Interactions between HA/GO/epoxy resin nanocomposites: optimization, modeling and mechanical performance using central composite design and genetic algorithm

Abstract

This study is focused on the role of hydroxyapatite (HA) nanoparticles and graphene oxide (GO) nanoplates on the flexural and compression properties of epoxy-based hybrid nanocomposites. In the first step, epoxy-based hybrid nanocomposites were reinforced by different HA nanoparticles and GO nanoplates up to 7 wt% and 0.5 wt%, respectively. Filler’s weight fractions that used as design parameters have been achieved by central composite design method in Minitab software. The experimental results showed different combinations of HA- and GO-enhanced mentioned mechanical properties in various states. In the second step, a statistical modeling has been done by response surface method (RSM), artificial neural network (ANN) and decision tree methods. Modeling results showed that ANN and decision tree methods have the best fitness. Finally, the mechanical properties are optimized by genetic algorithm. The optimum values are 21.54 MPa for flexural strength in 5.17 wt% HA and 0.38 wt% Go and 25.7 GPa for flexural modulus in 2.73 wt% HA and 0.195 wt% GO. Also, the optimum results for compression strength are 23.95 MPa in 7 wt% HA and 0.289 wt% GO, and also 690.5 MPa for compression modulus in 6.89 wt% HA and 0.007 wt% GO. Effective mechanisms of fillers have been analyzed by SEM and observed that debonding, crack path deflection, plastic void growth and pullout were dominant.