Prediction of temperature-dependent transverse strength of carbon fiber reinforced polymer composites by a modified cohesive zone model

A temperature-dependent micromechanical model was proposed in this study by the representative volume element model with the fibers randomly distributed, the law of temperature-dependent Young's modulus of polymer and the plasticity model of polymer. Considering the thermal effect, a modified cohesive zone model with temperature-dependent parameters was presented for simulating the fiber/polymer interface. Validations by other's experimental results indicate that the proposed model is reliable and accurate to estimate the transverse mechanical properties, including Young's modulus and strength of carbon fiber reinforced polymer composites. Compared with those at room temperature, values of the transverse tensile strength and transverse compressive strength at 80 degrees C are reduced by 9.1% and 4.8%, while both values at-196 degrees C greatly increase by 16.1% and 8.8%, respectively, showing that the temperature plays a more influencing role on the transverse tensile strength. Failure behaviors of composites were investigated for the temperature from-196 degrees C to 80 degrees C. Moreover, parametric sensitivity analysis was carried out to evaluate the effect of interface properties on the strengths of carbon fiber reinforced polymer composites. It was found that the interface strength significantly affects mechanical behavior of composites at different temperatures, while the influences of interface stiffness and fracture energy are fairly minimal.

Automated summary

A temperature-dependent micromechanical model was proposed to estimate the transverse mechanical properties of carbon fiber reinforced polymer composites. The model considered the random distribution of fibers, the temperature-dependent Young's modulus of the polymer, and the plasticity model of the polymer. A modified cohesive zone model with temperature-dependent parameters was presented for simulating the fiber/polymer interface. The model was validated by experimental results and found to be reliable and accurate.