Skip to content

Comparative molecular dynamics study of fcc-Al hydrogen embrittlement

Evidence and attribution

Authority of statements

This file is an Elsevier “Article in Press” PDF for the same study as 2015verners-corrosion-sc-comparative-molecular (DOI 10.1016/j.corsci.2015.05.008). Full Methods detail and primary citation anchors live on that page.

Summary

ReaxFF molecular dynamics in LAMMPS probes hydrogen embrittlement in fcc Al with oxidized surfaces and notched Al/Al₂O₃ interfaces, comparing H tied to initial Al vacancies versus random bulk H. This ingest is a publisher workflow duplicate of the curated version-of-record article; scientific content matches 2015verners-corrosion-sc-comparative-molecular. The duplicate PDF exists so the corpus can preserve “Article in Press” provenance while the wiki still points readers to the paginated sibling for stable citation anchors.

Methods

This Elsevier “Article in Press” PDF (pdf_path) is the same Corrosion Science study as DOI 10.1016/j.corsci.2015.05.008 documented on [[2015verners-corrosion-sc-comparative-molecular]] with the paginated file papers/Verners_AlH_CorrosionScience_2015.pdf. The protocol is LAMMPS ReaxFF on a 3D periodic oxidized, notched fcc Al slab with 162 H either on 162 Al vacancies (paired) or at random bulk sites, 573 K, 0.2 fs, NPT tensile cycling along [1̄10] (0.25% / 0.5 ps strain pulses, 2.5 ps relaxations, 5 × 10⁻⁶ fs⁻¹ nominal rate) with lateral stresses relaxed to 0 Pa via Nosé–Hoover thermostat (100 fs) and barostat (10,000 fs damping). Force-field training: targeted VASP PBE PAW Al/H augmentation (350 eV, 9×9×9 k-mesh) merged into published Al/O/H ReaxFF subsets, as on the canonical page. Static QM: same VASP PBE bulk references tabulated in the VOR article. No electric field or enhanced sampling is used.

Findings

Qualitative conclusions match [[2015verners-corrosion-sc-comparative-molecular]]: H placement shifts decohesion, voiding, and dislocation-mediated plasticity; vacancy-paired H lowers effective H diffusivity versus random bulk H in the supersaturated slab tests summarized in the abstract. ReaxFF metrics are validated against DFT and experiment on the canonical validation table. For figure citations, stress–strain detail, and discussion of nanoscale and ReaxFF limits, use the version-of-record PDF; this Article in Press file is retained for corpus provenance.

Limitations

Prefer the paginated Corrosion Science PDF (2015verners-corrosion-sc-comparative-molecular) for stable pagination and figures; nanoscale slabs and ReaxFF parameter bounds apply as on the canonical page. Hydrogen embrittlement in real Al alloys also involves precipitates, grain boundaries, and hydrogen trapping sites beyond the model notch geometries emphasized in the simulation campaign. Corrosion environments add chloride, pH, and electrochemical potential effects that are not the same problem as the mechanical loading cases tabulated here. Hydrogen entry from aqueous paths may follow different transport channels than bulk interstitial H scenarios emphasized in some slab models.

Relevance to group

Penn State / van Duin ReaxFF application to Al hydrogen embrittlement; this slug preserves non-primary ingest provenance.

Citations and evidence anchors

  • DOI: 10.1016/j.corsci.2015.05.008 — this artifact: papers/Verners_AlH_CorrosionScience_2015_online.pdf.

Reader notes (navigation)