1 of 60

Q-Pix: Kiloton-scale pixelated liquid noble TPCs

Jonathan Asaadi (on behalf of the Q-Pix consortium)

Q-Pix consortium would like the thank the DOE for its support via DE-SC0020065 award, �DE-SC 0000253485 award, and FNAL-LDRD-2020-027

Work based on original paper by Dave Nygren (UTA) and Yuan Mei (LBNL): arXiv:1809.10213

​

2 of 60

Why pixelated liquid noble TPCs?

  • Noble Element Time Projection Chambers offer access to very high quality information
  • Leveraging this information allows unprecedented access to detailed neutrino interaction specifics from MeV - GeV scales
  • Capturing this data w/o compromise and maintaining the intrinsic 3-D quality is an essential component of all readouts!
  • Conventional charge readout uses sets of wire planes at different orientations to reconstruct the 3D image
    • Challenge in reconstruction of some topologies
  • Pixel based charge readout is a natural solution

2

3 of 60

Liquid Argon Time Projection Chamber

3

Intrinsic reconstruction pathologies associated with charge deposited along the direction of the wires

Imagine a p→K𝜈 candidate event where you got unlucky and this occurred

4 of 60

Introduction

4

  • kiloTon scale LArTPC’s use “wrapped wire” geometries to reduce the number of readout channels
    • Challenging to engineer such massive structures
    • Possible ambiguities associated with the readout increase with the wrapped geometry
    • Wire failure poses risk to loss of readout of an entire APA
      • Requires extensive (expensive) QA/QC
  • The number of events in future large scale experiments are few and precious
    • Don’t want to lose any to readout/reconstruction

5 of 60

Big detectors = Big Data (but not all useful)

  • Kiloton scale LArTPC’s (such as DUNE) afford a huge “big data” challenge to extract all the details offered by LArTPC
    • 1 second of DUNE full stream data �~4.6 TB (for 1.5 million channels)
      • 1 year of full stream data ~ 145 EB (exabytes)
  • However, most of the time there is “nothing of interest” going on in the detector
    • But you must be ready “instantly” when something happens
    • Requires an “unorthodox” solution

5

6 of 60

What is to be gained? (3D vs 2D Readout)

​

6

  • Using pixels instead of wires can solve the shortcomings of projective wire readout
  • How much better is a pixel detector?
    • In a perfect world we would compare the complete readout and reconstruction of two detectors side by side to do this
      • Such a chain doesn’t exist!

​

Credit: arxiv: 1903.05663

  • Using modern machine learning techniques, we trained two parallel networks on identical simulated neutrino interactions
      • Network 1: 3D pixel based readout
      • Network 2: 2D projective wire readout
        • For both of these networks, we assumed ideal detectors with perfect response and no reconstruction pathologies

​

7 of 60

What is to be gained? (3D vs 2D Readout)

7

  • Simulation studies comparing the readout of 2D projective Liquid Argon TPCs (LArTPC’s) to 3D pixel LArTPC’s shows that 3D based readout offers significant improvement in all physics categories!
    • 𝝂e-CC inclusive: 17% gain in efficiency and 12 % gain in purity
    • 𝝂𝜇-CC inclusive: 10% gain in efficiency for 99% purity
    • NC𝜋0: 13% gain in efficiency and 6% gain in purity
    • Also offers gains in Neutrino-ID classification and final state topology ID

*** Improvements like these can lead to significantly shorter experimental running time required to meet desired physics goals!

JINST 15 P04009 (arXiv:1912.10133)

8 of 60

So pixelate them, what’s so hard?

8

  • Readout of a TPC using pixels instead of wires comes at the “cost” of many more channels (Example: 2 meter x 2 meter readout)
    • 3mm wire pitch w/ three planes = 2450 channels
    • 3mm pixel pitch = 422,000 channels
  • LArPix (JINST 13 P10007) readout has pioneered this frontier showing a low power pixel based readout can be done
    • Currently targeted to the DUNE near detector to allow a LArTPC to cope with the high event rates
    • Other solutions are being explored for kiloton scale underground LArTPCs

(JINST 13 P10007)

9 of 60

9

Why other solutions?: Scale of the detectors

One 10kT DUNE LArTPC Module (18 m x 19 m x 66 m)

¼ the total size of DUNE�𝓞 (130 million) 4mm pixels

One 300T DUNE-ND LArTPC Module (11m x 8 m x 7 m)

𝓞 (7 million) 4mm pixels

~18x more channels Far/Near

10 of 60

Scale of the detectors

10

Estimated event rates in the DUNE LArTPC Near Detector (ArgonCube) and a single DUNE 10kTon Far Detector Module

  • 105 - 106 difference in event rate from beam events near/far
  • Same number of events from the beam as from astrophysical sources
    • Spans 102 MeV energy range

Scaling pixel based readout to the multi-kiloton detector may require an “unorthodox” solution

11 of 60

An “unorthodox” solution

  • The Q-Pix pixel readout follows the “electronic principle of least action”
    • Don’t do anything unless there is something to do
      • Offers a solution to the immense data rates
        • Quiescent data rate 𝓞(50 Mb/s)
      • Allows for the pixelation of massive detectors
  • Q-Pix offers an innovation in signal capture with a new approach and measures time-to-charge:(ΔQ)
    • Keeps the detailed waveforms of the LArTPC
    • Opens the door to other innovative technical solutions
  • “Novelty does not automatically confer benefit”
    • Much remains to be explored

11

12 of 60

Q-Pix: The Charge Integrate-Reset (CIR) Block

  • Charge from a pixel (In) integrates on a charge sensitive amplifier (A) until a threshold (Vth~ΔQ/Cf) is met which fires the Schmitt Trigger which causes a reset (Mf) and the loop repeats

12

“reset” switch

Charge �sensitive Amp.

Schmitt Trigger

13 of 60

Q-Pix: The Charge Integrate-Reset (CIR) Block

  • Measure the time of the “reset” using a local clock (within the ASIC)
  • Basic datum is 64 bits
    • 32 bit time + pixel address + ASIC ID + Configuration + ...

13

14 of 60

Reset Time Difference

14

15 of 60

Toy Example

15

e

e

e

e

e

e

e

e

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

Reset Threshold

16 of 60

Toy Example

16

e

e

e

e

e

e

e

e

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

Reset Threshold

17 of 60

Toy Example

17

e

e

e

e

e

e

e

e

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

Reset Threshold

18 of 60

Toy Example

18

e

e

e

e

e

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

e

Reset Threshold

19 of 60

Toy Example

19

e

e

e

e

e

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

e

Reset Threshold

20 of 60

Toy Example

20

e

e

e

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

e

e

e

Reset Threshold

21 of 60

Toy Example

21

e

e

e

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

Reset Threshold

22 of 60

Toy Example

22

e

e

e

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

Reset Threshold

23 of 60

Toy Example

23

e

e

e

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

Reset Threshold

24 of 60

Toy Example

24

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

e

Reset Threshold

25 of 60

Toy Example

25

e

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

e

Reset Threshold

26 of 60

Toy Example

26

Voltage = Q/C

Time

e

e

e

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

e

e

Reset Threshold

27 of 60

Toy Example

27

Voltage = Q/C

Time

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

e

e

e

e

e

Reset Threshold

28 of 60

Toy Example

28

Voltage = Q/C

Time

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

Reset Threshold

29 of 60

Toy Example

29

Voltage = Q/C

Time

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

Reset Threshold

30 of 60

Toy Example

30

Voltage = Q/C

Time

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

Reset Threshold

31 of 60

Toy Example

31

Voltage = Q/C

Time

Time

Resets

Note: We’ll assume the RTD happens for 5 electrons, the reset happens faster than the drift of the next bunch, and this occurs without charge loss

e

e

Reset Threshold

32 of 60

Toy Example

32

Time

Resets

e

e

What I have here is a fixed amount of charge ΔQ (10 electrons in our toy example) during a time Δt

​

This gives me a current seen by the pixel during this time!

33 of 60

What is new here?

  • Take the difference between sequential resets
    • Reset Time Difference = RTD
  • Total charge for any RTD = ΔQ
  • RTD’s measure the instantaneous current and captures the waveform
    • Small average current (background) = Large RTD
      • Background from 39Ar ~ 100 aA
    • Large average current (signal) = Small RTD
      • Typical minimum ionizing track ~ 1.5 nA
  • One free running clock per ASIC (10-100 MHz)
    • Required precision for DUNE δf/f ~10-6 per second (should be easily achievable)
  • Time stamping routine has the ASIC asked once per second “what time is it?”
    • ASIC captures local time and sends it
    • Simple linear transformation to master clock synced to GMT
    • RTD’s calculated “off chip”

33

34 of 60

34

ΔQ~1.0 fC (~6000 e-)

Nygren & Mei arXiv:1809.10213

35 of 60

35

ΔQ~0.3 fC (~1800 e-)

Nygren & Mei arXiv:1809.10213

36 of 60

Q-Pix ASIC Concept

  • 16-64 pixels / ASIC
    • 1 Free-running clock/ASIC
    • 1 capture register for clock value, �ASIC, pixel subset
    • Necessary buffer depth for beam/burst events
    • State machine to manage dynamic network, token passing, clock domain crossing, data transfer to network (many details to be worked out)
  • Basic unit would be a “tile” of ASICs (4092 4mm x 4mm pixels)
    • Tile size 25.6 cm x 25.6 cm

36

37 of 60

Data Rates for 10 kTon

37

  • We imagine each tile is 16x16 ASICs and one readout plane (APA) has 11,136 tiles per APA

One 10kT DUNE LArTPC Module (18 m x 19 m x 66 m)

¼ the total size of DUNE

  • ���
  • We perform the clock calibration 1/second (perhaps less often)

-- This gives 16,384 bits / tile

  • The total data rate is thus set by the number of readout planes
    • 3.5 meter drift = 4 APA’s = 16,384 bits/tile x 44,552 tiles ~ 90 Mbytes/s
      • Full burst of supernova events w/ radiological backgrounds in simulation only adds ~24 Mbytes/s of data

38 of 60

Q-Pix Consortium

  • A consortium of universities and labs has formed to realize and test the Q-Pix concept
    • Being done in close collaboration with LArPix (JINST 13 P10007) readout for the DUNE near detector

38

39 of 60

Q-Pix Consortium

39

  • Five central ideas being worked on
    • Circuit Design: A prototype of the front-end + oscillator is expected to be completed in the next 6 months
    • Physics Simulations: Quantify the conferred benefit of pixel vs. wire readout, the requirements of the ASIC design, and study the capabilities enabled by the design
    • CIR Input: all extraneous leakage current at the input node needs to be small (aA)
    • Clock: δf/f ~10-6 per second
    • Light Detection: Exploring new ideas using photoconductors on the surface of the pixels

40 of 60

Opportunities for Q-Pix at ORNL

  • ORNL SNS offers a unique neutrino source for the study of low energy neutrino physics!
    • Combination of a unique beam and a growing facility makes this an attractive place for physics demonstrator for the Q-Pix readout
  • Enables unique neutrino measurements which DUNE will benefit from
    • 𝝂 - e scattering: Directionality with the SNS source
      • 17 events per year
      • Validates/tests our ability to do pointing with low energy electrons
      • First measurement of its kind on argon!
    • Low Energy 𝝂e - Ar Charged Current Interactions
      • 758 events per year
      • Experimental test of x-sections for supernova 𝜈’s
      • First measurement of its kind on argon!
    • Low Energy 𝝂 - Ar Neutral Current Interactions

40

Credit: Yun-Tse Tsai (SLAC) for help with these plots

3 MeV Electron 100 cm from the pixel plane

Instantaneous Current on the pixel�Reset Time Difference (RTD)

Cumulative Current on the pixel�Reconstructed Cumulative Current

Very interested in brainstorming ideas with interested parties!

41 of 60

View from above

41

50 cm

200 cm

25� cm

25� cm

25� cm

25� cm

LAr Active Volume

Cryostat

𝝂

𝝂

𝝂

𝝂

𝝂

𝝂

𝝂

𝝂

𝝂

𝝂

Beam from the SNS

  • 1m3 detector which might be able to fit into neutrino alley could be 2m (L) x 0.5m (w) x 1.0m (h) active volume
    • Cryostat would be 2.5m (L) x 1m (w) x 1.5 m (h)
  • Cathode in the middle giving 1 meter drift
    • 50 kV on the cathode (very doable)

This could be readout with 50 cm x 50 cm tiles with 4 mm pixel spacing

→ 62,500 pixels in total for both readout planes

42 of 60

What is the physics you can do with 1m3 at SNS

  • This obviously depends on what your detector is capable of….
  • But for some reasonable assumptions about the threshold charge an early version Q-Pix could achieve:
    • RTD 𝚫Q: ~ 1fC ~ 6000 electrons
      • Possible to go lower!
    • Minimum number of RTD’s to separate activity from radiogenic background (39Ar, Rn, Bi-Po): �3 RTD’s in 6 𝜇s
    • Assumed 100 aA leakage current
    • Assumed 500 ENC noise
  • MeV electrons easily reconstructable

42

1 fC (~6000 electrons)

⅔ fC (~4000 electrons)

43 of 60

Light Detection

  • Conventional LArTPC’s use the semi-transparent wires to their advantage and place their photon detectors behind the wire planes
    • Requires WLS
  • Pixel detectors have an opaque charge collection surface making use of this solution impossible
    • Alternative mounting schemes have been / are being explored
    • Charge only methods of reconstructing t0 also being explored
  • How do you turn a vice into a virtue?

43

44 of 60

Pixels which also are photo-sensitive?

  • What if the whole APA could collect light?
  • A pixel plane sensitive to UV photons and ionization charge SIMULTANEOUSLY would be a major breakthrough
    • Your effective instrumented area becomes enormous!
    • Even if the device has low efficiency you have a huge gain
    • Q-Pix could be an “enabling technology” to realize this for LArTPC’s

​

44

45 of 60

Light Detection (conceptual sketch)

​

45

46 of 60

Literature search suggests the absorption coefficient for a-Se at 128nm is 130 μm-1 (hasn’t been measured since the 1960’s)

  • This would suggest a 1𝜇m thick thin film would already be >99%

46

Amorphous Selenium

​

QE for converting 128 nm light to charge!

  • Moreover, gain in the a-Se is possible with the application of moderate E-field
    • Early calculations suggest 𝓞(100) - 𝓞(1000) electrons per pixel pad for 𝓞(MeV) levels of activity
  • The transport properties of a-Se also look favorable for mobilities over mm without significant leakage current
    • DEVIL LIVES IN THE DETAILS

47 of 60

Amorphous Selenium

​

Prototype board’s are being prepared at UTA and ORNL to test the viability of this idea in liquid argon and with VUV light

�Partnered with E. Gramellini (FNAL) with an LDRD and M. Febbraro (ORNL)

​

47

48 of 60

Opportunities for Q-Pix at ORNL

  • There are a number of other photoconductor candidates to be explored and working in the context of a neutrino detector at ORNL would allow further engagement with their photon detection / material science expertise
    • a-Se: Direct light to charge
    • ZnO: Pyro-photonic devices
    • CdTe: Nano-particle photoconductors
    • ….(many more being thought about)
  • If found to be viable, demonstration with low energy neutrinos (ala SNS) could be a critical step to their overall adoption

48

49 of 60

Conclusions

  • Q-Pix is a new technology which is suited for large detectors the principal of electronic least action is need
    • “Don’t do anything unless there is something to do”
    • “Be ready instantly and retain all the information”
  • Q-Pix will be capable of doing all of the same physics as “vanilla DUNE” with intrinsic 3D readout, significantly lower data rates, continuous untriggered readout
    • A more detailed study is underway and will be incorporated into the Q-Pix white paper.
  • The prospects for a Q-Pix based LAr detector at the SNS appears promising and studies are ongoing
  • The Q-Pix concept may afford a way to pixelize a kiloton scale LArTPC and retain all the details of data
    • The devil lives in the details, but an effort is underway with promising preliminary results

49

50 of 60

Thank you!

50

Q-Pix consortium would like the thank the DOE for its support via DE-SC0020065 award, �DE-SC 0000253485 award, and FNAL-LDRD-2020-027

​

51 of 60

Backup

51

52 of 60

Current light collection designs

  • Limited real estate for Light Collection System in the wire planes
  • How much of the available APA has photon detection capability?
    • X-Arapuca Design: 130 m2/10kT
      • Window area for each supercell (435.24 cm2) x 10 supercells/APA x 152 APA’s per 10kT x 2 (Double sided)
    • APA Active Area: ~200,000 m2/ 10kT
      • (135,700 cm2) * 152 APAs/10kT
  • Surface area instrumented is ~ 0.06%
    • Actually less when you take efficiency of the device into account

​

52

*** Not meant to disparage the current technology in any way...instead meant to give context to the problem

53 of 60

Physics Simulation

  • Measurement of Longitudinal Diffusion
    • Using a small sample muons a novel technique in Q-Pix can be seen

53

Calculation from: https://arxiv.org/pdf/1508.07059.pdf

54 of 60

Physics Simulation

  • Measurement of Longitudinal Diffusion
    • The average RTD versus the drift length yields a distribution which carries the diffusion information along with it
    • Allows for a fundamental measurement with few statistics
  • DLMeasured = 6.47土0.97 cm2/s
    • DLSimulation = 6.82 cm2/s

54

55 of 60

55

By looking at the RMS distribution of the RTD’s you can work the problem the other way!

  • Assume the diffusion constants
  • Drift speed is known (at a given field)
  • Solve for the reconstructed drift distance

​

This could (potentially) allow you to reconstruct an event’s t0 without a photon dector

56 of 60

Q-Pix response to the “higher limit” of the expected physics. This being a 500 MeV proton shot in the direction of the readout board.

56

57 of 60

Q-Pix digital logic test (running on an FPGA). Shows a uniform time response to readout all resets stored in the local buffers.

57

58 of 60

UTA/H TPC’s

The UTA/H TPC is designed to efficiently collect charge and light. This is done with an asymmetric design requiring a buffer region to protect the PMT’s.

​

58

59 of 60

UTA/H TPC’s

The UTA/H TPC is designed to efficiently collect charge and light. This is done with an asymmetric design requiring a buffer region to protect the PMT’s.

The drift region is 35cm long and is designed to handle drift fields up to 1.5kv/cm. Outfitted with HDPE reflector tubes which will be coated in TPB in order to maximize the light collection.

This TPC will allow is to test Q-Pix for functionality and explore how it responds to “low” energy (~100s of keV)

59

60 of 60

UTA/H TPC’s

The TPC is designed to efficiently collect charge and light. This is done with an asymmetric design requiring a buffer region to protect the PMT’s.

The drift region is 35cm long and is designed to handle drift fields up to 1.5kv/cm. It is also outfitted with HDPE reflector tubes which will be coated in TPB in order to maximize the light collection.

This TPC will allow is to test Q-Pix for functionality and explore how it responds to “low” energy (~100s of keV)

60