KINETIC STUDIES OF THE WATER - GAS SHIFT REACTION ON A SULFIDED COBALT - MOLYBDENA - ALUMINA CATALYST.

摘要

In this study, the applicability of low temperature oxygen chemisorption (LTOC) to measure the specific surface area of several rare-earth oxides (La, Ce, Pr, Nd, Tb) and the kinetics of the water-gas shift reaction over a sulfided cobalt-molybdena-alumina (AMOCAT 1A) catalyst are investigated.; The rare-earth oxides were studied after a "standard" (6 hours at 500{dollar}spcirc{dollar}C) pretreatment in hydrogen and also after re-oxidation and reduction. The pretreatments were carried out in-situ in an integrated microreactor system. The LTOC results indicate that oxygen is possibly adsorbed in the molecular form, O{dollar}sb2sp-{dollar}, as observed by others after heat treatment of these oxides in vacuum. Lanthana and ceria were found to have ratios of total surface area to LTOC similar to those of chromia and molybdena respectively, after a comparable pretreatment. Furthermore, ceria is deduced to exist as a monolayer on the alumina support at loadings below 12%. An additional hour of reduction after the 6 hours of reduction shows a significant increase in LTOC on lanthana, neodymia and terbia which may be due to phase changes exhibited by these polymorphic oxides.; The kinetics of the water-gas shift reaction has been extensively studied on iron oxide (high temperature shift) and copper oxide (low temperature shift) based catalysts. This investigation establishes the kinetics over a sulfided cobalt-molybdena-alumina (AMOCAT 1A) catalyst in the "medium temperature shift" range, 250-300{dollar}spcirc{dollar}C. The catalyst was sulfided in-situ in an high pressure integrated Berty reactor system. Reaction rates were measured for different CO/H{dollar}sb2{dollar}O feed ratios in the range 0.3-3.0, with and without CO{dollar}sb2{dollar} in the feed. The reaction was carried out at several pressures in the range 5-27 atm. and GHSV's in the range 4800-2400 hr{dollar}sp1{dollar}. The observed reaction rates were fit to a simple power rate law of the form: -{dollar}{lcub}rm rsb{lcub}CO{rcub}{rcub}{dollar} = k{dollar}sb0 cdot{dollar}e{dollar}sp{lcub}-rm E/RT{rcub}cdot{dollar}P{dollar}sb{lcub}rm CO{rcub}sp{lcub}rm l{rcub}cdot{dollar}P{dollar}sb{lcub}rm Hsb2 O{rcub}sp{lcub}rm m{rcub}cdot{dollar}P{dollar}sb{lcub}rm COsb2{rcub}sp{lcub}rm n{rcub}cdot{dollar}(1-{dollar}beta{dollar}). Where {dollar}beta{dollar} = (P{dollar}sb{lcub}rm COsb2{rcub}{dollar}){dollar}cdot{dollar}(P{dollar}sb{lcub}rm Hsb2{rcub}{dollar})/K{dollar}cdot{dollar}(P{dollar}{lcub}rm CO{rcub}{dollar}){dollar}cdot{dollar}(P{dollar}sb{lcub}rm Hsb2 O{rcub}{dollar}); with the following results: k{dollar}sb0{dollar} = 219; E = 9.3 kcal/mol. {dollar}spcirc{dollar}K; l = 0.52; m = 0.21; and n = {dollar}-0.10.{dollar} The deviation of the observed reaction rate to the reaction rate predicted by this model is less than 10%. The reaction possibly proceeds by a redox mechanism involving Mo(V) species as postulated by Hou et.al. (118).