KamLAND Neutrino Oscillation Results and Solar Future Patrick - - PowerPoint PPT Presentation

KamLAND Neutrino Oscillation Results and Solar Future

Patrick Decowski (UC Berkeley) for the KamLAND Collaboration

Neutrino 2008, Christchurch, NZ

1

Patrick Decowski / UC Berkeley

KamLAND Collaboration

2

Patrick Decowski / UC Berkeley

Reactors for Oscillation Studies

P(νe → νe) = 1 − sin2 2θ sin2 1.27∆m2L E

L

νe νe

Neutrino oscillation changes the overall normalization and shape of the spectrum

[MeV]

e

• E

2 4 6 8 10 Rate [au] 2 4 6 8 10

Simulation

No oscillation Δm2=7x10-5 eV2 Δm2=2x10-5 eV2 (sin22θ=0.8)

3

Patrick Decowski / UC Berkeley

Anti-Neutrino Detection Method

Inverse beta decay

νe + p → e+ + n n + p → d + γ

e n e+ 2.2MeV Liquid Scintillator

γ γ γ

Scintillator is both target and detector

• Distinct two step process:
• prompt event: positron
• delayed event: neutron capture after ~207μs
• 2.2 MeV gamma

Delayed coincidence: good background rejection

Eνe Eprompt + 0.8MeV

207 μs

4

Patrick Decowski / UC Berkeley

• In practice, only 1.5 neutrinos/fission detectable
• Calculated spectrum has been verified to 2%

accuracy in past reactor experiments

Detected Reactor Spectrum

1.8MeV threshold in Inverse Beta Decay

No near detector necessary!

Reactor from neutron rich fission fragments Detected Spectrum Cross section νe + p → e+ + n

νe

Zacek G. et al., Phys. Rev. D34, 2621 (1986).

Counts [MeV-1 h-1]

Gösgen

Ee+ (MeV)

5

Patrick Decowski / UC Berkeley

from 55 Reactor Cores in Japan

KL

Reactor neutrino flux: ~6x106 cm-2s-1

Japan Korean World

358 m 36◦2535.562 137◦1843.495

long. lat. alt.

Distance [km] 100 200 300 400 500 600 700 800 900 1000 Events/ year / kton 50 100 150 200 250 300 350 400 450 500

Effective baseline ~180km

70 GW (7% of world total) is generated at 130-220 km distance from Kamioka

• Mt. Ikenoyama

1000m rock = 2700 mwe

νe

6

Patrick Decowski / UC Berkeley

KamLAND detector

• 1 kton Scintillation Detector
• 6.5m radius balloon filled with:
• 20% Pseudocumene (scintillator)
• 80% Dodecane (oil)
• PPO
• 34% PMT coverage
• ~1300 17” fast PMTs
• ~550 20” large PMTs
• Multi-hit electronics
• Water Cherenkov veto counter

Water Cherenkov Outer Detector 1800 m3 Buffer Oil

20 m

}

3200 m3

7

Patrick Decowski / UC Berkeley

KL2002 reactor result PRL 90 021802 (2003). Solar PRL 92 071301 (2004).

0.4 1.0 2.6 8.5

Energy [MeV]

Neutrino Astrophysics Verification of SSM Neutrino Geophysics Study of earth heat model Neutrino Physics Precision measurement

• f oscillation parameters

Neutrino Cosmology Verification of universe evolution, SSM solar neutrino geo-neutrino reactor neutrino supernova, relic neutrino, solar anti-neutrinos etc.

Future Low background phase neutrino electron elastic scattering inverse beta decay

ν + e− → ν + e−

¯ νe + p → e+ + n

KamLAND Physics Capabilities

νe

Geoneutrinos Nature 436, 499 (2005). KL2004 reactor result PRL 94 081802 (2005).

Recent Results Accepted by PRL

8

Patrick Decowski / UC Berkeley

Analysis Improvements

6m

“Fuzzy” cut using event characteristics to distinguish signal from accidentals

S/B

Energy threshold from 2.6MeV → 1.0MeV (Inverse Beta Decay threshold)

accidentals

Max Radius(m) Lifetime(days) Exposure(ton-yr) Exposure Increase

KL2002 5 145 162 1x KL2004 5.5 515 766 4.7x Latest 6 1491 2881 17.8x

9

Patrick Decowski / UC Berkeley

R [cm]

100 200 300 400 500 600

Reconstructed Energy Deviation [%]

• 4
• 2

2 4 Co

Co

Co

Co

Ge - Configuration 4+1

Ge - Configuration 5+1

Ge - Configuration 5+1w

Ge - Configuration 6+1w

Energy deviation <2%

Range of radioactive sources (250keV to 6MeV):

203Hg, 137Cs, 68Ge, 65Zn, 60Co, 241Am9Be, 210Po13C

Full Volume Calibration System

R [cm]

100 200 300 400 500 600

Reconstructed Position Deviation [cm]

• 6
• 4
• 2

2 4 6 Co

60

Co

60

Co

60

Co

60

Ge - Configuration 4+1

68

Ge - Configuration 5+1

68

Ge - Configuration 5+1w

68

Ge - Configuration 6+1w

68

Position uncert. R<5.5m 3cm FV uncert. ~1.6%

Use 12B/12N spallation uniformity for 5.5m<R<6m → Total FV uncert R<6m: 1.8%

z-axis

Deployed in 2005-2006

10

Patrick Decowski / UC Berkeley

Systematic Uncertainties

}

Primarily affecting θ12 Sum: 4.1% Sum: 2.0% Systematic uncertainties between Δm212 and θ12 decouple to a large degree

11

Patrick Decowski / UC Berkeley

Balloon

• Inverse beta-decay selection:
• Rprompt, delayed < 6 m
• 0.9 MeV < Eprompt < 8.5 MeV
• 1.8 MeV < Edelayed < 2.6 MeV
• ΔR < 2m
• 0.5 μs < ΔT < 1000 μs
• L-selector: Use event characteristics

to limit effect of accidental backgrounds at high R

• Muon-induced spallation event cuts:
• 2 ms veto after every μ
• 2 s veto for showering/bad μ
• 2 s veto in a R = 3m tube along track

Event Selection

Prompt Delayed

Zp [m] Zd [m] ρ2 [m] ρ2 [m]

12

Patrick Decowski / UC Berkeley

L-selector: Signal/Accidentals Discrimination

(MeV)

prompt

E 1 2 3 4 5 6 7 8 Efficiency (%) 40 60 80 100

Efficiency

L = fνe fνe + facc

fνe(Ep, Ed, ∆R, ∆T, Rp, Rd) facc(Ep, Ed, ∆R, ∆T, Rp, Rd)

Use prompt-delayed event characteristics to distinguish Accidental BG from Signal

Generate Accidentals PDF from DATA (random pairs): Generate Signal PDF from MC (no-osc spectrum): L-selector (calculated EbE):

Establish L-selector cuts for different Ep bins, where FOM is maximal

FOM = S √S + Bacc

Lcut

(Ep bins of 0.1MeV)

If for candidate event pair L > Lcut → anti-neutrino Efficiency for Ep>~3MeV as expected from spatial cuts alone

13

Patrick Decowski / UC Berkeley

Dominant BG: 13C(α,n)16O

(MeV)

visible

E 1 2 3 4 5 6 7 8

• decay
• events/50keV/
• 7

10 1 2 3 4 (MeV)

visible

E 1 2 3 4 5 6 7 8

• decay
• events/50keV/
• 7

10 1 2 3 4

KamLAND PoC calib. source DATA Ground state (MC)

• exc. state (MC)

st

1

• exc. state (MC)

nd

2 Total (MC)

Cross sections tuned using detector MC

D.McKee et al., NIM A527, 272 (2008)

210Pb 210Bi 210Po 206Pb

T1/2=22yr 5d 138d

α, E=5.3MeV

210Po/13C

mixture

210Po13C source deployed

into the detector

S.Harissopulos et al., PRC 72, 062801 (2005) JENDL

}

From

222Rn chain:

1.1% abundance of 13C in LS →13C(α,n)16O

6.130 MeV 6.049 MeV 3- 0+ 0+

16O

γ

e+e-

Good match after scaling C.S.

14

Patrick Decowski / UC Berkeley

Backgrounds

}

Background from 222Rn chain

}

Cosmogenic Accidental Coincidences Geo-neutrinos are a background to the neutrino oscillation measurement →Talk by John Learned

Total excluding geo-neutrino

However, analysis is done by simultaneously fitting geo- and reactor neutrinos !

Using one geological model, which assumes 16TW of radiogenic heat from U+Th geo-neutrinos, expect 69.7 events

15

Patrick Decowski / UC Berkeley

(MeV)

p

E Events / 0.425 MeV 50 100 150 200 250 1 2 3 4 5 6 7 8

KamLAND data no oscillation best-fit osci. accidental O

16

,n)

• C(

13 e

• best-fit Geo

best-fit osci. + BG

e

• + best-fit Geo

Energy Spectrum

Fit to scaled no-oscillation spectrum excluded at 5.1σ

From Mar 9, 2002 to May 12, 2007 1491 live days, 2881 ton-year exposure (3.8x KL2004)

arXiv:0801.4589 / Accepted by PRL

16

Patrick Decowski / UC Berkeley

Neutrino Oscillation Parameters

• 1

10 1

• 4

10

KamLAND 95% C.L. 99% C.L. 99.73% C.L. best fit Solar 95% C.L. 99% C.L. 99.73% C.L. best fit

10 20 30 40

σ 1 σ 2 σ 3 σ 4 σ 5 σ 6

5 10 15 20

σ 1 σ 2 σ 3 σ 4

12

θ

2

tan

2

χ Δ )

2

(eV

21 2

m Δ

2

χ Δ

Best-fit light side: Best-fit dark side:

LMA-2 LMA-1 LMA-0

tan2 θ = 0.56+0.14

−0.09

tan2 θ = 1.84

∆m2 = 7.64 × 10−5 eV2 ∆m2 = 7.58+0.21

−0.20 × 10−5 eV2

KamLAND Global Solar

17

Patrick Decowski / UC Berkeley

Neutrino Oscillation Parameters

• 1

10 1

• 4

10

KamLAND 95% C.L. 99% C.L. 99.73% C.L. best fit Solar 95% C.L. 99% C.L. 99.73% C.L. best fit

10 20 30 40

σ 1 σ 2 σ 3 σ 4 σ 5 σ 6

5 10 15 20

σ 1 σ 2 σ 3 σ 4

12

θ

2

tan

2

χ Δ )

2

(eV

21 2

m Δ

2

χ Δ

Best-fit light side: Best-fit dark side:

LMA-2 LMA-1 LMA-0

tan2 θ = 0.56+0.14

−0.09

tan2 θ = 1.84

∆m2 = 7.64 × 10−5 eV2 ∆m2 = 7.58+0.21

−0.20 × 10−5 eV2

KamLAND Global Solar

17

Patrick Decowski / UC Berkeley 12

!

2

tan )

2

eV

• 5

(10

21 2

m " 0.2 0.3 0.4 0.5 0.6 0.7 0.8 5 6 7 8

KamLAND + Solar 95% C.L. 99% C.L. 99.73% C.L. best fit

Global Analysis

tan2 θ = 0.47+0.06

−0.05

∆m2 = 7.59+0.21

−0.21 × 10−5 eV2

Solar Experiments + KamLAND:

18

Patrick Decowski / UC Berkeley

[events/day]

e

• Expected no-osc

0.4 0.6 0.8 1 1.2 1.4 1.6 2002 2003 2004 2005 2006 2007

Calculated from detailed reactor data Monthly JAIF electric power data (scaled)

Reactor Signal Changes with Time

Many reactor inspections Steam pipe rupture New, nearby reactor being turned on & off Big earthquake July 16, 2007

rate [events/day]

e

• Expected no-osc

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 rate [events/day]

e

• Observed

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

(Rate for Ep>2.6MeV)

90% C.L. E x p e c t e d N

• s

c i l l a t i

• n

2007 2006

{

Upper limit on hypothetical georeactor at Earth’s center of 6.2TW at 90% C.L. Detailed operational records from all 55 reactors in Japan

19

Patrick Decowski / UC Berkeley

(km/MeV)

e

• /E

L 10 20 30 40 50 60 70 Survival Probability 0.2 0.4 0.6 0.8 1 1.2 1.4

CHOOZ data

L/E Plot

Pee = 1 − sin2 2θ sin2(∆m2 4 L E )

Oscillation pattern for a mono-energetic at one baseline

νe

20

Patrick Decowski / UC Berkeley

(km/MeV)

e

• /E

L 10 20 30 40 50 60 70 Survival Probability 0.2 0.4 0.6 0.8 1 1.2 1.4

CHOOZ data

(km/MeV)

e

• /E

L 10 20 30 40 50 60 70 Survival Probability 0.2 0.4 0.6 0.8 1 1.2 1.4

e

• Data - BG - Geo

CHOOZ data

Best-fit oscillation accounting for energy spectrum and reactor distribution

L/E Plot

Pee = 1 − sin2 2θ sin2(∆m2 4 L E )

Oscillation pattern for a mono-energetic at one baseline

νe

20

Patrick Decowski / UC Berkeley

(km/MeV)

e

• /E

L 10 20 30 40 50 60 70 Survival Probability 0.2 0.4 0.6 0.8 1 1.2 1.4

CHOOZ data

L/E Plot

(km/MeV)

e

• /E

L 10 20 30 40 50 60 70 Survival Probability 0.2 0.4 0.6 0.8 1 1.2 1.4

e

• Data - BG - Geo

CHOOZ data

Pee = 1 − sin2 2θ sin2(∆m2 4 L E )

LMA-1 LMA-2 LMA-0

20

Patrick Decowski / UC Berkeley

Best-fit Light and Dark Side

(km/MeV)

e

• /E

L 20 30 40 50 60 70 80 90 100 Survival Probability 0.2 0.4 0.6 0.8 1

e

• Data - BG - Geo

Best-fit Light Side Best-fit Dark Side

Difference in best-fit on the light and dark side is very small

• 1

10 1

• 4

10

KamLAND 95% C.L. 99% C.L. 99.73% C.L. best fit Solar 95% C.L. 99% C.L. 99.73% C.L. best fit σ 1

5 10 15 20

σ 1 σ 2 σ 3 σ 4

12

θ

2

tan )

2

(eV

21 2

m Δ

2

χ Δ

Small difference in 1st bin

Analysis includes Earth matter effects

21

Patrick Decowski / UC Berkeley

KL2002 reactor result PRL 90 021802 (2003). Solar PRL 92 071301 (2004).

0.4 1.0 2.6 8.5

Energy [MeV]

Neutrino Astrophysics Verification of SSM Neutrino Geophysics Study of earth heat model Neutrino Physics Precision measurement

• f oscillation parameters

Neutrino Cosmology Verification of universe evolution, SSM solar neutrino geo-neutrino reactor neutrino supernova, relic neutrino, solar anti-neutrinos etc.

Future Low background phase neutrino electron elastic scattering inverse beta decay

ν + e− → ν + e−

¯ νe + p → e+ + n

KamLAND Physics Capabilities

νe

Geoneutrinos Nature 436, 499 (2005). KL2004 reactor result PRL 94 081802 (2005).

Recent Results Accepted by PRL →Talk by Joshua Klein

22

Patrick Decowski / UC Berkeley

Low Background Phase

Liquid Scintillator from KamLAND is distilled into PC, MO and PPO, remixed and purged with N2 Reservoir tank

23

Patrick Decowski / UC Berkeley

1st Purification Campaign

First campaign May-August’07 Constraint: had to stop due to blasting in Kamioka

• Purified scintillator introduced at the top
• 1700 m3 of LS purified, avg. 1m3/hr
• Density instability: LS started to mix half-

way during the purification

• Near end of purification introduced 170m3
• f lighter LS, purified 2x
• Layers have remained separated

Selecting on 85Kr β energy

24

Patrick Decowski / UC Berkeley

Background Lower Volume Upper Volume

40K

0.29±0.03 0.29±0.03

85Kr

0.36±0.02 (2.8±0.8)x10-2

210Po

0.33±0.03 0.21±0.03

210Bi

0.24±0.05 (4.8±2.6)x10-3

Upper Volume

1st Purification BG Levels

Lower Volume

2x purified

Before Purification Before Purification

1x purified

Background reduction fractions:

Events/10keV/s/m3 Events/10keV/s/m3

10-1 10-9 10-2 10-9 1.2

Energy [MeV]

→ Main reactor & geo-neutrino BG 13C(α,n)16O already down

After Purification

25

Patrick Decowski / UC Berkeley

Towards the Low Background Phase

• Cavity blasting in Kamioka over
• Avoid mixing of Liquid Scintillator:
• Density and temperature control measures
• Introduce purified heavier LS into KamLAND

from the bottom

• Much time spent finding and fixing leaks
• Several found near top of detector and in the distillation system
• Upgrade of PPO distillation tower for better performance and stability
• Improved online reconstruction and monitoring tools
• Start 2nd purification campaign around end of May’08
• Electronics upgrade to do 11C tagging for CNO/pep solar neutrinos

→Poster by Tommy O’Donnell

26

Patrick Decowski / UC Berkeley

Summary

• Era of precision measurements of neutrino oscillation parameters
• KamLAND’s results with a data-set of 2881 ton-yr with lowered

threshold and significantly improved systematic uncertainties

• KamLAND:
• KamLAND+Solar:
• LMA-0 and LMA-2 excluded at > 4σ
• Future: Low background phase
• Solar 7Be and CNO/pep neutrinos
• Reactor and geo-neutrino measurements will continue

tan2 θ = 0.56+0.14

−0.09

∆m2 = 7.58+0.21

−0.20 × 10−5 eV2

tan2 θ = 0.47+0.06

−0.05

∆m2 = 7.59+0.21

−0.21 × 10−5 eV2

27

Patrick Decowski / UC Berkeley

Precision Neutrino Measurements

KL 2002

28

Patrick Decowski / UC Berkeley

Precision Neutrino Measurements

KL 2002

• 5

• 4

• 2

• 1

KL 2004

28

Patrick Decowski / UC Berkeley

Precision Neutrino Measurements

KL 2002

• 5

• 4

• 2

• 1

KL 2004

• 1

• 4

• 4
• 5
• 6
• 2

• )

• KL 2007

Neutrino Oscillation: A precision measurement!

28