Abstract Microscopy and materials characterization can be optimized through the application of techniques that enable nanoscale observations within large areas, to help understand the contextual relationship between the local nanostructure and global microstructure. In this work, we demonstrate that electron channeling contrast imaging (ECCI) can now be deployed in a controlled manner, exploiting advances in electron channeling pattern (ECP) formation and rapid interpretation using new software tools. We demonstrate this capability via four case studies: single-crystal silicon, dislocations in GaAs, a deformed nickel polycrystal, and deformed olivine.
Abstract Grain rotation under concurrent grain boundary (GB) migration can be observed in experiments, but poses a challenge in simulations. The need to resolve atom-scale structures while tracing GB migration over extended time scales defy most modeling approaches, including molecular dynamics (MD). In this work, phase field crystal (PFC) modeling is adopted as it permits the required combination of high spatial resolution and extended time spans, orders of magnitude longer than what is feasible in MD. The possibility to trace grain rotation by PFC is demonstrated and it is shown that grain rotation is increasingly pronounced as the grain size is reduced and while misorientation is small. In contrast, grains demarcated by high-angle GBs exhibit negligible rotation during migration. The ability of small grains to rotate is shown to depend on the extent of GB dislocation reactions. The results also highlight that classical grain growth kinetics fail to apply under rotation of small grains with low-angle GBs.
Abstract We use Dark‑Field X‑ray Microscopy to image dislocation ensembles in bulk Al1050 during the first 2 % of tensile deformation. A beam‑stop in the back focal plane suppresses bulk‑lattice diffraction, enhancing contrast from the strained regions surrounding dislocation cores. Layered rocking scans provide 3D point‑cloud representations of the dislocation configurations in two neighbouring grains. The measurements reveal directional dislocation arrangements, partial alignment with crystallographic slip‑plane traces, and the early formation of low‑angle boundaries.
Abstract The review presents how contact of solid interfaces to liquid recycled Al alloys contribute to purification but also contamination of Al alloys and how interfaces can affect the microstructure evolution during solidification. Interaction at interfaces as wetting, reactive layer formation and dissolution are especially considered. Benefits of targeted utilization of interfaces for dealing with impurity elements and inclusions in recycled Al alloys are discussed.
Abstract The evolution of microstructure and mechanical properties during annealing has been studied in a 90% rolled CoFeMnNi-5Cr-5Al-1C multi-principal element alloy. The rolled microstructure is observed to be strongly heterogeneous, comprising twinned regions and shear bands, layers of mixed orientations with high frequencies of high-angle boundaries, and extended regions dominated by low-angle boundaries. During annealing at 800 °C, shear bands and layers of mixed orientations act as preferential sites both for nucleation of recrystallized grains and for precipitation of B2 and carbide particles. Tensile tests conducted at room temperature reveal high yield strength (∼800 MPa) and good uniform elongation (14% and 17%) in partially recrystallized samples obtained by annealing for 30 min and 1 h.
Abstract This work characterized the microstructural evolution during annealing at high temperatures of 316L stainless steel samples manufactured by laser powder bed fusion. Electron back scattering diffraction was employed to analyse the microstructure in a series of samples annealed for different times. It was found that domains with low internal misorientation appear and grow into a matrix of higher internal misorientation. The microstructural evolution was analysed by typical stereological methodologies. The fraction of material consisting of the low internal misorientation (LIM) domains, as well as the un-impinged surface area of these, were quantified as a function of annealing time. Based on these parameters, the average growth rate of the LIM domains was determined. It was found that the growth rate decreases by about 2 orders of magnitude during the annealing. Reasons for this are discussed.
Abstract This paper reports the results of in situ FE-SEM observations of microstructure evolution during heating of an Fe-24 mass% Ni alloy, focusing on two distinct stages: (1) reverse martensitic transformation from BCC lath martensite to FCC austenite during step heating from 673 K (below A S ), and (2) recrystallisation of the reverted, highly dislocated FCC austenite during subsequent heating. Specifically, the reversion of coarse BCC martensite crystals was investigated in the former process, while the spatial preference of nucleation and growth was examined in the latter. The observations revealed that in the early stage of transformation, an individual coarse BCC martensite crystal transformed into multiple plate-like FCC crystals with varying orientations and higher Kernel Average Misorientation (KAM) values than in the martensite matrix. Furthermore, FCC recrystallised grains preferentially nucleated along prior-austenite grain boundaries and tended to grow into matrix regions with higher KAM values.
Abstract Residual stresses within recrystallizing grains have recently been demonstrated, but their role in local boundary migration remains insufficiently understood. Most existing studies are either surface sensitive or do not provide the full three-dimensional strain state. In the present work, a migrating recrystallization boundary in high-purity Al (99.996%) is selected from previously quantified annealing experiments based on two-dimensional (2D) electron microscopy and examined in the bulk by synchrotron three-dimensional (3D) Laue micro-beam diffraction. The approach combines time-resolved migration data with voxel-resolved orientation mapping and fitting of the deviatoric strain tensor before and after an annealing step. This enables determination of principal elastic strain directions within the recrystallized grain and correlation of these with local boundary migration pattern in 3D. The results show that residual elastic strains play an important role in governing local boundary migration during recrystallization. Together with previous findings, this work highlights the need to consider residual stresses as an intrinsic part of recrystallization boundary migration, alongside classical descriptors such as boundary mobility, boundary plane, and dislocation structures in the deformed matrix.
Abstract Low-alloy steels typically undergo an austenitisation, quench and temper heat treatment before entry into service. Following the quench, it is not uncommon for metastable retained austenite (RA) to be present. Tempering allows for sufficient carbon migration to facilitate the decomposition of the RA, which has been seen to decompose into a variety of microstructures such as ferrite and carbides, bainite, and martensite. Since the exact decomposition mechanisms for RA are not fully understood in some alloys, and the resulting microstructures are likely to influence mechanical properties, this work uses high temperature scanning electron microscopy (SEM) including electron backscatter diffraction (EBSD) to image RA decomposition during heating. A low-alloy pressure vessel steel is used as a case study. Particular interest is paid to the temperatures over which decomposition occurs, the order of the decomposition events, and the resulting products. The decomposition of individual grains of RA was observed and found to decompose over a 60°C temperature range when heating at a rate of ∼0.1°C s −1 , most likely transforming into bainite due to the continuation of IPF from the bainite grains to the decomposed RA and surface relief. Experiments were conducted that used either secondary electron imaging or EBSD imaging as the primary imaging method in-situ, so that observations of RA behaviour could be drawn from each method, and to contrast the relative advantages and disadvantages of each. SE imaging was able to discern changes in the surface during the decomposition, but contrast was weak and the behaviour was unclear. EBSD provided greater crystallographic detail for a tempering phase transformation process, but was limited by spatial resolution.
Abstract Three-dimensional polycrystalline microstructure datasets from non-destructive x-ray characterization techniques such as diffraction contrast tomography (DCT) and high-energy diffraction microscopy (HEDM) encode rich information about grain boundary geometry, crystallography, and topology. Translating this into a reduced-order (e.g., graph) representation requires careful choices about feature engineering. Here, we present a systematic workflow for converting 3D grain maps into graph representations where grain boundaries are encoded as edges carrying physically motivated feature vectors. Fifteen grain boundary descriptors spanning three categories are defined: crystallographic character, topological connectivity, and geometric shape. To arrive at a compact, non-redundant feature set, we apply a two-stage downselection procedure combining Pearson correlation analysis with Kruskal-Wallis testing based on time-resolved grain growth dataset. The retained descriptors reflect the physical mechanisms governing boundary migration. This feature engineering framework also serves as the foundation for a topology-aware graph neural network that learns directly from experimental microstructure data, offering a path toward data-driven prediction of grain boundary behavior during thermomechanical processing.
Abstract The effects of strong carbide-forming elements, i.e., Nb, Mo and V, on the low-temperature tempering kinetics of high-carbon martensite during continuous heating were investigated using calorimetry, dilatometry, neutron diffraction, and three-dimensional atom probe analysis. Particular attention was given to the reduction in martensite tetragonality during the 0th and 1st tempering stages. The results indicate that, while the addition of 0.03 at.% Nb and 1.0 at.% Mo exerts only minor effects, the addition of 1.0 at.% V significantly retards tempering kinetics. This retardation is associated with suppressed carbon redistribution, as evidenced by delayed reductions in both tetragonality and solute carbon content during tempering. This effect is attributed to strong attractive M-C interactions, which enhance carbon trapping and reduce its diffusivity in martensite.
Abstract Grain growth governs the properties of polycrystalline materials, and understanding it in bulk requires non-destructive access to the 3D morphology of individual grains as they evolve. Dark field X-ray microscopy (DFXM) provides the necessary spatial resolution, and its recent pink beam extension (pDFXM) has enabled time-resolved two-dimensional tracking of grain boundary migration through a ∼100-fold increase in bandwidth. Extending this temporal gain to three dimensions requires both the photon flux of pink beam illumination and a compatible tomographic scheme, as watching grains evolve in 3D in real time is essential for validating grain growth theories in bulk. Magnified topo-tomography (MTT), the tomographic variant of DFXM, offers a natural route to 3D grain shape reconstruction and is a candidate for pink beam operation, but the broader bandpass alters how reciprocal space is sampled at each tomographic angle, and its fidelity under these conditions must first be established. Here, we benchmark pink beam MTT for the first time against its monochromatic counterpart, comparing reconstructed grain volume and surface topology metrics on the same individual grain in fully recrystallized AA1050 aluminium after high-temperature annealing, imaged ex situ under both illumination conditions. Using a surface-normal anisotropy metric to quantify local grain topology, we find that the pink beam reconstruction yields a ∼10% larger volume and a systematically smoother envelope for the grain of interest. Pink beam MTT thus retains spatial resolution comparable to monochromatic MTT while increasing flux, laying the groundwork for time-resolved three-dimensional grain growth studies.
Abstract Dark-field X-ray microscopy (DFXM) maps crystal lattice distortions and defects in millimeter-sized samples by recording rocking curves with sub-micrometer spatial resolution. Quantitative analysis is complicated by overlapping diffraction contributions within single voxels, which limits the reliability of standard peak-fitting approaches. Here, we formulate rocking-curve decomposition as a Bayesian inverse problem, jointly inferring the shapes and spatial distributions of distinct diffraction components with uncertainty quantification. The method separates weak-beam and strong-beam contributions, distinguishing regions where defect contrast is enhanced from the central region dominated by dynamical diffraction. The resulting decomposition reduces data complexity and facilitates dislocation identification.
Abstract Investigating abnormal grain growth (AGG) mechanisms requires tracking grain evolution and neighboring environments during annealing in 3D. In this work, we apply synchrotron-based diffraction-contrast tomography (DCT) and phase-contrast tomography (PCT) to obtain time-resolved grain maps and quantify second-phase particle distributions in the commercial aluminum alloy AA5252. Utilizing a robust registration workflow, we tracked local particle densities within the volumes that growing abnormal grains exchanged with their neighbors. These densities were then related to the particle pinning pressure via the Zener drag model, establishing a framework for analyzing particle-grain interactions in 3D in future grain growth studies.
Abstract The coupled effects of second-phase precipitates and deformation microstructures in AA2024 were investigated. Cold rolling is found to accelerate the precipitation of S-phase in initial T3 state through enhanced diffusion and heterogeneous nucleation on dislocations, leading to refined precipitate spatial distributions resulting in increased hardness, while promoting faster coarsening of the precipitates in the initial T8 condition (with S-phase volume fraction close to equilibrium). In parallel, the presence of non-shearable S-phase precipitates strongly influences deformation microstructures by increasing sub-boundary density and stored energy, and by modifying the spatial distribution of dislocations. Combined 2D and 3D analyses highlight more pronounced and more heterogeneous orientation gradients in the deformed grains of the initial T8 state with S-phase precipitates as compared to those of the initial T3 state without S-phase, emphasizing the role of precipitate dislocation interactions.
Abstract Four-dimensional imaging, namely, monitoring in-situ mechanical tests via computed tomography, provides big datasets to probe constitutive models. This paper discusses recent advances in using such datasets to extract mechanically relevant information for quantifying in-situ properties, damage, and fracture. One way of analyzing 4D datasets is to measure displacement fields via Digital Volume Correlation (DVC), and various extensions thereof. Three different examples illustrate how mechanical information could be extracted from in-situ tests on lab tomographs, and how 4D imaging redefines how material microstructures and their behavior are characterized across scales. It is shown that additively manufactured microstructures, and in particular printing defects, lead to lower in-situ mechanical properties in comparison to more classical processing routes.
Abstract Under the growing demands of higher strength on structural materials, microstructures of advanced high-strength metallic alloys have become increasingly complex. The microstructures are in many cases composed of multi phases having different mechanical properties, i.e., strength and ductility. In the present study, we tackled to quantitatively evaluate stress/strain partitioning between hard phase and soft phase in multi-phased alloys using state-of-the-art experimental techniques. The micro-DIC (digital image correlation) analysis and neutron/synchrotron X-ray diffraction were applied to the dual-phase steels composed of soft ferrite and hard martensite phases for analysing strain and stress partitioning, respectively, between the phases during tensile deformation. From the average strain and stress amount in each of two phases obtained from the DIC and diffraction methods, we could reconstruct the stress-strain curves of ferrite and martensite separately for the Girst time. It was interestingly found that the stress-strain behaviour of two phases, especially that of the hard phase (martensite), greatly changed with increasing the density of inter-phase boundaries between two phases, leading to excellent strength-ductility synergy of the whole material having the particular morphology of the dual-phase microstructures. The result indicates that the deformation constraint through inter-phase boundaries plays a critical role to maximize deformation ability of hard phase(s) in alloys with multi-phase microstructures.
Abstract Plastic deformation in metals can manifest as intra-grain orientation partitioning, where initially continuous grains subdivide into domains separated by sharp orientation gradients. Dark-field X-ray microscopy (DFXM) is well suited to follow such processes in bulk crystals by imaging the diffracted intensity from selected grains or domains. However, we show that in the plastic regime, orientation partitioning can make the local DFXM signal multi-modal in reciprocal space. Classical first-moment analysis then reduces the local diffraction response to a single center-of-mass, which can obscure peak splitting, underestimate orientation spread, and dampen gradients across domain walls. We explain that the origin of this effect is linked to the finite diffracting gauge volume in DFXM and show that conventional analysis smooths domain boundaries and misrepresents the local lattice-normal field. To address this, we introduce a multi-peak diffraction analysis based on local-maximum labeling, which separates distinct local diffraction modes and extracts peak-specific lattice information. The method is demonstrated on DFXM data from a deformed Al1050 crystal and compared with conventional center-of-mass analysis. Synthetic DFXM simulations with known ground truth show that the segmentation-based approach gives lower mean absolute and root mean squared orientation errors. The algorithm is implemented in the open-source Python package Darling , which provides tools for filtering, moment analysis, peak segmentation and visualization of DFXM data.
Abstract Pearlitic steels exhibit a characteristic lamellar structure of alternating ferrite and cementite. These steels offer an excellent cost-performance ratio, balancing strength, ductility, and wear resistance. However, their fine lamellar microstructure is challenging to resolve, and conventional techniques such as scanning electron microscopy (SEM) or transmission electron microscopy (TEM) are limited to observations of small volumes below the surface. In this work, we explore the capabilities of three advanced synchrotron techniques for characterising pearlite: Dark-Field X-ray Microscopy (DFXM), scanning 3D X-ray Diffraction (s3DXRD), and nano-holotomography. While observing the dominant ferritic phase is generally achievable with precise orientation mapping, we emphasise both the potential and challenges of imaging the thinner cementite lamellae, which lie near the resolution limits of the techniques presented in this work.
Abstract This study investigates the effects of in-situ laser heat treatment (LHT) during laser powder bed fusion (LPBF) of 17-4 PH stainless steel. Firstly, the thermal stability of as-built material was evaluated through post-build direct aging. Further, various LHT durations were applied to assess their effectiveness in microstructural tailoring and hardness enhancement. The results reveal that LHT successfully promotes copper precipitation while preserving the as-built phase constitution, creating a hardness profile that reflects distinct thermal exposure conditions beneath the treated surface. These findings demonstrate that LHT is a viable strategy for in-situ microstructural engineering. While LHT can be time demanding, it offers a practical pathway to enhance local mechanical performance of 17-4 PH components at the meso-scale directly during manufacturing.
