SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Severe xerostomia is noted in the majority of patients irradiated for oropharyngeal cancer. Extracellular microRNAs (miRNAs) may serve as effective tools allowing prediction of radiation-related toxicity. The aim of this study was to create an efficient prognostic miRNA-based test for severe, patient-rated xerostomia 3 months after primary treatment.
Among all RNA viruses, coronavirus RNA transcription is the most complex and involves a process termed “discontinuous transcription” that results in the production of a set of 3'-nested, co-terminal genomic and subgenomic RNAs during infection. While the expression of the classic canonical set of subgenomic RNAs depends on the recognition of a 6- to 7-nt transcription regulatory core sequence (TRS), here, we use deep sequence and metagenomics analysis strategies and show that the coronavirus transcriptome is even more vast and more complex than previously appreciated and involves the production of leader-containing transcripts that have canonical and noncanonical leader-body junctions. Moreover, by ribosome protection and proteomics analyses, we show that both positive- and negative-sense transcripts are translationally active. The data support the hypothesis that the coronavirus proteome is much vaster than previously noted in the literature.
The thermal evolution of lattice parameters coupled with heat capacity data provide insight into tailorable magnetism-structure attributes in the orthorhombic compound AlFe 2 B 2 that was synthesized with and without small additions of gallium. Temperature-dependent X-ray powder diffraction experiments conducted through the magnetic phase transition reveal that the a- and b-parameters of both samples increase with increasing temperature while the c-parameter decreases. While a weak volumetric thermal expansion is noted over a range of temperatures well below and above the magnetic phase transition, anomalous behavior was observed within the phase transition region itself to reveal a magnetostructural phase transition with borderline first-order character in the Ga-modified sample but of more second-order character in the Ga-free sample. It is established that the nearest-neighbor Fe-Fe interatomic distance within the (ab)-plane plays a dominant role in influencing the magneto-functional response of these compounds. The magnetocaloric properties are discussed in the context of temperature-induced changes of the interatomic bonding that are influenced by the hypothesized presence of iron antisite defects in the AlFe 2 B 2 lattice.
In this work, we developed an effective U-Net based deep learning (DL) model for inversion of surface gravity data on a rectangular grid to predict 2-D high-resolution subsurface CO 2 distribution along a vertical cross-section due to CO 2 leakage through a wellbore within a deep CO 2 storage reservoir. We used synthetic data to model two types of CO 2 leakage scenarios: one CO 2 plume in a shallow aquifer (single plume case), and two plumes present at different depths (double plume case). The 3-D synthetic plume samples were created by sampling among predetermined CO 2 plume depths, saturations, and volumes. The corresponding surface gravity data on a rectangular grid were generated by a 3-D forward model. The U-Net model detected 72% of single-plume samples, and one or both plumes in 75% of double-plume samples. Most of the undetected single plumes have small gravity field strengths below the typical noise level of 5 μGal. This model generated reproducible, reliable predictions with acceptable errors and demonstrated improved spatial resolution over the conventional least-squares inversion. In contrast to the conventional least-squares inversion, which often overestimates the size of its target and underestimates its density, this U-Net model accurately delineated the boundary of a target. Furthermore, this DL inversion detected deep, small, or low saturation CO 2 plumes that are often more difficult to resolve with conventional gravity inversion methods. We note the limitations of this feasibility study, including the use of synthetic data with regular CO 2 plume shapes, and the prediction of a 2-D plume cross-section rather than the full 3-D plume, as well, we recognize the lower detection fraction for double-plume scenarios. Nevertheless, this study demonstrates that DL gravity inversion is a promising and potentially superior method to conventional least-squares inversion. Our U-Net based deep learning inversion approach may be adapted for inversion of other types of geophysical data. DL inversion can facilitate near real-time monitoring of geologic carbon sequestration to provide site operators with prompt information about subsurface CO 2 distribution for risk management and mitigation.
In the last decade, the widespread adoption of electronic health record documentation has created huge opportunities for information mining. Natural language processing (NLP) techniques using machine and deep learning are becoming increasingly widespread for information extraction tasks from unstructured clinical notes. Disparities in performance when deploying machine learning models in the real world have recently received considerable attention. In the clinical NLP domain, the robustness of convolutional neural networks (CNNs) for classifying cancer pathology reports under natural distribution shifts remains understudied. In this research, we aim to quantify and improve the performance of the CNN for text classification on out-of-distribution (OOD) datasets resulting from the natural evolution of clinical text in pathology reports. We identified class imbalance due to different prevalence of cancer types as one of the sources of performance drop and analyzed the impact of previous methods for addressing class imbalance when deploying models in real-world domains. Our results show that our novel class-specialized ensemble technique outperforms other methods for the classification of rare cancer types in terms of macro F1 scores. We also found that traditional ensemble methods perform better in top classes, leading to higher micro F1 scores. Based on our findings, we formulate a series of recommendations for other ML practitioners on how to build robust models with extremely imbalanced datasets in biomedical NLP applications.
Objective: Electronic health record (EHR) systems contain a wealth of clinical data stored as both codified data and free-text narrative notes (NLP). The complexity of EHR presents challenges in feature representation, information extraction, and uncertainty quantification. Here, to address these challenges, we proposed an efficient Aggregated naRrative Codified Health (ARCH) records analysis to generate a large-scale knowledge graph (KG) for a comprehensive set of EHR codified and narrative features. Methods: Using data from 12.5 million Veterans Affairs patients, ARCH first derives embedding vectors and generates similarities along with associated p-values to measure the strength of relatedness between clinical features with statistical certainty quantification. Next, ARCH performs a sparse embedding regression to remove indirect linkage between features to build a sparse KG. Finally, ARCH was validated on various clinical tasks, including detecting known relationships between entity pairs, predicting drug side effects, disease phenotyping, as well as sub-typing Alzheimer’s disease patients. Results: ARCH produces high-quality clinical embeddings and KG for over 60,000 codified and narrative EHR concepts. The KG and embeddings are visualized in the R-shiny powered web-API.3 ARCH achieved high accuracy in detecting EHR concept relationships, with AUCs of 0.926 (codified) and 0.861 (NLP) for similar EHR concepts, and 0.810 (codified) and 0.843 (NLP) for related pairs. It detected drug side effects with a 0.723 AUC, which improved to 0.826 after fine-tuning. Using both codified and NLP features, the detection power increased significantly. Compared to other methods, ARCH has superior accuracy and enhances weakly supervised phenotyping algorithms’ performance. Notably, it successfully categorized Alzheimer’s patients into two subgroups with varying mortality rates. Conclusion: The proposed ARCH algorithm generates large-scale high-quality semantic representations and knowledge graph for both codified and NLP EHR features, useful for a wide range of predictive modeling tasks.
In upscaling methods, closures for nonlinear problems present a well-known challenge. While a number of theoretical methods have been proposed for handling such closures, nonlinearities still remain a significant obstacle for many problems. In this work, we use a combination of formal upscaling and data-driven machine learning for explicitly closing a nonlinear transport and reaction process in multiscale tissues. The classical effectiveness factor model is used to formulate the macroscale reaction kinetics. We train a multilayer perceptron network using training data generated by direct numerical simulations over microscale examples. Once trained, the network is used in an algorithm for numerically solving the upscaled (coarse-grained) differential equation describing mass transport and reaction in two example tissues. The network is described as being explicit in the sense that the network is trained using macroscale concentrations and gradients of concentration as components of the feature space rather than incorporating them as part of a constraint in the optimization process. Network training and solutions to the macroscale transport equations were computed for two different tissues. The two tissue types (brain and liver) exhibit markedly different geometrical complexity and spatial scale (cell size and sample size). The upscaled solutions for the average concentration are compared with numerical solutions derived from the microscale concentration fields by a posteriori averaging. There are three outcomes of this work of particular note. 1) Our overall approach results in an upscaled nonlinear PDE. The PDE is closed using a neural network, and our approach results in the definition of the classical effectiveness factor for effecting closure. 2) We identify particular source terms for the closure problem that are important for representing the structure of the closure. These source terms involve macroscale concentrations and their gradients. We adopt these source terms to use as explicit features in the learning algorithm. We find the trained networks that include the macroscale source terms generate models that are able to predict the correction factor with increased fidelity over those that do not. 3) We find that the trained network exhibits good generalizability, and it is able to predict the effectiveness factor with high fidelity for realistically-structured tissues despite the significantly different scale and geometrical complexity of the two example tissue types. This latter result emphasizes our purposeful connection between conventional averaging methods with the use of machine learning for closure; this contrasts with some machine learning methods for upscaling where the exact form of the macroscale equation remains unknown.
Quantum optimal control plays a crucial role in quantum computing by providing the interface between compiler and hardware. Solving the optimal control problem is particularly challenging for multi-qubit gates, due to the exponential growth in computational complexity with the system's dimensionality and the deterioration of optimization convergence. To ameliorate the computational complexity of time-integration, this paper introduces a multiple-shooting approach in which the time domain is divided into multiple windows and the intermediate states at window boundaries are treated as additional optimization variables. Further, this enables parallel computation of state evolution across time-windows, significantly accelerating objective function and gradient evaluations. Since the initial state matrix in each window is only guaranteed to be unitary upon convergence of the optimization algorithm, the conventional gate trace infidelity is replaced by a generalized infidelity that is convex for non-unitary state matrices. Continuity of the state across window boundaries is enforced by equality constraints. A quadratic penalty optimization method is used to solve the constrained optimal control problem, and an efficient adjoint technique is employed to calculate the gradients in each iteration. We demonstrate the effectiveness of the proposed method through numerical experiments on quantum Fourier transform gates in systems with 2, 3, and 4 qubits, noting a speedup of 80x for evaluating the gradient in the 4-qubit case, highlighting the method's potential for optimizing control pulses in multi-qubit quantum systems.
Prediction of brittle fracture in silicate glasses requires a fundamental understanding of slow crack growth rates, far below the critical stress intensity (K IC ) of the material. Here, in this study, atomic force microscopy (AFM) was used to characterize slow crack propagation in soda lime silicate glasses. Using large-sample AFM systems with specialized tips and scanners, fractures less than 20 nm wide were measured as a function of relative humidity (RH). The results were found to be sensitive to the type of AFM tip used and to the dryness of the surface. Continuous scanning and image capture over several days allowed the determination of the crack tip velocity. The dependance of crack tip velocity on RH was measured by enclosing the sample and AFM head in a humidity-controlled chamber. The crack tip velocity was measured as low 4.0×10 -12 m/s at 15 % RH and as high as 3.7×10 -10 m/s at 40 % RH. Overall crack growth was smooth during the AFM scan. Additionally, recession of the crack tip was noted as the sample dried, indicating crack tip closure may be occurring. Further studies on alternative compositions will provide details on subcritical fracture processes in a range of different environmental conditions.
Ferritic-Martensitic steel welds typically require post weld heat treatment (PWHT) to restore toughness and high temperature performance. This off-line thermal process reduces disparities between weld and base metal, but can cause distortion, cracking, or simply be impractical due to assembly size and joint non-uniformity. Here we show integrated welding and thermal processing applied to modified 9Cr-1Mo (Grade 91) steel, favored for advanced power generation applications, performed in real time through the addition of a secondary heat source near the primary weld head. Optimal integrated processing reduces weld fusion and heat affected zone hardness by 125 HV, approaching performance of conventional 730 °C, 60 min PWHT processing. Microstructures and mechanical performance are compared for mechanized GTAW welds, with equivalent lifetimes noted in cross-weld creep rupture tests up to 234 MPa at 550 °C, and up to 104 MPa at 650 °C. The integrated process was validated on a Grade 91 pressure vessel with multipass cold wire feed GTAW. After 550 °C, 71.4 bar thermomechanical cyclic testing, the maximum weld hardness is <350 HV.
Despite decades of research, connecting the chemical and physical structure of thermoplastic polyurethanes to their mechanical properties remains highly challenging. Of particular note are their large-deformation and rate-dependent behaviors, which vary greatly with molecular chemistry, including the type and relative content of soft and hard segments. In this work, we develop a physically motivated mechanical theory for predicting the behavior of thermoplastic polyurethanes. The theory incorporates a representation of microstructural evolution during mechanical deformation, which captures the signatures of stress softening over cyclic loading (commonly referred to as the Mullins effect). There are only eight physically motivated fitting parameters, including a direct dependence on the hard segment fraction. The model predicts that increasing the hard segment fraction leads to higher stiffness and greater energy dissipation, in quantitative agreement with published experimental data. Furthermore, we provide a comprehensive analysis of the model and validate its predictions across several independent datasets focused on mechanical characterization. Direct comparisons to experimental data demonstrate its predictive capability on the effect of loading rate, cyclic deformations, and applied tension or compression. Altogether, this work establishes a predictive framework that connects polymer chemistry and microstructure to emergent mechanical behaviors.
Chromium-doped UO is a widely-studied near-term deployable accident-tolerant fuel concept because it results in a dense, large-grain structure that increases the fuel resistance to densification, swelling, and fission gas release. A new charged-interstitial mechanism was recently proposed to describe the behavior of dopants like chromium in sintered UO 2 . Based on that mechanism, manganese was suggested as an even stronger dopant than chromium. In the current work we use mesoscale sintering simulations in an effort to validate the new mechanism. We compare the relative behavior of Cr-doped and undoped UO 2 against experimental data in the literature. We also make predictions of the relative behavior of Mn-doped UO 2 . Simulations are done using the phase field-based grand potential sintering model. Dopants have two effects on sintered UO 2 . They marginally increase the densification rate and greatly increase the average grain size. Both of these effects are individually tested in small-scale simulations. Then large 3D sintering simulations are used to test the combined behavior of both effects. The results for the densification rate simulations are consistent with experiments for Cr-doped fuels. However, we note the grain growth rates are lower than what is found in the literature for Cr-doped fuels.
In light water reactor (LWR) UO 2 fuels, the evolution of volatile fission products is one of the critical areas of fuel behavior that is yet to be fully understood. In UO 2 irradiated to high burnups, it is well known that most released fission gases come from the central region of the fuel as opposed to the highly porous high burnup structure (HBS) on the periphery of the pellets. However, fuels with and without interconnected bubble networks at the fuel center showed high to moderate release fractions, which conceals the mechanisms responsible for the gas release in the latter scenario. In this work, focused ion beam tomography was used to investigate the three-dimensional bubble structure in an irradiated LWR UO 2 fuel pellet with high degree of fission gas retention so that the degree of bubble interconnection could be assessed. Six radial locations with different burnups and temperatures were serially sectioned and imaged to reconstruct the three-dimensional bubble structure. As expected, the highest porosity was observed at the periphery of the fuel (HBS). Here, the porosity then decreased towards the pellet center, except for the centermost location. This location had a slightly higher porosity than its adjacent mid-radial location, which was attributed to the temperature difference between the two locations. This study provides a first-time volumetric evaluation of the porosity at different radial locations on a UO 2 fuel pellet. During this investigation, no significant bubble interconnection was noted at any of the six radial locations.
Oxide dispersion strengthened (ODS) FeCrAl alloys are promising candidate materials for advanced nuclear reactor applications requiring high-temperature strength, corrosion resistance, and irradiation tolerance. As these alloys have increased in compositional complexity through attempts to use highly reactive elements such as Zr to refine particle sizes and optimize nanoprecipitate dispersion characteristics, much debate has ensued as to the effects of these alloying element additions on alloy properties. In an attempt to reconcile differences in nanoprecipitate distributions reported in the literature over the past decade, a detailed investigation of a recently developed ODS FeCrAl alloy with nominal composition Fe–10Cr-6.1Al-0.3Zr+0.3Y 2 O 3 is presented here using a combination of atom probe tomography (APT), scanning/transmission electron microscopy (S/TEM), and computational thermodynamics modeling. It is illustrated that based on the amount of Zr available in the lattice, Zr competes with Al and Cr to form carbides and nitrides as opposed to oxygen-rich precipitates. This alloy system has a high number density (>10 23 m -3 ) of ~2–4 nm diameter (Y,Al,O)-rich nanoprecipitates, but it is shown that due to the compositional spread and unknown partitioning of Al between the matrix and precipitates, significant challenges still exist for quantifying the exact compositions of these precipitates using APT. However, the noted compositional spread is supported by identified complex oxides yttrium aluminum monoclinic (YAM) and yttrium aluminum garnet (YAG) using S/TEM. As a result of these findings, researchers developing ODS FeCrAl with reactive element additions must pay careful attention to C and N impurities when optimizing reactive element additions.
This contribution investigates fuel constituent segregation and fuel-cladding chemical interactions (FCCI) in three U-Pu-Zr fuel pins irradiated to ~11% burnup in Experimental Breeder Reactor-II as part of the X441 DP-1 experiment. In examined pins, Zr content ranged from 6 to 10 and 14 wt.%, while Pu concentration was constant at 19 wt.% Pu. The primary goal of the investigation was to determine the role Zr content plays in fuel performance and this manuscript provides assessment of both constituent redistribution and FCCI as a function of axial position. FCCI was observed in all pins, though no cladding failure was noted. Optical and scanning electron microscopy (SEM) results show that variation in Zr content alters the number and relative size of constituent redistribution zones in the fuel. For example, larger Zr-rich central regions and smaller U-rich intermediate regions were observed in 14 wt.% Zr fuels. More importantly, all examined fuels have four or more discrete constituent redistribution regions, where each region has dissimilar morphological features. The existence of four to six distinctive regions deviates from traditionally accepted three region redistribution model, highlights complexity of constituent redistribution in this metal fuel, and identifies the need to conduct additional studies using state-of-the-art instrumentation.
We report an outstanding challenge in the manufacturing and joining of oxide dispersion strengthened steels is retaining the nanofeatures in the alloy throughout the fabrication and welding process. MA956 was friction stir welded with two different sets of welding parameters, resulting in a medium and high heat input. After welding, 5 MeV Fe ++ ion irradiations were performed at doses ranging from 50 to 200 dpa in the temperature range of 400 to 500°C. Post-irradiation characterization was performed with scanning transmission electron microscopy and energy-dispersive x-ray spectroscopy to investigate the Y-Al-O dispersoids, voids, and dislocations. After welding, the dispersoid microstructure coarsened, resulting in fewer and larger dispersoids regardless of heat input. After irradiation, the dispersoid behavior in the welded material was sensitive to temperature, exhibiting growth behavior attributed to Ostwald coarsening at 500 °C but a mixture of nucleation and more muted growth at 400 and 450 °C, attributed to competing mechanisms of radiation-enhanced diffusion and Ostwald coarsening. Void swelling correlated to heat input; being more prevalent in the welded conditions occurring at lower doses and in higher values relative to the base material. The low values of swelling despite microstructure coarsening caused by welding demonstrate the excellent swelling resistance of MA956, even after welding with the highest swelling values of 0.5% noted in the stir zone high heat input condition at 450 °C, 200 dpa. The dislocation behavior was inconsistent: the strongest trend was that network density was higher for welded versus base material, and an increase in loop diameter with temperature was observed. A rate theory model based on the observed microstructure suggests at high temperature interstitial loss to sinks was more likely to be dominant compared to mutual annihilation via point defect recombination, because of an increase of the radiation diffusion coefficient with temperature regardless of initial welded microstructure.
Density functional theory calculations are performed to explore the binding between He and alloying solutes, impurities, and transmutation products expected in neutron irradiated nanostructured ferritic alloys (NFAs), here 14YWT is taken as an example. Elements that exhibit significant binding (attraction) with an interstitial He are Y (binding energy = 0.46 eV), Mg (0.32), O (0.33), Ti (0.16), and C (0.15). Those that provide significant binding to a substitutional He are O (1.44), Y (1.24), N (0.73), H (0.56), Mg (0.52), Ti (0.34), Si (0.34), C (0.33), Al (0.32), Ni (0.26), Ta (0.23), and Mn (0.16). The presence of these elements in Fe matrix could reduce the transport of He towards oxide particles, dislocations, and internal boundaries, and could promote He bubble nucleation in the matrix. For convenience, we compile existing binding energy data of He with He n and He n V (He-vacancy) clusters. Dissociation pathway analysis reveals that, in general, the most likely dissociation of a He n V cluster is by a sequential emission of individual He atoms. Furthermore, larger bubbles are more prone to dissociation than smaller ones. In addition, we estimate the binding energy (segregation energy) of He in bulk Y 2 Ti 2 O 7 (YTO) single crystal, YTO/Fe interface, and YTO particle embedded in Fe, with respect to interstitial He in Fe, from existing formation energies of He in these structures. We also compile available data of He binding with Fe self-interstitial atom (SIA), SIA clusters, and edge and screw dislocations. Note that given the absence of DFT data, the binding with SIA clusters and dislocations are gathered from simulations with empirical potentials. Finally, the data presented in this paper is important to inform multiscale simulations of He bubble accumulation.
We report the joining process for oxide dispersion strengthened (ODS) alloys remains a key challenge facing the nuclear community. The microstructure and mechanical properties were characterized in the base material and friction stir welded ODS MA956 irradiated with 5 MeV Fe 2+ ions from 400 to 500°C up to 25 dpa. Nanoindentation was performed to assess changes in hardness and yield stress, and the dispersed barrier hardening (DBH) model was applied to described results. A combination of scanning transmission electron microscopy (STEM) and atom probe tomography (APT) were used to assess evolution of the microstructure including dispersoids, network dislocations and dislocation loops, nanoclusters, and solid solution concentrations. Overall, softening was observed as a result of increased dose, which was exacerbated at 500°C. The formation and coarsening of new dispersoids was noted while nanoclusters tended to dissolve in the base material, and were not observed in the stir zone. Solute nanocluster evolution was identified as a primary driver of the changes in mechanical properties.