Simulation of surface asperities on a carbon fiber using molecular dynamics and fourier series decomposition to predict interfacial shear strength in polymer matrix composites

Abstract

ABSTRACT The objective of this paper is to predict the fiber/matrix interfacial debond strength in composites. Atomic force microscopy (AFM) images of the surface topography of a de-sized carbon fiber reveal that there are surface asperities present at various length scales ranging from a nanometer to several microns. These asperities are likely caused by shrinkage of the polyacrylonitrile (PAN) precursor during the graphitization process. In order to bridge the length scales, a Fourier series-decomposition covering a range of asperity wavelengths and amplitudes is necessary to effectively capture the roughness of the fiber surface at different length scales. Further, once a surface asperity profile has been resolved into individual subcomponents using Fourier-decomposition, MD simulations can then be employed to obtain the interfacial shear strength of the subcomponent asperity of a given amplitude and wavelength. Finally, by recombining the peak interfacial shear force obtained from each of these subcomponents into the overall shear force for the fiber surface profile, the length-scale-averaged shear strength can be obtained for any given asperity. The objective of this paper is to use this novel approach to determine the interfacial shear strength of de-sized carbon fiber embedded in an epoxy matrix and compare predicted results with experimental data.