Skip to content

Effects of oxidation on tensile deformation of iron nanowires: Insights from reactive molecular dynamics simulations

Summary

Aral et al. couple oxidation chemistry with tensile deformation of iron nanowires at room temperature using variable-charge ReaxFF molecular dynamics, comparing inert/vacuum conditions to dry O\(_2\) exposure that builds oxide shells on the Fe surface. The study asks how oxide thickness modifies yield stress, critical strain, twinning, and the onset of plasticity in nanoscale Fe, where surface effects already dominate mechanical response even without oxidation. Adri C. T. van Duin is a coauthor, and the work exemplifies the group’s use of reactive MD to treat metal–oxygen bond formation concurrently with mechanical loading, rather than preoxidizing to a fixed stoichiometry and then deforming rigidly.

Methods

MD application (atomistic dynamics)

Force field: Variable-charge ReaxFF in LAMMPS (parallel implementation cited in §II) with EEM/QEq-style charge updates each step as described in the article.

Oxidation stage (build oxide shells on bcc Fe nanowires)

  • System: Cylindrical Fe nanowire, bcc lattice constant a = 2.86 Å, 24,050 Fe atoms, diameter 5.0 nm, length 14.3 nm along [001] (tensile axis in later mechanical tests).
  • Gas environment: 2000 O₂ molecules placed randomly above the wire surface to represent dry O₂ exposure.
  • Target oxide thicknesses: Growth is halted at three nominal oxide shell thicknesses quoted in the abstract (≈4.81 Å, ≈5.33 Å, ≈6.57 Å) by removing remaining gas-phase oxygen once the desired shell is achieved (§III).
  • Ensemble / thermostat: NVT at T = 300 K with a Nosé–Hoover thermostat chain applied to all atoms during oxidation; 3D periodic boundaries (§III).
  • Duration (example trajectory): The text reports an oxidation segment consuming 2870 O atoms over 3.35 ns for the illustrated core–shell formation case (figures in §III/SI).
  • Barostat / pressure: N/A — oxidation segment is NVT without documented hydrostatic pressure targeting beyond periodic gas/solid setup.

Mechanical testing (tensile pulls on pure and oxide-coated wires)

  • Preparation: Conjugate-gradient energy minimization for each wire; NPT relaxation at 300 K to relieve construction stresses (§III).
  • Tensile MD: NVT at 300 K with Nosé–Hoover thermostat; velocity Verlet integration with Δt = 0.5 fs; charges updated every MD step as in ReaxFF/EEM (§III).
  • Strain / duration: Uniaxial constant strain rate 0.01% ps⁻¹ (10⁸ s⁻¹) applied along [001] by rescaling the periodic box up to 16% engineering strain; total tensile segment 1.5 ns (§III).
  • Stress: Virial atomic stresses averaged and corrected to engineering stress using the true wire volume (§III).
  • Boundaries after oxidation: For tensile, PBC only along the axial (loading) direction with free lateral boundaries to reduce spurious lateral stress transmission (§III).
  • Electric field: N/A — not used.
  • Replica / enhanced sampling: N/A — not used.

Force-field training

N/A — not a parameterization paper; the manuscript applies an established ReaxFF Fe/O description for coupled oxidation + mechanics (see references in §II for parameter origins).

Static QM / DFT

N/A — no central DFT production results for the nanowire oxidation/tension trajectories in the sense of AGENTS block 3; QM appears only as background for ReaxFF validation references.

Findings

  • Mechanical softening with oxidation: Increasing oxide shell thickness on the modeled Fe nanowires reduces yield stress and critical strain relative to thinner shells / cleaner surfaces (abstract + §IV narrative).
  • Earlier plasticity: Oxidized wires yield at lower applied strain and require lower external stress to initiate plasticity, i.e., the oxide shell shifts the onset of plastic deformation earlier under the simulated conditions.
  • Twinning: Twinning remains important for both bare and coated wires, but twin nucleation occurs at lower strain when oxide thickness increases (abstract-level claim; see §IV for microstructural snapshots and stress–strain curves).
  • Oxidation kinetics context: The oxidation portion is described as showing fast initial oxidation followed by slow growth consistent with logarithmic-growth discussions in the article (§IV; see SI figures referenced there).

Limitations

Nanowire models omit grain boundaries and alloying common in structural steels; strain rates are MD-accessible and may exceed experiment. When comparing to oxidation-first experiments, note that the simulation protocol may pre-oxidize or co-oxidize relative to real atmospheric exposure schedules.

Relevance to group

Direct van Duin-group collaboration on ReaxFF modeling of metal oxidation coupled to nanomechanics.

Citations and evidence anchors