Vous êtes sur la page 1sur 10

Other books by Adrian Bejan:

Entropy Generation Through Heat and Fluid Flow, Wiley, 1982.

Convection Heat Transfer, Wiley, 1984.
Advanced Engineering Thermodynamics, Wiley, 1988.
Convection in Porous Media, with D. A. Nield, Springer-Verlag, 1992.
Heat Transfer, Wiley, 1993.
Convection Heat Transfer, Second Edition, Wiley, 1995.
Thermal Design and Optimization, with G. Tsatsaronis and M. Moran,
Wiley, 1996.
Entropy Generation Minimization, CRC Press, 1996.
To my Cristina and Teresa

This text is printed on acid-free paper.

Copyright 1997 by John Wiley & Sons, Inc.

All rights reserved. Published simultaneously in Canada.
Reproduction or translation of any part of this work beyond
that permitted by Section 107 or 108 of the 1976 United
States Copyright Act without the permission of the copyright
owner is unlawful. Requests for permission or further
information should be addressed to the Permissions Department,
John Wiley & Sons, Inc., 605 Third Avenue, New York, NY
This publication is designed to provide accurate and
authoritative information in regard to the subject
matter covered. It is sold with the understanding that
the publisher is not engaged in rendering legal, accounting,
or other professional services. If legal advice or other
expert assistance is required, the services of a competent
professional person should be sought.

Library of Congress Cataloging in Publication Data:

Bejan, Adrian, 1948-
Advanced engineering thermodynamics / Adrian Bejan. - 2nd ed.
p. cm.
Includes bibliographical references and index.
ISBN 0-471-14880-6 (cloth: alk. paper)
1. Thermodynamics. I. Title.
TJ265.B425 1997
621.402'1-{)c21 97-5543

Printed in the United States of Americil

10 9 8 7 6 5 4 3 2
It is a real privilege to be asked to prepare this second edition and to have
another opportunity to cast a bird's-eye view over the field of thermo-
dynamics, in both engineering and physics. For this I am deeply grateful to
the many professors and students who have used the first edition worldwide.
I am also indebted to my friends at John Wiley & Sons, who have "adopted"
me and my work, beginning with my first year as a professor.
In the first edition I urged the student not to regard thermodynamics as
finished, but to invest his or her creativity in the future growth of the field.
That was a call to action-a manifesto, really-to replace present-day
thermodynamics with something better and more useful. I was repeating a
call made in my first book (1982), where I noted that we already possess
deterministic means with which to attack realistic (irreversible) processes and
systems. By sketching Fig. 1, I predicted a merger of thermodynamics with
transport phenomena (e.g., heat transfer), to produce a more powerful
thermodynamics of irreversible devices by the year 2000.
nil rKbt'i\Lh

As I look back at Fig. 1 and the activity published since the first edition
(1988), I think it is time to claim a small victory and to accept a new and
greater challenge. The victory is that the combined method of thermo-
dynamics and heat transfer has sold itself over the wide spectrum of
engineering and physics. Today the method is best known as entropy
generation minimization (EGM), thermodynamic optimization, or finite-time
thermodynamics. This method brings systematically into thermodynamics
both modeling and optimization. The systems and processes that are analyzed
are realistic: Their irreversibilities are due to transport processes, which are
described in terms of practical (concrete) notions such as materials, shapes,
relative positions, and size and time constraints.
The simplest models and the most basic trade-offs (optima) revealed by
EGM have enriched the discipline of thermodynamics. These trade-offs are
fundamental: Since they rule the operation of the simplest model that is still
realistic, they are certainly present in the most complex (industrial R & D)
models, where they deserve to be identified and exploited.
The newer and greater challenge is to extend our deterministic powers to
the class of naturally organized systems, living and not living. Such systems
are all around us and inside ourselves. Their organization is in space and
in time. The networks visible in trees, roots, leaves, lungs, vascularized
tissues, dendrites in rapid solidification, axonal arbors, river basins, deltas,
lightning, streets, and other paths of telecommunication are spatially
organized. Figure 2, for example, makes us see all these phenomena and,
above all, beauty. Temporal organization is evident in the finely tuned
frequencies of respiration, circulation, and pulsating and meandering flows
(e.g., rivers, and many other turbulent flows).
My first steps in this new direction were purely by accident, as I now
recount in section 13.6. I saw this direction as a challenge to me (a
provocation) only after I did the work: It was then that I discovered the
voluminous material that physicists and biologists had published on "self-
organization," a huge and diverse ensemble of macroscopic phenomena that
they consider to be nondeterministic-that is, the result of chance. The
challenge was to construct a theory-a deterministic approach-to predict,
explain, and in this way unify the naturally organized phenomena. There had
to be a reason for all the geometric form and similarity that we see in
In Chapter 13, I show that the tree-shaped networks can in fact be
predicted in an amazingly simple and direct way, by geometrically optimizing
the access between one point and a finite volume (an infinite number of
points). The theoretical network has a definite time direction: It must be
constructed by proceeding from small to large-hence the name constructal
for the related theory. The existence of at least two access routes (flow
regimes) is essential: a slow regime without shape (diffusion, disorganization)
placed at the volumetric level, accompanied by a faster flow regime
with shape (streams, organization) along channels positioned optimally in

&ssembliesof large sizes. Through this geometric construction and the other dynamics was formulated and aimed at irreversible processes and systems and
results assembled in Chapter 13, life, purpose, and time are made a part of at ways of optimizing (i.e., improving) operation. The tools needed for this
our thermodynamics. work have been developed and used by engineers for the past 200 years. They
To attempt a deterministic theory of organization in Nature is to reach have been used with enormous success as separate disciplines, but they are
for things sacred: a better understanding of how we fit in this world and how now coming together (Fig. 1). Our standard of living today is a measure
the world holds together. t Organization and the beauty that it sets free are of this success (e.g., Fig. 8.1).
at the center of every religion. In science and philosophy, the organization Now, if we examine closely the problems solved in Chapter 13, we will
of Nature captivated man's imagination and served as centerpiece in the see that to predict natural organization we did not need thermodynamics.
dispute between randomness and determinism. The subject has experienced To minimize the resistance to heat or fluid flow was possible in the early
a resurgence in physics during the past two decades. It has become 1800s. To minimize the time of travel between a finite area and one point
fashionable to publish volumes of empirical ("Look! See?") material such was a problem for the time of Galilei and even earlier. This delay to roughly
as photographs, computer-generated images, and essays on the observation 150 years after the birth of thermal science (Fourier, Carnot) is due to a
that natural phenomena display geometric similarity. What had been missing coincidence on which I focus next.
were the hard facts: deterministic, predictive, black-on-white methods-that The development of principles of engineering science (including thermal
is, answers to questions such as "Why do flows possess geometric form?" science) began with the establishment of the modern engineering schools
and "Why is complexity increasing in time?" Instead, as Horgant has noted, (Paris, 1795; Prague, 1806; Vienna, 1815; Karlsruhe, 1825). The coincidence
the authors hid their lack of determinism behind metaphors such as "fractal," is that this was also the era in which differential calculus was beginning to
next to which I would list the word "disturbance." spread as the language of science and engineering. Even though Carnot and
This state of affairs is the reason why I placed the photographs of natural the other pioneers were stating their thermodynamics views with reference
phenomena near the end in Chapter 13-they are known to all of us anyway! to macrosystems of arbitrary size and unmentioned internal complexity, the
I placed in front of the chapter the nakedly simple analyses that allow the second generation of thermodynamicists sought to make its own contribution
reader to anticipate the shapes and organization that will spring out of the by using the newly learned language of infinitesimal calculus.
photographs. Constructal theory is accessible at the high-school level. I wrote The infinitesimal and microscopic facets of thermodynamics were almost
it intentionally for the pencil and paper, in a language that Euclid and exclusively the contribution of nonengineers (physicists, chemists, mathe-
Pythagoras would have liked: straight lines, circles,:j:and integers. maticians), at a time when engineers continued on the geometric and
It is time to put Fig. 1 and Fig. 2 together and to close this preface with macroscopic (finite-size system) path. The emphasis on the frontier shifted
a renewed manifesto addressed mainly to engineers. These two figures-the to the differential geometry of surfaces that relate the properties of simple
small victory and the new challenge-belong together because they both refer or nearly simple systems at equilibrium. Equilibrium (classical, Gibbsian, or
to (i) the irreversible operation of macroscopic systems and (ii) the analytical) thermodynamics is one lasting result of this emphasis (Chapters
thermodynamic optimization of the operation of such systems. They both 4-7). The steps made in this century away from equilibrium thermodynamics,
refer to "engineered" systems. in what has become known as Irreversible Thermodynamics or Nonequi-
Here is why engineers are the ones who should be paying attention to the librium Thermodynamics (Chapter 12), were also wedded to the infinitesimal,
thermodynamics of naturally organized systems. The history reviewed in zero-size approach. Unwittingly, these steps were a yearning for a return to
Chapters 1 and 2 shows that the thermodynamics pioneers were engineers, the realistic processes and systems targeted by the pioneers.
military men, doctors, and amateurs. The physicists contributed later. The The ISO-yeardelay to which I referred is the result of a common behavioral
reason is that the defining problem of thermodynamics-the heat engine- trend in science. It takes only one or two truly creative pioneers (e.g., Gibbs)
was a macroscopic system with purpose. From its very beginning, thermo- for an entire crowd to form and mimic these pioneers and to start believing
in its own material. Next, the even larger group that comes to be educated
tJ. Horgan, The twilight of science, Technology Review, July 1996, pp. 50-61.
by the crowd knows nothing-applauds nothing-other than the material
"It is necessary to be careful with the information presented by an experimentalist who lacks regurgitated by Gibbs' epigones. This is why today we read the claim that
theoretical principles ... [he] gathers at random several facts and presents them as proofs ... what we inherited from Carnot and his period is strictly a thermodynamics
scientific knowledge without reasoning [theory] does not exist" (J. Ie R. d'Alembert, Nouvelles of reversible processes. We also read that the engineers' interests and abilities
Experiences sur la Resistance des Fluides, Jambert, Paris, 1777; Personal communication by are limited to reversible phenomena and that in engineering, irreversibility
Profs. J. L. Lage and D. A. Nield, 1997).
'To see the circles beneath a construction such as Fig. 13.2, try to draw two perpendicular lines
is regarded as a "nuisance."
using no more than a ruler and a compass. Yes, most certainly, irreversibility is to be minimized when the construc-

tor's objective is to improve thermodynamic performance. The giant steps

(ideas) illustrated in Figs. 2.1, 8.1, and 10.29, however, did not occur "by
chance" to men who had neither interest in, nor an understanding of,
irreversibility. On the contrary. From Lazare Carnot, through to our own
century (e.g., Stodola, Claude, Keenan), irreversibility minimization has
been the main issue. That issue is even better known as efficiency increase,
performance improvement, or, simply, good engineering.
It is time that we engineers reclaim our own field-thermodynamics-so
that we may expand its deterministic powers in the direction of naturally
organized, living and not living systems. We are the ones to do this work
because Nature is engineered.

Durham, North Carolina
July 1996
I have assembled in this book the notes prepared for my advanced class in
engineering thermodynamics, which is open to students who have had
previous contact with the subject. I decided to present this course in book
form for the same reasons that I organized my own notes for use in the
classroom. Among them is my impression that the teaching of engineering
thermodynamics is dominated by an abundance of good introductory
treatments differing only in writing style and quality of graphics. For
generation after generation, engineering thermodynamics has flowed from
one textbook into the next, essentially unchanged. Today the textbooks
describe a seemingly "classical" engineering discipline, that is, a subject void
of controversy and references, one in which the step-by-step innovations in
substance and teaching method have been long forgotten.
Traveling back in time to rediscover the history of the discipline and
looking into the future for new frontiers and challenges are activities
abandoned by all but a curious few. This situation presents a tremendous
pedagogical opportunity at the graduate level, where the student's determina-
tion to enter the research world comes in conflict with the undergraduate
view that thermodynamics is boring and dead as a research arena. The few
textbooks that qualify for use at the graduate level have done little to alleviate
this conflict. On the theoretical side, the approach preferred by these
textbooks has been to emphasize the abstract reformulation of classical
thermodynamics into a sequence of axioms and corollaries. The pedagogical
drawback of overemphasizing the axiomatic approach in engineering is that
engineers do not live by axioms alone, and that the axiomatic reformulation
seems to change from one revisionist author to the next. Of course, there
is merit in the simplified phrasing and rephrasing of any theory: this is why
a comparative presentation of various axiomatic formulations is a component

t Abbreviated

of the present treatment. However, I see additional merit in proceeding to subject of classical thermodynamics. There are at least two ways in which
show how the theory can guide us through the everexpanding maze of every subject can be advanced by a second course such as this. One is a
contemporary problems. Instead of emphasizing the discussion of equilibrium "horizontal" expansion into the more remote fields intersected by the
states and relations among their properties, I see more value in highlighting subject; the other is a "vertical" expansion, that is, a deepening of our
irreversible processes, especially the kind found in practical engineering understanding of the most basic concepts that define the subject. In the
systems. present treatment, I have followed the second approach because I see it as
With regard to the presentation of engineering thermodynamics at the a more effective means of conveying a bird's-eye view of engineering
graduate level, I note a certain tendency to emphasize physics research thermodynamics. An exhaustive coverage of the horizontal type already
developments and to deemphasize engineering applications. I am sure that exists in the" handbooks"; and justice to each peripheral domain can be done
the engineering student-his t sense of self esteem-has not been well served only in specialized courses such as compressible fluid dynamics, combustion,
by the implication that the important and interesting applications are to be turbomachinery, refrigeration and air conditioning, cryogenics, etc.
found only outside the domain chosen by him for graduate study. If he, like I have followed the vertical approach in order to make a statement of what
Lazare and Sadi Carnot two centuries earlier, sought to improve his I consider effective as a pedagogical tool. Although it has become fashionable
understanding of what limits the "efficiency" of machines, then he finished to associate completeness and volume with "goodness," in this course I have
the course shaking his head wondering about the mechanical engineering made a conscious effort to focus on the structure of the field. I invite the
relevance of, say, negative absolute temperatures. research student to make his own contributions to this structure. For this last
These observations served to define my objective in designing the present reason, the more applied segments of the present treatment are dominated
treatment. My main objective is to demonstrate that engineering thermo- by the topics that have attracted my own interest as a researcher.
dynamics is an active and often controversial field of research, and to To summarize, the combined research and pedagogical mission of this
encourage the student to invest his creativity in the future growth of the effort is to take a second look at the field and to make this view accessible
field. in a one-semester course taken by individuals whose initial understanding
The other considerations that have contributed to defining the objective of the subject is by no means homogeneous. Depth is provided through a
of the present treatment are hinted at by the title Advanced Engineering comparative discussion of the various ways in which the fundamentals have
Thermodynamics. The focus is being placed on "engineering" thermo- been stated over the years, and by reestablishing the connection between
dynamics, that is, on that segment of thermodynamics that addresses the fundamentals and contemporary research trends such as the "exergy"
production of mechanical power and refrigeration in the field of engineering methodology.
practice. I use the word "thermodynamics" in spite of the campaign fought
on behalf of "thermostatics" as the better name for the theory whose subjects * * *
are either in equilibrium or, at least, in local equilibrium (more on this later,
pp. 68-71). I must confess that I feel quite comfortable using the word The preceding words are the true preface because I wrote them in 1984,
"thermodynamics" in the broad sense intended by its creator, William as I was starting the research for this book. I was then in the middle of a
Thomson (Lord Kelvin): this particular combination of the Greek words sabbatical leave at the University of Western Australia, which happened to
therme (heat) and dynamis (power) is a most appropriate name:!:for the field be my first official assignment as a professor at Duke. Upon my arrival at
that united the "heat" and "work" lines of activity that preceded it (Table Duke, I decided to use my enhanced freedom for the purpose of bettering
1.2, pp. 30-32). my research and my life in general. Thinking in depth about engineering
Finally, I view this as an "advanced" course in engineering thermo- thermodynamics was one result of that decision. The fact that large numbers
dynamics because it is the natural outcome of my own interaction with the of thermal engineers continued to regard the field as mature is precisely why
research arena and with students who were previously acquainted with the I picked engineering thermodynamics as a treatise topic: I not only saw merit
in questioning the established point of view, but I also knew that a true
research frontier is, quite often, the territory overlooked by the crowd.
tMasculine pronouns are used throughout this treatment only for succinctness. They are
As I look back at the past 3 to 4 years, I see a most gratifying project,
intended to refer to both males and females.
'The appetite for the "thermostatics" nomenclature is stimulated by comparisons with the a constant source of intellectual pleasure and new ideas. This project forced
dynamics/statics differentiation that is practiced in the field of mechanics: I believe that the me to think on my own about those areas-the gaps-of which I knew the
contemporary mechanics meaning of "dynamics" is being mistakenly viewed as the origin of least. It challenged me to be creative and produce my own version of what
"-dynamics" in "thermodynamics." fits best in any particular blank area. Overall, this book helped me diversify

and enrich my research, which is why during this period I was able personally While using those early drafts in the classroom, I collected many useful
to take steps in new directions, such as the axiomatic formulation of classical suggestions from the students, among whom I must mention: J. Gottwald,
thermodynamics (chapter 2), the graphic condensation of the relations J. L. Lage, P. A. Litsek, A. Mahajan, D. F. Mendivil, M. Wang, Z. Xia,
between thermodynamic properties (chapters 4 and 6), the design of power and Z. Zhang. Looking ahead, I will appreciate it very much if users of this
plants for maximum power (chapter 8), the theory of the ideal conversion book will write to call my attention to the imperfections that may have slipped
of solar radiation (chapter 9), and the design of refrigeration plants for into the final version.
maximum refrigeration effect per unit time (chapter 10). And, relative to
engineering thermodynamics as a whole, this book gave me the opportunity
to assemble in the same place many of the modern as well as the Durham, North Carolina
October 1987
long-forgotten references. I also used every opportunity to do what I like
best-produce original graphics.
Working on this book has been recreational. I did most of my thinking
while walking through the Duke Forest between my West Campus office and
our house in the Forest Hills section of Durham. I spent many hours
consulting the truly exceptional collection of books of the libraries of Duke
University. Ours is one university that from its early days in the 1800sinvested
in the important things. I made also many trips to the Library of Congress
in Washington, DC, where, while reading the original writings, I had a chance
to use the French, German, Latin, and Russian I learned in school.
The main contributor to the rewarding atmosphere of this project was
Mary. I have benefited from her wisdom, sense of strategy and intellectual
honesty during all my projects, big and small. This time, however, her
participation transcended a number of much more important projects: the
birth of child, the move from Colorado to North Carolina (via Western
Australia!), and the triumphant completion of her PhD in business ad-
ministration at the University of California, Berkeley. What I owe her is best
condensed in the dedication that opens my Convection Heat Transfer.
I also benefited from my year-long association with Dr. Peter Jany of the
Technical University of Munich, who generously contributed a most up-to-
date section on critical-point phenomena in chapter 6. I will always remember
the many conversations in which we compared notes on American engineer-
ing versus the German version, which had so much influence in Central
Europe and Russia.
I recognize also the contribution made by Linda Hayes, who not only
typed the manuscript, but also volunteered her rare talent of organization
and sense of symmetry to the raw material that I have produced. Her work
can be viewed directly in the Solutions Manual, which is available as a
separate book. This manual can be obtained by writing to Wiley-Interscience
(605 Third Avenue, New York, NY 10158-0012) or directly to me.
At various stages, I was helped by old friends, colleagues in academia,
and new students. Ren Anderson, Shigeo Kimura, Dimos Poulikakos, and
Osvair V. Trevisan kept me in touch with their respective corners of the
frontier and the literature. I am very grateful to my thermodynamics
colleagues at Duke, Prof. C. M. Harman, Prof. E. Elsevier, and Prof. J. B.
Chaddock, for commenting critically on early versions of the manuscript.


1.1 Elements of Thermodynamics Terminology, 1

1.2 The First Law for Closed Systems, 5
1.3 Work Transfer, 8
1.4 Heat Transfer, 13
1.5 Energy Change, 20
1.6 The First Law for Open Systems, 23
1.7 Historical Background, 29
1.8 The Structured Presentation of the First Law, 38
1.8.1 Poincare's Scheme, 38
1.8.2 Caratheodory's Scheme, 40
1.8.3 Keenan and Shapiro's Second Scheme, 40
References, 41
Problems, 44


2.1 The Second Law for Closed Systems, 49

2.1.1 Cycle in Contact with One Heat Reservoir, 50
2.1.2 Cycle in Contact with Two Heat Reservoirs, 52
2.1.3 Cycle in Contact with Any Number of Heat
Reservoirs, 60
2.1.4 Process in Contact with Any Number of Heat
Reservoirs, 63
~ ..~- '-' .......

2.2 The Second Law for Open Systems, 66 4.3 The Fundamental Relation, 157
2.3 The Local Thermodynamic Equilibrium Model, 68 4.3.1 Energy Representation, 158
2.4 The Entropy Maximum and Energy Minimum Principles, 71 4.3.2 Entropy Representation, 159
2.5 Caratheodory's Two Axioms, 77 4.3.3 Extensive Properties Versus Intensive Properties, 160
2.5.1 Reversible and Adiabatic Surfaces, 79 4.3.4 The Euler Equation, 161
2.5.2 Entropy, 84 4.3.5 The Gibbs-Duhem Relation, 162
2.5.3 Thermodynamic Temperature, 87 4.4 Legendre Transforms, 166
2.5.4 The Two Parts of the Second Law, 88 4.5 Relations Between Thermodynamic Properties, 175
2.6 A Heat Transfer Man's Two Axioms, 89 4.5.1 Maxwell's Relations, 176
2.7 Historical Background, 94 4.5.2 Relations Measured During Special Processes, 178
References, 97 4.5.3 Bridgman's Table, 187
Problems, 99 4.5.4 Jacobians in Thermodynamics, 188
4.6 Geometric Representations of Thermodynamic Relations, 193
4.7 Partial Molal Properties, 199
OF EXERGY 108 4.8 Ideal Gas Mixtures, 203
4.9 Real Gas Mixtures, 207
3.1 Lost Available Work, 109 References, 211
3.2 Cycles, 116 Problems, 213
3.2.1 Heat-Engine Cycles, 117
3.2.2 Refrigeration Cycles, 118
3.2.3 Heat-Pump Cycles, 121 5 EXERGY ANALYSIS
3.3 Nonftow Processes, 123
5.1 Nonftow Systems, 217
3.4 Steady-Flow Processes, 126
5.2 Flow Systems, 221
3.5 Mechanisms of Entropy Generation or Exergy Destruction, 133
5.3 Generalized Exergy Analysis, 224
3.5.1 Heat Transfer Across a Finite Temperature Difference, 134
5.4 Air-Conditioning Applications, 225
3.5.2 Flow with Friction, 136
5.4.1 Mixtures of Air and Water Vapor, 227
3.5.3 Mixing, 138
5.4.2 Total Flow Exergy of Humid Air, 229
3.6 Entropy Generation Minimization, 141
5.4.3 Total Flow Exergy of Liquid Water, 231
3.6.1 The Method, 141
5.4.4 Evaporative Cooling Process, 233
3.6.2 An Introduction: The Geometric Optimization of
5.5 Other Aspects of Exergy Analysis, 234
a Branching Fluid Network, 142
3.6.3 Entropy Generation Number, 145 References, 235
References, 147 Problems, 235
Problems, 148
6.1 The Energy Minimum Principle in U, H, F, and G
4.1 Simple System, 151 Representations, 239
4.2 Equilibrium Conditions, 153 6.1.1 The Energy Minimum Principle, 240