FILLER STRUCTURE ANISOTROPY AND ELECTRICAL...

10ο ΠΑΝΕΛΛΗΝΙΟ ΕΠΙΣΤΗΜΟΝΙΚΟ ΣΥΝΕΔΡΙΟ ΧΗΜΙΚΗΣ ΜΗΧΑΝΙΚΗΣ, ΠΑΤΡΑ, 4-6 ΙΟΥΝΙΟΥ, 2015.

FILLER STRUCTURE ANISOTROPY AND ELECTRICAL CONDUCTIVITY ENHANCEMENT

IN ELECTRIFIED CARBON NANOTUBE/THERMOPLASTIC POLYMER COMPOSITES

O. OSAZUWA, A.A. VASILEIOU, M. KONTOPOULOU, A. DOCOSLIS

Department of Chemical Engineering, Queen’s University, Kingston, Ontario K7L 3N6, Canada

ABSTRACT The effects of electric fields on the filler response dynamics and electrical percolation of poly(ethylene succinate)/multiwall carbon nanotube (MWCNT) composites are examined here for the first time. When subjected to AC electric fields in their melt state PESu/MWCNT composites exhibited dramatic improvements in their transverse electrical conductivity. Overall, the experimental results show that the electrified composites exhibit the same electrical conductivity levels as their non-electrified counterparts at approximately 3-fold less filler content. The dynamics of the insulator-to-conductor transition under an electric field are found to correlate reasonably well with operating parameters, such as electric field intensity, matrix viscosity and filler content through a relatively simple model. Such a model can serve as an enabling tool in the determination of process conditions for the manufacturing of electrically conducting MWCNT/polymer composites.

INTRODUCTION In the last decade, research has shown that polymer composites containing aligned multi-wall carbon nanotubes (MWCNTs) can exhibit enormous improvement in their electrical, thermal, and mechanical properties along the direction of MWCNT alignment at significantly lower filler concentration [2,3]. It has been demonstrated that externally applied alternating current (AC) or direct current (DC) electric fields can cause MWCNT alignment, end-to-end assembly, and formation of continuous networks inside liquid polymers or polymer melts [4-8].

Chen et al. [9] achieved an aligned single-wall carbon nanotube (SWCNT) array between two parallel, planar finger electrodes by applying an external AC electric field on the SWCNT suspension. They observed that the alignment of the SWCNT was dependent on several parameters of the applied electric field, such as the magnitude, frequency and duration of exposure. Many researchers used uncured epoxy resins, where alignment is more easily attainable due to low matrix viscosities [6,7,10]. Martin et al. [4] showed that an AC electric field was more efficient compared to a DC field in aligning CNT, and the resulting CNT networks were more uniformly distributed. It was observed that MWCNT were rapidly aligned in a polymer suspension [6,11] after the electric field was exerted, but then they gradually agglomerated over time and formed thick bundles. Some preliminary studies also exist on thermoplastics, such as poly(methyl methacrylate) PMMA [12]. In spite of all this promising research work, the full potential of electrical forces in structuring MWCNT-filled polymers has not been reached yet.

Recently [8], we reported on the effects of AC electric fields on the electrical properties of polyolefin-based carbon nanotube composites and derived a correlation between intensity (applied voltage), polymer viscosity, and filler concentration on the time required for an insulator-to-conductor transition during the electrification processes. We also proposed a selection criterion for processing conditions that permit carbon nanotube alignment and assembly under an electric field controlled regime, while limiting the effect of randomly driven nanotube aggregation.

In this work we focus our attention on the equally important, but largely unexplored, category of carbon nanotube/polyester composites. Aliphatic polyesters are an industrially relevant class of thermoplastics with physical properties comparable to those of polypropylene and low density polyethylene and the added advantage of biodegradability. Only limited work currently exists on the properties of isotropic blends of CNTs with, mainly, poly(ethylene terephthalate) [13-19], poly(butylene terephthalate) [20] and poly(butylene succinate) matrices. Finally, the only studies that are found on poly(ethylene succinate)/multi-wall carbon nanotube (PESu/ MWCNT) composites involve composites prepared with the method of solution casting. More importantly, we find no reports in the open literature on how the exposure to an external electric field can change the filler structure and physical properties of such composites.

This work is a first attempt to systematically study the effects of electric fields on the transient filler response and resulting electrical properties of polyester-based carbon nanotube composite melts. More specifically, the effects of important processing parameter, such as applied voltage and filler concentration on the transient filler response and solid state electrical conductivity of poly(ethylene succinate)/multi-wall carbon nanotube (PESu/MWCNT) composites are investigated. To achieve good initial MWCNT dispersion inside the polymer matrix, an in situ polymerization approach was employed to synthesize these composites.


EXPERIMENTAL 1. Materials Succinic acid (Su) (purum 99+ %), 1,2-ethanediol (Et) (Purity:>99.8 %) and titanium (IV) butoxide (≥97%) (TBT) were obtained from Sigma-Aldrich and used as received. Triphenylphosphate (TPP) (95%) was purchased from Fluka. MWCNTs (purity ≥98%, diameter 10 ± 1 nm and length 3- 5 µm) were purchased from Sigma-Aldrich and used as received. All other materials and solvents used in the analytical methods were of analytical grade.

2. In situ preparation of composites PESu/MWCNT nanocomposites were prepared in a glass batch reactor by the two-stage melt polycondensation of succinic acid and 1,2-ethanediol at a mole ratio of 1:1.1 .

The MWCNTs were pre-dried at 110 oC under vacuum for 24 h. Appropriate amount of MWCNTs corresponding to a final 0.1 – 2.0 wt% content in the composite was first dispersed in 1,2-ethanediol by intense stirring with a magnetic stirrer (300 rpm) and ultrasonication (25W) for 15 min prior to polymerization. The polymerization mixture was placed in a 250 cm3 round bottomed flask, degassed and purged with dry argon three times. Thereupon, the mixture was heated under an argon atmosphere for 3 h at 190 °C under constant stirring (350 rpm), with water removed as the reaction by-product of esterification.

For the second reaction stage of polycondensation, 0.3 wt% of triphenylphosphate (TPP) as heat stabilizer and 1.0×10-3 mol per mole of succinic acid of titanium (IV) butoxide (TBT) as polycondensation catalyst were added. The reaction was continued under increased mechanical stirring (700 rpm) and high vacuum (~2.0 Pa), which was applied slowly over a period of time of about 30 min, to avoid excessive foaming and to minimize oligomer sublimation, at 220 and 240 °C for 1.5 and 1 h intervals, respectively. The polymerization was stopped by fast cooling to room temperature.

3. Characterizations Intrinsic viscosity measurements of the prepared PESu and the composites were performed using an Ubbelohde viscometer at 25 oC in chloroform at a solution concentration of 1 wt%. The intrinsic viscosities [η] were calculated using the Solomon-Ciutâ equation [21]:

[η] = (1)

where C is the concentration of the solution, t the flow time of the solution and t0 the flow time of the solvent. Each sample was filtered through a nylon microfilter with a pore size of 0.2 µm before the measurements. Corrections were made to the concentration value used for the calculation of the intrinsic viscosities due to the presence of the filler. Number average molecular weights were estimated from the intrinsic viscosity values using the Berkowitz equation: 𝑀! = 3.29 x 10! 𝜂 !.!" [22].

Ultra-thin films of the composites for TEM characterization were prepared using a Leica ultra-microtome. A FEI Tecnai 20 instrument at an operating voltage of 200 keV was used for TEM imaging. The composite melts were also observed using an Olympus BX 51 optical microscope (Tokyo, Japan). Composite films were loaded on a Linkam SCC 450 Hot Stage (Surrey, UK) at 160 oC and pressed to a thickness of 25 µm. Images were recorded immediately once the desired thickness was reached (5 min), using transmitted light.

Volume resistivity was measured under DC current at room temperature. Samples were prepared by compression molding the melt-compounded composites in a Carver press at 140 oC and 10 MPa to get a thin film of 2.0 mm. Thin composite films with a diameter of 6 cm were placed inside the measuring chamber (Keithley 8009 Resistivity Test Fixture) of the Keithley 6517B Electrometer/High Resistance Meter (Keithley Instruments, Inc., Cleveland, Ohio, USA) for an electrification time of 1 minute. An Agilent 34401A 6 ½ Digit multimeter was used for resistivities lower than 105 Ω·m. Sample preparation included cutting test specimens with a dimension of 1 cm2 and thickness of 2 mm and gold sputtering the surface to reduce contact resistance between the sample and the electrodes of the multimeter. The edges of the samples were trimmed after gold sputtering to prevent short-circuit during the measurement of the resistance (R). The conductivity of the samples was determined from the inverse of resistivity, ρ:

ρ=RxA/l (2) where A the contact surface area and l the average sample thickness. Rheological characterization was carried out on a Reologica ViscoTech oscillatory rheometer using 20

mm parallel plate fixtures, with a gap of 1 mm at 120 oC. Compression molded disks with a diameter of 20 mm were prepared using the Carver press as described above. Stress sweep experiments were carried out from 1 to 103 Pa at a frequency of 0.1 Hz and temperature of 120 oC to identify the limits of linear viscoelasticity.

Film electrification was performed as described in our previous work [8]. In this case, electrification was carried out at 90 oC with field intensity ranging from 35 – 176 kV/m at 1 kHz.

Ctt

tt )1ln(2

00

−−×


RESULTS 1. Preparation of PESu/MWCNTs composites PESu synthesized in situ by the two-stage polycondensation method had a molecular weight of 31,800 Da, corresponding to a melt viscosity of 167 Pa.s. The presence of MWCNTs in the polymerization mixture affected the polycondensation reaction and led to small variations in the final molecular weights of the polymers. A major contributing factor to the observed variation was probably the overall increase in the viscosity of the polymerization mixture due to the presence of the MWCNTs. This can alter step change growth polymerization kinetics even at low conversion levels. The observed molecular weight variations are however small enough and should not invoke any significant effect over the properties of the final composites.

Optical microscopy images (Figures 1a and 1b) show a uniform filler distribution inside the matrix, while TEM imaging (Figures 1c) reveals that individual nanotubes co-exist with swollen MWCNT aggregates of submicron dimensions.

2. Solid state electrical conductivity The effect of filler content on the transverse electrical conductivity of the as prepared composites is shown in Fig. 2. It can be seen that the composites exhibit improvements in the transverse conductivity by 12-13 orders of magnitude compared to the neat polymer. Composite films with conductivities as high as 10-2 - 10-1 S/m were achieved, which fall in the range of static dissipative materials [23,24]. Higher MWCNT concentrations (up to 2.0 wt%) did not improve conductivity further.

The percolation threshold of the composites was estimated by fitting the conductivity data, using power-law relations above and below the critical concentration for percolation [24]:

σ= σmatrixϕc- ϕ

ϕc

-s ϕ<ϕc (3a)

σ=mϕ-ϕc1-ϕc

t ≈m ϕ-ϕc

t ϕ>ϕc (3b)

where 𝜎 is the electrical conductivity of the composite, 𝜎!"#$%& is the conductivity of the matrix, 𝜙 is the volume fraction of the filler, 𝜙! is critical volume fraction at percolation, s and t are the critical exponents below and above percolation, respectively and m is a constant. The percolation threshold of the as prepared composites was found to be 0.63 wt%. The resulting curve fits are included in Figure 2.

Fig. 2 also includes the solid state conductivity variation with respect to filler content for the composites exposed to an electric field of 176 kVm-1. A comparison between the two data sets reveals a dramatic increase in the conductivity of the electrified PESu-based composites. The increase is especially pronounced at filler concentrations below the percolation threshold of the as prepared composites (0.63 wt%), where conductivity differences of up to 11 orders of magnitude can be observed. From a practical standpoint, this means that between 2- and 3-fold higher filler amounts are need in the case of the as prepared composites in order to match the conductivity levels of their electrified counterparts.

Percolation threshold calculations performed for the case of electric field treated composites highlight an additional advantage of composite electrification. Specifically, the use of an electric field was found to decrease the composites’ percolation threshold from 0.63 wt% to 0.26 wt%, i.e., by almost 2.5 times.

Figure 1: (a, b) Optical microscopy and (c) TEM images of PESu/1.0 wt% MWCNT composite films.


3. Dynamic response of electrified composites The insulator-to-conductor transition dynamics of the MWCNT/polyester composites during melt electrification was studied as a function of filler content and electric field intensity. A representative dynamic response of the composites’ electrical conductivity during the application of an electric field is illustrated in Figure 4, where the results are displayed in terms of the potential drop across the electrode cell for PESu/0.1 wt% MWCNT. The graph depicts a typical response of electrified composites, whereby a sharp first drop in resistivity is followed by a second gradual drop and finally a plateau phase. Graphs such as this can be used for the calculation of a characteristic insulator-to-transition parameter, termed “percolation time” (tP). Percolation time here is defined as the time span between the first moment an electric field is applied to the composite film and the time point at which the first sharp drop in the composite’s resistivity (or, equivalently, electrical potential difference across the cell) occurs. An example illustrating the graphical calculation of tp is included in Figure 3.

The effect of applied voltage on the percolation time for composites with MWCNT concentrations ranging for 0.1 and 0.25 wt% is summarized in Figure 4a (percolation times could not be recorded for composites with higher MWCNT concentrations due to the extremely fast (< 1 𝑠) response to the electric field). The electric field intensity values correspond to the range of applied voltages between 100 and 500 Vpp, which translate to 35 and 176 kV/m. The plotted data show a power law relation of the type tp ∝ Ea (a<0), i.e., the percolation time decreases with increasing electric field intensity. These data do not obey a single power law relation with respect

Figure 2: Volume conductivity of PESu/MWCNT samples as a function of MWCNT concentration before and after electrification (conditions: 176 kV/m, 1 kHz, 90 oC). Solid lines are curve fits according to Eqns. 3a and 3b.

Figure 3: Voltage drop observed across the electrode cell as a function of time for PESu/0.1 wt% MWCNT composite melts at 176 kV/m. A graphical definition of percolation time is included.


to the field intensity; rather, the value of the exponent is decreasing with increasing filler concentration. Data regression yields an exponent value of -1.0 for the 0.1 wt% sample and approximately -2.0 for the 0.25 wt% sample [8].

Our previous study on the dynamics of MWCNT aggregation inside polyolefin composite melts showed that, under certain conditions, MWCNTs tend to align in the direction of the field and form long, electrically conducting pathways [8]. This preferred, anisotropic filler organization, which results in composites with enhanced electrical conductivity, can only be produced from composite melts subjected to an electric field controlled regime, i.e., to (E, φ, Τ) conditions that allow carbon nanotube dielectrophoresis to be much faster than Brownian motion. The desired (E, φ, Τ) combinations can be calculated using the following criterion: [8]:

ϕ0.5E2

T≥10 x

96kBR[ln(LR)-1]

2

εm,r ε0 π0.5 L4 (4)

where T is temperature of the melt, k! is Boltzmann’s constant, L and R denote the length and radius of the MWCNT, ε!,! is the relative permittivity of the melt and ε! is the permittivity of free space.

Moreover, the percolation times of all MWCNT/polyolefin composites subjected to an electric field controlled regime obeyed the following correlation:

tP ∝ ηE-2ϕ-1.5 (5) where η is the viscosity of polymer matrix. The present work aims to examine whether Expression (5) also applies in the case of MWCNT/polyester

composites. To accomplish this, a number of PESu/MWCNT composites with filler concentrations between 0.1 and 0.25 wt% were prepared. Composites with higher MWCNT concentrations could not be used since their respective percolation times were extremely short (< 1 𝑠) to be recorded accurately.

Figure 4: (a) Log-log plot of percolation time vs. field intensity (1 kHz, 90 oC) for PESu/0.1 wt% and 0.25 wt% MWCNT composite melts over a range of filler content. Solid lines represent the power law equation and circles the conditions under which MWCNT assembly occur under a mixed regime. (b) Experimentally measured electric field induced percolation times vs. their theoretical dependence on (η, E, φ) conditions. The solid markers correspond to data sets obtained under an electric-field controlled regime.


The experimentally calculated percolation times of the MWCNT/polyester composites were plotted against ηE!!ϕ!!.!, as shown in Figure 4b. As can be seen, the data are divided into two groups (solid and open squares). The solid squares correspond to samples treated under an electric field controlled regime. In the present case, the RHS of expression (4) was equal to 3.7 x 106 V2m-2 K-1; i.e., for all these samples, the processing conditions were chosen so as φ0.5E2T-1 > 3.7 x 106. On the other hand, the open squares encompass samples treated under a mixed or Brownian motion controlled regime (φ0.5E2T-1 < 3.7 x 106). Linear regression of data captured by the solid markers produces a trendline with a slope equal to 1.19, which indicates a relatively good agreement between the experimental results and the theoretical model (Expression 5). The data sets that did not meet the selection criterion are also plotted in Figure 4b (open squares). As seen in the graph, these points lie far away from the extrapolated trend.

It must be emphasized that the theoretical model assumes perfect filler dispersion. However, as evidenced in Figure 1, the MWCNTs inside the PESu are not fully disentangled and evenly distributed. The lack of perfect dispersion will also have an impact on the actual size and aspect ratio of the carbon nanotubes and, consequently, on the RHS value of the Expression (4), according to which the data were divided into two groups. A much better agreement between theory and experiments was found in our previous study, where the same batch of MWCNTs was first surface-compatibilized and then finely dispersed in a polyolefin-based matrix [8]. When the same selection criterion was applied, the slope of the corresponding trendline was found equal to 1.04.

In conclusion, the results of the present study reveal that the insulation-to-conductor transition times of electrified MWCNT/polyester melts, under an electric field-control regime, can be correlated with the processing conditions by means of a theoretically derived dynamic percolation model. In spite of the co-existence of finely distributed MWCNTs with MWCNT aggregates inside the polyester matrix, a relatively accurate correlation between percolation time and various (E, φ, η(Τ)) combinations could still be achieved. Such correlations are valuable tools as they give us the ability to predict the characteristic times of MWCNT assembly, which is a major advantage in experimental design or in the determination of process conditions for the manufacturing of electrically conducting MWCNT/polymer composites. CONCLUSIONS The application of an electric field on the transverse conductivity of PESu/MWCNT composite films was studied in this work. The use of in situ polymerization produced composites with isotropic filler distribution and relatively fine initial dispersion. PESu/MWCNT composites subjected to an electric field of 176 mVm-1 exhibited much higher level of conductivity compared with their untreated counterparts. Conductivity differences up to 11 orders of magnitude were recorded at concentrations around 0.5 wt%. It was shown that composite electrification can reduce the required filler amount up to 3 times.

A major finding of this work is that this method of preparation and electrification of the PESu/MWCNT composites produced electrically conductive composites at very low levels of filler concentration by lowering the percolation threshold by more than a half.

A correlation between percolation times and various (E, φ, η(Τ)) conditions that was previously shown to hold true for MWCNT/polyolefin composites is now found to also apply in the case of MWCNT dispersed in polyesters, albeit less accurately due to the incomplete filler disentanglement in the latter case. A generalized correlation that allows the a priori estimation of the experimental/processing conditions for a large number of composites is very enabling in experimental design, but more so in large-scale applications, where the composite’s exposure time to an electric field may be limited by other processing restrictions. In those cases, the operator will be able to explore applied voltage, filler composition and processing temperature combinations that are adaptable to an existing process.

ACKNOWLEDGEMENTS The authors would like to acknowledge the financial support provided by E.I. DuPont Canada and Natural Science and Engineering Research Council, Collaborative Research & Development grant.

REFERENCES [1]. Ajayan P, Stephen O, Colliex C, Trauth D. Aligned Carbon Nanotube Arrays Formed by Cutting a Polymer Resin-Nanotube Composite. Science 1994;265:1212. [2]. Lanticse L, Tanabe Y, Matsui K, Kaburagi Y, Suda K, Hoteida M, Endo M, Yasuda E. Shear-induced preferential alignment of carbon nanotubes resulted in anisotropic electrical conductivity of polymer composites. Carbon 2006;44(14):3078-3086. [3]. Tai X, Wu G, Tominaga Y, Asai S, Sumita M. An approach to one-‐‑dimensional conductive polymer composites. J Polym Sci B Polym Phys 2005;43(2):184-189. [4]. Martin C, Sandler J, Windle A, Schwarz M-, Bauhofer W, Schulte K, Shaffer M. Electric field-induced aligned multi-wall carbon nanotube networks in epoxy composites. Polymer 2005;46(3):877-886.


[5]. Park C, Wilkinson J, Banda S, Ounaies Z, Wise KE, Sauti G, Lillehei PT, Harrison JS. Aligned single-‐‑wall carbon nanotube polymer composites using an electric field. J Polym Sci B Polym Phys 2006;44(12):1751-1762. [6]. Ma C, Zhang W, Zhu Y, Ji L, Zhang R, Koratkar N, Liang J. Alignment and dispersion of functionalized carbon nanotubes in polymer composites induced by an electric field. Carbon 2008;46(4):706-710. [7]. Kovacs J, Mandjarov R, Blisnjuk T, Prehn K, Sussiek M, Müller J, Schulte K, Bauhofer W. On the influence of nanotube properties, processing conditions and shear forces on the electrical conductivity of carbon nanotube epoxy composites. Nanotechnology 2009;20(15):155703. [8]. Osazuwa O, Kontopoulou M, Xiang P, Ye Z, Docoslis A. Electrically conducting polyolefin composites containing electric field-aligned multiwall carbon nanotube structures: The effects of process parameters and filler loading. Carbon 2014;72(1):89-99. [9]. Chen C, Zhang Y. Manipulation of single-wall carbon nanotubes into dispersively aligned arrays between metal electrodes. J Phys D: Appl Phys 2006;39(1):172-176. [10]. Park C, Wilkinson J, Banda S, Ounaies Z, Wise K, Sauti G, Lillehei P, Harrison J. Aligned single-‐‑wall carbon nanotube polymer composites using an electric field. J Polym Sci B Polym Phys 2006;44(12):1751-1762. [11]. Yang X, Zhu Y, Ji L, Zhang C, Liang J. Influence of AC Electric Field on Macroscopic Network of Carbon Nanotubes in Polystyrene. J Dispersion Sci Technol 2007;28(8):1164-1168. [12]. Valentini L, Bon S, Kenny J. Anisotropic Electrical Transport Properties of Aligned Carbon Nanotube/PMMA Films Obtained by Electric-Field-Assisted Thermal Annealing. Macromol Mater Eng 2008;293(11):867-871. [13]. Logakis E, Pissis P, Pospiech D, Korwitz A, Krause B, Reuter U, Pötschke P. Low electrical percolation threshold in poly(ethylene terephthalate) / multi-walled carbon nanotube nanocomposites. European Polymer Journal 2010;46:928-936. [14]. Hu G, Zhao C, Zhang S, Yang M, Wang Z. Low Percolation Thresholds of Electrical Conductivity and Rheology in Poly(ethylene terephthalate) through the Networks of Multi Walled Carbon Nanotubes. 2006;47:480-488. [15]. Anoop AK, Agarwal US, Joseph R. Carbon nanotubes-reinforced PET nanocomposite by meltcompounding. J Appl Polym Sci 2007;104(5):3090-3095. [16]. Anoop AK, Agarwal US, Nisal A, Joseph R. PET-SWNT nanocomposites through ultrasound assisted dissolution-evaporation. European Polymer Journal 2007;43:2279-2285. [17]. Steinert BW, Dean DR. Magnetic field alignment and electrical properties of solution cast PET–carbon nanotube composite films. Polymer 2009;50(3):898-904. [18]. Hernández JJ, García-Gutiérrez MC, Nogales A, Rueda DR, Kwiatkowska M, Szymczyk A, Roslaniec Z, Concheso A, Guinea I, Ezquerra TA. Influence of preparation procedure on the conductivity and transparency of SWCNT-polymer nanocomposites. Composites Sci Technol 2009;69(11-12):1867-1872. [19]. Gupta A, Choudhary V. Electromagnetic interference shielding behavior of poly(trimethylene terephthalate)/multi-walled carbon nanotube composites. Composites Sci Technol 2011;71(13):1563-1568. [20]. Nogales A, Broza G, Roslaniec Z, Schulte K, Šics I, Hsiao BS, Sanz A, García-Gutiérrez MC, Rueda DR, Domingo C, Ezquerra TA. Low Percolation Threshold in Nanocomposites Based on Oxidized Single Wall Carbon Nanotubes and Poly(butylene terephthalate). Macromolecules 2004;37(20):7669-7672. [21]. Vassiliou AA, Bikiaris D, El Mabrouk K, Kontopoulou M. Effect of evolved interactions in poly(butylene succinate)/fumed silica biodegradable in situ prepared nanocomposites on molecular weight, material properties, and biodegradability. J Appl Polym Sci 2011;119(4):2010-2024. [22]. Achilias DS, Bikiaris DN, Karavelidis V, Karayannidis GP. Effect of silica nanoparticles on solid state polymerization of poly(ethylene terephthalate). Eur Polym J 2008;44(10):3096-3107. [23]. Al-Saleh M, Sundararaj U. An innovative method to reduce percolation threshold of carbon black filled immiscible polymer blends. Composites Part A 2008;39:284-293. [24]. Vasileiou AA, Docoslis A, Kontopoulou M, Xiang P, Ye Z. The role of non-covalent interactions and matrix viscosity on the dispersion and properties of LLDPE/MWCNT nanocomposites. Polymer 2013;54(19):5230-5240.

FILLER STRUCTURE ANISOTROPY AND ELECTRICAL...

Documents

Transcript of FILLER STRUCTURE ANISOTROPY AND ELECTRICAL...