Water adsorption on stepped ZnO surfaces from MD simulation

  • David Raymand
  • , Adri C.T. van Duin
  • , Daniel Spångberg
  • , William A. Goddard
  • , Kersti Hermansson

Research output: Contribution to journalArticlepeer-review

Abstract

This work presents a ReaxFF reactive force-field for use in molecular dynamics simulations of the ZnO-water system. The force-field parameters were fitted to a data-set of energies, geometries and charges derived from quantum-mechanical B3LYP calculations. The presented ReaxFF model provides a good fit to the QM reference data for the ZnO-water system that was present in the data-set. The force-field has been used to study how water is adsorbed, molecularly or dissociatively, at monolayer coverage on flat and stepped ZnO surfaces, at three different temperatures (10 K, 300 K, and 600 K). The stepped surfaces were created by introducing steps along the (0 0 0 1)-direction on the (1 0 over(1, ̄) 0)-surface. Equilibrium between molecular and dissociated water was observed on the (1 0 over(1, ̄) 0) terraces, resulting in a half dissociated, half molecular water monolayer. The equilibrium between dissociated and molecular water on the surface was found to be reached quickly (<10 ps). When water molecules desorb and the coverage falls, the 1:1 water-hydroxyl ratio is maintained on (1 0 over(1, ̄) 0) terraces, while steps remain largely hydroxylated. The results show that structures that promote hydrogen bonding are favored and that the presence of steps promotes an increased level of hydroxylation in the water monolayers.

Original languageEnglish
Pages (from-to)741-752
Number of pages12
JournalSurface Science
Volume604
Issue number9-10
DOIs
Publication statusPublished - 15 May 2010
Externally publishedYes

Keywords

  • Construction and use of effective interatomic interactions
  • Density-functional calculations
  • Molecular dynamics
  • Solid-gas interfaces
  • Water
  • Zinc oxide

Fingerprint

Dive into the research topics of 'Water adsorption on stepped ZnO surfaces from MD simulation'. Together they form a unique fingerprint.

Cite this