Abstract

The bond order conservation-Morse potential (BOC-MP) method proved to be versatile and useful in analyzing chemisorption and reactivity on transition metal surfaces [1–3]. Most recently, we have applied the BOC-MP method to analyze surface chemistry of sulfur oxides on the Cu and Ni group metals [4,5]. We have calculated the reaction energetics (heats of adsorption, reaction enthalpies and intrinsic activation barriers) of the decomposition and oxidation of sulfur dioxide on fcc(111) surfaces of Cu, Ag, Au, Ni, Pd and Pt. The accuracy of the BOC-MP heats of adsorption has been verified by high quality (RECP-HF-MP2) ab initio calculations of the heats of SO x adsorption ( x = 1, 2, 3, 4) on Ag and Pd surfaces, which also clarified SO x coordination modes. The BOC-MP analysis was made for low coverages without considering diffusional effects. The major model projections are: (1) The dissociation SO 2 → SO + O is unfavorable on all the metals considered; contrary to the common assumption, SO is unlikely to be a product of SO 2 dissociation. Also, on the Pt, Pd, Ni and Cu surfaces, SO is unstable. (2) The SO 2 dissociation is possible as SO 2 → S + O + O, showing distinct periodic trends. It is feasible on Cu and particularly on Ni. (3) In the presence of carbon monoxide, the dissociation SO 2 + CO → S + O + CO 2 may occur on all the metals examined. (4) The oxidation of SO 2 to SO 3 may be achieved with various-oxygen sources (e.g. O, O 2, H 2O 2 and NO) on Au, Ag, Pt and Pd surfaces. (5) Although adsorbed SO 3 may be readily obtained, it may be impossible to desorb SO 3 intact at low coverages because SO 3 will decompose to SO 2 + O before desorption. (6) At elevated temperatures, the most stable of adsorbed sulfur oxides appears to be tetraoxide. As shown by Sellers et al. [6], its state is ionic and close to sulfate. The relevant experimental data are discussed. Most of the model projections are in agreement with experiment. However, we suggest the reinterpretation of some of the reported experimental results and delineate model predictions to be verified in the future.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call