S.T. Kolaczkowski

Mass and heat transfer effects in catalytic monolith reactors

Abstract Numerical simulations of catalytic oxidation in monolith reactors are performed in order to develop criteria for mass transfer limitation. A two-dimensional finite-element simulator previously developed is used to examine previously reported studies of propane and carbon monoxide combustion in excess oxygen. The Sherwood and Nusselt numbers computed from the two-dimensional simulation results are compared to numbers derived experimentally. The results from the simulations are much higher than results which have been reported in the literature for experimental work. Simulation results agree well with numbers obtained analytically and experimentally for non-reacting flow in circular tubes, and also with other correlations for reacting flows based on numerical work. The reason for the discrepancy between experimental and simulated results is explained. For first-order reactions, a dimensionless catalytic reaction number is proposed, which may be used to evaluate whether or not the rate is mass transfer controlled. For the oxidation of CO, multiple steady states are possible and the variation in Nusselt and Sherwood numbers under transient conditions is discussed. The influence of diffusion in a real monolith washcoat is also examined. In square monolith channels of dimension 1 mm, low effectiveness factors are obtained for temperatures above 700 K, and much of the catalyst is not utilised. It is shown that care needs to be taken in the extension of relatively low-temperature kinetic data to the elevated temperatures encountered in real operating conditions.

The palladium catalysed oxidation of methane: reaction kinetics and the effect of diffusion barriers

Abstract The combustion of methane on a palladium catalyst was examined in a monolith reactor. The rate equation was determined and showed an approximately first order dependence in methane concentration and zero order dependence on oxygen concentration. Significant inhibition by water was observed, and inhibition by carbon dioxide was negligible. At high water concentrations the order with respect to water is approximately minus one. A significant reduction in both activity and activation energy was observed above temperatures of approximately 820 K with a dry feed. Significant diffusion limitation in the washcoat was observed. The intrinsic volumetric rate constant was found to be directly proportional to the palladium loading of the washcoat. The effect on the reaction rate of layers of inert washcoat placed on top of the active catalyst was investigated. These diffusion barriers reduced the reaction rate. The reactor performance was modelled using a two-dimensional finite element single channel model that included washcoat diffusion. The effect of diffusion barriers was compared to the effect of using a less active catalyst for steady state and transient modes of operation at values of the Lewis number. At low Lewis number the diffusion barrier was effective at reducing the temperature rise at the entrance to the reactor for large inlet reactant concentration.

Hydrogen peroxide promoted wet air oxidation of phenol: influence of operating conditions and homogeneous metal catalysts

The promoted wet air oxidation of phenol has been investigated through the addition of hydrogen peroxide as a source of free radicals. The reaction has been shown to proceed in two stages, an initial fast reaction associated with hydrogen peroxide consumption and a second slower step that occurs at a rate comparable with conventional wet air oxidation. An increase in temperature has a positive effect on both stages, while oxygen partial pressure only influences the second slower stage. The influence of pH on phenol oxidation is shown to be significant with the highest efficiency achieved at very alkaline conditions when phenol is completely dissociated. The catalytic activity of homogeneous metal salts was investigated in both the presence and absence of hydrogen peroxide. The combined addition of hydrogen peroxide and a bivalent metal (ie copper, cobalt or manganese) is shown to enhance the rate of phenol removal. However, in the absence of hydrogen peroxide only copper exhibited catalytic activity. Finally, a reaction mechanism involving different radical species has been proposed. From the experimental results the apparent activation energy (96.9 ± 3.5 kJ mol−1) and pre-exponential factor (1.6 ± 0.2 1010 s−1) were calculated for hydrogen peroxide decomposition into hydroxyl radicals. © 1999 Society of Chemical Industry

A new technique to measure the effective diffusivity in a catalytic monolith washcoat

A method is described for measuring the flux of a diffusing species through a multiple cell structure cut from a catalytic monolith honeycomb. One- and two-dimensional mathematical models are used to calculate the effective diffusivity in the catalyst/washcoat layer. This method is suitable for porous monolith supports, e.g. cordierite, but it is unsuitable for metal monoliths. To illustrate the technique the diffusion of CO in nitrogen is studied using a modified form of a Wicke–Kallenbach type of diffusion cell. The inlet concentration of the diffusing component is 2.4% CO in nitrogen, and experiments are performed at ambient temperature and pressures between 106 and 150 kPa on a catalytic monolith with 62 cells cm −2 . The technique can be applied in many areas where catalytic monoliths are used, e.g. catalytic converters, catalytic combustion reactors, SCR catalysts and many other applications. The method shows good agreement with the results obtained using other methods.

Wet Air Oxidation of Phenol: Factors that May Influence Global Kinetics

Wet air oxidation is presented as a technique for removal of organic pollutants found in waste water streams. The oxidation of phenol has been achieved at moderate temperatures ( T =473 K) and pressures (3.0 MPa total pressure) with up to 95% destruction in less than 30 minutes.An increase in either oxygen concentration or temperature has a positive influence on reaction rate, with temperature following an Arrhenius dependence. The influence of pH is shown to be complex. For a 0.01 mol l −1 phenol solution significant destruction was obtained; however, when the initial pH was modified to either less than two or between neutrality and ten, practically no change in phenol concentration was observed. When oxidation is carried out in strong alkali media (pH>12) reaction rate is enhanced significantly.Addition of small quantities of hydrogen peroxide resulted in enhanced rates of oxidation even at low temperatures (373 K). When hydrogen peroxide is added, oxygen plays a negligible role in the initial reaction, only becoming significant once the hydrogen peroxide has been consumed.The nature and geometry of the reactor has been found to play an important role in free radical termination steps, with metal surfaces likely to enhance significantly decomposition of organic radicals.

Modelling catalytic combustion in monolith reactors – challenges faced

Abstract When searching for a design concept in which a catalytic combustor is utilised, or looking for areas where improvements can be made to an existing design, then mathematical modelling is an important tool. However, models are only as good as the way in which the physico-chemical processes are modelled and the quality of the physical and chemical parameters (e.g. kinetic expressions, physical properties) acquired for use in the models. When selecting a basis for a model, there are many questions that need to be asked and answered by the developer of the chemical reaction engineering model of the catalytic combustor. Many challenges arise from having to make decisions on compromises that need to be made, and in recognising the consequences of such action. Examples of such challenges are outlined and, for some, clues are offered as to where the answers may lie. The examples include challenges in: the selection of appropriate kinetic expressions, recognition of the role that intraphase diffusion may play, the choice of pressure for catalytic kinetic and pilot scale studies, the selection of heat and mass transfer correlations, and the modelling of transients.

Modelling of heat transfer in non-adiabatic monolithic reactors

Abstract A model is presented to describe radial heat transfer in a honeycomb monolithic structure under non-adiabatic, non-reacting conditions. This is believed to be the first model where the walls of the channels in the radial and tangential directions are considered to have a finite thickness. The model is solved numerically and is tested on recently published experimental data. The rigorous analysis of temperature profiles in the monolith structure facilitates future extension of the model to include homogeneous and heterogeneous reaction terms. Initial conditions at the inlet and outer walls of the monolith may readily be varied in the model. It is shown that if these are oversimplified then this will lead to inaccuracies in the prediction of temperature profiles within the monolith.

Development of combustion catalysts for monolith reactors: a consideration of transport limitations

In the search for catalysts to be utilized in a new generation of catalytic gas turbine combustors it is not unusual to hear of catalysts that appear to have worked well in a laboratory environment but do not do so when installed in a high pressure pilot-scale rig. The influence that interphase and intraphase transport limitations may have on the rate of catalytic combustion of methane in a monolith reactor (with a 1.1 mm cell size) is investigated at a Reynolds number, Re = 103 (with) P = 2bar) and at Re = 104 (with P = 15bar), for bulk gas temperatures varying from 623–873 K and at a bulk methane mole fraction of 0.02. It is shown that it is not an easy task to evaluate chemical kinetic expressions at conditions which truly represent even the first stage (e.g., at gas temperatures from 623–973 K) in a catalytic gas turbine combustor, for as reaction rates become very fast, interphase and intraphase transport limitations also become significant, e.g., when comparing reaction rates evaluated at incorrectly assigned conditions, then even at a relatively low bulk fluid temperature= 773K, errors may vary from ca. 35% to 80% for Re = 104and103, respectively. In a laboratory test environment, where experiments may be performed close to atmospheric conditions, it is also shown that it is easy to over-temperature and hence damage the catalyst, and draw false conclusions about the catalyst being too active, e.g., considering interphase resistances alone, at a bulk fluid temperature of 773 K and Re = 103 the catalyst is damaged, whereas at Re = 104 this would not have occurred. As catalyst surface temperature exceeds 800 K, intraphase diffusion may become significant and the effectiveness factor soon becomes very much less than one. Therefore, reactor scale-up should be based on calculations where the combined effects of interphase and intraphase transport processes are coupled with reactions; scale-up based on simple multipliers, e.g., those derived from residence time or gas hourly space velocity, are unlikely to work. Modelling is shown to have an important role in catalyst and system design; if applied correctly, system development costs may be reduced.

Measurement of effective diffusivity in catalyst-coated monoliths

Abstract A review is provided about techniques that have been used to evaluate the effective diffusivity of gases in catalyst/washcoat layers, as used in catalytic monoliths. The importance of making such measurements is described, in order to ensure that the choice of model for effective diffusivity can be verified, and if necessary an appropriate value of tortuosity can be back-calculated. Based on methods described in the literature, it is concluded that, where possible experiments should be performed on actual monolith structures, rather than those that have been reformed. The chromatographic technique is applied to a catalytic monolith and preliminary results of unpublished work are presented. A method of using a cut section from a catalytic monolith in a modified form of ‘Wicke–Kallenbach diffusion cell’ is also described. Examples from the patent literature are provided showing, how interest in layered catalyst systems has started to grow, illustrating how diffusion in porous layers can be exploited to develop ‘designer catalyst systems’.

Manufacture and Characterisation of Silicalite Monoliths

Multichannel monoliths containing up to 90% silicalite by weight and with cell densities up to 28 cells/cm2, wall thicknesses down to 0.6 mm and an overall diameter of 40 mm have been prepared from silicalite powder and sodium bentonite (as a binder) by the unit operations of paste preparation, extrusion, drying and firing. The manufactured monoliths, which show good strength, retain the crystal structure and micropore size of the source silicalite powder, and adsorption measurements made by using a dynamic flow system confirm that the monoliths exhibit an equilibrium performance broadly similar to that of commercial silicalite pellets. In addition, the manufactured monoliths possess a higher macroporosity than the commercial pellets. Regeneration of the monoliths was found to be possible at mildly increased temperature. These features augur well for the recovery and/or separation of organic compounds in simple pressure swing and thermal swing processes.