|
Photonics Research
Aerosol Doping
Blue/UV LEDs with Very High Extraction and
Photon Conversion Efficiencies Using Textured Quantum Wells
Broadband Light Generation by Noncollinear
Parametric Down-conversion
Center for Subsurface Sensing &
Imaging Systems
Entangled Photons Generated via Optical
Parametric Downconversion in Periodically Poled Lithium
Niobate (PPLN)
High Resolution 4Pi Microscopy
The Holy Grail of silicon (Si) photonics: optical gain
in Si
Laboratory for Lightwave Technology
LCI – Loss Code Imager – DSX
Mission
Light in Deterministically Generated Aperiodic Optical
Materials
Microelectromechanical Flexure PZT Actuated
Optical Scanner: Static and Resonance Behavior
A Multi-Watt 980nm Single Mode Fiber Laser
III-Nitride Intersubband Optoelectronic
Devices
A Novel High Temperature Optical
"Thermocouple"
Numerical Aperture Increasing Lens Microscopy
Optical Pharmacokinetics: noninvasive
measurements of drug concentrations in tissue
Optical Properties of Carbon Nanotubes
Quantum Optical Coherence Tomography
Resonant Cavity Imaging Biosensor
A "Whispering Gallery"
Intra-cavity Biosensor
Aerosol Doping
Theodore Morse
Lightwave Technology Laboratory
A. Rare earth doped optical fibers
Initial inhomogeneities at the molecular level in solution
doping of rare earth doped optical fiber lasers, as a consequence
of the difference in the coordination number of silica and
the rare earth oxide, tend to foster clustering. Clustering
at high radiation levels can lead to radiation "darkening",
which can impact performance of fiber levels at high powers.
This is of particular importance for Q-switched devices.
We believe that our technique for using an aerosol of organo-metallic
precursors, is the best process for the fabrication of rare
earth doped fiber lasers, since it easily allows the introduction
of any glass forming and glass modifying constituent into
the preform. Below is a table presented by J. Simpson of
Bell Laboratories, which attests to these advantages. This
becomes more important when the additional aspect of radiation
darkening is considered.
B. Nanoscale oxide particles and polycrystalline
ceramics
Nanoscale oxide particles have many uses in a host of industries
as individual small oxide particles or as precursors to
the formation of polycrystalline ceramics. In the latter,
the ideal particle would be single nanosize crystal with
a particular size distribution and, most importantly, unagglommerated.
These are the characteristics that can lead to an easier
formation of non-scattering (i.e., transparent), bulk ceramic
objects with near net-shape properties. We will briefly
describe our efforts to achieve such transparent objects
for large ceramic lasers, as well as scintillating crystals
for MRI and PET scans. Such particles naturally have wide
applications as opacifiers.
The common techniques for obtaining small particles are
ball milling, spray pyrolysis, and sol-gel processing, all
of which have their deficiencies. Ball milling is limited
by the final obtainable size, as well as the introduction
of impurities. As the figures below indicate, Spray pyrolysis
and sol-gel processing leads to agglommerated particles
(unless the initial concentrations are so low as to lead
to a reduction in the processing rate). In the figures below,
we show typical results for nano-particle production using
spray pyrolysis and sol gel processing techniques.
 
Fig. 2: Nano-particles formed by spray pyrolysis
Fig. 3: Nano-particles formed by sol-gel process.
The high degree of agglommeration is evident in both examples.
We have learned that particle agglommeration occurs in spray
pyrolysis in regions of large temperature gradients. By
developing a patented burner design to nearly eliminate
radial temperature gradients in a combustion process, we
have been able to synthesize, using both organometallic
and organic precursors,  a
large array of nanoscale oxides that are completely unagglommerated.
These include YAG, yttria, many simple oxides, complex spinels
(magnesium aluminate), and perovskites (barium strontium
titanate). The particles are symmetric, solid single crystals,
unagglomerated with a size distribution in which half of
the particles are less than 30 nm, and the largest particles
are less than 140 nm. A typical TEM photo is shown in the
figure above, which should be contrasted with the above
two figures for sol-gel and spray pyrolysis. Note the absence
of "necking", which is the hallmark of agglommeration.
Fig. 4: Our patented process
without agglommeration.
Using nan oparticles
of silica as precursors for a modified sol-gel synthesis,
Lucent Technologies (formerly Bell Laboratories) has spent
150M to develop the following process.Taking up to 70% nanoparticles
of silica, and using a modified sol-gel reaction to "grow"
silica in the interstices between the particles, they were
able to obtain arbitrarily large bulk silica elements with
the final dimensions specified to within 500 microns. The
results are shown in the accompanying figure. We believe
that the same type of technique can be used, if our nanopowders
are employed as unagglommerated precursors, to synthesize
arbitrarily large transparent polycrystalline ceramic materials
for ceramic lasers, transparent ceramic armor, and scintillating
crystals for medical application. Work is in progress, and
we have developed a new aerosol torch that will allow us
to produce unagglomerated nanoparticles of almost any oxide
at a rate sufficient to process large bulk transparent ceramic
materials.
Fig. 5: 5kg SiO2 glass body by sol-gel method
for communications-grade optical fiber.
References
1. Bhandarkar, Journal of the American Ceramic Society,
87 (7), 1180-1199 (2004)
2. Killian, Morse; Aerosol Science and Technology, 34 (1),
227-235, (2001)
3. Sun, et al; Journal of Crystal Growth, 260 (2004), 171-175
Back to top
Blue/UV LEDs with
Very High Extraction and Photon Conversion Efficiencies
Using Textured Quantum Wells
Theodore D. Moustakas
Wide Band-gap Semiconductors Laboratory
http://www.bu.edu/nitrides/WBSL25.htm
Supported by the Department of Energy (DOE) under the Solid
State Lighting program.
This project studies a unique approach to
growing GaN-based LEDs on thick textured GaN quasi-substrates
grown by Hydride Vapor Phase Epitaxy (HVPE). The LED structures
grown on such templates employ III-Nitride textured multiple
quantum wells (MQWs) as the active region, grown by molecular
beam epitaxy (MBE). These LED structures were evaluated
by optical pumping and compared with similar structures
grown on smooth GaN templates [1-3]. The results indicate
that the extraction efficiency from the LED structures employing
textured MQWs is higher than 80%. Furthermore, the internal
quantum efficiency was found to increase very significantly
compared to those employing atomically smooth quantum wells.
The experimental results indicate that the improvement in
the internal quantum efficiency is due to minimization of
the quantum-confined Stark-effect (QCSE) since the planes
of the QWs are not perpendicular to the polarization direction.
Such devices have recently been fabricated into electrically
pumped structures and are currently under evaluation (see
Figure 1).
 
Figure 1. Photographs of a blue and a green
LED produced by the method described in the text.
References
1. Nitride LEDs Based on Flat and “Wrinkled”
Quantum Wells (Invited Paper), J. S. Cabalu, C. Thomidis,
I. Friel, T. D. Moustakas, Quantum Sensing and Nanophotonic
Devices, Proc. of the SPIE, Vol. 5732 (2005)
2. Enhanced light extraction through nano-textured GaN interfaces
via supercritical angle scattering, S. Riyopoulos, J. Cabalu
and T. Moustakas, Proc. of the SPIE, Vol. 6013 (2005).
3. Enhanced internal quantum efficiency and light extraction
efficiency from textured GaN/AlGaN quantum wells grown by
molecular beam epitaxy, J. S. Cabalu, C. Thomidis, T. D.
Moustakas, S. Riyopoulos, Lin Zhou and David J. Smith (submitted
to the Journal of Applied Physics).
Back to top
Broadband
Light Generation by Noncollinear Parametric Down-conversion
Silvia Carrasco, Magued Nasr, Alexander Sergienko,
Bahaa Saleh, Malvin Teich,
Juan Torres, and Lluis Torner
Quantum
Imaging Laboratory, Boston University and Universitat Politecnica
de Catalunya, Barcelona, Spain
Supported by NSF, CenSSIS, ARO (MURI), the David & Lucile
Packard Foundation, the
Fulbright Program, and the Government of Spain.
Broadband optical sources are needed for many
current scientific and technological applications, as diverse
as wavelength division multiplexing, signal generation and
amplification for telecommuncations, generation of short
pulses and phase stabilized pulses, optical metrology, tunable
high-precision spectroscopy, and optical coherence tomography
(OCT). A powerful route to generating broadband light is
supercontinuum generation. Currently, pulsed solid-state
lasers in combination with photonic-crystal fibers and tapered
fibers make supercontinuum generation a promising technique
that is being exploited to produce broadband spectral sources.
However, for most applications, a smooth rectangular or
quasi-Gaussian spectrum is desired together with a large
bandwidth. A salient example is OCT, a non-invasive imaging
technique with many applications in biology and medicine,
in which the axial resolution is inversely proportional
to the bandwidth of the source and is thus enhanced by the
use of a broadband sources. However, irregular spectral
profiles cause sidelobes to appear in the acquisition interferograms,
and thereby ad versely
affect the quality and precision of the measurements. To
solve this problem, we generate broadband light by noncollinear
spontaneous parametric down-conversion with a cw pump laser.
By using a suitable noncollinear phase-matching geometry
and a tightly focused pump beam, down-converted signals
that feature a bell-shaped spectral distribution with a
bandwidth approaching 200 nm are obtained. As an application
of the generated broadband light, submicron axial resolution
in an optical coherence tomography scheme is demonstrated;
a free-space resolution down to 0.8  m
was achieved.
Fig. 1: Experimental arrangement for broadband
light generation.
Back to top
Center for Subsurface
Sensing & Imaging Systems
Bahaa Saleh, Asst. Director; Michael Ruane,
Education Coordinator
http://www.censsis.neu.edu/
Supported by NSF (CenSSIS), ARO (MURI)
The Center for Subsurface Sensing and Imaging
Systems (CenSSIS) is a National Science Foundation (NSF)
Engineering Research Center (ERC), one of an elite group
of only nineteen ERCs in the nation.
Diverse Problems, Similar Solutions –
The CenSSIS mission is to revolutionize the existing technology
for detecting and imaging biomedical and environmental-civil
objects or conditions that are underground, underwater,
or embedded in the human body. The Center's unified, multidisciplinary
approach combines expertise in wave physics, sensor engineering,
image processing, and inverse scattering with rigorous performance
testing to create new sensing system prototypes that are
transitioned to our fourteen industry partners for further
development. A key element of the CenSSIS mission is to
immerse students in efforts to solve important real-world
problems such as noninvasive breast cancer detection or
underground pollution assessment.
CenSSIS overlaps with photonics in several research and
education projects. ECE groups have led CenSSIS research
developments in quantum imaging, photonic biosensing, and
biomedical imaging. We have collaborated on algorithmic
and tool development for mine detection and hyperspectral
imaging.
CenSSIS is creating a series of pathways that
will prepare students to meet the changing expectations
for engineers in the new work force, and connect directly
to the current practice and tools of engineering. Our goal
is to spark systematic change in engineering education.
We are using cross-disciplinary, real-world challenges to
inspire students and infuse them with a systems approach
to solving the com plex
technological and societal problems of the next century.
At ECE, the High Tech Tools & Toys Lab introduces freshmen
and senior project students to hands-on learning about imaging
and sensing systems.
Fig. 1 High Tech Tools & Toys Laboratory
LabVIEW controller development for student project on image-guided
control. This system controls a floating ball in a column
of air, based on real-time video sensing.
References:
1. P. LaPlume, M. Ruane, "Using imaging to introduce
engineering to freshmen", Session 1353, ASEE 2002 Annual
Conference, Montreal, PQ, Canada, June 2002
2. S.S. Chang, M. Ruane, “Detection of Land Mines
using GPR”, SPIE Aerosense Conference, Orlando, FL,
April 23-26, 2003.
Back to top
Entangled Photons
Generated via Optical Parametric Downconversion in Periodically
Poled Lithium Niobate (PPLN)
Hugues Guillet de Chatellus, Bahaa Saleh,
Alexander Sergienko, Malvin Teich
Quantum Imaging Laboratory
Supported by NSF, CenSSIS, ARO (MURI), and the David &
Lucile Packard Foundation.
Spontaneous parametric downconversion (SPDC)
has been widely used as sources of correlated and entangled
photon pairs. The applications of this source include experiments
in optical measurements, quantum imaging, quantum cryptography,
and fundamental studies of quantum mechanics. Among the
different types of entanglement, polarization entanglement
has been proven to be a reliable technique for implementing
non-local communication constructs such as teleportation,
dense coding, and quantum key distribution (QKD). There
has been a great deal of effort over the past decade toward
building compact and efficient sources of polarization entanglement.
Perhaps the best known scheme makes use of a beta-barium
borate (BBO) nonlinear optical crystal and enables polarization
entanglement to be achieved along two particular wave-vector
directions. This source has been extensively used in recent
years. However its efficiency in terms of entangled-pair
production is seriously limited as a result of the poor
overlap of the two orthogonally polarized SPDC cones. We
have demonstrated the possibility of using periodically
poled lithium niobate crystals (PPLN) as a direct and monolithic
source of entangled photons. Polarization entangled pairs
of photons are directly generated in a specially engineered
crystal by overlapping two different type-I nonlinear interactions,
thereby avoiding the need for birefringent compensation.
The output Bell state is tunable from  +
to -
by simply translating the crystal in the field of the pump
beam. We demonstrate efficient polarization entanglement
for the important wavelength pair 810 nm and 1550 nm, which
lies in the optical-fiber telecommunications band.
Fig. 1: Entangled-photon generation using
periodically poled lithium niobate (PPLN).
Reference:
Hugues Guillet de Chatellus, Giovanni Di Giuseppe, Alexander
V. Sergienko, Bahaa E. A. Saleh, and Malvin C. Teich, Engineering
Entangled-Photon States Using Two-Dimensional PPLN Crystals,
Proceedings of SPIE 5456, 75-80 (2004).
Back to top
High Resolution
4Pi Microscopy
Bennett Goldberg, Anna Swan, Selim Ünlü
Optical Characterization and
Nanophotonics Laboratory (OCN)
Confocal fluorescence microscopy has developed
into a standard tool in cell biology research. Light can
easily penetrate inside the cell and a fluorescent dye can
be made to interact with specific cellular components, for
example attach to an antibody that binds to a cellular protein.
The resolution of confocal microscopy is ca 0.5 um laterally
and 0.75 um axially.
The axial resolution of a conventional confocal
microscope can be improved by a factor of 3-5 in 4Pi microscopy.
A 4Pi confocal fluorescence microscope uses two opposing,
high numerical aperture objectives, shown in Fig 1. The
counter propagating wave fronts of the illumination form
interference fringes at the common focal point of the two
objectives. Likewise, the collected light is interfered
at the detector. This effectively reduces the axial focal
volume compared to conventional confocal microscope, illustrated
in Fig. 2. The measured point spread function for a fluorescent
100 nm bead is shown in figure 3. The improvement in axial
resolution using two objectives (b) compared to standard
confocal microscopy (1) can clearly be seen.
We are now combining 4Pi microscopy with
an interferometric method we have developed: spectral self-interference
fluorescent microscopy. The technique transforms the variation
in emission intensity for different path lengths used in
fluorescence interferometry to a variation in the intensity
for different wavelengths in emission, encoding the high-resolution
information in the emission spectrum. Using monolayers of
streptavidin, we have demonstrated better than 5nm axial
height determination for thin layers of fluorophores and
built successful models that accurately fit the data.
Reference:
L. Moiseev, C. R. Cantor, I. Aksun, M. Dogan,
B. B. Goldberg, A. K. Swan, and M. S. Ünlü, "Spectral
self-interference fluorescence microscopy," Journal
of Applied Physics, Vol. 96, 5311, 2004
Back to top
Silicon-based nanostructures for Silicon Photonics
Assistant Professor Luca Dal Negro
The
Holy Grail for silicon photonics has long been the development
of light-emitting devices, i.e. efficient LEDs, and ultimately
of silicon-based lasers [1]. The development of efficient
silicon-based light sources would in fact represent a
technology with enormous societal impact, enabling full
on-chip integration of optical functions within the existing
CMOS technology of microelectronics. Integration of light
emitters and compact lasers with Si microelectronics
will enable diverse technological applications ranging
from intra and inter chip optical interconnects, computing,
bio-sensors, to optical and spectroscopic detection,
and lab-on-a-chip diagnostics. The recent discovery of
optical gain and stimulated emission in silicon
nanocrystals [2] has reenergized the field of photonics
in nanostructured silicon (~350 citations at present),
and began the race towards the demonstration of a fully-silicon-based
laser [1,3].
 Figure
1: First demonstration of optical gain in nanocrystalline
silicon by the Variable Stripe Length technique. The
amplified spontaneous emission grows exponentially as
a result of optical gain when the pumping volume of the
sample is increased. Gain values as high as 80 cm-1 at
750 nm have been obtained under 390-nm ps optical pumping
(from L.Pavesi, L. Dal Negro, C.Mazzoleni, G.Franzò,
F.Priolo, Nature 408, 440, 2000).
 Figure
1: (a) Plan view TEM image and electron
diffraction pattern showing Si nanocrystals embedded
in SiO2 produced by magnetron sputtering deposition
(b) Cross section TEM image showing Si nanocrystals
(the dark spots) embedded in amorphous silicon nitride
by Plasma Enhanced Chemical Vapor Deposition
The
purpose of this research activity is: a) to engineer
novel, nanostructured-based materials solutions for
CMOS-compatible light sources and lasers, b) to characterize
and understand the physics of optical transitions in
Si-based nanostructures via optical spectroscopy and
light amplification techniques. The
full spectrum of the activities involves different
aspects of materials science engineering and characterization,
including thin film deposition, thermal annealing and
nanocrystals nucleation (see Fig. 1), infrared (FTIR)
and visible absorption, Transmission Electron Microscopy
(TEM), electrochemical etching; optical spectroscopy
of nanostructures, light amplification and non-linear
optical techniques along with optical device fabrication
and modelling.
References
- Towards the First Silicon Laser, L. Pavesi,
S. Gaponenko, L. Dal Negro, eds., NATO Science
Series, Kluwer Academic, Dordrecht, 2003
- L.Pavesi, L. Dal Negro, C.Mazzoleni, G.Franzò,
F.Priolo, “Optical gain in silicon nanocrystals”, Nature 408,
440, 2000
- Silicon Photonics, edited by L. Pavesi,
D. J. Lockwood, Springer-Verlag, Berlin (2004).
- L. Dal Negro, J. H. Yi, L.C. Kimerling, S. Hamel,
A. Williamson, G. Galli “Light
emission from silicon rich nitride nanostructures”, Appl.
Phys. Lett., 88, 183103,
2006
Back to top
Laboratory
for Lightwave Technology
Theodore Morse
The Laboratory for Lightwave Technology is
a completely equipped fiber optic laboratory with the capability
of designing, developing, and fabricating specialty optical
fibers. Our emphasis is on fiber lasers and fiber optical
sensors. Described below are a few of our current projects.
1. Q-Switched optical fiber lasers
Material processing applications of lasers dominate in a
total world-wide market of over one billion dollars. Optical
fiber lasers, although a small fraction of this total (100M)
are increasing at the rate of 80%/year. In particular, Q-switched
optical fiber lasers are increasingly supplanting Q-switched
solid state lasers (whether lamp or diode pumped). A Q-switched
laser in the milli-Joule range can mark and process materials
such as steel, glass, and ceramics, and the use of marking
is increasing at a singularly large rate. Q-switched fiber
lasers are more compact, more energy efficient, and, if
costs can be reduced, more desirable than either older flash
lamp pumped or even the more modern diode pumped devices.
An important component of the cost of a Q-switched
fiber laser is either an acousto-optical or electro-optical
switch that must be inserted in the laser cavity. We have
proposed a design in which these costly elements are absent.
In particular, we have demonstrated that it is possible
to Q-switch an optical fiber laser using a MEMS mirror.
This is demonstrated in the following figures.
Fig. 1: 20kHz, average power
230mW Fig. 2: FWHM 230 nm
2. Peak power 310 W
Professor Bifano, an expert in MEMS technology, estimates
that the cost of such a device in large quantities should
be about $50.00, significantly less costly than either an
E-O or A-O modulator.
Since it is not possible to have high power impinge on a
fragile MEMS mirror, it is necessary to use this concept
in a MOPA configuration, in which the output of the oscillator
section of a device would serve as a seed pulse for a fiber
optic amplifier. For example, the 310 W peak power can provide
a much larger input seed pulse than a diode laser, thus
requiring only one stage of amplification to achieve energies
in the mJ range. Such a device is of considerable commercial
interest. These MEMs devices can work up to 100k in air,
and up to several hundred kHz if hermetically sealed in
a vacuum.
In the next figure we illustrate the concept.
The Q-switched oscillator cavity is defined by the MEMs
mirror and a Bragg grating. The pumping of this section
is accomplished by a single mode laser diode at 980 nm and
a WDM coupler. Pulse amplification occurs as shown. The
1x6x1 pump combiner can be obtained from OFS Laboratories
(formerly the fiber part of Bell Laboratories). We are working
on a more efficient pump combiner for high power fiber lasers,
but this pump combiner is now a commercially available product.
Fig. 3: MOPA configuration for q-switched
fiber
Back to top
LCI –
Loss Cone Imager – DSX Mission
Theodore Fritz, Allyn Hubbard, Michael Ruane, Anton Mavreti
Supported by AFRL and USAF DSX Program.
The Air Force will launch the DSX mission
in 2009 and shall resolve critical feasibility issues for
Very Low Frequency (VLF) Wave-Particle Interaction to include
determining VLF antenna injection efficiency from ground-based
transmitters, characterizing the global distribution of
natural and manmade VLF waves in the inner magnetosphere,
and the detection of perturbations of particle populations
due to injected VLF. The Loss Cone Imager (LCI) is one of
several experiments collectively addressing Wave Particle
Interaction, and will provide a measurement of the three
dimensional energetic particle distributions with emphasis
on the measurement of the fluxes of energetic electrons
along the direction parallel and anti-parallel to the local
geomagnetic field vector. The LCI consists of two rotating
scan heads, a high sensitivity telescope, a data processing
unit, and supporting electronics. It is a high ly
sensitive imaging system for energetic particles.
Each scan head has a two-layer sensor in a pin hole camera
configuration. Incident particles pass through the thin
detector, and then strike one of six pixels in the thick
detector, where they are absorbed. Three pin holes on each
head yield 18 10° swaths per head. A field tracking
mode is also available when the spacecraft flies along the
filed lines.
Faculty and students in ECE and the Center for Space Physics,
led by Prof. Ted Fritz, are developing the detectors, rotating
head, FPGA-based DPU, and system level commands and interfaces
to the space craft computers.
Fig. 1 Loss cone image from prior mission,
showing full sphere of intensity of particles.
Light in Deterministically Generated
Aperiodic Optical
Materials
Assistant Professor Luca Dal Negro
The
control of light-matter interactions in complex dielectrics
without translational invariance offers the ultimate
potential for the creation and manipulation of light
states. Unlike periodically arranged dielectrics (photonic
crystals), aperiodic dielectric and metal/dielectric
fractal photonic structures show unique light localization
and transport properties related to an unprecedented
degree of structural complexity. Unlike aperiodic random
media, they can be generated according to simple mathematical
rules or deterministic prescriptions and therefore possess
perfect long-range order. This novel class of materials
is referred to as Deterministically Generated Aperiodic
Optical Materials (DGAOMs), and offers a big advantage
over aperiodic random media in terms of device reproducibility,
fabrication processing and design. For example, in one
dimension DGAOMs can be easily fabricated by stacking
together layers of different dielectric media, A and
B, according to simple mathematical rules, such as the
Fibonacci of the Thue-Morse sequences, encoding
a fascinating complexity (see Fig. 1). DGAOMs show a
non-periodic modulation of their optical constants, and
share distin ctive physical properties with both periodic
dielectric media, i.e. the formation of large energy
gaps, and disordered random media, i.e. the presence
of localized states with strong light-matter coupling
and large electric field enhancement effects. In addition,
DGAOMs exhibit new fascinating physical properties that
originate from the distinctive interplay between the
global lack of translational invariance and the presence
of well defined internal symmetries associated with the
long-range order.
Figure
1: Absolute value of the Fourier coefficients of: (top) a quasiperiodic Fibonacci structure; (center) Thue-Morse structure with singular continuous spectrum; (bottom) deterministic aperiodic structure with absolutely continuous Fourier spectrum (Rudin-Shapiro structure).
The focus of this
research activity is the understanding of light transpor
properties, localization and field enhancement effects
in dielectric and metal/dielectric DGAOMs This
research constitutes a theoretical/experimental joint
study of a new class of optical devices that have the
properties of random materials but can be deterministically
fabricated using the current toolsets of the microelectronics
industry.
The giant electric field enhancement effects and strong
group velocity dispersion (GVD) associated with critical
wavefunctions in DGAOMs lead to strong light-matter interactions
of significant impact for a variety of photonic active
devices, ranging from surface enhanced Raman scattering
(SERS) and single molecule detection to multi-frequency
light emitting devices and optical sensors. Theoretical
predictions and preliminary experimental results indicate
the potential of fabricating multi-wavelength optical
amplifiers and fractal lasers, novel sensors with high
sensitivity over very small probe volumes, non-linear
optical devices such as optical fans and switches.
References
L. Dal Negro,
C.J. Oton, Z. Gaburro, L. Pavesi, P. Johnson, A. Lagendijk,
M. Righini, L. Colocci, D.
Wiersma, Light Transport through the Band-Edge
States of Fibonacci Quasicrystals, Phys.Rev.Lett.,
90,
055501 (2003)
L. Dal Negro, M. Stolfi, Y. Yi, J. Michel,
X. Duan, L.C. Kimerling, J.LeBlanc, J. Haavisto, Photon
band-gap properties and omnidirectional
reflectance in Si/SiO2 Thue-Morse quasicrystals, Appl.
Phys. Lett., 84, 5186 (2004)
L. Dal Negro,
J. H. Yi, V. Nguyen, Y. Yi, J. Michel and L.C. Kimerling, Spectrally enhanced
light emission from aperiodic photonic structures, Appl.
Phys. Lett., 86, 261905 (2005).
Microelectromechanical
Flexure PZT Actuated Optical Scanner: Static and Resonance
Behavior
Vladimir F. Kleptsyn, Johannes G Smits1,
and Koji Fujimoto2
An optical mirror system has been built
using MEMS techniques while employing lead zirconate titanate
(PZT) thin films as strong piezoelectric material. The
system consists of two parallel silicon cantilever beams,
on which the PZT was deposited, which were arranged to
project inwards from a common frame. The beams were laterally
offset over a distance appropriate to fit a mirror between
them. One end of each beam is fixed to the frame, while
the other end was attached to a sideways projected bar,
directed to the other beam. To each sideways bar a connector
was attached, parallel with the original beam and pointing
towards the frame. Between both connectors a mirror was
attached. The application of voltage across the electrodes
on top and below the PZT causes the beams to bend as a
monomorph, and the mirror between the beams to rotate
around an axis through its center.
A variety of thicknesses have been chosen
to fabricate these devices. The mirror angle at a given
voltage is inversely proportional to the square of the
beam thickness, while the bandwidth is proportional to
the thickness.  Beams
with thickness of 5 microns were used to achieve optical
angles in static mode of up to 40 degrees using voltages
up to 13 V. In that case a bandwidth of around 700 Hz
was observed. In resonance optical angles of 10 degrees
with a driving voltage as low as 100 mV were achieved.
For devices with 30 micron thick beams, optical angles
of up to 30 degrees were observed in resonance at 17.4
kHz at driving voltages of around 6 V. Due to the proprietary
design a figure of merit, which may be defined as a product
of the resonance frequency and optical angle, is much
higher than one of the other PZT scanners.
Fig. 1 The 50x front view of the chip. Alignment
marks are top right. Alignment accuracy was 2-3 microns
Reference:
- J. Micromech. Microeng. 15 (2005) 1285–1293
1Present address: Scaldix B.V. Stationsstraat 13 4331
JA Middelburg, The Netherlands
2Present address: DNP Co., Ltd. Electronics System Laboratory
R and D Center, Wakashiba 250-1, Kashiwa, Chiba 2770871
Back to top
A Multi-Watt 980nm
Single Mode Fiber Laser
Theodore Morse
The most efficient pump source for an erbium
fiber laser is a 980 nm single mode diode pump that is
commercially available at a maximum .5 W. Pump powers
in excess of 2.0 W are needed for numerous Fiber to the
Home designs, and combining several .5 W diodes to achieve
the desired pump levels is not an ideal solution. The
possibility of a double-clad higher power pump is attractive,
and Ytterbium seems a natural for such a pump, with its
absorption/emission spectrum shown at left. This is particularly
true since 915 nm and 980 nm multimode fiber coupled pump
sources are commonly available. With Yb pumped by either
915 nm or 980 nm multimode diodes in a double-clad configuration,
it would seem that there would be a natural laser at the
sharp 980 nm. However, Yb is a three level lasing system
at 980 nm, and a quasi-four level system in the 1.07 micron
regime. Thus, the four-level system will dominate the
three level 980 system. If the laser is short enough,
in a double clad configuration, Southampton has demonstrated
that 2.0 W can be achieved. Working with Prof. R. Quimby
of Worce ster
Polytechnic Institute using a special doping to absorb
stimulated emission photons at wavelengths longer than
1.l0 microns, we have shown, through modeling, that if
the stimulated emission in the four level system is quenched
through an absorber in the glass fiber, then it should
be possible to obtain the order of 3 W from a longer length
of fiber. Work is in progress on experimental verification,
and we believe that we can fabricate a new type of multi-Watt
single mode pump source at 980 nm.
Back to top
III-Nitride Intersubband
Optoelectronic Devices
Roberto Paiella, Theodore D. Moustakas,
Enrico Bellotti
Intersubband transitions – i.e. optical
transitions between quantized conduction-band states in
semiconductor low-dimensional systems – offer several
attractive features for device applications, including
remarkable design flexibility through bandgap engineering,
ultrafast carrier lifetimes, and giant optical nonlinearities
[1]. These transitions form the basis of well-established
devices, such as the quantum-well infrared photodetector
(QWIP) and the quantum cascade (QC) laser, which so far
have been primarily demonstrated with arsenic-based semiconductors.
We are currently pursuing the development of similar devices
using GaN/AlGaN heterostructures [2], whose large conduction-band
offsets result in improved electron confinement in the
quantum wells and allow extending the intersubband transition
wavelength to the near infrared. Specific device applications
being investigated include: QC emitters and lasers for
the 3-5 m
atmospheric window; novel QWIP structures [3]; and ultrafast
all-optical switching devices at fiber-optic communication
wavelengths based on a variety of nonlinear optical interactions
(e.g. absorption saturation, optical Kerr effect, and
intersubband Raman scattering).
Fig.
1 Calculated conduction-band diagram (a) and measured
intersubband absorption spectrum (b) of a GaN/AlN quantum
well.
References
[1] R. Paiella, ed. Optoelectronic Devices Based on Intersubband
Transitions in Semiconductor Quantum Structures, McGraw-Hill,
to be published in March 2006.
[2] I. Friel, K. Driscoll, E. Kulenica, M. Dutta, R. Paiella,
and T.D. Moustakas, “Investigation of the design
parameters of AlN/GaN multiple quantum wells grown by
molecular beam epitaxy for intersubband absorption,”
J. Crystal Growth. 278, 387-392 (2005).
[3] S. Gunna, E. Bellotti, and R. Paiella, “Novel
single photon detector design based on intersubband transitions
at 1.55 m,”
presented at the 8th International Conference on Intersubband
Transitions in Quantum Wells (Cape Cod, September 2005).
Back to top
A Novel High
Temperature Optical "Thermocouple"
Theodore Morse
Lightwave Technology Laboratory
In the following, we describe research that
has led to an optical thermocouple capable of measuring
temperatures from -200C to 1,200C with an accuracy of
 1C.
Work is in progress to extend this to temperatures approaching
1,800C.
It is well known that a series of 1/4 wave
layers will reflect light over a narrow bandwidth, depending
on the number of layers and the refractive index differences
of the layers. Using a CVD reactor with ammonia/argon
and silane/argon to deposit silicon nitride and silicon-rich
silicon nitride on the end of an optical fiber, we are
able to build up a series of layers with a narrow band
reflectivity. This same technique can use the optical
fiber end as a "witness" sample in MOCVD to
measure deposition growth to within 10 Angstrom. If these
layers, which are only a few microns in height, are heated,
there is a change in the height of the stack due to thermal
expansion, and a variation in the refractive index with
temperature. These changes produce a shift in the reflection
characteristics of the stack, which can be readily monitored
with a compact solid state spectrometer. The results are
shown in the figure above. Work is in progress to employ
a sapphire single crystal fiber fused to a silica fiber,
or a thin disc of sapphire with a holographic tantalum
oxide layer to obtain high temperature measurements. The
final result would be an optical "thermocouple"
with a temperature range from liquid nitrogen temperatures
to temperatures in excess of 1,800 C. 
Back to top
Numerical Aperture
Increasing Lens Microscopy
Bennett Goldberg, Anna Swan, Selim Ünlü
Optical Characterization
and Nanophotonics Laboratory (OCN)
Numerical Aperture Increasing Lens (NAIL)
microscopy is a far-field subsurface imaging technique
that simultaneously enhances the light gathering power
and resolution of an optical microscope. When a NAIL is
placed on the backside of a sample, its convex surface
effectively transforms the NAIL and the planar sample
into an integrated solid immersion lens, capable of aberration-free
imaging of the structures underneath the substrate. Addition
of the NAIL to a standard microscope increases the numerical
aperture (NA) by a factor of the square of the optical
index n. The NAIL technology has had the greatest impact
in the field of optical failure analysis of Si integrated
circuits. In silicon, the NA is increased by a factor
of 13. Using an optimized confocal microscope, we have
already demonstrated a lateral resolution of 230 nm. Recently,
we have applied the technique to optical spectroscopy
of single quantum dots demonstrating an 8-fold improvement
in light collection from a single dot.
Fig. 1 The sample-NAIL assembly (left) under
an objective lens (right) 2-D spectral image of a single
quantum dot displaying 300 nm spatial resolution at high
collection efficiancy.
References
1. S.B. Ippolito, B. B. Goldberg, and M. S. Ünlü,
“High spatial resolution subsurface microscopy,”
Applied Physics Letters, 78, 4071 (2001).
2. Z. Liu, B. B. Goldberg, S. B. Ippolito, A.N. Vamivakas,
M.S. Ünlü and R.P. Mirin, “High resolution
and high collection efficiency in numerical aperture increasing
lens microspcopy of individual quantum dots,” Applied
Physics Letters, 87, 071905, (2005).
3. B. B. Goldberg, S. B. Ippolito, L. Novotny, Z. Liu,
and M. S. Ünlü, "Immersion Lens Microscopy
of Photonic Nanostructures and Quantum Dots," IEEE
Journal of Selected Topics in Quantum Electronics, Vol.
8, pp. 1051 -1059, (2002)
Back to top
Optical Pharmacokinetics:
noninvasive measurement of drug concentrations in tissue
Irving J Bigio
Biophotonics Laboratory
Supported by NIH, NSF (CenSSIS)
The method of optical pharmacokinetics (OP)
has been developed for noninvasive, localized, sub-surface
measurement of drug concentrations in tissue. The OP method
employs optical spectroscopy mediated by fiber-optic probes
and has general applicability for measuring concentrations
of chromophores in turbid media. Measurement is mediated
by optical fibers, and the details of the optical geometry
of the probe permit measurement of absorption coefficients
in a manner that is insensitive to variations in the scattering
coefficient. This technology can be valuable in the drug
discovery and pre-clinical testing in animals. A new extrapolation
of this technique may aid in the assessment of angiogenesis
(the stimulation of new vessel growth near tumors) and
the response of tumors to new anti-angiogenic chemotherapy
agents.
Fig. 1 Schematic of the OP system
References:
1. “Noninvasive measurement of chemotherapy drug
concentrations in tissue: preliminary demonstrations of
in vivo measurements,” Judith R. Mourant, Tamara
M. Johnson, Gerrit Los and Irving J. Bigio, J. Physics
in Medicine and Biology 44, pp. 1397-1417 (1999)
2. “Measuring absorbance in small volumes of highly-scattering
media: source-detector separations for which pathlengths
do not depend on scattering properties,” J.R. Mourant,
I.J. Bigio, D. Jack, T.M. Johnson and H.D. Miller, Applied
Optics 36, pp. 5655-5661(1997).
Back to top
Optical Properties
of Carbon Nanotubes
Bennett Goldberg, Anna Swan, Selim Ünlü
Optical Characterization
and Nanophotonics Laboratory (OCN)
Since the accidental discovery of a new
form of carbon in 1991, carbon nanotubes have attracted
wide-spread interest due to their remarkable properties.
It is the strongest, stiffest and toughest material known
as well as the best possible conductor of heat and electricity.
A nanotube can be thought of as a rolled up sheet of graphite,
with typical diameters around 1-2 nanometers for single
wall carbon nanotubes, and a length-to-diameter aspect
ratio of up to 108. The nanotubes can be either metallic
or semiconducting, depending on the chirality and diameter
of the nanotube, characterized by the roll-up vectors
(n,m). The variable and direct bandgap for the semiconducting
tubes makes for potentially powerful photonics applications.
In our lab we specialize in optical measurements
of individual carbon nanotubes so that we are able to
measure properties that are not possible to extract in
ensemble measurements. For example, a combination of applied
strain and strain measurements using resonant micro Raman
showed that previous strain measurements of CNTs in composites
overestimated the strain applied to the nanotubes by a
factor of 4, indicating problems with adhesion between
the composite and the nanotubes.
Due to the strong optical resonances typical
for a one dimensional material, we are also able to use
resonant Raman to map the optical resonance energies and
correlate the energies to the specific tube diameter and
chirality and gain understanding of the strong environmental
influence on the excitonic binding energies. See www.bu.edu/OCN
for further information.
Fig. 1 Stokes and anti-Stokes resonant Raman
measurement of a phonon mode mapping the optical resonance
for a single (9,4) nanotube.
References:
1.Y. Yin, S. Cronin, A. Walsh,
A. Stolyarev, M. Tinkam, W. Bacsa, M.S. Unlu, B.B. Goldberg,
A.K. Swan. “Intrinsic optical properties of carbon
nanotubes,” Nanotube 05 Gothenburg, Sweden, June
2005
2.S. B. Cronin, A. K. Swan, M. S. Ünlü, B. B.
Goldberg, M. S. Dresselhaus, and M. Tinkham, "Measuring
the Uniaxial Strain of Individual Single-Wall Carbon Nanotubes:
Resonance Raman Spectra of Atomic-Force-Microscope Modified
Single-Wall Nanotubes," Phys.Rev Lett.93 167401 (2004)
Back to top
Quantum
Optical Coherence Tomography
Bahaa Saleh, Alexander Sergienko, Malvin
Teich
Quantum Imaging Laboratory
Supported by NSF (CenSSIS), ARO (MURI)
Quantum Optical Coherence Tomography (QOCT)
is a novel technique for 3D imaging based on quantum interference
of entangled-photon beams. The interferometer uses a beam
splitter in the shown configuration. One beam travels
through the sample and the other through a controllable
delay 
before reflection through the interferometer. The rate
of coincidence of photons at the output ports of the beam
splitter is measured as a function of
by use of two photon-counting detectors and a coincidence
counter. Because of quantum destructive interference,
when the optical path lengths are equal, the coincidence
rate exhibits a sharp dip of width equal to that of the
photon wave packet, which is as small as 10 fs. This is
used to monitor the range (depth) with good resolution.
An image of the backscattered light is obtained by scanning
in the transverse direction. We have demonstrated theoretically
and experimentally that QOCT has a resolution greater
by a factor of 2 than conventional OCT for the same source
bandwidth, and that it is insensiti ve
to dispersion of the medium. This permits deeper subsurface
sensing in dispersive media. This remarkable property
is inherent in this type of quantum interferometry.
Fig. 1 Quantum OCT using light generated
from a nonlinear crystal by a process of spontaneous parametric
downconversion.
References
1. M. B. Nasr, B. E. A. Saleh, A. V. Sergienko, and M.
C. Teich “Dispersion-Cancelled and Dispersion-Sensitive
Quantum Optical Coherence Tomography,” Optics Express,
Vol. 12, pp. 1353-1362 (2004).
2. S. Carrasco, J. P. Torres, L. Torner, A. Sergienko,
B. E. A. Saleh, and M. C. Teich “Enhancing the axial
resolution of quantum optical coherence tomography by
aeriodic quasi-phase-matching”, Optics Letters,
29,2429 (2004).
Back to top
Resonant Cavity Imaging
Biosensor
Bennett Goldberg, Mike Ruane, Anna Swan,
Selim Ünlü
Optical Characterization
and Nanophotonics Laboratory (OCN)
The Resonant Cavity Imaging Biosensor (RCIB)
detects binding between target biomolecules from a sample
and probe biomolecules fixed to a microarray surface with
the potential for tens of thousands of simultaneous parallel
observation sites. Such ability yields information about
the affinity of the biomolecules under test for the molecules
on the capturing surface. Information about the affinity
between molecules of interest such as particular proteins
or DNA strands, yields great benefit to a number of applications
in biological research, medical diagnostics, and biohazard
detection.
Current high-throughput microarray technology
requires that the target molecules be labeled with a fluorescent
dye. At best, this preparation step adds an acceptably
small amount of time and money, but at its worse, can
be prohibitively difficult depending on the nature of
the application. RCIB operates label-free without the
need to add fluorescent labels or otherwise modify the
target molecules in any way. An optical IR beam couples
resonantly through a cavity constructed from Bragg mirrors
that contains the microarray surface; the wavelength of
the IR beam is swept using a tunable IR laser source;
and an IR camera monitors cavity transmittance at each
pixel, creating a highly parallel signature of transmittance
versus wavelength for the microarray surface. 
This novel technique is enabled by high
quality silicon substrates with buried Bragg reflectors
previously developed within our group for improved photodetectors.
The technique additionally relies on the use of commercial
telecommunications hardware that has become readily available
in recent years. In an alternative approach for microarray
detection, the reflection from the substrate is measured
with the varying wavelength. When binding occurs on the
surface of the wafer, the reflectivity vs. wavelength
curve shifts, from which the height information can be
extracted. This alternative approach is less sensitive
than RCIB, but it draws attention with its simplicity.
RCIB improves on existing label-free methods by offering
dramatically improved throughput necessary to meet the
needs of the microarray user community.
References:
D. A. Bergstein, R. J. Irani, M. F. Ruane, C. DeLisi,
M. S. Ünlü. “Resonant Cavity Imaging Biosensor,”
IEEE/LEOS 18th Annual Meeting for the Lasers and Electo-Optics
Society. Sydney Australia Oct. 23-27 2005
Back to top
A "Whispering
Gallery" Intra-cavity Biosensor
Theodore Morse
It is well known that if measurements in
the optical domain can be transferred into the frequency
domain, enormous increases in sensitivity occur. This
is particularly true for the measurement of relative changes
in wavelength that correspond to relative changes in frequency.
The relation between a change in frequency and a change
in frequency is given as:  ,
where the factor 
is of the order of 
in the infrared. Thus, a change in wavelength of 
picometer corresponds to a change in beat frequency of
10 kHz, which can be readily measured in a RFSA (Radio
Frequency Spectrum Analyzer).
A "stable" laser, even with a FWHM linewidth
in the kHz range, will still have a frequency instability
of the order of a MHZ. Thus, in order to obtain a reference
wavelength to be able to measure frequency changes in
the kHz range, it is necessary to obtain a reference signal
from within the laser cavity. We have employed this technique
to measure very small changes in laser wavelength by using
a laser with a slight anisotropy in the refractive indices
in orthogonal directions. The differences in wavelength
of these two components were not optically measurable;
however, when the laser output was projected on a 45-degree
polarizer and then into a RFSA, the beat signatures were
clearly discernable.
This is an ultrasensitive technique for the measurement
of wavelength changes, and this forms the basis for a
biosensor we are in the process of developing. The accompanying
figure clearly shows the PMB (Polarization Mode Beating)
technique. This sensitivity is the basis for our "whispering
gallery” biosensor that is under active investigation
in our laboratory. A whispering gallery biosensor functions
in the following manner: If a silica sphere (or planar
device) is used, then a Si-O-antigen can be functionalized
onto the surface such that only a specific antibody will
attach to the surface, thus changing the refractive index.
If laser light is coupled into the sphere, then there
will be a dip for that condition for which there is a
resonance. When the antibody attaches, there will be a
change in the refractive index and an associated change
in the resonance wavelength. Such a shift is shown in
the figure above.
It is clear that the use of a micro silica sphere with
a tapered fiber is not a suitable vehicle for any practical
biosensor. We have carried out experiments using a Little
Optics planar waveguide device that is, essentially, a
telecommunications add/drop filter. This is shown in the
figure below. The ring is 50 microns in diameter, and
we have inserted this within a laser cavity and have obtained
lasing.
Rings that are specific to our needs and design are being
fabricated at Cal Tech by Dr. Koby Scheuer, and we hope
to be able to incorporate these into a micro flow cell
in the near future. We believe this technique has the
possibility of being used in large parallel arrays, conveniently
and inexpensively, and that it will provide an unprecedented
degree of biosensitivity.
Back to top
 |
