J. Phys. Chem. B 2006, 110, 21857-21864
21857
Sulfur the Archetypal Catalyst Poison? The Sulfur-Induced Promotion of the Bonding of Unsaturated Hydrocarbons on Cu(111) G. B. D. Rousseau, N. Bovet, and M. Kadodwala* Department of Chemistry, Joseph Black Building, UniVersity of Glasgow, Glasgow G12 8QQ, U.K. ReceiVed: June 21, 2006; In Final Form: August 23, 2006
We have shown using a combination of temperature-programmed desorption and UV photoelectron spectroscopy that the presence of preadsorbed atomic sulfur promotes the bonding of cyclic unsaturated hydrocarbons (benzene and cyclohexene) to Cu(111). This promoting behavior of sulfur can be rationalized in terms of the ability of adsorbed sulfur to influence the balance between charge donation from the adsorbate to metal, and back-donation from the metal to adsorbate. The effects of sulfur on Cu(111) are dramatically different from those observed in previous studies on Pt(111), which found that it caused a downward shift in the desorption temperature of adsorbed benzene, through purely steric effects.
Introduction Sulfur is perceived as the archetypal poison of transition metal based heterogeneous catalysts. Indeed, considerable research effort, both academic and industrial, has been expended on the desulfurization of petroleum feedstocks to inhibit catalyst deactivation. However, the role of sulfur, and other electronegative elements, in noble metal catalysts is less clear-cut. Sulfur does poison the Cu-catalyzed water-gas shift reaction, but recently Hutchings and co-workers demonstrated promotional effects of sulfur on the selective hydrogenation reactions on supported noble metal catalysts. The modification of Cu/Al2O3 by very low levels of sulfur significantly enhanced the selectivity toward and the rate of formation of but-2-en-1-ol from the hydrogenation of but-2-enal. Analysis of their results showed that sulfur acted as a promoter for this selective hydrogenation reaction, rather than a poison as would have been expected.1-3 Similar promotional effects by sulfur were also observed with Au/ZnO for the same reaction.4-6 The promotional effects observed in these catalytic studies were assigned to electronic effects of preadsorbed sulfur. The aim of this work is to rationalize the promotional effects that sulfur has on Cu catalysts and establish if there are indeed electronic effects at play. To achieve this objective, we studied the influence of sulfur preadsorption on the bonding of a series of cyclic hydrocarbons (benzene, cyclohexene, and cyclohexane) on Cu(111). The presence of preadsorbed sulfur caused a stabilization in the bonding of unsaturated molecules, which could be rationalized using an electrostatic based model. No stabilization in bonding was observed for cyclohexane, which is also consistent with the electrostatic model. However, the coverage within the chemisorbed cyclohexane layer was increased by sulfur pre-coverage through the population of a new destabilized, relative to the clean surface, state. Experimental Section Experiments were performed in two separate UHV systems, temperature programmed desorption (TPD) data were collected in a system that was described in detail previously,8 and * Corresponding author. E-mail:
[email protected].
desorption profiles were collected using a heating rate of 0.5 K s-1. UV photoelectron spectroscopy (UPS) and X-ray photoelectron spectroscopy (XPS) measurements were performed in a second system, which will be briefly described. The system was equipped with the usual sample preparation facilities and low-energy electron diffraction (LEED) optics for confirming crystal quality. In addition, it has a discharge lamp (VG Ltd) that provides He(I) radiation and a concentric hemispherical analyzer (CHA) (CLAM 2 VG Ltd). All UPS spectra were collected in normal emission and with radiation incident at 47° to the surface normal. In both UHV systems, the Cu(111) surface was cleaned by cycles of Ar+ bombardment (1 keV, 40 min, ca. 16 µA) followed by annealing to 900 K. Surface cleanliness was monitored by electron beam AES (collected using an RFA) in the TPD system, while XPS was used in the second system. In both cases, LEED was used to monitor surface quality and adsorption was carried out at crystal temperatures of ca. 110 K. Results (A) Sulfur Overlayers. Experiments where performed on clean and θS ) 0.14, 0.33, and 0.43 ML Cu(111) surfaces. We have published a separate report on the structure of these sulfur surfaces, which includes UPS and work function data.8 At room temperature, the 0.14 and 0.33 ML surfaces are disordered, while the 0.43 ML displays a well-ordered (x7 × x7)R19°-S structure. The coverages of the disordered structures have been calibrated (using AES and XPS) against the ordered 0.43 ML surface. Upon cooling to
