Skip to content

Growth of stable surface oxides on Pt(111) at near-ambient pressures

Summary

Oxidation of Pt(111) under oxygen is a classic surface-science benchmark, but many prior mechanistic pictures lean heavily on ultrahigh-vacuum experiments that may miss kinetics accessible under more technologically relevant gas pressures. This Angew. Chem. Int. Ed. study combines near-ambient pressure XPS with ReaxFF grand-canonical Monte Carlo (ReaxFF-GCMC) to argue that stable, largely amorphous surface oxides can form on hour-long time scales at near-ambient \(p(\mathrm{O}_2)\), reconciling apparent contradictions between UHV expectations and catalytic/real-gas environments. The theoretical side seeks structures and coverages consistent with experiment for 430–680 K at 1 mbar \(p(\mathrm{O}_2)\), where brute-force DFT sampling of large disordered oxide motifs is expensive. The local corpus PDF is a galley/proof variant; for final pagination and figure quality, prefer the version-of-record sibling page 2017fantauzzi-venue-untitled when available in-tree.

Methods

A — Force-field training / fitting: Uses a published Pt–O ReaxFF suitable for surface oxide energetics in GCMC; no new parameter optimization summarized here.

B — Molecular dynamics / sampling: ReaxFF-GCMC with grand-canonical O moves evaluates amorphous oxide motifs on Pt(111) at T = 430–680 K and \(p(\mathrm{O}_2)=1\ \mathrm{mbar}\) (abstract-level summary). Equilibrium sampling; kinetic pathways may require dynamics beyond GCMC.

C — DFT / static QM: Not the main engine for large disordered cells—DFT cited where small-cluster or benchmark comparisons apply (full article).

D — Review / non-simulation framing: NAP-XPS tracks oxidation vs time under near-ambient O\(_2\); galley PDF—prefer 2017fantauzzi-venue-untitled for version-of-record alignment.

Atomistic sampling note. Same ReaxFF-GCMC protocol family as 2017fantauzzi-venue-untitled: Monte Carlo O insertion/deletion at fixed \(T\) and \(\mu_\mathrm{O}\) evaluated with ReaxFF, not production MD. Engine: ReaxFF-GCMC (see DOI PDF/SI). System / boundary / MD timestep / MD thermostat / barostat: N/A — GCMC replaces MD integration; PBC and slab sizes are documented in the article rather than duplicated here. Duration / staging: NAP-XPS tracks oxide growth over multi-hour laboratory exposures (abstract excerpt); GCMC sampling length in MC steps/cycles is defined in the article/SI, not as ps/ns MD trajectories. Temperature: 430–680 K (abstract-level pairing with NAP-XPS). Pressure: \(p(\mathrm{O}_2)=1\ \mathrm{mbar}\) in the abstract summary. Electric field: N/A — not used. Replica / enhanced sampling: N/A — not metadynamics/umbrella in the sense of MD enhanced sampling.

Findings

The combined dataset supports slow oxide development on laboratory hour scales, implying that equilibration-limited kinetics can matter when extrapolating from fast UHV protocols. ReaxFF-GCMC predicts stable amorphous oxide films whose coverage and trends align with NAP-XPS under the stated temperature and pressure window, extending prior UHV-focused interpretations toward near-ambient realism. The work positions GCMC as a practical complement to DFT for large-cell oxide disorder on metals, while acknowledging that kinetic growth pathways may require dynamics beyond pure thermodynamic sampling. For practitioners, the paper’s significance is methodological as much as material-specific: it demonstrates a tractable route to oxygen chemical potential–dependent oxide films on a close-packed metal surface where both experiment and simulation can be compared under identical T and p(O₂) statements, reducing the risk of comparing models calibrated in vacuum to measurements taken in reactive gases. The combined NAP-XPS plus ReaxFF-GCMC workflow is also a template for other noble metal oxidation problems where disorder and slow kinetics frustrate brute-force DFT enumeration. Practitioners should still treat GCMC snapshots as thermodynamic baselines: if oxidation is limited by O\(_2\) dissociation or place-exchange kinetics, pure grand-canonical oxygen sampling may need to be complemented by dynamics for fully predictive growth rates.

Limitations

GCMC emphasizes equilibrium sampling; full growth kinetics may require MD or kinetic Monte Carlo extensions. Galley PDFs can differ cosmetically from the version of record.

Relevance to group

van Duin-parameter ReaxFF with GCMC for Pt oxidation under realistic gas pressures; integrated surface catalysis line.

Citations and evidence anchors

Reader notes (navigation)