Slurry-phase CO2 hydrogenation to hydrocarbons over a precipitated Fe-Cu-Al/K catalyst: Investigation of reaction conditions

The hydrogenation of CO₂ to hydrocarbons over a precipitated Fe-Cu-Al/K catalyst was studied in a slurry reactor for the first time. Reducibility of the catalyst and effect of reaction variables (temperature, pressure and H₂/CO₂ ratio of the feed gas) on the catalytic reaction performance were investigated. The reaction results indicated that the Fe-Cu-Al/K catalyst showed a good CO₂ hydrogenation performance at a relatively low temperature (533 K). With the increase of reaction temperature CO₂ conversion and olefin to paraffin (O/P) ratio in C₂-C₄ hydrocarbons as well as the selectivity to C₂-C₄ fraction increased, while CO and CH₄ selectivity showed a reverse trend. With the increase in reaction pressure, CO₂ conversion and the selectivity to hydrocarbons increased, while the CO selectivity and O/P ratio of C₂-C₄ hydrocarbons decreased. The investigation of H2/CO₂ ratio revealed that CO₂ conversion and CH₄ selectivity increased while CO selectivity and O/P ratio of C₂-C₄ decreased with increasing H₂/CO₂ ratio.     

Pathways for CO2 Formation and Conversion During Fischer–Tropsch Synthesis on Iron-Based Catalysts

CO₂ reaction and formation pathways during Fischer–Tropsch synthesis (FTS) on a co-precipitated Fe–Zn catalyst promoted with Cu and K were studied using a kinetic analysis of reversible reactions and with the addition of ¹³C-labeled and unlabeled CO₂ to synthesis gas. Primary pathways for the removal of adsorbed oxygen formed in CO dissociation steps include reactions with adsorbed hydrogen to form H₂O and with adsorbed CO to form CO₂. The H₂O selectivity for these pathways is much higher than that predicted from WGS reaction equilibrium; therefore readsorption of H₂O followed by its subsequent reaction with CO-derived intermediates leads to the net formation of CO₂ with increasing reactor residence time. The forward rate of CO₂ formation increases with increasing residence time as H₂O concentration increases, but the net CO₂ formation rate decreases because of the gradual approach to WGS reaction equilibrium. CO₂ addition to synthesis gas does not influence CO₂ forward rates, but increases the rate of their reverse steps in the manner predicted by kinetic analyses of reversible reactions using non-equilibrium thermodynamic treatments. Thus the addition of CO₂ could lead to the minimization of CO₂ formation during FTS and to the preferential removal of oxygen as H₂O. This, in turn, leads to lower average H_2/CO ratios throughout the catalyst bed and to higher olefin content and C₅₊ selectivity among reaction products. The addition of ¹³CO₂ to H₂/¹²CO reactants did not lead to significant isotopic enrichment in hydrocarbon products, indicating that CO₂ is much less reactive than CO in chain initiation and growth. We find no evidence of competitive reactions of CO₂ to form hydrocarbons during reactions of H₂/CO/CO₂ mixtures, except via gas phase and adsorbed CO intermediates, which become kinetically indistinguishable from CO₂ as the chemical interconversion of CO and CO₂ becomes rapid at WGS reaction equilibrium.     

Preferential CO oxidation over Pt–SnO2/Al2O3 in hydrogen rich streams containing CO2 and H2O (CO removal from H2 with PROX)

The effects of reaction gases including CO₂ and H₂O and temperature on the selective low-temperature oxidation of CO were studied in hydrogen rich streams using a flow micro-reactor packed with a Pt–SnO₂/Al₂O₃ sol–gel catalyst that was initially designed and optimized for operation in the absence of CO₂ and H₂O. 100% CO conversion was achieved over the 1 wt% Pt–3 wt% SnO₂/Al₂O₃ catalyst at 110 °C using a feed composition of 1.0% CO, 1.5% O₂, 25% CO₂, 10% H₂O, 58% H₂ and He as balance at a space velocity of 24,000 cm³/(g h). CO₂ in the feed was found to decrease CO conversion significantly while the presence of H₂O in the feed increased CO conversion, balancing the effect of CO₂.     

Direct synthesis of dimethyl ether from carbon-monoxide-rich synthesis gas: Influence of dehydration catalysts and operating conditions

Various dehydration catalysts were studied in the synthesis of dimethyl ether (DME) directly from carbon-monoxide-rich synthesis gas under a series of different reaction conditions. The investigated catalyst systems consisted of combinations of a methanol catalyst (CuO/ZnO system) with catalysts for methanol dehydration based on γ-Al₂O₃ or zeolites and γ-Al₂O₃ was identified as the most favorable dehydration catalyst. Various reaction parameters such as temperature, H₂/CO ratio and space velocity were studied. The impact of water on Cu/ZnO/Al₂O₃-γ-Al₂O₃ catalysts was investigated and no deactivation could be observed at water contents below 10% during running times of several hours. A running time of several days and a water content of 10% led to a significant increase of CO conversion but the water gas shift reaction became dominating and CO₂ was the main product. After termination of water feeding significant deactivation of the catalyst system was observed but the system returned to high DME selectivity. Catalyst stability and the influence of CO₂ in the gas feed were studied in experiments lasting for about three weeks. The presence of 8% of CO₂ caused an approximately 10% lower CO conversion and an about 5% lower DME selectivity compared to the reaction system without CO₂.     

ALTERNATIVE GENERATION OF H2, CO AND H2, CO2 MIXTURES FROM STEAM-CARBON DIOXIDE REFORMING OF METHANE AND THE WATER GAS SHIFT WITH PERMEABLE (MEMBRANE) REACTORS

A new process is proposed which converts CO₂ and CH₄ containing gas streams to synthesis gas, a mixture of CO and H₂ via the catalytic reaction scheme of steam-carbon dioxide reforming of methane or the respective one of only carbon dioxide reforming of methane, in permeable (membrane) reactors. The membrane reformer (permreactor) can be made by reactive or inert materials such as metal alloys, microporous ceramics, glasses and composites which all are hydrogen permselective. The rejected CO reacts with steam and converted catalytically to CO₂ and H₂ via the water gas shift in a consecutive permreactor made by similar to the reformer materials and alternatively by high glass transition temperature polymers. Both permreactors can recover H₂ in permeate by using metal membranes, and H₂ rich mixtures by using ceramic, glass and composite type permselective membranes. H₂ and CO₂ can be recovered simultaneously in water gas shift step after steam condensation by using organic polymer membranes. Product yields are increased through permreactor equilibrium shift and reaction separation process integration.

CO and H₂ can be combined in first step to be used for chemical synthesis or as fuel in power generation cycles. Mixtures of CO₂ and H₂ in second step can be used for synthesis as well (e.g., alternative methanol synthesis) and as direct feed in molten carbonate fuel cells. Pure H₂ from the above processes can be used also for synthesis or as fuel in power systems and fuel cells. The overall process can be considered environmentally benign because it offers an in-situ abatement of the greenhouse CO₂ and CH₄ gases and related hydrocarbon-CO₂ feedstocks (e.g., coal, landfill, natural, flue gases), through chemical reactions, to the upgraded calorific value synthesis gas and H₂, H₂ mixture products.     

Mechanistic Study of Selective Oxidation of Dimethyl Ether to Formaldehyde over Alumina-supported Molybdenum Oxide Catalyst

XPS and IR spectroscopies were used to investigate the surface intermediates of dimethyl ether (DME) oxidation to formaldehyde over MoOx/Al₂O₃ catalyst. The reaction performances were tested by employing three typical reaction conditions, depending on the O₂/DME ratio and the reaction temperature. When there was sufficient oxygen present in the reaction media, a terminal or bridged CH₃O* species formed by DME dissociation was highly active and rapidly reacted with lattice oxygen to produce formaldehyde, leading to higher selectivity of HCHO. When oxygen was consumed completely or only DME was present in the reaction media, CH₃O species bonded to more than two Mo atoms (μ-OCH₃) and CH_x (x=1–3) species attached to the Mo atoms were observed and the relative ratio of (μ-OCH₃) /Mo–CH_x was significantly dependent on the reduction degree of MoO_x domains. The (μCH₃O) species was related to the formation of CH₃OH or CO_x, and the Mo–CH_x species led to the formation of CH₄.     

Effect of water pretreatment on CO<Subscript>2</Subscript> capture using a potassium-based solid sorbent in a bubbling fluidized bed reactor

A bubbling fluidized bed reactor was used to study CO₂ capture from flue gas by using a potassium-based solid sorbent, sorbKX35 which was manufactured by the Korea Electric Power Research Institute. A dry sorbent, sorbKX35, consists of K₂CO₃ for absorption and supporters for mechanical strength. To increase initial CO₂ removal, some amount of H₂O was absorbed in the sorbent before injecting simulated flue gas. It was possible to achieve 100% CO₂ removal for more than 10 minutes at 60°C and a residence time of 2 s with H₂O pretreatment. When H₂O pretreatment time was long enough to convert K₂CO₃ of sorbKX35 into K₂CO₃ · 1.5H₂O, CO₂ removal was excellent. The results obtained in this study can be used as basic data for designing and operating a large scale CO₂ capture process with two fluidized bed reactors. This work was presented at the 6 ^th Korea-China Workshop on Clean Energy Technology held at Busan, Korea, July 4–7, 2006.     

15N solid state NMR study of the reactions of propane or propene, 15NO and oxygen on Na‐, H‐, and CuZSM‐5

Ni- and Co-based catalysts derived from NiAl- and CoAl-layered double hydroxides were tested in four kinds of reactions of methanol, namely decomposition of methanol (DCM), partial oxidation of methanol (POM), steam reforming of methanol (SRM), and oxidative steam reforming of methanol (OSRM), for the purpose of H₂ production for fuel cells. H₂, CO and/or CO₂ were the predominant products with minor amounts of dimethyl ether (DME) and CH₄ depending on the reaction temperature. Among the four kinds of reactions tested, the OSRM reaction was found to be more effective in terms of MeOH conversion and H₂ selectivity over these catalysts. Higher selectivity of H₂ and CO₂ with only traces of CO could be obtained at about 100% methanol conversion around 300 °C in the OSRM reaction over the catalyst derived from CoAl-LDH. Substitution of a part of Al by Sn in the NiAl- and CoAl-LDH systems was found to be inhibiting the methanol conversion. On the other hand, the selectivities to DME and CH₄ were declined with a consequent increase in the selectivity to H₂. In addition, considerable amount of formaldehyde was also noticed, especially over the catalyst derived from CoAlSn-LDH at lower reaction temperatures. The observed difference in the catalytic performance upon Sn incorporation was attributed to an improved redox capability of the Ni- and Co-based oxide catalysts, as determined by temperature-programmed reduction (TPR) experiments. This revised version was published online in July 2006 with corrections to the Cover Date.     

Effective parameters for DME steam reforming catalysts for the formation of H2 and CO

Recently, DME has received attention as a clean fuel and is now considered an alternative fuel for diesel engines. DME diesels need de-NOx catalysts such as LNT (Lean NOx Trap) and SCR (Selective Catalytic Reduction) systems. DME is an attractive source of hydrogen because it can be stored easily and is a good transportation fuel. Hydrogen and CO enriched gas as a reductant was used with the LNT catalyst in order to reduce NOx emissions. The steam reforming catalyst of DME was used to formation of hydrogen. It has been reported that Cu-based catalysts have high selectivity and activity in the steam reforming of DME. This research used 600 cPsi cordierite as a catalyst, which was coated with copper. The catalysts were made via a sol–gel and impregnation methods. The formation of H₂ and CO under the prepared catalysts was tested by a model gas. Experimental parameters were considered; the space velocity (SV) and concentrations of H₂O, O₂, and CO₂ were evaluated. The Cu 30%/γ-Al₂O₃ catalyst from the sol–gel method exhibited high and stable activity in the production of hydrogen from the steam reforming of DME. Both DME conversion and the selectivity of hydrogen were affected by SV and the concentrations of H₂O, O₂, and CO₂.     

Oxy-CO2 reforming of methane to syngas over CoOx/MgO/SA-5205 catalyst

The oxy-CO₂ methane reforming reaction (OCRM) has been investigated over CoO_x supported on a MgO precoated highly macroporous silica–alumina catalyst carrier (SA-5205) at different reaction temperatures (700–900 °C), O₂/CH₄ ratios (0.3–0.45) and space velocites (20,000–100,000 cc/g/h). The reaction temperature had a profound influence on the OCRM performance over the CoO/MgO/SA-5205 catalyst; the methane conversion, CO₂ conversion and H₂ selectivity increased while the H₂/CO ratio decreased markedly with increasing reaction temperature. While the O₂/CH₄ ratio did not strongly affect the CH₄ and CO₂ conversion and H₂ selectivity, it had an intense influence on the H₂/CO ratio. The CH₄ and CO₂ conversion and the H₂ selectivity decreased while the H₂/CO increased with increasing space velocity. The O₂/CH₄ ratio and the reaction temperature could be used to manipulate the heat of the reaction for the OCRM process. Depending on the O₂/CH₄ ratio and temperature the OCRM process could be operated in a mildly exothermic, thermal neutral or mildly endothermic mode. The OCRM reaction became almost thermoneutral at an OCRM reaction temperature of 850 °C, O₂/CH₄ ratio of 0.45 and space velocity of 46,000 cc/g/h. The CH₄ conversion and H₂ selectivity over the CoO/MgO/SA-5205 catalyst corresponding to thermoneutral conditions were excellent: 95% and 97%, respectively with a H₂/CO ratio of 1.8.     

Effect of co-feeding carbon dioxide on Fischer–Tropsch synthesis over an iron–manganese catalyst in a spinning basket reactor

The effect of co-feeding CO₂ on the catalytic properties of an Fe–Mn catalyst during Fischer–Tropsch synthesis (FTS) was investigated in a spinning basket reactor by varying added CO₂ partial pressure in the feed gas. It was found that co-feeding CO₂ to syngas did not decrease the activity of the catalyst, on the contrary, a dramatic increase of the activity and an increase of methane selectivity were observed over the catalyst after removal of CO₂ from the feed gas. The addition of CO₂ led to an increase in olefin/paraffin ratios of low carbon hydrocarbons and a slight decrease in C₁₉⁺ selectivity. It also slightly decreased CO₂ formation rate on the catalyst by increasing the rate of reverse step of the water–gas shift (WGS) reaction and pushing the reaction towards equilibrium, and did not remarkably influence the hydrocarbon formation rate. However, the co-feeding CO₂ can significantly increase the water formation rate and the overall oxygenate formation rate under these reaction conditions.     

介质阻挡放电中气体成分对NOx脱除的影响

利用介质阻挡放电(DBD)产生低温等离子体进行烟气的脱硝实验,研究了在乙烯存在的条件下,温度和其他烟气成分对NO_x脱除率的影响。结果表明:随着温度的升高,NO脱除速率增快;模拟烟气中加入CO₂,在能量密度较低时,CO₂作为电负性分子会降低自由基的生成,导致NO的脱除率降低,随着能量密度的升高,CO₂对NO脱除的影响减小;模拟烟气中加入水后可以产生更多的OH、HO₂等自由基,促进NO的氧化;SO₂的加入会与自由基O反应,使初始反应中O与C₂H₄的反应速率减弱,从而影响了NO的氧化速率,但O₃、HO₂等强氧化自由基会优先与NO反应,因此SO₂的加入不会影响NO最终的脱除率。     

Methanol synthesis from carbon dioxide at atmospheric pressure over Cu/ZnO catalyst. Role of methoxide species formed on ZnO support

Methanol synthesis from CO₂ and H₂ was carried out over a Cu/ZnO catalyst (Cu/Zn = 3/7) at atmospheric pressure, and the surface species formed were analyzed by diffuse reflectance FT-IR spectroscopy and temperature programmed desorption method. Two types of formate species and zinc methoxide were formed in the course of the reaction. Zinc methoxide was readily hydrolyzed to methanol. H₂O formed through the reverse water gas shift reaction was suggested to be involved in the hydrolysis of zinc methoxide.     

Regeneration behaviors of Fe/Si‐2 and Fe–Mn/Si‐2 catalysts for C2H6 dehydrogenation with CO2 to C2H4

The catalytic performance of Fe/Si‐2 and Fe–Mn/Si‐2 catalysts for conversion of C₂H₆ with CO₂ to C₂H₄ was examined in a continuous‐flow and fixed‐bed reactor. The results show that the Fe–Mn/Si‐2 catalyst exhibits much better reaction activity and selectivity to C₂H₄ than those of the Fe/Si‐2 catalyst. Furthermore, the coking–decoking behaviors of these catalysts were studied through TG. The catalytic performances of the catalysts after regeneration for conversion of C₂H₆ or dilute C₂H₆ in FCC off‐gas with CO₂ to C₂H₄ were also examined. The results show that both activity and selectivity of the Fe–Mn/Si‐2 catalyst after regeneration reached the same level as those of the fresh catalyst, whereas it is difficult for the Fe/Si‐2 catalyst to refresh its reaction behavior after regeneration. This revised version was published online in July 2006 with corrections to the Cover Date.     

H2S–CO2 Separation Using Room Temperature Ionic Liquid [BMIM][Br]

Solubility and selective absorption of hydrogen sulfide (H₂S) over carbon dioxide (CO₂) in a room temperature ionic liquid, 1-butyl-3-methylimidazolium bromide ([BMIM][Br]) has been evaluated under ambient temperature and pressure. [BMIM][Br] demonstrated its potential as a solvent for selective removal of H₂S from CO₂/H₂S mixture. Our investigation indicated that H₂S solubility in [BMIM][Br] is comparable to or better than that in commercially available MDEA-based solvents. Meanwhile, CO₂ solubility in [BMIM][Br] is lower than that in the same amine resulting in H₂S/CO₂ absorption selectivity of within 3.5 to 3.75. The solubility behavior is relatively maintained after 4 times absorption-desorption cycles. A computational molecular study suggested that intramolecular hydrogen bonding interaction between anion Br and hydrogen atom of H₂S could stabilize the complex and resulted lower complexation energy than CO₂ interaction with [BMIM][Br]. Based on the experiment results, a separation process employing [BMIM][Br] is proposed to control the CO₂/H₂S ratio existing in a natural gas feed.     

Carbonation of fly ash in oxy-fuel CFB combustion

Oxy-fuel combustion of fossil fuel is one of the most promising methods to produce a stream of concentrated CO₂ ready for sequestration. Oxy-fuel FBC (fluidized bed combustion) can use limestone as a sorbent for in situ capture of sulphur dioxide. Limestone will not calcine to CaO under typical oxy-fuel circulating FBC (CFBC) operating temperatures because of the high CO₂ partial pressures. However, for some fuels, such as anthracites and petroleum cokes, the typical combustion temperature is above 900 °C. At CO₂ concentrations of 80-85% (typical of oxy-fuel CFBC conditions with flue gas recycle) limestone still calcines, but when the ash cools to the calcination temperature, carbonation of fly ash deposited on cool surfaces may occur. This phenomenon has the potential to cause fouling of the heat transfer surfaces in the back end of the boiler, and to create serious operational difficulties. In this study, fly ash generated in a utility CFBC boiler was carbonated in a thermogravimetric analyzer (TGA) under conditions expected in an oxy-fuel CFBC. The temperature range investigated was from 250 to 800 °C with CO₂ concentration set at 80% and H₂O concentrations at 0%, 8% and 15%, and the rate and the extent of the carbonation reaction were determined. Both temperature and H₂O concentrations played important roles in determining the reaction rate and extent of carbonation. The results also showed that, in different temperature ranges, the carbonation of fly ash displayed different characteristics: in the range 400 °C < T ? 800 °C, the higher the temperature the higher the CaO-to-carbonate conversion ratio. The presence of H₂O in the gas phase always resulted in higher CaO conversion ratio than that obtainable without H₂O. For T ? 400 °C, no fly ash carbonation occurred without the presence of H₂O in the gas phase. However, on water vapour addition, carbonation was observed, even at 250 °C. For T ? 300 °C, small amounts of Ca(OH)₂ were found in the final product alongside CaCO₃. Here, the carbonation mechanism is discussed and the apparent activation energy for the overall reaction determined.     

Catalytic syngas production from greenhouse gasses: Performance comparison of Ru-Al2O3 and Rh-CeO2 catalysts

In this work, 3% Ru-Al₂O₃ and 2% Rh-CeO₂ catalysts were synthesized and tested for CH₄-CO₂ reforming activity using either CO₂-rich or CO₂-lean model biogas feed. Low carbon deposition was observed on both catalysts, which negligibly influenced catalytic activity. Catalyst deactivation during temperature programmed reaction was observed only with Ru-Al₂O₃, which was caused by metallic cluster sintering. Both catalysts exhibited good stability during the 70 h exposure to undiluted equimolar CH₄/CO₂ gas stream at 750 °C. By varying residence time in the reactor during CH₄-CO₂ reforming, very similar quantities of H₂ were consumed for water formation. Reverse water-gas shift (RWGS) reaction occurred to a very similar extent either with low or high WHSV values over both catalysts, revealing that product gas mixture contained near RWGS equilibrium composition, confirming the dominance of WGS reaction and showing that shortening the contact time would actually decrease the H₂/CO ratio in the syngas produced by CH₄-CO₂ reforming, as long as RWGS is quasi equilibrated. H₂/CO molar ratio in the produced syngas can be increased either by operating at higher temperatures, or by using a feed stream with CH₄/CO₂ ratio well above 1.     

Epoxidation of Styrene by Anhydrous H2O2 over TS-1 and γ-Al2O3 Catalysts: Effect of Reaction Water, Poisoning of Acid Sites and Presence of Base in the Reaction Mixture

The styrene conversion and product (viz. styrene oxide, phenyl acetaldehyde, benzaldehyde) selectivity in the liquid-phase epoxidation of styrene by H₂O₂ (H₂O₂/styrene = 2) over TS-1 (Si/Ti = 80) and -Al₂O₃ are strongly influenced by the presence of water and/or base (viz. urea and pyridine) in the reaction mixture. The TS-1 showed high styrene conversion activity but no epoxide selectivity in the absence of any base. When anhydrous H₂O₂ (24% H₂O₂ in ethyl acetate), with the continuous removal of the reaction water (using the DeanStark trap), was used instead of 50% aqueous H₂O₂, both the conversion and epoxide yield are increased drastically for the -Al₂O₃, whereas for the TS-1, the increase in the conversion was quite small and there was also no improvement in the epoxide selectivity and/or yield. However, when urea or pyridine was added in the reaction mixture, the epoxide selectivity for both the catalysts was increased depending on the concentration of the base added; the increase in the selectivity was very large for the TS-1 but small for the -Al₂O₃. Poisoning of the acid sites of the -Al₂O₃ by the chemisorbed ammonia or pyridine (at 100 °C) caused a small decrease in the conversion, but it also caused a large decrease in the epoxide selectivity. However, the pyridine poisoning of the TS-1 caused a little beneficial effect, a small increase in the epoxide selectivity. The ammonia poisoning of the TS-1, however, resulted in a small decrease in the conversion with no improvement in the epoxide selectivity. As compared to the TS-1, the -Al₂O₃ catalyst showed a much better performance in the epoxidation by anhydrous H₂O₂ with the continuous removal of the reaction water. However, the reaction water, if not removed continuously, is detrimental to the -Al₂O₃, causing a large decrease in the catalytic activity and selectivity for styrene oxide but an increase in the selectivity for benzaldehyde.     

Experimental study of improved two step synthesis for DME production

Dimethyl ether (DME) has received growing attention due to its potential use as a multi-purpose fuel. A new technical route of improved two step synthesis is proposed for DME production, which is composed of methanol synthesis and methanol dehydration in a fixed-bed reactor. The influences of the operating conditions including reaction pressure, temperature, H₂/CO mole ratio in the syngas and space velocity on CO conversion, selectivity and yield of DME are investigated. CO conversion and DME yield both increase monotonically with the pressure increase. The optimal reaction temperatures for the synthesis and dehydration of methanol are different. CO conversion increases at first and keeps constant when the H₂/CO mole ratio is above 2. DME yield increases obviously and then decreases gradually with the space velocity increase. The optimal conditions are obtained to maximize the CO conversion and DME selectivity. The reaction temperatures of the top and bottom stage are in the range of 270-280 °C and 235-245 °C, respectively. The optimal ratio of H₂/CO is above 2, and the space velocity is in the range of 1000-1300 h^− 1.     

The Effects of CO2 and H2O on the NO x Destruction Performance of a Model NO x Storage/Reduction Catalyst

The effects of CO₂ and H₂O on the NO _x storage and reduction characteristics of a Pt/Ba/Al₂O₃ catalyst were investigated. The presence of CO₂ and H₂O, individually or together, affect the performance and therefore the chemistry that occurs at the catalyst surface. The effects of CO₂ were observed in both the trapping and reduction phases of the experiments, whereas the effect of H₂O seems limited to the trapping phase. The data also indicate that multiple types of sorption sites (or mechanisms for sorption) exist on the catalyst. One mechanism is characterized by a rapid and complete uptake of NO _x . A second mechanism is characterized by a slower rate of NO _x uptake, but this mechanism is active for a longer time period. As the temperature is increased, the effect of H₂O decreases compared to that of CO₂. At the highest temperatures examined, the elimination of H₂O when CO₂ is present did not affect the performance.