Quantification of gas permeability of epoxy resin composites with graphene nanoplatelets

Abstract

This paper presents the development and validation of a numerical simulation method using the Lattice Boltzmann Method (LBM) to predict the permeability of epoxy resin (ER) composites with graphene nanoplatelets (GNPs).Gas permeability tests were carried out for a series of GNP/ER nanocomposites with different loadings and diameters of GNPs. The experimental results confirm that inclusion of GNPs in ER significantly decreased the effective gas permeability, with the highest reduction of 66% when the GNP loading was 3 wt%. The effects of using different diameters of GNPs show that using GNPs of 25 μm in diameter achieved less reduction in gas permeability than using GNPs of smaller diameters of 5 and 15 μm at the same loading of 1 wt%. This unexpected result has now been explained by the developed numerical model.The microstructures of GNPs filled ER composites were numerically reconstructed for the relative gas permeability prediction model using LBM. The 3D X-ray CT scan images clearly show agglomeration of GNPs, in particular when the diameter of GNPs is large (25 μm), due to strong van der Waals forces. An agglomeration sub-model was thus incorporated when numerically constructing the microstructure of GNPs filled ER composites. Agglomeration of GNPs results in the formation of a small number of super-thick GNPs, leaving large spaces as ER-rich area without any GNP. This led the GNPs filled ER with 25 μm of GNP diameter to obtain a lower reduction in gas permeability than smaller GNPs filled ER.The results of numerical sensitivity studies on surface area, rotation, curling and folding of GNP flakes suggest that it is acceptable to use flat disk shaped flakes to represent amorphous GNPs with small degrees of deformation (less than 20° and 1.5 for folding angle and curling rate respectively). The results also show that the projection area perpendicular to the overall gas flow direction dominates the overall gas barrier effect of GNPs. The feasibility of using 2D models is demonstrated and it is acceptable to assume that the GNPs in the prepared samples are uniformly sized with a diameter equal to the nominal diameter.This numerical simulation model significantly improves the accuracy for prediction of reduction in gas permeability, over that of existing analytical models, when compared against the authors’ experimental results and experimental data from literature.