PSI - Issue 68

Soran Hassanifard et al. / Procedia Structural Integrity 68 (2025) 77–83 S. Hassanifard and K. Behdinan / Structural Integrity Procedia 00 (2025) 000–000

78

2

and porosity still persist and must be addressed during the design process (Khosravani et al., 2020). These factors can significantly affect the mechanical performance of AM parts, particularly under cyclic loading, which may lead to catastrophic fatigue failure (Liu et al., 2024). Predicting the fatigue life of AM components under various cyclic loading conditions remains a significant challenge for researchers. This complexity arises because numerous processing parameters can influence fatigue performance, and the problem is further complicated when these parameters are interdependent, affecting the part's overall quality and mechanical behavior (Shipley et al., 2018; Bakır et al., 2021). To address this, researchers have developed a variety of techniques aimed at accurately predicting the fatigue life of AM parts. However, most of these models focus on metallic AM components fabricated using techniques like selective laser melting (Li et al., 2024) and powder bed fusion (Foti et al., 2023; Sanaei and Fatemi, 2021; Kishore et al., 2023). Several approaches have garnered attention in this field, including data-driven machine learning methods (Zhan and Li, 2021; Hassanifard and Behdinan, 2023; Salvati et al., 2022), fracture mechanics-based models (Kishore et al., 2023), as well as stress, strain, and energy based fatigue damage models (Hu et al., 2024). Among these, critical plane approaches have been particularly prominent (Foti et al., 2023). Polymeric materials and polymer-based composites, on the other hand, exhibit different behaviors compared to metallic materials, as the processing parameters for these materials vary significantly. One of the most commonly used AM techniques for fabricating thermoplastics or polymer-based composites is fused filament fabrication (FFF). In this process, a raw filament is fed through a heated nozzle, where the material is extruded and deposited in a raster pattern onto a heated bed to build the desired part layer by layer (Rajan et al., 2022). During the FFF process, irregularities and discontinuities are likely to form between filaments due to factors such as the nature of filament deposition, gravitational effects, and material properties like viscosity and melt flow index (Acierno and Patti, 2023; Petersmann et al., 2020). These discontinuities are often visible on the outer surfaces of the printed part, and internal defects, such as voids and porosity, can form within the component. These internal and surface imperfections significantly reduce the mechanical properties and fatigue life of the printed part compared to that of the neat filament material (Hassanifard and Behdinan, 2024). As a result, 3D-printed components often behave similarly to notched parts, where stress concentrations are higher, compared to smooth, defect-free components typically produced through conventional methods such as injection molding. This behavior is a key factor in the reduced fatigue performance of FFF-printed parts. It has been proven that material degradation, particularly stiffness reduction, occurs in materials subjected to cyclic loads. This reduction, specifically in Young’s modulus, significantly impacts fatigue life predictions, as the material’s ability to bear applied loads diminishes over time, accelerating failure mechanisms. Therefore, it is essential to account for material degradation when predicting fatigue life. This study focuses on predicting the fatigue life of 3D-printed graphene nanoplatelet (GNP)/acrylonitrile butadiene styrene (ABS) composite parts by considering (i) varying GNP content, (ii) raster angle, and (iii) material degradation during cyclic loading. A computational model was developed in which 3D-printed samples are treated as homogeneous, defect-free parts with an imaginary notch characterized by a notch strength reduction factor. This factor was derived from experimental fatigue data for samples printed at 0º and 90º raster orientations, as well as for neat filament material. Interestingly, the notch strength reduction factor was found to be largely independent of GNP content but sensitive to the applied load. Using this relationship, a derived formula can be applied to other raster orientations and GNP contents with high accuracy. The complete loading-unloading cycles were modeled using Neuber and Ramberg-Osgood relations to determine the necessary mean stress and strain amplitude values. Fatigue life was then predicted using modified Morrow and Smith-Watson-Topper (SWT) equations and compared to experimental fatigue test data, showing reasonable agreement. 2. Methodology 2.1. Material and specimen In this study, 3D-printed ABS/GNP composite samples with varying GNP contents of 0.1, 0.5, and 1.0 wt.% GNPs at different raster orientations were selected for fatigue life predictions. Comprehensive details regarding the static and fatigue strength data, as well as the strain-based fatigue parameters of these composites, can be found in reference (Hassanifard and Behdinan, 2024).