Reactive molecular dynamics simulation of fullerene combustion synthesis: ReaxFF vs DFTB potentials

Hu Jun Qian, Adri Van Duin, Keiji Morokuma, Stephan Irle

Research output: Contribution to journalArticlepeer-review

78 Scopus citations


The dynamic fullerene self-assembly process during benzene combustion was studied using classical Reactive Force Field (ReaxFF) nonequilibrium molecular dynamics (MD) simulations. In order to drive the combustion process, the hydrogen to carbon (H/C) ratio was gradually reduced during the course of the MD simulations. Target temperatures of 2500 and 3000 K were maintained by using a Berendsen thermostat. Simulation conditions and hydrogen removal strategies were chosen to match closely a previous quantum chemical MD (QM/MD) study based on the density-functional tight-binding (DFTB) potential (Saha et al. ACS Nano 2009, 3, 2241) to allow a comparison between the two different potentials. Twenty trajectories were computed at each target temperature, and hydrocarbon cluster size, CxHy composition, average carbon cluster curvature, carbon hybridization type, and ring count statistics were recorded as a function of time. Similarly as in the QM/MD simulations, only giant fullerene cages in the range from 155 to 212 carbon atoms self-assembled, and no C 60 cages were observed. The most notable difference concerned the time required for completing cage self-assembly: Depending on temperature, it takes between 50 and 150 ps in DFTB/MD simulations but never less than 100 ps and frequently several 100s ps in ReaxFF/MD simulations. In the present system, the computational cost of ReaxFF/MD is about 1 order of magnitude lower than that of the corresponding DFTB/MD. Overall, the ReaxFF/MD simulations method paints a qualitatively similar picture of fullerene formation in benzene combustion when compared to direct MD simulations based on the DFTB potential.

Original languageEnglish (US)
Pages (from-to)2040-2048
Number of pages9
JournalJournal of Chemical Theory and Computation
Issue number7
StatePublished - Jul 12 2011

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Physical and Theoretical Chemistry


Dive into the research topics of 'Reactive molecular dynamics simulation of fullerene combustion synthesis: ReaxFF vs DFTB potentials'. Together they form a unique fingerprint.

Cite this