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 )
serial   ( sobekcm )
periodical   ( marcgt )


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
sobekcm - AA00000383_00075
lcc - TP165 .C18
ddc - 660/.2/071
System ID:

Full Text

c ileg ining e u a io

ac "dwle


wiAh a dao atiof a uids.


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
Ralph W. Pike
Louisiana State University
James Fair
University of Texas
Gary Poehlezn
Georgia Tech
Darsh T. Wasan
Illinois Institute of Technology
J. J. Martin
University of Michigan
Lowell B. Koppel
Purdue University
William H. Corcoran
California Institute of Technology
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. W. 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





107 Two Gentlemen From Yale,
J. J. Carberry

122 The Development of Communications Skills
Through a Laboratory Course, Curtis W. Frank,
George M. Homsy, Channing R. Robertson

Department of Chemical Engineering
102 ChE at Yale, Charles A. Walker
The Educator
98 C. 0. Bennett of Connecticut,
Stuart M. Case, Robert W. Coughlin
011 The Oscillating Sink, Thomas H. Leise,
Daniel J. Jenkins, John M. Tarbell

118 Digital Computer Application in Process
Control, M. F. Abd-El-Bary, Sriram Chari

Class and Home Problems
114 Melting Ice Cubes Problem,
Judge T. Sommerfeld, Peter A. Minderman
126 Trends in Biomedical Education,
Nicholas A. Peppas, Richard G. Mallinson
132 Recent Developments of ChE Education in
Mexico, Enrico N. Martinez, Roman Gomez
138 Chemical Process Synthesis, J. J. Siirola

125 In Memoriam

108-121 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 1982 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
0009-2479 for the identification of this periodical.


educator I

. BQ e4 l


University of Connecticut
Storrs, CT 06268

S C + 0 + B Many Active Sites

That equation is not about carbon, oxygen and
boron; it is about a man whose initials are COB,
whose friends and acquaintances often call him
CO and who does research on the catalytic
chemistry of CO.
"He's a rare blend of humanist and scientist. If fate
had rolled the dice a bit differently, he might have
been equally productive as an English professor in-
stead of as a chemical engineer."
These are the words of a professor of chemical
engineering at a midwestern university talking
about one of his former teachers, Carroll O.
Bennett of the University of Connecticut. Another
chemical engineering professor described Bennett
as "one of the most liked chemical engineers in
the country." Add to that the lasting influence of
his co-authored textbook, "Momentum, Heat, and
Mass Transfer," first published in 1962 and still
a world-wide standard classroom and reference
work, and it is no surprise that Bennett's se-
lection for the 1980 Warren K. Lewis Award was
greeted so enthusiastically.
For the modest, soft-spoken chemical engineer-

Because transient experiments
must be purged of the influences of transport
resistances, Bennett became an early designer of
gradientless reactors, which were forerunners
of the now well-known Berty reactor.

Copyright ChE Division, ASEE, 1982

ing professor, the path to the AIChE's highest
award for chemical engineering education is es-
pecially poignant. This is because it puts him
directly in the footsteps of one of the great in-
fluential teachers in his own life, Yale chemical
engineering professor, Barnett F. Dodge, who won
the first Lewis Award back in 1963.
Bennett's path to the prize led him through a
career that penetrated areas as diverse as meteor-
ology and engineering design, and led to journeys
on four continents. Along the way he became a
gourmet, a lover of fine art and music, and fluent
in four languages.
Born in New Britain, Connecticut's "hardware
city," Bennett grew up as an only child. Even in
high school, Bennett found he had an interest in
chemistry, but also found himself drawn to the
study of languages, doing well in the study of
Latin, French, and German. Influenced by the
reality of the depression, however, he continued
his studies at Worcester Polytechnic Institute and
by the end of his freshman year had chosen to



study chemical engineering. After receiving his
degree in an accelerated program in 1943, he
entered the Army Air Corps.
Returning to Connecticut after the war,
Bennett worked for a while as a metallurgist
before deciding to continue studies in chemical
engineering. Bennett spent three years at Yale,
receiving his doctorate in 1950. At Dodge's sug-
gestion, he did his thesis research in the area of
high pressure thermodynamics. "I studied the
compressibility of mixtures of hydrogen and nitro-
gen up to pressures of 3,000 atmospheres. It was
rather interesting work, with a lot of mechanical
technique involved." Yale, of course, was where
J. Willard Gibbs had done much of his pioneering
work on thermodynamics at the end of the 19th
century, and his influence was still strong at Yale
when Bennett was there.
At the same time, Bennett kept up his interest
in French, taking courses in French literature
with Henri Peyre, ". . who is very well known to
anybody who knows anything about French litera-
ture. I also went to lectures in art history by
Vincent Sculley, who is another shining light at
Yale," Bennett recalls. Both of these interests
were to reemerge later and influence his career
and life.
Bennett remembers Dodge "as a man of great
intelligence with a very special personality. He
was quite austere to students." He recalls the
weekly meetings he had with Dodge to discuss
his progress on his thesis; "I would go in there
and sit down in the chair, and he would look at
me and I was supposed to start telling him what
I had done. He would ask a few questions. I was
always afraid I would say something wrong. I
don't know whether Dodge did this purposely or
not, but it was very effective. I was always pre-
pared for my weekly meetings." While Dodge
seemed somewhat distant at the time, Bennett says
he realized later that the professor "was a man
of all sorts of human characteristics. Of course,
one never realizes that about one's professors."
Even though doctoral research in chemical
engineering was not at all common back in those
Yale days, Bennett knew he would stick with
chemical engineering rather than switch to
chemistry. "I didn't want to become a chemist
because I didn't want to stay in the lab," he says.
"I wanted to do things on a bigger scale, and the
scale-up process-going from the small scale you
have in the laboratory to the large scale you have
in a chemical plant-interested me." But Bennett

did not immediately go out into the field and scale
up chemical plants. Instead he went into teaching
at Purdue.
"When you go into a career as a teacher in
engineering, there are two parts to it," according
to Bennett. "One is the engineering part, which
emphasizes teaching undergraduates, so they can
go out and practice engineering the way MD's

Add to that the lasting influence of his
co-authored textbook, "Momentum, Heat, and Mass
Transfer, ". . and it is no surprise that Bennett's
selection for the 1980 Warren K. Lewis Award
was greeted so enthusiastically.

go out and practice medicine. The other part of
one's life in being a university teacher is doing
Looking back on that era, Bennett comments
that there were many engineering problems then
for which the solutions were not as clear as they
are now. "When I was just starting out in the
'50's, it became clear that chemical engineers
would probably become more interested in funda-
mental research which would be more similar to
research done in physical chemistry than to re-
search done in engineering." Thus he started his
research with Prof. Dodge in the basic area of
thermodynamics. While the work did not have an
immediate practical application in view, it was
related to the fact that Dodge, along with others,
was interested in solving problems connected with
synthesizing ammonia from nitrogen and hydro-
gen. Dodge wanted to find a way of running the
process at high pressure and temperature, so that
no catalyst would be needed. Ultimately, Dodge's
high temperature and high pressure process
proved uneconomic, and now Bennett's research
pursues the opposite approach: exploring low pres-
sure and low temperature reactions using
catalysts. Bennett finished his work with Dodge in
the summer of 1949, about the time he began to
realize that teaching might be interesting.
After considering various teaching offers,
Bennett selected Purdue University where he
started teaching thermodynamics, sharing the
course with Prof. Joe M. Smith. Smith had just
finished his book on the subject, and was even
then a well-known teacher and researcher in
chemical engineering. Smith had studied with
Warren K. Lewis at MIT and had picked up Lewis'
Socratic method of teaching.


Bennett spent three years at Yale, receiving his doctorate in 1950. At Dodge's suggestion he
did his thesis research in the area of high pressure thermodynamics. . At the same time, Bennett
kept up his interest in French, taking courses in French literature with Henri Peyre.

A year later another young instructor, Jack
Myers, came from Michigan to join the Purdue
faculty. Myers eventually became Bennett's co-
author on the now-famous textbook. Bennett and
Myers started thinking about the book in the mid
to late '50s, began it about 1959, and published
it in 1962, after Bennett had left Purdue for a
stint in private industry. The book represented a
new viewpoint on what the profession was really
In the old days, Bennett explains, engineers
were associated with specific industries. Thus,
there were sugar engineers, petroleum engineers,
sulfuric acid engineers, and so on. Then, Lewis and
Walker and others at MIT began to develop the
idea of organizing chemical engineering in terms
of "unit operations." These operations, such as
distillation, extraction, absorption, and others,

CO relaxing with his children Edward, Elizabeth and
Jonathan, after a skiing trip, in France.

were common to all sorts of chemical engineering
processes. Thus, training a person in "unit opera-
tions," would enable him to go out and work in
any one of the chemical industries.
That approach grew in strength and per-
suasiveness until it became increasingly clear that,
behind the unit operations, there were really only
a few basic divisions of the subject grouped
around the concepts of heat, mass, and, mo-
mentum transfer. "We wanted to unify the unit
operations which had previously unified the
various different processes in chemical engineer-

ing, but to unify them as being manifestations of
the principles of heat transfer, mass transfer, and
momentum transfer," Bennett says.
At the same time, the computer revolution
was just getting under way. "The first computers
of any practicality began to be used in the '50s,
and it became clear that it was practical to use
more mathematics than you could previously," he
notes. "You could solve certain differential equa-
tions or partial differential equations, or certain
multi-component problems that previously would
have been impractical to solve by hand because it
would take too long, and you probably would make
too many mistakes before you got through. The
computer created an atmosphere where the funda-
mental approach became more attractive or more
practical, and this, too, influenced the book."
While he was working on the book, and
continuing his research on high pressures, mass
transfer, and heat transfer, Bennett began to feel
he would like to be a practicing engineer for a
while. So in 1959 he left Purdue to join the
Lummus Co. in New York as a development engi-
neer. In the five years he was there, he worked
hard to develop an independent design capability
for the company, which previously had generated
much of its business by using licensed processes
developed by other companies.
"Dr. Bennett played a key role in establishing
the research and development department at the
Lummus Co., and the basic philosophies and ap-
proach which he established have been funda-
mental to the growth of this effort," according
to a senior vice president at CE-Lummus, who
added, "His spirit of free investigation and the
practical application of fundamentals have con-
tributed to the development and commercialization
of a number of important processes." A former
vice president for research and development at
CE-Lummus noted that Bennett's association with
Lummus was mutually beneficial and stated, "The
Lummus Co. benefited in the early adoption of
computer calculation techniques for chemical
engineering design. Professor Bennett benefited in
knowing how industry performs its work and
what industry needs. No doubt such a combination
has helped him to inspire his students and col-
leagues to high achievement."


Bennett's years at Lummus confirmed his be-
lief that he was on the right track in returning
to fundamentals as the basis of teaching chemical
engineering. "It convinced me that a person who
is intelligent and knows fundamental principles
can go right in and compete with people who are
experienced engineers, and who perhaps know how
to do certain things, but don't remember or never

CO and his wife, Jean, on a night out in Storrs.

had the fundamental principles, so that their
adaptability and flexibility are limited," he ob-
serves. "When improvements come along, or new
processes or new situations develop, often the
people with rusty fundamentals are not able to
cope with them and design something suitable."
He adds, too, that his teaching has been enhanced
by the practical examples he can draw from his
industrial experience, and he includes such
problems in his lectures, his exams, and in his
It was from Lummus that Bennett came back
to education, to the University of Connecticut, in
1964. And it was at UConn that Bennett's interest
began to shift into the design of chemical re-
actors and heterogeneous catalysis. The field was
not entirely new to Bennett, since some of his
work at Lummus involved catalytic reactions, and
many industrial processes are based on catalysts.
"Nevertheless, catalysis was always thought of as
a sort of black art and there seemed to be a lot of
room for increased understanding. It seemed to
be a good thing to study," Bennett reflected.
Bennett's elucidation and advocacy of the
transient method for catalytic studies has helped
to dispel some of the blackness of the catalytic

art. This approach allows simple experimentation
to shed light on the rich complexity of even the
simplest catalytic reactions.
Following pioneering experiments by Wagner
in 1938 and further work by Tamaru in 1963,
Bennett laid out a quantitative framework in
1967 [AIChE Journal 13, 890 (1967)]. Later
Kobayashi and Kobayashi and Yang et al.
implemented this approach, thereby bringing to
full fruition the earlier emphasis by Wagner and
Tamaru on measuring adsorption during catalysis.
Although transient experiments have long been
applied in other fields, only within about the last
10 years have they found wide acceptance in
heterogeneous catalysis. As recently as 1975,
chemical engineers modelled reactors using steady-
state rate expressions rather than models based
on elementary steps and their individual rates
which can change in response to changing con-
ditions during transients.
Because transient experiments must be purged
of the influences of transport resistances, Bennett
became an early designer of gradientless reactors,
which were forerunners of the now well-known
Berty reactor. Later, Bennett applied transient
methods to the study of Fischer-Tropsch catalysis
and his current research work is largely in this
In addition to teaching and research, Bennett
also served as acting department chairman on
several occasions, but it was not his first love, and
he was always happy to return to teaching and
research. While teaching has its drawbacks for

Bennett's years at Lummus confirmed
his belief that he was on the right track in
returning to fundamentals as the basis of
teaching chemical engineering.

Bennett (especially tasks like grading exams), it
still is one of the best parts of the job for him.
"Teachers should teach people," he believes. "I
suppose I enjoy the opportunity to make some-
thing clear. When I can describe something in
such a way that it is comprehensible maybe I can
remember the first time I looked at it, and I too,
thought, that it was incomprehensible."
Bennett teaches three undergraduate courses
and one graduate course at UConn. One of these
is a required 40-student section of the junior-year
course in transfer operations. He also teaches ad-
vanced courses in transfer operations in the fall
Continued on page 144.


P department


Yale University
New Haven, CT 06520

dustrial and Engineering Chemistry was
offered in earlier years, a program in Chemical
Engineering was first offered at Yale in 1922-23,
and it was heralded as follows in the Report of
the President for that year:
Professor Harry Alfred Curtis has been called to the
chair of Chemical Engineering, for which we have
provided admirable quarters in the new Sterling
Chemistry Laboratory. . We shall look forward to
the development of the new Department under his
charge in the most hopeful spirit.
Professor Curtis served as Chairman until 1931
and then left Yale to continue a distinguished
career as director of research for a major oil
company and later as Dean of Engineering at the
University of Missouri and a member of the Board
of Governors of the Tennessee Valley Authority.
Before leaving, however, he had provided well for
a successor in the person of Barnett F. Dodge, who
came to Yale in 1925 and served as chairman from
1931 until his retirement in 1964. Professor
Dodge's classic book on chemical engineering
thermodynamics reflected his research interests
in cryogenic engineering and phase equilibrium
and chemical equilibrium at high pressures. Others
who have served on the fulltime faculty in chemi-
cal engineering at Yale, in the order of the dates
of their first appointments, and with the names
of current faculty members in bold face type, are
Clifford C. Furnas, Melvin C. Molstad, Roger H.
Newton, Harold E. Graves, Winford B. Johnson,
R. Harding Bliss, James A. Johnston, E. E.

... at Yale a unique aspect is the fact that
chemical engineering students necessarily pursue
their studies in close association with students and
faculty members in many disciplines, including
those in . social sciences and liberal arts.

Copyright ChE Division, ASEE, 1982

Photo by William K. Sacco

Lindsey, Jr., Charles A. Walker, Raymond W.
Southworth, Homer F. Johnson, Dinwiddie C.
Reams, Randolph H. Bretton, Shen-Wu Wan,
John A. Tallmadge, Jr., Joshua Dranoff, John B.
Butt, Colin McGreavy, Lawrence H. Shendalman,
Reiji Mezaki, John B. Fenn, James B. Anderson,
Gary L. Haller, Daniel E. Rosner, William N.
Delgass, Csaba Horvath, Paul C. Nordine, Jimmie
Q. Searcy, Constantinos G. Vayenas, Bret L. Hal-
pern, and Subbarao B. Ryali. Comments on the
contributions of each of the former faculty
members are precluded by space considerations,
but Harding Bliss's service as the first editor of
AIChE Journal and the fact that the first course
in computer programming at Yale was taught by
a chemical engineer, Raymond W. Southworth,
deserve to be mentioned. Research interests of
current faculty members will be discussed later in
this article.
Until 1962 chemical engineering activities at
Yale were the responsibility of the Department of
Chemical Engineering in the School of Engineer-
ing. In that year Yale embarked on an experiment


in education by combining all engineering activi-
ties in a Department of Engineering and Applied
Science in the Faculty of Arts and Sciences. While
this experiment was successful in encouraging
interdisciplinary teaching and research, the range
of faculty interests was so broad as to make agree-
ment on educational philosophy and faculty ap-
pointments difficult to achieve. In the spring of
1981 the faculty and administration judged that
these difficulties were significant enough to justify
a return to smaller, more homogeneous units, in-
cluding a Department of Chemical Engineering.
During the years 1962-1981 chemical engi-
neering activities continued under the direction
of a faculty that varied in size from seven to ten
members. During some of these years there were
very few undergraduate students majoring in
chemical engineering, and much of our effort was
devoted to the graduate program and research.
In more recent years undergraduate interest in
chemical engineering has increased, as it has at
other institutions, and the undergraduate program
is again attracting reasonable numbers of
Thus, while we might appear to be a "new"
department of chemical engineering, we are in
fact a part of a continuing development of our
discipline at Yale.


engineering at Yale is conventional in many
respects and includes courses in the following
subjects: introductory thermodynamics; conserva-
tion of mass and energy; chemical engineering
thermodynamics; fluid mechanics; mass, energy,
and momentum transport processes; chemical
kinetics and chemical reactors; separation pro-
cesses; chemical engineering laboratory; and
chemical engineering process design, as well as
options in biochemical engineering, environmental
chemical engineering, and research projects.
The teaching of most of these subjects at Yale
is not significantly different from the way they are
taught in other universities and colleges. Each of
us feels, of course, as do all educators worthy
of the name, that he has developed an optimum
method for teaching his subject or that at least
he is now in position to teach it in optimum fashion
next year. There seems to be little point, therefore,
in outlining our approach to each of these courses,
but two courses deserve special mention. Our
course in chemical engineering process design is

taught, quite successfully, by engineers from Olin
Corporation. Dr. Herbert Grove, David Doonan,
Joseph Levitzky, and Howard Martin have par-
ticipated in teaching this course and have added
significantly to the education of our students. We
are grateful for their considerable efforts. Re-
search projects can be taken for credit at any time
in a student's undergraduate years. Each student
is affiliated with one of several research groups
and assigned a problem related to the ongoing re-
search effort. Almost every chemical engineering

Thus, while we might appear to be
a "new" department of chemical engineering,
we are in fact a part of a continuing development
of our discipline at Yale.

student takes at least one semester of research,
usually in the junior or senior year, and a few
continue for two or more semesters and have their
names appear on research publications.
The general requirements for graduation from
Yale College include a total of 36 term courses, 12
of which must be outside the area of the major.
Thus chemical engineering students must take
12 term courses in disciplines other than mathe-
matics, science, and engineering, a considerably
heavier requirement of humanities courses than
is typical for engineering students at other institu-
tions. We observe that some of our students use
these 12 courses to explore several disciplines,
such as literature, history, economics, philosophy,
and languages. Others satisfy a set of distri-
butional guidelines and then use the remaining
courses to concentrate in one discipline. One
student, for example, was able to arrange her
courses so that she satisfied requirements for
majors in both chemical engineering and English
literature. Another satisfied requirements for
majors in chemical engineering and economics.
The breadth of education that results from meet-
ing this requirement is an important and de-
sirable feature of the curriculum.
Recognizing the diversity of interests of
students, we offer undergraduate degrees in chemi-
cal engineering at three levels of intensity. For
those who plan to enter the profession, a B.S. in
Chemical Engineering, requiring a total of 40 term
courses and carefully structured courses in chemi-
cal engineering, is available. For those interested
in chemical engineering but desiring to take more
courses in other disciplines, a B.S. in Engineer-
ing Science (Chemical), requiring a total of 36


term courses and fewer courses in chemical engi-
neering, is offered. For students planning on
business school, law school, or medical school, a
still less intensive B.A. in Engineering Sciences
(Chemical) degree is offered. Most of our students
take one of the B.S. programs and, after gradu-

Photo by William K. Sacco
Dan Rosner at the blackboard with Suleyman Gokoglu.

ation, more than half go directly to industry and
others go to graduate school in chemical engineer-
ing or to schools of business, law, or medicine.
Class sizes are small but growing; in the past few
years we have awarded 10-12 undergraduate de-
grees each year.
In these Chemical Engineering Education
articles about education at the variety of types of
institutions that are the hallmark and the strength
of higher education in the United States, readers
deserve to learn what is unique about each institu-
tion as well as what they have in common. At the
undergraduate level at Yale a unique aspect is
the fact that chemical engineering students
necessarily pursue their studies in close association
with students and faculty members in many
disciplines, including those in the natural sciences,
the social sciences, and the liberal arts. This
necessity is imposed in part by the fact that a
large majority of undergraduates live on-campus
in one of twelve residential colleges. These colleges
are simply a basis for making dormitory life more
attractive by providing a sense of belonging to
well-identified units (Berkeley College, Silliman
College, etc.) that are the centers of much of the
social life of students, the intramural sports pro-
gram, and numerous informal activities in the fine
arts and the performing arts. The residential
colleges are not organized by disciplines; in fact,
considerable effort is expended in insuring that

each college includes students with varied aca-
demic and extracurricular interests. Faculty
members are also associated with these units, a
few as residents, a few as occupants of offices in
the colleges, a few as instructors in a limited
number of general-interest seminars, and a larger
number who take some of their meals in the college
dining halls. The results of this system include a
healthy mixing of students with varied interests
and the encouragement of informal student-faculty
Thus chemical engineering students at Yale
live in an atmosphere in which a majority of their
classmates are majors in the humanities. This ar-
rangement, the requirement of 12 term courses in
the humanities, and the general educational value
of studies in mathematics, science, and chemical
engineering provide them with an opportunity to
achieve a truly liberal education. As noted above,
most of our graduates enter the chemical engineer-
ing profession, which we regard as among the
most demanding and most satisfying of all pro-
fessions. We are inclined to regard those students
who go on to law, business, medicine, and other
professions as tributes to the breadth of chemical
engineering education rather than to bemoan the
loss to our own profession.

O UR CURRENT graduate program emphasizes re-
search and the PhD degree. This is a reason-
able allocation of resources in view of the fact
that current faculty members are research-
oriented and few in number. Topics for graduate
courses are selected in part on the basis of their
suitability as a preparation for research and they
are taught by research-oriented faculty members.
Current regular course offerings include chemical
engineering thermodynamics, chemical reaction
engineering, transport processes, separation pro-
cesses, spectroscopic surface analysis, and electro-
chemistry fundamentals and applications. Courses
in combustion science and technology, aerosol
science and technology, biochemical separation
processes, chromatography, heterogeneous cataly-
sis, and other topics are also offered periodically.
Because some of our graduate students enter with
undergraduate majors in subjects other than
chemical engineering, some upper-level under-
graduate chemical engineering courses are used
as mezzanine courses, i.e., graduate students can
take them for credit by satisfying requirements
in addition to those imposed on undergraduates.


At this time about 20 graduate students are in
residence, most of whom are studying for the
PhD degree.
Most of the graduate courses contain subject
matter relevant to development and design as well
as research. They are therefore appropriate for
students interested in a terminal M.S. degree and
careers in development and design, and we have
a few such students. In order to provide these
students with the opportunities that should be
available to them, however, we recognize that we
need to add courses in materials, computer-aided
design of separation processes, chemical process
control and optimization, and advanced chemical
process design. We are currently discussing such
possibilities, realizing that offering these topics
would require additional personnel, some of whom
might well be adjunct faculty members from in-
Yale's experiment with a Department of Engi-
neering and Applied Science was successful at the
graduate level, and we have retained the title
Engineering and Applied Science for our gradu-
ate program in collaboration with the Sections of
Applied Mechanics (later to be Mechanical Engi-
neering), Applied Physics, and Electrical Engi-
neering. This arrangement provides some signifi-
cant benefits. Courses common to these disciplines,
including applied mathematics and experimental
methods, are readily available to our students, as
are courses in electronics, control systems, fluid
mechanics, and a variety of other topics. Students
are provided with greater flexibility in their
choices of research topics, and interdepartmental
barriers to collaborative research programs simply
do not exist.
Course requirements for the PhD degree
depend on a student's background and interests.
A committee of three faculty members works with
each student to select courses that serve to advance
the student's knowledge of chemical engineering
and to prepare him or her for research. Some-
time during the first year or early in the second
year each student begins the process of develop-
ing a research topic by enrolling for research
with particular faculty members. The student is
then expected to collaborate with one or more
faculty members in developing a proposal for re-
search on a topic for which funding and equip-
ment are available or can be obtained.
Chemical engineering laboratories in Mason
Laboratory and Becton Center are lively with
the activities of faculty members, graduate

students, undergraduates, and postdoctoral re-
search associates. Expenditures on research from
grants and contracts amount to about $800,000
for the 1981-82 academic year. The varied re-
search interests of current faculty members are
described briefly below.
John Fenn applies molecular beam methods to
the study of a variety of scientific and techno-
logical problems. His well-equipped laboratory
provides the means for studying the distribution
of translational, vibrational, and rotational
energies of molecules during and after such events
as free jet expansion, collisions in the gas phase,
and collisions with surfaces. The results are of
interest in analyzing energy distributions in
heterogeneous catalysis processes, monitoring of
pollutants from internal combustion engines,
understanding the infrared radiation character-
istics of rocket exhaust plumes at high altitudes,
and elucidating structures and reactions in messy

Photo by William K. Sacco
S. P. Venkatenshan, Subbarao Ryali, and John Fenn in
the molecular beam laboratory.

mixtures such as biological fluids and coal-con-
version process streams.
Dan Rosner is engaged in research on mass
and energy transfer at fluid-solid and fluid-fluid
interfaces. His laboratories are equipped for
studies of nonequilibrium multiphase systems
under extreme conditions (high temperatures,
partially dissociated gases, particle-laden gases).
Current studies include the deposition of soot, ash,
and inorganic salts from combustion gases to heat
exchange surfaces and turbine blades, surface-
catalysed combustion, and the role of thermal
diffusion in high-temperature processes.
Gary Haller's research interests are in surface


chemistry and heterogeneous catalysis. Current
studies include reactions of olefins on chromium
oxide catalysts, analysis of binary alloy catalysts
using X-ray, photoelectron, chemisorption, and re-
action kinetics methods, and Fourier transform
internal reflection infrared spectroscopy applied
to metal-support interactions.
Csaba Horvath's research interests are in bio-
technology with particular regard to biochemical
separation processes and enzyme reactors. With
Wayne R. Melander, Research Associate and
Lecturer in the Department, he studies the
fundamentals of chromatography and the thermo-
dynamics of adsorption on non-polar surfaces.
High performance liquid chromatography is used
in fundamental studies of the interaction of bio-
logical substances with surfaces and the develop-
ment of linear free energy relationships of techno-
logical significance. In collaboration with phy-
sicians at Yale and at Roswell Park Memorial
Institute in Buffalo he is investigating the use of
hollow fiber enzyme reactors in extracorporeal
shunts in clinical applications. Other applications
of enzyme reactors are in the production of
precious biochemical substances and in food pro-
cessing. He is also studying the design and op-
timization of high performance displacement
chromatography, a potential industrial process.
Paul Nordine's research interests are in high-
temperature heterogeneous reaction kinetics. He
has developed techniques for jet levitation of solids
and laser heating and applied them to studies of
the preparation of crystals, reductive chlorination
of metal oxides, and fluorine-resistant refractory
Bret Halpern is also interested in the dynamics
of chemical reactions catalyzed by solid surfaces.
His studies on partitioning of chemical reaction
energy within product molecules and energy
dissipation in catalysts, monitoring of the depo-
sition of hydrocarbons and carbon on metals, and
oxidation of carbon at high temperatures comple-
ment John Fenn's interests. He is now engaged in
expanding his interest in research on electrochemi-
cal processes.
Charles Walker has collaborated with social
scientists during the past few years in studies of
social and political aspects of problems in environ-
ment and energy. He has worked with the Depart-
ment of Energy on technology assessment and is
currently working with colleagues on a book about
social and political problems in radioactive waste

Calculus 3
Ordinary Differential Equations
with Applications 1
Partial Differential Equations
with Applications 1
Comprehensive General Chemistry
(with laboratory) 2
Organic Chemistry 2
Physical Chemistry 2
Advanced General Physics
(with laboratory) 2
Computer Science 1
Introductory Thermodynamics 1
Introduction to Chemical Engineering 1
Chemical Engineering Thermodynamics 1
Fluid Mechanics 1
Energy, Mass, and Momentum
Transport Processes 1
Chemical Kinetics and Chemical Reactors 1
Separation Processes 1
Chemical Engineering Laboratory 1
Chemical Engineering Process Design 1
Chemical Engineering Projects 1
Technical electives in Engineering 2
Humanities courses 12

Subbarao Ryali uses molecular beam tech-
niques to study energy exchange translationall,
rotational, and vibrational relaxation processes)
and energy transfer (translation to rotation and
vibration) in gas-gas encounters under well-
defined conditions. Other research interests include
gas-surface interactions, nucleation and condensa-
tion, heat transfer, and combustion.

T HE COURSE of chemical engineering education
is influenced by the institutional setting in
which it occurs as well as by professional societies
and the needs of employers of chemical engineers.
There is, of course, a need for some degree of uni-
formity in the teaching of this discipline, but
chemical engineering education can and should
reflect the variety that is, as noted above, the
hallmark and the strength of higher education in
this country. As do other educators, we at Yale
seek to develop courses, curricula, and research
programs that are compatible with the long-range
interests of the chemical engineering profession
in its service to society and compatible with our
own institutional setting. O



J. J. CARBERRY, Yale '57
University of Notre Dame
Notre Dame, IN 46556

?. 'aGdiwn /k4s (1911-1971)

"Men," he remarked in the wake of one of our
disastrous exam performances, "once again the
Fourth Law is manifest-that Law in its original
form states, you will recall, that there is always
a parking spot on the other side of the street."
In an era which enshrines the illiterate, the
"you know, man" cult of the witless who equate
spasms with thought, he, R. Harding Bliss stands
out, even in death, as a giant.
He taught from a wheelchair, a consequence
of polio which was permanently visited upon him
in the prime of his career. Yet that terrible in-
firmity failed to dampen his wit, charm, dedica-
tion to teaching and research. His was perhaps
one of the first courses in Chemical Reaction
Engineering to be offered in this country, if not
in the world. It was a joy to participate in that
offering, in spite of our all too frequent encounters
with his Fourth Law. We did not "take" his
course; rather we participated, for such was his
zeal and humility that he wisely fashioned that
course as an intellectual adventure marked by
bilateral exchange-yet that classroom democracy
never descended to the thought-barren level of a
"rap" session. His was a classroom not a sandbox.
It was a tough, demanding and therefore a most
exciting arena. This was happily so because he
was a tough, demanding and therefore a most ex-
citing mentor. We loved him for those marvelous
qualities enriched as they were by his wit and
obvious love of his students. Such was that love
that his departure from our midst ends not the
grand affair. We love and are continually inspired
by the memories. And I, for one, am seized by the
intuitive vision of that grand gentleman of Yale
now reminding Plato of the Fourth Law, to the
everlasting joy of Dante and Bertrand Russell.


J. J. Carberry, Professor of Chemical Engineering at the University
of Notre Dame since 1961, received his doctorate at Yale in 1957,
following undergraduate work at Notre Dame ('50) for which G. I.
Bill-sponsored undertaking he prepped at Brooklyn Technical High
School and, at FDR's invitation, in WWII. He spent six years with the
du Pont Company, a few years at Cambridge (NSF-Sr. Fellow-1966;
Churchill Fellow-1979/1982); was Senior Fulbright Fellow (Italy 1974)
and in spite of his Irish-American heritage was elected Fellow of the
Royal Society of Arts in 1980. Thus his is now an Anglo-Irish in-
In 1968 he received the Yale Engineering Association Award for
the Advancement of Pure and Applied Sciences and in 1976 was
recipient of the R. H. Wilhelm Award (AIChE) in Chemical Reaction
Engineering. He is a member of the Advisory Council for Chemical
Engineering at Princeton University, an appointment apparently in-
spired by Nassau Hall's fond expectation that Princeton may now
have fond hopes of winning a football game against Yale.*
Author of the text "Chemical and Catalytic Reaction Engineering"
(McGraw-Hill) and co-editor of "Catalysis Reviews-Science and Engi-
neering" (M. Dekker, Inc.), Carberry is now contemplating co-author-
ship with Aris McPhearson Rutherford of a seminal opus "Isaac
Newton's Indebtedness to the Gill Report-Historical contrasts in
Mercury Poisoning."
*They did.

B. 0. 5obdze (f95-1972)

We were accustomed to setting our clocks not
by radio nor the Bureau of Standards, but by
Barney Dodge's arrival at Sterling Chemistry Lab,
Yale. A more secure standard did not exist. His
scholarly standards were as precise as his office
hours, and as consistent. His method of teaching
leaned heavily upon the "case method." While
some might question that as a philosophy of teach-
ing, none of us, in retrospect, can question the


Copyright ChE Division. ASEE, 1982

merits of Barney's "case method"; even the
kineticists in our midst matured via Dodge's
"Chemical Engineering Thermodynamics."
In this present age of devotion to "student-
teacher evaluations," I doubt that Barney would
fare very well-unless the students were required
to render their assessments of him five or ten
years after having suffered through his lectures.
They might then, as I have, realize that their
suffering was not in vain-indeed the fruits of
our labors are great. For Barney imposed realities
upon us while maintaining scholarly rigor with re-
spect to the principles of chemical engineering; in
particular, thermodynamics. And although his
text on that subject is a classic, his research inter-

ests were catholic. Before absorption and simul-
taneous chemical reaction was formally ac-
knowledged, B. F. Dodge directed seminal re-
search in that area.
He was a precise and candid man, virtues hope-
fully still with us. Should we become devoid of
these merits, it is solely because we suffer a
paucity of great men such as B. F. Dodge. While
we mourn his absence, his presence will not be
forgotten nor will our love of him diminish.
I have no doubt that he arrived in the here-
after precisely on time and immediately proceeded
to remind Plato of the first Three Laws to the
everlasting joy of, amongst others, his great
friend, R. Harding Bliss. 0

book reviews

By Aksel L. Lydersen
John Wiley and Sons, 1979; 357 pages, $53.95
Hardbound, $22.50 Paperback

Reviewed by Kenneth J. Bell
Oklahoma State University
This book surveys a wide variety of subjects
in fluid flow and heat transfer; in addition to the
more obvious topics, there are chapters on Particle
and Drop Mechanics, Liquid Filtration and Flota-
tion, and Atomization, Dispersion, Homogeniza-
tion, Crushing and Grinding. There is also a short
chapter on Energy Economy. The general level of
the treatment is at what might be termed the first
professional level: these are the pieces of informa-
tion and the equations that would be needed by a
process engineer carrying out preliminary plant
design. The need is to get reasonable answers to
a wide variety of problems quickly, leaving the
detailed design to be worked out later by special-
Little space is spent developing anything that
might be considered a theoretical base if it does
not contribute directly and immediately to
problem solving. On the other hand, all
working equations are there together with the
necessary charts, tables, and nomograms to permit
complete and consistent calculation of the answer
required. There is also enough description of the
various types of equipment to allow the non-
specialist to make intelligent selections. Also, there
are numerous completely worked-out examples

(some of them of considerable complexity) which
well illustrate the proper use of the design equa-
tions. Finally, the author includes a number of
comments concerning points frequently overlooked
or misunderstood by designers. If the book has a
technical weakness, it is that the references that
are given tend to be quite venerable so that anyone
seeking additional information is going to be about
ten years out of date.
So much for the technical content of the book.
Where does it fit into the engineering curriculum?
This is not so easy to answer. The book will not
do for the introductory courses in fluid mechanics
and heat transfer because of the almost total lack
of presentation of fundamental material and the
derivations of the working equations. It would
perhaps be suitable for those few curricula which
have advanced applied courses in these topics, but
the faculty member would want to do a lot of up-
dating with recent literature.
The book would be an excellent supporting
volume for an undergraduate (or even a gradu-
ate) design course, but it cannot take the place
of one of the books specifically oriented towards
that topic (e.g., Peters and Timmerhaus). It is
doubtful that it would be fair to expect a student
to pay as much money as this for a purely sub-
sidiary reference book, especially since much of
the material in this book is to some extent covered
in Perry's Handbook.
If it is hard to see where it fits into the chemi-
cal engineering curriculum, it is easy to recom-
mend this book to the practicing engineer, es-
pecially one just beginning his career in process
Continued on page 144.


Braun & Co
is a world leader
in the engineering and
construction industry. For
more than 70 years, we have
provided a wide range of services to the
process and power industries.
Our principal fields of activity are
chemical and petrochemical plants, oil
refineries, ore processing plants, coal
gasification facilities, and power

ating sta-
tions. Many of our
projects have been first-
of-a-kind, utilizing new
processes never before employed
on a commercial scale. We also have been
involved in the emerging synfuels
Our rapid growth has opened up many
challenging opportunities and
assignments for professional growth.
Positions are available at our engineering
headquarters in Alhambra, California, and
at our eastern engineering center in
Murray Hill, New Jersey.






Pennsylvania State University
University Park, PA 16802

SN THE SUMMER OF 1978, while cleaning glass-
ware in Room 3 of Fenske Laboratory, one of
us (THL) observed that the level of water in the
laboratory sink would periodically drop to a near
empty level and then slowly climb back to a moder-
ate height-all in the presence of what appeared
to be a steady inflow of water from the faucet.
These curious events, although memorable, were
not accorded any particular significance until the
following winter. At that time, THL attended a
graduate Process Dynamics Course in which the
subjects of Autonomous Oscillations and Bifurca-
tions were presented in some detail. In this course,
it was emphasized that systems with steady inputs
(autonomous systems) could produce oscillatory
outputs under appropriate conditions, and various
analytical and computer exercises, mainly in-
volving continuous stirred tank reactors, were
worked out while a series of papers from van
Heerden [1] and Bilous and Amundson [2] through
Uppal, Ray and Poore [3] and Schmitz, Graziani
and Hudson [4] were studied. This experience pro-

Copyright ChE Division. ASEE,. 1982

Thomas H.
in Chemical Er
1979. He is cu
a production s
Inc., Groton, C
Daniel J.
B.S. in Chemic

vided a framework for interpretation of the
strange events observed in Room 3. In fact, THL
may have been the only student in that class who
regarded autonomous oscillations as other than
an academic hoax, for although experimental
demonstrations had been alluded to in lectures,
(perhaps) only he had actually observed them.
To further satisfy his curiosity, as well as the
term paper requirement for his Process Dynamics
course, THL conducted primitive experiments on
the sink in Room 3. He measured level as a
function of time with a meter stick and stop
watch at various faucet flow rates which were
measured with a graduated cylinder. He observed
steady levels at low flow rates, oscillatory levels
at intermediate flow rates and a return to steady
levels at high flow rates. The oscillatory states
were not truly periodic as the amplitudes and
periods appeared to be randomly distributed, but
over a fairly narrow range. THL was inclined to
describe the oscillations as "chaotic" in light of
the paper by Schmitz, Graziani and Hudson [4].
The source of the strange oscillations was not
obvious, but the U-shaped trap in the drain line
at the bottom of the sink was highly suspect.
To gain better insight into the oscillatory phe-
nomena, a transparent model of a sink and drain
was constructed, and a more exhaustive and

Leise earned his B.S.
igineering from The
State University in
rrently employed as
supervisor by Pfizer,
onnecticut. (L)
Jenkins earned his
al Engineering from

The Pennsylvania State University in
1980. His presentation, "The
Oscillating Sink," won the Judges
Award for originality in research
at the 1980 Mid-Atlantic AIChE
Spring Conclave at Drexel Uni-
versity. He is presently working
toward his M.S. in Chemical Engi-

, .
- \

, c~iSl

neering at Carnegie-Mellon University. His graduate research deals
with synthetic fuels wastewater treatability and reuse. (C)
John M. Tarbell earned his Ph.D. in Chemical Engineering from
the University of Delaware in 1974. He is currently an Associate Pro-

fessor of Chemical Engineering at The Pennsylvania State University.
His research interests are in process dynamics and stability, cardio-
vascular fluid dynamics, and chemical reactor modeling. (R)


Our main purpose in communicating this experience is to
suggest a simple system, adaptable to a classroom or laboratory environment,
which demonstrates some basic dynamic phenomena (instability, bifurcation, and oscillation).

accurate set of data was obtained by another of
us (DJJ). We were amply rewarded for this effort
as additional curious phenomena attendant to the
oscillations were soon uncovered.
The apparatus, procedures and many of the
results of these experiments are described in the
following sections. Our main purpose in com-
municating this experience is to suggest a simple
system, adaptable to a classroom or laboratory
environment, which demonstrates some basic dy-
namic phenomena (instability, bifurcation, and

A schematic of the experimental apparatus is
shown in Fig. 1. A small, translucent, plastic
holding tank was fitted with a drain which con-
sisted of an elbow and two short pieces of metal
tubing (~ .375" i.d.) on the inside of the tank,

FIGURE 1. Experimental Apparatus.

with a length of flexible, transparent (Tygon)
tubing (-- .375" i.d.) on the outside. A bend of
variable height (X) was maintained in the exit
line with several clamps, and the line drained
directly into a bench top sink. For most of the
experiments, the drain inlet was oriented perpen-
dicular to the tank bottom and open upward. An
electronic pressure transducer was connected to
the side of the tank below the level of the drain
by means of a short piece of plastic tubing, and
the instantaneous pressure was continuously re-





S10 20 30 40 50 60 70 80
FLOW (ml/sec)

90 100

FIGURE 2. Height Versus Flow Rate (X = 0.0 cm.),
Smooth Curve is from Mechanical Energy
Balance with Friction Losses for Smooth
Tubing and Bends. Dashed Curve Includes
Drain Losses.

corded and assumed to be indicative of the in-
stantaneous fluid level. Metered water flow was
admitted to the tank through plastic tubing
(, .375" i.d.) with the inlet maintained below
the drain level in order to minimize splashing and
associated noise.
Steady inlet flows were varied over a range
from 0.5 ml/sec to 90 mi/sec resulting in Reynolds
numbers in the outlet line up to 9,280 under
steady level conditions. The bend height (X) was
varied between 0.0 cm and 90.0 cm to provide
conditions of: no bend, bend below the drain, and
bend above the drain.


A baseline case with no bend in the exit line
(X = 0.0) was characterized by the data shown
in Fig. 2. The ordinate is the height of the tank
level above the drain and the abscissa is the steady
inlet flow rate. The tank level was steady for all
flow rates, but a rather marked steady state
transition (bifurcation) occurred at -- 65 ml/sec
and a more subtle one at -- 3 ml/sec. At flow rates
between .- 3 ml/sec and 65 ml/sec, the exit
line was in a two-phase (air/water) flow con-
dition, while the flow became single-phase above
65 ml/sec. In the two-phase flow regime, there
was a swirling vortex above the drain which
sucked air into the exit line. As the flow rate was




S I **"*
10 20 30 40 50 60 70 80
FLOW (ml/sec)

9o0 o00

FIGURE 3. Height Versus Flow Rate (X = 41.4 cm.).

increased in this regime, the vortex contracted
leading to diminished entrainment of air. As the
flow rate was reduced below p- 3 ml/sec, two-
phase flow subsided and a slug of fluid remained
stationary in the bottom of the U in the exit line
with fluid trickling into and out of the slug.
The smooth curve in the high flow rate regime
is based on the Mechanical Energy Balance in-
corporating friction factors for the smooth tubing
and bends. When a loss coefficient of 2.0 is as-
signed to the drain, the dashed curve, which fits
the single-phase flow data quite well, is obtained.
Clearly, it is only in the high flow rate regime
that the simple tank draining model of elementary
process control [5] applies.
Fig. 3 shows the results of experiments with
a 41.4 cm bend in the exit line. The vertical lines
in the 4-25 ml/sec range represent oscillatory
states. The oscillations were not truly periodic
(perhaps chaotic?-see Fig. 4), and the heights
of the vertical lines represent only average ampli-
tudes. There are four bifurcations ( changes in
qualitative behavior) apparent in Fig. 3:
(1) Trickling steady states (with a stationary slug
at the bottom of the U) transform to large ampli-
tude oscillatory states at ~ 3 mi/sec;
(2) Large amplitude oscillatory states transform to
small amplitude oscillatory states at 12 mi/sec;
(3) Small oscillatory states become two-phase flow
steady states at 25 mi/sec; and
(4) Two-phase flow steady states become single phase
flow steady states at ~ 65 mi/sec.
The observed physical mechanism of large
amplitude oscillations is summarized in Fig. 5.
Consider that an empty tank (and exit line) is
being filled up, and the tank level has just reached
the mouth of the drain (A). Fluid spills over the
drain and fills up the U in the exit line as the
level gradually rises (B). Eventually, fluid com-
pletely covers the drain trapping a column of air

in the U (C). As the tank level rises, the air
column is forced further down the tube until it
passes around the bottom of the U. At this point
the buoyant force of the rising air is sufficient
to push fluid over the top of the bend (D). Once
fluid passes over the top of the bend, a siphon is
created and the tank quickly empties until the
level has descended to the drain (E). Now air is
sucked into the drain and a brief period of slug
flow follows (F) until condition (C) is re-
established. At this point the cycle (C-F) repeats
itself. The small amplitude oscillation mechanism
is quite similar. The major difference is that the
trapped air slug is smaller and the emptying is
triggered by fluid passing over the top of the
bend rather than air passing under the bottom
of the U as in (D).
The transition from oscillatory to steady flow
at 25 mi/sec can be understood if we realize that
the siphon flow which drains the tank has a fixed
rate determined by the configuration of the exit
line. If the inflow rate exceeds the siphon flow rate,
emptying is prohibited. The transition at 3 mi/sec

FIGURE 4. Height vs. Time Recording of Typical
Oscillations (X = 21.6 cm., Flow Rate =
16.5 ml./sec.).

is not so easily understood. In fact, with the bend
height set at X = 21.6 cm, oscillatory flows were
observed down to our lowest inflow rate (- 0.5
ml/sec) without a steady state transition (see
Fig. 6). The results of Figs. 3 and 6 suggested
the possibility of bi-stability (two stable flow
regimes) in the low flow rate range, but limited
hysteresis experiments did not reveal a range of
coexisting steady and oscillatory states.
The effect of bend height on flow regimes was
further investigated at X = 5.8, 30.1, 51.1, and
90.0 cm. Oscillations were observed for all bends,
with their amplitudes strongly proportional to
bend height. Steady states always existed at high
flow rates, but with sufficiently high bend height,
the two-phase flow regime disappeared. There
was no obvious pattern in the appearance of low
flow rate steady states. They appeared for X =


0.0, 5.8, 41.4 and 51.1 cm, but not for X = 21.6,
30.1 and 90.0 cm.
In a final desperate attempt to eliminate oscil-
lations with a finite bend, we inverted the drain
to an open downward configuration thinking that
this would eliminate the sucking of air which was
considered vital to the oscillation mechanism. To
our surprise, the siphon effect was strong enough
to suck air around the inverted drain, and we ob-
served oscillations following the mechanism of
Fig. 5. We conclude that the only way to com-
pletely eliminate the oscillatory regime is to com-
pletely eliminate the bend.


If you are interested in practical engineering
applications for our sink, you might consider it
as a primitive level controller when operated in
the two-phase flow regime, where level is nearly
independent of inflow rate, or, as a periodic feed
pump to a batch process when operated in the
oscillatory flow regime.
On the other hand, if you are looking for
theoretical significance, you should realize that it
is an unusual hydrodynamical system in which
intermediate time-dependent states are surrounded
by steady states at both lower and higher values
of a parameter. For example, in pipe flow there is
a transition from steady laminarr) flow to time-
dependent (turbulent) flow as the Reynolds
number is increased, but a return to steady flow
at higher Reynolds numbers is not observed. The
same may be said of the classical Taylor (circular
couette) and Rayleigh-Benard (heated layer)
flows [6]. As the Taylor or Rayleigh number is in-
creased, a bifurcation from steady to time de-
pendent motion occurs, but further parameter in-
crease does not result in steady motion. Inter-

o, t __o

M .o a0. 0^...

FIGURE 5. Schematic of the Oscillation Mechanism.


560 *0 0 0

0 10 20 30 40 50 60 70
FLOW (ml/sec)

90 100

FIGURE 6. Height Versus Flow Rate (X = 21.6 cm.).

mediate time dependent states are readily observed
in chemical reaction systems. For example, a
constant volume continuous stirred tank reactor
(CSTR) always produces stable steady states at
sufficiently high and low values of the flow rate
[7, 8], but time-dependent states may occur at
intermediate flow rates [4].
Finally, if your interests are pedagogical, "the
oscillating sink" provides a simple demonstration
of autonomous oscillations and a variety of bi-
furcation phenomena. As a parting note, the
demonstration value of the oscillating sink can be
improved by the use of a smaller diameter tank
than we have shown in Fig. 1. This will result in
shorter oscillation periods (we observed periods of
several minutes) and a more dynamic demonstra-
tion. The tank diameter should not affect the
oscillation mechanism. O

1. van Heerden, C., Ind. Eng. Chem., 45, 1242 (1953).
2. Bilous, 0., Amundson, N. R., A.I.Ch.E. J., V, 513
3. Uppal, A., Ray, W. H., Poore, A. B., Chem. Eng. Sci.
29, 967 (1974).
4. Schmitz, R. A., Graziani, K. R., Hudson, J. L., J. Chem.
Phys., 67, 3040 (1977).
5. Coughanowr, D. R., Koppel, L. B., "Process Systems
Analysis and Control," McGraw-Hill, New York
6. Fenstermacher, P. R., Swinney, H. L., in "Bifurcation
Theory and Applications in Scientific Disciplines,"
Eds., Gurel, O. K., Rossler, O. E., Ann. N.Y. Acad. Sci.,
316 (1979).
7. Plau, F., Tarbell, J. M., Chem. Eng. Commun., 4, 677
8. Gavalas, G. R., "Nonlinear Differential Equations of
Chemically Reacting Systems," Springer-Verlag, New
York (1968).



0 class and home problems

The object of this column is to enhance our readers' collection of interesting and novel problems in
Chemical Engineering. Problems of the type that can be used to motivate the student by presenting a
particular principle in class or in a new light or that can be assigned as a novel home problem are re-
quested as well as those that are more traditional in nature that elucidate difficult concepts. Please sub-
mit them to Professor H. Scot Fogler, ChE Department, University of Michigan, Ann Arbor, MI 48109.


Georgia Institute of Technology
Atlanta, GA 30332

A certain chemical reaction is being carried
out in the aqueous phase in a 2000-gal. reactor.
The reactor is half-full and its contents are
maintained at the reaction temperature of 100'F.
At a certain point in the reaction, it is desired
to quench the reaction as rapidly as possible. For
this purpose, the rapid dumping of 2000 lbs. of 1-
inch ice cubes (at 32F) into the reactor contents
has been proposed. Calculate 1) how long it will
take for all of the ice to melt and 2) what the final
solution temperature will be at this time.
Prior laboratory experiments under similar
conditions of agitation have shown that the overall
coefficient for heat transfer at the ice surface is
200 BTU/hr.-ft2- F. The latent heat of fusion of
ice is 144 BTU/lb., while its density is 57.5 lbs./
ft3. The physical properties of the aqueous solution
may be assumed to be the same as those for water.
A sketch of this system is presented in Fig. 1.

FIGURE 1. Sketch of the physical system in the melt-
ing ice cubes problem.

Jude T. Sommerfeld has been a professor of ChE at Georgia
Tech since 1970. He teaches courses on process control, distillation,
reactor design and process design, and his research interests include
energy conservation. He has also served as a consultant to numerous
industrial organizations. Prior to 1970 he had eight years of engi-
neering and management experience with the Monsanto Company and
BASF-Wyandotte Corp. Dr. Sommerfeld received his B.Ch.E. degree
from the University of Detroit, and his M.S.E. and Ph.D. degrees in
chemical engineering from the University of Michigan. (L)
Peter A. Minderman Jr. received his B.S. from Georgia Tech in
1979 and MS from the same institute in 1981. His prior industrial
experience includes work with DuPont and Tennessee Eastman Co.,
a division of Eastman Kodak. He has recently returned to Tennessee
Eastman and currently works in the Process Systems Engineering
group. (R)

The basis of the solution consists of an
enthalpy balance around the solid (ice) phase. The
liquid phase at the melting temperature of the
solid phase is chosen as the enthalpy reference
state. With this choice, the specific enthalpy
(BTU/lb) of the liquid phase is equal to cp (T-T,).
Similarly, the specific enthalpy of the solid phase
becomes the negative of the latent heat of fusion
(-X). Essential to the foregoing statement is the
assumption that the temperature of the solid phase
is always at its melting point. The rate of heat


Copyright ChE Division, ASEE, 1982

$ ~

input (BTU/hr) to the solid phase is then
UA (T T,)
while the rate of accumulation is
There is no heat output from the solid phase as
well as no generation or consumption terms.
Hence, the differential enthalpy balance equation

UA (T T,) = -X (1)
It is assumed that the ice cubes are truly
cubical in shape and further that they remain
cubical as they melt. That is, there exists no
preferential melting at any one face of the cube.
Assuming that all of the cubes behave identically,
the total number of cubes then also remains
constant as long as ice is present. It then follows
from these assumptions that


W, = pNx"-
A = 6Nx2

The result of substituting these qualities into
Eq. (1) is
dx (T T,) (2)
dt px
It remains then to relate the temperature of
the solution (T) to the size of the melting ice
cubes (x). This is achieved by performing
material and enthalpy balances around the com-
plete system. The material balance is

W, + W = W + Wo, = Wt (= const)
Wo = Wt W, (3)

And assuming adiabatic operation, the enthalpy
balance is
H, + H, = Ho, + Ho, = H, (= const)

Ht + XW,
T -T, Ht +W (5)
c, (Wt- W) (
Finally, substitution of the above result into Eq.
(2) yields

dx -2U Ht + XW.
dt pX p (Wt Ws)

dx -2U (Ht/XpN) + x
dt pc (Wt/pN) x (6)

Before proceeding to the integration of Eq.
(6), it is convenient to make the following
'p XpN

dx P3 p+ X3
dt x (7)

If one now defines a new dependent variable as

the result upon insertion into Eq. (7) is

dy a 1 +y3
dt y p 77,8 -_ y

And finally with the following additional defini-


a -= '3

H, = Ht- H,

Ho = WO, c, (To T)
H, = -XW%
H, W, c, (T T,)
H, =-XW,

Ht = W, c, (To-T,) -XW,

W,, c, (T- T,) = Ht + XW, (4)
Substitution of Eq. (3) for W, into Eq. (4) and
rearrangement of the result leads to

one has

dy 1 + y3
dy -c -1 (8)
dt a y3 (
The integration of Eq. (8) is given in any
comprehensive table of integrals. The result is

c(a+1) 1 (1 + y)2
3 (2 1 -y + y2

+ V3 TAN-1 2y- (9)
+33 VN /JYo





yo =

For the particular case wherein the upper limit
of integration is zero (that is, complete melting of
the solid phase), Eq. (9) assumes the following

a+1 1F (1+yo)
ct (y = 0) a+In (1 + y)
3 2 1 yo + yo'

+ V3 TAN-1 -y3 yo)-y
(2 yo


for which the following trigonometric identity
was invoked

TAN (u-v) =

TAN (u) -TAN (v)
1 + [TAN (u)].[TAN (v)]

The original problem concerned the quenching
of 1000 gallons of an aqueous solution originally
at 100F. The physical properties of the solution
were assumed to be the same as those for water.
It was proposed to dump 1 ton of 1-inch ice cubes
into the solution at quench time. The various
problem parameters then assume the following
Wt = (1000) (8.33) + 2000 = 10,330 lbs
HI = (1000) (8.33) (1.0) (100 32)
(144) (2000) = 278,000 BTU
= (2000) (12)3
(57.5) (1.0)3 =60,100
(2) (200)
a (57.5) (1.0) 6.96 ft/hr

Pf3 = 27,0- 0.558.10-3 ft3
(144) (57.5) (60,100)
p = 0.0823 ft
S= 0,0 = 2.995.10-3 ft3
(57.5) (60,100)
y = 0.1441 ft
a 0.558.10-3
c = 84.6 hr-1
yo = 1.012
(12) (0.0823)
Substitution of the last three quantities into
Eq. (10) then results in the following time re-
quired for complete melting of the ice (solid

w ,


FIGURE 2. Time histories for the size of the ice cubes
and the liquid-phase temperature.

t (x = 0) = 0.0514 hr = 3.08 min.
The temperature of the quenched aqueous phase
at this time is given via algebraic solution of Eq.
(5) with W, = 0; the result is 58.90F.
Complete time histories for the size of the ice
cubes and the liquid-phase temperature for this
particular system are presented in Fig. 2. D


a y/lP
A total surface area of unmelted solid
phase, ft2
c --a//
c, heat capacity of liquid phase, BTU/lb-F
H enthalpy, BTU
N total number of solid-phase cubes
t time, hr
T temperature, F
U overall coefficient for heat transfer from
the liquid phase to the solid phase,
W mass of material, lbs
x size of a solid-phase cube, ft
y x//
a 2U/pc,
Pj 3VHt/pN
y 3yWt/pN
X latent heat of fusion of solid phase,
p density of solid phase, lbs/ft3

1 liquid phase
o initial condition (t = 0)
s solid phase
t total (sum of liquid + solid phases)

o initial condition (t = 0)






O if --EN

Edith Whatley,
Industrial Hygienist,
Texas Division,
Dow Chemical U.S.A. J
(M.S. Chemistry, |
University of Virginia)

"Choosing the right offer was a big
decision. But the more I heard about
Dow, the easier it became.
"I was impressed by how big Dow
is. But more impressed when the
interviewer told me Dow decentrali-
zation keeps you from feeling lost
in a corporate maze. He told me that
I'd be on a first-name basis with
almost everyone in my department.
"I'd also be able to contribute right
away. Because Dow thinks you learn
best by doing, and not in some formal
training program. So you get hands-

on experience from your first day
on the job.
"Plus, they encourage movement
between divisions, departments, and
even functions. So I'd get a wide
range of experience-and a chance
to find the job that's right for me.
"But I guess the main reason I
chose Dow over any other job was
that I heard of Dow's strong commit-
ment to attracting the best people.
And to giving these people the
chance to develop and grow.
"I am developing. I am growing.

I'm pleased that I chose Dow."
If you know of qualified graduates
in engineering or the sciences, or
with an interest in marketing, finance
or computer science, we hope you
will encourage them to write us:
Recruiting and College Relations,
P.O. Box 1713-CE, Midland, Michigan
48640. Dow is an equal opportunity
employer- male/female.
Trademark of The Dow Chemical Company
1981, The Dow Chemical Company '


yft laboratory



New Jersey Institute of Technology
Newark, NJ 07102

use of Direct Digital Control (DDC) in in-
dustry, smaller versions of digital computers
equipped with laboratory modules are being
utilized increasingly in undergraduate Process
Control laboratories to train chemical engineers
in the theory and practice of digital control. It
was pointed out by Waller [1], however, that the
thrust of process control teaching is still directed
toward very basic facts only, and that old text-
books were being used to a large extent. Newer
textbooks and laboratory experiments should,
therefore, emphasize computers and digital
One of the chief reasons for the popularity of
digital computers is the "amplifier-limitation"

i i

M. F. Abd-EI-Bary is an Assistant Professor at the New Jersey
Institute of Technology. He holds degrees from Alexandria University,
MIT, and Lehigh University. He has a strong interest in the area of
digital control. His research work is primarily in the field of water
and air pollution. (L)
Sriram Chari is an MBA student at the Harvard Business School.
He has a B.S.ChE from Indian Institute of Technology, Delhi, and
an M.S.ChE from New Jersey Institute of Technology. He was a Shift
Engineer at Swadeshi Polytex Limited in Ghaziabad, India, before
working on this project as a Teaching Assistant at NJIT. (R)

P --O-P opoL oj Zt 0

FIGURE 1. Schematic of three CSTR's
disadvantage of analog computers. Complex
control systems, such as P-I-D Control, for
example, require a large number of amplifiers,
sometimes more than that which is available in
smaller analog computers. A digital computer can
resolve this difficulty by simulating the control
logic through its software. In the Process Control
laboratory at New Jersey Institute of Technology,
a digital computer has been hooked up with an
analog computer to give students an overview of
the advantages and disadvantages of both com-
puters in a single experiment.
An EAI-20 analog computer was used to simu-
late a system of three CSTR's in series. The first
CSTR is fed with a brine solution stream and a
fresh water stream. The salt concentration of the
third CSTR outlet stream is controlled by regu-
lating the flow of brine to the first CSTR, as shown
in Fig. 1. The original set up involved only pro-
portional control. The stability of the system is
studied by observing its response to a unit step
change in the set point. Writing a material
balance and performing magnitude scaling leads
to the final analog equation:
dt [20Y"] = 6 [20Y"] + 2.4 [100Y']
+ 0.32 [500Y] + 0.16 Ke [2(500Y)-1]

Copyright ChE Division. ASEE, 1982



IaTc1111E A .1ua,

where Y is the Laplace transform of the
controlled variable, and Ke is the Proportional
Controller gain (psi/mv). Fig. 2 shows the analog
patch diagram corresponding to the above equa-
tion. The proportional controller gain (Ke) was
adjusted by means of potentiometer number 15,
set at .016 K,. Marginal stability resulted at a
gain of 8 psi/mv, a value that was close to that
obtained by analytical stability calculations such
as the Routh test.
Recently, the department acquired a Digital
MINC-11 minicomputer equipped with four
laboratory modules: a preamplifier, an analog to
digital (a/d) converter, a clock, and a digital to
analog (d/a) converter. The two computers were
connected so that the use of potentiometer number
15 of the analog computer was done away with,
and the original input to the potentiometer be-
came the input to the digital computer, which
acted as the controller. For proportional control,
the input analog signal was sent in through one
of the a/d channels, converted to a digital signal
by the a/d converter, and then multiplied by a
constant (Kc). The resultant digital signal was
converted to its analog counterpart before the


I ,, S ---

3i -4--

-.. C,
S ~ o -I '-

FIGURE 2. Analog Patch Diagram

latter was sent out to the analog computer as a
substitute for the output of potentiometer number
Addition of other modes of control such as
Manual, Proportional-Integral, P-I-D, and P-D
were made easily. For P-I, for example, the input
signal was integrated by Simpson's Rule and the
necessary additions and multiplications performed.
For P-D control, on the other hand, the input
signal was differentiated. In addition to achieving
different modes of control, the digital computer
was also used to monitor the analog computer

One of the chief reasons for the
popularity of digital computers is the
"amplifier-limitation" disadvantage of analog
computers. Complex control systems, such as P.I.D
Control .. require a large number of amplifiers,
sometimes more than that which is available
in smaller analog computers.

output. Another channel of the analog/digital
converter served this purpose.


The frequency of sampling of the input analog
signal was made as high as possible by using a loop
in the BASIC program similar to a DO loop in
FORTRAN. The clock module in the digital com-
puter could have been used to obtain even higher
sampling rates, but its disadvantage is that pro-
cessing of the already collected data cannot simul-
taneously take place. Thus, the clock will allow
the collection of 50 sample points at a very high
rate, but will then wait for these data points to be
processed and for the output signal to be sent out
before collecting another 50 sample points. Be-
tween two processing enough time elapses to
render a fast process unstable.
In the sampling method adopted in this experi-
ment, one sample point (V2) was collected at the
start of each loop, and used along with the last
sample point (V,) to perform differentiation and
other operations. On the basis of these operations
an output signal was formulated and sent out, and
the next loop began with the collection of a new
data point. This new point was stored as V, and
the old value of V2 transferred to V1. This way,
the two latest sample points are processed in each
loop. A sampling rate of approximately one sample
every 0.12 seconds could be achieved by this
method. In addition, a PAUSE statement, was in-
corporated in the program to introduce a time
delay, and thus vary the sampling rate. Therefore,
the effect of the sampling frequency on the system
response could also be studied.

With the analog and the digital computers
hooked up together, the digital computer was
started first with the values of K,, Integral Time
(-1), and Derivative Time (To) specified. At a
preset time the analog computer was put on



.. -- system response
,.set point

Proportional Controller Gain =4 psi/mV
Sampling Interval =. seconds


S..system response
S' set point

Proportional Controller Gain 4 psi/mV
Sampling Interval =.25 seconds


,system response set point
set point

Proportional Controller Gain (Kc)=4psi/mV
Derivative Time (Td)= .I seconds
Sampling Interval = .12 seconds


S-system response

set point

Proportional Control Gain= 4 psi/mV
Integral Time = 5 seconds
Sampling Interval =.12 seconds


"Operate" mode. The response of the system
could then be monitored on the video terminal. A
Textronix Hard Copy Unit was used for immedi-
ate reproduction of the video projection.


The most striking (and only apparent) dis-
advantage of digital control is that, at best, the
sampling frequency only approximates continuous
analog sampling. This disadvantage is very pro-
nounced in controlling fast processes. For example,
if the residence time in each CSTR is 0.5 minutes,
the proportional controller gain (Kc) for marginal
stability is 5.25 psi/mv for a sampling period of
0.1 second, as compared to a value of 8.0 psi/mv
obtained with continuous analog sampling. In-
creasing the sampling frequency to 0.25 seconds
reduced the marginal stability value of Kc to 4.35
psi/mv. The use of discrete sampling through
the addition of digital computer, therefore, de-
creases the proportional controller gain for
marginal stability. Luyben [2] shows how this
addition introduces, in effect, a process lag and
hence increases instability.
This experiment allows the student to get a
better understanding of control theory since the
control is done through obvious software instead
of the (black box) analog controller. Students can
study the effect of varying sampling interval
(through the use of PAUSE statement) on the
performance of the system. By plotting the set
point and the controlled variable on the CRT, the
concept of offset, and oscillatory response can be
shown much more clearly. In Figs. 3, 4, 5, 6, and


7 the lower horizontal line represents the initial
value of the variable, while the upper horizontal
line represents the set point. Fig. 3 shows an off-
set, representing the behavior of the system with
proportional control, and Fig. 4 shows more
oscillatory response as the sampling frequency
was increased. Fig. 5 represents the response with
P-D control, showing better response over pro-
portional control. Fig. 6 shows no offset with P-I
control while Fig. 7 represents the response with

system response

set point

Proportional Controller Gain = 4psi/mV
Integral Time= 5 seconds Derivative Time =.I seconds
Sampling Interval= .12 seconds


P-I-D control. The best response is clearly ob-
tained with P-I-D control.
Another point to be taken into consideration
is that a complex mode of control requires more
amplifiers than can be provided by a small analog
computer. This difficulty can be overcome by hook-
ing up a digital computer together with it. In fact,
there is no end to the modes of control that can
be simulated by a digital computer used in con-
junction with even as small an analog computer
as an EAI-20.

Partial support for equipment purchased was
provided by Exxon. The authors acknowledge
helpful suggestions provided by Prof. E. C.
Roche. O

1. Waller, K. V., "Impressions of Process Control Edu-
cation and Research in the U. S." Chemical Engi-
neering Education, Vol. XV, Number 1, Winter 1981.
2. Luyben, W. L., Process Modeling, Simulation, and
Control for Chemical Engineers, McGraw-Hill (1973).

I Obook reviews

Edited by R. Aris and A. Varma
Pergamon Press, 1980. 830 pages
Reviewed by John H. Seinfeld
California Institute of Technology

Neal Amundson has exerted a profound in-
fluence on the course of modern chemical engi-
neering. As virtually a lone pioneer in the late
1940's and early 1950's, he opened the frontier
of the mathematical understanding of chemical
engineering systems. Although legions have
rushed in behind him, a vast number of the
Chief's own papers remain the seminal milestones
along the path that has led slowly and steadily to
a rigorous mathematical description of chemical
engineering processes. This volume, prepared
with love and care by Rutherford Aris and Arvind
Varma, is a Baedeker for the traveller retracing
that path.
Of over 2000 pages of Amundson's published
papers, Aris and Varma have selected 800 for re-
printing. For those papers not reproduced in
their entirety, leading pages, with the usual ab-
stract, and sometimes pages containing con-
clusions or summarizing statements, are included.
The scope of the papers is impressively broad, in-
cluding major contributions in ion exchange and
chromatography, distillation, chemical reactor
stability and control, polymerization reaction engi-
neering, the modeling of fixed and fluidized bed
reactors, steady state uniqueness and multiplicity
of catalyst particles and chemical reactors, and
the combustion of single carbon particles.
The modeling of physical systems is, as Aris
has noted, an art and a craft, and Neal Amundson
stands as the senior artist and craftsman in the
modeling of chemical engineering systems. In his
analysis of complex systems Amundson has al-
ways sought the mathematical description that
captures the fundamental essence of the system's
behavior, the art of selecting the proper balance
between simplicity and reality. Once a model has
been chosen, the elucidation of its properties can
be attacked with all the heavy machinery of
mathematics, the craft of the modeler. This
Continued on page 125.




Stanford University
Stanford, CA 94305

SEW WOULD DISPUTE THE obligation of depart-
ments of chemical engineering to provide
students with a sound technical education in the
fundamentals of thermodynamics, heat, mass and
momentum transfer, separations processes and
chemical reaction engineering. That schools in
this country have been successful in doing so is
indicated by the high level of industrial compe-
tition for new graduates. However, despite the
facility with which a young working engineer per-
forms design calculations or process analyses,
such efforts will prove fruitful only if they are
communicated effectively to others. In fact, in a

*This paper originally appeared, in somewhat different
form, as an article in the 1982 Compendium on Engineering
Laboratory Instruction.

George M. Homsy is a Professor
of Chemical Engineering at Stanford
University. He did his undergradu-
ate work at the University of Cali-
fornia at Berkeley, receiving his
B.S. in 1965. His graduate training
was at the University of Illinois
from which he received his M.S. in
1967 and PhD in 1969. His research
program consists of theoretical and
experimental studies on flow in-
stabilities at interfaces, convection
in fluids which is due to surface
tension or density gradients and in
fluid particle systems such as
fluidized beds and porous media.
Curtis W. Frank is an Associate Professor of Chemical Engineering
at Stanford University. He did his undergraduate work at the Uni-
versity of Minnesota, receiving his B. Chem. Eng. in 1967. His gradu-
ate studies were performed at the University of Illinois, from which
he received the M.S. in 1969 and PhD in 1972. His research program
in polymer physics includes projects in the thermodynamics of
amorphous polymer blends, intramolecular rotational motion and
polymer-solvent interactions in dilute solution and the synthesis and
morphology of poly (vinylidene fluoride). (M)

recent survey of prominent engineers, 95%o of the
respondents indicated that writing ability played
either a very important or a critically important
role in their work. [1] Unfortunately, training in
communications skills for engineers in most
colleges and universities is relegated to the two
or three courses in Freshman English required
of all students to graduate. Even if courses in
technical communication were available, the
typical chemical engineering curriculum is highly
structured and such an option, regardless of its
merits, faces stiff competition from other free
electives. A natural alternative to requiring an
additional writing course is the incorporation of
an emphasis on communication skills into the
laboratory course. Such an approach has been
followed in the Department of Chemical Engineer-
ing at Stanford since 1978.
The present configuration of our undergradu-
ate laboratory has evolved as a result of concerted
efforts to teach good communication. It was neces-

Copyright ChE Division, ASEE. 1982

Channing R. Robertson is Professor and Chairman of the Depart-
ment of Chemical Engineering at Stanford University. He completed
his B.S. at the University of California at Berkeley in 1965, and then
went to Stanford from which he obtained his M.S. in 1967 and
PhD in 1970. His research interests in bioengineering include projects
in transport phenomena in mammalian kidneys, artificial kidney com-
ponents, immobilization of whole cells, biocompatible materials and
membrane transport processes. (R)


sary, however, to first ensure that the experiments
were of uniformly high quality. In 1978 the de-
partment received an Instructional Scientific
Equipment grant from the National Science
Foundation. This grant, with additional funds
from the Stanford School of Engineering, the
Department of Chemical Engineering and a
Special Program Grant from the Dreyfus Founda-
tiontion, allowed new instrumentation to be pur-
chased so that new experiments in chemical re-
action engineering and polymer materials science
could be developed and existing ones upgraded.
Second, in a major commitment of department re-
sources, a new position was created for a tech-
nician whose primary responsibility would be the
undergraduate laboratory, his duties to include
routine maintenance, renovation of existing ap-
paratus and design and fabrication of new instru-
mentation. Third, three existing laboratory courses
were formally dropped from the curriculum, their
best experiments added to those developed under
NSF support in a new six unit two quarter se-
quence. These actions allowed strong emphasis on
the development of oral and written communica-
tion skills.
A key element in the new laboratory course has
been the assistance provided by the Communica-
tions Project of the School of Engineering at
Stanford University. [2] This is an innovative pro-
gram, initiated in September 1976, and designed
to assist engineering students in improving their
writing and speaking abilities. Among other
things, the Communications Project involves
person-to-person tutorials and the rewriting of
graded reports, features which have been made
integral parts of our laboratory course. The
unique aspect of the Communications Project is
that the communications tutors are not profes-
sional staff members; they are undergraduate
engineering students who have been specially se-
lected on the basis of their writing and speaking
talents and then given instruction to hone their
abilities further. The underlying philosophy is that
the tutors act as role models with whom the
students may closely identify.

N ORDER THAT the students have an appropri-
ate technical background for the experiments,
the course sequence is given in the winter and
spring quarters of the senior year after all
necessary lecture courses have been taken. A total
of twelve experiments in five different areas have

... in a recent survey of prominent
engineers, 95% of the respondents indicated that
writing ability played either a very important or a
critically important role in their work.

been developed to date. These include fluid me-
chanics (determination of flow profiles using
Laser-Doppler velocimetry, drag force on
spheres), heat transfer (transient and steady
conduction, forced convection, radiative energy
transport), mass transfer (steady diffusional mass
transfer), chemical reaction engineering (iso-
thermal batch reactor, tubular reactor, continuous
flow stirred tank reactor), and polymer materials
science viscoelasticc creep, dilatometric measure-
ment of glass transition temperature, differential
scanning calorimetry). Four faculty members are
associated with the course, with two sharing re-
sponsibility for the experiments each quarter. In
addition, there are two graduate teaching as-
sistants each quarter, one working with each
faculty member. Finally, there are one or more
peer tutors from the Communications Project in
both the written and oral skills areas.
The experiments are carried out by groups
composed of three students with each group per-
forming eight experiments during the two quarter
sequence. For each experiment one person acts as
group leader and the other two as assistants. Al-
though all group members should be conversant
in the underlying theory, the group leader bears
ultimate responsibility for the successful com-
pletion of the experiment. He must ensure that
the instrumentation is operating properly, that
the appropriate data are taken and that the calcu-
lations, shared among all group members, are done
correctly. The team concept is an important ele-
ment of the course. Efficient group operation in
the planning, execution and analysis of each ex-
periment requires effective intragroup communica-
tion and close cooperation. In its ideal form, this
experience provides a model for the student in
his later professional activities, e.g. as an engineer
on a process design team. Finally, since organiza-
tional responsibilities and reporting requirements
are rotated among group members throughout the
two quarters, each student receives multiple ex-
posure to several forms of expression, both
written and oral.

T HE PHILOSOPHY behind the reporting pro-
cedures (to be described shortly) is that de-


velopment of writing and speaking skills requires
extensive practice, i.e. learning by doing. More-
over, the student must make an active effort to
comprehend his errors and to correct the problems
through rewriting his reports until the presenta-
tion is clearly organized. An essential part of the
educational process is the tutorial session in which
the student meets individually with a writing and/
or speaking tutor from the Communications
Project, as appropriate. The tutorial session can
serve as a device to assist in preparation such as
a preview of an oral presentation or after the
fact as a means of evaluating clarity of presen-
tation in a report which has already been graded
for technical content. To assist the students
in understanding the level of faculty expectations

Perhaps the most important skill learned
is the ability to present results and conclusions
clearly and concisely in a short written
report or oral presentation.

for written and oral reports, a course syllabus is
provided in which guidelines are given. In addition,
during the first week of the course winter quarter,
before the experiments begin, some lectures are
devoted to fundamentals of communications skills.
The students are also advised to obtain and read
The Elements of Style, by William Strunk and
E. B. White, 3rd. ed., MacMillan Publishing, New
York, 1979 before the experiments begin. This
small text is concise and quite readable; in fact, it
serves as a good model of the objectives for the
written reports.
Since this course has only been offered a
limited number of times in its present form, it is
still evolving. The following description applies
to the course which was offered during the 1979-80
academic year in which four distinct forms of
communication were utilized. Three of these were
written and one oral. The most extensive written
document is the major technical report which is
required of the group leader for each of the first
six experiments. This report presents the theo-
retical background, objectives, laboratory pro-
cedure, results, conclusions and recommendations
in a style typical of a journal article. Although
this document is a major undertaking, the student
must exercise judgment in determining the depth
of coverage warranted for a particular experi-
ment. Since there are three persons to a group,
each student does two of these major reports

during the first six experiments. The major re-
port is graded twice: once for technical accuracy
and once for effectiveness of communication. The
first person to read the report is one of the gradu-
ate teaching assistants who prepares a critique
emphasizing technical accuracy. The report is
then examined by the faculty member in charge of
the experiment, again mainly for technical
purposes. However, if the report organization is
so confusing that it is not possible to determine
whether the student comprehended the intent of
the experiment, the report is returned to the
student for rewriting. Upon resubmission, the
faculty member and teaching assistant agree on
a grade on the technical content which is up to
75% of the total possible points. The report is
then returned to the student who must make an
appointment with the writing tutor from the
Communications Project. In this second evaluation
phase the tutor will go through the report in-
dividually with the student and examine it solely
for clarity of communication. After this tutorial
session, the student is usually required to rewrite
sections, or even the entire report. After re-
submission, the tutor assigns a grade, up to 25%
of the total possible points, on the basis of the
degree to which the student has complied with the
suggestions made during the tutorial session.
The second written form of communication is a
brief technical summary required of each of the
assistants. This is strictly limited to 300 words in
length and should provide a clear and concise de-
scription of how the objectives were met and the
major results. No supporting graphs or tables
are permitted and the use of equations is dis-
couraged. The grading of this report, both for
technical accuracy and for effectiveness of com-
munication, is done by the faculty member and
teaching assistant. Again, the report usually has
to be rewritten before a final grade is assigned.
Each student will do four of these technical
summaries over the first six experiments.
The third form of written report is used
during the last two experiments of the second
quarter. This is a short technical report or ex-
tended abstract which is prepared by each of the
group assistants. The report is three to five pages
long and includes a brief theoretical background,
objectives, presentation of results by means of
graphs and tables and discussion of the observa-
tions and conclusions. Each student does one or
two of these reports.
The oral communication exercise is a ten


minute talk which is required of one of the two
assistants for each of the first six experiments
and of the group leader for the last two. Thus,
each student presents at least three and perhaps
four talks over the two quarter sequence. Heaviest
use of the Communications Project tutors is made
in this phase of course activity, e.g. in 1979-80
there were three speaking tutors for a class of 29
students. The student presents his talk first to
the speaking tutor individually and then to the
entire class during one of the two weekly lecture
periods. Immediately after the presentation,
questions of a technical nature are posed by the
general audience. Then two previously selected
class members give oral critiques of the com-
munications aspects of the talk. Finally, each
faculty member, teaching assistant, speaking tutor
and student evaluator fill out grading forms on
the structure of the talk, use of visual aids, de-
livery and technical content. A summary of these
written evaluations is given to each student.
Although the course is only two years old, it
has become a focal point for the synthesis of other
elements of the undergraduate program. It differs
appreciably, however, from the orientation of the
traditional engineering courses, which typically
emphasizes mastery of the technical aspects of a
subject. Although development of sound experi-
mental technique is certainly one of the course ob-
jectives, the final course grade depends to a sub-
stantial degree on the ability of the student to
communicate the results of his laboratory efforts.
The selection of a laboratory course as the vehicle
to teach communication skills is particularly ap-
propriate for engineers. The class format is de-
signed to present students with varied require-
ments which are closely analogous to what they
will experience in their professional employment.
Perhaps the most important skill learned is the
ability to present results and conclusions clearly
and concisely in a short written progress report
or oral presentation. Rarely will supervisors or,
especially, managers have the time for more ex-
tensive discussions during the interim status re-
view of individual phases of an overall design
project, for example. However, in the final
documentation of the design for internal or ex-
ternal distribution, the ability to organize large
amounts of data, design calculations and recom-
mendations becomes essential. The major report
format is directed toward this objective. Two
pedagogical features of this course bear special
note. These are the emphasis on rewriting of

graded written reports, which is the rule rather
than the exception, and the use of tutorial sessions
for advance preparation on the oral talks and
followup analyses of the written reports. Only
through a clear understanding of his deficiencies
and extensive practice will the student develop
the desired facility in expressing his ideas and
describing his achievements. O
1. Davis, R. M., Engineering Education 68, 209-11 (1977).
2. For more information contact Ms. Ellen W. Nold, Di-
rector, Communications Project, School of Engineer-
ing, Stanford University, Stanford, CA 94305.

Continued from page 121.
volume contains many examples of Amundson
the artist and Amundson the craftsman.
Although a number of the papers in the col-
lection appear in a condensed form in textbooks,
the original papers are a preferable source. Many
are important references for graduate-level
courses in chemical engineering. The polymeriza-
tion papers with Liu, Warden, Zeman, and Gold-
stein are, for example, required reading in a course
on polymerization reaction engineering. (During
my graduate study in 1965, I spent a summer in
industry studying and applying these papers to
the modeling of a polymerization reactor.) Ma-
terial in the papers on the single catalyst particle
and on tubular and packed bed reactors has per-
meated most graduate-level courses in chemical
reaction engineering and is ideal supplementary
reading for students.
At a price of more than $100, the volume un-
fortunately lies outside the budget of many who
would greatly benefit by its presence in their
personal libraries. No academic or industrial
chemical engineering department should be with-
out at least one copy of this book. For those en-
gaged in or embarking on a career of research that
involves the mathematical modeling of chemical
engineering systems, the collected wisdom of
much of Neal Amundson's incomparable career is
well worth the personal investment. O

Yon Meurnooi.m
Lloyd A. Spielman, 42, Professor of Civil and
Chemical Engineering at the University of Dela-
ware, died on March 26 in Newark, DE.




Purdue University
West Lafayette, IN 47907

R RESULTS OF A RECENTLY conducted survey on
Teaching of Biomedical Engineering in Chemi-
cal Engineering Departments show that 28 schools
out of 79 replying to this questionnaire offer at
least one biomedical course for a total of 52
surveyed courses. Most of the courses reported are
general in nature although specialized courses in
biomedical fluid mechanics, biomedical mass
transport, biomaterials and other areas are also
taught. Five textbooks and instructors' personal
notes cover instructional needs. There does not
seem to be any shortage of teaching and research
professors in biomedical engineering and the
average class size is typical of most elective courses
in chemical engineering.

Biomedical engineering can be defined as the
application of the principles and practices of engi-
neering to the solution of problems in medical
research and health sciences. It was only very
recently that this research field was accepted as a
viable area of chemical engineering [1, 2], although
there are indications of related research activi-
ties in the late 50's. In terms of biomedical edu-
cation in chemical engineering, we have been able
to identify a relevant course at M.I.T. as the
earliest course of its type offered in a chemical
engineering department. It was introduced in the
curriculum by Professor Edward W. Merrill
during the 1962-63 academic year under the title
"Chemical Engineering in Biology and Medicine"
and it has been offered regularly since then [3].
Biomedical engineering is considered by many
as a "natural" for chemical engineers, since they
are qualified to handle problems related to kinetics,
fluid mechanics, mass and heat transfer in bio-

logical systems and artificial organs, as well as
problems related to synthetic materials used in
biomedicine. There are strong indications that
with the further expansion of major companies in
the area of artificial organs, qualified engineers
with basic background in the area will be needed.
Those who do not feel that biomedical courses
are necessary in chemical engineering point out
that there is a very small market for industrial
jobs in this area and that even the (rather limited
number of) biomedical engineering departments
and programs available in certain universities en-
counter problems in placing their graduates in
related jobs. It has also been noted that bio-
medical engineering is almost exclusively a gradu-
ate subject area and that the ratio of research to
other industrial positions is rather high. For these
reasons, introduction of even one biomedical
course in the chemical engineering curriculum is
a rather prohibitive "luxury", especially in view
of (i) the present popularity of other inter-
disciplinary areas such as energy, polymers, bio-

1 -

Nicholas A. Peppas is Professor of Chemical Engineering at
Purdue University. His research interests include macromolecular
networks, diffusion in polymers, biointerfacial phenomena and bio-
materials. He has coauthored over fifty publications in these areas
and he is director of the materials Division of AIChE and the Medical
Plastics Division of SPE. (L)
Richard G. Mallinson obtained B.S. degrees in chemical and bio-
medical engineering from Tulane University (1977) and his M.S. in
chemical engineering from Purdue University (1979). He is presently
a doctoral candidate at Purdue. (R)


Copyright ChE Division, ASEE. 1982

chemical and environmental engineering; (ii) the
alarming increase of undergraduate enrollment,
and (iii) the considerable teaching load of the
faculty, which has not increased proportionally to
the enrollments. Finally, unlike other inter-
disciplinary areas [4], biomedical engineering
suffers from an identity problem. The average
engineer confuses this area with biochemical
engineering. There is confusion, even within the
area, due to the development of more medically
rather than engineering oriented areas such as
clinical and hospital engineering [5].
A recently published British survey/report by
White [6] analyzes the research background of
some seventy U.S. and Canadian chemical engi-
neers in academic institutions, who are actively
involved in biomedical research. Although this re-
port is far from complete, White estimates that
there are about 400 chemical engineers involved
in medical research in the USA and Canada, ex-
cluding those employed by government agencies in
peripheral areas such as environmental medicine,
etc. Areas of research emphasis include blood
rheology and blood coagulation, heat and mass
transfer in biological systems, pharmacokinetics,
artificial internal and extracorporeal organs, bio-
materials, adsorption and separation from bio-
logical media, and cardiovascular and respiratory
White points out that in the United Kingdom
there are no more than 20 academic and a few in-
dustrial chemical engineers with biomedical
interests. A recent study by the Fed6ration
Europ6ene d' Associations Nationales d' Inge-
nieures (FEANI) [7] does not include biomedical
engineering as a recommended course in European
chemical engineering curricula. In West Germany,
the widely accepted "Erlangen alternative" does
not include any such courses either in the "Tech-
nische Chemie" or in the "Verfahrenstechnik"
program [7]. Therefore, it seems that chemical
engineering involvement in biomedical research is
an "American phenomenon"! It is exercised pre-
dominantly in academic research [6, 7] and its im-
pact in chemical engineering is not yet known [9].
Under the sponsorship of the AIChE Edu-
cational Projects Committee, a survey was carried
out during April 1980 in an effort to analyze the
trends of biomedical education, both graduate and
undergraduate, within chemical engineering de-

The main goal of this survey was to obtain
and analyze data which would show the trends of
biomedical education within chemical engineering.
Three types of data were used .. type and level of
course, subjects taught, and textbooks.

apartments. Questionnaires were mailed to 148 de-
partments reported in the AIChE publication
"Chemical Engineering Faculties" [10], 135 in
the U.S.A. and 13 in Canada. The questionnaires
were sent to faculty members previously identified
as having research interests in the area of bio-
medical engineering (about one fourth of the re-
cipients) or to the heads of the departments when
such information was not available. A total of 79
responses were received (53.4%) with 74 replies
coming from U.S. schools (55.2%) and five from
Canadian schools (38.5%).
Several schools designated that due to the
interdisciplinary nature of biomedical engineer-
ing, relevant courses are usually taught in other
departments. Other respondents noted that in
addition to courses offered in chemical engineer-
ing, there is a variety of courses in other depart-
ments. In processing the questionnaires we had to
eliminate a few courses in biochemical engineer-
ing, since the coverage of biomedical subjects was
not a major part of these courses (less than 30%
of the material covered).
Twenty-eight schools offer at least one course
in biomedical engineering. It must be noted, that
some of these courses are offered at irregular
intervals, when there is enough student interest
(graduate level), and that, sometimes, a course
outline may drastically change depending on the
instructor teaching the course. This number cor-
responds to 35.4% of the responses received or
18.9% of all schools where the questionnaire was
mailed. Of those that responded negatively (51),
four indicated that they teach some biomedical
engineering, six indicated that relevant courses
were taught in other departments only and four
said that they had discontinued previously offered
courses. In addition, five schools stated that they
had active research in this area, although no rele-
vant courses.
The survey shows that a total of 40 chemical
engineering faculty members are involved in bio-
medical teaching and 64 are involved in research,
for an average of 1.4 and 2.3 faculty members per
department respectively. These figures do not
include several schools known for their strong bio-


Biomedical Courses Offered in ChE Departments

# ChE




18 31 59.6
4 21
5 40.4

52 100.0

laboratory experiments in some or

medical program, either because no response was
received or because this program is administered
by another department (e.g. University of Utah,
Case Western Reserve University). For a more
accurate picture of biomedical education within
chemical engineering we were able to identify
through research publications, etc. [11], at least
14 more schools with active biomedical programs
which did not reply to this questionnaire. There-
fore, it can be said with some confidence that at
least 42 chemical engineering schools are active
in biomedical education and research.


The previous analysis is hardly encouraging
for the future of biomedical education in chemi-
cal engineering departments. As noted earlier,
biomedical engineering may be a "luxury" in these
days of high enrollments and high faculty teach-
ing loads. To investigate this notion we deter-
mined the total numbers of faculty members for
the 28 schools that offer a total of 52 biomedical
courses. Only six schools from this group have less
than ten faculty members, and the average size of
a school offering at least one biomedical course is
14 faculty members. The majority of the schools
offer one or two courses (Table I). Only six schools
(21%) offer 40% of the courses.
Table II shows the number of courses at each
educational level. There are 27 predominantly
undergraduate (52%) and 25 predominantly
graduate courses (48%). Rarities in this list
include a freshman general biomedical course (at
Washington University) and a sophomore course
on mechanics of animal motion (at John Hopkins).

All the courses meet at least two hours per week.
The only exception is a general biomedical course
at Purdue University which carries one credit
A more detailed analysis of the subjects
covered in various courses is offered in Tables III
and IV. Table III classifies the various courses by
area of emphasis according to the course title.
There are 29 general biomedical courses (56%).
Fluid mechanics and mass transport are major
areas of specialized courses. A number of bio-
medical courses with considerable emphasis on
mass transport, but also with additional cover-
age of other areas were judged by these authors
as "general" courses. The area of systems physi-
ology includes courses with significant emphasis
on biomedical control. Three courses in bio-
materials (at Rice University, C.U.N.Y. and Univ.
of Washington) offer the rare case of crossection
between two interdisciplinary areas, i.e. biomedical
and polymer engineering.
Table IV presents information on subjects
covered in the various biomedical courses as pro-
vided by the instructors for a total of 46 out of
52 surveyed courses. The cardiovascular and

Educational Level of Biomedical Courses

All levels



pulmonary systems, blood rheology, membrane
transport, pharmacokinetics and artificial organs
are the most "popular" subjects taught.
Class size and student participation are kept
at a reasonable level, typical of most elective
chemical engineering courses. The average class
had 16 students. The average graduate class had
10 students. The largest classes reported were at
Northwestern University for a physiology course
(60 students) and at Michigan State University
for a general course (40 students).
Chemical engineering faculties include
qualified researchers who can teach biomedical


79 100.0

* These schools offer
all of their courses.

Area of Emphasis and Level of Course

F So J J/S
1 3 2
- 1



- 1
- 1 1
- 2
.... 1

- 1 -
- 1
.... 2

- - -

- 1 -

8 7
- 6
1 1
1 -
- 1



Fluid Mechanics
Mass Transport
Systems Physiology
Artificial Organs
Animal Motion
Health Care Technology

courses. Schools reporting the highest number of
biomedical researchers were University of Wash-
ington (6), Carnegie-Mellon University (5), Uni-
versity of Minnesota (5) and C.U.N.Y. (5).


Additional information about trends in bio-
medical education is provided through textbook
preference by instructors of the various bio-
medical courses. A total of 5 books are used as re-
quired textbooks in more than one course, while
13 books were mentioned once in this survey
(Tables V and VI).
Textbook preferences have been classified in
Table V according to the educational level of the
course where they were used. These 68 preferences
do not represent an equal number of courses, since
in certain cases two books (or books and notes)

were required for a specific course. Perhaps the
most significant conclusion drawn from this Table
is that the "most cited" textbook was the in-
structors' "notes".
Cooney's, Lightfoot's and Middleman's text-
books capture 54% of this rather small market.
Cooney's textbook is used equally in under-
graduate and graduate courses. Middleman's and
Lightfoot's books are used predominantly in
graduate courses. The only other textbooks with
more than one preference were Seagrave's text
and Guyton's treatise of medical physiology.
Actually Guyton's text is the only one of the five
written by a non-ChE. All texts are relatively new,
having been published in the seventies. Reference
books are rarely used. There is, however, a prefer-
ence towards supplementary handouts such as re-
view papers etc.

Blood Rheology
Membrane Transport
Artificial Kidney
Pulmonary System
Heart & Lung
Other Artificial Organs
Body Composition
Thermal Effects
Biomedical Polymers
Others (one or two preferences)

Areas of Emphasis of 46 Biomedical Courses
F, So, J, J/S, S, all S/G, G



1 1 6 3 11 10 15

Textbook Preference for 50 Surveyed Biomedical

Biomed. Eng. Texts
Others (one preference)
Other texts
Others (one preference)
Literature Handouts

Undergraduate S/G, G


The main goal of this survey was to obtain and
analyze data which would show the trends of bio-
medical education within chemical engineering.
Three types of data were used for this analysis:
type and level of course, subjects taught, and text-
The number of chemical engineering depart-
ments that offer courses in biomedical engineering
is rather small. This subject seems to be of "second
priority" in most schools, and it is usually taught
as an elective, where there are qualified research-
ers to teach it. Only two schools of those surveyed
offer a biomedical course without having anyone
doing research in the area. Although there are
several courses for undergraduates only, most of
the courses seem to be open to graduate students
as well. The authors tend to believe (at least based
on the comments in the responses and on the text-
books used) that most of the surveyed courses are
of intermediate to advanced level. Courses which
are open to juniors must be (by necessity) mostly
descriptive in nature, simply because very little
mathematical analysis of biological phenomena
can be covered if the student has not finished at
least one course in transfer/transport phenomena.
As far as we know, in most schools, this is done in
the junior year.
Based on the analysis of the data of this survey
we believe that biomedical engineering has been
"accepted" by chemical engineers to the extent
that industrial needs warrant it, since the market
in this area is rather small. Biomedical engineer-

ing may not really be a "luxury", but it is worth
noting that most of the schools that offer bio-
medical courses had rather large faculties. We also
believe that this field is research-oriented and it
flourishes mostly through work at the graduate
level. This does not mean that there are no ap-
plications for the subjects taught or researched.
However, the market is limited, the competition
(from various disciplines) is high, and the results
and conclusions of biomedical research are "ap-
plicable" mostly to clinical cases (diseases etc.)

Textbooks Cited for Use in Biomedical Courses
(More than One Preference)

D. O. Cooney


Clarkson College

E. N. Lightfoot U. Wisconsin

S. Middleman

R. C. Seagrave

A. C. Guyton

U. Massachusetts

Iowa S. U.

U. Mississippi

Biomedical Engineer-
ing Principles
Dekker, 1976
Transport Phenomena
and Living Systems
Wiley, 1974
Transport Phenomena
in the Cardiovascular
Wiley, 1972
Biomedical Applica-
tions of Heat and
Mass Transfer
Iowa S.U. Press, 1971
Textbook of Medical
Saunders, 1976 (5th





100.0 61.8


rather than products. Significant misconceptions
concerning the goals of biomedical engineering
will have to be addressed by scientific societies.
Most of the biomedical courses offered in
chemical engineering are of general nature. They
usually include some anatomy and physiology,
principles and phenomena related to the cardio-
vascular and pulmonary systems, pharmaco-
kinetics and artificial organs. Specialized bio-
medical courses in fluid mechanics, mass transfer
and control seem to enjoy some popularity, es-
pecially when they address problems related to
microcirculation, diffusion in membranes and
systems dynamics.
Selection of an appropriate textbook is a rather
difficult task. This is not because of the lack of
good texts. Actually, in our opinion, all four
cited biomedical textbooks are outstanding in their
own way. However, biomedical engineering is an
area characterized by high individuality in in-
struction. Since many graduate courses are
tailored to the needs of graduate research as-
sistants working in biomedical engineering,
emphasis must be placed on one or two specific
areas. For example, instructors provided sugges-
tions on additional material that could be covered
in a new edition of the present textbooks. These
suggestions included: design of artificial organs,
thermal physiology applications, more controlled
release, more biomaterials, more physiology, more
quantitative texts, more and better pharmaco-
kinetics and more thermodynamics.
This simple listing also provides an indication
of present and future trends and needs of bio-
medical education in chemical engineering. It
continues to be a predominantly graduate subject.
More emphasis is given to quantitative and basic
aspects and there is a definite departure from em-
piricism. "New" areas in biomedical instruction
(not necessarily "new" to instructors, but defi-
nitely "new" for the textbooks) will have to be
added, such as pharmaceutical engineering in-
cluding controlled release systems, fundamentals
of biomaterials, natural membrane science, micro-
circulation, engineering aspects of cancer research
Class sizes are rather on the low side and some
courses are occasionally cancelled due to lack of
enough student interest. In recent years, bio-
chemical engineering has entered a flourishing
period, mainly due to excellent job opportunities
and the development of industrial processes in
waste treatment, biomass conversion, other

fermentations, food production, etc. However, bio-
medical engineering is not destined to have a simi-
lar role, at least not in the next few years, unless
there is a major research and industrial effort
towards health problems.

The survey findings were presented at the
73rd Annual AIChE Meeting, Chicago, IL, No-
vember 1979. The authors wish to acknowledge
the financial assistance of the School of Chemical
Engineering of Purdue University in the prepara-
tion and distribution of the questionnaires and
final reports of this survey. O

The following is the list of Departments of
Chemical Engineering that offer Biomedical
Engineering courses and responded to the survey.
In parentheses we have designated the research-
ers who replied. Univ. of Arizona (J. F. Gross),
Carnegie-Mellon Univ. (R. K. Jain), Clarkson
College (D. O. Cooney), Georgia Tech (A.
Yoganathan), Johns Hopkins (S. Corrsin), Univ.
of Kansas (K. Himmelstein), Kansas State Univ.
(W. P. Walawender), McMaster Univ. (I. A.
Feuerstein), Michigan State Univ. (D. K. Ander-
son), Univ. of Minnesota (K. H. Keller), M.I.T.
(C. K. Colton), C.U.N.Y. (H. Weinstein), SUNY
Buffalo (P. Stroeve), North Carolina State Univ.
(F. M. Richardson), Northwestern Univ. (T.
Goldstick), Univ. of Pennsylvania (D. Lauffen-
berger), Pennsylvania State Univ. (J. Ultman),
Univ. Pittsburgh (J. T. Cobb, Jr.), Purdue Univ.
(N. A. Peppas), Rice Univ. (L. V. McIntire),
Stanford Univ. (C. Robertson), Univ. of Ten-
nessee (W. W. Hsu), Univ. of Texas (R. Popo-
vich), Univ. of Toronto (M. V. Sefton), Tufts
Univ. (J. H. Meldon), Washington Univ. (R. E.
Sparks), Univ. of Washington (A. S. Hoffman),
W. Virginia Univ. (E. V. Cilento), Univ. of Wis-
consin (E. N. Lightfoot), and Yale Univ. (C.

1. O. A. Hougen, "Chemical Engineers and How They
Grow," Chem. Tech., 9, 10 January 1979.
2. 0. A. Hougen, "Seven Decades of Chemical Engi-
neering," Chem. Eng. Progress, 73 (1), 89, (1977).
3. "A Guide to Biomedical Engineering and Physics at
M.I.T. and Harvard University," Harvard-MIT Pro-
Continued on page 143.


o international



Universidad Autonoma Metropolitana-Iztapalapa
Mexico 13 D.F. Mexico

in Mexico have had an enormous growth rate
during the last three decades (averaging 10-12% /
year) and the demand for chemical engineers and
other related professionals has grown accordingly.
As a matter of fact, from 1973 to 1979 there was
an annual increase of 75% in the demand for
engineers in these industries.
Because of this demand, the Mexican Govern-
ment has promoted the establishment of under-
graduate programs in chemical engineering all
over the country. There are now 54 institutions
offering such programs, compared with only half
that number six years ago [1]. This has resulted
in an explosion of the number of incoming students
as well as in those receiving degrees, which oddly
enough has exceeded the real demand of chemical
engineers by about 25%
This rapid growth has been counterproductive
in terms of the quality of the programs offered
and of the performance by students and pro-
fessors. Because of this lack of quality, the larger
industries have had to develop training programs
to supplement the deficiencies of the new engi-
neers at the beginning of their careers. Also, the
most established institutions of higher education
have opened graduate programs (M.S. degree
only) in order to provide industry with highly
qualified engineers, as well as to supply the uni-
versities and research institutes with badly
needed teachers and scientists. These engineers

The most relevant single event that changed
the course of development of Mexican industry
was the nationalization of oil resources in 1938 ...
(which) increased the number of chemical
enterprises from 379 to 1710.

Copyright ChE Division, ASEE, 1982

Enrico N. Martinez, chemical engineer from Universidad Nacional
Autonoma de Mexico (UNAM), received his M.S. and Ph.D. (1972) in
Chemical Engineering from University of Notre Dame. He was an
Assistant Professor at UNAM, and is currently Associate Professor at
Universidad Autonoma Metropolitana-lztapalpa (UAM-I), where he has
just been appointed Head of Engineering. He consults in Process Re-
search and Development and his research interests are in Chemical
Reaction Engineering, Catalysis and Education. (L)
Roman Gomez-Vaillard received his B.Sc. degree in Chemical Engi-
neering at UNAM and his Ph.D. at Imperial College, London. He
joined UAM-I in 1976. He has recently become the Chairman of the
Chemical Engineering Group there. His research interests are Process
Design and Development and Computer Applications in Chemical
Engineering. (R)

are beginning to do research and development-
activities that, due to the nature of the Mexican
industry, were practically non-existent ten years
ago [2], and are only now starting to achieve their
just dimensions.

At this point we will analyze the development
of the chemical industry in Mexico, as it has
been determinant in the development of the chemi-
cal engineering profession and the curricula at the
universities that offer undergraduate and gradu-
ate programs.
The most relevant single event that changed the


I 1

course of development of Mexican industry was
the nationalization of oil resources in 1938. The
resulting growth (from 1940 to 1950) increased
the number of chemical enterprises from 379 to
1710. During this period the industry also started
to diversify into production of chemical inter-
mediates, pesticides, fertilizers, dies and inks,
paints and pharmaceuticals. Employment in the
chemical industry rose 13% annually, while it in-
creased by only 6 % in the manufacturing industry
as a whole. Furthermore, the investment rate had
a growth of 25.7% per year versus 10.5% for
the manufacturing industry [3].
The Mexican government established control
over the petrochemical industry through legisla-
tion giving Petroleos Mexicanos (PEMEX) the

Ton. x 10-3
4 500o

4 000o

a 500

2 000





Ton x 10- 3
---- Soondary


I \
I \
/ \
/ \



o I I I I I I yeoar
1965 6' 69 71 73 75 77

FIGURE 2. Petrochemical Industry: Imports

Petrochemical Commission [4]. It should be pointed
out that the deficit in the balance of payments has
been growing since 1960, and that most of the
equipment and technology for both the petro-
chemical industry and the chemical industry in
Mexico comes from abroad. This fact, as we will
see later, has tremendous importance in the de-
velopment of the chemical engineering profession
in the country.

1965 67 69 71 73 75 77

FIGURE 1. Petrochemical Industry: Production

right to produce and/or import all "basic" petro-
chemicals; "basic" petrochemicals being those
products derived through a first transformation
of oil and natural gas components. Moreover,
private enterprise can use those basic petro-
chemicals to produce "secondary" petrochemicals
if 60% of the share (minimum) corresponds to
Mexican capital. Thus, a company fulfilling that
requirement can obtain a "Petrochemical Permit"
to produce a given secondary petrochemical, such
as PVC, phenol, carbon black and other similar
An illustration of the development of the petro-
chemical industry (which by 1975 was already
41% of all basic chemical industry) is shown in
Figs. 1, 2, and 3, for the production, imports, and
exports of that sector of Mexican industry. All
the data were taken from reports by the Mexican


1965 67 69 Ti 73 75 1977

FIGURE 3. Petrochemical Industry: Exports





Because of this demand, the Mexican Government has promoted the establishment of
undergraduate programs in chemical engineering all over the country. There are now 54 institutions
offering such programs, compared with only half that number six years ago.

The Mexican Government also offered an extra-
ordinary stimulus for creation of the chemical
industry in Mexico, with the sole purpose of re-
ducing imports of foreign goods. This policy led,
in many cases, to inefficient and expensive pro-
duction of chemicals, resulting in products that
were highly priced internally, with no possibility
for export. This situation did not stimulate the
development nor the adaptation of technologies,
and only required chemical engineers for the
operation of plants, the administration of chemi-
cal firms and the commercialization of commodi-
At the beginning of the seventies, the govern-
ment changed its policy in order to reduce foreign
debt and to support a healthier industrial develop-
ment. Emphasis was put on technological inde-
pendence and the creation of appropriate tech-
nologies. The emphasis nowadays is on the creation
of enterprises that produce goods at competitive
international prices rather than solely for the
consumption by the internal Mexican market.
Several facts are indicative of the above; the
Mexican Petroleum Institute (IMP) was created
in the late sixties with the purpose of performing
research and development in oil and petrochemical
technology, as well as process and project engi-
neering mainly for PEMEX. The National
Council for Science and Technology was founded
in 1971; in 1972 the Law of Technology Transfer
was published, and in 1973 the Law for the Pro-
motion of Mexican Investment and Regulation of
Foreign Investment. Also, a major program for
exploration of oil and mineral resources was
launched, resulting in the discovery of important
oil reserves in Mexico.
Therefore, the challenge faced today by the
chemical engineering profession in Mexico is tre-
mendous. As we will see in paragraphs to follow,
the curricula of the main institutions have been
greatly influenced by the nature of the develop-
ment and the requirements of industry; but it has
not been intended to prepare professionals for the
main activities that they perform in developed
countries, such as process engineering and re-
search and development. Furthermore, since the
most established institutions are slow in changing

their programs, new engineers have been edu-
cated in sufficient numbers, but at the expense of a
loss in quality. A significant lag exists between the
real needs of the chemical industry and the supply
from the universities [5].

Since 1940, the demand for chemical engineers
has been growing in phase with the chemical
industry, and this has produced an explosion of
both students and professionals through the years.
Fig. 4 shows this situation graphically.
The first institution to offer a chemical engi-
neering degree was the National University
(UNAM), starting in 1918. In 1936 the National
Polytechnic Institute (IPN) opened the second
program in the country. Since then, these two
institutions have been the main suppliers of chemi-
cal engineers, due mainly to their location in the
middle of the densely populated and highly in-
dustrialized metropolitan area. Table 1 shows re-
vealing figures in this respect.
Therefore, it is not surprising that the two
main schools of chemical engineering (UNAM and
IPN) have had a strong influence on the programs
and development of most of the schools created
later. Actually, most State Universities in the
country have been founded by UNAM graduates
and their curricula have been, in many cases, the
same or very similar to that of the National Uni-

F-msng Chem E'.

Veorly Prodluct1on

FIGURE 4. Supply and Demand
in Mexico

of Chemical Engineers




versity. Even the private institutions had exactly
the same curriculum from 1943 to 1973. The
Ministry of Public Education has opened a good
number of Regional Technological Institutes
throughout the country in which chemical engi-
neering programs are offered, and these institutes
have traditionally followed the programs of IPN.

Graduates from Chemical Engineering
Schools in Mexico (1971-76)


I.P.N. 6086 35
U.N.A.M. 4560 26
Univ. Aut. de Puebla 1097 6
Univ. de Guadalajara 754 4.25
Univ. Aut. de Nuevo Le6n 546 3
I.T.E.S.M. 480 2.75
All Others 3901 23
Total 17424 100

Therefore, if we analyze the curricula of the two
main institutions, we will have a situation repre-
sentative of the whole country.
The curriculum at UNAM has been shaped
according to the needs of Mexican industry and
has remained practically unchanged for the last
40 years. The main features are a strong emphasis
on chemistry and a wide variety of subjects related
to activities that chemical engineers in Mexico
have traditionally been involved with, such as
aspects of civil, mechanical and electrical engineer-
ing. Activities that in developed economies are
handled by the respective specialist, but which in
Mexico (due to the nature and dimensions of the
industry) are given to the chemical engineers to
take care of.
Therefore, if we look at the curricula at UNAM
and IPN, we notice that they involve a large
number of subjects in general, but do not empha-
size the fundamental knowledge of basic sciences
such as physics and mathematics. Such important
disciplines as transport phenomena and process
dynamics and control are not a part of the cur-
riculum, but are offered as elective subjects. The
same happens with computer programming and
numerical methods.
Until 1967, a student had to complete five years
of studies in order to get a B.S. in chemical engi-
neering at UNAM; now the program is nine

semesters long. At that point, the curriculum was
changed from year courses to semester courses,
but there was no change in the content. At IPN
the program was also five years long and it was
only modified to a semester structure in 1975. Also,
in order to get a degree, the student had to write
a thesis and comply with a Social Service require-
ment that involves 400 hours of free work as a
service to society, since higher education is highly
subsidized by the state and is practically costless
to the student.
As mentioned in the introductory paragraphs,
the number of schools that offer chemical engineer-
ing degrees has grown tremendously in the last
ten years. Of the present 54, only 9 are private
institutions; of these, the most relevant are
ITESM at Monterrey and UIA in Mexico City.
Both these institutions have contributed sig-
nificantly to the education of chemical engineers,
mainly from the standpoint of quality graduates.
In the case of ITESM, its development has been
somewhat independent from the large institutions,
and in that respect it has pioneered in chemical
engineering education, following the patterns of
American universities such as Wisconsin. How-
ever, as we can see from Table 1, the contribution
of ITESM to the supply of chemical engineers
amounts to less than 3%.
Another characteristic of undergraduate edu-
cation in Mexico is that most of the professors
teach on a part-time basis and come from industry
to teach one or two courses a week. This contrasts
with education in developed countries where most
of the professors are fully devoted to teaching
and research. Table 2 shows figures for some of
the most representative schools of chemical engi-
The government answer to the demand for in-

Composition of Faculties


Univ. Aut. de Puebla
Univ. de Guadalajara
Univ. Aut. de Nuevo Le6n
I.T.R. Celaya
I.T.R. La Laguna
Univ. Veracruzana

Full Half
Time Time


35 8 330
186 102 142
6 18 20
9 37
24 2 47
13 1 14
6 2 12
4 3 12
10 2 28


creasing numbers of chemical engineers was the
creation of a large number of institutions offering
the program (22 in the last 6 years). Also, UNAM
has opened two new schools in the metropolitan
area of Mexico City and, more significantly, the
government created the Metropolitan University
(UAM) with two campuses offering programs in
chemical engineering in Mexico, D.F.
The characteristics of this new university are
different from those of other government spon-
sored universities in several respects. Outstanding
are the following facts: there is a tuition fee that
is considerably higher than the one charged at
UNAM, the academic structure is similar to that
of American universities (by departments), and
the ratio of students to professors is quite low. In
a way, we could say that UAM represents a model
of the type of university that the government
wants to have in this new stage of development
of the Mexican economy.
Thus far, thanks to support from the govern-
ment, the demand for chemical engineers has been
satisfied, in excess, in terms of quantity, as can
be seen in Fig. 4. However, the main problem in
the last ten years has been quality. According to
the opinion of recruiters, only 10 % of the gradu-
ates have "satisfactory" quality, about 30% are
"capable or able to be trained," and 50-60% are

The present problem is one of supplying engi-
neers capable of facing the challenge that the
development of modern chemical industry poses.
Quality should be emphasized, as well as a sound
formation in the fundamental principles of chemi-
cal engineering that are common to all chemical
industries. Also, in order to develop appropriate
technologies and to assimilate those imported from
abroad, the graduate programs should be
strengthened to provide researchers and teachers
capable of supporting the proposed new curricula.
Therefore, any attempt to develop a curriculum in
chemical engineering should be based on two
A knowledge of what chemical engineering is, and
A knowledge of the requirements of the chemical
The first condition seems to be easy to achieve.
However, experience has shown that in Mexico
most failures in curricula design can be blamed
on a lack of knowledge of what chemical engineer-

ing is. Although the discipline has reached a great
degree of maturity in places like the USA, in
Mexico there are very few people with enough
background to give a precise definition of the field.
A feasible proposal to solve this problem has
been set forth in a previous publication [7].
Basically, this recommendation consists of a
definition of those aspects of science that are most
important for sound development of chemical engi-
neering, in order to establish a basic structure of
knowledge for the first year of the curriculum.
Clearly, the fundamental core of any chemical

S. this basic structure of
knowledge, as well as the fundamental
core, should be established and developed
on a national basis through the
government sponsored schools.

engineering curriculum consists of mass and
energy balances, thermodynamics, applied mathe-
matics, transport phenomena and chemical re-
action engineering. The basic structure of
knowledge should prepare the student to face
these disciplines successfully, and should include
general physics, general chemistry, and mathe-
matics. All these should have a strong practical
support through carefully planned laboratory
We believe that this basic structure of
knowledge, as well as the fundamental core, should
be established and developed on a national basis
through the government sponsored schools. In
order to do this, an "Academic Commission"
should be formed by representatives with high aca-
demic standing from each area of the country,
calling upon nationally recognized researchers and
specialists to work as consultants.
The main purpose of these consultants would
be to supply the information needed for the second
item of our basic principles; i.e. the requirements
of our chemical industry. In this paper, we have
given a very brief review of the requirements from
our point of view. These have served for the design
of the curriculum at our own institution. UAM-I,
which we consider to be the model to follow.
Therefore in the next section we will deal with
it in some detail.

Fig. 5 shows a schematic of the general activi-
ties of a chemical engineer and the interconnec-


I) Thermodynamics 2) Applied Mathematics
3) Transport Phenomena 4) Chemical Reaction Engineering

FIGURE 5. General Activities of a Chemical Engineer

tions and relationships with the main industries
in Mexico. As can be seen, a wide variety of ac-
tivities is involved and the industries have large
differences among them. Therefore, it is very diffi-
cult for a program to attempt to cover every
aspect that the future engineer may require, be-
cause of the inherent risk of wasting resources
and effort. Rather, whatever resources we have
can be employed more efficiently if we follow the
guidelines given in the previous section of this
paper. This would provide a general program with
strong chemical engineering fundamentals, flexi-
bility, and adaptability to the different working
environments which will confront the engineer.
The objectives of the program at UAM-I are:
To educate chemical engineers with a high academic
standard, so as to enable them to contribute to the
development of the Mexican industry.
To provide the future professional with a strong
scientific background that will give him the cap-
ability to perform in any given field of chemical
engineering, and to solve the inherent problems with
the required depth and flexibility.
To provide the future engineer a continuous contact
with the problems and needs of the Mexican industry
in order to facilitate his finding a role within the
context of his future working environment.
To prepare professionals capable of continuing their
education in a graduate program, in Mexico or
abroad, to strengthen teaching and research in the
To prepare professionals with capabilities for de-
veloping new and appropriate technologies in ac-
cordance with the specific needs of our country.
Fig. 6 shows a schematic of the curriculum at

UAM-I. The foundation of the program is the
"General Core" of basic science and engineering.
This core is taken by all science and engineering
students in the first three quarters at the uni-
versity. The second stage of the program is a
"Common Core of Engineering," which is a group
of seven courses in applied mathematics, including
computer programming, numerical methods,
operations research, optimization and engineer-
ing economics. This second core is taken by all
engineering students between the fourth and ninth
quarters of the curriculum, while they are also
taking the fundamental chemical engineering
The last stage of the curriculum is what we
call a Major Area, and in particular at UAM-I
we offer a major in process design and develop-
ment. During this stage, the student takes a "Pro-
cess Design and Development Laboratory," where
he is given a comprehensive project with the pur-
pose of developing a technology for manufactur-
ing a product for the Mexican market, under the
supervision of our entire faculty. This course is
three quarters long and the student is supposed
to use bench scale and pilot plant facilities to
solve the problems posed by a real project.
The quarter structure of our program leaves
room for five elective courses, which should be
selected from those offered by other engineering,

Continued on page 143.

FIGURE 6. Chemical Engineering Curriculum at UAM-I




Eastman Kodak Company
Kingsport, TN 37662

Editor's Note: This is the second installment of Dr.
Siirola's two-part lecture. The first installment ap-
peared in Chemical Engineering Education, Vol.
XVI, No. 2, page 68.


Systematic generation approaches are vari-
ants of a problem-solving formalism known as the
state space paradigm. Problems framed in this
manner consist of an initial state (a set of streams
at their initial temperatures or a multicomponent
stream to be separated), a set of possible states
(streams at intermediate temperatures or streams
containing fewer components), a set of possible
transformations (changing temperatures by heat
transfer or separating multicomponent feeds into
two streams of different compositions), descrip-
tions of the states resulting from application of
a transformation to a given state (design equa-
tions for determining exit temperatures from an
exchanger or outlet compositions from a sepa-
rator), and a final state (the set of streams at
their desired temperatures or the set of separated
streams). The number of possible states generally
increases exponentially with problem size. A solu-
tion to the state space problem is any sequence of
transformations that leads from the initial state
to the final state subject to any constraints that
may be applicable. Different paths through the
state space may represent the same design. For
the heat integration and separations sequencing
synthesis problems considered here, the existence

Since the design that results from a
heuristic search often becomes the initial point
in an evolutionary structural improvement procedure,
it is important that the heuristically developed
design be as near to optimal as possible

Copyright ChE Division, ASEE, 1982

J. J. Siirola received his B.S. from the University of Utah and his
Ph.D. from the University of Wisconsin-Madison in 1970 where he
developed the AIDES process synthesis system and coauthored the
introductory text, Process Synthesis. Besides a continuing involve-
ment in synthesis technique development, implementation, and ap-
plication, his research interests also include simulation and optimiza-
tion aspects of computer-aided design, non-numeric programming,
artificial intelligence, and technology assessment. He is currently a
Research Associate in the Eastman Chemicals Division of Eastman Kodak

of a feasible path from the initial to the final
state is usually not in question. Rather, what is
sought are paths (designs) which arrive at the
final state in some optimal manner.
For sufficiently small problems, it may be
possible to search exhaustively through the state
space for solutions. Each feasible transformation
may be applied to the initial state producing
alternative intermediate states, followed by ap-
plication of transformations to these states
generating yet more states, and so on. A search
strategy that considers states for transformation
application in the same order in which they were
generated is termed 'breadth first.' Alternatively,
a search strategy can be implemented in which
some transformation is applied to the initial state,
followed immediately by another transformation
applied to the resultant state, and so on. When
the final state is reached, alternative transforma-
tions are applied to the penultimate state until


the final state is again reached. Only after all
feasible transformations have been applied to a
given state and followed through to the final
state are alternative transformations applied to
its predecessor, and so on, until all paths through
the state space have been generated. This type of
search is termed 'blind depth first.' However, if
costs can be computed independently for each
transformation, the smallest total cost for any
complete path generated from initial to final states
serves as an upper bound (if minimum cost is the
design objective) for other desirable designs.
Therefore, for the usual case where costs increase
with each transformation, the search can be
terminated without reaching the final state if
the total cost of a partial path being developed
exceeds the current bound.
Ideally, this 'branch-and-bound' technique is
more efficient (avoiding a search through more of
the feasible state space) if what proved to be the
best path is generated early in the search. This
might be accomplished by examining at each state
all feasible transformations but choosing first the
one with lowest cost. Searches which selectively
apply state transformations are termed 'ordered
depth first.' The technique may be refined by in-
cluding in the transformation evaluation not only
its cost but also an estimate of the costs of all
remaining transformations to the final state re-
sulting from its application. This 'predictor
ordered search' generally examines the smallest
subset of the state space in discovering optimum
In situations where the application of at least
some transformations are independent of each
other, several different paths through the state
space may result in the same design. As this is
the case for sharp separations sequencing, it is
sometimes more convenient to consider the dual
of the usual state space representation in which
the transformations become the nodes. This
focuses attention on the alternative transforma-
tions to be performed rather than on the inter-
mediate states. Each complete and consistent set
of transformations joined by arcs of the dual
graph represent a unique separations sequence.
Efficient use of state space search in syste-
matic generation synthesis approaches requires
at each state the identification of applicable trans-
formations and an evaluation of the value
(possibly considering effects on the remainder of
the problem) of its application. In the previous
examples, and in similar strategies applied to

Chemical process synthesis is an activity
concerned with the invention of structurally
and operationally superior design alternatives
... generally performed by experienced, creative
engineers assisted by several design aids
based primarily on analogy.

heat integration, as the objective function is
taken simply as the sum of the transformation
costs, the search is guaranteed to find the optimal
design for the classes of transformations avail-
able. Searches in this situation have been termed
However, in many cases involving for example
subjective multivariable design criteria, extremely
complex transformation evaluations, or very large
state spaces, it may not be desirable or even
possible to search analytically. The selection or
evaluation of transformations can then be made
on an empirical basis by using previous experience
embodied in design rule-of-thumb. These rules, or
'heuristics,' reflect generally successful strategies,
although they carry no guarantees of optimality.
Their value is that they often limit the growth
of the state space to only a polynomial function
of problem size. Heuristic searches are among
the oldest systematic process synthesis techniques,
having been applied very early to separations
selection and sequencing. For this problem, at
least 14 such rules are now in common use, in-
cluding 'remove the most volatile species next',
'remove the species in highest concentration next',
'perform the separation which results in an ap-
proximately equimolal split', 'perform separations
with high recover fractions last', and 'perform the
cheapest separation next.' Similar design heur-
istics have been proposed for other synthesis sub-
problems including complex refrigeration cycles
and heat integration.
Since the design that results from a heuristic
search often becomes the initial point in an evolu-
tionary structural improvement procedure, it is
important that the heuristically developed design
be as near to optimal as possible. Unfortunately,
the advice offered at each state by alternative
heuristics is sometimes contradictory. Adaptive
heuristic selection procedures can be modified on
the basis of the success of repeated heuristic
ordered depth first searches. However, for large
problems such iteration may not be practical. As
an alternative, situation dependent preference
ordering of the selection rules has been proposed


for once-through heuristic search procedures
which terminate when the final state is reached
and for guidance in altering the ordering during
any subsequent evolutionary modification.
Much synthesis research continues into the
search for more powerful design heuristics and
more precise definitions of conditions which de-
limit their applicability. For example, heuristics
have been developed which, given only a descrip-
tion of the initial state (and assumed for the
final state), completely specify the optimal path
without examining the intermediate states for
three-component separations that involve simple
ordinary and thermally coupled distillation. In
heat integration, utility and capital economic
factors have led to the realization that better
(from a cost criterion) designs generally result
from efforts to maximize heat transfer among pro-
cess streams, while simultaneously minimizing
the number of exchangers employed to accomplish
it. That maximum integration is consistent with
the thermodynamic desirability to minimize net-
work irreversibilities led to a minimum tempera-
ture driving force match heuristic. Methods to
minimize the number of exchangers have also
been investigated. The most significant recent
result in heat integration is the fact that maxi-
mum integration (and hence minimum utility re-
quirement) and the likely minimum number of
exchangers can both be calculated for a given
problem before the generation of any networks.
This analysis provides additional bounds and
targets which can be used to limit the size of the
search space, estimate the potential for evolution-
ary improvement of a candidate design, or
estimate the cost of accomplishing heat transfer
tasks for a process without actually synthesizing
any networks. Similar performance targets, often
the result of simplified thermodynamic analysis,
are being developed for other synthesis sub-
Another critical factor in the application of
state space problem-solving approaches in process
synthesis is the choice of representation for de-
scribing the states and the transformations
among them. It is desirable from search efficiency
considerations to develop representations which
do not include states that cannot possibly be part
of the optimal path. On the other hand, care must
be exercised so that the experience, prejudice,
heuristics, or assumptions embedded into a repre-
sentation for efficiency do not inadvertently con-
strain or otherwise exclude truly novel 'creative'

optimal solutions from consideration.
Although the previous discussion has drawn
examples only from heat integration and separa-
tions sequencing, state space systematic genera-
tion approaches have also been applied to other
synthesis subproblems including reaction paths,
control systems, and safety systems. Each of these
subproblems is fairly well defined and each has a
sizable state space. However, these synthesis sub-
problems are not necessarily independent. For
example, if thermal energy is employed to effect
a multicomponent separation as in distillation,
the separations sequencing and heat integra-
tion subproblems interact closely. The optimal
design and operation of a column (operating pres-
sure, degree of feed vaporization, number of
stages, reflux ratio, etc.) depend on energy costs
and, hence, on heat integration. Heat integration
opportunities, however, depend upon column
temperatures and heat loads, which are functions
of operating pressure and reflux ratio which is,
in turn, influenced by pressure effects on the ex-
ploited property difference, relative volatility.
Further, the identity of the streams available for
integration is not known until the complete
separation sequence has been generated.
Early procedures for solving this combined
problem required excessive effort, largely because
of the incompatibility of the dynamic pro-
gramming approach used for column sequencing
with the interactions resulting from heat inte-
gration and the large state space of even the re-
stricted integration problem. Later developments
in heuristic search improved both the capacity
and speed of heat integration syntheses. These
improvements allowed all sources and sinks in-
cluding sensible streams to be considered by an
integration procedure fast enough to be included
within a design variable optimization of each
synthesized separation sequence subject to the
same simplifying assumptions. Evaluations of
many case studies verified that, although the re-
sultant objective surfaces are very irregular, sig-
nificant improvements are achieved when designs
for these two subproblems are synthesized simul-
taneously in comparison with independent se-
quential synthesis. Current efforts, based on the
development of improved bounding strategies,
seek to generate optimal designs for this combined
problem without necessarily generating all paths
through the separations sequencing state space.


The design of complete flowsheets is another
complex synthesis problem that perhaps embodies
all the other synthesis subproblems discussed.
Ideally, based on experience with the combined
problems of separations sequencing and heat inte-
gration, it may be desirable to solve all the sub-
problems simultaneously. Development of such a
formulation is a formidable task and may never
be accomplished. Several simplified approaches,
however, have been suggested.
One scheme, like other systematic generation
approaches, decomposes the problem in such a
way that at least one task is immediately recogniz-
able as achievable with available technology such
as a reactor, distillation column, or heat exchang-
er. Given descriptions of the desired product, po-
tential raw materials, pertinent chemistry, and
available technology, the binary resolution pro-
cedure of predicate calculus was adapted to prove
constructively the existence of a path through the
state space. Since for most real design problems
the existence of a solution is not in question,
resolution was replaced with a goal-directed
depth first search (proceeding backwards from
the desired products) followed by evolutionary
improvements involving recycles and energy re-
covery to generate efficient overall designs.
Alternatively, a decomposition scheme is
possible that does not result in the immediate
recognition of available technology but rather a
multilevel series of subproblems similar to the
synthesis subproblems identified previously. The
innermost level subproblem involves the selection
of chemical transformations by which available
raw materials are to be converted into the desired
product. Once approximate conversions and yields
of the chosen reactions are determined, a rough
material balance can be calculated. In the next
subproblem, the fate of each chemical species in
every raw material and reactor effluent is as-
signed. Decisions to recycle unconverted reactants,
remove contaminants before or after reaction,
allow them in the product, send them to waste
treatment, or process them into byproducts, etc.
are made. The implementation of this species al-
location is addressed in the next subproblem by
using means-ends analysis to detect required
temperature-changing, pressure-changing, stream-
splitting, species-separating and or other physi-
cal transformation tasks. The outermost level sub-
problem specifies processing equipment and
control strategies to effect the required tasks,
integrating as appropriate consecutive or comple-

mentary tasks for capital and operating cost
efficiencies. Each of these subproblems interact.
Correct reaction choices depend upon how the re-
action path will be implemented. Optimal species
allocation to support the reaction path depends
in part upon the separation problems which arise
as a result of the allocation. The selection of
physical transformation tasks and their operating
conditions may depend upon the degree of task
integration, etc. At each level, choices are made
as a result of an analysis of the bounds defined
by actions taken in previous levels and by heuristic
estimates of the probable effects of the choice
selected on subsequent levels.

The level of sophistication exhibited by the
systematic chemical process synthesis techniques
so far developed ranges from primitive to
moderate. From an industrial point of view, heat
integration synthesis is the most mature, although
no single technique is universally superior in all
applications. The prediction of utility bounds,
minimum number of exchangers, and identifica-
tion of evolutionary structural improvements are
significant design tools. Some computer programs
are commercially available. Heat-integrated
separations sequencing also has industrial ap-
plicability, although at present is somewhat
limited by simplifying assumptions. The total
flowsheet synthesis computer programs are
strictly experimental. It is, of course, not neces-
sary that successful process synthesis techniques
be computerized, although in research the develop-
ment of computer implementations sometimes
forces more careful attention to logic detail.
Industrial interest in continued progress in
process synthesis techniques is strong. Well-at-
tended symposia sponsored by AIChE and others
throughout the world are becoming more frequent
and serve as a forum for discussion of latest re-
sults. Current research needs include relaxing the
restrictive assumptions particularly related to
type and component distribution in separations
sequencing techniques and developing improved
representations to account for phase equ-
libria and possibly reactions occurring within
separations equipment. Additional work is needed
in the development of complex synthesis strategies
and of heuristic transformation selection and
evaluation functions for efficient state space
searches which avoid unnecessary prejudice or
constraints. Improved performance targets and


bounding properties will be required for search
space reduction and dual-level design optimiza-
tions. Also work needs to continue on combined
synthesis problems such as control strategies and
heat integration.
As more difficult synthesis problems are
examined, more problem-solving techniques may
be required for their solution. Although the goal
of chemical process synthesis research is defi-
nitely not the computerization of the design in-
vention activity, some powerful paradigms have
been developed within the computer science
discipline of artificial intelligence. For example,
pattern recognition and semantic information pro-
cessing were once considered as a means for
coding and retrieving the historical experience
of the profession for application to new problems,
but were abandoned because of the feeling that
novel, creative solutions were unlikely. Likewise,
theorem proving was the foundation of resolution
based flowsheet synthesis but was ultimately re-
jected when structural optimization rather than
proof of feasible design existence was sought.
Learning techniques were incorporated into heur-
istic transformation selection functions and goal-
directed state space search techniques. It is ex-
pected that these and other artificial intelligence
strategies such as planning, reasoning, and game
playing may be an important part of the solution
procedure for more difficult synthesis problems.
Finally, synthesis is a fundamental part of the
synthesis-analysis-evaluation-optimization process
design activity. Although once the exclusive do-
main of the experienced practitioner, the results
of the research reviewed here indicate that at least
some concepts such as decomposition and state
space search can be formalized. And as such, it
seems definitely desirable to incorporate these as
well as some of the specific procedures for se-
lected subproblems such as heat integration, sepa-
rations selection and sequencing, or control
systems along with discussions of concept dis-
covery aids and descriptive chemical engineering
into traditional design courses. In less detailed
form, basic synthesis concepts would seem to be
an excellent introduction to the nature of chemi-
cal engineering, providing an overview and
foundation for the rest of the curriculum.

Chemical process synthesis is an activity con-
cerned with the invention of structurally and

operationally superior design alternatives. This
activity has generally been performed by ex-
perienced, creative engineers assisted by several
design aids based primarily on analogy.
Several approaches not necessarily patterned
after existing concept discovery practice have
been proposed for systematic process synthesis.
Optimization approaches attempt to apply tra-
ditional mathematical methods for operations op-
timization to structural optimization. Evolution-
ary approaches identify features for possible
modification or rearrangement. These approaches
require some preinvented design. This invention
can be accomplished with decomposition or
systematic generation approaches. Perhaps a
truly competent process synthesizer should
contain elements of all three approaches.
Systematic generation involves decomposition
and a subsequent search through very large state
spaces. For some synthesis subproblems, these
searches have become tractable through the de-
velopment of efficient state representations, search
heuristics, and bounds on the state space contain-
ing optimal designs.
Industrially significant results have been
obtained in the area of heat integration. Other
useful techniques have been demonstrated in re-
action path selection, separations sequencing
(particularly combined with heat integration),
control systems, and complete flowsheets. Tra-
ditional concept discovery aids and newer system-
atic synthesis approaches should be included in
process design instruction.
Significant research opportunities remain, par-
ticularly in the area of combined synthesis sub-
problems. Solution techniques for more difficult
problems may require greater application of
artificial intelligence techniques. Tradeoffs
between level of detail and search efficiency will
be difficult and representation and heuristic search
developments must be made carefully to avoid un-
necessary or unintentional constraints. O

1. J. E. Hendry, D. F. Rudd, and J. D. Seader, AIChE
J., 19, 1 (1973).
2. V. Hlavacek, Comput. Chem. Eng., 2, 67 (1978).
3. N. Nishida, G. Stephanopoulos, and A. Westerberg,
AIChE J., 27, 321 (1981).
4. A. W. Westerberg, "A Review of Process Synthesis,"
in Computer Applications to Chemical Process Design
and Simulation, ACS Symposium Series No. 124, 1980.


Continued from page 137.
or related, programs to form a coherent comple-
mentary, or minor, area. All students must spend
one summer in industry in a regular job after their
junior year, and they must also comply with the
Social Service requirement mentioned previously.
However, in contrast with other programs, there
is no thesis requirement for graduation.
At this point, we should mention that similar
programs have been proposed at UNAM, but are
still pending approval for their implementation
[8]. Also, private insittutions (such as UIA) are
already operating with a curriculum whose basic
philosophy coincides with ours, but with differ-
ences in the form of implementations. This is due
to the fact that their faculty is fundamentally
part-time, whereas at UAM-I most of the faculty
is on a full time basis. This brings up the point
that, in order for any program to be successful,
the level of preparation of the faculty members
should be the highest possible, and the composition
must be shifted from primarily part-time to mostly
full time teachers. It is in this respect that the
graduate programs in Mexico have become in-
creasingly important, and therefore should be


A brief review of the development of chemical
industry and of the chemical engineering profes-
sion in Mexico shows that they have been in phase
in terms of supplying the quantity of engineers re-
quired by industry. However, quality has been a
problem, particularly in the last five years.
Mexican industry now requires a different
type of chemical engineer; one capable of
assimilating the imported technologies and de-
veloping new processes more suitable for the
efficient utilization of our resources.
We propose the formation of an "Academic
Commission" on a national level, formed by highly
qualified professors from all parts of the country,
in order to coordinate the design of a curriculum
which could be implemented at all government
sponsored schools. This curriculum should contain
a fundamental core of basic science with a strong
interaction with practice through lab sessions.
The second stage of the curriculum must empha-
size the fundamentals of chemical engineering and,
finally, the third stage can be flexible and con-
centrate on several aspects, depending on the

region of the country or the strength of the faculty
at hand.
It is obvious that the implementation of the
proposed curriculum requires highly trained
teachers and researchers. These people should be
prepared through the graduate programs now
existing in Mexico. Such programs must be
strengthened and should be strongly supported
at the main government sponsored institutions.
An overview of graduate education in Mexico will
be published in a later issue of Chemical Engineer-
ing Education. F

1. Asociaci6n Nacional de Universidades e Institutos de
Estudios Superiores, "Anuario Estadistico," M6xico
2. Rosenblueth, I., "Historia de la Ingenieria Quimica en
M6xico," Research Project Sponsored by SEP,
Mexico (1978).
3. Giral, J., "La Industria Quimica en Mexico," Un-
published Material, M6xico (1977).
4. "Industria Petroquimica Secundaria, Comisi6n Petro-
quimica Mexicana, SEPAFIN, Mexico (1978).
5. Bucay, B., "Recursos Humanos Para la Industria
Quimica," VII Foro Nacional de la Industria Quimica,
Mexico (1974).
6. Diaz-Chavira, R., Aguilar, A., Montes-Paz, J. J., B.S.
Thesis, I.P.N., M6xico (1978).
7. Martinez, E. N., "ChE Education in M6xico-Method-
ology and Evaluation," Chem. Eng. Ed. 11, 78 (1977).
8. Barn6s, F., Bazila, E., Hernandez, M., Martinez, E. N.,
"Anteproyecto de Modificaciones al Plan de Estudios
de la Carrera de Ingenieria Quimica," UNAM (1976).

Continued from page 131.
gram in Health Sciences and Technology, Cambridge,
Mass., January 1976.
4. R. Roy, "Interdisciplinary Science on Campus-The
Elusive Dream," Chem. Eng. News, 55 (35), 28,
5. J. G. Llaurado, "Some Avenues of Training and Re-
search in Biomedical Engineering," Automedica, 1,
193 (1974).
6. P.A.F. White, "Chemical Engineering in Medicine in
North America," Chemical Engineer (London), 281
(April 1978).
7. G. Lindner, "Chemical Engineering i Europa,"
Kemisk Tidskrift, NR12, 44, (1976), (in Swedish).
8. R. Aris, "Academic Chemical Engineering in an
Historical Perspective," Ind. Eng. Chem., Fund., 16,
1, (1977).
9. R. L. Pigford, "Chemical Technology: The Past 100
Years," Chem. Eng. News, 54 (14), 190, (1976).
10. "Chemical Engineering Faculties, 1979-1980,"
A.I.Ch.E., New York, 1979.
11. "Directory of Graduate Research," American Chemi-
cal Society, Washington, D.C., 1979.


Continued from page 101.
semester; one for seniors and one for graduate
students. These tend to be more mathematically
oriented. "The graduate courses are always
pleasant because I usually get quite a few
questions from the students and we have good
discussions. But there is a lot to cover, so I wind
up talking quite a bit," he says.
Bennett's teaching has had great influence
around the world, since he has taught students
on four continents-North and South America,
Africa, and Europe. His first overseas teaching
venture (at the University of Nancy in France)
came in 1952, and reflected the influence of his
continued love of the French language and culture
that developed in his high-school and Yale days.
During his first visit to Nancy, Bennett helped
form that university's chemical engineering de-
partment, according to one of his French col-
leagues, who added, "Having Prof. Bennett with
us was extremely valuable. We benefited from
his American experience and he gave us good
advice on organizing courses and problems, on es-
tablishing laboratory experiments, as well as on
the construction of buildings, and above all, the
unit operations laboratory." During another visit
to France in 1970-71, Bennett participated in re-
search at Nancy on the design of catalytic re-
actors, and "made important contributions in the
conceptual design of laboratory reactors," accord-
ing to a colleague there, who also believes Bennett
played a role in developing chemical engineering
throughout France. He goes on to say, "Thanks to
his perfect knowledge of the French language,
Bennett has many times been consulted by aca-
demic authorities and even national ministries
concerning important decisions in the domain of
chemical engineering. Prof. Bennett's advice has
always been heeded and followed." It is not sur-
prising that the Bennett and Myers textbook is
one of the basic books used by French chemical
engineering students.
France also called to him during his 1977
sabbatical year, which he spent at the University
of Lyon. At Lyon he worked as a laboratory re-
searcher, read a great deal, and got to work with
some of the leaders in the field of catalysis, many
of whom were at Lyon. "I learned a lot there, and
it helped me quite a bit, and influenced my career,
too." Bennett reflects. He still maintains a co-
operative research relationship with the Uni-

versity of Lyon, and exchanges transatlantic
visits with some of its researchers.
Bennett's other extended foreign involvement
was with a country in a quite different situation,
Chile. He spent a period in 1964 at the University
of Santa Maria in Valparaiso (under an Agency
for International Development contract), returned
there in 1972 as an Organization of American
States lecturer, and last visited there in 1979.
Thus he was in a position to watch that country's
descent into turmoil, from the apparent normality
observed on his first visit.
Bennett's teaching visit to Vienna, was
shorter, only about a month in 1971. He also made
two trips to Algeria, where he taught natural
gas processing.
At home Bennett finds pleasure in many things
other than his work. Besides being a "gourmet
eater" (his son Johnathan is a professional chef),
Bennett also enjoys classical music and art. He
started playing piano at the age of eight and can
play Scott Joplin rags, "ineptly" he claims. Other-
wise, his tastes run more to Mozart, Beethoven,
and Brahms. His interest in architecture and art
now has professional guidance, since he married
a UConn art professor and Egyptologist, Jean
Keith, two years after his first wife, Elizabeth
Jane Balch, died. Bennett and Keith were married
August 24, 1979, on the anniversary of the erup-
tion of Mt. Vesuvius, which buried the Roman
city of Pompeii in 79 A.D., "a date quite suit-
able for us," Bennett notes. O

Continued from page 108.
engineering or returning to it. In its area of cover-
age, this book is more convenient to use and more
comprehensive than the handbooks and it has the
very great advantage of the worked-out example
problems. The references, if not last minute, are
at least solid and extensive. And I can also recom-
mend the book for the faculty member called upon
to teach design classes to have on his desk as a
handy reference. The comprehensive problems
given at the end of each chapter will also be useful
to the faculty member.
In summary, this is a solid contribution to the
chemical engineering literature, even if it does
not suggest itself as a vital component of the
undergraduate curriculum. I am certainly pleased
to have a copy where I can reach for it. E



Departmental Sponsors: The following 146 departments contributed
to the support of CHEMICAL ENGINEERING EDUCATION in 1982 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
Bucknell University
University of Calgary
California State Polytechnic
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
Cooper Union
Cornell University
University of Dayton
University of Delaware
U. of Detroit
Drexel University
Ecole Polytechnique (Canada)
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 Gas Technology
Institute of Paper Chemistry
University of Iowa
John Hopkins University
Iowa State University
Kansas State University
University of Kentucky
Lafayette College

Lamar 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
University of New Brunswick
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
University of Puerto Rico
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
Syracuse University
Teeside 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
University of Toronto
Tri-State University
Tufts University
Tulane University
University of Tulsa
University of Utah
Vanderbilt University
Villanova University
Virginia Polytechnic Institute
University of Virginia
Washington State University
University of Washington
Washington 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


Sometimes it's not all it's

cracked up to be.

However, at Union Carbide innovation continues to improve peoples' lives.
Union Carbide pioneered the petrochemicals industry. Today the Corporation's many hun-
dreds of chemicals are used in everything from automobile bumpers to shampoos. A leader in
the field of industrial gases, our cryogenic technology led to the development of the Oxygen
Walker System, which allows mobility for patients with respiratory diseases. Union Carbiders
are working on the frontiers of energy research-from fission to geothermal-at the world
renowned Oak Ridge National Laboratory in Tennessee. Our revolutionary Unipol process
produces polyethylene, the world's most widely used plastic, at one half the cost and one
quarter the energy of standard converting processes.
From sausage casings to miniature power cells, the Union Carbide tradition of innovation
extends beyond research and development activities to our engineering groups, manufactur-
ing operations, and sales forces.
Continued innovation will largely spring from the talents of the engineers and scientists who
join us in the 1980's.

We invite you to encourage qualified students
to see our representatives on campus-
or write to:

an equal opportunity employer

Manager, Professional Placement
Union Carbide Corporation
Old Ridgebury Road
Danbury, Conn. 06817


Full Text