The Hunt for Sterile Neutrinos. H. Ray, University of Florida

Transcription

1 The Hunt for Sterile Neutrinos 1

2 Particle Physics is Over!..finding the Higgs arguably the most important discovery in more than a generation has left physicists without a clear roadmap of where to go next. Wired, Oct 2013 Headline: Higgs boson discovery may signal the world s last physics experiment as scientists struggle to come up with next big question National Post, Dec

3 Over? Not Quite! 3

4 Higgs Doesn t Solve Everything! Need a dark matter candidate What about dark radiation? (2σ) Excess relativistic energy density at decoupling SM has no way for νs to acquire mass Anomalous results from neutrino sector Short baseline (SBL) oscillation expt. excesses ( σ) Reactor neutrino flux deficit (3σ) Radioactive source (Ga) deficit ( σ) IceCube flux deficit due to observed GRBs (3.7x lower) 4

5 Not One-Stop Shopping! Sterile mass Allow ν to acquire mass Dark Rad, SBL, Reactor, Ga Dark matter candidate Find via a direct search GeV YES NO NO NO GeV YES NO NO YES kev - GeV YES NO YES YES ev YES YES NO YES

6 Not One-Stop Shopping! Sterile mass Allow ν to acquire mass Dark Rad, SBL, Reactor, Ga Dark matter candidate Find via a direct search GeV YES NO NO NO GeV YES NO NO YES kev - GeV YES NO YES YES ev YES YES NO YES

7 Not All Compatible! Short baseline oscillation expt. excesses: ~1 ev Reactor + Ga deficits: ~1 ev Cosmology dark radiation candidate: ~1eV ex: SBL compatible with CMB in 3+1, 3+2 Incompatible with cosmological mass constraints from CMB, Large Scale Structure (sum of all ν masses < ev) Can be compatible with LSS if include initial lepton asymmetry Not just the mass, but proposed mixing with Standard Model νs, decoupling temperature, etc, also matters!,

8 What to Do? Propose experiments to further explore each anomaly Expts. to perform more precise measurements and searches for ev scale sterile neutrinos in reactors, radioactive decay, and SBL experiments My focus: SBL results and prospects for the future 8

9 SBL Anomalies: LSND 800 MeV proton beam + H 2 0 target, copper beam stop 167 ton tank, liquid scintillator, 25% PMT coverage E ν ~20-50 MeV L ~25-35 meters anti-ν e + p e + + n n + p d + γ (2.2 MeV) H. Ray 9

10 SBL Anomalies: LSND Fit to oscillation hypothesis Δm 2 = 1.2 ev 2 sin 2 2θ = Backgrounds P osc = sin 2 2θ sin Δm 2 L LSND observed excess of ν e in a ν µ beam E Excess: 87.9 ± 22.4 ± 6.0 (3.8σ) Phys. Rev. Lett. 77: (1996) Phys. Rev. C 58: (1998) Phys. Rev. D 64, (2001) 10

11 MiniBooNE vs LSND LSND (anti) Neutrino beam from accelerator (DAR, average E ν 35 MeV) MiniBooNE Neutrino beam from accelerator (DIF, average E ν 800 MeV) Detector placed at 500 m from neutrino beam creation point, preserve LSND L/E ν µ too low E to make µ or π New backgrounds: ν µ CCQE and NC π 0 mis-id for oscillation search Proton beam too low E to make K New backgrounds: intrinsic ν e from K decay (0.5% of p make K) 11

12 MiniBooNE Neutrino Beam π- π+ µ+ oscillations? FNAL booster (8 GeV protons) target and horn (174 ka) K 0 K + decay region (50 m) dirt (~500 m) detector Neutrinos from pions decaying in flight ~same L/E as LSND (~1) 12

13 MiniBooNE Detector 12.2 meter diameter sphere Pure mineral oil 2 regions Inner light-tight region, 1280 PMTs (10% coverage) Optically isolated outer veto-region, 240 PMTs Nucl. Instr. Meth. A599 (2009)

14 SBL Anomalies: MiniBooNE 2.8σ antineutrino mode 3.4σ neutrino mode 3.8σ combined excess All in MeV range 7σ stat so not a statistical fluctuation! x POT Antineutrino excess consistent with LSND Neutrino excess not so much 6.5 x POT All backgrounds fully constrained Need some new anomalous background process to explain low energy excess, if not invoking a sterile neutrino explanation

15 SBL Anomalies: Summary Δm 2 = 3.14 ev 2 sin 2 2θ = Δm 2 = ev 2 sin 2 2θ =

16 Resolving the SBL Mystery Need definitive experiments no more carving out small portions of the allowed sterile neutrino phase space No longer good enough to see an excess or deficit need to see those wiggles! Need to see wiggles as a function of energy! Need them to be cost-effective Preferably short-term, to use as input to longer-term projects 16

17 Proposed Experiments

18 Proposed Experiments

19 Proposed Experiments

20 Proposed Experiments

21 Proposed Experiments Same target nucleon as LSND, MB Shorter-term (3 months to ~2 yrs for construction, after funding)

22 MiniBooNE+ Proposal: arxiv

23 MiniBooNE Low-E Excess Largest backgrounds in region of excess are muon neutrino Neutral Current mis-id neutral pions and gammas that look identical to e + /e - in our detector 23

24 MiniBooNE Low-E Backgrounds Both NC backgrounds are constrained by in-situ measurements NC π 0 directly measured NC γ (radiative Δ decay) constrained to NC π 0 Also, recent theoretical calculations agree with MB

25 MiniBooNE Low-E Backgrounds Both NC backgrounds are constrained by in-situ measurements NC π 0 directly measured NC γ (radiative Δ decay) constrained to NC π 0 Also, recent theoretical calculations agree with MB

26 MiniBooNE+ Add scintillator to MiniBooNE to enable reconstruction of 2.2 MeV neutron-capture photons Reconstructed vs True Eν, Signal Re-run neutrino mode oscillation search Neutron-capture enables separation of CC oscillation events from NC backgrounds CC: ν e + n e - + p 1-10% of all interactions will produce a neutron Reconstructed vs True Eν, Backgrounds NC: ν µ + 12 C Δ γ or π 0 + p or n equal chances of getting n or p n + p d MeV γ 26

27 MiniBooNE+ Need to know the (1-10%) vs 50% very well for this analysis! These numbers come from previous data/models Will measure in MB+ Can measure n fraction in ν µ CC events (not the oscillation channel) Can measure n fraction in pristine NC π 0 events 27

28 MiniBooNE+ Same as previous analysis, same excess Require n-capture events Red: if excess is truly due to CC ν e events, excess disappears Blue: if excess is truly due to a NC process (ie not oscillations), excess remains Yields 3.5σ NC/CC separation for this test, for combined 5σ MB excess 28

29 OscSNS Proposal: arxiv:

30 OscSNS Spallation Neutron Source at Oak Ridge ~1 GeV protons+hg target (1.4 MW) Free source of neutrinos π - absorbed by target π + DAR Mono-Energetic! ν µ = 29.8 MeV E range up to 52.8 MeV 30

31 OscSNS Detector Homogeneous liquid scintillator detector Mineral oil + b-pbd 8 m diameter x 20.5 m length ~800 tons, 25% PMT coverage 205'tubes'per'end'cap' Flexible arm deployment system for 1 50 MeV calibration sources 16 N, 8 Li, 252 Cf Veto'barrier' Hamamatsu'R5912'assumed' 60'rows'(6 )'of'54'each'pmts'located' 14 '(.356m)'center'to'center.''Tube' center'located'3.4m'radially'from' detector'tank'center'line'(3240'tubes)' Proton beam OscSNS Detector Hall 60 m in the backward direction, ~150 degrees from incident proton beam 31

32 OscSNS ν µ -> ν e Experiment vs LSND More Detector Mass (x5) (Assuming Δm 2 < 1 ev 2 ) Higher Intensity Neutrino Source (x2) Lower Duty Factor (x1000) (less cosmic background) Separation of ν µ & ν e / anti-ν µ fluxes with timing Negligible DIF Background (backward direction) Lower Neutrino Background (~x2) (60m vs 30m) For LSND parameters, expect ~ ν e oscillation events & ~50 background events per year! 32

33 Oscillation Goals ν µ -> ν e appearance ν e -> ν s disappearance ν µ -> ν s disappearance ν all -> ν s disappearance 33

34 Appearance Sensitivity 50% detector efficiency, ~85% E e cut efficiency, oscillation probability of 0.26% l ν µ -> ν e appearance sensitivity for 1 & 3 years of running: ν e p à e + n; n p à d γ (2.2 MeV) CONTINUOUS! ) 2 (ev m e Confidence Level Curves for µ Oscillations (1 year) Already at 5σ! ) 2 (ev m e Confidence Level Curves for µ Oscillations (3 years) 10 1 LSND & KARMEN Allowed 10 1 LSND & KARMEN Allowed CL 3 CL 5 CL CL 3 CL 5 CL sin sin 2 34

35 Appearance Sensitivity 50% detector efficiency, ~85% E e cut efficiency, oscillation probability of 0.26% Statistical errors, 20% bgd CONTINUOUS! Assuming 5y of data & sin 2 2θ = 0.005, Δm 2 =1 ev 2 Assuming 5y of data & sin 2 2θ = 0.005, Δm 2 =4 ev 2 L/E (m/mev) L/E (m/mev) 35

36 Disappearance Sensitivity 50% detector efficiency Statistical errors, 1% bgd CONTINUOUS! ν e C à e - N gs, N gs à C gs e + ν e Assuming 5y of data & sin 2 2θ = 0.15, Δm 2 = 1 ev 2 Assuming 5y of data & sin 2 2θ = 0.15, Δm 2 = 4 ev 2 L/E (m/mev) L/E (m/mev) 36

37 Disappearance Sensitivity 50% detector efficiency Statistical errors, 30% bgd ν µ C à ν µ C*(15.11) CONTINUOUS! Assuming 5y of data & sin 2 2θ = 0.15, Δm 2 = 1 ev 2 Assuming 5y of data & sin 2 2θ = 0.15, Δm 2 = 4 ev 2 L/E (m/mev) L/E (m/mev) 37

38 Summary and Conclusions Many outstanding mysteries in the neutrino sector 38

39 Summary and Conclusions Many outstanding mysteries in the neutrino sector Even mysteries that indicate a similar solution aren t compatible 39

40 Summary and Conclusions Many outstanding mysteries in the neutrino sector Even mysteries that indicate a similar solution aren t compatible Need new era of 5σ, definitive, and cost-effective experiments to explore & resolve 40

41 Summary and Conclusions Many outstanding mysteries in the neutrino sector Even mysteries that indicate a similar solution aren t compatible Need new era of 5σ, definitive, and cost-effective experiments to explore & resolve MiniBooNE+, OscSNS are two such experiments to resolve the SBL oscillation mystery! 41

42 Backup 42

43 Sterile Information 43

44 SBL Oscillation Measurements Use beams of neutrino type X (e, µ, τ), search for neutrino oscillations, where ν X à ν Y by the time it reaches the detector Use Charged-Current interactions in an appearance analysis looking for ν Y to appear in a beam of ν X ν l + n à l - + p anti-ν l + p à l + + n 44

45 Enter Sterile Neutrinos! SM has no way for νs to acquire mass Need a dark matter candidate Neutrino Minimal Standard Model Introduce 3 Majorana singlet fermions 1 50 kev: dark matter particle 2 others are responsible for giving mass to neutrinos via see-saw mechanism. ~GeV scale

46 Enter Sterile Neutrinos! Anomalous results from neutrino sector Short baseline oscillation expt. excesses: σ IceCube flux deficit due to observed GRBs: 3.7x below prediction Need a dark matter candidate Introduce 3 sterile neutrinos Lightest one degenerate with one of the active νs (IceCube) ~ev mass (SBL) ~kev mass (dark matter)

47 Enter Sterile Neutrinos! Anomalous results from neutrino sector Reactor neutrino flux deficit: 3 σ Radioactive source (Ga) deficit: σ Fit these two anomalies in a 3+1 sterile framework Δm 2 new 1 ev2, sin 2 2θ new ~ 0.17 preferred

48 Enter Sterile Neutrinos! What about dark radiation? Most recent cosmological data, including Planck show indication in favor of an excess of relativistic energy density at the time of decoupling aka dark radiation Favored at 2 σ Sterile ν mass ~ev scale

49 Best Fit Points, ~1 ev Sterile ν MiniBooNE: antinu = delta m2 = ev2, sin22theta = 0.88 MiniBooNE: nu = delta m2 = 3.14, sin22theta = Ga only: delta m2 = 2.24 ev2, sin22theta = 0.50 Reactor only, rate: delta m2 = 0.5 ev2, sin22theta = 0.15 Reactor only, rate+shape: delta m2 = 2.4 ev2, sin22theta = 0.14 Ga+Reactor: delta m2 > 1.5 ev2, sin22theta = 0.17 CMB data: Have allowed regions plot of m_sterile as f(n_sterile). Appears between 1&2 sterile nus with ms of around 0.2 ev is best region KamLAND (spectral shape, independent of reactor flux normalization), solar, Daya Bay, RENO (dual baseline, independent of flux normalization), reactor anomaly: sin2theta14 = to

50 Global 3+N Fits arxiv:

51 SBL Anomalies: MiniBooNE P ΝΜ Νe or P ΝΜ Νe MiniBooNE Ν e LSND Ν e MiniBooNE Ν e Ν e or Ν e 3 2 best fit Ν e or Ν e 2Ν best fit L E Ν meters MeV

52 MiniBooNE+ Information 52

53 53

54 54

55 55

56 56

57 57

58 58

59 59

60 60

61 MiniBooNE+ Future PAC said go ahead with proposal and show them various things strength of collab, PID with KB+ FNAL wants the proposal The Indiana University (PI) group asked for and received 100K in funding from the NSF for equipment 61

62 OscSNS Information 62

63 63

64 SNS vs Other Beamlines LAMPF Booster SNS Energy 800 MeV 8000 MeV 1300 MeV Current 1 ma 1.1 ma Power 0.8 MW 42 kw 1.4 MW Pulse Length 600 µs 1600 ns 695 ns Rep Rate 120 Hz 6.5 Hz 60 Hz Duty Factor 7.2% 43% % Max delivery rate, from T. Kobilarcik 64

65 OscSNS Location Assumed' detector' location' Centered 60m upstream of the beam dump/target - removes DIF bgd 65

66 Beam Parameters Full beam power: 1.4 MW, 1.3 GeV protons protons per pulse 700 ns wide pulses Frequency of 60 Hz At 60 meters from the target, expect a total neutrino flux of ~1.6e14 ν/year/cm 2 66

67 Advantages Over Other ν Osc. Expts Well understood ν fluxes: ν µ, ν e, ν µ Well understood ν cross sections Low duty factor (4x10-5 ) Nuclear effects are not a problem Very low backgrounds (~0.1%) Beam comes for free from the SNS, which runs >1/2 the year Search for both appearance & disappearance oscillations Possibility of observing oscillations in the detector for Δm 2 > 0.5 ev 2! 67

68 68

69 69

70 Electronics area directly above tank ~6 m of earth fill placed on top of electronics room for overburden, will reach a height of 1.5 m above existing grade 70

71 Downstream Exit 3 m long x 12 m wide x 11.5 m high Upstream Exit 71

72 Expansion tank necessary for mineral oil overflow due to ambient temperature changes Located above top of tank 6 ft diameter x 16 ft long Vault: 3 m wide x 3 m high x 6 m long Electronics area 12 m wide x 4 m high x 25 m long 2 thick Tank enclosure 12 m wide x 12 m high x 25 m long 3 6 thick Winch access for tank vault 72

73 Particle Identification (with g/l of b-pbd) Electrons Protons Particle Identification depends on Cherenkov cone fit, position fit, and fraction of late light 73

74 2.2 MeV γ Identification 74

75 OscSNS Physics Plan Expect ~4e19 ν/year/species for a 8 m diameter cylinder, front face located 50 m from source Measure scattering cross sections Several channels to search for oscillations, sterile neutrinos 2 NC disappearance (ν µ, ν x 12 C to 12 C*) 1 CC disappearance (ν e 12 C -> e - 12 N gs ) 2 CC appearance (ν e, anti-ν e ) MiniBooNE low E excess (?) -> LSND check 75

76 Cross Sections: I NC: ν µ 12 C -> ν µ 12 C* (15.11) KARMEN KARMEN σ NC = (3.2 ± 0.5 ± 0.4) x cm 2 (B. Armbruster et al., Phys. Lett. B423 (1998) 15) KARMEN, 86 events, 20% total error (half due to stats) 76

77 Cross Sections: II CC: ν e C -> e - N gs LSND σ CC = (8.9 ± 0.3 ± 0.9) x cm 2 (L. B. Auerbach et al., Phys. Rev. C64, (2001)) LSND, 191 events, 17% total error 77

78 Flavor Oscillations: I CC Appearance: mono-energetic ν µ -> ν e ν e C -> e - N gs N gs -> C e + (~8 MeV) ν e Intrinsic ν e from μ decay mono-energetic ν e from ν µ 78

79 Event Rates per Year All estimates include 50% detector efficiency KARMEN, 86 events, 20% total error (half due to stats) LSND, 191 events, 17% total error 79

80 Event Rates per Year All estimates include 50% detector efficiency and are for 100% accelerator up-time S/B Efficiencies Applied ~3 30% bgd from Karmen measurement, under MeV peak ~3 (same as above) ~10 1% bgd from LSND ~3 20 MeV e selection cut eff applied to S, B, 0.26% osc ~ MeV e plus e+ cut, beam timing cut, 0.26% osc 80