sensors Letter Highly Sensitive Reflective-Mode Defect Detectors and Dielectric Constant Sensors Based on Open-Ended Stepped-Impedance Transmission Lines Pau Casacuberta 1 , Jonathan Muñoz-Enano 1, * , Paris Vélez 1 , Lijuan Su 1 , Marta Gil 2 and Ferran Martín 1 1 CIMITEC, Departament d’Enginyeria Electrònica, Universitat Autònoma de Barcelona, Bellaterra, 08193 Barcelona, Spain; Pau.CasacubertaO@e-campus.uab.cat (P.C.); Paris.Velez@uab.cat (P.V.); Lijuan.Su@uab.cat (L.S.); Ferran.Martin@uab.cat (F.M.) 2 Departamento Ingeniería Audiovisual y Comunicaciones, Universidad Politécnica de Madrid, 28031 Madrid, Spain; Marta.Gil.Barba@upm.es * Correspondence: Jonatan.Munoz@uab.cat  Received: 12 October 2020; Accepted: 30 October 2020; Published: 31 October 2020  Abstract: In this paper, reflective-mode phase-variation sensors based on open-ended stepped-impedance transmission lines with optimized sensitivity for their use as defect detectors and dielectric constant sensors are reported. The sensitive part of the sensors consists of either a 90◦ high-impedance or a 180◦ low-impedance open-ended sensing line. To optimize the sensitivity, such a sensing line is cascaded to a 90◦ transmission line section with either low or high characteristic impedance, resulting in a stepped-impedance transmission line configuration. For validation purposes, two different sensors are designed and fabricated. One of the sensors is implemented by means of a 90◦ high impedance (85 Ω) open-ended sensing line cascaded to a 90◦ low impedance (15 Ω) transmission line section. The other sensor consists of a 180◦ 15-Ω open-ended sensing line cascaded to a 90◦ 85-Ω line. Sensitivity optimization for the measurement of dielectric constants in the vicinity of that corresponding to the Rogers RO4003C substrate (i.e., with dielectric constant 3.55) is carried out. The functionality as a defect detector is demonstrated by measuring the phase-variation in samples consisting of the uncoated Rogers RO4003C substrate (the reference sample) with arrays of holes of different densities. Keywords: microwave sensors; stepped-impedance transmission lines; dielectric constant sensor; phase-variation sensors 1. Introduction Many approaches for the measurement of the dielectric constant of materials and other variables related to it, e.g., material composition, or for the detection of defects in samples, using microwaves have been reported. Of particular interest are those approaches based on fully planar structures, since planar sensors are low-cost; are compatible with other technologies (e.g., microfluidics); exhibit a low profile; and can be implemented on flexible substrates, including organic substrates, such as paper, and even on recyclable or compostable materials. Additionally, the associated electronics for the generation of the microwave signals, as well as the post-processing electronic module, and, eventually, the required circuitry for wireless communication, can be integrated with the sensitive part of the sensor, typically consisting of a transmission line configuration, a set of planar resonators, or a combination of both types of elements. Although the most typical strategy for sensing using microwaves is frequency variation [1–17] (the sensitive element being a resonant element), such an approach requires wideband signals for Sensors 2020, 20, 6236; doi:10.3390/s20216236 www.mdpi.com/journal/sensors Sensors 2020, 20, 6236 2 of 11 measurement (this applies also to frequency splitting sensors [18–27]), thereby increasing the cost of the associated electronics. The reason is the requirement of wideband voltage controlled oscillators (VCOs). To alleviate this limitation, single-frequency sensors constitute a good alternative. In such sensors, a single-tone harmonic signal suffices for sensing. Examples of single-frequency sensors based on coupling modulation have been reported [28–42]. In such sensors, the working principle is the variation of the transmission coefficient of a transmission line at the frequency of operation caused by a variation in the coupling level between such a line and a sensing resonant element (or set of resonant elements). Such sensors have been mainly devoted to the measurement of linear and angular displacements and velocities, as far as a relative displacement between the line and the resonant element modifies the coupling level between both elements. An alternative approach for sensing using a harmonic signal is phase-variation [43–50]. Phase-variation sensors are especially suited for material characterization, including dielectric constant measurements, determination of material composition, and defect detection, among others. In these sensors, the working principle is the variation of the electrical length of the sensing line caused by the presence of the so-called material (or sample) under test (MUT) on top of it. Such change in the electrical length of the sensing line in turn modifies the phase of the transmission coefficient (transmission-mode sensors) or the phase of the reflection coefficient (reflective-mode sensors) of the whole sensing structure, which are easily measurable quantities. Highly sensitive phase-variation sensors based on ordinary lines have been reported at the expense of a relatively large size for the sensing region [44,45] (it is well known that the sensitivity of the phase of a transmission line with the dielectric constant of the material surrounding it increases by increasing the length of the line and/or the frequency of the injected signal). As an alternative to ordinary lines, artificial lines exhibiting strong dispersion are also useful for the implementation of phase-variation sensors devoted to material characterization. Thus, composite right- and left-handed (CRLH) transmission lines [43], electro-inductive-wave (EIW) transmission lines [46], and slow-wave transmission lines [48] have been applied to the implementation of highly sensitive dielectric constant sensors. Despite the fact that with these sensors very good sensitivity with a relatively small sensing area was obtained, the design of artificial lines is not exempt of a certain complexity. Moreover, since the dispersive behavior of artificial lines is limited by the implementable values of the reactive elements loading the lines (CRLH or slow-wave lines), or by the effects of losses (EIW lines), increasing the sensitivity up to extraordinary values ultimately means to increase the length of the lines, that is, the size of the sensing region. Recently, a different approach for the implementation of phase-variation sensors operating in reflection has been reported [49] (other reflective-mode sensors operating differentially are presented in [51]). These sensors consist of an open-ended stepped-impedance transmission line, where the sensitive region is merely the open-ended line section, which should be either a high-impedance 90◦ line or a low-impedance 180◦ line in order to optimize the sensitivity. The step-impedance discontinuities further contribute to enhancing the sensitivity, provided the line sections with alternating high and low impedance exhibit electrical lengths of 90◦ , as demonstrated in [49]. Indeed, there is a multiplicative effect, and the sensitivity can be enhanced as one wishes, at the expense of a degradation in the linearity. Importantly, such sensitivity enhancement does not represent an increase in the size of the sensing region, since it is achieved by merely adding further 90◦ stages to the stepped-impedance transmission line structure. Moreover, sensor design is extremely simple, since these phase-variation sensors are made of a cascade of ordinary lines with alternating high and low impedance. A limited aspect, as mentioned, is the linearity. For this main reason, in the design of the sensor, a reference dielectric constant for the MUT should be chosen. Consequently, the sensitivity can be optimized in the vicinity of such a reference dielectric constant value, εREF . In [49], the reference dielectric constant was considered to be the one of air (εREF = 1). In this paper, it is demonstrated that the sensitivity can be optimized for arbitrary values of εREF . Therefore, not only variations in the dielectric constant of the MUT in the vicinity of the reference value can be accurately measured, but it is also possible to Sensors2020, Sensors 2020,20, 20,6236 6236 33 of of 11 12 measured, but it is also possible to use the sensor as defect detector. With such later functionality, it is expected use that the sensor asthe sensor defect is ableWith detector. to determine such lateriffunctionality, the so-calleditreference sample is expected is absent that the sensorof is defects able to or not (since, determine if thetypically, so-called such defects reference sampleare ismanifested as tinyorvariations absent of defects not (since, in the effective typically, dielectric such defects are constant of as manifested thetiny reference sample). variations in the effective dielectric constant of the reference sample). 2. 2. Sensor Sensor Design Design and and Fabrication Fabrication The The structure structure of of the the considered considered reflective-mode reflective-mode phase-variation phase-variation sensorssensors is is depicted depicted inin Figure Figure 1. 1. It It consists consists ofof an anopen-ended open-endedsensing sensingline linewith phaseφϕs sand withphase andcharacteristic impedanceZZs ,s,cascaded characteristicimpedance cascaded to to aa transmission line section (the design line) with phase φ and characteristic transmission line section (the design line) with phase ϕ and characteristic impedance Z. In [49], it impedance Z. In [49], itwas was demonstrated demonstrated that that forfor sensitivity sensitivity optimization, optimization, thethe phase phase of the of the design design lineline mustmust be to be set setϕto= φ = 90 90°, ◦ , whereas the phase of the sensing line must be set to either φ = 90◦ or φ = 180◦ . Moreover, whereas the phase of the sensing line must be set to either ϕs = s90° or ϕs = 180°. s Moreover, the the characteristic characteristic impedances impedances of of such such lineline sections sections shouldbebeset should settotoasasmuch muchextreme extremevalues values asas possible. possible. Specifically, ◦ Specifically, ifif the the phase phase ofofthe thesensing sensing line line isisset settoto90 90°,, then then the the characteristic characteristic impedance impedance of of such such aa line line must must bebe high, high, as as compared compared to to Z Z00, the reference impedance impedance of the ports, whereas whereas the the characteristic characteristic impedance ◦ sensing line, impedance of the design design line line must mustbe beset settotoa alow lowvalue. value.Conversely, Conversely, forfor thethe 180°180 sensing line, the the impedances impedances shouldshould satisfy Zs Z Z0,>inZorder 0 , in order to optimize to optimize the sensitivity. the sensitivity. Figure 1.1.Topology/schematic Figure Topology/schematic of step-impedance of the the step-impedance open-ended open-ended line reflective-mode line reflective-mode phase- phase-variation variation sensors. A single step-impedance discontinuity is considered, but further discontinuities sensors. A single step-impedance discontinuity is considered, but further discontinuities are possible. are possible. As it was reported in [49], a variation in the dielectric constant of a certain MUT in contact with As it was the sensing linereported inboth modifies [49], the a variation inlength electrical the dielectric and theconstant of a certain characteristic MUT in impedance ofcontact such a with line. the sensing Thus, line modifies the sensitivity, both theofelectrical or variation the phase length and of the the characteristic reflection φρ , the output coefficient,impedance of such a line. variable, Thus,the with thedielectric sensitivity, or variation constant of theεMUT of the MUT, phase ofinput , the the reflection variable, coefficient, ϕρ, theasoutput variable, can be expressed with the dielectric constant of the MUT, εMUT, the input variable, can be expressed as dφρ dφρ dφs dφρ dZs S = 𝑑 = 𝑑  · 𝑑  + 𝑑 · 𝑑𝑍 (1)  dεMUT dφs dεMUT dZs dεMUT 𝑆= =  +  (1) 𝑑 𝑑 𝑑 𝑑𝑍 𝑑  An exhaustive analysis carried out in [49] demonstrates that for the optimum phases of the line An of sections exhaustive the sensoranalysis carried (indicated out at the in [49] demonstrates beginning thatthe of this section), forlast the term optimum phases (1) in Equation of the line is null, sections since dφρof the ,/dZ s = sensor 0 for (indicated these at phases. the beginning Evaluation of of thethis section), sensitivity the for last these term in specific Equation phases (1) gives is null, since dϕρ,/dZs = 0 for these phases. Evaluation of the sensitivity for these specific phases gives Z0 Zs (1 − F) S = − 𝑍 𝑍 2(1 − 𝐹) φs (2a) 𝑆 = − 2Z εe f f  (2a) 2𝑍  for φs = 90◦ , and for ϕs = 90°, and Z2 (1 − F) S=− φs (2b) 2Z Zs ε− 𝑍 0(1 e f f𝐹) 𝑆=−  (2b)  εeff is the effective dielectric constant of the 2𝑍 𝑍(2a,b), for φs = 180◦ , in both cases with φ = 90◦ . In Equation sensing line covered by the considered MUT, and F is a form factor. In microstrip technology, adopted for ϕs = 180°, in both cases with ϕ = 90°. In Equation (2a,b), εeff is the effective dielectric constant of the in the present work, εeff and F are given by [52] sensing line covered by the considered MUT, and F is a form factor. In microstrip technology, adopted in the present work, εeff and F are given byεr[52] + εMUT εr − εMUT εe f f = + F (3)  +2  −2   = + 𝐹 (3) 2 2 and Sensors 2020, 20, 6236 4 of 12 / Sensors 2020, 20, 6236 ℎ 4 of 11 𝐹 = 1 + 12 (4a) 𝑊 and !−1/2 for Ws/h ≥ 1, or by h F = 1 + 12 (4a) Ws / for Ws /h ≥ 1, or by ℎ 𝑊 𝐹 = 1 + 12 !−1/2 + 0.04 1− (4b) 𝑊h  Ws ℎ 2 F = 1 + 12 + 0.04 1 − (4b) Ws h for Ws/h < 1. In Equation (3), εr is the dielectric constant of the substrate, and it has been assumed that for Ws /h < 1. In Equation (3), εr is the dielectric constant of the substrate, and it has been assumed the MUT is semi-infinite in the vertical direction. In Equation (4a,b), h and Ws are the substrate that the MUT is semi-infinite in the vertical direction. In Equation (4a,b), h and Ws are the substrate thickness and the width of the sensing line, respectively, and it is assumed that t << h, where t is the thickness and the width of the sensing line, respectively, and it is assumed that t << h, where t is the thickness of the metallic layer. thickness of the metallic layer. It is clear, in view of Equation (2a,b), that for sensitivity enhancement, the impedances should It is clear, in view of Equation (2a,b), that for sensitivity enhancement, the impedances should satisfy Zs > Z0 > Z for the 90° ◦sensing line, and Zs < Z0 < Z for the 180° sensing line. It should be satisfy Zs > Z0 > Z for the 90 sensing line, and Zs < Z0 < Z for the 180◦ sensing line. It should emphasized that for the sensing line, the required phase (90° or 180°) should be satisfied at the design be emphasized that for the sensing line, the required phase (90◦ or 180◦ ) should be satisfied at the frequency with the presence of the reference sample, with dielectric constant εREF on top of it. By design frequency with the presence of the reference sample, with dielectric constant εREF on top of contrast, the 90° phase◦ of the line section placed between the open-ended sensing line and the input it. By contrast, the 90 phase of the line section placed between the open-ended sensing line and the port (design line) should be satisfied with air on top of it (as far as the sensing region is merely input port (design line) should be satisfied with air on top of it (as far as the sensing region is merely restricted to the sensing line, as depicted in Figure 1). restricted to the sensing line, as depicted in Figure 1). With the design guidelines for sensitivity optimization, two different sensors were designed and With the design guidelines for sensitivity optimization, two different sensors were designed and fabricated. In one sensor, designated as sensor A, the structure consists in a Zs = 85 Ω open-ended fabricated. In one sensor, designated as sensor A, the structure consists in a Zs = 85 Ω open-ended sensing line with phase ϕs = 90° when the uncoated Rogers RO4003C substrate with εREF = 3.55 is on sensing line with phase φs = 90◦ when the uncoated Rogers RO4003C substrate with εREF = 3.55 is on top top of such a line, cascaded to a ◦ϕ = 90° design line with Z = 15 Ω characteristic impedance. The of such a line, cascaded to a φ = 90 design line with Z = 15 Ω characteristic impedance. The frequency frequency of operation was set to f0 = 2 GHz, and the considered substrate for sensor implementation of operation was set to f 0 = 2 GHz, and the considered substrate for sensor implementation was was the Rogers RO3010 substrate with dielectric constant εr = 10.2, thickness h = 1.27 mm and loss the Rogers RO3010 substrate with dielectric constant εr = 10.2, thickness h = 1.27 mm and loss tangent tanδ= 0.0023. For sensor B, the substrate and operation frequency are the same, whereas the tangent tan δ = 0.0023. For sensor B, the substrate and operation frequency are the same, whereas the line variables are Zs = 15 Ω and ϕs = 180°◦(when the same uncoated Rogers RO4003C substrate with line variables are Z = 15 Ω and φs = 180 (when the same uncoated Rogers RO4003C substrate with εREF = 3.55 is on topsof the open-ended sensing line), and ϕ = 90°, Z = 85 Ω for the cascaded design εREF = 3.55 is on top of the open-ended sensing line), and φ = 90◦ , Z = 85 Ω for the cascaded design line. line. With these parameters for sensors A and B, the two cases that optimize the sensitivity (high- With these parameters for sensors A and B, the two cases that optimize the sensitivity (high-impedance impedance 90° and low-impedance 180° sensing lines) are considered. The photographs of both 90◦ and low-impedance 180◦ sensing lines) are considered. The photographs of both sensors, fabricated sensors, fabricated by means of an LPKF H100 drilling machine, are depicted in Figure 2, where the by means of an LPKF H100 drilling machine, are depicted in Figure 2, where the relevant dimensions relevant dimensions are indicated (see caption). are indicated (see caption). (a) (b) Figure Figure 2.2. Photographs Photographs of of thethe designed designedandand fabricated reflective-mode fabricated reflective-modephase variation phase sensors. variation (a) sensors. Sensor (a) SensorA; (b) A; sensor B. For (b) sensor sensor B. For A, the sensor A, dimensions (in mm) the dimensions are as (in mm) arefollows: Wd = W as follows: d =ld7.5, 7.5, ld =W = 12.8, S = 12.8, WS =and 0.235 0.235 lS = = 13.5. lS and For sensor 13.5. For B, the dimensions sensor (in mm)(in B, the dimensions aremm) as follows: Wdld = Wd = 0.32, are as follows: = 0.32, 14.6, Wld S==14.6, 7.5 WS = and lS =7.5 and lS = 24.4. 24.4. Although the Although the main main interest interest in in the the present present paper paper isis the the sensitivity sensitivity analysis analysis and and optimization, optimization, providing a mathematical sensing model linking the phase variation to the dielectric providing a mathematical sensing model linking the phase variation to the dielectric constant of constant ofthe the MUTisisalso MUT alsoinsightful. insightful.Obviously, Obviously, Figures Figures 4a 4a and and 5a 5a of ofnext nextsection sectioncan canbebeused usedtotoinfer inferthe thedielectric dielectric constant of the MUT from the measured phase. Mathematically, the dependence of the phase of the reflection coefficient, the output variable, with the dielectric constant of the MUT, εMUT , is given by: Sensors 2020, 20, 6236 5 of 11 Z(Zs cot φs − Z tan φ) ! φρ = 2arctan (5) Z0 (Z + Zs tan φ cot φs ) where the dependence with εMUT is through the sensing line parameters, Zs and φs . Indeed, for the value of φ that optimizes the sensitivity (90◦ ), Equation (5) simplifies to Z2 ! φρ = 2arctan − tan φs (6) Z0 Zs Both φs and Zs depend on the effective dielectric constant of the sensing line according to ω0 ls p φs = εe f f (7) c and Z0 Zs = √ (8) εe f f respectively, where Z’ is the characteristic impedance of an hypothetical line with identical dimensions to the sensing line, but surrounded by air and with a substrate with dielectric constant εr = 1. In Equation (7), ω0 is the angular operating frequency, and c is the speed of light in vacuum. The dependence of Z’ with the line geometry can be found in several textbooks [52]. On the other hand, the effective dielectric constant depends on εMUT according to Equation (3). Thus, with the previous expressions, a mathematical model linking the input and output variable is provided. 3. Results For both designed and fabricated sensors, we obtained the phase variation in the reflection coefficient as a function of the dielectric constant of the MUT, taking as reference the phase of the reflection coefficient when the sensing line is covered by the reference substrate (indicated before). This has been done by means of electromagnetic simulation, using the Keysight ADS commercial software, theoretically, using Equations (6)–(8), and experimentally. In this latter case, several MUT samples with different dielectric constant were located on top of the sensing line, as Figure 3 illustrates. The thickness of the MUTs is roughly 3 mm (achieved by stacking two samples), sufficient to consider the substrate semi-infinite in the vertical direction. By these means, the experimental conditions are identical to those of the simulations (where the MUT is considered to be vertically semi-infinite, a necessary condition for the validity of Equation (2a,b), as discussed in [49]). The phase of the reflection coefficient was obtained by means of the Keysight 85072A vector network analyzer. It should also be mentioned that since the Keysight ADS software does not permit the simulation of structures with dielectric layers of different transverse dimensions, we first simulated solely the sensing line covered with the MUT. Then, we simulated the design line alone, and, finally, we imported the results to the Keysight ADS schematic simulator in order to obtain the phase response of the whole structure. Figure 4 depicts the phase variation as a function of the dielectric constant of the MUT, as well as the sensitivity (inferred from the simulated data points), for sensor A, whereas the corresponding values for sensor B are shown in Figure 5. The theoretical sensitivities, given by Equation (2a,b), evaluated at the dielectric constant of the reference MUT, are also given in the figures (and designated as Sth ). As can be appreciated, the agreement between the sensitivity inferred from the simulated data points and the one given by the theoretical expressions is good. The sensitivity is maximized for εMUT being equal to the dielectric constant of the REF sample, and, consequently, these sensors are especially suited for the measurement of tiny variations of the dielectric constant in the vicinity of 3.55, the dielectric constant of the REF sample. A very high value of the maximum sensitivity (S = −101.3◦ ) in the simulated sensitivity for sensor A was obtained, by virtue of the stepped-impedance configuration. Sensors 2020, 20, 6236 6 of 11 Figure 3. Detail of one of the sensors covered by the material under test (MUT). In order to avoid the Sensors 2020, 20, 6236 6 of 12 presence of an air gap between the MUT sample and the sensing line, the sample was pressured against the sensing line by means of Teflon screws. Figure 4 depicts the phase variation as a function of the dielectric constant of the MUT, as well as the sensitivity (inferred from the simulated data points), for sensor A, whereas the corresponding values for sensor B are shown in Figure 5. The theoretical sensitivities, given by Equation (2a,b), evaluated at the dielectric constant of the reference MUT, are also given in the figures (and designated as Sth). As can be appreciated, the agreement between the sensitivity inferred from the simulated data points and the one given by the theoretical expressions is good. The sensitivity is maximized for εMUT being equal to the dielectric constant of the REF sample, and, consequently, these sensors are especially suited for the measurement of tiny variations of the dielectric constant in the vicinity of 3.55, the dielectric constant of the REF sample. A very high value of the maximum sensitivity (S = Figure 3. Detail Detail of of one one of the the sensors ofsensitivity covered sensorsfor covered by byAthe the material material under underby test (MUT). testvirtue (MUT). In order to to avoid avoid the Figure −101.3°) in3.the simulated sensor was obtained, ofIn order the the stepped-impedance presence presence ofan of anair airgap gap between between thethe MUTMUT sample sample and and the the sensing sensing line, line, the the sample sample was was pressured pressured against configuration. against the sensing the sensing line byof line by means means Teflonofscrews. Teflon screws. Figure 4 depicts the phase variation as a function of the dielectric constant of the MUT, as well as the sensitivity (inferred from the simulated data points), for sensor A, whereas the corresponding values for sensor B are shown in Figure 5. The theoretical sensitivities, given by Equation (2a,b), evaluated at the dielectric constant of the reference MUT, are also given in the figures (and designated as Sth). As can be appreciated, the agreement between the sensitivity inferred from the simulated data points and the one given by the theoretical expressions is good. The sensitivity is maximized for εMUT being equal to the dielectric constant of the REF sample, and, consequently, these sensors are especially suited for the measurement of tiny variations of the dielectric constant in the vicinity of 3.55, the dielectric constant of the REF sample. A very high value of the maximum sensitivity (S = −101.3°) in the simulated(a)sensitivity for sensor A was obtained, by virtue of(b)the stepped-impedance configuration. Figure4.4. Dependence Figure Dependence of of the the differential differential phase phase ofof the thereflection reflectioncoefficient coefficient(i.e., (i.e.,∆ϕ ∆φρ = ϕ ρ =ρφ−ρϕ−ρ,REF) with φρ,REF ) the dielectric with constant the dielectric of the constant ofMUT the MUT(a) and sensitivity (a) and (b) for sensitivity (b)sensor A. The for sensor A. measured The measured data data points were points obtained were by covering obtained the sensing by covering line with the sensing line the withfollowing uncoated the following materials: uncoated RogersRogers materials: RO3010 (εMUT = RO3010 (εMUT = 10.2), FR4 (εMUT = 4.4), Rogers RO4003C (εMUT = εREF = 3.55), PLA (εMUT = 3.0, and fabricated 10.2), FR4 (ε MUT = 4.4), Rogers RO4003C (ε MUT = ε REF = 3.55), PLA (ε MUT = 3.0, and fabricated by means of means by a 3D printer). of a 3DAdditionally, the phase ofthe printer). Additionally, thephase reflection coefficient of the reflectionwas experimentally coefficient obtained with was experimentally Sensors 2020, 20, 6236 7 of 12 the sensing obtained withline theuncovered sensing line (εMUT, air = 1). (ε uncovered MUT, air = 1). (a) (b) Figure 4. Dependence of the differential phase of the reflection coefficient (i.e., ∆ϕρ = ϕρ − ϕρ,REF) with the dielectric constant of the MUT (a) and sensitivity (b) for sensor A. The measured data points were obtained by covering the sensing line with the following uncoated materials: Rogers RO3010 (εMUT = 10.2), FR4 (εMUT = 4.4), Rogers RO4003C (εMUT = εREF = 3.55), PLA (εMUT = 3.0, and fabricated by means (a) of a 3D printer). Additionally, (b) the phase of the reflection coefficient was experimentally obtained with the sensing Figure Figure line uncovered 5.5.Dependence Dependence ofofthe the air = 1). phase of the reflection coefficient (i.e., ∆ϕρ = ϕρ − ϕρ,REF) with (εdifferential differential MUT, phase of the reflection coefficient (i.e., ∆φ = φ − φ ) with ρ ρ ρ,REF thedielectric the dielectricconstant constantof ofthe theMUT MUT(a) (a)and andsensitivity sensitivity(b) (b)for forsensor sensorB. B.The TheMUTs MUTsare areidentical identicalto tothose those ofFigure of Figure4.4. Functionality as a defect detector of the proposed sensing structure relies on the variation of the effective dielectric constant of the sample under test, caused by the presence of defects or anomalies. In order to demonstrate the potential of the designed sensor to detect defects in samples, we generated defects in the reference sample, the uncoated slab of Rogers RO4003C substrate, consisting of arrays of holes across the samples, of different densities (Figure 6). (a) (b) Figure 5. Dependence of the differential phase of the reflection coefficient (i.e., ∆ϕρ = ϕρ − ϕρ,REF) with the dielectric constant of the MUT (a) and sensitivity (b) for sensor B. The MUTs are identical to those Sensors 2020, 20, 6236 7 of 11 of Figure 4. Functionalityasasa adefect Functionality defectdetector detectorofofthe theproposed proposedsensing sensingstructure structurerelies reliesononthe thevariation variationofofthe the effectivedielectric effective dielectricconstant constantofofthe thesample sampleunder undertest, test,caused causedbybythe thepresence presenceofofdefects defectsor oranomalies. anomalies. InInorder order to demonstrate to demonstrate the potential the potential of theof the designed designed sensor tosensor detectto detectindefects defects samples,inwesamples, generatedwe generated defects defects in the in the reference reference sample, sample, the the uncoated uncoated slab of Rogersslab of Rogers RO4003C RO4003C substrate, substrate, consisting of consisting arrays of of arrays holes acrossofthe holes across of samples, thedifferent samples, of different densities densities (Figure 6). (Figure 6). Figure6.6.Reference Figure Referencesample samplewith witharrays arraysofofholes holesofofdifferent differentdensities densitiesdrilled drilledacross acrossthe thesubstrate. substrate. From Fromsample sample(a–d), (a–d),the thedensity densityofofholes holesincreases. increases. The Themeasured measuredphase phase of of the reflection coefficient the reflection coefficient for forboth bothsensors sensorsAAand andB B corresponding corresponding to to the the different “defected” samples is depicted in Figure 7. The sensitivity different “defected” samples is depicted in Figure 7. The sensitivity is superior in sensor A, which is superior in sensor A, which demonstrates demonstrates its capability its capability of perfectly of perfectly detectingdetecting the presence the presence of theofsparser the sparser array array of holesof holes in the insample the sample (a) of Figure 6a, as revealed by the significant variation experienced by the phase of of (a) of Figure 6a, as revealed by the significant variation experienced by the phase the the reflection coefficient for that sample. Obviously, this high sensitivity in detecting reflection coefficient for that sample. Obviously, this high sensitivity in detecting tiny defects is due tiny defects is due to the fact that the sensors were designed by considering the “un-defected” to the fact that the sensors were designed by considering the “un-defected” sample as reference sample as reference material, ◦ ◦ material,i.e., i.e.,with withthe theoptimum optimumphasesphasesofofthe thesensing sensinglineline(90 (90°oror180 180°for forsensor sensorAAor orB,B,respectively) respectively) when such a line is covered with such material. In contrast, if the sensing when such a line is covered with such material. In contrast, if the sensing line is designed with line is designed with90°90◦or or180° ◦ 180 phase phasewhen whenititisiscovered coveredby by air, air, the the capability capability to to detect detect small smallchanges changes(“defects”) (“defects”)ininsamples samples with with dielectric constant substantially different to the one of air, decreases drastically. This isvisible dielectric constant substantially different to the one of air, decreases drastically. This is visible ininFigure Figure8,8,where where the the comparison comparison of of the theresponse responseofofsensorsensorAAwith with that that of of thethe equivalent equivalent sensor, sensor, but but tuned tuned bybyair,air, is depicted. is depicted. Thus, Thus, in in summary, summary, these these results results demonstratethat demonstrate thatdetecting detectingthe thepresence presenceof ofdefects defectsororabnormalities abnormalitiesin insamples, samples, in in comparison comparison to to aa well-known well-known reference, reference,isispossible possibleby byproperly properly designing the sensor (i.e., by taking into account the dielectric constant of the designing the sensor (i.e., by taking into account the dielectric constant of the reference sample, reference sample, and and by designing the sensing line with the required phase, either 90 ◦ or 180◦ , for sensitivity optimization). by designing the sensing line with the required phase, either 90° or 180°, for sensitivity optimization). It should be mentioned that, alternatively, the reported sensors could have been implemented in coplanar waveguide technology (CPW). In [50], it was demonstrated that the intrinsic sensitivity of CPWs is larger than that of microstrip lines. The reason is that the electric field generated by a CPW has a strong presence on top of the line, as compared to microstrip lines. However, microstrip line technology offers backside isolation. Moreover, sensitivity enhancement can be improved, if required, regardless of the specific technology, by adding further cascaded high/low impedance 90◦ line sections. Let us mention, to finish the present section, that the phase, like the frequency, is robust against the effects of electromagnetic interference and noise, at least as compared to the magnitude (e.g., the magnitude of a transmission or a reflection coefficient, a typical output variable in microwave sensors). In this regard, phase-variation sensors, as those reported in this paper, or frequency-variation sensors, are more tolerant than coupling-modulation sensors (or other sensors using either the magnitude of the reflection or transmission coefficient) to electromagnetic interferences and noise. Concerning the effects of cross sensitivities caused, e.g., by changes in temperature or humidity, phase-variation and frequency-variation sensors are not as good as differential-mode sensors, or sensors based on symmetry properties. This is because changes in environmental factors occur at scales much larger than the typical size of sensors, and, therefore, such changes are seen as common-mode stimulus by differential-mode sensors or by sensors based on symmetry properties. Nevertheless, the proposed phase-variation sensors exhibit relevant advantages, such as operation at a single frequency, robustness against electromagnetic interference and noise, one-port device operation (in reflection mode), and, most importantly, highly achievable sensitivities. Sensors 2020, 20, 6236 8 of 12 Sensors2020, Sensors 20,6236 2020,20, 6236 88ofof12 11 Figure 7. Differential phase of the reflection coefficient measured by covering the sensing line of Figure 7. Figure sensors A Differential 7. Differential and phase phase B with the ofofthe samples reflection the Figure coefficient ofreflection measured 6. coefficient by covering measured the sensing by covering line of line the sensing sensors of A and BAwith sensors and the samples B with of Figure the samples of 6. Figure 6. Comparisonofofthe Figure8.8.Comparison Figure themeasured measureddifferential differentialphase phaseofofthe thereflection reflectioncoefficient coefficientfor forsensor sensorAAand and Figure forthe 8.equivalent Comparison theequivalent of the sensor, butmeasured withthe differential thesensing sensing phase of the lineexhibiting exhibiting reflection ◦ when it for coefficient sensor A (i.e., and for sensor, but with line aaphase phaseofof90° 90 when it isisuncovered uncovered (i.e., for theby tuned tuned equivalent byair). air). sensor, but with the sensing line exhibiting a phase of 90° when it is uncovered (i.e., tuned by air). 4. Conclusions It should be mentioned that, alternatively, the reported sensors could have been implemented ItInshould in coplanar be mentioned waveguide conclusion, highly that, alternatively, technology sensitive (CPW). theitreported In [50], reflective-mode sensors sensors was demonstrated phase-variation could thathave the been to intrinsic devoted implemented sensitivity the accurate in of coplanar CPWs is larger measurement waveguide of thethan technology that of microstrip dielectric (CPW). constant oflines. In [50], MUTThe it was reason samples demonstrated inisthethat that the electric vicinity the intrinsic field generated of a certain sensitivity by a predetermined of CPW CPWs value has wereis larger a strong reported than thaton presence in this of top microstrip paper. of lines. the also It has line, as The been reasontoismicrostrip compared demonstrated that thethanks that, electric field suchgenerated lines.toHowever, by a highmicrostrip sensitivity, CPW line has a strong the technology sensors are also presence offers backside useful on top of detectors the line, isolation. as defect asorcompared Moreover, to microstrip sensitivity comparators, enhancement able lines. However, to discern can microstrip be improved, if a certain sampleifis line technology required, identical regardless offers backside of the specific or not to a reference isolation. sample. technology, Moreover, Due to thebyfact sensitivity adding that thefurther sensors enhancement cascaded merelyhigh/low can be improved, consist ofimpedance an open-ended 90°if required, line regardless of the specific technology, by adding further line (with either high characteristic impedance and 90 phase, or low impedance and 180◦ sections. sensing ◦ cascaded high/low impedance 90° line sections. Let us phase), mention, cascaded toto 90◦ design a finish the present line withsection, highthat the phase, impedance like theinfrequency, contrast regard toisthe robust against sensing line, the Let effects sensor us mention, to of electromagnetic design finish and fabrication the present interference section, and noise, are very simple. that Two the phase, at prototype like least as compared the frequency, sensors to were is the designed, robust against magnitudefabricated, (e.g., the the effects magnitude of electromagnetic of a transmission interference or a and reflection noise, at coefficient, leasta as compared typical and applied to the dielectric characterization of different MUT samples, as well as to the detection output to the magnitude variable in (e.g., microwave theof magnitude sensors). of a transmission In this regard, defects (consisting of drilled or a phase-variation reflection holes) in a reference coefficient, sensors, as those sample. a In one typical reported output variable in thisthe of the sensors, paper, maximum in microwave or frequency- sensitivity sensors). variation was found In to this sensors,be regard, are morephase-variation as high astolerant −101.3than◦ , but sensors, ascan coupling-modulation this value those reported sensors be further in this (or enhancedotherpaper, sensors by orusing cascadingfrequency- either further variation the ◦ magnitudesensors, of are the more tolerant reflection or than coupling-modulation transmission coefficient) sensors to 90 stages (with as much impedance contrast as possible) to the designed and fabricated structure. (or other electromagnetic sensors using interferences either and the noise. magnitude Concerning By redesigning of the thethe reflection effects sensor andof or transmission cross asensitivities adding coefficient) fluidic channel caused,on top to e.g.,electromagnetic of by the changes sensing line, interferences in temperature the applicationandor noise. humidity, Concerning of the structure the phase-variation effects of cross sensitivities and frequency-variation to the characterization of liquid samples, caused, sensors e.g., by are not astogood particularly changes in temperature as differential-mode the determination of small or humidity, sensors, variations or phase-variation sensors of solutebasedcontentonand frequency-variation symmetry in liquid properties. solutions, sensors is This are is because envisaged. In not as good changes particular, as differential-mode inthese environmental sensors may factors find sensors, occur application or sensors at scales based much larger in industrial on symmetry than thesuch processes, properties. typical size of as wine This is because sensors, and, fermentation and changes therefore, in environmental such changes medical diagnosis and are factors seen monitoring, as occur common-modeat scales much stimulus larger by than the typical differential-mode size of sensorssensors, e.g., the determination of variations of electrolyte content in urine or blood. or by and, sensors therefore, based such on changes symmetry are seen properties. as common-mode stimulus by differential-mode sensors or by sensors Nevertheless, the proposed phase-variation sensors exhibit relevant advantages, such as operation at based on symmetry properties. Author singleContributions: aNevertheless, the proposed frequency, P.C. and robustness J.M.-E. phase-variation againstcarried out the simulations sensors electromagnetic and experiments exhibitinterference relevant advantages, shown and noise, such in asthe paper. operation one-port deviceP.V., at L.S. and M.G. a single frequency, participated in the discussions. robustnessand, against F.M., P.C. and electromagnetic J.M.E. wrote the interference and paper. F.M. was the director noise, one-port device of the operation research. All (inauthors reflectionhavemode), read and agreed mosttoimportantly, the published highly versionachievable sensitivities. of the manuscript. operation (in reflection mode), and, most importantly, highly achievable sensitivities. Sensors 2020, 20, 6236 9 of 11 Funding: This work was supported by MICIIN-Spain (project TEC2016-75650-R and PID2019-103904RB-I00), by Generalitat de Catalunya (project 2017SGR-1159), by Institució Catalana de Recerca i Estudis Avançats (who awarded Ferran Martín), and by FEDER funds. J. Muñoz-Enano acknowledges Secreteraria d’Universitats i Recerca (Gen. Cat.) and European Social Fund for the FI grant. Paris Vélez acknowledges the Juan de la Cierva Program for supporting him through Project IJCI-2017-31339. Conflicts of Interest: The authors declare no conflict of interest. References 1. Mandel, C.; Kubina, B.; Schüßler, M.; Jakoby, R. Passive chipless wireless sensor for two-dimensional displacement measurement. In Proceedings of the 41st European Microwave Conference, Manchester, UK, 10–13 October 2011; pp. 79–82. 2. Puentes, M. Planar Metamaterial Based Microwave Sensor Arrays for Biomedical Analysis and Treatment; Springer: Heidelberg, Germany, 2014; ISBN 978-3319060408. 3. Ebrahimi, A.; Withayachumnankul, W.; Al-Sarawi, S.; Abbott, D. High-sensitivity metamaterial-inspired sensor for microfluidic dielectric characterization. IEEE Sens. J. 2013, 14, 1345–1351. [CrossRef] 4. Schüßler, M.; Mandel, C.; Puentes, M.; Jakoby, R. Metamaterial inspired microwave sensors. IEEE Microw. Mag. 2012, 13, 57–68. [CrossRef] 5. Boybay, M.S.; Ramahi, O.M. Material characterization using complementary split-ring resonators. IEEE Trans. Instrum. Meas. 2012, 61, 3039–3046. [CrossRef] 6. Lee, C.S.; Yang, C.L. Complementary split-ring resonators for measuring dielectric constants and loss tangents. IEEE Microw. Wirel. Compon. Lett. 2014, 24, 563–565. [CrossRef] 7. Yang, C.L.; Lee, C.S.; Chen, K.W.; Chen, K.Z. Noncontact measurement of complex permittivity and thickness by using planar resonators. IEEE Trans. Microw. Theory Tech. 2015, 64, 247–257. [CrossRef] 8. Withayachumnankul, W.; Jaruwongrungsee, K.; Tuantranont, A.; Fumeaux, C.; Abbott, D. Metamaterial-based microfluidic sensor for dielectric characterization. Sens. Actuators A Phys. 2013, 189, 233–237. [CrossRef] 9. Salim, A.; Lim, S. Complementary split-ring resonator-loaded microfluidic ethanol chemical sensor. Sensors 2016, 16, 1802. [CrossRef] 10. Su, L.; Mata-Contreras, J.; Vélez, P.; Fernández-Prieto, A.; Martín, F. Analytical method to estimate the complex permittivity of oil samples. Sensors 2018, 18, 984. [CrossRef] 11. Abdolrazzaghi, M.; Zarifi, M.H.; Daneshmand, M. Sensitivity Enhancement of Split Ring Resonator Based Liquid Sensors. In Proceedings of the 2016 IEEE SENSORS, Orlando, FL, USA, 30 October–3 November 2016. 12. Abdolrazzaghi, M.; Zarifi, M.H.; Pedrycz, W.; Daneshmand, M. Robust ultra-high resolution microwave planar sensor using fuzzy neural network approach. IEEE Sens. J. 2016, 17, 323–332. [CrossRef] 13. Zarifi, M.H.; Daneshmand, M. Monitoring solid particle deposition in lossy medium using planar resonator sensor. IEEE Sens. J. 2017, 17, 7981–7989. [CrossRef] 14. Zarifi, M.H.; Deif, S.; Abdolrazzaghi, M.; Chen, B.; Ramsawak, D.; Amyotte, M.; Vahabisani, N.; Hashisho, Z.; Chen, W.; Daneshmand, M. A microwave ring resonator sensor for early detection of breaches in pipeline coatings. IEEE Trans. Ind. Electron. 2017, 65, 1626–1635. [CrossRef] 15. Abdolrazzaghi, M.; Daneshmand, M.; Iyer, A.K. Strongly enhanced sensitivity in planar microwave sensors based on metamaterial coupling. IEEE Trans. Microw. Theory Tech. 2018, 66, 1843–1855. [CrossRef] 16. Zarifi, M.H.; Sadabadi, H.; Hejazi, S.H.; Daneshmand, M.; Sanati-Nezhad, A. Noncontact and nonintrusive microwave-microfluidic flow sensor for energy and biomedical engineering. Sci. Rep. 2018, 8, 139. [CrossRef] 17. Ebrahimi, A.; Scott, J.; Ghorbani, K. Ultrahigh-Sensitivity Microwave Sensor for Microfluidic Complex Permittivity Measurement. IEEE Trans. Microw. Theory Tech. 2019, 67, 4269–4277. [CrossRef] 18. Horestani, A.K.; Naqui, J.; Shaterian, Z.; Abbott, D.; Fumeaux, C.; Martín, F. Two-dimensional alignment and displacement sensor based on movable broadside-coupled split ring resonators. Sens. Actuators A Phys. 2014, 210, 18–24. [CrossRef] 19. Naqui, J.; Damm, C.; Wiens, A.; Jakoby, R.; Su, L.; Martín, F. Transmission lines loaded with pairs of magnetically coupled stepped impedance resonators (SIRs): Modeling and application to microwave sensors. In Proceedings of the IEEE MTT-S Int. Microwave Symposium, Tampa, FL, USA, 1–6 June 2014; pp. 1–4. 20. Su, L.; Naqui, J.; Mata-Contreras, J.; Martín, F. Modeling metamaterial transmission lines loaded with pairs of coupled split-ring resonators. IEEE Antennas Wirel. Propag. Lett. 2014, 14, 68–71. [CrossRef] Sensors 2020, 20, 6236 10 of 11 21. Su, L.; Naqui, J.; Mata, J.; Martín, F. Dual-band epsilon-negative (ENG) transmission line metamaterials based on microstrip lines loaded with pairs of coupled complementary split ring resonators (CSRRs): Modeling, analysis and applications. In Proceedings of the 2015 9th International Congress on Advanced Electromagnetic Materials in Microwaves and Optics (METAMATERIALS), Oxford, UK, 7–12 September 2015; pp. 298–300. 22. Su, L.; Naqui, J.; Mata-Contreras, J.; Vélez, P.; Martín, F. Transmission line metamaterials based on pairs of coupled split ring resonators (SRRs) and complementary split ring resonators (CSRR): A comparison to the light of the lumped element equivalent circuits. In Proceedings of the International Conference on Electromagnetics for Advanced Applications (ICEAA 2015), Torino, Italy, 7–11 September 2015; pp. 891–894. 23. Su, L.; Naqui, J.; Mata-Contreras, J.; Martín, F. Modeling and applications of metamaterial transmission lines loaded with pairs of coupled complementary split-ring resonators (CSRRs). IEEE Antennas Wirel. Propag. Lett. 2015, 15, 154–157. [CrossRef] 24. Naqui, J.; Damm, C.; Wiens, A.; Jakoby, R.; Su, L.; Mata-Contreras, J.; Martín, F. Transmission lines loaded with pairs of stepped impedance resonators: Modeling and application to differential permittivity measurements. IEEE Trans. Microw. Theory Tech. 2016, 64, 3864–3877. [CrossRef] 25. Su, L.; Mata-Contreras, J.; Vélez, P.; Martín, F. Splitter/combiner microstrip sections loaded with pairs of complementary split ring resonators (CSRRs): Modeling and optimization for differential sensing applications. IEEE Trans. Microw. Theory Tech. 2016, 64, 4362–4370. [CrossRef] 26. Vélez, P.; Su, L.; Grenier, K.; Mata-Contreras, J.; Dubuc, D.; Martín, F. Microwave microfluidic sensor based on a microstrip splitter/combiner configuration and split ring resonators (SRRs) for dielectric characterization of liquids. IEEE Sens. J. 2017, 17, 6589–6598. [CrossRef] 27. Ebrahimi, A.; Scott, J.; Ghorbani, K. Differential sensors using microstrip lines loaded with two split-ring resonators. IEEE Sens. J. 2018, 18, 5786–5793. [CrossRef] 28. Naqui, J.; Durán-Sindreu, M.; Martín, F. Novel sensors based on the symmetry properties of split ring resonators (SRRs). Sensors 2011, 11, 7545–7553. [CrossRef] [PubMed] 29. Naqui, J.; Durán-Sindreu, M.; Martín, F. On the symmetry properties of coplanar waveguides loaded with symmetric resonators: Analysis and potential applications. In Proceedings of the 2012 IEEE/MTT-S International Microwave Symposium Digest, Montreal, QC, Canada, 17–22 June 2012; pp. 1–3. 30. Naqui, J.; Durán-Sindreu, M.; Martín, F. Alignment and position sensors based on split ring resonators. Sensors 2012, 12, 11790–11797. [CrossRef] 31. Naqui, J.; Durán-Sindreu, M.; Martín, F. Transmission lines loaded with bisymmetric resonators and applications. In Proceedings of the IEEE MTT-S International Microwave Symposium Digest, Seattle, WA, USA, 2–7 June 2013; pp. 1–3. 32. Horestani, A.K.; Fumeaux, C.; Al-Sarawi, S.F.; Abbott, D. Displacement sensor based on diamond-shaped tapered split ring resonator. IEEE Sens. J. 2012, 13, 1153–1160. [CrossRef] 33. Horestani, A.K.; Abbott, D.; Fumeaux, C. Rotation sensor based on horn-shaped split ring resonator. IEEE Sens. J. 2013, 13, 3014–3015. [CrossRef] 34. Naqui, J.; Martı, F. Transmission lines loaded with bisymmetric resonators and their application to angular displacement and velocity sensors. IEEE Trans. Microw. Theory Tech. 2013, 61, 4700–4713. [CrossRef] 35. Ebrahimi, A.; Withayachumnankul, W.; Al-Sarawi, S.F.; Abbott, D. Metamaterial-inspired rotation sensor with wide dynamic range. IEEE Sens. J. 2014, 14, 2609–2614. [CrossRef] 36. Naqui, N.; Durán-Sindreu, M.; Martín, F. Transmission lines loaded with folded stepped impedance resonators (SIRs): modelling and applications. In Proceedings of the Sixth International Congress on Advanced Electromagnetic Materials in Microwaves and Optics (Metamaterials 2012), St. Petersburg, Russia, 17–22 September 2012. 37. Horestani, A.K.; Naqui, J.; Abbott, D.; Fumeaux, C.; Martín, F. Two-dimensional displacement and alignment sensor based on reflection coefficients of open microstrip lines loaded with split ring resonators. Electron. Lett. 2014, 50, 620–622. [CrossRef] 38. Naqui, J.; Martín, F. Angular displacement and velocity sensors based on electric-LC (ELC) loaded microstrip lines. IEEE Sens. J. 2013, 14, 939–940. [CrossRef] 39. Naqui, J.; Coromina, J.; Karami-Horestani, A.; Fumeaux, C.; Martín, F. Angular displacement and velocity sensors based on coplanar waveguides (CPWs) loaded with S-shaped split ring resonators (S-SRR). Sensors 2015, 15, 9628–9650. [CrossRef] Sensors 2020, 20, 6236 11 of 11 40. Naqui, J.; Martín, F. Application of broadside-coupled split ring resonator (BC-SRR) loaded transmission lines to the design of rotary encoders for space applications. In Proceedings of the IEEE MTT-S International Microwave Symposium, San Francisco, CA, USA, 22–27 May 2016; pp. 1–4. 41. Mata-Contreras, J.; Herrojo, C.; Martín, F. Application of split ring resonator (SRR) loaded transmission lines to the design of angular displacement and velocity sensors for space applications. IEEE Trans. Microw. Theory Tech. 2017, 65, 4450–4460. [CrossRef] 42. Mata-Contreras, J.; Herrojo, C.; Martín, F. Detecting the rotation direction in contactless angular velocity sensors implemented with rotors loaded with multiple chains of resonators. IEEE Sens. J. 2018, 18, 7055–7065. [CrossRef] 43. Damm, C.; Schüßler, M.; Puentes, M.; Maune, H.; Maasch, M.; Jakoby, R. Artificial transmission lines for high sensitive microwave sensors. In Proceedings of the IEEE Sensors, Christchurch, New Zealand, 25–28 October 2009; pp. 755–758. 44. Ferrández-Pastor, F.; García-Chamizo, J.; Nieto-Hidalgo, M. Electromagnetic differential measuring method: Application in microstrip sensors developing. Sensors 2017, 17, 1650. [CrossRef] [PubMed] 45. Muñoz-Enano, J.; Vélez, P.; Gil, M.; Martín, F. An Analytical Method to Implement High Sensitivity Transmission Line Differential Sensors for Dielectric Constant Measurements. IEEE Sens. J. 2019, 20, 178–184. [CrossRef] 46. Gil, M.; Vélez, P.; Aznar-Ballesta, F.; Muñoz-Enano, J.; Martín, F. Differential Sensor based on Electro-Inductive Wave (EIW) Transmission Lines for Dielectric Constant Measurements and Defect Detection. IEEE Trans. Antennas Propag. 2020, 68, 1876–1886. [CrossRef] 47. Muñoz-Enano, J.; Vélez, P.; Gil, M.; Mata-Contreras, J.; Martín, F. Differential-mode to common-mode conversion detector based on rat-race couplers: Analysis and application to microwave sensors and comparators. IEEE Trans. Microw. Theory Tech. 2020, 68, 1312–1325. 48. Coromina, J.; Muñoz-Enano, J.; Vélez, P.; Ebrahimi, A.; Scott, J.; Ghorbani, K.; Martín, F. Capacitively-Loaded Slow-Wave Transmission Lines for Sensitivity Improvement in Phase-Variation Permittivity Sensors. In Proceedings of the 50th European Microwave Conference, Utrecht, The Netherlands, 15–17 September 2020. 49. Muñoz-Enano, J.; Vélez, P.; Gil, M.; Martín, F. On the sensitivity of reflective-mode phase variation sensors based on open-ended stepped-impedance transmission lines: Theoretical analysis and experimental validation. IEEE Trans. Microw. Theory Tech. 2020. [CrossRef] 50. Su, L.; Muñoz-Enano, J.; Vélez, P.; Casacuberta, P.; Gil, M.; Martín, F. Highly sensitive phase variation sensors based on step-impedance coplanar waveguide (CPW) transmission lines for dielectric characterization. IEEE Sens. J. 2020. [CrossRef] 51. Ebrahimi, A.; Tovar-Lopez, F.; Scott, J.; Ghorbani, K. Differential microwave sensor for characterization of glycerole-water solutions. Sens. Actuators B Chem. 2020, 321, 128561. [CrossRef] 52. Pozar, D.M. Microwave Engineering, 4th ed.; John Wiley: Hoboken, NJ, USA, 2011. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).