Experimental characterization and numerical prediction for mechanical aging of epoxy and its carbon fiber-reinforced composite under hydrothermal conditions

The Aqueous environment impairs the mechanical performance of carbon fiber-reinforced polymers (CFRPs) and decreases the durability of engineering structures made of CFRPs primarily by degrading the resin matrix. Hence, the objective of this study is to investigate and accurately predict the influence of hydrothermal aging on the epoxy and its carbon fiber reinforced polymer (CFRP). Particularly, this work developed a modeling method that combined Fick's model, empirical aging model and elastoplastic constitutive model to predict the mechanical behavior of hydrothermally aged epoxy matrix. After the experimental characterization, this epoxy modeling method demonstrated its predictive capability on the mechanical response of the deteriorated epoxy under different hydro-thermal-salt conditions, with the mean absolute percent error (MAPE) of approximately 4.61%. Furthermore, the transverse compressive behavior of unidirectional (UD) CFRPs could be predicted by using the representative volume element (RVE) model with the degraded matrix behavior predicted via this epoxy modeling method. Integrated with this CFRP modeling, results from mechanical tests and fracture analysis further demonstrated that the hydrothermal aging effect on the transverse compressive performance of the UD CFRPs was dominated by the deterioration of their resin matrix. Therefore, this study demonstrates the feasibility of the finite element analysis (FEA) for mechanical aging of the epoxy matrix and its CFRP under hydrothermal conditions via employing the proposed integrated modeling method. Highlights: Degradation of epoxy polymer under hydro-thermal-salt aging conditions. Integrated model to predict mechanical response of hydrothermally aged epoxy. RVE model to predict the mechanical behavior of hydrothermally aged CFRP. Hydrothermal aging on the transverse compressive behavior of UD CFRP.