Three parallel algorithms for classical molecular dynamics are presented. The first assigns each processor a fixed subset of atoms; the second assigns each a fixed subset of inter-atomic forces to compute; the third assigns each a fixed spatial region. The algorithms are suitable for molecular dynamics models which can be difficult to parallelize efficiently—those with short-range forces where the neighbors of each atom change rapidly. They can be implemented on any distributed-memory parallel machine which allows for message-passing of data between independently executing processors. The algorithms are tested on a standard Lennard-Jones benchmark problem for system sizes ranging from 500 to 100,000,000 atoms on several parallel supercomputers--the nCUBE 2, Intel iPSC/860 and Paragon, and Cray T3D. Comparing the results to the fastest reported vectorized Cray Y-MP and C90 algorithm shows that the current generation of parallel machines is competitive with conventional vector supercomputers even for small problems. For large problems, the spatial algorithm achieves parallel efficiencies of 90% and a 1840-node Intel Paragon performs up to 165 faster than a single Cray C9O processor. Trade-offs between the three algorithms and guidelines for adapting them to more complex molecular dynamics simulations are also discussed.

Fast parallel algorithms for short-range molecular dynamics

The reactive force-field (ReaxFF) interatomic potential is a powerful computational tool for exploring, developing and optimizing material properties. Methods based on the principles of quantum mechanics (QM), while offering valuable theoretical guidance at the electronic level, are often too computationally intense for simulations that consider the full dynamic evolution of a system. Alternatively, empirical interatomic potentials that are based on classical principles require significantly fewer computational resources, which enables simulations to better describe dynamic processes over longer timeframes and on larger scales. Such methods, however, typically require a predefined connectivity between atoms, precluding simulations that involve reactive events. The ReaxFF method was developed to help bridge this gap. Approaching the gap from the classical side, ReaxFF casts the empirical interatomic potential within a bond-order formalism, thus implicitly describing chemical bonding without expensive QM calculations. This article provides an overview of the development, application, and future directions of the ReaxFF method.

https://biblio.ugent.be/publication/8514189/file/8514192.pdf

The ReaxFF reactive force-field: development, applications and future directions

Recent progress in the development of diesel surrogate fuels

Achieving high catalytic performance with the lowest amount of platinum is critical in fuel cell cost reduction. We describe a method of preparing highly active yet stable electrocatalysts containing ultralow Pt content using Co or Co/Zn zeolitic imidazolate frameworks as precursors. Synergistic catalysis between strained Pt-Co core-shell nanoparticles over a platinum-group-metal-free (PGM-free) catalytic substrate led to excellent fuel cell performance under 1 atmosphere of O 2 or air at both high voltage and high current domains. Two catalysts achieved the oxygen reduction reaction (ORR) mass activities of 1.08 A mg Pt −1 /1.77 A mg Pt −1 and retained 64%/15% of initial values after 30,000 voltage cycles in fuel cell. Computational modeling reveals that the interaction between Pt-Co and PGM-free sites improves ORR activity and durability.

https://science.sciencemag.org/content/sci/early/2018/11/07/science.aau0630.full.pdf

Ultralow-loading platinum-cobalt fuel cell catalysts derived from imidazolate frameworks.

The generalization of the Local Density Approximation (LDA) method for the systems with strong Coulomb correlations is presented which gives a correct description of the Mott insulators. The LDA+U method is based on the model hamiltonian approach and allows to take into account the non-sphericity of the Coulomb and exchange interactions. parameters. Orbital-dependent LDA+U potential gives correct orbital polarization and corresponding Jahn-Teller distortion. To calculate the spectra of the strongly correlated systems the impurity Anderson model should be solved with a many-electron trial wave function. All parameters of the many-electron hamiltonian are taken from LDA+U calculations. The method was applied to NiO and has shown good agreement with experimental photoemission spectra and with the oxygen Kα X-ray emission spectrum.

http://absimage.aps.org/image/MAR06/MWS_MAR06-2005-007233.pdf

First-principles calculations of electronic structure and spectra of strongly correlated systems: the LDA+U method

The use of dynamic adaptive chemistry in combustion simulation of gasoline surrogate fuels

We developed the ReaxFF force field for Pt/Ni/C/H/O interactions, specifically targeted for heterogeneous catalysis application of the Pt–Ni alloy. The force field is trained using the DFT data for equations of state of Pt3Ni, PtNi3 and PtNi alloys, the surface energy of the PtxNi1–x(111) (x = 0.67–0.83), and binding energies of various atomic and molecular species (O, H, C, CH, CH2, CH3, CO, OH, and H2O) on these surfaces. The ReaxFF force field shows a Pt surface segregation at x ≥ 0.67 for the (111) surface and x ≥ 0.62 for the (100) surface in vacuum. In addition, from the investigation of the preferential alloy component of the adsorbates, it is expected that H and CH3 on the alloy surface to induce a segregation of Pt whereas the oxidation intermediates and products such as C, O, OH, H2O, CO, CH, and CH2 are found to induce Ni segregation. The relative order of binding strengths among adsorbates is a function of alloy composition and the force field is trained to describe the trend observed in DFT c...

Development of a ReaxFF Reactive Force Field for the Pt-Ni Alloy Catalyst.

Reactive force field methods such as AIREBO, ReaxFF and COMB, are extremely useful for studying physical and chemical interactions between molecules and materials. However, many chemical reactions have relatively high activation energies, putting them beyond the times-scale of conventional molecular dynamics (MD) at modest temperatures. To capture the low-temperature long-lived radical chemistry in atomistic simulations, we have developed a new transition detection scheme for performing Reactive Parallel Replica Dynamics (RPRD) simulation enabling an extended MD time-scales, essentially up to a microsecond using ReaxFF. In the newly implemented event detection scheme, the transition events are identified whenever there is a change in connectivity of any atom. 1-Heptene pyrolysis is chosen as a model system, and RPRD simulations are performed at temperatures as low as 1350K for up to 1 μs for a system consisting of 40 heptene molecules. The chemical mechanism and the product distribution that were obtained...

Connectivity-Based Parallel Replica Dynamics for Chemically Reactive Systems: From Femtoseconds to Microseconds

Hydrogen embrittlement (HE) is a well-known material phenomenon that causes significant loss in the mechanical strength of structural iron and often leads to catastrophic failures. In order to provide a detailed atomistic description of HE we have used a reactive bond order potential to adequately describe the diffusion of hydrogen as well as its chemical interaction with other hydrogen atoms, defects, and the host metal. The currently published ReaxFF force field for Fe/C/H systems was originally developed to describe Fischer–Tropsch (FT) catalysis [C. Zou, A. C. T. van Duin and D. C. Sorescu, Top. Catal., 2012, 55, 391–401], and especially had been trained for surface formation energies, binding energies of small hydrocarbon radicals on different surfaces of iron and the barrier heights of surface reactions. We merged this force field with the latest ReaxFF carbon parameters [S. Goverapet Srinivasan, A. C. T. van Duin and P. Ganesh, J. Phys. Chem. A, 2015, 119, 1089–5639] and used the same training data set to refit the Fe/C interaction parameters. The present work is focused on evaluating the applicability of this reactive force field to describe material characteristics and study the role of defects and impurities in the bulk and at the precipitator interfaces. We study the interactions of hydrogen with pure and defective α-iron (ferrite), Fe3C (cementite), and ferrite–cementite interfaces with a vacancy cluster. We also investigate the growth of nanovoids in α-iron using a grand canonical Monte Carlo (GCMC) scheme. The calculated hydrogen diffusion coefficients for both ferrite and cementite phases predict a decrease in the work of separation with increasing hydrogen concentration at the ferrite–cementite interface, suggesting a hydrogen-induced decohesion behavior. Hydrogen accumulation at the interface was observed during molecular dynamics (MD) simulations, which is consistent with experimental findings. These results demonstrate the ability of the ReaxFF potential to elucidate various aspects of hydrogen embrittlement in α-iron and hydrogen interactions at a more complex metal/metal carbide interface.

Interactions of hydrogen with the iron and iron carbide interfaces: a ReaxFF molecular dynamics study

The reactivity of a metal catalyst depends strongly on the adsorbate coverage, making it essential for the reactivity models to account for the in situ structures and properties of the catalyst under reaction conditions. The use of first principle based thermodynamic approaches to describe adsorbate–adsorbate interaction though attractive for its technical rigor is tedious and computationally demanding especially for metal nanoparticles. With the advent of empirical reactive force fields (ReaxFF), there is a great deal of interest to advance simulation approaches like hybrid grand canonical Monte Carlo reactive molecular dynamics (GCMC/RMD) that enable efficient use of ReaxFF to model the adsorptive states. The predictive ability of GCMC/RMD relies upon the quality of the force field, which in turn depends upon the training set used for its parametrization. To this end, we investigate the adsorption behavior of O and H over the Pt catalysts using the newly developed Pt/O/H ReaxFF. We assess the thermodyna...

Sumathy Raman

Papers

The use of dynamic adaptive chemistry in combustion simulation of gasoline surrogate fuels

Development of a ReaxFF Reactive Force Field for the Pt-Ni Alloy Catalyst.

Connectivity-Based Parallel Replica Dynamics for Chemically Reactive Systems: From Femtoseconds to Microseconds

Interactions of hydrogen with the iron and iron carbide interfaces: a ReaxFF molecular dynamics study

Atomistic Adsorption of Oxygen and Hydrogen on Platinum Catalysts by Hybrid Grand Canonical Monte Carlo/Reactive Molecular Dynamics