SLIDE 1 1 / 14
Search of axions from a nuclear power reactor with a high-purity germanium detector
Hsi-Ming Chang Department of Physics, National Taiwan University Institute of Physics, Academia Sinica TEXONO Collaboration TAUP 2007
SLIDE 2
Outline
Outline
2 / 14
■
Introduction to Axions
■
Axion Production & Detection
■
Data analysis
■
Physics Results This work is published in PRD 75, 052004 (2007), a by-product of Taiwan EXperiment On NeutrinO.
SLIDE 3
Introduction to Axion
Introduction to Axion
3 / 14
■
Strong CP Problem: Neutron EDM < 10−25 e cm, Why QCD does not seem to break the CP-symmetry?
■
PQWW Axion (ma 100 keV):
◆
A hypothetical particle to solve stong CP problem.
◆
Excluded after extensive searches.
■
Invisible Axion:
◆
Evade previous experimental searches.
◆
Mass window 10−6 ma 10−2 (eV), from cosmological and astrophysical arguments.
◆
Popular models: DFSZ, KSVZ.
SLIDE 4
Reactor as Source
Axion Production & Detection Reactor as Source Branching Ratio Γa/Γγ Complications Reactor Building Detector & Shielding Axion Detection Event Rate Formula
4 / 14
■ Axions could be emitted via magnetic transition. ■ Inspired by F. T. Avignone III et al., PRD37, 618 (1988). ⇒ Radioactive 65Zn source & HPGe detector. ■ Reactor is the most powerful radioactive source we can control!
◆
Slow neuton capture: n + (Z, A) → (Z, A + 1) + γ (or axion).
◆
Nuclear de-excitation: (Z, A)∗ → (Z, A) + γ (or axion). ■ The photon fluxes φγ at detector: Energy Mode φγ (keV) (1010 cm−1s−1) np→dγ 2230 Isovector M1 22.1
7Li∗
478 M1 24.7
91Y∗
555 M4 2.10
97Nb∗
743 M4 4.81
135Xe∗
526 M4 0.85
137Ba∗
662 M4 0.37 ■ Axion flux is φa = φγ
Γa Γγ .
■ Solar axion flux ∼ 1012( ma
eV )2 cm−2sec−1, with average energy ∼4 keV.
SLIDE 5
Reactor as Source
Axion Production & Detection Reactor as Source Branching Ratio Γa/Γγ Complications Reactor Building Detector & Shielding Axion Detection Event Rate Formula
4 / 14
■ Axions could be emitted via magnetic transition. ■ Inspired by F. T. Avignone III et al., PRD37, 618 (1988). ⇒ Radioactive 65Zn source & HPGe detector. ■ Reactor is the most powerful radioactive source we can control!
◆
Slow neuton capture: n + (Z, A) → (Z, A + 1) + γ (or axion).
◆
Nuclear de-excitation: (Z, A)∗ → (Z, A) + γ (or axion). ■ The photon fluxes φγ at detector: Energy Mode φγ (keV) (1010 cm−1s−1) np→dγ 2230 Isovector M1 22.1
7Li∗
478 M1 24.7
91Y∗
555 M4 2.10
97Nb∗
743 M4 4.81
135Xe∗
526 M4 0.85
137Ba∗
662 M4 0.37 ■ Axion flux is φa = φγ
Γa Γγ .
■ Solar axion flux ∼ 1012( ma
eV )2 cm−2sec−1, with average energy ∼4 keV.
Kinematics Constraint!
SLIDE 6
Branching Ratio Γa/Γγ Complications
Axion Production & Detection Reactor as Source Branching Ratio Γa/Γγ Complications Reactor Building Detector & Shielding Axion Detection Event Rate Formula
5 / 14
The axion-to-photon branching ratio for M1 transition is: Γa Γγ = 1 2πα 1 1 + δ2 (pa ǫa )3 „ g0
aNNβ + g1 aNN
(µ0 − 1
2)β + (µ1 − η)
«2 . δ: E2/M1 mixing ratio ≈ 0. µ0 (µ1): Isoscalar (isovector) magnetic moment = 0.88 (4.71). η, β: Matrix elements from nuclear physics. ■ Numerical calculations of η and β are needed. ■ Even if η and β are known, two free parameters g0
aNN and g1 aNN still
remain. How to circumvent the complications? ■ It happens that np→dγ is an isovetor M1 transition: „Γa Γγ «
np
≡ Γa Γγ (np → dγ) ≈ 1 2πα(pa ǫa )3(g1
aNN
µ1 )2 ∝ (g1
aNN)2 .
■ In analysis, gaNN can be parametrized as a function of ma with axion models.
SLIDE 7 Reactor Building
Axion Production & Detection Reactor as Source Branching Ratio Γa/Γγ Complications Reactor Building Detector & Shielding Axion Detection Event Rate Formula
6 / 14
■
Power: 2.9 GW.
■
νe flux: 6 × 1012 cm−2·s−1 .
■
30 mwe overburden.
■
Data Size:
- ON: 459.0 days.
- OFF: 96.3 days.
in two ON/OFF periods.
SLIDE 8
Detector & Shielding
Axion Production & Detection Reactor as Source Branching Ratio Γa/Γγ Complications Reactor Building Detector & Shielding Axion Detection Event Rate Formula
7 / 14
HPGe detector:
■
Mass: 1 kg.
■
Threshold: 5 keV.
■
CsI and NaI: anti-Compton system.
■
28m from reactor core.
Outer Shielding: 1. Plastic scintillator: cosmic-ray veto. 2. Lead: block γ’s from outside. 3. Stainless steel: support the structure. 4. B-loaded polyethylene: neutron capturer. 5. OFHC copper: reduce the γ’s from lead or polyethylene.
SLIDE 9 Axion Detection
Axion Production & Detection Reactor as Source Branching Ratio Γa/Γγ Complications Reactor Building Detector & Shielding Axion Detection Event Rate Formula
8 / 14
a Ge γ Primakoff conversion (gaγγ) e a γ e a + γ Compton conversion (gaee)
500 1000 1500 2000
ma keV
0.5 1 1.5 2
Σ 1022 cm
2
gaΓΓ 1 GeV1 gaee 1 Primakoff 104 Compton
■
σPri = g2
aγγ · f(ma, ǫa)
⇒ sensitive at low ma
■
σCom = g2
aee · f(ma, ǫa)
⇒ sensitive at high ma (Here ǫa = 2230 keV)
SLIDE 10 Event Rate Formula
Axion Production & Detection Reactor as Source Branching Ratio Γa/Γγ Complications Reactor Building Detector & Shielding Axion Detection Event Rate Formula
9 / 14
The event rate in unit of day−1kg−1 is R = σ
Γa Γγ · Pdecay · Pmatter
Pdecay: Survival probability without decay. Pmatter: Survival probability without interaction. N: # of Ge atoms in kilogram target. ǫ: Efficiency of full-energy deposition at detector. R = R(ma , gaγγ/aee , gaNN) . Invoking the widely-used DFSZ model (gaNN ∝ ma) to reduce free parameter: R ∝ g2
aγγ/aeem2 a .
SLIDE 11 Energy Spectra
Data Analysis Energy Spectra ON-OFF Residual Results
10 / 14
10
10
1 10 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000
keV counts/day-kg-keV (a)
226Ra 214Pb 208Tl 228Ac 40K 208Tl
10
10
1 10 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000
keV counts/day-kg-keV (b)
ON OFF
SLIDE 12 ON-OFF Residual
Data Analysis Energy Spectra ON-OFF Residual Results
11 / 14
0.2 0.4 0.6 460 480 500
Energy (keV) Events / (day-kg-keV)
0.2 0.4 0.6 520 540
Energy (keV)
0.2 0.4 0.6 540 560
Energy (keV)
0.2 0.4 0.6 640 660 680
Energy (keV) Events / (day-kg-keV)
0.2 0.4 0.6 720 740 760
Energy (keV)
0.02 0.04 0.06 0.08 2200 2250
Energy (keV)
—: overlaid best-fit Gaussians
SLIDE 13 Results
Data Analysis Energy Spectra ON-OFF Residual Results
12 / 14
Statistical results:
Energy Period A Period B (keV) (day−1kg−1) (day−1kg−1) 478
0.14±0.41 526 0.26±0.67 0.38±0.16 555
662
743 0.14±0.55 0.22±0.37 2230
Energy P-A&P-B Combined Upper Bound (keV) (day−1kg−1) (day−1kg−1) 478
0.49 526 0.37±0.15 0.62 555
0.05 662
0.46 743 0.19±0.31 0.69 2230
0.02
7Li 91Y 97Nb 135Xe 137Ba
npdΓ
PA & PB combined
0.5 1
countsday1kg1
Systematics Uncertainty: < 20%, dominated by evaluation of φγ from np→dγ.
SLIDE 14 ma − ga Space
Physics Results ma − ga Space Summary
13 / 14
108 106 104 102 100 102 104 106 108
ma eV
1016 1014 1012 1010 108 106 104 102 100 102
gaΓΓ GeV1
HW HB Stars Microwave Cavity
CAST
SolarGermanium Laser Experiments Telescope Beam Dump R Zn TEXONO PQWW KSVZ DFSZ Int Kine Decay
108 106 104 102 100 102 104 106 108
ma eV
1016 1014 1012 1010 108 106 104 102 100 102
gaΓΓ GeV1
108 106 104 102 100 102 104 106 108
ma eV
1014 1012 1010 108 106 104 102 100
gaee
HW Red Giant Positronium Decay Beam Dump
Macro scopic Force
Zn TEXONO R DFSZ PQWW Kine Decay Int
108 106 104 102 100 102 104 106 108
ma eV
1014 1012 1010 108 106 104 102 100
gaee
SLIDE 15
Summary
Physics Results ma − ga Space Summary
14 / 14
■
There are 9 reactor axion papers among literature, and the latest one was in 1995. They all focused on standard (PQWW) axion.
■
We have new and more stringent results for general axions. PQWW, DFSZ and KSVZ models are excluded for ma ≈ 104eV - 106eV.
■
This approach defines exclusion boundary for ma ≈ 103 - 106eV among direct experiments.
■
The model-independent (not invoking the DFSZ gaNN − ma relation) upper bounds:
8 > < > : g2
aγγ ·
„ Γa Γγ «
np
< 5.9 × 10−17 GeV−2 gaγγ · g1
aNN < 7.7 × 10−9 GeV−1 ,
8 > < > : g2
aee ·
„ Γa Γγ «
np
< 1.7 × 10−20 gaee · g1
aNN < 1.3 × 10−10 .
