0% found this document useful (0 votes)

9 views

Exploring the Potentials of the Multi-modal Equivalent Circuit Approach for Stacks of 2-D Aperture Arrays-IEEE TAP.2021

This article discusses the multi-modal equivalent circuit approach (ECA) for analyzing and designing stacks of 2-D aperture arrays, highlighting the benefits of stacking conducting sheets and breaking alignment for enhanced performance in various applications. The ECA allows for rigorous analysis of scattering properties and dispersion characteristics of these structures, including the effects of glide symmetry. The paper aims to provide a systematic formulation that can be applied to both aligned and non-aligned configurations, facilitating the design of advanced electromagnetic devices.

Uploaded by

Vinay Shetty

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

9 views

Exploring the Potentials of the Multi-modal Equivalent Circuit Approach for Stacks of 2-D Aperture Arrays-IEEE TAP.2021

Uploaded by

Vinay Shetty

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 15

This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TAP.2021.3070150, IEEE
Transactions on Antennas and Propagation
IEEE , VOL. X, NO. X, 2021 1

Exploring the Potentials of the Multi-modal

Equivalent Circuit Approach for Stacks of 2-D
Aperture Arrays
Antonio Alex-Amor, Francisco Mesa, Fellow, IEEE, Ángel Palomares-Caballero, Carlos Molero, Member, IEEE,
and Pablo Padilla

Abstract—Many frequency selective surface (FSS) structures adjusting the geometrical parameters of the structure. This
are based on the use of a single periodic array of slot/apertures in key feature has found multiple applications in science and
a conducting sheet embedded in a layered medium. However, it is engineering, such as FSS [9]–[11], polarizers [12]–[14],
well known that stacking several conducting sheets and breaking
the alignment of the stack can bring multiple benefits to the absorbers [5], [15], [16], high-impedance surfaces [17], [18]
structure. In this paper, the analysis and design of stacks of 2-D and electromagnetic bandgap (EBG) devices [19], [20].
aperture arrays are carried out by exploiting as much as possible A subclass of structured surfaces that are of particu-
all the potentialities of a rigorous and systematic formulation lar interest in electromagnetism are one-dimensional (1-D)
based on the multi-modal equivalent circuit approach (ECA). A strip/slit gratings [21], [22] as well as two-dimensional (2-D)
key feature of the formulation is that linear transformations be-
tween the apertures of adjacent plates (rotation, translation, and periodic arrangements of metal patches and/or perforated
scaling) can be dealt with from a purely analytical perspective. apertures [23]–[30] in a layered medium. For these periodic
This fact is of potential interest for many practical applications, structures, the scattering properties associated with an inci-
such as the design of polarization converters, absorbers, filters, dent plane wave can be derived from a general waveguide
and thin matching layers. When the apertures have an arbitrary discontinuity problem where periodic boundary conditions are
geometry, it can be applied a hybrid approach that combines the
ability of commercial simulators to handle arbitrary geometries applied [31], [32]. From this fact, it directly follows that the
with the fast computation times and physical insight of the ECA. scattering problem can be analyzed in a rigorous manner from
In general, either the purely analytical or the hybrid approach a circuit model perspective [33]–[37].
can be applied in those many practical scenarios where the Most of the structures analyzed in the previous references
spatial profile of the electric field on the considered apertures consist of a single metal layer embedded in a layered dielectric
hardly changes with frequency. As an additional feature of the
approach, the dispersion properties (phase/attenuation constants medium. Nonetheless, it is well known that stacking several
and Bloch impedance) of infinite periodic stacks can be derived metal layers opens new possibilities to the design, such as
and, in particular, analytical expressions for mirror- and glide- the existence of transmission and rejection bands, increase
symmetric configurations are provided. of the operating bandwidth, appearance of negative-index
Index Terms—Equivalent circuit approach, 3-D periodic refraction bands, enhanced performance of polarization con-
stacks, dispersion analysis, frequency selective surface (FSS), verters, etc [27], [38]–[40]. Some equivalent circuits have been
glide symmetry, lossy materials, metamaterials, analytical treat- proposed to model the performance of stacked structures [14],
ment. [28], [29], [41]–[43]. However, some of these works (for
instance, [14], [28], [29], [43]) require a substantial assistance
I. I NTRODUCTION of previous full-wave simulations and the scope of some others
(for instance, [41], [42]) is focused to very particular con-
TRUCTURED surfaces have attracted a lot of attention
S both in microwaves [1]–[4], THz [5], [6] and the optical
range [7], [8] due to their versatility to control the reflection,
figurations and their formulation mainly based on a heuristic
rationale. In these latter works, the proposed circuit models fail
to take into account the strong coupling between the stacked
refraction, and diffraction of the impinging waves by simply layers when these are closely spaced. In the present work,
we are interested in an equivalent circuit approach (ECA)
Manuscript received X, 2020; revised X, 2020.
This work was supported by the Spanish Research and Development Na- that manages to include the coupling effects while offering
tional Program under Projects TIN2016-75097-P, RTI2018-102002-A-I00, B- a good physical insight on the scattering problem, and all of
TIC-402-UGR18, TEC2017-84724-P, and the predoctoral grant FPU18/01965; this by means of an analytical procedure [34]. This kind of
by Junta de Andalucı́a under project P18-RT-4830.
A. Alex-Amor, A. Palomares-Caballero, C. Molero and P. Padilla are insightful works starts from basic electromagnetic principles
with the Departamento de Teorı́a de la Señal, Telemática y Comunicaciones, and even/odd excitation techniques [44], [45] to study aligned
Universidad de Granada, 18071 Granada, Spain (email: [email protected], an- and symmetric configurations, which nonetheless is a limiting
[email protected]; [email protected].
A. Alex-Amor is also with the Information Processing and Telecommu- factor of the functionalities of the stacked structure.
nications Center, Universidad Politécnica de Madrid, 28040 Madrid, Spain Recently, the range of use of these more rigorous equivalent
(e-mail: [email protected]) circuits was successfully extended in [46] to model aligned
F. Mesa is with the Microwaves Group, Department of Applied Physics 1,
Escuela Técnica Superior de Ingenieria Informatica, Universidad de Sevilla, stacks of apertures. Interestingly, breaking the alignment of
41012 Sevilla, Spain; (e-mail: [email protected]) the stacked metal layers can lead to enhanced performances of

0018-926X (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: San Francisco State Univ. Downloaded on June 24,2021 at 23:18:57 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TAP.2021.3070150, IEEE
Transactions on Antennas and Propagation
IEEE , VOL. X, NO. X, 2021 2

the stacked structure, especially with the inclusion of some of

the revisited higher symmetries [47]. Glide symmetry is a kind
of higher symmetry particularly useful in planar and stacked
structures which involves a mirroring and a displacement of
half a period between adjacent layers. Their effects cannot be
modeled with the formulation presented in [46], as the layers
were required to be aligned along the vertical direction. The
implementation of glide symmetry makes it possible to sup-
press the lowest stopband of the first propagating modes [48],
reduce the frequency dispersion of the structure [49]–[51],
increase the equivalent refractive index [52], [53], and produce
wideband anisotropy [54].
(a) (b)
The above beneficial properties are expected to be efficiently
analyzed with the extension of the ECA proposed in this work.
To reach this goal, the multimodal equivalent-circuit method-
ology reported in [34] is now extended to accurately com-
pute the scattering properties of asymmetrical and nonaligned
stacks formed by slot-based 2-D periodic arrays with arbitrary
apertures under normal and oblique incidences. The proposed (c)
formulation is a non-trivial extension of 1-D case reported Fig. 1. (a) Example of a stack of three asymmetrical and nonaligned 2-D
in [55], which aims to explore the limits of applicability of periodic arrays composed of arbitrary apertures and (b) longitudinal view
the equivalent-circuit modelling by including a second spatial illustrating the incidence plane. (c) General circuit model for the stack of
three arbitrary coupled apertures.
dimension and by discussing new possibilities of analysis
and/or applications. Furthermore, stacked slot-based structures
possessing glide symmetry can be studied using the present frequency ω = 2πf is obliquely impinging with a wavevector
circuit perspective. This is a very appreciated feature since kinc = (kx0 , ky0 , kz0 ) given by
glide-symmetric structures can rarely be described by means q
of circuit models due to the strong and non-trivial interac- (0)
kx0 = εr k0 sin θ cos φ (1)
tion between adjacent layers [51], [56]. However, the strong q
(0)
couplings related to closely-spaced layers can be fully taken ky0 = εr k0 sin θ sin φ (2)
into account with the present approach. Thus, the proposed q
(0)
ECA will reveal itself as a very efficient tool for the design of kz0 = εr k0 cos θ (3)
wideband radomes, polarization converters, filters, absorbers, (0)
and many other devices based on stacked metallo-dielectric where εr is the relative permittivity of the incident medium,
layers, even in complex scenarios. k0 is the vacuum wavenumber, and θ and φ are the elevation
and azimuth angles of the incident wave, respectively. The
The work is organized as follows. Section II presents the
adjacent metallic screens of the stack are separated with
formulation applied to the computation of the scattering prop- (i)
erties of asymmetrical and nonaligned stacks of 2-D arbitrarily dielectrics of relative permittivity εr and thickness hi .
shaped apertures. Section III particularizes the study to canon- The tangential electric field on the aperture of a unit cell
ical geometries, such as rectangular and annular apertures. of one of the metal screens, Et (x, y, ω), is modeled in the
Section IV analyzes apertures of arbitrary shape. Section V following way:
illustrates how to carry out a dispersion analysis of 3-D Et (x, y, ω) = F (ω) Ea (x, y) (4)
periodic stacks for a frequency range only achieved before
by commercial simulators. Finally, Section VI summarizes the where F (ω) is a frequency-dependent complex factor and
main conclusions extracted from the work. Ea (x, y) is the assumed frequency-independent spatial profile.
This assumption, key for the development of the formulation,
is found to be applicable in a wide frequency band for many
II. A NALYSIS practical structures, even beyond the grating-lobe regime [34],
[46]. The spatial profile Ea (x, y) can be expressed in closed
A. Formal Derivation form for canonical geometries (such as the rectangular and
annular apertures considered in Appendix A), which helps
This section will first briefly outline the general procedure to reduce considerably the computational effort as it will be
already reported in previous works of some of the authors to exploited in Section III. For more complex aperture geome-
deal with a stack of N metallic screen periodically perforated tries, the use of any full-wave commercial software allows us
with arbitrary apertures. For simplicity, let us consider the to extract the spatial profile from the simulation of a single
stack of three nonaligned 2-D periodic arrays composed of (non-stacked) free-standing metallic screen at just one partic-
strongly-coupled arbitrary apertures displayed in Fig. 1(a), ular frequency value. With this limited use of the full-wave
upon which a time-harmonic incident plane wave of angular simulator, we can combine the advantages of the ECA with

Fig. 2. Detailed circuit model for a 3-layer asymmetrical and nonaligned stack of arbitrary apertures. Harmonics of different order are coupled together
through the associated parallel-connected transmission lines loaded with transformers.

the versatility of commercial software to deal with arbitrary sides of these transmission lines. The internal regions can then
geometries [25]. This advantageous assistance of commercial be modelled by the following admittance matrix (i = 1, 2):
simulators will be exploited in Section IV for the analysis of " #
(i) (i)
nonaligned stacks of arbitrary geometry. (i) Y11 Y12
Y = (i) (i) (6)
Based on the transfer (ABCD) matrix formalism [31], Y21 Y22
[57], the formulation derived in [55] for nonaligned stacks
of 1-D periodic slit arrays is now extended to cover cases the entries of which are calculated as (u, v = 1, 2)
of 2-D apertures of arbitrary geometry. As clearly shown ∞
X
(i) (i),TM (i),TE
in [55, Fig. 3(b)], the circuit topology found for the stacks Yuv = (Yuv,nm + Yuv,nm ) (7)
of slit/aperture arrays consists of blocks of parallel-connected n,m=−∞
transmission lines loaded with transformers. Thus, as schemat-
with (TX will stand indistinctly for either TM or TE)
ically shown in Fig. 1(b), the three-screen stacked structure
in Fig. 1(a) can be divided in four blocks: two associated (i),TX

(i),TX
2 h
(i),TX (i)
i
with the so-called external input and output regions, and other Y11,nm = Nnm,L −jYnm cot(kz,nm hi ) (8)
two regions associated with the internal part of each pair of
h i
(i),TX (i),TX (i),TX (i),TX (i)
Y12,nm = Nnm,L Nnm,R jYnm csc(kz,nm hi ) (9)
consecutive coupled arrays. The resulting network composed
of four parallel-connected blocks is depicted in Fig. 2. At
the light of this network, taking into account the parallel
h i
(i),TX (i),TX (i),TX (i),TX (i)
Y21,nm = Nnm,R Nnm,L jYnm csc(kz,nm hi ) (10)
nature of the connections, the external input/output regions are
2 h
completely characterized by the following single admittance:
i
(i),TX (i),TX (i),TX (i)
Y22,nm = Nnm,R −jYnm cot(kz,nm hi ) . (11)
∞ 2
(in)/(out)
X h
(1)/(2),TM (0),TM In the above derivations, it has been assumed that the internal
Yext = Nnm,L/R Ynm
region (i) only comprises a single dielectric (i). As reported
n,m=−∞
n,m6=(0,0) in [24], if the internal region (i) is composed of several

(1)/(2),TE
2 i dielectric layers, we should substitute the transmission lines
(0),TE
+ Nnm,L/R Ynm (5) associated with the harmonics inside the single dielectric by
the corresponding cascade of transmission lines that accounts
where the index (1)/(2) refers to the (i)-th internal region and for the layered environment (namely, the terms inside the
the index (0) refers to free space. The network topology of brackets in the above expressions should be substituted by the
the internal (i)-th block is formed by the parallel-connected corresponding ones associated with the cascade of dielectric
transmission lines associated with the harmonics correspond- layers). The indexes L and R in (5) and (8)–(11) refer
ing with the dielectric layers inside the corresponding pair of respectively to the left-side and right-side aperture arrays that
(i),TX
coupled arrays and the transformers at the left and right hand bound the dielectric (i). Ynm is the wave admittance of the

(n, m)-th (TX ≡ TM/TE) harmonic at dielectric (i), defined which implies that
as (i),TM (i),TM
Nnm,R = (−1)n+m Nnm,L . (20)
(i)
(i),TM 1 k
Ynm = (i) (i) (12) This particular relation between the transformer ratios is
η kz,nm
in very close correspondence with the discussion in [59]
(i)
(i),TE 1 kz,nm on the symmetry of the even/odd harmonics when dealing
Ynm = (13)
η (i) k (i) with glide-symmetric structures. Thus, it is found that each
(n, m) harmonic involves the presence of a magnetic/electric
with η (i) being the wave impedance of the i-th medium and wall in the middle plane of the sub-unit cell depending on
(i)
kz,nm the longitudinal wavenumber of the (m, n)-th harmonic whether n + m is even/odd. This interesting feature of Bloch
in such medium, given by modes in glide-symmetric structures is key for providing many
(i)
q of the beneficial properties of these periodic structures [60].
kz,nm = [k (i) ]2 − |kt,nm |2 . (14) It should be noted that all the previous expressions from
q (7) to (17) are frequency dependent. Thus, the double sum
(i)
In (14), k (i) = εr k0 and kt,nm is its associated transversal in (7) would have to be performed for every frequency value
wavevector, expressed as in an eventual frequency sweeping. However, for high-order
2 2
(ho) harmonics (kxn + kym εr k02 ), it is apparent that the
kt,nm = kxn x̂ + kym ŷ = (kx0 + kn )x̂ + (ky0 + km )ŷ (15) wavenumber and wave admittances can be well approximated
with as [24], [36]
2πn 2πm (i),ho
p
kn = , km = kz,nm ≈ −jαnm = −j kn2 + km 2 (21)
px py
and
and px , py the periods of the unit cell in the x and y directions.  (i)
(i),TX
From a circuit standpoint, the coefficients Nnm,L/R are jωε0 εr (i)
≡ jωCnm , TM harmonics



the turn ratios of transformers associated with the (m, n)-th (i),ho
Ynm ≈ αnm (22)
harmonics in region (i) [24]. Mathematically they stand for  αnm ≡ 1

, TE harmonics .

the projection of the (n, m)-th harmonic on the 2-D Fourier jωµ0 jωLnm
(i)
transform of the spatial profile Ea,L/R (x, y) at the correspond- It implies that a great deal of computational effort can be
ing left-/right-side apertures; that is, saved in the computation of (7) by splitting the double infinite
(i),TM e (i) (kt,nm ) · k̂t,nm sum into a low order (lo) contribution (|n, m| ≤ N ), which
Nnm,L/R = E a,L/R (16)
is frequency dependent but only comprises a few terms, plus
(i),TE e (i) (kt,nm )
Nnm,L/R = E a,L/R · (k̂t,nm × ẑ) (17) a higher order (ho) contribution (|n, m| ≥ N + 1) that is
frequency independent; namely,
where k̂t,nm is the unit vector associated with kt,nm and
N
e (i) (kt,nm ) is the 2-D Fourier transform of the spatial
E (i)
X h
(i),TM,lo (i),TE,lo
i
a,L/R Yuv (ω) = Yuv,nm (ω) + Yuv,nm (ω)
profile in the left/right apertures of the coupled arrays (i)
n,m=−N
calculated at kt,nm . ∞
(i),TX (i),TX X h i
In general, Nnm,L and Nnm,R have different and unrelated + (i),TM,ho
Yuv,nm (i),TE,ho
+ Yuv,nm . (23)
values, since Ee (i) (kt,nm ) can be arbitrarily different from |n,m|≥N +1
a,L
(i)
E
e (kt,nm ). However, in many practical situations, the left
a,R Therefore, the computational effort in the frequency-sweeping
and right apertures can be related by simple algebraic trans- computation of each of the admittance matrices Y(i) (ω) lies
formation (translation, rotation, reflection and dilation), which almost entirely in the obtaining of the reduced summation
then makes it possible to also find simple algebraic relations associated with Ylo (ω), with Yho needed to be computed
between the involved Fourier transforms. As an example, just once and stored for subsequent use. In most of the cases
the misalignment of consecutive periodic arrays is taken into studied in this paper, it suffices to take N . 6, although this
account by means of value can be smaller if (i) the upper frequency of analysis is
e (i) (kt,nm ) = E
e (i) (kt,nm )e jkt,nm ·d not close to the onset of the diffraction limit (f < c/p, with
E a,R a,L (18) c being the speed of light), and (ii) when the periods of the
where d = dx x̂ + dy ŷ represents the displacement of the unit cell are electrically small; feature that is found in many
aperture (this misalignment was already considered in [55] for applications of metasurfaces.
1-D periodic arrays). The case of stacks of rotated periodic For a straightforward computation of the scattering pa-
(in)/(out)
arrays will be treated in Section III-B. An interesting case rameters, the admittances Yext associated with the ex-
(i)
raises when d = px /2x̂ + py /2ŷ; namely, when the periodic ternal regions and the Y admittance matrices associated
stack has glide symmetry [47], [56]. Under normal incidence, with the internal regions are converted to transfer (ABCD)
it means that (18) turns into matrices [57]. Given the particular topology found for the
equivalent circuit of a generic stack of M ≥ 2 coupled
e (i) (kt,nm ) = (−1)n+m E
E e (i) (kt,nm ) (19) layers, the problem can be split up into M − 1 internal blocks
a,R a,L

plus 2 additional external blocks. Thus, the complete network

shown in Fig. 2 can be represented in terms of a single transfer
matrix computed as the product of the resulting four individual
transfer matrices. It should pointed out that the final transfer
matrix does contain all the relevant information about the
propagating mode and all higher-order harmonics, as well as
the possible couplings between them.

B. Additional Considerations
(a) (b)
Periodic arrays of patch-like scatterers can be regarded as
complementary to aperture-like ones and, thus, the application
of a similar procedure as above to patch-like arrays would
lead to a complementary equivalent circuit topology. In [24]
it was shown that the equivalent-circuit topology of a single
periodic array of patches has all the individual transmission
lines associated with the different harmonics connected in
series, unlike the parallel configuration found for aperture-like
arrays. If two or more patch-like arrays are to be stacked, the (c) (d)
corresponding transmission lines associated with harmonics of Fig. 3. (a) Pair of strongly coupled arrays with rectangular apertures arranged
the same order would have to be connected, as done in Fig. 2 in a glide-symmetric configuration. Transmission coefficient for a separation
for the aperture problem. However, the resulting patch-like between arrays of (b) h = p/5, (c) h = p/20, and (d) h = p/50. TM normal
incidence is assumed. Geometrical parameters of the unit cell: a1 = a2 =
connection gives rise to a very complicated network that, to the 6 mm, b1 = b2 = 3 mm, px = py = p = 10 mm, and εr = 1.
authors’ knowledge, cannot be simplified in a similar fashion
as in the concatenation of simple Y(i) blocks shown in Fig. 2.
The lack of such straightforward connection complicates the
Another limitation discussed in [58] concerns the variation
mathematical treatment of the problem enormously and makes
of the spatial profile in the different arrays of the stack when
it very difficult to have a simple and insightful physical
the apertures are strongly coupled. That possible variation
understanding of the structure from a circuit model standpoint.
might be a relevant limiting factor in many practical cases
For this reason, stacks of patch-like arrays are out of the scope
since we are implicitly assuming that all the apertures in the
of this paper.
stack have the same spatial profile as the one corresponding to
A relevant issue concerning the application of the ECA
each aperture taken isolated. In order to assess the relevance
previously proposed is the discussion of its limits of validity.
of this limitation, we will compare our ECA results with
As already mentioned and previously reported in [58], the
those provided by CST in some cases of strong coupling for
most relevant theoretical limitation of the approach comes
the configuration shown in Fig. 3(a). In our simulations with
from the validity of assumption (4); namely, that the spatial
commercial software CST, we select the Frequency solver,
profile of the tangential field in the apertures does not vary
configured with a maximum number of 20 cells per box
too much with frequency. From a practical point of view, this
model in the tetrahedral mesh, a maximum number of passes
assumption can be found satisfactory up to frequencies below
(finer mesh per iteration) of six, and 60 Floquet harmonics.
the second “excitable” resonance of the aperture. Thus, for
Figs. 3(b)-(d) show the transmission coefficient of a pair of
the case of an array of rectangular apertures of size a × b
strongly coupled arrays with rectangular apertures arranged in
(a > b) with a normally incident electric field directed along
a glide-symmetric configuration when the separation between
the shorter dimension, the second excitable resonance will
the arrays are p/5, p/20 and p/50, respectively (these cases
occur when a ≈ 3λ/2; that is, for frequencies satisfying
correspond to h = λ/5, h = λ/20 and h = λ/50, taking λ
f . 3c/(2a) (which includes a large frequency range well
at the onset of the diffraction regime; namely, f = 30 GHz).
inside the diffraction regime). In the case of oblique incidence,
The good agreement found between our ECA data with the
the limiting frequency can reduce to f . c/a since the second
ones provided by CST in all these cases makes it apparent that
excitable resonance may occur at a ≈ λ. For other non-
our assumption of taking the “isolated” spatial profile for the
canonical geometries of the aperture such as the Jerusalem
apertures work reasonably well, even in the extreme scenario
cross, the second excitable resonance may appear close to the
considered in Fig. 3(d).
first one, which would certainly reduce the range of appli-
cability of the present approach. However, despite this fact,
it can be stated that the ECA is found to work satisfactorily
for many practical cases where the numerically-intensive full- III. C ANONICAL A PERTURES
wave approach can be advantageously substituted by the much
simpler ECA. Actually, this consideration is one of the main In this section, the proposed ECA is used and tested to com-
goals of the present work, where we explore different scenarios pute the scattering properties of stacked structures formed by
that might be thought to be intractable by means of the present apertures for which the spatial dependence of their tangential
quasi-analytic ECA. electric fields can be expressed by closed-form expressions.

(a) (a)

(b)
(b)
Fig. 5. (a) Glide-symmetric stack of ten 2D arrays formed by subwavelength
Fig. 4. (a) Symmetrical and aligned stack of ten 2-D arrays formed by annular apertures. (b) Transmissivity versus frequency for TM normal inci-
annular apertures presented in [61]. (b) Transmissivity versus frequency for dence. Geometrical parameters of the unit cell: a = 3.8 mm, b = 4.8 mm,
TM normal incidence. Geometrical parameters of the unit cell: a = 3.8 mm, px = py = p = 10 mm, dx = dy = p/2, h = 1.575 mm, and εr = 2.65.
b = 4.8 mm, px = py = p = 10 mm, h = 1.575 mm, and εr = 2.65.

B. Glide-symmetric Stacks
A. Symmetrical and Aligned Stacks
As previously discussed, breaking the alignment between
As a first study case, it is considered the multilayered two consecutive layers can bring some advantages to conven-
structure studied in [61] and shown in Fig. 4(a). The stack is tional FSSs. In particular, next we will study the effect of
made up by ten perfectly aligned metallic screens of periodic the introduction of glide symmetry in the structure previously
annular apertures separated by a dielectric of permittivity εr . analyzed. The half-period displacement implicit in the glide
The metal is assumed to be a perfect electric conductor (PEC) symmetry is taken into account here in a fully-analytical form.
and no losses are considered in the dielectrics in this stage. To the authors’ knowledge, this is the first reported case where
High transmission can be achieved by stacking and alternating equivalent circuits can accurately model the strong higher-
identical metallic and dielectric layers. This is appreciated order coupling between Floquet harmonics in glide-symmetric
in Fig. 4(b), where the transmissivity is computed with the FSS structures.
proposed approach and then compared with the original results Fig. 5(a) presents the glide-symmetric version of the stacked
in [61]. For the computation, the electric field in the aperture structure previously presented in Fig. 4(a), named henceforth
is assumed to be well modeled by the function (43) (l = 1) as mirror-symmetric. For a fair comparison, the same geo-
in the Appendix, N = 5 has been considered in (23) and metrical parameters and number of layers have been kept.
the double infinite sum has been truncated to Nmax = 10. Fig. 5(b) illustrates the transmissivity of the glide-symmetric
A good agreement is found between our closed-form results staked structure. A good agreement is observed between the
and the set of data in [61]. The good agreement is somewhat proposed formulation and the finite element method (FEM)
expected since the frequency range analyzed covers the range of commercial software CST for such a complex transmis-
of validity of our approach discussed in Sec. II-B (in this sion spectra. However, our formulation is significantly more
case, f / 15 GHz). As the number of stacked metallic layers computationally efficient than the commercial software. Using
increases, more transmission peaks appear in the transmission the same computer, the ECA took less than 20 seconds in
spectra. For the structure under consideration, a passband the analysis of the whole frequency range while CST took
emerges from 6.5 GHz to 12.5 GHz with transmissivity values more than 30 minutes for the same analysis. It is also ob-
over 50%. It should be mentioned that all terms involved in the served that the passband of the glide-symmetric configuration
computation of our data are known in closed-form expressions, (8 GHz of bandwidth) is notably widened compared to the
which allows us to carry out the study shown in the figure with mirror-symmetric structure (6 GHz of bandwidth). This can be
a very reduced computational effort. attributed to the suppression of the stopband related to the first

Bloch mode in the glide-symmetric FSS. However, note that

the ripple level of the structure with glide symmetry is also
increased compared to the mirror-symmetric structure. This is
associated with the mismatching of the impinging free-space
wave and the impedance of the propagating Bloch mode in
the stacked structure. These facts will be discussed in greater
detail in Section V.

(a)
C. Asymmetrical and Nonaligned Stacks

(a)
(b)

Fig. 7. (a) Five-layer rotated stack formed by rectangular apertures. (b) Mag-
nitude of the transmission parameters for TM normal incidence. Geometrical
parameters of the unit cell: a1 = a5 = 7.25 mm, a2 = a4 = 8 mm,
a3 = 9 mm, b = 3 mm, px = py = p = 10 mm, h = 1.5 mm, and
εr = 1.

D. Rotated FSS
As previously mentioned, rotation is one of the possible
algebraic transformations that can relate the left and right aper-
tures of a coupled pair. Stacks of rotated periodic structures
(b) is a practical configuration, usually applied in the context of
polarization converters [14], [28], [29]. In the frame of our
Fig. 6. (a) Asymmetrical and nonaligned stack of two 2-D arrays formed analytical ECA, the spatial field profile of a single right- or
by annular and rectangular apertures. (b) Magnitude of the transmission
parameter for TM normal incidence. Geometrical parameters of the unit cell: left-side rotated aperture Erot
R/L (r) admits to be represented in
a1 = 3.8 mm, b1 = 4.8 mm, a2 = a3 = 0.4 mm, b2 = b3 = 0.2 mm, terms of the field profile in a non-rotated aperture EL/R (r)
px = py = p = 1 mm, dy = p/2, h1 = h2 = 0.6 mm, and through the rotation matrix R:
(1) (2)
εr = εr = 2.6.
−1
Erot
a,R/L (r) = REa,L/R (R r) (24)
The range of use of the circuit model is not limited to
multilayered structures with the same type of apertures, such where r = xx̂ + yŷ and
as those discussed above. Different types of apertures can

cos α − sin α
be combined, as shown in Fig. 6(a). Periodic annular and R= (25)
sin α cos α
rectangular apertures are stacked in this case, forming a 3-layer
structure where the second perforated plate is off-shifted half with α being the rotation angle of the apertures in the counter-
a period in y direction, dy = p/2. The spatial profile of the clockwise direction. Starting from (24), it can be demonstrated
electric field assumed to be excited on the rectangular aperture that the 2-D Fourier transform of the spatial profile in the
is given in (39) in the Appendix. Our closed-form results and rotated aperture can be written in terms of the profile with no
the ones obtained with CST are plotted in Fig. 6(b). Good rotation as [62], [63]
agreement is observed with CST in a wide frequency band, e a, L/R (R−1 kt,nm ) .
e rot (kt,nm ) = RE
E (26)
a, R/L
reaching a precision to the third and fourth decimal place
(-60 dB) in the rejection bands. It should be remarked the The obtaining of the 2-D Fourier transform of the rotated field-
fully-analytical nature of our results, in contrast to previous profile just demands a simple linear transformation in terms
approaches [46]. Fully-analytical results can be obtained as of the rotation angle α. The computation of the corresponding
long as the spatial profile of the considered apertures can TM- and TE- transformer ratios is finally achieved by intro-
be expressed in a closed form, regardless of the geometry ducing E e rot (kt,nm ) in (16) and (17).
a, R/L
of the apertures and the application of linear transformations An example of a stack comprising five rotated free-standing
(displacement, rotation, scaling, etc). FSSs (q = 1, 2, . . . , 5) is shown in Fig. 7(a), where it can be

seen that each single metallic screen is periodically perforated

with rectangular apertures of dimensions aq × bq . The stack
configuration is such that the rotation angle of the first and
fifth (last) layers are α1 = 0° and α5 = 90°, respectively.
This configuration, if efficiently optimized, may constitute a
potential polarization converter. The orientation of the first and
fifth apertures is suitable for a conversion from y-polarized to
(a)
x-polarized electric fields. To accomplish the conversion, the
intermediate layers (q = 2, 3, 4) have to be conveniently de-
signed. In the example of Fig. 7(a), the corresponding rotation
angles are α2 = 12°, α3 = 50°, and α4 = 78°, following a
growing trend from the first to the fifth screen. Along with
the optimum rotation angle for each screen, the dimensions
of each aperture are also optimized. In particular, an optimum
configuration has been found by keeping the shorter dimension
of the apertures (bq ) identical for all the screens whereas the
(b) (c)
larger dimension (aq ) is symmetrically distributed (a1 = a5 ,
a2 = a4 ). All the above derivations could be carried out with a Fig. 8. (a) Metal-backed stacked structure acting as an absorber. Absorption
very reduced computational effort due to the analytical nature coefficient for (b) TM oblique incidence and (c) TE oblique incidence.
Geometrical parameters of the unit cell: a = 8 mm, b = 1 mm, px = py =
of the employed ECA. (1)
p = 10 mm, h = 0.75 mm, hout = 0.75 mm, and ε̂r = 4 × (1 − j0.02).
Fig. 7(b) shows the magnitude of the transmission coeffi-
cient of the above structure when it is excited by a normally-
impinging plane wave with the electric-field vector directed
along the ŷ direction. The transmission coefficient is split the length of the shorted transmission lines (that is, the length
into two components: the component corresponding the co- of the grounded dielectric medium). Unlike (5), the summation
polarization term (Ey → Ey ) and the component corresponding in (27) has now to include the fundamental harmonics of order
to the cross-polarization term (Ey → Ex ). As it can be n = m = 0 for both TE and TM harmonics.
appreciated in the figure, almost full conversion Ey → Ex is Grounded dielectric FSSs can be employed for the design
achieved from 20 to 22 GHz, covering a fractional bandwidth of absorbers by introducing, for instance, a lossy dielectric
of 9.5% approximately (the co-pol level is below -20 dB). The substrate in the stack. The information about the losses is
agreement between results provided by CST and the results included in the permittivity of the lossy material, which
obtained by our analytical circuit model is very good. As a thus becomes a complex quantity. An example of a possible
comparison of the required computational effort, CST took absorber configuration is depicted in Fig. 8(a), where a stack
more than fifteen minutes (900 seconds) to compute 1001 of two perforated metallic screens with identical rectangular
equally-spaced frequency points while the proposed ECA took apertures is considered. A lossy dielectric slab of FR4 is
less than 15 seconds. sandwiched between both layers, having a relative permit-
(1)
tivity of εr = 4 and loss tangent tan(δ) = 0.02 —the
complex permittivity of the lossy substrate is then given by
E. Metal-backed FSS (1)
ε̂r = 4×(1−j0.02). A free-standing ground plane (ε(out) = 1)
r
The present ECA can easily deal with scenarios where the is placed at a distance hout . The structure is assumed to be
stack of aperture arrays are backed by a metallic screen. The excited by either a TM-polarized or a TE-polarized plane wave
back metallic screen is simply modeled as a short circuit; that impinges obliquely with an angle θ (normally incident
namely, the transmission lines associated with harmonics in waves are considered when θ = 0°). The geometry of the unit-
direct contact with the ground plane have to be terminated with cell aperture is chosen to have a high absorption rate in both
a short circuit. In the circuit representation shown in Fig. 2, polarizations up to a incidence angle of 20 degrees, though in
the outgoing dielectric medium was assumed semi-infinite and a narrow frequency band as shown in Fig. 8(b) and (c).
out
has an equivalent admittance, Yext , coming from the infinite The absorption coefficient (Ac ) is p
calculated in terms of
transmission lines in parallel shown in the yellow box of the reflection coefficient as Ac = 1 − |S11 |2 , where it
the equivalent circuit. When this medium is grounded, the is assumed that the reflected power is only carried by the
(out)
equivalent admittance Yext defined in (5) now becomes n, m = (0, 0) harmonic, whose reflection coefficient is then
∞
given by S11 . This assumption is valid below the onset of
h 2
(out)
Yext =
X (2),TM
− j Nnm,R (out),TM
Ynm (out)
cot(kz,nm hout ) grating lobes and absence of cross-pol effects, which is fully
n,m=−∞
satisfied in the present case (the symmetry of the rectangular
2 i apertures prevents the excitation of the cross-polarization
(2),TE (out),TE (out)
− j Nnm,R Ynm cot(kz,nm hout ) (27) term). The strong coupling induced by the proximity of the
perforated plates can result in high absorption amplitudes, as
(out),TM/TE (out)
where Ynm and kz,nm take into account the relative this case illustrates. The thickness of the whole absorber is
permittivity of the grounded dielectric medium, and hout is very reduced, with a total size of 1.5 mm, which means a

thickness of . λ0 /20 at the full-absorption peaks observed in

the figures. A slight frequency shift of less than 1% can be
appreciated between some results given by CST and the ECA.
The obtained good accuracy of the ECA demonstrates the
ability of the approach to cover scenarios with stacked arrays
including the presence of dielectric losses and a reflecting
ground plane. The ability of the proposed ECA to deal with
oblique incident waves is also validated in this example. (a)
Scenarios with θ = 60o are well represented for both TM
and TE incidence.
It should be noted that the present high accuracy of the
ECA is directly related to the good approximation that the
assumption made in (4) stands for in this case. The accuracy
is expected to be more limited in other situations where
the eventual excitation of higher resonances of the aperture
(b)
is more significant. The lack of symmetry in the problem,
the electrical size of the aperture as well as the number
of plates are factors that may affect the suitability of the
implicit approximation of the ECA in scenarios where oblique
incidence is considered. However, it should be pointed out
that many practical situations do satisfy the ECA validity
conditions, and that is the model’s application niche this work
is exploring.

IV. A RBITRARY A PERTURES

For non-canonical aperture geometries for which the spatial
profile in (4) is not easily expressed in closed form, we
can make use of a hybrid approach that advantageously (c)
combines the use of the equivalent circuit and commercial
simulators [25]. More specifically, we can benefit of the ability Fig. 9. Three-layer stack of arbitrary apertures. Geometrical parameters of
the unit cell: a = 4.1 mm, b1 = 1 mm, b2 = 0.5 mm, l = 1.5 mm, w =
of commercial simulators to deal with arbitrary geometries 0.4 mm, lb = 2 mm, wb = 2 mm, px = py = p = 8 mm, h1 = 1.575 mm,
to extract the spatial profile of the considered aperture when (1) (2)
h2 = 0.5 mm, εr = 4.7, and εr = 2.5. The bowtie-shaped apertures of
only a single periodic array is considered in free space. The the third layer are rotated α = 10o and scaled 1.2 times (Sx = Sy = 1.2)
extraction of this spatial profile has to be done at a single with respect to the apertures of the second layer.
frequency point, which is an operation far less computationally
demanding than the simulation of the complete stack in the it is only necessary to extract the spatial profile of the first
full frequency range. From this operation, the transformer (1,2)
and second plates Ea,L (x, y), since the Fourier transform
turn ratios in (16) and (17) are obtained after numerically of the spatial profile of the third plate can be expressed as
computing the Fourier transform of the aperture spatial profile. e (2) (kt,nm ) = R S E
E e (2) (R−1 S−1 kt,nm ), where R is the
a,R a,L
Moreover, linear transformations (rotation, scaling, displace- previously defined rotation matrix and S represents a scale
ment) can still be applied to this Fourier transform in order to matrix given by [62], [63]
find relations between the apertures of the perforated plates. As
previously discussed, this approach will be valid as long as the Sx 0
S= . (28)
aperture spatial profile does not greatly vary with frequency. 0 Sy
As an example, Fig. 9(a) presents a 3-layer stack formed Note that matrices R and S commute if Sx = Sy , as the
by a periodic array of Jerusalem-cross apertures and two scaling operator is then defined by a diagonal matrix. Fig. 9(c)
arrays of bowtie-shaped apertures, separated by two different illustrates the transmission parameter of the 3-layer stack of
(1)
dielectrics of permittivities, εr = 4.7 and εr (2) = 2.5. The arbitrary apertures. A good agreement is observed with CST,
spatial profiles of the different apertures are extracted with the even for such a complex structure. Slight differences are
simulation of single, free-standing layers in CST at the lowest observed beyond 26 GHz. This is due to the excitation of
operating frequency (20 GHz). In this example, the mesh of the second resonance in the Jerusalem cross, which reduces
the unit cell consists of 60×60 hexahedral elements. In order the range of validity of the ECA as the assumption of a
to show the potentialities of the approach also for arbitrary frequency-independent spatial profile in (4) is no longer valid
geometries, the apertures of the third layer are taken as rotated from this frequency. As a comparison, CST took more than
and scaled versions of the apertures that form the second layer. 20 minutes (more than 1200 seconds) to compute 501 equally-
This can be appreciated in Fig. 9(b), where the absolute value spaced frequency points while the present hybrid approach
of the aperture spatial profiles is shown. Thus, in this case took, in total, less than 40 seconds. It clearly proves that

efficiency of the present hybrid implementation to compute

the scattering parameters of stacked structures with arbitrary
apertures.

V. I NFINITE P ERIODIC S TACKS

A. Dispersion Diagram
As is well known, the dispersion relation of a periodic (a)
structure along the z direction can be expressed in terms of
the elements of the transfer matrix of the corresponding unit
cell (period pz ≡ p) as [31], [57]
Ap + Dp
cosh (γz pz ) = (29)
2 (b)
where γ = αz + jβz is the propagation constant of the Floquet
mode and the subindex p means that Ap and Dp are elements
of the transfer (ABCD) matrix Tp associated with the unit
cell of longitudinal period pz . Given that A = −Y22 /Y21 ,
D = −Y11 /Y21 [57], the dispersion relation can be expressed
in closed form by replacing (7)–(11) into (29) to give

cosh (γz pz )
∞ h i TM/TE
TM/TE 2 TM/TE 2
P
Nnm,L + Nnm,R Ynm cot(kz,nm h)
n,m=−∞
= ∞
TM/TE TM/TE TM/TE (c)
P
2 Nnm,R Nnm,L Ynm csc(kz,nm h)
n,m=−∞
(30)

where the sum above extends to both TE and TM modes.

The index (i) is removed in the above expression since
all the internal regions are exactly the same in the present
infinitely periodic structure under study. If there are more than
one dielectric between the pair of coupled arrays, the terms
inside the brackets in the second line of the above equation
should be appropriately modified. In many practical situations,
the unit cell of the periodic structure can be chosen to be
symmetric, which implies that Ap = Dp or, equivalently, (d)
TM/TE TM/TE
Nnm,R = Nnm,L .
As already mentioned in Sec. II-B, one of the main limita- Fig. 10. (a) Infinite periodic 3D array formed by annular apertures. (b) Spa-
tial distributions on the annular apertures analytically computed with Eq. (43).
tions of the ECA for the study of the scattering properties of Dispersion diagrams (phase shift and normalized attenuation constant) of
stacked structures comes from the inadequacy of the method the (c) mirror-symmetric and (d) glide-symmetric configurations. The results
extracted from CST (colored circles) are shown for comparison purposes. Pa-
to deal with frequency sweepings where the spatial profile of rameters of the unit cell: a = 3.8 mm, b = 4.8 mm, px = py = p = 10 mm,
the tangential field on the apertures significantly varies along pz = 1.575 mm, and εr = 2.65.
the considered frequency range [34]. This fact was physically
linked to the excitation of higher resonances in the aperture.
From a practical point of view, this limitation can be overcome Following this rationale, the dispersion relation of the peri-
by expressing the spatial profile with more than one basis odic mirror-symmetric stack of annular-aperture arrays shown
functions, as exploited, for instance, in [26]. However, at the in Fig. 10(a) has been computed by solving
light of (30), one can observe that this equation can still
∞ 2 TM/TE
be utilized to compute the dispersion behavior of high-order P TM/TE
Nnm Ynm cot(kz,nm h)

Floquet modes, despite using a single basis function to model n,m=−∞
cosh (γz h) = ∞ .
the spatial profile on the aperture, as long as the single spatial P 2
TM/TE ) [Y TM/TE csc(k
(Nnm nm z,nm h)]
profile employed to compute the transformer ratios in (30) can n,m=−∞
match the geometrical variations of the corresponding Floquet (31)
mode. In practice, it means that the function Ea (x, y) should The indexes L/R has been suppressed since its difference is not
be chosen with a spatial profile that closely resembles the necessary in this case. he different Floquet modes supported
different resonant modes of a single aperture. by the periodic stack are computed with (31) by imposing

the spatial profiles shown in Fig. 10(b). In the case of annular periodic stack, illustrated in Fig. 10(b), are the same as the
apertures, these spatial profiles correspond to the mathematical ones used for the mirror-symmetric periodic stack. The good
form given in Eq. (43) in the Appendix for l = 1, 2, 3, 4 agreement obtained for both the mirror-symmetric and the
and φ = 0, π/4. These closed-form expressions of the spatial glide-symmetric periodic stack of aperture arrays clearly sup-
profiles associated with the Floquet modes are found to match ports the ability of the present closed-form approach to obtain
quite well the actual tangential fields extracted with CST. A the dispersion diagram of periodic stacked structures even
comparison of the results for the phase shift (βz pz /π) obtained beyond the previous limits of validity of the analytical ECA
with the simplified procedure proposed in this work and data discussed in [34].
provided by the CST Eigensolver is shown in the left plot It is worth noting here the correlation between the re-
of Fig. 10(c), showing a good agreement between both sets of sults of transmissivity shown in Figs. 4(b) and 5(b) and the
results. dispersion behavior of the first Floquet mode in Figs. 10(c)
The dispersion diagram of the glide-symmetric version of and (d), respectively. The increase of the passband band-
the above periodic structure is shown in Fig. 10(d). For glide- width observed in Fig. 5(b) for the glide-symmetric structure
symmetric periodic structures, it should be taken into account completely agrees with the wider bandpass of the first mode
that the actual unit cell of the structure is symmetric (and of in Fig. 10(d) with respect to the one in Fig. 10(c). This effect
size pz = 2h in the particular case under study); namely, (29) has been widely reported as one of the advantages of glide-
reduces to symmetric structures [60], and is shown here to be also a
profitable characteristic of stacked FSS’s.
cosh (γz pz ) = Ap . (32)
A well-known relevant constraint of most electromagnetic
Since the actual unit cell now involves two pair of coupled commercial simulators comes from their inability to provide
arrays (and in the simplest case considered here, two dielectric the attenuation constants of the modes [65]. Fortunately, the
layers of size h), it means that Ap actually comes from Tp = present approach does not have this limitation since it directly
T0p/2 T00p/2 , where T0p/2 stands for the transfer matrix of one computes the complex propagation constants of the Floquet
of the two sub-unit cells of size p/2 that comprises the actual modes. The corresponding values for the normalized atten-
unit cell (T00p/2 is the transfer matrix of the remaining sub- uation constant (αz /k0 ) for the previously analyzed mirror-
unit cell). In similarity with the discussion in [64, Sec. 2.2], and glide-symmetric structures are shown in the right plots
the original dispersion relation (29) of the glide-symmetric of Figs. 10(c) and (d), respectively. For the sake of clarity, only
structure [with period pz = 2h, as shown in Fig. 10(a)] can the attenuation constant of the first Floquet mode is shown.
alternatively be rewritten as It can be appreciated that, in the common stopband regions
q (i.e., from 12.5 to 18.5 GHz), the attenuation constant of the
cosh (γz pz /2) = Ap/2 Dp/2 = Ap/2 . (33) mirror-symmetric configuration is greater than the one of the
For glide-symmetric structures, Ap/2 turns out to be equal to glide-symmetric case. This fact is in agreement with the results
Dp/2 , which follows after introducing (20) into (8) and (11). reported in [50] for waveguides loaded with holey structures.
It implies that the dispersion relation of glide-symmetric
structures can be obtained dealing only with the sub-unit cell B. Bloch-Floquet Impedance
of the structure (of size h in the present case); namely, the As is well known, the introduction of the Bloch-Floquet
dispersion equation can be written as impedance, given by
cosh (γz h) −2B
ZB± = p p (35)
∞
P TM/TE 2
TM/TE Ap − Dp ∓ (Ap + Dp )2 − 4
Nnm Ynm cot(kz,nm h)
n,m=−∞ is very helpful for the study of truncated periodic struc-
= ∞ . (34)
P TM/TE )2 tures [57]. Unfortunately, most of commercial eigenmode
(−1)n+m (Nnm TM/TE csc(k
[Ynm z,nm h)]
n,m=−∞ solvers are not able to directly compute the Bloch impedance.
The present formulation can overcome this weaknesses and
The comparison of the dispersion equations for the mirror- and
provides accurate information on the Bloch impedance.
glide-symmetric structures given in (31) and (34), respectively,
Figs. 11(a) and (b) illustrate the real part of the Bloch
clearly shows that the only difference comes from the factor
impedance for the mirror- and glide-symmetric infinite pe-
(−1)n+m in the denominator of (34).
riodic stacks already analyzed in Fig. 10. In the mirror-
The fact that only the sub-unit cell of the the glide-
symmetric structure, after noting that Ap = Dp , (35) reduces
symmetric structure has to be considered is in full corre-
to
spondence with the sub-unit cell concept introduced in the Bp
statement of the generalized Floquet’s theorem reported in [47] ZB± = ± q . (36)
A2p − 1
for periodic structures with higher symmetries.
The left plot in Fig. 10(d) shows that the results given by the For the glide-symmetric structure, it is found that the Bloch
closed-form expression in (34) are again in good agreement impedance can alternatively be computed as
with the data computed by CST, which have been obtained s s
by considering the actual unit cell of period p = 2h. Note ± Ap/2 Bp/2 Bp/2
ZB = ± =± (37)
that the considered spatial profiles for the glide-symmetric Cp/2 Dp/2 Cp/2

the inner radius of the annular-ring apertures varies. It can

be observed that the ripples reduce as the inner radius is
smaller; namely, as the width of the annular ring is wider. This
effect comes associated with a corresponding increase of the
Bloch impedance shown in Fig. 12(b), which is progressively
approaching the value of the free space impedance. As a result,
the mismatching between the impedance of the Bloch mode
and the free space is reduced and the transmission is enhanced,
(a) (b) although, in this case, at the cost of reducing the passband.

Fig. 11. Real part of the Bloch impedance for various Bloch modes in the (a)
mirror-symmetric, and (b) glide-symmetric infinite periodic 3-D stacks formed
by annular apertures. Geometrical parameters of the unit cell: a = 3.8 mm, VI. C ONCLUSION
b = 4.8 mm, px = py = p = 10 mm, h = 1.575 mm, and εr = 2.65.
This paper presents a rigorous formulation based on the
multi-modal equivalent circuit approach for the analysis and
design of stacked structures formed by 2-D periodic arrays of
arbitrary apertures. It is shown that a key potential of the ap-
proach comes from the fact that linear transformations between
adjacent layers (rotated, translated and scaled apertures) can
be modeled from a purely analytical perspective. This opens
new possibilities for the efficient design of polarizers, filters,
absorbers, thin matching layers, and other high-frequency
(a) (b) devices oriented to wireless communications. As an example,
Fig. 12. Reduction of the passband ripple in the glide-symmetric 10- we show the design of a broadband transparent structure
layer stack formed by annular apertures. (a) Transmissivity, and (b) Bloch formed by annular apertures, a polarization converter, and an
impedance for different inner radii values, a. Geometrical parameters of the absorber formed by rectangular apertures. Good agreement
unit cell: b = 4.8 mm, px = py = p = 10 mm, h = 1.575 mm, and
εr = 2.65. is observed between the present approach and the reference
results from CST for all the cases under study. Additionally,
the present formulation allows for the analysis of glide-
which means that only the sub-unit cell of size pz /2 should symmetric configurations from a circuit perspective. This is a
be considered in this case. remarkable feature, since the performance of glide-symmetric
It can be appreciated that the real part of the Bloch structures can rarely be described with circuit models due to
impedance in the first passband region of the l = 1 the strong interaction between adjacent layers.
mode in Fig. 11(a) for the mirror-symmetric configuration Purely analytical results are obtained from the circuit ap-
(around 195 Ω from 7 to 12.5 GHz) is closer to the free-space proach as long as the spatial profile of the apertures can be
impedance (η0 ≈ 377 Ω) than the one in Fig. 11(b) for the expressed in closed form, regardless of the geometry of the
glide-symmetric structure (around 145 Ω from 8 to 15 GHz). apertures and the applied linear transformations. However, a
This fact explains that the ripples in the passband of the hybrid approach that combines the use of commercial software
mirror-symmetric structure in Fig. 4 are lower than those for and the circuit model can be applied in those cases where the
the glide-symmetric structure in Fig. 5. In addition, the Bloch spatial profile cannot be expressed in closed form. This hybrid
impedance is progressively smaller in both configurations for approach integrates the ability of commercial simulators to
the second and third passbands of the first mode (l = 1), deal with arbitrary geometries with the reduced computational
which leads to an increased ripple level for the high frequency effort inherent to the equivalent circuit approach. The hybrid
passbands, in agreement with the results reported in [66] approach is validated with a three-layer stacked structure
for 1-D grating stacks. High-order modes (l = 2, 3, . . .) formed by Jerusalem-cross and bowtie-shaped apertures.
present a much smaller real part of the Bloch impedance Finally, it is shown that the dispersion properties of infinite
compared to the fundamental mode. Therefore, the resulting periodic stacks can be derived with the proposed formula-
mismatching causes these modes to be strongly reflected and tion. This is a remarkable feature, since most commercial
hardly transmitted in finite stacks. eigenmode software are unable to compute the attenuation
Bloch impedance can also be used to improve the perfor- constant and Bloch impedance of the modes. At the light
mance of multilayered FSS structures; in particular, filters of the present results, it is observed that the use of a single
and matching layers can be efficiently designed with the spatial profile suffices to compute the dispersion behavior of
proposed circuit approach by adjusting the Bloch impedance high-order Floquet modes, as long as the considered profile
of the passband regions to match the free space impedance. resembles the different resonant modes of the aperture. The
As an example, we will show that the passband ripple level good agreement shown supports the ability of our closed-form
of Fig. 5 can be reduced with this simple procedure. Thus, approach to obtain the dispersion diagram of periodic stacked
Fig. 12(a) illustrates the variation of the passband ripple level structures even beyond the previous limits marked by reference
of the glide-symmetric 10-layer stack with glide symmetry as works.

A PPENDIX where φ0 is the reference azimuth angle, and l stands for the
This appendix gives the analytical expressions of the spatial the order of the considered mode. As the variation in the radial
profiles Ea (x, y) [see Eq.(4)] considered for the computation direction has been suppressed, this approximation is valid as
of the stacked structures. It should be remarked that, although long as the slot width is narrow (1 ≤ b/a / 1.5). For the
the mathematical form of Ea (x, y) is assumed to be indepen- fundamental mode (l = 1), the expressions for the transformer
TM/TE
dent of the dielectric layers inserted in the stacked structure, turn ratios Nnm are found in [68, Eqs. (5) and (6)].
the information of the dielectric environment is fully accounted
for in the ECA by means of the characteristic admittances R EFERENCES
(12),(13) and wavenumbers (14) of the transmission lines [1] G. G. Macfarlane, “Quasi-stationary field theory and its application
to diaphragms and junctions in transmission lines and wave guides,”
associated with the Floquet harmonics. J. Inst. Electr. Eng. Part 3A, vol. 93, no. 4, pp. 703–719, Apr. 1946.
[2] J. Brown, “Artificial dielectrics having refractive indices less than unity,”
Proc. IEE Part IV Inst. Monogr., vol. 100, no. 62R, pp. 51–62, Oct. 1953.
Rectangular Apertures [3] C.-C. Chen, “Diffraction of electromagnetic waves by a conducting screen
Assuming that the tangential electric field at the aperture is perforated periodically with circular holes,” IEEE Trans. Microw. Theory
Techn., vol. 19, no. 5, pp. 475–481, May 1971.
oriented as follows: [4] V. S. Asadchy, A. Dıaz-Rubio, S. N. Tcvetkova, D.-H. Kwon, A. Elsakka,
o M. Albooyeh, and S. A. Tretyakov, “Flat engineered multi-channel reflec-
• TM polarization: φ = 90 , Ea (x, y) = Ea (x, y) ŷ
o tors,” Phys. Rev. X, vol. 7, p. 031046, 2017.
• TE polarization: φ = 0 , Ea (x, y) = Ea (x, y) x̂ , [5] B. Blázquez, N. Llombart, D. Cavallo, A. Freni and A. Neto, ”A rigor-
the following spatial profiles can be considered in the case of ous equivalent network for linearly polarized THz absorbers,” IEEE
Trans. Antennas Propag., vol. 62, no. 10, pp. 5077-5088, Oct. 2014.
a rectangular aperture of dimensions a × b [67]: [6] H. Yi, S. Qu, K. Ng, C. K. Wong and C. H. Chan, “Terahertz wavefront
πx x y control on both sides of the cascaded metasurfaces,” IEEE Trans. Antennas
Ea1 (x, y) ∝ cos rect rect (38) Propag., vol. 66, no. 1, pp. 209-216, Jan. 2018.
a a b [7] R. Ulrich, “Far-infrared properties of metallic mesh and its complemen-
tary structure,” Infrared Phys., vol. 7, pp. 37–55, 1967.
cos( πx

a )
x y
Ea2 (x, y) ∝ p rect rect (39) [8] C. Pfeiffer, C. Zhang, V. Ray, L. J. Guo, and A. Grbic, “High perfor-
1 − (2x/a)2 a b mance bianisotropic metasurfaces: asymmetric transmission of light,”
Phys. Rev. Lett., vol. 113, p. 023902, 2014.
where the rectangle function is defined as [9] F. J. Garcia-Vidal, L. Martin-Moreno, T. W. Ebbesen, and L. Kuipers,
( “Light passing through subwavelength apertures,” Rev. Mod. Phys., vol. 82,
x 1 , −u/2 ≤ x ≤ u/2 p. 729, 2010.
rect = (40) [10] R. S. Anwar, L. Mao, and H. Ning, “Frequency selective surfaces: A
u 0 , otherwise . review,” Appl. Sci., vol 8, p. 1689, 2018.
[11] A. Alu, and N. Engheta, “Achieving transparency with plasmonic and
A further study demonstrated that the spatial distribution (38) metamaterial coatings,” Phys. Rev. E, vol. 72, 016623, 2006.
offer more accurate results when the aperture size is narrow, [12] Y. Zhao, M.A. Belkin, and A. Alù, “Twisted optical metamaterials for
planarized ultrathin broadband circular polarizers,” Nat. Comm., vol. 3,
while (39) can be employed for bigger apertures. p. 870, 2012.
The Fourier transforms of these spatial profiles can be [13] C. Molero, M. Garcı́a-Vigueras, R. Rodrı́guez-Berral, F. Mesa, and
expressed in closed form as N. Llombart, “Equivalent circuit approach for practical applications of
meander-line gratings,” IEEE Antennas Wirel. Propag. Lett., vol. 16,
pp. 3088-3091, 2017.
#
cos kxn a2 sin kym 2b
"
[14] G. Perez-Palomino, J. E. Page, M. Arrebola, and J. A. Encinar, “A design
Ẽa1 (kxn , kym ) ∝ p (41) technique based on equivalent circuit and coupler theory for broadband
(π/a)2 − kxn
2 kym linear to circular polarization converters in reflection or transmission
mode,” IEEE Trans. Antennas Propag., vol. 66, no. 5, pp. 2428–2438, May
2018.
π a [15] Y. R. Padooru, A. B. Yakovlev, C. S. R. Kaipa, F. Medina, and F. Mesa,
Ẽa2 (kxn , kym ) ∝ J0 kxn + “Circuit modeling of multiband high-impedance surface absorbers in the
a 2 microwave regime,” Phys. Rev. B, vol. 84, no. 3, pp. 1–11, Jul. 2011.

sin k b [16] F. Costa, A. Monorchio, and G. Manara, “Analysis and design of ul-
π a ym 2 tra thin electromagnetic absorbers comprising resistively loaded high
+ J0 kxn − (42) impedance surfaces,” IEEE Trans. Antennas Propag., vol. 58, no. 5,
a 2 kym pp. 1551–1558, May 2010.
[17] P.-S. Kildal, and A. Kishk, “EM modeling of surfaces with STOP or GO
with J0 (·) being the Bessel function of the first kind and order characteristics - Artificial magnetic conductors and soft and hard Surfaces,”
TM/TE
zero. The transformer turn ratios Nnm can be directly ACES Journal, vol. 18, pp. 32-40, 2003.
calculated with (41) and (42) from [45, Eqs. (6) and (7)]. [18] M. A. Hiranandani, A. B. Yakovlev, and A. A. Kishk, “Artificial magnetic
conductors realised by frequency-selective surfaces on a grounded dielec-
tric slab for antenna applications,” IEE Proc.—Microw. Antennas Propag.,
vol. 153, no. 5, pp. 487-493, 2006.
Annular Apertures
[19] E. Rajo-Iglesias, Ó. Quevedo-Teruel and L. Inclán-Sánchez, “Mutual
The spatial profile of annular apertures shares a great coupling reduction in patch antenna arrays by using a planar EBG structure
and a multilayer dielectric substrate,” IEEE Trans. Antennas Propag.,
similarity with of rectangular apertures. In the case of annular vol. 56, no. 6, pp. 1648-1655, June 2008.
apertures, field variations occur in the azimuth direction φ, [20] Á. Palomares-Caballero, A. Alex-Amor, J. Valenzuela-Valdés and
rather than in x or y directions. According to [68], [69], the P. Padilla, “Millimeter-wave 3D-printed antenna array based on gap-
waveguide technology and split E-plane waveguide,” IEEE Trans. Antennas
following aperture field can be considered in an annular ring Propag., 2020.
whose inner and outer radii are a and b, respectively, [21] R. C. Compton, L. B. Whitbourn, and R. C. McPhedran, ”Strip gratings
at a dielectric interface and application of Babinet’s principle,” Appl. Opt.
Ea (φ) ∝ cos l(φ − φ0 ) ρ̂ (43) 23, pp. 3236-3242, 1984.

[22] R. Rodrı́guez-Berral, C. Molero, F. Medina and F. Mesa, “Analyti- surfaces,” IEEE Antennas Wirel. Propag. Lett., vol. 17, no. 10, pp. 1925-
cal Wideband Model for Strip/Slit Gratings Loaded With Dielectric 1929, Oct. 2018.
Slabs,” IEEE Trans. Microw. Theory Techn., vol. 60, no. 12, pp. 3908-3918, [46] M. D. Astorino, F. Frezza, and N. Tedeschi, “Equivalent-circuit model for
Dec. 2012. stacked slot-based 2D periodic arrays of arbitrary geometry for broadband
[23] M. Beruete et al., “Enhanced millimeter wave transmission through qua- analysis,” J. Appl. Phys., vol. 123, no. 10, Art. no. 103106, Mar. 2018.
sioptical subwavelength perforated plates,” IEEE Trans. Antennas Propag., [47] A. Hessel, M.H. Chen, R.C.M. Li, and A.A. Oliner, “Propagation in peri-
vol. 53, no. 6, pp. 1897-1903, June 2005. odically loaded waveguides with higher symmetries,” Proc. IEEE, vol. 61,
[24] R. Rodrı́guez-Berral, F. Mesa and F. Medina, “Analytical Multimodal no. 2, pp. 183–195, Feb. 1973.
Network Approach for 2-D Arrays of Planar Patches/Apertures Embedded [48] R. Quesada, D. Martı́n-Cano, F. J. Garcı́a-Vidal, and J. Bravo-Abad,
in a Layered Medium,” IEEE Trans. Antennas Propag., vol. 63, no. 5, “Deep-subwavelength negative-index waveguiding enabled by coupled
pp. 1969-1984, May 2015. conformal surface plasmons,” Opt. Lett., vol. 39, no. 10, pp. 2990-2993,
[25] F. Mesa, M. Garcı́a-Vigueras, F. Medina, R. Rodrı́guez-Berral, and 2014.
J. R. Mosig, “Circuit model analysis of frequency selective surfaces with [49] O. Dahlberg, R. C. Mitchell-Thomas, and O. Quevedo-Teruel, “Reducing
scatterers of arbitrary geometry,” IEEE Antennas Wirel. Propag. Lett., the dispersion of periodic structures with twist and polar glide symmetries,”
vol. 14, pp. 135–138, 2015. Sci. Rep., vol. 7, Aug. 2017.
[26] F. Mesa, R. Rodriguez-Berral, M. Garcı́a-Vigueras, F. Medina, and [50] A. Palomares-Caballero, A. Alex-Amor, P. Padilla, and J. F. Valenzuela-
J. R. Mosig, “Simplified modal expansion to analyze frequency-selective Valdés, “Dispersion and Filtering Properties of Rectangular Waveguides
surfaces: An equivalent circuit approach,” IEEE Trans. Antennas Propag., Loaded with Holey Structures,” IEEE Trans. Microw. Theory Techn., 2020.
vol. 46, no. 3, pp. 1106–1111, 2016. [51] Q. Chen, F. Ghasemifard, G. Valerio, and O. Quevedo-Teruel, “Modeling
[27] G. Xu, S. V. Hum, and G. V. Eleftheriades, “A technique for design- and dispersion analysis of coaxial lines with higher symmetries,” IEEE
ing multilayer multistopband frequency selective surfaces,” IEEE Trans. Trans. Microw. Theory Techn., vol. 66, no. 10, pp. 4338-4345, Oct. 2018.
Antennas Propag., vol. 66, no. 2, pp. 780–789, Feb. 2018. [52] T. Chang, J. U. Kim, S. K. Kang, H. Kim, D. K. Kim, Y.-H. Lee, and
[28] G. Perez-Palomino and J. E. Page, “Bimode Foster’s equivalent circuit J. Shin, “Broadband giant-refractive-index material based on mesoscopic
of arbitrary planar periodic structures and its application to design polar- space-filling curves,” Nat. Comm., vol. 7, Article number 12661, 2016.
ization controller devices,” IEEE Trans. Antennas Propag., vol. 68, no. 7, [53] F. Ghasemifard, M. Norgren, and O. Quevedo-Teruel, “Twist and polar
pp. 5308–5321, July 2020. glide symmetries: An additional degree of freedom to control the prop-
[29] M. Borgese and F. Costa, “A simple equivalent circuit approach for agation characteristics of periodic structures,” Sci. Rep., vol. 8, Jul. 2018,
anisotropic frequency selective surfaces and metasurfaces,” IEEE Trans. Art. no. 11266.
Antennas Propag., vol. 68, no. 10, pp. 7088–7098, Oct. 2020. [54] A. Alex-Amor, F. Ghasemifard, G. Valerio, M. Ebrahimpouri, P, Padilla,
[30] M. Camacho, F. Capolino, and M. Albani, “Asymmetric surface wave J. M. Fernández-González, and O. Quevedo-Teruel, “Glide-Symmetric
excitation through metasurface-edge diffraction,” Opt. Lett., 2020. Metallic Structures With Elliptical Holes for Lens Compression,” IEEE
[31] R. E. Collin. Field Theory of Guided Waves, 2nd ed. New York: IEEE Trans. Microw. Theory Techn., 2020.
Press, 1991. [55] C. Molero, R. Rodrı́guez-Berral, F. Mesa, and F. Medina, “Wideband an-
[32] J. E. Varela and J. Esteban, “Characterization of waveguides with a alytical equivalent circuit for coupled asymmetrical nonaligned slit arrays,”
combination of conductor and periodic boundary contours: Application to Phys. Rev. E, vol. 95, p. 023303 2017.
the analysis of bi-periodic structures,” IEEE Trans. Microw. Theory Techn., [56] G. Valerio, Z. Sipus, A. Grbic and O. Quevedo-Teruel, “Accurate
vol. 60, no. 3, pp. 419-430, March 2012. Equivalent-Circuit Descriptions of Thin Glide-Symmetric Corrugated
[33] K. Kurokawa, An Introduction to the Theory of Microwave Circuits, Metasurfaces,” IEEE Trans. Antennas Propag., vol. 65, no. 5, pp. 2695-
Ed. Academic Press, New York, 1969. 2700, May 2017.
[34] F. Mesa, R. RodrÍguez-Berral, and F. Medina, “Unlocking complexity [57] D. M. Pozar, Microwave Enginnering, 3rd Ed. Hoboken, NJ, USA: Wi-
using the ECA,” IEEE Microw. Mag., vol. 19, no. 4, pp. 44–65, Jun. 2018. ley, 2005.
[35] I. Palocz and A. A. Oliner, “Equivalent network of a multimode planar [58] F. Mesa, R. Rodrı́guez-Berral, and F. Medina, “On the limitations of
grating,” IEEE Trans. Microw. Theory Techn., vol. 18, no. 5, pp. 244–252, equivalent circuits for the modeling of periodic structures,” in 2017 IEEE
May 1970. MTT-S International Microwave Workshop Series on Advanced Materials
[36] S. Monni, G. Gerini, A. Neto, and A. G. Tijhuis, “Multi-mode equivalent and Processes for RF and THz Applications (IMWS-AMP), Pavia, 2017,
networks for the design and analysis of frequency selective surfaces,” IEEE pp. 1-3.
Trans. Antennas Propag., vol. 55, no. 10, pp. 2824–2835, Oct. 2007. [59] M. Bagheriasl, O. Quevedo-Teruel, and G. Valerio, “Bloch analysis
[37] C. Molero, R. Rodrı́guez-Berral, F. Mesa, and F. Medina, “Analytical of artificial lines and surfaces exhibiting glide symmetry,” IEEE
Circuit Model for 1-D Periodic T-Shaped Corrugated Surfaces,” IEEE Trans. Microw. Theory Techn., vol. 67, no. 7, pp. 2618–2628, 2019.
Trans. Antennas Propag., vol. 62, no. 2, pp. 794-803, Feb. 2014. [60] O. Quevedo-Teruel, Q. Chen, F. Mesa, G. Valerio, and N. Fonseca “On
[38] R. Marqués, L. Jelinek, F. Mesa, and F. Medina, “Analytical theory of the benefits of glide symmetry for microwave devices,” to appear in IEEE
wave propagation through stacked fishnet metamaterials,” Opt. Express, J. Microwaves, vol. 1, no. 1, 2021.
vol. 17, no. 14, pp. 11582–11593, July 2009. [61] Z. Wei, Y. Cao, Y. Fan, X. Yu, and H.Li, “Broadband transparency
[39] V. Torres, P. Rodrı́guez-Ulibarri, M. Navarro-Cı́a, and M. Beruete, “Fish- achieved with the stacked metallic multi-layers perforated with coaxial
net metamaterial from an equivalent circuit perspective,” Appl. Phys. Lett., annular apertures,” Opt. Express, vol. 19, pp. 21425-21431, 2011.
vol. 101, no. 24, p. 244101, Dec. 2012. [62] C. V. Jakowatz, D. Wahl, P. Eichel, D. C. Ghiglia, and P. A. Thompson,
[40] Y. He and G. V. Eleftheriades, “A Thin Double-Mesh Metamaterial Spotlight-Mode Synthetic Aperture Radar: A Signal Processing Approach.
Radome for Wide-Angle and Broadband Applications at Millimeter-Wave Norwell, MA: Kluwer Academic, 1996.
Frequencies,” IEEE Trans. Antennas Propag., vol. 68, no. 3, pp. 2176-2185, [63] B. S. Reddy and B. N. Chatterji, “An FFT-based technique for transla-
March 2020. tion, rotation, and scale-invariant image registration,” IEEE Trans. Image
[41] J. Carbonell, C. Croenne, F. Garet, E. Lheurette, J. L. Coutaz, and Process., vol. 5, no. 8, pp. 1266–1271, 1996.
D. Lippens, “Lumped elements circuit of terahertz fishnet-like arrays with [64] F. Mesa, R. Rodrı́guez-Berral, and F. Medina, “On the computation of the
composite dispersion,” J. Appl. Phys., vol. 108, no. 1, p. 014907, Jul. 2010. dispersion diagram of symmetric one-dimensionally periodic structures,”
[42] M. Kafesaki, I. Tsiapa, N. Katsarakis, T. Koschny, C. M. Soukoulis, and Symmetry, vol. 10, no. 8, p. 307, 2018.
E. N. Economou, “Left-handed metamaterials: The fishnet structure and its [65] F. Mesa, G. Valerio, R. Rodrı́guez-Berral, and O. Quevedo-Teruel,
variations,” Phys. Rev. B, vol. 75, no. 23, p. 235114, Jun. 2007. “Simulation-Assisted Efficient Computation of the Dispersion Diagram of
[43] X. Liu, W. Li, Z. Zhao, R. Lu, S. Zhu, Z. Xu, X. Chen, and A. Periodic Structures,” to appear in IEEE Antennas Propag. Mag. 2020, doi:
Zhang, “Tangential network transmission theory of reflective metasurface 10.1109/MAP.2020.3003210.
with obliquely incident plane waves,” IEEE Trans. Microw. Theory Techn., [66] C. Molero, R. Rodrı́guez-Berral, F. Mesa, F. Medina, and A. B. Yakovlev,
vol. 66, no. 1, pp. 64–72, Jan. 2018. “Wideband analytical equivalent circuit for one-dimensional periodic
[44] V. Torres, F. Mesa, M. Navarro-Cı́a, R. Rodrı́guez-Berral, M. Beruete and stacked arrays,” Phys. Rev. E, vol. 93, p. 013306, 2016.
F. Medina, ”Accurate circuit modeling of fishnet structures for negative- [67] Sembiam R. Rengarajan, “Choice of Basis Functions for Accurate Char-
index-medium applications,” IEEE Trans. Microw. Theory Techn., vol. 64, acterization of Infinite Array of Microstrip Reflectarray Elements,” IEEE
no. 1, pp. 15-26, Jan. 2016. Antennas Wirel. Propag. Lett., vol. 4, pp. 47-50, 2005.
[45] F. Mesa, R. Rodrı́guez-Berral, M. Garcı́a-Vigueras and F. Medina, “Effi- [68] P. Rodrı́guez-Ulibarri, M. Navarro-Cı́a, R. Rodrı́guez-Berral, F. Mesa,
cient hybrid full-wave/circuital approach for stacks of frequency selective F. Medina and M. Beruete, “Annular Apertures in Metallic Screens as

Extraordinary Transmission and Frequency Selective Surface Struc- Pablo Padilla was born in Jaén, Spain, in 1982.
tures,” IEEE Trans. Microw. Theory Techn., vol. 65, no. 12, pp. 4933-4946, He received the Telecommunication Engineering
Dec. 2017. degree and the Ph.D. degree from the Radiation
[69] R. Dubrovka, J. Vazquez, C. Parini, and D. Moore, “Equivalent cir- Group (Signal, Systems and Radiocommunications
cuit method for analysis and synthesis of frequency selective sur- Department) of the Technical University of Madrid
faces,” Proc. Inst. Elect. Eng.—Microw., Antennas Propag., vol. 153, no. 3, (UPM), Spain, in 2005 and 2009, respectively. In
pp. 213–220, Jun. 2006. 2007, he was with the Laboratory of Electromagnet-
ics and Acoustics, École Polytechnique Fédérale de
Lausanne, Switzerland, as an invited Ph.D. Student.
In 2009, he carried out a Postdoctoral stay at the
Helsinki University of Technology (AALTO-TKK).
In 2009, he became Assistant Professor at the Signal Theory, Telematics
Antonio Alex-Amor received the B.Sc. degree
and Communications Department of the University of Granada, where he is
in telecommunication engineering from Universi-
currently Associate Professor, since 2012. In 2017, he was an invited Visiting
dad de Granada in 2016, and the M.Sc. degree
Professor at the Royal Institute of Technology of Stockholm. He has authored
in telecommunication engineering from Universidad
more than 65 high-impact journal contributions and more than 60 contributions
Politécnica de Madrid (UPM) in 2018, where he is
to international symposia. His research interests include a variety of topics
currently pursuing the Ph.D. degree. Since 2016, he
related mainly to electromagnetism and communication issues (radiofrequency
has been with the Radiation Group, Signal, Systems
devices, antennas and propagation).
and Radiocommunications Department, UPM. From
2018-2019, he joined the Department of Language
and Computer Science, Universidad de Málaga. In
2020, he joined the Departamento de Teorı́a de
la Señal, Telemática y Telecomunicaciones, Universidad de Granada. He
received the Best Electromagnetics Paper Award at the 14th European
Conference on Antennas and Propagation (EuCAP 2020). His current research
interests include the use of liquid crystal as tunable dielectric, metamaterials,
structures with higher symmetries and radiofrequency energy harvesting
systems.

Francisco Mesa (Fellow, IEEE) received the Li-

cenciado and Ph.D. degrees in physics from the
Universidad de Sevilla, Seville, Spain, in 1989 and
1991, respectively. He is currently a Professor with
the Departamento de Fı́sica Aplicada 1, Universidad
de Sevilla. His research interest includes electromag-
netic propagation/radiation in planar structures.

Ángel Palomares-Caballero was born in Jaén,

Spain, in 1994. He received the B.Sc. and M.Sc.
degrees in telecommunication engineering from the
University of Granada (UGR), Granada, Spain, in
2016 and 2018, respectively, where he is currently
pursuing the Ph.D. degree at a National Pre-Doctoral
Fellowship. Since 2017, he has been with the Signal
Theory, Telematics and Communications Depart-
ment, University of Granada. His current research
interests include millimeter-wave antennas and phase
shifters, gap-waveguide technology, and structures
with higher symmetries.

Carlos Molero (Member, IEEE) was born in

Seville, Spain, in April 1987. He received the Li-
cenciado and Ph.D. degrees in physics from the
Universidad de Sevilla, Seville, in 2011 and 2017,
respectively. Since March 2017, he has been holding
a post-doctoral position at INSA Rennes, Rennes,
France. His research interests focus on the study of
periodic structures, both in planar and 3-D archi-
tectures, circuit models, full-metal devices, and new
conception of polarizers based on 3-D printable self-
supported cells. Dr. Molero was a recipient of some
prizes, such as the Best Engineer Prize in the European Microwave Conference
of 2015 in Paris, France.

2014-Reconfigurable Reflectarrays and Array Lenses For Dynamic Antenna Beam Control A Review
No ratings yet
2014-Reconfigurable Reflectarrays and Array Lenses For Dynamic Antenna Beam Control A Review
16 pages
Efficient Finite Element Analysis of Axially Symmetrical Waveguides and Waveguide Discontinuities - 41525
No ratings yet
Efficient Finite Element Analysis of Axially Symmetrical Waveguides and Waveguide Discontinuities - 41525
7 pages
Wave Scattering by A Cylindrical Metasur
No ratings yet
Wave Scattering by A Cylindrical Metasur
11 pages
11.Design_of_Perfect_Absorber_Based_on_Metagratings_Theory_and_Experiment
No ratings yet
11.Design_of_Perfect_Absorber_Based_on_Metagratings_Theory_and_Experiment
11 pages
Applsci 08 01689 v3
No ratings yet
Applsci 08 01689 v3
46 pages
Rapid_Seismic_Waveform_Modeling_and_Inversion_With_Neural_Operators
No ratings yet
Rapid_Seismic_Waveform_Modeling_and_Inversion_With_Neural_Operators
12 pages
Antenna Pattern Synthesis: A New General Approach
No ratings yet
Antenna Pattern Synthesis: A New General Approach
14 pages
Politecnico Di Torino Porto Institutional Repository
No ratings yet
Politecnico Di Torino Porto Institutional Repository
26 pages
Scattering of Spherically and Hemispherically Stratified Lenses Fed by Any Real Source
No ratings yet
Scattering of Spherically and Hemispherically Stratified Lenses Fed by Any Real Source
11 pages
1 s2.0 S0263822322010728 Main
No ratings yet
1 s2.0 S0263822322010728 Main
16 pages
Applied Acoustics: Marco Caniato, Federica Bettarello, Paolo Bonfiglio, Andrea Gasparella
No ratings yet
Applied Acoustics: Marco Caniato, Federica Bettarello, Paolo Bonfiglio, Andrea Gasparella
17 pages
Yuju
No ratings yet
Yuju
9 pages
Focusing On Bandwidth: Achromatic Metalens Limits: F P F M
No ratings yet
Focusing On Bandwidth: Achromatic Metalens Limits: F P F M
8 pages
Pstruct 2021 113645
No ratings yet
Pstruct 2021 113645
8 pages
IJMS
No ratings yet
IJMS
29 pages
Enhanced Method For Determining The Optical Response of Highly Complex Biological Photonic Structures
No ratings yet
Enhanced Method For Determining The Optical Response of Highly Complex Biological Photonic Structures
14 pages
Inductive Frequency Selective Surface An Applicati
No ratings yet
Inductive Frequency Selective Surface An Applicati
13 pages
On The Computation of The Dispersion Diagram of Symmetric One-Dimensionally Periodic Structures
No ratings yet
On The Computation of The Dispersion Diagram of Symmetric One-Dimensionally Periodic Structures
15 pages
1 s2.0 S0030401820310646 Main
No ratings yet
1 s2.0 S0030401820310646 Main
7 pages
Reconfigurable Bidirectional Beam-Steering Aperture With Transmitarray Reflectarray and Transmit-Reflect-Array Modes Switching
No ratings yet
Reconfigurable Bidirectional Beam-Steering Aperture With Transmitarray Reflectarray and Transmit-Reflect-Array Modes Switching
15 pages
2018 Design of Simplified Array Layouts For Preliminary Experimental Testing and Validation of Large AESAs
No ratings yet
2018 Design of Simplified Array Layouts For Preliminary Experimental Testing and Validation of Large AESAs
15 pages
2022 Advanced STEM Porous Surfaces
No ratings yet
2022 Advanced STEM Porous Surfaces
15 pages
Analysis and Characterization of A Wide-Angle Impedance Matching Metasurface For Dipole Phased Arrays
No ratings yet
Analysis and Characterization of A Wide-Angle Impedance Matching Metasurface For Dipole Phased Arrays
11 pages
Chuaqui 2017
No ratings yet
Chuaqui 2017
14 pages
1 s2.0 S1434841123003023 Main
No ratings yet
1 s2.0 S1434841123003023 Main
12 pages
Flexible Controls of Terahertz Waves Using Coding and Programmable Metasurfaces
No ratings yet
Flexible Controls of Terahertz Waves Using Coding and Programmable Metasurfaces
13 pages
2017 - Broadband Vortex Beam Generation
No ratings yet
2017 - Broadband Vortex Beam Generation
5 pages
Multipole and Multimode Engineering in Mie Resonan
No ratings yet
Multipole and Multimode Engineering in Mie Resonan
23 pages
Lightning Efficient Counterpoise Configurations For Transmission Line Grounding
No ratings yet
Lightning Efficient Counterpoise Configurations For Transmission Line Grounding
12 pages
A_Scalable_Physics-Based_Compact_Model_for_Terminal_Charge_Intrinsic_Capacitance_and_Drain_Current_in_Nanosheet_Field_Effect_Transistors
No ratings yet
A_Scalable_Physics-Based_Compact_Model_for_Terminal_Charge_Intrinsic_Capacitance_and_Drain_Current_in_Nanosheet_Field_Effect_Transistors
8 pages
Faceting For Direction-Dependent Spectral Deconvolution
No ratings yet
Faceting For Direction-Dependent Spectral Deconvolution
15 pages
Extending Spectral Factorization To Array Pattern Synthesis Including Sparseness, Mutual Coupling, and Mounting-Platform Effects
No ratings yet
Extending Spectral Factorization To Array Pattern Synthesis Including Sparseness, Mutual Coupling, and Mounting-Platform Effects
12 pages
Projet Hosseini2015
No ratings yet
Projet Hosseini2015
11 pages
1 s2.0 S0020740322008670 Main
No ratings yet
1 s2.0 S0020740322008670 Main
13 pages
A Broadband RCS Reduction Coating Using a Novel Ar
No ratings yet
A Broadband RCS Reduction Coating Using a Novel Ar
11 pages
Wang 2021
No ratings yet
Wang 2021
10 pages
1 s2.0 S0264127522001903 Main
No ratings yet
1 s2.0 S0264127522001903 Main
7 pages
Energies 15 04893 v2
No ratings yet
Energies 15 04893 v2
40 pages
A_Pattern-Reconfigurable_Cylindrical_Dielectric_Resonator_Antenna_With_Three_Switchable_Radiation_Patterns
No ratings yet
A_Pattern-Reconfigurable_Cylindrical_Dielectric_Resonator_Antenna_With_Three_Switchable_Radiation_Patterns
10 pages
Iftimie 2020 IOP Conf. Ser. Mater. Sci. Eng. 997 012023
No ratings yet
Iftimie 2020 IOP Conf. Ser. Mater. Sci. Eng. 997 012023
8 pages
Impedance Calculation of A Bobbin Coil I
No ratings yet
Impedance Calculation of A Bobbin Coil I
8 pages
A Laminated Structural Finite Element for the Behavior of Large Non-linear Reinforced Concrete Structures
No ratings yet
A Laminated Structural Finite Element for the Behavior of Large Non-linear Reinforced Concrete Structures
17 pages
Deep Learning 08988246
No ratings yet
Deep Learning 08988246
12 pages
sensors-21-02765
No ratings yet
sensors-21-02765
25 pages
A_Miniaturized_Frequency_Selective_Rasorber_With_Independently_Regulated_Selective_Dual-Transmission_Response
No ratings yet
A_Miniaturized_Frequency_Selective_Rasorber_With_Independently_Regulated_Selective_Dual-Transmission_Response
5 pages
Unit_Cell_Design_for_Aperiodic_Metasurfaces
No ratings yet
Unit_Cell_Design_for_Aperiodic_Metasurfaces
8 pages
Wave Propagation in Metamaterial Multi-Layered Structures: ST ND RD
No ratings yet
Wave Propagation in Metamaterial Multi-Layered Structures: ST ND RD
6 pages
Design and Analysis of 1-D Uniform and Chirped Electromagnetic Bandgap Structures in Substrate-Integrated Waveguides
No ratings yet
Design and Analysis of 1-D Uniform and Chirped Electromagnetic Bandgap Structures in Substrate-Integrated Waveguides
9 pages
1 s2.0 S0263822323009315 Main
No ratings yet
1 s2.0 S0263822323009315 Main
17 pages
1-s2.0-S0997753822002790-main
No ratings yet
1-s2.0-S0997753822002790-main
16 pages
2020-IEEE AWPL-Dual-band Tunable FSS-Saptarshi
No ratings yet
2020-IEEE AWPL-Dual-band Tunable FSS-Saptarshi
5 pages
Equivalent-Circuit Models For The Design of Metamaterials Based On Artificial Magnetic Inclusions
No ratings yet
Equivalent-Circuit Models For The Design of Metamaterials Based On Artificial Magnetic Inclusions
9 pages
2.
No ratings yet
2.
56 pages
Analysis, Design and Applications of Re Ectarrays: January 2008
No ratings yet
Analysis, Design and Applications of Re Ectarrays: January 2008
7 pages
Computational Model For Low Cycle Fatigue Analysis of Lattice Materials - Incorporating Theory of Critical Distance With Elastoplastic Homogenization
No ratings yet
Computational Model For Low Cycle Fatigue Analysis of Lattice Materials - Incorporating Theory of Critical Distance With Elastoplastic Homogenization
13 pages
Metasurfaces
No ratings yet
Metasurfaces
7 pages
Nano Materials
No ratings yet
Nano Materials
42 pages
1 s2.0 S0307904X15006058 Main
No ratings yet
1 s2.0 S0307904X15006058 Main
14 pages
Finite Element Method
From Everand
Finite Element Method
Gouri Dhatt
1/5 (1)
Mechanical Properties of Nanostructured Materials: Quantum Mechanics and Molecular Dynamics Insights
From Everand
Mechanical Properties of Nanostructured Materials: Quantum Mechanics and Molecular Dynamics Insights
Abdolhossein Fereidoon
No ratings yet
Gernsback 44 Basic Radio Course Frye
100% (1)
Gernsback 44 Basic Radio Course Frye
176 pages
Technical Bulletins Miglow Underwater12
No ratings yet
Technical Bulletins Miglow Underwater12
1 page
Physics (Code No. 042) Maximum Marks: 40 TIME: 2 Hours General Instructions
No ratings yet
Physics (Code No. 042) Maximum Marks: 40 TIME: 2 Hours General Instructions
4 pages
DITHER
No ratings yet
DITHER
9 pages
B.Sc. (H) Physics-2nd Semester-2019
No ratings yet
B.Sc. (H) Physics-2nd Semester-2019
9 pages
20-21 AS Prelim Ms
No ratings yet
20-21 AS Prelim Ms
27 pages
Transformers
No ratings yet
Transformers
25 pages
Link - MASM TOUR To MASM MTN - Proposal - Report
No ratings yet
Link - MASM TOUR To MASM MTN - Proposal - Report
5 pages
CH 12 Non Linear Optics
No ratings yet
CH 12 Non Linear Optics
26 pages
Xam Idea Physics - Chapter2
No ratings yet
Xam Idea Physics - Chapter2
52 pages
Us Grounding Bonding Applications Handbook
No ratings yet
Us Grounding Bonding Applications Handbook
28 pages
Installation Guide For PR311SE/PR311SE-BK Access Controllers
No ratings yet
Installation Guide For PR311SE/PR311SE-BK Access Controllers
15 pages
Power Your Signal: Antenna Specifications
No ratings yet
Power Your Signal: Antenna Specifications
3 pages
01 - Chapter 09 Sinusoids and Phasors
No ratings yet
01 - Chapter 09 Sinusoids and Phasors
33 pages
Class XII Level 1 Assignment
No ratings yet
Class XII Level 1 Assignment
72 pages
Classification of Spectros
No ratings yet
Classification of Spectros
2 pages
Waves For Communication P5
No ratings yet
Waves For Communication P5
2 pages
Tachyon Energy Easy Read
No ratings yet
Tachyon Energy Easy Read
27 pages
1lba4550 401
No ratings yet
1lba4550 401
13 pages
Physics Investigatory Project LDR
No ratings yet
Physics Investigatory Project LDR
19 pages
Syllabus For TGT Science
No ratings yet
Syllabus For TGT Science
3 pages
SL Paper2
No ratings yet
SL Paper2
26 pages
Experiment 1 - Group 6 PDF
No ratings yet
Experiment 1 - Group 6 PDF
4 pages
2021 780 Mock Paper2 Questions
No ratings yet
2021 780 Mock Paper2 Questions
6 pages
Internal Electrification Rate Analysis
No ratings yet
Internal Electrification Rate Analysis
16 pages
UENR8264UENR8264-01 - SIS Manual Deif m1 Marino
100% (1)
UENR8264UENR8264-01 - SIS Manual Deif m1 Marino
2 pages
Hand Crank Flashlight
No ratings yet
Hand Crank Flashlight
20 pages
ELP-01 Modern Physics
No ratings yet
ELP-01 Modern Physics
3 pages
Quantum Physics Final (1) .Docx Syllabus
No ratings yet
Quantum Physics Final (1) .Docx Syllabus
3 pages
Generators in Parallel Operation
No ratings yet
Generators in Parallel Operation
24 pages

Exploring the Potentials of the Multi-modal Equivalent Circuit Approach for Stacks of 2-D Aperture Arrays-IEEE TAP.2021

Uploaded by

Exploring the Potentials of the Multi-modal Equivalent Circuit Approach for Stacks of 2-D Aperture Arrays-IEEE TAP.2021

Uploaded by

This article has been accepted for publication in a future issue of this journal, but has not been

Exploring the Potentials of the Multi-modal

the stacked structure, especially with the inclusion of some of

plus 2 additional external blocks. Thus, the complete network

Bloch mode in the glide-symmetric FSS. However, note that

seen that each single metallic screen is periodically perforated

thickness of . λ0 /20 at the full-absorption peaks observed in

IV. A RBITRARY A PERTURES

efficiency of the present hybrid implementation to compute

V. I NFINITE P ERIODIC S TACKS

where the sum above extends to both TE and TM modes.

the inner radius of the annular-ring apertures varies. It can

Francisco Mesa (Fellow, IEEE) received the Li-

Ángel Palomares-Caballero was born in Jaén,

Carlos Molero (Member, IEEE) was born in

You might also like