measurement of low energy electronic recoil response and ... · s2[pe] 0 20 40 60 40
TRANSCRIPT
Measurement of low energy Electronic Recoil Response and Electronic/Nuclear Recoils
Discrimination in XENON100
Jingqiang Ye, UC San DiegoOn behalf on XENON Collaboration
LIDINE 2017, Sep. 22 - 24, 2017, SLAC
1
Introduction
2
ER,NR
𝑁"#
𝑁$
S1
S2
g1
g2
g1 =S1
Nph, g2 =
S2
Ne• Scintillation signal S1• Charge signal S2• Different S2/S1 for ER/NR
• Primary scintillation gain g1• Secondary scintillation gain g2• g1 is proportional to photon detection efficiency(PDE)
ER
NR
Data Extraction
0
20
40
60
80
100
120
140
160
180
s]µD
rift T
ime
[
0 20 40 60 80 100 120 140 160 180 200 220]2 at Liquid Surface [cm2Detected Radius
Rel
ativ
e Li
ght Y
ield
to C
ente
r0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
2.20
20
40
60
80
100
120
140
160
1800 20 40 60 80 100 120 140 160 180 200 220
FV#1FV#2FV#3FV#4FV#5FV#6FV#7
3
Calibration source & detector condition:• CH3T calibration (<18.6 keVee, ER) • AmBe calibration(NR)• 3 different drift and extraction fields
To compare with different PDE:• 7 sub-FVs(‘small detector’)
• 50% quantile in 𝑅&direction, equal in drift time• Enough statistics in each sub-FV• Avoid strong field distortion in top and bottom region
• No position dependent correction of S1 and S2• Small S1 and S2 variation in each sub-FV(6% for S1,5% for S2)• PDE increases from top part to bottom part
Drift field(V/cm)
Extraction field(kV/cm)
Electron lifetime(us)
Max drift time(us)
Events in sub-FV(10))
CH3T 400 10.0 1470 ± 190 182 43.4
CH3T 167 8.2 390 ± 160 202 11.9
CH3T 100 8.2 590 ± 30 220 8.9
AmBe 400 10.0 1490 ± 100 182 3.5
AmBe 167 8.2 490 ± 130 202 3.6
AmBe 100 8.2 550 ± 60 220 6.5
Detector calibration(g1, g2)
4
E = W · (S1
g1+
S2
g2),W = 13.7eV
S2
E= � g2
g1
S1
E+
g2W
Calibration principle: Doke method
Detector calibration(g1, g2)
Calibration result:• g1 z dependence due to geometry effect• g2 z dependence due to electron lifetime• g1 is consistent under different drift fields• g2 increases with larger extraction field
5
XeIncoming particle Xe*
Xe+
e-
Xe2+
e- e-
S1S2
Fano fluctuationGaussian
𝑁+~ N(E/W, 𝐹𝐸/𝑊� )F=0.059
Recombination fluctuationGaussian
r~ N(<r>,∆𝑟)Electron drift & extraction
Binomial~B(𝑁$, 𝜀4 ∗ 𝜀$6)
Photon detectionPoisson
RecombinationBinomial~ B(𝑁7,r)
r: recombination factor<r>: average recombination fraction(<r> is tuned)∆𝑟: recombination fluctuation(Δ𝑟/<r> is tuned) 6
Excimer productionBinomial
𝑁$6~ B(𝑁+, 𝛼/(1 + 𝛼))
Simulation model
MC-data matching
7
0500
100015002000250030003500
Cou
nts
S1 spectrum (data)15.4%-84.6% credible region
1.5
2.0
2.5
3.0
Log 1
0(S2
/S1) Point estimation MC
0 10 20 30 40 50 60 70 80S1[PE]
1.5
2.0
2.5
3.0
Log 1
0(S2
/S1) Data ER band medians (data)
ER band medians (mc)
0
1000
20000<S1<10 S2 spectrum (data)15.4%-84.6% credible region
0
500
1000
150010<S1<20
0
200
400
600
800
Cou
nts20<S1<30
0
100
200
30030<S1<40
0 1000 2000 3000 4000S2[PE]
0
20
40
60
8040<S1<50
100
101
102
Cou
nts
102
103
Cou
nts
Use Binned Maximum Likelihood Estimation(MLE) in Log10(S2/S1) vs S1 spaceto extract ER response
Light yield
8
2 4 6 8 10 12 14
10
20
30
40
50
hnphi/
E[p
h/ke
V]
a) 100 V/cm
Best estimation±1� fitting uncer.Credible regionNEST v0.98LUX @ 105 V/cm
2 4 6 8 10 12 14
b) 167 V/cm
LUX @ 180 V/cm
2 4 6 8 10 12 14
c) 400 V/cm
2 4 6 8 10 12 14
�4�2
024
Unc
.[ph
/keV
]
2 4 6 8 10 12 14Energy[keV]
2 4 6 8 10 12 14
• Lower light yield at higher drift field as expected• Consistent with LUX measurement• Light yield deviates from NEST at high energy, especially at high drift fields
Recombination fluctuation
2 4 6 8 10 12 14
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
�r
d) 100 V/cm
2 4 6 8 10 12 14
e) 167 V/cm
Best estimation±1� fitting uncer.Credible regionLUX @ 180 V/cm
2 4 6 8 10 12 14
f) 400 V/cm
2 4 6 8 10 12 14
�0.03�0.02�0.01
0.000.010.020.03
Abs
.unc
.
2 4 6 8 10 12 14Energy[keV]
2 4 6 8 10 12 14
No significant change observed between different drift fields
9
ER/NR discrimination
10
• Normalize S1 to photons generated to compare ER leakage under different g1• S2 is corrected for electron lifetime
ER/NR discriminationER leakage is smaller at larger g1
11
ER/NR discrimination for different g1 and drift fields
12
• S1 range(100-400 photons), energy range(11-34 keVnr)• ER leakage is smaller at larger g1• No significant difference for ER leakage between 100 V/cm and
400 V/cm drift field
1 g
0.04 0.05 0.06 0.07 0.08 0.09
ER L
eaka
ge F
ract
ion
3−10
2−10
400 V/cm
167 V/cm
100 V/cm
Summary
• Light yield and recombination fluctuation for low energy under three drift field are measured
• Light yield under 100 and 167 V/cm are consistent with LUX measurement
• ER leakage is smaller for larger photon detection efficiency
• No significant difference in ER leakage is observed between 100 V/cm and 400 V/cm drift field
• The paper will be available on arXiv next week
13
Back up
14
• Drift field increases:• ER/NR separation increases• ER band width increases
• g1 increases:• ER/NR separation doesn’t change significantly• ER band width decreases
Recombination factor
0 2 4 6 8 10 12 14 16Energy[keV]
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
hri
100 V/cm167 V/cm400 V/cm
Black: 180 V/cmBlue: 105 V/cm
LUX CollaborationarXiv: 1512.03133
15
16
0500
100015002000250030003500
Cou
nts
S1 spectrum (data)15.4%-84.6% credible region
1.5
2.0
2.5
3.0
Log 1
0(S2
/S1) Point estimation MC
0 10 20 30 40 50 60 70 80S1[PE]
1.5
2.0
2.5
3.0
Log 1
0(S2
/S1) Data ER band medians (data)
ER band medians (mc)
0
1000
20000<S1<10 S2 spectrum (data)15.4%-84.6% credible region
0
500
1000
150010<S1<20
0
200
400
600
800
Cou
nts20<S1<30
0
100
200
30030<S1<40
0 1000 2000 3000 4000S2[PE]
0
20
40
60
8040<S1<50
100
101
102
Cou
nts
102
103
Cou
nts
P-value = 0.01
P-value = 0.16
P-value = 0.10
P-value = 0.37
P-value = 0.38