Search

The microstructural origins of highly localized elastic strain concentrations in polycrystalline microstructures under monotonic loading are studied using grain-scale, in situ digital image correlation and crystal plasticity finite element method. It is shown that the locations of exceptionally high elastic strain concentrations in the microstructure depend on particular crystallographic and morphological orientations of grains and less so on crystalline details of their local neighborhood. Based on these results, we discuss how topological and crystallographic features of annealing twin boundaries can increase the likelihood of slip band initiation throughout the microstructure of polycrystalline Ni-base superalloys.

International audience; Fatigue crack nucleation in crystalline materials typically develops due to highly localized cyclic slip. During a fatigue cycle, reverse slip differs locally from slip in the forward direction particularly in precipitate-containing materials such as superalloys. In this paper we report the first direct measurements of irreversibility at the scale of individual slip bands by high-resolution digital image correlation (DIC) in a polycrystalline nickel-based superalloy. Quantitative measurements of the slip irreversibility are challenging for regions of material that have a size that captures the microstructure and its variability. High spatial resolution at the nanometer scale during experimental measurements is needed to observe slip localization during deformation. Moreover, large fields are also needed to obtain the material response over statistically representative populations of microstructural configurations. Recently, high resolution scanning electron microscope (SEM) digital image correlation (DIC) has been extended for quantitative analysis of discontinuities induced by slip events using the Heaviside-DIC method. This novel method provides quantitative measurements of slip localization at the specimen surface. In this paper, the Heaviside-DIC method is used to measure slip irreversibility and plastic strain accumulation in a nickel-based superalloy. The method detects bands with high levels of irreversibility early in cycling that ultimately form fatigue cracks upon further cycling. The local microstructural configurations that induce large amounts of plasticity and slip irreversibility are correlated to crack nucleation locations.

Micro-tensile specimens of nickel-based superalloy oligocrystals were tested in-situ in an scanning electron microscope in transmission mode (TSEM) enabling observation of dislocations. The dynamics of dislocation motion during tensile loading were captured and correlated with the measured intermittencies during plastic flow recorded by high load- and temporal-resolution sensors. This investigation in particular focused on the dislocation behavior near twin boundaries with different slip configurations. A multiplicity of deformation mechanisms at the dislocation scale were observed within individual slip bands, including precipitate shearing, dislocation decorrelation and antiphase boundary-coupled shearing. These processes affect strain localization near twin boundaries and provide new defect-level insights on plastic localization and fatigue crack initiation in these alloys.

This paper develops a Bayesian, probabilistic crack nucleation model in the Ni-based superalloy Ren\'e 88DT for fatigue loading. A data-driven, machine learning approach is developed to identify the underlying mechanics driving crack nucleation. An experimental set of fatigue-loaded microstructures is characterized near crack nucleation sites using scanning electron microscopy (SEM) and electron backscatter diffraction (EBSD) images to correlate grain morphology and crystallography to the spatial location of crack nucleation sites. A concurrent multiscale model that embeds polycrystalline microstructures, created from the EBSD images, in a self-consistent homogenized material is developed for low cycle fatigue simulations needed to create a database of state variables. The polycrystalline domain is modeled by a crystal plasticity finite element model (CPFEM), while a homogenized anisotropic plasticity model is used for the exterior domain. A Bayesian classification method is introduced to optimally select the most informative state variable predictors of crack nucleation and constructs a near-Pareto frontier of models with varying complexity. From this principal set of state variables, a simplified scalar crack nucleation indicator is formulated which encompasses all of the relevant components derived from the main discriminators. This Bayesian approach allows the micromechanical state variables responsible for causing crack nucleation events to come out naturally from existing data. The final result is a model that predicts the probability of nucleating a crack at a microstructural location, given the mechanical state of the material.

The advancement of materials science at the mesoscale requires improvements in both sampling volumes/areas and spatial resolution in order to make statistically significant measurements of microstructures that influence higher-order material properties, such as fatigue and fracture. Therefore, SEM-based techniques have become desirable due to improvements in imaging resolution, large sample handling capability, and flexibility for in-situ instrumentation. By using fast sampling of SEM electron detector signals, intrinsic beam scanning defects have been identified that are related to the response time of the SEM electron beam deflectors and electron detectors. Mitigation of these beam scanning defects using detector sampling approaches and an adaptive model for settling time is shown to produce higher resolution SEM images, at faster image acquisition times, with a means to quantify the different response functions for various beam deflectors and detectors including those for electrons and ions.

Quantification of mesoscale microstructures of polycrystalline materials is important for a range of practical tasks of materials design and development. The current protocols of quantifying grain size and morphology often rely on microstructure metrics (e.g., mean grain diameter) that overlook important details of the mesostructure. In this work, we present a quantification framework based on directionally resolved chord length distribution and principal component analysis as a means of extracting additional information from 2-D microstructural maps. Towards this end, we first present in detail a method for calculating chord length distribution based on boundary segments available in modern digital datasets (e.g., from microscopy post-processing) and their low-rank representations by principal component analysis. The utility of the proposed framework for capturing grain size, morphology, and their anisotropy for efficient visualization, representation, and specification of polycrystalline microstructures is then demonstrated in case studies on datasets from synthetic generation, experiments (on Ni-base superalloys), and simulations (on steel during recrystallization).

The microstructural configurations that favor early strain localization and fatigue crack initiation at intermediate and high temperature (400 °C–650 °C) have been investigated using novel experimental techniques, including high resolution digital image correlation and transmission scanning electron microscopy. Cyclic fatigue experiments in the high and low cycle fatigue regimes have been performed on a Rene 88DT polycrystalline nickel-base superalloy at temperatures up to 650 °C and compared to previous fatigue results obtained from tests in the very high cycle fatigue regime. Competing failure modes are observed along with an inversion in the temperature fatigue life dependence of fatigue strength from the low to high cycle fatigue regime. Oxidation-assisted processes are dominant at high applied stresses while cyclic plastic localization and accumulation govern fracture at low applied stresses. In addition, a second competing mode exists in the high and very high cycle fatigue regime from non-metallic inclusions as compared to internal intrinsic initiation sites. The grain-scale features that exhibit strain localization and crack initiation were investigated in detail. Transmission electron microscopy (TEM), transmission scanning electron microscopy (TSEM) and electron channeling contrast imaging have been conducted on samples removed from targeted regions with microstructural configurations that favor crack initiation to characterize the associated dislocation sub-structure and its evolution with temperature. Plasticity is observed to be less localized during cyclic loading at high temperature compared to room temperature. The microstructural features that drive initiation across the temperature range investigated are: twin-parent grains pairs that are at the upper end of the size distribution, are oriented for near maximum elastic modulus mismatch, and have high stresses along planes parallel to the twin boundaries.

In this work, we investigate the relationship between an intense slip band (ISB) and the zone of large lattice rotations that forms ahead of the tip of the ISB. We develop a crystal plasticity finite element model of a discrete ISB lying within an oligocrystalline assembly and calculate the local crystalline stress and lattice rotation fields generated by the ISB. The calculations demonstrate that, first, a region of severe lattice rotations, commonly referred to as a microvolume, does not form without the ISB, and second, large amounts of accumulated slip in the ISB are required to enlarge the microvolume to sizes and rotation magnitudes observed experimentally. Ahead of the ISB tip, the quintessential plastic zone always forms, but the atypical microvolume forms when non-concentrated and spatially diffuse slip is activated by the ISB-induced stress field. This result suggests that the detrimental ISB/microvolume pair will likely appear in pairs of crystals in which transmission of the slip from the ISB is severely blocked by the grain boundary, a hypothesis that we verify with a few target cases.

Fatigue crack nucleation at annealing twin boundaries (TBs) within polycrystal nickel-based superalloy Rene 88 D T is investigated with a microstructure-sensitive crystal plasticity (CP) model, digital image correlation strain measurements and experimental SEM crack nucleation observations. Strong slip localizations at TBs were experimentally observed and predicted by the CP model, which also showed high predicted geometrically necessary dislocation and corresponding stored energy densities, capturing experimental observations of crack nucleation. In a systematic study, elastic anisotropy was found to drive local elastic constraint and hence resolved shear stress, slip activation, GND density and stored energy density, demonstrating for this reason that TBs are preferential sites for crack nucleation in this alloy. The parent grain / twin pair crystallographic orientation with respect to remote loading was also demonstrated to be key to slip activation parallel to TBs and hence to stored energy density and fatigue crack nucleation, and the range of most damaging parent grain orientations has been identified.

The process of crack initiation has been investigated in three widely used titanium alloys with different microstructures and loading conditions. Using low-cycle fatigue tests, a unique crack nucleation mechanism involving strain localization at ( 0001 ) twist boundaries has been identified. In order to constitute a potential crack initiation site, the twist boundary must experience a high resolved shear stress and a high normal stress. Crack initiation at these boundaries is most frequently associated with twist angles spanning the 10° - 20° range. Deformation prior to crack initiation at these rare microstructural configurations has been characterized using transmission electron microscopy and high-resolution digital image correlation across large fields of view. The ( 0001 ) twist boundaries are preferential locations for early and intense strain localization. Prior to crack nucleation, deformation proceeds via shear along such boundaries where no β layer at the interface was evidenced. The presently discussed crack formation mechanism is believed to be of broad relevance as it is not significantly influenced by microstructural parameters such as the α grain size, the degree of microtexture, the β phase fraction or the surrounding microstructure as well as α and β compositions.

International audience; The slip localization behavior of the polycrystalline nickel base superalloy Inconel 718 during monotonic tensile loading at room temperature, is investigated for the first time in relation to the 3D microstructure. Multi-modal data merging tools are used to recombine high resolution digital image correlation (HR-DIC) data with 3D electron back-scatter diffraction tomography (3D EBSD), over a wide region of interest. This procedure enables reconstruction of the slip band planes in the 3D microstructure. Statistical analyses conducted over 500 individual slip bands reveal strong correlations between their location and specific microstructure configurations. In particular, over half of the slip bands emanate from triple junction lines (3D lines defined by the junction of three crystals). Moreover, the most intense and longest slip bands, which would become critical fatigue crack nucleation sites during cyclic loading, are located close and parallel to particular annealing twin boundaries and are simultaneously connected to triple junction lines. Crystal plasticity finite elements calculations are performed on the experimental microstructure to identify the slip activity that results in the formation of high intensity slip bands (localized plasticity) or zones of high lattice rotation (non-localized plasticity) in these particular microstructure regions.

A computational study of 3D virtual instantiations of microtextured Ti-6Al-4V with varying initial slip system strengths is presented. Electron backscatter diffraction (EBSD) scans of a rolled and heat-treated mill annealed plate were used in order to determine the approximate geometric morphology of both the grain structure and the microtextured regions. Data from the EBSD experiments were used to calculate representative orientation distribution functions (ODFs) and grain size distributions for the α (HCP) crystallographic phase. Laguerre tessellations were employed to create idealized geometric representations of the microstructure and microtextured regions, while orientations were sampled from the experimentally derived ODFs. A highly parallelized crystal plasticity finite element framework was used to model the deformation response of single phase polycrystals under uniaxial tension, with attention paid to the intragrain slip system activity. Simulations were conducted with changes in the orientations within microtextured regions, as well as with various sets of initial slip system strengths to reflect differences in reported values in literature. Results were compared to a strength-to-stiffness parameter designed to predict succession of yield as a function of orientation. Presented are slip activity trends as a function of microstructure and initial slip system strengths, as well as results concerning the development of long-range localization of plasticity as a function of the microstructure. Predictions are compared to slip system activity measured by scanning electron microscope based digital image correlation.

Fatigue crack initiation at high temperatures occurs at microscopic fatigue shear bands that form near twin boundaries in polycrystalline nickel-base superalloys that contain minimal metallurgical defects. The associated dislocation sub-structure is complex and dependent on thermal and mechanical conditions. Here we show the formation of fine-scale carbide precipitates along these fatigue shear bands in a nickel-base superalloy during fatigue above 650 °C. The formation of carbide precipitates is studied using calculations of bulk and pipe diffusion. The contribution of pipe diffusion is observed to be critical in the formation of fine-scale carbides and occurs coincident with the delocalization of slip during fatigue.

The equiaxed microstructure of Ti-6Al-4V contains nominally 10µm to 15µm α -phase grains. Depending on processing, these fine grains may aggregate into large, millimeter-scale regions of similar crystallographic orientation. These so-called microtextured regions are detrimental to quasi-static and fatigue properties. Their presence may facilitate the formation of long-range strain localization—bands of plastic strain that traverse grain boundaries and terminate at the boundaries of the microtextured region which compromises strength and ductility—and can cause early fatigue crack nucleation. Furthermore, the low angle boundaries within microtextured regions offer little resistance to crack growth, and corresponding increases in crack growth rates have been observed in these regions. Despite significant research into the effects of microtextured regions on macroscopic properties, there is still a lack of rigorous definition of what constitutes a microtextured region. To this end, we present a computational study in which deformation simulations are performed to study the effect of the intensity and character of microtexture on the localization of plastic strain. Mean orientation and intensity of microtexture are parameterized in a suite of simulations. Simulations indicate that strain localization is facilitated both by the presence of some amount of orientation spread within a microtextured region, and further by the intensity of microtexture in neighboring regions. Simulations are further discussed with respect to high resolution experimental measurements of intragrain strain in a microtextured Ti-6Al-4V specimen, which exhibit similar mechanical trends.

The local mechanical properties of a dissimilar friction stir welded AA-2024-T3/AA-2198-T3 joint were documented during a uniaxial tensile test. High-resolution digital image correlation was performed during monotonic tensile tests to capture the local in-plane strain fields of the heterogeneous macrostructure of the weld. In the shoulder-affected region, banded macrostructures with heterogeneous mechanical properties were found. They were related to pronounced textures regions, which can be associated to strain-rate gradient during one rotation of the tool. The banded macrostructures in the nugget region were observed to be responsible for early plasticity in the joint and ultimately to be the fracture location of the weld. The heterogeneous mechanical response of the joint was also investigated by microhardness measurements. Differences were found between hardness and local tensile properties, demonstrating microhardness measurements can be misleading and only direct high-resolution digital image correlation techniques can document the mechanical behavior of materials having complex and heterogeneous micro-/macrostructures. Keywords: Friction stir welding (FSW), Digital image correlation (DIC), Electron backscattered diffraction (EBSD), Texture, Aluminum alloys, Banded macrostructures, Microhardness, Dissimilar joint

High resolution scanning electron microscope digital image correlation (SEM DIC) was performed in situ during uniaxial loading on Ti-6Al-4V rolled titanium plate to determine the dependence of strain localization on microstructure and microtexture. Individual grains with preferred orientation for basal slip exhibited plastic localization along basal planes before macroscopic yielding. With additional strain, but still below macroscopic yielding, pyramidal and prismatic plastic activity was observed as slip bands transmitting across many grains and entire microtextured regions (MTRs). The localization of long range plastic strain occurred within MTRs that allowed for slip transmission across grains with low angle boundaries. The rolled titanium plate material having strong [ 0001 ] and [ 10 1 ¯ 0 ] texture components showed pyramidal and prismatic type slip extending across entire MTRs at strains well below macroscopic yielding. These strain localization processes occur much earlier in straining and over lengthscales much longer than observed with conventional slip offset imaging. The implications for properties are discussed.

The mechanical properties of titanium alloys result from their complex multi-scale microstructural features such as micron scale precipitates and millimeter scale microtextured regions (MTRs). Deformation processes that operate at the scale of the α grain are of critical importance to mechanical properties, especially to fatigue performance. However, previous investigations also highlighted that the mm-scale MTRs affect the mechanical properties of titanium alloys. Specifically, MTRs promote long-range strain localization due to the low intergranular misorientation within a MTR. Furthermore, the elastic anisotropy of the alpha phase and the non-random spatial distribution of crystallographic orientations within the MTRs produce complex mechanical effects. The present work is a mechanistic investigation of MTRs using crystal plasticity simulations of mm3-scale experimentally captured and synthetically generated 3D microstructure datasets. The explicit modeling of both the α grains and MTRs in a titanium alloy is used to determine the effect of the degree of microtexture on the deformation behavior and on the onset of plastic slip. The presence of MTRs with a dominant [0001] orientation results in both stress and plastic strain hotspots during the early stages of straining. The influence of MTRs on the local stress and strain fields are analyzed and discussed with regard to the monotonic tension, fatigue and dwell-fatigue behavior of titanium alloys.

Fatigue crack nucleation in crystalline materials typically develops due to highly localized cyclic slip. During a fatigue cycle, reverse slip differs locally from slip in the forward direction particularly in precipitate-containing materials such as superalloys. In this paper we report the first direct measurements of irreversibility at the scale of individual slip planes in a polycrystalline nickel base superalloy. Quantitative measurements of the slip irreversibility is challenging to be performed for regions of material that have a size that captures the microstructure and its variability. High spatial resolution at the nanometer scale during experimental measurements is needed to observe slip localization during deformation. Moreover, large fields are also needed to obtain the material response over statistically representative populations of microstructural configurations. Recently, high resolution scanning electron microscope (SEM) digital image correlation (DIC) has been extended for quantitative analysis of discontinuities induced by slip events using the Heaviside-DIC method. This novel method provides quantitative measurements of slip localization at the specimen surface. In this paper, the Heaviside-DIC method is used to measure slip irreversibility and plastic accumulation in a nickel base superalloy. The local microstructural configurations that induces large amounts of plasticity and slip irreversibility are captured and correlated to crack nucleation locations.

Damage initiation during cycling loading of polycrystalline metallic alloys involves localized damage at the scale of individual grains. To better understand damage processes and to build models for material behavior, there is a need for quantitative assessment of the microstructural configurations that favor fatigue crack initiation. In materials that form annealing twins during processing, these special interfaces are often locations of particular interest for their role in strain and damage accumulation. In the present study, fatigue experiments in the very high and low cycle fatigue regime on a Rene 88DT polycrystalline nickel-base superalloy were performed to statistically evaluate grain-scale features that favor crack initiation. Combined elastic and plastic criteria at the grain scale have been developed. A crack distribution function is defined to compare and assess the effect of the microstructural parameters for the two fatigue regimes.

Damage during cycling loading of polycrystalline metallic alloys involves localized plastic straining at the scale of individual grains. To better understand damage accumulation processes and to build models for material behavior there is a need for quantitative assessment of the heterogeneous strain fields at the grain and even more microscopic scales. In the present study, a digital image correlation (DIC) approach has been developed to measure the strains at the grain level and at finer scales where plastic strain localization is manifested as physical slip bands. Strain fields have been measured in situ and ex situ on a Rene 88DT polycrystalline nickel-based superalloy to assess the grain-scale deformation processes during monotonic straining in tension and compression. DIC analysis and transmission electron microscopy demonstrate that slip occurs in a highly localized manner. The highest localized strains developed in slip bands that formed on {1 1 1} planes parallel to, and slightly offset from, annealing twins. Enhanced local straining below yield was observed during compression loading. The degree of strain concentration caused by slip bands impinging on grain boundaries was also analyzed. The results are compared to predictions of plasticity models.

The anisotropic lattice rotation of individual grains induced by plasma nitriding of 316L austenitic stainless steel has been analyzed with the aim of identifying correlations between the initial grain’s orientation and the rotation behavior. Due to the quite large nitriding-induced strains (up to 20%), the Taylor–Bishop–Hill model has been chosen for the simulation of the lattice rotations. The model predicts the overall rotations, both amplitude and direction, reasonably well over the entire stereographic triangle. The magnitude of the rotations is in agreement with the level of deformation induced by insertion of nitrogen atoms into an austenitic lattice. With regard to plasticity, parallels between the nitriding process and tensile elongation along the normal surface can be drawn.

The impact of plasma nitriding at 400 °C on the monotonic mechanical behaviour of 316L austenitic stainless steel at room temperature has been investigated. It is shown that the residual stresses in the nitrided layer lead to a tension–compression anisotropy whose magnitude depends on the residual stresses intensity and extension (i.e. thickness of the nitrided layer). Using the stress differential technique, average residual stresses in the nitrided layer ranging from −1.5 up to −3 GPa were measured. The local mechanical behaviour of the nitrided layer has also been investigated through SEM in situ tensile tests. A quasi-brittle mechanical behaviour of the nitrided layer is observed with first evidences of crack initiation for plastic strains below 1%, whatever the nitrided layer. By increasing the total applied strain up to 20%, a progressive segmentation of the layer occurs. The crack initiation location mainly depends on the local nitrided thickness.

The nature of the near-surface γN phase produced by low-temperature (∼400 °C) plasma-assisted nitriding of an austenitic stainless steel 304L is studied. A combination of global probes (X-ray diffraction, nuclear reaction analysis, glow discharge optical emission spectroscopy) and local probes (field ion microscopy, conversion electron Mossbauer, X-ray absorption near edge structure and extended X-ray absorption fine structure spectroscopies) is employed to reveal the morphology, phase structure, atomic ordering and chemical state of the obtained γN phase. The results consistently reveal the heterogeneous nature of the nitrided layer consisting of nanometric CrN precipitates embedded in a Fe4N-like matrix. The size of the precipitates is found to be larger at the surface than at the nitrided layer–steel interface. The precipitates have irregular, sphere-like shapes. Moreover, X-ray spectroscopic investigation revealed three different intermetallic distances and different chemical environments for Fe, Cr and Ni, accompanied by a large static disorder. These findings suggest that the presence of the interstitial N destabilizes the homogeneous element distribution in 304L even at such low temperatures. This leads to the segregation into Cr-rich zones that are coherent with the Fe4N matrix. Possible atomistic decomposition mechanisms are discussed. Based on the heterogeneous nature of the γN phase revealed in 304L, an alternative view of its remarkable combination of properties such as large hardness, induced ferromagnetism and preserved corrosion resistance is considered.

Correlations between the grain orientations and elastic properties of plasma-nitrided polycrystalline 316L austenitic stainless steel are investigated. The grain orientations (h k l) in a delimited area were obtained from electron backscatter diffraction and related to hardness (Hhkl) and elastic modulus (Ehkl) maps obtained from large nanoindentation matrices. The influence of nitrogen concentration on the local mechanical properties has been studied by repeating these indentation matrices in the same area after successive partial removals of the nitrided layer. This nanoindentation tomography allowed the orientation, the shape and the surroundings of individual grains to be taken into account. The results show that plasma nitriding leads to a complete reversal of the elastic behaviour anisotropy: while the non-nitrided 316L austenitic stainless steel shows the typical elastic anisotropy of face-centred-cubic-type metals with a maximum of Ehkl for the 〈 1 1 1 〉 oriented grains, the maximum of Ehkl is observed for the 〈 0 0 1 〉 oriented grains in the nitrided layer. A similar anisotropy reversal is observed for the hardness Hhkl. These observations are discussed on the basis of the microstructural changes induced by the nitrogen incorporation.

The changes in anisotropic hardness and indentation modulus induced by plasma nitriding at 400 °C of a 316L polycrystalline austenitic stainless steel are analyzed. The dependence of hardness and elastic modulus modifications on the crystallographic orientation is investigated through instrumented indentation and electron backscattering diffraction. Both hardness and indentation modulus exhibit an inverted anisotropy compared to the untreated 316L, likely associated with the presence of the N atoms in interstitial sites.

The introduction at moderate temperature of nitrogen in the 316L austenitic stainless steel by plasma nitriding modifies the crystallographic texture in the very near surface region. The evolution of texture components has been quantitatively characterized by electron backscattered diffraction. The analysis of these experiments shows that the amplitude of the lattice rotation as well as the direction of rotation are directly related to the initial orientation of each grain. The retexturing behaviour is consistent with the lattice rotation upon tensile elongation of polycrystalline materials predicted by the Taylor model.

AISI 316L austenitic stainless steel is widely used in nuclear and chemical industries for structural components that are submitted to cyclic deformation and stresses. The influence of low temperature (~ 400 °C) plasma nitriding on the crystallographic texture and fatigue durability of 316L has been investigated. Electron back-scattering diffraction measurements show that nitriding enhances the and texture components while the one vanishes. Series of fatigue tests carried out in air at room temperature in the low cycle fatigue range show a significant improvement of the fatigue life. Results are discussed taking into account the residual compressive stresses in the nitrided layer.