Measurement of low energy Electronic Recoil Response and - PowerPoint PPT Presentation

Measurement of low energy Electronic Recoil Response and Electronic/Nuclear Recoils Discrimination in XENON100 Jingqiang Ye, UC San Diego On behalf on XENON Collaboration LIDINE 2017, Sep. 22 - 24, 2017, SLAC 1

Introduction g1 𝑂 "# S1 ER,NR g2 𝑂 $ S2 ER g 1 = S 1 , g 2 = S 2 • Scintillation signal S1 N ph N e • Charge signal S2 • Different S2/S1 for ER/NR NR • Primary scintillation gain g1 • Secondary scintillation gain g2 • g1 is proportional to photon detection efficiency(PDE) 2

Data Extraction Calibration source & detector condition: Drift Extraction Electron Max drift Events in sub-FV( 10 ) ) field(V/cm) field(kV/cm) lifetime(us) time(us) • CH 3 T calibration (<18.6 keVee, ER) 1470 ± 190 • AmBe calibration(NR) CH3T 400 10.0 182 43.4 • 3 different drift and extraction fields 390 ± 160 CH3T 167 8.2 202 11.9 CH3T 100 8.2 590 ± 30 220 8.9 0 0 2.2 1490 ± 100 AmBe 400 10.0 182 3.5 AmBe 167 8.2 490 ± 130 202 3.6 20 20 2 550 ± 60 AmBe 100 8.2 220 6.5 FV#1 Relative Light Yield to Center 1.8 40 40 FV#2 1.6 s] 60 60 µ FV#3 Drift Time [ To compare with different PDE: 1.4 80 80 • 7 sub-FVs(‘small detector’) FV#4 1.2 50% quantile in 𝑆 & direction, equal in drift time • 100 100 FV#5 • Enough statistics in each sub-FV 1 120 120 • Avoid strong field distortion in top and bottom region FV#6 0.8 • No position dependent correction of S1 and S2 140 140 • Small S1 and S2 variation in each sub-FV(6% for S1, 0.6 FV#7 5% for S2) 160 160 0.4 • PDE increases from top part to bottom part 180 1800 0.2 0 20 20 40 40 60 60 80 80 100 120 140 160 180 200 220 100 120 140 160 180 200 220 2 2 Detected Radius at Liquid Surface [cm ] 3

Detector calibration(g1, g2) Calibration principle: Doke method E = W · ( S 1 + S 2 ) , W = 13 . 7 eV g 1 g 2 S 2 E = − g 2 S 1 E + g 2 g 1 W 4

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

� Simulation model Excimer production Binomial 𝑂 $6 ~ B( 𝑂 + , 𝛽/(1 + 𝛽) ) Xe * Incoming particle Xe S1 S2 Xe + Xe 2+ Fano fluctuation Gaussian e - e - e - 𝑂 + ~ N(E/W, 𝐺𝐹/𝑋 ) F=0.059 Photon detection Poisson Recombination fluctuation Recombination Gaussian Electron drift & extraction Binomial r~ N(<r>, ∆𝑠 ) Binomial ~ B( 𝑂 7 ,r) ~B( 𝑂 $ , 𝜁 4 ∗ 𝜁 $6 ) r: recombination factor <r>: average recombination fraction(<r> is tuned) ∆𝑠 : recombination fluctuation( Δ𝑠 /<r> is tuned) 6

MC-data matching Use Binned Maximum Likelihood Estimation(MLE) in Log10(S2/S1) vs S1 space to extract ER response 3500 S2 spectrum (data) 0 < S1 < 10 S1 spectrum (data) 2000 3000 15.4%-84.6% credible region 15.4%-84.6% credible region 2500 Counts 1000 2000 1500 0 10 < S1 < 20 1000 1500 500 1000 0 500 3 . 0 Log 10 (S2/S1) Point estimation MC 0 10 3 800 20 < S1 < 30 Counts 2 . 5 600 Counts 400 2 . 0 10 2 200 0 1 . 5 30 < S1 < 40 300 200 3 . 0 ER band medians (data) Log 10 (S2/S1) Data 10 2 100 ER band medians (mc) 2 . 5 0 Counts 40 < S1 < 50 80 2 . 0 10 1 60 40 1 . 5 20 10 0 0 0 10 20 30 40 50 60 70 80 0 1000 2000 3000 4000 7 S1[PE] S2[PE]

Light yield • 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 a) 100 V/cm b) 167 V/cm c) 400 V/cm 50 h n ph i / E [ph/keV] 40 30 Best estimation ± 1 σ fitting uncer. 20 Credible region NEST v0.98 10 LUX @ 105 V/cm LUX @ 180 V/cm Unc. [ph/keV] 2 4 6 8 10 12 14 2 4 6 8 10 12 14 2 4 6 8 10 12 14 4 2 0 � 2 � 4 2 4 6 8 10 12 14 2 4 6 8 10 12 14 2 4 6 8 10 12 14 Energy[keV] 8

Recombination fluctuation No significant change observed between different drift fields 0 . 08 d) 100 V/cm e) 167 V/cm f) 400 V/cm 0 . 07 0 . 06 0 . 05 ∆ r 0 . 04 Best estimation 0 . 03 ± 1 σ fitting uncer. 0 . 02 Credible region LUX @ 180 V/cm 0 . 01 2 4 6 8 10 12 14 2 4 6 8 10 12 14 2 4 6 8 10 12 14 0 . 03 Abs. unc. 0 . 02 0 . 01 0 . 00 − 0 . 01 − 0 . 02 − 0 . 03 2 4 6 8 10 12 14 2 4 6 8 10 12 14 2 4 6 8 10 12 14 Energy[keV] 9

ER/NR discrimination • Normalize S1 to photons generated to compare ER leakage under different g1 • S2 is corrected for electron lifetime 10

ER/NR discrimination ER leakage is smaller at larger g1 11

ER/NR discrimination for different g1 and drift fields • 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 − 2 10 ER Leakage Fraction 400 V/cm − 3 10 167 V/cm 100 V/cm 0.04 0.05 0.06 0.07 0.08 0.09 g 12 1

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 • Drift field increases: • ER/NR separation increases • ER band width increases • g1 increases: • ER/NR separation doesn’t change significantly • ER band width decreases 14

Recombination factor LUX Collaboration arXiv: 1512.03133 0 . 8 0 . 7 0 . 6 0 . 5 h r i 0 . 4 0 . 3 100 V/cm 0 . 2 167 V/cm 0 . 1 400 V/cm 0 . 0 0 2 4 6 8 10 12 14 16 Energy[keV] Black: 180 V/cm Blue: 105 V/cm 15

3500 S2 spectrum (data) 0 < S1 < 10 S1 spectrum (data) 2000 3000 15.4%-84.6% credible region 15.4%-84.6% credible region 2500 Counts P-value = 0.01 1000 2000 1500 0 10 < S1 < 20 1000 1500 P-value = 0.16 500 1000 0 500 3 . 0 Log 10 (S2/S1) Point estimation MC 0 10 3 800 20 < S1 < 30 Counts 2 . 5 600 P-value = 0.10 Counts 400 2 . 0 10 2 200 0 1 . 5 30 < S1 < 40 300 P-value = 0.37 200 3 . 0 ER band medians (data) Log 10 (S2/S1) Data 10 2 100 ER band medians (mc) 2 . 5 0 Counts 40 < S1 < 50 80 P-value = 0.38 2 . 0 10 1 60 40 1 . 5 20 10 0 0 0 10 20 30 40 50 60 70 80 0 1000 2000 3000 4000 16 S1[PE] S2[PE]