Fatigue in Materials: Fundamentals, Multiscale Characterizations and Computational Modeling: Multiscale Modeling Approaches to Improve Fatigue Predictions
Sponsored by: TMS Structural Materials Division, TMS: Computational Materials Science and Engineering Committee, TMS: Integrated Computational Materials Engineering Committee, TMS: Mechanical Behavior of Materials Committee
Program Organizers: Garrett Pataky, Clemson University; Ashley Spear, University of Utah; Jean-Briac le Graverend, Texas A&M University; Antonios Kontsos, Drexel University; Brian Wisner, Ohio University

Wednesday 8:30 AM
February 26, 2020
Room: 11A
Location: San Diego Convention Ctr

Session Chair: Jean-Briac le Graverend, Texas A&M University


8:30 AM  Invited
Advances in Modeling of Fatigue Thresholds: Huseyin Sehitoglu1; 1University of Illinois
    The fatigue thresholds have been a topic of extensive research over the years mainly from the experimental viewpoint. From the theoretical side, there has been only handful of investigations. It is well known that the microstructure affects the fatigue crack growth thresholds. In many cases, the fatigue threshold is not unique and depends on the crack tip environment, ie. the presence of twins, their distance with respect to the crack tip, different grain boundary types and the local friction stress (resistance to flow) as influenced by the interfaces. We devise an atomistic-continuum model for determining the irreversibility of displacements at crack tips, hence fatigue crack growth rates, in metallic alloys. We show examples of our calculations for specific metals and explain the experimental results in the literature including our own work showing the role of microstructure on fatigue threshold behavior.

8:50 AM  
3D Discrete Dislocation Dynamic Modeling on Cyclic Behavior of Cu: Fanshi Meng1; Marc Fivel1; Emilie Ferrie1; Christophe Depres2; 1University Grenoble Alpes, CNRS, Grenoble INP, SIMaP; 2Laboratory SYMME, Université de Savoie, Annecy-Le-Vieux Cedex
    A 3D discrete dislocation dynamic simulation accompanied with a surface grain cyclically loaded is conducted in order to quantify the effect of different parameters such as the strain amplitude, grain diameter and stacking fault energy (SFE) on the formation of Persistent Slip Bands (PSBs) and extrusions/intrusions. The simulations unveil the mechanisms of formation and destruction of the PSBs with the cycles. As observed in experiments, after a few cycles, a saturation state is reached, during which the number of PSBs, dislocation densities and heights of extrusions and intrusions stay almost invariant. The number of PSBs at saturation state is demonstrated to be in a linear relation with the applied deformation and the distribution of PSBs can be explained by stress state on the cross-slip plane. Finally, the SFE effect is evidenced by comparing the simulation results performed on Cu with similar simulations performed on AISI 316L stainless steel.

9:10 AM  
Atomistic and Dislocation Dynamics Simulations of The Interaction of Dislocations with Twin Boundaries in FCC Alloys: Satish Rao1; Maxime Dupraz2; Jafaar El-Awady3; Christopher Woodward4; 1Ues Inc.; 2European Synchrotron Radiation Facility; 3Johns Hopkins University; 4Air Force Research Laboratory
    Annealing twins in polycrystalline Ni-based alloys produce stress concentrations during cyclic loading. Sub-grain digital image correlation of these boundaries under load show a factor of 4-10 local strain enhancement. A detailed understanding of the 3D dislocation-twin boundary interactions is required before dislocation dynamics simulations can be used to study this behavior. 3D atomistic simulations are used to study the interaction of screw dislocations with twin boundaries in Cu, Ni and Ni-12.5Al bipillars. The screw dislocation overcomes the low angle twin boundary at fairly low stresses by cross-slip in Cu, Ni and increases for Ni-12.5Al. Results are interpreted using a newly developed cross-slip model based on Escaig’s mechanism . Results of preliminary atomistically informed dislocation dynamics simulations on local strain enhancement in the gamma-matrix of superalloys near twin boundaries during cyclic loading is presented. These results are then used to explain fatigue crack nucleation in Ni based superalloys.

9:30 AM  
Effects of Surface Roughness on Microstructure-sensitive Computations of Fatigue Crack Formation Driving Force in Duplex Ti-6Al-4V and Al 7075-T6: Krzysztof Stopka1; David McDowell1; 1Georgia Institute of Technology
    Ensembles of duplex Ti-6Al-4V and Al 7075-T6 statistical volume elements (SVEs) undergo Crystal Plasticity Finite Element Method (CPFEM) simulations to examine the effects of a rough surface on the fatigue crack formation driving force. Surface roughness effects are incorporated into the CPFEM simulations by artificial slip intensification at surface grains, which decays toward the interior of the SVE. After elastic-plastic shakedown, mesoscale volume-averaged Fatigue Indicator Parameters (FIPs) are calculated within fatigue damage process zones of grains and fit to known Extreme Value Distributions (EVDs). SVE ensembles vary in crystallographic orientation distribution and grain shape. The surface proximity and elastic strain of the highest FIPs are examined, and multiaxial strain states are considered.

9:50 AM  
Exploring the Fundamental Role of Dislocation-twin Boundary Interactions in Fatigue: Orcun Koray Celebi1; Ahmed Sameer Khan Mohammed1; Francisco Andrade Chavez1; Jessica Anne Krogstad1; Huseyin Sehitoglu1; 1University of Illinois at Urbana Champaign
    Modeling microstructural dependence of fatigue crack growth behavior is vital in explaining variability in life predictions. In this regard, Twin Boundaries (TBs) play a decisive role in dictating fatigue resistance, particularly in the near threshold crack growth regime. Their influence is dependent on the nature of interaction between crack-tip emitted dislocations and the TB. However, this interaction is complex and can admit one of several possibilities depending on parameters such as the Schmid Factor of the twin interface and applied stress state. A model is proposed to explain these dependencies outlining the interplay of elastic and atomistic energy barriers in the process, particularly the core disregistry interactions when the dislocation approaches the twin. This is done through a combination of ab-initio DFT simulations and analytical tools of anisotropic elasticity. The critical influence of the TB is further evidenced through experimental characterization of crack growth behavior in carefully prepared twinned microstructures.

10:10 AM Break

10:30 AM  
Micromechanical Modeling of Copper under Very High Cycle Fatigue: Vahid Tari1; Michael Fitzka2; Herwig Mayer2; Jason W. Carroll1; 1Eaton Corporation Research & Technology; 2University of Natural Resource and Life Science
    Copper has received industrial attention as a fuse material due to its high conductivity. Fuses in new generation of electric vehicles experience vary random loads, therefore fuse materials need to show very good fatigue properties. Ultrasonic fatigue techniques are particularly useful for rapid qualification to examine the fuse material responses at very high frequency (20kHz). In this work, we employed a crystal plasticity framework to predict local heterogeneities in microstructure during ultrasonic fatigue tests, which are related to fatigue crack initiation. These modeling results can be used not only in prediction of hot spot regions causing fatigue damage, but also in microstructure optimization of fuse materials.

10:50 AM  
Micromechanics-based Effect of Defects Models for Ellipsoidal Anomalies: James Sobotka1; R. Craig McClung1; Michael Enright1; 1Southwest Research Institute
    This presentation introduces a novel methodology to characterize the effect of defects on fatigue lives. We approximate defects ellipsoidal anomalies in a homogenous, linear-elastic continuum loaded by arbitrarily far-field stresses. The geometric features of the anomaly concentrate stresses, and these stresses may be determined using Eshelby approaches. Furthermore, we combine these stresses with Taylor’s theory of critical distances to incorporate a material-dependent length scale into fatigue predictions. This presentation summarizes predictions from this framework, including exact analytical expressions for stresses near spherical pores, failure envelops analogous to Kitagawa diagrams, and fatigue-life reductions due to ellipsoidal anomalies. These results show the effect of stress state, pore size, and pore shape on fatigue life predictions. At this time, these predictions are limited to empty pores, but the framework presented here is easily extensible to anomalies due to stiffness mismatch and pores with entrapped gas as in additively manufactured components.

11:10 AM  
The Dislocation Configurational Energy for the Prediction of Fatigue Crack Nucleation: An Integrated Experimental and Computational Study: Nikoletta Prastiti1; Fionn Dunne1; Daniel Balint1; 1Imperial College London
     A new quantity termed as the dislocation configurational energy density was recently introduced and described as the elastically-stored energy density associated with the interaction of dislocations and their structures. This quantity has shown excellent correlation with the stored energy quantity which has been used extensively, to correctly identify the experimentally observed sites of fatigue crack nucleation in a series of engineering alloys. An integrated experimental, crystal plasticity and discrete dislocation plasticity study is presented to elucidate the mechanistic basis of fatigue crack nucleation in single crystal nickel-based superalloys. The role of geometrically necessary dislocations in fatigue crack nucleation at the micron scale is also discussed, along with other fatigue indicator parameters.

11:30 AM  
Quantitative Characterization and Multi-scale Modeling of the Effects of Porosity on Fatigue Life in Ni-based Single Crystal Superalloys: Keli Liu1; Junsheng Wang1; 1Beijing Institute of Technology
    Understanding the effects of porosity in Ni-based single crystal superalloys on fatigue damage is critical for predicting the lifetime of the aero-plane engine. In this research, the geometrical information of pores in as-cast and homogenized samples with different process conditions are obtained by X-ray tomography. Based on the reconstructed pore models as a function of casting conditions, simulations of the fatigue behavior is implemented as a subroutine in a fatigue life model which coupling of crystal plasticity and phase field fracture methods (the free and open source software DAMASK). Combining the experimental observation and numerical simulation results, the influences of the pore size, morphology and porosity connectivity on crack initiation and propagation behavior fatigue life can be quantitatively simulated. Therefore the fatigue damage mechanism of the pores is expected to be explained.