Searching for Dark Matter with PICO-40L Clarke Hardy on behalf of the PICO Collaboration EPS-HEP 2019 Ghent, Belgium July 11, 2019 Overview of PICO

• Merger of PICASSO and COUPP collaborations in 2012 • Direct detection of dark matter using bubble chambers • Located at SNOLAB 2 km underground in Creighton Mine near Sudbury, Canada

2019-07-11 Clarke Hardy 2 Components of a PICO Detector

Pressure vessel containing mineral Quartz jar oil/hydraulic fluid containing target fluid

Cameras and Acoustic sensors LEDs to image capture signal from active volume bubble formation

Temperature Bellows for control pressure control

2019-07-11 Clarke Hardy 3 Why BubbleBubble Chamber Physics Chambers? • Principle of operation • ExpandPut detector in metastable state and wait… under constant temperature to reach metastable superheated state • Energy deposition from nuclear recoils causes a phase transition

Superheated Liquid 4 • SomeDan Baxter, APS April 2016 requirements for dark matter direct detection: 1. Target sensitive to WIMP interactions 2. Low background rates 3. Capability for particle discrimination

2019-07-11 Clarke Hardy 4 Target • Fluorocarbon targets for sensitivity to spin-dependent interactions on 19F due to unpaired proton

Nucleus A Z Isotopic Fraction J 〈Sp〉 〈Sn〉 Xe 131 54 0.2129 3/2 -0.009 -0.227 Xe 129 54 0.264 1/2 0.028 0.359 I 127 53 1.0 5/2 0.309 0.075 Ge 73 32 0.0776 9/2 0.030 0.378 F 19 9 1.0 1/2 0.477 -0.004

• Variety of targets can be used 2 • CF3I for better SI sensitivity (scales with A )

• C3F8 for better SD sensitivity and electron recoil rejection • Variable energy threshold • Determined by setting temperature and pressure • Sensitive down to a few keV • Nucleation efficiency determined using calibration data

2019-07-11 Clarke Hardy 5 Detector Operation

• Event cycle • Set temperature then expand to desired pressure for a particular threshold • Event trigger from bubble visible in cameras or after timeout • Record pressure, temperature, acoustic signal, images, etc. • Recompress and then expand for next cycle • Live fraction of >80%

First start counting livetime

Expansion 32 Compression 30

28

26

0 20 40 60 80

2019-07-11 Clarke Hardy 6 Backgrounds

• Bubble nucleation mechanism differs between background particles • In general, two conditions for bubble nucleation: • Energy deposited must exceed threshold • Energy must be deposited within a critical radius • Seitz “heat-spike” model used to determine threshold for nuclear recoils

2019-07-11 Clarke Hardy 7 Nuclear Recoils

• Radiogenic neutrons from detector components and surrounding rock • Veto multiple bubble events • Coherent elastic neutrino-nucleus scattering (CEνNS) • Dominant background at proposed operating threshold • Cosmogenic muons • Highly suppressed at depth of SNOLAB (~0.29 μ/m2/day)

Operating [1] threshold Target background rate

2019-07-11 Clarke Hardy 8 Electron Recoils from β and γ

• Probability of nucleation from electron recoils in C3F8 given by exponential function of pressure and temperature Bf(P,t) = Ae P • Seitz model • Probability per photon interaction with f (P, t)=QSeitz

• New model [arXiv:1905.12522] Eion • Probability per energy deposited with f (P, t)= rl⇢l @ 4⇡ 4⇡ Q 4⇡r2 T + r3⇢ (h h ) r3 (P P ) Seitz ⇡ c @T 3 c b b l 3 c b l ✓ ◆ surface energy bulk energy reversible work @ 4⇡ E 4⇡r2 T + r3P ion ⇡ c @T 3 c l ✓ ◆ work done by liquid reservoir 2019-07-11 Clarke Hardy 9 Electron Recoils from β and γ -5.4 PICO-0.1 UdeM -5.8 70 6.3585 PICO-0.1 FNAL 5.755 3.3413 -6.1 8.7722 5.1516 • PICO-0.1 MINOS 8.1688 2.7379 New model (right) provides much better fit 4.5482 7.5654 -6.5 PICO-2L Run 2 6.9619 3.9447 Gunter (UofC) 6.3585 -6.8 60 2.1344 to existing data than Seitz model (left)* 8.7722 5.755 3.3413 Drexel 5.1516 1.531 -7.2 8.1688 4.5482 7.5654 2.7379 6.9619 -7.5 3.9447 6.3585 -7.8 50 5.755 • Seitz threshold decreases faster with 5.1516 3.3413 2.1344 -8.2 4.5482 1.531 2.7379 -8.5 3.9447 decreasing pressure than stopping power 40 -8.9 0.92754 3.3413 2.1344 -9.2 (Probability/keV) 10

Pressure (psia) 2.7379 1.531 -9.6 using ionization threshold log 30 -9.9

2.1344 -10.3 0.92754 -10.6 1.531 • Lower pressure maximizes ER rejection for 20 -11.0 -11.3 a given Q 0.92754 -11.7 Seitz 1.531 10 -12.0 14 16 18 20 22 24 26 Temperature (°C) 10-2 PICO-0.1 UdeM PICO-0.1 FNAL PICO-0.1 MINOS -4 PICO-2L Run 2 10 Gunter (UofC) Drexel UofC C3F8 CYRTE 2013 10-6 CYRTE 2014 PICO-60 PICO-2L Run 3

10-8

10-10 Nucleation Probability (per interaction)

10-12 01234567 Seitz Threshold (keV) * Iodine-contaminated data (diamonds) require an additional nucleation mechanism to be fully described (see arXiv:1905.12522) 2019-07-11 Clarke Hardy 10 Alpha Discrimination • Alphas deposit energy over much longer range than nuclear recoils • Emitted acoustic power scales with surface area of initial energy deposition region where target liquid is vaporized [arXiv:1906.04172] • Construct parameter “AP” from integrated acoustic power over a frequency band • Acoustic signals blinded for WIMP- search run to set unbiased cut • Machine learning techniques developed as an alternative to simple AP cut (C. Amole et al., PRL 118, 251301 (2017)) [arXiv:1811.11308] neutrons alphas Observable bubble ~mm Fluorine nucleus (~50 keV) incident neutron ~50 nm ~50 μm

~50 nm

Heavy daughter Helium nucleus (~5 MeV) Long radiating cylinder nucleus (~100 keV) 2019-07-11 Clarke Hardy 11 PICO-60 Detector

• In operation from 2013-2017

• First run with CF3I for 3415 kg-days • Added two cameras and increased frame rate

• Next run with C3F8 for 1167 kg-day at 3.3 keV, then 1404 kg-day at 2.45 keV • Water interface separates active region from stainless steel bellows • Background events observed around chamber walls and at water interface • Particulates from bellows falling into active volume causing background events

2019-07-11 Clarke Hardy 12 PICO-60 Results • World-leading limits on spin-dependent WIMP-proton cross section from first C3F8 run (C. Amole et al., PRL 118, 251301 (2017))

• Improved low-mass limits by including data from second C3F8 run at lower threshold (C. Amole et al., Phys. Rev. D 100, 022001 (2019))

• PICO-60 CF3I • PICO-2L • PICASSO • SIMPLE • XENON1T • PandaX-II

PICO-60 C3F8 2017 PRL

2019-07-11 Clarke Hardy 13 PICO-60 → PICO-40L

Bellows Active region

Water buffer Inner jar “piston”

Active region Bellows

• New “right-side-up” design • Eliminate water buffer and use second quartz jar as a piston • Superheated upper region; cold lower region • Active mass increased slightly from 52 kg to 58.5 kg C3F8 • Larger pressure vessel to reduce backgrounds from stainless steel • Proof-of-concept for future ton-scale detector

2019-07-11 Clarke Hardy 14 PICO-40L Installation Progress • Commissioning expected to be completed this month • Physics data collection to begin in October 2019

2019-07-11 Clarke Hardy 15 Current PICO-40L Run Plan • Calibrations using 252Cf and 60Co prior to first physics run • WIMP search at Seitz threshold of 2.8 keV • Operate at 18 psia instead of 30 psia as in PICO-60 in accordance with new ER nucleation model

PICO-60 PICASSO LUX XENON1T PandaX-II DarkSide-50 DEAP-3600 PICO-40L (projected) PICO-40L (projected) CDMSlite SuperCDMS SDn floor (Xe) SI floor (Xe) CRESST-III SDp floor (C F , no E) 3 8 SI floor (C3F8, no E) DAMIC-100

2019-07-11 Clarke Hardy 16 Next Steps: PICO-500

• ~260 L C3F8 with same “right-side-up” design as PICO-40L • Currently in design/procurement phase • Installation in SNOLAB cube hall to begin in 2020

• Sensitive to supernova neutrinos (T. Kozynets et al., Astroparticle Phys. 105 (2019) 25-30)

2019-07-11 Clarke Hardy 17 Conclusions

• PICO-60 set world-leading limits on SDp cross section • PICO-40L uses a new design to reduce background sensitivity from PICO-60 and provide proof-of-concept for PICO-500 • Commissioning in SNOLAB nearly complete • First physics data expected in fall 2019

2019-07-11 Clarke Hardy 18 C.E. Dahl, M. Jin, J. Zhang S. Priya, Y. Yan R. Filgas, I. Stekl

M. Ardid, M. Bou-Cabo, I. Felis M. Bressler, R. Neilson F. Flores, A. Gonzalez, E. Noriega-Benítez, I. Lawson E. Vázquez-Jáuregui P.S. Cooper, M. Crisler, W.H. Lippincott, A. Sonnenschein O. Harris

S. Ali, M. Das, S. Sahoo B. Broerman, G. Cao, K. Clark, G. Giroux, C. Hardy, H. Herrera, C. Moore, A. Noble, T. Sullivan D. Baxter, J.I. Collar, J. Fuentes

C. Coutu, N.A. Cruz-Venegas, S. Chen, M. Laurin, J.-P. Martin, A.E. Robinson, J. Farine, A. Le Blanc, T. Hillier, S. Fallows, T. Kozynets, K. Allen, E. Behnke, I. Arnquist , T. Grimes, N. Starinski, D. Tiwari, C. Licciardi, O. Scallon, C. Krauss, S. Pal, M.-C. Piro, I. Levine, B. Hackett, A. Hagen, V. Zacek, C. Wen Chao, U. Wichoski W. Woodley N. Walkowski, A. Weesner C.M. Jackson, K. Kadooka, B. Loer 2019-07-11 19 References

[1] "4. Science Potential of a Deep Underground Laboratory." National Research Council. 2003. Neutrinos and Beyond: New Windows on Nature. Washington, DC: The National Academies Press. doi: 10.17226/10583 [2] arXiv:1905.12522 [3] arXiv:1906.04172 [4] C. Amole et al., PRL 118, 251301 (2017) [5] arXiv:1811.11308 [6] C. Amole et al., Phys. Rev. D 100, 022001 (2019) [7] T. Kozynets et al., Astroparticle Phys. 105 (2019) 25-30

2019-07-11 Clarke Hardy 20