integrated chemical processes synthesis operation analysis and control

Preface XV Part I Integration of Heat Transfer and Chemical Reactions 1 1 Enhancing Productivity and Thermal Efficiency of High-Temperature Endothermic Processes in Heat-Integrated Fixed

Integrated Chemical Processes

Edited by Kai Sundmacher, Achim Kienle, Andreas Seidel-Morgenstern

Integrated Chemical Processes Edited by K Sundmacher, A Kienle and A Seidel-Morgenstern

Copyright © 2005 WILEY-VCH Verlag GmbH & Co KGaA, Weinheim

Further Titles of Interest

K Sundmacher, A Kienle (Eds.)

Reactive DistillationStatus and Future Directions 2003

ISBN 3-527-30579-3

J G Sanchez Marcano, T T Tsotsis

Catalytic Membranes and Membrane Reactors2002

ISBN 3-527-30277-8

T G Dobre, J G Sanchez Marcano

Chemical EngineeringModelling, Simulation and Similitude 2005

Integrated Chemical Processes

Synthesis, Operation, Analysis, and Control

Edited by

Kai Sundmacher,

Achim Kienle and

Andreas Seidel-Morgenstern

Prof Dr.-Ing Kai Sundmacher Prof Dr.-Ing Achim Kienle Prof Dr Andreas Seidel-Morgenstern

Max Planck Institute for Dynamics of Complex Technical Systems

Sandtorstr 1

39106 Magdeburg Germany

and

Otto-von-Guericke-University Magdeburg

Universitätsplatz 2

39016 Magdeburg Germany

All books published by Wiley-VCH are carefully produced Nevertheless, authors, editors, and publisher do not warrant the information contained

in these books, including this book, to be free of errors Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

Library of Congress Card No.: applied for.

British Library Cataloguing-in-Publication Data:

A catalogue record for this book is available from the British Library

Bibliographic information published by Die Deutsche Bibliothek

Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available in the Internet at

Printed in the Federal Republic of Germany.

Printed on acid-free paper

Typesetting Shiv e-Publishing Technologies Pvt Ltd.

Bangalore, India

Printing betz-druck GmbH, Darmstadt Bookbinding Schäffer GmbH i.G., Grünstadt

ISBN-13: 987-3-527-30831-6 ISBN-10: 3-527-30831-8

Preface XV

Part I Integration of Heat Transfer and Chemical Reactions 1

1 Enhancing Productivity and Thermal Efficiency of High-Temperature

Endothermic Processes in Heat-Integrated Fixed-Bed Reactors 3

Grigorios Kolios, Achim Gritsch,

Bernd Glöckler and Gerhart Eigenberger

1.3.2.1 Processes for Large-Scale Applications 29

1.3.2.2 Processes for Small-scale Applications 31

1.4 Conclusions 39

Symbols and Abbreviations 39

References 41

2 Conceptual Design of Internal Reforming in High-Temperature Fuel Cells 45

Peter Heidebrecht and Kai Sundmacher

2.4.3 Anode Exhaust Gas Recycling 63

2.5 Summary and Conclusions 65

Symbols 66

References 67

3 Instabilities in High-Temperature Fuel Cells due to Combined Heat

and Charge Transport 69

Michael Mangold, M Krasnyk, Achim Kienle and Kai Sundmacher

3.3.1 Cell with Infinite Length 74

3.3.2 Cell with Finite Length 76

Part II Integration of Separations and Chemical Reactions 85

4 Thermodynamic and Kinetic Effects on the Feasible Products of Reactive

Distillation: A-zeo-tropes and A-rheo-tropes 87

Kai Sundmacher, Zhiwen Qi, Yuan-Sheng Huang and Ernst-Ulrich Schlünder

4.1 Introduction 87

4.2 Azeotropes 89

4.2.1 Reactive Condenser and Reboiler 89

4.2.2 Conditions for Singular Points 90

4.2.2.1 Potential Singular Point Surface 90

4.2.2.2 Reaction Kinetic Surface 91

4.2.3 Examples 92

4.2.3.1 Hypothetical Ternary Systems 92

4.2.3.2 Real Ternary System: MTBE-Synthesis 97

4.2.3.3 Real Ternary System with Phase Splitting: Methanol Dehydration 101

4.2.3.4 Real Quaternary System: Isopropyl Acetate Hydrolysis 103

4.2.4 Application of Feasibility Diagram: Column Feasible Split 106

4.2.5 Remarks on Azeotropes 110

4.3 Arheotropes 110

4.3.1 Definition and Conditions 110

4.3.2 Illustrative Examples 111

4.3.2.1 Example 1: Stagnant Sweep Gas 111

4.3.2.2 Example 2: Flowing Sweep Gas 114

4.3.2.3 Example 3: Flowing Sweep Gas with Pervaporation 117

4.3.2.4 Example 4: Reactive Liquid Mixture 119

4.3.3 Remarks on Arheotropes 126

4.4 Kinetic Arheotropes in Reactive Membrane Separation 127

4.4.1 Model Formulation 127

4.4.1.1 Reaction Kinetics and Mass Balances 127

4.4.1.2 Kinetics of Vapor Permeation 129

4.4.2 Residue Curve Maps 130

4.4.2.1 Example 1: Ideal Ternary System 130

4.4.4 Remarks on Kinetic Arheotropes 144

4.5 Summary and Conclusions 144

Symbols and Abbreviations 145

References 147

5 Equilibrium Theory and Nonlinear Waves for

Reaction Separation Processes 149

Achim Kienle and Stefan Grüner

5.1 Introduction 149

5.2 Theoretical Background 150

5.2.1 Wave Phenomena 150

5.2.2 Mathematical Model 153

5.2.3 Prediction of Wave Patterns 157

5.3 Analysis of Reaction Separation Processes 161

5.3.1 Reactive Distillation 161

5.3.2 Chromatographic Reactors 163

5.3.2.1 Reactions of Type 164

5.3.2.2 Reactions of type 166

5.3.2.3 Binaphthol Separation Problem 168

5.3.3 Extension to Other Processes 171

5.4 Applications 172

5.4.1 New Modes of Operation 172

5.4.2 New Control Strategies 173

5.5 Conclusion 175

Acknowledgments 177

Symbols 178

References 179

6 Simulated Moving-Bed Reactors 183

Guido Ströhlein, Marco Mazzotti and Massimo Morbidelli

6.1 Introduction 183

6.2 Continuous Reactive Chromatography 185

6.2.1 Annular Reactive Chromatography 185

6.2.2 Simulated Moving-Bed Reactors 185

6.3 Design Parameters for Simulated Moving-bed Reactors 188

6.4 Modeling Simulated Moving-bed Reactors 195

6.5 Influence of the Stationary Phase Properties on SMBR Efficiency 197

7.2.1 The Claus Process 207

7.2.2 Direct Hydrogen Cyanide Synthesis 208

7.2.3 Water-gas Shift Reaction 210

7.2.4 The Deacon Process 211

7.3 Catalyst and Adsorbent 212

7.3.1 The Claus Process 212

7.3.2 Direct Hydrogen Cyanide Synthesis and Water-gas Shift Reaction 214

7.3.3 The Deacon Process 217

7.3.4 Other Adsorptive Catalysts 217

7.4 Reactor and Regeneration 218

7.4.1 Fixed-bed Reactors 218

7.4.2 Fluidized-bed Reactors 219

7.4.3 Pressure Swing Regeneration 220

7.4.4 Temperature Swing Regeneration 221

8 Reactive Stripping in Structured Catalytic Reactors: Hydrodynamics

and Reaction Performance 233

Tilman J Schildhauer, Freek Kapteijn, Achim K Heibel,

Archis A Yawalkar and Jacob A Moulijn

8.2.2 Hold-up, Pressure Drop, and Flooding Limits 242

8.2.3 Residence Time Distribution 244

8.2.4 Gas–Liquid Mass Transfer 247

8.3.3.1 Effect of Water Removal 253

8.3.3.2 Co-current versus Countercurrent 254

8.3.3.3 Selectivity 255

9.3.2 Equilibrium Stage Model 270

9.3.3 HTU/NTU-concept and Enhancement Factors 271

9.3.4 Rate-based Stage Model 272

11 Development of Reactive Crystallization Processes 339

Christianto Wibowo, Vaibhav V Kelkar,

Ketan D Samant, Joseph W Schroer and Ka M Ng

11.1 Introduction 339

11.2 Workflow in Process Development 339

11.3 Process Synthesis 341

11.4 Reactive Phase Diagrams 344

11.4.1 Projections and Canonical Coordinates 344

11.4.2 High-Dimensional Phase Diagrams 346

11.5 Kinetic Effects 351

11.5.1 Reactive Crystallization with a Solid Reactant 351

11.6 Asymmetric Transformation of Enantiomers 353

12.2.3 Selectivity and Yield 364

12.3 Mass Transfer through Membranes 366

12.4 Kinetic Compatibility in Membrane Reactors 368

12.5 Example 1: Product Removal with Membranes

(“Extractor”) 369

12.5.1 Model Reaction, Procedures and Set-up 370

12.5.2 Reaction Rates 370

12.5.3 Fixed-bed Reactor Experiments 372

12.5.4 Mass Transfer through the Membrane 373

12.5.5 Membrane Reactor Experiments and Modeling 374

12.5.6 Evaluation of the Extractor-type of Membrane Reactor 376

12.6 Example 2: Reactant Dosing with Membranes

Part III Integration of Mechanical Operations and Chemical Reactions 391

13 Reactive Extrusion for Solvent-Free Processing: Facts and Fantasies 393

Leon P B M Janssen

Abstract 393

13.1 Introduction 393

13.2 Advantages and Disadvantages 394

13.3 Main Reactions in Extruders 395

14.2 Mechanical Comminution of Solids 408

14.2.1 Reactivity of Mechanically Activated Solids 413

14.3 Equipment and Processes 415

14.3.1 Milling Designs 415

14.3.1.1 Crushers 415

14.3.1.2 Cutting Mills 415

14.3.1.3 Roller and Ring-roller Mills 415

14.3.1.4 Disk Attrition Mills 416

14.3.1.5 Grinding Media Mills 416

14.3.1.6 Impact Mills 419

14.3.1.7 Jet Mills 420

14.3.1.8 Pressure and Shear Devices 421

14.3.1.9 Special Milling and Activation Devices 421

14.3.1.10Dry and Wet Grinding 422

15.3 Separation of Particulates and Reaction of Solids 441

15.3.1 Diesel Soot Abatement 441

15.3.1.1 Filter Types and Catalyst Performance 443

15.3.1.2 Formal Kinetics and Modeling 445

15.3.2 Filtration Combustion of Solid Fuels 447

15.3.3 Reaction Cyclone 448

15.4 Conclusions 449

References 450

16 Reaction-Assisted Granulation in Fluidized Beds 453

Matthias Ihlow, Jörg Drechsler, Markus Henneberg, Mirko Peglow, Stefan Heinrich and Lothar Mörl

16.1 Introduction 453

16.1.1 Gas Cleaning 453

16.1.2 Application of the Gas-solid Fluidized Bed 455

16.1.3 Modeling and Experimental Set-up 456

16.2 Modeling 460

16.2.1 Model Assumptions 461

16.2.2 Phase and Reaction Equilibrium 462

16.2.3 Mass and Energy Flows 463

16.2.3.1 Gas Balance 463

16.2.3.2 Suspension Balance 467

16.2.3.3 Gas–Liquid Phase Boundary 471

16.2.3.4 Particle Balance 474

16.2.3.5 Apparatus Wall Balance 477

16.2.3.6 Solution of the Equations 477

16.3.2.2 Influence of the Liquid Injection Rate 481

16.3.2.3 Influence of Ca/S Ratio 488

16.3.2.4 Influence of the Gas Inlet Temperature 493

16.3.2.5 Influence of Gas Mass Flow 495

16.3.2.6 Influence of the Particle Diameter 497

16.3.3 Experiments for Reactive Absorption with Overlapped

Granulation of Solid 502

16.3.3.4 Experimental Realization 502

16.3.3.2 Batch Experiments 503

16.3.3.3 Continuous Experiments 510

16.3.4 Comparison between Measurement and Simulation 519

16.3.5 Simplified Stationary Balancing 519

16.4 Conclusions 523

Symbols and Abbreviations 526

References 529

Index 543

In the chemical industries, the pretreatment of educts, their chemical conversioninto valuable products, and the purification of resulting product mixtures in down-stream processes are carried out traditionally in sequentially structured trains of unitoperations In many cases, the performance of this classical chemical process struc-ture can be significantly improved by an integrative coupling of different processunits

The integration of unit operations to form multifunctional processes very oftengives rise to synergetic effects which can be technically exploited By suitable processdesign, an efficient and environmentally benign process operation can be achieved.Possible advantages of process integration include:

• higher productivity;

• higher selectivity;

• reduced energy consumption;

• improved operational safety; and

• improved ecological harmlessness by avoidance of auxiliary agents andchemical wastes

Due to the interaction of several process steps in one apparatus, the steady-stateand the dynamic operating behavior of an integrated process unit is often muchmore complex than the behavior of single, non-integrated units Therefore, suitablemethods for the design and control must be developed and applied, ensuring opti-mal and safe operation of the considered integrated process

The major objectives of current research activities in this highly interestingdomain of chemical engineering are to develop new concepts for process integration,

to investigate their efficiency, and to make them available for technical application.The importance of this field is reflected by the increasing number of articles in jour-nals and book contributions that have been published during the past three decades(Fig 1)

Among these published articles and books, some excellent reviews have appearedwhich focused on specific aspects of the process integration Agar and Ruppel [1]were among the first to investigate the whole area of integration of heat-exchangingfunctions in chemical reactors, whilst Agar [2] later also surveyed other innovativeintegration concepts in chemical reactor engineering According to the presenteditors’ knowledge, the first review which covered a broader range of integrationconcepts including heat exchange, separation and also mechanical unit operations,was published in 1997 by Hoffmann and Sundmacher [3] The cited works refer tointegrated chemical processes as “multifunctional reactors”, which is often used as a

Integrated Chemical Processes Edited by K Sundmacher, A Kienle and A Seidel-Morgenstern

Copyright © 2005 WILEY-VCH Verlag GmbH & Co KGaA, Weinheim

synonym Multifunctional reactors can be seen as a very important sub-class of thearea of “process intensification” which was summarized by Stankiewicz andMoulin [4].

A comprehensive volume covering all aspects of integrated chemical processesincluding heat exchange, separations and mechanical unit operations is still miss-ing, however, and as a consequence the present book was prepared to fill this gap.The book’s chapters have been authored by leading international experts, and pro-vide overviews on the present state of knowledge and on challenging future issues.The book is divided into three parts Part I surveys concepts for heat-integratedchemical reactors, with special focus on coupling reactions and heat transfer in fixedbeds and in fuel cells Part II is dedicated to the conceptual design, control and analysis

of chemical processes with integrated separation steps, whilst Part III focuses on howmechanical unit operations can be integrated into chemical reactors

Part I:

Integration of Heat Transfer and Chemical Reactions

Chapters 1 to 3 discuss two recent and important applications of heat-integratedchemical reactions Chapter 1, by Kolios et al., is concerned with high-temperature

endothermic processes in heat integrated fixed-bed reactors Emphasis is placed on

reforming processes, which are widely used for the production of basic chemicalsand fuels from fossil feed stocks These processes require large amounts of heat attemperatures up to 1000 °C In conventional solutions, only about half of the heatsupplied at high temperatures is transferred into the endothermic reaction Emerg-ing applications such as decentralized hydrogen production for residential andmobile power generation require considerable improvement in specific productivityand thermal efficiency Therefore, this topic is currently the subject of vivid researchactivities in industry and academia alike Chapter 1 also includes an introduction to

Fig 1. Journal publications on integrated chemical processes according to the Science Citation Index.

the fundamentals of heat-integrated processes, an overview on recent trends in cess and apparatus design, and an analysis of the state of the art, with specialemphasis on the steam reforming of methane.

pro-The focus in Chapters 2 and 3 is on high-temperature fuel cells with internal ing In particular, special attention is given to the Molten Carbonate Fuel Cell

reform-(MCFC) which is increasingly used for decentralized power generation In Chapter

2, Heidebrecht and Sundmacher use a simple model of an MCFC to discuss the prosand cons of alternative reforming concepts in high-temperature fuel cells

The temperature management in a fuel cell stack is a key issue in the operation of

high-temperature fuel cells In Chapter 3, prepared by Mangold and colleagues, it isshown that the temperature-dependence of the electrolyte’s electrical conductivity is

a potential source of instabilities, hot spots, and spatial temperature patterns

In Chapter 4, Sundmacher et al – in the first contribution – analyze in detail the

thermodynamic and kinetic effects relevant to an understanding of reactive tion processes Although a comprehensive volume on this type of process integration was published in 2003 [6], Chapter 4 focuses on the a priori determination of prod-

distilla-ucts that can be obtained using such processes

In exploiting the equilibrium theory, Kienle and Grüner present in Chapter 5 a

general analysis of the development and propagation of nonlinear waves in reaction

separation processes Besides considering reactive distillation as one example, theseauthors also analyze reactive chromatography

In Chapter 6, Morbidelli et al describe chromatographic separations combined with chemical reactions, the focus of their contribution being to present possibilities of

performing such processes in a continuous manner

An analysis of reactors where adsorbents are used as a regenerative source or sink for

one or several of the reactants is discussed systematically by Agar, in Chapter 7

In cases where reactive distillation cannot be applied because some of the

reac-tants are temperature-sensitive, reactive stripping might be an efficient alternative,

and the current state of the application of this technology is reviewed by Kapteijnand colleagues in Chapter 8

Another powerful concept is to combine absorption processes with chemical tions, and a large number of possible concepts for this approach is presented in

reac-Chapter 9 by Kenig and Górak

In addition, extraction processes can be performed with reacting species, and

several advantages of this technique may be realized compared to conventionalconsecutive processes, as discussed by Bart in Chapter 10

Based on a thorough analysis of reactive crystallization, Ng and colleagues,

in Chapter 11, demonstrate that such integrated processes can also be performedefficiently with solid phases involved

In the final chapter of Part II, Seidel-Morgenstern presents two examples of how

membrane reactors might become an alternative to conventional technology.

dates that reactive extrusion has emerged from a scientific curiosity to an industrial

process Nonlinear effects in this process can give rise to instabilities that are ofthermal, hydrodynamic, or chemical origin

In Chapter 14, Hoffmann and colleagues provide a survey on the status and

direc-tions of reactive comminution In this type of process integration, mechanical stress

exerted in mills is used to enhance the chemical reactions of solids with fluids.Simultaneously, chemical reactions can generate cracks in solid particles andthereby enhance their comminution

Filtration and chemical reactions can be usefully integrated in order to separate

diesel soot particles efficiently from motor exhaust gases, and this is illustrated byRieckmann and Völker in Chapter 15, together with a series of other examples ofreactive filtration processes which are realized in the chemical industries

In the final chapter, Mörl and coworkers analyze the complex interaction of

particle granulation and/or agglomeration with chemical reactions in fluidized beds.

For the description of the particle property distribution, a population balanceapproach is recommended which is mathematically challenging but which providesvaluable insight into the steady-state and dynamic process operating behavior

The Book’s History, and the Editors’ Acknowledgments

The present book is the outcome of the International Max Planck Symposium onIntegrated Chemical Processes held in Magdeburg, Germany, on 22–24 March,

2004 At this symposium, renowned scientists met to discuss the current state andfuture trends in the field of integrated chemical processes The conference wasorganized by this book’s editors and their colleagues at the Max Planck Institutefor Dynamics of Complex Technical Systems, with financial support from theKompetenznetz Verfahrenstechnik Pro3 e.V in Germany, which is gratefullyacknowledged

The present editors wish to thank their colleagues Kristin Czyborra, Anett Raasch,and Carolin Apelt for their excellent support in organizing the symposium, and in

collecting the manuscripts which form the basis of this book Last – but not least –

we are thankful to Dr Hubert Pelc and Rainer Münz from Wiley-VCH for theirhelpful assistance during the book’s preparation

References

1.D W Agar, W Ruppel,

Multifunktionelle Reaktoren

für die heterogene Katalyse,

Chem Ing Tech

1988, 60, 731–741.

2.D W Agar, Multifunctional Reactors:

Old Preconceptions and New

Dimensions, Chem Eng Sci

Process Intensification, Chem

Eng Progress, 2000, Jan., 22–34.

5.S Kulprathipanja, (Ed.) Reactive

Separation Processes, Taylor & Francis,

New York, 2002.

6.K Sundmacher, A Kienle, (Eds.),

Reactive Distillation, Status and Future Directions, Wiley-VCH,

Weinheim, 2003.

List of Contributors

Editors

Prof Dr.-Ing Achim Kienle

Max Planck Institute for Dynamics of

Complex Technical Systems

Prof Dr.-Ing Andreas Seidel-Morgenstern

Max Planck Institute for Dynamics of Complex

Sandtorstr 1

39106 Magdeburg Germany and Otto-von-Guericke-University Magdeburg Process Systems Engineering

Universitätsplatz 2

39106 Magdeburg Germany

Authors

Prof Dr David W Agar

University of Dortmund

Institute of Chemical Reaction Engineering

Department of Biochemical and

Chemical Engineering

Emil-Figge-Str 70

44227 Dortmund

Germany

Prof Dr Hans-Jörg Bart

Technische Universität Kaiserslautern

Steinfeldstr 5

39176 Barleben Germany Prof Dr.-Ing Gerhart Eigenberger University of Stuttgart

Institute for Chemical Process Engineering Böblinger Str 72

70199 Stuttgart Germany

Bernd Glöckler University of Stuttgart Institute for Chemical Process Engineering Böblinger Str 72

70199 Stuttgart Germany Prof Dr Andrzej Górak University of Dortmund Department of Biochemical and Chemical Engineering Emil-Figge-Str 70

44227 Dortmund Germany Achim Gritsch University of Stuttgart Institute for Chemical Process Engineering Böblinger Str 72

70199 Stuttgart Germany Stefan Grüner University of Stuttgart Institute for System Dynamics and Control Engineering

Pfaffenwaldring 9

70569 Stuttgart Germany Achim K Heibel Delft University of Technology Reactor and Catalysis Engineering Julianalaan 136

2628 BL Delft The Netherlands Dr.-Ing Peter Heidebrecht Otto-von-Guericke-University Magdeburg Process Systems Engineering

Universitätsplatz 2

39106 Magdeburg Germany Jun.-Prof Dr.-Ing Stefan Heinrich Otto-von-Guericke-University Magdeburg Institute of Process Equipment and Environmental Technology Universitätsplatz 2

39106 Magdeburg Germany Dr.-Ing Markus Henneberg AVA – Anhaltinische Verfahrens- und Anlagentechnik Ingenieurgesellschaft Henneberg & Partner Steinfeldstr 5

39176 Barleben

Prof Dr Ulrich Hoffmann Technische Universität Clausthal Institut für Chemische Verfahrenstechnik Leibnizstr 17

38678 Clausthal-Zellerfeld Germany

Dr Christian Horst Technische Universität Clausthal Institut für Chemische Verfahrenstechnik Leibnizstr 17

38678 Clausthal-Zellerfeld Germany

Yuan-Sheng Huang Max Planck Institute for Dynamics of Complex Technical Systems

Sandtorstr 1

39106 Magdeburg Germany Dr.-Ing Matthias Ihlow AVA – Anhaltinische Verfahrens- und Anlagentechnik

Ingenieurgesellschaft Henneberg & Partner Steinfeldstr 5

39176 Barleben Germany Prof Dr Leon P.B.M Janssen University of Groningen Department of Chemical Engineering Nijenborgh 4

9747 AG Groningen The Netherlands Prof Dr Freek Kapteijn Delft University of Technology Reactor and Catalysis Engineering Julianalaan 136

2628 BL Delft The Netherlands Vaibhav V Kelkar ClearWaterBay Technologies Inc.

20311 Valley Blvd., Suite C Walnut, CA 91789 USA

Dr Eugeny Y Kenig University of Dortmund Department of Biochemical and Chemical Engineering Emil-Figge-Str 70

44227 Dortmund Germany

Prof Dr.-Ing Achim Kienle

Max Planck Institute for Dynamics of

Complex Technical Systems

Dr.-Ing Grigorios Kolios

Christ Pharma & Life Science AG

Hauptstr 192

4147 Aesch

Switzerland

Mykhaylo Krasnyk

Max Planck Institute for Dynamics of

Complex Technical Systems

Sandtorstr 1

39106 Magdeburg

Germany

Prof Dr Ulrich Kunz

Technische Universität Clausthal

Institut für Chemische Verfahrenstechnik

Leibnizstr 17

38678 Clausthal-Zellerfeld

Germany

Dr.-Ing Michael Mangold

Max Planck Institute for Dynamics of

Complex Technical Systems

Sandtorstr 1

39106 Magdeburg

Germany

Prof Dr Marco Mazzotti

Swiss Federal Institute of Technology Zürich

Institut für Verfahrenstechnik

Sonneggstr 3

8092 Zürich

Switzerland

Prof Dr Massimo Morbidelli

Swiss Federal Institute of Technology Zürich

Institut für Chemie- und

Universitätsplatz 2

39106 Magdeburg Germany Prof Dr Jacob A Moulijn Delft University of Technology Reactor and Catalysis Engineering Julianalaan 136

2628 BL Delft The Netherlands Prof Dr Ka M Ng Hong Kong University of Science and Technology Department of Chemical Engineering

Clear Water Bay Kowloon, Hong Kong China

Mirko Peglow Fraunhofer Institute for Factory Operation and Automation IFF Magdeburg

Product Design and Modelling Group Sandtorstr 22

39106 Magdeburg Germany

Dr Zhiwen Qi Max Planck Institute for Dynamics of Complex Technical Systems Sandtorstr 1

39106 Magdeburg Germany Prof Dr.-Ing Thomas Rieckmann University of Applied Sciences Cologne Institute of Chemical Engineering and Plant Design

Betzdorfer Str 2

50679 Köln Germany

Dr Ketan D Samant ClearWaterBay Technology Inc.

20311 Valley Blvd., Suite C Walnut, CA 91789 USA

Dr Tilman J Schildhauer Delft University of Technology Reactor and Catalysis Engineering Julianalaan 136

2628 BL Delft

Prof Dr.-Ing Dr h c mult Ernst-Ulrich Schlünder (emeritus)

University of Karlsruhe Institute of Thermal Process Engineering Kaiserstr 12

76128 Karlsruhe Germany

Dr Joseph W Schroer ClearWaterBay Technology Inc.

20311 Valley Blvd., Suite C Walnut, CA 91789 USA

Prof Dr.-Ing Andreas Seidel-Morgenstern Max Planck Institute for Dynamics of Complex Technical Systems Sandtorstr 1

39106 Magdeburg Germany and Otto-von-Guericke-University Magdeburg Chair of Chemical Process Engineering Universitätsplatz 2

39016 Magdeburg Germany Guido Ströhlein Swiss Federal Institute of Technology Zürich Institut für Chemie- und

Bioingenieurwissenschaften ETH-Hönggerberg, HCI-F

8093 Zürich Switzerland

Prof Dr.-Ing Kai Sundmacher Max Planck Institute for Dynamics of Complex Technical Systems Sandtorstr 1

39106 Magdeburg Germany and Otto-von-Guericke-University Magdeburg Process Systems Engineering

Universitätsplatz 2

39106 Magdeburg Germany Dr.-Ing Susanne Völker

42 Engineering – Chemical Engineering Consulting Services

von-Behring-Str 9

34260 Kaufungen Germany

Dr Christianto Wibowo ClearWaterBay Technology Inc.

20311 Valley Blvd., Suite C Walnut, CA 91789 USA

Dr Archis A Yawalkar Delft University of Technology Reactor and Catalysis Engineering Julianalaan 136

2628 BL Delft The Netherlands

Part I

Integration of Heat Transfer and Chemical Reactions

Integrated Chemical Processes Edited by K Sundmacher, A Kienle and A Seidel-Morgenstern

Copyright © 2005 WILEY-VCH Verlag GmbH & Co KGaA, Weinheim

Enhancing Productivity and Thermal Efficiency

of High-Temperature Endothermic Processes

in Heat-Integrated Fixed-Bed Reactors

Grigorios Kolios, Achim Gritsch, Bernd Glöckler and Gerhart Eigenberger

Abstract

High-temperature endothermic processes (e.g., reforming of hydrocarbons) arewidely utilized in the production of basic chemicals and fuels from fossil feedstocks.These processes require large amounts of heat at temperatures up to 1000 ºC Inconventional solutions, only about half of the high-temperature heat supplied istransferred into the endothermic reaction Emerging applications such as decentral-ized hydrogen production for residential and mobile power generation require con-siderable improvement in specific productivity and thermal efficiency Thus, thistopic is currently the subject of exciting research activities in industry and academia.This chapter provides an introduction to the fundamentals of heat-integrated pro-cesses, an overview on conceptual trends in process and apparatus design, and ananalysis of the state of the art, with emphasis on steam reforming of methane

1.1

Introduction

Endothermic high-temperature processes stand at the beginning of the chemical duction chain – for example, syngas is produced mainly through steam reforming ofnaphtha or natural gas, ethylene and propylene through steam cracking, and styrenethrough dehydrogenation of ethylbenzene These processes are usually conducted inlarge furnaces and belong naturally to the largest fuel consumers At the same time,they have a significant heat surplus since typically only about 50 % of the heat genera-ted is consumed by the endothermic reaction In large product- and energy-integratedchemical sites, waste heat can be utilized in subsequent processes However, rigidthermal coupling throughout the plant imposes constraints regarding the heatbalance of individual processes and requires a considerable overhead in order toadjust plant-wide optimal operating conditions This problem could be significantlyrelaxed by reducing the surplus of the heat exporters This is one strong incentive forenhancing the thermal efficiency of endothermic high-temperature processes

pro-Integrated Chemical Processes Edited by K Sundmacher, A Kienle and A Seidel-Morgenstern

Copyright © 2005 WILEY-VCH Verlag GmbH & Co KGaA, Weinheim

Two emerging trends endorse the concept of heat-integrated processes: first, theproduction of basic chemicals is moved close to oil and gas wells where crude oil ornatural gas is processed in large stand-alone units [1] Second, fuel cell systemsrequire on-site and on-demand hydrogen production from primary fuels (i.e., natu-ral gas, liquid hydrocarbons or alcohols) [2] Net heat generation in these processes

is equivalent to raw material and energy loss, and is therefore undesirable

While many publications in the field of heat-integrated processes focus on specificprocesses such as dehydrogenation of paraffins or hydrogen production [3–5], thischapter is more focused on general conceptual trends in process and apparatus design

1.2 Heat-Integrated Processes for Endothermic Reactions

The intended purpose of heat-integrated processes for endothermic reactions isillustrated by the example of methane steam reforming for hydrogen generation,which is of high practical relevance [6] and features typical characteristics of the con-sidered process class The reaction is given by the following stoichiometric equation:

Figure 1.1 shows the heat fluxes required at different process stages for the ation of H2 equivalent to 1 kW combustion enthalpy based on the lower heatingvalue of hydrogen (LHV, H2) Typically, the feed composition of the technical processcorresponds to a steam-to-carbon molar ratio of S:C = 3:1 The reaction temperature

gener-is determined by chemical equilibrium The heat consumption of the ing reaction is equivalent to around 30 % of the lower heating value of the producedhydrogen The amount of heat required for heating the gaseous feed to reaction tem-

perature corresponds to almost 20 % of the heat of combustion On the other hand,

a considerable amount of sensible heat contained in the product stream is usable forfeed preheating

Heat-integrated processes comprise two additional functions besides the main(endothermic) reaction: process heat generation and heat recovery As shown above,both aspects are equally important, since the heat of reaction and the heat requiredfor feed preheating are in the same range Figure 1.2 shows, schematically, two dif-ferent configurations of heat-integrated processes The process comprises two ther-mally interconnected reaction stages for the endothermic reaction and for thecombustion Heat exchangers on both sides perform heat recovery by coupling feedpreheating with product cooling In the first configuration (Fig 1.2(a)) the endother-mic reaction mixture and the combustion gas flow countercurrently through all pro-cess stages such that the hot effluent from the exothermic reaction is used to heat upthe cold feed for the endothermic reaction, and vice versa In the second configura-tion (Fig 1.2(b)), thermal contact between the two process streams is limited to thereaction stage, while heat recovery is separated between the flows of the exothermicand the endothermic reactions

Fig 1.2. Schematic flow configurations of heat-integrated

processes for coupling endothermic and exothermic reactions

(a) Countercurrent flow of process streams (b) Cocurrent flow

of the process streams in the reactor stages and heat recovery

in separate circuits.

Optimality Conditions

Analysis of the overall heat balance yields basic conditions for optimal heat recovery.The optimizing condition is minimization of heat loss The feed temperatures of thetwo process streams, T0, and the outlet temperature of the endothermic stage, Tequ,are fixed (Fig 1.2) where Tequ is the temperature required for the desired conversionunder equilibrium conditions Obviously, Tequ is also a lower limit of the tempera-ture in the combustion stage The performance of the heat exchangers is given bythe mean heat transfer coefficient kh and the heat exchange area A Both heatexchangers are assumed to be identical The mass flow rate of the endothermic mix-ture and the combustion gas is and , respectively Crucial parameters for

the efficiency of heat recovery are the heat capacity ratio of the process streams (h)

and the number of transfer units of the heat exchangers (NTU):

Taking the amount of heat required for bridging the temperature gap (T equ – T0)for the endothermic mixture as a reference, the normalized heat loss is given by thefollowing expression:

A value of loss norm= 1 indicates that the total heat loss via the exit streams of theexothermic and the endothermic stage is equal to the amount of heat required toheat up the feed of the endothermic stage about 'T = Tequ – T0 loss norm can be

expressed explicitly as a function of h and NTU endo taking into account the theory ofideal countercurrent heat exchangers [7]

At this point, a distinction is required between the two configurations shown inFig 1.2 In the countercurrent configuration the heat capacity fluxes of the heatexchanging streams are generally differing, whereas in the counter-cocurrent config-uration they are equal to each other Hence, we obtain:

M c h

NTU h

h e

h e

Figure 1.3 illustrates the dependence of loss norm on h and NTU endo for both rations.

Fig 1.3. The effect of heat exchanger performance, NTU,

and heat capacity ratio, h, on the normalized heat loss for

heat-integrated processes in (a) countercurrent and

(b) counter-cocurrent flow configuration.

For the countercurrent configuration the optimal heat capacity ratio converges

asymptotically to h = 1 with increasing NTU values Even for an ideal heat exchanger

(NTU  ⬁), small deviations from the optimal heat capacity ratio cause significantheat losses On the other hand, heat losses could be minimized with decreasing heatcapacity ratio and increasing NTU values for the counter-cocurrent configuration Inthis case the heat recovery becomes complete for an ideal heat exchanger(NTU ⬁) independent of the heat capacity ratio h.

Conventional steam reformers are furnaces containing tubes filled with reformingcatalyst Radiation burners, which are usually installed at the top and the bottom ofthe furnace, generate the process heat (Fig 1.4(a)) Figure 1.4(b) shows a schematiclateral temperature profile inside a single reformer tube

Clearly, in the considered high-temperature processes the process conditions aredefined mainly through the thermal stability of the tubes A considerable tempera-ture difference across the tube wall is required as the driving force for the heat sup-ply to the reforming reaction The main heat transfer resistances occur at the innerand outer surface of the tube wall

The impact of heat transfer limitations is illustrated with a generic exampledescribing a best-case scenario (Fig 1.5) The allowed maximum temperature of thereformer tube is assumed at 950 ºC Hence, the wall temperature and inlet tempera-ture of the reaction gas are set to 950 ºC The reforming reaction is assumed to

be instantaneous – that is, at each axial position conversion is set equal to the

equi-librium conversion at the respective temperature X equ (T ) Dissipative effects (i.e.,

Fig 1.4. Simplified scheme of an industrial steam reformer furnace (a) Furnace with radiation burners adapted from [6]

(b) Lateral temperature profile inside a single reformer tube.

Fig 1.5. Effect of heat transfer limitation in industrial steam

reformers (a) Scheme of a single catalytic fixed-bed reformer

tube with typical heat transfer and operating parameters

(b) Computed temperature profile (solid line) and equilibrium

conversion profile (dashed line) in flow direction for a tubular

reformer (c) Measured fixed-bed temperature profiles at

different wall temperatures in a tubular reformer (D i = 8 mm)

filled with spherical catalyst pellets (d p = 1.5 mm).

diffusion and heat conduction) are neglected in the axial direction Finally, weassume a uniform fixed-bed temperature in radial direction Based on this model theaxial temperature profile of the fixed bed is given by the following relationship [8]:

Figure 1.5(b) shows the axial temperature profile of the packing for a favorableheat transfer coefficient at the tube wall of Dw = 200 W/m 2 /K and a small tube diame- ter of D i = 25mm The temperature at the inlet cross-section drops immediately to

the equilibrium value corresponding to the feed conditions Further downstream,the temperature increases gradually and approaches asymptotically the wall temper-ature A reactor length of approximately 10 m would be necessary in order to reach

an equilibrium conversion of 95 % According to this result, heat transfer limitationsbetween the catalyst packing and the heating tube wall result in poor catalyst utiliza-tion This has been verified experimentally with a laboratory-scale reactor of 8 mminner diameter (Fig 1.5(c)) Despite the significantly larger specific heat transferarea of this tube, a temperature drop of up to 200 K has been observed at theentrance to the catalytic bed Similar behavior – although not especially extreme – isobserved with endothermic reactions running under milder conditions For exam-ple, during styrene synthesis in a 2.5-cm tube a temperature difference up to 30 Khas been measured between the reactor wall and its center [9] This heat transferlimitation could be overcome by depositing the catalyst directly on the surface of thetube wall – that is, by using a wall reactor concept Charlesworth et al [10] estimatedthat such a reactor would be two orders of magnitude smaller than a conventionalsteam reformer

The above considerations indicate that, independent of implementation details,the space–time yield of endothermic reactions could be significantly enhanced byshifting the reaction site to the heat-exchanging surfaces This intention has led tothe production of a large variety of multifunctional reactor concepts for couplingendothermic and exothermic reactions In the following section the state of the art

in this area will be discussed for selected examples

1.3 Multifunctional Reactor Concepts

Figure 1.6 contains a classification of multifunctional reactor concepts with grated heat recovery for coupling endothermic and exothermic reactions [11] Cou-pling of methane steam reforming with methane combustion is displayed as arepresentative example Equivalent processes can be created based on recuperativeheat exchange in a stationary operation mode (left column) or based on regenerativeheat exchange in a cyclic operation mode (right column)

inte-The simultaneous mode is the simplest configuration where the feed streams

of the endothermic mixture and the combustion gas are mixed and react

simultaneously in the same volume The amount of oxygen is adjusted to generatesufficient excess heat, in order to compensate non-idealities of heat recovery Thesimultaneous mode has been widely applied in technical processes, for example, inautothermal reforming or in oxidative dehydrogenation processes However, mixing

of the process streams imposes substantial constraints to the process conditions,since they must be compatible to the endothermic and to the exothermic subsystem.For example, high-pressure operation, often desirable for steam reforming, would

Fig 1.6. Schematic classification of heat-integrated concepts

for the representative example of coupling methane steam

reform-ing and methane combustion.

require pressurization of the combustion air Finally, dilution of the product by theexhaust gas stream complicates subsequent purification steps.

In the asymmetric mode of operation the endothermic and exothermic reactionstake place separately either in different compartments (recuperative mode) or at dif-ferent time intervals (regenerative mode) The attractiveness of the asymmetricmode is related to the fact that the separation of the process streams allows an indi-vidual tuning of the operating conditions for the endothermic and the exothermicsubsystem

The symmetric mode aims at combining the advantages of the simultaneous andthe asymmetric mode: only the endothermic reaction takes place in the main reac-tor The process heat is added through a hot, inert side stream (e.g., the effluent gas

of an external combustion chamber) The side stream can be distributed along thereactor in order to adjust a specific temperature profile

The counter-cocurrent concept features a modular design: it provides separateheat exchanger loops for heat recovery within the endothermic mixture and the com-bustion gas Cocurrent flow of exothermic and endothermic process streams is prin-cipally favorable with respect to the controllability of heat release

As indicated by this schematic representation, different configurations vary withrespect to the degree of coupling between the process streams The strongest cou-pling occurs in the simultaneous mode, where chemical and thermal interactionoccurs between the process streams In contrast to this, the counter-cocurrent con-cept features only thermal interaction between the process streams localized in thereactor stage

As stated earlier in Section 1.2.1.1, equality of the heat capacity fluxes in theheat recovery sections is a crucial condition for efficient heat recovery This condi-tion is inherently fulfilled by the simultaneous and the counter-cocurrent concept

In the asymmetric concept the flow rates of the process streams must be adjustedaccordingly The requirement of equal heat capacity fluxes cannot be fulfilled inthe symmetric concept due to the continuous side stream addition However,besides these structural properties the design details are decisive for their specificperformance and efficiency of individual solutions

1.3.1

Regenerative Processes

Surprisingly, the majority of advanced heat-integrated reactor concepts employ acyclic mode of operation with regenerative heat exchange They have been proposedmainly for syngas and olefin production Their potential field of application is inlarge chemical and petrochemical processes, where a compact reactor could replace

a complex network of reactors and heat exchangers The following survey reflects theclassification introduced in Fig 1.6

1.3.1.1 Simultaneous Mode

A pioneering report on the simultaneous concept with integrated heat recovery poses a reverse-flow mode of operation for syngas production from natural gas [12].The process is implemented in an adiabatic fixed-bed reactor with inert end zonesand a catalytically active section in the middle The experiments confirm the feasibil-ity of the concept The results of this investigation, together with those of subse-quent studies [13, 14], indicate that local excess temperatures up to 1500 ºC are themajor problem of this concept One reason for this is the tendency for the reactionzones of combustion and reforming to separate from each other Due to a preferen-tial adsorption of oxygen on the catalyst surface, total oxidation of hydrocarbons isfavored [15] Steam reforming and the water-gas-shift reaction take off after com-plete depletion of oxygen Additionally, ignition of homogeneous combustion isunavoidable if the temperature exceeds 600 ºC Coke formation in the heat-exchangezones has been identified as an additional reason for the temperature runaway DeGroote et al [13] show that the major source of coke in the upstream section is meth-ane cracking and the Boudouard equilibrium in the downstream section The sud-den ignition of accumulated coke may lead to extreme local temperature peaks

The CATOFIN process developed by ABB-Lummus for dehydrogenation of C3–C4paraffins [16, 17] can be considered as the prototype process of coupling endother-mic and exothermic reactions Figure 1.7 shows the process scheme in schematicform Each cycle includes the production phase and two regeneration steps Theheat consumption of the endothermic reaction during the production phase is takenfrom the heat stored in the fixed bed The thermal reservoir is restored by a super-heated air purge during the second phase In addition to convective heating, heat isgenerated through the combustion of carbonaceous deposits Finally, hydrogen ispassed over the fixed bed in order to convert the catalyst back to its reduced, activeform, and this produces additional heat A similar concept has been proposed for

Fig 1.7. CATOFIN process: dehydrogenation of propane

adapted from Ullmann [54] (a) charge heater; (b) air heater;

(c) purge step; (d) production step; (e) regeneration step.

styrene synthesis [18] The attraction of this concept results from the simplicity ofthe reactor – a simple adiabatic fixed-bed type – and the process-integrated regenera-tion of the catalyst – that is, the removal of carbonaceous deposits Both, cocurrentflow and countercurrent flow have been investigated The authors concluded thatthe reverse-flow mode is superior with regard to selectivity because the endothermicreaction runs along an increasing temperature profile However, at a low air-to-hydrocarbon ratio the yield of the cocurrent-flow mode is higher than in the reverse-flow mode This is due to a wrong-way phenomenon caused by coupling the endoth-ermic reaction with countercurrent heat exchange.

Figure 1.8 illustrates the operating behavior of the reverse-flow CATOFIN process

in the limit of equal heat capacities during reaction and regeneration cycle (h = 1).

The inlet temperature of the regeneration gas is set approximately above theinlet temperature of the endothermic reaction feed In the periodic steady state, onlytwo narrow zones close to both reactor ends contribute considerably to the conver-sion, while the major part of the fixed bed cools down to a temperature level well

Fig 1.8. Reverse-flow CATOFIN process at equal heat capacity fluxes during production and regeneration cycle: periodic temp- erature profiles (top) and conversion profiles (bottom) at the end

of the endothermic semicycle (t = t cyc /2) and the regeneration cycle (t = 0).

∆T ad

below the adiabatic temperature drop of the endothermic reaction The tion of heat consumption of the endothermic reaction in the fixed bed is the reverseanalogy to the accumulation of heat of weakly exothermic reactions in autothermalreactors [19] This counter-intuitive effect indicates the complexity of coupling chem-ical reactions with countercurrent heat exchange.

accumula-The asymmetric reverse-flow operation mode as proposed by Levenspiel [20] is astraightforward extension of the CATOFIN process in order to integrate heat genera-tion and heat recovery in the reactor Its application in syngas production has beenstudied theoretically in a series of papers by Kulkarni and Dudukovi [21, 22] Thisstudy demonstrates the feasibility of coupling methane steam reforming with meth-ane combustion, but it also reveals the susceptibility of the process to severe over-heating Operation with preheated feed during the exothermic semicycle is proposed

Fig 1.9. Coupling of dehydrogenation of

ethylbenzene to styrene and hydrogen

combustion in a catalytic fixed-bed reverse flow

reactor [9] (a, b) Fixed-bed temperature profiles

during production and regeneration cycle

(c) Ethylbenzene conversion during production cycle (d) Hydrogen weight fraction during regeneration cycle (e) Ethylbenzene conversion during one period.

finding of this study is the importance of structuring the fixed bed in catalytic andinert sections In particular, a sufficient length of the right inert zone is decisive forthe establishment of a sufficient temperature level in the catalytic part However,heat release within a narrow zone at the right end of the catalytic section impliesinefficient heat storage The initial temperature profile of the endothermic semicycleenables indeed high conversion at a high selectivity, but subsequently the tempera-ture peak is shifted into the right inert section and becomes useless.

The major conclusions from the above-described studies are consistent: in theasymmetric mode of operation the reaction zones of the exothermic and endother-mic reactions inherently repel each other, leading either to an extreme maximumtemperature or to poor performance A noncontinuous heat supply and productionduring every other semicycle cause obviously strong fluctuations of operation More-over, reasonable states of operation are attainable only with an excess of gas duringthe exothermic semicycle This contradicts the condition of equal heat capacities foroptimal heat recovery (see Section 1.2.1.1) For example, the heat loss in the case dis-played in Fig 1.9 is equal to the heat demand of the endothermic reaction Differentstrategies have been assessed with regard to their potential to reduce hotspots dur-ing the exothermic semicycle and to improve thermal efficiency

Latent heat storage

It is clear that temperature oscillations during heating–cooling cycles depend on thefixed-bed heat capacity Figure 1.10 shows a simplified picture of the effect of phasechange on the effective heat capacity of pure substances Considerable amounts of

Fig 1.10. Temperature plotted versus specific enthalpy of a pure substance undergoing phase changes (melting, solidifying).

heat can be stored at a constant temperature level at the melting point The principle

of latent heat storage is discussed in detail in Ref [24] Figure 1.11 illustrates theeffect of adding aluminum-filled pellets to the catalytic fixed bed in the reverse-flowreactor for styrene synthesis The melting temperature of aluminum (664 ºC) iscompatible with the operating temperature of the styrene catalyst The amount ofheat required for melting aluminum would be sufficient to heat-up the sameamount of solid metal by approximately 400 K In Fig 1.11 the diagrams on the left-hand side show the conditions during the production cycle in the periodic steady

Fig 1.11. Coupling of dehydrogenation of ethylbenzene to

styrene and hydrogen combustion in a catalytic fixed-bed reverse-

flow reactor with a mixed bed of catalytic pellets and aluminum

powder (50:50) [23] (a, b) Temperature profile of the latent heat

storage (solid line) and of the fixed bed (dashed line) during

production and regeneration periods (c) Ethylbenzene conversion

during the production cycle (d) Hydrogen weight fraction during

the regeneration cycle (e, f) Molten fraction of the latent heat

storage during production and regeneration period.