Surface Buckling and Subsurface Oxygen: Atomistic Insights into the Surface Oxidation of Pt(111)
Summary¶
Pt(111) oxidation is a textbook surface-science problem, yet reconciling coverage-dependent structures, subsurface oxygen, and temperature-programmed desorption (TPD) experiments across literature datasets remains nontrivial. This ChemPhysChem article develops a unified ReaxFF picture for O/Pt(111) from low to ~1 monolayer-class coverages, including buckled surface oxides, subsurface oxygen, and place-exchange motifs informed by prior DFT surveys. Adri C. T. van Duin coauthors, connecting the work to the Penn State reactive force-field lineage. The authors compute reaction and diffusion barriers among key structures and feed those barriers into a first-principles-based TPD model compared to measured spectra for selected initial oxygen preparations.
Methods¶
MD application (atomistic dynamics)¶
N/A — the paper’s core computations are static ReaxFF energetics + barrier/kinetics analysis feeding a TPD model, not long finite-temperature molecular dynamics trajectories with reported timestep, thermostat, and ns/ps production lengths. ReaxFF energy and force evaluations are used on Pt(111) PBC slab supercells with oxygen at multiple coverages (atom counts in the PDF/SI), without LAMMPS-style production MD, NPT control, applied electric fields, or time-stamped dynamical trajectories.
Force-field training¶
N/A — uses a published Pt/O ReaxFF parameterization (“recently developed” in the sense of the article’s introduction) rather than deriving a new parameter set in this manuscript.
Static QM / DFT (validation layer)¶
- DFT spot checks: the Supporting Information contains DFT validations of formation energies for selected high-symmetry structures along the convex hull (as stated in the main text), used to anchor key ReaxFF energetics.
Kinetics + TPD modeling protocol (as authored)¶
- Structures / pathways: enumerate O/Pt(111) motifs spanning adsorbate-only, buckled high-coverage oxides, subsurface oxygen, and place-exchange-like configurations; construct a convex hull in coverage–energy space to identify stable phases and coexistence regions.
- Barriers / prefactors: perform ReaxFF geometry optimizations and transition-state searches for bottleneck steps; compute ReaxFF vibrational frequencies for TST prefactors (Eq. (4) and Table 1 in the article).
- TPD simulation: integrate a first-principles-based TPD model (Voter-type TST framework as cited) using a 1 K s⁻¹ heating rate and parameters tabulated in Table 1, then compare simulated O₂ TPD traces to experimental spectra for multiple initial coverages.
Findings¶
- Phase diagram / motifs: ReaxFF identifies multiple stable/regime-separated surface phases up to ~1 ML O-equivalent coverage, including buckled and subsurface-O arrangements that reorganize the Pt(111) surface beyond simple adsorbate lattices.
- TPD linkage: the barrier + prefactor network is used to simulate TPD peak structure (including multi-peak composition) in qualitative-to-quantitative agreement with selected experimental datasets referenced in the paper.
- Interpretive claim: buckling and subsurface oxygen are argued to be mechanistically important for reconciling STM/TPD phenomenology that is easy to misread from isolated DFT adsorption studies alone.
- Comparisons: simulated spectra are compared vs published TPD/STM experiments (explicit references in the article).
- Sensitivity: peak shapes depend on heating rate (1 K s⁻¹ in the model), coverage, and the assumed reaction network.
- Limitations / corpus honesty: DFT validation lives largely in SI; cite the PDF/SI for numerical barrier tables.
Limitations¶
ReaxFF accuracy is bounded by its training scope; electrolyte/electrochemical environments are outside the UHV TPD-centric validation shown here.
Relevance to group¶
Demonstrates van Duin-group ReaxFF deployment for noble-metal oxidation with explicit experimental desorption constraints.