Free Space Measurement Technique on Dielectric ...

Free Space Measurement Technique on Dielectric Properties of Agricultural Residues at Microwave Frequencies F.H.Wee, P.J.Soh, A.H.M Suhaizal, H.Nornikman, A.A.M Ezanuddin School of Computer and Communication Universiti Malaysia Perlis (UniMAP) Malaysia [email protected], {pjsoh, suhaizal}@unimap.edu.my, [email protected], [email protected] Abstract— Principles to determine dielectric properties measurement using microwave free-space transmission measurements technique is presented in this paper. The project’s main objective was to measure the dielectric properties of different agricultural waste material at microwave frequencies, which in turn, can serve as a potential alternative, replacing conventional printed circuit board (PCB) using FR4 material. A system, which includes a PNA vector network analyzer, horn antennas, and Agilent Technologies 85071E material measurement software were used and described in detail. Techniques and procedures essential in obtaining reliable permittivity data for agricultural waste products were also discussed. The waste products – rice husk, rice straw, and kenaf were studied in the frequency range of 2.2 to 3.3 GHz. These agricultural waste materials were mixed with two distinct resins, namely, Urea Formaldehyde (UF) and Phenol Formaldehyde (PF). Each type of resin was also investigated using different percentage of resin moisture content, and the results were analyzed thoroughly. Keywords-Free Space Measurement Technique, Dielectric Properties, Agricultural Residues, Microwave Frequencies

I. INTRODUCTION Dielectric properties measurement is an important factor in defining the physical and chemical properties related to storage and energy loss in various kind of materials [1-2]. Measurement of the dielectric properties of agricultural waste materials such as rice husk, rice straw and kenaf are essential for the understanding of their electrical behavior and the development of non-destructive methods for determining their physical characteristics. The dielectric properties are, by definition, a measure of the polarizability of a material when subjected to an electric field. Equation (1) shows the permittivity and loss factor respectively [3].

K=

=

=

-

(1)

= is the free space of permittivity Permittivity (ε), also known as a material’s dielectric constant, describes the interaction of a material with an

978-1-4244-5357-3/09/$26.00©2009IEEE

electric field. Dielectric constant (k) is equivalent to relative permittivity or the absolute permittivity (ε) relative to the permittivity of free space ( ). The real part of permittivity ( ) is a measure of how much energy from an external electric field is stored in a material. The imaginary part of permittivity ) is called the loss factor and is a measure of how ( dissipative or loss of a material is to an external electric field. Microwave is the alternating current signals with its electromagnetic wave frequencies between 300 MHz and 300 GHz, with a corresponding electrical wavelength, λ = c/ f between, 1 mm and 1 m, respectively. Microwave communication is widely used for long distance telephone communication and cell phones [4]. At microwave frequencies, various measurement techniques are available, which include transmission line techniques (waveguide, coaxial and free-space), impedance, coaxial probe and cavity methods. Amongst these techniques, free-space measurement technique has the advantages of allowing reflection and transmission measurements without any physical contact with the sample. The free space technique is best for thin flat faced materials, or other materials that can be formed into this shape. For the transmission line technique, the sample is deformed into a particular shape, depending on the transmission line used, either waveguide or coaxial type. On the other hand, for the coaxial probe technique, the sample tested should be in the form of liquid or powder, while for free space, samples are larger and in solid form [5]. The numerous types of agricultural cropping residues that exist are such as wheat straw, rice straw, barley straw, oat straw, seed grass straw, and rice husk as well as kenaf. These waste materials were gathered after the harvest season, as to ensure the three types of waste investigated were properly dried up. This research is also intended to be a catalyst in identifying alternative purposes for agricultural waste, while reducing hazard due to open burning [6].

183

II.

METHODOLOGY

A. Equipment and Measurement Setup The basic free space measurement system consists of: (1) a network analyzer, in this case is an Agilent PNA network analyzer, (2) a sample holder, (3) two horn antennas, and (4) Agilent 85071E software. Two horn antennas will be used as the transmitter and receiver, respectively, and the agricultural waste material boards will be placed in between them as the material under test (MUT) [1]. The horn antenna is shown in Figure 1. The antenna’s dimension used was 309 mm x 238.5 mm x 294 mm, with operating capability between 2.2 GHz to 3.3 GHz with VSWR is less or equal to 1.225.

Figure 1: Horn antenna

The Network Analyzer provided S- parameters such as and , and was installed with Agilent Technologies 85071E, the material measurement software. This software has the capability to determine the intrinsic electromagnetic properties of many dielectric and magnetic materials. The 85071E software controlled the network analyzer and calculated the complex permittivity ε’ (or dielectric constant) and permeability µ’, including the loss factor. Results were displayed as a function of frequency, with 1 to 2% accuracy (typical). Depending on the Agilent network analyzer and fixture used, test frequencies can also be extended to 325 GHz [3]. Figure 2 shows PNA network analyzer that used in the measurement.

Figure 2: PNA Network Analyzer

Free Space Measurement system setup is shown in Figure 3. The data gathered through this measurement will be able to determine a set of dielectric constant and loss factor, in order for characterization of these agricultural materials. These results are also important to investigate and study the properties of the materials which are planned to be used for the further RF microwave design. In addition to this, results graphs can also be viewed through the measurement software that was available as an option in PNA Network Analyzer – the Agilent 85071E Measurement software.

Figure 3: Free Space Measurement System

Rice Husk, Rice Straw, and Kenaf material boards were the Material under Test (MUT) for the Free Space Measurement Technique. These boards were placed perpendicularly with the bottom plane that was used to support horn antennas and agricultural material boards. Sample holders were used to hold the material boards such that it is non-contacting with the horn antennas. Without sample holder, it will cause the material board to be deviate to some degrees. Measurements without the sample holder have produced some deviation, which in turn will lead to inaccuracies [3]. The tested material boards are shown in Figure 4.

(a) (b) (c) Figure 4: Rice Husk, Rice Straw and Kenaf material boards

B. Calibration Measurement errors such as systematic, random and drift errors are most probably due to imperfections in the analyzer, and instrument noises (source phase noise, sampler noise, IF noise) [7]. These undesired errors must be removed by two port calibration, for both reflection and transmission measurements. Two-port calibration is done on four known standards (shortopen-load through or SOLT) [8]. In this calibration, an Agilent 85052 D calibration kit was used. C. Time Gating Setting The material board samples were placed in between the horn antennas and the time domain gating feature of the PNA was used to determine the actual placement distance of material under test [8]. The setting of time domain gating difference must feature can be seen in Figure 5. The average be more or equal to 40 dB when measuring a non-metal plate, compared to measuring a metal plate. The ‘a’ symbol shows the plot of the measurement with metal plate and non- metal plate. The two responses appeared at ‘b’ and ‘c’ as shown in figure 5 were the transmitting and receiving horn antenna responses. This was done across the frequency range of 2.2 GHz and 3.3 GHz, frequencies where the use of the horn antennas was optimized.

2009 SBMO/IEEE MTT-S International Microwave & Optoelectronics Conference (IMOC 2009)

184

b

from that, a software simulation setup was also introduced to provide an avenue of comparison of S- Parameter between hardware measurement and software simulation. Dielectric properties results that obtained through measurement were calculated averagely from 2.2 GHz until 3.3 GHz.

c a

A. Dielectric Properties of Rice Husk with the effect of both resin PF and UF.

Figure 5: Time Gating Setting

D. Simulation Computer Simulation Technology (CST) simulation program was then used in order to study the variation of the Free Space Measurement setup, which is shown in Figure 6. The measurement setup was remodeled in this software program along with the agricultural material board of exact dimensions. Waves were excited through the horn ports, and the scattering parameters characterizing the material were obtained. The parameters that needed to be specified in this simulation were the dielectric properties of the agricultural material boards (measured through free space measurement technique), range of frequencies in hardware measurement, actual dimensions of the equipment and it material under test.

Both Urea Formaldehyde (UF) and Phenol Formaldehyde (PF) are strong adhesives, which are popularly used to create strong bonds between adjacent material layers [9], thus was chosen as the agricultural waste material bonding agent. From Table I, the dielectric constant, ε’, of the rice husk bonded using PF composite was found to be higher than the UF composite. This trend was also observed in differing percentage (20% and 30%) contents of both resins. Higher moisture level in PF will lead to higher dielectric constant data compare to low moisture level contain in UF [11-12]. TABLE I. DIELECTRIC PROPERTIES OF RICE HUSK WITH EFFECT OF PF AND UF Resin

Percentage, %

PF

10 20 30 10 20 30

UF

(a) 29.5 mm 309 mm

46 mm 294 mm

(b)

215 mm

Figure 6: Free Space Measurement System in CST

III.

RESULT AND DISCUSSION

Several variations of MUT were tested to investigate the degree of distinction that the Free Space Measurement Technique was able to obtain. These MUTs include 1. Dielectric properties of rice husk with the effect of different resin Phenol Formaldehyde, PF and Urea Formaldehyde, UF. 2. Dielectric properties of three type agricultural waste material boards with the effect of different percentage of resin Phenol Formaldehyde, PF. 3. Dielectric properties of three type agricultural waste material boards with the distant effect between horn antennas with material under test (MUT). Apart

Dielectric Properties ε’

ε”

3.236 3.405 3.681 2.891 3.275 3.581

0.274 0.268 0.439 0.222 0.329 0.272

One of the reasons to this was due to UF, being less water absorbent, while its original liquid form is also denser compared to PF [9]. As can be seen across all percentages, the loss factor of the rice husk particle boards have been found to be increasing and declining unsteadily. This is due to the uneven distribution of the resin on the raw rice husk. Some part of the rice husk board was covered with a high amount of resin, while some other parts with a low volume of resin. Thus the loss factor for the part that high resin concentration was found to be higher compared to areas of low resin concentration [10]. B. Dielectric Properties of three types agricultural wastes material boards with the effect of different percentage of resin Phenol Formaldehyde, PF The agricultural waste material boards were fabricated with resin Urea Formaldehyde (UF) concentration varying from 10 % to 50% by volume and the thicknesses were 8 mm. Table II gives the ε’ and ε” values for three types of agricultural waste material boards and it was observed that ε’ increases as the concentration of PF was increased from 10%, 30%, and 50% respectively. This is due to higher volume fraction of the chemical resin in the composite. Water is characterized by high dielectric values and thus explains the higher dielectric constant values at elevated moisture contents


185

[11-13]. Thus, it showed that for similar measurement distances for all three materials, the dielectric constant, ε’ increased linearly with the increase in percentage of both resins. The inaccuracies of ε” measurement in free space technique are due to two main sources of errors; namely (1) the diffraction effects at the edges of agricultural residues material specimen and (2) multiple reflection during signal transmission occurring between the two horn antennas and mode transitions via surface of the sample. These propagation effects had lead to low signal strength, as well as a lengthened time for signal to reach the MUTs [14]. TABLE II. DIELECTRIC PROPERTIES OF AGRICULTURALWASTES WITH EFFECT OF DIFFERENT PERCENTAGE OF RESIN PF Material

Percentage, %

RH

10 30 50 10 30 50 10 30 50

RS

K

Dielectric Properties ε’

ε”

3.139 3.473 4.491 1.891 1.974 2.639 2.101 2.141 2.382

0.477 0.617 0.484 0.215 0.372 0.211 0.054 0.069 0.132

antennas. The results can be seen in Table III. This is due to the electromagnetic penetration into the material boards, which was less for larger distances and vice versa for the shorter distances during measurement [14]. Hence, the material boards have undergone more energy loss than it was absorbing. The reason of inconsistent value of ε” was similar to observation in the previous measurement [16]. D.

S- Parameter simulation vs measurement

Finally, a comparison between the hardware measurement and software simulation was carried out, generating a list of Sparameters ( and ). For the hardware measurement, and can be calculated automatically by using the Network Analyzer and software, while for the software simulation, the dielectric constant data from the hardware measurement was and results via simulation using used to produce the Computer Simulation Technology (CST) software. Scattering parameter of and of the kenaf boards with different composite percentage of Phenol Formaldehyde, PF resin, showed good agreement with the similar of graph obtained from hardware measurement, within the frequency band of 2.2 GHz to 3.3 GHz, as shown in Table IV and Figures 7,8, and 9.

C. Dielectric Properties of three types agricultural waste material boards with the effect of different sample distances

TABLE IV.

This part of the investigation was carried out to identify relationships between the dielectric properties and the three different distances established between horn antennas to the MUTs. Table III shows the variations of dielectric constant, ε’ and loss factor, ε” versus resin percentage (UF and PF) between the frequency range of 2.2GHz to 3.3GHz, at room temperature (27ºC). As can be seen in the table III, the dielectric constant, ε’ values dropped constantly from an initial measurement distance (215 mm between horn antenna to agricultural waste sample board) compared to larger measurement distances of 377 mm as well as measurement distances at 475.5 mm.

and

MEASUREMENT AND SIMULATION OF KENAF AT 10%, 30%, AND 50%PF S- Parameter

PF% (measurement)

(simulation)

(measurement)

(simulation)

-12.129 -12.127 -10.504

-13.680 -16.344 -15.694

-0.100 0.000 -0.013

-0.861 -0.188 -0.439

10 30 50

TABLE III. DIELECTRIC PROPERTIES OF THREE TYPES AGRICULTURAL WASTE MATERIAL BOARDS WITH DIFFERENT SAMPLE DISTANCE Materials

Rice Husk Rice Straw Kenaf

Resin volume (%) 10 30 50 10 30 50 10 30 50

Sample Distance (mm) 215

377

475.5

ε’

ε”

ε’

ε”

ε’

ε”

3.236 3.581 4.558 1.906 2.012 2.736 2.101 2.141 2.382

0.274 0.439 0.216 0.135 0.123 0.224 0.054 0.069 0.133

3.139 3.473 4.491 1.891 1.974 2.639 2.029 2.109 2.310

0.477 0.617 0.484 0.215 0.211 0.372 0.124 0.168 0.240

2.885 3.249 4.031 1.814 1.907 2.485 2.003 2.029 2.103

0.496 0.651 0.539 0.246 0.234 0.392 0.1278 0.284 0.210

The decreasing trend of dielectric constant was found when all the three type agricultural waste material boards (rice husk, rice straw, and kenaf) were placed further from the horn

Figure 7:

and

measurement and simulation of kenaf 10% PF

Figure 8:

and



186

Malaysian Ministry of Science, Technology and Innovation (MOSTI) (e-ScienceFund Grant 9005-00016) which enabled the production of this article. VI.

REFERENCES

[1]

Figure 9:

and


However, some uncertainty was detected between and when comparing between hardware result and software simulation, which is typically less than 50%. This can be seen in Table V. This was caused due to several uncertainty factors such as instrumentation, dimensional and geometrical uncertainty of MUTs, roughness and conductivity of the conduction surfaces [15]. Additional variations may also have been caused by the systematic uncertainties in the particular instrumentation, calibration standards and the dimensional imperfections of the implemented test fixture [16-18]. IV.

CONCLUSION

The Free Space Measurement was found to be a suitable method in determining the dielectric properties, especially when a non-destructive measurement of flat materials is desired. This method is simple, fast and offers superior accuracy, as materials were placed between antennas for a non-contacting measurement. The free space method works best for large flat solid materials, while granular and powdered materials can also be measured in a fixture. Calibration and gating techniques performed in the network analyzer can be used to reduce measurement errors. Dielectric constant, loss factor, and S-parameters were determined and tabulated for rice husk, rice straw, and kenaf boards at specified moisture contents or bonding structure of resins for the frequency range from 2.2 GHz to 3.3 GHz at room temperature of 27 ºC. From the result obtained, rice husk was found to possess a high dielectric constant compared to rice straw and kenaf, due to the natural properties of rice husk. The large surface area of rice straw has provided the ability to absorb more electromagnetic signal. V.

M.S. Venkatesh and G.S.V. Raghavan, “An overview of dielectric properties measuring techniques”, Canadian Biosystems Engineering, 2005 [2] B.-K. Chung, “Dielectric Constant Measurement For Thin Material At Microwave Frequencies,” Progress In Electromagnetics Research, PIER 75, 239–252, 2007. [3] Agilent Technologies, “Basic of Measuring the Dielectric Properties Meterials – Application notes”, 26 June 2006 [4] C.A. Balanis, “Antenna Theory: Analysis and Design”, 3rd Edition, Wiley Interscience, 2005 [5] Deepak Kumar Ghodgaonkar, Nor Azlin Ali,”Microwave Nondestructive Testing of Composite Materials using Free-Space Microwave Measurement Techniques.” 15th World Conference on Non Destructive Testing, Roma (Italy), 15-21 October 2000. [6] Suhardy Daud, Mohd Nazry Salleh, Farizul Hafiz Kasim and Saiful Azhar Saad, “Analysis of Chemical Elements in Major Perlis’s Agricultural Residue”, School of Materials Engineering, Kolej Universiti Kejuruteraan Utara Malaysia (KUKUM,2006) [7] Samir Trabelsi and Stuart O Nelson, “Free-Space Measurement Of Dielectric Properties Of Cereal Grain And Oilseed At Microwave Frequencies,” Meas. Sci. Technol. 14 (2003) 589–600 [8] Agilent Technologies, “Agilent 85071E Materials Measurement Software – Technical Overview”, 27 June 2006 [9] L.T. Chew, C.L. Ong, “Urea Formaldehyde Particle Boards From Yemane”, Forest Research Institute Malaysia, Kepong, 52109 Kuala Lumpur, Malaysia. [10] Russell Hornung “ Insertion Loss and Loss tangent” Arlon, Magnetic for electronic division. [11] T. Jayanthy, P.E. SankaranarayananMeasurement of Dry Rubber Content in Latex Using Microwave Technique,” Measurement Science Review, Volume 5, Section 3, 2005 [12] V. V. Navarkhele, S.T. Nakade and A.A. Shaikh, “A dielectric approach to determine water content in soil using microwave transmission technique,” J. Indian Inst. Sci., Nov.–Dec. 2006, 86, 723–729 [13] G. Ayalew, N.M. Holden and S.M. Ward,”Microwave dielectric properties of horticultural peat products” Mires and Peat, Volume 2, Article 03, 2007 [14] David M. Pozar, “Microwave Engineering”, 3rd Ed., Wiley, 2005 [15] M. Tabassum Afzal, Bruce Colpitts, Kurt Galik, “Dielectric Properties of Softwood Species Measured with an Open-ended Coaxial Probe”, 8th International IUFRO Wood Drying Conference – 2003 [16] Kamal Sarabandi and Fawwaz T. Ulaby “Technique for Measuring the Dielectric Constant of Thin Materials”, IEEE Transactions On Microwave Theory and techniques,VOL. 31, NO. 4, December 1988 [17] P. Kumar, P. Coronel, J. Simuvic, V.D. Truong, And K.P. Sandeep “Measurement of Dielectric Properties of Pumpable Food Materials under Static and Continuous Flow Conditions” Food Engineering and Physical Properties [18] J. Wang, T. Schmugge, and D. Williams “Dielectric Constant Of Soils At Microwave Frequencies “NASA Technical Paper 1238

ACKNOWLEDGEMENT

The authors would like to acknowledge Universiti Malaysia Perlis (Short Term Grant 9001-00110) and


187