Understanding neutrino background implications in LXe-TPC dark matter - PowerPoint PPT Presentation

Understanding neutrino background implications in LXe-TPC dark matter searches using 127 Xe electron captures Dylan J. Temples, Northwestern University Topics in Astroparticle and Underground Physics, Toyama, Japan September 9, 2019

LXe-TPC Dark Matter Searches Top PMT Advantages: Array ❖ Energy deposition reconstruction Sub-mm 3-D position reconstruction ❖ Anode S2 E ❖ Self-shielding / fiducialization Gate extraction Continually purify target ❖ e- ❖ Discrimination between NR (signal) and E drift ER (background) S1 ➢ Charge-to-light ratio Cathode Limitations (for DM searches) Insensitive to DM masses below ~5 GeV ❖ Bottom PMT Array Approaching CEνNS sensitivity ❖ (indistinguishable from dark matter signal) Observables: S1 (prompt scintillation), S2 (electroluminescence) Imperfect background discrimination ❖ E = W( n e + n γ ) = W( S1/g 1 + S2/g 2 ) September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 2

ER/NR Discrimination in LXe-TPCs Heat Dark matter searches use charge-to-light ratio to discriminate between NR (signal) and ER (background) Ionization Xe Xe Xe Xe Recoiling particle (e,N) Excitation ❖ NRs produce overall smaller signals Efficient energy loss to heat (High dE/dx) ➢ ❖ NRs produce proportionally less charge for the same amount of scintillation light Rejection efficiency “ quantified” by fraction of ER events below NR median -- nominally 99.5% arXiv:1703.09144 September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 3

Calibrating the Discrimination 5 keV 6 keV 1 keV 6 keV 3 He Tritium beta decay Valence 𝛿 / 𝜉 -e scatter Inner-shell 𝜉 -e scatter ❖ Typical low-energy ER Atomic deexcitation: Auger e - ❖ Similar energy deposition ❖ calibration source to beta-decay & x-ray Larger dE/dx than similar ❖ energy beta decays September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 4

X enon L -shell E lectron-recoil D iscrimination A nalyzer Based on design of MiX detector [arXiv:1507.0131] Photomultipliers: Top: Hamamatsu R8520 (4x) Bottom: Hamamatsu R11410 (1x) TPC volume: 40 mL (~117 g LXe) Operating conditions: Drift field: 275 V/cm Extraction field: 5650 V/cm (liq.) September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 5

Simulating Inner-shell Recoils with 127 Xe Electron Captures Activation of 127 Xe source at FNAL Excited nuclear state → 𝛿 s ❖ Inner-shell vacancy → x-rays and Auger e - s ❖ Isolate effects of secondary particles, select events where 𝛿 s escape without interacting in the detector. Xenon samples in neutron source Effect of interest strongest for L-shell capture Activity and isotopic content assayed by HPGe *Paper in preparation* ❖ Always results in emission of Auger electron(s) Detector response dominated by most ❖ energetic secondary particle September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 6

Detector Performance 2mm pitch hexagonal grids g1 = 0.255 phe/gamma g2 top = 18.005 phe/electron September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 7

Discrimination Parameter Space: EC vs. 𝛾 September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 8

Discrimination Parameter Space: EC vs. 𝛾 September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 9

Rejection Efficiency: EC vs. 𝛾 0.5% → 3.6% 0.5% → 2.4% Nominal ER rejection efficiency (LZ): 99.5% ❖ Leakage rate: 0.5% Reject tritium (valence) and EC (inner-shell) at same discrimination parameter Leakage rate for inner-shell ERs is 5x larger: 2.4% ❖ September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 10

Implications on Neutrino Backgrounds LUX-ZEPLIN Background Table [arXiv:1703.09144] 13 𝜉 -L scatters 20% of physics background from solar neutrinos Nominal rejection inefficiency: 0.5% ❖ 1.3 “leaked” neutrino ERs Inner-shell rejection inefficiency: ~2.5% 0.3 “leaked” 𝜉 -L ERs ❖ Looking forward: DARWIN [arXiv:1606.07001] ❖ 25% increase in neutrino burden on ❖ Will see 115 𝜉 -L scatters per year unrejected ER background Nailing down the rejection efficiency for these ❖ events is critical September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 11

The XELDA Team Northwestern University U. Chicago C. E. Dahl J. McLaughlin J. Phelan D. Baxter Fermilab / U.C. Santa Barbara Oxford D. Temples (Northwestern University) Summer students: W. H. A. Monte J. Bargemann A. Cottle M. Trask C. KlienStern M. Fassnacht Lippincott September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 12

Conclusions ● XELDA designed to calibrate effects of atomic deexcitation on ER/NR discrimination power in LXe-TPCs ● Neutrino-induced inner shell vacancies produce a different detector response than tritium 𝛾 -decays ● Preliminary results indicate this effect increases the neutrino background burden by ~25% ● PLR may interpret these events as signal if not included as a source of background ● This will be an important source of background to understand in larger xenon-based dark matter searches (DARWIN) ● Ongoing analysis work to refine measurement and beat down systematics, paper in the works Thank you to our funding sources: DE-SC0012161 DOE SCGSR Program September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 13

Backup Slides 14

Charge yields: EC vs. 𝛾 L-shell charge yield: 35.1 +/- 1.77 electron/keV September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 15

Alternative Rejection Efficiency Calculation Overlap of shifted along line of constant energy2D Gaussian with NR band median 0.5% → 6.3% Rejection efficiency: 93.7 % 16 September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan

127 Xe Electron Capture Decays Example atomic de-excitation process resulting from L-shell electron capture on 127 Xe, demonstrating full Auger cascade. September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 17

Inner-shell Vacancies & Auger Cascades Ejected electron ~ β Electronic transition Xe L-shell binding energy ~ 5 ❖ ❖ ❖ of same energy accompanied with keV x-ray or Auger In LZ dark matter search ❖ electron emission region Larger dE/dx in LXe ❖ 18 September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan

Particle Transport in Liquid Xenon September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 19

Impact on DM searches with DARWIN ❖ Nominal ER rejection: 99.98% [30% NR acceptance] WIMP search region: 2-10 keV ee ❖ Solar neutrino ER background rate: 5.2×10 −3 ❖ events/ton/year (after rejection) ➢ 26 events/ton/year (before rejection) Active volume: 40 ton LXe (5-yr exposure) ❖ ➢ 1040 total solar neutrino ERs per year in WIMP search ROI ➢ 115 of these will be 𝜉 -L ERs At 99.98%(99.5%) valence ER rejection efficiency, our ❖ data suggests an L-shell ER efficiency of 99.8%(97.0%) 0.18 (4.63) leaked neutrino valence ERs ➢ ➢ 0.23 (3.45) leaked 𝜉 -L ERs 20 September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan

127 Xe L-Shell Contamination of Tritium Sample K-shell Observed L:K shell ratio: 0.164 L-shell Expected: 0.157 (within 5% due to selection efficiency) Based on K-shell rate in tritium data set, estimate L-shell contamination of tritium in region of interest Contamination fraction: 3.2% September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 21

Effect of fluctuation of gains How much would detector gains have to change in order to move the tritium band to where the L-shell was observed? Good handle on g1 ❖ ➢ K-Shell appears in same S1 region in both data sets Fluctuation of g2: liquid level ❖ changes To move tritium to L-shell value, need a ➢ g2 below 15 No runs have calculated g2 values that ➢ low September 9, 2019 D. Temples , Northwestern U. TAUP, Toyama, Japan 22