Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “ab initio molecular dynamics (AIMD)”

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 OSTI2024

KCl-UCl 3 molten salts investigated by Ab Initio Molecular Dynamics (AIMD) simulations

Ab Initio Molecular Dynamics (AIMD) simulations are performed on molten KCl-UCl3 salt mixtures to determine energies, heat capacities, and densities. The density-dependent energy correction (DFT-dDsC), Grimme et al.'s DFT-D3, and Langreth & Lundqvist (vdW-cx) models are used for dispersion forces and combined with the Perdew-Burke-Ernzerhof (PBE) exchange-correlation potential with a Hubbard U parameter for the 5f electrons of uranium. After validating predictions for the end-member systems to literature data, KCl-UCl 3 mixtures are studied at select temperatures. Densities and energies both deviate from ideal solution behavior, with the maximum deviation occurring around 36% UCl 3 for mixing energies and slightly lower (29% UCl3) for densities. Compared to the NaCl-UCl 3 system, which was previously investigated using the same simulation methodologies, the KCl-UCl 3 density and mixing energy deviations from ideal solution behavior are larger by almost a factor of two. No deviation from ideal solution behavior for heat capacity was observed. The AIMD predictions for mixing energies and densities agree qualitatively with experimental data, though the spread in data obtained from the various dispersion force models utilized, measurements, and empirical estimates makes strong conclusions difficult. The dependence of thermodynamic and thermophysical properties on composition is correlated with the local chemistry of the solution phase, in particular, the tendency of UCl 3 to form network structures.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Ab initio molecular dynamics (AIMD) simulations of NaCl, UCl 3 molten salts

Ab initio molecular dynamics (AIMD) simulations are used to calculate select thermophysical and thermodynamic properties of NaCl, UCl and NaCl-UCl 3 molten salts. Following established approaches, the AIMD simulations include a model for Van der Waals interactions. The Langreth & Lundqvist (vDW-DF2), DFT-D3, and density-dependent energy correction (DFT-dDsC) dispersion models are first tested for molten NaCl in order to assess their accuracy for density and heat capacity predictions across a range of temperatures. Based on the NaCl results, the vdW-DF2 and DFT-dDsC methods are extended to the UCl 3 system and compared to available experimental data. Next, mixtures of NaCl-UCl 3 are investigated and analyzed with respect to the deviation from ideal solution behavior. For the DFT-dDsC simulations, density deviates by up to 2% from an ideal mixture, with the maximum occurring close to the eutectic composition. The mixing energy also deviates from an ideal solution and exhibits a minimum of about -0.074 to eV per formula unit, again close to the eutectic composition. Finally, the compressibility and species diffusivity of the pure and mixed salt systems are calculated. The diffusivities are slightly reduced in the mixed compared to the pure systems and the compressibilities loosely follow a linear correlation as a function of the UCl 3 composition. The trends observed for mixing properties are rationalized by correlating them to the coordination chemistry, which emphasizes the importance of maintaining the extended network formed by U and Cl ions as the NaCl concentration increases. The concentration at which breakdown of the extended network occurs, roughly coincides with the minimum of both the mixing energy and the deviation from ideal solution behavior for density.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

KCl-UCl 3 molten salts investigated by Ab Initio Molecular Dynamics (AIMD) simulations: A comparative study with three dispersion models

Ab Initio Molecular Dynamics (AIMD) simulations are performed on molten KCl-UCl 3 salt mixtures to determine energies, heat capacities, and densities. The density-dependent energy correction (DFT-dDsC), Grimme et al.’s DFT-D3, and Langreth & Lundqvist (vdW-cx) models are used for dispersion forces and combined with the Perdew-Burke-Ernzerhof (PBE) exchange-correlation potential with a Hubbard U param eter for the 5$f$ electrons of uranium. After validating predictions for the end-member systems to literature data, KCl-UCl 3 mixtures are studied at select temperatures. Densities and energies both deviate from ideal solution behavior, with the maximum deviation occurring around 36% UCl 3 for mixing energies and slightly lower (29% UCl 3 ) for densities. Compared to the NaCl-UCl 3 system, which was previously investigated using the same simulation methodologies, the KCl-UCl 3 density and mixing energy deviations from ideal solution behavior are larger by almost a factor of two. No deviation from ideal solution behavior for heat capacity was observed. The AIMD predictions for mixing energies and densities agree qualitatively with experimental data, though the spread in data obtained from the various dispersion force models utilized, measurements, and empirical estimates makes strong conclusions difficult. The dependence of thermodynamic and thermophysical properties on composition is correlated with the local chemistry of the solution phase, in particular, the tendency of UCl 3 to form network structures.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Anatomy of Local Structural Disorder of Ni(II) Species in MgCl 2 –KCl Molten Salts

Understanding the speciation of metal ions dissolved in molten salts (MS) is critical for enabling a broad range of high-temperature energy applications, including MS nuclear reactors and concentrated solar power plants. However, due to the inherent dynamicity of metal species in the MS environment and the strong temperature dependencies of their multiple coexisting forms, they are difficult to resolve structurally. Herein, we show that combining in situ X-ray absorption spectroscopy (XAS) with ab initio molecular dynamics (AIMD) simulations is necessary to uncover and quantify the coexisting coordination states of Ni(II) in molten MgCl 2 –KCl mixtures and explain how the temperature and salt composition control their relative populations. Furthermore, from the interionic angle and distance distributions of nickel in different coordination states obtained from AIMD simulations, it is evident that for each coordination state, the width and skewness of their bonding distributions increase with increasing coordination number. In conclusion, the combination of XAS with first-principles modeling to resolve metastable metal species in MS is critical for understanding their behavior over a wide range of temperatures and chemical environments in nuclear and solar applications.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Directed gas phase preparation of ethynylallene (H 2 CCCHCCH; X 1 A') via the crossed molecular beam reaction of the methylidyne radical (CH; X 2 Π) with vinylacetylene (H 2 CCHCCH; X 1 A')

The gas-phase bimolecular reaction of the methylidyne (CH; X 2 Π) radical with vinylacetylene (H 2 CCHCCH; X 1 A') was conducted at a collision energy of 20.3 kJ mol –1 under single collision conditions exploiting the crossed molecular beam experimental results merged with ab initio electronic structure calculations and ab initio molecular dynamics (AIMD) simulations. The laboratory data reveal that the bimolecular reaction proceeds barrierlessly via indirect scattering dynamics through long-lived C 5 H 5 reaction intermediate(s) ultimately dissociating to C 5 H 4 isomers along with atomic hydrogen with the latter predominantly originating from the vinylacetylene reactant as confirmed by the isotopic substitution experiments in the D1-methylidyne–vinylacetylene reaction. Combined with ab initio calculations of the potential energy surface (PES) and statistical Rice–Ramsperger–Kassel–Marcus (RRKM) calculations, the experimental determined reaction energy of –146 ± 26 kJ mol –1 along with the distribution minimum of T(θ) at 90° and isotopic substitution experiments suggest ethynylallene (p1; ΔrG = –230 ± 4 kJ mol –1 ) as the dominant product. The ethynylallene (p1) may be formed with extensive rovibrational excitation, which would result in a lower maximum translational energy. Further, AIMD simulations reveal that the reaction dynamics leads to p1 (ethynylallene, 75%) plus atomic hydrogen with the dominant initial complex being i1 formed by methylidyne radical addition to the double C[double bond, length as m-dash]C bond in vinylacetylene. Altogether, combining the crossed molecular beam experimental results with ab initio electronic structure calculations and ab initio molecular dynamics (AIMD) simulations, ethynylallene (p1) is expected to represent the dominant product in the reaction of the methylidyne (CH; X 2 Π) radical with vinylacetylene (H 2 CCHCCH; X 1 A').

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Accurate and scalable graph neural network force field and molecular dynamics with direct force architecture

Abstract Recently, machine learning (ML) has been used to address the computational cost that has been limiting ab initio molecular dynamics (AIMD). Here, we present GNNFF, a graph neural network framework to directly predict atomic forces from automatically extracted features of the local atomic environment that are translationally-invariant, but rotationally-covariant to the coordinate of the atoms. We demonstrate that GNNFF not only achieves high performance in terms of force prediction accuracy and computational speed on various materials systems, but also accurately predicts the forces of a large MD system after being trained on forces obtained from a smaller system. Finally, we use our framework to perform an MD simulation of Li 7 P 3 S 11 , a superionic conductor, and show that resulting Li diffusion coefficient is within 14% of that obtained directly from AIMD. The high performance exhibited by GNNFF can be easily generalized to study atomistic level dynamics of other material systems.

Chemistry↗
DOE OSTI2021

Insights into Spontaneous Solid Electrolyte Interphase Formation at Magnesium Metal Anode Surface from Ab Initio Molecular Dynamics Simulations

Spontaneous chemical reactivity at multivalent (Mg, Ca, Zn, Al) electrode surfaces is critical to solid electrolyte interphase (SEI) formation, and hence, directly affects the longevity of batteries. Here, we report an investigation of the reactivity of 0.5 M Mg(TFSI)2 in 1,2-dimethoxyethane (DME) solvent at a Mg(0001) surface using ab initio molecular dynamics (AIMD) simulations and detailed Bader charge analysis. Based on the simulations, the initial degradation reactions of the electrolyte strongly depend on the structure of the Mg(TFSI)2 species near the anode surface. At the surface, the dissociation of Mg(TFSI)2 species occurs via cleavage of the N-S bond for the solvent separated ion pair (SSIP) and via cleavage of the C-S bond for the contact ion pair (CIP) configuration. In the case of the CIP, both TFSI anions undergo spontaneous bond dissociation reactions to form atomic O, C, S, F, and N species adsorbed on the surface of the Mg anode. These products indicate that the initial SEI layer formed on the surface of the pristine Mg anode consists of a complex mixture of multiple components such as oxides, carbides, sulfides, fluorides, and nitrides. We believe that the atomic level insights gained from these simulations will lay the groundwork for the rational design of tailored and functional interphases that are critical for the success of multivalent battery technology.

Agarwal, Garvit↗
DOE OSTI2020

Mechanistic understanding of catalytic conversion of ethanol to 1-butene over 2D-pillared MFI zeolite

Ethanol is an important C2 platform molecule for producing value-added chemicals and distillate hydrocarbon fuels (e.g., jet and diesel). Among these, catalytic upgrading of ethanol to butenes can generate valuable commodity chemicals (e.g., 1-butene) and provide C4 olefin intermediates that can be further upgraded to jet/diesel fuels. 2D zeolites offer hierarchical mesoporous structures, leading to improved mass transport and reduced diffusion length, which can help to address the coking challenges faced by ethanol conversion to hydrocarbons over 3D zeolites. In this study, we investigate the acid-catalyzed conversion of ethanol to 1-butene over the Brønsted acid sites (BAS) in 2D-pillared MFI zeolite (2D-PMFI) using ab initio molecular dynamics (AIMD) simulations, in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), and calorimetric measurements. A detailed thermodynamic analysis, using quasi-harmonic approximation (QHA), on the Gibbs free energy pathway of ethanol conversion shows that the consideration of entropy is critical to accurately capture the detailed thermodynamic profiles. The formation of bulky ethoxy-ethene complex is found to be a potential rate-determining step (RDS), proceeding via stepwise mechanism. The reactivity of 2D-PMFI can be further tuned by manipulating RDS through a careful control of number of BAS and operating temperatures. The accuracy of computational model is confirmed through the comparison of calculated vibrational density of states (VDOS) against the experimental DRIFTS measurements. Overall, our study provides mechanistic insights into ethanol upgrading over the 2D-PMFI and shows the importance of evaluating entropic effects in such confined system.

Ethanol conversion, 2D-pillared zeolite, MFI, Anha↗
DOE OSTI2019

Beyond Badger’s Rule: The Origins and Generality of the Structure–Spectra Relationship of Aqueous Hydrogen Bonds

The structure of hydrogen bonded networks is intimately intertwined with their dynamics. Despite the incredibly wide range of hydrogen bond strengths encountered in water clusters, ion-water clusters and liquid water, we demonstrate that the previously reported correlation between the change in the equilibrium bond length of the hydrogen bonded OH covalent bond and the corresponding shift in its harmonic frequency in water clusters is much more broadly applicable. Surprisingly, this correlation describes the ratios for both the equilibrium OH bond length/harmonic frequency, and the vibrationally averaged bond length/anharmonic frequency in water, hydronium water and halide water clusters. Consideration of harmonic and anaharmonic data leads to a correlation of -19 ± 1 cm -1 /0.001 Å. The basic nature of this correlation is further confirmed through the analysis of ab initio Molecular Dynamics (AIMD) trajectories for liquid water. We show that this simple correlation for both harmonic and anharmonic systems can be modeled by the response of an OH bond to an external field. Treating the OH bond as a Morse oscillator, we develop analytic expressions, which relate the ratio of the shift in the vibrational frequency of the hydrogen-bonded OH bond to the shift in OH bond length, to parameters in the Morse potential and the ratio of the first and second derivatives of the field-dependent projection of the dipole moment of water onto the hydrogen-bonded OH bond. Based on our analysis, we develop a protocol for reconstructing the AIMD spectra of liquid water from the sampled distribution of the OH bond lengths. Our conclusions elucidate the origins of the relationship between the molecular structure of the fleeting hydrogen-bonded network and the ensuing dynamics, which can be probed by vibrational spectroscopy.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Effect of Fluoroethylene Carbonate Additives on the Initial Formation of the Solid Electrolyte Interphase on an Oxygen-Functionalized Graphitic Anode in Lithium-Ion Batteries

The formation of a solid electrolyte interphase (SEI) at the electrode/electrolyte interface substantially affects the stability and lifetime of lithium-ion batteries (LIBs). One of the methods to improve the lifetime of LIBs is by the inclusion of additive molecules to stabilize the SEI. To understand the effect of additive molecules on the initial stage of SEI formation, in this work we compare the decomposition and oligomerization reactions of a fluoroethylene carbonate (FEC) additive on a range of oxygen-functionalized graphitic anodes to those of an ethylene carbonate (EC) organic electrolyte. A series of density functional theory (DFT) calculations augmented by ab initio molecular dynamics (AIMD) simulations reveal that EC decomposition on an oxygen-functionalized graphitic ($11\bar{2}0$) edge facet through a nucleophilic attack on an ethylene carbon site (C E ) of an EC molecule (S2 mechanism) is spontaneous during the initial charging process of LIBs. However, decomposition of EC through a nucleophilic attack on a carbonyl carbon (C C ) site (S1 mechanism) results in alkoxide species regeneration that is responsible for continual oligomerization along the graphitic surface. In contrast, FEC prefers to decompose through an S1 pathway, which does not promote alkoxide regeneration. Including FEC as an additive is thus able to suppress alkoxide regeneration and results in a smaller and thinner SEI layer that is more flexible toward lithium intercalation during the charging/discharging process. In addition, we find that the presence of different oxygen functional groups at the surface of graphite dictates the oligomerization products and the LiF formation mechanism in the SEI.

25 ENERGY STORAGE↗
DOE OSTI2026

pH-Driven Restructuring of Hydration Layers and Cation Ad-sorption at the Alumina-Water Interface

Oxide-water interfaces underpin ion separation, catalysis, and electrochemical energy technologies, where the electrical double layer (EDL) controls adsorption, transport, and reactivity. Yet, the molecular-scale link between pH-dependent surface protonation, hydration-layer structure, and counter-ion adsorption remains poorly defined. Here, we combine in situ crystal truncation rod (CTR) and resonant anomalous X-ray reflectivity (RAXR) with streaming potential measurements and ab initio molecular dynamics (AIMD) simulations to resolve the chemical and structural evolution of the EDL at the single-crystal alumina (012)-water interface in 10 mM Rb+ over pH 3-12. CTR measurements reveal two distinct adsorbed water layers at ~2.2 and ~3.5 Å above the surface that each shift toward the substrate at transition pHs near 6.5 and 10.6, respectively, directly reflecting changes in primary hydration layer structure in response to the deprotonation of bridging and terminal aluminol groups. RAXR shows a 10-fold increase in Rb+ coverage and a decrease in mean adsorption height from ~3.5 to ~2.7 Å with increasing pH, indicating enhanced counter-ion binding accompanied by Stern layer contraction. Streaming potential measurements demonstrate that the zeta potential, i.e., potential at the hydrodynamic shear plane, is positive at pH 3 and becomes negative at pH ≥3.5. This negative charge magnitude increases with increasing pH, consistent with progressive surface deprotonation at higher pH. AIMD identifies inner- and outer-sphere Rb+ complexes whose adsorption heights and coordination geometries depend sensitively on the protonation state of surface oxygens, providing atomistic support for the experimentally inferred trends. These measurements establish two discrete, site-specific pH transitions in hydration-layer structure that track aluminol (de)protonation and quantitatively link them to a pH-driven contraction of the Stern layer (increasing Rb+ coverage and decreasing adsorption height). This provides a direct structural basis for connecting surface acid-base chemistry to ion binding distances at an oxide-water interface.

Electrical double layer (EDL), Surface protonation↗
DOE OSTI2022

Viscosity in water from first-principles and deep-neural-network simulations

Abstract We report on an extensive study of the viscosity of liquid water at near-ambient conditions, performed within the Green-Kubo theory of linear response and equilibrium ab initio molecular dynamics (AIMD), based on density-functional theory (DFT). In order to cope with the long simulation times necessary to achieve an acceptable statistical accuracy, our ab initio approach is enhanced with deep-neural-network potentials (NNP). This approach is first validated against AIMD results, obtained by using the Perdew–Burke–Ernzerhof (PBE) exchange-correlation functional and paying careful attention to crucial, yet often overlooked, aspects of the statistical data analysis. Then, we train a second NNP to a dataset generated from the Strongly Constrained and Appropriately Normed (SCAN) functional. Once the error resulting from the imperfect prediction of the melting line is offset by referring the simulated temperature to the theoretical melting one, our SCAN predictions of the shear viscosity of water are in very good agreement with experiments.

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

A hybrid calorimetry-simulation model of mixing enthalpy for molten salt

Calorimetric determination of enthalpies of mixing (ΔH mix ) in multicomponent molten salts is often interpreted using empirical models that lack physically meaningful parameters. However, for improving pyrochemical separation of spent nuclear fuel, where lanthanides are major fission products and critical elements, a deeper thermodynamic understanding of the link between excess thermodynamic properties and solvation structure is critically needed. In this work, we implement a hybrid and physics-informed framework, MIVM+Calorimetry+AIMD, which integrates experimentally measured ΔH mix (via high temperature drop calorimetry) with solvation structures from ab initio molecular dynamics (AIMD). This approach is demonstrated using LaCl 3 mixed with eutectic LiCl-KCl (58 mol% – 42 mol%) at 873 K and 1133 K. MIVM-derived parameters enable extrapolation of excess Gibbs energy and La 3+ activity across compositions. In contrast, direct ΔH mix predictions from AIMD and polarizable ion model simulations deviate significantly. By incorporating experimentally benchmarked solvation structures into an interpretable thermodynamic model, the MIVM+Calorimetry+AIMD formalism achieves higher accuracy and generalizable method for studying molten salts, offering a robust path for understanding and optimizing molten salt chemistry relevant to nuclear fuel cycles and separation science.

Goncharov, Vitaliy G. [Washington State Univ., Pul↗
DOE OSTI2020

Minimal Experimental Bias on the Hydrogen Bond Greatly Improves Ab Initio Molecular Dynamics Simulations of Water

Experiment Directed Simulations (EDS) is a method within a class of techniques seeking to improve molecular simulations by minimally biasing the system Hamiltonian to reproduce certain experimental observables. In a previous application of EDS to ab initio molecular dynamics (AIMD) simulation based on electronic density functional theory (DFT), the AIMD simulations of water were biased to reproduce its experimentally derived solvation structure. In particular, by solely biasing the O-O pair correlation functions, other structural and dynamical properties that were not biased were improved. In this work, the hypothesis is tested that directly biasing the OH pair correlation (and hence the H-O∙∙∙H hydrogen bonding), will provide an even better improvement of DFT-based water properties in AIMD simulations. The logic behind this hypothesis is that for most electronic DFT descriptions of water the hydrogen bonding is known to be deficient due to anomalous charge transfer and over polarization in the DFT. Using recent advances to the EDS learning algorithm, we thus train a minimal bias on AIMD water that reproduces the O-H radial distribution function derived from the highly accurate MB-pol model of water. Finally, it is then confirmed that biasing the O-H pair correlation alone can lead to improved AIMD water properties, with structural and dynamical properties in even closer to experiment than the previous EDS-AIMD model.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Machine-learned interatomic potentials by active learning: amorphous and liquid hafnium dioxide

We propose an active learning scheme for automatically sampling a minimum number of uncorrelated configurations for fitting the Gaussian Approximation Potential (GAP). Our active learning scheme consists of an unsupervised machine learning (ML) scheme coupled with a Bayesian optimization technique that evaluates the GAP model. We apply this scheme to a Hafnium dioxide (HfO 2 ) dataset generated from a "melt-quench" ab initio molecular dynamics (AIMD) protocol. Our results show that the active learning scheme, with no prior knowledge of the dataset, is able to extract a configuration that reaches the required energy fit tolerance. Further, molecular dynamics (MD) simulations performed using this active learned GAP model on 6144 atom systems of amorphous and liquid state elucidate the structural properties of HfO 2 with near ab initio precision and quench rates (i.e., 1.0 K/ps) not accessible via AIMD. The melt and amorphous X-ray structural factors generated from our simulation are in good agreement with experiment. In addition, the calculated diffusion constants are in good agreement with previous ab initio studies.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Molecular dynamics simulation of metallic Al–Ce liquids using a neural network machine learning interatomic potential

Al-rich Al-Ce alloys have the possibility of replacing heavier steel and cast-irons for use in high-temperature applications. Knowledge about the structures and properties of Al-Ce alloys at liquid state is vital for optimizing the manufacture process to produce desired allows. However, reliable molecular dynamics simulation of Al-Ce alloy systems remains a great challenge due to the lack of accurate Al-Ce interatomic potential. In this work, an artificial neural network (ANN) deep machine learning (ML) method is used to develop a reliable interatomic potential for Al-Ce alloy. Ab initio molecular dynamics (AIMD) simulation data on Al-Ce liquid with small unit cell (~200 atoms) and on the known Al-Ce crystalline compounds are collected to train the interatomic potential using ANN-ML. The obtained ANN-ML model reproduces well the energies, forces, and atomic structure of Al 90 Ce 10 liquid and crystalline phases of Al-Ce compounds in comparison with ab initio results. The developed ANN-ML potential is applied in molecular dynamics simulations to study the structures and properties of metallic Al 90 Ce 10 liquid, which would provide useful insight for guiding experimental process to produce desired Al-Ce allows.

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

First-Principles Molecular Dynamics of Monomethylhydrazine and Nitrogen Dioxide

The exploration of chemical reactions preceding ignition is essential for the development of ideal hypergolic propellants. Unexpected reaction pathways of a hypergolic mixture composed of monomethylhydrazine and nitrogen dioxide are predicted through a cooperative combination of (i) spin-unrestricted ab initio molecular dynamics (AIMD) and (ii) wave packet dynamics of protons. Ensembles of AIMD trajectories reveal a sequence of reaction steps for proton transfer and rupture of the C–N bond. The details of proton transfer are explored by wave packet dynamics on the basis of ab initio potential energy surfaces from AIMD trajectories. Finally, the possibility of spontaneous ignition of this hypergolic mixture at room temperature is predicted as a quantized feature of proton-transfer dynamics.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Using Machine Learning to Greatly Accelerate Path Integral Ab Initio Molecular Dynamics

Ab initio molecular dynamics (AIMD) has become one of the most popular and robust approaches for modeling complicated chemical, liquid, and material systems. However, the formidable computational cost often limits its widespread application in simulations of the largest-scale systems. Furthermore, the situation becomes even more severe in cases where the hydrogen nuclei may be better described as quantized particles using a path integral representation. Here, we present a computational approach that combines machine learning with recent advances in path integral contraction schemes, and we achieve a 2 orders of magnitude acceleration over direct path integral AIMD simulation while at the same time maintaining its accuracy.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗