Resumen de Fatigue growth of microstructurally short cracks in ni-based superalloys
INCONEL 718 (IN718) is a unique Ni-based superalloy that was developed many decades ago, but has continued to fulfill critical requirements for current and emerging products, especially for jet turbine engine components. This material has a balance of attributes that have made it one of the most utilized superalloys for high temperature applications. Such combination of outstanding properties is mainly due to the precipitation of γ’ and γ” strengthening nanoparticles. In addition, its microstructure includes the presence of carbides and δ phase particles.
After optimization of its strength, creep and low cycle fatigue (LCF) behavior during the last few years, high cycle fatigue (HCF) remains the main cause of failure of IN718 components. In fact, despite being operative since the 1960s, full fundamental understanding of the HCF behavior remains a major challenge. HCF is mainly governed by stage-I fatigue, which is determined by the incubation time required for the nucleation of the cracks and their propagation at the microstructural scale, until they become long cracks whose propagation is well understood through Paris type propagation laws. The transition from plastic strain accumulation to damage nucleation depends on cyclic softening and hardening effects that are controlled by the way dislocations interact with precipitates and grain boundaries, but these mechanisms are not completely understood in IN718 yet. In addition, microstructural heterogeneities, such as grain boundaries or carbides, play a major role in determining the sites where the first damage nucleates, but a systematic analysis and quantification of their effect is still missing in the literature. Finally, short crack propagation is thought to be governed by the interaction of the crack with microstructural obstacles, such as grain boundaries. However, systematic studies of crack transmission across grain boundaries are still lacking. In this context, the main objective of this investigation is the development and application of novel experimental techniques aimed at increasing our fundamental knowledge on fatigue crack initiation and propagation in Ni-based superalloys.
In order to achieve this goal, an experimental multiscale strategy was designed and applied to forged IN718 specimens subjected to HCF. First, novel nanomechanical xiv testing methodologies were developed based on cantilever bending and tensile testing of micrometer size specimens. The objective of these micromechanical tests was to study the transition from plastic strain accumulation to damage nucleation at the scale of individual grains. The specimens were milled out of individual grains from polycrystalline samples by focused ion beam (FIB) milling. The approach was aimed at analyzing cyclic softening effects, without the influence of cyclic hardening derived from dislocation pile-ups at grain boundaries. Subsequently, the interaction of dislocations with the hardening precipitates under cycling loading was analyzed by transmission electron microscopy (TEM). The results of the novel methodology were promising as they showed clear evidence of damage nucleation at preexisting slip traces and precipitate shearing effects by the gliding dislocations upon cyclic loading. However, no evidence of cyclic softening induced by the precipitate shearing was observed, because it was hampered by the hardening induced by the rapid increase in dislocation density associated with the small size of the specimens. In this condition, the crack propagation driving force was the cross-sectional area reduction associated with the accumulated slip. Future developments to assess cyclic softening effects should ensure the applications of fully reversible plastic cycles, i.e., stress ratios close to R = -1.
Moreover, a methodology to implement high resolution digital image correlation was applied on the surface of macroscopic specimens subjected to interrupted HCF tests, with the objective of studying the accumulation of plastic strain at local microstructural heterogeneities. The methodology is based on the application of a nanometer scale pattern on the surface of specimen based on a process of gold remodeling and the use of digital image correlation to determine full-field deformation maps from scanning electron microscopy (SEM) images taken from the same specimen at different numbers of loading cycles. The technique achieved a resolution below 1 µm, but better resolutions are possible optimizing the remodeling process.
The results showed that plastic strain was mainly accumulated at three specific microstructural features: twin boundaries with a high elastic incompatibility (a very different Young Modulus in the loading direction) of the adjacent grains; carbide particles, which are hard and brittle phases that act as stress concentrators; and large grains, which according to the Hall-Petch effect allow less constrained dislocation motion. However, the significance of the results obtained was limited because the technique lacks information on the sub-surface microstructure features that can lead to damage nucleation.
Finally, and to overcome this limitation, synchrotron assisted fatigue experiments, which included full volume phase contrast tomography (PCT) and diffraction contrast tomography (DCT) were carried out in miniaturized specimens of IN718, with the objective of studying crack nucleation and the interaction of microstructurally short cracks (MSC) with grain boundaries. These experiments corroborated that the main microstructural features that lead to fatigue crack initiation are large surface xv twin boundaries displaying a large elastic incompatibility between adjacent grains.
Moreover, it was found that the propagation of MSC occurred parallel to individual slip planes in each grain and that their transmission was controlled by easy dislocation slip transfer paths at each grain boundary.
All in all, the techniques developed and applied to forged IN718 in this work contributed to get a better understanding of the role of microstructure on the nucleation and propagation of fatigue cracks, which is a valuable asset to improve the microstructure and/or develop more accurate predictive models of fatigue behavior.
