Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “interpretability”

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.

At least 1,189 records · Page 66
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Analysis of Interpretable Data Representations for 4D-STEM Using Unsupervised Learning

Abstract Understanding the structure of materials is crucial for engineering devices and materials with enhanced performance. Four-dimensional scanning transmission electron microscopy (4D-STEM) is capable of mapping nanometer-scale local crystallographic structure over micron-scale field of views. However, 4D-STEM datasets can contain tens of thousands of images from a wide variety of material structures, making it difficult to automate detection and classification of structures. Traditional automated analysis pipelines for 4D-STEM focus on supervised approaches, which require prior knowledge of the material structure and cannot describe anomalous or deviant structures. In this article, a pipeline for engineering 4D-STEM feature representations for unsupervised clustering using non-negative matrix factorization (NMF) is introduced. Each feature is evaluated using NMF and results are presented for both simulated and experimental data. It is shown that some data representations more reliably identify overlapping grains. Additionally, real space refinement is applied to identify spatially distinct sample regions, allowing for size and shape analysis to be performed. This work lays the foundation for improved analysis of nanoscale structural features in materials that deviate from expected crystallographic arrangement using 4D-STEM.

Bruefach, Alexandra (ORCID:0000000209323477)↗
DOE OSTI2019

Human Glycerol 3-Phosphate Dehydrogenase: X-ray Crystal Structures That Guide the Interpretation of Mutagenesis Studies

Human liver glycerol 3-phosphate dehydrogenase (hlGPDH) catalyzes the reduction of dihydroxyacetone phosphate (DHAP) to form glycerol 3-phosphate, using the binding energy associated with the nonreacting phosphodianion of the substrate to properly orient the enzyme–substrate complex within the active site. Herein, we report the crystal structures for unliganded, binary E·NAD, and ternary E·NAD·DHAP complexes of wild type hlGPDH, illustrating a new position of DHAP, and probe the kinetics of multiple mutant enzymes with natural and truncated substrates. Mutation of Lys120, which is positioned to donate a proton to the carbonyl of DHAP, results in similar increases in the activation barrier to hlGPDH-catlyzed reduction of DHAP and to phosphite dianion-activated reduction of glycolaldehyde, illustrating that these transition states show similar interactions with the cationic K120 side chain. The K120A mutation results in a 5.3 kcal/mol transition state destabilization, and 3.0 kcal/mol of the lost transition state stabilization is rescued by 1.0 M ethylammonium cation. The 6.5 kcal/mol increase in the activation barrier observed for the D260G mutant hlGPDH-catalyzed reaction represents a 3.5 kcal/mol weakening of transition state stabilization by the K120A side chain and a 3.0 kcal/mol weakening of the interactions with other residues. The interactions, at the enzyme active site, between the K120 side chain and the Q295 and R269 side chains were likewise examined by double-mutant analyses. Here, these results provide strong evidence that the enzyme rate acceleration is due mainly or exclusively to transition state stabilization by electrostatic interactions with polar amino acid side chains.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Probabilistic Deep Learning Approach to Automate the Interpretation of Multi-phase Diffraction Spectra

Autonomous synthesis and characterization of inorganic materials requires the automatic and accurate analysis of X-ray diffraction spectra. For this task, we designed a probabilistic deep learning algorithm to identify complex multi-phase mixtures. At the core of this algorithm lies an ensemble convolutional neural network trained on simulated diffraction spectra, which are systematically augmented with physics-informed perturbations to account for artifacts that can arise during experimental sample preparation and synthesis. In this work, larger perturbations associated with off-stoichiometry are also captured by supplementing the training set with hypothetical solid solutions. Spectra containing mixtures of materials are analyzed with a newly developed branching algorithm that utilizes the probabilistic nature of the neural network to explore suspected mixtures and identify the set of phases that maximize confidence in the prediction. Our model is benchmarked on simulated and experimentally measured diffraction spectra, showing exceptional performance with accuracies exceeding those given by previously reported methods based on profile matching and deep learning. We envision that the algorithm presented here may be integrated in experimental workflows to facilitate the high-throughput and autonomous discovery of inorganic materials.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Mechanistic Interpretations and Insights for the Oxidative Dehydrogenation of Propane via CO 2 over Cr 2 O 3 /Al 2 O 3 Catalysts

Oxidative dehydrogenation (ODH) of alkanes using carbon dioxide as a soft oxidant has recently emerged as a potentially attractive alternative to steam cracking for the production of light olefins. To elucidate reaction pathways and their dependence on the operating conditions, CO 2 -assisted propane dehydrogenation over a redox-active Cr 2 O 3 /Al 2 O 3 catalyst was examined in a packed bed reactor as a function of temperature, Cr 2 O 3 /CO 2 feed ratio, and residence time. Previous ODH studies have largely focused on CO 2 -rich conditions with the aim of preventing coke formation. However, at T = 600 °C the present study finds that the use of propane-rich conditions (1 ≤ C 3 H 8 /CO 2 ≤ 2.5) maximizes propylene production and selectivity while maintaining catalyst stability. In this work, it is postulated that the selective Mars van Krevelen dehydrogenation process is optimized at these ratios. Excess CO 2 apparently promotes nonselective dehydrogenation and dry reforming pathways that generate additional CO, adversely impacting catalyst stability via the Bouduard reaction. This hypothesis is supported by complementary investigations of the reverse water gas shift reaction and thermodynamic analysis. The findings and methodology presented here are likely applicable to related ODH processes with other alkanes and redox-active catalysts.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Toward a Machine Learning Approach to Interpreting X-ray Spectra of Trace Impurities by Converting XANES to EXAFS

The fact that the photoabsorption spectrum of a material contains information about the atomic structure, commonly understood in terms of multiple scattering theory, is the basis of the popular extended X-ray absorption spectroscopy (EXAFS) technique. How much of the same structural information is present in other complementary spectroscopic signals is not obvious. Here we use a machine learning approach to demonstrate that within theoretical models that accurately predict the EXAFS signal, the extended near-edge region does indeed contain the EXAFS-accessible structural information. We do this by exhibiting deep operator neural networks (DeepONets) that have learned the relationship between the extended and near edge portions of the X-ray absorption spectrum to predict the former from the latter. We find that we can accurately predict the EXAFS spectrum between 6 and 14 Å –1 from the first 6 Å –1 (≈100 eV) of the absorption spectrum of Cu 2 + substitutional defects in the Fe 3+ mineral hematite (α-Fe 2 O 3 ). This surprising finding implies that theoretical analyses of X-ray absorption spectra could be implemented that extract the same conclusions as high-quality EXAFS studies from spectra collected over a much smaller range of photon energies. This relaxes a host of experimental limitations related to the X-ray source and measurement sample, including collection time, minimum dopant concentration, source brilliance, and energy range. We describe the theoretical data sets and DeepONet construction and show that the resulting DeepONets produce EXAFS that recovers linear combination fits to experimental data with accuracy approaching the original ab initio calculations. We discuss the implications of our findings for minor constituent characterization and for understanding the information content of spectroscopic data more broadly, including how this approach might be applied to measured experimental spectra. In conclusion, to encourage similar efforts, the simulated X-ray spectra, machine learning, and fitting code are publicly available.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2016

X-ray Emission Spectroscopy of Mn Coordination Complexes Toward Interpreting the Electronic Structure of the Oxygen-Evolving Complex of Photosystem II

X-ray emission (XES) spectroscopy is an attractive technique for analysis of the electronic structure of molecules, materials, and metalloproteins. However, a better understanding of XES results is required. Using a combination of experiment and ground-state density functional theory analysis, we rationalize differences in the X-ray emission spectra of multinuclear Mn complexes. Model compounds, including dinuclear [Mn 2 O 2 L' 4 ](ClO 4 ) 3 (L'= 2,2'-bipyridyl, [1]) and two examples from the Mn 4 O 4 L 6 “cubane” family of model compounds (L = (p-R-C 6 H 4 )PO 2 – , R = OCH 3 [2], CH3 [3] ), were compared with the Oxygen Evolving Complex of Photosystem II. Our analysis demonstrates that changes in the structure of the Mn complexes, resulting in changes to the spin polarization, can introduce significant spectral shifts in compounds of the same formal redox state. The implications of changes in spin polarization for understanding photosynthetic water-splitting catalysis is discussed.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2018

Interpreting Neutron Reflectivity Profiles of Diblock Copolymer Nanocomposite Thin Films Using Hybrid Particle-Field Simulations

Mixtures of block copolymers and nanoparticles (block copolymer nanocomposites) are known to microphase separate into a plethora of microstructures, depending on the composition, length scale and nature of interactions among its different constituents. Theoretical and experimental works on this class of nanocomposites have already high-lighted intricate relations among chemical details of the polymers, nanoparticles, and various microstructures. Confining these nanocomposites in thin films yields an even larger array of structures, which are not normally observed in the bulk. In contrast to the bulk, exploring various microstructures in thin films by the experimental route remains a challenging task. Here in this work, we construct a model for the thin films of lamellar forming diblock copolymers containing spherical nanoparticles based on a hybrid particle-field approach. The model is benchmarked by comparison with the depth profiles obtained from the neutron reflectivity experiments for symmetric poly(deuterated styrene-b-n butyl methacrylate) copolymers blended with spherical magnetite nanoparticles covered with hydrogenated poly(styrene) corona. We show that the model based on a hybrid particle-field approach provides details of the underlying microphase separation in the presence of the nanoparticles through a direct comparison to the neutron reflectivity data. This work benchmarks the application of the hybrid particle-field model to extract the interaction parameters for exploring different microstructures in thin films containing block copolymers and nanocomposites.

36 MATERIALS SCIENCE↗