Chemical engineering education ( Journal Site )

Material Information

Chemical engineering education
Alternate Title:
Abbreviated Title:
Chem. eng. educ.
Physical Description:
v. : ill. ; 22-28 cm.
American Society for Engineering Education -- Chemical Engineering Division
Chemical Engineering Division, American Society for Engineering Education
Place of Publication:
Storrs, Conn
Publication Date:
annual[ former 1960-1961]


Subjects / Keywords:
Chemical engineering -- Study and teaching -- Periodicals   ( lcsh )
periodical   ( marcgt )
serial   ( sobekcm )


Chemical abstracts
Additional Physical Form:
Also issued online.
Dates or Sequential Designation:
1960-June 1964 ; v. 1, no. 1 (Oct. 1965)-
Numbering Peculiarities:
Publication suspended briefly: issue designated v. 1, no. 4 (June 1966) published Nov. 1967.
General Note:
Title from cover.
General Note:
Place of publication varies: Rochester, N.Y., 1965-1967; Gainesville, Fla., 1968-

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 01151209
lccn - 70013732
issn - 0009-2479
lcc - TP165 .C18
ddc - 660/.2/071
System ID:

Full Text

chemic*al eninering edcati

achdmwledtei aid t~aias...


with aoj o


Department of Chemical Engineering
University of Florida
Gainesville, Florida 32611

Editor: Ray Fahien (904) 392-0857
Associate Editor: Mack Tyner
Editorial & Business Assistant:
Carole C. Yocum (904) 392-0861
Publications Board and Regional
Advertising Representatives:

Lee C. Eagleton
Pennsylvania State University

Past Chairman:
Klaus D. Timmerhaus
University of Colorado

Homer F. Johnson
University of Tennessee
Jack R. Hopper
Lamar University
James Fair
University of Texas
Gary Poehlemn
Georgia Tech

Darsh T. Wasan
Illinois Institute of Technology
Lowell B. Koppel
Purdue University

William B. Krantz
University of Colorado
C. Judson King
University of California Berkeley

Angelo J. Perna
New Jersey Institute of Technology
Stuart W. Churchill
University of Pennsylvania
Raymond Baddour
A. TV. Westerberg
Carnegie-Mellon University

Charles Sleicher
University of Washington
Leslie W. Shemilt
McMaster University
Thomas W. Weber
State University of New York

Chemical Engineering Education

58 waud .eeclae

Input Multiplicities in Process Control,
Lowell B. Koppel
50 The Educator
John A. Quinn of Pennsylvania, Eduardo D.
Glandt, Douglas A. Lauffenburger
54 Department of Chemical Engineering
Delft University of Technology,
Julia A. Little
64 Lecture
Reference States and Relative Values of
Internal Energy, Enthalpy, and Entropy,
A. G. Fredrickson
70 Laboratory
Laboratory Experiment for the Transient
Response of a Stirred Vessel, R. D. Noble,
R. G. Jacquot, L. B. Baldwin
74 Curriculum
The Infusion of Socio-Humanistic Concepts
into Engineering Courses via Horizontal
Integration of Subject Matter, Charles E.
Huckaba, Anne Griffin
ChE Education in the Third World-Two
78 North American Assistance, P. L.
79 Need for International Cooperation,
H. K. Abdul-Kareem
86 Classroom
Distillation Calculations with a Program-
mable Calculator, Charles A. Walker,
Bret L. Halpern
53 Stirred Pots
Ballad of Jack Weikart, Rutherford Aris
72 Letters to the Editor
73 In Memoriam Joseph J. Martin
77-85 Book Reviews

CHEMICAL ENGINEERING EDUCATION is published quarterly by Chemical
Engineering Division, American Society for Engineering Education. The publication
is edited at the Chemical Engineering Department, University of Florida. Second-class
postage is paid at Gainesville, Florida, and at DeLeon Springs, Florida. Correspondence
regarding editorial matter, circulation and changes of address should be addressed
to the Editor at Gainesville, Florida 32611. Advertising rates and information are
available from the advertising representatives. Plates and other advertising material
may be sent directly to the printer: E. O. Painter Printing Co., P. O. Box 877,
DeLeon Springs, Florida 32028. Subscription rate U.S., Canada, and Mexico is $15 per
year, $10 per year mailed to members of AIChE and of the ChE Division of ASEE.
Bulk subscription rates to ChE faculty on request. Write for prices on individual
back copies. Copyright 1983 Chemical Engineering Division of American Society
for Engineering Education. The statements and opinions expressed in this periodical
are those of the writers and not necessarily those of the ChE Division of the ASEE
which body assumes no responsibility for them. Defective copies replaced if notified
within 120 days.
The International Organization for Standardization has assigned the code US ISSN
U009-2479 for the identification of this periodical.


' -- I .


;oh^ 4I. 2ai:

of Pennsylvania

University of Pennsylvania
Philadelphia, PA 19104

T HE YEAR 1983 MARKS 25 years in the teaching
profession for John A. Quinn, the Robert D.
Bent Professor and Chairman of the Department
of Chemical Engineering at the University of
Pennsylvania. Coincidentally, he has also grad-
uated his 25th PhD student this year, and there-
fore it seems appropriate to reflect on his career
in chemical engineering education and research.
Although he now finds himself comfortably at
home on Philadelphia's Main Line, John was raised
in the midwestern heartland, in Springfield, Il-
linois. After graduating from the University of
Illinois in 1954, he went east to Dick Wilhelm's
chemical engineering department at Princeton for
graduate work. His PhD thesis work on fluidiza-
tion was supervised by Joe Elgin, and by Leon
Lapidus upon Elgin's ascension to the post of
Dean. During those years, John learned the joys of
commuting through frequent trips to Bryn Mawr,
just outside Philadelphia and not far from his
present home, to visit his fiance. He and Frances
were married at the completion of his degree, and
they headed back to Champaign-Urbana where
John rejoined his alma mater as assistant profes-
sor in 1958. Promotion to associate professor came
in 1964 and to full professor in 1966.
The department at Illinois was then under the
leadership of Harry Drickamer who, along with

... John was recognized by
election to the National Academy of
Engineering in 1978. In the same year he
received the Alpha Chi Sigma Award
from the AIChE.

O Copyright ChE Division, ASEE, 1983

Wilhelm, Elgin and Lapidus, are recognized by
John as the major influences on him in his develop-
ing years as a researcher and teacher. Moving
away from fluidization as a field of study, he began
a program of inquiry into unexplored territory in
interfacial transport phenomena, which he has
continued to this day. John's early work in this
area earned him the Allan P. Colburn Award in
1966, for excellence in publications by a young
AIChE member. Not long after, his stay at Illinois
was interrupted by what he likes to call a "pleasant
interlude" on sabbatical at Imperial College in
London. Beyond this brief description, he is
slightly vague when junior faculty members ask
about how research ideas are generated during a
sabbatical leave.
In 1971, Art Humphrey, then department
chairman at Penn, succeeded in attracting John to
join his faculty. In this, John followed the path of
Dan Perlmutter who had come from Illinois to
Penn a few years earlier. John's addition had great
impact on the growth of the department, both


through his teaching and his research. Within
three years from his arrival, his classroom excel-
lence was recognized with the S. Reid Warren
Award for distinguished teaching. Four years
later, he was chosen to be the first incumbent of
the Robert D. Bent Professorship, established at
Pennsylvania by the Atlantic Richfield Founda-
tion. The Bent chair seems singularly appropriate
for a man of the stature of "Big John" (as he is
affectionately, but discreetly, known by the staff).
The experimental research program in inter-
facial phenomena carried out by John and his
series of students (five currently in residence) has
followed several avenues. In the early years, a
major portion of this careful work dealt with thin
liquid films, monolayers, and Marangoni effects.
Later, closely related efforts included studies of
jet-drop fractionation. Another important line of
investigation is concerned with membrane trans-
port phenomena. John and several of his students,
John Anderson (now at Carnegie-Mellon) among
them, pioneered the use of track-etched mica mem-

Despite being chairman, John gets much of his satisfac-
tion from his active research program. Here he is
shown with Patricia Rodilosso, one of his PhD students,
in the interfacial transport laboratory.

branes as a powerful tool for learning about equi-
librium and transport phenomena in pores. In ad-
dition, John has studied nonporous and natural
membranes, making significant advances in the
understanding of carrier-mediated and active
transport. The conception and development of
compact "membrane reactors," consisting of the
juxtaposition of catalytic and semipermeable lay-
ers, capable of performing chemically-driven sep-
arations or of forcing thermodynamically unfavor-
able reactions is a major advance in the technol-
ogy of membrane applications. This area, which


This gathering of his graduate students (with spouses
and friends) at the Quinn's for Christmas typifies the
warm relationships John establishes with those who
work with him.

grew out of the thesis research of one of John's
recent PhDs, Steve Matson, is now being hotly
pursued. For his long list of novel research contri-
butions, John was recognized by election to the
National Academy of Engineering in 1978. In the
same year, he received the Alpha Chi Sigma
Award from the AIChE, for outstanding chemical
engineering research. Among many other honors,
he was the Sixth Mason Lecturer at Stanford in
While much of the aforementioned research
work has relevance to biological systems, other
projects he has been involved with have had even
stronger connections with biology and medicine.
Among them, in collaboration with David Graves
of the Penn chemical engineering faculty, is the
investigation of transport of gases through skin,
with applications to oxygen supply by the micro-
circulation, prevention of decompression sickness,
and analysis of gas concentration in the blood for
clinical purposes.
Since chemical engineering at Penn has tradi-
tionally been strong in biochemical and biomedical
applications, it is not surprising that John's re-
search interests are often labelled as "bioengineer-
ing." He is very careful, however, when asked to
identify this as his distinct area of work, and is
quick to explain the role that it must play within
the chemical engineering profession. An analogy
often heard from him is to modern electrical en-
gineering, a discipline that succeeded in providing
wonders like hand-held calculators and mini-com-
puters because its practitioners were aware of the
advances in solid-state physics and were ready to


use them. John feels that chemical engineers
should at the present time be carefully observing
the developments in biochemistry and biophysics.
An engineer can choose either to help understand
the ways by which nature carries out chemical re-
actions so beautifully, or to exploit such under-
standing to new benefits. He has focused on the
latter goal, and his membrane reactors are a
superb example of his philosophy: "All engineer-
ing disciplines prosper as they are linked to a basic
John particularly enjoys reminiscing on the
evolution of chemical engineering education and
the chemical engineering profession, preferably
over a cold Rolling Rock. His early career took
place during the "golden sixties." Not only was
this a time of rapid growth for scientific and
technological programs in academia, but he will
point out that it also was a turning point for re-
search in chemical engineering. Chemical engi-

~C-. .

John is highly popular with the students for the clarity
and enthusiasm of his teaching.

... his membrane reactors are a
superb example of his philosophy: "All engineering
disciplines prosper as they are linked
to a basic science."

neers, once and for all, ended the impression that
their research consisted of outmoded chemistry.
John is apt to emphasize that chemical engineers
of today are capable of doing scientific research
as well as those in any chosen field. Contributions
by chemical engineers in fluid mechanics, for ex-
ample, are now cited by all fluid mechanicians.
The same goes for surface chemistry, as another
example. Chemical engineers can now work on the
fundamental, as well as on the applied, aspects of
science, without any self-imposed limits. One of
John's favorite phrases is "chemical engineering
is what chemical engineers do."
This philosophy has guided John's leadership
role at Penn, particularly since he accepted the
chairmanship in 1980. He is a great encourage-
ment to faculty pursuing research in areas that
stretch beyond the traditional boundaries once
supposed to encircle the discipline of chemical en-
gineering. The overriding principle is that what-
ever is undertaken, the goal should be to perform
at the pioneering edge of the chosen field.
Chairmanship of a department is usually ac-
cepted under protest, and John is no exception to
this rule. "To enjoy being a chairman", he says,
"you must get your kicks from seeing other people
do well." This is undoubtedly true in his case, for
he takes pride and pleasure in any accomplishment
of his colleagues, especially the younger ones. Yet
he still gets much of his satisfaction from his own
research and teaching. Accordingly, John has ex-
hibited an administrative style which relies heavily
on delegating responsibilities. We had been given
notice of this attitude when, immediately after be-
coming chairman in the summer of 1980, he went
away for a half-year mini-sabbatical at the Woods
Hole Laboratory and at MIT. The motivation for
those blessed with various duties is certainly
linked to John's admirable leadership qualities.
However, he will frequently claim that the effect
of a chairman on his department is highly non-
linear, meaning to emphasize that there is a satu-
ration effect after some time, and that rotation of
the position is healthy. This doesn't fool us,
though, being the same old protest dressed in en-
gineering language.
It is easy to catch John's enthusiasm for the
future of chemical engineering at Penn. He har-


bors no doubts about the role of engineering at an
Ivy League university, especially given the long
tradition of interaction with the various physical
and life science departments here. He enjoys em-
phasizing that it is an ideal place for remaining
closely linked to the basics, and therefore is "fertile

ground" for marshalling a first-rate research and
teaching program. It is clear that we are also ex-
tremely fortunate to have "Big John" (protocol
being temporarily suspended here) as our mar-
shal. O

stirred pots


by Rutherford Aris
Editor's Note: The following ballad was read by Prof. Aris
in the Exxon suite at the 1983 AIChE annual meeting. CEE
joins the academic community in saluting Jack on his re-
tirement from Exxon.
Jack Weikart was a citizen
Of tact and of rapport.
An Exxon captain eke was he
From Jersey's eastern shore.
When Exxon wished to fill its ranks
Or its largesse bestow
They called for Jack, the chap with knack,
The man most in the know.
Whether the good ol' D. of E.
Were pulling back on shale,
Or Exxon'd made discovery
Beyond the farthest pale,
Whether the stocks of gasoline
And oil were running low,
Or whether shieks of Araby
Had dealt their latest blow,
Whether the oil profits drew
Apologetic coughs,
Or whether they turned negative
In Reagonomic troughs,
The Admirals forth their captain sent
To fetch a Chem E. crew.
They said to Jack, "You bring 'em back
We know what you can do."
So out would come Jack's little book
In which he kept the tot
Of which was where, and who was who,
And who was doing what.
Then off he'd go as was his wont
To cull each Inst. and U.
Sometimes he wanted all the crop
And sometimes precious few.
But stay! Though Jack is truly tops
At the recruiting game

That is not all; he has, you know,
Another claim to fame.
Whenever A.I.Ch.E.s
In learned conclave meet
Our Jack you'll find dispensing cheer
Within the Exxon suite.
For oft-with Peg-he has reserved
A penthouse in the sky,
Four-poster bed, plush carpeting
And mirrors set on high.
And from the living room of such
A truly classy job,
With wine & beer & sundry drinks,
He entertains the mob
Of academic types who swarm
Until the place near bursts
With shouting profs all trying to slake
Their monumental thirsts.
(For word is passed from prof to prof
As each one registers,
"Psst. Write this number on your card
Before your vision blurs.
The Exxon suite's a perfect square,
Its digits four are too,
What's more it is the least of such
And has a splendid view.")
O'er all our Jack-and with him Paul-
Preside with great panache,
And woe betide industrial types
If they should try to crash!
For Exxon suites are sacred as
The groves of Academe;
There hiring's done, ideas are swapped
And hatched is many a scheme.
Oh, woe are we, no more to see
Jack Weikart's hoary head,
And hear him boom across the room,
"Time's up! Be off to bed."

Sail on! Sail on! Exxon. Exxon.
You'll never see, alack,
For fetching you a Chem E. crew
As mighty a man as Jack.



i department |


Virginia Polytechnic Institute
and State University
Blacksburg, VA 24061

A RE YOU TRYING TO locate lecture room 123? This
may be relatively easy at an American uni-
versity, where all you need do is differentiate the
many classrooms by their room numbers. But at
the Delft University of Technology in Delft, The
Netherlands, your predicted easy search for room
123 will instead lead to your wandering about the
Scheikundige Technologie (Chemical Engineer-
ing) building. You will observe laboratory upon
laboratory, but not many classrooms with desks
and blackboards. You may not realize it at the

A large reactor located in the Physical Technology lab-
oratory, one of the required laboratories for 3rd and
4th years.

time, but you are seeing classrooms, because at
Delft, the laboratory itself is a classroom.
I had the unique opportunity to discover the
organization and activities of the Delft Chemical
Engineering and Chemistry Department as an
exchange student participating in the Interna-
tional Association for the Exchange of Students
for Technical Experience (IAESTE) program.
Through my discussions with twelve members of
the department-professors, graduate students,
laboratory technical staff, and a program coordi-
nator-I was able to translate departmental docu-
ments and catalogues, compiling an informational
report on the educational experience at the Delft
University of Technology.
Chemical engineering and chemistry are com-
bined into a single department at Delft, so there
are no "pure" chemistry majors or chemical engi-
neering students. The department is sub-divided
into the following ten workgroups:

Chemical Engineering
Organic Chemistry
Analytical Chemistry and Laboratory Automation
Inorganic and Physical Chemistry
General Chemistry
o General and Applied Microbiology
Biochemical Reactors
General and Technical Biology
Technical and Macromolecular Systems

This particular combination of so many fields
has a historical basis in the Delft fermentation in-
dustry and in the production of rubber, tea, and
coffee in the former Dutch colonies.
The department currently has 430 employees

16 full professors
6 associate professors
120 academically-degreed professionals, of which 80 are
tenured research associates
40 non-permanent graduate students
290 technicians, secretaries, and service employees

Each of the sixteen professors typically has
two staff members, several graduate students, and
a varying number of fourth- and fifth-year stu-

Copyright ChE Division, ASEE, 1983


In addition to the six
different required laboratories, two
research projects initiated by current research being
performed by a professor, or by some aspect of
a Ph.D. project, are required.

dents working with him. Professors are highly re-
spected and highly paid ($50,000 per year) for
conducting research, teaching, and representing
the department through international travel,
seminars, and industrial contacts.
Approximately 130 new students enter the de-
partment each year. Because forty percent of these
students do not pass the Propaedeutic Examina-
tion after the first year, only approximately 55
students graduate per year. Considering only the
full professors, the student-to-faculty ratio is 31:1.
However, because of the academically-degreed pro-
fessionals, the tenured research associates, and the
graduate students, who also teach courses and in-
struct laboratories, the student-to-teacher ratio is
closer to three students per "teacher." There is
sufficient space for all: the professors, staff, and
students in the ten workgroups have their labora-
tories located in four large chemical engineering
and chemistry buildings.
The state government completely finances the
Delft University of Technology with the exception
of some research money from the research founda-
tion Z.W.O. and less than one percent from indus-
try. The students pay very little for their uni-
versity education: only f900, or $360, per year. In
addition, for many students cost of living expenses
are provided by the government in the form of
interest free loans. Because of the huge sums of
money invested in Delft by the government during
the past fifteen to twenty years, the university is a
showcase of technology for The Netherlands.
In the Chemical Engineering and Chemistry
Department, a lump sum from the government
pays wages, equipment costs, travelling expenses,
and other operation expenses. During the 1980
fiscal year, this sum amounted to approximately
2.25 million dollars. This figure does not include
salaries, however, because the state pays the staff
as civil servants directly. Salary costs were ap-
proximately $12.5 million in 1980. The chemical
engineering workgroup received 24% of all the
funds that were divided between the ten different
work groups during the 1980 fiscal year.
Because of government reluctance to continue
increasing its expenditures for higher technical
education, more pressure is now being placed on

researchers to obtain research funds from indus-
try. For the 1982 fiscal year the department will
receive approximately fifty percent of that received
in the past to finance equipment costs, and expecta-
tions for future years are not very good. Also, the
government has required a reduction in the dura-
tion of study programs from five years to four
years starting in September 1982. Students will
have "student" status only for six years.
The present five-year study program leads to
the title Chemical Engineer for a student who
passes the Propaedeutic examination, the Candi-
date's examination, and the Final Engineer's ex-

Small scale production and control of a process forming
catalyst consisting of an alloy on a carrier, located in the
Process Technology laboratory.

amination. The first two years are identical for all
students, whereas the last three years are referred
to as the "free study", during which time a student
specializes in one major field and one minor field.
Because the study program lasts for five years-
with the fifth year devoted to research-the degree
obtained from Delft is more similar to the M.S.
chemical engineering degree from an American
The new four-year study program-begun
September 1982-is very similar to the present
five-year program. The only fundamental changes
are a reduction in the credits associated with the
research project, design project, and literature
study, and the inclusion of such work in the fourth
year. Table 1 gives the proposed curriculum, which
can now be directly compared with the four-year
chemical engineering programs in the United
States. Note that semesters at Delft last for thir-
teen weeks, and that course credits are designated
as "c.p. units", where 1 c.p. unit equals approxi-
mately 40 hours.


From my observations, Dutch students work
very hard in high school, and enter the university
at a higher level than do students in the United
States (perhaps with the equivalent of a year of
college in the U.S.). Once they enter the university,
the question of student attitudes and motivation
can be raised. Many students take it easy, typically
completing the five-year chemical engineering pro-
gram in six to seven years, for example.
One reason why the chemical engineering study
program requires more time is the number of re-
quired laboratories. A more important reason is
that students take their final examinations only
twice a year, and not necessarily immediately fol-
lowing the end of class lectures. This delayed ex-
amination schedule lends itself to class cutting and
the postponement of studying until several weeks
before the yearly exams. But the chemical engi-

neering students are not an exception. While the
average chemical engineering student completes
his five-year program in six to seven years, the
average student at Delft finishes his five-year pro-
gram in seven to eight years, with architecture stu-
dents finishing typically in nine years.
The most outstanding quality I encountered in
all of the students that I met was their realization
of the importance of research, both to gaining new
knowledge and applying this knowledge to current
problems. The students are thinkers and innova-
tors, mainly because of Delft's emphasis on open-
ended research, challenging student projects, and
laboratory experience.
Because Delft considers laboratory experience
and research to be such an important part of a
complete chemical engineering education, attention
should be drawn to the outstanding facilities and


1st Semester
Mathematics I
Mechanics of Solids and Fluids
Inorganic Chemistry
Organic Chemistry
Quantum Chemistry
Technical Writing

2nd Semester
Mathematics II
Computer Programming
Transport Phenomena
Bioscience I

First Year Laboratories
(divided over both semesters)
Chemistry (IR, UV, AAS, GLC)
Chemistry and Society
Excursions to Industry

1st Semester
Mathematics III
Phase Equilibria
Separation Processes
Physical Chemistry

2nd Semester
Particle Technology
Reactor Design
Organic Chemistry II
Bioscience II

c.p. units




1st Semester
Process Design
Analytical Chemistry
Environmental Studies
System Modeling

2nd Semester
Industrial Chemistry
Process Equipment
Product Technology
The Chemical Plant

Second and Third Year Laboratories
Basic skills in:
Organic Chemistry
Inorganic & Physical Chemistry
Physical Technology
Process Technology
Research Project

Electives (3)
Industrial work
Literature study & report
Design Project
Research Project & paper



2x 3

3x 3


Absorption apparatus for second year project #47, Re-
moval of CO2 from process gases.
program the department has developed.
First year laboratories stress development
of good experimental and research techniques,
thoroughness, and report writing ability. One lab-
oratory located in the Propaedeutic Chemistry
building is devoted solely to gas-liquid chromatog-
raphy and infrared equipment for first-year stu-
dents. An assistant specialized in analytical tech-
niques is responsible for this GLC/IR laboratory.
During the first week, the chemistry laboratory
reviews high school techniques. Twelve to fourteen
inorganic and physical chemistry laboratory ex-
periments follow during the next twenty-three
weeks, scheduled during Thursday and Friday
mornings (9 a.m. to 12:30 p.m.) and afternoons
(1:30 p.m. to 5:15 p.m.). During the second se-
mester, six or seven organic laboratory experi-
ments are completed. Thus, during the first year a
total of eighteen to twenty experiments in organic,
inorganic, and physical chemistry are conducted,
with each experiment having a duration of approx-
imately six to ten hours.
The typical laboratory experience is much more
than one of just following, step by step, a standard

university experiment handbook. To perform an
experiment, a student, working individually, must
get an outline from one of the two laboratory as-
sistants (there is one laboratory assistant for
every ten students). The student must first review
the cited literature, then speak to the lab assistant
privately and answer preliminary questions that
he may raise. If the student is adequately pre-
pared, he is allowed to set up his apparatus. This
process of setting up usually takes an entire morn-
ing, or at most, an entire day. The laboratory as-
sistant must then approve the apparatus before
the student can begin the experiment. The data,
results, and theoretical post-experiment questions
are handed in for grading in very concise form.
Longer reports are written only three or four
times per year.
While the first-year and most of the second-
year laboratories are located in the Propaedeutic
Chemistry building, some of the second- and all of
the third-year physics and methodology labora-
tories and the basic skill laboratories in organic
chemistry, inorganic and physical chemistry, bio-
science, automation, physical technology, and
process technology are located in different work-
group laboratories.
In addition to the six different required labora-
tories, two research projects initiated by current
research being performed by a professor, or by
some aspect of a Ph.D. project, are required. The
general reasons for having these projects are to
have students learn to cooperate and to distribute
work among themselves, to stimulate research in-
terest, and to combine several different study
Continued on page 92.

Following the absorption of CO2, shown above, the CO2
is desorbed. Recovery concentrations and measure-
ments are made with the analysis equipment shown.




The 1982 Chemical
Engineering Division
Lecturer is Lowell B.
Koppel of Purdue Uni-
versity. The 3M Com-
pany provides financial
support for this annual
lectureship award.
Born and raised in
Chicago, Lowell gradu-
ated with the B.S. from
Northwestern University
in 1957, the M.S. from
the University of Michigan in 1958, and the Ph.D.
from Northwestern in 1960.
His interest in process control began in 1954
when he was a co-op student working as an opera-
tor's assistant in pilot plants, and his academic in-
volvement with the subject began in 1960 when he
started teaching the undergraduate process con-
trol course at Caltech.
He began teaching at Purdue in 1961, and
teamed with Don Coughanowr to write an under-
graduate textbook on process control. Most of his
work was in the general area of developing control
theory for application to processes typical of the
chemical and petroleum industry. He accepted the
position of head of the school in 1973 and served
in that capacity until 1981 when yearnings for
full-time concentration on teaching and research
drew him back to the classroom.
Current research interests are focused on
multivariable systems, and particularly on the ef-
fects of input multiplicity. Although he has taught
virtually every required course in the curriculum,
his teaching emphasis is on process control. A sec-
ondary area of interest is process design, which he
views as a subject intimately related to process
control, and a rapidly developing interest in inter-
active computer graphics. Avocations include
music, hiking, and books of all sorts.

Purdue University
West Lafayette, IN 47907

sometimes contain a small number of loops
which are sporadically troublesome. These loops
operate satisfactorily most of the time, but peri-
odically suffer sudden and unexplained reduction
in their level of stability. The reduction is some-
times observed as excessively oscillatory transients
following disturbances, and is corrected by retun-
ing. It has also been observed as a relatively sud-
den change from stable to unstable operation, re-
quiring that operating personnel remove the of-
fending loops from automatic operation, allow the
plant to settle down, and then reclose and retune
the loops.
It is often hypothesized that the cause of the
behavior is a gradual change in one or more plant
characteristics, causing changes in the process dy-
namics, which in turn render the control tuning
less stable. A typical example of such a character-
istic is catalyst activity. There are three pieces of
evidence, in my experience, which are not con-
sistent with this hypothesis. First, the change in
stability level is too sudden, and too large. Second,
when the loops are opened, and retuned to bring
the process back under control, the new controller
settings are often not significantly different from
those which were in effect when the destabilization
occurred. And yet, the stability level is returned to
normal by the returning. Finally, I have never
observed the consistent pattern of movement in
controller settings which one would anticipate if
the cause were a unidirectional phenomenon such
as catalyst decay or heat exchanger fouling.
An hypothesis which is not inconsistent with
this evidence is that the plant has moved relatively
suddenly to a new steady state, at which the con-
trol system is significantly less stable. However,
the integral (reset) action used in the great ma-
jority of process control loops implies that any new


Copyright ChE Division, ASEE, 1983

The research described here investigates the potential effect of input multiplicity on
multivariable chemical process control systems. Several simple processes are first shown to exhibit the
possibility of input multiplicity. Some effects of the multiplicity . are then illustrated ...

steady state cannot involve new values of the out-
put variables, i.e., the variables being regulated by
the control loops. These can come to rest only at
the set-point values. If there is a new steady state,
it has to be in the values of the input, or manipu-
lated variables.
The phenomenon of multiple steady states is of
general concern in chemical engineering, and there
is much current research on the subject. From the
viewpoint of process control, this previous re-
search has assumed that a given set of manipulated
inputs can produce more than one output at steady-
state. This will be called output multiplicity. For
example, the classic work by Aris and Amundson
[2] shows that a given flow rate of coolant to a
stirred exothermic reactor can lead to three differ-
ent steady state temperatures. A likely control sys-
tem for this process will manipulate the coolant
flow to regulate the temperature at some desired
set point value. In most cases, reset (integral ac-
tion) will be used in the controller. This prevents
any temperature but the desired set point value
from being a steady-state. The control system has
effectively eliminated the steady-state multiplicity.
However, if more than one coolant flow could
produce the same reactor temperature, we would
then have input multiplicity. The control system,
even with reset action, could conceivably come to
steady state at any of these coolant flow rates, and
therefore does not eliminate the steady-state
The research described here investigates the
potential effect of input multiplicity on multivari-
able chemical process control systems. Several
simple processes are first shown to exhibit the
possibility of input multiplicity. Some effects of
the multiplicity on control systems for these
processes are then illustrated by example. Finally,
I will discuss recent theoretical developments on
input multiplicity and closely related phenomena.


FIGURE 1. Flash separation process

are the temperature and pressure of the flash. The
outputs to be regulated are the mole fraction of
component 3 in the vapor product, desired at 0.4,
and the molar split of the feed into liquid and
vapor products, desired at 50-50. For the following
set of Antoine constants




p = exp + B)

with p and T in arbitrary units, the following
three steady states exist



of 3
in Vapor


Fraction of
Feed Going
to Liquid


Flash Separator. The process is sketched in Fig.
1. It involves isothermal flash separation of a four-
component equimolar mixture. Each component is
assumed to obey Rauolt's law, the ideal gas law,
and Antoine's equation. The manipulated inputs

For clarity, we are regarding the set points of
the cascaded pressure and temperature loops to be
the manipulated variables, but the same observa-
tions result if vapor draw-off rate and coolant flow
rate are regarded as m, and m,. A control system


*T2 T2ol T4 T401 T6 T6ol
FIGURE 2. Heat exchanger process
designed to hold cl at 0.400 mole fraction, and c2 at
50 % of the original feed, has three different ways
to do it, and steady-state could be reached at any
of them by a control system containing reset ac-
Heat Exchange Network. The process, contain-
ing three process heat exchangers and a utility
heater, is sketched in Fig. 2. The flow rate of the
utility stream is manipulated to regulate at 487.50
the temperature T5 of the heated stream leaving
the third process exchanger. The utility heater
regulates To at 5560. The utility stream heats the
first process stream in the first exchanger, recov-
ers heat from the second process stream in the
second exchanger, and heats the third process
stream in the third exchanger. For ease of illustra-
tion, the exchangers have been modelled as lumped,
or perfectly mixed, on each side. However, pre-
cisely the same effect can be demonstrated for a
network modelled with the more usual log-mean
driving forces.
For the following parameter values (in any
internally consistent set of units)

of multiplicity exhibited by the flash separator,
and that by the heat exchange network. The latter
is structure dependent. It occurs because the tem-
perature TI of the hot stream entering exchanger
3 decreases as more flow is demanded by the con-
trol system on this exchanger. This behavior could
not occur in a heat exchanger without the rest of
the process, in this case without the action of the
first two exchangers. In contest, the flash sep-
arator exhibits input multiplicity independently of
its interconnection with other process units. This
difference in type of multiplicity may be important
in dealing with the phenomenon.
Sequence of Reactors. Another example of
structure-dependent input multiplicity occurs in
the work of Kubicek et al. [9]. A sequence of two
stirred reactors, with the exothermal reaction
A -> B, was studied by these authors for output


UA12 = 100
T2o = 80.50
wCp2 = 10

UAI4 = 100
T4o = 665.80
wCp4 = 10

UA56 = 10
T6o = 4500
wCp, = 5

there are three values of wCpo for the utility
stream which will satisfy the control system. The
corresponding steady states are
wCpo To T1 T2 T3 T, T5 T,
1 556.4 127.6 123.4 612.5 617.4 487.5 475.0
6 556.4 269.7 252.5 508.3 522.6 487.5 475.0
25 556.4 429.5 397.7 492.5 508.3 487.5 475.0
Structure-Dependent Multiplicity. There is
clearly a fundamental difference between the type

FIGURE 3. Two-reactor sequence

multiplicity. The process and its steady-state be-
havior are sketched in Fig. 3. A given coolant
temperature used in the two reactors could pro-
duce several different steady-state temperatures
of the product stream leaving the second reactor.
However, examination of their plot (see Fig. 3)
also shows that a given product stream tem-
perature could be produced by as many as three
different coolant temperatures. This input multi-
plicity has not been observed for a similar reactor
(A -> B) operating singly, suggesting again that


The input multiplicity in this case occurs in
part because of the chosen temperature dependence
of the kinetics. Fig. 6 illustrates this point, and
shows that the input multiplicity would likely be
anticipated before the control system is designed.
In the next case, this is not likely.
Single Isothermal Reactor. In his Ph. D. dis-
sertation, Rickard [10] studied the six-component
reaction network

FIGURE 4. 2 x 2 reactor control scheme

process structure can play an important role in
input multiplicity.
Single Exothermal Reversible Reactor. In a
previous paper [7] a single exothermal reactor, in
which the kinetics were of the form A-R-S, was
shown to exhibit input multiplicity. The process is
sketched in Fig. 4, and the multiplicity behavior
in Fig. 5. The 2 x 2 control scheme involves manip-
ulation of reactor temperature and residence time
to regulate two of the product concentrations, CA
and CR. Fig. 5 illustrates the input multiplicity by
showing that contours of CA values between 0.53
and 0.55 will thrice intersect the contour for c, =
0.35. This corresponds to three sets of residence
time and temperature values which will yield
steady state at, for example, cA = 0.54, CR = 0.35.
This shows that more complex kinetics can pro-
duce input multiplicity in a single reactor, with-
out structure dependence.



L 0.7






0.6 0.7 0.8 0.9
FIGURE 5. Multiple steady states in 2 x 2 c

with mass-action kinetics. The reaction is con-
ducted in a single, well-stirred, isothermal reactor.
The 3 x 3 control system manipulates the concen-
trations of A5 and A6 in the feed, and the residence

z 0.6
_ 0.5
z 0.4

0.0 ---- -- I
0.6 0.7 0.8 0.9

1.0 1.

SI FIGURE 6. Open-loop cross plot shows input multiplicity

time, to control concentrations of A1, A2, and A3 in
Sthe product. It is thus similar to the one sketched
S' in Fig. 4. Rickard demonstrated input multiplicity
by exhibiting three different sets of the three
manipulations which will produce the desired set
of three exit concentrations. This example differs
from the previous one primarily in being of higher
10 1.I dimension. However, it is introduced because it
also has been shown to exhibit an unconstrained
control system limit cycle, which is an important aspect of input


ci lc,=.3
C50 CI0


c,= 0.95

clo c50so

0.0 0.336
FIGURE 7. Open-loop cross plot doesn't show input
multiplicity possibility

multiplicity discussed below, and because detection
of the input multiplicity cannot be based on a
simple parametric plot such as Fig. 6.
Fig. 7 [10] illustrates one such plot. It shows a
surface of the concentration of A1, plotted against
the disturbance variable, concentration of A1, in
the feed, and one of the manipulations, concentra-
tion of As in the feed. The other two manipula-
tions are held constant. All similar plots are
equally innocent for this process.
However, Fig. 8 [10] shows the same behavior
under closed-loop conditions for the other two
manipulations. That is, the concentration of As in

' C!,. :i:

the feed and the residence time are both manipu-
lated at each point to hold the product concentra-
tions of A2 and A8 at the set point. This plot, which
represents the behavior under the chosen control
system, shows the input multiplicity possibility.
Such plots are seldom made for real processes, in
part because of inadequate models and in part
because the calculations are almost always difficult.


These example processes are quite simple, and
the alternate steady states are often catastroph-
ically removed from each other. Typical chemical
processes are far more complex, and should be
capable of exhibiting this phenomenon at more
closely spaced steady states.


Divergence to Alternate Steady State. Input
multiplicity is of concern in process control be-
cause the multiple steady states are not eliminated
by the control system. Therefore, one obvious pos-

07 -


/; r~

-1 0 2 3 4

5 6 7 8 9

1 36

.. . ..

, I- _

0.0 0.336
FIGURE 8. Closed-loop cross plot needed to show input
multiplicity possibility


sible effect of the phenomenon is divergence of the
system from the design steady state to one of the
other steady states. This behavior is illustrated in
Fig. 9. and Fig. 10 for the single exothermal re-
actor, with the 2 x 2 control system sketched in
Fig. 4. Following a temporary disturbance in feed
concentration, the product concentrations cA and
CR are returned to the desired values. However, the
manipulations, in this case temperature and resi-
dence time, diverge to values at an alternate steady


- CR

This divergence would be difficult to detect in
practice. First, because manipulations are not al-
ways monitored. Second, because even if the
changes in manipulations are observed, they could
well be due to a permanent change in a disturbance
variable, for which the reset action has compen-
sated by permanently changing (resetting) the

1.o0 Temperature
0 0.8
0.6 Residence Time
< 0.5 -
t o6e
0 0.3
S0.2 -

0 1 2 3 4

5 6 7 8 9

manipulations. Identification and monitoring of all
possible process disturbances, which is not a prac-
tical idea, would be required to make this distinc-
tion with certainty.
The reactor control system is much less stable
at the new steady state. In fact, it has been shown
[8] that one of the loops is in positive feedback at
this steady state. Its gain has the wrong sign.
Therefore, even though the interactions stabilize
the two loops together, the system is much more
vulnerable than before to disturbances. Detection
of the phenomenon in practice is more likely to
occur indirectly, by observation of the undesirable
sudden reduction in stability level. This phenom-
enon could explain the erratic behavior of some
loops in chemical process control systems: they
function well most of the time, but occasionally
suffer a sudden reduction in stability level.
Unconstrained Limit Cycles. Another effect of
input multiplicity, which has been demonstrated
on the six-component reaction system studied by
Rickard, is appearance of an unconstrained limit
cycle. This is a periodic oscillation, of fixed ampli-
tude and frequency, in which none of the variables
ever reach constraints. This contrasts with a limit
cycle in which, for example, at least one of the
control valves is wide open or fully shut during
some portion of the cycle. These constrained limit
cycles usually occur because the control system is


a 1.0-


S0.1 -


so al-1



lllllllllllla fAAA AAAAA A A .


FIGURE 11. Outputs reach limit cycle around original
steady state

unstable, but fortunately cannot exhibit the theo-
retical behavior of an ever-growing oscillation
because of physical constraints on (typically)
process flow rates.
An example of the unconstrained limit cycle
behavior is illustrated in Figs. 11 and 12 [10]. The
six-component reaction scheme discussed above is
being conducted isothermally in a stirred reactor.
The control system is using the feed concentration
Continued on page 89.


FIGURE 12. Inputs reach limit cycle around alternate
steady state





"""'"'""""""""""""" ""'U'L"1L'

11 111111111111111111 1111111111 1111111111 11111111111

f 1l A A A A A A Aft A A A A


RN lecture




University of Minnesota
Minneapolis, MN 55455

M ANY PROBLEMS IN chemical engineering ther- -
modynamics require for their solution the
application of the balance equations for moles of
various chemical species, (total) energy, and
entropy. Fairly general forms of these equations,
which are applicable to open, moving and deform-
ing, and unsteady state systems, are

SdN= w, x,,j + S aij Ri
dt q I

d (Ut + Kt + Dt) = Q + Wu + Wexp
+ 2 wq (h, + k, + 0,)
dS S + SG + Wqs
dt a

(1) A. G. Fredrickson, professor of chemical engineering at the Uni-
versity of Minnesota, received his BS and MS at that institution in 1954
and 1956, and his PhD from the University of Wisconsin in 1959. He
has been the American editor of Chemical Engineering Science since
1975 and is the author of "Principles and Applications of Rheology."
(2) His current research area is the dynamics of microbial populations, and
his non-professional interests lie in the fields of natural history, pre-
(3) history, and photography.

Here, the energy and entropy transfer rates asso-
ciated with heat transfer, Q and S, are related to
the local heat flux vector q and temperature T on
the boundary DB of the system B by
Q = (n-q)

S = -f- (n-q)
The second law of thermodynamics requires that
the entropy generation rate SG shall be non-nega-
tive, and further, shall be non-negative for every
subsystem into which B may be divided. The quan-
tity aij is the stoichiometric coefficient of the jth
chemical species in the it" reaction, Ri is the rate of
the it reaction in the system, and the summation
on i in Eq. (1) is over a set of independent ra-
actions [1]. The fact that all valid chemical re-

Copyright ChE Division, ASEE, 1983

action equations must be balanced is expressed by
the conditions

Saij ejk = 0 (4)

and this must be satisfied for all reactions (i) and
all elements (k) ; Ejk is the number of atoms of the
kt" kind of element in one molecule (or meric or
repeating unit) of the jth chemical species. The
quantity w, is the molar flow rate of material at
the qth port of the system; it is positive if material
enters the system at that port but negative if ma-
terial leaves the system there. The mole fraction
of the j chemical species in this stream is de-
noted by x,,j. Finally, -W, and Wexp are the rates
at which the system performs useful and expan-
sion work on the surroundings, and Kt and Dt de-
note kinetic and potential energy.
Eqs. (1-3) are not valid for systems which
exchange matter with their surroundings by dif-
fusion across a permeable boundary, but they are


easily generalized to include that case also. It can
be shown that the results to be deduced here apply
to these more general systems also, but in the
interests of simplicity such systems will not be
considered further here.
The quantities U, and St which appear in Eqs.
(2) and (3) are the absolute values of the internal
energy and entropy of the system, respectively,
and the quantities hq and s, which appear in these
equations are the absolute values of the enthalpy
and entropy of one mole of the qt' process stream,
respectively. Thermodynamics provides means for
calculating changes in Ut, St, h,, s, when the sys-
tem and the qth process stream experience pro-
cesses which change their states, but thermody-
namics does not identify the states in which these
quantities are zero. It is sometimes held that the
third law of thermodynamics identifies such a
state for the entropy but, within the context of
classical thermodynamics at any rate, it is clear
that this is not so: the third law, or rather the
version of it usually called Nernst's heat theorem,
asserts only that the entropy changes of certain
kinds of systems undergoing certain kinds of
isothermal processes approach zero as the tem-
peratures of the processes approach absolute zero
[2]. Hence, we do not know how to find the absolute
values of Ut, St, h,, and s,.
However, we need to be able to assign numer-
ical values to internal energy, entropy, etc. if we
want to tabulate what effects changes of tempera-
ture, pressure, and composition have on them.
Hence, reference states for internal energy,
entropy, and enthalpy are chosen, and values of
internal energy, entropy, and enthalpy relative to
the reference states are tabulated. The most fa-
miliar example of this is the steam tables, but the
utility of enthalpy-composition and related dia-
grams demonstrated in the well-known Hougen,
Watson, and Ragatz book [3] make such diagrams
nearly as familiar as the steam tables.
Let Utru be the (absolute) internal energy of
the matter of the system being considered when
this matter is in the reference state for internal
energy, and let Strs be the (absolute) entropy of
the matter of the system when this matter is in
the reference state for entropy. Then the relative
internal energy, Ut', and relative entropy, St', of
the system are defined by
Ut'- Ut UtrL. (5a)
St' St Sts (5b)
and the calculation of Ut' and St' involves only cal-

culation of changes of internal energy and entropy.
The reference quantities do not depend on the tem-
perature and pressure of the system, though they
do depend on its composition. Similarly, if we let
h,rH be the enthalpy of one mole of matter of the
qth process stream when this matter is in the ref-
erence state for enthalpy, then we can define a
relative enthalpy

hq' h hqr- h,.H
as well as a relative entropy

Sq Sq Sqrs



for the qth process stream. It is necessary to place
the subscript q on the reference state quantities
hqrH and Sqrs because these quantities depend on
the composition of the qth stream, and this is not in
general the same as the composition of other pro-
cess streams. However, hqr. and Sqrs do not depend
on the temperature and pressure of the qt1 process
If the relative properties Ut', St', hq', s,' are to
be useful then they should obey the same balance
equations as do the absolute properties Ut, St, hq,
s,; that is, the relative properties should obey

dt (Ut' + Kt + t)
= Q + Wu + Wexp + wq (hq' + kq + f,) (6)
dt -S+S G+ Wq Sq (7)
where Q, Wu, Wexp, S, and SG are the same quan-
tities that appear in Eqs. (2) and (3). Clearly, if
the relative properties satisfy these balance equa-
tions they then can be used (correctly) in the cal-
culations of heat and work effects that occur so fre-
quently in applied thermodynamics. I have been
unable to find a recent discussion of the choices of
reference states which will lead to Eqs. (6) and
(7), and so provision of such discussion is the ob-
jective of this paper.

The Pure Chemical Compounds at Specified Conditions
of T and p
Although it would be possible to choose refer-
ence states which are solutions we shall not do so
here. Instead, we shall consider only reference
states in which only pure substances, unmixed
with other substances, are present, and present in
states of aggregation that are at least metastable.


If the relative properties... are
to be useful then they should obey the same balance
equations as do the absolute properties ...

To begin with, we choose the pure substances to be
the pure compounds of which the system and the
process streams are composed. The reference state
for internal energy is assumed to involve in the
most general case a different combination of tem-
perature and pressure for each compound, and the
temperature and pressure of the reference state
for internal energy for a given compound are as-
sumed to be different from the temperature and
pressure of the reference state for entropy for that
compound. If it is not possible to choose reference
states in this very arbitrary way, we want to find
that out.
It is easy to show that Eqs. (6) and (7) follow
from Eqs. (2) and (3) and the definitions of Eqs.
(5) if, and only if, the reference properties satisfy

dUtr wq hqrH = 0
dt q

dStrs Wq Sqrs = 0

These equations are general; they must be satisfied
for every valid choice of reference states of Ut, Ht,
and St. It might be supposed at first that the time
derivatives in these equations are zero, but such is
not in general the case because, as pointed out
above, the reference state quantities Utru and Strs
depend on the composition of the system and this
in general is time varying.
Consider the second of these equations first.
Let sjrs be the (absolute) molar entropy of the
pure jth compound when it is in the reference state
for entropy. Then
Strs = Z Nj Srs
sqrS = 2Xq,j SjrS

and from these we find that

dSt,s q I dNj
dt q d dt q
= Z Sjrs Z aij Ri
3 i
= Z Ri aij Srs (10)
i i
where Eq. (1) has been used. If no chemical re-
actions occur, Eq. (10) shows that Eq. (9) is satis-


fled by arbitrary choices of the entropies Sirs, but
if reactions occur, Eq. (10) shows that Eq. (9)
will be satisfied if and only if we chose these en-
tropies such that

Saij Sirs = 0


for each and every independent reaction occurring.
This shows that if chemical reactions occur, the
conditions of temperature and pressure for the
entropy reference states of the compounds of
which the system is composed cannot be chosen
arbitrarily, but must be chosen so that the en-
tropies Sirs satisfy a number of constraints equal
to the number of independent reactions which oc-
Consider now Eq. (8). Let ujru (hjm) be the
(absolute) molar internal energy (enthalpy) of
the pure jth compound when it is in the reference
state for internal energy (enthalpy). Then
UtrU = 2 Nj ujru

hqrH = Xqj hjrH

and so

dUtrI dN )
dt -wqhqr j dt -har
= (u h dt
= (ujru hjr) --^

+ Z Ri aij hjrH
i j


where again Eq. (1) has been used. This result
shows that in the general case Eq. (8) will be sat-
isfied if the reference states for internal energy
are chosen so that

ujrU = hjrH
for all compounds present and
Y aij hjrH = 0



for all independent reactions occurring.
In summary of the case where the reference
states involve the pure compounds, we state the
following. If no reactions occur, the entropy refer-
ence states for the compounds can be chosen with
complete arbitrariness, and the same is true of
either the internal energy reference states or the
enthalpy reference states for the compounds. Once
one of these latter sets of states is chosen, however,
the other set of states must be chosen so as to sat-
isfy Eq. (13) ; if Eq. (13) is not satisfied, Eq. (6)
will not be true for unsteady state situations in-


volving open systems even though no reactions
occur. If chemical reactions occur, Eq. (13) must
still be satisfied, and in addition the reference
states for entropy and enthalpy must satisfy the
constraints imposed by Eqs. (11) and (14). These
constraints are inconvenient; Eq. (14) would
force us to choose reference state temperatures for
the enthalpies of gases to be different for different
gases, for example. Hence, we consider a reference
state which is not subject to such constraints.

The Pure Elements at Specified Condition of T and p
We assume now that material in a reference
state is present as the pure elements, unmixed
with other elements or compounds, in stable or
metastable states of aggregation and (in general)
at different conditions of temperatures and pres-
sure for different elements.
Let SEkrS be the (absolute) molar entropy of the
kt" kind of element in the reference state for en-
tropy. Then

Strs = I Nj Ejk SEkrS
j k
SqrS = 2 X,,j Ejk SEkrs
j k
From these, we get

dStrs SS dNj
dt s,,s = Ejk Srs w, Xj
tq j k\ q
= 2 Ri SEkrs aij Ejk = 0
i j j
because of Eq. (4). Hence, with this choice of ref-
erence state for entropy, Eq. (9) will always be
satisfied, even for systems in which chemical re-
actions occur.
Let UE kr (hEkrH) be the (absolute) molar in-
ternal energy (enthalpy) of the kth kind of ele-
ment in the reference state for internal energy
(enthalpy). Then

Utru = I Nj Ejk uEkrU
j k
hqrH =X I Xq,j Ejk hEkrH
J k
From these, we find that Eq. (8) will be satisfied
for all cases if we chose reference states for in-
ternal energy and enthalpy such that

UEkrU = hEkrH (15)
and the fact that chemical reactions occur imposes
no constraints of the types imposed by Eqs. (11)

and (14). We see, therefore, that the temperature
and pressure of the reference state for entropy of
a given element can be chosen with complete arbi-
trariness, as can the temperature and pressure of
the reference state for internal energy or enthalpy
of a given element.

... this... leads to a problem
that cannot be solved: we cannot know
how to adjust the heat transfer rate to the gas so
that its absolute total internal energy remains
constant as its mass changes.

The foregoing analysis shows that if Eqs. (6)
and (7) are to be correct in all circumstances, then
the temperatures and pressures of any pure sub-
stance (compound or element) in the reference
states for internal energy and enthalpy must be
such that

UrU = hrH


where we have dropped subscripts and super-
scripts identifying the particular element or com-
pound considered. If the mole numbers Nj are in-
dependent of time, Eq. (6) will be valid even if
reference states are not chosen so that Eq. (16) is
valid. That means that it is not necessary to chose
reference states so that this equation is true if we
are dealing with steady state open systems (in
which reactions may be occurring) or closed sys-
tems in which no reactions are occurring. In all
other cases, however, Eq. (16) must be satisfied if
Eq. (6) is to hold.
The internal energy and enthalpy in the refer-
ence state for internal energy must satisfy the
usual relation.
hu = ru, + Pr, v,. (17)
and combination of this with Eq. (16) shows that
hrH hru = Pru Vr (18)
This equation tells us how we must choose the ref-
erence state for enthalpy if the reference state for
internal energy is chosen. In the 1967 ASME
steam tables, for example, the reference states for
entropy and internal energy are chosen to be liquid
water at the triple point (273.16K, 611.2N/m2),
and the p-V product for liquid water at this condi-
tion is 11.0 J/kmol. It follows from Eq. (18) that
the reference state for enthalpy can be any state
for which the molar enthalpy of water is less than


the molar enthalpy of liquid water at the triple
point by 11.0 J/kmol. This difference could be
achieved by lowering the temperature of liquid
water by 0.0026K while keeping pressure constant
at 611.2 N/m2, for example. Liquid water at this
condition (metastable, by the way) has an en-
thalpy which is less than the enthalpy of liquid
water at the triple point by 11.0 J/kmol. However,
other states could also serve as the reference state.
One does not have to make an actual choice of
reference state for enthalpy once the reference
state for internal energy is decided upon. This is so
because when these reference states satisfy Eq.
(16) for every compound or element, then the rela-
tion between relative enthalpy and internal energy
is the same as the relation between absolute
enthalpy and internal energy. Thus, if the refer-
ence states involve the pure elements, we have for
the system
Ut = Ut' + I Z Nj Ejk uEkrU
j k
Ht = Ht' + S Z N, Ejk hEkrH
j k
Ht U = (Ht'- Ut')
+ Y Y Nj Ejk (hEkrH UEkrU) = Ht'- U' (19)
j k
if Eq. (16) is satisfied. Similarly, one can show
that h, u, = h,' u,' if Eq. (16) is satisfied. For
equilibrium systems a system pressure can be de-
fined and H, Ut = Ht' Ut' = pV so the relative
value of enthalpy is fixed once the relative value of
internal energy is stated, provided that Eq. (16)
is satisfied by the choice of reference states.

Problem 5.11 in the book of Modell and Reid [4]
provides an interesting application of the fore-
going ideas. In this problem the total internal
energy of a gas (helium) in a tank is to be held
constant as gas flows out of the tank, and this is to
be done by addition of heat to the gas in the tank
at the proper rate; one is to calculate how the
pressure and temperature of the gas vary with the
amount of gas remaining in the tank. Since no
chemical reactions occur, reference state 1 may be
used. Indeed, reference states 1 and 2 are identical
in this problem, but they would not be if the gas
involved was a chemical compound composed of
two or more elements.
The system to which mole and energy balances
are to be applied is the gas in the tank, and the
simplifying assumption that the gas has a spatially

uniform state at any time is made. Eqs. (5a) and
(16) then provide, for an interval of time t to t +
dUt = dUt' + u,,dN = dUt' + hrdN (20)
and in addition we have

Ht Ut= Ht'- Ut'= pV


when uru = hrH. Finally, the energy balance yields

dUt' = dQt + h'dN


when mechanical energy terms in the equation are
assumed to cancel one another. No expansion work
term appears in this equation since the system vol-
ume is constant and no useful work term appears
because we assume energy is added or removed by
heat transfer rather than by performance of use-
ful work.
If we interpret the statement of the problem in
the Modell and Reid book to mean that absolute
internal energy (total) is constant, then dUt = 0
and from Eqs. (20-22) we obtain after some

-hdN = Ndh'-Vdp (= dQ)


where the volume, V, of the system is constant.
Since N, h, and h' are functions of T, p, and V this
is an ordinary differential equation in the two state
variables T and p. Certainly, we can write equa-
tions for the functional dependence of N and h' on
these quantities for any given gas, once a choice of
reference state has been made, but we cannot do
the same for the absolute molar enthalpy, h. Hence,
this interpretation of the problem statement leads
to a problem that cannot be solved: we cannot
know how to adjust the heat transfer rate to the
gas so that its absolute total internal energy re-
mains constant as its mass changes.
On the other hand, if we interpret the state-
ment of the problem to mean that the relative in-
ternal energy (total) is constant, then dUt' = 0
and from Eqs. (20-22) we get

-h'dN = Ndh'-Vdp (= dQt)


Every term in this equation may be evaluated. If,
for example, we assume that helium is a perfect
gas with molar heat capacity cp independent of
temperature, then

h'= cp(T-TrH)

N pV



and Eq. (16) requires that the temperatures in the


reference states for internal energy and enthalpy

yTrH = Tr,


where y = c,/cy. One notes in passing that refer-
ence state temperatures for internal energy and
enthalpy are substantially different in the case of a
gas; as we noted, these temperatures were prac-
tically the same for a condensed phase because the
pressure-volume product is usually very small for
Substitution of Eqs. (25-27) into Eq. (24)
yields the differential equation
dp dT dT
+ (28)
p T + Tu- T (
and this may be integrated subject to an appropri-
ate initial condition to yield the relation between
T and p that must be satisfied if Ut' is to be con-
stant when N varies. The result contains the refer-
ence temperature Tru, and this is to be expected,
since if we change Tru, we change the relative in-
ternal energy of the system when it is in a given
state. O

1. R. Aris, "Introduction to the Analysis of Chemical Re-
actors," Chap. 2 (Englewood Cliffs, N.J.: Prentice-
Hall, Inc., 1965).
2. E. A. Guggenheim, "Thermodynamics, An Advanced
Treatment for Chemists and Physicists," pp. 158-63
(Amsterdam: North-Holland Publishing Company,
3. 0. A. Hougen, K. M. Watson, and R. A. Ragatz, "Chem-
ical Process Principles. Part I. Material and Energy
Balances," 2nd. ed., Chap. 9 (New York: John Wiley &
Sons, Inc., 1954).
4. M. Modell and R. C. Reid, "Thermodynamics and its
Applications," p. 141 (Englewood Cliffs, N.J.: Prentice-
Hall, Inc., 1974).

c,,cv Heat capacities at constant pressure and
volume, respectively, per mole
Ht Enthalpy, total
h Enthalpy, per mole
Kt Kinetic energy, total
k Kinetic energy, per mole
N Number of moles in system
n Unit outer normal vector to boundary aB
of system B
p Pressure
Q Rate of heat transfer into system
Qt Amount of heat transferred, total
q Heat flux vector

a special fall issue devoted to graduate education. This issue
consists of articles on graduate courses and research written by
professors at various universities, and of announcements placed
by ChE departments describing their graduate programs. Any-
one interested in contributing to the editorial content of the fall
1983 issue should write to the editor, indicating the subject of
the paper and the tentative date it can be submitted. Deadline
is June 15th.

R Rate of reaction
R, Gas constant
S Rate of entropy transfer into system
So Rate of entropy generation in system
St Entropy, total
s Entropy, per mole
T Absolute temperature
t Time
Ut Internal energy, total
u Internal energy, per mole
V Volume
v Volume, per mole
Wt Work, total
-Wexp Rate at which system does expansion work
on surroundings
-W, Rate at which system does useful (shaft,
electrical, etc.) work on surroundings
w Molar flow rate
x Mole fraction
aij Stoichiometric coefficient of jt' compound
in ith reaction
y Ratio of heat capacities
Ejk Number of atoms of kth element in one
molecule of j 1t compound
4t Potential energy, total
0b Potential energy, per mole

E Denotes property of a pure element
Denotes a relative property


G Generation
S Denotes itl' reaction
S Denotes jth chemical compound
k Denotes kth element
, Denotes qtl* process stream
t Total
rH, rS, rU Denote reference states for enthalpy, en-
tropy, and internal energy






A LABORATORY EXPERIMENT has been developed
to measure the transient response of a stirred
vessel. The experimental apparatus is simple in
construction, inexpensive to purchase, and gives
good quality data which demonstrate the phenom-
enon being tested. The apparatus can be used for
two different experiments which demonstrate
transient behavior, thus further reducing the cost
per experiment. Both experiments use salt dilution
as the method of demonstration.
In one experiment, the vessel is initially
charged with a known volume of water and known
weight of salt. A measured inlet flow of water is
started and the students determine salt concentra-
tion vs. time using a conductivity meter. They then
*Paper published in Proceedings of Frontiers in Educa-
tion Conference, Rapid City, SD (1981).
1. University of Colorado, Boulder, CO 80303. 2. Uni-
versity of Wyoming, Laramie, WY 82071. 3. University of
Wyoming, Laramie, WY 82071.

compare their experimental results with theoret-
ical predictions. This experiment demonstrates the
effect of volume change in the vessel on transient
The second experiment is set up in the same
fashion as the first except an outlet flow from the
vessel is used so the tank volume remains constant.
Students measure the salt concentration in the out-
let as a function of time and compare this to theo-
retical predictions. This experiment demonstrates
the effect of inflow and outflow on transient re-
The quality of the data obtained is very good
and allows the student to observe the phenomenon
and see how the theory is actually applied. Stu-
dents are also asked to comment on sources of
error in the experiment.


A macroscopic mass balance for a given species
(component) in a system is

Copyright ChE Division, ASEE, 1983

Richard D. Noble received his
B.E. degree in 1968 and M.E. degree
in 1969 from Stevens Institute of
Technology. In 1976, he received
his PhD degree from the University
of California, Davis. His current
research interests include facilitated
transport in liquid membranes,
transient heat transfer, and problem
solving skills. (L)
Raymond G. Jacquot is currently
Professor of Electrical Engineering
at the University of Wyoming where
he served in the various academic
ranks and served as Electrical En-
gineering Department Head for two
and one half years. Dr. Jacquot's education includes B.S. and M.S.
degrees in Mechanical Engineering from the University of Wyoming
and the PhD in M.E. from Purdue. His professional interests are in
dynamic systems and control and he is the author of Modern Digital
Control Systems. He is a member of ASEE, IEEE, ASME and Vice Chair-
man for Regional Activities for the CoEd division of ASEE. (C)
Leonard B. Baldwin is currently Professor of Civil Engineering at
the University of Wyoming. He holds B.S. (Physics) and M.S. (Civil

Engineering-Mechanics) degrees from Michigan State and the PhD in
Civil Engineering from Stanford. He has held academic appointments
at the University of Idaho, Michigan State, Tufts, Stanford and Mich-
igan Technological University. His professional interests are in hy-
draulics of pipelines, hydrology and fluid mechanics. He is currently
authoring text material in statics and dynamics with accompanying
audio-tutorial tapes and slides. He is a member of ASEE. (R)


d (CV) + QCo- QiCi-R = 0

where: V

= system volume
= species concentration
= volumetric flowrate
= generation term
= inlet
= outlet

The systems to be studied have no chemical re-
actions taking place (R= 0) and are contained in
a well-mixed vessel. Therefore, the outlet concen-
tration Co and the system concentration C are as-
sumed equal. Eq. (1) will be applied to the salt
contained in the vessel for each case. Also, the in-
let stream contains pure water so Ci = 0.
For the first experiment, the outlet flowrate Qo
is zero and Eq. (1) becomes

d (CV) = 0 (2)
The solution for this equation is

C CIV (3)
V1 + Qt
where the subscript 1 refers to the initial state of
the system.
For the second experiment, the volume in the
vessel remains constant and the outlet has a con-
stant flowrate. Eq. (1) becomes
V C + QC = 0 (4)
The solution to Eq. (4) becomes

C = Czexp(-Qot/V) (5)

Eqs. (3) and (5) are the theoretical predic-
tions against which the experimental results will
be compared.

A schematic of the experimental apparatus is
shown in Fig. 1. A cylindrical vessel (nominal
size: 10 gallons) has an outlet at the bottom with
a valve to adjust or shut off flow. A water line with
a valve for flow adjustment serves as the inlet. The
water line should have a flexible section at the end
so it can be removed from the vessel when neces-
sary. A stirrer is mounted so that the contents of
the vessel are well-mixed at all times. To further
reduce costs, students can provide the mixing
power. Not shown in the schematic is an electrical
conductivity meter which is used to measure salt

The apparatus can be used
for two different experiments which
demonstrate transient behavior, thus further reducing
the cost per experiment.

FIGURE 1. Schematic of experimental apparatus
concentration in liquid samples and a stop watch.
To perform the experiment with the outlet
valve closed (amount of salt in system is constant
and volume changes with time), a measured mass
of salt and volume of water is added to the tank
and well stirred. The initial concentration of salt
is measured with the conductivity meter. It is im-
portant that all the salt dissolve into solution. The
inlet line is turned on and the flowrate of water
measured. This is done while the line is not dis-
charging to the tank. At t=o, the inlet line is
placed into the tank and the stop watch started.
Samples are withdrawn periodically from the tank
and the salt concentration measured. These experi-
mental values are then compared to theory using
Eq. 3. Fig. 2 shows some experimental and theo-
retical results for this experiment.
To perform the experiment with the outlet
valve open (volume remains constant and amount
of salt decreases with time), the tank is initially
filled with water and the inlet flow rate is adjusted
until the volume remains constant in the tank with
the outlet valve open. The outlet valve is closed
and the inlet line removed from the tank. A meas-
ured quantity of salt is then mixed into the system
and an initial conductivity reading taken. At t= o,
the inlet line is replaced in the tank, the outlet is
reopened, and the stop watch started. Samples are
taken periodically and the salt concentration is
measured. Eq. 5 is then used to compare the experi-
mental results with theory. Fig. 3 shows a com-
parison of theoretical and experimental results.
For both experiments, the student is given the
macroscopic mass balance and asked to derive the
theoretical result and compare it to the experi-



V1 = 2 gals
Q = 0.61 gals/min
cr = 9.8 mU/cm
V /Q = 3.2 min

3 1 I I I I I
0 1 2 3 4 5

W TIME, min
FIGURE 2. Results for closed outlet experiment

mental result. Also, the student is asked to com-
ment on any sources of error in the experiment.


As seen by Figs. 2 and 3, the comparison be-
tween experimental and theoretical results is very
.d rTh d1,A4 -1-Z A 4d4 1 il d

The experimental procedure is also simple and
allows the student to focus attention on the phys-
ical phenomenon taking place and not get im-
mersed in the procedural detail. Students can per-
form multiple experiments since the procedure is
simple and short in time. This also reinforces the
validity of their results. Multiple tests could also
be used to instruct students in data analysis (mean
and variance, for example).
It is also quite simple to set up multiple experi-
mental apparatuses so that many different groups
of students could perform the experiments simul-
taneously. After performance, the apparatuses
could also be removed and stored so that the space
was available for other use.
In conclusion, the following points can be made.
* An experimental apparatus has been described
which demonstrates the transient response of a
stirred vessel. The apparatus is flexible in appli-
cation and can be used for two different experi-
* The experimental results are very good and con-
sistent with theory.
* The apparatus is inexpensive and simple to op-
erate. [

S letters

guuu. eIlt UiaIa b ubita ne s
consistent with theory. This
and the limitations of the
and removes it from a strict
The experimental appar:
and inexpensive. Yet, it pr
to perform two different ex]
further reduce storage spa



- <




FIGURE 3. Results for consta

very reprouuucLbJ alnu Sir:
reinforces the validity For some years I have used the films by Noel de Nevers
theory for the student entitled Phase Behavior, Parts I and II in my courses in
;ly textbook context. thermodynamics. Using high-pressure visualization equip-
atus is simple in design ment and time-lapse photography, Prof. de Nevers shows
examples of phase transitions in both pure and multicom-
ovides the opportunity ponent systems. The films demonstrate skillfully those
periments and thereby, aspects of fluid-phase behavior that are frequently dis-
ce and cost per experi- cussed by chemical engineers.
Because so much information is presented, however, I
have found that students often miss some of the subtleties.
In most sequences viewers must watch a moving phase
boundary along with temperature and pressure gauges, and
then correlate their observations with Prof. De Nevers'
0.45 gals/mn commentary. For the beginning student, this can be quite difficult.
3.7 gal
000 grains/gal To help solve this problem, I have prepared brief sum-
maries of the sequences. These may be discussed with
students both prior to and after showing the films and also
theory used as a basis for more lengthy study of phase behavior.
I believe that the films, together with such discussions,
have great pedagogical value in thermodynamics, and I
would be pleased to make these write-ups available to
others on request. The two films currently have a rental
II I cost of $14.50 each and may be obtained from the Uni-
8 12 16 versity of Utah, Instructional Media Services, 207 Milton
E, min Bennion Hall, Salt Lake City, UT 84112.
Kenneth R. Jolls
ant volume experiment Iowa State University


Js Memowamph

Joseph J. Martin

Joseph J. Martin, professor of chemical engi-
neering at the University of Michigan and associ-
ate director of its Institute of Science and Tech-
nology, died on December 13, 1982, just ten days
before his 66th birthday.
Joe had been a member of the University of
Michigan faculty since 1947, and in recognition of
his many contributions and excellence in teaching
was to have received the 1982 Outstanding
Teacher of the Year Award from its College of
Engineering. The award was presented to him
posthumously, and was only the latest addition to
a long list of honors, awards, and prizes given to
Joe over the past 40 years for his work and con-
tributions, primarily in the field of thermody-
A native of Anita, Iowa, Joe received his BS
from Iowa State University in 1939, his MS from
the University of Rochester in 1944, and his DSc
from Carnegie Mellon University in 1949. He was
acting director of the Institute of Science and
Technology at the University of Michigan from
1978 to 1981, founder and first chairman of the
Association for Cooperation in Engineering, and
former president of AIChE, ASEE, and the Engi-
neers Joint Council.
It was through the Association for Cooperation
in Engineering that Joe's goal of a unified voice
for engineering came to fruition. A visible result
of his efforts is this journal, CEE, which came into
being during Joe's tenure in ASEE. He also served
on the Engineers Council for Professional Devel-
opment and was chairman of the Education and
Accreditation Committee of the AIChE at the time
of his death. He was the author of two books and
more than 100 technical papers.
Joe's love and appreciation for thermodynam-
ics was unwavering and he and his students de-
voted over four decades to obtaining precise
thermodynamic data of substances so as to provide
the testing ground for the "Holy Grail" of thermo-
dynamics-a General Equation of State. Nearly a
third of his more than 100 publications were re-

lated to this effort. The remaining two-thirds may
have been overshadowed by his achievements in
thermodynamics, but in their own right were sig-
nificant landmarks in other disciplines. For ex-
ample, recognition of Joe's pioneering work in
radiation chemistry led to his election as chairman
of the Division of Nuclear Chemistry and Tech-
nology of the American Chemical Society and also
to the chairmanship of the Nuclear Engineering
Division of the AIChE.
Joe's attitude toward his profession and his
motivation for participation in society activities is
best expressed in his own words:
We have an unusual collection of talent in our member-
ships, drawn from industry, government, and education,
and are capable of directing it in a relevant manner for
the best interests of the individual, the specific group,
and the nation as a whole. Thus, a profession does not
exist in a vacuum-but derives its meaning, value and
goals through both its responsiveness to needs of so-
ciety and its influence on the direction of society. Being
an engineer carries with it a serious responsibility which
must be met in a considered, thoughtful manner by the
engineers who are developing the new technologies if
these advances are to play a positive role in our society.
Joe's energies and activities were not, however,
restricted to his scientific field. He was a longtime
member of the Ann Arbor Housing Commission,
having served from 1968 to 1978, and was an avid
tennis player. A boyhood interest in the game led
to his lifelong love of the sport and he played com-
petitively for over 35 years. He won, among other
titles, the "Ann Arbor Men's Singles: Over 40"
title in 1978, and continued to play even after sur-
gery replaced his right hip joint in 1981.
Survivors include his wife, Merrilyn (Terry) ;
two daughters, Judy Lee Martin and Jacque
Martin Downs; two sons, Joseph J. B. Martin and
Jon T. F. Martin; two sisters, and two grand-
children. E


V i
aSi ^ss^
r.jpp >"ai(y




via Horizontal Integration of Subject Matter*

736 West End Ave.
New York, NY 10025
The Cooper Union
New York, NY 10003

DURING THE PAST DECADE, many universities
have been critically reappraising the effective-
ness of some of their traditional elective courses in
view of the rapidly changing needs of professional
education in a wide variety of fields including sci-
ence, engineering, business administration, jour-
nalism, law and medicine. No longer is it possible
for professionals in any of these fields to function
in a comfortable milieu designed and limited
largely by its practitioners. A number of signifi-
cant trends are discernible in the interaction
among the professions, society, and higher educa-
tion. Irrespective of the particular professional
field, the same questions keep recurring. Promi-
nent among these are the sensitivity of the pro-
fessions to social needs, the concept of profession-
alism and professional competence, the problem of
licensure of professionals and of the maintenance
and upgrading of competence, the opening of
licensing boards to lay people, the emphasis on
ethics, values, motivations and the need for a
broader, more humanistic and humane view on the
part of decision makers and their advisors. As a
result, an urgent call is being heard from many
sectors for a "new breed" of professional whose
training and values reach beyond the cost-
efficiency considerations to include an assessment
of the political, social, and human dimensions of
the problem at hand.

*Paper presented at Annual Meeting of the AIChE,
Chicago (November 18, 1980).

Copyright ChE Division, ASEE, 1983

Charles Huckaba heads his own consulting firm in New York. He
has a varied background in engineering, medicine and applied mathe-
matics involving academic/industrial/government affiliations. The work
described in this paper was carried out during his tenure as Director of
Engineering Program Development at the Cooper Union. Dr. Huckaba
received academic training in chemical engineering at Vanderbilt, M.I.T.
and the University of Cincinnati and is a Fellow in the American In-
stitute of Chemical Engineers. He is also a member of ASEE.
Anne Griffin is Assistant Professor of Political Science at The Cooper
Union. A member of the Humanities faculty, she has taught courses on
public policy, technology and society, and urban politics, and has
played an active role in the development of curricula for the School of
Engineering. Dr. Griffin has also participated in projects dealing with
the interfaces between technology and politics. She holds a BA from
Wellesley College and an MA and PhD from New York University. (Not

Over the past quarter century, engineering
education has been evolving toward a more "lib-
eral" format, with an increasing emphasis upon
the humanities and social sciences to complement
the technical content of the curriculum. In 1951,
the Engineers Council for Professional Develop-
ment mandated that to meet minimum accredita-
tion requirements, all undergraduate engineering
curricula must contain 20%o non-technical (hu-
manities, social sciences) courses. Even though
this move constituted a significant break away
from the almost total preoccupation with the de-


velopment of technical expertise, the effect (al-
though salutary) has been of limited impact. Ex-
amination of the transcripts of recent engineering
graduates reveals little perceptible focus on, or
coordination among, the non-technical courses
elected; one strongly suspects that such criteria as
convenience of scheduling, word-of-mouth reports
on amount of work required, etc., were the primary
factors influencing the choice of those courses.
Whereas the resulting potpourri, no doubt, induced
some alternate viewpoints for the students, the re-
sult in most cases falls far short of the intended
Dr. Simon Ramo*, an engineer who was for-
merly an advisor to the White House, recently

Engineers must spend as much time learning about and
dealing with society as they do in applying science and
technology to society and its problems. The present
veneer of humanities and social sciences in university
engineering curricula is quite inadequate for this pur-

*Ramo, Simon. 75th Anniversary Convocation, National
Bureau of Standards (1976).

... an urgent call is being heard
from many sectors for a "new breed" of
professional whose training and values reach beyond
the cost-efficiency considerations to include
an assessment of the political, social,
and human dimensions of the
problem at hand.

Certainly we can agree that the "humanizing"
of professional education requires more than
merely encouraging students to read a few more
humanistic texts in the hope that a stronger aware-
ness of the human implications of their activities
will occur by magic contagion.


In an attempt to address this problem, in 1976
a group of educators from various disciplines at
the Cooper Union (originally under the initiative
of the Engineering School) launched a curricular
reform study with the firm intent to compose a
genuinely interdisciplinary (not just mutidiscipli-
nary) approach to engineering education. One of
the significant results of this study was a proposal

Social Aspects of the Technical Decision Process

Technological advances are increasingly shaping con-
temporary society and culture. Professionals in many fields
recognize that it is no longer possible to function in tradi-
tional contexts. This course will examine the social, ethical,
and humanistic dimensions of currently critical problem
areas, especially natural resource limitations, energy al-
ternatives, and environmental issues, wherein social and
human impact are equal in importance to technical/eco-
nomic criteria. Course format will encourage effective
horizontal integration of guiding concepts from the human-
ities and technical disciplines.
Science, engineering and technology.
Their place in the spectrum of human knowledge.
Interrelationships among them.
Case studies in engineering ethics. (See Table 2)
The current dialogue between technology and social
Independent projects involving social and technical
aspects. (See Table 3)
Setting standards: Values, valuation and applications.
Summary and conclusions.
Baum, Robert J. and Albert Flores, Ethical Problems
in Engineering. The Center for Human Dimensions,
Rensselaer Polytechnic Institute, Troy, NY (1978).

Florman, Samuel C., The Existential Pleasures of En-
gineering, St. Martins, NY (1976).
Kuhn, Thomas S., The Structure of Scientific Revolu-
tions, 2nd ed., University of Chicago Press, Chicago

Dubos, Ren6, So Human an Animal, Scribner, NY
(1968). pp. 181-242.
Ellul, Jacques, The Technological Society, Knopf, NY
(1964). pp 1-22; 65-79.
Hellman, Hal, Technophobia: Getting Out of the Tech-
nology Trap, M. Evans Co., NY (1961). pp 27-64;
Mumford, Lewis, Technics and Civilization, Harcourt,
Brace, NY (1934), pp 321-363.
Mumford, Lewis, The Myth of the Machine, Vol. II-The
Pentagon of Power, Harcourt, Brace, Jovanovitch,
NY (1970). pp 281-293.
Reich, Charles A., The Greening of America, Random
House, NY (1970). pp 6-9; 87-128.
Roszak, Theodore, Person/Planet-The Creative Dis-
integration of Industrial Society, Anchor Press/
Doubleday (1979). pp xix-xxx; 27-39; 132-137; 306-
Simons, J. H., A Structure of Science, Philosophical Li-
brary, New York (1966). pp 1-64.


to effect a horizonal integration (topical and
methodological) of humanities/social sciences sub-
ject matter with the engineering/design content of
the curriculum. This would be accomplished in a
team teaching format, involving both humanities
and engineering faculty members concerned with
the human and social implications of supposedly
neutral technologies and methodologies.

The initial attempt to implement this concept
involved a course entitled "Social Aspects of the
Technical Decision Process," which was team
taught by the authors.
Course content was built around an integrated
set of topics and related specific problem formula-
tions cutting across disciplinary boundaries, such
as alternative energy, ecology, urban planning,
etc., which represent traditional engineering con-
cerns but have built-in social and historical dimen-
sions. A guiding question was whether the latter
are mere additional concerns which one may or
may not deal with depending upon one's sense of
involvement, or whether such concerns might not
affect the very problem formulations themselves.
Table 1 indicates the course content. The cen-
tral concern of the "horizontally integrated" top-
ical core, as exemplified in this initial course, was
the relation between professional competence and
(supposedly extraneous) ethics and social impact,

Ethics Cases Considered from Baum and Flores

A case was chosen by each student for oral presentation
and critique to the class. The cases considered in this man-
ner were:
Engineering Ethics: A Blend of the Ideal and the
Practical (p 30)
Ethics, Engineering, and Publicity (p 91)
False Statements in Advertising (p 102)
Conflict of Interest-Related Work for Two Parties
(p 115)
Liability and the Engineer-Responsibility to Former
Employers (p 145)
Misuse of Confidential Information (p 152)
An Anatomy of Whistle Blowing (p 168)
The Case of the Three Engineers vs. BART (p 227)
Carl W. Houston and Stone and Webster (p 262)
The Life and Times of Lawrence Tate (p 288)
A Research Pirate in Action: The Case of Dr. Aries
(p 290)
Old Secrets in a New Job (p 292)
Reactor Safety: Independence of Rasmussen Study
Doubted (p 301)

Term Papers
Each student in consultation with the course instructors
prepared a report on a contemporary topic involving both
technical and social impact factors. Each topic was first
presented orally to the class to elicit suggestions and crit-
icisms and then a written report was prepared. Topics of
the papers were:
Recombinant DNA-Technology and Social Aspects
Dilemmas of the Technological Polity
Being There (Jerzy Kosinsky)-The Plight of the
Self-Made Man
The Supersonic Transport-Should it Continue?
Space Colonization
Fluoridation-Pros and Cons
Television and Society
Nuclear Wastes
Technology vs. the Environment, A Case Study: In-
Sperm Bank/Artificial Insemination (AID) Technol-
ogy: The Technique and its Social Implications
Government Interventions in the Business of IT&T

including an examination of the basic concept of
professionalism itself. To prevent these considera-
tions from appearing to be nothing more than
academic moralizing, this material was couched in
the context of presently critical technological prob-
lem areas such as those listed in Tables 2 and 3.
For example, in the initial layout of a new
processing plant or in the preliminary design of
an electrical power distribution system, students
can be led to recognize the necessity of considering,
in the early stages of planning, the impact upon
both the social and natural environments in addi-
tion to the usual technical and economic criteria.
Failure to do so may lead not only to costly delays
or expensive modifications of the operating system
but above all, to unacceptable human consequences.
In considering a series of examples of this type,
the students begin to recognize that it is expedi-
tious to consider the technical-social system as the
unified entity which it really is, rather than arti-
ficially subdividing it into components which, in
the final design, may not mesh in a compatible
manner. In short, the students are led to realize
that the understanding of social/political phenom-
ena is as necessary as the acquiring of technical

The work reported in this paper was supported
by a grant from the National Endowment for the
Humanities. O


Book reviews

By W. Brock Neely
Marcel Decker, Inc., New York, 1980. 245 Pages
Reviewed by Alfred J. Engel
Pennsylvania State University
The author of this book, Dr. W. Brock Neely,
is a biochemist by training and now works in the
Environmental Research Laboratory of the Dow
Chemical Company. It seems that the book is the
indirect result of his self-education in the area of
mathematical modeling of environmental phe-
nomena. Six of the book's seven chapters, as well
as most of the appendix, deal with a variety of
models, from the compartmental approach to eco-
systems, to dispersion methods in air and water
pollution. Unfortunately, none of this material is
of sufficient depth or comprehensiveness to allow
the untrained reader to gain much expertise in
the use of the models. On the other hand, the ex-
perienced reader will find much of the material
well worn and quite inadequate for further use.
Only the extensive bibliographies at the ends of
the chapters may prove of real value.
One of the stated aims of the book is to make
members ". . of the scientific community ....
more adept at predicting what will occur in the
environment as a result of some planned activity."
Such activities, of course, are principally the re-
lease of chemicals into the eco-system. Although
the book does an adequate job of describing
various models for making such predictions in
qualitative terms, it hardly lives up to its goals of
making us more adept. We gain appreciation,
rather than expertise.
On the positive side, the book presents
throughout a much needed industrial view of en-
vironmental regulations and public policy regard-
ing environmental impacts of industrial activity.
A fair and balanced account of the PCB problem
is presented, and the atmospheric flurocarbon con-
troversy is discussed in detail and then related
somewhat sketchily to possible models. On oc-
casion the author becomes a bit strident, but no
more so than most of us, in dealing with regula-
tory red tape.
Finally, the book is printed by photo-offset
from a typewritten manuscript. Although the

type is clear and easy to read, and the illustra-
tions and tables quite comprehensible, the proof
reading and editing leave much to be desired.
There are numerous typographical and spelling
errors which proved to be very annoying.
This is not a book for the average chemical
engineer. For those with training in environmental
matters, it is mostly too elementary to bother
with; for those with, little background in this
area, it may be a useful, though greatly simpli-
fied, introduction. Those willing to follow up on the
bibliography may well learn a great deal. ]

By Mason Benedict, Thomas Pigford, Hans Levi
McGraw-Hill Book Company
Reviewed by
Herbert S. Isbin
University of Minnesota
Impressive in scope, details, and thoroughness!
This text maintains a high concentration of re-
warding material per page for approximately
1000 pages. Not surprising when one considers
that three internationally recognized authorities
have pooled their expertise into a skillful account-
ing of theory and practice for the nuclear fuel
cycle. Manson Benedict (Professor Emeritus,
MIT), a most distinguished elder in nuclear chem-
ical engineering, has long been recognized for his
abilities to focus on significant features and prob-
lems and to write in a remarkably clear and stimu-
lating manner. Coauthors are Thomas H. Pigford
(Professor of Nuclear Engineering at the Uni-
versity of California, Berkeley) and Hans Wolf-
gang Levi (Director of the Hahn-Meitner-Insti-
tute fur Kernforschung in Berlin and APL-Pro-
fessor of Nuclear Chemistry, Technische Univer-
sitat Berlin). Further, the contributions of many
professional colleagues in the United States and
in Germany are acknowledged, reinforcing the
prestigious technical input for this text.
The first of the fourteen chapters designates
the chemical engineering needed to sustain the
nuclear fuel cycle for the fission power reactors.
Even though the purpose of the first chapter is to
establish an overall perspective, quantitative de-
tails are provided in the flow sheets. The next two
chapters, with emphasis on nuclear reactions and
specifics of the fuel cycle, develop degrees of so-
phistication seldom achieved in other complete
Continued on page 85.


P p international



University of Waterloo
Waterloo, Ontario, Canada

TOWARDS THE MIDDLE OF March about two years
ago, as night was beginning to fall, I stepped
from the pavillion of the spanking new guest house
at the University of Lagos and wandered in the
warm, fragrant evening air some 50 yards down a
dirt road that led to the lagoon. The horizon was
still an azure blue and silhouetted the palms that
bordered this broad body of water. The rising
moon dappled the undulating water; the lights of
Lagos flickered off to the north. Ripples lapped
gently at the shore. What a difference from Can-
ada where I had been just a few days before.
Toronto was then in the midst of a snow storm and
my colleagues probably were still digging out. It
wasn't to be the last snow storm of the year either.
Here I was in Nigeria, West Africa's most ex-
citing and exuberant country, to do my bit for the
Third World. I was to initiate a program of as-
sistance that had been agreed upon between my de-
partment in Canada and the department of chem-
ical engineering at the University of Lagos. Lagos
had that year begun a graduate program in chem-
ical engineering, even though its undergraduate
program had been underway for just four or five
years. The department needed staffing help for this
venture, and they also needed some assistance in
organizing their process design course. The plan
was to have a different Waterloo faculty member
spend a 4 month term there each year, offering at
least one graduate course and assisting with the
organization of research projects and the super-
vision of graduate students. Arrangements of this

The subcommittee on Assistance
to Chemical Engineering Education in the
Third World is looking for members and welcomes
those who will offer an active involvement in
projects and project initiation ...

Copyright ChE Division, ASEE, 1983

Peter Silveston, professor of chemical engineering at the University
of Waterloo for 18 years, is a 1951 graduate of M.I.T. He studied en-
gineering and hitch hiking in Germany during the 50's before choosing
the former as a career and the latter as an avocation, but in a guise
politely referred to as "academic travel". Thus, Professor Silveston has
accepted assignments in Nigeria, France, Germany, South Africa, Eng-
land and the United States. His research interests are primarily in the
field of reactor engineering, while his recreational passion is the
Canadian wilderness wherever it can be found in Southern Ontario.

type, though often supported by government
funds, have been made between other schools for
many years and are not at all unusual. Most par-
ticipants find assignments in the Third World sat-
isfying and enjoyable. Certainly, I found my four
months memorable and intellectually stimulating.
Perhaps I contributed in a small way to the Lagos
Notwithstanding such satisfaction, I was dis-
mayed at the cost and bothered by the suspicion
that there are better ways of assisting education
than bringing in a Westerner for a term, or even
two, of teaching. Although I've brought my sug-
gestions for improving the assistance program to
the attention to my colleagues, it seems to me that
some sharing of the experience would be useful on
an inter-institutional basis. Perhaps there are pro-
grams the AIChE could develop, promote or spon-
sor. Unquestionably, helping chemical engineering
grow in the Third World is a responsibility we
should take on. The question is how to help most
efficiently. These questions are behind the organi-
zation of the new subcommittee on Third World
Chemical Engineering Education undertaken this
Continued on page 82.




University of Lagos
Lagos, Nigeria

eeWE HAVE REACHED THE moon, but we have not
yet reached each other," U Thant, former
Secretary General of the United Nations, remarked
at the 25th anniversary of the United Nations. I
would like to believe that the AIChE's Chemical
Engineering Education Projects Subcommittee on
Third World ChE Education would like to help us
reach out to each other. However, I note that all
the members of the subcommittee are from North
America. This is hardly a good beginning. If
there is to be cooperation rather than charity,
there should certainly be members from the Third
World on the subcommittee.
Cooperation in education between the developed
and still-developing sections of the world is, I be-


Hussain K. Abdul-Kareem graduated B.Sc. (1973) Chemical Engineer-
ing at the University of Ife (Ile-Ife, Nigeria). After graduation, he was
posted to the Bendel Textile Mills (Asaba, Nigeria) as a Chemical En-
gineer/Quality Controller for a year for his National Youth Service, at
the end of which he worked, briefly, with the former I.C.I. (Nigeria)
Limited as a Chemical Sales Representative. He received his M.A.Sc.
(1975) and Ph.D. (1978) at the University of Waterloo and returned to
Nigeria as a Lecturer at the department of chemical engineering, Uni-
versity of Lagos. His research interests include forced cycling of
chemical processes and the cleaning of coal. He is currently in charge
of industrial training of chemical engineering students at Lagos. A
member of a number of professional associations, he is also a member
of the University of Lagos Senate.

My purpose in this paper
is to bring to the attention of chemical
engineering educators the problems that Third
World engineering teaching institutions face and also
to make some suggestions
toward solutions.

lieve, worthwhile. After all, two-thirds of the
global population live in the Third World. If we
are to achieve a secure and peaceful world, some-
thing must be done to raise the living standard
toward that now enjoyed by the developed nations.
The countries of the industrialized world have
the resources to help raise living standards. They
control about 88% of the world's wealth. Direct
aid is needed, but cooperation is also very, very
important. I believe cooperation is particularly im-
portant in education. The industrialized world will
certainly find that the Third World has one or two
things to teach them. Furthermore C. R. Fay
points out, in his book entitled "Co-operation At
Home And Abroad," that cooperation is to charity
(aid) as prevention is to cure. I believe this state-
ment is quite appropriate to engineering education
in the Third World.
My purpose in this paper is to bring to the at-
tention of chemical engineering educators the
problems that Third World engineering teaching
institutions face and also to make some suggestions
toward solutions.
One would think from reading the popular
press in the U.S. or in Canada that Third World
problems are only debated here. Nothing could be
further from the truth. We of the Third World
have not been idle, nor are we idle now. We are
striving hard to improve our lot. My country as
well as many others have enormously expanded our
educational resources since independence. Unfor-
tunately, some of this expansion has been mis-
directed. This misdirection in our educational pol-
icy has resulted in millions of unemployed educated
people and, at the same time, we are crying "shor-

Copyright ChE Division, ASEE, 1983


tage of manpower!" This definitely points to one
of the problems. There are various ways by which
this misdirection of education can be corrected.
Policies have to be reformed in such a way, as the
World Bank puts it, that they may better contrib-
ute towards economic development. This means
education must emphasize skills and technical
Chemical engineering has been introduced as a
program in many of the universities of Third
World countries now pursuing industrialization.
Nigeria now has 5 departments (Lagos, Ife, Port
Harcourt, Benin, Ahmadu Bello) and there are
jobs for all of our graduates. The discipline, there-
fore, is playing a role in our development.
Let me talk about some of the problems we face.
Although my experience is just with my own coun-
try, I believe it is also typical of other universities
in developing countries. Some of the problems that
face chemical engineering departments are:

shortage of faculty
inadequacy or lack of support services
shortage of teaching and research equipment
inadequacy of research funding
lack of administrative experience
negative attitude towards work
isolation from centers of technical activities

The problems are not necessarily listed in order of
The staffing problem is not limited to the Third
World; there are other challenging and better-
paying opportunities open to qualified chemical
engineers in the private sector everywhere. How-
ever, the problem is more acute in the developing
nations. In my own position I could certainly
double my salary by taking an industrial engineer-
ing job. Furthermore, benefits outside of salary
are much better in industry. Because of shortages,
benefits are more important in developing nations
than in the West. This situation makes it very dif-
ficult for universities to recruit faculty, although
once employed their turnover is no worse than else-
where in the world.
One solution to this recruitment problem is for
North American and European institutions of
higher learning to increase the intake of Third
World students into graduate programs. The trend
among Western countries to increase fees for
foreign students should be discouraged. The in-
crease in revenue for the university is very small
and is hardly worth the damage it does.
There is a problem with students' staying on in
their host countries. This can be handled by regu-

lations which discourage foreign students from
taking up academic appointments on completion of
their programs. The staffing problem can never be
solved by arrangements whereby faculty members
from universities in industrialized countries are
seconded or go on sabbatical to departments in the
Third World. For purposes of teaching, this is an
expensive exercise and the money would be much
better spent on support services. For research,
guest faculty do not serve any useful purpose be-
cause they lack basic knowledge of local problems
and working in the local environment. I believe
chemical engineering research in the Third World
should be geared to solving local industrial prob-
lems rather than for the purpose of scholarly pub-
lications. We cannot afford the luxury of scholar-
ship. Thus, the visiting professor who does not
know the industries in the country is not in a posi-
tion to provide suitable research problems or to
supervise graduate students.
The problem of lack of ChE staff is being
tackled by some institutions in the Third World
that have started their own graduate program.
This effort is, however, being thwarted by some or
all of the other enumerated problems. We can all
imagine a graduate program in chemical engineer-
ing without proper support services, inadequate or
obsolete equipment, a shortage of research funds
and the ancilliary problems. Such a program will
be plagued with problems.
Sabbatical arrangements can help, but they
should be in the opposite direction until our chem-
ical engineering departments are further devel-
oped. Sabbaticals in the West serve a dual purpose.
The visiting faculty from the Third World can
bring himself up-to-date in his field of chemical
engineering, he will have the opportunity to use
equipment that is not available in his home uni-
versity, and he will have the valuable opportunity
of exchanging ideas with faculty members of his
host institution. Access to the literature provided
by good libraries and the computer facilities are
also some of the advantages that he will enjoy.
During his stay, lecture notes can be brought up-
to-date and he can help his colleagues at home with
reprints or xerox copies, supplies that are often so
time-consuming to get in Africa or Asia.
The problem of inadequacy or lack of support
services is as serious as the previous one. Tech-
nicians capable of fabricating research or teaching
equipment are not available. Universities do not
recognize these skills and so we cannot compete
with industry for good people. Even if technicians


are available, there is the all-too-usual lack of an
adequate machine shop, or spare parts cannot be
obtained. A damaged piece of equipment is better
discarded. It is easier and often cheaper to replace
it because of lack of spare parts or the technical
know-how required to fix it. The problem of spare
parts is very acute. This is an area, I believe, where
the expertise of European and North American
chemical engineering departments can help a great
deal. The right spare parts to be stocked for spe-
cific pieces of equipment could be recommended by
your technicians and storekeepers. If we know
what we need, provision can be made when equip-
ment is purchased. A solution to the above prob-
lems which I strongly recommend is to set up
sabbaticals or apprenticeships for our technicians
and storekeepers in your chemical engineering de-
partments. Exchanging technicians would also be
useful, but it would suffer from the same problems
I have observed with visiting professors from the
The problem of inadequate equipment is really
a problem of funding. I will, therefore, discuss
these two problems together.
Most of the chemical engineering departments
in the developing world are inadequately equipped
either for teaching or for research. This is a result
of both insufficient grants to the universities at
large and the attitudes of the existing industry in
the country. One would expect developing nations
to encourage their institutions of higher learning
to expand for the purpose of producing the trained
manpower for nation-building by adequately fund-
ing these institutions. But this is not the case for
reasons which are, at times, beyond the control of
government itself, or simply due to the gross mis-
placement of priorities. Government is not the only
problem. A rather saddening observation of mine
and many others in both industrialized and indus-
trializing nations is the unpatriotic attitudes of
many foreign-owned industries operating in the
country. Despite the huge profits they declare-
never mind the undeclared ones-and the foreign
earnings they make for their home countries, these
foreign-owned companies never deem it fit to co-
operate with the educational or research institu-
tions in the nations where they garner their prof-
its. I strongly believe that the Chemical Engineer-
ing Education Projects Committee can and, in-
deed, should encourage companies with overseas
branches to fund research in those Third World
nations where they operate. Actually, I do not
think it will be out of place if our governments

made it mandatory for all companies operating in
our countries to fund research with a fixed per-
centage of their annual profit. It is equally sadden-
ing to see manufacturers of research equipment
selling outdated and or dilapidated equipment to
research institutions in the Third World. At times
we feel we are used as guinea pigs with untested
As the word "developing" implies we have not
yet arrived, so our administrative experience and
our attitudes towards work are not quite what they
ought to be. The universities and other advanced
teaching institutions, as citadels of learning, have
a sacred duty in this regard. We should be setting

Chemical engineering building, showing the high head
room laboratory and faculty office wing, University of

the standard in providing good administration.
Lack of administrative skill is acute at the tech-
nician/storekeeper level and training in North
America could really help. Even a better attitude
towards work could be fostered. Officers in the uni-
versity service departments such as grounds, fi-
nance and bookkeeping need training, but this is a
university job and not one for chemical engineer-
ing departments.
The development of one of the very important
support services-the library-can help solve our
problem of isolation from the centers of technical
activities of the world. Our libraries are poorly
stocked with the literature relevant to research
and teaching and they do not receive their peri-
odicals and journals on time due to poor communi-
cation between countries. Most universities do not
have sufficient funds to buy all the journals or
books which are needed. Perhaps the AIChE
through the ChE Education Projects Committee


could raise funds to provide subscriptions to chem-
ical engineering departments of developing na-
tions. Could the AIChE donate their publications
package to Third World departments? Third
World participation in the ChEEP. would also help
us stay in touch.
The location of most international chemical
engineering conferences, seminars, and the like,
are too far away from the Third World countries.
The exorbitant cost of transportation makes at-
tendance from the Third World virtually impossi-
ble. The importance of these meetings to the de-
velopment of engineering education cannot be
overemphasized. They provide, among other bene-
fits, opportunities for frequent exchange of ideas
and ensure up-to-date knowledge of current de-
velopments in the technical field. There are, and
can only be, very few such meetings in the Third
World because of the newness of the discipline and
the small number of institutions offering chemical
engineering as a course. I believe the AIChE,
through the ChE Education Projects Committee,
should consider arranging seminars at least an-
nually in a developing nation with the cooperation
of Third World chemical engineering departments
and perhaps UNESCO. Seminars could help solve
the problem of isolation. I am sure that Canadian
and U.S. professors would welcome a chance to
spend a week in Nigeria at a seminar instead of
going to New Jersey or Ohio.
In conclusion I will mention a "reaching out"
exercise that I am participating in. Canada, under
the auspices of the Natural Science and Engineer-
ing Research Council and the Canadian Inter-
national Development Agency (NSERC/CIDA),
has established a fellowship program whereby
faculty members from developing nations are
given the opportunity to work with faculty mem-
bers in Canadian universities on a joint research
project. These fellowships are valid for a period
of three months for three consecutive summers.
The United States also has such a scheme, known
as the Fullbright Award program. However, the
U.S. program is for a single period of four months
I have found participation in the Canadian pro-
gram very rewarding. It has provided an oppor-
tunity to discuss curriculum and textbooks with
my Canadian colleagues. I have also had the op-
portunity to review new books that have not yet
reached Nigeria and to order supplies that would
ordinarily take many months to reach us. Travel
funds are available so that other schools can be

visited and so that I can attend at least one tech-
nical meeting. The program also provides the op-
portunity to see what research is underway in the
host university and to discuss my research projects
with my hosts. Unfortunately, three months is just
too short a time to accomplish much on a research
project, even with equipment available and operat-
ing. This is true even though the Canadian pro-
gram usually sends the faculty member to the
school which he studied at, in order to avoid the
orientation period.
The problems with these two programs are the
interruptions in the Canadian fellowship and the
shortness of the U.S. fellowship. I believe an ar-
rangement whereby the recipient of the award and
his family spend a whole academic year in the host
country would better serve the purpose of these
programs. The programs are in the right direction,
however. Also in the right direction is a new co-
operative program established by the International
Development Research Center (IDRC) in Ottawa,
Canada. This program funds research in Canadian
universities on problems relative to development
in Third World countries. Researchers can be
Canadians or a team with membership from the
target developing country. Time will tell how suc-
cessful the program will be. If administered well,
it could make a real contribution. The Canadians
have set the ball rolling and we can and will reach
each other. E

Continued from page 78.
year by the AIChE's Chemical Engineering Edu-
cation Projects Commitee.
What are projects which could be undertaken
in place of, or to supplement visits of chemical en-
gineering educators? Let me begin discussion of
this question by outlining some of my observations.
The experience of the Lagos faculty lies heavily in
the areas of chemical reactor analysis, catalysis,
and kinetics; it's not surprising that many of the
initial projects selected for graduate study reflect
this experience. Modern work involves reasonably
sophisticated analytical equipment. Often the pro-
ductivity of a research student can be enhanced
enormously by semi-automated reactors. Nigeria's
oil wealth and the heavy emphasis the government
has put on education means money is available for
the purchase of quite sophisticated equipment.
During my short stay I had the frustrating experi-
ence of trying to help faculty and technicians set
up and shake down some of the units that had ar-


Engineering faculty buildings, University of Lagos

rived. Certainly there is no area of greater frustra-
tion for a Westerner in the Third World than try-
ing to build and operate experimental equipment.
Regular failures of electric power, water shut-
downs, and air conditioner failures coupled with
high humidity and dust are often second order
problems. In Lagos, primitive port facilities, inept
customs handling and road conditions mean that
the equipment arrives in the laboratory needing
immediate repair. There are, however, no main-
tenance facilities or parts depots in West Africa.
Not only must everything be shipped in from
Europe or North America, but clearance for ex-
penditures in foreign currency must be obtained
even for the most meager spare part. Parts cannot
be fabricated locally because shop facilities are in-
adequate and a good part of the technicians' day
must be spent ferreting out parts or supplies.
When something is found it goes into a file cabinet
and its existence is treated as a state secret.
The crying need in Lagos (and I suspect in
many new Third World chemical engineering de-
partments) is for reliable support services: a
storeroom stocked with basic parts, spares for key
components of departmental equipment, up-to-date
stock lists, a continuing budget for replacements,
and above all a well trained storekeeper. Lagos has
a basic shop adequate for fabricating supporting
structures, but it is inadequate for fine machining.
No electronics shop exists and virtually no equip-
ment is available for circuit testing and electronics
trouble shooting. Lagos has a good complement of
technicians, but they seem to interface poorly with
research students and faculty. One-on-one super-
vision was necessary. The Lagos department is
well supplied with office help, messengers and driv-

ers, but management is poor and supervision of
simple matters like note typing and duplicating
falls upon faculty. Assistance is desperately needed
for the training of the support staff. Probably
some sort of apprenticeship in our institutions
would be better than sending Western technicians
to train staff in Third World institutions. At
Waterloo we are searching for funds to bring
storekeepers and technicians from Lagos to Can-
ada for, say, a two to six month period to work as
assistants to their counterparts here.
Nigerian universities, concerned about estab-
lishing scholarly standards comparable to those in
the developed world, have promulgated advance-
ment standards in terms of research publications
that are laughably unrealistic in terms of the dif-
ficulties of initiating and maintaining research.
Promotion seems slow. This is unfortunate because
there are lucrative opportunities in the private
sector to tempt faculty. Indeed, faculty turnover is
One way of surmounting the problem of main-
taining research is to invite junior faculty mem-
bers to work in research groups in North Amer-
ican departments. Canada has already initiated an
effort in this direction: graduates of Canadian
institutions in Third World teaching positions are
given three month fellowships for each of three
years, which permit them to return to Canada
(usually their home institution) for further re-
The teaching of process design is enhanced by
using the technical literature of equipment manu-
facturers. For some reason, literature of this type
is virtually unattainable in the Third World. We
are tackling this problem at Waterloo by ordering
extra literature, beyond the number needed for our
own program, and providing it at cost to Lagos.
Are there alternatives to bringing faculty from
developed countries? In many areas, visiting fac-
ulty are high priced because of salary, or the size
of the family which must be transported and
housed, or because of living style expectations.
Some courses (process design, for example) should
have some relevance to the host country. It should
reflect the level of current technology or at least
the near technological horizon. Thus, the instruc-
tor must have a reasonably good knowledge of the
major industries of the country, their technologies,
and projects planned for the near future. It is rare
for visiting faculty to have such knowledge. Ad-
vising research students in Third World institu-
tions has its pitfalls also. Lack of simple parts,


equipment maintenance problems, and limited ex-
perience of technicians puts different constraints
on what experimental programs can be undertaken
than those met in North America or Europe. It is
difficult for visiting faculty to know what is a
reasonable problem for a student under such con-
A better alternative might be to bring junior
faculty to North American institutions and place
them as "interns" in team teaching situations or in
team research supervision. Lagos could have sent
two of their staff (with families) to Waterloo for
four months for the same cost to them of my visit
for the same length of time.
The number of North Americans who have
taken part in educational programs in the Third
World is surprising. These have either been sab-
batical leaves spent where the host country bears
all or part of the cost, or assignments undertaken

Proposals for Assisting ChE Education in the Third World
1. Apprenticeships for storekeepers, technicians in North
American ChE departments or research institutes
Objective: improve support services for teaching and
2. Sharing of company reference material used for process
design courses
Objective: enrich process design teaching
3. Summer research fellowships in North America for
junior faculty from Third World institutions
Objective: assist faculty development
4. Development of a final year/postgraduate course on de-
velopment problems, appropriate technologies, develop-
ment strategies for the Third World*
Objective: mitigate ignorance of our graduates of
the problems of this important part of
the world
5. Development of a co-op engineering program with multi-
national corporations operating in the Third World for
Third World engineering students*
Objective: improved engineering education, man-
power supply for multinationals
6. Preparation of course material on construction and op-
erating costs in Third World countries*
Objective: improve teaching materials for engineer-
ing economics courses, supply data for
North American organizations
7. Teaching internships for Third World ChE faculty in
North American departments
Objective: course development, teaching skill devel-

*Proposals originated by Professor W. H. Tucker, Tri-
State College, Indiana.

on behalf of United Nations agencies, or those
established through aid programs on intergovern-
mental levels. If you have not taken an overseas
assignment, its quite likely that one of your col-
leagues has. Certainly, a great deal of experience
has been collectively accumulated by North Amer-
ican faculty. The new subcommittee on Assistance
to Chemical Engineering Education in the Third
World hopes to tap this experience and, further, is
looking for your ideas. The subcommittee would
like to know about current as well as past pro-
grams. It is interested in those which were success-
ful and those which did not seem to work out.
Areas where the AIChE could help, of course, will
be sought. The subcommittee will also be looking
for programs that individual departments could
undertake, or even those which solicit contribu-
tions from individual faculty members.
A survey of North American faculty is planned
to identify and summarize the experience available
and to solicit ideas. Some proposals have already
come in. They are summarized in Table 1 along
with those just mentioned. It is hoped that the sub-
committee will eventually undertake projects. Per-
haps some will lead to AIChE activities; others
may stimulate action by our teaching institutions.
For the moment, the subcommittee plans to pursue
study, discussion of problems, and proposals, with
results emerging as reports.
The subcommittee has begun with a few en-
thusiasts-but in order to make a useful contribu-
tion, wider participation is needed. The subcom-
mittee on Assistance to Chemical Engineering
Education in the Third World is looking for mem-
bers and welcomes those who will offer an active
involvement in projects and project initiation, as
well as those whose role will be more passive,
through sharing of their experience and construc-
tive advice. O

Author's Note: The results of the ChE Education Sub-
committee survey begun in 1981 are now in, showing that
10% of the chemical engineering teaching staff in the U.S.
and Canada has participated in teaching research or train-
ing programs in Third World countries.
Survey results show that the proposal to assist develop-
ment of support staffs by providing apprentice type train-
ing to technicians, storekeepers, etc., for periods of a few
months to a year elicited 77 responses; 27 departments
would participate in such a program, 23 felt they did not
have suitable facilities or personnel to do so, and the re-
mainder wanted further details. Only one department
would contribute financially to the apprenticeship, and
forty indicated funding would have to come entirely from
external funds.
The second proposal was to invite Third World teaching


faculty to take sabbaticals in North America. 78% of the
respondents said they would participate in such a program,
but 45% of them would want the program to be funded
The third proposal, to invite PhD graduates with Third
World degrees to do one or more postdoctoral years in
North America, was less enthusiastically received. 57% of
the responses indicated participation and 17% would not
be interested. 27 departments would like to see separate
external funding and an equal number felt that some sup-
port could come from ongoing contracts or programs in
their departments.
The Third World Subcommittee is now chaired by Prof.
C. J. Barr, California State Polytechnic University,
Pomona. He would welcome your suggestions and inquiries.

Continued from page 77.
Chapter 4 lays the foundation for solvent ex-
traction and systems are selected for illustrating
principles and practice. Included is a brief account-
ing of the types of commercial equipment.
Uranium, thorium and zirconium have sperate
chapters devoted to the physical and chemical
properties of the element and its compounds and
the associated processing.
Chapters 8, 9 and 10 deal with the processing
of irradiated fuel and materials. The identifica-
tion of the radionuclides of interest and the time-
dependent behavior in and out of the reactor are
presented and illustrated in Chapter 8. Detailed
information on plutonium and other actinide ele-
ments (proactinium, neptunium, americium and
curium) is given in Chapter 9, and Chapter 10 is
devoted to fuel reprocessing. Both U.S. and over-
seas reprocessing plants are noted. The lack of an
operating commercial reprocessing plant in the
United States has perhaps placed an undue em-
phasis on the design calculations for the Barnwell
Nuclear Fuel Plant.
Rightfully, radioactive waste management is
singled out for Chapter 11. The focus, however, is
primarily on high-level wastes. A "most recent
estimate of the amount of solidified high-level
wastes to be accumulated at a federal repository
in the United States..." turns out to be a 1976
reference with an over-optimistic projection of
nuclear power and the achievement of the reposi-
tory. The authors have recognized the uncertain-
ties in the projections and sought to place in per-
spective the relative magnitudes of the volumes
and curies of the high-level wastes with respect to
the natural radioactivity in the earth's crust.
Nevertheless, this outdated reference serves to

illustrate the practical difficulties in updating the
material for this work.
Isotopic separations are the topics for the last
three chapters. Methods and principles are given
for stable isotopes in Chapter 12. Chapter 13 pre-
sents processes for the separation of isotopes of
hydrogen and other light elements. The final chap-
ter is on the separation of uranium isotopes.
In my judgment, the authors have provided a
wealth of technical information on the nuclear fuel
cycle and have been successful in highlighting im-
portant principles and in providing meaningful
illustrations. Each reviewer, however, may seek to
augment this outstanding contribution with still
more features. For example, elements of safety
and the protection of the public health have been
interwoven into the various chapters, but on too
modest a level from my point of view. What is
needed is an additional chapter on safety. What
are the special approaches used to cope with the
impact of this new industrial development upon
society? What are the guiding safety philosophies
and regulatory requirements? What are the tech-
niques that can be used to identify and evaluate
risks? Is there a new emerging role for the engi-
neer to be more responsive to societal issues
in providing inputs for evaluating acceptable
risks? O

By E. M. Goodger
John Wiley & Sons, NY. 1980
Reviewed by H. H. Lee
University of Florida
This is a useful reference for those who would
like to get a broad picture of alternative fuels as
well as for those who want some pertinent data
in certain specific areas. Fuel specialists may
find it useful in relating their area of specialty to
the overall picture of alternative fuels. As the
author adequately puts it, the book is offered
as one convenient format for the collocation and
analysis of relevant data scattered about the
The alternative fuels considered in this book
are those intended for oxidation only and so the
wider field of alternative energy sources such as
wind, tides, geothermal, direct solar radiation
and nuclear fusion is excluded from consideration.
It deals with alternative ways of using con-
ventional fuels by modifying the forms and ap-
Continued on page 88.





Yale University
New Haven, CT 06520

T HE CONCEPTS OF MINIMUM reflux ratio and op-
timum feed tray location in distillation prob-
lems involving more than two components are
somewhat more difficult for students to grasp than
are these concepts applied to two-component sys-
tems. These concepts can be developed from
fundamentals by considering zones of constant
composition and ratios of key components, using
either analytical solutions (such as the Underwood
equations) or stage-to-stage calculations. They can
also be developed with the use of computers by
calculating and plotting results for the distillation
of a particular multicomponent system for a spe-
cified separation of various reflux ratios and vari-
ous feed tray locations. The concept of minimum
reflux ratio becomes very real when mole fractions
<0 or >1 appear in printouts.
One reasonable way to approach the teaching
of multicomponent distillation to undergraduates
is in three steps:
1. Develop equilibrium relations .and operating line
equations for the simplest case of constant relative
volatility and constant molal flow rates of liquid and
vapor streams.
2. Assign students to study patterns of distribution by
performing stage-to-stage calculations for a specified
separation of a four-component mixture at several
reflux ratios (including at least one below the min-
imum) and several feed tray locations.
3. Generalize the results of student calculations and
develop criteria for minimum reflux ratio and op-
timum feed tray location.
The emphasis here on a study of patterns of

The concepts of minimum
reflux ratio and optimum feed tray
location in distillation problems involving more
than two components are somewhat more
difficult for students to grasp ...

0 Copyright ChE Division, ASEE, 1983

-------r IRIM--Hls
Charles A. Walker received his education at the University of Texas
at Austin and Yale. He is currently Chairman of the Department of
Chemical Engineering at Yale. He has recently edited a book on tech-
nology and social and political issues in radio-active waste manage-
ment, and he teaches courses in separation processes and conservation
of mass and energy. (L)
Bret L. Halpern received his PhD from the University of Chicago
and is currently an Assistant Professor. He has had experience in sur-
face science and catalysis at the University of Missouri (Rolla) and the
University de Lyon (France). His main research interest is the dynamics
of surface-catalyzed reactions, and he teaches courses in chemical en-
gineering laboratory, electrochemistry, and water quality control. (R)

distribution as an aid to understanding separation
processes is in accord with C. J. King's approach
as outlined in his book on separation processes.
Step 2 of this approach can also be carried out
handily with programmable calculators, providing
students with the convenience of studying any-
where they please rather than having to seek out a
computer terminal. The time required with a pro-
grammable calculator is quite reasonable. With the
programs developed here it is possible to perform
calculations for several cases in the course of one
Consider the distillation of a six-component
mixture.* Assume constant relative volatility and
constant molal overflow. The basic equations can
be written as follows:
*The TI-58C provides enough memories and program
steps for more than six components.


Yi =- aixi I iaix

L-b b
i,m +l Yi,m' yi,b
L L+d d
xi, + 1 L y'- x.d

Here the operating lines have been written in
the form convenient for calculations up a column,
beginning at the stillpot. These equations contain
20 constants for the six-component case for a
given feed composition, feed condition, desired
separation, and reflux ratio.
Register assignments and programs for stage-
to-stage calculations on the distillation of a six-
component mixture are presented in Table 1. Note
that registers 01-06 are reserved for storing suc-
cessive values of xi and yi. These values can be re-
called as needed for recording or plotting. Pro-
grams in Table 1 are also suitable for calculations
on 2-, 3-, 4-, or 5- component mixtures provided

zeroes are stored in registers designated for the
missing components.
The use of these programs for a case where the
bottoms product composition is known is straight-
forward. After placing values of xi,b in registers
01-06, depressing key A will result in calculation
of the composition of vapor leaving the stillpot.
Depressing key D then yields the composition of
liquid leaving the first tray above the stillpot. Al-
ternating between keys A and D is thus all that is
needed to calculate up the column. Calculations
above the feed tray consist of depressing keys A
and A' alternately.
As an example of the use of these programs
consider a simple case of distillation of a four-
component mixture.


mole % in feed


Six-Component Distillation Calculations
TI-58C Programmable Calculator
(Partition to 239/29)



Stripping Rectifying

2nd Lbl A
STO 13
STO 00
STO 28
2nd Lbl B
RCL 2nd Ind 28
2nd Prd 2nd Ind 00
RCL 2nd Ind 00
SUM 13
2nd Dsz 0
STO 00
2nd Lbl C
RCL 13
2nd Prd 2nd Ind 00
2nd Dsz 0

2nd Lbl D
STO 00
STO 13
2nd Lbl E
RCL 14
2nd Prd 2nd Ind 00
RCL 2nd Ind 13
SUM 2nd Ind 00
2nd Dsz 0

2nd Lbl A'
STO 00
STO 13
2nd Lbl B'
RCL 21
2nd Prd 2nd Ind 00
RCL 2nd Ind 13
INV SUM 2nd Ind 0
2nd Dsz 0




(L b)/L
(L + d)/L


The overhead product is to contain 99 mole %
A, and the recovery of A is to be 99%. With these
specifications it is reasonable to assume that the
overhead contains only A and B, and a good esti-
mate of the bottoms composition can be made:

Xl,b = 0.0067
X2,b = 0.3267
X3,b = 0.2500
X4,b = 0.4166

A first assignment for calculations in this case
could be as follows:
1. For a reflux ratio in the range L/d = 1 2, calculate
up the column until a zone of constant composition is
reached. Prepare a plot of liquid composition vs.
plate number. Select a reasonable tray for introduc-
ing the feed and calculate up the column to the prod-
uct composition. Try two more feed tray locations.
Which choice of feed tray requires the minimum
number of trays for the specified separation?
2. Study the effect of decreasing L/d. Try a series of
values until you reach a value where the specified
separation can not be achieved. What happens?
3. Rewrite the problem for a case where a split is to be
made between components 2 and 3. For example,
specify that recovery of B in the overhead is to be
99% and recovery of C in the bottoms is to be 99%.
With these exercises behind them, students are
well prepared for more formal developments of
multicomponent distillation calculations.
The results of calculations for a reflux ratio
(L/d) of 1.2 are summarized in the Table below
and in Fig. 1. Tray 8 was selected as the feed tray
after inspecting a graph of tray number vs. mole
fraction in which calculations were performed
from still to a pinch region (at about tray 18).

number [

Summary of Results for L/d = 1.2







1.7431 x 10-5
1.0564 x 10-6
(1.3858 x 10-7)

Values of x and y can be recorded for each
tray, of course, but liquid mole fractions on even-
numbered trays were chosen here as sufficient to
define a graph of tray numbers vs. liquid mole
fraction. O

Editorial Note: A more complete set of programs pro-
viding for calculations beginning at either the top or bot-
tom of a column is available and can be obtained by con-
tacting either of the authors of this article.


xi = mole fraction of ith component in liquid.
yi = mole fraction of ith component in vapor.
(zi = relative volatility of ith component.
L = Liquid flow rate in rectifying section,
L = Liquid flow rate in stripping section,
b = Flow rate of bottoms product, moles/hr.
d = Flow rate of overhead product, moles/hr.
m + 1 refers to plate above mth plate in strip-
ping section.
n + 1 refers to plate above nth plate in recti-
fying section.

BOOK REVIEW: Alternative Fuels
Continued from page 85.
plications of these fuels, and with the unconven-
tional fuels that can be derived from both con-
ventional and unconventional fuels. The author
then examines and compares each of the alterna-
tives in terms of combustion characteristics, and
combustion performance.
The last chapter is devoted to those fuels which
appear to hold particular promise, which are coal
conversion products, alcohol and hydrogen.
This book is informative and may be useful as
a reference in an introductory course on fuel. O


.1 .2 .3 4 .5 .6 .7 .8 .9 1.0

FIGURE 1. Results for L/d = 1.2

Continued from page 63.
of A5 to regulate the product concentration of A1,
the feed concentration of A6 to regulate the prod-
uct concentration of As, and the total feed rate to
regulate the product concentration of A,. Tran-
sients following a temporary disturbance, using
proportional-plus-integral control in each loop, are
shown for the outputs in Fig. 11, and the manipu-
lations in Fig. 12.
The limit cycle is clearly demonstrated by Figs.
11 and 12 to be unconstrained. Further, the same
ultimate period and amplitude is reached follow-
ing a variety of disturbances. Also, the "time-
averaged" behavior of the manipulated variables
shows that the system has diverged to an alternate

The ultimate goals are twofold: first,
we want to know whether indeed input multiplicity
occurs in industrial systems with sufficient
frequency to be of concern ...

steady-state, as discussed in the previous example.
At this alternate steady-state the control system is
less stable, causing an initial tendency to diverge,
but ultimately resulting in an unconstrained limit
The unconstrained limit cycle phenomenon
is potentially very important. In contrast to
"smooth" divergence to an alternate steady-state,
it is quite simple to detect in an operating plant.
Also, we have mathematical evidence, which will
be presented below, which suggests that the same
mathematical condition necessary for input multi-
plicity may also be necessary for an unconstrained
limit cycle. Therefore, if the sudden reduction in
stability is observed as a sustained oscillation, in
which none of the valves or transmitters is satu-
rated, this may be strong evidence of input multi-
plicity, and a good clue on how to eliminate the

Relations Between Input Multiplicity and Sta-
bility. Consider a process with steady-state be-
havior modelled by the equations
c=f(m) (1)
where c is a vector of n process outputs, m a vector
of n process inputs or manipulations, and f a vector
of n steady-state model equations. Input multi-
plicity occurs if more than one value of m can pro-

duce the desired value of c.
Under mild restrictions, catastrophe theory
(see for example Aris [1] or Calo and Chung [4,5,
6]) shows that a necessary condition for input
multiplicity is

Sac= 0 (2)
Syndicates the determinant, and
where I I indicates the determinant, and




ac ac2 DC2 3C2
am am, am2 amn

3Cn acn acn
aml am2 amn
is the Jacobian matrix of steady-state gains.
Roughly speaking, if Eq. (2) is satisfied anywhere
in the (c,m) space, the possibility of input multi-
plicity exists. The one-dimensional version of this
result is quite easy to interpret (see Fig. 14).
As shown previously [7,8], the condition in Eq.
(1) is closely related to the stability of the in-
tended control scheme. These results can be briefly
summarized as follows:
The components of c and m will each be ordered
according to the intended control scheme, which is
pairing cj with mj (for each j = 1,2,..., n) in a
typical single-input-single-output control loop.
This is the most widely practiced form of multi-
variable process control, and the majority of the
loops will contain proportional-plus-integral ac-
tion. Therefore, the diagonal elements of Eq. (3)
represent the open-loop process gains for the in-
tended pairing.
A plausible control scheme is defined as one in
which each of the n individual loops is operated in
strictly negative feedback. To my knowledge, im-
plausible control schemes are never knowingly op-
erated in practice, for at least two reasons. Any
loop with positive feedback cannot be individually
operated and tuned, since it is unstable. And, al-
though the entire system of n loops operating to-
gether may be stable, it is typically not sufficiently
stable to give the required robust behavior in the
presence of unmeasured disturbances.
The matrix ac'/am is formed from ac/am by
multiplying rows or columns, as desired, of ac/am
by (-1) as needed to produce a diagonal with all
positive elements. For example, if

ac _( 1
am -




c /1 1 or 1 1
'am 1
One can show [7,8] that, unless the following
condition is satisfied at the desired steady state

ac >0 (4)

there exists no set of plausible gains and reset
rates which will yield a stable control system at
that steady state.
The conditions in Eq. (2) and Eq. (4) are
closely related. The difference between a3c/3m[
and |Jc'/amm can be at most a sign. The locus of
all steady-states satisfying Eq. (2) therefore di-
vides the space into regions in which different
pairings can be stable and plausible. This observa-
tion was used [8] to map the steady-state space for
the 2 x 2 control scheme, on the single A-R S
exothermal reactor, into regions in which different
pairings could be stable and plausible. The results
are shown in Fig. 13.
These theoretical results demonstrate why the
stability level is apt to be significantly altered if
the control system diverges to an alternate steady
state, which is in a different stability/plausibility
region. The phenomenon is explained by the close

u 0.6

a 0.3




0.7 0.8 0.9

1.0 1.1

FIGURE 13. Stability/plausibility regions for 2 x 2 re-
actor control scheme

relation between the necessary conditions for in-
put multiplicity and for stability/plausibility.
Input Multiplicity and Unconstrained Limit
Cycles. The results to be discussed here are en-
couraging, but much more fragmentary than I
would like. They are currently worked out only for
a one-dimensional control system, i.e., with one
input and one output, with a specific form of dy-
We consider a process whose dynamics are rep-
resented by the single equation
d =f(m)-c (5)

The control will be proportional-plus integral
m = Ke (r-c) + f (r-c) dt (6)

where r is a constant set- point value. Differentiat-
ing Eq. (6) with respect to time, and substituting
Eq. (5) for dc/dt into the result, yield a second
differential equation

dt Ke -f (m) +- (7)

Eqs. (5) and (7) form two ordinary differential
equations in two unknowns, the case for which
there exists the largest body of theory on limit
cycle behavior.
A theorem of Bendixon [3] states the following:
If a system has dynamic equations x = P (x,y) and

y = Q (x,y), then no limit cycle can exist in any
region of the (x,y) phase plane within which
(aP/ax + DQ/ay) has an invariant sign, and is
not identically zero. Applied to Eqs. (5) and (7),
this says that no limit cycle can exist in any region
of (c,m) space unless
-1-K df
changes sign.
If we assume plausibility, and therefore K,
(df/dm) > 0 at the desired steady state, then there
can be no limit cycle unless df/dm changes sign
somewhere in (c,m) space. Eq. (5) shows that the
steady state is c = f(m). These observations to-
gether show that a necessary condition for a limit
cycle is that
0 (8)
somewhere in the steady-state (c,m) relationship.


This is the same condition needed for input multi-
The requirement for an unconstrained limit
cycle enters in the following manner: The proof of
Bendixon's theorem requires use of Green's

(Pdy Qdx) = + 3Q dx dy
SJ x D y )
which in turn requires that P and Q be continuous
and have continuous first partial derivatives. A
limit cycle operating with a saturated m or c would
not be described by functions in Eqs. (5) and (7)
which satisfied these conditions.
These fragmentary observations give encour-
agement about the relation between input multi-
plicity and unconstrained limit cycles. In this
specific example, similar conditions on the steady-
state derivatives and plausibility considerations
appear to be involved in both phenomena. More
investigation is warranted to determine the full
extent of the connection between input multiplicity
and unconstrained limit cycles, since this could
facilitate demonstration of the former by observa-
tion of the latter.
Adjacency and Connectedness. My experience
with the phenomenon has strongly suggested that
input multiplicity will not be a problem in process
control systems unless there exist at least three
possible steady states. Fig. 14 is sketched to sup-
port an approximate explanation of this assertion,
for a one-dimensional system. The sketches are
example steady-state relationships c = f (m).
In Fig. 14 (a), three possible steady states
exist for the chosen set-point value r. Of these,
only 1 and 3 can be stable if a controller with posi-
tive gain is used, and only 2 is stable if a negative
gain is used, because of the different signs of df/
dm at the steady states. Fig. 14 (b) then suggests
that, with only two steady-states, only one can be
stable for a particular choice of gain. Thus, we
might conclude that, unless there are at least three
steady states, we need not be concerned about hav-
ing the system come to rest at other than the de-
sign steady state. This reasoning explains why all
the illustrative examples discussed previously have
more than two steady states. We might also con-
clude from Figs. 14 (a) and (b) that adjacent
steady states cannot both be stable for the same
control settings, so that divergence could occur
only to a non-adjacent steady state. Fig. 14 (c) is
sketched to show that we have to amend this con-
clusion to state that two steady states which are
adjacent and connected cannot both be stable for


mI 2 m3 m
mI "



ml m2 m3
FIGURE 14. Illustrating adjacency and connectedness

the same control settings. These observations are
useful for dealing with the input multiplicity prob-
lem in practical situations.
Extending these observations to systems of
greater dimensionality requires further research.
The first complication is the difficulty of defining
and identifying adjacency and connectedness of
steady states in such systems. Partial results have
been obtained [10], but these are in need of sharp-
ening. Further, in higher dimensions the strongest
statement we can hope to prove is that adjacent
and connected steady states cannot both be stable
and plausible for the same controller settings. This
would still be a powerful statement, because a
stable but implausible system will exhibit a lower
level of stability.

At present, five main issues appear to be of
highest priority: (1) identification of process
structures likely to lead to input multiplicity; (2)
detection of input multiplicity from steady-state
process models; (3) detection of input multiplicity
from plant data; (4) identification of the residual
group of variables responsible for the input multi-
plicity; and (5) choice of control pairings to elimi-
nate the multiplicities.
The ultimate goals are twofold: First, we want
to know whether indeed input multiplicity occurs
in industrial systems with sufficient frequency to
be of concern. Assuming a positive answer, then
the second goal is development of approaches for
detection and elimination of the phenomenon, in


situations where it is a cause of operating prob-
lems. O

1. Aris, R., "Chemical Reactors and Some Bifurcation
Phenomena", Ann. N. Y. Acad. Sci., 1, 314 (1979).
2. Aris, R., and N. R. Amundson, "An Analysis of Chem-
ical Reactor Stability and Control", Chem. Eng. Sci., 7,
121 (1958).
3. Bendixon, I., "Sur les courbes defines par les equations
differentialles", Acta Mathematica, 24, 1 (1901).
4. Calo, J. M., and Chang, H.-S., "Catastrophe Theory
and Chemical Reactors: Exact Uniqueness Criteria
for the CSTR, Catalyst Particle, and Packed-Bed Re-
actor", Chem. Eng. Sci., 85, 264 (1980a).
5. Chang, H.-S., and Calo, J. M., "Exact Criteria for
Uniqueness and Multiplicity of an nth Order Chemical
Reaction via a Catastrophe Theory Approach", Chem.
Eng. Sci., 84, 285 (1979).
6. Chang, H.-S., and Calo, J. M., "Errata", Chem. Eng.
Sci., 35, 2377 (1980b).
7. Koppel, L. B., "Input Multiplicity in Nonlinear, Multi-
variable Control Systems", A.I.Ch.E. Journal, in press
8. Koppel, L. B., "Input-Output Pairing in Multivariable
Process Control", submitted to A.I.Ch.E. Journal
9. Kubicek, M., Hoffman, H., Hlavacek, V., and Sin-
kule, J., "Multiplicity and Stability in a Sequence of
Two Nonadiabatic Nonisothermal CSTR", Chem. Eng.
Sci., 85, 987 (1980).
10. Rickard, K. A., Ph.D. Dissertation in Chemical En-
gineering, Purdue University (1982).

Continued from page 57.
areas. Specifically, the department requires the
project to consist of six students who work five to
ten days as a team, to integrate at least two of the
six different laboratory areas, and to be scien-
tifically relevant or practical, remaining open-
ended and constantly requiring new research ideas
from the student group. Some of the research proj-
ects from a recent 27-project list include:

Heterogeneous catalytic decomposition of acids
Water purification
Removal of carbon dioxide from process gases

During the experiment work, each of the stu-
dents in the group is required to keep data, graphs,
and analyses. One of the students is then desig-
nated to write the group's report. The project as-
sistant grades each student based upon the report,
upon the individual reports written after each day
of experiments, and upon the skills the student
shows during his work.
The equipment used in these laboratories is

quite extensive, and well maintained. It is usually
associated with a professor's current research or
has been obtained after a research project has been
completed. Equipment is also available for new
projects when an old project is phased out after
three or four years. Thus, the research remains
open-ended, and ten-year-old equipment and proj-
ects do not exist.
By the fourth year of study, a student has
chosen major and minor fields of specialization,
and has a specific research aspect of the research
project of a professor or graduate student. Now
the student truly has access to the showcase of
technology and equipment available at Delft. For
example, in the reaction kinetics laboratory a stu-
dent may work with any of the following equip-

thermal analysis (TA)
differential thermal analysis (DTA)
differential scanning calorimetry (DSC)
evolved gas analysis (EGA)
advanced impregnation apparatus
high-temperature drying apparatus
variety of furnaces

There is a great variety of research to choose
from at Delft, and a student can review the direc-
tory that summarizes all research being conducted
in chemical engineering and chemistry before he
selects a professor. Some of the specific research
topics available, from a recent list of thirty-two
different research areas, include:

Inhibition of crystal growth in industrial processes.
Separation of hydrous and anhydrous calcium sulfate
in the production of phosphoric acid.
The influence of flow behavior on the separation of
solids from aluminum melts.
Automation of kinetic experiments involving the hy-
drogenation of carbon monoxide.
Desalination of sea water.
Recovery of metals from waste water.
Removal of hydrogen sulfide from process gases.
Production of manganese oxide for batteries.
Pressurized fluidized bed combustion of coal.
Process development on the hydroformylation of
olefins using immobilized catalysts.

Because of Delft chemical engineering's strong
emphasis on applying theory directly to research
and practical industrial process problems, a Delft
chemical engineer is valued highly by industry.
The Delft graduate has no difficulty in adapting to
"real world" chemical plants, since he has been
working with small scale reactors and processes
since his second year at the Delft University of
Technology. O



Departmental Sponsors: The following 144 departments contributed
to the support of CHEMICAL ENGINEERING EDUCATION in 1983 with bulk subscriptions.

University of Akron
University of Alabama
University of Alberta
Arizona State University
University of Arizona
University of Arkansas
Auburn University
Brigham Young University
University of British Columbia
Brown University
Ilucknell University
California State Polytechnic
California State U. at Long Beach
California Institute of Technology
University of California (Berkeley)
University of California (Davis)
University of California (Santa Barbara)
Carnegie-Mellon University
Case-Western Reserve University
University of Cincinnati
Clarkson College of Technology
Clemson University
Cleveland State University
University of Colorado
Colorado School of Mines
Columbia University
University of Connecticut
Cornell University
University of Dayton
University of Delaware
U. of Detroit
Drexel University
University of Florida
Florida Institute of Technology
Georgia Technical Institute
University of Houston
Howard University
University of Idaho
University of Illinois (Urbana)
Illinois Institute of Technology
Institute of Paper Chemistry
University of Iowa
John Hopkins University
University of Jordan
Iowa State University
Kansas State University
University of Kentucky
Lafayette College

Lamar University
Laval University
Lehigh University
Loughborough University
Louisiana State University
Louisiana Tech. University
University of Louisville
University of Maine
University of Maryland
University of Massachusetts
Massachusetts Institute of Technology
McMaster University
McNeese State University
University of Michigan
Michigan State University
Michigan Tech. University
University of Minnesota
University of Missouri (Columbia)
University of Missouri (Rolla)
Monash University
Montana State University
University of Nebraska
New Jersey Inst. of Tech.
University of New Hampshire
New Mexico State University
University of New Mexico
City University of New York
Polytechnic Institute of New York
State University of N.Y. at Buffalo
North Carolina State University
University of North Dakota
Northeastern University
Northwestern University
University of Notre Dame
Nova Scotia Tech. College
Ohio State University
Ohio University
University of Oklahoma
Oklahoma State University
Oregon State University
University of Ottawa
University of Pennsylvania
Pennsylvania State University
University of Pittsburgh
Princeton University
Purdue University
University of Queensland
Queen's University

Rensselaer Polytechnic Institute
University of Rhode Island
Rice University
University of Rochester
Rose-Hulman Institute
Rutgers U.
University of South Carolina
University of Saskatchewan
South Dakota School of Mines
University of South Alabama
University of South Florida
University of Southern California
Stanford University
Stevens Institute of Technology
University of Sydney
Syracuse University
Teesside Polytechnic Institute
Tennessee Technological University
University of Tennessee
Texas A&M University
Texas A&I University
University of Texas at Austin
Texas Technological University
University of Toledo
Tri-State University
Tufts University
Tulane University
University of Tulsa
University of Utah
Vanderbilt University
Villanova University
University of Virginia
Virginia Polytechnic Institute
Washington State University
University of Washington
Washmgton University
University of Waterloo
Wayne State University
West Virginia Inst. Technology
West Virginia University
University of Western Ontario
Widener College
University of Windsor
University of Wisconsin (Madison)
Worcester Polytechnic Institute
University of Wyoming
Yale University
Youngstown State University

TO OUR READERS: If your department is not a contributor, please ask your department chairman to write CHEMI-
CAL ENGINEERING EDUCATION, c/o Chemical Engineering Department, University of Florida, Gainesville, Florida




F. C. Jelen, Fellow of the American Association of Cost Engineers, Seond Edition
and James H. Black, University of Alabama and Fellow of the American Association of Cost Engineers
1983, 530 pages, (0-07-032331-3)
The only book that covers the entire breadth of cost engineering, this successful text is ideal for courses in engineer-
ing economics, cost engineering, and chemical plant design.
Featuring the work of contributors from both industry and education, Cost and Optimization Engineering, 2/e presents
up-to-date cost data plus complete coverage of important, timely topics. The new edition examines the new tax laws
and discusses all aspects of inflation in detail. A novel and illuminating analysis of productivity is provided, and a
chapter on risk analysis has been added.
Solutions Manual (0-07-032332-1)

Ray W. Fahien, University of Florida
1983, 640 pages (tent.), (0-07-019891-8)
Written by the widely respected'editor of Chemical Engineering minimal prior mathematical or scientific background proceeds
Education, this is the only text in the field to treat simultaneously gradually in a logical step-by-step approach illustrates both the
the one-dimensional transport of heat, momentum, and mass in de- similarities and dissimilarities in the equations used to describe the
tail before examining the more complex subject of multidimensional transport process and emphasizes the physical meaning of mathe-
transport. As a result of this approach, repetition is reduced, the matical quantities and operations such as the gradient, the diver-
analogies are fully exploited, and their limitations are exposed, gence, and the dyadic product.
The book meets the need for a beginning transport phenomena text'
that offers a careful explanation of the fundamentals assumes Solutions Manual (0-07-019892-6)

J. Frank Valle-Riestra, Dow Chemical U.S.A., and Adjunct Professor, University of California, Berkeley -'"-"
1983, 736 pages (tent.), (0-07-066840-X)
This is the first text to bring together and systematize all of the constituent themes of project evaluation and manage-
ment in the chemical process industries.
Intended for courses in project evaluation, plant design, and senior design, it gives students insight into how to apply '
acquired project evaluation tools, as well as assimilated academic disciplines, to "real world" industrial situations-- =. .'-
with the ultimate objective of promoting the commercial success of projects. Approximately 400 problems-often ., -
open-ended, unstructured, and typical to the industrial environment-allow students to practice problem-solving tech- p '
niques and learn methods of problem synthesis in realistic situations.
Solutions Manual (0-07-066841-8)
Charles D. Holland, Texas A&M University, and Anthanasios 1. Liapis, University of Missouri at Rolla
1983, 512 pages (tent.), (0-07-029573-5)
Now there is a book that provides an in-depth, unified presentation of modeling, numerical solutions of modeling equa-
tions, and the analysis of both staged and continuous separation processes.
The perfect text for courses in dynamics of separation processes, advanced unit operations, and applied numerical
methods, it features methods for setting up ordinary and partial differential equations for simple systems at the begin-
ning of the book, developing techniques useful for formulating more complex systems in subsequent chapters. Select-
ed numerical methods are presented in a programmed manner using clear, simple examples so that students with no
previous exposure to numerical methods can easily understand the material.
Solutions Manual (0-07-029574-3)

Nicholas Tsoulfanldis, University of Missouri-Rolla
1983, 571 pages, (0-07-065397-6)
Assuming no background in the subject, this new book teaches students how to select the proper detector, analyze
the results of counting experiments, and perform radiation measurements following proper health physics procedures,
It provides exceptionally comprehensive treatment of errors to help students understand the importance of reporting'
errors when analyzing experimental results. Coverage of data analysis methods is clear and concise, discussing meth-
ods of curve fitting, interpolation, and least square fitting as well as the tools needed for analyzing spectroscopic mea-
surements. The text also offers exceptionally detailed treatment of spectroscopy.
i College Division McGraw-Hill Book Company 1221 Avenue of the Americas New York, New York 10020 v

Full Text