Dielectric properties of oil palm-natural rubber biocomposites

2007 Annual Report Conference on Electrical Insulation and Dielectric Phenomena

Dielectric properties of oil palm-natural rubber biocomposites M. Marzinotto, C. Santulli, C. Mazzetti Electrical Engineering Department - Sapienza University of Roma, ITALY applications with a environmentally friendly end-of-life scenario. This can be considered an ambitious, albeit to be pursued, objective for studies such as the present one.

Abstract- In recent years, low cost composites made from plant oil sources proved of interest for dielectric applications, showing also some potential for future application as dielectric in microchips and circuit boards. In this work, the dielectric properties of oil palm/natural rubber composite materials were characterized. The basic material was formed with pure rubber. This was modified in different ways, by adding 20 or 30% wt. of untreated oil palm fibres, in the shape of 6 mm long macrofibers, or by treating the fibers with silane. Dielectric properties of the modified materials with respect to the original one were measured. In all cases, the effect due to the rubber modification on the dielectric properties was noticeable. In particular, a clear increase of the loss factor was obtained with fiber treatment. In contrast, the effect of the lower level of fiber content used (20%) on the dielectric properties was less evident.

II. SPECIMENS AND EXPERIMENTAL MEASUREMENTS Four type of rubber compounds have been tested in this study, in the following simply referred to as A, B, C and D. They are all natural rubber based compound (cis-polyisoprene from Hevea Brasiliensis), whose compositions are reported in Table I. Specimens B,C and D are also reinforced with fibers obtained from the leaves of oil palm (Elaeis Guineensis), a plant originally cropped in West Africa, but diffused also in some parts of India (Kerala), from where the fibers used in this study come from. TABLE I COMPOSITION OF THE RUBBER COMPOUNDS TESTED IN THIS STUDY

I. INTRODUCTION

Compound

In the more general field of bio-dielectrics, dielectric materials including natural fibres have recently attracted some interest [1]. These materials can also be easily intended for multi-function applications, as the hollow cellular structure of plant fibers proved effective in providing insulation against heat and noise. In addition, bio-composites present a more advantageous end-of-life scenario, as assessed from Life Cycle Analysis (LCA) studies, especially if little chemical tretament is required to have sufficient material properties for the envisaged application [2]. The use of plant fibers in them would represent an added value for developing countries, in that a local material, often an agricultural waste with inherently low-cost, can be employed for significant engineering purposes [3]. Some studies on the electrical properties of these materials suggested that the values of dielectric constant and volume resistivity in the material are somehow affected by the volume of reinforcement fibers introduced on dielectric constant [4]. In particular, it has been noted as in general the dielectric constant progressively increases with fiber loading and decreases with higher frequency [5]. In contrast, the effect of fibre treatment on dielectric properties has not yet been clearly assessed. It is noteworthy that treatment can be required to improve the mechanical and durability properties of natural fibers, and a number of treatments are available for this purposes [6]. Knowing the effect of chemical and physical modifications on the dielectric properties of the material would in the long run imply offering dielectric materials for electronic

1-4244-1482-2/07/$25.00 ©2007 IEEE

Composition

A

Pure rubber

B

80% rubber; 20% oil palm fibers (untreated)

C

70% rubber; 30% oil palm fibers (untreated)

D

70% rubber; 30% oil palm fibers (treated with 5% of silane solution)

Specimens in form of 2.1 mm thickness foils have been used to perform the measurements of AC conductivity σac versus frequency in accordance with [7,8]. In Fig. 1 the tubular electrode set-up (guarded, guards and unguarded electrodes) used for the measurements is reported. With the same set-up, the real and the imaginary parts of the dielectric constant, permittivity ε’ and loss factor ε’’ have been also registered. A digital RLC meter (50 Hz ÷ 100 kHz) – Fluke PM6304 has been used for the measurements of the above mentioned quantities. The measurements have been performed applying a voltage of 2 V rms between the electrodes no. 1 (guarded electrode) and electrode no. 3 (unguarded electrode) (Fig. 1). Electrode no. 2 (guard electrode) has been grounded [7]. Dissipation factor tan δ and power loss density pac have been calculated through the measurements of permittivity and loss factor as: tan δ =

and

584

ε ′′ ε′

(1)

p ac

⎛ ⎛r ω ⋅ ε ′′ ⋅ Co′ ⋅ ⎜⎜ E max ⋅ r1 ⋅ ln⎜⎜ 2 P ⎝ r1 ⎝ = ac = 2 2 ψ π ⋅ (r2 − r1 )

⎞⎞ ⎟⎟ ⎟ ⎟ ⎠⎠

constant at frequencies exceeding 1 kHz, whilst the introduction of more fibers (sample C) or the silane treatment increased it with respect to pure rubber in the whole frequency range considered (Fig. 2). From Figs 3, 4 and 5, referred to dielectric losses in terms of loss factor ε’’, dissipation factor tanδ and power loss density pac, it appears that the properties of a material including a larger amount (30%) of silane-treated fibers are very similar to those of a material with less (20%) untreated fibers. In other words, fiber treatment, which is intended to increase the fiber stiffness, reduced also considerably the losses. For as regards AC conductivity σac values (Fig. 6), the effect of the presence of oil palm fibres appears to prevail on both that of the quantity introduced and of the silane treatment, so that the values measured for the non-pure rubber samples are much closer between them than with the pure rubber sample. Here below some explanation is attempted of the observed results. A major difficulty in evaluating the dielectric properties of materials including plant fibers is that these strongly depend on fibre orientation, as suggested in [9] for sisal fiber reinforced composites. Fibers considered in the present study were inserted with statistical random orientation and therefore there might be significant variations of the dielectric properties from region to region of the materials on study, depending on the local orientation of fibers.

2

(2)

where • ω is the angular frequency of the applied voltage (2πf); • Co’ is the capacity in per unit length between the electrodes (guarded and unguarded) without the specimens (εr = 1); • Emax is the electric field at the unguarded electrode (electrode no. 3 in Fig. 1); • ψ is the specimen volume in per unit length between electrodes 1 and 3 (Fig. 1); • r1 is the inner radius of the electrode no. 3 (unguarded electrode); • r2 is the outer radius of the electrode no. 1 (guarded electrode).

10 9 8

Sample C

7

permettivity ε′

6 Sample D 5 4

3 Sample A Sample B 2

2

10

Fig. 1. Tubular electrode system used for the measurements (from [7]). Electrode no. 1: guarded electrode; electrode no. 2: guard electrode; electrode no. 3: unguarded electrode.

3

10 frequency [Hz]

4

10

5

10

Fig. 2. Permittivity ε’ vs. frequency for the four considered compounds.

A second possible effect is that of water absorption in the samples, which can be predicted to increase the conductivity of the samples [10]. It is also possible to suggest that treated fibers, having a better sizing with lesser defects, absorb less water and therefore have reduced losses with respect to the same material with the same amount of untreated fibers [11]. Some other aspects remain uncovered, which may be treated in further studies: in particular, the amount of chemical used for treatment may obviously increase the effect of reduced losses. In addition, the limited variation of the dielectric constant with respect to pure rubber observed in the material with the lower fiber content would suggest that the introduction of a very small amount of fibers may virtually

All tests have been performed at High Voltage Laboratory of the Sapienza University of Roma. III. RESULTS The discussion of the results obtained was aimed at clarifying two effects on the rubber dielectric material: the presence and the amount of plant fibers introduced (sample B and C), and the effect of chemical treatment (silane: sample D) on the fibers themselves. It is worthy to note that the introduction of a small amount of oil palm fibers, such as in sample B, reduced the dielectric

585

leave the dielectric properties unaffected, whilst still yielding considerable benefits in terms of lower cost and reduced environmental impact.

0

10

Sample C 1

-1

10

10 power loss density [W/m3]

Sample D Sample C Sample B 0

loss factor ε′′

10

Sample D

-2

10

-1

10

Sample B

-3

10

Sample A

-4

10

2

10 Sample A 2

10

3

10 frequency [Hz]

4

10

4

10

5

10

Fig. 5. Power loss density pac vs. frequency for the four considered compounds.

-2

10

3

10 frequency [Hz]

5

10

Fig. 3. Loss factor ε’’ vs. frequency for the four considered compounds.

-6

10

Sample C Sample B

0.35 -7

AC conductivity σ AC [S/m]

10 0.3

dissipation factor tanδ

0.25

Sample D

Sample C

0.2

0.15

Sample D

-8

10

-9

10

Sample A

Sample B 0.1

-10

10 Sample A

0.05

0

2

10

2

10

3

10 frequency [Hz]

4

10

3

10 frequency [Hz]

4

10

5

10

Fig. 6. AC conductivity σac vs. frequency for the four considered compounds.

5

10

Fig. 4. Dissipation factor tan δ vs. frequency for the four considered compounds.

ACKNOWLEDGMENT This work was supported in part by ISPESL (Istituto Superiore per la Prevenzione e la Sicurezza del Lavoro) Italian Focal Point in the information network of the European Agency for Safety and Health at Work – www.ispesl.it.

IV. CONCLUSIONS The effect of the introduction of oil palm fibers and of their chemical treatment with silanes on the dielectric properties of pure rubber has been investigated, as a part of a study on the application of bio-composites in this field. The effects obtained were in the sense of a clear increase of loss factors with both fibers introduction and chemical treatment. However, a number of other factors, such as e.g., fiber orientation and quantity of chemical used, are still to be assessed, which would be the object of further studies.

REFERENCES [1]

C.K Hong and R.P. Wool, “Low Dielectric Constant Material from Hollow Fibres and Plant Oil” J Nat Fibres, Vol. 1, pp.83-92, Oct. 2004. [2] S.V. joshi, L.T. Drzal, A.K. Mohanty, S. Arora, “Are natural fiber composites environmentally superior to glass fiber reinforced composites?” Compos. Part A Vol.35, pp.370-375, Apr. 2004. [3] P. Kandachar, R. Brouwer, “Applications of Bio-Composites in Industrial Products”, Materials Research Society Symposium–– Proceedings, 2002. [4] M. Jacob, K.T. Varughese, and S. Thomas, “Dielectric characteristics of sisal-oil palm hybrid biofibre reinforced natural rubber biocomposites”, J. Mat. Sci. Vol.41, pp.5538-5547, Sep. 2006.

586

[5]

A. Paul, and S. Thomas, “Electrical properties of natural-fiberreinforced low density polyethylene composites: A comparison with carbon black and glass-fiber filled low density polyethylene composites”, J. Appl. Polym. Sci., Vol.63, pp.247-266, Jan. 1997. [6] X. Li and L.G. Tabil LG, and S. Panigrahi, “Chemical treatments of natural fiber-reinforced composites: a review”, J. Polym. Environm. Vol.15, pp.25-33, Jan. 2007. [7] IEC 60093, “Methods of test for volume resistivity and surface resistivity of solid electrical insulating materials”, 1980. [8] M. Lisowski, R. Kacprzyk, “Changes proposed for the IEC 60093 Standard concerning measurements of the volume and surface resistivities of electrical insulating materials”, IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 13, No.1, pp. 139-145, 2006. [9] N. Chand, and D. Jain, “Effect of sisal fibre orientation on electrical properties of sisal fibre reinforced epoxy composites”, Compos. Part A, vol.36, pp.594-602, May 2005. [10] A.N. Fraga, E. Frulloni, O. de la Osa, J.M. Kenny, and A. Vazquez, “Relationship between water absorption and dielectric behaviour of natural fibre composite materials”, Polym. Testing, Vol.25, pp.181-187, Apr. 2006. [11] V.P. Cyras, C. Vallo, J.M. Kenny, and A. Vazquez, “Effect of chemical treatment on the mechanical properties of starch-based blends reinforced with sisal fibre”, J. Compos. Mat., Vol.38, pp.1387-1399, 2004.

587