Photodetector Calibrations: Gain & Quantum Efficiency M. Toups - - PowerPoint PPT Presentation

Photodetector Calibrations: 
 Gain & Quantum Efficiency

• M. Toups

Fermi National Accelerator Laboratory
 Workshop on Calibration and Reconstruction for LArTPC Detectors
 12/10/2018

Scintillation Light in a LArTPC

Photodetectors

12/10/18 2

• M. Toups—LArTPC Caleb Reco Workshop

LAr Light Detection Motivation

• Provides interaction time (t0) to

reconstruct drift coordinate

– Correct for e- lifetime, fiducialize volume

• Trigger detector readout

• Cosmic background rejection

– Essential for surface LArTPCs

• Event reconstruction/particle ID

TPC ν Cosmics during drift window

12/10/18 3

• M. Toups—LArTPC Caleb Reco Workshop

LAr Scintillation light

4

• LAr is a bright scintillator: O(10,000) γ/MeV
• LAr scintillates in the VUV at 127 nm
• LAr is very transparent to its own light
• Rayleigh scattering length: λ≈66 cm
• LAr light divided between two components
• Fast component: τ≈6 ns
• Slow component with τ≈1.5 us
• LAr scintillation light and charge anti-

correlation

E Morikawa et al, J Chem Phys vol 91 (1989) 1469

12/10/18 4

• M. Toups—LArTPC Caleb Reco Workshop

LAr Scintillation light

12/10/18 5

• LAr is a bright scintillator: O(10,000) γ/MeV
• LAr scintillates in the VUV at 127 nm
• LAr is very transparent to its own light
• Rayleigh scattering length: λ≈66 cm
• LAr light divided between two components
• Fast component: τ≈6 ns
• Slow component with τ≈1.5 us
• LAr scintillation light and charge anti-

correlation

• V. Gehman et al, NIM A 654, 116 (2011)

Tetraphenyl butadiene (TPB) evaporative coating

Standard SiPMs/PMTs not sensitive at this wavelength, so use wavelength shifters (WLS)

See also C. P . Benson et al, Eur. Phys. J. C (2018) 78: 329 5

• M. Toups—LArTPC Caleb Reco Workshop

TPB-coated acrylic plate

Present LArTPC Photon Detection Systems

6

127 nm light 430 nm light PMT

TPB coated acrylic light guide

e.g. MicroBooNE, SBND, ICARUS e.g. SBND, protoDUNE e.g. SBND, protoDUNE

ARAPUCA

e.g. protoDUNE

Commercial (EJ-280) light guide with TPB-coated radiator plates

12/10/18 6

• M. Toups—LArTPC Caleb Reco Workshop

Gain & Quantum Efficiency (QE) Calibrations

7

127 nm light 430 nm light

WLS 
 (e.g. TPB)

WLS conversion efficiency Photosensor QE/PDE Photosensor gain Incident VUV photon flux Acceptance Measured charge vs. time

=

Photodetector

Photodetector quantum efficiency

Goal: Infer incident VUV photon flux from measured charge

12/10/18 7

• M. Toups—LArTPC Caleb Reco Workshop

Photosensor Gain Calibration

• First step in PD calibration converts measured charge to PEs
• Function only of the photosensor and its operating conditions
• Depends on temperature, operating voltage
• Separates changes in photosensor response from other

changes in photodetector response or in the LArTPC response

12/10/18 8

• M. Toups—LArTPC Caleb Reco Workshop

Photosensor ex situ Gain Calibration

• Warm measurements of photosensor

gain often performed as part of a suite

• f acceptance tests
• Additional cold tests of photosensor

gain verify proper functioning at cryogenic temperatures

• Photosensor gain measurements at

cryogenic temperatures can determine nominal operating voltage for photosensors ahead of time

• DUNE-specific: ganging together

multiple MPPCs with similar gains may require pre-selection based on cold gain

JINST 13 (2018) no.10, P10030 12/10/18 9

• M. Toups—LArTPC Caleb Reco Workshop

Photosensor in situ Gain Calibration (visible light)

• MicroBooNE system uses individual

LEDs for each PMT

• Similar method but with a single laser

diode fanned out to all PMTs employed in ICARUS

• Both can also be used to verify proper

PMT functioning after installation

• Single PEs from late component of LAr

scintillation light can also be used to monitor gain over time while running

JINST 10 (2015) no.06, T06001

12/10/18 10

• M. Toups—LArTPC Caleb Reco Workshop

Photosensor in situ Gain Calibration (UV light)

• Used in 35-ton, protoDUNE, SBND, and DUNE
• Pulsed external UV LEDs (245-280 nm)

coupled to quartz fibers connected to diffusers

• n the cathode to uniformly illuminate PDs
• Reference photodiode monitors LED output
• Gain measurements, linearity, timing

resolution, and stability over time

• Does not measure (relative) QE due to different

PD response at 245 nm

• SiPMs/MPPCs often said to “self-calibrate”,

but situation may be more complicated when cold summing boards are used and clearly separated peaks are not visible

arXiv:1706.07081 [physics.ins-det] JINST 13 (2018) no.06, P06022

12/10/18 11

• M. Toups—LArTPC Caleb Reco Workshop

Photodetector QE

• Defined as QE =
• Depends on photon wavelength, incident angle (often ignored),

and hit location on photodetector

• Usually defined for a given WLS emission spectrum at a given

location on the photodetector or averaged over the entire surface.

• Convolution of 3 main processes:
• WLS conversion efficiency (depends on surface treatment,

temperature, environmental effects causing aging/degradation)

• Acceptance of WLS light at photosensor (depends on index
• f refraction, photosensor geometric coverage, surface quality)
• Photosensor QE/PDE (depends on temperature, operating

voltage)

N P E

meas/N ph pred

JINST 8 (2013) T07005 JINST 13 (2018) P02034

12/10/18 12

• M. Toups—LArTPC Caleb Reco Workshop

Photodetector QE Measurements: Three Methods

• Measure the convolution of process all at once, in situ
• Direct measurement of quantity of interest, but relies on poorly constrained quantities related to

light propagation (e.g. material reflectivity) and production (e.g. light yield, quenching)

• Measure/estimate each process individually, then combine
• Isolate and understand each effect individually, but brings in poorly understood quantities (e.g.

WLS conversion efficiency) and extrapolations (e.g. temperature)

• Cross-check/validation of the first method
• Measure relative QE w.r.t another absolutely-calibrated photodetector
• Difficult to use an identical flux of incident photons, so simulation is typically used to estimate

differences in the incident photon flux

• While the simulation still brings in poorly constrained quantities related to light propagation (e.g.

material reflectivity), this method does not depend on the predicted amount of light produced (unless it is brought in through the calibration of the reference photodetector)

12/10/18 13

• M. Toups—LArTPC Caleb Reco Workshop

14 Commercial Argon 1 ppm Oxygen, H2O Light Detection Test Stand (“Bo”)

Proton Assembly Building

12/10/18 14

• M. Toups—LArTPC Caleb Reco Workshop

PMT Calibration with Radioactive Source

2018 JINST 13 P02034 8” or 14.5” (5.3 MeV alpha) d = 8” d = 14.5”

averaged over TPB plate

Similar method used to measure ARAPUCA efficiency of 0.37 ± 0.01 (sys) ± 0.03 (stat)% in JINST 13 (2018) no.03, C03040 12/10/18 15

• M. Toups—LArTPC Caleb Reco Workshop

Light Guide Calibration with Cosmic Rays

NIMA 907 (2018) 9-21

“Blanche” test stand @ PAB

Commercial light guides with TPB-coated radiator plates

at readout end

12/10/18 16

• M. Toups—LArTPC Caleb Reco Workshop

Photodetector QE Take-aways

• Dedicated experiments requiring significant cryogenic infrastructure needed to

measure QE of just a few PD units (expensive)

• Surface and test beam LArTPCs can validate these PD QE measurements using

(tagged) cosmic rays or beam particles

• Scaling up in size to DUNE there are a few ways forward:
• Predict PD QE in LAr based on measurements of each process individually, then

validate with dedicated test stand experiments

• Similar to the above, but use relative measurement of convolution of all

processes to predict PD QE in LAr

• Need to identify appropriate physics sources (natural, radioactive sources, electron

beams?) to validate these PD QE measurements in DUNE (coupled to light yield)

• Can these be performed with drift HV off (removing HV integration requirements?)

12/10/18 17

• M. Toups—LArTPC Caleb Reco Workshop

Summary

• Gain measurements and stability monitoring is relatively

straightforward but important for understanding any variations in PD response seen over time

• Experiments to measure PD QE done in dedicated test stand

measurements (e.g. at PAB), but these are generally limited to just a few PD units

• Comparison with in situ measurements from currently operating

detectors (e.g. protoDUNE) important for understanding validity

• f these test stand measurements
• DUNE will require a large QC campaign of calibration

measurements to understand & measure PD QE before installation

12/10/18 18

• M. Toups—LArTPC Caleb Reco Workshop