Abstract
In this study, analytical models and numerical finite element analysis (FEA) were used to accurately determine the interphase properties and elastic modulus of both untreated and esterified cellulose nanocrystals (CNC)/epoxy bio-nanocomposites. This was achieved by incorporating an interphase region around the CNC particles, with properties that vary between those of the nanofiller and the matrix. Experimental techniques for measuring interphase properties, such as atomic force microscopy and nanoindentation, are useful but sensitive to sample preparation and limited to certain materials, which makes modelling a valuable alternative. Random 2D and 3D representative volume elements (RVEs) were generated using Digimat-FE and imported into ANSYS to determine the behaviour of nanocomposite under load. The elastic moduli values from analytical and FEA models, which included the interphase region, fell within the range of experimental variation. The results showed that the modulus increased with CNC loading from 2.5 wt% to 5 wt% and that nanofiller treatment enhanced the modulus and increased the interphase thickness. For excessive esterification, there was a weakening effect on the composite that the analytical models failed to accurately predict. ANSYS response surface optimization was used to determine the optimal geometries and material properties for the FEA model, leading to accurate behaviour. Latin hypercube sampling (LHS) was used to explore the design space, and the optimal geometry was identified with a 6 nm interphase thickness, an aspect ratio of 10, a CNC modulus of 0.66 MPa, and an interphase modulus of 1.28 GPa. The 2D FEA RVE models struck a balance between accuracy and computational costs for treated nanoparticles.