Search NASA⌕ Search

SEARCH · Search NASA

Results for “Program Synthesis”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Snowmass Computational Frontier: Topical Group Report on Quantum Computing

Quantum computing will play a pivotal role in the High Energy Physics (HEP) science program over the early parts of the 21$^{st}$ Century, both as a major expansion of our capabilities across the Computational Frontier, and in synthesis with quantum sensing and quantum networks. This report outlines how Quantum Information Science (QIS) and HEP are deeply intertwined endeavors that benefit enormously from a strong engagement together. Quantum computers do not represent a detour for HEP, rather they are set to become an integral part of our discovery toolkit. Problems ranging from simulating quantum field theories, to fully leveraging the most sensitive sensor suites for new particle searches, and even data analysis will run into limiting bottlenecks if constrained to our current computing paradigms. Easy access to quantum computers is needed to build a deeper understanding of these opportunities. In turn, HEP brings crucial expertise to the national quantum ecosystem in quantum domain knowledge, superconducting technology, cryogenic and fast microelectronics, and massive-scale project management. The role of quantum technologies across the entire economy is expected to grow rapidly over the next decade, so it is important to establish the role of HEP in the efforts surrounding QIS. Fully delivering on the promise of quantum technologies in the HEP science program requires robust support. It is important to both invest in the co-design opportunities afforded by the broader quantum computing ecosystem and leverage HEP strengths with the goal of designing quantum computers tailored to HEP science.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

The Data Mine model for accessible partnerships in data science

Abstract The Data Mine at Purdue University is a pioneering experiential learning community for undergraduate and graduate students of any background to learn data science. The first data‐intensive experience embedded in a large learning community, The Data Mine had nearly 1300 students in academic year (AY) 2022–2023 and nearly 1700 students for AY 2023–2024. The Data Mine embodies data‐infused education, research, and collaboration. Students learn Python, R, SQL, and shell‐scripting, while working on weekly projects within a high‐performance computing (HPC) cluster. In the Corporate Partners cohort, students work on teams of 5–15 students, led by a paid student team leader. Each cohort follows an Agile approach, working on data‐intensive projects provided by industry partners and mentored by company employees. Students develop professional and data skills throughout the academic year, from August through April. Many students return in subsequent years to the program, increasing their tenure with a Corporate Partner. Student teams are inherently interdisciplinary; students from 133 different majors are involved in the program, ranging from new incoming students through PhD level students. These interdisciplinary teams of students bring new perspectives to challenging problems in which data science is a key part of the solution. The interdisciplinary teams foster an environment of synthesis with ideas and solutions. Students come together with different life experiences, different levels of technical skill, but also varying ways they navigate paths to solutions because of the variety of majors represented, resulting in a more creative and robust solution than a traditional data science program. This article is categorized under: Applications of Computational Statistics > Education in Computational Statistics

Betz, Margaret A.↗
DOE OSTI2019

Fischer-Tropsch Synthesis: Cd, In and Sn Effects on a 15%Co/Al 2 O 3 Catalyst

The effects of 1% of Cd, In and Sn additives on the physicochemical properties and Fischer-Tropsch synthesis (FTS) performance of a 15% Co/Al 2 O 3 catalyst were investigated. The fresh and spent catalysts were characterized by BET, temperature programmed reduction (TPR), H 2 -chemisorption, NH 3 temperature programmed desorption (TPD), X-ray absorption near edge spectroscopy (XANES), and X ray diffraction (XRD). The catalysts were tested in a 1 L continuously stirred tank reactor (CSTR) at 220 °C, 2.2 MPa, H 2 /CO = 2.1 and 20–55% CO conversion. Addition of 1% of Cd or In enhanced the reduction degree of 15%Co/Al 2 O 3 by ~20%, while addition of 1% Sn slightly hindered it. All three additives adversely impacted Co dispersion by 22–32% by increasing apparent Co cluster size based on the H 2 -chemisorption measurements. However, the decreased Co active site density resulting from the additives did not result in a corresponding activity loss; instead, the additives decreased the activity of the Co catalysts to a much greater extent than expected, i.e., 82–93%. The additional detrimental effect on catalyst activity likely indicates that the Cd, In and Sn additives migrated to and covered active sites during reaction and/or provided an electronic effect. XANES results showed that oxides of the additives were present during the reaction, but that a fraction of metal was also likely present based on the TPR and reaction testing results. This is in contrast to typical promoters that become metallic at or below ~350 °C, such as noble metal promoters (e.g., Pt, Ru) and Group 11 promoters (e.g., Ag, Au) on Co catalysts in earlier studies. In the current work, all three additives remarkably increased CH 4 and CO 2 selectivities and decreased C 5+ selectivity, with the Sn and In additives having a greater effect. Interestingly, the Cd, In, or Sn additives were found to influence hydrogenation and isomerization activities. At a similar conversion level (i.e., in the range of 40–50%), the additives significantly increased 2-C 4 olefin content from 3.8 to 10.6% and n-C 4 paraffin from 50 to 61% accompanied by decreases in 1-C 4 olefin content from 48 to 30%. The Sn contributed the greatest impact on the secondary reactions of 1-olefins, followed by the In and Cd. NH 3 -TPD results suggest enhanced acid sites on cobalt catalysts resulting from the additives, which likely explains the change in selectivities for the different catalysts.

36 MATERIALS SCIENCE↗
DOE OSTI2023

At the interfaces of the hydrologic sciences: Connecting water, elements, ecosystems, and people through the major contributions of Dr. Emily Bernhardt

Here, in this paper, we describe the major contributions of Professor Emily Bernhardt to the hydrologic sciences. Dr. Bernhardt’s work addresses how carbon, nutrient, and contaminant dynamics respond to a wide range of environmental perturbations that alter hydrologic dynamics within and connectivity among ecosystems. Her research leverages intensive and extensive field sampling, experimental manipulations, macroscale data harmonization and exploration, and continental to global-scale synthesis activities to uncover key drivers and patterns of the impacts human perturbations have on water and elemental cycles. Dr. Bernhardt’s research program is defined by her ability to ask questions and use approaches that explicitly consider connectivity and interfaces in a variety of ways. Here, we highlight significant contributions from Dr. Bernhardt’s work, organized by connectivity, interfaces, and interactions among and across (1) elemental cycles, (2) ecosystems, (3) watersheds, (4) scales, and (5) disciplines. We conclude with a section on Dr. Bernhardt’s impact on the hydrologic sciences and beyond through her exceptional dedication to mentorship, engagement, and service.

58 GEOSCIENCES↗
DOE OSTI2019

Stable White Organic Light-emitting Diodes Enabled by New Materials with Reduced Excited State Lifetime (Final Report)

The Georgia Institute of Technology assembled a multidisciplinary team with complementary expertise in the areas of multi-level molecular simulations (Brédas), chemical synthesis and material characterization (Marder), and device physics and engineering (Kippelen). The team engaged in a two-year integrated theoretical and experimental program to develop blue-emissive layers that can yield simultaneously near 100% internal quantum efficiency (IQE) and short excited-state lifetimes to overcome the existing efficiency vs. stability tradeoff displayed by blue-emitting OLEDs and by WOLEDs, and to demonstrate stable WOLEDs that fulfill or exceed the 2020 target metrics (RDP 2016 section C.1.2) ahead of schedule. In this program, we investigated two general approaches involving the development of wide-bandgap ambipolar hosts with a low ΔEST. Ambipolar host achieved by intermolecular through-space D:A interactions and by intramolecular through-bond interactions between covalently linked D and A segments. Novel hosts were incorporated in to blue-emissive layers (EML) by combining them with suitable fluorescent guest emitters having high quantum yields of emission (Φem) to demonstrate blue-emitting OLEDs with high EQE and low roll-off. This final report describes the most important results achieved during this program

36 MATERIALS SCIENCE↗
DOE OSTI2021

Development of Passive HC/NOx Trap Catalysts for Low Temperature Gasoline Applications

This project aimed to develop fundamental understanding of the chemistry of NO adsorption and reaction in Pd/zeolites so as to facilitate the rational design of passive NOx adsorber catalysts. The approach adopted combined both experimental and computational methods, which together allow a deeper understanding of the governing chemistry than the use of either method alone. The workflow began with Pd/H-CHA and Pd/H-BEA catalyst synthesis and characterization, in which the Si/Al ratio and Al siting were systematically varied. This was followed by catalyst evaluation using temperature-programed adsorption/desorption methods, as well as in situ spectroscopic measurements to probe the chemistry of NO adsorption. In parallel, the adsorption of NO and other relevant species (H 2 O, CO, HCs) was studied by means of quantum chemical calculations in order to rationalize the experimental data and provide additional insights. Catalyst aging studies were also performed with the aim of elucidating the mechanism of catalyst degradation. Finally, the insights gained in this project were applied to the preparation of an optimized HC/NOx adsorber catalyst, the performance of which was studied using exhaust gas from an engine dynamometer.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Generalized optimization-based synthesis of membrane systems for multicomponent gas mixture separation

Synthesizing a membrane system to separate multicomponent gas mixture is challenging due to the combinatorial number of feasible configurations and the difficulties in describing the multicomponent permeators. Here we present a mixed-integer nonlinear programming (MINLP) model for synthesizing membrane systems for multicomponent gas mixture separation. The approach employs a richly connected superstructure to represent numerous potential system configurations, and different physics-based surrogate permeator models, such as countercurrent flow or crossflow, to be used in each stage. Moreover, to describe realistic systems, pressure drop equations can be included. We also present solution methods to accelerate the solution process. Through a case study of natural gas sweetening, we demonstrate that the proposed approach is able to obtain good solutions using an off-the-shelf global optimization solver. Finally, we expand the conventional membrane system synthesis problem by introducing feed variability in our model through a case study of an integrated reactor-separation system.

42 ENGINEERING↗
DOE OSTI2020

Programming Amphiphilic Peptoid Oligomers for Hierarchical Assembly and Inorganic Crystallization

Natural organisms make a wide variety of exquisitely complex, nano-, micro-, and macroscale structured materials in an energy-efficient and highly reproducible manner. During these processes, the information-carrying biomolecules (e.g., proteins, peptides, and carbohydrates) enable (1) hierarchical organization to assemble scaffold materials and execute high-level functions and (2) exquisite control over inorganic materials synthesis, generating biominerals whose properties are optimized for their functions. Inspired by nature, significant efforts have been devoted to developing functional materials that can rival those natural molecules by mimicking in vivo functions using engineered proteins, peptides, DNAs, sequence-defined synthetic molecules (e.g., peptoids), and other biomimetic polymers. Among them, peptoids, a new type of synthetic mimetics of peptides and proteins, have received particular attention because they combine the merits of both synthetic polymers (e.g., high chemical stability and efficient synthesis) and biomolecules (e.g., sequence programmability and biocompatibility). The lack of both chirality and hydrogen bonds in their backbone results in a highly designable peptoid-based system with reduced structural complexity and side chain-chemistry-dominated properties. Here in this Account, we present our recent efforts in this field by programming amphiphilic peptoid sequences for (1) the controlled self-assembly into different hierarchically structured nanomaterials with favorable properties and (2) manipulating inorganic (nano)crystal nucleation, growth, and assembly into superstructures. First, we designed a series of amphiphilic peptoids with controlled side chain chemistries that self-assembled into 1D highly stiff and dynamic nanotubes, 2D membrane-mimetic nanosheets, hexagonally patterned nanoribbons, and 3D nanoflowers. These crystalline nanostructures exhibited sequence-dependent properties and showed promise for different applications. The corresponding peptoid self-assembly pathways and mechanisms were also investigated by leveraging in situ atomic force microscopy studies and molecular dynamics simulations, which showed precise sequence dependency. Second, inspired by peptide- and protein-controlled formation of hierarchical inorganic nanostructures in nature, we developed peptoid-based biomimetic approaches for controlled synthesis of inorganic materials (e.g., noble metals and calcite), in which we took advantage of the substantial side chain chemistry of peptoids and investigated the relationship between the peptoid sequences and the morphology and growth kinetics of inorganic materials. For example, to overcome the challenges (e.g., complexity of protein- and peptide-folding, poor thermal and chemical stabilities) facing the area of protein- and peptide-controlled synthesis of inorganic materials, we recently reported the design of sequence-defined peptoids for controlled synthesis of highly branched plasmonic gold particles. Moreover, we developed a rule of thumb for designing peptoids that predictively enabled the morphological evolution from spherical to coral-shaped gold nanoparticles (NPs). With this Account, we hope to stimulate the research interest of chemists and materials scientists and promote the predictive synthesis of functional and robust materials through the design of sequence-defined synthetic molecules.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

DFSynthesizer: Dataflow-based Synthesis of Spiking Neural Networks to Neuromorphic Hardware

Spiking Neural Networks (SNNs) are an emerging computation model that uses event-driven activation and bio-inspired learning algorithms. SNN-based machine learning programs are typically executed on tile-based neuromorphic hardware platforms, where each tile consists of a computation unit called a crossbar, which maps neurons and synapses of the program. However, synthesizing such programs on an off-the-shelf neuromorphic hardware is challenging. This is because of the inherent resource and latency limitations of the hardware, which impact both model performance, e.g., accuracy, and hardware performance, e.g., throughput. We propose DFSynthesizer, an end-to-end framework for synthesizing SNN-based machine learning programs to neuromorphic hardware. The proposed framework works in four steps. First, it analyzes a machine learning program and generates SNN workload using representative data. Second, it partitions the SNN workload and generates clusters that fit on crossbars of the target neuromorphic hardware. Third, it exploits the rich semantics of the Synchronous Dataflow Graph (SDFG) to represent a clustered SNN program, allowing for performance analysis in terms of key hardware constraints such as number of crossbars, dimension of each crossbar, buffer space on tiles, and tile communication bandwidth. Finally, it uses a novel scheduling algorithm to execute clusters on crossbars of the hardware, guaranteeing hardware performance. We evaluate DFSynthesizer with 10 commonly used machine learning programs. Our results demonstrate that DFSynthesizer provides a much tighter performance guarantee compared to current mapping approaches.

Computer Science↗
DOE OSTI2024

Genomes of multicellular algal sisters to land plants illuminate signaling network evolution

Zygnematophyceae are the algal sisters of land plants. Here we sequenced four genomes of filamentous Zygnematophyceae, including chromosome-scale assemblies for three strains of Zygnema circumcarinatum . We inferred traits in the ancestor of Zygnematophyceae and land plants that might have ushered in the conquest of land by plants: expanded genes for signaling cascades, environmental response, and multicellular growth. Zygnematophyceae and land plants share all the major enzymes for cell wall synthesis and remodifications, and gene gains shaped this toolkit. Co-expression network analyses uncover gene cohorts that unite environmental signaling with multicellular developmental programs. Our data shed light on a molecular chassis that balances environmental response and growth modulation across more than 600 million years of streptophyte evolution.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Coupon synthesis status report

This letter report is in submission of completion of the Level 4 milestone number M4AT-23OR1101023 “Coupon synthesis status report” for the AT-23OR110102 “Beryllium Carbide as moderator for MSRs – ORNL” WBS number 2.04.11.01 within the larger Advanced Reactor Technologies program at Oak Ridge National Laboratory. This effort is an initial study of the viability of beryllium carbide as a future neutron moderator for molten salt reactors and other high temperature reactors. The research was awarded funding in September 2022 and this effort will be completed and reported on as a Level 2 milestone by the end of August 2023. This work is a collaboration between Oak Ridge National Laboratory and the University of Michigan. This report will discuss the status of this effort and planed work before the final reporting.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2017

Final Report: Archiving Data to Support Data Synthesis of DOE Sponsored Elevated CO 2 Experiments

Over the last three decades DOE made a large investment in field-scale experiments in order to understand the role of terrestrial ecosystems in the global carbon cycle, and forecast how carbon cycling will change over the next century. The Smithsonian Environmental Research Center received one of the first awards in this program and managed two long-term studies (25 years and 10 years) with a total of approximately $10 million of support from DOE, and many more millions leveraged from the Smithsonian Institution and agencies such as NSF. The present DOE grant was based on the premise that such a large investment demands a proper synthesis effort so that the full potential of these experiments are realized through data analysis and modeling. The goal of the this grant was to archive legacy data from two major elevated carbon dioxide experiments in DOE databases, and to engage in synthesis activities using these data. Both goals were met. All datasets deemed a high priority for data synthesis and modeling were prepared for archiving and analysis. Many of these datasets were deposited in DOE’s CDIAC, while others are being held at the Oak Ridge National Lab and the Smithsonian Institution until they can be received by DOE’s new ESS-DIVE system at Berkeley Lab. Most of the effort was invested in researching and re-constituting high-quality data sets from a 30-year elevated CO 2 experiment. Using these data, the grant produced products that are already benefiting climate change science, including the publication of new coastal wetland allometry equations based on 9,771 observations, public posting of dozens of datasets, metadata and supporting codes from long-term experiments at the Global Change Research Wetland, and publication of two synthetic data papers on scrub oak forest responses to elevated CO 2 . In addition, three papers are in review or nearing submission reporting unexpected long-term patterns in ecosystem responses to elevated CO 2 and nitrogen in a coastal wetland.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2018

Control of Reactivity in Nanoporous Metal/Ionic Liquid Composite Catalysts

In this program, we explored a new concept in nanostructured heterogeneous catalysis design – the nanoporous metal/ionic liquid (IL) composite. Over the course of this entire program from inception, we produced sixteen publications [DOE1-DOE16]. The idea behind this concept was to tailor the chemical environment within and near the pores of a nanoporous high surface area metal electrocatalyst in order to enhance the aggregate composite activity and selectivity. Most electrocatalyst design focuses upon lowering the activation barrier for a particular synthesis reaction. While we also try to minimize this quantity, one should also recognize that significant catalytic enhancements can be made by biasing the reaction to completion via control of the environment in which the reaction occurs. Specific ideas we explored in this program included (a) chemically biasing mass transport of reactants to the catalyst surface and products away from it, and (b) corralling of reactants and products to spatially separate them and reduce side reactions. We made a particular focus on electrocatalysis under kinetically limited conditions, and in which the catalyst is a nanoporous metal and the biasing environment is an ionic liquid impregnating the pores.

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

Synthesis of a planar, multicomponent catalytic surface of Na 2 CO 3 /MnO

Here, O 2 adsorption on MnO(100) precovered with sodium (Na) multilayers was investigated by X-ray photoelectron spectroscopy (XPS) and temperature programmed desorption (TPD). Deposition of Na multilayers leads to a first monolayer of oxidic Na followed by metallic Na island growth. XPS results for the oxidation of the metallic Na islands indicate an incomplete oxidation of Na at 125 K. Oxidation at 350 K completely oxidizes the metallic islands producing a mixture of Na 2 O and Na 2 O 2 . Thermal evolution of the oxidation products was examined. After oxidation at 350 K and flashing to 750 K, Na 2 O and Na 2 O 2 are the primary oxides on MnO(100). After flashing to 850 K, a solid state reaction of Na 2 O 2 /Na 2 O and the MnO(100) substrate forms a NaMnO 2 -like surface compound which decomposes above 850 K. Oxygen exchange between CO 2 and Na oxides is observed. The strong interaction between CO 2 and Na oxide islands forms Na 2 CO 3 on MnO(100). Heating the Na 2 CO 3 covered MnO(100) surface to 600 K for 10 min drives oxidic Na in the first monolayer into the MnO subsurface, and produces a surface exposing islands of Na 2 CO 3 on MnO(100).

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

A Gene Cluster That Encodes Histone Deacetylase Inhibitors Contributes to Bacterial Persistence and Antibiotic Tolerance in Burkholderia thailandensis

Persister cells are genetically identical variants in a bacterial population that have phenotypically modified their physiology to survive environmental stress. In bacterial pathogens, persisters are able to survive antibiotic treatment and reinfect patients in a frustrating cycle of chronic infection. To better define core persistence mechanisms for therapeutics development, we performed transcriptomics analyses of Burkholderia thailandensis populations enriched for persisters via three methods: flow sorting for low proton motive force, meropenem treatment, and culture aging. Although the three persister-enriched populations generally displayed divergent gene expression profiles that reflect the multimechanistic nature of stress adaptations, there were several common gene pathways activated in two or all three populations. These include polyketide and nonribosomal peptide synthesis, Clp proteases, mobile elements, enzymes involved in lipid metabolism, and ATP binding cassette (ABC) transporter systems. In particular, identification of genes that encode polyketide synthases (PKSs) and fatty acid catabolism factors indicates that generation of secondary metabolites, natural products, and complex lipids could be part of the metabolic program that governs the persistence state. We also found that loss-of-function mutations in the PKS-encoding gene locus BTH_I2366, which plays a role in biosynthesis of histone deacetylase (HDAC) inhibitors, resulted in increased sensitivity to antibiotics targeting DNA replication. Furthermore, treatment of multiple bacterial pathogens with a fatty acid synthesis inhibitor, CP-640186, potentiated the efficacy of meropenem against the persister populations. Altogether, our results suggest that bacterial persisters may exhibit an outwardly dormant physiology but maintain active metabolic processes that are required to maintain persistence.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Mediated Fuel Cells: Soluble Redox Mediators and their Applications to Electrochemical Reduction of O2 and Oxidation of H2, Alcohols, Biomass, and Complex Fuels

Mediated fuel cells are electrochemical devices that produce power in a manner similar to that of conventional proton exchange membrane fuel cells (PEMFCs). They differ from PEMFCs in their use of redox mediators dissolved in liquid electrolyte to conduct oxidation of the fuel or reduction of the oxidant, typically O2, in bulk solution. The mediators transport electrons (and often protons) between the electrode and the catalysts or chemical reagents in solution. This strategy can help overcome many of the challenges associated with conventional fuel cells, including managing complex multi-phase reactions (as in O2 reduction) or the use of challenging or heterogeneous fuels, such as hydrocarbons, polyols and biomass. Mediators are also commonly used in enzymatic fuel cells, where direct electron transfer from the electrode to the enzymatic active site can be slow. This review provides a comprehensive survey of historical and recent mediated fuel cell efforts, including applications using chemical and enzymatic catalysts. Our research on in this field has been exploring a number of different topics and has been supported different funding sources: the Center for Molecular Electrocatalysis, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences (molecular mediators for catalytic oxygen reduction); the Great Lakes Bioenergy Research Center, DOE Office of Science DE-SC0018409 (research on biomass-based fuels; U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Catalysis Program, DE-FG02-05ER15690 (copper-catalyzed oxidation reactions); and the NIH NIGMS, R01 GM100143 and R35 GM134929 (use of molecular mediators for organic chemical synthesis).

Anson, Colin W.↗
DOE OSTI2023

Architecture-Preserving Provable Repair of Deep Neural Networks

Deep neural networks (DNNs) are becoming increasingly important components of software, and are considered the state-of-the-art solution for a number of problems, such as image recognition. However, DNNs are far from infallible, and incorrect behavior of DNNs can have disastrous real-world consequences. This paper addresses the problem of architecture-preserving V-polytope provable repair of DNNs. A V-polytope defines a convex bounded polytope using its vertex representation. V-polytope provable repair guarantees that the repaired DNN satisfies the given specification on the infinite set of points in the given V-polytope. An architecture-preserving repair only modifies the parameters of the DNN, without modifying its architecture. The repair has the flexibility to modify multiple layers of the DNN, and runs in polynomial time. It supports DNNs with activation functions that have some linear pieces, as well as fully-connected, convolutional, pooling and residual layers. To the best our knowledge, this is the first provable repair approach that has all of these features. We implement our approach in a tool called APRNN. Using MNIST, ImageNet, and ACAS Xu DNNs, we show that it has better efficiency, scalability, and generalization compared to PRDNN and REASSURE, prior provable repair methods that are not architecture preserving.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Achieving Phase Control of Polymorphic Tungsten Carbide Catalysts

The polymorphism of tungsten carbide (W x C) and the challenge of selectively synthesizing pure phases have impeded a precise understanding of catalytic structure−property relationships. This study establishes a framework for phase-selective synthesis of W x C through controlling carburization kinetics. By maintaining particle sizes below 10 nm, β-W 2 C is selectively synthesized using gaseous carbon precursors (CH 4 /H 2 ) via temperature-programmed carburization (TPC). Our findings reveal that W 2 C stabilization is predominantly dictated by particle size and carburization kinetics rather than support interactions, providing a tunable approach to synthesize tungsten carbide catalysts. We elucidate the mechanistic pathway of WO x carburization, demonstrating that CH 4 activation occurs at mild temperatures via lattice oxygen. Our reactor studies establish ex situ synthesized β- W 2 C as an active and stable catalyst for the reverse water-gas shift (RWGS) reaction. However, the need for passivation and reduction pretreatment leads to a complex surface structure with diminished intrinsic activity. In contrast, our in situ synthesis protocol for β-W 2 C eliminates the need for passivation and exhibits increased CO STY during RWGS, illustrating the intrinsically higher activity compared to metallic W, WC 1−x (0.5 < x < 1), and stoichiometric WC.

CO2 conversion↗