Textile UHF-RFID antenna sensor for measurements of sucrose solutions in different levels of concentration

Textile UHF-RFID Antenna Sensor for Measurements of Sucrose Solutions in Different Levels of Concentration Chengyang Luo, Ignacio Gil and Ra´ul Fern´andez-Garc´ıa Department of Electronic Engineering, Universitat Politecnica de Catalunya, Barcelona 08222, Spain E-mail: [email protected] April 2021 Abstract. With the trend of textile antennas development, Ultra High Frequency (UHF, 865-868 MHz) Radio Frequency Identification (RFID) devices using textile materials are expected to be developed in many areas for replacing or simplifying some complex RFID devices on a PCB. In this paper, we present a textile UHF-RFID tag with two sensing positions (’radiation parts’ and ’loop part’) for exploring the feasibility of sucrose solutions measurements and the relationship between variables of the proposed design and the sucrose solutions. The simulated and measured resonance curves of the designs both match well (-20 dB in the simulation and -15.8 dB in the measurement at 868 MHz) and the read range measured by the RFID reader (M6e kit) is 1.71 m in air. Before the tests by the solutions on the proposed designs, a test board is developed as a preparation work for confirming the relative dielectric constants of the sensing substrate area in the real measurements. Compare the results of the simulation and the real tests, the proposed design shows good feasibility by comparing the simulated and measured results in confirmed relative dielectric constants. Moreover, its two sensing positions have different sensing features. The sensing ’radiation parts’ position shows a stable frequency operation performance but sensing range is from 1.71 m (dry) to 2.3 m, while the sensing ’loop part’ position has a wide sensing range from 0.4 m to 1.71 m (dry) but a lower frequency operation performance. Keywords: textile, Radio Frequency Identification (RFID), Utral High (UHF) frequency, sucrose solution, concentration, read range, relative dielectric constant 1. Introduction In modern society, Radio Frequency Identification (RFID) technology is essential in diverse applications [1][2][3] such as the goods classification, industrial process monitoring [4], transportation [5], localization and stuff management [21]. Most of them with popular integrated circuit (IC) chips are developed on a hard or flexible Printed Circuit Board (PCB). Due to the physical features of the commercial PCBs, most of RFID tags on a PCB are used as a ID tag, while when for sensing applications Author guidelines for IOP Publishing journals in L A T EX 2ε2 they are normally designed to connect to specifical sensors [6]. In addition, for sensing applications, RFID sensors on a PCB are often used in common fields such as temperature monitoring [7], brightness detection [8] and pressure measurements [9]. In order to expand application fields, some studies focused on RFID tags with chemical sensors [10][11], providing researchers more inspiration to explore the RFID technology on others materials with the similar features of physical or chemical sensors. In recent years, textile materials are obtaining more attention due to its special features such as flexibility, hygroscopicity and comfort according to the aforementioned background. Current RFID antenna designs deployed on textile materials involve two main approaches, one of which is a copper-based RFID antenna printed on a textile substrate [12][13][14][15], another of which is the metal-plated yarns sewed on a textile substrate [16][17][18][19]. Some of them focus on feasibility [13][14][15], reliability [17][18][19] and common applications such as tracking [16]. Especially for the work [12], the design is based on a copper antenna and a polyimide substrate which is applied for sensing solutions in the liquid through sensitivity (read power) of the RFID tag as the sensing parameter. It is a good idea and deserved to be referred for some RFID antenna sensor designs based on complete textile materials which offer the advantage of being more sensitive to liquids due to the conductive yarns in comparison with normal copper designs. Moreover, some works [18][19] are based on NFC technology operating at 13.5 MHz, which mainly uses capacitive or inductive coupling ways for sensing and is different with UHF-RFID technology based on backscattering [20]. Compared with the common Ultra High Frequency (UHF, 865-868 MHz) RFID tags and sensors on a PCB, the majority of textile UHF-RFID tags and sensors are embroidered by conductive yarns on textile substrates [20][22], which is more different than the copper on PCB. On the one hand, the textile tags can be deployed on common clothes which are more comfortable and lighter than those based on PCB inserted into clothes. On the other hand, the conductive yarns and textile substrates are sensitive to the environmental factors such as humidity [23], temperature, bending and washing [24][25]. As a result, textile UHFRFID sensors can be explored to replace or simplify some complex or expensive UHFRFID sensors on the PCB. Comparing with some studies using chemical sensors on a PCB for the target of wireless sweat-based monitoring [26], some textile materials with better hygroscopicity are expected to reach same target by supporting conductive-yarns-plated UHF-RFID tags. In addition, based on our previous study on solutions sensing [27], the textile UHF-RFID sensor based on only tags is promising to be explored gradually. In this paper, the textile UHF-RFID tag with two sensing positions (’radiation parts’ and ’loop part’) is proposed for sucrose solution measurements. Considering the possibility of expending future sensing applications, the function-extensible IC chip (ROCKY 100) is used in the proposed design. Before the tests on proposed textile UHF-RFID sensor, a test board is developed in the preparation work for confirming the relative dielectric constants of the sensing substrate area in the real measurements. The fundamental resonance analysis for the simulated and measurements is performed. Author guidelines for IOP Publishing journals in L A T EX 2ε3 Then in order to explore the relationship between variables of the proposed design and the sucrose solutions, two sensing positions are used to absorb the sucrose solutions with different levels of concentration. Then the read ranges of the design in different situations are tested by a RFID reader (M6e Kit) and analyzed combined with simulated results after obtaining the real relative dielectric constants of the sucrose solutions by the preparation work. 2. Design and Related Details 2.1. Structure of the proposed textile UHF-RFID antenna Figure 1. Geometry and configuration of the proposed design. (a) simulated design diagram, (b) embroidered pattern diagram, (c) photograph of the proposed design The geometry and configuration of the proposed textile UHF-RFID antenna sensor are shown in Fig. 1.The design mainly consists of two parts which are the ’loop part’ for impedance match and the ’radiation parts’ for improving the transmit-receive ability as shown in Fig. 1 (c). In Fig. 1 (a), the simulated design is presented and related size parameters are detailed in Table. 1. Note that when the design is embroidered by specific machines, the embroidery pattern makes certain influence on the impedance of the textile UHF-RFID antenna. As a result, the proposed design is converted to an embroidery pattern by ’satin fill’ mode as shown in Fig. 1 (b). In addition, for the UHF-RFID integrated circuit (IC) chip, a ROCKY 100 is selected due to its function-extensible feature. Comparing with some popular chips for textile RFID tags such as Monza R6, the ROCKY 100 can extend functions by connecting to various sensors. Certainly, in this paper, the proposed textile UHFRFID tag is explored to be an antenna sensor without sensors expansion. The complex impedance of the used chip (ROCKY 100) is 64-i*469 ohm and the minimum wake-up power is -10 dB. For conjugate matching, the impedance of the feed port in simulation software needs to be set as 64+i*469 ohm. 2.2. Materials The proposed textile UHF-RFID antenna sensor is embroidered using conductive yarns as the antenna sensor and polyester fibers as the substrate. The textile material of the Author guidelines for IOP Publishing journals in L A T EX 2ε4 Table 1. Size parameters of the design (Unit: mm) Parameter L1L2W1T1R1R2 V alue 23 14.5 18 2 4 10 Figure 2. Substrate parameters measurements. (a) Permittivity and loss tangent measurements, (b) Thickness measurement. antenna is a commercial conductive twisted yarn (Shieldex 117/17 dtex 2-ply) made of 99% pure silver-plated Nylon. Note that in order to reduce the impact from the difference between the pure silver and the real silver-plated Nylon[28][29], the most closed bulk conductivity of the yarns is redefined as 11500 siemens/m in the simulation software. Moreover, the polyester [30] is selected as the substrate due to its better hygroscopicity. Its relative dielectric constant and loss tangent in a dry situation are 1.24 and 0.000163, respectively, measured by a Microwave Frequency Q-Meter as shown in Figure 2 (a). On the other hand, its thickness is 0.62 mm measured by an Electronics Outside Micrometer (132-01-040A) as shown in Figure 2 (b). Considering the textile materials have the special feature (better hygroscopicity than PCBs), the solutions in different levels of concentration are expected to make an impact on substrate performance. In next section, this impact is evaluated. 3. Tests and Achieved Results Resistance and reactance curves for antennas have bigger slopes in the curve-rising stage, which means the resistance and reactance values change sensitively. Hence, when the design is used to absorb certain solutions causing change of substrate εr, some electrical properties such as the resonance frequency and read range are expected to change. In this section, the related tests are done for exploring the viability of the prototype as a sensor and relationship between variables. The measurement procedures of the whole work are divided into two parts linked by the special ‘medium’ as shown in Fig. 3. The first part is the preparation test and the second part is the UHF-RFID antenna sensor test. In detail, a simple capacitive structure is developed by means of simulation software and embroidered by Author guidelines for IOP Publishing journals in L A T EX 2ε5 the aforementioned embroidery machine. The S-parameter curves of the simulation and the real measurements are expected to be matched to obtain the corresponding values of εras the fitting parameter. The obtained values εrrelated to the solution concentration are used in the UHF-RFID sensor simulation of the second part and then the simulated results are compared with the measured results to confirm the feasibility of the measurement method by the textile UHF-RFID sensors. Note that the εris the bridge or medium to connect the first and second parts as shown in Fig. 3. Figure 3. Measurement procedures of the whole work 3.1. Preparation Test Figure 4. Geometry and configuration of the test board. (a) simulated test board diagram, (b) photograph of the test board In order to confirm the real relative dielectric constants of the sensing areas of the proposed design in different situations for validating the feasibility to be a wireless sensor for the concentration measurement of sucrose solutions, a preparation test is conducted in this section. As shown in Fig. 4, the same textile-based test board is designed by a professional embroidery machine (Singer Futura XL-550) to measure the real relative dielectric constants of the sensing areas with the sucrose solutions [32] in the different levels of concentration. The Fig. 4 (a) depicts the simulated test board model which is Author guidelines for IOP Publishing journals in L A T EX 2ε6 Table 2. Size parameters of the test board (Unit: mm) Parameter Lt1Lt2Tt1Tt2 V alue 44 12 2 2 embroidered by the mentioned method as shown in Fig. 4 (b). In addition, the related size parameters are given in the Table. 2. Moreover, the sucrose solutions are prepared by distilled water and sucrose (0, 125 mg/dl and 250 mg/dl). In order to obtain the sucrose solutions with different levels of concentration, accurate weighing is done for the sucrose by a precise balance (PCE-BS 300). Figure 5. Measurement setup for the preparation test. (a) Photograph of measurement setup, (b) Measurement setup configuration. The measurement setup for the preparation test is shown in Fig. 5 and the test procedure is listed as follows, (i) Connect the two ports of the test board to the Microwave Analyzer N9916A as shown in Figure 5 (a), and save the reflection coefficients of the test board in dry conditions. (ii) Compare the saved data from first step with the simulated curves of the test board model and confirm the both are matched. (iii) Drop the sucrose solutions in different levels of concentration (0, 125 mg/dl and 250 mg/dl) onto the sensing position of the test board, and save the reflection coefficients in each situation. (iv) Adjust the relative dielectric constants of the sensing area in the simulation software to find the matched curves with the measured curves. The corresponding relative dielectric constants are closed to the real relative dielectric constants of the sensing substrate area in the real measurements. By the above test procedure, the test results are shown in Fig. 6 and Fig. 7. From the experimental results, the relative dielectric constants of the sensing substrate area tuned to 62.24, 23.24 and 10.24 for 0, 125 mg/dl and 250 mg/dl, respectively. Therefore, Author guidelines for IOP Publishing journals in L A T EX 2ε7 Figure 6. Simulated and measured |S12|of the test board for measuring the relative dielectric constant of the tested position with solutions. Figure 7. Simulated and measured |S11|of the test board for measuring the relative dielectric constant of the tested position with solutions. the obtained relative dielectric constants can be used for the proposed design in next sensing tests. 3.2. Resonance analysis Fig. 8 shows the simulated and measured reflection coefficients (|S11|) of the embroidered UHF-RFID antenna, in which the measured curve is obtained by Microwave Analyzer N9916A with ’port extension’ function. The simulated |S11|reaches -22 dB at resonance Author guidelines for IOP Publishing journals in L A T EX 2ε8 Figure 8. Simulated and measured reflection coefficients of the embroidered designs in a dry condition. frequency 868 MHz and the bandwidth under -10 dB is 337 MHz from 646 MHz to 983 MHz. Comparing with the simulated result, the measured |S11|is a little different, which shows -15.8 dB at 868 MHz and 170 MHz of the bandwidth from 782 MHz to 952 MHz. From the results, the design can work at targeted ultra high resonance frequency (868 MHz). 3.3. Tests under sucrose solutions in different levels of concentration Considering the specifical features of the proposed textile UHF-RFID tag such as the sensitive impedance change and the better substrate hygroscopicity than that of a common PCB,the sucrose solution for the proposed textile UHF-RFID tag is selected to explore the possibility as a sensor and the relationship between the concentration and the read ranges. For the proposed design, there are two positions selected as the sensing areas, the ’radiation parts’ and the ’loop part’ as shown in Fig. 1 (c). The two positions are analyzed with respect to the Friis Transmission Formula as follows [13][31], dmax =λ 4π√PtGtGr·τ Pth (1) where dmax is the maximum value of the read range, λis the wave length at 868 MHz, Ptis the power fed into the reader antenna, Gtis the gain of the reader antenna, Gris the gain of the proposed antennas, τis the largest power transmission coefficient and Pth is the minimum wake-up power of the chip. From the equation 1, the maximum value of the read range (dmax) can be affected by the gain of the proposed antenna (Gr) and the largest power transmission coefficient (τ). Therefore, the ’radiation parts’ and the ’loop part’ related to the Grand τ, respectively, Author guidelines for IOP Publishing journals in L A T EX 2ε9 are good choices. When the solutions with different levels of concentration are dropped on the sensing areas, the relative dielectric constants of the sensing areas are expected to change. The measured relative dielectric constants of the sensing area in different levels of concentration are confirmed in the preparation test above. ·Tests at the ’radiation parts’ position The reflection coefficients (|S11|) are simulated by sweeping the relative dielectric constant of the substrate at only the ’radiation parts’ and the results at εr=1.24 (dry), εr=10.24 (250 mg/dl), εr=23.24 (125 mg/dl) and εr=62.24 (0 mg/dl) are selected to compare with the real measurement results. Figure 9. Simulated reflection coefficients of the proposed design with different relative dielectric constants at two ’radiation parts’ positions. Figure 10. Simulated realized gain at 868 MHz of the proposed design with different relative dielectric constants at the two ’radiation parts’ positions. As shown in Fig. 9, all |S11|curves are close, which means when using ’radiation Author guidelines for IOP Publishing journals in L A T EX 2ε16 [3] Yang M, Zhang W and Li L 2017 A Resistance-Type Sensor Based on Chipless RFID IEEE Trans. Antennas Propag. 65 3319-3325. [4] Shafiq Y, Gibson J S, Kim H, Ambulo C P, Ware T H and Georgakopoulos S V 2019 A Reusable Battery-Free RFID Temperature Sensor IEEE Trans. Antennas Propag. 67 6612-6626. [5] Wang Y, Pretorius A J and Abbosh A M 2016 Low-Profile Antenna With Elevated ToroidShaped Radiation for On-Road Reader of RFID-Enabled Vehicle Registration Plate IEEE Trans. Antennas Propag. 64 1520-1525. [6] Deng F, He Y and Wu X 2014 A CMOS pressure sensor with integrated interface for passive RFID applications Meas. Sci. Technol. 25(12) 125104. [7] Vaz A, Ubarretxena A and Zalbide I 2010 Full passive UHF tag with a temperature sensor suitable for human body temperature monitoring IEEE Trans. Circuits Syst. II Express Briefs 57 95-99. [8] Falco A, Salmer´on J F and Loghin F C 2017 Fully printed flexible single-chip RFID tag with light detection capabilities Sensors 17 534. [9] Kuhn M F, Breier G P and Dias A R P 2018 A novel RFID-based strain sensor for wireless structural health monitoring J. Nondestr. Eval. 37 1-10. [10] Zampolli S, Elmi I and Cozzani E 2018 Ultra-low-power components for an RFID Tag with physical and chemical sensors Microsyst. Technol. 14(4-5) 581-588. [11] Manzari S, Catini A and Pomarico G 2014 Development of an UHF RFID Chemical Sensor Array for Battery-Less Ambient Sensing IEEE Sens. J. 14 3616-3623. [12] Ennasar, M.A., El Mrabet, O., Mohamed, K. and Essaaidi, M. 2019 Design and characterization of a broadband flexible polyimide RFID tag sensor for NaCl and sugar detection. Prog. Electromagn. Res. 94 273-283. [13] Le D, Ahmed S and Ukkonen L 2021 A Small All-Corners-Truncated Circularly Polarized Microstrip Patch Antenna on Textile Substrate for Wearable Passive UHF RFID Tags. IEEE J. Radio Freq.. [14] Manzari S, Occhiuzzi C and Marrocco G 2012 Feasibility of body-centric systems using passive textile RFID tags. IEEE Trans. Antennas Propag. 54 49-62. [15] Ginestet G, Brechet N and Torres J 2016 Embroidered antenna-microchip interconnections and contour antennas in passive UHF RFID textile tags. IEEE Antennas Wirel. Propag. Lett. 16 1205-1208. [16] Liu Y, Yu M and Xia B 2021 E-Textile Battery-less Displacement and Strain Sensor for Human Activities Tracking. IEEE Internet Things J. [17] Liu Y, Xu L and Li Y 2019 Textile based embroidery-friendly RFID antenna design techniques IEEE International Conference on RFID 1-6. [18] Escobedo P, de Pablos-Florido J and Carvajal M A 2020 The effect of bending on laser-cut electrotextile inductors and capacitors attached on denim as wearable structures. Text. Res. J. 90 2355-2366. [19] Jiang Y, Xu L and Pan K 2019 e-Textile embroidered wearable near-field communication RFID antennas. IET Microw. Antennas Propag. 13 99-104. [20] Luo C, Gil I and Fern´andez-Garc´ıa R 2020 Wearable textile UHF-RFID sensors: A systematic review. Materials, 13 3292. [21] Ali Z, Perret E and Barbot N 2020 Authentication Using Metallic Inkjet-Printed Chipless RFID Tags IEEE Trans. Antennas Propag. 68 4137-4142. [22] Koski K, Syd¨anheimo L and Rahmat-Samii Y 2014 Fundamental Characteristics of ElectroTextiles in Wearable UHF RFID Patch Antennas for Body-Centric Sensing Systems IEEE Trans. Antennas Propag. 62 6454-6462. [23] He H, Chen X and Khan Z 2020 Textile-based Passive Sensor for Air Humidity 2020 IEEE 8th Electronics System-Integration Technology Conference (ESTC) 1-3. [24] Fu Y Y, Chan Y L and Yang M H 2015 Experimental Study on the Washing Durability of ElectroTextile UHF RFID Tags IEEE Antennas Wirel. Propag. Lett. 14 466-469. [25] Simorangkir R B V B, Le D and Bj¨orninen T 2019 Washing Durability of PDMS-Conductive Author guidelines for IOP Publishing journals in L A T EX 2ε17 Fabric Composite: Realizing Washable UHF RFID Tags IEEE Antennas Wirel. Propag. Lett. 18 2572-2576. [26] Gao W, Emaminejad S and Nyein H Y Y 2016 Fully integrated wearable sensor arrays for multiplexed in situ perspiration analysis Nature 529 509–514. [27] Luo C, Fern´andez-Garc´ıa R and Gil I 2020 Embroidered Textile Capacitive Sensor for Sucrose Solutions Measurement,” 2020 IEEE International Conference on Flexible and Printable Sensors and Systems (FLEPS) 1-4. [28] Zhang H 2015 Flexible textile-based strain sensor induced by contacts Meas. Sci. Technol. 26(10) 105102. [29] Logothetis I, Bayramol D V and Gil I 2020 Evaluating silver-plated nylon (Ag/PA66) e-textiles for bioelectrical impedance analysis (BIA) application Meas. Sci. Technol. 31(7) 075101. [30] http://www.urbatex.com/en/tessuto/polyester-fabric/ [31] Deleruyelle, T., P. Pannier, M. Egels, and E. Bergeret 2010 Dual band mono-chip HF-UHF tag antenna Proc. IEEE Antennas Propag. Soc. Int. Symp. 1–4. [32] Malmberg C G and Maryott A A 1950 Dielectric constants of aqueous solutions of dextrose and sucrose Journal of Research of the National Bureau of standards 45 299-303. [33] Harnsoongnoen S and Wanthong A 2015 Coplanar waveguides loaded with a split ring resonatorbased microwave sensor for aqueous sucrose solutions Meas. Sci. Technol. 27(1) 015103.