Get the full ebook with Bonus Features for a Better Reading Experience on ebookgate.

com

Separation Process Principles with Applications

Using Process Simulators 4th Edition J. D. Seader

https://ebookgate.com/product/separation-process-principles-
with-applications-using-process-simulators-4th-edition-j-d-
seader/

OR CLICK HERE

DOWLOAD NOW

Download more ebook instantly today at https://ebookgate.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Separation Process Engineering Second Edition Safari

https://ebookgate.com/product/separation-process-engineering-second-
edition-safari/

ebookgate.com

Process Heat Transfer Principles Applications and Rules of

Thumb 1st Edition Thomas Lestina

https://ebookgate.com/product/process-heat-transfer-principles-
applications-and-rules-of-thumb-1st-edition-thomas-lestina/

ebookgate.com

Process simulation using WITNESS including lean and Six

sigma applications 1st Edition Raid Al-Aomar

https://ebookgate.com/product/process-simulation-using-witness-
including-lean-and-six-sigma-applications-1st-edition-raid-al-aomar/

ebookgate.com

The Research Process 4th Candian Edition Lori Wilkinson

https://ebookgate.com/product/the-research-process-4th-candian-
edition-lori-wilkinson/

ebookgate.com
Process Dynamics and Control 4th Edition Dale E. Seborg

https://ebookgate.com/product/process-dynamics-and-control-4th-
edition-dale-e-seborg/

ebookgate.com

Real Estate Development Principles and Process Fifth

Edition Mike E. Miles

https://ebookgate.com/product/real-estate-development-principles-and-
process-fifth-edition-mike-e-miles/

ebookgate.com

Bayesian Hierarchical Models With Applications Using R

Second Edition Peter D. Congdon (Author)

https://ebookgate.com/product/bayesian-hierarchical-models-with-
applications-using-r-second-edition-peter-d-congdon-author/

ebookgate.com

Business Process Outsourcing Process Strategies and

Contracts Second Edition Halvey J.K.

https://ebookgate.com/product/business-process-outsourcing-process-
strategies-and-contracts-second-edition-halvey-j-k/

ebookgate.com

Applications of Random Process Excursion Analysis 1st

Edition Irina S. Brainina

https://ebookgate.com/product/applications-of-random-process-
excursion-analysis-1st-edition-irina-s-brainina/

ebookgate.com
Trim Size: 8.5in x 11in Seader f02.tex V2 - 11/16/2015 7:25 P.M. Page viii
Trim Size: 8.5in x 11in Seader b02.tex V2 - 09/23/2015 7:15 A.M. Page 1

Factors for Converting AE and CGS Units to SI Units

To convert from To Multiply by
Area
ft2 m2 0.0929
in2 m2 6.452 × 10−4
Acceleration
ft/h2 m/s2 2.352 × 10−8
Density
lbm ∕ft3 kg/m3 16.02
lbm ∕gal (US) kg/m3 119.8
g/cm3 kg/m3 = g/L 1000
Diffusivity, Kinematic Viscosity
ft2 /h m2 /s 2.581 × 10−5
cm2 /s m2 /s 1 × 10−4
Energy, Work, Heat
ft-lbf J 1.356
Btu (IT) J 1055
cal (IT) J 4.187
erg J 1 × 10−7
kW-h J 3.6 × 106
Enthalpy
Btu (IT)/lbm J/kg 2326
cal (IT)/g J/kg 4187
Force
lbf N 4.448
dyne N 1 × 10−5
Heat-Transfer Coefficient
Btu (IT)/h-ft2 -∘ F W/m2 -K 5.679
cal (IT)/s-cm2 -∘ C W/m2 -K 4.187 × 10−4
Interfacial Tension
lbf /ft kg/s2 14.59
dyne/cm kg/s2 1 × 10−4
Length
ft m 0.3048
in m 0.0254
Mass
lbm kg 0.4536
ton kg 907.2
tonne (metric ton) kg 1000
Trim Size: 8.5in x 11in Seader b02.tex V2 - 09/23/2015 7:15 A.M. Page 2

To convert from To Multiply by

Mass Flow Rate
lbm ∕h kg/s 1.26 × 10−4
lbm ∕s kg/s 0.4536
Mass Flux, Mass Velocity
lbm ∕h-ft2 kg/s-m2 1.356 × 10−3
Power
ft-lbf /h W = J/s 3.766 × 10−4
ft-lbf W = J/s 1.356
hp W = J/s 745.7
Btu (IT)/h W = J/s 0.2931
Pressure
lbf ∕ft2 Pa 47.88
lbf ∕in2 Pa 6895
atm Pa 1.013 × 105
Bar Pa 1 × 105
torr = mmHg Pa 133.3
in Hg Pa 3386
in H2 O Pa 249.1
Specific Heat
Btu (IT)/lbm -∘ F J/kg-K 4187
cal/g-∘ C J/kg-K 4187
Surface Tension
lbf /ft N/m 14.59
dyne/cm N/m 0.001
erg/cm2 N/m 0.001
Thermal Conductivity
Btu (I/T)-ft/h-ft2 -∘ F W/m-K = J/s-m-K 1.731
cal (IT)-cm/s-cm2 -∘ C W/m-K = J/s-m-K 418.7
Velocity
ft/h m/s 8.467 × 10−5
ft/s m/s 0.3048
Viscosity
lbm /ft-s kg/m-s 1.488
lbm /ft-h kg/m-s 4.134 × 10−4
cP kg/m-s 0.001
Volume kg/m-s 0.001
ft3 m3 0.02832
L m3 1 × 10−3
gal (US) m3 3.785 × 10−3
 Trim Size: 8.5in x 11in Seader ffirs.tex V1 - 11/05/2015 3:51 P.M. Page i

SEPARATION PROCESS
PRINCIPLES
with Applications Using
Process Simulators
FOURTH EDITION

J. D. Seader
Department of Chemical Engineering
University of Utah

Ernest J. Henley
Department of Chemical Engineering
University of Houston

D. Keith Roper
Ralph E. Martin Department of Chemical Engineering
University of Arkansas
 Trim Size: 8.5in x 11in Seader ffirs.tex V1 - 11/11/2015 8:00 P.M. Page ii

VICE PRESIDENT & DIRECTOR Laurie Rosatone

SENIOR DIRECTOR Don Fowley
EXECUTIVE EDITOR Linda Ratts
SPONSORING EDITOR Mary O’Sullivan
ASSISTANT DEVELOPMENT EDITOR Adria Giattino
SENIOR ASSISTANT Courtney Jordan
FREELANCE PROJECT MANAGER Ellen Keohane
PROJECT SPECIALIST Nichole Urban
FREELANCE PROJECT ASSISTANT Anna Melhorn
EXECUTIVE MARKETING MANAGER Daniel Sayre
ASSISTANT MARKETING MANAGER Puja Katariwala
ASSOCIATE DIRECTOR Kevin Holm
SENIOR CONTENT SPECIALIST Nicole Repasky
PRODUCTION EDITOR Linda Christina E
PHOTO RESEARCHER Mary Ann Price
COVER PHOTO CREDIT Courtesy J.D. Seader

This book was set in 10/12 Times LT Std Roman by SPi Global and printed and bound by Lightning Source Inc.

Founded in 1807, John Wiley & Sons, Inc. has been a valued source of knowledge and understanding for more than 200 years, helping people around the world
meet their needs and fulfill their aspirations. Our company is built on a foundation of principles that include responsibility to the communities we serve and
where we live and work.

In 2008, we launched a Corporate Citizenship Initiative, a global effort to address the environmental, social, economic, and ethical challenges we face in our
business. Among the issues we are addressing are carbon impact, paper specifications and procurement, ethical conduct within our business and among our
vendors, and community and charitable support. For more information, please visit our website: www.wiley.com/go/citizenship.

Copyright © 2016, 2011, 2006, 1998 John Wiley & Sons, Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or
transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except as permitted under Sections 107 or
108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate
per-copy fee to the Copyright Clearance Center, Inc. 222 Rosewood Drive, Danvers, MA 01923, website www.copyright.com. Requests to the Publisher for
permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030-5774, (201)748-6011, fax
(201)748-6008, website http://www.wiley.com/go/permissions.

Evaluation copies are provided to qualified academics and professionals for review purposes only, for use in their courses during the next academic year. These
copies are licensed and may not be sold or transferred to a third party. Upon completion of the review period, please return the evaluation copy to Wiley. Return
instructions and a free of charge return shipping label are available at www.wiley.com/go/returnlabel. Outside of the United States, please contact your local
representative.

ISBN: 978-1-119-23959-8 (PBK)

ISBN: 978-1-119-22412-9 (EVALC)

Library of Congress Cataloging-in-Publication Data

Names: Seader, J. D. | Henley, Ernest J. | Roper, D. Keith.

Title: Separation process principles with applications using process
simulators / J.D. Seader, Department of Chemical Engineering, University
of Utah, Ernest J. Henley, Department of Chemical Engineering, University
of Houston, D. Keith Roper, Ralph E. Martin Department of Chemical
Engineering, University of Arkansas.
Description: Fourth edition. | Hoboken, NJ : John Wiley & Sons, Inc., 2016. |
Includes bibliographical references and index.
Identifiers: LCCN 2015021981 (print) | LCCN 2015022980 (ebook) | ISBN
9781119239598 (pbk.) | ISBN 9781119141303 (pdf) | ISBN
9781119141297 (epub)
Subjects: LCSH: Separation (Technology)–Textbooks.
Classification: LCC TP156.S45 S364 2016 (print) | LCC TP156.S45 (ebook) | DDC
660/.2842–dc23
LC record available at http://lccn.loc.gov/2015021981

Printing identification and country of origin will either be included on this page and/or the end of the book. In addition, if the ISBN on this page and the back
cover do not match, the ISBN on the back cover should be considered the correct ISBN.

Printed in the United States of America

10 9 8 7 6 5 4 3 2 1
 Trim Size: 8.5in x 11in Seader f01.tex V2 - 10/16/2015 7:51 P.M. Page iii

About the Authors

J. D. Seader is Professor Emeritus of Chemical Engineering Ernest J. Henley is Professor of Chemical Engineering at
at the University of Utah. He received B.S. (summa cum laude) the University of Houston. He received his B.S. degree
and M.S. degrees from the University of California at Berkeley from the University of Delaware and his Dr. Eng. Sci. from
and a Ph.D. from the University of Wisconsin–Madison. From Columbia University, where he served as a professor from
1952 to 1959, he worked for Chevron Research in Richmond, 1953 to 1959. Henley also has held professorships at the
California, where he designed petroleum and petrochemical Stevens Institute of Technology, the University of Brazil,
processes and supervised engineering research, including Stanford University, Cambridge University, and the City
the development of one of the first computer-aided process University of New York. He has authored or coauthored
simulation programs and the first widely used vapor-liquid 72 technical articles and 17 books. For 17 years, he was a
equilibrium correlation. From 1959 to 1965, he conducted trustee of CACHE, serving as president from 1975 to 1976
rocket engine research on all of the engines that took man and directing the efforts that produced the seven-volume set
to the moon for the Rocketdyne Division of North American of Computer Programs for Chemical Engineering Education
Aviation in Canoga Park, California. Before joining the fac- and the five-volume set, AIChE Modular Instruction. An
ulty at the University of Utah, where he served for 37 years active consultant, Henley holds nine patents and has served
from 1966 until 2003, he was a professor at the University on the Board of Directors of 12 industrial corporations and
of Idaho. He has authored or coauthored 111 technical arti- three philanthropies. In 1998, he received the McGraw-Hill
cles, nine books, and four patents and also coauthored the Company Award for “Outstanding Personal Achievement in
section on distillation in the sixth and seventh editions of Chemical Engineering” and in 2002, received the CACHE
Perry’s Chemical Engineers’ Handbook. His co-authored Award of the ASEE for “recognition of his contribution to the
textbooks include FLOWTRAN Simulation—An Introduction use of computers in chemical engineering education.” He is
in three editions; Equilibrium-Stage Separation Operations president of the Henley Foundation and his biography appears
in Chemical Engineering, Product & Process Design Prin- in Who’s Who in America and Who’s Who in the World.
ciples in four editions; and the previous three editions of
Separation Process Principles. He was a founding member D. Keith Roper is the Charles W. Oxford Professor of
and trustee of CACHE for 33 years, serving as executive Emerging Technologies in the Ralph E. Martin Department
officer from 1980 to 1984. From 1975 to 1978, he served of Chemical Engineering at the University of Arkansas.
as chairman of the Chemical Engineering Department at the He is also Program Leader for the Engineering Research
University of Utah. For 20 years, he directed the use and Centers and Network for Computational Nanotechnology at
distribution of Monsanto’s FLOWTRAN process simulation the National Science Foundation. He received a B.S. degree
computer program to 190 chemical engineering departments (magna cum laude) from Brigham Young University in 1989
worldwide, helping to usher in the addition of instruction in and a Ph.D. from the University of Wisconsin–Madison in
computer-aided process design to the chemical engineering 1994. He developed processes for cell culture, fermentation,
curriculum. For 12 years, he served as an associate editor of biorecovery, and analysis of polysaccharide, protein, DNA,
the journal Industrial and Engineering Chemistry Research. and adenoviral-vectored antigens at Merck & Co. (West
He served as a director of AIChE from 1983 to 1985. In 1983, Point, PA); extraction of photodynamic cancer therapeutics
he presented the 35th Annual Institute Lecture of AIChE; in at Frontier Scientific, Inc. (Logan, UT); and virus binding
1988, he received the Computing in Chemical Engineering methods for Millipore Corp (Billerica, MA). Previously,
Award of the CAST Division of AIChE; in 2004, he received he held faculty appointments in chemical engineering and
the CACHE Award for Excellence in Chemical Engineering materials science and engineering at the University of Utah.
Education from the ASEE; and in 2004 he was a co-recipient, From 2010 to 2013, he served as Assistant Director of the
with Professor Warren D. Seider, of the Warren K. Lewis Microelectronics-Photonics graduate program at the Univer-
Award for Chemical Engineering Education of the AIChE. sity of Arkansas. From 2012 to 2014, he served as Program
In 2008, as part of the AIChE Centennial Celebration, he Director in the Education Engineering and Centers Division
was named one of 30 authors of groundbreaking chemical at the National Science Foundation. He is an active consul-
engineering books. tant with industries in biotechnology, biopharmaceuticals,

iii
 Trim Size: 8.5in x 11in Seader f01.tex V2 - 10/16/2015 7:51 P.M. Page iv

iv About the Authors

chemicals, optoelectronics, and energy as well as with uni- Biological Engineering and a member of the Arkansas
versities in translating discovery to disruptive innovation Academy of Science. He holds memberships in Tau Beta Pi,
through university–industry partnerships. He has authored ASEE, ACS, SPIE, and AIChE. His research examines elec-
or coauthored more than 65 technical articles, one book, trodynamics in biochemical and nano-optical material systems
two book chapters, three U.S. patents, and six U.S. patent important to advanced biomanufacturing, biomedicine, opto-
applications. He was instrumental in developing one viral and electronics, and sustainable energy, with recent advances
three bacterial vaccine products, 16 Good Manufacturing Pro- in computation, nanolithography, spectroscopy, and electron
cess documents, and multiple bioprocess equipment designs. microscopy of nanoplasmonic metamaterials and polymer thin
He is a fellow of the American Institute for Medical and films.
 Trim Size: 8.5in x 11in Seader f02.tex V2 - 11/16/2015 7:25 P.M. Page v

Preface to the Fourth Edition

INTRODUCTION STUDENT-FRIENDLY FEATURES

Separation Process Principles was first published in 1988 to Each chapter begins by stating an average of nine instructional
provide a rigorous and comprehensive treatment of industrial objectives to give the student an introduction to the author’s
separation processes in the form of a student-friendly textbook expectations. In chapters that introduce separation technology,
format. Both equilibrium-stage and mass-transfer (rate-based) this is followed by a flow sheet and description of an industrial
models were covered. Included also were chapters on thermo- process to demonstrate how the technology is used to create
dynamic and mass-transfer theory for separation operations. saleable products. The theory needed to design equipment or
Five of the 15 chapters began with an example of an indus- simulate a separation operation is developed and applications
trial separation process and descriptions of how and why are demonstrated by more than 200 examples, many of which
modern process design techniques evolved. The 2006 second make use of process simulators.
edition expanded the coverage of mass-transfer-based design Each chapter ends with study questions suitable for class
methods by adding three new chapters (16, 17, and 18) discussion (the inverted classroom) or qualitative examination
covering leaching, washing, crystallization, desublimation, questions. There are more than 600 homework exercises,
evaporation, and drying of solids. In the 2011 third edition, many of which require the use of a process simulator. Other
the content was enlarged to include approximately 175 pages examples and exercises involve nonlinear equations with rec-
on biochemical separations as well as a new Chapter 19 ommendations for using Matlab to solve them. In preparation
covering mechanical phase separations such as filtration of the fourth edition, text was clarified based on student feed-
and centrifugation. The title of the third edition, Separation back, and errors were eliminated. Following a suggestion by
Process Principles—Chemical and Biochemical Operations, some reviewers, chapter-specific nomenclature sections have
reflected this expansion. been added. A general nomenclature lists symbols common
This new condensed fourth edition focuses on material to many of the chapters. Each chapter-specific nomenclature
widely taught in undergraduate separations courses and keys symbols to equations, figures, or chapter sections. These
on process simulations using Aspen Plus, CHEMCAD, and revisions in the fourth edition reflect our desire to prepare an
ChemSep to design separation operations. Hence, the new title accessible introduction to chemical separations that includes
for the fourth edition: Separation Process Principles–with the methods and equations used by process simulators and the
Applications Using Process Simulators. Professor Seader problems and opportunities this has created.
began using process simulation programs in 1957 while
engaged in process design in the petroleum industry and was
TOPICAL ORGANIZATION
the AIChE Institute Lecturer on this subject in 1984. Professor
Henley founded what is now Chemstations and was its first Chapter 1 provides the student with a classification of
president (when it was named COADE). Professor Roper separation operations and an overview of industrial chemical
introduced simulation to vaccine process development in engineering separations. The next two chapters review ther-
1997. Authors of this fourth edition believe that students who modynamics and mass transfer, as applied to separation
learn to solve separation problems with process simulators as operations, at a level consistent with the level of this book.
juniors will be well prepared to tackle difficult process design Chapter 4, which introduces design specifications, and
problems in their senior year. The fourth edition provides com- equilibrium flash calculations are a prerequisite for all sub-
parisons of features of process simulators as well as examples sequent chapters. Chapter 5, which is also a prerequisite for
of how they are used to solve industrial-level design problems. all subsequent chapters, extends the single-stage concept
Much of the material removed from the third edition— to multiple-equilibrium-stage processing units, including a
biochemical separation process principles in: sections of degrees-of-freedom analysis for determining the number of
Chapters 1, 2, 3, 8, 14, and 15; Chapters 16, 17, and 18, design variables that can be specified for a given separation
and Chapter 19 on mechanical phase separations—is being process. A further prerequisite is Chapter 6, which describes
prepared by the authors for inclusion in a volume devoted to trays and packing, introduces general graphical methods,
bioseparation operations. Meanwhile, this content will remain stage efficiencies, and a transfer unit design method for
available from the publisher in the third edition. packed units.

v
 Trim Size: 8.5in x 11in Seader f02.tex V2 - 11/16/2015 7:25 P.M. Page vi

vi Preface to the Fourth Edition

Once the contents of the first six chapters have been combination of a web browser (e.g., Chrome, Firefox, and
mastered, the subsequent chapters can be studied in any Internet Explorer) and a general search engine (e.g., Google,
order preferred by the reader or the instructor, with a few Yahoo, Bing, and Ask) has made it so simple to search for
exceptions: (1) Concepts of minimum stage and minimum information that the number of internet users has increased
reflux are mentioned early in the text, but a quantitative elab- from about 400 million in 2000 to almost 3 billion (compared
oration, using McCabe–Thiele diagrams, resides in Section to a world population of 7.2 billion) in 2014.
7.2. McCabe–Thiele diagrams are used in three other chapters Some useful websites for the study of separation pro-
to help readers visualize concepts such as pinch points. cesses are presented in chapters of this book. However, such
(2) Unique technology such as Freundlich and Langmuir websites are sometimes removed from the Internet. Readers
isotherms is fully developed in Chapter 15 but applied earlier. of this book are encouraged to conduct their own searches by
(3) Mathematical tools such as solution methods for stiff using key words. Many subjects have articles in the Wikipedia,
differential equations and the Newton–Raphson method for a free, open-ended encyclopedia on the internet. An initial
solving nonlinear algebraic equations are described where search, where the word “wiki” is added to the key word(s) is
applied. In these three instances, the reader is referred to a often useful. For example, if the key words “distillation wiki”
book section where the technique is fully developed. are entered in the Google search engine, a 13-page article
To help students obtain an awareness of a section’s com- appears that provides excellent background material on a wide
plexity, section titles listed in the table of contents include the spectrum of the subject.
following symbols as prefixes: The 15 most-used websites are:
∗ Important for a basic understanding of separations and Search engines (Google, Yahoo, Bing, and Ask)
therefore recommended for presentation in class, unless Social networking websites (Facebook, Twitter, Instagram,
already covered in a previous course. LinkedIn, and Pinterest)
o Optional because the material is descriptive, probably News, Entertainment, Purchases, and Information
covered in a previous course, or can be read outside of (YouTube, Amazon, eBay, MSN)
class with little or no discussion in class. Blog creation (WordPress)
• Advanced material suitable for an undergraduate course
Students using this book are encouraged to communicate
if students are familiar with, and have access to, a process
with each other and their instructors via social networking
simulator.
websites to enhance their learning of separation processes.

NEW TO THIS EDITION

RESOURCES FOR INSTRUCTORS
The focus on using process simulators to design separa-
tion processes required some rearrangement and revision of The website www.wiley.com/college/seader furnishes the fol-
Chapters 1, 2, and 4 –13, which deal with operations covered lowing resources in the Instructor Companion Site:
in process simulators. Many of the examples and exercises 1. Instructor’s Solution Manual
have been revised and new ones added. Some chapters contain 2. Resources for Instructors-An introduction
correlations that stretch the state of the art. Chapter 7, for
3. Image Gallery
example, develops an improved method for obtaining plate
efficiencies for sieve-tray columns that is just beginning to 4. PowerPoint slides
appear in process simulators. 5. Sample Preliminary Examination
HELPUL WEBSITES In the 1980s, local computer net- 6. Suggestions for Completing Homework Exercise
works began to interconnect in a global system, called the
7. Suggested Review of Prerequisite Material
Internet. In late 1990, Tim Berners-Lee at the European
Organization for Nuclear Research (CERN), Switzerland, Instructors adopting this book must request a password from
completed all the tools required for sharing information over their Wiley sales representative.
the Internet. This included the HyperText Transfer Protocol
(HTTP), the HyperText Markup Language (HTML), and a
document-information system called the World Wide Web RESOURCES FOR STUDENTS
(WWW). In early 1991, the first-known web page was sent Students are encouraged to consult www.wiley.com/college/
over the internet by Lee to Paul Jones of UNC-Chapel Hill. seader for the following resources in the Student Companion
Physicists around the world could now quickly and efficiently Site:
share data. Later that year, the WWW became available to
the public, but its popularity was restrained until the release 1. Suggestions for Completing HW Exercises
of the Mosaic web browser in December 1992. Today the 2. Suggested Review of Prerequisite Material
 Trim Size: 8.5in x 11in Seader f02.tex V2 - 11/16/2015 7:25 P.M. Page vii

Preface to the Fourth Edition vii

ACKNOWLEDGEMENTS Dr. D. Keith Roper is currently on intergovernmental

personnel assignment with the National Science Foundation.
Since publication of the 3rd edition of this book, the authors
Any opinions, findings, and conclusions or recommenda-
have received valuable comments, corrections, and sug-
tions expressed in this book are those of the authors and
gestions from Professor Phillip R. Westmoreland of North
do not necessarily reflect the views of the National Science
Carolina University, Professor Shuguang Deng of New
Foundation.
Mexico State University, Professor Daniel E. Rosner of Yale
University, and particularly Professor N. Lawrence Ricker of
J. D. Seader
the University of Washington.
Ernest J. Henley
The selection of the content for the 4th edition was greatly
D. Keith Roper
influenced by the responses of reviewers to two surveys con-
ducted by Wiley.
Trim Size: 8.5in x 11in Seader f02.tex V2 - 11/16/2015 7:25 P.M. Page viii
 Trim Size: 8.5in x 11in Seader ftoc.tex V1 - 10/16/2015 7:52 P.M. Page ix

Contents

About the Authors iii 3. Mass Transfer and Diffusion 46

Preface to the Fourth Edition v 3.0∗ Instructional Objectives 46
General Nomenclature xiii 3.1∗ Steady-State, Ordinary Molecular
Dimensions and Units xvii Diffusion 47
∗
3.2 Diffusion Coefficients (Diffusivities) 51
1. Separation Processes 1
3.3∗ Steady-State and Unsteady-State Mass Transfer
1.0∗ Instructional Objectives 1
Through Stationary Media 58
1.1∗ Industrial Chemical Processes 1 ∗
3.4 Mass Transfer in Laminar Flow 60
1.2∗ Basic Separation Techniques 3
3.5∗ Mass Transfer in Turbulent Flow 68
1.3⚬ Separations by Phase Creation 4
3.6∗ Models for Mass Transfer in Fluids with a
1.4⚬ Separations by Phase Addition 6
Fluid–Fluid Interface 73
1.5⚬ Separations by Barrier 7 ∗
3.7 Two-Film Theory and Overall Mass-Transfer
1.6⚬ Separations by an External Field or
Coefficients 76
Gradient 7
Nomenclature, Summary, References, Study
∗
1.7 Brief Comparison of Common Separation
Questions, Exercises
Operations 8
∗
1.8 Separation Processes, Product Purity, 4. Single Equilibrium Stages and Flash
Component Recovery, and Separation Calculations 87
Sequences 9 4.0∗ Instructional Objectives 87
Summary, References, Study Questions, 4.1∗ Gibbs’ Phase Rule and Degrees of
Exercises Freedom 88
∗
4.2 Binary Vapor–Liquid Systems at
2. Thermodynamics of Separation Operations 16 Equilibrium 89
2.0∗ Instructional Objectives 16 ∗
4.3 Equilibrium Two-Phase Flash Calculations 93
2.1∗ Phase Equilibria 16 4.4∗ Ternary Liquid–Liquid Systems at
2.2∗ Ideal-Gas, Ideal-Liquid-Solution Model 20 Equilibrium 97
2.3⚬ Graphical Representation of Thermodynamic 4.5⚬ Multicomponent Liquid–Liquid Systems 101
Properties 21 4.6∗ Liquid–Solid Systems 102
⚬
2.4 Nonideal Thermodynamic Property 4.7∗ Gas–Liquid Systems 104
Models 23 4.8∗ Gas–Solid Systems 105
⚬
2.5 P-v-T Equation-of-State (EOS) Models 23 4.9⦁ Three-Phase Equilibrium Systems 107
2.6⚬ Highly Nonideal Liquid Solutions 27 Nomenclature, Summary, References, Study
2.7⚬ Gibbs Excess Free-Energy (gE ) Models 29 Questions, Exercises
2.8⚬ Predictive Models 34
2.9⚬ Electrolyte Solution Models 36 5. Multistage Cascades and Hybrid Systems 118
2.10⚬ Polymer Solution Models 36 5.0∗ Instructional Objectives 118
2.11∗ K-Value Methods in Process Simulators 36 5.1∗ Cascade Configurations 118
2.12∗ Exergy and Second-Law Analysis 37 5.2∗ Single-Section Liquid–Liquid Extraction
Nomenclature, Summary, References, Study Cascades 119
Questions, Exercises

ix
 Trim Size: 8.5in x 11in Seader ftoc.tex V1 - 10/16/2015 7:52 P.M. Page x

x Contents

5.3∗ Two-Section Distillation Cascades 121 8.2⚬ General Design Considerations 239
5.4⚬ Membrane Cascades 123 8.3∗ Hunter–Nash Graphical Equilibrium-Stage
5.5⚬ Hybrid Systems 125 Method 243
5.6∗ Degrees of Freedom and Specifications for 8.4⚬ Theory and Scale-Up of Extractor
Cascades 125 Performance 252
Nomenclature, Summary, References, Study Nomenclature, Summary, References, Study
Questions, Exercises Questions, Exercises

6. Absorption and Stripping 137 9. Approximate Methods for Multicomponent

6.0∗ Instructional Objectives 137 Distillation 267
6.1⚬ Equipment for Vapor–Liquid 9.0∗ Instructional Objectives 267
Separations 138 9.1∗ Fenske–Underwood–Gilliland (FUG)
⚬
6.2 General Design Considerations 143 Method 267
6.3∗ Graphical Method for Trayed Towers 144 9.2∗ Using the Shortcut (FUG) Method with Process
6.4∗ Kremser Group Method for Multicomponent Simulators 279
Absorption and Stripping 148 Nomenclature, Summary, References, Study
∗
6.5 Stage Efficiency and Column Height for Trayed Questions, Exercises
Columns 154
10. Equilibrium-Based Methods for Multicomponent
6.6∗ Flooding, Column Diameter, and Tray Layout
Absorption, Stripping, Distillation, and
for Trayed Columns 161
∗
Extraction 284
6.7 Rate-Based Method for Packed Columns 164
10.0∗ Instructional Objectives 284
6.8∗ Packed-Column Liquid Holdup, Diameter,
10.1∗ Simple Model for a Vapor–Liquid Equilibrium
Flooding, Pressure Drop, and Mass-Transfer
Stage 284
Efficiency 169 ⦁
⦁
10.2 Evolution of Methods for Solving the Mesh
6.9 Reactive (Chemical) Absorption 180
Equations 286
Nomenclature, Summary, References, Study ∗
10.3 Strategies for Applying Process-Simulator
Questions, Exercises
Methods 287
∗
10.4 Main Mathematical Procedures 291
7. Distillation of Binary Mixtures 191
7.0∗ Instructional Objectives 191 10.5∗ Bubble-Point (BP) and Sum-Rates (SR)
7.1⚬ Equipment and Design Considerations 193 Methods 294
∗
7.2∗ McCabe–Thiele Graphical Method for Trayed 10.6 Simultaneous-Correction Method 297
Towers 193 10.7∗ Inside-Out Method 304
⚬
7.3 Extensions of the McCabe–Thiele 10.8⦁ Rigorous Methods for Liquid–Liquid
Method 203 Extraction 309
7.4∗ Estimation of Tray Efficiency for Nomenclature, Summary, References, Study
Distillation 208 Questions, Exercises
∗
7.5 Column and Reflux-Drum Diameters 215 11. Enhanced Distillation and Supercritical
7.6∗ Rate-Based Method for Packed Distillation Extraction 320
Columns 216 11.0∗ Instructional Objectives 320
Nomenclature, Summary, References, Study 11.1∗ Use of Triangular Graphs 321
Questions, Exercises 11.2∗ Extractive Distillation 332
8. Liquid–Liquid Extraction with Ternary 11.3⦁ Salt Distillation 335
Systems 231 11.4⦁ Pressure-Swing Distillation 337
8.0∗ Instructional Objectives 231 11.5⦁ Homogeneous Azeotropic Distillation 339
8.1⚬ Equipment for Solvent Extraction 233 11.6∗ Heterogeneous Azeotropic Distillation 343
 Trim Size: 8.5in x 11in Seader ftoc.tex V1 - 10/16/2015 7:52 P.M. Page xi

Contents xi

11.7⦁ Reactive Distillation 352 14. Membrane Separations 408

11.8⦁ Supercritical-Fluid Extraction 357 14.0∗ Instructional Objectives 408
Nomenclature, Summary, References, Study 14.1⚬ Membrane Materials 410
Questions, Exercises 14.2⚬ Membrane Modules 414
14.3∗ Mass Transfer in Membranes 416
12. Rate-Based Models for Vapor–Liquid Separation
14.4∗ Dialysis 430
Operations 368
14.5⚬ Electrodialysis 432
12.0⦁ Instructional Objectives 368
14.6∗ Reverse Osmosis 434
12.1⦁ Rate-Based Model 370
14.7∗ Gas Permeation 438
12.2⦁ Thermodynamic Properties and Transport-Rate
14.8⚬ Pervaporation 441
Expressions 372
⦁ Nomenclature, Summary, References, Study
12.3 Methods for Estimating Transport Coefficients
Questions, Exercises
and Interfacial Area 375
⦁
12.4 Vapor and Liquid Flow Patterns 375 15. Adsorption, Ion Exchange, and
12.5⦁ Method of Calculation 376 Chromatography 451
Nomenclature, Summary, References, Study 15.0∗ Instructional Objectives 451
Questions, Exercises 15.1∗ Sorbents 453
15.2∗ Equilibrium Considerations 461
13. Batch Distillation 385
15.3∗ Kinetic and Transport Rate
13.0∗ Instructional Objectives 385
Considerations 470
13.1∗ Differential Distillation 385
15.4⚬ Equipment for Sorption Operations 475
13.2∗ Binary Batch Rectification 388
15.5∗ Slurry and Fixed-Bed Adsorption
13.3⦁ Batch Stripping and Complex Batch
Systems 479
Distillation 390
15.6∗ Continuous, Countercurrent Adsorption
⦁
13.4 Effect of Liquid Holdup 391
Systems 494
13.5∗ Stage-by-Stage Methods for Batch ⚬
15.7 Ion-Exchange Cycle 502
Rectification 391
15.8∗ Chromatographic Separations 503
∗
13.6 Intermediate-Cut Strategy 400
Nomenclature, Summary, References, Study
13.7⦁ Optimal Control by Variation of Reflux
Questions, Exercises
Ratio 401
Nomenclature, Summary, References, Study Answers to Selected Exercises 519
Questions, Exercises Index 521

∗ Suitable for an UG course

⚬ Optional
⦁ Advanced
Trim Size: 8.5in x 11in Seader ftoc.tex V1 - 10/16/2015 7:52 P.M. Page xii
 Trim Size: 8.5in x 11in Seader f03.tex V2 - 10/16/2015 7:52 P.M. Page xiii

General Nomenclature

All symbols are defined in the text where they first appear. UNIQUAC universal quasichemical theory method,
The list below contains only those symbols that are com- Table 2.8
mon to two or more chapters. VLE vapor–liquid equilibrium, Figure 4.6
Other symbols are listed in chapter-specific nomenclatures VLLE vapor–liquid–liquid equilibrium, §11.8
at the end of each chapter.
VOC volatile organic compounds
VTPR volume-translated P-R EOS, §2.8.3
Abbreviations and Acronyms Wilson Wilson EOS, Table 2.8

BWR Benedict–Webb–Rubin EOS, §2.5

CSTR completely stirred tank or reactor, §3.6.3 Latin Symbols
CW (cw) cooling water, Figure 2.11 Å Angstrom, §14.1
DAE differential-algebraic equations, §13.5.2 A mass-transfer area, (3-13)
DDB Dortmund data bank, §2.5.3 A absorption factor, L/KV, (5-38)
DDBST DDB software package, §2.5.3 a interfacial area per unit volume, (14-5);
DECHEMA data bank, §2.5.3 activity, Table 2.1
EM equimolar diffusion, Table 6.5 B identifier for bottoms, Figure 1.12
EOS equation of state (P-v-T relationship), §2.2 B bottoms molar flow rate, Figure 5.15
ESA energy-separation agent, Table 1.1 C identifier for condenser, Figure 1.12
FUG Fenske–Underwood–Gilliland method, §9.1 C components, (4-1)
HEPT height equivalent to a theoretical plate, CD drag coefficient, Table 3.9
(6-68) CP heat capacity at constant pressure, (2-38)
HETS height equivalent to a theoretical stage, c liquid concentration, moles/volume,
(6-68) (3-40)
HHK heavier than the heavy key, Figure 9.6 D an identifier for distillate, Figure 1.12
HK heavy key, §9.1 DAB molecular diffusivity of A in B, (3-3a)
HTU height of a transfer unit, §6.8.3 Di solute diffusivity, (14-14)
LK light key, §9.1 DK Knudsen diffusivity, (14-18)
LKP Lee–Kessler–Plöcker EOS, §2.11 De effective diffusivity, (3-49)
LLK lighter than the light key, Figure 9.6 D distillate flow rate, Figure 5.7; diameter,
LM log mean, (3-33) (3-121)
LW lost work, Table 2.10 dH hydraulic diameter, (14-5)
MSA mass-separation agent, Table 1.2 dP particle diameter, (6-58)
MW molecular weight E extract mass flow rate, Figure 5.2
NRTL nonrandom, two-liquid model, Table 2.8 E an identifier for extract phase, Figure 5.2
NTU number of transfer units, Table 6.5 E extraction factor, (4-35)
ODE ordinary differential equations, §13.5.2 EMV Murphree vapor tray efficiency, (6-53)
PR Peng–Robinson EOS, Table 2.8 Eo stage (plate efficiency), (6-41)
PSRK predictive SRK method, §2.8.2 EOV Murphree vapor point efficiency, (6-51)
RK Redlich–Kwong EOS, §2.8.2 F molar feed rate, Example 1.1
STP standard conditions, 1 atm, 0∘ C F identifier for feed, Figure 1.2
SRK Suave–Redlich–Kwong EOS, Table 2.8 F Faraday’s constant, (3-47)
UNIFAC functional group activity coefficient FD drag force, (6-57)
method, Table 2.8 f pure component fugacity, §2.1.1; Fanning
UM unimolecular diffusion, Table 6.5 friction factor, Table 3.9

xiii
 Trim Size: 8.5in x 11in Seader f03.tex V2 - 10/16/2015 7:52 P.M. Page xiv

xiv General Nomenclature

fo fugacity at standard state, (2-14) ND degrees of freedom, number of design

f partial fugacity, (2-7) variables, (4-1)
G Gibbs free energy, §2.1: mass velocity, NE number of independent equations, (4-3)
(3-145) NP number of phases, (4-1)
g Gibbs molar free energy, §2.1; gravitational NV number of variables, (4-1)
constant, (6-94) NFr Froude number, Table 3.9
gE excess free energy, (2-62) NLe Lewis number, Table 3.9
gc force to mass conversion factor in AE NNu Nusselt number, Table 3.9
system of units NPe Peclet number, Table 3.9
H Henry’s law constant (2-31); vessel height, NRe Reynolds number, Table 3.9
Figure 4.18
NSc Schmidt number, Table 3.9
ΔH vap enthalpy of vaporization, (2-43)
NSh Sherwood number, Table 3.9
HL height of a mass-transfer unit (L) liquid or
NSt Stanton number, Table 3.9
(V) vapor, Table 6.5
NWe Weber number, Table 3.9
HOL height of an overall mass-transfer unit (L or
V ), Table 6.5 NG number of gas phase (G) or liquid phase (L)
mass-transfer units, Table 6.5
h molar enthalpy, §2.2; heat transfer
coefficient, (15-18) NOG number of overall gas (G) or liquid phase
(L) mass-transfer units, Table 6.5
ho ideal gas molar enthalpy, §2.2
n molar flow rate, Example 1.1; mass transfer
J molar flux by molecular diffusion relative to
rate, §6.5.4
mixture, (3-3)
P pressure, (2-1); product flow rate, Figure 5.3
j mass flux relative to mixture, (3-5)
Ps vapor pressure, (2-13)
jD , jH , jM Chilton–Colburn factors for mass, heat, and
PM permeability, KD, (14-1), (14-27)
momentum transfer (3-147)
PM = PM ∕lM permeance, (14-1)
j-factor Table 3.9
p partial pressure (for designated component)
K vapor–liquid equilibrium ratio (K-value),
(2-18) Q rate of heat transfer, (4-17); volumetric flow
rate, (7-51)
K various subscripted overall mass-transfer
coefficients, Tables 6.5 and 3.12, §3.7.1 q heat flux, Q∕A, (3-2); q –line, (4-11)
KD distribution ratio, partition coefficient, R universal gas constant, (3-47): retentate,
(2-19) Example 1.1
K ′D liquid–liquid equilibrium ratio in mole or R reflux ratio L/D, Table 7.2; raffinate flow
mass ratios (4-33) rate, Figure 5.2
k various subscripted individual mass-transfer R identifier for reflux, Example 2.5: raffinate,
coefficients, Tables 6.5 and 3.12, §3.7.1 Figure 4.14
k thermal conductivity, (3-2) r radius, (3-54)
L liquid molar flow rate, Table 1.1; length, S entropy (2-2); solvent flow rate, (4-32); side
(3-82) stream, Table 5.4
LR liquid reflux flow rate, Figure 5.8 S stripping factor, (5-51)
L liquid molar flow rate in stripping section, ΔSirr irreversible change in entropy, Table 2.11
Figure 7.5 S identifier for solvent, §4.4
l component liquid flow rate, (6-15); packed s molar entropy, Table 2.3
height, Figure 6.25 T temperature, (2-1)
lM membrane thickness, (14-1) Tc critical temperature, Table 15.4
M molecular weight Tr reduced temperature, T/ Tc , (2-49)
m slope of equilibrium curve dy/dx, Figure t time
(3-16) u liquid velocity, (3-71)
N number of moles, (2.1); number of stages V molar vapor flow rate, Table 1.1; volume,
(5-7); flux, molar flow/area (3-1) (2-2)
molar trans-membrane flux, (14-1) V molar vapor flow in stripping section,
NA Avogadro’s number, (3-38) Figure 7.5
 Trim Size: 8.5in x 11in Seader f03.tex V2 - 10/16/2015 7:52 P.M. Page xv

General Nomenclature xv

v kinematic viscosity (6-98); vapor Subscripts

component flow rate, (6-13); liquid molar
A,B,C components, Figure 5.4
volume, (3-43); pore flow velocity, (14-2)
B bottoms, (7-4)
Ws shaft work, Table 2.11
C condenser, (7-37)
w mass fraction, (3-5)
D distillate, (7-4)
X mass or mole ratio in liquid, (4-32)
F feed, (7-6)
x liquid mole fraction
c critical point, §2.5
Y mass or mole ratio in vapor,
Figure 4.19 i component in a mixture
y vapor mole fraction irr irreversible
z mole fraction of feed component, min minimum
Figure 2.9 j stage, Figure 10.1
Z compressibility factor, (2-49) n stage, (6-11)
R reboiler, (7-37)
Greek Symbols r reference component, (9-14)
sat saturated
αAB relative volatility, KA ∕KB , (4-10)
x in the x direction
β relative selectivity, (2-21), (8-4)
y in the y direction
γ activity coefficient, Table 2.1
z in the z direction
Δ change in value of variable
o datum (reference state), 1 atm., 25∘ C
ε porosity, (15-2)
∞ infinity
η second-law efficiency, Table 2.10
μ viscosity, (3-38), chemical potential or partial
molar Gibbs free energy, (2-2) Superscripts
ρ mass density F feed, (4-32)
ρL liquid mass density, (2-41) N number of stages, (5-7): a stage, (5-15)
ρV vapor mass density, Table 2.3 R raffinate, Figure 5.4
σ liquid surface tension, (3-43) s saturation vapor pressure, (2-28)
τ tortuosity, (3-49) o reference state, Table 2.3
ϕ pure species fugacity coefficient, Table 2.1 * phase equilibrium value (with respect to
ϕs pure species fugacity coefficient at saturation another phase), Figure 3.16
pressure, (2-30) – average, (3-74); partial molar property, (2-8)
′
ϕ partial fugacity coefficient, Table 2.1 derivative, (4-28); mass instead of moles, for
̄
Φ local volume fraction, §2.7.1 K and D values, etc., (4-33); normalized
ω acentric factor, (2-48) values, §4.5; solute free gas or adsorbent, §6.3
Trim Size: 8.5in x 11in Seader f03.tex V2 - 10/16/2015 7:52 P.M. Page xvi
 Trim Size: 8.5in x 11in Seader f04.tex V2 - 10/16/2015 7:52 P.M. Page xvii

Dimensions and Units

C hemical engineers must be proficient in the use of three systems of units: (1) the Interna-
tional System of Units, SI System (Systeme Internationale d’Unites), which was established
in 1960 by the 11th General Conference on Weights and Measures and has been widely
adopted; (2) the AE (American Engineering) System, which is based largely upon an English
system of units adopted when the Magna Carta was signed in 1215 and is a preferred system
in the United States; and (3) the CGS (centimeter-gram-second) System, which was devised
in 1790 by the National Assembly of France, and served as the basis for the development
of the SI System. A useful index to units and systems of units is given on the website
http://www.sizes.com/units/index.php.
Engineers must deal with dimensions units, and values to express numerical quantities.
Thus, for 10 gallons of gasoline, the dimension is volume, the unit is gallons, and the value
is 10. As detailed in NIST (National Institute of Standards and Technology) Special Publi-
cation 811, Guide for the Use of the International System of Units (SI), 2008 edition, which
is available at the website http://www.nist.gov/physlab/pubs/sp811/index.cfm, units are base
or derived.

BASE UNITS
The base units are those that are independent, cannot be subdivided, and are accurately defined.
The base units are for dimensions of length, mass, time, temperature, molar amount, electrical
current, and luminous intensity, all of which can be measured independently. Derived units
are expressed in terms of base units or other derived units and include dimensions of vol-
ume, velocity, density, force, and energy. In this book we deal with the first five of the base
dimensions. For these, the base units are:

Base Dimension SI Unit AE Unit CGS Unit

Length meter, m foot, ft centimeter, cm
Mass kilogram, kg pound, lbm gram, g
Time second, s hour, h second, s
Temperature kelvin, K Fahrenheit, ∘ F Celsius, ∘ C
Molar amount gram-mole, mol pound-mole, lbmol gram-mole, mol

ATOM AND MOLECULE UNITS

atomic weight = atomic mass unit = the mass of one atom
molecular weight (MW) = molecular mass (M) = formula weight∗ = formula mass∗ = the
sum of the atomic weights of all atoms in a molecule ( ∗ also applies to ions)
1 atomic mass unit (amu or u) = 1 universal mass unit = 1 dalton (Da) = 1∕12 of the mass of
one atom of carbon-12 = the mass of one proton or one neutron
The units of MW are amu, u, Da, g∕mol, kg∕kmol, or lb∕lbmol (the last three are most
convenient when MW appears in a formula).
The number of molecules or ions in one mole = Avogadro’s number = 6.022 × 1023.

xvii
 Trim Size: 8.5in x 11in Seader f04.tex V2 - 10/16/2015 7:52 P.M. Page xviii

xviii Dimensions and Units

DERIVED UNITS
Many derived dimensions and units are used in chemical engineering. Some are listed in the
following table:

Derived Dimension SI Unit AE Unit CGS Unit

Area = Length2
m 2
ft2
cm2
Volume = Length3 m3 ft3 cm3
Mass flow rate = Mass/Time kg/s lbm /h g/s
Molar flow rate = mol/s lbmol/h mol/s
Molar amount/Time
Velocity = Length/Time m/s ft/h cm/s
Acceleration = Velocity/Time m/s2 ft/h2 cm/s2
Force = Mass ⋅ Acceleration newton, N = 1 kg-m/s2 lbf dyne = 1 g-cm/s2
Pressure = Force/Area pascal, Pa = lbf /in.2 atm
1 N/m2 =
1 kg/m-s2
Energy = Force ⋅ Length joule, J = ft-lbf , Btu erg = 1 dyne-cm =
1 N-m = 1 g-cm2 /s2 , cal
1 kg-m2 /s2
Power = Energy/Time = watt, W = hp erg/s
Work/Time 1 J/s =
1 N-m/s
1 kg-m2 /s3
Density = Mass/Volume kg/m3 lbm /ft3 g/cm3

OTHER UNITS ACCEPTABLE FOR USE WITH THE SI SYSTEM

A major advantage of the SI System is the consistency of the derived units with the base units.
However, some acceptable deviations from this consistency and some other acceptable base
units are given in the following table:

Dimension Base or Derived SI Unit Acceptable SI Unit

Time s minute (min), hour (h), day (d), year (y)
Volume m3 liter (L) = 10−3 m3
Mass kg metric ton or tonne (t) = 103 kg
Pressure Pa bar = 105 Pa

PREFIXES
Also acceptable for use with the SI System are decimal multiples and submultiples of SI units
formed by prefixes. The following table lists the more commonly used prefixes:

Prefix Factor Symbol

12
tera 10 T
giga 109 G
mega 106 M
kilo 103 k
deci 10−1 d
centi 10−2 c
milli 10−3 m
micro 10−6 μ
nano 10−9 n
pico 10−12 p
 Trim Size: 8.5in x 11in Seader f04.tex V2 - 10/16/2015 7:52 P.M. Page xix

Dimensions and Units xix

USING THE AE SYSTEM OF UNITS

The AE System is more difficult to use than the SI System because of the units for force,
energy, and power. In the AE System, the force unit is the pound-force, lbf , which is defined to
be numerically equal to the pound-mass, lbm , at sea-level of the earth. Accordingly, Newton’s
second law of motion is written,
g
F=m
gc

where F = force in lbf , m = mass in lbm , g = acceleration due to gravity in ft∕s2 , and, to
complete the definition, the constant gc = 32.174 lbm -ft∕lbf -s2 , where 32.174 ft∕s2 is the
acceleration due to gravity at sea-level of the earth. The constant gc is not used with the SI
System or the CGS System because the former does not define a kgf and the CGS System does
not use a gf .
Thus, when using AE units in an equation that includes force and mass, incorporate gc to
adjust the units.

EXAMPLE OF UNIT CONVERSIONS

A 5.000-pound-mass weight, m, is held at a height, h, of 4.000 feet above sea-level. Calculate its potential energy above sea-level, P.E. = mgh,
using each of the three systems of units. Factors for converting units are given on the inside front cover of this book.

SI System:

m = 5.000 lbm = 5.000(0.4536) = 2.268 kg

g = 9.807 m∕s2
h = 4.000 ft = 4.000(0.3048) = 1.219 m
P.E. = 2.268(9.807)(1.219) = 27.11 kg-m2∕s2 = 27.11 J

CGS System:

m = 5.000 lbm = 5.000(453.6) = 2268 g

g = 980.7 cm∕s2
h = 4.000 ft = 4.000(30.48) = 121.9 cm
P.E. = 2268(980.7)(121.9) = 2.711 × 108 g-cm2∕s2
= 2.711 × 108 erg

AE System:

m = 5.000 lbm
g = 32.174 ft∕s2
h = 4.000 ft
P.E. = 5.000(32.174)(4.000) = 643.5 lbm -ft2∕s2

However, the accepted unit of energy for the AE System is ft-lbf , which is obtained by dividing by gc . Therefore, P.E. = 643.5/32.174 =
20.00 ft-lbf .
Another difficulty with the AE System is the differentiation between energy as work and energy as heat. As seen in the preceding table,
the work unit is ft-lbf , while the heat unit is Btu. A similar situation exists in the CGS System with corresponding units of erg and calorie
(cal). In older textbooks, the conversion factor between work and heat is often incorporated into an equation with the symbol J, called Joule’s
constant or the mechanical equivalent of heat, where

J = 778.2 ft-lbf ∕Btu = 4.184 × 107 erg∕cal

Thus, in the previous example, the heat equivalents are

AE System:
20.00∕778.2 = 0.02570 Btu

CGS System:
2.711 × 108 ∕4.184 × 107 = 6.479 cal
 Trim Size: 8.5in x 11in Seader f04.tex V2 - 10/16/2015 7:52 P.M. Page xx

xx Dimensions and Units

In the SI System, the prefix M, mega, stands for million. However, in the natural gas and petroleum industries of the United States, when
using the AE System, M stands for thousand and MM stands for million. Thus, MBtu stands for thousands of Btu, while MM Btu stands for
millions of Btu.
It should be noted that the common pressure and power units in use for the AE System are not consistent with the base units. Thus, for
pressure, pounds per square inch, psi or lbf /in.2 , is used rather than lbf /ft2 . For power, Hp is used instead of ft-lbf /h, where the conversion
factor is
1 hp = 1.980 × 106 ft-lbf ∕h

CONVERSION FACTORS
Physical constants may be found on the inside back cover of this book. Conversion factors are
given on the inside front cover. These factors permit direct conversion of AE and CGS values
to SI values. The following is an example of such a conversion, together with the reverse
conversion.

EXAMPLE

1. Convert 50 psia (lbf /in.2 absolute) to kPa:

The conversion factor for lbf /in.2 to Pa is 6,895, which results in

50(6, 895) = 345,000 Pa or 345 kPa

2. Convert 250 kPa to atm:

250 kPa = 250,000 Pa. The conversion factor for atm to Pa is 1.013 × 105 . Therefore, dividing by the conversion factor,

250,000∕1.013 × 105 = 2.47 atm

Three of the units [gallons (gal), calories (cal), and British thermal unit (Btu)] in the list of conversion factors have two or more definitions.
The gallons unit cited here is the U.S. gallon, which is 83.3% of the Imperial gallon. The cal and Btu units used here are international (IT).
Also in common use are the thermochemical cal and Btu, which are 99.964% of the international cal and Btu.

FORMAT FOR EXERCISES IN THIS BOOK

In numerical exercises throughout this book, the system of units to be used to solve the problem
is stated. Then when given values are substituted into equations, units are not appended to the
values. Instead, the conversion of a given value to units in the above tables of base and derived
units is done prior to substitution into the equation or carried out directly in the equation, as
in the following example.

EXAMPLE
Using conversion factors on the inside back cover of this book, calculate a Reynolds number, NRe = Dvρ∕μ, given D = 4.0 ft, v = 4.5 ft∕s,
ρ = 60 lbm ∕ft3 , and μ = 2.0 cP (i.e., centipoise).
Using the SI System (kg-m-s),
Dvρ [(4.00)(0.3048)][(4.5)(0.3048)][(60)(16.02)]
NRe = = = 804,000
μ [(2.0)(0.001)]
Using the CGS System (g-cm-s),
Dvρ [(4.00)(30.48)][(4.5)(30.48)][(60)(0.01602)
NRe = = = 804,000
μ [(0.02)]
Using the AE System (lbm -ft-h) and converting the viscosity 0.02 cP to lbm /ft-h,

Dvρ (4.00)[(4.5)(3600)](60)
NRe = = = 804,000
μ [(0.02)(241.9)]
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 1

Chapter 1

Separation Processes
§1.0 INSTRUCTIONAL OBJECTIVES
After completing this chapter, you should be able to:
• Explain the role of separation operations in the chemical industries.
• Enumerate the four basic separation techniques.
• Explain the uses of energy-separating agents (ESA) and mass-separating agents (MSA).
• Calculate component material balances around a separation operation based on specifications of component recovery
and/or product purity.
• Understand the concept of sequencing of separation operations, particularly distillation.

S eparation processes for chemical mixtures have been known §1.1 INDUSTRIAL CHEMICAL PROCESSES
for millennia. Early civilizations developed processes for:
Chemical companies manufacture products that differ from
(1) extraction of metals from ores, perfumes from flowers,
dyes from plants, and potash from burnt plants; (2) evapora- those in the feedstocks. Included can be (1) naturally occurring
tion of seawater to obtain salt; (3) refining of rock asphalt; and living or nonliving materials; (2) chemical intermediates that
(4) distilling of liquors. The human body could not function are precursors for producing other chemicals; (3) “chemicals
if it had no kidney to separate water and waste products from in commerce” that can be purchased from the global market;
blood. or (4) waste products that can be processed into valuable
Chemical engineers design large-scale facilities that em- products. Especially common are oil refineries, which process
ploy separation methods that most often differ considerably crude oil, synthetic crude oil from tar sands, and tight oil from
from those used by chemists in laboratories. For example, che- the use of horizontal drilling followed by fracking of shale
mists separate light-hydrocarbon mixtures by chromatogra- deposits, to produce a variety of hydrocarbon-based products.
phy, while a manufacturing plant uses distillation to separate For example, starting in 1967, Canada began increasing its
the same mixture. This book discusses methods for the oil production dramatically by processing tar sands from the
design of large-scale separation operations for processes that huge Athabasca deposit in the province of Alberta. Figure 1.1
manufacture chemical products economically. Included are shows the initial products produced by Great Canadian Oil
all of the most common operations, including distillation (both Sands Ltd. (GCOS), now Suncor Energy, from its 1967 plant.
continuous and batch), gas absorption, stripping, liquid–liquid The products are the result of numerous separation opera-
extraction, membrane separations, and gas and liquid adsorp- tions within the plant. By 2013, several Canadian companies
tion. These design methods are incorporated into commercial produced more than 1.6 million barrels per day (bbl/day)
computer-aided process simulators such as Aspen Plus, of synthetic crude oil (syncrude). Because Canada produces
Aspen HYSYS, CHEMCAD, ChemSep, ProSimPlus, and more oil than it needs, it exports syncrude to the United
UniSim. States by the Keystone Pipeline. If phases 3 and 4 of the
Chemical engineers also design small-scale industrial sep- pipeline are completed, Canadian syncrude will flow more
aration systems involving manufacture of specialty chemicals than 2,000 miles from the Keystone Hardisty Terminal in
by batch processing; recovery of biological solutes; crystal Alberta, Canada, to the Gulf near Houston, Texas.
growth of semiconductors; recovery of chemicals from wastes; Chemical plants operate in a batchwise, continuous, or
and products such as lung oxygenators, espresso machines, semicontinuous manner. Plant operations may be key oper-
and hemodialysis devices. A companion book, Bioseparation ations unique to chemical engineering because they involve
Process Principles, covers these smaller-scale processes and changes in chemical composition, or auxiliary operations
the principles of bioseparations. Included in that book are that are necessary to the success of the key operations but
membranes and adsorption for bio applications, chromatog- are often designed by mechanical engineers because they
raphy, electrophoresis, leaching, crystallization, drying, and do not involve changes in chemical composition. The key
mechanical separation operations. Both large- and small-scale operations involve (1) chemical reactions and (2) separation
operations are illustrated in examples and homework exercises of chemical mixtures, such as the separation of a mixture of
in this book and in Bioseparation Process Principles. chemicals into pure, or nearly pure, species (components).

1
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 2

2 Chapter 1 Separation Processes

Plant fuel oil and fuel gas

Sulfur
5
2.85 × 10 kg/day (314 tons/day)

Naphtha

2560 m3/day
Tar sands (16,100 bbl/day)
Synthetic
8.41 × 107 kg/day crude
(92,700 tons/day) GCOS Kerosene
Plant
1685 m3/day 7154 m3/day
(10,600 bbl/day) (45,000 bbl/day)
Natural gas

Gas oil
Water 2909 m3/day
(18,300 bbl/day)

Coke
6
2.36 × 10 kg/day (2,600 tons/day)

Tailings
7.22 × 107 kg/day
(79,600 tons/day)

Figure 1.1 GCOS process for producing synthetic crude oil from Canadian Athabasca tar sands.

The auxiliary operations include phase separation (e.g., sep- product, 99% HCl, with small amounts of H2 , N2 , H2 O, CO,
aration of gas from liquid or separation of two immiscible and CO2 , requires no purification. However, simple process
liquid phases), heat addition or removal (heat exchangers), flowsheets that do not require separation operations are rare.
shaft-work (pumps, compressors, turbines), mixing or divid- In most process flowsheets, separation operations dominate.
ing of streams, solids agglomeration, size reduction of solids, Most industrial chemical processes involve at least one
and separation of solids by size. chemical reactor accompanied by one or more separation
Block-flow diagrams can represent key operations in operations. An example is the continuous hydration of ethy-
chemical processes. They indicate, by square or rectangular lene with water to produce ethyl alcohol, shown in Figure 1.3.
blocks, only the chemical reaction and separation steps and, Central to the process is a reactor packed with catalyst parti-
by connecting lines, the process streams. More detail is shown cles, in which the reaction C2 H4 + H2 O → C2 H5 OH occurs.
in process-flow diagrams, which also include auxiliary Due to chemical equilibrium limitations, conversion of ethy-
operations and utilize icons that depict the particular type of lene is only 5% per pass through the reactor. However, by
equipment employed. Figure 1.2 shows a block-flow diagram recovering unreacted ethylene, in a partial condensation sep-
for manufacturing hydrogen chloride gas from chlorine and aration step, and recycling the ethylene to the reactor, nearly
hydrogen. Central to the process is a chemical reactor, where complete overall conversion of the ethylene feed is achieved.
the gas-phase combustion reaction, H2 + Cl2 → 2HCl, occurs. Recycling is a common element of chemical processes.
In this process, no separation operations are necessary because If pure ethylene were available as a feedstock and no
of the complete conversion of chlorine and the absence of any side reactions occurred, the simple process in Figure 1.3
side reactions. A slight excess of hydrogen is used, and the could be realized. It utilizes a reactor, a partial condenser
for unreacted gaseous ethylene recovery, and distillation to
produce an overhead distillate of aqueous ethyl alcohol of
near-azeotropic composition (93 wt%), with a wastewater
bottoms product. Unfortunately, impurities in the ethylene
feed, together with side reactions involving ethylene and the
impurities, increase the number of separators required for the
process, as shown in Figure 1.4. Also, as shown, an additional
reactor may be necessary to convert impurities or by-products
to other chemicals that can be more easily separated from the
main product. Such an escalation of the need for separation
operations is common. Thus, most chemical processes include
Figure 1.2 Synthetic process for anhydrous HCl production. many more separation operations than chemical reactors.
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 3

§1.2 Basic Separation Techniques 3

Figure 1.3 Process for hydration of ethylene

to ethanol.

Figure 1.4 Industrial process for hydration of ethylene to ethanol.

Chemical engineers also design products that can involve §1.2 BASIC SEPARATION TECHNIQUES
separation operations. One such product is the espresso coffee
machine. Very hot water rapidly leaches desirable chemicals The separation of a chemical mixture into its components is
from the coffee bean, leaving behind ingredients responsible not a spontaneous process, like the mixing by diffusion of
for undesirable acidity and bitterness. The resulting cup of soluble components. Separations require energy in some form.
espresso has (1) a topping of creamy foam that traps the A mixture to be separated into its separate chemical species
extracted chemicals, (2) a fullness of body due to emulsifica- is usually a single, homogeneous phase. If it is multiphase, it
tion, and (3) a richness of aroma. Typically, 25% of the coffee is often best to first separate the phases by gravity or centrifu-
bean is extracted and the espresso contains less caffeine than gation, followed by the separation of each phase mixture.
filtered coffee. Cussler and Moggridge [1] and Seider, Seader, A schematic of a general separation process is shown in
Lewin, and Widagdo [2] discuss other examples of products Figure 1.5. The phase state of the feed can be a vapor, liquid,
designed by chemical engineers that involve the separation of or solid mixture. The products of the separation differ in
chemical mixtures. composition from the feed and may differ in the state of the
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 4

4 Chapter 1 Separation Processes

The extent of separation achieved depends on the exploita-

tion of differences in molecular, thermodynamic, and transport
properties of the species in the feed mixture. Of importance are
molecular weight, molecular shape, dipole moment, electric
charge, vapor pressure, solubility, adsorptivity, and diffusivity.
Values of these properties appear in handbooks, reference
books, textbooks, and journals. Many can be obtained from
Figure 1.5 General separation process. physical property models in process-simulation programs.
When properties are not available, they are estimated or
determined experimentally.
phase. The separation is accomplished by inducing the dif-
ferent chemical species in the feed to partition among the §1.3 SEPARATIONS BY PHASE CREATION
different product phases. Four basic methods for doing this
are shown in Figure 1.6. Most common is the phase creation The simplest and most widely used industrial separation tech-
technique in Figure 1.6a. For example, the feed may be a liquid nique is phase creation. The feed to the separator is a single-
mixture. By heating the mixture to partially vaporize the liq- phase vapor, liquid, or solid. The second phase is created by the
uid, the more-volatile components tend to move to the vapor, transfer of energy to or from the feed by an energy-separating
while less-volatile components tend to remain in the liquid agent (ESA). The ESA can be heat transfer or shaft work by
phase. Alternatively, the feed may be a vapor that is partially means of compression or by the reduction of pressure through
condensed. a turbine or across a valve. After sufficient time and agitation
A second technique, labeled phase addition, is shown in to approach phase equilibrium, the product phases are sepa-
Figure 1.6b. For example, a liquid solvent may contact a liq- rated. The size of the process vessel depends on the flow rate
uid feed to selectively dissolve certain species from the feed. in and out of the vessel and the time for the phases to approach
Less common is the third separation technique that uses a bar- equilibrium.
rier, such as a membrane, as shown in Figure 1.6c. The barrier The most common separation operations using phase
allows passage of certain species in the feed while exclud- creation are listed in Table 1.1. For these operations, design
ing or slowing other species. Finally, in a fourth technique, procedures are well established and are included as unit
an external force field or gradient is used to preferentially operation models in commercial process simulators.
attract certain species in the feed as indicated in Figure 1.6d. When the feed mixture is a vapor or a liquid with com-
The force field might be electrical or centrifugal. ponents that differ widely in volatility (e.g., hydrogen and
For the techniques of Figure 1.6, the rate of separation benzene), partial condensation or partial vaporization,
is governed by the rates of mass transfer of the different Operation (1) in Table 1.1, may be adequate to achieve the
components, while the extent of separation is limited by ther- desired separation. Heat is transferred to or from the feed in
modynamic equilibrium. These two topics are the subject of a heat exchanger followed by phase separation by gravity in a
previous chemical engineering courses, but are reviewed here vessel. Partial vaporization of a liquid occurs in flash vapor-
in Chapters 2 and 3. Also, fluid mechanics and heat trans- ization, Operation (2), by reducing the feed pressure with a
fer play important roles in the separation of mixtures, and valve or turbine, followed by phase separation. Both of these
their principles are applied, where appropriate, throughout separation operations are referred to as single equilibrium
this book. stages because experimental evidence shows that interphase

Figure 1.6 Basic separation process

techniques: (a) separation by phase
creation; (b) separation by phase addition;
(c) separation by barrier; (d) separation by
external force field or gradient.
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 5

§1.3 Separations by Phase Creation 5

Table 1.1 Common Separation Operations Based on Phase Creation

Separation Operation Symbol Feed Phase Created Phase Separating Agent(s)
V

(1) Partial condensation V/L Vapor and/or liquid Liquid or vapor Heat transfer (ESA)
or vaporization
L

(2) Flash L
Liquid Vapor Pressure reduction
vaporization
L

L
(3) Distillation V/L Vapor and/or liquid Vapor and liquid Heat transfer (ESA) and
sometimes shaft work (ESA)

mass transfer of species is so rapid that phase equilibrium and 1.7c, the phases are in continuous contact with other. In the
is closely approached. The liquid product is enriched with first case, randomly dumped ceramic, metal, or plastic rings
respect to the less volatile species, while the vapor product or saddles, called packing, promote turbulence as the vapor
is enriched with respect to the more volatile species. These follows a tortuous path through the liquid, which flows over
operations are discussed in Chapter 4. and through the packing. In the latter case, stacked, structured
When the degree of separation by a single equilibrium packing is used in the form of metal or plastic meshes, grids,
stage is inadequate because the volatility differences among or coils. Depending on the efficiency of the packing, inches or
key species are insufficiently large (e.g., benzene and toluene), feet of packed height are equivalent to an equilibrium stage.
distillation, Operation (3) in Table 1.1, with multiple stages, In the icon for distillation in Table 1.1, horizontal lines within
often becomes the separation method of choice. It is and the column indicate the stages. As vapor flows up the column,
has long been the most widely utilized industrial separa- it is increasingly enriched with respect to the more volatile
tion method. Distillation involves vigorous mixing during species. The liquid flowing down the column is increasingly
contacts between countercurrently flowing liquid and vapor enriched with respect to the less-volatile species.
phases. Figure 1.7 shows schematics of three widely used Feed to a distillation column enters at a stage somewhere
phase-contacting methods. Figure 1.7a depicts a series of between the top and bottom stages. Any vapor in the feed starts
trays, where the phases are mixed and then disengaged. Each up the column; feed liquid starts down. Liquid is also required
tray is some fraction of an equilibrium stage. In Figures 1.7b for making contacts with vapor above the feed stage, and vapor

Tray Dumped packing Structured packing

L L L
V V V V

Tray 1
Downcomer

Tray 2

Tray 3

L V V V
L V V V L
(a) (b) (c)

Figure 1.7 Phase-contacting methods in distillation columns.

 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 6

6 Chapter 1 Separation Processes

is also required for making contacts with liquid below the feed When the feed mixture is a vapor, and it is desired to remove
stage. Therefore, at the top of the column, vapor is condensed the higher molecular weight (heavier) components from the
to provide distillate product and down-flowing liquid (reflux). lower molecular weight (lighter) components, absorption,
Similarly, part of the liquid at the bottom is removed as bot- Operation (1) in Table 1.2, is used. The feed gas enters at
toms product, while the other part passes through a reboiler, the bottom of a multistage column and flows up the column
where it is heated to provide up-flowing vapor (boilup). Con- countercurrent to the MSA, called an absorbent, which enters
tinuous distillation is introduced in Chapter 5. The separation at the top of the column. Typically, absorbers operate at
of a binary mixture by continuous distillation is discussed in near-ambient temperature at an elevated pressure. A subse-
Chapter 7, while the continuous distillation of multicompo- quent separation, often distillation, separates the absorbate
nent mixtures is discussed in Chapters 9–12. Batch distillation from the absorbent, which is then recycled to the absorber.
is discussed in Chapter 13. The inverse of absorption is stripping, Operation (2) in
Table 1.2. Here, a liquid feed mixture is separated, at ele-
vated temperature and near-ambient pressure, by contacting
§1.4 SEPARATIONS BY PHASE ADDITION the feed, which enters at the top of the column, with a gas
The next most widely used industrial separation technique stripping MSA that enters at the bottom. A second separation
is phase addition. The feed to the separator is a single-phase operation may be needed if it is necessary to separate the
vapor, liquid, or solid. The second added phase is called a stripping agent from the components stripped from the liquid
mass-separating agent (MSA). Certain components in the feed feed and/or to recycle the stripping agent to the stripper.
move from the feed phase to the MSA phase. After sufficient Absorption and stripping are discussed in Chapter 6. Column
time and agitation to approach phase equilibrium, the product internals for absorption and stripping are those of Figure 1.7.
phases are separated. Disadvantages of using an MSA are Liquid–liquid extraction, Operation (3) in Table 1.2,
(1) the need for an additional separator to recover the MSA for using a solvent as the MSA, can be used when distillation is
recycle, (2) the need for MSA makeup, (3) possible product impractical, e.g., because the feed is temperature-sensitive. A
contamination by the MSA, and (4) more complex design solvent (MSA) selectively dissolves only certain components
procedures. Table 1.2 lists the most common separation in the feed. The products are an extract, LI , containing the
operations using phase addition. For these operations, design extracted components, and a raffinate, LII , containing the
procedures are well established and are included in process unextracted species. Several countercurrently arranged stages
simulators. may be necessary, either in a staged column with mechanical

Table 1.2 Common Separation Operations Based on Phase Addition

Separation Operation Symbol Feed Phase Added Phase Separating Agent(s)
V
MSA

(1) Absorption V
Vapor Liquid Liquid absorbent (MSA)
L

V
L

(2) Stripping MSA Liquid Vapor Stripping vapor (MSA)

L

LII

MSA

(3) Liquid–liquid extraction L

Liquid Liquid Liquid solvent (MSA)

LI

MSA

(4) Adsorption L
Vapor or liquid Solid Solid adsorbent (MSA)

S
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 7

§1.6 Separations by an External Field or Gradient 7

agitation or in a series of mixing and settling vessels. Addi- non-diffusible particles. Reverse osmosis, Operation (2), is
tional separation operations, often distillation, are needed to the selective transport of a solvent, for example, water, through
recover the solvent for recycle. Liquid–liquid extraction is a microporous membrane after the pressure of the feed is
described in Chapter 8. increased to a value higher than the osmotic pressure of the
The MSA may be a porous solid, in the form of granules, solution. Solutes in the solvent do not permeate the membrane.
that selectively removes one or more components from a It is widely used to produce drinkable water.
vapor or liquid feed mixture by adsorption, Operation (4) in Separation of gases by selective gas permeation through
Table 1.2. Adsorption is confined to the exterior and interior nonporous membranes, Operation (3), using a pressure driving
surfaces of the solid adsorbent, unlike absorption, which force, is a process first used by the U.S. government in the
occurs throughout the liquid absorbent. Adsorption separa- 1940s with porous fluorocarbon barriers to separate 235 UF6 and
tions are often conducted batchwise or semicontinuously in 238 UF . Today, nonporous polymer membranes are employed
6
vessels or columns. However, equipment is available to simu- to enrich mixtures containing hydrogen, recover hydrocarbons
late continuous operation, as in Table 1.2. When the adsorbent, from gas streams, and produce oxygen-enriched air.
S, becomes saturated with the adsorbed solutes (adsorbate), To achieve high purities, pervaporation, Operation (4), can
it must be regenerated to recover the adsorbate and reuse the be used. Certain species in the liquid feed diffuse through the
adsorbent. Alternatively, if the adsorbate is a waste product, nonporous membrane, where they are evaporated before exit-
the spent adsorbent may be discarded and replaced with fresh ing as permeate. This method uses low pressures to enhance
adsorbent. Adsorption is described in Chapter 15. vaporization and the heat of vaporization must be supplied. It
is used to separate azeotropic mixtures.
§1.5 SEPARATIONS BY BARRIER All four of the membrane separation operations in Table 1.3
are described in Chapter 14.
The use of microporous and nonporous membranes as semi-
permeable barriers for separating vapor or liquid mixtures §1.6 SEPARATIONS BY AN EXTERNAL FIELD
is rapidly gaining adherents. The products are the retentate
OR GRADIENT
(components that do not pass through the membrane) and
the permeate (components that do pass through the mem- External fields take advantage of differing degrees of response
brane). For microporous membranes, separation is effected of certain molecules. Centrifugation establishes a pressure
by differing rates of species diffusion through the membrane field that separates mixtures according to their size, shape, and
pores. For nonporous membranes, separation is controlled by density. It is used to separate 235 UF6 from 238 UF6 , and can
differences in solubility in the membrane and rates of species also separate large polymer molecules according to molecular
diffusion through the membrane material. The most complex weight.
and selective membranes are found in the trillions of cells in If a temperature gradient is applied to a homogeneous
the human body. solution, concentration gradients induce thermal diffusion. It
Table 1.3 lists four of the most common industrial has been used to enhance separation of isotopes in permeation
membrane separation operations. Dialysis, Operation (1) in processes.
Table 1.3, is the transport, by a concentration gradient, of small When water containing 0.000149 atom fraction deuterium
solute molecules through a porous membrane. The molecules is decomposed by electrolysis into hydrogen and oxygen, the
unable to pass through the membrane are small, insoluble, deuterium concentration in the hydrogen is lower than in water.

Table 1.3 Common Separation Operations Based on Barriers

Separation Operation Symbol Feed Phase Barrier Separating Agent(s)
L L

(1) Dialysis Liquid Microporous membrane Pressure (ESA)

L

L L

(2) Reverse osmosis Liquid Microporous membrane Pressure (ESA)

L

V V

(3) Gas permeation Vapor Nonporous membrane Pressure (ESA)

V

L L

(4) Pervaporation Liquid Nonporous membrane Pressure and heat transfer (ESA)
V
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 8

8 Chapter 1 Separation Processes

Until 1953, this process was the source of heavy water (D2 O). Table 1.4 Ease of Scale-Up of the Most Common Separations
In electrodialysis, cation- and anion-permeable membranes Operation in Decreasing Ease of Need for
carry a fixed charge that prevents migration of species with like Ease of Scale-Up Staging Parallel Units
charge. This phenomenon is applied to desalinate seawater. A
Distillation Easy No need
related process is electrophoresis, which exploits the different
Absorption Easy No need
migration velocities of charged colloidal or suspended species Liquid–liquid Easy Sometimes
in an electric field. extraction
These external field separation operations are not discussed Membranes Re-pressurization Almost always
further in this textbook, with the exception of electrodialysis, required between
which is described in Chapter 14. stages
Adsorption Easy Only for
regeneration
§1.7 BRIEF COMPARISON OF COMMON cycle
SEPARATION OPERATIONS
When selecting among feasible separation techniques for
a given application, some major factors to consider are applies. Table 1.4 lists operations ranked according to ease of
(1) technological maturity, which allows designers to apply scale-up. Those ranked near the top are frequently designed
prior knowledge; (2) cost; (3) ease of scale-up from labora- without pilot-plant or laboratory data. Operations near the
tory experiments; (4) ease of providing multiple stages; and middle usually require laboratory data, while those near
(5) need for parallel units for large capacities. A survey the bottom require pilot-plant tests. Included in the table is
by Keller [3], Figure 1.8, shows that the degree to which an indication of the ease of providing multiple stages and
a separation operation is technologically mature correlates whether parallel units may be required. Ultimately, the most
with its extent of commercial use. Operations based on mem- cost-effective process, based on operating, maintenance, and
branes are more expensive than those based on phase creation capital costs, is selected, provided it is controllable, safe,
(e.g., distillation) or phase addition (e.g., absorption, extrac- and nonpolluting.
tion, and adsorption). All separation equipment is limited Also of interest are studies by Sherwood, Pigford, and
to a maximum size. For capacities requiring a larger size, Wilke [4], Dwyer [5], and Keller [3] that show that the cost
parallel units must be provided. Except for size constraints of recovering and purifying a chemical depends strongly on
or fabrication problems, capacity of a single unit can be dou- its concentration in the feed. Keller’s correlation, Figure 1.9,
bled for an additional investment cost of about 60%. If two shows that the more dilute the feed in the product, the higher
parallel units are installed, the additional investment for the the product price. The five highest priced and most dilute
second unit is 100% of the first unit, unless a volume-discount chemicals shown are all proteins.

1,000,000,000
Use Urokinase
asymptote 100,000,000
Factor
Distillation
10,000,000 VIII

Gas absorption 1,000,000 Luciferase

Ext./azeo. dist.
Crystallization 100,000
Ion exchange Solvent ext.
Use maturity

Price, $/lb

Insulin
Adsorption: gas feed 10,000

1,000
Adsorption: liquid feed Rennin
Supercritical Membranes: gas feed 100
gas abs./ext. Ag
Membranes: liquid feed Penicillin
Liquid Chromatography: liquid feed 10 Co
membranes Citric Hg
Field-induced separations Ni
1 Acid
Affinity separations Cu
First Zn
application
0.10
Invention Technology 1 0.1 0.01 0.001 10–4 10–5 10–6 10–7 10–8 10–9
asymptote
Technological maturity Weight fraction in substrate

Figure 1.8 Technological and use maturities of separation Figure 1.9 Effect of concentration of product in feed material on
processes. price.
[Reproduced from [3] with permission of the American Institute of [Reproduced from [3] with permission of the American Institute of
Chemical Engineers.] Chemical Engineers.]
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 9

§1.8 Separation Processes, Product Purity, Component Recovery, and Separation Sequences 9

§1.8 SEPARATION PROCESSES, PRODUCT

PURITY, COMPONENT RECOVERY, AND
SEPARATION SEQUENCES
Separation processes usually consist of more than one oper-
ation and may produce more than one product. The process
is designed to meet product specifications, given as product
purities and compositions. The process strives to do this with
high component recoveries.
Consider the continuous hydrocarbon recovery process
shown in Figure 1.10, which consists of a sequence of three
multistage distillation operations (C1, C2, and C3). The feed
consists of propane (C3 H8 ), iso-butane (iC4 H10 ), normal-
butane (nC4 H10 ), iso- and normal-pentane (iC5 H12 and
nC5 H12 ), higher molecular weight (heavier) hydrocarbons
(C+6 ), and a small amount of ethane (C2 H6 ). The compo-
nents in the feed are rank-ordered by decreasing volatility
(increasing normal boiling point), with ethane being the most
volatile. The distillation columns were designed with the Figure 1.10 Hydrocarbon recovery process using distillation.
Aspen Plus simulator, using mathematical models described
in Chapter 10, to produce four products: a C+5 -rich bottoms
from C1, a C3 (propane)-rich distillate from C2, and an iC4 C+5 -rich component for Stream 2 indicates that all components
(isobutane)-rich distillate and an nC4 (normal butane)-rich less volatile than the pentanes are included.
bottoms from C3. A material balance for the process is given Each column separates two key components in the column
in Table 1.5. Streams 4, 6, and 7 are final products, while feed. In Column C1, the separation is between nC4 H10 (light
Stream 2 is an intermediate that requires further processing. key) and iC5 H12 (heavy key). This is indicated by the hori-
Specifications for the products are included in Table 1.6. The zontal line separating the flow rates of these two components

Table 1.5 Material Balance for Hydrocarbon Recovery Process of Figure 1.10
Flow rates in lbmol/h
Stream 1 2 3 4 5 6 7
Component Feed to C1 C+5 -rich Product Feed to C2 C3 -rich Product Feed to C3 iC4 -rich Product nC4 -rich Product
C2 H6 0.60 0.00 0.60 0.60 0.00 0.00 0.00
C3 H8 57.00 0.00 57.00 54.80 2.20 2.20 0.00
iC4 H10 171.80 0.10 171.70 0.60 171.10 162.50 8.60
nC4 H10 227.30 0.70 226.60 0.00 226.60 10.80 215.80
iC5 H12 40.00 11.90 28.10 0.00 28.10 0.00 28.10
nC5 H12 33.60 16.10 17.50 0.00 17.50 0.00 17.50
C+6 205.30 205.30 0.00 0.00 0.00 0.00 0.00
Totals 735.60 234.10 501.50 56.00 445.50 175.50 270.00

Table 1.6 Comparison of Process Product Purities with Product Specifications

mol% in Product
Propane-Rich Iso-Butane-Rich Normal-Butane-Rich
Component Data Spec Data Spec Data Spec
C2 H6 1.07 5 max 0 0
C3 H8 97.86 93 min 1.25 3 max { 0 { 1 max
iC4 H10 1.07 2 max 92.59 92 min
83.11 80 min
nC4 H10 0 6.16 7 max
C5 + 0 0 16.89 20 max
Total 100.00 100.00 100.00
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 10

10 Chapter 1 Separation Processes

in the “Feed to C1” column of Table 1.5. Column C1 pro- For the permeate: nP = 10.5 + 9.9 = 20.4 kmol∕h
duces a C+5 -rich product (Stream 2) as bottoms. The distillate mol% O2 = 10.5∕20.4 = 0.515 = 51.5%
from C1 is the feed to C2. Column C2 separates between C3 H8 mol% N2 = 100 − 51.5 = 48.5%
(light key) and iC4 H10 (heavy key) to produce a C3 -rich prod-
For the retentate: nR = 69.1 + 10.5 = 79.6 kmol∕h
uct (Stream 4) as distillate. The bottoms from C2 is the feed
to C3, which produces two products by separating between mol% O2 = 10.5∕79.6 = 0.132 = 13.2%
iC4 H10 (light key) and nC4 H10 (heavy key). mol% N2 = 100 − 13.2 = 86.8%
In Table 1.5, it should be noted that an overall material Case 2: O2 recovery is given; its distribution to the products is
balance equation is satisfied for each of the six components
and the C+6 multicomponent. For Figure 1.10, that equation in n1,P = 0.50(21) = 10.5 kmol∕h
terms of ni, j , the lbmol∕h flow rate of component i in stream j is n1,R = 21 − 10.5 = 10.5 kmol∕h

ni,1 = ni,2 + ni,4 + ni,6 + ni,7 Using the purity of O2 in the permeate, the total permeate flow rate is
For example, for propane (C3 H8 ), 57.00 = 0.00 + 54.80 + nP = 10.5∕0.5 = 21 kmol∕h
2.20 + 0.00.
From the data in Table 1.5, product purities and compo- By a total permeate material balance:
nent recoveries for the process can be computed. For example,
n2,P = 21 − 10.5 = 10.5 kmol∕h
the mol% purity of propane in the C3 -rich product, Stream 4,
is 54.80∕56.00 = 0.9786 = 97.86 mol%. The propane recov- By an overall N2 material balance:
ery in this product is 54.80∕57.00 = 0.9614 = 96.14%. Other
components in the propane product are ethane and isobutene. n2,R = 79 − 10.5 = 68.5 kmol∕h
Table 1.6 compares the product purities of the final products
of Streams 4, 6, and 7 with commercial product specifications. For the permeate: nP = 21 kmol∕h
Note that all product specifications are achieved. mol% O2 = 10.5∕21 = 0.50 = 50%
The following example illustrates the use of material bal- mol% N2 = 100 − 50 = 50%
ances for a separator when product purities and/or component For the retentate: nR = 100 − 21 = 79 kmol∕h
recoveries are specified. mol% O2 = 10.5∕79 = 0.133 = 13.3%
mol% N2 = 100 − 13.3 = 86.7%
EXAMPLE 1.1 Material balances around a separator
Case 3: Two material-balance equations, one for each component,
A feed, F, of 100 kmol∕h of air containing 21 mol% O2 (1) and can be written:
79 mol% N2 (2) is to be partially separated by a gas permeation mem- For nitrogen, with a purity of 0.85 in the retentate and 1.00 − 0.50 =
brane unit, Operation (3) in Table 1.3, according to each of three 0.50 in the permeate,
sets of specifications. Compute the flow rates (nP and nR ) in kmol∕h
and compositions in mol% of the two products (retentate, R, and n2,F = 0.85nR + 0.50nP = 79 kmol∕h (1)
permeate, P). In Figure 1.6(c), Phase 1 is the retentate while Phase 2
is the permeate. The membrane is more permeable to O2 than to N2 . For oxygen, with a purity of 0.50 in the permeate and 1.00 − 0.85 =
0.15 in the retentate,
Case 1: 50% recovery of O2 to the permeate and 87.5% recovery of
N2 to the retentate. n1,F = 0.50nP + 0.15nR = 21 kmol∕h (2)
Case 2: 50% recovery of O2 to the permeate and 50 mol% purity of
O2 in the permeate. Solving (1) and (2) simultaneously for the total flow rates of the prod-
Case 3: 85 mol% purity of N2 in the retentate and 50 mol% purity of ucts gives
O2 in the permeate.
nP = 17.1 kmol∕h and nR = 82.9 kmol∕h

Solution Therefore, the component flow rates are

The feed (F) rates of oxygen (1) and nitrogen (2) are n1,P = 0.50(17.1) = 8.6 kmol∕h
n2,R = 0.85(82.9) = 70.5 kmol∕h
n1,F = 0.21(100) = 21 kmol∕h
n1,R = 82.9 − 70.5 = 12.4 kmol∕h
n2,F = 0.79(100) = 79 kmol∕h
n2,P = 17.1 − 8.6 = 8.5 kmol∕h

Case 1: Because two recoveries are given: For the permeate: nP = 17.1 kmol∕h
mol% O2 = 8.6∕17.1 = 0.503 = 50.3%
n1,P = 0.50(21) = 10.5 kmol∕h
mol% N2 = 100 − 50.3 = 49.7%
n2,R = 0.875(79) = 69.1 kmol∕h
For the retentate: nR = 100 − 17.1 = 82.9 kmol∕h
n1,R = 21 − 10.5 = 10.5 kmol∕h
mol% O2 = 12.4∕82.9 = 0.150 = 15.0%
n2,P = 79 − 69.1 = 9.9 kmol∕h mol% N2 = 100 − 15.0 = 85.0%
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 11

§1.8 Separation Processes, Product Purity, Component Recovery, and Separation Sequences 11

§1.8.1 Purity and Composition Designations products when each column has a single feed that is separated
into one distillate and one bottoms. Consider a hydrocarbon
The product purities in Table 1.6 are given in mol%, a des-
feed that consists, in the order of decreasing volatility, propane
ignation usually restricted to gas mixtures, for which vol%
(C3 ), isobutane (iC4 ), n-butane (nC4 ), isopentane (iC5 ), and
is equivalent to mol%. For liquids, purities are more often
n-pentane (nC5 ). A sequence of distillation columns is used
specified in mass fractions or wt%. Although vol% is also
to separate the feed into three nearly pure products of C3 , iC4 ,
common for liquid mixtures, it is not easily calculated when
and nC4 ; and one multicomponent product of combined iC5
the liquid mixture is a nonideal solution. To meet environ-
and nC5 . Figure 1.11 shows the five alternative sequences.
mental regulations, the allowable concentration of an impurity
When two final products are desired, only a single distillation
in gas, liquid, or solids streams is typically specified in parts
column is needed. For three final products, there are two alter-
of impurity per million parts (ppm) or parts of impurity per
native sequences. As the number of final products increases,
billion parts (ppb), where if a gas, the parts are moles or
the number of alternative sequences grows rapidly as shown
volumes; if a liquid or solid, then mass or weight. For aqueous
in Table 1.7.
solutions, especially those containing acids and bases, com-
For the initial selection of a feasible sequence, the following
mon designations for composition are molarity (M), which is
the molar concentration in moles of solute per liter of solution; heuristics (plausible but not infallible rules) are useful and easy
molality (m) in moles of solute per kilogram of solvent; and to apply, and do not require an economic evaluation:
normality (N) in number of equivalent weights of solute per 1. Remove unstable, corrosive, or chemically reactive
liter of solution. Concentrations (c) in mixtures can be in components early in the sequence. Then the materials
units of moles or mass per volume (e.g., mol/L, g/L, kg/m3 , of construction used in later columns will be less expen-
lbmol/ft3 , lb/ft3 ). For some chemical products, an attribute, sive. Also remove very volatile components early in the
such as color, may be used in place of purity in terms of sequence so that column pressures can be reduced in
composition. later columns.
2. Remove final products one by one, in order of decreas-
ing volatility or increasing boiling point, as overhead
§1.8.2 Alternative Separation Sequences distillates.
The three-distillation-column recovery process shown in 3. Remove, early in the sequence, those components of
Figure 1.10 is only one of five alternative sequences of distil- greatest molar percentage in the feed. The remaining
lation operations that can separate the process feed into four columns will be smaller in diameter.

Figure 1.11 Alternative distillation sequences to produce four products.

 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 12

12 Chapter 1 Separation Processes

Table 1.7 Number of Alternative Sequences applied. For determining an optimal sequence, Seider et al. [2]
Number of Final Number of Number of Alternative
present rigorous methods that do require column designs and
Products Columns Sequences economic evaluations. They also consider complex sequences
that include separators of different types and complexity.
2 1 1
3 2 2
4 3 5 EXAMPLE 1.2 Selection of a Separation Sequence
5 4 14 using Heuristics.
6 5 42
A distillation sequence is to produce four final products from five
hydrocarbons. Figure 1.11 shows the five possible sequences. The
4. Make the most difficult separations in the absence of the molar percentages in the process feed to the sequence are C3 (5.0%),
other components. This will usually lower the diameter iC4 (15%), nC4 (25%), iC5 (20%), and nC5 (35%). The most difficult
of the tallest column. separation by far is that between the isomers, iC4 and nC4 . Use the
heuristics to determine the best sequence(s). All products are to be of
5. Leave later in the sequence those separations that pro- high purity.
duce final products of the highest purities. This will also
lower the diameter of the tallest column.
Solution
6. Select the sequence that favors near-equimolar amounts
of distillate and bottoms in each column. Then the Heuristic 1 does not apply. Heuristic 2 favors taking C3 , iC4 , and nC4
as distillates in Columns 1, 2, and 3, respectively, with the multicom-
two sections of the column will tend to have the same
ponent product of iC5 and nC5 taken as the bottoms in Column 3.
diameter.
Heuristic 3 favors the removal of the multicomponent product (55%
Unfortunately, these heuristics sometimes conflict with of the feed) in Column 1. Heuristic 4 favors the separation of iC4
from nC4 in Column 3. Heuristics 3 and 4 can be combined with C3
one another so that one clear choice may not be possible. If
taken as distillate in Column 2. Heuristic 5 does not apply. Heuristic 6
applicable, Heuristic 1 should always be employed. The most
favors taking the multicomponent product as bottoms in Column 1
common industrial sequence is that of Heuristic 2. When (45/55 mole split), nC4 as bottoms in Column 2 (20/25 mole split),
energy costs are high, Heuristic 6 is favored because of lower and C3 as distillate with iC4 as bottoms in Column 3. Thus, the heuris-
utility costs. When one of the separations is particularly diffi- tics lead to three possible sequences as most favorable.
cult, such as the separation of isomers, Heuristic 4 is usually

SUMMARY
1. Industrial chemical processes include equipment for sep- 4. Separation operations are designed to achieve product
arating chemical mixtures in process feed(s) and/or purity and to strive for high recovery.
produced in reactors within the process. 5. A sequence of separators is usually required when more
2. The more widely used separation operations involve trans- than two products are to be produced or when the required
fer of species between two phases, one of which is created product purity cannot be achieved in a single separator.
by an energy separation agent (ESA) or the introduction of 6. The cost of purifying a chemical depends on its concentra-
a mass-separating agent (MSA). tion in the feed. The extent of industrial use of a particular
3. Less commonly used operations employ a barrier to prefer- separation operation depends on its cost and technological
entially pass certain species or a force field to cause species maturity.
to diffuse to another location at different rates.

REFERENCES
1. Cussler, E.L., and G.D. Moggridge, Chemical Product Design, Cam- 3. Keller, G.E., II, AIChE Monogr. Ser, 83(17) (1987).
bridge University Press, Cambridge, UK (2001).
4. Sherwood, T.K., R.L. Pigford, and C.R. Wilke, Mass Transfer,
2. Seider, W.D., J.D. Seader, D.R. Lewin and S. Widagdo, Product & McGraw-Hill, New York (1975).
Process Design Principles 3rd ed., John Wiley & Sons, Hoboken, NJ 5. Dwyer, J.L., Biotechnology, 1, 957 (Nov. 1984).
(2009).

STUDY QUESTIONS
1.1. What are the two key process operations in chemical 1.3. What are the four general separation techniques and what do
engineering? they all have in common?
1.2. What are the main auxiliary process operations in chemical 1.4. Why is the rate of mass transfer a major factor in separation
engineering? processes?
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 13

Exercises 13

1.5. What limits the extent to which the separation of a mixture can 1.9. What is the difference between adsorption and absorption?
be achieved? 1.10. The degree of separation in a separation operation is often spec-
1.6. What are the most common methods used to separate two fluid ified in terms of product purities and component recoveries.
phases? How do these two differ?
1.7. What is the difference between an ESA and an MSA? Give 1.11. What is a key component? What is a multicomponent product?
three disadvantages of using an MSA. 1.12. Why are sequences of separators sometimes necessary to sep-
1.8. What is the most widely used industrial separation operation? arate a feed mixture?

EXERCISES
Section 1.1 1.9. Removal of VOCs from a wastewater stream.
1.1. Description of the ethanol process. Many waste gas streams contain volatile organic compounds
Considering possible side reactions, describe as best you can what (VOCs), which must be removed. Recovery of the VOCs may be
takes place in each block of the process shown in Figure 1.4. accomplished by (1) absorption, (2) adsorption, (3) condensation,
(4) freezing, (5) membrane separation, or (6) catalytic oxidation.
Section 1.2 Discuss the pros and cons of each method, paying particular attention
1.2. Mixing vs. separation. to the fate of the VOC. For the case of a stream containing 3 mol%
Explain, using thermodynamic principles, why mixing pure chem- acetone in air, draw a flow diagram for a process based on absorption.
icals to form a homogeneous mixture is a spontaneous process, while Choose a reasonable absorbent and include in your process a means
separation of that mixture into its pure species is not. to recover the acetone and recycle the absorbent.
1.10. Separation of air.
1.3. Separation of a mixture requires energy. Describe three methods suitable for the separation of air into nitro-
Explain, using the laws of thermodynamics, why the separation gen and oxygen.
of a mixture into pure species or other mixtures of differing compo-
sitions requires energy transfer to the mixture. Section 1.8
1.11. Material balance for a distillation sequence.
Sections 1.3 and 1.4
The feed to Column C3 in Figure 1.10 is given in Table 1.5. The
1.4. ESA vs. MSA. separation is to be altered to produce a distillate of 95 mol% pure
Compare the advantages and disadvantages of making separations isobutane with a recovery in the distillate of 96%. Because of the
using an ESA versus using an MSA. sharp separation in Column C3 between iC4 and nC4 , assume all
1.5. Differences among distillation and liquid–liquid propane goes to the distillate and C5 s to the bottoms.
extraction. (a) Compute the flow rates in lbmol∕h of each component in each of
Under what conditions should liquid–liquid extraction be consid- the two products leaving Column C3.
ered over distillation? (b) What is the percent purity of the n-butane bottoms product?
(c) If the isobutane purity in the distillate is fixed at 95%, what
Section 1.5 percent recovery of isobutane in the distillate will maximize the
1.6. Osmotic pressure. percent purity of normal butane in the bottoms product?
The osmotic pressure, π, of seawater is given by π = RTc∕M,
1.12. Material balance for a distillation sequence.
where c is the concentration of the dissolved salts (solutes) in
Five hundred kmol∕h of liquid alcohols containing, by moles,
g∕cm3 , M is the average molecular weight of the solutes as ions, T
40% methanol (M), 35% ethanol (E), 15% isopropanol (IP), and
is temperature in Kelvin, and R is the ideal gas constant. Consider
10% normal propanol (NP) is distilled in two distillation columns in
recovering pure water from seawater containing 0.035 g of salts∕cm3
series. The distillate from the first column is 98% pure M with a 96%
of seawater and M = 31.5, at 298 K. What is the minimum required
recovery of M. The distillate from the second is 92% pure E with a
pressure difference across the membrane in kPa to just overcome the
95% recovery of E from the process feed. Assume no propanols in
osmotic pressure?
the distillate from Column C1, no M in the bottoms from Column
1.7. Basic separation techniques. C2, and no NP in the distillate from Column C2.
For each of the following separation operations, state the basic (a) Compute flow rates in kmol∕h of each component in each feed,
separation process technique shown in Figure 1.6: absorption, adsorp- distillate, and bottoms. Draw a labeled block-flow diagram.
tion, dialysis, distillation, flash vaporization, gas permeation, liquid– Include the material balances in a table, similar to Table 1.5.
liquid extraction, pervaporation, reverse osmosis, and stripping. (b) Compute the mole-percent purity of the propanol mixture leaving
as bottoms from the second column.
Section 1.7
(c) If the recovery of ethanol is fixed at 95%, what is the maximum
1.8. Removing organic pollutants from wastewater. purity of the ethanol in the distillate from the second column?
The need to remove organic pollutants from wastewater is com-
(d) If instead, the purity of the ethanol is fixed at 92%, what is the
mon to many industrial processes. Separation methods to be consid-
maximum recovery of ethanol (based on the process feed)?
ered are: (1) adsorption, (2) distillation, (3) liquid–liquid extraction,
(4) membrane separation, (5) stripping with air, and (6) stripping with 1.13. Material balance for separation by pervaporation.
steam. Discuss the advantages and disadvantages of each method. Ethanol and benzene are separated in a network of distillation
Consider the fate of the organic material. and membrane separation steps. In one step, a near-azeotropic liquid
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 14

14 Chapter 1 Separation Processes

mixture of 8,000 kg∕h of 23 wt% ethanol in benzene is fed to a per- the adsorption step. The adsorbate, which exits the adsorber during
vaporation membrane consisting of an ionomeric film of perfluoro- regeneration at 100∘ F and 15 psia, is compressed to 800 psia and
sulfonic polymer cast on a Teflon support. The membrane is selective cooled to 100∘ F before being combined with non-permeate gas to
for ethanol, producing a vapor permeate containing 60 wt% ethanol, give the final pipeline natural gas.
while the non-permeate (retentate) liquid contains 90 wt% benzene. (a) Draw a process flow diagram of the process using appropriate
(a) Draw a flow diagram of the pervaporation using symbols from symbols. Include compressors and heat exchangers. Label the
Table 1.3 and include all process information. diagram with the data given and number all streams.
(b) Compute the component flow rates in kg∕h in the feed stream (b) Compute component flow rates of N2 , CH4 , and C2 H6 in lbmol∕h
and in the product streams, and enter these results in the diagram. and create a material-balance table similar to Table 1.5.
(c) What separation operation could be used to purify the vapor 1.17. Partial condensation of a reactor effluent.
permeate? Toluene (methylbenzene) is used as a solvent and a fuel and is
1.14. Material balance for an absorption-distillation sequence. a precursor for making benzene by hydrodealkylation; benzene and
1,000 kmol∕h of a gas mixture containing 90 mol% acetone and xylenes by disproportionation; and 2,4,6-trinitrotoluene (TNT)—an
10% nitrogen is to be processed to recover the acetone. First the gas explosive used by the military and construction companies—by
is fed to a multistage absorber where 95% of the acetone is absorbed nitration. Consider a hydrodealkylation process, where the main
with 500 kmol∕h of a liquid absorbent of methyl-isobutyl-ketone reaction is:
(MIBK). Two percent of the entering MIBK is lost to the gas leaving Toluene + H2 → Benzene + CH4
the top of the absorber. Nitrogen does not dissolve in the MIBK.
The liquid leaving the bottom of the absorber is sent to a distilla- with an undesirable side reaction:
tion column to separate the acetone from the MIBK. In this column, 2 Benzene ⇌ Biphenyl + H2
a distillate of 98 mol% acetone and a bottoms of 99 mol% MIBK is
obtained. The reactor effluent is cooled and partially condensed. The liquid
(a) Draw a flow diagram of the separation process and place the phase is then distilled in a sequence of three distillation columns to
above data on the diagram. give the following four products, one of which is recycled back to
the reactor.
(b) For each stream in the process, calculate by material balances
the flow rate of each component and enter your results in a table Benzene Toluene Biphenyl
similar to Table 1.5. Off-gas, Product, Recycle, By-product,
1.15. Separation by Gas Permeation. Component kmol/h kmol/h kmol/h kmol/h
The Prism gas permeation process developed by the Monsanto Hydrogen 1.48 0.00 0.00 0.00
Company is selective for hydrogen when using hollow-fiber mem- Methane 11.07 0.06 0.00 0.00
branes made of silicone-coated polysulphone. A feed gas at 16.7 MPa Benzene 0.80 264.72 1.33 0.00
and 40∘ C containing 42.4 H2 , 7.0 CH4 , and 0.5 N2 kmol∕h is sepa- Toluene 0.00 0.09 88.79 0.45
rated by the membrane into a retentate gas at 16.2 MPa and a permeate Biphenyl 0.00 0.00 0.02 4.59
gas at 4.56 MPa. Total 13.35 264.87 90.14 5.04
(a) Assume the membrane is non-permeable to nitrogen, the recov-
ery of H2 is 60.38%, and the mole ratio of CH4 to H2 in the
permeate is 0.0117. Calculate the total flow rate of the retentate (a) Calculate for the sequence the percent recoveries for benzene
and permeate gases and the flow rates of each component in them. and biphenyl.
(b) Compute the percent purity of the hydrogen in the permeate gas. (b) Calculate the mol% purities for benzene and biphenyl.
(c) Draw a process-flow diagram that displays pressure and compo- (c) Why is there a toluene recycle stream?
nent flow rates. (d) What would happen if the biphenyl was not separated from the
1.16. Separation by membrane separation and adsorption. toluene, but was recycled back to the reactor with the toluene?
Nitrogen is injected into oil wells to increase the recovery of 1.18. Separation by Gas Permeation.
crude oil (enhanced oil recovery). It mixes with the natural gas In a hydrodealkylation process of the type stated in Exercises 1.17
that is produced along with the oil. The nitrogen must be separated and 1.22, a gas rich in hydrogen and methane is separated from the
from the natural gas. A total of 170,000 SCFH (based on 60∘ F reactor effluent by partial condensation. The gas is then sent to a gas
and 14.7 psia) of natural gas containing 18% N2 , 75% CH4 , and permeation membrane separator (Operation 3 in Table 1.3) to sepa-
7% C2 H6 at 100∘ F and 800 psia is to be processed in two steps to rate the hydrogen from the methane. The permeate is recycled to the
reduce nitrogen content to 3 mol%: (1) membrane separation with reactor and the retentate is used for fuel. The component flow rates
a nonporous glassy polyimide membrane, followed by (2) adsorp- for the membrane feed, retentate, and permeate are as follows:
tion using molecular sieves to which the permeate gas is fed. The
membrane separator is highly selective for N2 (90.83% recovery),
and completely impermeable to ethane. The mole ratio of CH4 to Component Feed, kmol/h Retentate, kmol/h Permeate, kmol/h
N2 in the permeate is 1.756. The adsorption step selectively adsorbs Hydrogen 1575.7 157.6 1418.1
methane, giving 97% pure methane in the adsorbate, with an 85% Methane 2388.1 2313.5 74.6
recovery of CH4 fed to the adsorber. The non-permeate (retentate) Benzene 29.8 29.8 0.0
gas from the membrane step and adsorbate from the adsorption step Toluene 3.5 3.5 0.0
are combined to give a methane stream that contains 3.0 mol% N2 . Biphenyl 0.0 0.0 0.0
The pressure drop across the membrane is 760 psia. The permeate at Total 3997.1 2504.4 1492.7
20∘ F is compressed to 275 psia and cooled to 100∘ F before entering
 Trim Size: 8.5in x 11in Seader c01.tex V2 - 10/16/2015 10:38 A.M. Page 15

Exercises 15

(a) Calculate the percent recoveries and mol% purities of hydrogen (b) Complete a material balance for each column and summarize the
in the permeate and methane in the retentate. (b) Why not sepa- results in a table similar to Table 1.5. To complete the balance,
rate hydrogen from methane by distillation or absorption? you must make assumptions about the flow rates of: (1) isobutane
in the distillates for Columns 1 and 3 and (2) n-butane in the
1.19. Distillation sequences.
distillates for Columns 1 and 2, consistent with the specified
The feed stream in the table below is to be separated into four
recoveries. Assume no propane in the distillate from Column 3
nearly pure products. None of the components is corrosive and, based
and no pentanes in the distillate from Column 2.
on the boiling points, none of the three separations is difficult. As seen
in Figure 1.11, five distillation sequences are possible. (a) Determine (c) Calculate the mol% purities of the products and summarize your
a suitable sequence of three columns using the heuristics of §1.8.2. results as in Table 1.6, but without the specifications.
(b) If a fifth component were added to give five products, Table 1.7 1.22. Distillation sequence for a wide-boiling mixture.
indicates that 14 alternative distillation sequences are possible. Draw, The effluent from the reactor of a toluene hydrodealkylation pro-
in a manner similar to Figure 1.10, all 14 of these sequences. cess of the type stated in Exercise 1.17 is as follows:

Component Feed rate, kmol/h Normal boiling point, K Component Boiling Point, ∘ C Flow Rate, kmol/h

Methane 19 112 Hydrogen –252.8 1577.1

Benzene 263 353 Methane –169.0 2399.2
Toluene 85 384 Benzene 80.1 296.6
Ethylbenzene 23 409 Toluene 110.6 92.8
Biphenyl 255.2 4.6
Total 4370.3
1.20. Distillation sequence for a solution from
fermentation.
The reactor effluent at 1265∘ F and 500 psia is to be separated into
Starch from corn or sugarcane can be fermented to an aqueous
four products: (a) a gaseous mixture rich in hydrogen and methane;
solution of acetone (A), n-butanol (B), and ethanol (E). Typically, the
(b) a benzene-rich product of high purity; (c) a toluene-rich mixture of
mass ratio of bioproducts is 3(A):6(B):1(E). The solution contains
low purity that is recycled to the reactor; and (c) a biphenyl by-product
33 g of bioproducts per liter of water. After removal of solid particles
of high purity. Using the heuristics of §1.8.2, determine one or two
from the broth by centrifugation, the remaining liquid is distilled in a
favorable sequences. A sequence can contain one single-stage partial
sequence of distillation columns to recover (1) acetone with a maxi-
condenser in place of one multistage distillation column.
mum of 10 wt% water; (2) ethanol with a maximum of 10 wt% water;
(3) n-butanol (99.5 wt% purity with a maximum of 0.5% water); and 1.23. Distillation sequence for an alcohol mixture.
(4) water (W), which can be recycled to the fermenter. If the four A mixture of five alcohols is to be separated into nearly pure prod-
products distill according to their normal boiling points in ∘ C of 56.5 ucts in a sequence of distillation columns, each of which produces
(A), 117 (B), 78.4 (E), and 100 (W), devise a suitable distillation a distillate and a bottoms. According to Table 1.7, four columns
sequence using the heuristics of §1.8.2. are needed that can be arranged into 14 different sequences. Using
the heuristics of §1.8.2, determine one or two feasible sequences.
1.21. Separation by a distillation sequence. The following table lists the alcohols in the order of volatility. The
A light-hydrocarbon feed stream contains 45.4 kmol/h of propane, volatilities relative to the fifth (least volatile) alcohol, 1-hexanol, are
136.1 kmol∕h of isobutane, 226.8 kmol∕h of n-butane, 181.4 kmol∕h shown. Also included are the flow rates of the five alcohols in the
of isopentane, and 317.4 kmol∕h of n-pentane. This stream is to feed to the sequence.
be separated by a sequence of three distillation columns into four
products: (1) propane-rich, (2) isobutane-rich, (3) n-butane-rich,
and (4) combined pentanes–rich. The first-column distillate is the Feed flow rate, Volatility relative to
propane-rich product; the distillate from Column 2 is the isobutane- Alcohol kmol/h 1-hexanol
rich product; the distillate from Column 3 is the n-butane-rich
1-Butanol 360 3.6
product, and the combined pentanes are the Column 3 bottoms.
2-Methyl-1-Butanol 180 2.5
The recovery of the main component in each product is 98%. For
3-Methyl-2-Butanol 360 2.3
example, 98% of the propane in the feed stream appears in the
1-Pentanol 2520 1.8
propane-rich product.
1-Hexanol 3600 1.0
(a) Draw a process-flow diagram for this distillation sequence, Total 7020
similar to Figure 1.10.
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 16

Chapter 2

Thermodynamics of Separation Operations

§2.0 INSTRUCTIONAL OBJECTIVES

After completing this chapter, you should be able to:

• Explain phase equilibria in terms of Gibbs free energy, chemical potential, fugacity, fugacity coefficient, activity, and
activity coefficient.
• Understand the usefulness of phase equilibrium ratios (e.g., K-values and distribution ratios) for determining vapor
and liquid phase compositions.
• Derive K-value expressions in terms of fugacity coefficients and activity coefficients.
• Explain how computer programs use equation of state (EOS) models to compute thermodynamic properties of vapor
and liquid mixtures, including K-values.
• Explain how computer programs use liquid-phase activity-coefficients derived from Gibbs excess free-energy models
to compute thermodynamic properties, including K-values.
• Make energy, entropy, and exergy (availability) balances around a separation process.

T hermodynamic properties play a major role in designing data is from non-English sources, industry, and MS and PhD
and simulating separation operations with respect to energy theses. The DDB also presents comparisons of experimental
requirements, phase equilibria, and equipment sizing. This data with various estimation methods described in this chapter.
chapter reviews methods for calculating molar volume or den-
sity, enthalpy, entropy, exergy (availability), fugacities, activity §2.1 PHASE EQUILIBRIA
coefficients, and phase equilibria ratios of ideal and nonideal
vapor and liquid mixtures as functions of temperature, pres- Many separations are determined by the extent to which
sure, and composition. These thermodynamic properties are species are partitioned among two or more phases at equilib-
used for determining compositions at phase-equilibrium, and rium at a specified T and P. The distribution is determined by
for making energy balances, entropy balances, and exergy application of the Gibbs free energy, G. For each phase in a
balances to determine energy efficiency. Emphasis is on the multiphase, multicomponent system, the Gibbs free energy is
thermodynamic property methods most widely used in process G = G{T, P, N1 , N2 , . . . , NC } (2-1)
simulators.
Experimental thermodynamic property data should be where T = temperature, P = pressure, and Ni = moles of
used, when available, to design and analyze separation oper- species i. At equilibrium, the total G for all phases is a
ations. When not available, properties can often be estimated minimum, and methods for determining this are referred
with reasonable accuracy by methods discussed in this chapter. to as free-energy minimization techniques. Gibbs free
The most comprehensive source of thermodynamic proper- energy is also the starting point for the derivation of com-
ties for pure compounds and nonelectrolyte and electrolyte monly used equations for phase equilibria. From classical
mixtures—including excess volume, excess enthalpy, activity thermodynamics, the total differential of G is
coefficients at infinite dilution, azeotropes, and vapor–liquid,
liquid–liquid, and solid–liquid equilibrium—is the comput- ∑
C
dG = −S dT + V dP + μi dNi (2-2)
erized Dortmund Data Bank (DDB), described briefly at i=1
www.ddbst.com, and in detail by Gmehling, et al. [1]. It was
initiated by Gmehling and Onken in 1973. It is updated annu- where S = entropy, V = volume, and μi is the chemical poten-
ally and widely used by industry and academic institutions on tial or partial molar Gibbs free energy of species i. For a closed
a stand-alone basis or with process simulators via the DDB system consisting of two or more phases in equilibrium, where
software package (DDBST). In 2014, the DDB contained each phase is an open system capable of mass transfer with
more than 6.4 million data sets for more than 49,000 compo- another phase,
nents from more than 65,400 literature references. The DDB [C ]
∑
NP
∑ (p) (p)
contains openly available experimental data from journals, dGsystem = μi dNi (2-3)
which can be searched free of charge. A large percentage of the p=1 i=1 P,T

16
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 17

§2.1 Phase Equilibria 17

where superscript (p) refers to each of NP phases. Conserva- relationship between fugacity and pressure, it is convenient to
tion of moles of species, in the absence of chemical reaction, define a pure-species fugacity coefficient, ϕi , as
requires that
fi
∑NP
(p) ϕi = (2-11)
dNi(1) = − dNi (2-4) P
p=2
which is 1.0 for an ideal gas. For a mixture, partial fugacity
which, upon substitution into (2-3), gives coefficients for vapor and liquid phases, respectively, are
[C ] f̄
∑NP
∑ ( (p) )
(p) ϕ̄ iV ≡ iV (2-12)
μi − μ(1)
i dNi =0 (2-5) yi P
p=2 i=1
f̄
(p) ϕ̄ iL ≡ iL (2-13)
With dNi(1) eliminated in (2-5), each dNi
term can be varied xi P
(p)
independently of any other dNi term. But this requires that
(p) As ideal-gas behavior is approached, ϕ̄ iV → 1.0 and
each coefficient of dNi in (2-5) be zero. Therefore,
ϕ̄ iL → Psi ∕P, where Psi = vapor pressure.
(NP ) At a given temperature, the ratio of the partial fugacity of a
μ(1) (2) (3)
i = μi = μi = · · · = μi (2-6)
component to its fugacity in a standard state, fio , is termed the
Thus, the chemical potential of a species in a multicompo- activity, ai . If the standard state is selected as the pure species
nent system is identical in all phases at physical equilibrium. at the same pressure and phase condition as the mixture, then
This equation is the basis for the development of all phase-
f̄i
equilibrium calculations. ai ≡ (2-14)
fio

Since at phase equilibrium, the value of fio is the same for each
§2.1.1 Fugacities and Activity Coefficients phase, substitution of (2-14) into (2-8) gives another alterna-
Chemical potential is not an absolute quantity, and the numer- tive condition for phase equilibria,
ical values are difficult to relate to more easily understood (NP )
physical quantities. Furthermore, the chemical potential appro- a(1) (2) (3)
i = ai = ai = · · · = ai (2-15)
aches an infinite negative value as pressure approaches zero.
Thus, the chemical potential is not a favored property for For an ideal solution, aiV = yi and aiL = xi .
phase-equilibria calculations. Instead, fugacity, invented by To represent departure of activities from mole fractions
G. N. Lewis in 1901, is employed as a surrogate. when solutions are nonideal, activity coefficients based on
The partial fugacity of species i in a mixture is like a concentrations in mole fractions are defined by
pseudo-pressure, defined in terms of the chemical potential by aiV
γiV ≡ (2-16)
(μ ) yi
f̄i =  exp i (2-7) a
RT γiL ≡ iL (2-17)
xi
where  is a temperature-dependent constant. Regardless of
the value of , it is shown by Prausnitz, Lichtenthaler, and de For ideal solutions, γiV = 1.0 and γiL = 1.0
Azevedo [2] that (2-6) can be replaced with For convenient reference, thermodynamic quantities useful
(N )
in phase equilibria are summarized in Table 2.1.
f̄i(1) = f̄i(2) = f̄i(3) = · · · = f̄i P (2-8)

where, f̄i is the partial fugacity of species i. Thus, at equi- §2.1.2 Definitions of K-Values
librium, a given species has the same partial fugacity in
each phase. This equality, together with equality of phase A phase-equilibrium ratio is the ratio of mole fractions of
temperatures and pressures, a species in two phases at equilibrium. For vapor–liquid sys-
tems, the ratio is called the K-value or vapor–liquid equilib-
T (1) = T (2) = T (3) = · · · = T (NP ) (2-9) rium ratio:
(1) (2) (3) (NP ) y
P =P =P =···=P (2-10) Ki ≡ i (2-18)
xi
constitutes the well-accepted conditions for phase equilibria.
For a pure component, the partial fugacity, f̄i , becomes the For the liquid–liquid case, the ratio is a distribution ratio,
pure-component fugacity, fi . For a pure, ideal gas, fugac- partition coefficient, or liquid–liquid equilibrium ratio:
ity equals the total pressure, and for a component in an
ideal-gas mixture, the partial fugacity equals its partial pres- xi(1)
sure, pi = yi P, such that the sum of the partial pressures KDi ≡ (2-19)
xi(2)
equals the total pressure (Dalton’s Law). Because of the close
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 18

18 Chapter 2 Thermodynamics of Separation Operations

Table 2.1 Thermodynamic Quantities for Phase Equilibria

Limiting Value for Ideal Gas
Thermodynamic Quantity Definition Physical Significance and Ideal Solution
( )
∂G
Chemical potential μi ≡ Partial molar free energy, ḡ i μi = ḡ i
∂Ni P,T,Nj
(μ )
Partial fugacity f̄i ≡  exp i
Thermodynamic pressure f̄iV = yi P
RT
f̄iL = xi PSi

fi ϕiV = 1.0
Fugacity coefficient of a pure ϕi ≡ Deviation to fugacity due to pressure
P
species PS
ϕiL = i
P
f̄
Partial fugacity coefficient of ϕ̄ iV ≡ iV Deviations to fugacity due to pressure ϕ̄ iV = 1.0
yi P
a species in a mixture and composition PS
f̄ ϕ̄ iL = i
ϕ̄ iL ≡ iL P
xi P

f̄i aiV = yi
Activity ai ≡ Relative thermodynamic pressure
fio
aiL = xi
a
Activity coefficient γiV ≡ iV Deviation to fugacity due to γiV = 1.0
yi
aiL composition γiL = 1.0
γiL ≡
xi

For equilibrium-stage calculations, separation factors are defi- or

ned by forming ratios of equilibrium ratios. For the vapor– f̄iL = ϕ̄ iL xi P (2-24)
liquid case, relative volatility αi, j between components i and
j is given by and
K f̄iV = ϕ̄ iV yi P (2-25)
αi, j ≡ i (2-20)
Kj
If (2-24) and (2-25) are used with (2-18), an equation-of-
Separations are easy for very large values of αi, j , but become state (EOS) formulation of the K-value follows, where the
impractical for values close to 1.00. partial fugacity coefficients are obtained from an EOS, as
Similarly for the liquid–liquid case, the relative selectivity described later in this chapter.
βi, j is
ϕ̄
K Di Ki = ̄ iL (2-26)
βi, j ≡ (2-21) ϕiV
KDj
If (2-23) and (2-25) are used, an activity coefficient or
Equilibrium ratios are formulated in a variety of ways using gamma-phi formulation of the K-value is obtained:
the thermodynamic quantities in Table 2.1. The two of most
practical interest are formulated next. γ fo γ ϕ
Ki = ̄iL iL = iL̄ iL (2-27)
ϕiV P ϕiV
§2.1.3 Rigorous K-Value Formulations
For vapor–liquid equilibrium, (2-8) becomes, for each If the ideal gas EOS, Pv = RT applies—where v =
component, molar volume and R = the ideal-gas constant—and both the
f̄iV = f̄iL (2-22) vapor and liquid form ideal solutions, then γiL = 1.0, ϕiL =
Psi ∕P, and ϕ̄ iV = 1.0. If these are substituted into (2-27), the
To form an equilibrium ratio, partial fugacities are commonly result is Raoult’s law or ideal K-value expression:
replaced by expressions involving mole fractions. From the
definitions in Table 2.1: Psi
Ki = (2-28)
f̄iL = γiL xi f iL
o
(2-23) P
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 19

§2.1 Phase Equilibria 19

If, however, the liquid phase is a nonideal solution, For practical applications, the choice of K-value formulation
γiL ≠ 1.0 and (2-28) becomes the modified Raoult’s law is a compromise among accuracy, complexity, convenience,
K-value expression: and past experience.
For liquid–liquid equilibria, (2-8) becomes
γiL Psi
Ki = (2-29)
P f̄iL(1) = f̄iL(2) (2-32)
For moderate pressures, the Poynting correction is intro- where superscripts (1) and (2) refer to the immiscible liquid
duced into (2-27) by approximating the pure-component liquid phases. A rigorous formulation for the distribution coefficient
fugacity coefficient in (2-27) by is obtained by combining (2-23) with (2-19) to obtain an
( ) expression involving only activity coefficients:
s P
P 1
ϕiL = ϕsiV i exp v dP (2-30)
P RT ∫Ps iL γ(2) γ(2)
i xi(1) f o(2)
KDi = = iL iL
= iL
(2-33)
where ϕsiV = pure component vapor fugacity coefficient at xi(2) γ(1) f
iL iL
o(1) γ (1)
iL
the saturation pressure. The exponential term is the Poynting
correction. If the liquid molar volume, v, is reasonably con- For vapor–solid equilibria, if the solid phase consists of
stant over the pressure range, the integral in (2-28) becomes just one of the components of the vapor phase, combination
viL (P − Psi ). For moderate to high pressures, an EOS is used of (2-8) and (2-25) gives
to obtain ϕ̄ iV in (2-27).
For a low-molecular-weight gas species, whose tempera- fiS = ϕ̄ iV yi P (2-34)
ture at the critical point, Tc , is less than the system temperature,
At low pressures, ϕ̄ iV = 1.0 and the fugacity of the solid is
the Henry’s law form for the K-value is convenient, provided
approximated by its vapor pressure. Thus, for the vapor-phase
Hi , the Henry’s law coefficient, is available. It depends on
mole fraction of the component forming the solid phase:
composition, temperature, and pressure. At low to moderate
pressures, it replaces the vapor pressure in (2-28) to give (Psi )solid
yi = (2-35)
P
Hi
Ki = (2-31) For liquid–solid equilibria, if the solid phase is a pure com-
P ponent, the combination of (2-8) and (2-23) gives
Table 2.2 lists the above-mentioned vapor–liquid K-value
fiS = γiL xi f iL
o
(2-36)
expressions and includes recommendations for their appli-
cation. At low pressure, the fugacity of a solid is approximated by
Regardless of which thermodynamic formulation in vapor pressure to give, for a component in the solid phase,
Table 2.2 is used for estimating K-values, its accuracy depends
on the correlations used for the thermodynamic properties (Psi )solid
xi = (2-37)
(vapor pressure, activity coefficient, and fugacity coefficients). γiL (Psi )liquid

Table 2.2 Useful K-Value Expressions for Estimating Vapor–Liquid Equilibria (Ki = yi /xi )
Equation Recommended Application
Rigorous forms:
ϕ̄
(1) Equation-of-state Ki = ̄ iL Hydrocarbon and light gas mixtures from cryogenic temperatures to
ϕiV the critical region
γiL ϕiL
(2) Activity coefficient Ki = All mixtures from ambient to near-critical temperature
ϕ̄ iV
Approximate forms:
Psi
(3) Raoult’s law (ideal) Ki = Ideal solutions at near-ambient pressure
P
γiL Psi
(4) Modified Raoult’s law Ki = Nonideal liquid solutions at near-ambient pressure
P ( )
( s
) P
Pi 1
(5) Poynting correction Ki = γiL ϕsiV exp v dP Nonideal liquid solutions at moderate pressure and below the
P RT ∫Ps iL
i critical temperature
Hi
(6) Henry’s law Ki = Low-to-moderate pressures for species at supercritical temperature
P
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 20

20 Chapter 2 Thermodynamics of Separation Operations

Table 2.3 Thermodynamic Properties for Ideal Mixtures

EXAMPLE 2.1 K-Values from Raoult’s and Henry’s
Laws. Ideal gas and ideal-gas solution:

Estimate K-values and the relative volatility, αM,W , of a vapor–liquid V M RT ∑

C

mixture of water (W) and methane (M) at P = 2 atm, T = 20 and (1) vV = = = , M= yi Mi

∑C ρV P
80∘ C. What is the effect of T on the K-values? Ni
i=1

i=1
Solution
At these conditions, water exists mainly in the liquid phase and will ∑
C T ∑
C
(2) hV = yi (CPo )iV dT = yi hoiV
follow Raoult’s law (2-28) if little methane dissolves in the water. i=1
∫To i=1
Because methane has a critical temperature of −82.5∘ C, well below
the temperatures of interest, it will exist mainly in the vapor phase ( )
∑
C T
(CPo )iV P ∑
C
and follow Henry’s law (2-31). The Aspen Plus process simulator is (3) sV = yi dT − R ln −R yi ln yi ,
used to make the calculations using the Ideal Properties option with i=1
∫To T Po i=1
methane as a Henry’s law component. The Henry’s law constants for
the solubility of methane in water are provided in the simulator data where the first term is soV
bank. The results are as follows:
Ideal-liquid solution:

T, ∘ C KW KM αM,W V M ∑
C
∑
C
(4) vL = = = xi viL , M= xi Mi
20 0.01154 18,078 1,567,000 ∑
C ρL i=1 i=1
80 0.23374 33,847 144,800 Ni
i=1

For both temperatures, the mole fraction of methane in the water is ∑

C
vap
less than 0.0001. The K-values for H2 O are low but increase rapidly (5) hL = xi (hoiV − ΔHi )
with temperature. The K-values for methane are extremely high and i=1

change much less rapidly with temperature. [) ( ]

∑
C
CPo iV T
ΔHi
vap
(6) sL = xi dT −
i=1
∫To T T
§2.2 IDEAL-GAS, IDEAL-LIQUID-SOLUTION ( )
P ∑ C

MODEL − R ln − R xi ln xi
Po i=1
Classical thermodynamics provides a means for obtaining
fluid thermodynamic properties in a consistent manner from Vapor–liquid equilibria:
P–v–T EOS models. The simplest model applies when both
liquid and vapor phases are ideal solutions (all activity coef- Psi
(7) Ki =
ficients equal 1.0) and the vapor is an ideal gas. Then the P
thermodynamic properties of mixtures can be computed Reference conditions (datum): h, ideal gas at To and zero pressure; s, ideal
from pure-component properties of each species using the gas at To and Po = 1 atm.
equations given in Table 2.3. These ideal equations apply only Refer to elements if chemical reactions occur; otherwise refer to
components.
at low pressures—not much above ambient—for components
of similar molecular structure.
The vapor molar volume, vV , and mass density, ρV , are com- where the constants, a, depend on the species. Values of the
puted from (1), the ideal-gas law in Table 2.3. It requires only constants for hundreds of compounds, with T in K, are tabu-
the mixture molecular weight, M, and the gas constant, R. It lated by Poling, Prausnitz, and O’Connell [3]. Because CP =
assumes that Dalton’s law of additive partial pressures and dh∕dT, (2-38) can be integrated for each species to give the
Amagat’s law of additive volumes apply. ideal-gas species molar enthalpy:
The molar vapor enthalpy, hV , is computed from (2) in
Table 2.3 by integrating an equation in temperature for the T ∑
5
k−1 (T − Tok )R
a k
zero-pressure heat capacity at constant pressure, CPo V , starting hoV = CPo V dT = (2-39)
∫To k
k=1
from a reference (datum) temperature, To , to the temperature
of interest, and then summing the resulting species vapor
The molar vapor entropy, sV , is computed from (3) in
enthalpies on a mole-fraction basis. Typically, To is taken
Table 2.3 by integrating CPo V ∕T from To to T for each species;
as 25∘ C, although 0 K is also common. Pressure has no
summing on a mole-fraction basis; adding a term for the
effect on the enthalpy of an ideal gas. A number of empirical
effect of pressure referenced to a datum pressure, Po , which is
equations have been used to correlate the effect of temperature
on the zero-pressure vapor heat capacity. An example is the generally taken to be 1 atm (101.3 kPa); and adding a term for
fourth-degree polynomial: the entropy change of mixing. Unlike the ideal vapor enthalpy,
[ ] the ideal vapor entropy includes terms for the effects of pres-
CPo V = a0 + a1 T + a2 T 2 + a3 T 3 + a4 T 4 R (2-38) sure and mixing. The reference pressure is not zero, because
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 21

§2.3 Graphical Representation of Thermodynamic Properties 21

the entropy is infinity at zero pressure. If (2-38) is used for the Note that from (2-20), the ideal relative volatility using (2-47)
heat capacity, is pressure independent.
[ ]
T ( Co ) ( )
∑4
PV T ak (T k − Tok ) EXAMPLE 2.2 Thermodynamic Properties of an
dT = a0 ln + R
∫To T To k Ideal Mixture.
k=1
(2-40)
Styrene is manufactured by catalytic dehydrogenation of ethyl ben-
The liquid molar volume, vL , and mass density, ρL , are zene, followed by vacuum distillation to separate styrene from unre-
computed from the pure species using (4) in Table 2.3 and acted ethyl benzene [6]. Typical conditions for the feed are 77.9∘ C
assuming additive molar volumes (not densities). The effect of (351 K) and 100 torr (13.33 kPa), with the following flow rates:
temperature on pure-component liquid density from the freez-
ing point to the near-critical region at saturation pressure is
n, kmol/h
correlated well by the Rackett equation [4]:
Component Feed
A
ρL = (2-41) Ethyl benzene (EB) 103.82
B(1−T∕Tc )
2∕7

Styrene (S) 90.15

where values of constants A and B, and the critical temperature,
Tc , are tabulated for approximately 700 organic compounds by
Yaws et al. [5]. Assuming that the ideal-gas law holds and that vapor and liquid
The vapor pressure of a liquid species, Ps , is well repre- phases exist and are ideal solutions, use a process simulator to
sented over temperatures from below the normal boiling point determine the feed-stream phase conditions and the thermody-
to the critical region by an extended Antoine equation: namic properties listed in Table 2.2. Also, compute the relative
volatility, αEB,S .
ln Ps = k1 + k2 ∕(k3 + T) + k4 T + k5 ln T + k6 T k7 (2-42)
where constants k for hundreds of compounds are built into the Solution:
physical-property libraries of all process simulation programs. The Aspen Plus Simulator with the Ideal Properties option gives the
At low pressures, the molar enthalpy of vaporization is given following results where the datum is the elements (not the compo-
in terms of vapor pressure by classical thermodynamics: nents) at 25∘ C and 1 atm.
( )
d ln Ps
ΔH vap = RT 2 (2-43)
dT Property Vapor Liquid
If (2-42) is used for the vapor pressure, (2-43) becomes EB Flow rate, kmol/h 57.74 46.08
( ) S Flow rate, kmol/h 42.91 47.24
k2 k5 k7 −1
ΔH = RT − (
vap 2
)2 + k4 + + k7 k6 T Total Flow rate, kmol/h 100.65 93.32
k3 + T T Temperature, ∘ C 77.9 77.9
(2-44) Pressure, Bar 0.1333 0.1333
The molar enthalpy, hL , of an ideal-liquid mixture is Molar Enthalpy, kJ/kmol 87,200 56,213
obtained by subtracting the molar enthalpy of vaporization Molar Entropy, kJ/kmol-K –244.4 –350.0
from the ideal molar vapor enthalpy for each species, as given Molar Volume, m3 /kmol 219.0 0.126
by (2-39), and summing, as shown in (5) in Table 2.3. The Average MW 105.31 105.15
molar entropy, sL , of the ideal-liquid mixture, given by (6), is Vapor Pressure, Bar 0.1546 0.1124
K-Value 1.16 for EB 0.843 for S
obtained in a similar manner from the ideal-gas entropy by
Relative Volatility 1.376
subtracting the molar entropy of vaporization, ΔH vap ∕T.
The final equation in Table 2.3 gives the expression for
the ideal K-value, previously included in Table 2.2. It is the
K-value based on Raoult’s law, using
§2.3 GRAPHICAL REPRESENTATION
pi = xi Psi (2-45)
OF THERMODYNAMIC PROPERTIES
and Dalton’s law:
Plots of thermodynamic properties are useful not only for
pi = yi P (2-46)
the data they contain, but also for the graphical representa-
Combination of (2-45) and (2-46) gives the Raoult’s law tion, which permits the user to: (1) make general observations
K-value: about the effects of temperature, pressure, and composition;
y Ps (2) establish correlations and make comparisons with experi-
Ki ≡ i = i (2-47)
xi P mental data; and (3) make extrapolations. All process simula-
tors that calculate thermodynamic properties also allow the
where the extended Antoine equation, (2-42), is used to esti- user to make property plots. Handbooks and thermodynamic
mate vapor pressure. The ideal K-value is independent of com- textbooks contain generalized plots of thermodynamic proper-
position, but exponentially dependent on temperature because ties as a function of temperature and/or pressure. A typical plot
of the vapor pressure, and inversely proportional to pressure. is Figure 2.1, which shows vapor pressure curves of common
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 22

10,000
6,000
4,000

2,000 er
Critical points e at
id W
1,000 iox
rd
lfu
600 Su
400

200 lene de
E t hy ioxi ane
on d Eth
Vapor Pressure, psi

100 C a r b
int
60 po l
p le no
40 Tri ) ha
dC
O2 et

A
li ide M ne
lor

m
(So l ch ta

er
20 e y p
an eth He

th
rop M n-

w
P
e

Do
10
lo rid
6 ch
yl
4 Eth
ne

ry
cu
ze
2 e e
ne tan on
en

er
e
uta tan en et

M
ob

i-B u i-P Ac
1 n-B
itr

ol id ne
N

0.6 ne an Ac ze
e ic e
e

nz h et n
Et en be
an

e
0.4
tan
e Be Ac yl
ec

en an e X pyl
ex
D

n e -
n-P an o
n-

e o
0.2 n-H lu ct Pr
To O n-
0.1
–100 –75 –50 –25 0 25 50 75 100 150 200 250 300 350 400 500 600 700 800
Temperature, °F

Figure 2.1 Vapor pressure of some common industrial chemicals.

[Reproduced with permission from A.S. Faust, L.A. Wenzel, C.W. Clump, L. Maus, and L.B. Andersen, Principles of Unit Operations, John Wiley & Sons, New York (1960).]

22
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 23

§2.5 P-v-T Equation-of-State (EOS) Models 23

chemicals for temperatures from below their normal boiling reference state and (2) component reference state. For
point to their critical temperature where the vapor pressure both states, the reference temperature is 25∘ C (298.15 K).
curves terminate. These curves fit the extended Antoine equa- The default option is the elemental reference state, in which
tion (2-42) reasonably well and are useful in establishing the the enthalpy of a pure component is referenced to its stan-
phase of a pure species. If the pressure is above the curve, the dard elements by its standard enthalpy of formation. The
phase is liquid, if below it is vapor. Vapor pressures are also component reference state is the pure component as an ideal
used for estimating Raoult’s law K-values prior to using a pro- gas. The advantage of the default option is that an enthalpy
cess simulator. balance around a chemical reactor automatically accounts for
Before the advent of digital computers and process simula- the heat of reaction because components are referred to the
tors, various types of graphs were used for determining effects standard elements. The default option must be used for elec-
of temperature and pressure on vapor-liquid K-values for trolyte systems. As an example, the enthalpy of superheated
hydrocarbons and light gases. These graphs, often in nomo- steam at 300∘ C and 1 MPa, using the elemental reference
graph form, were correlations of experimental phase equilibria state, is −12,920 kJ∕kg. If the component reference state
data. They were not applicable when mixtures contained suffi- is used, the enthalpy becomes 508 kJ∕kg. The difference is
ciently different hydrocarbons such that liquid-phase activity −13,428 kJ∕kg or −241,900 kJ∕kmol, which is the value for
coefficients were not close to 1.0 and were strongly dependent the standard enthalpy of formation at 25∘ C for water vapor
on composition. With the advent of computers, empirical from the standard elements by the reaction, H2(g) + 1/2O2(g) →
models that account for effects of composition in addition to H2 O(g) .
temperature and pressure, and can be applied to mixtures of In the ChemSep process simulator, the reference tempera-
organic chemicals as well as hydrocarbons, displaced graph- ture is also 25∘ C, but the reference phase state can be vapor
ical correlations. These empirical models are available in all or liquid. The standard enthalpy of formation can be included
process simulators. Those most widely used are discussed in or excluded to achieve either the elemental or the component
subsequent sections of this chapter. Some of the models apply reference state for enthalpy.
to liquid–liquid equilibrium. In the CHEMCAD process simulator, the reference tem-
perature is also 25∘ C, but only the elemental reference state
is used.
§2.4 NONIDEAL THERMODYNAMIC
PROPERTY MODELS
Of importance to all separation processes are nonideal thermo- §2.5 P-v-T EQUATION-OF-STATE (EOS)
dynamic properties of mixtures. Models have been formulated MODELS
to estimate thermodynamic properties of nonideal, nonelec-
trolyte, electrolyte, and polymer mixtures. Many of these A relationship between molar volume, temperature, and pres-
models, together with the constants and parameters needed sure is a P-v-T equation of state. Numerous such equations
to apply them to the design and simulation of separation have been proposed. The simplest is the ideal-gas law, which
operations, are available in process simulators. In this chapter, applies only at low pressures or high temperatures because it
emphasis is on the most widely used models of three types: neglects intermolecular forces and the volume occupied by the
(1) P–v–T equation-of-state (EOS) models; (2) Gibbs excess molecules. All other equations of state attempt to correct for
free-energy (gE ) models from which liquid-phase activity these two deficiencies. The most widely used equations of state
coefficients can be calculated; and (3) predictive thermody- are listed in Table 2.4.
namic models. Their applicability depends on the nature of Not included in Table 2.4 is the van der Waals equation,
the components in the mixture, the degree of nonideality, the P = RT∕(v − b) − a∕v2 , where a and b are species-dependent
pressure and temperature, and the reliability of the equation constants. This equation was the first successful formulation
constants and parameters. of an equation of state for a nonideal gas. It is now rarely
used because of its narrow range of application. However, its
development suggested that all species have approximately
§2.4.1 Reference State (Datum) for Enthalpy equal reduced molar volumes, vr = v∕vc at the same reduced
It is important to note that enthalpy is not an absolute thermo- temperature, Tr = T∕Tc , and reduced pressure, Pr = P∕Pc ,
dynamic property. It is determined relative to a reference state where the subscript c refers to the critical point. This finding,
(datum) and no standard has been accepted. Instead, many referred to as the law of corresponding states, was utilized to
different reference states are used in practice. It is important develop the generalized equation-of-state, (2) in Table 2.4,
to be aware of the enthalpy reference states used in process which defines the compressibility factor, Z = Pv∕RT, where
simulators. Z is a function of Pr , Tr , and either the critical compress-
In the Aspen Plus process simulator, two options, discussed ibility factor, Zc , or the acentric factor, ω. The latter was
by Felder and Rousseau [47], are available: (1) elemental introduced by Pitzer et al. [7] to account for differences in
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 24

24 Chapter 2 Thermodynamics of Separation Operations

Table 2.4 Useful Equations of State

Name Equation Equation Constants and Functions
RT
(1) Ideal–gas law P= None
v
ZRT
(2) Generalized P= Z = Z{Pr , Tr , Zc or ω} as derived from data
v
RT a
(3) Redlich-Kwong (RK) P= − √ b = 0.08664RTc ∕Pc
v − b (v2 + bv) T
a = 0.42748R2 Tc2.5 ∕Pc
RT a
(4) Soave-Redlich-Kwong (SRK) P= − b = 0.08664RTc ∕Pc
v − b v2 + bv
[ ( )]2
a = 0.42748R2 Tc2 1 + fω 1 − Tr0.5 ∕Pc
fω = 0.48 + 1.574ω − 0.176ω2
RT a
(5) Peng–Robinson (PR) P= − b = 0.07780RTc ∕Pc
v − b v2 + 2bv − b2
[ ( )]2
a = 0.45724R2 Tc2 1 + fω 1 − Tr0.5 ∕Pc
fω = 0.37464 + 1.54226ω − 0.26992ω2

molecular shape and is determined from the vapor pressure §2.5.1 The Redlich–Kwong (RK) Model
curve by: [ ]
( s) In 1949, Redlich and Kwong [14] published the RK equation
P
ω = − log − 1.0 (2-48) of state which, like the van der Waals equation, contains only
Pc Tr =0.7 two constants, a and b, both of which can be determined from
Tc and Pc , by applying conditions at the critical point:
The value for ω is zero for symmetric molecules. Some typical
( ) ( 2 )
values of ω are 0.264, 0.490, and 0.649 for toluene, n-decane, ∂P ∂ P
and ethyl alcohol, respectively, as taken from the extensive tab- = 0 and =0
∂v Tc ∂v2 Tc
ulation of Poling et al. [3].
A common empirical P-v-T equation is the virial equation The RK model, given as (3) in Table 2.4 together with the
of state due to Thiesen [8] and Onnes [9]. It is a power series two parameters, is an improvement over the van der Waals
for compressibility factor, Z, in terms of 1∕v: equation. When applied to nonpolar compounds, its accuracy
is comparable to other EOS models containing many more
B C
Z =1+ + +··· constants. Furthermore, the RK equation can approximate the
v v2
liquid-phase region.
A modification of the virial equation is the Starling form [10] A cubic equation in v results when the RK equation is
of the Benedict–Webb–Rubin (BWR) equation for hydro- expanded to obtain a common denominator. Alternatively, (2)
carbons and light gases. Walas [11] presents a discussion of and (3) in Table 2.4 can be combined to eliminate v to give
BWR-type equations, which—because of the large number of the more useful compressibility factor, Z, form of the RK
terms and species constants (at least 8)—is not widely used equation:
except for pure substances at cryogenic temperatures. A more
useful modification of the BWR equation is a generalized Z 3 − Z 2 + (A − B − B2 )Z − AB = 0 (2-49)
corresponding-states form developed by Lee and Kesler [12]
where
with an extension to mixtures by Plöcker et al. [13]. All
aP
of the constants in the LKP equation are given in terms of A= (2-50)
R2 T 2
the acentric factor and reduced temperature and pressure, as
developed from P-v-T data for three simple fluids (ω = 0),
bP
methane, argon, and krypton, and a reference fluid n-octane B= (2-51)
RT
(ω = 0.398). The equations, constants, and mixing rules are
given by Walas [11]. The LKP equation describes vapor and Equation (2-49), a cubic in Z, can be solved for the three roots
liquid mixtures of hydrocarbons and/or light gases over wide using MATLAB with the Roots function. At supercritical
ranges of T and P. temperatures, where only one phase exists, one real root and a
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 25

§2.5 P-v-T Equation-of-State (EOS) Models 25

complex conjugate pair of roots are obtained. Below the crit- 100
ical temperature, where vapor and/or liquid phases can exist, Experimental data
three real roots are obtained, with the largest value of Z apply- of Yarborough
ing to the vapor and the smallest root corresponding to the N2 SRK correlation
liquid (ZV and ZL ). The intermediate value of Z is discarded. Temperature 250°F
To apply the RK model to mixtures, mixing rules are used
to average the constants a and b for each component. The rec- C1
CO2
ommended rules for vapor mixtures of C components are 10
[C ]
∑C
∑ ( )0.5 C2
a= yi yj ai aj (2-52) H 2S
i=1 j=1

∑
C
b= yi bi (2-53) C3

K-value
i=1
1.0

§2.5.2 The Soave–Redlich–Kwong (SRK)

nC5
Model
Following
√ the work of Wilson [15], Soave [16] replaced
the T term in the RK EOS with a third parameter, the
nC7
acentric factor, ω, to better represent nonspherical molecules. .1
The resulting Soave–Redlich–Kwong (SRK) or Redlich– Toluene
Kwong–Soave (RKS) equation, given as (4) in Table 2.4, was
quickly accepted for application to mixtures of hydrocarbons
and light gases because of its simplicity and accuracy. It makes
nC10
the constant a a function of ω and T, as shown in Table 2.4,
achieving a good fit to vapor pressure data, thereby improving
.01
the prediction of liquid-phase properties, and vapor-liquid 100 1000 10,000
K-values as discussed below. This is clearly shown in Pressure, psia
Figure 2.2 where experimental equilibrium data at 250∘ F
Figure 2.2 Comparison of experimental K-value data and the SRK
for a mixture of 10 components ranging in volatility from N2
equation.
to n-decane is compared to predictions by the SRK EOS over
a pressure range of about 250 to 2,500 psi. Note that at a pres-
sure somewhat above 2,500 psi, K-values of all 10 components
SRK equations to achieve improved agreement in the critical
appear to be approaching a value of 1.0. That point is called
region and for liquid molar volume. The Peng–Robinson (PR)
the convergence pressure for the mixture. It is analogous to
EOS is (5) in Table 2.4. The mixing rules for the PR model are
the critical pressure for a pure chemical.
(2-52), (2-53), and (2-54). Both the SRK and PR equations
The mixing rules for the SRK model are (2-52) and (2-53)
are used in process simulators for calculations of vapor–liquid
except that (2-52) is modified when light gases are present
phase equilibrium compositions for mixtures of hydrocarbons
in the mixture to include a binary interaction coefficient, kij ,
and light gases.
giving [C ]
∑C
∑ ( )0.5 ( )
a= yi yj ai aj 1 − kij (2-54) §2.5.4 Derived Thermodynamic Properties
i=1 j=1
from EOS Models
Equation (2-54) is particularly useful for gas mixtures con-
taining N2 , CO, CO2 , and H2 S. Values of kij back-calculated If a temperature-dependent, ideal-gas heat capacity or en-
from experimental data are used in process simulators. Gener- thalpy equation such as (2-38) or (2-39) is available, along
ally, kij is zero for hydrocarbons paired with hydrogen or other with an EOS, other vapor- and liquid-phase properties can
hydrocarbons. be derived using the integral equations listed in Table 2.5.
These equations, in the form of departures from the ideal-gas
equations of Table 2.3, apply to both vapor and liquid.
§2.5.3 The Peng–Robinson (PR) Model
When the ideal-gas law, P = RT∕v, is substituted into
Four years after the introduction of the SRK equation, Peng Equations (1) to (4) of Table 2.4, the results for the vapor are
and Robinson [17] presented a modification of the RK and (h − hV∘ ) = 0, ϕ = 1 and (s − s∘V ) = 0, ϕ = 1.
 Trim Size: 8.5in x 11in Seader c02.tex V2 - 10/16/2015 10:39 A.M. Page 26

26 Chapter 2 Thermodynamics of Separation Operations

When the RK equation is substituted into the equations of Table 2.5 Integral Departure Equations of Thermodynamics
Table 2.5, the results for the vapor phase are: At a given temperature and composition, the following equations
[ ( )] give the effect of pressure above that for an ideal gas.
∑ C
3A B
hV = (yi hoiV ) + RT ZV − 1 − ln 1 + (2-55)
i=1
2B ZV Mixture enthalpy:

( ) v [ ( )]
∑C ∂P
P (1) (h − hoV ) = Pv − RT − P−T dv
sV = (yi soiV ) − R ln o ∫∞ ∂T v
i=1
P
Mixture entropy:
∑
C
−R (yi ln yi ) + R ln(ZV − B) (2-56) v( ) v
∂P R
i=1 (2) (s − soV ) = dv − dv
∫∞ ∂T v ∫∞ v
[ ( )]
( ) A B
ϕV = exp ZV − 1 − ln ZV − B − ln 1 + Pure-component fugacity coefficient:
B ZV
[ P( ) ]
(2-57) 1 RT
(3) ϕiV = exp v− dP
[ RT ∫0 P
( )
̄ϕiV = exp ZV − 1 Bi − ln (ZV − B) [ ]
∞( )
( √ B ) RT
)]
1
( = exp P− dv − ln ZV + (ZV − 1)
A Ai Bi B RT ∫v v
− 2 − ln 1 + (2-58)
B A B ZV
Partial fugacity coefficient:
{ [ ] }
Similar results are obtained with the SRK and PR models. ∞ ( )
̄ 1 ∂P RT
The results for the liquid phase are identical if yi and ZV (4) ϕiV = exp − dV − ln ZV ,
(but not hoiV ) are replaced by xi and ZL , respectively. The RT ∫V ∂Ni T,V,Nj V
liquid-phase forms of (2-55) and (2-56) account for the
enthalpy and entropy of vaporization. This is because the RK ∑
C
where V = v Ni
equation, as well as the SRK and PR equations, are continuous i=1
functions through the vapor and liquid regions, as shown for
enthalpy in Figure 2.3. Thus, the liquid enthalpy, at tempera-
tures below the critical point, is determined by accounting for
four effects. From (1), Table 2.5, and Figure 2.3:
v[ ( )]
∂P P=0
hL = hoV + Pv − RT − P−T dv
∫∞ ∂T v hoV
1

2
This equation is then divided into four parts as shown in
Molar enthalpy, H

r
Vapo
Figure 2.3: P=P
0
1. Vapor at zero pressure = hoV 3

2. Pressure correction for vapor to saturation pressure =

vVsat [ ( )] hL
∂P 4
+(Pv)Vsat − RT − P−T dv
∫∞ ∂T v id
Liqu
3. Latent heat of vaporization =
( ) ( )
∂P T
−T vVsat − vLsat Temperature, T
∂T sat
Figure 2.3 Contributions to enthalpy.
4. Correction to liquid for pressure in excess of saturation
pressure =
[ ] vL [ ( )] The fugacity coefficient, ϕ, of a pure species from the RK
∂P
+ (Pv)L − (Pv)Lsat − P−T dv (2-59) equation, as given by (2-57), describes the vapor for P < Psi .
∫vL ∂T v
sat If P > Psi , ϕ is the liquid fugacity coefficient. Saturation
where the subscript “sat” refers to the saturation pressure. pressure corresponds to the condition of ϕV = ϕL . Thus, at
Discovering Diverse Content Through
Random Scribd Documents
The Project Gutenberg eBook of Ruy Blas:
Drame
 This ebook is for the use of anyone anywhere in
the United States and most other parts of the
world at no cost and with almost no restrictions
whatsoever. You may copy it, give it away or re-
use it under the terms of the Project Gutenberg
License included with this ebook or online at
www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of
the country where you are located before using
this eBook.

Title: Ruy Blas: Drame

Author: Victor Hugo

Release date: September 25, 2019 [eBook

#60354]
Most recently updated: October 17,
2024

Language: French

Credits: Produced by Ramon Pajares Box and the

Online Distributed
Proofreading Team at
http://www.pgdp.net (This file was
produced from images generously made
available by The
Internet Archive/Canadian Libraries)

*** START OF THE PROJECT GUTENBERG EBOOK

RUY BLAS: DRAME ***
Note de transcription
Table des Matières
 ŒUVRES
DE

VICTOR HUGO.

XXIV.
 REPRÉSENTÉ POUR LA PREMIÈRE FOIS
LE 8 NOVEMBRE 1838,
POUR L’OUVERTURE DU THÉÂTRE DE LA RENAISSANCE.

IMP. DE HAUMAN ET Ce.—DELTOMBE, GÉRANT.

Rue du Nord, no 8.
R U Y B LAS,
DRAME

Par Victor Hugo.

BRUXELLES.
SOCIÉTÉ BELGE DE LIBRAIRIE.
HAUMAN ET COMPe.
—
1839
 PRÉFACE.
Trois espèces de spectateurs composent ce qu’on
est convenu d’appeler le public: premièrement, les
femmes; deuxièmement, les penseurs;
troisièmement, la foule proprement dite. Ce que la
foule demande presque exclusivement à l’œuvre
dramatique, c’est de l’action; ce que les femmes y
veulent avant tout, c’est de la passion; ce qu’y
cherchent plus spécialement les penseurs, ce sont des
caractères. Si l’on étudie attentivement ces trois
classes de spectateurs, voici ce qu’on remarque: la
foule est tellement amoureuse de l’action qu’au
besoin elle fait bon marché des caractères et des
passions[1]. Les femmes, que l’action intéresse
d’ailleurs, sont si absorbées par les développements
de la passion, qu’elles se préoccupent peu du dessin
des caractères; quant aux penseurs, ils ont un tel
goût de voir des caractères, c’est-à-dire, des hommes
vivre sur la scène, que, tout en accueillant volontiers
la passion comme incident naturel dans l’œuvre
dramatique, ils en viennent presque à y être
importunés par l’action. Cela tient à ce que la foule
demande surtout au théâtre des sensations; la
femme, des émotions; le penseur, des méditations:
tous veulent un plaisir, mais ceux-ci, le plaisir des
yeux; celles-là, le plaisir du cœur; les derniers, le
plaisir de l’esprit. De là, sur notre scène, trois espèces
d’œuvres bien distinctes, l’une vulgaire et inférieure,
les deux autres illustres et supérieures, mais qui,
toutes les trois, satisfont un besoin: le mélodrame
pour la foule; pour les femmes, la tragédie qui
analyse la passion; pour les penseurs, la comédie qui
peint l’humanité.

[1] C’est-à-dire du style; car, si l’action peut, dans

beaucoup de cas, s’exprimer par l’action même, les
passions et les caractères, à très-peu d’exceptions près,
ne s’expriment que par la parole. Or, la parole au
théâtre, la parole fixée et non flottante, c’est le style.
Que le personnage parle comme il doit parler, sibi
constet, dit Horace. Tout est là.

Disons-le en passant, nous ne prétendons rien

établir ici de rigoureux, et nous prions le lecteur
d’introduire de lui-même dans notre pensée les
restrictions qu’elle peut contenir. Les généralités
admettent toujours les exceptions; nous savons fort
bien que la foule est une grande chose dans laquelle
on trouve tout, l’instinct du beau comme le goût du
médiocre; l’amour de l’idéal comme l’appétit du
commun; nous savons également que tout penseur
complet doit être femme par les côtés délicats du
cœur; et nous n’ignorons pas que, grâce à cette loi
mystérieuse qui lie les sexes l’un à l’autre aussi bien
par l’esprit que par le corps, bien souvent dans une
femme il y a un penseur. Ceci posé, et après avoir
prié de nouveau le lecteur de ne pas attacher un sens
trop absolu aux quelques mots qui nous restent à
dire, nous reprenons.
Pour tout homme qui fixe un regard sérieux sur
trois sortes de spectateurs dont nous venons de
parler, il est évident qu’elles ont toutes les trois
raison. Les femmes ont raison de vouloir être émues,
les penseurs ont raison de vouloir être enseignés, la
foule n’a pas tort de vouloir être amusée. De cette
évidence se déduit la loi du drame. En effet, au delà
de cette barrière de feu qu’on appelle la rampe du
théâtre et qui sépare le monde réel du monde idéal,
créer et faire vivre, dans les conditions combinées de
l’art et de la nature, des caractères, c’est-à-dire, et
nous le répétons, des hommes; dans ces hommes,
dans ces caractères, jeter des passions qui
développent ceux-ci et modifient ceux-là; et enfin du
choc de ces caractères et de ces passions avec les
grandes lois providentielles, faire sortir la vie
humaine, c’est-à-dire des événements grands, petits,
douloureux, comiques, terribles, qui contiennent pour
le cœur ce plaisir qu’on appelle l’intérêt, et pour
l’esprit cette leçon qu’on appelle la morale: tel est le
but du drame. On le voit; le drame tient de la tragédie
par la peinture des passions, et de la comédie par la
peinture des caractères. Le drame est la troisième
grande forme de l’art, comprenant, enserrant et
fécondant les deux premières. Corneille et Molière
existeraient indépendamment l’un de l’autre, si
Shakspeare n’était entre eux, donnant à Corneille la
main gauche, à Molière la main droite. De cette façon
les deux électricités opposées de la comédie et de la
tragédie se rencontrent, et l’étincelle qui en jaillit,
c’est le drame.
En expliquant, comme il les entend et comme il les
a déjà indiqués plusieurs fois, le principe, la loi et le
but du drame, l’auteur est loin de se dissimuler
l’exiguité de ses forces et la brièveté de son esprit. Il
définit ici, qu’on ne s’y méprenne pas, non ce qu’il a
fait, mais ce qu’il a voulu faire. Il montre ce qui a été
pour lui le point de départ. Rien de plus.
Nous n’avons en tête de ce livre que peu de lignes
à écrire et l’espace nous manque pour les
développements nécessaires. Qu’on nous permette
donc de passer, sans nous appesantir autrement sur
la transition, des idées générales que nous venons de
poser et qui, selon nous (toutes les conditions de
l’idéal étant maintenues du reste) régissent l’art tout
entier, à quelques-unes des idées particulières que ce
drame, Ruy Blas, peut soulever dans les esprits
attentifs.
Et premièrement, pour ne prendre qu’un des côtés
de la question, au point de vue de la philosophie de
l’histoire, quel est le sens de ce drame?—Expliquons-
nous.
Au moment où une monarchie va s’écrouler,
plusieurs phénomènes peuvent être observés. Et
d’abord la noblesse tend à se dissoudre. En se
dissolvant elle se divise, et voici de quelle façon:
Le royaume chancelle, la dynastie s’éteint, la loi
tombe en ruine; l’unité politique s’émiette aux
tiraillements de l’intrigue; le haut de la société
s’abâtardit et dégénère; un mortel affaiblissement se
fait sentir à tous au dehors comme au dedans; les
grandes choses de l’État sont tombées, les petites
seules sont debout: triste spectacle public; plus de
police, plus d’armée, plus de finances; chacun devine
que la fin arrive. De là, dans tous les esprits, ennui de
la veille, crainte du lendemain, défiance de tout
homme, découragement de toute chose, dégoût
profond. Comme la maladie de l’État est dans la tête,
la noblesse, qui y touche, en est la première atteinte.
Que devient-elle alors? Une partie des
gentilshommes, la moins honnête et la moins
généreuse, reste à la cour. Tout va être englouti, le
temps presse, il faut se hâter, il faut s’enrichir,
s’agrandir et profiter des circonstances. On ne songe
plus qu’à soi. Chacun se fait, sans pitié pour le pays,
une petite fortune particulière dans un coin de la
grande infortune publique. On est courtisan, on est
ministre, on se dépêche d’être heureux et puissant.
On a de l’esprit, on se déprave et l’on réussit. Les
ordres de l’État, les dignités, les places, l’argent, on
prend tout, on veut tout, on pille tout. On ne vit plus
que par l’ambition et la cupidité. On cache les
désordres secrets que peut engendrer l’infirmité
humaine sous beaucoup de gravité extérieure. Et
comme cette vie acharnée aux vanités et aux
jouissances de l’orgueil a pour première condition
l’oubli de tous les sentiments naturels, on y devient
féroce. Quand le jour de la disgrâce arrive, quelque
chose de monstrueux se développe dans le courtisan
tombé, et l’homme se change en démon.
L’état désespéré du royaume pousse l’autre moitié
de la noblesse, la meilleure et la mieux née, dans une
autre voie. Elle s’en va chez elle. Elle rentre dans ses
palais, dans ses châteaux, dans ses seigneuries. Elle a
horreur des affaires, elle n’y peut rien, la fin du
monde approche; qu’y faire et à quoi bon se désoler?
Il faut s’étourdir, fermer les yeux, vivre, boire, aimer,
jouir. Qui sait! a-t-on même un an devant soi? Cela
dit, ou même simplement senti, le gentilhomme prend
la chose au vif, décuple sa livrée, achète des chevaux,
enrichit des femmes, ordonne des fêtes, paye des
orgies, jette, donne, vend, achète, hypothèque,
compromet, dévore, se livre aux usuriers et met le feu
aux quatre coins de son bien. Un beau matin, il lui
arrive un malheur. C’est que, quoique la monarchie
aille grand train, il s’est ruiné avant elle. Tout est fini,
tout est brûlé. De toute cette belle vie flamboyante, il
ne reste pas même de la fumée; elle s’est envolée. De
la cendre, rien de plus. Oublié et abandonné de tous,
excepté de ses créanciers, le pauvre gentilhomme
devient alors ce qu’il peut, un peu aventurier, un peu
spadassin, un peu bohémien. Il s’enfonce et disparaît
dans la foule, grande masse terne et noire que,
jusqu’à ce jour, il a à peine entrevue de loin sous ses
pieds. Il s’y plonge, il s’y réfugie. Il n’a plus d’or, mais
il lui reste le soleil, cette richesse de ceux qui n’ont
rien. Il a d’abord habité le haut de la société, voici
maintenant qu’il vient se loger dans le bas, et qu’il
s’en accommode; il se moque de son parent
l’ambitieux, qui est riche et qui est puissant; il devient
philosophe, et il compare les voleurs aux courtisans.
Du reste, bonne, brave, loyale et intelligente nature;
mélange du poëte, du gueux et du prince; riant de
tout; faisant aujourd’hui rosser le guet par ses
camarades comme autrefois par ses gens, mais n’y
touchant pas; alliant dans sa manière, avec quelque
grâce, l’impudence du marquis à l’effronterie du
zingaro; souillé au dehors, sain au dedans; et n’ayant
plus du gentilhomme que son honneur qu’il garde,
son nom qu’il cache et son épée qu’il montre.
Si le double tableau que nous venons de tracer
s’offre dans l’histoire de toutes les monarchies à un
moment donné, il se présente particulièrement en
Espagne d’une façon frappante à la fin du dix-
septième siècle. Ainsi, si l’auteur avait réussi à
exécuter cette partie de sa pensée, ce qu’il est loin de
supposer, dans le drame qu’on va lire, la première
moitié de la noblesse espagnole à cette époque se
résumerait en don Salluste, et la seconde moitié en
don César. Tous deux cousins, comme il convient.
Ici, comme partout, en esquissant ce croquis de la
noblesse castillane vers 1695, nous réservons, bien
entendu, les rares et vénérables exceptions.—
Poursuivons.
En examinant toujours cette monarchie et cette
époque, au-dessous de la noblesse ainsi partagée, et
qui pourrait, jusqu’à un certain point, être
personnifiée dans les deux hommes que nous venons
de nommer, on voit remuer dans l’ombre quelque
chose de grand, de sombre et d’inconnu. C’est le
peuple: le peuple, qui a l’avenir et qui n’a pas le
présent; le peuple, orphelin, pauvre, intelligent et
fort; placé très-bas, et aspirant très-haut; ayant sur le
dos les marques de la servitude et dans le cœur les
préméditations du génie; le peuple, valet des grands
seigneurs, et amoureux, dans sa misère et dans son
abjection, de la seule figure qui, au milieu de cette
société écroulée, représente pour lui, dans un divin
rayonnement, l’autorité, la charité et la fécondité. Le
peuple, ce serait Ruy Blas.
Maintenant, au-dessus de ces trois hommes, qui,
ainsi considérés, feraient vivre et marcher aux yeux
du spectateur, trois faits, et dans ces trois faits toute
la monarchie espagnole au dix-septième siècle; au-
dessus de ces trois hommes, disons-nous, il y a une
pure et lumineuse créature, une femme, une reine.
Malheureuse comme femme, car elle est comme si
elle n’avait pas de mari; malheureuse comme reine,
car elle est comme si elle n’avait pas de roi; penchée
vers ceux qui sont au-dessous d’elle par pitié royale et
par instinct de femme aussi peut-être, et regardant en
bas pendant que Ruy Blas, le peuple, regarde en
haut.
Aux yeux de l’auteur, et sans préjudice de ce que
les personnages accessoires peuvent apporter à la
vérité de l’ensemble, ces quatre têtes ainsi groupées
résumeraient les principales saillies qu’offrait au
regard du philosophe historien la monarchie
espagnole il y a cent quarante ans. A ces quatre
têtes, il semble qu’on pourrait en ajouter une
cinquième, celle du roi Charles II. Mais, dans l’histoire
comme dans le drame, Charles II d’Espagne n’est pas
une figure, c’est une ombre.
A présent, hâtons-nous de le dire, ce qu’on vient
de lire n’est point l’explication de Ruy Blas. C’en est
simplement un des aspects. C’est l’impression
particulière que pourrait laisser ce drame, s’il valait la
peine d’être étudié, à l’esprit grave et consciencieux
qui l’examinerait, par exemple, du point de vue de la
philosophie de l’histoire.
Mais, si peu qu’il soit, ce drame, comme toutes les
choses de ce monde, a beaucoup d’autres aspects, et
peut être envisagé de beaucoup d’autres manières.
On peut prendre plusieurs vues d’une idée comme
d’une montagne. Cela dépend du lieu où l’on se place.
Qu’on nous passe, seulement pour rendre claire notre
idée, une comparaison infiniment trop ambitieuse: le
Mont-Blanc, vu de la Croix-de-Fléchères, ne ressemble
pas au Mont-Blanc vu de Sallenches. Pourtant, c’est le
Mont-Blanc.
 De même, pour tomber d’une très-grande chose à
une très-petite, ce drame, dont nous venons
d’indiquer le sens historique, offrirait une tout autre
figure si on le considérait d’un point de vue beaucoup
plus élevé encore, du point de vue purement humain.
Alors don Salluste serait l’égoïsme absolu, le souci
sans repos; don César, son contraire, serait le
désintéressement et l’insouciance; on verrait dans
Ruy Blas le génie et la passion comprimés par la
société et s’élançant d’autant plus haut que la
compression est plus violente; la reine enfin, ce serait
la vertu minée par l’ennui.
Au point de vue uniquement littéraire, l’aspect de
cette pensée, telle quelle, intitulée: Ruy Blas,
changerait encore. Les trois formes souveraines de
l’art pourraient y paraître personnifiées et résumées.
Don Salluste serait le Drame, don César la Comédie,
Ruy Blas la Tragédie. Le drame noue l’action; la
comédie l’embrouille, la tragédie la tranche.
Tous ces aspects sont justes et vrais, mais aucun
d’eux n’est complet. La vérité absolue n’est que dans
l’ensemble de l’œuvre. Que chacun y trouve ce qu’il y
cherche, et le poëte, qui ne s’en flatte pas du reste,
aura atteint son but. Le sujet philosophique de Ruy
Blas, c’est le peuple aspirant aux régions élevées; le
sujet humain, c’est un homme qui aime une femme;
le sujet dramatique, c’est un laquais qui aime une
reine. La foule qui se presse chaque soir devant cette
œuvre, parce qu’en France jamais l’attention publique
n’a fait défaut aux tentatives de l’esprit, quelles
qu’elles soient d’ailleurs, la foule, disons-nous, ne voit
dans Ruy Blas que ce dernier sujet, le sujet
dramatique, le laquais; et elle a raison.
 Et ce que nous venons de dire de Ruy Blas nous
semble évident de tout autre ouvrage. Les œuvres
vénérables des maîtres ont même cela de
remarquable qu’elles offrent plus de faces à étudier
que les autres. Tartufe fait rire ceux-ci et trembler
ceux-là. Tartufe, c’est le serpent domestique; ou bien
c’est l’hypocrite; ou bien c’est l’hypocrisie. C’est tantôt
un homme, tantôt une idée. Othello, pour les uns,
c’est un noir qui aime une blanche; pour les autres,
c’est un parvenu qui a épousé une patricienne; pour
ceux là, c’est un jaloux; pour ceux-ci, c’est la jalousie.
Et cette diversité d’aspects n’ôte rien à l’unité
fondamentale de la composition. Nous l’avons déjà dit
ailleurs: mille rameaux et un tronc unique.
Si l’auteur de ce livre a particulièrement insisté sur
la signification historique de Ruy Blas, c’est que dans
sa pensée, par le sens historique, et, il est vrai, par le
sens historique uniquement, Ruy Blas se rattache à
Hernani. Le grand fait de la noblesse se montre, dans
Hernani comme dans Ruy Blas, à côté du grand fait
de la royauté. Seulement, dans Hernani, comme la
royauté absolue n’est pas faite, la noblesse lutte
encore contre le roi, ici avec l’orgueil, là avec l’épée; à
demi féodale, à demi rebelle. En 1519, le seigneur vit
loin de la cour dans la montagne, en bandit comme
Hernani, ou en patriarche comme Ruy Gomez. Deux
cents ans plus tard, la question est retournée. Les
vassaux sont devenus des courtisans. Et, si le
seigneur sent encore d’aventure le besoin de cacher
son nom, ce n’est pas pour échapper au roi, c’est
pour échapper à ses créanciers. Il ne se fait pas
bandit, il se fait bohémien.—On sent que la royauté
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about books and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebookgate.com