8th June 2006 Rob Lambert - UoE PPE Seminar 1

Towards a Quantum Dot-Based Single Photon Source

Supervisor: R. J. Warburton

Assistant Supervisor: P. Dalgarno

Moderator: P. Barker

photon antibunching from a single charge-tuneable quantum dot

8th June 2006 Rob Lambert - UoE PPE Seminar 2

SET Award

Best Physics Student:

• National competition

• Round 1: Reference

• Round 2: Current Performance

• Round 3: Project Synopsis

• Round 4: Interview

Sponsors/Judges Include:

• NPL

• IOP

• UCL

• Imperial

8th June 2006 Rob Lambert - UoE PPE Seminar 3

Single Photon Applications

Quantum Computing:

• Spin-interaction and interference for binary 1 or 0 states

• Quantum Teleportation

• Ultrafast low-energy computation

Quantum Cryptography:

• Indeterministic nature of quanta

• Verifiably secure information transfer

8th June 2006 Rob Lambert - UoE PPE Seminar 4

Single Photon Applications

Quantum Cryptography:

• Imagine you’re an Octopus

8th June 2006 Rob Lambert - UoE PPE Seminar 5

Single Photon Applications

…… providing single photons can be reliably produced on demand.

Quantum Computing:

• Spin-interaction and interference for binary 1 or 0 states

• Quantum Teleportation

• Ultrafast low-energy computation

Quantum Cryptography:

• Indeterministic nature of quanta

• Verifiably secure information transfer

8th June 2006 Rob Lambert - UoE PPE Seminar 6

Single Photon Production

Highly Attenuated Laser Pulse Trains (RT):

• Multi-photon pulses (Poissonian probability)

Single Atom/Molecule emission (~RT):

• Difficult selection and microscopy <Å

• Bleaching and Blinking … Not Ideal

Single-QD regimes (4.2K):

• nm-scale island of semiconductor

• Artificial atom

• 3D confining potential

• ~Parabolic quantum well

• None of above problemsAtomic force micrograph image

Axel Lorke

8th June 2006 Rob Lambert - UoE PPE Seminar 7

Looking at Single QDs at 4.2K

Confocal Microscopy:

• high-stability, liquid-helium-cooled, solid-immersion-lens-enhanced, diffraction-limited, confocal microscope

Liquid He @ 4.2K

AR Window

To motor controller Liquid He @ 4.2K

AR Window

To motor controller

BS1

BS2

8° Angled Single-Mode (980nm) Fibre

IP

SampleSampleSample

Piezoelectric

X-Y-Z motors

ITO Gate

Vg

Objective lens NA=0.65

SIL

Annealed Contact

Lock-in Amplifier

I

Vg

vosc VDAC

AC+DC Mixer

Lock-in Amplifier

I

Vg

vosc VDAC

Lock-in Amplifier

I

Vg

vosc VDAC

AC+DC Mixer

826nm ExitationLaser Diode

CCD camera

Photodiode8° Angled Fibre

Collected PL

BandpassFilter

Holographic notch filter

8th June 2006 Rob Lambert - UoE PPE Seminar 8

Looking at Single QDs at 4.2K

Confocal Microscopy:

• Not as complicated as it looks…. Slightly

• Light from other dots thrown away.

826nm Laser

Up to 10MHz pulsed

Sample @ 4.2 K

~950nm PL

Delayed Emission

Collected by fibre

Fibre Delivered

High NA lensFibre Apertures

Motorized

Piezo

stage

Vg

8th June 2006 Rob Lambert - UoE PPE Seminar 9

MISFET Structure

gatesubstrate

backc onta ct

tun nelb arr ie r

A lAs/GaA s

blo ck ingb arrier

InAs dots

Vg1

Vg2

Fermi energy

R. J. Warburton, C. Schäflien, D. Haft et al., Nature 405, 926

(2000)Vg

2 > Vg1

n-GaAs

Metal - Insulator - Semiconductor Field Effect Transistor:

8th June 2006 Rob Lambert - UoE PPE Seminar 10

Charge-Tuneable PL

X0 X1-

2X0 2X1-

X2- 2X0h

Example State Configurations (spin omitted)

Exiton Spectroscopy:

8th June 2006 Rob Lambert - UoE PPE Seminar 11

Sample Preparation

Gate Contact:• Higher collection efficiency = better statistics

• Improve transmission, move to Indium Tin Oxide

15000 14000 13000 12000 11000 10000 90000.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0700 750 800 850 900 950 1000 10501100

ITO NiCr

Rel

ativ

e Tr

ansm

issi

on

Wavenumber (cm-1)

Wavelength (nm)

8th June 2006 Rob Lambert - UoE PPE Seminar 12

Sample Preparation

Solid Immersion Lens SIL:

• Lower Refraction at Semiconductor Interface

• Increase Refractive Index

• Wider angle = More Light + Better resolution = Easier selection

With SIL

Without SIL

8th June 2006 Rob Lambert - UoE PPE Seminar 13

Experiment

QD PL

0.001

Start

Stop

BEAM SPLITTER

SPAD 1

SPAD 2

FILTER

FILTER

Photon Counting Statistics:

• Hanbury-Brown Twiss Interferometer

8th June 2006 Rob Lambert - UoE PPE Seminar 14

Auto-Correlation

QD PL

0.001

Start

Stop

BEAM SPLITTER

SPAD 1

SPAD 2

FILTER

FILTER

Photon Counting Statistics:

• Hanbury-Brown Twiss Interferometer

• CW and Pulsed antibunching 2nd-order autocorrelation

-30 -20 -10 0 10 20 300.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Nor

mal

ized

cou

nts

τ (ns)

g(2)trough=0.261

τr = 1.00ns

g(2)(0)=0.24123τr = 0.9131ns

g(2)(0)=0.30437

30 -100 -50 0 50 100-10

0

10

20

30

40

50

60

70

80

X0 Vg= -0.9V

Coi

ncid

ence

s

τ (ns)

0.960.95

1.07

1.011.01

1.00

0.0763

1.00

1.02

1.01

0.98

1.08

0.941.02

8th June 2006 Rob Lambert - UoE PPE Seminar 15

Anti-Bunching

Unexpected results:

• Gate-voltage dependant antibunching

• Difference in dependance between X0 and X1- for the same dot

8th June 2006 Rob Lambert - UoE PPE Seminar 16

Results

Photon Counting Statistics:

X1-

8th June 2006 Rob Lambert - UoE PPE Seminar 17

Results

Photon Counting Statistics:

00.0

0.2

0.4

0.6

-0.75V

-0.76V

-0.80V

-0.84V

-0.90V

0.08

0.15

0.12

0.12

0.08

Sca

led

Cou

nts

t (ns)

g(2)(0)Vg

12.5 12.5

X0

8th June 2006 Rob Lambert - UoE PPE Seminar 18

Time Resolved Photo-Luminescence

Photon Counting Statistics:

• Direct comparison to TRPL data

• Neutral Exciton fine structure

0 10 20 30 40 50

10

100

1000

Coi

ncid

ence

s

time (ns)

X0 Dual Exponential

Decay

DarkBright

Spin-Flip

Fine structure discussed in J.M. Smith, P.A.Dalgarno, R.J. Warburton et. al PRL v. 94 n. 19 (2005) pp 197402

X0 X0

X1- X1-

Always Bright

8th June 2006 Rob Lambert - UoE PPE Seminar 19

Comparison with TRPL

Photon Counting Statistics:

• Decay lifetimes/amplitudes vary with gate voltage

• Coulomb Blockade Model

• Agreement with Autocorrelation

-0.85 -0.80 -0.75 -0.70 -0.65 -0.60

0.1

1

10

Life

time

(ns)

Gate Voltage Vg (V)

Primary

Secondary

Autocorrelation

-0.85 -0.80 -0.75 -0.70 -0.65 -0.600.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Primary % Secondary %

Com

pone

nt R

atio

Gate Voltage Vg (V)00.0

0.2

0.4

0.6

-0.75V

-0.76V

-0.80V

-0.84V

-0.90V

0.08

0.15

0.12

0.12

0.08

Sca

led

Cou

nts

t (ns)

g(2)(0)Vg

8th June 2006 Rob Lambert - UoE PPE Seminar 20

-30 -20 -10 0 10 20 300

40

80

120

160

200

0

0

0

1

1

2

2

Coi

ncid

ence

s

τ (ns)

Cross-Correlation

Photon Counting Statistics:

• Quantum Cascade

Peak g(2) = 2.50

Minimum g(2) = 0.41

exponential decay

τr=0.744ns

≡ τd(X0)=0.746ns

2X0 X0

START STOP

Vacuum

8th June 2006 Rob Lambert - UoE PPE Seminar 21

Summary

• ITO+SIL = Better collection efficiency = Higher statistics+Better Resolution

• Verified Antibunching @ 4.2K = Single Photons from QD

• Single Photons = Proof of Second Quantization

• Agreement with TRPL on the same QD for the first time

• Quantum cascade witnessed = Possible trigger for photons on demand

8th June 2006 Rob Lambert - UoE PPE Seminar 22

Outlook

QDs provide Single Photons

• Secure information

• Good news for Bob and Alice

QD Single photons are indistiguishable

• Quantum Teleportation

• Good news for Scotty/Kirk

QD single photon sources for supercomputers?

•Good news for Boffins everywhere

Possibly a lot more talks on this subject in the future

• Bad news for Jellybabies