Electrical Conductivity of Urea–Formaldehyde–CelluloseComposites Loaded with Copper

Gabriel Pinto,1 Abdel-Karim Maaroufi,2 Rosario Benavente,3 Jose M. Perena31Departamento de Ingenierıa Quımica Industrial y del Medio Ambiente, E.T.S.I. Industriales, UniversidadPolitecnica de Madrid, 28006 Madrid, Spain

2Laboratory of Composite Materials, Polymers and Environment, Department of Chemistry, Faculty ofSciences, P.B. 1014, Rabat Agdal, Morocco

3Instituto de Ciencia y Tecnologıa de Polımeros, Consejo Superior de Investigaciones Cientıficas (CSIC),Juan de la Cierva, 3, 28006 Madrid, Spain

This work is concerned with the preparation and char-acterization of composite materials prepared by com-pression molding of mixtures of copper powder and acommercial grade thermosetting resin of urea–formal-dehyde filled with a-cellulose in powder form. The elec-trical conductivity of the composites is <10212 S/cm,unless the metal content reaches the percolationthreshold of 24.0 vol %, beyond which the conductivityincreases markedly by as much as 11 orders of magni-tude, indicating an insulator–conductor phase transi-tion. The homogeneity of these composites waschecked by the morphologies of the constituents (fillerand matrix) and the composites characterized by opti-cal microscopy. The density of the composites wasmeasured and compared with values calculatedassuming different void levels within the samples todiscuss the porosity effect. Finally, the obtained resultson electrical conductivity have been well interpretedwith the statistical percolation theory. The deducedcritical parameters, such as the threshold of percola-tion, Vf*, the critical exponent, t, and the packing den-sity coefficient, F, were in good accord with earlierstudies. In addition, the hardness of samples remainedalmost constant with the increase of metal concentra-tion. POLYM. COMPOS., 32:193–198, 2011. ª 2010 Society ofPlastics Engineers

INTRODUCTION

Various methods of manufacture of composites of

polymers containing dispersed conductive fillers, as well

as information about their properties, have been reportedwidely in the literature for the last two decades [1–8], due

to their numerous technological applications in a variety

of areas. As it is well known, most polymers are ther-

mally and electrically insulating. The increase of thermaland electrical conductivities of polymers opens large new

markets. For example, King et al. [9] pointed out that theadvantages of conductive polymer composites when com-

pared with typically used metals includes improved corro-

sion resistance, lighter weight, and the ability to adapt theconductivity properties to suit the application needs.

Common technological applications of electrically

conductive resins are concerned with areas such as elec-

tromagnetic (EMI)/radio frequency interference shielding

for electronic devices (computer and cellular housings

for example), self regulating-heaters, overcurrent protec-

tion devices, photothermal optical recording, direction

finding antennas, chemical detecting sensors used in

electronic noses, and more [10–13]. The used fillers are

metal particles, metal coated particles, or carbon par-

ticles with different sizes (10 nm to some hundreds of

micrometers).

It is known that, in general, the percolation theory is

used to describe the electrical conductivity of conductive

polymer composites. Hence, the electrical conductivity for

polymer composites does not increase continuously with

increasing electroconductive filler content, but there is a

critical composition (percolation threshold) at which the

conductivity increases by some orders of magnitude from

the insulating range to values in the semiconductive or

conductive range [14]. For efficiency, to decrease the dif-

ficulty of the process and economic costs, the amount of

the conductive phase for achieving materials with high

conductivity should be usually as small as possible. A

huge number of different models have been proposed for

the estimation of the conductivity (or inverse resistivity)

versus filler concentration curves [15–18].

Correspondence to: Gabriel Pinto; e-mail: gabriel.pinto@upm.es

DOI 10.1002/pc.21032

Published online in Wiley Online Library (wileyonlinelibrary.com).

VVC 2010 Society of Plastics Engineers

POLYMER COMPOSITES—-2011

In the last few years, carbon nanofibers and nanotubes/

polymer conductive composites have been described as el-

egant materials that exhibit superior electrical, EMI

shielding effectiveness, and thermal properties compared

to conventional polymer composites [19, 20].

This article deals with further developments in previ-

ous reported investigations of preparation and characteri-

zation of electroconductive polymer composites [21–29].

We report an experimental study about the influence of

filler concentration on the electrical conductivity of com-

posites elaborated by hot compacting, i.e. compression

molding of mixtures of copper powder and urea–formal-

dehyde embedded in a-cellulose powder. Short fibers of

a-cellulose are usually used as reinforcing filler in urea–

formaldehyde-molding compounds. The homogeneity of

composites was controlled by morphological pictures

obtained by optical microscopy. Furthermore, to check

the void level within the samples, which influences

remarkably the electroconductivity, the porosity rate has

been calculated from densities of the composites. These

data, along with the reported previously [21–29], may be

helpful in developing theoretical models to better under-

stand the variation of electrical properties of such polymer

composites.

As in our previous works that are cited, to complete

the characterization of these materials, we also studied

the influence of filler concentration on the hardness of the

composites, as an example of mechanical property.

EXPERIMENTAL

Materials

The only matrix polymer used in our experiments was

a commercial grade urea–formaldehyde embedded in a-cellulose supplied in the form of powder by Aicar S.A.

with a density of 1.36 g/cm3 and an electrical conductiv-

ity of around 10213 S/cm. The content of a-cellulose in

the resin, according to the manufacturer, is 30 wt %. A

micrograph of this powder is shown in Figure 1, where

the longitudinal shape of particles can be observed.

The electrical conducting filler used was copper, deliv-

ered by Panreac with a purity around 99.9%, average par-

ticle size of 150–200 lm, density of �8.92 g/cm3, and

electrical conductivity, taken as the tabulated value [30]

of the order of 6.3 3 105 S/cm. The shape of the particles

of filler is illustrated in Figure 2.

Both the polymer and the metal powders were thor-

oughly dried before use at 608C during 48 h.

Composite Fabrication

Composites of urea–formaldehyde embedded in a-cel-lulose powder filled with copper were fabricated by mix-

ing the polymer matrix and the filler powders for 2 h in

an internal mixer, followed by compression molding in a

specially designed mold with three cavities of 30.0-mm

diameter and 3.0-mm thickness each one. The molding

parameters were 20 MPa and 1508C for 30 min. These

fabrication conditions were suggested by our previous ex-

perience with the same matrix [26–28]. Samples were

cooled as much as room temperature in �30 min.

Samples with filler contents on the range 0–80 wt %

(corresponding on the range 0–0.38 in volume fraction)

were prepared. To improve the finish of the sample and

ensure a better electrical contact for resistance measure-

ments, the surfaces were polished with sandpaper. Sample

thickness (necessary for the calculation of electrical con-

ductivity) was determined using a micrometer, Schmidt

Technology model J 50, to an accuracy of 0.01 mm.

Thickness measurements were taken at five locations and

averaged.

FIG. 1. Optical microscopy micrograph of the urea-formaldehyde em-

bedded in cellulose powder used as matrix. [Color figure can be viewed

in the online issue, which is available at wileyonlinelibrary.com.]

FIG. 2. Optical microscopy micrograph of copper powder used as filler.

[Color figure can be viewed in the online issue, which is available at

wileyonlinelibrary.com.]

194 POLYMER COMPOSITES—-2011 DOI 10.1002/pc

Composite Characterization Techniques

The microstructures of the powders used in the prepa-

ration of composites and of samples were observed by

reflection by means of a Nikon model Eclipse E200 opti-

cal microscope.

The electrical conductivity was determined through the

electrical resistance values that were measured using a

two-point arrangement as described elsewhere [22–26].

Three specimens of each composition were tested, taking

five data points on each sample. To decrease the contact

resistance, the sample surfaces were coated with silver

paint and dried during 24 h.

Measurements of volume electrical resistance higher

than 103 O were made, at 238C, using a programmable

megohmeter (Quadtech model 1865). Measurements of

low resistance (lower than 103 ohm) were made using a

digital multimeter (Leader model 856). A constant voltage

of 100 V was supplied to the samples, and the resistance

of the samples was measured after one minute, using a

test cycle consisting of 20-s charge, 20-s dwell, 20-s mea-

sure, and 20-s discharge. Before starting a new test, the

electrodes were short-circuited for 5 min to eliminate any

effect of the previous electrification. The procedure used

in this study to estimate the electrical conductivity, r,from electrical resistance, was similar to that reported ear-

lier [26–28].

The density of the composites was measured in ac-

cordance with ASTM D 792-91, by difference of weight

in the air or with the sample immersed in water as the

liquid of known density, at 238C, using a Mettler Toledo

AJ 100 balance equipped with a density determination

kit.

The hardness of the samples was determined at 238Cusing a Durotronic Instron model 1000 Shore D hardness

tester, in accord with ASTM D 2240-68. Five data points

were taken on each samples, and no difference was found

between hardness measurements on both faces of each

specimen.

RESULTS AND DISCUSSION

Figure 3 represents micrographs with the structure of a

composite sample before (Fig. 3a) and after (Fig. 3b) the

percolation threshold (24.0% v/v), which corresponds to

the insulating-conductive phase transition. These photos

show a distinction in contrast related to the different color

of the filler and matrix. The morphology of samples

remains similar, and the filler is uniformly dispersed, indi-

cating homogeneous composites.

The homogeneity of composites also was verified by

density measurements. As in previous studies with the

same matrix [26–28], this constitutes a confirmation of mi-

croscopic observation that the produced composites are

almost homogeneous, with a presence of air negligible.

Thus, by proceeding as explained in previous works [22–

25], where more detail was included, and by comparison

between experimental and theoretical densities of samples,

we obtained the composites’ porosity, as a function of the

filler volume fraction, as shown in Figure 4. It is to be

noted that, after an initial increase of porosity with the

FIG. 3. Optical microscopy micrograph of the copper-filled urea-formaldehyde and cellulose composites containing

9.2 vol % of copper (a) and 31.4 vol % of copper (b). [Color figure can be viewed in the online issue, which is

available at wileyonlinelibrary.com.]

FIG. 4. Porosity rate versus copper volume fraction.

DOI 10.1002/pc POLYMER COMPOSITES—-2011 195

increasing of filler fraction, the average fraction voids in

volume for the samples is almost constant with a value of

11% 6 2%, from a value of filler content of around 0.12.

Therefore, the quality of the obtained composites was good.

The hardness of samples remains approximately con-

stant, as 82 6 4 Shore D values, independently of the fil-

ler composition, as found for other composites prepared

with the same matrix [26–28].

The electrical conductivity of the composites as a func-

tion of filler content for the samples shows the typical S-

shaped dependency with three regions (dielectric, transi-

tion, and conductive) (see Fig. 5). As expected, samples

with low-filler content are almost nonconductive. How-

ever, the electrical conductivity of the composites

increases dramatically as the copper content reaches the

percolation threshold at 24.0 vol % of filler. The value of

the percolation threshold is obtained from the maximum

of the derivative of the conductivity as a function of filler

volume fraction. According to Flandin et al. [4], values of

20–40% (v/v) are typical for spherical particles of filler.

Above the percolation threshold, the conductivity of com-

posite has increased by much 11 orders of magnitude.

This behavior could be explained with the statistical

percolation theory. Such theory is usually used to relate the

electrical conductivity of composite to the existence of

clusters of connected particles; which give rise to the so-

called conducting infinite cluster above the threshold. In

this theory, the relationship between the electrical conduc-

tivity of the mixture and the volume fraction of the con-

ductive filler is given by a power-law relationship [15]:

r ¼ r0ðVf � V�f Þt (1)

where r is the electrical conductivity of the mixture, r0 isthe electrical conductivity of the filler’s particles, Vf is the

volume fraction of the filler, Vf* is the critical volume

concentration at the threshold of percolation, and t is an

exponent determining the increase of the conductivity

above Vf*. This theory gives a good description of experi-

mental results near the transition point. Nevertheless, dis-

crepancies were observed between critical parameters

(Vf*, t) resulting from Eq. 1 and experimental values [17]:

inasmuch as the basic classical statistical theory does not

take into consideration of several parameters. Whilst, the

experimental results show that the electrical conductivity

depends strongly on the viscosity and the surface tension

of the filled polymers. It also depends on the geometrical

parameters of the filler particles as well as on the filler/

matrix interactions. Mamunya et al. [17, 18] have devel-

oped a model in which specific parameters for each com-

posite have been introduced in the basic theory:

r ¼ r0 þ rm � r0ð Þ : Vf � V�f

F� V�f

8>>:

9>>;

teff

(2)

where r0 is the electrical conductivity at the percolation

threshold, rm is the maximum conductivity of composite,

F is the filler packing density coefficient (equivalent to

the maximum value of the filler volume fraction), and teffis given by the relation:

teff ¼ t1 þ t2 (3)

where t1 is equivalent to the t parameter in the basic Eq.1, which takes a value around 1.7, and t2 depends on the

specific composite. Thus, teff could have higher values

taking into account of the filler/polymer interactions.

The Eq. 2 was used with success in earlier studies to

interpret the experimental results [25, 28]. Therefore, the

fit, above the percolation threshold of the electrical con-

ductivity as function of volume fraction of Cu filled in

urea–formaldehyde embedded in cellulose power, is given

in Figure 6. It should be noted that the agreement between

FIG. 5. Variation of the electrical conductivity of urea-formaldehyde

embedded in cellulose powder/Cu composites with Cu content.

FIG. 6. Electrical conductivity of Cu/urea-formaldehyde and cellulose

composites as function of Cu volume fraction above the percolation

threshold (n). Solid line is the fit with Eq. 2.

196 POLYMER COMPOSITES—-2011 DOI 10.1002/pc

the experiment and the theory is fairly good. The deduced

parameters are Vf* � 0.24, teff � 2.21, and F � 0.62.

The determined packing density coefficient F value is

in good agreement with the prediction of Eq. 2 [31]. The

teff obtained value is slightly close to two, which represents

the accepted theoretical value for three-dimensional latti-

ces [32, 33]. This theoretical value is independent of the

exact composition of the random composites [32]. On the

other hand, the critical threshold percolation value

obtained is in good agreement with that determined by ex-

perience (see Fig. 5). Elsewhere, this result is also close to

0.19 found in Sn and Zn filled urea–formaldehyde and cel-

lulose microcomposites [26, 28], where the fillers had a

similar shape and size of particles. Indeed, the electrical

conductivity of random composites has already been shown

to depend on several parameters [25, 26, 34, 35], such as

the viscosity and the surface tension of the polymers, espe-

cially in the case of the mixtures in which the conductive

powder is dispersed; the size, the shape, and the surface

energy of the filling particles and the powder dispersion pro-

cedure, that is, type, duration, and strength of shear.

CONCLUSIONS

In this article, we have described an experimental

study about the effects of the copper content on the elec-

trical conductivity of composites of a urea–formaldehyde–

cellulose resin filled with that metal. From the obtained

results, the following conclusions could be made:

1 The electrical conductivity of composites increases as

much as 11 orders of magnitude for a given range of

filler concentration, showing the typical percolation

transition from dielectric to conductive region of such

polymer composite materials.

2 The percolation threshold concentration corresponds to

a volume fraction of copper of 0.24, in good agreement

with previous experiments.

3 The relation between filler content and electrical con-

ductivity is fairly fitted with the extended basic statisti-

cal percolation theory. The deduced critical parameters

are reasonable and coherent with experimental values

and the earlier predictions.

4 The average fraction voids in volume increases with

the increase of filler concentration but, for values of fil-

ler concentration higher than 12 vol % it is almost con-

stant, with a value of 11% 6 2%.

5 The Shore D hardness remain approximately constant

(82 6 4 Shore D values) with the increase of filler con-

centration.

ACKNOWLEDGMENTS

This work was realized in the frame of the collabora-

tion of Centre National pour la Recherche Scientifique etTechnique (CNRST) at Rabat (Morocco) and ConsejoSuperior de Investigaciones Cientıficas (CSIC) at Madrid

(Spain), under project 2007MA0043. We thank AICAR

S.A. for furnishing us the urea–formaldehyde embedded

in cellulose powder used as matrix in the samples and for

technical support. The authors are also very grateful to

Prof. Marıa Jose Molina for help in the microscope inves-

tigations.

REFERENCES

1. M. Thakur, Macromolecules, 21, 661 (1988).

2. S.H. Son, H.J. Lee, Y.J. Park, and J.H. Kim, Polym. Int., 46,308 (1998).

3. J. Bouchet, C. Carrot, and J. Guillet, J. Polym. Eng. Sci.,40, 36 (2000).

4. L. Flandin, A. Chang, S. Nazarenko, A. Hiltner, and E.J.

Baer, J. Appl. Polym. Sci., 76, 894 (2000).

5. R. Gangopadhyay and D. Amitabha, Sensors Actuators B,77, 326 (2001).

6. W. Zhang, A.A. Dehghani-Sanij, and R.S. Blackburn, J.Mater. Sci., 42, 3408 (2007).

7. R.T. Fox, V. Wani, K.E. Howard, A. Bogle, and L. Kempel,

J. Appl. Polym. Sci., 107, 2558 (2007).

8. G. Boiteux, Ye.P. Mamunya, E.V. Lebedev, A. Adamczew-

ski, C. Boullanger, P. Cassagnau, and G. Seytre, Synth.Met., 157, 1071 (2007).

9. J.A. King, K.W. Trucker, J.D. Meyers, E.H. Weber, M.L.

Clingerman, and K.R. Ambrosius, Polym. Comp., 22, 142

(2001).

10. J. Delmonte, Metal/Polymer Composites, Van Nostrand

Reinhold, New York(1990).

11. P. Lafuente, A. Fontecha, J.M. Dıaz, and A. Munoz-Esca-

lona, Rev. Plast. Modern., 447, 257 (1993).

12. V.E. Gul, Structure and Properties of Conducting PolymerComposites, VSP, New York(1996).

13. B.D. Mottahed, Polym. Eng. Sci., 40, 61 (2000).

14. S.K. Bhattacharya, Metal Filled Polymers, Marcel Dekker,

New York(1986).

15. D. Stauffer and A. Aharony, Introduction to the PercolationTheory, Taylor & Francis, London, 89–113, (1992).

16. F.J. Lux, J. Mater. Sci., 28, 285 (1993).

17. E.P. Mamunya, V.V. Davidenko, and E.V. Lebedev, Polym.Comp., 16, 319 (1995).

18. E.P. Mamunya, V.V. Davidenko, and E.V. Lebedev, Comp.Interf., 4, 169 (1997).

19. W. Bauhofer, and J.Z. Kovacs, Compos. Sci. Technol., 69,1486 (2009).

20. M.H. Al-Saleh, and U. Sundararaj, Carbon, 47, 2 (2009).

21. A. Larena, and G. Pinto, Polym. Comp., 16, 536 (1995).

22. G. Pinto, C. Lopez-Gonzalez, and A. Jimenez-Martın,

Polym. Comp., 20, 804 (1999).

23. G. Pinto and A. Jimenez-Martın, Polym. Comp., 22, 65

(2001).

24. G. Pinto and M.B. Maidana, J. Appl. Polym. Sci., 82, 1449(2001).

25. A. Maaroufi, K. Haboubi, A. El Amarti, and F. Carmona, J.Mater. Sci., 39, 265 (2004).

DOI 10.1002/pc POLYMER COMPOSITES—-2011 197

26. G. Pinto and A. Maaroufi, J. Appl. Polym. Sci., 96, 2011 (2005).

27. A. Maaroufi, G. Pinto, and I. Paz, J. Appl. Polym. Sci., 98,990 (2005).

28. G. Pinto and A. Maaroufi, Polym. Comp., 26, 401 (2005).

29. M. El Homrany, A. Maaroufi, R. Benavente, J.M. Perena, G.

Pinto, and M. Halim, J. Appl. Polym. Sci., in press.

30. D.R. Lide, Ed., CRC Handbook of Chemistry and Physics,CRC Press, Boca Raton, FL(1991).

31. Yu.N. Anisimov, L.P. Dobrova, and A. Yu. Anisimov, Russ.J. Appl. Chem., 71, 819 (1998).

32. M.B. Heaney, Phys. Rev. B, 52, 12477 (1995).

33. N.P. Berezina and L.V. Karpenko, Colloid J., 62, 676

(2000).

34. F. Carmona, Ann. Chim. Fr., 13, 395 (1988).

35. W.B. Genetti, W.L. Yuan, B.P. Grady, E.A. O’Rear, C.L.

Lai, and D.T. Glatzhofer, J. Mater. Sci., 33, 3085 (1998).

198 POLYMER COMPOSITES—-2011 DOI 10.1002/pc