An Overview of the Theory and Applications

of Metasurfaces: The Two-Dimensional

Equivalents of Metamaterials
Christopher L. Holloway1, Edward F. Kuester2, Joshua A. Gordon1, John O’Hara3,
Jim Booth1, and David R. Smith4
1
National Institute of Standards and Technology, Electromagnetics Division,
US Department of Commerce, Boulder Laboratories
325 Broadway, Boulder, CO 80305 USA
Tel: +1 (303)-497-6184; Fax: +1 (303) 497-6665; E-mail: [email protected]

University of Colorado
2

Boulder, CO 80309-0425 USA

3
Los Alamos National Labs
Los Alamos, NM USA

4
Duke University
Durham, NC USA

Abstract

Metamaterials are typically engineered by arranging a set of small scatterers or apertures in a regular array throughout
a region of space, thus obtaining some desirable bulk electromagnetic behavior. The desired property is often one that
is not normally found naturally (negative refractive index, near-zero index, etc.). Over the past ten years, metamaterials
have moved from being simply a theoretical concept to a field with developed and marketed applications. Three-
dimensional metamaterials can be extended by arranging electrically small scatterers or holes into a two-dimensional
pattern at a surface or interface. This surface version of a metamaterial has been given the name metasurface (the term
metafilm has also been employed for certain structures). For many applications, metasurfaces can be used in place
of metamaterials. Metasurfaces have the advantage of taking up less physical space than do full three-dimensional
metamaterial structures; consequently, metasurfaces offer the possibility of less-lossy structures. In this overview paper,
we discuss the theoretical basis by which metasurfaces should be characterized, and discuss their various applications.
We will see how metasurfaces are distinguished from conventional frequency-selective surfaces.

Metasurfaces have a wide range of potential applications in electromagnetics (ranging from low microwave to optical
frequencies), including: (1) controllable “smart” surfaces, (2) miniaturized cavity resonators, (3) novel wave-guiding
structures, (4) angular-independent surfaces, (5) absorbers, (6) biomedical devices, (7) terahertz switches, and (8) fluid-
tunable frequency-agile materials, to name only a few. In this review, we will see that the development in recent years of
such materials and/or surfaces is bringing us closer to realizing the exciting speculations made over one hundred years
ago by the work of Lamb, Schuster, and Pocklington, and later by Mandel’shtam and Veselago.

Keywords: Agile materials; effective media; engineered materials; generalized sheet transition conditions (GSTC);
metafilm; metamaterial; metasurface; polarizability densities; surface susceptibility; tunable materials

10 ISSN 1045-9243/2012/$26 ©2012 IEEE IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 10 6/3/2012 4:56:57 PM

 I. Introduction specific examples. Figure 1b shows a set of metallic scatterers
that generate an electrical response analogous to the magnetic

I
response of the classical split-ring structures. Figure 1c shows
n recent years, there has been a great deal of attention
an array of spherical particles (of which the three-dimensional
devoted to metamaterials [1-10]. Metamaterials are novel
metamaterial version was introduced in [7], based on earlier
synthetic materials engineered to achieve unique properties not
work by Lewin [21], but anticipated almost 100 years earlier
normally found in nature. In the context of electromagnetics,
by Gans and Happel [22]). Figure 1d shows an array of ceramic
early examples of these were artificial dielectrics. Later, we will
cubes. This surface version of a metamaterial was originally
see that the terms metamaterial and metasurface do not refer to
given the name metafilm [23, 24], by which is specifically
classical periodic structures, such as photonic bandgap (PBG)
meant a surface distribution of electrically small scatterers. It is
structures or frequency-selective surfaces (FSS). One particular
important to note that the individual scatterers constituting the
class of metamaterial that is being studied extensively consists
metafilm are not necessarily of zero thickness (or are even small
of the so-called double-negative (DNG) materials [1-10] (also
compared to the lattice constants): they may be of arbitrary
known as negative-index materials (NIM), backward-wave
shape, and can have dimensions that are required to be small
(BW) media, or left-handed materials (LHM)). Such materials
only in comparison to a wavelength in the surrounding medium.
have the property that their effective permittivity and effective
Like that of a metamaterial, the behavior of a metafilm can be
permeability are simultaneously negative in a given frequency
characterized by the electric and magnetic polarizabilities of its
band. Another property not normally found in nature that can be
constituent scatterers. Metafilms have also been referred to in
achieved with metamaterials is that of near-zero refractive index.
the literature as metasurfaces or single-layer metamaterials. We
In this type of material, either the permittivity or permeability
will attempt to clarify the terminology in Sections 1.1 and 1.2.
is designed to have its real part close to zero. Materials with
unique properties such as these have a wide range of potential
For many applications, metasurfaces can be used in place
applications in electromagnetics at frequencies ranging from the
of metamaterials. Metasurfaces have the advantage of taking
low microwaves to optical, including shielding, low-reflection
up less physical space than do full three-dimensional (3D)
materials, novel substrates, antennas, electronic switches,
metamaterial structures; consequently, metasurfaces offer the
“perfect lenses,” and resonators, to name only a few.
possibility of less-lossy structures. The use and application
of metasurfaces from microwave to optical frequencies has
Modern metamaterial research activities were stimulated
achieved great popularity in recent years [24-82]. In addition
by the theoretical work of Veselago [4], and later by the reali-
to the applications mentioned above for metamaterials,
zation of such structures by Pendry, Smith et al. [5]. However,
metasurfaces allow for controllable “smart” surfaces, minia-
many researchers in the field today fail to realize that the con-
turized cavity resonators, novel wave-guiding structures, com-
cept of negative-index materials and their interesting behavior
pact and wide-angle absorbers, impedance-matching surfaces,
date back much earlier. In fact, these types of materials were
and biomedical devices, to name only a few. Some of these will
anticipated by theory dating back more than a century. Some
be discussed in more detail later.
authors have attributed the first study of such media to Veselago
[4] in 1967, but Sivukhin [11] in 1957 briefly examined their
properties. Both authors, as well as Malyuzhinets [12] and Silin
[13-15], gave credit to much earlier work of L. I. Mandel’shtam 1.1 Metasurfaces Compared to
[16, 17]. Mandel’shtam himself referred to a 1904 paper of Frequency-Selective Surfaces
Lamb [18], who may have been the first person to suggest the
existence of backward waves (waves with phase and group A few comments are needed on (1) the difference between
velocities that are in opposite directions; his examples involved a metamaterial and a conventional photonic bandgap (PBG) or
mechanical systems rather than electromagnetic waves). In his electromagnetic bandgap (EBG) structure, and, in turn, (2) the
1904 book on optics, Schuster [19] briefly noted Lamb’s work, difference between a metasurface and a conventional frequency-
and gave a speculative discussion of its implications for optical selective surface (FSS). Consider first three-dimensional (bulk)
refraction should a medium with backward-wave properties ever metamaterials: these ideas will be extended below to
be found. In 1905, Pocklington [20] showed that a bicycle chain metasurfaces. To this end, it is important to understand the
suspended in a certain way could support backward waves, and electromagnetic behavior of a periodic composite material in
that a suddenly activated source does indeed produce a wave various frequency ranges or at different length scales. The
with a group velocity that is directed away from the source, behavior of such a composite material can be separated into
while the phase velocity moves towards it. three distinct regions (see Figure 2), with distinctive behaviors
in each region. For a three-dimensional bulk material, Region 1
Metamaterials are often engineered by arranging a set corresponds to the quasistatic region. This implies low
of small scatterers (e.g., metallic rings and rods, or spherical frequencies: more specifically, frequencies at which the
magneto-dielectric particles) in a regular array throughout a wavelength is much larger than the period of the structure (that
region of space, thus obtaining some desirable bulk behavior is, the periodicity of the scatterers that compose the composite
[1-10]. This concept can be extended by judiciously placing medium). These scatterers could have induced or permanent
electrically small scatterers into a two-dimensional pattern at a dipole moments, as is the case for atoms or molecules for
surface or interface, as shown in Figure 1. Figure 1a illustrates classical materials. Alternatively, these scatterers could be
a general set of scatterers, while Figures 1b-1d show more generic in shape and placed in a host matrix to obtain a man-

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 11
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 11 6/3/2012 4:56:57 PM

 Figure 1a. Illustrations of metasurfaces: a planar array of
arbitrarily shaped scatterers.

Figure 1c. Illustrations of metasurfaces: planar arrays of

spherical particles.

Figure 1b. Illustrations of metasurfaces: planar arrays of

metallic scatterers (electrical equivalent of a classical split- Figure 1d. Illustrations of metasurfaces: an array of ceramic
ring structure). cubes.

12 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 12 6/3/2012 4:56:59 PM

 made composite material designed to achieve some specified sometimes operated in this regime, but the distinction between
behavior. In this region, classical mixing formulas are used to this type of operation and that of Region 3 has not usually been
obtained equivalent effective-material properties (permittivity, made.
ε ; permeability, µ ; etc.). The determination of effective-
medium properties and the modeling of the electromagnetic We need to emphasize that this second region in Figure 2
response to inclusions embedded in a host material is a problem may not always occur. The scatterers need to be properly
with a long history going back to Maxwell and Rayleigh. designed, such that the scatterers’ resonances occur at a fre-
quency well below that where the next higher-order Floquet-
When the wavelength becomes comparable to or smaller Bloch mode can propagate. For example, if the bulk properties
than the period of the structure, the fields no longer “see” and the radius of a spherical-particle composite (see [7]) are
the composite as an effective medium. This corresponds to too small, or if the sizes or shapes of the scatterers used in the
Region 3 in Figure 2. At these frequencies, a more-compli- material are not properly chosen, the scatterers’ resonances
cated field behavior exists, and more-elaborate techniques for would be pushed toward the Floquet-Bloch-mode region,
analyzing the EM field interaction with the composite periodic and a double-negative material could not be realized [7].
structure must be used (i.e., full-wave approaches). The classi- The scatterers’ resonance region would be absorbed into the
cal analytical approach for this is the Floquet-Bloch mode Floquet-Bloch-mode region, and an effective-medium model
expansion [83-88], in which the fields are expanded into an would not adequately describe the behavior of the composite
infinite sum of plane waves propagating in various directions. material.
As the wavelength approaches the period, higher-order
Floquet-Bloch modes must be considered. These higher-order In summary, in Regions 1 and 2 of Figure 2 the electro-
modes then interfere with the fundamental wave propagating magnetic field behaves as if the composite material is some
through the composite, and we call the composite a photonic type of effective medium. In Region 1 (the classical mixing-
bandgap or electromagnetic bandgap material in this frequency theory region), the effective material properties are not fre-
range. We can say that the material is exhibiting array or quency dependent (except insofar as the constituent bulk
lattice dispersion in this case. At certain frequencies, photonic- properties have a frequency dependence). In Region 2 (the
bandgap and electromagnetic-bandgap structures block the scatterers’ resonant region), the material behaves as an effective
propagation of EM waves through the material. The frequency medium and has an inherent frequency dependence exhibited in
bands where this occurs are referred to as stop bands. At other the effective material properties. In this region, it is possible to
frequencies, the periodic structure allows energy to propagate achieve three-dimensional double-negative materials and other
through the structure: these frequency bands are referred to as interesting resonant behaviors (e.g., index near zero). In the
pass bands. Phenomena such as Bragg scattering are associated last region (Region 3 in Figure 2), the electromagnetic field’s
with this frequency region, and are the basis of many practical interaction with the periodic structure is very involved. We may
applications [1, 88]. no longer think of the composite material as behaving like an
effective medium. When the wavelength approaches the period,
Region 2 in Figure 2 corresponds to a region where the higher-order Floquet-Bloch modes must be considered, and one
period of the structure is still small compared to a wavelength, typically does not refer to these materials as metamaterials or
but the individual scatterers are designed in such a manner metasurfaces in this region.
(either via their shape or their constitutive properties) that the
scatterers themselves can resonate. When this occurs, a new
class of manmade materials (metamaterials) is realized, making 1.2 Types of Metasurfaces
possible a broad range of unique behavior not commonly found
in nature (double-negative or near-zero index materials, for
example). Region 2, where the scatters resonate (but not the We will call any periodic two-dimensional structure the
lattice, as is the case in Region 3), is where we call our medium thickness and periodicity of which are small compared to a
a metamaterial. The medium is dispersive here, but we may still wavelength in the surrounding media a metasurface. Within
characterize its electromagnetic behavior with the effective this general designation, we identify two important subclasses.
parameters ε and µ .

Similar behavior (three regions of behavior) will occur

for scatterers in a two-dimensional array (a metasurface). For
a two-dimensional lattice of scatterers, Region 1 corresponds
to classical thin-film materials, while Region 3 corresponds
to resonances associated with the periodicity of the scatterers.
The conventional frequency-selective surface [87] falls into
this third region. On the other hand, when we talk about a
metasurface, we are referring to an array of scatterers that falls
into Region 2. Resonances of the surface may be associated
with the resonances of the scatterers, but not with the perio- Figure 2. Three characteristic regions of a composite
dicity of the array. Ordinary frequency-selective surfaces are material.

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 13
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 13 6/3/2012 4:56:59 PM

 property characterization of a metasurface is incorrect at a
fundamental level. To the extent that classical algorithms for
bulk-parameter extraction give results that depend on sample
size, we must conclude that some localized effect is occurring
near the boundary of the sample, analogous to the effect of cutoff
modes near the junction between two different waveguides.

Figure 3. A metascreen. To illustrate the difficulties that arise, let us replace the
metasurface in Figure 1 by an equivalent material layer of
thickness d (see Figure 4). The effective material properties of
Metasurfaces that have a “cermet” topology – which refers this equivalent layer are determined from the same modified
to an array of isolated (non-touching) scatterers, as shown NRW approach used to analyze the bulk three-dimensional
in Figure 1a – are called metafilms, a term coined in [23] for metamaterial as discussed above. The problem is that effective
such surfaces. Metasurfaces with a “fishnet”‘ structure, as bulk material properties are not uniquely defined for a
shown in Figure 3, will be called metascreens. These are char- metasurface. While the geometry of the scatterers and the lat-
acterized by periodically spaced apertures in an otherwise tice constant, p, are uniquely defined, the thickness of the
relatively impenetrable surface. Other kinds of metasurfaces lie equivalent layer, d, is not. In fact, one could choose d to be any
somewhere between these two extremes. For example, a grating value, chosen within reasonable limits. Since d is not uniquely
of parallel conducting wires behaves like a metafilm in the defined, the effective material properties obtained in this way
direction perpendicular to the wires’ axes, but like a metascreen suffer from the same ambiguity, and hence are not uniquely
in the direction along these axes. defined. Therefore, contrary to what has been asserted elsewhere
in the literature [29-34], characterizing the metasurface as an
effective medium in this way is inappropriate and leads to
misinterpretations. In [30], Smith et al. addressed this point by
1.3 Modeling a Metasurface asserting that the quantity “ ε d ” is most likely constant. The
connection between the surface susceptibilities and bulk
The traditional and most convenient method by which to effective material properties was also discussed in [101], where
model metamaterials is with effective-medium theory. In this it was argued that for a metasurface, one cannot define bulk
approach, some type of averaging is performed on the electric parameters ( ε and µ ) that are independent of d. As such, it is
and magnetic fields over a given period cell composing the
the product of d and a bulk parameter ( ε or µ ) that must be
metamaterial. From these averages, the effective permittivity
and permeability (and hence, the refractive index) of the fixed and unique. This results in an arbitrariness in the choices
metamaterials are determined. It should be emphasized that the of d, ε , and µ . With that said, one can use these retrieved
averaging (also called homogenization) is valid only when the effective material properties as long as one also assumes that
wavelength is “large enough” compared to the lattice constant the slab thickness is held constant. One cannot necessarily use
of the period cell. While period-cell averaging for the fields is the same effective material properties for a slab of a different
the correct method for defining effective material properties thickness, let alone for any other non-planar shape of the
(see [61, 89-93] for this type of homogenization averaging), effective medium. As long as we are willing to use these
many researchers have in practice used an approach based upon effective material properties with a set thickness, they can be
the reflection and transmission coefficients of a metamaterial used without any lose of generality. On the other hand, if one
sample of some defined thickness. The Nicolson-Ross-Weir changes the thickness, but keeps the effective properties the
(NRW) approach [94, 95] or a variant of it is then used to obtain same, then the results would be meaningless.
the effective material properties of the bulk metamaterials.
Note that the standard NRW approach must be modified when This point is emphasized in Figures 5 and 6. Figure 5
negative material properties exist. Typically, the choice of the shows the real parts of the effective permittivity for an array of
sign of a square root is made unambiguous by ensuring positive spherical particles obtained in this manner for different values
power flow in the direction of propagation [96-98]. In some of d (where d is the effective thickness). Figure 6 shows the
situations, local effects near the boundaries of the sample must results for the refractive index for different thicknesses for an
also be taken into account [99]. If used properly, the effective- array of metallic scatterers. The effective properties shown in
medium approach can be a self-consistent and unique method Figures 5 and 6 are seen to be functions of the chosen thickness,
for characterizing a metamaterial. d, which illustrates that these effective bulk properties are not
an intrinsic property of the metasurface. We should also add
Attempts to use a similar bulk-parameter analysis for that the effective bulk properties of a metafilm/metasurface
metasurfaces have been less successful (see [100] for a detailed
discussion on this point). Indeed, some previous metasurface
studies have modeled the film as a single-layer metamaterial
in which effective bulk material properties of the metasurface
are obtained by arbitrarily introducing a nonzero thickness
parameter into the analysis. Several problems arise from the Figure 4. Representing a metafilm as a bulk effective
physically artificial character of this parameter: such a bulk- medium of thickness d.

14 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 14 6/3/2012 4:56:59 PM

 extracted in this way can display non-physical behaviors, as
well. For example, causality (as manifested in the Kramers-
Kronig relations) may be violated, or a passive material may
exhibit a bulk permittivity or permeability with gain (a positive
imaginary part for the e jωt time dependence assumed in this
paper).

In [100], it was shown that the surface susceptibilities are

unique properties of a metafilm, and as such, serve as the most-
appropriate manner of characterizing this type of metasurface.
Physically speaking, the metasurface acts like an infinitesimal
sheet: one that causes a phase shift (and possibly a change in
amplitude) in wave fields that interact with it. Scattering by
such a metasurface is best characterized by generalized sheet-
transition conditions (GSTCs) [23], in contrast to the effective-
medium description used for a metamaterial. The coefficients
appearing in the generalized sheet-transition conditions for
any given metasurface are all that are required to model the
metatsurface’s interaction with an electromagnetic field.

For a metafilm, it was shown in [23] that the generalized

sheet-transition conditions relate to the electromagnetic fields
on both sides of the metasurface (see Figure 1a):

0+  zz
a z × H z =0− jωχ ES  Et ,av
= − a z × ∇t  χ MS H z ,av 
z =0   z =0

0+  zz
E z =0− × a z jωχ MS  H t , av
= + a z × ∇t  χ ES Ez , av 
z =0   z =0
(1)

where the subscript “av” represents the average of the field on

either side of the metasurface, the subscript t refers to compo-

Figure 6. The real (top) and imaginary (bottom) parts of

the refractive index of a metafilm modeled as an effective
medium for different layer thicknesses. The metafilm in
this figure was composed of square metallic scatterers as in
Figure 1b.

nents transverse to z, and a z is the unit vector. The parameters

 
χ ES and χ MS are the dyadic surface electric and magnetic
susceptibilities. These have units of meters, and are related to
the electric and magnetic polarizability densities of the scatterers
per unit area. These dyadics vanish when the scatterers are
absent, and Equation (1) reduces to the ordinary conditions of
Figure 5. The effective permittivity of a metafilm modeled continuity of the tangential components of E and H. For
as an effective medium for different layer thicknesses. The simplicity, we consider here only the case where the scatterers
metafilm in this figure was composed of spherical particles and lattice have sufficient symmetry such that the surface
as seen in Figure 1c. susceptibility dyadics are diagonal:

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 15
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 15 6/3/2012 4:56:59 PM

  xx yy zz 0+ 
χ ES = χ ES a x a x + χ ES a y a y + χ ES azaz , a z × H z = 0− =
YES  Et ,av ,
z =0
(2) (5)
 xx yy zz 
χ MS = χ MS a xa x + χ MS a ya y + χ MS azaz . 0+
E z = 0− ×az =
Z MS  H t ,av ,
z =0

In this paper and in [101], we have chosen a sign convention for

in which the spatially dispersive (k-dependent) surface transfer
χ MS such that there is a minus sign in front of the second term
admittance and transfer impedance are given by
on the right-hand side (RHS) for the jump in the H field, and a
plus sign in front of the first term on the right-hand side for the
  j χ zz
jump in the E field (see Equation (1)). Note that this is a different YES = jωχ ES + MS ( a z × k )( a z × k )
sign convention than that used in [100]. We have chosen this ωµ
new sign convention because it ensures that Re ( χ MS ) > 0 (6)
when the surface magnetization density is in the same direction   j χ zz
Z MS = jωχ MS + ES ( a z × k )( a z × k ) .
as H (paramagnetic material), which is consistent with the ωε
definition of bulk susceptibilities of classical magnetic
materials. Such boundary conditions can also be interpreted as equivalent
transmission-line circuits [105].
The generalized sheet-transition conditions for the more-
general case of non-symmetric, bi-isotropic, and bianisotropic These generalized sheet-transition conditions allow a
surface susceptibility dyadics are possible, and the theory for metasurface to be modeled by a boundary condition that is
these has yet to be fully developed. We should emphasize that applied across an infinitely thin equivalent surface. The details
the generalized sheet-transition conditions of Equation (1) of the surface geometry are incorporated into this boundary
are appropriate only for metafilms. Metasurfaces with other condition through the surface susceptibilities or porosities. The
structures will require a different form of the generalized sheet- fields that appear in the generalized sheet-transition conditions
transition conditions. For example, metascreens consisting of are “macroscopic” fields, in the sense that they do not exhibit
periodic isolated apertures in a zero-thickness perfect conductor variations on a length scale comparable to scatterer or aperture
obey the Kontorovich form of generalized sheet-transition dimensions or the lattice spacing, but only on larger scales such
conditions [83], which can be written in the form as the wavelength in the surrounding medium. It should be
noted that the generalized sheet-transition conditions as written
 0+ 1 zz 0 +
here do contain (in a first approximation) spatial-dispersion
E
= × a z jωµπ MS  H z 0− + a z × ∇t π ES Dz  − effects via the presence of the surface gradients in Equations (1)
=z 0= ε   z =0
(3) and (3). A more-accurate account of these effects can be given
if the surface susceptibilities and porosities are allowed to be
where the tangential E field and the normal B field are con- functions of the vector wavenumber (i.e., of the direction of
  propagation of the waves).
tinuous across the metascreen. Here, the dyadics π ES and π MS
are what we call the electric and magnetic porosities of the
If the fine structure of the spatial field variation is not
metascreen, and, like the surface susceptibilities of a metafilm,
required, these generalized sheet-transition conditions along
these have dimensions of length. When the apertures are absent
with Maxwell’s equations are all that are required to analyze
and we have only the perfectly conducting plane, the electric
the interaction of the fields with the metasurface. The surface
and magnetic porosity dyadics vanish, and Equation (3) reduces
susceptibility and porosity dyadics that appear explicitly in the
to the vanishing of tangential E, as expected. The most general
generalized sheet-transition conditions are uniquely defined
form of metascreen boundary conditions has not yet been
(unlike in a bulk effective-parameter model), and, as such,
determined, and its development is the subject of ongoing work.
represent the physical quantities that uniquely characterize
The same is true for metasurfaces such as wire gratings, which
the metasurface. Retrieval techniques based on this model are
exhibit features of both metafilms and metascreens (cf. the
described in the next section.
generalized sheet-transition conditions for thin-wire gratings in
[102] and [103]).

The generalized sheet-transition conditions can also be 1.4 Characterization of Metasurfaces

cast in the form of impedance-type boundary conditions [83,
104-107]. For plane-wave fields the variation of which parallel As discussed in the last section, the bulk effective permit-
to a metafilm is of the form tivity and permeability (as well as the refractive index) of a
metasurface are not uniquely defined. In this section, we pre-
e − jk rt (where
= k k x a x + k y a y and=
rt xa x + ya y ), (4) sent an approach for the unique characterization of a metafilm
based on an inversion of its reflection and transmission coeffi-
we can use Maxwell’s equations to write Equation (1) as cients to obtain the surface susceptibilities, details of which
were discussed in [100] and [101].

16 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 16 6/3/2012 4:57:00 PM

 The reflection (R) and transmission (T) coefficients of a
metafilm for both TE and TM polarized plane waves (see Fig-
ure 7) were derived in [24]. The reflection coefficient, R, and
the transmission coefficient, T, for a TE-polarized plane wave
are given by

RTE (θ ) =
−j
k0
2 cos θ ES
(
χ yy + χ MS
zz xx
sin 2 θ − χ MS cos 2 θ
,
)
D1

2
 k  xx
1 +  0  χ MS
 2 
yy
χ ES zz
+ χ MS (
sin 2 θ )
TTE (θ ) = , (7)
D1

2
 k  xx
D1 ≡ 1 −  0  χ MS
 2 
yy
χ ES (
zz
+ χ MS sin 2 θ )
+j
k0
2 cos θ ES
(
χ yy + χ MS
xx zz
cos 2 θ + χ MS sin 2 θ , )
and for a TM polarized plane wave by

RTM (θ ) =
−j
k0
2 cos θ
xx
χ ES ( yy
cos 2 θ − χ MS zz
− χ ES sin 2 θ )
D2
Figure 7. A plane wave incident onto a metafilm: (top) TE
polarization, and (bottom) TM polarization.
2
 k  xx
1 +  0  χ ES
 2 
yy
χ MS zz
+ χ ES (
sin 2 θ ) 2 j RTE ( 0 ) − TTE ( 0 ) + 1
TTM (θ ) = , (8) xx
χ MS = ,
D2 k0 RTE ( 0 ) − TTE ( 0 ) − 1
(9)
2 j RTE ( 0 ) + TTE ( 0 ) − 1
2
D2 =
 k  xx
1 −  0  χ ES
 2
yy
χ MS zz
+ χ ES (
sin 2 θ ) yy
χ ES =
k0 RTE ( 0 ) + TTE ( 0 ) + 1
,

+j
ko
2 cos θ
yy
χ MS + χ ES(
xx
cos 2 θ + χ ESzz
sin 2 θ , ) and

where k0 is the free-space wavenumber. These reflection and zz χ yy 2 j cos (θ ) RTE (θ ) + TTE (θ ) − 1
χ MS − 2ES +
= ,
transmission coefficients apply to a metafilm the constituent sin (θ ) k0 sin 2 (θ ) RTE (θ ) + TTE (θ ) + 1
scatterers of which have sufficient symmetry that the surface (10)
susceptibility dyadics are diagonal. The more-general case of
non-symmetric, bi-isotropic surface susceptibility dyadics where R ( 0 ) and T ( 0 ) are the reflection and transmission
could also be handled in a similar way, as well as that of a
coefficients at normal incidence, and R (θ ) and T (θ ) are the
metascreen or more-general metasurface. These cases have yet
to be developed in full. reflection and transmission coefficients at some oblique inci-
dence angle, θ . For a TM-polarized wave, the three unknown
Once the reflection and transmission coefficients are surface susceptibilities are determined by
obtained (either from measurements or from numerical calcu-
lations), the surface susceptibilities can be determined. For xx 2 j RTM ( 0 ) + TTM ( 0 ) − 1
χ ES = ,
the present case, two different sets of R and T are required k0 RTM ( 0 ) + TTM ( 0 ) + 1
(for example, one at normal incidence and one at oblique inci- (11)
dence) for each polarization. For a TE-polarized wave, the yy 2 j RTM ( 0 ) − TTM ( 0 ) + 1
three unknown surface susceptibilities are determined from the χ MS = ,
k0 RTM ( 0 ) − TTM ( 0 ) − 1
following:
and

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 17
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 17 6/3/2012 4:57:00 PM

 χ yy 2 j cos (θ ) TTM (θ ) − 1 − RTM (θ ) formulas therein), we can then obtain the polarizabilities of the
zz
χ ES − 2MS +
= . individual scatterers in the metafilm. These are finally inserted
sin (θ ) k0 sin 2 (θ ) TTM (θ ) + 1 − RTM (θ ) into the Clausius-Mossottii relations to obtain the effective
(12) permittivity and permeability of a metamaterial realized by
stacking these metafilms in a third dimension. This technique
The sign errors that were printed in [100] are corrected in the is a useful alternative to standard parameter-extraction meth-
expressions given above. Corrections were also given in [101]. ods when transmission through a sample of many layers is very
small.
This approach can be applied equally well either to
numerically or experimentally determined values of R and T. In the remainder of the paper, we will discuss various
However, because of the difficulty of separating the incident applications of metasurfaces that have emerged over the last
and reflected components at normal incidence in measure- few years.
ments, it may be more beneficial to rewrite these equations for
two arbitrary incidence angles that both differ from zero, as
discussed in [100]. The interesting thing to notice is that this
retrieval approach for characterizing a metafilm by its surface
susceptibilities does not pose the problem of choosing the
appropriate sign of a square root, nor require the assumption of
an arbitrary layer thickness, as is the case when trying to infer
the effective properties for a bulk metamaterial model.

In order to provide a validation of these expressions for

retrieving the surface susceptibilities of a metafilm, we consider
the example of a metafilm consisting of a square array of
spherical particles, as in Figure 1c. In this example, we chose
a = 10 mm, p = 25.59 mm, εr = 2 , µr = 900 , and
tan δ = 0.04 . This structure is appealing because the polariza-
bilities of the spherical particles are known analytically, and
hence the surface susceptibilities can be calculated analytically
by use of expressions given in [100]. As suggested above, there
are various possible sets of the R and T that could be used to
determine the unknown susceptibilities, and several approaches
were discussed in [100].

yy
Figure 8 shows the real and imaginary parts of χ ES and
zz
χ ES . These results were obtained from numerically simulated
values of R and T for both polarizations at a 30° incidence
angle. The metafilm used in this example was a square array of
lossy spherical particles, and thus had symmetric transverse
xx yy xx yy
surface susceptibilities (or χ ES = χ ES and χ MS = χ MS ). Also
shown in the figure are the analytical results, details of which
were given in [100]. From this comparison, we saw that the
surface susceptibilities obtained from the retrieval approach
were virtually identical to those obtained from the approximate
analytical expressions.

The example given here and others presented in [100]

illustrate the validity of this retrieval approach. More impor-
tantly, these results illustrate that the surface susceptibilities are
unique properties of a metafilm, and as such serve as the most
appropriate way to characterize them.

Metafilms can also be used as a means of extracting bulk

properties of three-dimensional metamaterials [108]. In this
technique, plane-wave reflection and transmission coefficients
from a metafilm are used to extract the surface susceptibilities, Figure 8. The surface susceptibilities for a metafilm com-
as above. From [23] (or more properly, a refinement of the posed of spherical particles.

18 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 18 6/3/2012 4:57:00 PM

 2. Controllable Surfaces

Given a generic metasurface, one could use one of a

number of the commercial computational codes to analyze
the interaction of an electromagnetic field with a metasurface.
However, as we have seen, the generalized sheet-transition
conditions allow us to obtain closed-form expressions for
the reflection and transmission coefficients of a plane wave.
For a metafilm, they take the form of Equations (7) and (8).
These closed-form relationships are expressed in terms of the
surface susceptibilities of the metafilm, which are related to the
scatterers that make it up. The advantage of the closed-form
expressions is that one can get some physical insight into the
relationship of the surface susceptibilities to the reflection
and transmission behavior. Expressions (7) and (8) illustrate
that if the surface susceptibility (or the electric and magnetic
polarizability of the individual scatterers) can be changed, it is
possible to control the reflection and transmission behavior of
the surface.

As examples, from Equations (7) and (8), one can write Figure 9. The reflection coefficient as a function of the
down the relationship between the electric and magnetic sur- permeability, µrp , of the spherical particles comprising the
face susceptibilities needed to have either total reflection or
metasurface.
total transmission [24]. For total reflection, the following con-
ditions must be satisfied:

xx
k02 χ MS (
yy
χ ES zz
+ χ MS sin 2 θ = )
−4 for TE,
(13)
xx
k02 χ ES ( yy
χ MS zz
+ χ ES 2
)
sin θ =
−4 for TM,

while for total transmission, the required conditions are

yy xx zz
χ ES − χ MS cos 2 θ + χ MS sin 2 θ =
0,
(14)
yy xx 2 zz 2
χ MS − χ ES cos θ + χ ES sin θ =
0.

In principle, control of the reflection and transmission behav-

ior can be done in a number of ways, e.g., (1) by changing
the electrical or magnetic properties of the scatterers, or (2)
by changing the properties (either the material properties or
the geometry) of the substrate on which the scatterers lie. For
example, for a metafilm made up of spherical magnetic parti-
cles, Figure 9 shows R as a function of the permeability of the
scatterers [24]. From this figure, we see that the metafilm goes
from total reflection to total transmission as the permeability of
the spherical particles is varied. Such a controllable surface has
been realized by using a metafilm of spherical YIG particles,
and controlling the surface behavior with a dc magnetic bias
field [25]. Figure 10 shows the transmission behavior of such a
metafilm as a function of the external dc magnetic field. Other
approaches have been used to control the surface behavior of a
metasurface, and a significant amount of research is ongoing in
the area of dynamically controllable metasurfaces, as we will Figure 10. Controlling the transmission properties of a
discuss in Sections 7 and 9. surface with an external dc magnetic field.

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 19
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 19 6/3/2012 4:57:00 PM

 3. Anisotropic Metasurfaces: Angularly xx
χ MS yy
χ ES = −
4
for TE,
Independent Behavior, Absorbers, and k02
Impedance-Matching Surfaces (15)
xx yy 4
χ ES χ MS = − for TM.
The use of anisotropic scatterers to make up a metafilm k02
can be used to achieve a number of useful results not possible
with isotropic metasurfaces. One of the more interesting If a metasurface is thus designed such that the transverse com-
aspects of bulk metamaterials is the ability to engineer anisot- ponents of the surface susceptibilities are highly resonant as
ropic behavior into them. In fact, this is the idea behind so- compared to the normal components, then the metasurface may
called “cloaking” materials, which has generated so much exhibit angle-independent behavior. This behavior is illustrated
media interest. The cloaking concept is to design a specific in Figure 11, where we have plotted the reflection coefficient
anisotropic behavior into a metamaterial in such a way as to of a metasurface composed of the metallic structures shown in
cause electromagnetic energy to propagate or bend around an Figure 1b [60]. In this figure, we see that the surface exhibits
object that is covered with it [65, 66]. If the anisotropic mate- a nearly angle-independent behavior up to incidence angles as
rials are correctly designed, the coated object will neither scatter large as 60°.
nor absorb energy, and hence it appears to the electromagnetic
(EM) fields as if the object is not present. One could in principle One can extend this concept to many other types of struc-
thus achieve an object that appears invisible to the EM energy in tures. For example, it can be shown that highly anisotropic
any desired frequency band, from RF to optical. We might add slabs of materials will exhibit some type of angle-independent
that due to the inherent properties of materials and structures, behavior, as well. This can be seen by examining the reflec-
there are technical hurdles to overcome (some may appear to tion properties of an inhomogeneous anisotropic slab, as dis-
be impossible) in order to obtain the “holy grail” of broadband cussed in [61] (see Equations (31)-(38) therein). Additionally,
cloaking metamaterials. Only narrowband, single-polarization the mathematically constructed perfectly matched layer (PML)
examples have been experimentally demonstrated to date. [62] that was introduced into computational electromagnetics
for reducing non-physical reflections at radiation boundaries
This concept could be extended to metasurfaces, as well. also requires such angular independence. Such perfectly
In such a design, the scatterers composing a metasurface would matched layers may be physically realizable by use of the
be chosen and varied across the surface in order to achieve a concept of metasurfaces. The idea of designing a metasurface
desired surface behavior. For example, it is in principle possible structure to use as a perfectly matched layer in computational
to design a metasurface so as to focus an EM plane wave to a codes is under investigation.
desired region in space, much like a focusing antenna array. If
the scatterers of the metasurface were designed in a manner such This mechanism offers a unique way to match lossy mate-
that they could be changed at will, one could as a result have rials for the purpose of developing compact electromagnetic
a metasurface capable of changing the direction and frequency absorbers, and to make possible unique designs of impedance-
where the energy is focused, i.e., a frequency- and space-agile matching surfaces. Compact absorbers based on such ideas
surface. These concepts are currently being investigated. have recently begun to appear in the literature [53-58]. These
Another example deals with how to obtain (at least in
some parameter range) angularly independent behavior of the
reflection and transmission [60]. For a given incidence angle,
Equation (13) gives the relationship between the electric and
magnetic surface susceptibilities required for total reflection.
Equation (13) suggests a mechanism by which one could
achieve approximately the angular independence of such total
yy zz
reflection. For instance, in the TE case, when χ ES  χ MS , the
angular dependence becomes less significant, because the first
term in the parentheses of Equation (13) dominates over the
second angularly dependent term. Similarly, for the TM case,
yy zz
when χ MS  χ ES , the angular dependence likewise becomes
weaker. These conditions could occur, for example, at certain
resonance frequencies of the scatterers that constitute a meta- Figure 11. The reflection coefficient for a TE-polarized
film. When the indicated terms are dominant, the total reflection incident plane wave for an electrical resonator metasurface
conditions, Equation (13), become structure.

20 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 20 6/3/2012 4:57:01 PM

 structures typically are designed with a metasurface on the front resonator needed to obtain a desired resonant frequency. The
side of a lossy substrate backed by a metal plate (see the second reduction in resonant frequency of the resonator for the simple
images in Figure 1b). These structures may be narrowband, but square-patch metafilm resonator is shown in Table 1 for
are quite compact in size. An additional benefit of using the p = 500 µm. In this table, the percent reduction (which is also a
metasurface is the fact that these structures show substantial measure of the size reduction possible for a given resonant
angular independence, which has been demonstrated both frequency) is listed for different values of l p . The results in
numerically and experimentally [53, 55, 57]. The physical
explanation of this angularly independent behavior is discussed
above and in [60].

Various groups have also investigated the use of metasur-

faces to develop impedance-matching surfaces. An array of
metallic structures similar to the one shown in Figure 12a was
used to design and fabricate a wide-angle impedance-matching
surface [59]. This structure had highly anisotropic properties,
which as we saw above can enable angularly independent
behavior. The reflection properties of such a thin anisotropic
metasurface are shown in Figure 12b.

4. Resonator Size Reduction

Engheta [63, 64] demonstrated that the λ 2 classical

lower bound on the size of a resonant structure can be reduced
if a cavity is partially filled with a negative-index material.
Extending this idea, it was shown in [35, 36] that the same thing
could be accomplished with a metasurface. The advantage of a
metasurface is that because it requires less physical space than
a three-dimensional metamaterial, a cavity or resonator with a
metasurface can in principle be made smaller than those that
use three-dimensional metamaterials. The phase-matching
conditions required at resonance for the separation distance, d,
between two metal plates with a metasurface placed in the Figure 12. An impedance-matching metasurface: (a) a
center is given by [35] typical electric-resonator structure; (b) the reflection
behavior as a function of angle, after Smith [56].
λ 2 π 
d= 1 + nπ − tan −1  χ E   ,
2 π λ 
(16)
λ 2 −1  π 
=d 2nπ − tan  χM  ,
2  π λ 

for n = 0,1, 2,3,... ,

where n = 0 is not allowed if χ M ≥ 0 (see [35] for details).

From these expressions it is seen that by judiciously choosing
the metasurface, it is possible to have a resonator that over-
comes the λ 2 size limit. Various cases for different conditions
on the scatterers were given in [35].

As an example, consider a metafilm composed of square

patches placed between two metal plates. Figure 13 shows the
resonant frequency as a function of l p (where p is the period
and l is the length of the side of one of the squares) for three
different plate separations. For reference, the classical result
d = λ 2 occurs when l p = 0 . It is seen that the capacitive Figure 13. The resonant frequency of an air-filled parallel-
metasurface can significantly reduce the resonant frequency for plate resonator loaded with a square-patch capacitive
a given resonator size d, or equivalently reduce the size of the metafilm.

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 21
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 21 6/3/2012 4:57:01 PM

 Table 1. The reduction in resonant frequency as a function It is desirable to find conditions for which the imaginary part of
of l p for a parallel-plate resonator with a metasurface ne is as small as possible, and we need to ensure that Im ( ne ) < 0
consisting of square metal patches with p = 500 µm. and 0 < Re ( ne ) < 1 (the latter conditions because otherwise the
mode will basically be a surface wave localized near the two
l p d = 0.5 mm d = 1.5 mm d = 2.0 mm metasurfaces, and will likely suffer increased attenuation as a
0 0% 0% 0% result). Similar sets of expressions for the TM modes along
0.4 7.3% 3.8% 1.9% with an example were given in [37].
0.6 20.9% 11.9% 6.4% This waveguide shows the promise of being compact,
0.8 38.4% 25.2% 14.9% with low material and radiation losses. If the metasurfaces
0.96 56.1% 42.6% 29.0% are constructed in a polymer type of material, then it should
be possible to develop a flexible wave-guiding structure. If the
scatterers composing the metafilm are then chosen properly, a
flexible low-loss waveguide could be developed, which could
have potential applications at terahertz frequencies. It is also
this table show that a square-patch metafilm placed at the center possible to control the properties of the scatterers. This could
of a resonator can reduce the size by as much as 56%. If be used to design controllable or smart frequency-agile wave-
metasurfaces are designed with scatterers having more elaborate guiding structures.
polarizability characteristics (e.g., judiciously chosen resonant
behavior), it should in principle be possible to achieve even
greater reductions in size. In fact, by controlling the properties
6. Complex Modes and Surface Waves
of the metasurface, a frequency-agile resonator could be
realized. on a Metasurface

As with the case of a classical dielectric slab, metasur-

faces may also support surface waves under suitable condi-
5. Waveguides tions. However, unlike a conventional dielectric slab, forward
and backward surface waves, as well as complex modes, can be
Because metasurfaces can be designed to have total reflec- simultaneously excited on the metafilm, a direct consequence
tion of an incident wave, it should be possible to trap and guide of engineering the properties of the constituent scatterers. In
electromagnetic energy in a region between two metasurfaces. fact, the scatterers that compose a metafilm can be judiciously
This class of waveguides is illustrated in Figure 14 (note the chosen such that surface waves and/or complex modes will be
change in coordinate system compared to Figure 7, in order to present only at desired frequencies. Details of when surface
have the waveguide mode propagate along z) [37]. Assuming a waves can be supported on a metasurface were given in [109].
waveguide mode with the propagation constant in the z There, it was shown that the poles of the reflection coefficient for
direction, β = k0 sin θ , from Equation (13) we find that for a a line source (either electric or magnetic) above a metasurface
TE mode, the normalized propagation constant must obey are functions of the surface susceptibilities of the metasurface.
Once these poles are determined, the propagation constant for
zz yy 4 a wave traveling along the surface under various conditions on
χ MS χ ES + the electrical and magnetic surface susceptibilities is given by
β k02
ne = zz xx
(17)
k0 − χ MS χ MS
βs
= k02 − k x2 , (20)
in order for total reflection to occur. This quantity will generally
be complex for leaky modes. Once the scatterers that compose
the metasurface are chosen to meet the above criterion and β
is determined, the transverse wavenumber in the x direction is
given by

kx = k02 − β 2 = k0 1 − ne2 , (18)

and the separation distance, d, between the two metasurfaces

is given by

 zz 
 1 −1 k x χ MS
 n +  π + 2 tan  

 2  2  for n = 1, 2,3, ... . (19)
d=
kx Figure 14. An illustration of a metasurface waveguide.

22 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 22 6/3/2012 4:57:02 PM

 Table 2. Conditions for surface waves and guided modes, as well as expressions for the propagation constant.

Number Type of
yy
Conditions on χ ES zz
, χ MS xx
, and χ MS of Guided Guided-Wave Poles Guided
Waves Wave

zz
χ MS >0 zz
χ MS ≥
1
zz
zz
− χ MS
2
β1,2 =
zz
− j ± k 2 χ MS (yy
χ ES zz
+ χ MS ) complex
mode
k 2 χ MS zz
χ MS

zz
χ MS >0 zz
− χ MS yy
< χ ES ≤
1
zz
zz
− χ MS
2
β1,2

= −j
2 zz yy zz
1 ± 1 − k χ MS χ ES + χ MS( )  surface
wave
k 2 χ MS zz 
 χ MS 
 

zz
χ MS >0 yy
χ ES zz
≤ − χ MS 1

β1 = − j 
2 zz yy
(zz
1 + 1 − k χ MS χ ES + χ MS )  surface
wave
zz 
 χ MS 
 

 
( ) surface
2 zz 2
zz
χ MS >0 yy
χ ES =0 2 1 ± 1 − k χ MS  wave
β1,2 = −j zz

 χ MS 
 
 

zz
χ MS <0 yy
χ ES zz
< − χ MS 1

β1 = − j 
2 zz yy
(zz
1 − 1 − k χ MS χ ES + χ MS )  surface
wave
zz 
 χ MS 
 

j k 2 yy surface
zz yy 1 β1 = − χ
χ MS =0 χ ES >0 2 ES wave

where β s is given in Table 2 for various conditions for an altering the electrical landscape, configuration, or environment
electric line source; values for a magnetic line source were of the elements composing the surface. When the electrical
given in [109]. When the surface susceptibilities of the landscape is changed – usually by means of a semiconductor
metasurface meets one of these conditions, surface waves or inclusion in a scatterer – these are sometimes termed hybrid
complex modes can be excited. Figure 15 illustrates a metamaterials [38]. The terahertz (THz) frequency regime is
metasurface that has either a surface wave or complex modes, one area in which controllable metasurfaces have made a big
depending on the frequency of excitation. This figure shows the impact. This is due mainly to the dearth of practical THz tech-
magnitude of the E field for a magnetic line source placed nology. Devices considered commonplace in microwave and
above an array of spherical particles (see [109] for the geome- photonic areas, such as switches and modulators, largely don’t
try). Figure 15a illustrates the presence of one surface wave, exist in the THz regime. A major challenge has been to find
and Figure 15b illustrates the case for a complex mode. By natural materials that respond strongly and controllably to THz
varying the properties of the scatterers, the surface suscepti- radiation, yet do not suffer high loss.
bilities will change. Hence, from the results in Table 2, it is in
Numerous demonstrations have now shown that THz
principle possible to have a surface wave and/or complex mode
metasurfaces offer these advantages. THz metamaterials lev-
exist at any desired frequency, i.e., to realize a frequency-agile
erage the benefits of a high ability to be tuned, relatively
guided-wave structure.
easy fabrication, low losses, and dynamic control. The first
dynamic control over a THz metasurface utilized a near-infra-
red laser that actively switched off the resonant response by
7. Terahertz Devices increasing the metasurface’s substrate conductivity [39]. The
concept utilized copper split-ring resonators (SRRs), fabri-
cated on high-resistivity GaAs. When illuminated by ampli-
Controllable surfaces may prove to be beneficial for the fied laser light, conduction electrons were excited across the
development of components and devices at terahertz frequen- bandgap of the GaAs, making it semi-metallic. This shorted
cies. In essence, controllable metasurfaces are created by out the split-ring resonator’s capacitive gap, turning off the

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 23
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 23 6/3/2012 4:57:02 PM

 that amplitude modulation in this method is always accompanied
by phase modulation at different frequencies, according to the
Kramers-Kronig relations. As such, even though the split-ring
resonator resonance is narrowband, a modulation effect (in one
form or another) exists across a wide band. This led to the first
demonstration of a solid-state THz chopper, the modulation
frequency of which can be as high as 30 kHz [42], a significant
improvement over mechanical devices that are limited to
~1 kHz for THz beams. These examples vividly illustrated
dramatic improvements over existing technologies in switching
and modulation of freely propagating THz waves at room
temperature.

Many other THz modulators have been developed on the

controllable metasurface paradigm. These include a frequency-
agile metasurface, in which the resonance frequency shifts upon
stimulation by near-infrared light [43]. Here, a small region
of silicon is fabricated into the gap of a split-ring-resonator
structure (see Figure 16). In its unperturbed state, the silicon

Figure 15. The magnitude of the E field (on a linear scale)

from a magnetic line source placed 45.49 mm above an
array of spherical particles: (a) f = 1.42 GHz, one surface
wave; and (b) f = 1.5 GHz, complex mode.

fundamental resonance and changing the metasurface’s mac-

roscopic response. The concept was also demonstrated with
short-carrier-lifetime semiconductors, enabling ultra-fast
modulation [40]. These types of “switches” have demonstrated
switching speeds of 20 ps. Optical control was soon supple-
mented by electronic control schemes. Here, an array of gold
split-ring resonators formed a Schottky junction with a moder-
ately doped GaAs layer grown by molecular-beam epitaxy
(MBE). The entire metamaterial array was surrounded by an
ohmic contact, providing the second electrical connection.
In the unperturbed state, the metasurface had no resonance,
because the resonators were shorted by the doped GaAs. Upon
application of a reverse voltage bias, the split-ring resonators
formed an increasingly large depletion region in the GaAs near
their gap. This formed an insulating region around the split-
ring resonators, reestablishing their resonant response, and thus Figure 16. A split-ring resonator with silicon on the gap
changing the macroscopic properties [41]. It was realized later arms.

24 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 24 6/3/2012 4:57:02 PM

 is insulating, and thereby contributes only a small amount of of freely tailoring the electric and magnetic response of a
capacitance to the overall resonance. Upon optical stimulation, material at optical frequencies holds promise for new and
the region becomes semi-metallic and effectively increases the exotic optical phenomena, including optical magnetism [68],
size of the split-ring-resonator “capacitor,” thus lowering the a negative index of refraction [4, 69, 70], and ideal lensing
resonance frequency. The novelty here was that one need not [70]. At optical frequencies, plasmonic resonances, typically
destroy the resonance to alter it, as in the previous case. Instead, excited in gold and silver nanostructures, provide a means for
the resonance can be tuned. Not only did this demonstrate a controlling the electric and magnetic moments that collectively
dynamically tunable resonance, but it also enabled a new form are responsible for the behavior of a metamaterial. Examples
of frequency-shift keying, wherein broadband incident THz of such passive structures include coupled plasmonic nanorods
radiation is controllably converted to a narrowband output [69, 70] and spheres [72], voids in metal films [69], metallic
occupying one of two narrow frequency bins. fishnet structures [73], and double and single split-ring reso-
nators [68, 71]. Due to the inherent high loss associated with
Tunable THz metasurfaces continue to be developed plasmonic materials at optical frequencies, optical metamate-
for new applications and using new physical mechanisms. rials are severely limited in their practical application. Struc-
Amplitude modulators have been adapted to more-specific tures that incorporate gain media to overcome plasmonic
applications, such as spatial light modulators [44] and quan- losses have therefore also been investigated. Examples include
tum-cascade laser modulators [45]. Tunable metamaterials active plasmonic-coated nanoparticles [75], passive plasmonic
have also been created by other methods, such as the MEMS particles embedded in gain media [69], and gain-coated plas-
[46], wherein split-ring resonators are fabricated on bi-mate- monic nanoparticles [76]. Such active structures show the
rial cantilevers that reorient the split-ring resonator with potential for new and interesting properties and devices, such
respect to the incident wave upon a change in temperature. as optically-switchable frequency-selective surfaces [77], and
Other thermal tuning methods have been investigated. In one, the surface-plasmon amplification by stimulated emission of
temperature variations changed the intrinsic carrier density of a radiation (SPASER) [76]. Another investigating area is the use
semiconductor, thus altering its relative permittivity in the THz of optical metasurfaces and metamaterials for so-called nano-
range. Metasurfaces fabricated on such a substrate would be transmission lines [78, 79].
frequency-agile as a function of temperature [47]. In another
variation, vanadium dioxide served as the metasurface substrate Due to fabrication obstacles, optical metamaterials are
[48]. Upon temperature variations, vanadium dioxide undergoes often limited in construction to single-layer two-dimensional
a metal-insulator transition, dramatically changing its dielectric arrays of scatterers [69, 70, 80], i.e., metasurfaces. Much of the
properties, and thus the macroscopic metamaterial resonance. A published work on optical metamaterials thus actually describes
very interesting development called the memory metamaterial optical metasurfaces. As discussed above, defining bulk
[49] was developed from this work. Here, the hysteretic properties for these optical metasurfaces must be approached
nature of the vanadium dioxide metal-insulator transition with caution, and generalized sheet-transition conditions
enabled a persistence in the dielectric properties over a certain provide a preferred means for uniquely describing a two-
temperature range. The THz metasurface, being fabricated on dimensional array of scatters used for a large range of optical
this substrate, acquired this same persistent behavior in terms applications. They are therefore a more appropriate approach
of its resonance frequency, thus allowing a form of memory in for describing optical metasurfaces, rather than attempting to
the electromagnetic response. define and assign properties more appropriate for describing a
bulk material.
In still other work, THz metasurfaces have been dynami-
cally tuned on a much slower timescale by changing the envi- Even though most so-called optical metamaterials to date
ronment around the resonators through the application of a have in fact been metasurfaces, we should mention that recent
dielectric coating to the metasurface’s surface [50-52]. This advances have been made towards realizing truly three-dimen-
was investigated as a form of remote sensing, based on the idea sional optical metamaterials [80, 81], where bulk quantities
that the metasurface resonance is very sensitive to changes in such as the permittivity, permeability, and index of refraction
the dielectric environment, particularly around the split-ring can be appropriately and uniquely defined. Improved fabrica-
resonator’s gap. This idea has been extending microwave tion techniques have made possible the layering and stacking
frequencies where fluids have been used for controllable of metasurfaces into volume arrays of scatterers. For example,
surfaces, bio-sensing, and microwave-assisted chemistry. This negative refraction of light was recently observed in an optical
is discussed in detail below. prism made of negative-index material [81].

8. Optical Metasurfaces 9. Microfluidics for Controllable Surfaces,

Assisted Chemistry, and
Up to this point, we have discussed only RF, microwave, Biomedical Sensors
and THz applications of metasurfaces. A few comments are
in order on the use of optical-frequency metasurfaces. As at One of the disadvantages often mentioned about
RF and microwave frequencies, optical metamaterials have metamaterials (and metasurfaces) is that they have a very nar-
gained much attention over recent years. The realization row bandwidth of operation (less than a few percent) in the

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 25
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 25 6/3/2012 4:57:02 PM

 desired frequency range. While true, rather than being a limi- Experiments on a 3 x 6 array of these inclusions, fabri-
tation, this narrowband behavior can be utilized to advantage in cated from Duroid 6002 High-Frequency Laminate (mention
some applications. Three such applications are in the areas of of this product is not an endorsement, but serves only to spec-
fluid-controllable surfaces, microwave-assisted chemistry, and ify the material used) with 17 μm-thick copper laminated to a
biomedical sensors. 0.508 mm-thick substrate, was used to investigate fluid tuning.
Polymer tubing was used for fluid channels. The polymer
fluid channels were bonded to the metasurface by use of a
cyanoacrylate-based glue. A picture of the metasurface with the
9.1 Fluid-Controllable Surfaces

The highly resonant nature of metamaterials and metasur-

faces provides a means for tuning the frequency response of
these structures. Perturbations to the electric or magnetic
response of the inclusions making up a metasurface can be
modified in real time, thereby altering the effective material
response. This was discussed above in Section 2, where a
magnetic-bias field was used to change the polarizabilities of
magneto-dielectric spheres in a metasurface, and in Section 7
for THz applications. However, many metamaterials and
metasurfaces are constructed from metallic inclusions that
have plasma resonances governed largely by the choice of their
geometry. Aside from the specific geometry, the polarizabilities
of these metallic inclusions may also be controlled by affecting
the capacitive and or inductive properties that dictate the
plasma resonances. The electric-coupled resonator is a structure
that provides a means for directly perturbing the capacitive
response by altering the electrical properties of the material
occupying the electrical gap. One way to accomplish this is
to cause different fluids to flow in a channel over the gap: for Figure 17. An electric-coupled resonator (OE2 structure).
details, see [82].

Figure 17 shows a diagram of the type of electric-field-

coupled-resonator inclusion (referred to as an OE2 structure
[29]) that was used in designing a fluid-controllable metasur-
face for operation in the S band over 2.6 GHz to 3.9 GHz, with
the following dimensions: = t w= 0.5 mm, d = 9.5 mm, l = 5
mm, and g = 0.15 mm. This type of inclusion is easily excited
by the electric field of the incident wave, if the inclusions are
suitably oriented.

Being able to orient the inclusions in the plane of the

metasurface has several attractive features: (1) it is consistent
Figure 18. A fluid-tunable metafilm, showing the metafilm
with the planar fabrication techniques used for microwave
array and fluid-channel orientation.
circuits and polydimethylsilozane (PDMS) microfluid chan-
nels; (2) it allows the capacitive gaps for each inclusion to be
simultaneously aligned with each other as well as with the fluid
channel, therefore allowing multiple inclusions to share the
same fluid-channel path; and (3) it allows the fluid channels to
be in direct contact with the capacitive gaps of the inclusions,
which facilitates strong coupling of the fluid channels to the
electric field that is excited in the gaps.

A diagram of a fluid-tunable metasurface made of a 3 × 6

square array of inclusions is shown in Figure 18. The period
of the inclusions was 11 mm. The simulated response of this
metasurface placed in a 72 x 34 mm S-band waveguide is
shown in Figure 19. The simulations discussed here were per-
formed by use of Ansoft HFSS (mention of this product is not Figure 19. The simulated response of a metafilm composed
an endorsement, but serves only to identify the software used). of a 3 × 6 array of OE2 inclusions in an S-band waveguide.

26 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 26 6/3/2012 4:57:02 PM

 polymer fluid channels bonded and aligned across the gaps in
the inclusions is shown in Figure 20a. Figure 20b illustrates
a second structure fabricated from gold on glass with PDMS
(polydomethykiloxane) fluid-channel sections.

The 3 x 6 metasurface array was supported by a block of

polystyrene foam that filled two-thirds of the waveguide cross
section, and was used as a handle so that the metasurface could
be easily taken in and out, in order to facilitate quick filling of
the fluid channels. The input and output ports of the waveguide
were then connected to a calibrated vector network analyzer.
Employing the TRL (thru-reflect-line) calibration method
[110], we measured the scattering parameters of the waveguide
over the frequency range of 2.6 GHz to 3.95 GHz. The
measurement uncertainties in reflection coefficient resulting
from this calibration method were ∆ S11 = 0.02 (one standard
deviation).

Measurements with and without fluid in the channels were

performed to determine the tuning of the transmission
resonance. De-ionized water with a relative permittivity of
ε r = 81 was chosen to test the ability to tune the frequency of
the metasurface, due to its ease of handling, low volatility, high
dielectric constant, and low loss in the S band. Fluid channels
were manually filled by use of a syringe. Figure 21 shows the
metasurface inside the S-band waveguide during the experiment.

Experimental results are shown in Figure 22. They demon-

strated the ability to tune over a frequency range of 150 MHz.
The reflection dip was clearly shifted from 3.75 GHz down to
3.6 GHz between the cases without and with water in the fluid
channels, respectively. The disagreement between the positions
of the resonance dip for the case without water and that of the Figure 20. A fluid-tunable metafilm composed of a 3 × 6
simulated result in Figure 19 is believed to be due to inclusion- array of OE2 resonators: (a) fabricated from copper on
fabrication errors. The additional oscillations in the response are Duroid with three polymer-tubing fluid-channel sections;
also a result of fabrication nonuniformity and periodicity errors (b) fabricated from gold on glass with PDMS fluid-channel
of the inclusions, as well nonuniform perturbations across the sections.
metasurface due to errors in bonding the fluid channels. In the
future, fabrication improvements will be explored to mitigate
these errors.

In addition to using fluids for specifically tuning micro-

wave characteristics as focused on here, changes in resonance
characteristics may also be used as a means to monitor proc-
esses tied to the changes in the fluid flowing in the channels.
This approach has applications to manufacturing, industrial,
biomedical, and chemical processes. The use of metamaterials/
metasurfaces for influencing and sensing external process is
discussed in greater detail below.

9.2 Microwave-Assisted
Chemistry Applications

A metafilm operating in the resonant state stores electro-

magnetic energy in the modes of the inclusions. This property Figure 21. A 3 × 6 array of OE2 resonators fabricated from
can be exploited to increase the interactions experienced copper on Duroid with three polymer-tubing fluid-channel
between the electromagnetic fields and liquids carried by the sections placed in an S-band waveguide.

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 27
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 27 6/3/2012 4:57:03 PM

 Figure 22. Experimental results comparing S11 for the
copper on Duroid microfluidic metafilm for empty fluid
channels, and for fluid channels filled with de-ionized water
over the S band (2.6-3.95 GHz).

Figure 24. A split-ring-resonator coplanar-waveguide bio-

sensor concept.

Figure 23. The electric-field structure in the gap of the Figure 25. A simulated response, showing S11 for the split-
inclusions of the metasurface showing the enhancement of ring-resonator coplanar-waveguide bio sensor as the rela-
the electric field. The maximum field amplitude was tive permittivity of the fluid in the channel was reduced
1.25 × 105 V/m at 3.29 GHz. from 81 to 58.

28 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 28 6/3/2012 4:57:03 PM

 fluidic channels. There is currently an explosion of research erties of the fluid in the channels, the transmission character-
into processes that use microwave energy to catalyze chemical istics of the coplanar-waveguide structure can be tuned.
or biochemical reactions [111-113], and many of these types of
processes could benefit from the use of metafilms to increase Simulation results for this structure are shown in Fig-
the coupling between electromagnetic fields and fluids. The ure 25, where the relative permittivity of the fluid in the fluid
metafilm approach described here is particularly valuable for channel was changed from ε r = 81 (that of water) to lower
reactions that are implemented by use of a continuous-flow permittivities, down to ε r = 58 . As the fluid was changed, the
geometry, as chemical composition can be precisely controlled reflection coefficient was shifted in frequency over several tens
by controlling reactant flow rates, while the amount of energy of MHz. This ability to affect the resonance characteristics of a
transferred to the fluid can be precisely controlled by adjusting metamaterial-inspired structure using a fluid channel provides a
the frequency and power of the microwave excitation, in means for realizing a compact sensor, suitable for biomedical-
addition to controlling the flow rate of the fluid across the sensing applications such as counting cells or particles in a
metafilm. In addition, because the resonant frequency of the fluid. Future work includes experimental verification of this
metafilm array can be tailored by the geometry and arrangement type of sensor, as well as determining the minimum detectable
of the inclusions, different metafilms with different resonant change in permittivity, and the feasibility of counting particles
frequencies can be stimulated by a single variable-frequency with such a device. Other types of compact resonator structures
source within a single waveguide. The metasurface structure will be explored as well, which may be suitable for integrated-
used above to demonstrate fluid tuning was also investigated as sensor devices. We are currently investigating the possibility of
a means for concentrating electric fields in a waveguide [82]. using this concept for detecting and counting small particles in
fluids, e.g., counting white blood cells.
Simulations of the electric-field strength inside the
waveguide when the fundamental mode of the S-band
waveguide was excited were conducted. For a 1 W input at
10. Conclusion
3.29 GHz, the maximum field strength calculated was 8 × 102
V/m. For the same waveguide, but with the metasurface placed Metasurfaces offer an alternative to bulk three-dimen-
in the center, the maximum electric-field strength calculated in sional metamaterials. Because of the two-dimensional nature
the gaps of the electric-field coupled inclusions was 1.25 × 105 of the metasurface structures, they occupy less physical space
V/m. This demonstrated a field enhancement of at least two and can exhibit lower loss. We have pointed out that while
orders of magnitude, and an enhancement in the specific effective-property models are valid for three-dimensional
absorption rate (SAR) of at least four orders of magnitude. The metamaterials, the more appropriate way to characterize
field structure in the central gap of the inclusion is shown in metasurfaces/metafilms is by their effective surface electrical
Figure 23. and magnetic susceptibilities, where these surface suscepti-
bilities are related to the scatterers that compose the metasur-
The localized field structure shown in Figure 23 demon- face. In this overview paper, we have discussed various appli-
strated the possibility of using a metamaterial structure to cations of metasurfaces, ranging from microwave to optical
precisely deliver electromagnetic energy. The fluid-tunable frequencies. The applications discussed here are by no means
metasurface discussed above clearly demonstrated the interac- the only applications possible. However, they offer a few ideas
tion of a fluid channel with the field in the gaps of the inclusions of where metasurfaces can be used, and hopefully the paper
comprising the metasurface. It can therefore be envisioned that opens the door to the development of other new applications.
such a metasurface coupled to a fluidic network could be used The analysis tools presented here give one the means to model,
to precisely deliver microwave fields for assisting in chemical conceive, and generate new applications in the future.
processes confined within the fluid channels.
What about a one-dimensional metamaterial? The two-
dimensional metamaterial (i.e., a metasurface) concept can
9.3 Biosensor Applications be extended even further to the concept of using only a linear
unit cell, rather than a square, a rectangle, etc., that is, using
The concept of the fluid-tunable metasurface discussed only a single sub-wavelength resonant structure for some
above can be extended to realize highly resonant integrated desired effect or behavior. In fact, we have already begun to
and chip-level structures for sensor applications. Integrating see a few applications of this concept. One in particular is the
single inclusions used in realizing a metamaterial into a circuit use of a unit cell in the design of electrically small antennas:
or network may allow for compact sensors for use in biomedi- for details, see [114-121]. In the antenna application, the unit-
cal applications to count cells or monitor reactions in fluids. cell acts like a parasitic element to the radiating element of the
An example of such a device is shown in Figure 24. Two split- antenna, and serves as a means to match the electrically small
ring resonators are coupled to a coplanar-waveguide (CPW) radiating element to both (1) the feeding transmission line
structure. A fluid channel is coupled to the gaps in the split- and (2) free space. Such designs have been shown to achieve
ring resonators in much the same way as the fluid channels of efficient electrically small antennas [114-121]. Nanoparticles
the fluid-tunable metasurface above were used to affect the have also been used for tuning so-called optical nanoantennas
capacitance of the inclusions. By altering the electrical prop- [122]. Another example is using a one-dimensional unit cell as

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 29
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 29 6/3/2012 4:57:03 PM

 a tuning structure for planar transmission lines [123], as well 10. M. Lapine and S. Tretyakov, “Contemporary Notes on
as the sensor shown in Figure 24. Another emerging area of Metamaterials,” IET Microwaves, Antennas, and Propagation,
application is the use of one-dimensional chains of nanoparti- 1, 1, 2007, pp. 3-11.
cles as waveguides supporting surface waves, of which exam-
ples could be found in [124-130]. 11. D. V. Sivukhin, “The Energy of Electromagnetic Fields in
Dispersive Media” [in Russian], Opt. Spektrosk., 3, 1957, pp.
While much more work is still needed in the understand- 308-312.
ing, analysis, design, and fabrication of both metamaterials and
metasurfaces, we see that the development of such materials 12. G. D. Malyuzhinets, “A Note on the Radiation Principle”
and surfaces in recent years is bringing us closer to realizing [in Russian], Zh. Tekh. Fiz., 21, 1951, pp. 940-942.
the exciting predictions made over one hundred years ago by
the work of Lamb, Schuster, and Pocklington [18-20]. The 13. R. A. Silin, “Optical Properties of Artificial Dielectrics
potential of these materials has changed forever the landscape (Review)” [in Russian], Izv. VUZ Radiofiz., 15, 1972, pp. 809-
of RF, microwaves, optics, and photonics for the future. 820; English translation in Radiophys. Quantum Electron., 15,
1972, pp. 615-624.

14. R. A. Silin, “Possibility of Creating Plane-Parallel Lenses,”

11. References [in Russian], Opt. Spektrosk., 44, 1978, pp. 189-191; English
translation in Opt. Spectrosc., 44, 1978, pp. 109-110.
1. S. Zouhdi, A. Sihvola and M. Arsalane (eds.), Advances
in Electromagnetics of Complex Media and Metamaterials, 15. R. A. Silin and I. P. Chepurnykh, “On Media with Nega-
Boston , Kluwer Academic Publishers, 2002. tive Dispersion” [in Russian], Radiotekh. Elektron., 46, 2001,
pp. 1212-1217; English translation in J. Commun. Technol.
2. N. Engheta and R. W. Ziolkowski, Electromagnetic Meta­ Electron., 46, 2001, pp. 1121-1125.
materials: Physics and Engineering Explorations, Hoboken,
NJ, John Wiley & Sons, 2006. 16. L. I. Mandel’shtam, “Lectures on Certain Problems of
Oscillation Theory: Lecture 4” [in Russian], Polnoe Sobraniye
3. G. V. Eleftheriades and K. G. Balmain, Negative Refraction Trudov, 5, Leningrad, Izdat. Akad. Nauk SSSR, 1950, pp.
Metamaterials: Fundamental Principles and Applications, 461-467; also in his Lektsii po Optike Teorii Otnositel’nosti i
Hoboken, NJ, John Wiley & Sons, 2005. Kvantovoi Mekhanike, Moscow, Nauka, 1972, pp. 431-437.

4. V. G. Veselago, “The Electrodynamics of Substances with 17. L. I. Mandel’shtam, “Group velocity in crystalline arrays”
Simultaneously Negative Values of ε and µ ” [in Russian], [in Russian], Zh. Eksp. Teor. Fiz., 15, 1945, pp. 475-478; also
Usp. Fiz. Nauk, 92, 1967, pp. 517-526; English translation in in Polnoe Sobraniye Trudov, 2, Leningrad, Izdat. Akad. Nauk
Sov. Phys. Uspekhi, 10, 1968, pp. 509-514. SSSR, 1947, pp. 334-338.

5. D. R. Smith, W. J. Padilla, D. C. Vier, S. C. Nemat-Nasser 18. H. Lamb, “On Group-Velocity,” Proc. London Math. Soc,
and S. Schultz, “Composite Medium with Simultaneously Ser. 2, 1, 1904, pp. 473-479.
Negative Permeability and Permittivity,” Phys. Rev. Lett., 84,
2000, pp. 4184-4186. 19. A. Schuster, An Introduction to the Theory of Optics. Lon-
don, Edward Arnold, 1904, pp. 313-318.
6. R. Marques, J. Martel, F. Mesa and F. Medina, “A New
2D Isotropic Left-Handed Metamaterial Design: Theory and 20. H. C. Pocklington, “Growth of a Wave-Group when the
Experiment,” Micr. Opt. Technol. Lett., 35, 5, 2002, pp. 405- Group Velocity is Negative,” Nature, 71, 1905, pp. 607-608.
408.
21. L. Lewin, “The Electrical Constants of a Material Loaded
7. C. L. Holloway, E. F. Kuester, J. Baker-Jarvis and P. Kabos, with Spherical Particles,” J. IEE (London), III, 94, 1947, pp.
“A Double Negative (DNG) Composite Medium Composed of 65-68.
Magneto-Dielectric Spherical Particles Embedded in a Matrix,”
IEEE Transactions on Antennas and Propagation, AP-51, 10, 22. R. Gans and H. Happel, “Zur Optik kolloidaler Met-
2003, pp. 2596-2603. allösungen,” Ann. Physik, 29, 7, 1909, pp. 277-300.

8. A. Sihvola, “Metamaterials in Electromagnetics,” Metama­ 23. E. F. Kuester, M. A. Mohamed, M. Piket-May and C. L.

terials, 1, 1, 2007, pp. 2-11. Holloway, “Averaged Transition Conditions for Electromag-
netic Fields at a Metafilm,” IEEE Transactions on Antennas
9. E. Shamonina and L. Solymar, “Metamaterials: How the and Propagation, AP-51, 10, 2003, pp. 2641-2651.
Subject Started,” Metamaterials, 1, 1, 2007, pp. 12-18.

30 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 30 6/3/2012 4:57:03 PM

 24. C. L. Holloway, M. A. Mohamed, E. F. Kuester and A. 36. M. Caiazzo, S. Maci, and N. Engheta, “A Metamaterial
Dienstfrey, “Reflection and Transmission Properties of a Surface for Compact Cavity Resonators,” IEEE Antennas and
Metafilm: With an Application to a Controllable Surface Wireless Propagation Letters, 3, pp. 261-264, 2004.
Composed of Resonant Particles,” IEEE Trans. Electromag.
Compat., 47, 4, 2005, pp. 853-865. 37. C. L. Holloway, E. F. Kuester and D. Novotny, “Waveguides
Composed of Metafilms: The Two-Dimensional Equivalent
25. C. L. Holloway, P. Kabos, M. A. Mohamed, E. F. Kuester, of Metamaterials,” IEEE Antenna and Wireless Propagation
J. Gordon, M. D. Janezic and J. Baker-Jarvis, “Realization of Letters, 8, 2009, pp. 525-529.
a Controllable Metafilm/Metasurface Composed of Resonant
Magnetodielectric Particles: Measurements and Theory,” IET 38. H.-T. Chen, S. Palit, T. Tyler, C. M. Bingham, J. M. O. Zide,
Microwaves, Antennas, and Propagation, 4, 8, 2010, pp.1111- J. F. O’Hara, D. R. Smith, A. C. Gossard, R. D. Averitt, W. J.
1122. Padilla, N. M. Jokerst and A. J. Taylor, “Hybrid Metamaterials
Enable Fast Electrical Modulation of Freely Propagating
26. J. F. O’Hara, E. Smirnova, A. K. Azad, H.-T. Chen and A. Terahertz Waves,” Appl. Phys. Lett., 93, 2008, art. 091117.
J. Taylor, “Effects of Microstructure Variations on Macroscopic
Terahertz Metafilm Properties,” Active and Passive Electronic 39. W. J. Padilla, A. J. Taylor, C. Highstrete, M. Lee and R.
Components, 2007, 2007, art. 49691. D. Averitt, “Dynamical Electric and Magnetic Metamaterial
Response at Terahertz Frequencies,” Phys. Rev. Lett., 96, 2006,
27. W. J. Padilla, D. N. Basov and D. R. Smith, “Negative art.107401.
Refractive Index Metamaterials,” Materials Today, 9, 7, 2006,
pp. 28-35. 40. H.-T. Chen, W. J. Padilla, J. M. O. Zide, S. R. Bank, A.
C. Gossard, A. J. Taylor and R. D. Averitt, “Ultrafast Optical
28. H. O. Moser, B. D. F. Casse, O. Wilhelmi and B. T. Saw, Switching of Terahertz Metamaterials Fabricated on ErAs/
“Terahertz Response of a Microfabricated Rod-Split-Ring- GaAs Nanoisland Superlattices,” Opt. Lett., 32, 12, 2007, pp.
Resonator Electromagnetic Metamaterial,” Phys. Rev. Lett., 94, 1620-1622.
2005, art. 063901.
41. H.-T. Chen, W. J. Padilla, J. M. O. Zide, A. C. Gossard,
29. W. J. Padilla, M. T. Aronsson, C. Highstrete, M. Lee, A. A. J. Taylor and R. D. Averitt, “Active Terahertz Metamaterial
J. Taylor and R. D. Averitt, “Electrically Resonant Terahertz Devices,” Nature, 444, 2006, pp. 597-600.
Materials: Theoretical and Experimental Investigations,” Phys.
Review B, 75, 2007, 041102. 42. H.-T. Chen, W. J. Padilla, M. J. Cich, A. K. Azad, R. D.
Averitt and A. J. Taylor, “A Metamaterial Solid-State Terahertz
30. D. R. Smith, D. Schurig and J. J. Mock, “Characterization Phase Modulator,” Nature Photonics, 3, 2009, pp. 148-151.
of a Planar Artificial Magnetic Metamaterial Surface,” Phys.
Review E, 74, 2006, art. 036604. 43. H.-T. Chen, J. F. O’Hara, A. K. Azad, A. J. Taylor, R. D.
Averitt, D. B. Shrekenhamer and W. J. Padilla, “Experimental
31. C. Highstrete, M. Lee and W. Padilla, “Complementary Demonstration of Frequency-Agile Terahertz Metamaterials,”
Planar Terahertz Metamaterials,” Optics Express, 15, 3, 2007, Nature Photonics, 2, 2008, pp. 295-298.
pp. 1084-1095.
44. W. L. Chan, H.-T. Chen, A. J. Taylor, I. Brener, M. J. Cich
32. Z. Q. Li, G. M. Wang, K. J. Mikolaitis, D. Moses, A. J. and D. M. Mittleman, “A Spatial Light Modulator for Terahertz
Heeger and D. N. Basov, “An Infrared Probe of Tunable Beams,” Appl. Phys. Lett., 94, 2009, art. 213511.
Dielectrics in Metal-Oxide-Semiconductor Structures,” Appl.
Phys. Lett., 86, 2005, art. 223506. 45. X. G. Peralta, I. Brener, W. J. Padilla, E. W. Young, A. J.
Hoffman, M. J. Cich, R. D. Averitt, M. C. Wanke, J. B. Wright,
33. W. J. Padilla, A. J. Taylor, C. Highstrete, M. Lee and R. H.-T. Chen, J. F. O’Hara, A. J. Taylor, J. Waldman, W. D.
D. Averitt, “Dynamical Electric and Magnetic Metamaterial Goodhue, J. Li and J. Reno, “External Modulators for Terahertz
Response at Terahertz Frequencies,” Phys. Review Lett., 96, Quantum-Cascade Lasers Based on Electrically-Driven Active
2006, art. 107401. Metamaterials,” Metamaterials, 4, 2, 2010, pp. 83-88.

34. H-T Chen, W. J. Padilla, J. M. O. Zide, S. R. Bank, A. 46. H. Tao, A. C. Strikwerda, K. Fan, W. J. Padilla, X. Zhang
C. Gossard, A. J. Taylor and R. D. Averitt, “Ultrafast Optical and R. D. Averitt, “Reconfigurable Terahertz Metamaterials,”
Switching of Terahertz Metamaterials Fabricated on ErAs/ Phys. Rev. Lett., 103, 2009, art. 147401.
GaAs Nanoisland Superlattices,” Optics Lett., 32, 12, 2007, pp.
1620-1622. 47. J. Han and A. Lakhtakia, “Semiconductor Split-Ring
Resonators for Thermally Tunable Terahertz Metamaterials,” J.
35. C. L. Holloway, D. C. Love, E. F. Kuester, A. Salandrino Mod. Optics, 56, 2009, pp. 554-557.
and N. Engheta, “Sub-Wavelength Resonators: On the Use of
Metafilms to Overcome the λ/2 Size Limit,” IET Microwaves, 48. T. Driscoll, S. Palit, M. M. Qazilbash, M. Brehm, F.
Antennas, and Propagation, 2, 2, 2008, pp. 120-129. Keilmann, B.-G. Chae, S.-J. Yun, H.-T. Kim, S. Y. Cho, N. M.

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 31
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 31 6/3/2012 4:57:04 PM

 Jokerst, D. R. Smith and D. N. Basov, “Dynamic Tuning of 60. J. Gordon, C. L. Holloway and A. Dienstfrey, “A Physical
an Infrared Hybrid-Metamaterial Resonance Using Vanadium Explanation of Angle-Independent Reflection and Transmis-
Dioxide,” Appl. Phys. Lett., 93, 2008, art. 024101. sion Properties of Metafilms/Metasurfaces,” IEEE Antennas
Wireless Propag. Lett., 8, 2009, pp. 1127-1130.
49. T. Driscoll, H.-T. Kim, B.-G. Chae, B.-J. Kim, Y.-W. Lee, N.
M. Jokerst, S. Palit, D. R. Smith, M. Di Ventra and D. N. Basov, 61. E. F. Kuester and C. L. Holloway, “A Low-Frequency
“Memory Metamaterials,” Science, 325, 2009, pp. 1518-1521. Model for Wedge or Pyramid Absorber Arrays – I: Theory,”
IEEE Trans. Electromag. Compat., 36, 4, 1994, pp. 300-306.
50. T. Driscoll, G. O. Andreev, D. N. Basov, S. Palit, S. Y.
Cho, N. M. Jokerst and D. R. Smith, “Tuned Permeability in 62. J. Fang, “Generalized Perfectly Matched Layers for the
Terahertz Split-Ring Resonators for Devices and Sensors,” Absorption of Propagating and Evanescent Waves in Lossless
Appl. Phys. Lett., 91, 2007, art. 062511. and Lossy Media,” IEEE Trans. Microwave Theory Tech., 44,
12, 1996, pp. 2216-2222.
51. J. F. O’Hara, R. Singh, I. Brener, E. Smirnova, J. Han, A. J.
Taylor and W. Zhang, “Thin-Film Sensing with Planar Terahertz 63. N. Engheta, “An Idea for Thin Subwavelength Cavity
Metamaterials: Sensitivity and Limitations,” Opt. Express, 16, Resonators Using Metamaterials with Negative Permittivity
2, 2008, pp. 1786-1795. and Permeability,” IEEE Antennas and Wireless Propagation
Letters, 1, 2002, pp. 10-13.
52. S. Y. Chiam, R. Singh, J. Gu, J. Han, W. Zhang and A.
A. Bettiol, “Increased Frequency Shifts in High Aspect Ratio 64. T. Hand, S. Cummer, and N. Engheta, “The Measured
Terahertz Split Ring Resonators,” Appl. Phys. Lett., 94, 2009, Electric Field Spatial Distribution within a Metamaterial Sub-
art. 064102. Wavelength Cavity Resonator,” IEEE Transactions on Anten­
nas and Propagation, AP-55, 6, 207, pp. 1781-1788.
53. S. A. Tretyakov and S. I. Maslovski, “Thin Absorbing
Structure for all Incident Angles Based on the Use of a High- 65. J. B. Pendry. D. Schurig and D. R. Smith, “Controlling
Impedance Surface,” Microwave and Optical Technology Electromagnetic Fields,” Science, 312, 2006, pp. 1780-1782.
Letters, 38, 3, 2003, pp. 175-178.
66. D. Schurig, J. J. Mock, B. J. Justice, S. A. Cummer, J. B.
54. Y. Kotsuka, K. Murano, M. Amano and S. Sugiyama, Pendry, A. F. Starr and D. R. Smith, “Metamaterial Electro-
“Novel Right-Handed Metamaterial Based on the Concept of magnetic Cloak at Microwave Frequencies,” Science, 314,
‘Autonomous Control System of Living Cells,’ and its Absorber 2006, pp. 977-980.
Applications,” IEEE Trans. Electromag. Compat., 52, 3, 2010,
pp. 556-565. 67. C. Enkrich, M. Wegener, S. Linden, S. Burger, L. Zschiedrich,
F Schmidt, J. F. Zhou, Th. Koschny and C. M. Soukoulis,
55. Hu Tao, C. M. Bingham, A. C. Strikwerda, D. Pilon, C. “Magnetic Metamaterials at Telecommunications and Visible
Shrekenhamer, N. I. Landy, K. Fan, X. Zhang, W. J. Padilla Frequencies,” Phys. Rev. Lett., 95, 2005, art. 203901.
and R. D. Averitt, “Highly Flexible Wide Angle of Incidence
Terahertz Metamaterial Absorber: Design, Fabrication, and 68. T. A. Klar, A. V. Kildishev, V. P. Drachev and V. M. Shalaev,
Characterization,” Phys. Rev. B., 78, 2008, art. 241103. “Negative-Index Metamaterials: Going Optical,” IEEE Journal
of Selected Topics in Quantum Electronics, 12, 6, 2006, pp.
56. N. I. Landy, S. Sajuyigbe, J. J. Mock, D. R. Smith and W. J. 1106-1115.
Padilla, “Perfect Metamaterial Absorber,” Phys. Rev. Lett., 100,
2008, art. 207402. 69. A. V. Kildishev, W. Cai, K. Chettiar, H. Yuan, A. K. Sarychev,
V. P. Drachev and V. M. Shalaev, “Negative Refractive Index in
57. O. Luukkonen, F. Costa, C. R. Simovski, A. Monorchio and Optics of Metal-Dielectric Composites,” J. Opt. Soc. Am. B, 23,
S. A. Tretyakov, “A Thin Electromagnetic Absorber for Wide 3, 2006, pp. 423-433.
Incidence Angles and Both Polarizations,” IEEE Transactions
on Antennas and Propagation, AP-57, 10, 2009, pp. 3119-3125. 70. J. B. Pendry, “Negative Refraction Makes a Perfect Lens,”
Phys. Rev. Lett., 85, 2000, pp. 3966-3969.
58. F. Bilotti, A. Toscano, K. B. Alici, E. Ozbay, and L. Vegni,
“Design of Miniaturized Narrowband Absorber Based on 71. A. Alù, A. Salandrino and N. Engheta, “Negative Effective
Resonant-Magnetic Inclusions,” IEEE Trans. Electromag. Permeability and Left-Handed Materials at Optical Frequen-
Compat., 53, 1, 2011, pp. 63-72. cies,” Opt. Exp., 14, 4, 2006, pp. 1557-1567.

59. S. Sajuyigbe, M. Ross, P. Geren, S. A. Cummer, M. H. 72. G. Dolling, C. Enkrick, M. Wegener, C. M. Soukoulis and
Tanielian and D. R. Smith, “Wide Angle Impedance Matching S. Linden, “Low-Loss Negative-Index Metamaterials at Tele-
Metamaterials for Waveguide-Fed Phased-Array Antenna,” IET communication Wavelengths,” Opt. Lett., 31, 12, 2006, pp.
Microwaves, Antennas, and Propagation, 4, 8, 2010, pp. 1063- 1800-1802.
1072.

32 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 32 6/3/2012 4:57:04 PM

 73. T. J. Yen, W. J. Padilla, N. Fang, D. C. Vier, D. R. Smith, 88. C. M. Soukoulis, Photonic Band Gap Materials, Dordrecht,
J. B. Pendry, D. N. Basov and X. Zhang, “Terahertz Magnetic The Netherlands, Kluwer Academic Publishers, 1996.
Response from Artificial Materials,” Science, 303, 2004, pp.
1494-1496. 89. E. Sanchez-Palencia, Non­Homogeneous Media and
Vibration Theory, Berlin, Springer-Verlag, 1980 (Lecture Notes
74. J. A. Gordon and R. W. Ziolkowski, “The Design and in Physics No. 127).
Simulated Performance of a Coated Nano-Particle Laser,”
Optics Express, 15, 5, 2007, pp. 2622-2653 90. G. Bouchitté and B. Schweizer, “Homogenization of
Maxwell’s Equations in a Split Ring Geometry,” Multiscale
75. M. I. Stockman, “Spasers Explained,” Nature Photonics, 2, Modeling Simulation, 8, 3, 2010, pp. 717-750.
2008, pp. 327-329.
91. N. A. Khizhnyak, “Artificial Anisotropic Dielectrics: I,
76. J. A. Gordon and R. W. Ziolkowski, “CNP Optical Meta- II and III,” Zh. Tekh. Fiz., 27, pp. 2006-2013, 2014-2026 and
materials,” Optics Express, 16, 9, 2008, pp. 6692-6716. 2027-2037, 1957 [in Russian]; English translation in Sov. Phys.
Tech. Phys., 2, pp. 1858-1864, 1865-1876 and 1877-1886, 1957.
77. V. M. Shalaev, “Optical Negative-Index Metamaterials,”
Nature Photonics, 1, 2007, pp. 41-48. 92. M. J. Dignam and M. Moskovits, “Optical Properties of
Sub-Monolayer Molecular Films,” J. Chem. Soc. Faraday
78. A. Alu and N. Engheta, “Theory of Linear Chains of Trans. II, 69, 1973, pp. 56-64.
Metamaterials/Plasmonic Particles as Subdiffraction Optical
Nanotransmission Lines,” Physical Review B, 74, 2006, art. 93. M. Johansson, C. L. Holloway, and E. F. Kuester, “Effec-
205436. tive Electromagnetic Properties of Honeycomb Composites,
and Hollow Pyramidal and Alternating Wedge Absorbers,”
79. A. Alu and N. Engheta, “Three-Dimensional Nanotrans- IEEE Transactions on Antennas and Propagation, AP-53, 2,
mission Lines at Optical Frequencies: A Recipe for Broadband February 2005, pp. 728-736.
Negative-Refraction Optical Metamaterials,” Physical Review
B, 75, 2007, art. 024304. 94. A. M. Nicolson and G. Ross, “Measurement of the Intrin-
sic Properties of Materials by Time Domain Techniques,” IEEE
80. N. Liu, H. Guo, L. Fu, S. Kaiser, H. Schweizer and H. Trans. Instrum. Meas., 19, 4, 1970, pp. 377-382.
Giessen, “Three-Dimensional Photonic Metamaterials at Opti-
cal Frequencies,” Nature Materials, 7, 2008, pp. 31-37. 95. W. B. Weir, “Automatic Measurements of Complex
Dielectric Constant and Permeability at Microwave Frequen-
81. J. Valentine, S. Zhang, T. Zentgraf, E. Ulin-Avila, D. A. cies,” Proc. IEEE, 62, 1, 1974, pp. 33-36.
Genov, G. Bartal and X. Zhang, “Three-Dimensional Optical
Metamaterial with a Negative Refractive Index,” Nature, 455, 96. D. R. Smith, S. Schultz, P. Markos and C. M. “Soukoulis,
2008, pp. 376-379. “Determination of Effective Permittivity and Permeability of
Metamaterials from Reflection and Transmission Coefficients,”
82. J. Gordon, C. L. Holloway and J. Booth, “Fluid Interac- Phys. Review B, 65, 2002, art. 195104.
tions with Metafilms/Metasurfaces for Tuning, Sensing, and
Microwave Assisted Chemical Processes,” Physical Review B, 97. R. W. Ziolkowski, “Designs, Fabrication, and Testing
83, 2011, art. 205130. of Double Negative Metamaterials,” IEEE Transactions on
Antennas and Propagation, AP-51, 7, 2003, pp. 1516-1529.
83. S. Tretyakov, Analytical Modeling in Applied Electromag­
netics, Norwood, MA, Artech House, 2003. 98. X. Chen, T. M. Grezegorczyk, B.-I. Wu, J. Pacheco and J.
A. Kong, “Robust Method to Retrieve the Constitutive Effec-
84. L. Brillouin, Wave Propagation in Periodic Structures, New tive Parameters of Metamaterials,” Phys. Review E, 70, 2004,
York, Dover, 1953, Chapter 7. art. 016608.

85. A. Ishimaru, Electromagnetic Wave Propagation, Radia­ 99. S. Kim, E. F. Kuester, C. L. Holloway, A. D. Scher and
tion, and Scattering, Englewood Cliffs, NJ, Prentice Hall, 1991, J. Baker-Jarvis, “Boundary Effects on the Determination of
Chapter 7. Metamaterial Parameters from Normal Incidence Reflection
and Transmission Measurements,” IEEE Transactions on
86. J. A. Kong, Electromagnetic Wave Theory, New York, John Antennas and Propagation, AP-59, 6, 2011, pp. 2226-2240.
Wiley & Sons, 1986, Chapter 6.
100. C. L. Holloway, A. Dienstfrey, E. F. Kuester, J. F. O’Hara,
87. T. K. Wu, Frequency Selective Surface and Grid Array, A. K. Azad and A. J. Taylor, “A Discussion on the Interpretation
New York, John Wiley & Sons, 1995. and Characterization of Metafilms/Metasurfaces: The Two-
Dimensional Equivalent of Metamaterials,” Metamaterials, 3,
2, 2009, pp. 100-112.

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 33
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 33 6/3/2012 4:57:04 PM

 101. C. L. Holloway, E. F. Kuester, and A. Dienstfrey, “Char- 113. C. Fermér, P. Nilsson, and M. Larhed, “Microwave-
acterizing Metafilms: The Connection Between Surface Sus- Assisted High-Speed PCR,” European Journal of Pharma­
ceptibilities and Effective Material Properties.” IEEE Antennas ceutical Sciences, 18, 2, 2002, pp. 129-132.
and Wireless Propagation Letters, 2011.
114. R. W. Ziolkowski, P. Jin and C.-C. Lin, “Metamaterial-
102. L. A. Wainstein, “On the Electrodynamic Theory of Inspired Engineering of Antennas,” Proc. IEEE, 99, 10, 2011,
Grids,” in P. L. Kapitsa and L. A. Wainstein (eds.), High­Power pp. 1720-1731.
Electronics 2, Oxford, UK, Pergamon Press, 1966, pp. 14-48.
115. P. Jin and R. W. Ziolkowski, “Metamaterial-Inspired,
103. C. L. Holloway, E. F. Kuester, and A. Dienstfrey, “On Electrically Small, Huygens Sources,” IEEE Antennas and
the Use of the Homogenization Technique to Determine a Wireless Propagation Letters, 9, 2010, pp. 501-505.
Generalized Boundary Conditions for Arbitrary Shaped Coated
Wire Grading,” Radio Science, 2011 (in preparation). 116. P. Jin and R. W. Ziolkowski, “Multiband Extensions of the
Electrically Small Metamaterial-Engineered Z Antenna,” IET
104. T. B. A. Senior and J. L. Volakis, Approximate Boundary Microwaves, Antennas & Propagation, 4, 8, 2010, pp. 1016-
Conditions in Electromagnetics, London, UK, IEE, 1995. 1025.

105. M. I. Oksanen, S. A. Tretyakov and I. V. Lindell, “Vector 117. C.-C. Lin and R. W. Ziolkowski, “Dual-Band 3D Mag-
Circuit Theory for Isotropic and Chiral Slabs,” J. Electromag. netic EZ Antenna,” Micr. Opt. Technol. Lett., 52, 4, 2010, pp.
Waves Appl., 4, 1990, pp. 613-643. 971-975.

106. S. A. Tretyakov, A. J. Viitanen, S. I. Maslovki and I. E. 118. P. Jin and R. W. Ziolkowski, “Broadband, Efficient,
Saarela, “Impedance Boundary Conditions for Regular Dense Electrically Small Metamaterial-Inspired Antennas Facilitated
Arrays of Dipole Scatterers,” IEEE Transactions on Antennas by Active Near-Field Resonant Parasitic Elements,” IEEE
and Propagation, AP-51, 8, 2003, pp. 2073-2078. Transactions on Antennas and Propagation, AP-58, 2, 2010,
pp. 318-327.
107. O. Luukkonen, C. Simovski, G. Granet, G. Goussetis, D.
Lioubtchenko, A. V. Räisänen and S. A. Tretyakov, “Simple 119. P. Jin and R. W. Ziolkowski, “Low Q, Electrically Small,
and Accurate Analytical Model of Planar Grids and High- Efficient Near Field Resonant Parasitic Antennas,” IEEE
Impedance Surfaces Comprising Metal Strips or Patches,” Transactions on Antennas and Propagation, AP-57, 9, 2009,
IEEE Transactions on Antennas and Propagation, AP-56, 6, pp. 2548-2563.
2008, pp. 1624-1632; “Correction,” IEEE Transactions on
Antennas and Propagation, AP-58, 2010, p. 2162. 120. A. Erentok and R. W. Ziolkowski, “Metamaterial-Inspired
Efficient Electrically-Small Antennas,” IEEE Transactions on
108. A. D. Scher and E. F. Kuester, “Extracting the Bulk Antennas and Propagation, AP-56, 3, 2008, pp. 691-707.
Effective Parameters of a Metamaterial via the Scattering from
a Single Planar Array of Particles,” Metamaterials, 3, 1, 2009, 121. R. W. Ziolkowski and A. Erentok, “Metamaterial-Based
pp. 44-55. Efficient Electrically Small Antennas,” IEEE Transactions on
Antennas and Propagation, AP-54, 7, 2006, pp. 2113-2130.
109. C. L. Holloway, D. Love, E. F. Kuester, J. Gordon,
and D. A. Hill, “Use of GSTCs to Model Guided Waves on 122. A. Alu and N. Engheta, “Tuning the Scattering Response
Metasurfaces/Metafilms,” IEEE Transactions on Antennas and of Optical Nanoantennas with Nanocircuit Loads,” Nature
Propagation, submitted 2011. Photonics, 2, May 2008, pp. 307-310.

110. J. Baker-Jarvis, M. D. Janezic, J. H. Grosvenor and R. 123. I. A. Ibrahheem and M. Koch, “Coplanar Waveguide
G. Geyer, “Transmission/Reflection and Short-Circuit Line Metamaterials: The Role of Bandwidth Modifying Slots,” Appl.
Methods for Measuring Permittivity and Permeability,” NIST Phys. Lett., 91, 2007, art. 113517.
Technical Note 1355-R, NIST, US Department of Commerce,
1993. 124. M. Quinten, A. Leitner, J. R. Krenn and F. R. Aussenegg,
“Electromagnetic Energy Transport via Linear Chains of Silver
111. B. Wathey, J. Tierney, P. Lidström and J. Westman, “The Nanoparticles,” Opt. Lett., 23, 17, 1998, pp. 1331-1333.
Impact of Microwave-Assisted Organic Chemistry on Drug
Discovery,” Drug Discovery Today, 7, 6, 2002, pp. 373-380. 125. S. A. Tretyakov and A. J. Viitanen, “Line of Periodically
Arranged Passive Dipole Scatterers,” Elec. Eng., 82, 6, 2000,
112. C. Oliver Kappe and D. Dallinger, “The Impact of pp. 353-361.
Microwave Synthesis on Drug Discovery,” Nature Reviews
Drug Discovery, 5, 2006, pp. 51-63.

34 IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 34 6/3/2012 4:57:04 PM

 126. M. L. Brongersma, J. W. Hartman and H. A. Atwater,
“Electromagnetic Energy Transfer and Switching in Nanopar-
ticle Chain Arrays Below the Diffraction Limit,” Phys. Rev. B,
Changes of Address
62, 2000, pp. 16356-16359.
or Delivery Problems
127. W. H. Weber and G. W. Ford, “Propagation of Optical
Excitations by Dipolar Interactions in Metal Nanoparticle Information regarding subscriptions and addresses is
Chains,” Phys. Rev. B, 70, 2004, art. 125429. man­aged by IEEE headquarters. It is not maintained, nor can
it be changed, by any member of the Magazine staff. If you
128. R. A. Shore and A. D. Yaghjian, “Traveling Electromag- are a mem­ber of the IEEE, your subscription is sent to the
netic Waves on Linear Periodic Arrays of Lossless Spheres,” address in your IEEE mem­ber record. Your address can be
Electron. Lett., 41, 10, 2005, pp. 578-580. confirmed or updated by visiting the Web page dealing with
delivery of IEEE publica­tions:
129. R. A. Shore and A. D. Yaghjian, “Traveling Electromag-
netic Waves on Linear Periodic Arrays of Lossless Penetrable http://www.ieee.org/about/help/subscriptions.html
Spheres,” IEICE Trans. Commun., E88-B, 6, 2005, pp. 2346-
This page also has information about publication delivery,
2352.
and a link to an online form that can be used to inquire about
missing or delayed publications:
130. M. Guasoni and C. De Angelis, “Analytical Approxima-
tions of the Dispersion Relation of a Linear Chain of http://www.ieee.org/about/help/publication_inquiry.html
Nanoparticles,” Opt. Commun., 284, 7, 2011, pp. 1822-1827.
You can also update your address information by con­
tacting IEEE headquar­ters: Member Address Records, IEEE
Headquarters, 445 Hoes Lane, Piscataway NJ 08855-1331
USA; Tel: +1 (908) 981-0060 or +1 (800) 678-4333; Fax:
+1 (908) 981-9667; E-mail: [email protected]. If you
are an institutional or other non-member subscriber, contact
IEEE Customer Service at the above address, telephone, and
fax numbers; E-mail: [email protected].

Please do not send requests related to the above items

to any member of the Magazine Staff.

Getting the Magazine

by Air Freight
Air freight delivery of the IEEE Antennas and
Propagation Magazine is available to subscribers in IEEE
Regions 8-10. The cost was recently $44.00 per year. This
can be added to a member’s subscription (and the current cost
verified) via the Web at http://www.ieee.org/web/aboutus/
help/member_support.html.

IEEE Antennas and Propagation Magazine, Vol. 54, No. 2, April 2012 35
Authorized licensed use limited to: University Town Library of Shenzhen. Downloaded on March 22,2024 at 05:58:37 UTC from IEEE Xplore. Restrictions apply.

AP_Mag_Apr_2012_Final.indd 35 6/3/2012 4:57:04 PM