Single-photon detection system for quantum optics …cms2.unige.ch/gap/optics/wiki/_media/publications:bib:sspdinqo.pdf · quantum communication and quantum optics, for example antibunching-type

> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <

1

Abstract— We describe the design and characterization of a

fiber-coupled, double-channel single-photon detection system based on superconducting single-photon detectors (SSPD), and its application for quantum optics experiments on semiconductor nanostructures. When operated at 2 K temperature the system shows 10% quantum efficiency at 1.3 µµµµm wavelength with dark count rate below 10 counts per second and timing resolution <100 ps. The short recovery time and absence of afterpulsing leads to counting frequencies as high as 40 MHz. Moreover, the low dark count rate allows operation in continuous mode (without gating). These characteristics are very attractive – as compared to InGaAs avalanche photodiodes – for quantum optics experiments at telecommunication wavelengths. We demonstrate the use of the system in time-correlated fluorescence spectroscopy of quantum wells and in the measurement of the intensity correlation function of light emitted by semiconductor quantum dots at 1300 nm.

Index Terms— superconducting single-photon detectors, superconducting thin films, time-correlated single-photon counting, antibunching, single-photon sources

I. INTRODUCTION UPERCONDUCTING single-photon detectors (SSPDs)

are a nanowire devices based on long strip of ultrathin superconducting NbN film operated well below critical temperature Tc but in the presence of subcritical bias current [1]-[3]. The prospects of the SSPD have been recently

Manuscript received January 9, 2007. This work was supported in part by the grant "Non-equilibrium processes after IR photon absorption in thin-film superconducting nanostructures" of Russian Agency on education and the grant 02.445.11.7434 of Russian Ministry of education and science for support of leading scientific schools, by the European Commission under project "SINPHONIA", contract number NMP4-CT-2005-16433, and project “QAP”, contract number 15848, and by the Swiss National Science Foundation through the NCCR Quantum Photonics and Professeur boursier programs.

A. Korneev, O. Minaeva, A. Divochiy, Yu. Vachtomin, K. Smirnov, O. Okunev, G. Gol'tsman are with the Department of Physics, Moscow State Pedagogical University, Moscow 119992 Russia.

C. Zinoni, N. Chauvin, L. Balet, F. Marsili, D. Bitauld, B. Alloing, L.H. Li, A. Fiore, are with the Ecole Polytechnique Fédérale de Lausanne, Institute of Photonics and Quantum Electronics, Station 3, CH-1015 Lausanne, Switzerland.

L. Lunghi and A. Gerardino are with the Institute of Photonics and Nanotechnology, CNR, via del Cineto Romano 42, 00156 Roma, Italy

M. Halder, C. Jorel and H. Zbinden,are with .Group of Applied Physics, University of Geneva, 20, Rue de l’Ecole de Médecine, 1211 Geneva 4, Switzerland.

demonstrated: high quantum efficiency in the visible light and near infrared spectrum, single-photon sensitivity up to 5.6 µm wavelength, negligibly low level of dark counts, picosecond pulse-to-pulse timing jitter. Recently subnanosecond dead time was demonstrated on SSPDs consisting of several small meanders connected in parallel and covering total area of 10 µm x 10 µm [4] thus opening the way to gigahertz counting rates. By a number of characteristics SSPD outperforms traditional single-photon detectors based on avalanche photodiodes and photomultiplier tubes as well as other superconducting single-photon detectors such as transition-edge sensors (TES) and tunnel junctions.

One of the major difficulties related to SSPDs is their cryogenic operating temperatures. Typical SSPDs have 5%-10% QE at 4.2 K and reach 30% only when cooled below 2 K. This fact, together with the need to couple the SSPDs to single-mode optical fibers, led us to develop a two-channel single-photon detection system with fiber coupling and an easy-to-operate cryogenic system. This detection system is optimized for quantum cryptography and quantum optics experiments at telecommunication wavelengths, where the poor performance of InGaAs avalanche photodiodes represents a major limitation. In particular, we report a system QE of 10% with a dark count rate below 10 s-1 (while previous attempts at fiber-coupled SSPDs [5], [6] have resulted in a QE<1%), a temporal resolution <100 ps, and a counting frequency as high as 40 MHz. The application of this system to time-correlated photon counting for fluorescence spectroscopy and intensity-correlation measurements on semiconductor nanostructures is then discussed. We first present in Section II the fabrication and characteristics of SSPD chips. In Sections III and IV, the design and characterization of the cryogenic SSPD system including two fiber-coupled SSPD chips are described. Finally, in Section V the use of the SSPD system and its advantage over avalanche photodiodes is demonstrated in time-correlated single-photon counting (TCSPC) on semiconductor quantum wells and intensity correlation experiments on single semiconductor quantum dots (QDs) emitting at 1300 nm.

II. SUPERCONDUCTING SINGLE-PHOTON DETECTOR The active element of the SSPD is a narrow (100-120 nm

Single-photon detection system for quantum optics applications

A. Korneev, Yu. Vachtomin, O. Minaeva, A. Divochiy, K. Smirnov, O. Okunev, G. Gol'tsman, C. Zinoni, N. Chauvin, L.Balet, F. Marsili, D. Bitauld, B. Alloing, L.H. Li, A. Fiore, L. Lunghi,

A. Gerardino, M. Halder, C. Jorel, H. Zbinden

S


2

wide) superconducting strip patterned from 4-nm-thick NbN film deposited on double-side polished sapphire substrate. For better coupling with the incident radiation the strip has a meander shape and covers the square area of 10 µm ×10 µm with filling factor (the ratio of the area occupied by the superconducting strip to the device nominal area) as high as 0.6-0.7. The length of the strip reaches up to 500 µm. The NbN meander is connected to gold contact pads designed as a 50 Ω coplanar line. Fig. 1 presents the chip topology with contact pads for coplanar line and the SEM image of the meander. The meanders are fabricated using a process based on direct electron beam lithography and reactive-ion etching, contacts pads are formed with optical lithography and chemical etching [7]

Although the ultimate limit in quantum efficiency (QE) is given by the absorption in the thin film (around 30%), typical devices only reach this limit at temperatures around 2K, while at 4.2 K the QE is reduced to below 10% (see the QE vs bias current characteristics on Fig. 2). This is attributed to slight nonuniformities in the wire patterning process, and could be

overcome with improved fabrication processes, In this case we used an e-beam writer with higher resolution. In practice it is difficult to reproducibly realize extremely uniform meanders. This fact encourages using SSPDs in cryostats or cryocoolers capable to provide 2 K temperature level.

We also note that the QE can be further increased at a particular wavelength by the integration of the SSPD with a λ/4 microcavity [8], although that results in a strong spectral dependence of the QE [9].

When the SSPD is completely shielded from room temperature background radiation it exhibits negligibly low dark counts: 2x10-4 s-1, a value which was actually limited by the duration of the experiment [3]. Such a value of dark counts rate was measured at 2 K temperature and a ratio of bias current to critical current equal to 0.88. In fig. 2 it corresponds to bias current of about 20 µA, where QE is above 10%. This should be compared to InGaAs single-photon avalanche photodiodes (APDs), which present a dark count rate in the tens of kHz range, depending on the gate width and frequency.

The SSPD has a nanosecond-range photoresponse time which is mostly determined by the kinetic inductance of the long, narrow and thin superconducting strip. As it was shown by A. Kerman et al [10] a 10 µm x10 µm active area SSPD with 500 µm total length has kinetic inductance of 415 nH. As shown below, this limits the maximum counting frequency to few tens of MHz. On the other hand, the internal mechanism of energy relaxation in thin superconducting NbN film has an ultrafast (picosecond) characteristic time, which allows one to realize up to 1GHz counting rate if a low kinetic inductance topology of SSPD is applied. In contrast, InGaAs APDs are limited to much lower counting frequencies (typically <<1 MHz), due to the need of imposing a deadtime in order to reduce the afterpulsing probability.

III. SINGLE-PHOTON DETECTION SYSTEM DESIGN We developed and implemented a two-channel single-

photon infrared detection system designed for research in quantum communication and quantum optics, for example antibunching-type correlation studies of near-infrared photons emitted by semiconductor quantum dots.

As noted above, typical SSPDs reach the highest 30% QE at a temperature of 2 K. Thus, the basic idea was to create a system that would allow reaching a temperature of 2 K and below using standard liquid-He storage dewars, rather than cryostats or expensive cryocoolers. The system is designed as a double-wall insert with vacuum insulation (fig. 3). Liquid helium from the storage dewar penetrates into the inner volume of the insert through the filter and the capillary. To achieve a temperature below 4.2 K the liquid helium vapor is pumped from the inner volume. The capillary limits the speed of liquid helium penetration and allows obtaining a vapor pressure of 80-120 Pa and a temperature of 1.7 K.

The system has two identical photon detecting channels. Two SSPDs (one in each channel) are coupled to the single-

3 mm

3 m

m

80 nm

Fig. 1. Detector contact pads design and SEM image of the superconducting meander covering 10 µm x 10 µm area. Strip width is 100-120 nm with spacing of 100-80 nm respectively.

12 16 20 2410-4

10-3

10-2

10-1

100

101

102

9%

30%

Qua

ntum

effi

cien

cy %

Bias curent, µA

T=4.2 K T=1.8 K

Fig. 2. Typical dependence of SSPD quantum efficiency at 1.26 µm wavelength on bias current. Although at 4.2 K SSPD has QE below 10% at 1.8 K temperature it increases to 30% saturation level limited by absorption of NbN nanostructure.


3

mode optical fibers and mounted at the bottom of the inner volume of the insert in the liquid helium. The SSPDs are aligned against the fibers with the standalone micromechanical positioner and then mechanically fixed with metal springs on the fibers [11]. When SSPDs are fixed in this way the positioner is removed. This structure does not exhibit any subsequent noticeable displacement of SSPDs when they are cooled to liquid helium temperature. Electrical contact is realized through co-planar waveguides and coaxial cables connected to two room temperature bias-Ts. For DC bias we use a home-built current/voltage source. The absorption of a photon leads to the appearance of a voltage pulse with about 1mV amplitude on the terminals of the SSPD. This voltage pulse is transmitted by the coaxial cable to the room-temperature amplifiers, one in each channel (60 dB gain) producing pulses larger than 0.5 V. Then the signals from the channels are fed to the discriminators and to the counter and oscilloscope or to the coincidence circuit depending on the goals of experiment.

The temperature is controlled by the efficiency of pumping and is measured with the thermometer mounted on the same holder as the SSPDs. The minimum achievable temperature is mostly determined by the capabilities of the pump and can be

as low as 1.6 K. Temperature below 2 K can be achieved with a pump performance above 5 m3/hour.

IV. SYSTEM CHARACTERIZATION To achieve maximum sensitivity SSPDs are cooled to

temperatures below 2 K and biased with current above 0.8Ic. Upon photon absorption SSPD produces a voltage pulse of several nanoseconds with a very sharp front edge. The amplitude of the pulse is proportional to the bias current. The dependence of the pulse amplitude on the photon energy was not studied thoroughly yet. Anyway if such dependence exists it is very weak. The rise and fall times are limited by the kinetic inductance of the NbN meander and can be described (similarly to the effects of geometric inductances) in terms of a characteristic time τ=Lk/R. Here Lk is kinetic inductance and R is the equivalent resistance seen by the meander. In the case of the rise time (i.e. just after the superconducting-to-normal transition) R is the resistance of the normal part of the meander which appears after photon absorption, plus the transmission line resistance. In case of the fall time (i.e. when the wire has become superconducting again) R is given by the transmission line resistance only. As the normal part of the meander has resistance about an order of magnitude higher than the resistance of the transmission line, the rise time is in subnanosecond range, whereas fall time can be up to several nanoseconds long. Fig. 4 shows the SSPDs response waveform transient taken from the screen of the oscilloscope and measured in amplifiers band of 1 - 500 MHz. The rise time is limited by the amplifier bandwidth in this experiment.

An important parameter is the system quantum efficiency (SQE) which includes quantum efficiency of the SSPDs themselves and coupling losses. System quantum efficiency is defined as the ratio of photoresponse pulses from the SSPD to the number of photons fed to the input of the fiber, in the single-photon regime where each pulse contains less than one photon in average. The number of photons is derived from the measurement of the power fed to the fiber input and the known light source wavelength. The dependencies of the system quantum efficiency at 1.3 µm wavelength on the SSPD bias current for both channels is presented in the fig. 5 (open symbols and left axes). Different critical currents of the SSPDs are caused by the difference in meander strip width or film thickness and is usually not a problem for practical application. At the same time the difference in system quantum efficiency between the channels is mainly connected with different quantum efficiencies of the installed SSPDs and in coupling efficiencies due slight misalignment with the fiber output which occurs during SSPD installation.

Together with QE, the dark count rate plays an important role for practical application of the detection system. We define the dark counts as spontaneous detection events when the input of the fiber is completely blocked. For SSPDs, the dark count rate varies exponentially with the bias current. Fig. 5 (closed symbols and right axes) shows the dependence

Fig. 3. Two-channel single-photon detection system design.

Fig. 4. SSPD photoresponse waveform transient


4

of the dark count rate on the bias current for both channels. One can see that both channels have similar characteristics. The deviation from the exponential slope is connected with a parasitic illumination and the thermal background radiation. It is worth noting that at a dark count rate of 1-2 counts per second the system still have quantum efficiency about 10%. In many applications, this fact allows one to use SSPDs without gating, a method widely used with APD for dark counts reduction.

For time-correlated single-photon counting, pulse to pulse timing jitter is an essential parameter of the detector, determining the temporal resolution of the entire setup. It was shown that the intrinsic jitter of the SSPD (when light is delivered to SSPD in free space) is 18 ps including jitter of the read-out electronics [3]. This result was obtained by direct measurement of delay between the laser trigger pulse and the SSPD response. Here we present the results of jitter measurement performed on the fiber-coupled system in a different way, i.e. by a photon correlation technique. We are using correlated photon pairs at 1560 nm generated by parametric downconversion in a periodically poled LiNbO3 waveguide [12]. The two photons of a pair are emitted simultaneously, within their coherence time of about 1 ps. Therefore, the histogram of the measured difference in the detection time of the two photons gives us the information on

the timing jitter of the detectors. We made two measurements: The first one with two detectors, one giving the start signal and the other the stop signal for a time-amplitude converter (see

fig. 6). Thus, we obtain the total jitter of the two detectors jtotal

2=j12+j2

2. In the second measurement both photons are detected by the same detector, the stop photon being delayed by about 50 ns (10 m of fiber) (see fig. 7). In this case we obtain jtotal

2 = 2j12 Both measurements give almost identical

results. The FWHM of the histogram showing the measured time differences has a width of 105 ps. Hence, we obtain a timing jitter of 74 ps for one detector. This value includes all electronic jitter of the pulse discriminators and TAC. The total electronic jitter have been evaluated to be under 30 ps with a pulse generator giving the start and stop signal to the discriminators. In comparison, the timing jitter of InGaAs APD is typically in the order of 200-300 ps. This timing jitter can be reduced for some diodes by increasing the biasing voltage, however in return, the dark count rate is increased considerably.

We then set to investigate the maximum counting frequency of the SSPD system. To this aim, an attenuated 750 nm diode laser with repetition frequency variable between 10 and 80 MHz was coupled in the SSPD input fiber, with more than one photon per pulse to ensure SSPD transition for almost every pulse. The voltage pulses from the SSPD were filtered through

28 30 32 34 36 38 40 42 44 4610-4

10-3

10-2

10-1

100

101

102

Channel 1

Current, µA

Syst

em q

uant

um e

ffici

ency

, %

10-1

100

101

102

103

104

105

Dar

k co

unts

, s-1

12 14 16 18 20 22 24 26 2810-3

10-2

10-1

100

101

102

Syst

em q

uant

um e

ffici

ency

, %

Current, µA

100

101

102

103

104

105

Channel 2D

ark

coun

ts, s

-1

Fig 5. System quantum efficiencies at 1.3 µm wavelength (open symbols, left axes) and dark counts rates (closed symbols, right axes) measured as functions of bias currents at temperature below 2 K.

Fig. 6. Histogram of measured detection time differences between two detectors.

Fig. 7 Histogram of measured time delay between two photons from the same pair detected by the same detector.


5

a 2.5 MHz - 6 GHz bias tee, amplified using two 20 dB, 2 MHz - 3 GHz amplifiers and measured on a 1 GHz bandwidth oscilloscope. Fig. 8 and 9 show the oscilloscope traces taken with 10 and 40 MHz repetition rates, respectively. Both measured rise time from 20% to 80% of 530 ps and the 80%-20% 5.1 ns fall time are limited by the SSPD kinetic inductance - indicates a maximum intrinsic detection rate of this system in the order of 100 MHz. However, a slower recovery with a time constant of few tens of ns is observed (see e.g. fig. 8) after the pulse, which can be related to the amplifier bandwidth, as it has been checked separately. For the 10 MHz rate, the 100 ns separation between two subsequent pulses is sufficient to allow signal relaxation to zero. At much higher rates (see e.g. the 40 MHz trace), the slow recovery introduces a memory (or pattern) effect: The pulse amplitude depends on previous pulses. Nevertheless, counting of almost all pulses is still possible. A test performed at 80 MHz (not shown) showed up to 30% fluctuation in peak intensity due to

the memory effect and significant undercounting. It should be noted that in APDs a deadtime of few µs must be imposed after a detection event, in order to allow release of trapped carriers in the device and reduce the afterpulsing probability. This typically reduces the maximum counting frequency of APDs to well below 1 MHz. SSPD hence present a maximum counting frequency about two orders of magnitude larger than APDs. This could be further improved by a better matching of the amplifier bandwidth, and by using meander structures with lower kinetic inductance.

V. TIME-CORRELATED SINGLE-PHOTON MEASUREMENTS As a first demonstration of the application of the SSPD

system, we performed a time-resolved photoluminescence measurement by time-correlated single-photon counting of a GaInNAs quantum well (QW) emitting at 1300 nm at room temperature [13]. The channel 2 of the SSPD is used with a bias current of 19 µA and a corresponding SQE of 5.5%. The QW is excited at room temperature by a pulsed diode laser emitting at 750 nm with a repetition rate of 10 MHz and an average power of 50 nW. The collected photoluminescence is coupled into a single-mode fiber and a band pass filter is used for selecting the QW emission. The fiber is then connected to the input of the SSPD system. The experiment is also performed with an id Quantique InGaAs APD operated in Geiger mode with the same excitation power, a repetition rate of 400 kHz, a gate width of 100 ns and a dead time of 10 µs. The APD has a 10% SQE and a dark count rate of 610 s-1 in these operating conditions. The detector output and the laser trigger signal are fed to the start-stop inputs of a Picoquant (TimeHarp 200) correlation card (with a 35 ps time bin) to produce a histogram of photon detection events. A short lifetime of 50 ps at room temperature has been previously measured for the studied QW [13], which is faster than the temporal resolution of the detectors and thus allows us to

-100 0 100 200 300 400 500 600

0

50

100

150

Inte

nsity

(mV

)

Time, ns

Fig. 8. SSPD response to an attenuated pulsed laser with 10 MHz repetition frequency.

0 200 400 600

-40

0

40

80

120

Inte

nsity

(mV

)

Time, ns

Fig. 9. SSPD response to an attenuated pulsed laser with 40 MHz repetition frequency.

-2 -1 0 1 2 3 41E-4

1E-3

0.01

0.1

1

APD SSPD

Inte

nsity

(nor

mal

ized

)

Time, ns

Fig. 10. Photoluminescence transient of a GaInNAs QW measured by time-correlated single-photon counting using the SSPD and APD detectors.


6

measure the resolution of the SSPD and APD detector. The results are shown on fig. 10.

The full-width at half maximum (FWHM) of the peaks after dark noise subtraction are of 590 ps and 410 ps for the APD and the SSPD, respectively, showing an improved temporal resolution using the SSPD. The residual jitter in the experiment is mainly related to the jitter of the laser and of the correlation card. Moreover, an impressive improvement (over two orders of magnitude) is obtained in the dynamic range using the SSPD, due to the extremely low dark count rate. This shows that SSPDs can be used to provide unprecedented sensitivity and resolution in NIR time-correlated spectroscopy.

In order to validate the application of SSPDs to the characterization of non-classical light states at telecom wavelengths, we measured the intensity autocorrelation function g(2)(τ), at any delay τ, of the luminescence of a single InAs/GaAs quantum dot emitting at 1300nm. The g(2)(τ) is exceedingly difficult to measure using InGaAs technology since the high dark count rate will drive the signal-to-noise ratio below unity. Only a few measurements of g(2)(0), using a pulsed pump laser and gating the APDs, have been reported at these wavelengths [14], [15].

The details of the growth methods and optical characterization of the QDs used in these experiments can be found in [16]. Fig. 11 shows the experimental setup used for measuring single photon correlations at 1300nm. A cw diode laser emitting at ~1220nm is used to excite quasi-resonantly electron-hole pairs in the excited state of the charged exciton. The laser beam is focused by a 0.5NA objective onto the sample which is held at 10K in a liquid helium flow cryostat. The photoluminescence from a single QD is collected by the same objective and the signal can be directed to a monochromator equipped with an InGaAs detector for spectroscopy measurements or coupled to a single mode fiber. The correlation measurements are made in a fiber coupled Hanbury-Brown Twiss (HBT) setup. The signal from the charged exciton transition (X+ see fig. 12a) is spectrally isolated by a tunable band pass filter, and fed to the two SSPD detectors through a 50/50 beam splitter. The high frequency

components of the detector output are fed through a bias-T to a 60 dB-gain amplifier with a 2.5 GHz bandwidth. Correlations are measured with the correlation card using a time binning of 139 ps. In this experiment another system with two SSPDs having slightly lower efficiencies is used. The SSPDs are biased to give a SQE=2-5% for dark count rate of 10-30Hz.

The histogram from the correlation measurement, shown in Fig. 12b, was measured on the X+ emission (Fig. 12a). For time delays greater than 10ns (not shown) a decrease of the correlation function is observed: this bunching behaviour, already studied for short-wavelength QDs, indicates that after emission of a photon from the positive trion the QD remains charged allowing the re-excitation of the charged exciton state. For short time delays τ≤2ns, an antibunching dip is observed, confirming the sub-Poissonian statistics of the light emitted by the trion line. Fitting the correlation function with the expression reported in [17], we obtain a g(2)(0 )=0.29±0.02 demonstrating single photon emission. We underline that this type of experiment has never been demonstrated at telecommunication wavelengths using APDs, which directly shows the potential of application of SSPDs, with much lower dark count rate than APDs, in quantum optics.

START STOP

Diode laser cw 1220nm

BS 50/50

M O N O C H R O M A T O R

InGaAs detector

f=100mm f=2mm NA=0.5

f=8mm NA=0.5

PCI bus

Correlation card:PicoquantTimeHarp200

~2K

4.2K

DC bias RF amp

RF amp

Flip mirror SMF fiber 50/50 BS SMA cable

BP filter

QD

LHe

LHe cryostat 10K Capillary

SSPDs

Bias - T

Bias - T

Vacuum

PUMP

START STOP

Diode laser cw 1220nm

BS 50/50

M O N O C H R O M A T O R

InGaAs detector

f=100mm f=2mm NA=0.5

f=8mm NA=0.5

PCI bus

Correlation card:PicoquantTimeHarp200

~2K

4.2K

DC bias RF amp

RF amp



BP filter BP filter

QD

LHe

LHe cryostat 10K Capillary

SSPDs

Bias - T

Bias - T

Vacuum

PUMP

Fig. 11. Diagram of the micro-photoluminescence setup and fiber

coupled HBT setup used to measure the intensity correlation function at 1300nm.

-10 -5 0 5 100

20

40

0.00.51.01.52.02.53.03.5

1270 1275 1280 1285 1290

Coi

ncid

ence

cou

nt

Time, ns

Cor

rela

tion

func

tion(b)

(a)

g(2)(0) = 0.29

X

PL in

tens

ity (a

rb. u

n.)

Wavelength, nm

X+

Fig. 12 (a) Single QD spectroscopy under quasi-resonant excitation showing the exciton (X) and charged exciton (X+) emission lines. (b) Intensity correlation measurement on the filtered X+ emission, demonstrating single photon emission at 1300nm.


7

VI. CONCLUSION We have reported a two-channel, fiber-coupled single-photon detection system which provides at the same time a system quantum efficiency of 10% at 1300 nm, a dark count rate <10 s-1, a timing resolution <100 ps and a maximum counting frequency >40 MHz. For each of these aspects – apart from the efficiency, which is comparable – the SSPDs outperform InGaAs APDs, in some cases by several orders of magnitude. The system can be conveniently operated with a standard liquid-He dewar, and its application to single-photon correlation experiments has been demonstrated. Future improvements include the optimization of quantum efficiency by monolithic integration with optical microcavities, a better matching of the amplification bandwidth and the reduction of the kinetic inductance to achieve even higher counting rates, and the integration with a closed-cycle cryogen-free cooler. Such a system is expected to represent an invaluable tool in the fields of quantum communication (a successful application of the SSPD for quantum key distribution over 200 km distance has been recently reported [18]), near-infrared spectroscopy and quantum optics, and any other application requiring ultimate sensitivity and temporal resolution.

ACKNOWLEDGEMENTS We are grateful to Dr. H. Riechert, Infineon Technologies, for providing a GaInNAs quantum well sample.

REFERENCES [1] G. Gol'tsman, O. Okunev, G. Chulkova, A. Lipatov, A. Semenov, K.

Smirnov, B. Voronov, A. Dzardanov, C. Williams, R. Sobolewski, "Picosecond superconducting single-photon optical detector" Applied Physics Letters vol.79, no.6, pp.705-707, 2001

[2] A.Verevkin, J.Zhang, R.Sobolewski, A.Lipatov, O.Okunev, G.Chulkova, A.Korneev, K.Smirnov, G.Gol'tsman, A. Semenov, "Detection efficiency of large-active-area NbN single-photon superconducting detectors in ultraviolet to near-infrared range", Applied Physics Letters, vol.80, no.25, pp.4687-4689, 2002.

[3] A. Korneev, P. Kouminov, V. Matvienko, G. Chulkova, K. Smirnov,B. Voronov, and G. N. Gol’tsman, M. Currie, W. Lo and K. Wilsher, J. Zhang, W. Slysz, A. Pearlman, A. Verevkin, and Roman Sobolewski, "Sensitivity and gigahertz counting performance of NbN superconducting single-photon detectors", Applied Physics Letters vol 84, no. 26, pp 5338-5340, 2004

[4] G. Gol'tsman, A. Korneev, O. Minaeva, A. Antipov, A. Divochiy, N. Kaurova, B. Voronov, D. Pan, A. Cross, A. Pearlman, I. Komissarov, W. Slysz, and R. Sobolewski, "Middle-Infrared to visible-light ultrafast superconducting single-photon detector", presented at the Applied Superconductivity Conference 2006, Seattle, USA, August 27, 2006

[5] R. H. Hadfield, M. J. Stevens, S. S. Gruber, A. J. Miller, R. E. Schwall, R. P. Mirin, Sae Woo Nam, " Single photon source characterization with a superconducting single photon detector", Optics Express 13, 10846, 2005

[6] W. Slysz, M. Wegrzecki, J. Bar, M. Gorska, V. Zwiller, C. Latta, P. Bohi, I. Milostnaya, O. Minaeva, A. Antipov, O. Okunev, A. Korneev, K, Smirnov, B. Voronov, N. Kaurova, G. Gol’tsman, A. Pearlman, A. Cross, I. Komissarov, A. Verevkin, R. Sobolewski, "Fiber-coupled single-photon detectors based on NbN superconducting nanostructures for practical quantum cryptography and photon-correlation studies", Appl. Phys. Lett. Vol. 88, 261113, 2006.

[7] G. N. Gol’tsman, K. Smirnov, P. Kouminov, B. Voronov, N. Kaurova, V. Drakinsky J. Zhang, A. Verevkin, and R. Sobolewski, "Fabrication of Nanostructured Superconducting Single-Photon Detectors," IEEE

Transactions On Applied Superconductivity, Vol. 13, No. 2, pp. 192-195, 2003

[8] K. M. Rosfjord, J. K. W. Yang, E. A. Dauler, A. J. Kerman, V. Anant, B. M. Voronov, G. N. Gol'tsman, K. K. Berggren, "Nanowire single-photon detector with an integrated optical cavity and anti-reflection coating", Optics Express, Vol. 14, Issue 2, pp. 527-534, 2006

[9] I. Milostnaya, A. Korneev, I. Rubtsova, V. Seleznev, O. Minaeva, G. Chulkova, O. Okunev, B. Voronov, K. Smirnov, G. Gol'tsman, W. Słysz, M. Wegrzecki, M. Guziewicz, J. Bar, M. Gorska, A. Pearlman, J. Kitaygorsky, A. Cross and R. Sobolewski "Superconducting single-photon detectors designed for operation at 1.55-µm telecommunication wavelength" Journal of Physics: Conference Series 43, pp. 1334–1337, 2006

[10] A. J. Kerman, E. A. Dauler, W. E. Keicher, J. K. W. Yang, K. K. Berggren, G. Gol'tsman, and B. Voronov, "Kinetic-inductance-limited reset time of superconducting nanowire photon counters", Appl. Phys. Lett. 88, 111116, 2006

[11] O. Okunev; G. Chulkova; I. Milostnaya; A. Antipov; K. Smirnov; D. Morozov; A. Korneev; B. Voronov; G. Gol'tsman; W. Stysz; M. Wegrzecki; J. Bar; P. Grabiec; M. Gorska; A. Pearlman; A. Cross; J. Kitaygorsky; R. Sobolewski, "Registration of infrared single photons by a two-channel receiver based on fiber-coupled superconducting single-photon detectors", Proceedings of the CAOL 2005. 2nd International Conference on Advanced Optolectronics and Lasers. IEEE Cat No 05TH8809. 2005: 282-5 VOL. 2

[12] S. Tanzilli, H. de Riedmatten, W.Tittel, H. Zbinden, P. Baldi, M.de Micheli, D. Ostrowski, N. Gisin. PPLN Waveguide for Quantum Communication. Eur. Phys. J. D 18, 155-160 (2002)

[13] A. Markus, A. Fiore, J. D. Ganière, U. Oesterle, J. X. Chen, B. Deveaud, M. Ilegems, and H. Riechert, Appl. Phys. Lett. 80, 911 (2002)

[14] M. B. Ward, O. Z. Karimov, D. C. Unitt et al., "On-demand single-photon source for 1.3µm telecom fiber," Appl. Phys. Lett. 86 (20), 201111, 2005

[15] C. Zinoni, B. Alloing, C. Monat et al., "Time-resolved and antibunching experiments on single quantum dots at 1300 nm," Appl. Phys. Lett. 88 (13), 131102, 2006

[16] B. Alloing, C. Zinoni, V. Zwiller et al., "Growth and characterization of single quantum dots emitting at 1300 nm," Appl. Phys. Lett. 86, 101908 (2005)

[17] S. C. Kitson, P. Jonsson, J. G. Rarity et al., "Intensity fluctuation spectroscopy of small numbers of dye molecules in a microcavity," Phys Rev A 58 (1), 620-627 (1998)

[18] H. Takesue, S. W. Nam, Q. Zhang, R.H. Hadfield, T.Honjo, K.Tamaki, and Y. Yamamoto "Quantum key distribution over 40 dB channel loss using superconducting single photon detectors" Nature Photonics Vol. 1, 343 (2007)

Alexander Korneev graduated from the Moscow State

Pedagogical University (MSPU) in 2000. He received his Ph.D. degree in radio-physics from MSPU in 2006. Now he is a senior scientist at the Radio-Physics Research and Educational Center of MSPU. His research interests are in studies of non-equilibrium processes in superconducting

single-photon detectors. Yury Vachtomin defended his Ph.D. degree in radio-physics at Moscow

State Pedagogical University (MSPU) in 2005. Now he is a senior researcher at the Radio-Physics Research and Educational Center of MSPU. His research interests are in the research and development of superconducting single-photon detectors and superconducting terahertz direct detectors and mixers.

Olga Minaeva received her M.S. degree from the Moscow State

Pedagogical University (MSPU) in 2005. Now she is a Ph.D. student there. Her scientific interest is in the study of non-equilibrium phenomena in superconducting single-photon detectors. She also performs development and characterization of fast superconducting single-photon detectors.

Alexander Divochiy received his M.S. degree from the Moscow State

Pedagogical University (MSPU) in 2006. Now he is a Ph.D. student there. Currently he is working on the development of superconducting single-photon detectors and their characterization. His scientific interest is in the research


8

on the non-equilibrium processes in the superconducting single-photon detectors.

Konstantin Smirnov received his Ph.D. degree in radio-

physics from Moscow State Pedagogical University (MSPU) in 2000. He is a leading research scientist at the Department of Phisics of MSPU. His research interests are in the fields of electron-phonon interactions in two-dimensional semiconductor structures, the quantum Hall effect, and

design and fabrication of superconducting thin-film devices. Oleg Okunev defended his Ph.D. thesis in radio-physics

and optics at Moscow State Pedagogical University (MSPU) in 2004. Currently he is a senior research scientist in MSPU. His scientific interests are in the areas of kinetic phenomena in superconducting films and millimeter-wave techniques.

Gregory Gol'tsman received his Ph.D. degree in radio-

physics and Doctor of Science (Sc.D.) degree in semiconductor and dielectric physics from Moscow State Pedagogical University (MSPU), Moscow, Russia, in 1973 and 1985, respectively. Currently he is a professor at the Physics Department of MSPU and a head of the Radio-

Physics Research and Educational Center at MSPU. His scientific interests are in the areas of superconductivity, nonequilibrium phenomena in superconductors, semiconductors, far-infrared spectroscopy, as well as terahertz direct detectors and mixers. He is an author and co-author of more than 200 papers and conference presentations.

Andrea Fiore graduated in Electronic Engineering (1994) and in Physics

(1996) at University of Rome "La Sapienza". From 1994 to 1997 he carried out his PhD thesis on nonlinear frequency conversion in semiconductor waveguides at Thomson CSF Central Research Laboratory (Orsay, France). He then worked at University of California at Santa Barbara (1997-98) on multiple-wavelength arrays of vertical-cavity surface-emitting lasers, at the Ecole Polytechnique Fédérale de Lausanne (1998-2001) on quantum dot lasers, and at the Institute of Photonics and Nanotechnology of the Italian CNR (2001-2002) on nanostructured photonic devices. Since September 2002, he is leading the activity on Quantum Devices at the Ecole Polytechnique Fédérale de Lausanne as Assistant Professor of the Swiss National Science Foundation. Dr. Fiore has co-authored over 65 journal publications, 35 contributions to conferences and given 18 invited talks.

Matthäus Halder is a PhD student since 2003 at Geneva

University in the Group of Prof. Gisin. After having done his diploma theses at the University of Munich (LMU) in quantum cryptography, he is now working in the field of quantum communication.

Corentin .Jorel received an engineering degree at the

Grenoble physics engineering school (ENSPG). After that he did his PhD (2004) on the development of Superconducting Tunnel Junctions for photon counting in astronomy between the condensed matter research department of the CEA-Grenoble and the Grenoble

astrophysics laboratory (LAOG). He is now in charge of the SSPD (Superconducting Single-Photon detectors) thematics at the Geneva University.

Hugo Zbinden received his diploma in physics in

1987 for his work on radiocarbon dating and his PhD in 1991 for his work on rare-earth solid-state lasers from the University of Berne. In 1993 Hugo Zbinden joined the group of applied physics of the University of Geneva. He is leading the experiments in the field of quantum

information. In 2001 he co-founded id Quantique, a spin-off committing to

commercialize quantum cryptography and other quantum technologies.

Single-photon detection system for quantum optics …cms2.unige.ch/gap/optics/wiki/_media/publications:bib:sspdinqo.pdf · quantum communication and quantum optics, for example antibunching-type

Documents