This study examines the constitutive behavior of natural silica sand specimens subjected to one-dimensional (1D) high-strain-rate (HSR) loading through confined compression experiments, grain survival analysis, finite element simulations, and force-chain characterization. Sands with different gradations, particle sizes, and morphologies were tested to evaluate how particle-scale properties influence macroscopic constitutive response. Although all specimens exhibited the characteristic sequence of initial compaction, strain hardening, and post-peak softening, their responses differed significantly depending on packing characteristics and grain morphology. The fully graded F50 sand exhibited the highest peak stress and the smoothest stress evolution, indicating enhanced packing efficiency and more effective stress transmission. In contrast, poorly graded (uniform) specimens and the angular GS40 sand showed lower strength, greater stress fluctuations, and increased susceptibility to grain breakage. Grain survival analysis further confirmed that the fully graded specimen had the highest survival percentage, whereas the angular GS40 sand had the lowest. Force-chain analyses revealed that specimens with more distributed load-bearing networks tend to preserve a greater fraction of grains, while localized stress-transfer paths promoted fracture and fragmentation. Overall, the results demonstrate that well-graded sand develop more stable internal force networks and exhibit greater resistance to grain breakage under high-strain-rate loading.

Additive manufacturing (AM) has the potential to transform component production, but its widespread adoption is constrained by defects, such as pores, that compromise structural integrity. This study investigates the influence of porosity on the micromechanical and fatigue response of AM Inconel 718 (IN718), a widely used aerospace superalloy. One baseline specimen with minimal stochastic porosity and another intentionally seeded with lack of fusion (LOF) pores were examined using high-energy X-ray diffraction microscopy (HEDM) and micro-computed tomography during cyclic loading. Fatigue cracks in the LOF specimen initiated more frequently and at lower cycle counts than in the baseline specimen. The number fraction of fatigue cracks that grew was comparatively higher in the LOF specimen. Grain-level metrics, including stress and diffraction spot widths, were quantified using far-field HEDM. Across the thousands of grains detected in both specimens, the pores in the LOF specimen increased the variability of stress and spot width evolution in the azimuthal direction, the latter serving as a surrogate measure of plastic deformation. However, no correlations emerged between these metrics and grain proximity to fatigue crack initiation sites or pores, underscoring the limitations of grain-averaged metrics for predicting fatigue. Nonetheless, the rich dataset reported here provides a foundation for future modeling efforts.

Additively manufactured (AM) metals offer the potential for customizable, cost-effective components, but qualification and certification are crucial. Key to this process is understanding pore dynamics under stress, typically analyzed using micro-computed tomography. This study introduces laboratory-scale “betatron” x-rays from laser wakefield acceleration as a high-throughput alternative for x-ray tomography of advanced materials, such as AM AlSi10Mg alloys. Coupled with 3D finite element modeling, this method provides detailed insights into stress-porosity interactions. The approach delivers high-resolution scans, revealing that pore shape and local triaxiality significantly influence fracture dynamics, supporting advanced material characterization. This work also demonstrates the potential and versatility of laser-betatron x-ray [Formula: see text]CT for generating large datasets to accelerate our understanding of the stochastic, process-specific nature of pore formation in AM alloys.

X-ray computed tomography (CT) is a widely used imaging technique that provides detailed examinations into the internal structure of an object with synchrotron CT (SR-CT) enabling improved data quality by using higher energy, monochromatic X-rays. While SR-CT allows for improved resolution, time-resolved experimentation, and reduced imaging artifacts, it also produces significantly larger datasets than conventional CT. Accurate and efficient evaluation of these datasets is a critical component of these workflows; yet is often done manually representing a major bottleneck in the analysis phase. While deep learning has emerged as a powerful tool capable of providing a wide range of purely data-driven solutions, it requires a substantial amount of labeled data for training and manual annotation of SR-CT datasets is impractical in practice. In this paper, we introduce a novel framework that enables automatic segmentation of large, high-resolution SR-CT datasets by eliminating the need to hand label images for deep learning training. First, we generate pseudo labels by clustering on the voxel values identifying regions in the volume with similar attenuation coefficients producing an initial semantic map. Afterwards, we train a segmentation model on the pseudo labels before utilizing the Unbiased Teacher approach to self-correct them ensuring accurate final segmentations. We find our approach improves pixel-wise accuracy and mIoU by 13.31

The increasing complexity of in situ high-energy diffraction microscopy (HEDM) experiments demands a quantitative understanding of the data analysis pipeline to ensure reproducible science. However, the influence of key analysis parameters on the accuracy and precision of microstructural reconstructions is often not well quantified, creating a barrier to progress. This paper addresses this critical gap by presenting a rigorous, systematic validation of the HEDM data reduction methodology as implemented in the MIDAS software suite. Using a new, dedicated Ti-7 Al dataset, we investigate both far-field (FF) and near-field (NF) HEDM. Our results reveal critical sensitivities, demonstrating that grain position accuracy in FF-HEDM is highly dependent on the diversity of sampled diffraction vectors, while orientation precision in NF-HEDM improves dramatically with increased detector separation. We demonstrate the methodology's robustness against common experimental challenges, such as severe diffraction peak overlap, which is effectively filtered by requiring crystallographic consistency. Based on these quantitative findings, we establish a framework of best practices for HEDM data acquisition and analysis to guide the community towards more accurate and reliable results.

Reconstruction quality in Far-field High-Energy Diffraction Microscopy (FF-HEDM) is limited by the spatial reso lution of area detectors and the frequent occurrence of overlapping diffraction spots. To address these challenges, we developed a super-resolution (SR) framework using convolutional neural networks (CNNs) to recreate 2D diffraction peaks at up to & times;8 resolution from raw detector data. A specialized simulation tool was created to generate synthetic training datasets with varying degrees of peak overlap. Integrated into the Microstructural Imaging using Diffraction Analysis Software (MIDAS), the SR model improves the spatial accuracy and precision of 3D grain reconstruction by an order of magnitude. This approach provides a robust solution for investigation of complex micromechanical states and material classes where the analysis is limited by the presence of overlap ping peaks. Furthermore, the methodology developed here can potentially be extended to other techniques that require sub-pixel accuracy for high-fidelity data analysis.

Brittle fracture is difficult to study in situ due to the speed of a growing crack and the often-catastrophic nature of failure in brittle materials. As a result, the influence of microstructural considerations, such as orientation, grain boundary locations, and strain field, on the crack path remains poorly understood. Presented in this study is a method addressing this knowledge gap, which utilizes the double-cleavage drilled compression geometry to achieve quasi-stable fracture in aluminum oxynitride (AlON). Synchrotron X-ray micro-computed tomography is used to characterize the crack shape and length, while high-energy diffraction microscopy provides information on the strains, orientations, and shapes of grains in the microstructure surrounding the crack tip. During testing, the crack grew in discrete and irregular jumps while the fracture toughness falls within reported ranges. The crack in AlON is found to have no greater tendency to crack intergranularly as compared to transgranularly, and grains which are cracked transgranularly do not display a trend in orientation or stress when compared to those around which the crack followed a grain boundary. The high resolution of the crack path and microstructural data provides a path forward for modeling and understanding 3D brittle fracture.

The inversion of large-scale diffraction datasets from modern synchrotron sources presents a fundamental challenge in computational crystallography. This paper presents a unified algorithmic framework for the analysis of both near-field (morphological) and far-field (orientational and strain) high-energy diffraction microscopy (HEDM) data. We detail the mathematical formalisms and physical models that form the foundation of this methodology. Key aspects include a generalized model for detector distortion correction, robust algorithms for peak identification in noisy and overlapping patterns, a computationally efficient indexing formalism based on Friedel pair symmetry, and a decoupled iterative refinement scheme that exploits the differing sensitivities of position, orientation and lattice parameters to diffraction observables. We also describe the synergistic integration of near-field and far-field data streams, a critical feature of a truly comprehensive approach. The framework is validated in Part II of this series [Sharma et al. (2026). Acta Cryst. A82, 305-320] using both experimental Ti-7 Al datasets and synthetic reconstructions with known ground truth, achieving orientation accuracy of ∼0.05° and position accuracy of ∼10 µm on experimental data, and a 190× improvement in lattice parameter precision over conventional simultaneous parameter refinement on synthetic data. This integrated framework provides a powerful and extensible solution for turning raw diffraction images into actionable microstructural and micromechanical information.

Point-focused high-energy diffraction microscopy (pf-HEDM), also known as scanning 3D X-ray diffraction, is used to measure the intragranular elastoplastic deformation inside a polycrystalline commercially pure titanium sample during in-situ tensile loading. A pf-HEDM reconstruction framework is developed to accurately resolve intragranular orientations and local elastic strain tensors at each load step. Based on these measurements, we present a method for determining intragranular critical resolved shear stress (CRSS) distributions in polycrystalline materials. The intragranular CRSS method yields CRSS values for prismatic, basal, pyramidal < a >, and pyramidal < c + a > slip systems, and the resulting intragranular CRSS ratios are compared with room-temperature values reported in the literature for CP-Ti. The results reveal a distribution of slip activation strengths, attributed to pre-existing dislocations and spatially heterogeneous hardening; on average, the response follows Taylor hardening, with scatter arising from complex dislocation activity. We also use these results to demonstrate that performing Schmid factor analysis using grain-averaged or macroscopic stresses can lead to inaccuracies, including incorrect inference of active slip systems. Finally, we discuss the origin of local stress concentrations that govern plastic deformation in this material as coming from a combination of grain-grain interactions, elastic anisotropy, and static equilibrium. These findings highlight the importance of spatially resolved stress measurements for accurately capturing intragranular plasticity and ultimately improving the predictive fidelity of micromechanical models.

Two critical questions in brittle rock mechanics are how rocks developed localized strains and to what extent internal stress heterogeneity controls this localization and subsequent macroscopic failure. Definitive answers have not yet emerged, but would provide insight into rock fracture mechanics as relevant to hydrocarbon extraction and sequestration. Here, we use synchrotron X-ray tomography (XRT) and 3D X-ray diffraction (3DXRD) during uniaxial and triaxial tests on Nugget and Bentheimer sandstones to examine strain and stress localization prior to mechanical failure. 3DXRD was used to measure intra-granular lattice strains which were used to compute elastic stress tensors of each grain. Digital volume correlation (DVC) was applied to XRT images to determine the strain field in the sample. Both samples featured marked spatial heterogeneity, localization, and temporal persistence of elevated stresses and strains during their mechanical deformation toward failure. Both samples featured a majority of grains with at least one principal stress component that was tensile, a signature of the influence of heterogeneity on stress transmission. Measurements further revealed that compressive stress orientations and statistics evolved in a similar manner to those of inter-particle forces in loose granular materials, with triaxially-compressed rock exhibiting enhanced grain stress heterogeneity compared to uniaxially-compressed rock. Our results complement recent work by others who employed XRT and scanning 3DXRD to study triaxially-compressed sandstone, but extend those results to uniaxial compression, sandstones of varied porosity, and grain stress measurements throughout the 3D full extent of the samples rather than in a single layer examined with scanning 3DXRD.

Variation of strain rate sensitivity among different families of slip systems in the hexagonal close-packed ( ) phase of titanium (Ti) alloys has the potential to alter microscale load redistribution during both creep and dwell fatigue loading. However, existing literature contains conflicting reports regarding the degree of anisotropy present in the strain rate sensitives between slip system families across Ti alloys. Here, we quantify the strain rate sensitivity of slip system families in Ti-6Al-4V using high-energy X-ray diffraction microscopy (HEDM). We present a novel procedure in which we utilize the HEDM-measured grain-scale stress states during stress relaxation to determine the strain rate sensitivities of the basal ⟨ a ⟩ , prismatic ⟨ a ⟩ , and first-order pyramidal ⟨ c+a ⟩ slip system families. In addition, we measure the strain rate sensitivity exponent of the macroscopic response and the phase for comparison. We find rate sensitivities for the different slip system families to range from 0.02 to 0.04, which—while varied—are relatively isotropic in comparison with some values presented in the literature. We also demonstrate the effects of the measured anisotropic rate sensitivities using crystal plasticity finite element simulations.

Synchrotron-based x-ray tomographic imaging enables the examination of the internal structure of materials at high spatial and temporal resolution. Experimental constraints can impose dose and time limits on the measurements, introducing a higher level of noise and artifacts in the reconstructed images. Deep learning has emerged as a powerful tool to remove noise from reconstructed images. Recently, the Noise2Inverse method was designed specifically for denoising reconstructed images without requiring paired noisy and clean images. This method creates multiple statistically independent reconstructions used to pair the data in which training involves transforming one reconstruction into the other, and vice versa. Originally designed to be used after a fixed number of epochs, we see in practice that this approach may not produce the optimal model and may unnecessarily waste computational resources. Therefore, we propose an alternative method of identifying the best model during training that aligns with the Noise2Inverse method. During validation, we compare the model output of the multiple reconstructions among each other. We hypothesize that the best model is the one that produces images with the highest similarity, implying a convergence in the predicted material properties and absorption values. To compare model outputs, we consider the absolute error, square error, structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), and cosine similarity. We evaluate our method on two simulated tomography datasets and two, real-world, low-contrast, high-energy x-ray tomography datasets. We show our approach is more effective at determining the best model, up to an increase of 12.50% and 12.53% in SSIM and PSNR, respectively, while only requiring a fifth of the training time compared to the original approach.

Alkali-activated slag (AAS) is a promising low-CO2 alternative cement consisting of several disordered phases of similar composition. Although their local atomic arrangements are known to influence macroscopic behavior, determination of structural changes in response to external stimuli remains a challenge. Here, X-ray diffraction-computed tomography (XRD-CT), pair distribution function-CT (PDF-CT), and nanoprobe X-ray fluorescence (nano-XRF) have been used to uncover how an increase of magnesium in AAS affects the atomic structure and spatial arrangement of phases after aggressive carbonation (100% dry CO2), conditions experienced in applications such as oil and gas wells and geological storage of CO2. From PDF-CT it is found that a higher magnesium content decreases the average nanoscale crystallite size of disordered calcium carbonate. At the same time, higher magnesium content is correlated with a less decalcified C-(N)-A-S-H gel, as determined via analysis of Ca-Si atom-atom correlations from PDF-CT and Ca/Si ratios from nano-XRF. Finally, nano-XRF reveals that the disordered (i.e., amorphous) calcium carbonate is stabilized by the presence of silicates.

High temperature ceramic protective coatings, also called thermal barrier coatings (TBCs), are used to protect components, which are internally or back-side cooled. Calcium-magnesium-alumino-silicate (CMAS) particulates, such as sand, are ingested by these aero-engines and travel through the hot gas stream, eventually depositing onto the TBCs. The subsequent degradation is detrimental to the lifetime of the coatings with the risk of engine failure. In this work, synchrotron X-ray diffraction measurements were used to capture effects CMAS has on a 7-8 wt% yttria-stabilized zirconia (YSZ) TBC applied on an internally cooled hollow dogbone sample under thermal gradient conditions. Thermochemical degradation of the YSZ coating was observed, after 1 hour of CMAS-infiltration, through the formation of monoclinic phase. The monoclinic phase concentration was observed to be around 18% at the surface of the YSZ coating and declining towards zero halfway through the coating. This monoclinic concentration is roughly 2 times less than under isothermal conditions at similar annealing temperatures and timescales. These results showcase the successful capture and monitoring of coating degradation throughout a replicated aircraft flight cycle. These results can be further expanded upon with other coating systems, to elucidate the transitory processes of in-situ coating degradation throughout an operational cycle. The results may be used in strategies to mitigate the detrimental effects of CMAS on high temperature ceramics.

High-Energy Diffraction Microscopy (HEDM) is a powerful technique for in-situ characterization of metallic micro-structures, but traditional analysis methods are too computationally intensive for real-time experimental steering, often taking hours for a single scan. While machine learning frameworks like Rare Event Indicator (REI) achieve significant speedups, they face critical performance bottlenecks that render them insufficient for next-generation detectors or light sources like the APS Upgrade, which will increase data rates by over 100-fold. The primary challenge for accelerating the REI framework stems from a fundamental data scale disparity: the workflow must ingest massive, I/O-intensive images (e.g., 2048x2048) from disk while processing tiny, computationally inefficient patches (e.g., 15x15) on the GPU. This mismatch leads to high I/O costs and severe CPU-GPU load imbalance. To address these limitations, we present FastREI, a high-performance framework designed for CPU-GPU systems. FastREI implements a suite of optimizations, including strategic device placement based on workload profiling, parallel data processing via array partitioning, advanced load-balancing techniques to saturate the GPU, and kernel fusion to reduce launch overhead. Our integrated approach reduces the end-to-end workflow time by approximately 90%, achieving a 10-fold speedup over the baseline REI framework. This acceleration enables real-time data analysis at the extreme data rates of modern light sources, paving the way for adaptive, high-throughput materials science experiments.

Computed tomography (CT) is an essential imaging technique that utilizes x-ray measurements taken from different angles of the object allowing for detailed investigations into the internal structure of the material. During the experimental setup, factors can arise that corrupt the produced images with varying levels of background noise. Methods for denoising have been extensively studied with recent approaches favoring deep learning-based methods due to their superior performance. However, a limitation of these methods is their poor generalizability. At synchrotron based light source facilities, researchers regularly bring distinct samples to perform specific investigations. Therefore, models unable to generalize to new samples must be trained from scratch for each use case potentially slowing down the CT workflow. In this work, we take the first step towards developing a universal model for denoising CT data by iteratively fine-tuning models on new samples. To ensure models don’t suffer from catastrophic forgetting, we incorporate knowledge distillation using a student-teacher framework. We evaluate our approach on a simulated CT dataset showing an improvement of 24.64% and 62.56% in the SSIM and PSNR, respectively, on past data when using knowledge distillation. When applied to real-world experimental datasets, our method shows a substantial improvement in recovered visual features.

Advances in deep learning have seen widespread success in the field of computed tomography (CT) enabling automatic segmentation while reducing variability and subjectivity associated with manual evaluation. While these models have reached or surpassed human level performance, they require a substantial amount of labeled data for training and accurately annotating datasets is tedious, labor-intensive and costly. Recently, numerous studies have explored pseudo labeling as an alternative that overcomes manual curation yet require robust training methods to limit the effects of label noise. In this paper, we propose a novel framework that derives and incorporates a per-pixel uncertainty score of the pseudo labels for self-training the model. Specifically, we generate the pseudo labels using the K-means algorithm clustering on the pixel value and calculate the uncertainty of the label by measuring the absolute difference between the pixel and its distance between the two nearest cluster centroids. During training, we re-weight the standard cross-entropy loss by element-wise multiplication of the uncertainty score to reduce the weight of noisy regions in the image. We find our method improves pixel-wise accuracy and mIoU by 2.28% and 4.13%, respectively, on a simulated CT dataset. When applied to real-world experimental datasets our method produces segmentations that are considerably better than the original pseudo labels.

Subsurface processes in sandstones are controlled by porosity, permeability, and deformation mechanisms, all of which are controlled by a complex interplay of crystallographic rock texture, structure, and micromechanics. Texture, structure, and micromechanics have historically been studied using optical and electron microscopy of thin-sections. We employed a new combination of in situ X-ray tomography and ray diffraction microscopy to study crystallographic texture, structure, and grain stresses in 3D. We examined these features in a sample of Nugget sandstone, a sandstone constituting hydrocarbon reservoirs across the American West. Our aims are threefold. First, we demonstrate the utility of X-ray diffraction microscopy probes for revealing texture, structure, and stress transmission in 3D. Second, we apply these techniques to Nugget sandstone and discuss findings in the context of prior work. Third, we study grain stress tensor evolution during mechanical compression to examine whether their heterogeneity and orientation evolution reflect that of inter-particle forces in granular materials. Our results show: (a) larger grains featured higher intra-granular misorientations, possibly from an increased prevalence of cements; (b) pores closed parallel to the loading direction and opened normal to loading; (c) grain stresses featured heterogeneity and orientations similar to inter-particle forces in non-cohesive granular materials; (d) grains featured compressive stresses in the loading direction and tensile stresses orthogonal to the loading direction, the latter resisting sample dilation and grain separation. Our work demonstrates the first known application of multi-modal X-ray tomography and diffraction microscopy to sandstone, providing new 3D insight into the nature of quartz cement and stress evolution.