Skip to content

Unveiling the intricacies of steel corrosion induced by chloride: Insights from reactive molecular dynamics simulation

Scope

Reactive molecular dynamics with a Fe/Cl-augmented ReaxFF (trained against DFT) is used to interpret chloride-induced corrosion of reinforcing steel at the atomistic scale, emphasizing initiation, charge transfer, and corrosion-product formation.

Summary

Chloride-induced corrosion of reinforcing steel in concrete is difficult to resolve experimentally at atomic resolution. Shen and colleagues develop a reactive force field (ReaxFF) incorporating Fe/Cl chemistry using density functional theory (DFT) reference data, then apply ReaxFF molecular dynamics in LAMMPS to follow chloride-driven attack on an Fe(100) surface in alkaline aqueous electrolyte. The introduction frames longstanding debates—whether chloride acts primarily as a catalyst versus a direct initiator—and argues for atomistic models that capture electron transfer, oxide formation, and dissolution dynamics beyond macroscopic coupon tests. This slug is a duplicate ingest of DOI 10.1016/j.conbuildmat.2024.137839 (alternate pdf_path versus 2024shen-construction-unveiling-intricacies); the scientific claims are the same peer-reviewed article.

Methods

1 — MD application (ReaxFF, §2.3). Engine / code: LAMMPS molecular dynamics with ReaxFF. System: Fe(100) in ~34.4 × 34.4 × 66.3 ų with 3168 Fe, 1160 H₂O, 27 Na⁺, 20 Cl⁻, 7 OH⁻ (pH 13.5, 1 mol/L Cl⁻). Boundaries: PBC in x,y; fixed z; reflective upper wall. Ensemble / T: NVT at 300 K with NVT thermostat per §2.3 (see VOR for full coupling). Timestep / duration: Δt = 0.1 fs; 500 ps; snapshots each 1000 steps; OVITO visualization. Barostat, pressure, E-field, enhanced sampling: N/A for the summarized NVT slab run.

2 — Force-field training. Fe/Cl ReaxFF refit against B3LYP-D3(BJ)/6-311++G(2df,2p) cluster and CASTEP GGA-PW91 periodic Fe(100)+Cl data (see 2024shen-construction-unveiling-intricacies for the same equation-of-state style summary).

3 — Static QMtraining/validation roles only; not a stand-alone DFT application report beyond the fit.

(Duplicate DOI and alternate pdf_path: prefer 2024shen-construction-unveiling-intricacies for full parameter tables in one place.)

Findings

Corrosion sequence. Early dynamics show oxide initiation from OH/O interaction at the metal surface, followed by Cl⁻ accumulation that weakens Fe–Fe bonding and promotes Fe dissolution and vacancy formation—consistent with a catalytic chloride role emphasized in the discussion. Charge analysis tracks Fe oxidation, O reduction, and Cl⁻-mediated electron transfer patterns across the 5–500 ps window.

Mechanistic detail (§3.2.2). The paper traces short-lived intermediates (including Fe(OH)⁺ formation and Cl substitution steps) toward Fe(OH)₂-like products, arguing that Cl⁻ can remain surface-catalytic rather than requiring bulk chloride intercalation—supporting catalyst-style pictures over naive bulk chlorination.

RDF and morphology. Fe–O and Fe–Cl radial distribution functions align with literature Fe–Cl bond lengths near ~2.35 Å versus ~2.32 Å experimental averages cited in the text. Surface roughening and pitting progress through nucleation–growth–stabilization stages by 500 ps in the reported trajectory analysis.

Corpus honesty — this duplicate slug mirrors the same DOI; numeric citations should follow 2024shen-construction-unveiling-intricacies when the two PDFs diverge.

Limitations

Prefer 2024shen-construction-unveiling-intricacies for evidence-grounded numbers; this duplicate page exists for manifest/PDF tracking. Atomistic models omit full concrete pore transport and long-time passivation kinetics.

Relevance to group

Co-authorship by Adri C. T. van Duin links the work to the group’s ReaxFF development and corrosion-related applications.

Citations and evidence anchors

Reader notes (navigation)

Canonical Methods/Findings with full protocol: 2024shen-construction-unveiling-intricacies. Theme retrieval: paper-index-by-domain.