Skip to content

Atomistic-scale analysis of carbon coating and its effect on the oxidation of aluminum nanoparticles by ReaxFF molecular dynamics simulations

Summary

A ReaxFF parametrization for Al/C/H/O is developed using VASP DFT and Jaguar QM training sets, then applied in ReaxFF molecular dynamics (implemented in ADF) to model hydrocarbon-derived carbon coating on aluminum nanoparticles (ANPs) and subsequent O₂ oxidation, comparing to prior experiments on laser/plasma carbon coating.

Methods

  • QM training (periodic): VASP GGA-PBE with PAW potentials, 400 eV cutoff; Al(111) slabs (6 layers, bottom layers fixed) for hydrocarbon adsorption/decomposition; k-meshes include 8×8×8 (bulk Al₄C₃) and 5×5×1 (slab adsorbates); NEB (L-BFGS, 3 images) for barriers.
  • QM clusters: Jaguar B3LYP/6-311G for nonperiodic Al/C/H/O geometries; bond/angle scans for Al–C dissociation and distortion energies.
  • ReaxFF optimization: Sequential refinement of Al–C, Al/H, and angle/off-diagonal terms to match QM targets; ReaxFF-NEB cross-checks vs DFT-NEB for selected pathways (e.g., C–C cleavage unfavorability on Al(111)).
  • MD protocol (coatings / oxidation): NVT with Berendsen thermostat (100 fs damping); time step 0.1 fs (motivated by high T up to ~3000 K); executed with ADF ReaxFF molecular dynamics (4–8 processors stated). Hydrostatic pressure / NPT segments: N/A — not used in the summarized coating/oxidation stages (constant-volume NVT).
  • Coating cycles: 864 Al ANP in 45×45×45 ų box with 350 gas molecules per cycle; ANPs at 300 K, hydrocarbon gas at 2500 K for 15 ps, cool to 300 K in 8.5 ps; repeat cycles; precursors ethylene, ethane, acetylene compared.
  • Oxidation: Carbon-coated ANP + 600 O₂ in 60×60×60 ų; 300 K vs 3000 K snapshots to 150 ps; elevated O₂ density ~0.15 g/cm³ to accelerate kinetics within short MD windows.

Findings

  • ReaxFF reproduces key QM interaction energies/barriers and shows hydrocarbon deposition proceeds primarily by H transfer to Al sites with no C–C bond breaking in the ethylene coating runs (up to six cycles).
  • Carbon coverage and C/Al ratio trends vs precursor follow binding strength ordering (acetylene > ethylene > ethane), modulating coating thickness.
  • Oxidation MD: At 300 K, oxidation products remain limited; at 3000 K, gas-phase H₂O, H₂, CO, CO₂ form as the carbonaceous layer is stripped, increasing exposed Al and O₂ uptake—consistent with temperature-dependent protection described experimentally.

Limitations

  • Corpus PDF is an ACS author proof (Hong_AlCOx_JPCC_2016_proof.pdf); cite the journal version-of-record for pagination and final copy edits.
  • Short oxidative timescales require accelerated O₂ density; reported oxidation kinetics are qualitative indicators within the MD window.

Relevance to group

van Duin group ReaxFF parametrization + application on Al combustion nanoparticles with explicit coating chemistry.

Citations and evidence anchors

Reader notes (navigation)