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

A potential function is presented that can be used to model both chemical reactions and intermolecular interactions in condensed-phase hydrocarbon systems such as liquids, graphite, and polymers. This potential is derived from a well-known dissociable hydrocarbon force field, the reactive empirical bond-order potential. The extensions include an adaptive treatment of the nonbonded and dihedral-angle interactions, which still allows for covalent bonding interactions. Torsional potentials are introduced via a novel interaction potential that does not require a fixed hybridization state. The resulting model is intended as a first step towards a transferable, empirical potential capable of simulating chemical reactions in a variety of environments. The current implementation has been validated against structural and energetic properties of both gaseous and liquid hydrocarbons, and is expected to prove useful in simulations of hydrocarbon liquids, thin films, and other saturated hydrocarbon systems.

https://www.usna.edu/Users/chemistry/jah/_files/Papers/AIREBO_JCPversion.pdf

A reactive potential for hydrocarbons with intermolecular interactions

A second-generation potential energy function for solid carbon and hydrocarbon molecules that is based on an empirical bond order formalism is presented. This potential allows for covalent bond breaking and forming with associated changes in atomic hybridization within a classical potential, producing a powerful method for modelling complex chemistry in large many-atom systems. This revised potential contains improved analytic functions and an extended database relative to an earlier version (Brenner D W 1990 Phys. Rev. B 42 9458). These lead to a significantly better description of bond energies, lengths, and force constants for hydrocarbon molecules, as well as elastic properties, interstitial defect energies, and surface energies for diamond.

/pdf/a-second-generation-reactive-empirical-bond-order-rebo-3nq9vxvup4.pdf

A second-generation reactive empirical bond order (REBO) potential energy expression for hydrocarbons

Recent advances in graphene based polymer composites

Electrochemical capacitors, also called supercapacitors, store energy in two closely spaced layers with opposing charges, and are used to power hybrid electric vehicles, portable electronic equipment and other devices¹. By offering fast charging and discharging rates, and the ability to sustain millions of ²⁻⁵, electrochemical capacitors bridge the gap between batteries, which offer high energy densities but are slow, and conventional electrolytic capacitors, which are fast but have low energy densities. Here, we demonstrate microsupercapacitors with powers per volume that are comparable to electrolytic capacitors, capacitances that are four orders of magnitude higher, and energies per volume that are an order of magnitude higher. We also measured discharge rates of up to 200 V s⁻¹, which is three orders of magnitude higher than conventional supercapacitors. The microsupercapacitors are produced by the electrophoretic deposition of a several micrometre-thick layer of nanostructured carbon onions⁶‚⁷ with diameters of 6-7 nm. Integration of these nanoparticles in a microdevice with a high surface-to-volume ratio, without the use of organic binders and polymer separators, improves performance because of the ease with which ions can access the active material. Increasing the energy density and discharge rates of supercapacitors will enable them to compete with batteries and conventional electrolytic capacitors in a number of applications.

/pdf/ultrahigh-power-micrometre-sized-supercapacitors-based-on-28alaqlic6.pdf

Ultrahigh-power micrometre-sized supercapacitors based on onion-like carbon

Molecular dynamics simulations have been used to examine the friction between the hydrogen-terminated (111) faces of diamond with small hydrocarbon (third-body) molecules trapped between them. In general, the presence of the trapped third-body molecules reduced the friction between the diamond surfaces with the most pronounced reduction at high loads. The size and shape of the third-body molecule, as well as the alignment of atoms on opposing diamond surfaces, were found to be paramount in determining the magnitude of the friction. These results are compared to results from previous simulations that examined the effects of chemically bound hydrocarbons on the friction between diamond surfaces and to available experimental data.

Friction between Diamond Surfaces in the Presence of Small Third-Body Molecules

Several facets are observed when diamond films are produced by chemical vapor deposition methods. It is unknown, however, if all of these facets exhibit the same atomic-scale frictional behavior. In an effort to determine whether a facet or surface topological dependence of the atomic-scale friction of diamond exists, we have used molecular dynamics simulations to examine the friction which occurs when the (100)-(2 x 1) reconstructed surfaces of two diamond lattices are placed in sliding contact. Calculations are performed as a function of applied load, crystallographic sliding direction, and sliding velocity. Results from these calculations are compared with previous computations performed on a diamond (111) surface. We find that the general dependence of the friction coefficint, {mu}, on the applied load is similar, regardless of the facet. Indeed, this conclusion is supported by atomic force microscope experiments which have examined the friction of diamond (100) and (111) surfaces. While the friction coefficients, and therefore the amount of energy dissipation, are similar regardless of facet, the atomic-scale motions which lead to energy dissipation differ slightly depending on the facet. 47 refs., 9 figs.

Universal Aspects of the Atomic-Scale Friction of Diamond Surfaces

The elastic constants of diamond between 100 and 1100 K have been calculated for the first time using molecular dynamics and the second-generation, reactive empirical bond-order potential (REBO). This version of the REBO potential was used because it was redesigned to be able to model the elastic properties of diamond and graphite at 0 K while maintaining its original capabilities. The independent elastic constants of diamond, C(11), C(12), and C(44), and the bulk modulus were all calculated as a function of temperature, and the results from the three different methods are in excellent agreement. By extrapolating the elastic constant data to 0 K, it is clear that the values obtained here agree with the previously calculated 0 K elastic constants. Because the second-generation REBO potential was fit to obtain better solid-state force constants for diamond and graphite, the agreement with the 0 K elastic constants is not surprising. In addition, the functional form of the second-generation REBO potential is able to qualitatively model the functional dependence of the elastic constants and bulk modulus of diamond at non-zero temperatures. In contrast, reactive potentials based on other functional forms do not reproduce the correct temperature dependence of the elastic constants. The second-generation REBO potential also correctly predicts that diamond has a negative Cauchy pressure in the temperature range examined.

https://www.usna.edu/Users/chemistry/jah/_files/Papers/Gaoetal_JPhysCMatter_July06.pdf

Elastic constants of diamond from molecular dynamics simulations

A parametrization for silicon is presented that is based on the second-generation reactive empirical bond-order (REBO) formalism [Brenner, Shenderova, Harrison, Stuart, Ni, and Sinnott J. Phys.: Condens. Matter 14, 783 (2002)]. Because it shares the same analytic form as Brenner's second-generation REBO, this new potential is a step toward a single potential that can model many atom systems that contain C, Si, and H, where bond breaking and bond making are important. The widespread use of Brenner's REBO potential, its ability to model both zero-Kelvin elastic constants of diamond and the temperature dependence of the elastic constants, and the existence of parameters for many atom types were the motivating factors for obtaining this parametrization for Si. While Si-C-H classical bond-order potentials do exist, they are based on Brenner's original formalism. This new parametrization is validated by examining the structure and stability of a large number of crystalline silicon structures, by examining the relaxation energies of point defects, the energies of silicon surfaces, the effects of adatoms on surface energies, and the structures of both liquid silicon and amorphous silicon. Finally, the elastic constants of diamond-cubic and amorphous silicon between 0 and $1100\phantom{\rule{0.3em}{0ex}}\mathrm{K}$ are calculated with this new parametrization and compared to values calculated using a previously published potential.

https://www.usna.edu/Users/chemistry/jah/_files/Papers/Schalletal_SiPot_PRBV77.pdf

Elastic constants of silicon materials calculated as a function of temperature using a parametrization of the second-generation reactive empirical bond-order potential

In this study, we explore the wear behavior of amplitude modulation atomic force microscopy (AM-AFM, an intermittent-contact AFM mode) tips coated with a common type of diamond-like carbon, amorphous hydrogenated carbon (a-C:H), when scanned against an ultra-nanocrystalline diamond (UNCD) sample both experimentally and through molecular dynamics (MD) simulations. Finite element analysis is utilized in a unique way to create a representative geometry of the tip to be simulated in MD. To conduct consistent and quantitative experiments, we apply a protocol that involves determining the tip–sample interaction geometry, calculating the tip–sample force and normal contact stress over the course of the wear test, and precisely quantifying the wear volume using high-resolution transmission electron microscopy imaging. The results reveal gradual wear of a-C:H with no sign of fracture or plastic deformation. The wear rate of a-C:H is consistent with a reaction-rate-based wear theory, which predicts an exponential d...

Judith A. Harrison

Papers

Friction between Diamond Surfaces in the Presence of Small Third-Body Molecules

Universal Aspects of the Atomic-Scale Friction of Diamond Surfaces

Elastic constants of diamond from molecular dynamics simulations

Elastic constants of silicon materials calculated as a function of temperature using a parametrization of the second-generation reactive empirical bond-order potential

Atomic-scale wear of amorphous hydrogenated carbon during intermittent contact: a combined study using experiment, simulation, and theory.