- Email: [email protected]
Design of a metamaterial inspired omega shaped resonator based sensor for industrial implementations
Physica E 116 (2020) 113734
Contents lists available at ScienceDirect
Physica E: Low-dimensional Systems and Nanostructures journal homepage: http://www.elsevier.com/locate/physe
Design of a metamaterial inspired omega shaped resonator based sensor for industrial implementations Olcay Altıntas¸ a, b, *, Murat Aksoy b, Emin Ünal a a b
Iskenderun Technical University, Department of Electrical and Electronics Engineering, Iskenderun, Hatay, 31200, Turkey Cukurova University, Department of Electrical and Electronics Engineering, Saricam, Adana, 01330, Turkey
A R T I C L E I N F O
A B S T R A C T
Keywords: Metamaterial Dielectric Fuel adulteration Oil condition
In this study, the omega shaped resonator based sensor structure by inspiring of metamaterial (MTM) concept is presented at X-band frequency regime for industrial purpose both numerically and experimentally. Four dis tinguishing applications which are (i) authentic and inauthentic gasoline samples, (ii) authentic and inauthentic diesel samples, (iii) clean and waste lubricant samples and (iv) clean and waste transformer oil samples have been realized by determining dielectric properties of them. Although there are close dielectric constant values between the samples (about 0.65 for (i), 0.25 for (ii), 0.10 for (iii) and 0.15 for (iv)), the proposed sensor sensitively discriminates all sample groups with about a frequency shift of 350 MHz for (i), 180 MHz for (ii), 60 MHz for (iii) and 70 MHz for (iv). Sensor structure can be efficiently used many industrial applications since the results are sensitively and precisely obtained for proposed studies. In addition, the proposed sensor is du rable, suitable for real-time applications and long-term stability considering the fabrication technique and nondestructive measurement method.
1. Introduction Recently, the scientists focus on the high-tech sensors since the In dustry 4.0 concept needs the sensor devices which have real-time operated, without requiring human resources and more precise. More over, some sectors ask for the sensing mechanism which can be achieved more sensitive, durable and long term stability with non-destructive methods. Hence, the researchers have investigated the metamaterial (MTM) based sensor structures to obtained novel and improved sensing characteristics. The MTM concept has been a popular research area among the scientists at the last decade. It has non-natural properties which are negative refraction index, reverse Snell Law and reverse Doppler Effect etc. Many applications and studies have been realized by the researchers in the topics such as absorbers [1,2], energy harvesters [3–5], polarization converters [6,7], super lenses [8,9], antennas [10, 11] and sensors [12–29]. Lots of MTM based sensor studies in the literature have been carried out in different approaches. Albishi et al. proposed a highly sensitive MTM based sensor structure which can be detected the cracks in the metallic materials [12]. Gargari et al. studied a wireless MTM inspired rotation sensor. Their sensing mechanism has sub milliradian resolution
to measure elastic-region bending in materials such as steel [13]. Altintas et al. designed a circular split ring resonator for multipurpose sensing applications such as dielectric characterization of liquids, rota tion detection of rods and strain sensing [14]. Saghati et al. improved a transmission line integrated sensor structure for dielectric spectroscopy of liquid chemicals in a wideband frequency regime both numerically and experimentally [15]. Wu et al. implemented a rectangular complementary-ring resonator based sensor to monitor permittivity of liquids at a working frequency of 900 MHz [16]. He et al. realized a thin film sensing application by using a tip shaped split ring resonator (SRR) MTM at microwave frequency regime [17]. Ni et al. proposed an MTM absorber based sensor structure to observe humidity of environment [18]. Keshavarz et al. realized an ultrahigh sensitive temperature sensor structure based on graphene semiconductor MTM in terahertz (THz) frequency regime [19]. Tumkaya et al. designed a MTM based sensor structure at X-band to discriminate authentic and inauthentic fuel samples [20]. Bakir et al. proposed a microfluidic and fuel adulteration sensing mechanism by using chiral MTM based structure in microwaves [21]. Tamer et al. investigated a transmission line based sensor for determining authentic and inauthentic gasoline samples [22]. Shah marvandi et al. implemented a CMOS MTM absorber based gas sensing
* Corresponding author. Cukurova University, Department of Electrical and Electronics Engineering, Saricam, Adana, 01330, Turkey. E-mail address: [email protected] (O. Altıntas¸). https://doi.org/10.1016/j.physe.2019.113734 Received 8 July 2019; Received in revised form 29 August 2019; Accepted 21 September 2019 Available online 27 September 2019 1386-9477/© 2019 Elsevier B.V. All rights reserved.
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
[23]. Rawat et al. studied a sensor structure which can achieve to sense the toxic and hazardous materials for human health [24]. Sethi et al. improved a blood glucose sensing mechanism by using a photonics based MTM [25]. Keshavarz et al. presented a novel water-based ter ahertz MTM and a semiconductor film to detect skin cancer [26]. Altintas et al. designed a meander line based sensor structure for permittivity characterization by using dielectric materials with the flat surface [27]. The proposed resonator structure has been inspired by the omega shaped metamaterial structures with strong resonance property which are highly used among the researchers. Aydin et al. suggested that the band gap of a single omega unit cell is due to electric and magnetic resonance since the omega type structures has bianisotropic properties [30]. Lheurette et al. discussed the left-handedness of the periodic omega type resonators at X and Ku band regime [31]. Li et al. studied different strong resonance conditions in transmission spectra with planar metamaterials with omega shaped metallic inclusions [32]. Basiry et al. investigated that the electromagnetic performance of the omega type metamaterials for planar radome applications [33]. Bal makou et al. present ground-plane-less bidirectional absorber applica tions with metamaterial based omega resonator structure at terahertz regime [34]. Labidi et al. designed a multi-band bowtie antenna based on omega-shaped resonator to improve return loss and radiation pattern parameters [35]. Many studies had been realized by researchers about microwave sensors with different kinds of resonator geometry for liquid detection applications. Gordon et al. proposed metasurface consisting of metallic electric-field coupled resonators operating in S band for chemicals [36]. Abduljabar et al. used a double split ring resonator to determine the dielectric properties of common liquids at around 3 GHz [37]. Miyamaru et al. designed a rectangular complementary split ring resonator to achieve sensitive sensing application for glycerol-water mixtures in terahertz regime [38]. Sadeqi et al. presented metamaterial on low cost and ubiquitous paper substrate by implementing an array of the circular disk for detection of dielectric properties of liquids [39]. Chen et al. realized terahertz sensing of highly absorptive water-methanol mixtures by using asymmetric dual wire resonator [40]. Lin achieved volatile organic solutions sensing application with complementary U-shaped resonator at terahertz regime [41]. Studies about MTM absorber based sensor applications had been achieved by the scientists. A new type dual band terahertz MTM absorber with a patterned metallic strip and a dielectric layer on a metallic ground plane is investigated for biological monitoring and sensing application [45]. A triple-band perfect absorber is realized by employing multiple different-sized metallic patterns and its sensing performance is examined in terms of the surrounding index [46]. Quad-band terahertz MTM absorber based on a common sandwich structure model is examined by focusing LC resonance. The structure has the ability to tune the frequencies of the absorption peaks and it has potential application perspectives in imaging and sensing [47]. Pressure, density, and volumetric moisture sensing applications have been ach ieved by using composed of a split ring resonator topology in microwave regime [48]. Swastika shaped MTM absorber structure is investigated in X-band and the sensing ability of the structure is tested by using some chemical liquids such as acetone, methanol, ethanol, polyethylene gly col [49]. In this paper, a metamaterial inspired omega shaped resonator based sensor structure is proposed for industrial implementations which are detections of fuel adulteration, lubricant condition and transformer oil condition. The samples studied in this paper are highly used in industry and they have very important role while utilizing from the industrial devices in energizing and lubrication phases. In addition, conditions and qualities of these materials is crucial issue to obtain maximum efficiency and minimum failure from the machines. The study is realized at X-band frequency regime both numerically and experimentally and the pro posed structure is by considering the proper waveguide sizes. CST
Microwave Studio simulation software is used in the numerical studies to observe the equivalent circuit diagram, electric field and surface current distribution, impedance matching and parametrical analysis of the sensor structure. Moreover, the measured dielectric characteristics of the sample by using a coaxial dielectric probe kit has imported to the simulation program and the sensor has been then numerically tested. After that, the experimental studies have been achieved by means of the PNA-L Agilent vector network analyzer (VNA) which has the operating frequency band between 10 MHz and 43.5 GHz. An X-band waveguide, two WR90 waveguide adapter and a proper sample holder are used during the measurements of the samples. Finally, simulated and measured results are compared and evaluated. The proposed sensor structure is a strong candidate for industrial implementations since it is sensitive, durable and can be used in real time applications. The novelty of the study is that four applications in one MTM based sensor structure are successfully tested comparing the other sensor studies in literature given above. The sensor structure is clearly discriminate the samples with respect to very close dielectric constant and loss values in X band. The study is more sensitively detect the liquid samples considering the other sensor structures which are operating frequency of X band. In addition, the MTM based sensor highly satisfy in terms of high Q-factor especially as experimental. 2. Design and theory The design and numerical studies of omega shaped resonator (OSR) are achieved by using a full wave electromagnetic solver finite inte gration technique (FIT) based CST Microwave Studio simulation soft ware. The OSR-based sensor structure has been designed at the X-band frequency regime. The overall size of the structure has 22.86 mm of width and 10.16 mm of length as illustrated in Fig. 1. This size has ar ranged by considering an X band waveguide to realize experimental work in the waveguide measurement method. The proposed structure has FR4 type dielectric substrate with a 1.6 mm of thickness, 4.3 of dielectric constant and 0.02 loss tangent values. The OSR consists of the copper type metal having a conductivity of 5.80001 � 107 S/m and thickness of 35 μm and it has been placed both front and back side of the FR4 substrate. The OSR can be defined by the parameters which are
Fig. 1. (a) The parametric dimensions of the OSR and (b) perspective view of the sensor structure. 2
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
internal radius (r), line width (w), gap width (g) and the horizontal arm line (Hline) as shown in Fig. 1(a). The values of these parameters are 2.2 mm, 0.8 mm, 0.8 mm and 13 mm, respectively. Sensor layer has been determined as the back side of the sensor structure with the thickness of 10 mm as shown in Fig. 1(b). This thickness value corresponds to the depth of the sample holder in the experimental studies. The equivalent circuit diagram has been analyzed to understand the working and the physical mechanism of the proposed OSR as shown in Fig. 2. The OSR can be divided into three parts such as ring, horizontal arms and gap demonstrated in Fig. 2(a). Basically, ring and horizontal arms have created inductive effects which can be defined as Lring and Larm respectively. The gap illustrated as Cgap creates a capacitive effect. Hence, the equivalent circuit diagram of the OSR has been obtained as an RLC circuit shown in Fig. 2(b) with these circuit parameters. Besides these parameters, the sensor layer which can be varied with the different samples having different electrical properties also creates a capacitive effect as Csens . Thus, the total capacitance Ct of the system can be written as; Ct ¼ C0 þ Cgap þ εsamp Csens
working principle for the engineering perspective and approach. The calculation of circuit parameters is another topic for a comprehensive project as a future study. Because, this calculation is highly affected by many properties of the incident wave such as angle, direction, polarization. Furthermore, the equivalent impedance of the structure can be rep resented as; � (4) Zeq ðωÞ ¼ Rt þ jωLt þ 1 jωCt where the Rt is overall resistance. The impedance of the free space; qffiffiffiffiffiffiffiffiffiffiffi Z0 ¼ μ0 =ε0 ¼ 377Ω (5) The reflection calculation between these two mediums can be writ ten as; �� � RðωÞ ¼ Zeq ðωÞ Z0 Zeq ðωÞ þ Z0 (6) The impedance matching between the proposed structure and free space is numerically studied as demonstrated in Fig. 3. Considering Eq (5), while the impedance of the structure Zeq ðωÞ approaches the impedance of the free space Z0 , the reflection RðωÞ is approachesto zero. This relation can be seen in Fig. 3. The refection magnitude is going to zero at the frequency of 11.70 GHz while the impedance of the structure is at maximum. Moreover, the imaginary part of the impedance of the structure is also zero and the transmission magnitude of the signal is at the maximum level at this frequency point. The relation between the impedance characteristics and maximum transmission ratio for the resonator has been explained Fig. 3. It can be concluded that the proposed resonator structure has the resonance fre quency point at X band with respect to the maximum transmission ratio. On the other side, the study is focused on zero transmission ratios since the sensor structure has been sensitively discriminated of the samples at X band. Hence the electric field and surface current distribution of the sensor is numerically presented at operating frequency point of 9.85 GHz with zero transmission coefficient in Fig. 4. The electric field is highly concentrated around the resonator part of the structure. Espe cially, it is located at the end of the horizontal arms and capacitive gap of the omega shaped resonator as shown in Fig. 4(a). These concentrations can describe how the capacitive effects occur in the resonator part of the sensor structure. The surface current with opposite directions is driven by capacitive effects. It has occurred at both left and right horizontal inductive arm of the resonator with the opposite directions as shown in Fig. 4(b). Hence, the zero transmission coefficient is obtained at the working frequency point of 9.85 GHz with these opposite surface currents.
(1)
where C0 is the capacitive effects which are originated from the sur rounding space and the εsamp Csens describes the capacitive effect of the sensor layer. The permittivity of the sample loaded to the sensor layer can be expressed as εsamp ¼ ε’samp þ jε’’samp . Hence the total capacitance ðCt Þ of the proposed structure is a function of the dielectric constant ðε’samp Þ and dielectric loss ðjε’’samp Þ of the sample placed in the sensor layer. � � (2) Ct ¼ Fc ε’samp ; ε’’samp The resonance frequency f0 of the structure can be written as; . � pffiffiffiffiffiffiffiffi � f0 ¼ 1 2π Ct Lt
(3)
where the Lt total inductance value of the OSR. Hence, the sensing characteristics of the structure can be easily monitored by the resonance frequency point shifts. This diagram presentation of the omega-shaped resonator as shown in Fig. 2(b) has been carefully prepared due to understanding the
3. Numerical and experimental studies The numerical setup has been prepared by using a full wave solver CST Microwave Simulation software and it is demonstrated in Fig. 5. The omega shaped resonator parts are placed in the front and back side of the dielectric substrate. The sensor layer with a thickness of 10 mm which correspond to experimental conditions has been arranged along the z direction. Two waveguide port is placed to the front and back side of the structure to monitor the transmission coefficient (S12). The reference planes are set with respect to both end plane of the structure. The fre quency range has arranged between 8 GHz–12 GHz which corresponds to X band. The boundary conditions are set to the perfect electric conductor (PEC) along x- and y-axis which corresponds to waveguide boundary conditions as shown in Fig. 5. The z-axis has determined as propagation direction. The parametric studies have been realized to observe the effects of the dimensions of the resonator on the resonance frequency of the structure as demonstrated in Fig. 6. The dimension parameters of gap width (g), line width (w) and horizontal arm length (Hline) are
Fig. 2. (a) Parts of the proposed sensor structure and (b) equivalent circuit diagram of OSR. 3
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
Fig. 3. Normalized real impedance, imaginary impedance, reflection magnitude and transmission magnitude of the proposed structure.
Fig. 4. (a) Electric field and (b) surface current distribution of the proposed sensor at 9.85 GHz.
investigated by varying from certain value ranges. The effect of the gap width parameter (g) is investigated between 0.7 mm and 0.9 mm by increasing 0.1 mm in Fig. 6(a). The resonance frequency is decreasing while the gap width is increased. The variation is linearly observed about with 20 MHz decrements. The resonance frequency point is highly affected by the line width parameter (w). This parameter is changed from 0.2 mm to 1.0 mm by increasing 0.2 mm point. This time, the line width parameter increased while the resonance frequency is increased too. This increment has linearly changed about with 180 MHz. Lastly, the horizontal arm length (Hline) parameter has been analyzed between 12.0 mm and 14.0 mm by increased 0.5 mm. The variation of the reso nance frequency has monitored as about 160 of decrements for each step. It can be concluded that the resonance frequency of the structure is highly affected by the resonator dimensions. The proposed structure is manufactured by using a CNC based LPKF E33 PCB prototyping machine. The fabricated sensor structure is illus trated in Fig. 7(a). Experimental studies are realized via an X-band
Fig. 5. Numerical study setup for proposed sensor structure.
4
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
placed to the front side of the waveguide. The fabricated sensor structure has embedded to the waveguide adapter coincided with the sample holder as shown in Fig. 7(b). After the configuration the waveguide has connected to PNA-L Agilent vector network analyzer (VNA) with a fre quency range between 10 MHz and 43.5 GHz as shown in Fig. 7(c). Before the measurements, the frequency range is arranged between 8 GHz and 12 GHz. The VNA has then calibrated with the way of special open circuit, short circuit and load connectors for both two ports. In addition, an open ended coaxial probe is used to determine the electrical properties of the liquid samples. The coaxial probe is connected to VNA and is calibrated by the way of air, water at 25� and special short circuit calibration kit. The dielectric characteristics are then measured by immerging the Keysight 85070 dielectric probe to the liquid samples. The dielectric values of the samples are calculated via the Keysight 85070 software embedded the PNA-L Agilent VNA. After that the sensing measurement setup are prepared and the transmission coeffi cient (S12) are measured with the waveguide, adapters, sensor structure and liquid sample holder as shown in Fig. 7. The liquid sample is firstly loaded by an injector via the channel placed in the top side inside the sample holder as shown in Fig. 7(c). After that, the transmission coef ficient is measured and in the same way, the sample is discharged from the holder. This method is applied for each measurement steps of the liquid samples. 3.1. Gasoline sample distinguishing study Gasoline is one of the highly used energy sources by many countries. The producer companies or governments are added to their own addi tives to mark the fuels. However, the unmarked gasoline fuels are sold by illegal ways as tax free. Hence, the purpose of this study is to solve this problem by distinguishing the authentic and inauthentic gasoline samples from each other. The electrical properties of the authentic and inauthentic samples are obtained between the 8 GHz and 12 GHz as illustrated in Fig. 8. Clear discrimination has been monitored between them. The dielectric con stant values of the authentic gasoline samples and inauthentic gasoline samples are about 2.4 and 1.75, respectively. The dielectric loss values of them are about 0.45 and 0.40, respectively. The differences in the dielectric characteristics of these two samples are clearly seen in Fig. 8. After the dielectric property determining study, the samples are defined in the CST Microwave simulation software as new materials. The numerical studies are realized by placing these materials into the sensor layer. The transmission coefficients for authentic and inauthentic gas oline samples are monitored between 8.5 GHz and 10.5 GHz as shown in Fig. 9(a). The resonance frequencies for authentic and inauthentic samples are 9.364 GHz and 9.772 GHz, respectively. About 408 MHz of frequency shift is obtained between these two resonance frequency points. The measured results for the authentic and inauthentic gasoline samples is shown in Fig. 9. The measured resonance frequency points are
Fig. 6. The parametric study plots for OSR structure; (a) gap width parameter (g), (b) line width parameter (w) and (c) horizontal arm length param eter (Hline).
waveguide, two WR90 waveguide adapter and a sample holder as shown in Fig. 7 (b). The sample holder having 10 mm of depth dimension is loaded with the liquid sample at every measurement phase and it is
Fig. 7. (a) Fabricated sensor structure, (b) waveguide configuration and (c) experimental setup. 5
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
Fig. 8. Dielectric constant and dielectric loss values of the authentic and inauthentic gasoline samples at X-band.
gasoline samples from each other. 3.2. Diesel sample distinguishing study The other highly preferred fossil fuel as an energy source is diesel. This fuel (like gasoline) is an important commercial tool for interna tional companies. To detection the inauthentic diesel is a crucial duty for the governments. This unbranded fuels also poses a great danger for the users with respect to engine healthiness. Hence, the authentic and inauthentic diesel sample distinguishing study is realized by using the proposed sensor structure. The dielectric characteristics of the authentic and inauthentic diesel samples are measured the X-band frequency regime by using a dielectric probe kit as shown in Fig. 10. At the frequency range of 8 GHz–12 GHz, the dielectric constant value for authentic diesel sample is between 2.79 and 2.52, and for inauthentic diesel sample is between 2.46 and 2.18. The dielectric loss values are almost at the same values for the samples as illustrated in Fig. 10. The numerical and experimental studies to distinguish the authentic and inauthentic diesel samples are presented in Fig. 11. The results are plotted at between the 8.5 GHz and 10.5 GHz. In the numerical studies, the lowest transmission coefficients are observed at the frequency points of 9.119 GHz and 9.311 GHz for authentic and inauthentic diesel sam ples, respectively as shown in Fig. 11(a). As to experimental studies, mentioned lowest transmissions are monitored at the frequency points of 9.496 GHz and 9.669 GHz, respectively as shown in Fig. 11(b). The frequency difference between the authentic and inauthentic diesel samples is 192 MHz in numerical and 173 MHz in experimental. The numerical and experimental studies of the sensor structure for diesel samples are given almost the same response except about the resonance frequency shift of 350 MHz. Hence, it can be concluded that the pro posed sensor structure can be sensitively distinguished the authentic and inauthentic diesel samples from each other.
Fig. 9. (a) Numerical and (b) experimental study plots for authentic and inauthentic gasoline samples.
monitored at 9.486 GHz and 9.847 GHz, respectively. The resonance frequency difference between these two frequency points is 361 MHz experimentally. It can be seen from Fig. 9 that the simulated and measured results are in a good agreement and the proposed sensor structure has the ability to discriminating the authentic and inauthentic
Fig. 10. Dielectric constant and dielectric loss values of the authentic and inauthentic diesel samples at X-band. 6
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
Fig. 11. (a) Numerical and (b) experimental study plots for authentic and inauthentic diesel samples.
Fig. 13. (a) Numerical and (b) experimental study plots for clean and waste lubricant samples.
3.3. Lubricant sample condition detection study
for clean and waste oil, respectively. The lowest level of the transmission ratio in the experimental study is obtained at 9.502 GHz and 9.441 GHz for clean and waste oil, respectively. The frequency shifts are obtained as 72 MHz in simulation and 61 MHz in experimental. Although there is 200 MHz of frequency shift, the numerical and experimental results are in a good agreement such a detection study with very close electrical characteristics of the samples. Therefore, it can be said that the sensor structure is able to discriminate the clean and waste lubricant sample from each other, precisely and sensitively.
In this study, the lubricant samples which are used in the mecha nisms such as engine systems, robotic machines, mechanics etc. have been investigated. It is very important to detect the machine or engine failures originated from the lube oil for safety and cost. Hence, this section of the study has been focused on the discrimination between the clean and waste lubricant sample. The lubricant samples have taken from the machines which are operated in hydraulic systems. The dielectric properties of the clean and waste lubricant samples are measured at X-band by means of the dielectric probe as shown in Fig. 12. The dielectric constant values and dielectric loss values for the clean and waste lubricant samples have been observed very closely. The dielectric constant values are between about 2.5 and 2.2. Dielectric loss values are between about 0.5 and 0.7 at X-band. It is hard to discriminate the clean lubricant sample from the waste oil since the dielectric properties of them is almost the same. However, the proposed sensor structure has precisely detected of the condition of the oils. The numerical and experimental studies of the lubricant sample condition sensing are plotted in Fig. 13. In the numerical studies. The minimum level of transmission is occurred at 9.257 GHz and 9.185 GHz
3.4. Transformer oil condition detection study The last study is carried out for detection of the transformer oil condition at X-band both numerically and experimentally. Transformers have a key role in the industry for the utilization of electrical energy and controlling the energy flow. Thus, protection of the transformers is a very important topic for health and safety management and system failures. In some type of transformers, special transformer oils are used for protection. Hence, this part of the study is focused on the transformer oil condition detection with clean and waste oils. The dielectric properties
Fig. 12. Dielectric constant and dielectric loss values of the clean and waste lubricant samples at X-band. 7
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
of the clean and waste transformer oil samples are measured at X-band regime as illustrated in Fig. 14. The dielectric constant of the clean and waste oil samples are close to each other and they are about 2.9 and 2.7. The dielectric loss values of them are monitored at almost the same value which is about between 0.5 and 0.8. Hence, the dielectric constant parameter is the key to detect the condition of the transformer oil. The numerical and experimental results for clean and waste trans former oils are plotted in Fig. 15 between the frequencies of 8.5 GHz and 10.5 GHz. In the numerical studies, the lowest transmission coefficient values at this frequency range have been observed at the frequency points of 9.026 GHz and 9.103 GHz for clean and waste oil samples, respectively. The difference between these frequency points is 77 MHz as shown in Fig. 15(a). In the experimental studies, these values have been monitored at the frequency points of 9.381 GHz and 9.446 GHz for clean and waste oil samples, respectively. The frequency shift between them is 65 MHz as demonstrated in Fig. 15(b). The similar frequency shift (about 350 MHz) between the numerical and experimental results has occurred in the transformer oil condition study as well. However, the results are in a good agreement considering the frequency difference between the clean and waste oil. It can be concluded that the proposed structure can determine the transformer oil condition sensitively. 4. Results and discussions The operating frequency range of the study has been carefully chosen since the most important issue is to monitor sufficient differentiation between the samples. The X band frequency regime gives the precise results to observe the difference sensitively with the sensor structure. Moreover, this higher frequency regime provides to get compact size sensor structure. Although the sensor including waveguide and adapters is expensive as price, the sensing implementation results for the four different study is very sufficient, sensitive and precise. Some of these applications are about the adulteration to prevent the smuggling and to save a huge amount of money. Some of them are about the improvement machine/transformer safety of systems and decreasing the costs by detecting failures instantly. Hence, the cost of the measurement setup with waveguide and the price of the sensor structure can be negligible. In the experimental process, transmission coefficients for every liquid sample have been measured by 10 times with the proposed sensor structure. The observed results remain constant. Hence it can be said that the sensor structure has high durability with substrate material having high flexural strength, thermal strength, and thermal stress. Moreover, proposed sensor sensitively distinguishes the samples from each other considering the other studies in the literature. In Ref. [20], the sensor structure provided about 100 MHz frequency shift between the authentic and inauthentic diesel samples. In the same study, about 50 MHz frequency shift occurred between authentic and inauthentic gasoline samples. In Ref. [22], the transmission line based sensor structure experimentally discriminated authentic and inauthentic diesel
Fig. 15. (a) Numerical and (b) experimental study plots for clean and waste transformer oil samples.
samples with 50 MHz of frequency shift considering the transmission coefficient. In Ref. [44], the researcher proposed a high sensitive sensor structure which can detect the fuel adulteration by 27 MHz of frequency shift in resonance frequency. A similar study with kerosene had been realized in Ref. [21] with 32 MHz of resonance frequency shift. In Ref. [42], the researchers presented a transmission line based sensor structure to detect transformer oil conditions by observing 40 MHz of resonance frequency shift between clean and waste transformer oil. In this study, the proposed sensor provides 350 MHz of frequency shift for authentic and inauthentic gasoline samples, 180 MHz of frequency shift for authentic and inauthentic diesel samples, 60 MHz of frequency shift for clean and waste lubricant samples and 70 MHz of frequency shift for clean and waste transformer oil samples. Hence it can also be concluded that the proposed sensor structure has high sensitivity ratio. The difference between two liquid samples for four applications is directly proportional with the dielectric constant value. The dielectric constant and resonance frequency differences are about 0.65 and 361 MHz, 0.27 and 173 MHz, 0.12 and 61 MHz, 0.13 and 65 MHz for gasoline, diesel, lubricant and transformer oil samples, respectively. It
Fig. 14. Dielectric constant and dielectric loss values of the clean and waste transformer oil samples at X-band. 8
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
can be seen from these values that some differences exist in sensitivity. However, these sensitivity differences are originated from the irregu larity of the dielectric loss values of the samples. Although the dielectric constant values differences are the same for lubricant and transformer oil samples, the dielectric loss values differences are about 0.05 and zero for these samples, respectively as shown in Figs. 12 and 14. Hence the sensitivity difference is observed as about 4 MHz. The same sensitivity issue can be observed between results for gasoline and diesel samples illustrated in Figs. 9 and 11. The direct relations can be observed be tween the dielectric constant difference and frequency shifts. However, the dielectric loss values of them play key role by determining the sensitivity. Although the simulations and experiments share similar resonance frequencies, their waveforms are different from each other. It can be seen that the quality factor (Q-factor) of the experimental results is higher than the simulated ones. The difference is originated from the waveguide calibration errors, probe loss and probe connection mis matches. This causes lower transmission ratio at resonance frequencies in experimental process. The reason why the omega-shaped resonator is chosen is that it can response both electric and magnetic component of the incident elec tromagnetic wave [30]. Since the split ring resonator based sensor structures only response to the magnetic component of the electro magnetic waves and sense the changes occurring in the magnetic field, they cannot be sufficient in some applications. On the other hand, the omega shaped resonator based sensor structure has the capability for more sensitive observations due to response both component of elec tromagnetic wave although it is polarization dependent like split ring resonator. In addition, the proposed sensor structure have compact size considering the operating frequency range, which can be considered as an advantageous for the implementations. Another advantageous side of the study is that the sensor promises a multi-functional sensing appli cation which can distinguish the four different samples groups. In literature, the diesel/gasoline adulteration sensing study and trans former oil condition study has been realized by different sensor struc tures [20–22,42,43].
Acknowledgment The authors would like to acknowledge the Scientific Project Unit of Cukurova University (FDK-2018-10488). References [1] H. Zhu, F. Yi, E. Cubukcu, Nat. Photonics 10 (11) (2016) 709. [2] B.X. Wang, G.Z. Wang, J. Mater. Sci. Mater. Electron. 28 (12) (2017) 8487. [3] H. Wang, V.P. Sivan, A. Mitchell, G. Rosengarten, P. Phelan, L. Wang, Sol. Energy Mater. Sol. Cells 137 (2015) 235. [4] M. Karaaslan, M. Ba� gmancı, E. Ünal, O. Akgol, O. Altıntas¸, C. Sabah, Opt. Quant. Electron. 50 (5) (2018) 225. [5] E. Unal, F. Dincer, E. Tetik, M. Karaaslan, M. Bakir, C. Sabah, J. Mater. Sci. Mater. Electron. 26 (12) (2015) 9735. [6] O. Akgol, O. Altintas, E. Unal, M. Karaaslan, F. Karadag, Int. J. Microwave Wireless Technol. 10 (1) (2018) 133. [7] T. Zhang, L. Huang, X. Li, J. Liu, Y. Wang, J. Phys. D Appl. Phys. 50 (45) (2017) 454001. [8] N. Kaina, F. Lemoult, M. Fink, G. Lerosey, Nature 525 (7567) (2015) 77. [9] N. Fang, X. Zhang, in: Proceedings of the 2nd IEEE Conference on Nanotechnology, vol. 225, 2002. [10] J.K. Ji, G.H. Kim, W.M. Seong, IEEE Antennas Wirel. Propag. Lett. 9 (2010) 36. [11] S.A. Rezaeieh, M.A. Antoniades, A.M. Abbosh, IEEE Antennas Wirel. Propag. Lett. 15 (2016) 1893. [12] A.M. Albishi, O.M. Ramahi, IEEE Trans. Microw. Theory Tech. 65 (5) (2017) 1864. [13] A.M. Gargari, B. Ozbey, H.V. Demir, A. Altintas, U. Albostan, O. Kurc, V.B. Ertürk, IEEE Sens. J. 18 (11) (2018) 4482. [14] O. Altintas, M. Aksoy, O. Akgol, E. Unal, M. Karaaslan, C. Sabah, J. Electrochem. Soc. 164 (12) (2017) B567. [15] A.P. Saghati, J.S. Batra, J. Kameoka, K. Entesari, IEEE Trans. Microw. Theory Tech. 65 (7) (2017) 2558. [16] J. Wu, P. Wang, X. Huang, F. Rao, X. Chen, Z. Shen, H. Yang, Sens. Actuators A Phys. 280 (2018) 222. [17] X.J. He, Y. Wang, J.M. Wang, T.L. Gui, Microsyst. Technol. 16 (10) (2010) 1735. [18] B. Ni, Z.Y. Wang, R.S. Zhao, X.Y. Ma, Z.Q. Xing, L.S. Yang, L.J. Huang, Y.Y. Lin, D. B. Zhang, Opt. Quant. Electron. 49 (1) (2017) 33. [19] A. Keshavarz, A. Zakery, Appl. Phys. A 123 (12) (2017) 797. [20] M.A. Tümkaya, F. Dinçer, M. Karaaslan, C. Sabah, J. Electron. Mater. 46 (8) (2017) 495. € Alkurt, O. Altıntas¸, S. Dalgac, [21] M. Bakır, M. Karaaslan, E. Unal, F. Karadag, F.O. C. Sabah, J. Electrochem. Soc. 165 (11) (2018) B475. [22] A. Tamer, F.O. Alkurt, O. Altintas, M. Karaaslan, E. Unal, O. Akgol, F. Karadag, C. Sabah, J. Electrochem. Soc. 165 (7) (2018) B251. [23] E.K. Shahmarvandi, M. Ghaderi, P. Ayerden, G. de Graaf, R.F. Wolffenbuttel, Procedia Eng. 168 (2016) 1241. [24] V. Rawat, R. Kitture, D. Kumari, H. Rajesh, S. Banerjee, S.N. Kale, J. Magn. Magn. Mater. 415 (2016) 77. [25] K.K. Sethi, G. Palai, P. Sarkar, Optik 168 (2018) 296. [26] A. Keshavarz, Z. Vafapour, IEEE Sens. J. 19 (4) (2019) 1519. [27] O. Altintas, M. Aksoy, E. Unal, F. Karakasli, M. Karaaslan, J. Electron. Mater. 47 (10) (2018) 6185. [28] A. Vivek, K. Shambavi, Z.C. Alex, Sensor review. https://doi.org/10.1108/SR-062018-0152, 2018. [29] A. Salim, S. Lim, Biosens. Bioelectron. 117 (2018) 398. [30] K. Aydin, Z. Li, S. Bilge, E. Ozbay, Photon. Nanostruct. Fund. Appl. 6 (1) (2008) 116. [31] E. Lheurette, G. Houzet, J. Carbonell, F. Zhang, O. Vanbesien, D. Lippens, IEEE Trans. Antennas Propag. 56 (11) (2008) 3462. [32] Z. Li, K. Aydin, E. Ozbay, Opt. Commun. 283 (12) (2010) 2547. [33] R. Basiry, H. Abiri, A. Yahaghi, Int. J. RF Microw. Computer-Aided Eng. 21 (6) (2011) 665. [34] A. Balmakou, M. Podalov, S. Khakhomov, D. Stavenga, I. Semchenko, Opt. Lett. 40 (9) (2015) 2084. [35] M. Labidi, R. Salhi, F. Choubani, Appl. Phys. A 123 (5) (2017) 313. [36] J.A. Gordon, C.L. Holloway, J. Booth, S. Kim, Y. Wang, J. Baker-Jarvis, D. R. Novotny, Phys. Rev. B 83 (20) (2011) 205130. [37] A.A. Abduljabar, D.J. Rowe, A. Porch, D.A. Barrow, IEEE Trans. Microw. Theory Tech. 62 (3) (2014) 679. [38] F. Miyamaru, K. Hattori, K. Shiraga, S. Kawashima, S. Suga, T. Nishida, M. W. Takeda, Y. Ogawa, J. Infrared, Millim. Terahertz Waves 35 (2) (2014) 198. [39] A. Sadeqi, H.R. Nejad, S. Sonkusale, Opt. Express 25 (14) (2017) 16092. [40] M. Chen, L. Singh, N. Xu, R. Singh, W. Zhang, L. Xie, Opt. Express 25 (13) (2017) 14089. [41] Y.S. Lin, Mater. Lett. 195 (55) (2017). [42] O. Altintas¸, M. Aksoy, E. Ünal, M. Karaaslan, J. Electrochem. Soc. 166 (6) (2019) B482. [43] M.A. Tümkaya, M. Karaaslan, C. Sabah, Chin. J. Phys. 56 (5) (2018) 1872. [44] V. Rawat, V. Nadkarni, S.N. Kale, Def. Sci. J. 66 (4) (2016) 421. [45] B.X. Wang, X. Zhaia, G.Z. Wang, W.Q. Huang, L.L. Wang, J. Appl. Phys. 117 (2015), 014504. [46] B.X. Wang, G.Z. Wang, T. Sang, J. Phys. D 49 (2016) 165307.
5. Conclusion In this study, a metamaterial inspired sensor structure has been presented for different industrial applications both numerically and experimentally. The electrical characteristics of the samples are firstly investigated at X-band frequency regime by using a coaxial type dielectric probe kit. The numerical studies are then realized for gasoline, diesel, lubricant and transformer oil samples. After that the proposed structure is manufactured and experimental analyses are carried out by PNA-L Agilent vector network analyzer and proper waveguide setup. Finally, numerical and experimental results are compared. The proposed sensor structure is able to clearly distinguish the inauthentic and authentic gasoline samples from each other since their dielectric values are highly different. The other inauthentic fuel detec tion study has been achieved by the diesel samples. The sensor structure has successfully discriminated the inauthentic diesel sample from authentic diesel sample. The last two studies are about lubricant and transformer oil condition detection. Dielectric properties of the clean and waste lubricant samples is observed as very similar. However, the sensor structure precisely exhibits the difference between the clean and waste samples with the frequency shift of 61 MHz in experimental. Accordingly, the dielectric values of the clean and waste transformer oil is very close to each other. The dielectric constant difference between these samples is about 0.15. The proposed sensor structure distinguishes sensitively the waste transformer oil sample from the clean sample with the 65 MHz in experimental. In conclusion, the sensor structure can be efficiently used in the industrial implementations at the X-band fre quency regime. Moreover, the structure is low cost, durable and reconfigurable at any desired frequency range.
9
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
[47] B.X. Wang, IEEE J. Sel. Top. Quantum Electron. 23 (2017) 4700107. [48] M. Bakir, M. Karaaslan, O. Akgol, O. Altintas, E. Unal, C. Sabah, Optik 168 (2018) 741.
[49] Y.I. Abdulkarim, L. Deng, O. Altintas, E. Unal, M. Karaaslan, Phys. E Low-dimens. Syst. Nanostruct. 114 (2019) 113593.
10
Contents lists available at ScienceDirect
Physica E: Low-dimensional Systems and Nanostructures journal homepage: http://www.elsevier.com/locate/physe
Design of a metamaterial inspired omega shaped resonator based sensor for industrial implementations Olcay Altıntas¸ a, b, *, Murat Aksoy b, Emin Ünal a a b
Iskenderun Technical University, Department of Electrical and Electronics Engineering, Iskenderun, Hatay, 31200, Turkey Cukurova University, Department of Electrical and Electronics Engineering, Saricam, Adana, 01330, Turkey
A R T I C L E I N F O
A B S T R A C T
Keywords: Metamaterial Dielectric Fuel adulteration Oil condition
In this study, the omega shaped resonator based sensor structure by inspiring of metamaterial (MTM) concept is presented at X-band frequency regime for industrial purpose both numerically and experimentally. Four dis tinguishing applications which are (i) authentic and inauthentic gasoline samples, (ii) authentic and inauthentic diesel samples, (iii) clean and waste lubricant samples and (iv) clean and waste transformer oil samples have been realized by determining dielectric properties of them. Although there are close dielectric constant values between the samples (about 0.65 for (i), 0.25 for (ii), 0.10 for (iii) and 0.15 for (iv)), the proposed sensor sensitively discriminates all sample groups with about a frequency shift of 350 MHz for (i), 180 MHz for (ii), 60 MHz for (iii) and 70 MHz for (iv). Sensor structure can be efficiently used many industrial applications since the results are sensitively and precisely obtained for proposed studies. In addition, the proposed sensor is du rable, suitable for real-time applications and long-term stability considering the fabrication technique and nondestructive measurement method.
1. Introduction Recently, the scientists focus on the high-tech sensors since the In dustry 4.0 concept needs the sensor devices which have real-time operated, without requiring human resources and more precise. More over, some sectors ask for the sensing mechanism which can be achieved more sensitive, durable and long term stability with non-destructive methods. Hence, the researchers have investigated the metamaterial (MTM) based sensor structures to obtained novel and improved sensing characteristics. The MTM concept has been a popular research area among the scientists at the last decade. It has non-natural properties which are negative refraction index, reverse Snell Law and reverse Doppler Effect etc. Many applications and studies have been realized by the researchers in the topics such as absorbers [1,2], energy harvesters [3–5], polarization converters [6,7], super lenses [8,9], antennas [10, 11] and sensors [12–29]. Lots of MTM based sensor studies in the literature have been carried out in different approaches. Albishi et al. proposed a highly sensitive MTM based sensor structure which can be detected the cracks in the metallic materials [12]. Gargari et al. studied a wireless MTM inspired rotation sensor. Their sensing mechanism has sub milliradian resolution
to measure elastic-region bending in materials such as steel [13]. Altintas et al. designed a circular split ring resonator for multipurpose sensing applications such as dielectric characterization of liquids, rota tion detection of rods and strain sensing [14]. Saghati et al. improved a transmission line integrated sensor structure for dielectric spectroscopy of liquid chemicals in a wideband frequency regime both numerically and experimentally [15]. Wu et al. implemented a rectangular complementary-ring resonator based sensor to monitor permittivity of liquids at a working frequency of 900 MHz [16]. He et al. realized a thin film sensing application by using a tip shaped split ring resonator (SRR) MTM at microwave frequency regime [17]. Ni et al. proposed an MTM absorber based sensor structure to observe humidity of environment [18]. Keshavarz et al. realized an ultrahigh sensitive temperature sensor structure based on graphene semiconductor MTM in terahertz (THz) frequency regime [19]. Tumkaya et al. designed a MTM based sensor structure at X-band to discriminate authentic and inauthentic fuel samples [20]. Bakir et al. proposed a microfluidic and fuel adulteration sensing mechanism by using chiral MTM based structure in microwaves [21]. Tamer et al. investigated a transmission line based sensor for determining authentic and inauthentic gasoline samples [22]. Shah marvandi et al. implemented a CMOS MTM absorber based gas sensing
* Corresponding author. Cukurova University, Department of Electrical and Electronics Engineering, Saricam, Adana, 01330, Turkey. E-mail address: [email protected] (O. Altıntas¸). https://doi.org/10.1016/j.physe.2019.113734 Received 8 July 2019; Received in revised form 29 August 2019; Accepted 21 September 2019 Available online 27 September 2019 1386-9477/© 2019 Elsevier B.V. All rights reserved.
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
[23]. Rawat et al. studied a sensor structure which can achieve to sense the toxic and hazardous materials for human health [24]. Sethi et al. improved a blood glucose sensing mechanism by using a photonics based MTM [25]. Keshavarz et al. presented a novel water-based ter ahertz MTM and a semiconductor film to detect skin cancer [26]. Altintas et al. designed a meander line based sensor structure for permittivity characterization by using dielectric materials with the flat surface [27]. The proposed resonator structure has been inspired by the omega shaped metamaterial structures with strong resonance property which are highly used among the researchers. Aydin et al. suggested that the band gap of a single omega unit cell is due to electric and magnetic resonance since the omega type structures has bianisotropic properties [30]. Lheurette et al. discussed the left-handedness of the periodic omega type resonators at X and Ku band regime [31]. Li et al. studied different strong resonance conditions in transmission spectra with planar metamaterials with omega shaped metallic inclusions [32]. Basiry et al. investigated that the electromagnetic performance of the omega type metamaterials for planar radome applications [33]. Bal makou et al. present ground-plane-less bidirectional absorber applica tions with metamaterial based omega resonator structure at terahertz regime [34]. Labidi et al. designed a multi-band bowtie antenna based on omega-shaped resonator to improve return loss and radiation pattern parameters [35]. Many studies had been realized by researchers about microwave sensors with different kinds of resonator geometry for liquid detection applications. Gordon et al. proposed metasurface consisting of metallic electric-field coupled resonators operating in S band for chemicals [36]. Abduljabar et al. used a double split ring resonator to determine the dielectric properties of common liquids at around 3 GHz [37]. Miyamaru et al. designed a rectangular complementary split ring resonator to achieve sensitive sensing application for glycerol-water mixtures in terahertz regime [38]. Sadeqi et al. presented metamaterial on low cost and ubiquitous paper substrate by implementing an array of the circular disk for detection of dielectric properties of liquids [39]. Chen et al. realized terahertz sensing of highly absorptive water-methanol mixtures by using asymmetric dual wire resonator [40]. Lin achieved volatile organic solutions sensing application with complementary U-shaped resonator at terahertz regime [41]. Studies about MTM absorber based sensor applications had been achieved by the scientists. A new type dual band terahertz MTM absorber with a patterned metallic strip and a dielectric layer on a metallic ground plane is investigated for biological monitoring and sensing application [45]. A triple-band perfect absorber is realized by employing multiple different-sized metallic patterns and its sensing performance is examined in terms of the surrounding index [46]. Quad-band terahertz MTM absorber based on a common sandwich structure model is examined by focusing LC resonance. The structure has the ability to tune the frequencies of the absorption peaks and it has potential application perspectives in imaging and sensing [47]. Pressure, density, and volumetric moisture sensing applications have been ach ieved by using composed of a split ring resonator topology in microwave regime [48]. Swastika shaped MTM absorber structure is investigated in X-band and the sensing ability of the structure is tested by using some chemical liquids such as acetone, methanol, ethanol, polyethylene gly col [49]. In this paper, a metamaterial inspired omega shaped resonator based sensor structure is proposed for industrial implementations which are detections of fuel adulteration, lubricant condition and transformer oil condition. The samples studied in this paper are highly used in industry and they have very important role while utilizing from the industrial devices in energizing and lubrication phases. In addition, conditions and qualities of these materials is crucial issue to obtain maximum efficiency and minimum failure from the machines. The study is realized at X-band frequency regime both numerically and experimentally and the pro posed structure is by considering the proper waveguide sizes. CST
Microwave Studio simulation software is used in the numerical studies to observe the equivalent circuit diagram, electric field and surface current distribution, impedance matching and parametrical analysis of the sensor structure. Moreover, the measured dielectric characteristics of the sample by using a coaxial dielectric probe kit has imported to the simulation program and the sensor has been then numerically tested. After that, the experimental studies have been achieved by means of the PNA-L Agilent vector network analyzer (VNA) which has the operating frequency band between 10 MHz and 43.5 GHz. An X-band waveguide, two WR90 waveguide adapter and a proper sample holder are used during the measurements of the samples. Finally, simulated and measured results are compared and evaluated. The proposed sensor structure is a strong candidate for industrial implementations since it is sensitive, durable and can be used in real time applications. The novelty of the study is that four applications in one MTM based sensor structure are successfully tested comparing the other sensor studies in literature given above. The sensor structure is clearly discriminate the samples with respect to very close dielectric constant and loss values in X band. The study is more sensitively detect the liquid samples considering the other sensor structures which are operating frequency of X band. In addition, the MTM based sensor highly satisfy in terms of high Q-factor especially as experimental. 2. Design and theory The design and numerical studies of omega shaped resonator (OSR) are achieved by using a full wave electromagnetic solver finite inte gration technique (FIT) based CST Microwave Studio simulation soft ware. The OSR-based sensor structure has been designed at the X-band frequency regime. The overall size of the structure has 22.86 mm of width and 10.16 mm of length as illustrated in Fig. 1. This size has ar ranged by considering an X band waveguide to realize experimental work in the waveguide measurement method. The proposed structure has FR4 type dielectric substrate with a 1.6 mm of thickness, 4.3 of dielectric constant and 0.02 loss tangent values. The OSR consists of the copper type metal having a conductivity of 5.80001 � 107 S/m and thickness of 35 μm and it has been placed both front and back side of the FR4 substrate. The OSR can be defined by the parameters which are
Fig. 1. (a) The parametric dimensions of the OSR and (b) perspective view of the sensor structure. 2
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
internal radius (r), line width (w), gap width (g) and the horizontal arm line (Hline) as shown in Fig. 1(a). The values of these parameters are 2.2 mm, 0.8 mm, 0.8 mm and 13 mm, respectively. Sensor layer has been determined as the back side of the sensor structure with the thickness of 10 mm as shown in Fig. 1(b). This thickness value corresponds to the depth of the sample holder in the experimental studies. The equivalent circuit diagram has been analyzed to understand the working and the physical mechanism of the proposed OSR as shown in Fig. 2. The OSR can be divided into three parts such as ring, horizontal arms and gap demonstrated in Fig. 2(a). Basically, ring and horizontal arms have created inductive effects which can be defined as Lring and Larm respectively. The gap illustrated as Cgap creates a capacitive effect. Hence, the equivalent circuit diagram of the OSR has been obtained as an RLC circuit shown in Fig. 2(b) with these circuit parameters. Besides these parameters, the sensor layer which can be varied with the different samples having different electrical properties also creates a capacitive effect as Csens . Thus, the total capacitance Ct of the system can be written as; Ct ¼ C0 þ Cgap þ εsamp Csens
working principle for the engineering perspective and approach. The calculation of circuit parameters is another topic for a comprehensive project as a future study. Because, this calculation is highly affected by many properties of the incident wave such as angle, direction, polarization. Furthermore, the equivalent impedance of the structure can be rep resented as; � (4) Zeq ðωÞ ¼ Rt þ jωLt þ 1 jωCt where the Rt is overall resistance. The impedance of the free space; qffiffiffiffiffiffiffiffiffiffiffi Z0 ¼ μ0 =ε0 ¼ 377Ω (5) The reflection calculation between these two mediums can be writ ten as; �� � RðωÞ ¼ Zeq ðωÞ Z0 Zeq ðωÞ þ Z0 (6) The impedance matching between the proposed structure and free space is numerically studied as demonstrated in Fig. 3. Considering Eq (5), while the impedance of the structure Zeq ðωÞ approaches the impedance of the free space Z0 , the reflection RðωÞ is approachesto zero. This relation can be seen in Fig. 3. The refection magnitude is going to zero at the frequency of 11.70 GHz while the impedance of the structure is at maximum. Moreover, the imaginary part of the impedance of the structure is also zero and the transmission magnitude of the signal is at the maximum level at this frequency point. The relation between the impedance characteristics and maximum transmission ratio for the resonator has been explained Fig. 3. It can be concluded that the proposed resonator structure has the resonance fre quency point at X band with respect to the maximum transmission ratio. On the other side, the study is focused on zero transmission ratios since the sensor structure has been sensitively discriminated of the samples at X band. Hence the electric field and surface current distribution of the sensor is numerically presented at operating frequency point of 9.85 GHz with zero transmission coefficient in Fig. 4. The electric field is highly concentrated around the resonator part of the structure. Espe cially, it is located at the end of the horizontal arms and capacitive gap of the omega shaped resonator as shown in Fig. 4(a). These concentrations can describe how the capacitive effects occur in the resonator part of the sensor structure. The surface current with opposite directions is driven by capacitive effects. It has occurred at both left and right horizontal inductive arm of the resonator with the opposite directions as shown in Fig. 4(b). Hence, the zero transmission coefficient is obtained at the working frequency point of 9.85 GHz with these opposite surface currents.
(1)
where C0 is the capacitive effects which are originated from the sur rounding space and the εsamp Csens describes the capacitive effect of the sensor layer. The permittivity of the sample loaded to the sensor layer can be expressed as εsamp ¼ ε’samp þ jε’’samp . Hence the total capacitance ðCt Þ of the proposed structure is a function of the dielectric constant ðε’samp Þ and dielectric loss ðjε’’samp Þ of the sample placed in the sensor layer. � � (2) Ct ¼ Fc ε’samp ; ε’’samp The resonance frequency f0 of the structure can be written as; . � pffiffiffiffiffiffiffiffi � f0 ¼ 1 2π Ct Lt
(3)
where the Lt total inductance value of the OSR. Hence, the sensing characteristics of the structure can be easily monitored by the resonance frequency point shifts. This diagram presentation of the omega-shaped resonator as shown in Fig. 2(b) has been carefully prepared due to understanding the
3. Numerical and experimental studies The numerical setup has been prepared by using a full wave solver CST Microwave Simulation software and it is demonstrated in Fig. 5. The omega shaped resonator parts are placed in the front and back side of the dielectric substrate. The sensor layer with a thickness of 10 mm which correspond to experimental conditions has been arranged along the z direction. Two waveguide port is placed to the front and back side of the structure to monitor the transmission coefficient (S12). The reference planes are set with respect to both end plane of the structure. The fre quency range has arranged between 8 GHz–12 GHz which corresponds to X band. The boundary conditions are set to the perfect electric conductor (PEC) along x- and y-axis which corresponds to waveguide boundary conditions as shown in Fig. 5. The z-axis has determined as propagation direction. The parametric studies have been realized to observe the effects of the dimensions of the resonator on the resonance frequency of the structure as demonstrated in Fig. 6. The dimension parameters of gap width (g), line width (w) and horizontal arm length (Hline) are
Fig. 2. (a) Parts of the proposed sensor structure and (b) equivalent circuit diagram of OSR. 3
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
Fig. 3. Normalized real impedance, imaginary impedance, reflection magnitude and transmission magnitude of the proposed structure.
Fig. 4. (a) Electric field and (b) surface current distribution of the proposed sensor at 9.85 GHz.
investigated by varying from certain value ranges. The effect of the gap width parameter (g) is investigated between 0.7 mm and 0.9 mm by increasing 0.1 mm in Fig. 6(a). The resonance frequency is decreasing while the gap width is increased. The variation is linearly observed about with 20 MHz decrements. The resonance frequency point is highly affected by the line width parameter (w). This parameter is changed from 0.2 mm to 1.0 mm by increasing 0.2 mm point. This time, the line width parameter increased while the resonance frequency is increased too. This increment has linearly changed about with 180 MHz. Lastly, the horizontal arm length (Hline) parameter has been analyzed between 12.0 mm and 14.0 mm by increased 0.5 mm. The variation of the reso nance frequency has monitored as about 160 of decrements for each step. It can be concluded that the resonance frequency of the structure is highly affected by the resonator dimensions. The proposed structure is manufactured by using a CNC based LPKF E33 PCB prototyping machine. The fabricated sensor structure is illus trated in Fig. 7(a). Experimental studies are realized via an X-band
Fig. 5. Numerical study setup for proposed sensor structure.
4
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
placed to the front side of the waveguide. The fabricated sensor structure has embedded to the waveguide adapter coincided with the sample holder as shown in Fig. 7(b). After the configuration the waveguide has connected to PNA-L Agilent vector network analyzer (VNA) with a fre quency range between 10 MHz and 43.5 GHz as shown in Fig. 7(c). Before the measurements, the frequency range is arranged between 8 GHz and 12 GHz. The VNA has then calibrated with the way of special open circuit, short circuit and load connectors for both two ports. In addition, an open ended coaxial probe is used to determine the electrical properties of the liquid samples. The coaxial probe is connected to VNA and is calibrated by the way of air, water at 25� and special short circuit calibration kit. The dielectric characteristics are then measured by immerging the Keysight 85070 dielectric probe to the liquid samples. The dielectric values of the samples are calculated via the Keysight 85070 software embedded the PNA-L Agilent VNA. After that the sensing measurement setup are prepared and the transmission coeffi cient (S12) are measured with the waveguide, adapters, sensor structure and liquid sample holder as shown in Fig. 7. The liquid sample is firstly loaded by an injector via the channel placed in the top side inside the sample holder as shown in Fig. 7(c). After that, the transmission coef ficient is measured and in the same way, the sample is discharged from the holder. This method is applied for each measurement steps of the liquid samples. 3.1. Gasoline sample distinguishing study Gasoline is one of the highly used energy sources by many countries. The producer companies or governments are added to their own addi tives to mark the fuels. However, the unmarked gasoline fuels are sold by illegal ways as tax free. Hence, the purpose of this study is to solve this problem by distinguishing the authentic and inauthentic gasoline samples from each other. The electrical properties of the authentic and inauthentic samples are obtained between the 8 GHz and 12 GHz as illustrated in Fig. 8. Clear discrimination has been monitored between them. The dielectric con stant values of the authentic gasoline samples and inauthentic gasoline samples are about 2.4 and 1.75, respectively. The dielectric loss values of them are about 0.45 and 0.40, respectively. The differences in the dielectric characteristics of these two samples are clearly seen in Fig. 8. After the dielectric property determining study, the samples are defined in the CST Microwave simulation software as new materials. The numerical studies are realized by placing these materials into the sensor layer. The transmission coefficients for authentic and inauthentic gas oline samples are monitored between 8.5 GHz and 10.5 GHz as shown in Fig. 9(a). The resonance frequencies for authentic and inauthentic samples are 9.364 GHz and 9.772 GHz, respectively. About 408 MHz of frequency shift is obtained between these two resonance frequency points. The measured results for the authentic and inauthentic gasoline samples is shown in Fig. 9. The measured resonance frequency points are
Fig. 6. The parametric study plots for OSR structure; (a) gap width parameter (g), (b) line width parameter (w) and (c) horizontal arm length param eter (Hline).
waveguide, two WR90 waveguide adapter and a sample holder as shown in Fig. 7 (b). The sample holder having 10 mm of depth dimension is loaded with the liquid sample at every measurement phase and it is
Fig. 7. (a) Fabricated sensor structure, (b) waveguide configuration and (c) experimental setup. 5
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
Fig. 8. Dielectric constant and dielectric loss values of the authentic and inauthentic gasoline samples at X-band.
gasoline samples from each other. 3.2. Diesel sample distinguishing study The other highly preferred fossil fuel as an energy source is diesel. This fuel (like gasoline) is an important commercial tool for interna tional companies. To detection the inauthentic diesel is a crucial duty for the governments. This unbranded fuels also poses a great danger for the users with respect to engine healthiness. Hence, the authentic and inauthentic diesel sample distinguishing study is realized by using the proposed sensor structure. The dielectric characteristics of the authentic and inauthentic diesel samples are measured the X-band frequency regime by using a dielectric probe kit as shown in Fig. 10. At the frequency range of 8 GHz–12 GHz, the dielectric constant value for authentic diesel sample is between 2.79 and 2.52, and for inauthentic diesel sample is between 2.46 and 2.18. The dielectric loss values are almost at the same values for the samples as illustrated in Fig. 10. The numerical and experimental studies to distinguish the authentic and inauthentic diesel samples are presented in Fig. 11. The results are plotted at between the 8.5 GHz and 10.5 GHz. In the numerical studies, the lowest transmission coefficients are observed at the frequency points of 9.119 GHz and 9.311 GHz for authentic and inauthentic diesel sam ples, respectively as shown in Fig. 11(a). As to experimental studies, mentioned lowest transmissions are monitored at the frequency points of 9.496 GHz and 9.669 GHz, respectively as shown in Fig. 11(b). The frequency difference between the authentic and inauthentic diesel samples is 192 MHz in numerical and 173 MHz in experimental. The numerical and experimental studies of the sensor structure for diesel samples are given almost the same response except about the resonance frequency shift of 350 MHz. Hence, it can be concluded that the pro posed sensor structure can be sensitively distinguished the authentic and inauthentic diesel samples from each other.
Fig. 9. (a) Numerical and (b) experimental study plots for authentic and inauthentic gasoline samples.
monitored at 9.486 GHz and 9.847 GHz, respectively. The resonance frequency difference between these two frequency points is 361 MHz experimentally. It can be seen from Fig. 9 that the simulated and measured results are in a good agreement and the proposed sensor structure has the ability to discriminating the authentic and inauthentic
Fig. 10. Dielectric constant and dielectric loss values of the authentic and inauthentic diesel samples at X-band. 6
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
Fig. 11. (a) Numerical and (b) experimental study plots for authentic and inauthentic diesel samples.
Fig. 13. (a) Numerical and (b) experimental study plots for clean and waste lubricant samples.
3.3. Lubricant sample condition detection study
for clean and waste oil, respectively. The lowest level of the transmission ratio in the experimental study is obtained at 9.502 GHz and 9.441 GHz for clean and waste oil, respectively. The frequency shifts are obtained as 72 MHz in simulation and 61 MHz in experimental. Although there is 200 MHz of frequency shift, the numerical and experimental results are in a good agreement such a detection study with very close electrical characteristics of the samples. Therefore, it can be said that the sensor structure is able to discriminate the clean and waste lubricant sample from each other, precisely and sensitively.
In this study, the lubricant samples which are used in the mecha nisms such as engine systems, robotic machines, mechanics etc. have been investigated. It is very important to detect the machine or engine failures originated from the lube oil for safety and cost. Hence, this section of the study has been focused on the discrimination between the clean and waste lubricant sample. The lubricant samples have taken from the machines which are operated in hydraulic systems. The dielectric properties of the clean and waste lubricant samples are measured at X-band by means of the dielectric probe as shown in Fig. 12. The dielectric constant values and dielectric loss values for the clean and waste lubricant samples have been observed very closely. The dielectric constant values are between about 2.5 and 2.2. Dielectric loss values are between about 0.5 and 0.7 at X-band. It is hard to discriminate the clean lubricant sample from the waste oil since the dielectric properties of them is almost the same. However, the proposed sensor structure has precisely detected of the condition of the oils. The numerical and experimental studies of the lubricant sample condition sensing are plotted in Fig. 13. In the numerical studies. The minimum level of transmission is occurred at 9.257 GHz and 9.185 GHz
3.4. Transformer oil condition detection study The last study is carried out for detection of the transformer oil condition at X-band both numerically and experimentally. Transformers have a key role in the industry for the utilization of electrical energy and controlling the energy flow. Thus, protection of the transformers is a very important topic for health and safety management and system failures. In some type of transformers, special transformer oils are used for protection. Hence, this part of the study is focused on the transformer oil condition detection with clean and waste oils. The dielectric properties
Fig. 12. Dielectric constant and dielectric loss values of the clean and waste lubricant samples at X-band. 7
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
of the clean and waste transformer oil samples are measured at X-band regime as illustrated in Fig. 14. The dielectric constant of the clean and waste oil samples are close to each other and they are about 2.9 and 2.7. The dielectric loss values of them are monitored at almost the same value which is about between 0.5 and 0.8. Hence, the dielectric constant parameter is the key to detect the condition of the transformer oil. The numerical and experimental results for clean and waste trans former oils are plotted in Fig. 15 between the frequencies of 8.5 GHz and 10.5 GHz. In the numerical studies, the lowest transmission coefficient values at this frequency range have been observed at the frequency points of 9.026 GHz and 9.103 GHz for clean and waste oil samples, respectively. The difference between these frequency points is 77 MHz as shown in Fig. 15(a). In the experimental studies, these values have been monitored at the frequency points of 9.381 GHz and 9.446 GHz for clean and waste oil samples, respectively. The frequency shift between them is 65 MHz as demonstrated in Fig. 15(b). The similar frequency shift (about 350 MHz) between the numerical and experimental results has occurred in the transformer oil condition study as well. However, the results are in a good agreement considering the frequency difference between the clean and waste oil. It can be concluded that the proposed structure can determine the transformer oil condition sensitively. 4. Results and discussions The operating frequency range of the study has been carefully chosen since the most important issue is to monitor sufficient differentiation between the samples. The X band frequency regime gives the precise results to observe the difference sensitively with the sensor structure. Moreover, this higher frequency regime provides to get compact size sensor structure. Although the sensor including waveguide and adapters is expensive as price, the sensing implementation results for the four different study is very sufficient, sensitive and precise. Some of these applications are about the adulteration to prevent the smuggling and to save a huge amount of money. Some of them are about the improvement machine/transformer safety of systems and decreasing the costs by detecting failures instantly. Hence, the cost of the measurement setup with waveguide and the price of the sensor structure can be negligible. In the experimental process, transmission coefficients for every liquid sample have been measured by 10 times with the proposed sensor structure. The observed results remain constant. Hence it can be said that the sensor structure has high durability with substrate material having high flexural strength, thermal strength, and thermal stress. Moreover, proposed sensor sensitively distinguishes the samples from each other considering the other studies in the literature. In Ref. [20], the sensor structure provided about 100 MHz frequency shift between the authentic and inauthentic diesel samples. In the same study, about 50 MHz frequency shift occurred between authentic and inauthentic gasoline samples. In Ref. [22], the transmission line based sensor structure experimentally discriminated authentic and inauthentic diesel
Fig. 15. (a) Numerical and (b) experimental study plots for clean and waste transformer oil samples.
samples with 50 MHz of frequency shift considering the transmission coefficient. In Ref. [44], the researcher proposed a high sensitive sensor structure which can detect the fuel adulteration by 27 MHz of frequency shift in resonance frequency. A similar study with kerosene had been realized in Ref. [21] with 32 MHz of resonance frequency shift. In Ref. [42], the researchers presented a transmission line based sensor structure to detect transformer oil conditions by observing 40 MHz of resonance frequency shift between clean and waste transformer oil. In this study, the proposed sensor provides 350 MHz of frequency shift for authentic and inauthentic gasoline samples, 180 MHz of frequency shift for authentic and inauthentic diesel samples, 60 MHz of frequency shift for clean and waste lubricant samples and 70 MHz of frequency shift for clean and waste transformer oil samples. Hence it can also be concluded that the proposed sensor structure has high sensitivity ratio. The difference between two liquid samples for four applications is directly proportional with the dielectric constant value. The dielectric constant and resonance frequency differences are about 0.65 and 361 MHz, 0.27 and 173 MHz, 0.12 and 61 MHz, 0.13 and 65 MHz for gasoline, diesel, lubricant and transformer oil samples, respectively. It
Fig. 14. Dielectric constant and dielectric loss values of the clean and waste transformer oil samples at X-band. 8
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
can be seen from these values that some differences exist in sensitivity. However, these sensitivity differences are originated from the irregu larity of the dielectric loss values of the samples. Although the dielectric constant values differences are the same for lubricant and transformer oil samples, the dielectric loss values differences are about 0.05 and zero for these samples, respectively as shown in Figs. 12 and 14. Hence the sensitivity difference is observed as about 4 MHz. The same sensitivity issue can be observed between results for gasoline and diesel samples illustrated in Figs. 9 and 11. The direct relations can be observed be tween the dielectric constant difference and frequency shifts. However, the dielectric loss values of them play key role by determining the sensitivity. Although the simulations and experiments share similar resonance frequencies, their waveforms are different from each other. It can be seen that the quality factor (Q-factor) of the experimental results is higher than the simulated ones. The difference is originated from the waveguide calibration errors, probe loss and probe connection mis matches. This causes lower transmission ratio at resonance frequencies in experimental process. The reason why the omega-shaped resonator is chosen is that it can response both electric and magnetic component of the incident elec tromagnetic wave [30]. Since the split ring resonator based sensor structures only response to the magnetic component of the electro magnetic waves and sense the changes occurring in the magnetic field, they cannot be sufficient in some applications. On the other hand, the omega shaped resonator based sensor structure has the capability for more sensitive observations due to response both component of elec tromagnetic wave although it is polarization dependent like split ring resonator. In addition, the proposed sensor structure have compact size considering the operating frequency range, which can be considered as an advantageous for the implementations. Another advantageous side of the study is that the sensor promises a multi-functional sensing appli cation which can distinguish the four different samples groups. In literature, the diesel/gasoline adulteration sensing study and trans former oil condition study has been realized by different sensor struc tures [20–22,42,43].
Acknowledgment The authors would like to acknowledge the Scientific Project Unit of Cukurova University (FDK-2018-10488). References [1] H. Zhu, F. Yi, E. Cubukcu, Nat. Photonics 10 (11) (2016) 709. [2] B.X. Wang, G.Z. Wang, J. Mater. Sci. Mater. Electron. 28 (12) (2017) 8487. [3] H. Wang, V.P. Sivan, A. Mitchell, G. Rosengarten, P. Phelan, L. Wang, Sol. Energy Mater. Sol. Cells 137 (2015) 235. [4] M. Karaaslan, M. Ba� gmancı, E. Ünal, O. Akgol, O. Altıntas¸, C. Sabah, Opt. Quant. Electron. 50 (5) (2018) 225. [5] E. Unal, F. Dincer, E. Tetik, M. Karaaslan, M. Bakir, C. Sabah, J. Mater. Sci. Mater. Electron. 26 (12) (2015) 9735. [6] O. Akgol, O. Altintas, E. Unal, M. Karaaslan, F. Karadag, Int. J. Microwave Wireless Technol. 10 (1) (2018) 133. [7] T. Zhang, L. Huang, X. Li, J. Liu, Y. Wang, J. Phys. D Appl. Phys. 50 (45) (2017) 454001. [8] N. Kaina, F. Lemoult, M. Fink, G. Lerosey, Nature 525 (7567) (2015) 77. [9] N. Fang, X. Zhang, in: Proceedings of the 2nd IEEE Conference on Nanotechnology, vol. 225, 2002. [10] J.K. Ji, G.H. Kim, W.M. Seong, IEEE Antennas Wirel. Propag. Lett. 9 (2010) 36. [11] S.A. Rezaeieh, M.A. Antoniades, A.M. Abbosh, IEEE Antennas Wirel. Propag. Lett. 15 (2016) 1893. [12] A.M. Albishi, O.M. Ramahi, IEEE Trans. Microw. Theory Tech. 65 (5) (2017) 1864. [13] A.M. Gargari, B. Ozbey, H.V. Demir, A. Altintas, U. Albostan, O. Kurc, V.B. Ertürk, IEEE Sens. J. 18 (11) (2018) 4482. [14] O. Altintas, M. Aksoy, O. Akgol, E. Unal, M. Karaaslan, C. Sabah, J. Electrochem. Soc. 164 (12) (2017) B567. [15] A.P. Saghati, J.S. Batra, J. Kameoka, K. Entesari, IEEE Trans. Microw. Theory Tech. 65 (7) (2017) 2558. [16] J. Wu, P. Wang, X. Huang, F. Rao, X. Chen, Z. Shen, H. Yang, Sens. Actuators A Phys. 280 (2018) 222. [17] X.J. He, Y. Wang, J.M. Wang, T.L. Gui, Microsyst. Technol. 16 (10) (2010) 1735. [18] B. Ni, Z.Y. Wang, R.S. Zhao, X.Y. Ma, Z.Q. Xing, L.S. Yang, L.J. Huang, Y.Y. Lin, D. B. Zhang, Opt. Quant. Electron. 49 (1) (2017) 33. [19] A. Keshavarz, A. Zakery, Appl. Phys. A 123 (12) (2017) 797. [20] M.A. Tümkaya, F. Dinçer, M. Karaaslan, C. Sabah, J. Electron. Mater. 46 (8) (2017) 495. € Alkurt, O. Altıntas¸, S. Dalgac, [21] M. Bakır, M. Karaaslan, E. Unal, F. Karadag, F.O. C. Sabah, J. Electrochem. Soc. 165 (11) (2018) B475. [22] A. Tamer, F.O. Alkurt, O. Altintas, M. Karaaslan, E. Unal, O. Akgol, F. Karadag, C. Sabah, J. Electrochem. Soc. 165 (7) (2018) B251. [23] E.K. Shahmarvandi, M. Ghaderi, P. Ayerden, G. de Graaf, R.F. Wolffenbuttel, Procedia Eng. 168 (2016) 1241. [24] V. Rawat, R. Kitture, D. Kumari, H. Rajesh, S. Banerjee, S.N. Kale, J. Magn. Magn. Mater. 415 (2016) 77. [25] K.K. Sethi, G. Palai, P. Sarkar, Optik 168 (2018) 296. [26] A. Keshavarz, Z. Vafapour, IEEE Sens. J. 19 (4) (2019) 1519. [27] O. Altintas, M. Aksoy, E. Unal, F. Karakasli, M. Karaaslan, J. Electron. Mater. 47 (10) (2018) 6185. [28] A. Vivek, K. Shambavi, Z.C. Alex, Sensor review. https://doi.org/10.1108/SR-062018-0152, 2018. [29] A. Salim, S. Lim, Biosens. Bioelectron. 117 (2018) 398. [30] K. Aydin, Z. Li, S. Bilge, E. Ozbay, Photon. Nanostruct. Fund. Appl. 6 (1) (2008) 116. [31] E. Lheurette, G. Houzet, J. Carbonell, F. Zhang, O. Vanbesien, D. Lippens, IEEE Trans. Antennas Propag. 56 (11) (2008) 3462. [32] Z. Li, K. Aydin, E. Ozbay, Opt. Commun. 283 (12) (2010) 2547. [33] R. Basiry, H. Abiri, A. Yahaghi, Int. J. RF Microw. Computer-Aided Eng. 21 (6) (2011) 665. [34] A. Balmakou, M. Podalov, S. Khakhomov, D. Stavenga, I. Semchenko, Opt. Lett. 40 (9) (2015) 2084. [35] M. Labidi, R. Salhi, F. Choubani, Appl. Phys. A 123 (5) (2017) 313. [36] J.A. Gordon, C.L. Holloway, J. Booth, S. Kim, Y. Wang, J. Baker-Jarvis, D. R. Novotny, Phys. Rev. B 83 (20) (2011) 205130. [37] A.A. Abduljabar, D.J. Rowe, A. Porch, D.A. Barrow, IEEE Trans. Microw. Theory Tech. 62 (3) (2014) 679. [38] F. Miyamaru, K. Hattori, K. Shiraga, S. Kawashima, S. Suga, T. Nishida, M. W. Takeda, Y. Ogawa, J. Infrared, Millim. Terahertz Waves 35 (2) (2014) 198. [39] A. Sadeqi, H.R. Nejad, S. Sonkusale, Opt. Express 25 (14) (2017) 16092. [40] M. Chen, L. Singh, N. Xu, R. Singh, W. Zhang, L. Xie, Opt. Express 25 (13) (2017) 14089. [41] Y.S. Lin, Mater. Lett. 195 (55) (2017). [42] O. Altintas¸, M. Aksoy, E. Ünal, M. Karaaslan, J. Electrochem. Soc. 166 (6) (2019) B482. [43] M.A. Tümkaya, M. Karaaslan, C. Sabah, Chin. J. Phys. 56 (5) (2018) 1872. [44] V. Rawat, V. Nadkarni, S.N. Kale, Def. Sci. J. 66 (4) (2016) 421. [45] B.X. Wang, X. Zhaia, G.Z. Wang, W.Q. Huang, L.L. Wang, J. Appl. Phys. 117 (2015), 014504. [46] B.X. Wang, G.Z. Wang, T. Sang, J. Phys. D 49 (2016) 165307.
5. Conclusion In this study, a metamaterial inspired sensor structure has been presented for different industrial applications both numerically and experimentally. The electrical characteristics of the samples are firstly investigated at X-band frequency regime by using a coaxial type dielectric probe kit. The numerical studies are then realized for gasoline, diesel, lubricant and transformer oil samples. After that the proposed structure is manufactured and experimental analyses are carried out by PNA-L Agilent vector network analyzer and proper waveguide setup. Finally, numerical and experimental results are compared. The proposed sensor structure is able to clearly distinguish the inauthentic and authentic gasoline samples from each other since their dielectric values are highly different. The other inauthentic fuel detec tion study has been achieved by the diesel samples. The sensor structure has successfully discriminated the inauthentic diesel sample from authentic diesel sample. The last two studies are about lubricant and transformer oil condition detection. Dielectric properties of the clean and waste lubricant samples is observed as very similar. However, the sensor structure precisely exhibits the difference between the clean and waste samples with the frequency shift of 61 MHz in experimental. Accordingly, the dielectric values of the clean and waste transformer oil is very close to each other. The dielectric constant difference between these samples is about 0.15. The proposed sensor structure distinguishes sensitively the waste transformer oil sample from the clean sample with the 65 MHz in experimental. In conclusion, the sensor structure can be efficiently used in the industrial implementations at the X-band fre quency regime. Moreover, the structure is low cost, durable and reconfigurable at any desired frequency range.
9
O. Altıntas¸ et al.
Physica E: Low-dimensional Systems and Nanostructures 116 (2020) 113734
[47] B.X. Wang, IEEE J. Sel. Top. Quantum Electron. 23 (2017) 4700107. [48] M. Bakir, M. Karaaslan, O. Akgol, O. Altintas, E. Unal, C. Sabah, Optik 168 (2018) 741.
[49] Y.I. Abdulkarim, L. Deng, O. Altintas, E. Unal, M. Karaaslan, Phys. E Low-dimens. Syst. Nanostruct. 114 (2019) 113593.
10