Loss Monitoring Detectors Photon Detection and Silicon Photomultiplier Technology in accelerator and particle physics Sergey Vinogradov QUASAR group Department of Physics University ID: 365374
Download Presentation The PPT/PDF document "Beam" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Beam Loss Monitoring – Detectors Photon Detection and Silicon Photomultiplier Technology in accelerator and particle physics
Sergey Vinogradov QUASAR groupDepartment of Physics, University of Liverpool, UK Cockcroft Institute of Accelerator Science and Technology, UKP.N. Lebedev Physical Institute of the Russian Academy of Sciences, Russia
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide2
Content1. Introduction: the best photodetectors2. Silicon Photomultipliers (SiPM) as new photon number resolving detectors3. Benefits, drawbacks, and typical applications of SiPM4. Evaluation
studies of SiPMs for Beam Loss Monitoring5. Modelling and analysis of comparative performance: SiPM vs PMT and APD6. Trends and prospects of SiPM technology for BLM and accelerator applicationsSergey Vinogradov oPAC Advanced School on Accelerator Optimization Royal Holloway University, London, UK, July 9th, 2014
2Slide3
Introduction1. Introduction: the best photodetectors your choice?2. Silicon Photomultipliers (SiPM) as new photon number resolving detectors
3. Benefits, drawbacks, and typical applications of SiPM4. Evaluation studies of SiPMs for Beam Loss Monitoring5. Modelling and analysis of comparative performance: SiPM vs PMT and APD6. Trends and prospects of SiPM technology for BLM and accelerator applications
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014
3Slide4
Photodetector #1 *Sergey Vinogradov oPAC Advanced School on Accelerator Optimization Royal Holloway University, London, UK, July 9th, 2014 4
Adaptive focusing & trichromatic / monochromatic vision High sensitivity due to 100 million rod cells (10-40 photons)High resolution & double dynamic range due to 5 million cone cellsHigh readout rate of 30 frame/s
Internal signal processing (100M cells to 1M nerves @30fps)
540
million years
old design
(*) Yu.
Musienko
, NDIP 2011Slide5
CCD/CMOS approach – toward to #1Sergey Vinogradov oPAC Advanced School on Accelerator Optimization Royal Holloway University, London, UK, July 9th, 2014 5
Trichromatic
/ monochromatic
vision
Number of pixels up to ~ 50 M
Sensitivity
from
~ 10 -100 photons
Dynamic range up to ~ 50K
R
eadout up to ~ 1000 frame/s
40 years old designSlide6
SiPM approach – toward to ideal low photon detectionSergey Vinogradov oPAC Advanced School on Accelerator Optimization Royal Holloway University, London, UK, July 9th, 2014 6
20
th
Anniversary ~ now!Slide7
Silicon Photomultipliers (SiPM) as new photon number resolving detectors1. Introduction: the best photodetectors2. Silicon Photomultipliers (SiPM) as new photon number resolving detectors3. Benefits
, drawbacks, and typical applications of SiPM4. Evaluation studies of SiPMs for Beam Loss Monitoring5. Modelling and analysis of comparative performance: SiPM vs PMT and APD6. Trends and prospects of SiPM technology for BLM and accelerator applications
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014
7Slide8
Concept of ideal detector:first step to SiPMIdeal detector: conversion of any input signal starting from single photon to recognizable output without
noise and distortion in amplitude and timing of the signal
W. Farr, SPIE LEOS 2009
Ideal photon detector
R
eal photon detector
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide9
Single photon detectionin 1 GHz BW with electronic noise 104 e
σ
(N
out
)
σ
noise
Gain =1
ENF ~ 1
Gain ~ 100
ENF ~ 3…10
Gain ~ 1 M
ENF ~ 1.2
Gain ~ 1 M
ENF ~ 1.01
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide10
Avalanche with negative feedback: main step to SiPMStrong negative feedback = fast quenching & small charge fluctuations
Higher Field
V.
Shubin, D. Shushakov, Avalanche Photodetectors, 2003
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide11
Multi-pixel design & feedback resistor:final step to SiPM
Fig. 1-4: Sadygov, NDIP 2005 Fig. 5-7: B. Dolgoshein et al., 2001-2005
SiPM – 1996 / 2000s
MRS APD – 1990s
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide12
SiPM: photon number resolution
B.
K
ardinal et al., Nat. Photonics, 2008
R. Mirzoyan et al., NDIP, 2008
A. Barlow and J. Schilz, SiPM matching event, CERN, 2011
APD (
self-differencing
mode)
VLPC
SiPM (MEPhI/Pulsar)
SiPM (Excelitas)
PMT (Hamamatsu R5600)
I.
Chirikov-Zorin
et al, NIMA 2001
MPPC (Hamamatsu)
S. Vinogradov, SPIE 2011
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide13
Sergey Vinogradov Seminars on SiPM at the Cockcroft Institute 2 December 2013
13
G. Collazuol, PhotoDet, 2012Slide14
Benefits, drawbacks, and typical applications of SiPM1. Introduction: the best photodetectors2. Silicon Photomultipliers (SiPM) as new photon number
resolving detectors3. Benefits, drawbacks, and typical applications of SiPM4. Evaluation studies of SiPMs for Beam Loss Monitoring5. Modelling and analysis of comparative performance: SiPM vs PMT and APD
6. Trends
and prospects of SiPM technology for BLM and accelerator applications
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014
14Slide15
SiPM: photon number resolutionSiPM looks like ~ ideal detectorNear-ideal amplification: Gain > 105, ENF < 1.01Room temperature
Low bias (<100 V)Large area (6x6 mm2)Good timing (jitter < 200 ps)Fast response (rise <1 ns, fall~20 ns)In fact, not a photon spectrumPhotoelectronsDark electronsCrosstalk & Afterpulses
In fact, non-Poissonian distribution
Why?
How much?
Distribution?
Resolution?
P. Finocchiaro et al., IEEE TNS, 2009
A. Barlow and J. Schilz, SiPM matching event, CERN, 2011
SiPM (Excelitas)
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide16
SiPM drawbacks: crosstalkCrosstalk: hot carrier photon emission + detection = false event
A. Lacaita et al., IEEE TED, 1993
R. Mirzoyan, NDIP, 2008
Yu. Musienko, NDIP, 2005
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide17
SiPM drawbacks: afterpulsingAfterpulsing: trapping + detrapping + detection = false event
primary avalanche
afterpulses
Δ
time
Output
G. Collazuol, PhotoDet, 2012
C. Piemonte et al.,
Perugia,
2007
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide18
SiPM drawbacks: nonlinearityLimited number of pixels = losses of photonsDead time of pixels during recovery = losses of photons
Plot details:
Npixel
=100
PDE=100%
No Noise (DCR, CT, AP)
Ideal 100 pixel SSPM
Ideal photon detector
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide19
Application typesBinary detection of light
pulses – “events” (bit error rate) - NAPhoton number resolution (noise-to-signal ratio, σn/μ
n
) -
Calorimetry
Time-of-flight detection (transit time spread,
σ
t
) – TOF PET
Detection of arbitrary signals starting from photon counting -
I
ph
(t)
- Beam
Loss Monitoring
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide20
SiPM application examplesCalorimetrySiPM (MEPhI) small HCAL (MINICAL), DESY, 2003MPPC (Hamamatsu), T2K, 2005-2009MPPC at LHC CMS HCALRICH for ALICE (LHC)
FermiLab, Jefferson Lab calorimeter upgrade projectsAstrophysicsSiPM cosmic ray detection in space (MEPhI, 2005)Cherenkov light detection of air showers (CTA, 2013)Medical imaging Positron Emission Tomography:
TOF-PET
PET / MRI
Telecommunication
Quantum
cryptography
Deep space laser
link
(Mars
exploring program)
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide21
Evaluation studies of SiPMs for Beam Loss Monitoring1. Introduction: the best photodetectors2. Silicon Photomultipliers (SiPM) as new photon number
resolving detectors3. Benefits, drawbacks, and typical applications of SiPM4. Evaluation studies of SiPMs for Beam Loss Monitoring5. Modelling and analysis of comparative performance: SiPM vs PMT and APD
6. Trends
and prospects of SiPM technology for BLM and accelerator applications
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014
21Slide22
Beam Loss Monitoring (ref. E. Nebot talk 08-07-14)
Objectives:
Protect
Monitor
Adjust
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide23
BLM: first evaluation of SiPMD. Di Giovenale et al., NIMA, 2011 SPARC accelerator, Frascati
, INFN FERMI@Elettra, Synchrotrone TriesteMPPC, 1mm2, 400 pixelsQuartz fiber 300 μm, 100 mDark count noise: negligibleElectronic noise: negligible
Spectral dispersion in fiber:
n(𝜆) →∆t(𝜆)
~ 3 ns @100 m
τ
fall
~ 10 ns
→
deconvolution
Compact low cost BLM
1m-scale resolution @100 m
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide24
SiPM performance metrics for BLM
Loss scenarios
reconstruction
Amplitude → # photons → # particles per location (PNR)
Transit time to rising edge → single loss location (Time Res.)
Resolution of multiple loss locations & # particles
Modulation transfer function (MTF) ?
Nonlinearity has to be accounted !
PNR
TTS
New metrics ?
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide25
Challenges for SiPM in BLM:saturation, recovering, duplicationsTransient nonlinearity of SiPM responseLarge rectangular light pulse: Nph > Npix; Tpulse >
TrecPeak – initial avalanche events in ready-to-triggering pixelsPlateau – repetitive recovering and re-triggering of pixelsFall – final recovering (without photons, but with afterpulses!) 4 us pulse 50 ns pulse
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide26
MPPC response on rectangular pulse
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide27
Modelling and analysis of comparative performance: SiPM vs PMT and APD1. Introduction: the best photodetectors2. Silicon Photomultipliers (SiPM) as new photon number resolving detectors
3. Benefits, drawbacks, and typical applications of SiPM4. Evaluation studies of SiPMs for Beam Loss Monitoring5. Modelling and analysis of comparative performance: SiPM vs PMT and APD
6. Trends
and prospects of SiPM technology for BLM and accelerator applications
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014
27Slide28
Photon Number ResolutionPhoton Number Resolution & Excess Noise FactorBurgess variance theorem
Ideal 100 pixel SSPM
Ideal photon detector
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide29
Schematics of AP & CT stochastic processes
Branching Poisson Crosstalk Process
Geometric Chain Afterpulsing
Process
Poisson number of primaries <N>=
μ
e.g. SSPM Photon Spectrum
Single primary event N≡1
e.g. SSPM Dark Spectrum
Duplication Models
Non-random (Dark) event
Primary
1
st
CT
2
nd
CT
No CT
Random CT events
…
Random CT events
Random Photo events
Random primary (Photo) events
Random CT events
…
…
…
Non-random (Dark) event
Random CT events
…
…
…
…
…
…
Random primary (Photo) events
Random CT events
…
…
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide30
Analytical results for CT & AP statistics
CT & AP
model results
[1] S. Vinogradov et al., NSS/MIC 2009
λ
is a mean number of successors in one branch generation
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide31
A few photon detection spectrum of Hamamatsu MPPC
(S. Vinogradov, SPIE 2012)
Dark event complimentary cumulative distribution – DCR vs. threshold
(S. Vinogradov, NDIP 2011; experiment B. Dolgoshein et al., NDIP 2008)
Pct=40%
Pct=10%
P. Finocchiaro et al., IEEE TNS, 2009
A.N. Otte, JINST 2007
Crosstalk models and experiments
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide32
<Ns>=f(<Nph>)
SiPM binomial nonlinearity
Intrinsic Resolution in
σ
units; Npix=506
Photons per pulse, Nph
B. Dolgoshein et al., 2002
Output signal, Ns
E.B. Johnson
, NSS/MIC 2008
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide33
SiPM recovery nonlinearity
Plot details:
PDE=100%
Npixel=500
Tpulse=100 ns
Nonparalizible
dead time model
Probability distribution (~ Gaussian)
W. Feller,
An Introduction to Probability Theory and Its Applications,
Vol. 2, Ch. XI, John Willey & Sons, Inc., 1968
Recovery non-linearity → ENF
S. Vinogradov et al., IEEE NSS/MIC 2009
Exponential recovery of Gain m(t)
accounting for
Pdet
(m)
S.
Vinogradov,
SPIE DSS, 2012
M.Grodzicka, NSS 2011
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide34
Performance metrics: ENF and DQE
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide35
Performance in DQE – various detectors
Fm
PDE
Gain
Gain
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide36
Time resolutionTime resolution is combined as a sum of contributionsTransit time spread of photon arrival, avalanche triggering, avalanche development, and single electron response timesJitter of signal amplitude fluctuations in a time scale
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide37
Filtered point process approach to amplitude fluctuations & time resolution
Sergey Vinogradov oPAC Advanced School on Accelerator Optimization Royal Holloway University, London, UK, July 9th, 2014 Slide38
Clastered filtered point process modelTime resolution includes all essential factors and combines performance in time response and PNR (ENF)
Sergey Vinogradov oPAC Advanced School on Accelerator Optimization Royal Holloway University, London, UK, July 9th, 2014 Slide39
Time resolution:scintillation model & experiment
Most popular & demanded case study: LYSO+MPPC
LYSO
: 0.09 ns rise, 44 ns decay; 9% resolution
MPPC:
Npe
=3900,
ENFgain
=1.015,
Pct
=0.14; SPTR=0.124 ns,
Vnoise
=0.32 mV
S. Seifert et al, “A Comprehensive Mode to Predict the Timing Resolution ”, TNS, 2012.
MPPC SER pulse shape – analytical expression
(~ 1 ns rise, ~ 25 ns decay)
D.
Marano
et al, “Silicon Photomultipliers Electrical Model: Extensive Analytical Analysis” TNS 2014
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide40
Arbitrary signal detection:rectangular pulse response modelSergey Vinogradov oPAC Advanced School on Accelerator Optimization Royal Holloway University, London, UK, July 9th, 2014 40Slide41
Trends and prospects of SiPMs for BLM and accelerator applications1. Introduction: the best photodetectors2. Silicon Photomultipliers (SiPM) as new photon number
resolving detectors3. Benefits, drawbacks, and typical applications of SiPM4. Evaluation studies of SiPMs for Beam Loss Monitoring
5. Modelling
and analysis of comparative performance: SiPM vs PMT and APD
6. Trends
and prospects of SiPM technology for BLM and accelerator applications
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014
41Slide42
SiPM trends and advancesMarket leadersHamamatsuKETEKSensL
FBK / AdvanSiDExcelitas / Perkin ElmerPhilips (digital SiPM)Design improvements (~ in a few year time scale)Higher Photon Detection Efficiency (30% → 70%)Lower crosstalk, lower afterpulsing (30% → 3%)
Lower dark count rate (1000 → 40
Kcps
/mm
2
)
Faster SER, smaller pixel size (25 → 10 um)
Larger area, larger arrays (3x3→10x10mm
2
,
4x4 →
16x16 channels)
Latest
news
from
2
nd
SiPM Advanced
Workshop
and
Conf
. on New Development s in Photodetection,
2014
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide43
Hamamatsu: Through Silicon Vias
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide44
Hamamatsu: Low Crosstalk & Afterpulsing
Sergey Vinogradov oPAC Advanced School on Accelerator Optimization Royal Holloway University, London, UK, July 9th, 2014 Slide45
Performance in DQE - MPPC series
Pulse duration & detection time = 10 ns
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide46
KETEK
Highest PDE @50 um pixels
Various geometries
15 … 100 um pixels
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide47
SensLFast capacitive outputFWHM < 3.2 ns @ 6x6 mm2Large arrays / modulesLow cost
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide48
Philips: Digital SiPM (Modern active quenching SPAD array)
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide49
Summary on BLM with SiPMBLM is one of the most challenging application for SiPMBenefitsPractical & efficient (cost, compactness, Si reliability…)Perfect Transit Time Resolution (as is for now)
Acceptable DQE within dynamic range (may be better)DrawbacksUpper margin of dynamic range is low (design improvement)Number / density of pixels (↑ 10 times)Pixel recovery time (↓ 10 times)Time response (bandwidth) (external measures)Analog / digital SiPM output signal processingBLM with SiPM: big problem with a chance to winAnd with a lot of space for new ideas, designs, and fun
Sergey Vinogradov Seminars on SiPM at the Cockcroft Institute 3 February 2014
49Slide50
Summary on SiPM
SiPM technology: breakthrough in photon detectionPhoton number resolution at room temperatureSilicon technology / mass production / reliability / priceHighly competitive in short (< μs) pulse detectionFast progress in improvements: DQE, Dynamic range, TimingWelcome to SiPM applications
Scintillation
Cherenkov
Laser pulse
And much more…
Sergey Vinogradov
oPAC
Advanced School on Accelerator Optimization Royal Holloway University, London, UK,
July 9
th
, 2014 Slide51
The endThank you for your attentionQuestions?Sergey.Vinogradov@liv.ac.uk
Sergey Vinogradov oPAC Advanced School on Accelerator Optimization Royal Holloway University, London, UK, July 9th, 2014 51