Coordination modes and bonding of sulfur oxides on transition metal surfaces: combined ab initio and BOC-MP results

Abstract

Binding energies for sulfur oxides, SO x , x = 1–3, have been determined for several coordination modes on silver, gold and palladium surfaces employing ab initio quantum chemical methods and the bond order conservation Morse potential (BOC-MP) method. SO 2 coordination was studied in the most detail. In general the agreement between the BOC-MP and ab initio binding energies is good for the (111) surfaces of silver and palladium with both methods predicting that, in the zero coverage limit, di-coordination via S,O and O,O will be more favorable energetically than mono-coordination via S. In the case of chemisorption on the Pd (110) surface the two methods agree well for the cases in which there are formulas for the BOC-MP binding energies. In going from the (111) surfaces to the (110) surfaces of silver and palladium the ab initio calculations predict that the preferred chemisorption site shifts from the bridge site to the hollow site. On the silver surfaces the net charge transferred to the adsorbate as judged from the Mulliken populations correlates roughly with the binding energy. No significant charge transfer was found on the palladium surfaces. Our SO 2 chemisorption calculations indicate that the work functions of the metal surfaces examined should increase upon mono-S adsorption, increase to a lesser extent upon di S,O adsorption and may even decrease upon di O,O adsorption. Ab initio calculations provide evidence of the existence of SO 2 surface dimers. The binding energy predicted by the BOC-MP model for SO 3 in the bridging site agrees well with the ab initio result for SO 3 di-coordinated in the long bridge of the Ag(110) surface. The methods yield similar predictions for the case of SO on silver. Our modeling provides a coherent picture consistent with many aspects of the experimental literature. We present some model predictions, particularly the di O,O coordination mode for SO 2, that require verification experimentally.