Session A19: Invited Session: Fifty Years of Molecular Dynamics Simulations I: Past, Present and Future

Combining Molecular Dynamics and Density Functional Theory

The time evolution of a system consisting of electrons and ions is often treated in the Born-Oppenheimer approximation, with electrons in their instantaneous ground state. This approach cannot capture many interesting processes that involved excitation of electrons and its effects on the coupled electron-ion dynamics. The time scale needed to accurately resolve the evolution of electron dynamics is atto-seconds. This poses a challenge to the simulation of important chemical processes that typically take place on time scales of pico-seconds and beyond, such as reactions at surfaces and charge transport in macromolecules. We will present a methodology based on time-dependent density functional theory for electrons, and classical (Ehrenfest) dynamics for the ions, that successfully captures such processes. We will give a review of key features of the method and several applications. These illustrate how the atomic and electronic structure evolution unravels the elementary steps that constitute a chemical reaction.   

Material Discovery and Design with Dynamic Charge Reactive Potentials

Atomic scale computational simulations of multi-phase systems is increasingly important as our ability to simulate nanometer-sized systems becomes routine. The recently developed charge optimized many body potential (COMB) potentials have significantly enhanced our ability to carry out atomic-scale simulations of heterogeneous material systems. The formalism of this potential combines variable charge electrostatic interactions with a classical analytical bond-order potential. It therefore has the capacity to adaptively model metallic, covalent, ionic, and van der Waals bonding within the same simulation cell and dynamically determine the charges on individual atoms according to the local environment. The utility of the COMB potentials is illustrated for materials design and discovery by exploring the structure, stability, mechanical properties, and thermal properties of intermetallic systems and oxide-metal interfaces. They are also used to address key questions associated with corrosion, thin film growth, and heterogeneous catalysis.   

Massively Parallel Reactive and Quantum Molecular Dynamics Simulations

In this talk I will discuss two simulations: Cavitation bubbles readily occur in fluids subjected to rapid changes in pressure. We use billion-atom reactive molecular dynamics simulations on a 163,840-processor BlueGene/P supercomputer to investigate chemical and mechanical damages caused by shock-induced collapse of nanobubbles in water near silica surface. Collapse of an empty nanobubble generates high-speed nanojet, resulting in the formation of a pit on the surface. The gas-filled bubbles undergo partial collapse and consequently the damage on the silica surface is mitigated. Quantum molecular dynamics (QMD) simulations are performed on 786,432-processor Blue Gene/Q to study on-demand production of hydrogen gas from water using Al nanoclusters. QMD simulations reveal rapid hydrogen production from water by an Al nanocluster. We find a low activation-barrier mechanism, in which a pair of Lewis acid and base sites on the Al$_{n}$ surface preferentially catalyzes hydrogen production. I will also discuss on-demand production of hydrogen gas from water using and LiAl alloy particles. Research reported in this lecture was carried in collaboration with Rajiv Kalia, Aiichiro Nakano and Ken-ichi Nomura from the University of Southern California, and Fuyuki Shimojo and Kohei Shimamura from Kumamoto University, Japan.   

Small is Different: Nanoscale Computational Microscopy

Finite materials systems of reduced sizes exhibit discrete quantized energy level spectra and specific structures and morphologies, which are manifested in unique, nonscalable, size-dependent physical and chemical properties. Indeed, when the scale of materials structures is reduced to the nanoscale, emergent phenomena often occurs, that is not commonly expected, or deduced, from knowledge learned at larger sizes. Characterization and understanding of the size-dependent evolution of the properties of materials aggregates are among the major challenges of modern materials science. Computer-based classical and quantum computations and simulations are tools of discovery of nanoscale emergent behavior [1]. We highlight such behavior in diverse systems, including: (i) Atomistic simulations of nanoscale liquid jets and bridges and the stochastic hydrodynamic description of their properties [2]; (ii) Metal nanoclusters and their self-assembled superlattices exhibiting stabilities and properties originating from superatom electronic shell-closing, atom packing, and interactions between protecting ligands [3]; (iii) Electric-field-induced shape-transitions and electrocrystallization of liquid droplets [4], and (iv) Symmetry-breaking and formation of highly-correlated Wigner molecules between electrons in 2D quantum dots and bosons in traps [5]. \\[4pt] [1] (a) U. Landman, ``Materials by Numbers,'' \textit{Proc. Nat. Acad. Sci. (USA)} \textbf{102}, 6671 (2005). (b) U. Heiz {\&} U. Landman, \textit{Nanocatalysis}(Springer, 2006). \\[0pt] [2] M. Moseler, U. Landman, \textit{Science} \textbf{289}, 1165 (2000).\\[0pt] [3] A. Desireddy, et al., \textit{Nature} \textbf{501}, 399 (2013); B. Yoon, \textit{Nat. Mater.,} \textbf{13}, 807 (2014).\\[0pt] [4] W.D. Luedtke, J. Gao, U. Landman, \textit{J. Phys. Chem. C}\textbf{ 115}, 20343 (2011).\\[0pt] [5] C. Yannouleas, U. Landman, \textit{Rep. Prog. Phys.} \textbf{70}, 206 (2007); \textit{PRB } \textbf{84}, 165327 (2011).