Skip to main navigation Skip to search Skip to main content

Small gold species supported on alumina. A computational study of α-Al2O3(0001) and γ-Al2O3(001) using an embedded-cluster approach

  • Vladimir A. Nasluzov
  • , Tatyana V. Shulimovich
  • , Aleksey M. Shor
  • , Valery I. Bukhtiyarov
  • , Notker Rösch
  • Institute of Chemistry and Chemical Technology of the Siberian Branch of the RAS
  • Siberian Federal University
  • Boreskov Institute of Catalysis SB RAS

Research output: Contribution to journalArticlepeer-review

11 Scopus citations

Abstract

We calculated the structures of and analyzed the bonding in adsorption complexes of small gold species Aun on α-Al2O3(0001), n = 1-6, and γ-Al2O3(001), n = 1-5. We applied a scalar-relativistic gradient-corrected density functional (DF) method to cluster models of the support that were embedded in an extended elastic polarizable environment (EPE). The shortest Au-O distances, 204-211 pm, are consistent with extended X-ray adsorption fine structure (EXAFS) data for gold clusters on alumina surfaces. The calculated total adsorption energies increase with cluster nuclearity, up to n=4, but drop for larger adsorbed species. In the gas phase, these small gold clusters exhibit a planar structure which they keep, oriented parallel to the surface, as adsorbates on α-Al2O3(0001). Unfavorable energy contributions result for larger clusters as their planar shape is notably distorted by the interaction with the support which amounts to 0.5-1.5 eV. On γ-Al2O3(001), also the larger gold clusters retain their intrinsic planar structure as they adsorb oriented perpendicular to the surface. The corresponding adsorption energies are slightly smaller, 0.3-1.2 eV.

Original languageEnglish
Pages (from-to)1023-1031
Number of pages9
JournalPhysica Status Solidi (B) Basic Research
Volume247
Issue number5
DOIs
StatePublished - May 2010

Fingerprint

Dive into the research topics of 'Small gold species supported on alumina. A computational study of α-Al2O3(0001) and γ-Al2O3(001) using an embedded-cluster approach'. Together they form a unique fingerprint.

Cite this