A study of Cl adsorption on Pt(111) and Pt(100) using Ab Initio Grand-canonical Monte Carlo

We used ab initio grand-canonical Monte Carlo (AIGCMC) simulations based on plane-wave density-functional theory to probe the structures and surface energies of Pt(100) and Pt(111) with adsorbed chlorine. For Pt(100), we considered both the (1 × 1) surface and a (5 × 1) reconstruction, as a model for the experimentally observed “hex” reconstruction of Pt(100). We constructed phase diagrams of the surface energies as function of the Cl chemical potential and identified the most relevant surfaces. For Pt(100), we find the hex reconstruction is favored at low Cl chemical potentials and that Cl adsorption lifts the reconstruction. The progression of ordered structures predicted for this surface is: bare (5 × 1) Pt(100), Θ = 1/2 (1 × 1) Pt(100), and Θ = 2/3 (1 × 1) Pt(100), where Θ is the fractional surface coverage of Cl. All these structures are seen experimentally. We also observe a structure with Θ = 3/4 and intermixing between Pt and Cl on Pt(100) that is related to the structure at Θ = 2/3. For Pt(111), we find a progression of (3 × 3) unit cells at Θ = 1/9, 1/3, 4/9, 5/9, and 2/3. The structures at Θ = 1/3 and 4/9 have been proposed experimentally and most experiments predict a series of (3 × 3) unit cells with increasing Cl coverage. If intermixing between Cl and Pt does not occur in experiment, then we find a (4 × 2) Cl structure at Θ = 1/2 is energetically favored, as is observed in experiment. A strength of AIGCMC is the capability to identify relevant structures, including disordered structures, without predefined input. This increases the chance of having high fidelity to experiment and identifying relevant substrates for applications.