sipms for solar neutrino detector
play

SiPMs for solar neutrino detector? J. Kaspar, 6/10/14 1 SiPM is - PowerPoint PPT Presentation

SiPMs for solar neutrino detector? J. Kaspar, 6/10/14 1 SiPM is Geiger photodiode APD Mode APD full depletion V APD take a photo-diode reverse-bias it above breakdown voltage (Geiger mode avalanche photo diode) add


  1. SiPMs for solar neutrino detector? J. Kaspar, 6/10/14 1

  2. SiPM is Geiger photodiode APD Mode APD full depletion V APD • take a photo-diode • reverse-bias it above breakdown voltage 
 (Geiger mode avalanche photo diode) • add quenching resistor • repeat many thousand times 2

  3. 
 Huge variety of products Hamamatsu 
 KETEK 
 SensL 
 Philips 
 ST Microelectronics 1x1 – 26 x 26 mm 10 – 200 um pixels 3

  4. Theory of operation • each pixel is an independent detector • when pixel fires, it delivers charge (~1e6 electrons) 
 (regardless how many photons hit the pixel) • then the pixel is dead for ~100 nsec (recovery time) • ~1000 times more pixels on the device than photons 4

  5. Advantages • fast like PMT, compact, cheap • runs in magnetic fields, is non-magnetic • high photo detection efficiency • low voltage (typ. 40 or 70 V, diode orientation) • much lower radioactivity than PMT 5

  6. � Disadvantages • temperature dependence 
 photo-effect (sec. order, cryo) 
 probability electron triggers avalanche discharge 
 gain (charge delivered by a pixel when it fires) 
 � (M=1.25 × 10 6 ) 3000 breakdown voltage 
 2500 Frequency (number of events) -> temperature monitoring 
 2000 -> in situ calibration 
 1500 (0,1,2 ph comb, or laser) 1000 500 • batteries not included 
 0 0 1 2 3 4 5 6 7 8 9 Number of detected photons no standard pre-amplifier (think PMT without a base) 6

  7. Cryo compatible • photon absorption length depends on temperature • easy down to 100 K (SiPMs like that) • modified runs well in LXe, LAr 
 anti-reflective coating (UV eff), cryo comp package 
 K.Sato NIM A 732, 2013 (427–430) • charge carrier freeze-out < 50 K 7

  8. Historical artifacts • cross-talk -> optical trenches • after-pulsing -> Si wafer purity • high dark rate -> Si wafer purity • slow pulses, pulse dependent on temperature -> 
 metal (Ni) based quenching resistor • high cost -> now much cheaper than PMT 8

  9. cross-talk • real photons emitted 
 during avalanche discharge • problem for stat properties of pulses 
 e.g. mean over sigma squared proxy 
 for number of pixels fired • optical trenches 9

  10. after-pulsing • incomplete discharge • part trapped • delayed release 10

  11. metal quench resistor pulse decay time: 
 R (quench resistor) * 
 C (diode) poly-crystalline Si (old) Ni based (new) 11

  12. examples of use • Cerenkov telescopes (CTA) 
 single photons, shaping, clipping, 
 pole-zero correction • hadron calorimeters 
 ~1000 photons 
 Cerenkov, fast scint, or both • positron emission tomography 
 TOF 12

  13. pre-amplifier batteries not included (like a PMT without a base) 3 possible designs: • voltage amplifier with a shunt resistor 
 pulse shapes ~40 nsec • discrete trans-impedance amplifier 
 pulse shapes ~10 – 20 nsec • integrated trans-imp. amplifier 
 pulse shapes ~5 nsec 13

  14. g-2 example • anomalous dipole moment 
 of muon • segmented lead fluoride 
 calorimeter (Cerenkov) • readout by SiPMs 
 25 – 4000 photons 
 per event 14

  15. SiPM board design goals • energy scale (gain) stability 
 comes from pulse amplitude 
 0.1 % (short time stability) • timing resolution, time scale accuracy 
 comes from the leading edge 
 ~30 ps (different crystals), ~50 ps (pile-up) • pulse width 
 leaked energy vs. direct hit, lost muon 
 SiPM board should preserve light profile 15

  16. Monolithic design S12642-0404PA-50 
 16 channel array 
 12x12 mm 2 (active) area 
 50 µm pixels 
 Ni-based quench. resistor 
 through silicon vias 
 optical trenches 
 high purity Si wafer 25 – 4000 photons 16

  17. LMH6881 fully differential op-amp 
 variable gain (6–26dB), SPI 
 thermal coupling to crystal AC coupled output 
 2 MMCX connectors feeding 
 twinax cable 2x THS3202 dual op-amp 
 each sums 4 SiPM channels THS3201 at unity gain 
 sums four 4-sums 17

  18. 1 2 3 4 5 A A Knobs to turn 1 C1 1 C2 1 C3 Bias Return 1 C4 C6 1 100p 0.1u 6.8u 2 1 2 2 2 1 C7 1 C8 0.01u 1 C9 C5 1u +5V 0.01u 2 2 R1 0.01u 1u 0.01u 2 2 2 Bias 1 C11 2 1 +5V 4.99 C10 Vs+ 0.01u 5 THS3201 0p Vcc 2 19 R2 Vcc 20 +In U2 R4 Out + SOT23-5 + Vcc 2 1 SiPM anode to 23 3 49.9 24.9 R3 Vcc 2 1 24 Vout INMS 49.9 1 8 C12 0.01u - -In OUT+ C13 LMH6881 0.1u INMD 4 22 ground 9 OUT- INPD 21 R5 R6 R7 D1 U1 10 Vs- 2 Out - GND 2 1 25 2 1 GND INPS 2 49.9 13 49.9 +5V GND 11 R8 12 GND SD 7 C14 0.01u S12572-025 GND +5V 17 6 49.9 OCM B B SPI 2 1 C15 PGA SDO 5 D2 D0 -5V SDI 4 2 1 D1 3.0K R9 CLK 3 0.1u D2 2 1 C16 S12572-025 PGA CS- 15 D3 16 SiPM anode to 1 C17 1 C18 1 C19 D3 0.1u 2 1 2 virtual ground 100p 0.1u 6.8u S12572-025 2 2 2 1 C20 R10 D4 1.0K 2 1 0.1u 2 S12572-025 18 PGA SDO Zero ohms to connect shield, HDMI Bias Open for shield connected to Bias supply only. 1 R11 TDMS_Data_2+ R12 2 270 0 TDMS_Data_2_Shield Bias Return 3 TDMS_Data_2- Out + 4 TDMS_Data_1+ 5 TDMS_Data_1_Shield Out - 6 TDMS_Data_1- PGA SDO 7 TDMS_Data_0+ 8 TDMS_Data_0_Shield PGA CS- 9 TDMS_Data_0- CLK 10 TDMS_Clock+ 11 TDMS_Clock_Shield SDI 12 TDMS_Clock_- C C 13 CEC 14 NC 15 DDC_Clock 16 DDC_Data 17 +5V -5V Ground 18 +5V_Power 19 Hot_Plug_Det P1 File: SiPM 4 Ch Test Board Mk 9.sch Sheet: / Title: Size: A4 Date: 12 mar 2014 Rev: KiCad E.D.A. Id: 1/1 1 2 3 4 5

  19. Pulse shape intrinsic pulse 
 (no pole zero correction) response to 2.5 nsec LED jitter from the pulse gen 10 ns 19

  20. High rate capable ~5 MHz laser shots 
 2000 photons per shot jitter from the pulse gen 20

  21. pileup resolution 21

  22. pileup resolution 22

  23. Summary Geiger mode avalanche photodiodes Advantages: • fast, compact, cheap, low-voltage devices, high detection efficiency • much lower radioactivity than PMT • cryo friendly Disadvantages and Challenges • requires custom readout board • gain is sensitive to temperature; must control environment 23

Recommend


More recommend