Detailed kinetic modeling of NOx adsorption and NO oxidation over Cu-ZSM-5

Detailed kinetic modeling was used in combination with flow reactor experiments to investigate the NO_x adsorption/desorption and NO oxidation over Cu-ZSM-5. NO oxidation is likely an important step for selective catalytic reduction (SCR) using urea and hydrocarbons, and thus was investigated separately. First the NO₂ adsorption on Brönstedt acid sites in H-ZSM-5 was modeled using an NO₂ temperature programmed desorption (TPD) experiment. These results, together with the results of the NO₂ TPD and NO oxidation experiments, were used in developing the model for Cu-ZSM-5. A substantial amount of NO₂ was adsorbed on the catalyst. However, the results from a corresponding NO TPD experiment showed that only very small amounts of NO were adsorbed on the catalyst and therefore this step was not included in the model. The model consists of reversible steps for NO₂ and O₂ adsorption, O₂ dissociation, NO oxidation and two steps for nitrate formation. The first nitrate formation step was disproportionation of NO₂ to form NO and nitrates. This step enabled us to describe the NO production during NO₂ adsorption. Further, in the reverse step the NO reacts with the nitrates and decreased their stability. Without this step the nitrates blocked the surface resulting in to low NO oxidation activity. However, we observe that nitrates can be decomposed also without the presence of NO and in the second reversible step were the nitrates decomposed to form NO₂ and oxygen on the copper. These steps enabled us to describe both the TPD and activity measurement results. NO oxidation was observed even at room temperature. Interestingly, the NO₂ decreased when increasing the temperature up to 100 °C and then increased as the temperature increased further. We suggest that this low-temperature NO oxidation occurs with species loosely bound on the surface and that is included in the detailed mechanism. An additional NO₂ TPD at 30 °C was also modeled to describe the loosely bound NO₂ on the surface. The detailed model correctly describes NO₂ storage, NO oxidation and low-temperature NO oxidation.     

A Study on the Properties and Mechanisms for NO x Storage Over Pt/BaAl2O4-Al2O3 Catalyst

The NO_x storage catalyst Pt/BaAl₂O₄-Al₂O₃ was prepared by a coprecipitation--impregnation method. For fresh sample, the barium mainly exists as the BaAl₂O₄ phase except for some BaCO₃ phase. The BaAl₂O₄ phase is the primary NO _x storage phase of the sample. EXAFS and TPD were used for investigating the mechanism of NO _x storage. It is found that two kinds of Pt sites are likely to operate. Site 1 is responsible for NO chemisorption and site 2 for oxidizing NO to nitrates and nitrites. When NO adsorbs on the sample below 200 °C, it mainly chemisorbs in the form of molecular states. Such adsorption results in an increase of the coordination magnitude of Pt-O, and a decrease of that of Pt-Pt and Pt-Cl. The coordination distance of Pt-Pt, Pt-Cl and Pt-O also increases. When the adsorption occurs above 200 °C, NO can be easily oxidized by O₂, and stored as nitrites or nitrates at the basic BaAl₂O₄. Site 2 is regenerated quickly. A high adsorption temperature is favorable for nitrate formation.     

A novel continuous reactor for catalytic reduction of NOx—fixed bed simulations

A novel dual‐zone fluidized bed reactor was proposed for the continuous adsorption and reduction of NO_x from combustion flue gases. The adsorption and reaction behaviour of such a reactor has been simulated in a fixed bed reactor using Fe/ZSM‐5 catalyst and propylene reductant with model flue gases. Fe/ZSM‐5 exhibited acceptable activity at T = 350°C and GHSV = 5000 h^?1 when O₂ concentration was controlled at levels lower than 1% with a HC to NO molar ratio of about 2:1. XPS and BET surface area measurement revealed the nature of the deactivation of the catalyst. Those performance data demonstrated the feasibility of a continuous dual‐zone fluidized bed reactor for catalytic reduction of NO_x under lean operating conditions.     

The Effect of a Changing Lean Gas Composition on the Ability of NO2 Formation and NOx Reduction over Supported Pt Catalysts

Three model catalysts (Pt/Al₂O₃, Pt/TiO₂, Pt/V₂O₅/TiO₂) were examined in regard to their NO₂ formation ability under a changing lean gas composition. The results show that the NO to NO₂ oxidation function as well as the NO _x reduction under lean gas conditions is affected by a change in the lean gas atmosphere. The NO oxidation activity also decreased with time, for Pt/Al₂O₃ and Pt/TiO₂, and a possible explanation may be platinum oxide formation. This deactivation was not observed for Pt/V₂O₅/TiO₂.     

NO x uptake mechanism on Pt/BaO/Al2O3 catalysts

The NO _x adsorption mechanism on Pt/BaO/Al₂O₃ catalysts was investigated by performing NO _x storage/reduction cycles, NO₂ adsorption and NO + O₂ adsorption on 2%Pt/(x)BaO/Al₂O₃ (x = 2, 8, and 20 wt%) catalysts. NO _x uptake profiles on 2%\Pt/20%BaO/Al₂O₃ at 523 K show complete uptake behavior for almost 5 min, and then the NO _x level starts gradually increasing with time and it reaches 75% of the inlet NO _x concentration after 30 min time-on-stream. Although this catalyst shows fairly high NO _x conversion at 523 K, only ~2.4 wt% out of 20 wt% BaO is converted to Ba(NO₃)₂. Adsorption studies by using NO₂ and NO + O₂ suggest two different NO _x adsorption mechanisms. The NO₂ uptake profile on 2%Pt/20%BaO/Al₂O₃ shows the absence of a complete NO _x uptake period at the beginning of adsorption and the overall NO _x uptake is controlled by the gas–solid equilibrium between NO₂ and BaO/Ba(NO₃)₂ phase. When we use NO + O₂, complete initial NO _x uptake occurs and the time it takes to convert ~4% of BaO to Ba(NO₃)₂ is independent of the NO concentration. These NO _x uptake characteristics suggest that the NO + O₂ reaction on the surface of Pt particles produces NO₂ that is subsequently transferred to the neighboring BaO phase by spill over. At the beginning of the NO _x uptake, this spill-over process is very fast and so it is able to provide complete NO _x storage. However, the NO _x uptake by this mechanism slows down as BaO in the vicinity of Pt particles are converted to Ba(NO₃)₂. The formation of Ba(NO₃)₂ around the Pt particles results in the development of a diffusion barrier for NO₂, and increases the probability of NO₂ desorption and consequently, the beginning of NO _x slip. As NO _x uptake by NO₂ spill-over mechanism slows down due to the diffusion barrier formation, the rate and extent of NO₂ uptake are determined by the diffusion rate of nitrate ions into the BaO bulk, which, in turn, is determined by the gas phase NO₂ concentration.     

A kinetic study of NOx reduction over Pt/SiO2 model catalysts with hydrogen as the reducing agent

Flow reactor experiments and kinetic modeling have been performed in order to study the mechanism and kinetics of NO_x reduction over Pt/SiO₂ catalysts with hydrogen as the reducing agent. The experimental results from NO oxidation and reduction cycles showed that N₂O and NH₃ are formed when NO_x is reduced with H₂. The NH₃ formation depends on the H₂ concentration and the selectivity to NH₃ and N₂O is temperature dependent. A previous model has been used to simulate NO oxidation and a mechanism for NO_x reduction is proposed, which describes the formation/consumption of N₂, H₂O, NO, NO₂, N₂O, NH₃, O₂ and H₂. A good agreement was found between the performed experiments and the model.     

Formation and reactions of isocyanic acid during the catalytic reduction of nitrogen oxides

Reactions which can produce and consume isocyanic acid (HNCO) over two types of catalysts active for the reduction of nitrogen oxides have been investigated. More than 1000 ppm HNCO can be produced by the reduction of 3000 ppm NO with H₂/CO mixtures over a Pt/SiO₂ catalyst. Complete hydrolysis of HNCO to ammonia and carbon dioxide occurs if even weakly catalytic materials, such as CeO₂/SiO₂ and BaO/SiO₂, are placed downstream. Isocyanic acid is also involved as an intermediate in the reaction of nitromethane over CoZSM5 and CuZSM5 under the conditions of hydrocarbon SCR. In the initial stages of reaction there is complete conversion through to N₂ with CuZSM5 but the process stops at ammonia with CoZSM5 at temperatures below 350°C. In both cases, but especially with CoZSM5, isocyanic acid becomes observable as the catalyst deactivates during continuous exposure at temperatures below about 290°C. In situ FTIR measurements indicate that deactivation is due to a reaction between isocyanic acid and ammonia which generates cyclic striazine compounds.     

Fundamental studies of NOx storage at low temperatures

A commercial NO_x storage catalyst (Pt, BaO and alumina containing) was investigated by temperature programmed desorption (TPD) experiments in the temperature range 100–400 °C. The catalyst stored a substantial amount of NO_x at 100 °C using NO + O₂. Nitrites or loosely bound NO species are suggested for this storage, since no NO was oxidised at this low temperature. In addition, the released NO_x during the temperature ramp consisted of mainly NO and at lower temperatures the NO₂ dissociation is limited. Water and CO₂ was found to decrease the storage substantially, 92% for the NO + O₂ adsorption at 100 °C. The total storage for 60 min using NO₂ + O₂ at 200 °C was similar when introducing CO₂ and H₂O. However, the initial total uptake of NO_x was decreased. Initially we probably formed loosely bound NO_x species, which likely are strongly influenced by water and CO₂. After longer time periods are barium nitrates probably formed and they can remove the carbonates by forming stable nitrates, thus resulting in the same total uptake of NO_x.     

Copper active sites for the selective reduction of nitrogen monoxide by propane on Cu-MFI catalysts and non-zeolitic supported-copper solids

The selective reduction of NO by C₃H₈ is performed on copper-based catalysts: Cu/Al₂O₃, Cu/SiO₂, Cu/SiO₂–Al₂O₃ solids, fresh and hydrothermally-treated Cu-MFI with various Si/Al ratios. For all the Cu-MFI solids and for the non-zeolitic supported-copper solids with low copper loadings, O₂ promotes the reduction of NO. The C₃H₈–NO reaction rate correlates with the number of accessible and isolated Cuⁿ⁺ ions deduced from the infrared spectroscopy of adsorbed CO. When only one type of sites is detected, the FTIR spectroscopy of adsorbed CO allows an estimation of the copper dispersion. This revised version was published online in July 2006 with corrections to the Cover Date.     

Migration of Potassium in an Fe2O3/H‐ZSM‐5 Composite Catalyst

The migration of potassium in an iron/H‐ZSM‐5 bifunctional system was investigated by pressing K/Fe₂O₃ and H‐ZSM‐5 in a pellet using 2 t of pressure. These pellets were heated at 350 °C in air for a number of days. Migration of potassium was visualized using energy‐dispersive X‐ray profiling. The distribution of potassium in the Fe₂O₃ phase and the H‐ZSM‐5 phase was approximately constant, with a step change over the phase boundary. The step change varied as a function of the heating time. The amount of potassium migrated from the Fe₂O₃ phase to the H‐ZSM‐5 phase was quantified using NH₃‐TPD. It is shown that an equilibrium distribution between potassium in the Fe₂O₃ phase and the H‐ZSM‐5 phase is obtained after ca. 7 days of heating.     

NO2 and N2 sorption in MFI films with varying Si/Al and Na/Al ratios

MFI crystals or films with controlled thicknesses and different Si/Al ratios were grown on seeded cordierite monoliths using a clear synthesis mixture with template or a template-free gel. The materials were analyzed by scanning electron microscopy, X-ray diffraction, inductively coupled plasma-atomic emission spectrometry, X-ray photoelectron spectroscopy, thermogravimetric analysis and sorption experiments using N₂ or NO₂ adsorbates. The films were uniformly distributed over the support surface. As expected, the specific monolayer N₂ adsorption capacity (mol/g_zeolite) was constant and independent of film thickness. The specific molar NO₂ adsorption capacity was significantly lower than the specific molar monolayer N₂ adsorption capacity, indicating that NO₂ is adsorbed at specific sites rather than evenly distributed in a monolayer. A number of NO₂ adsorption sites with varying strengths were observed by TPD experiments. At 30 °C, the amount of adsorbed NO₂ in the MFI films increased with increasing Al and Na content as opposed to the N₂ adsorption capacity, which was independent of these parameters. At 200 °C, the adsorbed amount of NO₂ was lower than at 30 °C and apparently independent on Al concentration in the Na-MFI films. These results indicate that different mechanisms are involved in NO₂ adsorption. NO₂ may adsorb weakly on Na⁺ cations and also react with silanol groups and residual water in the zeolite, the latter two results in more strongly bound species. Upon NO₂ adsorption, formation of NO was observed. This work represents the first systematic study of the effects of Al and Na content on NO₂ adsorption in MFI films.     

The Remarkable Effect of In2O3 on the Catalytic Activity of In/HZSM-5 for the Reduction of NO with CH4

In/HZSM-5/In₂O₃ catalyst that contained two different kinds of In induced by the impregnating and the physical mixing method respectively has shown remarkable activity for the CH₄-SCR of NO _x comparing with In/HZSM-5. The addition of In₂O₃ into In/HZSM-5 improved the NO conversion through enhancing the adsorption of NO _x over In/HZSM-5.     

An XRD, FTIR and TPD Investigation of NO2 Surface Adsorption Sites of δ, γ Al2O3 and Barium Supported δ, γ Al2O3

In this paper an XRD, FTIR and TPD investigation of NO₂ surface adsorption sites of , Al₂O₃ and barium supported , Al₂O₃ is reported. Aim of this study is to bring additional light on the surface structures involved in NOx adsorption. Two samples of barium supported aluminas have been prepared and aged at 800 °C. These samples were characterised in comparison with the relative alumina support. The XRD characterisation of these samples shows the presence of barium carbonate and barium aluminate supported on alumina. The comparison of the FTIR spectra, before and after NO₂ adsorption, has revealed the formation, upon NO₂ contact, of a complex variety of nitrate and nitrite groups. The thermal desorption of nitrate and nitrite species has been simultaneously studied by means of FTIR spectroscopy and by TPD technique. By comparing the structural, adsorptive and spectroscopic results obtained on alumina and on barium supported alumina samples, a hypothesis on the basic sites active in NO₂ adsorption and of the possible decomposition paths induced by thermal heating are proposed.     

Modelling of NOx adsorption over NOx adsorbers

Modelling of the phenomena involved during the adsorption of NO_x on NO_x trap catalysts was developed. The aim of the model is the prediction of the quantity of stocked barium nitrate as well as the emissions of NO and NO₂, as a function of time and temperature. The mechanism of the process is sounded on the adsorption of gas species (NO, NO₂, O₂) on platinum sites, equilibrium reaction between adsorbed species followed by the formation of Ba(NO₃)₂. This formation of barium nitrate is limited by the thermal decomposition reaction which liberates NO in the gas phase. The kinetic constant of decomposition of barium nitrate was determined by temperature programmed thermogravimetry on pure Ba(NO₃)₂, using the method of Freeman and Carroll. Other kinetic constants bound to the mechanism were estimated by fitting the results of the model to experimental results.The mechanism was validated for various values of the molar fraction of O₂, the molar fraction of NO and various values of the NO/NO₂ ratio in the gas entering the reactor. It was also tested with different catalyst compositions (variation of the platinum and BaO concentrations). The importance of oxygen in the process was clearly demonstrated as well as the promoting role of NO₂.     

Catalytic reduction of NO by propene in the presence of oxygen over mechanically mixed metal oxides and Ce-ZSM-5

The mechanical mixing of Mn₂O₃ or CeO₂ to Ce-ZSM-5 considerably enhanced the rate of the reduction of NO by propene in the low to medium temperature region, although Mn₂O₃ or CeO₂ itself was much less active for this reaction. In contrast, Mn₂O₃ was highly active and CeO₂ was moderately active for the oxidation of NO to NO₂. On the basis of the comparison of the rates of the C₃H₆ + O₂, NO + C₃H₆ + O₂ and NO₂ + C₃H₆ + O₂ reactions over these catalysts, a bifunctional mechanism is proposed, in which Mn₂O₃ and CeO₂ accelerate the oxidation of NO and the subsequent reaction steps between NO₂ and propene proceed on Ce-ZSM-5.     

Support modification to improve the sulphur tolerance of Ag/Al2O3 for SCR of NOx with propene under lean-burn conditions

Ag/Al₂O₃ catalysts with 1 wt% SiO₂ or TiO₂ doping in alumina support have been prepared by wet impregnation method and tested for sulphur tolerance during the selective catalytic reduction (SCR) of NO_x using propene under lean conditions. Ag/Al₂O₃ showed 44% NO_x conversion at 623 K, which was drastically reduced to 21% when exposed to 20 ppm SO₂. When Al₂O₃ support in Ag/Al₂O₃ was doped with 1 wt% SiO₂ or TiO₂ the NO_x conversion remained constant in presence of SO₂ showing the improved sulphur tolerance of these catalysts. Subsequent water addition does not induce significant deactivation. On the contrary, a slight promotional effect on the activity of NO conversion to nitrogen is observed after Si and Ti incorporation. FTIR study showed the sulphation of silver and aluminum sites of Ag/Al₂O₃ catalysts resulting in the decrease in the formation of reactive intermediate species such as –NCO, which in turn decreases NO_x conversion to N₂. In the case of Ag/Al₂O₃ doped with SiO₂ or TiO₂, formation of silver sulphate and aluminum sulphate was drastically reduced, which was evident in FTIR resulting in remarkable improvement in the sulphur tolerance of Ag/Al₂O₃ catalyst. These catalysts before and after the reaction have been characterized with various techniques (XRD, BET surface area, transmittance FTIR and pyridine adsorption) for physico-chemical properties.     

Adsorption and temperature programmed desorption of NO,NO + O2, NO2 and NO + NO2 over CoH-ZSM-5

The NO₂, NO (O₂) adsorption and temperature programmed desorption (TPD) were studied systematically to probe into the selective catalytic reduction of NO by methane (CH₄–SCR) over CoH-ZSM-5 (SiO₂/Al₂O₃ = 25, Co/Al = 0.132–0.312). Adsorption conditions significantly affect the adsorption of NO, NO₂ and NO + O₂. Adsorbed NO species are unstable and desorbed below the reactive temperature 523 K. Increasing adsorption temperature results in the decrease of the adsorbed NO species amount. The amount of –NO_y species formed from NO₂ adsorption increases with the increase of NO₂ concentration in the adsorption process, while decreases significantly with the increase of adsorption temperature. Though NO species are adsorbed weakly on CoH-ZSM-5, competitive adsorption between NO and –NO_y species decreases the amount of adsorbed –NO_y species. Similar desorption profiles of NO₂ were obtained over CoH-ZSM-5 while they were contacted with NO₂ or NO + O₂ followed by TPD. If NO₂ was essential to form adsorbed –NO_y species, the amount of adsorbed –NO_y species for NO + O₂ adsorption should be the least among the adsorptions of NO₂, NO + O₂ and NO + NO₂ because of the lowest NO₂ concentration and highest NO concentration. In fact, the amount of adsorbed –NO_y species is between the other two adsorption processes. These indicate that the formation of adsorbed –NO_y species may not originate from NO₂.     

A TPD study of H2 ON Ru/SiO2 and Ru-Cu/SiO2

In H₂ TPD from Ru/SiO₂, two desorption peaks were observed. Both exchanged H for D in sequential dosing experiments. These hydrogen adsorption states were also found for Ru-Cu/SiO₂, along with another, higher temperature state at 400–500 K. This last state was neither exchangeable with nor replaceable by deuterium subsequently dosed at 150 K. The three chemisorption states are attributed to hydrogen held at the interface between Ru and SiO₂ (< 300 K), adsorbed on Ru particles (310–380 K), and held at the Ru-Cu interface (> 400 K).     

Deterioration Mode of Barium-Containing NOx Storage Catalyst

Barium-containing NO _x storage catalyst showed serious deactivation under thermal exposure at high temperatures. To elucidate the thermal deterioration of the NO _x storage catalyst, four types of model catalyst, Pt/Al₂O₃, Ba/Al₂O₃, Pt–Ba/Al₂O₃, and a physical mixture of Pt/Al₂O₃ + Ba/Al₂O₃ were prepared and their physicochemical properties such as BET, NO TPD, TGA/DSC, XRD, and XPS were evaluated while the thermal aging temperature was increased from 550 to 1050°C. The fresh Pt–Ba/Al₂O₃ showed a sorption capacity of 3.35 wt%/g-cat. but the aged one revealed a reduced capacity of 2.28 wt%/g-cat. corresponding to 68% of the fresh one. It was found that this reduced sorption capacity was directly related to the deterioration of the NO _x storage catalyst by thermal aging. The Ba on Ba/Al₂O₃ and Pt–Ba/Al₂O₃ catalysts began to interact with alumina to form Ba–Al solid alloy above 600°C and then transformed into stable BaAl₂O₄ having a spinel structure. However, no phase transition was observed in the Pt/Al₂O₃ catalyst having no barium component, even after aging at 1050°C.     

An in situ infrared study of NO reduction by C3H8 over Fe‐ZSM‐5

The interactions of NO, O₂ and NO₂ with Fe‐ZSM‐5, as well as the reduction of NO by C₃H₈ in the presence of O₂, have been investigated using in situ infrared spectroscopy. The sample of Fe‐ZSM‐5 (Fe/Al =0.56) was prepared by solid‐state ion exchange. NO adsorption in the absence of O₂ produces only mono‐ and dinitrosyl species associated with Fe²⁺ cations. Adsorbed NO₂/NO₃ species are formed via the reaction of adsorbed O₂ with gas‐phase NO or by the adsorption of gas‐phase NO₂. The reduction of NO in the presence of O₂ begins with the reaction of gas‐phase C₃H₈ with adsorbed NO₂/NO₃ species to form a nitrogen‐containing polymeric species. A reaction pathway is proposed for the catalyzed reduction of NO by C₃H₈ in the presence of O₂. This revised version was published online in July 2006 with corrections to the Cover Date.