KamLAND: Measuring Neutrino-Oscillation with...
Transcript of KamLAND: Measuring Neutrino-Oscillation with...
KamLAND: Measuring Neutrino-Oscillation with
Reactors
Patrick DecowskiUC Berkeley
UCSC November 15, 2004
Patrick Decowski / UC Berkeley
L
Reactor Experiments
νe
νe
νx
νe
Few MeV anti-neutrinos, energy too low to produce μ or τ➡ disappearance experiments
?
?
P (νe → νe) = 1 − sin2 2θ sin21.27∆m2L
E
Patrick Decowski / UC Berkeley
• Many different experiments
• Baselines up to 1km
• No evidence for disappearanceνe
Oscillation searches with Reactors
Reactors have played an important role in the early history of neutrinos and in neutrino-oscillation searches: 1953 - Present
Poltergeist(Reines & Cowan 1955)
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Nobs/
Nex
p
101
102
103
104
105
Distance to Reactor (m)
ILL Savannah River Bugey Rovno Goesgen Krasnoyarsk Palo Verde Chooz
Reactor Anti-Neutrinos
Patrick Decowski / UC Berkeley
Reactor Neutrinos
235U fission
The stable products most likelyfrom Uranium fission:
Together 98 protons and 136 neutrons
235
92 U + n → X1 + X2 + 2n
9440Zr
14058 Ce
6 neutrons have to -decay to reach stable matter, producing 6 / fissionνe
94 140
Patrick Decowski / UC Berkeley
Calculating Neutrino Spectra
• Fission rates provided by reactor companies
• Chiefly function of thermal power
• Weak function of inlet T: 10% -> ~0.15% rate change
10-5
10-4
10-3
10-2
10-1
1
0 1 2 3 4 5 6 7 8 9 10
Energy (MeV)
ne
utr
ino
s/M
eV
/fis
sio
n
238U
239Pu
235U
241Pumeasured at ILL
K.Schreckenbach et al., Phys.Lett.B160(1985)325
A.A.Hahn et al., Phys.Lett.B218(1989)365
theoretical 744 tracesP.Vogel et al., Phys. Rev. C24(1981)1543
Date
3/3 1/30
fis
sio
n/s
)1
9F
iss
ion
ra
te (
10
0
1
2
3
4
5
6
7
819
x10
Mar Apr May Jun
2002Jul Aug Sep Oct Nov Dec Jan Feb Mar
2003Apr May Jun Jul Aug Sep Oct Nov Dec
A typical 1.3GWe class BWR in Japan
0
1
2
3
4
5
6
7
8
U235
U238Pu239
Pu241
Data provided according to the special agreement betweenTohoku Univ. and a Japanese nuclear power reactor operator.
Only 4 isotopes relevant
Patrick Decowski / UC Berkeley
Detected Reactor Spectrum
1.8MeV threshold in Inverse Beta Decay
In practice, only 1.5 neutrinos/fission detectable
Patrick Decowski / UC Berkeley
• Reactors produce:
• Only anti-neutrinos
• On average around 6 per fission
• Reactor anti-neutrino spectra measured at short baseline
• Gösgen (1986)
• Bugey (1996)
• Chooz (1999)
• Palo Verde (2001)
Reactor Spectra
Gösgen
νe
Agreement of measured with expected~2% - no near detector necessary!
Patrick Decowski / UC Berkeley
Distortion of Spectrum
Neutrino oscillation changes both the overall normalization and the shape of the spectrum
P (νe → νe) = 1 − sin2 2θ sin21.27∆m2L
E
Patrick Decowski / UC Berkeley
Anti-neutrino Detection Method
Reaction process: Inverse beta decay
νe + p → e+
+ n
n + p → d + γ
e n
e+
2.2MeV
Scintillatorγ γ
γ
Scintillator is both target and detector
• Distinct two step process:
• prompt event: positron
• delayed event: neutron capture after ~210μs• 2.2 MeV gamma
Delayed coincidence: good background rejection
Eνe! Eprompt + 0.8MeV
Patrick Decowski / UC Berkeley
Long Baseline Means Large Detectors
1m
PoltergeistChooz4 m
KamLAND20 m
Log 1
0(M
uon
Flux
) (m
-2s-
1 )Depth (mwe)
The KamLAND Experiment
Patrick Decowski / UC Berkeley
from 53 Reactor Cores in Japan
~95.5% from Japan~3.5% from Korea
70 GW (7% of world total) is generated at 130-220 km distance from Kamioka.
Effective distance ~180km
∼ 6 × 106/cm2/secReactor neutrino flux,
358 m36◦25′35.562′′
137◦18′43.495′′long.lat.alt.
1000m rock= 2700 mwe
νe
KamLAND Collaboration
Patrick Decowski / UC Berkeley
Building KamLAND
1998
September 2001 January 2002
Data Taking
September 2000
October 1999
Patrick Decowski / UC Berkeley
KamLAND detector
• 1 kton Scintillation Detector
• 6.5m radius balloon filled with:
• 20% Pseudocumene (scint)
• 80% Dodecane (oil)
• 34% PMT coverage
• ~1300 17” fast PMTs
• ~550 20” large PMTs
• Multi-hit, deadtime-less electronics
• Water Cherenkov veto counter
Water Cherenkov Outer Detector
Buffer Oil1800 m3
20m
3200 m3
Patrick Decowski / UC Berkeley
0.4 1.0 2.6 8.5 Visible energy [MeV]
Neutrino Astrophysicsverification of SSM
Neutrino Geophysicsverification of earth
heat model
Neutrino PhysicsPrecision measurement of oscillation parameters
Neutrino Cosmologyverification of
universe evolution
7Be solar neutrino geo-neutrino reactor neutrino supernova relic neutrinoetc.
1st reactor result PRL 90, 021802 (2003)2nd reactor result hep-ex/0406035
Solar PRL 92, 071301 (2004)
ν̄e“soon”future2nd phase
neutrino electron elastic scatteringinverse beta decayν + e− → ν + e− ν̄e + p → e+ + n
KamLAND Physics Capabilities
Detector Performance
Patrick Decowski / UC Berkeley
Trigger Efficiency
0.9MeVPrompt Trigger:200 PMT hits (~0.7MeV)
Delayed Trigger:120 hits for 1msec after prompt trigger
Reactor analysis threshold: 2.6MeV, trigger is 100% efficient
(To study radioactive backgrounds)
Patrick Decowski / UC Berkeley
0
500
1000
1500
2000
2500
visible energy (MeV)4 6 8 10 12 14 16
12
B/12
N
0
0.2
0.4
0.6
0.8
1
0 1 2 3 4 5
68
Ge65
Zn
n-p n-12
C
60
Co
visible energy (MeV)
0 1 2 3 4 5 6 7 8 9
-0.04
-0.02
0
0.02
0.04
Energy (MeV)
E/E
∆
-ray sourcesγ decayβ
Ge68
Co
60
Zn65
n-p
C12n-
N12
B/12
Energy scale error at 2.6 MeVCherenkov/BirksTime dependencePosition dependence20” PMT non-linearityTotal
1.0%1.3%1.0%0.8%2.0%
σ
E∼ 6.2%√
EEnergy Resolution:
Patrick Decowski / UC Berkeley
0 5 10 15 20 25 30 35 40 45 50-8
-6
-4
-2
0
2
4
6
8prompt
)2
(m2
+y2x
z (m
)
0 5 10 15 20 25 30 35 40 45 50-8
-6
-4
-2
0
2
4
6
8delayed
)2
(m2
+y2x
z (m
)
Inverse beta-decay selection:
● Rprompt, delayed < 5.5 m
● ΔR < 2 m
● 0.5 μs < ΔT < 1000 μs● 1.8 MeV < Edelayed < 2.6 MeV
● 2.6 MeV < Eprompt < 8.5 MeV
• 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
Tagging efficiency 89.8%
Dead time 9.7%Balloon
Patrick Decowski / UC Berkeley
02 03 04 05 06 07 08 09 10 1112 01 02 03 04 05 06 07 08 09 10 1112 01
ho
ur
0
2
4
6
8
10
12
14
16
18
20
22
24
2002 2003 2004
Liv
eT
ime
[d
ay
]
0
100
200
300
400
500
600
Physics Run Calibration Run (source,laser and LED)
Test/Bad Run
1st result4 Mar 2002 - 6 Oct 2002
145.1 days
2nd result9 Mar 2002 - 11 Jan 2004
515.1 days
X 3.55 live time
Patrick Decowski / UC Berkeley
Quite a few Reactors were Off
Jul/02 Jan/03 Jul/03 Jan/04
(ev
ents
/day
)no-o
scil
exp
N
0
0.2
0.4
0.6
0.8
1
1.2
Wei
gh
ted
dis
tan
ce(k
m)
150
160
170
180
190
200
210
no-oscil
expN
Weighted distance
Period for first result
In 2002 it was discovered that some of the Japanese reactor companies had falsified safety records
Anti-neutrino flux was 20% lower
Backgrounds
Patrick Decowski / UC Berkeley
Accidental Backgrounds
0.0086 ± 0.000051st result (5m fiducial 162 ton-yr)
2.69 ± 0.022nd results (5.5m fiducial 766.3 ton-yr)
prompt energy [MeV]
0 1 2 3 4 5 6 7 8 9 10
co
un
ts/2
0k
eV
10-1
1
10
102
103
104
105
106
208Tl from wall
210Bi in LS
analysis threshold
Fiducial volume is limited by accidental backgrounds.
energy [MeV]0 2 4 6 8 10 12 14
co
un
ts/b
in
0
10
20
30
40
50
60
9Li > 85% (90%CL)
detection eff. 98.7%
Spallation events
2 sec VETO
for all volume
350 → 0.1
Eextra > 3GeV
2 sec VETO
for 6mφ cylinder
93.8% eff.
86 → 5.7
Eextra < 3GeV
9LiT1/2 = 0.18 sec
9Be8Be + n
β ntotal 50%
4.8 ± 0.9 events (E > 2.6 MeV)(2nd result)
Patrick Decowski / UC Berkeley
Dealing with muons
0 to 2 ms after muon outside scint: veto entire volume.
>2 ms after muon outside scint:all of fiducial volume is okay.
0 to 2 ms after muon in scint: veto entire volume
2 ms to 2 s after muon in scint: veto events within 3 m of track
0 to 2 ms after shower in scint: veto entire volume
2 ms to 2 s after shower in scint: also veto entire volume
3 Typical scenarios of muons interacting in KamLAND
Patrick Decowski / UC Berkeley
with Radioactive Sourceswith Muon Spallation
z-axis[cm]-600 -400 -200 0 200 400 600
z d
ev
iati
on
[cm
]
-10
-5
0
5
10Co
Zn
GeAm/Be(4.4MeV)
Am/Be(~7.6MeV)
0 0.2 0.4 0.6 0.8 1 1.2 1.40
500
1000
1500
2000
2500
3000
3500
4000
fiducial
balloonradius
3(R/6.5m)
Events/Bin
0 0.2 0.4 0.6 0.8 1 1.2 1.40
200
400
600
800
1000
fiducial
balloonradius
3(R/6.5m)
3(delayed R/6.5m)0 0.2 0.4 0.6 0.8 1 1.2p
rom
pt
R -
del
ayed
R [
cm]
-150
-100
-50
0
50
100
150
Li (prompt energy>4MeV)9He8
fiducial
12B
delayed coincidence
Fiducial/Total Volume Ratios
Geometrical
12Bp(n, γ)d
0.595 ± 0.013
0.607 ± 0.006 ± 0.006
0.587 ± 0.0139Li relative < 2.7%
Fiducial Volume uncertainty 4.7%
p(n, γ)d
9Li
(=
696.9 m3
1171 ± 25 m3
)
β + n
Fiducial Volume Uncertainty
Fast neutron
OD miss-taggingrock penetrating
Total
<0.4<0.5<0.9
recoil proton+
neutron capture
sec]µ T [∆500 1000 1500 2000
0
10
20
30
40
50
60
70
80
delayed energy [MeV]1.6 1.8 2 2.2 2.4 2.6 2.8 3
0
10
20
30
40
50
60
70
prompt radius [cm]
0 100 200 300 400 500 600 700 800
0
10
20
30
40
50
5m 5.5m
prompt energy [MeV]0 2 4 6 8 10 12 14
0
5
10
15
20
25
30
35
40
fiducial
Tagged eventsR-distribution prompt energy time difference delayed energy
Patrick Decowski / UC Berkeley
(α,n) Background
222Rn (3.8d)218Po214Pb
214Bi214Po210Pb (22.3y)
210Bi (5.013d)
210Po (138.4d)206Pb (stable)
equilibrium
(Eα = 5.3MeV )
10.6 events (E>2.6MeV)
1.1% abundance (measured)
13C(α,n)16O ∼ 10−7
13C(α,n)16O(g.s.)
13C(α,n)16O∗(6.05)
13C(α,n)16O∗(6.13)
→12 C(n,nγ)12C
13C(α,n)16O(g.s.)
low energy
~4.4 MeV
~6 MeV
Visible Energy (MeV)
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Ev
en
ts/M
eV
/se
c
10-4
10-3
10-2
10-11
10
102
103
104
85Kr210Po
210Bi
Low energy spectrum
Patrick Decowski / UC Berkeley
Uncertainty %Fiducial volume 4.7
Energy threshold 2.3Cuts efficiency 1.6
Live time 0.1Reactor thermal power 2.1
Fuel composition 1.0Antineutrino spectra 2.5
Cross section 0.2Total uncertainty 6.5
Systematic Uncertainties
Results
Patrick Decowski / UC Berkeley
1st Reactor Result
Data Summaryfrom March 4 to October 6, 2002
145.1 live days, 162 ton-year exposure
Analysis threshold 2.6 MeVexpected signal
BGobserved
86.8 ± 5.6
54Neutrino disappearance at 99.95% CL.
R = 0.611 ± 0.085(stat) ± 0.041(syst)
KamLAND collaboration, Phys.Rev.Lett.90(2003)021802
1 ± 1
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Nob
s/N
no_osc
illa
tion
101
102
103
104
105
Distance to Reactor (m)
ILL Savannah River Bugey Rovno Goesgen Krasnoyarsk Palo Verde Chooz
KamLAND
Evidence for reactor neutrino disappearance
Patrick Decowski / UC Berkeley
Before KamLAND: After KamLAND 1st result:
RSFPFCNCdecay
decoherenceetc.
tan2(!)
"m
2 in
eV
2
10-12
10-11
10-10
10-9
10-8
10-7
10-6
10-5
10-4
10-3
10-4
10-3
10-2
10-1
1 10 102
Ga
Cl
SuperK
SNO
LMA
SMA
LOW
VAC
LMA
SMA
LOW
RSFPFCNCdecay
decoherenceetc.
VAC
KamLAND reactor disappearance excluded all but LMA solution
Data Summaryfrom 9 Mar 2002 to 11 Jan 2004
515.1 live days, 766.3 ton-year exposure
expected signalBG
observed
Neutrino disappearance at 99.998% CL.
KamLAND collaboration, hep-ex/0406035
365.2 ± 23.7
258
for Mar to Oct 2002is consistent with first results
2nd resultData Summaryfrom March 4 to October 6, 2002
145.1 live days, 162 ton-year exposure
expected signalBG
observed
86.8 ± 5.6
54Neutrino disappearance at 99.95% CL.
R = 0.611 ± 0.085(stat) ± 0.041(syst)
KamLAND collaboration, Phys.Rev.Lett.90(2003)021802
1 ± 1
1st result
×4.7 exposure (×3.55 live time, ×1.33 fiducial)
17.8 ± 7.3
R = 0.658 ± 0.044(stat) ± 0.047(syst)
R = 0.601 ± 0.069(stat) ± 0.042(syst)
Caveat: ratio does not have an absolute meaning in KamLAND, since, with oscillations, it depends on which reactors are on/off
Patrick Decowski / UC Berkeley
Delayed Coincidence
(M
eV)
del
ayed
E 2
3
4
5
(MeV)prompt
E0 1 2 3 4 5 6 7 8
0
12C(n, γ) expected1.5
accidental
1.8 < Edelayed < 2.6 MeV
2.6 < Eprompt < 8.5 MeV
Clear delayed coincidence events
Patrick Decowski / UC Berkeley
A fit to a simple rescaled reactor spectrumis excluded at 99.6% CL
Energy Spectrum
Best-fit oscillation:
∆m2
= 7.9+0.6−0.5 × 10
−5eV
2
tan2θ = 0.46
Patrick Decowski / UC Berkeley
Unbinned Likelihood: 2 nu Oscillation
∆m2
= 7.9+0.6−0.5 × 10
−5eV
2
LMA2 is excluded at 98.0%
LMA0 is excluded at 97.5%
tan2θ = 0.46
Patrick Decowski / UC Berkeley
Global Analysis
Solar Experiments are sensitive to
tan2θ = 0.40
+0.10−0.07
∆m2
= 7.9+0.6−0.5 × 10
−5eV
2
Most precise to date!
θ
∆m2KamLAND is most sensitive to
Patrick Decowski / UC Berkeley
No
Osc
illatio
n
Current statistics not enough to make firm statements
Correlation with Reactor Power
Recent extensive inspection may provide another chance to investigate the correlation.
04
05
06
07
08
09
10
11
12
01
02
03
04
05
06
07
08
09
10
11
12
01
02
03
04
05
06
07
08
)2
MW
/cm
-12
Ele
ctr
ic p
ow
er
flu
x (
10
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
-1010!
Electric power flux at KamLAND
0
1
2
3
4
5
6
7
8
9
10
JAIF data
2002.
2003.
2004.
?
2nd result
Patrick Decowski / UC Berkeley
Ratio of measured to expected spectrum
Oscillation pattern for one of a given energy
νe
Best-fit oscillation accounting for energy spectrum and reactor distribution
Pee = 1 − sin2 2θ sin2(∆m2
4
L
E)
Patrick Decowski / UC Berkeley
V.D.Bager et al.,PRL82,2640(1999)
E.Lisi et al.,PRL85,1166(2000)
oscillation Pee = 1 − sin2 2θ sin2(∆m2
4L
E)
decay Pee = (cos2 θ + sin2 θ exp(−m2
2τ
L
E))2
decoherence Pee = 1 − 12
sin2 2θ(1 − exp(−γL
E))
Future activities at KamLAND
Patrick Decowski / UC Berkeley
Karsten Heeger, LBNL Sendai, September 20, 2003
Development and Testing of an Off-Axis
Calibration System for KamLAND
KamLAND 4! GroupLawrence Berkeley National Laboratory
It’s being built!z-axis calibration
full volume calibration
Future Improvements
Reduce Fiducial Volume and Energy Uncertainty
3% rate error 1% scale error 3kt-yr data accumulation
KamLAND onlyrate+shape sensitivity
(rough estimation)
capability to reject full mixing
mixing angle determination comparable with current solar data
fact
or ~
2 im
prov
emen
t?
Patrick Decowski / UC Berkeley
What is the Earth’s heat source?
0.0
20.0
50.0
80.0
110.0
mW/m2Surface Heatflux at 2x2 degrees
Surface heat flux measurement
Total flow of 44 TWis 40 times larger than total world reactor power.
Combining all the available geo-chemical knowledge:
Radioactivity 20 TWUranium 8TW, Thorium 8TW,
Potassium 4TW
U,Th are condensed in the
crust
Patrick Decowski / UC Berkeley
Neutrinos from radioactivity provide direct information on the earth’s interior
Uranium seriesThorium series
Analyzing data E< 2.6MeVanalysis forthcoming
Patrick Decowski / UC Berkeley
Verification of the SSM with more abundant neutrinos is important.
Real time measurement of low energy solar neutrinos has never been done.
KamLAND
Real time measurements
(pp-chain) 98.5%
€
D p→3Heγ
€
3He 4He→7Beγ
€
3He 3He→4He p p
13.8%
84.7%
€
7Bee−→7Liν e (γ)(7Be)
€
7Be p→8Bγ
€
p p→De+ν e
€
pe− p→ Dν e(pp) (pep)
99.77% 0.23%
~2×10-5%
13.78% 0.02%
€
7Li p→4He 4He
€
3He p→4He e+ν e(hep)
€
8B→8Be* e+ν e
€
4He 4He(8B)
7Be ~14% 8B ~0.02%
Fusion reaction in the Sun
Towards 7Be Solar Neutrino Observation
Patrick Decowski / UC Berkeley
Detecting Solar Neutrinos
0
500
1000
1500
2000
2500
3000
0 200 400 600 800 1000
cou
nts
/25
keV
Energy [keV]
KamLAND future goal
14C
(assume 14
C/12
C=10-18
)
232Th(5.2x10
-17g/g)
11C
(1.0x103 day
-1kton
-1)
7Be !(SSM)
210Pb
(assume 1!Bq/m3)
39Ar,
85Kr
(assume 1!Bq/m3)
All
All ! (SSM)
Detecting through elastic scattering:
Challenging experiment: Reducing backgrounds is crucial!
νeνe
e- e-
1104∼5
110∼6
requiredfurther improvement
For future physics
Background now goal
238U(by Bi-Po) 3.5× 10−18g/g OK!!
238U(by 234Pa) O(10−15g/g)(Max.) 10−18g/g
232Th (by Bi-Po) 5.2× 10−17g/g OK!!
40K 2.7× 10−16g/g(max.) < 10−18g/g
210Pb ∼ 10−20g/g 5× 10−25g/g ∼ 1µBq/m3
85Kr, 39Ar 85Kr =0.7Bq/m3 1µBq/m3
222Rn 238U = 3.5× 10−18g/g OK!! (1µBq/m3)
(after purification) = 3.3× 10−8Bq/m3
222Rn 1mBq/m3
(during purification) 210Pb = 0.5µBq/m3 after decay
4
Visible Energy (MeV)
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Ev
en
ts/M
eV
/se
c
10-4
10-3
10-2
10-11
10
102
103
104
Current background level
7Be νpp ν
14C
11C
85Kr210Bi210Po
daughter of 210Pb
SSM
Patrick Decowski / UC Berkeley
• KamLAND’s updated results strengthen support for “neutrino disappearance”
• Best-fit KamLAND+Solar oscillation parameters are:
• No-oscillation hypothesis rejected at 99.6%
• Other disappearance mechanisms strongly disfavored
• Neutrino oscillation observed in:
• solar, atmospheric, accelerator and reactor experiments
• Look out for many other things to come out of KamLAND!
Summary
∆m2
= 7.9+0.6−0.5 × 10
−5eV
2tan
2θ = 0.40
+0.10−0.07 Best to date!