Probabilistic analysis of solid state photomultiplier performance

Abstract

New generation of photon detectors - so-called Solid State (Silicon) Photomultipliers (SSPM, SiPM), Negative Feedback APDs (NFAD), and a few more names - is widely recognized to be competitive with PMTs and conventional APDs in various low light level applications. SSPM designs are mostly associated with multi-pixel Geiger mode APD with built-in negative feedback elements. Strong negative feedback applied to Geiger avalanche breakdown enables near ideal single electron multiplication with very high gain and ultra-low excess noise. Multi-pixel architecture provides capability of multi-photon pulse detection with remarkable photon number resolution starting from single photons at room temperature. On the other hand, the SSPM design concept results in low dynamic range due to limited number of pixels with some recovery time and in considerable excess noises of crosstalk and afterpulsing. These specific drawbacks affect signal-tonoise ratio and complicate estimation of the SSPM applicability and competitiveness with other detectors and within the SSPM generation. This study presents probabilistic analysis of the SSPM and analytical results on probability distributions of the output signals with crosstalk and afterpulsing, accounting of the saturation and non-linearity effects, representation of the key photodetection processes in terms of excess noise factors (ENF). Results of the study seem to be useful for the improvements in designing, characterization, and application-specific optimization of the SSPM.