The Effects of Cofeeding Chlorinated Hydrocarbons in the Direct Epoxidation of Propylene by Molecular Oxygen

Propylene oxide (PO) is a valuable intermediate used in the production of a large variety of valuable consumer products, such as polyurethane foams, polymers, propylene glycol, cosmetics, food emulsifiers and as fumigants and insecticides. 2] Over 8 million tons of PO are produced annually from propylene. The technology, economics, and environmental impacts of current as well as alternate propylene epoxidation processes have recently been reviewed. Recently, we reported the discovery of a new class of silicasupported RuO2–CuOx–NaCl catalysts for the direct epoxidation of propylene by using molecular oxygen under atmospheric pressure. This trimetallic catalyst, at its optimal composition of Ru/Cu/Na=4:2:1(metal weight ratio, or about 3:4:4 atom ratio) at 12.5 wt% total metal loading, exhibited PO selectivities in the range 40–50% at propylene conversions of 10–20% at 240–270 8C and 1 atm (1 atm=1.0133 10 Pa), and it maintained this activity for up to 4–8 h. However, we subsequently observed a slow, but steady decrease in PO selectivity in experiments over longer time periods. This degradation in performance is not acceptable from a practical standpoint if the RuO2–CuOx–NaCl/SiO2 system is to be exploited commercially. Here, we report that the introduction of chlorinated hydrocarbon (CHC) additives to the C3H6/O2 feed in the range 1–100 parts per million (ppm by volume) ameliorates the performance degradation problem and enables the steady production of PO, albeit at a decrease in propylene conversion. The beneficial effect of chlorinated hydrocarbons on propylene epoxidation in the RuO2–CuOx–NaCl/SiO2 system appears to be different than the promotional effects observed in Ag catalyzed ethylene oxide (EO) production, although some similarities also exist. The promotion of EO by chlorine on silver has been studied in considerable detail in the past and has been attributed to a combination of geometric/ensemble and electronic effects. For example, surface adsorbed Cl atoms (Cls) have been suggested to decrease the number of sites for oxygen adsorption, thereby decreasing catalyst activity. Additionally, by site blocking, Cls has been proposed to reduce the number of neighboring active sites needed for the dissociative adsorption of O2. This leads to increased molecular O2 adsorption (i.e. , Os O) as opposed to Os (surface oxygen), thus increasing EO selectivity at the expense of a decrease in activity. In the electronic models, an increase in EO selectivity by chlorine has been attributed to its higher electronegativity. This has been suggested to result in the weakening of the Ag O bond, which leads to increased EO selectivity. However, the same mechanism also results in an increased activation energy for the dissociative chemisorption of O2, thereby decreasing the overall activity. In another proposal, EO promotion by chlorine has been attributed to the formation of both surface and subsurface Cl that collectively alter the energetics of the oxymetallacycle (OMC) mechanism towards EO synthesis. Recent computational studies on Ag2O (001) also point to surface adsorption of Cl to vacant sites next to Os (thereby favoring O2 adsorption as Os O) as the primary reason for EO promotion. These investigators also predicted a direct EO synthesis route from Ag2O when oxygen vacancies are not present. This pathway was shown to switch to the less selective OMC mechanism when oxygen vacancies are introduced at low oxygen coverages, which promote dissociative O2 chemisorption. As evident from the above summary, in spite of decades of research, a consensus has not yet been reached regarding the mechanism(s) of chlorine promotion of EO over silver catalysts. Considering the more complex physical and chemical nature of the RuO2–CuOx–NaCl/SiO2 system and the presence of alyllic hydrogen atoms in propylene, the development of molecular level insights to account for the beneficial effects of chlorine on propylene epoxidation remains a formidable intellectual challenge. We hope that the experimental results provided in this communication will promote thinking and inspire others to undertake both computational and experimental surface science studies directed towards clarifying the mechanism(s) of propylene epoxidation and the effects of chlorine on PO synthesis. The direct epoxidation of propylene is described by Equation (1). Yet a variety of byproducts, such as acetone (AT), acrolein (AC), acetaldehyde (AD), and propanal (PaL) could form together with the total oxidation products of CO and CO2. [3]