Journal of ELECTRONIC MATERIALS, Vol. 49, No.

1, 2020
https://doi.org/10.1007/s11664-019-07761-y
 2019 The Minerals, Metals & Materials Society

Dual-Band Microwave Sensor for Investigation of Liquid

Impurity Concentration Using a Metamaterial Complementary
Split-Ring Resonator

YOGITA KHANNA1 and Y.K. AWASTHI 2,3

1.—Department of Electronics and Communication Engineering, Manav Rachna University,

Faridabad 121002, India. 2.—Department of Electronics and Communication Engineering, Manav
Rachna International Institute of Research and Studies, Faridabad 121002, India. 3.—e-mail:
yogendra@mru.edu.in

In this article, a dual-band microwave sensor using a complementary split-

ring resonator (CSRR) is presented that determines the concentration of any
liquid bi-mixture like water in ethanol and urea in whole milk. The proposed
sensor is unique as it is designed, simulated and fabricated using a single-
metamaterial cell structure to operate at dual frequency, i.e. 2.45 GHz and
5.8 GHz using an industrial, scientific and medical (ISM) band. The sensor is
fabricated on FR4 substrate using a typical photolithography technique, and
simulated results are in agreement with the measured results. Investigation
of the sensing mechanism for impurity concentration (i.e. water in ethanol or
urea in milk) is performed by placing the sample mixture in the pipette placed
across from the sensor. As microwave sensors respond to the change in the
dielectric constant of the nearby materials, when the liquid mixture concen-
tration varies, there is shifting in the resonant frequency at which the sensor
is designed. The proposed sensor is unique due to dual-band resonance,
reusability, compactness (12 mm 9 20 mm), low cost, noninvasiveness, non-
destructiveness, and a user-friendly approach.

Key words: Bi-mixture liquid, CSRR, dual-band, resonator, dielectric

characterization, sample under test

to that pure material. So, dielectric characterization

INTRODUCTION
becomes significant for the investigation of the
Sensors are one of the most essential components concentration of bi-mixtures which can be accom-
in the current technology scenario and are also plished by using a microwave sensor.3 In the
inevitably used for the detection of numerous microwave domain, the design of artificially tailored
adulterations in food items using noninvasive tech- materials, i.e. metamaterials (MTM), is a noncon-
niques.1,2 Although there have long been invasive ventional approach to design sensors that have
techniques as well, nevertheless, noninvasive meth- enabled the precise measurement of permittivity
ods using electromagnetic detection have become using planar technology.4–8 MTM sensors have been
popular due to their fast response and reusability. designed for quality control in the food industry,
The principle for electromagnetic detection is that bimolecular material analysis, subsurface detection,
every material possesses certain electromagnetic and chemical concentration measurement.9–11 MTM
properties, and these properties tend to change sensors have been utilized to find the dielectric
when there is an addition of some foreign material constant as well as the loss tangent of chemicals by
dielectric characterization. It is evident from prior
research that dielectric spectroscopy is the method
considered appropriate for research on food items as
(Received July 3, 2019; accepted October 22, 2019;
published online November 4, 2019) it has high sensitivity towards molecular

385
386 Khanna and Awasthi

polarization.12,13 Sensors implemented using the acts as a narrow-band reject filter, and the CSRR
MTM unit, i.e. split-ring resonator (SRR), or com- resonator etched on the ground plane provides an
plementary split-ring resonator (CSRR), are found expanded fringing area for the sample interaction in
suitable for chemical categorization and biomedical comparison with an SRR.23 As far as ring geometry
sensing for having high accuracy, compactness, and is concerned, there can be circular, square, or
low fabrication cost.11 It is also worthwhile men- octagonal geometry that can be used; however,
tioning that designing of sensors resonating at circular geometry is preferred because of its high
multiple discrete frequencies is important to sensitivity.18 The proposed sensor is unique as it
address the dispersive nature of materials where resonates with the dual-frequency band due to the
the electromagnetic properties are dependent on unique design and optimization of the CSRR. Res-
frequency.14 The multiband property of sensors can onance at multiple bands makes the sensor appro-
also be utilized to sense two liquids simultaneously priate for critical observations as the dielectric
using folded microstrip lines.15 Design and fabrica- constant is dispersive in nature, having variation
tion of multiband operation of CSRR can be done with the frequency of operation. The proposed
with a single board through appropriate isolation as sensor is utilized to measure the dielectric proper-
well as multiple boards.16–18 The measurement ties of a material.
methods adopted for dielectric characterization of The sensor structure is resonating at dual fre-
micro-fluids established so far are complex, incor- quency, i.e. 2.4 GHz and 5.8 GHz. The design of the
porating several steps and proper setup.19,20 proposed sensor is initiated with the study of the
The proposed sensor has a single CSRR on a transmission line as a wideband filter. The design
single board that operates on a dual-frequency parameters and their dimensional details are indi-
band, i.e. 2.4 GHz and 5.8 GHz. This design of the cated in Fig. 1a and b and Table I, respectively.
sensor shows its compactness and also eases the This section presents the design route of the pro-
measurement process as the position of the sample posed sensor to achieve the dual-frequency response
is fixed for measurement on both frequencies. Thus, through Fig. 2. The structural analysis begins with
the proposed sensor can be utilized for dielectric a 50-X transmission line where the width of 26the
characterization of dispersive liquid mixtures at transmission line can be calculated using Eq. 1
dual frequencies in a single step. Moreover, the  pffiffiffiffiffiffiffiffiffi
Z0 er þ1:41
setup of the sensor ensures that it is contactless and  87
reusable, as it uses a pipette attached to its center. w ¼ 7:48  h  e ð1Þ
The modeling and electromagnetic simulations of where w, h, and er values are mentioned in Table I.
the proposed sensor are accomplished using CST From Fig. 2a, it is seen that the transmission line
Design Studio. acts as a wideband filter where the insertion loss is
This paper is organized as follows. The Sensor below 20 dB. The effective permittivity is calcu-
Design and Theoretical Analysis section explains lated for the microstrip line26 when the width of the
the sensor design description, in which the theoret- line (w) is 1.47 mm and height of substrate (h) is
ical concept and its performance are supported by 0.8 mm by Eq. 2
finite element method (FEM) simulations. The
 
Fabrication and Sample Preparation section er þ 1 er  1 12h 1=2
describes the sensor fabrication and process of eeff ¼ þ þ 1 ð2Þ
2 2 w
precise sample preparation for measurement pur-
poses. The Sensor Measurement and Experimental
Analysis section gives the description of various In the proposed CSRR, the ring structure induces
sample measurements and estimation results using a magnetic field in a perpendicular direction which
OriginPro 2019. gives the required inductance and, similarly, a split
gap between the rings generates appropriate capac-
itance. Hence, the combined effects of inductance
SENSOR DESIGN AND THEORETICAL and capacitance generated play an important role in
ANALYSIS the optimization of resonant frequency in the ISM
Victor Veselago projected that such material can band.
be fabricated with a negative refractive index, i.e. Figure 2b and c show the inner and outer ring
metamaterial in 1968.21 J. B. Pendry demonstrated structures with split gaps acting as resonators such
metamaterial using SRRs and thin wire structures that they resonate at different frequencies. It indi-
in 1999,22 whereas the CSRR was validated by cates that the ring radius can be optimized so as to
Falcone in 2004 and explained by Babinet’s princi- obtain the resonance on the desired frequency.
ple.23,24 The mentioned structures are used to Figure 2d shows that when both rings are inte-
design filters having a high Q-factor.25 The pro- grated, resonance at dual frequency exists with a
posed sensor is designed using a CSRR which is minor shift, as compared to the single-ring struc-
etched on the ground plane and a 50-X transmission ture. The circular patch, when incorporated in the
line with a centered circular patch on the opposite structure, provides more stable results, as shown in
side of the printed circuit board (PCB). This layout Fig. 2e. The circular hole indicated in Fig. 2f in the
Dual-Band Microwave Sensor for Investigation of Liquid Impurity Concentration Using a 387
Metamaterial Complementary Split-Ring Resonator

Fig. 1. Structural parameters of the proposed sensor (a) indicate the CSRR etched on the ground plane with its dimensions and (b) indicate the
transmission line with a center patch optimized for a dual-band sensor structure.

Table I. Design parameters of proposed dual-band sensor using a CSRR

Name Value (mm) Description

G 0.30 Split-ring gap

A 0.62 Gap between the consecutive rings
d1 0.75 Width of the outer ring
d2 0.80 Width of the inner ring
L 20.0 Length of substrate
W 12.0 Width of substrate
R1 4.80 Outer ring radius
R 4.80 Radius of patch
er 4.30 Relative permittivity of a dielectric
T 0.035 Thickness of copper
H 0.80 Substrate height
W1 1.47 Width of transmission line
R2 1.50 Outer radius of the pipette/hole radius
tp 0.17 Wall thickness of pipette
er glass 4.82 Relative permittivity of the pipette (glass)

center of the circular patch and CSRR is used to conducting strip of the CSRR. The losses in the
place the liquid under test in a pipette. This is the ring are given by resistance RR, whereas CR and LR
final proposed sensor design with resonance at are the equivalent capacitance and inductance,
2.4 GHz and 5.8 GHz. respectively, for the proposed sensor structure.
The electric field intensity distribution is shown The resonance frequencies of the equivalent cir-
in Fig. 3 to study the behavior of CSRRs on both cuit for both ring resonators are defined by Eq. 3
resonant frequencies using the field monitoring tool given as27
available in CST Design Studio. It is evident from 1
Fig. 3a and b that the electric field concentration at fR ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð3Þ
2.4 GHz and 5.8 GHz can be observed and that each 2p LR ðCR þ CC Þ
CSRR has its own electric field concentration in the
ground plane corresponding to a resonant fre- The quality factor of the resonator should be high,
quency. This indicates that the proposed design and it is indicated by Eq. 4
has expanded over the existing CSRR designs used sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
for the single-band operation. CR þ CC
The equivalent structure of a CSRR as shown in Q ¼ RR ð4Þ
LR
Fig. 4 is etched on a ground plane., The unit cell is
represented by lumped elements resistance (R),
inductance (L), and capacitance (C) in the electrical Substrate material was fixed while designing the
analogy. Inductance L/2 is by virtue of the trans- ring resonator as FR4, which clearly indicates CC
mission line structure, and capacitance CC arises remains constant when sensing any change near
due to the coupling of the ground plane and the sensor. A change in the resonant frequency of
388 Khanna and Awasthi

Fig. 2. Scattering results of structural variation of the proposed sensor that resonates at 2.4 GHz and 5.8 GHz frequency (a) indicate the planar
transmission line and (b) show the variation of S11 and S21 with respect to inner ring independently. (c) Variation of S11 and S21 with respect to the
outer ring independently. (d) Changes when both rings are integrated. (e) Changes due to the circular patch. (f) Changes due to a drilled hole.

the sensor in the air is observed when the sample CR ¼ Ceff þ esut  Cp ð5Þ
under test (SUT) is placed. Theoretically, for pure
dielectric materials, the inductance LR of the ring is where Ceff represents the effective capacitance due
not affected; only the ring capacitance CR varies to a substrate in near space and the packaging of
according to the dielectric constant of the sample. the sample, and Cp is the capacitance offered by the
For the microwave measurement at more than one glass pipette.
frequency simultaneously, a unique CSRR unit cell
is etched, resulting in dual-band resonance. For the FABRICATION AND SAMPLE PREPARATION
proposed dual-band sensor, the working principle Standard photolithography and chemical etching
can be explained as the sample complex permittivity technique are used to fabricate the proposed CSRR-
is given by esut ¼ e0sut þ e00sut . Then the effect on the based metamaterial sensor on the FR4 substrate
ring capacitance CR due to sampling can be calcu- with a thickness of 0.8 mm and a relative permit-
lated by Eq. 5 as tivity of 4.3 with a loss tangent of 0.0023. A micro-
Dual-Band Microwave Sensor for Investigation of Liquid Impurity Concentration Using a 389
Metamaterial Complementary Split-Ring Resonator

Fig. 3. Distribution of electric field intensity of the proposed sensor at (a) 2.4 GHz and (b) 5.8 GHz.

Fig. 4. Equivalent electrical circuit of the proposed sensor having a width (W) of 12 mm and length (L) of 20 mm.

Table II. Comparison of published sensors with the proposed dual-band sensor
No. of Resonant Fabrication/
resonant frequency measurement Resonator
bands (GHz) Size (mm2) Characterization process type

Single-band 2.30 ISM 20 9 28 Water–ethanol solution Simple CSRR7

Single-band 1.30 10 9 50 Air, Teflon, RO3003, FR4 Simple CSRR16
Single-band 2.70 40 9 60 Teflon, poly-ethylene, plexiglass, Simple CSRR17
PVC, rubber, wood
Single-band 2.10 – Water–ethanol solution Complex CSRR17
Single-band 2.00 26 9 26 Water–ethanol solution Complex SRR18
single-band 1.90 110.6 9 54.5 Water–NaCl solution Complex Rectangular
cavity20
Dual-band 6.6 24 9 15 Water–ethanol, water–methanol Simple SRR14
8.1 solution
Dual-band 1.6 – Air, water, isopropyl, ethanol, Moderate Folded
3.5 methanol microstrip
line15
Dual-band 2.45 12 9 20 Milk–urea solution Very simple CSRR
5.80 ISM water–ethanol, etc. [proposed]

fluidic glass pipette is used in the hole of 1.5-mm sensors having a dual-band sensor used to investi-
outer diameter and 1.33-mm inner diameter. The gate different materials.
hole cut inside the sensor is the same as the pipette Sensor results are taken when it is bare and
outer diameter so that no additional setup is loaded with the SUT in the capillary. Figure 5a
required for the liquid insertion. Table II indicates shows the measured and simulated results of the
various attributes of the prior research for the sensor with no capillary or SUT. It can be noted that
390 Khanna and Awasthi

Fig. 5. Results of the proposed sensor (a) show the simulated and measured S21 parameter results of the bare sensor and (b) show the
fabricated sensor indicating the ground layer and its transmission line with a patch and hole for the pipette.

there is a slight divergence of measured results from

the simulated results due to the fabrication errors
and the parasitic components.
For testing, two sets of samples have been
prepared. The first sample set consists of binary
liquid, i.e. ethanol and water, that varies with the
percentage concentration initiated with 0% water
(100% ethanol) to 100% water (0% ethanol). The
ethanol was AR grade and received from West India
Chemicals (Gujarat, India). The second sample set
was prepared for urea and milk, where urea is in
powdered form and milk in liquid form. Urea and Fig. 6. Measurement arrangement for the sensor using a pipette
milk were mixed 30 min before the measurement in indicating the hole drilled and pipette fitted into the hole.
a glass pipette. Samples for the test were prepared
by mixing urea in the proportion of 75 mg, 300 mg, SENSOR MEASUREMENT
750 mg, and 1500 mg with 100 mL of milk each AND EXPERIMENTAL ANALYSIS
time. As per the Food Safety and Standards Author- For measurement purposes, the pipette was filled
ity of India (FSSAI) Act of 2006 and Prevention of completely with the sample and placed across the
Food Adulteration (PFA) Act of 1954, the maximum sensor hole, and a glass beaker was placed under-
limit of urea that should be present in milk is neath. For each sample, a fresh pipette and beaker
70 mg/100 mL.28 Therefore, the sample set was were used for maximum accuracy in the results.
prepared taking into consideration the maximum Whenever a new sample was filled in the pipette,
level as per the establishments. To get accurate there was a change in the permittivity of samples
results from samples, urea particles must be fully that caused a change in the resonant frequency and
dissolved in milk. For measurement, the milk that Q-factor of the sensor. The resonance frequency and
has been utilized is one having high-fat content-6,28 Q-factor are highest when the sensor was kept in
often termed as whole milk. The milk sample was free space without any sample, as shown in Fig. 5a.
collected from local Indian brands with a current When the pipette was filled with the liquid mixture,
date of packaging so as to reduce the effects of the resonance frequency bands shifted towards the
contamination. Urea or carbamide of industrial lower frequency, as shown in Fig. 7a.
technical grade was bought from Milton Chemical The measured values from the experimental
Pvt., Ltd. having a particle size of 1–3 mm and 46% setup can be analyzed using a simple technique
nitrogen. The arrangement of the sensor package where complex permittivity, frequency shift, and Q-
used to measure the S-parameters is outlined in factor are related to the function given by Eqs. 6
Fig. 6. The sample was prepared in a glass beaker and 7.
shaken well before taking measurements. During
measurement, the response of the sensor was De0 ¼ F 0 ðDfr ; DQÞ ð6Þ
monitored in real time with a vector network
analyzer (VNA, N5222A) that can sweep across
the microwave frequency range of interest. De00 ¼ F 00 ðDfr ; DQÞ ð7Þ
Dual-Band Microwave Sensor for Investigation of Liquid Impurity Concentration Using a 391
Metamaterial Complementary Split-Ring Resonator

Fig. 7. Measured results of the proposed sensor: (a) S21 parameter for the sensor with variation in water concentration at 2.4 GHz; (b) change in
resonant frequency and cumulative frequency shift with change in water concentration at 2.4 GHz; (c) change in Q-factor and amplitude variation
at 2.4 GHz; (d) S21 parameter for the sensor with variation in water concentration at 5.8 GHz; (e) change in resonant frequency and cumulative
frequency shift with change in water concentration at 5.8 GHz; (f) change in Q-factor and amplitude variation at 5.8 GHz; (g) reference and
estimated complex permittivity at 2.4 GHz; (h) reference and estimated complex permittivity at 5.8 GHz.

where Dfr ¼ fref  fsam , DQ ¼ Qref  Qsam , De0 ¼ sensitivity of the proposed sensor is 1.92 MHz/
e0ref  e0sam and De00 ¼ e00ref  e00sam . The reference value percentage of water concentration in ethanol. Fig-
and sample value are indicated with ref and sam, ure 7c indicates the cumulative Q-factor in refer-
respectively, and the reference value for analysis ence to the change in resonant frequency that has a
purposes is taken at 50% water. The D value in each nonlinear trend for dielectric variations at 2.4 GHz.
term indicates the changes with respect to the Similarly, Fig. 7d shows the sample measurements,
reference value and also indicates the trend in whereas Fig. 7e and f indicate extraction of fre-
which the frequency is shifting due to the change in quency shift and Q-factor at 5.8 GHz, respectively.
the percentage of the water content in ethanol. In Figure 7g and h show the final results indicating
Fig. 7a and b, the resonant frequency shift has been the permittivity attained from the reference and the
extracted at 2.4 GHz based upon the variation in estimations for 2.4 GHz and 5.8 GHz, respectively.
water concentration. The sensitivity of the sensor is Selectivity characterizes the capability of a sensor to
defined as the slope of the curve fitting equation, i.e. determine the concentration of the detected compo-
y = 1.92x + 23.36. It indicates that resonant fre- nents in a mixture. Since this sensor is analyzed on
quency variation is decreasing linearly with an the basis of dielectric capability, it is worthwhile
increase in water concentration, and so the mentioning that the proposed sensor is selective for
392 Khanna and Awasthi

the dielectric mixtures. The sensor is selective for To validate the dielectric model fitting, another
the change in the permittivity values from 9 (0% nonconventional mixture is taken, i.e. milkurea
water) to 79 (100% water). where milk is in liquid form and urea is powder
Equations 6 and 7 are used to find the estimated form, and upon shaking well, urea dissolves in milk,
results, and these are expanded for different values thus altering the permittivity as it increases the
of resonant frequency and the Q-factor as given in non-protein nitrogen content. The reference val-
Eq. 10 and solved using OriginPro 2019 using the ues12 for complex permittivity for pure milk and
fitting function in Eqs. 8 and 9 for 2.4 GHz and milk with urea are observed to be in concurrence
5.8 GHz. with the estimated values.
The curve fitting coefficients indicated in Table
X
3 X
3
III are solved using third order equations given by
De0n ¼ ai ðDFn Þi þ bi ðDQn Þi ð8Þ
(8) and (9) for estimating the real and imaginary
i¼1 i¼1
permittivity for urea-milk mixture. The square of
the correlation coefficient (R2) is greater than 0.99
X
3 X
3 at 2.4 GHz and 5.8 GHz. The sensor is designed to
De00n ¼ ci ðDFn Þi þ di ðDFn Þi ð9Þ sense 0 mg urea/100 mL milk (i.e. pure whole milk)
i¼1 i¼1 to a saturated mixture of urea/milk. However,
during the experimental analysis, the sensor
DFn ¼ Dfr1 ; Dfr2 . . . Dfr7 started sensing a 75-mg urea sample in 100 mL of
milk due to the measuring equipment’s resolution.
DQn ¼ DQ1 ; DQ2 . . . DQ7 Moreover, beyond 1500 mg of urea in 100 mL of
ð10Þ
De0n ¼ De001 ; De02 . . . De07 milk, no change was observed in the results. Based
on the measurement results, it can be speculated
De00n ¼ De001 ; De002 . . . De007
that the minimum limit of a sensor to sense urea in
whole milk is 75 mg as shown in Fig. 8.
The curve fitting coefficients indicated in Table
III fit into the third-order equations given by (8) and CONCLUSION
(9) for estimating the real and imaginary permit-
In this paper, a microwave dual-band sensor has
tivity for water-ethanol mixture. The square of
been designed to determine the concentration of
correlation coefficient (R2) for the equations solved
water in ethanol, urea in whole milk, water in whole
is greater than 0.98 for 2.4 GHz and 5.8 GHz. So,
milk, and any liquid bi-mixture. This detection
the estimated values are replicas of the values
technique of a liquid mixture is noninvasive and
taken from Ref. 29.

Fig. 8. Measured results for the milk–urea mixture: (a) S21 parameter at 2.4 GHz; (b) S21 parameter at 5.8 GHz; (c) complex permittivity results
at 2.4 GHz; (d) complex permittivity results at 5.8 GHz.
Dual-Band Microwave Sensor for Investigation of Liquid Impurity Concentration Using a 393
Metamaterial Complementary Split-Ring Resonator

Table III. Coefficients of curve fitting for a water–ethanol dielectric model using OriginPro 2019
Water–ethanol Milk–urea

2.4 GHz 5.8 GHz 2.4 GHz 5.8 GHz

Coefficients De¢ De¢¢ De¢ De¢¢ De¢ De¢¢ De¢ De¢¢

a1 0.6059 0.03708 2.0560 1.34123 0.57758 0.73735 0.19671 2.4177

a2 0.00112 0.00124 0.0079 0.01211 0.03273 0.20319 0.06202 0.41951
a3 0.00002 0.000017 0.0005 0.00041 – – – –
b1 0.08881 0.10189 3.3645 3.02246 1.26861 4.52566 0.0128 4.31444
b2 0.0176 0.00088 0.0526 0.00128 1.09337 0.65853 0.42894 1.21549
b3 0.00027 0.000015 0.0054 0.00313 – – – –

Table IV. Comparison of the proposed complementary split-ring resonator (CSRR) sensor with other food
sensors
Sensor type Adulteration Methodology

Copper electrode with poly- Detergents, ammonium sulphate [(NH4)2SO4], sodium Invasive, sensor impedance
methyl-methacrylate hydroxide (NaOH), sodium-bi-carbonate (NaHCO3) and variation1
(PMMA) common salt (NaCl) in milk
Optomechanical sensor Animal fat in clarified butterfat Invasive, time period
variation of simple
harmonic motion2
Electrical impedance sensor Drinking water, deionized water, hydrogen peroxide Invasive, electrical
(H2O2), sodium hydroxide (NaOH), and formaldehyde impedance spectroscopy30
(CH2O)
Microwave sensor Water in milk Noninvasive, signal
attenuation through
sample31
Microwave metamaterial Fiber content and ratio in construction materials Noninvasive, signal
sensor attenuation through
sample32
Proposed CSRR sensor Water in ethanol, urea in milk Noninvasive, permittivity
variation proposed

nondestructive, as shown in Table IV. The sensor has 2. Pallavi Gupta, Anwar Sadat, and M.J.R. Khan, IEEE Sens.
been fabricated on abundantly available substrate J. 14, 2930 (2014).
3. Z. Akhter and M.J. Akhtar, IEEE Trans. Instrum. Meas. 65,
FR4 using a typical photolithography technique, and 2394 (2016).
simulated results also have better agreement with 4. S.P. Chakyar, S.K. Simon, C. Bindu, J. Andrews, V.P. Jo-
the measured results. The measured results have seph, S.P. Chakyar, S.K. Simon, C. Bindu, J. Andrews, and
been validated, and the permittivity of unknown V.P. Joseph, J. Appl. Phys. 121, 054101 (2017).
5. C. Lee and C. Yang, IEEE Sens. J. 14, 695 (2014).
substrates is also calculated by using complex per- 6. M.A.H. Ansari, A.K. Jha, Z. Akhter, and M. Jaleel Akhtar,
mittivity functions and nonlinear curve-fitting equa- IEEE Sens. J. 18, 6596 (2018).
tions. The sensor’s performance is sufficient enough 7. E.L. Chuma, S. Member, Y. Iano, G. Fontgalland, S. Mem-
to support the proposed work. However, a sensor ber, L. Lorenzo, and B. Roger, IEEE Sens. J. 18, 9978 (2018).
might be improved for multiband sensing which is 8. M.A.H. Ansari, A.K. Jha, and M.J. Akhtar, IEEE Sens. J.
15, 7181 (2015).
used for determining the accurate complex permit- 9. L. Benkhaoua, S. Mouissat, M.T. Benhabiles, Y. Yakhlef,
tivity/permeability of magneto-dielectric materials M.L. Riabi, International Microwave Bio Conference (IM-
so that the permittivity of unknown substrates is BIOC) (2017), pp. 1–3.
calculated by using complex permittivity functions 10. D.-K. Lee, J.-H. Kang, J. Kwon, J.-S. Lee, S. Lee, D.H. Woo,
J.H. Kim, C.-S. Song, Q.-H. Park, and M. Seo, Sci. Rep. 7,
and nonlinear curve-fitting equations. The designed 8146 (2017).
sensor is found to be reusable, compact, low-cost, 11. S. Wang, L. Xia, H. Mao, X. Jiang, S. Yan, H. Wang, D. Wei,
noninvasive, nondestructive, and user-friendly. H. Cui, and C. Du, IEEE Photonics Technol. Lett. 28, 986
(2016).
REFERENCES 12. K. Zhao, Y. Liu, and Q. Zhang, J. Mol. Liq. 273, 37 (2019).
1. M. Chakraborty, S. Member, and K. Biswas, IEEE Sens. J. 13. S.N. Jha, K. Narsaiah, A.L. Basediya, R. Sharma, P. Jais-
18, 2395 (2018). wal, R. Kumar, and R. Bhardwaj, J. Food Sci. Technol. 48,
387 (2011).
394 Khanna and Awasthi

14. H. Zhou, D. Hu, C. Yang, C. Chen, J. Ji, M. Chen, Y. Chen, 25. J. Bonache, I. Gil, J. Garcı́a-Garcı́a, and F. Martı́n, IEEE
Y. Yang, and X. Mu, Sci. Rep. 8, 14801 (2018). Trans. Microw. Theory 54, 265 (2006).
15. Nikolina Jankovic and Vasa Radonic, Sensors (Basel) 17, 26. Y.K. Awasthi, H. Singh, M. Sharma, S. Kumari, and A.K.
2713 (2017). Verma, J. Eng. 9, 512 (2017).
16. M.S. Boybay and O.M. Ramahi, IEEE Trans. Instrum. 27. J. Bonache, M. Gil, and I. Gil, IEEE Microw. Wirel. Compon.
Meas. 61, 3039 (2012). Lett. 16, 543 (2006).
17. S. Ramya and I. Srinivasa Rao, Prog. Electromagn. Res. 50, 28. T. Azad and S. Ahmed, Int. J. Food Contam. 3, 1 (2016).
23 (2016). 29. J.Z. Bao, M.L. Swicord, and C.C. Davis, J. Chem. Phys. 104,
18. A. Ebrahimi, W. Withayachumnankul, S. Al-Sarawi, and D. 4441 (1996).
Abbott, IEEE Sens. J. 14, 1345 (2014). 30. G. Durante, W. Becari, F.A.S. Lima, and H.E.M. Peres,
19. W. Withayachumnankul, K. Jaruwongrungsee, A. IEEE Sens. J. 16, 861 (2016).
Tuantranont, C. Fumeaux, and D. Abbott, Sens. Actuators A 31. D. Agranovicha, I. Renhartb, P.B. Ishaia, G. Katzc, D.
Phys. 189, 233 (2013). Bezmanc, and Y. Feldman, Elsevier Food Control 63, 195
20. G. Gennarelli, S. Romeo, M.R. Scarfı̀, and F. Soldovieri, (2016).
IEEE Sens. J. 13, 1857 (2013). 32. O. Akgol, E. Unal, M. Bagbanci, M. Karaaslan, U.K. Sevim,
21. V. Veselago, L.S. Braginsky, V. Shklover, and C. Hafner, J. M. Ozturk, and A. Bhadauria, J. Electron. Mater. 48, 2469
Comput. Theor. Nanosci. 3, 189 (2006). (2019).
22. J.B. Pendry, A.J. Holden, D.J. Robbins, and W.J. Stewart,
IEEE Trans. Microw. Theory Tech. 47, 2075 (1999).
23. F. Falcone, T. Lopetegi, and J.D. Baena, IEEE Microw.
Wirel. Compon. Lett. 14, 280 (2004). Publisher’s Note Springer Nature remains neutral with
24. F. Falcone, T. Lopetegi, and M.A.G. Laso, Phys. Rev. Lett. regard to jurisdictional claims in published maps and institu-
93, 197401 (2004). tional affiliations.