Transcript ppt
Study of the Multi-Pixel Photon
Counter for ILC calorimeter
Satoru Uozumi (Kobe University)
BNM2008 @ Atami
• Introduction of ILC and MPPC
• The MPPC performance
• Calorimeter Prototype with MPPC
• Summary
The International Linear Collider and ILD
• e+e- collider with center-of-mass energy
at 500 ~ 1000 GeV.
• ILD (International Large Detector) is one of the
detector concepts proposed for the ILC experiment.
• Various precision measurements expected:
– e+e- g H, W, Z, tt, SUSY, etc …
g Multi-jets final states.
• Particle Flow Algorithm (PFA) allows precise
jet-energy measurement
.
ETOT = pe+ p + pcharged hadron + E + Eneutral hadron
[ tracks only]
[calorimeter only]
• Separation of jet particles in the calorimeter is required for the PFA
g Fine granular calorimeter is necessary.
The ILC Scintillator-Strip Calorimeter
• One approach for the fine granular calorimeter.
other approaches : silicon strip cal, digital cal
• Sampling calorimeter with W/Pb - scintillator sandwich structure.
• Scintillator stirp structure to achieve fine granularity
(strip size ~ 1 x 4.5 x 0.2~0.3 cm).
• Signal of all the strips are read out individually..
• Therefore the number of channels is huge
(~10M for ECAL, ~4M for HCAL).
• The calorimeter is placed inside 3 T
magnetic field.
Need small, cheap,
magnetic-field tolerant
photon sensor
while having high performance
comparable with conventinal PMTs.
The Multi-Pixel Photon Counter (MPPC)
- A Geiger-mode avalanche photo-diode with multi-pixel structure ~ 1 mm
Substrate
• Belongs to Pixelated Photon
Detector family (same as SiPM)
• Manufactured by Hamamatsu
Photonics.
• High Gain (105~106)
• Good Photon Detection
Efficiency (~15% with 1600 pixel)
• Compact
(package size ~ a few mm)
• Low Cost
• Insensitive to magnetic field
• Dark noise exists ( ~100 kHz)
• Input vs output is non-linear
We are developing and studying the
1600-pixel MPPC with Hamamatsu
for the ILD calorimeter readout.
What are required to the MPPC ?
• Gain, Photon Detection Efficiency (P.D.E.) comparable to PMTs.
– Gain at least 105
– P.D.E. ~ 20%
• Dark noise rate (due to thermal electrons) and
inter-pixel cross-talk probability as low as possible.
– Dark noise rate < 1 MHz, Cross-talk probability ~ a few per cent.
• Uniform performance over many pieces.
• Dynamic range enough to measure EM shower max.
– Electromagnetic shower is quite dense.
– Need dynamic range corresponds to 2000~5000 photoelectrons.
• Stability & Robustness.
Tolerance to temperature change, long-term use,
magnetic field and radiation.
• Low cost, compactness.
– Price order of $1~5, package size ~ 2 x 2 mm2.
• Time resolution ~ 1 ns
– Useful for bunch-ID, neutron separation
Gain, Dark Noise Rate, Inter-pixel Cross-talk
•30oC
•25oC
•20oC
•15oC
•10oC
•0oC
•-20oC
•
•
•
•
•
•
•
DV0/DT
= (56.0±0.1) mV/oC
1600 pixel
Over-voltage
– C … Pixel capacity
– V0 … Breakdown voltage
• Gain comparable to conventional PMTs.
• Dark noise rate ~100 kHz.
• Performance is temperature sensitive.
gtemperature control / monitoring is
important.
•
•
•
•
•
•
•
30 oC
25 oC
20 oC
15 oC
10 oC
0 oC
-20 oC
30 oC
25 oC
20 oC
15 oC
10 oC
0 oC
-20 oC
Gain – 800 pieces
Noise Rate (kHz)
Piece-by-piece Variation
Noise Rate
450 pieces
400 kHz
200 kHz
1
2
3
4
5
Over-voltage (V)
• Piece-by-piece variation is acceptably small.
g No need for further selection or categorization on massive use !
Just a small tuning of operation voltages is necessary.
• Further effort is ongoing by Hamamatsu to make the variation
even smaller.
Photon Detection Efficiency (PDE)
Measured by njecting same light pulse into both MPPC and PMT,
and comparing light yield.
WLSF LED
MPPC
PMT
0.5 mm f hole
PDEMPPC
N pMPPC
.e .
N
PMT
p .e .
PDE PMT
~ 16 %
MPPC
1600 pixel
PMT
The 1600-pixel MPPC has comparable P.D.E.
with normal photomultipliers (15~20%).
w
Response Curve
•If the recovery time is very long, MPPC
output is defined only by number of pixels.
•However if the recovery time is shorter than
input light, dynamic range may be enhanced.
1600 pix
Results
LED
PMT
MPPC
w = 50 ns
24 ns
1600
8 ns
1600pix
Simulation
Slow recovery
16 ns
1600
• Linearity of 1600 pixel MPPC is not limited by number of pixels
thanks to quick recovery time (~4ns).
• No significant influence from changing bias voltage.
• Time structure of the light pulse gives large effects
in non-linear region.
• Knowing time structure of input light is important.
Things done / not yet done
Performance
status
Gain
105~106
OK
Photon Detection Eff.
~0.2 for 1600 pix. MPPC
OK
Dark Noise Rate
~ 100 kHz
OK
Photon counting
Great
OK
Bias voltage
~ 70 V
OK
Size
Compact
OK
Dynamic range
Determined by # of pixels and recovery
time
underway
Cost
Expected to be < $10
Negotiating
Long-term Stability
Unknown
To be checked
Robustness
Unknown, presumably good
underway
Radiation hardness
Concerned
underway
B field
Expected to be Insensitive
Looks OK
Timing resolution
Expected to be 0.1~1 ns
To be checked
468 channels
In total
ECAL Prototype Performance
MPPCs (1600 pixels)
Tungsten
(3.5 mm thick)
Scintillator layer
The
calorimeter
with
Full MPPC readout
+
(3
mm
thick)
e
Frame
is proven toScintillator
work !strip
(1-6 GeV)
Energy Resolution for e+
(1 x 4.5 x 0.3 cm)
WLS fibre
Linearity
+ 1%
Summary & Prospects
• For the ILC calorimeter readout, study of the MPPC is
extensively ongoing collaborating with Hamamatsu.
• Measured performance of 1600 pixel MPPC is almost satisfactory
for the requirement:
–
–
–
–
Comparable gain / P.D.E. with photomultipliers.
Low noise rate (~100kHz) comparing with SiPMs.
Small piece-by-piece variation.
Short recovery time enhances the dynamic range for scintillator signal.
• The first EM calorimeter prototype with MPPC readout shows good and
reasonable performance.
• We are still working on further study and improvement.
– More number of pixels for more dynamic range.
– Need to check long-term stability, robustness, radiation hardness.
• The MPPC is a promising device which has lots of
excellent features !
The MPPC Line-up
Comparing with other Pixelated
Photon Detectors (PPD),
the MPPC has,
• Low dark noise
• High sensitivity to blue light
• Small device-by-device variation
Number of pixels
From HPK catalog
100
1600
1 x 1 mm2
Sensor size
Nominal Bias Volt.
400
70 10 V
77 10 V
Gain (x 105)
24.0
7.5
2.75
Noise Rate (kHz)
400
270
100
Photon Detection Efficiency
65 %
50 %
25 %
Temperature dependence (DV0/DT)
50 mV / oC
MPPC New Release Timeline
(informed at NSS Nov 2007 by Hamamatsu)
• 2007 Dec : 3x3 mm2 commercial sample
1x1mm2 SMD small package mechanical sample
2x2, 1x4 Array (3x3mm2) mechanical sample
• 2008 Jan : SMD small package commercial samples
• 2008 Apr : 3x3 mm2 product release
1x4 array
Array commercial samples
2x2 array
Backups
Recovery Time Measurement
Oscilloscope view
(with x63 amp)
Dt
1600 pixel
t (nsec)
Black … MPPC output for 1st Laser
Green … MPPC output for 2nd Laser
Red … Laser + LED
Blue … (Laser+LED) – Laser
= net response to 2nd laser
Ratio of Blue / Green gives
recovery fraction.
• Recovery time of the 1600-pixel MPPC
is measured to be ~ 4 ns.
• This number is consistent with
RC time constant of a pixel
(C ~ 20 fC, R ~ 200 kW, RC ~ 4 ns).
Excellent photon counting ability
0,1,2,3,4,5,6,7, . . . Photoelectrons !
1 photoelectron
2 photoelectrons
The MPPC has lots of advantages
Photomultiplier
MPPC
Gain
~106
105~106
Photon Detection Eff.
0.1 ~ 0.2
~0.2 for 1600 pix. MPPC
Response
fast
fast
Photon counting
Yes
Great
Bias voltage
~ 1000 V
~ 70 V
Size
Small
Compact
B field
Sensitive
Insensitive
Cost
Very expensive !
Not very expensive
Dynamic range
Good
Determined by # of pixels
Long-term Stability
Good
Unknown
Robustness
decent
Unknown, presumably good
Noise (fake signal by
thermions)
Quiet
Noisy (order of 100 kHz)
The MPPC is a promising photon sensor,
and feasible for the GLD Calorimeter readout !
Radiation hardness of MPPC (100 / 400 pixels)
Proton irradiation
(400 pixel MPPC)
Gamma-ray
100 pixel MPPC
Neutron
3.3 x 107 n /cm
1 x 1010 n /cm