SEARCH · Engineering Papers
Results for “performance prediction”
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.
Fundamental Thermodynamic, Kinetic, and Mechanical Properties of Lithium and Its Alloys
Lithium alloying reactions are beneficial in promoting uniform plating and stripping of lithium metal in all-solid-state batteries. First-principles calculations are performed to predict thermodynamic, kinetic, and mechanical properties of lithium and several important Li–M alloys (M = Mg, Ag, Zn, Al, Ga, In, Sn, Sb, and Bi). While the Li–Mg binary system forms a solid solution, most other lithium–metal alloys prefer stoichiometric intermetallic compounds with common local motifs that enable fast Li diffusion. Lithium and Li-rich alloys exhibit an unusually flat energy landscape along paths that connect BCC to close-packed structures like FCC and HCP, with important implications for mechanical properties. Very low migration barriers for Li diffusion that rival those of superion conductors are predicted, both in pure Li and in Li–M intermetallics. However, vacancy concentration, which is crucial for substitutional diffusion, is predicted to be low in metallic Li and most Li–M intermetallics. Compounds such as B32 LiAl and LiGa as well as D0 3 Li 3 Sb and Li 3 Bi exhibit structural vacancies at higher ends of their voltage windows, which together with low migration barriers leads to exceptionally high Li mobilities. In the Li–Mg solid solution, the addition of Mg is found to decrease the vacancy tracer diffusion coefficient by an order of magnitude.
Three-Dimensional Pore Networks in Miocene Stevens Sandstone of California: Implications for CO 2 Geologic Storage
The Miocene Stevens Sandstone in the San Joaquin Basin of California is increasingly recognized as a promising candidate for CO 2 geological storage due to the enormous storage capacity, proven sealing, and existing infrastructure. In this study, computed microtomography imaging and pore network modeling were employed to investigate the influence of pore geometry and wettability on the CO 2 injectivity and residual trapping. Image analysis revealed that a significant fraction of the cement and matrix consists of microporous regions. The microporosity can substantially increase the overall pore space, yet its contribution to permeability remains modest, particularly in samples with low permeability. The intrinsic heterogeneity of turbidite reservoirs further complicates the reservoir properties among different layers. Two-phase flow simulations under varying wettability conditions (water-wet, weak water-wet, and neutral-wet) demonstrated that the CO 2 injection is predominantly controlled by macropores. CO 2 invades microporous regions only after these larger pores are filled. The presence of microporosity leads to a decrease in both initial and residual CO 2 saturations, with the magnitude of the reduction being influenced by wettability. Neutral-wet scenarios exhibit higher CO 2 mobility and thus lower residual trapping than water-wet scenarios. The results imply that heterogeneity in pore geometry and cement distribution across different layers can result in stratified CO 2 flow pathways, complicating efforts to predict injection performance. Overall, the Stevens Sandstone shows considerable promise for CO 2 geologic storage, but effective implementation will require detailed characterization of the pore structure as well as the integration of reactive fluid flow to account for potential mineral dissolution and fines migration.
Invariant Molecular Representations for Heterogeneous Catalysis
Catalyst screening is a critical step in the discovery and development of heterogeneous catalysts, which are vital for a wide range of chemical processes. In recent years, computational catalyst screening, primarily through density functional theory (DFT), has gained significant attention as a method for identifying promising catalysts. However, the computation of adsorption energies for all likely chemical intermediates present in complex surface chemistries is computationally intensive and costly due to the expensive nature of these calculations and the intrinsic idiosyncrasies of the methods or data sets used. This study introduces a novel machine learning (ML) method to learn adsorption energies from multiple DFT functionals by using invariant molecular representations (IMRs). To do this, we first extract molecular fingerprints for the reaction intermediates and later use a Siamese-neural-network-based training strategy to learn invariant molecular representations or the IMR across all available functionals. Our Siamese network-based representations demonstrate superior performance in predicting adsorption energies compared with other molecular representations. Notably, when considering mean absolute values of adsorption energies as 0.43 eV (PBE-D3), 0.46 eV (BEEF-vdW), 0.81 eV (RPBE), and 0.37 eV (scan+rVV10), our IMR method has achieved the lowest mean absolute errors (MAEs) of 0.18 0.10, 0.16, and 0.18 eV, respectively. These results emphasize the superior predictive capacity of our Siamese network-based representations. The empirical findings in this study illuminate the efficacy, robustness, and dependability of our proposed ML paradigm in predicting adsorption energies, specifically for propane dehydrogenation on a platinum catalyst surface.
Machine Learning Potentials with the Iterative Boltzmann Inversion: Training to Experiment
Methodologies for training machine learning potentials (MLPs) with quantum-mechanical simulation data have recently seen tremendous progress. Experimental data have a very different character than simulated data, and most MLP training procedures cannot be easily adapted to incorporate both types of data into the training process. Here, we investigate a training procedure based on iterative Boltzmann inversion that produces a pair potential correction to an existing MLP using equilibrium radial distribution function data. By applying these corrections to an MLP for pure aluminum based on density functional theory, we observe that the resulting model largely addresses previous overstructuring in the melt phase. Interestingly, the corrected MLP also exhibits improved performance in predicting experimental diffusion constants, which are not included in the training procedure. The presented method does not require autodifferentiating through a molecular dynamics solver and does not make assumptions about the MLP architecture. Our results suggest a practical framework for incorporating experimental data into machine learning models to improve the accuracy of molecular dynamics simulations.
COCOMO2: A Coarse-Grained Model for Interacting Folded and Disordered Proteins
Biomolecular interactions are essential in many biological processes, including complex formation and phase separation processes. Coarse-grained computational models are especially valuable for studying such processes via simulation. Here, we present COCOMO2, an updated residue-based coarse-grained model that extends its applicability from intrinsically disordered peptides to folded proteins. This is accomplished with the introduction of a surface exposure scaling factor, which adjusts interaction strengths based on solvent accessibility, to enable the more realistic modeling of interactions involving folded domains without additional computational costs. COCOMO2 was parametrized directly with solubility and phase separation data to improve its performance on predicting concentration-dependent phase separation for a broader range of biomolecular systems compared to the original version. COCOMO2 enables new applications including the study of condensates that involve IDPs together with folded domains and the study of complex assembly processes. COCOMO2 also provides an expanded foundation for the development of multiscale approaches for modeling biomolecular interactions that span from residue-level to atomistic resolution.
Correction to “COCOMO2: A Coarse-Grained Model for Interacting Folded and Disordered Proteins”
Biomolecular interactions are essential in many biological processes, including complex formation and phase separation processes. Coarse-grained computational models are especially valuable for studying such processes via simulation. Here, we present COCOMO2, an updated residue-based coarse-grained model that extends its applicability from intrinsically disordered peptides to folded proteins. This is accomplished with the introduction of a surface exposure scaling factor, which adjusts interaction strengths based on solvent accessibility, to enable the more realistic modeling of interactions involving folded domains without additional computational costs. COCOMO2 was parametrized directly with solubility and phase separation data to improve its performance on predicting concentration-dependent phase separation for a broader range of biomolecular systems compared to the original version. COCOMO2 enables new applications including the study of condensates that involve IDPs together with folded domains and the study of complex assembly processes. COCOMO2 also provides an expanded foundation for the development of multiscale approaches for modeling biomolecular interactions that span from residue-level to atomistic resolution.
Measurement of Three Transport Coefficients and the Thermodynamic Factor in Block Copolymer Electrolytes with Different Morphologies
The design and engineering of composite materials is one strategy to satisfy the materials needs of systems with multiple orthogonal property requirements. In the case of rechargeable batteries with lithium metal anodes, the system requires a separator with fast lithium ion transport and good mechanical strength. In this work, we focus on the system polystyrene-block-poly(ethylene oxide) (SEO) with bis(trifluoromethane)sulfonimide lithium salt (LiTFSI). Ion transport occurs in the salt-containing poly(ethylene oxide)-rich domains. Mechanical rigidity arises due to the glassy nature of polystyrene (PS). If we assume that the salt does not interact with the PS-rich domains, we can describe ion transport in the electrolyte by three transport parameters (ionic conductivity, κ, salt diffusion coefficient, D, and cation transference number, t + 0 ) and a thermodynamic factor, T f . By systematically varying the volume fraction of the conducting phase, Φ c between 0.29 and 1.0, and chain length, N between 80 and 8000, we elucidate the role of morphology on ion transport. We find that κ is the strongest function of morphology, varying by three full orders of magnitude, while D is a weaker function of morphology. To calculate t + 0 and T f , we measure the current fraction, ρ + , and the open circuit potential, U, of concentration cells. We find that ρ + and U follow universal trends as a function of salt concentration, regardless of chain length, morphology, or Φ c , allowing us to calculate t + 0 for any SEO/LiTFSI or PEO/LiTFSI mixture when κ and D are known. Finally, the framework developed in this paper enables predicting the performance of any block copolymer electrolyte in a rechargeable battery.
Exceptional Radiation Absorption in a Pentagon-Based Si Allotrope
Excellent photovoltaic performance is predicted in a pentagonal covalent network of Si in a hollow structure exhibiting both thermal and dynamical stability. Consisting of a combination of sp 2 and sp 3 hybridized Si atomic orbitals, the GW0 computed band structure shows an indirect band gap near the zone edge and also a manifold of directly absorbing transitions at frequencies in the window of visible light, in distinction with conventional Si. Hydrogenation of a single sp 2 site is predicted to lead to a robust local magnetic moment. In this study, we find a low formation energy at low pressure that is compatible with other experimentally known phases, suggesting that a stable phase might be obtained.
Above-Room-Temperature Ferromagnetism in Thin van der Waals Flakes of Cobalt-Substituted Fe 5 GeTe 2
Two-dimensional (2D) magnetic van der Waals materials provide a powerful platform for studying the fundamental physics of low-dimensional magnetism, engineering novel magnetic phases, and enabling thin and highly tunable spintronic devices. To realize high-quality and practical devices for such applications, there is a critical need for robust 2D magnets with ordering temperatures above room temperature that can be created via exfoliation. Here, in this paper, the study of exfoliated flakes of cobalt-substituted Fe 5 GeTe 2 (CFGT) exhibiting magnetism above room temperature is reported. Via quantum magnetic imaging with nitrogen-vacancy centers in diamond, ferromagnetism at room temperature was observed in CFGT flakes as thin as 16 nm corresponding to 16 layers. This result expands the portfolio of thin room-temperature 2D magnet flakes exfoliated from robust single crystals that reach a thickness regime relevant to practical spintronic applications. The Curie temperature T c of CFGT ranges from 310 K in the thinnest flake studied to 328 K in the bulk. To investigate the prospect of high-temperature monolayer ferromagnetism, Monte Carlo calculations were performed, which predicted a high value of T c of ~270 K in CFGT monolayers. Pathways toward further enhancing monolayer T c are discussed. These results support CFGT as a promising platform for realizing high-quality room-temperature 2D magnet devices.
Too Many Materials and Too Many Applications: An Experimental Problem Waiting for a Computational Solution
Finding the best material for a specific application is the ultimate goal of materials discovery. However, there is also the reverse problem: when experimental groups discover a new material, they would like to know all the possible applications this material would be promising for. Computational modeling can aim to fulfill this expectation, thanks to the sustained growth of computing power and the collective engagement of the scientific community in developing more efficient and accurate workflows for predicting materials' performances. We discuss the impact that reproducibility and automation of the modeling protocols have on the field of gas adsorption in nanoporous crystals. We envision a platform that combines these tools and enables effective matching between promising materials and industrial applications.
Dynamic Promotion of the Oxygen Evolution Reaction via Programmable Metal Oxides
Hydrogen gas is a promising renewable energy storage medium when produced via water electrolysis, but this process is limited by the sluggish kinetics of the anodic oxygen evolution reaction (OER). Herein, we used a microkinetic model to investigate promoting the OER using programmable oxide catalysts (i.e., forced catalyst dynamics). We found that programmable catalysts could increase current density at a fixed overpotential (100–600× over static rates) or reduce the overpotential required to reach a fixed current density of 10 mA cm –2 (45–140% reduction vs static). In our kinetic parametrization, the key parameters controlling the quality of the catalytic ratchet were the O*-to-OOH* and O*-to-OH* activation barriers. Furthermore, our findings indicate that programmable catalysts may be a viable strategy for accelerating the OER or enabling lower-overpotential operation, but a more accurate kinetic parametrization is required for precise predictions of performance, ratchet quality, and resulting energy efficiency.
Probing the Radial Chemistry of Getter Components in Light Water Reactors via Controlled Electrochemical Dissolution
Getters are among the key functional components in the Tritium Producing Burnable Absorber Rods (TPBARs) of Light Water Reactors, and are used to capture the released tritium gas. They are cylindrical tubes comprised of the alloy zircaloy-4 which is coated with nickel plating both internally and externally. Their exposure to tritium as well as the irradiation environments in the light water reactors changes their structure, chemical composition and chemistry. Understanding the radial tritium distribution is key to gaining insight into the evolution of new chemistry upon irradiation in order to predict getter performance. Consequently, efforts are underway to devise a method that can allow selective probing into the getter core radially to probe the chemistry akin to carefully and selectively peeling of the layers of an onion. We have recently devised an electrochemical dissolution technique for controlled radial dissolution of zirconium based liners; this work describes the first proof-of-concept demonstration of the applicability of this technique on the controlled radial dissolution of nickel plated getters. The technique is complemented with periodic analysis of the electrolyte solution to measure the rates of release of the various elemental components and therefore glean information on their radial distribution across the getter; further complementary microscopy on the exposed surfaces allows us insight into the chemistry at the getter core.
Investigating the Variation in Solvation Interactions of Choline Chloride-Based Deep Eutectic Solvents Formed Using Different Hydrogen Bond Donors
Choline chloride-based deep eutectic solvents (DESs) have gained widespread recognition as green solvents in catalysis and separation science as their designer properties permit solvation properties to be modulated by the choice of the hydrogen bond donor (HBD) as well as its molar ratio. Despite being one of the most popular classes of DESs, very little is currently understood regarding how their solvation characteristics differ among themselves as well as from other classes of DESs. Previous studies have shown that the catalytic activity, extraction efficiency, and solubilities of natural compounds can be influenced by the type of choline salt and HBD, but no study has systematically related these parameters with its solvation properties. In this study, inverse gas chromatography was employed for the first time to study the individual solvation characteristics of a diverse range of choline chloride and acetate-based DESs composed of amides, diols, organic acids, and carbohydrates. Furthermore, hydrogen bond acidity was found to be a dominant interaction for all DESs, especially those composed of organic acid HBDs. DESs composed of isomers of butane diol and hexane diol as HBDs exhibited higher hydrogen bond basicity and dispersive-type interactions compared to those composed of urea, acetamide, and organic acids with the position of the hydroxyl functional group influencing their solvation interactions. Choline acetate-based DESs exhibited lower dipolarity and stronger hydrogen bond basicity and dispersive-type interactions compared to choline chloride DESs. The solvation models developed in this study were used to interpret and explain DES behavior in previously reported studies involving catalysis, organic synthesis, and extractions demonstrating that the measured solute–solvent interactions can serve to predict the performance of choline chloride-based DESs when used in various applications.
A Scale‐Adaptive Urban Hydrologic Framework: Incorporating Network‐Level Storm Drainage Pipes Representation
Abstract Below‐ground urban stormwater networks (BUSNs) significantly influence urban flood dynamics, yet their representation at the watershed or larger scales remains challenging. We introduce a scalable urban hydrologic framework that centers on a novel network‐level BUSN representation, balancing the needs for physical basis, parameter parsimony, and computational efficiency. Our framework conceptualizes an urban watershed into four interacting zones: hillslopes (natural), storm‐sewersheds (urban), a sub‐network channel (tributaries), and a main channel. We develop an innovative Graph Theory‐based algorithm to derive network‐level BUSN parameters from publicly available datasets, enabling efficient, scalable parameterization. We demonstrate this framework's applicability at nine representative watersheds in the Houston metropolitan region, USA, with urban imperviousness ranging from 0% to 64% and drainage areas ranging from 24 to 302 . Our model achieves satisfying computational efficiency, completing hourly time step simulations for 18 years in less than 5 sec per watershed on a standard PC. Validation against observed daily streamflow confirms that the model can capture small‐to‐large flood peaks and seasonal and annual water balance over these watersheds. Comparisons with the National Water Model show better performance in predicting flood peaks and overall water balance, underscoring the promises of our new framework for urban hydrologic modeling at large scales. Furthermore, analysis reveals nonlinear relationships between BUSNs' designed capacities and flood reduction effects. Our approach bridges the gap between detailed hydraulic and large‐scale hydrologic models, providing a valuable tool for urban flood prediction and management across broader spatial and temporal scales.
Converting tabular data into images for deep learning with convolutional neural networks
Abstract Convolutional neural networks (CNNs) have been successfully used in many applications where important information about data is embedded in the order of features, such as speech and imaging. However, most tabular data do not assume a spatial relationship between features, and thus are unsuitable for modeling using CNNs. To meet this challenge, we develop a novel algorithm, image generator for tabular data (IGTD), to transform tabular data into images by assigning features to pixel positions so that similar features are close to each other in the image. The algorithm searches for an optimized assignment by minimizing the difference between the ranking of distances between features and the ranking of distances between their assigned pixels in the image. We apply IGTD to transform gene expression profiles of cancer cell lines (CCLs) and molecular descriptors of drugs into their respective image representations. Compared with existing transformation methods, IGTD generates compact image representations with better preservation of feature neighborhood structure. Evaluated on benchmark drug screening datasets, CNNs trained on IGTD image representations of CCLs and drugs exhibit a better performance of predicting anti-cancer drug response than both CNNs trained on alternative image representations and prediction models trained on the original tabular data.
Predicting wind-driven spatial deposition through simulated color images using deep autoencoders
Abstract For centuries, scientists have observed nature to understand the laws that govern the physical world. The traditional process of turning observations into physical understanding is slow. Imperfect models are constructed and tested to explain relationships in data. Powerful new algorithms can enable computers to learn physics by observing images and videos. Inspired by this idea, instead of training machine learning models using physical quantities, we used images, that is, pixel information. For this work, and as a proof of concept, the physics of interest are wind-driven spatial patterns. These phenomena include features in Aeolian dunes and volcanic ash deposition, wildfire smoke, and air pollution plumes. We use computer model simulations of spatial deposition patterns to approximate images from a hypothetical imaging device whose outputs are red, green, and blue (RGB) color images with channel values ranging from 0 to 255. In this paper, we explore deep convolutional neural network-based autoencoders to exploit relationships in wind-driven spatial patterns, which commonly occur in geosciences, and reduce their dimensionality. Reducing the data dimension size with an encoder enables training deep, fully connected neural network models linking geographic and meteorological scalar input quantities to the encoded space. Once this is achieved, full spatial patterns are reconstructed using the decoder. We demonstrate this approach on images of spatial deposition from a pollution source, where the encoder compresses the dimensionality to 0.02% of the original size, and the full predictive model performance on test data achieves a normalized root mean squared error of 8%, a figure of merit in space of 94% and a precision-recall area under the curve of 0.93.
Tin-based ionic chaperone phases to improve low temperature molten sodium–NaSICON interfaces
High temperature operation of molten sodium batteries impacts cost, reliability, and lifetime, and has limited the widespread adoption of these grid-scale energy storage technologies. Poor charge transfer and high interfacial resistance between molten sodium and solid-state electrolytes, however, prevents the operation of molten sodium batteries at low temperatures. In this study, in situ formation of tin-based chaperone phases on solid state NaSICON ion conductor surfaces is shown in this work to greatly improve charge transfer and lower interfacial resistance in sodium symmetric cells operated at 110 °C at current densities up to an aggressive 50 mA cm –2 . It is shown that static wetting testing, as measured by the contact angle of molten sodium on NaSICON, does not accurately predict battery performance due to the dynamic formation of a chaperone NaSn phase during cycling. This work demonstrates the promise of sodium intermetallic-forming coatings for the advancement of low temperature molten sodium batteries by improved mating of sodium–NaSICON surfaces and reduced interfacial resistance.