Elsevier

Polymer

Volume 46, Issue 19, 8 September 2005, Pages 8222-8232
Polymer

Polyethylene multiwalled carbon nanotube composites

https://doi.org/10.1016/j.polymer.2005.06.094Get rights and content

Abstract

Polyethylene (PE) multiwalled carbon nanotubes (MWCNTs) with weight fractions ranging from 0.1 to 10 wt% were prepared by melt blending using a mini-twin screw extruder. The morphology and degree of dispersion of the MWCNTs in the PE matrix at different length scales was investigated using scanning electron microscopy (SEM), transmission electron microscopy (TEM), atomic force microscopy (AFM) and wide-angle X-ray diffraction (WAXD). Both individual and agglomerations of MWCNTs were evident. An up-shift of 17 cm−1 for the G band and the evolution of a shoulder to this peak were obtained in the Raman spectra of the nanocomposites, probably due to compressive forces exerted on the MWCNTs by PE chains and indicating intercalation of PE into the MWCNT bundles. The electrical conductivity and linear viscoelastic behaviour of these nanocomposites were investigated. A percolation threshold of about 7.5 wt% was obtained and the electrical conductivity of PE was increased significantly, by 16 orders of magnitude, from 10−20 to 10−4 S/cm. The storage modulus (G′) versus frequency curves approached a plateau above the percolation threshold with the formation of an interconnected nanotube structure, indicative of ‘pseudo-solid-like’ behaviour. The ultimate tensile strength and elongation at break of the nanocomposites decreased with addition of MWCNTs. The diminution of mechanical properties of the nanocomposites, though concomitant with a significant increase in electrical conductivity, implies the mechanism for mechanical reinforcement for PE/MWCNT composites is filler-matrix interfacial interactions and not filler percolation. The temperature of crystallisation (Tc) and fraction of PE that was crystalline (Fc) were modified by incorporating MWCNTs. The thermal decomposition temperature of PE was enhanced by 20 K on addition of 10 wt% MWCNT.

Introduction

The identification in 1991 of carbon nanotubes (CNTs) by Iijima [1] has stimulated intense research interest in the structure [2], [3], [4], [5], [6], [7], [8], properties [9], [10], [11], [12], [13] and applications [14], [15], [16], [17] of these unique materials. The intrinsic superconductivity [9], field emission behaviour [10], potential as molecular quantum wires [11], ability to store hydrogen [12], unusually high thermal conductivity [13], use as sensors for gas detection [16] and the biocompatibility and potential for biomolecular recognition [17] of carbon nanotubes has been reported. However, it is the combination of exceptional conductivity (electrical and thermal), low density and mechanical properties [16] of CNTs that has resulted in their use in filled composites. Both theoretical and experimental studies have shown CNTs to have extremely high tensile moduli (>1 TPa for single walled carbon nanotubes, SWCNTs) and tensile strengths of the order of 500 GPa [18], [19]. Carbon nanotubes are thermally stable up to 2400 °C in vacuo, have a thermal conductivity about double that of diamond and electric-current-carrying capacity 1000 times higher than copper wire [19]. The reported exceptional properties of carbon nanotubes have stimulated several groups to investigate both experimentally and theoretically the preparation and properties of polymer carbon nanotube composite materials. Polymer carbon nanotube composites will find many applications, most important of these will as structural materials due to their low density, where mechanical reinforcement or increased electrical conduction are required, as selectively permeable membranes, in electromagnetic induction shielding and in bio-molecule and drug delivery.

Irrespective of the method of preparation there are two fundamental and critical issues associated with translating or transferring the unique properties of carbon nanotubes to a polymer matrix. Firstly, the nanotubes must be uniformly distributed and dispersed throughout the polymer matrix, and secondly, there must be enhanced interfacial interaction/wetting between the polymer and the nanotubes. For example, any load applied to the polymer matrix should be transferred to the nanotube. This load relies on the effective interfacial stress transfer at the polymer–nanotube interface, which tends to be polymer dependent [20]. Three general approaches have been adopted in attempts to modify the surface of CNTs to promote such interfacial interactions; chemical, electrochemical and plasma treatment. For example, Castaño et al. [21] placed different organo-functional groups on MWCNTs using an oxidation and silanization process. Haiber et al. [22] modified the surface of CNTs using low-pressure oxygen plasma treatment and using X-ray photoelectron spectroscopy (XPS) detected hydroxide, carbonyl and carboxyl functionality on the surface layers of the CNTs.

There are several challenges to overcome with regard the processing of polymer carbon nanotube composites. Nanotubes, whether bundles of SWCNTs or aggregates of MWCNTs tend to agglomerate and its difficult to separate individual nanotubes during mixing. While high power ultrasonic mixers [23], surfactants, solution mixing [24] and in situ polymerisation have all been used to produce polymer carbon nanotube composites, these techniques have many limitations, including that they may not be commercially viable and are environmentally contentious. However, to date the preparation of polymer carbon nanotube composites using melt blending/extrusion has not been widely reported.

Advani et al. [25] reported some improvements in stiffness and work to failure of HDPE/MWCNT composites that were prepared using a multi-step twin-screw extrusion procedure. Andrews et al. [26] using poor shear melt mixing, obtained modest increases in elastic modulus and a decrease in tensile strength for polypropylene, polystyrene and acrylonitrile/butadiene/styrene (ABS)/MWCNT composites. CNTs have also been shown to alter the crystallisation kinetics of semi-crystalline polymers [27], [28]. More recently, Shaffer et al. [29] melt blended polyamide-12 with MWCNTs and carbon fibres using a twin-screw microextruder, then produced fibres of the blends prepared. They highlighted that both the intrinsic crystalline quality of the nanocomposite and the orientation of the embedded CNTs are major factors controlling the reinforcing capability of CNTs. CNT reinforced Nylon-6 composites prepared by melt compounding having 120% improvements in tensile modulus and strength compared to virgin Nylon 6 was reported by Zhang et al. [30] Pötschke et al. have studied the morphology [31], electrical [32] and rheological [33] properties of polycarbonate (PC) based MWCNT composites and most notably obtained a 107 Ω cm reduction in volume resistivity for a PC/PE blend on addition of 0.41 vol% MWCNT. In this paper we report the preparation of PE/MWCNT composites using melt blending and investigate the morphology and dispersion of MWCNTs in the PE matrix using a combination of microscopy, WAXD and Raman spectroscopy. The electrical conductivity of PE/MWCNT composites is measured, and combined with melt state oscillatory rheology measurements the percolation threshold for this system is determined. The mechanical, crystallisation and thermal decomposition properties of the nanocomposites are also discussed.

Section snippets

Materials

The polyethylene (PE) used in this study was a third generation linear medium density metallocene catalysed polyethylene (Borecene RM8343) kindly provided by Boreaslis A/S in powder form, with a MFR=6 g/10 min (190 °C/2.16 kg) and ρ=934 kg/m3. The multiwalled carbon nanotubes (MWCNTs) were supplied by Sun Nanotech Co. Ltd, People's Republic of China. The MWCNTs were prepared using a chemical vapour deposition (CVD) process, using acetylene as a carbon source, a Fe and Ni catalyst system and a

Results and discussion

The morphology and the degree of dispersion of MWCNTs in the polyethylene matrix at different length scales were studied using a combination of SEM, HRTEM, AFM and WAXD. Fig. 1A–D shows the SEMs of bundled MWCNTs and typical cryo-fractured surfaces of PE with 10 wt% MWCNT taken parallel and perpendicular to the flow direction, at ×5000 and ×30,000 magnification. The MWCNTs can be clearly identified and are uniformly dispersed as single nanotubes and as aggregates of varying dimensions. In some

Conclusions

PE/MWCNT nanocomposites were prepared using twin-screw melt compounding. Microscopic observations across the length scales and WAXD indicate that the MWCNTs are very well distributed and dispersed in the PE matrix. Both individual MWCNTs and agglomerations of MWCNTs were evident. The PE intercalated into the MWCNT bundles, detected by an up-shift of 17 cm−1 in the G band and the evolution of a shoulder to this peak. Remarkably, the electrical conductivity of PE, the most electrically insulating

Acknowledgements

The authors acknowledge the financial support of the Royal Society (574006/G503/24135) and the Nuffield Foundation (NAL/00696/G). We thank Borealis for providing the PE and Cormac Byrne, Peter Boyd, Michael Lewis, Bronagh Millar, Jacqueline Patrick and Stephen McFarland for technical assistance.

References (51)

  • S. Iijima

    Phys B

    (2002)
  • E.T. Thostenson et al.

    Compos Sci Technol

    (2001)
  • H. Bubert et al.

    Diamond Relat Mater

    (2003)
  • L. Valentini et al.

    Diamond Relat Mater

    (2003)
  • Y. Zou et al.

    Carbon

    (2004)
  • A.R. Bhattacharyya et al.

    Polymer

    (2003)
  • J.W.K. Sandler et al.

    Polymer

    (2004)
  • P. Pötschke et al.

    Polymer

    (2003)
  • P. Pötschke et al.

    Eur Polym J

    (2004)
  • P. Pötschke et al.

    Polymer

    (2002)
  • L. Valentini et al.

    Compos Sci Technol

    (2003)
  • C.A. Martin et al.

    Compos Sci Technol

    (2004)
  • O. Meincke et al.

    Polymer

    (2004)
  • P. Pötschke et al.

    Polymer

    (2004)
  • A.R. Bhattacharyya et al.

    Chem Phys Lett

    (2004)
  • S. Iijima

    Nature

    (1991)
  • S. Iijima et al.

    Nature

    (1993)
  • B. Vigolo et al.

    Science

    (2000)
  • J. Sloan et al.

    Chem Commun

    (1998)
  • A. Thess et al.

    Science

    (1996)
  • N. Hamada et al.

    Phys Rev Lett

    (1992)
  • J.W. Mintmire et al.

    Phys Rev Lett

    (1992)
  • M. Kociak et al.

    Phys Rev Lett

    (2001)
  • A.G. Rinzler et al.

    Science

    (1995)
  • S.J. Tans et al.

    Nature

    (1997)
  • Cited by (789)

    • A critical review on intrinsic conducting polymers and their applications

      2023, Journal of Industrial and Engineering Chemistry
    View all citing articles on Scopus
    View full text