A continuum model with a percolation threshold and tunneling-assisted interfacial conductivity for carbon nanotube-based nanocomposites

Abstract

A continuum model that possesses several desirable features of the electrical conduction process in carbon-nanotube (CNT) based nanocomposites is developed. Three basic elements are included: (i) percolation threshold, (ii) interface effects, and (iii) tunneling-assisted interfacial conductivity. We approach the first one through the selection of an effective medium theory. We approach the second one by the introduction of a diminishing layer of interface with an interfacial conductivity to build a "thinly coated" CNT. The third one is introduced through the observation that interface conductivity can be enhanced by electron tunneling which in turn can be facilitated with the formation of CNT networks. We treat this last issue in a continuum fashion by taking the network formation as a statistical process that can be represented by Cauchy's probability density function. The outcome is a simple and yet widely useful model that can simultaneously capture all these fundamental characteristics. It is demonstrated that, without considering the interface effect, the predicted conductivity would be too high, and that, without accounting for the additional contribution from the tunneling-assisted interfacial conductivity, the predicted conductivity beyond the percolation threshold would be too low. It is with the consideration of all three elements that the theory can fully account for the experimentally measured data. We further use the developed model to demonstrate that, despite the anisotropy of the intrinsic CNT conductivity, it is its axial component along the CNT direction that dominates the overall conductivity. This theory is also proved that, even with a totally insulating matrix, it is still capable of delivering non-zero conductivity beyond the percolation threshold.