Instrumentation & Measurement Magazine 24-5 - 71

that it easily absorbs oxygen, resulting in a deterioration of the
material purity over time, and hence the critical temperature is
lowered (as well as the critical current), affecting the detector
performances. To prevent the material oxidation, we deposited
a thin Al film on the SNSPD surface.
In [17], the authors demonstrate that the deposition of a
weak ferromagnetic layer on the SNSPD enhances the signal
amplitude, increases the critical current and reduces the dark
counts, probably due to pinning effect. In this scenario, we observed
that the presence of the Al additional layer, along with
preventing the oxidation, raised the critical temperature up to
T ~ 6 K, which made it possible to detect the single photon at
T = 3.3 K.
Device Fabrication and Characterization
We deposited by magnetron sputtering a 9 nm thick Mo0.70
Si0.30
film. By x-ray analysis we could estimate that the final stoichiometry
is reliable within a factor of 5%. We chose a Si substrate,
and the pressure in the deposition chamber was kept below
10−7
Fig. 3. MoSi/Al SNSPD IV curves at T = 3.2 K and T = 4.2 K.
Torr. In the same vacuum chamber, we deposited a 2 nm
Al thin film to protect the superconducting material and prevent
the oxidation. Such a thin Al thickness also ensures that
the transition temperature of the film is not modified due to the
proximity effect, resulting in a working temperature that can
be easily reached with a He closed-cycle cryostat. We also deposited
by evaporation technique a Ti/Au layer of 70 nm for
the contact pads.
The thin film was then patterned trough electron beam lithography
(EBL), and we realized a 150 nm wide strip, folded
in a meander shape to cover a square area of 5 μm × 5 μm (Fig.
1). We estimated the SNSPD critical temperature by the R(T)
curve and we obtained Tc
current density Jc are 0.7 MA/cm2
that the Al layer prevented the oxidation and protected the
superconductor.
In Fig. 2 we present the result of a measurement of critical
current from T = 7 mK to T = 6 K, after two years from the
SNSPD fabrication and many applied thermal cycles.
We also measured the IV curves at T = 3.2 K and T = 4.2 K
= 6.1 K. We found that the critical
at T = 4.2 K and 2.1 MA/cm2
at T = 2.1 K. These values are among the highest found in literature
for SNSPD based on MoSi. After two years and many
thermal cycles, the critical current density is reduced just by
about 20%. Similar damages are reported usually just after a
single thermal cycle for unprotected MoSi devices. This means
(Fig. 3). Interestingly, we observe that the IV curves are also
hysteretic at T = 4.2 K, due to the enhancement of the critical
current provided by the Al layer. The hysteresis in the IV curve
is a key parameter in SNSPDs since the voltage pulses arising
after a photon detection are proportional to its amplitude (as
also shown in [14]). The fact that we can observe a deep current
hysteresis is an indicator of the possibility to perform single
photon counting with MoSi/Al also inside a liquid helium
dewar, that is at T = 4.2 K, or in relatively inexpensive closed
cycle cryostats (operating between 2 K and 4 K).
Fig. 2. Critical current vs temperature in the temperature range 0.01 K < T < 6 K.
August 2021
Single Photon Detection at 1550 nm
After noticing that the Al layer improved the SNSPD performances,
we wanted to verify that the device still acts as a single
photon detector. To do that, we anchored the device onto the
cold plate of a cryostat kept at the temperature T = 3.3 K, and
we illuminated the device with an optical fiber at λ = 1550 nm.
The distance between the device and a standard single mode
SMF 28 fiber is about 1 cm, providing a uniform illumination
on the detector area. A CMT CITLF3 SiGe low noise cryo-amplifier
(Cosmic Microwave Technology, Inc., 0.01 to 2 GHz,
37 dB gain) kept at T = 40 K is used to amplify the output pulse
of the detector. The amplifier is equipped with a cryogenic integrated
DC bias tee used to dc bias the detector by means of a
voltage generator and a 100 kΩ resistor. A coaxial cable leads
the amplified output voltage pulse up to room temperature
where it is analysed with a fast oscilloscope (4 GHz Lecroy waverunner
640zi) or a counter (keysight 53220A). The counting
rate and the dark count rate as a function of the bias current are
reported in Fig. 4.
To verify the single photon detection regime, we fixed
the bias current and measured the counting rate at different
IEEE Instrumentation & Measurement Magazine
71
Instrumentation & Measurement Magazine 24-5

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 24-5
