SEARCH · Engineering Papers

Results for “Variational principle”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Alloying multiple halide perovskites on the same sublattice in search of stability and target band gaps

Single-component halide perovskites (HPs) rarely satisfy all the necessary criteria for optoelectronic applications, such as achieving an optimal band gap while maintaining high chemical and structural stability. Alloying halide perovskites has emerged as a promising strategy, not only to enhance stability but also to fine-tune their electronic and optical properties. In this work, we explore multiple degrees of freedom in alloy design, considering different substitution sublattices sites (A, B, or X in ABX3 perovskites), various chemical species (isovalent and hetero-valent elements), and multi-component compositions on a given sublattice. Using first-principles calculations based on density functional theory (DFT), we investigate how compositional variations influence the electronic (band gap) and structural properties (mixing enthalpy) of HP alloys. Our approach employs the polymorphous cell model, allowing full local relaxation which breaks local symmetry while preserving global cubic symmetry—an essential framework for accurately modeling HPs. Our results reveal that X-site mixing (halogen substitution) primarily affects the valence band maximum, allowing target band gap engineering. Additionally, variations in halogen radii introduce internal strain through octahedral distortions, influencing the mixing enthalpy. A-site substitution, while not directly contributing to the band edge states, modifies structural stability via volume effects, indirectly impacting the band gap. B-site alloying plays a dominant role in band gap modulation, leading to either positive or negative band gap bowing. Specifically, isovalent B-site mixing (Sn–Pb) induces strong positive bowing, where the alloy band gap is smaller than the average gap of parent compounds, whereas hetero-valent mixing (Cd–Pb) results in pronounced negative bowing. As an aside, we investigate the competition between the excess energy of disordered alloys vs. that of long-range ordered double perovskites of the same compositions, seeking examples of ordered phases emerging from disordered alloys. Furthermore, our findings provide fundamental insights into the electronic and structural behavior of HP alloys, offering valuable design principles for the development of stable and efficient materials for next-generation photovoltaic and optoelectronic devices.

14 SOLAR ENERGY↗

DOE OSTI2025

Applying Linear Optics from Closed Orbits Modulation for finding beam-based alignment of harmonic sextupoles

A fast and accurate beam-based alignment (BBA) method for harmonic sextupoles has been devel oped at NSLS-II using Linear Optics from Closed Orbit Modulation (LOCOM). The approach excites the beam with simultaneous sine-wave signals at two fast correctors, chosen with an appropriate phase advance to span the full betatron phase space. This strategy suppresses systematic errors from hysteresis, while additional errors from orbit drift and power-supply calibration are minimized by the short measurement time (a few minutes) and reliance solely on beam-based current-to-field conver sion of the sextupoles, with hysteresis explicitly included. Simulations indicate that Linear Optics from Closed Orbits (LOCO) combined with 0.5 mm local orbit bumps can resolve relative sextupole field offsets (∆k₂) with precision better than 10% of k₂, reflecting to beam-based alignment (BBA) accuracy finer than 50 µm. Moreover, employing machine-learning-optimized local orbit bumps en hances sextupole-induced quadrupole signals in a deterministic manner and maintains orbit stability under large sextupole strength variations (±40%). The proposed method is experimentally validated through proof-of-principle measurements at NSLS-II, demonstrating its potential as a fast, precise, and robust tool for harmonic sextupole alignment.

43 PARTICLE ACCELERATORS↗

DOE OSTI2025

Mechanistic insights into superionic thioarsenate argyrodite solid electrolytes via machine learning interatomic potentials

The lithium argyrodite sulfide solid electrolyte Li 6 PS 5 Cl has attracted considerable interest for all-solid-state batteries owing to its high ionic conductivity, which can be further enhanced through ionic substitution. Although a variety of substitutions have been investigated, thioarsenate argyrodites remain comparatively underexplored. Here, we systematically investigate the phase stability and Li-ion conduction mechanisms in superionic Br-incorporated thioarsenate argyrodites using first-principles calculations and molecular dynamics simulations based on machine learning interatomic potentials (MLIPs). Systematic variation of S/Br site inversion reveals that an optimal degree of anion disorder significantly enhances inter-cage connectivity and facilitates long-range Li-ion diffusion. Configurational entropy serves as an effective quantitative descriptor of anion disorder, exhibiting a strong correlation with ionic conductivity. While greater anion disorder induced by site inversion and higher Br content enhances ionic conductivity up to 50 mS cm −1 , it simultaneously reduces structural stability. This trade-off results in an optimal window in which a moderate level of disorder yields conductivities exceeding 20 mS cm −1 while maintaining synthetic feasibility. In conclusion, this work highlights the reliability and efficiency of MLIPs for elucidating ion-transport mechanisms and accelerating the design of novel superionic argyrodites.

Jang, Myeongcho [Korea Institute of Science and Te↗

DOE OSTI2025

Designing an Optimal Sensor Network via Minimizing Information Loss

Optimal experimental design is a classic topic in statistics, with many well-studied problems, applications, and solutions. The design problem we study is the placement of sensors to monitor spatiotemporal processes, explicitly accounting for the temporal dimension in our modeling and optimization. We observe that recent advancements in computational sciences often yield large datasets based on physics-based simulations, which are rarely leveraged in experimental design. We introduce a novel model-based sensor placement criterion, along with a highly-efficient optimization algorithm, which integrates physics-based simulations and Bayesian experimental design principles to identify sensor networks that “minimize information loss” from simulated data. Our technique relies on sparse variational inference and (separable) Gauss-Markov priors, and thus may adapt many techniques from Bayesian experimental design. We validate our method through a case study monitoring air temperature in Phoenix, Arizona, using state-of-the-art physics-based simulations. Our results show our framework to be superior to random or quasi-random sampling, particularly with a limited number of sensors. We conclude by discussing practical considerations and implications of our framework, including more complex modeling tools and real-world deployments.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures

Generative AI (genAI) has emerged as a powerful tool for synthesizing diverse and complex image data, offering new possibilities for scientific imaging applications. This review presents a comprehensive comparative analysis of leading generative architectures, ranging from Variational Autoencoders (VAEs) to Generative Adversarial Networks (GANs) on through to Diffusion Models, in the context of scientific image synthesis. We examine each model's foundational principles, recent architectural advancements, and practical trade-offs. Our evaluation, conducted on domain-specific datasets including microCT scans of rocks and composite fibers, as well as high-resolution images of plant roots, integrates both quantitative metrics (SSIM, LPIPS, FID, CLIPScore) and expert-driven qualitative assessments. Results show that GANs, particularly StyleGAN, produce images with high perceptual quality and structural coherence. Diffusion-based models for inpainting and image variation, such as DALL-E 2, delivered high realism and semantic alignment but generally struggled in balancing visual fidelity with scientific accuracy. Importantly, our findings reveal limitations of standard quantitative metrics in capturing scientific relevance, underscoring the need for domain-expert validation. We conclude by discussing key challenges such as model interpretability, computational cost, and verification protocols, and discuss future directions where generative AI can drive innovation in data augmentation, simulation, and hypothesis generation in scientific research.

Generative Adversarial Networks↗

DOE OSTI2026

Integrated machine learning-molecular dynamics framework for electrolyte property prediction

Electrochemical stability windows determine the operating range of battery electrolytes, yet accurate prediction remains challenging because stability emerges from statistical ensembles of local solvation environments rather than single ground-state molecular structures. Traditional density functional theory calculations on energy-minimized clusters cannot capture the thermal variations in local coordination environments and geometries that govern decomposition, while SMILES-based machine learning methods lack explicit representation of three-dimensional solvation structure and ion pairing. Here, we introduce a structure-aware machine learning framework that predicts frontier orbital energies (HOMO and LUMO) directly from molecular dynamics-sampled solvation configurations, achieving sub-0.6 eV accuracy at computational costs 3–4 orders of magnitude lower than first-principles methods. Across twelve representative battery electrolytes, we demonstrate that solvent-separated and contact ion pairs exhibit strong size- and local chemistry dependent electronic stability, with variations in coordination shifts of HOMO or LUMO level by 2–3 eV, and that extended solvation structure and partially desolvated environment further modulate stability by up to 3 eV. By encoding the statistical nature of electrochemical failure through ensemble sampling of explicit solvation geometries, our approach enables high-throughput screening and rational design of next-generation battery electrolytes with mechanistic understanding of structure–property relationships.

Energy - Storage↗

DOE OSTI2025

Long-read sequencing transcriptome quantification with lr-kallisto

RNA abundance quantification has become routine and affordable thanks to high-throughput “short-read” technologies that provide accurate molecule counts at the gene level. Similarly accurate and affordable quantification of definitive full-length, transcript isoforms has remained a stubborn challenge, despite its obvious biological significance across a wide range of problems. “Long-read” sequencing platforms now produce data-types that can, in principle, drive routine definitive isoform quantification. However some particulars of contemporary long-read datatypes, together with isoform complexity and genetic variation, present bioinformatic challenges. We show here, using ONT data, that fast and accurate quantification of long-read data is possible and that it is improved by exome capture. To perform quantifications we developed lr-kallisto, which adapts the kallisto bulk and single-cell RNA-seq quantification methods for long-read technologies.

Loving, Rebekah K. (ORCID:0000000187250376)↗

DOE OSTI2024

Constructing nested coordinates inside strongly shaped toroids using an action principle

A new approach for constructing polar-like boundary-conforming coordinates inside a toroid with strongly shaped cross-sections is presented. A coordinate mapping is obtained through a variational approach, which involves identifying extremal points of a proposed action in the mapping space from [0,2π] 2 ×[0,1] to a toroidal domain in $\mathbb{R}$ 3 . This approach employs an action built on the squared Jacobian and radial length. Extensive testing is conducted on general toroidal boundaries using a global Fourier–Zernike basis via action minimisation. The results demonstrate successful coordinate construction capable of accurately describing strongly shaped toroidal domains. The coordinate construction is successfully applied to the computation of three-dimensional magnetohydrodynamic equilibria in the GVEC code where the use of traditional coordinate construction by interpolation from the boundary failed.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2025

Mapping the three-dimensional fermiology of the triangular lattice magnet EuAg 4 Sb 2

In this paper, we report the temperature-field phase diagram as well as present a comprehensive study of the electronic structure and three-dimensional fermiology of the triangular-lattice magnet EuAg 4 ⁢Sb 2 , utilizing quantum oscillation measurements, angle-resolved photoemission spectroscopy, and first-principles calculations. The complex magnetic phase diagram of EuAg 4 ⁢Sb 2 highlights many transitions through nontrivial AFM states. Shubnikov-de Haas and de Haas-van Alphen oscillations were observed in the polarized ferromagnetic state of EuAg 4 ⁢Sb 2 , revealing three pairs of distinct spin-split frequency branches with small effective masses. A comparison of the angle-dependent oscillation data with first-principles calculations in the ferromagnetic state and angle-resolved photoemission spectra shows good agreement, identifying tubular hole pockets and hourglass-shaped hole pockets at the Brillouin zone center, as well as diamond-shaped electron pockets at the zone boundary. As the temperature increases, the frequency branches of the tiny hourglass pockets evolve into a more cylindrical shape, while the larger pockets remain unchanged. This highlights that variations in exchange splitting, driven by changes in the magnetic moment, primarily impact the small Fermi pockets without significantly altering the overall band structure. As a result, this is consistent with first-principles calculations, which show minimal changes near the Fermi level across ferromagnetic and simple antiferromagnetic states or under varying on-site Coulomb repulsion.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Biocrust Mosses and Cyanobacteria Exhibit Distinct Carbon Uptake Responses to Variations in Precipitation Amount and Frequency

ABSTRACT Dryland organisms exhibit varied responses to changes in precipitation, including event size, frequency, and soil moisture duration, influencing carbon uptake and reserve management strategies. This principle, central to the pulse‐reserve paradigm, has not been thoroughly evaluated in biological soil crusts (biocrusts), essential primary producers on dryland surfaces. We conducted two experiments to investigate carbon uptake in biocrusts under different precipitation regimes. In the first, we applied a gradient of watering amounts to biocrusts dominated by moss or cyanobacteria, hypothesising distinct pulse‐response strategies. The second experiment extended watering treatments over three months, varying pulse size and frequency. Our results revealed distinct carbon uptake patterns: moss crusts exhibited increased CO 2 uptake with larger, less frequent watering events, whereas cyanobacteria crusts maintained similar carbon uptake across all event sizes. These findings suggest divergent pulse‐response strategies across biocrust types, with implications for modelling dryland carbon dynamics and informing land management under changing precipitation regimes.

Young, Kristina E. [USDA‐ARS Jornada Experimental ↗

DOE OSTI2025

Organic Electrochemical Transistor Channel Materials: Copolymerization Versus Physical Mixing of Glycolated and Alkoxylated Polymers

Organic electrochemical transistors (OECTs) feature a polymer channel capable of conducting both ions and electronic charges. The choice of the channel material is critical for OECT performance. Many efforts have focused on improving performance via the chemical tunability of conjugated polymers – through backbone, side chain, and molar mass engineering – leading to useful design principles for accumulation-mode OECT materials. However, tuning the chemical structure of conjugated polymers often requires time-consuming optimization of the synthesis route. Meanwhile, variations in molar mass, dispersity, structural defects, and metal content present challenges when attempting to analyze the detailed effects of structural modifications, as multiple performance-determining factors are often (unintentionally) changed at the same time. Therefore, this study explores blended channel materials obtained by physically mixing glycolated and alkoxylated polymers in different ratios, and compares their OECT performance with the corresponding statistical copolymers. It is shown that mixing two well-performing materials creates blends that enable rational tuning of the transistor properties without compromising on performance. Thus, channels based on blends of alkoxylated and glycolated polymers hold promise for OECT technology with tailored response, as only two materials are needed to achieve any desired side chain ratio, simplifying the optimization of OECT characteristics.

copolymerization↗

DOE OSTI2025

Applications of Nickelate perovskites for neuromorphic computing from electronic structure and Machine Learning

While the limit of Moore's law is presently being reached with current microelectronic technologies, we need to develop new paradigms that overcome this limitation. In that respect, neuromorphic computing is a concept that emulates the neural behavior and response of the human brain, and it has been recognized as a promising alternative approach. In this research project, we will perform multi-fidelity scale bridging to explore the potential use of materials with metal to insulator transition for neuromorphic applications. In particular, rare earth nickelates are promising for such purposes, as the transition in these materials is quite sensitive to a broad set of different external stimuli. Our multi-fidelity approach will bridge the high-fidelity electronic structure calculations with classical potentials. We will bridge dynamical mean field theory with a classical atomistic representation via a deep learning force field. The neural network is trained with energies, charges, and forces obtained by accurate electronic structure theories based on Dynamical Mean Field Theory. The configurational space is generated from known crystal phases, ab initio molecular dynamics with exchange-correlation functionals corrected with the Hubbard model, disordered phases with different concentrations of oxygen vacancies, and nonsymmetrical positions and induced strain by grain interfaces or contact with a substrate. Strategies to train the model with a reduced number of training examples are obtained from active learning methods, and new structures for improving the learning process are generated by using machine learning autoencoders. This classical potential will be validated through a diversity of electronic structure methods and represents an important step to combine the flexibility and accuracy of first-principles with the speed of classical potentials. The generated multi-fidelity surrogate model will be used to understand the role of strain, oxygen vacancies, proton doping, the variation of the crystal phase, substrate effects, vibrational effects as the octahedral rotation, grain boundaries and defect effects on the response of a Metal to Insulator Transition (MIT) in correlated materials. Long time and large-scale simulations will help understand the role of different stimuli to control the hysteresis of the MIT, as it has been experimentally suggested. Selected configurations will be analyzed with higher-level theories to provide an accurate electronic description and to study how the orbitals and charges are rearranged under different conditions.

36 MATERIALS SCIENCE↗

DOE OSTI2025

A Representation Fusion Framework for Decoupling Diagnostic Information in Multimodal Learning

Modern medicine increasingly relies on multimodal data, ranging from clinical notes to imaging and genomics, to guide diagnosis and treatment. However, integrating these heterogeneous data sources in a principled and interpretable manner remains a major challenge. We present MODES (Multi-mOdal Disentangled Embedding Space), a representation fusion framework that explicitly separates shared and modality-specific factors of variation, offering a structured latent space for multimodal information that improves both prediction and interpretability. By leveraging pre-trained unimodal foundation models, MODES mitigates the dependency on extensive paired datasets, crucial in data-scarce clinical settings. We introduce a masking strategy that optimizes representation dimensionality by eliminating low-information dimensions, to achieve compact, information-rich representations. Our framework demonstrates superior performance in predicting diagnoses and phenotypes compared to unimodal and conventional fusion models. MODES also enables robust diagnostic inference in missing data scenarios, offering an opportunity toward interpretable and efficient multimodal diagnostics in personalized healthcare.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2025

Evidence for Ga clusters in β-Ga 2 O 3 from Raman spectroscopy and density functional theory

Monoclinic gallium oxide (β-Ga 2 O 3 ) single crystals have a Raman mode at ∼250 cm −1 that is strongly correlated with free-electron density. Prior work attributed this peak to an electronic excitation of a shallow donor impurity band. However, heavily n-type thin films grown by metalorganic chemical vapor deposition or molecular beam epitaxy do not have the peak. In the present work, an alternate model is proposed: the 250 cm −1 Raman peak arises from Ga clusters, defined as two or more Ga atoms that form Ga–Ga bonds. Raman mapping reveals variations in the frequency that are consistent with a distribution of cluster sizes. The intensity of the peak decreases as the temperature is raised, attributed to melting of the Ga clusters. First-principles calculations indicate that the 250 cm −1 mode is due to Ga–Ga bond-stretching vibrations. As the Fermi energy is raised, the formation of Ga–Ga dimers becomes energetically favorable, explaining the correlation between n-type conductivity and the appearance of the Raman peak.

36 MATERIALS SCIENCE↗

DOE OSTI2025

In‐situ Analysis of Paste Properties in Resonant Acoustic Mixers for Quality Monitoring

Formulation control is key to achieving consistent target properties of energetic materials, as feedstock variations and slight deviations in the ratios of different ingredients can have major effects on final product properties, particularly in dense pastes with high particle loading >65 vol.%. In large‐scale operations, it is imperative to either correct or remove batches of material that perform outside baseline property specifications as early as possible to avoid unnecessary processing of suboptimal material. Quality monitoring is the practice of measuring material properties during processing using process analytical technologies as opposed to only testing the properties of the final product; it is a key principle in the quality‐by‐design frameworks used for designing formulations and manufacturing processes. Herein, a process analytical technology method for correlating material properties of dense pastes directly after mixing in a Resonant Acoustic Mixer to motor data is developed and used to detect differences in the particle content of dense paste formulations. This method was also capable of detecting variations in powder feedstock properties, such as particle packing efficiency, and is sensitive enough to detect changes of 2 wt.% in the total solids content of the formulation. The techniques presented herein show excellent promise for use as a process analytical technology capable of quantifying formulation effects on material movement modes during resonant acoustic mixing.

Materials science↗

DOE OSTI2026

Anisotropic Ferromagnetism in CrAu 3 Sb 6

The crystal structure and properties of CrAu 3 Sb 6 are presented, determined by measurements on single crystal and polycrystalline samples and first-principles calculations. The trigonal structure (space group P$\overline{3}$1m) comprises a CdI 2 -like sublattice of AuSb 2 with Cr occupying octahedral holes in a fully ordered triangular array. It can be viewed as a variation of the interesting and well-known families of partially intercalated transition metal dichalcogenides, but with stronger interactions along the stacking direction evidenced by short Cr–Au distances. The compound is metallic and ferromagnetic with a Curie temperature of 164 K. A strong anomalous contribution to the Hall effect is seen in the ferromagnetic state, and quantum oscillations are observed in magnetization at 2 K. Magnetization measurements show that the ordered moments of 1.5 μ B per Cr are oriented along the c-axis with relatively strong magnetocrystalline anisotropy. Electronic structure calculations confirm this uniaxial anisotropy and the important role of spin–orbit coupling in CrAu 3 Sb 6 and reveal strongly favored ferromagnetic ground state consistent with the measured Curie temperature. Through combined experiment and theory, this work provides a detailed picture of the basic properties and behaviors of this uniquely structured, Cr-based, anisotropic ferromagnet.

Crystal structure↗

DOE OSTI2025

Learning nuclear cross sections across the chart of nuclides with graph neural networks

We explore the use of deep learning techniques to learn how nuclear cross sections change as we add or remove protons and neutrons. As a proof of principle, we focus on the neutron-induced reactions in the fast energy regime. Our approach follows a two-stage learning framework. First, we apply representation learning to encode cross section data into a latent space using either variational autoencoders (VAEs) or implicit neural representations (INRs). Then, we train graph neural networks (GNNs) on the resulting embeddings to predict missing values across the nuclear chart by leveraging the topological structure of neighboring isotopes. We demonstrate accurate cross section predictions within a 9 × 9 block of missing nuclei. We also find that the optimal GNN training strategy depends on the type of latent representation used, with VAE embeddings performing best under end-to-end optimization in the original space, while INR embeddings achieve better results when the GNN is trained only in the latent space. Furthermore, using clustering algorithms, we map groups of latent vectors into regions of the nuclear chart and show that VAEs and INRs can discover some of the neutron magic numbers. These findings suggest that deep-learning models based on the representation encoding of cross sections combined with graph neural networks hold significant potential in augmenting nuclear theory models, e.g., by providing reliable estimates of covariances of cross sections, including cross-material covariances.

Machine learning↗

DOE OSTI2025

𝐴𝑏 initio density-matrix approach to exciton coherence: Phonon scattering, Coulomb interactions, and radiative recombination

Relaxation processes following light excitation in semiconductors are key in materials-based quantum technology applications. These processes are broadly studied in atomically thin transition-metal dichalcogenides, quasi-two-dimensional excitonic semiconductors in which atomistic design allows for tunable excited-state properties, such as relaxation lifetimes and photoinduced coherence. In this work, we present a density-matrix-based approach to compute exciton relaxation within a many-body ab initio perspective. We expand our previously developed Lindblad density-matrix formalism to capture multichannel electron-hole pair relaxation processes, including phonon and Coulomb scattering as well as radiative recombination, and we study their effect on the time-resolved excited-state propagation. Using monolayer MoSe 2 as a prototypical example, we examine many-body effects on the time-dependent dynamics of photoactive excitations, exploring how the electron-hole pair interactions are reflected in variations of the excitation energy, spectral signature, and state coherence. In conclusion, our method supplies a detailed understanding of exciton relaxation mechanisms in realistic materials, offering a previously unexplored pathway to study excited-state dynamics in semiconductors from first principles.

Band structure methods↗