NaI, CsI ๏CS2, CF4, CHF3 Heat

nuclear recoils

Charge Light

4 Dark Matter “Gold Rush”

Target Materials Techniques ๏Xe, Ar, Ne, He ๏Noble liquids ๏Ge, Si, He ๏Low-threshold ๏C3F8, CF3I ๏Bubble chambers ๏Ge, Si, CaWO4, NaI ๏Cryogenic bolometers ๏NaI, CsI ๏Scintillating crystals ๏CS2, CF4, CHF3 Heat ๏Directional detectors

nuclear recoils

Charge Light

4 Dark Matter “Gold Rush”

Target Materials SIMPLE Techniques ๏Xe, Ar, Ne, He PICASSO ๏Noble liquids ๏Ge, Si, He COUPP ๏Low-threshold ๏C3F8, CF3I PICO ๏Bubble chambers ๏Ge, Si, CaWO4, NaI ๏Cryogenic bolometers ๏NaI, CsI ๏Scintillating crystals ๏CS2, CF4, CHF3 Heat ๏Directional detectors

nuclear recoils

Charge Light

4 Dark Matter “Gold Rush”

CoGeNT CDEX nuclear DAMIC recoils SENSEI NEWS-G Charge Light DRIFT MIMAC DMTPC NEWAGE

4 Dark Matter “Gold Rush”

CoGeNT CDEX nuclear DAMIC recoils DAMA SENSEI DM-Ice NEWS-G Charge Light COSINE DRIFT SABRE MIMAC ANAIS DMTPC DEAP NEWAGE XMASS

4 Dark Matter “Gold Rush”

SuperCDMS CRESST EDELWEISS COSINUS CoGeNT CDEX nuclear DAMIC recoils DAMA SENSEI DM-Ice NEWS-G Charge Light COSINE DRIFT SABRE MIMAC ANAIS DMTPC DEAP NEWAGE XMASS

4 Dark Matter “Gold Rush”

SuperCDMS CRESST EDELWEISS COSINUS CoGeNT CDEX nuclear DAMIC recoils DAMA SENSEI DM-Ice NEWS-G Charge Light COSINE DRIFT SABRE MIMAC LUX/LZ ANAIS ArDM DMTPC PandaX DEAP DarkSide NEWAGE XENON XMASS

4 # of researchers in this ﬁeld increased by ~10 times in the last 20 years

5 # of researchers in this ﬁeld increased by ~10 times in the last 20 years

Detector sensitivity improved by 5 orders of magnitude in the last 20 years

5 DAMA/LIBRA so far the only experiment claimed detection of “dark matter”

arXiv:1805.10486 Nucl. Phys. At. Energy 19, 307 (2018)

• phase-1: 1.33 ton x yr (14 annual cycles) • phase-2 • 250 kg radio-pure NaI crystals • 1.13 ton x yr (6 annual cycles) • Sm = 0.0103 ± 0.0008 cpd/kg/keV • conﬁrms phase-1 modulation at 9.5σ C.L.

6 Many experiments have already ruled our DAMA signals in various dark matter models.

Xe Target

XENON100: 5.7σ discrepancy with DAMA LUX: 9.2σ tension with DAMA arXiv:1701.00769 arXiv:1807.07113 PRL 118, 101101 (2017) PRD 98, 062005 (2018)

7 Many experiments have already ruled our DAMA signals in various dark matter models.

Xe Target

XENON100: 5.7σ discrepancy with DAMA LUX: 9.2σ tension with DAMA arXiv:1701.00769 arXiv:1807.07113 PRL 118, 101101 (2017) PRD 98, 062005 (2018)

Ge Target

CDEX modulation analysis: excluded DAMA/LIBRA & CoGeNT modulation signals at >99.99% & 98% C.L. arXiv:1904.12889 7 Experiments using NaI target in confront with DAMA COSINE-100

Nature 564, 83-86 (2018) analysis on absolute rate

• 106 kg NaI crystals • ﬁrst 59.5 days of data • ruled out DAMA signal as from WIMP SI interaction with NaI

8 Experiments using NaI target in confront with DAMA COSINE-100 arXiv:1903.10098 Nature 564, 83-86 (2018) PRL (2019) analysis on absolute rate model independent modulation analysis

• 1.7 years of data for modulation analysis • analysis for 2-6 keV region • 106 kg NaI crystals • R = 2.7 cpd/kg/keV ﬁrst 59.5 days of data • ruled out DAMA signal as from • Sm = 0.0092 ± 0.0067 cpd/keV/kg • so far not yet in conﬂict with DAMA WIMP SI interaction with NaI • • also consistent with NO modulation

8 Experiments using NaI target in confront with DAMA ANAIS-112

• 112.5 kg NaI crystals • 1.5 years of data arXiv:1903.03973 • analysis for 1-6 keV region PRL 123, 031301 (2019) • R = 2.94 ± 0.87 exp(−t/217.8) cpd/kg/keV • Sm = − 0.0044 ± 0.0057 cpd/kg/keV • Consistent with NO modulation ﬁve years running to reach 3 sensitivity of DAMA • σ 9 Comments on the annual modulation search

• All these results rely on signal rates only. • Most analyses do NOT distinguish signal types: electronic (ER) or nuclear recoils (NR) • Most electronic recoil events of these experiments are coming from radioactive background

• no claim is valid without fully understanding the background

• Need experiments with the capability to distinguish ER/NR signals to detect WIMPs.

10 Technology with ER/NR discrimination and better background control

• Bubble chambers • Cryogenic bolometers • Liquid Argon • Liquid Xenon: two-phase time projection chamber (TPC)

• Good 3D position resolution (~mm) • S2/S1 to distinguish ERs & NRs • Large target mass (ton-scale) • Low background • Low threshold • keV with both S1+S2 • <100 eV with S2-only

11 The Era of Liquid Xenon TPCs

100 cm 48 cm 60 cm

LUX PandaX-II Active Target: ~250 kg Active Target: ~580 kg XENON1T Active Target: 2000 kg

12 Reducing the dominant internal electronic recoils background

Kr85 dominant

XENON1T

Kr/Xe~0.66 ppt

13 XENON1T placed the most stringent constraints for WIMPs above 6 GeV/c2

arXiv:1805.12562 PRL 121, 111302 (2018)

• 279 days data in 1.3 ton (1.0 ton yr) • Energy region: 5-41 keVnr (1.4-10.6 keVee) • ER background: 82 evts/ton/yr/keVee • Best SI limit: 4.1 x10-47 cm2 at 30 GeV/c2

14 XENON1T placed the most stringent constraints for WIMPs above 6 GeV/c2

arXiv:1805.12562 arXiv:1902.03234 PRL 121, 111302 (2018) PRL 122, 141301 (2019)

• 279 days data in 1.3 ton (1.0 ton yr) • same data as for the SI search • Energy region: 5-41 keVnr (1.4-10.6 keVee) • Xe129 (29.5%), Xe131 (23.7%) • ER background: 82 evts/ton/yr/keVee • best SD-neutron limit: 6.3x10-42 cm2 at • Best SI limit: 4.1 x10-47 cm2 at 30 GeV/c2 30 GeV/c2

14 Technology with ER/NR discrimination and better background control: Liquid Argon PRD 2018, arXiv:1802.07198

DarkSide-50 DarkSide-50, PSD-only

DarkSide-50, PSD+radial+S2/S1 cut

• Pulse-shape discrimination (PSD) to reject ERs • Target: 46 kg liquid argon (Ar-39 depleted) • Exposure: 532 live-days 15 Technology with ER/NR discrimination and better background control: Liquid Argon

DEAP-3600

arXiv:1902.04048

• Use PSD to reject ERs, dominated by Ar-39 • 3.3-ton liquid argon target • 231 live-days DM search • best SI limit with Ar: 3.9e-45 cm-2 for 100 GeV WIMPs 16 SuperCDMS at Soudan Underground Lab

PRL 120, 061802 (2018) (arXiv: 1708.08869)

• interleaved Z-sensitive Ionization and Photon-mediated (iZIP) detectors • Each detector: 76-mm diameter, 25-mm • 1690 kg days exposure thick Ge (0.6 kg) • single candidate observed, consistent with bkg • Cryogenic operation: O(10) mK • best limits for WIMP-germanium-nucleus interaction > 12 GeV/c2

17 PICO-60: bubble chamber technology

arXiv:1902.04031 SI PRD 100, 022001 (2019)

• bubble chamber: 52 kg C3F8 • excellent electron recoil and alpha rejection

• 1404-kg-day exposure at 2.45 keV threshold SD-proton • previous: 1167-kg-day exposure at 3.3 keV threshold

• larger ﬁducial volume • most stringent SD WIMP-proton limit: 2.5e-41 cm2 at 25 GeV/c2

18 Prospect for heavy WIMP Searches

XENON1T

Discovery(Ruppin, Limits Billard(Xe) due et toal.) CEvNS

✦Major on-going G2 experiments with Noble Liquids, with Liquid Xenon/Argon ✦Start operational as early as 2020 and for ~5 years ✦Next generation (G3) experiment DARWIN is under preparation to start ~2025 and run for 10 years. ✦Expect another 15 years of “neutrino-background free” search for heavy WIMPs 19 Low-mass (GeV-scale) dark matter 2 2 Mχ = 5 GeV/c , σχ-n = 1e-42 cm 4

10th Xe (A=131) Ge (A=73) Ar (A=40) - Kinematic energy transfer is low 3 Ne (A=20) 10 (favors lighter nuclei) - Requires lower energy threshold

102 Rate [Events/Ton/Year] > E - detectors lose ER/NR discrimination - background/noise high and sometime

10 not well understood

1 0 1 2 3 4 5 6 7 8 9 10 Nuclear Recoil Threshold [keV ] nr Low-mass (GeV-scale) dark matter 2 2 Mχ = 5 GeV/c , σχ-n = 1e-42 cm 4

10th Xe (A=131) Ge (A=73) Ar (A=40) - Kinematic energy transfer is low 3 Ne (A=20) 10 (favors lighter nuclei) - Requires lower energy threshold

102 Rate [Events/Ton/Year] > E - detectors lose ER/NR discrimination - background/noise high and sometime

10 not well understood

1 0 1 2 3 4 5 6 7 8 9 10 Nuclear Recoil Threshold [keV ] nr Low-mass (1-10 GeV) dark matter: low-threshold counting

CDEX-10 at China Jinping Lab • 10kg Ge detector in liquid nitrogen • 102.8 kg-days exposure • analysis threshold: 160 eVee • residual bkg rate: ~2.5 evt/keVee/kg/day • improved SI & SD-n limits at 5 GeV/c2

21 Low-mass (1-10 GeV) dark matter: low-threshold counting

CDEX-10, PRL 120, 241301 (2018) CDEX-10 at China Jinping Lab (arXiv:1802.09016) • 10kg Ge detector in liquid nitrogen

• 102.8 kg-days exposure CDMSlite Run2 • analysis threshold: 160 eVee • residual bkg rate: ~2.5 evt/keVee/kg/day • improved SI & SD-n limits at 5 GeV/c2

21 Low-mass (1-10 GeV) dark matter: cryogenic bolometers

CDMSlite, Phys. Rev. D 97, 022002 (2018), arXiv:1707.01632 CDMSlite, Phys. Rev. D 99, 062001 (2019), arXiv:1808.09098

CDMSlite: HV operation SI • no ER/NR discrimination: higher bkg • but lowering the threshold: ~100 eVee • Bkg rate: 1~2 evts/keVee/kg/day • gain sensitivity for low-mass WIMPs

SD-neutron

22 First results from CRESST-III (arXiv:1904.00498)

• 23.6 g CaWO4 • 30.1 eV threshold • total exposure: 3.64 kg days • phonon & scintillation readout • silicon-on-sapphire light absorber • lowest DM mass: 160 MeV/c2

23 Low-mass (1-10 GeV) dark matter: liquid argon

Ionization-only (S2-only) Low-mass DM search in DarkSide-50

Phys.Rev.Lett. 121, 081307 (2018) (arXiv:1802.06994)

• no ER/NR discrimination • Ultra-low threshold: ~100 eVee (a few electrons) • bkg: ~1.5 event/keVee/kg/d at 0.5 keVee • 6786 kg day exposure • Constrains low-mass DM between 2-5 GeV/c2

24 LUX: lower threshold using single scintillation photon signals

arXiv:1907.06272

dark counts xenon scint. light

• Use single photon induced double photoelectron emission signal to reduce threshold • Improving the sensitivity for WIMP mass down to 2.5 GeV/c2

25 LUX: lower threshold using single scintillation photon signals

arXiv:1907.06272

dark counts xenon scint. light

• Use single photon induced double photoelectron emission signal to reduce threshold • Improving the sensitivity for WIMP mass down to 2.5 GeV/c2

25 XENON1T “S2-only” Results • threshold: ~200 eVee • bkg: ~1 event/keVee/tonne/d (>400 eVee) arXiv:1907.11485 • Exposure: 22 tonne-days

26 XENON1T “S2-only” Results • threshold: ~200 eVee • bkg: ~1 event/keVee/tonne/d (>400 eVee) arXiv:1907.11485 • Exposure: 22 tonne-days

2 XENON1T constrains new DM parameter space down to 3~4 GeV/c 26 Status of the Low-mass (GeV-scale) Dark Matter Searches

CRESST-III

DarkSide-50 S2-Only

XENON1T S2-Only

XENON1T

S1+S2

Discovery Limits due to CEvNS (Ruppin, Billard et al.)

Future improvement relies on suppression of known/unknown background with a reasonable large target mass.

27 Probing even Lighter (sub-GeV) Dark Matter

• Need to look at electron signals • Electrons emitted as a secondary process (Migdal, Bremsstrahlung) • Dark matter interacts with electrons • Require: detect down to a few or even single electrons • Reducing the “few-electron” background to improve sensitivity • Existing large experiments (LUX/LZ, DarkSide, XENON, SuperCDMS) can perform simultaneous search while searching for heavy WIMPs • Smaller target experiments: DAMIC, SENSEI, LBECA … with goals to achieve negligible single-to-few electron background Light DM Search with the Migdal effect

• Migdal eﬀect in dark matter detection • Masahiro Ibe et al, arXiv:1707.07258, Dolan et al., 1711.09906 • atomic electrons lags behind recoil nucleus resulting in ionization/excitation of atoms

CDEX, arXiv:1905.00354

LUX, arXiv:1811.11241 PRL 122, 131301 (2019) Light (sub-GeV) dark matter scattering on electrons

Essig et al., arXiv:1703.00910 DarkSide-50, arXiv:1802.06998 PRD 96, 043017 (2017) PRL 121,111303 (2018) SuperCDMS, arXiv:1804.10697 PRL 121,051301 (2018) Xe Ar Si

SENSEI, arXiv:1901.10478 PRL 122, 161801 (2019)

Si Light (sub-GeV) dark matter scattering on electrons

Essig et al., arXiv:1703.00910 DarkSide-50, arXiv:1802.06998 PRD 96, 043017 (2017) PRL 121,111303 (2018) SuperCDMS, arXiv:1804.10697 PRL 121,051301 (2018) Xe Ar Si

SENSEI, arXiv:1901.10478 PRL 122, 161801 (2019) XENON1T S2-only arXiv:1907.11485

Xe Si Dark photon dark matter absorption

An et al., arXiv:1412.8378 SuperCDMS, arXiv:1804.10697 PLB 747 (2015) 331 PRL 121,051301 (2018)

Xe SENSEI, arXiv:1901.10478 PRL 122, 161801 (2019)

XENON1T S2-only arXiv:1907.11485 Si

Xe also see DAMIC, arXiv:1611.03066 PRL 118, 131803 (2017) Summary

32 Summary

• Although there is NO clear direct detection of dark matter, sensitivity has been improved dramatically in recent years for both heavy WIMPs and light dark matter. For future:

32 Summary

• Although there is NO clear direct detection of dark matter, sensitivity has been improved dramatically in recent years for both heavy WIMPs and light dark matter. For future: • Heavy WIMPs: • large experiments are coming online soon: PandaX-4T, XENONnT, LZ, DarkSide-20k, PICO-500, SuperCDMS • direct detection of WIMPs by 2025? • Light dark matter: • simultaneous search in existing large experiments • dedicated search in smaller “zero background” experiments 32