Experimental and multiscale modeling investigations of cryo-thermal cycling effects on the mechanical behaviors of carbon fiber reinforced epoxy composites

2022 
Abstract In cryotank applications, carbon fiber reinforced epoxy composites are required to face the challenge of extreme cryo-thermal cycles while rare relevant research work has been conducted. This paper reports cryo-thermal cycling effects on the mechanical behaviors of carbon fiber reinforced epoxy composites via experimental and multiscale finite element method (FEM) investigations. Four types of cryo-thermal cycling are designed to explore the roles of the cycling number and the cycling period under normal and severe conditions. Unidirectional and orthotropic laminates are prepared to demonstrate intralaminar and interlaminar expansion mismatch phenomena. For unidirectional laminates, the transverse tensile strength shows a dramatic initial decrease of 18.7%, 21.2%, 4.1% and 17.4% while the interlaminar shear strength is lowered by 4.9%, 14.2%, 5.1% and 14.0% after cryo-thermal cycling for four cycle types, respectively. Meanwhile, it is exhibited that the interlaminar shear strength of orthotropic laminates is reduced by 17.1%, 18.1%, 13.1% and 17.9%, respectively. Fourier transform infrared (FTIR) and differential scanning calorimetry (DSC) measurements are executed to detect the secondary curing. Based on the constitutive relationship affected by temperature and moisture absorption, a microscale unit cell with random fibers and a specimen-sized model with cohesive elements are established to reveal the intralaminar and interlaminar damage evolutions during cryo-thermal cycling. After elucidating typical damage modes, such as interface debonding, matrix microcracks, surface frost heave and interlaminar gap generating, the multiscale FEM model is modified to evaluate mechanical degradations quantitatively, which gives reasonable explanations for experimental results and provides an efficient and convenient method for composite design.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    42
    References
    0
    Citations
    NaN
    KQI
    []