> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
1
Near Field Enhancement and Subwavelength
Imaging Using Resonantly Loaded Apertures
Oleksandr Malyuskin, Member, IEEE, Vincent Fusco, Fellow, IEEE

Abstract— It is demonstrated that the electromagnetic (EM)
transmission through a subwavelength or non-resonant aperture
in a conductive screen can be dramatically enhanced by loading it
with folded metallic strips exhibiting resonant properties. When
illuminated by an EM plane wave these loaded apertures enable
very tight, subwavelength, collimation of the EM power in the
near field zone. We propose planar and quasi-planar resonant
insertion geometries that should allow, for the first time, twodimensional dual-polarization subwavelength field confinement
along with ability to focus both electric and magnetic fields.
The proposed technique for resonance transmission
enhancement and near field confinement forms a basis for a new
class of microwave near field imaging probe with subwavelength
resolution capable of operating over a wide range of imaging
distances (0.05-0.25λ). Measurement results demonstrate the
possibility of high contrast (more than 3dB in amplitude and 40
degrees in phase) near field subwavelength imaging of 2D and 3D
resonant and non-resonant metallic and dielectric targets in free
space and in moderately lossy layered media.
Index Terms— Enhanced transmission, focusing, near field
enhancement, resonance, sub-diffraction imaging, subwavelength
aperture.
I. INTRODUCTION
H
RESOLUTION near field electromagnetic (EM)
imaging is important in a number of application including
materials characterization [1], biomolecular and medical noninvasive probing [2], non-destructive testing [3], etc.
At present, there are a number of different near-field
imaging techniques applicable across different spectral ranges.
These techniques can be broadly divided into two categories:
aperture-based and apertureless near field probes. In the first
case the probe consists of a small, typically subwavelength,
aperture in a metallic screen [4]. Alternatively it can be an
open waveguide loaded with a dielectric insert [5]. The object
to be imaged is positioned in close vicinity of the aperture and
illuminated by an EM wave. The radiation scattered off the
object is transmitted through the aperture and is collected by
an EM radiation detector where the properties of the imaged
object are estimated by measuring the detector output current.
In this scenario, the spatial resolution of an aperture-based
probe is determined by the size of the aperture [6], [7].
IGH
Date submitted. This work was supported in part by the Leverhulme Trust,
UK .
The authors are with The Institute of Electronics, Communications and
Information Technology, Queen's University Belfast, Queen's Rd, Belfast,
BT3 9DT, UK. e-mails:o.malyuskin@qub.ac.uk; v.fusco@ecit.qub.ac.uk.
In the apertureless imaging scheme EM radiation is
scattered by a sharp tip of a probe located in the vicinity of an
imaged object and collected in the far or near field [8].
Knowing the scattering characteristics of a probe it is possible
to reconstruct the object properties [8]. In this scenario image
spatial resolution depends on the probe tip dimensions,
geometry and the probe-to-object separation distance.
These techniques can also be used for the active source
characterization where the radiation from the source can be
collected through the aperture or by a small coaxial probe [9].
The above mentioned techniques have key limitations. The
main limiting factor that restricts the use of small apertures is
a dramatic reduction of the transmitted power as aperture size
decreases, proportional to the third power of the aperture size
[6], [7]. This leads to the necessity of using sources of
considerable power or large signal amplification. The
availability of these imparts practical difficulties especially at
millimeter and sub-millimeter wavelengths. In the apertureless
scenario the operating distance is extremely small, typically
λ/50- λ/100 [8], λ is a wavelength of radiation, leading to the
fast deterioration of the image as the probe tip is located
farther away from the sample. This compounds the
interpretation of the images obtained, which in general, is not
a trivial task [10] requiring some prior knowledge of the
imaged scene and de-embedding of the probe tip intrinsic
scattering properties.
In this paper we propose a simple method which overcomes
the main drawbacks of previous aperture-based near field
probe solutions by using resonantly loaded apertures. It was
first noted in [11] that the electrically small apertures can
exhibit resonance transmission properties in the presence of a
conductive object. Several structures that can enhance the
transmission through a small hole based on this principle have
been proposed in recent years, particularly metal gratings [12]
or metamaterials [13], [14]. In [15] a very simple arrangement
consisting of a rectangular slot aperture loaded with strip has
been proposed. It was shown that such a loaded aperture
enables up to 25dB transmitted field enhancement with respect
to the unloaded slot transmission in the L or S band.
Additionally superior subwavelength near field confinement
was observed with transmitted beam full width at half
maximum (FWHM) measured approximately one tenth of a
wavelength at 0.1λ distance from the slot. This probe
demonstrated very high-contrast (with more than 10dB
discrimination) near field subwavelength resolution of
metallic scatterers separated by less than quarter-wavelength
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
at 0.1λ to 0.2λ imaging distance.
In the work to be presented here we significantly extend the
results of our previous studies [15] in two ways.
First, we propose several new insert and aperture
geometries that result in significant further improvement of
loaded aperture probe performance in terms of the near field
enhancement, spatial squeezing and enhanced imaging
contrast. In particular we propose planar and quasi-planar
resonant insertion geometries that allow, for the first time,
two-dimensional dual-polarization subwavelength field
confinement along with the ability to focus both electric and
magnetic fields. Secondly, the experimental results we present
demonstrate significant sensitivity of the resonantly loaded
aperture probe to the dielectric properties of the imaged
samples. This makes the approach a promising candidate for
dielectric material characterization.
It should be noted that while papers such as [16]-[18] deal
with the problem of slot–strip interaction and the transmission
properties of such arrangements, these works only consider the
case of orthogonal complementary structures. The geometries
considered below lead to fundamentally new scattering effects
not studied before.
2
some important points have to be mentioned.
(a)
II. PRINCIPLE OF OPERATION OF RESONANTLY LOADED
APERTURE
A resonantly loaded slot aperture in a metallic screen
illuminated by a plane wave is shown in Fig.1(a). In this
arrangement, the lengths of the slot and the strip are 0.6λ and
0.5λ and respective widths are 0.1λ and 0.05λ. In the absence
of the strip insert, the transmitted EM field, when co-polarized
to the slot, is significantly attenuated, Fig. 1(b), dotted line. In
this graph the transmitted electric field amplitude is plotted as
a function of the transversal range x, at z = -0.1λ, i.e. in the
near field zone. This transmitted field attenuation occurs due
to the unloaded slot radiation mechanism [17] which in this
case is cut-off for the radiating magnetic currents. Insertion of
a half-wavelength strip results in electric currents being
supported that can re-radiate, generating resonantly enhanced
EM field transmission whose electric field component is
polarized along the slot, Fig.1(b), solid line. Additional near
field spatial squeezing of the EM field transmitted through the
loaded aperture, solid line, as compared to the scattered by a
standalone strip in free space, dashed line, Fig.1(b) is also
observed. This can be explained in simple terms by the fact
that at the metal boundaries of the slot the tangential field is
zero which confines the spatial signature of the transmitted
near field in transversal directions and leads to its amplitude
enhancement in the “hot spot”. In Fig.1(b), the FWHM of the
transmitted by the resonantly loaded aperture near field is
0.18λ at z = -0.1λ, while the FWHM of the near field due to
the unloaded slot aperture and strip in free space are both
0.26λ at the same distance z = -0.1λ.
The scattering properties of the slot can be qualitatively
understood on the basis of the system of coupled integral
electric field equations which can be represented in terms of
the equivalent circuit network [19]. The detailed analysis of
this problem is out of the scope of the present paper however
(b)
Fig.1 Resonant near field enhancement and spatial confinement due to loaded
aperture. (a) Loaded aperture geometry; (b) FEKO simulated near field
amplitude across the transversal range at the z = -0.1λ offset distance.
First, it should be noted that both the Ex and Ey components
of the electric field inside a gap between the loaded narrow
slot and a strip are significantly enhanced (compared to the
incident field of unit amplitude), due to strong near field
coupling, Fig.2.
Fig.2. Simulated (FEKO) electric field magnitude in the gap between the strip
of width 0.02λ and slot (0.1λ wide). The field is plotted near the strip end at
y=0.2λ, z=0 where the electric field intensity is large. Red lines show strip and
slot edges.
However, in the narrow slots (width<0.2λ) the transversal Ex
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
3
component is dominating over the longitudinal component Ey
everywhere in the gap area. This can be explained by the fact
that the last term in the mixed potential integral equation [19]
 
 exp  jkR   
E r    j  0
J r dr 
4R
S
exp  jkR  
 
 r dr 
40 R
S
(1)
contains spatial derivatives resulting in the terms with
~ 1 R 2 , ~ 1 R 3 spatial dependence as opposed to the ~ 1 R
 
 
dependence of the first term. In (1) J r  and  r  are the
current and electric charge density on the strip, the first term
defines the equivalent inductance while the second term
contributes to the equivalent capacitance [19] of the loaded
aperture which can be viewed as a capacitance of the strip
insert in free space and additional series quasi-static
capacitance due to near field coupling. When the gap width is
reduced the quasi-static capacitance is also reduced
proportionally to the ratio K 1   2 K   , where ν is the
ratio of strip width to the gap width and K   is the complete
elliptic integral of the first kind [20], [21]. The considered
loaded aperture EM transmission properties will be studied
experimentally in the next section.

(a)

III. NEAR FIELD PROBE BASED ON THE LOADED APERTURE
AND ITS CHARACTERISTIC RESOLUTION
A. Near Field Imaging Probe Description
A microwave probe can be composed using a resonantly
loaded aperture backed by a cavity or horn antenna, Fig.3(a).
In our experiment the measurement setup consists of a
pyramidal horn antenna excited with a vertical E-field
polarized signal and a conductive plate with resonantly loaded
aperture, covering a pyramidal horn with aperture dimensions
140mm x 105mm. Microwave imaging is based on the
measurements of the reflection coefficient S11 at the coaxial
excitation port of the horn antenna. Resonant EM transmission
due to the loaded aperture leads to significant near field
enhancement in a tight focal spot in the vicinity of the probe
insert. This, in its turn, leads to very strong near field coupling
of the probe to objects positioned within the imaging scene,
Fig. 3(b). The geometry and EM properties of these objects
can be characterized by the variation of the reflection
coefficient S11 as the probe scans the scene.
B. Microwave Probe Transmission Enhancement and Near
Field Collimation
In the previous section it has been shown that the EM field is
dramatically enhanced both outside the loaded slot and inside
the gap between the strip insert and slot. The latter effect leads
to the quasi-static capacitive coupling between the strip insert
and slot aperture.
(b)
Fig.3. (a) Microwave near field probe based on the horn antenna terminated
with metal plate with resonantly loaded aperture; (b) imaging scene and probe
set-up geometry.
Fig.4.Measured EM reflection and transmission of the straight strip loaded
aperture with slot width as a parameter. Strip with is 2mm, strip length is
52mm, slot length is 62.5mm. Green hollow dotted lines show the
transmission and reflection parameters for the case of unloaded aperture.
The measured resonance transmission properties of the striploaded aperture geometry are shown in Fig. 4, where the slot
width varies as a parameter. The transmitted field is measured
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
by the wideband standard horn (positioned 300mm away from
the probe) with a 140mm x 240mm aperture. It can be seen
that for narrow slots the resonance frequency increases due to
abovementioned equivalent capacitance reduction. It is also
interesting to note that wider slots ensure higher level of
transmission and increased frequency bandwidth of operation.
It is expected that wider slots should result, however, to poorer
E-field collimation in the near field zone. To verify this, the
near electric field transversal distribution FWHM has been
calculated (at y = 0.2λ, i.e. close to the strip ends) as a
function of slot width and distance from the loaded aperture
using FEKO solver, Fig.5.
4
using a canonical model of two parallel y-oriented Hertzian
dipoles located in the imaging plane and separated by the
subwavelength distance d across the range x.
(a)
Fig.5.Calculated FWHM for the straight strip loaded aperture vs. distance
with slot width as a parameter. Wavelength λ=0.125m (@2.4GHz). Other
geometry parameters are the same as in Fig.4. Parameter wsl stands for the slot
width.
From Fig.5 it can be concluded that for very narrow slots
(~0.05λ wide) the E-field is tightly collimated in the vicinity
of the strip, -z ranges from 0.05 to 0.10λ, and then rapidly
diverges as the imaging distance increases (-z>0.15λ). For
wide slots (~0.2λ) the divergence of the E-field at larger
imaging distances is slower, however, as expected in the
vicinity of the aperture (0.05÷0.10λ) the FWHM is ~0.05λ
wider than for the narrower slot. Apparently there is some
optimal slot width, viz. 10mm (~0.1λ) that can be chosen in
microwave probe design. It also can be noted that based on the
FWHM properties the probes with different slot widths should
ensure very similar imaging resolution (with ~λ/20 difference)
in the imaging distances range 0.05÷0.15λ. Fig.5 also gives a
characteristic resolution of the loaded aperture probe based on
the FWHM as a function of imaging distance. Another
important aspect affecting the ability of the probe to resolve
multiple targets is discussed below.
C. Characteristic Transversal Resolution of Two Electric
Hertzian Dipoles
Imaging multiple targets in the cross range, Fig.3(b) is
associated with the near field cross-coupling between these
targets which affects the microwave probe imaging resolution.
Since EM coupling is caused by the induced currents on the
targets, the characteristic resolution limits can be estimated
(b)
Fig.6. (a) Calculated normalized near field distribution of two Hertzian
dipoles in free space separated by quarter wavelength at various distances z;
(b) resolution contrast as a function of dipoles cross-range separation and
imaging distance.
Fig.6(a) shows the near field distribution of two y-oriented
Hertzian point-like dipoles separated by λ/4 in cross range at
various stand-off distances z. It can be seen that the resolution
between the dipole sources decreases very rapidly with the
distance – it is noteworthy that two λ/4 separated dipoles
cannot be resolved at 0.15λ stand-off distance. Next, Fig. 6(b)
shows the resolution contrast of two Hertzian dipoles,
measured as 20log of the ratio of a peak-to-dip field
magnitude for a range of dipoles separation and stand-off
distances. It can be seen that the resolution contrast for the
considered dipole sources decreases exponentially with the
stand-off distance. The results provided in Figs. 5, 6 can be
used to estimate the characteristic resolution of the considered
near field probe.
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
IV. LOADED APERTURE NEAR FIELD PROBE PERFORMANCE
ENHANCEMENT
For subwavelength resolution imaging in two dimensions
(in the x-y plane) it is essential to reduce the dimensions of the
insert, especially its length. The major drawback of the
straight strip insert loaded aperture considered in the previous
Sections is its wide near E-field pattern in the longitudinal y
range [22] which allows subwavelength imaging only in one
direction. Additionally, in many applications dual polarization
probes are required. Insert configurations fulfilling these
requirements will now be discussed.
A. Insertion of Folded Strip in Rectangular Slot
The geometry of the folded strip insert (whose total unfolded
length is ~ λ/2) is shown in Fig.7(a) and its measured
reflection S11 parameter in Fig.7(b).
5
screen aperture should lead to better imaging contrast as
compared to a single screen configuration(~10dB difference in
the S11 levels for these two structures).
Fig.7(b) shows the measured S11 parameter and Fig.8
displays calculated (with FEKO) near field spatial distribution
for the folded strip loaded aperture with geometric parameters:
total unfolded length of the strip is λ/2; top horizontal segment
is λ/4; vertical segments are λ/20 each; bottom horizontal
segments are λ/13.3 each; strip and slot widths are λ/30 and
λ/10 correspondingly. The dual layers are separated by 3mm,
the wavelength λ=13.15cm corresponds to the frequency
2.28GHz.
The simulated field intensity distribution in the vicinity of
the dual-screen loaded aperture at z = -0.1λ demonstrates that
the folded strip insertion results in a dominant dipole mode in
the near electric field, Fig. 8(a), with a subwavelength average
FWHM approximately equal to the length of folded strip in
longitudinal direction (0.25λ in this case). The FWHM in the
transversal directions depends on the strip and slot widths and
intrinsically is very narrow (Table I).
(a)
(a)
(b)
Fig.7. (a) Slot aperture loaded with folded strip. The aperture in dual slotted
screen is shown. Small red arrows show the current density flow path (FEKO
simulated) at the resonance frequency. (b) S11 parameters of the folded strip
loaded aperture. Vertical separation between the screens dscr = 3mm.
A single or dual screen configuration as shown in Fig. 7(b)
can be used. Experimental results and numerical simulations
not discussed here in detail for the brevity’s sake show that the
dual slotted screen loaded aperture leads to better near field
confinement as compared to the single screen aperture. Also,
enhanced transmission, Fig.7(b), through the slotted dual
(b)
Fig. 8. (a)Simulated electric field intensity distribution in the near field
vicinity of the probe at z = -0.1λ; (b) Simulated magnetic field intensity
distribution at the plane z = -0.1λ.
It should also be noted that the magnetic field intensity (with
dominant Hx component) forms a tightly localized circular
spot with FWHM ~ 0.1λ or less at offset distance z = -0.1λ,
Fig. 8(b).
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
6
B. Folded Cross Dipole in Circular Subwavelength Aperture
The folded strip insertion considered in the previous section
offers considerable near electric and magnetic field
collimation as compared to the straight metallic insert.
However, again, it can operate only for single EM field
polarization. A compacted cross dipole insertion in a circular
subwavelength aperture, Fig. 9(a) is now proposed as planar
probe geometry with dual-polarization capability.
(a)
(a)
(b)
Fig. 10. (a) Simulated electric field intensity distribution in the near field
vicinity of the folded cross loaded aperture at z = -0.1λ; (b) simulated
magnetic field intensity distribution at the plane z = -0.1λ.
A comparison between the near field patterns of folded strip
and compacted cross inserts loaded apertures is summarized in
Table I.
TABLE I
FWHM (in λ) OF THE LOADED APERTURE NEAR FIELD
(b)
Fig.9. (a) Geometry of the compacted dipole cross insertion in circular hole
and (b) its measured S11 parameter. Red arrows in Fig.9(a) show the current
density flow path at the resonance frequency.
Here resonant transmission occurs due to the electric currents
along the crosses dipole arms whose unfolded length is
approximately λ/2, λ is the wavelength corresponding to
resonance. Fig. 10 shows the simulated electric and magnetic
field intensity distribution in the xy plane at the offset distance
z = -0.1λ. The geometric parameters of the structure are:
circular aperture diameter Dh is λ/5; strip width is λ/50; gap g
is λ/35 and the incident field is polarized along the y axis.
It can be seen that the normally incident Ey polarized plane
wave induces a dominant dipole mode in the electric response
of the compacted cross inclusion; both the electric and
magnetic components of the near field intensity distribution
are tightly collimated with FWHM data detailed in Table I.
0.05
Folded strip,
|E|
0.11
Folded strip,
|H|
0.14
Cross insert,
|E|
0.23
Cross insert,
|H|
0.28
0.1
0.18
0.19
0.26
0.37
0.15
0.23
0.26
0.34
0.46
0.2
0.30
0.4
0.46
0.54
0.25
0.42
0.53
0.58
0.70
-z/λ
In this table the FWHM data are provided for the slotted dual
screen and compacted cross loaded apertures discussed above.
It can be seen from Table I that the folded strip loaded
aperture enables much tighter (~0.1λ smaller FWHM) near
field collimation than the compacted cross geometry. Also the
near field pattern of the circular compacted cross loaded
aperture diverges fast at larger stand-off distances
(FWHM>0.5λ at |z|>0.25λ) which makes sub-diffraction
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
7
resolution impossible for more than quarter-wavelength
imaging distance.
V. NEAR FIELD SUBWAVELENGTH RESOLUTION IMAGING OF
METALLIC AND DIELECTRIC OBJECTS
In this section we study the imaging properties of dual screen
aperture arrangements of the folded strip and compacted cross
resonant insert probe. Characteristic targets are represented by
resonant and non-resonant metallic 1D and 2D shapes in free
space and in moderately lossy layered media. Also, imaging of
non-resonant dielectric 2D and 3D targets is experimentally
studied in free space and in the presence of a lossy absorbing
screen. In all experiments the targets have subwavelength
dimensions or are separated by the subwavelength distance in
cross-range.
A. Planar Metallic Targets in Free Space
The first target to be imaged consists of a pair of resonant
half-wavelength strips, separated along the x-axis by a quarter
wavelength (at 2.28GHz), Fig. 11(a). The imaging distance
is 0.1λ for all experiments. Fig. 11 shows the measured S11
amplitude and phase obtained using the folded strip insert
probe. It can be seen that the probe reconstructs the EM
properties (near field distribution) of the strips with very high
contrast (~10dB in amplitude and ~200o in phase).
(c)
Fig. 11. Microwave image of two sub-wavelength spaced resonant strips on a
20mm thick layer of foam εr=1.1 in free space. (a)Imaged target; (b) measured
S11 magnitude; (c) S11 phase. The results are obtained with folded strip insert
probe. The scale in Fig.11(a) is in millimeters.
(a)
(a)
(b)
Fig. 12. Microwave image of two resonant strips in free space. (a) measured
S11 magnitude; (c) S11 phase. The results are obtained with folded cross insert
probe.
(b)
Fig.12 demonstrates the microwave image of the same twostrip structure obtained with a probe using folded cross insert
in circular aperture described in Section IV B. From Figs.11,
12 it can be seen that the coupling mechanism of these two
probes is substantially different; the loaded circular hole probe
tends to average the EM coupling effect across its aperture.
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
Even in this case the obtained image has a very high amplitude
and phase contrast, however the shape of the wire scatterers
represented not as clearly as in Fig.11.
It is important to note at this stage that in this paper we
show “raw” imaging data with probe coupling effect
embedded into images. The images could be potentially
improved using numerical post-processing de-embedding the
probe coupling effect, this will be reported in future work.
(a)
(b)
8
obtained with folded strip insert in rectangular slot probe. The scale in
Fig.13(a) is in millimeters.
Figs.13, 14 demonstrate the near field microwave image of
two non-resonant planar metallic shapes in free space as
obtained using the folded strip and compacted cross loaded
aperture probes correspondingly.
(a)
(b)
Fig. 14. Microwave image of two planar metal shapes in free space. (a)
Measured S11 magnitude; (c) S11 phase. The results are obtained with
compacted cross insert probe.
(c)
Fig. 13. Microwave image of two different metal shapes in free space. (a)
Imaged target; (b) measured S11 magnitude; (c) S11 phase. The results are
The shapes are separated by ~λ/3 in the cross-range and
have ~λ/4 characteristic size. It can be seen that the folded
strip loaded slot probe recovers (3dB amplitude and ~50
degrees phase contrast) the EM near field distribution of these
targets. The compacted crossed probe results in much higher
amplitude contrast (~6dB) image. The features of these shapes
are less obvious due to the abovementioned averaging of the
coupling effects by the loaded circular aperture, however in
both cases – folded strip probe, Fig.13 and compacted cross
loaded aperture probe, the recovered difference in the near
field properties between the shapes is quite significant (up to
5dB in amplitude and about 30 degrees in phase) even though
the characteristic dimensions of the shapes differ by less than
0.1λ.
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
B. Planar Metallic Targets in Lossy Medium
In the next experiment the planar metal shapes depicted in Fig.
13(a) were positioned behind a 2mm thick sheet of microwave
absorbing magnetodielectric material Eccosorb FGM-U40
[23] which has ~10dB/cm attenuation at around 2.3GHz.
9
This can be explained by the fact that the folded strip in
rectangular slot has its dominant horizontal magnetic dipole
component parallel to the sample surface. The compacted
cross in circular aperture has dominant vertical magnetic
dipole component, perpendicular to the sample surface. The
electric field (with dominant horizontal Ey component) of the
vertical magnetic dipole decays very much faster than the
electric field of the horizontal magnetic dipole in the near field
zone, Fig. 16. This electric field behavior leads to the strong
EM coupling of the compacted cross probe to the sheet of
absorbing material and inability to resolve targets behind the
lossy material.
C. Dielectric Target Imaging
In this section microwave imaging of dielectric targets with
moderate value of permittivity is studied.
Fig. 15. Microwave image of two different metal shapes behind the 2mm layer
of FGM-U-40 material [23]. Microwave probe based on the folded probe
insert in rectangular slot.
Fig. 15 demonstrates the amplitude near field image of the
structure obtained with the folded strip insert probe. It can be
seen that the absorbing material in front of the shapes leads to
much higher contrast imaging (~6dB) than when the shapes
were positioned in free space. This is believed to be
principally due to two reasons. Firstly, the absorbing material
increases the resonance frequency of the shapes which are
non-resonant in free space because it behaves as a superstrate
with higher than unity relative permittivity and permeability.
Secondly, the absorber leads to higher attenuation of the
Fresnel’s components of the near field (responsible for image
smearing) due to the better collimation of the near field by the
loaded aperture in the presence of lossy layer. It was noticed
that the probe based on the folded cross in circular aperture
cannot resolve these targets at z=-0.1λ or even closer distance
z=-0.05λ.
(a)
(b)
Fig. 17. Microwave image of two identical rubber balls, permittivity ~3, of
λ/4 diameter whose centers separated by λ/3 in the x range. (a) Folded strip
loaded aperture; (b) folded cross loaded aperture probe.
Fig. 16. Electric field of the horizontal and vertical magnetic dipoles of
1µAm2 dipole moment amplitude as a function of vertical offset distance in
the near field.
Fig. 17 demonstrates the amplitude near field image of two
solid rubber balls (permittivity ~3) of 25mm diameter (λ/4 at
2.28GHz) immersed in foam (permittivity is ~1.1). The
centers of the balls are separated by 44mm (λ/3) in the x range.
Fig. 17(a) shows the near field image obtained with folded
strip insert probe, while Fig.17(b) demonstrates the image
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
collected by compacted cross loaded aperture. It can be seen
that the use of the folded strip loaded aperture results in a clear
image with ~3dB maximum contrast. Near field coupling
leads to a dipole-like response of the dielectric targets. The
compacted cross loaded aperture results in a slightly better
image contrast, ~5dB. Asymmetry observed in the image
between left and right targets is due to 2 mm difference in the
offset positions z of the balls (the right ball is 2mm farther
away from the probe than the left one).
(a)
(b)
(c)
Fig. 18. Near field amplitude image of a dielectric tile (εr~10) on a lossy 1mm
sheet of absorber Eccosorb FGM-U40, (a). Near field image obtained with
folded strip (b) and compacted cross (c) loaded aperture
In the next experiment a 20mm x 20mm dielectric tile (with
εr ~ 10 and thickness 2mm) was positioned on 1mm thick
10
Eccosorb FGM-U40 layer, Fig. 18(a). This scenario is relevant
to a number of applications, e.g. microwave skin cancer
detection, non-destructive surface probing and quality control,
etc. It can be seen that both probes, based on folded strip and
compacted cross loaded apertures, collect a reasonably highcontrast (~3dB) image of a tile. S11 phase images demonstrate
very high contrast in phase (~35-40 degrees).
Fig. 19. Near field amplitude image of a water-filled plastic ball behind a
lossy sheet of absorbing material. The imaging offset distance is ~ λ/4.
Finally, Fig. 19 demonstrates a near field amplitude image of a
plastic ball of 25mm diameter (λ/4) wall thickness 1mm filled
with water (εr ~ 80) and obtained with folded strip loaded
aperture probe. The ball is positioned behind a 1mm layer of
Eccosorb FGM-U40 material at 30mm offset distance (~λ/4).
It can be seen that very high contrast image of a
subwavelength dielectric target is possible at larger imaging
distances when high permittivity dielectric targets are being
imaged. The compacted cross probe has not been able to
resolve this target due to mentioned behavior of the electric
field component of the vertical magnetic dipole, Fig.16.
From the experiments in Section V it is possible to conclude
that the folded insert in rectangular slot and compacted cross
in circular aperture probes behave in similar fashion when
non-resonant metallic objects are being imaged. In such cases
the increased level of reflection amplitude S11 corresponds to
the presence of metal surface, cf. Figs. 11(a)-12(a) and 13(a)14(a).
In the case of dielectric objects in free space, the coupling
mechanisms of these two probes are opposite. In general, the
folded strip couples to the inhomogeneity in such a way that
the EM transmission is enhanced by the presence of dielectric
object, Figs.17(a), 18(b) while the compacted cross probe
detects the dielectric object through the increased level of
reflection as compared to the background, Figs.17(b)-18(c).
Therefore using data obtained simultaneously by both probe
types should result in better object characterization.
It should be noted that in many practical imaging scenarios,
especially involving dielectric media, frequency tuning of the
probe resonance is required [15]. In the microwave range this
can be accomplished using above considered inserts loaded
with electrically or mechanically tunable varactors [15].
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
VI. CONCLUSIONS
It has been shown that the EM transmission through the
subwavelength or non-resonant apertures in conductive
screens can be dramatically enhanced by loading them with
folded metallic strips exhibiting resonant properties. We
propose planar and quasi-planar resonant insertion geometries
that enable two-dimensional dual-polarization subwavelength
field confinement along with the ability to focus both electric
and magnetic fields. When illuminated by an EM beam these
loaded apertures enable very tight, subwavelength, collimation
of the EM power in the near field zone. This effect forms a
basis for a class of microwave near field imaging probes with
subwavelength resolution operating in a wide range of
imaging distances (0.05-0.25λ). Measurement results
demonstrate a possibility of high-contrast (more than 3dB in
amplitude and 40 degrees in phase) near field subwavelength
imaging of two-and three-dimensional resonant and nonresonant metallic and dielectric targets in free space and
moderately lossy layered media. The results of the paper
should be useful for applications of microwave imaging in
medical applications, non-destructive testing and non-invasive
surface probing.
ACKNOWLEDGMENT
The Authors are most thankful to the Reviewers of this
paper whose valuable and insightful comments allowed to
improve the original manuscript.
REFERENCES
[1] L. Markley, G.V. Eleftheriades, “Two-dimensional subwavelengthfocused imaging using a near-field probe at a λ/4 working distance”,
Journ. Appl. Physics, vol. 107, no.9, pp. 093102 - 093102-5, 2010.
[2] Y. Kawano, “Highly sensitive detector for on-chip near-field THz
imaging”, IEEE Journ. Sel. Topics Quantum Electronics, vol.17, no 1,
pp. 67-78, 2011.
[3] X. Zhuge, A.G. Yarovoy, “Sparse multiple-input multiple-output arrays
for high-resolution near-field ultra-wideband imaging”, IET Microw.
Antennas Propag., vol. 5, no. 13, pp. 1552–1562, 2011.
[4] J. Szelc, H. Rutt, “Near-field THz imaging and spectroscopy using a
multiple subwavelength aperture modulator”, IEEE Trans. Terahertz
Science Technology, vol.3, no.2, pp. 165-171, 2013.
[5] S. Dieter, W. Menzel, “High-resolution probes for near-field
measurements of reflectarray antennas”, IEEE Trans Antennas Propagat.
Letters, vol. 8, pp. 157-160, 2009.
[6] O. Mitrofanov, et al, “Terahertz near-field microscopy based on a
collection mode detector”, Appl. Phys. Lett. vol. 77, pp. 3496-3499,
2000.
[7] H. A. Bethe, “Theory of diffraction by small holes”, Phys. Rev. vol. 66,
pp. 163-182, 1944.
[8] G. C. Cho, H-T Chen, S. Kraatz, et al., “Apertureless terahertz near-field
microscopy”, Semicond. Sci. Technol., vol. 20, pp. S286–S292, 2005.
[9] A. Shitvov, D. Zelenchuk, A. Schuchinsky, V. Fusco, “Passive
intermodulation generation on printed lines: near-field probing and
observations, IEEE Trans. Microwave Theory Techniques, vol.56, no.12,
pp.3121-3128, 2008.
[10] M. Misra, S. R. Andrews, S. A. Maier, “Waveguide artefacts in terahertz
near field imaging”, Appl. Phys. Lett., vol. 100, pp. 191109-191112, 2012
[11] R. Harrington, “Resonant behavior of a small aperture backed by a
conducting body”, IEEE Trans. Antennas Propagat., vol. AP-30, no. 2,
pp. 205-212, March 1982.
[12] A. Azad, J. O’Hara, R. Singh, H.-T. Chen and A. Taylor, “A Review of
terahertz plasmonics in subwavelength holes on conducting films”, IEEE
Journ. Selected Topics Quant. Electron., vol. 19, no.1, 8400416, Jan/Feb.
2013.
11
[13] D. Ates, A.Cakmak, E.Ozbay, “Near-field light localization using
subwavelength apertures incorporated with metamaterials”, Optics
Communications, 285, pp. 3390-3396, 2012.
[14] A. Cakmak et.al., “Enhanced transmission through a subwavelength
aperture using Metamaterials”, Appl. Phys. Lett. 95, 052103, 2009.
[15] T. Pochiraju, O. Malyuskin, V. Fusco, “Tunable near-field subwavelength
microwave imaging”, Microwave Opt. Technology Lett. vol. 53, no.6, pp.
1229-1231, 2011.
[16] C. Harrison, “Receiving characteristics of impedance loaded slot
configurations”, IEEE Trans. Electromagn. Compat., vol. EMC-15, no.3,
pp. 131-136, 1973.
[17] C. Butler, Ya. Rahmat-Samii, R. Mittra, “Electromagnetic penetration
through apertures in conducting surfaces”, IEEE Transactions Antennas
Propagat., vol. AP-26, no.1, pp. 82-93, 1978.
[18] C. M. Butler, K. R. Umashankar, “Electromagnetic excitation of a wire
through an aperture-perforated, conducting screen,” IEEE Trans.
Antennas Propagat., vol. AP-24, pp. 456-462, 1976.
[19] L. Yeung, Ke-Li Wu, “Generalized partial element equivalent circuit
(PEEC) modeling with radiation effect”, IEEE Trans. Microwave Theory
Techniques, vol. 59, no.10, pp. 2377-2384, Oct. 2011.
[20] R.E. Collin, Foundations for microwave engineering, Wiley 2007.
[21] K.J. Binns and P.J. Lawrenson, Analysis and Computation of Electric and
Magnetic Field Problems, Pergamon Press, 1973.
[22] T. Pochiraju, O. Malyuskin, V. Fusco, “Sub-wavelength near field
imaging using a cross polarised slot with wire insert”, 2010 URSI Int.
Symp. Electromagnetic Theory (EMTS), pp. 559 – 561, 2010.
[23] http://www.eccosorb.eu/products/eccosorb/eccosorb-fgm-u-40
Oleksandr Malyuskin (M’04) received the
MSc degree in Radiophysics and
Electronics and the PhD degree in
Radiophysics from Kharkiv National
University, Ukraine, in 1997 and 2001
respectively. He joined the Institute of
Electronics,
Communications
and
Information
Technology,
Queens
University Belfast, in March 2004 as a Post
Doctoral Research Fellow involved in the
development of novel composite materials
for advanced EM applications. His research
interests include analytic and numerical methods in electromagnetic wave
theory, characterization and application of complex and nonlinear materials,
antenna arrays and time reversal techniques. Dr Malyuskin has published over
50 scientific papers in major journals and in refereed international
conferences, and has acted as a reviewer for the IEEE publications.
Vincent Fusco (S’82–M82–SM’96–
F’04) received the Bachelors degree
(1st class honors) in electrical and
electronic engineering, the Ph.D.
degree in microwave electronics, and
the D.Sc. degree, for his work on
advanced front end architectures with
enhanced functionality, from The
Queens University of Belfast (QUB),
Belfast, Northern Ireland, in 1979,
1982, and 2000, respectively.
His
research
interests
include
nonlinear microwave circuit design, and active and passive antenna
techniques. He is the Director of the High Frequency Laboratories, Queens
University of Belfast, and is also Director of the International Centre for
Research for System on Chip and Advanced MicroWireless Integration,
SoCaM. He has published over 420 scientific papers in major journals and
international conferences, and is the author of two textbooks. He holds several
patents on active and retrodirective antennas and has contributed invited
chapters to books in the fields of active antenna design and EM field
computation.
Prof. Fusco is a Fellow of the Royal Academy of Engineering and a Member
of the Royal Irish Academy. In 1986, he was awarded a British
Telecommunications Fellowship and 1997 he was awarded the NI
Engineering Federation Trophy for outstanding industrially relevant research.