High-cycle fatigue analysis of laser-based directed energy deposition maraging steels: Combined phase field and experimental studies

This work presents an efficient phase field fracture methodology for modeling high-cycle fatigue crack growth in laser-based Directed Energy Deposition (LDED) produced components. The model incorporates a fatigue history variable that accounts for fatigue crack propagation with sensitivity to the mean load variations by introducing an accumulative energy term. Two distinct available degradation functions are implemented to investigate their influence on fatigue crack evolution. An efficient solution strategy combining a hybrid Newton–Raphson method with an adaptive switching between full and delayed stiffness updates and envelope loading scheme is developed to handle the computational challenges of high-cycle fatigue simulations. The model’s performance regarding fatigue crack evolution and computational efficiency is evaluated through representative numerical benchmarks and sensitivity analyses. Further validation is conducted by comparing predicted fatigue crack growth rates against experimental data obtained from 18Ni300 compact tension specimens. Experimental measurements reveal that specimens with cracks propagating through build layers exhibit higher fatigue resistance than those with cracks propagating along layer interfaces. The integrated experimental-numerical methodology incorporates specimens manufactured via both conventional and LDED processes. Through adjusting of fatigue-related model parameters, the framework efficiently reproduces fatigue crack growth rate curves for AM-fabricated CT samples across different build orientations, demonstrating computational efficiency and accuracy in high-cycle fatigue analysis. • Phase field model for high-cycle fatigue captures both crack initiation and propagation. • The model features a fatigue history variable that accounts for mean load effects. • An adaptive scheme reduces the computational cost for high-cycle fatigue simulations. • The model accurately predicts FCGR in DED-produced maraging steel specimens.