Atomic-scale Simulations Research Articles

A novel mechanism and model of hydrogen isotope exchange in vacancy considering nuclear quantum effects

Abstract Tritium (T) retention in plasma-facing materials (PFMs) raises significant radiological safety concerns and adversely affects the self-sustained burning of T in fusion reactors. Therefore, the removal of retained T from PFMs has become an urgent task. Hydrogen isotope (HI) exchange has proven to be an effective method for T removal. However, the microscopic mechanisms, particularly the role of isotope effects, remain insufficiently understood. For the first time, this work employs path integral molecular dynamics (PIMD) to account for the nuclear quantum effects of HIs and explore the microscopic mechanisms of HI exchange in tungsten (W) vacancies. Atomic-scale simulations reveal that the fundamental principle of HI exchange is the reduction in binding energy between HIs and vacancies as the defect filling level increases, involving two key processes: de-trapping and trapping. These processes are influenced by isotope effects, with pronounced differences observed at low temperatures. Notably, T exhibits a higher probability of de-trapping from vacancies compared to hydrogen (H), while vacancies demonstrate a stronger affinity for trapping H. In contrast, the isotope effect is less pronounced between deuterium (D) and H, leading to different exchange efficiencies between H and T, as well as between D and T. Based on these isotope effects, we proposed a detailed microscopic mechanism for HI exchange in vacancies and developed a T evolution model that accounts for variations in HI concentrations. This work advances our understanding of HI exchange for T removal applications and offers a more accurate assessment of T retention in future D-T environments.

Atomistic-scale simulation of multilayer graphene etching induced by Hx+ (x = 1−3) ions irradiation in hydrogen plasma

Atomistic-scale simulation of multilayer graphene etching induced by H<i>x</i>+ (<i>x</i> = 1−3) ions irradiation in hydrogen plasma

Au/CeO2 Nanozyme Scaffold Boosts Electron and Hydrogen Transfer for NIR-Enhanced Chemodynamic Therapy.

CeO2 nanozymes have demonstrated the potential to enhance biological scaffolds with chemodynamic therapy. However, their catalytic efficacy is limited by the slow conversion of Ce4+ to Ce3+ and the lack of substrates like H2O2 and H+. To address these challenges, we adopted a dual-pronged strategy that utilized the plasmonic resonance of Au nanoparticles and their glucose-oxidase mimicry to boost electron and hydrogen transfer. Specifically, we integrated Au/CeO2 nanozymes into poly-l-lactic acid scaffolds via selective laser sintering. This conversion of Ce3+ to Ce4+ in the scaffolds enhanced the reduction of H2O2 to a hydroxyl radical, inducing oxidative stress in tumor cells. The Au nanoparticles played a crucial role in boosting the Ce3+/Ce4+ catalytic cycle by providing both the energy and the catalytic substrates. They recycled Ce4+ back to Ce3+ by exploiting plasmonic-induced hot electrons and catalyzed glucose oxidation to supply H2O2 and H+. Our nanoscale and atomic-scale simulations confirmed that the Au/CeO2 hybrid structure utilized near-field coupling to amplify the plasmonic resonance and the Au-O-Ce bridge reduced the electron transfer barrier. Consequently, the Au/CeO2 scaffold decreased the activation energy from 22.57 to 9.92 kJ/mol. These findings highlight the significant promise of the Au/CeO2 nanozyme scaffold for NIR-enhanced chemodynamic therapy.

Accelerated Alzheimer's Aβ-42 secondary nucleation chronologically visualized on fibril surfaces.

Protein fibril surfaces tend to generate toxic oligomers catalytically. To date, efforts to study the accelerated aggregation steps involved with Alzheimer's disease-linked amyloid-β (Aβ)-42 proteins on fibril surfaces have mainly relied on fluorophore-based analytics. Here, we visualize rare secondary nucleation events on the surface of Aβ-42 fibrils from embryonic to endpoint stages using liquid-based atomic force microscopy. Nanoscale imaging supported by atomic-scale molecular simulations tracked the adsorption and proliferation of oligomeric assemblies at nonperiodically spaced catalytic sites on the fibril surface. Upon confirming that fibril edges are preferential binding sites for oligomers during embryonic stages, the secondary fibrillar size changes were quantified during the growth stages. Notably, a small population of fibrils that displayed higher surface catalytic activity was identified as superspreaders. Profiling secondary fibrils during endpoint stages revealed a nearly threefold increase in their surface corrugation, a parameter we exploit to classify fibril subpopulations.

Glass transition temperature of asphalt binder based on atomistic scale simulation

Glass transition is one of the most crucial physical properties for polymerical materials. As a typical complex polymerical material, the glass transition phenomenon in asphalt binder is directly related to their temperature-related properties. To investigate the glass transition characteristics, this study delves into the glass transition temperature of asphalt binder based on molecular dynamics simulations. It is found that the calculation range for the glass transition temperature sits between 100 and 400 K. The evolution of asphalt binder structure is influenced by different cooling rates, where lower cooling rates allow sufficient microstructural rearrangement, resulting in a smaller volume at the lower temperature. Model size is closely associated with the glass transition region. As the size increases, the transition region significantly expands. Increasing the model size also reduces volume fluctuations after isothermal relaxation, providing more stable volume changes. It is observed that higher cooling rates with a model size over 100 Å can well reproduce the glass transition process of asphalt binders. This work provides atomic-scale insights for the glass transition phenomenon in asphalt binder, which could be beneficial for the design of high-performance asphalt binder.

An understanding of the segregation and migration mechanism of point defects in tungsten grain boundaries: An atomic scale simulation

In this study, molecular dynamics (MD) simulations are employed to investigate the interactions between point defects and grain boundaries (GBs) in tungsten. Firstly, the segregation energy of vacancies (Vs) and self-interstitial atoms (SIAs) to GBs is examined. The results indicate that both Vs and SIAs tend to segregate towards GBs. Notably, the different types of GBs exhibit varying attraction to Vs and SIAs. The interaction radius is applied to investigate the degree of segregation, which is a significant index of GBs’ attraction to Vs and SIAs. The segregation radius of SIAs is typically larger than that of Vs. High-angle grain boundaries (HAGBs) show strong segregation for SIAs, facilitating the aggregation at GBs. Additionally, elastic theory is applied to qualitatively discuss the factors influencing the segregation energy distribution. The differences in atomic free volume caused by hydrostatic stress and the lattice distortion induced by point defects lead to Vs and SIAs occupying different sites. Finally, the nudged elastic band (NEB) method is employed to study the migration of Vs and SIAs near GBs, revealing that migration energy barriers for Vs and SIAs in GBs are much lower than in bulk. GBs facilitate the migration of point defects. Low-angle grain boundaries (LAGBs) present a higher energy barrier for V migration. Vs tend to occupy the compressive region, while SIAs tend to occupy the tensile region. Particularly, in comparison to Vs, SIAs are more likely to segregate to GBs due to higher binding energy, wider interaction radius, and extremely lower diffusion energy barrier. This provides some insights into the segregation and migration of point defects in GBs.

On the stability of nanovoids in fcc metals and the influence of hydrogen

Atomic-scale simulations are used to examine the stability of nanometric vacancy clusters in variety of face centered cubic metals (Au, Ag, Al, Cu and Ni), according to their shape, their size and the presence of hydrogen. It is shown that whilst in most of metals the stacking fault tetrahedron is the most stable defect in absence of hydrogen, cavities becomes more stable when the concentration of hydrogen is sufficiently large to cover the internal surfaces. The (100) surfaces are found to be the less energetic, which favors vacancy clusters with cuboid shape in the metals concerned here. The simulations are shown to agree well with theoretical predictions provided that the theory is parametrized with surface and bulk energies conformed to the atomistic models. The affinity between hydrogen and cuboid cavities has also been confirmed by the computation of hydrogen solubility, which is neatly enhanced in the vicinity of these cavities. Our study paves the way towards the comprehensive atomistic-based predictions for micro-structure in hydrogen-rich metals.

Data-driven modeling of dislocation mobility from atomistics using physics-informed machine learning

Dislocation mobility, which dictates the response of dislocations to an applied stress, is a fundamental property of crystalline materials that governs the evolution of plastic deformation. Traditional approaches for deriving mobility laws rely on phenomenological models of the underlying physics, whose free parameters are in turn fitted to a small number of intuition-driven atomic scale simulations under varying conditions of temperature and stress. This tedious and time-consuming approach becomes particularly cumbersome for materials with complex dependencies on stress, temperature, and local environment, such as body-centered cubic crystals (BCC) metals and alloys. In this paper, we present a novel, uncertainty quantification-driven active learning paradigm for learning dislocation mobility laws from automated high-throughput large-scale molecular dynamics simulations, using Graph Neural Networks (GNN) with a physics-informed architecture. We demonstrate that this Physics-informed Graph Neural Network (PI-GNN) framework captures the underlying physics more accurately compared to existing phenomenological mobility laws in BCC metals.

Oxygen ion diffusion in RE3TaO7: Why long-range migration of O2− is prohibited in the defective-fluorite structure?

Oxides with low O2− lattice diffusion rate are critical for the development of topcoat materials for next generation thermal barrier coatings (TBCs) working at > 1600 °C to prevent the fast growth of thermally grown oxide (TGO) at the bondcoat/topcoat interface. Here we delve into a comprehensive analysis of the oxygen ion transport characteristics of the recently developed low-, medium- and high-entropy rare earth tantalates (RE3TaO7) for TBCs by a combination of impedance spectroscopy, 18O isotope diffusion tests and atomic-scale simulations. Results show that the RE3TaO7 series display remarkably low oxygen ion diffusion rate (> 3 orders of magnitude lower than the state-of-the-art topcoat material yttria stabilized zirconia, YSZ) despite the presence of abundant oxygen vacancies in the structure. Molecular dynamic (MD) simulations combined with first principles calculations reveal that oxygen ions are strongly trapped by the potential well at the Ta-Ta edge in the tantalates. In addition, the high electrostatic potential (EP) surrounding oxygen vacancies and the high diffusion barriers between rare earth elements also impede oxygen ion diffusion. With low oxygen ion diffusion rate, along with exceptional thermal and mechanical properties reported previously, RE3TaO7 are promising topcoat materials for next-generation TBCs working at higher temperatures. Moreover, this study also provides valuable insights for understanding the O2− lattice diffusion behaviour in RE3TaO7 with a defective fluorite structure, which may benefit the exploration of oxide-ion conductors.

I-PI 3.0: A flexible and efficient framework for advanced atomistic simulations.

Atomic-scale simulations have progressed tremendously over the past decade, largely thanks to the availability of machine-learning interatomic potentials. These potentials combine the accuracy of electronic structure calculations with the ability to reach extensive length and time scales. The i-PI package facilitates integrating the latest developments in this field with advanced modeling techniques thanks to a modular software architecture based on inter-process communication through a socket interface. The choice of Python for implementation facilitates rapid prototyping but can add computational overhead. In this new release, we carefully benchmarked and optimized i-PI for several common simulation scenarios, making such overhead negligible when i-PI is used to model systems up to tens of thousands of atoms using widely adopted machine learning interatomic potentials, such as Behler-Parinello, DeePMD, and MACE neural networks. We also present the implementation of several new features, including an efficient algorithm to model bosonic and fermionic exchange, a framework for uncertainty quantification to be used in conjunction with machine-learning potentials, a communication infrastructure that allows for deeper integration with electronic-driven simulations, and an approach to simulate coupled photon-nuclear dynamics in optical or plasmonic cavities.

A Multiscale Computational Modeling Framework to Quantify Microstructural Effects on Nucleation of Li Metal on Carbon Scaffold Architectures

Carbon materials with scaffold architecture have been shown to mitigate the non-uniform Li plating/stripping issues during cycling, which is critical for achieving significantly enhanced energy densities and improved safety of Li metal batteries. However, optimal design of the architecture and microstructure of the carbon materials requires fundamental understanding of the Li metal nucleation behavior at different length scales. In this poster, we present a multiscale computational modeling framework to quantify the microstructural effects of carbon materials on the tendency of Li metal nucleation during plating by integrating atomic-scale simulation, mesoscale microstructure-based homogenization, continuum-scale electrochemistry modeling, and classical nucleation theory. We modeled the architecture of the carbon scaffold and the electrochemistry of Li transport across the electrode and electrolyte by finite element simulation to determine boundary conditions for our mesoscale models. We then generated different porous microstructures of various carbon materials by stochastic and physics-based simulations informed from experiments and performed multiphysics-based simulations at mesoscale to identify the electro-chemo-mechanical hotspots which are sensitive to the microstructure morphology. We calculated the energy barriers for Li to form clusters on graphene surfaces as a function of local Li concentrations and electric fields by ab initio molecular dynamics simulations. The synergistic deployment of different models across scales allows us to theoretically estimate the Li metal nucleation tendency based on a modified classical nucleation theory for different microstructures under various electroplating conditions. Comparison of the theoretical predictions and experimental results will be briefly discussed. The insights obtained from this multiscale framework offer new understanding and strategies for the design of carbon scaffold architectures to mitigate the non-uniformity of Li plating and issues in Li metal anode deployment in solid-state batteries.This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 and 22-ERD-043.

Crystal structure identification with 3D convolutional neural networks with application to high-pressure phase transitions in SiO2

Efficient, reliable and easy-to-use structure recognition of atomic environments is essential for the analysis of atomic scale computer simulations. In this work, we train two neuronal network (NN) architectures, namely PointNet and dynamic graph convolutional NN (DG-CNN) using different hyperparameters and training regimes to assess their performance in structure identification tasks of atomistic structure data. We show benchmarks on simple crystal structures, where we can compare against established methods. The approach is subsequently extended to structurally more complex SiO2 phases. By making use of this structure recognition tool, we are able to achieve a deeper understanding of the crystallization process in amorphous SiO2 under shock compression. Lastly, we show how the NN based structure identification workflows can be integrated into OVITO using its python interface.

Hybrid van der Waals Epitaxy.

The successful growth of non-van der Waals (vdW) group-III nitride epilayers on vdW substrates not only opens an unprecedented opportunity to obtain high-quality semiconductor thinfilm but also raises a strong debate for its growth mechanism. Here, combining multiscale computational approaches and experimental characterization, we propose that the growth of a nitride epilayer on a vdW substrate, e.g., AlN on graphene, may belong to a previously unknown model, named hybrid vdW epitaxy (HVE). Atomic-scale simulations demonstrate that a unique interfacial hybrid-vdW interaction can be created between AlN and graphene, and, consequently, a first-principles-based continuum growth model is developed to capture the unusual features of HVE. Surprisingly, it is revealed that the in-plane and out-of-plane growth are strongly correlated in HVE, which is absent in existing growth models. The concept of HVE is confirmed by our experimental measurements, presenting a new growth mechanism beyond the current category of material growth.

CHARMM-GUI Multicomponent Assembler for modeling and simulation of complex multicomponent systems

Atomic-scale molecular modeling and simulation are powerful tools for computational biology. However, constructing models with large, densely packed molecules, non-water solvents, or with combinations of multiple biomembranes, polymers, and nanomaterials remains challenging and requires significant time and expertise. Furthermore, existing tools do not support such assemblies under the periodic boundary conditions (PBC) necessary for molecular simulation. Here, we describe Multicomponent Assembler in CHARMM-GUI that automates complex molecular assembly and simulation input preparation under the PBC. In this work, we demonstrate its versatility by preparing 6 challenging systems with varying density of large components: (1) solvated proteins, (2) solvated proteins with a pre-equilibrated membrane, (3) solvated proteins with a sheet-like nanomaterial, (4) solvated proteins with a sheet-like polymer, (5) a mixed membrane-nanomaterial system, and (6) a sheet-like polymer with gaseous solvent. Multicomponent Assembler is expected to be a unique cyberinfrastructure to study complex interactions between small molecules, biomacromolecules, polymers, and nanomaterials.

Atomic insight into nanoindentation response of nanotwinned FeCoCrNiCu high entropy alloys

FeCoCrNiCu high-entropy alloys (HEAs) exhibit extraordinary mechanical properties and have the capability to withstand extreme temperatures and pressures. Their exceptional attributes make them suitable for diverse applications, from aerospace to chemical industry. We employ atomic-scale simulations to explore the effects of twinning boundary and twinning thickness on the mechanical behavior of nanotwinned FeCoCrNiCu during nanoindentation. The findings suggest that as the twinning thickness decreases within the range of 19.3–28.9 Å, both twinning partial slips (TPSs) and horizontal TPSs gradually become dominant in governing the plastic behaviors of the nanotwinned FeCoCrNiCu, thereby resulting in an inverse Hall–Petch effect. Remarkably, when the twinning thickness is compressed below 19.3 Å, a shift in the plastic deformation mechanism emerges, triggering the conventional Hall–Petch relation. The observed Hall–Petch behavior in nanotwinned FeCoCrNiCu is attributed to the strengthening effect imparted by the twinning boundaries. Consequently, the twinning boundary play an instrumental role in steering the plastic deformation mechanism of the nanotwinned FeCoCrNiCu when the twinning thickness descends beneath 19.3 Å. This study contributes significant insights towards the design of next-generation high-performance HEAs, underpinning their potential industrial utilization.

Improved Highly Mobile Membrane Mimetic Model for Investigating Protein-Cholesterol Interactions.

Cholesterol (CHL) plays an integral role in modulating the function and activity of various mammalian membrane proteins. Due to the slow dynamics of lipids, conventional computational studies of protein-CHL interactions rely on either long-time scale atomistic simulations or coarse-grained approximations to sample the process. A highly mobile membrane mimetic (HMMM) has been developed to enhance lipid diffusion and thus used to facilitate the investigation of lipid interactions with peripheral membrane proteins and, with customized in silico solvents to replace phospholipid tails, with integral membrane proteins. Here, we report an updated HMMM model that is able to include CHL, a nonphospholipid component of the membrane, henceforth called HMMM-CHL. To this end, we had to optimize the effect of the customized solvents on CHL behavior in the membrane. Furthermore, the new solvent is compatible with simulations using force-based switching protocols. In the HMMM-CHL, both improved CHL dynamics and accelerated lipid diffusion are integrated. To test the updated model, we have applied it to the characterization of protein-CHL interactions in two membrane protein systems, the human β2-adrenergic receptor (β2AR) and the mitochondrial voltage-dependent anion channel 1 (VDAC-1). Our HMMM-CHL simulations successfully identified CHL binding sites and captured detailed CHL interactions in excellent consistency with experimental data as well as other simulation results, indicating the utility of the improved model in applications where an enhanced sampling of protein-CHL interactions is desired.

Thermal degradation of greenhouse gas SF6 at realistic temperatures: Insights from atomic-scale CVHD simulations

Sulfur hexafluoride (SF6), recognized as a potent greenhouse gas with significant contributions to climate change, presents challenges in understanding its degradation processes. Molecular dynamics simulations are valuable tools for understanding modes of decomposition while the traditional approaches face limitations in time scale and require unrealistically high temperatures. The collective variable-driven hyperdynamics (CVHD) approach has been introduced to directly depict the pyrolysis process for SF6 gas at practical application temperatures, as low as 1600 K for the first time. Achieving an unprecedented acceleration factor of up to 107, the method extends the simulation time scale to milliseconds and beyond while maintaining consistency with experimental and theoretical models. The differences in the reaction process between simulations conducted at actual and elevated temperatures have been noted, providing insights into SF6 degradation pathways. The work provides a basis for the further studies on the thermal degradation of pollutants.

Data-Driven Path Collective Variables.

Identifying optimal collective variables to model transformations using atomic-scale simulations is a long-standing challenge. We propose a new method for the generation, optimization, and comparison of collective variables that can be thought of as a data-driven generalization of the path collective variable concept. It consists of a kernel ridge regression of the committor probability, which encodes a transformation's progress. The resulting collective variable is one-dimensional, interpretable, and differentiable, making it appropriate for enhanced sampling simulations requiring biasing. We demonstrate the validity of the method on two different applications: a precipitation model and the association of Li+ and F- in water. For the former, we show that global descriptors such as the permutation invariant vector allow reaching an accuracy far from the one achieved via simpler, more intuitive variables. For the latter, we show that information correlated with the transformation mechanism is contained in the first solvation shell only and that inertial effects prevent the derivation of optimal collective variables from the atomic positions only.

Tensorial Properties via the Neuroevolution Potential Framework: Fast Simulation of Infrared and Raman Spectra.

Infrared and Raman spectroscopy are widely used for the characterization of gases, liquids, and solids, as the spectra contain a wealth of information concerning, in particular, the dynamics of these systems. Atomic scale simulations can be used to predict such spectra but are often severely limited due to high computational cost or the need for strong approximations that limit the application range and reliability. Here, we introduce a machine learning (ML) accelerated approach that addresses these shortcomings and provides a significant performance boost in terms of data and computational efficiency compared with earlier ML schemes. To this end, we generalize the neuroevolution potential approach to enable the prediction of rank one and two tensors to obtain the tensorial neuroevolution potential (TNEP) scheme. We apply the resulting framework to construct models for the dipole moment, polarizability, and susceptibility of molecules, liquids, and solids and show that our approach compares favorably with several ML models from the literature with respect to accuracy and computational efficiency. Finally, we demonstrate the application of the TNEP approach to the prediction of infrared and Raman spectra of liquid water, a molecule (PTAF-), and a prototypical perovskite with strong anharmonicity (BaZrO3). The TNEP approach is implemented in the free and open source software package gpumd, which makes this methodology readily available to the scientific community.

Enhancing understanding metal matrix composites through molecular dynamics simulation: A comprehensive review

Metal-matrix composites (MMCs) attracts have gained significant attention in recent years and are extensively utilized in aerospace field due to their exceptional properties. Molecular dynamics simulation are suitable for describing interface behavior and predicting the mechanical properties of MMCs. This paper provides a review of mechanical properties prediction, physical processes and strengthen mechanisms of MMCs through molecular dynamics simulations. Firstly, the potential functions and establishment methods of composite models are introduced. Subsequently, commonly used simulation software and post-processing methods are discussed. Various mechanical predictions are summarized to evaluate the effect of microstructure on the mechanical properties, and the physical processes and functionalized composites formed through self-assembly and encapsulation are introduced. Additionally, dislocation evolution and crack propagation are analyzed to investigate the strengthening mechanisms of MMCs. The reviews also covers multi-scale simulations that elucidate structure-property relationship, and proposes research direction for atomic-scale simulation.