Chemical Reactor Analysis and Design Fundamentals James B. Rawlings Department of Chemical Engineering University of Wisconsin Madison, Wisconsin John G. Ekerdt Department of Chemical Engineering The University of Texas Austin, Texas wey Publishing Madison, WisconsinThis book was set in Lucida using WIyX, and printed and bound by Worzalla Cover design by Cheryl M. and James B. Rawlings, and John W. Eaton. Copyright © 2002 Nob Hill Publishing, All rights reserved, Nob Hill Publishing Cheryl M. Rawlings, publisher Madison, WI 53705 nobhi11@g1obaldiatog.com http://www. global dialog. com/~nobhi1 No part of this book may be reproduced, in any form or by any means, without permission in writing from the publisher, Library of Congress Control Number: 2001135832 Rawlings, Janes B, Chemica) Reactor Analysis and Design Fundamentals / by James 8. Rawlings and John G, Ekerdt. p.m, Includes bibliographical references (p.) and index. ISBN 0-615-11884-4 (Toth) 1. Chemical reactors. I. Ekerdt, John G. IE. Title. Printed in the United States of America w98 7654324 To Cheryl, Melanie and Blake, and Carol, Alice and Barbara, for their love and encouragement.Preface Chemical reactors are at the core of the chemical engineering discipline, and chemical reactor analysis and design is one of the distinguishing courses that clearly separates the chemical engineers from the other engineering professionals. Given that chemical reactor analysis and design is a mature and stable topic in the curriculum of chemical engi: neering, however, it is natural to ask what is the motivation for a new text on this topic. We offer our motivations here. This book grew out of the combined experience of the two authors teaching this subject to undergradu- ates for more than 30 years. Given the rapidly changing landscape of scale and type of reactors of interest to practicing chemical engi- neers (chemical vapor deposition reactors, pharmaceutical fermentors, micro-reactors, as well as traditional catalytic crackers, bulk polymer- ization reactors, etc), it seems unwise to emphasize one industrial sec- tor and treat its reactor types in detail. Practicing chemical engineers work in a broad array of industrial sectors, and many will change sec tors during their carcers. IF chemical engineering has any important distinguishing characteristic, it is a set of fundamentals that apply to all scales and all types of reaction and transport processes involving ‘chemical change. ‘This bookiis all about reactor fundamentals. Rather than presenting ‘many facts about reactors, we focus on the framework for how to think about reactors — a framework for thinking that enables one, with some experience, to establish any of these facts for oneself, and discover new facts given new situations. All engineering and science textbooks do this to some extent; in this text, we will do it to a rather large extent. Computations matter in this subject. Reactor fundamentals, like the fundamentals in any subject, are few in number. But the diversity of the consequences of these fundamentals is enormous, Computational approaches provide a powerful and general approach to systematically investigating these conseque' hout making unrealistic simplify- ing assumptions. We attempt to exploit the significant advances in computing algorithms, software, and hardware in order to revise and n, Octave andManias, Octave Is freely av or a v an can be downloade rom mas eae a aerators cially available from The MathWorks, Inc, and is becoming a commonly available to] of industrial engineering practice, These laggucees alla us 10 focus on essential and gore programming deta whichis the Students should not feel compelled to recall the detatled informa- tion in the figures, but recall only the concepts, principles and results, Students will have the comput os to recreate the fig. ures in this or any other textbook on this subject. For example, al calculations required for the figures in this text were performed with Octave, Students should not feel compelled to memorize design equa tions for reactor types. The goal 1s to develop sufficient expertise so that students can set up appropriate models from the basic principles each nev problem they encounter. That is the time-tested way to till confidence that one can analyze a new situation, which we full expect to be the experience of practicing engineers. i Newly practicing engineers certainly will ned to learn the economics of manufacturing their main products, main reactor configurations, di talled energy-recovery schemes, contacting patterns, new catalysts, and Teactor monitoring and control systems. Rather than make inadequate efforts to include al of this information as a survey during the reactor design course, we feel this material deserves separate coverage, or can be learned during the early years of engineering practice. We do not ard it as poor preparation that some of th versity erica; rater, we reed earning art of the rocess gn te job a par ofthe maturation process ofthe practic profes Finally, we hope the text conveys some of the excitement that chem engineers feel for chemical reactions and chemical reactors. Under: eee oe ipitie to investigate new situations. Grounding he ud es you to explore with confidence JBR Madison, Wisconsin nia Austin, Texas Acknowledgments Greeters sec e cae EEE ECE ee We benefited greatly from the help and advice of many colleagues dur- ing the preparation of this text. Taking Warren Stewart's graduate reac tor modeling course greatly influenced the first author's way of thinking about this subject. More recently, Warren also provided-thany helpful comments on computational issues involving reaction-diffusion, and parameter estimation, Harmon Ray contributed many insightful and helpful comments on a variety of topics, especially on issues of reac tor mixing and industrial polymerization reactors. Bob Bird provided ‘many helpful comments on presenting the material related to transport phenomena, He also provided detailed feedback on a review copy of the text. ‘Marty Feinberg read an early version of the notes, and provided de- tailed feedback on many issues including reaction rates and presenting computational material. Yannis Kevrekidis used a preliminary version of the notes as part of a graduate reactor modeling course at Prince ton. His feedback and encouragement are very much appreciated. John Falconer also provided many helpful comments. ‘Tom Edgar provided much-needed departmental support and en- couragement when the authors began writing. Roger Bonnecaze, Bud. die Mullins and C. D. Rao used early versions of the notes and Oc- tave software in teaching the undergraduate course at The University of Texas, Their feedback was very helpful Several graduate students contributed to the software, which is an essential part of this text. Ken Muske developed some examples for the chapter on mixing. John Campbell worked on programs for man- aging thermochemical data, Sankash Venkatash contributed programs on parameter estimation and reaction-diffusion, and Dominik Nagel contributed programs to calculate potential-energy surfaces. Chris Rao provided helpful feedback on mixing issues, and Matt Tenny and Eric Haseltine provided helpful input on the cover design. prepare the solution manual. We have benefited from the feedback of many classes of undergraduates both at the University of Wisconsin and The University of Texas. We appreciate the enthusiasm with which they received some very early and incomplete drafts of the notes. ur publisher deserves special recognition for providing enoughflexibility so that we could realize our ing that we actually finish the project. Finally, we would like to acknowledge the essential contributions of John Eaton. When the two authors first outlined the goals of this writing Project, we envisioned creating in parallel a convenient computational environment. The original goal of the computational environment was fo enable students to quickly and conveniently define and solve new reactor design problems as a means to gain deeper understanding and appreciation of chemical reactor fundamentals. Starting from this sim ple idea, John Eaton created the Octave software project. We could not have written this text in the way we planned without Octave. Every fig. ure in the text requiring calculation was prepared with Octave. John contributed in a myriad of other ways. He designed the typesetting macros and the text version control system. We maintained more than. 10 versions of the source files on more than 7 different computing sys- tems at various stages of production. With John's expert guidance, we easily managed this complexity; in fact, the computing tools worked so well, that the computational aspects of the book production were enjoyable. original vision, while still insist- Contents 1 Setting the Stage 1.1 Introduction ....-. ++... paaeentid 1.2. Some Classifications and Terminology . . . . .-- 1.3 Scale . : 14 Some Examples... . 0-20. 0- 1.4.1. Chemical Vapor Deposition 14.2 Hydrodesulfurization 1.4.3. Olefin Polymerization 1.4.4. Hepatitis B Virus Modeling 1,5. An Overview of the Text... - 2. The Stoichiometry of Reactions PH Introduction vs sesecrvcee=: 2.2 Chemical Reactions and Stoichiometry 23 Independent Reactions 24 Reaction Rates and Production a 7 5 \spects of Stoichiometry . O51 Computing Production Rats irom Reaction Kt 2.5.2 Computing Reaction Rates from Production Rates 25:3 Measurement Errors and Least-Squares Estimation 2.6 Summary and No 27 Exercises - 3. Review of Chemical Equilibrium 3.1 Introduction . eee .2 Condition for Reaction Equilibr! ; °° SLL Evaation ofthe Gibbs Energy Change of Reaction 3.2.2. Temperature Dependence of the Gibbs Energy 3.3. Condition for Phase Equilibrium . i : 3.3.1 Ideal Mixtures . 3.3.2 Nonideal Mixtures 3.4 Heterogeneous Reactions 3.5. Multiple Reactions 7 3.5.1 Algebraic Approach 3.5.2. Optimization Approach a7 27 28 35 41 46 47 48 49 53 55 63 63 64 7 76 79 80 82 a2 88 88 92te Contents 3.6 Summary and Notation 94 3.7 Exercises. 98 4 The Material Balance for Chemical Reactors co 4.1 General Mole Balance . Feed ate PT 4.2. The Batch Reactor sees 112 4.2.1 Analytical Solutions for Simple Rate Laws ||| 113 4.3 The Continuous-Stirred-Tank Reactor (CSTR) . 129 44 The Semi-Batch Reactor... 2.0.0... .0.2.0..... 133 4.5 Volume Change Upon Reaction . . 133 4.6. The Plug-Flow Reactor (PFR) . Se eee ate IAG: 4.6.1 Thin Disk Volume Element ............... 146 4.6.2 Steady-State Operation 7 cesses 147 4.6.3, Volumetric Flowrate for Gas-Phase Reactions... 147 4.6.4 Volumetric Flowrate for Liquid-Phase Reactions .. 148 4.6.5. Single Reaction Systems ................. 149 4.6.6. Multiple-Reaction Systems 151 4.7 Some PFR-CSTR Comparisons . . e +. 159 4.8 Stochastic Simulation of Chemical Reactions sees 164 4.9 Summary and Notation 173 4.10 Exercises 76 5. Chemical Kinetics 189 5.1 Introduction . fa Sottatd sees 189 5.2 Elementary Reaction ... . 7 seeeees 190 5.3. Elementary Reaction Kinetics... settee eee 195 54 Fast and Slow Time Scales ......... 207 54.1 The Reaction Equilibrium Assumption ........ 209 5.4.2. The Quasi-Steady-State Assumption.......... 218 5.5. Rate Expressions eae 7 + 226 5.6 Reactions at Surfaces»... E se + 237 5.7 Summary and Notation 255 5.8 Exercises . EEE eer eect . 257 6 The Energy Balance for Chemical Reactors 275 6.1 General Energy Balance oar cence 275 6.1.1 Work Term. Sete eee eect a 6.1.2 EnergyTerms ...... Ra etary, 6.2 The Batch Reactor ................, wees 297 6.3 The CSTR 7 286 6.3.1 Steady-State Multiplicity 290 Contents 6.3.2 Stability of the Steady State . 7 6.3.3 Sustained Oscillations, Limit Cycles . 64 The Semi-Batch Reactor 6.5. The Plug-Flow Reactor 66 67 6.5.1. Plug-Flow Reactor Hot Spot and Runaway 6.5.2. The Autothermal Plug-Flow Reactor... .-- Summary and Notation . Exercises 26-5 + 7. Fixed-Bed Catalytic Reactors, 71 72 73 74 78 76 77 78 79 Introduction Catalyst Properties . . 7.2.1. Effective Diffusivity ‘The General Balances in the Catalyst Particle Single Reaction in an Isothermal Particle 7.4.1 First-Order Reaction in a Spherical Particle . 7.4.2. Other Catalyst Shapes: 7.4.3 Other Reaction Orders . 7.44 Hougen-Watson Kinetics 7.45. External Mass Transfer 7/416 Observed versus Intrinsic Kinetic Parameters. Nonisothermal Particle Considerations . Multiple Reactions . . Fixed-Bed Reactor Design . 7.7.1 Coupling the Catalyst and Fluid ‘Summary and Notation Exercises, 8 Mixing in Chemical Reactors 81 82 83 Introduction . Residence-Time Distribution 8.2.1 Definition 8.2.2. Measuring the RTD 823 Comtinuous-Stirred-Tank Reactor (CSTR) 2.5 CSTRs in Series 8.2.6 Dispersed Plug Flow Limits of Reactor Mixing. 8.3.1 Complete Segregation . 8.3.2 Maximum Mixedness 83.3 Mass Transfer and Limits of Reactor Mi 'ylinders and Slabs 8.2.4 Plug-Flow Reactor (PFR) and Batch Reactor nexiv Contents 84 Limits of Reactor Performance... .. 469 8.4.1 A Single Convex (Concave) Reaction Rate 369 84.2. The General Case 474 8.5. Examples in Which Mixing is Critical 479 8.6 Summary and Notation 486 87 Exereees ‘490 9 Parameter Estimation for Reactor Models 501 9.1. Experimental Methods . 501 9.1.1 Analytical Probes for Concentration 501 9.1.2. Experimental Reactors for Kinetics Studies 306 9.1.3. Characterizing Catalysts and Surfaces . . + 510 9.2 Data Modeling and Analysis. . 313 9.2.1. Review of the Normal Distribution ili 9.2.2 Kigenvalues and Eigenvectors 515 9.2.3. Least-Squares Estimation . a 2317 9.24 Least Squares with Unknown Variance... . 523 9.25 Nonlinear Least Squares ...... 4... © 587 9.26 Design of Experiments 1... 1 532 92.7 Parameters imation with Differential Equation fodels 5 9.3. An Industrial Case Study... “551 9A Summary and Notation ‘362 9.5. Exercises 565 ‘A Computational Methods S75 Ad Linear Algebra and Least Squares. ; 375 A.2 Nonlinear Algebraic Equations and Optimization 378 A2.1. Defining Functions (function) 578 A22 Solving Nonlinear Algebraic Equations (Fsolve) .. 579 2.3 Solving Nonlinear Optimization Problems (npsol) 580 A3 Differential Equations (1sode) 582 A3.1_ Differential Algebraic Equations (dass1) 384 3.2 Automatic Stopping Times (ddasrt) 584 A Sensitivities of Differential Equations (ddasac) 585 A5 Boundary-Value Problems and Collocation (col1ec).. .. 589 ‘395 Index 599 List of Figures ee ee {1 Schematic diagram of a batch reactor. « 4 112. Expanded view of the internals of a batch reactor... 4 13. Schematic diagram of aCSTR.... + : 5 1.4 Schematic diagram of five CSTRs in werles sesso 6 6 7 1.5 Schematic diagram of a plug-flow reactor. : 16 Cross-sectional view of a1 yl combinatorial screening reactor 17 Assembled view of 256 I-uL reactors built into a 3-1n by 3.in support wafer ais 18 Top view of a combinatorial reactor gas-distribution network 9 1.9 Polyethylene reactor 10 Top view ofa singlewafer CVD reactor... eet LLL Representative organosulfur compounds . . - Eee: 1.12 Hydrodesulfurization process. . - Eee 13 Cross-sectional view ofa commercial HDS reactor 15 1.14 Expanded view of a multi-bed HDS reactor 16 1.15 Simplified polymerization process... +++ +--+ v7 1.16 The chemical events comprising the reproduction cycle of the hepatitis B viru 9 2: Defining the reaction rate, 7, for the Siz reaction ...... 42 22 Estimated reaction rates from six production-ate mea surements subject to measurement noise. sz 2.3. Estimated reaction rates from 500 production-rate mea- surements subject to measurement noise. 32 3.1. ‘Thermodynamic system. - oie ea 3.2. Gibbs energy versus reaction extent at constant T and P... 65 3.3. Gibbs energy G versus reaction extent &". - - - - 75 3.4 Partial pressures of components A and C versus ‘guid: 84 phase composition in a nonideal solution « 3.5. Gibbs energy versus two react mnextents at constant T and P. 89 3.6. Gibbs energy contours for the pentane reactions as a func:avi List of Figures 3.7 Two systems with identical intensive properties and dif- ferent extensive properties. 3.8 Vessel with semipermeable membrane and chemical re- action 2A=B.. 4.1 Reactor volume element, 4.2 Batch reactor volume element. 43. Firstorder, irreversible kinetics in a batch reactor, . 45 Birstorder, reversible kinetics ina batch reactor, ky = 1, ky = 0.5, c40~ 1, 50 =0.. 4.6 Second-order and first-order kinetics in a batch reactor, cao=1k=1 Reaction rate versus concentration for nth order kinetics, r=kelnz0.. ‘ 48 Batch reactor with nth-order kinetics, y = ke’, ko ~ kets} nel. 4.9 Batch reactor with nth-order kineties,y = ke ky ~ke’sl = nel, : 4410 Reaetion rateversus ‘oncentration for ntivorder kinetics, = keh, n <0. 4.11 Two fist order reactions in series in bate reactor 4.12 Two first-order reactions in parallel in a batch reactor 4.13 CSTR volume element. 4.14 Reaching steady state in a CSTR. 4.15 Semi-batch reactor volume for different monomer addi tion policies 4.16 Semi-batch reactor feed flowrate for different monomer addition policies . 4.17 Semi-batch reactor monomer content for different monomer addition policies ; 4.18 Semi-batch reactor polymer content for different monomer addition policies : Plug-flow reactor volume element. ; Benzene conversion versus reactor volume: Component mole fractions versus reactor volume Molar flowrate of ethane, ethylene and NO versus reactor volume for ethane pyrolysis example, . 4.23 Molar flowrate of ethane versus reactor volume for inlet temperatures of 1000, 1050 and 1100 k - 100 - 102 ua 113 i - 5 4A First-order, irreversible kinetics in a batch reactor, log scale, 115, List of Figures 4.24 The PER is smaller than the CSTR for irreversible, nth- order kinetics, positive order, n > 0 4425 The CSTR is smaller than the PFR for ireversible, mth- order kinetics, negative order, <0 ......2+ +5 4.26 PER versus CSTR with recycle and separation. . . 4.27 Overall and per-pass conversion of Aas a function of frac- tional recycle, o « . 7 4.28 Choosing a reaction with appropriate probability +” 4.29 Stochastic simulation of the first-order reactions A-B—-C starting with 100 A molecules. 4,30 Stochastic simulation of the first-order reat starting with 1000 A molecules. ..---. 0.0. 002255 4.31 Stochastic simulation of the first-order reactions A—B—C starting with 4000 A molecules. ....-. « 44.32 Species cecDNA versus time for hepatitis B virus model; deterministic and average stochastic models. . . 4.33 Species rcDNA versus time for hepatitis B virus model, deterministic and average stochastic models. 1434 Envelope versus time for hepatitis B virus model; deter ministic and average stochastic models. . . 4.35 Species cccDNA versus time for hepatitis B virus model; ‘two representative stochastic trajectories = 4,36 An alternative CSTR-separator design. : : 4.37 Deterministic simulation of reaction A + B = C com: pared to stochastic simulation . 5.1 Morse potential for Hy and HE. 5.2 Potential-energy surface for the F, Hy re 5.3 Contour representation of the potential 5.4 Reaction-coordinate diagram. 55 Comparison of measured and calculated rate constant versus temperature for oxirane decomposition. 5.6 Full model solution for ky = 1,k-1 =0.5,k2 = k.2 = W 5.7 Full model solution for ky ~ 1,k-1 = 0.5,kz = k.2 = 10, 5.8 Concentrations of B and C versus time for full model with increasing kz with Ko = ko/k.2 = 1. 5.9 Comparison of equilibrium assumption to full model for ky = 1-1 = 0.5,k2 = ko = 10. 5.10 Normalized concentration of versus dimensionless time for the series reaction A ~ B = C nergy surface. ail - 160 ~ 161 162 163 167 168 . 169 169 . 7 7 172 183 185 196 . 197 198 199 + 208 ~ 210 210 . 2 ais 221