Search NASA⌕ Search

SEARCH · Search NASA

Results for “compositional”

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 1,135 records · Page 63
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

Study of microstructure, dielectric and ferromagnetic properties of the (1-x)Ba{sub 0.95}Sr{sub 0.05}TiO{sub 3}-(x)CoFe{sub 1.8}Mn{sub 0.2}O{sub 4} multiferroic composites

In the present work, (1-x)Ba{sub 0.95}Sr{sub 0.05}TiO{sub 3}-(x)CoFe{sub 1.8}Mn{sub 0.2}O{sub 4} composites are prepared by standard solid state reaction method. The X-ray diffraction measurement of the composites shows that both the phases coexist in the composite, individually. The morphology of the composites were examined by field emission scanning electron microscopy and reveals homogeneous microstructure with two types of grains, smaller grains of the Ba{sub 0.95}Sr{sub 0.05}TiO{sub 3} (BST) and bigger grains of the CoFe{sub 1.8}Mn{sub 0.2}O{sub 4} (CFM). The dielectric studies show that all the composites exhibit dispersion in the lower frequency region attributable to the interfacial polarization. In addition, at lower frequencies, the dielectric constant (ε´) is found to increase with increase in CFM content in the composites. The ferromagnetic properties of the composites improve with the increase in the CFM content.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2015

Compositional modulated atomic layer stacking and uniaxial magnetocrystalline anisotropy of CoPt alloy sputtered films with close-packed plane orientation

An atomic layer stacking structure in hexagonal close packed (hcp) Co{sub 100−x}Pt{sub x} alloy films with c-plane sheet texture was directly observed by a high-angle annular dark-field imaging scanning transmission electron microscopy. The analysis of sequential and/or compositional atomic layer stacking structure and uniaxial magnetocrystalline anisotropy (K{sub u} = K{sub u1} + K{sub u2}) revealed that (1) integrated intensity of the superlattice diffraction takes the maximum at x = 20 at. % and shows broadening feature against x for the film fabricated under the substrate temperature (T{sub sub}) of 400 °C. (2) Compositional separation structure in atomic layers is formed for the films fabricated under T{sub sub} = 400 °C. A sequential alternative stacking of atomic layers with different compositions is hardly formed in the film with x = 50 at. %, whereas easily formed in the film with x = 20 at. %. This peculiar atomic layer stacking structure consists of in-plane-disordered Pt-rich and Pt-poor layers, which is completely different from the so-called atomic site ordered structure. (3) A face centered cubic atomic layer stacking as faults appeared in the host hcp atomic layer stacking exists in accompanies with irregularities for the periodicity of the compositional modulation atomic layers. (4) K{sub u1} takes the maximum of 1.4 × 10{sup 7 }erg/cm{sup 3} at around x = 20 at. %, whereas K{sub u2} takes the maximum of 0.7 × 10{sup 7 }erg/cm{sup 3} at around x = 40 at. %, which results in the maximum of 1.8 × 10{sup 7 }erg/cm{sup 3} of K{sub u} at x = 30 at. % and a shoulder in compositional dependence of K{sub u} in the range of x = 30–60 at. %. Not only compositional separation of atomic layers but also sequential alternative stacking of different compositional layers is quite important to improve essential uniaxial magnetocrystalline anisotropy.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2015

Effects of magnetic field treatment on dielectric properties of CCTO@Ni/PVDF composite with low concentration of ceramic fillers

Using melt mixing, we produced a ceramic/polymer composite with a matrix of polyvinylidene fluoride (PVDF) and a filler of 5 vol.% Ni-deposited CaCu{sub 3}Ti{sub 4}O{sub 12} core-shell ceramic particles (CCTO@Ni), and studied its prominent dielectric characteristics for the first. Its phase composition and morphology were analyzed by X-ray diffraction and scanning electron microscopy, respectively. After treating the composite films with various durations of a magnetic field treatment, we compared their dielectric properties. We found that the CCTO@Ni ceramic had a typical urchin-like core-shell structure, and that different durations of the magnetic field treatment produced different distributions of ceramic particles in the PVDF matrix. The dielectric permittivity of the untreated CCTO@Ni/PVDF composite was 20% higher than that of neat PVDF, and it had a low loss tangent. However, only the composite treated for 30 min in the magnetic field had an ultra-high dielectric permittivity of 1.41 × 10{sup 4} at 10 Hz, three orders of magnitude higher than the untreated composite, which declined dramatically with increasing frequency, accompanied by an insulating-conducting phase transition and an increase in loss tangent. Our results demonstrate that changes in the dielectric properties of PVDF composites with magnetic field treatment are closely related to the percolation effect and interfacial polarization.

36 MATERIALS SCIENCE↗
DOE OSTI2017

Methodology for optimizing composite towers for use on floating wind turbines

In this work, a methodology for the design and optimization of a composite wind turbine tower for use on a floating offshore platform is presented. A composite turbine tower on a floating offshore platform not only has the potential to reduce maintenance and upkeep costs associated with the use of steel offshore but also has potential to reduce the tower mass and subsequently the support platform mass. The optimization problem is formulated to obtain a turbine tower that meets all strength and serviceability criteria and minimizes the tower mass. The optimization and design process link a number of dynamic analyses and finite element routines using a genetic algorithm. This work documents the optimization and design software and illustrates its use in various case studies for a 6 MW floating wind turbine system. Case studies include the optimization of a steel tower as a comparison to a composite material tower and the use of a cored, sandwich panel composite tower versus a solid composite tower. The results demonstrate that significant reductions in the tower mass are likely when comparing a steel tower with a composite tower. Also demonstrated for the platform and turbine configuration considered in this work is that the minimum mass tower design is driven towards a solid shell composite configuration, rather than a sandwich panel tower shell, in the face of reasonable design constraints.

17 WIND ENERGY↗
DOE OSTI2019

Guided Optimization of Phase-Unstable III–V Compositionally Graded Buffers by Cathodoluminescence Spectrum Imaging

Compositionally graded buffers (CGBs) are essential components of high-efficiency III-V inverted metamorphic solar cells, and high-quality CGBs with low defect densities are a requirement for high device efficiencies. Here, we use plan-view and cross-sectional cathodoluminescence spectrum imaging (CLSI) to investigate the effect of growth conditions on the microstructure of Al y Ga 1-x-y In x AS CGBs with x In > 0.30. CLSI reveals significant compositional fluctuation in these materials that correlates with the distribution of threading dislocations in the CGBs. Threading dislocations are observed to collect in regions between abrupt transitions in emission energy, implying that concomitant strain fluctuations in these regions restrict dislocation motion. Thus, the compositional fluctuation leads to elevated threading dislocation densities that degrade CGB quality. We also correlate compositional fluctuation in the CGB with compositional fluctuation in subsequently grown device layers. The effectiveness of specific growth conditions at suppressing compositional fluctuation and limiting threading dislocation density is evaluated. CGBs grown at high temperatures with high Al content exhibit the lowest defect densities. The use of a high V/III ratio suppresses compositional fluctuation and defect density. In conclusion, these insights lead directly to improvements in the efficiency of metamorphic Ga 1-x In x As solar cells grown on AlyGa 1-x-y In x As CGBs.

14 SOLAR ENERGY↗
DOE OSTI2018

Scalable measurements of tow architecture variability in braided ceramic composite tubes

Abstract The performance of braided ceramic matrix composites has been shown to depend on the spatial arrangement of tows; therefore, a new class of tools is required to measure irregularities in the composite architecture for components with intricate geometries. We report a scalable and robust reconstruction technique built upon stereoscopic digital image correlation that is able to efficiently measure the position of tows in arbitrarily shaped composites. This method was applied to triaxially braided ceramic matrix composite tubes intended for use as nuclear fuel cladding, which revealed both long‐range “systematic” tow packing defects associated with the manufacturing process and short‐range “intrinsic” defects due to the braid architecture. These findings suggested that the character of tow spacing variation in braided composite tubes was substantially more complex than in planar woven composites. These measurements are expected to lead to improved processing of braided composites and to facilitate the design of statistically representative virtual specimens for finite element modeling.

Heim, Frederick M.↗
DOE OSTI2024

Above room temperature multiferroic and magnetoelectric properties of (1‐Φ) PZTFT‐Φ CZFMO particulate composites

Magnetoelectric (ME) composites of suitable ferroelectric and magnetic materials can display elevated magnetic and ferroelectric operational temperatures, along with substantial ME coupling, compared to conventional single-phase multiferroics. Herein, we describe the synthesis of (1-Φ) PZTFT-Φ CZFMO, Φ = 0.1, 0.2, 0.3 (PZTFT: [0.6(PbZr 0.53 Ti 0.47 O 3 )–0.4(PbFe 0.5 Ta 0.5 )O 3 ] CZFMO: Co 0.6 Zn 0.4 Fe 1.7 Mn 0.3 O 4 )] particulate (0–3) composites and report on the magnetic as well as ferroelectric phase transitions and magnetoelectric coupling. The phase formation and the induced strain in these composites are investigated via Raman spectroscopy. A large bifurcation of the zero-field cooled-field cooled magnetization curves confirms the highly anisotropic behavior of the composites. These curves also identify spin glass behavior in the CZFMO phase at ≈230 K. The magnetic phase transition of the composite (Φ = 0.2) is reported to be ≈532 (± 10) K. The temperature dependent dielectric data displays the ferroelectric phase transitions from the PZTFT phase and the broad relaxation peak from the CZFMO phase. The quadratic relationship between the magneto-capacitance and the magnetization confirms the existence of biquadratic magnetoelectric coupling in the systems. The collective results are consistent with the presence of a direct magneto-electric effect in the composites, i.e., by the application of magnetic field, the magnetic phase is strained, and this induced strain is responsible for changes of ferroelectric order parameter in the piezoelectric phase. As a result, this attribute makes the current composite structure a promising candidate for multiferroic data storage and processing technologies.

Bhoi, Krishnamayee↗
DOE OSTI2021

Passive sensing of a microparticle modified hybrid, fiber-reinforced composite

The widespread commercial adoption of high-performance, fiber-reinforced composites has pushed research interests toward the next generation of composites. These new composites are tasked with integrating additional functionalities into the structures without causing a trade-off in mechanical performance. One such functionality that has received significant interest is sensing. This is especially important for composites using high-performance fibers (e.g., carbon fiber) because their strain-to-failure is relatively low, resulting in brittle fracture. Besides, fiber damage can be hidden within the composite, potentially leading to premature catastrophic failure if not detected. In prior research, we demonstrated continuous feed-through deposition of ceramic nanoparticles on carbon fiber’s surface that simultaneously enhanced both the piezoresistive response and interlaminar shear strength. In this work, a similar continuous feed-through deposition process was used to demonstrate passive sensing and energy harvesting by integrating ferroelectric microparticles on the surface of electrically nonconductive fibers. The sensing and energy harvesting capabilities were characterized by mechanically straining composite beams and measuring the power generated. The improvements in mechanical properties are shown through interlaminar shear strength tests. Therefore, this research aims to demonstrate a high throughput, commercially scalable approach to coat fibers with ferroelectric microparticles that enable passive sensing as well as improved mechanical performance when fabricated into a fiber-reinforced composite.

Bowland, Christopher↗
DOE OSTI2024

Multiscale modeling-enabled design of multifunctional composites

This study aims to create a comprehensive model that considers multiple scales and physics for predicting the electromechanical behavior of fiber-reinforced composites enhanced with barium titanate (BaTiO3). In our earlier work, we have demonstrated that depositing BaTiO3 microparticles of 200-nm-diameter, on fiber surfaces during fiber-reinforced composite fabrication enhances mechanical strength, passive self-sensing, and energy harvesting properties. The key is to carefully control the microparticle concentration to prevent agglomeration. Since the particles are micron-sized, understanding how agglomeration affects the composites' electromechanical properties is crucial for guiding such multifunctional materials’ design. This study introduces a micromechanics-based approach to explore the impact of microparticle dispersion on the bulk composites' electromechanical properties. Insights gained from this investigation are applied in experiments, enabling accurate predictions of mechanical and self-sensing responses in BaTiO3-enhanced fiber-reinforced composites. Micro-level findings from this computational approach can be integrated into larger continuum models to comprehensively capture the electromechanical behavior of the composite structures at bulk scale. The proposed model is validated by comparing predictions with experimental results, accounting for the nonlinear mechanical and electromechanical behaviors of constituent materials. Consequently, this computational model serves as a digital platform for efficiently designing multifunctional composites.

Gupta, Sumit↗
DOE OSTI2025

Tensile and fatigue characterization of multifunctional composites

This research is part of a larger effort to develop advanced self-sensing multifunctional polymer composites that are both lightweight and high-strength, while also enabling structural damage detection, fatigue cycle monitoring, and service life prediction. These multifunctional composites are particularly sought after in the automotive industry for their potential to significantly reduce vehicle weight and simultaneously provide additional functionality like condition monitoring to enhance safety. This study examines the tensile and fatigue properties of a composite material composed of acrylonitrile butadiene styrene (ABS) polymer embedded with piezoelectric barium titanate (BaTiO3) nanoparticles. The integration of BaTiO3 nanoparticles not only supplies the material with self-sensing capabilities but also influences its mechanical properties. While a high content of BaTiO3 nanoparticles is desired to enhance sensing capacity, the brittle nature of such materials causes concerns of decreased strength characteristics. To explore this, various composite samples were fabricated with nanoparticle contents ranging from 0 wt% to 20 wt%. These samples underwent tensile testing to measure their ultimate tensile strengths and Young’s moduli. Following this, fatigue tests were conducted to generate S-N curves, which are essential for understanding the material's durability under cyclic loading. The findings from these tests assess the impact of nanoparticle content on the composite’s tensile strength and fatigue life, providing essential insights that can guide the optimization and design of future self-sensing multifunctional composites. The results suggest that 5 wt% BaTiO3 provides an optimal balance between mechanical properties and nanoparticle concentration, making it a promising composition for semi-structural applications.

Bowland, Christopher [ORNL] (ORCID:000000021229431↗
DOE OSTI2019

Experimental Evidence that Stochasticity Contributes to Bacterial Composition and Functioning in a Decomposer Community

ABSTRACT Stochasticity emerging from random differences in replication, death, mutation, and dispersal is thought to alter the composition of ecological communities. However, the importance of stochastic effects remains somewhat speculative because stochasticity is not directly measured but is instead inferred from unexplained variations in beta-diversity. Here, we performed a field experiment to more directly disentangle the role of stochastic processes, environmental selection, and dispersal in the composition and functioning of a natural bacterial decomposer community in the field. To increase our ability to detect these effects, we reduced initial biological and environmental heterogeneity using replicate nylon litterbags in the field. We then applied two treatments: ambient/added precipitation and bacterial and fungal dispersal using “open” litterbags (made from 18.0-μm-pore-size mesh) (“open bacterial dispersal”) versus bacterial and fungal dispersal using “closed” litterbags (made from 22.0-μm-pore-size mesh) (“closed bacterial dispersal”). After 5 months, we assayed composition and functioning by the use of three subsamples from each litterbag to disentangle stochastic effects from residual variation. Our results indicate that stochasticity via ecological drift can contribute to beta-diversity in bacterial communities. However, residual variation, which had previously been included in stochasticity estimates, accounted for more than four times as much variability. At the same time, stochastic effects on beta-diversity were not attenuated at the functional level, as measured by genetic functional potential and extracellular enzyme activity. Finally, dispersal was found to interact with precipitation availability to influence the degree to which stochasticity contributed to functional variation. Together, our results demonstrate that the ability to quantify stochastic processes is key to understanding microbial diversity and its role in ecosystem functioning. IMPORTANCE Randomness can alter the diversity and composition of ecological communities. Such stochasticity may therefore obscure the relationship between the environment and community composition and hinder our ability to predict the relationship between biodiversity and ecosystem functioning. This study investigated the role of stochastic processes, environmental selection, and dispersal in microbial composition and its functioning on an intact field community. To do this, we used a controlled and replicated experiment that was similar to that used to study population genetics in the laboratory. Our study showed that, while the stochastic effects on taxonomic composition are smaller than expected, the degree to which stochasticity contributes to variability in ecosystem processes may be much higher than previously assumed.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Compositional zoning of the Otowi Member of the Bandelier Tuff, Valles caldera, New Mexico, USA

The Otowi Member of the Bandelier Tuff erupted at ca. 1.60 Ma from the Valles caldera (New Mexico, USA). It consists of as much as 400 km 3 (dense rock equivalent) of strongly differentiated high-silica rhyolite and shows systematic upward variations in crystallinity, mineral chemistry, and trace element concentrations through its thickness, but the major element composition is almost constant and is near the low-pressure granite minimum. Incompatible trace elements in whole pumice fragments and glasses show well-correlated linear covariations. Upward zoning to lower abundances of incompatible trace elements is accompanied by development of overgrowths on quartz and alkali feldspar, although earlier-formed interiors of quartz and feldspar have near-constant compositions throughout the tuff, modified by cation diffusion in the case of feldspar. Melt inclusions in remnant quartz cores show diverse Pb isotope ratios, pointing to a wide range of distinct protoliths that contributed rhyolitic melt to the Otowi magma. Mineral thermometers suggest a modest temperature gradient through the melt body, perhaps of 40 °C, at the time of eruption. Chemical, textural, and mineralogical variations and volume-composition relations through the tuff are consistent with an origin for zoning by melting of a high-crystallinity cumulate layer beneath cognate supernatant liquid to produce denser, remobilized liquid of accumulative composition (i.e., the “modified mush model”). Melting may have occurred in several episodes. The latest of these episodes, probably thousands of years prior to eruption, introduced new rhyolitic liquid into the system and was associated with a thermal excursion, recorded in core compositions of pyroxene, during which much of the earlier crystal mass was dissolved. This left inherited cores and interiors of accumulated quartz and feldspar mantled with new growth having less-evolved compositions (higher Ti, Sr, and Ba). Changing solubility of zircon during cumulate melting produced a reversal of Zr concentrations. There is no clear petrologic evidence of a recharge eruption trigger; nonetheless, compositional zoning resulted mainly from repeated recharge-induced remobilization of quartz-feldspar cumulate. The Otowi system was built, evolved, and modified by several events over the course of a few hundred thousand years.

Wolff, J. A.↗
DOE OSTI2017

Understanding the Effect of Preparative Approaches in the Formation of “Flower-like” Li 4 Ti 5 O 12 —Multiwalled Carbon Nanotube Composite Motifs with Performance as High-Rate Anode Materials for Li-Ion Battery Applications

Herein we highlight the significance of nanoscale attachment modality as an important determinant of observed electrochemical performance. Specifically, controlled loading ratios of multi-walled carbon nanotubes (MWNTs) have been successfully anchored onto the surfaces of a unique “flower-like” Li 4 Ti 5 O 12 (LTO) micro-scale sphere motif, for the first time, using a number of different and distinctive preparative approaches, including (i) a sonication method, (ii) an in situ direct-deposition approach, (iii) a covalent attachment protocol, as well as (iv) a π-π interaction strategy. In terms of structural characterization, the composites generated by physical sonication as well as non-covalent π-π interactions retained the intrinsic hierarchical “flower-like” morphology and exhibited a similar crystallinity profile as compared with that of pure LTO. By comparison, the composite prepared by an in situ direct deposition approach yielded not only a fragmented LTO structure, likely due to the possible interfering presence of the MWNTs themselves during the relevant hydrothermal reaction, but also a larger crystallite size, owing to the higher annealing temperature associated with its preparation. Finally, the composite created via covalent attachment was covered with an amorphous insulating linker, which probably led to a decreased contact area between the LTO and the MWNTs and hence, a lower crystallinity in the resulting composite. In addition electrode tests suggested that the composite generated by π-π interactions out-performed the other three analogous heterostructures, due to a smaller charge transfer resistance as well as a faster Li-ion diffusion. In particular, the LTO-MWNT composite, produced by π-π interactions, exhibited a reproducibly high rate capability as well as a reliably solid cycling stability, delivering 132 mA h g -1 at 50 C, after 100 discharge/charge cycles, including 40 cycles at a high (>20 C) rate. To conclude, such data denote the highest electrochemical performance measured to date as compared with any LTO-carbon nanotube-based composite materials previously reported, under high discharge rate conditions, and tangibly underscore the correlation between preparative methodology and the resulting performance metrics.

25 ENERGY STORAGE↗
DOE OSTI2015

Intermediate-scale Fire Performance of Composite Panels under Varying Loads

New aircraft are being designed with increasing quantities of composite materials used in their construction. Different from the more traditional metals, composites have a higher propensity to burn. This presents a challenge to transportation safety analyses, as the aircraft structure now represents an additional fuel source involved in the fire scenario. Most of the historical fire testing of composite materials is aimed at studying kinetics, flammability or yield strength under fire conditions. Most of this testing is small-scale. Heterogeneous reactions are often length-scale dependent, and this is thought to be particularly true for composites which exhibit significant microscopic dynamics that can affect macro-scale behavior. We have designed a series of tests to evaluate composite materials under various structural loading conditions with a consistent thermal condition. We have measured mass-loss, heat flux, and temperature throughout the experiments. Several types of panels have been tested, including simple composite panels, and sandwich panels. The main objective of the testing was to understand the importance of the structural loading on a composite to its behavior in response to fire-like conditions. During flaming combustion at early times, there are some features of the panel decomposition that are unique to the type of loading imposed on the panels. At load levels tested, fiber reaction rates at later times appear to be independent of the initial structural loading.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Increasing Compositions In NBSR Neutronics Model – Part 2

Calculations in support of the NIST research reactor (NBSR) use a Monte Carlo neutronics model in conjunction with a burnup capability. The representation of compositions within the fuel plates for the current reactor has assumed that with burnup there is a single uniform composition within all plates in each half fuel element. That model has been validated using operational data and leads to acceptable conservative results for safety analyses. However, for the conversion of the reactor to the use of low-enriched uranium (LEU) fuel, there has been an effort to change the modeling for several reasons. The use of large composition zones is inconsistent with the rigor in the rest of the neutronics model. There is also the possibility that the conservatism in maximum fission density may lead to excessive requirements for LEU fuel qualification. A new composition model has been proposed that would break each half fuel element into 15 composition zones improving the rigor while still allowing reasonable running time. This approach was tested, and results show that although it is an improvement over the previous model that had one composition zone (and an interim model with six zones) in each half fuel element, it falls short of being the ideal composition model. This report documents the objectives for the new model and the analysis of the resulting calculations.

42 ENGINEERING↗
DOE OSTI2022

Composite Solid Ion Conductor with Engineered Lithium Interface

Successful widespread commercialization of electric vehicles is contingent upon development of safe high energy density batteries capable of long cycle life. Lithium metal affords the highest theoretical capacity (3,860 mAh/g) and lowest electrochemical potential (-3.04V vs SHE), which offers the highest specific energy density of anode materials today. While Li-ion batteries are capable of delivering energy densities of 400-600 Wh/kg, the development of lithium metal batteries such as Li-S and Li-air may boost this number up to 650 and 950 Wh/kg, respectively. However, significant progress towards the passivation of lithium metal must occur before the energy density benefit can be realized. Issues with lithium dendrite formation, anode volume expansion, and continuous solid electrolyte interphase (SEI) build-up often result in significant safety concerns, high cell resistance, and poor cycle life. The intrinsic high reactivity between lithium metal with conventional lithium ion electrolytes (organic carbonate-based solvents) makes it extremely difficult to overcome these problems. In this project, Wildcat performed focused, fundamental research and development on composite polymer/ceramic electrolytes and protected lithium metal anodes to develop an all solid state lithium metal battery targeting the DOE requirements to enable commercialization. Wildcat successfully leveraged its high throughput battery platform and explored a broad composite electrolyte compositional space. Additionally, Wildcat screened large numbers of inorganic and organic coatings for lithium metal protection using in situ liquid methods and translated the best results to all solid cells. The proposed composite polymer/ceramic electrolyte and a protected lithium metal anode enabled a solid-state lithium metal battery. The ultimate targets from this effort will deliver a safe all solid-state lithium metal pouch cell with over 350 Wh/kg and over 1,000 cycles (C/3) with the cost estimate below $100/kWh. During the course of the project, we demonstrated stable cycling of 10’s of cycles with the composite SSE and our protected lithium. The work has continued after the conclusion of the DOE funding, demonstrating further improvements in catholyte composition and thinner composite solid electrolytes.

25 ENERGY STORAGE↗
DOE OSTI2023

An ICME Modeling Framework for Titanium/Tungsten-Carbide Metal Matrix Composites

This report describes a collaborative project to develop a validated, predictive model for the high temperature mechanical properties of a titanium-matrix, tungsten-carbide/cobalt-reinforced metal matrix composite. The modeling approach was to first develop a detailed, microstructural model linking the material structure and the interfacial debonding properties to the effective properties of the material. The project then completed a throughput simulation campaign to generate a large number of simulations for discrete microstructures and different debonding parameters. Finally, the project trained a fast, Gaussian process surrogate model against this simulation database to provide a quick model linking the material compositions, structure, and processing parameters to the resulting material properties. This model was validated against high temperature tensile test data on a few particular composite compositions. The tests validate the model predictions for ultimate tensile strength and uniform elongation/ductility, meaning the final surrogate model can now be used to tune the material composition and processing parameters to identify optimal composite compositions for particular applications.

36 MATERIALS SCIENCE↗
DOE OSTI2022

MetaText: Compositional Generalization in Deep Language Models

Compositional generalization, the ability to infer and reason using novel combinations of previously encountered entities and structures, is a trait of great utility across a variety of deep learning tasks; it implies a certain ability to reason using consistent and therefore interpretable rules, to understand the constituent parts of its input at multiple levels of granularity, to be robust to spurious differences between semantically equivalent inputs, and more. We seek to understand the degree to which existing natural language models achieve or fall short of compositional generalization across a variety of tasks, whether there are distinct types of compositional generalization in practice, and to identify avenues of intervention through which we can improve models’ compositional generalization ability. After exploring these questions using an array of models, tasks, and potential interventions we see that large, pretrained language models have encountered sufficient training data to account for a variety of fine-grained compositional behavior but still struggle to reason at the level of phrases or larger language structures. Non-intrusive, data-based interventions in the form of augmenting individual sequences with compressed versions of themselves or deriving new examples from induced grammars prove insufficient to encourage greater levels of compositional reasoning, indicating that future work might benefit most from focusing on changes to a model’s inductive bias at the architectural or loss level, or by integrating compositionality-boosting data interventions into the large-scale pretraining process itself.

97 MATHEMATICS AND COMPUTING↗