Microwave Dielectric Properties of Glass-Reinforced Polymers

http://www.e-polymers.org e-Polymers 2005, no. 004. ISSN 1618-7229

Short communication: Microwave dielectric properties of glass-reinforced polymers

Luís Cadillon Costa 1 *, Susana Devesa 1, Paulo André 1,2

1 Physics Department, University of Aveiro, 3810-193 Aveiro, Portugal; [email protected] 2 Telecommunications Institute, 3810-193 Aveiro, Portugal; [email protected]

(Received: October 15, 2004; published: January 21, 2005)

This work has been presented at the 12th Annual POLYCHAR World Forum on Advanced Materials, January 6-9, 2004, in Guimaraes, Portugal

Abstract: The well-balanced combination of properties of some polymers makes them good materials for industrial microwave applications, such as telecommunications or microwave ovens. Particular electrical properties are usually needed and can be controlled by additives. In this work, we present the results of complex dielectric permittivity measurements, ε* = ε’ - iε’’, in the microwave frequency region (2.7 and 12.8 GHz), at constant temperature 300 K, on different glass-reinforced and pigmented plastics (poly(butylene terephthalate), polypropylene and acrylonitrile-butadiene-styrene), using the resonant cavity method. We measured the shift in the resonant frequency of the cavity, ∆f, caused by the insertion of the sample, which can be related to the real part of complex permittivity, ε’, while the change in the inverse of the quality factor of the cavity, ∆(1/Q), gives the imaginary part, ε’’.

1. Introduction A very versatile family of thermoplastics includes acrylonitrile-butadiene-styrene (ABS), polypropylene (PP) and poly(butylene terephthalate) (PBT). They present interesting properties, as low thermal expansion coefficient, high dielectric strength, and very low dielectric losses, and have good resistance to chemical attack [1-3]. The addition of particulate fillers has been used for a long time in polymer industry, primarily to give mechanical reinforcement and then to control electrical properties [4]. In general, glass fibre reinforced polymers show an important increase of the performance of mechanical properties. Carbon black is also often applied as an additive [5], because it is a compatible material, mixing and adhering rather well with the matrix, and it is a cheap material. It is used as a conductivity controller, properly changing the doping quantities. For small quantities of carbon black particles no percolation threshold is observed and the composite is maintained as an insulator [6] but with higher dielectric constant, 1 which can be important for industrial applications. Another conducting particle frequently used is polypyrrole, because it can be easily introduced in a polymeric matrix and the critical percolation concentration is well controlled [7]. A crucial part of a microwave oven is the door, the purpose of which is to provide access to the oven space and to confine energy inside it, and where the microwave leakage must be quite controlled. In a non-contacting configuration, a spacer between the door and the oven flange is used to create a gap [8]. Providing a cavity having a terminating surface prevents the escape of energy through this gap. Filling that cavity with a polymer prevents entrance of particles and reduces the dimensions of the choke cavity by the square root of the dielectric constant of the filler [9]. Dielectric antennas can also use polymer materials to reduce the secondary lobes, and simultaneously to increase the main lobe. Frequently, this is made using the dielectric in the form of a rod in the horn, improving matching [10]. To avoid appre- ciable dielectric losses, maintaining high values of dielectric constant, the electrical properties must be carefully controlled, eventually adding different concentrations of doping charges.

2. Experimental part The studied materials are acrylonitrile-butadiene-styrene (ABS), polypropylene (PP) and poly(butylene terephthalate) (PBT), reinforced with 15% of fibreglass, and pigmented with 1.5% of carbon black particles. Glass fibres are manufactured in the range 10 to 25 µm, and could be observed in a PBT matrix using scanning electron microscopy (SEM) as shown in Fig. 1. The samples used were in the form of cylinders of length 3 and 5 mm, and diameter 0.8 and 1 mm.

Fig. 1. Scanning electron micrograph of glass-reinforced PBT

2 To calculate the complex permittivity, we have used the resonant cavity method, with two cavities operating at 2.7 and 12.8 GHz. The lower frequency was chosen because the obtained variations were rather accurate and it was the closest to 2.45 GHz, the licensed frequency for use in microwave ovens, and the higher frequency is important for telecommunication applications. We measured the cavity transmission using HP 8753D and HP E8361A Network Analysers. When we introduce a material inside the cavity, a perturbation of the electric field is observed, and consequently a different transmission is obtained. This is shown, as an example, in Fig. 2.

60 ∆ f

50 empty u. . 40 a on / i

s loaded

s 30 i m 1/Q

ans 0 r 20 T 1/Q l

0 f f 12700 l 12750 0 12800 Frequency / MHz Fig. 2. Perturbation of the cavity transmission by the insertion of a sample (12.8 GHz)

If we consider only the first-order perturbation caused by the sample, the relations between the shifts in the resonant frequency of the cavity, ∆f, and in the inverse of the quality factor of the cavity, ∆(1/Q), with the complex permittivity are simple [11], ∆f ()ε' − 1 v = K V (1) f0 K  1  ε' ' v = ∆ V (2) 2 Q  where K is a constant related to the depolarisation factor, which depends upon the geometric parameters, and  1  1 1 ∆  = − (3) Q  Ql Q0

∆f = f0 − fl (4) where the indices 0 and l refer to the empty and loaded cavity, respectively. Using a sample of known complex permittivity we can determine the depolarising factor. In our case we used a polytetrafluoroethylene (PTFE) sample, with the same size and shape as the other samples.

3 3. Discussion In Fig. 3 we present the transmission in one of the cavities when a PBT sample was inserted. The experimental data and the Lorentzian fit used to calculate the resonant frequency and the quality factor are shown. The obtained correlation factor was 0.989. In Fig. 4 we show the measured transmission, for the empty cavity, and filled with PTFE and with the different polymers, at a constant temperature of 300 K. PBT is the most perturbating sample, which indicates higher complex permittivity.

2x10-2 experimental data Lorentzian fit ) . u

. -2 a 1x10 ( n o i s s i m s

n -3

a 5x10 Tr

2.780x109 2.784x109 Frequency (Hz) Fig. 3. Transmission in the 2.7 GHz cavity, with experimental data and Lorentzian fit

PP ABS PTFE -2 1,6x10 empty PBT

-2 u.

. 1,2x10 a

on / 8,0x10-3 ansmissi r T 4,0x10-3

0,0 2,780x109 2,784x109 2,788x109 Frequency / Hz Fig. 4. Transmission of the 2.7 GHz cavity, for the empty cavity, and filled with PTFE and with the different polymers

Tab. 1 resumes the calculated ε’ and ε’’ values, for the different polymers, at both measured frequencies. As we can observe, PP has the lowest values for the real and 4 imaginary parts of the complex permittivity. This fact can be explained because PP is essentially a non-polar polymer, and consequently should not exhibit dielectric relax- ation [12], and dielectric constant and loss factors are small at all frequencies.

Tab. 1. ε’ and ε’’ for PP, ABS and PBT, at two different frequencies, and constant temperature 300 K 2.45 GHz 12.8 GHz ε’ ε’’ (10-4) ε’ ε’’ (10-3)

PP 2.46 11 2.40 11 ABS 2.96 23 2.73 53 PBT 3.68 45 3.24 109

ABS, in which the monomers have different contributions for the physical properties, presents higher dielectric constant and loss factor. The conductivity increases with increasing ε’. The interpretation of this relationship is that if more charge carriers are present then conductivity increases [13]. A small polarity is also present in PBT, explaining higher complex permittivity. The obtained values make then a good choice for applications in microwave ovens and telecommunications products. The higher ε’ permits to reduce the dimensions of the chokes in the oven doors and in the rods in dielectric antennas. Despite the higher ε’’, the values are still quite low, avoiding the heating of the material.

4. Concluding remarks The resonant cavity technique has proved very useful for dielectrics measurements on polymers. Under the chosen conditions, the calculation of complex permittivity is quite simple. An additional advantage is the ease of sample preparation, and the absence of contacts to measure, which provides a good accuracy. A principal disadvantage is the need for a calibration material to calculate the depolarisation factor. Glass-reinforced PBT with carbon black as additive is the most suited amongst the studied polymers to be used as filler to the chokes in oven doors and to dielectric telecommunications antennas.

[1] Johnson, C.; Hilton, G.; “Acrylonitrile-Butadiene-Styrenes”, Engineered Materials Handbook, AMS International, 1988, vol. 2, 107. [2] “Handbook of Chemistry and Physics”, ‘Polypropylene’, CRC Press, Cleveland 1975, pp. C788. [3] Vishn, S.; “Handbook of Plastics Testing Technologies”, Wiley Intersciences, New York 1984, p. 334. [4] O'Brien, T. K.; Chawan, A. D.; DeMarco, K.; J. Comp. Tech. Res. 2003, 25, 3. [5] Krupa, I.; Chodak, I.; Eur. Polym. J. 2001, 37, 2159.

5 [6] Valente, M. A.; Costa, L. C.; Mendiratta, S. K.; Henry, F.; Ramanitra, L.; Sol. Stat. Commun. 1999, 112, 67. [7] Costa, L. C.; Henry, F.; Valente, M. A.; Mendiratta, S. K.; Sombra, A. S.; Eur. Polym. J. 2002, 38, 1495. [8] Decareau, R. V.; Peterson, R. A.; “Microwave processing and engineering”, VHC, London 1986, p. 141. [9] Copson, D. A.; “Microwave heating”, AVI Publishing Company, Connecticut 1975, p. 439. [10] Williams, H. P.; “Antenna theory and design”, Sir Isaac Pitman and sons, London 1950, 190. [11] Henry, F.; PhD Thesis, France 1982. [12] McCrum, N.; Read, B.; Williams, G.; “Anelastic dielectric effects in polymer solids”, Dover Pub. Inc., New York 1991, p. 123. [13] Saito, S.; Macromolecules 1968, 17, 672.