Computational studies of nonstoichiometric sodium auride clusters

Area of Science:

• Computational Chemistry
• Materials Science
• Physical Chemistry

Background:

• Sodium auride clusters are of interest due to their unique electronic and structural properties.
• Understanding the molecular structures of these clusters is essential for predicting their behavior and potential applications.
• Previous studies have faced challenges in correlating theoretical predictions with experimental observations for these systems.

Purpose of the Study:

• To computationally investigate the molecular structures of low-lying isomers of anionic and neutral sodium auride clusters.
• To determine the first vertical detachment energies and photodetachment energies for these clusters.
• To compare theoretical calculations with experimental photoelectron spectroscopy data to identify the structures of experimentally observed clusters.

Main Methods:

• Second-order Møller-Plesset perturbation theory (MP2) with quadruple-ζ basis sets and polarization functions.
• Linear response approximate coupled-cluster singles and doubles (LR-CC2) for calculating electronic excitation energies.
• Spin-component-scaled MP2 (SCS-MP2) calculations to re-evaluate isomer ordering.

Main Results:

• Calculated ionization energies were compared with experimental photoelectron spectra for various sodium auride clusters (e.g., NaAu(-), NaAu2(-)).
• The energetically lowest structures from initial calculations did not consistently match experimentally observed clusters.
• SCS-MP2 calculations improved the agreement between calculated and experimental isomer ordering, but did not enhance photodetachment energy accuracy.

Conclusions:

• The potential energy surface of sodium aurides is soft, necessitating accurate electron correlation treatments for structural determination.
• Energetic criteria alone are insufficient for identifying experimentally observed sodium auride cluster structures.
• Combining theoretical calculations with experimental data is vital for accurate assignment of cluster structures.