Search NASA⌕ Search

SEARCH · Search NASA

Results for “atomic structure generation”

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 163 records · Page 9
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Study of self-assembly behavior and ionic conductivity of conjugated liquid crystals with T-shaped facial-polyphilic structure

The unique self-assembly of liquid crystals (LCs), combined with their potential application as organic semiconductors, has become a focus of recent research. Here, a joint experimental and computational study of the self-assembly and ionic conduction was carried out on a series of T-shaped conjugated LCs consisting of three incompatible components. By extending the EOn side-chain length, several experimental evaluations confirmed a decrease of the order-disorder transition temperature, while coarse-grained simulations revealed a structural evolution from a smectic phase to a columnar phase. Ionic conductivity of these molecules was achieved by adding Li salt, leading to a maximum conductivity of 1.1 × 10 −3 siemens per centimeter observed at 120°C. All-atom simulations were performed to examine the Li-ion solvation environment and to evaluate the intrachain and interchain Li-ion hopping mechanisms. The molecule with a long EOn side chain was found to generate a densely distributed network of Li-ion solvation sites, which can facilitate effective interchain hopping to promote ion transport.

Liu, Ziwei [Cornell Univ., Ithaca, NY (United Stat↗
DOE OSTI2025

Reweighting configurations generated by transferable, machine learned models for protein sidechain backmapping

Multiscale modeling requires the linking of models at different levels of detail, with the goal of gaining accelerations from lower fidelity models while recovering fine details from higher resolution models. Communication across resolutions is particularly important in modeling soft matter, where tight couplings exist between molecular-level details and mesoscale structures. While multiscale modeling of biomolecules has become a critical component in exploring their structure and self-assembly, backmapping from coarse-grained to fine-grained, or atomistic, representations presents a challenge, despite recent advances through machine learning. A major hurdle, especially for strategies utilizing machine learning, is that backmappings can only approximately recover the atomistic ensemble of interest. We demonstrate conditions for which backmapped configurations may be reweighted to exactly recover the desired atomistic ensemble. By training separate decoding models for each sidechain type, we develop an algorithm based on normalizing flows and geometric algebra attention to autoregressively propose backmapped configurations for any protein sequence. Critical for reweighting with modern protein force fields, our trained models include all hydrogen atoms in the backmapping and make probabilities associated with atomistic configurations directly accessible. We also demonstrate, however, that reweighting is extremely challenging despite state-of-the-art performance on recently developed metrics and generation of configurations with low energies in atomistic protein force fields. Through detailed analysis of configurational weights, we show that machine-learned backmappings must not only generate configurations with reasonable energies, but also correctly assign relative probabilities under the generative model. These are broadly important considerations in generative modeling of atomistic molecular configurations.

Monroe, Jacob I. [Univ. of Arkansas, Fayetteville,↗
DOE OSTI2025

Abbr. Final report: Self-assembled molecular containers as artificial water channels: towards biomimetic desalination membranes

Global water scarcity demands advances in desalination technologies that can deliver more fresh water with less energy. Current reverse osmosis membranes are fundamentally limited by a trade-off between how much water they can pass and how well they block salts. To address this challenge, we developed a bottom-up strategy to design and test artificial water channels that mimic the efficiency of biological proteins but are built from robust synthetic molecules. Over two years, we synthesized and evaluated more than twenty molecular channel candidates, including supramolecular macrocycles and nanographene pores with atomically precise structures. We showed that small chemical modifications allow direct control over pore size and chemistry, which in turn govern water permeability and salt rejection. In collaboration with university partners, we reported the first experimental demonstration of water transport through a nanographene pore, bridging a long-standing gap between simulation and experiment. Several of the artificial channels we developed achieved water–salt selectivity beyond conventional polymer membranes, highlighting their potential for next-generation desalination and precision separations.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Small Molecule Inhibitor-Modulated Al 2 O 3 Atomic Layer Deposition on Monolayer MoS 2 for Controlled Nucleation

The integration of ultrathin dielectrics on two-dimensional (2D) semiconductors is essential for advancing beyond-Si electronics. However, the intrinsic inertness of van der Waals 2D basal planes remains a primary bottleneck to achieving uniform dielectric nucleation and growth. Here, in this study, we introduce a small molecule inhibitor (SMI)-modulated thermal atomic layer deposition (ALD) strategy, exemplified by aluminum oxide (Al 2 O 3 ) ALD on monolayer molybdenum disulfide (1L MoS 2 ) with acetic acid (HAc) SMI. The ABC-type sequence comprises HAc inhibitor (A), trimethylaluminum (TMA) precursor (B), and deionized H 2 O coreactant (C). In situ quartz crystal microbalance (QCM) studies reveal robust HAc adsorption on Al 2 O 3 and suppression of subsequent oxide growth on HAc-passivated surfaces. When applied to 1L MoS 2 , this inhibitory pathway enables HAc to selectively passivate nascent Al 2 O 3 nuclei formed on the MoS 2 surface, limiting their three-dimensional (3D) island coarsening and redirecting precursor adsorption toward the uncovered basal plane. Consequently, nearly continuous ultrathin (∼1.5 nm) Al 2 O 3 films are achieved on 1L MoS 2 with markedly improved uniformity compared to standard Al 2 O 3 ALD using TMA and H 2 O, as validated by atomic force microscopy (AFM), cross-sectional scanning transmission electron microscopy (STEM), and energy-dispersive X-ray spectroscopy (EDS). Density functional theory (DFT) calculations further provide atomistic insight into HAc-modulated Al 2 O 3 nucleation, corroborating the energetic preference of HAc for Al 2 O 3 over MoS 2 and attenuated TMA adsorption on HAc-passivated surfaces. Spatially resolved Raman spectroscopy also confirms that the HAc-modulated process preserves the structural integrity of 1L MoS 2 , with only minimal strain and doping perturbations observed after dielectric deposition. This SMI-modulated approach offers a broadly applicable framework for controlling ALD nucleation across various inhibitors, ALD chemistries, and 2D materials, opening opportunities for reliable dielectric integration in next-generation nanoelectronics.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Crystalline Order Yet Glass-Like Heat Transport Driven by Hidden Local Distortions as the Structural Origin of Ultralow Thermal Conductivity in AgErTe 2

Given their rich chemical diversity and the interplay among the p-, d-, and f-orbitals of chalcogens, transition metals, and lanthanides, respectively, rare-earth transition-metal chalcogenides exhibit a wide variety of structural, magnetic, and transport phenomena. As a result, they form a particularly appealing platform for investigating structure–property relationships, emergent electronic and magnetic behaviors, and thermal transport. Here we investigate AgErTe 2 as a model system to understand phonon-glass behavior in ordered crystalline solids, which establishes the design principles for thermal barrier coatings and next-generation thermoelectrics. The local bonding asymmetry and lattice softness suppress the inherently low lattice thermal conductivity, resembling the characteristics of amorphous materials. This suppression is significantly influenced by local off-centering of Ag atoms, which breaks lattice periodicity while maintaining global crystallinity. The presence of antibonding states just below the Fermi level, arising from Ag 4d and Te 5p orbital interactions, leads to lattice softening and destabilizes ideal tetrahedral coordination, resulting in a pseudo Jahn–Teller distortion. Furthermore, the coexistence of weaker, more polarizable Ag–Te bonds and stronger Er–Te bonds creates a complex vibrational landscape enriched with low-frequency modes and enhanced phonon scattering. A pronounced disparity in interatomic force constants gives rise to highly localized, low-energy optical phonons linked to Ag rattling. These flat vibrational modes exhibit strong coupling with transverse acoustic phonons, resulting in ultrashort phonon lifetimes and mean free paths approaching interatomic distances. These features collectively enhance phonon scattering across a broad range of length and energy scales. This work offers a framework for engineering suppressed thermal conductivity in crystalline systems without the introduction of alloying elements.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Additive manufacturing of amorphous metal soft magnetic composites

Soft magnet alloys are used as magnetic cores for electric motors, transformers, wind turbines and other power generation systems. Soft magnetic cores are expensive and time consuming to manufacture in the complex shapes required for next-generation devices using conventional press and sinter powder metallurgy. The objective of this effort is to additively manufacture high performance soft magnets, with reduced cost and reduced material waste and 10x lower energy (core) loss at the high operating frequencies of many electric machines. Laser powder bed fusion additive manufacturing is used as the fabrication method. Electrical steel and amorphous alloys atomized powders are used as a feedstock materials. Magnetic cores are printed in topology optimized structures such as the Hilbert curve because this has been shown to reduce energy losses by minimizing the eddy currents that circulate within the magnet at high frequencies. In our project, we succeed in printing FeSi 3.5wt% and FeSi 6.5wt% electrical steels, and iron-based soft magnetic amorphous alloys in the shape of Hilbert and Peano curve topology optimized structures. We found that the Peano curve has a higher cut-off frequency than the Hilbert curve, and that amorphous alloys have high cut-off frequencies and higher mechanical hardness than electrical steels. Processing conditions such as laser power, scan speed, and hatching pattern were optimized to achieve high density prints, and optimize magnetic performance. We find that the printing of amorphous alloy soft magnetic cores may be technoeconomically feasible for large scale applications such as transformers, for which supply chain issues and the labor costs of manual fabrication of magnetic cores is prohibitive in some cases.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Absence of hydrogen insertion into highly crystalline superconducting infinite layer nickelates

Here, the discovery of superconductivity in the infinite layer nickelates introduced a materials system analogous to the cuprates for the study of unconventional superconductivity. The synthesis of infinite layer nickelates, (RNiO 2 , R = lanthanide) often uses calcium hydride (CaH 2 ) to facilitate the deintercalation of apical site oxygen atoms from a precursor perovskite (RNiO 3 ) phase via topotactic reduction. However, it remains uncertain whether the use of CaH 2 results in the insertion of hydrogen into the infinite layer structure, and if it does, what the implications are for superconductivity. To quantify the hydrogen composition of highly crystalline infinite layer nickelates, we synthesized Nd 1-x⁢ Sr x ⁢NiO 2 thin films on LSAT substrates and conducted time-of-flight secondary ion mass spectroscopy measurements to generate hydrogen depth profiles. We compare the hydrogen density of nickelates prepared with and without a SrTiO 3 capping layer. Additionally, we measure the hydrogen content in nickelate samples at various doping levels spanning the superconducting phase space, including the underdoped, optimally doped, and overdoped regime. We report no significant increase in hydrogen density between the perovskite and infinite layer phases in any of the measured samples. Furthermore, we put an upperbound on the hydrogen concentration of our nickelate samples to Nd 1-x Sr x ⁢NiO 2 ⁢H 0.05 . Our results imply that hydrogen is not responsible for the emergence of superconductivity in the infinite layer nickelates.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2026

Prediction and Experimental Verification of Electrolyte Solvation Structure from an OMol25-Trained Interatomic Potential

A molecular-level understanding of electrolyte solvation structure and ion–ion correlations is critical to developing next-generation battery chemistries. Atomistic simulation capabilities with sufficient accuracy, speed, and transferability to deliver reliable structural insights while avoiding arduous system-specific reparameterization are thus highly desirable. Machine learning interatomic potentials (MLIPs) trained on large, chemically diverse data sets are revolutionizing computational chemistry, enabling molecular dynamics simulations of battery electrolytes with near-DFT accuracy over 10,000× faster than DFT. While previous MLIP training data sets with suitable elemental coverage for electrolytes have been based on inorganic materials, the Open Molecules 2025 (OMol25) data set provides large-scale molecular DFT MLIP training data with broad elemental coverage and specifically samples tens of millions of electrolyte configurations. Here, we integrate computational modeling with experimental validation to systematically assess the ability of large-scale MLIPs pretrained on materials data or on OMol25 to accurately resolve nanoscale structural organization and ion-solvation characteristics in Na-ion battery electrolytes across diverse physicochemical conditions and compositional regimes. We find that the OMol25-trained Universal Model of Atoms (UMA-OMol) predicts experimentally measured densities and X-ray structure factors in substantially better agreement compared to state-of-the-art models trained only on inorganic materials data. Using UMA-OMol, we further analyze systematic trends in solvation structure as a function of cation identity, anion chemistry, salt concentration, and solvent topology. We observe that increasing system temperature amplifies the heterogeneity within the solvation environment, perturbing cation–solvent interactions and promoting the formation of contact ion pairs (CIPs). Moreover, subtle variations in the solvent topology of glyme-based electrolytes cause pronounced changes in ion correlations and solvation structure. The experimental agreement and microscopic insights shown here position OMol25-trained MLIPs as a practical route to predictive, high-throughput electrolyte simulations beyond the limits of classical force fields and direct DFT molecular dynamics, serving as a powerful tool for accelerating the design of next-generation Na-ion battery electrolytes and beyond.

MLIPs↗
DOE OSTI2025

Understanding how defects and dopant atoms in copper surface oxides affect reactivity

Copper and its oxides are key catalytic materials, on which reactions often occur at the metal/oxide interface. Here, in this work, we directly connect the induction period observed during methanol-driven reduction of thin-film copper oxides to their atomic-scale structural order. Using temperature-programmed desorption (TPD) methanol titrations combined with scanning tunneling microscopy, we show that highly ordered oxide phases – particularly the “29” structure with its low defect density – exhibit long induction periods and initially low reactivity. The induction period, defined as the number of methanol TPD cycles required to reach half of the maximum formaldehyde yield, scales with oxide order and oxygen coverage. Enhanced reactivity of well-ordered oxides emerges only after repeated methanol adsorption/desorption cycles generate oxygen vacancies and new Cu(111)/Cu x O interfacial sites. In contrast, disordered or sub-stoichiometric oxides, which contain more intrinsic defects and interfaces, are active from the first TPD cycle. We further examine how dilute Pt and Rh dopants influence oxide order and reactivity: 1% Pt increases defect density and catalytic activity, while 1% Rh promotes oxide ordering and longer induction periods. These findings demonstrate that dilute alloying provides a potential method for tuning the structure and reactivity of Cu(111)/Cu x O interfaces.

Cu(111)Methanol oxidation↗
DOE OSTI2024

Melting and transport properties of Al 2⁢ O 3 at extreme conditions

The high-pressure, temperature phase diagram and transport properties of materials are of broad interest to planetary sciences and high-energy-density sciences and applications. Alumina (Al 2 O 3 ) or its various forms (e.g., solutions with other oxides or silicates) are important constituents in Earth’s and super-Earths’ mantles, common window materials in dynamic compression experiments, and standard pressure calibrators in diamond-anvil-cell experiments. Its structures and transport properties are of particular importance but have not been well studied at above 100 GPa pressures. Based on extensive first-principles molecular dynamics calculations, we obtain atomic level insights on structural differences and accurate results on electrical conductivity (σ dc ), thermal conductivity (κ), and reflectivity of Al 2 O 3 in several phases at pressures of ~130–1300 GPa and temperatures of 4000–20,000 K. We find the solid-to-liquid state changes of Al 2 O 3 are accompanied by an insulator-to-semimetal transition with enhanced σ dc and κ, which is similar to other silicates and oxides and can facilitate magnetic field generation in deep interiors of rocky planets. We also find the solid-to-solid transition (from Rh 2 O 3 (II) to CaIrO 3 ) is accompanied by negligible changes in σ dc and κ at 160 GPa, contrary to previous expectations, whereas another higher pressure phase transition (CaIrO 3 to U 2 S 3 at 450 GPa) is associated with increases in σ dc and κ. Furthermore, we show the transport properties can be enhanced by oxygen vacancy in Al 2 O 3 , but are not significantly affected by forming solutions with MgSiO 3 , particularly when in the solid states.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2024

Constrained patterning of orientated metal chalcogenide nanowires and their growth mechanism

One-dimensional metallic transition-metal chalcogenide nanowires (TMC-NWs) hold promise for interconnecting devices built on two-dimensional (2D) transition-metal dichalcogenides, but only isotropic growth has so far been demonstrated. Here we show the direct patterning of highly oriented Mo 6 Te 6 NWs in 2D molybdenum ditelluride (MoTe 2 ) using graphite as confined encapsulation layers under external stimuli. The atomic structural transition is studied through in-situ electrical biasing the fabricated heterostructure in a scanning transmission electron microscope. Atomic resolution high-angle annular dark-field STEM images reveal that the conversion of Mo 6 Te 6 NWs from MoTe 2 occurs only along specific directions. Combined with first-principles calculations, we attribute the oriented growth to the local Joule-heating induced by electrical bias near the interface of the graphite-MoTe 2 heterostructure and the confinement effect generated by graphite. Using the same strategy, we fabricate oriented NWs confined in graphite as lateral contact electrodes in the 2H-MoTe 2 FET, achieving a low Schottky barrier of 11.5 meV, and low contact resistance of 43.7 Ω µm at the metal-NW interface. Our work introduces possible approaches to fabricate oriented NWs for interconnections in flexible 2D nanoelectronics through direct metal phase patterning.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Mapping the structural–mechanical landscape of amorphous carbon with ReaxFF molecular dynamics

We use ReaxFF molecular dynamics (MD) to investigate the relationship between structural and mechanical properties in bulk and nanostructured amorphous carbon (a-C). The liquid-quench MD method is used to generate isotropic bulk samples with mass densities ranging from 0.96 to 3.29 g/cm3. Structural analysis identifies two types of structures with distinct short- and medium-range order: lower-density sp2-dominated a-C, which is characterized by a bimodal ring-size distribution, and higher-density sp3-dominated tetrahedral amorphous carbon (ta-C), exhibiting a unimodal ring-size distribution. Stress–strain MD simulations and analysis reveal how an atomistic structure impacts elastic properties and post-yield atomic rearrangements. All stretched structures demonstrate elastic isotropy and plasticity driven by a ring-size expansion mechanism reflected in changes in ring statistics. The plastic region is substantially larger in ta-C than in a-C due to the post-yield shift from sp3 to sp2 C dominant bonding. In both a-C and ta-C, ultimate failure occurs when a reactive crack, traversed by long sp chains, forms and propagates predominantly perpendicular to the direction of the applied strain. Oxygen infiltration into the fractured region significantly reduces stress resistance, primarily through the early rupture of long sp chains. MD simulations and analysis are extended to a-C slabs, a-C nanotubes, and partially a-C nanotubes. The latter nanostructure highlights the differences between the elastically isotropic a-C walls, which develop circumferential cracking, and the crystalline walls, which tear along crystallographic directions. These results provide a strong foundation for further computational characterization of a-C materials.

Dernov, A. (ORCID:0009000004220973)↗
DOE OSTI2026

Polymer-Derived Amorphous Aluminosilicate Nanomembranes for H 2 Purification

Aluminosilicate zeolite membranes with robust microporous crystalline structures are attractive for the molecular separation of H 2 from light gases, but their large-scale fabrication is complicated and costly, hindering their practical applications. Herein, we present polymer-derived amorphous aluminosilicate nanomembranes that combine the exceptional processability of polymers with the superior gas separation properties of aluminosilicates. Specifically, thin-film composite membranes comprising 150 nm polydimethylsiloxane were first treated with oxygen plasma to generate 10 nm polyorganosilica (POSi) on the surface, which were then subjected to few-cycle atomic layer deposition (ALD) using trimethylaluminum as a metal precursor and water vapor as a coreactant. This scalable two-step process yields few-nanometer amorphous aluminosilicates with strong size-sieving ability. For example, three-cycle ALD treatment of POSi increases H 2 /CO 2 selectivity from 39 to 200 and H 2 /CH 4 selectivity from 190 to 500, while decreasing H 2 permeance from 990 to 210 GPU at 150 °C, superior to the state-of-the-art membranes. In conclusion, rapid and scalable manufacturing of amorphous aluminosilicate nanolayers can also be of interest for catalysis and adsorption applications.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Crystalline 1D Coordination Polymer Inhibitor Layer Leads to Vertical Sidewalls in Selectively Deposited ZnO on Nanoscale Patterns

Area-selective atomic layer deposition (AS-ALD) is a promising technique for the fabrication of next-generation nanoelectronics. There are two main challenges in AS-ALD: (1) achieving high selectivity of deposition on the growth regions, and (2) preventing mushrooming of the growth material onto the nongrowth regions and achieving well-defined interfaces. In this work, we use benzenethiol (BT) as an inhibitor in the selective deposition of ZnO on SiO 2 in the presence of copper with and without a native oxide (Cu/CuO x ). We observe that BT forms a monolayer on the Cu surface and a Cu-thiolate multilayer structure on CuO x . Using grazing incidence X-ray diffraction combined with simulations, we find that the multilayer structure is crystalline and composed of 1D coordination polymers of Cu-thiolate. Here, using ellipsometry and X-ray photoelectron spectroscopy, we show that the BT consumes the entirety of the CuO x during multilayer formation, allowing the multilayer thickness to be tuned by the thickness of the original oxide. Both the monolayer BT and the multilayer BT prove to be effective inhibitors of ZnO ALD, blocking nearly 500 ALD cycles, which is more than twice that achieved with other thiol inhibitors. Finally, we demonstrate that the multilayer structure can prevent mushrooming of the ALD material onto the nongrowth surface of nanoscale patterns, creating vertical sidewalls with well-defined material interfaces and providing excellent pattern transfer, even for a relatively thick deposited film. As such, these results demonstrate that BT is not only an effective inhibitor but also that its ability to form tunable multilayers makes it well-suited for highly precise nanopatterning applications.

Layers↗
DOE OSTI2025

Quick-and-Easy Validation of Protein–Ligand Binding Models Using Fragment-Based Semiempirical Quantum Chemistry

Electronic structure calculations in enzymes converge very slowly with respect to the size of the model region that is described using quantum mechanics (QM), requiring hundreds of atoms to obtain converged results and exhibiting substantial sensitivity (at least in smaller models) to which amino acids are included in the QM region. As such, there is considerable interest in developing automated procedures to construct a QM model region based on well-defined criteria. However, testing such procedures is burdensome due to the cost of large-scale electronic structure calculations. Here, we show that semiempirical methods can be used as alternatives to density functional theory (DFT) to assess convergence in sequences of models generated by various automated protocols. The cost of these convergence tests is reduced even further by means of a many-body expansion. We use this approach to examine convergence (with respect to model size) of protein–ligand binding energies. Fragment-based semiempirical calculations afford well-converged interaction energies in a tiny fraction of the cost required for DFT calculations. Two-body interactions between the ligand and single-residue amino acid fragments afford a low-cost way to construct a “QM-informed” enzyme model of reduced size, furnishing an automatable active-site model-building procedure. This provides a streamlined, user-friendly approach for constructing ligand binding-site models that needs neither a priori information nor manual adjustments. Extension to model-building for thermochemical calculations should be straightforward.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Similarity Metric for Data Optimization and Efficient Training of Reactive Machine Learning Force Fields for Hydrocarbon Radiolysis

Radiolysis is a common approach to sterilize polymers, chemically modify them for upcycling, and accelerate their decomposition for recycling purposes. Reactive molecular dynamics (MD) simulations provide a powerful tool to generate atomic-level trajectories of the reactive processes and quantify radiolytic chemical degradation pathways. For this, machine learning (ML) surrogate models for reactive force fields with quantum mechanical accuracy are now widely used, which require ML training data sets that can provide information on atomic environments for target chemical systems. However, radiolysis chemistry can be highly complex and diverse, which poses significant challenges for generating training data to parametrize ML models. In this regard, we developed a method for optimizing the training data set using a cosine similarity metric to help guide training set selection for radiolysis of polyethylene, a model hydrocarbon polymer, as well as to enhance the transferability of our reactive ML force field (MLFF) to a variety of molecular and polymeric systems. Our approach performs atom-by-atom comparisons between local atomic environments to pinpoint important data points associated with rare and localized events, such as radiolysis damage within structures. We apply this approach to train the Chebyshev Interaction Model for Efficient Simulation (ChIMES) MLFF model, which expresses the atomic interaction potentials in terms of linear combinations of many-body Chebyshev polynomials. We first show that our method can reduce our training set size by ∼70% while improving overall accuracy compared to more standard MD model fitting approaches. We then validate our optimum model against diverse hydrocarbon simulation data, including simple alkanes and systems with unsaturated carbon bonds, over a wide range of thermodynamic conditions. Finally, we use our ChIMES model to perform MD simulations of radiolytic damage with large-scale systems that help avoid system size effects. Overall, our approach yields an MD force field that retains most of the accuracy of the underlying quantum method while yielding many orders of improvement in computational efficiency. In conclusion, our efforts will have impact on future hydrocarbon polymer radiolysis studies, where the chemical details of the polymer–radiation interactions can have a strong effect on the resulting products observed in experiments.

Hydrocarbons↗
DOE OSTI2026

Hierarchical Epoxy Structures via Tunable Polymerization-Induced Phase Separation Combined with Additive Manufacturing

Polymerization-induced phase separation (PIPS) allows for the control of thermoset morphologies and properties, enabling the tuning of domain sizes and thermomechanical response. However, its use in generating substructural features in additively manufactured materials has been limited. In this work, we combine epoxy PIPS with UV curable acrylate and rheological modifiers to print nano- to macro-phase separating materials via a two-step, dual-cure approach. This method enables direct ink write printing of hierarchical structures with both controlled morphologies through phase separation and macroscale architecture through print design. We find that formulations for phase-separating materials require judicious incorporation of additives to enable printability and to provide sufficient green strength. Atomic force microscopy-nano infrared mapping reveals tunable, reticulated nano- to micron-scale domains of the resultant multiphase materials and their morphology changes due to additives, resulting in alterations to thermomechanical and tensile properties. Shape memory behavior is also demonstrated through multimaterial additive manufacturing of epoxies with functionally graded internal morphology using active mixing techniques, highlighting this method’s ability to fabricate complex architectures with controlled morphologies and thermomechanical response.

Van Meter, Kylie E [Organic Materials Science, San↗
DOE OSTI2026

Tuning the Coordination Environment of Rh Single Atoms on Highly Dispersed Reducible Oxides for Enhanced Reverse Water-Gas Shift Performance

Controlling the dynamic mobility of catalyst surface active sites and their interactions with the surrounding environment is critical in generating active surfaces that directly influence the catalytic activity and selectivity. Here, we report a strategy for tailoring the dispersion and electronic environment of single atom Rh catalysts by decorating the alumina support with highly dispersed (HD) cerium and molybdenum oxides. The resulting catalysts exhibit markedly different behavior in the Reverse Water Gas Shift (RWGS) reaction. In particular, Rh/MoOx(HD)/Al2O3 maintains atomically dispersed Rh even at elevated temperatures (up to 400 °C), achieving CO selectivity of up to 100% and resists sintering via the formation of a newly developed structure featuring Rh single atoms embedded in MoOx clusters. In situ spectroscopy and microscopy analyses confirm the stabilization of Rh and the dynamic evolution of Rh–Mo coordination under reaction conditions. Our findings highlight the power of support modification in steering active site structure and activity, offering a pathway toward enhanced and tunable single atom catalysts for CO2 valorization.

CO selectivity↗