Experimental and numerical investigations of crack growth of hot-rolled steel Q420C using cohesive zone model

In this paper, the application problem of cyclic cohesive zone models at different load ratios is thoroughly studied based on test data of hot-rolled steel Q420C. Firstly, the fracture toughness of Q420C steel was measured and load-crack mouth opening displacement (F-CMOD) curves were recorded. Secondly, with the help of F-CMOD curves, a comparative study of monotonic cohesive zone models was performed to calibrate model parameters. Finally, cyclic cohesive zone models with different unloading–reloading paths were used to simulate the high-cycle fatigue crack growth behaviour of Q420C steel, and their performances were compared. Research results show that plane stress assumption is more sensible when the finite element model is simplified from 3D to 2D. Rather than the conditional fracture toughness Κ_Q, the elastic–plastic fracture toughness CTOD should be used to calculate the fracture energy of the monotonic cohesive zone model. Both cyclic cohesive zone models show good robustness towards the mesh size. When the linear scaling method is used to reduce simulation time, the cyclic cohesive zone model with an unloading–reloading path passing through the origin of coordinates is the better choice. To improve the simulation accuracy of cyclic cohesive zone models at different load ratios, the load ratio must be incorporated in the damage evolution law, and a linear relationship between accumulated cohesive length δ_∑ and load ratio R was proposed for engineering applications.