All entries for Francesco Marsili

• Print This Page
• Download as PDF

Single-Photon Detection with Ultranarrow Superconducting Nanowires

• Authors: F. Marsili, F. Najafi, E. Dauler, F. Bellei, X. Hu, R. Molnar, K. K. Berggren
• Sponsorship: National Science Foundation, Department of Energy, Lincoln Laboratory, Department of Electrical Engineering and Computer Science (MIT)

Superconducting nanowire single-photon detectors (SNSPDs) ^[1] perform single-photon counting in the near‑infrared with outstanding performance. The main limitations of standard SNSPDs, based on ~ 4-nm-thick, 100-nm-wide NbN nanowires, are: (1) fragility with respect to constrictions; and (2) substantially reduced sensitivity beyond 2 µm wavelength (λ). We developed SNSPDs based on ultra-narrow (30- to 10-nm-wide) superconducting nanowires ^[2] , which showed improved robustness to constrictions and higher sensitivity to near-infrared photons with respect to standard SNSPDs.

As shown in Figure 1, at λ = 1550 nm our 30 nm nanowire‑width SNSPDs could be biased far from the device critical current (I_C) with minimal loss in detection efficiency (η), so even heavily‑constricted devices could reach the same efficiency as constriction‑free ones.

As shown in Figure 2 a, varying λ from 700 to 2100 nm, the η vs bias current (I_B) curves of 30 nm nanowire‑width SNSPDs kept a sigmoidal shape, with the cut-off current (I_co, taken to be at the inflection point of the η vs I_B curves) increasing from 0.31 I_C to 0.41 I_C. For 90 nm nanowire‑width detectors (shown in Figure 2 b), I_co increased from 0.64 I_C at λ = 500 nm to 0.89 I_C at λ = 1400 nm. This behavior indicates that ultra-narrow-nanowire SNSPDs are more sensitive to low-energy photons than standard devices and suggests that their sensitivity may extend to mid-infrared wavelengths.

The MIT Lincoln Laboratory portion was sponsored by the Department of the Air Force under Air Force Contract #FA8721-05-C-0002. Opinions, interpretations, recommendations and conclusions are those of the authors and are not necessarily endorsed by the United States Government.



Figure 1: η as a function of the normalized bias current (I_B / I_C) for a constricted SNSPD (red curve) and a constriction-free SNSPD (blue curve) based on 30 nm wide nanowires. Inset: Scanning electron microscope image of an SNSPD hydrogen silsesquioxane mask on NbN. The nanowires are 30 nm wide and the pitch is 100 nm.



Figure 2: a. η vs I_B / I_C curves of a 30 nm nanowire width SNSPD for λ varying from 700 nm (purple curve) to 2100 nm (black curve). The detection efficiency curves at each wavelength were normalized by the value of η at the switching current (I_SW). b. η vs I_B / I_C curves of a 90 nm nanowire width SNSPD for λ varying from 700 nm (purple curve) to 2100 nm (black curve).




1. G. N. Gol’tsman, O. Okunev, G. Chulkova, A. Lipatov, A. Semenov, K. Smirnov, B. Voronov, A. Dzardanov, C. Williams, and R. Sobolewski, “Picosecond superconducting single-photon optical detector,” Applied Physics Letters, vol. 79, no. 6, pp. 705-707, 2001. [↩]
2. F. Marsili, F. Najafi, E. Dauler, X. Hu, M. Csete, R. Molnar, and K. Berggren, “Single-photon detectors based on ultra-narrow superconducting nanowires,” Nano Letters, vol. 11, no. 9, pp. 2048-2053, 2011. [↩]

Timing Performance of Superconducting Nanowire Avalanche Photodetectors

• Authors: F. Najafi, F. Marsili, K. K. Berggren



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.

1. M. Ejrnaes et al., Appl. Phys. Lett. vol. 91, p. 262509, 2007. [↩]
2. G. N. Gol’tsman et al., Appl. Phys. Lett. vol. 79, p. 705, 2001. [↩]
3. F. Marsili et al., Appl. Phys. Lett. 98, p. 093507, 2011. [↩] [↩]
4. F. Marsili et al., Nano Lett., 2011, 11 (5), pp 2048–2053. [↩]
5. M. Ejrnaes et al., Appl. Phys. Lett. 95, p. 132503, 2009. [↩]
6. E. A. Dauler et al., IEEE Trans. Appl. Supercond. vol. 17, p. 279, 2007. [↩]