Skip to content

Nanoscale oxidation and complex oxide growth on single crystal iron surfaces and external electric field effects

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

This PCCP study uses ReaxFF reactive MD to investigate early-stage oxidation and nanoscale oxide growth on Fe(100), Fe(110), and Fe(111) across temperature and with an optional external electric field (~10 MV/cm). At 300 K over ~1 ns, oxidation kinetics decreases (110) > (111) > (100); higher temperature accelerates oxidation. The mechanism narrative in the excerpt emphasizes early oxygen interstitial transport producing non-stoichiometric wüstite-like regions, evolving toward more stoichiometric wüstite/hematite character as the film thickens, with increasing cation outward transport. Post-growth relaxation between 600–1500 K yields Arrhenius estimates for oxygen diffusion activation energies ~0.32, 0.26, 0.28 eV on (100), (110), (111), respectively. The field accelerates early oxidation via interstitial oxygen transport but approaches a self-limiting thickness; effects on activation energy are described as minimal while slightly promoting cation outward migration.

Introduction motivation ties iron oxidation to corrosion and passivation across technologies and notes that electric fields can appear near surfaces in electrochemical or charged environments, so including ~MV/cm-class fields alongside thermal oxidation tests probes whether field-assisted transport changes early film growth kinetics on low-index facets.

Methods

1 — MD application (atomistic dynamics)

ReaxFF is used for dry oxidation of bcc Fe slabs with ~32 Å metal thickness, vacuum along the surface normal (z), and 3D PBC (\(a=2.870\) Å lattice constant stated). Fe(100) (2662 Fe, 31.57×31.57×31.57 ų), Fe(110) (2816 Fe, 32.47×31.57×32.47 ų), and Fe(111) (3120 Fe, 32.47×35.15×32.31 ų) cells are reported (Sec. 3.2, pdf_path). Substrates are thermalized 10 ps at the oxidation temperature before O₂ insertion (Sec. 3.2).

  • Engine / code: ReaxFF reactive MD; N/A — MD software package not named on the indexed excerpt pages used for this wiki pass.
  • System size & composition: Stoichiometries and supercells as above; O₂ supply via a controlled insertion protocol (pairs inserted when prior oxygen is consumed; randomized x/y placement; initial normal velocities scaled to temperature) (Sec. 3.2, pdf_path).
  • Boundaries / periodicity: Slab + vacuum with PBC (Sec. 3.2).
  • Ensemble: NVT (indexed excerpt + Sec. 4.1 context on pdf_path).
  • Timestep: 1 fs (Sec. 4.1, pdf_path).
  • Duration / stages: Oxidation trajectories run to 1 ns at 300 K and 900 K (Sec. 4.1, pdf_path); additional 600–1500 K relaxation stages are used for diffusion/Arrhenius extraction (Sec. 4.2–4.3, pdf_path).
  • Thermostat / barostat: N/A — thermostat/barostat algorithm names are not recovered from the short indexed excerpt pages—verify pdf_path.
  • Temperature: 300 K and 900 K oxidation; 600–1500 K for post-growth diffusion analysis ( pdf_path).
  • Pressure / stress: N/A — hydrostatic pressure / stress control not emphasized in the excerpted protocol summary.
  • Electric field: 10 MV/cm along the surface normal at 300 K (Sec. 4.2), implemented via an additional energy term \(E_i=\mu q_i r_i\) and forces \(-\nabla E_i\) coupled into charge equilibration (Eqs. (6)–(9), pdf_path).
  • Replica / enhanced sampling: N/A — not indicated in the indexed excerpt.

Electrostatics / cutoffs: ReaxFF nonbonded cutoff 10 Å is stated explicitly (Sec. 3.2, pdf_path).

2 — Force-field training

N/A — this paper applies an Fe–O ReaxFF parameterization and benchmarks reference oxide cells (wüstite, magnetite, hematite) with short NVT relaxations (10 ps, 1 fs) at 300 and 900 K for structural/charge comparisons (Table 1, Sec. 3.1, pdf_path) rather than reporting a new general refit here.

3 — Static QM / DFT-only

N/A — not the paper’s primary methodology beyond citations/context in the full article.

Findings

  • Kinetic stages: fast early oxidation (O ingress through interstitial channels) transitions to slower growth / saturation after ~300 consumed O atoms (transition ~300 ps at 300 K, ~100 ps at 900 K or with field—Sec. 4–5).
  • Facet ordering (1 ns, thermal oxidation): more oxide on (110)/(111) than (100) under the reported conditions; quantitative % differences tabulated in Sec. 4.1.
  • Structure/charge: early films are non-stoichiometric Fe\(_{1-x}\)O-like by PDF/charge analysis; Mott–Cabrera-style field language used to interpret field-assisted early anion motion (Sec. 4.2–5).
  • Diffusion barriers (no field): \(E_a \approx 0.32\), 0.26, and 0.28 eV on (100)/(110)/(111) from 600–1500 K fits (Fig. 11).
  • Diffusion barriers (10 MV/cm): \(E_a \approx 0.33\), 0.24, and 0.23 eV on (100)/(110)/(111)—slightly lower than the no-field fits (Sec. 4.2 / Fig. 11 caption).
  • Field vs no-field at 1 ns: field accelerates early uptake but Fe(100)/(110) can show slightly less total oxide at 1 ns than no-field runs due to later-stage cation reordering/charge coupling (Sec. 4.2).

Limitations

  • Classical reactive FF uncertainty for defect-rich oxides and long-time phase evolution; field representation is model-dependent.

Relevance to group

Strong example of ReaxFF for metal oxidation and electrochemically biased oxide growth—useful alongside other Fe/Ni aqueous oxidation entries.

Citations and evidence anchors

  • Abstract and Sec. 1–2 opening: kinetics ordering, mechanism narrative, diffusion activation energies, field effects (Phys. Chem. Chem. Phys., 2013, 15, 1821–1830; PDF pp. 1–2 per extract).