Skip to content

Development and validation of a ReaxFF reactive force field for Fe/Al/Ni alloys: Molecular dynamics study of elastic constants, diffusion, and segregation

This J. Phys. Chem. A article parameterizes ReaxFF for Fe–Al–Ni alloys and validates it against elastic constants, diffusion, and high-temperature surface segregation in model intermetallic structures.

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 work develops a ReaxFF parameterization for Fe/Al/Ni binary alloys using QM-derived training data, including bulk phases, (100)/(110)/(111) surface energies, and adatom binding energies. Validation includes: (i) elastic constants of FeAl, FeNi₃, and Ni₃Al from 300–1100 K with temperature-dependent softening consistent with experiment; (ii) diffusion in Al/Ni composition-gradient samples (1000 K), finding much larger Al diffusivity at the Al-rich end versus trace layers, while Ni shows only modest end-vs-trace contrast below Ni melting; and (iii) surface segregation in L1₂ clusters at 2500 K, reporting strongest Al segregation in Fe₃Al and weakest in Ni₃Al, plus subsurface depletion layers.

Methods

Force-field training

The authors develop a ReaxFF description for Fe/Al/Ni binary alloys. The QM-derived training set includes bulk alloy phases, (100), (110), and (111) surface energies, and adatom binding energies (abstract; normalized/extracts/2012yun-venue-jp-2012-08507x_p1-2.txt). N/A — program, functional, basis, and k-mesh for those QM references are not recovered from the short local extract; use pdf_path for full computational details. Parameters are optimized so the field reproduces that reference data; validation MD studies below exercise the fitted field.

MD application (atomistic dynamics)

Engine / code: ReaxFF molecular dynamics is used for validation trajectories (abstract). N/A — the integrator package is not named on the indexed excerpt pages; confirm in papers/Yun_JPCA_2012_Alloys_ASAP.pdf. The corpus also holds a galley registration at [[2012yun-venue-research]] (same DOI) whose Methods section names ADF/ReaxFF for part of the protocol when cross-checking numerics.

System size & composition: Supercell sizes and atom counts for elastic, diffusion, and segregation models are given in the full article (not in the p1–2 excerpt); N/A — exact counts from the excerpt alone.

Boundaries / periodicity: Bulk and surface/cluster validation tasks are described at the abstract level; N/A — explicit PBC vs open-face details are not in the indexed excerpt—see pdf_path.

Ensemble: N/A — NVE/NVT/NPT labels are not stated in 2012yun-venue-jp-2012-08507x_p1-2.txt; the galley sibling [[2012yun-venue-research]] documents NPT for elastic constants—verify ensemble labels in this slug’s pdf_path.

Timestep: N/A — \(\Delta t\) not stated in the indexed excerpt.

Duration / stages: N/A — equilibration/production schedules are not stated in the indexed excerpt.

Thermostat: N/A — thermostat type and coupling constants are not stated in the indexed excerpt (the galley Methods on [[2012yun-venue-research]] cite Berendsen with stated damping for some blocks).

Barostat: N/A — not stated in the indexed excerpt.

Temperature: 300–1100 K for elastic constants; 1000 K for Al/Ni diffusion in composition-gradient cells; 2500 K for segregation on L1₂ clusters (abstract).

Pressure / stress: N/A — not stated in the indexed excerpt.

Electric field: N/A — not used (not mentioned).

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

Findings

Outcomes and mechanisms: Elastic constants of FeAl, FeNi₃, and Ni₃Al decrease with temperature from 300 K to 1100 K for C₁₁, C₁₂, and C₄₄, matching experimental trends in the abstract’s summary. Al diffusivity at the pure Al end of an Al/Ni gradient at 1000 K is about two orders of magnitude larger than in Al trace layers, while Ni diffusivity differs only slightly between pure Ni and Ni trace regions at temperatures well below Ni melting (abstract). Surface segregation on L1₂ Fe₃Al, Fe₃Ni, and Ni₃Al clusters at 2500 K shows strongest Al segregation in Fe₃Al and weakest in Ni₃Al, with depletion layers of the segregating species next to the enriched surface (abstract).

Comparisons: Elastic trends and segregation behavior are described as qualitatively consistent with experiment where cited in the article (abstract).

Sensitivity / design levers: Trends are tied to temperature, composition (gradient endpoints vs trace layers), and alloy stoichiometry (which L1₂ cluster).

Limitations (authored / implied): The abstract frames segregation as supporting interplay between bulk phase stability and surface reconstruction; ReaxFF accuracy for alloys remains bounded by the training set.

Corpus / KB honesty: normalized/extracts/2012yun-venue-jp-2012-08507x_p1-2.txt covers abstract and introduction only; timestep, barostat, supercell sizes, and integrator branding require the full pdf_path (or cross-check [[2012yun-venue-research]] for galley Methods numerics at the same DOI).

Limitations

  • High-temperature segregation study conditions are extreme; careful interpretation for near-melting phenomena.

Relevance to group

Key alloy ReaxFF capability paper enabling later oxidation/corrosion studies on complex Fe/Al/Ni systems.

Citations and evidence anchors

  • Abstract and introduction: training set scope, validation pillars (J. Phys. Chem. A; ASAP manuscript header in extract).

Reader notes (navigation)

  • Same DOI as [[2012yun-venue-research]] (galley vs ASAP path); prefer this slug for version-of-record PDF when hashes differ.