DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution Air Force Research Laboratory Integrity Service Excellence DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution Optoelectronics & Photonics Gernot S. Pomrenke, PhD Program Manager AFOSR/RTD Air Force Research Laboratory MARCH 2014
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Air Force Research Laboratory
Integrity Service Excellence
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Optoelectronics &
Photonics
Gernot S. Pomrenke, PhD
Program Manager
AFOSR/RTD
Air Force Research Laboratory
MARCH 2014
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 2 DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
2014 AFOSR SPRING REVIEW
BRIEF DESCRIPTION OF PORTFOLIO (3001C): Explore light-matter interactions at the subwavelength- and nano-scale between
metals, semiconductors, & insulators.
Explore optoelectronic information processing, integrated photonics, and
associated optical & photonic device components for air and space platforms to
transform AF capabilities in computing, communications, storage, sensing and
surveillance … with focus on nanotechnology approaches. Explore chip-scale
optical networks, signal processing, and novel-sensing .
LIST SUB-AREAS IN PORTFOLIO: - Nanophotonics & Plasmonics: Plasmonics, Photonic Crystals, Metamaterials,
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 7 DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
60
Years
Why did silicon win for electronics?
It was the best platform for integrated systems!
An
opportunity
to support
shared
fabrication
for silicon
photonics
Optoelectronic Systems Integration in Silicon
Prof Michael Hochberg, Univ of Delaware, OpSIS Foundry
opsisfoundry.org/
http
://n
an
op
ho
ton
ics.e
ce
.ud
el.ed
u/a
bo
ut_
the
_la
b.h
tml
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 8 DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
OpSIS - Scaling toward complex systems
• We’re seeing a Moore’s Law-like growth in
system complexity
• Doubling time is around a year
• Filling a reticle with photonic devices of
~500 square microns gets us to ~1.7M
devices
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 9 DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
OpSIS Institute at UD An opportunity to support shared fabrication for silicon photonics
OpSIS Objective:
•Make integrated photonic fabrication flows easily and cheaply accessible to the research and development community through MPW shared-shuttle processes
•Drive process and tool development and standardization
•Provide educational resources and support to the community
•Develop an ecosystem of service and equipment providers to help move the silicon photonics community forward
~150 users around the world
Half corporate, half academic
0
50
100
150
2011 2012 2013 2014(est)
pre-run 001 002,3,4 005,6,7,8
Active OpSIS Users
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 10 DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
OpSIS Research Activities
• Development of design tools & design elements
• Demonstrations of complex systems
• Development of methodology & measurement tools/techniques
• Design automation
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 11 DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
• High-efficiency waveguide-coupled photodiodes at 1.2 A/W (not yet published)
• World-record low-loss silicon modulators at 30 GHz (MZI) and 45 GHz (ring)
• Ultra-low loss passives library – crossings, couplers, junctions, etc at both 1550 and 1310 nm
• Hybridized lasers at 200 kHz linewidths
• World-record bipolar amplifiers at 80+ GHz for electronic-photonic integration
OpSIS
320 Gbit/second (40Gx8) Over 1 fiber with WDM
2 km reach <10 mm2 silicon
World record transmitter (with Bergman Group, Columbia)
2.4 Tbit/second
transceiver
demonstration
in progress now
86 GHz amplifier
Telecom-grade laser
40G data
8 channel
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Qianfan Xu, Rice University
Reconfigurable Optical Directed-Logic Circuits
Objective:
• Develop a scalable and reconfigurable directed
logic architecture for low-latency optical
computing.
• Demonstrate a direct logic circuit with the highly
integrated silicon photonic technology.
• Demonstrate a reconfigurable optical switch with
high speed, high extinction ratio and low
insertion loss as the basic building block.
• Develop a complete circuit simulation tool.
Approach:
• A cellular geometry with a regular row-column layout of
reconfigurable optical switches.
• Each switch has three distinctive operation modes.
• Direct mapping the truth table of an arbitrary logic function
to the operation modes of the optical switches.
• Double-ring based optical switches as the unit cell.
• A multi-spectral implementation will be realized.
• Both electro-optic and all-optical logic will be investigated.
Impact:
• Overcome the limitation of
conventional optical logic.
• Boost the performance of digital
systems for real-time applications
such as video analysis, object
recognition, missile guidance,
visualization and battle management.
• Enable highly efficient packet-switched
optical networks on chip.
• Provide a scalable platform for various
applications and for future expansion.
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 13 DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Experimental Demonstration
Circuits based on the multi-spectral
implementation are fabricated in CMOS
photonics foundry at IME Singapore.
Each switch has an embedded p-i-n
junction for logic input and a micro-heater
for reconfiguration.
y
l1
l2
l1
l
x1 x2
l1& l2
A 2×2 switch array can
calculate any logic function
between two logic inputs.
0 1
0 X X
1 X X
A B
Q. Xu and R. Soref, Opt. Express 19, 5244-5259 (2011) [email protected] http://www.ece.rice.edu/ece/xugroup/
fiber
ring
optical logic circuits fabricated through OPSIS program
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 14 DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Objective:
•Investigate the best architectural modalities to insert silicon photonic interconnect technology into systems of interest to the DOD to overcome performance and energy bottlenecks in emerging SoCs
Impact:
•This research will aim to meet aggressive ITRS data rates of multiple TB/s and a power goal of less than 1pJ/bit in multi-core SoCs, far exceeding what is possible with conventional electronics today
Approach:
•Design innovative heterogeneous network topologies and protocols that effectively combine multiple stacked layers of optical links with electrical wires
•Create new techniques for enhancing memory-access performance with optically connected DRAM and mechanisms for energy-efficient reconfiguration of opto-electronic components at run-time
Integrated Optoelectronic Networks for
Application-Driven Multicore Computing
PI: Sudeep Pasricha, Colorado State University
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 15
SiGeSn: Basic Features and Properties
• CMOS - compatible (with SiGe); strain control,
stressors for high velocity MOSFET channels
• Variable bandgap of SiGeSn: ≈ 1.1 to 0.11 eV; (1.1 to
11 μm)
• Lattice/ strain control (aSn = 6.5 Å; aGe = 5.6 Å: aSi =
5.43 Å) (strain engineering)
• SiGeSn: highly conductive; covalent bonds; no
scattering by polar optical phonons, or Reststrahlen
absorption (limitation of III-V’s); direct bandgap (from
0.2 to 0.6 eV); enhanced luminescence; high optical
absorption; high speed (small effective mass)
Group IV
elements
Beyond Silicon Photonics - SiGeSn: a promising material system
The first group-IV material with a widely tunable 2D
compositional space, SiGeSn makes it possible to
decouple band gap and lattice constant, enabling
wide-range applications from thermal imaging to
photovoltaics to lasers
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 16
-Integrated circuits are moving to Ge-rich SiGe
materials; this requires the larger lattice constant of
GeSn for stressor layers to increase active channel
carrier mobility
-Experiments and theory suggest that Ge1-ySny has
a direct bandgap for Sn contents above 6 %
(wavelengths above 2 μm) for efficient detectors
and emitters in the mid-IR region, and compatibility
with SiGe integrated circuits. (Beyond Si Photonics)
-Mid-IR devices: GeSn provides excellent mid-IR
performance (beyond 1.8 μm limit of Ge) with possibility of
multi-pixel arrays compatible with integrated circuits and
lasers
Three prime motivators for
SixGeySn1-x-y Research
Wide Area
Motion
Imagery
Full
Motion
Video
Hyperspectral
Imagery
Enhanced
Resolution at
Range
Higher Altitude
and
Greater Standoff
Day and NIGHT
TRADITIONAL
ASYMMETRIC
HME/IED/CBRNE
Missile Warning
Active Sensor/Seeker
Detection
Pre-LaunchSensor/SeekerCountermeasure
Expendable
Post-Launch Laser
Countermeasure
Counter EO/IR Adjuncts on AAA, RF SAM, LBR, DE Weapons…
RF/EO/IR Threat Warning
and ProtectionPassive EO/IR Sensing in
Contested Environments
Laser Radar Sensing in Contested Environments
Phenomenology and
Innovative Concepts
Modeling and
Analysis
Laser Source
Research
Detector/FPA
Research
Optics/Aperture
Research
Cross-Range [m]
Ran
ge [m
]
F-4 1D Ladar Back-Projection Reconstrunction
-8 -6 -4 -2 0 2 4 6 8 10
-8
-6
-4
-2
0
2
4
6
8
10
0
20
40
60
80
100
120
140
160
180
80 100 120 140 160 180 200 220
NET
D a
t R
ange
(m
K)
Detector Operating Temperature (K)
nBn
XBn
MCT Rule 7
BLIP
InSb, 10 mm
NS4 InSb, 20 mm, 7500 ft AGL
NS25 InSb, 20 mm, 7500 ft AGL
NS25, InSb, 20 mm, 20000 ft AGL
MCT/Si
NETD Goal: 75 mK
Long Term
Concepts
[note that previous detector
arrays using III-V’s such as
InGaAs are incompatible with
SiGe circuits]
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 17
Challenges of SiGeSn
• Increase Sn content in GeSn above 14 atomic %, to study the
direct bandgap compositions, and for mid-IR operation
(present reported devices use about 4 % Sn, ~45% Sn in SiSn)
• Improve thermal stability (above 400 °C) of high Sn content
alloys for subsequent device processing (contacts, etching,
annealing)
• Identify a robust Sn source for CVD: Sn hydrides are unstable
after weeks: SnCl4 is more stable, but needs study
• Investigate fundamental properties of SiGeSn (bandgap,
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Electrical Interfaces and Modulation for
Plasmon Laser (optically pumped)
WEB plasmon laser :
• Constructed by crossing semiconductor
nanostrips over metal nanostrips with an 5-
10nm insulator gap layer
• A square shaped plasmon laser cavity at each
intersection
• Both metal and semiconductor strips can be
used as electrical contacts
Direct laser amplitude modulation with low
power consumption:
•16 dB modulation depth for a peak bias of 4V
•Opposite operation - injecting electrons and
holes into the active cavity region - can lead
to an electrically pumped semiconductor
plasmon laser
2 mm
Future: develop efficient electrical injection strategies and materials & Multiple WEB plasmon lasers at different wavelengths
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
FY11 MURI: Hybrid Nanophotonics
Team: Prof Mark Brongersma, Stanford Univ, PI & team lead; Zhang – UCBerkeley; Miller
& Fan – Stanford; Shalaev - Purdue; Atwater & Painter – CalTech; Lukin & Park - Harvard
Objective: Explore the full potential of hybrid nanophotonic components for on
chip optical communication by combining the best aspects of metal and
semiconductor photonics
- a suite of high performance hybrid
nanophotonic devices and systems
- new simulation tools that can deal
with the hybrid/quantum nature of the
components
- new materials and “meta” building
blocks for hybrid devices and systems
-new optical characterization tools to
analyze hybrid devices with nano-
scale resolution
-new fabrication techniques that
enable scalable fabrication of
complex hybrid devices
- Plasmonics for extremely light concentration and
enhancing light matter-interaction at nanoscale
- Semiconductors for active functions, quantum
behavior
scaling of photonics will enable devices and circuits that offer lower power operation, higher speed,, denser integration
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Objective: Electrically-driven plasmon source
Objective – explore & develop electrically-
driven plasmon sources that via short
plasmonic waveguides can be coupled to
low loss dielectric waveguides
Key findings/Implications
Technical Approach
Embed p-GaAs /InGaAs/n-GaAs quantum well source
(lem = 970 nm) at in a plasmonic slot waveguide
Use same metals for waveguide as for electrical contacts
Exploit Purcell effect to enable effective coupling to the
emission to a single plasmonic mode
Optical simulations show possibility to effectively
couple QW emission into a plasmon slot
waveguide (>80%) Using e-beam lithography, plasmonic QW sources
coupled to plasmonic slot waveguides have been
realized
(a)
Basic, ultra-compact circuit elements have been
implemented
Brongersma Group, Stanford
(b)
(d) (e) ENZ modulator
Electrically-driven plasmon source
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Objective: Hybrid Nanophotonic Photodetectors
demonstrate efficient photodetection into
subwavelength structures compatible with CMOS
processing techniques
Technical Approach
Embed a semiconductor detector in a nanometallic slit
Use same metals for contacts and optical confinement
Exploit resonances in the combined metal/dielectric
resonance to enhance absorption efficiency and allow
tunability of spectral response
Silicon fin detectors showing strong absorption near 850 nm
Good quantum efficiency in only 170 nm Si thickness
Key findings
Germanium fin detectors extend absorption beyond 1550 nm
Silicon two-fin detectors for simultaneous two color detection
David Miller Group, Stanford
1.5mm
v1
v2
Hybrid Nanophotonic
Photodetectors
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 29
Cavity-Free, Matrix-Addressable Quantum Dot Architecture for On-Chip Optical Switching
-- Rashid Zia and Arto V. Nurmikko, Brown University
Device integration enabled by the precise placement of Quantum Dots in Waveguide-Addressable Matrix.
Schematic Design of
Colloidal Quantum Dot
Optical Circuit
Architecture exploits strong localized fields from plasmonic & nanophotonic waveguides to:
(1)Direct quantum dot emission for scalable on-chip single photon sources, and
(2) Mediate strong interaction between quantum dots for low-photon number optical switches
Objective: Develop a cavity-free architecture for waveguide-integrated single
photon sources and optical signal processing at the few to single photon level.
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 30
The Scalable Approach builds on the recent development of highly-ordered colloidal quantum dot arrays in Nurmikko Lab through a two-step process of dielectric encapsulation and electrostatic self assembly.
50 nm
TEM Image of Silica-clad QD
Process allows one to exploit precision of E-Beam Lithography to position single QDs in large-scale arrays
Cavity-Free, Matrix-Addressable Quantum Dot Architecture for On-Chip Optical Switching
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Cuong Dang et al., Nature Nanotechnology, Vol: 7, 335–339, (2012)
DBRs, R>99%
Threshold ~
60µJ/cm2 <N> ~
0.53 exciton
• First colloidal QD-based vertical-cavity surface-emitting laser (VCSEL)
• Very low threshold, well defined coherent beam at R,G, & B wavelengths
Demonstration of Colloidial
QD-VCSEL
Development of high-quality colloidal QDs enables new applications for these scalable materials in photonic devices.
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
LDOS Engineering in 1-D
Waveguides
1D waveguide thickness allows the direct emission into a desired mode.
Theory
Experiment
Pol
TE mode TM mode TE mode
Fabrication: electrostatic deposition and e-beam evaporation
Randomly oriented QDs present challenge for directing light emission.
Demonstrated that engineering the local optical environment can preferentially couple QD emission into specific modes.
Optimized a fabrication technique to safely embed QDs within high index dielectric layers for 1-D waveguides.
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Robust (Statistical) Nanophotonics
by Repeatable, Scalable Emitter Integration
• Waveguide-Integrated Single Photon Sources
• Demonstrating a scalable process for embedding colloidal QDs in dielectric
waveguides
Method easily integrated with any nanofab process.
Reusable templates allow for repeated study of the same resonant nanostructure with different single emitters -- i.e. building statistical data sets without variations of fabrication errors or irregularities.
SEM images of QD integrated with gold rods
Scanning image of QD embedded waveguide
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution
Impact and Implication for Future Work
Quantum Emitter Substrate
Silica-clad Quantum dots: 1.7-3.5 NV center nano-diamond: ~3 Defects in ZnO nano-crystal: 8.7-10.3 Transition metal ions doped MgO nano-cube: 12-13
The electrostatic self-assembly technique can be easily extended to other quantum emitters and material systems. By choosing proper materials for the patterns and substrates, one can precisely place various single quantum emitters.
DISTRIBUTION STATEMENT A – Unclassified, Unlimited Distribution 35
Technology Transitions
Technology Transitions
• Students to Intel, Professor to Max Planck Institute