Investigation of Ethylene Oxide on Clean and Oxygen-Covered Ag(110) Surfaces

Abstract

Temperature-programmed desorption (TPD) and Density Functional Theory (DFT) were used to investigate the reactions of oxametallacycles derived from ethylene oxide on clean and oxygen-covered Ag (110) surfaces. Ethylene oxide ring-opens following adsorption at 250 K on both clean and O-covered Ag(110) to form a stable oxametallacycle. On the clean Ag(110) surface, the oxametallacycle reacts to reform the parent epoxide at 280 K during TPD, while the aldehyde isomer, acetaldehyde, is observed at higher oxametallacycle coverages. In the presence of coadsorbed oxygen atoms, a portion of the oxametallacycles dissociate to release ethylene. However, of those that react to form oxygen-containing products, the fraction forming ethylene oxide is similar to that on the clean surface. The acetaldehyde product of oxametallacycle reactions combusts via formation of acetate species; the acetates react to form CO 2 at temperatures as low as 360 K on the O-covered surface. No evidence was observed for other combustion channels. This work provides experimental evidence for the connection of oxametallacycles to combustion via acetaldehyde formation as well as to ringclosure to form ethylene oxide.

Comments on “Investigation of Ethylene Oxide on Clean and Oxygen-Covered Ag(110) Surfaces”

“…The desorption activation energy so obtained was 18 k cal mol -1 , a value which the authors stated was consistent with density functional theory (DFT) calculations and with previous results on Ag(111) [1,2].…”

“…The authors acknowledge that the m/z = 29 peaks shown in Fig. 1 are ''less consistent'' with a first order process than those reported for epoxybutene and styrene oxide derived oxametallacycles, but justify their first order kinetic analysis of them by citing the consistency of the energies derived from them with both DFT calculations and with the desorption energies derived from dosing EO on to Ag(111) at 250 K [1,2].The authors also suggest that the lack of consistency of the peak shapes with a first order process is due to their method of dosing the EO at 250 K on to the Ag (110) …”

“…They were reported in reference [1]. Lukaski and Barteau concluded in that paper that adsorption of EO at 250 K on to Ag(110) resulted in the opening of the epoxide ring, producing an adsorbed oxametallacycle [1]. They concluded there that heating the adsorbate from 250 K caused the oxametallacycle to re-cyclise to EO which desorbed at a peak maximum temperature of 290 K for the lowest coverage of the adsorbate (curve d).…”

“…They were reported in reference [1]. Lukaski and Barteau concluded in that paper that adsorption of EO at 250 K on to Ag(110) resulted in the opening of the epoxide ring, producing an adsorbed oxametallacycle [1].…”

“…The desorption process therefore involved: (1) first flashing the crystal to 230 K to remove the molecularly held EO, (2) cooling briefly and (3) initiating the temperature programming., The aim of the process was to isolate the 280 K peak. The authors, however, caution that ''the partial flash step may influence both the leading edge of the peak and the resultant peak position'', making ''inferences about peak shape or peak shifts dubious'' [1].…”