conductive polymer composites with ultralow ...

JOURNAL OF NANOSTRUCTURED POLYMERS AND NANOCOMPOSITES 4/1 21-27

CONDUCTIVE POLYMER COMPOSITES WITH ULTRALOW PERCOLATION THRESHOLD CONTAINING CARBON NANOTUBES YE.P. MAMUNYA1*,N.I. LEBOVKA2, M.O. LLSUNOVA2, E.V. LEBEDEV1, A. RYBAK3 , G. BOITEUX3 Institute of Macromolecular Chemistry, National Academy of Sciences of Ukraine, 48 Kharkovskoe chaussee, 02160, Kiev, Ukraine. 2 Institute of Biocolloidal Chemistry named after F.D. Ovcharenko, NAS of Ukraine, 2, blvr. Vernadskogo, Kyiv 03142, Ukraine. 3 Laboratoire des Materiaux Polymeres et des Biomateriaux, UMR CNRS 5627 “Ingenierie des Materiaux Polymeres”, Universite Claude BernardLyonl, 15 Boulevard A. Latarjet, 69622 Villeurbanne Cedex, France. 1

Received 7 May 2007; accepted 21 September 2007 Abstract Electrical conductivity σ and dielectric characteristics ε’, tan δ were investigated as function of the multiwalled carbon nanotubes (CNT) content in the PE/CNT and PVC/CNT composites prepared by hot compacting method. The composites show a presence of the ultralow values of percolation threshold, φc = 0.00036 in PE/CNT and φc = 0.00047 in PVC/CNT. It is caused by two reasons: high anisotropy of CNT particles with aspect ratio length/diameter ~ 1000 and the presence of segregated structure of CNT within the polymer matrix with distribution of nanotubes on the boundaries between the polymer grains. The dielectric characteristics demonstrate the percolation behavior as well. In the vicinity of percolation threshold the values of ε’ and tan δ sharply increase by orders of magnitude, for PE/CNT composite such effect is essentially higher. Wide range of tan δ variations at low content of carbon nanotubes evidences that such composites are promising for EMI shielding materials.

1. Introduction The study of conductive polymer composites attracts large interest because of wide possibilities of their industrial applications. They display a broad spectrum of properties useful for production of sensitive electrodes [1], sensors for chemical vapours [2], electromagnetic radiation shielding materials [3], electrical heaters [4], as well as pressure, deformation and temperature sensors [5, 6]. Such composites are thermoplastic or thermoset matrix filled with dispersed fillers, such as metals [7, 8] or carbon [9, 10]. They possesses both polymer (insulating) and metallic (conducting) properties depending on content of the conductive filler and structure of the conductive phase. Specific feature of composites with conductive dispersed fillers is the presence of, so-called, percolation threshold, and a sharp transition from insulating to conductive state occurs when the filler content φ exceeds some threshold value φc. _____

*Corresponding author. E-mail address: [email protected] (Ye. Mamunya)

It seems promising to prepare composite material with small filler content. Such composite can preserve mechanical properties of a polymer and possess a high electrical conductivity. The percolation threshold φc is typically observed within 5-30% for particles of near-spherical shape, but can be noticeably smaller for needle-like particles [7, 11]. Also, a reduced percolation threshold is usually observed for segregated systems with ordered spatial distribution of dispersed filler in a polymer matrix [12-14], particularly in polymer blends [15, 16]. Most currently, the carbon nanotubes (CNTs) started to cause ever increasing interest as a conductive filler for polymeric composites. Owing to high aspect ratio (˜ 1000), the percolation threshold in a polymer filled with the CNT can be achieved at a very low concentration of the CNTs, nevertheless the values of percolation threshold are revealed in very wide range, for instance within 0.0025%-7.5% for MWCNT [17-21] and within 0.005%-4% for SWCNT [22-24]. The experimental data clearly indicate that the percolation threshold value depends 21

ΥΕ.P. Mamunya et al.: JNPN 4/1 (2008) 21-27

on vigorous mixing of the components, on method used for embedding the CNTs into a polymer matrix, as well as on agglomeration and alignment of the CNTs. It was demonstrated that the percolation thresholds, experimentally observed in the polymer/CNTs composites, were much lower than those predicted by the statistical percolation theory [25]. A possible explanation is that the observed percolation is not purely geometrically random, and there exists segregation with selective localization of the conductive CNT particles inside a polymeric matrix [18]. But many details of the observed discrepancies between the theoretical and experimental results, as well as relations between the process conditions and percolation behavior in the polymer/CNTs composites, are not well understood yet. The aim of this work was to investigate the percolation behavior in the thermoplastic polymer/multiwalled carbon nanotubes hot compacted composites in the terms of electrical conductivity and dielectric parameters. 2. Experimental The CNTs used in this work were produced by TMSpetsmash Ltd (Kyiv, Ukraine, e-mail: [email protected]) and have the diameter about 12-20 nm and length about tens of microns (μm). The density of the CNT walls was assumed to be the same as the pure graphite density, ρf =2.045 g/cm3. As polymer component, the ultrahigh molecular weight polyethylene (PE) and the polyvinyl chloride (PVC) in a powdered form with the average particle size D of 100 μm were used. The density (ρp) of PE and PVC was equal to 0.95 and 1.37 g/cm3, respectively. The mechanical mixture of PE or PVC with CNTs was homogenized by thorough grinding of the mixed polymer/CNT powders in a porcelain mortar to the visually homogeneous state. The mixture was placed into a hot die heated to 155°C and then pressed (hot compacted). The samples were produced as discs with 30 mm diameter and 1.5 mm height. The DC electrical conductivity a was measured 22

using a two-electrode scheme . The sample was placed between two electrodes made from 20 jam thick soft (annealed) alumina foil with 1 mm thick rubber bottom layer to ensure a perfect compositeelectrode contact. The values of σ were calculated using the following equation:

σ=

(1)

1 l ⋅ R S

where R is a electrical resistance measured experimentally using E6-13 teraohmmeter, l and S are thickness and area of the sample, respectively. The conductivity σf of the CNT powder was measured using a cell described in [26]. It has been found that σf = 10.0 S/cm. The dielectric parameters (dielectric constant ε’ and dielectric loss tangent tan δ) were measured at frequency 1 kHz by bridge P5083. 3. Results Figs 1a and 2a show electrical conductivity σ versus volume content φ of CNT in the PVC/CNT and PE/CNT composites. The electrical conductivity of composites sharply increases by many order of magnitude when the filler content φ reaches the threshold concentration φc. It can be seen that the non-conducting - conducting state transition takes place at a rather small volume content of CNTs. The electrical conductivity behavior above the percolation threshold can be described by the well-known scaling relation

σ = σ 0 (ϕ − ϕ c )

t

(2)

where t is the critical exponent, and σ0 is the adjustable parameter. The conductivity exponent t reflects dimensionality of the system and universality class of the problem [25]. The theoretical random percolation value is close to 2.0 for the three-dimensional system. Figs. 1b and 2b show scaling of σ versus φ-φc dependences in the double logarithmic presentation for calculation of the parameters of eq.(2). Line corresponds to the least square fitting


Fig. 1a - electrical conductivity σ versus volume content of nanotubes φ; b - scaling of σ versus φ - φc dependences in the double logarithmic presentation for PVC/CNTs composites.

Fig. 2a - electrical conductivity σ versus volume content of nanotubes φ; b - scaling of σ versus φ - φc dependences for PE/CNTs composites.

of the experimental data with CNT fraction at φ =φc, the slope of this line defines the value of t. The values of parameters were φc = 0.00047, t = 2.4 for PVC/CNTs composites and φc= 0.00036, t = 2.0 for PE/CNT composites. The values of t are slightly higher than theoretical value, which is close to 2.0. Fig. 3 demonstrates dependence of the dielectric parameters (dielectric constant ε΄ and tan δ) on the volume fraction of CNTs. It can be seen that a slight increase of ε΄ and tan δ takes place below the percolation threshold at φ φc. The structure of CNTs and polymer/CNTs composites was examined by electron microscopy (transmission type for the CNTs and scanning type for the composites). Fig. 4-a demonstrates the tangled structure of pure nanotubes. The granular structure of the polymer matrix contains nanotubes on the surface between grains (Fig. 4-b). Such distribution of nanotubes is a result of the procedure of composite preparation.

23


Fig. 3. Dependence of the dielectric constant ε΄ and loss-factor tan δ on content of CNTs for the PVC/CNTs and PE/CNTs composites.

Fig. 4. a- electron microscope image of multiwalled carbon nanotubes; b - morphology of PE/CNTs composite. 24


Discussion Dependence of electrical conductivity on the content of nanotubes in the PVC/CNTs and PE/CNTs composites demonstrates very low values of the percolation threshold, which is equal to 0.00047 and 0.00036, respectively. The high anisotropy of CNTs with the aspect ratio length/diameter l/2r ˜ 1000 is an obvious reason of such an effect. Higher aspect ratio of the filler particles leads to lower value of the percolation threshold [7, 11]. On the other hand, the distribution of CNTs in the polymer matrix is not uniform. Ordered distribution of filler is a result of the composite preparation procedure. The mechanical mixture of PVC and CNT powders creates such a structure that big particles of PVC appear to be covered by the powdered nanotubes, and conductive phase of CNTs is on the surface of the polymer particles. Hot pressing deforms the polymer particles and results in formation of a compacted continuous polymer phase, where conductive patterns of filler are located on the boundaries between polymer grains [12, 13, 27]. Fig. 4-b demonstrates a granular structure of the polymer matrix with the presence of nanotubes on the surface between the polymer particles. Hence, we can conclude that very low percolation threshold concentrations φq>c can be attributed not only to the high aspect ratio of the nanotubes, but also to their segregation inside the polymer matrix. The models of segregated conductive polymerfiller composites were described elsewhere [11-14, 27]. The models [12, 13] predict decrease of the percolation threshold φcs in the segregated system comparing with the value of φc at random distribution of the conductive filler in a polymer matrix. The value of φcs depends on the ratio D/d under condition D>>d, where D is size of the polymer particles and d is size of the filler particles. The effective nanotube size d can be estimated as an average distance between the linkages of crossing nanotubes. This value can be taken from Fig. 4a, approximately, as d ˜ 1 μm. Assuming the average size of a polymer particle D = 100 m, we obtain the value D/d = 100. In the segregated polymer/CNTs system, the value of percolation threshold can be calculated as [13]:

ϕ cs ≈

3n

l D ⋅ 2r d

(3)

where n is the number of layers of filler on the boundary between the polymer grains, n is small, n~1, l is length of nanotubes and 2r is their diameter. This calculation gives the predicted value of the percolation threshold φcs ≈ 310-5, whereas the experimental values of c for the PVC/CNTs and PE/ CNTs composites lie in the range of (3-5).10-4. Such a discrepancy can be explained by conditions of distribution of the CNTs between the polymer grains. Fig. 4b shows that filler can form thick layers and aggregates in some places on the intergrain surfaces. In this case, increase of parameter n results in increase of φcs. Moreover, uncertainty in determination of the parameter d in eq. (3) can also influence the value of φcs. On the other hand, the value of φc should be lower for better filler distribution on the surface of polymer particles. Hence, correspondence between the model φcs and experimental φc values can be considered as rather satisfactory. It follows from these data that the observed ultralow values of the percolation threshold φc in PVC/ CNTs and PE/CNTs composites can be explained by both very high aspect ratio (˜1000) of the nanotubes and segregation of CNTs inside the polymer matrix. By contrast, the authors [14] have obtained much higher value of the percolation threshold in segregated system of polyethylene/single-walled CNTs, φc ≈ 0.01 (weight fraction). The critical exponents in eq. (2) t =2.0 for PE/ CNTs and t=2.4 and PVC/CNTs are close to the theoretical value t ≈ 2. The dielectric properties demonstrate the percolation behaviour as well. Above the percolation threshold φc the sharp increase of ε΄ and tan δ takes place, than the values of dielectric parameters rich the plateau. Such behavior was predicted by the model for two-phase insulating/conducting systems [28] and was observed in polymer/dispersed metal composites [28,29]. Rise of ε΄ above the percolation threshold φc ≈ 0.019

25


was observed for the LDPE/CNT composites in [30]. It is necessary to note that there exist some differences for PE/CNTs and PVC/CNTs composites. The value of ε΄ on the plateau is higher for PE/CNTs than for PVC/CNTs by the order of magnitude. It can be caused by the fact that conductivity of PE/ CNTs composite is approximately ten times higher that that of PVC/CNTs composite for samples with maximal content of CNTs. The reason of such behavior is not clear, since geometry of the conductive phase (the values of D and ratio D/d) is equivalent for both composites. The Fig. 3 shows that the interval of tan δ alteration with variation of the content of CNTs is much wider for PE/CNTs composites. The maximum values of tan δ are approximately the same for both composites, while initial value of tan δ is lower by two orders of magnitude for PE in comparison to PVC. It is known that PE is a polymer with one of the lowest values of dielectric losses. Very small content of carbon nanotubes, slightly higher than value of φc, leads to sharp increase of dielectric losses. It can be caused by creation of a segregated structure of CNTs in the volume of hot compacted composite, which forms “shielding skeleton” of conductive phase for electromagnetic flow. Such ability is promising for creation of shielding materials for electromagnetic irradiation. 4. Conclusions Investigation of electrical conductivity of the PVC/CNTs and PE/CNTs composites depending on the content of CNTs enabled to reveal the ultralow values of percolation threshold, φc = 0.00047 and φc = 0.00036, respectively. It is caused by two reasons: high anisotropy of CNTs with the aspect ratio (length/diameter) ~ 1000 and the presence of segregated structure of CNTs within the polymer matrix with distribution of nanotubes on the boundaries between the polymer grains. The dielectric characteristics (ε΄ and tan δ) demonstrate the percolation behavior. In the vicinity of the percolation threshold, the values of ε΄ and tan δ sharply increase by orders of magnitude. Such effect 26

is essentially higher for PE/CNTs composites. Wide range of the tan δ variations at low content of carbon nanotubes points to great potential of such composites for EMI shielding materials. Acknowlegments This research was supported by Project 18959 (2006-2007) “Nanostructured multicomponent polymer systems for conductive materials” of CNRSNASU cooperation. References: 1. Ramirez-Garcia S., Alegret S., Cespedes S., Forster R.J. “Carbon composite microelectrodes: charge percolation and electroanalytical performance”. Anal. Chem, 76 (2004) 503-512. 2. Tillman E.S, Lewis N.S. “Mechanism of enhanced sensitivity of linear poly (ethylenimine) - carbon black composite detectors to carboxylic acid vapors. Sensors & Actuators”, B-Chem, 96 (2003) 329-342. 3. Ma C.C, Hu A.T, Chen D.K. “The processability, electrical and mechanical properties of EMI shielding ABS composites”. Polym. Compos., 1 (1993) 93-99. 4. El-Tantawy F., Kamada K., Ohnabe H. “In situ network structure, electrical and thermal properties of conductive epoxy resin-carbon black composites for electrical heater applications”. Mater. Lett., 56 (2003) 112-126. 5. Chelidze T, Gueguen Y. “Pressure-induced percolation transitions in composites”, J. Phys. D: Appl. Phys., 31 (1998)2877-2885. 6. Knite M., Teteris V., Polyakov B., Erts D. “Electric and elastic properties of conductive polymeric nanocomposites on macro- and nanoscales”, Mater. Sci. Eng.- C, 19 (2002) 15-19. 7. Metal-Filled Polymers (Properties and Applications) / Ed. by S.K. Bhattacharya, Marcel Dekker, NewYork, Basel, 1986. 8. Delmonte J. Metal/Polymer Composites. Van Nostrant Reinhold, New York, 1990. 9. Carbon black-polymer composites. The physics of electrically conducting composites / Ed. by E.K. Sichel, Marcel Dekker, New-York, Basel, 1982. 10. Huang J.C. Review. “Carbon black filled conducting polymers and polymer blends”, Advances in Polymer Technology, 21 (2002) 299-313. 11. He D., Ekere N.N. “Effect of particle size ratio on the conducting percolation threshold of granular conductive-insulating composites”, J. Phys. - D: Appl. Phys., 37 (2004)1848-1852. 12. Mamunya Y.P., Davydenko V.V., Pissis P., Lebedev E.V. “Electrical and thermal conductivity


13.

14.

15.

16.

17.

18.

19. 20.

21.

22.

23.

24.

25.

of polymers filled with metal powders”, Europ. Polym. J., 38 (2002) 1887-1897. Lebovka N., Lisunova M., Mamunya Y.P., Vygornitskii N. “Scaling in percolation behaviour in conductive-insulating composites with particles of different size”, J. Phys.- D: Appl. Phys., 39 (2006) 2264—2271. Mierczynska A., Friedrich J., Maneck H.-E., Boiteux G., Jeszka J.K. “Segregated network polymer/carbon nanotubes composites”, Central Europ. J. Chem, 2 (2004) 363-370. Mironi-Harpaz I., Narkis M. “Electrical behaviour and structure of polypropylene/ultrahigh molecular weight polyethylene/carbon black immiscible blends”, J. Appl. Polym. Sci., 81 (2001) 104-115. Gubbels F., Jerome R., Teyssie P.H. “Selective localization of carbon black in immiscible polymer blends: a useful tool to design electrical conductive composite”, Macromolecules, 27 (1994) 1972-1974. Sandier J.K.W., Kirk J.E., Kinloch I.A., Shaffer M.S.P., Windle A.H. “Ultra-low electrical percolation threshold in carbon-nanotube-epoxy composites”, Polymer, 44 (2003) 5893-5899. Martin C.A., Sandier J.K.W., Shaffer M.S.P., Schwarz M.K., Bauhofer W.,. Schulte K., Windle A.H. “Formation of percolating networks in multiwall carbon-nanotube-epoxy composites”, Compos. Sci. Technol., 64 (2004) 2309-2316. Zou Y., Feng Y., Wang L., Liu X. “Processing and properties of MWCNT/HDPE composites”, Carbon, 42 (2004)271-277. McNally T, Potschke P, Halley P, Murphy M, Martin D, Bell SEJ, Brennan GP, Bein D, Lemoine, Quinn JP. “Polyethylene multiwalled carbon nanotube composites”, Polymer 46 (2005), 8222-32. Li Z, Luo G, Wei F, Huang Y. “Microstructure of carbon nanotubes/PET conductive composites fibers and their properties”, Compos Sci and Technol 66 (2006),1022-9. Smith J.G., Connell J.W., Delozier D.M., Lillehei P.T., Watson K.A., Lin Y., Zhou B., Sun Y.P. “Space durable polymer/carbon nanotube films for electrostatic charge mitigation”, Polymer, 45 (2004) 825-836. Zhang Q, Rastogi S, Chen D, Lippits D, Lemstra PJ. “Low percolation threshold in single-walled carbon nanotube/high density polyethylene composites prepared by melt processing technique”, Carbon 44 (2006),778-85. McLachlan D.S., Chiteme C., Park C., Wise K.E., Lowther S.E., Lillehei P.T., Siochi E.J., Harrison J.S. “AC and DC percolative conductivity of single wall carbon nanotube polymer composites”, J. Polym. Sci. Part B: Polym Phys., 43 (2005) 3273-3287. Stauffer D, Aharony A. “Introduction to Percolation Theory”, London: Taylor and Francis; 1992.

26. Mamunya Y.P., Zois H., Apekis L., Lebedev E.V. “Influence of pressure on the electrical conductivity of metal powders used as fillers in polymer composites”, Powder Technology, 140 (2004) 49-55. 27. Kusy R.P., Corneliussen R.D. “The thermal conductivity of nickel and copper dispersed in poly(vinyl chloride)”, Polym Eng. Sci., 15 (1975) 107-112. 28. Mamunya Ye.P., Davydenko V.V., Zois H., Apekis L., Snarskii A.A., Slipchenko K.V. “Dielectric properties of polymers filled with dispersed metals” Polymers & Polymer Composites, 10 (2002) 219-228. 29. Zois H., Apekis L., Mamunya Ye.P. “Dielectric properties and morphology of polymer composites filled with dispersed iron”, J. Appl. Polym. Sci., 88 (2003) 3013-3020. 30. Liang G.D., Tjong S.C. “Electrical properties of low-density polyethylene/multiwalled carbon nanotube nanocomposites”, Mater. Chem. & Phys., 100 (2006), 132-137.

27