Effect of Imperfections on Elastic Stiffness of Polymers Reinforced with Long Aligned Single-Walled Carbon Nanotubes

Abstract

Theoretically, the polymers reinforced with long aligned single-walled carbon nanotubes (LASWCNTs) must have one order of magnitude larger stiffness and strength than the classical carbon fiber reinforced polymers. However, imperfections such as vacancy defects in the single-walled carbon nanotubes (SWCNTs), undulation, and clustering of SWCNTs in the polymer matrix are known to adversely affect the otherwise superior properties of LASWCNT-reinforced polymer composites. The determination of unbiased relative importance of various forms of imperfections is important to determine the most efficient strategies to produce polymers reinforced with LASWCNTs. An investigation of stochastic effects of imperfections in the form of vacancy defects in the nanotubes, undulation of nanotubes, and clustering of nanotubes on the initial elastic stiffness of LASWCNT-reinforced polymer composites is presented. To this end, first the effect of vacancy defects on axial elastic stiffness of imperfect LASWCNTS is determined by using a structural analysis of LASWCNTs with various randomly distributed defect ratios. Next, the effect of undulation in 2D of the SWCNTs on the axial stiffness of LASWCNT-reinforced polymer composites is determined. It is shown that, if the only random variables considered are axial stiffness of LASWCNTs with vacancy defects and their undulation amplitude, then the undulation amplitude turns out to be much more important than the axial stiffness of LASWCNTs with vacancy defects. A parallel coupling of LASWCNTs with nanostructured polymer matrix is shown to grossly overestimate the axial stiffness of the LASWCNT-reinforced polymer composite even in the presence of experimentally observed undulation. To fit the experimental results, a simple parallel coupling model that takes into account the clustering of the nanotubes is proposed. The relative importance of all random variables involved in the model is estimated. In general, the most important variable turns out to be the effective diameter of the bundle of LASWCNTs, and the next most important parameter is the undulation amplitude of the LASWCNTs.