Abstract
This study discusses the water absorption behavior of three different types of nanocomposites. Silica, carbon and alumina nanoparticles are used as fillers incorporated in epoxy resin with different contents of 0.5, 1.5, and 3 wt%. A new nonlinear least-squares regression approach in conjunction with a gradient optimization was developed using symbolic manipulator MAPLE software to find the best fit. The equilibrium moisture content of the nanocomposites is measured as a function of time and the analytical modelling is successfully conducted. Results indicate that the addition of nanoparticles generally lower moisture content in contrast to neat epoxy. Compositions of 1.5 and 3 wt% alumina as well as silica nanocomposites however show a different behaviour. Diffusion coefficients for different types of nanocomposites were computed using Fickian and Langmuir models of diffusion. The least and highest diffusion coefficients correspond to the addition of silica 0.5 wt% and alumina 3 wt%, respectively.