| dc.description.abstract | The study is a comprehensive dive into the fatigue behavior and related failure mechanisms of additively manufactured (AM) and commercially available wrought Ti-6Al-4V under realistic loading conditions, including mean stress, random, and multiaxial loading. Wrought titanium alloy has been broadly used in the biomedical and aerospace fields, specifically for implants and structural components. AM titanium parts aim to become a suitable alternative that may eventually be incorporated into fatigue critical applications. Additive Manufacturing (AM) technologies, specifically laser powder bed fusion (L-PBF), have made progress towards their full fledge adoption in aerospace, medical and several other applications. Although L-PBF Ti-6Al-4V has been heavily researched, there are fatigue performance aspects under realistic loadings that are not well explored. The detrimental effect and high fatigue life scatter, induced by volumetric defects generated from the AM processes and metal powder recycling, have proven challenging to address. In addition to the conventional challenges of fatigue characterization under realistic loading, AM components could encounter multiaxial stress states due to their inherently complex design; these stress states are also accompanied by residual stresses and/or fluctuating external loading. Moreover, it is known that the resultant fatigue performance can be affected by the metal powder condition used during L-PBF; therefore, it has been always preferred to use new (virgin) powder. The virgin powder use practice is not desirable due to its high cost and wastefulness. Hence, part of the research hinges on the effects of powder recycling on the fatigue performance of L-PBF Ti-6AL-4V parts. As a result, powder characteristics and mechanical performance of unmachined and machined specimens fabricated from new and heavily used Ti-6Al-4V powder are compared. For realistic loading conditions, mean stress effects are investigated under strain-controlled constant amplitude loading at different strain ratios, Rε. The generated data is used to compare several mean stress fatigue life prediction models such as Morrow, Smith-Watson-Topper, and Walker. Variable amplitude loading conditions include high-low (H-L), low-high (L-H), periodic overload (PO), and a randomly generated variable amplitude (VA) loading. Furthermore, effects of surface roughness are also investigated by comparing the fatigue performance of unmachined and machined specimens. Lastly, the effects of layer orientation on the multiaxial fatigue behavior are studied. Specimens are tested under axial, torsional, in-phase axial/torsional, and 90° out-of-phase axial-torsional cyclic loadings. Upon failure, fracture surface (of uniaxial specimens) and the crack orientation of vertical and diagonal (multiaxial) specimens is investigated to find a correlation between type of loading and the failure mechanisms. Understanding the effects of intrinsic AM properties (e.g., geometry, surface roughness, porosity, build orientation) on the fatigue behavior of AM metals under realistic loadings is one of the most important steps to facilitate the adoption of this technology in fatigue critical applications. | en_US |
