Search NASA⌕ Search

SEARCH · Search NASA

Results for “Complex structure”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Electronic structure complexity and extremely large magnetoresistance in antiferromagnetic semimetal SmAgSb 2

SmAgSb 2 , a layered magnetic semimetal in the tetragonal 𝑅⁢𝑇⁢ Sb 2 family (𝑅 = Y, Sc, rare earth; 𝑇 = transition metal), is known to exhibit extremely large magnetoresistance (XMR) below its antiferromagnetic (AFM) transition temperature. Here, in this work, we present a comprehensive investigation combining magnetotransport measurements, density functional theory calculations accounting for electron correlation, and angle-resolved photoemission spectroscopy. Our results reveal a complex electronic structure characterized by a multiband Fermi surface and intricate magnetic ground states. We demonstrate that simple two-band models, previously employed in the literature, fail to consistently describe the observed transport phenomena. Notably, we report an XMR of approximately 25200% at 2 K under a 14 T magnetic field, significantly exceeding earlier reports for this material family and rivaling the performance of prominent nonmagnetic XMR systems. This pronounced enhancement below 𝑇 𝑁 suggests that the XMR originates from a combination of multiband electron-hole compensation and enhanced magnetic scattering in this correlated AFM semimetal.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2026

Stabilizing homogeneous CaMn6Sn6via oscillatory crystal growth: structural complexity in a kagome metal

CaMn6Sn6 is a member of the large family of Mn-based kagome metals derived by filling voids of the prototypical CoSn structure. We observed Ca-deficiency and structural complexity, with approximately 10% deficiency of calcium relative to the ideal 166 stoichiometry. These features are common in related Ca-based materials. We find that growth conditions stabilize different phases with varying Curie temperatures. Diffraction and magnetization measurements reveal that conventional growth profiles yield inhomogeneity both within and between crystals in a given batch, whereas an oscillatory temperature profile promotes homogeneous crystals that adopt a superstructure beyond the parent hexagonal structure type. We present the physical properties of crystals grown using the oscillatory approach. The crystals are highly conductive, with a c-axis residual resistivity ratio ρ(300 K)/ρ(2 K) exceeding 100 and a positive, linear magnetoresistance at base temperature. Magnetically, the crystals are quite similar to MgMn6Sn6, with easy-plane ferromagnetic behavior and a Curie temperature of TC ≈ 237 K. These results highlight the critical role of growth conditions in stabilizing homogeneous crystals of complex phases such as Ca0.9Mn6Sn6. The oscillatory growth approach provides an effective route for accessing the intrinsic properties of this complex material and may be broadly useful for the growth of related kagome systems.

May, Andrew [ORNL] (ORCID:0000000307778539)↗
DOE OSTI2025

Residual stress distribution in an additively manufactured complex structure by neutron diffraction measurement

Residual stress in an aerodynamically shaped Ni-based superalloy airfoil fabricated by laser powder bed fusion was measured by neutron diffraction. The experiment was conducted by considering the complex shape, implementing computer aided experiment planning, and automatic alignment at each rapid measurement. The 3-dimensional (3D) residual stress distribution in the airfoil is presented in this work, which lacks symmetry due to the complex geometry of the airfoil. In conclusion, the results provide theoretical thermal processing models a complete residual stress dataset of simulation validation on 3D shape complex structure.

Residual stress↗
DOE OSTI2025

Comparative Analysis of TCR and TCR-pMHC Complex Structure Prediction Tools

The rapid development of computational approaches for predicting the structures of T cell receptors (TCRs) and TCR-peptide-major histocompatibility (TCR-pMHC) complexes, accelerated by AI breakthroughs such as AlphaFold, has made it feasible to calculate these structures with increasing accuracy. Although these tools show great potential, their relative accuracy and limitations remain unclear due to the lack of standardized benchmarks. Here, we systematically evaluate seven tools for predicting isolated TCR structures together with six tools for predicting TCR-pMHC complex structures. The methods include homology-based approaches, general prediction tools using AlphaFold, TCR-specific tools derived from AlphaFold2, and the newly developed tFold-TCR model. The evaluation uses a post-training data set comprising 40 αβ TCRs and 27 TCR-pMHC complexes (21 Class I and 6 Class II). Model accuracy is assessed at global, local, and interface levels using a variety of metrics. We find that each tool offers distinct advantages in various aspects of its predictions. AlphaFold2, AlphaFold3, and tFold-TCR excel in overall accuracy of TCR structure prediction, and TCRmodel2 and AlphaFold2 perform well in overall accuracy of TCR-pMHC structure prediction. However, TCR-specific tools derived from AlphaFold2 show lower accuracy in the framework region than both homology-based methods and general-purpose tools such as AlphaFold, and challenges remain for all in modeling CDR3 loops, docking orientations, TCR-peptide interfaces, and Class II MHC-peptide interfaces. Furthermore, these findings will guide researchers in selecting appropriate tools, emphasize the importance of using multiple evaluation metrics to assess model performance, and offer suggestions for improving TCR and TCR-pMHC structure prediction tools.

Chemical structure↗
DOE OSTI2026

Fyn–Saracatinib Complex Structure Reveals an Active State-like Conformation

Fyn is a Src-family tyrosine kinase implicated in synaptic dysfunction and neuroinflammation across multiple neurodegenerative disorders, including Alzheimer’s disease (AD) and Parkinson’s disease (PD). Saracatinib (AZD0530) is a potent Src-family inhibitor that has been explored as a repurposed therapeutic; however, its clinical utility is limited by poor kinase selectivity caused by high sequence conservation within Src-family ATP-binding sites. Here, we combine surface plasmon resonance (SPR) and X-ray crystallography to define saracatinib recognition by the Fyn kinase domain (KD). SPR single-cycle kinetics shows that saracatinib binds the isolated Fyn KD and full-length Fyn with low-nanomolar affinity, whereas dasatinib binds with subnanomolar affinity and markedly slower dissociation. We determined the crystal structure of the Fyn KD-saracatinib complex at 2.22 Å resolution. The kinase adopts an active-like conformation with the DFG motif and αC-helix in the ‘in’ state and a conserved β3 αC Lys-Glu salt bridge. Saracatinib occupies the adenine and ribose pockets, and engages the hinge through direct and water-mediated hydrogen bonding while complementing a hydrophobic back pocket by van der Waals contacts. Comparison with reported saracatinib-bound structures of other kinases suggests that the active-state geometry observed for Fyn creates a pocket not observed in inactive-like complexes, providing a structural handle for designing Fyn-selective inhibitors. Comparison with all saracatinib-bound kinase co-structures currently available in the PDB (ALK2 and PKMYT1) indicates a conserved monodentate hinge binding mode but kinase-dependent αC-helix conformations, providing a structural rationale for designing Fyn-selective analogues.

AZD0530↗
DOE OSTI2025

Accurate and rapid acoustic damage characterization in complex structures using sparse sensor networks and deep learning models

Damage diagnosis in critical components is essential for ensuring the safety and reliability of operations across industries, spanning manufacturing, aerospace, and energy. Traditional acoustic nondestructive testing methods primarily focus on detecting defects through the direct scattering of single-mode incident waves from the damage, which limit their applicability to simple structures and small inspection areas. Our earlier research demonstrated that machine learning algorithms combined with sparse sensor networks can identify critical defect signatures even from multiply scattered, multi-mode acoustic signals, indicating the potential for improved defect inspection in complex, real-world structures. In this work, we demonstrate the successful implementation of this approach in a fixed sensor configuration to rapidly and accurately detect simulated defects in a geometrically complex, real-world structure, a brake rotor hub. Three different types of defects were physically simulated on the surface of the hub, and the collected data were used to train an autoencoder-based deep learning model. Two models were tested, one using single measurements and the other using multiple measurements taking advantage of the spatial distribution of the sensor network. After training, the multi-measurement model achieved 100 % accuracy in identifying, classifying, and locating unseen, unique damages. This work illustrates the potential of the proposed method for a wide range of industrial applications.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Dissecting Disorder: Defect-Driven Structural Complexity in Layered Li3InCl6 Solid Electrolyte

Halide solid electrolytes have emerged as promising candidates for solid-state batteries owing to their high oxidative stability and ionic conductivity. Among them, Li3InCl6 (LIC) has attracted significant attention. However, diffraction patterns of LIC synthesized via different methods exhibit distinct differences particularly at low-angle reflectionsindicative of underlying structural disorder. These variations are attributed to deviations from ideal crystallographic order, especially stacking faults, whose impact on structure and ion transport remains poorly understood. Here, we identify and quantify stacking faults in LIC samples prepared under different synthetic conditions. Using X-ray diffraction and time-of-flight neutron diffraction, we construct and refine stacking fault models that accurately reproduce the experimental diffraction features. LIC samples with higher degrees of stacking faults exhibit only negligible differences in ionic conductivities and activation energies. This indicates that stacking faults have a limited impact on altering the Li+ diffusion pathway along the c-axis, likely due to the high concentration of vacancies in the In layers, while Li+ diffusion remains nearly unchanged in the ab-plane. Our results account for the observed differences in diffraction patterns across samples and provide a quantitative assessment of faulting probabilities and stacking sequences. The insights gained from this study are expected to be broadly applicable to other layered halide solid electrolytes and contribute to a deeper understanding of the role of structural disorder in ion transport.

Liu, Jue [ORNL] (ORCID:000000024453910X)↗
DOE OSTI2025

Structural complexity in the f -block: small deviations of the complexation of lanthanides by O,Oʹ -diethylmonothiophosphate

Dithiophosphinic acids undergo radiolytic degradation during the extraction of actinides in used nuclear fuel. These will degrade into monothiophosphinic acids and then to phosphinic acids. To elucidate how the complexes that are formed during these radioactive separations change as the ligand degrades, the mixed donor ligand O,O′- diethylmonothiophosphate is chosen as an analog for the monothiophosphinic intermediate. Herein, the monothiophosphate complexes Ln 2 (OPS(OEt) 2 ) 6 (H 2 O) 8 (Ln = La) (La 2 L 6 H 2 O), Ln 2 (OPS(OEt) 2 ) 6 (EtOH) 4 (Ln = La) (La 2 L 6 EtOH), K 2 [Ln(OPS(OEt) 2 ) 5 (H 2 O) 2 ]·H 2 O·CH 2 Cl 2 , (Ln = Ce) (CeL 5 -α), K 2 [Ln(OPS(OEt) 2 ) 5 (H 2 O) 2 ]·H 2 O·CH 2 Cl 2 , (Ln = Pr) (PrL 5 -β), K[Ln(OPS(OEt) 2 ) 4 (H 2 O) 3 ], (Ln = Pr, Sm-Er) (ML 4 ), and K 3 [Ln(OPS(OEt) 2 ) 6 ], (Ln = Dy) (DyL 6 ) were synthesized and characterized using single-crystal X-ray diffraction and optical spectroscopy. Although the lanthanides contract in a nearly linear fashion, the structural changes observed as the f-block is traversed in these compounds are not necessarily a hard line but more so a blend of different structure types possible for each f-element. Furthermore, comparison of the Ln−O bond lengths shows a nearly linear contraction, but the Ln−S bond lengths do not monotonically decrease because of the hard Lewis acidity of the Ln 3+ cations.

37 - INORGANIC, ORGANIC, PHYSICAL AND ANALYTICAL C↗
DOE OSTI2025

Selective deuteration of an RNA:RNA complex for structural analysis using small-angle scattering

The structures of RNA:RNA complexes regulate many biological processes. Despite their importance, protein-free RNA:RNA complexes represent a tiny fraction of experimentally determined structures. Here, we describe a joint small-angle X-ray and neutron scattering (SAXS/SANS) approach to structurally interrogate conformational changes in a model RNA:RNA complex. Using SAXS, we measured the solution structures of the individual RNAs and of the overall RNA:RNA complex. With SANS, we demonstrate, as a proof of principle, that isotope labeling and contrast matching (CM) can be combined to probe the bound state structure of an RNA within a selectively deuterated RNA:RNA complex. Furthermore, we show that experimental scattering data can validate and improve predicted AlphaFold 3 RNA:RNA complex structures to reflect its solution structure. In conclusion, our work demonstrates that in silico modeling, SAXS, and CM-SANS can be used in concert to directly analyze conformational changes within RNAs when in complex, enhancing our understanding of RNA structure in functional assemblies.

HIV-1 dimerization initiation site↗
DOE OSTI2026

Artificial intelligence methods for protein structure and interaction prediction: Recent advances and challenges

Recent advances in artificial intelligence have introduced novel methods for high-accuracy prediction of protein tertiary structures, protein complex structures, and interactions between proteins and other biomolecules, such as small molecules and nucleic acids. Such advancements are accelerating biomedical research and the development of new protein design and bioengineering methods among many other important biotechnology applications. Here, in this review, we outline the recent advances in protein-centric biomolecular structure and interaction prediction, highlight some major challenges in the field, and discuss potential directions to address them.

Morehead, Alex [Lawrence Berkeley National Laborat↗
DOE OSTI2026

Structure of Complex Liquid–Liquid Extraction Organic Phases for Rare Earth Separations

Complex, multicomponent liquids with hierarchical structure and phase transitions are encountered in many natural and industrial processes, including in chemical separations. One notable example is aggregation and organic phase splitting in liquid–liquid extraction (LLE) of metal ions. While these two phenomena that have long been closely associated, a mechanistic link between mesoscale structure and the capacity-limiting organic phase splitting remains elusive due to complexity of these systems. Here, in this study, we combine small-angle X-ray scattering (SAXS), X-ray photon correlation spectroscopy (XPCS), and molecular dynamics simulation to reveal a comprehensive picture of structure at the nano- and mesoscale in these complex solutions. For the representative case of rare earth extraction from an acidic aqueous phase by a malonamide extractant in dodecane, we investigate a wide range of process-relevant extractant and acid concentrations to provide a complete picture of how aggregation depends on composition. We decompose organic phase structure from SAXS into two contributions, which together can capture the scattering at all compositions: composition fluctuations described by the Ornstein–Zernike equation at low wavenumber Q, and nanostructure modeled by a “pre-peak” at intermediate Q. The former contains information about the thermodynamics of demixing, while the latter reflects nanoscopic self-assembly of the extractant and extracted solutes. While fluctuations have typically not been considered in the literature, we find they in fact dominate the total structure for nearly all practical conditions. As only the fluctuations have a strong temperature response, we confirm this attribution with temperature-dependent SAXS measurements, including for extracted europium nitrate complexes. SAXS and XPCS measurements near the critical point find static and dynamic scaling consistent with theory. Overall, this new paradigm for understanding LLE organic phases connects composition, nanoscale, and mesoscale structuring to phase behavior, providing both a comprehensive picture of solution structure and a quantitative link between aggregation and third phase formation.

Peroutka, Allison A. [Argonne National Laboratory ↗
DOE OSTI2026

Complex spin structure in co-trimer-chain Li 2 Co 3 Se 4 O 12

Complex magnetic materials are extremely attractive for revealing unconventional spin states and novel magnetic excitations. Here, we report the structural, thermodynamic, and magnetic properties of a novel magnetic material Li 2 Co 3 Se 4 O 12 based on x-ray and neutron diffraction, specific heat, magnetization, and x-ray photoelectron spectroscopy measurements. X-ray and neutron diffraction refinements reveal two Co sites Co (1) and Co (2) even though both are in the octahedral environment. While they are not connected along the b and c directions, these octahedra are edge-shared forming the Co (2) – Co (1) – Co (2) trimer chain along the a direction. The magnetic susceptibility exhibits the Curie-Weiss (CW) temperature dependence at high temperatures (above ∼50 K) with the negative CW temperature, a dip centered at T ⁎ ∼ 8.0 K, and an antiferromagnetic transition at T N = 3.3 K. The specific heat confirms that there is a phase transition at T N and a hump at T ⁎ . The long-range magnetic transition at T N implies that, in addition to the intra-chain interaction, there is strong inter-chain interaction, which is likely due to polarized SeO 3 bridging between chains. Single crystal neutron diffraction refinement reveals a complex magnetic structure with the angle between Co (1) and Co (2) moments ∼105°. Within the Co (2) – Co (1) – Co (2) trimer, two Co (2) moments are parallelly aligned. Surprisingly, the Co (1) moment (1.92μB) is only half of the Co (2) moment (3.96μB). There is likely the spin-state change for Co (1) from the high-spin state at T > T ⁎ to the low-spin state at T < T ⁎ , causing a dip in the magnetic susceptibility and a hump in the specific heat. When the magnetic field is applied, multiple metamagnetic transitions are found in all directions, implying field-driven magnetic excitations. Our results demonstrate rich magnetic properties of Li 2 Co 3 Se 4 O 12 that are sensitive to the external stimuli such as the magnetic field.

Antiferromagnetism↗
DOE OSTI2025

Equivariant Graph Attention Network - 3D Conformers & Feature Fusion

EGAN-3F (Equivariant Graph Attention Network - 3D Conformers & Feature Fusion) presents an innovative approach for predicting binding affinity between small molecules and protein targets, a fundamental task in drug discovery. Traditional structure-based methods often depend on protein-ligand complex structures obtained from crystallography or molecular docking. In contrast, ligand-only machine learning models using 1D or 2D representations such as SMILES have been developed to predict binding affinity without structural information about the target; however, their accuracy is often limited due to the lack of 3D ligand information. EGAN-3F addresses this limitation by integrating spatially aware graph learning with traditional descriptor-based features. We systematically investigate how combining 2D and 3D molecular representations enhances binding affinity prediction from SMILES strings. This approach underscores the importance of modeling conformational diversity and incorporating chemically meaningful descriptors to improve predictive accuracy. The key innovation of EGAN-3F lies in its ability to achieve robust ligand-based binding affinity predictions without requiring protein-ligand complex structures, effectively bridging the gap between purely structural and ligand-only modeling paradigms.

Shim, Heesung [Lawrence Livermore National Laborat↗
DOE OSTI2025

Constrained GAN-Generated X-Ray CT Data For Self-Supervised And Foundation-Model Segmentation Of Concrete Microstructures

Three-dimensional characterization of materials using X-ray computed tomography (XCT) is challenging due to the complexity of internal structures, noise, and variations in resolution. Traditional computer vision models often struggle to accurately segment these images, particularly in domain-specific applications like materials science. While supervised deep learning approaches have been developed to address the limitations of conventional algorithms, they typically require large amounts of labeled training data and often fail to generalize across different datasets. Self-supervised, few-and zero-shot learning methods have gained prominence in natural image processing and segmentation tasks, but their application to scientific imaging remains limited due to the unique structural complexity, noise, and textural artifacts present in materials science data. In this work, we investigate how domain adaptation, leveraging physics-based and GAN-generated synthetic data, impacts segmentation performance. We introduce a modified Contrastive Unpaired Translation (CUT) model designed to generate realistic labeled data, which can be used for training, pre-training, and fine-tuning segmentation models for real XCT microstructure data. We evaluate the performance of two segmentation approaches: a self-supervised network (SSL-ALPNet) and a foundation model (Segment Anything Model), assessing their improvements when pre-trained and/or fine-tuned on the synthesized data. Our results demonstrate that leveraging synthetic data significantly enhances segmentation performance, particularly in challenging materials science applications.

Ziabari, Amir [ORNL] (ORCID:000000034776457X)↗
DOE OSTI2025

Trans-Influence in Dinuclear Pt(III) Complexes: Electronic Structure, σ-Donation, and Pt–Pt Spin–Spin Coupling

This study investigates the trans influence in dinuclear platinum(III) complexes using a combined approach of ab initio molecular dynamics and natural localized molecular orbital (NLMO) analysis. Focusing on pivalamidate-bridged Pt III complexes with axial ligands of varying σ-donation strength, it is quantified how ligand−metal interactions propagate through the Pt−Pt bond, and how they affect bond polarization, axial water coordination, and 1 J PtPt spin−spin coupling constants. NLMO analysis reveals quantitatively that strong σ-donating ligands polarize the Pt−Pt bond, shifting the electron density toward the opposite platinum center. The polarization mechanism is identified as the primary reason for the observed reduction of 1 J PtPt , because the bond polarization diminishes the transmission of the nuclear magnetic spin-induced electron spin density through the Pt−Pt bond. Additionally, the destabilization of axial water coordination at the opposite Pt site can be rationalized through a polarizationinduced Pt IV − Pt II -like mixed-valence character.

Ab initio molecular dynamics↗
DOE OSTI2025

Protein-ligand binding affinity prediction using multi-instance learning with docking structures

Recent advances in 3D structure-based deep learning approaches demonstrate improved accuracy in predicting protein-ligand binding affinity in drug discovery. These methods complement physics-based computational modeling such as molecular docking for virtual high-throughput screening. Despite recent advances and improved predictive performance, most methods in this category primarily rely on utilizing co-crystal complex structures and experimentally measured binding affinities as both input and output data for model training. Nevertheless, co-crystal complex structures are not readily available and the inaccurate predicted structures from molecular docking can degrade the accuracy of the machine learning methods. We introduce a novel structure-based inference method utilizing multiple molecular docking poses for each complex entity. Our proposed method employs multi-instance learning with an attention network to predict binding affinity from a collection of docking poses. We validate our method using multiple datasets, including PDBbind and compounds targeting the main protease of SARS-CoV-2. The results demonstrate that our method leveraging docking poses is competitive with other state-of-the-art inference models that depend on co-crystal structures. This method offers binding affinity prediction without requiring co-crystal structures, thereby increasing its applicability to protein targets lacking such data.

97 MATHEMATICS AND COMPUTING↗