Session C25: Advances in Molecular Dynamics Simulation: From Atomistic to Coarse Grained Models - II

Many-Body Molecular Dynamics: Chemical and Spectroscopic Accuracy from the Gas to the Condensed Phase

Two of the most challenging problems at the intersection of electronic structure theory and molecular dynamics simulations are the accurate representation of intermolecular interactions and the development of reduced-scaling algorithms applicable to large systems. To some extent, these two problems are antithetical, since the accurate calculation of non-covalent interactions typically requires correlated, post-Hartree-Fock methods whose computational scaling with respect to system size precludes the application of these methods to large systems. Our many-body molecular dynamics (MB-MD) methodology for aqueous systems that overcomes these limitations and enables computer simulations from the gas to the condensed phase, with chemical and spectroscopic accuracy. MB-MD is a unified molecular dynamics framework that combines many-body representations for potential energy, dipole moment, and polarizability surfaces, derived entirely from correlated electronic structure data using supervised learning techniques, with quantum dynamics methods that explicitly account for nuclear quantum effects. The accuracy of the MB-MD methodology is assessed through the analysis of several properties of aqueous systems across different phases with a particular focus on nuclear quantum effects and vibrational spectra.   

Rate Theory on Ion Pairings at the Liquid/Vapor Interface of Water

There is overwhelming evidence that ions are present near the vapor\textbf{--}liquid interface of aqueous salt solutions. Despite their importance in many physical phenomena, ways in which ion\textbf{--}ion interactions and related properties, such as the kinetics and thermodynamics, are affected by interfaces have not been quantified. We use molecular simulations to exam the thermodynamics and kinetics of small alkali halide ions in the bulk and near the water vapor\textbf{--}liquid interface. We will calculate dissociation rates using transition state theory, and weight them with transmission coefficients determined by the reactive flux$^{\mathrm{\thinspace }}$method and Grote-Hynes theory. Our results will show that, in addition to affecting the free energy of ions into solution, the interfacial environments (i.e., redistribution of the hydrogen bonds network) and the anion type should significantly influence the kinetics of ion pairings. These results will increase our understanding of thermodynamic and kinetic properties of ion pairing in different environments.   

Friction at ice-Ih / water interfaces

We present evidence that the prismatic and secondary prism facets of ice-I$_{h}$ crystals possess structural features that alter the effective hydrophilicity of the ice / water interface. This is shown through molecular dynamics simulations of solid-liquid friction, where the prismatic $\{10\bar{1}0\}$, secondary prism $\{11\bar{2}0\}$, basal $\{0001\}$, and pyramidal $\{20\bar{2}1\}$ facets are drawn through liquid water. We find that the two prismatic facets exhibit differential solid-liquid friction coefficients when compared with the basal and pyramidal facets. These results are complemented by a model solid/liquid interface with tunable hydrophilicity. These simulations provide evidence that the two prismatic faces have a significantly smaller effective surface area in contact with the liquid water. The ice / water interfacial widths for all four crystal facets are similar (using both structural and dynamic measures), and were found to be independent of the shear rate. Additionally, decomposition of orientational time correlation functions show position-dependence for the short- and longer-time decay components close to the interface.   

First principles molecular dynamics of heterogenous materials

The development of accurate and efficient computational frameworks to predict multiple properties of heterogeneous systems is crucial to building a robust strategy for the scientific design of novel materials. I will present some progress in developing first principles methods for the calculation of spectroscopic and transport properties of heterogeneous materials, by coupling ab initio molecular dynamics with many body perturbation theory calculations. I will discuss their application to specific systems, e.g. materials for solar energy conversion and defective solids for quantum information applications.   

A Gaussian Approximation Potential for Silicon

We present an interatomic potential for silicon using the Gaussian Approximation Potential (GAP) approach, which uses the Gaussian process regression method to approximate the reference potential energy surface as a sum of atomic energies. Each atomic energy is approximated as a function of the local environment around the atom, which is described with the smooth overlap of atomic environments (SOAP) descriptor. The potential is fit to a database of energies, forces, and stresses calculated using density functional theory (DFT) on a wide range of configurations from zero and finite temperature simulations. These include crystalline phases, liquid, amorphous, and low coordination structures, and diamond-structure point defects, dislocations, surfaces, and cracks. We compare the results of the potential to DFT calculations, as well as to previously published models including Stillinger-Weber, Tersoff, modified embedded atom method (MEAM), and ReaxFF. We show that it is very accurate as compared to the DFT reference results for a wide range of properties, including low energy bulk phases, liquid structure, as well as point, line, and plane defects in the diamond structure.   

Computational photoelectron spectroscopy of liquid water

We present an extensive computational study of the electronic properties of liquid water simulated with MB-pol potential [1]. Electronic properties were determined using density-functional and many-body perturbation theory methods. We find that the $G_0W_0$ approximation starting from wavefunctions obtained with dielectric-dependent hybrid functionals [2] provides the best agreement of computed photoelectron spectra with experiment. The inclusion of nuclear quantum effects softens the structure of water, broadens the lines in the spectra and yields narrower band gaps; the use of different methods for modeling water in the absence of quantum effects, such as density-functional approximations [3--4], has a weaker effect on the liquid electronic properties. Remarkably, the positions of the valence band maximum and conduction band minimum are rather sensitive to the presence of the air/liquid interface, underlining the importance of modeling realistic systems for comparison with experiment. \\[8pt] [1]~G.~R. Medders, V. Babin, F. Paesani, JCTC 10, 2906 (2014); [2]~J.~H. Skone, M. Govoni, G. Galli, PRB 89, 195112 (2014); [3]~T.~A. Pham, C. Zhang, E. Schwegler, G. Galli, PRB 89, 060202 (2014); [4]~A.~P. Gaiduk, M. Govoni, R. Seidel, J. Skone, B. Winter, G. Galli, JACS 138, 6912 (2016)   

Molecular dynamics simulations of bimetallic nanocatalysts, guided by X-ray Absorption Fine Structure data

Small bimetallic nanoparticles (with sizes below 4 nm) are a promising material for various applications, e.g., in the field of catalysis, in particular, due to the possibility to tailor their properties by changing their size, shape, chemical composition and intra-particle distributions of both metals. To fully exploit these properties, it is necessary to gain a deep understanding of relation between these degrees of freedom and nanoparticle (NP) atomistic structure and dynamics. It was demonstrated during the last decade that theoretical simulations, such as molecular dynamics (MD) simulations, may be very helpful for this purpose. The structure models, obtained in MD simulations, however, need to be validated by experimental data. In this study we demonstrate that Extended X-ray Absorption Fine Structure (EXAFS) spectroscopy can be successfully used for such validation, since it is a unique, chemically sensitive method that is able to provide detailed information on the distributions of atoms within NPs and on their dynamics. The potentiality of such combined MD-EXAFS approach is illustrated in this study on the example of bimetallic PdAu NPs, synthesized using peptide template method.