Nanolasers: The current status of the trailblazer of synergetics

Transcription

Nanolasers: The current status of the trailblazer of synergetics
Nanolasers:
Current Status of Trailblazer of Synergetics
Cun-Zheng Ning
[email protected], http://nanophotonics.asu.edu
School of Electrical, Computer, and Energy Engineering
Arizona State University
Support: ARO, AFOSR, DARPA, NASA, SFAz
[email protected], http://nanophotonics.asu.edu
Miniaturization of Semiconductor Lasers
mm ~ cm scale
(1962)
0.5 mm
First laser diode (GaAs) Lincoln Lab
100 nm ~ µm scale
(2012)
[email protected], http://nanophotonics.asu.edu
Engineering Photon-Semiconductor Interaction
Radiative coupling between light and semiconductor
rcv ∝ ρ ph (ω ) •ρ e (E )
3D cavity:
1 1
κ
(
)
ρ cav
ω
=
ph
Vc π ω − ω0 2 + κ 2
(
)
Decreasing Vc
Purcell Enhancement
3


λ
(
)
ρ cav
ω
3
ph
0
 Q
=
FP = 3 D
2 
ρ ph (ω0 ) 4π  Vc 
Density of Electronic States
Size Quantization
2
ω
3D
Free space: ρ ph (ω0 ) = 2 0 3
π c
Cavity Size Reduction
Density of Photonic States
1  2m
 2
2 
2π  h
Bulk
ρ e3 D (E ) =
QW:
*
m
ρ e2 D (E ) = 2
πh
*
2 m*
QWR: ρ (E ) =
πh
1D
e
QD:



3
2
E − Eg
1
E − En
ρ e0 D = 2δ ( E − En )
[email protected], http://nanophotonics.asu.edu
Engineering the Densities of States
Density of Photonic States
ρ cav
ph (ω0 )
Density of Electronic States
ω02
ρ (ω0 ) = 2 3
π c
More efficiently use of photons
Size Quantization
Cavity Size Reduction
3D
ph
More efficiently use of electrons/holes
More efficiently coupling photons and semiconductors
[email protected], http://nanophotonics.asu.edu
Why Nanolasers?
From Application Point of View
• Optical and electronic integration, size compatibility with
electronic devices
• VLSI photonics: more functions in smaller volume
• On-chip light sources (e.g., micro and nano-fluidic)
• General trends in nanotechnology development: the
smaller the better
• Other new applications not envisioned yet, but will be
enabled once smaller and smaller lasers are available
[email protected], http://nanophotonics.asu.edu
Moore’s Law in Photonics
M.K. Smit, Moore’s law in photonics
[email protected], http://nanophotonics.asu.edu
Moore’s Law in Photonics
Technology Breakup
M.K. Smit, Moore’s law in photonics
[email protected], http://nanophotonics.asu.edu
Moore’s Law for Microelectronics
[email protected], http://nanophotonics.asu.edu
Challenges for Nanophotonics
• Size, Size, and Size
a) Passive devices (waveguides): > λ / 2n , single mode fiber:
5 µm; silicon wire or other semiconductor nanowire: 100-200 nm
b) Active devices: (lasers): gain length required to achieve threshold:
1-100 µm, large footprint, difficult for integrate
• Complexity, diversity, and cost: diversity of devices and
materials, small market share of each device, expensive
manufacturing
• Compatibility with silicon for integration with electronics
light emitting materials: non-silicon (III-V, II-VI) such as
GaAs, InP
• No silicon light source (external to CMOS)
• …
[email protected], http://nanophotonics.asu.edu
Examples of Smallest Lasers…( before 2007)
(what is in common: pure dielectric waveguide structures)
nanowire
InAs/AlGaAs single layer of QD, 60 nW output
(Painter group, Opt. Exp. 2006)
substrate
55 nm-think ZnO nanocrystal layer is
dispersed on a SiO2 disk of 10 microns in
diameter, Liu et al, APL (2004)
Erbium doped silica disk of 60 microns in diameter on a
silicon stem (Kippenberg, PRA 2006) (optically pumped)
(Optically pumped, RT-CW, smallest, PC laser,
Baba’s group, InGaAsP/InP, Opt. Exp.2007)
Park et al. Science 305, 1444
(2004).
[email protected], http://nanophotonics.asu.edu
Questions
• Can lasers be made even smaller?
• What is the ultimate size limit?
• How about electrical injection, rather than optical?
• Can you make a laser that is smaller than vacuum
wavelength in all three dimensions (DARPA
NACHOS program)?
NACHOS (Nanoscale Architectures for Coherent Hyper-Optic Sources)
Goals: Electrical injection, room temperature, subwavelength in all 3dimensions
[email protected], http://nanophotonics.asu.edu
How to Make Smaller Cavities?
• Pure dielectric cavities are not adequate
• Metallic, especially plasmonic structures offer
potential hope
Plasmonics, Spasers, Before 2007….
• Bergman and Stockman, PRL 2003
• Stockman and Bergman, Laser Phys, 2004
• Nezhad, Tedz, and Fainman, Opt. Exp. 2004
• Maier, Opt. Comm. 2006
• Miyazaki and Kurokawa, PRL 2006
[email protected], http://nanophotonics.asu.edu
Plasmon Photon Coupling
Plasma/Plasmon: Longitudinal excitation of electron motion
(in metals or doped semiconductors)
Drude model:
ε (ω ) = 1 −
ω p2
ω 2 + iγω
1/ 2
 Ne 2 
ωp = 

m
ε
 0 
Surface Plasmon or Surface Plasmon Polariton: Coupled
EM wave and plasmon excitation at the interface of a
dielectric layer and a metallic layer.
ε1
ε2
kz =
ω
c
ε1ε 2
ε1 + ε 2
[email protected], http://nanophotonics.asu.edu
Surface Plasmon Polariton (SPP)
2π
k z′
λ eff =
λ0 540 nm
=
= 15.4
35 nm
λeff
k z′
Silver
Semiconductor
Silver
SPP wave along the interface
k z′′
I = I 0 e 2i ( k ′+ik ′′) z
(eV)
Near SPP Resonance:
~ nm
1) Huge wave compression (35 nm)
2) Strong localization ( few nm)
3) Huge loss (3.6 million 1/cm)
[email protected], http://nanophotonics.asu.edu
Lasers, Spasers, and Photon-Plasmon Coupling
k
k=
ω
c
n
BPP
SPP
Plasmonicity
SPASERS: Bergman and Stockman,
Phys. Rev. Lett. 90, 027402 (2003)
SP
ωp
2
BP
ωp
ω
[email protected], http://nanophotonics.asu.edu
Light Coupling to SPP Mode:
Dramatic Purcell Enhancement
InGaN QW-Silver (8nm)
by GaN thickness:
(Neogi et al, PRB66,
153305(2002)
Neogi et al, PRB, 2002
[email protected], http://nanophotonics.asu.edu
Feasibility of a Semiconductor-Core Metal-Shell
(Jan 2007 SPIE Paper)
Maslov-Ning , 2007
[email protected], http://nanophotonics.asu.edu
First Experimental Demonstration of
the Semiconductor-Metal Core-Shell Laser
M. Hill et al. Nat. Photonics, 1, (2007),589
[email protected], http://nanophotonics.asu.edu
A Zoo of Nanolaser Designs… after 2007
(What is in Common? Everyone Likes Metals)
Hill , 2007
250 nm
Aluminum
Alumina
p-GaN
PMMA
MQW
Ti/Au
230 nm
Ni/Au
Chuang-Bimberg Group
Hill -Ning 2009
Maslov-Ning , 2007
n-GaN
(Wu Group, Berkeley)
Sapphire
(Zhang Group, 2009
Yang Group
Fainman UCSD)
Noginov/Shalaev, 2009)
(Lieber, Harvard, Park, Korea)
Painter Group 2009
[email protected], http://nanophotonics.asu.edu
Summary of Short History and Status
• Design and theoretical study: Maslov and Ning, Proc. SPIE 6468, (2007)64680I
• 1st experimental demonstration: M. Hill et al. Nat. Photonics, 1, (2007),589
• Electrical injection sub-half-wavelength laser: Hill et al, Opt. Exp., 2009
• Metal encased in a doped shell: Noginov et al., 2009
• Wire on a metal surface: Oulton et al., 2009
• Metal-semiconductor disk laser, Parahia et al, APL, 95 (2009) 201114
• Optically pumped lasing at RT: Nezhad et al, Nat. Phontonics, 4, (2010),395
• Nano patch laser: Yu et al., Opt. Exp. , 18 (2010) 8790
• Nano pan laser: Kwon et al. (2010), Nano. Lett, 10, (2010),3679
• Metallic cavity VCSEL, RT operation, Lu et al, Appl. Phys. Lett, 96, 251101 (2010)
•
Goals: Sub-wavelength, CW RT operation, electrical injection
[email protected], http://nanophotonics.asu.edu
Semiconductor-Metal Core-Shell Nanolaser
500
n-contact
polyamide
nm
Ag
Si3N4
500
300
n-InP
InGaAs
p-InP
Ti/Pt/Au p-cntct
InP Subs.
Circular pillars: diameters ~280nm to 500nm
Rectangular pillars: 6 and 3 micron long; core width
~80nm +/- 20nm to ~340nm
Hill, Marell, Leong, Smalbrugge, Zhu, Sun, Veldhoven, Geluk, Karouta, Oei, Nötzel, Ning, Smit, Opt. Exp.,17, 11107 (2009)
[email protected], http://nanophotonics.asu.edu
Lasing in a Silver-Coated 90+40 nm-Thick Pillar:
(thickness below half-wavelength limit)
4
7
x 10
Run6 row 1 dev #17, 10K, 130uA
Total light output vs current
Run 6 row 1 dev #18 10K
5
6
x 10
90nm
Run6 row 1 dev #17, 10K
6
40 microamps
60 microamps
80 microamps
100 microamps
450
400
4
3
Intensity (counts)
350
Intensity (counts)
Intensity (counts)
5
5
300
250
200
150
100
50
2
3
2
0
1300
1350
1400
1450
wavelength (nm)
1500
1
1
0
1250
4
1300
1350
1400
1450
Wavelength (nm)
1500
1550
0
0
50
100
150
current (microamps)
200
250
Optical thickness = 3.1X90 + 2X20X2 + 2X10X2 = 400 nm < DL = λ / 2 = 670 nm
(Semicond.) (Dielectric)
(Metal)
The thinnest electrical injection laser ever demonstrated !
Hill, Marell, Leong, Smalbrugge, Zhu, Sun, Veldhoven, Geluk, Karouta, Oei, Nötzel, Ning, Smit, Opt. Exp.,17, 11107 (2009)
[email protected], http://nanophotonics.asu.edu
More Recent Progress on Nanolasers with
V < λ3
2009, pulse, LT
10
8
2012, CW, RT
wide linewidth
7
6
5
6
4
3
4
2
linewidth (nm)
2012, CW, RT
final goal!
A
Integrated intensity (a.u.)
2011, CW 260K
2
1
0
0.0
0
0.5
1.0
1.5
2.0
Current (mA)
[email protected], http://nanophotonics.asu.edu
[email protected], http://nanophotonics.asu.edu
[email protected], http://nanophotonics.asu.edu
1E15
1e-4
1e-3
1e-2
0.1
0.2
0.4
0.6
0.8
1.0
5.00E+014
β
L ∝ I 18
1E12
1E11
0.0004
0.0008
current (uA)
1
1e-3
1e-2
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1E13
1E10
0.00E+000
0.0000
β=
1E14
Intensity
photon density
1.00E+015
Disappearance of Threshold in Nanolasers
β=
L∝I
∝V
0.0012
1E9
1E-7
L∝I
1E-6
1E-5
Current (A)
1E-4
Similar to disappearance of phase transitions in
finite or lower dimensional systems, threshold
becomes increasingly soft and disappears
eventually in nanolasers as size decreases
Thresholdless? Yes.
But approaching zero (lasers)? or infinity (LED)?
[email protected], http://nanophotonics.asu.edu
Photon Statistics Near Threshold
g ( 2) (τ ) =
I (t ) I (t + τ )
I (t )
thermal light
2
There seems to be a
more fundamental
limitation (beyond gain
requirement and cavity
mode etc) to how small
laser can be made:
When size becomes too
small, we cannot make
a laser with the coherent
state emission
g(2) (0) = 2
Threshold
infinity?
laser light
g(2)(0) =1
Gies et al, PRA ,75, 013803,2007
[email protected], http://nanophotonics.asu.edu
Further on Nanolasers…
[email protected], http://nanophotonics.asu.edu
Summary
• Nanolsaer research is driven both by the engineering
of photonic and electronic densities of states and by
future applications in nanophotonic integrated
systems
• Plasmonic structures provide an interesting means
for the cavity miniaturization to reach nanoscale
• There seems to be a fundamental limit in terms of
how small a laser cavity can be: when the cavity is
so small that spontaneous emission coupling to the
lasing mode is approaching 100%, the threshold
becomes increasingly high!
[email protected], http://nanophotonics.asu.edu
Thank You!
[email protected], http://nanophotonics.asu.edu