Genetic Algorithm Driven Force Field Parameterization for Molten Alkali-Metal Carbonate and Hydroxide Salts

Structure and dynamics of Li1.24K0.76CO3 molten carbonate electrolyte from molecular simulations with explicit polarization

Molten carbonate electrolysis cells represent a key technology for harnessing surplus energy from renewable sources and converting it into gaseous energy carriers. To optimize their efficiency, a comprehensive understanding of each step in the operational process is essential. Here, we focus on the electrolyte of choice in molten carbonate cells: the Li1.24K0.76CO3 melt. Utilizing molecular dynamics with explicit polarization, we demonstrate that the structure of this molten mixture is characterized by a dense network of lithium-carbonate complexes, with K+ ions loosely embedded within this network. This structural insight enables us to rationalize from an atomistic perspective the conductivity trends observed experimentally in molten carbonates. Moreover, our work highlights the importance of including polarization for the simulations of dense liquid carbonates. It also acts as a foundational step towards more advanced theoretical studies for elucidating the role of the electrolyte in these devices.

FFParam-v2.0: A Comprehensive Tool for CHARMM Additive and Drude Polarizable Force-Field Parameter Optimization and Validation

Developing production quality CHARMM force-field (FF) parameters is a very detailed process involving a variety of calculations, many of which are specific for the molecule of interest. The first version of FFParam was developed as a standalone Python package designed for the optimization of electrostatic and bonded parameters of the CHARMM additive and polarizable Drude FFs by using quantum mechanical (QM) target data. The new version of FFParam has multiple new capabilities for FF parameter optimization and validation, with an emphasis on the ability to use condensed-phase target data in optimization. FFParam-v2 allows optimization of Lennard-Jones (LJ) parameters using potential energy scans of interactions between selected atoms in a molecule and noble gases, viz., He and Ne, and through condensed-phase calculations, from which experimental observables such as heats of vaporization and free energies of solvation may be obtained. This functionality serves as a gold standard for both optimizing parameters and validating the performance of the final parameters. A new bonded parameter optimization algorithm has been introduced to account for simultaneously optimizing multiple molecules sharing parameters. FFParam-v2 also supports the comparison of normal modes and the potential energy distribution of internal coordinates towards each normal mode obtained from QM and molecular mechanics calculations. Such comparison capability is vital to validate the balance among various bonded parameters that contribute to the complex normal modes of molecules. User interaction has been extended beyond the original graphical user interface to include command-line interface capabilities that allow for integration of FFParam in workflows, thereby facilitating the automation of parameter optimization. With these new functionalities, FFParam is a more comprehensive parameter optimization tool for both beginners and advanced users.

Sensitivity Analysis for ReaxFF Reparametrization Using the Hilbert-Schmidt Independence Criterion

We apply a global sensitivity method, the Hilbert-Schmidt independence criterion (HSIC), to the reparametrization of a Zn/S/H ReaxFF force field to identify the most appropriate parameters for reparametrization. Parameter selection remains a challenge in this context, as high-dimensional optimizations are prone to overfitting and take a long time but selecting too few parameters leads to poor-quality force fields. We show that the HSIC correctly and quickly identifies the most sensitive parameters and that optimizations done using a small number of sensitive parameters outperform those done using a higher-dimensional reasonable-user parameter selection. Optimizations using only sensitive parameters (1) converge faster, (2) have loss values comparable to those found with the naive selection, (3) have similar accuracy in validation tests, and (4) do not suffer from problems of overfitting. We demonstrate that an HSIC global sensitivity is a cheap optimization preprocessing step that has both qualitative and quantitative benefits which can substantially simplify and speed up ReaxFF reparametrizations.

Integrating Machine Learning in the Coarse-Grained Molecular Simulation of Polymers

Machine learning (ML) is having an increasing impact on the physical sciences, engineering, and technology and its integration into molecular simulation frameworks holds great potential to expand their scope of applicability to complex materials and facilitate fundamental knowledge and reliable property predictions, contributing to the development of efficient materials design routes. The application of ML in materials informatics in general, and polymer informatics in particular, has led to interesting results, however great untapped potential lies in the integration of ML techniques into the multiscale molecular simulation methods for the study of macromolecular systems, specifically in the context of Coarse Grained (CG) simulations. In this Perspective, we aim at presenting the pioneering recent research efforts in this direction and discussing how these new ML-based techniques can contribute to critical aspects of the development of multiscale molecular simulation methods for bulk complex chemical systems, especially polymers. Prerequisites for the implementation of such ML-integrated methods and open challenges that need to be met toward the development of general systematic ML-based coarse graining schemes for polymers are discussed.

Modeling Chemical Reactions in Alkali Carbonate-Hydroxide Electrolytes with Deep Learning Potentials

We developed a deep potential machine learning model for simulations of chemical reactions in molten alkali carbonate-hydroxide electrolyte containing dissolved CO₂, using an active learning procedure. We tested the deep neural network (DNN) potential and training procedure against reaction kinetics, chemical composition, and diffusion coefficients obtained from density functional theory (DFT) molecular dynamics calculations. The DNN potential was found to match DFT results for the structural, transport, and short-time chemical reactions in the melt. Using the DNN potential, we extended the time scales of observation to 2 ns in systems containing thousands of atoms, while preserving quantum chemical accuracy. This allowed us to reach chemical equilibrium with respect to several chemical species in the melt. The approach can be generalized for a broad spectrum of chemically reactive systems.

The effect of a mixture of an ionic liquid and organic solvent on oxygen reduction reaction kinetics

Li-O₂ batteries attract great attention due to their promising theoretical energy density. One of the main obstacles on the way to achieving high energy density and good cyclability is positive electrode passivation by the Li₂O₂ discharge product as well as the presence of parasitic reactions that degrade electrode and electrolyte materials. To overcome these issues new electrolytes are being extensively searched for to ensure the bulk-mediated mechanism of the oxygen reduction reaction and inhibition of parasitic reactions. Different additives to organic solvents can significantly change the properties of electrolytes. This work is devoted to the effect of ionic liquids (ILs), which are proposed as an additive to the solvent due to their excellent solvation properties, high stability, low volatility and flammability. Using molecular dynamics simulations we investigate mixtures of the Pyr₁₄TFSI ionic liquid and dimethoxyethane (DME) with different volume fractions of the IL. Our calculations show that the presence of the ionic liquid in the electrolyte stabilises solvation shells around the ions, both involved in the oxygen reduction and parasitic reactions, slowing down the kinetics of Li⁺ and O₂^- association. This makes the usage of such mixtures promising for electrolyte design for Li-O₂ batteries.

Sintering Rate and Mechanism of Supported Pt Nanoparticles by Multiscale Simulation

Thermal stability is the key issue in the industrial application of supported metal nanocatalysts. A combination method of density functional theory calculations, machine learning, and molecular dynamics simulation is adopted to study the sintering behavior of supported platinum (Pt) nanoparticles on graphene or TiO₂ nanosheet, and analyze sintering mechanisms under different temperatures, particle sizes, and metal support interactions (MSIs). The results show that the agglomeration of supported nanoparticles is mainly based on the mechanism of small particle migration and growth. Small-sized particles with high surface energy determine the sintering rate. In addition, the increase of temperature is conducive to the agglomeration of particles, especially for systems with strong MSI. Based on the analysis of the sintering process, a sintering kinetic model of supported Pt nanoparticles related to particle size, temperature, and MSI is established, which provides theoretical guidance for the design of supported metal catalysts with high thermal stability.

Efficient Parametrization of Force Field for the Quantitative Prediction of the Physical Properties of Ionic Liquid Electrolytes

The prediction of transport properties of room-temperature ionic liquids from nonpolarizable force field-based simulations has long been a challenge. The uniform charge scaling method has been widely used to improve the agreement with the experiment by incorporating the polarizability and charge transfer effects in an effective manner. While this method improves the performance of the force fields, this prescription is ad hoc in character; further, a quantitative prediction is still not guaranteed. In such cases, the nonbonded interaction parameters too need to be refined, which requires significant effort. In this work, we propose a three-step semiautomated refinement procedure based on (1) atomic site charges obtained from quantum calculations of the bulk condensed phase; (2) quenched Monte Carlo optimizer to shortlist suitable force field candidates, which are then tested using pilot simulations; and (3) manual refinement to further improve the accuracy of the force field. The strategy is designed in a sequential manner with each step improving the accuracy over the previous step, allowing the users to invest the effort commensurate with the desired accuracy of the refined force field. The refinement procedure is applied on N,N-diethyl-N-methyl-N-(2-methoxyethyl)ammonium bis(trifluoromethanesulfonyl)imide (DEME-TFSI), a front-runner as an electrolyte for electric double-layer capacitors and single-molecule-based devices. The transferability of the refined force field is tested on N,N-dimethyl-N-ethyl-N-methoxyethoxyethylammonium bis(trifluoromethanesulfonyl)imide (N_112,2O2O1-TFSI). The refined force field is found to be better at predicting both structural and transport properties compared to the uniform charge scaling procedure, which showed a discrepancy in the X-ray structure factor. The refined force field showed quantitative agreement with structural (density and X-ray structure factor) and transport properties-diffusion coefficients, ionic conductivity, and shear viscosity over a wide temperature range, building a case for the wide adoption of the procedure.