Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No.

3, e2024280724 Aug 2024

DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 1
Article

Methods of Measurements of Electromagnetic

Characterization of Free Space Materials in
the X-Band
Rafaela Real Passos1,2 , Cecília Maia Corsato1 , Bruno Ferraz Donati1 ,
Nicholas Eras Fonseca1 , Valdirene Aparecida da Silva1,3
1
Research and Development Institute, University of Vale do Paraiba, São José dos Campos, São Paulo, Brazil.
[email protected], [email protected], [email protected],
[email protected]
2
Department of Aerospace Science and Technology, Institute for Development and Industrial Coordination, IFI,
DCTA, São José dos Campos, São Paulo, Brazil.
3
Institute of Technological Research, IPT/LEL, São José dos Campos, São Paulo, Brazil.
[email protected]

Abstract— The technique of electromagnetic characterization of

materials in free space allows for the measurement of constitutive
properties without making contact with the material being
measured. Characterization methods in Free Space are naturally
suited for Electromagnetic Radiation Absorbing Materials (RAM)
but are also important in applications where the material sample to
be measured does not undergo major interventions. They consist of
evaluating a received signal resulting from a free-space transmission
through two horn antennas with beam collimating lenses and a
Vector Network Analyzer (VNA). The characteristics of the
transmitted signal are modified by a material positioned between the
transmitting and receiving antennas. This work presents methods for
electromagnetic characterization in Free Space, in the frequency
range of 8.2 to 12.4 GHz (X-Band), in a semi-anechoic chamber,
seeking the development of measurement processes that present
results with repeatability and reliability. The measurements were
made with standard and electromagnetic radiation absorbing
materials. The measurement technique presented 95% of efficiency.

Index Terms— Electromagnetic Characterization in Free Space, RAM, S-

Parameters, X-Band.

I. INTRODUCTION

Measurements of the dielectric properties of materials can provide valuable information about their
electromagnetic behavior and thus their applications. These measures allow the proper application of
the material or to monitor a manufacturing process for quality improvement. The interaction of the
electromagnetic wave with matter results from the electrical and/or magnetic properties of the material
and the characteristics of the incident wave, such as the frequency and intensity range in which they are
submitted [1], [2], [3], [4].
In recent years, the free space characterization technique has gained increasing popularity for

Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 2

measuring electromagnetic properties of materials [2]. The Measurement System in Free Space is non-
destructive and non-contact; therefore, it is especially suitable for measuring permittivity and
permeability [3]-[7].
Accurate assessment of a material's complex permittivity is an essential requirement in many
applications. Among many established techniques for electromagnetic characterization of materials, the
free space technique provides the most realistic non-destructive measurements suitable for applications
in the aerospace and automotive industries [3], [4], [5].
The technique of electromagnetic characterization of materials in free space, or insertion loss between
antennas, evaluates in a non-destructive way and, in a wide band of frequencies and temperature, the
properties that constitute the material under test [1]-[6]. Such properties determine the response of the
material when exposed to electromagnetic fields, as they are extracted by algorithms through
measurements. This technique allows the characterization of flat material samples when positioned
between two sources, transmitter and receiver. However, the evaluation of material characteristics is
made with the results of measurements performed at a single point [3]-[6]-[7].
The method called NRL arc is a free space measurement system for testing wideband microwave
absorbing materials. It is a well-established method in the literature where it allows the characterization
of radar absorber materials (RAM) with angular dependence, that is, it allows the electromagnetic
characterization of the electromagnetic wave absorption as a function of the angle of incidence and
reception, called a bistatic reflection method. In this configuration, two antennas are used to transmit
and receive signals respectively, and the reflectivity of the electromagnetic wave can be measured at
different incident angles. Using this configuration, material properties in different directions can be
characterized. It should be noted that, in bistatic reflection measurements, the reflection is dependent
on the polarization of the incident wave. Incident waves with parallel and perpendicular polarization
generally result in different reflection coefficients. In addition, a special calibration is required for free
space bistatic reflection measurements [6], [7], [8].
Measurements performed using NRL arc (Naval Research Laboratory) and RCS (Radar Cross
Section) methods show discrepancies that can be explained by the influence of electromagnetic wave
diffraction on the sample edges. The literature [7], [8], [9] shows that different methods decrease the
discrepancy and that, observed through high frequency electromagnetic wave scattering calculations,
these diffraction phenomena depend on the distance to the sample, even in the scattering area, that is,
at high frequency, different diffraction phenomena are influenced by the sample distance.
These experimental measurements aim to explore the concept of electromagnetic characterization of
materials with oblique incidence and are carried out with focus conical horn antennas. This work was
carried out using a setup originally applied for measurements of permittivity and permeability in an
angular positioning system in the form of a compass. In this system, the antennas were positioned in a
far-field, allowing for angles variation from their normal position (around 5° between higher frequency
antennas) and 45°, as the upper limiting angle [6]. To verify the system, measurements were taken of
Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 3

materials that absorb electromagnetic radiation with oblique incidences of 30°, 45° and 60°.
Electromagnetic radiation absorbing materials prepared with ferrocarbonyl at different concentrations
were measured to evaluate the system with reliable data.
As a justification, it is shown that using this setup for studies of oblique measurements and radar
straight section, both of plane surfaces and other geometries, may present strong indicators of
application. It is noteworthy that measurements in this type of setup are not performed in Brazil.
Thus, with a setup positioned in a far-field and that allows its variation in angles up to 45º as a
compass, this work seeks to develop a methodology for measuring the characterization of materials that
are safe in free space, with repeatability and reliability procedures, and that can meet the various areas
that use the electromagnetic characterization of materials. Fig. 1 show Schematic setup in a far-field
that allows variation in angles up to 45º as a compass, described in [9].

Fig. 1. Schematic setup in a far-field that allows variation in angles up to 45º as a compass [9].

I. METHODOLOGY
A. Materials
Reflective material: aluminum plate in the dimensions of 25 cm x 25 cm and 3 mm thick; Standard
Material: 25 cm x 25 cm Teflon® plates with 1, 3, 5, 7 and 12 mm thick; RAM: absorbent mats of 25
cm x 25 cm 5 mm thick, added with commercial ferrocarbonyl from the Basf Ltda., in concentrations
from 20 to 70%, ranging in 5%, shown in Table I. The mats were made in an elastomeric rubber matrix
of Butadiene Styrene (SBR), and provided by KBF CHEM Indústria Comércio e Representação Ltda.

Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 4

TABLE I. FERROCARBONYL CONCENTRATION IN THE STUDIED RAM SAMPLES.

Sample Concentration
(% in large scale)
A 20
B 25
C 30
D 35
E 40
F 45
G 50
H 55
I 60
J 65
L 70

B. Calibration
For the characterization of free space materials, measurement methods often require complicated
calibration procedures due to the unavailability of accurate information from the standards. This method
involves systematic errors such as return loss at the antenna input and wave propagation up to the
reference plane, defined on the surface of the samples.
The reflectivity measurement first consists on performing a calibration of the measurement system
(VNA, cables and connectors), then measuring the reflection reference with the reflector plate, that is,
the value that will be considered as zero absorption in the system.
Fig. 2 (a) and Fig. 2 (b) show a measurement system of the Electromagnetic Measurements
Laboratory, IFI/DCTA, for electromagnetic characterization in NRL arc. Two different calibrations are
required [9]. The first process, as shown in Fig. 2 (a), begins by performing the calibration of the
measurement system (VNA, cables and connectors) up to the entrance of the antennas, using an
instrument called ECAL (Electronic Calibration Module). Then, with the measurement system already
connected to the antennas, the windowing interval is defined, positioning the antennas equidistant from
the sample holder (Fig. 2(b)), with a metal plate of known thickness, preferably close to the thickness
of the samples to be measured. Afterwards, the reflection curves in the time domain of parameters S 11
and S22 are superimposed. Generally, the second peak determines the windowing interval that lies
between the time intervals, t1 and t2.

(a) (b)
Fig. 2. Measurement system of the Electromagnetic Measurements Laboratory, IFI/DCTA, for electromagnetic
characterization in NRL arc. (a) angular positioning of the RX horn antenna and sampler (b).

Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 5

C. Measurement Uncertainty
The results were obtained through the relative values between measurements. In order to simplify the
uncertainty calculation, the dispersion analysis of the test values of the measurement system was
performed during the equivalent period in the measurements. This procedure allows to obtain an
approximation of the stability levels in the results, due to the influence of all system components on the
measurement result.
For obtaining an estimate of the uncertainty of ε' r due to the uncertainties of the measurements, a
differential analysis was used as shown by the equation (1):

2 2 2
𝜕ℇ′ 𝜕ℇ′ 𝜕ℇ′
𝑢𝜀2𝑟′ = 𝑢𝑒𝑠𝑡
2
+ (𝜕𝐺 𝑟 𝑢𝐺𝑟𝑥 ) + (𝜕𝐺 𝑟 𝑢𝐺𝑡𝑥 ) + (𝜕𝑉𝑁𝐴
𝑟
𝑢𝑣𝑛𝑎 ) (1)
𝑟𝑥 𝑡𝑥

Where:
2
𝑢𝑒𝑠𝑡 = Equipment stability, 𝑢𝐺𝑟𝑥 = Receiving antenna uncertainty, 𝑢𝐺𝑡𝑥 = Transmitting antena,
uncertainty, 𝑢𝑣𝑛𝑎 = VNA calibration uncertainty.

D. Electromagnetic Characterization
a. S-Parameters
Two resulting waves can be created during the interaction between an EM wave and a material: the
transmitted and the reflected. The transmitted wave travels through the material towards its second face.
During its journey, some of the energy can be absorbed and transformed into heat so that its amplitude
decreases exponentially. When the wave hits the opposite surface, part can be transmitted and some
reflected. If the shield thickness is greater than the skin depth, the wave reflected in the inner wall is
absorbed and therefore multiple reflections can be ignored. Both the reflected (R) and transmitted (T)
powers are measured by the reflection coefficient (S 11) and the reverse transmission coefficient (S 21),
respectively,
The measurements of the S parameters that allow obtaining the values of electrical permittivity and
magnetic permeability of the materials were performed using the Transmission and Reflection Method
and Nicolson Ross Weir (NRW) method.
The method developed by Nicolson, for the electric permittivity calculation, is based on the reflection
and transmission occurring at the boundaries of the free space and the material under analysis. Its
theoretical derivation is well explained in references and is briefly presented here [10]. Defining V1 =
S21 + S11 and V2 = S21 - S11, the reflection coefficient is given by (2):
1−𝑉 𝑉 1− 𝑉 𝑉 2
Γ1 = ( 𝑉 −1𝑉 2 ) ± √( 𝑉 −1𝑉 2 ) −1 (2)
1 2 1 2

The correct root is chosen by requiring |Г1|≤ 1 and the transmission coefficient Z1 for the NRW is
written as (3):
𝑆11 + 𝑆21 −Γ1
𝑍1 = (3)
1−(𝑆11 + 𝑆21) Γ1
Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 6

Considering the length of the sample as, it is defined: H (4)

1 1 2
= − [2 𝜋𝐻 𝑙𝑛(𝑍1 )] (4)
Λ2

Calling μr the relative permeability, λ0 the free space wavelength and λc the cutoff wavelength, then
the relative electric permittivity εr is given by (5):
𝜆20 1 1
𝜀𝑟 = [ + Λ2 ] (5)
𝜇𝑟 𝜆2𝑐

b. Setup assembly
The measurements were performed in the semi-anechoic chamber of IFI/DCTA. The measurement
technique in free space adopted was the VSWR (Voltage Standing Wave Ratio), which analyzes
amplitude variations in the frequency range of the X-band, 8.2 to 12.4 GHz.
The set up used, shown in Fig. 3, is composed by a RAM lined anechoic chamber (1), microwave
signal generator (2) that feeds, via low loss cable (3), the antenna input terminals, establishing the
incident energy transmission to the receiving antenna (4). On signal reception, a spectrum analyzer (5)
is used that collects the signals coming from the receiving antenna.
The system used for the displacement of antennas, shown in Fig. 3, has two approximation engines
(with a horizontal useful stroke of 2,980 m) and two elevation engines (with a vertical useful stroke of
1.12 m) of the transmitting and receiving antenna towers, a motor for polarization changes of the
transmitting antenna and an azimuth rotation engine of the receiving antenna.
The angular orientation system of the anechoic chamber uses the indication system through
synchronous, which provide voltage levels for each direction of the receiving antenna.

Fig. 3. Measurement setup: equipment and accessories used in the free space material characterization measurement
system in IFI/DCTA.

III. RESULTS AND DISCUSSION

A. Calibration Measures
In order to confirm a correct calibration, it is necessary to validate the measurement using a material
with known constitutive properties for characterization. For this purpose, air and a Teflon® plate were
chosen as standards [10]-[11]-[12].
The results of measurements with an empty sample holder are the standards for calibration
knowledge. The theoretical values of the real and imaginary parts of the air permittivity are,
Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 7

respectively, ε'=1 and ε''= 0 and for Teflon®, ε'= 2.1 and ε''=0 [7]. Fig. 4(a) and 4(b) show the
measurements of the real and imaginary permittivity curves as a function of the frequency of air and
Teflon® patterns.

(a) (b)
Fig. 4: Measurements of real and imaginary permittivity as a function of the frequency of air (a) and Teflon® (b)
standards.
The values of real permittivity, ε', and imaginary permittivity, ε”, of air have values of 1 and 0,
respectively, and are in accordance with the literature in the entire frequency range 8.2 – 12.4 GHz, as
well as the results presented for Teflon®, ε’= 2,1 e ε’’= 0 [7], [8]. The curves in Fig. 4(b) indicate a
variation in real permittivity of 0.6 in the frequency range from 12.3 to 12.4 GHz, keeping practically
at the same value in the entire frequency range, 2,0. The maximum value in the variation of the values
of the imaginary part of the permissiveness was 0.22. These variations in the measured values results
can be disregarded, as the greatest distortions always occur at the beginning and end of the frequency
range, due to the time domain filtering process occurring in broadband [8].

B. Uncertainty Calculation
Although errors have been reduced by GRL calibration, they still exist and should be considered in
the measurement. To assess the uncertainty of each result, a spreadsheet was created to reach a 95%
confidence interval. The expanded uncertainty was obtained by calculating the sum of the contributions
from all individual sources, using a coverage factor of (k = 2).
Uncertainty extracted from measurements on Teflon® plates, with dimensions of 25 cm x 25 cm and
thicknesses of 1 mm, 3 mm, 5 mm, 7 mm and 12 mm, in the frequency range of 8.2-12.4 GHz, was
named uest. Plate thickness measurements were taken at the center of the sample where most of the
energy is focused. The edges of the samples are subject to warping and may have errors associated with
thickness, being then disregarded.
In order to obtain the standard uncertainty in the permittivity measurements, ten relative permittivity
readings were taken under the same test conditions and environmental control. The errors due to the
gains of the uRx and uTx antennas must be considered, as well as the mismatch error that occurs in the
system represented by the VNA (Vector Network Analyzer), since its uncertainties have significant
values. Table II represents, in a simple way, an uncertainty spreadsheet of the error sources pointed out
Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 8

in the permittivity measurements, based on Equation 1. The Ui2 term represents the uncertainty in each
source.
TABLE II. CALCULATION WORKSHEET FOR SOURCES OF UNCERTAINTY IN PERMITTIVITY MEASURES
Uncertainty Standard Distribution Divider Standard Ci Hey Ui²
component deviation Uncertainty
uestab Stability of --- U-Type A 1.41 --- 1 --- ---
Measures
utx Tx antenna 0.2(dB) Rectangular- 1.73 0.11 0.5 0.055 0.003025
Type A
urx Rx antenna 0.2(dB) Rectangular- 1.73 0.11 0.5 0.055 0.003025
Type A
Uvna VNA 0.2(dBm) U-TypeB 1.41 0.14 0.71 0.01 0.001
ΣUi² --
Uc Combined Normal
Uncertainty

Performed during the calibration step of the antenna input measurement system. As well as
uncertainties related to reflections, antenna position and sample holder position are suppressed due to
the GRL calibration step. Fig. 5 presents the expanded uncertainty calculated as the sum of the
contribution of all individual sources, for K=2.

Amostras
Samples
1mm
3mm
0,0691 5mm
8mm
12mm
0,0690
Expandida
Uncertainty

0,0689
Incerteza

0,0688
Expanded

0,0687

0,0686

0,0685
8 9 10 11 12 13
Frequency (GHz)
Frequência (GHz)

Fig. 5. Expanded Uncertainty of measurements of Teflon plates with thicknesses of 1, 3, 5, 7 and 12 mm in relation to the
frequency range of 8.2-12.4 GHz.

C. Electromagnetic Characterization
a. S-Parameters
When the electromagnetic wave propagates through the material, the wave-matter interaction occurs
at electronic levels and can result in heat in the RAM. The evaluation of this interaction is made possible
by measurements of S-parameters [13], [14], [15], [16]. During this interaction, two waves can be
created, a transmitted and a reflected. The first one travels through the material towards the opposite
face, and, in this path, part of this energy can be absorbed and transformed into heat. When the wave
hits the other face of the material, part of the wave that hit that other face can be reflected or transmitted.
The Reflection (S11) and Transmission (S21) coefficients are obtained.
Parameter S11 represents the energy that is reflected on the first surface of the material when the
Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 9

electromagnetic wave collides with the material. Thus, the closer the measured S 11 value is to 0 dB, the
greater the behavior of the material reflector. On the other hand, the larger the further measured value
is from 0 dB, the less energy is reflected from the first material surface. This means that the lower the
energy that returns to the origin from the material hit by the incident wave [13], [14]. Parameter S21
represents the energy that propagates through the material that is favored by the impedance matching,
energy that crosses the material and is detected on the opposite side of the incidence wave. In this case,
the closer this value is to 0 dB, the more energy is transmitted through the material. Based on the S 21
and S11 values it is possible to assess how much energy is absorbed in RAM [13], [14], [15], [16].
Fig. 6 (a), Fig. 6 (b), Fig 6 (c) and Fig. 6 (d) shows the measurement curves of the reflection
parameters, S11. The curves of parameter S11 of the A-L samples shown in Fig. 6 (a) refer to the 8.2-
10.4 GHz frequency range, in Fig. 6 (b) to the 10.2-12.4 GHz frequency range, in Fig. 6 (c) to the 10.2-
11.4 GHZ band and in Fig. 6 (d) refer to the 11.4 GHz to 12.4 GHz frequency band. The curves of the
S11 parameters present a resonant behavior, with values ranging from -5 dB to -30 dB across the entire
frequency range, with resonance peaks reaching -50 dB. Analyzing the curves, one can see that as the
concentration of ferrocarbonyl in the RAM increases RAM, the values of S11 increase, as can be seen
in curve L, which refers to the RAM with 70% of iron carbonyl. The results show that RAM performs
well as a microwave absorber.

Fig. 6: Measured values of S11 Scattering Parameters in the X band of samples with ferrocarbonyl concentrations ranging
from 20 to 70% by mass. (a) 8.2-9.4 GHz, (b) 9.2-10.4 GHz (c) 10.2-11.4 GHz and (d) 11.4-12.4 GHz.

Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 10

Parameter S21 represents the energy transmitted by the material. The closer to the reference the
measured S21 value is, the greater the transmitted energy, that is, the greater the energy that penetrated
the material, radiated from the emitter and propagated through the material and was transmitted to the
receiver [2]-[12]-[15]-[16].
The graphs in Fig. 7 (a), Fig. 7 (b), Fig. 7 (c) and Fig. 7(d) show the measurement curves of the S 21
parameters. The curves of parameter S 21 of the A-L samples, shown in Fig. 7 (a), refer to the 8.2-10.4
GHz frequency range, in Fig. 7 (b) to the 10.2-12.4 GHz frequency range, in Fig. 7 (c) to the 10.2-11.4
GHz band and in Fig. 7 (d) refer to the 11.4 GHz to 12.4 GHz frequency band. The values of S 21, as
well as S11, increase with increasing concentration of ferrocarbonyl in the RAM. Considering the curve
of sample L, it is noted that the transmission values vary from -7 dB to -12 dB in the studied frequency
range. These values show that the studied RAM has a good transmission of the incident electromagnetic
wave.

Fig. 7: Measured values of S21 Scattering Parameters in the X band of samples with ferrocarbonyl concentrations ranging
from 20 to 70% by mass. (a) 8.2-9.4 GHz, (b) 9.2-10.4 GHz (c) 10.2-11.4 GHz and (d) 11.4-12.4 GHz.

D. Complex parameters of permittivity and permeability

The complex parameters of electrical permittivity and magnetic permeability are closely linked to
the loss attenuation mechanisms of the electromagnetic wave in absorbing materials, allowing
prediction of the behavior of RAM. Real parts are associated with stored electrical and magnetic energy,
Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 11

while imaginary components represent energy losses. For the characterization of RAM, it is possible to
design more efficient and effective materials, in tuned frequency bands, with the reduction of costs and
processing time. These values are dependent on the transmission and reflection coefficients, S 21 and S11,
respectively, obtained in the vector network analyzer. The range of frequencies of interest and the nature
of the material are important conditions in choosing the method for measuring these parameters [2]-[9].
The graphs in Fig. 8 show the values obtained in the measurements of real electrical permittivity, ε',
and imaginary, ε”. It can be seen from the graphs in Fig. 8 (a), Fig. 8 (b), Fig. 8 (c) and Fig. 8 (d) that
the values of real and imaginary permittivity increase with the increase in the concentration of
ferrocarbonyl. As the permissiveness is an extrinsic property of the material, this increase is expected
[2]. Considering the increased concentration of the absorber center in the RAM, this higher proportion
generates denser networks resulting in more efficient electrical conduction mechanisms [17].
The linear behavior of the curves presented at all studied frequencies is observed, both for the real
permittivity, ε', and for the imaginary permittivity, ε”, with the exception of the measurement made
between 11.3 and 12.4 GHz (Fig. 8 (d)), which presents an ε” rise at the end of the frequency range.

Fig. 8. Measures of real to imaginary complex permittivity in the X band of samples with ferrocarbonyl concentrations
ranging from 20 to 70% by mass. (a) 8.2-9.4 GHz, (b) 9.2-10.4 GHz (c) 10.2-11.4 GHz and (d) 11.4-12.4 GHz.
The real and imaginary permeability values remained constant, without any variations, throughout
the studied frequency range. The values obtained in the real, μ', and imaginary, μ” magnetic
permeability measurements are, respectively, 1 and 0, in accordance with the literature, with the free
space measurement methods in the Reflection and Transmission Method [17], [18].
Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 12

E. Reflectivity Measures
NRL arc measurement method allows the characterization of electromagnetic radiation absorbing
materials (RAM) as a function of angles of incidence and reception. The measurement system consists
of two standard horn-type antennas positioned according to the far-field condition and with an incidence
angle and reflection normal to the reflecting plane [9]-[18]-[19]. The measurements were performed
with the NRL arc electromagnetic characterization measurement system with angular positioning of the
RX horn antenna, as shown in Fig. 4. In this configuration, the material's reflectivity can be measured
at different angles of incidence.
Thus, if the reflectivity indicates a level of -40 dB, the material will have an efficiency of 99.99% of
electromagnetic field absorption, at the indicated frequency [20].
Fig. 9 shows the curve of the reflectivity results for the incidence angle of 30° in the frequency range
of 8.2 to 12.4 GHz. For samples A, B, C, E, F, G, H, I and J the attenuation shows an increase with the
addition of ferrocarbonyl in the material throughout the X band. In the frequency range from 8.2 GHz
to 9.3 GHz the curve of sample D shows a decrease in attenuation from -15 dB to -9 dB. There was an
increase in attenuation from -8 dB to -15 dB in the frequency range of 10.2 GHz to 12.4 GHz in the L
sample curve. The behavior of the attenuation curves in the samples ensures that the ferrocarbonyl
interferes with the electromagnetic field that focuses on it and its absorption varies from 50% to
approximately 96.9%. The sample L can be highlighted, since it presents greater attenuation in almost
the entire range of analyzed frequencies, and sample D, as well, with a lower concentration of
ferrocarbonyl, but with attenuation of up to -15 dB at the beginning of the range. The attenuation of the
electromagnetic wave by an absorber varies according to the variation in the concentration and thickness
of the RAM [21].

Fig. 9: Reflectivity curves for 30° angle of incidence of samples with ferrocarbonyl concentrations ranging from 20 to
70% by mass in the X band.

Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 13

A similar behavior occurs in the curve of the reflectivity results for the incidence angle of 45° in the
frequency range of 8.2 to 12.4 GHz (Fig. 10). The attenuation is significant for the L sample, increasing
from -9 dB to -13 dB in the frequency range 10.2 GHz to 12.4 GHz. For samples A, B, C, D, E, F, G,
H, I and J attenuation increases with the addition of ferrocarbonyl in the material across the band.

Fig. 10: Reflectivity curves for 45° angle of incidence of samples with ferrocarbonyl concentrations ranging from 20 to 70%
by mass in the X band.

Fig. 11 shows the curve of the reflectivity results for the incidence angle of 60° in the frequency
range of 8.2 to 12.4 GHz and 3201 points of resolution. The attenuation curves show a homogeneous
behavior of the samples over the entire frequency range. The curve of sample A had an attenuation of -
19 dB at 8.3 GHz, an efficiency of 99% at this frequency.

Fig.11: Reflectivity curves for 60° angle of incidence of samples with ferrocarbonyl concentrations ranging from 20 to
70% by mass in the X band.

For comparison purposes, samples of absorbers were prepared for measurements in a waveguide,
with dimensions of 22.9 mm x 10.2 mm, 4 mm thick.

Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 14

The electromagnetic characterization of the obtained RAM was performed using the waveguide
technique, in the X-band frequency range (8.2 to 12.4 GHz), using an Agilent Technologies brand
Vector Network Analyzer, model PNA- L-N5230C, at the Electronic Warfare Laboratory of the
Technological Institute of Aeronautics (ITA).
The graph in Fig. 12 shows the results of measurements performed on the waveguide. As in free
space measurements, the increase in the concentration of ferrocarbonyl in the RAM promotes an
increase in the reflectivity of the electromagnetic wave.

Fig. 12. Reflectivity values measured in the waveguide in the X band of samples with ferrocarbonyl concentrations ranging
from 20 to 70% by mass.
Table III shows the reflectivity values measured in the waveguide, the reflectivity in the NRL arc at
angles of 30º, 45º and 60º and the parameters S 11 and S21, in the frequency of 10 GHz. Analyzing this
table, it is possible to verify that the measured values are very close, with differences of less than 2
units.
TABLE III. REFLECTIVITY VALUES MEASURED IN THE WAVEGUIDE, THE REFLECTIVITY IN THE NRL ARC AT
ANGLES OF 30º, 45º AND 60º AND THE PARAMETERS S11 AND S21, IN THE FREQUENCY OF 10 GHZ
Samples Waveguide Arc 30o Arc 45o Arc 60o S11 S21
A -2.207 -4,443 -4,298 -5,091 -6.531 -6,914
B -7.531 -4,542 -4,748 -4,297 -6.810 -6,457
C -7.763 -4,669 -4,816 -6,226 -5.592 -7,095
D -7.319 -6,178 -5,334 -4,715 -6.671 -7,115
E -7.790 -4,496 -4,854 -5,402 -6.540 -7,161
F -3.230 -4,566 -4,341 -4,366 -6.021 -7,049
G -5.384 -4,565 -4,243 -4,388 -5.242 -6,914
H -5.816 -4,705 -5,119 -6,015 -5.195 -7,515
I -4.085 -5,277 -5,890 -6,088 -5.549 -7,463
J -4.319 -4,698 -5,083 -5,889 -7.549 -6,445
L -4.348 -6,252 -8,041 -7,39 -7.852 -7,635
According to the literature, [9]-[10]-[18], the measures performed using the NRL arc method may
present discrepancies that can be explained by the influence of electromagnetic wave diffraction on the

Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 15

sample edges. The different characterization methods reduce these discrepancies that depend on the
distance to the sample. The small variations that can be observed in the results in Table III suggest these
discrepancies. Thus, in these antennas measuring systems with angular positioning in the form of a
compass and antennas with focal lenses, it suggests a better acuity in measurements with greater
effectiveness and less scattering and reflections in the vicinity of the analyzed samples.

IV. CONCLUSION
The Electromagnetic Characterization of materials in Free Space is a simple and fast method to
determine the electromagnetic properties of flat solid materials.
The suggested setup allows the measurements of S parameters scattering and constitutive properties
of dielectric/magnetic materials and reflectivity of the material under study, just adjusting the angle of
the transmitting antenna to the angle of the sample holder containing the sample in a metal plate.
To validate the method, an uncertainty value with 95% confidence was presented, dependent on the
intrinsic properties of the frequency and thickness of a dielectric material, showing high repeatability.
For verification and validation, the measurement system was configured, calibrated and characterized
by microwave absorbers, prepared with ferrocarbonyl, in concentrations ranging from 20 to 70%.
Comparison of results suggests consistency between measurements.
The methodologies used for measurements are considered to be easy to handle as they allow testing
with absorbing elements of different materials with different sizes and geometries.
With this work, DCTA institutes and universities and companies in the aerospace industry will have
a laboratory infrastructure for the development of materials that absorb electromagnetic radiation
necessary for aerospace applications, as well as expand the training and integration of teams, enabling
the exchange of knowledge and promotion in the availability of laboratory resources specialized in
sensitive technology items.

ACKNOWLEDGMENT
The authors acknowledge CNPq (Processes no. 408560/2016-9) for the financial support and Eng.
Newton Adriano dos Santos Gomes of Electronic Warfare Laboratory (ITA) for his help
electromagnetic measurements.

REFERENCES

[1] D. A. Ronsó, “Design of Dielectric Lenses for Electromagnetic Field Collimation in the S-Band (in portuguese),”
Dissertation (master’s degree in Space Science and Technology) - Instituto Tecnológico de Aeronáutica, p. 135 fl, 2017.
[2] V. A. Silva, J. J. Pereira, E. L. Nohara and M. C. Rezende. “Electromagnetic behavior of microwave absorbing materials
based on Ca hexaferrite modified with CoTi ions and doped with La (in portuguese).” Journal of Aerospace Technology
Management. v. 1, n. 2, 2009. https://doi.org/10.5028/jatm.2009.0102255263
[3] J-M., Hyun, A. Hasan and J-R., Lee, “Development of dual port scanning free space measurement system”. Measurement
Science and Technology. v. 29, n 7, 2018. DOI 10.1088/1361-6501/aac7fd
[4] Keysightkeysight, “Basics of Measuring the Dielectric Properties of Materials.,” Application note, pp. 5989-2589. Z. 2018.
[5] Z. A. E. M. Akhter, “Free-Space Time Domain Position Insensitive Technique for Simultaneous Measurement of Complex
Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074
Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 23, No. 3, e2024280724 Aug 2024
DOI: https://dx.doi.org/10.1590/2179-10742024v23i3280724 16
Permittivity and Thickness of Lossy Dielectric Samples,” IEEE Transactions on Instrumentation and Measurement, vol.
65, pp. 2394-2405, 2016. DOI: 10.1109/TIM.2016.2581398
[6] D. Micheli, C. Apollo, R. Pastore, Marchetti, M., “Modeling of Microwave Absorbing Structure using Winning Particle
Optimization applied on electrically Conductive Nanostructured composite material”. XIX International Conference on
Electrical Machines - ICEM 2010, Rome. DOI: 10.1109/ICELMACH.2010.5607881
[7] N. Anyutin, I. Malay, A. Titarenko and A. Malyshev, "Synthesis of Reflectivity Measurement Conditions for Pyramidal
Radio Absorbing Materials in Free Space," 2018 IEEE East-West Design & Test Symposium (EWDTS), pp. 1-4, 2018.
DOI: 10.1109/EWDTS.2018.8524646
[8] S. Narayan, J. Sreeja, V.V. Surya, et al. “Radar Absorbing Structures Using Frequency Selective Surfaces: Trends and
Perspectives”. Journal of Electronic Materials. 49, pp. 1728–1741, 2020. https://doi.org/10.1007/s11664-019-07911-2
[9] Keysight, “Option 100 - Freespace Calibration: User Manual,” 2016. [Online]. Available:
<http://na.support.keysight.com/materials/help/85071webhelp/85071_opt100_grl_calibration.htm>. Access: Jun 2019.
[10] S. L. Rocha, C. C. Junqueira, E. Gambin, A. N. Vicente, A. E. Culhaoglu, E. Kemptner. “A Free Space Measurement
Approach for Dielectric Material Characterization”. Conference Paper. Source: DLR. 2013 SBMO/IEEE MTT-S
International Microwave & Optoelectronics Conference (IMOC). Rio de Janeiro, Brazil, pp. 1-5, 2013. DOI:
10.1109/IMOC.2013.6646474
[11] Keysight, “Keysight Technologies Time Domain Analysis Using a Network Analyzer”, Application Note, Literature
Number 5989-5723EN,” 2017.
[12] S. Batista and M. El-Shenawee, "Free-Space Characterization of Radar Absorbing Non-Magnetic Materials in the W-
Band," 2020 94th. ARFTG Microwave Measurement Symposium (ARFTG), San Antonio, TX, USA, pp. 1-4, 2020. DOI:
10.1109/ARFTG47584.2020.9071679
[13] A. E. M. Akhter, “Free-Space Time Domain Position Insensitive Technique for Simultaneous Measurement of Complex
Permittivity and Thickness of Lossy Dielectric Samples,” IEEE Transactions on Instrumentation and Measurement, vol.
65, pp. 2394-2405, 2016. DOI: 10.1109/TIM.2016.2581398
[14] D. A. N. A. Ghodgaonkar, “Materials using Free-Space Microwave Measurement Techniques,” 15a WCNDT, 2000. 15th.
World Conference on Nondestructive Testing. Roma (Italy), pp. 15-21, 2000
[15] S.D. Pinto, M.C. Rezende. “Morphological, Electromagnetic, and Absorbing Properties of POMA and PAni/Carbon
Black Composites”. Journal of Electronic Materials. 46, pp. 4939–4947, 2017. https://doi.org/10.1007/s11664-017-5492-y
[16] V. A. Silva and, M. C. Rezende. “S-parameters, electrical permittivity, and absorbing energy measurements of carbon
nanotubes-based composites in X-band”. Journal Applied Polymer Science, pp. 1-10. 2020. DOI: 10.1002/app.49843
[17] F. C. Reis, M. C. Rezende and B. Ribeiro, “The influence of the transparent layer thickness on the absorption capacity
of epoxy/carbon nanotube buckypaper at X-band”. Journal Applied Polymer Science. v. 138, pp. 47, 2021.
https://doi.org/10.1002/app.51407
[18] F. Gonçalves et al. “Free-Space Materials Characterization by Reflection and Transmission Measurements using
Frequency-by-Frequency and Multi-Frequency Algorithms.” Electronics. v. 7(10), pp. 260, 2018.
https://doi.org/10.3390/electronics7100260
[19] A. Delfini et al., “Experimental Reflection Evaluation for Attitude Monitoring of Space Orbiting Systems with NRL Arch
Method”. Applied Sciences, v. 11, 8632, 2021. https://doi.org/10.3390/app11188632
[20] S. M. LEE, “International encyclopedia of composites”. v.6. New York. VHC Publ., 1991
[21] S. S. Pinto, N. A. S. Gomes, V. A. Silva e M. C. Rezende. “Investigation of the Effect of Thickness and Concentration of
Ferrocarbonyl-Based Composites Applied as X-Band Microwave Absorbers (in Portuguese)”. SIGE 2019. XIX Simpósio de
Aplicações Operacionais em Áreas de Defesa (SIGE), pp. 187-190, 2019.

Brazilian Microwave and Optoelectronics Society-SBMO received 23 Nov 2023; for review 10 Jan 2024; accepted 25 Jul 2024
Brazilian Society of Electromagnetism-SBMag © 2024 SBMO/SBMag ISSN 2179-1074