Self-aligned multi-channel superconducting nanowire avalanche photodetector

We describe a micromachining process to allow the coupling of an array of single-mode telecommunication fibers to individual superconducting nanowire single photon detectors (SNSPDs). As proof of principle, we show the integration of four detectors on the same silicon chip, including two standard single-section nanowire detectors and two superconducting nanowire avalanche photodetectors (SNAPs) with modified series structure without external inductor, and their performances are compared. The SNAP shows saturated system detection efficiency of 16% while the dark count rate is less than 20 Hz, without the use of photon-recycling reflectors. The SNAP also demonstrates doubled signal-to-noise ratio, reduced reset time (~ 4.9 ns decay time) and improved timing jitter (62 ps FWHM) compared to standard SNSPDs.

2 Multiple fibers are installed in an array of patterned pits on a single detector chip. This approach enables back-illuminated detector structure which separates the optical access and electronic readout circuits on two sides of the chip and allows for compact integration of multi-channel detectors.  is deposited. Then, arrays of rectangular-shaped windows are opened in the nitride layer at the backside via optical lithography and following reactive ion etching (RIE). Using these nitride patterns as mask, the silicon substrate can be anisotropically etched into trapezoidal pits in heated KOH solution. In principle, the etching stops automatically when the silicon substrate is completely etched through, leaving free-standing nitride membranes. However, we stop the reaction in advance by accurate timing and thus leaving 20 m -thick residue silicon layer, which serves as supporting structure to enhance the mechanical strength of the membranes. By controlling the dimension of the windows patterned in the nitride mask layer, we are able to make the bottom of silicon pits slightly smaller than a standard SMF-28 fibers (125 m in diameter) so that fibers can be stuck at certain height without breaking the membrane as shown in Fig. 1(a). After the wet-etching process, a thin layer of NbTiN superconducting film is deposited by means of DC magnetron sputtering and the silicon wafer is then diced into several smaller chips (8 mm × 11 mm size), which contain an array of 4 × 5 = 20 pits, for further device fabrication. A total of 20 nanowire detectors are fabricated on each chip via two e-beam lithography steps. The first e-beam lithography exposes the nanowire pattern in 6% HSQ (~100nm thickness) negative resist, which is accurately aligned with the backside silicon pits so that the active nanowire area of the detector is centered in the nitride membrane with a misalignment less than 1 m. The second ebeam lithography defines Ti/Au metal pads using a standard PMMA/MMA bilayer lift-off process.
Finally, the nanowire pattern of HSQ is transferred to underlying NbTiN layer by RIE using CF4 plasma.
The completed detector chip is mounted on a printed circuit board (PCB) and fixed upside down on an inverted microscope for fiber alignment and packaging. The targeted pit is first filled with a tiny drop of UV-curable epoxy. A cleaved fiber is manipulated by a 3-axis motorized stage with a minimum step size of 30nm to approach the bottom of the pit. Precision alignment to the active nanowire area is guided by a red laser spot launched through the fiber as shown in Fig. 1 Among the total 20 nanowire detectors fabricated on one silicon chip, 10 detectors are standard SNSPDs and the other 10 are superconducting nanowire avalanche photodetectors (SNAPs). Different from the n-SNAPs reported by Ref. 20 and Ref. 21 , which employ current-limiting inductors serially connected with n parallel active nanowires, we adopted the modified structure of series-2-SNAP similar to Ref 22 but with much larger detection area. In this way, the external inductor could be entirely folded into the active detection area and thus the total inductance of the detector can be significantly reduced.
Two standard SNSPDs and two series-2-SNAPs were randomly selected among the 20 detectors for fiber connectorization.  NbTiN film deposited on Si3N4-on-Si substrate is 6.5nm covered by 2nm-thick native oxide (confirmed by TEM). The nanowire width, spacing and the diameter of the circular-shaped active nanowire area is 40nm, 80nm and 10 μm, respectively. The floating nanowires outside the circular area is for proximity effect correction during e-beam exposure, which constitutes 20 μm × 20 μm rectangular together with the active nanowires. The 180 degree bending between nanowire elements is especially designed as round corner for relieving current crowding. 23 Figure 2 (b) shows an equivalent electrical circuit model for series-2-SNAP. The basic 2-SNAP element consists of two parallel nanowires which are modeled as two inductors (L0), and the series-2-SNAP consists of an array of serially connected 2-SNAPs. When a single photon is absorbed by any nanowire section and thus create a hotspot, the current has to be diverted into the parallel nanowire first because of the current-limiting serial inductor Ls. If the bias current IB is high enough (larger than avalanche current IAV), the diverted current will switch the secondary nanowire section to normal state. Therefore, most of the current flowing through the device, which is about two time the current carried by a single nanowire section, will be diverted to the read-out RL, providing two times signal-to-noise ratio (SNR) compared to standard SNSPDs. In previous n-SNAP structure, 20,21 the current-limiting series inductor Ls is connected externally and typically designed as 10 times larger than L0 for ensuring stable operation without afterpulsing, 24 which limits the reset time and also timing jitter. However, in this modified series-2-SNAP structure, all the unfired nanowire pairs serve as current-limiting inductor Ls until the avalanche happens. If the active area of the detector is large, e.g. 10 μm in diameter, the total length of the nanowire is more than 650 μm. This is equivalent to Ls > 16 × L0, so that dedicated external inductor is no longer needed and hence the reset time can be shortened considerably. In order to characterize the performance of our detectors, the detector package is mounted on the cold plate of a closed-cycle refrigeration cryostat 25 and cooled down to ~1.7K. 1550nm laser light is sent through single-mode fibers installed in the cryostat, which are fusion-spliced with the fibers in the detector package. Polarization controller is used for optimizing the polarization status of incident photons and the photon flux is fixed at 100 kHz via a series of variable attenuators. Figure 3 (a) shows the plot of the system detection efficiency (SDE) and dark count rate (DCR) as a function of normalized bias current for the best detector (series-2-SNAP). We measure the maximum SDE of 16.1% for parallel polarized photons with the DCR (without room light) lower than 20 Hz. However, we find the DCR increases to ~1.6 kHz with room light switched on, which we believe arises from the stray light that leak into fibers and cryostat and can be removed by better shielding of fibers and cryostat. In order to investigate the crosstalk between neighboring detectors we sent 1 MHz rates of photons to the adjacent detector and measured the dark count again, the curve of which perfectly overlaps with the DCR curve without room light, indicating the absence of crosstalk. Figure 3(b) and (c) show averaged signal traces and timing jitter measured by oscilloscope for series-2-SNAP and standard SNSPD having the same detection area, respectively. As expected, the SNR of series-2-SNAP is almost doubled compared with standard SNSPD and the decay time is shortened from 10.2 ns to 4.9 ns, which are extracted from exponential fitting. Series-2-SNAP also shows better time performance with reduced jitter of 62 ps compared to standard SNSPD's 79 ps owing to the improved SNR. We also measured the performance of the other three detectors and they demonstrate 12% (series-2-SNAP) and 3-5% (standard SNSPDs) system efficiency at similar dark count level.
In conclusion, we demonstrated a new robust self-aligned packaging scheme with multi-channel detectors on a single chip. We also showed saturated SDE for the first time, to the best of our knowledge, in series-2-SNAP with detection area comparable with the core size of single-mode fibers.
We demonstrate good success with all four connected detectors. In principle, larger array of detectors can be assembled if sufficient coaxial readout cables are available. Furthermore, by integrating the detectors with optical cavities combined with pre-screening before packaging, it is possible to realize more efficient multi-channel single-photon detectors in a very compact size. Recently, our multi-channel detector package has been already used in characterizing fiber-coupled photon-pair sources.