This action might not be possible to undo. Are you sure you want to continue?
Photodetectors and DoubleDisk Cavity Optomechanics
Thesis by
Jessie Rosenberg
In Partial Fulﬁllment of the Requirements
for the Degree of
Doctor of Philosophy
California Institute of Technology
Pasadena, California
2010
(Defended December 8, 2009)
ii
c 2010
Jessie Rosenberg
All Rights Reserved
iii
To my mother, father, and all of the friends, near and far, who have supported me.
iv
Acknowledgments
First and foremost, I want to thank my advisor, Professor Oskar Painter. Your scientiﬁc brilliance,
creativity, and boundless energy will forever serve as an inspiration.
Thanks, also, to Qiang Lin. You have always been so generous with your knowledge, and never
balked at answering any question, seemingly trivial or otherwise. Our collaboration over these past
months has been an experience of immeasurable value for me.
Thanks to Raviv Perahia, for your companionship through ﬁve years and three different ofﬁces;
for all the conversations, and questions, and answers. It’s been a delight. Thanks, also, to my other
ofﬁcemates, past and present: Orion Crisafulli, Qiang Lin, Thiago Alegre.
Thanks to the senior members of the group, Kartik Srinivasan, Paul Barclay, Matt Borselli, and
Tom Johnson, for the truly phenomenal amount of work you put in to get the group started, and for
sharing that knowledge and experience with those of us who followed. Thanks to Darrick Chang,
for so many fruitful discussions.
Thanks to the recently graduated and current members of the group, Raviv Perahia, Chris
Michael, Matt Eichenﬁeld, Thiago Alegre, Ryan Camacho, Jasper Chan, Amir SafaviNaeini, Jeff
Hill, Alex Krause, Daniel Chao, Chaitanya Rastogi, and Justin Cohen, for making the group such
a dynamic and vibrant scientiﬁc environment. Thanks to the old guard for sharing knowledge and
discoveries and stories along the way, and thanks to the new students for carrying everything on into
the future  and hopefully doing it all even better than we did.
Thanks to Professor Sanjay Krishna’s group at the University of New Mexico, for such a valu
able collaboration, and for being so welcoming to a visitor. In particular, thanks to Rajeev Shenoi,
for the work we did together, and for being a great host.
Thanks to all the excellent staff at Caltech, who always were willing to contribute their time to
help out.
Thanks to all of my friends, Caltech students or otherwise, nearby or far away. I can’t even
begin to name everyone who had such an impact, but know that you are valued. In particular, a
v
heartfelt thanks to Eve Stenson, Neil Halelamien, Megan Nix, Jen Soto, and Jay Daigle. Thanks to
every person on the organizational team of the Caltech Ballroom Dance Club.
Thanks, above all, to my mother and father, for being the best and most supportive parents I
could have asked for.
Thanks, everyone.
vi
Abstract
Optical resonators present the potential to serve vital purposes in many emergent technologies that
require spectral ﬁltering, high optical intensities, or optical delays. By scaling down the optical res
onators to the micro or nanoscale, the relevant phenomena can increase signiﬁcantly in magnitude,
while the device geometries become suitable for chipscale and integrated processing. In this the
sis, research is presented on several valuable resonator geometries and implementations, beginning
with a more standard alloptical design, and continuing on to investigate the novel phenomena and
applications which are made possible when optical and mechanical structures can be synergistically
combined.
First, the design and experimental implementation of a plasmonic photonic crystal spectral and
polarization ﬁltering element is presented. This resonator scheme, in addition to allowing for a tai
lorable frequency and polarization response for single detector pixels, also increases the absorption
of a thin layer of detector material by utilizing the unique optical properties of metal to conﬁne light
more tightly within the detector active region. Demonstrated in the valuable midinfrared regime,
this method of producing pixelintegrated multispectral detectors could ﬁnd application in biologi
cal sensing and spectroscopy, missile tracking and guidance, and night vision.
Following this discussion, progress is presented in the relatively new ﬁeld of cavity optome
chanics: utilizing mechanically compliant optical resonators to couple to, control, and read out me
chanical motion via optical forces. The use of optical resonators allows the generally weak optical
forces to be increased in strength by orders of magnitude due to the many passes light makes within
the resonator, while miniaturizing optomechanical devices into a convenient form factor for onchip
applications. Using a fully siliconcompatible doublediskgeometry optomechanical resonator, ex
tremely large optomechanical coupling and very high optical quality factors are shown, enabling the
demonstration of regenerative mechanical ampliﬁcation, high compression factor optomechanical
cooling, coherent mechanical mode mixing, and widebandwidth alloptical wavelength routing.
Applications to groundstate cooling of mesoscopic devices, tunable optical buffering, photonic
vii
phononic quantum state transfer, channel routing/switching, pulse trapping/release, and tunable las
ing are discussed.
viii
Contents
Acknowledgments iv
Abstract vi
Preface xiii
1 Plasmonic Resonators for Multispectral MidInfrared Detectors 1
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Photonic Crystal Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 SingleMetal Plasmon Resonator Design . . . . . . . . . . . . . . . . . . . . . . . 8
1.4 SingleMetal Experimental Demonstration . . . . . . . . . . . . . . . . . . . . . . 13
1.5 DoubleMetal Plasmon Resonator Design . . . . . . . . . . . . . . . . . . . . . . 15
1.6 Critical Coupling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2 Double Disk Optomechanical Resonators 24
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2 Optomechanical coupling and dynamic backaction . . . . . . . . . . . . . . . . . 25
2.3 Doubledisk fabrication, optical, and mechanical design . . . . . . . . . . . . . . . 26
2.4 Optical and mechanical characterization . . . . . . . . . . . . . . . . . . . . . . . 32
2.5 Regenerative oscillation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.5.1 Ambient pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.5.2 Vacuum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
2.6 Optomechanical cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.7 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3 Coherent Mechanical Mode Mixing in Optomechanical Nanocavities 52
ix
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.2 Zipper cavity and doubledisk design, fabrication, and optical characterization . . . 53
3.3 Theory of optomechanical effects in the presence of mode mixing . . . . . . . . . 55
3.3.1 Intracavity ﬁeld in the presence of optomechanical coupling . . . . . . . . 56
3.3.2 The power spectral density of the cavity transmission . . . . . . . . . . . . 57
3.3.3 The mechanical response with multiple excitation pathways . . . . . . . . 58
3.3.4 The mechanical response with external optical excitation . . . . . . . . . . 60
3.4 Mechanical mode renormalization in zipper cavities . . . . . . . . . . . . . . . . . 61
3.5 Coherent mechanical mode mixing in doubledisks . . . . . . . . . . . . . . . . . 65
3.6 Coherent mechanical mode mixing in zipper cavities . . . . . . . . . . . . . . . . 68
3.7 Analogy to electromagneticallyinduced transparency . . . . . . . . . . . . . . . . 69
3.8 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4 Mechanically Pliant Double Disk Resonators 75
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.2 Spiderweb resonator design and optical characterization . . . . . . . . . . . . . . . 76
4.3 Static ﬁlter response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
4.4 Dynamic ﬁlter response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
4.5 Discusssion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
5 Conclusion 96
x
List of Figures
1.1 Simulated bandstructure for a doublemetal photonic crystal . . . . . . . . . . . . . 4
1.2 Simulated bandstructure for a square and rectangular lattice plasmonic photonic crystal 5
1.3 Schematic and bandstructure for singlemetal detectors . . . . . . . . . . . . . . . . 7
1.4 Plasmonic photonic crystal detector simulations . . . . . . . . . . . . . . . . . . . . 9
1.5 Singlemetal detector enhancement factor and active region absorption . . . . . . . . 11
1.6 Detector devices and measurement setup . . . . . . . . . . . . . . . . . . . . . . . 12
1.7 DWELL detector measurement results . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.8 Waveguide thickness dispersion for doublemetal waveguides . . . . . . . . . . . . 15
1.9 Field proﬁles for doublemetal waveguides . . . . . . . . . . . . . . . . . . . . . . 16
1.10 Comparison of FDTD and group theory . . . . . . . . . . . . . . . . . . . . . . . . 17
1.11 Farﬁeld proﬁles for stretchedlattice double metal resonators . . . . . . . . . . . . . 18
1.12 Schematic of doublemetal detector loss mechanisms . . . . . . . . . . . . . . . . . 19
1.13 Vertical and substrate coupling vs. metal thickness and hole size for the singlemetal
structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.14 Vertical coupling vs. metal thickness and hole size for the doublemetal structure . . 22
1.15 Doublemetal focal plane array schematic . . . . . . . . . . . . . . . . . . . . . . . 23
2.1 Doubleslab waveguide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.2 Cavity Optomechanical Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.3 Double disk ﬂapping mode displacement . . . . . . . . . . . . . . . . . . . . . . . 31
2.4 Fabrication and characterization of doubledisk NOMS . . . . . . . . . . . . . . . . 33
2.5 Transmission scans for a doubledisk microcavity . . . . . . . . . . . . . . . . . . . 34
2.6 Optical and mechanical mode spectroscopy . . . . . . . . . . . . . . . . . . . . . . 42
2.7 Regenerative oscillation in a doubledisk microcavity . . . . . . . . . . . . . . . . . 44
2.8 Dynamical backaction: damping and ampliﬁcation of mechanical motion . . . . . . 45
xi
3.1 Schematics and optical modes of two optomechanical systems . . . . . . . . . . . . 54
3.2 Mode mixing in zipper cavities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
3.3 Mode mixing measurements in double disks . . . . . . . . . . . . . . . . . . . . . . 67
3.4 Zipper cavity mechanical mode mixing . . . . . . . . . . . . . . . . . . . . . . . . 70
3.5 Mechanical mode mixing analogues to optical systems and to EIT . . . . . . . . . . 71
4.1 Spiderweb microresonator images and simulations . . . . . . . . . . . . . . . . . . 77
4.2 Pumpprobe experimental setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
4.3 Static tuning of a spiderweb microresonator . . . . . . . . . . . . . . . . . . . . . . 81
4.4 Thermomechanical deﬂection of a spiderweb resonator . . . . . . . . . . . . . . . . 83
4.5 Dynamic response of a spiderweb microresonator . . . . . . . . . . . . . . . . . . . 85
xii
List of Tables
1.1 Point Group character tables for the square and rectangular lattice. . . . . . . . . . . 5
xiii
Preface
The startup of a research lab, especially one with such an extensive range of facilities, is a truly
overwhelming proposition requiring a daunting amount of effort. I was fortunate enough to join the
group at a rather fortuitous time, when the group was mature enough to have essentially all fully
functioning labs, but all of the original group members, with their tremendous experience born from
setting up everything in the lab, were still around and generously willing to share their knowledge.
I immediately began work on a project started by Kartik Srinivasan and Raviv Perahia, that I
imagined would occupy my entire graduate career: the patterning of photonic crystals in metal ﬁlms,
to create spectrally sensitive pixels in midinfrared detector material grown by our collaborators,
Prof. Sanjay Krishna’s group at the University of New Mexico. Though undoubtedly fraught with
unforeseeable complexities, the path to working devices seemed straightforward, and dovetailed
well with the many other photonic crystal efforts then under investigation in the lab.
My ﬁrst major task, occupying nearly the entirety of my ﬁrst summer in the lab, was to become
a fully functional user of our group cleanroom. We patterned some initial devices, and I also got
started learning about theory and doing some modeling in Matlab, as well as becoming familiar with
our group’s homebuilt ﬁnite difference time domain code. With some initial dubious successes on
the processing front, I also spent time working with Orion Crisafulli in doing a thorough investiga
tion of the properties of metalinsulatormetal waveguides, which became quite useful later.
With a full plate of classwork interspersed, this occupied the ﬁrst few years of my graduate
career, splitting my time between experiment and theory. We made several sets of photonic crystal
devices, but never saw clear resonant spectral enhancement correlating with the photonic crystal
patterns, which bafﬂed us for quite some time. In the end, we ﬁnally realized that, due to the
particulars of the device fabrication, the apertures that we were patterning photonic crystals on
were, in fact, only a small fraction of the open area of the devices: there was a large outer area
of exposed detector material that had no photonic crystal patterning or covering of metal, and this
background scattering was completely overwhelming our signal.
xiv
After that discovery, and another round of device processing followed by my ﬁrst visit to the
testing facilities at the University of New Mexico, we ﬁnally had our ﬁrst clear success: spectrally
sensitive detector pixels with a resonance frequency having a direct correspondence to the designed
photonic crystal patterns. We celebrated, and I ﬂew home to continue doing modeling. One more
processing run followed, on a new optimized detector material grown by the Krishna group to have
more quantum dot layers in a stable conﬁguration, and we ﬁnally had data that we were satisﬁed
with.
Unfortunately, the simulations proved to be somewhat more difﬁcult. The detector devices,
containing both metals and thick waveguide layers, required a very ﬁne simulation mesh, proving a
strain on computational power. While I attempted to ﬁnd methods to extract the numbers we wanted,
we also encountered processing difﬁculties in attempting to fabricate the doublemetal detector
devices that we predicted would give us better results. In the end, although further progress in this
area seemed achievable with a fair amount of additional effort, we decided to focus on ﬁnishing up
what we already had accomplished.
As that project ended, then, I was freed up to begin something else. At the time, Qiang Lin
was intensely busy with the ﬁrst demonstrations of the doubledisk optomechanical resonators, so
the obvious thing for me to do was to join him to provide some additional manpower. With the
process optimization I had done working on the photonic crystal detectors, I was wellsituated to
help optimize the doubledisk processing, and I started to learn about cavity optomechanics at the
same time. The project started out as a crash course in ﬁberoptic testing for me, since the detector
testing I had done had all been freespacebased, but in the end we came out with some very nice
results and a world of possibilities to investigate.
The next task we tackled was modifying the doubledisk geometry into a very ﬂexible structure
that could achieve large static displacements with only a small applied optical force. Developing
the “spiderweb” geometry devices – doublering structures with inner spokes, and rings to stabilize
the mechanics – went surprisingly quickly, with few fabrication setbacks, and we were able to begin
testing those devices with a minimum of delay. In the end, we had demonstrated a new type of
alloptical tunable resonator, with applications to a variety of ﬁelds in optical communications, as
well as nonlinear and quantum optics.
Emboldened by our successes in both of those endeavors, we returned to complete the work on
coherent mechanical mode mixing Qiang had been working on in the original doubledisk geometry.
With the help of fruitful discussions with Darrick Chang, we completed the theoretical underpin
xv
nings of the work, and expanded the original idea to encompass the potential for slowlight effects
relying on the long phononic timescale rather than the relatively quite short photonic one, as well as
the eventual possibility for quantum state transfer between phonons and photons.
Looking into the future, there is still a lot of work to be done: cavity optomechanics remains
a wideopen ﬁeld. I see the possibilities presented by this and all the other innovative work being
done in the ﬁeld, and I can only be excited for what will undoubtedly come next.
This thesis begins by presenting the work on plasmonic photonic crystal midinfrared photode
tectors, beginning with the theory and an experimental demonstration of singlemetal waveguide
devices, and continuing to expand the theory to discuss doublemetal devices. Continuing on, work
on dynamic backaction in doubledisk resonators is presented, followed by the demonstration of
coherent mechanical mode mixing in the same device structure. Finally, the work on the ﬂexible
doublering “spiderweb” optomechanical resonators is discussed, ﬁnishing with several appendices
on the experimental and mathematical details of the optomechanical work presented in this thesis.
1
Chapter 1
Plasmonic Resonators for Multispectral
MidInfrared Detectors
1.1 Introduction
Optical sensors in the midinfrared wavelength range are extremely important in a wide variety of
areas, such as night vision, missile guidance, and biological spectroscopy [1]. Currently, the best
midinfrared detectors are based on mercurycadmiumtelluride (MCT). MCT detectors are very
efﬁcient, but largearea focalplane arrays are difﬁcult and expensive to grow due to difﬁculties with
the epitaxial growth of mercurybased compounds [2, 3]. More recently, other detector materials
have become more common, but they have various limitations in the required direction of incoming
light, such as in quantum well detectors, or in detector efﬁciency, as in quantum dot or dotsina
well (DWELL) detectors [4, 5]. With the use of a resonant cavity, it becomes possible to increase
the detector efﬁciency many times over by greatly extending the interaction length between the
incoming light and the active material. Instead of incoming light making only one pass through the
active region, in a resonant detector the light can make hundreds of passes.
The presence of a resonator can also make each pixel frequency and polarization speciﬁc [6–
8], allowing for a hyperspectral and hyperpolarization sensor without the need for any external
grating or prism. There are many applications for frequency and polarizationsensitive detectors. A
hyperspectral detector array could function as a spectrometer on a chip, ﬁltering incoming signals
through the use of hundreds of highly sensitive detector pixels. A hyperpolarization detector could
be used in a camera to provide an additional layer of information which can be combined with
frequency and intensity data to better distinguish between different objects in an image.
The current dominant technologies in the ﬁeld of multispectral imaging rely on the use of either
2
a broadband focal plane array (FPA) with a spinning ﬁlter wheel in front of it [9], or a bank of FPAs
with a dispersive element such as a grating or prism to separate light of different frequencies. These
methods are limited by the often high cost and complexity of such systems. However, when spectral
sensitivity is encoded at the pixel level within a single focal plane array, multispectral detection
becomes much more practical for use in a wide range of applications. In addition, the use of pixel
integrated resonators to provide spectral sensitivity can dramatically increase the efﬁciency of the
detector due to the many passes light makes within the resonator.
The resonator system we investigate here is composed of a photonic crystal cavity for inplane
conﬁnement, and a plasmonic waveguide [10, 11], composed of either a single or a doublelayer of
metal, for the vertical conﬁnement. This resonator design, combining the beneﬁts of a plasmonic
waveguide and a photonic crystal cavity, has a number of advantages. The plasmonic waveguide
serves multiple purposes: it serves as a superior top contact (or in the doublemetal case, top and
bottom contact) for the detector device providing enhanced extraction efﬁciency; it provides strong
vertical conﬁnement (nearly total conﬁnement, for the doublemetal structure) of the resonator mode
within the active region; and it increases the index contrast in the photonic crystal, enhancing the in
plane conﬁnement of the resonator mode and enabling strong conﬁnement even with a very shallow
photonic crystal etch extending only through the top metal layer [12, 13]. The photonic crystal
patterning also serves a dual purpose: it provides inplane conﬁnement to the resonator mode, serves
as a grating coupler to couple normalincidence light into the inplane direction of the detector, and
provides a mechanism for freely adjusting the polarization response of the detector pixel.
In the past, many promising schemes have been proposed and/or demonstrated illustrating vari
ous aspects of these concepts: optical resonators to provide spectral [14] or spectral and polarization
ﬁltering [15–17], enhanced conﬁnement of light to increase material absorption [14, 16–18], and
metallic gratings to enable strong conﬁnement without the necessity for deep etching [15, 17, 18].
We have also demonstrated plasmonic photonic crystal designs with the maximum ﬁeld intensity
at the top metal interface to allow for thinner devices and increase ﬁeld overlap with the active re
gion versus metallic Fabry Perotbased structures, in both deepetched [19] and shallowetched [13]
singlemetallayer implementations. Here we detail the design and experimental demonstration of
the shallowetch singlemetal resonators, and expand those design principles to propose a highly
efﬁcient shallowetch doublemetal cavity design for hyperspectral and hyperpolarization, strongly
enhanced midinfrared detection. Both singlemetal and doublemetal designs are detector material
agnostic, are easily incorporated into current FPA processing techniques, and do not involve the
3
damage or removal of any detector active region material, providing signiﬁcantly increased ﬂexibil
ity and functionality with a minimal increase in complexity. This work was originally presented in
Refs. [13, 20].
1.2 Photonic Crystal Design
A signiﬁcant obstacle to using resonant cavities to enhance detector absorption and provide spectral
and polarization sensitivity is achieving sufﬁcient input coupling from freespace light. Commonly
such resonators, with their high conﬁnement, have only very poor phasematching to a normal inci
dence freespace beam such as that which we would ideally like to detect for imaging applications.
However, with suitable design and optimization of the plasmonic photonic crystal structure, it be
comes possible to achieve signiﬁcant freespace coupling, and indeed, even move towards achieving
critical coupling (as will be discussed in Section 1.6).
We used group theory to design a frequency and polarization sensitive photonic crystal struc
ture suitable for coupling efﬁciently to normal incidence light. The simplest polarizationsensitive
resonator design would be a onedimensional grating. However, it is beneﬁcial to choose a fully
connected photonic crystal design in order to take full advantage of the increased current extraction
efﬁciency from the plasmonic metal layer serving as the top contact of the detector device, as well
as allowing for continuous variation between polarizationsensitive and polarizationinsensitive de
vices. Therefore, we analyze a squarelattice structure here, and describe how stretching the lattice
in one direction can split the degenerate modes of the structure and create a strong polarization
sensitivity for use in imaging applications.
The bandstructure of a square lattice photonic crystal is shown in Fig. 1.1(a), calculated using
plane wave expansion. The ratio of circular hole radius r to lattice spacing a used was r/a = 0.32,
and the index of the material was taken to be the effective index of the double metal plasmon waveg
uide, n
eff
= 3.24. Details of the plasmon effective index calculation are discussed in Section 1.5.
There are no band gaps for this structure, however there are several ﬂatband regions. The group
velocity of these bandedge modes is close to zero, therefore the light travels very slowly and is
effectively conﬁned within the patterned region. Bandedge modes are ideal for applications such
as detectors, as the mode volume is large, allowing more of the active region to be contained within
the resonator. There are a number of ﬂatband regions within the bandstructure in Fig. 1.1(a), but
we are interested in the modes at the Γpoint. The Γpoint corresponds to normalincidence modulo
4
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
N
o
r
m
a
l
i
z
e
d
F
r
e
q
u
e
n
c
y
(
a
/
λ
0
)
Γ X Γ M
(a) (b)
Γ
0
Γ
0
M
X
Γ
1
Γ
1
Γ
2
Γ
2
Γ
4
Γ
4
Γ
3
Γ
3
G
(c)
Γ
0
Γ
0
M
X
1
Γ
1
Γ
1
Γ
2
Γ
2
Γ
4
Γ
4
Γ
3
Γ
3
G
2
X
2
G
1
Figure 1.1: (a) Inplane guided mode TMlike bandstructure plot of a squarelattice photonic crystal
(n
eff
=3.24) with r/a =0.32. The light line is shown in yellow, and the modes of interest are circled.
(b) Reciprocal lattice for an unstretched photonic crystal. (c) Reciprocal lattice for a stretched
photonic crystal.
a reciprocal lattice vector, so the Γpoint modes are capable of coupling normalincidence light into
the inplane direction of the detector. In addition, the Γpoint modes are above the light line, and
therefore leak into the air, enabling them to couple more easily to an input free space beam. We
investigated the four lowestorder Γpoint modes, circled in Fig. 1.1(a), using group theory [21, 22].
The point group symmetry of the square photonic crystal lattice, with reciprocal lattice shown
in Fig. 1.1(b), is C
4v
. The inplane ﬁeld of the unperturbed waveguide is given by E
k
(r
) =
ˆ ze
(k
)r
, with k
and r
representing the inplane wavenumber and spatial position, respectively.
When the structure is patterned, coupling will occur between waveguide modes with similar unper
turbed frequencies, and propagation constants that differ by a reciprocal lattice vector G.
There is one Γpoint within the ﬁrst Brillouin zone (IBZ), at (0, 0)k
Γ
, with k
Γ
= 2π/a. Since
we are interested in modes with nonzero kvectors in the inplane direction, we will consider the
nearest Γpoints in the surrounding Brillouin zones, at ( (1, 0)k
Γ
, (0, 1)k
Γ
). These points are
labeled in Fig. 1.1(b). The group of the wave vector, the symmetry group of a plane wave modulo
G, is C
4v
at the Γpoint. The character table of C
4v
is shown in Table 1.1.
The star of k (k) at the Γpoint is the set of independent Γpoints within the region. In this
case, k is given, not uniquely, by k
Γ
1
. This will be our seed vector. We ﬁnd the symmetry basis
for the modes at that satellite point by applying the symmetry operations of the group of the wave
vector to the seed vector. In this case, the basis is (E
G
1
, E
G
1
, E
G
2
, E
G
2
). Projecting this symmetry
5
Table 1.1: Point Group character tables for the square and rectangular lattice.
C
4v
E C
2
2C
4
2σ
v
2σ
d
C
2v
E C
2
σ
x
σ
y
A
1
1 1 1 1 1 A
1
1 1 1 1
A
2
1 1 1  1 1 A
2
1 1 1 1
B
1
1 1 1 1 1 B
1
1 1 1 1
B
2
1 1 1 1 1 B
2
1 1 1 1
E 2 1 0 0 0
(a) (b)
0.2
0.3
0.4
0.5
Γ Γ X X
N
o
r
m
a
l
i
z
e
d
F
r
e
q
u
e
n
c
y
(
a
/
λ
0
)
Figure 1.2: 2D bandstructure plots near the gamma point of (a) a square lattice and (b) a rectangular
lattice, stretched by 10%. The dipolelike modes are shown in bold.
6
basis onto the irreducible representation (IRREP) spaces of C
4v
, we ﬁnd the modes:
E
A
1
= ˆ z(cos(k
G
1
r) +cos(k
G
2
r)),
E
B
1
= ˆ z(cos(k
G
1
r) cos(k
G
2
r)),
E
E,1
= ˆ z(sin(k
G
2
r)),
E
E,2
= ˆ z(sin(k
G
1
r)),
(1.1)
where A
1
, B
1
, and E are IRREP spaces of C
4v
(see Table 1.1), and r has its origin at the center of the
air hole. Considering that modes with more electric ﬁeld concentrated in areas with high dielectric
constant tend to have lower frequency than those with electric ﬁeld concentrated in low dielectric
regions [23], we can order the modes by frequency. E is a two dimensional IRREP, so generates two
degenerate modes. We associate this pair of modes E
E,1
, E
E,2
with the second and third frequency
bands, which is in agreement with the bandstructure in Fig. 1.1(a). These degenerate modes, with
dipolelike symmetry and the spatial pattern given in Eqs. 1.1, radiate with a farﬁeld pattern which
is uniform: in the case of a ﬁnite structure, a Gaussianlike farﬁeld without antinodes.
In order to achieve polarization sensitivity, we need to split these two degenerate dipolelike
modes. To do this, we stretch the photonic crystal lattice (not the photonic crystal holes) in one
direction, giving the reciprocal lattice shown in Fig. 1.1(c). The effect of stretching the lattice on
the four lowestorder Γpoint modes is shown in Fig. 1.2. The symmetry group of this perturbation
is C
2v
; the character table for C
2v
is shown in Table 1.1. Using the compatibility relations between
C
4v
and C
2v
, we ﬁnd the new set of modes:
E
A
1
,1
= ˆ z(cos(k
G
1
r) +cos(k
G
2
r)),
E
A
1
,2
= ˆ z(cos(k
G
1
r) cos(k
G
2
r)),
E
B
1
= ˆ z(sin(k
G
2
r)),
E
B
2
= ˆ z(sin(k
G
1
r)).
(1.2)
These modes are plotted in Fig. 1.10. The C
4v
twodimensional representation E decomposes into
B
1
B
2
under C
2v
, therefore the dipolelike modes are no longer degenerate. This is in agreement
with what we see in the bandstructure of the stretched lattice, Fig. 1.2(b).
7
E
z
Intensity of
Guided Mode
(a)
0.25
0.35
ΓΓ
M
X
4
3
2
1
0
(d) (c)
X Γ XΓ
M
X
X
ΓΓ
0.3
0.4
F
r
e
q
u
e
n
c
y
(
a
/
λ
)
0 2 4 6 8 10
0
4
8
12
16
20
R
e
f
r
a
c
t
i
v
e
I
n
d
e
x
/
E
z
I
n
t
e
n
s
i
t
y
(
a
.
u
.
)
Depth (µm)
Active Region
(b)
z
y
Figure 1.3: (a) A crossectional image of several lattice constants of the singlemetal DWELL detec
tor design. (b) E
z
intensity proﬁle of the fundamental plasmon waveguide mode (blue) and the real
part of the refractive index of the layers (red), with the detector active region highlighted. (c) FDTD
bandstructure for the unstretched singlemetal photonic crystal structure shown in (a) in the region
between the Γ and X points. (d) FDTD bandstructure between the Γ and X
1
points for a singlemetal
photonic crystal structure stretched and compressed by 10% in the x and y directions, respectively.
8
1.3 SingleMetal Plasmon Resonator Design
In addition to the inplane conﬁnement provided by the photonic crystal pattern discussed in Sec
tion 1.2, it is necessary to conﬁne the light in the vertical direction as well. We begin with a single
metal design suitable for straightforward fabrication (as experimentally demonstrated in [13]), and
then expand the discussion to consider a doublemetal design that mimics the top and bottom con
tact layers in detector focal plane arrays for easy integration. As we operate in the midinfrared
frequency range, we are far from the plasmon resonance frequency of metals, typically in the ultra
violet; operating in this regime avoids the very high metal losses that occur at frequencies closer to
the plasmon frequency, and allows the mode to extend farther into the active region of the detector.
The singlemetal resonant cavity consists of a single layer of metal with etched square holes in
a square lattice periodic array. A representation of several lattice constants of the device structure is
shown in Fig. 1.3(a). The plasmonic layer provides the vertical conﬁnement, conﬁning the optical
mode with a maximum at the surface of the metal (Fig. 1.4(b,c)), while the etched airholes create a
PC pattern to conﬁne the light inplane. Combined together, this resonator design provides full 3D
conﬁnement, signiﬁcantly increasing the amount of time light spends within the detector active re
gion, and therefore enhancing the probability of detection. Due to the strong index contrast between
the surface plasmon [10, 11, 24] mode beneath the metal regions and the dielectricconﬁned mode
beneath the air holes, this plasmonic PC grating is strong enough to generate an inplane conﬁned
resonant mode without etching into the detector active material [12, 25], allowing a resonator to be
fabricated without damaging or removing active material. The numerical and symmetry analysis
presented in Sec. 1.2 and Ref. 20 shows that the two degenerate dipolelike inplane modes of the
structure (Fig. 1.4(bd)) couple most easily from free space. Further improvements in the freespace
coupling efﬁciency were performed by optimizing the top metal thickness and hole size. In addition,
as the two dipolelike modes couple to orthogonal polarizations of incoming light, a stretch of the
PC lattice breaks the degeneracy of the two modes, splitting their resonance frequencies and thus
achieving high polarization selectivity [6, 16].
The singlemetal resonant DWELL detector structure we study is shown in Fig. 1.3(a), along
with the 1D E
z
intensity proﬁle of the fundamental plasmon waveguide mode in Fig. 1.3(b) (not
including the effects of the photonic crystal holes). The vertical conﬁnement factor of this mode
within the active region of the detector is η = 91%, in strong contrast to the generally much lower
conﬁnement factor of purely dielectric waveguides. This extremely high conﬁnement, even for a
9
x (µm)
z
(
µ
m
)
2
0
2
4
6
8
1
0.5
0
0.5
1
0.5
1
0
0.5
1
x (µm)
y
(
µ
m
)
1 0 1
1
0
1
1
0
0.5
semiconductor
air
active region
(d)
a
x
a
y
W
(a) (b)
y (µm)
z
(
µ
m
)
active region
2
0
2
4
6
8
0 1 1 0.5 0.5
x10
3
0.5
1
x10
3
(c)
0.5
0.5
0.5
0.5
1 0 1 0.5 0.5
x10
3
Figure 1.4: (a) A diagram of the unstretched PC structure showing relevant dimensions. The ex
panded plots show the E
z
mode proﬁle for one lattice constant of one of the two dipole modes for an
unstretched PC lattice in (b) the xz plane along the hole edge, (c) the yz plane along the hole edge,
and (d) the xy plane just beneath the metalsemiconductor interface, for a structure with lattice
constant a = 2.38 µm,
¯
W = 0.6, and metal thickness t = 150 nm.
10
waveguide utilizing only a single layer of metal, immediately showcases the beneﬁts of choosing a
plasmonbased photonic crystal design.
Figure 1.4(a) shows relevant dimensions of the structure; we deﬁne the normalized hole width
as
¯
W = 2W/(a
x
+a
y
). Field proﬁles along different planes for the square lattice dipolelike mode
are plotted in Fig. 1.4(bd) over a single unit cell; the other dipole mode has the same ﬁeld pattern,
rotated by 90 degrees in the xy plane. The simulated active region absorption corresponding to
this mode is 10.9%, given an approximate 2% singlepass absorption in the DWELL material, and
corresponds to an expected resonant responsivity enhancement of 56 times that of a control sample
with no plasmonic layer or PC patterning.
The TM square lattice singlemetal plasmonic photonic crystal bandstructure for the region
near the Γpoint is shown in Fig. 1.3(c), calculated using ﬁnite difference time domain (FDTD)
methods with metal material properties from Ref. 26. The simulated structure has a lattice constant
of a = 2.939 µm, a hole width W vs. lattice constant ratio of
¯
W = 0.567, and a metal thickness
of t
m
= 150 nm, and shows Γpoint modes which are in good agreement with the group theory
predictions in the previous section. Figure 1.3(d) shows the same bandstructure region for a lattice
stretched and compressed by 10% along the ˆ x and ˆ y axes respectively, with the two dipole modes
showing a signiﬁcant frequency splitting, also as predicted. These two 3D FDTD simulations match
up well with the 2D planewave expansion bandstructure predictions shown in Fig. 1.2, with the
addition of visible higherorder vertical modes from the singlemetal plasmon waveguide. The
doublemetal waveguide structure can be designed such that these higherorder vertical modes are
eliminated.
The overall quality factor of the two degenerate dipole modes in the unstretchedlattice case (for
resonator parameters given in the caption to Fig. 1.5) is calculated to be Q
per
= 48 for a perfectly
periodic structure, not including the inplane quality factor, Q
xy
, which can be increased indeﬁnitely
by adding more lattice constants to the resonator structure. By simulating the structure as excited
by an incoming, normal incidence plane wave, we can measure the percentage of the incoming light
absorbed within the active region. The simulated DWELL active region absorption corresponding
to this mode is A
t
= 11.5% (Fig. 1.5(a)), with material parameters speciﬁed so as to reproduce
the approximate DWELL material singlepass absorption, A
DWELL
= 2%. This corresponds to an
expected responsivity enhancement factor of E A
t
/A
DWELL
= 5.75 at the resonant wavelength,
versus a sample with no top patterned plasmonic metal layer at the same wavelength (Fig. 1.5).
Note that this enhancement factor is not normalized to the area of the holes in the plasmonic metal,
11
2
4
6
8
10
12
Wavelength (µm)
E
n
h
a
n
c
e
m
e
n
t
F
a
c
t
o
r
,
E
6 7 8 9 10 11
A
c
t
i
v
e
R
e
g
i
o
n
A
b
s
o
r
p
t
i
o
n
(
%
)
1
2
3
4
5
6
(a)
P
WG
WG
P
P
P
5
(b)
2
0
2
4
6
8
1 1 0
y (µm)
z
(
µ
m
)
0.04
0.02
0
0.02
0.04
2
0
2
4
6
8
1 1 0
y (µm)
z
(
µ
m
)
0.015
0.005
0
0.005
0.015
0.01
0.01
2
0
2
4
6
8
1 1 0
y (µm)
z
(
µ
m
)
0.02
0
0.01
0.02
0.01
P WG P
Figure 1.5: (a) FDTD simulated enhancement factor and active region absorption vs. wavelength,
based on a 2% singlepass absorption, using a structure with lattice constant a = 2.939 µm,
¯
W =
0.567, and metal thickness t
m
= 150 nm. (b) E
z
mode proﬁles in the yz plane at the hole edge
for the three longerwavelength peaks in (a), for one lattice constant. The three shorterwavelength
peaks have similar vertical ﬁeld proﬁles, but are higherorder in the xy plane.
as is typically done in the case of ’extraordinary’ transmission through thin metal layers [24]; the
absorption values are compared over the same physical region of detector material.
The simulated active region absorption, A
t
, and enhancement factor, E, are plotted versus fre
quency in Fig. 1.5(a), showing the fundamental plasmon mode at 9.6 µm and a series of higherorder
modes at shorter wavelengths. The mode P
xy
corresponds to the x
th
order inplane and y
th
order
vertical plasmon waveguide mode. Thus P
00
corresponds to the fundamental plasmon mode as dis
cussed above. WG
1
and WG
2
are the ﬁrst and secondorder TM waveguide modes of the structure,
and Fig. 1.5(b) shows that they have a lower overlap with the metal surface than the plasmon modes.
Thus their Q
per
is higher than the plasmon modes for the perfectly periodic structure simulated, as
can be seen in Fig. 1.5(a), but their inplane Qfactor, Q
xy
, will be very small due to their minimal
interaction with the photonic crystal grating. As the overall absorption enhancement factor is lower
for these modes even in the inﬁnitestructure limit, we conclude that the response of this resonator
structure will be dominated by the surface plasmonguided modes.
12
bottom metal
top metal
plasmon metal
5 µm
(a)
(c)
5 µm
GaAs substrate
AlGaAs
DWELL active region
plasmon metal
(b)
Blackbody
Current Amplifier
Trigger
LN
2
Dewar
(d)
ESA
Chopper
Controller
+
_
Figure 1.6: (a) Optical image of the fabricated device indicating the top, bottom and plasmon met
allizations. (b) Crosssectional SEM of the fabricated squarelattice device indicating the plasmon
metal and device layers. (c) SEM image of the fabricated squarelattice PC pattern on the plasmon
metallization. (d) Schematic of the setup used for measuring responsivity and detectivity of the
fabricated devices. Responsivity measurements were performed by illuminating the sample with
a calibrated Mikron M365 blackbody at T = 800 K. The blackbody radiation was modulated at a
frequency of 400 Hz using a chopper and this signal was used as a trigger for the SRS 760 fast
Fourier transform (FFT) spectrum analyzer. The photocurrent was ampliﬁed using a SRS 570 low
noise ampliﬁer and then measured in the spectrum analyzer.
13
R
e
s
p
o
n
s
i
v
i
t
y
(
A
/
W
)
(d)
0.10
0.05
0.00
6 4 2 0 2 4 6
Voltage (V)
D
e
t
e
c
t
i
v
i
t
y
(
c
m
H
z
1
/
2
/
W
)
10
10
10
9
10
8
10
7
6 4 2 0 2 4 6
Voltage (V)
(e)
5 6 7 8 9 10
Wavelength (µm)
S
p
e
c
t
r
a
l
R
e
s
p
o
n
s
e
(
a
.
u
.
)
unpatterned
a=1.90µm
a=1.97µm
a=2.04µm
a=2.10µm
a=2.17µm
a=2.24µm
a=2.31µm
a=2.38µm
a=1.83µm
(a) (b)
unpolarized
0º
polarized
90º
polarized
5 5.5 6 6.5
Wavelength (µm)
S
p
e
c
t
r
a
l
R
e
s
p
o
n
s
e
(
a
.
u
.
)
unpatterned
unpatterned
response
unprocessed
control
divided 2
0
1
0.5
x 10
x 10
(c)
5 6 7 8 9 10
Wavelength (µm)
S
p
e
c
t
r
a
l
R
e
s
p
o
n
s
e
(
a
.
u
.
)
0
5
0
0
0.5
1
normalized
5.5 6 6.5 7 7.5 8 8.5
Wavelength (µm)
5
4
3
2
1
R
e
s
p
o
n
s
i
v
i
t
y
E
n
h
a
n
c
e
m
e
n
t
,
E
(f )
Sample B 5V bias
Sample A −5V bias
Sample A 5V bias
Sample B −5V bias
Figure 1.7: (a) Normalized spectral response from square lattice devices at a bias of 5 V, indicating
tuning of peak wavelength with the lattice constant. (b) Normalized spectral response from a rectan
gular lattice device (a
x
=1.65 µm, a
y
=2.02 µm) at a bias of 5 V. The response to unpolarized light
is shown in green, and beneath is the response to light polarized at 0 degrees (red) and 90 degrees
(blue) relative to the shorter lattice constant dimension of the lattice. (c) Data processing of the
a = 2.38 µm device response. The unprocessed data (green) is divided by the unpatterned DWELL
response to show the resonances independently of the base detector response. The background
scattering from other regions of the sample (yellow) is then subtracted from the control divided
data (blue) and normalized, to give the ﬁnal spectral response (red) that is plotted in Fig. 1.7(a,b).
(d) Measured responsivity and (e) detectivity of a stretched lattice device with a
x
= 1.78 µm and
a
y
= 2.16 µm (red), and a control device (black). (f) Measured peak responsivity enhancement E
versus resonant device wavelength for two samples at positive and negative bias.
1.4 SingleMetal Experimental Demonstration
To test these predictions, we fabricated two detector samples: sample A with a square lattice PC,
and sample B with a rectangular lattice PC having a lattice constant stretching ratio a
y
/a
x
= 1.2.
Figure 1.6(ac) shows representative images of the fabricated devices. All of the PC patterns had the
same normalized hole width (
¯
W 0.6), but different lattice constant values, and therefore different
resonant wavelengths determined by the scaling of the pattern. The spectral response of the surface
plasmon resonant detectors was measured at 30 K using a Nicolet 870 Fourier transform infrared
spectrometer (FTIR) with the patterned detector sample used in place of the standard FTIR detector.
Responsivity and detectivity measurements were performed at 77 K using the experimental setup
shown in Fig. 1.6(d). To separate out the background scattering and the frequency response of
the DWELL material from the resonant enhancement, we perform data processing as shown for a
typical spectral response measurement in Fig. 1.7(c).
As predicted, by varying the lattice constant and symmetry of the patterned grating, we are
able to tailor the wavelength and polarization response of each detector pixel. Figure 1.7(a) shows
the resonant spectral response from a set of representative detector pixels on sample A, varying
14
the peak wavelength response from 5.5 µm to 7.2 µm by choosing PC lattice constants in a range
from 1.83 µm to 2.38 µm. The linewidth of these resonances is 0.9 µm, providing strong spectral
sensitivity within the broad background DWELL response which covers more than 5 µm. In addition
to the fundamental surface plasmon resonant mode, we also observe a higherorder plasmon mode
as predicted by the FDTD simulations in Fig. 1.4(b), at a wavelength in good agreement with the
theory. In order to generate a polarizationsensitive response, we stretch the lattice constant in one
direction (sample B), splitting the resonant detector response into two wellseparated peaks as shown
in the green curve of Fig. 1.7(b). By varying the polarization of the light incident on the detector,
we show that these two peaks correspond to orthogonal linear polarization directions of incoming
light, as represented by the blue and red curves of Fig. 1.7(b). The high polarization extinction
between the two curves indicates clearly the strong polarization dependence in our device. The
experimentally measured spectral peaks are broadened relative to the FDTD simulated values in
Fig. 1.4(b) due to the ﬁnite extent of the PC pattern ( 50 µm in diameter), and therefore the limited
inplane conﬁnement.
To characterize the efﬁciency of the detector response and the resonant enhancement, we deﬁne
and measure the responsivity and detectivity of samples A and B as follows. The peak responsivity
was computed using the expression
R
p
=
I
0
λ
2
λ
1
R
N
(λ)L
e
(λ, T)A
s
A
d
tF
F
r
2
dλ
(1.3)
where R
N
(λ), I
0
, L
e
, T, A
s
, and A
d
are the normalized spectral response, measured photocurrent, the
black body spectral excitance, the black body source temperature, the area of the source, and the area
of the detector, and r, t, F
F
are the distance between the source and the detector, the transmission of
the window and geometrical form factor, respectively. The lower and upper wavelength bounds of
the detector response are given by λ
1
and λ
2
. The detectivity D
is then
D
=
A
d
∆f
i
n
R
p
(1.4)
where A
d
is the detector area, ∆f is the noise equivalent bandwidth ofour measurement, and i
n
is
the noise current.
Compared to a control (unpatterned) sample, the plasmonic PC patterned devices provide a
strong enhancement of responsivity and a corresponding increase in detectivity across an applied
15
0 200 400 600 800 1000
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
Thickness, t (nm)
3
4
5
6
7
8
E
f
f
e
c
t
i
v
e
I
n
d
e
x
,
n
e
f
f
(a) (b)
0 200 400 600 800 1000
Thickness, t (nm)
E
f
f
e
c
t
i
v
e
I
n
d
e
x
,
n
e
f
f
Figure 1.8: The (a) real (red) and (b) imaginary (blue) parts of the dielectric constant dispersion
relation of an Ag/GaAs/Ag waveguide vs. waveguide thickness t, for a freespace wavelength of
λ = 10 µm. The GaAs core index is indicated by a dotted black line.
bias range from 5 V to 5 V, as shown in Fig. 1.7(d) and (e). The enhancement factor E is deﬁned
as E = R(λ
i
)/R
c
(λ
i
), where R(λ
i
) is the responsivity of the patterned detector at the resonant wave
length and R
c
(λ
i
) is the responsivity of the control sample at the same wavelength. In Fig. 1.7(f),
we show enhancement factors across a range of wavelengths reaching as high as 5X for sample A,
and 4X for sample B.
1.5 DoubleMetal Plasmon Resonator Design
Though we have shown, theoretically and experimentally, that a singlemetal plasmonic device
can have high activeregion conﬁnement and reasonable Qfactors, we can move to a doublemetal
design to increase both of these quantities even further. The doublemetal structure brackets the
active region with a thin layer of plasmonic metal on either side, and the photonic crystal holes are
etched only into the top metal layer, as before. All of the advantages of the singlemetal device
are preserved, while the substrate loss can be essentially eliminated and the detector active region
vertical conﬁnement can approach 100%. These are achieved at the price of higher plasmonic metal
loss, but in the midinfrared region, this loss is not prohibitive.
In choosing the ideal waveguide thickness, t, for the doublemetal (or metalinsulatormetal,
MIM) plasmonic waveguide, there are several considerations. Figure 1.8 shows the variation of
mode effective index n
eff
with waveguide thickness for a Ag/GaAs/Ag waveguide at a freespace
wavelength of λ = 10 µm. As the waveguide thickness decreases and more energy moves into the
metal regions, both the real (Fig. 1.8(a)) and the imaginary (Fig. 1.8(b)) parts of the effective index
16
100 50 0 50 100 150
x 10
5
z (nm)
a
b
s
(
E
)
(
N
/
C
)
n
e
= 7.64+0.2584i
(a)
0
1
2
3
4
5
0 200 400 600
z (nm)
(b)
n
e
= 3.44+0.0112i
a
b
s
(
E
)
(
N
/
C
)
0
1
2
3
4
5
x 10
4
0 10 20 30 40 50 60
z (µm)
(c)
n
e
= 3.30+0.0005i
0
2
4
6
8
10
12
a
b
s
(
E
)
(
N
/
C
)
x 10
3
Figure 1.9: The ﬁeld proﬁle for Ag/GaAs/Ag plasmon waveguides with a thickness t of (a) 10 nm,
(b) 500 nm, and (c) 50 µm are shown, for a freespace wavelength of λ =10 µm. The effective index
n
eff
for each plasmon waveguide is also given.
increase. A high real part of the effective index is beneﬁcial, because it increases the index contrast
of the photonic crystal by increasing the contrast between the photonic crystal holes and the metal
covered regions (the doublemetal waveguide). A higher index contrast increases the strength of the
photonic crystal perturbation, improving the inplane conﬁnement Q
xy
of the resonator. If the index
contrast is high enough, the photonic crystal holes can be etched only into the top metal, without
removing material from the detector active region. In the doublemetal case, this condition is even
easier to achieve than in the singlemetal case. However, the imaginary part of the effective index
is proportional to the loss in the waveguide, and must be minimized. Therefore a waveguide width
must be chosen to balance the competing factors of index contrast and loss.
The ﬁeld proﬁles of three Ag/GaAs/Ag plasmon waveguides are shown in Fig. 1.9, with the
calculated effective index n
eff
, for a freespace wavelength of λ =10 µm. Though the effective index
values shown here would seem to generate only a low index contrast with the core dielectric material
(n
GaAs
= 3.3) for reasonable waveguide thicknesses, it must be considered that these simulations do
not take into account the effect of the photonic crystal holes etched in the metal. In fact, it can be
shown [12, 25] that the presence of holes in a metal layer has the effect of lowering the effective
plasmon frequency of that layer without signiﬁcantly raising losses, increasing the real but not the
imaginary part of the waveguide effective index. Thus, the actual combined photonic crystal and
plasmon structure will have a considerably higher index contrast than would be expected from these
effective index values. This is demonstrated via simulations of the full 3Dstructure which showthat,
even with the photonic crystal holes etched only into the top metal layer, we still achieve signiﬁcant
17
(d)
(b)
(c)
(a)
Figure 1.10: A comparison of the FDTD simulated and group theory predicted E
z
ﬁeld proﬁles for
the four lowest gammapoint modes of a 20% stretched lattice, in order of increasing frequency. The
modes, labeled according to C
2v
designations, are (a) A
1,1
, (b) B
2
, (c) B
1
, (d) A
1,2
. The group theory
predictions are shown on the left, while the FDTD results, including the effects of the photonic
crystal air hole (overlaid white square), are shown on the right. The FDTD ﬁelds shown are 2D
slices of the full simulations taken just below the top metal layer, inside the active region.
inplane conﬁnement and vertical coupling.
Full structure simulations (photonic crystal plus plasmon waveguide) were performed using the
FDTD method on a perfectly periodic lattice (as before, due to computational constraints), with
a lattice stretched by 20%. These FDTD simulations conﬁrm the results of our separate photonic
crystal and plasmon waveguide simulations. Figure 1.10 shows the FDTD ﬁeld plots (left) in com
parison with the group theory mode plots (right). Though the presence of the airhole distorts the
shape of the modes in the center of the FDTD images, at the outside of the simulation region it can
be seen that the simple group theory calculations have accurately predicted the mode shapes given
by the more complex FDTD simulations.
We have also investigated the farﬁeld proﬁles of the four Γpoint modes, through examining
the spatial fourier transform of E
z
. Due to timereversal symmetry, the ﬁeld proﬁle of a mode that
can be coupled into the resonator is equivalent to the ﬁeld proﬁle of the resonator mode propagated
18
y
(
µ
m
)
x (µm)
(a)
100 50 0 50 100
150
100
50
0
50
100
150
1
2
3
4
5
Py/Px > 10
9
(b)
100 50 0 50 100
150
100
50
0
50
100
150
1
2
3
4
5
Px/Py > 10
9
x (µm)
y
(
µ
m
)
Figure 1.11: Farﬁeld plots at z =90 µm of the two Γpoint dipole modes in a stretchedlattice struc
ture with
¯
W = 0.5309 and a
y
/a
x
= 1.2. (a) B
2
mode power density, with dominant ˆ ypolarization.
(b) B
1
mode power density, with dominant ˆ xpolarization. For both B
1
and B
2
modes, the polar
ization selectivity is calculated to be greater than 10
9
, limited entirely by error in the numerical
simulation.
out into the farﬁeld, therefore these farﬁeld plots indicate the modeshapes and polarizations that
couple most strongly from free space to the resonator mode. Farﬁeld plots of the two fundamental
stretchedlattice dipole modes, B
1
and B
2
, are shown in Fig. 1.11, generated from an 10X10 tiled
array of the FDTD simulated ﬁeld proﬁle (itself one lattice constant in size) and apodized using
a Gaussian function with a standard deviation of two lattice constants (a
x
and a
y
, respectively)
in the x and y directions. We can see from Fig. 1.11(a,b) that the B
1
and B
2
modes are well
suited for coupling to incident freespace light, since the farﬁeld proﬁle has a single lobe at normal
incidence and does not contain any antinodes, in agreement with the group theory predictions from
Section 1.2.
1.6 Critical Coupling
After optimizing the largescale resonator design and choosing photonic crystal modes which have
the largest coupling to normalincident light, it still remains to ﬁnd the best values for the design
parameters to increase detector absorption, and to determine the fundamental limits on absorption
enhancement for these two (singlemetal and doublemetal) resonator designs. We ﬁnd that, for the
doublemetal resonator, the absorption is greatest at the point of critical input coupling, whereas we
ﬁnd a different optimal point for the singlemetal resonator, as the parasitic substrate loss increases
along with the detector absorption as the input coupling is increased.
19
Q
xy
+V
V
Q
D
Q
e
Q
metal
Figure 1.12: The dominant loss mechanisms within a doublemetal plasmonic photonic crystal
resonant detector.
Critical coupling occurs when the external coupling to the resonator (the resonator “coupling
loss”) is equal to the total internal cavity loss from all other loss mechanisms. When that condition
occurs, the reﬂection coefﬁcient goes to zero, and all of the incident light at the resonance frequency
is absorbed in the resonator [27]. Figure 1.12 shows the various scattering and absorption processes
that are involved in near normal incidence resonant detection. The reﬂection from the cavity is given
by
R =
∆
2
+
_
γ
0
γ
e
2
_
2
∆
2
+
_
γ
t
2
_
2
, (1.5)
where ∆ = ω ω
0
is the frequency detuning from the resonance frequency ω
0
, γ
0
is the intrinsic
cavity loss rate, γ
e
is the vertical coupling rate to free space, and γ
t
= γ
0
+γ
e
. The loss rates γ are
related to the Qfactors given previously by γ = nλQ/2πc. It is clear from Eq. 1.5 that, when the
cavity is excited onresonance (∆ = 0), the reﬂection goes to zero when γ
0
= γ
e
, when the rate at
which the cavity can be fed from free space is equal to the sum of all the internal cavity loss rates.
This is the critical coupling condition.
From the expression for the reﬂection in Eq. 1.5, we can write the power dropped into the cavity
20
(not only the absorbed power, but all of the power not reﬂected):
P
d
= P
in
(1 R) = P
in
γ
0
γ
e
∆
2
+
_
γ
t
2
_
2
, (1.6)
where P
in
represents the power incident on the cavity. Therefore the fractional absorption efﬁciency
into the ith loss channel is
p
i
=
γ
i
γ
0
P
d
P
in
=
γ
i
γ
e
∆
2
+
_
γ
t
2
_
2
. (1.7)
We can enumerate the loss mechanisms in Fig. 1.12, such that γ
0
= γ
D
+γ
metal
+γ
xy
+γ
sub
, cor
responding to the (beneﬁcial) detector absorption, the metal absorption, the inplane loss, and the
substrate loss, respectively. The inplane loss can always be made negligible, by adding more lat
tice constants to the photonic crystal patterning region to increase the inplane conﬁnement strength
relative to the other loss mechanisms. In the singlemetal case, we can consider γ
sub
= mγ
e
, rep
resenting a mode coupling into the substrate that is m times larger than that into the air due to the
higher substrate refractive index; in the doublemetal case, m 0 due to the thick bottom layer of
plasmon metal.
The fractional absorption into the DWELL detector material is then, at resonance, given by
p
D
=
4γ
D
γe
[(m+1)γ
e
+γ
metal
+γ
D
]
2
. (1.8)
From this expression, we see that the maximum fractional absorption occurs at
γ
e
=
γ
D
+γ
metal
1+m
. (1.9)
As the input coupling γ
e
can be adjusted by varying resonator parameters, this maximal condition
should be readily achievable, corresponding to a fractional absorption into the detector material of
p
D,max
=
γ
D
(1+m)(γ
D
+γ
metal
)
. (1.10)
For the doublemetal case in which m 0, the maximal detector absorption occurs at the pure critical
coupling condition, γ
e
= γ
D
+γ
metal
. In this case, the fractional power absorbed is primarily limited
by the relatively small metal losses in the midinfrared region. For the silver plasmon waveguides
simulated in this work, we ﬁnd a metal loss quality factor of Q
metal
= 149, in comparison to the
21
10
10
10
10
100 200 300 400 500
Metal Thickness, t
m
(nm)
Q
u
a
l
i
t
y
F
a
c
t
o
r
(a)
10
10
10
10
10
0.4 0.5 0.6 0.7 0.8
(b)
Q
u
a
l
i
t
y
F
a
c
t
o
r
W
Figure 1.13: (a) Variation of external coupling and substrate loss quality factors, Q
e
and Q
sub
, with
metal thickness t
m
, and (b) with
¯
W for the fundamental (blue) and higherorder (red) modes of
the unstretched singlemetal photonic crystal lattice. Open circles represent Q
e
and ﬁlled triangles
represent Q
sub
. A dotted line marks the value of the parameter held constant in the opposing plot.
estimated DWELL detector absorption quality factor of Q
D
= 188. This indicates that 55.8% of
the incoming light will be absorbed in the active material for the optimal external coupling quality
factor of Q
e
= 83.
There are many free parameters in this resonator structure which can be optimized in order to
achieve the optimal input coupling value. Choosing two of the most signiﬁcant, the normalized hole
width,
¯
W, and the top metal thickness, t
m
, we investigate their effect on Q
e
for both the singlemetal
unstretched lattice (Fig. 1.13) and doublemetal stretchedlattice (Fig. 1.14, with a lattice stretching
ratio of 1.2) structures. With variation of
¯
W, shown in Figs. 1.13(b) for singlemetal and 1.14(b)
for doublemetal, the overall trend for both structures is the same, showing a curve most likely
due to a combination of factors: the increased hole size provides a larger aperture through which
light can escape, decreasing the Q
e
; but the larger air hole also distorts the shape of the mode,
causing it to generate a less pure farﬁeld proﬁle which does not match as well with a freespace
beam. In contrast, the variations in quality factor with changes in the top metal thickness, shown in
Figs. 1.13(a) for singlemetal and 1.14(a) for doublemetal, illustrate vertical quality factors Q
e
of
both dipolelike modes increasing monotonically as the top metal becomes thicker. Varying the top
metal thickness t
m
is an effective way to change Q
e
to better match the internal loss, and thus more
closely approach critical coupling, without changing the mode frequency.
22
100
200
300
400
500
600
(b)
0.54 0.58 0.62 0.66 0.7
W
0 200 400 600 800 1000 1200
0
2000
4000
6000
8000
10000
Metal Thickness, t
m
(nm)
E
x
t
e
r
n
a
l
Q
u
a
l
i
t
y
F
a
c
t
o
r
,
Q
e
(a)
E
x
t
e
r
n
a
l
Q
u
a
l
i
t
y
F
a
c
t
o
r
,
Q
e
Figure 1.14: Variation of vertical coupling quality factor Q
e
of the B
1
(blue square) and B
2
(green
circle) modes of the doublemetal photonic crystal lattice with changing (a) t
m
and (b)
¯
W.
Though the behavior as hole size and metal thickness are varied is similar for both single
metal and doublemetal structures, the doublemetal structure has a lower achievable Q
e
, indicating
more favorable external coupling conditions; the stretching of the photonic crystal lattice does not
signiﬁcantly decrease Q
e
.
1.7 Conclusion
We have designed a plasmonic photonic crystal resonator utilizing either a singlemetal or double
metal plasmon waveguide for use in midinfrared photodetectors, and experimentally demonstrated
singlemetal devices with responsivity enhancement of up to 5X. This resonator design shows good
frequency and polarization selectivity for use in hyperspectral and hyperpolarization detectors. We
theoretically analyzed the conditions for optimal detector absorption enhancement, and by varying
the photonic crystal hole size and top metal thickness, we adjusted the vertical coupling efﬁciency
to more closely match the resonator loss, moving towards achieving critical coupling. Additional
increases in coupling efﬁciency or reductions in loss could bring the system to near 100% absorption
in the detector. This resonator can be optimized for use at any wavelength from the terahertz to the
visible with suitable scaling of the photonic crystal holes and waveguide width, and can easily be
modiﬁed to suit any detector material, since no photonic crystal holes are etched into the active
region itself.
23
Infrared Light In
Silicon Nitride
Indium Bump
Epoxy Underfill
Photonic Crystal Resonator
Metal
Active Region
Highly Doped Layer
Highly Doped Layer
Metal
Figure 1.15: A design schematic for a resonant doublemetal plasmonic photonic crystal FPA.
The ﬂipchip bonding method of focal plane array (FPA) fabrication naturally lends itself to
use with a doublemetal resonant cavity, with only the top metal photonic crystal lithography step
differing from standard process techniques. DWELL FPAs have already been demonstrated with
hybridization to a readout integrated circuit [4, 28]. In Fig. 1.15, a proposed FPA schematic is
shown, illustrating the ease with which doublemetal plasmonic photonic crystal resonators can
be incorporated into current FPA designs and presenting the possibility to achieve highly sensitive
midinfrared spectral and polarization imaging at low cost.
24
Chapter 2
Double Disk Optomechanical
Resonators
2.1 Introduction
Many precision position measurement devices involve the coupling of mechanical degress of free
dom to an electromagnetic interferometer or cavity [29, 30]. Today, cavitymechanical systems span
a wide range of geometries and scales, from multikilometer long gravitationalwave detectors [31]
to coupled nanomechanicalmicrowave circuits [32]. For the sensitive detection and actuation of
mechanical motion, each of these systems depend upon “dynamical backaction” [33, 34] resulting
from the positiondependent feedback of electromagnetic wave momentum. Recent work in the
optical domain has used the scattering radiation pressure force to both excite and dampen oscilla
tions of a micromechanical resonator [35–41], with the intriguing possibility of selfcooling the
mechanical system down to its quantum groundstate. As has been recently proposed [42, 43] and
demonstrated [44, 45], the optical gradient force within guidedwave nanostructures can be orders
ofmagnitude larger than the scattering force. In this work we combine the large perphoton optical
gradient force with the sensitive feedback of a high quality factor whisperinggallery microcavity.
The cavity geometry, consisting of a pair of silica disks separated by a nanoscale gap, shows ex
tremely strong dynamical backaction, powerful enough to excite giant coherent oscillations even
under heavily damped conditions (mechanical Q 4). In vacuum, the threshold for regenerative
mechanical oscillation is lowered to an optical input power of only 270 nanoWatts, or roughly 1000
stored cavity photons, and efﬁcient cooling of the mechanical motion is obtained with a temperature
compression factor of 13 dB for 4 microWatt of dropped optical input power. These properties of
the doubledisk resonator make it interesting for a broad range of applications from sensitive force
25
and mass detection in viscous environments such as those found in biology [46, 47], to quantum
cavityoptomechanics in which a versatile, chipscale platform for studying the quantum properties
of the system may be envisioned. This work was initially presented in Ref. 48.
2.2 Optomechanical coupling and dynamic backaction
The per photon force exerted on a mechanical object coupled to the optical ﬁeld within a resonant
cavity is given by g
OM
, where g
OM
dω
c
/dx is a coefﬁcient characterizing the dispersive nature
of the cavity with respect to mechanical displacement, x. In a FabryPerot (Fig. 2.2a) or microtoroid
resonator (Fig. 2.2b), the optical force manifests itself as a socalled scattering radiation pressure
due to direct momentum transfer from the reﬂection of photons at the cavity boundary [49, 50]. As
the momentum change of a photon per round trip is ﬁxed inside such cavities, while the roundtrip
time increases linearly with the cavity length, the radiation pressure per photon scales inversely with
the cavity size. In contrast, for the gradient optical force the cavity length and the optomechanical
coupling can be decoupled, allowing for photon momentum to be transfered over a length scale ap
proaching the wavelength of light [42, 43]. This method was recently employed in a silicon photonic
circuit to manipulate a suspended waveguide [44]. However, without the feedback provided by an
optical cavity or interferometer, the optical force only provides a static mechanical displacement.
In the case of a cavity optomechanical system, dynamical backaction can be quantiﬁed by con
sidering the magnitude of the damping/ampliﬁcation that an input laser has on the mechanical mo
tion. For a ﬁxed absorbed optical input power in the badcavity limit (κ Ω
m
), the maximum rate
is given by
Γ
m,opt
_
3
3g
2
OM
κ
3
i
ω
c
m
x
_
_
1
(1+K)
2
_
P
d
, (2.1)
where ω
c
is the optical cavity resonance frequency, m
x
is the motional mass of the optomechanical
system, P
d
is the optical power dropped (absorbed) within the cavity, and K κ
e
/κ
i
is a cavity
loading parameter (κ
i
, the intrinsic energy loss rate of the optical cavity; κ
e
, the energy coupling
rate between external laser and internal cavity ﬁelds). The effectiveness of the coupling between
the optical and mechanical degrees of freedom can thus be described by a backaction parameter,
B = g
2
OM
/
_
κ
3
i
ω
c
m
x
_
, which depends upon the motional mass, the perphoton force, and the optical
cavity Qfactor.
26
2.3 Doubledisk fabrication, optical, and mechanical design
Here we describe the design, fabrication, and characterization of a nanooptomechanical system
(NOMS) consisting of a pair of optically thin disks separated by a nanoscale gap. The double
disk structure (Fig. 2.2c) supports highQ whisperinggallery resonances, and provides backaction
several orders of magnitude larger than in previously demonstrated gradient force optomechanical
systems [44, 45] (very recent work [51] involving the versatile coupling of external nanomechanical
elements to the nearﬁeld of a highFinesse microtoroid has realized very strong dynamical back
action, although still roughly twoorders of magnitude smaller than in our integrated device).
Fabrication of the doubledisk whisperinggallery resonator began with initial deposition of
the cavity layers. The two silica disk layers and the sandwiched amorphous silicon (αSi) layer
were deposited on a (100) silicon substrate by plasmaenhanced chemical vapor deposition, with
a thickness of 340 4 nm and 158 3 nm for the silica and αSi layers, respectively. The wafer
was then thermally annealed in a nitrogen environment at a temperature of T = 1050 K for 6 hours
to drive out water and hydrogen in the ﬁlm, improving the optical quality of the material. The
disk pattern was created using electron beam lithography followed by an optimized C
4
F
8
SF
6
gas
chemistry reactive ion etch. Release of the doubledisk structure was accomplished using a SF
6
chemical plasma etch which selectively (30, 000 : 1) attacks the intermediate αSi layer and the
underlying Si substrate, resulting in a uniform undercut region between the disks extending radially
inwards 6 µm from the disk perimeter. Simultaneously, the underlying silicon support pedestal
is formed. The ﬁnal gap size between the disks was measured to be 138 8 nm (shrinkage having
occurred during the anneal step). Two nanoforks were also fabricated near the doubledisk resonator
to mechanically stabilize and support the ﬁber taper during optical coupling; the geometry was
optimized such that the forks introduce a total insertion loss of only 8%.
The ﬁnal doubledisk structure, shown in Fig. 2.4, consists of 340nmthick silica disks sepa
rated by a 140 nm air gap extending approximately 6 µm in from the disk perimeter (the undercut
region). Two different sized cavities are studied here, one large (D = 90 µm; Sample I) and one
small (D=54 µm; Sample II) in diameter. The small diameter cavity structure represents a minimal
cavity size, beyond which radiation loss becomes appreciable (Q
r
10
8
).
Finite element method (FEM) simulations of the whisperinggallery optical modes of the double
disk structure shows substantial splitting of the cavity modes into even and odd parity bonded and
antibonded modes (Fig. 2.2(ef)). Due to its substantial ﬁeld intensity within the air gap, the bonded
27
mode tunes rapidly with changing gap size as shown in the inset to Fig. 2.2(g). As the mode conﬁne
ment in a doubledisk NOMS is primarily provided by the transverse boundaries formed by the two
disks, the doubledisk structure can be well approximated by a symmetric doubleslab waveguide
shown in Fig. 2.1.
For the bonding mode polarized along the ˆ e
y
direction, the tangential component of the electric
ﬁeld is given by:
E
y
=
_
¸
¸
¸
¸
¸
¸
¸
¸
¸
_
¸
¸
¸
¸
¸
¸
¸
¸
¸
_
Ae
γx
, x > h+x
0
/2
Bcosκx +Csinκx, x
0
/2 < x < h+x
0
/2
Dcoshγx, x
0
/2 < x < x
0
/2
Bcosκx Csinκx, x
0
/2 > x > h x
0
/2
Ae
γx
, x < h x
0
/2
(2.2)
where κ is the transverse component of the propagation constant inside the slabs and γ is the ﬁeld
decay constant in the surrounding area. They are given by the following expressions:
κ
2
= k
2
0
n
2
c
β
2
, γ
2
= β
2
k
2
0
n
2
s
, (2.3)
where k
0
=ω
0
/c is the propagation constant in vacuum and β =k
0
n
eff
is the longitudinal component
of the propagation constant of the bonding mode. n
eff
is the effective refractive index for the guided
mode. Accordingly, the tangential component of the magnetic ﬁeld can be obtained through H
z
=
i
µω
0
∂E
y
∂x
. The continuity of E
y
and H
z
across the boundaries requires κ and γ to satisfy the following
equation:
κγ[1+tanh(γx
0
/2)] =
_
κ
2
γ
2
tanh(γx
0
/2)
¸
tanκh, (2.4)
which reduces to tanκh = γ/κ when x
0
0, as expected.
The circular geometry of the double disk forms the whisperinggallery mode, in which the res
onance condition requires the longitudinal component of the propagation constant, β, to be ﬁxed as
2πRβ = 2mπ, where R is the mode radius and m is an integer. Thus, any variation on the disk spac
ing x
0
transfers to a variation on the resonance frequency ω
0
through Eqs. (2.3) and (2.4), indicating
that ω
0
becomes a function of x
0
. By using these two equations, we ﬁnd that the optomechanical
28
h
h
x
0
z
x
n
c
n
s
n
s
n
s
n
c
Figure 2.1: Schematic of a symmetric doubleslab waveguide. h and x
0
are the slab thickness and
the slab spacing, respectively. n
c
and n
s
are the refractive indices for the slab and surrounding area,
respectively.
coupling coefﬁcient, g
OM
=
dω
0
dx
0
, is given by the general form
g
OM
(x
0
) =
cχγ
2
k
0
sech
2
_
γx
0
2
_
4(n
2
c
n
2
s
)tanκh+n
2
s
x
0
χsech
2
_
γx
0
2
_
+2ξ
_
(n
2
c
γhcsc
2
κh+2n
2
s
)tanκh+
n
2
s
κ
γ
n
2
c
γ
κ
_, (2.5)
where χ κ+γtanκh and ξ 1+tanh(
γx
0
2
).
When x
0
0, Eq. (2.5) leads to the maximum optomechanical coupling of
g
OM
(0) =
ω
0
γ
3
2β
2
+2k
2
0
n
2
c
γh
. (2.6)
In analogy to FabryPerot cavities and microtoroids, the magnitude of the optomechanical coupling
can be characterized by an effective length, L
OM
, deﬁned such that g
OM
ω
0
L
OM
. Equation (2.6) infers
a minimum effective length
L
0
=
2
γ
_
1+
k
2
0
γ
2
(n
2
s
+n
2
c
γh)
_
=
λ
0
π
n
2
eff
+k
0
hn
2
c
_
n
2
eff
n
2
s
_
n
2
eff
n
2
s
_
3/2
, (2.7)
which is approximately on the order of the optical wavelength λ
0
.
Physically, as the two slabs are coupled through the evanescent ﬁeld between them with ampli
tude decaying exponentially with slab spacing at a rate γ [see Eq. (2.2)], the resulting optomechani
29
cal coupling can be well approximated by an exponential function
g
OM
(x
0
) g
OM
(0)e
γx
0
, (2.8)
where g
OM
(0) is given by Eq. (2.6). As indicated by the red curve in Fig. 2.2(g), Eq. (2.8) provides
an excellent approximation for the optomechanical coupling coefﬁcient in a doubledisk NOMS.
Therefore, the approximate effective length, L
OM
ω
0
g
OM
(0)
e
γx
0
, agrees well with the results simu
lated by the ﬁnite element method, shown in Fig. 2.2(g), and the effective length decreases roughly
exponentially with decreasing disk spacing, reaching a minimum value of 3.8 µm at a resonance
optical wavelength of λ
c
1.5 µm. For the air gap of 138 nm used in this work, the optomechanical
coupling is estimated to be g
OM
/2π = 33 GHz/nm (L
OM
= 5.8 µm), equivalent to 22 fN/photon.
The doubledisk structure also supports a number of different micromechanical resonances,
ranging from radial breathing modes to whisperinggallerylike vibrations of the disk perimeter.
The most strongly coupled mechanical resonance is that of the symmetric (i.e., azimuthal mode
number, m = 0) ﬂapping motion of the disks. With a clamped inner edge and a free outer edge,
the mechanical displacement of a double disk exhibiting a ﬂapping mode is generally a function
of radius (Fig. 2.3). What matters for the optomechanical effect, however, is the disk spacing at
the place where the whisperinggallery mode is located, as that determines the magnitude of the
splitting between the bonding and antibonding cavity modes.
As the mechanical displacement actuated by the gradient force is generally small compared with
the original disk spacing x
0
, we can assume it is uniform in the region of the whisperinggallery
mode and deﬁne the effective disk spacing x
m
(r
0
) at the mode center, where r
0
is the radius of the
whisperinggallery mode. The effective mechanical displacement is then given by x
eff
=x
m
(r
0
) x
0
,
corresponding to an effective mechanical potential energy of E
p
= m
x
Ω
2
m
x
2
eff
/2, where m
x
is the
corresponding effective motional mass and Ω
m
is the resonance frequency of the ﬂapping mode.
Note that x
eff
is twice the real displacement at the mode center for a single disk, x
eff
= 2d(r
0
). E
p
reaches its maximum value when the double disk is at rest at its maximum displacement, at which
point all of the mechanical energy is stored in the strain energy U
s
. Therefore, E
p
= U
s
and the
effective motional mass is given by
m
x
=
2U
s
Ω
2
m
[x
m
(r
0
) x
0
]
2
=
U
s
2Ω
2
m
d
2
(r
0
)
, (2.9)
30
L
C
R
g
OM
= ω
0
/R
x
h
x
SiO
2
SiO
2
αSi
(
G
H
z
/
n
m
)
g
O
M
2
π
0
10
20
30
40
50
200 400 600 800
Disk spacing x (nm)
0
E
e
c
t
i
v
e
l
e
n
g
t
h
L
O
M
(
µ
m
)
0
10
20
30
40
50
60
196
200
204
R
e
s
o
n
a
n
c
e
f
r
e
q
u
e
n
c
y
(
T
H
z
)
208
400 800
Disk spacing x (nm)
0
194
2
1
0
1
2
H
e
i
g
h
t
(
µ
m
)
40 42 44 46 48
Radius (µm)
2
1
0
1
2
H
e
i
g
h
t
(
µ
m
)
b
c a
g
d
e
f
g
OM
= ω
0
/L
C
g
OM
= ω
0
/L
OM
Figure 2.2: Schematic of the corresponding (a) FabryParot and (b) microtoroid optomechanical
cavities. (c) Schematic of the doubledisk NOMS structure, showing the mechanical ﬂapping mo
tion of the disks. FEMsimulated optical mode proﬁles of the radial component of the electric ﬁeld
for the (d) bonded mode at λ = 1520 nm and (e) antibonded mode at λ = 1297.3 nm. (f) FEM
simulated tuning curve of the bonded mode. (g) Optomechanical coupling coefﬁcient and effective
length (blue curves) for the bonded mode. g
OM
and L
OM
are both wellapproximated by exponential
functions (red curves).
31
x
0
x
m
(r
0
)
r
0
whisperinggallery mode
d(r)
r
a
r
b
Figure 2.3: Illustration of the disk displacement. x
0
is the disk spacing in the absence of the optical
ﬁeld. r
0
is the radius of the whisperinggallery mode. x
m
(r
0
) corresponds to the effective disk
spacing at the mode center. r
a
and r
b
are the inner and outer radii of the disk region involved in the
ﬂapping motion. d(r) is the mechanical displacement at radius r.
where both U
s
and d(r
0
) can be obtained from the mechanical simulations by the ﬁnite element
method.
The relationship between the effective mass and the physical mass of the doubledisk NOMS
can be found by examining the mechanical potential energy. With a mechanical displacement d(r)
for each single disk [Fig. 2.3], we can ﬁnd the total mechanical potential energy by integrating over
the disk regions involved in the ﬂapping motion:
E
p
=
r
b
r
a
Ω
2
m
d
2
(r)ζ2πrhdr, (2.10)
where ζ is the material density, h is the thickness for a single disk, and r
a
and r
b
are the inner and
outer radii of the disk region involved in the ﬂapping motion (see Fig. 2.3). Note that E
p
is the
total potential energy for the two disks, which is simply two times that of a single one because of
the symmetry between the two disks. As the physical mass of a single disk region involved in the
ﬂapping motion is given by m
p
= πζh(r
2
b
r
2
a
), using Eq. (2.10), we ﬁnd that the effective mass is
related to the physical mass through the following expression:
m
x
=
4m
p
_
r
2
b
r
2
a
_
[x
m
(r
0
) x
0
]
2
r
b
r
a
rd
2
(r)dr =
m
p
_
r
2
b
r
2
a
_
d
2
(r
0
)
r
b
r
a
rd
2
(r)dr. (2.11)
As the whisperinggallery mode is generally located close to the disk edge (i.e., the mode radius
r
0
= 44 µm in a double disk with r
b
= 45 µm), d
2
(r)/d
2
(r
0
) 1 for most of the region between
32
r
a
and r
b
, and Eq. (2.11) shows that m
x
m
p
/2. Therefore, the effective mass is signiﬁcantly
less than half the physical mass of a single disk region. In practice, the effective mass is much
smaller than this value because of the real displacement function d(r). For the 90µm device used
in our experiment, with an 6 µm undercut air gap region involved in the ﬂapping motion (Fig. 2.2c
and Fig. 2.4), the effective mass is 0.264 nanogram, only about one ﬁfth of the physical mass of a
single disk region m
p
= 1.18 nanogram. The effective mass decreases to 0.145 nanogram for the
54µm device, due to the decrease in the disk radius. Note that both these values are more than two
orders of magnitude smaller than commonly used micromirrors and microtoroids [36–41, 49], and
in combination with the large perphoton force, provide a signiﬁcant enhancement to the dynamic
backaction parameter which scales as g
2
OM
/m
x
.
2.4 Optical and mechanical characterization
Optical and mechanical measurements were initially performed at room temperature in a one atmo
sphere nitrogen environment. Fig. 2.6(a) shows the wavelength scan of a large diameter doubledisk
cavity (Sample I). Several radialorder whisperinggallery modes are evident in the spectrum, all of
them of TElike polarization and bonded mode character. The fundamental TElike bonded optical
mode at λ = 1518.57 nm is shown in the Fig. 2.6(a) inset, from which an intrinsic optical Qfactor
of 1.75 10
6
is inferred, taking into account the mechanical perturbations.
Unlike other microcavities in which the linear transmission is determined only by the cavity loss
and dispersion, for the doubledisk NOMS, even the small thermal Brownian motions of the ﬂapping
mode introduce signiﬁcant perturbations to the cavity resonance due to the large optomechanical
coupling, leading to considerably broadened cavity transmission. Figure 2.5(a) shows an example
of the cavity transmission of Sample I. With a small input power of 5.8 µWwell belowthe oscillation
threshold, the cavity transmission exhibits intense ﬂuctuations when the laser frequency is scanned
across the cavity resonance. As a result, the averaged spectrum of the cavity transmission (red
curve) is signiﬁcantly broader than the real cavity resonance. A correct description of the cavity
transmission requires an appropriate inclusion of the optomechanical effect, which is developed in
the following.
When the optical power is well below the oscillation threshold and the ﬂapping mode of the
double disk is dominantly driven by thermal ﬂuctuations, the mechanical motion can be described
33
VOA
14951565 nm
Tunable Laser
MZ I
10 : 90
splitter
Reference
Detector 1
Polarization
Controller
double disk
vacuum chamber
fiber taper
TBPF
10 : 90
splitter
EDFA
VOA
Highspeed
Detector
Oscilloscope
Reference
Detector 2
20 µm
1 µm c b
a
pedestal
undercut
nanoforks
Figure 2.4: (a) Schematic of the experimental setup for optical testing of the doubledisk cavity. The
cavity input and transmission are both transported through a singlemode silica ﬁber taper, which is
supported by two nanoforks for stable operation. A tunable laser source is used to optically probe
and actuate the doubledisk structure, with input power controlled by a variable optical attenuator
(VOA) and wavelength calibrated by a MachZehnder interferometer (MZI). For experiments per
formed in a nitrogen environment, the cavity transmission is sent directly to the photodetectors,
while it is ﬁrst ampliﬁed by an erbiumdoped ﬁber ampliﬁer (EDFA) for the experiments performed
in vacuum. (b,c) Scanning electron microscope images of the 54µm doubledisk NOMS. False
color is used to indicate different relevant regions of the device.
34
1516 1520 1524 1528 1532
0
0.2
0.4
0.6
0.8
1
Wavelength (nm)
N
o
r
m
a
l
i
z
e
d
T
r
a
n
s
m
i
s
s
i
o
n
(b)
0.94
0.96
0.98
1
−30 −20 −10 0 10 20 30
[λ−1518.57nm] (pm)
Q
i
= 1.75 10
6
−10 −5 0 5 10
0.4
0.6
0.8
1
Wavelength Detuning (pm)
N
o
r
m
a
l
i
z
e
d
T
r
a
n
s
m
i
s
s
i
o
n
(a)
Figure 2.5: (a) The cavity transmission of Sample I in a nitrogen environment, when the laser is
scanned across the cavity resonance at 1518.57 nm with an input power of 5.8 µW. The blue curve is
the instantaneous signal collected by the highspeed detector and the red curve is the average signal
collected by the slow reference detector 2. The slight asymmetry in the transmission spectrum is
due to the static component of mechanical actuation when the laser is scanned from blue to red. The
dashed line indicates the laser frequency detuning used to record the power spectral density shown
in the top panel of Fig. 2.6(b). (b) Linear scan of the averaged cavity transmission of Sample I at an
input power of 2.9 µW. The inset shows a detailed scan for the bonding mode at 1518.57 nm, with
the experimental data in blue and the theoretical ﬁtting in red.
by the following equation:
d
2
x
dt
2
+Γ
m
dx
dt
+Ω
2
m
x =
F
T
(t)
m
x
, (2.12)
where Ω
m
, Γ
m
, and m
x
are the resonance frequency, damping constant, and effective mass of the
ﬂapping mode, respectively. F
T
is the Langevin force driving the mechanical Brownian motion, a
Markovin process with the following correlation function:
F
T
(t)F
T
(t +τ)= 2m
x
Γ
m
k
B
Tδ(τ), (2.13)
where T is the temperature and k
B
is Boltzmann’s constant. It can be shown easily from Eqs. (2.12)
and (2.13) that the Brownian motion of the ﬂapping mode is also a Markovin process with a spectral
correlation given by ¯ x(Ω
1
)¯ x
(Ω
2
)= 2πS
x
(Ω
1
)δ(Ω
1
Ω
2
), where ¯ x(Ω) is the Fourier transform
of the mechanical displacement x(t) deﬁned as ¯ x(Ω) =
+∞
∞
x(t)e
iΩt
dt, and S
x
(Ω) is the spectral
intensity for the thermal mechanical displacement with the following form:
S
x
(Ω) =
2Γ
m
k
B
T/m
x
(Ω
2
m
Ω
2
)
2
+(ΩΓ
m
)
2
. (2.14)
35
The time correlation of the mechanical displacement is thus given by
x(t)x(t +τ)=
1
2π
+∞
∞
S
x
(Ω)e
iΩτ
dτ x
2
ρ(τ) x
2
e
Γ
m
τ/2
cosΩ
m
τ, (2.15)
where x
2
= k
B
T/(m
x
Ω
2
m
) is the variance of the thermal mechanical displacement and ρ(τ) is the
normalized autocorrelation function for the mechanical displacement.
To be general, we consider a doublet resonance in which two optical ﬁelds, one forward and
the other backward propagating, circulate inside the microcavity and couple via Rayleigh scattering
from the surface roughness. The optical ﬁelds inside the cavity satisfy the following equations:
da
f
dt
= (i∆
0
κ/2 ig
OM
x)a
f
+iηa
b
+i
κ
e
A
in
, (2.16)
da
b
dt
= (i∆
0
κ/2 ig
OM
x)a
b
+iηa
f
, (2.17)
where a
f
and a
b
are the forward and backward whisperinggallery modes (WGMs), normalized
such that U
j
= a
j
2
( j = f , b) represents the mode energy. A
in
is the input optical wave, normalized
such that P
in
= A
in
2
represents the input power. κ is the photon decay rate for the loaded cavity,
and κ
e
is the photon escape rate associated with the external coupling. ∆
0
=ω ω
0
is the frequency
detuning from the input wave to the cavity resonance and η is the mode coupling coefﬁcient. In the
case of a continuouswave input, Eqs. (2.16) and (2.17) provide a formal solution of the forward
WGM:
a
f
(t) = i
κ
e
A
in
+∞
0
cos(ητ) f (τ)e
ig
OM
τ
0
x(t τ
)dτ
dτ, (2.18)
where f (τ) e
(i∆
0
κ/2)τ
represents the cavity response. Using Eq. (2.15), we ﬁnd that the statisti
cally averaged intracavity ﬁeld is given as:
a
f
(t)= i
κ
e
A
in
+∞
0
cos(ητ) f (τ)e
ε
2
h(τ)
dτ, (2.19)
where ε g
2
OM
x
2
and h(τ) is deﬁned as
h(τ)
τ
0
ρ(τ
1
τ
2
)dτ
1
dτ
2
. (2.20)
36
Similarly, we can ﬁnd the averaged energy for the forward WGM as:
U
f
(t)= κ
e
P
in
+∞
0
f (τ
1
) f
(τ
2
)cos(ητ
1
)cos(ητ
2
)e
ε
2
h(τ
1
τ
2
)
dτ
1
dτ
2
=
κ
e
P
in
2κ
κ iη
κ 2iη
+∞
0
e
ε
2
h(τ)
[ f
c
(τ) + f
s
(τ)] dτ +c.c., (2.21)
where f
j
(τ) e
(i∆
j
κ/2)τ
( j = c, s), with ∆
c
=∆
0
+η and ∆
s
=∆
0
η. c.c. denotes complex conju
gate.
As the transmitted power from the double disk is given by
P
T
(t) = P
in
+κ
e
U
f
(t) +i
κ
e
_
A
in
a
f
(t) A
in
a
f
(t)
¸
, (2.22)
the averaged cavity transmission, T P
T
/P
in
, thus takes the form
T= 1
κ
e
κ
i
2κ
__
1
iηκ
e
κ
i
(κ 2iη)
_
+∞
0
e
ε
2
h(τ)
[ f
c
(τ) + f
s
(τ)] dτ +c.c.
_
. (2.23)
In the case of a singlet resonance, η = 0 and Eq. (2.23) reduces to the simple form expression
T= 1
κ
e
κ
i
κ
+∞
0
e
ε
2
h(τ)
[ f (τ) + f
(τ)] dτ. (2.24)
In the absence of optomechanical coupling, g
OM
= 0 and Eq. (2.24) reduces to the conventional
form of
T = 1
κ
e
κ
i
∆
2
0
+(κ/2)
2
, (2.25)
as expected.
Using the theory developed above and ﬁtting the experimental averaged cavity transmission
spectrum, we obtain the optical Q factor of the resonance, as shown in Fig. 2.5(b) for Sample I. The
same approach is used to describe the cavity transmission of Sample II, given in Fig. 2.6(a).
The radiofrequency (RF) power spectrum of the optical signal transmitted through the cavity
(Fig. 2.6(b), top panel) exhibits three clear frequency components at 8.30, 13.6, and 27.9 MHz
corresponding to thermallyactuated resonances of the doubledisk structure. These values agree
well with FEM simulations of the differential ﬂapping mode (7.95 MHz), and the ﬁrst (14.2 MHz)
and second (28.7 MHz) order radial breathing modes (Fig. 2.6c). The strong dynamic backaction
of the ﬂapping mode (under thermal excitation) also produces a broadband spectral background in
37
the RF spectrum with a shoulder at the second harmonic frequency. A correct description of the
power spectrum (Fig. 2.6(b), red curve) shows that the ﬂapping mode has a 3dB linewidth of 2.1
MHz (mechanical Qfactor, Q
M
= 3.95), limited by the squeezeﬁlm process of the nitrogen gas
between the disks [52].
We can describe the power spectral density of the cavity transmission in the presence of mechan
ical Brownian motion using a linearperturbation approximation when the optomechanical effects
are small, and a nonperturbation theory, accurate for arbitrarily strong optomechanical effects,
when the effects are larger. Both analyses are presented here.
If the induced optomechanical perturbations are small, Eq. (2.18) can be approximated as
a
f
(t) i
κ
e
A
in
+∞
0
cos(ητ) f (τ)
_
1 ig
OM
τ
0
x(t τ
)dτ
_
dτ. (2.26)
In this case, the transmitted optical ﬁeld can be written as A
T
(t) = A
in
+i
κ
e
a
f
(t) A
0
+δA(t),
where A
0
is the transmitted ﬁeld in the absence of the optomechanical effect and δA is the induced
perturbation. They take the following forms:
A
0
= A
in
_
1 κ
e
+∞
0
cos(ητ) f (τ)dτ
_
A
in
ˆ
A
0
, (2.27)
δA(t) = ig
OM
κ
e
A
in
+∞
0
dτcos(ητ) f (τ)
τ
0
x(t τ
)dτ
. (2.28)
The transmitted power then becomes P(t) = A
T
(t)
2
A
0
2
+A
0
δA(t) +A
0
δA
(t). It is easy to
show that δA(t) = 0 and P
T
(t) = A
0
2
. As a result, the power ﬂuctuations, δP(t) P
T
(t)
P
T
(t), become
δP(t) g
OM
P
in
+∞
0
dτu(τ)
τ
0
x(t τ
)dτ
, (2.29)
where u(τ) iκ
e
cos(ητ)[
ˆ
A
0
f (τ)
ˆ
A
0
f
(τ)]. By using Eq. (2.15), we ﬁnd the autocorrelation func
tion for the power ﬂuctuation to be
δP(t)δP(t +t
0
) εP
2
in
+∞
0
dτ
1
dτ
2
u(τ
1
)u(τ
2
)ψ(t
0
, τ
1
, τ
2
), (2.30)
where ψ(t
0
, τ
1
, τ
2
) is deﬁned as
ψ(t
0
, τ
1
, τ
2
)
τ
1
0
dτ
1
τ
2
0
dτ
2
ρ(t
0
+τ
1
τ
2
). (2.31)
38
Taking the Fourier transform of Eq. (2.30), we obtain the power spectral density S
P
(Ω) of the cavity
transmission to be
S
P
(Ω) g
2
OM
P
2
in
H(Ω)S
x
(Ω), (2.32)
where S
x
(Ω) is the spectral intensity of the mechanical displacement given in Eq. (2.14) and H(Ω)
is the cavity transfer function given by
H(Ω) =
¸
¸
¸
¸
1
Ω
+∞
0
u(τ)(e
iΩτ
1)dτ
¸
¸
¸
¸
2
. (2.33)
In the case of a singlet resonance, the cavity transfer function takes the form:
H(Ω) =
κ
2
e
_
∆
2
0
+(κ/2)
2
¸
2
4∆
2
0
(κ
2
i
+Ω
2
)
[(∆
0
+Ω)
2
+(κ/2)
2
] [(∆
0
Ω)
2
+(κ/2)
2
]
. (2.34)
In most cases, the photon decay rate inside the cavity is much larger than the mechanical damp
ing rate, κ Γ
m
. For a speciﬁc mechanical mode at the frequency Ω
m
, the cavity transfer function
can be well approximated by H(Ω) H(Ω
m
). In particular, in the sidebandunresolved regime, the
cavity transfer function is given by a simple form of
H =
4κ
2
e
κ
2
i
∆
2
0
_
∆
2
0
+(κ/2)
2
¸
4
. (2.35)
Therefore, Eq. (2.32) shows clearly that, if the optomechanical effect is small, the power spectral
density of the cavity transmission is directly proportional to the spectral intensity of the mechanical
displacement.
The situation becomes quite complicated when the optomechanical effects are large. From
Eq. (2.22), the autocorrelation function for the power ﬂuctuation of the cavity transmission, δP(t)
P
T
(t) P
T
, is given by
δP(t
1
)δP(t
2
)= κ
2
e
U
f 1
U
f 2
κ
e
_
A
in
a
f 1
A
in
a
f 1
__
A
in
a
f 2
A
in
a
f 2
_
+ iκ
3/2
e
_
U
f 1
_
A
in
a
f 2
A
in
a
f 2
_
+U
f 2
_
A
in
a
f 1
A
in
a
f 1
_
¸
_
κ
e
U
f
+i
κ
e
_
A
in
a
f
A
in
a
f
_¸
2
, (2.36)
where U
f j
= U
f
(t
j
) and a
f j
= a
f
(t
j
) ( j = 1, 2). Equation (2.36) shows that the autocorrelation
function involves various correlations between the intracavity energy and ﬁeld, all of which can
39
be found using Eqs. (2.15) and (2.18). For example, we can ﬁnd the following correlation for the
intracavity ﬁeld:
_
A
in
a
f 1
A
in
a
f 1
__
A
in
a
f 2
A
in
a
f 2
_
= κ
e
P
2
in
+∞
0
dτ
1
dτ
2
C
1
C
2
e
ε
2
(h
1
+h
2
)
_
f
1
f
2
e
εψ
+ f
1
f
2
e
εψ
+c.c.
¸
, (2.37)
where, in the integrand, C
j
= cos(ητ
j
), h
j
= h(τ
j
), f
j
= f (τ
j
) (with j = 1, 2), and ψ = ψ(t
2
t
1
, τ
1
, τ
2
). h(τ) and ψ(t
2
t
1
, τ
1
, τ
2
) are given by Eqs. (2.20) and (2.31), respectively.
Equations (2.20) and (2.31) show that h(τ) and ψ(t
2
t
1
, τ
1
, τ
2
) vary with time on time scales
of 1/Ω
m
and 1/Γ
m
. However, in the sidebandunresolved regime, κ Γ
m
and κ Ω
m
. As the
cavity response function f (τ) decays exponentially with time at a rate of κ/2, the integrand in
Eq. (2.37) becomes negligible when τ
1
2/κ or τ
2
2/κ. Therefore, ψ(t
2
t
1
, τ
1
, τ
2
) can be well
approximated as
ψ(t
2
t
1
, τ
1
, τ
2
) =
1
2πx
2
+∞
∞
S
x
(Ω)
Ω
2
e
iΩ(t
2
t
1
)
_
e
iΩτ
1
1
__
e
iΩτ
2
1
_
dΩ
τ
1
τ
2
2πx
2
+∞
∞
S
x
(Ω)e
iΩ(t
2
t
1
)
dΩ = τ
1
τ
2
ρ(t
2
t
1
). (2.38)
Similarly, h(τ) τ
2
, since h(τ) = ψ(0, τ, τ). Therefore, Eq. (2.37) becomes
_
A
in
a
f 1
A
in
a
f 1
__
A
in
a
f 2
A
in
a
f 2
_
κ
e
P
2
in
Φ(∆t,C
1
C
2
), (2.39)
where ∆t =t
2
t
1
and Φ(∆t,C
1
C
2
) is deﬁned as
Φ(∆t,C
1
C
2
)
+∞
0
dτ
1
dτ
2
C
1
C
2
e
ε
2
(τ
2
1
+τ
2
2
)
_
f
1
f
2
e
ετ
1
τ
2
ρ
+ f
1
f
2
e
ετ
1
τ
2
ρ
+c.c.
¸
, (2.40)
with ρ =ρ(∆t). Following a similar approach, we can ﬁnd the other correlation terms in Eq. (2.36).
Using these terms in Eq. (2.36), we ﬁnd that the autocorrelation function of the power ﬂuctuations
is given by
δP(t
1
)δP(t
2
) κ
2
e
P
2
in
Φ(∆t, σ
1
σ
2
)
_
κ
e
U
f
+i
κ
e
_
A
in
a
f
A
in
a
f
_¸
2
, (2.41)
40
where σ
j
= σ(τ
j
) ( j = 1, 2) and σ(τ) is deﬁned as
σ(τ)
_
1
κ
e
(κ
2
+2η
2
)
κ(κ
2
+4η
2
)
_
cos(ητ) +
ηκ
e
κ
2
+4η
2
sin(ητ). (2.42)
Moreover, Eq. (2.19) and (2.21) show that, in the sidebandunresolved regime, a
f
and U
f
are well approximated by
a
f
(t) i
κ
e
A
in
+∞
0
cos(ητ) f (τ)e
ε
2
τ
2
dτ, (2.43)
U
f
(t) κ
e
P
in
+∞
0
f (τ
1
) f
(τ
2
)cos(ητ
1
)cos(ητ
2
)e
ε
2
(τ
1
τ
2
)
2
dτ
1
dτ
2
. (2.44)
Therefore, we obtain the ﬁnal term in Eq. (2.41) as
κ
e
U
f
+i
κ
e
_
A
in
a
f
A
in
a
f
_
κ
e
P
in
+∞
0
σ(τ)[ f (τ) + f
(τ)] e
ε
2
τ
2
dτ. (2.45)
Using this term in Eq. (2.41), we obtain the ﬁnal form for the autocorrelation of the power ﬂuctua
tions:
δP(t
1
)δP(t
2
) κ
2
e
P
2
in
[Φ(∆t, σ
1
σ
2
) Φ(∞, σ
1
σ
2
)] . (2.46)
It can be further simpliﬁed if we notice that the exponential function e
ετ
1
τ
2
ρ(∆t)
in Eq. (2.40) can
be expanded in a Taylor series as
e
ετ
1
τ
2
ρ(∆t)
=
+∞
∑
n=0
( ετ
1
τ
2
)
n
n!
ρ
n
(∆t). (2.47)
Substituting this expression into Eq. (2.40) and using it in Eq. (2.46), we obtain the autocorrelation
function for the power ﬂuctuation in the following form
δP(t)δP(t +t
0
) κ
2
e
P
2
in
+∞
∑
n=1
ε
n
ρ
n
(t
0
)
n!
G
n
+( 1)
n
G
n
2
, (2.48)
where G
n
is deﬁned as
G
n
+∞
0
τ
n
σ(τ) f (τ)e
ε
2
τ
2
dτ. (2.49)
In the case of a singlet resonance, η = 0 and σ(τ) simpliﬁes considerably to σ =κ
i
/κ. The autocor
relation function for the power ﬂuctuation is still described by Eq. (2.48).
In general, the power spectral density of the cavity transmission is given by the Fourier transform
41
of Eq. (2.48):
S
p
(Ω) = κ
2
e
P
2
in
+∞
∑
n=1
ε
n
S
n
(Ω)
n!
G
n
+( 1)
n
G
n
2
, (2.50)
where S
n
(Ω) is deﬁned as
S
n
(Ω) =
+∞
∞
ρ
n
(τ)e
iΩτ
dτ. (2.51)
Eq. (2.14) shows that the spectral intensity of the mechanical displacement can be approximated
by a Lorentzian function, resulting in an approximated ρ(τ) given as ρ(τ) e
Γ
m
τ/2
cosΩ
m
τ [see
Eq. (2.15)]. As a result, Eq. (2.51) becomes
S
n
(Ω)
1
2
n
n
∑
k=0
n!
k!(n k)!
nΓ
m
(nΓ
m
/2)
2
+[(2k n)Ω
m
+Ω]
2
. (2.52)
Combining Eq. (2.50) and (2.52), we can see that, if the optomechanical coupling is signiﬁcant, the
thermal mechanical motion creates spectral components around the harmonics of the mechanical
frequency with broader linewidths. As shown clearly in Fig. 2.6(b), the second harmonic is clearly
visible. In particular, if the fundamental mechanical linewidth is broad, various frequency com
ponents on the power spectrum would smear out, producing a broadband spectral background, as
shown in the top panel of Fig. 2.6(b) for Sample I. This phenomenon is similar to the randomﬁeld
induced spectral broadening in nuclear magnetic resonance [53] and atomic resonance ﬂuorescence
[54].
This theory can be extended easily for the case with multiple mechanical frequencies. In this
case, the power spectrum only only exhibits harmonics of each mechanical frequency, but also
their frequency sums and differences. As shown in the bottom panel of Fig. 2.6(b), the frequency
components near 0 MHz are the differential frequencies and those near 1820 MHz are the second
harmonic and sum frequencies.
2.5 Regenerative oscillation
2.5.1 Ambient pressure
Despite the nearunity mechanical quality factor of the ﬂapping mode, the powerful dynamic back
action in the doubledisk structure provides sufﬁcient compensation of mechanical loss to excite
regenerative mechanical oscillation. As shown in Fig. 2.6(d), with an input optical power of 760 µW
launched at the blue detuned side of the resonance, the induced parametric mechanical instability
42
1516 1520 1524 1528 1532
0
0.2
0.4
0.6
0.8
1
Wavelength (nm)
N
o
r
m
a
l
i
z
e
d
T
r
a
n
s
m
i
s
s
i
o
n
0.94
0.96
0.98
1
−30 −20 −10 0 10 20 30
[λ−1518.57nm] (pm)
Q
i
= 1.75 10
6
1 2 3 4
120
110
100
90
P
o
w
e
r
S
p
e
c
t
r
a
l
D
e
n
s
i
t
y
(
d
B
m
/
H
z
)
130
0 10 20 30 40 50
Frequency (MHz)
1
2
3 4
c
110
100
90
Q
m
= 3.95
b
0 50 100 150 200
0.5
1
Time (ns)
50
0
M
e
c
h
a
n
i
c
a
l
d
i
s
p
l
a
c
e
m
e
n
t
(
p
m
)
0
0.5
1
N
o
r
m
a
l
i
z
e
d
t
r
a
n
s
m
i
s
s
i
o
n
Time (µs)
0 0.2 0.4 0.6 0.8 1
0
25
8.2 8.6 9 9.4 9.8
110
100
90
80
Frequency (MHz)
Q
m
= 4.07 10
3
d
e
a
Figure 2.6: (a) Optical transmission spectrumof a large diameter (D=90 µm; Sample I) doubledisk
cavity. The inset shows the fundamental TElike bonded mode at λ =1518.57 nm. (b) Upper panel:
optical transmission power spectral density (PSD) of a Sample I doubledisk in the 1 atm. nitrogen
environment for P
i
= 5.8 µW. Experimental data in blue, theoretical modeling in red, and detector
noise background in yellow. Lower panel: transmission PSD of a small diameter (D = 54 µm;
Sample II) doubledisk cavity in vacuum for P
i
=44 nW. The inset shows a zoomin of the spectrum
around the fundamental ﬂapping mode frequency. (c) FEM simulated mechanical modes indicated
in (b). (d) Recorded transmission waveform of Sample I for P
i
= 0.76 mW. (e) Comparison of
experimental (blue curve) and simulated (red curve) waveforms, with the corresponding simulated
mechanical displacement (green curve).
43
causes the cavity transmission to oscillate over the entire coupling depth with a fundamental fre
quency of 13.97 MHz (this value is about 68% larger than the intrinsic mechanical frequency due
to the optical spring effect [55]). A zoomin of the recorded time waveform (Fig. 2.6e) agrees well
with our numerical simulation which shows that the gradient force actuates an extremely large
(50 pm) mechanical displacement amplitude, dragging the cavity resonance over more than 10
cavitylinewidths and leaving distinctive features of the Lorentzian cavity transfer function. In par
ticular, two sequential passes of the cavity resonance across the laser frequency can be seen, along
with an overshoot and oscillation of the transmitted optical power resulting from the quick release
of Doppler shifted photons from the cavity.
The optomechanical oscillations are simulated through the following coupled equations govern
ing the intracavity optical ﬁeld and mechanical motions, respectively:
da
dt
= (i∆
0
κ
2
ig
OM
x)a+i
κ
e
A
in
, (2.53)
d
2
x
dt
2
+Γ
m
dx
dt
+Ω
2
m
x =
F
T
(t)
m
x
+
F
o
(t)
m
x
, (2.54)
where we have counted in both the thermal Langevin force F
T
and the optical gradient force F
o
=
g
OM
a
2
ω
0
for actuating mechanical motions.
The threshold for regenerative oscillation depends sensitively upon the optical input power and
the average lasercavity resonance detuning, a map of which can be used to quantify the strength of
the dynamic backaction. An estimate of the threshold detuning (∆
th
), for a given input power, can
be determined from the abrupt kink in the cavity transmission that marks the onset of regenerative
oscillation (Fig. 2.8(a) and Fig. 2.7).
Figure 2.7 shows an example of the cavity transmission of Sample I. The mechanical ﬂapping
mode starts to oscillate when the input laser frequency is scanned across a certain detuning. Within
this detuning value, the same magnitude of optomechanical oscillation is excited over a broad range
of laser blue detuning. The intense transmission oscillations cover the entire coupling depth, leaving
an abrupt kink on the transmission spectrum. The coupling depth at the kink point, ∆T
th
, corresponds
to the threshold coupling at the given power level, from which we can obtain the threshold frequency
detuning ∆
th
.
The detuning dependence of the optomechanical ampliﬁcation coefﬁcient can be lumped into a
44
−30 −20 −10 0 10
Wavelength Detuning (pm)
N
o
r
m
a
l
i
z
e
d
T
r
a
n
s
m
i
s
s
i
o
n
1.0
0.8
0.6
0.4
0.2
0
Figure 2.7: Scan of the cavity transmission of Sample I at an input power of 0.76 mW, with the in
stantaneous and averaged signals shown in blue and red, respectively. The dashed line indicated the
laser frequency detuning used to record the timedependent cavity transmission given in Fig. 2.6(d).
single detuning function,
f (∆)
_
∆
2
+(κ/2)
2
κκ
e
κ
3
i
∆
__
(∆+Ω
m
)
2
+
_
κ
2
_
2
__
(∆ Ω
m
)
2
+
_
κ
2
_
2
_
. (2.55)
where κ = κ
i
+κ
e
is the total photon decay rate of the loaded cavity. The right panel of Fig. 2.8(b)
shows a map of f (∆
th
) versus optical input power for the 90 µm diameter doubledisk cavity in
the heavily damped nitrogen environment. The data in Fig. 2.8(b), as expected, shows a linear
dependence of f (∆
th
) on input power, and is well described in the unresolved sideband regime [50]
by
f (∆
th
) =
2g
2
OM
P
i
ω
c
m
x
Γ
m
κ
3
i
=
_
2B
Γ
m
_
P
i
, (2.56)
where Γ
m
= 2.1 MHz is the bare mechanical damping rate of the ﬂapping mode. Fitting of eq.
(2.56) to the data in Fig. 2.8(b) yields a dynamic backaction parameter of B = 0.061 MHz/µW,
corresponding to an optomechanical coupling factor of g
OM
/2π = 33.8 0.4 GHz/nm, in good
agreement with the simulated result of 33 GHz/nm.
2.5.2 Vacuum
In order to eliminate the squeezeﬁlm damping of the nitrogen environment, measurements were
also performed in vacuum (P < 5 10
4
Torr). The signiﬁcantly reduced mechanical linewidth
in vacuum shows that the ﬂapping mode consists of a small cluster of modes (Fig. 2.6(b), bottom
45
I II
a
0.2
0.05
0.6
0.8
l
0.05
0.2
0.6
6 4 2 0 2 4
wavelength detunlng (pm)
I
f ¬ 8.53 MHz
f ¬ 9.63 MHz
N
o
r
m
a
l
l
z
e
d
t
r
a
n
s
m
l
s
s
l
o
n
8 l0 6
III
0.5 l l.5 2 0 200 400 600 800 0
vacuum alr
nput power (µw)
0
l0
20
30
40
50
60
70
80
T
h
r
e
s
h
o
l
d
d
e
t
u
n
l
n
g
f
u
n
c
t
l
o
n
f
(
∆
t
h
)
b
Prequency (MHz)
D
l
s
p
l
a
c
e
m
e
n
t
s
p
e
c
t
r
a
l
d
e
n
s
l
t
y
(
m
´
/
H
z
)
l0`"
l0´'
l0´"
l0´´
7 7.2 7.4 7.6 7.8 8 8.2 8.4 8.6 6.8
c
ln
cre
a
sln
g
o
p
t
l
c
a
l
p
o
w
e
r
l0
l00
l
nput power (µw)
0.l l l0
L
n
e
c
t
l
v
e
t
e
m
p
e
r
a
t
u
r
e
(
K
)
d
S
p
e
c
t
r
a
l
d
e
n
s
l
t
y
(
m
2
/
H
z
)
l0`"
l0´'
l0`'
l0`´
Prequency (MHz)
3 4 5 6 7 8
Figure 2.8: (a) Top panel: Normalized cavity transmission for Sample II in vacuum and P
i
= 11
µW. Blue and red traces show the instantaneous and lowpassﬁltered signals, respectively. Middle
panel: the transduction amplitude of the frequency component at 8.53 MHz and its higherorder
harmonics. Bottom panel: the transduction amplitude of the frequency component at 9.63 MHz and
its higherorder harmonics. (b) f (∆
th
) as a function of optical input power. Right panel: Sample
I in a 1 atm. nitrogen environment. Left panel: Sample II in vacuum (inset shows the minimum
achievable threshold (green arrow)). (c) Spectral intensity of the thermallydriven fundamental
ﬂapping mode at various input powers, recorded for Sample II in vacuum, with a laser detuning
of ∆ = −1.45(κ/2) (inset shows the displacement sensitivity at the highest input power with the
second optical attenuator removed), and (d) the corresponding effective temperature. In (d), the red
curve is a ﬁt to the data, the solid green (dashed black) curve is a theoretical curve obtained using
the estimated Bparameter from the left panel of (b) and the experimental (optimal) detuning of
∆ = −1.45(κ/2)
_
∆ = −(κ/2)/
5
_
.
46
panel). These modes are a mixture of the lowerlying azimuthal modes, coupled together due to
deviations in circularity of the undercut region and support pedestal.
Because of the extremely short roundtrip time of the cavity mode, the optical wave is sensitive
only to the variations of averaged disk spacing around the whole disk. As a result, the optome
chanical coupling for the fundamental ﬂapping mode, which has a ﬂapping amplitude uniformly
distributed around the disk perimeter, is maximum, but it is nearly zero for ﬂapping modes with
higherorder azimuthal mode numbers. However, due to the asymmetry in practical devices, the net
variations in the average disk spacing induced by the higherorder ﬂapping modes (with azimuthal
mode number 1) is not zero, and their thermal motion is visible in the transmission power spec
trum. In general, their optomechanical coupling is weak and does not provide efﬁcient dynamic
back action.
Measurements of the optical spring effect indicates that the optical ﬁeld renormalizes the cluster
of modes, with the lowestfrequency mode at 8.53 MHz transforming into the fundamental ﬂapping
mode with uniformly distributed displacement along the disk perimeter (the rest of the modes de
couple from the light ﬁeld). With an invacuum Q
M
= 4070 (Fig. 2.6(b), inset), the fundamental
ﬂapping mode has an extremely lowthreshold input power for regenerative oscillation. Figure 2.8(a)
shows a transmission spectrum when the laser is scanned across the cavity resonance. Three differ
ent regimes can be clearly seen: (I) transduction of thermal motion, (II) onset of opticallydriven
oscillation, and (III) optically damped motion. The onset of regenerative oscillation coincides with
a frequency shift in the fundamental ﬂapping mode to 9.63 MHz as shown in the bottom two panels
of Fig. 2.8(a). The left panel of Figure 2.8(b) shows a plot of the invacuum f (∆
th
) versus input
power, with a measured minimum threshold power of P
i
= 267 nW. Extrapolation of the experi
mental data using Eqs. (2.55) and (2.56) to the optimal detuning point shows a minimum threshold
power of only 40 nW.
2.6 Optomechanical cooling
In general, the optomechanical effect is governed by Eqs. (2.53) and (2.54). However, the op
tomechanical effect during mechanical cooling is well described by linear perturbation theory since
the thermal mechanical motion is signiﬁcantly suppressed. The intracavity ﬁeld can thus be ap
proximated as a(t) a
0
(t) +δa(t), where a
0
is the cavity ﬁeld in the absence of optomechanical
coupling and δa is the perturbation induced by the thermal mechanical motion. From Eq. (2.53),
47
they are found to satisfy the following equations:
da
0
dt
= (i∆
0
κ/2)a
0
+i
κ
e
A
in
, (2.57)
dδa
dt
= (i∆
0
κ/2)δa ig
OM
xa
0
. (2.58)
In the case of a continuouswave input, Eq. (2.57) gives a steadystate value given as:
a
0
=
i
κ
e
A
in
κ/2 i∆
0
, (2.59)
and Eq. (2.58) provides the spectral response for the perturbed ﬁeld amplitude,
δ¯ a(Ω) =
ig
OM
a
0
¯ x(Ω)
i(∆
0
+Ω) κ/2
, (2.60)
where δ¯ a(Ω) is the Fourier transform of δa(t) deﬁned as δ¯ a(Ω) =
+∞
∞
δa(t)e
iΩt
dt. Similarly, ¯ x(Ω)
is the Fourier transform of x(t).
The optical gradient force, F
o
=
g
OM
a
2
ω
0
, is given by
F
o
(t) =
g
OM
ω
0
_
a
0
2
+a
0
δa(t) +a
0
δa
(t)
¸
. (2.61)
The ﬁrst term is a static term which only affects the equilibrium position of the mechanical motion,
and can be removed simply by shifting the zeropoint of the mechanical displacement to the new
equilibrium position. Therefore, we neglect this term in the following discussion. The second and
third terms provide the dynamic optomechanical coupling. From Eq. (2.60), the gradient force is
given by the following equation in the frequency domain:
¯
F
o
(Ω) =
2g
2
OM
a
0
2
∆
0
¯ x(Ω)
ω
0
∆
2
0
Ω
2
+(κ/2)
2
+iκΩ
[(∆
0
+Ω)
2
+(κ/2)
2
] [(∆
0
Ω)
2
+(κ/2)
2
]
. (2.62)
As expected, the gradient force is linearly proportional to the thermal mechanical displacement.
Equation (2.54) can be solved easily in the frequency domain, which becomes
(Ω
2
m
Ω
2
iΓ
m
Ω)¯ x =
¯
F
T
m
x
+
¯
F
o
m
x
. (2.63)
Equation (2.63) together with (2.62) provides the simple form for the thermal mechanical displace
48
ment,
¯ x(Ω) =
¯
F
T
m
x
1
(Ω
m
)
2
Ω
2
iΓ
m
Ω
, (2.64)
where Ω
m
and Γ
m
are deﬁned as
(Ω
m
)
2
Ω
2
m
+
2g
2
OM
a
0
2
∆
0
m
x
ω
0
∆
2
0
Ω
2
+(κ/2)
2
[(∆
0
+Ω)
2
+(κ/2)
2
] [(∆
0
Ω)
2
+(κ/2)
2
]
Ω
2
m
+
2g
2
OM
a
0
2
∆
0
m
x
ω
0
∆
2
0
Ω
2
m
+(κ/2)
2
[(∆
0
+Ω
m
)
2
+(κ/2)
2
] [(∆
0
Ω
m
)
2
+(κ/2)
2
]
, (2.65)
Γ
m
Γ
m
2g
2
OM
a
0
2
κ∆
0
m
x
ω
0
1
[(∆
0
+Ω)
2
+(κ/2)
2
] [(∆
0
Ω)
2
+(κ/2)
2
]
Γ
m
2g
2
OM
a
0
2
κ∆
0
m
x
ω
0
1
[(∆
0
+Ω
m
)
2
+(κ/2)
2
] [(∆
0
Ω
m
)
2
+(κ/2)
2
]
. (2.66)
Equations (2.64)(2.66) show clearly that the primary effect of the optical gradient force on the
mechanical motion is primarily to change its mechanical frequency (the socalled optical spring
effect) and energy decay rate to the new values given by Eqs. (2.65) and (2.66). The efﬁciency of
optomechanical control is determined by the ﬁgure of merit g
2
OM
/m
x
. On the red detuned side, the
optical wave damps the thermal mechanical motion and thus increases the energy decay rate. At the
same time, the mechanical frequency is modiﬁed, decreasing with increased cavity energy in the
sidebandunresolved regime.
Using Eqs. (2.13) and (2.64), we ﬁnd that the spectral intensity of the thermal displacement is
given by a form similar to Eq. (2.14):
S
x
(Ω) =
2Γ
m
k
B
T/m
x
[(Ω
m
)
2
Ω
2
]
2
+(ΩΓ
m
)
2
, (2.67)
which has a maximum value S
x
(Ω
m
) =
2Γ
m
k
B
T
m
x
(Ω
m
Γ
m
)
2
. The variance of the thermal mechanical displace
ment is equal to the area under the spectrum,
(δx)
2
=
1
2π
+∞
∞
S
x
(Ω)dΩ =
k
B
TΓ
m
m
x
(Ω
m
)
2
Γ
m
. (2.68)
Cooling the mechanical motion reduces the spectral magnitude and the variance of thermal dis
placement.
The large mechanical ampliﬁcation of the doubledisk NOMS implies a correspondingly efﬁ
cient cooling of mechanical motion on the reddetuned side of the cavity resonance. As shown in
Fig. 2.8(c) for Sample II in vacuum, the spectral intensity of the fundamental ﬂapping mode de
49
creases dramatically with increased input power, accompanied by a signiﬁcant broadening of the
mechanical linewidth. Even for the strongest damping levels, the inset to Fig. 2.8(c) shows good
signal to noise for the transduced motion due to the high displacement sensitivity of the doubledisk
(7 10
17
m/Hz
1/2
, as limited by the background level).
A measure of the optical cooling can be determined from the integrated area under the displace
ment spectrum [56]. For a mechanical mode in thermal equilibrium, the effective temperature can
be inferred from the thermal mechanical energy using the equipartition theorem:
k
B
T
eff
= m
x
(Ω
m
)
2
(δx)
2
. (2.69)
The area under the displacement spectrum thus provides an accurate measure of the effective tem
perature. In practice, ﬂuctuations on the laser frequency detuning may cause the mechanical fre
quency and damping rate to ﬂuctuate over a certain small range [Eq. (2.65) and (2.66)], with a
probability density function of p(Ω
m
). As a result, the experimentally recorded displacement spec
trum is given by the averaged spectrum
S
x
(Ω) =
S
x
(Ω)p(Ω
m
)dΩ
m
, (2.70)
where S
x
(Ω) is given by Eq. (2.67) and we have assumed
p(Ω
m
)dΩ
m
= 1. The experimentally
measured spectral area is thus
1
2π
+∞
∞
S
x
(Ω)dΩ =
(δx)
2
p(Ω
m
)dΩ
m
(δx)
2
. (2.71)
Therefore, the integrated spectral area obtained from the experimental spectrum is the averaged
variance of thermal mechanical displacement, from which, according to the equipartition theorem,
we obtain the effective average temperature
k
B
T
eff
= m
x
(Ω
m
)
2
(δx)
2
, (2.72)
where Ω
m
Ω
m
p(Ω
m
)dΩ
m
is the center frequency of the measured displacement spectrum S
x
(Ω).
Compared with the room temperature, the effective temperature is thus given by
T
eff
T
0
=
(Ω
m
)
2
(δx)
2
Ω
2
m
(δx)
2
0
, (2.73)
50
where (δx)
2
0
is the displacement variance at room temperature, given by the spectral area at T
0
.
Figure 2.8(d) plots the inferred temperaure, T
eff
, which drops down to 12.5 K for a maximum
input power of P
i
= 11 µW (P
d
= 4.4 µW). In principle, the effective temperature is related to the
optical damping rate (Γ
m,opt
) through the relation T
0
/T
eff
= 1+Γ
m
/Γ
m,opt
, where T
0
= 300 K is the
bath temperature. In Fig. 2.8d the red curve is a ﬁt of the measured cooling curve using the relation
T
0
/T
eff
= 1 +αP
i
, whereas the green curve represents the expected cooling curve for the dynamic
backaction parameter (B = 0.032 MHz/µW) determined from the threshold plot in the right panel
of Fig. 2.8(b) and the experimental lasercavity detuning (∆ = 1.45(κ/2)). For comparison, we
have also plotted (dashed black line) the theoretical cooling curve in the case of optimal lasercavity
detuning (∆ = (κ/2)/
5). The difference between the two theoretical curves and the measured
data, along with the limited range of optical input power studied, can largely be attributed to issues
associated with the limited bandwidth and range of our current cavity locking scheme (a problem
exacerbated by the very large transduction of even the Brownian motion of the disks). As the
dashed black curve indicates, technical improvements in the cavity locking position and stability
should enable temperature compression factors of 20 dB for less than 1 µW of dropped power.
2.7 Discussion
The large dynamic backaction of the doubledisk cavity, primarily a result of the large perphoton
force and small motional mass of the structure, opens up several areas of application outside the
realm of more conventional ultrahighQ cavity geometries. This can be seen by considering not
only the efﬁciency of the cooling/ampliﬁcation process, but also the maximum rate of effective
cooling/ampliﬁcation, the scale of which is set by the optical cavity decay rate [57, 58]. In the
doubledisk cavities presented here, the dynamic backaction parameter is B 0.06 MHz/µW for a
cavity decay rate of κ/2π 100 MHz. The combination allows for higher mechanical frequencies
of operation, where the bare damping is expected to scale with frequency, and makes possible enor
mous temperature compression ratios. A quantum mechanical analysis of the optical selfcooling
process [57, 58], indicates that the sideband resolved regime (κ
32Ω
m
) is necessary to reduce
the phonon occupancy below unity. Having already achieved optical Qfactors in excess of 10
6
, and
planar silica microdisks having already been demonstrated with Q>10
7
[59], we expect that further
optimization of the doubledisk NOMS will be able to extend its operation well into the sideband
resolved regime. The combination of large dynamic backaction parameter and large maximum
51
ampliﬁcation rate also present intriguing possibilites for sensitive, high temporal resolution force
detection [60], particularly in heavily damped environments such as ﬂuids for biological applica
tions [46, 47]. Other application areas enabled by the chipscale format of these devices include
tunable photonics [42–44], optical wavelength conversion [61], and RFoveroptical communica
tion.
52
Chapter 3
Coherent Mechanical Mode Mixing in
Optomechanical Nanocavities
3.1 Introduction
The coherent mixing of multiple excitation pathways provides the underlying mechanism for many
physical phenomena. Wellknown examples include the Fano resonance [62] and electromagnet
ically induced transparency (EIT) [63], arising from the interference between excitations of dis
crete states and/or a continuum background. In the past few decades, Fanolike or EITlike reso
nances have been discovered in a variety of physical systems, such as electron transport in quan
tum wells/dots [64, 65], phonon interactions in solids [66, 67], inversionfree lasers [68, 69], cou
pled photonic microcavities [70–73], and plasmonic metamaterials [74]. Here we report a new
class of coherent excitation mixing which appears in the mechanical degree of freedom of nano
optomechanical systems (NOMS). We use two canonical systems, coupled microdisks and cou
pled photoniccrystal nanobeams, to show that the large optical stiffening introduced by the optical
gradient force actuates signiﬁcant coherent mixing of mechanical excitations, not only leading to
renormalization of the mechanical modes, but also producing Fanolike and EITlike optomechan
ical interference, both of which are fully tunable by optical means. The demonstrated phenomena
introduce the possibility for classical/quantum information processing via optomechanical systems,
providing an onchip platform for tunable optical buffering, storage, and photonicphononic quan
tum state transfer. This work was initially presented in Ref. 75.
Optical forces within micromechanical systems have attracted considerable interest of late due
to the demonstration of alloptical ampliﬁcation and selfcooling of mesoscopic mechanical res
onators [35–39]. This technique for sensing and control of mechanical motion relies on the radia
53
tion pressure forces that build up in a mechanically compliant, highFinesse optical cavity, resulting
in strong dynamical backaction between the cavity ﬁeld and mechanical motion. More recently
[44, 45, 48, 76–78], it has been realized that guided wave nanostructures can also be used to gener
ate extremely large perphoton optical forces via the gradient optical force [79]. The combination of
tailorable mechanical geometry, small motional mass, and large perphoton force in such nanostruc
tures results in a regime of operation in which the dynamic response of the coupled optomechanical
system can signiﬁcantly differ from that of the bare mechanical structure. In particular, the me
chanical motion can be renormalized by the optical spring effect [29, 33, 45, 55, 80, 81], creating a
highly anistropic, intensitydependent effective elastic modulus of the optomechanical structure.
3.2 Zipper cavity and doubledisk design, fabrication, and optical char
acterization
We have focused on two speciﬁc implementations of nanoscale cavity optomechanical systems,
shown in Fig. 3.1, in which dynamical backaction effects are particularly strong. The ﬁrst system
consists of two patterned nanobeams in the nearﬁeld of each other, forming what has been termed a
zipper cavity [45, 82]. In this cavity structure the patterning of the nanobeams localizes light through
Braggscattering, resulting in a series of high Finesse (F 3 10
4
), nearinfrared (λ 1550 nm)
optical supermodes of the beam pair. Clamping to the substrate at either end of the suspended
beams results in a fundamental inplane mechanical beam resonance of frequency 8 MHz. The
second cavity optomechanical system is based upon the whisperinggallery microdisk optical cavity
structure presented in Chapter 2 and Ref. 48. By creating a pair of microdisks, one on top of the
other with a nanoscale gap in between, strong optical gradient forces may be generated between
the microdisks while maintaining the beneﬁts of the lowloss, highQ (Q 10
6
) character of the
whisperinggallery cavity. As shown schematically in Fig. 3.1(a), the doubledisk structure [48]
is supported and pinned at its center, allowing the perimeter of the disks to vibrate in myriad of
different ways.
The zipper cavity is formed from a thinﬁlm (400 nm) of tensilestressed, stoichiometric Si
3
N
4
deposited by lowpressure chemical vapor deposition on a silicon substrate. Electron beamlithography,
followed by a series of plasma and wet chemical etches, are used to form the released nanobeam
structure. The doubledisk structure is formed from a 158 nm sacriﬁcal amorphous silicon layer
sandwiched in between two 340 nm thick silica glass layers, all of which are deposited via plasma
54
b
a
c
a
d
f
1 µm
w
s
1 µm
a
e
Figure 3.1: (a) Schematic and (b) zoomedin scanning electron microscopic (SEM) image of the
doubledisk NOMS. (c) FEMsimulated electric ﬁeld intensity of a transverseelectric (TE) polar
ized, bonded (even parity) whisperinggallery supermode between the two microdisks (shown in
crosssection and for resonance wavelength λ
c
1550 nm). The doubledisk bonded supermode
has an optomechanical coupling coefﬁcient of g
OM
/2π 33 GHz/nm. The device studied here has
a measured resonance wavelength of λ
c
= 1538 nm and an intrinsic and loaded quality (Q) factor
of 1.07 10
6
and 0.7 10
6
, respectively. (d) Schematic, (e) SEM image, and (f) FEMsimulated
bonded (even parity) optical supermode of the zipper cavity. The zipper cavity bonded supermode
has an optomechanical coupling coefﬁcient of g
OM
/2π 68 GHz/nm, a measured resonance wave
length of λ
c
=1545 nm, and an intrinsic and loaded Qfactor of 3.0 10
4
and 2.8 10
4
, respectively.
Additional details for both devices are in Refs. 45, 48.
55
enhanced chemical vapor deposition. A high temperature (1050 K) thermal anneal is used to im
prove the optical quality of the asdeposited silica layers. The microdisk pattern was fabricated by
reactive ion etching, and the sandwiched αSi layer was undercut by 6 µm from the disk edge using
a sulfur hexaﬂuoride dry release etch. This etch simultaneously undercuts the silicon substrate to
form the underlying silicon pedestal. The ﬁnal airgap between the silica disks size is measured to
be 138 nm due to shrinkage of the amorphous silicon layer during annealing.
Aﬁbertaper optical coupling technique is used to incouple and outcouple light from the zipper
and doubledisk cavities. The ﬁber taper, with extremely lowloss (88% transmission efﬁciency),
is put in contact with the substrate near the cavities in order to mechanically anchor it during all
measurements (thus avoiding powerdependent movement of the taper due to thermal and/or optical
forces). An optical ﬁber polarization controller, consisting of a series of circular loops of ﬁber, is
used to selectively excite the transverseelectric polarized optical modes of both cavities.
RF spectra are measured by direct detection of the optical power transmitted through the cavities
using a 125 MHz bandwidth photoreceiver (noiseequivalentpower NEP= 2.5 pW/Hz
1/2
from 0
10 MHz and 22.5 pW/Hz
1/2
from 10200 MHz, responsivity R = 1 A/W, transimpedance gain
G = 4 10
4
V/A) and a highspeed oscilloscope (2 Gs/s sampling rate and 1 GHz bandwidth). A
pair of “dueling” calibrated optical attenuators are used before and after the cavities in order to vary
the input power to the cavity while keeping the detected optical power level constant. The measured
electrical noise ﬂoor is set by the circuit noise of the photodetector for the optical power levels
considered in this work, corresponding to 125 dBm/Hz near 10 MHz.
3.3 Theory of optomechanical effects in the presence of mode mixing
Of particular interest in both the zipper and doubledisk systems are two types of motion: the differ
ential motion of the nanobeams or disks, in which the changing gap between the elements creates
a large dispersive shift in the internally propagating cavity light ﬁeld; and the common motion, in
which both nanobeams or disks move together, and the gap remains approximately constant, result
ing in mechanical motion that is decoupled fromthe light ﬁeld. Due to the strong lightﬁeld coupling
and dynamical backaction of the differential mode, and the correspondingly weak coupling of the
common mode, we term these two motional states opticallybright and opticallydark, respectively.
The theory for gradientforce optomechanical systems in which there is coupling between these two
types of mechanical excitations is presented in the next sections.
56
3.3.1 Intracavity eld in the presence of optomechanical coupling
In the presence of optomechanical coupling, the optical ﬁeld inside the cavity satisﬁes the following
equation:
da
dt
= (i∆
0
Γ
t
/2 ig
om
x
b
)a+i
_
Γ
e
A
in
, (3.1)
where a is the optical ﬁeld of the cavity mode, normalized such that U = a
2
represents the mode
energy, and A
in
is the input optical wave, normalized such that P
in
= A
in
2
represents the input
power. Γ
t
is the photon decay rate for the loaded cavity and Γ
e
is the photon escape rate associated
with the external coupling. ∆
0
=ω ω
0
is the frequency detuning from the input wave to the cavity
resonance. g
om
is the optomechanical coupling coefﬁcient associated with the optically bright mode,
with a mechanical displacement given by x
b
. In Eq. (3.1), we have neglected the optomechanical
coupling to the optically dark mode because of its negligible magnitude.
Well below the threshold of mechanical oscillation, the mechanical motion is generally small,
and its impact on the intracavity optical ﬁeld can be treated as a small perturbation. As a result, the
intracavity ﬁeld can be written as a(t) a
0
(t) +δa(t), where a
0
is the cavity ﬁeld in the absence of
optomechanical coupling and δa is the perturbation induced by the mechanical motion. They satisfy
the following two equations:
da
0
dt
= (i∆
0
Γ
t
/2)a
0
+i
_
Γ
e
A
in
, (3.2)
dδa
dt
= (i∆
0
Γ
t
/2)δa ig
om
x
b
a
0
. (3.3)
In the case of a continuouswave input, Eq. (3.2) leads to a steady state given by
a
0
=
i
Γ
e
A
in
Γ
t
/2 i∆
0
, (3.4)
and Eq. (3.3) provides a spectral response for the perturbed ﬁeld amplitude of
δ¯ a(Ω) =
ig
om
a
0
¯ x
b
(Ω)
i(∆
0
+Ω) Γ
t
/2
, (3.5)
where δ¯ a(Ω) is the Fourier transform of δa(t) deﬁned as δ¯ a(Ω) =
+∞
∞
δa(t)e
iΩt
dt. Similarly, ¯ x
b
(Ω)
is the Fourier transform of x
b
(t).
57
3.3.2 The power spectral density of the cavity transmission
From the discussion in the previous section, the transmitted optical power from the cavity is given
by
P
T
=
¸
¸
¸A
in
+i
_
Γ
e
a
¸
¸
¸
2
A
0
2
+i
_
Γ
e
(A
0
δa A
0
δa
), (3.6)
where A
0
is the steadystate cavity transmission in the absence of optomechanical coupling. It is
given by
A
0
= A
in
(Γ
0
Γ
e
)/2 i∆
0
Γ
t
/2 i∆
0
, (3.7)
where Γ
0
is the photon decay rate of the intrinsic cavity. It is easy to show that the averaged cavity
transmission is given by P
T
= A
0
2
, as expected. By using Eqs. (3.5), (3.6), and (3.7), we ﬁnd the
power ﬂuctuations, δP
T
(t) P
T
(t) P
T
, are given in the frequency domain by
δ
¯
P
T
(Ω) =
iΓ
e
P
in
g
om
¯ x
b
(Ω)
(Γ
t
/2)
2
+∆
2
0
_
(Γ
0
Γ
e
)/2+i∆
0
Γ
t
/2 i(∆
0
+Ω)
(Γ
0
Γ
e
)/2 i∆
0
Γ
t
/2+i(∆
0
Ω)
_
, (3.8)
where δ
¯
P
T
(Ω) is the Fourier transform of δP
T
(t). By using Eq. (3.8), we obtain a power spectral
density (PSD) for the cavity transmission of
S
P
(Ω) = g
2
om
P
2
in
S
x
b
(Ω)H(Ω), (3.9)
where S
x
b
(Ω) is the spectral intensity of the mechanical displacement for the optically bright mode
which will be discussed in detail in the following sections. H(Ω) is the cavity transfer function
deﬁned as
H(Ω)
Γ
2
e
_
∆
2
0
+(Γ
t
/2)
2
¸
2
4∆
2
0
(Γ
2
0
+Ω
2
)
[(∆
0
+Ω)
2
+(Γ
t
/2)
2
] [(∆
0
Ω)
2
+(Γ
t
/2)
2
]
. (3.10)
In general, when compared with S
x
b
(Ω), H(Ω) is a slowly varying function of Ω and can be well
approximated by its value at the mechanical resonance: H(Ω) H(Ω
mb
). Clearly then, the power
spectral density of the cavity transmission is linearly proportional to the spectral intensity of the
mechanical displacement of the optically bright mode.
58
3.3.3 The mechanical response with multiple excitation pathways
When the optically bright mode is coupled to an optically dark mode, the Hamiltonian for the
coupled mechanical system is given by the general form:
H
m
=
p
2
b
2m
b
+
1
2
k
b
x
2
b
+
p
2
d
2m
d
+
1
2
k
d
x
2
d
+κx
b
x
d
, (3.11)
where x
j
, p
j
, k
j
, and m
j
( j = b, d) are the mechanical displacement, kinetic momentum, the spring
constant, and the effective motional mass for the j
th
mechanical mode, respectively, and κ repre
sents the mechanical coupling between the bright and dark modes. The subscripts b and d denote
the optically bright and optically dark modes, respectively. With this system Hamiltonian, including
the optical gradient force on the optically bright mode and counting in the mechanical dissipation
induced by the thermal mechanical reservoir, we obtain the equations of motion for the two me
chanical modes:
d
2
x
b
dt
2
+Γ
mb
dx
b
dt
+Ω
2
mb
x
b
+
κ
m
b
x
d
=
F
b
m
b
+
F
o
m
b
, (3.12)
d
2
x
d
dt
2
+Γ
md
dx
d
dt
+Ω
2
md
x
d
+
κ
m
d
x
b
=
F
d
m
d
, (3.13)
where Ω
2
mj
k
j
m
j
is the mechanical frequency for the j
th
mode. F
j
( j = b, d) represents the Langevin
forces from the thermal reservoir actuating the Brownian motion, with the following statistical prop
erties in the frequency domain:
¯
F
i
(Ω
u
)
¯
F
j
(Ω
v
)= 2m
i
Γ
mi
k
B
Tδ
i j
2πδ(Ω
u
Ω
v
), (3.14)
where i, j = b, d, T is the temperature and k
B
is the Boltzmann constant.
¯
F
i
(Ω) is the Fourier
transform of F
i
(t).
In Eq. (3.12), F
o
=
g
om
a
2
ω
0
represents the optical gradient force. From the previous section, we
ﬁnd that it is given by
F
o
(t) =
g
om
ω
0
_
a
0
2
+a
0
δa(t) +a
0
δa
(t)
¸
. (3.15)
The ﬁrst term is a static term which only changes the equilibrium position of the mechanical mo
tion. It can be removed simply by shifting the mechanical displacement to be centered at the new
equilibrium position. Therefore, we neglect this term in the following discussion. The second and
59
third terms provide the dynamic optomechanical coupling. From Eq. (3.5), the gradient force is
found to be given in the frequency domain by
¯
F
o
(Ω) f
o
(Ω)¯ x
b
(Ω) =
2g
2
om
a
0
2
∆
0
¯ x
b
(Ω)
ω
0
∆
2
0
Ω
2
+(Γ
t
/2)
2
+iΓ
t
Ω
[(∆
0
+Ω)
2
+(Γ
t
/2)
2
] [(∆
0
Ω)
2
+(Γ
t
/2)
2
]
, (3.16)
which is linearly proportional to the mechanical displacement of the optically bright mode.
Equations (3.12) and (3.13) can be solved easily in the frequency domain, in which the two
equations become
L
b
(Ω)¯ x
b
+
κ
m
b
¯ x
d
=
¯
F
b
m
b
+
¯
F
o
m
b
, (3.17)
L
d
(Ω)¯ x
d
+
κ
m
d
¯ x
b
=
¯
F
d
m
d
, (3.18)
where L
j
(Ω) Ω
2
mj
Ω
2
iΓ
mj
Ω ( j = b, d). Substituting Eq. (3.16) into Eq. (3.17), we ﬁnd that
Eq. (3.17) can be written in the simple form,
L
b
(Ω)¯ x
b
+
κ
m
b
¯ x
d
=
¯
F
b
m
b
, (3.19)
where L
b
(Ω) is now deﬁned with a new mechanical frequency Ω
mb
and energy decay rate Γ
mb
as
L
b
(Ω) = Ω
2
mb
Ω
2
iΓ
mb
Ω
f
o
(Ω)
m
b
(Ω
mb
)
2
Ω
2
iΓ
mb
Ω, (3.20)
and the new Ω
mS
and Γ
mS
are given by
(Ω
mb
)
2
Ω
2
mb
+
2g
2
om
a
0
2
∆
0
m
b
ω
0
∆
2
0
Ω
2
+(Γ
t
/2)
2
[(∆
0
+Ω)
2
+(Γ
t
/2)
2
] [(∆
0
Ω)
2
+(Γ
t
/2)
2
]
Ω
2
mb
+
2g
2
om
a
0
2
∆
0
m
b
ω
0
∆
2
0
Ω
2
mb
+(Γ
t
/2)
2
[(∆
0
+Ω
mb
)
2
+(Γ
t
/2)
2
] [(∆
0
Ω
mb
)
2
+(Γ
t
/2)
2
]
, (3.21)
Γ
mb
Γ
mb
2g
2
om
a
0
2
Γ
t
∆
0
m
b
ω
0
1
[(∆
0
+Ω)
2
+(Γ
t
/2)
2
] [(∆
0
Ω)
2
+(Γ
t
/2)
2
]
Γ
mb
2g
2
om
a
0
2
Γ
t
∆
0
m
b
ω
0
1
[(∆
0
+Ω
mb
)
2
+(Γ
t
/2)
2
] [(∆
0
Ω
mb
)
2
+(Γ
t
/2)
2
]
. (3.22)
Clearly, the effect of the optical gradient force on the optically bright mode is primarily to change
its mechanical frequency (the optical spring effect) and energy decay rate (mechanical ampliﬁcation
or damping).
60
Equations (3.18) and (3.19) can be solved easily to obtain the solution for the optically bright
mode,
¯ x
b
(Ω) =
¯
F
b
(Ω)
m
b
L
d
(Ω)
κ
m
b
¯
F
d
(Ω)
m
d
L
b
(Ω)L
d
(Ω) η
4
, (3.23)
where η
4
κ
2
m
b
m
d
represents the mechanical coupling coefﬁcient. By using Eq. (3.14) and (3.23),
we obtain the spectral intensity of the mechanical displacement for the optically bright mode,
S
x
b
(Ω) =
2k
B
T
m
b
η
4
Γ
md
+Γ
mb
L
d
(Ω)
2
L
b
(Ω)L
d
(Ω) η
4
2
, (3.24)
where L
b
(Ω) is given by Eq. (3.20). The mechanical response given by Eq. (3.24) is very similar to
the atomic response in EIT.
3.3.4 The mechanical response with external optical excitation
The previous section focuses on the case in which the mechanical excitations are primarily intro
duced by the thermal perturbations from the environmental reservoir. However, the mechanical
motion can be excited more intensely through the optical force by modulating the incident optical
wave. In this case, the input optical wave is composed of an intense CW beam together with a small
modulation: A
in
= A
in0
+δA(t). As a result, Eq. (3.3) now becomes
dδa
dt
= (i∆
0
Γ
t
/2)δa ig
om
x
b
a
0
+i
_
Γ
e
δA. (3.25)
This equation leads to the intracavity ﬁeld modulation given in the frequency domain as:
δ¯ a(Ω) =
ig
om
a
0
¯ x
b
(Ω) i
Γ
e
δ
¯
A(Ω)
i(∆
0
+Ω) Γ
t
/2
, (3.26)
where δ
¯
A(Ω) is the Fourier transform of δA(t). By use of this solution together with Eq. (3.15), the
gradient force now becomes
¯
F
o
(Ω) = f
o
(Ω)¯ x
b
(Ω) +
¯
F
e
(Ω), (3.27)
where f
o
(Ω) is given by Eq. (3.16) and
¯
F
e
(Ω) represents the force component introduced by the
input modulation. It is given by the following form:
¯
F
e
(Ω) =
i
Γ
e
g
om
ω
0
_
a
0
δ
¯
A(Ω)
i(∆+Ω) Γ
t
/2
+
a
0
δ
¯
A
( Ω)
i(∆ Ω) +Γ
t
/2
_
. (3.28)
61
In particular, in the sidebandunresolved regime, Eq. (3.28) can be well approximated by
¯
F
e
(Ω)
i
Γ
e
g
om
ω
0
(i∆ Γ
t
/2)
_
a
0
δ
¯
A(Ω) +a
0
δ
¯
A
( Ω)
_
. (3.29)
In the case that the mechanical excitation is dominated by the external optical modulation, the
thermal excitation from the reservoir is negligible and Eqs. (3.12) and (3.13) become
d
2
x
b
dt
2
+Γ
mb
dx
b
dt
+Ω
2
mb
x
b
+
κ
m
b
x
d
=
F
o
m
b
, (3.30)
d
2
x
d
dt
2
+Γ
md
dx
d
dt
+Ω
2
md
x
d
+
κ
m
d
x
b
= 0. (3.31)
Using Eqs. (3.26) and (3.27), following a similar procedure as the previous section, we ﬁnd that the
mechanical displacement for the optically bright mode is now given by
¯ x
b
(Ω) =
¯
F
e
(Ω)
m
b
L
d
(Ω)
L
b
(Ω)L
d
(Ω) η
4
, (3.32)
where L
b
(Ω) and L
d
(Ω) are given in the previous section. Clearly, the mechanical response given
in Eq. (3.32) is directly analogous to the atomic response in EIT systems [83].
3.4 Mechanical mode renormalization in zipper cavities
We begin with an analysis of the zipper cavity, in which the strong opticallyinduced rigidity as
sociated with differential inplane motion of the nanobeams results in a dressing of the mechanical
motion by the light ﬁeld. Optical excitation provides both a means to transduce mechanical motion
(which is imparted on the transmitted light ﬁeld through phase and intensity modulation) and to
apply an opticalintensitydependent mechanical rigidity via the strong optical gradient force. By
ﬁtting a Lorentzian to the two lowestorder inplane mechanical resonances in the radiofrequency
(RF) optical transmission spectrum, we display in Fig. 3.2(a) and (b) the resonance frequency and
resonance linewidth, respectively, of the two coupled mechanical modes of the nanobeam pair
as a function of lasercavity detuning. At large detuning (low intracavity photon number) the
nanobeams’ motion is transduced without inducing signiﬁcant optical rigidity, and the measured
mechanical resonances are split by 200 kHz, with similar linewidths (damping) and transduced
amplitudes (Fig. 3.2(c)). As the laser is tuned into resonance from the blueside of the cavity, and
the intracavity photon number increases (to 7000), the higher frequency resonance is seen to
62
7.4
7.6
7.8
8
8.2
8.4
2 1.5 1 0.5 0 0.5 1 1.5 2
0
0.2
0.25
0.05
0.1
0.15
Normalized Detuning (∆
ο
/Γ
t
)
M
e
c
h
a
n
i
c
a
l
L
i
n
e
w
i
d
t
h
(
M
H
z
)
M
e
c
h
a
n
i
c
a
l
F
r
e
q
u
e
n
c
y
(
M
H
z
)
a
b
N
photon
NNN
photon photon photon photon
e
f
7.2 7.6 8 8.4 8.8
70
80
60
P
o
w
e
r
S
p
e
c
t
r
a
l
D
e
n
s
i
t
y
(
d
B
m
/
H
z
)
Frequency (MHz)
100
110
90
P
o
w
e
r
S
p
e
c
t
r
a
l
D
e
n
s
i
t
y
(
d
B
m
/
H
z
)
c
d
Figure 3.2: (a) Mechanical frequency and (b) linewidth of the fundamental inplane mechanical
resonances of the zipper cavity’s coupled nanobeams as a function of laser frequency detuning. The
input power for these measurements is 127 µW, corresponding to a maximum cavity photon number
of 7000 on resonance. The circles show the experimental data and the solid curves correspond to
a ﬁt to the data using Eq. (3.37). Opticallytransduced RF spectrum at a lasercavity detuning of (c)
∆
0
/Γ
t
=2.1 and (d) ∆
0
/Γ
t
=0.32. The two nanobeams vibrate independently when the lasercavity
detuning is large, but are renormalized to the cooperative (e) differential and (f) common motions
near resonance.
signiﬁcantly increase in frequency while the lower frequency mode tunes to the average of the in
dependent beam frequencies with its transduced amplitude signiﬁcantly weaker. The linewidth of
the high frequency resonance also tends to increase, while that of the lower frequency mode drops.
Tuning from the redside of the cavity resonance reverses the sign of the frequency shifts and the
roles of the high and low frequency modes.
Aqualitative understanding of the lightinduced tuning and damping of the zipper cavity nanobeam
motion emerges if one considers the effects of squeezeﬁlm damping [84]. Squeezeﬁlm effects, a
result of trapped gas inbetween the beams (measurements were performed in 1 atm. of nitrogen),
tend to strongly dampen differential motion of the beams and should be negligible for common mo
tion of the beams. Similarly, the optical gradient force acts most strongly on the differential beam
motion and negligibly on the commonmode motion. The sign of the resulting optical spring is pos
itive for blue detuning and negative for red detuning from the cavity resonance. Putting all of this
63
together, a consistent picture emerges from the data in Fig. 3.2 in which the nanobeams start out at
large detuning moving independently with similar damping (the frequency splitting of 200 kHz is
attributable to fabrication assymetries in the beams). As the detuning is reduced, and approaches the
cavity halflinewidth, the motion of the nanobeams is dressed by the internal cavity ﬁeld into differ
ential motion with a large additional optical spring constant (either positive or negative) and large
squeezeﬁlm damping component, and common motion with reduced squeezeﬁlm damping and
minimal coupling to the light ﬁeld. Due to the strong lightﬁeld coupling of the differential mode
and the correspondingly weak coupling of the common mode, we term these dressed motional states
opticallybright and opticallydark, respectively.
In general, the motion of individual disks or nanobeams satisﬁes the following equations:
d
2
x
1
dt
2
+Γ
m1
dx
1
dt
+Ω
2
m1
x
1
=
F
1
m
1
+
F
o
m
1
+
F
q
m
1
, (3.33)
d
2
x
2
dt
2
+Γ
m2
dx
2
dt
+Ω
2
m2
x
2
=
F
2
m
2
F
o
m
2
F
q
m
2
, (3.34)
where F
q
is the viscous force from the squeeze ﬁlm damping, and m
j
, x
j
, Ω
mj
, Γ
mj
, F
j
( j = 1, 2)
are the effective mass, the mechanical displacement, resonance frequency, damping rate, and the
Langevin force for individual disks (or beams), respectively.
The optically bright mechanical mode corresponds to the differential motion of the two disks/beams,
with a mechanical displacement given by x
b
x
1
x
2
. By transferring Eqs. (3.33) and (3.34) into
the frequency domain, it is easy to ﬁnd that the mechanical displacement of the optically bright
mode is given by
¯ x
b
(Ω) =
¯
F
1
(Ω)
m
1
L
1
(Ω)
¯
F
2
(Ω)
m
2
L
2
(Ω)
+
_
1
m
1
L
1
(Ω)
+
1
m
2
L
2
(Ω)
_
_
¯
F
q
(Ω) +
¯
F
o
(Ω)
_
, (3.35)
where L
j
(Ω) = Ω
2
mj
Ω
2
iΓ
mj
Ω ( j = 1, 2). The squeezeﬁlm effect is produced by the pressure
differential between the gap and the outer region introduced by the differential mechanical motion,
and thus has a magnitude linearly proportional to the differential displacement. In general, it can be
described by
¯
F
q
(Ω) = f
q
(Ω)¯ x
b
(Ω), where f
q
(Ω) represents the spectral response of the squeeze gas
ﬁlm [84]. Using this form together with Eq. (3.16) in Eq. (3.35), we obtain the spectral intensity of
64
the optically bright mode displacement,
S
x
b
(Ω) =
2k
B
T
_
Γ
m1
m
1
L
2
(Ω)
2
+
Γ
m2
m
2
L
1
(Ω)
2
_
¸
¸
¸L
1
(Ω)L
2
(Ω) [ f
o
(Ω) + f
q
(Ω)]
_
L
1
(Ω)
m
2
+
L
2
(Ω)
m
1
_¸
¸
¸
2
. (3.36)
As the squeezeﬁlm effect primarily damps the differential motion, its spectral response can be
approximated as f
q
(Ω) iα
q
Ω. Moreover, since the two disks or nanobeams generally have only
slight asymmetry due to fabrication imperfections, they generally have quite close effective masses
and energy damping rates: m
1
m
2
= 2m
b
and Γ
m1
Γ
m2
Γ
m
, where we have used the fact
that the effective motional mass of the differential motion is given by m
b
= m
1
m
2
/(m
1
+m
2
). As a
result, Eq. (3.36) can be well approximated by
S
x
b
(Ω)
k
B
TΓ
m
m
b
L
1
(Ω)
2
+L
2
(Ω)
2
¸
¸
L
1
(Ω)L
2
(Ω)
1
2
[ f
o
(Ω)/m
b
+iΓ
q
Ω] [L
1
(Ω) +L
2
(Ω)]
¸
¸
2
, (3.37)
where Γ
q
α
q
/m
b
represents the damping rate introduced by the squeeze gas ﬁlm, and the spectral
response of the gradient force f
o
(Ω) is given by Eq. (3.16).
The intrinsic mechanical frequencies of 7.790 and 7.995 MHz for the two individual nanobeams
are measured from the experimental recorded PSD with a large lasercavity detuning. The optome
chanical coupling coefﬁcient is 68 GHz/nm and the effective mass is 10.75 pg for the fundamental
differential mode, both obtained from FEM simulations (note that these values are different than
those quoted in Ref. 45 due to the different deﬁnition of mode amplitude for x
b
). The intrinsic and
loaded optical Q factors are 3.0 10
4
and 2.8 10
4
, respectively, obtained from optical charac
terization of the cavity resonance. By using these values in Eqs. (3.37) and (3.16), we can easily
ﬁnd the mechanical frequencies and linewidths for the two renormalized modes, where we treat
the intrinsic mechanical damping rate Γ
m
and the squeezeﬁlminduced damping rate Γ
q
as ﬁtting
parameters. As shown in Fig. 3.2, this theoretical model provides an accurate description of the
mechanical mode renormalization, with a ﬁtted intrinsic mechanical and squeezeﬁlm damping rate
of 0.03 and 0.2 MHz, respectively.
Similarly, we can obtain the spectral intensity of x
d
x
1
+x
2
for the opticallydark mechanical
65
mode, which is given by the following form:
S
x
d
(Ω) = 2k
B
T
Γ
m2
m
2
¸
¸
¸L
1
(Ω)
2
m
1
[ f
o
(Ω) + f
q
(Ω)]
¸
¸
¸
2
+
Γ
m1
m
1
¸
¸
¸L
2
(Ω)
2
m
2
[ f
o
(Ω) + f
q
(Ω)]
¸
¸
¸
2
¸
¸
¸L
1
(Ω)L
2
(Ω) [ f
o
(Ω) + f
q
(Ω)]
_
L
1
(Ω)
m
2
+
L
2
(Ω)
m
1
_¸
¸
¸
2
. (3.38)
Similar to the opticallybright mode, with m
1
m
2
= 2m
b
and Γ
m1
Γ
m2
Γ
m
, Eq. (3.38) can be
well approximated by
S
x
d
(Ω)
k
B
TΓ
m
m
d
L
1
(Ω) h(Ω)
2
+L
2
(Ω) h(Ω)
2
¸
¸
L
1
(Ω)L
2
(Ω)
1
2
h(Ω)[L
1
(Ω) +L
2
(Ω)]
¸
¸
2
, (3.39)
where m
d
= m/2 is the effective mass of the common mode and h(Ω) [ f
o
(Ω)/m
b
+iΓ
q
Ω] repre
sents the total spectral response of the optical gradient force and squeeze ﬁlm damping. In partic
ular, when the opticalspringinduced frequency shift is much larger than the intrinsic mechanical
frequency splitting, the spectral intensities of these two modes reduce to
S
x
b
(Ω)
2k
B
TΓ
m
/m
b
L
o
(Ω) h(Ω)
2
, S
x
d
(Ω)
2k
B
TΓ
m
/m
d
L
o
(Ω)
2
(3.40)
where L
0
(Ω) = (Ω
m1
+Ω
m2
)
2
/4 Ω
2
iΓ
m
Ω. Equation (3.40) indicates that the optically bright
and dark modes reduce to pure differential and common modes, respectively.
3.5 Coherent mechanical mode mixing in doubledisks
A similar opticallyinduced renormalization mechanism applies to the doubledisk cavity structure
shown in Fig. 3.1(ac). In this case, the large optical spring effect for the differential motion of
the two microdisks excites another, more intriguing form of coherent optomechanical mixing with
the optically dark common mode of the disks. Unlike in the zipper cavity, FEM modeling of the
mechanics of the doubledisk structure indicates a signiﬁcant frequency splitting between the differ
ential and common modes of motion of the double disk (shown in Fig. 3.3(b) and (c)), primarily due
to the difference in the extent of the undercut between the disk layers and the extent of the central
pedestal which pins the two disk layers. The result is that the differential, or “ﬂapping” motion,
of the undercut disk region has a lower frequency of 7.95 MHz, whereas the common motion of
the disks results in a higher frequency (14.2 MHz) “breathing” motion of the entire doubledisk
structure.
66
The RFspectrum of the transmitted optical intensity through a doubledisk cavity, measured
using the same ﬁber probing technique as for the zipper cavity, is shown in Fig. 3.2(c) versus laser
cavity detuning. For the largest detuning (in which the optical spring is negligible) the spectrum
shows a broad (2.1 MHz) resonance at 8.3 MHz and a much narrower (0.11 MHz) resonance at
13.6 MHz, in good corresponce with the expected frequencies of the ﬂapping and breathing modes,
respectively. The difference in damping between the two resonances can be attributed to the strong
squeezeﬁlm damping of the differential ﬂapping motion of the disks. As shown in Fig. 3.2(c), the
ﬂapping mode can be tuned in frequency via the optical spring effect from its bare value of 8.3
MHz all the way out to 15.7 MHz (optical input power of P
i
= 315 µW). In the process, the ﬂapping
mode is tuned across the breathing mode at 13.6 MHz. Although the opticallydark breathing mode
is barely visible in the tranduced spectrum at large lasercavity detunings, its spectral amplitude is
considerably enhanced as the opticallybright ﬂapping mode is tuned into resonance. In addition,
a strong Fanolike lineshape, with 13 dB antiresonance, appears in the power spectrum near
resonance of the two modes (Fig. 3.3(fh)).
As shown schematically in Fig. 3.5(a), the Fanolike interference in the opticallybright power
spectral density can be attributed to an internal mechanical coupling between the ﬂapping and
breathing mechanical modes. This is quite similar to the phononphonon interaction during the
structural phase transition in solids [66, 85–89], in which the internal coupling between phonon
modes produces Fanolike resonances in the Ramanscattering spectra.
The power spectral density (PSD) of the cavity transmission is linearly proportional to Eq. (3.24).
Equation (3.24) together with (3.9) is used to ﬁnd the theoretical PSD shown in Fig. 3.3, by using an
optomechanical coupling coefﬁcient of g
om
/2π =33 GHz/nm and an effective mass of m
b
=264 pg
for the ﬂapping mode, both obtained from FEM simulations. The intrinsic and loaded optical qual
ity factors of 1.07 10
6
and 0.7 10
6
are obtained from optical characterization of the cavity
resonance, and are also given in the caption of Fig. 3.1. The intrinsic mechanical frequencies and
damping rates of the two modes (Ω
mb
, Ω
md
, Γ
mb
, and Γ
md
) are obtained from the experimentally
recorded PSD of cavity transmission with a large lasercavity detuning, as given in the caption of
Fig. 3.3. The mechanical coupling coefﬁcient η is treated as a ﬁtting parameter. Fitting of the PSDs
results in η = 3.32 MHz, indicating a strong internal coupling between the two mechanical modes.
As shown clearly in Fig. 3.3(d, fh), our theory provides an excellent description of the observed
phenomena.
67
10 12 14 16 18 8 6 0 1
) z H M ( y c n e u q e r F ) z H M ( y c n e u q e r F ∆
0
/Γ
t
P
o
w
e
r
S
p
e
c
t
r
a
l
D
e
n
s
i
t
y
(
2
0
d
B
/
d
i
v
)
10 12 14 16 18 8 6 3 2
h
g
f
16 8 10 12 14
Frequency (MHz)
6 1 8 1 8 10 12 14
Frequency (MHz)
6 1 8 1 8 10 12 14
Frequency (MHz)
18
80
90
100
75
P
S
D
(
d
B
m
/
H
z
)
85
95
c
d
e
f
g
h
a
b
Figure 3.3: (a,b) FEM simulated mechanical motion of the differential ﬂapping mode (a) and
the common breathing mode (b), with simulated frequencies of 7.95 and 14.2 MHz. The color
map indicates the relative magnitude (exaggerated) of the mechanical displacement. (c) Recorded
power spectral density (PSD) of the cavity transmission for the doubledisk, with an input power of
315 µW. Each curve corresponds to a normalized lasercavity frequency detuning, ∆
0
/Γ
t
indicated
in (e). For display purposes, each curve is relatively shifted by 10 dB in the vertical axis. (d) The
corresponding theoretical PSD. (fh) Detailed PSD at three frequency detunings indicated by the
arrows in (e), with the experimental and theoretical spectra in blue and red, respectively.
68
3.6 Coherent mechanical mode mixing in zipper cavities
The coherent mixing of mechanical excitation is universal to gradientforcebased NOMS with a
giant optical spring effect. Similar phenomena to that presented for doubledisks were also observed
in the zipper cavity. However, due to the device geometry, the coupled nanobeams have more
complex mechanical mode families in which all the evenorder mechanical modes are optically
dark, because they exhibit a mechanical node at the beam center where the optical mode is located.
As the sameorder common and differential motions of the two beams have similar mechanical
frequencies, they can simultaneously couple to the same optically bright mode, leading to multiple
excitation interferences on the mechanical response.
In the case when the optically bright mode is coupled to two optically dark modes, the Hamil
tonian for the mechanical system is given by the following general form:
H
m
=
∑
i=b,1,2
_
p
2
i
2m
i
+
1
2
k
i
x
2
i
_
+κ
1
x
b
x
1
+κ
2
x
b
x
2
, (3.41)
where i = b, 1, 2 corresponds to the optically bright mode and optically dark modes 1 and 2, respec
tively. With this Hamiltonian, counting in both the optical gradient force and the Langevin forces
from the thermal reservoir, we obtain the equations of motions for the three modes:
d
2
x
b
dt
2
+Γ
mb
dx
b
dt
+Ω
2
mb
x
b
+
κ
1
m
b
x
1
+
κ
2
m
b
x
2
=
F
b
m
b
+
F
o
m
b
, (3.42)
d
2
x
1
dt
2
+Γ
m1
dx
1
dt
+Ω
2
m1
x
2
+
κ
1
m
1
x
b
=
F
1
m
1
, (3.43)
d
2
x
2
dt
2
+Γ
m2
dx
2
dt
+Ω
2
m2
x
3
+
κ
2
m
2
x
b
=
F
2
m
2
, (3.44)
where the gradient force F
o
is given by Eq. (3.15), and the statistical properties of the Langevin
forces are given by Eq. (3.14). Following the same analysis as in Section 3.3.3, we can obtain the
spectral intensity for the mechanical displacement of the optically bright mode as
S
x
b
(Ω) =
2k
B
T
m
b
η
4
1
Γ
m1
L
2
(Ω)
2
+η
4
2
Γ
m2
L
1
(Ω)
2
+Γ
mb
L
1
(Ω)L
2
(Ω)
2
¸
¸
L
b
(Ω)L
1
(Ω)L
2
(Ω) η
4
1
L
2
(Ω) η
4
2
L
1
(Ω)
¸
¸
2
, (3.45)
where η
4
j
κ
2
j
m
b
m
j
( j = 1, 2) represents the mechanical coupling coefﬁcient. L
j
(Ω) = Ω
2
mj
Ω
2
iΓ
mj
Ω ( j = 1, 2) and L
b
(Ω) is given by Eq. (3.20) with Ω
mb
and Γ
mb
given in Eqs. (3.21) and (3.22),
respectively. As the optical wave is coupled to the optically bright mode only, the power spectral
69
density of the cavity transmission is still given by Eq. (3.9), with the mechanical response S
x
b
given
in Eq. (3.45).
Figure 3.4 shows the PSD of the cavity transmission by launching a continuous wave into a res
onance of the coupled nanobeams with an intrinsic and loaded Q factor of 3.0 10
4
and 2.8 10
4
,
respectively. Three mechanical modes are clearly visible, where mode I is the fundamental dif
ferential mode [Fig. 3.4(h)I], and mode II and III correspond to the secondorder common and
differential modes [Fig. 3.4(h)II and III], respectively. Similar to the doubledisk NOMS, the gi
gantic optical spring effect shifts the frequency of the optically bright mode I from its intrinsic
value of 8.06 MHz to 19 MHz, crossing over both optically dark modes II and III closely located
at 16.54 and 17.04 MHz and resulting in complex interferences on the power spectra [Fig. 3.4(a)].
Equation (3.45) provides an accurate description of the observed phenomena, as shown clearly in
Fig. 3.4(b), (d)(f). Fitting of the PSD results in mechanical coupling coefﬁcients of η
1
=3.45 MHz
and η
2
=3.48 MHz, implying that the two optically dark modes couple to the fundamental optically
bright mode with a similar magnitude.
3.7 Analogy to electromagneticallyinduced transparency
The mechanical response given by Eq. (3.24) is directly analogous to the atomic response in EIT [83].
Just as in EIT, one can understand the resulting Fano lineshape in two different ways. The ﬁrst
perspective considers the interference associated with multiple excitation pathways. In the optome
chanical system, the mechanical motion of the ﬂapping mode is thermally excited along two dif
ferent pathways, either directly into the broadband (lossy) ﬂapping mode, or indirectly, through the
ﬂapping mode, into the longlived breathing mode, and then back again into the ﬂapping mode. The
two excitation pathways interfere with each other, resulting in the Fanolike resonance in the spec
tral response of the optically bright ﬂapping mode. An alternative, but perfectly equivalent view of
the coupled optomechanical system considers the dressed states resulting from the internal mechan
ical coupling. In this picture the internal mechanical coupling renormalizes the broadband ﬂapping
mode and the narrowband breathing mode into two dressed mechanical modes, both broadband and
opticallybright. In particular, when the ﬂapping and breathing mechanical frequencies coincide,
the two dressed modes are excited with equal amplitude and opposite phase at the center frequency
between the split dressed states. Destructive interference results, suppressing excitation of the me
chanical system at the line center. Consequently, the mechanical motion becomes purely a trapped
70
80
90
100
110
P
S
D
(
d
B
m
/
H
z
)
22 14 16 18 20
Frequency (MHz)
22 14 16 18 20
Frequency (MHz)
22 14 16 18 20
Frequency (MHz)
(e) (f ) (d)
ΙΙ ΙΙΙ Ι
Ι ΙΙ ΙΙΙ
(h)
Frequency (MHz)
P
o
w
e
r
S
p
e
c
t
r
a
l
D
e
n
s
i
t
y
(
2
0
d
B
/
d
i
v
)
10 12 14 16 18 20
(a)
0 1
∆
0
/Γ
t
(c)
f
e
d
Frequency (MHz)
10 12 14 16 18 20
(b)
Figure 3.4: (a) Experimentally recorded power spectral densities of the cavity transmission for the
zipper cavity of Fig. 3.1(df), with an input power of 5.1 mW. Each curve corresponds to a laser
frequency detuning indicated in (c). Each curve is relatively shifted by 5 dB in the vertical axis for
a better vision of the mechanical frequency tuning and the induced mechanical interference. The
optically dark mode II and III have a fullwidth at half maximum (FWHM) of 0.16 and 0.15 MHz,
respectively. The optically bright mode I has an intrinsic FWHM of 0.30 MHz. (b) The corre
sponding theoretical spectra of the power spectral density. (d)(f) The detailed spectra of the power
spectral density at three frequency detunings indicated by the three arrows in (c). The blue and red
curves show the experimental and theoretical spectra, respectively. (h) FEM simulated mechanical
motions for the fundamental differential mode (I), the secondorder common (II) and differential
(III) modes, whose frequencies are indicated by the arrows in (a). The color map indicates the
relative magnitude (exaggerated) of the mechanical displacement.
71
input output
1〉
3〉
2〉
−〉
+〉
probe
control
1〉
3〉
2〉
RF/microwave
photon
coupling
Stokes
pump
antiStokes
pump
d
a
c
b
Γ
2
Γ
3
γ
2
γ
3
k
1
m
1
k
2
k
12
m
2
x
b
x
d
input output
2
1
coupling
Figure 3.5: (a) Schematic of an equivalent FabryPerot cavity system showing mechanical mode
mixing. The mechanical motion of the cavity mirror (m
1
, equivalent to the opticallybright ﬂapping
mode) is primarily actuated by the spring k
1
and the optical force. It is internally coupled to a second
massspring system (m
2
, equivalent to the breathing mode) actuated by the spring k
2
which is de
coupled from the optical wave. The two masses are internally coupled via spring k
12
. (b) A photonic
analogue to the optomechanical system involving coupled resonators. Microcavity 1 is directly cou
pled to the external optical waveguide (equivalent to the opticallybright ﬂapping mode) and also
internally coupled the narrowband cavity 2 (equivalent to the opticallydark breathing mode). (c)
State diagram of an EITlike medium. The excited state (2) is split by the optical control beam into
two broadband dressed states (+ and −). The dipole transition between groundstates 1 and
3 is forbidden. (d) The state diagram corresponding to the optomechanical system of (a) where
1 is the phonon vacuum state, and 2 and 3 correspond to the ﬂapping and breathing modes,
respectively.
72
mechanicallydark state, transparent to external excitation. As shown in Fig. 3.3(c), this induced
mechanical transparency is a direct analogue to EIT in atomic systems [63, 83, 90, 91], in which
the quantum interference between the transition pathways to the dressed states of the excited elec
tronic state, through either 1 + or 1 , leads to an induced spectral window of optical
transparency.
Despite the intriguing similarities between the optomechanical system studied here and EIT in
atomic media, there are some important, subtle differences. For instance, in the optomechanical
system, rather than the linear dipole transition of EIT, the interaction corresponds to a secondorder
transition. The dynamic backaction between the cavity ﬁeld and mechanical motion creates Stokes
and antiStokes optical sidebands, whose beating with the fundamental optical wave resonates with
the mechanical motion to create/annihilate phonons (see Fig. 3.3(d)). Functionally, this is like
coherent Stokes and antiStokes Raman scattering, albeit with unbalanced scattering amplitudes
resulting from the coloring of the electromagnetic density of states by the optical cavity.
The system Hamiltonian of an optomechanical cavity is given by the following general form:
H = ω
0
a
†
a+Ω
m
b
†
b+g
om
x
b
a
†
a, (3.46)
where a and b are the annihilation operators for photon and phonon, respectively, normalized such
that a
†
a and b
†
b represent the operators for photon and phonon number. x
b
is the mechanical
displacement for the optically bright mode, related to b by
x
b
=
¸
2m
b
Ω
mb
_
b+b
†
_
. (3.47)
Therefore, the interaction Hamiltonian between the optical wave and the mechanical motion is given
by
H
i
= hga
†
a
_
b+b
†
_
, (3.48)
where the factor g
_
g
2
om
3
2m
b
Ω
mb
_
1/2
.
The mechanical motion modulates the intracavity ﬁeld to create two optical sidebands. As a
result, the optical ﬁeld can be written as
a = a
p
+a
s
e
iΩ
mb
t
+a
i
e
iΩ
mb
t
, (3.49)
73
where a
p
is the ﬁeld amplitude of the fundamental wave, and a
s
and a
i
are those of the generated
Stokes and antiStokes wave, respectively. As the magnitudes of the Stokes and antiStokes side
bands are much smaller than the fundamental wave, when we substitute Eq. (3.49) into Eq. (3.48)
and leave only the ﬁrstorder terms of a
s
and a
i
, under the rotatingwave approximation, the inter
action Hamiltonian becomes
H
i
= g
_
b+b
†
_
a
†
p
a
p
+gb
†
_
a
†
s
a
p
+a
†
p
a
i
_
+gb
_
a
†
p
a
s
+a
†
i
a
p
_
. (3.50)
In Eq. (3.50), the ﬁrst term describes the static mechanical actuation, which changes only the equi
librium position of mechanical motion and is neglected in the current analysis, as discussed previ
ously. The second and third terms show clearly that the process corresponds directly to coherent
Stokes and antiStokes Raman scattering as shown in Fig. 3.5(d).
Therefore, in analogy to EIT, it is the modulation signal carried by the incident optical wave
(radiofrequency or microwave photons) that fundamentally probes/excites the mechanical motion
and to which the trapped mechanicallydark state becomes transparent. Moreover, rather than tuning
the Rabisplitting through the intensity of a control beam resonant with the 3 2 electronic
transition (Fig. 3.3(c)), this opticallyinduced mechanical transparency is controlled via optical
spring tuning of the resonance frequency of the optically bright ﬂapping mechanical mode. Perhaps
the most apt analogy to the optomechanical system can be made to the photonic resonator system
shown in Fig. 3.3(d). The interference in this case is between the two optical pathways composed of
the waveguidecoupled lowQoptical resonator 1, and the waveguidedecoupled highQresonator 2.
This interference again leads to a Fanolike resonance, or what has been termed coupledresonator
induced transparency, in the optical cavity transmission [70–73].
3.8 Discussion
Although the studies considered here involve thermal excitation of the optomechanical system, the
same phenomena can be excited more efﬁciently, and with greater control, using external optical
means (Sec. 3.3.4). As such, beyond the interesting physics of these devices, exciting application
in RF/microwave photonics and quantum optomechanics exist. Similar to the information storage
realized through EIT [83, 92, 93], optical information can be stored and buffered in the dark me
chanical degree of freedom in the demonstrated NOMS. This can be realized through a procedure
74
similar to that recently proposed for coupled optical resonators [94, 95] in which dynamic, adiabatic
tuning of optical resonances are used to slow, store, and retrieve optical pulses. The correspond
ing optomechanical system would consist of an array of doubledisk resonators, all coupled to a
common optical bus waveguide into which an optical signal carrying RF/microwave information
would be launched. In this scheme, a second control optical beam would adiabatically tune the
frequency of the opticallybright ﬂapping mode of each resonator, allowing for the RF/microwave
signal to be coherently stored in (released from) the longlived breathing mode through adiabatic
compression (expansion) of the mechanical bandwidth [94, 95]. In comparison to the allphotonic
system, optomechanical systems have several advantages, primarily related to the attainable life
time of the dark mechanical state. For example, the radial breathing mechanical mode of a similar
whisperinggallery cavity has been shown to exhibit a lifetime of more than 2 ms [96], a timescale
more than seven orders of magnitude longer than that in demonstrated photoniccoupledresonator
systems [97] and comparable with EIT media [92, 93]. Moreover, mechanical lifetimes of more
than one second have recently been demonstrated using stressed silicon nitride nanobeam [98] and
nanomembrane [40] mechanical resonators operating in the MHz frequency regime. In the quantum
realm, such a system operating in the goodcavity or sidebandresolved regime (by increasing either
the optical Q factor [59] or the mechanical frequency), would reduce the simultaneous creation and
annihilation of Stokes and antiStokes photons, enabling efﬁcient information storage and retrieval
at the singlequantalevel suitable for quantum state transfer.
75
Chapter 4
Mechanically Pliant Double Disk
Resonators
4.1 Introduction
Optical information processing in photonic interconnects relies critically on the capability for wave
length management [99, 100]. The underlying essential functionalities are optical ﬁltering and
wavelength routing, which allow for precise selection and ﬂexible switching of optical channels
at high speeds over a broad bandwidth [99, 101–103]. In the past two decades, a variety of tech
nologies have been developed for this purpose [104–106]; those based on micro/nanoresonators
are particularly attractive because of their great potential for future onchip integrated photonic ap
plications [107–119]. In general, reconﬁgurable tuning of cavity resonances is realized through
thermooptic [109, 110, 117, 119], electrooptic [112, 118, 119], photochemical [111], optoﬂuidic
[115], or microelectricalmechanical approaches [108, 114, 120]. However, all of these tuning mech
anisms have intrinsic limitations on their tuning speed [108–111, 114, 115, 117], tuning bandwidth
[112, 118, 119], routing efﬁciency [108, 109, 118, 119], and/or routing quality [112, 118, 119].
Here we propose and demonstrate an alloptical wavelengthrouting approach which combines the
advantages of various approaches into one nanophotonic device. By using a tuning mechanism
based upon the optical gradient forces in a speciallydesigned nanooptomechanical system, we are
able to realize seamless wavelength routing over a range about 3000 times the channel intrinsic
linewidth, with a tuning efﬁciency of 309 GHz/mW, a switching time of less than 200 ns, and 100%
channelquality preservation over the entire tuning range. The demonstrated approach and device
geometry indicates great prospects for a variety of applications such as channel routing/switching,
buffering, dispersion compensation, pulse trapping/release, and tunable lasing, with easy onchip
76
integration on a siliconcompatible platform. This work was initially presented in Ref. 78.
The physics of electromagnetic forces within mechanicallycompliant resonant cavities is by
now well established, with some of the early experimental considerations being related to the
quantumlimited measurement of weak, classical forces [29]. In the optical domain, experiments
involving optical FabryPerot “pendulum cavities” were ﬁrst explored [121], with more recent
studies having measured radiation pressure forces in micro and nanomechanical structures [35–
38, 40, 44, 45, 48, 51]. In each of these systems, whether it be gravitational wave observatory
[31] or photonic crystal nanomechanical cavity [45], the same fundamental physics applies. A nar
rowband laser input to the system, of ﬁxed frequency, results in a “dynamical backaction” [33]
between mechanical ﬂuctuations and the internal electromagnetic ﬁeld. This dynamical backaction
modiﬁes both the real and imaginary parts of the frequency of the mechanical motion, yielding an
opticallycontrollable, dynamic mechanical susceptibility. A separate effect occurs when the laser
frequency is swept across the cavity resonance, pushing on the mechanical system as the internal
light ﬁeld builds up near cavity resonance. The more compliant the mechanical system, the larger
the static displacement and the larger the tuning of the optical cavity. Here we utilize both the static
and dynamic mechanical susceptibilities of a coupled optomechanical system to realize a chip
based optical ﬁlter technology in which wideband tuning and fast switching can be simultaneously
accomplished.
4.2 Spiderweb resonator design and optical characterization
The optomechanical system we consider here is a simple modiﬁcation to the common microring
whisperinggallery cavity that has found widespread application in microphotonics. As shown in
Fig. 4.1(d), it consists of a pair of planar microrings, one stacked on top of the other [122, 123]. The
resulting nearﬁeld modal coupling forms a “supercavity,”, with a resonance frequency ω
0
strongly
dependent on the vertical cavity spacing, x.
Fabrication of the spiderweb whisperinggallery resonator began with initial deposition of the
cavity layers. The two silica web layers and the sandwiched amorphous silicon (αSi) layer were
deposited on a (100) silicon substrate by plasmaenhanced chemical vapor deposition, with a thick
ness of 400 4 nm and 150 3 nm for the silica and αSi layers, respectively. The wafer was
then thermally annealed in a nitrogen environment at a temperature of T = 1050 K for 10 hours to
drive out water and hydrogen in the ﬁlm, improving the optical quality of the material. The spider
77
1 µm 20 µm
b
25 µm
c a
d
2
1
0
1
2
H
e
i
g
h
t
(
µ
m
)
40 42 44 46 48
Radius (µm)
e
f
αSi
SiO
2
SiO
2
x
Si
1
5
0
n
m
g
a
p
g
Ring spacing (nm)
200 800 600 400 0
P
e
r

p
h
o
t
o
n
f
o
r
c
e
(
f
N
)
0
30
25
20
15
10
5
Ring spacing (nm)
200 800 600 400 0
W
a
v
e
l
e
n
g
t
h
(
n
m
)
1460
1500
1540
1580
0
10
20
30
40
50
1000
g
O
M
/
2
π
(
G
H
z
/
n
m
)
Figure 4.1: Scanning electron microscope images of (a) the betweenring gap, (b) the 54 µm spi
derweb resonator, and (c) the 90 µm spiderweb resonator. (d) Schematic of a crosssection of the
resonator, showing the bending of the two silica rings under the inﬂuence of the optical force. (e)
Mechanical FEM simulation of the bending of the 90 µm spiderweb resonator. The outward bend
ing motion is shown for ease of viewing, and is exaggerated for clarity. (f) FEM simulation of the
radial component of the electric ﬁeld for the fundamental TE bonding mode of the 90 µm spider
web structure. (g) The theoretical wavelength tunability, perphoton force, and wavelength (inset)
of the spiderweb cavity as a function of the ring spacing. The vertical dashed lines represent the
experimentallyrealized ring spacing of 150 nm.
78
web pattern was created using electron beam lithography followed by an optimized C
4
F
8
SF
6
gas
chemistry reactive ion etch. Release of the web structure was accomplished using a SF
6
chemical
plasma etch which selectively (30, 000 : 1) attacks the intermediate αSi layer and the underlying Si
substrate, resulting in a uniform undercut region which extends radially inwards 4 µm on all bound
aries, fully releasing the web. Simultaneously, the underlying silicon support pedestal is formed.
Two nanoforks were also fabricated near the doubledisk resonator to mechanically stabilize and
support the ﬁber taper during optical coupling; the geometry was optimized such that the forks
introduce a total insertion loss of only 4%.
The optomechanical coupling coefﬁcient, g
OM
dω
0
/dx, determines both the tunability and
perphoton optical force [43, 76]. Finiteelementmethod (FEM) simulation shows that, for two
400nmthick planar silica whisperinggallery microcavities placed 150 nm apart (Fig. 4.1(fg)),
the resonance tunability is as large as g
OM
/2π = 31 GHz/nm (corresponding to a 21 fN/photon
force). The corresponding static mechanical displacement for N photons stored inside the cavity is
∆x
static
=Ng
OM
/k, where k is the intrinsic spring constant of the mechanical structure. The overall
magnitude of the cavity resonance tuning is then,
∆ω
0
= g
OM
∆x
static
=
Ng
2
OM
k
=
g
2
OM
P
d
kω
0
Γ
0
, (4.1)
where P
d
is the power dropped into the cavity and Γ
0
is the intrinsic photon decay rate, inversely
proportional to the optical quality factor.
As the optical gradient force stems from the evanescent ﬁeld coupling between the two near
ﬁeldspaced cavities, it is completely independent of the roundtrip length of the cavity. This feature
enables independent control of the optical and mechanical properties, allowing us to freely engi
neer the intrinsic mechanical rigidity through the scalability of the structure without changing the
perphoton force. In order to minimize the mechanical stiffness while also providing mechanical
stability, we utilize a spiderweblike support structure consisting of an arrangement of spokes and
inner rings [96]. The zerothorder spiderweb cavity (Fig. 4.1(b)) has a 54 µm diameter outer ring
supported by ﬁve spokes, while the ﬁrstorder structure (Fig. 4.1(c)) has a 90 µm outer diameter
ring with six spokes and one supporting inner ring. FEM simulations show that these structures
have spring constants of 9.25 N/m and 1.63 N/m for the smaller and larger resonators, respectively.
In addition to the favorable mechanical properties, the whisperinggallery nature of the spi
derweb resonator provides for highQ optical resonances. Optical spectroscopy of the devices is
79
OSA
MZ I
Reference
Detector 1
VOA
50 : 50
splitter
Polarization
Controller
EOM
DC Bias
Polarization
Controller
VOA
10 : 90
splitter
Reference
Detector 2
Oscilloscope Network Analyzer
R
F
D
r
i
v
e
DEMUX
MUX
10 : 90
splitter
Reference
Detector 3
Highspeed
Detector
fiber taper
Pump Laser
EDFA
VOA
BPF
Polarization
Controller
Probe Laser
Figure 4.2: The pump and probe lasers are coupled to the spiderweb resonator via a singlemode
silica ﬁber taper stabilized by two nanoforks fabricated near the device. The pump laser power is
boosted by an erbiumdoped ﬁber ampliﬁer (EDFA) and passed through a bandpass ﬁlter (BPF).
The two lasers are split into separate wavelength channels using a mux/demux system (providing
greater than 120 dB pumpprobe isolation). For modulation experiments, the pump laser wavelength
is modulated using an electrooptic modulator (EOM) driven by a network analyzer. The laser
power levels are controlled by several variable optical attenuators (VOAs), the probe wavelength
is calibrated by a MachZehnder interferometer (MZI), and the pump wavelength is monitored by
an optical spectrum analyzer (OSA). The spiderweb device itself is contained within a nitrogen
environment at atmospheric pressure.
80
performed using the experimental setup shown in Fig. 4.2. Figure 4.3(a) shows the low power, in
plane polarized, wavelength scan of a 54µm diameter resonator. The excited family of resonances,
corresponding to the fundamental transverseelectriclike (TElike) modes, has a freespectral range
(FSR) of 9.7 nm, with resonances at λ = 1529 nm and λ = 1549 nm exhibiting intrinsic quality fac
tors of Q
i
= 1.04 10
6
and Q
i
= 0.90 10
6
, respectively.
The extremely small intrinsic spring constant of the spiderweb resonator leads to signiﬁcant
thermal Brownian mechanical motion and introduces considerable ﬂuctuations on the cavity trans
mission spectrum, as shown in Fig. 4.5(a). This makes it difﬁcult to measure the optical Q factor
of a cavity resonance. As discussed previously, the thermal Brownian mechanical motion can be
signiﬁcantly suppressed through the optical spring effect. This feature provides an elegant way to
accurately characterize the optical Q of a cavity mode, by launching a relatively intense wave at a
different resonance to suppress the perturbations induced by the thermal mechanical motion. More
over, a complete theory developed previously in Sec. 2.4 and Ref. 48 was used to describe the linear
cavity transmission with the inclusion of the optomechanical effect.
4.3 Static lter response
The combination of high cavity Qfactor, large g
OM
, and ﬂoppy spiderweb structure result in the
large optomechanical bistability shown in Fig. 4.3(b). With a power of 1.7 mW dropped into the
cavity, the cavity resonance initially at λ = 1549 nm is shifted by 4.4 nm (a little more than 0.5
THz), corresponding to a static mechanical displacement of ∆x
static
=17.7 nm. We observed similar
performance from the larger 90 µm spiderweb structures, although device yield (20%) and a slow
change in device properties over time (despite devices being tested in a nitrogen environment to
avoid water adsorption), indicate that further mechanical design optimization may be necessary for
the larger structures. By comparison, the smaller 54µm diameter structures had near100% yield
and maintained their properties over the entire period of testing.
As the mechanical displacement is universally experienced by all doublering cavity modes,
the displacement actuated by one cavity mode can be used to control the wavelength routing of
an entire mode family, indicating a great potential for broad waveband translation and switching
in the wavelengthdivision multiplexing conﬁguration. This is demonstrated in Fig. 4.3(d), where
the mechanical displacement actuated by the “pump” mode at λ = 1549 nm is used to control the
wavelength of a “probe” mode initially located at λ = 1529 nm. With increased dropped pump
81
probe
a
Wavelength (nm)
1530 1540 1550 1560 1520 1510 1500
N
o
r
m
a
l
i
z
e
d
t
r
a
n
s
m
i
s
s
i
o
n
0.8
1
0.6
0.4
pump
0.2
0
0 8
Detuning (pm)
8 0 8
Detuning (pm)
8
1 1
0.5
0.8
0.6
1549 1550 1551 1552 1553
low power
h
ig
h
p
o
w
e
r
b
N
o
r
m
a
l
i
z
e
d
t
r
a
n
s
m
i
s
s
i
o
n
0.4
0.6
0.8
1
Wavelength (nm)
0.5
0
Pump dropped power (mW)
0
0.2
0.4
0.6
0.8
1
[
l
p

1
5
3
1
]
(
n
m
)
0
0.5
0.5
e
0.1 0.2 0.3 0.4
P
r
o
b
e
t
u
n
i
n
g
(
G
H
z
)
0 1 0.5 1.5
d
0
200
300
400
500
100
c
0
0.5
1
1.5
P
u
m
p
d
r
o
p
p
e
d
p
o
w
e
r
(
m
W
)
1529 1530 1531 1532 1533
Probe wavelength (nm)
3
0
9
G
H
z
/
m
W
14 GHz/mW
Figure 4.3: (a) Broadband optical transmission spectrum of the 54 µm spiderweb cavity. Inset:
ﬁne frequency scan of pump (probe) mode, highlighed in red (green), with Lorentzian ﬁt to the
lineshape. (b) Overcoupled pumpmode transmission spectrum at P
d
= 275 nW (blue) and P
d
=
1.7 mW (red). (c) Probemode transmission curves for a selection of dropped powers in d, with
P
d
indicated by the baseline of each transmission curve. (d) Measured (blue circles) and linear
ﬁt (red curve) to the probe resonance wavelength tuning versus P
d
. Green curve corresponds to
thermooptic component of tuning. (e) Intensity image of the optical transmission spectrum near
the anticrossing of two TE and TMlike probe modes.
82
power, the probe wavelength is tuned linearly and continuously by 4.2 nm, approximately 3000
times the probe resonance intrinsic channel linewidth (or 500 times the loaded linewidth). This
factor is at least one order of magnitude larger than any other conventional approach previously
reported [108–115, 117–119]. The tuning range shown in Fig. 4.3(d) is about 43% of the FSR. In
principle, it is possible to tune over the entire freespectral range with a moderate dropped pump
power of only 4 mW. Importantly, this wavelengthrouting approach is purely dispersive in nature
and completely preserves the channel quality during the wavelength routing process as can be clearly
seen in Fig. 4.3(c). This is in contrast to other tuning mechanisms such as the electrooptic approach
via carrier injection [112, 118, 119], in which the accompanying carrier absorption degrades the
quality of the switched channel and thus limits the ultimate tuning bandwidth.
A linear ﬁt to the probe resonance tuning data data in Fig. 4.3(d) gives a tuning efﬁciency of
309 GHz/mW. This value agrees reasonably well with the theoretically predicted value of 393 GHz/mW,
inferred from optical and mechanical FEM simulations and the measured optical Qfactor (see
eq. (4.1)). Independent measurements show that the thermooptic effect contributes only a small
component to the overall tuning rate (13.8 GHz/mW; green curve in Fig. 4.3(d)), and FEM simula
tions indicate a negligible thermomechanical component ( 0.06%).
The thermooptical effect on the resonance tuning was calibrated by using another identical de
vice on the same sample. To isolate the thermooptic effect from the optomechanical effect, we
caused the two rings to stick together through the van der Waals force, so the ﬂapping mechanical
motion was completely eliminated. Testing was performed on a cavity mode at 1552 nm using ex
actly the same conditions as for the wavelength routing measurements. A power of 2.1 mW dropped
into the cavity introduces a maximum resonance red tuning by 0.23 nm, corresponding to a tuning
rate of 0.11 nm/mW (13.8 GHz/mW), about 4% of the total tuning rate recorded experimentally.
The thermooptic resonance tuning indicates a maximum temperature change of 21 K in the
resonator. FEM simulations show that such a temperature variation of the resonator introduces
a ringgap change by only about 10 pm, shown by the differential displacement of the top and
bottom rings in Fig. 4.4. Therefore, thermally induced static mechanical deformation has only a
negligible contribution of 0.06% of the experimentally recorded wavelength tuning. The negligible
contributions of both thermooptic and thermomechanical effects are conﬁrmed by the pumpprobe
modulation spectra shown in Fig. 4.5(d) and (e). This, and properties of the dynamical response of
the system (see below), show that the wavelength routing is indeed a result of the optical gradient
force. The difference between theoretical and experimental optical force tuning rates ( 25%) can
83
0
10
20
z

d
i
s
p
l
a
c
e
m
e
n
t
(
p
m
)
z
Figure 4.4: FEM simulation illustrating the zdisplacement of a 54 µm spiderweb resonator under
the 21 K temperature differential between substrate and ring induced by 2.1 mW dropped optical
power.
likely be attributed to the uncertaintity in the Young’s modulus of the annealed PECVD silica used
to form the spiderweb structure.
In addition to the TElike modes, the spiderweb doublering resonator also supports a family
of highQ transversemagneticlike (TMlike) modes with a FSR of 10 nm. FEM simulations show
that the perphoton force is slightly larger for the TM modes (26.5 fN/photon, or a 59% larger tuning
efﬁciency), due primarily to the enhanced electric ﬁeld strength in the nanoscale gap between the
rings for polarization normal to the plane of the rings. Figure 4.3(e) shows the mode hybridization
between a pair of TE and TMlike modes (the slight angle in the outer sidewall of the two rings
breaks the vertical symmetry, allowing for modemixing) induced by the optical force tuning of the
two mode families.
The anticrossing between the two probe modes when they approach each other is primarily due
to the internal coupling between the two cavity modes, which can be described by a simple theory
as follows. Assume two cavity resonances located at ω
01
and ω
02
. For an input probe wave at ω,
the two cavity modes are excited through the following equations:
da
1
dt
= (i∆
1
Γ
t1
2
)a
1
+iβa
2
+i
_
Γ
e1
A
in
, (4.2)
da
2
dt
= (i∆
2
Γ
t2
2
)a
2
+iβa
1
+i
_
Γ
e2
A
in
, (4.3)
where ∆
j
=ω ω
0j
represents the cavity detuning of the j
th
mode, and β is the optical coupling co
efﬁcient between the two cavity modes. With a continuouswave input, the steady state of Eqs. (4.2)
84
and (4.3) is given by the following solution
a
1
=
iA
in
_
(i∆
2
Γ
t2
/2)
Γ
e1
iβ
Γ
e2
¸
(i∆
1
Γ
t1
/2)(i∆
2
Γ
t2
/2) +β
2
, (4.4)
a
2
=
iA
in
_
(i∆
1
Γ
t1
/2)
Γ
e2
iβ
Γ
e1
¸
(i∆
1
Γ
t1
/2)(i∆
2
Γ
t2
/2) +β
2
. (4.5)
As the transmitted ﬁeld from the cavity is given by A
T
= A
in
+i
Γ
e1
a
1
+i
Γ
e2
a
2
, the cavity
transmission thus has the following equation
T
A
T
2
A
in
2
=
¸
¸
¸
¸
¸
(i∆
1
Γ
01
Γ
e1
2
)(i∆
2
Γ
02
Γ
e2
2
) +(β i
Γ
e1
Γ
e2
)
2
(i∆
1
Γ
t1
/2)(i∆
2
Γ
t2
/2) +β
2
¸
¸
¸
¸
¸
2
. (4.6)
The experimental observation agrees well with this simple theory (dashed curve in Fig. 4.3(e)),
giving a tuning efﬁciency for the TM modes which is 42% larger than that of the TE modes. This
precisely tunable channel coupling may ﬁnd applications in polarization switching/multiplexing/demultiplexing
in optical signal processing, or carriersideband ﬁltering in microwave photonics [124].
4.4 Dynamic lter response
In addition to the static mechanical actuation of the spiderweb structure, the optical gradient force
also introduces dynamical back action which alters the dynamic response of the mechanical motion
[33, 50, 125]. The inphase component of the optical force leads to a modiﬁed mechanical resonance
frequency and effective dynamical spring constant of
k
= k +
2g
2
OM
P
d
∆
ω
0
Γ
0
[∆
2
+(Γ
t
/2)
2
]
, (4.7)
where ∆ = ω
l
ω
o
is the detuning of the input laser (ω
l
) from the cavity resonance (ω
o
) frequency
and Γ
t
is the photon decay rate of the loaded cavity. As the intrinsic spring constant of the spiderweb
resonator is small (9.5 N/m), the dynamical spring can be greatly modiﬁed optically. The alteration
of the effective dynamic spring is clearly seen in the resonance spectra of the cavity resonances
(left panel of Fig. 4.5(a)). For the ﬂoppy spiderweb structure, thermal Brownian motion introduces
signiﬁcant ﬂuctuations in the cavity resonances. As pump power is dropped into the cavity, however,
the dynamic spring stiffens and strongly suppresses the magnitude of the thermal ﬂuctuations (right
panel of Fig. 4.5(a)).
85
d
M
o
d
u
l
a
t
i
o
n
s
p
e
c
t
r
u
m
(
d
B
)
25
20
15
10
5
0
5
10
15
20
25
0 5 10 15 20 25 30 35
Frequency (MHz)
a
N
o
r
m
a
l
i
z
e
d
t
r
a
n
s
m
i
s
s
i
o
n
1
0.8
0.6
0.4
0.2
Wavelength (nm)
1529.12 1529.08 1529.28 1529.32
e
0 200 400 600 800 1000 1200 1400
Frequency (MHz)
M
o
d
u
l
a
t
i
o
n
s
p
e
c
t
r
u
m
(
d
B
)
0
20
20
40
60
80
100
78 dB
0.8
0
0.2
0.4
0.6
Time(µs)
0 1.2 0.2 0.4 0.6 0.8 1
0.8
0
F
r
a
c
t
i
o
n
a
l
M
o
d
u
l
a
t
i
o
n
0.2
0.4
0.6
pump
probe
c
1
0.8
0.6
1
0.8
0.6
1
0.8
0.6
Time(µs)
0 0.1 0.2 0.3 0.4 0.5
N
o
r
m
a
l
i
z
e
d
t
r
a
n
s
m
i
s
s
i
o
n
b
0
Figure 4.5: (a) Undercoupled probe transmission spectra recorded at low (P
d
= 0 mW (blue)) and
high (P
d
= 0.20 mW (red)) pump power (timeaveraged trace in black). (b) Time waveforms of the
probe transmission for sinusoidally modulated (22.3 MHz; vertical dashed line in (d)) pump mode
with modulation depths of 1.9% (blue), 14.9% (green), and 20.5% (red) at average P
d
= 0.85 mW.
(c) Pulsed modulation of the pump (top) and corresponding probe response (bottom). The fractional
modulation for the pump is deﬁned relative to the average dropped power, while that for the probe is
deﬁned relative to the onresonance probemode coupling depth. (d) Normalized probe modulation
spectra for P
d
= 14, 110, 210, 430, 850 µW. The dashed black curves show the corresponding
model. (e) Probe modulation spectrum (P
d
= 0.85 mW) shown over a wide frequency span. The
green curve shows the modeled response including only the dominant ﬂapping mechanical mode
(the red curve includes other, breathinglike, mechanical resonances). The orange curve shows the
measured noise ﬂoor.
86
Optical control of the dynamic response is most clearly demonstrated through the pumpprobe
modulation response of the spiderweb structure. In general, the pump wave inside the cavity satisﬁes
the following equation:
da
p
dt
= (i∆
p
Γ
tp
2
)a
p
ig
OM
xa
p
+iγa
p
2
a
p
+i
_
Γ
ep
A
p
, (4.8)
where a
p
and A
p
are the intracavity and input ﬁeld of the pump wave, respectively, normalized such
that U
p
a
p
2
and P
p
A
p
2
represent the intracavity energy and input power. ∆
p
= ω
p
ω
0p
represents the detuning of pump frequency ω
p
to the cavity resonance ω
0p
and Γ
tp
is the photon
decay rate of the loaded cavity for the pump mode. In Eq. (4.8), the third term represents the back
action of mechanical motion on the cavity resonance, where g
OM
is the optomechanical coupling
coefﬁcient and x(t) is the mechanical displacement of the cavity structure. The fourth term describes
the selfphase modulation introduced by the Kerr nonlinearity, where the nonlinear parameter γ =
cω
p
n
2
n
2
0
V
eff
, n
2
= 2.6 10
20
m
2
/W is the Kerr nonlinear coefﬁcient of silica, n
0
= 1.44 is the silica
refractive index, and V
eff
= 370 µm
2
(from FEM simulation) is the effective mode volume [126–
128]. However, compared with the dominant optomechanical effect, the selfphase modulation on
the pump wave is negligible in the spiderweb ring resonator. The ﬁnal term in Eq. (4.8) represents
the external ﬁeld coupling with a photon escape rate of Γ
ep
.
Assume that the input pump wave consists of an intense continuous wave together with a small
timevarying modulation, A
p
= A
p0
+δA
p
(t). The intracavity ﬁeld can be written as a
p
= a
p0
+
δa
p
(t), governed by the following equations:
da
p0
dt
= (i∆
p
Γ
tp
2
)a
p0
+i
_
Γ
ep
A
p0
, (4.9)
dδa
p
dt
= (i∆
p
Γ
tp
2
)δa
p
ig
OM
xa
p0
+i
_
Γ
ep
δA
p
, (4.10)
where we have neglected the negligible selfphase modulation for the pump wave. Equation (4.9)
provides a steadystate solution of
a
p0
=
i
_
Γ
ep
A
p0
Γ
tp
/2 i∆
p
, (4.11)
from which we obtain the average pump power dropped into the cavity, P
pd
, given by
P
pd
=
P
p0
Γ
0p
Γ
ep
∆
2
p
+(Γ
tp
/2)
2
, (4.12)
87
where P
p0
= A
p0
2
is the averaged input pump power and Γ
0p
is the intrinsic photon decay rate of
the pump mode. Clearly, to the zeroth order, the relative magnitude of the dropped pump power
modulation is directly equal to that of the input modulation:
δP
pd
(t)
P
pd
=
δP
p
(t)
P
p0
, (4.13)
where δP
p
= A
p0
δA
p
+A
p0
δA
p
is the timevarying component of the input pump power.
Eq. (4.10) leads to a pumpﬁeld modulation in the frequency domain of
δ¯ a
p
(Ω) =
ig
OM
a
p0
¯ x(Ω) i
_
Γ
ep
δ
¯
A
p
(Ω)
i(∆
p
+Ω) Γ
tp
/2
, (4.14)
where δ¯ a
p
(Ω), ¯ x(Ω), and δ
¯
A
p
(Ω) are Fourier transforms of δa
p
(t), x(t), and δA
p
(t), respectively,
deﬁned as
¯
B(Ω) =
+∞
∞
B(t)e
iΩt
dt. Physically, the ﬁrst term in Eq. (4.14) represents the perturba
tion induced by the mechanical motion, while the second term represents the effect of direct input
modulation.
The optical gradient force is linearly proportional to the cavity energy as F
o
=
g
OM
U
p
ω
p
. With
modulation of the pump energy, the gradient force thus consists of two terms, F
o
= F
o0
+δF
o
(t),
where F
o0
=
g
OM
U
p0
ω
p
is the static force component introduced by the averaged pump energy U
p0
=
a
p0
2
, and δF
o
(t) is the dynamic component related to the pump energy modulation δU
p
(t), given
by
δF
o
(t) =
g
OM
δU
p
ω
p
=
g
OM
ω
p
_
a
p0
δa
p
(t) +a
p0
δa
p
(t)
¸
. (4.15)
Substituting Eq. (4.14) into Eq. (4.15), we ﬁnd the force modulation is described by this general
form in the frequency domain:
δ
¯
F
0
(Ω) = f
o
(Ω)¯ x(Ω) +
i
_
Γ
ep
g
OM
ω
p
_
a
p0
δ
¯
A
p
(Ω)
i(∆
p
+Ω) Γ
tp
/2
+
a
p0
δ
¯
A
p
( Ω)
i(∆
p
Ω) +Γ
tp
/2
_
, (4.16)
where the ﬁrst term represents the back action introduced by the mechanical motion, with a spectral
response f
0
(Ω) given by
f
o
(Ω)
2g
2
OM
a
p0
2
∆
p
ω
p
∆
2
p
Ω
2
+(Γ
tp
/2)
2
+iΓ
tp
Ω
_
(∆
p
+Ω)
2
+(Γ
tp
/2)
2
¸_
(∆
p
Ω)
2
+(Γ
tp
/2)
2
¸. (4.17)
Figure 4.5(d) shows the spectral response of a probe resonance to smallsignal sinusoidal pump
88
modulation for several different (average) pump dropped powers. When the pump dropped power
is low, the pump backaction on mechanical motion is negligible and the probe response is given
by a combination of the intrinsic mechanical stiffness and the squeezeﬁlm effect [84] of trapped
gas in between the rings. When the pump power is increased, however, the mechanical resonance
frequency increases correspondingly, reaching a value of 22.3 MHz at a dropped power of 0.85 mW.
This value is about 32 times larger than the intrinsic mechanical frequency, and implies a dynamical
stiffness more than 1000 times that of the silica rings.
The spiderweb ring resonators are separated by a 150 nm gap, which is only about 2.2 times
the mean free path in a nitrogen environment ( 68 nm). As the ring is 6.3 µm wide, much
larger than the ring gap, the nitrogen gas sandwiched in the gap is highly conﬁned by the two silica
layers and cannot move freely during the ﬂapping motion of the two rings. The resulting signiﬁcant
pressure differential between the internal and external regions of the paired silica rings functions as a
viscous force to damp the mechanical motion. This phenomenon is wellknown as the squeezeﬁlm
effect, which has a profound impact on the dynamic response of micro/nanomechanical systems
[84]. Apart from the optical gradient force, the squeezeﬁlm effect is the dominant mechanism
responsible for the dynamic mechanical response of our devices. The associated damping force
can be described by a general form of
¯
F
sq
(Ω) = f
sq
(Ω)¯ x(Ω), where f
sq
(Ω) represents the spectral
response of the squeeze ﬁlm.
In general, the squeezeﬁlm effect is typically described by two theories which work in quite
different regimes, depending on the Knudsen number K
n
characterizing the ratio between the mean
free path and the gap [84]. In the classical regime with K
n
1 where the gas can be considered
a continuum, the squeezeﬁlm viscous force for a rectangular plate is well described by f
sq
(Ω) =
k
e
(Ω)+iC
d
(Ω), where k
e
and C
d
represent the spring constant and damping, respectively, induced
by the squeeze ﬁlm. They are given by the following equations [129]
k
e
(Ω) =
64σ
2
P
a
L
0
W
0
π
8
h
0
∑
m,n odd
1
m
2
n
2
[(m
2
+(n/η)
2
)
2
+σ
2
/π
4
]
, (4.18)
C
d
(Ω) =
64σP
a
L
0
W
0
π
6
h
0
∑
m,n odd
m
2
+(n/η)
2
m
2
n
2
[(m
2
+(n/η)
2
)
2
+σ
2
/π
4
]
, (4.19)
where P
a
is the ambient gas pressure, W
0
and L
0
are the width and length of the plate, h
0
is the gap,
89
η = L
0
/W
0
is the aspect ratio of the plate, and σ is the squeeze number given by
σ(Ω) =
12µ
eff
W
2
0
Ω
P
a
h
2
0
, (4.20)
where µ
eff
= µ/(1 +9.638K
1.159
n
) is the effective value of the viscosity coefﬁcient µ [130]. Under
this model, the squeeze ﬁlm functions primarily as a damping (or elastic) force when the modulation
frequency is below (or above) the cutoff frequency given by
Ω
c
=
π
2
P
a
h
2
0
12µ
eff
_
1
W
2
0
+
1
L
2
0
_
. (4.21)
In contrast, in the freemolecule regime with K
n
1 where the interaction between gas molecules
is negligible, the squeeze ﬁlm approximately behaves like a damping force, f
sq
(Ω) = iC
r
Ω, with C
r
given by the following equation [131, 132]
C
r
=
_
S
16πh
0
_
4P
a
L
0
W
0
¸
2M
m
πR T
, (4.22)
where M
m
is the molar mass of gas, T is the temperature, R is the ideal gas constant, S is the
perimeter length of the gap region.
However, our devices have a Knudsen number of K
n
= 0.45, falling in the crossover regime
where neither theory adequately describes the squeezeﬁlm effect [133]. As the device works in
the regime between the continuum and freemolecule limit, we heuristically propose that the damp
ing/elastic force of the squeeze ﬁlm is effectively described by a composite of the two theories:
f
sq
(Ω) = k
e
(Ω) +iC
d
(Ω) +iη
r
C
r
Ω, (4.23)
with a modiﬁed effective coefﬁcient of viscosity µ
eff
=η
µ
µ
eff
, where η
r
and η
µ
are parameters used
for a best description of the squeezeﬁlm response in our devices. Detailed analysis shows that
η
µ
= 0.7 and η
r
= 0.03 provides the best ﬁt for our devices. As our devices have a spiderweb
geometry, we approximate it with an equivalent rectangular shape with W
0
given by the ring width,
L
0
given by the circumference at the ring center, and S 2L
0
. As shown by the experimental results
and theoretical ﬁts, this model provides an accurate description of the squeezeﬁlm effect in our
devices.
Although the intrinsic mechanical frequency of the 54 µm spiderweb structure is 694 kHz (in
90
dicated by FEM simulation), Fig. 4.5(d) shows a minimum dynamic frequency response of 6 MHz,
dominated by the squeezeﬁlm damping. Interestingly, although squeezeﬁlm damping is generally
detrimental in other micro/nanomechanical systems [84, 134], it is beneﬁcial in this case, as it helps
to extend the modulation bandwidth for wavelength routing.
With the optical gradient force and the squeezeﬁlm damping force, the mechanical motion of
the cavity satisﬁes the following equation:
d
2
x
dt
2
+Γ
m
dx
dt
+Ω
2
m
x =
1
m
eff
(F
o
+F
sq
+F
T
) =
1
m
eff
(F
o0
+δF
o
+F
sq
+F
T
), (4.24)
where m
eff
is the effective motional mass of the ﬂapping mechanical mode, and Ω
m
and Γ
m
are
intrinsic mechanical frequency and damping rate, respectively. F
T
is the thermal Langevin force
responsible for the thermal Brownian motion, a Markovin process with the following correlation
function:
F
T
(t)F
T
(t +τ)= 2m
eff
Γ
m
k
B
Tδ(τ), (4.25)
where k
B
is Boltzmann’s constant.
As the squeezeﬁlm viscous force is zero at Ω = 0, the squeeze gas ﬁlm impacts only the dy
namic response of mechanical motion. Equation (4.24) shows clearly that the static mechanical
displacement is actuated only by the static component of the optical force given by
x
0
=
F
o0
m
eff
Ω
2
m
=
g
om
U
p0
k
m
ω
p
=
g
om
P
pd
k
m
ω
p
Γ
0p
, (4.26)
where k
m
= m
eff
Ω
2
m
is the intrinsic spring constant of the spiderweb structure. With a speciﬁcally
designed extremely small spring constant, x
0
can be quite signiﬁcant for a given dropped power.
As a result, the cavity resonance can be tuned by a signiﬁcant magnitude of g
OM
x
0
. This is the
primary mechanism responsible for the resonance tuning. On the other hand, this static mechanical
displacement primarily changes the equilibrium position of the mechanical motion. It is convenient
to remove this component in Eq. (4.24) by deﬁning x
=x x
0
, since both the squeezeﬁlm damping
force and dynamic component of the optical force affect only the dynamics of x
.
Substituting Eqs. (4.16), (4.17), (4.23) into Eq. (4.24) in the frequency domain, we ﬁnd that the
squeezeﬁlm damping force and the backaction term of the optical force primarily change the values
91
of the resonant frequency and damping rate of the mechanical motions. Deﬁning
L(Ω) Ω
2
m
Ω
2
iΓ
m
Ω
f
o
(Ω)
m
eff
f
sq
(Ω)
m
eff
, (4.27)
the mechanical displacement is thus given by
¯ x(Ω) =
¯
F
T
(Ω)
m
eff
L(Ω)
+
i
_
Γ
ep
g
OM
m
eff
ω
p
L(Ω)
_
a
p0
δ
¯
A
p
(Ω)
i(∆
p
+Ω) Γ
tp
/2
+
a
p0
δ
¯
A
p
( Ω)
i(∆
p
Ω) +Γ
tp
/2
_
, (4.28)
where we have dropped the prime notation of x
for simplicity.
The ﬁrst term in Eq. (4.28) represents the thermal Brownian motion while the second term
describes the motions actuated by the pump modulation. In the absence of pump modulation, the
mechanical motion is dominated by the Brownian motion. By using Eq. (4.25), we ﬁnd the spectral
density of thermal mechanical displacement has the form
S
x
(Ω) =
2Γ
m
k
B
T
m
eff
L(Ω)
2
. (4.29)
Equations (4.16), (4.17), and (4.27) show that one dominant effect of the pump energy inside the
cavity is to increase the mechanical rigidity, the socalled optical spring effect. In most cases, L(Ω)
can be well approximated by L(Ω) (Ω
m
)
2
Ω
2
iΓ
m
Ω with a new mechanical resonance Ω
m
and damping rate Γ
m
affected by the optical force. Equation (4.29) thus leads to a variance of the
thermal mechanical displacement given by
(δx)
2
=
1
2π
+∞
∞
S
x
(Ω)dΩ =
k
B
TΓ
m
k
m
Γ
m
k
B
T
k
m
, (4.30)
where k
m
= m
eff
(Ω
m
)
2
is the effective spring constant and the approximation in the ﬁnal term as
sumes a negligible change in the mechanical linewidth. Clearly, the increase of the mechanical
resonance frequency through the optical spring effect dramatically suppresses the magnitude of
the thermal mechanical displacement and its perturbation of the cavity resonance, as shown in
Fig. 4.5(a). In the presence of pump modulation, the mechanical motion is primarily dominated
by the dynamic optical force rather than the actuation from the thermal Langevin force, and the
ﬁrst term is negligible compared with the second term in Eq. (4.28). Thus, we neglect the thermal
Brownian term in the following discussion.
92
The probe wave inside the cavity is governed by a dynamic equation similar to Eq. (4.8):
da
s
dt
= (i∆
s
Γ
ts
2
)a
s
ig
OM
xa
s
+2iγa
p
2
a
s
+i
_
Γ
es
A
s
, (4.31)
except that the Kerrnonlinear term now describes the crossphase modulation from the pump wave.
With the perturbations induced by the pump modulation, similar to the previous discussion of the
pump wave, the intracavity probe ﬁeld can be written as a
s
=a
s0
+δa
s
(t), governed by the following
equations:
da
s0
dt
= (i∆
s
Γ
ts
2
)a
s0
+2iγU
p0
a
s0
+i
_
Γ
es
A
s
, (4.32)
dδa
s
dt
= (i∆
s
Γ
ts
2
)δa
s
+2iγU
p0
δa
s
ig
OM
xa
s0
+2iγδU
p
a
s0
, (4.33)
where we have assumed the probe input is a continuous wave with a power of P
s
= A
s
2
. The
second terms of Eqs. (4.32) and (4.33) represent the static cavity tuning introduced by crossphase
modulation, which can be included in the cavity tuning term ∆
s
for simplicity. In general, it is
negligible compared with the cavity linewidth at the power level used for exciting optomechanical
effects, leading to 2γU
p0
Γ
tp
, Γ
ts
.
Equation (4.32) provides a steadystate solution of
a
s0
=
i
Γ
es
A
s
Γ
ts
/2 i∆
s
, (4.34)
and Eq. (4.33) results in a probeﬁeld modulation in the frequency domain of
δ¯ a
s
(Ω) =
ia
s0
_
g
OM
¯ x(Ω) 2γδ
¯
U
p
(Ω)
_
i(∆
s
+Ω) Γ
ts
/2
, (4.35)
where δ
¯
U
p
(Ω) is the Fourier transform of δU
p
(t). As the transmitted ﬁeld of the probe is given by
A
Ts
= A
s
+i
Γ
es
a
s
, the modulation of the transmitted probe power thus takes the form
δP
Ts
= i
_
Γ
es
(A
0s
δa
s
A
0s
δa
s
), (4.36)
where A
0s
= A
s
+i
Γ
es
a
s0
is the transmitted probe wave in the absence of modulation. By use
of Eqs. (4.14), (4.28), (4.34) and (4.35), we ﬁnd that the power spectrum of the transmitted probe
93
modulation is given by the following equation:
δ
¯
P
Ts
(Ω)
2
P
2
s
=
¸
¸
¸
¸
g
2
OM
m
eff
ω
p
L(Ω)
+2γ
¸
¸
¸
¸
2
P
2
pd
Γ
2
0p
δ
¯
P
pd
(Ω)
2
P
2
pd
4Γ
2
es
Γ
2
0s
∆
2
s
[∆
2
s
+(Γ
ts
/2)
2
]
4
, (4.37)
where Γ
0s
is the intrinsic photon decay rate of the probe mode, and δ
¯
P
Ts
(Ω) and δ
¯
P
pd
(Ω) are
the Fourier transforms of δP
Ts
(t) and δP
pd
(t), respectively. To obtain Eq. (4.37), we have used
Eq. (4.13) to relate the dropped pump power to the input, and have also taken into account the
fact that the Kerr effect is relatively small, such that 2γU
p0
Γ
tp
. We also assume the cavity is in
the sidebandunresolved regime with Ω
m
Γ
tp
, Γ
ts
. The modulation spectra given in Fig. 4.5 are
deﬁned as
ρ(Ω)
δ
¯
P
Ts
(Ω)
2
/P
2
s
δ
¯
P
pd
(Ω)
2
/P
2
pd
. (4.38)
For a better comparison of the dynamicbackaction induced variations on the probe modulation, the
modulation spectra shown in Fig. 4.5(d) are normalized by a factor corresponding to the ratio of
the dropped power for each curve relative to the maximum dropped power. Therefore, the plotted
modulation spectra are given by
ρ
(Ω) ρ(Ω)
P
2
pd0
P
2
pd
=
δ
¯
P
Ts
(Ω)
2
/P
2
s
δ
¯
P
pd
(Ω)
2
/P
2
pd0
, (4.39)
where P
pd0
= 0.85 mW is the maximum drop power used in Fig. 4.5(d).
The derivations above take into account only the ﬂapping mechanical mode, since it is most
strongly actuated by the optical gradient force. In general, there are many mechanical resonances
for the spiderweb resonators, but weakly coupled to the optical waves inside the cavity. In this
case, following the same procedure above, it is easy to show that the spectral response of probe
modulation now becomes
ρ(Ω) =
¸
¸
¸
¸
¸
2γ +
∑
j
g
2
j
m
j
ω
p
L
j
(Ω)
¸
¸
¸
¸
¸
2
P
2
pd
Γ
2
0p
4Γ
2
es
Γ
2
0s
∆
2
s
[∆
2
s
+(Γ
ts
/2)
2
]
4
, (4.40)
where g
j
, m
j
, and L
j
(Ω) are optomechanical coupling coefﬁcient, effective motional mass, and
the spectral response of mechanical motions, respectively, for the j
th
mechanical mode. For those
weakly actuated mechanical modes, L
j
(Ω) = Ω
2
mj
Ω
2
iΓ
mj
Ω where Ω
mj
and Γ
mj
are the reso
nance frequency and damping rate of the j
th
mechanical mode. Equation (4.40) was used to describe
94
the modulation spectrum shown in Fig. 4.5(d).
As shown in the dashed curves in Fig. 4.5(d), eq. (4.37) provides an accurate description of
the pumpprobe modulation response (the Fanolike resonance seen at P
d
= 0.85 mW is due to
intrinsic mechanical coupling between different types of motion, and is discussed in Chapter 3). In
general, the small modulation of the pump wave which actuates the mechanical oscillation is greatly
magniﬁed on the probe resonance. Figure 4.5(d) shows that for P
d
= 0.85 mW there is a resonant
modulation “gain” of greater than 20 dB. This can also be seen in the time waveform of the probe in
Fig. 4.5(b), where a 1.9% modulation of the pump power is large enough to introduce considerable
fractional modulation in the probe time waveform (top panel). Increasing the pump modulation to
14.9% (Fig. 4.5(b), middle panel) results in a probe modulation of larger than a halflinewidth (full
contrast modulation). Further increase in the pump modulation depth actuates ﬂapping mechanical
motion so intense it begins to excite a second mechanical mode (the Fanolike feature in Fig. 4.5(d)),
resulting in a beat signal with a period of 0.36 µs on the probe time waveform.
One metric for characterizing the response time of the spiderweb optomechanical cavity is the
resonant oscillation period [120]. Figure 4.5(d) shows that the optical spring effect enables a modu
lation time as fast as 44.8 ns. This can be further enhanced by using the transduction “gain” to push
the probe modulation into the nonlinear regime, where in the lower panel of Fig. 4.5(b) the probe
wavelength (10%–90%) onoff switching time is reduced to 7 ns, roughly 3 orders of magnitude
faster than modulation schemes based upon thermooptic, optoﬂuidic, photochemical, or micro
electricalmechancial approaches [108–111, 113–115, 117, 120]. For many switching applications,
however, one is more interested in the impulse response of the system. The pulsed response of the
probe is shown in Fig. 4.5(c). As is common in micro/nanomechanical systems [114, 120], the res
onant response causes ringing during switching, with a settling time determined by the mechanical
linewidth. The measured settling time constant of the probe response is 196 ns, consistent with the
mechanical linewidth of 23 MHz (see Fig. 4.5(d)).
In addition to the optomechanical nonlinearity, other optical (material, etc.) nonlinearities
can also contribute to the probe modulation. As shown in the expanded modulation spectrum of
Fig. 4.5(e), the resonant optomechanical nonlinearity is dominant out to a frequency of 500 MHz,
after which the response plateaus due to the ultrafast Kerr nonlinearity of silica. The Kerr nonlin
earity is measured to be 78 dB below the resonant optomechanical response. This ratio agrees well
with the theoretical value of 81 dB given by
_
g
2
OM
2γm
eff
ω
p
Ω
m
Γ
m
_
2
, where Ω
m
and Γ
m
are the effective
mechanical resonance frequency and damping rate, respectively. The Kerr nonlinearity in silica has
95
been extensively studied over more than three decades for optical signal processing [126, 135, 136],
and the excellent agreement between the theoretical and experimental spectra provides yet another
indication that the optical gradient force is the dominant tuning mechanism in the spiderweb cavity
structure.
4.5 Discusssion
The versatility of the gradient optical force tuning approach described here provides considerable
room for future improvement of device performance. An increase in the tuning range and efﬁciency
(actuation power) can be expected with further engineering of the mechanical stability of the spi
derweb structure. For example, the 90 µm diameter ﬁrstorder spiderweb cavities should allow for
a sixfold increase in tuning efﬁciency to approximately 15 nm/mW. There are also many well
established methods for managing the dynamical response, in particular the ringing, of resonant
micro and nanomechanical systems [114, 120]. In contrast to cavityoptomechanical applications
such as cooling and ampliﬁcation of mechanical motion [50, 125], a reduction in the mechanical Q
factor, which can be obtained through elevated gas pressure or incorporation of damping materials,
is sought to improve the switching time. Given the similarity of the doublering spiderweb stucture
to other more conventional planar microring technologies, one can also incorporate other chipbased
optical components such as waveguides, lasers, and modulators to enable full control and function
ality of the optomechanics. One example technology would be an onchip reconﬁgurable optical
add/drop multiplexer or wavelength selective switch/crossconnect, which could be accomplished
by integrating an array of doublering cavities into a parallel or cascaded conﬁguration. In addition
to the demonstrated wavelength routing, other prospective applications for optomechanical devices
include tunable optical buffering [137], dispersion compensation [138], tunable lasers [139], and
nonlinear signal processing [126].
96
Chapter 5
Conclusion
Here has been presented work on several optical resonator systems: the single and doublemetal
plasmonic photonic crystal resonator, the doubledisk whisperinggallery cavity, and the double
ring spiderweb cavity. Each of these resonator designs was developed and optimized for a particular
range of applications, and each has been shown to be effective in at least initial demonstrations.
Multispectral midinfrared resonant detectors were demonstrated with enhanced responsivity
and detectivity, and tailorable polarization and spectral sensitivity. These devices were fabricated
using a very simple singleetch process, and are detector agnostic, with design principles that are
easily transferrable to any other detector material or frequency range with a minimum of difﬁculty.
Expanding to a doublemetal device structure, this method could be used to easily and inexpensively
impart frequency and polarization selectivity, as well as absorption enhancement, to current detector
focal plane array processing.
A novel optomechanical device structure has also been presented, consisting of two stacked
microdisks with an optically narrow gap between them. This device has a very large optomechanical
coupling and a high quality factor, giving rise to extremely large dynamical backaction in the form
of both regenerative mechanical oscillation and optomechanical cooling. Due to the large optical
spring effect in these structures, we also demonstrate tunable coherent mechanical mode mixing
with an analogy to electromagnetically induced transparency, showing the possibility for slowlight
effects on the very long phononic timescale, and the potential for phononphoton quantum state
transfer.
Finally, an extremely ﬂexible doublering optomechanical device is shown, demonstrating all
optical wavelength routing with unprecedented range and efﬁciency, and 100% channel quality
preservation. As this device can be easily integrated onchip, it shows great promise for optical
communications applications, as well as for more fundamental physicsbased applications such as
97
in cavity quantum electrodynamics or for dispersion compensation in nonlinear optics.
98
Bibliography
[1] A. Rogalski, J. Antoszewski, and L. Faraone, “Thirdgeneration infrared photodetector ar
rays,” Journal of Applied Physics 105, 091101 (2009).
[2] J. Phillips, “Evaluation of the fundamental properties of quantum dot infrared detectors,”
Journal of Applied Physics 91, 4590–4594 (2002).
[3] Y. Sidorov, “Peculiarities of the MBE growth physics and technology of narrowgap IIVI
compounds,” Thin Solid Films 306, 253–265 (1997).
[4] S. Krishna, “Quantum dotsinawell infrared photodetectors,” Journal of Physics D: Applied
Physics 38, 2142–2150 (2005).
[5] R. V. Shenoi, R. S. Attaluri, A. Siroya, J. Shao, Y. D. Sharma, A. Stintz, T. E. Vandervelde,
and S. Krishna, “Lowstrain InAs/InGaAs/GaAs quantum dotsinawell infrared photode
tector,” Journal of Vacuum Science and Technology B 26, 1136–1139 (2008).
[6] O. Painter and K. Srinivasan, “Polarization properties of dipolelike defect modes in photonic
crystal nanocavities,” Optics Letters 27, 339–341 (2002).
[7] M. Lon car, T. Yoshie, A. Scherer, P. Gogna, and Y. Qiu, “Lowthreshold photonic crystal
laser,” Applied Physics Letters 81, 2680–2682 (2002).
[8] O. Painter, J. Vu ckovi c, and A. Scherer, “Defect modes of a twodimensional photonic crystal
in an optically thin dielectric slab,” Journal of the Optical Society of America B: Optical
Physics 16, 275–285 (1999).
[9] M. L. Althouse and C. I. Chang, “Chemical vapor detection with a multispectral thermal
imager,” Optical Engineering 30, 1725–1733 (1991).
[10] H. Raether, Surface Plasmons on Smooth and Rough Surfaces and on Gratings (Springer
Tracts in Modern Physics) (Springer, Berlin, 1988).
99
[11] B. Prade, J. Y. Vinet, and A. Mysyrowicz, “Guided optical waves in planar heterostructures
with negative dielectric constant,” Physical Review B: Condensed Matter 44, 13556–13572
(1991).
[12] M. Bahriz, V. Moreau, R. Colombelli, O. Crisafulli, and O. Painter, “Design of midIR and
THz quantum cascade laser cavities with complete TM photonic bandgap,” Optics Express
15, 5948–5965 (2007).
[13] J. Rosenberg, R. V. Shenoi, T. E. Vandervelde, S. Krishna, and O. Painter, “A multispectral
and polarizationselective surfaceplasmon resonant midinfrared detector,” Applied Physics
Letters 95, 161101 (2009).
[14] E. Homeyer, J. Houel, X. Checoury, F. Delgehier, S. Sauvage, P. Boucaud, R. Braive,
L. Le Gratiet, L. Leroy, A. Miard, A. Lemaˆtre, and I. Sagnes, “Resonant coupling of quan
tum dot intersublevel transitions with midinfrared photonic crystal modes,” Applied Physics
Letters 95, 041108 (2009).
[15] E. Laux, C. Genet, T. Skauli, and T. W. Ebbesen, “Plasmonic photon sorters for spectral and
polarimetric imaging,” Nature Photonics 2, 161–164 (2008).
[16] J. K. Yang, M. K. Seo, I. K. Hwang, S. B. Kim, and Y. H. Lee, “Polarizationselective reso
nant photonic crystal photodetector,” Applied Physics Letters 93, 211103 (2008).
[17] X. Hu, M. Li, Z. Ye, W. Y. Leung, K. M. Ho, and S. Y. Lin, “Design of midinfrared photode
tectors enhanced by resonant cavities with subwavelength metallic gratings,” Applied Physics
Letters 93, 241108 (2008).
[18] J. S. White, G. Veronis, Z. Yu, E. S. Barnard, A. Chandran, S. Fan, and M. L. Brongersma,
“Extraordinary optical absorption through subwavelength slits,” Optics Letters 34, 686–688
(2009).
[19] R. V. Shenoi, D. A. Ramirez, Y. Sharma, R. S. Attaluri, J. Rosenberg, O. J. Painter, and
S. Krishna, “Plasmon assisted photonic crystal quantum dot sensors,” vol. 6713, p. 67130P
(SPIE, 2007).
[20] J. Rosenberg, R. V. Shenoi, S. Krishna, and O. Painter, “Design of plasmonic photonic crystal
resonant cavities for polarization sensitive infrared photodetectors,” Opt. Express 18, 3672–
3686 (2010).
100
[21] M. Tinkham, Group Theory and Quantum Mechanics (Dover Publications, New York, NY,
2003).
[22] O. Painter, K. Srinivasan, and P. E. Barclay, “Wannierlike equation for the resonant cavity
modes of locally perturbed photonic crystals,” Physical Review B: Condensed Matter 68,
035214 (2003).
[23] J. D. Joannopoulos, R. D. Meade, and J. N. Winn, Photonic Crystals (Princeton University
Press, Princeton, New Jersey, 1995).
[24] W. L. Barnes, A. Dereux, and T. W. Ebbesen, “Surface plasmon subwavelength optics,” Na
ture 424, 824–830 (2003).
[25] J. B. Pendry, L. MartinMoreno, and F. J. GarciaVidal, “Mimicking Surface Plasmons with
Structured Surfaces,” Science 305, 847–848 (2004).
[26] P. B. Johnson and R. W. Christy, “Optical Constants of the Noble Metals,” Physical Review
B: Condensed Matter 6, 4370–4379 (1972).
[27] M. Cai, O. Painter, and K. J. Vahala, “Observation of Critical Coupling in a Fiber Taper to
a SilicaMicrosphere WhisperingGallery Mode System,” Physical Review Letters 85, 74–77
(2000).
[28] T. E. Vandervelde, M. C. Lenz, E. Varley, A. Barve, J. Shao, R. V. Shenoi, D. A. Ramirez,
W. Jan, Y. D. Sharma, and S. Krishna, “Quantum DotsinaWell Focal Plane Arrays,” IEEE
Journal of Selected Topics in Quantum Electronics 14, 1150–1161 (2008).
[29] V. B. Braginsky and A. B. Manukin, Measurements of Weak Forces in Physics Experiments
(The University of Chicago Press, Chicago, IL, 1977).
[30] C. Caves, K. S. Thorne, R. W. P. Drever, V. D. Sandberg, and M. Zimmermann, “On the
measurement of a weak classical force coupled to a quantummechanical oscillator,” Reviews
of Modern Physics 52, 341–392 (1980).
[31] A. Abramovici, W. E. Althouse, R. W. P. Drever, Y. Gursel, Seiji, D. Shoemaker, L. Sievers,
R. E. Spero, K. S. Thorne, R. E. Voigt, R. Weiss, S. E. Whitcomb, and M. E. Zucker, “LIGO:
The Laser Interferometer GravitationalWave Observatory,” Science 256, 325 (1992).
101
[32] C. A. Regal, J. D. Tuefel, and K. W. Lehnert, “Measuring nanomechanical motion with a
microwave cavity interferometer,” Nature Physics 4, 555–560 (2008).
[33] V. B. Braginsky, F. Y. Khalili, and K. S. Thorne, Quantum Measurement (Cambridge Univer
sity Press, Cambridge, UK, 1995).
[34] T. J. Kippenberg and K. J. Vahala, “Cavity Optomechanics: BackAction at the Mesoscale,”
Science 321, 1172–1176 (2008).
[35] T. J. Kippenberg, H. Rokhsari, T. Carmon, A. Scherer, and K. J. Vahala, “Analysis of
RadiationPressure Induced Mechanical Oscillation of an Optical Microcavity,” Physical Re
view Letters 95, 033901 (2005).
[36] S. Gigan, H. R. Bohm, M. Paternostro, F. Blaser, G. Langer, J. B. Hertzberg, K. C. Schwab,
D. Bauerle, M. Aspelmeyer, and A. Zeilinger, “Selfcooling of a micromirror by radiation
pressure.” Nature 444, 67–70 (2006).
[37] O. Arcizet, P. F. Cohadon, T. Briant, M. Pinard, and A. Heidmann, “Radiationpressure cool
ing and optomechanical instability of a micromirror,” Nature 444, 71–73 (2006).
[38] D. Kleckner and D. Bouwmeester, “Subkelvin optical cooling of a micromechanical res
onator,” Nature 444, 75–78 (2006).
[39] A. Schliesser, P. Del’haye, N. Nooshi, K. J. Vahala, and T. J. Kippenberg, “Radiation Pressure
Cooling of a Micromechanical Oscillator Using Dynamical Backaction,” Physical Review
Letters 97, 243905 (2006).
[40] J. D. Thompson, B. M. Zwickl, A. M. Jayich, F. Marquardt, S. M. Girvin, and J. G. E. Har
ris, “Strong dispersive coupling of a highﬁnesse cavity to a micromechanical membrane,”
Nature 452, 72–75 (2008).
[41] A. Schliesser, R. Riviere, G. Anetsberger, O. Arcizet, and T. J. Kippenberg, “Resolved
sideband cooling of a micromechanical oscillator,” Nature Physics 4, 415–419 (2008).
[42] M. L. Povinelli, S. G. Johnson, M. Loncar, M. Ibanescu, E. J. Smythe, F. Capasso, and
J. D. Joannopoulos, “HighQ enhancement of attractive and repulsive optical forces between
coupled whisperinggallerymode resonators,” Optics Express 13, 8287–8295 (2005).
102
[43] P. T. Rakich, M. A. Popovic, M. Soljacic, and E. P. Ippen, “Trapping, corralling and spectral
bonding of optical resonances through optically induced potentials,” Nature Photonics 1,
658–665 (2007).
[44] M. Li, W. H. P. Pernice, C. Xiong, T. BaehrJones, M. Hochberg, and H. X. Tang, “Harnessing
optical forces in integrated photonic circuits,” Nature 456, 480–484 (2008).
[45] M. Eichenﬁeld, R. Camacho, J. Chan, K. J. Vahala, and O. Painter, “A picogram and
nanometrescale photoniccrystal optomechanical cavity,” Nature 459, 550–555 (2009).
[46] J. K. H. Horber and M. J. Miles, “Scanning Probe Evolution in Biology,” Science 302, 1002–
1005 (2003).
[47] J. Arlett, “Properties of piezoresistive silicon nanoscale cantilevers with applications to
BioNEMS,” Ph.D. thesis, California Institute of Technology (2006).
[48] Q. Lin, J. Rosenberg, X. Jiang, K. J. Vahala, and O. Painter, “Mechanical Oscillation and
Cooling Actuated by the Optical Gradient Force,” Physical Review Letters 103, 103601
(2009).
[49] P. Meystre, E. M. Wright, J. D. Mccullen, and E. Vignes, “Theory of radiationpressure
driven interferometers,” Journal of the Optical Society of America B: Optical Physics 2,
1830–1840 (1985).
[50] T. J. Kippenberg and K. J. Vahala, “Cavity OptoMechanics,” Optics Express 15, 17172–
17205 (2007).
[51] G. Anetsberger, O. Arcizet, Q. P. Unterreithmeier, E. M. Weig, J. P. Kotthaus, and T. J. Kip
penberg, “Nearﬁeld cavity optomechanics with nanomechanical oscillators,” Nature Physics
5, 909–914 (2009).
[52] S. S. Verbridge, R. Illic, H. G. Craighead, and J. M. Parpia, “Size and frequency dependent
gas damping of nanomechanical resonators,” Applied Physics Letters 93, 013101 (2008).
[53] R. Kubo, A Stochastic Theory of Line Shape (John Wiley & Sons, New York, NY, 1969).
[54] H. J. Kimble and L. Mandel, “Resonance ﬂuorescence with excitation of ﬁnite bandwidth,”
Physical Review A: Atomic, Molecular, and Optical Physics 15, 689–699 (1977).
103
[55] B. S. Sheard, M. B. Gray, C. M. Mow Lowry, D. E. McClelland, and S. E. Whitcomb, “Ob
servation and characterization of an optical spring,” Physical Review A: Atomic, Molecular,
and Optical Physics 69, 051801 (2004).
[56] P. F. Cohadon, A. Heidmann, and M. Pinard, “Cooling of a Mirror by Radiation Pressure,”
Physical Review Letters 83, 3174–3177 (1999).
[57] F. Marquardt, J. P. Chen, A. A. Clerk, and S. M. Girvin, “Quantum theory of cavityassisted
sideband cooling of mechanical motion,” Physical Review Letters 99, 093902 (2007).
[58] I. W. Rae, N. Nooshi, W. Zwerger, and T. J. Kippenberg, “Theory of ground state cooling
of a mechanical oscillator using dynamical backaction,” Physical Review Letters 99, 093901
(2007).
[59] T. J. Kippenberg, J. Kalkman, A. Polman, and K. J. Vahala, “Demonstration of an erbium
doped microdisk laser on a silicon chip,” Physical Review A: Atomic, Molecular, and Optical
Physics 74, 051802 (2006).
[60] T. D. Stowe, K. Yasumura, T. W. Kenny, D. Botkin, K. Wago, and D. Rugar, “Attonew
ton force detection using ultrathin silicon cantilevers,” Applied Physics Letters 71, 288–290
(1997).
[61] M. Notomi, H. Taniyama, S. Mitsugi, and E. Kuramochi, “Optomechanical Wavelength and
Energy Conversion in HighQ DoubleLayer Cavities of Photonic Crystal Slabs,” Physical
Review Letters 97, 023903 (2006).
[62] U. Fano, “Effects of Conﬁguration Interaction on Intensities and Phase Shifts,” Physical
Review Online Archive (Prola) 124, 1866–1878 (1961).
[63] S. E. Harris, J. E. Field, and A. Imamo glu, “Nonlinear optical processes using electromag
netically induced transparency,” Physical Review Letters 64, 1107–1110 (1990).
[64] J. Faist, F. Capasso, C. Sirtori, K. W. West, and L. N. Pfeiffer, “Controlling the sign of
quantum interference by tunnelling from quantum wells,” Nature 390, 589–591 (1997).
[65] M. Kroner, A. O. Govorov, S. Remi, B. Biedermann, S. Seidl, A. Badolato, P. M. Petroff,
W. Zhang, R. Barbour, B. D. Gerardot, R. J. Warburton, and K. Karrai, “The nonlinear Fano
effect,” Nature 451, 311–314 (2008).
104
[66] J. F. Scott, “Softmode spectroscopy: Experimental studies of structural phase transitions,”
Reviews of Modern Physics 46, 83 (1974).
[67] M. Hase, J. Demsar, and M. Kitajima, “Photoinduced Fano resonance of coherent phonons
in zinc,” Physical Review B: Condensed Matter 74, 212301 (2006).
[68] S. E. Harris, “Lasers without inversion: Interference of lifetimebroadened resonances,”
Physical Review Letters 62, 1033–1036 (1989).
[69] D. E. Nikonov, A. Imamo glu, and M. O. Scully, “Fano interference of collective excitations in
semiconductor quantum wells and lasing without inversion,” Physical Review B: Condensed
Matter 59, 12212–12215 (1999).
[70] S. Fan, “Sharp asymmetric line shapes in sidecoupled waveguidecavity systems,” Applied
Physics Letters 80, 908–910 (2002).
[71] D. D. Smith, H. Chang, K. A. Fuller, A. T. Rosenberger, and R. W. Boyd, “Coupled
resonatorinduced transparency,” Physical Review A: Atomic, Molecular, and Optical Physics
69, 063804 (2004).
[72] Q. Xu, S. Sandhu, M. L. Povinelli, J. Shakya, S. Fan, and M. Lipson, “Experimental Real
ization of an OnChip AllOptical Analogue to Electromagnetically Induced Transparency,”
Physical Review Letters 96, 123901 (2006).
[73] K. Totsuka, N. Kobayashi, and M. Tomita, “SlowLight in CoupledResonatorInduced Trans
parency,” Physical Review Letters 98, 213904 (2007).
[74] N. Liu, L. Langguth, T. Weiss, J. Kastel, M. Fleischhauer, T. Pfau, and H. Giessen, “Plas
monic analogue of electromagnetically induced transparency at the Drude damping limit,”
Nature Materials 8, 758–762 (2009).
[75] Q. Lin, J. Rosenberg, D. Chang, R. Camacho, M. Eichenﬁeld, K. J. Vahala, and O. Painter,
“Coherent mixing of mechanical excitations in nanooptomechanical structures,” Nature
Photonics (2010). doi:10.1038/nphoton.2010.4.
[76] M. L. Povinelli, M. Loncar, M. Ibanescu, E. J. Smythe, S. G. Johnson, F. Capasso, and J. D.
Joannopoulos, “Evanescentwave bonding between optical waveguides,” Optics Letters 30,
3042–3044 (2005).
105
[77] M. Eichenﬁeld, C. P. Michael, R. Perahia, and O. Painter, “Actuation of micro
optomechanical systems via cavityenhanced optical dipole forces,” Nature Photonics 1,
416–422 (2007).
[78] J. Rosenberg, Q. Lin, and O. Painter, “Static and dynamic wavelength routing via the gradient
optical force,” Nature Photonics 3, 478–483 (2009).
[79] A. Ashkin, “History of optical trapping and manipulation of smallneutral particle, atoms,
and molecules,” IEEE Journal of Selected Topics in Quantum Electronics 6, 841–856 (2000).
[80] M. H. Zadeh and K. J. Vahala, “Observation of optical spring effect in a microtoroidal op
tomechanical resonator,” Optics Letters 32, 1611–1613 (2007).
[81] T. Corbitt, C. Wipf, T. Bodiya, D. Ottaway, D. Sigg, N. Smith, S. Whitcomb, and N. Maval
vala, “Optical Dilution and Feedback Cooling of a GramScale Oscillator to 6.9 mK,” Physi
cal Review Letters 99, 160801 (2007).
[82] J. Chan, M. Eichenﬁeld, R. Camacho, and O. Painter, “Optical and mechanical design of a
“zipper” photonic crystal optomechanical cavity,” Optics Express 17, 3802–3817 (2009).
[83] M. Fleischhauer, A. Imamoglu, and J. P. Marangos, “Electromagnetically induced trans
parency: Optics in coherent media,” Reviews of Modern Physics 77, 633–673 (2005).
[84] M. Bao and H. Yang, “Squeeze ﬁlm air damping in MEMS,” Sensors and Actuators A: Phys
ical 136, 3–27 (2007).
[85] A. S. Barker and J. J. Hopﬁeld, “Coupledopticalphononmode theory of the infrared disper
sion in BaTiO
3
, SrTiO
3
, and KTaO
3
,” Physical Review Online Archive (Prola) 135, A1732–
A1737 (1964).
[86] D. L. Rousseau and S. P. S. Porto, “AugerLike Resonant Interference in Raman Scatter
ing from One and TwoPhonon States of BaTiO
3
,” Physical Review Letters 20, 1354–1357
(1968).
[87] J. F. Scott, “Hybrid Phonons and Anharmonic Interactions in AlPO
4
,” Physical Review Let
ters 24, 1107–1110 (1970).
[88] A. Zawadowski and J. Ruvalds, “Indirect Coupling and Antiresonance of Two Optic
Phonons,” Physical Review Letters 24, 1111–1114 (1970).
106
[89] A. Chaves, R. S. Katiyar, and S. P. S. Porto, “Coupled modes with A1 symmetry in tetragonal
BaTiO
3
,” Physical Review B: Condensed Matter 10, 3522–3533 (1974).
[90] C. L. G. Alzar, M. A. G. Martinez, and P. Nussenzveig, “Classical analog of electromagneti
cally induced transparency,” American Journal of Physics 70, 37–41 (2002).
[91] P. R. Hemmer and M. G. Prentiss, “Coupledpendulum model of the stimulated resonance
Raman effect,” Journal of the Optical Society of America B: Optical Physics 5, 1613–1623
(1988).
[92] C. Liu, Z. Dutton, C. H. Behroozi, and L. V. Hau, “Observation of coherent optical informa
tion storage in an atomic medium using halted light pulses,” Nature 409, 490–493 (2001).
[93] M. Bajcsy, A. S. Zibrov, and M. D. Lukin, “Stationary pulses of light in an atomic medium,”
Nature 426, 638–641 (2003).
[94] M. Fatih and S. Fan, “Stopping Light All Optically,” Physical Review Letters 92, 083901
(2004).
[95] M. F. Yanik and S. Fan, “Stopping and storing light coherently,” Physical Review A: Atomic,
Molecular, and Optical Physics 71, 013803 (2005).
[96] G. Anetsberger, R. Riviere, A. Schliesser, O. Arcizet, and T. J. Kippenberg, “Ultralow
dissipation optomechanical resonators on a chip,” Nature Photonics 2, 627–633 (2008).
[97] Q. Xu, P. Dong, and M. Lipson, “Breaking the delaybandwidth limit in a photonic structure,”
Nature Physics 3, 406–410 (2007).
[98] S. S. Verbridge, H. G. Craighead, and J. M. Parpia, “A megahertz nanomechanical resonator
with room temperature quality factor over a million,” Applied Physics Letters 92, 013112
(2008).
[99] R. G. Beausoleil, P. J. Kuekes, G. S. Snider, S.Y. Wang, and R. S. Williams, “Nanoelectronic
and Nanophotonic Interconnect,” Proceedings of the IEEE 96, 230–247 (2008).
[100] J. Berthold, A. A. M. Saleh, L. Blair, and J. M. Simmons, “Optical networking: past, present,
and future,” Journal of Lightwave Technology 26, 1104–1118 (2008).
107
[101] J. Gripp, M. Duelk, J. E. Simsarian, A. Bhardwaj, P. Bernasconi, O. Laznicka, and M. Zirn
gibl, “Optical switch fabrics for ultrahighcapacity IP routers,” Journal of Lightwave Tech
nology 21, 2839 (2003).
[102] J. Strand and A. Chiu, “Realizing the advantages of optical reconﬁgurability and restoration
with integrated optical crossconnects,” Journal of Lightwave Technology 21, 2871 (2003).
[103] D. T. Neilson, C. R. Doerr, D. M. Marom, R. Ryf, and M. P. Earnshaw, “Wavelength selective
switching for optical bandwidth management,” Bell Labs Technical Journal 11 (2006).
[104] D. Sadot and E. Boimovich, “Tunable optical ﬁlters for dense WDM networks,” IEEE Com
munications Magazine 36, 50–55 (1998).
[105] L. Eldada, “Optical communication components,” Review of Scienti c Instruments 75, 575–
593 (2004).
[106] W. J. Tomlinson, “Evolution of passive optical component technologies for ﬁberoptic com
munication systems,” Journal of Lightwave Technology 26, 1046–1063 (2008).
[107] B. E. Little, J. S. Foresi, G. Steinmeyer, E. R. Thoen, S. T. Chu, H. A. Haus, E. P. Ippen,
L. C. Kimerling, and W. Greene, “Ultracompact SiSiO
2
microring resonator optical channel
dropping ﬁlters,” IEEE Photonics Technology Letters 10, 549–551 (1998).
[108] Y. Yi, P. Bermel, K. Wada, X. Duan, J. D. Joannopoulos, and L. C. Kimerling, “Tunable
multichannel optical ﬁlter based on silicon photonic band gap materials actuation,” Applied
Physics Letters 81, 4112 (2002).
[109] P. Rabiei and W. H. Steier, “Tunable polymer double microring ﬁlters,” IEEE Photonics
Technology Letters 15, 1255–1257 (2003).
[110] T. Asano, W. Kunishi, M. Nakamura, B. S. Song, and S. Noda, “Dynamic wavelength tuning
of channeldrop device in twodimensional photonic crystal slab,” Electronics Letters 41,
37–38 (2005).
[111] J. K. S. Poon, Y. Huang, G. T. Paloczi, A. Yariv, C. Zhang, and L. R. Dalton, “Widerange
tuning of polymer microring resonators by the photobleaching of CLD1 chromophores,”
Optics Letters 29, 2584–2586 (2004).
108
[112] Q. Xu, B. Schmidt, S. Pradhan, and M. Lipson, “Micrometrescale silicon electrooptic mod
ulator,” Nature 435, 325–327 (2005).
[113] V. S. Ilchenko and A. B. Matsko, “Optical resonators with whisperinggallery modesPart II:
applications,” IEEE Journal on Selected Topics in Quantum Electronics 12, 15–32 (2006).
[114] M. C. Wu, O. Solgaard, and J. E. Ford, “Optical MEMS for lightwave communication,”
Journal of Lightwave Technology 24, 4433–4454 (2006).
[115] C. Monat, P. Domachuk, and B. J. Eggleton, “Integrated optoﬂuidics: A new river of light,”
Nature Photonics 1, 106–114 (2007).
[116] F. Xia, M. Rooks, L. Sekaric, and Y. Vlasov, “Ultracompact high order ring resonator ﬁlters
using submicron silicon photonic wires for onchip optical interconnects,” Optics Express 15,
11934–11941 (2007).
[117] N. S. Droz, H. Wang, L. Chen, B. G. Lee, A. Biberman, K. Bergman, and M. Lipson, “Optical
4x4 hitless silicon router for optical networksonchip (NoC),” Optics Express 16, 15915–
15922 (2008).
[118] Y. Vlasov, W. M. J. Green, and F. Xia, “Highthroughput silicon nanophotonic wavelength
insensitive switch for onchip optical networks,” Nature Photonics 2, 242–246 (2008).
[119] I. Fushman, E. Waks, D. Englund, N. Stoltz, P. Petroff, and J. Vu ckovi´ c, “Ultrafast nonlinear
optical tuning of photonic crystal cavities,” Applied Physics Letters 90, 091118 (2007).
[120] M. Yano, F. Yamagishi, and T. Tsuda, “Optical MEMS for photonic switchingcompact and
stable optical crossconnect switches for simple, fast, and ﬂexible wavelength applications
in recent photonic networks,” IEEE Journal on Selected Topics in Quantum Electronics 11,
383–394 (2005).
[121] A. Dorsel, J. D. Mccullen, P. Meystre, E. Vignes, and H. Walther, “Optical Bistability and
Mirror Conﬁnement Induced by Radiation Pressure,” Physical Review Letters 51, 1550–1553
(1983).
[122] M. Eichenﬁeld and O. J. Painter, “Optomechanics of Strongly Coupled Stacked Monolithic
Microdisks,” in Conference on Lasers and ElectroOptics/Quantum Electronics and Laser
109
Science Conference and Photonic Applications Systems Technologies (Optical Society of
America, 2008).
[123] G. S. Wiederhecker, L. Chen, A. Gonarenko, and M. Lipson, “Controlling photonic structures
using optical forces,” Nature 462, 633–636 (2009).
[124] A. J. Seeds and K. J. Williams, “Microwave Photonics,” Journal of Lightwave Technology
24, 4628–4641 (2006).
[125] I. Favero and K. Karrai, “Optomechanics of deformable optical cavities,” Nature Photonics
3, 201–205 (2009).
[126] G. P. Agrawal, Nonlinear Fiber Optics 4th ed (Academic Press, New York, NY, 2007).
[127] Q. Lin, O. J. Painter, and G. P. Agrawal, “Nonlinear optical phenomena in silicon waveguides:
modeling and applications,” Optics Express 15, 16604–16644 (2007).
[128] Q. Lin, T. J. Johnson, R. Perahia, C. P. Michael, and O. J. Painter, “A proposal for highly
tunable optical parametric oscillation in silicon microresonators,” Optics Express 16, 10596–
10610 (2008).
[129] J. J. Blech, “On isothermal squeeze ﬁlms,” Journal of Lubrication Technology 105, 615–620
(1983).
[130] T. Veijola, H. Kuisma, J. Lahdenper¨ a, and T. Ryh¨ anen, “Equivalentcircuit model of the
squeezed gas ﬁlm in a silicon accelerometer,” Sensors and Actuators A: Physical 48, 239–
248 (1995).
[131] R. G. Christian, “The theory of oscillatingvane vacuum gauges,” Vacuum 16, 175–178
(1966).
[132] M. Bao, H. Yang, H. Yin, and Y. Sun, “Energy transfer model for squeezeﬁlm air damping
in low vacuum,” Journal of Micromechanics and Microengineering 12, 341–346 (2002).
[133] R. B. Bhiladvala and Z. J. Wang, “Effect of ﬂuids on the Q factor and resonance frequency
of oscillating micrometer and nanometer scale beams,” Physical Review E: Statistical, Non
linear, and Soft Matter Physics 69, 036307 (2004).
110
[134] S. S. Verbridge, J. M. Parpia, R. B. Reichenbach, L. M. Bellan, and H. G. Craighead, “High
quality factor resonance at room temperature with nanostrings under high tensile stress,”
Journal of Applied Physics 99, 124304 (2006).
[135] P. Del’haye, A. Schliesser, O. Arcizet, T. Wilken, R. Holzwarth, and T. J. Kippenberg, “Opti
cal frequency comb generation from a monolithic microresonator,” Nature 450, 1214–1217.
[136] M. Ferrera, L. Razzari, D. Duchesne, R. Morandotti, Z. Yang, M. Liscidini, J. E. Sipe, S. Chu,
B. E. Little, and D. J. Moss, “Lowpower continuouswave nonlinear optics in doped silica
glass integrated waveguide structures,” Nature Photonics 2, 737–740 (2008).
[137] N. K. Fontaine, J. Yang, Z. Pan, S. Chu, W. Chen, B. E. Little, and S. J. B. Yoo, “Continuously
Tunable Optical Buffering at 40 Gb/s for Optical Packet Switching Networks,” Journal of
Lightwave Technology 26, 3776–3783 (2008).
[138] C. K. Madsen, G. Lenz, A. J. Bruce, M. A. Cappuzzo, L. T. Gomez, and R. E. Scotti, “Inte
grated allpass ﬁlters for tunable dispersion and dispersion slope compensation,” IEEE Pho
tonics Technology Letters 11, 1623–1625 (1999).
[139] M. C. Y. Huang, Y. Zhou, and C. J. C. Hasnain, “A nanoelectromechanical tunable laser,”
Nature Photonics 2, 180–184 (2008).
ii
c 2010 Jessie Rosenberg All Rights Reserved
iii
To my mother, father, and all of the friends, near and far, who have supported me.
for such a valuable collaboration. nearby or far away. Thanks to the senior members of the group. also. Chaitanya Rastogi. and for being so welcoming to a visitor. a . Caltech students or otherwise. Jasper Chan. and for being a great host.iv Acknowledgments First and foremost. also. Thanks to the recently graduated and current members of the group. Jeff Hill. Your scientiﬁc brilliance.and hopefully doing it all even better than we did. Kartik Srinivasan. I want to thank my advisor. and for sharing that knowledge and experience with those of us who followed. for all the conversations. Thanks to Raviv Perahia. and Tom Johnson. Thanks to Darrick Chang. to Qiang Lin. and questions. and answers. for making the group such a dynamic and vibrant scientiﬁc environment. Thiago Alegre. and never balked at answering any question. thanks to Rajeev Shenoi. Paul Barclay. In particular. who always were willing to contribute their time to help out. Daniel Chao. Thanks to all of my friends. for the truly phenomenal amount of work you put in to get the group started. creativity. Ryan Camacho. and Justin Cohen. Amir SafaviNaeini. for your companionship through ﬁve years and three different ofﬁces. Our collaboration over these past months has been an experience of immeasurable value for me. It’s been a delight. for so many fruitful discussions. Matt Borselli. Thanks to the old guard for sharing knowledge and discoveries and stories along the way. Qiang Lin. but know that you are valued. past and present: Orion Crisafulli. Thanks. Raviv Perahia. Chris Michael. You have always been so generous with your knowledge. and thanks to the new students for carrying everything on into the future . Professor Oskar Painter. seemingly trivial or otherwise. Thanks. to my other ofﬁcemates. Matt Eichenﬁeld. Alex Krause. I can’t even begin to name everyone who had such an impact. Thanks to Professor Sanjay Krishna’s group at the University of New Mexico. Thanks to all the excellent staff at Caltech. for the work we did together. In particular. and boundless energy will forever serve as an inspiration. Thiago Alegre.
Jen Soto. Thanks. Megan Nix. everyone. Neil Halelamien.v heartfelt thanks to Eve Stenson. to my mother and father. and Jay Daigle. for being the best and most supportive parents I could have asked for. . Thanks. Thanks to every person on the organizational team of the Caltech Ballroom Dance Club. above all.
while the device geometries become suitable for chipscale and integrated processing. By scaling down the optical resonators to the micro or nanoscale. The use of optical resonators allows the generally weak optical forces to be increased in strength by orders of magnitude due to the many passes light makes within the resonator.vi Abstract Optical resonators present the potential to serve vital purposes in many emergent technologies that require spectral ﬁltering. also increases the absorption of a thin layer of detector material by utilizing the unique optical properties of metal to conﬁne light more tightly within the detector active region. Demonstrated in the valuable midinfrared regime. extremely large optomechanical coupling and very high optical quality factors are shown. coherent mechanical mode mixing. in addition to allowing for a tailorable frequency and polarization response for single detector pixels. or optical delays. and continuing on to investigate the novel phenomena and applications which are made possible when optical and mechanical structures can be synergistically combined. beginning with a more standard alloptical design. Applications to groundstate cooling of mesoscopic devices. missile tracking and guidance. and read out mechanical motion via optical forces. this method of producing pixelintegrated multispectral detectors could ﬁnd application in biological sensing and spectroscopy. and night vision. Using a fully siliconcompatible doublediskgeometry optomechanical resonator. the design and experimental implementation of a plasmonic photonic crystal spectral and polarization ﬁltering element is presented. high compression factor optomechanical cooling. In this thesis. high optical intensities. enabling the demonstration of regenerative mechanical ampliﬁcation. First. progress is presented in the relatively new ﬁeld of cavity optomechanics: utilizing mechanically compliant optical resonators to couple to. the relevant phenomena can increase signiﬁcantly in magnitude. research is presented on several valuable resonator geometries and implementations. This resonator scheme. control. Following this discussion. photonic . while miniaturizing optomechanical devices into a convenient form factor for onchip applications. tunable optical buffering. and widebandwidth alloptical wavelength routing.
. pulse trapping/release. channel routing/switching.vii phononic quantum state transfer. and tunable lasing are discussed.
. . . . . . Optomechanical coupling and dynamic backaction . . . . . . . . . . . . . . . . . . . .6 2. . . .2 2. . . . . 2. Vacuum . . . . . . . . . . . . . . . . Critical Coupling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 2. . . . . . . . . . . . . . . . . . . . . . .7 2 Introduction . . . . . . . SingleMetal Plasmon Resonator Design . . . . . . . . . . . . . . . . . . . . iv vi xiii 1 1 3 8 13 15 18 22 24 24 25 26 32 41 41 44 46 50 52 Double Disk Optomechanical Resonators 2. . . . . . . . . . . . . . . . . . . . 3 Coherent Mechanical Mode Mixing in Optomechanical Nanocavities . . . . . . . .2 1. . . optical. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . .1 2. . . . . . . . . . . Regenerative oscillation . . . . . . . . . . . . . . . . . . . . .4 1. . . . . . . . . . . . . . Optomechanical cooling . . . Doubledisk fabrication. . . . . .7 Ambient pressure . . . . . . . . SingleMetal Experimental Demonstration . . . . .5. . . . . . DoubleMetal Plasmon Resonator Design . . . . . . . . . . .3 1. . . . . . . . .viii Contents Acknowledgments Abstract Preface 1 Plasmonic Resonators for Multispectral MidInfrared Detectors 1. . . and mechanical design . . . . . . . . . . . . . . . . . . . Photonic Crystal Design . . . .3 2. . .1 2. . . . . . . . . .4 2. . . . . . . .5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6 1. . . . . . . . .5 Introduction . . . . . . . . . . . . . . . . . . .5 1. . . . . . . . . . . . . Conclusion . . . .1 1. . . . . . . . . . . . . . . . . . Optical and mechanical characterization . . . . . . . . . . . . . . . . .
. . .6 3. . . . . . . . . . . . . . . . . . . . . . . . .7 3. . . . . . . . . . . The power spectral density of the cavity transmission . . . . . . . . . . . . . . . . . .3. . . . . . Analogy to electromagneticallyinduced transparency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 3. . . . . . . . . . . . . . . . . . .3. . . . . Mechanically Pliant Double Disk Resonators 4. . . . . . . . . . . . . . . . . . . . . . . . .3. . . . .5 3. . . . . . . . . . . . . . . . . . . . .2 3. . . . . . . . . . .8 4 Intracavity ﬁeld in the presence of optomechanical coupling . . . Dynamic ﬁlter response . . . . 3.1 3. . . . . . . .1 3. . Coherent mechanical mode mixing in zipper cavities . . . . . 52 53 55 56 57 58 60 61 65 68 69 73 75 75 76 80 84 95 96 Mechanical mode renormalization in zipper cavities . .3 Introduction . . . . . . . . . . . Spiderweb resonator design and optical characterization . Zipper cavity and doubledisk design. . . . .4 3. Discussion . . . . . . . . . . . . . . Static ﬁlter response . . . . . . . . . Discusssion . . . . . . . . . . .3 4. . . . . . . . . and optical characterization . . . . . . . . .2 3. . . . . . .5 Introduction . . . . Theory of optomechanical effects in the presence of mode mixing . . . . . .4 4. . .3. . . . . . . . . . .ix 3. . . . . . . . fabrication. . . . . . . . . . . . The mechanical response with multiple excitation pathways . . . .2 4. . .1 4. . . . . . Coherent mechanical mode mixing in doubledisks . . . . . . . . . . . . . . . The mechanical response with external optical excitation .3 3. . . . . . . . . . 5 Conclusion . . . . . . . .
. . . . . . .2 2. . . . Detector devices and measurement setup . . . . . . . . . . . . . . . . . . .1 2. .8 1. . . . 1. . . . . . . . . Waveguide thickness dispersion for doublemetal waveguides . . . . . . .1 1. . .6 1. . . . . Plasmonic photonic crystal detector simulations . . . . . . .5 2. . . . . . . . . . Doublemetal focal plane array schematic . . . . . Comparison of FDTD and group theory . . Singlemetal detector enhancement factor and active region absorption . . . . . metal thickness and hole size for the singlemetal structure . . . . . . . . . .12 1. . . . . . . . . . . . Vertical and substrate coupling vs. . Dynamical backaction: damping and ampliﬁcation of mechanical motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DWELL detector measurement results . . . . .5 1. . . . . . . . . . . .10 1. . . . .13 Simulated bandstructure for a doublemetal photonic crystal . . metal thickness and hole size for the doublemetal structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 1. . . . . . . Doubleslab waveguide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 1. . . .3 2. . . . . . . . . . . . . . . . . .8 Vertical coupling vs. . . . . . . . . .14 1. . . . . . .4 2. . . . . . . . . . . . . . . . . . . Field proﬁles for doublemetal waveguides . . . . . . . . . . . . Optical and mechanical mode spectroscopy . . . . . . 21 22 23 28 30 31 33 34 42 44 45 4 5 7 9 11 12 13 15 16 17 18 19 . . . . . . .15 2. . . . . . . . . . . . .x List of Figures 1. . . . Fabrication and characterization of doubledisk NOMS . . Double disk ﬂapping mode displacement . . Regenerative oscillation in a doubledisk microcavity . Cavity Optomechanical Structures . . . .11 1. .6 2. . . . . Farﬁeld proﬁles for stretchedlattice double metal resonators . . .9 1. . . . . . . . . . . .7 2. . . . . . . Simulated bandstructure for a square and rectangular lattice plasmonic photonic crystal Schematic and bandstructure for singlemetal detectors . . . . . . . . . . Transmission scans for a doubledisk microcavity . . . . . . . . . . . . . . . . . . . . . Schematic of doublemetal detector loss mechanisms . . . . . . . . . . . . . . .7 1. . . . . . . . . . . .4 1. . . . .
. . . . . .5 4. . . . . Dynamic response of a spiderweb microresonator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Spiderweb microresonator images and simulations . . . . . . . . . . . . . . . . . . . . . . . . . .1 4. .3 3. . . . . . . . . . Mechanical mode mixing analogues to optical systems and to EIT . . . . . . . 54 62 67 70 71 77 79 81 83 85 . . . . . . . . . . . . . . . . . . . . . . Static tuning of a spiderweb microresonator .4 4. Zipper cavity mechanical mode mixing . .5 Schematics and optical modes of two optomechanical systems . . . . . . . . . Pumpprobe experimental setup . .xi 3. . . . . . Thermomechanical deﬂection of a spiderweb resonator . . .4 3. . . . . . . . . . . . . . .2 4. . . . . . . . . . . . . . . . . . . . . . . . . . .2 3. . . . . . . Mode mixing measurements in double disks . . . . . . . . .1 3. . . . . . Mode mixing in zipper cavities . . .3 4. . . .
5 . . . . .1 Point Group character tables for the square and rectangular lattice.xii List of Tables 1. . . . . . .
we ﬁnally realized that. splitting my time between experiment and theory. but never saw clear resonant spectral enhancement correlating with the photonic crystal patterns. were still around and generously willing to share their knowledge. due to the particulars of the device fabrication. the apertures that we were patterning photonic crystals on were. . In the end. My ﬁrst major task. is a truly overwhelming proposition requiring a daunting amount of effort. but all of the original group members. We made several sets of photonic crystal devices. Sanjay Krishna’s group at the University of New Mexico. With a full plate of classwork interspersed. occupying nearly the entirety of my ﬁrst summer in the lab. which became quite useful later. the path to working devices seemed straightforward. only a small fraction of the open area of the devices: there was a large outer area of exposed detector material that had no photonic crystal patterning or covering of metal. as well as becoming familiar with our group’s homebuilt ﬁnite difference time domain code. I immediately began work on a project started by Kartik Srinivasan and Raviv Perahia. this occupied the ﬁrst few years of my graduate career.xiii Preface The startup of a research lab. that I imagined would occupy my entire graduate career: the patterning of photonic crystals in metal ﬁlms. With some initial dubious successes on the processing front. Though undoubtedly fraught with unforeseeable complexities. and I also got started learning about theory and doing some modeling in Matlab. with their tremendous experience born from setting up everything in the lab. I also spent time working with Orion Crisafulli in doing a thorough investigation of the properties of metalinsulatormetal waveguides. and this background scattering was completely overwhelming our signal. was to become a fully functional user of our group cleanroom. especially one with such an extensive range of facilities. in fact. when the group was mature enough to have essentially all fully functioning labs. Prof. We patterned some initial devices. I was fortunate enough to join the group at a rather fortuitous time. and dovetailed well with the many other photonic crystal efforts then under investigation in the lab. which bafﬂed us for quite some time. to create spectrally sensitive pixels in midinfrared detector material grown by our collaborators.
Developing the “spiderweb” geometry devices – doublering structures with inner spokes. In the end. proving a strain on computational power. we had demonstrated a new type of alloptical tunable resonator. As that project ended. and I ﬂew home to continue doing modeling. as well as nonlinear and quantum optics. and rings to stabilize the mechanics – went surprisingly quickly. we ﬁnally had our ﬁrst clear success: spectrally sensitive detector pixels with a resonance frequency having a direct correspondence to the designed photonic crystal patterns. we decided to focus on ﬁnishing up what we already had accomplished. with few fabrication setbacks. and we ﬁnally had data that we were satisﬁed with. At the time. since the detector testing I had done had all been freespacebased. Qiang Lin was intensely busy with the ﬁrst demonstrations of the doubledisk optomechanical resonators. with applications to a variety of ﬁelds in optical communications. and another round of device processing followed by my ﬁrst visit to the testing facilities at the University of New Mexico. required a very ﬁne simulation mesh. While I attempted to ﬁnd methods to extract the numbers we wanted. containing both metals and thick waveguide layers. With the process optimization I had done working on the photonic crystal detectors. on a new optimized detector material grown by the Krishna group to have more quantum dot layers in a stable conﬁguration.xiv After that discovery. The project started out as a crash course in ﬁberoptic testing for me. and we were able to begin testing those devices with a minimum of delay. we completed the theoretical underpin . One more processing run followed. Unfortunately. I was freed up to begin something else. Emboldened by our successes in both of those endeavors. the simulations proved to be somewhat more difﬁcult. With the help of fruitful discussions with Darrick Chang. The detector devices. We celebrated. so the obvious thing for me to do was to join him to provide some additional manpower. and I started to learn about cavity optomechanics at the same time. we returned to complete the work on coherent mechanical mode mixing Qiang had been working on in the original doubledisk geometry. although further progress in this area seemed achievable with a fair amount of additional effort. In the end. I was wellsituated to help optimize the doubledisk processing. we also encountered processing difﬁculties in attempting to fabricate the doublemetal detector devices that we predicted would give us better results. but in the end we came out with some very nice results and a world of possibilities to investigate. then. The next task we tackled was modifying the doubledisk geometry into a very ﬂexible structure that could achieve large static displacements with only a small applied optical force.
beginning with the theory and an experimental demonstration of singlemetal waveguide devices. work on dynamic backaction in doubledisk resonators is presented. followed by the demonstration of coherent mechanical mode mixing in the same device structure. Looking into the future. and expanded the original idea to encompass the potential for slowlight effects relying on the long phononic timescale rather than the relatively quite short photonic one. Finally. as well as the eventual possibility for quantum state transfer between phonons and photons.xv nings of the work. . and I can only be excited for what will undoubtedly come next. This thesis begins by presenting the work on plasmonic photonic crystal midinfrared photodetectors. and continuing to expand the theory to discuss doublemetal devices. ﬁnishing with several appendices on the experimental and mathematical details of the optomechanical work presented in this thesis. Continuing on. the work on the ﬂexible doublering “spiderweb” optomechanical resonators is discussed. there is still a lot of work to be done: cavity optomechanics remains a wideopen ﬁeld. I see the possibilities presented by this and all the other innovative work being done in the ﬁeld.
as in quantum dot or dotsinawell (DWELL) detectors [4. 3]. the best midinfrared detectors are based on mercurycadmiumtelluride (MCT). but they have various limitations in the required direction of incoming light. other detector materials have become more common. such as in quantum well detectors. or in detector efﬁciency. More recently.1 Chapter 1 Plasmonic Resonators for Multispectral MidInfrared Detectors 1. A hyperpolarization detector could be used in a camera to provide an additional layer of information which can be combined with frequency and intensity data to better distinguish between different objects in an image.1 Introduction Optical sensors in the midinfrared wavelength range are extremely important in a wide variety of areas. in a resonant detector the light can make hundreds of passes. it becomes possible to increase the detector efﬁciency many times over by greatly extending the interaction length between the incoming light and the active material. but largearea focalplane arrays are difﬁcult and expensive to grow due to difﬁculties with the epitaxial growth of mercurybased compounds [2. MCT detectors are very efﬁcient. There are many applications for frequency and polarizationsensitive detectors. With the use of a resonant cavity. ﬁltering incoming signals through the use of hundreds of highly sensitive detector pixels. allowing for a hyperspectral and hyperpolarization sensor without the need for any external grating or prism. The current dominant technologies in the ﬁeld of multispectral imaging rely on the use of either . such as night vision. Currently. and biological spectroscopy [1]. The presence of a resonator can also make each pixel frequency and polarization speciﬁc [6– 8]. A hyperspectral detector array could function as a spectrometer on a chip. 5]. Instead of incoming light making only one pass through the active region. missile guidance.
enhancing the inplane conﬁnement of the resonator mode and enabling strong conﬁnement even with a very shallow photonic crystal etch extending only through the top metal layer [12. strongly enhanced midinfrared detection. Here we detail the design and experimental demonstration of the shallowetch singlemetal resonators. The resonator system we investigate here is composed of a photonic crystal cavity for inplane conﬁnement. or a bank of FPAs with a dispersive element such as a grating or prism to separate light of different frequencies. has a number of advantages. This resonator design. combining the beneﬁts of a plasmonic waveguide and a photonic crystal cavity. In addition. and expand those design principles to propose a highlyefﬁcient shallowetch doublemetal cavity design for hyperspectral and hyperpolarization. 18]. 13]. We have also demonstrated plasmonic photonic crystal designs with the maximum ﬁeld intensity at the top metal interface to allow for thinner devices and increase ﬁeld overlap with the active region versus metallic Fabry Perotbased structures. and it increases the index contrast in the photonic crystal.2 a broadband focal plane array (FPA) with a spinning ﬁlter wheel in front of it [9]. Both singlemetal and doublemetal designs are detector material agnostic. However. These methods are limited by the often high cost and complexity of such systems. multispectral detection becomes much more practical for use in a wide range of applications. In the past. when spectral sensitivity is encoded at the pixel level within a single focal plane array. for the vertical conﬁnement. 17. it provides strong vertical conﬁnement (nearly total conﬁnement. and a plasmonic waveguide [10. 16–18]. and metallic gratings to enable strong conﬁnement without the necessity for deep etching [15. The photonic crystal patterning also serves a dual purpose: it provides inplane conﬁnement to the resonator mode. are easily incorporated into current FPA processing techniques. and provides a mechanism for freely adjusting the polarization response of the detector pixel. composed of either a single or a doublelayer of metal. in both deepetched [19] and shallowetched [13] singlemetallayer implementations. for the doublemetal structure) of the resonator mode within the active region. and do not involve the . many promising schemes have been proposed and/or demonstrated illustrating various aspects of these concepts: optical resonators to provide spectral [14] or spectral and polarization ﬁltering [15–17]. the use of pixelintegrated resonators to provide spectral sensitivity can dramatically increase the efﬁciency of the detector due to the many passes light makes within the resonator. The plasmonic waveguide serves multiple purposes: it serves as a superior top contact (or in the doublemetal case. top and bottom contact) for the detector device providing enhanced extraction efﬁciency. enhanced conﬁnement of light to increase material absorption [14. 11]. serves as a grating coupler to couple normalincidence light into the inplane direction of the detector.
2 Photonic Crystal Design A signiﬁcant obstacle to using resonant cavities to enhance detector absorption and provide spectral and polarization sensitivity is achieving sufﬁcient input coupling from freespace light. The bandstructure of a square lattice photonic crystal is shown in Fig. it becomes possible to achieve signiﬁcant freespace coupling. There are no band gaps for this structure. The ratio of circular hole radius r to lattice spacing a used was r/a = 0. however there are several ﬂatband regions. We used group theory to design a frequency and polarization sensitive photonic crystal structure suitable for coupling efﬁciently to normal incidence light. providing signiﬁcantly increased ﬂexibility and functionality with a minimal increase in complexity. However. 1. The simplest polarizationsensitive resonator design would be a onedimensional grating. and describe how stretching the lattice in one direction can split the degenerate modes of the structure and create a strong polarization sensitivity for use in imaging applications. as well as allowing for continuous variation between polarizationsensitive and polarizationinsensitive devices. it is beneﬁcial to choose a fullyconnected photonic crystal design in order to take full advantage of the increased current extraction efﬁciency from the plasmonic metal layer serving as the top contact of the detector device. with their high conﬁnement. even move towards achieving critical coupling (as will be discussed in Section 1. This work was originally presented in Refs. but we are interested in the modes at the Γpoint. Bandedge modes are ideal for applications such as detectors. with suitable design and optimization of the plasmonic photonic crystal structure. and the index of the material was taken to be the effective index of the double metal plasmon waveguide. Commonly such resonators. therefore the light travels very slowly and is effectively conﬁned within the patterned region. The Γpoint corresponds to normalincidence modulo . Details of the plasmon effective index calculation are discussed in Section 1.3 damage or removal of any detector active region material. calculated using plane wave expansion.1(a). as the mode volume is large.5. The group velocity of these bandedge modes is close to zero. neff = 3. and indeed. 20].6). allowing more of the active region to be contained within the resonator. Therefore.1(a). we analyze a squarelattice structure here. There are a number of ﬂatband regions within the bandstructure in Fig. [13. 1.32. However. 1.24. have only very poor phasematching to a normal incidence freespace beam such as that which we would ideally like to detect for imaging applications.
(0.4 0. 22]. at (f (1. so the Γpoint modes are capable of coupling normalincidence light into the inplane direction of the detector. a reciprocal lattice vector. 0)kΓ g. respectively. The star of k ( k) at the Γpoint is the set of independent Γpoints within the region. EG2 .4 0.7 0. we will consider the nearest Γpoints in the surrounding Brillouin zones.3 0.24) with r/a = 0. circled in Fig. coupling will occur between waveguide modes with similar unperturbed frequencies. Projecting this symmetry .1(b). and the modes of interest are circled. (c) Reciprocal lattice for a stretched photonic crystal. is C4v . with k? and r? representing the inplane wavenumber and spatial position. not uniquely. E G2 ). and propagation constants that differ by a reciprocal lattice vector G. (b) Reciprocal lattice for an unstretched photonic crystal. The inplane ﬁeld of the unperturbed waveguide is given by Ek? (r? ) = ze ˆ (k? ) r? . using group theory [21. the Γpoint modes are above the light line. E G1 . with reciprocal lattice shown in Fig. 1)kΓ g).1 0 Γ X M Γ Γ2 G M (c) Γ2 G2 X2 M X1 Γ1 G1 Γ3 Γ0 X Γ1 Γ3 Γ0 Γ4 Γ4 Figure 1. Since we are interested in modes with nonzero kvectors in the inplane direction. the basis is (EG1 . The character table of C4v is shown in Table 1. is C4v at the Γpoint. In addition. This will be our seed vector. at f(0. 1. k is given. When the structure is patterned. and therefore leak into the air. These points are labeled in Fig. 1.6 0.32. There is one Γpoint within the ﬁrst Brillouin zone (IBZ). 1. The group of the wave vector. with kΓ = 2π/a.5 0. by fkΓ1 g.1(b).2 0.1(a).1: (a) Inplane guided mode TMlike bandstructure plot of a squarelattice photonic crystal (neff = 3.8 Normalized Frequency (a/λ0) (a) (b) 0. We ﬁnd the symmetry basis for the modes at that satellite point by applying the symmetry operations of the group of the wave vector to the seed vector. The point group symmetry of the square photonic crystal lattice. 0)kΓ . In this case. the symmetry group of a plane wave modulo G.1. The light line is shown in yellow. enabling them to couple more easily to an input free space beam. In this case. We investigated the four lowestorder Γpoint modes.
.2 Γ X Γ X Figure 1.5 Table 1.5 (a) (b) 0. C4v E C2 2C4 2σv 2σd C2v E C2 σx σy A1 1 1 1 1 1 A1 1 1 1 1 1 1 1 A2 1 1 1 1 A2 1 1 B1 1 1 1 1 1 B1 1 1 1 1 B2 1 1 1 1 1 B2 1 1 1 1 E 2 1 0 0 0 Normalized Frequency (a/λ0) 0. stretched by 10%.1: Point Group character tables for the square and rectangular lattice.2: 2D bandstructure plots near the gamma point of (a) a square lattice and (b) a rectangular lattice.3 0. The dipolelike modes are shown in bold.4 0.
The symmetry group of this perturbation is C2v . This is in agreement cos(kG2 r)).1. The effect of stretching the lattice on the four lowestorder Γpoint modes is shown in Fig. ˆ EE. we stretch the photonic crystal lattice (not the photonic crystal holes) in one direction.1(a).1 . ˆ EA1 .1(c).1. radiate with a farﬁeld pattern which is uniform: in the case of a ﬁnite structure. To do this. Using the compatibility relations between C4v and C2v . a Gaussianlike farﬁeld without antinodes. (1. we need to split these two degenerate dipolelike modes.2 = z(cos(kG1 r) ˆ EB1 = z(sin(kG2 r)). so generates two degenerate modes. ˆ EB1 = z(cos(kG1 r) ˆ EE.6 basis onto the irreducible representation (IRREP) spaces of C4v . 1.1 = z(cos(kG1 r) + cos(kG2 r)). . 1.2. 1. ˆ These modes are plotted in Fig. 1.1). EE. E is a two dimensional IRREP. 1. the character table for C2v is shown in Table 1. and r has its origin at the center of the air hole. ˆ where A1 .2(b).2) cos(kG2 r)). These degenerate modes.2 = z(sin(kG1 r)).1 = z(sin(kG2 r)).10. The C4v twodimensional representation E decomposes into B1 B2 under C2v . In order to achieve polarization sensitivity. we ﬁnd the new set of modes: EA1 .2 g with the second and third frequency bands. therefore the dipolelike modes are no longer degenerate. 1. (1.1) with what we see in the bandstructure of the stretched lattice. B1 . which is in agreement with the bandstructure in Fig. ˆ EB2 = z(sin(kG1 r)). we can order the modes by frequency. giving the reciprocal lattice shown in Fig. Fig. and E are IRREP spaces of C4v (see Table 1. We associate this pair of modes fEE. with dipolelike symmetry and the spatial pattern given in Eqs. we ﬁnd the modes: EA1 = z(cos(kG1 r) + cos(kG2 r)). Considering that modes with more electric ﬁeld concentrated in areas with high dielectric constant tend to have lower frequency than those with electric ﬁeld concentrated in low dielectric regions [23].
(c) FDTD bandstructure for the unstretched singlemetal photonic crystal structure shown in (a) in the region between the Γ and X points.7 (a) Ez Intensity of Guided Mode Refractive Index/Ez Intensity (a.u. (b) Ez intensity proﬁle of the fundamental plasmon waveguide mode (blue) and the real part of the refractive index of the layers (red). .3: (a) A crossectional image of several lattice constants of the singlemetal DWELL detector design.3 1 0 0 0 2 Depth (µm) 4 6 8 10 0.25 Γ XΓ X Figure 1.) 20 16 (b) z y 0. (d) FDTD bandstructure between the Γ and X1 points for a singlemetal photonic crystal structure stretched and compressed by 10% in the x and y directions. with the detector active region highlighted.4 (c) M Γ X (d) X M Γ X 4 3 2 12 8 4 Active Region Frequency (a/λ) 0.35 0. respectively.
1.2 and Ref. typically in the ultraviolet. Combined together. The vertical conﬁnement factor of this mode within the active region of the detector is η = 91%. 11. even for a . as the two dipolelike modes couple to orthogonal polarizations of incoming light. and allows the mode to extend farther into the active region of the detector. a stretch of the PC lattice breaks the degeneracy of the two modes. The singlemetal resonant cavity consists of a single layer of metal with etched square holes in a square lattice periodic array. allowing a resonator to be fabricated without damaging or removing active material. A representation of several lattice constants of the device structure is shown in Fig.4(bd)) couple most easily from free space. Further improvements in the freespace coupling efﬁciency were performed by optimizing the top metal thickness and hole size. operating in this regime avoids the very high metal losses that occur at frequencies closer to the plasmon frequency.3(a).3(a). 25]. conﬁning the optical mode with a maximum at the surface of the metal (Fig.8 1. and therefore enhancing the probability of detection. Due to the strong index contrast between the surface plasmon [10. 20 shows that the two degenerate dipolelike inplane modes of the structure (Fig. this resonator design provides full 3D conﬁnement.4(b.2. 1. 1. As we operate in the midinfrared frequency range.3 SingleMetal Plasmon Resonator Design In addition to the inplane conﬁnement provided by the photonic crystal pattern discussed in Section 1. this plasmonic PC grating is strong enough to generate an inplane conﬁned resonant mode without etching into the detector active material [12. splitting their resonance frequencies and thus achieving high polarization selectivity [6. The plasmonic layer provides the vertical conﬁnement. and then expand the discussion to consider a doublemetal design that mimics the top and bottom contact layers in detector focal plane arrays for easy integration. 1. 1. 16]. signiﬁcantly increasing the amount of time light spends within the detector active region. The singlemetal resonant DWELL detector structure we study is shown in Fig.c)). in strong contrast to the generally much lower conﬁnement factor of purely dielectric waveguides. In addition. along with the 1D Ez intensity proﬁle of the fundamental plasmon waveguide mode in Fig. This extremely high conﬁnement. We begin with a singlemetal design suitable for straightforward fabrication (as experimentally demonstrated in [13]).3(b) (not including the effects of the photonic crystal holes). it is necessary to conﬁne the light in the vertical direction as well. The numerical and symmetry analysis presented in Sec. 1. we are far from the plasmon resonance frequency of metals. 24] mode beneath the metal regions and the dielectricconﬁned mode beneath the air holes. while the etched airholes create a PC pattern to conﬁne the light inplane.
W = 0. and (d) the xy plane just beneath the metalsemiconductor interface.5 z (µm) 2 active region 4 6 0 0.9 (a) 2 0 (b) x103 1 0.5 1 1 0.5 0.5 0. (c) the yz plane along the hole edge.6. for a structure with lattice ¯ constant a = 2.5 0 (c) (d) z (µm) y (µm) 0.5 0 W ay ax 2 0 2 active region 4 6 8 0 8 x103 1 1 0.5 1 1 0.5 1 x103 1 0.4: (a) A diagram of the unstretched PC structure showing relevant dimensions.5 1 1 1 air semiconductor 0 1 0.5 1 Figure 1.38 µm. and metal thickness t = 150 nm.5 0 y (µm) 0.5 0 x (µm) 0.5 x (µm) 0.50. . The expanded plots show the Ez mode proﬁle for one lattice constant of one of the two dipole modes for an unstretched PC lattice in (b) the xz plane along the hole edge.
.75 at the resonant wavelength. Field proﬁles along different planes for the square lattice dipolelike mode are plotted in Fig. and corresponds to an expected resonant responsivity enhancement of 56 times that of a control sample with no plasmonic layer or PC patterning. versus a sample with no top patterned plasmonic metal layer at the same wavelength (Fig. given an approximate 2% singlepass absorption in the DWELL material. 26. and a metal thickness of tm = 150 nm. This corresponds to an expected responsivity enhancement factor of E At /ADWELL = 5. calculated using ﬁnite difference time domain (FDTD) methods with metal material properties from Ref. 1. The overall quality factor of the two degenerate dipole modes in the unstretchedlattice case (for resonator parameters given in the caption to Fig.2. rotated by 90 degrees in the xy plane.4(a) shows relevant dimensions of the structure. the other dipole mode has the same ﬁeld pattern.axes respectively. with material parameters speciﬁed so as to reproduce the approximate DWELL material singlepass absorption.567.4(bd) over a single unit cell. we deﬁne the normalized hole width ¯ as W = 2W /(ax + ay ). ADWELL = 2%. which can be increased indeﬁnitely by adding more lattice constants to the resonator structure. The simulated active region absorption corresponding to this mode is 10. 1. Figure 1.3(c). 1.and y. and shows Γpoint modes which are in good agreement with the group theory predictions in the previous section. The simulated DWELL active region absorption corresponding to this mode is At = 11. lattice constant ratio of W = 0. we can measure the percentage of the incoming light absorbed within the active region. also as predicted. Figure 1.5). Qxy .939 µm. with the two dipole modes ˆ ˆ showing a signiﬁcant frequency splitting. The simulated structure has a lattice constant ¯ of a = 2. a hole width W vs.9%. The doublemetal waveguide structure can be designed such that these higherorder vertical modes are eliminated.3(d) shows the same bandstructure region for a lattice stretched and compressed by 10% along the x. with the addition of visible higherorder vertical modes from the singlemetal plasmon waveguide. By simulating the structure as excited by an incoming. 1. 1. Note that this enhancement factor is not normalized to the area of the holes in the plasmonic metal. The TM square lattice singlemetal plasmonic photonic crystal bandstructure for the region near the Γpoint is shown in Fig.5% (Fig. 1. immediately showcases the beneﬁts of choosing a plasmonbased photonic crystal design. These two 3D FDTD simulations match up well with the 2D planewave expansion bandstructure predictions shown in Fig.5) is calculated to be Qper = 48 for a perfectly periodic structure.5(a)).10 waveguide utilizing only a single layer of metal. not including the inplane quality factor. normal incidence plane wave.
The mode Pxy corresponds to the xth order inplane and yth order vertical plasmon waveguide mode. will be very small due to their minimal interaction with the photonic crystal grating. but their inplane Qfactor. as is typically done in the case of ’extraordinary’ transmission through thin metal layers [24]. 1.5(a). 1.5: (a) FDTD simulated enhancement factor and active region absorption vs. (b) Ez mode proﬁles in the yz plane at the hole edge for the three longerwavelength peaks in (a). W = 0.005 2 11 0. but are higherorder in the xy plane.005 0 2 0 2 4 6 8 P 0.939 µm. W G1 and W G2 are the ﬁrst and secondorder TM waveguide modes of the structure.02 0. using a structure with lattice constant a = 2. Thus their Q per is higher than the plasmon modes for the perfectly periodic structure simulated. ¯ based on a 2% singlepass absorption. we conclude that the response of this resonator structure will be dominated by the surface plasmonguided modes. and metal thickness tm = 150 nm. . for one lattice constant. and Fig.5(a).01 0 0. and enhancement factor. are plotted versus frequency in Fig.015 0.01 0. as can be seen in Fig. The three shorterwavelength peaks have similar vertical ﬁeld proﬁles.02 0.04 2 0 WG 0.6 µm and a series of higherorder modes at shorter wavelengths.015 1 0 1 y (µm) 1 0 1 y (µm) z (µm) 2 2 1 0 1 y (µm) Figure 1. Thus P00 corresponds to the fundamental plasmon mode as discussed above.01 0. As the overall absorption enhancement factor is lower for these modes even in the inﬁnitestructure limit.02 0 WG P P P 7 8 9 Wavelength (µm) 10 4 6 8 z (µm) 4 6 8 0. Qxy . 1. E.11 Active Region Absorption (%) (a) 6 Enhancement Factor. showing the fundamental plasmon mode at 9.02 0. E 5 4 3 2 1 5 6 P WG 12 10 8 6 4 (b) z (µm) 2 0 P 0. At .04 0. the absorption values are compared over the same physical region of detector material. The simulated active region absorption.01 0. wavelength.567.5(b) shows that they have a lower overlap with the metal surface than the plasmon modes.
(d) Schematic of the setup used for measuring responsivity and detectivity of the fabricated devices. Responsivity measurements were performed by illuminating the sample with a calibrated Mikron M365 blackbody at T = 800 K. bottom and plasmon metallizations. The photocurrent was ampliﬁed using a SRS 570 low noise ampliﬁer and then measured in the spectrum analyzer. (c) SEM image of the fabricated squarelattice PC pattern on the plasmon metallization.12 (a) top metal plasmon metal (b) plasmon metal DWELL active region AlGaAs bottom metal 5 µm GaAs substrate (c) (d) LN2 Dewar Blackbody + Chopper Controller _ Trigger ESA Current Amplifier 5 µm Figure 1. . (b) Crosssectional SEM of the fabricated squarelattice device indicating the plasmon metal and device layers. The blackbody radiation was modulated at a frequency of 400 Hz using a chopper and this signal was used as a trigger for the SRS 760 fast Fourier transform (FFT) spectrum analyzer.6: (a) Optical image of the fabricated device indicating the top.
(f) Measured peak responsivity enhancement E versus resonant device wavelength for two samples at positive and negative bias. 1. to give the ﬁnal spectral response (red) that is plotted in Fig.38 µm device response. ay = 2.b). indicating tuning of peak wavelength with the lattice constant.4 SingleMetal Experimental Demonstration To test these predictions.5 Spectral Response (a. Figure 1. As predicted.) Spectral Response (a.04µm a=2. To separate out the background scattering and the frequency response of the DWELL material from the resonant enhancement.) 0.02 µm) at a bias of 5 V. 1.5 7 7.10µm a=2.7(a.6).u.38µm Spectral Response (a.05 5 x 10 unpatterned response 0 Responsivity Enhancement. we are able to tailor the wavelength and polarization response of each detector pixel.5 6 6.2.7(c). and beneath is the response to light polarized at 0 degrees (red) and 90 degrees (blue) relative to the shorter lattice constant dimension of the lattice.16 µm (red).83µm a=1.5 Wavelength (µm) Figure 1. and therefore different resonant wavelengths determined by the scaling of the pattern.97µm a=2.5 5 5.17µm a=2. but different lattice constant values. we perform data processing as shown for a typical spectral response measurement in Fig.u. Figure 1. and sample B with a rectangular lattice PC having a lattice constant stretching ratio ay /ax = 1.5 0 5 6 7 8 9 10 Wavelength (µm) 5 6 7 8 9 10 (e) 4 2 0 2 4 6 Sample A −5V bias Sample A 5V bias Sample B −5V bias Sample B 5V bias Wavelength (µm) Wavelength (µm) Voltage (V) 5. by varying the lattice constant and symmetry of the patterned grating.) a=1. The background scattering from other regions of the sample (yellow) is then subtracted from the control divided data (blue) and normalized.90µm a=1. The unprocessed data (green) is divided by the unpatterned DWELL response to show the resonances independently of the base detector response.24µm a=2.10 5 4 3 2 1 (f ) unpolarized 90º polarized 0º polarized 0.65 µm.31µm a=2.6(d).7(a) shows the resonant spectral response from a set of representative detector pixels on sample A.u.7: (a) Normalized spectral response from square lattice devices at a bias of 5 V. Responsivity and detectivity measurements were performed at 77 K using the experimental setup shown in Fig. 1.00 (d) 4 0 6 Detectivity (cm Hz1/2/W) control divided 2 1010 109 108 107 6 Voltage (V) 2 0 2 4 6 1 normalized 0 0.5 8 8. (c) Data processing of the a = 2. varying .5 6 6. we fabricated two detector samples: sample A with a square lattice PC.6(ac) shows representative images of the fabricated devices. (b) Normalized spectral response from a rectangular lattice device (ax = 1. The response to unpolarized light is shown in green.13 0. E unpatterned Responsivity (A/W) (a) (b) unpatterned (c) unprocessed 1x 10 0. The spectral response of the surfaceplasmon resonant detectors was measured at 30 K using a Nicolet 870 Fourier transform infrared spectrometer (FTIR) with the patterned detector sample used in place of the standard FTIR detector. 1.78 µm and ay = 2. and a control device (black). All of the PC patterns had the ¯ same normalized hole width (W 0. (d) Measured responsivity and (e) detectivity of a stretched lattice device with ax = 1.
4(b) due to the ﬁnite extent of the PC pattern ( 50 µm in diameter).5 µm to 7. the black body spectral excitance. splitting the resonant detector response into two wellseparated peaks as shown in the green curve of Fig. at a wavelength in good agreement with the theory. and the area of the detector. and in is the noise current.7(b). ∆ f is the noise equivalent bandwidth ofour measurement. 1. providing strong spectral sensitivity within the broad background DWELL response which covers more than 5 µm. measured photocurrent.3) where RN (λ). T )As Ad r2 dλ (1.38 µm. 1.4) where Ad is the detector area. and therefore the limited inplane conﬁnement. the plasmonic PC patterned devices provide a strong enhancement of responsivity and a corresponding increase in detectivity across an applied . respectively. The detectivity D is then p Ad ∆ f Rp D = in (1. 1. The high polarization extinction between the two curves indicates clearly the strong polarization dependence in our device. as represented by the blue and red curves of Fig. T . By varying the polarization of the light incident on the detector. t. 1. The lower and upper wavelength bounds of the detector response are given by λ1 and λ2 .2 µm by choosing PC lattice constants in a range from 1. As . The experimentally measured spectral peaks are broadened relative to the FDTD simulated values in Fig. the transmission of the window and geometrical form factor. FF are the distance between the source and the detector. the black body source temperature. we also observe a higherorder plasmon mode as predicted by the FDTD simulations in Fig.4(b). and r. The linewidth of these resonances is 0.7(b). Le . To characterize the efﬁciency of the detector response and the resonant enhancement. and Ad are the normalized spectral response. we deﬁne and measure the responsivity and detectivity of samples A and B as follows.83 µm to 2. we show that these two peaks correspond to orthogonal linear polarization directions of incoming light. the area of the source. In addition to the fundamental surface plasmon resonant mode.9 µm. we stretch the lattice constant in one direction (sample B). The peak responsivity was computed using the expression Rp = I0 λ2 tFF λ1 RN (λ)Le (λ. I0 . Compared to a control (unpatterned) sample. In order to generate a polarizationsensitive response.14 the peak wavelength response from 5.
but in the midinfrared region. t. and the photonic crystal holes are etched only into the top metal layer. The GaAs core index is indicated by a dotted black line. 1. t (nm) 200 Thickness. 1. there are several considerations. this loss is not prohibitive. as shown in Fig. The enhancement factor E is deﬁned as E = R(λi )/Rc (λi ).15 0.2 0. and 4X for sample B.8: The (a) real (red) and (b) imaginary (blue) parts of the dielectric constant dispersion relation of an Ag/GaAs/Ag waveguide vs. theoretically and experimentally.05 0 0 (b) Effective Index. The doublemetal structure brackets the active region with a thin layer of plasmonic metal on either side. In Fig. As the waveguide thickness decreases and more energy moves into the metal regions.25 0. that a singlemetal plasmonic device can have high activeregion conﬁnement and reasonable Qfactors. neff Thickness. we show enhancement factors across a range of wavelengths reaching as high as 5X for sample A.8(a)) and the imaginary (Fig.7(f).5 DoubleMetal Plasmon Resonator Design Though we have shown. These are achieved at the price of higher plasmonic metal loss. for the doublemetal (or metalinsulatormetal. t (nm) 400 600 800 1000 Figure 1. for a freespace wavelength of λ = 10 µm.8 shows the variation of mode effective index neff with waveguide thickness for a Ag/GaAs/Ag waveguide at a freespace wavelength of λ = 10 µm. All of the advantages of the singlemetal device are preserved. while the substrate loss can be essentially eliminated and the detector active region vertical conﬁnement can approach 100%.8(b)) parts of the effective index . neff 200 400 600 800 1000 0. 1. In choosing the ideal waveguide thickness. as before. 1.1 0. both the real (Fig. we can move to a doublemetal design to increase both of these quantities even further.7(d) and (e). where R(λi ) is the responsivity of the patterned detector at the resonant wavelength and Rc (λi ) is the responsivity of the control sample at the same wavelength. 1. bias range from 5 V to 5 V. MIM) plasmonic waveguide. Figure 1. waveguide thickness t.3 0.35 0.15 8 7 6 5 4 3 0 (a) Effective Index.
even with the photonic crystal holes etched only into the top metal layer. we still achieve signiﬁcant . A high real part of the effective index is beneﬁcial. This is demonstrated via simulations of the full 3D structure which show that. Thus. In fact.16 x 105 5 4 (a) abs(E) (N/C) 5 4 3 2 1 0 x 104 (b) ne = 3.2584i abs(E) (N/C) 0 200 400 600 8 6 4 2 0 0 100 50 z (nm) 0 50 100 150 z (nm) 10 z (µm) 20 30 40 50 60 Figure 1. for a freespace wavelength of λ = 10 µm. it can be shown [12. 25] that the presence of holes in a metal layer has the effect of lowering the effective plasmon frequency of that layer without signiﬁcantly raising losses.9. the imaginary part of the effective index is proportional to the loss in the waveguide. and (c) 50 µm are shown. this condition is even easier to achieve than in the singlemetal case. Though the effective index values shown here would seem to generate only a low index contrast with the core dielectric material (nGaAs = 3. (b) 500 nm.44+0. However.30+0. The effective index neff for each plasmon waveguide is also given. and must be minimized.0112i 12 10 x 103 (c) ne = 3. Therefore a waveguide width must be chosen to balance the competing factors of index contrast and loss.3) for reasonable waveguide thicknesses. because it increases the index contrast of the photonic crystal by increasing the contrast between the photonic crystal holes and the metalcovered regions (the doublemetal waveguide). increase.0005i abs(E) (N/C) 3 2 1 0 ne = 7. improving the inplane conﬁnement Qxy of the resonator. with the calculated effective index neff . In the doublemetal case.64+0. it must be considered that these simulations do not take into account the effect of the photonic crystal holes etched in the metal. for a freespace wavelength of λ = 10 µm. A higher index contrast increases the strength of the photonic crystal perturbation. If the index contrast is high enough. 1. increasing the real but not the imaginary part of the waveguide effective index. The ﬁeld proﬁles of three Ag/GaAs/Ag plasmon waveguides are shown in Fig. the actual combined photonic crystal and plasmon structure will have a considerably higher index contrast than would be expected from these effective index values. the photonic crystal holes can be etched only into the top metal.9: The ﬁeld proﬁle for Ag/GaAs/Ag plasmon waveguides with a thickness t of (a) 10 nm. without removing material from the detector active region.
with a lattice stretched by 20%. due to computational constraints). The modes.2 . (d) A1. (b) B2 . through examining the spatial fourier transform of Ez . inplane conﬁnement and vertical coupling. The group theory predictions are shown on the left. inside the active region. are (a) A1. labeled according to C2v designations. in order of increasing frequency. are shown on the right. The FDTD ﬁelds shown are 2D slices of the full simulations taken just below the top metal layer. (c) B1 .1 . Full structure simulations (photonic crystal plus plasmon waveguide) were performed using the FDTD method on a perfectly periodic lattice (as before. at the outside of the simulation region it can be seen that the simple group theory calculations have accurately predicted the mode shapes given by the more complex FDTD simulations. We have also investigated the farﬁeld proﬁles of the four Γpoint modes.10 shows the FDTD ﬁeld plots (left) in comparison with the group theory mode plots (right). Due to timereversal symmetry. Figure 1.17 (a) (b) (c) (d) Figure 1. These FDTD simulations conﬁrm the results of our separate photonic crystal and plasmon waveguide simulations. including the effects of the photonic crystal air hole (overlaid white square). while the FDTD results.10: A comparison of the FDTD simulated and group theory predicted Ez ﬁeld proﬁles for the four lowest gammapoint modes of a 20% stretched lattice. Though the presence of the airhole distorts the shape of the modes in the center of the FDTD images. the ﬁeld proﬁle of a mode that can be coupled into the resonator is equivalent to the ﬁeld proﬁle of the resonator mode propagated .
18
150 100 50
(a)
Py/Px > 109
150 5 4 3 2 1 100 50
(b)
Px/Py > 109
5 4 3 2 1
y (µm)
y (µm)
0
0 50
50 100 150 100 50 0 50 100
100 150 100 50 0 50 100
x (µm)
x (µm)
Figure 1.11: Farﬁeld plots at z = 90 µm of the two Γpoint dipole modes in a stretchedlattice struc¯ ture with W = 0.5309 and ay /ax = 1.2. (a) B2 mode power density, with dominant ypolarization. ˆ (b) B1 mode power density, with dominant xpolarization. For both B1 and B2 modes, the polarˆ ization selectivity is calculated to be greater than 109 , limited entirely by error in the numerical simulation.
out into the farﬁeld, therefore these farﬁeld plots indicate the modeshapes and polarizations that couple most strongly from free space to the resonator mode. Farﬁeld plots of the two fundamental stretchedlattice dipole modes, B1 and B2 , are shown in Fig. 1.11, generated from an 10X10 tiled array of the FDTD simulated ﬁeld proﬁle (itself one lattice constant in size) and apodized using a Gaussian function with a standard deviation of two lattice constants (ax and ay , respectively) in the x and y directions. We can see from Fig. 1.11(a,b) that the B1 and B2 modes are wellsuited for coupling to incident freespace light, since the farﬁeld proﬁle has a single lobe at normal incidence and does not contain any antinodes, in agreement with the group theory predictions from Section 1.2.
1.6
Critical Coupling
After optimizing the largescale resonator design and choosing photonic crystal modes which have the largest coupling to normalincident light, it still remains to ﬁnd the best values for the design parameters to increase detector absorption, and to determine the fundamental limits on absorption enhancement for these two (singlemetal and doublemetal) resonator designs. We ﬁnd that, for the doublemetal resonator, the absorption is greatest at the point of critical input coupling, whereas we ﬁnd a different optimal point for the singlemetal resonator, as the parasitic substrate loss increases along with the detector absorption as the input coupling is increased.
19
Qe Qmetal
V
+V
QD
Qxy
Figure 1.12: The dominant loss mechanisms within a doublemetal plasmonic photonic crystal resonant detector.
Critical coupling occurs when the external coupling to the resonator (the resonator “coupling loss”) is equal to the total internal cavity loss from all other loss mechanisms. When that condition occurs, the reﬂection coefﬁcient goes to zero, and all of the incident light at the resonance frequency is absorbed in the resonator [27]. Figure 1.12 shows the various scattering and absorption processes that are involved in near normal incidence resonant detection. The reﬂection from the cavity is given by R= where ∆ = ω ∆2 +
γ0 γe 2 2 , 2 ∆2 + γt 2
(1.5)
ω0 is the frequency detuning from the resonance frequency ω0 , γ0 is the intrinsic
cavity loss rate, γe is the vertical coupling rate to free space, and γt = γ0 + γe . The loss rates γ are related to the Qfactors given previously by γ = nλQ/2πc. It is clear from Eq. 1.5 that, when the cavity is excited onresonance (∆ = 0), the reﬂection goes to zero when γ0 = γe , when the rate at which the cavity can be fed from free space is equal to the sum of all the internal cavity loss rates. This is the critical coupling condition. From the expression for the reﬂection in Eq. 1.5, we can write the power dropped into the cavity
20 (not only the absorbed power, but all of the power not reﬂected): Pd = Pin (1 R) = Pin γ0 γe ∆2 +
γt 2 2
,
(1.6)
where Pin represents the power incident on the cavity. Therefore the fractional absorption efﬁciency into the ith loss channel is pi = γi Pd γi γe = 2 + γt γ0 Pin ∆ 2
2
.
(1.7)
We can enumerate the loss mechanisms in Fig. 1.12, such that γ0 = γD + γmetal + γxy + γsub , corresponding to the (beneﬁcial) detector absorption, the metal absorption, the inplane loss, and the substrate loss, respectively. The inplane loss can always be made negligible, by adding more lattice constants to the photonic crystal patterning region to increase the inplane conﬁnement strength relative to the other loss mechanisms. In the singlemetal case, we can consider γsub = mγe , representing a mode coupling into the substrate that is m times larger than that into the air due to the higher substrate refractive index; in the doublemetal case, m plasmon metal. The fractional absorption into the DWELL detector material is then, at resonance, given by pD = 4γD γe . [(m + 1)γe + γmetal + γD ]2 (1.8) 0 due to the thick bottom layer of
From this expression, we see that the maximum fractional absorption occurs at γe = γD + γmetal . 1+m (1.9)
As the input coupling γe can be adjusted by varying resonator parameters, this maximal condition should be readily achievable, corresponding to a fractional absorption into the detector material of pD,max = For the doublemetal case in which m γD . (1 + m)(γD + γmetal ) (1.10)
0, the maximal detector absorption occurs at the pure critical
coupling condition, γe = γD + γmetal . In this case, the fractional power absorbed is primarily limited by the relatively small metal losses in the midinfrared region. For the silver plasmon waveguides simulated in this work, we ﬁnd a metal loss quality factor of Qmetal = 149, in comparison to the
With variation of W .7 0. . but the larger air hole also distorts the shape of the mode. Qe and Qsub . tm (nm) 0.13(b) for singlemetal and 1. and (b) with W for the fundamental (blue) and higherorder (red) modes of the unstretched singlemetal photonic crystal lattice. we investigate their effect on Qe for both the singlemetal unstretched lattice (Fig. tm . without changing the mode frequency. the variations in quality factor with changes in the top metal thickness. and the top metal thickness. illustrate vertical quality factors Qe of both dipolelike modes increasing monotonically as the top metal becomes thicker. Open circles represent Qe and ﬁlled triangles represent Qsub .8% of the incoming light will be absorbed in the active material for the optimal external coupling quality factor of Qe = 83. 1.8 Figure 1.4 0.2) structures. the normalized hole ¯ width. showing a curve most likely due to a combination of factors: the increased hole size provides a larger aperture through which light can escape. shown in Figs.13(a) for singlemetal and 1. Varying the top metal thickness tm is an effective way to change Qe to better match the internal loss. and thus more closely approach critical coupling.13) and doublemetal stretchedlattice (Fig. shown in Figs. This indicates that 55. with ¯ metal thickness tm . W . In contrast. with a lattice stretching ¯ ratio of 1. the overall trend for both structures is the same. 1.5 W 0. A dotted line marks the value of the parameter held constant in the opposing plot.14(b) for doublemetal. 1. causing it to generate a less pure farﬁeld proﬁle which does not match as well with a freespace beam. decreasing the Qe . estimated DWELL detector absorption quality factor of QD = 188. There are many free parameters in this resonator structure which can be optimized in order to achieve the optimal input coupling value.21 10 (a) Quality Factor 100 200 300 400 500 10 10 10 10 10 (b) Quality Factor 10 10 10 Metal Thickness.6 0.13: (a) Variation of external coupling and substrate loss quality factors.14(a) for doublemetal. Choosing two of the most signiﬁcant. 1.14.
the stretching of the photonic crystal lattice does not signiﬁcantly decrease Qe .62 W 0.22 External Quality Factor. . moving towards achieving critical coupling. Qe 10000 (a) 600 500 400 300 200 (b) 100 0. We theoretically analyzed the conditions for optimal detector absorption enhancement. Qe 8000 6000 4000 2000 00 Metal Thickness.58 0. This resonator can be optimized for use at any wavelength from the terahertz to the visible with suitable scaling of the photonic crystal holes and waveguide width. since no photonic crystal holes are etched into the active region itself. and experimentally demonstrated singlemetal devices with responsivity enhancement of up to 5X.66 0. we adjusted the vertical coupling efﬁciency to more closely match the resonator loss. 1. and can easily be modiﬁed to suit any detector material.7 Conclusion We have designed a plasmonic photonic crystal resonator utilizing either a singlemetal or doublemetal plasmon waveguide for use in midinfrared photodetectors. the doublemetal structure has a lower achievable Qe . indicating more favorable external coupling conditions.7 Figure 1. Additional increases in coupling efﬁciency or reductions in loss could bring the system to near 100% absorption in the detector.54 0. Though the behavior as hole size and metal thickness are varied is similar for both singlemetal and doublemetal structures.14: Variation of vertical coupling quality factor Qe of the B1 (blue square) and B2 (green ¯ circle) modes of the doublemetal photonic crystal lattice with changing (a) tm and (b) W . tm (nm) 200 400 600 800 1000 1200 External Quality Factor. This resonator design shows good frequency and polarization selectivity for use in hyperspectral and hyperpolarization detectors. and by varying the photonic crystal hole size and top metal thickness.
15. 1.15: A design schematic for a resonant doublemetal plasmonic photonic crystal FPA. a proposed FPA schematic is shown. The ﬂipchip bonding method of focal plane array (FPA) fabrication naturally lends itself to use with a doublemetal resonant cavity. In Fig. 28]. . illustrating the ease with which doublemetal plasmonic photonic crystal resonators can be incorporated into current FPA designs and presenting the possibility to achieve highly sensitive midinfrared spectral and polarization imaging at low cost.23 Infrared Light In Photonic Crystal Resonator Highly Doped Layer Metal Silicon Nitride Metal Active Region Highly Doped Layer Indium Bump Epoxy Underfill Figure 1. DWELL FPAs have already been demonstrated with hybridization to a readout integrated circuit [4. with only the top metal photonic crystal lithography step differing from standard process techniques.
from multikilometer long gravitationalwave detectors [31] to coupled nanomechanicalmicrowave circuits [32]. each of these systems depend upon “dynamical backaction” [33. These properties of the doubledisk resonator make it interesting for a broad range of applications from sensitive force . consisting of a pair of silica disks separated by a nanoscale gap. cavitymechanical systems span a wide range of geometries and scales. and efﬁcient cooling of the mechanical motion is obtained with a temperature compression factor of 13 dB for 4 microWatt of dropped optical input power.24 Chapter 2 Double Disk Optomechanical Resonators 2. As has been recently proposed [42. Today. Recent work in the optical domain has used the scattering radiation pressure force to both excite and dampen oscillations of a micromechanical resonator [35–41]. shows extremely strong dynamical backaction. the threshold for regenerative mechanical oscillation is lowered to an optical input power of only 270 nanoWatts. 45]. In vacuum. the optical gradient force within guidedwave nanostructures can be ordersofmagnitude larger than the scattering force. 34] resulting from the positiondependent feedback of electromagnetic wave momentum.1 Introduction Many precision position measurement devices involve the coupling of mechanical degress of freedom to an electromagnetic interferometer or cavity [29. 43] and demonstrated [44. with the intriguing possibility of selfcooling the mechanical system down to its quantum groundstate. For the sensitive detection and actuation of mechanical motion. powerful enough to excite giant coherent oscillations even under heavily damped conditions (mechanical Q 4). In this work we combine the large perphoton optical gradient force with the sensitive feedback of a high quality factor whisperinggallery microcavity. 30]. or roughly 1000 stored cavity photons. The cavity geometry.
47]. the energy coupling rate between external laser and internal cavity ﬁelds).2b). 50]. However.1) where ωc is the optical cavity resonance frequency. B = g2 / κ3 ωc mx . . x. without the feedback provided by an optical cavity or interferometer. As the momentum change of a photon per round trip is ﬁxed inside such cavities. the optical force only provides a static mechanical displacement. The effectiveness of the coupling between the optical and mechanical degrees of freedom can thus be described by a backaction parameter. In a FabryPerot (Fig. 43]. the maximum rate Pd . 2. In contrast. This method was recently employed in a silicon photonic circuit to manipulate a suspended waveguide [44].2 Optomechanical coupling and dynamic backaction The per photon force exerted on a mechanical object coupled to the optical ﬁeld within a resonant cavity is given by gOM . the intrinsic energy loss rate of the optical cavity. the optical force manifests itself as a socalled scattering radiation pressure due to direct momentum transfer from the reﬂection of photons at the cavity boundary [49. 48. κe . and the optical i OM cavity Qfactor.2a) or microtoroid resonator (Fig. 2. the perphoton force. while the roundtrip time increases linearly with the cavity length. allowing for photon momentum to be transfered over a length scale approaching the wavelength of light [42.25 and mass detection in viscous environments such as those found in biology [46. which depends upon the motional mass. to quantum cavityoptomechanics in which a versatile.opt p 3 3g2 OM κ3 ωc mx i 1 (1 + K)2 Ωm ). the radiation pressure per photon scales inversely with the cavity size. This work was initially presented in Ref. dynamical backaction can be quantiﬁed by considering the magnitude of the damping/ampliﬁcation that an input laser has on the mechanical motion. (2. For a ﬁxed absorbed optical input power in the badcavity limit (κ is given by Γm. 2. chipscale platform for studying the quantum properties of the system may be envisioned. where gOM dωc /dx is a coefﬁcient characterizing the dispersive nature of the cavity with respect to mechanical displacement. for the gradient optical force the cavity length and the optomechanical coupling can be decoupled. mx is the motional mass of the optomechanical system. and K κe /κi is a cavity loading parameter (κi . Pd is the optical power dropped (absorbed) within the cavity. In the case of a cavity optomechanical system.
The doubledisk structure (Fig. 45] (very recent work [51] involving the versatile coupling of external nanomechanical elements to the nearﬁeld of a highFinesse microtoroid has realized very strong dynamical backaction. 2. fabrication. and characterization of a nanooptomechanical system (NOMS) consisting of a pair of optically thin disks separated by a nanoscale gap.4. respectively. Sample II) in diameter.2(ef)). Sample I) and one small (D = 54 µm. The two silica disk layers and the sandwiched amorphous silicon (αSi) layer were deposited on a (100) silicon substrate by plasmaenhanced chemical vapor deposition. 2. Release of the doubledisk structure was accomplished using a SF6 chemical plasma etch which selectively (30. beyond which radiation loss becomes appreciable (Qr 108 ). Simultaneously. resulting in a uniform undercut region between the disks extending radially inwards 6 µm from the disk perimeter. the bonded . The disk pattern was created using electron beam lithography followed by an optimized C4 F8 SF6 gas chemistry reactive ion etch. and provides backaction several orders of magnitude larger than in previously demonstrated gradient force optomechanical systems [44. The ﬁnal doubledisk structure.2c) supports highQ whisperinggallery resonances. one large (D = 90 µm. 2. Two nanoforks were also fabricated near the doubledisk resonator to mechanically stabilize and support the ﬁber taper during optical coupling. consists of 340nmthick silica disks separated by a 140 nm air gap extending approximately 6 µm in from the disk perimeter (the undercut region). improving the optical quality of the material.3 Doubledisk fabrication. Fabrication of the doubledisk whisperinggallery resonator began with initial deposition of the cavity layers. optical. The wafer was then thermally annealed in a nitrogen environment at a temperature of T = 1050 K for 6 hours to drive out water and hydrogen in the ﬁlm. and mechanical design Here we describe the design. The ﬁnal gap size between the disks was measured to be 138 8 nm (shrinkage having occurred during the anneal step). shown in Fig. Finite element method (FEM) simulations of the whisperinggallery optical modes of the doubledisk structure shows substantial splitting of the cavity modes into even and odd parity bonded and antibonded modes (Fig. 000 : 1) attacks the intermediate αSi layer and the underlying Si substrate. with a thickness of 340 4 nm and 158 3 nm for the silica and αSi layers. the underlying silicon support pedestal is formed. Two different sized cavities are studied here. although still roughly twoorders of magnitude smaller than in our integrated device). the geometry was optimized such that the forks introduce a total insertion loss of only 8%.26 2. Due to its substantial ﬁeld intensity within the air gap. The small diameter cavity structure represents a minimal cavity size.
Ey = D cosh γx. Aeγx .4).3) where k0 = ω0 /c is the propagation constant in vacuum and β = k0 neff is the longitudinal component of the propagation constant of the bonding mode. x > h + x0 /2 x0 /2 < x < h + x0 /2 x0 /2 < x < x0 /2 x0 /2 > x > h x< h x0 /2 x0 /2 (2. (2. The circular geometry of the double disk forms the whisperinggallery mode. The continuity of Ey and Hz across the boundaries requires κ and γ to satisfy the following equation: κγ [1 + tanh(γx0 /2)] = κ2 γ2 tanh(γx0 /2) tan κh. indicating that ω0 becomes a function of x0 . For the bonding mode polarized along the ey direction.2(g). we ﬁnd that the optomechanical . as expected. They are given by the following expressions: 2 κ2 = k0 n2 c β2 . β. Accordingly. any variation on the disk spacing x0 transfers to a variation on the resonance frequency ω0 through Eqs. B cos κx +C sin κx. to be ﬁxed as 2πRβ = 2mπ. Thus.1. 2. (2.2) where κ is the transverse component of the propagation constant inside the slabs and γ is the ﬁeld decay constant in the surrounding area.3) and (2. As the mode conﬁnement in a doubledisk NOMS is primarily provided by the transverse boundaries formed by the two disks.4) which reduces to tan κh = γ/κ when x0 ! 0. By using these two equations. γ2 = β2 2 k0 n2 . the doubledisk structure can be well approximated by a symmetric doubleslab waveguide shown in Fig. neff is the effective refractive index for the guided mode. 2. the tangential component of the electric ˆ ﬁeld is given by: Ae γx . s (2.27 mode tunes rapidly with changing gap size as shown in the inset to Fig. B cos κx C sin κx. where R is the mode radius and m is an integer. the tangential component of the magnetic ﬁeld can be obtained through Hz = i ∂Ey µω0 ∂x . in which the resonance condition requires the longitudinal component of the propagation constant.
Equation (2. (2.7) which is approximately on the order of the optical wavelength λ0 . as the two slabs are coupled through the evanescent ﬁeld between them with amplitude decaying exponentially with slab spacing at a rate γ [see Eq. Eq. c 3/2 γ γ s π n2 n2 eff s ω0 LOM . deﬁned such that gOM a minimum effective length 2 2 2 2 2 k0 2 λ0 neff + k0 hnc neff ns 1 + 2 (n2 + n2 γh) = . the magnitude of the optomechanical coupling can be characterized by an effective length. 2 2β2 + 2k0 n2 γh c (2. respectively. Physically.6) infers L0 = (2.2)]. respectively. is given by the general form gOM (x0 ) = 4(n2 c cχγ2 2 γx0 k0 sech 2 2 2 γx0 n2 ) tan κh + n2 x0 χsech 2 + 2ξ (n2 γh csc2 κh + 2n2 ) tan κh + nsγκ s s s c n2 γ c κ (2.5) . coupling coefﬁcient. 2 When x0 ! 0. nc and ns are the refractive indices for the slab and surrounding area. gOM = dω0 dx0 . LOM . where χ κ + γ tan κh and ξ 1 + tanh( γx0 ).28 x ns h x0 h nc ns nc ns z Figure 2. (2.5) leads to the maximum optomechanical coupling of gOM (0) = ω0 γ3 . h and x0 are the slab thickness and the slab spacing. the resulting optomechani .6) In analogy to FabryPerot cavities and microtoroids.1: Schematic of a symmetric doubleslab waveguide.
For the air gap of 138 nm used in this work. reaching a minimum value of 3. What matters for the optomechanical effect.8) where gOM (0) is given by Eq.9) x0 ]2 . 2 corresponding to an effective mechanical potential energy of E p = mx Ω2 xeff /2. The doubledisk structure also supports a number of different micromechanical resonances. E p = Us and the effective motional mass is given by mx = 2Us 2 [x (r ) Ωm m 0 = Us . LOM ω0 γx0 gOM (0) e . Therefore.2(g). as that determines the magnitude of the splitting between the bonding and antibonding cavity modes. the mechanical displacement of a double disk exhibiting a ﬂapping mode is generally a function of radius (Fig. the optomechanical coupling is estimated to be gOM /2π = 33 GHz/nm (LOM = 5.8 µm).6).8 µm at a resonance optical wavelength of λc 1. at which point all of the mechanical energy is stored in the strain energy Us . The most strongly coupled mechanical resonance is that of the symmetric (i. As indicated by the red curve in Fig.. equivalent to 22 fN/photon. however. azimuthal mode number. Note that xeff is twice the real displacement at the mode center for a single disk. 2. the approximate effective length.5 µm.3). agrees well with the results simu lated by the ﬁnite element method. (2. The effective mechanical displacement is then given by xeff = xm (r0 ) x0 . As the mechanical displacement actuated by the gradient force is generally small compared with the original disk spacing x0 .8) provides an excellent approximation for the optomechanical coupling coefﬁcient in a doubledisk NOMS. (2. m = 0) ﬂapping motion of the disks. is the disk spacing at the place where the whisperinggallery mode is located. E p reaches its maximum value when the double disk is at rest at its maximum displacement.29 cal coupling can be well approximated by an exponential function gOM (x0 ) gOM (0)e γx0 .2(g). Eq. where r0 is the radius of the whisperinggallery mode. shown in Fig. With a clamped inner edge and a free outer edge. where mx is the m corresponding effective motional mass and Ωm is the resonance frequency of the ﬂapping mode. Therefore. 2 d 2 (r ) 2Ωm 0 (2. xeff = 2d(r0 ). ranging from radial breathing modes to whisperinggallerylike vibrations of the disk perimeter. we can assume it is uniform in the region of the whisperinggallery mode and deﬁne the effective disk spacing xm (r0 ) at the mode center. (2. 2. and the effective length decreases roughly exponentially with decreasing disk spacing. 2.e.
(f) FEMsimulated tuning curve of the bonded mode.3 nm.2: Schematic of the corresponding (a) FabryParot and (b) microtoroid optomechanical cavities. showing the mechanical ﬂapping motion of the disks. FEMsimulated optical mode proﬁles of the radial component of the electric ﬁeld for the (d) bonded mode at λ = 1520 nm and (e) antibonded mode at λ = 1297. gOM and LOM are both wellapproximated by exponential functions (red curves).30 a gOM= ω0/LC c gOM= ω0/LOM SiO2 αSi SiO2 x LC x h b R gOM= ω0/R 2 Height (µm) Resonance frequency (THz) 1 0 d 208 204 200 196 f 50 40 gOM 2π (GHz/nm) 30 20 10 0 0 g 60 50 40 30 20 10 200 400 600 Disk spacing x (nm) 800 0 E ective length LOM (µm) 1 2 2 1 0 Height (µm) e 1 2 40 42 44 46 Radius (µm) 48 194 0 400 800 Disk spacing x (nm) Figure 2. (g) Optomechanical coupling coefﬁcient and effective length (blue curves) for the bonded mode. (c) Schematic of the doubledisk NOMS structure. .
Note that E p is the total potential energy for the two disks. 2. which is simply two times that of a single one because of the symmetry between the two disks. (2. xm (r0 ) corresponds to the effective disk spacing at the mode center. 2.3].e. where both Us and d(r0 ) can be obtained from the mechanical simulations by the ﬁnite element method. As the physical mass of a single disk region involved in the 2 ﬂapping motion is given by m p = πζh(rb 2 ra ). h is the thickness for a single disk.10) where ζ is the material density. m (2. we ﬁnd that the effective mass is related to the physical mass through the following expression: mx = 4m p 2 rb 2 ra [xm (r0 ) rb x0 ]2 rd 2 (r)dr = ra 2 rb mp 2 ra d 2 (r0 ) rb ra rd 2 (r)dr.31 ra d(r) xm(r0) x0 r0 rb whisperinggallery mode Figure 2. d 2 (r)/d 2 (r0 ) 1 for most of the region between .11) As the whisperinggallery mode is generally located close to the disk edge (i.3: Illustration of the disk displacement. using Eq.3).10). the mode radius r0 = 44 µm in a double disk with rb = 45 µm). r0 is the radius of the whisperinggallery mode. The relationship between the effective mass and the physical mass of the doubledisk NOMS can be found by examining the mechanical potential energy. we can ﬁnd the total mechanical potential energy by integrating over the disk regions involved in the ﬂapping motion: rb Ep = ra Ω2 d 2 (r)ζ2πrhdr. (2.. ra and rb are the inner and outer radii of the disk region involved in the ﬂapping motion. and ra and rb are the inner and outer radii of the disk region involved in the ﬂapping motion (see Fig. With a mechanical displacement d(r) for each single disk [Fig. x0 is the disk spacing in the absence of the optical ﬁeld. d(r) is the mechanical displacement at radius r.
all of them of TElike polarization and bonded mode character. the mechanical motion can be described .4). with an 6 µm undercut air gap region involved in the ﬂapping motion (Fig. provide a signiﬁcant enhancement to the dynamic backaction parameter which scales as g2 /mx . Several radialorder whisperinggallery modes are evident in the spectrum. leading to considerably broadened cavity transmission. 49]. the cavity transmission exhibits intense ﬂuctuations when the laser frequency is scanned across the cavity resonance. Fig.8 µW well below the oscillation threshold.4 Optical and mechanical characterization Optical and mechanical measurements were initially performed at room temperature in a one atmosphere nitrogen environment.6(a) shows the wavelength scan of a large diameter doubledisk cavity (Sample I). the averaged spectrum of the cavity transmission (red curve) is signiﬁcantly broader than the real cavity resonance. A correct description of the cavity transmission requires an appropriate inclusion of the optomechanical effect. the effective mass is 0. Therefore. 2. The effective mass decreases to 0. taking into account the mechanical perturbations. even the small thermal Brownian motions of the ﬂapping mode introduce signiﬁcant perturbations to the cavity resonance due to the large optomechanical coupling. With a small input power of 5. Unlike other microcavities in which the linear transmission is determined only by the cavity loss and dispersion.32 ra and rb .264 nanogram.57 nm is shown in the Fig. OM 2. For the 90µm device used in our experiment. the effective mass is signiﬁcantly less than half the physical mass of a single disk region. 2. only about one ﬁfth of the physical mass of a single disk region m p = 1. and in combination with the large perphoton force.11) shows that mx m p /2.2c and Fig. (2. for the doubledisk NOMS. from which an intrinsic optical Qfactor of 1. due to the decrease in the disk radius. which is developed in the following.75 106 is inferred. When the optical power is well below the oscillation threshold and the ﬂapping mode of the double disk is dominantly driven by thermal ﬂuctuations. and Eq.145 nanogram for the 54µm device. the effective mass is much smaller than this value because of the real displacement function d(r).18 nanogram.5(a) shows an example of the cavity transmission of Sample I. As a result. 2. Figure 2. In practice. Note that both these values are more than two orders of magnitude smaller than commonly used micromirrors and microtoroids [36–41.6(a) inset. 2. The fundamental TElike bonded optical mode at λ = 1518.
False color is used to indicate different relevant regions of the device. . while it is ﬁrst ampliﬁed by an erbiumdoped ﬁber ampliﬁer (EDFA) for the experiments performed in vacuum. the cavity transmission is sent directly to the photodetectors. with input power controlled by a variable optical attenuator (VOA) and wavelength calibrated by a MachZehnder interferometer (MZI).c) Scanning electron microscope images of the 54µm doubledisk NOMS. A tunable laser source is used to optically probe and actuate the doubledisk structure. The cavity input and transmission are both transported through a singlemode silica ﬁber taper. (b.4: (a) Schematic of the experimental setup for optical testing of the doubledisk cavity.33 VOA 14951565 nm Tunable Laser vacuum chamber a 10 : 90 splitter Polarization Controller Reference Detector 1 MZ I Highspeed Detector Oscilloscope double disk Reference Detector 2 10 : 90 splitter TBPF EDFA VOA fiber taper b 1 µm c nanoforks pedestal undercut 20 µm Figure 2. For experiments performed in a nitrogen environment. which is supported by two nanoforks for stable operation.
57 nm. a Markovin process with the following correlation function: hFT (t)FT (t + τ)i = 2mx Γm kB T δ(τ).96 0. The blue curve is the instantaneous signal collected by the highspeed detector and the red curve is the average signal collected by the slow reference detector 2. where x(Ω) is the Fourier transform +∞ iΩt ∞ x(t)e dt.75 106 −30 −20 −10 0 10 20 30 [λ−1518.13) that the Brownian motion of the ﬂapping mode is also a Markovin process with a spectral correlation given by hx(Ω1 )x (Ω2 )i = 2πSx (Ω1 )δ(Ω1 of the mechanical displacement x(t) deﬁned as x(Ω) = Ω2 ). 2. m 2 dt dt mx (2.5: (a) The cavity transmission of Sample I in a nitrogen environment.6 0. It can be shown easily from Eqs. (2.6(b).34 Normalized Transmission Normalized Transmission 1 1 0. damping constant.4 0. FT is the Langevin force driving the mechanical Brownian motion.94 0. Γm .4 (a) −10 −5 0 5 10 Qi = 1.8 µW. The inset shows a detailed scan for the bonding mode at 1518. and effective mass of the ﬂapping mode.57 nm with an input power of 5.14) (Ω2 m .13) where T is the temperature and kB is Boltzmann’s constant.12) and (2. and Sx (Ω) is the spectral intensity for the thermal mechanical displacement with the following form: Sx (Ω) = 2Γm kB T /mx . The slight asymmetry in the transmission spectrum is due to the static component of mechanical actuation when the laser is scanned from blue to red.8 0. when the laser is scanned across the cavity resonance at 1518. respectively.98 0.6 0. (2. Ω2 )2 + (ΩΓm )2 (2. (b) Linear scan of the averaged cavity transmission of Sample I at an input power of 2. The dashed line indicates the laser frequency detuning used to record the power spectral density shown in the top panel of Fig. and mx are the resonance frequency.2 0 0. with the experimental data in blue and the theoretical ﬁtting in red.12) where Ωm .9 µW. by the following equation: d2x FT (t) dx + Γm + Ω2 x = .57nm] (pm) (b) 1532 1516 1520 1524 1528 Wavelength Detuning (pm) Wavelength (nm) Figure 2.8 1 0.
17) provide a formal solution of the forward WGM: p a f (t) = i κe Ain where f (τ) e(i∆0 κ/2)τ +∞ τ 0 x(t cos(ητ) f (τ)e 0 igOM τ0 )dτ0 dτ. Using Eq. normalized such that Pin = jAin j2 represents the input power.35 The time correlation of the mechanical displacement is thus given by hx(t)x(t + τ)i = 1 2π +∞ Sx (Ω)e ∞ iΩτ dτ hx2 iρ(τ) hx2 ie Γm jτj/2 cos Ωm τ.19) h(τ) 0 ρ(τ1 τ2 )dτ1 dτ2 . (2. (2.15) where hx2 i = kB T /(mx Ω2 ) is the variance of the thermal mechanical displacement and ρ(τ) is the m normalized autocorrelation function for the mechanical displacement. igOM x)ab + iηa f . Ain is the input optical wave. one forward and the other backward propagating. In the case of a continuouswave input. ∆0 = ω ω0 is the frequency detuning from the input wave to the cavity resonance and η is the mode coupling coefﬁcient. normalized such that U j = ja j j2 ( j = f . we ﬁnd that the statisti cally averaged intracavity ﬁeld is given as: p ha f (t)i = i κe Ain where ε g2 hx2 i and h(τ) is deﬁned as OM τ +∞ cos(ητ) f (τ)e 0 ε 2 h(τ) dτ.16) (2. (2. (2. b) represents the mode energy. (2. circulate inside the microcavity and couple via Rayleigh scattering from the surface roughness. κ is the photon decay rate for the loaded cavity.18) represents the cavity response. (2.15). To be general.17) κ/2 κ/2 where a f and ab are the forward and backward whisperinggallery modes (WGMs). (2.20) . Eqs. The optical ﬁelds inside the cavity satisfy the following equations: da f = (i∆0 dt dab = (i∆0 dt p igOM x)a f + iηab + i κe Ain . and κe is the photon escape rate associated with the external coupling. we consider a doublet resonance in which two optical ﬁelds.16) and (2.
The strong dynamic backaction of the ﬂapping mode (under thermal excitation) also produces a broadband spectral background in κe κi .95 MHz). top panel) exhibits three clear frequency components at 8. These values agree well with FEM simulations of the differential ﬂapping mode (7.21) κe Pin κ iη 2κ κ 2iη e 0 ε 2 h(τ) [ fc (τ) + fs (τ)] dτ + c.c. as shown in Fig. 2. and 27. Using the theory developed above and ﬁtting the experimental averaged cavity transmission spectrum. (2. we can ﬁnd the averaged energy for the forward WGM as: +∞ hU f (t)i = κe Pin = where f j (τ) gate. 2.25) .36 Similarly.. 2.2 MHz) and second (28.23) In the case of a singlet resonance. κ/2)τ ( j = c.22) hPT i/Pin . η = 0 and Eq.6c).c. 2 + (κ/2)2 ∆0 (2.23) reduces to the simple form expression hT i = 1 κe κi κ +∞ e 0 ε 2 h(τ) [ f (τ) + f (τ)] dτ. (2.5(b) for Sample I. The same approach is used to describe the cavity transmission of Sample II. gOM = 0 and Eq. hT i hT i = 1 κe κi 2κ 1 Ain a f (t) . . we obtain the optical Q factor of the resonance.24) In the absence of optomechanical coupling.7 MHz) order radial breathing modes (Fig. (2.30.c. e(i∆ j 0 f (τ1 ) f (τ2 ) cos(ητ1 ) cos(ητ2 )e +∞ ε 2 h(jτ1 τ2 j) dτ1 dτ2 (2. c. with ∆c = ∆0 + η and ∆s = ∆0 η.6(a). The radiofrequency (RF) power spectrum of the optical signal transmitted through the cavity (Fig. thus takes the form +∞ iηκe κi (κ 2iη) e 0 ε 2 h(τ) [ fc (τ) + fs (τ)] dτ + c. denotes complex conju As the transmitted power from the double disk is given by p PT (t) = Pin + κeU f (t) + i κe Ain a f (t) the averaged cavity transmission.6.24) reduces to the conventional form of T =1 as expected. (2. 13.9 MHz corresponding to thermallyactuated resonances of the doubledisk structure. given in Fig. and the ﬁrst (14. (2. 2.6(b). s).
τ1 . we ﬁnd the autocorrelation func tion for the power ﬂuctuation to be hδP(t)δP(t + t0 )i where ψ(t0 . They take the following forms: +∞ A0 = Ain 1 κe cos(ητ) f (τ)dτ 0 +∞ τ ˆ Ain A0 . τ2 ). the power ﬂuctuations. limited by the squeezeﬁlm process of the nitrogen gas between the disks [52].31) . (2. red curve) shows that the ﬂapping mode has a 3dB linewidth of 2. τ2 ) 0 dτ0 1 τ2 0 dτ0 ρ(t0 + τ0 2 1 τ0 ).29) ˆ A0 f (τ)]. A correct description of the power spectrum (Fig. accurate for arbitrarily strong optomechanical effects.95).6(b).37 the RF spectrum with a shoulder at the second harmonic frequency. The transmitted power then becomes P(t) = jAT (t)j2 jA0 j2 + A0 δA(t) + A0 δA (t). τ1 .27) (2. and a nonperturbation theory. QM = 3. It is easy to PT (t) show that hδA(t)i = 0 and hPT (t)i = jA0 j2 .28) δA(t) = igOM κe Ain dτ cos(ητ) f (τ) 0 τ0 )dτ0 .15). We can describe the power spectral density of the cavity transmission in the presence of mechanical Brownian motion using a linearperturbation approximation when the optomechanical effects are small. τ1 . By using Eq. (2. (2. become +∞ τ δP(t) where u(τ) ˆ iκe cos(ητ)[A0 f (τ) gOM Pin dτu(τ) 0 0 x(t τ0 )dτ0 . 2. p In this case. x(t 0 (2. If the induced optomechanical perturbations are small.30) ψ(t0 . τ2 ) is deﬁned as τ1 2 εPin +∞ 0 dτ1 dτ2 u(τ1 )u(τ2 )ψ(t0 . (2. (2. when the effects are larger.26) A0 + δA(t). 2 (2.1 MHz (mechanical Qfactor. As a result. Both analyses are presented here. δP(t) hPT (t)i.18) can be approximated as a f (t) p i κe Ain +∞ τ cos(ητ) f (τ) 1 0 igOM x(t 0 τ0 )dτ0 dτ. the transmitted optical ﬁeld can be written as AT (t) = Ain + i κe a f (t) where A0 is the transmitted ﬁeld in the absence of the optomechanical effect and δA is the induced perturbation. Eq.
14) and H(Ω) is the cavity transfer function given by 1 Ω +∞ 0 2 H(Ω) = u(τ)(eiΩτ 1)dτ . the can be well approximated by H(Ω) cavity transfer function is given by a simple form of H= 4κ2 κ2 ∆2 e i 0 ∆2 + (κ/2)2 0 4 .36) shows that the autocorrelation function involves various correlations between the intracavity energy and ﬁeld. all of which can . (2. the power spectral density of the cavity transmission is directly proportional to the spectral intensity of the mechanical displacement. 2). (2. is given by hδP(t1 )δP(t2 )i = κ2 hU f 1U f 2 i e + iκe 3/2 κe h Ain a f 1 Ain a f 1 Ain a f 2 Ain a f 2 i Ain a f 1 i (2. From Eq. in the sidebandunresolved regime. The situation becomes quite complicated when the optomechanical effects are large. (2. if the optomechanical effect is small. the autocorrelation function for the power ﬂuctuation of the cavity transmission.30). (2.35) Therefore. Eq.33) In the case of a singlet resonance. For a speciﬁc mechanical mode at the frequency Ωm . the photon decay rate inside the cavity is much larger than the mechanical damping rate. 2 + (κ/2)2 2 [(∆0 + Ω)2 + (κ/2)2 ] [(∆0 Ω)2 + (κ/2)2 ] ∆0 κ2 e (2. κ Γm . (2. where U f j = U f (t j ) and a f j = a f (t j ) ( j = 1.34) In most cases. the cavity transfer function takes the form: H(Ω) = 4∆2 (κ2 + Ω2 ) 0 i .32) where Sx (Ω) is the spectral intensity of the mechanical displacement given in Eq.22). the cavity transfer function H(Ωm ).32) shows clearly that. OM (2. In particular. (2. we obtain the power spectral density SP (Ω) of the cavity transmission to be SP (Ω) 2 g2 Pin H(Ω)Sx (Ω). δP(t) PT (t) hPT i.36) hU f 1 Ain a f 2 Ain a f 2 i + hU f 2 Ain a f 1 p 2 κe hU f i + i κe Ain ha f i Ain ha f i . Equation (2.38 Taking the Fourier transform of Eq.
Following a similar approach. ψ(t2 t1 . f j = f (τ j ) (with j = 1.41) . (2. Eq.37) where. we can ﬁnd the other correlation terms in Eq.C1C2 ).39 be found using Eqs.C1C2 ) 0 dτ1 dτ2C1C2 e ε 2 (τ2 +τ2 ) f f e 1 2 1 2 ετ1 τ2 ρ + f1 f2 eετ1 τ2 ρ + c. As the Equations (2. since h(τ) = ψ(0.37) becomes negligible when τ1 approximated as t1 .39) where ∆t = t2 t1 and Φ(∆t. we ﬁnd that the autocorrelation function of the power ﬂuctuations is given by hδP(t1 )δP(t2 )i 2 κ2 Pin Φ(∆t. respectively. τ. the integrand in Eq. (2.31). τ1 . (2. τ). (2. in the integrand. and ψ = ψ(t2 t1 . Using these terms in Eq. Therefore. we can ﬁnd the following correlation for the intracavity ﬁeld: h Ain a f 1 2 = κe Pin Ain a f 1 +∞ 0 Ain a f 2 Ain a f 2 i ε 2 (h1 +h2 ) dτ1 dτ2C1C2 e f1 f2 e εψ + f1 f2 eεψ + c.15) and (2. h(τ) τ2 . τ2 ) are given by Eqs. t1 . in the sidebandunresolved regime.20) and (2. τ1 . For example. (2.40) with ρ = ρ(∆t). Therefore.20) and (2.c. . (2. τ2 ) = 1 2πhx2 i τ1 τ2 2πhx2 i +∞ ∞ +∞ ∞ 2/κ or τ2 2/κ. κ cavity response function f (τ) decays exponentially with time at a rate of κ/2.C1C2 ) is deﬁned as +∞ Φ(∆t. 1 dΩ (2. τ2 ) can be well ψ(t2 Sx (Ω) e Ω2 Sx (Ω)e iΩ(t2 t1 ) e iΩτ1 1 eiΩτ2 t1 ). τ1 . (2.36).37) becomes h Ain a f 1 Ain a f 1 Ain a f 2 Ain a f 2 i 2 κe Pin Φ(∆t. 2). C j = cos(ητ j ). h j = h(τ j ). However.36). (2. τ1 . τ1 .18). (2. .38) iΩ(t2 t1 ) dΩ = τ1 τ2 ρ(t2 Similarly.c. σ1 σ2 ) e p κe hU f i + i κe Ain ha f i Ain ha f i 2 .31) show that h(τ) and ψ(t2 of 1/Ωm and 1/Γm . h(τ) and ψ(t2 t1 . τ2 ). τ2 ) vary with time on time scales Γm and κ Ωm . (2.
46). we obtain the ﬁnal form for the autocorrelation of the power ﬂuctuations: hδP(t1 )δP(t2 )i 2 κ2 Pin [Φ(∆t.21) show that. (2. n! n=0 +∞ ∑ (2. σ1 σ2 )] .42) σ(τ) Moreover.45) Using this term in Eq.48). (2.40) and using it in Eq.40 where σ j = σ(τ j ) ( j = 1.41) as p κe hU f i + i κe Ain ha f i +∞ Ain ha f i κe Pin σ(τ) [ f (τ) + f (τ)] e 0 ε 2 2τ dτ. Eq.43) τ2 )2 0 f (τ1 ) f (τ2 ) cos(ητ1 ) cos(ητ2 )e dτ1 dτ2 . η = 0 and σ(τ) simpliﬁes considerably to σ = κi /κ. ε 2 (τ1 (2. (2. (2. (2. we obtain the autocorrelation function for the power ﬂuctuation in the following form hδP(t)δP(t + t0 )i where Gn is deﬁned as +∞ 2 κ2 Pin ∑ e εn ρn (t0 ) jGn + ( 1)n Gn j2 . in the sidebandunresolved regime. σ1 σ2 ) e Φ(∞. ha f i and hU f i are well approximated by ha f (t)i hU f (t)i p i κe Ain κe Pin +∞ cos(ητ) f (τ)e 0 +∞ ε 2 2τ dτ.49) 0 In the case of a singlet resonance. In general.47) Substituting this expression into Eq.48) Gn τn σ(τ) f (τ)e ε 2 2τ dτ. (2.44) Therefore.19) and (2. 2) and σ(τ) is deﬁned as 1 κe (κ2 + 2η2 ) ηκe sin(ητ). (2. (2.46) in Eq. n! n=1 +∞ (2. the power spectral density of the cavity transmission is given by the Fourier transform . (2. we obtain the ﬁnal term in Eq. cos(ητ) + 2 2 + 4η2 ) κ(κ κ + 4η2 (2. The autocorrelation function for the power ﬂuctuation is still described by Eq. ετ1 τ2 ρ(∆t) (2.40) can It can be further simpliﬁed if we notice that the exponential function e be expanded in a Taylor series as e ετ1 τ2 ρ(∆t) = ( ετ1 τ2 )n n ρ (∆t). (2.41).
if the fundamental mechanical linewidth is broad. (2. As shown in Fig.6(b) for Sample I. the thermal mechanical motion creates spectral components around the harmonics of the mechanical frequency with broader linewidths.51) Eq.5 2. Eq.15)]. (2.14) shows that the spectral intensity of the mechanical displacement can be approximated by a Lorentzian function. (2. but also their frequency sums and differences. This theory can be extended easily for the case with multiple mechanical frequencies.50) where Sn (Ω) is deﬁned as Sn (Ω) = ∞ ρn (τ)eiΩτ dτ. if the optomechanical coupling is signiﬁcant. we can see that. As shown clearly in Fig. 2. producing a broadband spectral background. the induced parametric mechanical instability . as shown in the top panel of Fig. 2.52) e Γm jτj/2 cos Ω τ m [see Combining Eq. As a result. various frequency components on the power spectrum would smear out. with an input optical power of 760 µW launched at the blue detuned side of the resonance.52).1 Regenerative oscillation Ambient pressure Despite the nearunity mechanical quality factor of the ﬂapping mode. 2. (2.5.6(b). the second harmonic is clearly visible. 2.6(b).51) becomes Sn (Ω) 1 n n! nΓm .41 of Eq. the power spectrum only only exhibits harmonics of each mechanical frequency. the frequency components near 0 MHz are the differential frequencies and those near 1820 MHz are the second harmonic and sum frequencies. This phenomenon is similar to the randomﬁeldinduced spectral broadening in nuclear magnetic resonance [53] and atomic resonance ﬂuorescence [54]. In this case.50) and (2.48): 2 S p (Ω) = κ2 Pin ∑ e εn Sn (Ω) jGn + ( 1)n Gn j2 . the powerful dynamic backaction in the doubledisk structure provides sufﬁcient compensation of mechanical loss to excite regenerative mechanical oscillation. resulting in an approximated ρ(τ) given as ρ(τ) Eq. As shown in the bottom panel of Fig. 2. (2. n! n=1 +∞ +∞ (2. In particular.6(d). (2. ∑ 2n k=0 k!(n k)! (nΓm /2)2 + [(2k n)Ωm + Ω]2 (2.
76 mW.5 50 0 0 0. (d) Recorded transmission waveform of Sample I for Pi = 0. (c) FEM simulated mechanical modes indicated in (b).4 0.94 2 Qm = 4.4 0.42 1 Power Spectral Density (dBm/Hz) Normalized Transmission 90 100 110 120 130 90 100 110 0 1 Qm = 3.07 103 80 90 100 110 8.75 106 −30 −20 −10 0 10 20 30 [λ−1518. Sample I) doubledisk cavity. (b) Upper panel: optical transmission power spectral density (PSD) of a Sample I doubledisk in the 1 atm. (e) Comparison of experimental (blue curve) and simulated (red curve) waveforms. The inset shows the fundamental TElike bonded mode at λ = 1518. Lower panel: transmission PSD of a small diameter (D = 54 µm.2 0.95 b 3 4 0.8 µW. theoretical modeling in red.57 nm. . Experimental data in blue.6 0.4 9.6: (a) Optical transmission spectrum of a large diameter (D = 90 µm. nitrogen environment for Pi = 5.98 0.57nm] (pm) a 1532 1516 1520 c 1 Wavelength (nm) 1524 1528 10 Frequency (MHz) 20 30 40 50 2 3 4 d Normalized transmission 1 1 e Mechanical displacement (pm) 0 25 0. The inset shows a zoomin of the spectrum around the fundamental ﬂapping mode frequency.6 0.5 0.6 9 9.8 1 0 0 50 100 150 200 Time (µs) Time (ns) Figure 2. and detector noise background in yellow.2 0 1 0.2 8.8 0. with the corresponding simulated mechanical displacement (green curve). Sample II) doubledisk cavity in vacuum for Pi = 44 nW.8 Frequency (MHz) Qi = 1.96 0.
The coupling depth at the kink point. In particular.6e) agrees well with our numerical simulation which shows that the gradient force actuates an extremely large (50 pm) mechanical displacement amplitude.7 shows an example of the cavity transmission of Sample I. two sequential passes of the cavity resonance across the laser frequency can be seen. The mechanical ﬂapping mode starts to oscillate when the input laser frequency is scanned across a certain detuning. The optomechanical oscillations are simulated through the following coupled equations governing the intracavity optical ﬁeld and mechanical motions. An estimate of the threshold detuning (∆th ). Figure 2. the same magnitude of optomechanical oscillation is excited over a broad range of laser blue detuning. 2. a map of which can be used to quantify the strength of the dynamic backaction. The detuning dependence of the optomechanical ampliﬁcation coefﬁcient can be lumped into a . A zoomin of the recorded time waveform (Fig.54) where we have counted in both the thermal Langevin force FT and the optical gradient force Fo = gOM jaj2 ω0 for actuating mechanical motions. The intense transmission oscillations cover the entire coupling depth. m dt 2 dt mx mx (2. Within this detuning value.43 causes the cavity transmission to oscillate over the entire coupling depth with a fundamental frequency of 13. dt 2 dx d2x FT (t) Fo (t) + Γm + Ω2 x = + . 2. dragging the cavity resonance over more than 10 cavitylinewidths and leaving distinctive features of the Lorentzian cavity transfer function. ∆Tth .7). along with an overshoot and oscillation of the transmitted optical power resulting from the quick release of Doppler shifted photons from the cavity. from which we can obtain the threshold frequency detuning ∆th . The threshold for regenerative oscillation depends sensitively upon the optical input power and the average lasercavity resonance detuning. 2.97 MHz (this value is about 68% larger than the intrinsic mechanical frequency due to the optical spring effect [55]). leaving an abrupt kink on the transmission spectrum. for a given input power. corresponds to the threshold coupling at the given power level. can be determined from the abrupt kink in the cavity transmission that marks the onset of regenerative oscillation (Fig. respectively: p da κ = (i∆0 igOM x)a + i κe Ain .53) (2.8(a) and Fig.
7: Scan of the cavity transmission of Sample I at an input power of 0.56) where Γm = 2. as expected. Fitting of eq.4 0. respectively. 0.8(b). 2.6(d). Γm (2.8(b) shows a map of f (∆th ) versus optical input power for the 90 µm diameter doubledisk cavity in the heavily damped nitrogen environment. The dashed line indicated the laser frequency detuning used to record the timedependent cavity transmission given in Fig.4 GHz/nm.1 MHz is the bare mechanical damping rate of the ﬂapping mode.55) where κ = κi + κe is the total photon decay rate of the loaded cavity. 2. (2. shows a linear dependence of f (∆th ) on input power. and is well described in the unresolved sideband regime [50] by f (∆th ) = 2g2 Pi OM = ωc mx Γm κ3 i 2B Pi .6(b). The data in Fig.8 agreement with the simulated result of 33 GHz/nm. in good 2.061 MHz/µW.2 0 −30 −20 −10 0 10 Wavelength Detuning (pm) Figure 2. f (∆) ∆2 + (κ/2)2 κκe κ3 ∆ i (∆ + Ωm )2 + κ 2 2 (∆ Ωm )2 + κ 2 2 .44 1. measurements were also performed in vacuum (P < 5 10 4 Torr). 2.6 0.2 Vacuum In order to eliminate the squeezeﬁlm damping of the nitrogen environment.56) to the data in Fig. with the instantaneous and averaged signals shown in blue and red. 2. The signiﬁcantly reduced mechanical linewidth in vacuum shows that the ﬂapping mode consists of a small cluster of modes (Fig. (2.76 mW.8(b) yields a dynamic backaction parameter of B = 0. single detuning function.0 Normalized Transmission 0.5. The right panel of Fig. 2. bottom .8 0. corresponding to an optomechanical coupling factor of gOM /2π = 33.
05 I II III I 80 70 60 50 40 30 20 10 00 vacuum air f = 8. .45 Normalized transmission Threshold detuning function f(∆th) 1 0.5 2 0 200 400 600 800 Displacement spectral density (m /Hz) 10 10 10 10 10 3 4 5 6 7 Frequency (MHz) 8 o sing increa ca p ti lp we 10 o E ective temperature (K) 10 c d 100 r 10 10 6.8 8 8. nitrogen environment. the solid green (dashed black) curve is a theoretical curve obtained using the estimated Bparameter from the left panel of (b) and the experimental (optimal) detuning of p ∆ = −1. (b) f (∆th ) as a function of optical input power. Bottom panel: the transduction amplitude of the frequency component at 9.53 MHz and its higherorder harmonics.45(κ/2) (inset shows the displacement sensitivity at the highest input power with the second optical attenuator removed).45(κ/2) ∆ = −(κ/2)/ 5 . Right panel: Sample I in a 1 atm.4 8. with a laser detuning of ∆ = −1.53 MHz 0.6 1 0.63 MHz a 10 Spectral density (m2/Hz) b 0. In (d).8 7 7. recorded for Sample II in vacuum.6 f = 9. and (d) the corresponding effective temperature.4 7.2 7.8 0. Middle panel: the transduction amplitude of the frequency component at 8. Left panel: Sample II in vacuum (inset shows the minimum achievable threshold (green arrow)).05 0.6 7.2 0.2 8. respectively.1 Frequency (MHz) Input power (µW) 1 10 Figure 2.2 0. the red curve is a ﬁt to the data.8: (a) Top panel: Normalized cavity transmission for Sample II in vacuum and Pi = 11 µW.6 0.63 MHz and its higherorder harmonics. Blue and red traces show the instantaneous and lowpassﬁltered signals. (c) Spectral intensity of the thermallydriven fundamental ﬂapping mode at various input powers.5 1 Wavelength detuning (pm) 8 6 4 2 0 2 4 6 Input power (µW) 1.
the net variations in the average disk spacing induced by the higherorder ﬂapping modes (with azimuthal mode number 1) is not zero. The intracavity ﬁeld can thus be approximated as a(t) a0 (t) + δa(t). 2. Figure 2. Extrapolation of the experimental data using Eqs. the optomechanical coupling for the fundamental ﬂapping mode.63 MHz as shown in the bottom two panels of Fig. The onset of regenerative oscillation coincides with a frequency shift in the fundamental ﬂapping mode to 9.46 panel). the optomechanical effect during mechanical cooling is well described by linear perturbation theory since the thermal mechanical motion is signiﬁcantly suppressed. Measurements of the optical spring effect indicates that the optical ﬁeld renormalizes the cluster of modes.8(a). Three different regimes can be clearly seen: (I) transduction of thermal motion. However. due to the asymmetry in practical devices. where a0 is the cavity ﬁeld in the absence of optomechanical coupling and δa is the perturbation induced by the thermal mechanical motion. (2. but it is nearly zero for ﬂapping modes with higherorder azimuthal mode numbers. Because of the extremely short roundtrip time of the cavity mode. (2. is maximum. coupled together due to deviations in circularity of the undercut region and support pedestal. However. .8(b) shows a plot of the invacuum f (∆th ) versus input power. with a measured minimum threshold power of Pi = 267 nW. With an invacuum QM = 4070 (Fig. inset). their optomechanical coupling is weak and does not provide efﬁcient dynamic back action. the optomechanical effect is governed by Eqs. From Eq. with the lowestfrequency mode at 8. 2. As a result.56) to the optimal detuning point shows a minimum threshold power of only 40 nW. In general. the fundamental ﬂapping mode has an extremely low threshold input power for regenerative oscillation. and their thermal motion is visible in the transmission power spec trum. The left panel of Figure 2. (2. (II) onset of opticallydriven oscillation.6 Optomechanical cooling In general. which has a ﬂapping amplitude uniformly distributed around the disk perimeter.6(b).54).53 MHz transforming into the fundamental ﬂapping mode with uniformly distributed displacement along the disk perimeter (the rest of the modes decouple from the light ﬁeld). 2.53). the optical wave is sensitive only to the variations of averaged disk spacing around the whole disk.8(a) shows a transmission spectrum when the laser is scanned across the cavity resonance. These modes are a mixture of the lowerlying azimuthal modes.55) and (2. and (III) optically damped motion.53) and (2.
ω0 [(∆0 + Ω)2 + (κ/2)2 ] [(∆0 Ω)2 + (κ/2)2 ] (2.63) together with (2. and can be removed simply by shifting the zeropoint of the mechanical displacement to the new equilibrium position. we neglect this term in the following discussion. Therefore. a0 = κ/2 i∆0 and Eq. (2.62) As expected. is given by (2.60).47 they are found to satisfy the following equations: da0 = (i∆0 dt dδa = (i∆0 dt p κ/2)a0 + i κe Ain . x(Ω) where δa(Ω) is the Fourier transform of δa(t) deﬁned as δa(Ω) = is the Fourier transform of x(t). Fo = Fo (t) = gOM jaj2 ω0 . i(∆0 + Ω) κ/2 +∞ iΩt ∞ δa(t)e dt. δa(Ω) = igOM a0 x(Ω) . Eq.63) Equation (2. (2.54) can be solved easily in the frequency domain. which becomes (Ω2 m Ω2 iΓm Ω)x = FT Fo + . κ/2)δa igOM xa0 .57) (2.58) In the case of a continuouswave input.62) provides the simple form for the thermal mechanical displace . (2. mx mx (2. From Eq. The second and third terms provide the dynamic optomechanical coupling.61) gOM ja0 j2 + a0 δa(t) + a0 δa (t) . ω0 The ﬁrst term is a static term which only affects the equilibrium position of the mechanical motion.60) Similarly. the gradient force is given by the following equation in the frequency domain: Fo (Ω) = 2g2 ja0 j2 ∆0 x(Ω) ∆2 Ω2 + (κ/2)2 + iκΩ 0 OM . Equation (2.59) (2. (2. (2.58) provides the spectral response for the perturbed ﬁeld amplitude. the gradient force is linearly proportional to the thermal mechanical displacement.57) gives a steadystate value given as: p i κe Ain . The optical gradient force.
(2. the OM optical wave damps the thermal mechanical motion and thus increases the energy decay rate. x(Ω) = where Ω0 and Γ0 are deﬁned as m m (Ω0 )2 m Ω2 + m 2g2 ja0 j2 ∆0 ∆2 Ω2 + (κ/2)2 0 OM mx ω0 [(∆0 + Ω)2 + (κ/2)2 ] [(∆0 Ω)2 + (κ/2)2 ] 2g2 ja0 j2 ∆0 ∆2 Ω2 + (κ/2)2 m 0 Ω2 + OM . As shown in Fig. the mechanical frequency is modiﬁed.66) show clearly that the primary effect of the optical gradient force on the mechanical motion is primarily to change its mechanical frequency (the socalled optical spring effect) and energy decay rate to the new values given by Eqs. Using Eqs.67) [(Ω0 )2 m 2Γm kB T . h(δx)2 i = 1 2π +∞ Sx (Ω)dΩ = ∞ kB T Γm . On the red detuned side.68) Cooling the mechanical motion reduces the spectral magnitude and the variance of thermal displacement. At the same time.64)(2.48 ment.64) (2. (2.8(c) for Sample II in vacuum.65) and (2.66) Equations (2. The large mechanical ampliﬁcation of the doubledisk NOMS implies a correspondingly efﬁcient cooling of mechanical motion on the reddetuned side of the cavity resonance.14): Sx (Ω) = which has a maximum value Sx (Ω0 ) = m 2Γm kB T /mx . m mx ω0 [(∆0 + Ωm )2 + (κ/2)2 ] [(∆0 Ωm )2 + (κ/2)2 ] 2g2 ja0 j2 κ∆0 1 OM Γm 2 + (κ/2)2 ] [(∆ mx ω0 [(∆0 + Ω) Ω)2 + (κ/2)2 ] 0 2g2 ja0 j2 κ∆0 1 OM Γm .66). Ω2 ]2 + (ΩΓ0 )2 m (2. The efﬁciency of optomechanical control is determined by the ﬁgure of merit g2 /mx . 2.65) Γ0 m (2.13) and (2. mx ω0 [(∆0 + Ωm )2 + (κ/2)2 ] [(∆0 Ωm )2 + (κ/2)2 ] FT mx (Ω0 )2 m 1 Ω2 iΓ0 Ω m .64). mx (Ω0 Γ0 )2 m m The variance of the thermal mechanical displace ment is equal to the area under the spectrum. we ﬁnd that the spectral intensity of the thermal displacement is given by a form similar to Eq. the spectral intensity of the fundamental ﬂapping mode de . decreasing with increased cavity energy in the sidebandunresolved regime. (2. mx (Ω0 )2 Γ0 m m (2. (2.
from which. the integrated spectral area obtained from the experimental spectrum is the averaged variance of thermal mechanical displacement.71) Therefore. the experimentally recorded displacement specm trum is given by the averaged spectrum Sx (Ω) = Sx (Ω)p(Ω0 )dΩ0 .69) The area under the displacement spectrum thus provides an accurate measure of the effective temperature. with a probability density function of p(Ω0 ). In practice. T0 Ωm h(δx)2 i0 0 (2. (2. m m (2.73) . Even for the strongest damping levels.70) where Sx (Ω) is given by Eq. where Ωm 0 0 (2. the effective temperature is thus given by T eff (Ωm )2 h(δx)2 i = 2 . (2. the effective temperature can be inferred from the thermal mechanical energy using the equipartition theorem: kB Teff = mx (Ω0 )2 h(δx)2 i.67) and we have assumed measured spectral area is thus 1 2π +∞ p(Ω0 )dΩ0 = 1. m (2.8(c) shows good signal to noise for the transduced motion due to the high displacement sensitivity of the doubledisk (7 10 17 m/Hz1/2 .49 creases dramatically with increased input power. we obtain the effective average temperature kB T eff = mx (Ωm )2 h(δx)2 i. The experimentally m m Sx (Ω)dΩ = ∞ h(δx)2 ip(Ω0 )dΩ0 m m h(δx)2 i.66)]. A measure of the optical cooling can be determined from the integrated area under the displacement spectrum [56]. accompanied by a signiﬁcant broadening of the mechanical linewidth. For a mechanical mode in thermal equilibrium. ﬂuctuations on the laser frequency detuning may cause the mechanical frequency and damping rate to ﬂuctuate over a certain small range [Eq.72) Ω0 p(Ω0 )dΩ0 is the center frequency of the measured displacement spectrum Sx (Ω). (2.65) and (2. As a result. m m m Compared with the room temperature. 2. as limited by the background level). the inset to Fig. according to the equipartition theorem.
50 where h(δx)2 i0 is the displacement variance at room temperature, given by the spectral area at T0 . Figure 2.8(d) plots the inferred temperaure, Teff , which drops down to 12.5 K for a maximum input power of Pi = 11 µW (Pd = 4.4 µW). In principle, the effective temperature is related to the optical damping rate (Γm,opt ) through the relation T0 /Teff = 1 + Γm /Γm,opt , where T0 = 300 K is the bath temperature. In Fig. 2.8d the red curve is a ﬁt of the measured cooling curve using the relation T0 /Teff = 1 + αPi , whereas the green curve represents the expected cooling curve for the dynamic backaction parameter (B = 0.032 MHz/µW) determined from the threshold plot in the right panel of Fig. 2.8(b) and the experimental lasercavity detuning (∆ = 1.45(κ/2)). For comparison, we
have also plotted (dashed black line) the theoretical cooling curve in the case of optimal lasercavity p detuning (∆ = (κ/2)/ 5). The difference between the two theoretical curves and the measured data, along with the limited range of optical input power studied, can largely be attributed to issues associated with the limited bandwidth and range of our current cavity locking scheme (a problem exacerbated by the very large transduction of even the Brownian motion of the disks). As the dashed black curve indicates, technical improvements in the cavity locking position and stability should enable temperature compression factors of 20 dB for less than 1 µW of dropped power.
2.7
Discussion
The large dynamic backaction of the doubledisk cavity, primarily a result of the large perphoton force and small motional mass of the structure, opens up several areas of application outside the realm of more conventional ultrahighQ cavity geometries. This can be seen by considering not only the efﬁciency of the cooling/ampliﬁcation process, but also the maximum rate of effective cooling/ampliﬁcation, the scale of which is set by the optical cavity decay rate [57, 58]. In the doubledisk cavities presented here, the dynamic backaction parameter is B cavity decay rate of κ/2π 0.06 MHz/µW for a
100 MHz. The combination allows for higher mechanical frequencies
of operation, where the bare damping is expected to scale with frequency, and makes possible enormous temperature compression ratios. A quantum mechanical analysis of the optical selfcooling p process [57, 58], indicates that the sideband resolved regime (κ 32Ωm ) is necessary to reduce the phonon occupancy below unity. Having already achieved optical Qfactors in excess of 106 , and planar silica microdisks having already been demonstrated with Q > 107 [59], we expect that further optimization of the doubledisk NOMS will be able to extend its operation well into the sideband resolved regime. The combination of large dynamic backaction parameter and large maximum
51 ampliﬁcation rate also present intriguing possibilites for sensitive, high temporal resolution force detection [60], particularly in heavily damped environments such as ﬂuids for biological applications [46, 47]. Other application areas enabled by the chipscale format of these devices include tunable photonics [42–44], optical wavelength conversion [61], and RFoveroptical communication.
52
Chapter 3
Coherent Mechanical Mode Mixing in Optomechanical Nanocavities
3.1 Introduction
The coherent mixing of multiple excitation pathways provides the underlying mechanism for many physical phenomena. Wellknown examples include the Fano resonance [62] and electromagnetically induced transparency (EIT) [63], arising from the interference between excitations of discrete states and/or a continuum background. In the past few decades, Fanolike or EITlike resonances have been discovered in a variety of physical systems, such as electron transport in quantum wells/dots [64, 65], phonon interactions in solids [66, 67], inversionfree lasers [68, 69], coupled photonic microcavities [70–73], and plasmonic metamaterials [74]. Here we report a new class of coherent excitation mixing which appears in the mechanical degree of freedom of nanooptomechanical systems (NOMS). We use two canonical systems, coupled microdisks and coupled photoniccrystal nanobeams, to show that the large optical stiffening introduced by the optical gradient force actuates signiﬁcant coherent mixing of mechanical excitations, not only leading to renormalization of the mechanical modes, but also producing Fanolike and EITlike optomechanical interference, both of which are fully tunable by optical means. The demonstrated phenomena introduce the possibility for classical/quantum information processing via optomechanical systems, providing an onchip platform for tunable optical buffering, storage, and photonicphononic quantum state transfer. This work was initially presented in Ref. 75. Optical forces within micromechanical systems have attracted considerable interest of late due to the demonstration of alloptical ampliﬁcation and selfcooling of mesoscopic mechanical resonators [35–39]. This technique for sensing and control of mechanical motion relies on the radia
3. 48. creating a highly anistropic. strong optical gradient forces may be generated between the microdisks while maintaining the beneﬁts of the lowloss. nearinfrared (λ 1550 nm) optical supermodes of the beam pair. 76–78].2 Zipper cavity and doubledisk design. all of which are deposited via plasma . The zipper cavity is formed from a thinﬁlm (400 nm) of tensilestressed. fabrication. 45. 45. and optical characterization We have focused on two speciﬁc implementations of nanoscale cavity optomechanical systems. are used to form the released nanobeam structure. followed by a series of plasma and wet chemical etches. Clamping to the substrate at either end of the suspended beams results in a fundamental inplane mechanical beam resonance of frequency 8 MHz. The ﬁrst system consists of two patterned nanobeams in the nearﬁeld of each other. intensitydependent effective elastic modulus of the optomechanical structure.1(a). Electron beamlithography. it has been realized that guided wave nanostructures can also be used to generate extremely large perphoton optical forces via the gradient optical force [79]. 80. resulting in strong dynamical backaction between the cavity ﬁeld and mechanical motion. By creating a pair of microdisks. allowing the perimeter of the disks to vibrate in myriad of different ways. 55. shown in Fig. The combination of tailorable mechanical geometry. In this cavity structure the patterning of the nanobeams localizes light through Braggscattering. small motional mass. the doubledisk structure [48] is supported and pinned at its center. one on top of the other with a nanoscale gap in between. highFinesse optical cavity. forming what has been termed a zipper cavity [45. The second cavity optomechanical system is based upon the whisperinggallery microdisk optical cavity structure presented in Chapter 2 and Ref. 3. 33. In particular. and large perphoton force in such nanostructures results in a regime of operation in which the dynamic response of the coupled optomechanical system can signiﬁcantly differ from that of the bare mechanical structure. highQ (Q 106 ) character of the whisperinggallery cavity. The doubledisk structure is formed from a 158 nm sacriﬁcal amorphous silicon layer sandwiched in between two 340 nm thick silica glass layers. As shown schematically in Fig. stoichiometric Si3 N4 deposited by lowpressure chemical vapor deposition on a silicon substrate.1.53 tion pressure forces that build up in a mechanically compliant. in which dynamical backaction effects are particularly strong. 82]. 48. 81]. resulting in a series of high Finesse (F 3 104 ). the mechanical motion can be renormalized by the optical spring effect [29. More recently [44. 3.
The doubledisk bonded supermode has an optomechanical coupling coefﬁcient of gOM /2π 33 GHz/nm. The device studied here has a measured resonance wavelength of λc = 1538 nm and an intrinsic and loaded quality (Q) factor of 1. . (d) Schematic. a measured resonance wavelength of λc = 1545 nm. Additional details for both devices are in Refs. respectively. 45. and (f) FEMsimulated bonded (even parity) optical supermode of the zipper cavity. (c) FEMsimulated electric ﬁeld intensity of a transverseelectric (TE) polarized.07 106 and 0. bonded (even parity) whisperinggallery supermode between the two microdisks (shown in crosssection and for resonance wavelength λc 1550 nm).54 a b 1 µm c d e 1 µm s a f a w Figure 3.8 104 . 48. (e) SEM image.7 106 . and an intrinsic and loaded Qfactor of 3.1: (a) Schematic and (b) zoomedin scanning electron microscopic (SEM) image of the doubledisk NOMS. The zipper cavity bonded supermode has an optomechanical coupling coefﬁcient of gOM /2π 68 GHz/nm.0 104 and 2. respectively.
corresponding to 125 dBm/Hz near 10 MHz. is used to selectively excite the transverseelectric polarized optical modes of both cavities. This etch simultaneously undercuts the silicon substrate to form the underlying silicon pedestal. A ﬁbertaper optical coupling technique is used to incouple and outcouple light from the zipper and doubledisk cavities. and the correspondingly weak coupling of the common mode. The microdisk pattern was fabricated by reactive ion etching.55 enhanced chemical vapor deposition. in which the changing gap between the elements creates a large dispersive shift in the internally propagating cavity light ﬁeld. and the common motion. Due to the strong lightﬁeld coupling and dynamical backaction of the differential mode. An optical ﬁber polarization controller. The measured electrical noise ﬂoor is set by the circuit noise of the photodetector for the optical power levels considered in this work. with extremely lowloss (88% transmission efﬁciency). responsivity R = 1 A/W. transimpedance gain G=4 104 V/A) and a highspeed oscilloscope (2 Gs/s sampling rate and 1 GHz bandwidth). The theory for gradientforce optomechanical systems in which there is coupling between these two types of mechanical excitations is presented in the next sections. consisting of a series of circular loops of ﬁber. is put in contact with the substrate near the cavities in order to mechanically anchor it during all measurements (thus avoiding powerdependent movement of the taper due to thermal and/or optical forces). A pair of “dueling” calibrated optical attenuators are used before and after the cavities in order to vary the input power to the cavity while keeping the detected optical power level constant. in which both nanobeams or disks move together. . we term these two motional states opticallybright and opticallydark. and the gap remains approximately constant. RF spectra are measured by direct detection of the optical power transmitted through the cavities using a 125 MHz bandwidth photoreceiver (noiseequivalentpower NEP= 2.3 Theory of optomechanical effects in the presence of mode mixing Of particular interest in both the zipper and doubledisk systems are two types of motion: the differential motion of the nanobeams or disks.5 pW/Hz1/2 from 10200 MHz. The ﬁnal airgap between the silica disks size is measured to be 138 nm due to shrinkage of the amorphous silicon layer during annealing. resulting in mechanical motion that is decoupled from the light ﬁeld. A high temperature (1050 K) thermal anneal is used to improve the optical quality of the asdeposited silica layers. and the sandwiched αSi layer was undercut by 6 µm from the disk edge using a sulfur hexaﬂuoride dry release etch. The ﬁber taper. 3. respectively.5 pW/Hz1/2 from 010 MHz and 22.
normalized such that U = jaj2 represents the mode energy. Γt is the photon decay rate for the loaded cavity and Γe is the photon escape rate associated with the external coupling. gom is the optomechanical coupling coefﬁcient associated with the optically bright mode.2) (3.3) provides a spectral response for the perturbed ﬁeld amplitude of δa(Ω) = igom a0 xb (Ω) .1). the optical ﬁeld inside the cavity satisﬁes the following equation: da = (i∆0 dt Γt /2 igom xb )a + i Γe Ain .1) where a is the optical ﬁeld of the cavity mode. In Eq. Well below the threshold of mechanical oscillation. and Ain is the input optical wave.1 Intracavity eld in the presence of optomechanical coupling In the presence of optomechanical coupling.3) igom xb a0 .2) leads to a steady state given by p i Γe Ain . As a result. ∆0 = ω ω0 is the frequency detuning from the input wave to the cavity resonance. (3. a0 = Γt /2 i∆0 and Eq. i(∆0 + Ω) Γt /2 +∞ iΩt ∞ δa(t)e dt.3.5) Similarly. In the case of a continuouswave input. the mechanical motion is generally small. (3. where a0 is the cavity ﬁeld in the absence of optomechanical coupling and δa is the perturbation induced by the mechanical motion. with a mechanical displacement given by xb . the intracavity ﬁeld can be written as a(t) a0 (t) + δa(t). xb (Ω) where δa(Ω) is the Fourier transform of δa(t) deﬁned as δa(Ω) = is the Fourier transform of xb (t). we have neglected the optomechanical coupling to the optically dark mode because of its negligible magnitude. normalized such that Pin = jAin j2 represents the input power.4) (3. . (3. and its impact on the intracavity optical ﬁeld can be treated as a small perturbation. Eq. (3. (3. (3.56 3. They satisfy the following two equations: da0 = (i∆0 dt dδa = (i∆0 dt Γt /2)a0 + i Γt /2)δa Γe Ain .
(3.6) where A0 is the steadystate cavity transmission in the absence of optomechanical coupling. Γt /2 i∆0 (3.8) iΓe Pin gom xb (Ω) (Γ0 Γe )/2 + i∆0 Γt /2 i(∆0 + Ω) (Γt /2)2 + ∆2 0 where δPT (Ω) is the Fourier transform of δPT (t).3.6).8). we obtain a power spectral density (PSD) for the cavity transmission of 2 SP (Ω) = g2 Pin Sxb (Ω)H(Ω). Clearly then. 2 2 2 2 2 ∆2 + (Γt /2)2 [(∆0 + Ω) + (Γt /2) ] [(∆0 Ω) + (Γt /2) ] 0 Γ2 e (3. It is given by A0 = Ain (Γ0 Γe )/2 i∆0 . when compared with Sxb (Ω). we ﬁnd the power ﬂuctuations. By using Eqs. om (3. By using Eq. (3. Ω) (3.57 3. H(Ω) is a slowly varying function of Ω and can be well approximated by its value at the mechanical resonance: H(Ω) H(Ω0 ). as expected. H(Ω) is the cavity transfer function deﬁned as H(Ω) 4∆2 (Γ2 + Ω2 ) 0 0 . It is easy to show that the averaged cavity transmission is given by hPT i = jA0 j2 .7) where Γ0 is the photon decay rate of the intrinsic cavity. and (3. . (3. (3. the power mb spectral density of the cavity transmission is linearly proportional to the spectral intensity of the mechanical displacement of the optically bright mode.2 The power spectral density of the cavity transmission From the discussion in the previous section. are given in the frequency domain by (Γ0 Γe )/2 Γt /2 + i(∆0 i∆0 . δPT (t) δPT (Ω) = PT (t) hPT i.10) In general.7).9) where Sxb (Ω) is the spectral intensity of the mechanical displacement for the optically bright mode which will be discussed in detail in the following sections. the transmitted optical power from the cavity is given by 2 PT = Ain + i Γe a jA0 j2 + i Γe (A0 δa A0 δa ) .5).
k j . The second and . p j . j = b. (3. Fo = ﬁnd that it is given by Fo (t) = gom ja0 j2 + a0 δa(t) + a0 δa (t) . including the optical gradient force on the optically bright mode and counting in the mechanical dissipation induced by the thermal mechanical reservoir. d) are the mechanical displacement. and κ represents the mechanical coupling between the bright and dark modes.3 The mechanical response with multiple excitation pathways When the optically bright mode is coupled to an optically dark mode. With this system Hamiltonian.11) where x j . d) represents the Langevin forces from the thermal reservoir actuating the Brownian motion.12) (3.3. we neglect this term in the following discussion.14) where i. the Hamiltonian for the coupled mechanical system is given by the general form: Hm = p2 p2 1 2 1 2 b + kb xb + d + kd xd + κxb xd . T is the temperature and kB is the Boltzmann constant. with the following statistical properties in the frequency domain: hFi (Ωu )Fj (Ωv )i = 2mi Γmi kB T δi j 2πδ(Ωu Ωv ). we obtain the equations of motion for the two mechanical modes: dxb Fb Fo d 2 xb κ + Γmb + Ω2 xb + xd = + . ω0 (3. It can be removed simply by shifting the mechanical displacement to be centered at the new equilibrium position. From the previous section. Fi (Ω) is the Fourier transform of Fi (t).15) gom jaj2 ω0 represents the optical gradient force. The subscripts b and d denote the optically bright and optically dark modes.58 3. Fj ( j = b. the spring constant. md 2 dt dt md md where Ω2 j m kj mj (3.13) is the mechanical frequency for the jth mode. and the effective motional mass for the jth mechanical mode. kinetic momentum. d. mb dt 2 dt mb mb mb d 2 xd dxd Fd κ + Γmd + Ω2 xd + xb = . and m j ( j = b. (3. In Eq. respectively. Therefore. we The ﬁrst term is a static term which only changes the equilibrium position of the mechanical motion. respectively.12). 2mb 2 2md 2 (3.
Equations (3.5). (3. (3. mb mb (3. (3. From Eq. in which the two equations become Fb Fo κ xd = + . mb mb ω0 [(∆0 + Ωmb )2 + (Γt /2)2 ] [(∆0 Ωmb )2 + (Γt /2)2 ] 2g2 ja0 j2 Γt ∆0 1 om Γmb 2 + (Γ /2)2 ] [(∆ m b ω0 [(∆0 + Ω) Ω)2 + (Γt /2)2 ] t 0 1 2g2 ja0 j2 Γt ∆0 om . mb mb mb κ Fd Ld (Ω)xd + xb = .59 third terms provide the dynamic optomechanical coupling.20) and the new Ω0 and Γ0 are given by mS mS (Ω0 )2 mb ∆2 Ω2 + (Γt /2)2 2g2 ja0 j2 ∆0 om 0 mb ω0 [(∆0 + Ω)2 + (Γt /2)2 ] [(∆0 Ω)2 + (Γt /2)2 ] ∆2 Ω2 + (Γt /2)2 2g2 ja0 j2 ∆0 0 mb Ω2 + om . Γmb m b ω0 [(∆0 + Ωmb )2 + (Γt /2)2 ] [(∆0 Ωmb )2 + (Γt /2)2 ] Ω2 + mb (3. the gradient force is found to be given in the frequency domain by Fo (Ω) fo (Ω)xb (Ω) = ∆2 Ω2 + (Γt /2)2 + iΓt Ω 2g2 ja0 j2 ∆0 xb (Ω) om 0 .12) and (3.16) into Eq. Lb (Ω)xb + κ Fb xd = . the effect of the optical gradient force on the optically bright mode is primarily to change its mechanical frequency (the optical spring effect) and energy decay rate (mechanical ampliﬁcation or damping).17) can be written in the simple form.21) Γ0 mb (3. we ﬁnd that Eq.18) iΓm j Ω ( j = b.17) (3.17).13) can be solved easily in the frequency domain. . mb (3. (3. md md Lb (Ω)xb + where L j (Ω) Ω2 j m Ω2 (3.16) ω0 [(∆0 + Ω)2 + (Γt /2)2 ] [(∆0 Ω)2 + (Γt /2)2 ] which is linearly proportional to the mechanical displacement of the optically bright mode.19) where Lb (Ω) is now deﬁned with a new mechanical frequency Ω0 and energy decay rate Γ0 as mb mb Lb (Ω) = Ω2 mb Ω2 iΓmb Ω fo (Ω) mb (Ω0 )2 mb Ω2 iΓ0 Ω. d).22) Clearly. Substituting Eq. (3.
24) where Lb (Ω) is given by Eq. mb jLb (Ω)Ld (Ω) η4 j2 (3. The mechanical response given by Eq.60 Equations (3.4 The mechanical response with external optical excitation The previous section focuses on the case in which the mechanical excitations are primarily introduced by the thermal perturbations from the environmental reservoir.26) where δA(Ω) is the Fourier transform of δA(t).3. (3.18) and (3. By using Eq. (3.24) is very similar to the atomic response in EIT. the gradient force now becomes Fo (Ω) = fo (Ω)xb (Ω) + Fe (Ω). However.27) where fo (Ω) is given by Eq. i(∆0 + Ω) Γt /2 (3. In this case. (Ω) η4 Lb (Ω)Ld (3.23).14) and (3.25) This equation leads to the intracavity ﬁeld modulation given in the frequency domain as: p igom a0 xb (Ω) i Γe δA(Ω) δa(Ω) = . Eq. (3.15). By use of this solution together with Eq. we obtain the spectral intensity of the mechanical displacement for the optically bright mode.28) . (3. Sxb (Ω) = 2kB T η4 Γmd + Γmb jLd (Ω)j2 .23) represents the mechanical coupling coefﬁcient.20). (3.16) and Fe (Ω) represents the force component introduced by the input modulation. (3. the input optical wave is composed of an intense CW beam together with a small modulation: Ain = Ain0 + δA(t).3) now becomes dδa = (i∆0 dt Γt /2)δa igom xb a0 + i Γe δA. (3. the mechanical motion can be excited more intensely through the optical force by modulating the incident optical wave. (3. As a result. It is given by the following form: p a0 δA(Ω) a0 δA ( Ω) i Γe gom .19) can be solved easily to obtain the solution for the optically bright mode. Fe (Ω) = + ω0 i(∆ + Ω) Γt /2 i(∆ Ω) + Γt /2 (3. xb (Ω) = where η4 κ2 mb md Fb (Ω) mb Ld (Ω) κ Fd (Ω) mb md . 3.
in which the strong opticallyinduced rigidity associated with differential inplane motion of the nanobeams results in a dressing of the mechanical motion by the light ﬁeld.29) In the case that the mechanical excitation is dominated by the external optical modulation. ω0 (i∆ Γt /2) 0 (3. the mechanical response given in Eq. As the laser is tuned into resonance from the blueside of the cavity.2(a) and (b) the resonance frequency and resonance linewidth.2(c)). the higher frequency resonance is seen to .13) become dxb Fo κ d 2 xb + Γmb . (3. (3. and the measured mechanical resonances are split by 200 kHz. Optical excitation provides both a means to transduce mechanical motion (which is imparted on the transmitted light ﬁeld through phase and intensity modulation) and to apply an opticalintensitydependent mechanical rigidity via the strong optical gradient force. + Ω2 xb + xd = mb 2 dt dt mb mb d 2 xd dxd κ + Γmd + Ω2 xd + xb = 0. we display in Fig. 3.26) and (3. we ﬁnd that the mechanical displacement for the optically bright mode is now given by xb (Ω) = Fe (Ω) Ld (Ω) mb Lb (Ω)Ld (Ω) . 3. the thermal excitation from the reservoir is negligible and Eqs.30) (3.12) and (3. Eq. 3.4 Mechanical mode renormalization in zipper cavities We begin with an analysis of the zipper cavity.32) is directly analogous to the atomic response in EIT systems [83]. in the sidebandunresolved regime.61 In particular. Clearly. with similar linewidths (damping) and transduced amplitudes (Fig. of the two coupled mechanical modes of the nanobeam pair as a function of lasercavity detuning.28) can be well approximated by Fe (Ω) p i Γe gom a δA(Ω) + a0 δA ( Ω) .31) Using Eqs. md 2 dt dt md (3.27). (3. By ﬁtting a Lorentzian to the two lowestorder inplane mechanical resonances in the radiofrequency (RF) optical transmission spectrum. (3. and the intracavity photon number increases (to 7000). At large detuning (low intracavity photon number) the nanobeams’ motion is transduced without inducing signiﬁcant optical rigidity. respectively. following a similar procedure as the previous section.32) η4 where Lb (Ω) and Ld (Ω) are given in the previous section. (3.
25 0.2: (a) Mechanical frequency and (b) linewidth of the fundamental inplane mechanical resonances of the zipper cavity’s coupled nanobeams as a function of laser frequency detuning.4 8. of nitrogen). the optical gradient force acts most strongly on the differential beam motion and negligibly on the commonmode motion. A qualitative understanding of the lightinduced tuning and damping of the zipper cavity nanobeam motion emerges if one considers the effects of squeezeﬁlm damping [84].5 Normalized Detuning (∆ο/Γt) 1 0.5 1 1.15 0. Squeezeﬁlm effects. (3.5 2 7.2 Frequency (MHz) 7.37).32. corresponding to a maximum cavity photon number of 7000 on resonance.6 7.8 7.2 0. Tuning from the redside of the cavity resonance reverses the sign of the frequency shifts and the roles of the high and low frequency modes. The input power for these measurements is 127 µW. The sign of the resulting optical spring is positive for blue detuning and negative for red detuning from the cavity resonance. Similarly. Opticallytransduced RF spectrum at a lasercavity detuning of (c) ∆0 /Γt = 2. Putting all of this . signiﬁcantly increase in frequency while the lower frequency mode tunes to the average of the independent beam frequencies with its transduced amplitude signiﬁcantly weaker.4 8. a result of trapped gas inbetween the beams (measurements were performed in 1 atm.5 0 0.05 0 Mechanical Linewidth (MHz) Power Spectral Density (dBm/Hz) b 60 d f Nphoton 70 80 2 1.4 Power Spectral Density (dBm/Hz) Mechanical Frequency (MHz) a 90 c e Nphoton 100 110 0.1 0. The two nanobeams vibrate independently when the lasercavity detuning is large. tend to strongly dampen differential motion of the beams and should be negligible for common motion of the beams.6 8 8.1 and (d) ∆0 /Γt = 0. The linewidth of the high frequency resonance also tends to increase.2 8 7.8 Figure 3. The circles show the experimental data and the solid curves correspond to a ﬁt to the data using Eq. while that of the lower frequency mode drops. but are renormalized to the cooperative (e) differential and (f) common motions near resonance.62 8.
35) where L j (Ω) = Ω2 j m iΓm j Ω ( j = 1. respectively.34) into the frequency domain.33) (3.63 together.16) in Eq. 3. and approaches the cavity halflinewidth. where fq (Ω) represents the spectral response of the squeeze gas ﬁlm [84]. we term these dressed motional states opticallybright and opticallydark.35). and m j . it can be described by Fq (Ω) = fq (Ω)xb (Ω). damping rate. Using this form together with Eq. The optically bright mechanical mode corresponds to the differential motion of the two disks/beams. x j . the motion of individual disks or nanobeams satisﬁes the following equations: Fq d 2 x1 F1 Fo dx1 + Γm1 + Ω2 x1 = + + . Ωm j . and thus has a magnitude linearly proportional to the differential displacement. Γm j . it is easy to ﬁnd that the mechanical displacement of the optically bright mode is given by xb (Ω) = F1 (Ω) m1 L1 (Ω) Ω2 F2 (Ω) 1 1 + + m2 L2 (Ω) m1 L1 (Ω) m2 L2 (Ω) Fq (Ω) + Fo (Ω) . (3. respectively. By transferring Eqs. m2 2 dt dt m2 m2 m2 (3. (3.2 in which the nanobeams start out at large detuning moving independently with similar damping (the frequency splitting of 200 kHz is attributable to fabrication assymetries in the beams). 2). Due to the strong lightﬁeld coupling of the differential mode and the correspondingly weak coupling of the common mode. (3. we obtain the spectral intensity of . As the detuning is reduced. (3. 2) are the effective mass. The squeezeﬁlm effect is produced by the pressure differential between the gap and the outer region introduced by the differential mechanical motion. with a mechanical displacement given by xb x1 x2 . the mechanical displacement. In general.33) and (3. Fj ( j = 1. the motion of the nanobeams is dressed by the internal cavity ﬁeld into differential motion with a large additional optical spring constant (either positive or negative) and large squeezeﬁlm damping component. and common motion with reduced squeezeﬁlm damping and minimal coupling to the light ﬁeld. resonance frequency. and the Langevin force for individual disks (or beams). a consistent picture emerges from the data in Fig. m1 dt 2 dt m1 m1 m1 Fq dx2 d 2 x2 F2 Fo + Γm2 + Ω2 x2 = . In general.34) where Fq is the viscous force from the squeeze ﬁlm damping.
(3. (3. this theoretical model provides an accurate description of the mechanical mode renormalization. 3. they generally have quite close effective masses and energy damping rates: m1 m2 = 2mb and Γm1 Γm2 Γm .2.37) where Γq αq /mb represents the damping rate introduced by the squeeze gas ﬁlm.36) As the squeezeﬁlm effect primarily damps the differential motion. Similarly. The intrinsic and loaded optical Q factors are 3. As shown in Fig.16).75 pg for the fundamental differential mode.16). respectively. we can obtain the spectral intensity of xd x1 + x2 for the opticallydark mechanical . Moreover. since the two disks or nanobeams generally have only slight asymmetry due to fabrication imperfections.995 MHz for the two individual nanobeams are measured from the experimental recorded PSD with a large lasercavity detuning. both obtained from FEM simulations (note that these values are different than those quoted in Ref.64 the optically bright mode displacement.8 104 .790 and 7. obtained from optical charac terization of the cavity resonance. (3. (3. where we have used the fact that the effective motional mass of the differential motion is given by mb = m1 m2 /(m1 + m2 ). 2kB T Sxb (Ω) = L1 (Ω)L2 (Ω) Γm1 m1 jL2 (Ω)j2 + Γm2 jL1 (Ω)j2 m2 L1 (Ω) m2 [ fo (Ω) + fq (Ω)] + L2 (Ω) m1 2 . where we treat the intrinsic mechanical damping rate Γm and the squeezeﬁlminduced damping rate Γq as ﬁtting parameters. we can easily ﬁnd the mechanical frequencies and linewidths for the two renormalized modes. and the spectral response of the gradient force fo (Ω) is given by Eq.03 and 0. As a result.2 MHz. (3. Eq. By using these values in Eqs. with a ﬁtted intrinsic mechanical and squeezeﬁlm damping rate of 0. respectively. The optomechanical coupling coefﬁcient is 68 GHz/nm and the effective mass is 10.37) and (3.36) can be well approximated by Sxb (Ω) kB T Γm mb L1 (Ω)L2 (Ω) jL1 (Ω)j2 + jL2 (Ω)j2 1 2 [ fo (Ω)/mb + iΓq Ω] [L1 (Ω) + L2 (Ω)] 2 . its spectral response can be approximated as fq (Ω) iαq Ω. 45 due to the different deﬁnition of mode amplitude for xb ).0 104 and 2. The intrinsic mechanical frequencies of 7.
In this case. (3. which is given by the following form: Γm2 m2 L1 (Ω) Sxd (Ω) = 2kB T 2 m1 [ fo (Ω) + fq (Ω)] + Γm1 L2 (Ω) m1 [ fo (Ω) + fq (Ω)] L1 (Ω) m2 2 2 m2 2 [ fo (Ω) + fq (Ω)] 2 . 3.38) L1 (Ω)L2 (Ω) (Ω) + L2m1 Similar to the opticallybright mode. The result is that the differential. In particular.1(ac).40) indicates that the optically bright and dark modes reduce to pure differential and common modes. respectively.40) where L0 (Ω) = (Ωm1 + Ωm2 )2 /4 iΓm Ω. Sxd (Ω) 2kB T Γm /md jLo (Ω)j2 (3. whereas the common motion of the disks results in a higher frequency (14. FEM modeling of the mechanics of the doubledisk structure indicates a signiﬁcant frequency splitting between the differential and common modes of motion of the double disk (shown in Fig. Eq. primarily due to the difference in the extent of the undercut between the disk layers and the extent of the central pedestal which pins the two disk layers. md L1 (Ω)L2 (Ω) 1 h(Ω) [L1 (Ω) + L2 (Ω)] 2 2 (3. (3.2 MHz) “breathing” motion of the entire doubledisk structure. more intriguing form of coherent optomechanical mixing with the optically dark common mode of the disks. Unlike in the zipper cavity. the spectral intensities of these two modes reduce to Sxb (Ω) 2kB T Γm /mb jLo (Ω) Ω2 h(Ω)j 2 .39) where md = m/2 is the effective mass of the common mode and h(Ω) [ fo (Ω)/mb + iΓq Ω] repre sents the total spectral response of the optical gradient force and squeeze ﬁlm damping. Equation (3. .5 Coherent mechanical mode mixing in doubledisks A similar opticallyinduced renormalization mechanism applies to the doubledisk cavity structure shown in Fig. when the opticalspringinduced frequency shift is much larger than the intrinsic mechanical frequency splitting.3(b) and (c)).38) can be kB T Γm jL1 (Ω) h(Ω)j2 + jL2 (Ω) h(Ω)j2 . 3.95 MHz. or “ﬂapping” motion.65 mode. with m1 well approximated by Sxd (Ω) m2 = 2mb and Γm1 Γm2 Γm . the large optical spring effect for the differential motion of the two microdisks excites another. of the undercut disk region has a lower frequency of 7. 3.
3. measured using the same ﬁber probing technique as for the zipper cavity.1.6 MHz. 3. our theory provides an excellent description of the observed phenomena. 3.2(c) versus lasercavity detuning. fh). by using an optomechanical coupling coefﬁcient of gom /2π = 33 GHz/nm and an effective mass of mb = 264 pg for the ﬂapping mode.3(d. As shown schematically in Fig. and are also given in the caption of Fig.66 The RFspectrum of the transmitted optical intensity through a doubledisk cavity. 3. its spectral amplitude is considerably enhanced as the opticallybright ﬂapping mode is tuned into resonance. (3.24). Although the opticallydark breathing mode is barely visible in the tranduced spectrum at large lasercavity detunings. As shown clearly in Fig. For the largest detuning (in which the optical spring is negligible) the spectrum shows a broad (2.9) is used to ﬁnd the theoretical PSD shown in Fig. Fitting of the PSDs results in η = 3. a strong Fanolike lineshape. the ﬂapping mode is tuned across the breathing mode at 13.07 106 and 0.1 MHz) resonance at 8. This is quite similar to the phononphonon interaction during the structural phase transition in solids [66. As shown in Fig.7 MHz (optical input power of Pi = 315 µW). in good corresponce with the expected frequencies of the ﬂapping and breathing modes.32 MHz.3. Equation (3. 3. the Fanolike interference in the opticallybright power spectral density can be attributed to an internal mechanical coupling between the ﬂapping and breathing mechanical modes.3 MHz and a much narrower (0. and Γmd ) are obtained from the experimentally recorded PSD of cavity transmission with a large lasercavity detuning. in which the internal coupling between phonon modes produces Fanolike resonances in the Ramanscattering spectra. The difference in damping between the two resonances can be attributed to the strong squeezeﬁlm damping of the differential ﬂapping motion of the disks.3 MHz all the way out to 15. The intrinsic mechanical frequencies and damping rates of the two modes (Ωmb . 3. appears in the power spectrum near resonance of the two modes (Fig. In the process. . The power spectral density (PSD) of the cavity transmission is linearly proportional to Eq.3.24) together with (3. with 13 dB antiresonance. respectively.11 MHz) resonance at 13. Γmb . indicating a strong internal coupling between the two mechanical modes. 3. the ﬂapping mode can be tuned in frequency via the optical spring effect from its bare value of 8. is shown in Fig. In addition.2(c).5(a). both obtained from FEM simulations. as given in the caption of Fig. Ωmd .3(fh)).6 MHz. 3.7 106 are obtained from optical characterization of the cavity resonance. 85–89]. The intrinsic and loaded optical quality factors of 1. The mechanical coupling coefﬁcient η is treated as a ﬁtting parameter.
95 and 14. ∆0 /Γt indicated in (e). (c) Recorded power spectral density (PSD) of the cavity transmission for the doubledisk. with simulated frequencies of 7. Each curve corresponds to a normalized lasercavity frequency detuning. each curve is relatively shifted by 10 dB in the vertical axis. . For display purposes. respectively.2 MHz. with the experimental and theoretical spectra in blue and red.67 a b Power Spectral Density (20 dB/div) c d e h g f 6 75 8 10 12 14 16 Frequency (MHz) 18 6 8 10 12 14 16 Frequency (MHz) 18 f g 01 2 3 ∆0/Γt h PSD (dBm/Hz) 80 85 90 95 100 8 10 Frequency (MHz) 12 14 16 18 8 10 Frequency (MHz) 12 14 16 18 8 10 Frequency (MHz) 12 14 16 18 Figure 3. with an input power of 315 µW. (d) The corresponding theoretical PSD. The color map indicates the relative magnitude (exaggerated) of the mechanical displacement. (fh) Detailed PSD at three frequency detunings indicated by the arrows in (e).b) FEM simulated mechanical motion of the differential ﬂapping mode (a) and the common breathing mode (b).3: (a.
6 Coherent mechanical mode mixing in zipper cavities The coherent mixing of mechanical excitation is universal to gradientforcebased NOMS with a giant optical spring effect.44) where the gradient force Fo is given by Eq. 2) and Lb (Ω) is given by Eq.3. However. the Hamiltonian for the mechanical system is given by the following general form: Hm = ∑ i=b.45) where η4 j κ2 j mb m j ( j = 1. (3. we obtain the equations of motions for the three modes: d 2 xb κ1 κ2 Fb Fo dxb + Γmb + Ω2 xb + x1 + x2 = + . (3. mb dt 2 dt mb mb mb mb dx1 d 2 x1 κ1 F1 + Γm1 + Ω2 x2 + xb = . we can obtain the spectral intensity for the mechanical displacement of the optically bright mode as Sxb (Ω) = 2kB T η4 Γm1 jL2 (Ω)j2 + η4 Γm2 jL1 (Ω)j2 + Γmb jL1 (Ω)L2 (Ω)j2 1 2 .68 3. (3. m2 dt 2 dt m2 m2 (3. the coupled nanobeams have more complex mechanical mode families in which all the evenorder mechanical modes are optically dark. Similar phenomena to that presented for doubledisks were also observed in the zipper cavity.3. With this Hamiltonian.42) (3. m1 2 dt dt m1 m1 dx2 d 2 x2 κ2 F2 + Γm2 + Ω2 x3 + xb = . (3. L j (Ω) = Ω2 j m Ω2 iΓm j Ω ( j = 1. due to the device geometry.21) and (3. they can simultaneously couple to the same optically bright mode. As the sameorder common and differential motions of the two beams have similar mechanical frequencies.22).43) (3.14).20) with Ω0 and Γ0 given in Eqs.41) where i = b.15). mb mb respectively.2 p2 1 i + ki x2 + κ1 xb x1 + κ2 xb x2 . counting in both the optical gradient force and the Langevin forces from the thermal reservoir. 2 mb Lb (Ω)L1 (Ω)L2 (Ω) η4 L2 (Ω) η4 L1 (Ω) 1 2 (3. and the statistical properties of the Langevin forces are given by Eq. 1. the power spectral . leading to multiple excitation interferences on the mechanical response. Following the same analysis as in Section 3. respectively. 2) represents the mechanical coupling coefﬁcient.1. In the case when the optically bright mode is coupled to two optically dark modes. 2 corresponds to the optically bright mode and optically dark modes 1 and 2. 2mi 2 i (3. because they exhibit a mechanical node at the beam center where the optical mode is located. As the optical wave is coupled to the optically bright mode only.
Fitting of the PSD results in mechanical coupling coefﬁcients of η1 = 3. (3. into the longlived breathing mode. both broadband and opticallybright.06 MHz to 19 MHz. An alternative.7 Analogy to electromagneticallyinduced transparency The mechanical response given by Eq. respectively. or indirectly. one can understand the resulting Fano lineshape in two different ways. Similar to the doubledisk NOMS. 3.4(b). as shown clearly in Fig. (d)(f). the two dressed modes are excited with equal amplitude and opposite phase at the center frequency between the split dressed states.0 104 and 2. Consequently. 3. either directly into the broadband (lossy) ﬂapping mode. Equation (3. The ﬁrst perspective considers the interference associated with multiple excitation pathways.4 shows the PSD of the cavity transmission by launching a continuous wave into a resonance of the coupled nanobeams with an intrinsic and loaded Q factor of 3.4(h)I]. through the ﬂapping mode. respectively. and mode II and III correspond to the secondorder common and differential modes [Fig. resulting in the Fanolike resonance in the spectral response of the optically bright ﬂapping mode.45). and then back again into the ﬂapping mode. In the optomechanical system.04 MHz and resulting in complex interferences on the power spectra [Fig.8 104 . 3. suppressing excitation of the mechanical system at the line center.45) provides an accurate description of the observed phenomena. (3. In this picture the internal mechanical coupling renormalizes the broadband ﬂapping mode and the narrowband breathing mode into two dressed mechanical modes.69 density of the cavity transmission is still given by Eq.4(a)]. 3. Figure 3. the gigantic optical spring effect shifts the frequency of the optically bright mode I from its intrinsic value of 8. where mode I is the fundamental differential mode [Fig. implying that the two optically dark modes couple to the fundamental optically bright mode with a similar magnitude. In particular. the mechanical motion of the ﬂapping mode is thermally excited along two different pathways. Three mechanical modes are clearly visible.45 MHz and η2 = 3. Destructive interference results.4(h)II and III]. The two excitation pathways interfere with each other. Just as in EIT. (3. with the mechanical response Sxb given in Eq. the mechanical motion becomes purely a trapped .9).24) is directly analogous to the atomic response in EIT [83]. when the ﬂapping and breathing mechanical frequencies coincide. crossing over both optically dark modes II and III closely located at 16. but perfectly equivalent view of the coupled optomechanical system considers the dressed states resulting from the internal mechanical coupling.48 MHz.54 and 17. 3.
The optically dark mode II and III have a fullwidth at half maximum (FWHM) of 0. .4: (a) Experimentally recorded power spectral densities of the cavity transmission for the zipper cavity of Fig. (d)(f) The detailed spectra of the power spectral density at three frequency detunings indicated by the three arrows in (c).15 MHz. (h) FEM simulated mechanical motions for the fundamental differential mode (I). respectively.30 MHz. Each curve corresponds to a laser frequency detuning indicated in (c). the secondorder common (II) and differential (III) modes.1 mW. whose frequencies are indicated by the arrows in (a).70 ΙΙ (a) Power Spectral Density (20 dB/div) ΙΙΙ Ι (b) (c) f e d 10 80 12 14 16 18 Frequency (MHz) 20 10 12 14 16 18 Frequency (MHz) 20 0 1 ∆0/Γt PSD (dBm/Hz) 90 100 110 (d) 14 (e) Frequency (MHz) 16 18 20 22 14 (f ) Frequency (MHz) 16 18 20 22 14 Frequency (MHz) 16 18 20 22 (h) Ι ΙΙ ΙΙΙ Figure 3. Each curve is relatively shifted by 5 dB in the vertical axis for a better vision of the mechanical frequency tuning and the induced mechanical interference. with an input power of 5. respectively. 3. (b) The corresponding theoretical spectra of the power spectral density. The color map indicates the relative magnitude (exaggerated) of the mechanical displacement. The blue and red curves show the experimental and theoretical spectra. The optically bright mode I has an intrinsic FWHM of 0.1(df).16 and 0.
equivalent to the breathing mode) actuated by the spring k2 which is decoupled from the optical wave. . (b) A photonic analogue to the optomechanical system involving coupled resonators. The mechanical motion of the cavity mirror (m1 . (d) The state diagram corresponding to the optomechanical system of (a) where j1i is the phonon vacuum state. The excited state (j2i) is split by the optical control beam into two broadband dressed states (j+i and j−i). It is internally coupled to a second massspring system (m2 . Microcavity 1 is directly coupled to the external optical waveguide (equivalent to the opticallybright ﬂapping mode) and also internally coupled the narrowband cavity 2 (equivalent to the opticallydark breathing mode).5: (a) Schematic of an equivalent FabryPerot cavity system showing mechanical mode mixing. respectively. (c) State diagram of an EITlike medium. The two masses are internally coupled via spring k12 . The dipole transition between groundstates j1i and j3i is forbidden. equivalent to the opticallybright ﬂapping mode) is primarily actuated by the spring k1 and the optical force.71 a input xb m1 k1 b input output 1 output coupling k12 m2 xd k2 2 c +〉 2〉 −〉 probe γ3 γ2 d pump antiStokes Stokes 2〉 coupling 3〉 control 3〉 pump Γ2 1〉 1〉 RF/micr RF/mic RF/microwave pho photon 1〉 Γ3 Figure 3. and j2i and j3i correspond to the ﬂapping and breathing modes.
(3. there are some important.72 mechanicallydark state. The system Hamiltonian of an optomechanical cavity is given by the following general form: H = ω0 a† a + Ωm b† b + gom xb a† a. 91]. The mechanical motion modulates the intracavity ﬁeld to create two optical sidebands. For instance. where the factor g g2 3 om 2mb Ωmb 1/2 (3. The dynamic backaction between the cavity ﬁeld and mechanical motion creates Stokes and antiStokes optical sidebands. As shown in Fig. 90. (3. whose beating with the fundamental optical wave resonates with the mechanical motion to create/annihilate phonons (see Fig. subtle differences. albeit with unbalanced scattering amplitudes resulting from the coloring of the electromagnetic density of states by the optical cavity. transparent to external excitation. respectively. xb is the mechanical displacement for the optically bright mode.3(d)). in which the quantum interference between the transition pathways to the dressed states of the excited electronic state.46) where a and b are the annihilation operators for photon and phonon. rather than the linear dipole transition of EIT. leads to an induced spectral window of optical transparency. this induced mechanical transparency is a direct analogue to EIT in atomic systems [63. 83. As a result. normalized such that a† a and b† b represent the operators for photon and phonon number. related to b by xb = 2mb Ωmb b + b† . in the optomechanical system. the optical ﬁeld can be written as a = a p + as e iΩmb t + ai eiΩmbt . the interaction Hamiltonian between the optical wave and the mechanical motion is given by Hi = hga† a b + b† . the interaction corresponds to a secondorder transition. this is like coherent Stokes and antiStokes Raman scattering. 3. 3.49) . Functionally. through either j1i $ j+i or j1i $ j i.47) Therefore. Despite the intriguing similarities between the optomechanical system studied here and EIT in atomic media.48) . (3.3(c).
8 Discussion Although the studies considered here involve thermal excitation of the optomechanical system. (3.48) and leave only the ﬁrstorder terms of as and ai . 92. This interference again leads to a Fanolike resonance. 3. using external optical means (Sec. which changes only the equilibrium position of mechanical motion and is neglected in the current analysis. As such.73 where a p is the ﬁeld amplitude of the fundamental wave. or what has been termed coupledresonatorinduced transparency. in the optical cavity transmission [70–73]. and with greater control. The second and third terms show clearly that the process corresponds directly to coherent Stokes and antiStokes Raman scattering as shown in Fig. and as and ai are those of the generated Stokes and antiStokes wave. 3. as discussed previously.3(c)). beyond the interesting physics of these devices. Similar to the information storage realized through EIT [83. rather than tuning the Rabisplitting through the intensity of a control beam resonant with the j3i $ j2i electronic transition (Fig. under the rotatingwave approximation.5(d). and the waveguidedecoupled highQ resonator 2.49) into Eq. this opticallyinduced mechanical transparency is controlled via optical spring tuning of the resonance frequency of the optically bright ﬂapping mechanical mode. p s p p i (3. 93]. 3.50). when we substitute Eq. it is the modulation signal carried by the incident optical wave (radiofrequency or microwave photons) that fundamentally probes/excites the mechanical motion and to which the trapped mechanicallydark state becomes transparent. (3. Moreover. the ﬁrst term describes the static mechanical actuation. Therefore. The interference in this case is between the two optical pathways composed of the waveguidecoupled lowQ optical resonator 1. optical information can be stored and buffered in the dark mechanical degree of freedom in the demonstrated NOMS. (3. As the magnitudes of the Stokes and antiStokes sidebands are much smaller than the fundamental wave. the same phenomena can be excited more efﬁciently. Perhaps the most apt analogy to the optomechanical system can be made to the photonic resonator system shown in Fig. respectively. the interaction Hamiltonian becomes Hi = g b + b† a† a p + gb† a† a p + a† ai + gb a† as + a† a p . 3.3(d). This can be realized through a procedure .3.50) In Eq.4). exciting application in RF/microwave photonics and quantum optomechanics exist. 3. in analogy to EIT.
. adiabatic tuning of optical resonances are used to slow. enabling efﬁcient information storage and retrieval at the singlequantalevel suitable for quantum state transfer. Moreover. In this scheme. In the quantum realm. and retrieve optical pulses. would reduce the simultaneous creation and annihilation of Stokes and antiStokes photons. a timescale more than seven orders of magnitude longer than that in demonstrated photoniccoupledresonator systems [97] and comparable with EIT media [92. In comparison to the allphotonic system. such a system operating in the goodcavity or sidebandresolved regime (by increasing either the optical Q factor [59] or the mechanical frequency).74 similar to that recently proposed for coupled optical resonators [94. The corresponding optomechanical system would consist of an array of doubledisk resonators. allowing for the RF/microwave signal to be coherently stored in (released from) the longlived breathing mode through adiabatic compression (expansion) of the mechanical bandwidth [94. all coupled to a common optical bus waveguide into which an optical signal carrying RF/microwave information would be launched. 95] in which dynamic. 93]. a second control optical beam would adiabatically tune the frequency of the opticallybright ﬂapping mode of each resonator. primarily related to the attainable lifetime of the dark mechanical state. store. 95]. the radial breathing mechanical mode of a similar whisperinggallery cavity has been shown to exhibit a lifetime of more than 2 ms [96]. For example. mechanical lifetimes of more than one second have recently been demonstrated using stressed silicon nitride nanobeam [98] and nanomembrane [40] mechanical resonators operating in the MHz frequency regime. optomechanical systems have several advantages.
118. buffering.75 Chapter 4 Mechanically Pliant Double Disk Resonators 4. photochemical [111]. 115. 114.1 Introduction Optical information processing in photonic interconnects relies critically on the capability for wavelength management [99. In general. all of these tuning mechanisms have intrinsic limitations on their tuning speed [108–111. those based on micro/nanoresonators are particularly attractive because of their great potential for future onchip integrated photonic applications [107–119]. 119]. dispersion compensation. 118. 119]. By using a tuning mechanism based upon the optical gradient forces in a speciallydesigned nanooptomechanical system. 110. 109. 117]. 119]. 119]. 114. 117. 101–103]. The underlying essential functionalities are optical ﬁltering and wavelength routing. and/or routing quality [112. a switching time of less than 200 ns. pulse trapping/release. routing efﬁciency [108. and tunable lasing. tuning bandwidth [112. and 100% channelquality preservation over the entire tuning range. Here we propose and demonstrate an alloptical wavelengthrouting approach which combines the advantages of various approaches into one nanophotonic device. 119]. However. a variety of technologies have been developed for this purpose [104–106]. with a tuning efﬁciency of 309 GHz/mW. The demonstrated approach and device geometry indicates great prospects for a variety of applications such as channel routing/switching. 118. 118. or microelectricalmechanical approaches [108. we are able to realize seamless wavelength routing over a range about 3000 times the channel intrinsic linewidth. 120]. which allow for precise selection and ﬂexible switching of optical channels at high speeds over a broad bandwidth [99. electrooptic [112. with easy onchip . optoﬂuidic [115]. 100]. reconﬁgurable tuning of cavity resonances is realized through thermooptic [109. In the past two decades.
2 Spiderweb resonator design and optical characterization The optomechanical system we consider here is a simple modiﬁcation to the common microring whisperinggallery cavity that has found widespread application in microphotonics. of ﬁxed frequency. The spider . one stacked on top of the other [122. with more recent studies having measured radiation pressure forces in micro. with some of the early experimental considerations being related to the quantumlimited measurement of weak. In each of these systems. Fabrication of the spiderweb whisperinggallery resonator began with initial deposition of the cavity layers. 4. 123]. x. 40. classical forces [29]. the same fundamental physics applies. dynamic mechanical susceptibility. the larger the static displacement and the larger the tuning of the optical cavity. A narrowband laser input to the system. pushing on the mechanical system as the internal light ﬁeld builds up near cavity resonance. with a thickness of 400 4 nm and 150 3 nm for the silica and αSi layers. The more compliant the mechanical system. respectively. 51]. The resulting nearﬁeld modal coupling forms a “supercavity. As shown in Fig. 44. improving the optical quality of the material.1(d). yielding an opticallycontrollable. 48. The wafer was then thermally annealed in a nitrogen environment at a temperature of T = 1050 K for 10 hours to drive out water and hydrogen in the ﬁlm. A separate effect occurs when the laser frequency is swept across the cavity resonance.76 integration on a siliconcompatible platform. This work was initially presented in Ref. experiments involving optical FabryPerot “pendulum cavities” were ﬁrst explored [121]. Here we utilize both the static and dynamic mechanical susceptibilities of a coupled optomechanical system to realize a chipbased optical ﬁlter technology in which wideband tuning and fast switching can be simultaneously accomplished.”. 45. In the optical domain. 78.and nanomechanical structures [35– 38. whether it be gravitational wave observatory [31] or photonic crystal nanomechanical cavity [45]. 4. results in a “dynamical backaction” [33] between mechanical ﬂuctuations and the internal electromagnetic ﬁeld. it consists of a pair of planar microrings. This dynamical backaction modiﬁes both the real and imaginary parts of the frequency of the mechanical motion. with a resonance frequency ω0 strongly dependent on the vertical cavity spacing. The two silica web layers and the sandwiched amorphous silicon (αSi) layer were deposited on a (100) silicon substrate by plasmaenhanced chemical vapor deposition. The physics of electromagnetic forces within mechanicallycompliant resonant cavities is by now well established.
(g) The theoretical wavelength tunability. The outward bending motion is shown for ease of viewing. (e) Mechanical FEM simulation of the bending of the 90 µm spiderweb resonator. perphoton force. Perphoton force (fN) gOM/2π (GHz/nm) . (d) Schematic of a crosssection of the resonator.77 1 µm a 20 µm b c 150 nm gap 25 µm Wavelength (nm) d SiO2 e αSi SiO2 2 50 40 30 20 10 0 1580 1540 1500 1460 0 200 400 600 800 30 25 20 x Height (µm) 1 0 f Ring spacing (nm) 15 10 Si 1 2 40 42 44 46 48 0 g 200 5 400 600 800 0 1000 Radius (µm) Ring spacing (nm) Figure 4.1: Scanning electron microscope images of (a) the betweenring gap. The vertical dashed lines represent the experimentallyrealized ring spacing of 150 nm. (b) the 54 µm spiderweb resonator. (f) FEM simulation of the radial component of the electric ﬁeld for the fundamental TE bonding mode of the 90 µm spiderweb structure. and is exaggerated for clarity. and (c) the 90 µm spiderweb resonator. showing the bending of the two silica rings under the inﬂuence of the optical force. and wavelength (inset) of the spiderweb cavity as a function of the ring spacing.
Optical spectroscopy of the devices is . respectively. The overall magnitude of the cavity resonance tuning is then. Simultaneously. fully releasing the web. inversely proportional to the optical quality factor. resulting in a uniform undercut region which extends radially inwards 4 µm on all boundaries. In addition to the favorable mechanical properties. k kω0 Γ0 (4. determines both the tunability and The optomechanical coupling coefﬁcient. the underlying silicon support pedestal is formed.1(c)) has a 90 µm outer diameter ring with six spokes and one supporting inner ring. The corresponding static mechanical displacement for N photons stored inside the cavity is ∆xstatic = N gOM /k. dω0 /dx.63 N/m for the smaller and larger resonators. for two 400nmthick planar silica whisperinggallery microcavities placed 150 nm apart (Fig.78 web pattern was created using electron beam lithography followed by an optimized C4 F8 SF6 gas chemistry reactive ion etch. This feature enables independent control of the optical and mechanical properties. the geometry was optimized such that the forks introduce a total insertion loss of only 4%. 4. where k is the intrinsic spring constant of the mechanical structure. Finiteelementmethod (FEM) simulation shows that. Release of the web structure was accomplished using a SF6 chemical plasma etch which selectively (30. 4. The zerothorder spiderweb cavity (Fig. In order to minimize the mechanical stiffness while also providing mechanical stability. ∆ω0 = gOM ∆xstatic = N g2 g2 Pd OM = OM .1(fg)). allowing us to freely engineer the intrinsic mechanical rigidity through the scalability of the structure without changing the perphoton force. it is completely independent of the roundtrip length of the cavity.1) where Pd is the power dropped into the cavity and Γ0 is the intrinsic photon decay rate. FEM simulations show that these structures have spring constants of 9. Two nanoforks were also fabricated near the doubledisk resonator to mechanically stabilize and support the ﬁber taper during optical coupling. gOM perphoton optical force [43. As the optical gradient force stems from the evanescent ﬁeld coupling between the two nearﬁeldspaced cavities. 000 : 1) attacks the intermediate αSi layer and the underlying Si substrate. we utilize a spiderweblike support structure consisting of an arrangement of spokes and inner rings [96]. while the ﬁrstorder structure (Fig. 4. the whisperinggallery nature of the spiderweb resonator provides for highQ optical resonances. the resonance tunability is as large as gOM /2π = 31 GHz/nm (corresponding to a 21 fN/photon force). 76].25 N/m and 1.1(b)) has a 54 µm diameter outer ring supported by ﬁve spokes.
the probe wavelength is calibrated by a MachZehnder interferometer (MZI).2: The pump and probe lasers are coupled to the spiderweb resonator via a singlemode silica ﬁber taper stabilized by two nanoforks fabricated near the device. The spiderweb device itself is contained within a nitrogen environment at atmospheric pressure. The pump laser power is boosted by an erbiumdoped ﬁber ampliﬁer (EDFA) and passed through a bandpass ﬁlter (BPF). The two lasers are split into separate wavelength channels using a mux/demux system (providing greater than 120 dB pumpprobe isolation). For modulation experiments. The laser power levels are controlled by several variable optical attenuators (VOAs). .79 EDFA Pump Laser BPF VOA MUX Probe Laser 50 : 50 splitter Polarization Controller EOM VOA Polarization Controller Polarization Controller RF Drive DC Bias Reference Detector 1 MZ I Oscilloscope Network Analyzer Reference Detector 2 fiber taper VOA 10 : 90 splitter 10 : 90 splitter Highspeed Detector Reference Detector 3 OSA DEMUX Figure 4. and the pump wavelength is monitored by an optical spectrum analyzer (OSA). the pump laser wavelength is modulated using an electrooptic modulator (EOM) driven by a network analyzer.
4. a complete theory developed previously in Sec. by launching a relatively intense wave at a different resonance to suppress the perturbations induced by the thermal mechanical motion. indicate that further mechanical design optimization may be necessary for the larger structures. the cavity resonance initially at λ = 1549 nm is shifted by 4. the smaller 54µm diameter structures had near100% yield and maintained their properties over the entire period of testing. 4. This is demonstrated in Fig. As discussed previously.3(b). indicating a great potential for broad waveband translation and switching in the wavelengthdivision multiplexing conﬁguration. The excited family of resonances.04 106 and Qi = 0. 4. the displacement actuated by one cavity mode can be used to control the wavelength routing of an entire mode family. the thermal Brownian mechanical motion can be signiﬁcantly suppressed through the optical spring effect. corresponding to a static mechanical displacement of ∆xstatic = 17. where the mechanical displacement actuated by the “pump” mode at λ = 1549 nm is used to control the wavelength of a “probe” mode initially located at λ = 1529 nm. Figure 4. By comparison.3(a) shows the low power. With a power of 1.7 mW dropped into the cavity. 4.3(d). 4. with resonances at λ = 1529 nm and λ = 1549 nm exhibiting intrinsic quality factors of Qi = 1. has a freespectral range (FSR) of 9. respectively.7 nm. 2. although device yield (20%) and a slow change in device properties over time (despite devices being tested in a nitrogen environment to avoid water adsorption).7 nm. This makes it difﬁcult to measure the optical Q factor of a cavity resonance.4 nm (a little more than 0. large gOM . Moreover. As the mechanical displacement is universally experienced by all doublering cavity modes. inplane polarized. wavelength scan of a 54µm diameter resonator. and ﬂoppy spiderweb structure result in the large optomechanical bistability shown in Fig. This feature provides an elegant way to accurately characterize the optical Q of a cavity mode. The extremely small intrinsic spring constant of the spiderweb resonator leads to signiﬁcant thermal Brownian mechanical motion and introduces considerable ﬂuctuations on the cavity transmission spectrum. With increased dropped pump .5 THz). 48 was used to describe the linear cavity transmission with the inclusion of the optomechanical effect. corresponding to the fundamental transverseelectriclike (TElike) modes. We observed similar performance from the larger 90 µm spiderweb structures. as shown in Fig.3 Static lter response The combination of high cavity Qfactor.80 performed using the experimental setup shown in Fig.2.4 and Ref.5(a).90 106 .
8 Probe tuning (GHz) W [ lp .6 30 0. . with Lorentzian ﬁt to the lineshape. (c) Probemode transmission curves for a selection of dropped powers in d.5 0 0.1531] (nm) /m 9G Hz 300 200 100 0 0 0. (b) Overcoupled pumpmode transmission spectrum at Pd = 275 nW (blue) and Pd = 1.5 0 0.81 1 1 Normalized transmission 0.8 0.5 1 Wavelength (nm) 1530 1540 500 400 1550 d 0.5 0.2 0 1 0. highlighed in red (green). with Pd indicated by the baseline of each transmission curve.1 0.3 0. (d) Measured (blue circles) and linear ﬁt (red curve) to the probe resonance wavelength tuning versus Pd .2 0. (e) Intensity image of the optical transmission spectrum near the anticrossing of two TE.5 0.3: (a) Broadband optical transmission spectrum of the 54 µm spiderweb cavity.5 Normalized transmission 0.4 0.8 high probe 1 0. Green curve corresponds to thermooptic component of tuning.6 low power a 1500 Detuning (pm) Detuning (pm) 8 1560 0.7 mW (red).5 0. Inset: ﬁne frequency scan of pump (probe) mode.2 1529 Probe wavelength (nm) 1530 1531 1532 1533 0 Pump dropped power (mW) Figure 4.6 0.8 8 1510 0 8 0.5 1 1.and TMlike probe modes.4 1549 1550 1551 b e Wavelength (nm) 1552 1553 1 1520 Pump dropped power (mW) c 1.4 0.6 8 0 pow er pump 0.4 0 14 GHz/mW 0.
5(d) and (e). approximately 3000 times the probe resonance intrinsic channel linewidth (or 500 times the loaded linewidth).2 nm. In principle.06%). 117–119].3(c).3(d) is about 43% of the FSR.3(d) gives a tuning efﬁciency of 309 GHz/mW. it is possible to tune over the entire freespectral range with a moderate dropped pump power of only 4 mW. 4. and properties of the dynamical response of the system (see below).06% of the experimentally recorded wavelength tuning.11 nm/mW (13. FEM simulations show that such a temperature variation of the resonator introduces a ringgap change by only about 10 pm. A linear ﬁt to the probe resonance tuning data data in Fig. 118. This. This factor is at least one order of magnitude larger than any other conventional approach previously reported [108–115.8 GHz/mW. 4. 4. 4. This value agrees reasonably well with the theoretically predicted value of 393 GHz/mW. corresponding to a tuning rate of 0. in which the accompanying carrier absorption degrades the quality of the switched channel and thus limits the ultimate tuning bandwidth.8 GHz/mW). This is in contrast to other tuning mechanisms such as the electrooptic approach via carrier injection [112. To isolate the thermooptic effect from the optomechanical effect. Therefore. The negligible contributions of both thermooptic and thermomechanical effects are conﬁrmed by the pumpprobe modulation spectra shown in Fig. 4. Importantly.23 nm. The thermooptic resonance tuning indicates a maximum temperature change of 21 K in the resonator. The difference between theoretical and experimental optical force tuning rates ( 25%) can . The thermooptical effect on the resonance tuning was calibrated by using another identical device on the same sample. thermally induced static mechanical deformation has only a negligible contribution of 0. Independent measurements show that the thermooptic effect contributes only a small component to the overall tuning rate (13. (4. the probe wavelength is tuned linearly and continuously by 4. so the ﬂapping mechanical motion was completely eliminated. show that the wavelength routing is indeed a result of the optical gradient force. and FEM simulations indicate a negligible thermomechanical component ( 0.3(d)).1)). inferred from optical and mechanical FEM simulations and the measured optical Qfactor (see eq. The tuning range shown in Fig. about 4% of the total tuning rate recorded experimentally.82 power.1 mW dropped into the cavity introduces a maximum resonance red tuning by 0. green curve in Fig. this wavelengthrouting approach is purely dispersive in nature and completely preserves the channel quality during the wavelength routing process as can be clearly seen in Fig. 119]. shown by the differential displacement of the top and bottom rings in Fig. Testing was performed on a cavity mode at 1552 nm using exactly the same conditions as for the wavelength routing measurements. A power of 2. we caused the two rings to stick together through the van der Waals force.4. 4.
allowing for modemixing) induced by the optical force tuning of the two mode families.4: FEM simulation illustrating the zdisplacement of a 54 µm spiderweb resonator under the 21 K temperature differential between substrate and ring induced by 2. and β is the optical coupling co efﬁcient between the two cavity modes. likely be attributed to the uncertaintity in the Young’s modulus of the annealed PECVD silica used to form the spiderweb structure. (4. the spiderweb doublering resonator also supports a family of highQ transversemagneticlike (TMlike) modes with a FSR of 10 nm. Figure 4. FEM simulations show that the perphoton force is slightly larger for the TM modes (26. Assume two cavity resonances located at ω01 and ω02 . For an input probe wave at ω. The anticrossing between the two probe modes when they approach each other is primarily due to the internal coupling between the two cavity modes.83 0 10 z 20 Figure 4.3(e) shows the mode hybridization between a pair of TE and TMlike modes (the slight angle in the outer sidewall of the two rings breaks the vertical symmetry.5 fN/photon. the steady state of Eqs. due primarily to the enhanced electric ﬁeld strength in the nanoscale gap between the rings for polarization normal to the plane of the rings.3) ω0j represents the cavity detuning of the jth mode.2) (4. (4. Γe2 Ain . or a 59% larger tuning efﬁciency).2) zdisplacement (pm) . In addition to the TElike modes. the two cavity modes are excited through the following equations: da1 = (i∆1 dt da2 = (i∆2 dt where ∆ j = ω Γt1 )a1 + iβa2 + i 2 Γt2 )a2 + iβa1 + i 2 Γe1 Ain . which can be described by a simple theory as follows. With a continuouswave input.1 mW dropped optical power.
5 N/m). 4. (4. 4. (i∆1 Γt1 /2)(i∆2 Γt2 /2) + β2 (4.5(a)).4 Dynamic lter response In addition to the static mechanical actuation of the spiderweb structure. thermal Brownian motion introduces signiﬁcant ﬂuctuations in the cavity resonances. The alteration of the effective dynamic spring is clearly seen in the resonance spectra of the cavity resonances (left panel of Fig.3) is given by the following solution a1 = a2 = p p iAin (i∆2 Γt2 /2) Γe1 iβ Γe2 .84 and (4. As the intrinsic spring constant of the spiderweb resonator is small (9. 50.5) p p As the transmitted ﬁeld from the cavity is given by AT = Ain + i Γe1 a1 + i Γe2 a2 . 4. the cavity transmission thus has the following equation (i∆1 jAT j2 = 2 jAin j Γ01 Γe1 )(i∆2 2 Γ02 Γe2 ) + (β 2 T (i∆1 Γt1 /2)(i∆2 p i Γe1 Γe2 )2 Γt2 /2) + β2 2 . the dynamic spring stiffens and strongly suppresses the magnitude of the thermal ﬂuctuations (right panel of Fig. however. 4. giving a tuning efﬁciency for the TM modes which is 42% larger than that of the TE modes. ω0 Γ0 [∆2 + (Γt /2)2 ] k0 = k + where ∆ = ωl (4. For the ﬂoppy spiderweb structure. 125]. the optical gradient force also introduces dynamical back action which alters the dynamic response of the mechanical motion [33.6) The experimental observation agrees well with this simple theory (dashed curve in Fig. (i∆1 Γt1 /2)(i∆2 Γt2 /2) + β2 p p iAin (i∆1 Γt1 /2) Γe2 iβ Γe1 . . The inphase component of the optical force leads to a modiﬁed mechanical resonance frequency and effective dynamical spring constant of 2g2 Pd ∆ OM .5(a)).4) (4. or carriersideband ﬁltering in microwave photonics [124].3(e)). This precisely tunable channel coupling may ﬁnd applications in polarization switching/multiplexing/demultiplexing in optical signal processing.7) ωo is the detuning of the input laser (ωl ) from the cavity resonance (ωo ) frequency and Γt is the photon decay rate of the loaded cavity. the dynamical spring can be greatly modiﬁed optically. As pump power is dropped into the cavity.
5: (a) Undercoupled probe transmission spectra recorded at low (Pd = 0 mW (blue)) and high (Pd = 0.12 25 20 15 10 5 0 5 0.3 0.6 0. The dashed black curves show the corresponding model.20 mW (red)) pump power (timeaveraged trace in black). (e) Probe modulation spectrum (Pd = 0. 14.6 1 0.32 Time(µs) 0.4 0.85 1 Normalized transmission Normalized transmission Fractional Modulation 1 0.8 0.8 0.3 MHz.6 0. vertical dashed line in (d)) pump mode with modulation depths of 1. while that for the probe is deﬁned relative to the onresonance probemode coupling depth.2 0 1529.85 mW) shown over a wide frequency span.4 Time(µs) 0. The orange curve shows the measured noise ﬂoor.2 b 0.08 1529.8 0.6 0.1 0.5 0. (c) Pulsed modulation of the pump (top) and corresponding probe response (bottom).8 0. 850g µW.2 0 0. (b) Time waveforms of the probe transmission for sinusoidally modulated (22. . The green curve shows the modeled response including only the dominant ﬂapping mechanical mode (the red curve includes other.8 0.6 1 0. (d) Normalized probe modulation spectra for Pd = f14. The fractional modulation for the pump is deﬁned relative to the average dropped power.9% (green).28 1529.2 0 0 0.8 0.4 0.2 pump c probe a Wavelength (nm) 1529.9% (blue).4 0. 210.4 0. and 20. 430.6 0.2 Modulation spectrum (dB) Modulation spectrum (dB) d 20 0 20 40 60 80 0 200 400 600 800 1000 1200 1400 e 78 dB 10 15 20 25 0 5 10 Frequency (MHz) 15 20 25 30 35 100 Frequency (MHz) Figure 4.85 mW. 110. mechanical resonances).8 1 1.5% (red) at average Pd = 0. breathinglike.6 0 0.
The intracavity ﬁeld can be written as a p = ap0 + δa p (t). In general. (4.9) provides a steadystate solution of ap0 = i Γep Ap0 . The ﬁnal term in Eq. Γtp /2 i∆ p (4. The fourth term describes the selfphase modulation introduced by the Kerr nonlinearity. the pump wave inside the cavity satisﬁes the following equation: da p = (i∆ p dt Γtp )a p 2 igOM xa p + iγja p j2 a p + i Γep Ap .8). ∆ p = ω p ω0p represents the detuning of pump frequency ω p to the cavity resonance ω0p and Γtp is the photon decay rate of the loaded cavity for the pump mode.9) Γep δAp . respectively. the selfphase modulation on the pump wave is negligible in the spiderweb ring resonator. the third term represents the back action of mechanical motion on the cavity resonance.11) from which we obtain the average pump power dropped into the cavity. n2Veff 0 n2 = 2. (4. compared with the dominant optomechanical effect. Equation (4. (4.12) . given by Ppd = Pp0 Γ0p Γep . 2 Γtp )δa p igOM xap0 + i 2 (4. (4. However. where gOM is the optomechanical coupling coefﬁcient and x(t) is the mechanical displacement of the cavity structure.6 10 20 m2 /W is the Kerr nonlinear coefﬁcient of silica.86 Optical control of the dynamic response is most clearly demonstrated through the pumpprobe modulation response of the spiderweb structure. where the nonlinear parameter γ = cω p n2 .8) represents the external ﬁeld coupling with a photon escape rate of Γep . and Veff = 370 µm2 (from FEM simulation) is the effective mode volume [126– 128]. governed by the following equations: dap0 = (i∆ p dt dδa p = (i∆ p dt Γtp )ap0 + i Γep Ap0 . n0 = 1. Assume that the input pump wave consists of an intense continuous wave together with a small timevarying modulation. normalized such that Up ja p j2 and Pp jA p j2 represent the intracavity energy and input power.8) where a p and A p are the intracavity and input ﬁeld of the pump wave. Ppd . In Eq.44 is the silica refractive index.10) where we have neglected the negligible selfphase modulation for the pump wave. 2 + (Γ /2)2 ∆p tp (4. A p = Ap0 + δA p (t).
87 where Pp0 = jAp0 j2 is the averaged input pump power and Γ0p is the intrinsic photon decay rate of the pump mode. Clearly, to the zeroth order, the relative magnitude of the dropped pump power modulation is directly equal to that of the input modulation: δPpd (t) δPp (t) = , Ppd Pp0 where δPp = Ap0 δA p + Ap0 δA p is the timevarying component of the input pump power. Eq. (4.10) leads to a pumpﬁeld modulation in the frequency domain of igOM ap0 x(Ω) i Γep δA p (Ω) , i(∆ p + Ω) Γtp /2 (4.13)
δa p (Ω) =
(4.14)
where δa p (Ω), x(Ω), and δA p (Ω) are Fourier transforms of δa p (t), x(t), and δA p (t), respectively, deﬁned as B(Ω) =
+∞ iΩt ∞ B(t)e dt.
Physically, the ﬁrst term in Eq. (4.14) represents the perturba
tion induced by the mechanical motion, while the second term represents the effect of direct input modulation. The optical gradient force is linearly proportional to the cavity energy as Fo =
gOMUp0 ωp gOMU p ωp .
With
modulation of the pump energy, the gradient force thus consists of two terms, Fo = Fo0 + δFo (t), where Fo0 = is the static force component introduced by the averaged pump energy Up0 =
jap0 j2 , and δFo (t) is the dynamic component related to the pump energy modulation δUp (t), given by δFo (t) = gOM δUp = ωp gOM a δa p (t) + ap0 δa p (t) . ω p p0 (4.15)
Substituting Eq. (4.14) into Eq. (4.15), we ﬁnd the force modulation is described by this general form in the frequency domain: i ap0 δA p (Ω) ap0 δA p ( Ω) Γep gOM + , ωp i(∆ p + Ω) Γtp /2 i(∆ p Ω) + Γtp /2
δF0 (Ω) = fo (Ω)x(Ω) +
(4.16)
where the ﬁrst term represents the back action introduced by the mechanical motion, with a spectral response f0 (Ω) given by fo (Ω) ∆2 Ω2 + (Γtp /2)2 + iΓtp Ω 2g2 jap0 j2 ∆ p p OM . 2 + (Γ /2)2 (∆ ωp (∆ p + Ω) Ω)2 + (Γtp /2)2 tp p
(4.17)
Figure 4.5(d) shows the spectral response of a probe resonance to smallsignal sinusoidal pump
88 modulation for several different (average) pump dropped powers. When the pump dropped power is low, the pump backaction on mechanical motion is negligible and the probe response is given by a combination of the intrinsic mechanical stiffness and the squeezeﬁlm effect [84] of trapped gas in between the rings. When the pump power is increased, however, the mechanical resonance frequency increases correspondingly, reaching a value of 22.3 MHz at a dropped power of 0.85 mW. This value is about 32 times larger than the intrinsic mechanical frequency, and implies a dynamical stiffness more than 1000 times that of the silica rings. The spiderweb ring resonators are separated by a 150 nm gap, which is only about 2.2 times the mean free path in a nitrogen environment ( 68 nm). As the ring is 6.3 µm wide, much
larger than the ring gap, the nitrogen gas sandwiched in the gap is highly conﬁned by the two silica layers and cannot move freely during the ﬂapping motion of the two rings. The resulting signiﬁcant pressure differential between the internal and external regions of the paired silica rings functions as a viscous force to damp the mechanical motion. This phenomenon is wellknown as the squeezeﬁlm effect, which has a profound impact on the dynamic response of micro/nanomechanical systems [84]. Apart from the optical gradient force, the squeezeﬁlm effect is the dominant mechanism responsible for the dynamic mechanical response of our devices. The associated damping force can be described by a general form of Fsq (Ω) = fsq (Ω)x(Ω), where fsq (Ω) represents the spectral response of the squeeze ﬁlm. In general, the squeezeﬁlm effect is typically described by two theories which work in quite different regimes, depending on the Knudsen number Kn characterizing the ratio between the meanfree path and the gap [84]. In the classical regime with Kn 1 where the gas can be considered
a continuum, the squeezeﬁlm viscous force for a rectangular plate is well described by fsq (Ω) = ke (Ω) + iCd (Ω), where ke and Cd represent the spring constant and damping, respectively, induced by the squeeze ﬁlm. They are given by the following equations [129] ke (Ω) = Cd (Ω) = 64σ2 Pa L0W0 1 ∑ m2 n2 [(m2 + (n/η)2 )2 + σ2 /π4 ] , π8 h0 m,n odd 64σPa L0W0 m2 + (n/η)2 , ∑ 22 2 2 2 2 4 π6 h0 m,n odd m n [(m + (n/η) ) + σ /π ] (4.18) (4.19)
where Pa is the ambient gas pressure, W0 and L0 are the width and length of the plate, h0 is the gap,
89 η = L0 /W0 is the aspect ratio of the plate, and σ is the squeeze number given by σ(Ω) = 12µeffW02 Ω , Pa h2 0 (4.20)
1.159 where µeff = µ/(1 + 9.638Kn ) is the effective value of the viscosity coefﬁcient µ [130]. Under
this model, the squeeze ﬁlm functions primarily as a damping (or elastic) force when the modulation frequency is below (or above) the cutoff frequency given by Ωc = π2 Pa h2 0 12µeff 1 1 + 2 2 W0 L0 . (4.21)
In contrast, in the freemolecule regime with Kn
1 where the interaction between gas molecules
is negligible, the squeeze ﬁlm approximately behaves like a damping force, fsq (Ω) = iCr Ω, with Cr given by the following equation [131, 132] S 16πh0 2Mm , πR T
Cr =
4Pa L0W0
(4.22)
where Mm is the molar mass of gas, T is the temperature, R is the ideal gas constant, S is the perimeter length of the gap region. However, our devices have a Knudsen number of Kn = 0.45, falling in the crossover regime where neither theory adequately describes the squeezeﬁlm effect [133]. As the device works in the regime between the continuum and freemolecule limit, we heuristically propose that the damping/elastic force of the squeeze ﬁlm is effectively described by a composite of the two theories: fsq (Ω) = ke (Ω) + iCd (Ω) + iηrCr Ω, (4.23)
with a modiﬁed effective coefﬁcient of viscosity µ0 = ηµ µeff , where ηr and ηµ are parameters used eff for a best description of the squeezeﬁlm response in our devices. Detailed analysis shows that ηµ = 0.7 and ηr = 0.03 provides the best ﬁt for our devices. As our devices have a spiderweb geometry, we approximate it with an equivalent rectangular shape with W0 given by the ring width, L0 given by the circumference at the ring center, and S 2L0 . As shown by the experimental results
and theoretical ﬁts, this model provides an accurate description of the squeezeﬁlm effect in our devices. Although the intrinsic mechanical frequency of the 54 µm spiderweb structure is 694 kHz (in
this static mechanical displacement primarily changes the equilibrium position of the mechanical motion. as it helps to extend the modulation bandwidth for wavelength routing.26) (4. (4.17). a Markovin process with the following correlation function: hFT (t)FT (t + τ)i = 2meff Γm kB T δ(τ). On the other hand. m 2 dt dt meff meff (4.90 dicated by FEM simulation). respectively. With the optical gradient force and the squeezeﬁlm damping force. and Ωm and Γm are intrinsic mechanical frequency and damping rate. (4. Equation (4. meff Ω2 km ω p km ω p Γ0p m (4. FT is the thermal Langevin force responsible for the thermal Brownian motion. (4. dominated by the squeezeﬁlm damping. With a speciﬁcally m designed extremely small spring constant. 134]. since both the squeezeﬁlm damping force and dynamic component of the optical force affect only the dynamics of x0 . 4. Fig.24) shows clearly that the static mechanical displacement is actuated only by the static component of the optical force given by x0 = gomUp0 gom Ppd jFo0 j = = . It is convenient to remove this component in Eq. it is beneﬁcial in this case. (4.24) in the frequency domain. the mechanical motion of the cavity satisﬁes the following equation: d2x dx 1 1 + Γm + Ω2 x = (Fo + Fsq + FT ) = (Fo0 + δFo + Fsq + FT ). the cavity resonance can be tuned by a signiﬁcant magnitude of gOM x0 .24) by deﬁning x0 = x x0 . Interestingly. Substituting Eqs. This is the primary mechanism responsible for the resonance tuning.23) into Eq. As the squeezeﬁlm viscous force is zero at Ω = 0.25) where km = meff Ω2 is the intrinsic spring constant of the spiderweb structure. we ﬁnd that the squeezeﬁlm damping force and the backaction term of the optical force primarily change the values .5(d) shows a minimum dynamic frequency response of 6 MHz. (4. x0 can be quite signiﬁcant for a given dropped power. where kB is Boltzmann’s constant. As a result.16).24) where meff is the effective motional mass of the ﬂapping mechanical mode. although squeezeﬁlm damping is generally detrimental in other micro/nanomechanical systems [84. the squeeze gas ﬁlm impacts only the dynamic response of mechanical motion.
28) where we have dropped the prime notation of x0 for simplicity.25).30) 0 where km = meff (Ω0 )2 is the effective spring constant and the approximation in the ﬁnal term asm sumes a negligible change in the mechanical linewidth.16). we ﬁnd the spectral density of thermal mechanical displacement has the form Sx (Ω) = 2Γm kB T . (4. the mechanical motion is primarily dominated by the dynamic optical force rather than the actuation from the thermal Langevin force. The ﬁrst term in Eq. In the absence of pump modulation.27) show that one dominant effect of the pump energy inside the cavity is to increase the mechanical rigidity.91 of the resonant frequency and damping rate of the mechanical motions. In most cases.29) thus leads to a variance of the m thermal mechanical displacement given by h(δx)2 i = 1 2π +∞ Sx (Ω)dΩ = ∞ kB T Γm 0 km Γ0 m kB T .17).29) Equations (4. the mechanical motion is dominated by the Brownian motion.5(a). Thus.27) the mechanical displacement is thus given by ap0 δA p (Ω) ap0 δA p ( Ω) i Γep gOM FT (Ω) + + . 0 km (4. L (Ω) can be well approximated by L (Ω) (Ω0 )2 m Ω2 iΓ0 Ω with a new mechanical resonance Ω0 m m and damping rate Γ0 affected by the optical force. meff L (Ω) meff ω p L (Ω) i(∆ p + Ω) Γtp /2 i(∆ p Ω) + Γtp /2 x(Ω) = (4. as shown in Fig. and (4. (4.28) represents the thermal Brownian motion while the second term describes the motions actuated by the pump modulation. (4. Equation (4. the increase of the mechanical resonance frequency through the optical spring effect dramatically suppresses the magnitude of the thermal mechanical displacement and its perturbation of the cavity resonance. 4.28). meff (4. (4. and the ﬁrst term is negligible compared with the second term in Eq. In the presence of pump modulation. the socalled optical spring effect. By using Eq. Deﬁning L (Ω) Ω2 m Ω2 iΓm Ω fo (Ω) meff fsq (Ω) . Clearly. meff jL (Ω)j2 (4. we neglect the thermal Brownian term in the following discussion. .
2 Γts )δas + 2iγUp0 δas igOM xas0 + 2iγδUp as0 . (4.32) (4. it is negligible compared with the cavity linewidth at the power level used for exciting optomechanical effects. The second terms of Eqs. (4.14).8): das = (i∆s dt Γts )as 2 igOM xas + 2iγja p j2 as + i Γes As . the intracavity probe ﬁeld can be written as as = as0 +δas (t).34) and (4. we ﬁnd that the power spectrum of the transmitted probe . the modulation of the transmitted probe power thus takes the form δPTs = i Γes (A0s δas A0s δas ). governed by the following equations: das0 = (i∆s dt dδas = (i∆s dt Γts )as0 + 2iγUp0 as0 + i Γes As .35). (4.31) except that the Kerrnonlinear term now describes the crossphase modulation from the pump wave.28). leading to 2γUp0 Γtp . In general. (4. (4. Γts /2 i∆s and Eq. which can be included in the cavity tuning term ∆s for simplicity. (4. By use of Eqs.34) where δUp (Ω) is the Fourier transform of δUp (t). Γts . (4.36) p where A0s = As + i Γes as0 is the transmitted probe wave in the absence of modulation.92 The probe wave inside the cavity is governed by a dynamic equation similar to Eq. (4. With the perturbations induced by the pump modulation. As the transmitted ﬁeld of the probe is given by p ATs = As + i Γes as .32) and (4. (4.33) results in a probeﬁeld modulation in the frequency domain of ias0 gOM x(Ω) δas (Ω) = i(∆s + Ω) 2γδUp (Ω) Γts /2 .35) (4.33) represent the static cavity tuning introduced by crossphase modulation. 2 (4. Equation (4.32) provides a steadystate solution of p i Γes As as0 = .33) where we have assumed the probe input is a continuous wave with a power of Ps = jAs j2 . similar to the previous discussion of the pump wave.
4. (4. For those weakly actuated mechanical modes.5(d) are normalized by a factor corresponding to the ratio of the dropped power for each curve relative to the maximum dropped power.5 are . effective motional mass. since it is most strongly actuated by the optical gradient force. we have used Eq.39) where Ppd0 = 0. Equation (4. respectively. Γts . We also assume the cavity is in Γtp . and δPTs (Ω) and δPpd (Ω) are the Fourier transforms of δPTs (t) and δPpd (t).40) was used to describe . To obtain Eq. but weakly coupled to the optical waves inside the cavity. such that 2γUp0 the sidebandunresolved regime with Ωm deﬁned as ρ(Ω) jδPTs (Ω)j2 /Ps2 2 jδPpd (Ω)j2 /Ppd Γtp . (4. Therefore. the plotted modulation spectra are given by ρ0 (Ω) 2 Ppd0 2 Ppd ρ(Ω) = jδPTs (Ω)j2 /Ps2 2 jδPpd (Ω)j2 /Ppd0 .37) where Γ0s is the intrinsic photon decay rate of the probe mode. following the same procedure above.13) to relate the dropped pump power to the input. In this case.37).5(d). for the jth mechanical mode. it is easy to show that the spectral response of probe modulation now becomes g2 j mj ω p L j (Ω) 2 2 Ppd ρ(Ω) = 2γ + ∑ j 4Γ2 Γ2 ∆2 es 0s s Γ2 [∆2 + (Γts /2)2 ]4 0p s . (4. and have also taken into account the fact that the Kerr effect is relatively small. and the spectral response of mechanical motions. The modulation spectra given in Fig. the modulation spectra shown in Fig. respectively. L j (Ω) = Ω2 mj Ω2 iΓmj Ω where Ωmj and Γmj are the reso nance frequency and damping rate of the jth mechanical mode. 4.85 mW is the maximum drop power used in Fig.93 modulation is given by the following equation: 2 2 Ppd jδPpd (Ω)j2 4Γ2 Γ2 ∆2 g2 jδPTs (Ω)j2 es 0s s OM = + 2γ . m j . In general. (4. there are many mechanical resonances for the spiderweb resonators.38) For a better comparison of the dynamicbackaction induced variations on the probe modulation. 2 2 + (Γ /2)2 ]4 Ps2 meff ω p L (Ω) Γ2 Ppd [∆s ts 0p (4.40) where g j . and L j (Ω) are optomechanical coupling coefﬁcient. 4. (4. The derivations above take into account only the ﬂapping mechanical mode.
where a 1.5(d) shows that the optical spring effect enables a modulation time as fast as 44. other optical (material.) nonlinearities can also contribute to the probe modulation. This ratio agrees well with the theoretical value of 81 dB given by g2 OM 2γmeff ω p Ω0 Γ0 m m 2 .9% modulation of the pump power is large enough to introduce considerable fractional modulation in the probe time waveform (top panel).5(b). respectively. 4. middle panel) results in a probe modulation of larger than a halflinewidth (full contrast modulation). 117. Further increase in the pump modulation depth actuates ﬂapping mechanical motion so intense it begins to excite a second mechanical mode (the Fanolike feature in Fig.85 mW there is a resonant modulation “gain” of greater than 20 dB. however.5(d) shows that for Pd = 0.37) provides an accurate description of the pumpprobe modulation response (the Fanolike resonance seen at Pd = 0.5(b). or microelectricalmechancial approaches [108–111. roughly 3 orders of magnitude faster than modulation schemes based upon thermooptic. As is common in micro/nanomechanical systems [114. This can also be seen in the time waveform of the probe in Fig. 4. As shown in the expanded modulation spectrum of Fig. One metric for characterizing the response time of the spiderweb optomechanical cavity is the resonant oscillation period [120].36 µs on the probe time waveform.9% (Fig. the small modulation of the pump wave which actuates the mechanical oscillation is greatly magniﬁed on the probe resonance. one is more interested in the impulse response of the system. Increasing the pump modulation to 14. where Ω0 and Γ0 are the effective m m mechanical resonance frequency and damping rate. consistent with the mechanical linewidth of 23 MHz (see Fig. 4. 4. The Kerr nonlinearity in silica has . 120]. 4. Figure 4. This can be further enhanced by using the transduction “gain” to push the probe modulation into the nonlinear regime. where in the lower panel of Fig. and is discussed in Chapter 3). The measured settling time constant of the probe response is 196 ns.85 mW is due to intrinsic mechanical coupling between different types of motion. eq. (4.5(d). the resonant optomechanical nonlinearity is dominant out to a frequency of 500 MHz.94 the modulation spectrum shown in Fig. The Kerr nonlinearity is measured to be 78 dB below the resonant optomechanical response. resulting in a beat signal with a period of 0. As shown in the dashed curves in Fig.5(e).5(d). 4. 4. optoﬂuidic. Figure 4. 4. after which the response plateaus due to the ultrafast Kerr nonlinearity of silica. etc. photochemical. The pulsed response of the probe is shown in Fig.5(d)). 120].5(c). with a settling time determined by the mechanical linewidth. For many switching applications.5(d)). In addition to the optomechanical nonlinearity.8 ns. the resonant response causes ringing during switching. In general.5(b) the probe wavelength (10%–90%) onoff switching time is reduced to 7 ns. 4. 113–115.
95 been extensively studied over more than three decades for optical signal processing [126. and modulators to enable full control and functionality of the optomechanics. and the excellent agreement between the theoretical and experimental spectra provides yet another indication that the optical gradient force is the dominant tuning mechanism in the spiderweb cavity structure. is sought to improve the switching time. one can also incorporate other chipbased optical components such as waveguides. 136]. 120]. dispersion compensation [138]. For example. Given the similarity of the doublering spiderweb stucture to other more conventional planar microring technologies. An increase in the tuning range and efﬁciency (actuation power) can be expected with further engineering of the mechanical stability of the spiderweb structure. 125]. lasers. a reduction in the mechanical Qfactor. There are also many wellestablished methods for managing the dynamical response. of resonant micro.and nanomechanical systems [114.5 Discusssion The versatility of the gradient optical force tuning approach described here provides considerable room for future improvement of device performance. . One example technology would be an onchip reconﬁgurable optical add/drop multiplexer or wavelength selective switch/crossconnect. other prospective applications for optomechanical devices include tunable optical buffering [137]. which can be obtained through elevated gas pressure or incorporation of damping materials. and nonlinear signal processing [126]. In contrast to cavityoptomechanical applications such as cooling and ampliﬁcation of mechanical motion [50. 135. In addition to the demonstrated wavelength routing. tunable lasers [139]. 4. which could be accomplished by integrating an array of doublering cavities into a parallel or cascaded conﬁguration. in particular the ringing. the 90 µm diameter ﬁrstorder spiderweb cavities should allow for a sixfold increase in tuning efﬁciency to approximately 15 nm/mW.
and each has been shown to be effective in at least initial demonstrations. we also demonstrate tunable coherent mechanical mode mixing with an analogy to electromagnetically induced transparency. showing the possibility for slowlight effects on the very long phononic timescale. A novel optomechanical device structure has also been presented. and are detector agnostic. and 100% channel quality preservation. to current detector focal plane array processing. Each of these resonator designs was developed and optimized for a particular range of applications. This device has a very large optomechanical coupling and a high quality factor.96 Chapter 5 Conclusion Here has been presented work on several optical resonator systems: the single and doublemetal plasmonic photonic crystal resonator. and tailorable polarization and spectral sensitivity. Multispectral midinfrared resonant detectors were demonstrated with enhanced responsivity and detectivity. and the doublering spiderweb cavity. an extremely ﬂexible doublering optomechanical device is shown. and the potential for phononphoton quantum state transfer. demonstrating alloptical wavelength routing with unprecedented range and efﬁciency. as well as for more fundamental physicsbased applications such as . with design principles that are easily transferrable to any other detector material or frequency range with a minimum of difﬁculty. consisting of two stacked microdisks with an optically narrow gap between them. As this device can be easily integrated onchip. it shows great promise for optical communications applications. as well as absorption enhancement. the doubledisk whisperinggallery cavity. Expanding to a doublemetal device structure. this method could be used to easily and inexpensively impart frequency and polarization selectivity. Finally. These devices were fabricated using a very simple singleetch process. giving rise to extremely large dynamical backaction in the form of both regenerative mechanical oscillation and optomechanical cooling. Due to the large optical spring effect in these structures.
97 in cavity quantum electrodynamics or for dispersion compensation in nonlinear optics. .
” Journal of the Optical Society of America B: Optical Physics 16. Painter. Chang. A.” Journal of Physics D: Applied Physics 38. Scherer. Attaluri. Vandervelde. Sharma. and Y. Vuckovic. Rogalski.” Journal of Vacuum Science and Technology B 26. Surface Plasmons on Smooth and Rough Surfaces and on Gratings (Springer Tracts in Modern Physics) (Springer. Shenoi. . T. “Lowstrain InAs/InGaAs/GaAs quantum dotsinawell infrared photodetector. A. Qiu. [7] M. 1725–1733 (1991). Gogna.” Journal of Applied Physics 105. L. “Peculiarities of the MBE growth physics and technology of narrowgap IIVI compounds. 339–341 (2002). [2] J. P. 2142–2150 (2005). Krishna. [10] H. [3] Y. 1136–1139 (2008). A. Y. Berlin.” Applied Physics Letters 81. Sidorov. R. V. [5] R. T. J. “Polarization properties of dipolelike defect modes in photonic crystal nanocavities. 253–265 (1997). Siroya. 4590–4594 (2002). J. [8] O. E. and A. “Lowthreshold photonic crystal laser. “Evaluation of the fundamental properties of quantum dot infrared detectors. Loncar.” Optical Engineering 30. 1988). S. J. 091101 (2009).” Thin Solid Films 306. Stintz.” Optics Letters 27. Raether. 275–285 (1999). “Quantum dotsinawell infrared photodetectors. and L. Antoszewski. and S. D. Scherer. Phillips. [9] M. Faraone. I. [6] O. [4] S. Althouse and C. “Thirdgeneration infrared photodetector arrays. Srinivasan. 2680–2682 (2002). “Defect modes of a twodimensional photonic crystal in an optically thin dielectric slab. “Chemical vapor detection with a multispectral thermal imager. Krishna. Yoshie.” Journal of Applied Physics 91. Shao. Painter and K.98 Bibliography [1] A.
[13] J.” Applied Physics Letters 95. 686–688 (2009). [15] E. Yu. A. Delgehier. Krishna. Ebbesen. [17] X. Painter. S. F. Y. X. “Polarizationselective resonant photonic crystal photodetector. Miard. Bahriz. R. S. Rosenberg. Li. Rosenberg. M. V. E. K. Chandran. G. Sauvage. and O. Shenoi. [12] M. J. Homeyer. Attaluri. A. S. Lin. J. S. and S. [18] J. W. p. and T. Houel. Lee. Sagnes. 3672– 3686 (2010). Shenoi. J. White. L. [14] E. Painter. “Plasmonic photon sorters for spectral and polarimetric imaging. W. “Design of midIR and THz quantum cascade laser cavities with complete TM photonic bandgap. Shenoi. Prade. Express 18.” Optics Express 15. Krishna. Z. and S. O. [20] J. Ramirez. C. Checoury. I. Moreau. M. R. 6713. Painter. K. 241108 (2008). Crisafulli.” Opt. [16] J. “Guided optical waves in planar heterostructures with negative dielectric constant. S. Genet. 5948–5965 (2007). B. 2007). H. Colombelli. Y. Skauli. M. Le Gratiet. Y. Hu. Seo. “Plasmon assisted photonic crystal quantum dot sensors. Laux. Leung. S. and O. and A. and O.” Applied Physics Letters 93. K. R. L. A. Boucaud. 13556–13572 (1991). 161–164 (2008). “Design of plasmonic photonic crystal resonant cavities for polarization sensitive infrared photodetectors. Kim. D.” Optics Letters 34. Braive. P.” Nature Photonics 2. . Vinet. J. Ye. L. 211103 (2008). R. “A multispectral and polarizationselective surfaceplasmon resonant midinfrared detector. Krishna. “Design of midinfrared photodetectors enhanced by resonant cavities with subwavelength metallic gratings.” vol. “Extraordinary optical absorption through subwavelength slits.99 [11] B. Lemaˆtre. Brongersma. S. V.” Applied Physics Letters 95. Yang. Barnard. Ho. S. V. R. V. A. and I. 041108 (2009). Veronis. 161101 (2009).” Physical Review B: Condensed Matter 44. Leroy. Vandervelde. 67130P (SPIE. O. Mysyrowicz. Y. Hwang. Painter. [19] R. and Y. K. Z. T. E.” Applied Physics Letters 93. “Resonant coupling of quantum dot intersublevel transitions with midinfrared photonic crystal modes. Sharma. and M. T. Rosenberg. Fan.
“Quantum DotsinaWell Focal Plane Arrays. NY. New Jersey. . and T. Johnson and R. Sandberg. Krishna. Zucker. M. D.” IEEE Journal of Selected Topics in Quantum Electronics 14. W. W. Barnes.” Nature 424. Varley. S. R. GarciaVidal. W. V. D. P. MartinMoreno. R. D. Srinivasan. and M. S. K. [26] P. Weiss. Sievers. Chicago. Tinkham. Shao. “Wannierlike equation for the resonant cavity modes of locally perturbed photonic crystals. “Mimicking Surface Plasmons with Structured Surfaces. Lenz. E. L. 325 (1992). S. Spero. 1977).” Physical Review B: Condensed Matter 68. Drever. Manukin. Thorne. 035214 (2003). Braginsky and A. Cai. E. Painter. E. Barve. V. K. “Optical Constants of the Noble Metals. Ramirez. O. Christy. Caves. “Observation of Critical Coupling in a Fiber Taper to a SilicaMicrosphere WhisperingGallery Mode System. [24] W. C. 1995). A. B. Thorne. R. W. J. J. A. [29] V. “LIGO: The Laser Interferometer GravitationalWave Observatory. and F. Vandervelde. E. W. D. [25] J. 824–830 (2003). B.” Physical Review Letters 85.” Reviews of Modern Physics 52. R. R. Abramovici. “Surface plasmon subwavelength optics. R. Meade. B. Voigt. L. Shoemaker. Photonic Crystals (Princeton University Press.” Physical Review B: Condensed Matter 6. P. R. Pendry. and P. L. Sharma. N. IL. A. “On the measurement of a weak classical force coupled to a quantummechanical oscillator. Dereux. Shenoi. W. E.” Science 256. Drever. Painter. E. and K. B. [22] O. Y. Joannopoulos. 2003). [28] T. K. Whitcomb. Seiji. New York. [31] A. Gursel.” Science 305. 1150–1161 (2008). Vahala. 341–392 (1980). 4370–4379 (1972). Ebbesen. and J. Measurements of Weak Forces in Physics Experiments (The University of Chicago Press. Althouse. Princeton. Jan. [30] C.100 [21] M. Y. E. E. [23] J. 74–77 (2000). Winn. Barclay. 847–848 (2004). and M. J. [27] M. and S. D. Group Theory and Quantum Mechanics (Dover Publications. D. Zimmermann.
Pinard.101 [32] C. B. Tuefel. M. J. Nooshi. Y. Capasso. K. J. Arcizet. Aspelmeyer. 243905 (2006). [34] T. M. “Cavity Optomechanics: BackAction at the Mesoscale. Povinelli. N. M. B. 555–560 (2008). “Analysis of RadiationPressure Induced Mechanical Oscillation of an Optical Microcavity. “Radiation Pressure Cooling of a Micromechanical Oscillator Using Dynamical Backaction.” Nature Physics 4. J. K. M. Del’haye. M. “Measuring nanomechanical motion with a microwave cavity interferometer. Braginsky. “Resolvedsideband cooling of a micromechanical oscillator. Girvin. 8287–8295 (2005). W. [36] S. Cohadon. Hertzberg.” Science 321. F. 71–73 (2006). 033901 (2005).” Nature 444. Bauerle. G. Schliesser. Regal. [39] A. Smythe. “Subkelvin optical cooling of a micromechanical resonator. Vahala. and A. [37] O. E.” Physical Review Letters 95. Bouwmeester. [38] D. Paternostro. H. 72–75 (2008). 1995). Bohm. Anetsberger. Zeilinger. F. “Radiationpressure cooling and optomechanical instability of a micromirror. Kippenberg and K. D. P.” Nature 444. and K. G. A. 75–78 (2006). Lehnert. UK. F. Loncar. and K. [41] A. 67–70 (2006). S. Ibanescu. Blaser. S. M. Marquardt. D. R. S. L. C. Zwickl. [35] T. [42] M. J. Schliesser. Cambridge. Kippenberg. T. [40] J. H.” Nature 452. Scherer. Kippenberg. Thorne. Kippenberg. and T. T. A. Joannopoulos.” Physical Review Letters 97. 1172–1176 (2008). “HighQ enhancement of attractive and repulsive optical forces between coupled whisperinggallerymode resonators. 415–419 (2008). Johnson. Khalili. D. F. Gigan. J. Langer. B. F.” Nature 444. and J. Thompson. O. J. P. [33] V. Kleckner and D. Vahala. and J.” Optics Express 13. Harris. D. and T. A.” Nature Physics 4. G. Schwab. Rokhsari. Arcizet. G. R. M. Carmon. “Selfcooling of a micromirror by radiation pressure. Riviere. Jayich. and K. Briant. J. E. J. J. M. and A. “Strong dispersive coupling of a highﬁnesse cavity to a micromechanical membrane. Heidmann. Vahala. J. Quantum Measurement (Cambridge University Press. .
“Trapping. . Arcizet. California Institute of Technology (2006). Craighead.and nanometrescale photoniccrystal optomechanical cavity.” Physical Review A: Atomic. [44] M. Parpia. K. Wright. M. T. 103601 (2009). J. P.” Ph. Hochberg. Horber and M. 658–665 (2007). “Properties of piezoresistive silicon nanoscale cantilevers with applications to BioNEMS. 1002– 1005 (2003).D. W. Pernice. H. and J. D. Painter. O. Mccullen.” Nature 459.102 [43] P. J. [45] M. [47] J. R. and Optical Physics 15. “Mechanical Oscillation and Cooling Actuated by the Optical Gradient Force. and E. J. E. 013101 (2008). Meystre. Miles. A. “Cavity OptoMechanics. 689–699 (1977). Vahala. C. H. Kippenberg and K. A Stochastic Theory of Line Shape (John Wiley & Sons. “Scanning Probe Evolution in Biology. “Size and frequency dependent gas damping of nanomechanical resonators. M. M. M. Q. 1969). Painter. M. S.” Science 302. J. P. J. “Theory of radiationpressuredriven interferometers. and O. P. Eichenﬁeld. and T. T. NY. Lin. Arlett. E. M.” Optics Express 15.” Applied Physics Letters 93. and O. Kippenberg. Rosenberg. Ippen. H. 909–914 (2009). Chan. “Nearﬁeld cavity optomechanics with nanomechanical oscillators. “Resonance ﬂuorescence with excitation of ﬁnite bandwidth. Li. and E. [48] Q. 480–484 (2008). X. Unterreithmeier. Verbridge. “A picogram. and H. corralling and spectral bonding of optical resonances through optically induced potentials.” Nature Photonics 1. “Harnessing optical forces in integrated photonic circuits. [53] R. Vahala. Soljacic.” Physical Review Letters 103. [52] S.” Journal of the Optical Society of America B: Optical Physics 2. Kotthaus. [51] G. [49] P. thesis. [50] T.” Nature Physics 5. Kubo. 17172– 17205 (2007). BaehrJones. Rakich. J. Vahala. Jiang. Weig. Tang. K. P. [54] H. Camacho. Kimble and L. Popovic. J. Mandel. 1830–1840 (1985). Anetsberger. 550–555 (2009). Illic. Xiong. K. [46] J. J. G. Vignes. X.” Nature 456. R. Molecular. New York. J. J. J.
Stowe. Seidl. C. Badolato. and A. [63] S. F.” Physical Review Letters 99. K. and E. D. [58] I. and K. Karrai. S. S. and S. Vahala. N. B. Sheard. “Demonstration of an erbiumdoped microdisk laser on a silicon chip. West. and Optical Physics 69. Mitsugi. 051802 (2006).” Nature 451. N. E. J. Yasumura. Field. Remi. D. “Quantum theory of cavityassisted sideband cooling of mechanical motion. Wago. S. W. Whitcomb. 023903 (2006). Heidmann. W.” Physical Review Letters 97. R.” Applied Physics Letters 71. Pinard.” Physical Review A: Atomic. E. P. [60] T. Taniyama. Kippenberg. J. “Cooling of a Mirror by Radiation Pressure. M. K. O. Polman. Molecular. “Nonlinear optical processes using electromagnetically induced transparency. F. “Observation and characterization of an optical spring. Chen. [59] T. Pfeiffer. and S. E. [62] U. McClelland. A. 311–314 (2008). Govorov. and D.” Physical Review Online Archive (Prola) 124. A.” Nature 390. J. Sirtori.103 [55] B. Zhang. Warburton. Marquardt. J.” Physical Review Letters 64. Rugar. “The nonlinear Fano effect. 1866–1878 (1961). Nooshi. Gerardot. and Optical Physics 74. B. Capasso. B. Clerk. S. Cohadon. Biedermann. and T. [65] M. 288–290 (1997). R. Molecular. W. and K. H. Gray. Notomi. A.” Physical Review Letters 83. Botkin. “Controlling the sign of quantum interference by tunnelling from quantum wells. 093902 (2007). Kippenberg. Girvin. Rae. Barbour. A. Fano. D. M. W. and M. 3174–3177 (1999). J. D. 093901 (2007). [64] J. J. 589–591 (1997). “Theory of ground state cooling of a mechanical oscillator using dynamical backaction. Kalkman. Imamoglu. “Effects of Conﬁguration Interaction on Intensities and Phase Shifts. [61] M. . J. Kuramochi. C. “Attonewton force detection using ultrathin silicon cantilevers. Faist. Kroner.” Physical Review A: Atomic. M. 051801 (2004). Petroff. “Optomechanical Wavelength and Energy Conversion in HighQ DoubleLayer Cavities of Photonic Crystal Slabs. Zwerger. A.” Physical Review Letters 99. E. 1107–1110 (1990). P. W. [56] P. K. Harris. Mow Lowry. A. and L. M. T. [57] F. Kenny.
D. Sandhu. Shakya. M. Johnson. E. and Optical Physics 69. Kitajima. Painter. E.1038/nphoton. 3042–3044 (2005). . [72] Q. Camacho. “Evanescentwave bonding between optical waveguides. doi:10. G. [68] S. 213904 (2007). Vahala. Molecular. O.” Nature Materials 8. “Experimental Realization of an OnChip AllOptical Analogue to Electromagnetically Induced Transparency. Loncar. Hase. [67] M. and M.” Optics Letters 30. Langguth. and J.4. T. S. Liu. Fan. D. T. and M. Smith. Chang. and H. and M. H. E. Nikonov.” Reviews of Modern Physics 46. Kobayashi. 063804 (2004). Povinelli. Boyd. [75] Q. Weiss. Kastel. S.” Physical Review Letters 98. Tomita. A. Giessen. M. F. W. “Photoinduced Fano resonance of coherent phonons in zinc. Ibanescu. J. Smythe. and R. J. [70] S.” Physical Review B: Condensed Matter 74. “Lasers without inversion: Interference of lifetimebroadened resonances.” Physical Review A: Atomic. M. [71] D. “Coherent mixing of mechanical excitations in nanooptomechanical structures. Pfau. and O. Fleischhauer. Demsar. J. 908–910 (2002). Imamoglu. 1033–1036 (1989). Fan. M. Chang. R. Harris.” Physical Review Letters 62. L. [69] D. 12212–12215 (1999). Lipson. [73] K. A. 83 (1974). J. Rosenberger. “Coupledresonatorinduced transparency. 123901 (2006). A. N. Joannopoulos. D. and M. Povinelli.104 [66] J. “Softmode spectroscopy: Experimental studies of structural phase transitions. 758–762 (2009). “Plasmonic analogue of electromagnetically induced transparency at the Drude damping limit. 212301 (2006). Totsuka. Xu. J. Lin. S.” Physical Review Letters 96. K. M. Eichenﬁeld. L.2010. “Slow Light in CoupledResonatorInduced Transparency. “Sharp asymmetric line shapes in sidecoupled waveguidecavity systems. [76] M. “Fano interference of collective excitations in semiconductor quantum wells and lasing without inversion. Fuller. L. Capasso. T. Scott. Scully. [74] N.” Nature Photonics (2010). Rosenberg.” Physical Review B: Condensed Matter 59. K. F. J.” Applied Physics Letters 80.
9 mK. A1732– A1737 (1964). 478–483 (2009). “Indirect Coupling and Antiresonance of Two Optic Phonons. Porto. S.” Physical Review Letters 24. N. S. A. and N.” Nature Photonics 3. Ottaway.” Physical Review Online Archive (Prola) 135. “Squeeze ﬁlm air damping in MEMS. “Actuation of microoptomechanical systems via cavityenhanced optical dipole forces.” Nature Photonics 1. [83] M. Bodiya. Sigg. Smith.” Reviews of Modern Physics 77. P. F. Perahia.” Optics Letters 32. Rousseau and S. S. Q. Wipf. Whitcomb. T. [80] M. and O.105 [77] M. C. Camacho. “Static and dynamic wavelength routing via the gradient optical force.” Physical Review Letters 99. Imamoglu. J. and KTaO3 . “Electromagnetically induced transparency: Optics in coherent media.” Sensors and Actuators A: Physical 136. and O. Yang. Hopﬁeld. J. Bao and H. “Optical and mechanical design of a “zipper” photonic crystal optomechanical cavity. “History of optical trapping and manipulation of smallneutral particle. D. [85] A. D. [88] A. R. P. Eichenﬁeld. H. L. “AugerLike Resonant Interference in Raman Scattering from One.” Physical Review Letters 20. M. Scott. and J. Rosenberg.” IEEE Journal of Selected Topics in Quantum Electronics 6. 416–422 (2007). Vahala. Painter.and TwoPhonon States of BaTiO3 . [84] M. Ruvalds. . Corbitt. Zadeh and K. “Coupledopticalphononmode theory of the infrared dispersion in BaTiO3 . C. [82] J. Marangos. 3–27 (2007). Painter. Barker and J. 1611–1613 (2007). and O. R. [78] J. and molecules. Eichenﬁeld. [87] J. Mavalvala. P. 160801 (2007). atoms. 1111–1114 (1970).” Physical Review Letters 24. “Observation of optical spring effect in a microtoroidal optomechanical resonator. [79] A. [86] D. Lin. Fleischhauer. SrTiO3 . [81] T. 841–856 (2000). Ashkin. “Optical Dilution and Feedback Cooling of a GramScale Oscillator to 6. 3802–3817 (2009). Chan. 1354–1357 (1968). Painter. 633–673 (2005). “Hybrid Phonons and Anharmonic Interactions in AlPO4 . 1107–1110 (1970). Zawadowski and J.” Optics Express 17. Michael.
P.” Nature 409. A. Behroozi. Riviere. [98] S. and J. S.106 [89] A. M. and M. . [90] C. Snider. and S. Prentiss. S. M. Xu. Anetsberger. Simmons. and J. S.” Nature Physics 3. G. D. [100] J. R. “Coupledpendulum model of the stimulated resonance Raman effect. G. and L.” Physical Review A: Atomic. 37–41 (2002). Chaves. “Stopping and storing light coherently.Y. Lukin. Lipson. [93] M. G. A. G. “Optical networking: past. 1613–1623 (1988). Fatih and S. V. Parpia. H. S. Berthold. “Coupled modes with A1 symmetry in tetragonal BaTiO3 . Verbridge. Katiyar. G. Wang. Craighead. “A megahertz nanomechanical resonator with room temperature quality factor over a million.” Nature 426. G. A. [91] P. “Ultralowdissipation optomechanical resonators on a chip. Nussenzveig. Dutton. 638–641 (2003).” Journal of Lightwave Technology 26. Porto. Martinez. O. “Classical analog of electromagnetically induced transparency. “Nanoelectronic and Nanophotonic Interconnect. P. L. R. [95] M. M. 1104–1118 (2008). A. Kuekes. Zibrov. 083901 (2004). [99] R. Blair. [97] Q. Williams. Fan. [92] C. J. P. 013112 (2008). A. Molecular. Bajcsy. and T. present. and R. Schliesser.” American Journal of Physics 70. Dong. [94] M.” Proceedings of the IEEE 96. “Stopping Light All Optically. Alzar. [96] G. Liu. Beausoleil. and M. Kippenberg.” Physical Review B: Condensed Matter 10. “Observation of coherent optical information storage in an atomic medium using halted light pulses. 3522–3533 (1974). C. “Breaking the delaybandwidth limit in a photonic structure. Yanik and S. “Stationary pulses of light in an atomic medium. 406–410 (2007). and future. Z. H. Saleh. F. L. R. M. S. Hemmer and M. Hau. and Optical Physics 71.” Nature Photonics 2.” Applied Physics Letters 92. 013803 (2005).” Physical Review Letters 92. S. 490–493 (2001). J.” Journal of the Optical Society of America B: Optical Physics 5. 230–247 (2008). Fan. and P. 627–633 (2008). Arcizet. S.
Steier.” Bell Labs Technical Journal 11 (2006). “Tunable polymer double microring ﬁlters. J. [107] B. and M. Neilson. C. Gripp. Earnshaw. [102] J. J. Steinmeyer. B. [111] J. 4112 (2002).” IEEE Photonics Technology Letters 10. J. 37–38 (2005). P. [109] P.” Journal of Lightwave Technology 21. Kimerling. 1046–1063 (2008). Song. Sadot and E. 2839 (2003). G. Eldada. “Widerange tuning of polymer microring resonators by the photobleaching of CLD1 chromophores. S. R. . Rabiei and W. Bhardwaj. Dalton. E. “Tunable optical ﬁlters for dense WDM networks. Duelk. T. M. Tomlinson. O. M. E. [104] D. A. Thoen. “Realizing the advantages of optical reconﬁgurability and restoration with integrated optical crossconnects. E. and M. 2584–2586 (2004). M. Laznicka. C. Poon. S. J. Zirngibl. R. Bermel. “Optical communication components. Nakamura. Yi.” Applied Physics Letters 81. 549–551 (1998).” IEEE Communications Magazine 36. Kunishi. T.” IEEE Photonics Technology Letters 15. P. C. Chu. K. and S. A. “Dynamic wavelength tuning of channeldrop device in twodimensional photonic crystal slab. H. 1255–1257 (2003). A.” Optics Letters 29. “Optical switch fabrics for ultrahighcapacity IP routers. W. [103] D. Doerr. Haus. Wada. 2871 (2003).” Electronics Letters 41. T. P. Chiu. Kimerling. Strand and A. Boimovich. R. Duan. Huang. K.” Journal of Lightwave Technology 26. 575– 593 (2004). Noda. R. 50–55 (1998). C. Marom. P. H. [108] Y. Yariv. and L. Little. G. and W. Ryf. Ippen. [105] L. Greene. Foresi. S. Y.107 [101] J. L. E. Paloczi. Zhang. Bernasconi. X. Asano. D. “Tunable multichannel optical ﬁlter based on silicon photonic band gap materials actuation. [106] W. D. “Wavelength selective switching for optical bandwidth management. [110] T.” Journal of Lightwave Technology 21. and L.” Review of Scienti c Instruments 75. “Ultracompact SiSiO2 microring resonator optical channel dropping ﬁlters. Joannopoulos. Simsarian. “Evolution of passive optical component technologies for ﬁberoptic communication systems. S.
F. and ﬂexible wavelength applications in recent photonic networks. “Optical MEMS for photonic switchingcompact and stable optical crossconnect switches for simple.” Optics Express 16.” Applied Physics Letters 90. Englund. Vlasov. Xia. “Optical Bistability and Mirror Conﬁnement Induced by Radiation Pressure. Walther. Yano. 1550–1553 (1983). Eichenﬁeld and O. [113] V. “Highthroughput silicon nanophotonic wavelengthinsensitive switch for onchip optical networks. J. E. Chen. B. Painter. E. and F. Lee. Eggleton. L. S. and T. Stoltz. Sekaric.” Physical Review Letters 51. C. Petroff. Tsuda. Vuckovi´ . J. “Ultrafast nonlinear c optical tuning of photonic crystal cavities.” Nature Photonics 1.” in Conference on Lasers and ElectroOptics/Quantum Electronics and Laser . and Y. fast. S. [122] M. M. L. Pradhan. [119] I.” Optics Express 15. O. “Optical 4x4 hitless silicon router for optical networksonchip (NoC). “Optical MEMS for lightwave communication. “Optomechanics of Strongly Coupled Stacked Monolithic Microdisks. A. and M. [117] N. [116] F. Biberman. W. J. Monat. B. Green. [118] Y. 242–246 (2008). “Micrometrescale silicon electrooptic modulator. H. “Ultracompact high order ring resonator ﬁlters using submicron silicon photonic wires for onchip optical interconnects. G. Lipson. 11934–11941 (2007). 106–114 (2007). P.108 [112] Q. and H. S. Droz. N.” IEEE Journal on Selected Topics in Quantum Electronics 11. Bergman. Rooks. J. and J. Waks. D. Fushman. M. 091118 (2007). Yamagishi. [121] A. Ilchenko and A. Meystre. “Integrated optoﬂuidics: A new river of light. Schmidt. Lipson. and B. Wang. 15915– 15922 (2008).” Nature Photonics 2. [115] C. 325–327 (2005). K. Domachuk. B.” Nature 435. “Optical resonators with whisperinggallery modesPart II: applications. E. Xia. [120] M.” Journal of Lightwave Technology 24. Mccullen. and M. Wu. Vignes. 383–394 (2005). P. Xu. [114] M. Dorsel. and J. P. Matsko. 4433–4454 (2006). 15–32 (2006).” IEEE Journal on Selected Topics in Quantum Electronics 12. Vlasov. Solgaard. D. Ford.
A. 2007). Michael.” Sensors and Actuators A: Physical 48. Lipson. Veijola. O. and Soft Matter Physics 69. Bhiladvala and Z. Williams. 4628–4641 (2006). “Equivalentcircuit model of the a a squeezed gas ﬁlm in a silicon accelerometer.” Vacuum 16. Painter. [124] A. Painter. Blech. [128] Q. [129] J. J.109 Science Conference and Photonic Applications Systems Technologies (Optical Society of America.” Journal of Lubrication Technology 105. “A proposal for highly tunable optical parametric oscillation in silicon microresonators. [133] R. “Nonlinear optical phenomena in silicon waveguides: modeling and applications. . Wang. T. Christian. J. New York. Yin. R. Lin. “Controlling photonic structures using optical forces. 341–346 (2002). 10596– 10610 (2008). [126] G. [131] R. Chen. NY. and O. [130] T. Bao. J. “Optomechanics of deformable optical cavities. Gonarenko. H. 036307 (2004).” Journal of Micromechanics and Microengineering 12. 16604–16644 (2007). [132] M. and G.” Optics Express 15. J. J. and T. 239– 248 (1995). 633–636 (2009). [125] I. Seeds and K. H. Lin. Kuisma. Agrawal.” Physical Review E: Statistical. P.” Nature 462. P.” Journal of Lightwave Technology 24. P. “The theory of oscillatingvane vacuum gauges. Nonlinear. Ryh¨ nen. “Microwave Photonics. 201–205 (2009).” Nature Photonics 3. S. Johnson. and M. 2008). Nonlinear Fiber Optics 4th ed (Academic Press. L. Perahia. and Y. Lahdenper¨ .” Optics Express 16. Wiederhecker. Sun. J. [123] G. Agrawal. “Energy transfer model for squeezeﬁlm air damping in low vacuum. [127] Q. B. “On isothermal squeeze ﬁlms. J. “Effect of ﬂuids on the Q factor and resonance frequency of oscillating micrometer and nanometer scale beams. H. Karrai. C. Yang. Favero and K. 615–620 (1983). 175–178 (1966). J. G.
Zhou. Little. and D. Arcizet. Scotti. L. Duchesne. [138] C. Reichenbach. B. Cappuzzo. R. M. A. Kippenberg. C. J. Del’haye.” Journal of Applied Physics 99. J. T. C. Y. 737–740 (2008). M. Razzari.” Nature Photonics 2.” Nature 450. B. Z. E. and T.” Nature Photonics 2. J. Gomez. “High quality factor resonance at room temperature with nanostrings under high tensile stress. B. Little. and R. Liscidini. Sipe. Pan. G. Hasnain. Verbridge. Yang. 1214–1217. S. “Lowpower continuouswave nonlinear optics in doped silica glass integrated waveguide structures. Craighead. 3776–3783 (2008). Lenz. J. 124304 (2006). J. L. Schliesser. Chu. [139] M. Yoo. Fontaine. and S. J. “Integrated allpass ﬁlters for tunable dispersion and dispersion slope compensation. R. Moss. A. Huang. Chen. . W. J.” Journal of Lightwave Technology 26. [136] M. S. Bellan. Chu. [137] N. L. 180–184 (2008). Wilken. Ferrera. S. Yang. “A nanoelectromechanical tunable laser. R. E. Madsen.110 [134] S. K. M. B. A. E. “Continuously Tunable Optical Buffering at 40 Gb/s for Optical Packet Switching Networks. K. Y. and H. M. J. and C. G. O. Holzwarth. D. “Optical frequency comb generation from a monolithic microresonator. 1623–1625 (1999). Parpia. Morandotti. Z. Bruce. [135] P. E. T.” IEEE Photonics Technology Letters 11.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.