Ni-Cr-Mo alloys are widely used in the nuclear industry as structural materials due to their high temperature strength and corrosion resistance. Ni-based alloys containing around 33 at.% (Cr+Mo) developed a long-range ordered Ni2(Cr,Mo) phase after thermal aging and/or irradiation. The ordering mechanism for thermally-aged Ni2(Cr,Mo) phase is well-understood, characterized to be sluggish, homogeneous, and isotropic. The ordering mechanism for irradiation-induced Ni2(Cr,Mo) phase is not fully understood, characterized as having rapid formation and demonstrating anisotropic precipitation. This work elucidates the anisotropic precipitation and anisotropic precipitation mechanism of Ni2(Cr,Mo) after proton irradiation in Ni-Cr-Mo alloys. Selected area electron diffraction and bright-field scanning transmission electron microscopy imaging are used to image superlattice reflections from the ordered phase and irradiation-induced defects, respectively. A higher degree of anisotropic precipitation is observed with increasing dislocation loop and void size; a phenomenon not observed in thermally aged samples.

Analyzing microstructural defects in transmission electron microscopy (TEM) images, particularly in irradiated metal alloys, is often limited by the availability of high-quality, labeled data. To address this, we introduce a generative data augmentation approach using a mask-conditioned latent diffusion model (LDM) for synthesizing realistic TEM images with controllable, automatically labeled multi-class defect masks. Without requiring manual annotations for generation, our method enables the creation of synthetic image-mask pairs by sampling distributions learned from experimental masks. These generated data were used to augment small experimental datasets of varying sizes (10, 50, and 100 labeled experimental images) to train a Mask Regional Convolutional Neural Network (R-CNN) model for defect detection and classification. Our results show that generative augmentation yields small overall model performance improvements, with up to a 0.02 gain in the harmonic mean of detection and classification F1 scores. However, we also find that the relative contributions to detection and classification improvement depend on the specific train/test data split. These findings highlight the potential of targeted generative models to enhance deep learning performance in data-scarce microscopy-based image quantification tasks.

Quantifying prediction uncertainty when applying object detection models to new, unlabeled datasets is critical in applied machine learning. This study introduces an approach to estimate the performance of deep learning-based object detection models for quantifying defects in transmission electron microscopy (TEM) images, focusing on detecting irradiation-induced cavities in TEM images of metal alloys. We developed a random forest regression model that predicts the object detection F1 score, a statistical metric used to evaluate the ability to accurately locate and classify objects of interest. The random forest model uses features extracted from the predictions of the object detection model whose uncertainty is being quantified, enabling fast prediction on new, unlabeled images. The mean absolute error (MAE) for predicting F1 of the trained model on test data is 0.09, and the R^2 score is 0.77, indicating there is a significant correlation between the random forest regression model predicted and true defect detection F1 scores. The approach is shown to be robust across three distinct TEM image datasets with varying imaging and material domains. Our approach enables users to estimate the reliability of a defect detection and segmentation model predictions and assess the applicability of the model to their specific datasets, providing valuable information about possible domain shifts and whether the model needs to be fine-tuned or trained on additional data to be maximally effective for the desired use case.

Radiation-induced swelling in structural materials is a significant challenge for the safe operation of nuclear power plants. One of the promising solutions to accelerate the understanding of swelling is the development of machine learning tools for accelerated characterization of irradiated microstructures, focusing on cavities which contribute to a materials swelling response. In this paper, we examine an object detection model for cavities, YOLOv8, for cavity detection using the datasets from the Canadian Nuclear Laboratory (CNL) and Nuclear Oriented Materials & Examination (NOME). Specifically, we explored the use of Image Super-Resolution (ISR) techniques to enhance the detection performance either in the underfocused or overfocused condition for improved small cavity detection. The overall F1-score increase of 6.8% via ISR, highlights the effectiveness of ISR techniques in enhancing the model’s performance in detecting irradiation-induced cavities across all modalities. Furthermore, this study presents a comparative evaluation of YOLOv8 and Faster R-CNN, discussing their detection trade-offs and suitability for different imaging conditions. In addition to evaluating overall detection accuracy, this work assesses the impact of ISR on both models’ precision and recall for cavity detection.

Metallic thin films offer a platform to experimentally study the dynamics of microstructural evolution, but the required transmission electron microscopy (TEM)-based imaging generates complex images that are challenging to segment and quantify. This work provides a comparative analysis of a new YOLOv8 model and an established U-Net model for bright-field TEM images of polycrystals, employing a framework leveraging physical observables to evaluate performance against two hand-traced benchmark datasets. This methodology obviates the comparison of large, diversely structured, and manually labeled datasets that are required to assess performance on a per-image/per-pixel basis. It is found that the YOLOv8 model, adapted for real-time instance segmentation, has up to 43× faster inferencing (NVIDIA GeForce RTX 4090) compared to U-Net and reconstructs hand-traced grain size distributions (GSDs) with excellent fidelity, finding mean diameter within 3% for grains near an optimal magnification; for grains that deviate from the optimal pixel-diameter, the size of small- (large)-diameter grains is systematically over- (under)-estimated. This is partially mitigated by including scale-aware augmentations during training. Moreover, when the bias is corrected post-inference by a rigid shift in distribution, the YOLOv8 model reproduces ground truth GSDs with exceptional fidelity, with statistical tests indicating <5% probability that the distributions are distinct. Based on ground truth data, calibration curves pertaining to this shift can be constructed for a given model. This issue is not present in the U-Net model’s results, indicating that for quantitative measurements where the true size of objects is of interest, special procedures must be implemented for YOLO-based models.

The development of a Pb-shielded fixture for the execution of a small-angle neutron scattering (SANS)-based workflow for interrogation of highly irradiated nuclear materials has been explored. The Pb shielding was specially designed to reduce the detected radioactivity from the specimen during SANS experiments, and the overall configuration is termed shielded magnetic SANS (SM-SANS). Two FeCrAl-based alloys, C35M and 125YF, were examined with the SM-SANS technique using a free-form size distribution locally monodisperse model in both the as-received and irradiated states. Quantitative values derived from the free-form size distribution were compared with atom probe tomography experiments. Microstructural and compositional parameters determined using the two characterization techniques were complements of each other. The results demonstrate that the SM-SANS technique is an effective means of characterizing nanoscale clustering in irradiated material systems and provides new avenues for investigating radioactive material microstructures.

The integration of machine learning (ML) models enhances the efficiency, affordability, and reliability of feature detection in microscopy, yet their development and applicability are hindered by the dependency on scarce and often flawed manually labeled datasets with a lack of domain awareness. We addressed these challenges by creating a physics-based synthetic image and data generator, resulting in an ML model that achieves comparable precision (0.86), recall (0.63), F1 scores (0.71), and engineering property predictions (R2 = 0.82) to a model trained on human-labeled data. We enhanced both models by using feature prediction confidence scores to derive an image-wide confidence metric, enabling simple thresholding to eliminate ambiguous and out-of-domain images, resulting in performance boosts of 5–30

Whilst there is a clear scientific and technological need for the technical capabilities of transmission electron microscopes with in-situ ion irradiation, it also requires a collaborative community of international researchers to support such facilities in successfully meeting this demand. Instruments of this type serve to provide fundamental understanding of the mechanisms which drive changes in materials important to nuclear fission and fusion energy, the semiconductor industry, quantum information systems, space travel, astronomy, geology and many more applications. As these areas continue to evolve and the instrumentation possibilities expand, the capacity of in-situ ion irradiation facilities must also develop hand-in-hand with the user community to deliver an ever-greater diversity of high-fidelity extreme-environment experimentation. Future directions for the field, such as miniaturization from MEMS/microfluidic devices and advanced controls with ML-based analysis, continuously emerge to advance both the hardware and software which support the coupling of TEMs with ion beams. This review sets out to provide up-to-date insights into the community and advancement of current, and development of future, facilities which have the potential to further unlock access to the nanoscale exploration of coupled extreme environments crucial to many of the important science and engineering challenges we face today.

Precipitates are main microstructural features to provide high temperature creep strength and radiation resistance in structural materials for fusion energy systems. However, the mechanisms of precipitate stability under irradiation in candidate structural materials for fusion first-wall and blanket components are poorly understood. In particular, the dual effects of helium transmutation and irradiation-induced damage on precipitate evolution have not been systematically studied in candidate materials, the leading of which are Fe-9Cr reduced activation ferritic/martensitic (RAFM) alloys. To fill this knowledge gap, a fundamental understanding of the single and combined interactions of helium (0-25 appm He/dpa), temperature (300-600C), and atomic displacements (15-100 dpa) on the behavior of MX (M=metal, X=C and/or N) precipitates in an advanced Fe-9Cr RAFM alloy were studied through the use of dual ion irradiation experiments. It was found that helium suppressed the diffusion-mediated mechanisms of precipitate stability (i.e. radiation-enhanced growth) at elevated temperatures and intermediate damage levels but had no effect on precipitate dissolution in the high dose conditions (>50 dpa). A precipitate stability model was used to rationalize the impacts of helium on ballistic dissolution and radiation-enhanced diffusion which are key contributors to overall precipitate stability. This is the second paper in a series of three to provide a systematic evaluation of MX precipitate behavior in RAFM steels under various fusion-relevant ion irradiation conditions.

This work is the third and final part in an initial series on addressing the behavior of MX precipitate stability in an advanced Fe-9Cr reduced activation ferritic/martensitic (RAFM) alloy under fusion-relevant ion irradiation conditions. Here, the helium trapping properties of MX precipitates are investigated across varying damage levels (15-100 dpa), temperatures (400-600 degrees C), and helium doses (10-25 appm He/dpa) using sophisticated dual ion beam experiments and electron microscopy. Results indicate that individual MX precipitates efficiently sequester helium in the form of nanoscale bubbles at the precipitate-matrix interfaces near the peak swelling temperature (similar to 5 bubbles/precipitate at 500 degrees C). Swelling was primarily due to matrix cavities. The Fe-9Cr alloy reached 2% swelling by 100 dpa, suggesting a shift to steady-state swelling around 50 dpa at 500 degrees C. However, MX precipitate dissolution beginning at 15 dpa did not coincide with this onset of steady-state swelling.

Austenitic 316 stainless steel flux thimble tubes (FTTs) removed from a commercial pressurized water reactor (PWR) were re-irradiated using nickel ions to evolve the irradiated microstructure to higher damage levels than those achieved in reactor. The microstructures of the neutron-plus-ion irradiated samples were compared with those of neutron irradiated only samples at the same doses to evaluate the effectiveness of ion irradiation to extend the damage range and match that created in reactor. Ion irradiations effectively captured, both qualitatively and semi-quantitatively, the evolution of dislocation loops, nanocavities, nanocluster size and density, and radiation-induced segregation (RIS) with dose. Only Ni-Si clusters were observed in ion irradiated FTT samples while Ni-Si-Mn clusters that were frequently observed in reactor irradiated samples above ∼41 dpa were absent in ion irradiated samples probably due to ballistic mixing at the high ion irradiation damage rate. Two samples were ion irradiated to ∼160 dpa to predict the irradiated microstructure that may develop due to an increase of the damage accumulation by extending a PWR life from 40 years to 60 years. The extended ion irradiation results indicated that significant microstructure changes were unlikely to occur even to very high doses near common PWR relevant temperature conditions. Overall, the microstructure resulting from ion irradiation following neutron irradiation matched that of neutron to the same dose reasonably well in most cases.

This article investigates the fuel-cladding chemical interaction (FCCI) behavior of two commercial FeCrAl alloys, APMT composition (Fe-21Cr-5Al-3Mo wt.%) and C35M (Fe-13Cr-5Al-2Mo-0.2Si-0.03Y wt.%), after neutron irradiation. "H-cup" diffusion multiples of FeCrAl alloys and ceramic UO2 fuel were irradiated at a temperature of similar to 300 degrees C to a total estimated burnup of 26 GWd/tHM. Post-irradiation Examination results demonstrate the excellent degradation resistance of FeCrAl alloys as accident tolerant fuel (ATF) cladding materials in light water reactor conditions. The study concludes that there was no irradiation-induced defects observed in either of the two commercial FeCrAl claddings. The formation of amorphous Al/U mixed oxide was observed at the fuel-clad interface, which can serve as a tritium permeation barrier and protect against potential chemical attack from the fuel. The study attributed the formation of amorphous Al/U mixed oxide to the low temperature and limited time of neutron irradiation. APMT forms more distinct Cr and Cr-Fe intermetallic at the FeCrAl-UO2 interface than C35M due to the higher bulk Cr:Al ratio.

This work systematically reports grain boundary radiation-induced segregation (RIS), and insights into RIS mechanisms, in FeCrAl alloys. Recently, FeCrAl alloys have received significant attention as direct replacement accident tolerant fuel (ATF) claddings for light water cooled nuclear reactors, because of their corrosion and high-temperature oxidation resistance characteristics. One degradation method not studied in detail is RIS, which could sufficiently alter grain boundary chemistry, potentially compromising the superior aqueous corrosion resistance. This study focuses on candidate ATF FeCrAl alloys C06M, C35M, C36M, and C37M that are neutron irradiated to 1.8 displacements per atom (dpa) at 357°C. Grain boundary Cr enrichment and Al depletion are observed in all irradiated alloys that span the 10-13 wt.% Cr and 5-6 wt.% Al composition space. Fe will either enrich or deplete at grain boundaries to balance the grain boundary composition. The Cr enrichment is attributed to interstitial diffusion, consistent with that in body centric cubic 9-12 wt.% Cr steels. The depletion of Al occurs through the same mechanisms as α−α′ phase partitioning previously observed in FeCrAl alloys. This study underscores the need for FeCrAl alloy design to consider RIS implications on corrosion and oxidation susceptibility.

Accurately quantifying swelling of alloys that have undergone irradiation is essential for understanding alloy performance in a nuclear reactor and critical for the safe and reliable operation of reactor facilities. However, typical practice is for radiation-induced defects in electron microscopy images of alloys to be manually quantified by domain-expert researchers. Here, we employ an end-to-end deep learning approach using the Mask Regional Convolutional Neural Network (Mask R-CNN) model to detect and quantify nanoscale cavities in irradiated alloys. We have assembled a database of labeled cavity images which includes 400 images, > 34 k discrete cavities, and numerous alloy compositions and irradiation conditions. We have evaluated both statistical (precision, recall, and F1 scores) and materials property-centric (cavity size, density, and swelling) metrics of model performance, and performed targeted analysis of materials swelling assessments. We find our model gives assessments of material swelling with an average (standard deviation) swelling mean absolute error based on random leave-out cross-validation of 0.30 (0.03) percent swelling. This result demonstrates our approach can accurately provide swelling metrics on a per-image and per-condition basis, which can provide helpful insight into material design (e.g., alloy refinement) and impact of service conditions (e.g., temperature, irradiation dose) on swelling. Finally, we find there are cases of test images with poor statistical metrics, but small errors in swelling, pointing to the need for moving beyond traditional classification-based metrics to evaluate object detection models in the context of materials domain applications.

Journal Article Evaluation of Human-Bias in Labeling of Ambiguous Features in Electron Microscopy Machine Learning Models Get access Gabriella Bruno, Gabriella Bruno Department of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, United States Corresponding author: gbruno@umich.edu Search for other works by this author on: Oxford Academic Google Scholar Matthew J Lynch, Matthew J Lynch Department of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, United States Search for other works by this author on: Oxford Academic Google Scholar Ryan Jacobs, Ryan Jacobs Department of Materials Science and Engineering, University of Wisconsin – Madison, Madison, WI, United States Search for other works by this author on: Oxford Academic Google Scholar Dane D Morgan, Dane D Morgan Department of Materials Science and Engineering, University of Wisconsin – Madison, Madison, WI, United States Search for other works by this author on: Oxford Academic Google Scholar Kevin G Field Kevin G Field Department of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI, United States Search for other works by this author on: Oxford Academic Google Scholar Microscopy and Microanalysis, Volume 29, Issue Supplement_1, 1 August 2023, Pages 1493–1494, https://doi.org/10.1093/micmic/ozad067.767 Published: 22 July 2023

This work investigates the micromechanical deformation and failure mechanisms of advanced FeCrAl alloys developed for use in nuclear reactors as cladding material. Three different FeCrAl alloys (82-X)Fe-13Cr-5Al-X (X=Nb, TiC, 0 for base alloy) were investigated both in their as-received and welded states. In-situ neutron diffraction with simultaneous digital image correlation was used to determine micromechanical deformation mechanisms not only as a function of elemental composition but also as a condition of state (as-received vs. welded). Ex-situ X-ray computed tomography was used on as-deformed samples to help determine the failure mechanisms and void initiation strains.

Face-centered-cubic (FCC) based materials have long been used for core materials in existing light-water reactors, and are proposed to be used in various next-Gen nuclear reactor systems. The potential extensive adoption of FCC-based materials, such as austenitic stainless steels, nickel-based superalloys and high-entropy alloys, is due to their acceptable radiation resistance, high-temperature corrosion resistance, ductility and creep resistance. The analysis of dislocation loops in these FCC materials under irradiation – of ½ ⟨ 110 ⟩ {110} perfect and ⅓ ⟨ 111 ⟩ {111} faulted types – is of particular interest because dislocation loops are known to contribute to the mechanical property degradation through hardening and embrittlement. Yet, there has not been a way to effectively characterize both types of dislocation loops of all variants in irradiated FCC materials using transmission electron microscopy (TEM) based techniques. The traditional Rel-Rod conventional TEM (CTEM) dark-field imaging has been extensively used for imaging two variants of faulted loops (out of four in total) that appear edge-on at [011] zone axis. However, this technique fails to image the other two faulted loop variants, nor any of the six perfect loop variants. All these overlooked features contribute significantly to irradiation hardening. Another traditionally used method, called two-beam condition imaging in both CTEM and scanning TEM (STEM) modes requires

FeCrAl alloys are promising candidate materials for the accident tolerant fuel (ATF) cladding application due to their exceptional resistance to oxidation in elevated temperature steam environments. Currently, limited fracture toughness data are available for the FeCrAl alloys, including the FeCrAl alloys newly developed at Oak Ridge National Laboratory (ORNL) under the U.S. Department of Energy’s Advanced Fuels Campaign (AFC) program. In this study, two Generation II candidate FeCrAl alloys, i.e., C06M (81.8Fe-10Cr-6Al-0.03Y-2Mo-0.2Si) and C36M (78.8Fe-13Cr-6Al-0.03Y-2Mo-0.2Si), were irradiated in the High Flux Isotope Reactor (HFIR) at ORNL to assess the fracture characteristics of these alloys after neutron irradiation. A total of six rabbit capsules were irradiated in HFIR at target temperatures of 200°C, 330°C, and 500°C up to target damage doses of 8 displacements per atom (dpa) and 16 dpa. Post-irradiation fracture toughness testing was performed following the Master Curve method in the ASTM E1921 standard. The main findings of this study are: 1) Both the C06M and C36M alloys exhibited a similar response to irradiation concerning irradiation hardening and embrittlement. 2) The irradiation temperature played different roles in terms of irradiation hardening and embrittlement for both C06M and C36M: after irradiation between 166°C and 204°C, both materials exhibited significant irradiation hardening and embrittlement; after irradiation between 315°C and 343°C, both materials showed small irradiation hardening without irradiation embrittlement. After irradiation between 501°C and 507°C, however, the irradiation softening without irradiation embrittlement was observed in both materials. 3) Comparing the microhardness and Master Curve reference temperature T0q before and after neutron irradiation, we did not observe a linear correlation between the two parameters for both C06M and C36M steels. This should be mainly due to a flat response of the Master Curve reference temperature T0q to the irradiations at 166–204°C and 315–343°C ranges 4) C06M showed a lower T0q, meaning better toughness, than C36M at the unirradiated condition, and such trend was kept even after neutron irradiation except for the 166–204°C irradiation after which both materials had similar T0q. 5) In terms of hardening and embrittlement, the irradiation effect on both C06M and C36M appeared to saturate after an irradiation dose of 7 dpa.

Post-neutron irradiation examination is performed on advanced accident-tolerant fuel (ATF) cladding iron-chromium-aluminum (FeCrAl) alloys with similar to 10-13at. % Cr, similar to 10-12 at. % Al, similar to 1 at. % Mo, and minor alloying elements including Y irradiated to a damage level of 7 displacements per atom (dpa) at irradiation temperatures of 267-282 degrees C. A compositional dependency of the Cr and Al content is observed on the ratio of sessile and glissile dislocation loops, where the density of a (100) type loops is some-what higher than the a/2(111) type loops. The alpha' precipitate number density is inversely correlated to the starting Cr concentration of the alloys of interest. The irradiation to a higher dose of 7 dpa results in a higher density of dislocation loops and alpha' precipitates for the same alloys at a lower irradiation dose, such as 1.8 dpa. In this work, the effect of alpha' precipitates on the dislocation loop density is discussed, and the presence of alpha' appears to inhibit the nucleation of loops. Compared with first-generation FeCrAl alloys, these advanced alloys with heterogeneous structure exhibit a lower Cr concentration in alpha' precipitation at the same dose level; they act as weaker obstacles deviating from the primary hardening contribution from the mature alpha'. Hence, the overall irradiation-induced hardening decreases; our alloys show improved radiation resistance because of their stronger sink strengths. The results presented in this paper could provide insights for the design and optimization of ATF cladding materials for future fission and space applications. (C) 2022 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.