guillaumeplante columbiauniversity …zentner/guillaume_plante_files/... · 2011. 11. 14. · pmti...

Status of the XENON100 Dark Matter Search

Guillaume Plante

Columbia University

on behalf of the XENON Collaboration

Exploring Low-Mass Dark Matter Candidates - PITT PACC - November 14-16, 2011

XENON100 Collaboration

Guillaume Plante Exploring Low-Mass Dark Matter Candidates - PITT PACC - November 14-16, 2011 2 / 37

Columbia Rice UCLA Zurich Coimbra LNGS SJTU

Mainz Bologna MPIK NIKHEF Subatech Munster WIS

Direct Detection


Heat

ScintillationIonization

ZEPLIN II, IIIXENONLUXWARPArDMDarkSide

CRESST-ISIMPLEPICASSOCOUPP

IGEXGEDEONGENIUSGERDACoGeNT

ZEPLIN IDEAPCLEANXMASSDAMA/LIBRAKIMS

CDMS I, IISuperCDMSEDELWEISS I, II

CRESST-IIROSEBUD

Why Xenon?


• Large mass number A (∼131), expecthigh rate for SI interactions (σ ∼ A2)if energy threshold for nuclear recoilsis low

• ∼50% odd isotopes (129Xe, 131Xe)for SD interactions

• No long-lived radioisotopes, Kr can bereduced to ppt levels

• High stopping power (Z = 54, ρ =3g cm−3), active volume is self shield-ing

• Efficient scintillator (∼80% light yieldof NaI), fast response

• Scalable to large target masses

• Nuclear recoil discrimination with si-multaneous measurement of scintilla-tion and ionization

10−5

10−4

10−3

Tot

alR

ate

[evt/

kg/

day

]

0 10 20 30 40 50

Eth - Nuclear Recoil Energy [keV]

Si (A = 28)

Ar (A = 40)

Ge (A = 73)

Xe (A = 131)

σχN = 1 × 10−44 cm2

Principle


Eionization

excitation

Xe++ e−

+Xe

Xe+

2

+e−

Xe∗∗+XeXe∗

+Xe

Xe∗2

2Xe

178 nmsinglet (3 ns)

2Xe

178 nmtriplet (27 ns)

• Bottom PMT array below cathode, fully immersed in LXeto efficiently detect scintillation signal (S1).

• Top PMTs in GXe to detect the proportional signal (S2).

• Distribution of the S2 signal on top PMTs gives xy coordi-nates (∆r < 3mm) while drift time measurement providesz coordinate (∆z < 300µm) of the event.

• Ratio of ionization and scintillation (S2/S1) allows dis-crimination between electron and nuclear recoils.

XENON100: Design


PMT HV and Signal Lines ❛❛❛❛❛❛❛❛❛❛❛❛

❆❆❆❆Anode Feedthrough ❝

❝❝❝❝❝❝❝❝❝❝

Tube to

Cooling

Tower

Top Veto

PMTs

Top Array

PMTs✭✭✭✭✭✭

PTFE

Panels✭✭✭✭✭✭✭✭✭

Side Veto

PTFE Lining✱✱✱

Lower Side

Veto PMTs✏✏✏✏✏

Bottom Array PMTs ✑✑✑✑✑✑✑✑✑✑✑✑

Double-Wall

Cryostat

✔✔

✔✔

✔✔

✔✔

✔✔

✔✔

✔✔

✄✄✄✄✄✄✄✄✄✄✄✄✄✄✄✄✄✄✄

Upper Side

Veto PMTs��

��

Bell✭✭✭✭✭✭✭✭

Top Mesh✘✘✘✘✘✘✘✘Anode

Bottom Mesh❳❳❳❳❳❳❳

Field

Shaping

Rings

✭✭✭✭✭✭✭✭

Cathode❤❤❤❤❤❤❤❤❤

Bottom Veto PMTs

❧❧

❧❧

❧❧

• Goal was to build a detector with a ×10 increasein fiducial mass and a ×100 reduction in back-ground compared to XENON10.

• All detector materials and components werescreened in a dedicated low-background count-ing facility

• 161 kg LXe total mass consisting of a 62 kg tar-get surrounded by a 99 kg active veto. 15 cmradius, 30 cm drift length active volume.

• TPC inner volume defined by 24 interlockingPTFE panels. Drift field uniformity ensured by40 double field shaping wires, inside and outsidethe panels.

• Cathode at -16 kV, drift field of 0.533 kV/cm.Anode at 4.5 kV, proportional scintillation re-gion with field ∼12 kV/cm. Custom-made lowradioactivity HV feedthroughs.

• Aprile et al., arXiv:1107.2155

XENON100: PMT Arrays


• 242 low activity Hamamatsu R8520-06-Al 1” squarePMTs,

• 98 tubes on top (QE ˜23%), in concentric circles andenclosed in a PTFE structure,

• 80 high QE (˜33%) tubes on bottom, on a rectangulargrid to maximize photocathode coverage,

• 64 tubes in the LXe veto, in two rings, alternatinginward and down (up) to allow them to view the top,bottom and sides of the veto volume.

XENON100: Shield


• XENON100 installed in a passive shield to suppress external backgrounds:

• 20 cm of H2O to moderate neutrons produced in the cavern rock

• 20 cm Pb (inner 5 cm with low radioactivity Pb) to stop gamma rays

• 20 cm polyethylene to moderate neutrons produced in the Pb

• 5 cm Cu to stop gamma rays from the polyethylene

• Shield cavity continuously purged with N2 to keep 222Rn level < 1Bq/m3.

XENON100: Gran Sasso


✑✑✑

❤❤❤❤❤❤❤❤❤❤❤❤

Outside Buildings

Tunnel

Lab

XENON100: Typical Low Energy Event


s]µDrift Time [0 20 40 60 80 100 120 140 160 180

Am

plitu

de [V

]

0.0

0.5

1.0

1.5

Veto

Am

plitu

de [V

]

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

TPC

-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6

0.00

0.05

0.10

0.15S1: 14.1 pe

170 171 172 173 174 175 176 177

0.0

0.1

0.2

0.3 S2: 926.1 pe

XENON100: Typical Low Energy Event


1

2

3

4

5

67

8 910

11

12

13

14

15

16

17

18

19

20

2122232425

26

27

28

29

30

31

32

33

34

3536 37 38

39

40

41

42

43

44

45

46

47484950

51

52

53

54

55

56

57

5859 60 61

62

63

64

65

66

67

68697071

72

73

74

75

76

7778 79

80

81

82

83

848586

87

88

89

9091

92

93

9495

96

97 98179

180

181

182

183

184

185186 187 188

189

190

191

192

193

194

195

196

197

198

199

200

201202203204

205

206

207

208

209

210

99 100 101 102

103 104 105 106 107 108 109

110 111 112 113 114 115 116 117 118

119 120 121 122 123 124 125 126 127 128

129 130 131 132 133 134 135 136 137 138

139 140 141 142 143 144 145 146 147 148

149 150 151 152 153 154 155 156 157 158

159 160 161 162 163 164 165 166 167

168 169 170 171 172 173 174

175 176 177 178

211

212

213

214

215

216

217218 219 220

221

222

223

224

225

226

227

228

229

230

231

232

233234235236

237

238

239

240

241

242

[pe]

PM

Ti

S2

0

20

40

60

80

100

120

140

XENON100: ER Background


Energy [keV]0 500 1000 1500 2000 2500 3000

]-1

keV

-1 d

ay-1

Rat

e [e

vent

s kg

-410

-310

-210

data (Fall 2009, no veto cut)

MC (total)

Kr, 120 ppt)85MC (

Bq/kg)µRn, 21 222MC (

y)2110×=2.111/2

, Tββ νXe 2136MC (

Pb214

Ac228

Bi214

Bi214

Cs137 Co60

Co60

K40Tl208

Tl208

Mn54

Energy [keVee]1 2 3 4 5 6 7 8 910 20 30 40 50 100

]-1

d-1

kg

-1R

ate

[eve

nts

keV

ee

-310

-210

-110

1

10

210

CRESST

DAMA/LIBRA

CDMSCoGeNT

IGEX

XENON10

XENON100(target volume)

XENON100(fiducial volume)

XENON100 keVee scale needed below 9 keV

• Measured and predicted ER background without veto cut and for a 30 kg fiducial volume.

• Measured ER background (9.6× 10−3 events keV−1 kg−1d−1) is in good agreement with MonteCarlo predictions (no tuning). Background from materials is dominated by PMTs.

• ER background level in fiducial volume (< 6× 10−3 events keV−1 kg−1d−1, with vetocoincidence cut) is ×100 lower than XENON10 (0.6 events keV−1 kg−1d−1).

• Details in Aprile et al., Phys. Rev. D 83, 082001, 2011

XENON100: Calibration


• Electronic recoil band calibration per-formed with high energy gammasfrom 60Co (1.17 Mev, 1.33 MeV).

• Background in the energy region ofinterest is due to low energy Comp-ton scatters from high energy gammarays or β decays.

• Nuclear recoil band calibration per-formed with 3.7 MBq (220 n/s)AmBe neutron source.

• Since WIMPs are expected to elasti-cally scatter off of nuclei understand-ing the behavior of single elastic nu-clear recoils in Xe is essential.

]nr

Nuclear Recoil Equivalent Energy [keV0 5 10 15 20 25 30 35 40

(S2/

S1)

10lo

g

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.00 5 10 15 20 25 30 35 40

(S2/

S1)

10lo

g

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

60Co

AmBe

Nuclear Recoil Equivalent Energy


• Nuclear recoil equivalent energy Enr isobtained from the S1 signal

Enr =S1

Ly,er

1

Leff(Enr)

Ser

Snr

• Ly,er, light yield of electron recoilsfrom 122 keV γ rays

• Ser, Snr, scintillation light quenchingdue to drift field

• Relative scintillation efficiency Leff

Leff (Enr) =Ly,nr (Enr)

Ly,er (Eee = 122 keV) Nuclear Recoil Energy [keV]1 10 100

eff

L

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

Sorensen 2009

Lebedenko 2009

Arneodo 2000

Akimov 2002

Aprile 2005

Aprile 2009

Bernabei 2001

Chepel 2006

Manzur 2010

• The uncertainty in Leff at low energies is the largest systematic uncertainty in the reported resultsfrom LXe WIMP searches at low WIMP masses.

• Recent measurement performed at Columbia University, lowest energy measured 3 keV.

New Measurement of Leff : Detector


• Built a new special purpose LXe detec-tor with maximized scintillation light de-tection efficiency

• Cubic sensitive volume with six 2.5 x 2.5 cm Hama-matsu R8520-406 SEL High QE PMTs

• Calibration with 122 keV γ rays from a 57Cosource gives a light yield of Ly = 24.14 ±0.09(stat)± 0.44(sys) pe/keVee with a resolution(σ/E) of 5%

• Very high light yield, enables a measurement witha low energy threshold

New Measurement of Leff : Setup


LXe

d

2H(d, n)3Heϕ = π

2

EJ301

EJ301

n

θ

θ

• Record fixed-angle elastic scatters of monoenergeticneutrons tagged by organic liquid scintillators with n/γdiscrimination

• Use 2H(d, n)3He 2.5 MeV neutrons from a compactsealed-tube neutron generator

• Recoil energy is fixed by kinematics

Er ≈ 2EnmnMXe

(mn +MXe)2(1− cos θ)

Time of Flight [ns]-50 0 50 100 150

]-1

d-1

Rat

e [e

vent

s ns

0

50

100

150

200

250

300

n

γ o = 34.5θ 1.0 keV±6.5

Accidentals Accidentals

Scintillation Signal [pe]0 10 20 30 40 50 60 70

]-1

d-1

Rat

e [e

vent

s pe

0

20

40

60

80

100o = 34.5θ

1.0 keV±6.5

New Measurement of Leff : Result


Nuclear Recoil Energy [keV]1 10 100

eff

L

0.00

0.05

0.10

0.15

0.20

0.25

0.30Sorensen 2009Lebedenko 2009Horn 2011 (FSR)Horn 2011 (SSR)Aprile 2005Aprile 2009Chepel 2006

Manzur 2010Plante 2011

• Lowest energy (3 keV) and most precise Leff direct measurement achieved to date.

• For details see Plante et al., Phys. Rev. C 84, 045805, 2011

XENON100: Data Taking 2009/2010


Date [mm/dd]10/27 11/26 12/26 01/25 02/24 03/26 04/25 05/25

Live

Day

s

0

20

40

60

80

100

120Science Data (run_08)

Science Data (run_07) Co)60Calibration (Calibration (AmBe)

• Results from 11.2 days unblinded data in Aprile et al., Phys. Rev. Lett. 105, 131302, (2010).

• Data taken in the first half of 2010, blinded region of interest, 100.9 live days

• Results from the blind 100.9 days in Aprile et al., Phys. Rev. Lett. 107, 131302, 2011

XENON100: Data Selection


Basic quality cuts

Designed to remove noisy events, events with un-physical parameters. Very high acceptance.

• S1 coincidence cut

• S2 threshold cut

• S2 saturation cut

• Signal/Noise cut

Fiducial volume cut

Because of the high stopping power of LXe, fidu-cialization is an extremely effective way of reduc-ing background. Fiducial volume chosen:

• 48 kg super-ellipsoid

Scatter cuts

Designed to remove events with multiple inter-actions (multiple S2s), with delayed coincidences(multiple S1s) or misidentified S1s.

• S1 single peak cut

• S2 single peak cut

• Veto cut

Advanced cuts

Designed to remove events which fail consistencychecks, e.g. mismatch in positions from differentalgorithms, or with S1 PMT patterns not consis-tent with their position in the TPC

• S1 PMT pattern cut

• S2 width consistency cut

• Position reconstruction cut

XENON100: Leff Parametrization


Energy [keVnr]1 2 3 4 5 6 7 8910 20 30 40 100

Leff

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

Energy [keVnr]1 2 3 4 5 6 7 8910 20 30 40 100

Leff

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35 Arneodo 2000Bernabei 2001Akimov 2002Aprile 2005Chepel 2006Aprile 2009Manzur 2010Plante 2011

• Leff parametrization is an average of the direct fixed angle neutron scattering Leff measurements

• New measurement of Leff at low energies added to the data used to obtain the parametrization

XENON100: Nuclear Recoil Acceptance


• Keep acceptance high (think discovery)

• Data quality cuts acceptance estimatedfrom AmBe and 60Co calibration data, MCsimulations, and ERs outside the WIMPsearch energy range

• Decided a priori to use Profile Likelihoodapproach and test both background onlyand signal+background hypotheses

• No S2/S1 rejection cut in the Profile Like-lihood approach

• Define a benchmark WIMP region for a par-allel cuts-based analysis

• 99.75% ER rejection line

• S2 threshold

• NR band lower 3-σ

Energy [keVnr]10 20 30 40 50

Acc

epta

nce

0

0.2

0.4

0.6

0.8

1

S1 [PE]5 10 15 20 25 30 35

Energy [keVnr]10 20 30 40 50

/S1)

-ER

mea

nb

(S2

10lo

g

-1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

99.75% rejection

NR acceptanceσ-3

S1 [PE]2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38

mχ ≥ 50GeV

mχ = 10GeV

mχ = 7GeV

XENON100: Nuclear Recoil Acceptance


WIMP spectrum

3 pe 4 pe

Photoelectron spectrum

Efficiency corrected

• Resolution near threshold is dominated bystatistics in photon detection efficiency.

• For each WIMP mass, convert the energyspectrum to a S1 spectrum in photoelec-trons, apply the S2 acceptance function,convolve the spectrum with a Poisson dis-tribution.

Energy [keVnr]10 20 30 40 50

Acc

epta

nce

0

0.2

0.4

0.6

0.8

1

S1 [PE]5 10 15 20 25 30 35

cS1 [pe]

0 5 10 15 20 25 30

cS

2 [p

e]

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

Preliminary

mχ ≥ 50GeV

mχ = 10GeV

mχ = 7GeV

XENON100: Background Prediction


• Expected background: 48 kg fiducial mass,99.75% electronic recoil rejection, 100.9live days

• Statistical leakage (from electronic recoilevents)

• 1.14 ± 0.48 events, estimated from thenon-blinded electronic recoil band frombackground

• Dominated by 85Kr (Kr concentration∼700 ppt) due to a previous leak

• Anomalous leakage

• 0.56± 0.25 events, estimated using dataand MC from 60Co and background

• Neutron prediction from MC

• 0.11 ± 0.08 events, muon-induced fastneutrons and neutrons from (α, n) reac-tions and spontaneous fission

• Total 1.8± 0.6 events in 100.9 days

Energy [keVnr]10 20 30 40 50

Acc

epta

nce

0

0.2

0.4

0.6

0.8

1

S1 [PE]5 10 15 20 25 30 35

Energy [keVnr]10 20 30 40 50

/S1)

-ER

mea

nb

(S2

10lo

g

-1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

99.75% rejection

NR acceptanceσ-3

S1 [PE]2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38

XENON100: Unblinding


• 6 events in the signal region after unblinding, inspection reveals 3 are due to electronic noise

• Population of noise events near threshold, leaks into signal region

XENON100: Unblinding


Energy [keVnr]10 20 30 40 50

/S1)

-ER

mea

nb

(S2

10lo

g

-1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

S1 [PE]2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38

• 6 events in the signal region after unblinding, inspection reveals 3 are due to electronic noise

• Population of noise events near threshold, leaks into signal region

• Remove population with noise post-unblinding cut, 3 candidate WIMP events remain

XENON100: Event Distribution


Energy [keVnr]10 20 30 40 50

/S1)

-ER

mea

nb

(S2

10lo

g

-1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

S1 [PE]2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38

]2 [cm2Radius0 50 100 150 200 250

z [c

m]

-30

-25

-20

-15

-10

-5

0

Radius [cm]2 4 6 8 10 12 14 15.3

• 3 candidate events clearly inside the 48 kg fiducial volume

• Probability (Poisson) to observe 3 or more events when expecting 1.8± 0.6 is 28%

• Profile Likelihood analysis does not yield a significant signal excess either, background-onlyhypothesis has a p-value of 31%

XENON100: Profile Likelihood Approach


• Construct the Likelihood function

L = L1(σ,Nb, ǫs, ǫb,Leff , vesc;mχ)

× L2(ǫs)× L3(ǫb)

× L4(Leff)× L5(vesc)

• Main term contains only one parameter ofinterest, the signal cross-section σ, otherparameters are nuisance parameters andprofiled out

• Additional terms constrain the nuisanceparameters in the main term

• Makes use all observed events in theWIMP search data, no sharp S2/S1 dis-crimination cut, energy distribution

• Allows systematic uncertainties to be in-corporated in a consistent manner

S1 [PE]0 2 4 6 8 10 12 14 16 18 20 22 24

S2

[PE

]0

1000

2000

3000

4000

5000

6000

• More details in

Aprile et al., Phys. Rev. D 84, 052003, 2011

XENON100: Limit


]2WIMP Mass [GeV/c6 7 8 910 20 30 40 50 100 200 300 400 1000

]2W

IMP

-Nuc

leon

Cro

ss S

ectio

n [c

m

-4510

-4410

-4310

-4210

-4110

-4010

-3910

]2WIMP Mass [GeV/c6 7 8 910 20 30 40 50 100 200 300 400 1000

]2W

IMP

-Nuc

leon

Cro

ss S

ectio

n [c

m

-4510

-4410

-4310

-4210

-4110

-4010

-3910

]2WIMP Mass [GeV/c6 7 8 910 20 30 40 50 100 200 300 400 1000

]2W

IMP

-Nuc

leon

Cro

ss S

ectio

n [c

m

-4510

-4410

-4310

-4210

-4110

-4010

-3910

DAMA/I

DAMA/Na

CoGeNT

CDMS (2010)

CDMS (2011)

EDELWEISS (2011)

XENON10 (S2 only, 2011)

XENON100 (2010)

XENON100 (2011)observed limit (90% CL)

Expected limit of this run:

expectedσ 2 ± expectedσ 1 ±

Buchmueller et al.

Trotta et al.

• Strongest limit to date over a large WIMP mass range, challenges the interpretation ofCoGeNT and DAMA signals as being due to low mass WIMPs

• Results recently published Aprile et al., Phys. Rev. Lett. 107, 131302, 2011

XENON100: Current Data Taking Status


• Kr concentration reduced by more than ×5 (∼50% background reduction)

• Improved S2 trigger with lower trigger threshold

• New dark matter run started early 2011, ∼160 live days accumulated

• Much more ER calibration data, already ∼16 live days 60Co and ∼18 live days 232Th

Date [mm/dd]03/01 05/01 07/01 08/31 10/31

Live

Day

s

0

20

40

60

80

100

120

140

160

Science Data (run_10)

Co)60Calibration (Th)232Calibration (

Calibration (AmBe)

XENON1T: Future


• 1m3 TPC, 2.4t LXe, 1t fiducial mass

• ×100 background reduction compared toXENON100

• Low radioactivity photosensors

• 10 m water shield

• Currently in design phase, construction 2012

• Approved for construction in Hall B at LNGS

Summary


Date [mm/dd]03/01 05/01 07/01 08/31 10/31

Live

Day

s

0

20

40

60

80

100

120

140

160

Science Data (run_10)

Co)60Calibration (Th)232Calibration (

Calibration (AmBe)

• XENON100 background goal achieved, detector is taking dark matter data

• Accumulated 100 days exposure in 2010, 3 events observed with an expectation of 1.8± 0.6

• Set the most stringent limit to date on the WIMP-nucleon spin-independent cross section

• Already ∼160 days exposure accumulated with lower background level and improved triggerthreshold, stay tuned for new results

Extras

XENON100: Position Dependent Corrections


Z [mm]

-300-250

-200-150

-100-50

0Radius [mm]

0 20 40 60 80 100 120 140

rela

tive

LCE

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

histEntries 2888Mean x -178.9

Mean y 72.85

RMS x 86.59

RMS y 43.28

histEntries 2888Mean x -178.9

Mean y 72.85

RMS x 86.59

RMS y 43.28

x [mm]-150 -100 -50 0 50 100 150

y [m

m]

-150

-100

-50

0

50

100

150

0.75

0.80

0.85

0.90

0.95

1.00

1.05

1.10

• Corrections from measurements with 137Cs,AmBe (40 keV inelastic), 131mXe (164 keV),with agreement better than 3%.

• S1 rz and S2 xy correction due to spatial de-pendence of the light collection.

• S2 z correction due to finite electron lifetime.

Date [mm/dd/yy]12/31/10 03/02/11 05/02/11 07/02/11 09/01/11

s]µE

lect

ron

Life

time

[

300

350

400

450

500

550

600

eq.]2

Concentration [ppb O

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

New Measurement of Leff : Extracting Leff


Time of Flight [ns]10 20 30 40 50 60 70 80 90 100 110

]-1

d-1

Rat

e [e

vts

ns

1

10

210

310

410

Time of Flight [ns]10 20 30 40 50 60 70 80 90 100 110

]-1

d-1

Rat

e [e

vts

ns

1

10

210

310

410

o = 34.5θ 1.0 keV±6.5

Energy [keV]0 2 4 6 8 10 12 14 16 18 20

]-1

d-1

Rat

e [e

vts

keV

1

10

210

310

410

Energy [keV]0 2 4 6 8 10 12 14 16 18 20

]-1

d-1

Rat

e [e

vts

keV

1

10

210

310

410

o = 34.5θ 1.0 keV±6.5

• Transform the MC recoil energy spectrum into a simu-lated scintillation spectrum g using S1 = Ly ·Leff,j ·Er,

with gaussian energy resolution σ = R√Er, PMT gain

fluctuations, and applying trigger efficiency

• Extract the energy dependence of Leff by minimizingthe χ2 between the measured and simulated spectra,h and g

χ2(

Leff,j , Rj

)

=

N∑

i=0

[

hi − gi(

Leff,j , Rj

)]2

σ2

h,i+ σ2

g,i

(

Leff,j , Rj

)

effL0.105 0.110 0.115 0.120 0.125 0.130 0.135 0.140

Res

olut

ion

Par

amet

er R

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

4049

5973

89109

133162

198

o = 34.5θ 1.0 keV±6.5

Scintillation Signal [pe]0 10 20 30 40 50 60 70

]-1

d-1

Rat

e [e

vent

s pe

0

20

40

60

80

100o = 34.5θ

1.0 keV±6.5 = 32.12χ

Uncertainty in Leff


σ2

Leff= σ2

Leff ,fit+

( ∂Leff

∂Ly

)2σ2

Ly+

( ∂Leff

∂Enr

)2σ2

Enr

+(

∆Leff

∆ǫ

)2σ2ǫ +

(

∆Leff

∆rg

)2σ2rg

+(

∆Leff

∆rs

)2σ2rs

• The total uncertainty on Leff is computed from

• σLeff ,fit, uncertainty from the fit

• σLy, uncertainty 57Co light yield

• σEnr, spread in nuclear recoil energies

• σǫ, liquid scintillator cut efficiency uncertainty

• σg , neutron generator position uncertainty

• σs, liquid scintillator position uncertainty

• ∂Leff/∂Enr is computed from a logarithmic fit to the measured values

• ∆Leff/∆ǫ, ∆Leff/∆rg , and ∆Leff/∆rs are calculated from MC simulations

• At all energies, the dominant contribution is from σEnrBelow 6.5 keV, the second largest

contribution is from σǫ. At 6.5 keV and above, the second largest contribution is fromσLeff ,fit

XENON100: Anomalous Leakage


• Two types of leakage from the ER band, statis-tical leakage and anomalous leakage

• We assume the ER band is Gaussian inlog (S2/S1), fixed discrimination at 99.75%gives the expected statistical leakage

• Events with low non-Gaussian S2/S1 also ingamma calibration data, e.g. 60Co

• One source for those “anomalously leaking”events is multiple scatter events where one ormore scatter occurs in a charge insensitive regionof the detector

• Use the S1 Pattern likelihood cut to removeevents likely due to two energy deposits

• Compute the expected anomalous leakage inbackground using 60Co as reference

GXe

LXe

Eg

Ed

S1

S2

e-e-e-

S11

2

Example Noise Event


s]µDrift Time [0 2 4 6 8 10 12 14

Am

plitu

de [V

]

0.0

0.1

0.2

0.3

0.4 TPC

s]µDrift Time [-200 -150 -100 -50 0 50 100 150

Am

plitu

de [V

]

0.0

0.2

0.4TPC

-10 -5 0 5 10

0.00

0.01

S1: 3.7 pecS1: 5.6 pe

guillaumeplante columbiauniversity …zentner/guillaume_plante_files/... · 2011. 11. 14. · pmti...

Documents