Search

One of the challenges of full-scale computer simulation of a catalytic reactor is to consider the different scales involved in the problem in a practical fashion. In a monolith catalytic converter, these scales range from the molecular scale for the reactions, through the pore scale, washcoat scale, channel scale, and finally the full converter scale. This paper describes the implementation of a model reduction methodology using look-up tables to perform a consistent comparison of six different catalytic converters used for the catalytic combustion of methane. A detailed mechanistic model for methane combustion is used. Diffusion in the non-uniform washcoat is considered. The converters have different cell densities and wall thicknesses. Steady state and transient light-off simulations are performed. Efficient computational speed is achieved by successive model reduction, which allows the preservation of detailed small-scale information. The results obtained show that there is a non-intuitive relationship between the various operating parameters, which can only be deduced from a comprehensive model.

It has recently been shown that the oxygen content of a three-way catalyst can be observed directly by a microwave-assisted technique. In this contribution, several control strategies based on this microwave measurement approach are proposed. In the first step, the lambda probe downstream of the catalyst is replaced by the microwave-based state observer. In the second step, the microwave-based state observer replaces both lambda probes, upstream and downstream of the catalyst. A model-based rapid control prototyping approach is used to test the selected control schemes and to compare their performance to the conventional air-fuel ratio control. This comparison is completed using a complex catalyst model and measured data for a realistic driving cycle. It is shown that under idealized conditions the new microwave-based control strategies can outperform the conventional lambda control based on two lambda probes. One important result is that improving the time resolution of the current microwave measurement system is essential to use the full potential of the new approach.

This paper describes an experimental and theoretical investigation of the oxidation of carbon monoxide, nitrogen monoxide and propene, and the reduction of nitrogen monoxide by propene over two diesel oxidation catalysts. Synthetic exhaust gas mixtures are used. The catalysts were commercial monolith samples of 400 CPSI with a total PGM loading of 95 g/ft 3 . One monolith had platinum catalyst only, and the other was a bimetallic catalyst of 4:1 by mass platinum and palladium. The bimetallic catalyst was more active, and showed a different behaviour for propene oxidation than the Pt only catalyst. Global kinetic models for the oxidation of CO, C 3 H 6 , and NO, and the selective catalytic reduction of NO to N 2 and N 2 O are developed. A good agreement between experimental results and simulation is observed.

With growing, accessible natural gas reserves, there is renewed interest in natural gas engines for the wider automotive market. A potential issue, if legislated, is controlling (reducing) the un-combusted CH 4 emissions. The key challenge in catalytic CH 4 oxidation is the high temperature required, relative to other hydrocarbon species. In this context, a monolith-supported Pt/Al 2 O 3 catalyst was evaluated for CH 4 combustion under fuel lean and fuel rich mixtures using temperature programmed reaction (TPRxn) and step-change temperature and oxygen level experiments. The experiments included performance evaluation during both ignition (increasing temperature) and extinction (decreasing temperature after ignition). Conversion hysteresis was observed, with the conversions during extinction higher than those during ignition under fuel lean or stoichiometric combustion reaction conditions. Results obtained demonstrate that this hysteresis effect can be used to achieve high CH 4 oxidation conversions at temperatures lower than that required for ignition, admittedly first through using high temperatures to obtain ignition, then lowering the temperature to take advantage of the hysteresis. Results also suggest that changing O 2 levels can lead to similar benefits. With the assumption that lean conditions and lower exhaust temperatures are associated with improved fuel economy, while higher exhaust temperatures and fuel rich conditions lead to higher CH 4 oxidation rates over the catalyst, the findings presented clearly demonstrate the potential to achieve both via a cyclic operating approach, with the frequency on the order of 10 s of minutes. Data obtained when cycling between temperatures above and below the ignition point and between excess O 2 and stoichiometric O 2 levels proved this hypothesis.

This paper describes an experimental and theoretical investigation of the selective catalytic reduction of NO by propene over two diesel oxidation catalysts. Both catalysts were commercial monolith samples of 400 CPSI with a PGM loading of 95 g/ft3. In one monolith the catalyst was platinum only, and in the other a bimetallic catalyst of 4:1 by mass platinum and palladium. Both catalysts showed some activity for the SCR reaction, although relatively large amounts of N2O were also formed. Global kinetic models for the oxidation of propene, NO and the SCR reactions are developed. Several models from the literature were tested and compared. New models based on earlier work are proposed. A good agreement between experimental results and simulation was observed.

Three-way catalyst material was deposited inside the pores of a ceramic particulate filter and the pore geometry as well as the distribution of the catalyst in the pores was determined by X-ray microtomography. On the resulting 3D geometry, the flow field through the pores was computed and the convection-diffusion-reaction equation in the open pores and the catalyst particles was solved assuming a first order model reaction taking place in the catalyst. The conversion in the filter wall was compared to a homogeneous model with the same dimensions and catalyst content and it was found that the conversion in the pore network is lower than predicted by the homogeneous model, indicating the presence of some kind of in-pore micro-scale transport limitation. Analysis of the flow field showed channeling of the flow through a few large pores and blocking of many other pores by the catalyst that leads to a broadening of the residence time distribution. While the broadened residence time distribution has some negative effect on the conversion, diffusion limitation in the catalyst particles was identified as the main reason for the reduced conversion compared to the homogeneous model. When diffusion in the washcoat was described by a standard effectiveness factor based on a generalized Thiele modulus with an effective diffusion length fitted to the results of the full pore-scale simulation, a very good agreement between the pore-scale simulation and the homogeneous model was found. The effective diffusion length obtained by this fit is surprisingly large, compared to the apparent size of the washcoat particles. This can be explained by the limited accessibility of the washcoat due to the confinement in the pore structure. Finally, the 3D micro-scale model was coupled to a channel-scale model representing one pair of inlet and outlet channels. It is shown that micro-scale diffusion limitations have a higher impact on the overall reactor performance than the channel-scale mass transport. This suggests that for filter design and for reactor modeling more emphasis should be placed on micro-scale transport effects.

This paper presents a detailed analysis of global kinetic models for the oxidation of carbon monoxide and propene, in the presence of water, hydrogen, carbon dioxide, excess oxygen and nitrogen. The context is the automotive catalytic converter for a lean burn engine. Experimental results are presented for a commercial platinum diesel oxidation catalyst used in the form of a 400 CPSI washcoated monolith. Models are developed based on the classical LHHW format presented by Voltz et al., Ind. Eng. Chem. Prod. Res. Dev. 1973;12:294. The models are developed stepwise for experiments for each reactant alone and the predictive ability of the results is examined. It is found that it is necessary to use experiments with both reactants present to develop a reliable model.

Upcoming legislation will most likely require the introduction of particulate filters for gasoline engines. One attractive technical solution combines the three way catalytic functionality and the filter in one device, the so called ‘catalysed Gasoline Particulate Filter’. The current study uses temperature step experiments and CO oxidation as a test reaction to compare the catalysed particulate filter and the conventional flow-through monolith with respect to their dynamic cold-start behaviour. Despite the fact that the two reactor configurations are tested with identical washcoat formulation, precious metal loading and thermal mass, experiments show a significantly delayed cold-start for the particulate filter. The resulting cumulated CO emissions of the catalysed filter exceed those of the open monolith by 190–300% in the temperature range between 250 °C and 325 °C. The experimental results are analysed by means of numerical simulation. In a first step a kinetic model of the CO oxidation is parameterised using only experimental data obtained for the conventional flow-through catalyst. The resulting kinetics are implemented in a model of the wall-flow filter. Without further modification of the kinetic parameters, this model quantitatively predicts the cold-start behaviour of the catalysed filter. Finally, the numerical model is used in a sensitivity analysis to identify and quantify the individual physical effects contributing to the experimentally observed difference in the light-off behaviour. It is shown that a part of the observed difference in the cold-start performance can be traced back to differences in cell density and the heat capacity of the plugs. Even at identical cell density and without the plug effect the filter shows significantly higher CO emissions. It is shown that this intrinsic difference between the filter and the conventional monolith can be quantitatively explained by differences in heat transfer, internal mass transfer and external mass transfer.

This paper presents a methodology for multi-scale modelling using pre-computed data for the small scales. By pre-computing the required data at the micro- and meso-scales and storing them in a look-up table, these small-scale effects can be captured in a macro-scale model that executes with practicable execution times. The methodology is presented in the context of modelling a washcoated monolith reactor for the catalytic combustion of methane using a detailed multistep mechanistic model. The modelling is illustrated as a sequence of steps of increasing model simplification. In the first instance, only the reaction rates are pre-computed. Then the washcoat diffusion is included using a 2D axisymmetric model with a washcoat of uniform thickness, then with a non-symmetrical channel with a washcoat of non-uniform thickness. It is shown how a simple 1D single channel model can accurately account for the presence of a non-uniform washcoat. Finally, the external transport resistance is included in the pre-computation step, and it is shown that a simple one dimensional pseudo-homogeneous single channel model can be used provided the effective average rates are appropriately calculated. Execution times are shown to decrease by several orders of magnitude at minimal loss of solution accuracy. The use of pre-computed data allows for the inclusion of complex heat and mass transfer phenomena, as well as complex kinetics in full scale converter models at relatively small computational cost. The implementation of look-up tables into full scale CFD models is described, and optimal strategies proposed.

Three-way catalyst-based automotive exhaust gas aftertreatment is of high importance to meet today’s emission standards. To determine in situ the oxygen loading state of three-way catalysts, a microwave cavity perturbation method is used. In this study, it is investigated whether this measurement setup that had originally been described for full-sized catalysts can be transferred to a lab test bench using cores of 1″ diameter. The initial tests were successful and a high correlation between the oxygen loading degree dependent resonance frequency and the conversion was found. As an application example of the new in situ characterization technique, the steady state degree of oxidation of a three way catalyst was measured as a function of the exhaust stoichiometry. The experimental results are compared with the prediction of a recently published improved kinetic model that takes into account the oxidation of reduced ceria by H2O and CO2. It is shown that the experimental observations agree very well with this improved model. This result provides evidence that under typical operating conditions, the degree of oxidation of the three way catalyst is controlled by equilibrium effects.

The oxygen loading degree in TWCs, the amount of stored ammonia in SCR catalysts, the NOx loading degree in LNTs, or the soot loading of DPFs play a key role in automotive exhaust gas aftertreatment. Today’s methods determine the catalyst state indirectly. They utilize gas sensors installed up- or downstream of the catalysts and the catalyst state is inferred from the sensor signals. This overview reports on the status of an alternative approach based on the interaction of electromagnetic microwaves with the catalyst material. Since the catalyst state is strongly correlated with the electrical properties of the catalyst material itself, this concept shows a great potential.

It is well known that conversion as a function of temperature hysteresis can occur during ignition and extinction exothermic reaction experiments, such as CO oxidation over Pt/Al2O3, with the activity during the ignition process not matching that during the extinction process. Conversions being higher during extinction than that during ignition are often observed. Several explanations have been proposed in which heat effects, different catalyst surface states, and different Pt oxidation states are the most common. In this work CO oxidation hysteresis behavior, when in a mixture with C3H6, was investigated. The results show that when C3H6 was absent, CO oxidation followed normal hysteresis behavior; however, when C3H6 was added to the mixture, the catalytic activity during the extinction phase decreased. As the C3H6 concentration in the mixture increased, the hysteresis loop became smaller and ultimately reverse hysteresis was observed. The decrease in catalytic activity during extinction was due to the formation of C3H6 oxidation intermediate species. These species competed with CO for active sites, thus inhibiting CO oxidation, and were not present during ignition as CO was the dominant adsorbed species when starting at low temperature.

This paper describes a computational study of a proposed recuperative catalytic converter intended for the catalytic combustion of lean methane streams. The design is based on the principle of alternating layers of washcoated monolith channels through which the fluid passes in counter-current flow. Heat transfer occurs continuously across the boundary separating the layers, thus the exiting hot gas exchanges energy with the incoming cold gas. Monoliths composed of cordierite and silicon carbide are compared, each of which had 12% by volume washcoat applied. The effects of layer thickness, monolith length and space time are compared. It is found that the monolith comprised of higher thermal conductivity material is more effective in enhancing the temperature in the reactor. Conversion of methane entering the reactor at relatively low temperatures can be oxidised with some added thermal energy.

This paper demonstrates that the use of a solution-mapping approach can accelerate the simulation of a dual-layer ammonia oxidation catalyst by several orders of magnitude. In a first step, a 1-D + 1-D model of the dual-layer catalyst is set up. The results of this model are compared to the numerical solution of a 2-D model, and it is shown that the 1-D + 1-D model provides an accurate approximation of the full 2-D solution. In a second step, spline interpolation functions are constructed that map the solution of the mass balance equations for one axial volume element of the 1-D + 1-D model as a function of the inlet concentrations, temperature and residence time. During the runtime of the simulation, concentration profiles can be computed by simple successive calls of the interpolation function without the need for a numerical solver. The 5-dimensional spline functions reproduce the output concentrations of a reactor model with 35 axial volume elements with an accuracy of better than 1.25%. The use of the solution mapping reduces the computation time for a steady-state solution of the monolith model from 97.5 s down to 10 ms.

Today, platinum based ammonia oxidation catalysts are applied in automotive exhaust systems to avoid ammonia slip from the SCR catalyst. In this paper we present a dual layer catalyst design with an additional SCR layer on top of the Pt/Al2O3 layer. Laboratory experiments show that the additional SCR layer significantly reduces NO formation, improves N2 selectivity but also reduces the overall NH3 conversion. A mechanistically based kinetic model for NH3 oxidation on platinum is presented. The model is parameterised using experiments with a single layer Pt/Al2O3 catalyst and well describes the NH3 conversion and the selectivities for the products N2, N2O and NO. Additionally NO oxidation to NO2 is included in the model. Based on the mechanistic model for the platinum layer and a previously published model for the SCR layer, a 2-D model of the dual layer catalyst is set up. This model takes into account the diffusion and reaction in the two washcoat layers of one monolith channel. The 2-D model reproduces the experimental data well, with the mechanisms for the two layers parameterised in separate experiments and no additional parameters adjusted for the dual layer model. Finally, the model is used to study the effect of some design parameters such as catalyst size, NO oxidation capacity of the Pt-layer and the thickness or diffusion coefficient of the SCR layer.

The oxidation of NO, CO and hydrocarbons (HC) individually, in mixtures, and with NO2 were investigated over a monolith-supported Pt/Al2O3 catalyst under oxidising conditions. Although competitive adsorption and inhibition by other species on oxidation reactions is a relatively well-known phenomenon, this study represents a more comprehensive examination of such effects between key components in vehicle exhaust gases. NO2 was completely reduced by CO and C3H6, under NO2 limited conditions, at temperatures as low as 110°C and at temperatures above 140°C with dodecane and m-xylene. NO2 was then again observed once the extent of oxidation of the other species by oxygen was significant. Under the conditions tested, NO, CO and HC oxidation was inhibited by NO2 in the feed gas mixture. HC were also found to inhibit the oxidation of NO and other HC species due to site adsorption competition. For CO, HC did not change the onset of oxidation, but did inhibit the extent after their light off began. At low temperatures, CO was initially found to inhibit NO oxidation, but at higher temperatures, once CO oxidation was significant, CO promoted NO conversion to NO2. The observed inhibition effects of the different gases on HC oxidation were not additive, indicating one species would cause inhibition, but once its inhibition ended, another species could still then cause inhibition. The combined effect of C3H6, NO and NO2 on CO conversion was found to be additive. This is because CO oxidation started prior C3H6. © 2011 Canadian Society for Chemical Engineering

A highly efficient single channel monolith reactor model for the oxidation of ammonia is implemented. A mechanistic model for ammonia oxidation on platinum is used, and all internal and external mass transfer effects are included. The model efficiency is derived from the use of pre-computed solutions of the mass balance equations. Spline interpolation functions are constructed from pre-computed solutions of the mass balances of one volume element of the reactor model. It is shown that these spline functions reproduce the exact outlet concentrations of the volume element with a relative error of less than 3%. This solution mapping approach allows computation of concentration profiles in the reactor by simple successive calls of the interpolation function without any numerical solution. Outlet concentrations of the complete reactor computed in this way show a relative error of less than 2%. Application of the solution mapping approach speeds up the solution of the concentration profiles by a factor of more than 3200, compared to the numerical solution of the mass balances. In this way one steady state concentration profile in an ammonia oxidation catalyst can be computed in less than 0.003 s, despite the fact that the model uses mechanistic surface kinetics and includes the radial diffusion in the washcoat.

Eine neue Methode zur modellbasierten Optimierung der zeitabhangigen Harnstoffdosierung fur SCR-Katalysatoren (Selective Catalytic Reduction) wird prasentiert. Es werden Dosierprofile fur zwei verschiedene Katalysatoren (Fe- und Cu-Zeolith) berechnet. Dabei zeigt sich, dass die relative Leistungsfahigkeit der beiden Katalysatoren vom gewahlten Dosierprofil abhangt. Dies bedeutet, dass ein aussagekraftiger Vergleich zwischen verschiedenen Katalysatoren nur moglich ist, wenn jeder Katalysator mit einer individuell optimierten Harnstoffdosierung getestet wird. A new method for the model-based optimization of the time-dependent urea dosing for SCR catalysts (selective catalytic reduction) is presented. Optimized dosing profiles are computed for two different catalysts (Fe- and Cu zeolite). It is shown that the relative performance of the two catalysts depends on the choice of the dosing profile. This means that a meaningful comparison between different catalysts is only possible if each catalyst is tested with an individually optimized urea dosing strategy.

A microkinetic model is developed to simulate the promoting effect of H 2 on the light-off of CO oxidation on platinum. It is shown that the coupling of the CO and the H 2 oxidation sub-mechanisms, via the hydroxyl intermediate and its reaction with CO to carboxyl, is crucial to describe the process. This carboxyl-path was not considered in previous modeling studies.

Hydrogen production via hydrocarbon steam reforming and water gas shift reactions was investigated over a monolith-supported Pt-based diesel oxidation catalyst. The evaluation included comparison between constantly rich gas composition conditions and cycling between rich gas conditions and an inert stream. Analysis was performed along the catalyst length at temperatures ranging from 200 to 500 °C. During the constant inlet composition experiments, C3H6 steam reforming started at 375 °C, while dodecane steam reforming began at 450 °C, and resulted in less hydrogen produced. With a mixture of C3H6 and dodecane, hydrogen production originated solely from C3H6 steam reforming and, under otherwise identical conditions, was less than that observed with only C3H6, but higher than that with only dodecane. Hydrogen production from the water gas shift reaction was higher than that observed with hydrocarbon steam reforming and started at 225 °C. During cycling experiments, hydrogen production via hydrocarbon steam r...

This paper demonstrates how pre-computed rate data can be used to enable the numerically efficient implementation of mechanistic kinetics in a reactor model for an automotive ammonia slip catalyst. In a pre-processing step the source terms for the gas species are mapped as a function of gas composition and temperature. From this map (80,000 data points), a spline interpolation function is constructed. The spline function approximates the numerical steady state solution of the kinetic model with an average error of less than 1%. Application of the rate mapping procedure results in a speedup of about two orders of magnitude. The spline approximation is implemented in a 2-dimensional model of a monolith reactor channel that includes diffusion in the washcoat layer. Depending on the channel diameter and the washcoat thickness, diffusion limitations in the open channel and in the washcoat can have a significant influence on the NH 3 conversion. Surprisingly, the selectivity for the products N 2 , N 2 O and NO is not sensitive to diffusion effects in the gas phase or in the washcoat.

It is shown that a platinum catalyst operated under a transient temperature profile exhibits a hysteresis behavior. The catalytic activity during heat-up phases exceeds the activity of the cool down phases. This effect is the opposite of the well-known hysteresis that can be caused by the existence of multiple steady states or the thermal inertia of the catalyst. It is therefore referred to as ‘inverse hysteresis’. If the catalyst is operated at one temperature for a longer time, a steady state activity is observed that is in-between the higher activity of the transient heat-up and the lower activity of the transient cool down. The most likely explanation for the observed hysteresis behavior is a reversible oxidation of the platinum surface by NO2 and a reduction of the platinum surface by NO. A micro-kinetic surface mechanism including the reversible surface oxidation has been developed and implemented in a monolith reactor model. The observed hysteresis effects can be quantitatively reproduced by the numerical model. If the feed contains CO/NO/O2 an inverse hysteresis is observed for NO and CO oxidation. This means that in the presence of NO the hysteresis behavior of the CO oxidation switches from the typical standard hysteresis to an inverse hysteresis.

We demonstrate in this paper that the effective diffusion coefficient in the wall of a particulate filter can be determined by measuring the diffusion of NO between two adjacent channels of a filter segment in a simple apparatus based on a modified experimental method of Beeckman. The effective diffusion coefficient in the walls of an uncoated SiC particulate filter is determined as 2.8 × 10−6 m2/s at room temperature. Coating the filter with 30 or 65 g/L of wall-integrated washcoat leads to an insignificant increase in the diffusion coefficient, whereas coating the same filter with a high washcoat loading of 140 g/L leads to a decrease of the diffusion coefficient to 2.0 × 10−6 m2/s. All of the determined diffusion coefficients increase with temperature proportional to T1.5. This indicates that the diffusion in the wall is mainly molecular diffusion and that Knudsen diffusion plays a minor role. Fitting the parallel pore model to the experimental diffusion coefficient of the uncoated filter results in a ...

The setup of a detailed model for the NH3-SCR process on iron exchanged zeolite catalysts is described, which takes into account the reactions of NH3, NO and NO2 occurring under typical exhaust conditions. Validation experiments for a standard SCR gas mixture show that the model is able to describe the dynamic behaviour of this system at transient temperature conditions. The model is used for the parameterization of an ammonia dosing controller. To prove the functionality of this controller it was implemented on a synthetic gas test bench and tested by a dynamic temperature experiment. This setup allows the evaluation of the parameterized controller without any disturbances from engine or injection effects.

Existing reaction mechanisms for three-way catalysts ignore the oxidation of Ce2O3 by H2O or CO2 and therefore treat oxygen storage on ceria as a kinetically controlled non-equilibrium process. Such mechanisms do not correctly reproduce a number of practically important phenomena. Based on transient concentration step experiments, this paper demonstrates that the oxidation of reduced ceria by H2O or CO2 plays an important role for the understanding of the oxygen storage dynamics. The results of the step experiments are well reproduced by a simple three-reaction kinetic model that takes into account the equilibrium character of oxygen storage. There are a number of effects that cannot be understood based on the conventional non-equilibrium mechanisms: • The oxygen storage capacity is lowered by the presence of H2O and CO2. • The oxygen storage capacity depends on the amplitude of the lambda-oscillations. • Following a rich-lean step the catalyst emits CO/H2 when there is no CO/H2 in the inlet. • Delayed CO/H2 emissions in the so-called fuel cut-off scenario. • The capability of the oxygen storage to dump lambda-oscillations, even if the average lambda is slightly rich. All these effects are explained and quantitatively predicted by the equilibrium based oxygen storage model.

A microkinetic model is developed for the NO-oxidation on platinum. The approach we followed employs semi-empirical methods, optimization of the parameters of elementary-like reactions, and thermodynamic constraints. The model is capable of capturing reasonably light-off experimental data.

The oxidation of ammonia on a Pt/Al2O3 coated monolith has been studied under automotive NH3-slip catalyst conditions. Ammonia conversion as well as the selectivities towards the products N2, N2O and NO are well described by a mechanistic model that is based on reaction mechanisms originally developed for NH3 oxidation in nitric acid production plants.

This paper describes the development and validation of a global kinetic model for a diesel oxidation catalyst. The model is based on an extension of the classical Voltz model, and incorporates CO, H2, NO and HC oxidation, and NO reduction by HC. The model is tested against experimental light-off curves for a variety of scenarios. A one dimensional single channel model for the reactor was used. Parameter optimization was performed using a general pattern search algorithm. The model is shown to be able to capture the ignition curves reasonable well.

It is shown that a mapping of precomputed rate data can be used to implement efficiently detailed surface mechanisms into reactor simulations. The procedure is demonstrated using a surface mechanism for CH 4 oxidation on platinum (11 surface species, 19 reactions). It is found that a spline representation based on 7000 data points reproduces the effective CH 4 conversion rates with an error of less than 0.5%. To demonstrate the potential of the approach a 3D model of a monolith channel including the irregularly shaped washcoat is implemented in a commercial CFD program. A steady state solution of this model takes 20 s on a 1.5 GHz Pentium computer. This corresponds to a speed up of three orders of magnitude compared to the solution of the same model coupled to a full numerical solution of the surface chemistry.

This paper reports the results of a study of the combined oxidation of hydrogen and carbon monoxide on a platinum diesel oxidation catalyst. Experimental results for the light-off curves obtained in a monolith supported catalyst are shown for different concentrations. The presence of hydrogen is shown to promote the oxidation of CO, with the largest effect shown for the first small addition of hydrogen, then a progressively decreasing effect as the hydrogen concentration is increased. The enhancement effect can be successfully simulated using a kinetic model in which the activation energy for the desorption of CO is decreased as a function of hydrogen adsorption. Hydrogen is allowed to adsorb on sites only accessible to hydrogen and not CO or oxygen.

The dynamic behaviour of the NO SCR-reaction on Fe-exchanged zeolites was investigated by the transient response method (TRM). The results were used to develop a mathematical model for catalysts containing Fe-exchanged zeolites capable of describing the dynamic phenomena observed.

Different global kinetics for the oxidation of NO on platinum are compared in this work. The general form of the equation for the reaction rate is $r = k \cdot c_{NO}^\alpha \cdot c_{O_2 }^\beta \cdot c_{NO_2 }^\gamma \cdot TD^\delta$ with $TD = \left( {1 - c_{NO_2 } \cdot c_{NO}^{ - 1} \cdot c_{O_2 }^{ - 0.50} \cdot K^{ - 1} } \right)$ . Furthermore steady-state and temperature-programmed experiments were performed. The best results of simulation coupled with parameter estimation were obtained using α = 0.28, β = 0.49, γ = 0 and δ = 1, along with an activation energy of 47.5 kJ mol−1.

The use of the Diesel engine is increasing, while at the same time emission limits are becoming more stringent. To reduce emissions, novel catalyst converter designs have been proposed, including the placement of a small converter before the turbocharger. The role of this catalyst is to provide some reduction of the pollutant level prior to the main catalyst. This catalyst is typified by high flow rates. In this paper the performance of small catalytic converters operated at high flowrate is examined using computer simulation. The oxidation of CO is used as a model reaction. It is shown that for a typical oxidation catalyst, conversions of the order of 30% can be achieved at typical temperatures and a GHSV of 5 million h−1.

This paper describes an approximation method for the estimation of effectiveness factors in porous catalysts that have reactions with non-linear kinetic models. The reaction rate expression is expressed as a combination of two asymptotic solutions based on a local concentration, leading to an approximate rate function containing four parameters. The effectiveness factors computed with the approximated rate are shown to be close to the values calculated from the real rate function for a variety of non-linear LHHW type models. The advantage of the approximate method is that a given set of four parameters, which can easily be calculated, uniquely identifies an effectiveness factor. This observation enables the construction of a table from which the effectiveness factor can be readily extracted during reactor simulation, thus resulting in a decrease in computational effort.

Simulation models of three-way catalysts (TWC) are very useful for the design and tests of catalyst controllers. When applying such models to controller design, accurate descriptions of the emissions restricted by legislation and of the emissions driving the λ sensor (NOx, CO, HC, H2 and O2) are important. A kinetic one-channel model of a TWC, capable of calculating the demanded output, was recently published. That model was developed based on measurements from an engine test bench. Since, the dynamic behaviour of the catalyst is very complex and the variation of the exhaust gas composition is very limited on an engine test bench, various phenomena can hardly be analysed separately on an engine test bench. This fact led to the choice of the model-gas test bench, which allows the study of the individual phenomena under idealised conditions. This contribution now is to validate that model using measurement results obtained on a model-gas test bench. To analyse the performance of the TWC model, the response of the TWC to a defined input is measured under various conditions and compared with results of simulations. The main goals of this investigation include the analysis of dynamic effects, such as the deactivation of the catalyst under rich conditions, the role of water and hydrogen (water–gas shift reaction), and the storage of components other than oxygen.

Diesel soot abatement via diesel particulate filters composed of so-called wall-flow monoliths is well established. Today, due to recent improvements in the production technology full-featured catalyst functionality can be implemented in the filter walls. This work focuses on a comparison of the reactor performance of the wall-flow filter and the conventional flow-through monolith. To this end a two-dimensional numerical model is set up for each of the two reactor configurations. Concentration profiles in the wall-flow filter systematically change as a function of flow velocity. At high flow velocities transport from the inlet channel into the porous wall is nearly entirely dominated by convection. This leads to uniform axial concentration profiles in the inlet and outlet channel and a steep gradient in the porous wall. At low velocities radial transport into the porous wall is dominated by diffusive transport. This leads to a negligible radial concentration gradient between the inlet and the outlet channel. Under most operating conditions relevant for an automotive exhaust catalyst the flow velocity is in an intermediate range with contributions of diffusive and convective transport. The transition from entirely convection dominated transport at high space velocities to increasingly diffusion dominated transport at lower flow velocities is similarly found for first order kinetics and a generalized Langmuir–Hinshelwood–Hougen–Watson (LHHW) rate law. Wall-flow filters show systematic conversion advantages over the conventional monolith for a first order reaction. For a reaction with LHHW-type kinetics this effect is not generally observed. It is one major result of this work that the relative performance of the two reactor configurations depends on the kinetics of the catalyzed reaction.

This paper presents experiments and model predictions for the oxidation of CO over a platinum catalyst in a monolith reactor. Experimental behavior is broadly consistent with previously reported work on CO oxidation. Ignition–extinction (light-off) curves demonstrated the presence of multiple steady states with a hysteresis effect. The two branches corresponding to the two states of predominantly CO covered or oxygen covered. Admission of CO pulses to an oxygen covered surface results in reaction, indicating the occurrence of adsorption of CO on an oxygen covered surface. A model based on adsorption and surface reaction using the classical Langmuir–Hinshelwood (LH) approach qualitatively was able to reproduce the light-off behavior. The best model assumes dissociative chemisorption of oxygen on two surface sites. It was superior to a model proposing molecular adsorption of oxygen followed by rapid dissociation. Addition of steps allowing CO adsorption on an oxygen filled surface via an “oxygen compression” mechanism enable the qualitative description of the reactor response to step inputs of CO to an oxygen rich feed stream. Some parameter adjustment remains to allow a better fit between experiment and model predictions.

This paper presents experimental and modelling results for the oxidation of mixtures of hydrogen and carbon monoxide in a lean atmosphere. Transient light-off experiments over a platinum catalyst (80 g/ft 3 loading) supported on a washcoated ceramic monolith were performed with a slow inlet temperature ramp. Results for CO alone agree with earlier results that predict self-inhibition of CO; that is an increasing light-off temperature with increasing CO concentration. Addition of hydrogen to the feed causes a reduction in light-off temperature for all concentrations of CO studied. The most significant shift in light-off temperature occurs with the addition of small amounts of hydrogen (500 ppm, v/v) with only minor marginal enhancement occurring at higher hydrogen concentrations. Hydrogen alone in a lean atmosphere will oxidise at room temperature. In mixtures of hydrogen and CO, the CO was observed to react first until a conversion of about 50% was observed, at which point the conversion of hydrogen rapidly went from 0 to 100%. Simulations performed using literature mechanistic models for the oxidation of these mixtures predicted that hydrogen ignites first, followed by CO, a direct contradiction of the experimental evidence. Upon changing the activation energy between adsorbed hydrogen and oxygen, the CO was observed to oxidise first, however, no enhancement of light-off was predicted. The effect cannot be explained by the mechanistic model currently under discussion.

This paper presents the results of a series of simulations of the light-off of carbon monoxide in excess oxygen in a single channel monolith reactor. The catalytic washcoat geometry is tested for its effect on the light-off position and shape. The simulations used a combination of two and three-dimensional modelling with non-linear oxidation kinetics. Different temperature ramp rates were employed. It is seen that the shape of the washcoat has an influence on the shape of the light-off curve, especially for high values of gas hourly space velocity and larger temperature ramp rates. The washcoat geometry that gives the fastest rise to complete conversion is the one that is the most non-uniform in shape; that is, the one that has thin sections.

Diesel particulate filters composed of so-called wall-flow monoliths are well established devices for diesel particulate abatement. Recent improvements in production technology allow implementation of full-featured catalyst functionality in the filter walls. From a reactor engineering point of view such wall-flow reactors with wall-integrated catalyst show fundamental differences compared to conventional flow-through monoliths. The complex interactions of convection, diffusion and reaction in the wall-flow monolith are studied by means of numerical simulation. A two-dimensional model for the flow in one pair of inlet/outlet channels with a generic first order reaction in the catalytic filter wall is developed. Concentration profiles in the reactor and a conventional flow-through catalyst are compared. It is found that in the range of moderate reactor conversion concentration gradients along the inlet channel of the filter are small. Thus the reactor can be described by an approximate one-dimensional model, taking into account only the radial flux through the filter wall and assuming a constant inlet concentration in axial direction along the inlet channel. Light-off curves are computed for the wall-flow and for the conventional flow-through monolith. Significantly better conversion is found for the wall-flow configuration. This can be explained by mass transfer limitation in the conventional flow-through monolith.

This paper describes a method for the determination of effectiveness factors in a monolith washcoat of non-uniform thickness. The method is based on the previously reported slice concept, in which the monolith is divided into a series of slices, each of which is analysed using a one-dimensional approach. In each one-dimensional slice, an apparent characteristic length is used to compute the Thiele modulus. In the approach described here, the apparent characteristic length is determined by conducting a single two-dimensional simulation for a first-order reaction in the kinetically controlled region. The characteristic length is then calculated from the flux at the surface using an analytical solution. Results are shown for the idealized circular washcoat in a square channel, as well as for a real washcoat in a square channel and a sinusoidal shape. Results for first-order and LHHW kinetics show good agreement between the one-dimensional approximation and the true solution. Both average and local effectiveness factors can be calculated within acceptable range for monolith reactor simulations.

This paper reports the results of a numerical investigation of the diffusion and reaction in the washcoat of a catalytic monolith reactor. Both idealized shapes and real geometries are tested. It is shown that for non-uniform washcoat thickness, the Sherwood number varies along the channel/washcoat interface. The internal diffusion resistance, expressed in terms of an effectiveness factor, cannot be represented in terms of a unique curve using the generalized Thiele modulus approach. The most significant deviation occurs when the washcoat has the greatest variation in thickness. The internal diffusion affects the magnitude of the external mass transfer resistance. Three factors significantly affect the rates and role of mass transfer. These are (1) the washcoat thickness, (2) the channel radius, including its non-uniformity around the channel, and (3) the angular diffusion in the washcoat caused by variable thickness.

The product branching ratio of [NH2 + NO → N2H + OH]:[NH2 + NO → N2 + H2O] has been determined in the temperature range of 1340−1670 K in a shock tube study with laser photolytic generation of the NH2 radicals. Sensitive frequency-modulation detection of the NH2 radical enables experiments with very low initial radical concentrations and, hence, virtually no interference from secondary reactions or dependence on the overall rate coefficient of the reaction. The branching ratio, α, defined as α = k1a/(k1a+k1b), increases from 0.42 at 1340 K to 0.53 at 1670 K. Over this temperature range our results can be expressed as α = 0.5 + (3.36 × 10-4)(T/K − 1600). A detailed error analysis yields an uncertainty in the values of α of ±0.02 near 1500 K, increasing to ±0.05 at the temperature extremes.

Frequency modulation detection of NH2 in shock tube kinetic experiments is demonstrated with sensitivities of 0.5 ppm in a single pass and 0.25 ppm in a double pass configuration (1500 K, 1.3 atm, detection bandwidth 1 MHz, 15 cm shock tube diameter). This corresponds to a minimum detectable absorption of 0.01% and 0.005%, respectively, which represents an improvement of more than a factor of 20 when compared to conventional laser absorption detection. The feasibility of quantitative absolute concentration measurements is demonstrated using CH3NH2 as a precursor for the preparation of known NH2 concentrations. The uncertainty for absolute concentration measurements is estimated to be ±10% if a suitable precursor for direct shock tube calibration measurements can be used, and ±15% if an alternative calibration scheme based on the detection of the signal generated by a scanning etalon in reflective mode is used. FM detection has been applied to determine the rate coefficient of the thermal decomposition of CH3NH2: CH3NH2 + M CH3+NH2+M over the temperature range 1530–1975 K and at pressures near 1.3 atm. The rate coefficient was found to be: k1=8.17×1016 exp(−30710/T) (±20%) [cm3 mol−1 s−1] This is in good agreement with a recent determination using conventional laser absorption detection of the NH2 radical. © 1999 John Wiley & Sons, Inc. Int J Chem Kinet 31: 445–453, 1999

The collisional deactivation of vibrationally highly excited azulene was studied from the gas to the compressed liquid phase. Employing supercritical fluids like He, Xe, CO2, and ethane at pressures of 6–4000 bar and temperatures ≥380 K, measurements over the complete gas–liquid transition were performed. Azulene with an energy of 18 000 cm−1 was generated by laser excitation into the S1 and internal conversion to the S0*‐ground state. The subsequent loss of vibrational energy was monitored by transient absorption at the red edge of the S3←S0 absorption band near 290 nm. Transient signals were converted into energy‐time profiles using hot band absorption coefficients from shock wave experiments for calibration and accounting for solvent shifts of the spectra. Under all conditions, the decays were monoexponential. At densities below 1 mol/l, collisional deactivation rates increased linearly with fluid density. Average energies 〈ΔE〉 transferred per collision agreed with data from dilute gas phase experiments. For Xe, CO2, and C2H6, the linear relation between cooling rate and diffusion coefficient scaled collision frequencies ZD turned over to a much weaker dependence at ZD≳0.3 ps−1. Up to collision frequencies of ZD=15 ps−1 this behavior can well be rationalized by a model employing an effective collision frequency related to the finite lifetime of collision complexes.

The article contains sections titled: 1. Introduction 2. Pollutant Formation and Limitation 2.1. Carbon monoxide 2.2. Hydrocarbons 2.3. Oxides of Nitrogen (NOx) 2.4. Particulate Emissions 2.5. Carbon Dioxide 2.6. Sulfur Compounds 3. Catalytic Exhaust Aftertreatment, General Concepts 3.1. The Physical Design of the Catalytic Converter 3.2. The Washcoat 3.3. The Catalytic Material 3.4. Production of Catalysts 4. Catalytic Aftertreatment of Stoichiometric Exhaust Gas 4.1. Three-Way Catalysis 4.2. Oxygen Storage in Three-Way Catalysts 4.3. Precious Metals in Three-Way Catalysis 5. Exhaust Aftertreatment for Diesel Vehicles 5.1. The Diesel Oxidation Catalyst 5.2. The Particulate Filter 5.3. NOx Treatment of Oxygen-Rich Exhaust 5.3.1. HC-DeNOx 5.3.2. The NOx Adsorber Catalyst 5.3.3. Selective Catalytic Reduction (SCR) with Ammonia 6. Exhaust Aftertreatment for Lean-Burn Gasoline Engines 7. Conclusion and Outlook

CH radicals are important precursors in prompt-NO formation and they also play a role in NO removal via reburning. Little information exists, however, on the absolute concentrations of CH during fuel-rich methane oxidation and their dependence on NO concentrations. Here the authors report results of CH-radical measurements for such conditions, using a shock tube in combination with a sensitive CH absorption diagnostic. These measurements provide useful data for the further development of large-scale reaction mechanisms, such as the recently developed Gas Research Institute methane oxidation and NO reburning mechanism (GRI-Mech 2.11). CH-radical profiles were measured at high temperatures for fuel-rich methane oxidation with and without addition of NO. Comparison of these data with predictions based on the GRI-Mech 2.11 mechanism shows significant deviations, suggesting the need for additional high-temperature rate data on important CH reactions and for further optimization of the detailed mechanism.

Several diesel passenger car manufacturers in the European Union recently announced the future use of catalyzed diesel particulate filter systems on their vehicles. The major technical challenge is the periodical regeneration of the filters loaded with the retained diesel particulates. In order to promote filter regeneration, catalytic activation of the accumulated soot is advantageous. Therefore, the first serial application of diesel particulate filter systems uses catalytically active fuel additives. These systems were introduced about four years ago. Since that time, other systems, using a dedicated catalytically activated diesel particulate filter combined with an upstream diesel oxidation catalyst, have been introduced as well. This allows filter regeneration without extra fuel additives. In the past, adding catalytic coating to a filter substrate has often resulted in increasing the pressure drop over the filter to an unacceptable level. Using a newly developed, dedicated processing technology, coatings with outstanding catalytic functions even on standard filter substrates can be applied without a severe backpressure increase. This so-called "macrowashcoating" technology enables the possibility to integrate the diesel oxidation catalyst's functions directly in the filter substrate. The HC and CO oxidation performance of a filter with integrated oxidation catalyst functionality is compared to a conventional flow-through monolith by computer simulation and experiment. Simulation and experiment both show a systematic advantage of the catalyst filter over the conventional monolith configuration. Computer simulation provides some explanation for this observation. Finally, the technology's suitability for application to passenger cars is proven by intensive engine bench and vehicle tests.