First-Principles-Based Multiscale, Multiparadigm Molecular Mechanics and Dynamics Methods for Describing Complex Chemical Processes - Multiscale Molecular Methods in Applied Chemistry - page 20

lifts its center off the particle surface in the experimentally observed Yarmulke
mechanism [ 78 , 88 ]. Currently, ReaxFF RD simulations beginning with several
hundred gas-phase acetylene molecules surrounding a nickel nanoparticle support
the early stages of this picture. Initially, as acetylene chemisorbs and decomposes
on the Ni nanoparticle, the C atoms formed migrate into the bulk of the catalyst and
forming carbide. After a couple nanoseconds of dynamics, the chemical potential
gradient reverses and carbon begins segregating to the surface, forming carbon
chains. As more carbon moves to the surface, ring structures form and clump
together to form larger ring structures, resulting in multi-ring structures with tens
of rings formed from a couple of hundred carbon atoms (see Fig. 7 ). Thus, the
trajectories from these RD simulations provide an atomistically detailed picture of
the early stages of CNT growth.
Following nucleation is the nanotube growth stage in which carbon is added to
the end of the growing nanotube. This stage likely lasts significantly longer than the
previous stages, which means that ReaxFF RD simulations of the entire growth
stage are probably not computationally feasible at present. Nevertheless, a couple
different strategies are available for overcoming this difficulty. The first is to use an
already growing nanotube as the initial structure for ReaxFF simulations, and study
just a part of the growth process. As a simple model we have used ReaxFF to
consider the barriers for adding small hydrocarbon species to the edge of a graphene
sheet laying on a Ni(111) surface. These simulations find the lowest carbon addition
barriers for C 2 hydrocarbon species, suggesting that C 2 may be the activate form of
carbon responsible for CNT growth. Unconstrained ReaxFF RD on a full-scale
model of a growing CNT will provide further validation for this hypothesis.
The second option for circumventing the time limitations on ReaxFF RD is the
use of a kinetic Monte Carlo procedure to bypass long periods of quasi-equilibrium
dynamics between reaction events using principles from statistical mechanics and
Fig. 7 ReaxFF RD simulations of acetylene adsorption and decomposition on a 468-atom nickel
particle [not shown]: (a) after 1 ns a limited number of structured rings have formed and (b) after
2 ns of ReaxFF RD simulations a clear ring pattern formation appears. Simulations were
performed using the parallel prototype reaxFF implementation from collaboration with H.M.
Aktulga and A. Grama at Purdue, and A.C.T van Duin at Penn State