Search NASA⌕ Search

SEARCH · Search NASA

Results for “Atomistic models”

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 OSTI2022

Te/CdTe and Al/CdTe Interfacial Energy Band Alignment by Atomistic Modeling

A synergistic approach that incorporates first-principles atomistic modeling with numerical device simulations is used to systematically evaluate the role of heterointerfaces within metal-chalcogenide-based photovoltaic technologies. Two interfaces involving either a tellurium back contact or aluminum back electrode combined with a cadmium telluride absorber layer within cadmium-telluride-based thin-film solar cells are investigated on an atomic scale to determine the mechanisms contributing to variations in device performance. Electronic structures and predicted charge transport behavior with respect to cadmium and tellurium termination of the absorber layer are studied along the polar oriented CdTe{111} facets. The computational methodology reveals a noticeable contrast between the Schottky barrier forming Al/CdTe interface versus the Type I Te/CdTe heterojunction. Greater band bending features are exhibited by the cadmium termination as opposed to the tellurium termination for each interface case. Subsequent device modeling suggests that 3.6% higher photovoltaic conversion efficiency is achievable for the cadmium termination relative to the tellurium termination of the Te/CdTe interface. Based strictly on an idealistic representation, both interface models show the importance of atomic-scale interfacial properties for cadmium telluride solar cell device performance with their bulk properties being validated in comparison to published experimental data. Here, the synergistic approach offers a suitable method to analyze solar cell interfaces through a predictive computational framework for the engineering and optimization of metal-chalcogenide-based thin-film photovoltaic technologies.

Green’s function↗
DOE OSTI2021

Atomistic modeling of meso -timescale processes with $\mathrm{SEAKMC}$: A perspective and recent developments

On-the-fly kinetic Monte Carlo (kMC) methods have recently garnered significant attentions after successful applications to various atomic-scale problems using a timescale outside the reach of classical molecular dynamics. These methods play a critical role in modeling atomistic meso-timescale processes, and it is therefore essential to further improve their capabilities. In this report we review one of the on-the-fly kMC methods, Self-Evolving Atomistic kinetic Monte Carlo (SEAKMC) and propose two schemes that considerably enhance the efficiency of saddle point searches (SPSs) during the simulations. The performance of these schemes is tested using the diffusion of point defects in bcc Fe. In addition, we discuss approaches to significantly mitigate limitations of these schemes, which further improves their efficiencies. Importantly, these schemes improve the SPS efficiency not only for SEAKMC but also for other on-the-fly kMC methods, broadening the applications of on-the-fly kMC simulations to complex meso-timescale problems.

36 MATERIALS SCIENCE↗
DOE OSTI2013

The putative liquid-liquid transition is a liquid-solid transition in atomistic models of water. II

This paper extends our earlier studies of free energy functions of density and crystalline order parameters for models of supercooled water, which allows us to examine the possibility of two distinct metastable liquid phases [D. T. Limmer and D. Chandler, J. Chem. Phys.135, 134503 (2011) and preprint http://arxiv.org/abs/arXiv:1107.0337 (2011)]. Low-temperature reversible free energy surfaces of several different atomistic models are computed: mW water, TIP4P/2005 water, Stillinger-Weber silicon, and ST2 water, the last of these comparing three different treatments of long-ranged forces. In each case, we show that there is one stable or metastable liquid phase, and there is an ice-like crystal phase. The time scales for crystallization in these systems far exceed those of structural relaxation in the supercooled metastable liquid. We show how this wide separation in time scales produces an illusion of a low-temperature liquid-liquid transition. The phenomenon suggesting metastability of two distinct liquid phases is actually coarsening of the ordered ice-like phase, which we elucidate using both analytical theory and computer simulation. For the latter, we describe robust methods for computing reversible free energy surfaces, and we consider effects of electrostatic boundary conditions. We show that sensible alterations of models and boundary conditions produce no qualitative changes in low-temperature phase behaviors of these systems, only marginal changes in equations of state. On the other hand, we show that altering sampling time scales can produce large and qualitative non-equilibrium effects. Recent reports of evidence of a liquid-liquid critical point in computer simulations of supercooled water are considered in this light.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2019

Machine Learning Applied to a Variable Charge Atomistic Model for Cu/Hf Binary Alloy Oxide Heterostructures [Machine Learning a Variable Charge Atomistic Model for Cu/Hf Binary Alloy Oxide Heterostructures]

Alloy oxide heterostructures are essential for a wide variety of energy applications, including anticorrosion coatings, bifunctional catalysis, energy storage, as well as emerging platforms for multilevel nonvolatile memory and neuromorphic devices. The key role played by oxide composition, density, and stoichiometry in governing electrochemical reactions, interface structure, and ionic transport in alloy oxides remains largely unknown. This is primarily due to the lack of variable charge interatomic potential models that can adequately describe the various atomic/ionic interactions in alloy oxides within a unified framework. Here, we introduce a charge transfer ionic potential (CTIP) model for Cu/Hf/O alloy system and demonstrate its ability to accurately capture the complex potential energy surface owing to dynamically varying interactions, including metallic (Cu/Hf), ionic (Cu/Hf/Ox), mixed environment (interfaces). We leverage supervised machine learning methods powered by genetic algorithms coupled with local simplex optimization to efficiently navigate through a high-dimensional parameter space and identify an optimum set of an independent set of CTIP parameters. We train our model against an extensive first-principles based data set that includes lattice constants, cohesive energies, equations of state, and elastic constants of various experimentally observed polymorphs of hafnia, copper oxide, hafnium, as well as copper. Our machine-learned CTIP model captures the structure, elastic properties, thermodynamics, and energetic ordering of various polymorphs. It also accurately predicts the surface properties of both oxides and their metals. To demonstrate the suitability of our CTIP model for investigating dynamic processes, we employed it to identify the atomistic scale mechanisms associated with the initial nanoscale oxidation and surface oxide growth on Cu-doped hafnia surfaces. Here, our machine-learned CTIP model can be used to probe the dynamic response of Cu/Hf/O alloy interfaces subjected to external stimuli (e.g., electric field, pressure, temperature, strain, etc.) and a variety of atomistic phenomena including the dynamics of switching in emerging neuromorphic platforms.

36 MATERIALS SCIENCE↗
DOE OSTI2013

Atomistic Modeling of the Electrode–Electrolyte Interface in Li-Ion Energy Storage Systems: Electrolyte Structuring

The solid electrolyte interface (SEI) forms as a result of side reactions between the electrolyte and electrode surfaces in Li-ion batteries and can adversely impact performance by impeding Li-ion transport and diminishing the storage capacity of the battery. To gain a detailed understanding of the impact of the SEI on electrolyte structure, atomistic molecular dynamics simulations of the electrode/electrolyte interface were performed in the presence and absence of the SEI under applied voltages. The composition of the SEI was guided by a wealth of data from experiments and allowed to vary across the simulations. A novel computational approach was implemented that showed significant computational speedup compared to fully polarizable electrode simulations, yet, retained the correct qualitative physics for the electrolyte. A force-matching algorithm was used to construct a new force field for the pure electrolyte, LiPF6 in ethylene carbonate, which was developed from ab initio molecular dynamics simulations. The electrode/electrolyte interface was included using a simple, physically motivated model, which includes the polarization of the conducting graphitic electrode by the electrolyte and the application of an external voltage. Changes in the structure of the electrolyte at the interface as a function of applied voltage, the thickness of the SEI layer, and composition of the SEI provide molecular level insight into the species present at these interfaces and potential clues to the effect of the SEI on transport. It is noted that, with increasing SEI thickness and LiF content, lithium ions are drawn closer to the SEI surface, which implies that these interfaces favor desolvation and promote more rapid lithium transport.

Jorn, Ryan↗
DOE OSTI2025

Scaling Laws of Graph Neural Networks for Atomistic Materials Modeling

Atomistic materials modeling is a critical task with wide-ranging applications, from drug discovery to materials science, where accurate predictions of the target material property can lead to significant advancements in scientific discovery. Graph Neural Networks (GNNs) represent the state-of-the-art approach for modeling atomistic material data thanks to their capacity to capture complex relational structures. While machine learning performance has historically improved with larger models and datasets, GNNs for atomistic materials modeling remain relatively small compared to large language models (LLMs), which leverage billions of parameters and terabyte-scale datasets to achieve remarkable performance in their respective domains. To address this gap, we explore the scaling limits of GNNs for atomistic materials modeling by developing a foundational model with billions of parameters, trained on extensive datasets in terabytescale. Our approach incorporates techniques from LLM libraries to efficiently manage large-scale data and models, enabling both effective training and deployment of these large-scale GNN models. This work addresses three fundamental questions in scaling GNNs: the potential for scaling GNN model architectures, the effect of dataset size on model accuracy, and the applicability of LLM-inspired techniques to GNN architectures. Specifically, the outcomes of this study include (1) insights into the scaling laws for GNNs, highlighting the relationship between model size, dataset volume, and accuracy, (2) a foundational GNN model optimized for atomistic materials modeling, and (3) a GNN codebase enhanced with advanced LLM-based training techniques. Our findings lay the groundwork for large-scale GNNs with billions of parameters and terabyte-scale datasets, establishing a scalable pathway for future advancements in atomistic materials modeling.

Li, Chaojian [ORNL] (ORCID:0000000340309777)↗
DOE OSTI2026

Multi-task Parallelism for Robust Pre-training of Graph Foundation Models on Multi-source, Multi-fidelity Atomistic Modeling Data

Graph foundation models using graph neural networks promise sustainable, efficient atomistic modeling. To tackle challenges of processing multi-source, multi-fidelity data during pre-training, recent studies employ multi-task learning, in which shared message passing layers initially process input atomistic structures regardless of source, then route them to multiple decoding heads that predict data-specific outputs. This approach stabilizes pre-training and enhances a model’s transferability to unexplored chemical regions. Preliminary results on approximately four million structures are encouraging, yet questions remain about generalizability to larger, more diverse datasets and scalability on supercomputers. We propose a multi-task parallelism method that distributes each head across computing resources with GPU acceleration. Implemented in the open-source HydraGNN architecture, our method was trained on over 24 million structures from five datasets and tested on the Perlmutter, Aurora, and Frontier supercomputers, demonstrating efficient scaling on all three highly heterogeneous super-computing architectures.

Lupo Pasini, Massimiliano [ORNL] (ORCID:0000000249↗
DOE OSTI2012

Atomistic Modeling of Corrosion Events at the Interface between a Metal and Its Environment

Atomistic simulation is a powerful tool for probing the structure and properties of materials and the nature of chemical reactions. Corrosion is a complex process that involves chemical reactions occurring at the interface between a material and its environment and is, therefore, highly suited to study by atomistic modeling techniques. In this paper, the complex nature of corrosion processes and mechanisms is briefly reviewed. Various atomistic methods for exploring corrosion mechanisms are then described, and recent applications in the literature surveyed. Several instances of the application of atomistic modeling to corrosion science are then reviewed in detail, including studies of the metal-water interface, the reaction of water on electrified metallic interfaces, the dissolution of metal atoms from metallic surfaces, and the role of competitive adsorption in controlling the chemical nature and structure of a metallic surface. Some perspectives are then given concerning the future of atomistic modeling in the field of corrosion science.

Taylor, Christopher D.↗
NASA NTRS2002

Atomistic Modeling of RuAl and (RuNi) Al Alloys

Atomistic modeling of RuAl and RuAlNi alloys, using the BFS (Bozzolo-Ferrante-Smith) method for alloys is performed. The lattice parameter and energy of formation of B2 RuAl as a function of stoichiometry and the lattice parameter of (Ru(sub 50-x)Ni(sub x)Al(sub 50)) alloys as a function of Ni concentration are computed. BFS based Monte Carlo simulations indicate that compositions close to Ru25Ni25Al50 are single phase with no obvious evidence of a miscibility gap and separation of the individual B2 phases.

Gargano, Pablo↗
DOE OSTI2014

Experimentally driven atomistic model of 1,2 polybutadiene

We present an efficient method of combining wide angle neutron scattering data with detailed atomistic models, allowing us to perform a quantitative and qualitative mapping of the organisation of the chain conformation in both glass and liquid phases. The structural refinement method presented in this work is based on the exploitation of the intrachain features of the diffraction pattern and its intimate linkage with atomistic models by the use of internal coordinates for bond lengths, valence angles, and torsion rotations. Atomic connectivity is defined through these coordinates that are in turn assigned by pre-defined probability distributions, thus allowing for the models in question to be built stochastically. Incremental variation of these coordinates allows for the construction of models that minimise the differences between the observed and calculated structure factors. We present a series of neutron scattering data of 1,2 polybutadiene at the region 120–400 K. Analysis of the experimental data yields bond lengths for Cî—¸C and C î—» C of 1.54 Å and 1.35 Å, respectively. Valence angles of the backbone were found to be at 112° and the torsion distributions are characterised by five rotational states, a three-fold trans-skew± for the backbone and gauche± for the vinyl group. Rotational states of the vinyl group were found to be equally populated, indicating a largely atactic chan. The two backbone torsion angles exhibit different behaviour with respect to temperature of their trans population, with one of them adopting an almost all trans sequence. Consequently, the resulting configuration leads to a rather persistent chain, something indicated by the value of the characteristic ratio extrapolated from the model. We compare our results with theoretical predictions, computer simulations, RIS models and previously reported experimental results.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗