MTL Annual Research Report 2011

Figure 1: (a) Instrument response function (IRF) of a SNAP based on two 30-nm-wide nanowires (2-SNAP). (b) Jitter of a 2-, 3-, and 4-SNAP based on 30 nm wide nanowires as a function of the normalized bias current (IB / ISW). (c) IRF asymmetry of the same devices as in (b).

Superconducting nanowire avalanche photodetectors (SNAPs) ^[1] are based on a parallel architecture that performs single-photon counting with higher signal-to-noise ratio (up to a factor ~4) than traditional superconducting nanowire single-photon detectors (SNSPDs) ^[2] . Although the understanding of the operation mechanism of SNAPs was recently improved ^{[3] [4]} , a comprehensive study of the timing performance is still lacking. In the only study reported so far ^[5] , SNAPs showed significantly higher timing jitter (>250 ps FWHM) than SNSPDs (~30ps FWHM ^[6] ).

We report a study of the SNAP timing performance for several device architectures and bias regimes. Our main findings were: (1) the instrument response function (IRF) shifted to longer delay times when the bias current was decreased (Figure 1.a); (2) when properly biased, SNAPs achieved the same jitter as SNSPDs (Figure 1.b); and (3) the IRF became more asymmetric when the bias current approached the avalanche current (Figure 1.c).

We simulated the electro-thermal dynamics of SNAPs using the model described in ^[3] . While we could not explain the origin of the IRF asymmetry, the IRF shift to longer delay times was shown to be caused by a change in the electro-thermal behavior of the detectors at decreasing bias currents. We conclude that SNAPs are suitable for timing-sensitive single-photon experiments while offering a higher signal-to-noise ratio than standard SNSPDs.