Scintillation and Ionization yield of

Liquid ArgonRichard Saldanha

On behalf of the SCENE collaboration

24th September 2015

Postdoc position available at UChicago

Work with Prof. Luca Grandi on Xenon1T

For more details please seehttps://inspirehep.net/record/1389800

or talk directly to Luca !

Prior status of Ar Nuclear Recoil calibration

3

!"#$%&%'()*#%+"#,-".#/"#012*"3#!4'("#5%6"#789#:;<=;>#<=;<=?#

Few measurements of scintillation yield in liquid

argon, especially at low (< 25 keV) energies, with no studies of the effect of

the drift field

One recent measurement of ionization yield

(Joshi, Sangiorgio et. al arXiv: 1402.2037) at 6.7 keVr

7Li(p,n)7Be

4 C.A.Burke et al., Phys. Rev. C 10, 1299 (1974)ATOMIC DATA AND NUCLEAR DATA TABLES 15, 57-84 (1975)

Produces monoenergetic (< 2 MeV) neutronsProton Energy Threshold: 1.881 MeVCross Section Peak: Ep = 2.25 MeV

Experimental Setup

5

6

Proton Beam

• Pulsed Beam(Period = n x 101.5 ns)

• Beam Energy: 10 MeV max.

• Beam Energy Uncertainty(± 1 keV mean, ± 2 keV spread)

• Beam angle at target < 0.006 deg

SCENE parameters1 ns pulses203 ns period (S1 only)406 - 812 ns period (S2)

2.376 - 2.930 MeV

6.3 x 104 p/pulse3 x 10-4 n/pulse (@TPC)

FN Tandem Van de Graaff accelerator at the University of Notre Dame

LiF target

7

0.20 mg/cm2 LiF on 1 mm aluminium backing

0.1 mm tantalum layer between LiF target and aluminium to fully stop protons (Oct run)

~ 20 keV energy loss in target

8

Two-phase liquid argon TPC

2 x 3” R11065 PMTs top and bottom3M foil reflector

ITO-coated fused silica windows

9

Two-phase liquid argon TPC

2 x 3” R11065 PMTs top and bottom3M foil reflector

ITO-coated fused silica windows

10

Energy [keV]0 5 10 15 20 25 30

Counts/0.5 PE

0

200

400

600

800

1000

1200

1400

1600

1800

2000

All scatters

Mult. scatters

Single scatter

median: 10.3 keV

Two-phase liquid argon TPCDesigned to minimize neutron scatters outside active volume

Geant4 simulations of neutron beam predict a clear single scatter peak (after coincidence cuts) above the

multiple scatter background

Simulations of electric field uniformity indicate a ~ 8-10% variation of electric field in the

neutron interaction region

Neutron Detectors

11

EJ-301 Organic Liquid Scintillator5” x 5” coupled to PMT

Provides both timing and γ/n pulse shape discrimination

EJ-301’s mounted on two-axis goniometric stand

Background Discrimination

12

Proton Beam RF Timing

Triggered on coincidence of TPC PMTs (OR/AND)

and any one EJ detector

TPC S1 TimeTPC S1 Pulse Shape

EJ TimeEJ Pulse Shape

Experimental Setup

13

S1 S2[V/cm ] [V/cm ]

0 049.596.4 96.4193 193293 293486 486970

71 cm

73.1 cm (Jun)82.4 cm (Oct)

14Ntof [ns]-20 0 20 40 60 80 100 120 140 160 180

Npsd

0

0.05

0.1

0.15

0.2

0.25

0.3

0

1

2

3

4

5

6

7

8

Neutrons

Photons

(c)

Neutron DetectorTPCtof [ns]

-40 -20 0 20 40 60 80 100 120

f90

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0

1

2

3

4

5

6

7

Photons

Neutrons

(b)7 us after S1Require S2 starts later than

LAr-TPC

Selection Cuts

Application of 2 TOF and 2 pulse shape cuts

allows for extremely clean sample of data

Selection Cuts (57.3 keV, 193 V/cm)

15

Before Cuts After Cuts

S1 S2

S1 Trigger Efficiency 16

Signal amplitude [PE]0 2 4 6 8 10 12 14

Trig

ger

Effi

cien

cy

0

0.2

0.4

0.6

0.8

1

Two PMT OR

Two PMT AND

S1 [PE]0 10 20 30 40 50 60

Arbi

trar

y Un

its

0

0.01

0.02

0.03

0.04

0.05 Two PMT OR

Two PMT AND

20.5 keV (1000 V/cm)

Two methods used to measure trigger efficiency

Simultaneously digitize signal and trigger logic, while triggering on

22Na source with external coincidence detector

(Plante et. al. Phys. Rev. C 84, 045805)

Nuclear recoils at 20.5 keV and 970 V/cm were acquired using

both OR and AND trigger

S1 OR Analysis Threshold: 4 PES1 AND Analysis Threshold: 12 PE

Fitting S1 Spectra

• To determine the scintillation yield, the S1 spectra are fit with a Monte Carlo generated recoil spectrum.

• Starting from the neutron production, the entire geometry of the TPC and neutron detectors is simulated

• The same coincidence and timing cuts as used in the analysis are implemented in the simulation to obtain the final sample of nuclear recoils

• The data were then fit to the MC spectra using an independent fixed Leff and resolution for each geometrical setup

17

Fitting S1 Spectra

18

Example of fits to 14.8 keV

nuclear recoils

Kr Reference

19S1 [PE]0 50 100 150 200 250

Counts/2 PE

1

10

210

310All Coincidence Events

After Ntof & Npsd Cuts

After TPCtof & f90 Cuts

(a)

Leff measurements are usually made with respect to electron recoils at null field

83mKr is extremely useful as a reference calibration since it can

easily injected, distributes within the active volume, and decays to a

stable isotope within a few hours

Kr peak

83mKr was continuously injected into the detector to

monitor the light yield stability during the beam runs

Null Field LY:6.3 ± 0.3 PE/keV (June)4.8 ± 0.2 PE/keV (Oct)

Leff Results

20

Recoil Energy [keV]10 20 30 40 50 60

Kr(0V/cm)

83m

S1 Relative to

0.14

0.16

0.18

0.2

0.22

0.24

0.26

0.28

0.3

0.32

0 V/cm

290 V/cm

970 V/cm

Drift Field [V/cm]0 200 400 600 800 1000

Kr (0V/cm)

83m

S1 Relative to

0.14

0.16

0.18

0.2

0.22

0.24

0.26

0.28

0.3

0.3210.3 keV25.4 keV36.1 keV57.3 keV

Previously believed that the scintillation yield of nuclear recoils is unaffected by external fields

due to the dense nature of the track

SCENE measurements show lower scintillation yield at high fields

Advantageous to run liquid argon TPCs at low fields to improve pulse shape discrimination

Leff Systematics

21

Example of systematics studied at null field

Leff Results

22

0 20 40 60 80 1000

0.1

0.2

0.3

0.4

0.5

Recoil energy [keV]

Leff a

t 0

V/c

m

This workGastlerRegenfus

Gastler et. al. Phys. Rev. C 85, 065811 (2012) Regenfus et. al. J. Phys.: Conf. Ser. 375, 012019 (2012) Creus et. al arXiv:1504.07878v2

Creus

Energy [keV]0 20 40 60 80 100 120 140 160 180 200

S2 [PE]

0

100

200

300

400

500

600

700

800

Quadratic Qy vs Energy

Logarithmic S2 vs Energy

Power Law S2 vs Energy

PRD Submission

Fit Range

500 V/cm486

S2 fitting

Threshold for S2 fitting set by translating S1 threshold using data at each drift field setting

23

S1 [PE]0 20 40 60 80 100 120 140 160 180 200

Mean S2 [PE]

0

100

200

300

400

500

600

S2 OR THRESHOLD: 123 PE

S2 AND THRESHOLD: 196 PE

16.9 keV25.4 keV36.1 keV57.3 keVOR ThresholdAND Threshold

193 V/cm

S2 yield can vary rapidly with energy, significantly even within a

given geometrical setup

Data was fit using several different global functions:

Quadratic, Logarithmic, Power Law

S2 fitting

24

Results for data acquired at 486 V/cm

Qy results

25

Recoil energy [keV]10 20 30 40 50 60

S2 yield [PE/keV]

2

4

6

8

10

12

14

16

96.4 V/cm

193 V/cm

293 V/cm

486 V/cm

/keV]

-S2 yield [e

1

2

3

4

5

Model of energy deposition for electron recoils

26

LET Dependence of Scintillation Yields in Liquid Argon Doke, T. et al. Nucl.Instrum.Meth. A269 (1988) 291-296

Energy (E)

Nex Ni

Nph Ne-

S1 S2

1/W1/Wph -1/W

r (dE/dx,F)1 - r (dE/dx, F)

g1 g2

Excited Ar atoms

Ar ions + electrons

UV Photons

S1 signal

Drift Electrons

S2 Signal

Recombination(depends of track density, drift field)

W : Energy required to produce electron-ion pairWph : Energy required per quanta (Nex + Ni) = Energy required to produce a photon when r = 1r : Recombination fractiong1: Scintillation detection efficiency [PE/UV photon]g2: Electron detection efficiency [PE/electron]

Only recombination (r) depends on the drift field (F)

S1 and S2 are linearly anti-correlated, independent of drift field

27

E

Nex Ni

Nph Ne-

S1 S2

1/W

g1 g2

1/Wph -1/W

r (dE/dx,F)1 - r (dE/dx, F)

S1

E= −g1

g2· S2E

+g1Wph

Is this also true for nuclear recoils ?

S2 vs S1

28

S2 yield [PE/keV]0 5 10 15 20 25 30 35 40 45

S1 yield [PE/keV]

3

3.5

4

4.5

5

5.5

6(a) / ndf 7 / 222!

0.08±(max) [PE/keV] 5.33 phW/1

g

0.003± 0.033 2

g/1

g

Kr during:83m

16.9 keV

25.4/57.3 keV

36.1 keV

S2 yield [PE/keV]0 2 4 6 8 10 12 14 16

S1 yield [PE/keV]

0.8

1

1.2

1.4

1.6

1.8(b)

16.9 keV

25.4 keV

36.1 keV

57.3 keV

(max) [PE/keV]phW/1

g 0.04±16.9 keV: 1.33

0.04±25.4 keV: 1.40

0.04±36.1 keV: 1.49

0.03±57.3 keV: 1.54

Krypton Nuclear Recoils

S1

E= −g1

g2· S2E

+g1Wph

Both electron and nuclear display an S1-S2 anti-correlation, with the same slope (-g1/g2)

g1: 0.104 ± 0.006 PE/photon g2: 3.1 ± 0.3 PE/e-

Assuming Wph(Kr) = 19.5 ± 0.1 eV

S2 vs S1

29

S2 yield [PE/keV]0 5 10 15 20 25 30 35 40 45

S1 yield [PE/keV]

3

3.5

4

4.5

5

5.5

6(a) / ndf 7 / 222!

0.08±(max) [PE/keV] 5.33 phW/1

g

0.003± 0.033 2

g/1

g

Kr during:83m

16.9 keV

25.4/57.3 keV

36.1 keV

S2 yield [PE/keV]0 2 4 6 8 10 12 14 16

S1 yield [PE/keV]

0.8

1

1.2

1.4

1.6

1.8(b)

16.9 keV

25.4 keV

36.1 keV

57.3 keV

(max) [PE/keV]phW/1

g 0.04±16.9 keV: 1.33

0.04±25.4 keV: 1.40

0.04±36.1 keV: 1.49

0.03±57.3 keV: 1.54

Krypton Nuclear Recoils

S1

E= −g1

g2· S2E

+g1Wph

Intercept (g1/Wph) changes with energy for nuclear recoils

Total Quenching

30Recoil energy [keV]0 10 20 30 40 50 60 70

Total quenching factor

0.15

0.2

0.25

0.3

0.35

0.4

This work

Fit to Mei et al.

Lindhard Theory

Nuclear Recoil Energy (Enr)

Nex Ni

Nph Ne-

1/W1/Wph -1/W

r (dE/dx,F)1 - r (dE/dx, F)

Electron Recoil Energy (E)

η (Enr) Some nuclear recoil energy is lost to atomic motion

Lindhard Theory: Quenching attributed to loss of energy to atomic motion

Mei et. al: Quenching attributed to Lindhard + Birk’s quenching

(kB = 5.0 ± 0.2 x10-4 MeV/g/cm2)

Recombination ModelNumber of ions can also be determined using a

recombination model

31

Drift field [V/cm]0 100 200 300 400 500 600

S2 y

ield

[PE

/keV

]

0

5

10

15

20

25

30

35

40

45

16.9 keV25.4 keV36.1 keV57.3 keV

Kr during:m83

16.9 keV25.4/57.3 keV36.1 keV

/keV

]-

S2 y

ield

[e

0

2

4

6

8

10

12

14

Modified Thomas-Imel Box Model (arXiv: 1402.2037)

B: Constant - same for all species, energies

(0.61 ± 0.03)

C, Ni: Constant - different for each energy

Recoil Energy [keV] C [(V/cm)B/e−] Ni/E[keV]16.9 0.58±0.17 8.2±1.925.4 0.50±0.23 7.0±2.536.1 0.45±0.19 5.9±2.057.3 0.42±0.16 4.8±2.64.8± 1.8

Summary• We have performed a measurement of the scintillation and

ionization yield of liquid argon

• Measurement spans energy ranges from 10 - 57 keV recoils with drift fields from 0 - 970 kV/cm

• Simultaneous measurement of the S1 and S2 signals allows for a comparison with quenching and recombination models

• For additional details, including measurements of pulse shape discrimination and directionality, please see publications:

Observation of the dependence on drift field of scintillation from nuclear recoils in liquid argon T. Alexander et. al. (SCENE Collaboration) Phys.Rev. D88 (2013) 9, 092006

Measurement of Scintillation and Ionization Yield and Scintillation Pulse Shape from Nuclear Recoils in Liquid ArgonCao et. al. (SCENE Collaboration) Phys. Rev. D 91, 092007 (2015)

32

Thank you

Energy [keV]0 20 40 60 80 100 120 140 160 180 200

Mea

n S2

[PE

]

0

100

200

300

400

500

600

700

800293 V/cm

16.9 keV25.4 keV

36.1 keV57.3 keVFit to MC Spectrum

293 V/cm

35

Take profile of means in each

S1 slice

Convert S1 axis to energy using Leff

measurement:

S1 = E x LYx Leff

293 V/cm

Assume all events after cuts are single scatter

nuclear recoils in TPC

S1-S2 Consistency Check

36