Page 1

Recent advances in transformation optics

Yongmin Liuaand Xiang Zhang*ab

Received 9th May 2012, Accepted 27th June 2012

DOI: 10.1039/c2nr31140b

Within the past a few years, transformation optics has emerged as a new research area, since it provides

a general methodology and design tool for manipulating electromagnetic waves in a prescribed manner.

Using transformation optics, researchers have demonstrated a host of striking phenomena and devices;

many of which were only thought possible in science fiction. In this paper, we review the most recent

advances in transformation optics. We focus on the theory, design, fabrication and characterization of

transformation devices such as the carpet cloak, ‘‘Janus’’ lens and plasmonic cloak at optical

frequencies, which allow routing light at the nanoscale. We also provide an outlook of the challenges

and future directions in this fascinating area of transformation optics.

1. Introduction

Nobody would disagree that the better understanding, manipu-

lation and application of light, or electromagnetic waves in a

more general respect, play a crucial role in advancing science and

technology. The underlying driving force is the long-standing

interest and attention of human beings concerning novel elec-

tromagneticphenomenaand devices. Without persistent

pursuits, it is impossible to develop a more efficient and direc-

tional radar antenna, a brighter light source, or an instrument

with higher imaging resolution. One of the central aims of these

devices is to control and direct electromagnetic fields. For

instance, by optimizing the curvature of glass lenses in a micro-

scope, we intend to focus light to a geometrical point with less

aberration so that the imaging resolution could be improved.

Alternatively, the technique of gradient index (GRIN) optics has

been applied to design lenses by shaping the spatial distribution

of the refractive index of a material rather than the interface of

lenses. The resulting lenses can be flat and avoid the typical

aberrations of traditional lenses.

Yongmin Liu

Yongmin Liu received his Ph.D.

degree from the University of

California, Berkeley in 2009,

under the supervision of Prof.

Xiang Zhang. Currently he is a

postdoctoral researcher in the

same group. Dr. Liu will join the

faculty of Northeastern Univer-

sity in August 2012, with a joint

appointment in the departments

of Electrical & Computer Engi-

neering and

Industrial Engineering.

Liu’s research interests include

nanoscale materials and engi-

neering, nano photonics, nano

quantum optics

Mechanical&

Dr.

devices,

nanostructures.

andnonlinear andof metallic

Xiang Zhang

Xiang Zhang received his Ph.D.

degree from the University of

California, Berkeley in 1996. He

is Ernest S. Kuh Endowed Chair

Professor at UC Berkeley and

the Director of NSF Nano-scale

Science and Engineering Center.

He is also a Faculty Scientist at

Lawrence Berkeley

Laboratory. Prof. Zhang is an

elected member

Academy of

(NAE) and Fellow of four

scientific societies:

Association for the Advance-

ment of Science

National

of National

Engineering

American

(AAAS),

American Physical Society (APS), Optical Society of America

(OSA), and the International Society of Optical Engineering

(SPIE). His research interests are nano-scale science and tech-

nology, materials physics, photonics and bio-technologies.

aNSF Nanoscale Science and Engineering Center (NSEC), 3112

Etcheverry Hall, University of California, Berkeley, CA 94720, USA.

E-mail: xiang@berkeley.edu

bMaterials Science Division, Lawrence Berkeley National Laboratory, 1

Cyclotron Road, Berkeley, CA 94720, USA

This journal is ª The Royal Society of Chemistry 2012Nanoscale, 2012, 4, 5277–5292 | 5277

Dynamic Article LinksC

<

Nanoscale

Cite this: Nanoscale, 2012, 4, 5277

www.rsc.org/nanoscale

FEATURE ARTICLE

Page 2

In fact, it has been long known that a spatially changing

refractive index modifies light propagation characteristics. Based

on the early work of the ancient Greek mathematician Hero of

Alexandria (10–70 AD) and the Arab scientist Ibn al-Haytham

(965–1040 AD), Pierre de Fermat formulated the famous Fer-

mat’s principle to determine how light propagates in materials.1

This principle states that light follows the extremal optical paths

(shortest or longest, although mostly shortest), where the optical

path is measured in terms of the refractive index n integrated

along the light trajectory. If we replace one material with another

one with a different refractive index in the space where light

propagates, the light path will be bent or even curved instead of a

straight line. Many optical phenomena, such as refraction of a

straw at the interface of air and water, and the mirage effect in a

desert due to the air density (refractive index) variation, can be

explained by Fermat’s principle.

Fermat’s principle tells us how light propagates, if we know

the distribution of the refractive index in space. The emerging

field of transformation optics enables us to solve the inverse

problem, that is, how to realize a specific light path by designing

the variation of material properties.2–4Apparently, this is one

significant step moving forward. With transformation optics, we

have the most general and powerful method to realize almost all

kinds of novel optical effects and devices, some of which only

existed in science fiction and myths. Tremendous progress has

been achieved in the field of transformation optics during the

past a few years, thanks to the new electromagnetic theory and

modelling software, state-of-the-art fabrication tools as well as

greatly improved characterization and analysis techniques. In

this review article, we will first outline the general theory of

transformation optics and metamaterials that allow for the

realization of transformation optical designs. Then we will focus

on the most recent advances, both theory and experiment, in

transformation optics at optical frequencies and at the nano-

scale. Finally, the perspective of transformation optics will be

presented.

2. Theory of transformation optics

The fundamental of transformation optics arises from the fact

that Maxwell’s equations, the governing equations for all elec-

tromagnetic effects, are form invariant under coordinate trans-

formations. Assuming no free current densities, in a Cartesian

coordinate system Maxwell’s equations can be written as

?V ? E ¼ ?m$vH=vt

where E (H) is the electric (magnetic) field, and 3 (m) is the electric

permittivity (magnetic permeability) of a medium that can be a

tensor in general. It can be rigorously proved that after applying

a coordinate transformation x0¼ x0(x), Maxwell’s equations

maintain the same format in the transformed coordinate

system,2–4that is,

?V0? E0¼ ?m0$vH0=vt

In eqn (2), the new permittivity tensor 30and permeability tensor

m0in the transformed coordinate system are related to the orig-

inal 3 and m given by5,6

V ? H ¼ 3$vE=vt

(1)

V0? H0¼ 30$vE0=vt

(2)

8

>

>

>

>

>

>

:

<

30¼L3LT

detjLj

m0¼LmLT

detjLj

(3)

where L is the Jacobian matrix with components defined as Lij¼

vx

i/vxj. The Jacobian matrix characterizes the geometrical vari-

ation in the original space x and the transformed space x0. The

corresponding electromagnetic fields in the new coordinate are

given by

(

0

E0¼?LT??1E

H0¼?LT??1H

(4)

Eqn(1)–(4)formthebasisoftransformationoptics.Wecandesign

and manipulate the light trajectory by an arbitrary coordinate

transformation. Consequently, the material properties and field

components need to be rescaled according to the form invariance

ofMaxwell’sequations.Thisguaranteesthephysicalcharacteristic

oflightpropagationtobepreservedatdifferentscales.Infact,such

a correspondence between coordinate transformations and mate-

rials parameters has been noticed for a long time.

Probably the most remarkable transformation optical device is

the invisibility cloak, which can render an object unperceivable

although the object physically exists. One seminal design of such

a cloak was proposed by Sir John Pendry et al.2They considered

the hidden object to be a sphere of radius R1and the cloaking

region to be contained within the annulus R1# r # R2. By

applying a very simple coordinate transformation

8

:

<

r0¼ R1þ rðR2? R1Þ=R2

q0¼ q

f0¼ f

(5)

the initial uniform light rays in the central region (0 # r # R2) are

squeezed into a shell (R1#r0# R2), while the rest of the light rays

(in the region r > R2) are maintained. Waves cannot penetrate

into and hence interact with the core region (0 # r0# R1),

because it is not part of the transformed space. No matter what

object is placed inside the core, it appears to an observer that

nothing exists; that is, the object is concealed or cloaked. Based

on eqn (3), we can calculate the required material properties for

the cloaking device. In the region of r0# R1, 30and m0can

take any values and do not cause any scattering. In the region of

R1# r0# R2,

8

>

>

>

Finally, for r0$ R2the properties of materials are unchanged.

Under the short wavelength limit (R1, R2[ l), the ray tracing

results confirm the performance of the invisibility cloak as shown

in Fig. 1(a). The rays, which represent the Poynting vector or

energy flow, are numerically obtained by integration of a set of

Hamilton’s equations taking into account the anisotropic,

>

>

>

>

>

>

>

:

>

<

>

>

>

>

>

>

>

>

>

3

0

r0 ¼ m

0

r0 ¼

R2

R2? R1

?r0? R1

R2

R2? R1

R2

R2? R1

r0

?2

3

0

q0 ¼ m

0

q0 ¼

3

0

f0 ¼ m

0

f0 ¼

(6)

5278 | Nanoscale, 2012, 4, 5277–5292This journal is ª The Royal Society of Chemistry 2012

Page 3

inhomogeneous material properties in the compressed region

(R1# r0# R2).2,6Light is smoothly wrapped around the core,

and the propagation characteristic is preserved outside the cloak.

This implies that any object placed in the interior region appears

to be concealed, since there is no diffracted or scattered light in

the presence of the object. Full-wave simulations without

geometric optics approximation also verify the cloaking effect.7

It is worth mentioning that different approaches have been

proposed to realize invisibility cloaks. One example is the

conformal mapping technique to design the refractive index

profile that guides light around an object.8,9We can introduce a

new coordinate w described by an analytic function w(z) that

does not depend on z*, where a complex number z ¼ x + i ? y is

used to describe the spatial coordinate in a two-dimensional (2D)

plane and z* stands for the conjugate of z. Such a function

defines a conformal mapping that preserves the angles between

the coordinate lines. For a gradually varying refractive index

profile, both the electricand magnetic fields satisfy the Helmholtz

equation. In the new coordinate w, the Helmholtz equation has

the same format with a transformed refractive index profile that

is related to the original one as n0¼ n$

Helmholtz equation in the coordinates w as the Schr€ odinger

equation of a quantum particle in the Kepler potential, Leon-

hardt designs a dielectric invisibility cloak with refractive index

ranging from 0 to about 36 (Fig. 1(c)). Different from the

transformation optics approach, the conformal mapping tech-

nique is strictly two-dimensional. However, the conformal

mapping idea can be extended to non-Euclidean geometry to

realize three-dimensional (3D) cloaks, and eliminate the extreme

values of materials parameters that often appearin the method of

transformation optics.10The conformal mapping method has

been used for designing a variety of devices in addition to the

invisibility cloak.11–15Moreover, it has been shown that

conformal mapping performs well even in the regime beyond

geometrical optics.16Another type of approximate invisibility

cloaking is a core–shell structure. It has been shown that a

negative-permittivity shell can significantly reduce the scattering

cross-section of a small positive-permittivity core in the quasi-

static limit.17,18By exploiting the frequency dispersion of metals

and their inherent negative polarizability, it is shown that

????

dw

dz

????

?1

. By interpreting the

covering a dielectric or conducting object of a certain size with

multilayered metallic shells may reduce the ‘‘visibility’’ of the

object by several orders of magnitude simultaneously at multiple

frequencies.19

Meanwhile,researchers

exploring the interesting physics associated with invisibility

cloaks,20–24or trying to detect an invisibility cloak.25,26

The invisibility cloak has triggered widespread interest in

transformation optics. Many other novel effects and devices,

such as illusion optics,27–30optical black holes,31–34beam shifters

and rotators,35–37lossless waveguide bends38–41as well as various

lenses42–50have been proposed. In particular, combining the

concept of complementary medium51with transformation optics,

Yang et al. proposed a superscatter which can enhance the

electromagnetic wave scattering cross section, so that it appears

as a scatter with a larger dimension.52Subsequently, Chan’s

group theoretically conceived and numerically demonstrated a

general concept of illusion optics: making an arbitrary object

appear like another object with a completely different shape and

material constituent.27Cloaking can be considered as the crea-

tion of an illusion in free space. The principle behind illusion

optics is not light bending but rather the cancellation and

restoration of the optical path of light by using negative-index

materials. The key of an illusion device lies in two distinct pieces

of materials, that is, a complementary medium and a restoring

medium. The complementary medium annihilates the adjacent

space and cancels any light scattering from an object itself. Then

the restoring medium recovers the cancelled space with a new

illusion space that embraces another object chosen for the illu-

sion. Numerical simulations confirm the performance of the

illusion device, which transforms the field distribution scattered

from a dielectric spoon into the scattering pattern from a metallic

cup. More interestingly, the illusion device can work at a distance

from the object. It is shown that this ‘‘remote’’ feature enables the

opening of a virtual aperture in a wall so that one can peep

through the wall. Lai et al. also numerically demonstrate a

remote invisibility cloak that can cloak an object at a certain

distance outside the cloaking shell rather than encircled by the

cloaking shell.53

Unlikeprevious

devices,2–9the constitutive parameters of illusion devices do not

need a complex spatial distribution. However, materials with a

negative refractive index are required in the design, which are not

obtainable in nature.

havebeen actively

light-bendingcloaking

3.

optical designs

Metamaterials for realizing transformation

Although transformation optics provides the most general means

to design exotic optical effects and elements, the experimental

realization of them is far from trivial. As shown in eqn (3), both

electric permittivity and magnetic permeability need to be

spatially and independently tailored. Moreover, the resulting

material properties are anisotropic in general, and may require

unusual values (negative, zero or infinity). We are limited in

natural materials to fulfil such demands. For example, natural

materials only show magnetism (m/m0 s 1) up to terahertz

frequencies. Fortunately, the emerging field of metamaterials

offers an entirely new route to design material properties at will,

so that the transformation optical design could be experimentally

realized.54–61Different from natural materials, the physical

Fig. 1

from ref. 2 with permission. The rays, representing the Poynting vector,

divert within the annulus of the cloak region (R1 < r < R2), while

emerging on the far side without any scattering and distortion. (b) Ray

tracing results for an invisibility cloak in a three-dimensional view,

reprinted from ref. 2 with permission. (c) Ray propagation in the

dielectric invisibility device, reprinted from ref. 8 with permission. The

light rays (shown in yellow) smoothly flow around the interior cloak

region (shown in black). The brightness of the green background indi-

cates the refractive index profile taken from the Kepler profile.

(a) A schematic of a cloak in a two-dimensional view, reprinted

This journal is ª The Royal Society of Chemistry 2012 Nanoscale, 2012, 4, 5277–5292 | 5279

Page 4

properties of metamaterials are not primarily dependent on the

chemical constituents, but rather on the internal, specific struc-

tures of the building blocks of metamaterials. These building

blocks function as artificial ‘‘atoms’’ and ‘‘molecules’’, in analogy

to those in natural materials. Through regulated interactions

with electromagnetic waves, they can produce extraordinary

properties that are difficult or impossible to find in naturally

occurring or chemically synthesized materials.

Metamaterials consist of periodically or randomly distributed

artificial structures, whose size and spacing are much smaller

than the wavelength of electromagnetic waves. As a result, the

microscopic detail of individual structures cannot be sensed by

electromagnetic waves. What matters is the average result of the

collective response of the whole assembly. In other words, we can

homogenize such a collection of inhomogeneous objects and

define effective material properties at the macroscopic level. This

is effective media approximation, which has been well known.62

The most attractive aspect of metamaterials, however, is that the

material properties can be controlled by properly engineering the

structures. For instance, metallic wire arrays63and metallic split-

ring structures64can produce effective 3 and m, respectively, with

tunable values ranging from positive to negative within a certain

wavelength range. By combining the two basic structures with

simultaneously negative 3 and m, we can even create materials

possessing a negative refractive index that enable negative

refraction65–69and perfect imaging.70–74Furthermore, meta-

materials allow us to achieve unusual anisotropy75–77and

chirality.78–80We refer readers to recent review papers and books

for more insights in the field of metamaterials.54–61

The complete control over electric permittivity and magnetic

permeability offered by metamaterials turns transformation

optical design into reality. In 2006, Smith’s group demonstrated

the first invisibility cloak in the microwave region.81To mitigate

the fabrication and measurement challenges, a 2D cylindrical

cloak instead of a 3D spherical one was implemented. Since the

electric field is polarized along the z axis of the cylindrical

coordinate, in the transformed 30and m0tensors only 3

are relevant. After a further renormalization, the reduced

material parameters are

?

0

z, m

0

rand m

0

q

3

0

z¼

R2

R2? R1

?2

;m

0

r¼

?r0? R2

r0

?2

;m

0

q¼ 1(7)

where the interior and exterior radium of the cloaking device is

R1and R2, respectively. The advantage of using reduced material

properties is that only one parameter (m

the other two are constant throughout the structure. This

parameter set is realized in a metamaterial structure consisting of

split-ringresonators withcarefully

(Fig. 2(a)). In the experiment, a field-sensing antenna is used to

record the field amplitude and phase inside the cloak and in the

surrounding free-space region. The experimental results show

that the cloak can significantly decrease scattering from the

hidden object and also reduce its shadow. From Fig. 2(b) and (c),

one can clearly see that electromagnetic waves smoothly flow

around the cloak, and propagate to the far side with only a

slightly perturbed phase front,which is mainly dueto the reduced

parameter implementation. In comparison, a bare Cu cylinder

without the cloak produces much stronger scattering in both the

forward and backward directions.

0

r) spatially varies while

designed geometries

4.

optical wavelengths

Broadband transformation optical design at

The pioneering work on transformation optics in 2006 (ref. 2, 3

and 6) stimulated the global attention of researchers in different

disciplines. Ever since then, tremendous effort has been devoted

to the field of transformation optics. Considering the great

application potential, one prime direction of transformation

optics is to implement designs working in the optical regime.82

However, most transformation optical devices rely on meta-

materials, in which the building blocks are normally much

smaller than the wavelength of interest. This indicates that the

feature size of the device should be precisely controlled at the

scale of a few hundred or even below one hundred nanometers.

More importantly, metamaterials are usually resonant structures

with narrow operation bandwidth and high loss. These two

factors impose severe challenges on the implementation of

transformation optical devices with broad bandwidth and low

loss at near-infrared and visible frequencies. New designs and

creative fabrication techniques are imperative to tackle the

challenges.

In the following, we will concentrate on the discussion of the

carpet cloak introduced by Jensen Li and John Pendry,83

although other designs, such as one-dimensional (1D) cloaks84,85

Fig. 2

the cloak. The split-ring resonator of layer 1 (inner) and layer 10 (outer) are shown in the transparent square insets. (b) Simulated and (c) experimentally

mapped field patterns of the cloak. Reprinted from ref. 81 with permission.

(a) An image of a 2D microwave cloak made of split-ring resonators. The background plots the values of the prescribed material properties for

5280 | Nanoscale, 2012, 4, 5277–5292 This journal is ª The Royal Society of Chemistry 2012

Page 5

and cloaks using non-Euclidean geometries,10can also operate

over a relatively wide range of wavelengths. Different from the

original complete cloak that essentially crushes the object to a

point and works for arbitrary incident angles, the carpet cloak

crushes the object to a sheet and the incident angle is limited

within the half space of a 2D plane. However, the carpet cloak

does not require extreme values for the transformed material

properties. Moreover, by applying the quasi-conformal mapping

technique, the anisotropy of the cloak can be significantly

minimized. Consequently, only isotropic dielectrics are needed to

construct the carpet cloak (Fig. 3(a)), implying the device could

be broadband and practically scalable to operate in the optical

regime. Full-wave simulations confirm that the carpet cloak

successfully imitates a flat reflecting surface. As shown in

Fig. 3(b), the light reflected from a curved reflecting surface, on

top of which is covered with the carpet cloak, well maintains the

flat wavefront without any distortion. It seems that the light is

reflected by a flat ground plane. Therefore, it renders an object

placed underneath the curved bump invisible. In contrast, if the

cloak is absent, the incident beam is deflected and split into two

different angles (Fig. 3(c)).

Soon after the demonstration of a microwave carpet cloak

based on non-resonant metallic metamaterials,86three groups

independently realized the carpet cloak in the near-infrared

region.87–89Interestingly, all of them utilized the same dielectric

platform (silicon-on-insulator (SOI) wafer) to achieve the broad-

band and low-loss carpet cloak, although the configurations are

different. In the design of Zhang’s group,87the carpet cloaking

device consists of two parts (Fig. 4(a) and (b)): a triangular

region with a uniform hole pattern that acts as a background

medium with a constant effective index (1.58), and a rectangular

region with varying hole densities to realize the spatial index

profile similar to Fig. 3(a). The holes with a constant diameter

(110 nm) were made through the Si layer by focused ion beam

(FIB) milling. Under the effective medium approximation, the

desired spatial index profile can be achieved by controlling the

density of holes through the relation 3eff¼ 3airrair+ 3SirSi, where

r is the volumetric fraction and 3 is the effective dielectric

constant of each medium. In addition, two gratings were fabri-

cated in order to couple light into and out of the Si slab wave-

guide. Finally, directional deposition of 100 nm gold was carried

out using electron beam evaporation to create the reflecting

surface. In the experiments, the authors characterize the reflected

beam profile of a Gaussian beam in three scenarios: (1) a flat

surface without a cloak, (2) a curved surface with a cloak and (3)

a curved surface without a cloak. It is observed that in both case

(1) and (2), the reflected beam preserves the Gaussian profile,

similar to the incident waves. In a sharp contrast, the light

Fig. 3

(refractive index n). The grey lines represent the transformed grid after

the quasi-conformal mapping. All of the square cells in the original

Cartesian coordinate are transformed to nearly squares of a constant

aspect ratio after the quasi-conformal mapping. Consequently, the

anisotropy of the material property is minimized to a negligible degree.

Comparing with the background (SiO2with n ¼ 1.45), the resulting

refractive index is higher in the region above the curved bump, while it is

lower at the two shoulders of the bump. (b) The electric field pattern for a

Gaussian beam launched at 45?towards the ground plane from the left,

with the spatial index distribution given in (a). (c) The electric field

pattern when only the curved bump is present without the cloak. The

wavelength is 500 nm for simulation in (b) and (c).

(a) The colour maps show the transformed material properties

Fig. 4

of a near-infrared carpet cloak, which is realized by milling holes with

different densities in a SOI wafer. Reprinted from ref. 87 with permission.

In the schematic figure (a), the rectangular cloak region marked as C1has

a varying index profile given by the transformation design, and the

triangular region marked as C2has a uniform hole pattern, serving as a

background medium with a constant effective index of 1.58. (c) and (d)

SEM images of another near-infrared cloaking device by etching silicon

posts in a SOI wafer. Light is coupled into the device via an input

waveguide and reflected by the Bragg mirror towards the x–z plane.

Reprinted from ref. 88 with permission. (e) Schematic and (f) cross-

sectional SEM image of a 3D carpet-cloak structure working in the near-

infrared region, reprinted from ref. 90 with permission. The 3D cone of

light in (e) corresponds to the NA ¼ 0.5 microscope lens.

(a) Schematic and (b) scanning electron microscope (SEM) image

This journal is ª The Royal Society of Chemistry 2012 Nanoscale, 2012, 4, 5277–5292 | 5281

Page 6

reflected from the uncloaked bump (case (3)) shows three distinct

spots at the output grating due to the scattering of the bump.

These results unambiguously verify the performance of the

carpet cloak. Furthermore, since the device is composed of

dielectric materials rather on resonant elements, it is expected to

operate over broad wavelengths. Indeed, for wavelengths

ranging from 1400 to 1800 nm, the reflected beam from the

cloaked curve surface shows a single peak at the output grating.

In comparison, the carpet cloak demonstrated by Lipson’s

group is a complementary structure.88Instead of milling holes in

the SOI wafer, they etched silicon posts of subwavelength 50 nm

in diameter with spatially varying density and cladded the device

with SiO2medium (Fig. 4(c) and (d)). In addition, the reflective

surface is composed of a dielectric distributed Bragg reflector

(DBR) with a deformation that covers the cloaked region rather

than a metallic film. It is experimentally observed that the output

of the light propagating through the cloak and incident on the

curved DBR mirror resembles the image from a flat mirror

without any distortion. In another similar work, light propaga-

tion inside a silicon-nanorod-based carpet cloak is imaged by

near-field scanning optical microscopy (NSOM), providing a

direct visualization of the cloaking effect.89

The aforementioned work represents a major step towards

general transformation optics at optical frequencies. They also

show potentials to realize a variety of transformation optical

devices in on-chip silicon photonic footprints. However, the

demonstrated carpet cloaks were essentially based on a 2D

waveguide configuration, implying that the cloaking effect only

works in the plane. In other words, the devices are visible in the

third dimension. Tolga Egin and his colleagues extrude the third

dimension of the original carpet cloak design, rendering the

cloaking to work in a 3D setting for reasonably large viewing

angles (Fig. 4(e)).90They implement a design based on tailored,

dielectric face-centered-cubic (FCC) wood-file photonic crystals.

The diamond-symmetry woodpile geometry is chosen for its

nearly isotropic optical properties. The technique of direct laser

writing via multiphoton polymerization of a negative photoresist

was used to fabricate the 3D photonic crystals. By properly

controlling the position and intensity of the writing laser beam,

an arbitrary three-dimensional connected pattern, either periodic

or non-periodic, can be created. Fig. 4(f) shows the interior of the

fabricated carpet cloak after FIB milling. The background is a

homogeneous woodpile structure. Within the cloak region, the

local effective refractive index is controlled by the volume filling

ratio of the polymerand air void. In the optical measurement, the

samples are illuminated by unpolarized light from an incandes-

cent lamp. The carpet plane is imagined through the glass

substrate on an image plane. A single reflective Cassegrain lens

with numerical aperture NA ¼ 0.5 is used to avoid chromatic

aberrations. This NA corresponds to a 3D illumination and full

viewing angle of about 60?. A multimode optical fibre is scanned

across the image plane in order to measure the spatial and

spectral dependence. The light emerging from the other end of

the fibre is collimated and sent into a home-made Fourier

transform spectrometer. In both bright-field and dark-field

optical spectroscopy measurements, it is observed that scattered

light from the cloaking device is drastically suppressed over a

broad wavelength spectrum (1.5–2.6 mm) in comparison with the

uncloaked bump,whichis consistentwith ray-tracing

calculations.91Interestingly, this result implies that the effective

medium approximation may even work as approaching the

Wood anomaly wavelength, since the lattice constant (0.8 mm) of

the photonic crystals is already comparable with the operation

wavelength. It is theoretically shown that transformation optical

devices can be achieved in the photonic crystal platform, by

either searching different types of constant frequency contours to

approximate a specific effective medium profile or manipulating

Bloch waves in curved and gradient photonic crystals.92–94

Recently, Fischer et al. reduced the lattice constant of the

previous photonic crystal structure by a factor of more than 2,

and successfully realized a 3D carpet cloak for unpolarized light

at visible wavelengths.95Inspired by stimulated-emission-deple-

tion (STED) fluorescence microscopy, the authors significantly

improved the lithography resolution of direct laser writing.96By

overlapping a femtosecond excitation beam spot and a contin-

uous-wave depletion beam, the effective exposure volume for the

photoresist (0.25 wt% 7-diethylamino-3-thenoylcoumarin in

pentaerythritol tetraacrylate) can be greatly reduced. Conse-

quently, the lithography resolution could overcome the diffrac-

tion-limit. Fig. 5(b) shows SEM images of the cross-section of the

homogeneous woodpile photonic crystal (the reference sample)

Fig. 5

and 3D carpet cloak (bottom) structures working at visible wavelengths,

which are fabricated on a glass substrate and coated with 100 nm gold.

The scale bar corresponds to 10 mm. (b) SEM images showing the cross-

section of the reference and cloak after FIB cut. The scale bar corre-

sponds to 2 mm. (a) and (b) are reprinted from ref. 95 with permission. (c)

A schematicof the cloakdeviceimplemented in a SiN waveguide ona low

index nanoporous silicon oxide substrate. The SiN layer and the nano-

porous oxide layer are 300 nm and 5–10 mm thick, respectively. The

diameterof holes variesin sizefrom65 to 20 nm.The insetshowsanSEM

image of the low-index nanoporous silicon oxide substrate. (d) An atomic

force microscope (AFM) image of the hole pattern as transferred to the

electron beam resist after development. (e) An SEM image of the fabri-

cated carpet cloak device, consisting of roughly 3000 holes. (c)–(e) are

reprinted from ref. 98 with permission.

(a) A false-coloured SEM image of the polymer reference (top)

5282 | Nanoscale, 2012, 4, 5277–5292This journal is ª The Royal Society of Chemistry 2012

Page 7

as well as of the carpet cloak device. The lattice constant of the

photonic crystal is 350 nm. Reflection optical measurements at

different visible wavelengths and incident angles, along with the

interferometric phase measurement97strongly confirm the 3D

polarization-independent cloaking effect in the visible region.

A carpet cloak in the waveguide geometry for visible light has

also been demonstrated,98with two important modifications

compared with the Si carpet cloak working at the near-infrared

frequency.87First, a silicon nitride (SiN) slab is used as the

footprint, since silicon becomes lossy due to absorption at visible

wavelengths. Second, nanoporous silicon oxide with very low

index (n < 1.25) is developed for the substrate, which increases

the available index modulation and enables the realization of

transformation optics for guided visible light in the SiN slab. By

repeating electrochemical etching and oxidation, a solid silicon

layer is slowly consumed, leaving solid filling fractions as low as

15%. Once the desired porosity is reached, the entire silicon

network is converted to a porous silicon oxide medium by

oxidizing at hightemperature (800?C). The poresize ranges from

2 to 20 nm (shown in the inset of Fig. 5(c)), and the surface

roughness is less than 3 nm rms. A 300 nm SiN slab waveguide is

deposited on the low index substrate using plasma-enhanced

chemical vapour deposition (PECVD). Finally, a two-step

pattern transfer process is applied to fabricate the hole pattern

with a fixed hexagonal lattice constant (130 nm) but varied hole

sizes (20–65 nm). In the optical measurement scheme similar to

ref. 87, it is shown that at three different wavelengths (480, 520,

and 700 nm), the reflection from the uncloaked bump produces a

clear perturbation in the wavefront in comparison with the

reflection from a flat mirror. The cloaking device, on the other

hand, reconstructs the wavefront and results in a beam profile

identical to the original Gaussian beam reflected from a flat

mirror. This confirms that the designed transformation effec-

tively cloaks the uneven surface throughout the entire visible

spectrum. The platform based on the SiN slab and low-index

porous silicon oxide substrate leads to a general implementation

of optical transformation structures in the visible range.

The aforementioned carpet cloaks in the planar Si or SiN

geometry can be readily extended to other transformation optical

designs. In particular, it is feasible and desirable to combine

transformation optics with on-chip photonics for much broader

functionalities.99–102Traditional integrated photonics collects

individual, discrete optical elements, such as the light source,

waveguide, modulator, detector, etc. on a footprint. Employing

the transformation optics method, Zentgraf et al. have

successfully designed and demonstrated a photonic ‘‘Janus’’

device which simultaneously possesses multiple functions within

one single optical element.99This opens up a new avenue to

achieving a high density of functionalities, effectively scaling

down the size of integrated photonic circuits. As shown in

Fig. 6(a), we can combine a lens and a beam-shifter into the same

device, while working along the horizontal and vertical direction,

respectively. The theoretically designed permittivity profile of the

‘‘Janus’’ device is translated into a pattern of 75 nm air holes with

a spatially varying density (inset of Fig. 6(a)). In the optical

measurements, the excitation of the slab TM waveguide mode,

which passes the lens along the horizontal direction, is performed

using a Gaussian beam with a spot diameter of 13 mm at the input

grating, while a small beam spot with a diameter of 2 mm is used

for the beam-shifter in the vertical direction. From Fig. 6(b), one

can see that the large beam spot is strongly reduced after passing

the ‘‘Janus’’ device along the x-axis. In contrast, if the beam is

propagating through the device along the y-axis, it is shifted at

the output grating from left to right and vice versa (Fig. 6(c)). The

measurement of the device shows that the element works over a

range of 100 nm for a center wavelength of 1.5 mm.

5. Macroscopic transformation optical devices

One ultimate goal of transformation optics is to realize practical

devices at the macroscopic level. For instance, we want to hide

realistic, large objects using invisibility cloaks. The demonstrated

cloaks at microwave and terahertz frequencies are physically

large, and range from centimetres to millimetres.81,86,103–105In

other words, they are in the order of around 100 wavelengths or

less. However, how to observe the cloaking effect with naked eye,

i.e., to cloak a macroscopic object in the visible regime was

thought extremely challenging. As we have discussed, almost all

transformation optical design operating at visible wavelengths

are fabricated by state-of-the-art micro-/nano-manufacturing

techniques, including electron-beam lithography, FIB milling

and direct laser writing, in order to realize the spatially complex

material properties. If we rely on these top-down methods, it is

very difficult and time-consuming to realize macroscopic trans-

formation optical devices for visible light, because we need to

precisely control the feature size at the nanoscale over a large

domain that may be 1000 times larger than the operation

wavelength in all three dimensions.

In early 2011, two groups independently reported the

demonstration of a macroscopic volumetric cloaking device

Fig. 6

a beam-shifter. The inset shows a magnified view of the air holes in the silicon waveguide slab. (b) and (c) represent the optical microscope images with

the intensity distribution at the in-couple and out-couple gratings for the lens and the shifter, respectively.

A photonic ‘‘Janus’’ device for integrated photonics,reprinted from ref. 99 with permission. (a)SEM images of the device consisting of a lens and

This journal is ª The Royal Society of Chemistry 2012 Nanoscale, 2012, 4, 5277–5292 | 5283

Page 8

based on a very similar design and optical characterization

scheme.106,107The cloaking design uses calcite, a natural bire-

fringent crystal,thus eliminating

manufacturing processes and enabling one to hide objects at the

scale of millimetres. The demonstrated cloaking device can be

still regarded as a carpet cloak. However, it is achieved with

spatially homogeneous, anisotropic dielectric materials, in

contrast to the original proposed one with inhomogeneous,

isotropic material properties. Furthermore, it has been pointed

out that the original carpet cloak based on the quasi-conformal

mapping method generally gives rise to a lateral shift of the

scattered wave, which may make the object detectable.108

Fig. 7(a) shows an illustration of the carpet cloak design in ref.

106. A triangular cross-section (blue) in a virtual space is mapped

toaquadrilateralregion(brown).Asmalltriangularregion(grey)

isopenedupwhereinobjectscanbeplacedandrenderedinvisible.

The transformation can be mathematically described by

time-consumingnano-

x0¼ x;

y0¼H2? H1

H2

$y þd ? x$sgnðxÞ

d

$H1;

z0¼ z

(8)

where (x, y, z) and (x0, y0, z0) correspond to the coordinates of the

virtual space and physical space, respectively. Following eqn (3)

and assuming that the virtual space is filled with an isotropic

material property of 3 and m (m ¼ 1), it is straightforward to

obtain reduced material properties of the quadrilateral cloaking

region for transverse-magnetic (TM) polarized light propagating

in the x–y plane,

0

B

@

m

3

0

x?y¼ 3

B

B

B

B

?

H2

H2? H1

H1H2

ðH2? H1Þ2dsgnðxÞ

?2

?

H1H2

2

ðH2? H1Þ2dsgnðxÞ

?

?

2

1 þ

H2

H2? H1

?2?H1

d

?2

1

C

A

C

C

C

C

0

z¼ 1

(9)

Here, 3

permeability element along the z-axis after the transformation.

Eqn (9) indicates that such a triangular cloak only requires

spatially invariant, anisotropic materials, such as calcite and

calomel, which are readily available in macroscopic sizes.

The macroscopic carpet cloak of ref. 106 is realized by gluing

two calcite prisms together with the protruding bottom surface of

the cloak serving as a deformed reflecting mirror (Fig. 7(b)).

Calcite is a uniaxial birefringent crystal whose refractive indices

are about 1.66 and 1.49 for ordinary and extraordinary light,

respectively, at the wavelength of 590 nm. To meet the material

requirement give by eqn (9), the optical axis orientation angle

with respect to the y-axis, as well as other geometrical parame-

ters, need to be carefully designed. The cloak region has a

triangular cross-section formed by thetwo bottom facets, and the

height of the triangle is close to 1.2 mm, which is more than three

orders of magnitude larger than the visible wavelength.

To directly visualize the cloaking effect, a mask with an arrow

pattern is placed in front of a green laser (532 nm wavelength), so

that the emitted laser beam contains the same pattern (Fig. 7(c)).

When light is reflected from the triangular bottom bump, the

distortion of the reflected image tells us whether the cloaking

effect is achieved or not. In addition, a linear polarizer is used to

control the light polarization. The image of the laser beam

reflected by a flat mirror is shown in the top panel of Fig. 7(d),

which is a horizontally flipped arrow pattern. Since the cloak

does not work for TE polarized light, the reflection from the

bottom bump splits the laser beam into two (middle panel). In

contrast, the reflected beam for TM polarization almost

completely conserves the arrow pattern, except for a small dark

stripe in the center due to the imperfection in the alignment of the

two calcite crystals (bottom panel). Other characterizations, such

as different incident angles and white light illuminations, further

verify the performance of the cloak.

0

x–yis the permittivity tensor in the x–y plane and m

0

zis the

Fig. 7

quadrilateral region (brown) with uniform and anisotropic optical properties. The cloaked region is the small triangle area (grey) wherein objects can be

renderedinvisible. (b)A photograph of the triangularcloakwith the geometricalparameters indicatedin the figure.The dimensionof the cloakalong the

z-axisis 2cm.The opticalaxisof the calcitecrystal,representedbyredarrows,is orientedwith anangle of30?relativeto the y-axis.(c)Aschematicof the

experimental setup. The laser goes through a mask with an arrow pattern and then a polarizer. Subsequently, it is reflected by the calcite cloak and

projected on a screen. (d) The pattern of the laser beam reflected from a flat mirror, and from the calcite cloak for TE and TM polarizations are shown in

the top, middle and bottom panel, respectively. The laser beam reflected by the triangular protruding surface for TM polarization resembles the

reflection from a flat surface. (a)–(d) are reprinted from ref. 106 from permission. (e) A schematic diagram of the experimental setup for another calcite

carpetcloakdesignedbythe MIT group.(f) Opticalimages capturedonaCCDcamera.Thethreepanelsfromlefttorightshowthe reflectedimageswith

an uncloaked wedge, a flat mirror on top of the wedge and a cloaked wedge, respectively, at the wavelength of 561 nm. (e) and (f) are reprinted from ref.

107 with permission.

(a) A schematic of the macroscopic carpet cloak design, in which a triangular cross-section (blue) filled with isotropic materials is mapped to a

5284 | Nanoscale, 2012, 4, 5277–5292 This journal is ª The Royal Society of Chemistry 2012

Page 9

In another work conducted by a research group at MIT,107the

cloak is designed in a similar approach and also made of two

pieces of calcite crystals with specific orientations of the optical

axis. The optical characterization scheme also resembles ref. 106.

A hollow transmission pattern reading ‘‘MIT’’ is printed on an

opaque plastic plate. The pattern is then illuminated by a

continuous wave laser with TM polarization. The illumination

condition is carefully chosen such that the light transmitted

through the inverted ‘‘M’’ goes through the cloak device with the

hidden wedge underneath, while the light through the inverted

‘‘IT’’ is directly reflected by the mirror surface. If the cloak works

and hides the wedge object successfully, the CCD camera should

capture an undistorted ‘‘MIT’’ as if there is nothing on top of the

flat mirror. As shown in the sequential panels of Fig. 7(f) from

left to right, when light illuminates the wedge only without the

cloak placed on top of the mirror, the letter ‘‘M’’ in the reflected

laser beam is far away from ‘‘IT’’ and missed in the CCD image.

When there is a flat reflecting plane on top of the wedge, the letter

‘‘M’’ is undistorted, but it is shifted upwards compared with

‘‘IT’’. Only for the cloaking device, does the CCD image show the

correct ‘‘MIT’’ pattern, as if the cloaked wedge does not exist.

Using calcite crystals, a simplified hexagonal cloak which works

for six incident directions has been demonstrated very recently.109

Although these works show the possibility of realizing practical

transformation optical devices without suffering complicated

micro-/nano-fabrication fabrication processes, the demonstrated

cloak is essentially limited in 2D geometry, because light must

propagate in one plane and polarize in one direction. It is

possible to extend such 2D cloaks into truly 3D geometries

working in the visible spectrum.

6. Transformation optics for plasmonics

Transformation optics in principle embraces all forms of elec-

tromagnetic phenomena at all length scales. Although most work

is devoted to manipulating propagating waves in free space,

recently there has been a keen interest in transforming near-field

optical waves, such as surface plasmon polaritons (SPPs).

Surface plasmon polaritons are collective charge oscillations

existing at the interface between a metal and a dielectric.110They

are driven by and coupled with the electric field of external

electromagnetic waves, behaving as propagating or localized

optical surface waves at the metal–dielectric interface. Due to the

tight confinement and strong field enhancement, SPPs are widely

used for various purposes at the subwavelength scale, ranging

from nano optical circuitry,111–114

raphy,116,117and data storage,118,119to biosensing120,121and

photovoltaics.122Such a new research paradigm, called plas-

monics, has become a very active branch in nano optics. Merging

transformation optics with plasmonics is expected to give rise to

a host of fascinating near-field optical phenomena and devices.

SPPs are bound surface waves at metal–dielectric interfaces,

implying that the entire domain, both the metal and dielectric

materials, needs to be transformed if we rigorously follow the

transformation optics approach. In practice, it is extremely

difficult, if not impossible, to spatially modify the metal property

at the deep subwavelength scale. Fortunately, we can overcome

this problem via prudent designs. For example, as pointed out by

Liu et al.123and Huidobro et al.124independently, one cancontrol

microscopy,74,115

lithog-

SPPs by solely modifying the dielectric material based on the

transformation optics technique, since a significant portion of

SPP energy is carried in the dielectric medium at optical

frequencies. Moreover, the transformed dielectric materials can

be isotropic and nonmagnetic if an advanced transformation

technique, such as conformal or quasi-conformal trans-

formation, is performed.

We take the similar geometry of the carpet cloak as an example

to show how the propagation characteristic of SPPs can be

modified and the scattering of SPPs due to surface topology can

be considerably suppressed.123Scattering of SPPs exists when-

ever there is a variation in geometries or material properties.125In

addition to the intrinsic Ohmic losses of metals, scattering can be

a major loss factor that limits the propagation length of SPPs.

Fig. 8(a) shows full-wave simulation at 633 nm wavelength,

where SPPs at the air–silver interface are launched from the left-

hand site and then pass a surface protrusion. One can clearly see

that the protrusion gives rise to forward scattering into free

space. Quantitatively, about 26% of the SPP energy is radiated to

the far field in this scattering process. This is a fairly big loss,

considering that the energy attenuation due to the Ohmic loss is

only about 4% for SPPs propagating the same lateral distance. In

contrast, once we apply the refractive index profile on top of the

metal surface following the coordinate transformation to map a

Fig. 8

which can significantly suppress the scattering of SPPs due to the uneven

metal surface (top panel). Reprinted from ref. 123 with permission. (b)

Powerflowdistribution(colormap) andstreamlinesof a3D carpet cloak,

reprinted from ref. 124 with permission. (c) An SEM image of the

designed plasmonic cloak. The cloak is made of TiO2nano-pillars, as

shown in the inset. (d) and (e) are leakage radiation images of SPPs for a

bare curved Bragg-reflector and a plasmonic cloak, respectively. When

the incident SPPs (yellow arrow) hit a curved Bragg mirror (red curves),

the backreflectedSPPs (green arrows)have different directions due to the

curved shape of the reflector. If the cloak (purple dotted line) is placed in

front of the curved Bragg mirror, the beating pattern in reflection has an

almost flat wavefront, similar to the reflection from a straight Bragg

reflector. (c)–(e) are reprinted from ref. 126 with permission.

(a) A 2D transformation plasmonic structure (bottom panel),

This journal is ª The Royal Society of Chemistry 2012Nanoscale, 2012, 4, 5277–5292 | 5285

Page 10

protruded surface to flat one, the scattering of SPPs is almost

completely eliminated. The surface appears virtually flat for the

SPPs, although physically the surface protrusion exists. Since

solely isotropic and nondispersive materials are used to realize

the transformed dielectric material, one major advantage of the

transformation plasmonic structure is the broadband perfor-

mance. For wavelengths from 850 to 450 nm, scattering loss

increases from 14% to 43% before the transformation. After the

transformed dielectric cladding is applied, strikingly, the scat-

tering loss of the SPPs is below 4.5% over the entire wavelength

region. A 3D carpet cloak designed from transfinite mapping can

also work effectively for bounding SPPs at the uneven metal

surface, although in this case the transformed material is aniso-

tropic.124The color map and the streamlines in Fig. 8(b) repre-

sent the density and the direction of the power flow of SPPs,

respectively. SPPs are guided around the bump and continue

traveling along the air–gold interface with only slight scattering.

Almost at the same time of the publication of ref. 123 and 124,

Quidant’s group experimentally demonstrated the plasmonic

carpet cloak.126The configuration is shown in Fig. 8(c), where a

gold surface is structured with TiO2nano-pillar structures to

realize the required refractive index profile. A curved Bragg-type

reflector, consisting of 15 gold lines periodically separated by half

of the SPP wavelength, is employed to act as the object to be

hidden behind the carpet. Leakage radiation microscopy (LRM)

images map the distribution of the SPPs propagating at the air–

gold interface. In the case of a bare curved Bragg-reflector, the

reflected SPPs are propagating into different directions depend-

ing on their relative angles to the normal of the mirror lines,

leading to a curved wave front (Fig. 8(d)). Conversely, incor-

porating TiO2nano-pillar structures recovers a fringe pattern in

the reflected SPPs with a nearly straight wave front, similar to the

reflection from a flat Bragg-mirror (Fig. 8(e)). The remaining

small lateral modulations are attributed to imperfections in the

manufacturing. Data analysis further quantifies that the wave

front curvature is reduced by a factor of 3.7 in the presence of the

crescent-moon-like TiO2carpet.

Ref. 126 clearly demonstrates how transformation optics can

be applied to mold the flow of SPPs. If following the traditional

approach, we need to place dielectric nanostructures on metals or

structure metal surfaces to realize the transformation plasmonic

devices, similar to the employment of other various plasmonic

elements. However, the abrupt discontinuities in the material

properties or geometries of these elements lead to considerable

scattering of SPPs, which significantly limits the device perfor-

mance. Instead of spatially modifying the refractive index of the

dielectric material, the thickness of a homogeneous dielectric

cladding layercan bevaried to change the effective modeindex of

SPPs. It provides an alternative method to realize transformation

plasmonic devices. Using grey-scale electron-beam lithography

(EBL) to adiabatically tailor the thickness of a thin dielectric (3 ¼

2.19) poly(methyl methacrylate) (PMMA) film adjacent to a

metal surface, Zhang’s group has demonstrated a plasmonic

Luneburg lens to focus SPPs and a plasmonic Eaton lens to bend

SPPs.127Fig. 9(a) shows the SEM image of one fabricated plas-

monic Luneburg lens. A PMMA cone structure on top of a gold

surface can achieve the required index profile of the traditional

Luneburg lens given by nðrÞ ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

2 ? ðr=RÞ2

q

, where R is the

radius of the lens and r is the distance to the center. The 3D full

wave simulation of the plasmonic Luneburg lens is presented in

Fig. 9(b). In analogy to the traditional Luneburg lens, plane-

wave-like SPPs launched from the left-hand side can be focused

to a point on the opposite side of the perimeter of the PMMA

cone base. Furthermore, because the optical properties are

changed gradually rather than abruptly in the geometry, losses

due to scattering can be significantly reduced in comparison with

previously reported plasmonic elements. Fluorescence imaging

and leakage radiation microscopy are applied to characterize the

performance of the plasmonic Luneburg lenses, confirming the

focusing effect over a relative broad wavelength range (Fig. 9(c)).

The approach introduced in ref. 127 has the potential to achieve

low-loss functional plasmonic elements with a standard fabri-

cation technology based on the grey-scale electron-beam

lithography, and could enable more complex 2D plasmonic

elements using transformation optics.

On the basis of conformal transformation, Pendry’s group has

provided an elegant framework for designing plasmonic nano-

structures with remarkable properties over a broadband spec-

trum.128The general strategy is to start with an infinite plasmonic

geometry that naturally shows a broadband spectrum, and then

apply a conformal coordinate transformation that converts the

infinite structure into a finite one while preserving the continuous

spectrum.

One simple example is presented in Fig. 10(a). A point dipole is

located between two semi-infinite slabs of metal. Due to the near-

field components of the dipole radiation, SPPs can be excited and

propagate along the metal surface. Now we apply a conformal

mapping

z0¼ g2/z*(10)

where g is a constant, z ¼ x + i ? y is the complex number

notation and z* represents the conjugate of z. Such a coordinate

transformation translates the infinity points in z to the origin in

Fig. 9

PMMA on top of a gold film. The diameter and height of the lens are

about 13 mm and 200 nm, respectively. (b) The plot of the magnetic field

for SPPs propagating along the z-axis, which is focused to a point on the

perimeter of the plasmonic Luneburg lens. (c) The intensity image

obtained by leakage radiation microscopy for SPPs passing a Luneburg

lens for wavelengths of 770 nm. SPPs are launched from a gold grating

(dashed box) towards the Luneburg lens (dashed circle). Reprinted from

ref. 127 with permission.

(a) An SEM image of a plasmonic Luneburg lens made of

5286 | Nanoscale, 2012, 4, 5277–5292 This journal is ª The Royal Society of Chemistry 2012

Page 11

z0, and translates planes into cylinders. The resulting structure is

two kissing cylinders with the diameters of D1¼ g2/a1and D2¼

g2/a2, respectively. The dipole source very close to the origin in z

is translated to near infinity in z0, giving rise to a uniform electric

field excitation with respect to the kissing cylinders. Assuming

the original dipole has strength D, the electric field at the origin in

the transformed frame is then given by

E0ðz0¼ 0Þ ¼

1

2p30

D

g2

(11)

If the dimension of the kissing cylinders is sufficiently small

compared with the wavelength of interest, the quasi-static

approximation can be applied. In this case, the magnetic field is

decoupled with the electric field, and the dielectric properties of

the transformed geometry (cylinders and the surrounding

medium) are the same as the original ones from which they are

derived. This is a very intriguing property of deep subwavelength

structures after conformal mapping, which closely links the

physics at work in each of the very different geometries.

Although the material property is unchanged, the plasmon mode

behaves rather differently in the two geometries. Before the

transformation, the dipole excites surface plasmon modes on the

metallic slabs that transport the energy out to the infinity without

reflection. After the transformation, the same modes are excited

by a uniform electric field E0(eqn (11)), and propagate to the

origin in an adiabatic manner. Approaching the structure

singularity (touching point), the wavelength of surface plasmons

is shortened and the field is significantly enhanced due to the

geometry folding (Fig. 10(b)). For an ideal lossless metal, surface

plasmons accumulate energy toward the touching point but

never reach it, since the plasmon modes excited in the original

slab never reach infinity in a finite time. In practice, the finite loss

introduces energy dissipation, but the maximum field enhance-

ment can be still over 104times in the vicinity of the touching

point. Such a giant field enhancement will be extremely useful in

a Raman scattering experiment at the single molecule level. The

enhancement decreases due to a larger damping approaching the

surface plasmon frequency, because surface plasmons are

absorbed before having reached the touching point. Neverthe-

less, the simulation shows the enhancement over a broadband

spectrum (Fig. 10(c)).

Applying different conformation transformations, researchers

have explored a number of novel plasmonic geometries, such as

nano crescents,128sharp wedges129and touching spheres,130which

exhibit broadband response and prominent field enhancement at

the geometry singularity. Moreover, detailed studies have been

conducted to elucidate other relevant properties associated with

plasmonic nanostructures. For instance, by taking into account

radiation damping, Aubry et al. have extended the conformal

transformation approach to predict the optical response of the

plasmonic nanostructures beyond the quasi-static limit.131It is

found that the radiative losses can be mapped directly onto the

power dissipated by a fictive absorbing particle in the original

frame. Radiative losses limit the maximum light enhancement

capability but improve its broadband feature. The field

enhancement is shown to decrease with the structure dimension,

while still remaining in the order of 103over the near-infrared

and visible spectra. In addition, an insightful transformation

optics approach has been developed to investigate the influence

of the nonlocal effect on the optical properties of plasmonic

nanostructures.132The light-harvesting performance of a dimer

of touching nanowires is studied by using the hydrodynamical

Drude model, which reveals nonlocal resonances not predicted

by previous local calculations. Based on the hydrodynamical

Drude model, the interplay between radiative and nonlocal

effects is explored in touching nanowires, allowing us to optimize

the geometry formaximizing

enhancement.

Besides metals, graphene, a one-atom-thick planar sheet of

sp2-bonded carbon atoms, is able to support SPPs in the ter-

ahertz and infrared region.133–135Such a SPP wave is tightly

confined to a single graphene layer, with a guided wavelength

much smaller than free space wavelength, whereas its propa-

gation distance could be large. The most important advantage

of graphene for plasmonic applications over noble metals, such

as silver and gold, is the capability to tune the conductivity and

hence the permittivity of graphene by chemical doping or bias

voltage dynamically and locally (Fig. 11(a) and (b)). As a

result, we can realize the desired permittivity profiles across the

graphene layer to achieve many flatland plasmonic devices.136

As an example of transformation plasmonic devices, a ‘‘flat’’

version of a Luneburg lens is presented in Fig. 11(c). With a

special configuration of bias arrangement in the manner of

several concentric rings, we can create, approximately, a

gradient conductivity pattern that provides the required effec-

tive index for the graphene-based Luneburg lens. Specifically,

the conductivity follows the expression si,n¼ si,out{2 ? [(rn+

rn?1)/D]2}?1/2, where si,nand rnrepresent the imaginary part of

the conductivity and the radius of the nthsection, and si,outis

the imaginary part of the conductivity of the background

the absorption and field

Fig. 10

infinitemetallicslabstotwokissing-cylinders.Meanwhile,adipolesource

attheoriginistransformedtoauniformelectricfield.Thesurfaceplasmon

modes initially propagating along the metal surfaces are folded and

squeezedtowardthetouchingpointofkissingcylinders,whilemaintaining

the continuous and broadband spectrum. (b) The amplitude of the x0-

component of the electric field normalized by the uniform excitation field

for silver at u ¼ 0.9usp. The field amplitude around the touching point of

thekissingcylindersisextremelyhigh,althoughthecolorscaleisrestricted

to[?55].(c)Theabsolutevalueofthefieldenhancement,|E0|/E0,alongthe

cylindersurfaceasafunctionoftheangleqandfrequencyforaplanewave

incident normal to the axis of the cylinders. Considerable field enhance-

ment and confinement can be achieved over a broad spectrum range.

Reprinted from ref. 128 with permission.

(a)Anexampleofconformalmappingthattransformstwosemi-

This journal is ª The Royal Society of Chemistry 2012Nanoscale, 2012, 4, 5277–5292 | 5287

Page 12

graphene. The simulation reveals that the SPPs generated from

a point source is evolved into an approximately ‘‘collimated

beam’’ of SPPs on the one-atom-thick graphene, as a conven-

tional 3D Luneburg lens collimates wavefronts generated from

a point source into a 3D beam. The diameter of the ‘‘flat’’

Luneburg lens is about 1.5 mm, which is only one-tenth of the

wavelength. We can also design other subwavelength graphene-

based optical devices, including lenses,137,138nanoribbon plas-

monic waveguides,139,140and surface cloaks,141implying that

graphene provides a versatile platform for electro-optics and

transformation optics at the atom scale. In ref. 141, it is

numerically demonstrated that an atomically thin graphene

monolayer may drastically suppress the scattering of a cylin-

drical object over a moderately broad bandwidth in the ter-

ahertz regime (Fig. 11(d)). In addition, the working frequency

of the surface cloak may be largely tuned by varying the

chemical potential, realizing a tunable and switchable cloaking

device.

7. Future directions

So far, transformation optics has been focusing on the spatial

control of the light path. From a mathematical point of view,

the spatial and temporal evolutions of light share certain

similarities. Therefore, it will be feasible and intriguing to

extend the concept of transformation optics to the temporal

domain. For instance, instead of rendering an object invisible

by a spatial invisibility cloak, we may create a temporal cloak

to hide the occurrence of an event within a well-controlled time

gap as recently proposed by McCall et al.142Such a time gap

can be opened by manipulating the dispersion of materials in

time, so that the front and rear parts of a probe light beam are

accelerated and slowed down, respectively. An event, such as

the incidence of a pump beam, which occurs within the

resulting temporal gap, would not modify the probe beam in

any case. Subsequently, the time gap can be closed by the

reverse manipulation of the dispersion and thus the speed light.

When the restored probe light reaches an observer, it appears

as a continuous, uniform light as if the event has never

occurred. The experimental demonstration of temporal cloak-

ing utilizes time lenses and dispersive media.143A time lens can

change the colour of the probe light beam at different moments

in time, via electro-optic modulation or parametric nonlinear

optical processes such as four-wave mixing with a chirped

pump wave. Then the ‘‘red’’ and ‘‘blue’’ parts of the chirped

probe beam are transferred to the edges of the time window

through an optical fibre due to dispersion. Consequently, a

time gap is generated and any event within this gap that might

produce a temporal or spectral change to the probe beam will

have no effect. Finally, a dispersive-compensating optical fibre

together with another time lens is used to close the time gap.

The result is that neither the occurrence of the event nor the

presence of the time-lenses is perceivable to an observer. The

experimental demonstration in ref. 143 is essentially a 1D

temporal cloak. Future directions may include temporal

cloaking working for arbitrary incident angles, or even full

spatial–temporal transformation optical devices.

The geometric transformation approach can be extended to

other systems beyond electromagnetic waves, as long as the

governing equations for these systems are invariant under

coordinate transformations. Along this direction, significant

effort has been devoted to transforming acoustic waves144–148

elastic waves in thin plates,149,150linear surface liquid waves151,152

and fluid flows.153Even in the quantum mechanics regime, it is

possible to transform matter waves,154–156the wave description of

particles like electrons and neutrons. For instance, Zhang et al.

theoretically designed an invisibility cloak for matter waves

basedon time-invariantcoordinate

Schr€ odinger’s equations.154The cloak is a 3D optical lattice of

laser beams, forming optical standing waves with gradually

varying amplitude along the radial direction. The varying

amplitude of the optical lattice changes the effective mass and

band energy of the incident particles, which cause the stream of

particles or incident matter waves to bend and come out on the

exit side as if no objects were present in the center. The experi-

mental demonstration of 3D quantum wave cloaks will be

extremely challenging. However, a 2D version of such cloaks

may be possible via electric bias to control the effective mass of

electrons on a graphene sheet. By examining a graphene p–n

junction, Cheianov et al. numerically show that a point source of

electron currents in the n-type region radiates electrons to the

interface, where they are negatively refracted into the p-type

transformationsof

Fig. 11

of a free-standing graphene as a function of the chemical potential and

frequency. (c) The simulated phase of SPPs at f ¼ 30 THz along the

graphene-based Luneburglens(D ¼ 1.5 mm, w ¼ 75 nm,L ¼ 1.6 mm). The

black triangle indicates the position of the point source. (a)–(c) are

reprinted from ref. 136 with permission. (d) The scattering width of an

infinitedielectric cylinder with diameterD ¼ l0/5 and relativepermittivity

3d¼ 3.9 under different conditions. The red line represents an ideal

lossless mantle cloak with surface reactance Xs¼ 313 U (red line). Gra-

phene surface cloaks with a chemical potential of 0.51 eV and different

values of momentum relaxation time s are shown in blue, green, and

yellow lines. The momentum relaxation time is inverse to the electron–

phonon scattering rate. The dashed line corresponds to the bare cylinder

without a cloak. Reprinted from ref. 141 with permission.

(a) The real part and (b) the imaginary part of the conductivity

5288 | Nanoscale, 2012, 4, 5277–5292 This journal is ª The Royal Society of Chemistry 2012

Page 13

region and brought to a focus. This is just in analogy to light

focused by a slab of negative index materials.157

The recent progress on nonlinear and tunable metamaterials

promises the further development of transformation optical

structures. Up to now, almost all the implementations of

transformation optics have relied on passive and linear meta-

materials. It has been proposed that using active sources rather

than passive materials could achieve cloaking, similar to the

active control of sound for noise suppression.158,159Although

active-source cloaking has certain advantages in terms of

fabrication and bandwidth, but the technique is very chal-

lenging at optical frequencies. In contrast, nonlinear and

tunable metamaterials may be the ultimate approach for real-

izing active transformation optical devices, such as invisibility

cloaks which can be turned on and off by external fields. Since

the early stage of metamaterial research, nonlinear meta-

materials have attracted continuous attention due to their novel

properties and phenomena.64,160–164The experimental demon-

strations associated with nonlinear metamaterials, including

tunable split-ring resonators,165

tion,166,167negative refraction arising from phase conjugation168

and four-wave mixing169as well as magnetoelastic meta-

materials,170manifest a very bright future towards actively

tunable transformation optical devices. The proof-of-principle

experiments at microwave wavelengths should be feasible. In

the optical region, the major issue of material losses could be

overcomeby incorporating

materials.171–176Very recently, Yang et al. have demonstrated

that a laminar liquid flow in an optofluidic channel exhibits

spatially variable dielectric properties depending on the flow

rate, allowing for chirped focusing of light and distinctive

discrete diffraction.177In addition, it has been shown that

electric or magnetic fields can control the spatial distribution

and orientation of metallic nanostructures suspended in

fluids.178,179These results indicate that optofluidic systems may

provide a new platform for controllable or even reconfigurable

transformation optical devices.

second harmonic genera-

gain mediainto meta-

8. Conclusions

Rooted in electromagnetism, an ancient subject existing over

centuries, transformation optics has opened an unprecedented

avenue towards the ultimate control over light flow at will.

Driven by the rapid development of metamaterials and start-of-

the-art nanofabrication techniques, many remarkable trans-

formation optical devices have been realized in the optical

domain soon after proof-of-concept demonstrations at low

frequencies. Nowadays we can really visualize the invisibility

effect, which had been thought magical for a long time, even

with the naked eye. Many other fascinating aspects of trans-

formation optics and its extensions are rising from the horizon,

such as the temporal control of light waves, tunable and

reconfigurable transformation optical devices and the manipu-

lation of other waves including quantum waves. The ideas in

transformation optics are far from exhausted. As nano-

photonics and nanotechnology are moving forward, we can

make more seemingly impossible things into reality with the

versatile methodology of transformation optics.

Acknowledgements

The authors are grateful for the financial support from the

United States Army Research Office MURI program under

grant number W911NF-09-1-0539 and the NSF Nano-scale

Science and Engineering Center (NSEC) under grant number

CMMI-0751621.

References

1 M. Born and E. Wolf, Principles of Optics, Cambridge University

Press, Cambridge, UK, 1999.

2 J. B. Pendry, D. Schurig and D. R. Smith, Controlling

electromagnetic fields, Science, 2006, 312, 1780–1782.

3 U. Leonhardt and T. G. Philbin, Transformation optics and the

geometry of light, Prog. Opt., 2009, 53, 69–152.

4 H. Y. Chen, C. T. Chan and P. Sheng, Transformation optics and

metamaterials, Nat. Mater., 2010, 9, 387–396.

5 G. W. Milton, M. Briane and J. R. Willis, On cloaking for elasticity

and physical equations with a transformation invariant form, New J.

Phys., 2006, 8, 248.

6 D. Schurig, J. B. Pendry and D. R. Smith, Calculation of material

properties and ray tracing in transformation media, Opt. Express,

2006, 14, 9794–9804.

7 S. A. Cummer, B.-I. Popa, D. Schurig, D. R. Smith and J. B. Pendry,

Full-wave simulations of electromagnetic cloaking structures, Phys.

Rev. E: Stat., Nonlinear, Soft Matter Phys., 2006, 74, 036621.

8 U. Leonhardt, Optical conformal mapping, Science, 2006, 312,

1777–1780.

9 U. Leonhardt, Notes on conformal invisibility devices, New J. Phys.,

2006, 8, 118.

10 U. Leonhardt and T. Tyc, Broadband invisibility by non-euclidean

cloaking, Science, 2009, 323, 110–112.

11 N. I. Landy and W. J. Padilla, Guiding light with conformal

transformations, Opt. Express, 2009, 17, 14872–14879.

12 J. P. Turpin, A. T. Massoud, Z. H. Jiang, P. L. Werner and

D. H. Werner, Conformal mappings to achieve simple material

parameters for transformation optics devices, Opt. Express, 2010,

18, 244–252.

13 B. Vasic, G. Isic, R. Gajic and K. Hingerl, Controlling

electromagnetic fieldswith

metamaterial regime, Opt. Express, 2010, 18, 20321–20333.

14 M. Schmiele, V. S. Varma, C. Rockstuhl and F. Lederer, Designing

optical elements from isotropic materials by using transformation

optics, Phys. Rev. A: At., Mol., Opt. Phys., 2010, 81, 033837.

15 K. Yao and X. Y. Jiang, Designing feasible optical devices via

conformal mapping, J. Opt. Soc. Am. B, 2011, 28, 1037–1042.

16 T. Tyc, H. Y. Chen, C. T. Chan and U. Leonhardt, Non-euclidean

cloaking for light waves, IEEE J. Sel. Top. Quantum Electron.,

2010, 16, 418–426.

17 A. Al? u and N. Engheta, Achieving transparency with plasmonic and

metamaterial coating, Phys. Rev. E: Stat., Nonlinear, Soft Matter

Phys., 2005, 72, 016623.

18 B. Edward, A. Al? u, M. Silveirinha and N. Engheta, Experimental

verification of plasmonic cloaking at microwave frequencies with

metamaterials, Phys. Rev. Lett., 2009, 103, 153901.

19 A. Al? u and N. Engheta, Phys. Rev. Lett., 2008, 100, 113901.

20 H. S. Chen, B. I. Wu, B. L. Zhang and J. A. Kong, Electromagnetic

wave interactions with a metamaterial cloak, Phys. Rev. Lett., 2007,

99, 063903.

21 Z. C. Ruan, M. Yan, C. W. Neff and M. Qiu, Ideal cylindrical cloak:

perfect but sensitive to tiny perturbations, Phys. Rev. Lett., 2007, 99,

113903.

22 M. Yan, Z. C. Ruan and M. Qiu, Cylindrical invisibility cloak with

simplified material parameters is inherently visible, Phys. Rev. Lett.,

2007, 99, 233901.

23 B. L. Zhang, H. S. Chen, B. I. Wu and J. A. Kong, Extraordinary

surface voltage effect in the invisibility cloak with an active device

inside, Phys. Rev. Lett., 2008, 100, 063904.

24 H. Hashemi, B. L. Zhang, J. D. Joannopoulos and S. G. Johnson,

Delay-bandwidth and delay-loss limitations for cloaking of large

objects, Phys. Rev. Lett., 2010, 104, 253903.

gradedphotonic crystalsin

This journal is ª The Royal Society of Chemistry 2012 Nanoscale, 2012, 4, 5277–5292 | 5289

Page 14

25 H. Y. Chen, X. D. Luo, H. R. Ma and C. T. Chan, The anti-cloak,

Opt. Express, 2008, 16, 14603–14608.

26 B. L. Zhang and B. I. Wu, Electromagnetic detection of a perfect

invisibility cloak, Phys. Rev. Lett., 2009, 103, 243901.

27 Y. Lai, J. Ng, H. Y. Chen, D. Z. Han, J. J. Xiao, Z. Q. Zhang and

C. T. Chan, Illusion optics: the optical transformation of an object

into another object, Phys. Rev. Lett., 2009, 102, 253902.

28 W. X. Jiang, H. F. Ma, Q. Cheng and T. J. Cui, Illusion media:

generating virtual objects using realizable metamaterials, Appl.

Phys. Lett., 2010, 96, 121910.

29 X. D. Luo, T. Yang, Y. W. Gu, H. Y. Chen and H. R. Ma, Conceal

an entrance by means of superscatter, Appl. Phys. Lett., 2009, 94,

223513.

30 C. Li, X. Meng, X. Liu, F. Li, G. Y. Fang, H. Y. Chen and

C. T. Chan, Experimental

broadband illusion-optics analogue, Phys. Rev. Lett., 2010, 105,

233906.

31 U. Leonhardt and P. Piwnicki, Optics of nonuniformly moving

media, Phys. Rev. A: At., Mol., Opt. Phys., 1999, 60, 4301–4312.

32 D. A. Genov, S. Zhang and X. Zhang, Mimicking celestial

mechanics in metamaterials, Nat. Phys., 2009, 5, 687–692.

33 E. E. Narimanov and A. V. Kildishev, Optical black hole:

broadband omnidirectional light absorber, Appl. Phys. Lett., 2009,

95, 041106.

34 Q. Cheng, T. J. Cui, W. X. Jiang and B. G. Cai, An omnidirectional

electromagnetic absorber made of metamaterials, New J. Phys.,

2010, 12, 063006.

35 M. Rahm, S. A. Cummer, D. Schurig, J. B. Pendry and

D. R. Smith, Optical design of reflectionless complex media by

finite embedded coordinate transformations, Phys. Rev. Lett.,

2008, 100, 063903.

36 H. Y. Chen and C. T. Chan, Transformation media that rotate

electromagnetic fields, Appl. Phys. Lett., 2007, 90, 241105.

37 H. Y. Chen, B. Hou, S. Y. Chen, X. Y. Ao, W. J. Wen and

C. T. Chan, Design and experimental realization of a broadband

transformation media field rotator at microwave frequencies, Phys.

Rev. Lett., 2009, 102, 183903.

38 D. R. Roberts, M. Rahm, J. B. Pendry and D. R. Smith,

Transformation-optical design of sharp waveguide bends and

corners, Appl. Phys. Lett., 2008, 93, 251111.

39 D. H. Kwon and D. H. Werner, Transformation optical designs for

wave collimators, flat lenses and right-angle bends, New J. Phys.,

2008, 10, 115023.

40 J. T. Huangfu, S. Xi, F. M. Kong, J. J. Zhang, H. S. Chen,

D. X. Wang, B. I. Wu, L. X. Ran and J. A. Kong, Application of

coordinate transformation in bent waveguides, J. Appl. Phys.,

2008, 104, 014502.

41 Z. L. Mei and T. J. Cui, Arbitrary bending of electromagnetic waves

using isotropic materials, J. Appl. Phys., 2009, 105, 104913.

42 D. Schurig, J. B. Pendry and D. R. Smith, Transformation-designed

optical elements, Opt. Express, 2007, 15, 14772–14782.

43 A. V. Kildishev and V. M. Shalaev, Engineering space for light via

transformation optics, Opt. Lett., 2008, 33, 43–45.

44 M. Yan, W. Yan and M. Qiu, Cylindrical superlens by a coordinate

transformation, Phys. Rev. B: Condens. Matter Mater. Phys., 2008,

78, 125113.

45 M. Tsang and D. Psaltis, Magnifying perfect lens and superlens

designed by coordinate transformation, Phys. Rev. B: Condens.

Matter Mater. Phys., 2008, 77, 035122.

46 Y. K. Ma, C. K. Ong, T. Tyc and U. Leonhardt, An omnidirectional

retroreflector based on the transmutation of dielectric singularities,

Nat. Mater., 2009, 8, 639–642.

47 D. A. Roberts, N. Kundtz and D. R. Smith, Optical lens

compression via transformation optics, Opt. Express, 2009, 17,

16535–16542.

48 N. Kundtzand D.R. Smith,

metamaterial lens, Nat. Mater., 2010, 9, 129–132.

49 H. F. Ma and T. J. Cui, Three-dimensional broadband and broad-

angle transformation-optics lens, Nat. Commun., 2010, 1, 124.

50 Z. H. Jiang, M. D. Gregory and D. H. Werner, Experimental

demonstration of a broadband transformation optics lens for

highly directive multibeam emission, Phys. Rev. B: Condens.

Matter Mater. Phys., 2011, 84, 165111.

51 J. B. Pendry and S. A. Ramakrishna, Focusing light using negative

refraction, J. Phys.: Condens. Matter, 2003, 15, 6345–6364.

realizationofa circuit-based

Extreme-anglebroadband

52 T. Yang, H. Y. Chen, X. D. Luo and H. R. Ma, Superscatter:

enhancement of scattering with complementary media, Opt.

Express, 2008, 16, 18545–18550.

53 Y. Lai, H. Y. Chen, Z. Q. Zhang and C. T. Chan, Complementary

media invisibility cloak that cloaks objects at a distance outside

the cloaking shell, Phys. Rev. Lett., 2009, 102, 093901.

54 D. R.Smith,J.B. PendryandM.C.K.Wiltshire,Metamaterialsand

negative refractive index, Science, 2004, 305, 788–792.

55 V.M. Shalaev, Optical negative-index

Photonics, 2007, 1, 41–48.

56 M. Wegenerand S.Linden,

metamaterials, Phys. Today, 2010, 63, 32–36.

57 Y. M. Liu and X. Zhang, Metamaterials: a new frontier of science

and technology, Chem. Soc. Rev., 2011, 40, 2494–2507.

58 C. M. Soukoulis and M. Wegener, Past achievements and future

challenges in the development of three-dimensional photonic

metamaterials, Nat. Photonics, 2011, 5, 523–530.

59 N. Engheta and R. W. Ziolkowski, Electromagnetic Metamaterials:

Physics and Engineering Explorations, Wiley-IEEE Press, 1st edn,

2006.

60 W. S. Cai and V. M. Shalaev, Optical Metamaterials: Fundamentals

and Applications, Springer, New York, 1st edn, 2009.

61 T. J. Cui, D. R. Smith and R. P. Liu, Metamaterials: Theory, Design,

and Applications, Springer, 1st edn, 2009.

62 A. Sihvola, Electromagnetic Mixing Formulas and Applications,

Institution of Electrical Engineers, 1999.

63 J. B. Pendry, A. J. Holden, W. J. Stewart and I. Youngs, Phys. Rev.

Lett., 1996, 76, 4773–4776.

64 J. B. Pendry, A. J. Holden, D. J. Robbins and W. J. Stewart, IEEE

Trans. Microwave Theory Tech., 1999, 47, 2075–2084.

65 D. R. Smith, W. J. Padilla, D. C. Vier, S. C. Nemat-Nasser and

S. Schultz, Phys. Rev. Lett., 2000, 84, 4184–4187.

66 R. A. Shelby, D. R. Smith and S. Schultz, Science, 2001, 292, 77–79.

67 C. G. Parazzoli, R. B. Greegor, K. Li, B. E. C. Koltenbah and

M. Tanielian, Experimental verification and simulation of negative

index of refraction using Snell’s law, Phys. Rev. Lett., 2003, 90,

107401.

68 S. Zhang, W. J. Fan, N. C. Panoiu, K. J. Malloy, R. M. Osgood and

S. R. J. Brueck, Experimental demonstration of near-infrared

negative-index metamaterials, Phys. Rev. Lett., 2005, 95, 137404.

69 J. Valentin, S. Zhang, T. Zentgraf, E. Ulin-Avila, D. A. Genov,

G. Bartal and X. Zhang, Three-dimensional optical metamaterial

with a negative refractive index, Nature, 2008, 455, 376–379.

70 J. B. Pendry, Negative refraction makes a perfect lens, Phys. Rev.

Lett., 2000, 85, 3966–3969.

71 A. N. Lagarkov and V. N. Kissel, Near-perfect imaging in a focusing

system based on a left-handed-material plate, Phys. Rev. Lett., 2004,

92, 077401.

72 N. Fang, H. Lee, C. Sun and X. Zhang, Sub-diffraction-limited

optical imaging with a silver superlens, Science, 2005, 308, 534–537.

73 T. Taubner, D. Korobkin, Y. Urzhumov, G. Shvets and

R. Hillenbrand, Near-field microscopy through a SiC superlens,

Science, 2006, 313, 1595.

74 X. Zhang and Z. W. Liu, Superlenses to overcome the diffraction

limit, Nat. Mater., 2008, 7, 435–441.

75 D. R. Smith, D. Schurig, J. J. Mock, P. Kolinko and P. Rye, Partial

focusing of radiation by a slab of indefinite media, Appl. Phys. Lett.,

2004, 84, 2244–2246.

76 A. J. Hoffman, et al., Negative refraction in semiconductor

metamaterials, Nat. Mater., 2007, 6, 946–950.

77 J. Yao, Z. W. Liu, Y. M. Liu, Y. Wang, C. Sun, G. Bartal,

A. M. Stacy and X. Zhang, Optical negative refraction in bulk

metamaterials of nanowires, Science, 2007, 321, 930.

78 S. Zhang, Y. Park, J. S. Li, X. C. Lu, W. L. Zhang and X. Zhang,

Negative refractive index in chiral metamaterials, Phys. Rev. Lett.,

2009, 102, 023901.

79 E. Plum, J. Zhou, J. Dong, V. A. Fedotov, T. Koschny,

C. M. Soukoulis and N. I. Zheludev, Metamaterial with negative

index due to chirality, Phys. Rev. B: Condens. Matter Mater.

Phys., 2009, 79, 035407.

80 J. K. Gansel, et al., Gold helix photonic metamaterial as broadband

circular polarizer, Science, 2009, 325, 1513–1515.

81 D. Schurig, J. J. Mock, B. J. Justice, S. A. Cummer, J. B. Pendry,

A. F. Starr and D. R. Smith, Metamaterial electromagnetic cloak

at microwave frequencies, Science, 2006, 314, 977–980.

metamaterials,

Nat.

Shapingoptical spacewith

5290 | Nanoscale, 2012, 4, 5277–5292This journal is ª The Royal Society of Chemistry 2012

Page 15

82 W. S. Cai, U. K. Chettiar, A. V. Kildishev and V. M. Shalaev,

Optical cloaking with metamaterials, Nat. Photonics, 2007, 1, 224–

227.

83 J. S. Li and J. B. Pendry, Hiding under the carpet: a new strategy for

cloaking, Phys. Rev. Lett., 2008, 101, 203901.

84 H. Y. Chen and C. T. Chan, Electromagnetic wave manipulation by

layered systems using the transformation media concept, Phys. Rev.

B: Condens. Matter Mater. Phys., 2008, 78, 054204.

85 S. Xi, H. S. Chen, B. I. Wu and J. A. Kong, One-directional perfect

cloak created with homogeneous material, IEEE Microwave and

Wireless Components Letters, 2009, 19, 131–133.

86 R. Liu, C. Ji, J. J. Mock, J. Y. Chin, T. J. Cui and D. R. Smith,

Broadband ground-plane cloak, Science, 2009, 323, 366–369.

87 J. Valentine, J. Li, T. Zentgraf, G. Bartal and X. Zhang, An optical

cloak made of dielectrics, Nat. Mater., 2009, 7, 568–571.

88 L. H. Gabrielli, J. Cardenas, C. B. Poitras and M. Lipson, Silicon

nanostructurecloak operating

Photonics, 2009, 43, 461–463.

89 J. H. Lee, J. Blair, V. A. Tamma, Q. Wu, S. J. Rhee, C. J. Summers

and W. Park, Direct visualization of optical frequency invisibility

cloak based on silicon nanorod array, Opt. Express, 2009, 17,

12922–12928.

90 T. Ergin, N. Stenger, P. Brenner, J. B. Pendry and M. Wegener,

Three-dimensionalinvisibility

Science, 2010, 328, 337–339.

91 T. Ergin, J. C. Halimeh, N. Stenger and M. Wegener, Optical

microscopy of 3D carpet cloaks: ray-tracing calculations, Opt.

Express, 2010, 18, 20535–20545.

92 Y. A. Urzhumov and D. R. Smith, Transformation optics with

photonic band gap media, Phys. Rev. Lett., 2010, 105, 163901.

93 Z. X. Liang and S. J. Li, Scaling two-dimensional photonic crystals

for transformation optics, Opt. Express, 2011, 19, 16821–16829.

94 E. Cassan and K.-V. Do, Analytic design of graded photonic

crystals in the metamaterial regime, J. Opt. Soc. Am. B, 2011,

28, 1905.

95 J.Fischer, T.Erginand

polarization-independent visible-frequency carpet invisibility cloak,

Opt. Lett., 2011, 36, 2059–2061.

96 J. Fischer and M. Wegener, Three-dimensional direct laser writing

inspired by stimulated-emission-depletion microscopy, Opt. Mater.

Express, 2011, 1, 614–624.

97 T. Ergin, J. Fischer and M. Wegener, Optical phase cloaking of 700

nm light waves in the far field by a three-dimensional carpet cloak,

Phys. Rev. Lett., 2011, 107, 173901.

98 M. Charghi, C. Gladden, T. Zentgraf, Y. M. Liu, X. B. Yin,

J. Valentine and X. Zhang, A carpet cloak for visible light, Nano

Lett., 2011, 11, 2825–2828.

99 T. Zentgraf,J. Valentine, N. Tapia, J. S. Li and X. Zhang,An optical

"Janus"devicefor integratedphotonics,Adv. Mater.,2010,22,2561–

2564.

100 A. Di Falco, S. C. Kehr and U. Leonhardt, Luneburg lens in silicon

photonics, Opt. Express, 2011, 19, 5156–5162.

101 J. Hunt, T. Tyler, S. Dhar, Y. J. Tsai, P. Bowen, S. Larouche,

N. M. Jokerst and D. R. Smith, Planar, flattened Luneburg lens at

infrared wavelengths, Opt. Express, 2012, 20, 1706–1713.

102 L. H. Gabrielli and M. Lipson, Integrated Luneburg lens via ultra-

strong index gradient on silicon, Opt. Express, 2011, 19, 20122–

20127.

103 H. F. Ma and T. J. Cui, Three-dimensional broadbandground-plane

cloak made of metamaterials, Nat. Commun., 2010, 1, 21.

104 F. Zhou, Y. J. Bao, W. Cao, C. T. Stuart, J. Q. Gu, W. L. Zhang and

C. Sun, Hiding realistic object using a broadband terahertz

invisibility cloak, Sci. Rep., 2011, 1, 78.

105 D. H. Liang, J. Q. Gu, J. G. Han, Y. M. Yang, S. Zhang and

W. L. Zhang, Robust large dimension terahertz cloaking, Adv.

Mater., 2012, 24, 916–921.

106 X.Z. Chen,Y. Luo, J. J. Zhang,K. Jiang, J. B. PendryandS. Zhang,

Macroscopic invisibility cloaking of visible light, Nat. Commun.,

2011, 2, 176.

107 B. L. Zhang, Y. Luo, X. G. Liu and G. Barbastathis, Macroscopic

invisibility cloak for visible light, Phys. Rev. Lett., 2011, 106, 033901.

108 B. L. Zhang, T. Chan and B. I. Wu, Lateral shift makes a ground-

plane cloak detectable, Phys. Rev. Lett., 2010, 104, 233903.

109 H. S. Chen and B. Zheng, Broadband polygonal invisibility cloak for

visible light, Sci. Rep., 2012, 2, 255.

atopticalfrequencies,

Nat.

cloak at optical wavelengths,

M. Wegener,Three-dimensional

110 H. Raether, Surface Plasmons: On Smooth and Rough Surfaces and

on Gratings, Springer, Berlin, 1988.

111 W. L. Barnes, A. Dereux and T. W. Ebbesen, Surface plasmon

subwavelength optics, Nature, 2003, 424, 824–830.

112 N. Engheta, Circuits with light at nanoscales: optical nanocircuits

inspired by metamaterials, Science, 2007, 317, 1698–1702.

113 D. K. Gramotnev and S. I. Bozhevolnyi, Plasmonics beyond the

diffraction limit, Nat. Photonics, 2010, 4, 83–91.

114 J. A. Schuller, E. S. Barnard, W. S. Cai, Y. C. Jun, J. S. White and

M. L. Brongersma, Plasmonics for extreme light concentration and

manipulation, Nat. Mater., 2010, 9, 193–204.

115 S. Kawata, Y. Inouye and P. Verma, Plasmonics for near-field nano-

imaging and superlensing, Nat. Photonics, 2009, 3, 388–394.

116 W. Srituravanich, N. Fang, C. Sun, Q. Luo and X. Zhang,

Plasmonic nanolithography, Nano Lett., 2004, 4, 1085–1088.

117 W. Srituravanich, L. Pan, Y. Wang, C. Sun, D. B. Bogy and

X. Zhang, Flying plasmonic lens in the near field for high-speed

nanolithography, Nat. Nanotechnol., 2008, 3, 733–737.

118 W. A. Challener, et al., Heat-assisted magnetic recording by a near-

field transducer with efficient optical energy transfer, Nat. Photonics,

2009, 3, 220–224.

119 B. C. Stipe, et al., Magnetic recording at 1.5 Pb m?2using an

integrated plasmonic antenna, Nat. Photonics, 2010, 4, 484–488.

120 J. N. Anker, W. P. Hall, O. Lyandres, N. C. Shah, J. Zhao and

R. P. Van Duyne, Biosensing with plasmonic nanosensors, Nat.

Mater., 2008, 7, 442–453.

121 N. J. Halas, Plasmonics: an emerging field fostered by Nano Letters,

Nano Lett., 2010, 10, 3816–3822.

122 H. A. Atwater and P. Albert, Nat. Mater., 2010, 9, 205–213.

123 Y. M. Liu, T. Zentgraf, G. Bartal and X. Zhang, Transformational

plasmon optics, Nano Lett., 2010, 10, 1991–1997.

124 P. A. Huidobro, M. L. Nesterov, L. Mart? ın-Moreno and

F. J. Garc? ıa-Vidal, Transformation optics for plasmonics, Nano

Lett., 2010, 10, 1985–1990.

125 R. F. Oulton, D. F. P. Pile, Y. M. Liu and X. Zhang, Scattering of

surface plasmon polaritons at abrupt surface interfaces: implications

for nanoscale cavities, Phys. Rev. B: Condens. Matter Mater. Phys.,

2007, 76, 035408.

126 J. Renger, M. Kadic, G. Dupont, S. S. A? cimovi? c, S. Guenneau,

R. Quidant and S. Enoch, Hidden progress: broadband plasmonic

invisibility, Opt. Express, 2010, 18, 15757–15768.

127 T. Zentgraf, Y. M. Liu, M. H. Mikkelsen, J. Valentine and

X.Zhang, PlamonicLuneburg

Nanotechnol., 2011, 6, 151–155.

128 A. Aubry, D. Y. Lei, A. I. Fern? andez-Dom? ınguez, Y. Sonnefraud,

S. A. Maier and J. B. Pendry, Plasmonic light-harvesting devices

over the whole visible spectrum, Nano Lett., 2010, 10, 2574–2579.

129 Y. Luo, J. B. Pendry and A. Aubry, Surface plasmons and

singularities, Nano Lett., 2010, 10, 4186–4191.

130 A. I. Fern? andez-Dom? ınguez, S. A. Maier and J. B. Pendry,

Collection and concentration of light by touching spheres: a

transformation optics approach, Phys. Rev. Lett., 2010, 105,

266807.

131 A. Aubry, D. Y. Lei, S. A. Maier and J. B. Pendry, Conformal

transformation applied to plasmonics beyond the quasistatic limit,

Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 82, 205109.

132 A. I. Fern? andez-Dom? ınguez, A. Wiener, F. J. Garc? ıa-Vidal,

S. A. Maier and J. B. Pendry, Transformation-optics description

of nonlocal effects in plasmonic nanostructures, Phys. Rev. Lett.,

2012, 108, 106802.

133 M. Jablan, H. Buljan and M. Soljacic, Plasmonics in graphene at

infrared frequencies, Phys. Rev. B: Condens. Matter Mater. Phys.,

2009, 80, 245436.

134 L. Ju,

et al.,Graphene plasmonics

metamaterials, Nat. Nanotechnol., 2011, 6, 630–634.

135 H. G. Yan, X. S. Li, B. Chandra, G. Tulevski, Y. Q. Wu, M. Freitag,

E. J. Zhu, P. Avouris and F. N. Xia, Tunable infrared plasmonic

devices using graphene/insulator stacks, Nat. Nanotechnol., 2012,

7, 330–334.

136 A. Vakil and N. Engheta, Transformation optics using graphene,

Science, 2011, 332, 1291–1294.

137 H. J. Xu, W. B. Ju, Y. Jiang and Z. G. Dong, Beam-scanning planar

lens based on graphene, Appl. Phys. Lett., 2012, 100, 051903.

138 A. Vakil and N. Engheta, Fourier optics on graphene, Phys. Rev. B:

Condens. Matter Mater. Phys., 2012, 85, 075434.

and Eaton lenses,

Nat.

for tunableterahertz

This journal is ª The Royal Society of Chemistry 2012 Nanoscale, 2012, 4, 5277–5292 | 5291

Page 16

139 A. Yu. Nikitin, F. Guinea, F. J. Garc? ıa-Vidal and L. Mart? ın-

Moreno, Edge and waveguide terahertz surface plasmon modes in

graphene microribbons, Phys. Rev. B: Condens. Matter Mater.

Phys., 2011, 84, 161407.

140 J.Christensen,A. Manjavacas,

F. H. L. Koppens and F. J. Garc? ıa de Abajo, Graphene plasmon

waveguidingand hybridization

nanoribbons, ACS Nano, 2012, 6, 431–440.

141 P. Y. Chenand A. Alu, Atomicallythin surfacecloakusing graphene

monolayers, ACS Nano, 2011, 5, 5855–5863.

142 M. W. McCall, A. Favaro, P. Kinsler and A. Boardman, A

spacetime cloak, or a history editor, J. Opt., 2011, 13, 024003.

143 M. Fridman, A.Farsi,Y.

Demonstration of temporal cloaking, Nature, 2012, 481, 62–65.

144 S. A. Cummerand D. Schurig,One pathto acoustic cloaking, New J.

Phys., 2007, 9, 45.

145 H. Chen and C. T. Chan, Acoustic cloaking in three dimensions

using acoustic metamaterials, Appl. Phys. Lett., 2007, 91, 183518.

146 S. A. Cummer, B. I. Popa, D. Schurig, D. R. Smith, J. Pendry,

M. Rahm and A. Starr, Scattering theory derivation of a 3D

acoustic cloaking shell, Phys. Rev. Lett., 2008, 100, 024301.

147 S. Zhang, C. G. Xia and N. Fang, Broadband acoustic cloak for

ultrasound waves, Phys. Rev. Lett., 2011, 106, 024301.

148 X. F. Zhu, B. Liang, W. W. Kan, X. Y. Zou and J. C. Cheng,

Acousticcloaking by a superlens with sing-negative materials,

Phys. Rev. Lett., 2011, 106, 014301.

149 M. Farhat, S. Guenneau and S. Enoch, Ultrabroadband elastic

cloaking in thin plates, Phys. Rev. Lett., 2009, 103, 024301.

150 N. Stenger, M. Wilhelm and M. Wegener, Experiments on elastic

cloaking in thin plates, Phys. Rev. Lett., 2012, 108, 014301.

151 M. Farhat, S. Enoch, S. Guenneau and A. B. Movchan, Broadband

cylindrical acoustic cloak for linear surface waves in a fluid, Phys.

Rev. Lett., 2008, 101, 134501.

152 H. Y. Chen, J. Yang, J. Zi and C. T. Chan, Transformation media

for linear liquid surface waves, Europhys. Lett., 2009, 85, 24004.

153 Y. A. Urzhumov and D. R. Smith, Phys. Rev. Lett., 2011, 107,

074501.

154 S. Zhang, D. A. Genov, C. Sun and X. Zhang, Cloaking of matter

waves, Phys. Rev. Lett., 2008, 100, 123002.

155 A.Greenleaf,Y. Kurylev,M.Lassas andG. Uhlmann,Approximate

quantum cloaking and almost-trapped states, Phys. Rev. Lett., 2008,

101, 220404.

156 D. H. Lin and P. G. Luan, Cloaking of matter waves under the

global Aharonov–Bohm effect, Phys. Rev. A: At., Mol., Opt.

Phys., 2009, 79, 051605.

157 V. V. Cheianov, V. Fal’ko and B. L. Altshuler, The focusing of

electron flow and a Veselago lens in graphene p–n junctions,

Science, 2007, 315, 1252–1255.

158 D. A. B. Miller, On perfect cloaking, Opt. Express, 2006, 14, 12457–

12466.

159 F. G. Vasquez, G. W. Milton and D. Onofrei, Active exterior

cloaking for the 2D Laplace and Helmholtz equations, Phys. Rev.

Lett., 2009, 103, 073901.

160 A.A.Zharov,I.V.ShadrivovandY.S.Kivshar,Nonlinearproperties

of left-handed metamaterials, Phys. Rev. Lett., 2003, 91, 037401.

161 V.M.Agranovich,Y. R.

A. A. Zakhidov, Linear and nonlinear wave propagation in

S.Thongrattanasiri,

in individualand paired

Okawachiand A. L.Gaeta,

Shen, R. H. Baughmanand

negative refraction metamaterials, Phys. Rev. B: Condens. Matter

Mater. Phys., 2004, 69, 165112.

162 Y. M. Liu, G. Bartal, D. A. Genov and X. Zhang, Subwavelength

discrete solitons in nonlinear metamaterials, Phys. Rev. Lett., 2007,

99, 153901.

163 J. B. Pendry, Time reversal and negative refraction, Science, 2008,

322, 71–73.

164 P. Y. Chen, M. Mohamed and A. Alu, Bistable and self-tunable

negative-index metamaterial at optical frequencies, Phys. Rev.

Lett., 2011, 106, 105503.

165 I. V. Shadrivov, S. K. Morrison and Y. S. Kivshar, Tunable split-

ring resonators for nonlinear negative-index metamaterials, Opt.

Express, 2006, 14, 9344–9339.

166 M. W. Klein, C. Enkrich, M. Wegener and S. Linden, Second-

harmonic generation from magnetic metamaterials, Science, 2006,

313, 502–504.

167 A. Rose, D. Huang and D. R. Smith, Controlling the second

harmonic in a phase-matched negative-index metamaterial, Phys.

Rev. Lett., 2011, 107, 063902.

168 A. R. Katko, S. Gu, J. P. Barrett, B. I. Popa, G. Shvets and

S. A. Cummer, Phase conjugation and negative refraction using

nonlinear active metamaterials, Phys. Rev. Lett., 2010, 105,

123905.

169 S. Palomba, S. Zhang, Y. Park, G. Bartal, X. B. Yin and X. Zhang,

Optical negative refraction by four-wave mixing in thin metallic

nanostructures, Nat. Mater., 2012, 11, 34–38.

170 M. Lapine, I. V. Shadrivov, D. A. Powell and Y. S. Kivshar,

Magnetoelastic metamaterials, Nat. Mater., 2011, 11, 30–33.

171 A. Fang, Th. Koschny, M. Wegener and C. M. Soukoulis, Self-

consistent calculation of metamaterials with gain, Phys. Rev. B:

Condens. Matter Mater. Phys., 2009, 79, 241104.

172 S. Wuestner, A. Pusch, K. L. Tsakmakidis, J. M. Hamm and

O. Hess, Overcoming losses with gain in a negative refractive index

metamaterial, Phys. Rev. Lett., 2010, 105, 127401.

173 S. M. Xiao, V. P. Drachev, A. V. Kildishev, X. J. Ni, U. K. Chettiar,

H. K. Yuan and V. M. Shalaev, Loss-free and active optical

negative-index metamaterials, Nature, 2010, 466, 735–738.

174 M. T. Hill, et al., Lasing in metallic-coated nanocavities, Nat.

Photonics, 2007, 1, 589–594.

175 M. A. Noginov, G. Zhu, A. M. Belgrave, R. Bakker, V. M. Shalaev,

E. E. Narimanov, S. Stout, E. Herz, T. Suteewong and U. Wiesner,

Demonstration of a spaser-based nanolaser, Nature, 2009, 460,

1110–1112.

176 R. F. Oulton, V. J. Sorger, T. Zentgraf, R. M. Ma, C. Gladden,

L. Dai, G. Bartal and X. Zhang, Plasmon lasers at deep

subwavelength scale, Nature, 2009, 461, 629–632.

177 Y. Yang, A. Q. Liu, L. K. Chin, X. M. Zhang, D. P. Tsai, C. L. Lin,

C. Lu, G. P. Wang and N. I. Zheludev, Optofluidic waveguide as a

transformationopticsdevice

manipulation, Nat. Commun., 2012, 3, 651.

178 Q. K. Liu, Y. X. Cui, D. Gardner, X. Li, S. L. He and I. I. Smalyukh,

Self-alignment of plasmonic gold nanorods in reconfigurable

anisotropic fluids for tunable bulk metamaterial applications,

Nano Lett., 2010, 10, 1347–1353.

179 A. B. Golovin and O. D. Lavrentovich, Electrically reconfigurable

optical metamaterial based on colloidal dispersion of metal

nanorods in dielectric fluid, Appl. Phys. Lett., 2009, 95, 254104.

for lightwave bendingand

5292 | Nanoscale, 2012, 4, 5277–5292 This journal is ª The Royal Society of Chemistry 2012