Optical Crosstalk in SiPM Nepomuk Otte Max-Planck-Institut für Physik, München Humboldt Universität, Berlin light emission during Geiger breakdown (Sciacca, 2003)

Outline optical crosstalk light emission in avalanches what is SiSi? tuning SiSi photons responsible for optical crosstalk simulating optical crosstalk in a back side illuminated SiPM conclusions

Optical Crosstalk photons can trigger additional cells Sketch from Cova et al. NIST 2003 Workshop on single photon detectors  called optical crosstalk artificial increase in signal Excess Noise Factor of SiPM can be quite significant and problem in applications

Light Emission in Avalanches I will show: This is the interesting region several people tried their best inconclusive results I will show measured spectra do not show similar behavior emission mechanisms not clear very few absolute measurements W. J. Kindt do different technologies emphasize different emission mechanisms?

SiPM-Simulator* full geometrical description of a SiPM: number of cells active volume … simulation of avalanche photons: black body radiation with free parameters: temperature intensity isotropic emission For this purpose I developed SiSi full geometrical simulation of a SiPM photoelectrons in non-depleted bulk are subject to simple diffusion model; lifetime of electrons is free parameter *Elisabeth ”Sis(s)i”von Wittelsbach was the empress consort of Emperor Franz Joseph of Austria. She was born 1853 in Munich, Bavaria and murdered 1898 in Geneva, Switzerland

Tuning of Model Parameters SiSi has three free parameters which need to be determined temperature of photon spectrum intensity of photon spectrum / probability that avalanche carrier emits photon lifetime of electrons in non-depleted bulk

Model parameters tuned by reproducing measured optical-crosstalk behavior of a SiPM with SiSi. by MEPhI/Pulsar optical crosstalk (dark noise) spectrum (measured)

three examples of simulated distributions of optical crosstalk: free parameters: temperature and intensity (efficiency) of the photon spectrum lifetime of electrons in non-depleted bulk: 0 nsec Goodness of agreement between simulation and measurement is tested with a χ²-Test

Simulated optical crosstalk: Temperature 4500K Efficiency >1.015eV: 1.45x10-4 photons/electron electron lifetime 60 nsec integration time 18 nsec simulated and measured crosstalk distribution residuals Residuals can be explained by dark counts which are not simulated in SiSi

 no unique solution of model parameters but χ² -distribution of a scan in: temperature of the photon spectrum intensity of photon spectrum log scale  no unique solution of model parameters but

Characteristics of Photons responsible for Optical Crosstalk Intensity (1.15 … 1.40 eV): ~2.8*10-5 photons / avalanche electron (uncertain by a factor of ~2) Peak: ~1.26eV clear correlation between temperature and photon emission efficiency FWHM: ~0.21 eV very narrow distribution

No need to precisely now spectral shape! 2000 K No need to precisely now spectral shape! 4500 K simulated photon spectrum

~10µm – 1mm absorption length note the log-scale Narrow energy distribution explained by strong dependence of absorption lengths on energy

Optical Crosstalk in Back Side illuminated SiPM photon entrance window active volume simulated structure: pitch between cells 100 µm avalanche region 10 µm diameter 100% active volume device proposed by G. Lutz et al.

Crosstalk Probability in Back Side illuminated SiPM assume breakdown probability = 0.9 For operation of back side illuminated SiPM need gain << 105 ~20% crosstalk at gain 104 But: absolute measurements always difficult! e.g. Lacaita (1993) give five times lower intensity  ~4% crosstalk at gain 104 instead of 20%

Conclusions SiSi is a nice tool to study SiPM crosstalk behavior well described after tuning three free parameters in the simulation of which two are strongly correlated. only photons within a narrow energy interval (1.15eV-1.40eV) give rise to optical crosstalk measured intensity of photons: ~2.8•10-5 photons / avalanche electron (1.15eV-1.40eV) optical crosstalk is a serious problem for back side illuminated SiPM  need low gain ~104; but: inferred photon intensity could be too high

Two Examples of Crosstalk Events

SiPM that was used to tune SiSi produced by MEPhI/Pulsar 576 cells 42µm sensitive volume 21µm 2.5µm 390µm avalanche region non-depleted bulk One cell of the SiPM

Electron Lifetimes 4500K 2000K Intensity of the photons is reduced by ~30% if lifetime of the electrons in the non-depleted volume is non zero