Evaluating Solvent Effects at the Aqueous/Pt(111) Interface

• Published in: Chemphyschem 2017-08, Vol.18 (16), p.2171-2190
• Main Authors: Iyemperumal, Satish Kumar, Deskins, N. Aaron
• Format: Article
• Language: English
• Subjects: Adsorbates; Adsorption; Artificial neural networks; Catalysis; Chemical bonds; density functional calculations; Density functional theory; Dipole moments; Hydrogen bonding; interfaces; Neural networks; Platinum; Solvation; Solvent effect; solvent effects; Solvents; Surface chemistry
• ISSN: 1439-4235; 1439-7641
• DOI: 10.1002/cphc.201700162

Liquid–metal interfaces occur in a number of surface processes, and it is fundamentally important to accurately model them. Herein, it is systematically determined how the presence of water affects several processes by using density functional theory with implicit solvation models. Adsorption of 41 common adsorbates and four catalytic reactions in both vacuum and water over the Pt(111) surface were modeled. The results show that adsorption energies for some species can change significantly in the presence of water (by up to 0.44 eV). It is further shown that solvation effects can be explained and predicted by analyzing simple chemical descriptors such as dipole moment and adsorbate charge. Models from artificial neural networks involving several potential descriptors, including gas‐phase solvation energy, adsorbate charge, dipole moment, and surface area, are also reported. When water is present, reaction energies change by up to 0.23 eV, although it appears that water solvent negligibly affects several elementary reaction steps. The results show that hydrogen bonding can be important for a number of reactions, but is largely absent in the implicit solvation models. Furthermore, other solvents besides water were also modeled, and if a solvent has a small dielectric constant, then small solvation effects occur. This work provides guidelines on when solvation effects may be important for surface chemistry, and also provides valuable insights into modeling such effects. Solvating the surface: Implicit solvation models within the DFT framework can be used to describe liquid–metal interfaces (see figure) in a computationally cheap manner. Solvation effects were found for many surface‐chemistry processes, including adsorption and several catalytic reactions, and it was also determined which chemical descriptors may be important for predicting solvation effects.