Simulation of Metal-Supported Metal-Nanoislands: A Comparison of DFT Methods

Area of Science:

• Computational materials science
• Surface science
• Quantum chemistry

Background:

• Accurate simulation of metal nanoparticles on surfaces is crucial for catalysis.
• Density functional theory (DFT) is a primary tool, but method choice impacts results.
• Understanding the performance of various DFT functionals is essential for reliable predictions.

Purpose of the Study:

• To evaluate and compare different DFT methods for simulating supported metal nanoparticles.
• To assess the accuracy of various functionals for geometric, energetic, electronic, and adsorption properties.
• To identify the most suitable DFT approaches for modeling platinum (Pt) and palladium (Pd) nanoislands on Au(111).

Main Methods:

• Employed a range of DFT functionals: GGA (PW91, PBE, RPBE, revPBE, PBESol), vdW-corrected GGA (PBE-D3, revPBE-vdW), meta-GGA (SCAN, MS2), and machine learning (BEEF-vdW).
• Utilized Pt and Pd nanoislands on Au(111) as model systems.
• Investigated geometric, energetic, electronic, and hydrogen adsorption properties.

Main Results:

• Most DFT methods provided consistent results for geometric and electronic properties of Pt and Pd nanoislands on Au(111).
• Relative energetics of small Pt and Pd clusters showed similar trends across methods, with specific conformational preferences identified (linear for triatomic Pt, triangular for triatomic Pd, non-linear for larger clusters).
• Significant discrepancies among DFT methods were observed solely for hydrogen adsorption energies.

Conclusions:

• The choice of DFT functional has minimal impact on the simulation of geometric and electronic properties of supported Pt and Pd nanoislands.
• Conformational preferences of small metal clusters on surfaces are generally well-captured by various DFT methods.
• Accurate prediction of hydrogen adsorption energies requires careful selection of DFT functionals, as results vary considerably between methods.