Skip to content

Molecular dynamics based simulations to study the fracture strength of monolayer graphene oxide

Evidence and attribution

Authority of statements

Prose below summarizes the Nanotechnology article identified by doi, title, and pdf_path.

Summary

Mode-I fracture toughness of monolayer graphene oxide (GO) is estimated with ReaxFF MD (LAMMPS) using Chenoweth-type CHO parameters. Central cracks along zigzag vs armchair directions are studied with staged oxidation: hydroxyl, epoxide, and carboxyl groups separately, non-functionalized crack tips, and combined functionalization. \(K_{IC}\) is extracted from virial stress and crack geometry at the first bond rupture; results emphasize strong dependence on coverage and spatial distribution of oxygen groups.

Methods

  • Engine / potential: LAMMPS ReaxFF (Chenoweth et al. parameters cited); neighbor cutoff 4.5 Å, hydrogen-bond cutoff 6 Å; QEq-style charge updates every 10 timesteps.
  • Geometry: Monolayer sheet ~270 × 270 Å (~27 468 atoms), in-plane PBC; central crack 2a = 2.7 nm after convergence tests; sheet outer dimensions ~10× crack length to mitigate finite-size effects.
  • Mechanical test: Relax 12.5 ps at 1 K (NPT), then uniaxial tension along x (ZZ) or y (AC) at strain rate 10\(^{-3}\) ps\(^{-1}\) with NPT (zero transverse stress); effective thickness 7.5 Å (functionalized) vs 3.35 Å (pristine) for stress intensity normalization.
  • Timestep: N/A — integration timestep in fs is not recovered cleanly from the Nanotechnology PDF text extraction used here; confirm the value printed in the Methods of papers/ReaxFF_others/Verma_2018_Nanotechnology_29_115706.pdf (the relaxation and tensile segments above are stated in ps).
  • Thermostat / barostat: NPT is used for relaxation and tensile loading with zero lateral stress as described in the article; the PDF body does not spell out a thermostat brand in the excerpted simulation paragraph—cite the issue text if a specific algorithm is required for reproduction.

  • Post-processing: Virial stress formula in paper; OVITO for visualization.

  • Electric field / replica / metadynamics: N/A — not used (mode-I tensile NPT fracture protocol only).

Findings

  • Validation: Pristine graphene \(K_{IC}\) (~3.69 MPa·m\(^{1/2}\) ZZ; ~3.26 MPa·m\(^{1/2}\) AC) aligns with cited MD and experimental ranges.
  • Hydroxyl coverage: Tabled \(K_{IC}\) decreases/increases with coverage and direction (e.g. 25–75% OH examples in the paper); failure morphologies show blunting, chain formation, and water formation near crack tips at high strain.
  • Epoxide / carboxyl / combined: Trends differ by group type and placement (detailed per §3); overall conclusion: distribution and concentration of functional groups strongly alter fracture response.

Limitations

Single-temperature (1 K) fracture runs reduce thermal broadening but omit finite-temperature noise; ReaxFF uncertainty for oxidized carbon chemistry remains. Crack-tip chemistry can be sensitive to water and environmental species not included in the dry GO models discussed in the article. Finite system sizes and periodic images can also perturb stress fields near cracks compared to macroscopic notch tests. Functional-group distributions in real GO may be spatially correlated at scales larger than the simulation cell used for fracture benchmarks. Nanotechnology tables list orientation-resolved toughness metrics that should be cited when comparing to other oxidized graphene studies.

Relevance to group

Mechanical benchmark for oxidized graphene using ReaxFF in LAMMPS—useful cross-reference for defect and functionalization effects on carbon nanostructure toughness in the broader corpus.