Development and Study of the Multi Pixel Photon Counter Satoru - PowerPoint PPT Presentation

Development and Study of the Multi Pixel Photon Counter Satoru Uozumi (Shinshu University, Japan) for the KEK Detector Technology Project / Photon Sensor Group Vienna Conference on Instrumentation 2007, Feb 19-24 , 2007 1. Introduction 2. Fundamental Performance 3. Application to High Energy Physics 4. Summary

KEK Detector Technology Project Photon Sensor Group (http://rd.kek.jp/) (KEK, Kobe, Kyoto, Nagoya, Nara-WU, NDA, Niigata, Shinshu, Tokyo/ICEPP, Tsukuba) S. Gomi, H. Hano,T. Iijima, S. Itoh, K. Kawagoe, S. H. Kim, T. Kubota, T. Maeda, T. Matsumura, Y. Mazuka, K. Miyabayashi, T. Murakami, T. Nakadaira, T. Nakaya, H. Otono, E. Sano, T. Shinkawa, Y. Sudo, T. Takeshita, M. Taguchi, T. Tsubokawa, S. Uozumi, M. Yamaoka, H. Yamazaki, M. Yokoyama, K. Yoshimura, T. Yoshioka And special thanks to Hamamatsu photonics K.K.

The Multi Pixel Photon Counter (MPPC) - A silicon avalanche photo-diode with multi-pixel structure - Si Resistor 1 mm V bias Guard ring n - p + n + p - Substrate • Each pixel works as a Geiger-mode APD, substrate p + • One pixel can count only one photon. � need multi - pixel structure for photon counting 400 pixel • Electric charges from all the fired pixels are summed up 50 µ m and read out as a signal. • There are 4 different types available: # of pixels Sensor size Pixel size Geometrical eff. 1600 pixel 100 ~ 0.65 100 µ m 1x1 mm 2 400 ~ 0.5 50 µ m 1600 ~ 0.25 25 µ m

Excellent Photon Counting Ability 0,1,2,3,4,5,6,7, . . photoelectrons ! 1600 pixel 1600 pixel 1 0.8 1 photoelectron Entries 0.6 2 photoelectrons 0.4 0.2 0 100 200 300 400 500 600 MPPC signal (ADC counts)

The MPPC has lots of advantages Photomultiplier MPPC Gain ~10 6 10 5 ~10 6 Photon Detection Eff. 0.1 ~ 0.2 0.2 (1600pix.) ~ 0.5 (100pix.) Response fast fast Photon counting Yes Great Bias voltage ~ 1000 V ~ 80 V Size Small Compact B field Sensitive Insensitive Cost Expensive Not expensive Dynamic range Good Determined by # of pixels Long-term Stability Good Unknown Robustness decent Unknown Noise (fake signal by Quiet 1 pixel noise exist thermions) (order of 100 - 500 kHz)

Fundamental performance • Gain • Dark Noise Rate • Inter-pixel Cross-talk • Photon Detection Efficiency • Uniformity in a pixel

Fundamental Performance - Gain •30 o C •25 o C 22 ] 5 100 pixel •20 o C Gain [ x10 9 •15 o C 18 Gain (x10 ) 8 5 •10 o C 7 •0 o C 14 Gain 6 •-20 o C =10 6 10 5 •25 o C 4 •20 o C 6 •15 o C 3 2 68.6 69 69.4 1600 pixel Bias Voltage (V) 1 74 75 76 77 78 79 80 81 Bias voltage [V] 400 pixel 14 Gain (x10 ) 12 5 Gain = 10 6 10 Over-voltage – C … Pixel capacity 8 – V 0 … Breakdown voltage •25 o C 6 •20 o C 4 • V 0 temperature ( Δ V 0 / Δ T ~ 50 mV/C o ) •15 o C • Larger pixel size results in larger gain. 69.2 69.6 70 70.4 70.8 Bias Voltage (V)

Variation of V 0 and C over 750 MPPCs (Measured at 15 o C) 60 60 Variation Variation Number of samples Number of samples ~ 0.45 V < 4% 40 40 20 20 0 0 72 73 74 75 0.02 0.022 0.024 0.026 Breakdown voltage (V) Pixel capacity (pC) • ~750 pieces of 1600 pixel MPPCs have been tested. • Device-by-device variation is less than a few %. � No need for further selection or categorization on massive use ! Just need a small tuning of operation voltages.

Fundamental Performance – Dark Noise Rate • 30 o C 1600 pixel 500 100 pixel Noise Rate (kHz) • 25 o C 400 400 • 20 o C Noise Rate (kHz) 300 kHz • 15 o C 300 300 • 10 o C 300 kHz 200 • 0 o C 200 • -20 o C 100 100 0 •25 o C 0.2 0.6 1 1.4 •20 o C V bias - V 0 (V) 1.5 2 2.5 3 3.5 4 4.5 5 •15 o C 400 pixel V - V (V) 600 bias 0 Noise Rate (kHz) • The dark noise is caused by thermal electrons. 400 • Its rate depends on both over-voltage and temperature. 300 kHz 200 • More number of pixels � smaller active area 0 0.5 1 1.5 2 2.5 � fewer noise rate V bias - V 0 (V)

Fundamental Performance - Inter-pixel Cross-talk - 0.5 Cross-talk probability 100 pixel 0.4 0.3 • Inter-pixel cross-talk is caused by •25 o C a photon created in an avalanche 0.2 •20 o C • Probability of the cross-talk has been 0.1 •15 o C measured using dark noise rates: 0 0.2 0.6 1 1.4 V bias - V 0 (V) 0.5 400 pixel Cross-talk probability 0.4 • 30 o C 0.3 1600 pixel • 25 o C Cross-talk probability •25 o C • 20 o C 0.2 0.3 •20 o C • 15 o C 0.1 •15 o C • 10 o C 0.2 0 • 0 o C 0.5 1 1.5 2 2.5 V bias - V 0 (V) • -20 o C 0.1 • Cross-talk probability is affected by over-voltage, but not affected by 0 1.5 2 2.5 3 3.5 4 4.5 5 temperature. V - V (V) bias 0

Fundamental Performance - Photon Detection Efficiency (P.D.E) - • Q.E. (~ 0.9) … Quantum Efficiency • ε Geiger (~ up to 0.9 , depends on bias voltage) … Probability to cause avalanche • ε geom (0.25 ~ 0.65 , depends on pixel size) … Fraction of sensitive region in a sensor Measurement of relative P.D.E. • Inject same light pulse into both the MPPC and the PMT, and compare light yield measured by both: LED ( λ =450~550nm) MPPC ~15 % PMT hole

Fundamental Performance – Photon Detection Efficiency - 2.2 PDE ratio (= MPPC/PMT) 100 pixel 2 1 . 4 1600 pixel Relative PDE 1.8 •25 o C 1 . 2 •20 o C 1 1.4 1 •15 o C 0 . 8 1 1 0 . 6 0.6 0 . 4 0.2 0.6 1 1.4 0 . 2 V bias - V 0 (V) 0 2 2 . 5 3 3 . 5 4 4 . 5 5 5 . 5 2.2 V bias - V 0 (V) 400 pixel 2 1.8 Relative PDE • PDE of the MPPC is x1~2 of the PMT ! •25 o C • Larger pixel size •20 o C 1.4 � less dead space •15 o C � larger PDE 1 1 • The PDE also depends on over-voltage, and slightly affected by temperature 0.6 change. 0.5 1 1.5 2 2.5 V bias - V 0 (V)

1600 pixel Laser Scan in One Pixel Microscopic view • Pin-point scan has been done using YAG laser ( λ = 532 nm) with spot size ~ 1 µ m. • Variation of photon sensitivity and gain in one pixel are evaluated. • Observed variation is 2 ~ 5 % in a sensitive area for the 100 / 400 / 1600 pixel MPPCs. Gain (x 10 5 ) 1600 pixel Sensitivity 1600 pixel Gain Sensitivity (arbitrary) y-point (1 µ m pitch) y-point (2 µ m pitch) One pixel Variation ~ 3 % x-point (2 µ m pitch) x-point (1 µ m pitch)

Practical Applications

Application to High Energy Physics - Calorimeter for linear collider experiment - Light yield of scintillator strip (1 x 4.5 x 0.2 cm) EM-Scintillator-layer model for beta-ray absorber plate TT 10Aug05 WLS fiber readout Direct readout MPPC MPPC T-Layer Light yield Light yield 4cmx4cmx2mm ~ 21 p.e. ~ 13 p.e. MPC R/O with WLSF X-Layer 1cmx4cmx2mm MPC R/O with WLSF Light yield (ADC counts) • The MPPC is feasible for strip-type Z-Layer scintillator calorimeter. 1cmx4cmx2mm (size,cost, performance…) MPC R/O with WLSF • Dynamic range is the key issue. particles • First scintillator-ECAL beam test will start in next week at DESY !

Application to High Energy Physics - T2K near detectors - OffAxis detector OnAxis Scintillator + WLS fiber detector With MPPC readout • Need to read out large number of signals from WLS fibers in limited space • Used in 0.2 Tesla magnetic field • The MPPC is the perfect solution ! - Belle Ring Imaging Cerenkov Detector - • Capture Cerenkov ring image for particle ID • For the ring imaging, • Sufficient photon detection efficiency • Position resolution (~5 mm) are required for photon sensor. • MPPC is a powerful candidate for this purpose, • Larger sensor area (~3x3 mm 2 ) is desired Aerogel in future development. radiators

The MPPC is still evolving … Stay tuned for future development ! Near future Jan. 2006 2006-2007 Mar. 2005 ● Improved ● 100/400/1600 ● 100/400/1600 ● 100/400 pixels performance pixels pixels ● First sample ● Larger sensor area ● Larger PDE commercialized from Hamamatsu ● More pixels ● More pixels ● Improved Gain and dark noise ● Tests of massive use ?

Summary • The MPPC is a promising photon sensor which has many remarkable features. – High gain, compact size, low-cost, excellent P.D.E., etc …. • Extensive R&D of the MPPC is ongoing in KEK DTP group collaborating with Hamamatsu photonics. – Study and improvement of basic properties … underway – Evaluation of variation over many samples … underway – Study radiation hardness (for γ -ray, neutron) … just started – Evaluate robustness and long-term stability … start soon – Test magnetic-field tolerance … near future • Tests for actual use at several high energy physics experiments are also underway. • Applications in various other fields are being explored. – Positron Emission Tomography, etc… • The improvement of the performance will be continued toward the “Perfect Photon Sensor” !

Novel Photon Detector Workshop Novel Photon Detector Workshop June- -2007 Kobe, Japan 2007 Kobe, Japan June The KEK Detector Technology Project group will host an international workshop for the future photon sensors. Check the KEKDTP web site: http://rd.kek.jp/ Contact : Takeshi.Nakadaira@kek.jp

Backups