Copyright © 2008, 1997, 1984, 1973, 1963, 1950, 1941, 1934 by The McGraw-Hill Companies, Inc All rights reserved Manufactured in the United States of America Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written permission of the publisher 0-07-154226-4 The material in this eBook also appears in the print version of this title: 0-07-151142-3 All trademarks are trademarks of their respective owners Rather than put a trademark symbol after every occurrence of a trademarked name, we use names in an editorial fashion only, and to the benefit of the trademark owner, with no intention of infringement of the trademark Where such designations appear in this book, they have been printed with initial caps McGraw-Hill eBooks are available at special quantity discounts to use as premiums and sales promotions, or for use in corporate training programs For more information, please contact George Hoare, Special Sales, at george_hoare@mcgraw-hill.com or (212) 904-4069 TERMS OF USE This is a copyrighted work and The McGraw-Hill Companies, Inc (“McGraw-Hill”) and its licensors reserve all rights in and to the work Use of this work is subject to these terms Except as permitted under the Copyright Act of 1976 and the right to store and retrieve one copy of the work, you may not decompile, disassemble, reverse engineer, reproduce, modify, create derivative works based upon, transmit, distribute, disseminate, sell, publish or sublicense the work or any part of it without McGraw-Hill’s prior consent You may use the work for your own noncommercial and personal use; any other use of the work is strictly prohibited Your right to use the work may be terminated if you fail to comply with these terms THE WORK IS PROVIDED “AS IS.” McGRAW-HILL AND ITS LICENSORS MAKE NO GUARANTEES OR WARRANTIES AS TO THE ACCURACY, ADEQUACY OR COMPLETENESS OF OR RESULTS TO BE OBTAINED FROM USING THE WORK, INCLUDING ANY INFORMATION THAT CAN BE ACCESSED THROUGH THE WORK VIA HYPERLINK OR OTHERWISE, AND EXPRESSLY DISCLAIM ANY WARRANTY, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE McGraw-Hill and its licensors not warrant or guarantee that the functions contained in the work will meet your requirements or that its operation will be uninterrupted or error free Neither McGraw-Hill nor its licensors shall be liable to you or anyone else for any inaccuracy, error or omission, regardless of cause, in the work or for any damages resulting therefrom McGraw-Hill has no responsibility for the content of any information accessed through the work Under no circumstances shall McGraw-Hill and/or its licensors be liable for any indirect, incidental, special, punitive, consequential or similar damages that result from the use of or inability to use the work, even if any of them has been advised of the possibility of such damages This limitation of liability shall apply to any claim or cause whatsoever whether such claim or cause arises in contract, tort or otherwise DOI: 10.1036/0071511423 This page intentionally left blank Section 19 Reactors* Carmo J Pereira, Ph.D., MBA DuPont Fellow, DuPont Engineering Research and Technology, E I du Pont de Nemours and Company; Fellow, American Institute of Chemical Engineers Tiberiu M Leib, Ph.D Principal Consultant, DuPont Engineering Research and Technology, E I du Pont de Nemours and Company; Fellow, American Institute of Chemical Engineers REACTOR CONCEPTS Reactor Types Classification by Mode of Operation Classification by End Use Classification by Phase Reactor Modeling Modeling Considerations Chemical Kinetics Pressure Drop, Mass and Heat Transfer Reactor Dynamics Reactor Models 19-4 19-4 19-7 19-7 19-7 19-7 19-9 19-10 19-11 19-13 RESIDENCE TIME DISTRIBUTION AND MIXING Tracers Inputs Types of Responses Reactor Tracer Responses Understanding Reactor Flow Patterns Connecting RTD to Conversion Segregated Flow Early versus Late Mixing—Maximum Mixedness Reaction and Mixing Times 19-14 19-15 19-15 19-15 19-16 19-17 19-18 19-18 19-20 SINGLE-PHASE REACTORS Liquid Phase Homogeneous Catalysis Gas Phase Supercritical Conditions Polymerization Reactors 19-20 19-20 19-21 19-21 19-21 FLUID-SOLID REACTORS Heterogeneous Catalysts Catalytic Reactors Wire Gauzes Monolith Catalysts Fixed Beds Moving Beds Fluidized Beds Slurry Reactors Transport Reactors Multifunctional Reactors Noncatalytic Reactors Rotary Kilns Vertical Kilns 19-25 19-27 19-27 19-27 19-30 19-33 19-33 19-36 19-36 19-36 19-36 19-36 19-36 *The contributions of Stanley M Walas, Ph.D., Professor Emeritus, Department of Chemical and Petroleum Engineering, University of Kansas (Fellow, American Institute of Chemical Engineers), author of this section in the seventh edition, are acknowledged The authors of the present section would like to thank Dennie T Mah, M.S.Ch.E., Senior Consultant, DuPont Engineering Research and Technology, E I du Pont de Nemours and Company (Senior Member, American Institute of Chemical Engineers; Member, Industrial Electrolysis and Electrochemical Engineering; Member, The Electrochemical Society), for his contributions to the “Electrochemical Reactors” subsection; and John Villadsen, Ph.D., Senior Professor, Department of Chemical Engineering, Technical University of Denmark, for his contributions to the “Bioreactors” subsection We acknowledge comments from Peter Harriott, Ph.D., Fred H Rhodes Professor of Chemical Engineering (retired), School of Chemical and Biomolecular Engineering, Cornell University, on our original outline and on the subject of heat transfer in packed-bed reactors The authors also are grateful to the following colleagues for reading the manuscript and for thoughtful comments: Thomas R Keane, DuPont Fellow (retired), DuPont Engineering Research and Technology, E I du Pont de Nemours and Company (Senior Member, American Institute of Chemical Engineers); Güray Tosun, Ph.D., Senior Consultant, DuPont Engineering Research and Technology, E I du Pont de Nemours and Company (Senior Member, American Institute of Chemical Engineers); and Nitin H Kolhapure, Ph.D., Senior Consulting Engineer, DuPont Engineering Research and Technology, E I du Pont de Nemours and Company (Senior Member, American Institute of Chemical Engineers) 19-1 Copyright © 2008, 1997, 1984, 1973, 1963, 1950, 1941, 1934 by The McGraw-Hill Companies, Inc Click here for terms of use 19-2 REACTORS FLUID-FLUID REACTORS Gas-Liquid Reactors Liquid-Liquid Reactors Reactor Types Agitated Stirred Tanks Bubble Columns Tubular Reactors Packed, Tray, and Spray Towers 19-38 19-41 19-42 19-42 19-44 19-46 19-46 SOLIDS REACTORS Thermal Decomposition Solid-Solid Reactions Self-Propagating High-Temperature Synthesis (SHS) 19-48 19-48 19-49 MULTIPHASE REACTORS Bioreactors Electrochemical Reactors Reactor Types Agitated Slurry Reactors Slurry Bubble Column Reactors Fluidized Gas-Liquid-Solid Reactors Trickle Bed Reactors Packed Bubble Columns (Cocurrent Upflow) Countercurrent Flow SOME CASE STUDIES 19-49 19-50 19-53 19-53 19-56 19-57 19-57 19-60 19-60 REACTORS 19-3 Nomenclature and Units In this section, the concentration is represented by C Mass balance accounting in terms of the number of moles and the fractional conversion is discussed in Sec and can be very useful The rate of reaction is r; the flow rate in moles is Na; the volumetric flow rate is V′; reactor volume is Vr Several equations are presented without specification of units Use of any consistent unit set is appropriate Following is a listing of typical nomenclature expressed in SI and U.S Customary System units Specific definitions and units are stated at the place of application in this section Symbol a Ak C C0 cp CSTR d D Deff De E E(t) E(tr) fa F(t) h H He ∆Hr k km L m n Nu N Ha pa P Pe PFR q Q r R Re Sc Sh t⎯ t tr T TFR u u(t) U v vij Definition Surface area per volume Heat-transfer area Concentration of substance Initial mean concentration Heat capacity at constant pressure Ideal continuous stirred tank reactor Diameter Diameter, diffusivity Effective diffusion coefficient Effective dispersion coefficient Activation energy Residence time distribution Normalized residence time distribution Fraction of A remaining unconverted, Ca /Ca0 or na/a0 Age function of tracer Heat-transfer coefficient Height of tank Henry constant Heat of reaction Specific rate constant for first-order reaction Mass-transfer coefficient Length of path in reactor Magnitude of impulse Number of stages in a CSTR battery or parameter of Erlang or gamma distribution Nusselt number Speed of agitator Hatta number Partial pressure of substance A Total pressure Peclet number for dispersion Plug flow reactor Heat flux, reaction order, or impeller-induced flow Volumetric flow rate Rate of reaction per unit volume, radius Radius Reynolds number Schmidt number Sherwood number Time Mean residence time ⎯ Reduced time, tրt Temperature Tubular flow reactor Linear velocity Unit step input Overall heat-transfer coefficient Volumetric flow rate during semibatch operation Stoichiometric coefficients SI units U.S Customary System Units Symbol Definition SI units 1/m m2 kg⋅molրm3 kg⋅molրm3 kJր(kg⋅K) 1/ft ft2 lb⋅molրft3 lb⋅molրft3 Btuր(lbm⋅°F) V′ Vr w x Volumetric flow rate Volume of reactor Catalyst loading Axial position in a reactor, conversion m ft α m2/s m2/s kJ/(kg⋅mol) ft2/s ft2/s Btu/(lb⋅mol) β λ Λ(t) µ ν ρ σ2(τ) σ2(tr) ξ τ φ φσ φε⎯ Fraction of feed that bypasses reactor Fraction of reactor volume that is stagnant, Prater number Unit impulse input, Dirac delta function Distance or film thickness Void fraction in a packed bed, particle porosity Effectiveness factor of porous catalyst Thermal conductivity Intensity function Viscosity Kinematic viscosity, µրρ Density Variance Normalized variance Fractional conversion Tortuosity Thiele modulus Shape factor Local rate of energy dissipation a b c cir d f G i L ma me mi p r R s t u w δ Agitator, axial, species A Bed, species B Critical value, catalyst, coolant, continuous phase Circulation Dispersed phase Fluid, feed Gas phase Interface Liquid Macro Meso Micro Pellet Reaction, reduced Reactor Surface Tank Step function Wall Inlet Delta function Initial condition m3/s m3 U.S Customary System Units ft3/s ft3 Greek letters δ(τ) δL ε η kJր(s⋅m2 ⋅°C) m Pa⋅m3ր(kg⋅mol) kJր(kg⋅mol) 1/s Btuր(h⋅ft2 ⋅°F) ft atm⋅ft3ր(lb⋅mol) Btuր(lb⋅mol) 1/s m/s m kg⋅mol ft/s ft lb⋅mol rpm rpm Pa Pa psi psi m3/s ft3/s m ft s s s s °C °F m/s ft/s kJր(s⋅m2 ⋅°C) Btuր(h⋅ft2 ⋅°F) m /s m ft kJր(s⋅m⋅°C) Btuր(h⋅ft⋅°F) Pa⋅s m2/s kg/m3 lbmր(ft⋅s) ft2/s lbm/ft3 Subscripts Superscripts ft3/s GENERAL REFERENCES: The General References listed in Sec are applicable for Sec 19 References to specific topics are made throughout this section A chemical reactor is a controlled volume in which a chemical reaction can occur in a safe and controllable manner A reactor typically is a piece of equipment; however, it can also be a product (such as a coating or a protective film) One or more reactants may react together at a desired set of operating conditions, such as temperature and pressure There may be a need for appropriate mixing, control of flow distribution and residence time, contacting between the reactants (sometimes in the presence of a catalyst or biocatalyst), removal (or addition) of heat, and integration of the reactor with the rest of the downstream process Depending on the nature of the rate-limiting step(s), a reactor may serve primarily as a holding tank, a heat exchanger, or a mass-transfer device Chemical reactions generate desired products and also by-products that have to be separated and disposed A successful commercial unit is an economic balance of all these factors A variety of reactor types are used in the chemical, petrochemical, and pharmaceutical industries Some of these reactors are listed in Table 19-1 They include gas, liquid, or multiphase batch reactors, stirred tank reactors, and tubular rectors There are a number of textbooks on chemical reaction engineering Davis and Davis (Fundamentals of Chemical Reaction Engineering, McGraw-Hill, 2003) provide a lucid discussion of kinetics and principles A more comprehensive treatment together with access to CD-ROM and web resources is in the text by Fogler (Elements of Chemical Reaction Engineering, 3d ed., Prentice-Hall, 1999) A chemistry-oriented perspective is provided by Schmidt (The Engineering of Chemical Reactions, Oxford University Press, 1999) The book by Froment and Bischoff provides a thorough discussion of reactor analysis and design A practical manual on reactor design and scale-up is by Harriott (Chemical Reactor Design, Marcel Dekker, 2003) Levenspiel (Chemical Reaction Engineering, 3d ed., Wiley, 1999) was among the first to present a phenomenological discussion of fundamentals The mathematical underpinnings of reactor modeling are covered by Bird et al (Transport Phenomena, 2d ed., Wiley, 2002) This section contains a number of illustrations and sketches from books by Walas (Chemical Process Equipment Selection and Design, Butterworths, 1990) and Ullmann [Encyclopedia of Chemical Technology (in German), vol 3, Verlag Chemie, 1973, pp 321–518] Mathematical models may be used to design reactors and analyze their performance Detailed models have mainly been developed for large-scale commercial processes A number of software tools are now available This chapter will discuss some of the reactors used commercially together with how mathematical models may be used For additional details, a number of books on reactor analysis cited in this section are available The discussion will indicate that logical choices aimed at maximizing reaction rate and selectivity for a given set of kinetics can lead to rational reactor selection While there has been progress in recent years, reactor design and modeling are largely an art REACTOR CONCEPTS Since a primary purpose of a reactor is to provide desirable conditions for reaction, the reaction rate per unit volume of reactor is important in analyzing or sizing a reactor For a given production rate, it determines the reactor volume required to effect the desired transformation The residence time in a reactor is inversely related to the term space velocity (defined as volumetric feed rate/reactor volume) The fraction of reactants converted to products and by-products is the conversion The fraction of desired product in the material converted on a molar basis is referred to as selectivity The product of conversion and the fractional selectivity provides a measure of the fraction of reactants converted to product, known as yield The product yield provides a direct measure of the level of (atom) utilization of the raw materials and may be an important component of operating cost A measure of reactor utilization called space time yield (STY) is the ratio of product generation rate to reactor volume When a catalyst is used, the reactor has to make product without major process interruptions The catalyst may be homogeneous or heterogeneous, and the latter can be a living biological cell A key aspect of catalyst performance is the durability of the active site Since a chemical or biochemical process has a number of unit operations around the reactor, it is often beneficial to minimize the variability of reactant and product flows This typically means that the reactor is operated at a steady state Interactions between kinetics, fluid flow, transport resistances, and heat effects sometimes result in multiple steady states and transient (dynamic) behavior Reactor dynamics can also result in runaway behavior, where reactor temperature continues to increase until the reactants are depleted, or wrong-way behavior, where reducing inlet temperature (or reactant flow rate) can result in temperature increases farther downstream and a possible runaway Since such behavior can result in large perturbations in the process and possibly safety issues, a reactor control strategy has to be implemented The need to operate safely under all conditions calls for a thorough analysis to ensure that the reactor is inherently safe and that all possible unsafe outcomes have been considered and addressed Since various solvents may be used in chemical processes and reactors generate both products and by-products, solvent and by-product emissions can cause emission and environmental footprint issues that must be considered Reactor design is often discussed in terms of independent and dependent variables Independent variables are choices such as reactor type and internals, catalyst type, inlet temperature, pressure, and fresh feed composition Dependent variables result from independent variable selection They may be constrained or unconstrained Con19-4 strained dependent variables often include pressure drop (limited due to compressor cost), feed composition (dictated by the composition of the recycle streams), temperature rise (or decline), and local and effluent composition The reactor design problem is often aimed at optimizing independent variables (within constraints) to maximize an objective function (such as conversion and selectivity) Since the reactor feed may contain inert species (e.g., nitrogen and solvents) and since there may be unconverted feed and by-products in the reactor effluent, a number of unit operations (distillation, filtration, etc.) may be required to produce the desired product(s) In practice, the flow of mass and energy through the process is captured by a process flow sheet The flow sheet may require recycle (of unconverted feed, solvents, etc.) and purging that may affect reaction chemistry Reactor design and operation influence the process and vice versa REACTOR TYPES Reactors may be classified according to the mode of operation, the end-use application, the number of phases present, whether (or not) a catalyst is used, and whether some other function (e.g., heat transfer, separations, etc.) is conducted in addition to the reaction Classification by Mode of Operation Batch Reactors A “batch” of reactants is introduced into the reactor operated at the desired conditions until the target conversion is reached Batch reactors are typically tanks in which stirring of the reactants is achieved using internal impellers, gas bubbles, or a pumparound loop where a fraction of the reactants is removed and externally recirculated back to the reactor Temperature is regulated via internal cooling surfaces (such as coils or tubes), jackets, reflux condensers, or pump-around loop that passes through an exchanger Batch processes are suited to small production rates, to long reaction times, to achieve desired selectivity, and for flexibility in campaigning different products Continuous Reactors Reactants are added and products removed continuously at a constant mass flow rate Large daily production rates are mostly conducted in continuous equipment A continuous stirred tank reactor (CSTR) is a vessel to which reactants are added and products removed while the contents within the vessel are vigorously stirred using internal agitation or by internally (or externally) recycling the contents CSTRs may be employed in series or in parallel An approach to employing CSTRs in series is to have a large TABLE 19-1 Residence Times and/or Space Velocities in Industrial Chemical Reactors* Product (raw materials) Type Reactor phase Catalyst Acetaldehyde (ethylene, air) Acetic anhydride (acetic acid) Acetone (i-propanol) Acrolein (formaldehyde, acetaldehyde) Acrylonitrile (air, propylene, ammonia) Adipic acid (nitration of cyclohexanol) Adiponitrile (adipic acid) FB TO MT FL FL TO FB L L LG G G L G Alkylate (i-C4, butenes) Alkylate (i-C4, butenes) Allyl chloride (propylene, Cl2) Ammonia (H2, N2) CST CST TO FB L L G G Cu and Pd chlorides Triethylphosphate Ni MnO, silica gel Bi phosphomolybdate Co naphthenate H3BO3 H3PO4 H2SO4 HF NA Fe Ammonia (H2, N2) FB G Fe Ammonia oxidation Aniline (nitrobenzene, H2) Aniline (nitrobenzene, H2) Aspirin (salicylic acid, acetic anhydride) Benzene (toluene) Flame B FB B TU G L G L G Benzene (toluene) Benzoic acid (toluene, air) Butadiene (butane) Butadiene (1-butene) TU SCST FB FB Butadiene sulfone (butadiene, SO2) i-Butane (n-butane) i-Butane (n-butane) Butanols (propylene hydroformylation) T, °C P, atm 50–100 700–800 300 280–320 400 125–160 370–410 0.3 1 4–20 5–10 25–38 500 450 2–3 8–11 150 450 225 Pt gauze FeCl2 in H2O Cu on silica None None 900 95–100 250–300 90 740 1 38 G LG G G None None Cr2O3, Al2O3 None 650 125–175 750 600 CST FB FB FB L L L L 34 40–120 370–500 150–200 Butanols (propylene hydroformylation) Calcium stearate Caprolactam (cyclohexane oxime) FB B CST L L L Carbon disulfide (methane, sulfur) Carbon monoxide oxidation (shift) Furn TU G G t-Butyl catechol AlCl3 on bauxite Ni PH3-modified Co carbonyls Fe pentacarbonyl None Polyphosphoric acid None Cu-Zn or Fe2O3 Portland cement Chloral (Cl2, acetaldehyde) Chlorobenzenes (benzene, Cl2) Coking, delayed (heater) Coking, delayed (drum, 100 ft max height) Kiln CST SCST TU B S LG LG LG LG None Fe None None Cracking, fluid catalytic Cracking, hydro (gas oils) Cracking (visbreaking residual oils) Cumene (benzene, propylene) Cumene hydroperoxide (cumene, air) Cyclohexane (benzene, H2) Cyclohexanol (cyclohexane, air) Cyclohexanone (cyclohexanol) Cyclohexanone (cyclohexanol) Cyclopentadiene (dicyclopentadiene) DDT (chloral, chlorobenzene) Dextrose (starch) Dextrose (starch) Dibutylphthalate (phthalic anhydride, butanol) Diethylketone (ethylene, CO) Dimethylsulfide (methanol, CS2) Diphenyl (benzene) Riser FB TU FB CST FB SCST CST MT TJ B CST CST B TO FB MT G LG LG G L G LG L G G L L L L L G G Dodecylbenzene (benzene, propylene tetramer) Ethanol (ethylene, H2O) Ethyl acetate (ethanol, acetic acid) Ethyl chloride (ethylene, HCl) Ethylene (ethane) CST FB TU, CST TO TU Ethylene (naphtha) Ethylene, propylene chlorohydrins (Cl2, H2O) Ethylene glycol (ethylene oxide, H2O) Ethylene glycol (ethylene oxide, H2O) Ethylene oxide (ethylene, air) Ethyl ether (ethanol) Fatty alcohols (coconut oil) Formaldehyde (methanol, air) Glycerol (allyl alcohol, H2O2) TU CST TO TO FL FB B FB CST Residence time or space velocity Source and page† 12 18–36 20–50 1,000 6–40 0.25–5 s 2.5 h 0.6 s 4.3 s 2h 3.5–5 s 350–500 GHSV 5–40 5–25 0.3–1.5 s 28 s 7,800 GHSV 33 s 10,000 GHSV 0.0026 s 8h 0.5–100 s >1 h 48 s 815 GHSV 128 s 0.2–2 h 0.1–1 s 0.001 s 34,000 GHSV 0.2 LHSV 0.5–1 LHSV 1–6 WHSV 100 g L⋅h [1] 192 [4] 239, [7] 683 [4] 239 [1] 373 110 180 80–110 10 1h 1–2 h 0.25–2 h [7] 125 [7] 135 [1] 73, [7] 139 500–700 390–220 26 [1] 322, [7] 144 [6] 44 1,400–1,700 20–90 40 490–500 500–440 1 15–4 Zeolite Ni, SiO2, Al2O3 None H3PO4 Metal porphyrins Ni on Al2O3 None N.A Cu on pumice None Oleum H2SO4 Enzyme H2SO4 Co oleate Al2O3 None 520–540 350–420 470–495 260 95–120 150–250 185–200 107 250–350 220–300 0–15 165 60 150–200 150–300 375–535 730 2–3 100–150 10–30 35 2–15 25–55 48 1 1–2 1 1 200–500 L G L G G AlCl3 H3PO4 H2SO4 ZnCl2 None 15–20 300 100 150–250 860 82 6–20 G LG LG LG G G L G L None None 1% H2SO4 None Ag WO3 Na, solvent Ag gauze H2WO4 550–750 30–40 50–70 195 270–290 120–375 142 450–600 40–60 2–7 3–10 13 2–100 1 1.0 s 4.5 s 7,000 GHSV 10 h 140 h 24 h 250 s 0.3–0.5 ft/s vapor 2–4 s 1–2 LHSV 450 s, LHSV 23 LHSV 1–3 h 0.75–2 LHSV 2–10 0.75 h 4–12 s 0.1–0.5 LHSV 8h 20 100 1–3 h 0.1–10 h 150 GHSV 0.6 s 3.3 LHSV 1–30 1,800 GHSV 0.5–0.8 LHSV 2s 1.03 s 1,880 GHSV 0.5–3 s 0.5–5 30 1h 1s 30 2h 0.01 s 3h 35 9–13 0.25 [2] 1, [7] [2] [1] 314 [1] 384, [7] 33 [3] 684, [2] 47 [2] 51, [7] 49 [1] 152, [7] 52 [4] 223 [4] 223 [1] 416, [7] 67 [6] 61 [6] 61 [6] 115 [1] 289 [7] 82 [7] 89 [6] 36, [9] 109 [1] 183, [7] 98 [7] 101 [7] 118 [3] 572 [11] [7] 158 [1] 122 [1] 10 [1] 10 (14) 353 [11] [11] [11] [7] 191 [7] 201 [7] 203 [8] (1963) [8] (1963) [7] 212 [7] 233 [8] (1951) [7] 217 [7] 227 [7] 243 [7] 266 [7] 275, [8] (1938) [7] 283 [2] 356, [7] 297 [10] 45, 52, 58 [7] 305 [3] 411, [6] 13 [7] 254 [7] 310, 580 [2] 398 [2] 398 [2] 409, [7] 322 [7] 326 [8] (1953) [2] 423 [7] 347 19-5 TABLE 19-1 Residence Times and/or Space Velocities in Industrial Chemical Reactors (Concluded) Product (raw materials) Type Reactor phase Catalyst T, °C P, atm Hydrogen (methane, steam) MT G Ni 790 13 Hydrodesulfurization of naphtha TO LG Co-MO 315–500 20–70 Hydrogenation of cottonseed oil Isoprene (i-butene, formaldehyde) Maleic anhydride (butenes, air) Melamine (urea) Methanol (CO, H2) Methanol (CO, H2) o-Methyl benzoic acid (xylene, air) SCST FB FL B FB FB CST LG G G L G G L Ni HCl, silica gel V2O5 None ZnO, Cr2O3 ZnO, Cr2O3 None 130 250–350 300–450 340–400 350–400 350–400 160 2–10 40–150 340 254 14 Methyl chloride (methanol, Cl2) Methyl ethyl ketone (2-butanol) Methyl ethyl ketone (2-butanol) FB FB FB G G G Al2O3 gel ZnO Brass spheres 340–350 425–475 450 2–4 Nitrobenzene (benzene, HNO3) Nitromethane (methane, HNO3) Nylon-6 (caprolactam) Phenol (cumene hydroperoxide) Phenol (chlorobenzene, steam) Phosgene (CO, Cl2) CST TO TU CST FB MT L G L L G G H2SO4 None Na SO2 Cu, Ca phosphate Activated carbon 45–95 450–700 260 45–65 430–450 50 5–40 2–3 1–2 5–10 Phthalic anhydride (o-xylene, air) Phthalic anhydride (naphthalene, air) Polycarbonate resin (bisphenol-A, phosgene) MT FL B G G L 350 350 30–40 1 Polyethylene Polyethylene Polypropylene Polyvinyl chloride i-Propanol (propylene, H2O) Propionitrile (propylene, NH3) Reforming of naphtha (H2/hydrocarbon = 6) TU TU TO B TO TU FB L L L L L G G V2O5 V2O5 Benzyltriethylammonium chloride Organic peroxides Cr2O3, Al2O3, SiO2 R2AlCl, TiCl4 Organic peroxides H2SO4 CoO Pt 180–200 70–200 15–65 60 70–110 350–425 490 Starch (corn, H2O) Styrene (ethylbenzene) B MT L G SO2 Metal oxides 25–60 600–650 1 Sulfur dioxide oxidation FB G V2O5 475 t-Butyl methacrylate (methacrylic acid, i-butene) Thiophene (butane, S) Toluene diisocyanate (toluene diamine, phosgene) Toluene diamine (dinitrotoluene, H2) Tricresyl phosphate (cresyl, POCl3) Vinyl chloride (ethylene, Cl2) Aldehydes (diisobutene, CO) Allyl alcohol (propylene oxide) Automobile exhaust Gasoline (methanol) Hydrogen cyanide (NH3, CH4) Isoprene, polymer NOx pollutant (with NH3) Automobile emission control Nitrogen oxide emission control CST TU B B TO FL CST FB FB FB FB B FB M M L G LG LG L G LG G G G G L G G G H2SO4 None None Pd MgCl2 None Co Carbonyl Li phosphate Pt-Pd: 1–2 g/unit Zeolite Pt-Rh Al(i-Bu)3⋅TiCl4 V2O5⋅TiO2 Pt/Rh/Pd/Al2O3 V2O5-WO3/TiO2 25 600–700 200–210 80 150–300 450–550 150 250 400–600+ 400 1150 20–50 300–400 350–500 300–400 1 2–10 200 1 20 1–5 1–10 1 Carbon monoxide and hydrocarbon emission control Ozone control from aircraft cabins Vinyl acetate (ethylene + CO) M G Pt-Pd/Al2O3 500–600 M MT G LG Pd/Al2O3 Cu-Pd 130–170 130 30 1,000–1,700 20–50 10–20 10 2–14 70–200 30–35 Residence time or space velocity 5.4 s 3,000 GHSV 1.5–8 LHSV 125 WHSV 6h 1h 0.1–5 s 5–60 5,000 GHSV 28,000 GHSV 0.32 h 3.1 LHSV 275 GHSV 0.5–10 2.1 s 13 LHSV 3–40 0.07–0.35 s 12 h 15 WHSV 16 s 900 GHSV 1.5 s 5s 0.25–4 h 0.5–50 0.1–1,000 s 15–100 5.3–10 h 0.5–4 h 0.3–2 LHSV LHSV 8,000 GHSV 18–72 h 0.2 s 7,500 GHSV 2.4 s 700 GHSV 0.3 LHSV 0.01–1 s 7h 10 h 0.5–2.5 h 0.5–5 s 1.7 h 1.0 LHSV WHSV 0.005 s 1.5–4 h 20,000 GHSV 4–10,000 GHSV 80–120,000 GHSV ~106 GHSV h L, 10 s G Source and page† [6] 133 [4] 285, [6] 179, [9] 201 [6] 161 [7] 389 [7] 406 [7] 410 [7] 421 [3] 562 [3] 732 [2] 533 [7] 437 [10] 284 [7] 468 [7] 474 [7] 480 [7] 520 [7] 522 [11] [3] 482, 539, [7] 529 [9] 136, [10] 335 [7] 452 [7] 547 [7] 549 [7] 559 [6] 139 [7] 393 [7] 578 [6] 99 [7] 607 [5] 424 [6] 86 [1] 328 [7] 652 [7] 657 [7] 656 [2] 850, [7] 673 [7] 699 [12] 173 [15] 23 [13]3 383 [15] 211 [15] 82 [14] 332 [16] 69 [16] 306 [16] 334 [16] 263 [12] 140 *Abbreviations: reactors: batch (B), continuous stirred tank (CST), fixed bed of catalyst (FB), fluidized bed of catalyst (FL), furnace (Furn.), monolith (M), multitubular (MT), semicontinuous stirred tank (SCST), tower (TO), tubular (TU) Phases: liquid (L), gas (G), both (LG) Space velocities (hourly): gas (GHSV), liquid (LHSV), weight (WHSV) Not available, NA To convert atm to kPa, multiply by 101.3 †1 J J McKetta, ed., Encyclopedia of Chemical Processing and Design, Marcel Dekker, 1976 to date (referenced by volume) W L Faith, D B Keyes, and R L Clark, Industrial Chemicals, revised by F A Lowenstein and M K Moran, John Wiley & Sons, 1975 G F Froment and K B Bischoff, Chemical Reactor Analysis and Design, John Wiley & Sons, 1979 R J Hengstebeck, Petroleum Processing, McGraw-Hill, New York, 1959 V G Jenson and G V Jeffreys, Mathematical Methods in Chemical Engineering, 2d ed., Academic Press, 1977 H F Rase, Chemical Reactor Design for Process Plants, Vol 2: Case Studies, John Wiley & Sons, 1977 M Sittig, Organic Chemical Process Encyclopedia, Noyes, 1969 (patent literature exclusively) Student Contest Problems, published annually by AIChE, New York (referenced by year) M O Tarhan, Catalytic Reactor Design, McGraw-Hill, 1983 10 K R Westerterp, W P M van Swaaij, and A A C M Beenackers, Chemical Reactor Design and Operation, John Wiley & Sons, 1984 11 Personal communication (Walas, 1985) 12 B C Gates, J R Katzer, and G C A Schuit, Chemistry of Catalytic Processes, McGraw-Hill, 1979 13 B E Leach, ed., Applied Industrial Catalysts, vols., Academic Press, 1983 14 C N Satterfield, Heterogeneous Catalysis in Industrial Practice, McGraw-Hill, 1991 15 C L Thomas, Catalytic Processes and Proven Catalysts, Academic Press, 1970 16 Heck, Farrauto, and Gulati, Catalytic Air Pollution Control: Commercial Technology, Wiley-Interscience, 2002 REACTOR CONCEPTS (a) (d) (b) (e) (c) (f) Stirred tank reactors with heat transfer (a) Jacket (b) Internal coils (c) Internal tubes (d) External heat exchanger (e) External reflux condensor (f) Fired heater (Walas, Reaction Kinetics for Chemical Engineers, McGraw-Hill, 1959.) FIG 19-1 cylindrical tank with partitions: feed enters the first compartment and over (or under) flows to the next compartment, and so on The composition is maintained as uniform as possible in each individual compartment; however, a stepped concentration gradient exists from one CSTR to the next When the reactants have limited solubility (miscibility) and a density difference, the vertical staged reactor with countercurrent operation may be used Alternatively, each CSTR in a series or parallel configuration can be an independent vessel Examples of stirred tank reactors with heat transfer are shown in Fig 19-1 A tubular flow reactor (TFR) is a tube (or pipe) through which reactants flow and are converted to product The TFR may have a varying diameter along the flow path In such a reactor, there is a continuous gradient (in contrast to the stepped gradient characteristic of a CSTR-inseries battery) of concentration in the direction of flow Several tubular reactors in series or in parallel may also be used Both horizontal and vertical orientations are common When heat transfer is needed, individual tubes are jacketed or a shell-and-tube construction is used The reaction side may be filled with solid catalyst or internals such as static mixers (to improve interphase contact in heterogeneous reactions or to improve heat transfer by turbulence) Tubes that have 3- to 4-in diameter and are several miles long may be used in polymerization service Large-diameter vessels, with packing (or trays) used to regulate the residence time in the reactor, may also be used Some of the configurations in use are axial flow, radial flow, multishell with built-in heat exchangers, and so on A reaction battery of CSTRs in series, although both mechanically and operationally more complex and expensive than a tubular reactor, provides flexibility Relatively slow reactions are best conducted in a stirred tank reactor battery A tubular reactor is used when heat transfer is needed, where high pressures and/or high (or low) temperatures occur, and when relatively short reaction times suffice Semibatch Reactors Some of the reactants are loaded into the reactor, and the rest of the reactants are fed gradually Alternatively, one reactant is loaded into the reactor, and the other reactant is fed continuously Once the reactor is full, it may be operated in a batch mode to complete the reaction Semibatch reactors are especially favored when there are large heat effects and heat-transfer capability is limited Exothermic reactions may be slowed down and endothermic reactions controlled by limiting reactant concentration In bioreactors, the reactant concentration may be limited to minimize toxicity Other situations that may call for semibatch reactors include control of undesirable by-products or when one of the reactants is a gas of limited solubility that is fed continuously at the dissolution rate 19-7 Classification by End Use Chemical reactors are typically used for the synthesis of chemical intermediates for a variety of specialty (e.g., agricultural, pharmaceutical) or commodity (e.g., raw materials for polymers) applications Polymerization reactors convert raw materials to polymers having a specific molecular weight and functionality The difference between polymerization and chemical reactors is artificially based on the size of the molecule produced Bioreactors utilize (often genetically manipulated) organisms to catalyze biotransformations either aerobically (in the presence of air) or anaerobically (without air present) Electrochemical reactors use electricity to drive desired reactions Examples include synthesis of Na metal from NaCl and Al from bauxite ore A variety of reactor types are employed for specialty materials synthesis applications (e.g., electronic, defense, and other) Classification by Phase Despite the generic classification by operating mode, reactors are designed to accommodate the reactant phases and provide optimal conditions for reaction Reactants may be fluid(s) or solid(s), and as such, several reactor types have been developed Singlephase reactors are typically gas- (or plasma- ) or liquid-phase reactors Two-phase reactors may be gas-liquid, liquid-liquid, gas-solid, or liquidsolid reactors Multiphase reactors typically have more than two phases present The most common type of multiphase reactor is a gas-liquidsolid reactor; however, liquid-liquid-solid reactors are also used The classification by phases will be used to develop the contents of this section In addition, a reactor may perform a function other than reaction alone Multifunctional reactors may provide both reaction and mass transfer (e.g., reactive distillation, reactive crystallization, reactive membranes, etc.), or reaction and heat transfer This coupling of functions within the reactor inevitably leads to additional operating constraints on one or the other function Multifunctional reactors are often discussed in the context of process intensification The primary driver for multifunctional reactors is functional synergy and equipment cost savings REACTOR MODELING As discussed in Sec 7, chemical kinetics may be mathematically described by rate equations Reactor performance is also amenable to quantitative analysis The quantitative analysis of reaction systems is dealt with in the field of chemical reaction engineering The level of mathematical detail that can be included in the analysis depends on the level of understanding of the physical and chemical processes that occur in a reactor As a practical matter, engineering data needed to build a detailed model for some new chemistry typically are unavailable early in the design phase Reactor designers may use similarity principles (e.g., dimensionless groups), rules of thumb, trend analysis, design of experiments (DOE), and principal-component analysis (PCA) to scale up laboratory reactors For hazardous systems in which compositional measurements are difficult, surrogate indicators such as pressure or temperature may be used As more knowledge becomes available, however, a greater level of detail may be included in a mathematical model A detailed reactor model may contain information on vessel configuration, stoichiometric relationships, kinetic rate equations, correlations for thermodynamic and transport properties, contacting efficiency, residence time distribution, and so on Models may be used for analyzing data, estimating performance, reactor scale-up, simulating start-up and shutdown behavior, and control The level of detail in a model depends on the need, and this is often a balance between value and cost Very elaborate models are justifiable and have been developed for certain widely practiced and large-scale processes, or for processes where operating conditions are especially critical Modeling Considerations A useful reactor model allows the user to predict performance or to explore uncertainties not easily or cost-effectively investigated through experimentation Uncertainties that may be explored through modeling may include scale-up options, explosion hazards, runaway reactions, environmental emissions, reactor internals design, and so on As such, the model must contain an optimal level of detail (principle of optimal sloppiness) required to meet the desired objective(s) For example, if mixing is critical to performance, the model must include flow equations that reflect the role of mixing If heat effects are small, an isothermal model may be used 19-48 REACTORS SOLIDS REACTORS Reactions of solids are typically feasible only at elevated temperatures High temperatures are achieved by direct contact with combustion gases Often, the product of reaction is a gas The gas has to diffuse away from the reactant, sometimes through a solid product Thermal and mass-transfer resistances are major factors in the performance of solids reactors There are a number of commercial processes that utilize solid reactors Reactor analysis and design appear to rely on empirical models that are used to fit the kinetics of solids decomposition Most of the information on commercial reactors is proprietary General references on solids reactions include Brown, Dollimore, and Galwey [“Reactions in the Solid State,” in Bamford and Tipper (eds.), Comprehensive Chemical Kinetics, vol 22, Elsevier, 1980], Galwey (Chemistry of Solids, Chapman and Hall, 1967), Sohn and Wadsworth (eds.) (Rate Processes of Extractive Metallurgy, Plenum Press, 1979), Szekely, Evans, and Sohn (Gas-Solid Reactions, Academic Press, 1976), and Ullmann (ed.) (Enzyklopaedie der technischen Chemie, “Uncatalyzed Reactions with Solids,” vol 3, 4th ed., Verlag Chemie, 1973, pp 395–464) THERMAL DECOMPOSITION Thermal decompositions may be exothermic or endothermic Solids that decompose on heating without melting often form gaseous products When the product is a gas, the reaction rate can be affected by diffusion so particle size can be important Aging of solids can result in crystallization of the surface Annealing reduces strains and slows the decomposition rate The decomposition of some fine powders follows a first-order rate law Otherwise, empirical rate equations are available (e.g., in Galwey, Chemistry of Solids, Chapman and Hall, 1967) A few organic compounds decompose before melting These decomposition processes are highly exothermic and may cause explosions Decomposition kinetics may follow an autocatalytic law The temperature range for decomposition is 100 to 200°C (212 to 392°F) The decomposition of oxalic acid (m.p 189°C) obeyed a zero-order law at 130 to 170°C (266 to 338°F) The decomposition of malonic acid has been measured for both the solid and the supercooled liquid Exothermic decompositions are nearly always irreversible When several gaseous products are formed, the reverse reaction would require that these products all combine together, which is unlikely Commercial interest in such materials lies more in their energy storage properties than as a source of desirable products These are often nitrogen-containing compounds such as azides, diazo compounds, and nitramines Ammonium nitrate, an important explosive, decomposes into nitrous oxide and water In the solid phase, decomposition begins at about 150°C (302°F) but becomes extensive only above its melting point (170°C) (338°F) The reaction is first-order, with activation energy of about 40 kcal/(g⋅mol) [72,000 Btu/(lb⋅mol)] Traces of moisture and Cl− lower the decomposition temperature Many investigations have reported on the decomposition of azides of barium, calcium, strontium, lead, copper, and silver in the range of 100 to 200°C (212 to 392°F) Activation energies were found to be 30 to 50 kcal/(g⋅mol [54,000 to 90,000 Btu/(lb⋅mol)] or so Some difficulties with data reproducibility were encountered with these hazardous materials Lead styphnate (styphnic acid contains nitrogen) monohydrate was found to detonate at 229°C (444°F) The course of decomposition could be followed at 228°C and below Sodium azide is a propellant in most motor vehicle SRS systems (airbags) Silver oxalate decomposes smoothly and completely in the range of 100 to 160°C (212 to 320°F) Ammonium chromates and some other solids exhibit aging effects Material that has been stored for months or years follows a different decomposition rate than a fresh material Examples of such materials are available in the review by Brown et al (“Reactions in the Solid State,” in Bamford and Tipper, Comprehensive Chemical Kinetics, vol 22, Elsevier, 1980) Endothermic decompositions are generally reversible Hydroxides (which give off water) and carbonates (which give off CO2) have been the most investigated compounds Activation energies are nearly the same as reaction enthalpies As the reaction proceeds, the rate of reaction may be limited by diffusion of the water through the product layer Since a particular compound may have several hydrates, the level of dehydration will depend on the partial pressure of water vapor in the gas For example, FeCl2 combines with 4, 5, 7, or 12 molecules of water with melting points ranging from about 75 to 40°C (167 to 104°F) The dehydration of CuSO4 pentahydrate at 53 to 63°C (127 to 145°F) and of the trihydrate at 70 to 86°C (158 to 187°F) obeys the AvramiErofeyev equation [−ln(1 − x) = ktn, n = 3.5, 4] The rate of water loss from Mg(OH)2 at lower temperatures is sensitive to the partial pressure of water Its decomposition above 297°C (567°F) yields appreciable amounts of hydrogen and is not reversible Carbonates decompose at relatively high temperatures, e.g., 660 to 740°C (1220 to 1364°F) for CaCO3 When deep beds are used, the rate of heat transfer or the rate of CO2 removal controls the decomposition rate Some ammonium salts decompose reversibly and release ammonia, e.g., (NH4)2SO4 ⇔ NH4HSO4 + NH3 at 250°C (482°F) Further heating can release SO3 irreversibly The decomposition of silver oxide was one of the earliest solid reactions studied It is smoothly reversible below 200°C (392°F) The reaction is sensitive to the presence of metallic silver at the start (indicating autocatalysis) and to the presence of silver carbonate, which was accidentally present in some investigations SOLID-SOLID REACTIONS In solid-solid reactions, ions or molecules in solids diffuse to the interface prior to reaction This diffusion takes place through the normal crystal lattices of reactants and products as well as in channels and fissures of imperfect crystals Solid diffusion is slow compared to liquids even at the elevated temperatures at which these reactions have to be conducted Solid-solid reactions are conducted in powder metallurgy Typical particle sizes are 0.1 to 1000 µm and pressures are 138 to 827 MPa (20,000 to 60,000 psi) Reactions of solids occur in ceramic, metallurgical, and other industries Even though cement manufacture has been discussed in the gas-solid reactor section, solid-solid reactions take place as well Large contact areas between solid phases are essential These may be obtained by forming and mixing fine powders and compressing them Reaction times are to h at 1200 to 1500°C (2192 to 2732°F) even with 200-mesh particles The literature reports several examples of laboratory solid-solid reactions The mechanism of zinc ferrite formation (ZnO + Fe2O3 ⇒ ZnFe2O4) has been studied up to temperatures of 1200°C (2192°F) At lower temperatures, ZnO is the mobile phase that migrates and coats the Fe2O3 particles Similarly, MgO is the mobile phase in the MgO + Fe2O3 ⇒ MgFe2O4 reaction Smaller particles (< µm) obey the power law x = k ln t, but larger ones have a more complex behavior In the reaction 2AgI + HgI2 ⇒ Ag2 HgI4, nearly equivalent amounts of the ions Ag+ and Hg2+ were found to migrate in opposite directions and arrive at their respective interfaces after 66 days at 65°C (149°F) Several reactions that yield gaseous products have attracted attention because their progress is easily followed Examples include MnO3 + 2MoO3 ⇒2MnMoO4 + 0.5O2 (where MoO3 was identified as the mobile phase) and Ca3 (PO4)2 + 5C ⇒ 3CaO + P2 + 5CO For the reaction KClO4 + 2C ⇒KCl + CO2, fine powders were compressed to 69 MPa (10,000 psi) and reacted at 350°C (662°F), well below the 500°C (932°F) melting point The reaction CuCr2O4 + CuO ⇒ Cu2Cr2O4 + 0.5O2 eventually becomes diffusion-controlled and is described by the relationship [1 − (1 − x)1ր3]2 = k ln t In the reaction, CsCl + NaI⇒CsI + NaCl, two solid products are formed The ratecontrolling step is the diffusion of iodide ion in CsCl Carbothermic reactions are solid-solid reactions with carbon that apparently take place through intermediate CO and CO2 The reduction of iron oxides has the mechanism FexOy + yCO ⇒ xFe + yCO2, CO2 + C⇒2CO The reduction of hematite by graphite at 907 to 1007°C in the presence of lithium oxide catalyst was correlated by the equation − (1 − x)1ր3 = kt The reaction of solid ilmenite ore and carbon has the mechanism FeTiO3 + CO ⇒Fe + TiO2 + CO2, CO2 + C ⇒2CO A similar case is the preparation of metal carbides from metal and carbon, C + 2H2 ⇒ CH4, Me + CH4 ⇒ MeC + 2H2 Self-Propagating High-Temperature Synthesis (SHS) Conventional methods of synthesizing materials via solid reactions involve MULTIPHASE REACTORS multiple grinding, heating, and cooling of suitable precursor compounds Reactions need extended time periods mainly because interdiffusion in solids is slow, even at high temperatures By contrast, in SHS, highly reactive metal particles ignite in contact with boron, carbon, nitrogen, and silica to form boride, carbide, nitride, and silicide ceramics Since the reactions are extremely exothermic, the reaction fronts propagate rapidly through the precursor powders Usually, the 19-49 ultimate particle size can be controlled by the particle size of the precursors In recent years, several commercial and semicommercial facilities have been built (in Russia, the United States, Spain, and Japan) to synthesize TiC powders, nitrided ferroalloys, silicon nitride (β-phase) and titanium hydride powders, high-temperature insulators, lithium niobate, boron nitride, etc (e.g., Weimer, Carbide, Nitride and Boride Materials Synthesis and Processing, Chapman & Hall, 1997) MULTIPHASE REACTORS Multiphase reactors include, for instance, gas-liquid-solid and gas-liquid-liquid reactions In many important cases, reactions between gases and liquids occur in the presence of a porous solid catalyst The reaction typically occurs at a catalytic site on the solid surface The kinetics and transport steps include dissolution of gas into the liquid, transport of dissolved gas to the catalyst particle surface, and diffusion and reaction in the catalyst particle Say the concentration of dissolved gas A in equilibrium with the gas-phase concentration of A is CaLi Neglecting the gas-phase resistance, the series of rates involved are from the liquid side of the gas-liquid interface to the bulk liquid where the concentration is CaL, and from the bulk liquid to the surface of catalyst where the concentration is Cas and where the reaction rate is ηwkCasm At steady state, = kLa(CaLi − CaL) = ksas(CaL − Cas) = ηwkCasm (19-78) where w is the catalyst loading (mass of catalyst per slurry volume) For a first-order reaction, m = 1, the catalyst effectiveness η is independent of Cas , so that after elimination of CaL and Cas the explicit solution for the observed specific rate is 1 ra,observed = CaLi ᎏ + ᎏ + ᎏ kLa ksas ηwk −1 (19-79) More complex chemical rate equations will require numerical solution Ramachandran and Chaudhari (Three-Phase Chemical Reactors, Gordon and Breach, 1983) apply such rate equations to the sizing of plug flow, CSTR, and dispersion reactors They list 75 reactions and identify reactor types, catalysts, temperature, and pressure for processes such as hydrogenation of fatty oils, hydrodesulfurization, Fischer-Tropsch synthesis, and miscellaneous hydrogenations and oxidations A list of 74 gas-liquid-solid reactions with literature references has been compiled by Shah (Gas-Liquid-Solid Reactions, McGraw-Hill, 1979), classified into groups where the solid is a reactant, a catalyst, or an inert Other references include de Lasa (Chemical Reactor Design and Technology, Martinus Nijhoff, 1986), Gianetto and Silveston (eds.) (Multiphase Chemical Reactors, Hemisphere, 1986), Ramachandran et al (eds.) (Multiphase Chemical Reactors, vol 2, Sijthoff & Noordhoff, 1981) and Satterfield [“Trickle Bed Reactors,” AIChE J 21: 209–228 (1975)] Some contrasting charac- TABLE 19-13 teristics of the main kinds of three-phase reactors are summarized in Table 19-13 BIOREACTORS Bioreactors use live cells or enzymes to perform biochemical transformations of feedstocks to desired products Bioreactor operation is restricted to conditions at which these biological systems can function Most plant and animal cells live at moderate temperatures and not tolerate extremes of pH The vast majority of microorganisms also prefer mild conditions, but some thrive at temperatures above the boiling point of water or at pH values far from neutral Some can endure concentrations of chemicals that most other cells find highly toxic Commercial operations depend on having the correct organisms or enzymes and preventing death (or deactivation) or the entry of foreign organisms that could harm the process The pH, temperature, redox potential, and nutrient medium may favor certain organisms and discourage the growth of others In mixed culture systems, especially those for biological waste treatment, there is an ever-shifting interplay between microbial populations and their environments that influences performance and control Although open systems may be suitable for hardy organisms or for processes in which the conditions select the appropriate culture, many bioprocesses are closed and elaborate precautions including sterilization and cleaning are taken to prevent contamination The optimization of the complicated biochemical activities of isolated strains, of aggregated cells, of mixed populations, and of cell-free enzymes or components presents engineering challenges Performance of a bioprocess can suffer from changes in any of the many biochemical steps functioning in concert, and genetic controls are subject to mutation Offspring of specialized mutants, especially bioengineered ones that yield high concentrations of product, tend to revert during propagation to less productive strains—a phenomenon called rundown Developments such as immobilized enzymes and cells have been exploited partially, and genetic manipulations through recombinant DNA techniques are leading to practical processes for molecules that could previously be found only in trace quantities in plants or animals Bioreactors may have either two phases (liquid-solid, e.g., in anaerobic processes) or three phases (gas-liquid-solid, e.g., aerobic processes) The solid phase typically contains cells that serve as the biocatalyst The Characteristics of Gas-Liquid-Solid Reactors Property Trickle bed Gas holdup Liquid holdup Solid holdup Liquid distribution RTD, liquid phase 0.25–0.45 0.05–0.25 0.5–0.7 Good only at high liquid rate Narrow RTD, gas phase Interfacial area Nearly plug flow 20–50% of geometrical MTC, gas/liquid MTC, liquid/solid Radial heat transfer Pressure drop High High Slow High with small dp Flooded Small High Narrower than for entrained solids reactor Like trickle bed reactor RTD = residence time distribution; MTC = mass-transfer coefficient Stirred tank Entrained solids Fluidized bed 0.2–0.3 0.7–0.8 0.01–0.10 Good Wide Good Wide Backmixed 100–1500 m2/m3 Backmixed 100–400 m2/m3 Narrow Less than for entrained solids reactor Fast Fast Intermediate High Fast Hydrostatic head 0.5–0.7 Good Narrow 19-50 REACTORS solid can be either the free biocatalyst (bacteria, fungi, algae, etc.), also called the biotic phase (with density close to water), or an immobilized version, in which case the cells are immobilized on a solid structure (e.g., porous particles) The liquid is primarily water with dissolved feed (usually a sugar together with mineral salts and trace elements) and products (referred to as metabolites) In aerobic bioreactors, the gas phase is primarily air with the product gas containing product CO2 produced by the organism and evaporated water Bioreactors are mainly mechanically agitated tanks, bubble columns and air lift reactors For low biomass concentrations (e.g., less than 60 g/L) bioreactor design is similar to that of a gas-liquid reactor For some specialized applications, such as in some wastewater treatment processes, packed beds or slurry reactors with immobilized biocatalyst are used Figure 19-32 shows some typical bioreactors While bioreactors not differ fundamentally from other two- and three-phase reactors, as indicated above, there are more stringent requirements regarding control of temperature, pH, contamination (presence and growth of other microorganisms or phage), and toxicity (that may result from high feed and product concentrations) In aerobic processes, since O2 is required for respiration, it must be properly distributed and managed Whereas bacteria and yeast cells are very robust, cultivations of filamentous fungi and especially animal cell cultures and plant cell cultures are quite shear-sensitive To maintain a robust culture of animal and plant cells, very gentle stirring either by a mechanical stirrer or by gas sparging is usually necessary Unlike chemical catalysis, one of the (main) bioreaction products is biomass (new cells), leading to autocatalytic behavior; i.e., the rate of production of new cells per liquid volume is proportional to the cell concentration Section of this Handbook presents more details on the kinetics of bioreactions Bioreactors mainly operate in batch or semibatch mode, which allows better control of the key variables However, an increasing number of bioprocesses are operated in continuous mode, typically processes for treating wastewater, but also large-scale processes such as lactic acid production, conversion of natural gas to biomass (single-cell protein production), and production of human insulin using genetically engineered yeast Continuous operation requires good process control, especially of the sterility of the feed, but also that the biocatalyst be robust and its traits (especially for bioengineered strains) persist over many generations Several special terms are used to describe traditional reaction engineering concepts Examples include yield coefficients for the generally fermentation environment-dependent stoichiometric coefficients, metabolic network for reaction network, substrate for feed, metabolite for secreted bioreaction products, biomass for cells, broth for the fermenter medium, aeration rate for the rate of air addition, vvm for volumetric airflow rate per broth volume, OUR for O2 uptake rate per broth volume, and CER for CO2 evolution rate per broth volume For continuous fermentation, dilution rate stands for feed or effluent rate (equal at steady state), washout for a condition where the feed rate exceeds the cell growth rate, resulting in washout of cells from the reactor Section discusses a simple model of a CSTR reactor (called a chemostat) using empirical kinetics The mass conservation equations for a batch reactor are as follows: Cells: dC Vr ᎏx = (rg − rd)Vr dt (19-80) dCs Substrate: Vr ᎏ = Yxs(−rg)Vr − rsmVr dt (19-81) dCp Product: Vr ᎏ = Yxp(rg)Vr dt (19-82) Several of the terms above have been discussed in Sec 7: rg and rd are the specific rates (per broth volume) for cell growth and death, respectively; rsm is the specific rate of substrate consumed for cell maintenance, and Yxi are the stoichiometric yield coefficient of species i relative to biomass x The maintenance term in Eq (19-81) can result also in an increased production of product p [additional term required in Eq (19-82)] for metabolites such as lactic acid, but not for protein production In many cases, a semibatch reactor is used, where the reactants are added with an initial cells and sugar concentration, and a certain feed profile or recipe is used—this is also called fed batch operation mode Further modeling details are available in the books by Nielsen, Villadsen, and Liden (Bioreaction Engineering Principles, Kluwer Academic/ Plenum Press, 2003) and Fogler (Elements of Chemical Reaction Engineering, 3d ed., Prentice-Hall, 1999) Bioreactors and bioreaction engineering are discussed in detail by Bailey and Ollis (Biochemical Engineering Fundamentals, 2d ed., McGraw-Hill, 1986), Clark (Biochemical Engineering, Marcel Dekker, 1997), and Schugerl and Bellgardt (Bioreaction Engineering, Modeling and Control, Springer, 2000) ELECTROCHEMICAL REACTORS Electrochemical reactors are used for electrolysis (conversion of electric energy to chemicals, e.g., chlor-alkali), power generation (conversion of chemicals to electric energy, e.g., batteries or fuel cells), or for chemical separations (electrodialysis) An electrochemical cell contains at least two electronically conducting electrode phases and one ionic conducting electrolyte phase The electrolyte phase separates the two electrode phases The electrode phases are also connected to each other through an electronically conducting pathway, typically external of the electrochemical cell; but in the case of corrosion, the electrode phases may be localized regions on the same piece of metal, the bulk metal allowing electron flow between the regions Thus a series electric circuit is completed beginning at one electrode through the electrolyte to the second electrode and then out of the reactor through the external circuit back into the starting electrode An electrochemical cell reaction involves the transfer of electrons across an electrode/electrolyte interface There are two types of electrochemical cell reactions In one reaction the electron transfer is from an electrode to a chemical species within the electrolyte, resulting in a reduction process, and in this case the electrode is defined as the cathode The second electrochemical reaction involves the electron transfer from a chemical species within the electrolyte to an electrode, resulting in an oxidation process; in this case the electrode is defined as the anode Each of these cathode (reduction) or anode (oxidation) electrochemical reactions is considered a half-cell reaction Since an electrochemical cell requires a complete series electric circuit, the overall electrochemical cell reaction is the stoichiometric sum of the electrochemical half-cell reactions, and all electrochemical cell reactions are close-coupled to maintain the conservation of electric charge Electrochemical cell reactions are considered heterogeneous reactions since they occur at the interface of the electrode surface and electrolyte Sometimes the electrochemical product species is employed, in turn, as a reducing or oxidizing species, either in the bulk electrolyte or in a separate external process vessel Subsequently, the spent reducing or oxidizing species is regenerated within the electrochemical reactor This augmentation is known as a mediated (or indirect) electrochemical process More details on the mechanism and kinetics of electrochemical reactions are given in Sec An electrochemical reactor is a controlled volume containing the electrolyte and two electrodes The electrode phases may be a solid, e.g., carbon or metal, or a liquid, e.g., mercury The geometry of the electrodes is optimized to maximize energy efficiency and/or cell life and usually consists of parallel plates or concentric cylinders The electrolyte may be a liquid (such as concentrated brine in the production of caustic or a molten salt in the production of aluminum) or a solid (such as a protonconducting Nafion® membrane in fuel cells) As the electric current passes through the electrolyte, a voltage drop occurs that represents an energy loss; therefore, the gap or spacing between the electrodes is usually minimized The electrodes may also be separated by a membrane, a diaphragm, or a separator so as to prevent the unwanted mixing of chemical species, ensure process safety, and maintain product purity and yield One or both of the electrodes may evolve a gas (e.g., chlorine); or alternatively, one or both of the electrodes may be fed with a gas (e.g., hydrogen or oxygen) to reduce cell voltage or utilize gaseous fuels Examples of electrochemical reactors are shown in Fig 19-33 MULTIPHASE REACTORS (a) (2) 19-51 (b) (1) (3) (4) (6) (5) FIG 19-32 Some examples of fermenters (1) Conventional batch fermenter (2) Air lift fermenters: (a) Concentric cylinder or bubble column with draft tube; (b) external recycle (3) Rotating fermenter (4) Horizontal fermenter (5) Deep-shaft fermenter (6) Flash-pot fermenter 19-52 REACTORS b PLATE AND FRAME TANK a c CAPILLARY GAP DIPOLAR ELECTRODE DISKS e d SWISS ROLL f FLUID BED FIXED BEDS A A SECTION A A COOLANT g h GAS DIFFUSION SLURRY I SPE A GAS ELECTROLYTE A j DIPOLAR PARTICLES ELECTROLYTE k SECTION A A ELECTRODIALYSIS GAS Electrochemical reactor configurations [From Oloman, Electrochemical Processing for the Pulp and Paper Industry, The Electrochemical Consultancy, 1999, p 79, Fig 2.10; printed in Great Britain by Alresford Press Ltd Referring to “Tutorial Lectures in Electrochemical Engineering and Technology” (D Chin and R Alkire, eds.), AIChE Symposium Series 229, vol 79, 1983; reproduced with permission.] FIG 19-33 MULTIPHASE REACTORS The size of an electrochemical reactor may be determined by evaluating the capital costs and the operating costs (on a dollar per unit mass basis) as a function of the operating current density (production rate per unit electrode area basis) Typically, the capital costs decrease with increasing current density, and the operating cost increase with current density, thus, a minimum in the total costs may be observed and serve as a basis for the sizing of the electrochemical reactor Given an optimal current density, the electrochemical reactor design is refined to minimize voltage losses and maximize current efficiency This is done by taking into consideration the component availability (e.g., membrane widths), the management of the excess heat removal, the minimization of pressure drops (due to liquid and gas traffic within the electrochemical reactor), and the maintenance costs (associated with reactor rebuilding) The largest, cost-effective reactor size is then replicated to meet production capacity needs An electrochemical reactor usually has shorter operating life than the rest of the plant facility, requiring the periodic rebuilding of the reactors In electrochemical engineering, several terms share similar definitions to traditional reaction engineering These include fractional conversion, yield, selectivity, space velocity, and space time yield Several terms are unique to electrochemical reaction engineering such as cell voltage (the electric potential difference between the two electrodes within the electrochemical cell) and cell overpotentials (voltage losses within the electrochemical cell) Voltage losses include (1) ohmic overpotential (associated with passage of electric current in the bulk of the electrolyte phase and the bulk electrode phases and the electrical conductors between the electrochemical cell and the power supply or electrical load); (2) activation overpotential (associated with the limiting rates at which some steps in the electrode reactions can proceed); and (3) concentration overpotential (generated from the local depletion of reactants and accumulation of products at the electrode/electrolyte interface relative to the bulk electrolyte phase due to mass transport limitations) The current density is the current per unit surface area of the electrode Typically, the geometric or projected area is utilized since the true electrode area is usually difficult to estimate due to surface roughness and/or porosity It is related to the production rate of the electrolytic cell through the Faraday constant The current efficiency is the ratio of the theoretical electric charge (coulombs) required for the amount of product obtained to the total amount of electric charge passed through the electrochemical cell Many of these and other terms are discussed in Sec 7, in Plectcher and Walsh (Industrial Electrochemistry, 2d ed., Chapman and Hall, 1984) and in Gritzner and Kreysa [“Nomenclature, Symbols and Definitions in Electrochemical Engineering,” Pure & Appl Chem 65: 5, 1009–1020 (1993)] A discussion of electrochemical reactors is available in books by Prentice (Electrochemical Engineering Principles, Prentice-Hall, 1991), Hine (Electrode Processes and Electrochemical Engineering, Plenum Press, 1985), Oloman (Electrochemical Processing for the Pulp and Paper Industry, The Electrochemical Consultancy, 1996), and Goodridge and Scott (Electrochemical Process Engineering: A Guide to the Design of Electrolytic Plant, Plenum, 1995) REACTOR TYPES Multiphase reactors are typically mechanically agitated vessels, bubble columns, trickle bed, flooded fixed beds, gas-liquid-solid fluidized beds, and entrained solids reactors Agitated reactors keep solid catalysts in suspension mechanically; the overflow may be a clear liquid or slurry, and the gas disengages from the vessel Bubble column reactors keep the solids in suspension as a result of agitation caused by the sparging gas In trickle bed reactors both gas and liquid phases flow down through a packed bed of catalyst The reactor is gas continuous with liquid “trickling” as a film over the solid catalyst In flooded reactors, the gas and liquid flow upward through a fixed bed The reactor is liquid continuous As the superficial velocity is increased, the solids first become suspended (as a dense fluidized bed) and may eventually be entrained and the effluent separated into its phases in downstream equipment When the average residence time of solids approaches that of the liquid, the reactor becomes an entrained solids reactor Agitated Slurry Reactors The gas reactant and solid catalyst are dispersed in a continuous liquid phase by mechanical agitation using stirrers Most issues associated with gas-liquid-solid stirred tanks are analogous to the gas-liquid systems In addition to providing good 19-53 gas-liquid contacting, the agitation has to be sufficient to maintain the solid phase suspended Catalytic reactions in stirred gas-liquid-solid reactors are used in a large number of applications including hydrogenations, oxidations, halogenations, and fermentations The benefits of using a mechanically agitated tank include nearly isothermal operation, excellent heat transfer, good mass transfer, and use of high-activity powder catalyst with minimal intraparticle diffusion limitations The reactors may be operated in a batch, semibatch, or continuous mode; and catalyst deactivation may be managed by on-line catalyst makeup Scale-up is relatively straightforward through geometric similarity and by providing the agitator power/volume required to produce the same volumetric mass-transfer coefficient at different scales The hydrodynamics are decoupled from the gas flow rate Some downsides of stirred gas-liquid-solid reactors include difficulty with catalyst/liquid product separation and lower volumetric productivity than fixed beds (due to lower catalyst loading per reactor volume) In addition, the reactor size may be limited due to high power consumption (due to horsepower limitations on agitator motor)—typically the limit is at around 50 m3 Sealing of the agitator system can also be challenging for large reactors (magnetic coupling is used for small to midrange units) These result in increased capital and operating costs Solid particles are in the range of 0.01 to 1.0 mm (0.0020 to 0.039 in), the minimum size limited by filterability Small diameters are used to provide as large an interface as possible to minimize the liquid-solid mass-transfer resistance and intraparticle diffusion limitations Solids concentrations up to 30 percent by volume may be handled; however, lower concentrations may be used as well For example, in hydrogenation of oils with Ni catalyst, the solids content is about 0.5 percent In the manufacture of hydroxylamine phosphate with Pd-C, the solids content is 0.05 percent The hydrodynamic parameters that are required for stirred tank design and analysis include phase holdups (gas, liquid, and solid); volumetric gas-liquid mass-transfer coefficient; liquid-solid mass-transfer coefficient; liquid, gas, and solid mixing; and heat-transfer coefficients The hydrodynamics are driven primarily by the stirrer power input and the stirrer geometry/type, and not by the gas flow Hence, additional parameters include the power input of the stirrer and the pumping flow rate of the stirrer The reactant gas either is sparged below the stirrer or is induced from the vapor space by a gas-inducing agitator which has a hollow shaft with suction orifices on the shaft and discharge orifices on the impeller Impellers vary with applications For low-viscosity applications, flat-bladed Rushton turbines are widely used and provide radial mixing and gas dispersion Pitched-blade turbines may also be used to induce axial flow Often multiple impellers are provided on one shaft, sometimes with a mix of flat blade and pitched-blade type agitators Additional information may be obtained from the corresponding section in this Handbook and from Baldyga and Bourne (Turbulent Mixing and Chemical Reactions, Wiley, 1998) As the stirrer speed is increased, different flow regimes are observed depending on the stirrer type/geometry and the nature of the gas-liquid system considered For example, for a Rushton turbine with a low-viscosity liquid, three primary flow regimes are observed (Fig 19-34) Regime I (Fig 19-34a) has single bubbles that rise, and the gas is not dispersed uniformly Regime II (Fig 19-34b) has the gas dispersed radially as the bubbles ascend Regime III (Fig 19-34c) has the gas recirculated to the stirrer in an increasingly complex pattern [see, e.g., Baldi, Hydrodynamics and Gas-Liquid Mass Transfer in Stirred Slurry Reactors, in Gianetto and Silveston (eds.), Multiphase Chemical Reactors, Hemisphere, 1986] For gas-liquid systems, the power dissipated by the stirrer at the same stirrer speed N is lower than the corresponding power input for liquid systems due to reduced drag on the impeller The power of the gassed system PG is related to that of the ungassed system P0 by using the power number NP correlation with the aeration number Na: PG NP = ᎏ P0 (19-83) QG Na = ᎏ3 NDI (19-84) 19-54 REACTORS Increasing QG Constant N (b) N NF CD (a) N (c) H C/H= 1/4 h T Constant Q G Increasing N H =T FIG 19-34 Gas circulation as a function of stirrer speed (From Nienow et al., 5th European Conference on Mixing, Wurzburg, 1985; published by BHRA, The Fluid Engineering Centre, Cranfield, England; Fig 1.) The power number is a decreasing function of the aeration rate, as shown in Fig 19-35 For instance, Hughmark [Ind Eng Chem Proc Des Dev 19: 638 (1980)] developed a correlation for the power number of Rushton turbines that correlates a large database: D3I NP = 0.1Na−0.25 ᎏ VL −0.25 N2D4I −0.2 ᎏ gH V 2/3 (19-85) I L Increasing the solids content increases the power number, as indicated, e.g., by Wiedman et al [Chem Eng Comm 6: 245 (1980)] With solids present, a minimum agitator speed is required to suspend all the solids, e.g., the correlation of Baldi et al [Chem Eng Sci 33: 21 (1978)]: 0.42 0.14 0.125 β2 µ0.17 dp w L [g(ρp − ρL)] Nm = ᎏᎏᎏ 0.58 0.89 ρL DI where w is the catalyst loading in weight percent and parameter β2 depends on reactor/impeller ratio, e.g., from Nienow [Chem Eng J 9: 153 (1975)], β2 = 2(dRրDI).1.33 Gas holdup and volumetric gas-liquid mass-transfer coefficients are correlated with the gassed power input/volume and with the aeration rate (actual gas superficial velocity), e.g., the correlation of van’t Riet [Ind Eng Chem Proc Des Dev 18: 357 (1979)] for the volumetric mass-transfer coefficient of coalescing and noncoalescing systems: kLa = e 1.0 PG / P 0.8 u0.5 G for coalescing nonviscous liquids u0.2 G (19-87) for noncoalescing nonviscous liquids PG 2.6 × 10−2 ᎏᎏ VL PG 2.0 × 10−3 ᎏᎏ VL 0.4 0.7 For the gas holdup a similar correlation was developed by Loiseau et al [AIChE J 23: 931 (1977)]: 0.6 (19-86) εG = e P′G 0.011σ −0.36µL−0.056 ᎏᎏ VL P′G 0.0051 ᎏᎏ VL 0.57 u0.24 G 0.27 u0.36 G for nonfoaming systems for nonfoaming system (19-88) QGρGRT P1 P′G PG QGρGu20 ᎏ = ᎏ + 0.03 ᎏ + ᎏ ln ᎏ VLMWG P2 VL VL VL 0.4 0.02 0.04 Q G / ND3 0.06 0.08 FIG 19-35 Effect of aeration number and stirrer speed on the power number— N increases in order of N1 < N2 < N3 < N4 [Adapted from Baldi, “Hydrodynamics and Mass Transfer in Stirred-Slurry Reactors,” in Gianetto and Silveston (eds.), Multiphase Chemical Reactors, Hemisphere Publishing Corp., 1986, Fig 14.8.] The last two terms of the power/volume equation include the power/volume from the isothermal expansion of the gas through the gas distributor holes having a velocity u0 and the power/volume to transfer the gas across the hydrostatic liquid head Increasing the solids loading leads to a decrease in gas holdup and gas-liquid volumetric mass-transfer coefficient at the same power/volume [e.g., Inga and Morsi, Can J Chem Eng 75: 872 (1997)] MULTIPHASE REACTORS 19-55 Liquid-solid mass transfer is typically not limiting due to the small particle size resulting in large particle surface area/volume of reactor, unless the concentration of the particles is very low, and or larger particles are used In the latter case, intraparticle mass-transfer limitations would also occur Ramachandran and Chaudhari (Three-Phase Catalytic Reactors, Gordon and Breach, 1983) present several correlations for liquid-solid mass transfer, typically as a Sherwood number versus particle Reynolds and Schmidt numbers, e.g., the correlation of Levins and Glastonbury [Trans Inst Chem Engrs 50: 132 (1972)]: 0.38 + 0.44Re0.5 p Sc Sh = e ᎏk dᎏ s p D ρLucdp Rep = ᎏ ᎏ µL νL Sc = ᎏᎏ D (19-89) Here uc is a characteristic velocity, and the velocity terms composing it are estimated from additional correlations There is good heat transfer in agitated gas-liquid-solid slurry reactors; see, e.g., van’t Riet and Tramper for correlations (Basic Bioreactor Design, Marcel Dekker, 1991) Additional information on mechanically agitated gas-liquid-solid reactors can be obtained in van’t Riet and Tramper (Basic Bioreactor Design, Marcel Dekker, 1991), Ramachandran and Chaudhari (ThreePhase Catalytic Reactors, Gordon and Breach, 1983), and Gianetto and Silveston (Multiphase Chemical Reactors, Hemisphere, 1986) Examples • Liquid benzene is chlorinated in the presence of metallic iron turnings or Raschig rings at 40 to 60°C (104 to 140°F) • Carbon tetrachloride is made from CS2 by bubbling chlorine into it in the presence of iron powder at 30°C (86°F) • Substances that have been hydrogenated in slurry reactors include nitrobenzene with Pd-C, butynediol with Pd-CaCO3, chlorobenzene with Pt-C, toluene with Raney® Ni, and acetone with Raney® Ni • Some oxidations in slurry reactors include cumene with metal oxides, cyclohexene with metal oxides, phenol with CuO, and n-propanol with Pt • Aerobic fermentations For many hydrogenations, semibatch operations often are preferred to continuous ones because of the variety of feedstocks or product specifications, or long reaction times, or small production rates A sketch of a batch hydrogenator is shown in Fig 19-36 The vegetable oil hydrogenator, which is to scale, uses three impellers The best position for inlet of gas is at a point of maximum turbulence near the impeller, or at the bottom of the draft tube A sparger is desirable; however, an open pipe is often used A two-speed motor is desirable to prevent overloading Since the gassed power requirement is significantly less than ungassed, the lower speed is used when the gas supply is cut off but agitation is to continue In tanks of 5.7 to 18.9 m3 (1500 to 5000 gal), rotation speeds are from 50 to 200 rpm and power requirements are to 75 hp; both depend on superficial velocities of gas and liquid [Hicks and Gates, Chem Eng., pp 141–148 (July 1976)] As a rough guide, power requirements and impeller tip speeds are shown in Table 19-14 Edible oils are mixtures of unsaturated compounds with molecular weights in the vicinity of 300 The progress of the hydrogenation reaction is expressed in terms of iodine value (IV), which is a measure of unsaturation The IV is obtained by a standardized procedure in which the iodine adds to the unsaturated double bond in the oil IV is the ratio of the amount of iodine absorbed per 100 g of oil To start a hydrogenation process, the oil and catalyst are charged, then the vessel is evacuated for safety and hydrogen is continuously added and maintained at some fixed pressure, usually in the range of to 10 atm (14.7 to 147 psi) Internal circulation of hydrogen is provided by axial and radial impellers or with a hollow impeller that throws the gas out centrifugally and sucks gas in from the vapor space through the hollow shaft Some plants have external gas circulators Reaction times are to h For edible oils, the temperature is kept at about 180°C (356°F) Since the reaction is exothermic and because space for heat-transfer coils in the vessel is limited, the process is organized to give a maximum IV drop of about 2.0/min The rate of Stirred tank hydrogenator for edible oils (Votator Division, Chemetron Corporation.) FIG 19-36 reaction drops off rapidly as the reaction proceeds, so a process may take several hours The endpoint of a hydrogenation is a specified IV of the product Hardness or refractive index also can be measured to follow reaction progress Saturation of the oil with hydrogen is maintained by agitation The rate of reaction depends on agitation and catalyst concentration Beyond a certain agitation rate, resistance to mass transfer is eliminated, and the rate becomes independent of pressure The effect of catalyst concentration also reaches limiting values The effects of pressure and temperature on the rate are indicated by Fig 19-37 A supported nickel catalyst (containing 20 to 25 weight percent Ni on a porous silica particle) is typically used The pores allow access of the reactants to the extended pore surface, which is in the range of 200 to 600 m2/g (977 × 103 to 2931 × 103 ft2րlbm) of which 20 to 30 percent is catalytically active The concentration of catalyst in the slurry can vary over a wide range but is usually under 0.1% Ni After the reaction is complete, the catalyst can be easily separated from the product Catalysts are subject to degradation and poisoning, particularly by sulfur compounds Accordingly, 10 to 20 percent of the recovered catalyst is replaced by fresh catalyst before reuse Other catalysts are applied in TABLE 19-14 Guidelines Power Requirements and Impeller Tip Speed Operation Homogeneous reaction With heat transfer Liquid-liquid mixing Gas-liquid mixing *1 hp/1000 gal = 0.197 kW/m3 hp/1000 gal* 0.5–1.5 1.5–5 5–10 Tip speed, ft/s 7.5–10 10–15 15–20 15–20 19-56 REACTORS (a) (b) Hydrogenation of soybean oil (a) Effect of reaction pressure and temperature on rate (b) Effect of catalyst concentration and stirring rate on hydrogenation [Swern (ed.), Bailey’s Industrial Oil and Fat Products, vol 2, Wiley, 1979.] FIG 19-37 special cases Expensive palladium has about 100 times the activity of nickel and is effective at lower temperatures A case study of the hydrogenation of cottonseed oil was made by Rase (Chemical Reactor Design for Process Plants, vol 2, Wiley, 1977, pp 161–178) Slurry Bubble Column Reactors As in the case of gas-liquid slurry agitated reactors, bubble column reactors may also be used when solids are present Most issues associated with multiphase bubble columns are analogous to the gas-liquid bubble columns In addition, the gas flow and/or the liquid flow have to be sufficient to maintain the solid phase suspended In the case of a bubble column fermenter, the sparged oxygen is partly used to grow biomass that serves as the catalyst in the system Many bubble columns operate in semibatch mode with gas sparged continuously and liquid and catalyst in batch mode The benefits of using slurry bubble columns include nearly isothermal operation, excellent heat transfer, good mass transfer, and use of highactivity powder catalyst with minimal intraparticle diffusion limitations The reactors may be operated in a batch, semibatch, or continuous mode and require less power input than mechanically agitated reactors Catalyst deactivation may be managed by on-line catalyst makeup The reactor (essentially an empty shell with a sparger grid at the bottom) is easy to design, and the capital investment can be low Some downsides of slurry bubble column reactors include catalyst/liquid product separation difficulty and lower volumetric productivity than fixed beds (due to lower catalyst loading per reactor volume), and catalyst distribution can be skewed with higher concentration at the bottom than at the top of the reactor Also, accounting for the effect of internals (e.g., heat exchange tubes) and of increased diameter on the hydrodynamics is not well understood Hence gradual scale-up is often required over multiple intermediate scales before commercialization Cold flow models can also be useful in determining hydrodynamics in the absence of reaction As is the case for reactors with two or more mobile phases, a variety of flow regimes exist depending primarily on the gas superficial velocity (the driver for bubble column hydrodynamics) and column diameter A qualitative flow regime map is shown in Fig 19-38 In the homogeneous flow regime at low gas superficial velocity, bubbles are relatively small and rise at constant rate (about 20 to 25 cm/s) As the flow rate is increased, bubbles become larger and irregular in shape, they frequently coalesce and break up, and the transition to churn turbulent regime is obtained In small-diameter columns, the larger bubbles may bridge the column, creating slugs—hence the slug flow regime The large transition zones in Fig 19-38 are indicative of the lack of accurate knowledge and of the dependence of the transition region on conditions (temperature, pressure) and physical properties of the gas and liquid Hydrodynamic parameters that are required for bubble column design and analysis include phase holdups (gas, liquid, and solid for slurry bubble columns); volumetric gas-liquid mass-transfer coefficient; liquid-solid mass-transfer coefficient; liquid, gas, and solid axial and radial mixing; and heat-transfer coefficients These parameters depend strongly on the prevailing flow regime Correlations for gas holdup and the volumetric gas-liquid masstransfer coefficient can have the general form εG = αuβG kLa = γuδG (19-90) where uG is the superficial gas velocity, εG is the gas holdup (fraction of gas volume), kL is the liquid-side gas-liquid mass-transfer coefficient, and a is the interfacial area per volume of either the liquid or the expanded liquid (liquid + gas) The exponents are β,δ∼1 for the homogeneous bubbly flow regime and β,δ < for heterogeneous turbulent flow regime The correlations depend on the gas-liquid-solid system properties Gas-liquid systems can be classified as coalescing leading to increased bubble size, and noncoalescing, leading to larger gas holdup and volumetric mass-transfer coefficients for the latter There is a voluminous literature for these parameters, and there is substantial variability in estimated values—one should be careful to validate the parameters with data applicable to the real system considered For instance, for gas holdup see the correlation of Yoshida FIG 19-38 Flow regime map for gas-liquid bubble columns [Fig 16 in Deckwer et al., Ind Eng Chem Process Des Dev 19:699–708 (1980).] MULTIPHASE REACTORS and Akita [AIChE J 11: (1965)] εG ρLgd2R ᎏ4 = α ᎏ (1 − εG) σ α = e 0.2 0.25 ρ2Lgd3R ᎏ µ 1ր8 1ր12 L uG ᎏ ͙ෆ gdR for pure liquids and nonelectrolytes for salt solutions (19-91) and for volumetric gas-liquid mass-transfer coefficient, see the correlation of Akita and Yoshida [I&EC Proc Des Dev, 12: 76 (1973)]: µL D kLa = 0.6 ᎏ ᎏ d2R ρLD ρLgd2R ρ2Lgd3R ᎏ ᎏ σ µ 0.5 0.62 L 0.31 1.1 εG (19-92) More recent correlations for gas holdup and mass transfer include the effect of pressure and bimodal bubble size distribution (small and large bubbles), in a manner analogous to the treatment of dilute and dense phases in fluidized beds [see, e.g., Letzel et al., Chem Eng Sci., 54: (13): 2237 (1999)] Increasing the catalyst loading decreases the gas holdup and the volumetric gas-liquid mass transfer coefficient [see, e.g., Maretto and Krishna, Catalysis Today, 52: 279 (1999)] Axial mixing in the liquid, induced by the upflow of the gas bubbles, can be substantial in commercial-scale bubble columns, especially in the churn turbulent regime Due to typically small particle size, the axial dispersion of the solid catalyst in slurry bubble columns is expected to follow closely that of the liquid; exceptions are high-density particles The liquid axial mixing can be represented by an axial dispersion coefficient, which typically has the form DaL = αuβG d γR (19-93) Based on theoretical considerations (Kolmogoroff’s theory of isotropic turbulence), β = 1ր3 and γ = 4ր3 For example, Deckwer et al [Chem Eng Sci 29: 2177 (1973)] developed the following correlation: 1.4 DaL = 2.7u0.3 G dR (19-94) It is expected that the strong dependence on reactor diameter only extends up to a maximum diameter beyond which there is no effect of diameter; however, there is disagreement among experts as to what that maximum diameter may be There are a large number of correlations for liquid axial dispersion with widely different predictions, and care must be exerted to validate the predictions with data at some significant scale, even if only in a cold flow mockup The gas axial mixing is due to the bubble size distribution resulting in a distribution of bubble rise velocities, which varies along the column due to bubble breakup and coalescence There are a variety of correlations in the literature, with varying results and reliability, for instance, the correlation of Mangartz and Pilhofer [Verfahrenstechn., 14: 40 (1980)] uG 1.5 DaG = × 10−4 ᎏ dR εG (19-95) This equation is dimensional, and cm/s for uG, cm for dR, and cm2/s for DaG should be used The radial mixing can be represented by radial dispersion coefficients for the gas and the liquid For instance, the liquid radial dispersion coefficient is estimated at less than one-tenth of the axial one Correlations for the heat-transfer coefficient have the general form St = f(Re Fr Pr2) u3GρL Re Fr = ᎏ µLg cpLµL Pr = ᎏ λL hw St = ᎏ uGρLcpL (19-96) For instance, see the correlation of Deckwer et al [Chem Eng Sci 35(6): 1341–1346 (1980)] St = 0.1(Re Fr Pr2)−1ր4 (19-97) 19-57 Additional information on hydrodynamics of bubble columns and slurry bubble columns can be obtained from Deckwer (Bubble Column Reactors, Wiley, 1992), Nigam and Schumpe (Three-Phase Sparged Reactors, Gordon and Breach, 1996), Ramachandran and Chaudhari (Three-Phase Catalytic Reactors, Gordon and Breach, 1983), and Gianetto and Silveston (Multiphase Chemical Reactors, Hemisphere, 1986) Computational fluid mechanics approaches have also been recently used to estimate mixing and mass-transfer parameters [e.g., see Gupta et al., Chem Eng Sci 56(3): 1117–1125 (2001)] Examples There are a number of examples including FischerTropsch synthesis in the presence of Fe or Co catalysts, methanol synthesis in the presence of Cu/Zn solid catalyst, and hydrocracking in the presence of zeolite catalyst Fermentation reactions are conducted in bubble column reactors when there is a benefit for increased scale and for reduced cost The oxygen is sparged from the bottom, and the liquid reactants are added in a semibatch mode The absence of reactor internals is an advantage as it prevents contamination Heat transfer has to be managed through a cooling jacket If heat removal is an issue, cooling coils may be installed Fluidized Gas-Liquid-Solid Reactors In a gas-liquid-solid fluidized bed reactor, only the fluid mixture leaves the vessel Gas and liquid enter at the bottom Liquid is continuous, gas is dispersed Particles are larger than in bubble columns, 0.2 to 1.0 mm (0.008 to 0.04 in) Bed expansion can be small Bed temperatures are uniform within 2°C (3.6°F) in medium-size beds, and heat transfer to embedded surfaces is excellent Catalyst may be bled off and replenished continuously, or reactivated continuously Figure 19-39 shows examples of gas-liquid-solid fluidized-bed reactors Figure 19-39a illustrates a conventional gas-liquid-solid fluidized bed reactor Figure 19-39b shows an ebullating bed reactor for the hydroprocessing of heavy crude oil A stable fluidized bed is maintained by recirculation of the mixed fluid through the bed and a draft tube Reactor temperatures may range from 350 to 600°C (662 to 1112°F) and 200 atm (2940 psi) An external pump sometimes is used instead of the built-in impeller shown Such units were developed for the liquefaction of coal A biological treatment process (Dorr-Oliver Hy-Flo) employs a vertical column filled with sand on which bacterial growth takes place while waste liquid and air are charged A large interfacial area for reaction is provided, about 33 cm2/cm3 (84 in2/in3) BOD removal of 85 to 90 percent is claimed in 15 compared with to h in conventional units In entrained beds, the three-phase mixture flows through the vessel and is separated downstream These reactors are used in preference to fluidized beds when catalyst particles are very fine or subject to disintegration or if the catalyst deactivates rapidly in the process Trickle Bed Reactors Reactant gas and liquid flow cocurrently downward through a packed bed of solid catalyst particles The most common use of trickle bed reactors is for hydrogenation reactions The solubility of feed hydrogen in the liquid even at the higher pressure is insufficient to provide the stoichiometric needs of the reaction, and a gas flow exceeding the need is fed into the reactor High hydrogen partial pressures can prevent catalyst deactivation due to undesirable reactions, such as coking Cooling (or heating) is typically done between stages either with heat transfer to a coolant outside the reactor or through direct cooling with a cold reactant gas or liquid Advantages of a trickle bed are ease of installation, low liquid holdup (and therefore less undesirable homogeneous reactions), minimal catalyst handling issues, low catalyst attrition, and catalyst life of to years The liquid and gas flow in trickle beds approaches plug flow (leading to higher conversion than slurry reactors for the same reactor volume) Downsides of trickle beds include flow maldistribution (bypassing), sensitivity to packing uniformity and prewetting (leading to hot spots), incomplete contacting/wetting, intraparticle diffusion resistance, potential for fouling and bed plugging due to particulate matter in the feed, and high pressure drop A significant fraction of the flow is gas that has to be compressed and recycled (i.e., increased compressor costs) A schematic of a trickle bed reactor is shown in Fig 19-40 The reactor is a high-pressure vessel equipped with a drain and a manhole for vessel entry Typical vessel diameters may range from to 30 ft with height from to 100 ft The liquid enters the reactor and is 19-58 REACTORS (a) (b) Gas-liquid-solid reactors (a) Three-phase fluidized-bed reactor (b) Ebullating bed reactor for hydroliquefaction of coal (Kampiner, in Winnacker-Keuchler, Chemische Technologie, vol 3, Hanser, 1972, p 252.) FIG 19-39 distributed across the cross-section by a distributor plate The liquid feed flows downward due to gravity helped along by the drag of the gas at such a low rate that it is distributed over the catalyst as a thin film The gas enters at the top and is distributed along with the liquid In the simplest arrangement, the liquid distributor is a perforated plate with about 10 openings/dm2 (10 openings/15.5 in2), and the gas enters FIG 19-40 Trickle bed reactor for hydrotreating 20,000 bbl/d of light catalytic cracker oil at 370ЊC and 27 atm To convert atm to kPa, multiply by 101.3 (Baldi in Gianetto and Silveston, Multiphase Chemical Reactors, Hemisphere, 1986, pp 533–563.) through several risers about 15 cm (5.9 in) high More elaborate distributor caps also are used Uniform distribution of liquid across the reactor is critical to reactor performance The aspect ratio of the reactor can vary between and 10 depending on the pressure drop that can be accommodated by the compressor It is not uncommon to redistribute the liquid using a redistribution grid every to 15 ft The catalyst is often loaded on screens supported by a stainless steel grid near the bottom of the reactor Often, large inert ceramic balls are loaded at the very bottom, with slightly smaller ceramic balls above the first layer, and then the catalyst Smaller inert ceramic balls can also be loaded above the catalyst bed and topped off with the larger balls The layer of inert balls can be in to ft in depth The balls restrict the movement of the bed and distribute the liquid across the catalyst As is the case when two or more mobile phases are present, cocurrent gas-liquid downflow through packed beds produces a variety of flow regimes depending on the gas and liquid flow rates and the physical properties of the gas and the liquid In Fig 19-41, a flow regime map for trickle beds of foaming and nonfoaming systems is presented Here L and G are the liquid and gas fluxes (mass flow rate per total flow cross-sectional area) In the low interaction or trickle flow regime, gas is the continuous phase and the liquid is flowing as rivulets Increasing the liquid and gas flow results in high interaction or pulse flow, with the liquid and gas alternatively bridging the bed voids At high liquid flow and low gas flow, the liquid becomes the continuous phase and the gas is the dispersed phase, called dispersed bubble flow Finally at high gas flow and low liquid flow, the spray flow regime exists with liquid being the dispersed phase The literature contains a number of references to other flow regime maps; however, there is no clear advantage of using one map versus another Wall effects can also have a major effect on the hydrodynamics of trickle bed reactors Most of the data reported in the literature are for small laboratory units of 2-in diameter and under Hydrodynamic parameters that are required for trickle bed design and analysis include bed void fraction, phase holdups (gas, liquid, and solid), wetting efficiency (fraction of catalyst wetted by liquid), volumetric gas-liquid mass-transfer coefficient, liquid-solid mass-transfer coefficient (for the wetted part of the catalyst particle surface), gas-solid MULTIPHASE REACTORS FIG 19-41 19-59 Trickle bed flow regime map [From Gianetto et al., AIChE J 24(6):1087–1104 (1978); reproduced with per- mission.] mass-transfer coefficient (for the unwetted part of the catalyst particle surface), liquid and gas axial mixing, pressure drop, and heat-transfer coefficients These parameters vary with the flow regime (i.e., for the low and high interaction regimes) There are a number of pressure drop correlations for two-phase flow in packed beds originating from the Lockhart-Martinelli correlation for two-phase flow in pipes These correlate the two-phase pressure drop to the single-phase pressure drops of the gas and the liquid obtained from the Ergun equation See, for instance, the Larkins correlation [Larkins, White, and Jeffrey, AIChE J 7: 231 (1967)] The static holdup can be correlated with the Eotvos number NEo as it results from a balance of surface tension and gravity forces on the liquid held up in the pores in absence of flow: gravity force ρLgdp2 NEo = ᎏᎏᎏ = ᎏ surface tension force σL (19-100) For instance Fig 19-42 illustrates the dependence of the static holdup on the Eotvos number for porous and nonporous packings ∆PGL 5.0784 ln ᎏᎏ = ᎏᎏ2 ∆PL + ∆PG 3.531 + (ln X) where X = ∆P ᎏ Ί ∆P L 0.05 ≤ X ≤ 30 (19-98) G Since some of the published pressure drop correlations can differ by an order of magnitude, it is best to verify the relationship with actual data before designing a reactor Other approaches to two-phase pressure drop include the relative permeability method of Saez and Carbonell [AIChE J 31(1): 52–62 (1985)] The bed void volume available for flow and for gas and liquid holdup is determined by the particle size distribution and shape, the particle porosity, and the packing effectiveness The total voidage and the total liquid holdup can be divided into external and internal terms corresponding to interparticle (bed) and intraparticle (porosity) voidage The external liquid holdup is further subdivided into static holdup εLs (holdup remaining after bed draining due to surface tension forces) and dynamic holdup εLd Additional expressions for the liquid holdup are the pore fillup Fi and the liquid saturation SL: εt = εB + εp(1 − εB) εL = εLe + εLi εLe = εLd + εLs εLi = Fiεp(1 − εB) εL SL = ᎏ εB total voidage total liquid holdup external liquid holdup internal liquid holdup (19-99) The static liquid holdup for porous and nonporous solids (Fig 7.7 in Ramachandran and Chaudhari, Three-Phase Catalytic Reactors, Gordon and Breach, 1983.) FIG 19-42 liquid saturation 19-60 REACTORS A variety of correlations have been developed for the total and the dynamic liquid holdup For instance, the total liquid holdup has been correlated with the Lockhardt-Martinelli parameter X for spherical and cylindrical particles [Midou, Favier, and Charpentier, J Chem Eng Japan, 9: 350 (1976)] εL 0.66X 0.81 ᎏ = ᎏᎏ εb + 0.66X 0.81 (19-101) Correlations for the dynamic liquid holdup have also been developed as function of various dimensionless numbers including the liquid and gas Reynolds number, and the two-phase pressure drop [see, e.g., Ramachandran and Chaudhari, Three-Phase Catalytic Reactors, Gordon and Breach, 1983; and Hofmann, Hydrodynamics and Hydrodynamic Models of Fixed Bed Reactors, in Gianetto and Silveston (eds.), Multiphase Chemical Reactors, Hemisphere 1986] The various volumetric mass-transfer coefficients are defined in a manner similar to that discussed for gas-liquid and fluid-solid mass transfer in previous sections There are a large number of correlations obtained from different gas-liquid-solid systems For more details see Shah (Gas-Liquid-Solid Reactor Design, McGraw-Hill, 1979), Ramachandran and Chaudhari (Three-Phase Catalytic Reactors, Gordon and Breach, 1983), and Shah and Sharma [Gas-Liquid-Solid Reactors in Carberry and Varma (eds.), Chemical Reaction and Reactor Engineering, Marcel Dekker, 1987] Axial mixing of the liquid is an important factor in the design of trickle bed reactors, and criteria were proposed to establish conditions that limit axial mixing Mears [Chem Eng Sci 26: 1361 (1971)] developed a criterion that when satisfied, ensures that the conversion will be within percent of that predicted by plug flow: uL L Pe = ᎏ > 20n ln ᎏ D 1−x (19-102) where n is the order of the reaction with respect to the limiting reactant and x is the fractional conversion of that reactant Correlations for axial dispersion can be found in Ramachandran and Chaudhari, ThreePhase Catalytic Reactors, Gordon and Breach, 1983 Incomplete wetting can be also a critical factor in reactor design and analysis, leading usually to lower performance due to incomplete utilization of the catalyst bed In a few select cases, the opposite may be the case, e.g., when a volatile reactant reacts faster than its liquid phase because it is not limited by the gas-liquid mass-transfer resistance and higher gas diffusivity Correlations for the fraction of catalyst surface wetted are available, although not very reliable and strongly system-dependent (e.g., Shah, Gas-Liquid-Solid Reactor Design, McGraw-Hill, 1979) Due to the complex hydrodynamics and the dependence of the hydrodynamic parameters on the flow regime, trickle beds are notoriously difficult to scale up Laboratory units (used for kinetics and process development) and commercial units typically are operated at the same liquid hourly space velocity (LHSV) Since the LHSV represents the ratio of the superficial liquid velocity to the reactor length, the superficial velocity in a laboratory reactor will be lower than in a commercial reactor by the ratio of reactor lengths, which is often well over an order of magnitude This means that heat and mass transport parameters may be considerably different in laboratory reactors operated at the target LHSV This also shifts the flow regime from trickle flow (low interaction) in the lab and small pilot plants to the high-interaction regime in large-scale commercial reactors Wall effects in lab units of 50-mm (1.97-in) diameter can be important while these are negligible for commercial reactors of m or more diameter Wall effects in the lab can be reduced by using reactor/particle diameter ratios greater than If that is not possible, inert fines are added in the lab to reduce wall effects Also, in large-diameter beds, uniform liquid distribution is difficult, even with a large number of distributor nozzles, and unless the flow is redistributed, the nonuniformity can persist along the bed, leading to potential hot spots that can cause by-products and fast catalyst deactivation In trickle beds that are not prewetted, a hysteresis phenomenon related to wetting occurs, where the behavior with increasing flow of the liquid phase is not retraced with decreasing liquid flow This can often be avoided by prewetting the reactor before start-up In practice, the thickness of liquid films in trickle beds has been estimated to vary between 0.01 and 0.2 mm (0.004 and 0.008 in) The dynamic liquid holdup fraction is 0.03 to 0.25, and the static fraction is 0.01 to 0.05 The high end of the static fraction includes the liquid that partially fills the pores of the catalyst The effective gas-liquid interface is 20 to 50 percent of the geometric surface of the particles, but it can approach 100 percent at high liquid loading This results in an increase of reaction rate as the amount of wetted surface increases (i.e., when the gas-solid reaction rate is negligible) Examples Hydrodesulfurization of petroleum oils was the first large-scale application of trickle bed reactors commercialized in 1955 In this application, organosulfur species contained in refinery feeds are removed in the presence of hydrogen and a catalyst and released as hydrogen sulfide Conditions depend on the quality and boiling range of the oil The reactor pressure is optimized to increase the solubility of the hydrogen and minimize catalyst deactivation due to coking Over the life of the catalyst, the temperature is increased to maintain a constant conversion Temperatures are in the range of 345 to 425°C (653 to 797°F) with pressures of 34 to 102 atm (500 to 1500 psi) A large commercial reactor may have 20 to 25 m (66 to 82 ft) of total depth of catalyst, and may be up to 3-m (9.8-ft) diameter or above in several beds of 3- to 6-m (9.8- to 19.7-ft) depth Bed depth is often limited by pressure drop, the catalyst crush strength, and the maximum adiabatic temperature increase for stable operation The need to limit pressure drop is driven by the capital and operating costs associated with the hydrogen recycle compressor Catalyst granules are 1.5 to 3.0 mm (0.06 to 0.12 in), sometimes a little more Catalysts are 10 to 20 percent Co and Mo (or Ni and W) on alumina The adiabatic temperature rise in each bed usually is limited to 30°C (86°F) by injection of cold hydrogen between beds Since the liquid trickles over the catalyst, the wetting efficiency of the catalyst is important in determining the volumetric reaction rate As expected, wetting efficiency increases with increasing liquid rate Catalyst effectiveness of particles to mm (0.12 to 0.20 in) in diameter has been found to be about 40 to 60 percent Packed Bubble Columns (Cocurrent Upflow) These reactors are also called flooded-bed reactors In contrast to trickle beds, both gas and liquid flow up cocurrently A screen is needed at the top to retain the catalyst particles Such a unit has been used for the hydrogenation of nitro and double-bond compounds and nitriles [Ovcinnikov et al., Brit Chem Eng 13: 1367 (1968)] High gas rates can cause movement and attrition of the particles Accordingly, such equipment is restricted to low gas flow rates, for instance, where a hydrogen atmosphere is necessary but the consumption of hydrogen is slight The liquid is the continuous phase, and the gas, the dispersed phase Benefits of cocurrent upflow versus trickle (cocurrent downflow) include high wetting efficiency (resulting in good liquid-solid contacting), good liquid distribution, and better heat and mass transfer Disadvantages include higher pressure drop and liquid backmixing, the latter resulting in increased extent of undesirable homogeneous reactions A number of flow regime maps are available for packed bubble columns [see, e.g., Fukushima and Kusaka, J Chem Eng Japan, 12: 296 (1979)] Correlations for the various hydrodynamic parameters can be found in Shah (Gas-Liquid-Solid Reactor Design, McGrawHill, 1979), Ramachandran and Chaudhari (Three-Phase Catalytic Reactors, Gordon and Breach, 1983), and Shah and Sharma [GasLiquid-Solid Reactors in Carberry and Varma (eds.), Chemical Reaction and Reactor Engineering, Marcel Dekker, 1987] Countercurrent Flow The gas flows up countercurrent with the downflow liquid This mode of operation is not as widely used for catalytic reactions since operation is limited by flooding at high gas velocity: at flooding conditions increasing the liquid flow does not result in increase of the liquid holdup For more details see Shah (Gas-Liquid-Solid Reactor Design, McGraw-Hill, 1979) and Hofmann [Hydrodynamics and Hydrodynamic Models of Fixed Bed Reactors, in Gianetto and Silveston (eds.), Multiphase Chemical Reactors, Hemisphere 1986] SOME CASE STUDIES 19-61 SOME CASE STUDIES The literature contains case studies that may be useful for analysis or design of new reactors Several of these are listed for reference Rase (Case Studies and Design Data, vol of Chemical Reactor Design for Process Plants, Wiley, 1977): • Styrene polymerization • Cracking of ethane to ethylene • Quench cooling in the ethylene process • Toluene dealkylation • Shift conversion • Ammonia synthesis • Sulfur dioxide oxidation • Catalytic reforming • Ammonia oxidation • Phthalic anhydride production • Steam reforming • Vinyl chloride polymerization • Batch hydrogenation of cottonseed oil • Hydrodesulfurization Rase (Fixed Bed Reactor Design and Diagnostics, Butterworths, 1990) has several case studies and a general computer program for reactor design: • Methane-steam reaction • Hydrogenation of benzene to cyclohexane • Dehydrogenation of ethylbenzene to styrene Tarhan (Catalytic Reactor Design, McGraw-Hill, 1983) has computer programs and results for these cases: • Toluene hydrodealkylation to benzene and methane • Phthalic anhydride by air oxidation of naphthalene • Trickle bed reactor for hydrodesulfurization Ramage et al (Advances in Chemical Engineering, vol 13, Academic Press, 1987, pp 193–266): • Mobil’s kinetic reforming model Dente and Ranzi [in Albright et al (eds.), Pyrolysis Theory and Industrial Practice, Academic Press, 1983, pp 133–175]: • Mathematical modeling of hydrocarbon pyrolysis reactions Shah and Sharma [in Carberry and Varma (eds.), Chemical Reaction and Reaction Engineering Handbook, Marcel Dekker, 1987, pp 713–721]: • Hydroxylamine phosphate manufacture in a slurry reactor Exploration for an acceptable or optimum design for a new reactor may require consideration of several feed and product specifications, reactor types, catalysts, operating conditions, and economic evaluations Modifications to an existing process likewise may need to consider many cases Commercial software may be used to facilitate examination of options A typical package can handle a number of reactions in various ideal reactors under isothermal, adiabatic, or heattransfer conditions in one or two phases Outputs can provide profiles of composition, pressure, and temperature as well as vessel size Thermodynamic software packages may be used to find equilibrium compositions at prescribed temperatures and pressures Such calculations require knowledge of feed components and products and their thermodynamic properties and are based on Gibbs free energy minimization techniques Examples of thermodynamic packages may be found in Smith and Missen (Chemical Reaction Equilibrium Analysis Theory and Algorithms, Wiley, 1982) and in Walas (Phase Equilibria in Chemical Engineering, Butterworths, 1985) For some widely practiced processes, especially in the petroleum industry, computer models are available from a number of vendors or, by license, from proprietary sources Such processes include fluid catalytic cracking, hydrotreating, hydrocracking, alkylation with HF or H2SO4, reforming with Pt or Pt-Re catalysts, tubular steam cracking of hydrocarbon fractions, noncatalytic pyrolysis to ethylene, and ammonia synthesis Catalyst vendors may sometimes also provide simple process models The reader is advised to peruse some of the process simulation packages listed for sale in the CEP Software Directory (e.g., AIChE, 1994) that gets periodically updated with new offerings This page intentionally left blank ... Vertical Kilns 19- 25 19- 27 19- 27 19- 27 19- 30 19- 33 19- 33 19- 36 19- 36 19- 36 19- 36 19- 36 19- 36 *The contributions of Stanley M Walas, Ph.D., Professor... Engineers) 19- 1 Copyright © 2008, 199 7, 198 4, 197 3, 196 3, 195 0, 194 1, 193 4 by The McGraw-Hill Companies, Inc Click here for terms of use 19- 2 REACTORS FLUID-FLUID REACTORS Gas-Liquid Reactors. .. Countercurrent Flow SOME CASE STUDIES 19- 49 19- 50 19- 53 19- 53 19- 56 19- 57 19- 57 19- 60 19- 60 REACTORS 19- 3 Nomenclature and Units In this section, the concentration is