This study investigates the early propagation of microstructurally small fatigue cracks from micro-defects in polycrystalline Inconel 718. A 3D conformal global/local finite-element framework with adaptive remeshing is employed to explicitly resolve grain boundaries, and the energy release rate (ERR) is extracted along the crack front using the domain integral. By simulating statistically representative aggregates under linear elastic assumptions, the relative influence of microstructural features, including grain size, orientation, and topology, on the fracture driving force is quantified and ranked. Results demonstrate that the host-grain crystallographic orientation exerts a critical first-order influence: The [101] orientation increases the local ERR by +172% on average compared to the [001] orientation. This significant sensitivity affects early propagation behavior, inducing pronounced scatter in fatigue life driven by the uncontrolled local microstructural neighborhood. However, this strong microstructural forcing is shown to be confined to the first few grains. Once the crack front length exceeds the characteristic grain size by an order of magnitude, local fluctuations spatially average out, and the homogenized approximation effectively recovers the propagation kinetics of randomly oriented microstructures.

A computational framework is developed for full-scale crystal plasticity simulations of polycrystalline Ni-based superalloy turbine blades, explicitly accounting for microstructural heterogeneity. Realistic polycrystalline microstructures with prescribed grain size and orientation are embedded into complex geometries through a conformal meshing procedure, allowing the rapid generation of models with variable grain counts. The constitutive behavior is described by a dislocation-density-based crystal plasticity model incorporating geometrically necessary dislocations under finite deformations and non-isothermal loading conditions. The resulting high-fidelity nonlinear problems are solved using a grain-wise domain decomposition combined with an Adaptive Multipreconditioned FETI solver. The robustness and computational efficiency of the approach with respect to increasing microstructural resolution are demonstrated on a turbine blade geometry containing up to 40,000 grains under representative thermomechanical loading conditions. Finally, the generated simulation database is leveraged to train graph neural networks for predicting grain-averaged mechanical fields from morphological and crystallographic descriptors. The proposed methodology enables systematic investigation of microstructure-property relationships in components with realistic complexity.

Megaearthquakes (Mw > 8) cause continental-scale, long-lasting surface deformation, mainly due to viscoelastic relaxation of the asthenosphere. To investigate the links between this deformation and the slip history along subduction interfaces—including earthquakes, postseismic slip, and interseismic coupling—large 3D spherical finite-element meshes are required. This technical report introduces the various steps to build Chile_Mesh_v1.0, a customizable mesh for the Chilean subduction zone, designed as a robust platform for testing various viscoelastic rheologies. It spans ~8500 km in longitude, ~7300 km in latitude, encompassing the entire South American plate, and from the surface to 2900 km depth. Special care was taken to reproduce the complex slab geometry, especially in flat-slab regions such as the Pampean and Peruvian segments, following the Slab2 model. We show that accurately modeling both coseismic and postseismic deformation over large scales requires realistic meshed domains, extending down to the Core-Mantle boundary and thousands of kilometers from the trench. In some cases, depth-reduced meshes can be used to model viscoelastic postseismic deformation, but they fail to simultaneously capture coseismic deformation accurately. We hope this open-access mesh proves valuable for researchers studying subduction dynamics in Chile and supports the development of similar models for other regions.

The influence of a helicopter gearbox lubricant on Mode I or Mode II fatigue crack growth in 16NCD13 steel was characterized through tests performed on single-edge notched tensile samples loaded in tension-compression and on cruciform samples submitted to reversed shear plus static biaxial compression, respectively. In Mode I, the lubricant reduced the growth rate at low Delta KI and increased the threshold Delta K$$ \varDelta K $$, while in Mode II, it accelerated crack growth at low Delta KIIeffective, which was not only due to a reduction in crack face friction. The upward convective flow of lubricant carrying debris exuding from the crack, a modification in oil aspect and properties, and chemical analyses near the crack front suggest that a temperature-induced degradation of the lubricant leads to a corrosive attack of the metal, which accelerates crack growth. A tribologically transformed structure is observed along the lips of cracks grown in Mode II with normal compression in oil.

Gear and bearing failures are most often caused by rolling contact fatigue (RCF). Understanding the growth of a main surface-initiated crack into the depth of the piece, as well as the growth of subsurface-initiated branches towards the surface is necessary, in order to improve fatigue life prediction and reduce fatigue damage repair or component replacements before catastrophic failure. The purpose of this work is to analyze, based on 3D finite element simulations, the initiation and mode I growth of secondary branches in a case hardened gear, their influence on the main crack growth in mode II, and the competition between mode I and mode II crack growth. The residual stress field issued by the case hardening is taken into account and its role is analyzed. The reasons why branch cracks develop only from the upper main crack face and why only those initiated when the main crack is still shallow can reach the surface, inducing spalling are explained.

Fatigue crack growth under large-scale yielding condition is studied for high-temperature loading. The applied strains are so important that diffuse damage phenomena are visible as a network of micro-cracks in front of the major crack. The survey of a macroscopic cracked surface is nevertheless possible, and numerical simulations with explicit representation of this crack are carried out to evaluate crack driving forces. The proposed numerical scheme takes into account plastic wake in the course of crack growth in a 3D model. A non-local model of fatigue crack growth rate, based on partition of strain energy density into elastic and plastic terms, yields improved results as compared to classical assessment of ∆J by numerical methods.

Mode II fatigue crack growth under reversed shear and static biaxial compression was investigated in two bearing steels. Many aborted branches, quasi-orthogonal to the main crack, were observed along the crack face. The compressive stress parallel to the main crack hindered the growth of these branches and favored coplanar mode II crack growth. The crack face sliding displacement profiles measured by DIC were used to derive Delta KII,eff, at the main crack tip, using elastic-plastic FE simulations with crack face friction, by an inverse method. Frictioncorrected crack growth kinetics were obtained for mode II crack growth in both steels.

The long-standing problem of arbitrarily-shaped discrete dislocation loops in three-dimensional heterogeneous material structures is addressed by introducing novel singularity-free elastic field solutions as well as developing adaptive finite element computations for dislocation dynamics simulations. The first framework uses the Stroh formalism in combination with the biperiodic Fourier-transform and dual variable and position techniques to determine the finite-valued Peach–Koehler force acting on curved dislocation loops. On the other hand, the second versatile mixed-element method proposes to capture the driving forces through dissipative energy considerations with domain integrals by means of the virtual extension principle of the surfacial discontinuities. Excellent agreement between theoretical and numerical analyses is illustrated from simple circular shear dislocation loops to prismatic dislocations with complicated simply-connected contours in linear homogeneous isotropic solids and anisotropic elastic multimaterials, which also serves as improved benchmarks for dealing with more realistic boundary-value problems with evolving dislocations. Examples of sophisticated dislocation applications include the short-range core reaction between intersecting dislocation loops in interaction with a spherical cavity, as well as the Orowan dislocation-precipitate bypass mechanism in a compressed micropillar of polycrystalline copper. The latter multiscale investigation spans three orders of magnitude in size scale, and is thus enabled by computationally efficient and robust adaptive mesh generation procedures for explicit dislocation propagation, interaction, and coalescence in three-dimensional materials and material structures.

A fully coupled thermoelastic framework is formulated to cope with the free vibration response of anisotropic multilayered plates in three dimensions. The laminated structure consists of homogeneous laminae of arbitrary thickness and width under simply supported edge conditions in thermal environment. The general and exact field expressions of the temperature, heat flux, displacement and stress components are expressed in terms of double Fourier series expansions in any rectangular plate, which lead to the extended Stroh formalism with thermomechanical coupling effects in a concise and compact matrix form. Different imperfect interface conditions are introduced to characterize specific structural and thermal contact properties at the bounding interfaces, and further to determine the finite complex valued coefficients in the suitable series relations. The complete time-harmonic solutions in the laminated composites in the presence of perfect/imperfect interfaces are recursively obtained by means of the modified dual variable and position technique with explicit layer-to-layer transfer matrices. Results are obtained for different layups, length-to-thickness ratios and interfacial boundary conditions for two application examples, namely the graphite/epoxy cross-ply composites and the thermal barrier coatings on superalloys, without suffering from numerical exponential instability. These investigations reveal that the natural frequencies and first and higher vibration mode shapes of the multilayered structures can be considerably affected by increasing the environmental temperature and the severity of the interfacial imperfections. Since the through-thickness stress distribution in 2, 5, and 10 layered composites appears to be strongly correlated to the layups, such modal stress analysis could be exploited to locate the fatigue hotspots operated in dynamic structures and to guide the structural design of aircraft and spacecraft composite laminates subjected to residual vibrations.

The aim of this paper is to test the influence of different crack path models, from oversimplification currently used in literature of a straight crack front orthogonal to the specimen sides to a realistic 3D crack path. On the basis of experimental features observed for Ti17 and Ti6242 alloys under multiaxial loading, a sensitivity analysis is proposed to address the impact of realistic 3D crack path on SIF and FCGR assessment.

The dataset presents a synthesis of an experimental benchmark performed in the context of the French national research network GDR FATACRACK . A sample has been designed to produce mixed-mode crack propagation and variation of small scale yielding conditions. Two geometries and two maximum load levels are defined for the two tested materials: a stainless steel and an aluminum alloy. Around ten participants performed experiments using their usual instrumentation. Among the eight possible parameter sets, three are selected for which detailed results are presented. A satisfying overall agreement is obtained. But, some discrepancies are evidenced due either to limitations of the instrumentation or simply because from one lab to the other the applied load is not exactly the same. The results obtained by all the participants as well as the description of their testing setups and measurement tools are collected in this dataset. It includes crack path, stress intensity factor range and crack growth rate measurement for the selected configurations.

The design of reliable structures and the estimation of the residual fatigue life of industrial parts containing flaws or cracks rely on our ability to predict the propagation of fatigue cracks. Whereas in industrial component cracks might have a complex path due to geometry and loading, lab experiments used for identifying crack propagation law are often in pure mode I. The paper presents a synthesis of an experimental benchmark performed in the

Une comparaison de modelisations de l’amorcage et de la propagation de la fissuration transverse et des decohesions au sein d’une cellule elementaire representative d’un composite tisse est realisee. Celle ci utilise d’une part un critere couple (CC) en energie et en contrainte et d’autre part un modele des zones cohesives (CZM). De maniere globale les deux approches reposant sur les meme grandeurs physiques sont comparables. Cependant les contraintes imposees sur l’evolution de l’endommagement avec le CC ralentissent celui-ci par rapport au CZM qui ne les impose pas. L’observation en details des geometries de fissures obtenues avec chaque modele montre que le CZM permet une propagation moins guidee que dans le cas avec le CC et les hypotheses contraignantes.

A wide range of numerical methods have been developed, in the last decades, to efficiently simulate 3D crack propagation: X-FEM, GFEM, BEM or FEM with adaptive remeshing. These approaches have been successful in many situations, yet many issues are still encountered when dealing with complex aspects such as crack lips contact in finite strain, highly non-linear elastic-plastic material behaviors, dealing with both crack initiation and propagation stages... Recent developments have been carried out in order to address some of these issues: variational formulation of fracture, phase-field, or thick levelset, are some promising strategies to deal with the complete damage to failure problem. However, mixing both complex non-linear material behavior in finite strain and non-trivial damage model, to simulate the complete ductile failure process remains a challenging problem. In this context, the approach herein developed is focused on the capability to mix very complex highly non-linear elastic-plastic material models (for ductile or thermo-mechanical-fatigue failure) with a numerical strategy suitable to simulate a complete scenario from the finite strain plastic deformation of a structure through damage, crack initiation, propagation and up to complete failure. Concerning the material behavior part, a wide range of models can be used in the context of fatigue or critical failure. Sharing some aspects with a previously developed technique [1], we perform local remeshing in identified critical zones where the material dissipation is maximal, and a specific mesh transfer algorithm is applied to ensure further calculation. However, in order to deal with both damage and failure aspects, this new approach aims to insert, in the identified damage zones, a cohesive surface whose size and orientation will be dynamically controlled, in order to satisfy a maximal dissipation criterion during the failure process (using some algorithms previously presented in [2]). To highlight the performance and robustness of this method, various numerical assessments will be presented. In particular, the case of the Second Sandia Fracture Challenge [3] will be addressed, and these new results will be compared to the original blind predictions obtained during our participation to this round-robin.

La suite Z-set est un ensemble d'outils dedies a la mecanique des materiaux et au calcul de structures. Elle integre differents modules tels qu'un solveur elements finis, une bibliotheque d'inte-gration de loi materiaux, une collection de post-traitements dedies a l'analyse mecanique, un optimiseur permettant de determiner les parametres materiaux d'une loi a partir de donnees d'essais. Certains de ces modules sont independants et peuvent etre utilises par des codes commerciaux sous la forme de plu-gins. On presentera lors de la session logicielle des elements developpes recemment sur le calcul haute performance, la fissuration et le remaillage adaptatif, la simulation du contact, le couplage de codes et l'interoperabilite. Mots cles-code elements finis, bibliotheque materiau, duree de vie, calcul parallele, couplages.

This study aims at describing fatigue crack growth in dissimilar welding of Ti based alloys under macroscopic multiaxial loading. The proposed methodology involves the experimental analysis of fatigue crack behavior under equibiaxial tension and macroscopic combination of mode I and II for Ti17, Ti6242 and laser welded specimen of both base metals. Based on these experiments, crack path, fatigue crack growth rate and crack interaction with microstructure have been addressed. The 3D finite element analysis of cracks shapes has enabled to derive stress intensity factor (SIF) investigated for opening, in-plane and out-of-plane shear modes based on linear elastic fracture mechanics assumptions. Finally, an equivalent SIF has been proposed to take into account the local mode mixity induced by both macroscopic shear and 3D crack shape. As a conclusion, the dissimilar welding of Ti based alloys increase the fatigue crack growth rate (FCGR) for any macroscopic loading with or without shear. Moreover, the microstructure of Ti6242 alloy, is well known to inhibit FCGR by multiples local bifurcation of crack path induced by the coarse microstructure of this alloy. This point was confirmed during equibiaxial tension but anomalous and very high FCGR was observed for macroscopic mode I + II loading. For the welded material, the fatigue crack to microstructure interactions have shown that the FCGR was clearly limited by coarse a needles inducing local bifurcation and conversely that in both fusion zone and heat affected zone, local refinement of a needles could not slow down the crack propagation.