Skip to content

Atomistic-scale simulations of the chemomechanical behavior of graphene under nanoprojectile impact

Evidence and attribution

Authority of statements

Prose sections below (Summary, Methods, Findings, etc.) are curated summaries of the publication identified by doi, title, and pdf_path in the front matter above. They are not new primary claims by this wiki.

For definitive numerical values, reaction schemes, and interpretations, use the peer-reviewed article (and optional records under normalized/papers/ when present)—not this page alone.

Summary

The study uses ReaxFF to simulate supersonic impact of nanoscale silica and nickel projectiles on single-layer graphene, enabling bond-breaking chemistry at the projectile–graphene interface beyond fixed-bond carbon potentials. Reported penetration energies \(E_p^*\) are related to pre-crack deformation, defect content (mono-vacancies, grain boundaries), and projectile chemistry, including crack-edge structures such as pentagon–heptagon pairs during penetration. Simulated \(E_p^*\) values are compared with multilayer graphene experiments (Lee et al., Science 2014), supporting large energy absorption under extreme strain-rate loading.

Methods

LAMMPS ReaxFF (papers/Yoon_Carbon_2015.pdf, §2) merges C-2013 carbon parameters with published Si/O/C/H and Ni/O/C/H subsets for projectile chemistry. Hydrogen-terminated single-layer graphene sheets (~380 Å × 360 Å) are pristine or contain mono-vacancies or polycrystalline patches (~25 Å grain size). Spherical amorphous silica (~48 Å diameter) and fcc Ni (~38 Å) projectiles impact along the surface normal. In-plane PBC, H-passivated edges with edge C atoms fixed mimic supported samples. After 0.1 K minimization and 300 K equilibration (including defective models where noted), production is NVE microcanonical penetration with 0.05 fs timestep and initial projectile speed 5 km s⁻¹ (authors state this exceeds typical ~1 km s⁻¹ experiments to remain tractable at the modeled scale). Pressure in the shock direction is not a fixed GPa setpoint but emerges from the NVE piston; lateral stress relaxes according to the periodic slab constraints in §2. No thermostat or barostat during NVE impact; no electric field or enhanced sampling. Specific penetration energy \(E_p^\*\) comes from projectile kinetic-energy loss per mass; armchair vs zigzag edges and 5|7 rings are tracked (abstract, §3).

Force-field training: N/A — literature merges are used.

Static QM / DFT: N/A — not the primary modality.

Findings

ReaxFF captures reactive oxidation and fracture mechanisms for Ni or silica impact on graphene, beyond fixed-bond carbon models (papers/Yoon_Carbon_2015.pdf). Simulated \(E_p^\*\) is the same order of magnitude as Lee et al., Science 2014 experimental multilayer-graphene microparticle impacts—a benchmark comparison noted in the abstract. Pentagon–heptagon defects appear at crack edges; pre-crack deformability and defect content (mono-vacancy, grain boundary) sensitivity shifts \(E_p^\*\) and morphology, as does projectile chemistry (Ni vs silica). Limitations in §2 include single-layer models versus multilayer tests and higher simulated impact speed than typical experiment. Corpus honesty: tabulated \(E_p^\*\) values and time-resolved reaction counts belong in the journal PDF figures; this page stays at abstract precision.

Limitations

  • ReaxFF C-2013 tensile response differs from REBO families at moderate strain; the reported impact studies operate in large-strain fracture regimes where this parametrization was judged applicable in the original work.
  • Direct comparison to experiment is indirect (multilayer experimental graphene vs simulated single-layer).

Relevance to group

Demonstrates ReaxFF for 2D carbon under extreme mechanical loading with van Duin group leadership—useful reference for reactive simulations of graphene in contact with metals/oxides.

Citations and evidence anchors