Afterpulsing studies of low-noise InGaAs/InP single-photon negative-feedback avalanche diodes

Abstract

We characterise the temporal evolution of the afterpulse probability in a free-running negative-feedback avalanche diode (NFAD) over an extended range, from ns to ms. This is possible thanks to an extremely low dark count rate on the order of 1 cps at 10% efficiency, achieved by operating the NFAD at temperatures as low as 143 K. Experimental results in a large range of operating temperatures (223–143 K) are compared with a legacy afterpulsing model based on multiple trap families at discrete energy levels, which is found to be lacking in physical completeness. Subsequently, we expand on a recent proposal which considers a continuous spectrum of traps by introducing well-defined edges to the spectrum, which are experimentally observed

Low noise InGaAs/InP single-photon negative feedback avalanche diodes: characterization and applications

“…The NFAD devices that we have used belong to this new generation of single-photon detectors and offer the possibility to lower temperatures as an effective mean to reduce DCR. The only limiting factor is now the increase of the afterpulsing probability for a given hold-off time 18,22,23 . Fig.…”

Systematic afterpulsing-estimation algorithms for gated avalanche photodiodes

“…Afterpulsing is a complex stochastic self-interacting phenomenon, which is proportional to the incoming light intensity. A number of techniques have been used to study it, including: time interval analysis [47][48][49], double gate method [14,50,51], temporal distribution (background decay) [37,52,53], in-gate effect of afterpulsing [9,10,31], corrections in g (2) (τ) correlation measurements [23,30,54,55], modified double-gate method (in order to study higher-order afterpulsing) [10,56], and other studies of the dependence of afterpulsing on operation settings [11,22].…”

“…Due to the electric field anisotropy in the internal structure of APDs, there are ensembles of carrier-traps with associated energy distributions. In the multiple exponential decay function (MEDF) approach [14,19,48,52,57] each time constant is related with a particular carrier trap energy [52,58]. More sophisticated models introducing carrier-trap energy distributions are discussed in [52,59,60].…”

“…In the multiple exponential decay function (MEDF) approach [14,19,48,52,57] each time constant is related with a particular carrier trap energy [52,58]. More sophisticated models introducing carrier-trap energy distributions are discussed in [52,59,60]. The most widely used model to study afterpulsing is an effective single exponential decay function (SEDF) [14,19,48,52,57,58,60], with parameters determined by the corresponding mean values over the carrier-trap ensemble.…”

Afterpulsing model based on the quasi-continuous distribution of deep levels in single-photon avalanche diodes

“…Though these imperfections cannot nowadays be overcome completely, in many applications, especially in ones related to security, like quantum cryptography and quantum random number generators, it is sufficient to provide a precise quantitative characterisation of the imperfections, thus enabling the users to distinguish the regular non-ideal detector operation from an intrusion of a malevolent adversary. And though at present the quantum efficiency, deadtime and dark count rate of SPADs can be measured with acceptable precision and are typically included into the detector specifications by the manufacturer, the situation with afterpulses is far less optimistic, notwithstanding considerable efforts in this area [4,5,6,7,8,9,10,11,12]. In particular the exact shape of the distribution function for the apterpulse waiting time is not generally known and the methods for its precise characterisation are still in the process of development.…”