Neutrino Studies with the KamLAND detector. Yuri Efremenko University of Tennessee

Transcription

1 Neutrino Studies with the KamLAND detector Yuri Efremenko University of Tennessee Seminar Lab-344 MEPI, Moscow April 20, 2012

2 Outline I. Intro (Neutrino Properties) II. III. KamLAND Detector Results

3 Neutrino History n = p + ve + e The neutrino was first proposed by Pauli to explain the continuous electron energy spectrum in nuclear β decay "I have done a terrible thing. I have postulated a particle that cannot be detected." W.Pauli First neutrinos from nuclear reactors were detected by Reines and Cowan Detected via inverse beta decay same method used today ν e + p = n + e + E thresh =1.8 MeV

4 Neutrino Properties The muon neutrino discovery (and later tau neutrino) combined with the study of the Z boson decay conclude that there are three light neutrinos.

5 Motivation for The KamLAND experiment Was to Solve Solar Neutrino Problem Chlorine Experiment Homestake Mine p-p chain: 4p 4 He + 2e + + 2ν e 1968: Ray Davis pioneers the radiochemical experiment, Chlorine, and observes 1/3 of predicted solar neutrino flux. difficulty of Chlorine experiment and uncertainties in solar model led to speculation that either one or both were wrong on: Ray Davis refines his experiment and John Bahcall refines his theory, no errors found 1989: Kamiokande, real-time water Cherenkov detector, also observes deficit of solar neutrinos s: Sage and Gallex radiochemical experiments confirm deficit in solar neutrino flux By 2000 Borexino, SNO, and KamLAND was in the race to give definite answer to this puzzle

6 Neutrino Oscillations The idea of neutrino oscillations existed before Davis experiment: Pontecorvo (1958), Maki, Nakagawa, and Sakata (1962), and Pontecorvo and Gribov (1969) If m ν is non-zero, then mixing between different neutrino flavors is possible What is produced and detected is weak eigenstate ν j U jl ν = j U jl ν l j is a 3 x 3 unitary matrix (like the CKM matrix for quarks) What propagates is the mass eigenstate l ν U jl = cosθ 12 sinθ sinθ 12 cosθ cosθ 23 sinθ sinθ 23 cosθ 23 cosθ 13 0 e -iδ sinθ e -iδ sinθ 13 0 cosθ 13 e -iα 1 / e iα 2 /

7 Neutrino Mixing Direct Inverted? 0 ev

8 Two Neutrinos Oscillations: Neutrino oscillations are analogous to beats in the sound waves ν 1 wavefunction ν 2 wavefunction Neutrino with given flavor Simplified expression for two flavor oscillations in a vacuum: P(ν l ν l ) = sin 2 2θ sin 2 (1.27 m 2 (ev 2 )L(m)/E ν (MeV))

9 Two types of Oscillations Vacuum m L P( ν e ν µ, L) = sin 2θ sin E Matter enhanced (MSW effect) Out of three leptons, only electrons are present in our regular matter. Because of that ν e has both NC(Z 0 ) and CC(W ± ) interactions. However ν µ, ν τ have only NC. Mixing angle and oscillation distance should be modified: Sin 2 2θ m = sin 2 2θ /{1-2(l V /l 0 )cos2θ + (l V /l 0 ) 2 } l m =l V / {1-2(l V /l 0 )cos2θ + (l V /l 0 ) 2 } l 0 ~ρ/y e, l V ~E/ m 2 Resonance conditions occurs when Y e m 2 /ρe = cos2θ Sin 2 2θ m ~ 1 l m ~ 0

10 Status Before KamLAND SMA LMA Compilation of all solar data: Cl+H 2 O+Ga LOW Vacuum tan 2 θ It was a gamble because KamLAND has sensitivity to LMA only

11 Outline I. Intro (Neutrino Properties) II. III. KamLAND Detector Results

12 KamLAND -Kamioka Liquidscintillator Anti-Neutrino Detector - Detecting reactor ν e 1km beneath Mt. Ikeyama Inside the Kamioka Mine Surrounded by 55 Japanese Reactor Units

13 The KamLAND Detector Balloon & support ropes calibration device & operator Target LS Volume (1 kton, 13m diameter) Glove box Buffer Oil Zone Photomultiplier Tubes (34% coverage of ID) Stainless Steel Inner Vessel (18m diameter) Chimney (access point) Outer Detector (3.2 kton Water Cherenkov)

14 The Target Volume Liquid Scintillator: proton rich: > protons 20% Pseudocume + 80% Mineral Oil g/l PPO Optimal light yield while maintaining long attenuation length (~15 m). Welding the Balloon Balloon: Separates target LS volume from buffer oil 135 µm Nylon/EVOH (ethylene vinyl alcohol copolymer) Supported by braided kevlar ropes and buffer oil 14

15 KamLAND Photomultipliers PMT and acrylic panel installation (2000) tubes tubes (since Feb. 03) ~ hits for 1 MeV energy deposit Transit time spread < 3 ns on 17 tubes acrylic panels protect against radioactive backgrounds Cables and Custom Build Electronics 15

16 Cables and Electronics 12 ch. per board, 400 MHz sampling Generates trigger Connected to GPS (Custom build at LBNL) In 2010 second set of electronics was commissioned. Both working in parallel now.

17 Detector top with a Glove box

18 Event Reconstruction σ E E = 6.5% E(MeV) How much energy deposited and where? Energy Reconstruction: Energy Number of Hit PMT s Correction for Vertex Position Correction for Quenching and Cherenkov Radiation KamLAND Event Display Vertex Reconstruction Determined by Very Precise Timing of Hits (~few ns resolution) Inherent Detector Resolution ~12cm/ E(MeV) 18

19 Muon Tracking Source of cosmogenic backgrounds Rate of Muons hitting KamLAND OD: ~1 Hz KamLAND ID: 0.3 Hz Timing of inner detector hits Good agreement with simulation of muons passing through detailed mountain topography 19

20 Antineutrino candidate (colour is time) Prompt Signal E = 3.20 MeV t = 111 µs R = 34 cm Delayed Signal E = 2.22 MeV ν e + p e + + n n + p d +γ(2.2 MeV)

21 Calibration with Off-Axis System Off-axis calibration to improve energy and vertex estimation Reduce fiducial volume uncertainty (dominant uncertainty on ν- e rate) Position bias from relative distance between sources Simple in concept, difficult in practice! swappable source embedded 60 Co Monitor Risks to Detector: balloon safety (135 µm thick), cleanliness, LS compatibility, no pieces left behind 21

22 Example Off-Axis Calibration Results Poles of fixed length swept through zenith angle +z Deviation of Energy [%] energy bias w/ 60 Co (E=2.5 MeV) 1.5% 60 Co/ 68 Ge composite source deployments Deployments with different sources check for energy dependent systematic effects: 60 Co, 68 Ge, 241 Am 9 Be, 210 Po 13 C, 203 Hg Deviation of Position [cm] vertex bias w/ 60 Co < 3cm biases exhibit zenith dependence 22

23 Accidental Backgrounds Steel in the chimney region Contamination concentrated on balloon and in support ropes High rate of single gammas from natural background radiation (U, Th, K, ) can accidentally mimic promptdelayed signal candidate events background events removed by selection cuts Varies greatly with energy and location within the detector. Reduced by time ( T[0.5,1000]µs) and spatial ( R<2m, R<6m) cuts. z y x 23

24 Correlated Backgrounds: Spallation Products Cosmogenic Muons interact with material producing: fast neutrons - removed with 2ms veto after any detected muon delayed neutron β emitters ( 9 Li) - removed with 2 second veto around µ-track He 8 thought to be a negligible contribution Cutting events correlated with muons removes almost all cosmogenic bg <10% deadtime introduced by all muon cuts 24

25 Correlated Background: 13 C(α,n) 16 O Originating from Rn contamination, discovered after first publication low energy 4.4 MeV ~6 MeV S.Harissopulos et al., Phys Rev. C72, Background Prompt E (MeV) 25

26 Outline I. Intro (Neutrino Properties) II. III. KamLAND Detector Some Results

27 Antineutrino Production At Reactors 4 main fuel components Time history of reactor reload Calculated Neutrino Spectrum N(ν) = f(e,t)

28 Antineutrino Spectra Primary Fissioning Isotopes (representative ratio) 235 U: 238 U: 239 Pu: 241 Pu = 0.61:0.13:0.20: ν e Flux: deduced from measurements of cumulative daughter β-decay specta. Phys Lett B 160, 325 (1985) Phys Lett B 218, 365 (1989) X-section: precise calculation, O(1/M n ) Phys Rev C 24, 1543 (1981) (baseline 1) (baseline 2) Detectable Spectrum (unoscillated) (baseline 3) Phys Rev D 34, 2621 (1986) Predicted spectrum shown to have good agreement by earlier reactor experiments 28

29 Expected Signal from Reactors m L P( ν e ν µ, L) = sin 2θ sin E

30 Antineutrino candidate (colour is time) Prompt Signal E = 3.20 MeV t = 111 µs R = 34 cm Delayed Signal E = 2.22 MeV ν e + p e + + n n + p d +γ(2.2 MeV)

31 Data Analysis from MC simulation L ratio (E prompt ) = f ν -/(f ν - + f accidental ) Example for E prompt [2.2,2.3] 2. Accidental Background - ν e signal from off-timing accidental data At fixed E prompt : L ratio depends on E delayed, R prompt, R delayed, R, T

32 Results Data from March 2002 till November 2009

33 Oscillation Parameters (observed - bg) / expected L 0 = 180km flux-weighted average reactor distance m L P( ν e ν µ, L) = sin 2θ sin E Combined Best-Fit Parameters: m 2 21 = ev tan 122 θ =

34 Inverse beta decay Neutrino uniqueness ν e + p n + e ν Electron Scattering ν e ' + e v + e Coherent Scattering ' e + A v + Nuclear Interactions ν + e + N e + ' Extremely low interaction cross section A X ' As a result neutrinos are carriers of information from sources inaccessible by other means σ σ = σ = σ = = 9.3 ( E /1MeV ) 10 cm ( E /1MeV ) 10 cm N ( E /1MeV ) 10 cm 39 2 ) ( E /1MeV 10 cm

35 Geo Neutrinos Dear Fred, Just accured to me that your background neutrinos my just be comming from high energy β- decaying members of U and Th families in the crust of the Earth George Gamov to Fred Reines Antineutrinos can do for study of the earth what neutrinos can do for the sun. Krauss, Glashow, Schramm. Nature 310 (1984)

36 Anti-neutrinos from the Earth 238 U 206 Pb He + 6e - + 6ν e + 52 [MeV] 235 U 207 Pb He + 4e - + 4ν e + 47 [MeV] 232 Th 208 Pb He + 4e - + 4ν e + 43 [MeV] 40 K 40 Ca + e - + ν e [MeV] (89.3%) 40 K + e - 40 Ar + ν e [MeV] (10.7%) Isotope Abundance, relative T ½, By. Heat * production, TW 232 Th U U K * Based on the Bulk silicate Earth model

37 Earth s Total Surface Heat Flow Conductive heat flow measured from bore-hole temperature gradient and conductivity 40,000 data points mw m -2 Surface heat flow 46±3 TW (1) 47±2 TW (2) (1) Jaupart et al (2008) Treatise of Geophys. (2) Davies and Davies (2010) Solid Earth

38 One Slide History of the Earth Heterogeneous mixtures of components with different formation temperatures and conditions Planet mix of metal, silicate, volatiles

39 Plate Tectonics, Convection and Cooling of the Mantle Differentiation of initially homogeneous Earth

40 Chromatographic separation Mantle melting & crust formation BSE-Bulk Silicate Earth Differentiation ~13 ng/g U in the Earth Metallic sphere (core) <<<1 ng/g U Silicate sphere 20 ng/g U *Javoy et al (2010) predicts 12 ng/g *Turcotte & Schubert (2002) 31 ng/g Continental Crust 1300 ng/g U Mantle ~12 ng/g U

41 One of the BSE models Total Earth s surface heat flow 46 ± 3 (47 ± 2) Crust R* (7 TW) Mantle cooling (18 TW) Mantle R* (13 TW) Core (9 TW) *R radiogenic heat after Jaupart et al 2008 Treatise of Geophysics (0.4 TW) Tidal dissipation Chemical differentiation

42 Heat Production in the Earth/Mantle There is factor of ~3 differences in BSE models Mantle BSE Turcotte & Schubert (2002) Anderson (2007) Palme & O Neill (2003) Allegre et al (1995), McD & Sun ( 95) Lyubetskaya & Korenaga (2007) Javoy et al. (2010) TW in Mantle (minus crust contribution and only Th & U flux) U content (ng/g) (Bulk Silicate Earth)

43 Radiogenic Heat Production History

44 What Geo-neutrinos can tell us: Measure total radiogenic heat production Distinguish heat generation in mantle vs. crust Help evaluate different geo models Provide input to better understand geological history of the Earth

45 Geo-neutrino Flux Based on the BSE model geo-neutrino flux can be predicted at every point on the Earth surface Effect of neutrino oscillations Terrestrial Neutrino Unit (TNU) N geo-neutrino interactions per year at free protons Geo-neutrino Spectra

46 Geo-neutrino and Nuclear Power Plants

47 Effects of Local Geology Average Uranium 2.32 ppm <500km 50% Average Thorium 8.3 ppm Effect of local geology < 10% uncertainty of total flux 50% of the total flux originates from a distance > 500 km!!!

48 Anti-neutrinos at the KamLAND Geoneutrinos KamLAND Reactor Background with oscillation KamLAND was designed to measure reactor anti-neutrinos and they are the most significant background for geo-neutrinos.

49 Event rate time variation: 0.9 MeV MeV Rate, events/day before-purification After correlation Data Reactor + BG Expected reactor Reactor + BG + geo Best fit Reactor + BG Reactor+BG+geo We see constant excess above the estimated reactor neutrino + non-neutrino background at 0.9 < E < 2.6 MeV region

50 First KamLAND result 2005 (Nature) Total: 152 events (α,n) 42 ev Reactor 80.4 ev Random 2.4 ev Geo from U Geo from Th Background 127 ±13 Excess - 25 events BSE prediction 18.7 events

51 Latest result (2011, Geo Science) 4126 ton-yr data-set (2135 days) Rate analysis (0.9 < E < 2.6 MeV) 841 candidates 9 Li 2.0 ± 0.1 Accidental 77.4 ± 0.1 Fast neutron < 2.8 (α, n) ± 18.2 Reactor ν ± 26.5 BG total ± 32.3 excess events

52 Fixing U/Th ratio U/Th mass ratio fixed to 3.9 best-fit (U, Th) (65, 33) model w/o neutrino osc. U/Th ratio fixed Nature 436, 28 (2005) PRL 100, (2008) Nature geoscience 4, 647 (2011) Earth model prediction EPSL 258, 147 (2007) Ngeo = 106 Fgeo = events 10 6 /cm 2 /sec This is a conformation that radiogenic is responsible to up to ~50% of the total heat emitted by the Earth

53 KamLAND and Borexino Constrainting U & Th in the Earth MODELS Cosmochemical: uses meteorites Javoy et al (2010) EPSL Geochemical: uses terrestrial rocks McD & Sun, Palme & O Neil, Allegre et al Geophysical: parameterized convection Schubert et al; Davies; Korenaga, Jaupart et al

54 Post Fukushima Nuclear Energy in Japan On February of 2011 about 70% on nuclear energy capacity were in operation Every 13 month every nuclear unit should be stopped for regular maintenance During the last year none of the units get permission to resume operation after planned shutdown. At the end of January 2012 only 3 units were in operation at: Shimane, Kashiwazaki, Tomari Feb 20 th Shimane off March 26 th Kashiwazaki off Starting from the beginning of this May there will be no nuclear power plant in Japan in operation Nuclear power plants Tomari 783 km Shika 88km Wakasa 146~192km Kashiwazaki 159 km Shimane 401 km 180km Hamaoka 200km

55 Changes in Reactor Anti-neutrino Flux Beginning of 2011 Now

56 KamLAND, Japan 1 kt Worldwide Efforts Borexino, Italy 0.3 kt SNO+, Canada 1 kt Hanohano, Hawaii 10kt LENA, EU 50 kt Location Reactor rate <3.3 MeV TNU Geo rate TNU* Detector N geo per year Status KAMIOKA 5.2 (now) 34.5 KamLAND 20.7 Running FREJUS SUDBURY SNO+ ~40 About to start GRAN SASSO Borexino 4.2 Running PYHASALMI LENA 1500 Proposal BAKSAN DUSEL HAWAII Hanohano 75 Proposal * Fiorentini at all, Phys Rep. 2007

57 Geo Reactor? M.Herndon and D.Hollenbach Most of U and Th are in the core! This hypotheses is not on the main stream of geology. Challenges for detection Similar spectra as for man made reactors Background from nuclear power plants No directionality in ν e + p n + e + reaction Based on the fluctuations of energy production by nuclear power plants and background subtraction upper limits are: P geo-reacto < 3TW (Borexino), P geo-reacto < 5.2 TW (KamLAND) If nuclear power plants in Japan will stay off for entire 2012, expected sensitivity for Geo-reactor at one sigma for KamLAND is ~2 TW

58 Perspectives for Potassium E tr for IBD Isotope Abundance % Threshold, MeV Product Product life time Q kev 3 He H y N C 5730 y S P d Cl S Stable 63 Cu Ni 100 y Cd Ag 24min 2965

59 Why Potassium is Interesting Th & U K from McDonough & Sun, Chem. Geol., 120, , 1995