Low Energy Neutrino Physics at the Kuo-Sheng Reactor ...
1
Low Energy Neutrino Physics at the Kuo-Sheng Reactor Laboratory Hau-Bin Li (on behalf of the TEXONO Collaboration) Institute of Physics, Academia Sinica, Taipei 115, Taiwan 1 Introduction 1.1 TEXONO collaboration • TEXONO established in 1996. • First big collaboration between China and Taiwan. • Magnetic moment results published. • Period II → CsI detector data. • Next → νN scattering with ULE-HPGe detector. 1.2 ¯ ν e e - → ¯ ν X e - Scattering The searches for neutrino magnetic moments are performed by study- ing the recoil electron spectrum of ¯ ν e e - → ¯ ν X e - scattering. Both both diagonal and transition moments are allowed. ¯ νe ¯ νe e - e - + Z 0 W - ¯ νe ¯ νe e - e - g ¯ νe g e A ,g e V 2 + ¯ νe ¯ νe e - e - γ 2 The differential cross-section of this process could be written as a sum of a standard model term and a magnetic moment term [3], ( dσ dT ) SM = G F 2 m e 2π [(g V - g A ) 2 +(g V + g A ) 2 (1 - T E ν ) 2 +(g 2 A - g 2 V ) m e T E 2 ν ], ( dσ dT ) MM = πα 2 μ 2 ν m 2 e ( 1 T - 1 E ν ), with T is the energy of recoil electron, μ ν is the neutrino magnetic mo- ment in units of μ B . When T → 0, ( dσ dT ) SM → constant, while ( dσ dT ) MM → 1/T . In very low recoil energy, magnetic moment term will dominated over standard model term. The value of μ 2 eff we are interested in is ∼10 -10 μ B , which is consis- tent with solar data(before KamLand results), and could be reached by present day laboratory experiment. 1.3 ¯ ν e Spectrum and Recoil Energy Spectrum The spectrum of ¯ ν e at detector site: with a total flux ∼ 6×10 12 cm -2 s -1 This yield a recoil electron spectrum: ν _ e N q.f.=0.25 ν _ e N q.f.=1.0 ν _ e e(SM) ν _ e e(MM) backgroud level In the experiment, we focus at 10 - 100 keV range, in which magnatic moment related event rate is ”decoupled” from standard model ”back- ground” at μ 2 ≈10 -10 μ B . The spectrum will compare with the reactor-on subtract reactor-off spec- trum to search for neutrino magnetic moment bound. The magnetic moment bound provide a limit on neutrino radiative decay constant in the process: ν 1 → ν 2 γ. The neutrino radiative decay constant is related to magnetic moment in this equation [4]: Γ= 1 2π (Δm 2 ) 3 m 3 μ ν 2 1.4 νN coherent scattering The differential cross section of νN coherent scattering could be de- scribed by [5]: ( dσ dt ) SM = G F 2 m N 4π [Z(1 - 4sin 2 θ W ) - N ] 2 [1 - M N T N 2E 2 ν ] ( dσ dt ) MM = πα 2 μ 2 ν m 2 e Z 2 ( 1 T - 1 E ν ) The magnetic moment term enhanced by Z 2 . However, due to the nuclei mass and quenching factor, the recoil energy of nuclei has a very low energy at ∼ 100eV. The figure shown rate of νN scattering with quenching factor 1.0 and 0.25 respectively, as well as ¯ ν e e - scattering and background level of period I at 5 - 8 keV. 2 The Experiment 2.1 Location The Kuo-Sheng nuclear power plant locate at the nothern shore of Taiwan, with two 2.9 GW reactor core. The de- tector is 28 m from first core. Diagram of experiment site and reactor core is shown in figure at right: Experiment Site Primary Containment Building Auxiliary Reactor Pressure Vessel Dry Well Suppression Pool Nuclear Power Plant II : Reactor Building Reactor Core The experiment site is overburden by 10 m of concrete(30 mwe), which effectively shield hadronic component of cosmic ray. 2.2 Shielding Inner Target Volume 100(W) x 80(D) x 75(H) cm3 Copper : 5cm Stainless Steel Frame : 5 cm Lead : 15 cm Veto Plastic Scintillator : 3 cm Boron-loaded Polyethylene : 25 cm The shielding include plastic scintillator as cosmic veto, lead, steel(structure frame), boron loaded polyetheylene and copper. 2.3 Period I Configuration HPGe detector(1 kg) CsI detector (186 kg, Period II) In the period I(June 2001 - April 2002) experiment, the main detector is a 1 kg HPGe detector, along with a 46 kg CsI array detector, which is upgraded to 186 kg in period II. PMT N from dewar 2 Lead Pre - Amplifier HPGe Radon purge plastic bag OFHC Copper 28 cm 70 cm Liquid nitrogen dewar CsI(Tl) NaI(Tl) The HPGe detector is surrounded by a NaI detector and a CsI detector at bottom as anti-Compton detector, and copper as passive shielding. The whole volume is flushed with nitrogen. 3 Data and Event Selections 3.1 Data Period I data: • 4712 hours of reactor-on, 1250 hours of reactor-off data. • Detector mass 1.06 kg. • Energy resolution of 0.4 keV(RMS) at 10 keV. • Detector threshold 5 keV. • Background at 12 - 60 keV is at O(1 cpd), comparable to under- ground Dark Matter experiment. An efficiencied normalization accurated to 0.2% is achieved by: • DAQ book keeping(hardware status, deadtime). • Monitoring of random trigger events. • Stability of 40 K peaks. • Monitoring of 10 keV Ga X-rays peak(decaying with time). 3.2 Event Selections Arrival time of signal on veto scintillator vs. Energy deposit in HPGe: 0 20 40 60 80 100 120 -10 -5 0 5 10 15 20 Rise Time of cosmic veto (μs) Energy deposit in HPGe (keV) only events uni- formly distributed at right had been seleted. Energy deposit in NaI vs. Energy deposit in HPGe: events without de- posit energy in NaI detector had been selected. Pulse shape discrimination: Pulse height vs. Energy: select those events with proper pulse height to pulse area(energy) ratio, also shown in the plot is 66.712 keV double pulses events. 3.3 Efficiency & Uncertainties The event selections give us a total suppression factor of 5% with effe- ciency 94%. The uncertainties is came, mainly, from uncertainties of reactor ¯ ν e spec- trum, which give us a final systematic error of <0.4×10 -20 μ 2 B . The sources of effeciency and uncertainties is summarized as follow: Event selection Suppression Efficiency Raw data 1.0 1.0 Anti-Compton (AC) 0.06 0.99 Cosmic-ray veto(CRV) 0.96 0.95 Pulse shape analysis 0.86 1.0 Combined efficiency 0.05 0.94 Sources Uncertainties σ(κ 2 e )10 -20 μ 2 B DAQ live time ON/OFF <0.2% <0.30 Efficiencies for magnetic scattering <0.2% <0.01 Rates for magnetic scattering 24% 0.23 SM background subtraction 23% 0.03 Combined systematic error ··· <0.4 4 Results 4.1 Neutrino Magnetic Moment The following figure show the reactor-on/off spectrum. The spectrum are calibrated by peaks from Ga x-rays, 73 Ge * and 234 Th, 10 -2 10 -1 1 10 10 2 0 20 40 60 80 100 120 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 10 20 30 40 50 60 The reactor-off spectrum is fitted with a polynomial φ OFF (χ 2 /dof = 80/96), with use as an input to fit the reactor-on spec- trum φ ON with φ OFF + φ SM + κ 2 φ MM [10 -10 μ B ] The best fit value of κ 2 = -0.4 ± 1.3(stat.) ± 0.4(sys.)[10 -10 μ B ] with χ 2 /dof = 48/49 is obtained. Adopting unified approach [6], yield μ ν < 1.3(1.0) × 10 -10 μ B 90(68)%C.L. 2 - σ best fit region of κ 2 is plotted in (b). 4.2 Limits from Other Experiments The limits quoted by PDG [6] is μ ν < 1.5×10 -10 μ B , from the ν e e - scat- tering of SuperK data. Limits from other reactor ¯ ν e e - scattering experiment [6]: • Savannah River(plastic scintillator), μ ν ≈ 2 - 4×10 -10 μ B • Kurtchatoc(fluorocarbon scintillator), μ ν < 2.4×10 -10 μ B • Rovno(Si(Li)), μ ν < 1.9×10 -10 μ B • MUNU(CF 4 ), threshold ∼ 1 MeV. These reactor ¯ ν e e - scattering experiment limits is around 1 - 2×10 -10 μ B . Astrophysics bound is more stringent, at 10 -12 μ B order [6], however those limits is depends on stellar model and interaction model between neutrino and stellar objects. 4.3 Sensitivity Our limits slightly better than previous ¯ ν e e - scattering exper- iment. The most important thing is that our experiment sit- ting at a very low threshold, at which magnetic moment contri- bution is decoupled from stan- dard model ”background”, and thus the uncertainties of ex- pected ¯ ν e e - scattering rate play a less significant role. 0.5 1 1.5 2 2.5 3 3.5 4 4.5 1 10 10 2 10 3 Derive from the bound on neu- trino magnetic moment, the bound on neutrino radiative de- cay is τ ν m ν 3 > 2.8×10 18 eV 3 s which is more stringent than di- rect search. -5 0 5 10 15 20 25 30 35 40 -4 -3.5 -3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 5 νN coherent scattering and ULE-HPGe detector 5.1 ULE-HPGe Calibration Data: Threshold With ULE-HPGe, measurement of νN coherent scattering is pos- sible. Figure at right show 55 Fe spectrum with Ti as back- scattering source on a 5 g ULE- HPGe detector with Threshold ∼ 60 eV . Figure at right show that noise and signal have different PSD, the noise edge is at ∼ 60 eV. However, with such a small detector, integral count rate of νN scattering events is ∼0.05 counts per days. A larger detector is needed. 6 CsI: Period II The purpose of the CsI detector is to study electro-weak parameter, g V , g A , sin 2 (θ W ), at MeV range [7] [8]. The period II CsI array detec- tor consist 93×2 kg CsI crystals with PMT readout at both end, as shown in figure at right: • The energy resolution is 10% at 660 keV. • Position resolution is 2 cm at 660 keV. • Calibration by 137 Cs, 40 K and 208 Tl. 6.1 Background spectrum of Period II CsI After Veto NO cut After all Cuts 221 kg-days OFF-period Standard Model After PSD After events selections and reactor-on/off subtration, the residual event rate is close to expected ¯ ν e e - scattering rate at above 3 MeV. The event selections include: • Cosmic veto cut → one order suppression factor. • Single crystal event and PSD → two order suppression factor. • Reactor-on subtract reactor-off → one order suppression factor. Goal: σ(¯ ν e e - ) accurate to 20%. 7 Summary Period I HPGe: • μ ν analysis : results published Period II: HPGe • addtional 1400/790 hours reactor-ON/OFF data → background and analysis improvement. Period II: CsI(Tl) • Measure electro-weak parameter at MeV range with σ(¯ ν e e - ). • Data analysing. Period III and ULE-HPGe: • continue with HPGe and CsI(Tl) configuration at period II. • explore potentials on ¯ ν e N coherent scattering with ULE-HPGe. • study quenching factor and pulse shape with neutron beam exper- iment. • study on-site ULE-HPGe background. References [1] Home Page http://hepmail.phys.sinica.edu.tw/˜texono, also hep-ex/0307001. [2] H. B. Li, et al., TEXONO Coll., Phys. Rev. Lett. 90, 131802 (2003). [3] P. Vogel and J. Engel, Phys. Rev. D 38, 3378(1989). [4] G. G. Raffelt, Phys. Rev. D 39, 2066(1989). [5] A. C. Dodd, et al., Phys. Lett. B 266 434(1991). [6] See the respective sections in Review of Particle Physics, Phys. Rev. D 66 (2002). [7] H. B. Li, et al., Nucl. Instrum. Methods A 459 93(2001). [8] Y. Liu, et al., Nucl. Instrum. Methods A 482 125(2002).
Transcript of Low Energy Neutrino Physics at the Kuo-Sheng Reactor ...
Low Energy Neutrino Physics at the Kuo-Sheng Reactor LaboratoryHau-Bin Li (on behalf of the TEXONO Collaboration)
Institute of Physics, Academia Sinica, Taipei 115, Taiwan
1 Introduction
1.1 TEXONO collaboration• TEXONO established in 1996.
• First big collaboration between China and Taiwan.
• Magnetic moment results published.
• Period II → CsI detector data.
• Next → νN scattering with ULE-HPGe detector.
1.2 ν̄ee−
→ ν̄Xe− Scattering
The searches for neutrino magnetic moments are performed by study-ing the recoil electron spectrum of ν̄ee
−→ ν̄Xe− scattering. Both both
diagonal and transition moments are allowed.
ν̄e ν̄e
e− e−
+Z0W−
ν̄e ν̄e
e− e−
gν̄e
geA, ge
V
2
+
ν̄e ν̄e
e− e−
γ
2
The differential cross-section of this process could be written as a sum ofa standard model term and a magnetic moment term [3],
(dσ
dT)SM =
GF2me
2π[(gV − gA)2 + (gV + gA)2(1 −
T
Eν
)2 + (g2
A − g2
V )meT
E2ν
],
(dσ
dT)MM =
πα2µ2ν
m2e
(1
T−
1
Eν
),
with T is the energy of recoil electron, µν is the neutrino magnetic mo-ment in units of µB .When T → 0, ( dσ
dT)SM → constant, while ( dσ
dT)MM → 1/T .
In very low recoil energy, magnetic moment term will dominated overstandard model term.The value of µ2
eff we are interested in is ∼10−10µB , which is consis-tent with solar data(before KamLand results), and could be reached bypresent day laboratory experiment.
1.3 ν̄e
Spectrum and Recoil Energy SpectrumThe spectrum of ν̄e at detector site:
with a total flux ∼ 6×1012 cm−2s−1
This yield a recoil electron spectrum:
ν_
eN q.f.=0.25
ν_
eN q.f.=1.0
ν_
ee(SM)
ν_
ee(MM) backgroud level
In the experiment, we focus at 10 - 100 keV range, in which magnaticmoment related event rate is ”decoupled” from standard model ”back-ground” at µ2
≈10−10µB .The spectrum will compare with the reactor-on subtract reactor-off spec-trum to search for neutrino magnetic moment bound.The magnetic moment bound provide a limit on neutrino radiative decayconstant in the process:
ν1 → ν2γ.
The neutrino radiative decay constant is related to magnetic moment inthis equation [4]:
Γ=1
2π
(∆m2)3
m3µν
2
1.4 νN coherent scatteringThe differential cross section of νN coherent scattering could be de-scribed by [5]:
(dσ
dt)SM =
GF2mN
4π[Z(1 − 4sin2θW ) − N ]2[1 −
MNTN
2E2ν
]
(dσ
dt)MM =
πα2µ2ν
m2e
Z2(1
T−
1
Eν
)
The magnetic moment term enhanced by Z2. However, due to the nucleimass and quenching factor, the recoil energy of nuclei has a very lowenergy at ∼ 100eV.The figure shown rate of νN scattering with quenching factor 1.0 and 0.25respectively, as well as ν̄ee
− scattering and background level of period Iat 5 - 8 keV.
2 The Experiment
2.1 Location
The Kuo-Sheng nuclear power plantlocate at the nothern shore of Taiwan,with two 2.9 GW reactor core. The de-tector is 28 m from first core. Diagramof experiment site and reactor core isshown in figure at right:
ExperimentSite
PrimaryContainment
BuildingAuxiliary
ReactorPressureVessel
DryWell
SuppressionPool
Nuclear Power Plant II : Reactor Building
ReactorCore
The experiment site is overburden by 10 m of concrete(30 mwe), whicheffectively shield hadronic component of cosmic ray.
2.2 Shielding
Inner Target Volume 100(W) x 80(D) x 75(H) cm 3
Copper : 5cm
Stainless Steel Frame : 5 cm
Lead : 15 cm
Veto Plastic Scintillator : 3 cm
Boron-loaded Polyethylene : 25 cm
The shielding include plastic scintillator as cosmic veto, lead,steel(structure frame), boron loaded polyetheylene and copper.
2.3 Period I Configuration
HPGe detector(1 kg) CsI detector (186 kg, Period II)
In the period I(June 2001 - April 2002) experiment, the main detector isa 1 kg HPGe detector, along with a 46 kg CsI array detector, which isupgraded to 186 kg in period II.
PMT
N from dewar2
Lead
Pre − Amplifier
HPGe
Radon purge plastic bag
OFHC Copper
28 cm
70 cm
Liquid nitrogendewar
CsI(Tl)
NaI(Tl)
The HPGe detector is surrounded by a NaI detector and a CsI detector atbottom as anti-Compton detector, and copper as passive shielding. Thewhole volume is flushed with nitrogen.
3 Data and Event Selections
3.1 DataPeriod I data:
• 4712 hours of reactor-on, 1250 hours of reactor-off data.
• Detector mass 1.06 kg.
• Energy resolution of 0.4 keV(RMS) at 10 keV.
• Detector threshold 5 keV.
• Background at 12 - 60 keV is at O(1 cpd), comparable to under-ground Dark Matter experiment.
An efficiencied normalization accurated to 0.2% is achieved by:
• DAQ book keeping(hardware status, deadtime).
• Monitoring of random trigger events.
• Stability of 40K peaks.
• Monitoring of 10 keV Ga X-rays peak(decaying with time).
3.2 Event SelectionsArrival time of signal on veto scintillator vs. Energy deposit in HPGe:
0
20
40
60
80
100
120
-10 -5 0 5 10 15 20
Rise Time of cosmic veto (µs)
Ene
rgy
depo
sit i
n H
PG
e (k
eV)
only events uni-formly distributedat right had beenseleted.
Energy deposit in NaI vs. Energy deposit in HPGe:
events without de-posit energy in NaIdetector had beenselected.
Pulse shape discrimination: Pulse height vs. Energy:
select those eventswith proper pulseheight to pulsearea(energy) ratio,also shown inthe plot is 66.712keV double pulsesevents.
3.3 Efficiency & UncertaintiesThe event selections give us a total suppression factor of 5% with effe-ciency 94%.The uncertainties is came, mainly, from uncertainties of reactor ν̄e spec-trum, which give us a final systematic error of <0.4×10−20µ2
B .The sources of effeciency and uncertainties is summarized as follow:
Event selection Suppression EfficiencyRaw data 1.0 1.0Anti-Compton (AC) 0.06 0.99Cosmic-ray veto(CRV) 0.96 0.95Pulse shape analysis 0.86 1.0Combined efficiency 0.05 0.94
Sources Uncertainties σ(κ2e)10−20µ2
B
DAQ live time ON/OFF <0.2% <0.30Efficiencies for magnetic scattering <0.2% <0.01Rates for magnetic scattering 24% 0.23SM background subtraction 23% 0.03Combined systematic error · · · <0.4
4 Results
4.1 Neutrino Magnetic MomentThe following figure show the reactor-on/off spectrum. The spectrumare calibrated by peaks from Ga x-rays, 73Ge∗ and 234Th,
10-2
10-1
1
10
10 2
0 20 40 60 80 100 120
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
10 20 30 40 50 60
The reactor-off spectrum is fitted with a polynomialφOFF (χ2/dof = 80/96), with use as an input to fit the reactor-on spec-trum φON with
φOFF + φSM + κ2φMM [10−10µB ]
The best fit value of
κ2 = −0.4± 1.3(stat.) ± 0.4(sys.)[10−10µB ]
with χ2/dof = 48/49 is obtained.Adopting unified approach [6], yield
µν < 1.3(1.0)× 10−10µB 90(68)%C.L.
2 − σ best fit region of κ2 is plotted in (b).
4.2 Limits from Other ExperimentsThe limits quoted by PDG [6] is µν < 1.5×10−10µB , from the νee
− scat-tering of SuperK data.Limits from other reactor ν̄ee
− scattering experiment [6]:
• Savannah River(plastic scintillator), µν ≈ 2 − 4×10−10µB
• Kurtchatoc(fluorocarbon scintillator), µν < 2.4×10−10µB
• Rovno(Si(Li)), µν < 1.9×10−10µB
• MUNU(CF4), threshold ∼ 1 MeV.
These reactor ν̄ee− scattering experiment limits is around
1 − 2×10−10µB .Astrophysics bound is more stringent, at 10−12µB order [6], howeverthose limits is depends on stellar model and interaction model betweenneutrino and stellar objects.
4.3 SensitivityOur limits slightly better thanprevious ν̄ee
− scattering exper-iment. The most importantthing is that our experiment sit-ting at a very low threshold, atwhich magnetic moment contri-bution is decoupled from stan-dard model ”background”, andthus the uncertainties of ex-pected ν̄ee
− scattering rate playa less significant role.
0.5
1
1.5
2
2.5
3
3.5
4
4.5
1 10 102
103
Derive from the bound on neu-trino magnetic moment, thebound on neutrino radiative de-cay is
τνmν3 > 2.8×1018eV3s
which is more stringent than di-rect search.
-5
0
5
10
15
20
25
30
35
40
-4 -3.5 -3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1
5 νN coherent scattering and ULE-HPGedetector
5.1 ULE-HPGe Calibration Data: Threshold
With ULE-HPGe, measurementof νN coherent scattering is pos-sible. Figure at right show55Fe spectrum with Ti as back-scattering source on a 5 g ULE-HPGe detector with Threshold∼ 60 eV .
Figure at right show that noiseand signal have different PSD,the noise edge is at ∼ 60 eV.
However, with such a small detector, integral count rate of νN scatteringevents is ∼0.05 counts per days. A larger detector is needed.
6 CsI: Period IIThe purpose of the CsI detector is to study electro-weak parameter, gV,gA, sin2(θW), at MeV range [7] [8].
The period II CsI array detec-tor consist 93×2 kg CsI crystalswith PMT readout at both end,as shown in figure at right:
• The energy resolution is10% at 660 keV.
• Position resolution is 2 cmat 660 keV.
• Calibration by 137Cs, 40Kand 208Tl.
6.1 Background spectrum of Period II CsI
After Veto
NO cutAfter all Cuts
221 kg-days OFF-period
Standard Model After PSD
After events selections and reactor-on/off subtration, the residual eventrate is close to expected ν̄ee
− scattering rate at above 3 MeV.The event selections include:
• Cosmic veto cut → one order suppression factor.
• Single crystal event and PSD → two order suppression factor.
• Reactor-on subtract reactor-off → one order suppression factor.
Goal: σ(ν̄ee−) accurate to 20%.
7 Summary
Period I HPGe:
• µν analysis : results published
Period II: HPGe
• addtional 1400/790 hours reactor-ON/OFF data→ background and analysis improvement.
Period II: CsI(Tl)
• Measure electro-weak parameter at MeV range with σ(ν̄ee−).
• Data analysing.
Period III and ULE-HPGe:
• continue with HPGe and CsI(Tl) configuration at period II.
• explore potentials on ν̄eN coherent scattering with ULE-HPGe.
• study quenching factor and pulse shape with neutron beam exper-iment.
• study on-site ULE-HPGe background.
References[1] Home Page http://hepmail.phys.sinica.edu.tw/˜texono,
also hep-ex/0307001.
[2] H. B. Li, et al., TEXONO Coll., Phys. Rev. Lett. 90, 131802 (2003).
[3] P. Vogel and J. Engel, Phys. Rev. D 38, 3378(1989).
[4] G. G. Raffelt, Phys. Rev. D 39, 2066(1989).
[5] A. C. Dodd, et al., Phys. Lett. B 266 434(1991).
[6] See the respective sections in Review of Particle Physics, Phys. Rev. D 66 (2002).
[7] H. B. Li, et al., Nucl. Instrum. Methods A 459 93(2001).
[8] Y. Liu, et al., Nucl. Instrum. Methods A 482 125(2002).
