Molecular dynamics simulation of chemical vapor deposition: graphene growth on Ni(111)

Evidence and attribution¶

Authority of statements

Prose below summarizes the publication identified by doi, title, and pdf_path in the front matter. For exact numerical values and figures, use the peer-reviewed article.

Summary¶

Classical molecular dynamics with a ReaxFF parametrization for C/Ni is used to study graphene evolution and growth kinetics on Ni(111) at several temperatures. The abstract reports that low carbon concentration favors dissolution of C into Ni, while higher concentration leads to graphene island formation; defect annealing near ~1000 K improves island quality; islands can grow by incorporating deposited C and forming hexagons at edges with self-healing—framed as guidance for CVD control.

Corpus note: another ingest of the same JPCC article is registered as [[2012meng-venue-jp212149c]] (alternate PDF path).

Methods¶

Force field and integration: ReaxFF for C/Ni, using parameters developed for that system and validated against quantum calculations in the cited training work. Velocity Verlet integration with Δt = 0.25 fs (chosen for stable energy conservation near T ~ 1000 K). Berendsen thermostat with 100 fs damping.

Surface model: Four-layer Ni(111) slab, 256 Ni total (64 per layer); in-plane orientation uses x ∥ [110] and y along the [1̄12] direction (Miller indices as printed in the article); ~20 Å vacuum along z; periodic boundaries in-plane; bottom slab layer fixed to mimic a semi-infinite bulk.

Deposition / conditions: 16, 32, or 64 C atoms (⅛, ¼, ½ of a full graphene monolayer over the cell area), randomly deposited on the surface. Annealing temperatures 800, 1000, 1200, and 1400 K; each (n C atoms, T) labeled C_n@T in the article. Production trajectories reported in the article use 100 ps annealing segments at the stated temperatures for structure evolution (e.g., Figure 1 at 1000 K).

MD application (ReaxFF CVD-inspired annealing)¶

Engine / code: Classical molecular dynamics with ReaxFF (normalized/extracts/2012lijuan-meng-j-phys-chem-jp212149c_p1-2.txt); N/A — standalone program name not recovered from the indexed excerpt—verify pdf_path.

System size & composition: Four-layer Ni(111) slab with 256 Ni atoms (64 per layer) plus 16, 32, or 64 deposited C atoms.

Boundaries / periodicity: Periodic boundary conditions in-plane; ~20 Å vacuum along z; bottom Ni layer fixed to mimic a semi-infinite catalyst.

Ensemble: NVT-style thermal control via a Berendsen thermostat (canonical-sampling intent as described in the article).

Timestep: 0.25 fs (velocity Verlet).

Duration / stages: 100 ps segments at 1000 K for the C-structure evolution example in Figure 1; other C_n@T combinations follow the article’s schedule.

Thermostat: Berendsen thermostat with 100 fs damping constant (Section II of the article).

Barostat / pressure control: N/A — NPT barostat not stated for these slab annealing runs.

Temperature: 800, 1000, 1200, and 1400 K labels for the C_n@T survey.

Pressure / stress: N/A — external pressure control not described in the indexed excerpt.

Electric field: N/A — not used.

Replica / enhanced sampling: N/A — not used.

Force-field training¶

N/A — this work applies an existing C/Ni ReaxFF parametrization with prior QM validation as cited in the article; it does not report a new ReaxFF fit in pdf_path.

Findings¶

At low C loading (16 atoms at 1000 K in the discussion excerpt), C monomers favor subsurface sites because the surface → subsurface barrier is only ~0.6 eV, so dimers on the surface are scarce, whereas dimers and trimers on the surface can lock Ni into bridging motifs with large effective diffusion barriers. At higher C loadings, the paper’s abstract-level summary still applies: low concentration tends toward dissolution into Ni, high concentration drives graphene island formation, ~1000 K is highlighted for defect annealing, and islands can grow by capturing added C and forming hexagons at edges with self-healing-like behavior—framed as kinetic insight for CVD control versus prior static nucleation studies.

Limitations¶

ReaxFF is noted in the article as less accurate than quantum methods for the same observables; finite slab size, time span, and deposition protocol bound transferability to industrial CVD.

Relevance to group¶

Independent ReaxFF application to graphene-on-Ni growth kinetics; useful alongside group work on reactive carbon and metal interfaces.

Citations and evidence anchors¶

DOI 10.1021/jp212149c — J. Phys. Chem. C 116, 6097–6102 (2012).
PDF: papers/ReaxFF_others/Meng_Ding_Ni_graphene_JPCC_2012.pdf; extract: normalized/extracts/2012lijuan-meng-j-phys-chem-jp212149c_p1-2.txt.