Effect of fabrication process on electrical properties of polymer/multi-wall carbon nanotube nanocomposites

Abstract

Polymer/carbon nanotubes nanocomposites were fabricated by an in situ polymerization process using multi-wall carbon nanotubes (MWNT) as filler in an epoxy polymer. Effects of curing process, mixing speed, mixing time, addition of ethanol, timing of hardener addition, etc., in the fabrication process on the electrical properties of nanocomposites have been investigated. In the fabrication process, the effective formation of macroscopic conducting network in matrix is most important to enhance the electrical properties of nanocomposites. It was found that the curing temperature and the mixing conditions are key factors in the fabrication process, which influence the formation of conducting network significantly. Therefore, careful design of these factors in the fabrication process is required to achieve high electrical performances of nanocomposites. The experimental percolation threshold of the resultant nanocomposites was around 0.1 wt%. Moreover, a statistical percolation model was built up to numerically investigate the percolation threshold. The experimental electrical conductivity increases from the percolation threshold following a percolation-like power law with the identified critical exponent t as 1.75.

Introduction

Recently, much attention has been paid to the fabrication of nanocomposites with the use of carbon nanotubes (CNT) in polymer materials to harness the exceptional intrinsic properties of CNT. In particular, polymers with the incorporation of CNT show great potential for electronic device applications, such as organic field emitting displays, photovoltaic cells, highly sensitive strain sensors, electromagnetic interference materials, etc. In the recent decade, numerous studies on the electrical properties of nanocomposites made from insulating polymers filled by CNT have been carried out.

Currently, melt mixing compounding [1], [2], [3], [4], curing/in situ polymerization [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17] and coagulation [18], [19] are widely used to prepare the nanocomposites. Depending on the polymer matrix and processing technology as well as the type of nanotube material used, as reviewed in [12], percolation thresholds ranging from less than 1.0% to over 10.0 wt% of CNT loading have been observed experimentally. For example, for single-wall carbon nanotubes (SWNT), Nogales et al. [5] applied in situ polycondensation reaction to prepare PBT/SWNT nanocomposites and achieved electrical percolation threshold as low as 0.2 wt% of SWNT loading. Ounaies et al. [6] have investigated the electrical properties of SWNT reinforced polyimide (CP2) composites. The obtained conductivity obeys a percolation-like power law with a low percolation threshold of around 0.1 wt%. The bundling phenomenon of SWNT within the matrix has been identified in experimental analysis. Park et al. [7] have shown that it is possible to control the electrical properties of polymer/SWNT composites through a field alignment technique for SWNT. Kymakis et al. [8] studied the electrical properties of SWNT and the soluble polymer poly (3-octylthiophene) (P3OT). The reported percolation threshold was around 11 wt%. In their latter work [9], purified SWNT were used to get a much lower percolation threshold at around 4 wt%.

For multi-wall carbon nanotubes (MWNT), Sandler [10] have employed MWNT with a bisphenol-A epoxy resin and an aromatic hardener, and they got the low percolation threshold of 0.04 wt%. The formation of aggregates was also identified. Sandler et al. [11] reported the lowest percolation threshold to date, i.e., 0.0025 wt% using MWNT. To obtain the lower percolation threshold, using MWNT and epoxy, Martin et al. [12] investigated the influence of process parameters in an in situ polymerization fabrication process, such as stirring rate, resin temperature and curing temperature. It was found that the electrical properties of composites strongly depend on the choice of these parameters. Using the in situ polymerization process, the polymer/MWNT nanocomposites were prepared in [13], [14], [15], and the obtained percolation threshold was found to be lower than 1.0 wt%. Hu et al. [19] prepared the PET/MWNT nanocomposites by means of coagulation process. Uniform dispersion of MWNT throughout PET matrix was confirmed by transmission electron microscopy (TEM) and scanning electron microscopy (SEM). The percolation threshold was around 0.9 wt%.

Generally, there are two issues addressed in many previous studies: dispersion of CNT in polymer matrix and interaction between CNT and polymer. For the first issue, due to the high surface-to-mass ratio of CNT, molecular scale forces and interactions should be considered among CNT. Van der Waals forces usually promote flocculation of CNT, whilst electrostatic charges or steric effects lead to a stabilization of the dispersion through repulsive forces [6], [12]. As a consequence, by considering the nature of percolating network formed by very fine filler, e.g., CNT, the balance of the two factors of reverse effects outlined above should be taken into account. For the second issue, the fact that the nanotubes in the composites are coated or encapsulated with a thin insulating polymer layer was identified for SWNT [9] and MWNT [19]. This encapsulation acts as a barrier to the electrical charge transfer between nanotubes [9].

As mentioned above, although a lot of studies have been performed recently, whereas, except for the study [12], there is little literature covering the detailed influences of various factors in the fabrication process on the electrical properties of polymer/CNT nanocomposites using the in situ polymerization method to our knowledge. In this study, we used the in situ polymerization method to prepare polymer/MWNT nanocomposites. The reasons to choose MWNT as filler are that they are generally conducting and their dispersion is comparatively easier due to their much lower absorption energy compared with that of SWNT (around one order of magnitude lower). Furthermore, the price of MWNT is much cheaper than that of SWNT. For the case of 2 wt% of MWNT loading, the effects of various factors in the fabrication process on the electrical behaviors of nanocomposites have been studied. It was found that the bulk conductivity of the nanocomposites is significantly sensitive to the curing temperature, the mixing speed and the mixing time in the fabrication process, which ranges from the order of 10⁻⁶ S/m to the order of 1.0 S/m depending on the different process conditions used. The experimental percolation threshold was obtained as 0.1 wt%. Moreover, a statistical percolation numerical model was developed to obtain the numerical percolation thresholds for comparison. The experimental electrical conductivity of nanocomposites obeys a percolation-like power law with the identified critical exponent t as 1.75.