A new compounding method for exfoliated graphite–polypropylene nanocomposites with enhanced flexural properties and lower percolation threshold

COMPOSITES

www.elsevier.com/locate/compscitech

Composites Science and Technology 67 (2007) 2045–2051

SCIENCE ANDTECHNOLOGY

A new compounding method for exfoliated graphite–polypropylenenanocomposites with enhanced flexural properties and

lower percolation threshold

Kyriaki Kalaitzidou *, Hiroyuki Fukushima, Lawrence T. Drzal

The Composite Materials and Structures Center, Department of Chemical Engineering and Materials Science, Michigan State University,

East Lansing, MI 48823, United States

Received 1 September 2006; received in revised form 16 November 2006; accepted 28 November 2006Available online 23 January 2007

Abstract

Nanocomposites made of polypropylene reinforced with exfoliated graphite nanoplatelets (xGnPTM), are fabricated by melt mixing,polymer solution and coating. Coating is a new compounding method proposed in this research, where xGnP and PP powder are pre-mixed in isopropyl alcohol using sonication to disperse the xGnP by coating individual PP powder particles. It is found that the coatingmethod is more effective than the polymer solution method widely used, in terms of lowering the percolation threshold of thermoplasticnanocomposites, and enhancing the probability that the large platelet morphology of xGnP can be preserved in the final composite. Theresearch reported here provides an understanding on how the compounding method used during the fabrication of nanocomposites isimportant to achieving the optimal flexural properties, electrical conductivity and percolation threshold. This method should have wideapplicability to all thermoplastic matrix nanocomposite systems.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Exfoliated graphite; Particle-reinforced composites (A); Electrical properties (B); Mechanical properties (B); Extrusion (E)

1. Introduction

Nanocomposites are defined as composite materialswhere the reinforcement has at least one dimension in therange of 1–100 nm. The most common nanoreinforcementsused are layered silicate nanoclays and carbon nanotubes,however graphite platelets are also among the leadingnanoscale fillers in research and development and commer-cial projects [1]. As demonstrated recently by the Drzalgroup [2,3] exfoliated graphite nanoplatelets, (xGnPTM),which combine the lower price and layered structure ofclays with the superior thermal and electrical propertiesof carbon nanotubes, can be effective alternative to both

0266-3538/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.compscitech.2006.11.014

* Corresponding author. Present address: Department of PolymerScience and Engineering, University of Massachusetts, 120 GovernorsDrive, Amherst, MA 01003, United States.

E-mail address: [email protected] (K. Kalaitzidou).

clays and nanotubes and provide excellent competitivefunctional properties.

Expanded graphite was developed and proposed byAylsworth [4,5] as reinforcement of polymers, phenolic res-ins in particular, in 1910s. The incorporation of interca-lated graphite into an organic using conventionalprocessing techniques such as extrusion, lay-up, injectionmolding and pressing, was proposed by Lincoln andClaude [6] in 1980s. Since that time, research has been con-ducted on exfoliated graphite reinforced polymers usinggraphite particles of various dimensions and a wide rangeof polymers. In all the cases the objective is to find the opti-mum processing method that will utilize graphite’s superiorproperties and lead to nanocomposites with the desiredproperties.

The processing methods used for graphite-polymernanocomposites are similar to the ones used for clays sinceboth materials have a layered structure. However, because

2046 K. Kalaitzidou et al. / Composites Science and Technology 67 (2007) 2045–2051

they are chemically different some modifications arerequired. Once the graphite is exfoliated (ex situ process)then the nanocomposites can be made by:

(i) Direct mixing, often used in case of low viscositythermoset matrices [2,7].

(ii) Melt compounding, a method used mainly with poly-olefins i.e., HDPE [8], PE exfoliated graphite and PS–graphite using a Brabender mixer [9], HDPE–graph-ite using a Haake mixer and twin screw extruder[10], nylon 6,6-graphite and polycarbonate–graphite,using a twin screw extruder [11] and HDPE–graphitenanocomposites made using a two-roll mill [12].

(iii) Solution intercalation, a method utilizing a solvent todissolve the polymer and disperse the graphite. Thesolvent is evaporated once the mixing is completed.Nanocomposites made by the solution approach arePMMA/graphite using chloroform as solvent[13,14], and maleic anhydride grafted polypropyl-ene/graphite in the presence of xylene [15]. This pro-cessing method results in nanocomposites with higherelectrical conductivity and lower percolation thresh-old compared to nanocomposites made from theexactly same materials using the melt mixing tech-nique [15].

(iv) In situ polymerization where the monomer is poly-merized in the presence of graphite nanosheets, forexample nylon 6/graphite nanocomposites via inter-calation polymerization of e-caprolactam in the pres-ence of expanded graphite [16], graphite–polystyrene,starting with styrene–graphite–benzoyl peroxide mix-ture [17–19], and graphite–polyacrylonitrile nano-composites [20].

(v) Electrospinning of nanoscale fibers using polymersolution and melts at ambient conditions i.e., fabrica-tion of graphite nanofibers starting from xGnP dis-persed in (PAN)/N,N DMF solution [21] byapplying electric field between the polymer reservoirand a collection plate that is oppositely charged. Ifthe graphite is not exfoliated prior to fabricatingthe composite, then an in situ exfoliation processcan be used such as:

(vi) Polymerization filling technique, where in situ poly-merization occurs in the presence of initiator-interca-lated graphite, i.e. intercalated graphite was mixedwith an epoxy resin and exfoliated in situ duringthe curing process [22]. Polystyrene–graphite nano-composites [23], also fabricated by this method,found to have better properties compared to compos-ites made by melt mixing of polystryrene withexpanded graphite.

In summary, composites made by in situ processing havebetter mechanical properties compared to composites madeby melt-mixing or other ex situ fabrication methods due tobetter dispersion, prevention of agglomeration and stron-ger interactions between the reinforcement and the poly-

mer. In situ exfoliation can also be achieved during meltmixing, however since the temperature required for exfoli-ation is �230 �C only polymers which can be processed athigh temperatures without being degraded can be used forin situ fabrication of nanocomposites.

The focus of this research is to explore how the com-pounding method employed to fabricate xGnP–PP nano-composites alter their properties. The nanocomposites arefabricated by melt mixing or polymer solution method fol-lowed by compression or injection molding and their flex-ural properties, electrical conductivity and percolationthreshold are determined. A new compounding method,the coating of PP with xGnP by sonication in presence ofisopropyl alcohol, (IPA) is proposed as an alternative tothe traditional melt mixing method, that results in poor dis-persion and agglomeration of the reinforcements; and tothe commonly used polymer solution method that is usedto improve dispersion and obtain electrically conductivepolymer composites with low percolation threshold.

2. Experimental

2.1. Materials

The polymer used is polypropylene powder (Pro-fax6301, melt flow index 12 g/10 min, ASTM D1238) kindlyprovided by Basell [24]. The exfoliated graphite nanoflakesare produced from sulfuric acid based intercalated graph-ite, in this case obtained from UCAR International Inc.,using a cost and time effective exfoliation process initiallyproposed by Fukushima [2]. Two types of xGnP are used,xGnP-15 which have a diameter of �15 mm and xGnP-1with an average diameter of �1 mm. Both xGnP have aplatelet less than 10 nm. Details on the exfoliation processas well as on the morphology of xGnP can be found else-where [2].

2.2. Processing of xGnP–PP nanocomposites

Three compounding methods were used to fabricate thexGnP–PP nanocomposites. Melt mixing, which is simpleand compatible with existing polymer processing tech-niques such as extrusion and injection or compressionmolding; using a DSM Micro 15cc Compounder, (vertical,co-rotating twin-screw microextruder). The processing con-ditions used (3 min, Tbarrel = 180 �C, Tmold = 80 �C and245 rpm) were the optimum based on a design of experi-ments study (23 factorial design) [3].

The second compounding method used is the solutionapproach, which while feasible, in the case of PP requireslarge amounts of solvents such as toluene or xylene andhigh temperatures that are neither practical nor safe. How-ever, in order to understand the effect of fabricationmethod on the electrical conductivity and percolationthreshold of xGnP–PP composites a modified version ofthe solution method proposed by Shen et al. [15] wasemployed. The xGnP were dispersed in xylene using sonica-

Fig. 1. ESEM image of (a) PP particle (scale bar 15 lm), (b) 0.2 wt.%xGnP-15 coated PP (scale bar 20 lm) and (c) 0.2 wt.% xGnP-1 coated PP(scale bar 15 lm).

K. Kalaitzidou et al. / Composites Science and Technology 67 (2007) 2045–2051 2047

tion for 2 h and the PP was dissolved in refluxing xylene at130 �C for 0.5 h. The graphite suspension was added dropwise to the PP solution and after refluxing for 1.5 h it wasfiltered. When the temperature dropped to about 70 �C thesolution precipitated by addition of acetone, filtered, anddried in vacuum oven. The resulting composite powderwas used for compression or injection molding.

Coating of polypropylene with graphite in presence ofisopropyl alcohol is a new compounding method developedby the Drzal group [3]. The xGnP is dispersed in isopropylalcohol by sonication for 1 h at room temperature. The PPpowder is added to the solution and sonication is continuedfor 0.5 h. Finally, the solvent is evaporated at 80 �C result-ing in complete coverage of the powder particles with thexGnP. Alternatively, the isopropyl alcohol can be recycledby using filtration and reused. Thus, this new premixingmethod can be environmental friendly and more cost andtime effective compared to the solution approach. Themain advantage of this method is that sonication breaksdown the xGnP agglomerates and the thick xGnP–IPAsolution covers the PP particles very efficiently resultingin a homogeneous xGnP coated PP powder that can beused for compression or injection molding. Micrographsof the uncompressed neat polymer powder, xGnP-15coated PP and xGnP-1 coated PP at 0.2 wt.% are shownin Fig. 1a–c, respectively.

Once the compounding was completed the compositeswere made by injection or compression molding. A DacaMicro Injector system was used for injection molding.The cylinder and mold temperature used were 180 �C and80 �C, respectively, and the pressure used was 160 psi.The compression-molded samples were made using thecomposite pellets or powder obtained during compound-ing. The conditions used are at 200 �C for 20 min with nopressure applied and 200 �C for 20 min under pressure�20,000 psi. During the compression molding vacuumwas applied to remove any trapped air.

2.3. Characterization techniques

Flexural tests were performed with a UTS SFM-20machine [United Calibration Corp.] at room temperatureby following the ASTM D790 standard test method (3-point bending mode). The samples were made in standardbar shape and the span was set to 5.08 cm. The test wasperformed at flexural rate of 0.127 cm/min.

The morphology of the nanocomposites was investi-gated by environmental scanning electron microscopy(Electroscan 2020). The samples were gold coated to avoidcharging and the voltage used was 20–30 kV.

The resistivity of xGnP-1/PP and xGnP-15/PP compos-ites was measured along the flow direction, in case of theinjection-molded samples, using impedance spectroscopyby applying the two-probe method at room temperature.Samples with dimensions of 5 · 3 · 12 mm3 were cut fromthe middle portion of flexural bars, and the resistivity wasmeasured along the thickness direction (5 mm). The two

surfaces that were connected to the electrodes were firsttreated with O2 plasma (10 min, 550 W) in order to removethe top surface layers which are rich in polymer and thengold coated to a thickness of 1–2 nm to ensure good con-

2048 K. Kalaitzidou et al. / Composites Science and Technology 67 (2007) 2045–2051

tact of the sample surface with the electrodes. The resis-tance of samples was measured in frequency range of0.1–100,000 Hz and converted to conductivity by takinginto account the sample dimensions.

3. Results and discussion

3.1. Effect of compounding on flexural properties

The flexural strength and modulus of elasticity of xGnP-15/PP nanocomposites at 0, 1, 3, 10 vol.% made by (i) meltmixing and injection molding and (ii) coating followed bymelt mixing and injection molding are shown in Figs. 2and 3, respectively. The coated samples show 8% higherstrength at 3 and 10 vol.% and an improved modulus upto 60% at a loading of 10 vol.% compared to the samplesmade by melt mixing a strong indication of the improveddispersion conditions due to the extra coating step. It isexpected that coating will not have such a strong effect athigher xGnP-15 contents because the number of graphiteplatelets will be so large that the amount of polymer pres-ent will not be sufficient to prevent agglomeration.

The morphology of melt mixed and premixed PP com-posites at 10 vol.% of xGnP-15 was studied by ESEM.The fracture surfaces examined were obtained during flex-ural testing. Agglomerates of xGnP-15 are present in the

35

39

43

47

51

55

0 2 4 6 8 10vol%

MP

a

Melt mixingCoating

Fig. 2. Effect of compounding on the flex strength of xGnP-15/PPcomposites.

1

1.5

2

2.5

3

3.5

0 2 4 6 8 10vol%

GP

a

Melt mixingCoating

Fig. 3. Effect of compounding on flex modulus of xGnP-15/PPcomposites.

samples made by melt mixing as shown by the arrows inFig. 4a whereas coating the PP powder with xGnP-15breaks the graphite agglomerates and results in a well dis-persed system free of particle agglomerations with thegraphite platelets being well embedded in the PP matrixas shown in Fig. 4b. In addition, the ESEM study revealedthat there are two distinguished types of morphology in themelt mixed samples: (i) areas with big xGnP-15 agglomer-ates, shown in Fig. 4a, and large flat graphite plateletsbuckled and deformed that are shown in Fig. 5a and band (ii) areas that are away from the specimen edges whereno xGnP-15 or only very small graphite particles can beseen on the surface as indicated by the arrows in Fig. 5c.Based on Figs. 4a, and 5 which are typical ESEM imagesof melt mixed xGnP-15/PP morphology it is concluded thatduring melt mixing there is not enough shear to breakdown the xGnP-15 agglomerates and homogeneously dis-perse the graphite platelets.

3.2. Effect of compounding on electrical conductivity

The important properties in electrical conductive com-posites are the electrical conductivity, reported as bulk con-ductivity (S/cm), and the percolation threshold, defined asthe minimum volume content of the conductive reinforce-ment above which the polymer composite becomes electri-cally conductive. It is desirable for the conductive fillercontent to be as low as possible in order to achieve goodprocessability, low cost and satisfactory mechanical perfor-mance. Both the conductivity and the percolation thresh-old are affected by various factors such as the volumefraction and geometric characteristics of the conductive fil-ler [25–27], the filler orientation and spacing within thepolymer matrix [28,29] as well as the crystallinity of thematrix [30]. The fabrication method and processing condi-tions of the composites play an important role in the perco-lation threshold and conductivity since they affect theorientation, dispersion and interparticle spacing withinthe polymer matrix and they may alter the filler’s aspectratio or enhance the interactions between filler and matrix[9,11,20,21,25,31] and change the matrix crystallinity.

The effect of the three compounding methods; (i) meltmixing, (ii) polymer dissolution and (iii) coating the PPpowder with xGnP-1, on the percolation threshold andelectrical conductivity of xGnP-1/PP nanocomposites isshown in Fig. 6. All the samples were compression moldedand the electrical conductivity was measured in the direc-tion parallel to the sample’s length. As shown the conduc-tivity of xGnP-1/PP nanocomposites made by thepremixing compounding method is as high as 10�4 S/cmat a loading of 3 vol.%, indicating that the percolationthreshold which can be achieved by this method is muchlower than the others. For both xGnP-1 loadings usedi.e., 3 and 5 vol.% the proposed coating method results inconductivity higher than the conductivity of the polymersolution processed samples which is the method commonlyused when composites with lower percolation threshold are

Fig. 4. ESEM of fracture surface of 10 vol.% xGnP-15/PP composites made by (a) melt mixing and (b) coating and melt mixing (scale bar 150 lm).

Fig. 5. ESEM of fracture surface of 10 vol.% xGnP-15/PP composites made by melt mixing: (a) and (b) (scale bar: 50 lm and 150 lm, respectively) andmade by coating: (c) (scale bar: 150 lm).

K. Kalaitzidou et al. / Composites Science and Technology 67 (2007) 2045–2051 2049

required. This indicates that the coating method is at leastas efficient in facilitating the formation of conductive net-work as the commonly used solution method but avoidsthe problems with recovery of solvent.

In the case of coating there are no agglomerates of xGnPdue to the use of sonication as a result of the PP powderbeing homogeneously coated by xGnP. When the polymermelts under compression molding conditions, the xGnPplatelet network formed on the exterior of the PP particles

remains largely intact. However during the solution processxGnP platelets have the opportunity to re-aggregate anddo not efficiently participate in the conductive network.

Composites were made by coating and compressionmolding using both 1 and 15 lm xGnP in order to deter-mine the percolation threshold. As indicated in Fig. 7,xGnP-1 has a percolation threshold of 0.1 vol.% whilethe corresponding value for xGnP-15 is 0.3 vol.%. The lar-ger percolation threshold for the larger 15 lm xGnP are

1.E-131.E-121.E-111.E-101.E-091.E-081.E-071.E-061.E-051.E-041.E-031.E-02

0 3 5

vol%

S/cm

Coating+CM

Solution+CM

Melt Mixing+CM

Fig. 6. Effect of compounding on the percolation threshold and conduc-tivity of xGnP-1–PP nanocomposites made by compression molding.

1.E-13

1.E-11

1.E-09

1.E-07

1.E-05

1.E-03

1.E-01

0 5 10 15 20 25 30 35 40vol%

S/cm

Melt mixing

Coating+Melt mixing

Fig. 8. Effect of compounding on the percolation threshold and conduc-tivity of xGnP-15–PP made by injection molding.

2050 K. Kalaitzidou et al. / Composites Science and Technology 67 (2007) 2045–2051

surprising i.e., the larger the aspect ratio should produce alower the percolation threshold. However, at the same vol-ume fraction, the smaller 1 lm xGnP would have approxi-mately 200 times the number of particles that the larger15 lm xGnP so the possibility that the PP powder will becoated more effectively during coating with the smallerplatelets.

It is evident that combination of the coating and com-pression molding yields composites with lower percolationthreshold and higher conductivity. However, it is of practi-cal interest to explore what is the effect of coating in thecase of injection molded samples since injection moldingis widely used both in research labs and large scale produc-tion. Composites were made using (i) melt mixing andinjection molding and (ii) coating, melt mixing and injec-tion molding, since it is not practical to injection moldthe coated composite powder without passing it firstthrough the extruder.

The percolation threshold and electrical conductivity ofxGnP-15/PP nanocomposites are shown in Fig. 8. The pre-mixed samples have a percolation threshold less than5 vol.% while the melt mixed ones �7 vol.%. As the xGnPcontent increases the difference in electrical conductivity ofthe composites made with the two methods decreases. Thereason is that coating breaks down the xGnP agglomeratesand allows for formation of a continuous network at lowerloadings, however, as the graphite concentration increases

1E-13

1E-11

1E-09

1E-07

1E-05

0.001

0.1

0 1 2 3 4 5vol%

S/cm

xGnP-15

xGnP-1

Fig. 7. Effect of xGnP’s aspect ratio on the percolation threshold andelectrical conductivity of xGnP/PP made by coating and compressionmolding.

the platelets can re-agglomerate. The significant effect of fil-ler orientation during injection molding on the percolationthreshold is outlined by comparing Figs. 7 and 8. Thecoated and compression molded xGnP-15/PP have a perco-lation threshold of �0.3 vol.% (Fig. 7), whereas the coatedinjection molded composites have a percolation thresholdof �5 vol.% (Fig. 8).

4. Conclusions

A new, efficient compounding method has been devel-oped for optimizing the dispersion of nanoparticles in ther-moplastic polymers. It consists of coating polymer particleswith nanoparticles using a liquid phase non-solvent underultrasonication. In this research, xGnP nanoparticles weredispersed in isopropyl alcohol using sonication in the pres-ence of PP powder at room temperature to produce a uni-form coating on the PP. Subsequent melt mixing andinjection molding produced nanocomposites withimproved flexural strength and modulus. The enhancementof the flexural properties is attributed to the uniform dis-persion of the nanoparticles in the solid state before meltprocessing.

The coating method is more effective at lowering thepercolation threshold of nanocomposites than the widelyused polymer solution method. It may result that this isone of the only methods to insure that the large plateletmorphology of xGnP can be preserved in the final compos-ite. Additional advantages of this method are that theexperimental set up is very simple, no solvents are used,there is no need for high temperatures and the non-solventcan be easily recycled making for a method that is practi-cal, safe, cost and time effective and environmentallyfriendly.

Composites made by melt mixing and injection moldingshow a higher percolation threshold because of limitationsin the ability of the melt mixing equipment to disperse thexGnP and maintain their platelet type morphology. Fur-thermore, injection molding creates morphology with pref-erential alignment the platelets along the flow direction. Asa result, no improvement in electrical conductivity resultingfrom the effect of larger xGnP aspect ratio was detected.The lowest percolation threshold measured was less than

K. Kalaitzidou et al. / Composites Science and Technology 67 (2007) 2045–2051 2051

0.1 vol.% for samples made by coating and compressionmolding while composites with similar compositions madeby melt mixing and injection molding had a percolationthreshold of �7 vol.%.

Acknowledgements

Partial support for this research was provided by a grantfrom NASA LaRC, ‘‘Graphite Nanoreinforcements forAerospace Nanocomposites’’ NAG1-01004, ThomasGates, Program Director. The authors also wish to expresstheir thanks to Basell for providing the polypropyleneresin.

References

[1] Sherman LM. Chasing nanocomposites. plastics technology. Avail-able from: http://www.plasticstechnology.com/articles/200411fa2.html.

[2] Fukushima H. Graphite nanoreinforcements in polymer nanocom-posites. PhD thesis, Michigan State University, Department ofChemical Engineering and Materials Science, 2003.

[3] Kalaitzidou K. Exfoliated graphite nanoplatelets as reinforcement formultifunctional polypropylene nanocomposites. PhD thesis, Michi-gan State University, Department of Chemical Engineering andMaterials Science, 2006.

[4] Aylsworth JW. Expanded graphite and composition thereof. USpatent 1,137373, 1915.

[5] Aylsworth JW. Expanded graphite. US patent 1,191,383, 1916.[6] Lincoln VF, Claude Z. Organic matrix composites reinforced with

intercalated graphite. US patent 4,414,142, 1983.[7] Celzard A, McRae E, Marache JF, Furdin G, Dufort M, Deleuze C.

Composites based on micron-sized exfoliated graphite particles:electrical conduction, critical exponents and anisotropy. J Phys ChemSolid 1996;57(6–8):715–8.

[8] Wang YS, O’Gurkis MA, Lindt JT. Electrical properties of exfoli-ated-graphite filled polyethylene composites. Polym Compos1986;7(5):349–54.

[9] Krupa I, Chodak I. Physical properties of thermoplastic-graphitecomposites. Eur Polym J 2001;37(11):2159–68.

[10] Zheng W, Lu X, Wong S-C. Electrical and mechanical properties ofexpanded graphite-reinforced high-density polyethylene. J ApplPolym Sci 2004;91(5):2781–8.

[11] Clingerman ML, King JA, Schulz KH, Meyers JD. Evaluation ofelectrical conductive models for conductive polymer Composites. JAppl Polym Sci 2002;83(6):1341–56.

[12] Weng WG, Chen G-H, Wu D-J, Yan W-L. HDPE/Expandedgraphite electrically conducting composite. Compos Interfaces2004;11(2):131–43.

[13] Zheng W, Wong S-C, Sue H-J. Transport behavior of PMMA/expanded graphite nanocomposites. Polymer 2002;43(25):6767–73.

[14] Zheng W, Wong S-C. Electrical conductivity and dielectric propertiesof PMMA/expanded graphite composites. Comp Sci Tech2003;63(2):225–35.

[15] Shen J-W, Chen X-M, Huang W-Y. Structure and electrical prop-erties of grafted polypropylene/graphite nanocomposites prepared bysolution intercalation. J Appl Polym Sci 2003;88(7):1864–9.

[16] Pan Y-X, Yu Z-Z, Ou Y-C, Hu G-H. A new process of fabricatingelectrically conducting Nylon 6/graphite nanocomposites via interca-lation polymerization. J Polym Sci, Part B: Polym Phys2000;38(12):1626–33.

[17] Zou J-F, Yu Z-Z, Pan Y-X, Fang X-P, Ou Y-C. Conductivemechanism of polymer/graphite conducting composites with lowpercolation threshold. J Polym Sci, Part B: Polym Phys2002;40(10):954–63.

[18] Chen G-H, Wu D-J, Weng W-G, He B, Yan W-L. Preparation ofpolystyrene–graphite conducting nanocomposites via intercalationpolymerization. Polym Int 2001;50(6):980–4.

[19] Chen G-H, Wu D-J, Weng W-G, Yan W-L. Preparation of polymer/graphite conducting nanocomposite by intercalation polymerization.J Appl Polym Sci 2001;82(10):2506–13.

[20] Du XS, Xiao M, Meng YZ. Synthesis and characterization ofpolyaniline/graphite conducting nanocomposites. J Polym Sci, Part B:Polym Phys 2004;42(10):1972–8.

[21] Mack J, Viculis L, Luoh R, Kaner R, Hahn T, Ali A, et al.Nanocomposite fibrils from graphite nanoplatelets. In: Conferenceproceedings, international conference on composite materials, SanDiego (CA), USA, 2003.

[22] Chung DDL. Composites of in-situ exfoliated graphite. US patent,4,946,892 1990.

[23] Xiao P, Xiao M, Gong K. Preparation of exfoliated graphite/polystyrene composite by polymerization-filling technique. Polymer2001;42(11):4813–6.

[24] http://www.basell.com/portal/site/basell/.[25] Banerjee P, Mandal BM. Conducting polyaniline nanoparticle blends

with extremely low percolation thresholds. Macromolecules1995;28:3940–3.

[26] Jing X, Zhao W, Lan L. The effect of particle size on electricconducting percolation threshold in polymer/conducting particlecomposites. J Mater Sci Lett 2000;19:377–9.

[27] Bigg DM. The effect of compounding on the conductive properties ofEMI shielding compounds. Adv Polym Tech 1984;4:255–6.

[28] Gokturk HS, Fiske TJ, Kalyon DM. Effects of particle shape and sizedistributions on the electrical and magnetic properties of nickel/polyethylene composites. J Appl Polym Sci 1993;50:1891–901.

[29] Fiske TJ, Gokturk HS, Kalyon DM. Percolation in magneticcomposites. J Mater Sci 1997;32:5551–60.

[30] Chodak I, Krupa I. Percolation effect and mechanical behavior ofcarbon black filled polyethylene. J Mater Sci Lett 1999;18:1457–9.

[31] Chodak I, Omastova M, Pionteck J. Relation between electrical andmechanical properties of conducting polymer composites. J ApplPolym Sci 2001;82:1903–6.