Study of radiation damage induced by 82 MeV protons on multipixel Geiger-mode avalanche photodiodes

Transcription

1 Study of radiation damage induced by 82 MeV protons on multipixel Geiger-mode avalanche photodiodes Y. Musienko*, S. Reucroft, J. Swain (Northeastern University, Boston) D. Renker, K. Dieters (PSI, Villigen) Z. Charifoulline (INR RAS, Moscow) *on leave from INR RAS (Moscow) NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 1

2 Outline Radiation damage of silicon detectors G-APDs studied and their parameters Parameters of G-APDs after 82 MeV proton irradiation Discussion NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 2

3 Radiation damage in semiconductors (Si) Radiation damage in silicon is strongly dependent on the type and energy of the radiation Two types of radiation damage: Surface damage (ionizing damage in the Si/SiO 2 interface) Bulk damage (crystal lattice defects: displacement of silicon atoms) NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 3

4 Surface damage SiO 2 is a very good insulator (or a semiconductor with a large band gap of 8.8 ev). Electron/hole pairs created by ionizing particles can be trapped into very deep levels associated with the defects in oxide from which the emission back into conduction/valence band is very unlikely at room temperature Ionizing radiation (charged particles, gammas) produces surface damage (damage in the Si/SiO 2 interface) due to accumulation of positive charges in the oxide (SiO 2 ) and the Si/SiO 2 interface This may cause: breakdown voltage shift (early APD breakdown) QE reduction surface current increase NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 4

5 Bulk damage and NIEL function Bulk damage scales linearly with the amount of Non Ionizing Energy Loss (NIEL hypothesis), which is very dependent on the particle type and its energy NIEL(1 MeV gammas) ~ 10-5 * NIEL(1 MeV neutrons) NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 5

6 Bulk damage effects Increase of the dark current generated in the silicon bulk (multiplied current): Φ eq 1 MeV neutron equivalent total flux V silicon active volume α dark current damage constant (~4*10-17 A/cm for 1 MeV neutrons after 80 min annealing at 60 C or ~10-16 A/cm few day annealing at room temperature) Changes in the effective doping concentration (creation of acceptor-like states) a few weeks after irradiation: NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 6

7 Multipixel Geiger-mode APDs (G-APDs) Model Active G-APD volume: V= S*G*L=s*L S - total area s - active area (total area minus non-sensitive area between pixels) G - geometric factor L - depletion layer thickness Expected G-APD dark current increase after irradiation: ΔI=α M P G Φ eq V α - dark current damage constant M G-APD gain P G Geiger discharge probability (is a f[v-vb]) Φ eq 1 MeV neutron equivalent total flux NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 7

8 G-APDs irradiation studies at PSI (82 MeV protons) 23 G-APDs from 5 different producers irradiated at PSI at the end of last year 1* MeV/c (82 MeV kinetic energy) protons/cm 2 in 4 steps NIEL factor is ~2 times of 1 MeV neutrons, Total flux: 1*10 10 protons/cm 2, equivalent to ~2* MeV neutrons/cm 2 Gain, PDE, Id, Dark count vs. voltage, forward G-APD current were measured before and 4 month after irradiation Signals from G-APDs were amplified with fast (1 ns rise time) transconductance amplifier, green (515nm) LED was used for PDE and gain measurements Keithley-487 picoammeter/voltage source, Picoscope 5200 (digital oscilloscope) and Labview DAQ were used in these measurements measurements after irradiation were performed at CERN APD Lab, temperature during the measurements was stabilized using air conditioner (dt<1 C) Here I will report results on 5 G-APDs, received from different producers: CPTA/Photonique FBK Hamamatsu (HPK) MEPhI/Pulsar Zecotek NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 8

9 G-APDs and some of their geometric and electrical parameters before irradiation (T=22 C) G-APDs Substrate Area [mm 2 ] # of pixels VB [V] R pix [MOhm] CPTA/Photonique p-type MEPhI/PULSAR p-type FBK p-type Zecotek (MAPD-2A) p-type ? HPK S C n-type NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 9

10 Dark current increase during irradiation Dark Current [mka] "HPK_1mm_#535" (U=69.5V) "HPK_1mm_#535"(U=70.1V) "HPK_1mm_#535"(U=70.7V) Irradiation # First result: Dark current increased linearly with the integrated flux for all 5 G-APDs NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 10

11 Change of the breakdown voltage after irradiation (T=22 C) G-APDs VB [V] before irr. VB [V] after irr. Estimated measurements error [V] CPTA/Photonique MEPhI/PULSAR FBK Zecotek (MAPD-2A) HPK S C Second result: No VB change within the accuracy of our measurements (~50 mv) NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 11

12 Forward bias measurements before and after irradiation Third result: No cell resistor change within the accuracy of our measurements (~5%) NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 12

13 Cell recovery Double LED pulse was used to measure the cell recovery time. LED signal had length of ~10 ns and was sufficient to saturate the G-APD response (~ photons/pulse) NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 13

14 Dark current vs. voltage before and after irradiation NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 14

15 Dark count vs. voltage before and after irradiation NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 15

16 Normalized LED amplitude vs. voltage before and after irradiation Measured with low intensity, fast (~50 photons, 10 ns ) green LED pulse. The LED pulse was normalized using XP2020 PMT. NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 16

17 PDE/F vs. voltage before and after irradiation It is difficult to measure PDE after irradiation. We measured: PDE*= PDE/F=(A-ped)/SQRT(RMS A2 -RMS ped2 ) NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 17

18 Dark Count Increase/PDE/Area-I G-APDs studied have different area, geometric factor, depletion volume, etc. How to compare the dark count increase produced by radiation in different G-APDs? Expected G-APD dark count increase after irradiation: ΔN=ΔI/q/M=α*M*P G *Φ eq *V/q/M=α*P G *Φ eq *S*G*L/q, q electron charge Assuming that PDE is proportional to P G *G: ΔN ~α*pde*φ eq *S*L/q ΔN/PDE/S ~α*φ eq *L/q This ratio is expected to have weak dependence on the G-APD PDE, geometric factor and sensitive area. Dependence on the depletion thickness remains. NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 18

19 Dark Count Increase/PDE/Area-II NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 19

20 The effective thickness 4*10-17 A/cm/n damage constant ~640 Hz/cm/n dark count 2*10 10 n/cm 2 ~13 MHz/μm/mm 2 or 130 khz/μm/mm 2 /% We measured khz/μm/mm 2 /% Effective thickness: μm NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 20

21 Summary of the results G-APDs from 5 different producers irradiated at PSI at the end of last year up to 1* MeV/c (82 MeV kinetic energy) protons/cm 2 5 G-APDs were characterized in details 4 month after irradiation The results of our study: - No change of VB (with 50 mv accuracy) - No change of Rcell (with 5% accuracy) - Dark current and dark count significantly increased for all the devices - No change of PDE/F ratio (resolution) (with ~15% accuracy) - The gain of some G-APD decreased by % (probably due to cell occupation) - The G-APDs operate as a Si device with μm effective thickness NDIP-08, June 17, 2008, Aix-les-Bains Y. Musienko 21