Overcoming the Chemical Complexity Bottleneck in on-the-Fly Machine Learned Molecular Dynamics Simulations

We develop a framework for on-the-fly machine learned force field molecular dynamics simulations based on the multipole featurization scheme that overcomes the bottleneck with the number of chemical elements. Considering bulk systems with up to 6 elements, we demonstrate that the number of density functional theory calls remains approximately independent of the number of chemical elements, in contrast to the increase in the smooth overlap of the atomic positions scheme.

Advances in the Dynamics of Adsorbate Diffusion on Metal Surfaces: Focus on Hydrogen and Oxygen

Adsorbates on metal surfaces are typically formed from the dissociative chemisorption of molecules occurring at gas-solid interfaces. These adsorbed species exhibit unique diffusion behaviors on metal surfaces, which are influenced by their translational energy. They play crucial roles in various fields, including heterogeneous catalysis and corrosion. This review examines recent theoretical advancements in understanding the diffusion dynamics of adsorbates on metal surfaces, with a specific emphasis on hydrogen and oxygen atoms. The diffusion processes of adsorbates on metal surfaces involve two energy transfer mechanisms: surface phonons and electron-hole pair excitations. This review also surveys new theoretical methods, including the characterization of the electron-hole pair excitation within electronic friction models, the acceleration of quantum chemistry calculations through machine learning, and the treatment of atomic nuclear motion from both quantum mechanical and classical perspectives. Furthermore, this review offers valuable insights into how energy transfer, nuclear quantum effects, supercell sizes, and the topography of potential energy surfaces impact the diffusion behavior of hydrogen and oxygen species on metal surfaces. Lastly, some preliminary research proposals are presented.

Solvent structure and dynamics over Brønsted acid MWW zeolite nanosheets

In the liquid phase of heterogeneous catalysis, solvent plays an important role and governs the kinetics and thermodynamics of a reaction. Although it is often difficult to quantify the role of the solvent, it becomes particularly challenging when a zeolite is used as the catalyst. This difficulty arises from the complex nature of the liquid/zeolite interface and the different solvation environments around catalytically active sites. Here, we use ab initio molecular dynamics simulations to probe the local solvation structure and dynamics of methanol and water over MWW zeolite nanosheets with varying Brønsted acidity. We find that the zeolite framework and the number and location of the acid sites in the zeolite influence the structure and dynamics of the solvent. In particular, methanol is more likely to be in the vicinity of the aluminum (Al3+) at the T4 site than at T1 due to easy accessibility. The methanol oxygen binds strongly to the Al at the T4 site, weakening the Al-O for the bridging acid site, which results in the formation of the silanol group, significantly reducing the acidity of the site. The behavior of methanol is in direct contrast to that of water, where protons can easily propagate from the zeolite to the solvent molecules regardless of the acid site location. Our work provides molecular-level insights into how solvent interacts with zeolite surfaces, leading to an improved understanding of the catalytic site in the MWW zeolite nanosheet.

Reference-Quality Free Energy Barriers in Catalysis from Machine Learning Thermodynamic Perturbation Theory

For the first time, we report calculations of the free energies of activation of cracking and isomerization reactions of alkenes that combine several different electronic structure methods with molecular dynamics simulations. We demonstrate that the use of a high level of theory (here Random Phase Approximation-RPA) is necessary to bridge the gap between experimental and computed values. These transformations, catalyzed by zeolites and proceeding via cationic intermediates and transition states, are building blocks of many chemical transformations for valorization of long chain paraffins originating, e.g., from plastic waste, vegetable oils, Fischer-Tropsch waxes or crude oils. Compared with the free energy barriers computed at the PBE+D2 production level of theory via constrained ab initio molecular dynamics, the barriers computed at the RPA level by the application of Machine Learning thermodynamic Perturbation Theory (MLPT) show a significant decrease for isomerization reaction and an increase of a similar magnitude for cracking, yielding an unprecedented agreement with the results obtained by experiments and kinetic modeling.

Catalytic Ring-Opening Polymerisation of Cyclic Ethylene Carbonate: Importance of Elementary Steps for Determining Polymer Properties Revealed via DFT-MTD Simulations Validated Using Kinetic Measurements

The production of CO2-containing polymers is still very demanding in terms of controlling the synthesis of products with pre-defined CO2 content and molecular weight. An elegant way of synthesising these polymers is via CO2-containing building blocks, such as cyclic ethylene carbonate (cEC), via catalytic ring-opening polymerisation. However, to date, the mechanism of this reaction and control parameters have not been elucidated. In this work, using DFT-metadynamics simulations for exploiting the potential of the polymerisation process, we aim to shed more light on the mechanisms of the interaction between catalysts (in particular, the catalysts K3VO4, K3PO4, and Na2SnO3) and the cEC monomer in the propagation step of the polymeric chain and the occurring CO2 release. Confirming the simulation results via subsequent kinetics measurements indicates that, depending on the catalyst's characteristics, it can be attached reversibly to the polymeric chain during polymerisation, resulting in a defined lifetime of the activated polymer chain. The second anionic oxygen of the catalyst can promote the catalyst's transfer to another electrophilic cEC monomer, terminating the growth of the first chain and initiating the propagation of the new polymer chain. This transfer reaction is an essential step in controlling the molecular weight of the products.

A Combined Reaction Path Search and Hybrid Solvation Method for the Systematic Exploration of Elementary Reactions at the Solid-Liquid Interface

We present a combined simulation method of single-component artificial force induced reaction (SC-AFIR) and effective screening medium combined with the reference interaction site model (ESM-RISM), termed SC-AFIR+ESM-RISM. SC-AFIR automatically and systematically explores the chemical reaction pathway, and ESM-RISM directly simulates the precise electronic structure at the solid-liquid interface. Hence, SC-AFIR+ESM-RISM enables us to explore reliable reaction pathways at the solid-liquid interface. We applied it to explore the dissociation pathway of an H2O molecule at the Cu(111)/water interface. The reaction path networks of the whole reaction and the minimum energy paths from H2O to H2 + O depend on the interfacial environment. The qualitative difference in the energy diagrams and the resulting change in the kinematically favored dissociation pathway upon changing the solvation environments are discussed. We believe that SC-AFIR+ESM-RISM will be a powerful tool to reveal the details of chemical reactions in surface catalysis and electrochemistry.

Hydrogen-migration governed dynamic magnetic coupling characteristics in nitrogen-vacancy-hydrogen nanodiamonds

The nitrogen-vacancy center doped with hydrogen (NVH) is one of the most common defects in diamonds, and the doping of hydrogen is known to enable mobility among three equivalent C-radicals in the defect, which noticeably affects the spin coupling among the radicals. Here, we for the first time uncover the dynamic nature of magnetic coupling induced by H-migration in the NVH center of nanodiamonds, using spin-polarized density functional theory calculations and enhanced sampling metadynamics simulations. The mobility of doping H enables the interior NVH region to become a variable magnetic space (antiferromagnetic/AFM versus ferromagnetic/FM). That is, the dynamic H has three frequently reachable binding C sites where H enables the center to exhibit variable AFM coupling (high up to J = -1282 cm-1) and that in other H-reachable regions including N sites, it enables the center to exhibit FM coupling (high up to J = 598 cm-1). The magnetic switching (AFM ↔ FM) and strength fluctuation strongly depend on the H-position which can adjust the ratio of the C radical orbitals in their mixing orbitals for a special three-electron three-center covalent C⋯H⋯C H-bonding and radical orbital distributions. Clearly, this work provides insights into the dynamic switching of magnetic coupling in such multi-radical centers of defect nanodiamonds.

Operando Modeling of Zeolite-Catalyzed Reactions Using First-Principles Molecular Dynamics Simulations

Within this Perspective, we critically reflect on the role of first-principles molecular dynamics (MD) simulations in unraveling the catalytic function within zeolites under operating conditions. First-principles MD simulations refer to methods where the dynamics of the nuclei is followed in time by integrating the Newtonian equations of motion on a potential energy surface that is determined by solving the quantum-mechanical many-body problem for the electrons. Catalytic solids used in industrial applications show an intriguing high degree of complexity, with phenomena taking place at a broad range of length and time scales. Additionally, the state and function of a catalyst critically depend on the operating conditions, such as temperature, moisture, presence of water, etc. Herein we show by means of a series of exemplary cases how first-principles MD simulations are instrumental to unravel the catalyst complexity at the molecular scale. Examples show how the nature of reactive species at higher catalytic temperatures may drastically change compared to species at lower temperatures and how the nature of active sites may dynamically change upon exposure to water. To simulate rare events, first-principles MD simulations need to be used in combination with enhanced sampling techniques to efficiently sample low-probability regions of phase space. Using these techniques, it is shown how competitive pathways at operating conditions can be discovered and how broad transition state regions can be explored. Interestingly, such simulations can also be used to study hindered diffusion under operating conditions. The cases shown clearly illustrate how first-principles MD simulations reveal insights into the catalytic function at operating conditions, which could not be discovered using static or local approaches where only a few points are considered on the potential energy surface (PES). Despite these advantages, some major hurdles still exist to fully integrate first-principles MD methods in a standard computational catalytic workflow or to use the output of MD simulations as input for multiple length/time scale methods that aim to bridge to the reactor scale. First of all, methods are needed that allow us to evaluate the interatomic forces with quantum-mechanical accuracy, albeit at a much lower computational cost compared to currently used density functional theory (DFT) methods. The use of DFT limits the currently attainable length/time scales to hundreds of picoseconds and a few nanometers, which are much smaller than realistic catalyst particle dimensions and time scales encountered in the catalysis process. One solution could be to construct machine learning potentials (MLPs), where a numerical potential is derived from underlying quantum-mechanical data, which could be used in subsequent MD simulations. As such, much longer length and time scales could be reached; however, quite some research is still necessary to construct MLPs for the complex systems encountered in industrially used catalysts. Second, most currently used enhanced sampling techniques in catalysis make use of collective variables (CVs), which are mostly determined based on chemical intuition. To explore complex reactive networks with MD simulations, methods are needed that allow the automatic discovery of CVs or methods that do not rely on a priori definition of CVs. Recently, various data-driven methods have been proposed, which could be explored for complex catalytic systems. Lastly, first-principles MD methods are currently mostly used to investigate local reactive events. We hope that with the rise of data-driven methods and more efficient methods to describe the PES, first-principles MD methods will in the future also be able to describe longer length/time scale processes in catalysis. This might lead to a consistent dynamic description of all steps-diffusion, adsorption, and reaction-as they take place at the catalyst particle level.

Deep Reinforcement Learning Environment Approach Based on Nanocatalyst XAS Diagnostics Graphic Formalization

The most in-demand instrumental methods for new functional nanomaterial diagnostics employ synchrotron radiation, which is used to determine a material's electronic and local atomic structure. The high time and resource costs of researching at international synchrotron radiation centers and the problems involved in developing an optimal strategy and in planning the control of the experiments are acute. One possible approach to solving these problems involves the use of deep reinforcement learning agents. However, this approach requires the creation of a special environment that provides a reliable level of response to the agent's actions. As the physical experimental environment of nanocatalyst diagnostics is potentially a complex multiscale system, there are no unified comprehensive representations that formalize the structure and states as a single digital model. This study proposes an approach based on the decomposition of the experimental system into the original physically plausible nodes, with subsequent merging and optimization as a metagraphic representation with which to model the complex multiscale physicochemical environments. The advantage of this approach is the possibility to directly use the numerical model to predict the system states and to optimize the experimental conditions and parameters. Additionally, the obtained model can form the basic planning principles and allow for the optimization of the search for the optimal strategy with which to control the experiment when it is used as a training environment to provide different abstraction levels of system state reactions.

Iterative multiscale and multi-physics computations for operando catalyst nanostructure elucidation and kinetic modeling

Modern heterogeneous catalysis has benefitted immensely from computational predictions of catalyst structure and its evolution under reaction conditions, first-principles mechanistic investigations, and detailed kinetic modeling, which are rungs on a multiscale workflow. Establishing connections across these rungs and integration with experiments have been challenging. Here, operando catalyst structure prediction techniques using density functional theory simulations and ab initio thermodynamics calculations, molecular dynamics, and machine learning techniques are presented. Surface structure characterization by computational spectroscopic and machine learning techniques is then discussed. Hierarchical approaches in kinetic parameter estimation involving semi-empirical, data-driven, and first-principles calculations and detailed kinetic modeling via mean-field microkinetic modeling and kinetic Monte Carlo simulations are discussed along with methods and the need for uncertainty quantification. With these as the background, this article proposes a bottom-up hierarchical and closed loop modeling framework incorporating consistency checks and iterative refinements at each level and across levels.

A DFT-metadynamics study disclosing key properties of ring-opening polymerization catalysts to produce polyethercarbonate polyols from cyclic ethylene carbonate as part of an emerging CCU technology

The ring opening polymerization of cyclic carbonates made from epoxide and CO₂ to CO₂-containing polymers constitutes an emerging technology of particular industrial interest. Considering the reaction of ring-opening polymerization of cyclic ethylene carbonate to produce polyethercarbonate polyols, several types of catalysts were tested experimentally and mechanistic pathways were proposed, but a detailed analysis of structure property relationship including the CO₂-liberation pathways is still lacking. This contribution is using computational methods to investigate reported benchmark catalysts with the lead structure A_xM_yO_z (A: alkali metal or alkyl, M: main group element or transition metal) that are particularly approved as effiecient catalysts for industrial purpose. Employing DFT-metadynamics simulations, free energy surfaces (FESs) for the key-steps in the catalytic polymerization of cyclic ethylene carbonate (cEC) are generated. Important structural criteria and characteristics of the catalysts that influence the catalytic performance and (side)reaction pathways are determined. It turns out that less nucleophilicity of the catalyst anion and more labile cations remain major criteria for prohibiting CO₂ liberation during polymerization. The key learnings of this contribution currently serve as a basis to develop the next generation of catalysts to bring this emerging carbon capture and use (CCU) technology into industrial application.

Challenges in modelling dynamic processes in realistic nanostructured materials at operating conditions

The question is addressed in how far current modelling strategies are capable of modelling dynamic phenomena in realistic nanostructured materials at operating conditions. Nanostructured materials used in applications are far from perfect; they possess a broad range of heterogeneities in space and time extending over several orders of magnitude. Spatial heterogeneities from the subnanometre to the micrometre scale in crystal particles with a finite size and specific morphology, impact the material's dynamics. Furthermore, the material's functional behaviour is largely determined by the operating conditions. Currently, there exists a huge length-time scale gap between attainable theoretical length-time scales and experimentally relevant scales. Within this perspective, three key challenges are highlighted within the molecular modelling chain to bridge this length-time scale gap. Methods are needed that enable (i) building structural models for realistic crystal particles having mesoscale dimensions with isolated defects, correlated nanoregions, mesoporosity, internal and external surfaces; (ii) the evaluation of interatomic forces with quantum mechanical accuracy albeit at much lower computational cost than the currently used density functional theory methods and (iii) derivation of the kinetics of phenomena taking place in a multi-length-time scale window to obtain an overall view of the dynamics of the process. This article is part of a discussion meeting issue 'Supercomputing simulations of advanced materials'.

Understanding the selection of catalytic pathway on graphene-supported nitrogen coordinated Ru-atom by ab initio molecular dynamics simulation

CONTEXT

In the paper, the ORR/OER on graphene-supported nitrogen coordinated Ru-atom (Ru-N-C) is simulated. We discuss nitrogen coordination influences electronic properties, adsorption energies, and catalytic activity in a single-atom Ru active site. The over potentials on Ru-N-C are 1.12 eV/1.00 eV for ORR/OER. We calculate Gibbs-free energy (ΔG) for every reaction step in ORR/OER process. In order to gain a deeper understanding of the catalytic process on the surface of single atom catalysts, the ab initio molecular dynamics (AIMD) simulations show that Ru-N-C has a structural stability at 300 K and that ORR/OER on Ru-N-C can occur along a typical four-electron process of reaction. AIMD simulations of catalytic processes provide detailed information about atom interactions.

METHODS

In this paper, we use density functional theory (DFT) with PBE functional to study the electronic properties and adsorption properties of graphene-supported nitrogen coordinated Ru-atom (Ru-N-C) Gibbs-free energy and Gibbs-free energy for very reaction step. The structural optimization and all the calculations are carried out by Dmol3 package, adopting the PNT basis set and DFT semicore pseudopotential. Ab initio molecular dynamics simulations (AIMD) were run for 10 ps. The canonical (NVT) ensemble, massive GGM thermostat, and a temperature of 300 K are taken into account. The functional of B3LYP and the DNP basis set are chosen for AIMD.

Selecting High-Dimensional Representations of Physical Systems by Reweighted Diffusion Maps

Constructing reduced representations of high-dimensional systems is a fundamental problem in physical chemistry. Many unsupervised machine learning methods can automatically find such low-dimensional representations. However, an often overlooked problem is what high-dimensional representation should be used to describe systems before dimensionality reduction. Here, we address this issue using a recently developed method called the reweighted diffusion map [J. Chem. Theory Comput. 2022, 18, 7179-7192]. We show how high-dimensional representations can be quantitatively selected by exploring the spectral decomposition of Markov transition matrices built from data obtained from standard or enhanced sampling atomistic simulations. We demonstrate the performance of the method in several high-dimensional examples.

Reweighted Manifold Learning of Collective Variables from Enhanced Sampling Simulations

Enhanced sampling methods are indispensable in computational chemistry and physics, where atomistic simulations cannot exhaustively sample the high-dimensional configuration space of dynamical systems due to the sampling problem. A class of such enhanced sampling methods works by identifying a few slow degrees of freedom, termed collective variables (CVs), and enhancing the sampling along these CVs. Selecting CVs to analyze and drive the sampling is not trivial and often relies on chemical intuition. Despite routinely circumventing this issue using manifold learning to estimate CVs directly from standard simulations, such methods cannot provide mappings to a low-dimensional manifold from enhanced sampling simulations, as the geometry and density of the learned manifold are biased. Here, we address this crucial issue and provide a general reweighting framework based on anisotropic diffusion maps for manifold learning that takes into account that the learning data set is sampled from a biased probability distribution. We consider manifold learning methods based on constructing a Markov chain describing transition probabilities between high-dimensional samples. We show that our framework reverts the biasing effect, yielding CVs that correctly describe the equilibrium density. This advancement enables the construction of low-dimensional CVs using manifold learning directly from the data generated by enhanced sampling simulations. We call our framework reweighted manifold learning. We show that it can be used in many manifold learning techniques on data from both standard and enhanced sampling simulations.

Improving the Accuracy of Atomistic Simulations of the Electrochemical Interface

Atomistic simulation of the electrochemical double layer is an ambitious undertaking, requiring quantum mechanical description of electrons, phase space sampling of liquid electrolytes, and equilibration of electrolytes over nanosecond time scales. All models of electrochemistry make different trade-offs in the approximation of electrons and atomic configurations, from the extremes of classical molecular dynamics of a complete interface with point-charge atoms to correlated electronic structure methods of a single electrode configuration with no dynamics or electrolyte. Here, we review the spectrum of simulation techniques suitable for electrochemistry, focusing on the key approximations and accuracy considerations for each technique. We discuss promising approaches, such as enhanced sampling techniques for atomic configurations and computationally efficient beyond density functional theory (DFT) electronic methods, that will push electrochemical simulations beyond the present frontier.