Jan J. Lerou

Chemical reaction engineering : A multiscale approach to a multiobjective task

Abstract A strategy for synthesizing a reaction system to meet various business and technical objectives is outlined. It replaces the conventional paradigm — the sequential evolution of the microreactor, bench-scale reactor, pilot plant reactor and commercial reactor. Starting with a reaction proposed for commercialization, this approach examines the synthesis problem from the viewpoint of the plant, the reactor, hydrodynamics, transport phenomena, catalyst design, reaction chemistry, and molecular modeling concurrently. Some of the design issues, methods and tools for each of the above areas are discussed.

Discovering the role of Au and KOAc in the catalysis of vinyl acetate synthesis

Publisher Summary One of the most studied applications of Catalytic Membrane Reactors (CMRs) is the dehydrogenation of alkanes. For this reaction, in conventional reactors and under classical conditions, the conversion is controlled by thermodynamics and high temperatures are required leading to a rapid catalyst deactivation and expensive operative costs. In a CMR, the selective removal of hydrogen from the reaction zone through a perm selective membrane will favor the conversion and then allow higher olefin yields when compared to conventional (non-membrane) reactors. Following this principle, when using a membrane, lower temperatures can be used leading to a longer catalyst lifetime and energy and cost saving. However, such a membrane should be stable at high temperature (ca 500oC) highly perm selective (loss of reactant should be of course avoided) and permeable enough (the permeation rate should be in the same range as the reaction rate). Dense Pd-based membranes have been first used for CMRs applications. They are highly selective for H 2 permeation but are expensive, sensitive to ageing and poisoning, and are strongly limited by their low permeabilities. This chapter describes the morphological and transport properties of a composite zeolite (silicalite)—alumina membrane. Results in CMRs applications are also briefly given in the chapter.

First-principles-based kinetic Monte Carlo simulation of nitric oxide decomposition over Pt and Rh surfaces under lean-burn conditions

First-principles-based kinetic Monte Carlo simulation was used to track the elementary surface transformations involved in the catalytic decomposition of NO over Pt(100) and Rh(100) surfaces under lean-burn operating conditions. Density functional theory (DFT) calculations were carried out to establish the structure and energetics for all reactants, intermediates and products over Pt(100) and Rh(100). Lateral interactions which arise from neighbouring adsorbates were calculated by examining changes in the binding energies as a function of coverage and different coadsorbed configurations. These data were fitted to a bond order conservation (BOC) model which is subsequently used to establish the effects of coverage within the simulation. The intrinsic activation barriers for all the elementary reaction steps in the proposed mechanism of NO reduction over Pt(100) were calculated by using DFT. These values are corrected for coverage effects by using the parametrized BOC model internally within the simulation. This enables a site-explicit kinetic Monte Carlo simulation that can follow the kinetics of NO decomposition over Pt(100) and Rh(100) in the presence of excess oxygen. The simulations are used here to model various experimental protocols including temperature programmed desorption as well as batch catalytic kinetics. The simulation results for the temperature programmed desorption and decomposition of NO over Pt(100) and Rh(100) under vacuum condition were found to be in very good agreement with experimental results. NO decomposition is strongly tied to the temporal number of sites that remain vacant. Experimental results show that Pt is active in the catalytic reaction of NO into N2 and NO2 under lean-burn conditions. The simulated reaction orders for NO and O2 were found to be +0.9 and −0.4 at 723 K, respectively. The simulation also indicates that there is no activity over Rh(100) since the surface becomes poisoned by oxygen.

First principle analysis of the catalytic reaction pathways in the synthesis of vinyl acetate

Abstract Acetoxylation of ethylene over supported palladium and palladium/gold is a well established commercial route for the formation of vinyl acetate. While the overall reaction chemistry for the synthesis of vinyl acetate was uncovered some thirty years ago (Eur. Chem. News 1967; World Pet. Cong. Proc. 1968, U.S. Patent 1967 & 1977), the active catalytic surface ensembles, key reaction intermediates, and mechanism are still poorly understood. Issues such as the oxidation state of the active centers (Pd 0 vs Pd 2+ ), particle ensemble size (small clusters versus large particles, rate-determining elementary steps, secondary decomposition routes, and the structural and/or electronic role of Au, have yet to be resolved. Herein, we employ first-principle quantum chemical techniques to model a series of proposed elementary steps representative of vinyl acetate synthesis. Calculations using palladium and oxidized palladium particles of varying size provide a fundamental understanding of the elementary physicochemical steps in the oxidative coupling of ethylene and acetic acid in route to the formation of vinyl acetate.

Gas holdup in a trayed cold-flow bubble column

An experimental study was performed to investigate the effect of sieve trays on the time-averaged gas holdup profiles and the overall gas holdup in a cold-flow bubble column that was scaled-down from a commercial unit. γ-ray computed tomography (CT) was used to scan the column at several axial locations in the presence and absence of trays from which the local variation of the gas holdup was extracted. The overall gas holdup was also determined using the same configuration by comparing the expanded and static liquid heights. Air and water were used as the gas-liquid system. The superficial gas and liquid velocities were selected to span the range of the commercial system using gas spargers having multiple lateral distributors that were also scaled-down from the commercial design. To investigate the impact of sparger hole density on the local and overall gas holdup, two difference sparger designs were used in which the hole density per lateral was varied. The gas hole velocity was maintained constant at ca. 245 m/s, which approached that used in the commercial reactor. It is shown that the local gas holdup determined by CT is generally higher in the tray down comer region and exhibits an asymmetric pattern when trays are present. The use of increased sparger hole density at a constant gas superficial velocity leads to steeper gradient in the gas holdup near the column centerline and a higher overall gas holdup. These findings suggest that the performance of bubble column reactors for various applications is sensitive to both sparger and tray design.

Supported liquid phase catalysis in selective oxidation

Abstract The application of supported liquid-phase catalysis (SLPC) to selective oxidation examples is discussed to provide valuable alternatives to the purely homogeneous or heterogenous processes. Specific examples discussed are the Wacker type ethylene oxidation on heterogenous catalysts and the vinyl acetate production by ethylene acetoxylation. The potential industrial application of SLPC for these oxidations and the possible extensions of this technology to other oxidation processes are also discussed.

TRANSIENT RESPONSE METHODS FOR ASSISTED DESIGN OF GAS PHASE HETEROGENEOUS CATALYSTS: EXPERIMENTAL TECHNIQUES AND MATHEMATICAL MODELING

3 L INTRODUCnON 3 A. Background.............................................................................. 3 B. Industrial Catalyst Research 5 C Role of Mathematical Models 8 D. Objectives 10 IL TRANSIENT RESPONSE METHODS 11 A. Introductory Comments . .. 11 B. Laboratory Reactor Types 12 1. Reactor Classification 12 2. Fixed-Bed Pulse Reactors 16 Vol. 9. Nas. 1-2.1993 Transient Response Methods for Assisted Design of Gas Phase Heterogeneous Catalysts: Experimental Techniques and Mathematical Modeling a. Microcataly tic Pulse Reactor 16 b. Gas Chromatographie Pulse Reactor 18 c. Temporal Analysis of Products (TAP) Reactor 21 3. Backmixed Reactors 23 a. Carberry Spinning Basket Reactor 26 b. Berty Internally Recycled Reactor 31 c. Internally Recycled Reactor for Low Pressures 31 4. Single Pellet Reactors 34 a. Wicke-Kallenback Type Single Pellet Reactors 35 b. Single Pellet Diffusion Reactor 37 c. Weisz Single Pellet Reactor 38 C Laboratory Reactor Modeling for Selected Reactor Types 41 1. Basic Concepts „ 41 2. Fixed-Bed Pulse Reactors 44 a. Microcataly tic Pulse Reactor 44 b. Gas Chromatographie Pulse Reactor SO c. Temporal Analysis of Products (TAP) Reactor 57 3. Backmixed Reactors 65 a Conventional Transient Operation 66 b. Inlet Disturbances Based on Feedback 70 4. Single Pellet Reactors 73 a. Weisz Single Pellet Reactor 73 III. SUMMARY AND CONCLUSIONS ., 78 IV. NOMENCLATURE 82 V. REFERENCES 89 P.L· Mills and JJ.Lerou Reviews in Chemical Engineering

Fast response distributed parameter fluidized bed reactor model for propylene partial oxidation using feed-forward neural network methods

Abstract The use of a neural network model (NNM) to simulate the performance of a fluidized-bed reactor for the partial oxidation of propylene to acrolein is investigated. The training set needed to generate the NNM is obtained from a two-phase cell model of the fluidized-bed where the flow patterns for the bubble and emulsion phases in each cell are assumed to be plug-flow and perfectly mixed, respectively. The intrinsic kinetics, which are taken from the literature, are based upon a single site redox type model that exhibits a nonlinear dependence on both molecular oxygen and propylene. The formation of acrolein, acetaldehyde, and total combustion products is described by a series-parallel reaction network. The fluidized bed model accounts for variable gas velocity as well as finite transport resistance between the bubble ane emulsion phases. To perform the required NNM training, output responses predicted from the cell model are first generated by using all possible combinations of eleven key input parameters varied over practical ranges of interest. The axial variation of the nine output responses is represented by a recurrent NNM. The NNM parameters are then identified using a special-purpose computer software package that implements both training and analysis of the input data and corresponding output responses. To simulate the behavior of a real reactor, the output responses are corrupted with random noise. Comparisons between the output responses obtained from the NNM trained to noisy data to those from the cell model with no noise indicate that the NNM is capable of providing filtering. Furthermore, a sensitivity analysis indicates that the NNM captures the dependence of the output variables on the input ones.

The continuously-stirred decanting reactor: Steady state and dynamic features

The steady-state and dynamic behavior of an exothermic reaction in a cooled, continuously-stirred decanting reactor (CSDR) is examined. For the case considered, two phases are fed but only one is removed, as reaction occurs in the lighter phase and the products are soluble only in the heavier, decanted phase. A number of distinguishing features of the CSDR are identified. Ignition and extinction phenomena do not occur because the heat generation rate is a constant, a consequence of complete conversion of the lighter phase reactant. The buildup of the reactive phase as the reactor temperature is decreased may make the operation infeasible. Thus, either zero or one feasible steady-state solutions exist. The thermal feedback and variable phase volumes may destabilize the unique steady state at a supercritical Hopf bifurcation point, making the CSDR a natural oscillator during nonisothermal operation.

Kinetics of the Redox Reactions of the O2: Propylene:γ-Bismuth Molybdate System: A TAP Reactor Study

Publisher Summary In an attempt to better understand the interaction of oxygen and propylene with γ-bismuth molybdate (γ-Bi 2 MoO 6 ), the approach of using a TAP (Temporal Analysis of Products) reactor to study these processes has been investigated. Previous approaches to the study of this system have involved processes that resulted in significant disruptions of the catalyst surfaces. In all cases, the redox processes involved amounts of oxidant or reductant equivalent to about a monolayer or more of coverage. In addition, the time scales of the processes examined were of the order of minutes. Several groups have reported that the reductions of γ-Bi2MoO6 by olefins proceed by different pathways. Recently, Glaeser and coworkers reported a model, based on a Raman IR study of reduced γ-Bi2MoO6, for these reductions where different oxygens of γ-Bi2MoO6 are involved in the oxidations of propylene and 1-butene.