superconducting nanowire single-photon detectorsridl.cfd.rit.edu/products/talks/dvw/karl...

111114-rochester

Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology,

Cambridge, Massachusetts 02139, USA

Superconducting Nanowire Single-Photon Detectors

K. K. Berggren

berggren@mit.edu 1

111114-rochester

http://messenger.jhuapl.edu/news_room

111114-rochester

3

Free-Space Optical Communications

receiver 109 km

Boroson, Biswas, Edwards, "Overview of NASA's mars laser communications demonstration system," FREE-SPACE LASER COMM. TECH. XVI 5338: 16-28, 2004

Toyoshima, et. al., “Comparison of microwave and light wave communication systems in space applications,” Opt. Eng. 46 015003 (2007)

transmitter

111114-rochester

4

Free-Space Optical Communications

0000

0001

time

0010

0011

0100

0101

0110

0111

1000

1001

1010

1011

1100

1101

1110

1111

0000

0001

0010

0011

0100

0101

0110

0111

1000

1001

1010

1011

1100

hv hv

Encodes 01011001 (4 bits/photon)

receiver 109 km

Boroson, Biswas, Edwards, "Overview of NASA's mars laser communications demonstration system," FREE-SPACE LASER COMM. TECH. XVI 5338: 16-28, 2004

Toyoshima, et. al., “Comparison of microwave and light wave communication systems in space applications,” Opt. Eng. 46 015003 (2007)

transmitter

111114-rochester

5

Photons

• Illumination in eye from 1 pixel of laptop at 100 km in 1 sec

• Energy inversely proportional to wavelength

– Longer wavelength => harder to detect

5

Berggren

111114-rochester

Superconductive Nanowire Single-Photon Detector

6

Superconductive Nanowire

Single-Photon Detector

90 nm

3 µm

proximity-effect-correction features

electrode

electrode

Gol’tsman et al., APL, (2001).

Yang et al., IEEE TAS (2005)

111114-rochester

7

lens

cornea

retina

optic

nerve

vitreous

humor

Eye Anatomy

111114-rochester

8

lens

cornea

retina

optic

nerve

vitreous

humor

sapphire

substrate

antireflection

coating

cavity

NbN

mirror contact pad

light

Eye Anatomy

sapphire niobium

nitride

gold….

111114-rochester

9

Tapetum Lucidum of a Cat

9

Berggren

Bernstein and Pease, “Electron Microscopy of the Tapetum Lucidum of

the Cat” Journal of Cell Biology, Vol. 5, 35-39, (1959)

~ 100 nm

retina

tapetum

lucidum

111114-rochester

10

lens

cornea

retina

optic

nerve

vitreous

humor

sapphire

substrate

antireflection

coating

cavity

NbN

mirror contact pad

light

Eye Anatomy

tapetum

lucidum

sapphire niobium

nitride

gold….

111114-rochester

Characteristics of Photon Detectors

• Efficiency

• Reset time

• Jitter

• Dark count rate

11

time

hν incident photons

with no signal

time

incident photons

blocked by

earlier signal

time t1 t3

varying delay between

photons and signals

voltage

time

voltage pulses with no

corresponding photon

t2

111114-rochester

Nanowire Single-Photon Detector

superconducting nanowire

single-photon detector

photomultiplier tube

avalanche photodiode

transition edge sensor

104

104 100

1 / jitter

[MHz]

1 / reset time

[MHz] device

efficiency

[%]

• VLSI device evaluation

• LIDAR

• Communication

– quantum

– interplanetary

APPLICATIONS

Comparison at 1.55 µm

111114-rochester

Device Operation

DEVICE OPERATION

111114-rochester

Superconductive Nanowire Behavior

I

V

critical current, Ic

Berggren

superconducting

111114-rochester

< 100 nm

niobium nitride

Detection Mechanism

Explanation

4 nm

superconductor is biased near its transition

detector

Ibias

Rload

L

111114-rochester

Detection Mechanism

Explanation

photon-induced hotspot forces bias current above critical density

< 100 nm

niobium nitride 4 nm

detector

Ibias

Rload

L A. D. Semenov, G. N. Gol’tsman, and A. A. Korneev, “Quantum detection by current carrying superconducting film,” Physica C, vol. 351, pp. 349–356, 2001.

111114-rochester

Detection Mechanism

Explanation

resistive barrier spans nanowire

< 100 nm

niobium nitride 4 nm

detector

Ibias

Rload

L

111114-rochester

Detection Mechanism

Explanation

4 nm

< 100 nm

niobium nitride

resistance grows from heating

detector

Ibias

Rload

L

111114-rochester

Detection Mechanism

Explanation

4 nm

< 100 nm

niobium nitride

current is diverted

detector

Ibias

Rload

L

111114-rochester

< 100 nm

niobium nitride

Detection Mechanism

Explanation

4 nm

superconductivity is restored

detector

Ibias

Rload

L

111114-rochester

< 100 nm

niobium nitride

Detection Mechanism Explanation

4 nm

superconductivity is restored

detector

Ibias

Rload

L

21

111114-rochester

22

111114-rochester

< 100 nm

niobium nitride

Detection Mechanism

Explanation

4 nm

bias current is restored

detector Rload

L

Ibias

Kerman, Dauler, Keicher, Yang, KB, Gol'tsman, Voronov, Applied Physics Letters, (2006)

111114-rochester 24

Cross-section transmission-electron

micrograph

90 nm

sapphire

cavity dielectric

Au mirror

NbN

24 Rosfjord et al., Opt. Express (2006)

sapphire fib

er

gold

~λ/4

anti-reflection

coating

NbN

SNSPD integrated in an optical cavity

111114-rochester

Summary of Device Performance

25

1 ps, 1550 nm

optical pulse

28.5 ps FWHM

-100 -80 -60 -40 -20 0 20 40 60 80

0.01

0.1

1

No

rma

lize

d C

ou

nts

(a

.u.)

Time (ps)

Time (ns)

Voltage (mV)

3 ns

Kerman et al. in APL (2006)

Background Counts

~100 s-1

(97.5% bias, 2.7K)

70-nm-wide detector, 140-nm pitch

210-nm-thick cavity

0

0.4

0.8

0 0.5 1

~ 65 %

~ 70 %

Dete

ctio

n e

ffic

iency

bias current / Ic

Collaboration with

Lincoln Laboratory

Time (ps)

Norm

aliz

ed C

ou

nts

(a.

u.)

111114-rochester

Fabrication

FABRICATION

111114-rochester

Basic Fabrication Flow hydrogen silsesquioxane (HSQ)

(electron-beam resist)

gold contact pads

sapphire substrate

niobium nitride (from collaborators)

111114-rochester

Basic Fabrication Flow

niobium nitride nanowire

hydrogen silsesquioxane (HSQ)

(electron-beam resist)

gold contact pads

sapphire substrate

niobium nitride (from collaborators)

111114-rochester

29

Basic Fabrication Flow

HSQ

anti-reflection coating

niobium nitride nanowire

hydrogen silsesquioxane (HSQ)

(electron-beam resist)

gold contact pads

sapphire substrate

niobium nitride (from collaborators)

111114-rochester

30

Cavity Fabrication Flow

HSQ

Au contact pads

sapphire substrate NbN

detector

1nm/

120nm

Ti/Au

mirror

HSQ cavity

HSQ

ARC

111114-rochester

Photon Statistics

APPLICATION

31

111114-rochester

10 µm

32

111114-rochester 33

Pseudo-Thermal Source Setup

Ground

Glass

4-element

SNSPD

Cryostat

Fiber

12 V DC Fan (up to 6000 rpm)

Grating-Stabilized

CW Diode Laser

1070 nm

Scattered Light

Speckle

PC

6-channel

FPGA

Time-stamp each channel

~7 ns bins

111114-rochester

3-photon coincidences

t1

t2

2-photon coincidence conditions:

t1 = 0

t2 = 0

t1 = t2

3-photon coincidence :

t1 = t2 = 0

t

34

10 µm

Ele

ment

1

Ele

ment

2

Ele

ment

3

Ele

ment

4

111114-rochester 35

3rd-order Intensity Correlations

“High-order temporal coherences ofchaotic and laser light”, Stevens, Baek, Dauler, Kerman, Molnar, Hamilton, Berggren, Mirin, and Nam, Optics Express, 18, 1430 (2010)

g(3)(0,0) = 5.87 ~ 3!

111114-rochester

Photon Statistics

NANO-ANTENNAE

111114-rochester

The Three Keys to Efficiency

active

area

coupling

efficiency absorptance

v

resistive state

formation

r

t

× × ×

37

v

threshold

detection

× ×

v

111114-rochester

Antenna – Integrated Detector

small pitch

top view

cross section

large pitch fill the gap with gold

v

t

v

t

v

t

111114-rochester

Optical Nano-Antenna

Au … …

TM

k

NbN

39

Final Version

111114-rochester

Optical Nano-Antenna

Au … …

TM

k

NbN

40

111114-rochester

Fabrication

ion-beam cross-section

fabrication challenges

• e-beam writing on thick HSQ

• gold migration in evaporation

sapphire

Au

Pt

HSQ

Au

600 nm

80 nm

9 mm

Au

pad

Au

pad

SiO2/

Ti/Au

111114-rochester

• Nano-antennae improve collection, permitting 3 x area, with same reset time

Nano-Optical Antenna for Nanowire Detectors

• 600-nm pitch, 9-mm-by-9-mm area

• 47% device efficiency • 5 ns reset time

Hu et. al. Optics Express, 2010

111114-rochester

Photon Statistics

SUB-30-NM DEVICES

43

80 nm

60 nm 40 nm 30 nm

20 nm 10 nm

44

80 nm

60 nm 40 nm 30 nm

20 nm 10 nm

45

80 nm

60 nm 40 nm 30 nm

20 nm 10 nm

46

111114-rochester

30 nm

1 µm

30-nm-wide-nanowire SNSPD

47

111114-rochester

w = 90 nm

Standard vs Ultranarrow-Nanowire SNSPDs

Constrictions

48

111114-rochester

constricted

constriction-free

more constricted

Standard vs Ultranarrow-Nanowire SNSPDs

Constrictions

49

w = 90 nm

111114-rochester

constriction-free

w = 30 nm

Standard vs Ultranarrow-Nanowire SNSPDs

Constrictions

50

w = 90 nm

constricted

more constricted

111114-rochester

constriction-free

w = 30 nm

Standard vs Ultranarrow-Nanowire SNSPDs

Constrictions

51

w = 90 nm

constricted

more constricted

A. D. Semenov, G. N. Gol’tsman, and A. A. Korneev, “Quantum detection by current carrying superconducting film,” Physica C, vol. 351, pp. 349–356, 2001.

111114-rochester

constriction-free

w = 30 nm

Standard vs Ultranarrow-Nanowire SNSPDs

Constrictions

52

w = 90 nm

constricted

more constricted

A. D. Semenov, G. N. Gol’tsman, and A. A. Korneev, “Quantum detection by current carrying superconducting film,” Physica C, vol. 351, pp. 349–356, 2001.

111114-rochester

constriction-free

constriction-free

w = 30 nm

Standard vs Ultranarrow-Nanowire SNSPDs

Constrictions

53

w = 90 nm

constricted

more constricted

constricted

111114-rochester

Negligible responsivity above 2 μm

w = 90 nm

54

Standard vs Ultranarrow-Nanowire SNSPDs

Gol’tsman et al., IEEE Trans. Appl. Supercond (2011)

w = 55 nm

w = 100 nm

111114-rochester

Negligible responsivity above 2 μm

w = 90 nm

55

Standard vs Ultranarrow-Nanowire SNSPDs

Gol’tsman et al., IEEE Trans. Appl. Supercond (2011)

w = 55 nm

w = 100 nm

111114-rochester

λ = 0.5 μm

w = 85 nm

56

Detection efficiency vs IB and wavelength

λ = 2 μm

111114-rochester

λ = 0.5 μm

w = 30 nm

57

λ = 2 μm

Detection efficiency vs IB and wavelength

111114-rochester

58

0

3.6

7.2

η (%)

w = 85 nm

w = 50 nm

w = 30 nm

Detection efficiency vs IB, wavelength and width

111114-rochester

Fabrication

SIGNAL TO NOISE RATIO

59

111114-rochester

60

Low signal to noise ratio

111114-rochester

61

Low signal to noise ratio

111114-rochester 62

“A Cascade Switching Superconducting Single Photon Detector,”

M. Ejrnaes, R. Cristiano, O. Quaranta, S. Pagano, A. Gaggero, F.

Mattioli, R. Leoni, B. Voronov, and G. Gol'tsman,

Appl. Phys. Lett. 91, 262509 (2007)

111114-rochester

Superconducting Nanowire Avalanche Photodetectors

(SNAPs)

63

111114-rochester

Superconducting Nanowire Avalanche Photodetectors

(SNAPs)

1.4 µm

30 nm

64

111114-rochester

IB RA

VO

UT

+

- Basic model of SNAP operation

65

LS

111114-rochester

IB RA

VO

UT

+

- Basic model of SNAP operation

66

LS

111114-rochester

IB RA

VO

UT

+

- Basic model of SNAP operation

67

LS

111114-rochester

IB RA

VO

UT

+

- Basic model of SNAP operation

68

LS

111114-rochester

SNAPs improve the SNR

4-SNAP

3-SNAP

2-SNAP

SNSPD

69

111114-rochester

Experimental results

λ = 1550 nm

SNSPD

2-SNAP

70

111114-rochester

Experimental results

λ = 1550 nm

SNSPD

2-SNAP

71

threshold current

111114-rochester

Experimental results

λ = 1550 nm

3-SNAP

SNSPD

2-SNAP

72

111114-rochester

Experimental results

λ = 1550 nm

3-SNAP

SNSPD

2-SNAP

73

In which bias range are 3-SNAPs working as single-photon detectors?

111114-rochester

IB RA

VO

UT

+

- SNAP operation below the avalanche current

74

LS

111114-rochester

IB RA

VO

UT

+

-

75

LS

arm

SNAP operation below the avalanche current

111114-rochester

IB RA

VO

UT

+

-

76

LS

SNAP operation below the avalanche current

111114-rochester

IB RA

VO

UT

+

-

77

LS

SNAP operation below the avalanche current

111114-rochester

IB RA

VO

UT

+

-

78

LS

trigger

SNAP operation below the avalanche current

111114-rochester

IB RA

VO

UT

+

-

79

LS

SNAP operation below the avalanche current

111114-rochester

IB RA

VO

UT

+

-

80

LS

SNAP operation below the avalanche current

111114-rochester

IB RA

VO

UT

+

-

81

LS

SNAP operation below the avalanche current

111114-rochester

IB RA

VO

UT

+

-

82

LS

SNAP operation below the avalanche current

111114-rochester

Experimental results

λ = 1550 nm

3-SNAP

SNSPD

2-SNAP

83

threshold current

avalanche arm-trigger

111114-rochester 84

increasing

power

IAV

η Power Dependence Explained

111114-rochester

η Power Dependence Explained

85

photon arm +

photon trigger

photon arm +

dark count trigger

IAV

111114-rochester

Electro-thermal simulation of SNAPs

86

initiating

secondary

Iout

avalanche superconducting

initiating:

secondary:

normal / superconducting boundary

photon

n

s

n

s

J. Yang et al. IEEE TAS (2007)

F. Marsili et al., APL (2011)

111114-rochester

3-SNAP in “arm-trigger” regime

87

Pseudo 2-SNAP

trigger

superconducting

I3

I2

I1

Iout

arm

111114-rochester

Inter-arrival time measurements

time

hν voltage

Avalanche mode:

88

111114-rochester

Inter-arrival time measurements

time

hν voltage

Avalanche mode:

tD1 tD2 tD3 tD4

τΦ1 τΦ2 τΦ3 τΦ4

89

Dt

f t f tt

111114-rochester

Inter-arrival time measurements

time

T voltage

Arm-trigger mode:

A T A T

90

111114-rochester

Inter-arrival time measurements

time

T voltage tD1 tD2

Arm-trigger mode:

A T A T

t2Φ1 t2Φ2

91

D 2t

f t f tt

111114-rochester

Inter-arrival time histograms

92

IB < IAV

IB > IAV

F. Marsili et al., Nanolett. (2011)

111114-rochester

20-nm-wide-nanowire SNAPs

20 nm

1.1 µm

93

111114-rochester

20-nm-nanowire-width SNAPs

2-SNAP

4-SNAP

3-SNAP

94

λ = 1550 nm

111114-rochester

The Future

• Scaling sensitivity out to longer wavelengths

– (is high-efficiency single-photon detection possible at 10 um?)

• Understanding the source of jitter

– Intrinsic to material? Electronic? Thermal? Electro-thermal?

– Dependent on design/architecture? Engineerable?

• Can we break the 1 ns speed limit?

– Need to investigate new materials

– Need to investigate new device designs

• Is near-100% efficiency possible?

95

111114-rochester

NbN-LL-171-A: 05-4-1

W=11 p=100

10 nm

1.1 µm

96

111114-rochester

Acknowledgements

MIT Quantum Nanostructures and

Nanofabrication Group

Vikas Anant (now at PhotonSpot)

Francesco Bellei

Andrew Dane

Eric Dauler (Lincoln Lab Fellow, now at

Lincoln Laboratory)

Charles Herder (undergraduate)

Xiaolong Hu (now at U. of Michigan)

Hasan Korre

Francesco Marsilli (post-doc, now at

NIST)

Faraz Najafi

Kristen Sunter

Joel Yang (ASTAR Fellowship, now at

A*STAR)

MIT Lincoln Laboratory

Eric Dauler

Andrew J. Kerman

Richard Molnar

Bryan Robinson

Scott Hamilton

NIST Martin J. Stevens Burm Baek Sae Woo Nam Richard P. Mirin

Other Collaborators

M. Csete (U. of Szeged)

Boris Voronov (MSPU)

Gregory Gol’tsman (MSPU)

Sponsors

AFOSR, IARPA, DARPA, NASA, NSF

111114-rochester

Photon Statistics

END OF PRESENTATION

98