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 )


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

Record Information

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

Full Text

chemical engineering education


Graduate Program Advertisements Begin on Page 321

Award Lectures
BSL Lecture ConocoPhilips Lecture
The Equations (of Change) Don't Change: Future Directions in ChE Education:
But the Profession of Engineering Does A New Path to Glory
to (p. 242) Ip. 284)
W. R. Schowahter Arvind airma

Random Thoughts: Learning by Doing (p. 282)
Felder, Brent
.Exceptions to the Le Chatelier Principle (P. 290)
Corti, Franses
Learning in Industr): Returning as a Professor (p. 310)
Blau. Wankat
A Fluid-Mixing Laboratory for ChE Undergraduates (p. 296)
Ascanzo, Legros, Tanguy
Mixing Writing with First-Year Engineering: An Unstable Solution? (p. 248)
Lebduska, DiBiasio
Factors Influencing the Selection of Chemical Engineering as a Career (p. 268)
Particle Technology Demonstrations for the Classroom and Laboratory I p. 274)
Iveson, Franks
', Development and Implementation of an Educational Simulator: GLUCOSIM (p. 300)
Erzen, Birol, (inar
Sensitivity Analysis in ChE Education: Part 2. Application to Implicit Models (p. 254)
Smith, Missen
A Batch Fermentation Experiment for L-lysine Production in the Senior Laboratory (p. 262)
Shonnard, Fisher, Caspary
Simulation and Experiment in an Introductory Process Control Laboratory Experience (306)
Using Spreadsheets and Visual Basic Applications as Teaching Aids for a Unit Operations Course (p. 316)
Hinestroma, Papadopoulos



on the



http ://

Chemical Engineering Education
Department of Chemical Engineering
University of Florida Gainesville, FL 32611
PHONE and FAX: 352-392-0861

Tim Anderson

Phillip C. Wankat

Carole Yocum

James 0. Wilkes, U. Michigan

William J. Koros, Georgia Institute of Technology


E. Dendy Sloan, Jr.
Colorado School of Mines

Pablo Debenedetti
Princeton University
Dianne Dorland
Rowan University
Thomas F. Edgar
University of Texas at Austin
Richard M. Felder
North Carolina State University
Bruce A. Finlayson
University of Washington
H. Scott Fogler
University of Michigan
Carol K. Hall
North Carolina State University
William J. Koros
Georgia Institute of Technology
John P. O'Connell
University of Virginia
David F. Ollis
North Carolina State University
Ronald W. Rousseau
Georgia Institute of Technology
Stanley I. Sandler
University of Delaware
Richard C. Seagrave
Iowa State University
C. Stewart Slater
Rowan University
Donald R. Woods
McMaster University

Chemical Engineering Education

Volume 37

Number 4

Fall 2003

242 The Equations (of Change) Don't Change: But the Profession of
Engineering Does,
W. R. Schowalter
284 Future Directions in ChE Education: A New Path to Glory,
Arvind Varma

248 Mixing Writing with First-Year Engineering: An Unstable Solution?
Lisa Lebduska, David DiBiasio

254 Sensitivity Analysis in ChE Education: Part 2. Application to Implicit
William R. Smith, Ronald W Missen
274 Particle Technology Demonstrations for the Classroom and Laboratory,
Simon M. Iveson, George V Franks
290 Exceptions to the Le Chatelier Principle,
David S. Corti, Elias I. Franses
300 Development and Implementation of an Educational Simulator:
Fetanet Ceylan Erzen, Gillnur Birol, Ali tinar
316 Using Spreadsheets and Visual Basic Applications as Teaching Aids for a
Unit Operations Course,
Juan P. Hinestroza, Kyriakos Papadopoulos

262 A Batch Fermentation Experiment for L-lysine Production in the Senior
David R. Shonnard, Edward R. Fisher David W Caspary
296 A Fluid-Mixing Laboratory for ChE Undergraduates,
Gabriel Ascanio, Robert Legros, Philippe A. Tanguy
306 Simulation and Experiment in an Introductory Process Control Labora-
tory Experience,
Kenneth R. Muske

268 Factors Influencing the Selection of Chemical Engineering as a Career,
David C. Shallcross

282 Learning by Doing,
Richard M. Felder, Rebecca Brent

310 Returning as a Professor,
Gary Blau, Phillip Wankat

CHEMICAL ENGINEERING EDUCATION (ISSN 0009-2479) is published quarterly by the Chemical Engineering
Division, American Society for Engineering Education, and is edited at the University of Florida. Correspondence
regarding editorial matter, circulation, and changes of address should be sent to CEE, Chemical Engineering Department,
University of Florida, Gainesville, FL 32611-6005. Copyright 2003 by the Chemical Engineering Division, 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, ASEE, which body assumes no responsibility for them. Defective copies replaced if
notified within 120 days of publication. Write for information on subscription costs and for back copy costs and availability.
POSTMASTER: Send address changes to Chemical Engineering Education, Chemical Engineering Department., University
of Florida, Gainesville, FL 32611-6005. Periodicals Postage Paid at Gainesville, Florida and additional post offices.

Fall 2003

R1 lecture

The following is the first BSL Lecture, given at the
University of Wisconsin on October 2, 2001.



But the Profession of Engineering Does

University of Illinois at Urbana-Champaign Urbana, IL 61801

At the outset, I wish to express my thanks for the honor
to be associated with this celebration marking the
appearance of the 2nd edition of Transport Phenom-
ena. The enormous influence of the first edition on chemical
engineering education is so obvious and well known that it
would be gratuitous for me to spend time talking about how
the book transformed chemical engineering education
throughout the world. That is not to say its adoption was im-
mediate or uniform across the country. Similar to Feynman's
Lectures on Physics, the instructors often learned more than
the students. With apologies for the terribly mixed meta-
phors, one could describe BSL as a stew that didn't al-
ways fly well with lightweights.
This talk is meant to be forward-looking rather than back-
ward-reminiscing, but it is clear that included in my qualifi-
cations for delivering this lecture are (a) my pedigree as an
alumnus of this department and (b) a date of birth that puts
me in an ever-contracting pool of candidates who were edu-
cated in pre-BSL history. I therefore choose a few prelimi-
nary stories, which (beyond being amusing) have, I hope,
some value as reminders of the reach and the lasting influ-
ence, often unintended, those of us who teach have on a
large body of students.
Until late in my undergraduate experience, the most in-
fluential instructors I had were, with one exception, graduate
students. Names such as Ednie, Garver, Woods, and Kirk come
to mind. Bob Kirk was an assistant professor. The others were
instructors and World War II veterans, several of them mar-
ried with young families. They were no-nonsense people who
took their teaching duties seriously and, for the most part,

explained the material well. There is a lesson in this bit of
history. In spite of our recoil when we are told that students
are forced to learn from graduate students rather than senior
faculty, instruction from graduate students isn't necessarily a
bad experience. In fact, it has the advantage of being taught
by someone more apt to appreciate student difficulties, often
similar to those endured just a few years earlier by the in-
structor, than is the case when a full professor is in charge.
The "big names" in the department in those days were
Hougen and Marshall. As an undergraduate, I had no contact
whatsoever with either one. In later years, however, I came
to know both gentlemen well-Bob Marshall through AIChE
committees and Olaf Hougen when he joined a Madison re-
tirement center to which my parents had moved. He was a
truly remarkable person.
The second vignette has to do with undergraduate advis-
ing. In my sophomore year I became frustrated because it
seemed all I was doing was rushing from one assignment and
exam to another, without time to reflect on what I was learn-
ing. I went to my adviser, Professor C.C. Watson, and told

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

William R. Schowalter, professor and dean
emeritus of the College of Engineering at the
University of Illinois at Urbana-Champaign, is
currently senior advisor to the president of the
National University of Singapore. He received
his BS from the University of Wisconsin, and
his MS and PhD from the University of Illinois,
all in chemical engineering. He is an authority
in the field of fluid mechanics, especially as it
applies to the processing of polymer melts,
polymer solutions, and colloidal dispersions.


him I wanted to cut back on the number of
courses and stretch out my residence time to
five years. He looked at my grades, then
looked at me, and said there was no reason
for me to take longer than four years. His an-
swer was no-end of discussion, end of ap-
pointment with adviser. Years later I could say
with conviction that his decision was abso-
lutely correct. I can only guess why he re-
fused me, but I suspect he believed I would
use the extra time for anything but "reflect-
ing" on what I was learning, and I was prob-
ably too immature to know what to reflect
about. Beyond that, one needs a balance be-
tween thoroughness and efficiency. Part of
engineering education is learning how to
find that balance.


Engineering education is, by its very na- s
ture, in a continuous state of flux. Appearance
every decade of a definitive report on the fu-
ture of engineering education is as predictable as a sighting
of the first crocuses in Madison near the end of March. The
constant uncertainty over what we teach should not be sur-
prising. Engineers solve problems. If they are successful, those
problems disappear. Then we find new problems to solve.
Although the principles used to solve successive generations
of problems change very slowly, the problems themselves
have different emphases and different details that require
continual fine tuning, and, as with BSL, occasional step
changes in our approaches.
In my remarks, I shall first offer comments on the purpose
of undergraduate engineering education within the context
of a large, selective university such as UW Madison. From
there I wish to specialize the discussion to challenges and
opportunities relevant to chemical engineering departments
within such institutions, looking successively at undergradu-
ate, professional, and graduate (PhD) curricula. Finally, a few
generalizations are offered regarding challenges for compre-
hensive public research universities.

A popular current claim of engineering educators, and one
to which I subscribe, is that our subject is the liberating art of
the 21st century. That claim lays upon the departments making
it a set of curricular responsibilities which, I believe, includes
1) A program of study sufficient for entry-level positions
in engineering practice and engineering-related fields
2) Exposure to "shoulder areas" of engineering
3) Provision for an understanding of professional and
personal ethics
4) Mastery of fundamentals sufficient to pursue graduate

its very n
in a continue
The con
what we
should n

those p
Then we

Fall 2003

study in engineering or related fields
isng These features of an undergraduate engi-
is, by neering education are not without contro-
atue versy, so I should explain why I believe they
ous state are important and valid.

tant .First, we are not in the business of pro-
viding solely a "pre-professional" education,
ty over in spite of the efforts of generations of edu-
teach cators to do so. The educators have failed
ot be because the market has dictated otherwise.
ing. In good economic times, eager companies
solve line up at the nation's best engineering
If l ey schools to nab graduates of four-year accred-
0ssf9 cited programs. Sometimes those graduates
b4kms enter highly focused technical areas in the
ea electronics, chemical, or automotive indus-
fitd tries, and they will probably receive a gen-
fid erous dose of in-house training to sharpen
-0vs ~the generalities learned in their BS educa-
,* tion. Nevertheless, they are hired as engi-
neering graduates, not as products of a gen-
eralized pre-engineering curriculum now
ready for finishing school. Alternatively, other companies are
anxious to hire them for jobs in which the analytical and rea-
soning skills of the graduates will be applied to a broad rather
than narrow and highly disciplinary context. During the past
decade, the consulting firm Accenture has been one of the
largest employers of entry-level engineers graduating with a
four-year BS degree from Illinois.
Second, I refer above to the "shoulder areas" of engineer-
ing. This is an acknowledgment of the fact that engineering
is a profession that cannot be practiced in isolation-it re-
quires a context. Here is where the "liberating art" label can
be argued. As technology forms an ever-deepening influence
on the lives of everyone, it should be expected that an educa-
tion in engineering must provide a foundation for future spe-
cialization in business, law, or (as we are beginning to see)
even the arts. Those connections should not be left to chance.
On the contrary, opportunities to link strictly technical issues
with economic, social, and political factors should be sought
and exploited. Note that this implies more than a simple re-
quirement for students to take x credits of subject y during
their undergraduate education.
The third ingredient referred to pertains to professional
and personal ethics. If an engineering education does not help
us to understand and promote a civilized society, it is not a
liberating art. Again, this emphasizes that engineering is prac-
ticed in a context. Precisely because engineering is so impor-
tant in today's world, a so-called "engineering decision" is
seldom without consequences far beyond the realm of engi-
neering itself. Those decisions have far-reaching economic,
social, and ethical implications, and the choices among alter-
natives are seldom clearly right or clearly wrong.


Fourth, I have required that an undergraduate curriculum
properly serve the person who wishes to compete at the high-
est level in graduate study in engineering or a related field.
This is the true proof of principle when we claim to be edu-
cating engineers. It separates the schools preparing engineers
from the schools preparing dilettantes. My opinions here have
been shaped by many years on the faculty at Princeton. At
that institution, the largest number of undergraduate majors
over a period of several decades were in either English or
history. Clearly, very few of those majors went on to make a
living in either of those fields. Some did, however, and they
became distinguished in their specialty. We should be able to
do the same. It is not necessary for a school of engineering to
choose between producing either specialists or persons so
well-rounded they have no peaks of excellence.
I have laid out a daunting assignment for an undergraduate
engineering program. How can one provide the technical
depth required to excel at subsequent graduate courses in fast-
moving fields, while at the same time give proper attention
to the shoulder areas and gen-
eral intellectual maturation im-
plied in the above remarks? 1. Chemical Change
Let me be the first to admit that Reactor design
it is not easy, and it is prob- Material and energy balance
ably not possible for any but 2. Separations
2. Separations
the most selective of institu- The concept of staging
tions. To accomplish the ends Unit operations
stated here, one needs entering
students with a rare combina- 3. Therocess
tion of mental ability, prepa- Optimization
ration, and maturity. Those Economics
young people surely exist in Figure 1. Distinctive
our society. It is up to the en- Figure ng Distinctri
gineering profession to con-
vince them that study of engi-
neering is worth their atten-
tion. It is when we do not attract a sufficient number of these
students that we say the curriculum is too difficult, too stuffed
with requirements for completion in four years, or too bor-
ing. I should add that this can become a vicious circle. It is
possible to design a curriculum with all of these undesirable
features-one that will dissuade from the outset attracting
the type of students we wish to see.


Undergraduate Education Everything said so far per-
tains to all of engineering. What can one say that is unique to
chemical engineering? This is, in fact, becoming an increas-
ingly difficult question to answer-especially at the gradu-
ate level, as we shall see shortly. There are some distinguish-
ing, if not unique, features of chemical engineering practice
and education, however. For the moment, let us concentrate
on ramifications for undergraduate education. I believe we


e fea
ng e

distinguish ourselves through the following three character-
A focus on chemical change
Well-developed methodologies for describing separation of
mixtures into their components
A systems approach to the design and description of
These three features can be expanded as shown in Figure 1.
Will these characteristics ensure a healthy chemical engineer-
ing profession into the next generation? In my opinion, they
will not, because we have forgotten a critically important in-
gredient: the student. More than anything, a preponderance
of high-quality students has sustained us ever since World
War II. We have operated more-or-less successfully under
the paradigm, "Build a sound curriculum, and they will come."
I don't believe that approach will work in the future. In order
to engage the cream of the student body, we need to educate
students who, upon graduation, will find jobs with satisfac-
tory financial compensation, intellectual challenge (something
that really is rocket science),
and a sense of excitement and
mission, i.e., conviction that
one's work will make the world
Transport phenomena a better place.
(rates of transfer of mass,
momentum, and energy) How do we measure up?
Relative to other engineering
disciplines, job opportunities
for chemical engineers have
been plentiful, if not as legend-
ary as they were for several
years in computer engineering
rtures of a chemical and science. On the criterion of
education. intellectual challenge, we mea-
sure up very well indeed. We
teach fundamental reasoning
rather than rote application of
facts, and students are stretched accordingly. It is with re-
spect to the last item on the list that we fall short. Faculty
members are excited about what they are doing, but the num-
ber of links in a chain of inference between that excitement
and the excitements of engineering practice is perilously large.
This is not true with engineering in general. It happened in
our field because the interests of faculty members and the
needs of practitioners have diverged. Such does not have
to be the case if one is to retain the intellectual challenge
of the second point.
As an example, I have compared in Table 1 the citations of
some academics recently elected to the National Academy of
Engineering in the chemical engineering section (Section 3)
and the electronics engineering section (Section 7). I believe
one finds a much closer identification of the latter group with
the current interests and proclaimed needs of the industry they
represent. Nevertheless, they are presumably conducting re-

Chemical Engineering Education

On the criterion of intellectual challenge, [engineering education] measures] up
very well indeed. We teach fundamental reasoning rather than rote
application of facts, and students are stretched accordingly.

search considered by their peers to be at the highest level
among their cohorts. I don't mean to imply a value judgment
by the comparison, but our profession is at some risk when
there is lack of identification with the areas of commerce we
claim to serve. The link between academic research and com-
mercial needs is much stronger in the bio-related areas of
chemical engineering, and that is one reason for the current
vitality of that subject.
It is universally agreed that bioengineering, biotechnology,
and bioscience are important to the future of chemical engi-
neering. Exactly how that importance is to be acknowledged
in our curricula and research is still an open question, and
indeed, there is no single answer. Pluralism has been a pillar
of strength for the U.S. educational system, and we shall no
doubt see many successful models. Before leaving this sub-
ject, I do wish to provide an often-forgotten historical per-
spective. Fifty years ago, programs in biochemical engineer-
ing, or its equivalent, already existed at Wisconsin and Illi-
nois, and probably at other schools as well. I do not believe
any of them had a large following. Timing is everything!
Bioengineering is not the only cross-cutting subject on
which chemical engineering should have an important influ-
ence. An emphasis on chemical change implies a special in-

Comparison of Citations for Members Recently Elected
to the National Academy of Engineering

Chemical Engineering Section
For advancing our understanding
of the behavior of block copoly-
mers and other polymeric and
complex fluids
For advancing our understanding
of electrokinetic and electro-
hydrodynamic processes ...
For pioneering contributions in
defining and advancing metabolic
engineering ...
For advancing our understanding
of the mechanisms and modeling
of processes that control radiation
in and pollutant control of com-
For elucidating the flow proper-
ties of complex fluids at the
molecular and continuum
levels ...

Electronics Engineering Section
For advances in optoelectronic
devices, detectors for fiber optics,
and efficient LEDs for displays

For contributions to the develop-
ment of CMOS technology .

For contributions to signal and
image processing .. .

For contributions to and leader-
ship in research on micro-
electromechanical systems

For introducing photonic band-
gap engineering and applying
semiconductor concepts to
electromagnetic waves in
artificial periodic structures

terest and competence in things at the molecular and near-
molecular length scale, a current area of great promise in tech-
nology. At the other end of the length scale are macro-projects
associated with a systems approach. Although chemical en-
gineering education has profited from its close-knit structure,
perhaps the time has come to consider multiple tracks to a
degree, an approach successfully followed for decades by
electrical engineers.
Professional Education I will say the least about this
because here, for the most part, chemical engineering depart-
ments are involved at the margins. Put differently, if we ex-
celled in this arena, the profession would benefit but the im-
pact would not be overwhelming. Likewise, if we turn our
backs on professional education, the damage done is not life
threatening to our profession or to the universities. To a large
extent, one can say this because other stakeholders have a
firm grasp on professional education: the professional soci-
eties, private firms, and the internal education programs of
most large technology companies.
Having said that, I believe there are significant opportuni-
ties for universities to gain from professional education ini-
tiatives. Stanford is probably the most outstanding example.
Its electrical engineering and computer science departments
and the technology culture in the Palo Alto area have been
inseparable. From that example as well as others, distinct ad-
vantages for an academic department and its inhabitants include
Discretionary income for the providing unit (a for-profit
Opportunity for faculty members to be closer to the firing
line (see above criticism aimed at many chemical engineer-
ing departments)
A chance for students and practitioners to mingle
Chemical engineers may have to accomplish this virtually
rather than actually because of distance and time constraints.
Historically, we have declined to be heavily involved in pro-
fessional education, but there are several interesting examples
that have some features of professional education. They in-
An often-called "Master of Engineering" program, such as
the one at Cornell. These involve some type of project work
in place of a Master's thesis. Typical features of the project
include design, teamwork, and liaison with an industrial
The joint MS degree in chemical engineering between the
University of Illinois at Urbana-Champaign (UIUC) and the
National University of Singapore (NUS). Here, students
from both institutions combine the features of study abroad
and industrial experience in both countries, leading in
approximately 18 months to a Master's degree.

Fall 2003

The MIT Practice School. This venerable program, also
leading to a Master's degree, has been in existence for
generations and is arguably the most influential and effective
of all advanced work-study programs.
These examples indicate that it is possible for profession-
ally oriented education to be both financially and intellectu-
ally rewarding for chemical engineering departments. Issues
of resource allocation, faculty interest and talent, and geogra-
phy, however, all indicate a local decision on the importance of
a professional education component to a department's welfare.
Graduate Education This is the arena that has the deep-

est effect on faculty careers at
a research-oriented university
such as Wisconsin. Research
funding, publications, graduate
students, and professional ad-
vancement and rewards all
hinge on graduate education. It
is also the arena where, I be-
lieve, the distinctiveness of
chemical engineering is rapidly
disappearing-as is the distinc-
tiveness of any brand of engi-
neering. We are witnessing a
secularization of the disciplines
of engineering. When I was a
Wisconsin undergraduate, the

Some Consequences of Ta




Students (and faculty!) become
technically multilingual
Solutions to important research
problems are more likely to
be found
Silo mentality is neutralized

Attractive research topics for

auto industry held sway over mechanical engineers. The
atrium of the Mechanical Engineering Building was deco-
rated with a huge banner hanging from the ceiling and pro-
claiming the latest doings of the Society of Automotive En-
gineers. Electrical engineering graduates went to work for
Wisconsin Electric Power or one of the electronics compa-
nies that at that time had a high concentration in the Midwest
(if I remember correctly, names such as Zenith, Haseltine,
and Collins were prominent), and chemical engineers worked
for chemical or petroleum companies, with a local emphasis
on the paper industry. Contrast that with today: ME's design
disk drives, ChE's work on prosthetic devices, and EE's work
on just about everything.
These examples are drawn primarily from the BS/MS lev-
els, but it is perhaps even more evident at the PhD level. By
probing more deeply into our sub-specialties, we have gone
our separate ways, a bit like swimming through fish traps of
ever-narrower pore size, and now we are finding ourselves
all in the same large trap. It has the name "bio/info/nano." As
a consequence, we are entering a period of great excitement
and opportunity at the frontiers of knowledge. It is a devel-
opment we should welcome rather than fear, but it does pose
challenges for our system of doctoral education. I make no
claim to having all the prescriptions for meeting those chal-
lenges, but I do have some suggestions.
To look for some answers, I recently sought advice from a

colleague, Michael Heath, in our Department of Computer
Science. Mike is director of a large DOE-sponsored effort
housed in the Center for the Simulation of Advanced Rock-
ets (CSAR) and known as the Accelerated Strategic Comput-
ing Initiative (ASCI). The mission of CSAR is to simulate
the behavior of rockets from a systems point-of-view, mean-
ing the problem involves issues of mechanics, combustion,
materials, and aerodynamics. Specific thrust areas are the
province of faculty members from departments of physics
and most of the departments of engineering at Illinois.
Mike, whose own specialty is scientific computing, de-
scribes the work of the Cen-
ter as "nonlinear everything."
_E 2 I asked him how it was pos-
isk-Oriented Research sible to make rational progress
when everything depends on
everything. His response was
[ore difficult to do academic not surprising. He stressed the
-ministration importance of a tight network
disciplinary core can become an for efficient and multichannel
endangered species
communication. It is impera-
tive that people talk with, lis-
nclear lines of reporting and re-
)onsibility for junior faculty ten to, and understand each
[embers other. This requires effort on
evidence time to degree can all sides. The materials people
increase need to appreciate the prob-
lems of the propulsion people,
who in turn need to know the problems and constraints of the
guidance and control people, etc. There must also be an over-
all goal and periodic evaluations of how the group is reach-
ing that goal. As we talked, I gained a new appreciation for
the importance on our campuses of interdepartmental labo-
ratories, centers, and institutes. They have become more than
desirable-they are now essential.
I conclude from these experiences that research themes must
go beyond the "Professor X group" mentality and that inter-
penetration among groups must be real and substantial. This
means that PhD research will need to be "managed" more
effectively than before. Perhaps more of the funding for our
research should be structured along the lines of NIH and be
project- or goal-oriented, as are most of the individual insti-
tutes of NIH. There has been periodic discussion of a similar
structure for NSF, but to the best of my knowledge, serious
consideration has not taken place.
In Table 2 I have shown a balance sheet for task-oriented
research conducted with graduate students. The capability to
be technically multilingual and to work in cross-functional
groups is a skill deemed critical to survival in contemporary
professional life, and the sooner students become adept at it,
the more valuable their service will be to an employer. This,
by the way, is no less true in academe than in industry. The
reason for this added-value is, of course, because most im-
portant problems today cannot be solved in isolation. Mike

Chemical Engineering Education

Heath's "nonlinear everything" applies
far beyond CSAR. Working against the
assets of Table 2 are the potential liabili-
ties. Note, however, that with the pos-
sible exception of the second item, the
liabilities are driven by structure rather
than by substance. That is not to claim

that alteration of a structure is a simple
matter, or that the present structure does not serve a purpose.
Overcoming the challenges posed in the liabilities column is
perhaps the major task today of the graduate schools of the
major research universities.
There is an additional contemporary issue not identified in
Table 2. Few will deny that our nation's research capability
has been put at risk by the demise of most of our major in-
dustrial research laboratories. In the past, they often carried
out the important task-oriented but fundamental engineering
research essential to a technically advanced civilization's tech-
nology base. Much of that research has now been transplanted
to university campuses. But if corporate shareholders are
unwilling to pay for these admittedly necessary results, who
should? The federal government? A consortium of federal
governments? (The European Union represents one model
of the latter.) This is a question that deserves a better answer
than either industry or government has provided to date. A
step in the right direction would be clearer articulation by
government research-supporting agencies of the relative im-
portance to them of research results and the advanced educa-
tion of graduate students.

Moving to the final item on my agenda, we must not forget
that chemical engineering education is often conducted within
the environment of a research university, which itself is a
dynamic institution. Much has been written about the shape
of research universities in the future, the increasing role of
corporate and philanthropic support, and the need to preserve
excellence in subjects not directly related to economic needs.
For a more global view than is appropriate here, I refer you
to a recent book[31 by James Duderstadt, an engineer and
former president of the University of Michigan.
I do wish to voice a somewhat parochial concern about
universities such as Wisconsin and Illinois, and that is the
ever-widening gap in resources between the top-tier private
and the top-tier public research universities in this country.
The former had extraordinary endowment growth during the
1990s. That growth will, of course, erode in down-markets,
but the miracle of compound interest is such that I fear the
public will never catch up.
A few years ago, Illinois made a comparison of faculty sala-
ries at different ranks among leading public and private uni-
versities. The most dramatic result of this comparison was a

... an undergraduate curriculum [should] properly serve the
person who wishes to compete at the highest level in
graduate study in engineering or a related field.
This ... principle .. separates the schools preparing
engineers from the schools preparing dilettantes.

shift that has occurred during the past twenty years. Twenty
years ago there was a healthy mix of publics and privates
among the top performers. Today, private institutions domi-
nate those providing the highest faculty salaries. Salaries, of
course, do not alone reflect the quality of a research univer-
sity, but they are, over time, an important indicator.
Midwestern universities, in particular, need to reaffirm their
desire to compete with the best and to convince their citizens
of the value these universities add to their states. We need a
new articulation of the land-grant idea, probably in a con-
certed way across several states. This is too large an issue to
be appropriate for more than a passing comment in a talk of
this type, but it will surely affect chemical engineering edu-
cation at universities such as ours.
So what does all of this mean for Transport Phenomena II
and the University of Wisconsin's place in the history of
chemical engineering education? The clear and illuminating
developments of momentum, energy, and mass transfer found
in Transport Phenomena I are intact in the second edition.
Those concepts and the tight coupling between them will
surely remain in what we consider the chemical engineering
canon. But people toting the successor to that familiar red (or
in later printings, green) classic must find applications we
haven't dreamed of. If they don't, chemical engineering will
deserve to be devoid of bright, ambitious, competitive, and
interesting students. My own bet will be on the side of a fu-
ture in which Transport Phenomena II will follow its own
laws of diffusion. The subject will penetrate into new areas
of application and enrich them, and it will be students edu-
cated through Transport Phenomena II who will be the agents
of change for diffusion of the subject into the broad sweep of
modern technology.

1. Bird, R. Byron, "Mass, Momentum, and Heat Transfer: The Impact on
Engineering Education," in Recent Advances in the Engineering Sci-
ences, McGraw-Hill, New York, NY (1958)
2. Colton, Clark K., ed., Advances in Chemical Engineering, Vol. 16,
"Perspectives in Chemical Engineering Research and Education,"
Academic Press, Boston, MA (1991)
3. Duderstadt, James J., A University for the 21st Century, The Univer-
sity of Michigan Press, Ann Arbor, MI (2000)
4. Hougen, O.A., "Seven Decades of Chemical Engineering," Chem. Eng.
Prog., 73, 89 (January, 1977)
5. Servos, John W., "The Industrial Relations of Science: Chemical En-
gineering at MIT, 1900-1939," Isis, 71, 531 (1980)
6. Servos, John W., Physical Chemistry from Ostwald to Pauling,
Princeton University Press, Princeton, NJ (1990) 0

Fall 2003

rje -1curriculum



An Unstable Solution?

Worcester Polytechnic Institute Worcester, MA 01609

Most first-year students have little in-depth knowl-
edge of their chosen profession-particularly in
engineering, which has so few high school experi-
ences connected to it. Moreover, chemical engineering de-
partments rarely offer core courses until the sophomore year
and hence have little contact with first-year students inter-
ested in chemical engineering. Recently, more departments
have begun offering seminars or other career-oriented activi-
ties for first-year students,M' recognizing that early engage-
ment with the profession can increase motivation for learn-
ing and improve retention in the major.[2,31 Improving student
understanding of engineering should certainly allow students
to make informed, rational decisions about their academic
and professional careers, but providing them with such an
understanding can be challenging and too often devolves into
passive activities such as seminars and introductory techni-
cal courses. By contrast, a process that engages students ac-
tively in learning about and identifying with engineering
would benefit both them and the profession.
Students' ability to identify with their chosen profession
improves both motivation for learning and retention in the
major and also seems to influence their ability to write effec-
tively. Science writing is often influenced by "a student's in-
adequate sense of self as scientist,"141 and a similar rhetorical
struggle would be expected for students in engineering disci-
plines. If engineering students do not view themselves as
engineers, they cannot become fully aware of the audience to
which they are writing and the specific needs of that audi-
ence. Consequently, they approach engineering writing with-
out adequate knowledge of the language practices that define
their discipline. Traditional writing assignments such as lab
write-ups, while helpful in shaping students' thinking and
identifying what is new knowledge to them, may not help

*Address: Wheaton College, Norton, MA 01766

them adopt professional roles. Lab reports typically are writ-
ten to document completion and understanding of the engi-
neering process. For the most part (and with good reason),
first-year labs do not ask students to write as professionals
but as novices demonstrating skills and knowledge.J15
Educators have addressed engineering students' writing
abilities for over a hundred years, with varying degrees of
success and satisfaction.6'1 Institutions have adopted a range
of approaches to improve students' writing skills, such as
writing-across-the-curriculum (WAC) courses that integrate
technical content with rhetorical analysis. Despite good in-
tentions, however, some of these WAC approaches have nev-
ertheless failed to adequately prepare engineering students
for the types of writing tasks that they will encounter aca-
demically and in their careers. As technologists and human-
ists often use different techniques to teach writing, it may be
difficult for students to incorporate lessons from the humani-
ties into their engineering coursework.[71 Engineers may also
lack the language and understanding of composition stud-
ies to effectively teach the writing process. Offering a
pedagogical balance between engineering and rhetoric is
thus a challenging problem.
At Rensselaer Polytechnic Institute, the chemistry depart-
ment employed writing consultants from the Department of

David DiBiasio is Associate Professor of Chemical Engineering at Worces-
ter Polytechnic Institute. He received his BS, MS, and PhD degrees in
chemical engineering from Purdue University. His educational work focuses
on active and cooperative learning and educational assessment. His other
research interests are in biochemical engineering, specifically biological
reactor analysis.
Lisa Lebduska is the Director of College Writing at Wheaton College in
Norton, MA, where here she is designing a writing-across-the curriculum pro-
gram and contributing to the development of Wheaton's new College Learn-
ing Center. With research interests in computer-mediated literacies and
peer tutoring, she has contributed work to Writing Center Journal and the
anthology Student-Assisted Teaching and Learning.

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

Language, Literature, and Communication
to work with junior-level chemistry ma-
jors on their lab reports in two required
"writing intensive" courses. These con-
sultants met with chemistry faculty to dis-
cuss writing practices in that discipline
before they began offering feedback to stu-
dents, who produced multiple drafts of
their reports before submitting final ver-
sions for grading. The writing focus in this
WAC effort targeted upper-class students
and formal lab writing and resulted in bet-
ter quality lab reports.171 A WAC effort in
the Department of Animal Sciences at the
University of Kentucky similarly targeted
upper-class students through a senior-level
course, but by contrast it emphasized more
"real world" assignments that would help
students recognize the importance of writ-
ing in their discipline-an achievement
that is often sought by WAC endeavors in
engineering and technical programs. The
Kentucky course stressed the importance
of rhetorical context in writing assign-
ments to improve student interest and to
clarify assignment objectives.8'
A much broader, more programmatic
approach to WAC has been undertaken by the Materials Sci-
ence and Engineering Department at Virginia Polytechnic
Institute, which integrates writing and speaking into eight core
courses that students take over a three-year period. The se-
quence used a combination of formal and informal ("inter-
personal") communication assignments, peer writing consult-
ants, and supplemental writing workshops. Their efforts seem
to have contributed to the establishment of a required zero-
credit class for majors that asks students to create a writing
portfolio containing their best work in a variety of modes
from their required classes.[91
Historically, attempts to understand these varying ap-
proaches to writing have resulted in two groups: in one, the
expressivist model, writing is used as a means of teaching
and learning, employing free writing and journals, and in the
other, the "social constructionist model," writing pedagogy
emphasizes disciplinary or workplace conventions. Such cat-
egorization oversimplifies the WAC process, with some re-
searchers turning to an "interactionalist" approach that com-
bines elements of both models. "An interactional approach...
emphasizes that learning is a social process that necessitates
active involvement on the part of both the learner and the
teacher while also emphasizing the contribution of disciplin-
ary knowledge in the transaction.""''"
At WPI, we attempted to adopt a scaled-down version of
this "interactionalist" approach, which had been developed

through a successful collaboration be-
tween humanities and engineering faculty
at Michigan Tech University.o01 Our
interactionalist approach involved using
some writing activities that taught students
to use writing as a means of understand-
ing what they wanted to say and were ex-
ploratory. Other activities, by contrast, in-
troduced them to conventions within the
discipline and encouraged them to learn
and reproduce those conventions. The bal-
ance, in part, is between teaching students
what they need to learn to become practi-
tioners of an inherited discourse while also
giving them the critical thinking skills they
need to question and challenge conven-
tions. Leadership in any field requires in-
dividuals who can go beyond the mere re-
production of knowledge by continually
reexamining the discipline and, when
needed, reshaping it.

Students often think of writing and
speaking strictly in terms of evaluation,
e.g., the lab report or presentation that they
must produce to "prove" that they com-
pleted and understood the science. They have a fairly limited
understanding of what "communication" can be used for. At
the same time, their knowledge of what chemical engineers
actually do is equally limited. Because WPI does not offer
freshman chemical engineering courses or require writing
courses, we wanted to design a course that would actively
engage students in the profession while improving their ap-
proach to and understanding of communication as a prob-
lem-solving tool. Additionally, we needed to recognize that
although first-year chemical engineering majors do not take
any chemical engineering courses, they carry one of the
heaviest academic course loads on campus, a fact that
challenged us to design a one-credit class that would
achieve our pedagogical goals but still attract students.

Jointly taught by a chemical engineering professor and a
writing professor, the course stressed collaboration between
chemical engineering and communication in its design and
its execution. We reasoned that the best way to teach that
communication and chemical engineering should inform each
other was to demonstrate the integration, so we collaborated
on the design and delivery of every assignment. Both instruc-
tors attended every class, so the students would again see the
connection between the two disciplines and not think of
"communciation days" versus "chemical engineering days."

Fall 2003

Course development was funded through a WPI grant (itself
supported by NSF's Institute-Wide Reform Program) the first
year and a Davis Educational Foundation grant the second year.
We offered the course three times over two academic years,
revising it after each offering. About one-third of the declared
majors took the course each offering (7-10 students per se-
mester). We required portfolios each time we taught the
course, but in the second offering we required the students to
submit all of the assignments from the course. Ideally (in
keeping with writing portfolio pedagogy), we would have
allowed the students to select what they felt were their stron-

To assess writing gains and to assess the
reliability of our portfolio assessment, we
used an external writing specialist. A final
evaluation measure involved student
self-assessment as expressed in
their portfolio cover letters.

gest pieces, but because we met only once weekly and the
course was "low-stakes" (only a single credit), there weren't
enough assignments from which to choose. We nevertheless
were able to design assignments about chemical engineering
that would give the students an awareness of audience, intro-
duce them to group writing, peer response, and revision, and
give them practice writing reflective cover letters that would
initiate a metacognitive approach to writing-that is, get them
to think about the process of writing. Additionally, we stressed
class discussion so that students would receive practice com-
municating ideas, responding to others' ideas, and learning
the language needed to participate in the discipline.

The course had several activities that covered a variety of
engineering topics integrated with communication issues. For
the purposes of this paper, we summarize a few of the activi-
ties, then follow with a detailed discussion of two. Our em-
phasis in this paper is on portions of the course dealing with
ethics/professionalism and understanding audience.
We started the course with a scavenger hunt that sent stu-
dent teams to various faculty, the writing center, and some
research facilities such as the electron microscope facility.
Teams collected some technical information from each visit
and gave an informal presentation on their findings.
A visual-rhetoric activity had students describe an assigned
visual element that was related to chemical engineering (e.g.,
a pump) to a partner who had to draw it without looking at it.
This activity gave students experience with precise verbal
communication and active listening, while illustrating some
basic chemical engineering principles. We then debriefed the
class with their sometimes-humorous drawings, their guesses
about what the devices were and what led them to their con-

clusions, and then an explanation about the real function of
the visual element.
To connect visual and verbal skills, students went to the
Unit Operations lab for a demonstration of a pilot-scale dis-
tillation column. Prior to the lab visit, they were asked to
develop and sketch a process for production of fuel-grade
ethanol from a fermentation broth. This exercise introduced
basic separation principles, including staging. The lab dem-
onstration was combined with a quantitative problem assign-
ment and a writing task that integrated all the elements.
This was the first time any of the students had observed
the operation of a larger-than-bench scale piece of chemi-
cal processing equipment.
The follow-up activity to the laboratory visit involved vis-
its to actual industrial facilities. We wanted students to expe-
rience chemical engineering in the workplace and to have an
opportunity to talk with practicing engineers in a more active
way than a standard plant tour allowed. Each team visited a
different site and spent several hours with a WPI alumnus
during a major part of their workday. Companies visited in-
cluded an environmental consultant's site visit, membrane
separations (Sepracor), and stem cell production (Viacell). After
the trips, each group wrote a trip summary and gave a brief oral
presentation to the rest of the class about the experience.
Although the activities described above provided some in-
teresting exercises and opportunities for writing within a tech-
nical context, we really wanted to engage students at a deeper
level. Course logistics and student background prevented
going too far into the details of chemical engineering funda-
mentals, so we took a different route. Two activities, described
below, resulted in some interesting issues and posed some
particularly challenging problems for the instructors. Details
about the course syllabus, assignments, and portfolios can be
obtained directly from the authors at or

0 Ethics, Racism, and Engineering Practice
Civic responsibility, the interaction of technology and so-
ciety, and professional and ethical responsibility are all part
of WPI's educational philosophy, so in the first offering of
the course we attempted to engage the class in issues of work-
place racism. Wanting our students to realize that ethics and
race issues have a place in chemical engineering and in their
education as engineers, we used a campus event featuring a
documentary about racism in Japan and a discussion with its
director, and a real case-study involving a chemical company
and allegations of racism. This exercise provided important
data that only a collaboration would have provided.
The racial homogeneity of WPI, this class, and its instruc-
tors contributed to the impression that racism is something
that occurs elsewhere and is perhaps not a real problem, and
our all-too-brief treatment of the issue did little to counter

Chemical Engineering Education


Our conclusion is that mixing writing and first-year engineering is certainly a stable solution
when the experiment is properly conducted....Ensuring stability takes energy, time,
and commitment from the faculty, however-it's a challenging and
difficult process, but it is rewarding and fun.

that impression. Because the film examined racism in Japan,
our students responded to the issue as if it were a symptom of
Japanese culture in particular. Focusing on the lives of Afri-
can-Americans in Japan and their isolation there, the film
was interpreted by students as an instance of something that
occurs outside the racial democracy of the United States. Our
shift of the discussion to the Texaco racial-discrimination law-
suit['3] did little to alter students' perception that racism was
something that occurred "out there." Although we pointed
out that the Texaco executives who had been accused of mak-
ing racist remarks might have been trained in chemistry or
engineering professions, our students nevertheless discussed
the issue as if it were something that couldn't happen here.
When we shifted the discussion to subtle forms of racism
that we have witnessed, such as unofficial segregation in the
cafeteria or in fraternities, several students offered anecdotes
about their best friends who were of color. We seemed to
have created an atmosphere in which students felt the need to
testify against racism and to represent themselves as among
the enlightened, but our goal had been more to get students
to consider the complexities of racism and to examine how
they operate in the workplace. The exercise suffered from a
larger cultural constraint in which "racism seems always to
be an appendage to the classroom curriculum, something
loosely attached to a course but not quite integral, even when
race is the issue."[141
We have not yet resolved the race issue to our satisfaction
and will continue to explore ways to address it. We might
consider, for example, having students explore how "white-
ness" is often understood as a "non-race" or universal in the
workplace. We might also consider examining race in the non-
managerial levels of the workplace. At the same time, we
consider the exercise successful because it provided us with
information about our students' perceptions that traditional
lab activities cannot provide. Additionally, because the exer-
cise was presented within the context of a chemical engi-
neering class, it sent the message that racism is something
that concerns chemical engineers.
Scheduling logistics and the issues described above caused
us to reconsider our approach to introducing the grayer areas
of professional decision-making. We assumed that a shift from
the larger but harder-to-concretize issue of racism to other
more clearly defined ethical dilemmas might be easier for
students to grasp as an entry point into the profession's com-
plexities. So, in subsequent course offerings, we decided to
focus on a very specific well-defined problem. Using an On-
line Ethics Center web site ,

we designed an assignment to introduce students to common
chemical engineering ethical dilemmas. We used a case study
on "Request to Falsify Data" to generate in-class discussion
about how the engineer in the case study might have responded
if her manager wanted her to falsify data about an environ-
mental oil spill. The writing assignment followed up on this
discussion by asking students to evaluate the problem from
the perspectives of a member of the state's environmental
protection agency, the CEO of the company, company attor-
neys, and members of the community.
Some students seemed surprised that engineering had an
ethical component. As one student noted, "I never expected
[a discussion about ethics] in a department other than Hu-
manities. We discussed a dilemma between one's future ca-
reer and morality as part of the human community. From this
discussion, I learned how ethical issues were involved with
chemical engineering ... I liked the idea that we had to give
opinions from different perspectives."
Another student found himself challenged by a situation
that did not offer any moral certitude. By the end of the course,
he described his dilemma: "It was hard to decide how other
people would react and what they would do ... Why would
they want to jeopardize their career or the company and what
qualities are needed to stand up for what is right?"
Because these exercises did not offer the students any an-
swers, they introduced them to a significant but seldom-dis-
cussed component of chemical engineering as well as a lan-
guage by which to begin considering the issues involved. The
exercises provided practice in understanding and articulat-
ing multiple perspectives of the same scenario as well as
the subjective context in which professional life across
disciplines is situated.

N Understanding Audience
A major group-writing piece involved describing a current
field of chemical engineering research to a general audience.
Student teams were assigned a research area and provided
with at least one technical article describing that research,
major benefits that might come from it, and problems associ-
ated with it. Each team did additional reading and produced
an article written for the campus newspaper that described
the role of chemical engineers in the specific research area.
Some groups interviewed appropriate faculty with expertise
in the area. The writing process allowed us to introduce tech-
niques for collaborative writing, revision, and peer review.
The difficulties of understanding audience in an educational

Fall 2003

context emerged as the students struggled to write to an audi-
ence of peers while recognizing that their professors would
be reading and commenting on drafts. One group was as-
signed the research area of obesity drugs-a topic involving
an interesting combination of medicine, biology, engineer-
ing, and patient treatment. In an effort to engage their pro-
spective peer audience early in the piece and to be funny, the
first draft of their paper appeared with the title in large, bold
font: "What's Up FATTY?" and a lead sentence of "Are you
Fat? If so, read on." Other examples of their humor included
statements such as "The diseases related to obesity include
heart disease, stroke, diabetes, hypertension, and gall BLAD-
DER disease (ooooh!)....Scientists were exstatic [sic] when
they discovered that the drug accts [sic] on the brain like CO-
CAINE!!! Fortunately, it does not have the harmful side af-
fects [sic] (you dope fiend)....Some people who are slightly
overweight (not obese) are very emotionally disturbed be-
cause of society around them projecting the image that to be
thin is better. They could then abuse the drug to become overly
thin. Drugs for the MASSES. New drugs: Fad or PHAT."
To a certain extent, the students' article demonstrated a kind
of "institutional under life," which, in the writing classroom,
is a productive assertion of identity against the one being
taught. Robert Brooke, who adapted the sociological con-
cept to explain student behavior in writing classes, notes that
contrary to teacher responses that see such behavior as detri-
mental to instruction, such rebellion is actually productive
because it indicates that students are acquiring a necessary
critical distance from roles that are imposed on them. Ac-
cording to Brooke, such critical distance helps to form a more
self-aware professional identity: "If the student in a chemistry
class grew to think of herself as someone who thinks in certain
ways to solve certain problems rather than as someone who
must 'learn' equations to pass tests, then the student would be-
gin to see herself as a chemist, and to act accordingly."'I5s
The review process included in-class peer revision and in-
structor comments. Both of those audiences suggested the
writers consider the effect of their language on readers. The
student team needed to recognize that their article's message
could be undermined by inappropriate humor. While some
of the students' peers might have been attracted to an article
designed to entertain them, some of their peers would have
been offended rather than entertained. Additionally, many
newspaper readers seek information rather than amusement.
We tried to point out that the campus newspaper ultimately
serves the entire community and that student writing should
reflect an understanding of that community. Their final ver-
sion was titled "Obesity No More?" and led with "Have you
ever wondered why someone can pig out and stay thin, while
someone else can never seem to maintain a healthy weight?
If so, read on." The subsequent article replaced the earlier
joking tone with one that was more formal: "If the drugs are
approved, chemical engineers will be responsible for design-
ing the necessary processes to produce the drug for the masses.

Chemical engineers would also be working to produce the
drugs more efficiently....Obese people could abuse the drug to
become overly thin because of the influence of society. Society
projects an overwhelming image that being thin is better."
This group's end-of-course portfolios indicated that they
realized, in reflection, their initial drafts were offensive to
some readers, but they also felt the revision process had taken
the life out of the paper. Their cover letters pointed out that
they were not interested in the topic from the beginning and
had tried to find a way to make it interesting to each other:
"Todd and I wanted to make it goofy enough for a college
student, yet we all knew that some of our jokes would go
over badly....We managed to put together a pretty crude pa-
per full of stupid remarks." Rather than reflecting a lack of
understanding of audience, these remarks suggest a kind of
rebellion against it. Hence, their first draft was written suitably
for their intended audience: their group. This draft also suited
their purpose, which was to entertain and be entertained.
The subsequent revisions indicate a kind of capitulation to
the educational system. As that same student noted in his
portfolio letter, "The group got together again and took out
all of the brazen humor to make what I thought was a dry
article." His comments reflect an understanding of the edu-
cational game in which the faculty audience is the final arbi-
ter as well as his refusal or perhaps inability to identify with
that audience. At this stage, he knows what his audience wants,
and given that a grade is at stake, he will give that audi-
ence what it wants, but he will not identify with it. Also,
he cannot fathom how someone would find the subject of
obesity drugs relevant or interesting, but he is willing to
play the language game.
This activity also made us question our experiences with
the racism discussion. Again, those activities reflect the stu-
dents' desire to play the language game, which they inter-
preted as testifying against racism but did not reflect an un-
derstanding of what they themselves did not experience di-
rectly. These students could not imagine racism's existence
any more than they could imagine how someone would want
to read an article about obesity that did not make jokes about
it. Ultimately, both exercises attested to the need for educa-
tion that requires students to imagine conditions and groups
other than themselves as part of their intellectual maturation.

We used several measures to assess student gains in knowl-
edge of the chemical engineering profession and writing ap-
proach. To assess student gains in knowledge of the profes-
sion, an external evaluator administered questionnaires and
conducted focus groups that categorized "knowledge" in three
dimensions: "activities of chemical engineers, industries
employing them, and issues faced by them." To assess writ-
ing gains and to assess the reliability of our portfolio assess-
ment, we used an external writing specialist. A final evalua-

Chemical Engineering Education

tion measure involved student self-assessment as expressed
in their portfolio cover letters.
After the first iteration of the course, the evaluator com-
pared first-year chemical engineering majors who had taken
the course to a control group of first-year chemical engineer-
ing students who had not. These pre- and post-comparisons
were not useful due to the relatively small sample sizes. As a
result, the evaluator turned to focus groups to provide a fuller
understanding of what had happened.
We did not conduct any longitudinal studies, but it has been
clear that students who took this course remained in the de-
partment. Many became active in the student AIChE society
and others were academically outstanding. We believe this
probably has more to do with the students' predisposition for
chemical engineering as a major than the effects of a one-
credit course.

0 Gains in Knowledge
of the Engineering Profession
The evaluator concluded that the project had succeeded in
producing gains in student knowledge of the activities in
which chemical engineers engage. One of the greatest
struggles for the students involved the group writing assign-
ments, which they found difficult to complete because of in-
compatible schedules. Some also felt the course required too
much writing for a single-credit course. In the second itera-
tion of the course we addressed the group logistics problem
by giving them more instruction in collaborative writing,
fewer collaborative writing assignment, and more in-class
time to write collaboratively. We did not decrease the fre-
quency of writing assignments as we felt they were crucial to
achieving our objectives.

10 Gains in Approach to Writing
To evaluate gains in student writing approaches, we de-
signed a portfolio evaluation rubric that we provided to stu-
dents at the beginning of the course. The rubric identified
nine key criteria, each of which was ranked "Superior,"
Good," "Acceptable," or "Unacceptable." A majority of "Su-
perior" rankings earned the portfolio an "A"; a majority of
"Good" earned a "B;" a majority of "Acceptable" earned a
"C," and a majority of "Unacceptable" earned an "NR" ("Not
Recorded," which is equivalent to a fail grade; WPI does not
have a "D" or "F" grade). The portfolio review criteria were
Demonstrates a robust understanding of the chemical
engineering profession
Shows sustained original, logical thinking
Has strong organization at the paragraph and global level
Demonstrates a strong sense of audience and voice; language
is creative and appropriate; uses active voice wherever
Uses grammar and mechanics to enhance meaning; has an
interesting, credible voice
Supports points thoroughly

Takes risks that challenge the reader
Is professionally presented
Is complete and on time
DiBiasio and Lebduska then evaluated each portfolio in-
dependently. That is, we did not share our evaluations until
we had ranked all of the portfolios. Although there was some
disagreement over the ranking of specific criteria for certain
portfolios, our overall rankings of the portfolios corresponded
exactly, suggesting reliability. To further assess the reliabil-
ity of our measures, the external writing specialist evaluated
the portfolios using the same rubric and without knowledge
of our evaluations. With the exception of one portfolio, her
assessments correlated with ours, again suggesting a fair
amount of portfolio assessment reliability. In the case of the
exception, the evaluator assessed a grade of "NR," while
we had each assessed it as a "C." In reviewing the materi-
als, we concluded that our assessments had been influ-
enced by our knowledge of the student, his participation
in class, and the effort we assumed he had devoted to a
low-credit, voluntary course.
The external evaluator of the portfolios concluded that "this
course experience, as reflected in the student portfolios [was]
valuable in contributing to student learning,"['21 but noted that
although the students' portfolio cover letters did reflect on
their learning, they did not demonstrate an understanding of
how the course's various assignments were related. We at-
tempted to address this deficiency by giving clearer letter-
writing guidelines in the second iteration of the course.
Perhaps the greatest insights about the course came from
the students themselves. Most of them recognized the mar-
ketability of the skills the course provided. The following
quotes, which validate our interactionalist approach, are rep-
resentative of what students wrote in their portfolio cover
letters. One student, for example, wrote
Unless an engineer is involved in solitary research and development,
he or she cannot expect to survive in the job market without superior
communication skills. These skills are needed to get hired via an
interview, to coherently and precisely express problems to the brass of
the company, and to write technical reports that management can read
without first acquiring an engineering degree.
Another wrote
On the field trip day I was very excited....The plant tour was
unexpectedly amazing. It was nothing like those I saw in the movies.
Another interesting fact was that the whole building was designed to
be explosion proof even inside the elevator... Chemical Engineering
and Communications class was a very unique opportunity offered to
me. It was nothing like other classes in WPI where I took notes on the
lectures and discussed them in groups, I felt that I learned something
new every class meeting. It was like a combination of different subjects
that would help prepare a future Chemical Engineer for the real world
out there.
And finally
What did I learn from this course? Well, I was exposed to environmen-
tal conservation organizations and I saw equipment used at the
industrial level being implemented to be environmentally friendly....!
Continued on page 261.

Fall 2003

[ME, classroom



Part 2. Application to Implicit Models

University of Ontario Institute of Technology Oshawa, Ontario, Canada LIH 7K4

In Part 1 of this series,t1" we emphasized the importance
of sensitivity analysis (SA) in chemical engineering peda-
gogy and described its application to the class of engi-
neering models expressible in explicit form, y = f(x;p). Here,
in Part 2, we consider applications of SA to the more com-
plex class of models expressed in the implicit form,
f(y, x;p) 0 (1)
where y is the vector of N outputs, x is the vector of J system
variables, and p is the vector of K constitutive parameters.
Implicit models can take many forms; their distinguishing
property is that Eq. (1) cannot be "solved analytically" for y
in terms of the inputs (although we typically assume that the
solution of the equations is unique).
In Part 1, we showed how to use SA to determine and em-
ploy the sensitivity coefficients of the output quantities with
respect to x and to p. In this paper, we similarly discuss SA in
relation to several types of implicit models, including sets of
nonlinear equations, systems of ordinary differential equa-
tions, and unconstrained optimization problems (including
regression analysis). For an explicit model, determining the
sensitivity coefficients is relatively straightforward; for an
implicit model, this is usually a more complex task.
We then demonstrate the use of SA for a particular implicit
model arising in thermodynamics concerning two-phase equi-
librium of a pure substance, for which the underlying model
is a set of nonlinear equations with one system variable and
several constitutive parameters. Since calculation of the sen-
sitivity coefficients for an implicit model is a more complex
task, we focus here on their calculation and use for the sys-

* Part 1 appeared in Chem. Eng. Ed., 37(3), 222, 2003
**University of Toronto, Toronto, Ontario, Canada M5S 3E5

tern variable and for the constitutive parameters. We use the
former to illustrate the use of SA as a unifying theme, in this
case involving thermodynamics; we use the latter to address
items la and lb of Part 1.111 Thus,
1. We show the application of SA to the set of nonlinear
equations for vapor-liquid equilibrium (pure sub-
stance) arising from equating the chemical potentials
and pressures of the coexisting phases. The resulting
implicit model determines the coexistence properties
(output quantities) {p,vg,v e } in conjunction with an
EOS involving the three constitutive parameters:
critical temperature, T., critical pressure, P., and
acentric factor, o. Here, p" is the vapor pressure, and
v9 and vt are the molar volumes of the vapor and
liquid phases, respectively. From this model, we
calculate the first- and second-order sensitivity
coefficients of the output quantities with respect to
the single system variable, T.
2. We show how SA can be used to calculate the
uncertainties of the outputs {pvg,v } in terms of the
uncertainties of the constitutive parameters {T, Pc,

William R. Smith is Professor and Dean of Science at the University of
Ontario Institute of Technology. He received his BASc (Eng. Sci.) and
MASc (Chem. Eng.) degrees from the University of Toronto, and his MSc
and PhD degrees in applied mathematics from the University of Waterloo.
His research is in classical and statistical thermodynamics. He is co-au-
thor of Chemical Reaction Equilibrium Analysis (1982, 1991).
Ronald W. Missen is Professor Emeritus (chemical engineering) at the
University of Toronto. He received his BSc and MSc degrees in chemical
engineering from Queen's University and his PhD in physical chemistry
from the University of Cambridge. He is co-author of Chemical Reaction
Equilibrium Analysis (1982, 1991) and Introduction to Chemical Reaction
Engineering and Kinetics (1999).

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

w } of an underlying three-parameter EOS employed
in the nonlinear equation model for pure-fluid vapor-
liquid equilibrium.

As discussed in Part 1, the implementation of SA requires
calculation of sensitivity coefficients. For an explicit model,
their calculation is relatively straightforward; for an implicit
model, their calculation depends on the particular type of
model. We briefly sketch how sensitivity coefficients are cal-
culated for several implicit models arising in chemical engi-
neering: sets of nonlinear equations, systems of ordinary dif-
ferential equations, and unconstrained optimization. The re-
sulting expressions are scattered in the literature, and it is
useful to present them all here.
For an implicit model defined by a set of nonlinear alge-
braic or transcendental equations
fi(y,x;p)= 0 i=1,2,...,N (2)

the first-order sensitivity coefficients of y with respect
or p are obtained by partial differentiation using the cl
rule. Thus, for the system variables x

@Y~ (af, '
axji axj

k=1 Yk

i = 1,2,..., N; j = 1,2,...,

where denotes evaluation at the solution to Eqs. (2). E
tions (3) are a set of NJ linear algebraic equations in the
sitivity coefficients 3yk/8x.. The result for the sensitivity
efficient ayk/8pj is analogous to Eqs. (3). We illustrate
low the use of Eqs. (3) by means of a numerical example
An implicit model defined by a system of first-order o
nary differential equations (ODEs) is expressed as

dyi gi(y,t; x,p)

i = l,2,..., N

to x

SA can serve as a unifying theme
for various topics involving engineering
models since, among other things, it can
show the relative importance of changes in
input quantities as they affect output
quantities ("the solution").

The corresponding equations for ayax. are obtained by re-
placing pj with x,.
We can also consider the initial conditions of Eq. (5) as
additional constitutive parameters; the sensitivity coefficients
with respect to these are given by the analogs of Eqs. (6) and

d (ayj ) N _i 1( aYk,

(a i
SayJ J(O -

i, j = 1, ,2,..., N

i,j = 1,2,..., N

where 8 is the Kronecker delta.
Equations { (6)(7) } and { (8)(9) } are initial-value problems
J for sets of first-order nonlinear ODEs for the sensitivity co-
efficients, which may be solved numerically simultaneously
(3) with the model, Eqs. {(4)(5)}. They appear in the literature
in various places relating to differential equations; a relatively
qua- early treatment is given by Cukier, et al.121
co- For an implicit model defined by an unconstrained optimi-
be- zation problem


min f(y; p)

the outputs are the values at the optimal solution, y*. The
first-order necessary conditions for optimization are

yi(0)= yoi i= 1,2,...,N (5)
The first-order sensitivity coefficients of Eq. (4) with respect
to the constitutive parameters p are obtained by differentia-
tion of Eq. (4) to give

d y (@iy+ N_ (__gi )(Yk
Sdt=pkj .Y..pj
i = 1,2,...,N; j = 1,2,..., K

~-(a () = 0

(y; p)= 0

i= 1,2,..., N

To calculate the sensitivity coefficients of the optimal solu-
tion to changes in the constitutive parameters p, we can treat
Eqs. (11) as a set of nonlinear equations and apply Eqs. (3) to

N ( 2f *(,y* ( a2f

j=1y,.yj P k a J 1y PkJ
i =1, 2,..., N; k = 1, 2,..., K (12)

Equations (12) are a set of linear algebraic equations for
(7) ay,*/8pk involving the second-order coefficients of f at the
optimum, (82f/yiayj)* and (82f/ayiPk)*'

Fall 2003

We can also consider changes involving functions of the
output variables y in an implicit model. For example, a typi-
cal parameter-estimation problem involving an engineering
model can be viewed as an implicit model in which the pa-
rameters are outputs y obtained by minimizing the sum of
squares of deviations of a model from a set of observed data.
The values of the objective function at parameter values near
the optimal solution are important in determining their joint
confidence regions03J (their uncertainties in a statistical sense).
Thus, the change in the residual-sum-of-squares objective
function, Af, from the optimal value is given approximately
by the Taylor expansion

Af 6

+N N( a~ *8iy
+ = i j= 1 y1a y jJ y y =

N N ( 2 Y
| |_-)l 8yi8YJ (13)
i=1 j=J ayiyj

where the first-order term vanishes because of Eqs. (11).
Equation (13), involving the second-order sensitivity coeffi-
cients, defines an ellipsoidal confidence region for the pa-
rameters for a specified value of Af. This region, defined by
the set of all parameter values such that the right side of Eq.
(13) is less than or equal to the left side, can be viewed as the
parameter region that yields an acceptable uncertainty in the
residual sum of squares.



For a pure fluid, {pI,vg,v } are output quantities arising
from a model consisting of a set of three nonlinear equations
involving an EOS that is assumed to be applicable to both
liquid and gas/vapor phases. The model equations result from
equating the chemical potentials and the pressures of the co-
existing phases (at a given T). The former equality gives rise
to Maxwell's equal-area rule[4] (first enunciated independently
by Maxwell151 and by Clausiust61)

p [T;p](vg[T;p]- v'[T;p])= J P(v,T;p)dv (14)

where P(v,T;p) represents the EOS, and we explicitly denote
the dependence of the outputs on the system variable T and
the constitutive parameters of the EOS, p. (Equation 14 was
given, in effect, by Planck.M7t) The pressure equality results in
two additional equations involving the EOS

p[T;p] = P(vg[T;p], T)

p'[T;p] = P(v[T; p], T)

The numerical solution of Eqs. (14) to (16) is part of the cal-
culations described in the following example.
We now turn our attention to the sensitivity coefficients for
this implicit model. Equations (14) to (16) are three equa-
tions in the three outputs {pf,vg.v e }, with the system vari-
able T, for a given set of constitutive parameters, p (which
we consider to be fixed in this section, and for simplicity
suppress their appearance in the following equations). We
carry out a first-order sensitivity analysis of the model by
differentiating the equations with respect to T, to give a set of
three linear equations for the sensitivity coefficients

(dpa / dT), (dvs / dT) and (dv' / dT)

in the form of Eqs. (3). The notation signifies evaluation at
the solution of Eqs. (14) to (16). These are ordinary (as op-
posed to partial) derivatives, since there is a single system
variable, T.
Differentiation of Eq. (14) (involving differentiation of the
integral) gives
dTP g -vT)= (v,T)dv (17)

dT 'v
Differentiation of Eqs. (15) and (16) (involving application
of the chain rule) gives

dT aT v V ~vTL

dp0 (aP) + (aPY' dv~
dT a T v~ a v )TdT)0a

In these equations, with respect to the derivatives on the right
side, uT denotes "along the saturation curve," and superscripts
g and e refer to evaluation at (vg,T) and (v 1,T) respec-
tively; all quantities are evaluated at the saturation conditions
corresponding to {p",vg,v 1,T}.
The sensitivity coefficients are available analytically from
Eqs. (17) to (19) as

v( v,T)dv
dT (20)


dp0 (rapg
dT (BT), ( v g dp+ ( v)
ap (-P )T dT aT p
,aV JT

Chemical Engineering Education

(ae dT YT )v av jdpa aj
e~T aP dT aT
dT T (a

where the cyclic derivative rule has been used to obtain the
final terms in each of Eqs. (21) and (22). (Equations of the
type of 21 and 22 were obtained by Planck.171)
Equations (20) through (22) can be differentiated again to
obtain the second-order sensitivity coefficients on the satu-
ration curve, which are given by

d2pa I ap ___'dvg) ap dv_
dT2 v 7 T) JdT (aT) AvdT

dpa (dv, (dv) ] vg (a2p, (
dT L -+ r(v, Tr)dv (23)
ddT JdT dT jv aT2, vT2


d2p (02p (2p (dv )2 (2P g(dvg
dT2 12 av2 dT avaT dT,

p )T
(av j


d 2p- (a2p)' (a2p~ ( dvf )2 ( 2p Wdcvt
d1 ~2 2TdT Ja LavaTr J LdT~

aV )T


The sensitivity coefficients of y, with respect to the consti-
tutive parameters p, can be obtained by differentiating Eqs.
(14) through (16) at each T. The normalized sensitivity coef-

ficients {3p'/3pj, avg/pj, 3v I/ap } are given by the analogs
of Eqs. (20) through (22) as

( aP 1
aIn y_ a inp! PJvi tJJ(V'T)dv (26)
aInpj aInp pj ep va-vp

apa ( ap g
a1ny2 lnvg) f V pj Pj j (27)p
a1npj 1alnpj (vS apg

'Dp0 apY
alny3 r(alnfl2 aPJ~ a
In pj I~n pj9ve r afl
( aV )T

The derivatives aP/a3p on the right side are evaluated from
the EOS holding fixed v and T, and all parameters other than
pJ. The derivatives (3P/av)T are evaluated at fixed values of
all constitutive parameters, as in Eqs. (20) through (22).
If we denote the relative uncertainties in the three constitu-
tive parameters by u(ln pJ), the relative uncertainties in the
three output quantities, u(ln yi), are given by the analog of
Eq. (9) of Part 1 '

u2 (lnyi)=J In p u(np) (29)

where we assume that the input uncertainties are uncorrelated.
The upper and lower (95%) uncertainty limits for y, are then
calculated from

yi (upper)= yi exp[2u(ln yi)]; yi (lower)= yi exp[-2u(ln yi)]

As a numerical example of items la and Ic of Part 11 or 1
and 2 above, we consider the calculation of {pFvg,v I} for
5) toluene, their sensitivity coefficients with respect to {T,Pc,() }
as functions of T from the triple-point temperature, Tt, to TO,
and the use of the latter in uncertainty analysis in conjunc-
tion with the Peng-Robinson EOS181

RT a(T)
v-b v(v+b)+b(v-b)
2 2
a(T) = 0.45724 c a(T)

Fall 2003

b = 0.07780 RT

oa(T) = (1+ K(W)[1-(T/Tc)+5l)2

K(Wo) = 0.37464 + 1.54226w 0.26992 X2

The derivatives required in Eqs. (26) through (28) are given
from Eqs. (31) through (35) by

aP RT ( 31nb ) ( alnb 2ba(T)(v-b)
--b + ____
3pj (v-b)2 1npjj l1npj [v(v+b)+b(v-b)]2

a(T) (alna(T) 36)
v(v+b)+b(v-b)l ainp J

a 1na(T)2+ a In a(T) 2+(T)]-5 (37)
SIn Tc 3InT c TcI

a In a(T)_ 1 (38)
n P, 1 (38)
a In Pc

alna(T) alna(T)_ (T '5I In Kc(o)
a 1n o 3ano ) Tc) jL 3lno J

2) _1- (1.54226-0.53984 o) (39)
[x(T)]o" Tc

a-ln =1 (40)

-=-1 (41)
a In Pc

a = 0 (42)

CP) RT 2a(T)(v+b)
lav T (v-b)2 [v(v+ b)+ b(v-b)]2

We have used Maple[91 to calculate the coexistence proper-
ties {p,vg,v I } from Eqs. (14) through (16) and their sensi-
tivity coefficients from Eqs. (26) through (28), with pj =
T ,P,w in turn. A Maple script is available on the web site at
Figure 1 shows the normalized sensitivity coefficients with
respect to the constitutive parameters {P,To,w} as functions
of T, from the triple-point temperature, T, to T., using the
nominal parameter values for toluene of'01 {42.365 kPa,
593.95 K, 0.26141 } (wo was calculated from the vapor pres-
sure equation given by Goodwin"t01). The ordinate values can

Figure 1. Normalized sensitivity coefficients with respect
to {Pc, T,,w for toluene from PR EOS over entire liquid
range (Tt = 293.15 K to T = 593.95 K):
(a) for p"; (b) for v8; (c) for ve.

Chemical Engineering Education


-20 /O

, -30


100 200 300 400 500 40 700

a ina va/ 1n u




-50 i
100 200 300 400 500 600 700
T/K T-



be interpreted as the % change in the output for a 1% change
in the input.1'1
The most important parameter for all three output variables
is TC. For p, (Figure la), the sensitivity coefficient with re-
spect to Tc is negative, and increases in magnitude from about
7% near Tc to over 28.5% at the triple point (T, = 178.15
K1101). The corresponding coefficients of vs (Figure lb) and
v (Figure 1 c) both become infinite in magnitude at Tc. At
lower temperatures, the coefficient of v9 is much larger in
magnitude than that of v 1. The latter coefficient decreases
from 0.9 at Tt to become negative at T = 461 K, and increases
rapidly in magnitude as T approaches Tc. The former coeffi-

Figure 2a. Vapor pressure (p) for toluene (178.15 K(T,) to
593.95 K (T)); central curve (nominal value) obtained from
PR EOS; points are experimentalilol outer curves define 95%
uncertainty bands (see text).

10-3 10-2 101 100 101 102 103 104 105 106 107 10.
v/L mol"
Figure 2b. Liquid-vapor binodal curve for toluene (T-v co-
ordinates); central curve (nominal value) obtained from PR
EOS; points are experimental;(ooJ outer curves define 95%
uncertainty bands (see text); the inset shows the breakdown
of the first-order SA expansion for the uncertainty bands
near TC (see text).

Fall 2003

cient is always positive, starting at 28.5 at Tt, going through a
minimum at T = 486 K, and rapidly increasing in magnitude
as T approaches Tc. Voulgaris, et al.,1"1t have also reported on
the "extreme sensitivity of p to To" for various fluids, al-
though they did not use SA in their investigation.
The sensitivity coefficients with respect to Pc are much
smaller in magnitude than those with respect to T., and all
have constant numerical values (+1 for p" and -1 for v9 and
v K), results not anticipated prior to the numerical calcula-
tions. In retrospect, however, it was realized that this follows
from the fact that the Peng-Robinson EOS is, in effect, a two-
parameter EOS with the acentric factor w incorporated into
the parameter a, via a function of the reduced temperature,
T = T/T For all such EOS, including a strictly two-param-
eter EOS such as the van der Waals, the reduced vapor pres-
sure, p'/Pc, is a universal function 01 of T, and w

pa = P,0l(Tr,o) (44)

Differentiation of Eq. (44) then gives

lnp-1 (45)

as in Figure la.
Similarly, as in Eq. (44), the liquid and vapor saturation
volumes are also universal functions of T and w

vg = v42 (Tr,O));

v = Vc03(Tr,O)

Also, the Peng-Robinson or similar EOS each has a univer-
sal value of the compressibility factor at the critical point

zc-= P(47)

Taking logarithms and differentiating both members of Eq.
(46) and Eq. (47) with respect to In Pc, we obtain

aInv 1ainvy
-1 (48)
aIn P aInPc

as in Figures lb and Ic.
The sensitivity coefficients with respect to w are also much
smaller in magnitude than those with respect to T with that
of p" being the largest. This coefficient (Figure 1a) is nega-
tive and increases from -5.2% at T, to zero at T. The coeffi-
cient of v9 with respect to w (Figure lb) is positive and de-
creases from 5.2% at T, to zero at Tc; for v it is negative and
very small (of magnitude less than 0.06%).
Figure 2 shows the output quantities p" (Figure 2a), and vt
and vs (Figure 2b) in the form of the T-v binodal curve, to-
gether with uncertainty bands calculated from Eqs. (29) and
(30), assuming 2.5% uncertainties in each of the constitutive
parameters, {P,,T,wo}, i.e., with 2u(ln pj) = 0.05. The experi-
mental points in Figure 2a, for comparison with the calcu-

lated PR EOS results (nominal values), are given by
Goodwin."101 The relatively large uncertainty of 2.5% in the
parameters was chosen to illustrate relatively large uncertainty
bands for the outputs. For a species such as toluene, 2.5% is
a much greater uncertainty than arises from experimental
measurements.tl21 If the critical constants must be estimated,
however, the uncertainties may be of comparable or greater
In Figure 2a, the uncertainty bands for p" arise primarily
from the uncertainty in T., as indicated by Figure la. The
spread of the uncertainty bands increases as T -) Tc. This is
because, although the relative sensitivity coefficient decreases
as T increases (Figure la), the value of pa increases more
rapidly. At low temperatures (below about 350 K), the rela-
tively large sensitivity coefficient applies to very small val-
ues of p", and the resulting spread is imperceptible on Figure
2a. For CO2, a substance with a much larger vapor pressure
at T,, the uncertainty bands are more prominent, even for 0.5 %
uncertainty (not shown here).
In Figure 2b (in which the abscissa is a logarithmic scale),
below T. (up to about 550 K), there is a considerable uncer-
tainty spread in the vg values, but this is barely perceptible in
the v e values. The reasons for this correspond to those for
the corresponding spreads in the p6 values. As T. is ap-
proached, the uncertainty bands approach infinite magnitude,
because of the similar behavior of the sensitivity coefficients
of v9 and v e with respect to Tc; these coefficients completely
dominate the determination of the uncertainty bands. In this
region, the first-order SA expansion corresponding to Eq. (30)
gives these as

vg= vg(593.95)exp[2u(ln T) (a In v9

V =vI~ti.~~texpLuI n ve) I
v =V(59.95)xp12u~l~c) aIn T, Ij



The inset shows the breakdown of this expansion near the
critical point, using an even smaller uncertainty in To of 0.5%
(2u(ln T) = 0.01). The solid curves are the actual binodal
curves calculated using parameter values for TC of 1% above
and below the nominal value of T = 593.95 K. The dashed
and dotted curves are the results for vg and v t, respectively,
from Eq. (49a,b); a few degrees below (the nominal) T., they
agree with the solid curves, but near T they become increas-
ingly inaccurate.
This example shows that one must beware when sensitiv-
ity coefficients become infinite in magnitude. The first-order
SA expansion cannot be applied to situations when the out-
put variable is not analytic for some values of the input quan-
tities, i.e., does not have a Taylor series expansion. The phe-
nomenon we have discovered numerically here is related to

the non-analyticity of the quantity (vsat- vc) as a function of
(T- T-).J141

CONCLUSIONS (Parts 1 and 2)
1. Sensitivity analysis (SA) is an important pedagogical
topic that should be explicitly included in the chemi-
cal engineering curriculum in many courses.
2. SA can serve as a unifying theme for various topics
involving engineering models, since, among other
things, it can show the relative importance of changes
in input quantities as they affect output quantities
("the solution").
3. SA is applicable to both explicit engineering models
(Part 1) and implicit ones (Part 2), whether they
involve algebraic, transcendental, or differential
equations, or optimization problems.
4. The elementary aspects involving sensitivity coeffi-
cients can be introduced into the undergraduate
curriculum, since they only require a background in
multivariable calculus. This includes an introduction
to uncertainty analysis.
5. Additional aspects of SA may have to be deferred to
the graduate curriculum. These include overall effects
of changes in system variables on the solution, which
needs a background in linear algebra, which, in turn,
may not be required of all undergraduates.

Financial assistance has been received from the Natural
Sciences and Engineering Research Council of Canada. C.
Stuart assisted with the graphics.

1. Smith, W.R., and R.W. Missen, Chem. Eng. Ed., 37(3), 222 (2003)
2. Cukier, R.I., H.B. Levine, and K.E. Shuler, J. Comput. Phys., 26, 1
3. Draper, R.N., and H. Smith, Applied Regression Analysis, 3rd ed., John
Wiley & Sons, New York, NY (1998)
4. Hanif, N.S.M., G.-S. Shyu, K.R. Hall, and P.T. Eubank, Ind. Eng. Chem.
Res., 35, 2431 (1996)
5. Maxwell, J.C., J. Chem. Soc., 28, 493 (1875)
6. Clausius, R., Phil. Mag., Ser. 5, 9, 393 (1880); 12, 381 (1881)
7. Planck, M., Ann. Physik u. Chem., Ser. 3, 15, 446 (1882)
8. Peng, D.-Y., and D.B. Robinson, Ind. Eng. Chem. Fundam., 15, 59
9. MAPLE is a registered trademark of Waterloo Maple, Inc.
10. Goodwin, R.D., J. Phys. Chem. Ref Data, 18, 1565 (1989)
11. Voulgaris, M., S. Stamatakis, K. Magoulas, and D. Tassios, Fluid Phase
Equilib., 64, 73 (1991)
12. Tsonopoulos, C., and D. Ambrose, J. Chem. Eng. Data, 40, 547 (1995)
13. Poling, B.E., J.M. Prausnitz, and J.P. O'Connell, The Properties of
Liquids and Gases, 5th ed., pp. 2.2-2.26, McGraw-Hill, New York,
NY (2001)
14. Rowlinson, J.S., Liquids and Liquid Mixtures, 2nd ed., Chapter 3,
Butterworth, London, U.K. (1969) 0

Chemical Engineering Education

Mixing Writing with First-Year Engineering

Continued from page 253.
was subjected to morally stimulating situations which made me think,
which is novel and frightening. And finally I was presented with two
projects that would be assigned to everyday chemical engineers. In my
opinion I feel that I have learned something about the chem. eng.
profession and that I must remember to communicate my ideas to
others succinctly and clearly as I take the roller coaster ride of
education towards the tunnel of real life working environments.

For both students and faculty, this course experiment
seemed to move in a promising direction. On a professional
development level, the activity lessened the widening "gulf
of mutual incomprehension" between scientists and human-
ists that C.P. Snow said threatened the quality of intellectual
life."'61 DiBiasio and Lebduska each gained insight into how
the other half lived, into the priorities informing engineering
and humanities education, and on how the two sides, too of-
ten thought of dichotomously, might speak to each other in
the classroom. Equally important was the opportunity to al-
low students to hear the conversation-that is, to experience
chemical engineering as a practice that is informed by hu-
manities values, including clear and ethical communication.
Our conclusion is that mixing writing and first-year engi-
neering is certainly a stable solution when the experiment is
properly conducted. In our opinion, the unstable solution,
represented by segregated technical writing courses and en-
gineering writing that emphasizes only lab reports, is not as
productive. Ensuring stability takes energy, time, and com-
mitment from the faculty, however-it's a challenging and
difficult process, but it is rewarding and fun. The students
will also be challenged, not just by trying to understand a
profession they think they want to pursue, but also by being
engaged in thinking through writing. Generally, that's a new
concept for most of them.
For the most part, the activities we designed accomplished
our original goals while providing us with greater insight into
first-year students. In her evaluation of the portfolios, the
external writing specialist noted

Such opportunities for students to reflect on their learning-what
they learned, what it means, why it is important, etc.-are critical
components of effective portfolios, and they distinguish portfolios
from other kinds of student learned assessment (tests, essays, and
so on)."'21

The course experience, in other words, not only provided stu-
dents with information about chemical engineering, but it
offered them an opportunity to gain knowledge about it-
that is, a means by which they could reflect about the infor-
mation and place it within the context of their overall lives.
Despite problems such as course logistics, students' time
constraints, and a kind of cultural resistance to writing, most
students demonstrated growth in their knowledge of the pro-

fession and their use of communication as a learning tool.
Additionally, we discovered that a collaboration between
seemingly unrelated disciplines aids in faculty development
(an opportunity to see how the other half thinks), but to be
truly effective this approach needs to be transported be-
yond the two involved faculty members to a more global-
ized WAC endeavor.

Recently, the chemical engineering department voted to
expand the course and now offers a full 3-credit introduction
to chemical engineering on a two-year trial basis. The course
counts toward graduation requirements and it is expected to
become a permanent part of the department's curriculum.


1. Roberts, S.C., "A Successful Introduction to Chemical Engineering
First-Semester Course Focusing on Connection, Communication, and
Preparation," Proceedings of 2000 Annual Meeting of AIChE, Los
Angeles, Chemical Engineering in the New Millenium, 406 (2000)
2. Young, V.L., "Technical Communication and Awareness of Social Is-
sues for Sophomores," Proceedings of 2000 Annual Meeting of AIChE,
Los Angeles, Chemical Engineering in the New Millenium, 399 (2000)
3. Yokomoto, C.F., M. Rizkalla, C. O'Laughlin, M. El-Sharkawy, and N.
Lamm, "Developing a Motivational Freshman Course in Using the
Principle of Attached Learning," J. Eng. Ed., 88(2), 99
4. Balley, R., and C. Gelsler, "An Approach to Improving Communica-
tion Skills in a Laboratory Setting," J. Chem. Ed., 68(2), 150 (1991)
5. Herrington, A., reprinted from 1985, "Writing in Academic Settings:
A Study of the Contexts for Writing in Two College Chemical Engi-
neering Courses," Landmark Essays on Writing Across the Curricu-
lum, Charles Bazerman and David Russell, eds., Hermagoras Press,
Davis, CA (1994)
6. Mair, D., and J. Radovich, "Developing Industrial Cases for Technical
Writing on Campus," JAC 6,
7. Lablanca, D.A., "Writing Across the Curriculum: A Heretical Perspec-
tive," J. Chem. Ed., 62(5), 400 (1985)
8. Aaron, D.K., "Writing Across the Curriculum: Putting Theory into
Practice in Animal Science Courses," J. Animal Sci., 74(11), 2810
9. Hendricks, R.W., and E.C. Pappas, "Advanced Engineering Educa-
tion: An Integrated Writing and Communication Program for Materi-
als Engineers," J. Eng. Ed., 85(4), 343 (1996)
10. Flynnm, E.A., K. Remlinger, and W. Bulleit, "Interaction Across the
Curriculum," JAC 17.3,
11. Gurland, S.T., "Bridge Project: Communications and Chemical Engi-
neering," unpublished external evaluation (2000)
12. Williams, Julia, "Final Evaluation: Chemical Engineering and Com-
munication Bridge Course," unpublished external evaluation (2000)
13. "Texaco Settles Bias Suit," Posted 15 November 1996; Accessed 13
October, 1999: texaco_settle/a/index.htm>
14. Villanueva, Victor, "On the Rhetoric and Precedents of Racism," Col-
lege Composition and Communication, 50(4), 645 (1999)
15. Brooke, Robert, "Underlife and Writing Instruction," College Com-
position and Communication, 38(2), 141 (1987)
16. Snow, C.P., The Two Cultures, Cambridge University Press, Cambridge,
England (1959) 0

Fall 2003

[,1=1 laboratory



In the Senior Laboratory

Michigan Technological University Houghton, MI 49931-1295

Biochemical processes are finding increasing applica-
tion in the chemical industry for the production of a
wide variety of products from renewable resources.
These products include pharmaceuticals, consumer and food
products, fuel additives, industrial enzymes, and many oth-
ers. They are typically created using batch processing, a
marked departure from the more traditional continuous pro-
cesses for commodity chemicals. Recent graduates from chemi-
cal engineering programs are finding more opportunities for
employment in industries that use biochemical processes and
perhaps fewer opportunities on a percentage basis in traditional
commodity chemical and petrochemical production.1[1
Biochemical processes are complex, involving multiple
steps in converting raw material into products. In addition,
preparation steps and downstream separations are not typical
of traditional chemical processing. Examples of chemical en-
gineering laboratory experiments using biochemical processes
have recently appeared.[2-41 In these experiments, ethanol is
typically produced in short-duration experiments that are, by
necessity, abbreviated and less complex than most industrial
fermentations. In order to prepare undergraduates for oppor-
tunities in biochemical processing and to provide a labora-
tory experience with a complexity similar to a commercial
process, we have developed a batch fermentation experiment
to produce L-lysine for the senior laboratory.

In this experiment, student groups produce L-lysine, an
essential amino acid, from a glucose minimal defined media
in batch culture (fermentation). In this article, we will de-
scribe the pedagogical approach, the objectives for a semes-
ter-long design of experiments, and key results from the fer-
mentation experiment.

Figure 1.
Medi fermentation
Sterilizer Air Filter experiment
for L-lysine
product from
a defined
A- .. 5-L Batch glucose

David R. Shonnard is Associate Professor in the
Department of Chemical Engineering at Michigan
Technological University. His research and teach-
ing interests are in the areas of environmentally
conscious design of chemical and biochemical pro-
cesses, optimization, life-cycle assessment, and
cell-based in-vitro toxicology. He is coauthor of 2
books in environmentally-conscious design and
over 40 peer-reviewed publications.

Edward R. Fisher received his BSc from Berkeley
in 1961 and his PhD from Johns Hopkins Univer-
sity in 1965, both in chemical engineering. After
teaching for 35 years at Wayne State University and
Michigan Technological University, he recently re-
tired as Professor Emeritus in Chemical Engineer-
ing and Chemical Engineering Technology.

David W. Caspary is Manager of Laboratory Fa-
cilities in the Department of Chemical Engineer-
ing at Michigan Tech University He received his
BSc from Michigan Tech in 1982 and has held
several engineering positions with the ChE de-
partment over the past nineteen years. He is cur-
rently co-instructing the Chemical Plant Opera-
tions and Unit Operations Laboratory courses.

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

N2 Air 02

The L-lysine batch fermentation experiment is shown sche-
matically in Figure 1. It is conducted using a 5-liter bioreac-
tor (New Brunswick Scientific BioFlow3000) and a data ac-
quisition and control system (New Brunswick Scientific
BioCommand). With this system, students study the kinetics
of microbial growth and L-lysine production under controlled
conditions of temperature, pH, dissolved oxygen (DO), and
agitation. Auxiliary equipment includes a mobile autoclave
sterilizer (New Brunswick Scientific) and a media
microfiltration unit (Fisher Scientific).
Approximately 60 to 100 senior-year chemical engineer-
ing students annually conduct the batch fermentation experi-
ment in the "Chemical Plant Operations Laboratory" course.
Due to the complexity of this experiment, students work in

L-aspartic aspartyl aspartyl @ L-homoserine L-threonine
acid 4 phosphate semialdehyde

dihydrodipicolinate a-ketobutyrate


----------- L-lysine L-methionine L-isoleucine

Figure 2. Feedback inhibition for regulation of L-lysine syn-
thesis within the cell. Dashed lines indicate feedback inhi-
bition of key enzymes in the metabolic pathway (solid lines).

Experiment Plan for Cell Growth and L-lysine Production
(Amino acid base case values are L-threonine (150 mg/L),
L-methionine (40 mg/L), and L-leucine (100 mg/L)

Glucose Concentration (g/L)
Amino Acid Concentration 20 30
1. Low (50% lower) Team 2 Team 5
2. Base case Team 1 Team 4
3. High (50% higher) Team 3 Team 6

teams comprised of two 4-member groups. The fermentation
experiment requires two to three days of continuous opera-
tion to complete due to the slow kinetics of cell growth and
L-lysine production. Table 1 shows the sequence of events
for this experiment.

The first pedagogical objective for the fermentation experi-
ment is to introduce the students to biochemical process equip-
ment and to explain the key steps for production of a bio-
chemical product. Because most of the graduating seniors have
little or no biochemistry or biochemical engineering experi-
ence, the experiment objectives are geared toward an intro-
ductory treatment. Prior to conducting the experiments, we
give a 2-hour orientation and provide background informa-
tion on L-lysine production using Corynebacterium
glutamicum (American Type Culture Collection, ATCC No.
21253), we conduct a tour of the laboratory, and hold a dis-
cussion of experiment objectives.
We give background information in an oral presentation to
the 8-member student team and describe cell growth in the
context of the major growth stages: lag, exponential, decel-
eration, stationary, and death. Specific metabolic character-
istics of C. glutamicum are described as shown in Figure 2. 51
We further explain that due to a mutation in the cellular DNA
by chemical treatment, this cell cannot convert aspartyl
semialdehyde to L-homoserine. In order to grow the cells on
a glucose minimal medium, L-methionine, L-isoleucine, and
L-threonine must be added in trace amounts. Once these
supplemented amino acids are consumed by growth, any re-
maining glucose is converted to L-lysine rather than cell mass.
We explain that concerted feedback inhibition of the enzy-
matic conversion of L-aspartic acid to a aspartyl phosphate is
relaxed as L-threonine is consumed, thus allowing overpro-
duction of L-lysine. When these concepts are understood, we
tell the students that cell growth and product formation are
expected to occur separately in the batch culture. One of
the objectives for the student teams is to test this hypoth-
esis and also to determine if the amount of supplemented
amino acids controls the maximum concentration of cells
in the fermentation.
The second part of the orientation is a tour of the labora-
tory facilities. We describe each piece of equipment and ex-
plain its purpose in the production of L-lysine. We empha-
size the importance of maintaining sterile conditions and show
the students the two methods of sterilization used; steam au-
toclaving for the bioreactor and microfiltration for the growth
media. We discuss scale-up and the need for coordinating
processes at smaller scales to support production at a larger
scale (e.g., flask-scale cultures for inoculating the fermenter
and the associated equipment). Finally, we explain that the
safety aspects of the laboratory are consistent with Biosafety
Level I requirements (Center for Disease Control, CDC). The

Fall 2003

Schedule for L-lysine Experiment

Orientation Week 1
Proposal preparation Weeks 1 and 2
Pre-laboratory check-in Start of week 3
Laboratory experiment During week 3
Post-laboratory oral presentation During week 4
Final report preparation Weeks 4 and 5

last part of the orientation is a discussion of handout materi-
als (available upon request by e-mail from
) and a schedule for meeting the re-
quirements as outlined in Table 1.
Another pedagogical objective is to test the effects of ini-
tial glucose and amino acid concentrations on L-lysine pro-
duction and cell growth in a design of experiments. As shown
in Table 2, this design of experiments involves six teams dur-
ing the semester. The goal is to involve the student teams in a

Safety is integrated into all aspects
of the undergraduate chemical engineering
laboratory experience... In the design
phase ... a thorough safety review
of the bioprocessing equipment,
procedures, chemicals, and
biological organisms
was conducted.

continuous improvement exercise and to increase their un-
derstanding of how fermentation parameters affect cellular
growth and L-lysine production. Each team conducts an ex-
periment at different initial glucose and amino acid concen-
trations. During the semester, as experiments are completed
and results become available, sharing the data with the other
student teams is intended to increase the level of understand-
ing about this fermentation process for the entire class. Stu-
dent teams share their results by attaching reports and pre-
sentations to an e-mail to the instructor-the cumulative re-
sults (as shown later in Table 6) are then organized and dis-
seminated by the instructor to the student teams (by e-mail
attachment) during the final days of the semester.

Following the orientation, each team prepares and submits
a proposal in which students demonstrate their familiarity
with the process equipment, the objectives, laboratory safety
(chemical, physical, and biological hazards), sample calcu-
lations, and the market aspects of their product. Because of
scheduling limitations, during the 52-hour experiment the
teams are split into two groups. One group from the team
initiates the fermentation over a 4-hour period. This involves
formulating the growth medium, assembling and autoclav-
ing the bioreactor, sterilizing the medium and transferring it
to the bioreactor using microfiltration, calibrating 02 and pH
probes, and finally inoculating with flask-grown cells. Dur-
ing the next 48 hours, all students in the team periodically
sample for cell growth, glucose consumption, and L-lysine
production (no sampling is done between midnight and 8 a.m.,
Each run in the experiment plan is conducted under identi-
cal conditions of temperature (300C), pH (7.0), dissolved

oxygen (50% of saturation with air), and duration (52 hours).
The experiment objectives given to each team are shown in
Table 3. The maximum specific growth rate is obtained by
applying the Monod equation6]' to the definition of the spe-
cific growth rate, g, as
1 dX
=--- =(1)
X dt
where X is the concentration of cells in the medium (g/L).
The Monod equation is
S= maxK SK A (2)

where tmax is the maximum specific growth rate constant
(hr'), S and A are the concentration of glucose and supple-
mented amino acids, respectively (g/L), and Ks and KA are
the half saturation constants (g/L). At the start of the fermen-
tation, S>>Ks, A>>KA, and therefore 9= imax in Eq. (1). The
solution to Eq. (1) for exponential growth is

n --X =maxt (3)

For cell growth, samples from the bioreactor are taken at
2-hour intervals on the first day and at 4-hour intervals on
the second and third days. Mass concentrations are obtained
by first measuring the absorbance (at 500 nm wavelength,
A0, Milton Roy Spectronic 21D) and converting those val-
ues using the conversion factor, y (g dry cell wt./L) = 0.5x,
where x is A00. Every 8 hours, samples are taken for glucose

Fermentation Experiment Objectives

1. Determine maximum specific growth rate, i.t (hr')
2. Measure glucose consumption (g/L)
3. Measure L-lysine production (g/L)
4. Determine cell growth yield, Yxs
5. Determine L-lysine production yield, Ypis

Major Steps in the Experiment Procedure
for L-lysine Production in Batch Culture

1. Assembly of fermenter and microfilter for steam sterilization
2. Steam sterilization of fermenter and microfilter
3. Media preparation
4. Filter sterilization of culture media
5. Calibration of pH and dissolved oxygen probes
6. Initialize data acquisition
7. Fermentation
8. Sampling for cell, glucose, and L-lysine
9. Analysis of glucose and L-lysine samples
10. Shutdown and clean-up of fermenter

Chemical Engineering Education

and L-lysine analysis by filtering 5 ml of cell culture solution
through a 0.2 pim (polycarbonate, 25 mm dia. Millipore
GTTP02500) membrane and into a closed capped vial (20
ml) to remove cells. These samples are then stored in a re-
frigerator (4C) until the end of the experiment, when they
were analyzed together by the second group of the team. Glu-
cose concentration is analyzed using the hexoskinase/glucose-

Composition of Defined Minimal Media for L-lysine
Production using C. glutamicum.
(All values are per liter of final solution)

20 grams D-glucose
5 g (NH4)2SO4
8 g KHPO,4
4 g KH2PO4
0.2 g MgSO4 7 H20
*1.0 g NaCI
0.5 g citric acid
20 mg FeSO4 7 H,O
50 mg CaCl2 2 H20
150 mg L-threonine
40 mg L-methionine
100 mg L-leucine
I mg biotin
1 mg thiamine HCI
10 ml 100x trace salts

100x Trace Salts Solution: per liter of distilled water
200 mg MnSO4
6 mg HBO3
4 mg (NH4)6Mo070 *4 H2O
100 mg FeC13 6 HO
1 mg ZnSO4 .7 HO0
30 mg CuSO4 5 H20
(pH of this solution adjusted to 2 to avoid precipitation)

25 I deceleration I stationary decline
.. I exponential -
20 2-
C 15 1,E+00 -

1,E-01 o
0 13
0 1,E-02
0 5 10 15 20 25 30 35 40 45 50
Time (hours)
-Glucose Lysine -- -Cells

Figure 3. Results for cell growth, glucose consumption, and
L-lysine production for initial concentrations of 20 g/L of
glucose and base case amounts of amino acid supplements.

Fall 2003

6-phosphate dehydrogenase method (INFINITY Glucose
Reagent, Sigma Scientific) and L-lysine concentration by
using the saccharopine dehydrogenase assay (Sigma Scien-
tific S-9383). The yield of cell growth on glucose consumed
(Yxis) is calculated as Yxis = AX/AS and the data are taken
over the exponential and deceleration growth stages. The yield
of L-lysine produced on glucose consumed (Ypls) is calcu-
lated as Yp/s = AP/AS and the data are taken over the entire
fermentation period, but especially during deceleration and
stationary stages of cell growth (when L-lysine produc-
tion occurs). Although different student groups conducted
the initiation and sample analyses portions of the experi-
ment, the group that was not "on duty" was encouraged
to drop into the laboratory to observe the activities of the
other group, and many students did so when their class
schedules permitted.
The major steps in the fermentation procedure are shown
in Table 4. Table 5 shows the composition of the defined
medium for the fermentation per liter of solution. Handout
materials for this experiment can be obtained in electronic
format (PDF file) by contacting the author at
. Materials available include an over-
view of the semester-long experiment plan, an introduction
to bioprocess safety issues, and detailed steps in the fermen-
tation preparation, start up, and sample analysis.


Figure 3 shows a set of results for the cell growth, glucose
consumption, and product formation for these experiments
using Corynebacterium glutamicum. Cell data shows four
stages of batch growth: exponential, deceleration, stationary,
and declining. Glucose is consumed fastest during the expo-
nential and deceleration stages and more slowly during the
stationary and declining stages. L-lysine production is most
rapid during the deceleration stage and increases to the greatest
amount during the decline stage. This observation is consis-
tent with the metabolic pathway shown in Figure 2, with L-
lysine production in Corynebacterium glutamicum being
greatest after the added amino acids are largely consumed
and cell growth ceases, and during the period that concerted
feedback inhibition of the L-lysine metabolic pathway is re-
leased. The students are made aware of the difference be-
tween growth-associated versus non-growth-associated
product formation. Figure 3 provides an example of mixed
growth-associated product formation-that is, intermedi-
ate between the two types. Results from the remaining
experiments (for the most part) showed similar trends for
the batch culture data.
Table 6 shows the results for all six teams from the semes-
ter-long experiment plan. For the 20 g/L initial glucose con-
centration experiments, the maximum cell concentration de-
creased (from 9.5 to 4.0 g/L) when the initial amino acid con-
centration was decreased by 50%, but cell concentration did

not increase (it decreased slightly from 9.5 to 8.5 g/L) as ex-
pected from the metabolism shown in Figure 2, when the
initial amino acid concentration was increased by 50%. The
absence of this additional cell growth may be due to the in-
crease in L-lysine production. An increase in the initial amino
acid concentration of 50% did increase the ultimate L-lysine
concentration (from 2.09 to 7.5 g/L), whereas a decrease in
the initial amino acid concentration did not significantly
change the L-lysine concentration.
For the 30 g/L initial glucose experiments, again the maxi-
mum cell concentration decreased (from 8.0 to 3.9 g/L) when
the initial amino acid concentration was decreased by 50%,
but (contrary to the 20 g glucose/L results) the ultimate L-
lysine concentration increased (from 2.55 to 10.0 g/L). The
results for 30 g glucose/L and 150% amino acid concentra-
tion were compromised because the dissolved oxygen probe
failed during the run, causing the culture to become anaero-
bic and changing the cell growth and L-lysine production
characteristics. This team proceeded in the same manner as
the other teams. They measured cell concentration, plotted a
cell growth curve, measured glucose consumption and lysine
production, and calculated all growth and yield parameters.
The purpose for doing this in this case was to measure effects
of anaerobic growth conditions on fermentation performance.
The cell growth yield, YxIs, varied from 0.27 to 0.99 for
these experiments, with the exception of the last experiment,
which became anaerobic, as mentioned previously. These
values are in the range typically found for aerobic culture on
glucose and similar growth substrates.161 The highest value
violates a carbon mass balance, however, which predicts a
maximum biomass yield of

Yxis = g bio. 72gC -0.8gbio./gsugar
x/s 0.5 gC 180gsugar

for typical values for biomass dry weight fraction carbon of
49-51%. Most likely, this erroneous result came from mea-
surement error on glucose, as the cell mass measurement is
more accurately obtained. The L-lysine production yield var-
ied over the range of 0.14
to 0.60 for the various ex-

The results from this ex-
periment plan for cell
growth and L-lysine pro-
duction confirm the
student's prior understand-
ing regarding metabolism
for this culture, as shown in
Figure 2. Maximum cell
growth did decline ap-
proximately in proportion
to the decrease in the ini-
tial amino acid concentra-

tion, although it did not increase with increasing amino acid
concentration. Additional experiments are needed to reduce
uncertainty in measured results, which may help explain the
higher-than-possible biomass yield observed in one of the
experiments. Enhanced L-lysine production was observed
compared to the basecase conditions for two experiments, 20
g glucose/L, 150% amino acid concentration and 30 g glu-
cose/L, 50% amino acid concentration. From the results thus
far, however, the exact mechanism for this enhanced produc-
tion is not yet understood.
Table 6, along with a summary narrative of the results from
the entire set of experiments, was developed by the instruc-
tor and disseminated by e-mail attachment at the end of the
semester to the students who participated in the fermentation
experiments. The narrative contained a summary of key re-
sults for these fermentation experiments:
1. The supplemental amino acids limit the maximum cell
concentration that is achieved during fermentation.
2. Cell growth and L-lysine production appear to occur
in separate stages of the fermentation.
3. It is possible to increase L-lysine concentration by the
end of the fermentation by altering initial glucose and
amino acid concentrations.
This end-of-semester summary provided the cumulative
results needed to address the two most important experiment
objectives: testing the hypothesis that maximum cell concen-
tration in the fermentation is affected by the initial concen-
tration of supplemented amino acids and identifying whether
initial glucose and amino acid concentrations could be al-
tered to enhance L-lysine production.
The Department of Chemical Engineering at Michigan Tech
has an assessment program for the evaluation of student learn-
ing outcomes. As required by ABET 2000 Criteria, we use
these assessments to monitor student proficiency in master-
ing chemical engineering fundamentals and for improving
faculty teaching effectiveness.
In this assessment program there are eight major efforts,


Summary of Student Team Results from the Fermentation Experiment Plan
(Base case concentrations of amino acids [L-threonine, L-methionine, and L-leucine) are given in Table 5)

Initial Glucose Concentration (g/L)
Initial Amino Acid Concentration
Maximum L-lysine Concentration (g/L)
Maximum Cell Concentration (g/L)
pm. (1/hr), Max. Specific Growth Rate
Td (hr) Doubling Time
Yxs (g cells/g glucose)
Ys (g L-lysine/g glucose)

Team 1

Team 2
1/2 Basecase

Team 3
150% Basecase

1.82 1.39 2.09

Team 4

Team 5
1/2 Basecase

1.6 1.64
0.27 0.99
0.23 0.60

Team 6
150% Basecase

Chemical Engineering Education

one of which is an assessment of student outcomes in the
Senior Laboratory. From a critical reading of student team
reports by members of the faculty, we evaluate how well stu-
dents communicate in writing, the thoroughness of data analy-
sis and discussion of results, how well they function in teams,
and how proficient they are in understanding the experimen-
tal system, in developing an experimental plan, and in con-
ducting that plan. We include this fermentation experiment
in the assessment plan for the Senior Laboratory.
It is apparent from reading these reports that the students
understand the basic concepts of microbial growth and growth
stages during batch fermentation, concerted feedback inhibi-
tion of enzymes for amino acid production, and growth-as-
sociated and nongrowth-associated product formation. Thus,
from the one-hour orientation and out-of-class readings from
the handout materials, the students appear to be assimilating
and retaining the biochemical concepts needed to interpret
the experimental results. Also, the majority of student teams
have demonstrated that they are up to the task of carefully
executing the detailed experimental procedures provided to
them, although admittedly a good deal of faculty and teach-
ing assistant supervision is required to achieve good results.


Safety is integrated into all aspects of the undergraduate
chemical engineering laboratory experience at MTU. In the
design phase, before any equipment was purchased, a thor-
ough safety review of the bioprocessing equipment, proce-
dures, chemicals, and biological organisms was conducted.
The physical and chemical hazards in this laboratory are com-
mon to other chemistry or chemical engineering laborato-
ries: contact or ingestion of concentrated HCl and NaOH;
flammability hazards; hazards of high-pressure bottled air,
02, and N2; and hazards of poor housekeeping.
In addition to the chemistry laboratory safety concerns,
Corynebacteria glutamicum is classified as a Level 1 bio-
hazard (the lowest biohazard classification). To mitigate the
additional hazards of biological agents, the Center for Dis-
ease Control (-search for biosafety) recom-
mends specific standard practices including
Restricting access to the laboratory
Washing hands with antimicrobial soap prior to
leaving the laboratory
Disinfecting all work surfaces with 75% ethanol after
any spill
Decontamination (by autoclaving or use of 3% bleach)
of all cultures, growth media, equipment, and
disposables after use

Proper preparation of growth media, sterilization of equip-
ment before use, sterile transfer of growth media into the re-
actor, and proper inoculation techniques are critical to the
success of fermentation, and all these aspects expand the stu-

dents' awareness beyond the traditional chemical engineer-
ing experience. Couple this with the biosafety program, and
students are well prepared to enter this exciting area of the
chemical engineering profession.

A batch fermentation experiment to produce L-lysine was
developed for the Chemical Engineering Senior Laboratory
at MTU. The experiment objectives and procedures are ap-
propriate for an introductory treatment of batch fermentation
processes, microbial growth, and metabolism. A semester-
long experiment plan has been implemented to test for the
effects of initial amino acid and glucose concentrations
on cell growth and L-lysine production in long-term ex-
periments (52 hours).
The distribution of tasks between the two student groups
in each team appears to result in a reasonable level of student
effort in this long-term experiment. Judging from the oral
and written reports, the students appear to understand the fun-
damental biochemical principles (provided during a pre-labo-
ratory one-hour orientation and from handout materials) at a
level sufficient to interpret experimental results. Consider-
ing that most students had little or no prior biochemistry edu-
cation, this outcome is viewed as positive.

Funding to develop this experiment was provided by a
National Science Foundation Instrumentation and Laboratory
Improvement grant (Proposal No. 97-50570), by the James
and Loma Mack Endowment Fund, and by the Davis W.
Hubbard Memorial Fund at Michigan Technological Univer-
sity. Several graduate and undergraduate students were in-
volved in the development of these experiments, including
Dale Clark, Amber Kemppainen, Renu Chandrasekaran,
Geoffrey Roelant, and Eileen Kim. Helpful comments by three
anonymous reviewers were greatly appreciated.

1. Cussler, E.L., "Do Changes in the Chemical Industry Imply Changes
in Curriculum?" Chem. Eng. Ed., 33(1), 12 (1999)
2. Shuler, M.L., N. Mufti, M. Donaldson, and R. Taticek, "A Bioreactor
Experiment for the Senior Laboratory," Chem. Eng. Ed., 28(1) 24
3. Badino Jr., A.C., and C.O. Hokka, "Laboratory Experiment in Bio-
chemical Engineering: Ethanol Fermentation," Chem. Eng. Ed., 33(1),
4. Brown, W.A., "Developing the Best Correlation for Estimating the
Transfer of Oxygen from Air to Water," Chem. Eng. Ed., 35(2), 134
5. Araki, K., Amino acids (survey), Kirk-Othmer Encyclopedia of Chemi-
cal Technology, 4th ed., Vol. 2, p. 504, John Wiley & Sons, New York,
NY (1992)
6. Shuler, M.L., and F. Kargi, Bioprocess Engineering: Basic Concepts,
2nd ed., Prentice Hall, Upper Saddle River, NJ (2002)
7. AIChE, Chemical Engineering Graduates-What's Happened to the
Class of 2000?, American Institute of Chemical Engineers, New York,
NY (2000) 1

Fall 2003


rdMe survey




University of Melbourne Melbourne, Victoria 3010, Australia

round the world a range of strategies has been pro-
posed and adopted in an effort to attract more stu-
dents into chemical engineering. These strategies
range from distributing brochures, videos, and interactive CD-
ROMs to secondary school careers teachers, to running ac-
tivities for either the school students or their mathematics
and science teachers.
In Australia in the early 1990s, the Joint Victorian Chemi-
cal Engineering Committee commissioned a short video,
"This is Chemical Engineering," that was later distributed to
all secondary school careers teachers,'11 and in 1997, to cel-
ebrate the 75th anniversary of the Institution of Chemical
Engineering, that UK-based body prepared and distributed a
CD-ROM aimed at secondary school students."t2 More re-
cently the Institution of Chemical Engineers has established
an innovative website aimed at attracting secondary school
students into the profession (found at>).13' The American Institute of Chemical En-
gineers has a similar, but less interactive, site at org/careers/>.'41 Some universities, such as North Carolina
State University, run summer engineering camps for school
students and their teachers.'5'
Rather than targeting the students, another strategy involves
working with secondary school math and science teachers to
raise the profile of the profession in the secondary school
community. The Faculty of Engineering at the University of
Melbourne has followed this strategy since 1994. In that year,
the faculty began running one-day seminars for secondary
school math and science teachers to introduce them to engi-
neering.i6'7' More recently, a book (jointly written by a chemi-
cal engineering academic and four practicing secondary
school math teachers) has been published that introduces
teachers and students of years 9 to 11 (ages 15 to 17) to real
engineering applications of mathematics.181 The problems
presented to the readers relate to the design of a bulk liquid

chemical storage facility, i.e., a tank farm. As another strat-
egy, the Tufts University Center for Engineering Educational
Outreach has been involved in a project pairing graduate-
level engineering and computer science students with sec-
ondary school classroom teachers.191
But, which of these strategies is most effective? While en-
gineering graduates have been surveyed to identify the fac-
tors that led them to study engineering at the undergraduate
level1101 or at the postgraduate level,"11 and to identify the main
work activities in their professional careers,'121 no studies have
been reported in the literature that investigate the career
choices of currently enrolled chemical engineering under-
graduate students.
This paper reports on the results of a survey aimed in part
at identifying the most effective strategies. Between October
2000 and October 2001, over 2,500 undergraduate chemical
engineering students studying at 15 universities in seven coun-
tries were surveyed. The survey sample was drawn from all
year levels and included students who had left their home
country to study. The aims of the two-page survey were three-
To investigate student perceptions of the chemical
engineering profession
To investigate the key factors that influenced the
student's decision to become a chemical engineer

David Shallcross is Associate Professor in the
Department of Chemical and Biomolecular En-
gineering at the University of Melbourne and is
s Associate Dean (International) of the Faculty of
Engineering. The author of three books, he is
active in the secondary school community de-
veloping teaching material aimed at raising the
profile of the engineering profession for school

@ Copyright ChE Division of ASEE 2003

Chemical Engineering Education

Survey Questions

Indicate the extent to which you agree
or disagree with these statements:
* Chemical engineering is a well-paid
* Chemical engineering offers scope to
express my creativity.
I am happy with my choice of chemical
engineering as a career.
* Chemical engineers are concerned with
sustaining/enhancing the quality of our
* Chemical engineering is important to
the well-being of society.
* Chemical engineering will allow me to
work and travel internationally.
* Chemical engineering is different to
what I thought it was when I applied to
enter the course.
* Chemical engineering is a well-
respected profession.
* Chemical engineering is of more value
to society than other forms of
* Chemical engineers need communica-
tion skills of a high standard.
* I would recommend others to study
chemical engineering.
* I expect that within ten years of
graduating, I will have moved out of
engineering into a management role.

I chose chemical engineering because
* I was inspired by a member of my
family to study chemical engineering.
* I was inspired by a "role model"
chemical engineer (not a family
member) whom I admire.
* I really liked chemistry at school.
* Chemical engineering is involved in a
range of diverse industries.
* I wanted to do engineering, but didn't
like/take physics at school.
* I wanted to do engineering, but the
other engineering disciplines didn't
appeal to me.
* A chemistry teacher at my school
triggered my interest in chemical
* A careers teacher at my school
suggested that I consider chemical
* I was inspired by visiting a tertiary
information session/event.
* I attended an engineering camp/summer
school type event.
* Chemical engineering is a "clean" form
of engineering.
* I was told to study chemical
engineering by a family member.
* I will be able to make a positive
difference in caring for the environ-

To determine which of a list offifteen industrial sectors the students most and least
want to work in upon completion of their degree

This paper examines the key factors that influenced the students to choose chemical engi-
neering as their profession. The results from the other sections of the survey are published


The survey consisted of a single-sheet, two-page form prepared in English, German,
Russian, and Vietnamese. It was only given to students currently enrolled in an under-
graduate chemical engineering course. The students were asked to identify their gender,
their grade level, and whether or not they were studying in their own country. The ques-
tions asked of students are shown in Table 1.

Some thirty universities in a range of countries that included Australia, Canada, Ger-
many, New Zealand, Russia, Thailand, the United Kingdom, the United States, and Viet-
nam were contacted and asked to participate in the survey. A number of universities de-
clined for a variety of reasons, including university policies against conducting external
surveys and concerns over the privacy rights of their students. Fifteen of the sixteen uni-
versities that agreed to participate are listed in Table 2 (the University of Hanoi also par-
ticipated in the survey, completing about 500 forms, but they were lost by the Vietnamese
postal service and were not received for processing.) Table 3 summarizes the number of
respondents by gender and national origin. In all countries except Vietnam, the
English-language version of the survey was used.
There is a total of eleven university chemical engineering departments in Australia and
New Zealand, so the four participating in the survey provided a statistically significant
sample of the student population from this region. The same is true for the United King-

Summary of Survey Respondents

Gender Student Origin
Not Not
Country Total Male Female Stated Local Foreign Stated
Total 2584 1538 1037 9 1940 459 185

University of Melbourne Australia 300 150 150 0 230 69 1

McMaster University Canada 82 44 38 0 72 7 3

University of Canterbury New Zealand 65 39 25 1 57 2 6

Imperial College London UK 337 247 90 0 195 138 4

University of Loughborough UK 53 34 19 0 45 6 2

University of Surrey UK 65 43 21 1 32 31 2

Iowa State University USA 235 140 94 1 213 12 10

Fall 2003

dom, in which five of the twenty undergraduate chemical
engineering programs were sampled, and for both Thailand
and Vietnam, which each have only three major programs.
Within the United States there are approximately 160 under-
graduate programs, of which only two were sampled by the
survey, and only two of Canada's twenty programs were sur-
veyed. As a consequence, while the data for Australia, New
Zealand, Thailand, the United Kingdom, and Vietnam can be
considered representative of the countries, the same is not
true for the North American data.
To preserve the anonymity of most of the universities, the
survey responses are grouped together by country or region.
Thus, the University of Canterbury is grouped with the three
Australian universities, the five United Kingdom universi-
ties are grouped together, and the two Canadian universities
are grouped together, as are the two United States universi-
ties. Because of the large number of responses from Ho Chi
Minh City University of Technology, that university is con-
sidered separately. As there is no logical grouping into which
to place the Prince of Songkla University responses, and since
the number of responses is so few as to have little statistical
value, those responses play no further part in this analysis.
In most of the countries surveyed, the decision to study
chemical engineering is usually made while the student is
still enrolled in secondary school. In countries such as Aus-
tralia and Vietnam, students apply for specific courses at their
desired universities while they are in their last year of sec-
ondary school. Because this present study seeks to identify
the different local factors that influence students to take chemi-
cal engineering, this study only considers the responses from
students who are native to the country of their institution.
The responses from students who traveled internationally to
study chemical engineering are not considered. Thus, for ex-
ample, the response of students originally from outside the
United Kingdom who have traveled to study at one of the
UK institutions participating in the survey will not be con-
sidered. Table 3 summarizes the number of respondents for
each of the five university groupings for local (i.e., non-in-
ternational) students. It is the responses from these 1858 stu-
dents upon which the present analysis is based.
In the survey, students were given
statements on why they chose to
study chemical engineering (such
as "I really liked chemistry at Summary of Respon
school") and were asked to indicate
the level to which this influenced
their decision. The survey re- Country 1
sponses were scored "strongly in- Australia/New Zealand
fluenced" 3, "some influence" 2, Canada
"no influence at all" 1, no response
0, and more than one response 0.
To illustrate how the survey results United States
might be analyzed, consider the Vietnam

following example. Of the 220 local Canadian students sur-
veyed, 216 non-zero responses were recorded to the state-
ment above. Of these, 76.4% indicated no influence, 19.9%
indicated some influence, and just 3.7% indicated a strong
influence. The average score for this statement is 1.27 and is
based only on the non-zero responses.
The questions and statements used in the survey were care-
fully selected after consultation with students enrolled at Aus-
tralian universities. The thirteen statements relating to the
choice of chemical engineering as a career are not exhaus-
tive, and other statements such as "My prospects for even-
tual employment will be enhanced by have a degree in chemi-
cal engineering" could have been included. The number of
statements was limited by the desire to keep the survey as
short as possible.

The average scores for each of the thirteen statements pre-
sented in the survey are presented diagrammatically in Fig-
ure 1 for each of the five university groupings. These state-
ments relate to the selection of chemical engineering as a
career and Table 4 presents the equivalent information nu-
merically, but also classified by gender.
Several important points arise from the results:
0- A third of students admitted to being inspired by a mem-
ber of their family. Responses to the statement "I was in-
spired by a member of my family" are shown in Figure 2.
Overall, no more than 11% in each country admitted to being
strongly influenced to take chemical engineering by a family
member. In the U.S., however, significantly more female stu-
dents were influenced by this factor (15.4%) than were male
students (7.4%). Over half the female students admitted to
some form of influence, while two-thirds of the male stu-
dents were not influenced at all by family members.
I- Very few students were influenced by non-family role mod-
els. In Australia and New Zealand, nearly 90% of respon-
dents said they were not influenced at all by this factor, while
in Vietnam, nearly 40% were influenced to some degree.
10 One of the two most important factors identified by the

dent Classes Grouped by Country for Students of Local Origin
Gender Engineering Year Level
Not Not
totall Male Female Stated Year 1 Year 2 Year 3 Year 4 Year 5 Stated
458 288 168 2 56 139 137 123 0 3
220 127 93 0 0 72 59 59 29 1
448 345 103 0 159 111 106 68 2 2
254 149 104 1 55 42 41 83 31 2
478 239 234 4 1 82 168 159 64 4

Chemical Engineering Education

Inspired by family member

Inspired by role model

Liked chemistry at school

Diverse range of industries .

Didn't like/take physics

Didn't like other disciplines

Chemistry teacher J

Careers teacher

Tertiary information event
0 Australia & New Zealand
Attended engineering camp =0 Canada
0 United Kingdom
Chemical engineering is clean l0 United States

Told to do chemical engineering

Positive difference for environment

1.00 1.20 1.40 1.60 1.80 2.00 2.20 2.40 2.60 2.80 3.00
Average Response Score
Figure 1. Average scores for each university groupings regarding the level of influence
in selecting chemical engineering. A response of "No influence at all" corresponds to
a score of 1 and "Strongly influences" to a score of 3.

Summary of Average Scores Regarding the Level of Influence
of Factors in Selecting Chemical Engineering

and New
Gender Zealand

Inspired by a family member

Really like chemistry at school

Didn't like/take physics at school

Chemistry teacher triggered interest

Inspired by visiting tertiary information session/,

Chemical engineering is a "clean" form of engin

Able to make a positive difference to environme



_. P. m -faal,
event Male
S Fe l .

eering Male

nt Male

- 20t0





United United
Canada Kingdom States Vietnam

- 1.82


- ..25






1 19


_- .17 .



.- _.32


survey was, not surprisingly, that
the students liked chemistry at
school. Nearly 70% of the U.S. re-
spondents said they were strongly
influenced by their school chem-
istry experiences, with just 5.1%
admitting that they were not in-
fluenced at all. The percentages of
Australian and New Zealand, Ca-
nadian, UK, and Vietnamese stu-
dents who were not influenced at
all by this factor were 12.3%,
18.0%, 12.7%, and 17.4%, respec-
tively (see Figure 3). There was
little difference in responses be-
tween genders.

- The second major factor iden-
tified by the survey was that stu-
dents perceived that chemical en-
gineering is involved in a diverse
range of industries. Care must be
taken, however, in interpreting the
responses to this particular state-
ment. Of the thirteen statements
in this section of the survey, all
but two (including this one) were
framed directly in response to the
opening statement, "I chose
chemical engineering because..."
Thus, it is possible that students
responded to this statement not in
terms of whether or not they were
influenced by it to study chemi-
cal engineering, but whether they
believed the statement to be true.
Nonetheless, the fact that chemi-
cal engineering has application
across such a diverse range of in-
dustries should be emphasized
when recruiting students into the
profession's courses. Across the
five university groupings, the per-
centage of students not influenced
at all by this perception ranged
from 5.7% in Vietnam to 11.5%
in the U.S. Outside Vietnam, more
females were strongly influenced
by this perception than were

I- Across the five groupings, the
percentage of students who iden-
tified that the statement "I wanted
to do engineering but didn't like/

Fall 2003

take physics at school" played no part in the selection of
chemical engineering as a career ranged from 58.0% in Viet-
nam to 74.3% in the UK. In Australia, New Zealand, Canada,
and Vietnam, around 15% of students were strongly influ-
enced by this factor (see Figure 4). There were differences
between the genders in all countries, but particularly in the
UK, where just 8.1% of males were strongly influenced by
this factor, while for females it was 23.0%
1 More female students than males were strongly influenced
by the fact that they wanted to study engineering but the other
disciplines didn't appeal to them. In Canada, 17.6% of males
and 34.1% of females were strongly influenced by this fac-
tor, while the corresponding figures for the UK were 19.8%
and 31.0% for males and females, respectively. Of the 70
students from Australia and New Zealand who were strongly
influenced in this respect, only five indicated they were not
influenced at all. When coupled with the responses to the
preceding statement regarding physics, it is apparent that
chemical engineering owes a significant proportion of its
appeal to the fact that of all the major engineering disciplines,
it is the one in which a sound foundation in physics is the
least important. Kumagai conducted focus group meetings

with over 500 female undergraduate students at eight differ-
ent universities1"31 and found that women who had chosen
chemical or environmental engineering did so because of
negative experiences in physics at secondary schools. Fig-
ure 5 presents the distribution of responses for the five
regional groupings.
- In Australia, New Zealand, and the United Kingdom, the
influence of a chemistry teacher was relatively low compared
to a much greater influence in Vietnam (see Figure 6). Just
one-third of all Australian and New Zealand students re-
sponded that they were influenced to some extent by their
chemistry teachers. This is all the more surprising because
nearly 90% of these students said that they really liked chem-
istry at school. These statistics suggest that in all countries
other than Vietnam, significant opportunities exist to work
with chemistry teachers to raise the profile of chemical engi-
neering as a profession. This could be achieved by running
professional development sessions for chemistry teachers
where chemical engineering is showcased, or by developing
material for the secondary school chemistry classroom that
illustrates how chemical engineers use basic concepts taught
in chemistry in real-life applications.

Distribution of Responses to Statements

Figure 2. "I was inspired by a
member of my family."


Aust/NZ Canada UK US Vietnam
D Strong influence M Some influence D No influence
Figure 5. "I wanted to do engineer-
ing, but the other disciplines
didn't appeal to me."

Figure 4. I wanted to do engineering,
but didn't like/take physics."

Aust/NZ Canada LK US Vietnam
O Strong influence Z Some influence ONo influence
Figure 6. A chemistry teacher at my
school triggered my interest."

Aust/NZ Canada UK US Vietnam
E Strong influence Some influence ONo influence
Figure 7. "A careers teacher
suggested I consider
chemical engineering."

Chemical Engineering Education

Perhaps the most important conclusion that can be drawn from this survey relates
to the different extent that an enjoyment of chemistry at school and the
role of chemistry teachers influences students to
study chemical engineering.

0 School careers teachers play a relatively insignificant role
in steering students toward chemical engineering (see Figure
7). In North America, 87% of all respondents were not influ-
enced at all by them, while in the UK they were more effec-
tive, having had some level of influence over 36% of the stu-
dents. It is possible that many of the North American respon-
dents misunderstood the term "careers teachers" as this is
not a term in common use there. In other parts of the world,
the term is well understood. One conclusion that can be drawn
from these results is that the institutions and professional
bodies in all countries should work more closely with ca-
reers teachers.
> Many educational institutions put a lot of effort into events
such as university open days. The results presented in Figure
1 and Table 4 show that in Vietnam, in particular, these events
are effective, with 19.8% of Vietnamese respondents being
strongly influenced and 34.6% influenced to some degree.
At the other extreme, only 5.1% of U.S. students were strongly
influenced by such events, compared to 76.4% who were not
influenced at all. It should be noted, however, that the term
used in the survey, "tertiary information session/event," may
have been misinterpreted so that respondents did not con-
sider it to encompass university open days and the like.
- One of the biggest differences in responses between gen-
ders was observed for the role of the engineering camp/sum-
mer school-type event. Just 20% of male UK students indi-
cated they were influenced to some extent by such events,
while nearly 45% of the female UK students were influenced.
A similar difference in responses between genders was ob-
served for U.S. students. In Australia, Canada, and New
Zealand, such events had very little influence.
D "Chemical engineering is a clean form of engineering" is
the second of the two statements not directly framed in re-
sponse to the opening statement. The responses indicate that
this perception has relatively little influence on the selection
of chemical engineering as a career, with less than 7% of
students in each of the groupings being strongly influenced
by this factor. Of those few who were strongly influenced by
this perception, however, some two-thirds were also strongly
influenced by the perception that as chemical engineers they
will be able to make a positive difference in the environment.
- Very few respondents chose to study chemical engineer-
ing because they were told to by a family member. The great-
est degree of influence occurred among Vietnamese respon-
dents. Across all countries, females admitted to being more
strongly influenced than males.

- The perception that the respondents will be able to make a
positive difference in caring for the environment had rela-
tively high average scores across all five groupings. In Viet-
nam, nearly 40% of the respondents indicated that they were
strongly influenced by the perception. Across most country
groupings, females were influenced significantly more than
male students by this perception.
In Australia, high proportions of students are enrolled in
combined degree programs in which they can pursue two
degrees simultaneously. These programs have been described
more fully elsewhere."4' No statistically significant differences
in the factors leading to the selection of chemical engineer-
ing were observed between students currently enrolled in
single and combined degrees.
No statistically significant differences were observed be-
tween students enrolled in different year levels. This is as
expected since the factors leading to a student's selection of
a particular course should not vary significantly in the space
of five years.

The results of this international survey clearly show the
factors that influence a student to study chemical engineer-
ing differ between countries. Some of these differences may
arise due to cultural factors and historical influences. Viet-
namese students are more strongly influenced in their choice
of chemical engineering than students in other countries by
their chemistry teachers, by tertiary information events, and
by the perception that they will be able to make a positive
difference to the environment. In the UK, the role of careers
teachers is much more important than in Australia, New
Zealand, or Canada.
Gender issues are also important, with the responses from
male and female students differing considerably in several in-
stances. A number of workers in the past have studied the gen-
der issues related to course selection at school. Lewis'17' stated
It is at the crucial adolescent age when females seek interrelat-
edness and males seek independence that we ask students to
make their subject choices. Girls who choose the physical
sciences or engineering not only have to show a strong sense of
independence by choosing a nontraditional subject, they are
also asked to choose a set of math and science subjects which
are characterized as abstract laws disconnected from their
social and physical worlds. Boys, on the other hand, can make a
decision in tune with their peer group, and overlapping their
need for emotional separation through disconnected abstract
Continued on page 281.

Fall 2003


S classroom



For The Classroom and Laboratory

University of Newcastle Callaghan, NSW 2308, Australia

ne of the joys of teaching a course on powder tech-
nology is the abundance of quick and simple experi-
ments that can be used in lectures to demonstrate
the fundamental phenomena being discussed. These can be
used as breaks part way through a lecture or as an interest-
arousing introduction to a new topic. Demonstrations can be
used to highlight the often counter-intuitive behavior of pow-
ders by asking the students to break into groups and try to
predict a priori how they expect a given system to behave. The
often quite different behavior that they subsequently observe
will then challenge them to understand the causes of their mis-
conceptions and arouse their interest in the lecture material."'21
There are more than enough such demonstrations to fill a
spot in every lecture of a one-semester introductory course
on powder technology. Most, however, are referred to only in
passing in references scattered throughout the literature1eig,3-6J
or are passed on by word-of-mouth from one practitioner to
another. Klinzingt7' has provided a partial list of such demon-
strations, and a recent CD by Rhodes and Zakharil81 contains
video clips of many others.
This paper seeks to provide a comprehensive compilation
of demonstrations to act as a reference for new instructors in
particle technology. Demonstrations related to wet-powder
systems are presented first, followed by dry-powder systems.
Wet-powder system behavior covered includes single-particle
settling, hindered and lamella settling, sedimentation, the ef-
fect of surface chemistry on slurry rheology, powder wet-
ting, and wet-granule coalescence. Dry-powder system be-
havior covered includes flow from hoppers, percolation and
elutriation segregation, the "Brazil-nut" effect, surface fric-
tion, and powder compaction. Where possible, the source of
the ideas presented is acknowledged, either by reference to a
publication or mention of the person who first told the au-
thors. Many of these ideas have been around so long, how-
ever, that it is difficult to identify their origin, and we apolo-
gize in these cases for not acknowledging their original source.

Single-Particle Settling (in-class demonstration)
Most courses on fluid-particle interactions begin by exam-
ining the settling of a single spherical particle. The effect of
fluid viscosity can be demonstrated by using glass marbles
in two identical perspex tubes about 40 cm long, one filled
with water and the other with glycerol (or any other transpar-
ent viscous fluid).[91 Start by asking the students in groups to
estimate the settling time of each marble. Most will correctly
guess that the marble will settle more slowly in the glycerol.
When asked why, they will probably refer to either glycerol's
greater density or its greater viscosity compared to water.
If students think that density difference is the cause, then a
simple buoyant force balance can be used to calculate how
long it would take the marble to fall under the influence of
gravity alone. For water = 1 g/cm3, pglycerol = 1.25 g/cm3 and
Pglass = 2.5 g/cm3, the increase in fall time in the glycerol due
to its greater density would be only 10%. The calculated set-

Simon Iveson completed his Bachelor of
Chemical Engineering in 1992 and his PhD in
1997, both at the University of Queensland.
Since then he has been a research fellow and
lecturer in the Department of Chemical Engi-
neering at the University of Newcastle. His re-
search interests are in the field of particle tech-
nology, with his focus being on the agglomera-
tion of fine particles by the addition of liquid bind-

George Franks completed his BS in Materials
Science and Engineering at MIT in 1985 and his
PhD in Materials at the University of California
at Santa Barbara in 1997. He has been a senior
lecturer in the Chemical Engineering Department
at the University of Newcastle since 1999. His
research interests include colloidal processing
of ceramics, mechanical behavior of wet pow-
der bodies, and mineral processing processes
such as flocculation.

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

tling times for a 40-cm long tube are of the order 0.4 s, which
is less than the approximate 1 sec and 10 sec observed in
practice. Clearly, buoyancy effects alone do not explain the
speed at which the marbles fall.
At this stage (if they have not already done so), some stu-
dents may recall the concept of viscosity and fluid drag from
their previous fluids courses. This can lead to a brief review
of drag coefficients, terminal velocities, etc. For the marble
in glycerol, the Reynolds number is low and Stokes' law (CD
= 24/Re) applies, Hence, the terminal settling velocity VT of
the marble is given by

VT (pp -L)d2g (1)
where pp and pL are the density of the particle and liquid, d is
the marble diameter, g is gravitational acceleration, and [L is
the fluid viscosity.
For the marble in water, the Reynolds number is of order
105, giving a drag coefficient of approximately CD = 0.44.
The terminal settling velocity is given by

VT- =4(pp-pL)dg (2)
Thus, the difference in viscosity between glycerol (R 1
Pa-s) and water (R. = 0.001 Pa-s) can be shown to be the
major cause of the difference in their settling velocities.
Some students may also think of another cause for the
slower than expected fall of the marbles, namely hindering
caused by the back flow of displaced liquid up the tube walls
as the marble moves past. This is illustrated in the next dem-
Hindered and Lamella Settling (in-class demonstration)
Hindered settling can be illustrated using a pair of perspex
tubes filled with the water, the first containing only a single
small bead and the second a large group of identical beads.

(a) (b) (c)

g ;'

Figure 1. (a) Hindered settling and ways to speed up
settling by (b) tilting the tube or (c) shaking it in a
circular motion.

The settling rates of the two systems can then be compared,
to illustrate how the presence of many particles reduces the
settling speed. Explain how settling is hindered by the need
for the displaced water to flow back up through the bed of
particles (see Figure la); then ask the class to think of ways
to increase the rate of particle settling.
One way to accelerate settling is to tilt the tube slightly so
that particles have only a short distance to fall to reach the
tube wall, where they can then slide down quickly as the dis-
placed water flows unhindered above (Fibure lb).101 This il-
lustrates how lamella settlers operate. Another method is to
shake the tube in a horizontal circular motion while keeping
it vertical. The centrifugal force moves particles to the wall,
leaving the center of the tube free for displaced water to flow
upward, thus allowing the particles to settle more quickly
down the sides (Figure Ic). This is similar to what happens
inside a hydro-cyclone.
Sedimentation and Flocculation (in-class demonstration)
The different types of sedimentation behavior can be eas-
ily demonstrated by filling three tall jars with Type I, Type II,
and colloidal (non-settling) particle suspensions. Shake the
jars at the start of the lecture and point out the different be-
haviors. Type I suspensions are those that form three zones
during settling-a clear liquor above the settling particles
(zone A), a suspension of particles of the same concentration
as the initial suspension (zone B), and a settled bed at the
base (zone S). During settling, the interface between zones A
and B falls and the interface between zones B and S rises,
until the two meet and zone B disappears.
Type II suspensions form four zones during settling. In
addition to zones A, B, and S, there is a zone of variable
concentration (zone E) that forms between zones B and S.
Colloidal particles do not settle out at all. Toward the end of
the lecture, once students are convinced that the colloidal
particles are not going to settle, a flocculant can be mixed
into the suspension to demonstrate the resultant dramatic im-
provement in settling behavior.
Sedimentation and Flocculation (laboratory module)
A laboratory module based on particle settling is also pos-
sible. Give the students three or four samples of silica of dif-
ferent average particle size ranging from about 1 micron to
about 250 microns. The size distributions should be mono-
modal and less than one decade in breadth. Prepare 250 ml of
a suspension of 3 wt% solids for each powder in distilled
water in 150-ml graduated cylinders. Adjust the pH to 8.0
with NaOH so that the silica is well dispersed. Shake the
cylinders and observe the sedimentation. Measure the time
required for the first noticeable formation of the sediment
bed. Measure the height of the interface between the clear
supernatant and the suspension as a function of time.
Students will notice that the micron-sized silica does not
settle appreciably in the time available. Mention that for par-

Fall 2003

tiles smaller than about 0.1 microns,
Brownian motion dominates gravita-
tional settling so that a stable suspen-
sion results. Use a suitable cationic poly-
mer to flocculate the suspension so a
clear supernatant results.
The students should calculate the size of
the largest particles in the sample assum-
ing that the time for the first noticeable
sediment bed to form corresponds to the
time that the largest particles settle the dis-
tance from the top of the tube to the bot-
tom. Using that velocity and Stokes' law
or Newton's law, the size of the largest
particles can be calculated. They should
also calculate the size of the smallest par-
ticles based on the velocity of the suspen-
sion/clear supernatant interface. Then pro-
vide them with the measured particle size
distributions of the silica samples and ask
them to compare their calculated largest
and smallest particle sizes with the size
distribution data provided. The compari-
son is surprisingly good.
Interparticle Force Effects on Colloidal
Suspension Rheology
(laboratory module)
Many chemical engineers are not trained
to consider how the chemical nature of the
fluid medium can influence the rheologi-
cal behavior of a suspension. pH is one of
the easiest properties of a slurry to mea-
sure on-line and it can also dramatically
affect suspension properties. A simple
laboratory project that illustrates this ef-
fect by comparing the slumping behavior
of zircon suspensions as pH is varied is
shown in Table 1.
Wetting Behavior of Dry Powders
(in-class activity or laboratory module)
The wetting behavior of liquids on dry
powders is important in applications such
as mixing pigments into paints and the for-
mation of agglomerates in agitated granu-
lators. If the paint pigments do not wet
well, then they will not disperse and in-
stead form clumps of dry powder with
trapped air inside. This detrimentally af-
fects the paint quality. In granulation, the
initial wetting behavior can have a large
effect on the final product size produced
in a granulator. Drops that penetrate the

Laboratory Module: Interparticle Force Effects on Colloidal Suspension Rheology

Few chemical engineers are trained to consider how the chemical nature of the fluid medium can
influence the rheological behavior of a suspension. pH is one of the easiest properties of a slurry to
measure on-line and it can also have a dramatic effect. The students measure the yield stress of a
0.40 volume fraction of solids zircon suspension over a range of pH values. The average size of
the zircon is about 6 microns, so the interparticle surface forces are important in determining the
rheological behavior. The density of zircon is 4400 kg/m3. The yield stress can be measured by the
slump method."" In this method, the paste-like suspension is filled into a cylinder on a flat surface
and the cylinder is lifted off the suspension. The resulting slump height is measured (see Figure
Al). The yield stress is related to the slump height by

y = pg9 1 (A1)

where Ty is the yield stress, p is the suspension density, g is the gravitational acceleration (9.8 m/
s2), and H and s are indicated in Figure A1.
The students should measure the yield stress of the suspension at pH values of approximately
pH 7, pH 6, pH 5, pH 4, and pH 3. Use HCI and NaOH to adjust the pH, being careful not to
overshoot the pH and come back since this will add salt to the suspension and thus affect the
interparticle forces and thus the yield stress. Make sure the suspension is well mixed. The zeta
potentials of the powder as a function of pH can be provided to the students as shown in Figure
A2. Ask them to compare the measured yield stress values with the zeta potentials. They should
comment on the behavior in their report.

Abbreviated Laboratory Report:
Figure A3 is a photo of the slump test being performed by one of the authors. The density of the
suspension can be calculated as
Psus = OPzircon + EPH20
Psus = 0.4(4400 kg / m3) + 0.6 (1000 kg / m3) = 2360 kg / m3

The initial cylinder height (H) was 0.103 m. The slump (s) was measured with a ruler over a range
of pH values from 3 to 7. The measured slump was used to calculate the yield stress (using Eq.
Al). The yield stress is plotted against pH in Figure A4. The maximum yield stress correlates with
the isoelectric point (where the zeta potential is zero). At this pH, only van der Waals attraction is
operating between the particles creating a strong attraction and thus a high yield stress. The yield
stress decreases as the pH is moved away from the isoelectric point. This is because as the charge
on the surface of the particles increases, the electrical double layer repulsion also increases-thus
reducing the magnitude of the attraction and thus the yield stress. See Shaw,"121 Hunter,"3' or
Johnson, et al.,"4] for more details.

Figure Al. Dimensions used in
calculation of yield stress from
slump test.

Figure A2. Zeta potentials of

Figure A3. Slump test in progress.

6 300


3 4 5 6 7
pHFigure A4. Yield stresses of zircon.
Figure A4. Yield stresses of zircon.

Chemical Engineering Education


bed surface quickly are more likely to form individual nu-
clei-hence controlling the drop size controls the granule size.
Slow penetration can lead to pooling of liquid on the powder
surface, resulting in widely sized initial nuclei and widely
sized final product."011
The rate of penetration of a liquid into the pores of a pow-
der bed can be estimated by equating the capillary pressure
driving force from the Young-Laplace equation

A 2y LV COS (3)
cap r
with the viscous resistance to laminar flow predicted from
the Hagen-Poiseuille equation
APvis = 8 ull
AP (4)

to give a form of the Washburn equation

dl ryLvCose (5)
dt 41j
where u is the liquid velocity, r is the effective pore radius,
LV is the liquid-vapor surface tension, 0 is the solid-liquid
contact angle, 1 is the length of pore filled, and [t is the liq-
uid viscosity.
The effects of the parameters in Eq. (5) can be demonstrated
by asking students to measure the penetration times of drops
of water, honey, and alcohol onto a number of different pow-
der beds, e.g., coarse and fine sugar, ground pepper, and
parmesan cheese (see Figure 2).[I61 The coarse and fine sugar
demonstrate the effect of pore size r. The rate of liquid pen-
etration is approximately proportional to the particle size.
Hence, the water penetrates the fine sugar more slowly than
the coarse sugar. (Note: if an alternative powder that is in-
soluble in water is available in two different particle sizes,

Figure 2. The non-wetting behavior of drops of water (front
row) and sugar solution (middle row) compared with the
rapid wetting of alcohol (back row) on a bed of grated
parmesan cheese. Dye added to liquids to enhance visibility.

this may be preferable to using sugar.) The water and honey
demonstrate how increasing viscosity pi slows down the rate
of penetration. The water and alcohol on the cheese and pep-
per demonstrate the important effect of contact angle 0. Wa-
ter does not wet or penetrate into either of these two pow-
ders, but alcohol wets both powders because it has a lower
contact angle due to its lower surface tension, as seen by a
force balance at the contact line between the three phases
(the Young-Dupre equation)

COS VS -YLs (6)
where the subscripts V, S, and L refer to the vapor, solid, and
liquid phases, respectively.
A more comprehensive predictive model for the penetra-
tion time of a liquid drop onto a powder surface is presented
by Hapgood, et al."51 This could form the basis of a labora-
tory module for students in more advanced powder technol-
ogy subjects where they would be required to measure the
powder size and bed porosity.
Granulation Coalescence Behavior (in-class demonstration)
Wet granulation is performed by spraying a liquid binder
onto an agitated powder mass. There is great interest in being
able to predict the rate at which these granules grow as they
are agitated. This depends on how likely it is for granules to
coalesce during collisions of varying velocity.
In more advanced powder technology subjects, students
may be introduced to two of the models available for predict-
ing wet-granule coalescence. The Ennis model considers the
collision of two equi-sized elastic spheres of radius r collid-
ing head-on at a relative speed of 2u.'171 Each sphere is sur-
rounded by a layer of fluid of viscosity R. and thickness h.
The surface of each sphere has a roughness of ha, which lim-
its how close they can approach together. The spheres have a
coefficient of restitution, e, and a density, p. Solving Newton's
second law of motion, it is predicted that coalescence will
occur when the viscous Stokes number, Stv, is less than some
critical viscous Stokes number, St,*, where

St = 8p and St* =, I+ln) 4+ (7)

This model predicts that reducing the impact speed acts to
increase the likelihood of coalescence. This behavior can be
demonstrated by dropping a rubber ball from different heights
onto a flat surface coated with a layer of honey. Below a thresh-
old impact velocity (release height), the ball will not rebound.
Liu, et al.,[181 model colliding granules as elastic-plastic
spheres that are initially surface dry, but then have liquid
squeezed to the surface during collisions. This model pre-
dicts that low-velocity collisions are less likely to result in
coalescence. This is because very little permanent plastic de-
formation occurs, and hence the area of contact formed be-

Fall 2003


tween the two granules is very small and weak. This behav-
ior can be demonstrated by dropping round balls of plasticine
from different heights onto a flat surface (it helps if the
plasticine is first warmed up by vigorously hand rubbing to
make it softer). When the surface is inverted, the plasticine
dropped from a low height will drop off because it did not
deform, whereas plasticine dropped from a large height will
remain stuck on for some time due to the greater amount of
deformation forming a strong bond.
These two demonstrations serve to illustrate how coales-
cence behavior varies as consolidation changes granules from
being low-density and deformable to high-density and non-
deformable during the granulation process.

Hopper Flow (in-class activity)
A good hands-on introduction to a set of lectures on hop-
per flow is to split the class into groups of 2-4 students each
and supply each group with some thin cardboard, paper, over-
head transparencies, masking tape, scissors, a beaker of sand,
and a pan (to prevent spilling sand all over the classroom
floor). Give them 10 to 15 minutes to build a funnel (hopper)
that must discharge a set mass of sand in a set time (starting
full). Make up a hopper beforehand to check that the time
limit is reasonable-try to set a required time that is long
enough so that in trying to slow down the flow, the students
will encounter problems such as arching or rat-holing. Offer-
ing an incentive such as a large chocolate bar as a prize for
the group that gets closest to the set time adds some competi-
tive spirit and fun to the exercise.
Wander between the groups as they try different designs
and ask them what the design parameters are (hopper angle,
opening size, and possibly the wall material), how the mate-
rial is flowing (mass or funnel flow), and what problems they
are encountering (e.g., arching and rat-holing). Some groups
may resort to things such as tapping or stirring the hopper in
order to promote flow-discuss the practicalities and costs
associated with this in an industrial setting.
Funnel Flow and Mass Flow (in-class demonstration)
The two main types of flow from hoppers are mass and
funnel flow. In mass flow, the entire powder bed is in motion
as the bin discharges. The first material put into the bin is
also the first to come out. In funnel flow, material slips from
the top surface down through a rat-hole in the center of the
bin. The material at the bin walls is static-hence the first
material put into the bin is often the last out of it. Jenike and
Johanson[119 demonstrate the difference between mass and
funnel flow by using an hourglass arrangement with a differ-
ent hopper angle in each half (photograph can be found in
reference 7). If the hopper angles and material are chosen
correctly, the material will flow through the steep-angled hop-
per in mass flow and through the shallow-angled hopper in

funnel flow. Fan151 extends this idea by connecting the two
hoppers by a long straight pipe. This column then acts as a
standpipe, exhibiting regions of both moving bed transport
and suspension transport of the solids.
Consolidation Effects of Powder Flow
(in-class demonstration)
One important aspect of powder technology that should be
stressed to students is that, unlike most fluids, the behavior
of powder systems is history-dependent. The effect of con-
solidation on flow behavior can be demonstrated by using
some dish-washing powder and a funnel (the end of a plastic
drink bottle works fine). Pour the powder into the funnel and
when the exit is opened, then it will flow out easily. But if the
powder is poured into the funnel and tapped before the exit is
opened, it will have consolidated and no flow will occur when
the exit is opened.1201 Mention should be made that other factors
besides consolidation can also cause powder properties to change
with time, such as capillary condensation, re-crystallization, and
solid-state diffusion causing bonding at interparticle contacts.
Particle Dilation (in-class demonstration)
Osborne Reynolds[21] demonstrated shear-induced particle
dilation using a manometer attached to a rubber bag filled
with saturated sand. This experiment can be repeated using a
clear plastic drink bottle (the type with the straw built into
the cap so that it is easy to use while running or bike riding).
Tightly pack the main bottle cavity with saturated sand and
then top off with water until the water level reaches part-way
up the tube. Ask students what will happen when the bottle is
squeezed lightly. Counter to intuition, the water level actu-
ally drops. This is because the sand must dilate in order for
particles to slide over one another. Water flows back into the
powder bed as it dilates. This dilation behavior explains why
sand "dries up" around your foot as you walk along the beach
near the water's edge. Water is sucked away from the sur-
roundings into the dilated sand matrix. When you lift your
foot, this excess water then causes the sand to temporarily
liquefy as the load is relaxed.[eg .,22]
Wall Friction (in-class demonstration)
Powder bed behavior is different from that of a Newtonian
fluid. In a fluid, some flow is always initiated when a shear
force is applied, but powder beds offer a finite resistance to
shear forces. This ability of a powder bed to support large
loads can be demonstrated by asking students to push or pull
a plunger up a tube that is gradually filled with more and
more particles. Eventually, a stage of filling is reached be-
yond which they can no longer move the plunger-the force
they are exerting is totally transferred by the particles to the
wall of the tube. The force being exerted can be made visu-
ally evident by either including a spring balance on the pull
cable"71 or by attaching a large spring on the shaft used to
push the plunger up the column.'81 The implications of this
behavior for the distribution of stresses on hopper walls and

Chemical Engineering Education


the difficulty of achieving uniform compaction in presses and
dies can then be discussed.
A variation of this demonstration is to tape a thin sheet of
tissue paper over one end of the tube, fill the tube with the
powder, and then ask for a volunteer who thinks he or she is
strong enough to push the powder bed through the tissue paper.
Segregation During Hopper Flow (in-class demonstration)
Another counter-intuitive behavior of powders is that flow
and agitation often cause segregation, rather than mixing.
Segregation during discharge of material onto a stockpile or
into a hopper is a well-known phenomena. Many workers
have used transparent 2D hoppers to demonstrate the "her-
ring-bone" pattern formed due to a combination of percola-
tion of fine particles and the lower angle of repose of the
coarse particles (see Figure 3).ie.g.'7-8.19.23] The small particles
percolate between the larger ones and this causes the fine ones
to become concentrated in the center of the bed. During the
periodic avalanches, large particles tend to roll further down
the sloped surface of the bed because of their higher inertia and
lower angle of repose. During these avalanches, the fine par-
ticles tend to settle out along the way. This causes the large
particles to become concentrated at the base of the pile and also
gives rise to the alternate bands of fine and coarse material.
A third, and less-often demonstrated mechanism of segre-
gation is the elutriation or fluidization of ultra-fine particles
in the upflowing air displaced by the downflowing solids.
This can result in the ultra-fine particles settling out after the
other particles and forming a layer on top of the heap. With
an airtight 2-D hourglass arrangement and correct choice of
particles, elutriation and percolation segregation phenomena
can both be demonstrated simultaneously in the same appa-
ratus. Figure 3 shows the demonstration midway through the


Figure 3. Elutriation segregation of-20 micron hollow glass
spheres (fluidized bed in the upper chamber) and forma-
tion in the lower chamber of a segregated "herring-bone"
pattern of 200-400 micron beach sand (light color) and 50-
100 micron hematite/iron-ore particles (darker color) dur-
ing discharge from a sealed hopper.

Fall 2003

discharge process. The back flow of air has elutriated the ul-
tra-fines from the material flowing through the opening. The
ultra-fines have instead formed a fluidized bed in the upper
hopper. As a result, they are the last particles to flow from the
hopper and hence they deposit on the top surface of the heap
and flow down to the base at each edge.
Before performing this demonstration, the students should
be asked to predict where in the heap they think the different
size fractions of material will be preferentially deposited. Then
they can compare their predictions with the final result. Dis-
cuss the difficulties this behavior causes in obtaining repre-
sentative samples from a poured heap of granular material.
Representative samples can only be obtained by sampling at
random intervals of time the full cross-section of a powder
stream when it is in motion.
Vibrational Segregation (in-class demonstration)
The well-known "Brazil nut" effect can be easily demon-
strated by covering a steel ball bearing with sand and then
vertically tapping the container. The steel ball will rise to the
surface, in spite of its greater density. The cause of this phe-
nomena is not fully understood, but is believed to be linked
to the inertia of the object, causing it to "punch through" the
expanded bed during the upstrokes, whereas the packing of
the powder prevents it from descending during the
downstroke. [824-25]
Shinbrot and Muzzio1251 suggest a variation to this demon-
stration. If a low-density object is also added to the container,
then the behavior of the two objects varies depending on
whether the container is shaken horizontally or vertically.
Under vertical vibrations, the steel ball rises and the low-
density object sinks. Under vigorous horizontal vibrations,
the steel ball sinks and the light object rises! The cause of
this reversal is unclear, but is probably due to the bed dilating
and becoming fluidized during horizontal vibration. The class
can be asked to predict beforehand which of the two objects
will rise or sink when the jar is "shaken" (without specifying
how). Then the instructor can deliberately shake the jar in a
direction that gives a result counter to the majority class opin-
ion, in order to arouse their interest.
Fluidization (in-class demonstration)
Fluidization can be demonstrated in the classroom using a
small bed connected to a portable compressor, or if the bed is
small enough, a willing volunteer's lungs.[]l Behavior that
can be displayed includes the way the fluidized bed remains
level as the bed is tilted and the floating and sinking of ob-
jects of different density when the bed is fluidized. This can be
contrasted with the behavior of these objects in the bed when it
is vertically vibrated (see Vibrational segregation above).
Bubbling behavior can be demonstrated by filling a long
tube most of its length with a Geldart Group A powder. In-
verting the tube will result in a slug slowly rising up the length
of the column.161 Again, students could be asked beforehand

to predict what will happen when the tube of fine powder is
inverted. Many may expect the powder to move as a solid
plug from one end to the other.
Flow Improvement Due to Powder Agglomeration
(in-class demonstration)
The dramatic improvement in the flow properties of granu-
lated versus ungranulated materials can be demonstrated by
setting two hoppers side-by-side, one with the raw fine pow-
der and the other with the same powder after it has been granu-
lated. When inverted, the raw powder arches and does not
flow without tapping, whereas the granulated product flows
freely (see Figure 4). Small batches of granules for use in
this demonstration can easily be prepared at home in a do-
mestic food processor.

If you do not have the resources or time to build and per-
form these demonstrations, many of them are shown as video
clips on a CD produced by Rhodes and Zakhari.181 An ex-
panded version of this CD is due out soon that will include
interactive problems.
The Particle Technology Forum of the American Institute
of Chemical Engineers has established a website with many
good educational resources for particle technology.1261 For
ideas on how to construct and structure an introductory course
on powder technology, we suggest reading the papers by
Chase and Jacob'31 and Donnelly and Rajagopalan,[4' and also
the textbook by Rhodes1231 that was written specifically with
the purpose of being an introductory undergraduate textbook.

Instructors of powder technology courses have no excuse
for not using visual, hands-on demonstrations to introduce a
little more variety and interest to their teaching. Most of the
demonstrations mentioned in this paper can be built at little
cost using materials readily available in most engineering de-
partments. No expensive or hazardous chemicals are needed,
and most of the powders can be found at your local beach or
supermarket. Asking students to guess the powder behavior
before the demonstration is performed is an effective tool for
engaging their interest.

1. Felder, R., "How About a Quick One?" Chem. Eng. Ed., 26(1), 18
2. McKeachie, W. J., Teaching Tips, 8th ed., D.C. Heath & Co., Lexing-
ton, MA (1986)
3. Chase, G.C., and K. Jacob, "Undergraduate Teaching in Solids Pro-
cessing and Particle Technology: An Academic/Industrial Approach,"
Chem. Eng. Ed., 32, 118 (1998)
4. Donnelly, A.E., and R. Rajagopalan, "Particle Science and Technol-
ogy: Educational Initiatives at the University of Florida," Chem. Eng.
Ed., 32 122 (1998)
5. Fan, L.-S., "Particle Dynamics in Fluidization and Fluid-Particle Sys-

Figure 4. Granulated powder (left-hand side) flows easily
into lower hopper, whereas raw powder arches and does
not flow (right-hand side).

teams: Part 1. Educational Issues," Chem. Eng. Ed., 34, 40 (2000)
6. Fan, L.-S., "Particle Dynamics in Fluidization and Fluid-Particle Sys-
tems: Part 2. Teaching Examples," Chem. Eng. Ed., 34, 128 (2000)
7. Klinzing, G., "Experiments, Demonstrations, Software Packages and
Videos for Pneumatic Transport and Solid Processing Studies," Chem.
Eng. Ed., 32, 114 (1998)
8. Rhodes, M., and A. Zakhari, "Laboratory Demonstrations in Particle
Technology," CD, Monash University, Melbourne, Australia (1999)
9. Idea introduced to the authors by Professor Nafis Ahmed, formerly of
the University of Newcastle, Australia (1998)
10. Idea introduced to the authors by Professor Kevin Galvin, University
of Newcastle, Australia (1998)
11. Pashias, N., and D.V. Boger, "AFifty-Cent Rheometer for Yield Stress
Measurement," J. Rheol., 40, 1179 (1996)
12. Shaw, D.J., Introduction to Colloid and Surface Chemistry, 4th ed.,
Butterworth Heinemann (1992)
13. Hunter, R.J., Introduction to Modern Colloid Science, Oxford Science
Publications (1992)
14. Johnson, S.B., G.V. Franks, P.J. Scales, D.V. Bogher, and T.W. Healy,
"Surface-Chemistry-Rheology Relationships in Concentrated Mineral
Suspensions," Int. J. Miner. Process., 58, 267 (2000)
15. Hapgood, K.P., J.D. Litster, S.T. Biggs, and T. Howes, "Drop Penetra-
tion into Porous Powder Beds," J. Colloid & Interface Sci., 253, 353
16. Idea introduced to the authors by Dr. Bryan Ennis and Professor Jim
Litster during one of their industrial short courses on granulation (1999)
17. Ennis, B.J., G.I. Tardos, and R. Pfeffer, "A Micro-Level Based Charac-
terization of Granulation Phenomena," Powder Technol., 65,257 (1991)
18. Liu, L.X., S.M. Iveson, J.D. Litster, and B.J. Ennis, "Coalescence of
Deformable Granules in Wet Granulation Processes," AIChE J., 46,
529 (2000)
19. Jenike & Johansen, Westford, Massachusetts
20. Idea mentioned to the authors by Professor Martin Rhodes, Monash
University, Australia (2000)
21. Reynolds, 0., "Experiments Showing Dilatancy, a Property of Granu-
lar Materials, Possibly Connected with Gravitation," Proc. Roy. Inst.,
11, 354 (1886)
22. Nagel, S., "Shifting Sands," New Scientist, 53, July (2000)
23. Rhodes, M., Introduction to Particle Technology," Wiley (2000)
24. Liffman, K., D. Gutteridge, M.J. Rhodes, and G. Metcalfe, "The Bra-
zil Nut Effect," CHEMECA '98, Paper #122
25. Shinbrot, T., and F.J. Muzzio, "Non-Equilibrium Patterns in Granular
Mixing and Segregation," Physics Today, 53, 25, March (2000)
26. 0

Chemical Engineering Education

Chemical Engineering as a Career
Continued from page 273.

Perhaps the most important conclusion that can be drawn
from this survey relates to the different extent that an enjoy-
ment of chemistry at school and the role of chemistry teach-
ers influences students to study chemical engineering. It is
worth restating that across all country groupings, 90% of the
respondents admitted to being influenced to some degree be-
cause they liked chemistry at school, whereas 60% of the
respondents outside of Vietnam were not influenced at all by
their chemistry teachers. This suggests that educational and
professional institutions should work together with chemis-
try teachers to raise the profile of the chemical engineering
profession in the secondary school chemistry classroom. At
the same time, careers teachers have little influence on stu-
dents selecting chemical engineering. Further work, possi-
bly including the use of focus groups, needs to be done to
identify the reasons for this. It may be that there exists con-
siderable scope for working with careers teachers to promote
the profession.
By its very nature, this limited survey cannot gauge the
effects that localized programs and activities such as those
run by North Carolina State and Tufts have had on increas-
ing interest in the profession. Nonetheless, this survey pro-
vides the basis of an international benchmark for com-
paring factors that influence students to select our pro-
fession for their future.
This survey has identified and to some extent quantified
the important influences that acted on students currently en-
rolled in undergraduate chemical engineering degree pro-
grams. This work could be extended by surveying not only
students before they have finally selected their courses, but
also students currently enrolled in other engineering disci-
plines. The survey could also be extended to more non-En-
glish speaking countries.
The author welcomes contact from academics who might
be interested in participating in another, more comprehen-
sive survey.

The author gratefully acknowledges the cooperation of the
students who participated in the survey. Special thanks are
owed to those people who administered the surveys at the
participating institutions: Dr. Adisa Azapagic (University of
Surrey), Dr. Charun Buyakan (Prince of Songkla University),
Dr. Caroline Crosthwaite (University of Queensland), Pro-
fessor Fraser Forbes (University of Alberta), Professor Phan
Van Ha (University of Hanoi), Dr. Graham Harrison (Clemson
University), Professor Andrew Hrymak (McMaster Univer-
sity), Professor Lester Kershenbaum (Imperial College of
Science, Technology and Medicine), Professor Peter Reilly
(Iowa State University), Professor Martin Rhodes (Monash

University), Professor Jonathan Seville (University of Bir-
mingham), Professor Phan Minh Tan (Ho Chi Minh City Uni-
versity of Technology), Dr. J. Keith Walters (University of
Nottingham), Professor Laurence Weatherley (University of
Canterbury), Dr. Robin Wilcockson (University of
Loughborough). In addition, the assistance of Mr. Vo Son
Binh of the University of Melbourne is gratefully acknowl-
edged for his assistance in conducting the surveys in Hanoi
and Ho Chi Minh City, and to Mr. Duong Minh Hai of the
University of Melbourne for his translation of the survey into


1. "This is Chemical Engineering," Institution of Chemical Engineers in
Australia, Melbourne, Australia (1993)
2. "Chemical Engineering: One Profession, Many Careers," Institute of
Chemical Engineers, Rugby, UK (1997)
3. Whynotchemeng home page at
accessed April of 2003
4. "Career Choices for Chemical Engineers," at careers/> accessed April of 2003
5. Bottomley, L.J., and E.A. Parry, "Engineering Alive: A Summer Engi-
neering Camp for Middle School Students and Teachers," Proc. ASEE
Ann. Conf, Montreal, Canada (2002)
6. Shallcross, C.D., D. Novak, C. West, C.F. Duffield, and R.L. Hughes,
"Engineering! For Secondary School Science and Maths Teachers,"
Proc. Australian Assn. for Eng. Ed. 7th Conv., Melbourne, Australia,
December, 394 (1995)
7. Shallcross, D.C., J. Anderson, and D. Schaffner, "Introducing Engi-
neering into Secondary Schools: A Collaboration Between University
Academics and School Teachers," Proc. 3rd UICEE Ann. Conf. Eng.
Ed., 214 (2000)
8. Shallcross, D.C., D. Dell'Oro, D. Lamson, M. Schaffner, and J. Vincent,
Investigative Projects in Engineering: Designing a Bulk Liquid Chemi-
cal Storage Facility, Mathematical Association of Victoria, Melbourne,
Australia (1999)
9. Rushton, E., M. Cyr, B. Gravel, and L. Prouty, "Infusing Engineering
into Public Schools," Proc ASEEAnn. Conf., Montreal, Canada (2002)
10. Isaacs, B., "Mystery of the Missing Women Engineers: A Solution,"
ASCE J. Prof. Issues in Eng. Ed Pract., 127(2), 85 (2001)
11. Grandy, J., "Graduate Enrollment Decisions of Undergraduate Sci-
ence and Engineering Majors: A Survey of GRE Test Takers," GRE
Board Professional Report No. 85-01 P, ETS Research Report 92-51,
Educational Testing Service, Princeton, NJ (1992)
12. Burton, L., L. Parker, and W.K. LeBold, "U.S. Engineering Career
Trends," ASEE Prism, 5/98, 18 (1998)
13. Shallcross, D.C., "Perceptions of the Chemical Engineering Profes-
sion: Results of an International Survey," Internat. Conf. Eng. Ed.,
Manchester, UK (2002)
14. Shallcross, D.C., and G.H. Covey, "Undergraduate Chemical Engi-
neering Student Perceptions of the Pulp and Paper Industries," 6th
World Congress of Chem. Eng., Melbourne, Australia (2001)
15. Kumagai, J., "Physics Anxiety in Engineering," Physics Today (1999)
16. Shallcross, D.C., and D.G. Wood, "Combined Degree: A New Para-
digm in Engineering Education," Proc. ASEE Ann. Conf., Montreal,
Canada (2002)
17. Lewis, S., "Intervention Programs in Science and Engineering Educa-
tion: From Secondary Schools to University," in Equity in the Class-
room Towards Effective Pedagogy for Girls and Boys, P.E Murphy
and C.F. Gipps, eds., The Falmer Press, London, 192 (1996) J

Fall 2003

Random Thoughts...



North Carolina State University Raleigh, NC 27695

Thanks to some excellent research in recent decades,
we know a great deal about how learning happens
and how little of it happens in lectures.01 As fasci-
nated as professors think students should be with an hour of
material like
dA = PdV SdT -> dA = (dA/dV)TdV + (dA/dT)vdV &
dG = VdP SdT -- dG = (dG/dP)7P + (oG/dT)p dV
& dH = (dH/dS)pdS + (OH/9P)sdP -- V = (dH/dP)s = (dG/
dP)r _> S = (dA/dT)V = (dG/dT), & (dP/dT)v = (dS/dV)r

there's no mistaking the dazed stupor that falls over class-
rooms after even just a few minutes of it. Numbed minds
can't learn. The students who decide that their interests lie in
cutting that 8 a.m. class and getting more sleep may be right
on target.
You have roughly 40 contact hours in a typical course. If
all you do in them is lecture, you might as well just hand out
your notes and let the students find something more produc-
tive to do with all that time. The only way a skill is devel-
oped-skiing, cooking, writing, critical thinking, or solving
thermodynamics problems-is practice: trying something,
seeing how well or poorly it works, reflecting on how to do it
differently, then trying it again and seeing if it works better.
Why not help students develop some skills during those con-
tact hours by giving them some practice in the tasks they'll
later be asked to perform on assignments and tests?
Which is to say, why not use active learning? At several
points during the class,
1. Give the students something to do (answer a ques-
tion, sketch a flow chart or diagram or plot, out-
line a problem solution, solve all or part of a prob-
lem, carry out all or part of a formula derivation,
predict a system response, interpret an observa-
tion or an experimental result, critique a design,

troubleshoot, brainstorm, come up with a ques-
2. Tell them to work individually, in pairs, or in groups
of three or four; tell them how long they'll have
(anywhere from 10 seconds to two minutes); and
turn them loose.
3. Stop them after the allotted time, call on a few in-
dividuals for responses, ask for additional volun-
teered responses, provide your own response if
necessary, and continue teaching.
You may also occasionally do a think-pair-share, in which
the students work on something individually and then pair
up to compare and improve their responses before you call
on them.
As little as five minutes of that sort of thing in a 50-minute
class session can produce a major boost in learning. For start-
ers, it wakes students up: we have seen some of them elbow-
ing their sleeping neighbors when an active learning task was
assigned. Academically weak students get the benefit of be-
ing tutored by stronger classmates, and stronger students get
the deep understanding that comes from teaching something
to someone else. Students who successfully complete a task
own the knowledge in a way they never would from just
watching a lecturer do it. Students who are not successful are

Richard M. Felder is Hoechst Celanese Professor Emeritus of Chemical
Engineering at North Carolina State University. He received his BChE
from City College of CUNY and his PhD from Princeton. He is coauthor of
the text Elementary Principles of Chemical Processes (Wiley, 2000) and
codirector of the ASEE National Effective Teaching Institute
Rebecca Brent is an education consultant specializing in faculty devel-
opment for effective university teaching, classroom and computer-based
simulations in teacher education, and K- 12 staff development in language
arts and classroom management. She co-directs the SUCCEED Coali-
tion faculty development program and has published articles on a variety
of topics including writing in undergraduate courses, cooperative learn-
ing, public school reform, and effective university teaching.

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

put on notice that they don't know something they may need
to know, so when the answer is provided shortly afterwards
they are likely to pay attention in a way they never do in
traditional lectures.
The number of possible active learning tasks is limitless.121
At a minimum, you can ask the same questions you would
normally ask in your lectures, only now you'll get the whole
class trying to answer them and not just the same two stu-
dents who always answer them. You can also use any of the
activities suggested in Item 1 of the list several paragraphs
back, and you might occasionally run a TAPPS ("thinking-
aloud pair problem solving") exercise, arguably the most
powerful classroom instructional technique for promoting
understanding.E3] Have the students work in pairs through a
complex derivation or worked-out problem solution in the
text or on a handout, with one of them explaining the solu-
tion step-by-step and the other questioning anything unclear
and giving hints when necessary. Periodically stop them, call
on several of them for explanations, provide your own when
necessary, and have the students reverse roles in their pairs
and proceed from a common starting point. It may take most
or all of a class period to work through the entire solution,
but the students will end with a depth of understanding they
would be unlikely to get any other way.
Here are several techniques to make active learning as ef-
fective as possible.
At the beginning of the course, announce that
you'll be assigning short exercises during class and
explain why you're doing it (research shows
students learn by doing, and the exercises will give
them a head start on the homework and tests). The
explanation can help defuse the resistance some
students feel toward any teaching approach other
than the instructor telling them just what they need
to know for the exam.
After an active learning exercise, call on a few
individuals for responses before opening the floor to
volunteers. The knowledge that you might call on
them gets active participation from students who
would normally just sit passively and let others do
the work.
Go for variety. Vary the type of activity (answer-
ing questions, solving problems, brainstorming,
etc.), the activity duration (10 seconds-2 minutes),
the interval between activities (1-15 minutes), and
the size of the groups (1-4 students). Mixing things
up keeps active learning from becoming as stale as
straight lecturing.
As many as half of the participants in our recent teaching

workshops report using active learning in their classes, but
nonusers often have concerns about the approach. (1) If I use
active learning, will I still be able to cover my syllabus? (2)
Can I do it in a really large class? (3) What should I do if
some of my students refuse to participate?
We have offered detailed answers to the first two questions
in another column14 and so will just give the short versions
here. (1) Yes. (See Reference 4 for details on how.) (2) Yes,
and in fact, the larger the class, the more important it is to use
active learning. Try finding another way to get students ac-
tively engaged when there are 150 of them in the room.
What about students who refuse to participate? There may
indeed be several who just sit staring straight ahead when
groupwork is assigned, even after the awkwardness of the
first few times has passed. We never see more than two or
three of them in our classes, but for the sake of discussion
let's say it's as many as 10% in yours. That means that while
you're doing an active learning exercise, 90% of the students
are actively engaged with the material and getting practice in
the skills you're trying to teach them, and 10% are out to
lunch. On the other hand, at any given moment in a tradi-
tional lecture, if as many as 10% of your students are ac-
tively involved with the lecture material you're doing very
well. No instructional technique works for all students at all
times: the best you can do is reach as many as possible, and
90% is more than 10%. If some students opt out, don't let it
bother you-it's their loss, not yours.
In short, if you start using active learning in your classes,
you can expect to see some initial hesitation among the stu-
dents followed by a rapidly increasing comfort level, much
higher levels of energy and participation, and above all, greater
learning. Check it out.

1. (a) McKeachie, W.J., P.R. Pintrich, Y-G Lin, D.A. Smith, and R.
Sharma, Teaching and Learning in the College Classroom: A Review
of the Research Literature, 2nd ed., University of Michigan, Ann Ar-
bor, MI (1990); (b) Bransford,J.D., A.L. Brown, and R.R. Cocking,
Eds. How People Learn: Brain, Mind, Experience, and School, Na-
tional Academy Press, Washington, DC (2000)
2. (a) Johnson, D.W., R.T. Johnson, and K.A. Smith, Active Learning:
Cooperation in the College Classroom, 2nd edn., Interaction Book Com-
pany, Edina, MN, (1998); (b) Felder, R.M., and R. Brent, "Coopera-
tive Learning in Technical Courses: Procedures, Pitfalls, and Payoffs,"
ERIC Document Reproduction Service, ED 377038 (1994),

3. Lochhead, J., and A. Whimbey, "Teaching Analytical Reasoning
through Thinking Aloud Pair Problem Solving," in J.E. Stice (Ed.),
Developing Critical Thinking and Problem-Solving Abilities, New
Directions for Teaching and Learning, No. 30, Jossey-Bass, San Fran-
cisco, CA (1987)
4. Felder, R.M., and R. Brent, FAQs-II. Chem. Engr Ed., 33(4), 276-277
(1999) D

Fall 2003

All of the Random Thoughts columns are now available on the World Wide Web at and at




A New Path to Glory

This is the 2003 ConocoPhillips Lecture, presented at Oklahoma
State University, Stillwater, Oklahoma, on April 25, 2003.

University of Notre Dame Notre Dame, IN 46556
he chemical engineering profession is in the midst of
great change. Chemical engineers used to focus on
making large quantities of small, relatively simple
molecules (commodity products). With increasing frequency,
in the future they will have to make smaller quantities of more
complex, possibly biologically active, molecules and
nanostructured materials (specialty products). Further, we
used to only scale things up; now we must also scale down,
as in lab-on-a-chip devices and portable fuel cells. In addi-
tion, developments in science and other engineering disci-
plines-such as nanoscale synthesis and characterization tech-
niques, molecular biology and information technology-in-
fluence progress in our field. There is also a continuing need
to consider what will be the energy sources for the future-
conventional fuels such as oil, gas, and coal, or others such
as nuclear, biomass, and solar? Finally, growing environmen-
tal considerations in society make us aware of the long-term
and global implications of our manufacturing practices.
I would like to discuss how all these factors, currently at
play, will impact the education of chemical engineers pri-
marily at the undergraduate level, although some remarks
will also be made toward graduate education and research

Before turning toward the future, it is instructive to first
examine how the discipline of chemical engineering evolved.
Fascinating detailed accounts of early developments in the
curriculum and the profession have been presented in many
sources,1-61 so I will keep this discussion brief.
It is generally agreed that chemical engineering as a dis-
tinct discipline began in January of 1888 when George E.
Davis gave a series of twelve lectures on the subject at the

Manchester Technical School in England. He had previously
coined the term "chemical engineer" in 1880 and promoted
it (unsuccessfully) to found a society of chemical engineers.
The first four-year undergraduate chemical engineering de-
gree program was established at MIT by the chemistry pro-
fessor Lewis Mills Norton in 1888. It was soon followed by
those at the University of Pennsylvania (1892), Tulane (1894),
Michigan (1898), and others, including our own at Notre
Dame (1909). Most early curricula had their origin in chem-
istry departments, although there are examples of some evolv-
ing from mechanical (e.g., Colorado, 1904) and electrical (e.g.,
Wisconsin, 1905) engineering departments as well.
The early chemical engineering curricula included an amal-
gam of courses taken by chemists and mechanical engineers,
with those in industrial and applied chemistry in the third
and fourth years being unique to the field. The discipline re-
ceived its first unifying theme with development of the con-
cept of "unit operations," which is often called the first para-
digm of chemical engineering. It grew out of the realization
that purely physical operations of chemical processing,
whether to produce smaller quantities of fine or larger amounts
of heavy chemicals, all depended on certain common prin-
ciples of physics and chemistry. As first noted by Arthur D.
Little (1915) in the Chemical Engineering Visiting Commit-

Arvind Varma is the Arthur J. Schmitt Profes-
sor of Chemical and Biomolecular Engineer-
ing, and Director of the Center for Molecularly
Engineered Materials at the University of Notre
Dame. Author or coauthor of more than 230
archival journal research articles and three
books, he has received a variety of recogni-
tion for his teaching and research, including
the Wilhelm Award of AlChE (1993) and the
Chemical Engineering Lectureship Award of
ASEE (2000).

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

tee report to the President and Corporation of MIT
Any chemical process, on whatever scale conducted, may
be resolved into a coordinated series of what may be termed
'unit actions,' as pulverizing, mixing, heating, roasting, ab-
sorbing, condensing.... The number of these basic unit op-
erations is not very large and relatively few of them are in-
volved in any particular process.
The first significant textbook for the discipline, Principles of
Chemical Engineering by Walker, Lewis, and McAdams of
MIT, appeared in 1923.'71 It showed that by combining a few

The discipline received its first unifying
theme with development of the concept
of "unit operations," which is often
called the first paradigm of
chemical engineering...

chemical industries followed, and in turn catalyzed, devel-
opments in chemical engineering.
The 1950s also saw a greater emphasis on the use of analy-
sis and applied mathematics in solving chemical engineering
problems that can be traced to three separate events.t1"I First,
it was recognized that the individual unit operations involve
a combination of the same basic principles in microscopic
momentum, heat, and mass transport, each with similar math-
ematical descriptions. Thus, a study of the individual trans-
port processes as a unified subject "Transport Phenomena"
can lead to a greater understanding of chemical processes;
this concept was greatly aided by the appearance of a fa-
mous book in 1960 with that same title, written by Bird,
Stewart, and Lightfoot.""I
Second, applications of sophisticated mathematical tech-
niques were yielding strong results for the design and opera-

principles of momentum, mass, and heat transfer, it is pos-
sible to understand the unit operations. It was an extremely
influential textbook that charted the education, development,
and practice of chemical engineering for decades.
Soon after the introduction of unit operations, attention
turned to developing procedures for overall material and en-
ergy balances in processes, including single or multiple reac-
tions, recycle, and bypass, and curricula in the 1930s included
courses in industrial chemical calculations. In the 1940s,
courses in thermodynamics were introduced that included
properties of gases and liquids and applications of both the
first and second laws. This decade also saw the development
of courses in equipment and process design.
Although there were important efforts in German, notably
by Damkohler,1sI the systematic development of chemical and
catalytic reaction engineering principles in the English lan-
guage, using information on reaction rates and catalysis,
waited until the appearance in 1947 of Chemical Process
Principles: Part III. Kinetics and Catalysis by Hougen and
Watson.[9' By the end of the 1950s, most chemical engineer-
ing undergraduates took formal courses in reaction engineer-
ing and courses in process control were initiated.
All through this period, since the early days of the profes-
sion, a synergistic relationship existed between academia and
industry. Much of university research was supported by in-
dustry, and the graduates were readily employed by the grow-
ing petroleum refining, petrochemical, and chemical indus-
tries. The oil companies refined petroleum crude to produce
gasoline and other fuels for automobiles and airplanes, the
petrochemical complexes produced bulk chemicals, while the
chemical companies produced these as well as polymers, fer-
tilizers, paints, and other specialty chemicals. All these prod-
ucts satisfied a rising demand from a society that had an in-
creasing standard of living. The growth of the petroleum and

. the 1950s and 60s saw the emergence of
the so-called engineering science approach
in the discipline-the second paradigm
in chemical engineering.

tion of separation processes and chemical reactors, as exem-
plified in the works ofAmundson.1 21 Finally, the general avail-
ability of computers, whereby it became possible to conduct
numerical simulations of process models to identify optimal
design and operating conditions, also accelerated the appli-
cation of analytical and numerical techniques.
Thus, the 1950s and 60s saw the emergence of the so-called
engineering science approach in the discipline-the second
paradigm in chemical engineering. This approach led to a
curriculum at both the undergraduate and graduate levels that
is a unique blend of chemistry, physics, and mathematics.
The chemical engineers educated in this manner could effec-
tively develop, design, and operate complex chemical pro-
cesses that typically produced commodity products.

I will now examine the current status of the discipline as it
relates to the education and employment of chemical engi-
neers. I will be brief since the topic has been addressed well
in a lecture by Ed Cussler last yearE3' and elsewhere.[141 Fur-
ther, there is much discussion of these issues in the context of
undergraduate curriculum revitalization as a result of the
"New Frontiers in ChE Education" workshops organized
through the Council for Chemical Research (CCR) and spon-
sored by the National Science Foundation (NSF).1"1

A typical undergraduate chemical engineering curriculum
consists of foundation courses in mathematics, physics, chem-

Fall 2003

istry, and engineering in the early years, as well as courses in
humanities and social sciences that serve to provide a broad
education. The chemical engineering courses typically include
offerings in mass and energy balances, thermodynamics,
transport processes, separations, reaction engineering, pro-
cess design, process control, and laboratories where principles
learned in the lecture courses are reinforced and include ele-
ments of both written and oral communication of experimen-
tal results and analysis. Finally, there are generally several
electives to choose from, in chemical engineering as well as
in other science and engineering disciplines.
A striking fact is that while the discipline of chemical engi-
neering has evolved significantly over the last forty or so years
(as I shall detail next), the undergraduate curriculum has re-
mained essentially unchanged. The engineering science para-
digm continues to dominate the core curriculum as well as
the textbooks that are used. The examples used in courses
continue to come primarily from the petroleum refining and
bulk chemicals production industries.


Between 50% and 60% of the BS degree chemical engi-
neering graduates in the U.S. seek industrial employment im-
mediately upon graduation-Figure 1 shows their distribu-
tion during the 2000-01 year by nature of the industry. It is
remarkable that the skills learned from understanding engi-
neering principles and processes, based largely on physical
and chemical transformations, are considered to be valuable
by a large number of industries. If we consider chemical and
energy companies as the traditional employers, however, then
only about 40% of chemical engineers find their initial em-
ployment there. About an equal number go to the electronic,
food/consumer products, biotechnology, and materials-related

Figure 1.
Initial industrial Engrg. Svcs.- Environmental 2.4
employment of Engrg. Svcs.- Research
BS chemical & Testing 1.8%
engineers, Engrg. Svcs. Design
y& Cnstrctn. 5.6%
2000-01 year.
(ArChE, Pulp & Paper 2.1%
of Career
Services) Biotech./Related
Services) -Inh.+i- Dharmn \ l /

industries, which were not significant employers some years
ago. Overall, as noted elsewhere,1141 only about 25% of the
graduates are hired by companies that manufacture com-
modity chemicals emphasized in the curriculum, while
about 50% go to those with a product orientation-in
contrast with approximately 80% and 15%, respectively,
twenty-five years ago.
In addition to the increasingly wider spectrum of indus-
tries where chemical engineers now find employment, sev-
eral other factors currently at play, even with the traditional
energy and chemical companies, are

The companies are becoming more global, with a
greater fraction of their manufacturing and research
conducted overseas
*Many companies are merging into larger ones, with
significant reductions in the workforce
Chemical companies are increasingly incorporating
life sciences into their manufacturing and products
Chemical engineers cannot expect to work with a
single company or industry type and must now accept
several job changes over their professional careers

Other Driving forces

There are also other driving forces currently operative, and
I would like to enumerate some of them, without claiming
completeness. First, biology is rapidly developing as a mo-
lecular-based science so that its connections can now be made
more readily to chemical engineering. There are numerous
opportunities for coupling molecular-level understanding of
biological reactions and interactions with chemical engineer-
ing concepts and processes, that can result in products of tre-

Chemical Engineering Education

mendous value. Some examples are bioprocessing for pro-
duction of pharmaceuticals and even commodity chemicals,
metabolic engineering, controlled drug delivery, biomaterials,
tissue engineering, functional genomics, gene therapy, drug
design and discovery, nano and micro biotechnology for lab-
on-a-chip devices, etc.
Second, there is a current trend toward establishment of
bioengineering and biomedical engineering departments,
driven by the Whitaker Foundation grants (see, for example,
Ref. 16). Owing to the closest fit, these new programs com-
pete for students and resources that in many instances would
otherwise come to chemical engineering.
Third, there is growing awareness of the pressures that cur-
rent manufacturing practices place on the environment in
terms of pollutants that require remediation and waste gen-
eration that demands disposal and diminishes resource uti-
lization. Thus, environmentally benign processing and sus-
tainable development is receiving increased attention, both
to satisfy environmental regulations and to increase prof-
Fourth, new educational tools and methods are being de-
veloped that can be used to enhance the quality of chemical
engineering education. Examples include use of web-based
educational materials that can be shared across institutions,
web interfaces to run actual laboratory experiments, and
simulations to explore influence of parameters or to learn
about cases too dangerous to conduct in the laboratory,
such as explosions.

Based on the current status of the discipline, some sugges-
tions can now be made for future directions of chemical en-
gineering education, especially at the undergraduate level.
The intent is to provide a framework that takes advantage of
both the unique aspects of the present curriculum and the
changing scene related to employment and developments in
other science and engineering disciplines. My basic premise
is that the defining characteristics of chemical engineers, i.e.,
the ability to apply molecular level understanding to convert
raw materials into more valuable products by physical, chemi-
cal and biological transformation using economic and safe
processes, should remain unaltered. Thus, the core subjects
in the curriculum involving mass and energy balances, ther-
modynamics, transport processes, reaction engineering, sepa-
rations, laboratories, and design should continue in the fu-
ture, but with structural modifications as discussed below.

Expanded Examples of Applications

As noted earlier, chemical engineers now find employment
in a wide variety of industries. It is apparent that their skill
set, which includes chemistry in addition to physics and math-

.. molecular engineering of products
and processes is emerging as
a new paradigm for
the discipline.

ematics also available to other engineering disciplines, makes
them uniquely qualified to impact a diverse set of technolo-
gies. The curriculum, however, continues to include examples
primarily from the petroleum refining, petrochemical, and
bulk chemicals industries. It is important to broaden the scope
by including examples from areas such as materials process-
ing, biotechnology, pharmaceuticals, food processing, and en-
vironment. Similarly, when discussing design, considerations
of product and not merely process should be included. These
movements will require new textbooks and teaching mod-
ules. Some steps in this direction are already being taken.

Modern Biology as an
Underlying Fundamental Science

Recent developments in molecular and cellular biology, the
similarity of using biological and chemical reactions at the
molecular level for design of new products and processes,
and the growth of biotechnology industries where chemical
engineers are currently employed (and from all indications
will be in greater numbers in the future) all suggest that biol-
ogy will soon reach an almost equal status with chemistry as
a basic science in defining chemical engineering. Thus, it is
now timely to include one or two formal courses in biology
and biochemistry in the early years of the undergraduate cur-
riculum. This requires two types of actions: one, working to-
gether with the relevant disciplines to arrive at suitable
courses, and two, incorporating elements of biology within
all the chemical engineering courses just as chemistry is to-
day. Thus, for example, in the reaction engineering course,
building upon knowledge of biochemistry and biology gained
earlier, connections between molecular mechanisms and mac-
roscopic kinetics could be made and related to modeling of
cells and bioreactors, similar to what is done today with chemi-
cal catalysis and diffusion-reaction in catalyst pellets leading
to fixed-bed reactor design. Similarly, based on biological
understanding, separations courses can readily include liv-
ing systems and processing of biomolecules.
Numerous opportunities also exist in other core courses,
such as mass and energy balances, thermodynamics, trans-
port processes, and design. These developments are likely to
take some time to materialize, but movement in this direc-
tion is critical for chemical engineers to contribute effectively
and exercise leadership in the biotechnology areas that offer
tremendous potential for growth.

Fall 2003

Recruitment of Talented and Motivated Students

People are our greatest asset, so for the vitality and future
of the discipline we must attract the best and the brightest to
chemical engineering. This will occur naturally if we offer
imaginative courses and programs involving new technolo-
gies, use newer methods and tools in our teaching, and pro-
vide intellectual challenges for our stu-
dents, so that they have promise of a
bright future while solving important
problems facing society.
A specific method I have found to
be effective in challenging students
intellectually is to involve them in un-
dergraduate research. The opportunity
to do an independent project with only
general overall guidelines provided,
often using equipment assembled on
their own, is stimulating for most stu-
dents. Over the last ten years, when I
began to keep a record of this activity,
25 undergraduates have conducted re-
search in my laboratory, many starting
in their junior year, and some 15 have
gone on to attend graduate school else-
where, most for PhD degrees. (In a
lighthearted vein, I sometimes say that
I have saved a large number of bril-
liant chemical engineers from leaving
our profession for careers such as in
medicine or law-of course, I do not
say this in front of my daughters, one a
lawyer and the other studying to be-
come one!) Many work closely with a
graduate student or a postdoctoral as-
sociate, to mutual benefit, and I have a
number of journal papers with undergraduates as coauthors.
Undergraduate research exposes students to the frontiers of
the field and provides the intellectual challenges that are dif-
ficult to match in typical lecture or laboratory courses.

Name Change of Departments

As noted earlier, the chemical engineering profession is
changing rapidly and faces many new challenges. The most
impressive movement appears to be the emergence of mod-
em biology as a fundamental science, on an almost equal
footing with chemistry, in defining the field. Further, all indi-
cations are that its role will continue to grow in the future.
For this and other pragmatic reasons, including the facts that
students are attracted to biological departments and degree
names and that we face new competition for students and
resources from new bioengineering and biomedical engineer-
ing departments (some 90 such departments already existed

at the end of 2001Ei5]), many chemical engineering depart-
ments are changing their names to include some biological
term. Among several that are possible, chemical and
biomolecular engineering seems to be gaining acceptance,
as adopted recently by departments at Cornell, Illinois, and
ours at Notre Dame. It connects with the scientific base of
the discipline, is more inclusive of modern biotechnology as
compared with alternatives, and owing
to its molecular focus, it offers more
potential for collaborations with bio-
chemists and biologists. Thus, while
William Shakespeare's Juliet asks,
"What's in a name? That which we call
a rose, by any other name would smell
as sweet," for the reasons cited, I fa-
vor departmental name changes.

Graduate Education
and Research
Although I have limited my remarks
so far to undergraduate education, I
would like to say a few words about
graduate education and research.
Graduate education also started in the
early 1900s, at both the MS and PhD
levels. The core graduate curriculum
has essentially mirrored the curriculum
at the undergraduate level, with the
former always being more fundamen-
tal and mathematical in content. Thus,
courses in thermodynamics, kinetics
and reaction engineering, transport pro-
cesses, and mathematical analysis,
based on the engineering science ap-
proach, are currently required in most
graduate programs. They are augmented by other courses in
chemical engineering, various sciences, and other engineer-
ing disciplines, to suit the student's research needs and inter-
ests. Similar to the undergraduate courses, the graduate
courses also need to include examples in newer application
areas and incorporation of biology, particularly as it is intro-
duced in the earlier years.
In research, chemical engineering graduate programs have
moved forward rapidly to embrace all areas of new technolo-
gies, including biological, materials, environmental, infor-
mation, and energy. This movement was promoted by the
National Research Council's "Frontiers in Chemical Engi-
neering" report published in 1988,E1'1 whose recommenda-
tions were recently reinforced and updated.[sJ Further, there
is a growing trend toward interdisciplinary research involv-
ing faculty members and students from different fields work-
ing together to solve research problems. This trend has its
origin in at least two related facts: one, the cutting-edge prob-

Chemical Engineering Education

By offering imaginative courses that use new teaching methods and tools, and by providing
intellectual challenges, we will be able to attract the best and brightest to chemical engi-
neering and educate them to become leaders in industry, academia, and society.

lems are often at the interface between disciplines, and two,
funding agencies (now primarily federal and state, as com-
pared to mainly industrial prior to the 1950s) seem to favor
this approach. In turn, universities have responded by estab-
lishing research centers, typically involving colleges of sci-
ence and engineering but sometimes also business or public
policy, that facilitate interdisciplinary interactions. While the
coexistence of traditional departments and centers can lead
to tension, I believe that organization along these lines is re-
quired and that this structure is here to stay for some time.
Finally, there is another movement currently occurring in
the chemical engineering discipline, particularly at the gradu-
ate education and research levels. On one hand, in addition
to a molecular-level description of chemical and biological
transformations and processes, there is growing feasibility
now to also conduct molecular-scale simulations to compute
thermodynamic, transport, and other properties of fluids and
materials. On the other hand, owing to the strengths of analy-
sis inherent in the engineering science approach, along with
a systems view, it is possible to analyze complex systems
and their interactions. These directions are changing the na-
ture of chemical engineering such that it could be claimed
that molecular engineering of products and processes is
emerging as a new paradigm for the discipline. This move-
ment will take some time to significantly influence the edu-
cation of chemical engineers at the undergraduate level, and
there is current discussion ongoing in this regard."5'

Chemical engineering as a distinct discipline started with
applications primarily in petroleum refining and bulk chemi-
cals production industries, but skills developed as a result of
a solid foundation in the fundamental sciences (chemistry,
physics, mathematics, and now increasingly, biology), along
with a quantitative engineering science approach, have per-
mitted chemical engineers to move rapidly into many of the
emerging technologies. Their impact in the newer areas will
be enhanced by continuing the core curriculum and augment-
ing it by expanding examples of applications, incorporating
biology in all core courses, and including orientation toward
both product and process design. By offering imaginative
courses that use new teaching methods and tools, and by pro-
viding intellectual challenges, we need to attract the best and
brightest to chemical engineering and educate them to be-
come leaders in industry, academia, and society.
I hope that these remarks, along with the current discus-
sion ongoing in the NSF/CCR workshops,1I51 will lead to in-

Fall 2003

novative chemical engineering programs that involve new
technologies and provide a bright future for our students while
solving important problems facing society.

I have benefited much from discussions with my colleague
Mark McCready. Bob Armstrong and Barry Johnston of MIT
helped immensely by providing information about the NSF/
CCR workshops. The Edison Lectures of Bob Brown, also
of MIT, provided a valuable perspective in the evolution of
chemical engineering. Finally, by collecting data from sources
and designing slides, Chris Norfolk and Alexander Mukasyan
helped prepare this lecture.

1. Aris, R., "Academic Chemical Engineering in an Historical Perspec-
tive," I&EC Funds., 16,1 (1977)
2. Hougen, O.A., "Seven Decades of Chemical Engineering," Chem. Eng.
Prog., 73(1), 89 (1977)
3. Pigford, R.L., "Chemical Technology: The Past 100 Years," C&ENews,
54(15), 190 (1976)
4. Furter, W.F., Editor, "History of Chemical Engineering," Adv. in Chem.
Series, 190 (1980)
5. Furter, W.F., Editor, A Century of Chemical Engineering, Plenum Press,
New York, NY (1982)
6. Scriven, L.E., "On the Emergence and Evolutiion of Chemical Engi-
neering," Adv. in Chem. Eng., 16, 3 (1991)
7. Walker, W.H., W.K. Lewis, and W.H. McAdams, Principles of Chemi-
cal Engineering, McGraw-Hill, New York, NY (1923)
8. Damk6hler, G., "Einfluss von Diffusion, Str6mung und Warmetransport
auf die Ausbeute bei chemisch-technischen Reaktionen," Der Chemie-
ingenieur, A. Euken and M. Jakob, eds., 3, 359 (1937)
9. Hougen, O.A., and K.M. Watson, Chemical Process Principles: Part
III. Kinetics and Catalysis, John Wiley, New York, NY (1947)
10. Varma, A., "Some Historical Notes on the Use of Mathematics in Chemi-
cal Engineering," pages 353-387 in A Century of Chemical Engineer-
ing, W.F. Furter, ed., Plenum Press, New York, NY (1982)
11. Bird, R.B., W.E. Stewart, and E.N. Lightfoot, Transport Phenomena,
John Wiley & Sons, New York, NY (1960)
12. Amundson, N.R., The Mathematical Understanding ofChemical Engi-
neering Systems: Selected Papers ofNeal R. Amundson, R. Aris and A.
Varma, eds, Pergamon Press, Oxford (1980)
13. Cussler, E.L., "What Happens to Chemical Engineering Education?"
Phillips Lecture, Oklahoma State University (2002)
14. Cussler, E.L., D.W. Savage, A.P.J. Middelberg, and M. Kind, "Refocus-
ing Chemical Engineering," Chem. Eng. Prog., 98(1), 26S (2002)
15. New Frontiers in Chemical Engineering Education, a series of work-
shops on the Chemical Engineering Undergraduate Curriculum; docu-
ments available following the link at
16. Katona, P.G., "The Whitaker Foundation: The End Will be Just the Be-
ginning," IEEE Trans. Med. Imaging, 21, 845 (2002)
17. Frontiers in Chemical Engineering: Research Needs and Opportuni-
ties, National Academy Press, Washington, DC (1988)
18. Beyond the Molecular Frontier: Challenges for Chemistry and Chemi-
cal Engineering, The National Academies Press, Washington, DC (2003)





Purdue University West Lafayette, IN 47907-2100

When studying chemical reactions within a single
phase, chemical engineers require knowledge of
the equilibrium constants. For a given tempera-
ture and pressure, equilibrium compositions may then be cal-
culated for all relevant reactions. If the temperature, pres-
sure, or composition of one of the components changes, how-
ever, the equilibrium position usually shifts. The direction of
such shifts can be calculated by direct computation of the
new equilibrium state.
Observations of the direction of shifts in the equilibrium
position led to the formulation of a general statement referred
to as the "Principle of Le Chatelier,"E'1 or sometimes as the
"Principle of Le Chatelier and Braun."'21 Le Chatelier's prin-
ciple can be stated as follows:'1i
In a system at equilibrium, a change in one of the
variables that determines the equilibrium will shift the
equilibrium in the direction counteracting the change
in that variable.
The above statement is useful in inferring, without direct
calculation, the effects of changes in a system initially at equi-
librium. Yet, still not widely known, particularly in the chemi-
cal engineering literature, is that Le Chatelier's principle is
not universally valid, and exceptions are known to occur. (See,
however, Sandler[2] and Tester and Modell131 as examples of
current chemical engineering textbooks that highlight the limi-
tations of the above statements. Exceptions to Le Chatelier's
principle appear to be more widely known in the physical
chemistry literature and have been discussed for some time.
See, for example, de Heer141 and Liu, et al.,E15 for an historical
account of Le Chatelier's principle.)
Consider, for example, the ammonia synthesis reaction
N2 +3H2 <.- 2NH3
in which equilibrium has been established at a given tem-

perature, T, and pressure, P. Le Chatelier's principle predicts
that the reaction will shift to the right (i.e., more ammonia
will be produced) upon the addition of more nitrogen to the
reaction vessel. If the initial mole fraction of nitrogen ex-
ceeds 0.5 and the given T and P are held fixed, however, the
reaction instead proceeds to the left, producing more nitro-
gen, as predicted from rigorous equilibrium constant calcu-
lations (the value of 0.5, as shown later, is calculated assum-
ing ideal gas behavior). This shift to the left is a clear excep-
tion to the principle of Le Chatelier, which has not been rig-
orously proven[41
Proofs of this unexpected shift have been given before.14-61
Most chemical engineering texts do not provide a proof, ex-
cept, for example, Tester and Modell,i31 which does provide a
detailed proof. The most widely referenced and reproduced
proof is by Katz[16 (the procedure followed in Liu, et al.,15t is
nearly the same as the approach by Katz, although the au-

David S. Cortl is Assistant Professor of Chemi-
cal Engineering at Purdue University. His re-
search interests include molecular thermody-
namics of liquids (both stable and metastable),
glasses, and complex fluids, droplet conden-
sation and bubble nucleation, and the devel-
opment of molecular simulation algorithms. He
teaches courses on Thermodynamics and Sta-
tistical Mechanics.

Elias I. Franses is Professor of Chemical Engi-
neering at Purdue University. His research in-
terests include adsorption equilibria and dynam-
ics of surfactants and proteins at air/water inter-
faces, with applications to lung surfactants, and
the surface chemistry andphysics ofadsorbents
at liquid/solid interfaces, for bioseparations. He
teaches courses on Colloidal and Interfacial
Phenomena, Thermodynamics, and Chemical
Reaction Engineering.

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

To address the technical and educational issues of Le Chatelier's principle,
we therefore present in this paper a new and conceptually more straightforward
analysis of the direction of the equilibrium shift for the
ammonia synthesis reaction as an example.

thors were apparently unaware of Katz). This proof makes
use of a "reaction quotient" that has the same functional form
as the ratio of mole fractions at equilibrium and is applicable
whether or not equilibrium has been established. The value
of this reaction quotient, defined in Eq. (3), varies if a change
occurs, but must equal the equilibrium constant when the sys-
tem returns to an equilibrium state. The direction that the re-
action quotient takes to restore itself to the equilibrium value
determines the direction of reaction for the given change.
The use of a reaction quotient can be confusing to students,
particularly to students exposed to reaction equilibria for the
first time. To address the technical and educational issues of
Le Chatelier's principle, we therefore present in this paper a
new and conceptually more straightforward analysis of the
direction of the equilibrium shift for the ammonia synthesis
reaction as an example. Our approach is, however, more gen-
eral. In contrast to the other methods, changes at constant T
and P are now considered in which the value of the reaction
quotient is strictly held fixed and equal to the equilibrium
constant. Hence, the analysis makes no use of a separately
defined reaction quotient (that is applicable whether equilib-
rium is or is not established) and should be easier for stu-
dents to understand. The analysis also involves finite, as
well as infinitesimal, changes, which can be the basis of
future experimental tests that may demonstrate more viv-
idly the key thermodynamic laws (see del Pino, et al.,171
for an example of a simple experiment concerning shifts
of chemical equilibrium).
Le Chatelier's principle can be reformulated in a more gen-
eral way that becomes universally valid,14,51 although it bears
little resemblance to the statement given earlier. For peda-
gogical reasons, we briefly discuss this new general state-
ment in the last section of this paper. An excellent overview,
and proof, of this new general statement is given by de Heer.141
It is, however, only valid for infinitesimal changes from the
initial equilibrium state.151 In this paper, we also consider the
ammonia synthesis reaction for the case of adding nitrogen
in finite amounts (Liu, et al.,s15 considered finite additions as
well, but the present analysis provides a more straightfor-
ward and quantitative discussion). The value of 0.5 for the
mole fraction of nitrogen, above which the reaction proceeds
to the left while below the reaction proceeds to the right, is
shown to be true for infinitesimal additions of nitrogen.
For finite changes, no universally valid statement on the
direction in which the reaction shifts can be formulated, and
thus each case must be considered individually. In such cases,

instructors should advise ignoring the reformulated Le
Chatelier's principle and instead should calculate, in general,
the shift in the equilibrium state directly from the relations of
chemical equilibrium.

Exception to the Principle of Le Chatelier
Let us consider the ammonia synthesis reaction and assume
for simplicity that the components comprise an ideal-gas mix-
ture. Analyses for nonideal mixtures, although possible, have
not been reported. Let species 1 represent nitrogen, species 2
hydrogen, and species 3 ammonia. The chemical potential of
each species i, i, in the ideal-gas mixture is given by181

i = Fi(T)+RTfinyP (1)
where F,(T) is the chemical potential of pure component i, as
an ideal gas, at the temperature T (and a fixed reference pres-
sure P.), R is the ideal gas constant, P is the system pressure,
and y, is the mole fraction of species i. At equilibrium, the
chemical potentials of the components participating in the
chemical reaction must satisfy"31

4Vii = 0 (2)

where vi is the stoichiometrically balanced coefficient of
species i in the reaction (vy = -1, v2 = -3, v3 = 2). Upon sub-
stituting Eq. (1) into Eq. (2), rearrangement yields

y3- P2K(T)= Kp(T,P) (3)
where K(T) is the equilibrium constant and Kp(T,P) is a func-
tion ofT and P. The ratio of mole fractions on the far left side
of Eq. (3) is the "reaction quotient."
Now, let the system be at equilibrium at a given T and P At
the initial equilibrium state, there are no, no, and no moles
of each species with mole fractions y?, y', and y' satisfy-
ing Eq. (3). Next, we consider the addition of A moles of
nitrogen (1), while keeping T and P constant. As the system
re-equilibrates, the reaction proceeds so that the final mole
numbers of each species will be given byE81

n = n + A -

n2 = no 3

n3= no + 2

Fall 2003

where i is the extent of reaction starting from the above
initial equilibrium state; i is defined to be positive if the
reaction proceeds to the right, i.e., nitrogen and hydro-
gen are consumed while ammonia is produced, and nega-
tive if the reaction proceeds to the left. The above rela-
tions imply that the final mole fractions are given by

= n++A yo +A'- '
n+A-24 1+A'-24'

Y2= = 2
1 1+A'-2 '

Y3 = y +23 (5)
1+A'- 2'

where n =no +no +n', A'-A/n, and 4'--4/n; A'
and i' are dimensionless quantities. Since T and P are
held constant, Eq. (3) implies that

33 = Kp(T,P)= (y) (6)
(y? +A',- y 3')3 y(y4 )

Equation 6 is valid for all values of A' and can be used to
determine the value of 4' for a given choice of A'. The
sign of i', however, determines the direction in which
the reaction shifts, and its value determines the extent of
the reaction triggered by the addition of nitrogen.
First, we focus on how the extent of the reaction varies
when an infinitesimal amount of nitrogen, i.e., A' -+ 0,
is added. One can solve Eq. (6) for various values of A'
and then determine the sign of 4' as A' -) 0. But since
-' -> 0 as A' -> 0, we instead determine d4' / dA' ana-
lytically for A' -> 0. To proceed, and for ease of further
manipulations, we first rewrite Eq. (6) as

We now differentiate both sides of Eq. (7) with respect to
A' for constant y?, y', and yo, letting
n = d' / dA' = d' / dA. We then take the limit for A' -4
0, with -> 0 as well, and finally solve for nT. After a
few lines of algebra, we obtain

d yoy(1-2y1)
d- 0 (8)
dA 4yy (1-y )+yjy +9yly3

Since the denominator in Eq. (8) is always positive, the
sign of il is given by the sign of (1 2y'). Therefore, for
infinitesimal additions of nitrogen, we conclude that

when y0 <2-, >0:
reaction proceeds to the right,
consistently with the Le Chatelier Principle (9a)

when y0 >-, 1<0:
reaction proceeds to the left,
against the Le Chatelier Principle (9b)
0 1
when y? = -,= O:
no reaction takes place,
against the Le Chatelier Principle (9c)
Hence, the exception to Le Chatelier's principle occurs when the
initial equilibrium mole fraction of nitrogen is equal to or ex-
ceeds 0.5 and is independent of the mole fractions of the other
species or of the values of the temperature and pressure.
Why the reaction reverses direction can be understood qualita-
tively1m by considering the form of the reaction quotient in Eq.
(3). The addition of nitrogen increases the mole fraction of nitro-
gen, y', but also causes a decrease in y2 (hydrogen) and Y3 (am-
monia). Since y2 is cubed in the denominator of the reaction quo-
tient, the decrease in y2 may have a more significant effect on the
reaction quotient than the increase in y, or decrease in Y3" When
the mole fraction of nitrogen is small, the change in y, upon addi-
tion of some nitrogen yields a proportionally larger change in y,
as compared to the decreases in y2 and Y3 To compensate, and so
ensuring that the reaction quotient must remain equal to K,(T,P),
the reaction proceeds to the right, reducing some of the added
nitrogen and producing more ammonia. When the mole fraction
of nitrogen is large, so that the mole fraction of hydrogen is small,
the proportional decrease in y2 is greater than the increase in y,.
The decrease in y2 is magnified by the appearance of y23, and so
the reaction proceeds to the left, generating more nitrogen and

Figure 1. Extent of reaction versus the amount of
nitrogen added for yo = 0.5.

Chemical Engineering Education


0.6 0.7 0.8 0.9 1.0


yo)2 (Yo )3
A'- ')(y 34

0.0 0.1 0.2

hydrogen (thereby offsetting, to some degree, the decrease in y2)
Why the reaction shifts direction when y, exceeds 0.5 is, however,
not readily apparent from the above analysis. The composition at
which the reaction changes direction is, in general, dependent on
the form of the reaction quotient. 61
The above exceptions, relations (9b) and (9c), do not occur when
hydrogen or ammonia are added to the system or when the tem-
perature and volume are held constant. 1161 The addition of an inert
species does not alter these conclusions. Many other reactions can
exhibit such exceptions to Le Chatelier's principle, including liq-
uid-phase reactions. Katz161 considers in general the conditions un-
der which a reaction with several reactants and products shifts to
the left. The results depend on the stoichiometric coefficients of
the given species and can be analyzed similarly as above. Such
examples can provide useful teaching assignments, and use of

Figure 2. Extent of reaction versus the amount of
nitrogen added for yo = 0.7.

Figure 3. Extent of reaction versus the amount of
nitrogen added for yo = 0.25.

nonideal gas models may provide further enrichment.
Equation (8) and the conclusions of (9) are valid only
for infinitesimal additions of nitrogen at constant T and
P. Equation (6) or (7), however, is valid for finite addi-
tions of nitrogen at fixed T and P. To determine how the
reaction shifts upon addition of a finite amount of nitro-
gen, one must use Eq. (6) or (7) to determine the value
(and sign) of !' for a given value of A'. Given that y3
= 1 y,O y2o, Eq. (7) can be rewritten as

(1- y y + 2 ')2(1+ A'- 2 ')2y(y)3

= 1- y- y2)2 (y + A'- '(y- 3!') (10)

This equation is fourth-order in 4' with roots that de-
pend on A', y1o, and y,. The physically relevant value
of i' must be such that each of the terms in parentheses
in Eq. (10) and the final mole fractions of Eq. (5) lie
between 0 and 1. For small to moderate values of A',
the physically relevant solution of Eq. (10) is therefore
small. Equation (10) was solved by the standard New-
ton-Raphson's method with an initial guess of = 0.
Convergence to the one physical root was readily
achieved. Values of were generated for a range of A'
at a given yo and y,. One may vary y2 independently
from y o, with the constraint that 0 < y20 < 1 y, Equa-
tion (10), like Eqs. (6) and (7), also does not require that
one specify the pressure and temperature explicitly. The
value of y2 for a given yo must be consistent with the
choice of T and P in Eq. (3), but the specific value of T
and P is not required for the determination of W'.
Figure 1 displays a plot of 4' versus A' for yo0 = 0.5
and different values of y20. In all cases, !' -4 0 as A'
-* 0, and the slope of approaches zero as A' -> 0,
consistent with relation (9c). Nonetheless, for any finite
addition of nitrogen, the reaction proceeds to the left,
i.e., 4' < 0. The value of is quite small, remaining so
as A' -* 1. The reaction proceeds to the left, but only
by a relatively small extent, even if nitrogen is added in
an amount equal to the total number of moles of all the
species initially present (e.g., = 0.006 for A' = 1.0
and y20 = 0.30). The extent of reaction, though always
negative, also depends on y2.
Figures 2 and 3 display plots of 4' versus A' for y, =
0.7 and y,0 = 0.25, respectively, at different values of
y2. When y = 0.7, the limiting slopes are all negative,
consistent with relation (9b), and a finite addition of ni-
trogen, at least up to A' = 1, causes the reaction to pro-
ceed to the left (t' < 0). When y,0 = 0.25, the limiting
slopes are all positive, again as required by relation (9a),
and a finite addition of nitrogen up to A' = 1 causes the
reaction to proceed to the right (! > 0). Yet, the curves
in Figure 3 display maxima, suggesting that for suffi-

Fall 2003


0.0 0.1 0.2 0.3 0.4 0.5

0.6 0.7 0.8 0.9 1.0

ciently large A', the curves will eventually yield nega-
tive values of 4'.
To illustrate this effect further, we consider Figure 4
in which y, = 0.45. As required by relation (9a), all the
curves initially have a positive slope, so that the reac-
tion proceeds to the right for small values of A'. At
some critical value of A' (which depends slightly on
y20), 4' becomes negative. Therefore, at y, = 0.45, the
reaction proceeds to the right for small additions of ni-
trogen, but shifts to the left when a sufficient quantity
of nitrogen is added. The figures suggest that for all
values of y,, the reaction will eventually proceed to the
left if a large enough amount of nitrogen is added to the
reaction vessel. These calculations also suggest ways
for experimentally testing the predictions and demon-
strating the thermodynamic laws.


The Principle of Moderation

We close this paper by briefly discussing the general-
ized statement of Le Chatelier as stated and proved by
de Heer,141 which is more appropriately called a Prin-
ciple of Moderation (the proof given by de Heer is be-
yond the scope of an undergraduate course). Since gen-
erality often causes one to sacrifice simplicity, the more
general principle of moderation may be given as'14
The change of an intensive variable caused by
changing the corresponding (conjugate) extensive
variable is smaller if chemical equilibrium is
maintained than if no reaction could take place in the
The change of an extensive variable caused by
changing the corresponding (conjugate) intensive
variable is larger if chemical equilibrium is
maintained than if no reaction could take place in the

The above statement has been shown to be valid only
for infinitesimal changes from the initial equilibrium
state.14,51 It does not necessarily hold for finite changes.511
The principle of moderation applies, for example, to
the change of the chemical potential of nitrogen (inten-
sive variable) upon the addition of nitrogen (conjugate
extensive variable). With the ammonia synthesis reac-
tion prevented from occurring, the addition of nitrogen
at constant T and P yields an increase in the value of the
chemical potential of nitrogen. If nitrogen were instead
added while maintaining chemical equilibrium, i.e., the
reaction was allowed to proceed, the resultant increase
in the chemical potential of nitrogen would be smaller

than the increase obtained when the reaction was prohibited. In
other words, the change in the chemical potential of nitrogen is
moderated, or lessened in this example, by the reaction.
Since at constant T and P the chemical potential of a component
in an ideal-gas mixture increases with increasing mole fraction, we
instead analyze the change in the mole fraction of nitrogen upon
the addition of nitrogen to illustrate in more detail the principle of
moderation (this is one aspect of the ammonia synthesis reaction
that was not directly discussed by de Heer). We consider the addi-
tion of nitrogen with the ammonia synthesis reaction taking place
and with the reaction prevented from occurring. From Eq. (5), we
see that in the limit of A' -> 0

dyl- y +n(2y ) (11)

If no reaction is allowed to take place, then the final mole fraction
of y, is equal to

Sy+A' (12)
Taking the derivative of y, with respect to A', we find in the limit
of A' -> 0 that

dy1 n

=l- yo > 0

Equations (11) and (13) imply that
dy ( dy (2y 1)
dA IdA'-)no rxn + 1

Since Eq. (8) indicates that il(2y1 1) < 0, then

dyl <(dyl )
dA' -dA'no rxn

0.0 0.1 0.2 0.3

0.4 0.5 0.6 0.7 0.8 0.9 1.0

Figure 4. Extent of reaction versus the amount of
nitrogen added for y1, = 0.45.

Chemical Engineering Education

or the change in the mole fraction of nitrogen is always less
(or equal) when the reaction proceeds than when no reaction
takes place. The reaction is said to "moderate" the mole frac-
tion of nitrogen, i.e., the reaction decreases the final mole
fraction of nitrogen as compared to the case when no reac-
tion occurs. This result is counter-intui-
tive, since Eq. (15) is also valid when
the reaction shifts to the left, thereby pro-
ducing more nitrogen. In this instance,
the additional hydrogen that is pro-
duced by the reaction shifting to the _wof y I
left offsets the increase in the mole
fraction of nitrogen. .
Similar arguments, as discussed by de -
Heer, show that the partial pressure and .
chemical potential of nitrogen are also '
moderated by the reaction: the change
in these intensive quantities is always -
less when the reaction occurs than when J
no reaction takes place. Equation (15)
also supports these conclusions since, at
constant T and P, the partial pressure and i
chemical potential of a component in an
ideal gas mixture increase with increas- s
ing mole fraction. .
For finite additions of nitrogen, and
for most conditions, calculations show Igct z
that Eq. (15) is still satisfied. There are
ranges of finite A', however, in which prob
the reaction does not moderate the
change in mole fraction of nitrogen (a
value of yo0 = 0.45 provides an example
for A' between approximately 0.12 and
0.22). In this case, the final mole fraction of nitrogen after
reaction is in fact greater than the final mole fraction of nitro-
gen without reaction. The ratio of the final mole fraction of
nitrogen with reaction to the mole fraction without reaction
is, however, only slightly greater than unity. Thus, in this case,
the violation of the principle is minor for finite values of A'.


When a gas-phase system is at chemical reaction equilib-
rium at constant temperature and pressure, and some extra
reactant or product is added, the system, upon reestablishing
equilibrium does not always respond in a way qualitatively
consistent with the traditional Le Chatelier principle. If there
is a change in the number of moles, as in the ammonia syn-
thesis reaction, then adding one reactant (N,) may cause the
reaction to proceed in a direction that produces more of the
added ingredient (N2). These results are perfectly consistent
with the laws of thermodynamics. The direction of the reac-
tion depends on whether the added amount is infinitesimal or
finite. For infinitesimal additions, a new principle of mod-

eration, first suggested formally by de Heer,14 does apply.
Even this principle does not apply for finite additions of re-
actant (N,). These results indicate that the principle of Le
Chatelier should be taught in its more general form. In addi-
tion, instructors should emphasize that even the more gen-
eral formulation is valid for infinitesimal
changes only. (Nonetheless, the principle
does appear to be valid for finite changes
t- in temperature and pressure.) The present
analysis provides for a deeper understand-
- VPi-- S ing of chemical reaction equilibria and can
form the basis of several stimulating lec-
tures and problem-solving sessions.
-t .. As a final note, a general statement con-
cerning the direction of shift for changes
S...-- in temperature of arbitrary amounts at
constant pressure can be formulated. As-
- suming that the heat of reaction is always
_ positive or negative, then an increase in
temperature of any amount will cause the
m- equilibrium to be displaced in the direc-
tion of the heat of reaction. A similar state-
ment appears to be possible for pressure
_C changes at constant temperature and de-
pends on which direction the volume
changes upon reaction. The addition of
uzn1 reactants and/or products, however, re-
quires care, and in these cases it is not
S possible to formulate a general statement
that is universally valid for any addition
of products or reactants.

The work described in this paper was partially supported
by an Academic Reinvestment Proposal, Purdue Research

1. Levine, I.N., Physical Chemistry, 3rd ed., McGraw-Hill Book Co.,
New York, NY (1988)
2. Sander, S.I., Chemical and Engineering Thermodynamics, 3rd ed.,
John Wiley & Sons, Inc., New York, NY (1999)
3. Tester, J.W., and M. Modell, Thermodynamics and Its Applications,
3rd ed., Prentice Hall PTR, Upper Saddle River, NJ (1997)
4. de Heer, J., "The Principle of Le Chatelier and Braun," J. Chem. Ed.,
34(8), 375 (1957)
5. Liu, Z.-K., J. Agren, and M. Hillert, "Application of the Le Chatelier
Principle on Gas Reactions," Fl. Phase Equil., 121(1-2), 167 (1996)
6. Katz, L., "A Systematic Way to Avoid Chatelier's Principle in Chemi-
cal Reactions," J. Chem. Ed., 38(7), 375 (1961)
7. Plaza del Pino, I.M., and J.M. Sanchez-Ruiz, "A Simple, Experimen-
tal Illustration of the Le Chatelier Principle," J. Chem. Ed., 68(11),
8. Smith, J.M., H.C. Van Ness, and M.M. Abbott, Introduction to Chemi-
cal Engineering Thermodynamics, 6th ed., McGraw-Hill Book Co.,
New York, NY (2001) 0

Fall 2003







re M. laboratory




Ecole Polytechnique Montreal, Quebec, Canada H3C 3A7

Mixing is a common operation in the process indus-
tries and is generally performed by a rotating im-
peller in a vessel. Products obtained from food,
petroleum, mining, pharmaceutical, pulp and paper, and
chemical industries would not be available without fluid mix-
ing equipment and technology. Mixing also plays a vital role
in industrial waste treatment and in environmental cleaning,
such as in sulfur dioxide absorption for treatment of acid
A wide range of mixing situations can be found in practice,
which may involve high- or low-viscosity fluids, suspending
solids in liquids, dispersing gas or solids in liquids, etc. Mix-
ing operations at the industrial level are increasingly carried
out at low to moderate Reynolds numbers, leading to segre-
gated or dead regions and resulting in long mixing times.
The simplest way used to improve mixing efficiency con-
sists of increasing the rotational speed, which unfortunately
leads to higher energy consumption. Mixing times in small-
scale stirred tanks are commonly measured by non-intrusive
techniques such as colorimetry. This technique also allows
observation of the aforementioned segregated regions and
how they tend to disappear as the impeller speed increases.
The objective of the mixing laboratory is to give students
practical experience in the fluid mechanics of mixing by ana-
lyzing power consumption and mixing times associated with
radial and axial flow impellers with Newtonian and non-
Newtonian fluids.
The mixing laboratory is part of an undergraduate unit op-
erations course offered by the Department of Chemical En-
gineering at Ecole Polytechnique of Montreal for senior-year
students. Groups composed of a maximum of three students
perform the required laboratory work in a period of four hours.
The group prepares a preliminary report after finishing the
experiments, and the students either hand over a full report

or give an oral presentation the following week. Both the full
reports and the oral presentations consist of a description of
the experiment's objectives, its theoretical basis, the engi-
neering method used, the experimental setup, and the operat-
ing conditions. Then they present the experimental data, dis-
cuss the results, and make recommendations to improve the

The mixing system used in all the experiments is a modi-
fied Turbotest (VMI Rayneri) laboratory mixer, shown in
Figure 1. It consists of a transparent polycarbonate vessel of
165-mm inner diameter and 230-mm height, with an open
top fixed to a rigid table for safe operation. Two classical
impellers are tested-a radial-flow impeller (6-blade Rushton
turbine) and an axial-flow impeller (marine propeller). The
impellers are mounted on a rigid shaft driven by a DC motor,
with the speed carefully regulated in a range from 10 to 2500
rpm by means of a DC controller. The motor is mounted on a
rigid structure that can be moved to adjust the vertical posi-

Gabriel Ascanio received his BS and MS from the National University of
Mexico in 1988 and 1995, respectively, and his PhD from Ecole
Polytechnique of Montreal in 2003. He is currently a postdoctoral fellow at
URPEI in the Department of Chemical Engineering. Some of his research
interests are in coating processes and mixing of rheology complex fluids.
Robert Legros is Professor of Chemical Engineering at Ecole
Polytechnique of Montreal. He received his BS from Ecole Polytechnique
in 1983 and his PhD from the University of Surrey in 1987. His academic
research involves solids thermal treatments in fluid beds, modeling of com-
bustion reactors, heat and mass transfer, and hydrodynamics of spouted
beds. Some of his current research interests are related to pharmaceutical
engineering, namely in powder technology and downstream processes.
Philippe A. Tanguy is Professor of Chemical Engineering at Ecole
Polytechnique of Montreal. He received his BSc in 1976 and his Doctorat
de spdcialitd in 1979, both from Universit6 de Paris, and his PhD from
Laval University in 1982. His research interests are in non-Newtonian fluid
mechanics, CFD and process engineering involving complex fluids, in par-
ticular coating processes, and in agitation and mixing operations.

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

The purpose of this experiment consists of determining the mixing time with two impellers
providing different flow patterns. The mixing time, defined as the time needed to reach
a specified degree of homogeneity, can be determined by various techniques ...

tion of the impeller. As can be seen in Figure 1, a standard
mixing configuration is used as a starting point, with the im-
peller placed on the vessel centerline at 1/3 of the liquid height.
The agitation torque is measured by a non-contact type
torquemeter (range between 0.1 and 1.42 N.m) fitted between
the motor and the agitation shaft.
Newtonian fluids consist of aqueous solutions of corn syrup
having a viscosity of 1.5 Pa.s, while aqueous solutions of
carboxy methyl cellulose (CMC) are employed as non-New-
tonian fluids. The mixing times are measured with a colored
tracer consisting of Methylene Blue diluted in both solutions.
The two solutions, together with the tracer, are prepared prior
to the experiments and allowed to settle at least 24 hours in
order to eliminate air bubbles. The rheological properties of
the fluids are determined
by a Bohlin Visco 88V
viscometer using a con- MOTOR_-
centric cylinder configu- MOTOR
ration. Rheological mea-
surements and the ex- TORQUEMETER
periments are performed
at room temperature
(around 24oC). LIQUID LEVEL

The cost of the labora-
tory mixer and the solu-
tions used for the experi-
ments is about $5,000 and
$20, respectively.


Figure 1. Experir

Power Consumption
This experiment consists of determining the power con-
sumption for both radial- and axial-flow impellers with New-
tonian fluids. For that purpose, the fluid under study must be
added to the H level of the tank (see Figure 1) and then the
mixer speed is set to zero rpm and the torquemeter to zero
N.m. The impeller speed is gradually changed from 15 to
700 rpm, and the torque reading for each speed is used to
calculate the power consumption by means of

P = 2 TirNT (1)

where r is the impeller radius in m, N is the rotational speed
in rps, and T is the torque in N.m.
The power consumption is correlated to the impeller speed
by means of the dimensionless power number (Np) and the
Reynolds number (Re), defined by

Np= pND

and Re = -pND

where P is the power in Watts, p is the fluid density in kg/m3,
D is the impeller diameter in m, and [L is the dynamic viscos-
ity in Pa.s.
The laminar and transition regimes must be identified after
plotting Np as a function of Re on a log-log scale;13'41 the
constant K for each impeller can be calculated by

Kp = Np Re (3)
Mixing Times
The purpose of this experiment consists of determining the
mixing time with two im-
pellers providing differ-
ent flow patterns. The
mixing time, defined as
the time needed to reach
a specified degree of ho-
RUSHTON TURBINE mogeneity, can be deter-
mined by various tech-
niques based on the mea-
surements of concentra-
HT = 1 PROPELLER tion, density, electrical
S= 1/3 conductivity, tempera-
ture, or by colorimetry,
optical methods, thermal
method, etc. The colorim-
etry technique is a quali-
zental setup, tative method to deter-
mine the mixing time by
adding a small amount of a color tracer to the fluid that is
being mixed. The overall fluid color will change, and mixing
time corresponds to the time when the tracer is judged to
have completely dispersed within the fluid. The detailed pro-
cedure for measuring the mixing time is

1. Use the configuration shown in Figure 1 with the Rushton
2. Add fluid (aqueous corn syrup or aqueous CMC) up to the H
3. Prepare the color tracer solution by dissolving 10 mL of
Methylene Blue in 100 mL of fluid to be studied.
4. With the mixer at rest, add 15 mL of the color tracer solution
to the tank containing the fluid.
5. Set impeller speed at 100 rpm and switch the mixer on.
6. Measure the mixing time at this speed.
7. Repeat Steps 3 to 5, using different speeds. The speed range

Fall 2003


for this experiment is from 100 rpm to 600 rpm, with
increments of 100 rpm.
8. Repeat the experiment for the marine propeller.
As proposed by Moo-Young, et al.,E 1 mixing time can be
correlated with the impeller speed by means of a dimension-
less mixing time defined as

Ntm = a(Re)' (4)
where tm is the mixing time in s, N is the impeller speed in
rps, Re is the Reynolds number, and oa and P3 are adjustable
Shear Rate of non-Newtonian Fluids
The purpose of this experiment is to find out the effective
shear rate for non-Newtonian fluids in the vicinity of the im-
peller by the Metzner-Otto correlation.[61 They developed a
general relationship to correlate the impeller speed and the
shear rate of a non-Newtonian fluid in the laminar regime.
Based on the single knowledge of the power curve for
Newtonian fluids, this relationship can be used to interpret
and correlate power draw data for non-Newtonian fluids. This
method assumes that there exists an average mixer shear rate
developed in the vicinity of the impeller, which corresponds
to the power consumption. This shear rate is directly propor-
tional to the impeller speed through

7A = ksN (5)
where k is the mixer shear rate constant.
The average shear rate, Y'A, defines an apparent viscosity,
which is used in the definition of the Reynolds number for
power consumption prediction for non-Newtonian fluids. The
apparent viscosity is determined from viscometric measure-
ments at the appropriate shear rate and used directly for plot-
ting the power curve. The determination of the average shear
rate, YA involves the following steps (see Figure 2):


1. For a given impeller speed, a power number (Np') is
calculated from the P vs. N for non-Newtonian fluids.
2. Using this power number, Np', a Reynolds number (Re')
is obtained from the power number-Reynolds number corre-
lation for Newtonian fluids.
3. Finally, the average shear rate can be determined from
the viscometric data and, using the impeller speed, the mixer
shear rate constant, k,, can be calculated from Eq. (5).
The procedure for this experiment consists of the following
1. Mount the Rushton turbine at the end of the shaft and
locate it in the center of the vessel, as in the first
2. Add the aqueous CMC to the H level.
3. Gradually change the speed, record the torque for
each speed, and calculate the impeller power con-
sumption from Eq. (1).
4. Plot the power consumption (P) vs. impeller speed
5. By using the viscometer with the same fluid, record
the apparent viscosity for each shear rate and plot the
PIA vs. YA curve.
6. Following the steps mentioned above, determine the
average shear and calculate the mixer shear rate
constant ks from the Metzner-Otto correlation (Eq. 5).
7. Repeat the experiment using the propeller.

As mentioned before, students are asked to prepare a pre-
liminary report after finishing the experimental work. A week
later they must deliver a full report or give an oral presenta-
tion. The full reports must contain

N' N ie log Re YA

2nN'P' D2 N'p ks
D (NY p A Re* N
(a) (b) (c)

Figure 2. Determination of the shear rate constant ks: a) non-Newtonian power consump-
tion, b) Newtonian power consumption, c) non-Newtonian viscometry.

Chemical Engineering Education


10An abstract, including the objectives, the methodology
used to achieve the objectives, and results and conclu-
sions in relation to the proposed objectives.
0 The objectives must be clearly stated.
1,A theoretical perspective different from the one presented
in the laboratory manual.
0-Main results for discussion and analysis, including
graphs and tables. An example of a set of experimental
data obtained by students is shown in Figures 3 and 4.
The power curves in terms of the dimensionless power
number (Np) as a function of the Reynolds number (Re)
are shown in Figure 3. After performing linear regres-
sion with the experimental data, a good correlation can
be observed between Np and Re. It must be noted that a
slope of -1 should be obtained between Np and Re cor-
responding to the laminar region. An error of 5.47% and
0.53% in the slope is obtained for the Rushton turbine

10 --- -- --- i' -
S32.705Re5" -*- Rushton turbine
= 0.9948 --0-- Propeller


Np = 12.649ReW'"7
R2 = 0.9997

Figure 3. Experimental power curves for the Rushton
turbine and the propeller.

S-0-- Rushton turbine ;
2500 N Nt =3x106Re'13 --0- Propeller
R'= 0.9939


500 Ntm = 6xl 0a Re. '
R2= 0.997
0 -

4 6

10 12 14

and the propeller, respectively. Figure 4 shows the di-
mensionless mixing time as a function of the Reynolds
number for both impellers with a Newtonian fluid. From
the linear regression results, it can be observed that the
larger coefficients o and 3 correspond to the Rushton
turbine, which is in good agreement with the results re-
ported in the literature.121
Interpretation, analysis, and discussion. These elements
should be presented in great detail in a quantitative way,
including the experimental error encountered. In the case
of the experiments of power consumption and shear rate
of non-Newtonian fluids, the torque should be measured
at least three times in order to determine the experimen-
tal error.
I Recommendations. This feature is used as feedback chan-
nel, so the students should suggest another experiment
to perform or modifications to the experimental setup in
order to improve the experiments.
1'Appendix. All the raw data must be presented so the re-
viewer can verify if the data were well processed.
On the other hand, the oral presentation is evaluated in terms
of the form and the content. The introduction and objectives,
presentation structure, illustrations, conclusions, and ques-
tions are all considered in the form. The subject knowledge,
theoretical basis, and references and analysis capability are
considered as parts of the presentation.

Because mixing is a unit operation involved in many in-
dustrial applications, a good understanding of this operation
is central for a successful process. The proposed experiments
give the students a general introduction to the fluid mechan-
ics of mixing with Newtonian and non-Newtonian fluids,
using impellers that provide different types of flow. In fluid
mixing technology, as in other process design areas, dimen-
sionless groups are used to correlate scale-up parameters. For
that reason, experimental results must be presented in terms
of these dimensionless numbers to be useful to the process
designers. The proposed mixing experiments enable engi-
neering students to gain excellent insight into the use of
dimensionless groups.

1. Coulson, J.M., J.F. Richardson, J.R. Backhurst, and J.H. Harker, Chemi-
cal Engineering, Vol. 1, Pergamon Press, p. 225 (1990)
2. Harnby, N., M.F. Edwards, and A.W. Nienow, Mixing in the Process
Industries, 2nd ed., Butterworth Heinemann (1992)
3. Rushton, J.H., "The Use of Pilot Plant Mixing data," Chem. Eng. Prog.,
47, No. 9, p. 485 (1951)
4. Rushton, J.H., E.W. Costich, and H.J. Everett, "Power Characteristics
of Mixing Impellers, Part 1." Chem. Eng. Prog., 46, No 8. p. 395 (1950)
5. Moo-Young, M., K. Tichar, and FA.L. Dullien, "The Blending Effi-
ciencies of Some Impellers in Batch Mixing,"AIChE J., 54,139 (1976)
6. Metzner, A.B., and R.E. Otto, "Agitation of Non-Newtonian Fluids,"
AIChE J., 3(1), 3 (1957) 17

Fall 2003

Figure 4. Dimensionless mixing time as a function of
the Reynolds number for Newtonian fluids.

BI, -classroom




Illinois Institute of Technology Chicago, IL 60616

he quality of student learning can be enhanced sig-
nificantly by simulation of complex systems with user-
friendly software. Complex real-world problems and
solutions can be introduced to students by using simulation
systems to conduct virtual experiments. These virtual experi-
ments are especially useful when their real-world analogs are
expensive and/or dangerous.
Simulation involves the use of the model of a system to
observe the system's response to changes in properties and
inputs to the system. Simulations can introduce realistic prob-
lem situations and support a particularly active form of learn-
ing by letting students manipulate the conditions of the sys-
tem and observe the consequences of those variations. The
availability of a reliable and realistic mathematical model
is essential to conduct simulations and understand the be-
havior of the system.
We have developed a dynamic simulator for glucose-insu-
lin interactions in a healthy person and a Type- 1 diabetic pa-
tient. The aims of this public domain simulator are to provide
assistance to bioengineering students in learning glucose-in-
sulin interactions in the human body, to offer a tool for engi-
neering students in learning system dynamics, and to pro-
vide an illustrative tool to diabetic patients. The simulator
cannot be used for adjusting a patient's insulin dosage regu-
lation in real life, but may be helpful for patient education.
Both MATLAB-based-stand-alone and web-based graphi-
cal user interfaces (GUI) are software designs that yield in-
teractive systems. Despite many similarities between the two
designs, there are also many differences. Simulators with web-
based GUIs are more accessible over the internet, but stand-
alone software can also be distributed widely while giving

* Pharm Tech, Inc., 14048 Petronella Dr., Libertyville, IL 60048
** Northwestern Uiversity, Biomedical Engg., Evanston, IL 60208

the developer/designer an opportunity to keep track of the
users. Furthermore, since the stand-alone GUIs are in a de-
fined frame, the designer controls where the user goes when
browsing among the links, while in web-based GUIs, the user
has all the control and in many typical situations there is a
high possibility that he or she may branch out to an arbitrary
website while browsing.
Modeling glucose-insulin interactions requires an under-
standing of the physiological and metabolic processes that
determine observable behavior.111 Mathematical models de-
scribing carbohydrate metabolism are available in the litera-
ture.12-41 We have used and extended two mathematical mod-
els based on pharmacokinetic diagrams of glucose and insu-
lin in the human body. Mass balances for both glucose and
insulin resulted in a set of ordinary differential equations,
and the models are implemented in a computer program writ-
ten in MATLAB 5.3. The mathematical details of these mod-
els are available elsewhere. 1l51 ODE23 (low-order Runge-
Kutta routine) is used for solving differential equations. The
models are then integrated with a GUI that is responsible for
presenting information to the user in a clear and friendly way.

Fetanet Ceylan Erzen received a BS degree in chemical engineering
from the Middle East Technical University (1999) and an MS degree from
Illinois Institute of Technology (2000). Her thesis studies included model-
ing and simulation of glucose-insulin interaction in the human body and
graphical user interface development.
GOInur Birol received BSc, MSc, and PhD degrees in chemical engi-
neering from Bogazici University in 1990, 1992, and 1997, respectively.
Her current research interests include glucose biosensors, investigation
of retinal vascular occlusion, and the relationships between oxidative and
glycolytic metabolism in the retina on animal models.
All Cinar received a BS degree in chemical engineering from Robert Col-
lege, Turkey (1970), and MEng (1973) and PhD (1976) degrees from Texas
A&M University. His teaching and research interests are process model-
ing and control, statistical process monitoring and fault diagnosis, and
use of knowledge-based systems for real-time process supervision and

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

EL System Availability
GLUCOSIM was originally developed in MATLAB 5.3.1 on
a PC platform. It requires three MB of hard disk space. A
demonstration of the package is available on the web at />. The simulator can be
obtained by writing to Ali Cinar (e-mail at
Ea User Interface Design
A computer is limited not by its power to compute, but rather
by its power to communicate with its human users.16] The
main requirement for wide acceptance and use by the stu-


Tutorial & Information Files

Demo I <> Simulator

I .

User Inputs

Simulation Models
Normal Man
--* Diabetes Patient


dents is an easy-to-learn, easy-to-use efficient interface.7' A
simple and natural dialogue for modem computer systems
with GUIs can be achieved by a good graphic designt'8 and
consistent screen layouts. Several guidelines are followed for
this purpose while designing the GUI in this work:
Consistency of the user interface. Similar objects and colors
are used to perform similar functions throughout the
simulator to facilitate recognition. If users know that the
same command or the same action will always have the
same effect, they will feel more confident in using the
system.'91 Also, the design is limited to a small number of
consistently applied colors.
Ease of navigation. The user is able to navigate without
getting lost or worrying about causing harm.
Importance of help and documentation is kept in mind. If
students need to refer to documentation for help or for
background information, there is sufficient and comprehen-
sive, but brief, documentation throughout the simulator.
Navigation is also available between the documentation; for
example, users can return to the tutorial while reading a help
Dialog boxes have a quit and/or back button. This gives
users a feeling of being in control since the user rather than
the computer decides where to go, what to see, and when to
The structure of the simulator is illustrated in Figure 1 and
its capabilities are outlined in Table 1. There are three cat-
egories of windows in the simulator: information windows,
transition windows, and input/output windows. A detailed de-
scription of these windows is presented later in this paper.

El Model Equations
The pharmacokinetic models for glucose and insulin are
based on mass balance equations on various physiological
compartments such as heart, lungs, and arteries (H), nervous
system (NS) (for glucose), or subcutaneous tissue (SC) (for
insulin), liver (L), pancreas (PN), gastrointestinal tract (GT),
kidney (K), and periphery (PR) (skeletal muscle and adipose
tissue). For example, the circulating blood insulin concentra-
tion, I is described by

where Q denotes the blood-flow rate (dl/min), V denotes the
volume (dl), t denotes time (min), and I denotes the insulin
concentration (mg/dl). Subscripts B and HA denote blood and
hepatic artery, respectively. The mass balance in subcutane-
ous tissue is

VsC =Qsc(IB Isc) + rIA (2)

where r denotes a metabolic source or sink rate, and the sub-
script IA denotes insulin absorption. The detailed model con-
sists of a system of ordinary differential equations represent-


Figure 1. Structure of the simulator.

Program Features

Large Database Related internet sites
Diabetes dictionary
Carbohydrate values
Over 100 references
Operational Modes Demo
Experimental Modes Oral glucose tolerance test
Healthy person
Type-1 diabetic patient
Inputs Carbohydrate content
Time of meal and injection
Insulin type and dose
Body weight
Duration of exercise and simulation
Save Options and Outputs Continuous graphical display
Saving in ASCII and graphic modes
Recall/display profiles from previous runs

Fall 2003

ing mass balance equations in all compartments. The overall
model is derived by assuming that the mass balances in
each tissue are in quasi-steady state (i.e., dl/dt = dG/dt =
0). The resulting algebraic equations for glucose and in-
sulin concentrations are combined into the glucose and
insulin balances in the blood, yielding an overall model
with two differential equations.
Two models of insulin release were taken from the literature,
modified, and used in the current simulator for healthy hu-
mans. The first one is based on islet insulin secretion model
developed by Nomura, et al., E10 for rat islets, and the second
one is based on the pancreatic insulin release model devel-
oped by Carson and Cramp.E""
EL Features
A MATLAB-based, user-friendly GUI was designed and
integrated with the MATLAB code written for the mathemati-
cal model. The interaction of the user with the software has
been kept as simple as possible. Menus, buttons, and sliders
are widely used as controlling elements. Values are displayed
graphically with a "save" option. "Help" windows through-
out the program are available and the user can quit the pro-
gram at any time. Furthermore, the simulation can be stopped
at any time by using the "stop" button on the simulation
progress bar. The main window (see Figure 2) is designed to
familiarize the user with the environment.
There are three buttons-"About," "Tutorial," and "Back-
ground Information." The "About" button gives a brief in-
troduction to the program; "Tutorial" provides information
about the model used, along with a short literature review;
and the "Background Information" button is linked to an-
other window where it is possible to search for the definition
of a word related to diabetes from the database created (Fig-
ure 3) to view the relevant web links (Figure 4) or to get
information about the references used on the development of
both the mathematical model and the simulator.

Figure 2. Main window.

By using the "NEXT" button placed in the bottom left-
hand side of the main window (see Figure 2), the user can
choose between "DEMO" and "SIMULATOR" modes. The
purpose of the DEMO, which consists of snapshots, is to give
the user a general idea of the simulator's capabilities and a
preview of how the simulator functions. By selecting the
SIMULATOR mode, the user goes from the "information"
mode to an "experiment" mode. Here, there are three options
for the virtual experiments. The first option performs the "Oral
Glucose Tolerance Test" (OGTT), the second and third op-

Figure 3. Dictionary

M,110 (hH 8 1 4 10H 3i- *Ofjf Ad~

7, Tb.-h.I.d-fl

flq ~ I ib..4 I

h.. -1-4 Al..dlc. M ib.. pK".,.~

sm61 "01k4- .6

P W. .ahm. I ,....hAul. I,. PIk., %ibb D~"

100%b %t~ -. .41.1 A"4 i,.tr~~~ .i-.

Figure 4. Links

Chemical Engineering Education

tions simulate a "Healthy Person" and a "Diabetic Patient,"
respectively. The user has the flexibility to choose between
the two different models for the healthy mode (Model I and
Model II) and between the two different models for the Type-
1 diabetes mode (detailed model and overall model).
For OGTT, the only input is the weight of the person. There
is also an option where the user can load his or her own pre-
viously saved data. Inputs for the other two modes (Figures 5
and 6) are
1. Carbohydrate content of the meal. There is also a
nutritional database where the user can find the
carbohydrate content of a specific meal.
2. Time of meal and injection. The user can enter a
value between 0-24 hours for time of meal and
insulin injection.
3. Insulin type and dose. Two types of insulin are
available, i.e., regular and ultralente.
4. Body weight
5. Duration of exercise. The exercise option, which is

Figure 5. Main window for Type-1 diabetes mode.

Figure 6. Main window for healthy person mode.

Fall 2003

specifically designed for moderate exercise,1121 is
available for only Type-1 diabetes mellitus mode.
6. Duration of simulation. It is possible to simulate the
dynamics of the diabetic patient and a normal person
for a maximum of 24 hours with up to four injec-
tions in diabetes mode.
The outputs of the simulation are displayed by continu-
ously updating the figures displayed on the screen (Figure
7). Once the simulation is finished, data can be saved in ASCII
and/or graphic form to make the recall and display of the
profiles possible for further analysis.

EN Overview of the Course
The course focuses on application of engineering principles
to biochemical and biomedical systems. Biochemical engi-
neering topics include biological systems, enzymes and mi-
crobial kinetics, and design and analysis of biological reac-
tors. Biomedical engineering topics include flow properties
of blood, transport in human cardiovascular systems, and
analysis and design of artificial organs. Half of the semester
is spent on biomedical engineering, while the other half is
used for biochemical engineering. Details of the course are
documented elsewhere.[131 The average number of students
registering for this class is around twelve every semester.
The primary learning goals of the course are to provide
students with basic principles in cellular biology of micro-
bial cells, bioreactor operations and transport phenomena in
living systems, and enzyme and microbial kinetics and phar-
macokinetics-in short, to provide them with a working
knowledge of bioengineering applications.
The course was designed to achieve these learning objec-
tives that were assessed using fairly traditional methods (i.e.,
homework assignments, examinations, and term projects)

Figure 7. Output window.

throughout the semester. The class has been
taught by the same instructor for three semes-
ters at IIT and has been updated from semes-
ter to semester to better meet the learning ob-
jectives and the needs of the students in an
effective learning environment. During this
time, it has become a popular course. The
overall rating of the instructor and the course
increased 18% and 31%, respectively, since
its inception. It was rated 15% and 12%
higher when compared to the average instruc-
tor and average course ratings of the depart-
ment (Chemical and Environmental Engi-
neering Department), respectively.

aE Implementation of the Simulator

Previously, the beta version of the simula-
tor had been tested in the course in the Fall
'99 semester.1131 Based on the feedback pro-
vided by the students, MATLAB codes were
updated and the gamma version was inte-
grated into course material in the Fall '00

This was one of the term projects in which
students were expected to use the simulator
for a period of two weeks, and it formed 7.5 %
of the class grade. The simulation package,
which was distributed to students on a CD,
was introduced immediately after the phar-
macokinetics topic had been covered in a se-
ries of lectures. Distributing the software on
CDs helped students work from home as well
as from different PCs as long as they had
MATLAB software installed.

The students were asked to run a series of
simulations at different conditions, and the
choice of models was left up to the students
so they could have the opportunity to inves-
tigate the parts) they wanted and were most
interested in. Their interests varied signifi-
cantly. Some tried all the combinations (de-
scribed in the Features section), while others
focused on investigating a single issue (e.g.,
the effect of body weight on glucose levels).

At the end of the two-week exposure to the
simulation package, a class discussion was
organized where students could share their
experiences with the simulator, talk about
their findings, and make conclusions in an
informal discussion setting. During the two-
week period, the instructor and a graduate
student were available outside of class to as-

Survey Questions for the Project

1. How could the simulator and GUI be improved?
2. How computer literate do people need to be in order to use the simulator:?
3. How much do people need to know about human physiology in order to benefit from the simulator?
4. In your opinion, does the product have any educational (or other) value?
5. What did you learn from the project? What else would have been interesting to learn?

Survey Questions and Responses for the Simulator

iStd. Dev,


Design of Screens
1. The design of the Home Page (main page) of the simulator is: poor(l)...good(5) 4.22
2. Page colors are chosen to help concentrate: poor(l)...well(5) 3.66 +
3. Navigation is: easy(l)...difficult(5) 2.11
4. During the navigation from one page to the next, you get: lost(l)...well oriented(5) 4.11
5. During the navigation from simulator to links, you get: lost(l)...well oriented(5) 4.00
6. Learning how to navigate is: easy(l)...difficult(5) 1.44
7. Reading the text is: easy(l)...difficult(5) 1.88 +
8. Text facilitates hypertext and branching: poor(l)...well(5) 3.55 +
9. Simulation outputs are: poor(l)...good(5) 4.11 +
10. Using the output for further analysis is: easy(l)...difficult(5) 2.77 +
11. To read the characters on the computer screen: hard(l)...easy(5) 4.22
12. Screen layouts were helpful: never(l)...always(5) 4.11
13. Sequence of screens: confusing(l)...clear(5) 4.00
14. Your comments related to design of screens
System Experience
1. Terrible(l)...Wonderful(5) 3.55
2. Frustrating(l)...Satisfying(5) 4.00
3. Dull(l)...Stimulating(5) 3.88
4. Difficult(l)...Easy(5) 4.11
5. Inadequate power(l)...Adequate power (5) 3.55
6. Rigid(l)...Flexible(5) 4.22
7. Your comments related to system experience:
Terminology and System Information
Please rate the following
1. Use of terminology throughout system: confusing(l)...clear(5) 4.11
2. Terminology relates well to the work you are doing: never(l)...always(5) 4.00
3. Learning the simulator: difficult(1)...easy(5) 4.00 +
4. Tasks can be performed in a straightforward manner: never(l)...always(5) 3.88
5. Your comments related to terminology and system information
System Capabilities
1. System speed: too slow(l) enough(5) 3.11
2. The system is reliable: never(l)...always(5) 3.44
3. Correcting your mistakes is: difficult(l)...easy(5) 3.66
4. Ease of operation depends on your level of experience: never(l)...always(5) 3.00
5. Your comments related to system capabilities.
Background and Technical Info
1. Have you looked at background info? Yes/No Yes 100%
2. Have you looked at demo? Yes/No Yes 100 %
3. Have you looked at technical info? Yes/No Yes 100%
4. Technical info provided as a guidance in INFO sections are confusing(l)...clear(5) 3.66 +
5. Information from INFO is easily understood: never(l)...always(5) 4.00 +
6. Information from INFO is easily applied: never(l)...always(5) 3.88
7. Information from BACKGROUND is: poor(l)...satisfactory(5) 4.44 +
8. Your comments related to Background and Technical Info
1. This simulator is a helpful learning tool: disagree(l)...strongly agree(5) 4.55
2. Your comments related to simulator







Chemical Engineering Education

sist students. Some students had difficulty uploading the soft-
ware, so extra meetings were arranged to overcome and mini-
mize any technical difficulties. The instructor also encour-
aged students to visit her outside of class and to discuss the
progress of their projects. Most students showed a high inter-
est in the course during this period because they were able to
integrate textbook topics with real-life situations.
In order to assess the educational benefits that the simula-
tor provided, a discussion session that took one lecture hour
was conducted by the instructor, students handed in individu-
ally prepared written reports at the end of two weeks, and a
survey was prepared and administered to assess the use-
fulness of the project as a means for achieving the learn-
ing goals (see Table 2).

Furthermore, to assess the simulator's design capabilities, a
questionnaire based on earlier works on software and
courseware design and evaluation'14] was developed and ad-
ministered at the end of the project. Table 3 lists the ques-
tions that were targeted to serve as a tool for refinement
of the simulator and summarizes students' responses to
the survey questions.

Students commented favorably on the project and the simu-
lator. Most of them thought they better appreciated the appli-
cation of theories that they learned in the biomedical section
of the course, that they learned the difference between a
healthy person and a diabetic person, that the project was
stimulating because it allowed them to do something other
than calculations or a report, and that it gave them an oppor-
tunity to experience a practical application of pharmacoki-
netics in a real-life problem. They felt that even though it
took them some time to start working with the simulator,
its clarity made the project interesting. Some of the stu-
dents provided feedback on typographical errors and im-
provement of the GUI.
Furthermore, students identified the simulator as a great
learning tool for open-ended real-life problems. They found
it interesting to input their own daily diet and see how their
glucose rates changed throughout the day. They also found
the exercise option interesting. Some of them were even
amazed at how MATLAB software could be turned into such a
user-friendly form and could be used in such a practical way.


The purpose of this article is to share an educational tool
based on glucose insulin interactions in the human body. A
simulation package was integrated into an Introduction to
Bioengineering course and was assessed for its efficacy
through assigning a term project that allowed students to ex-
plore glucose insulin interactions in the human body with the
aid of a simulator. Further assessment of the simulator has
been carried out via surveys for additional improvement and
refinement. These data are in the form of questionnaires com-

pleted at the end of the term project. The course is being re-
fined based on student and expert feedback, and examination
problems that address the learning objectives of the simula-
tor are being developed to better assess the simulator's effec-
tiveness as a learning tool. A web version of the software has
also been developed that will provide an opportunity to serve
and obtain feedback from other student communities.

The authors are grateful to Drs. J. Abbasian, I. Birol, V.
Perez-Luna, and C. Undey of IIT and Peter J. Reilly of Iowa
State University for their valuable suggestions. Financial sup-
port provided by NSF (EEC-0080527) is gratefully ac-
knowledged. Special thanks to the Fall 1999 and 2000
students for their feedback.

1. Erzen, F.C., G. Birol, and A. Cinar, "Simulation Studies on the Dy-
namics of Diabetes Mellitus," in proc. EEE Internat. BIBE Simp., p.
231 (2000)
2. Guyton, J.R., R.O. Foster, J.S. Soeldner, M.H. Tan, C.B. Kahn, L.
Konez, and R.E. Gleason, "A Model of Glucose-Insulin Homeostasis
in Man that Incorporates the Heterogeneous Fast Pool Theory of Pan-
creatic Insulin Release," Diabetes, 27, 1027 (1978)
3. Puckett, W.R., "Dynamic Modeling of Diabetes Mellitus," PhD the-
sis, Chemical Engineering Department, University of Wisconsin-Madi-
son, (1992)
4. Sorensen, J.T., "A Physiologic Model of Glucose Metabolism in Man
and Its Use to Design and Assess Improved Insulin Therapies for Dia-
betes," PhD thesis, Department of Chemical Engineering, Massachu-
setts Institute of Technology (1985)
5. Erzen, F.C., "Studies on Modeling Glucose Insulin Interaction in Hu-
man Body and Development of a Simulation Package," Master's the-
sis, Department of Chemical Engineering, Illinois Institute of Tech-
nology (2000)
6. Hartson, H.R., and D. Hix, "Human-Computer Interface Development:
Concepts and Systems for its Management," ACM Computing Sur-
veys, p. 5 (1989)
7. Neilsen, J., Useability Engineering, American Press Limited, London,
UK (1993)
8. Marcus, A., Graphic Design for Electronic Documents and User In-
terfaces, Addison-Wesley, Reading, MA (1992)
9. Lewis, C., D. Hair, and V. Schoenberg, "Generalization, Consistency,
and Control," in Proc. ACM CHI'89, Seattle, WA, p. 1 (1989)
10. Nomura, M., M. Shichiri, R. Kawamori, Y. Yamasaki, N. Iwama, and
H. Abe, "A Mathematical Insulin-Secretion Model and Its Validation
in Isolated Rat Pancreatic Islets Perifusion," Comput. Biomed. Res.,
11. Carson, E.R., and D.G. Cramp, "A Systems Model of Blood Glucose
Control," Inter J. Bio-Med. Comput., 7, 21 (1976)
12. Berger, M., P. Berchtold, H.J. Cuppers, H. Drost, H.K. Kley, W.A.
Muller, W. Wiegelmann, H. Zimmermann-Telschow, F.A. Gries, H.L.
Kruskemper, and H. Zimmermann, "Metabolic and Hormonal Effects
of Muscular Exercise in Juvenile Type Diabetics," Diabetologia, 13,
355 (1977)
13. Birol, G., I Birol, and A. Cinar, "Student Performance Enhancement
by Cross-Course Project Assignments: A Case Study in Bioengineer-
ing and Process Modeling," Chem. Eng. Ed., 35(2), 128 (2001)
14. Gallagher, T., "A Proposed Software Evaluation Form or Toll and Its
Use in the Evaluation of First Verbs from Laureate," Software and
Courseware Design and Evaluation, (1999) found at> I

Fall 2003

M &^ laboratory

Simulation and Experiment in




Villanova University Villanova, PA 19085-1681

here are several advantages to integrating classroom
and laboratory exposure. For many students, under-
standing concepts taught in the classroom improves
significantly when they have the opportunity to gain hands-
on experience in a laboratory. A laboratory exercise also pro-
vides an opportunity to apply the theory they have learned in
the classroom to an actual engineering problem. Finally, com-
paring experimental data and dynamic simulation results is
an effective way to reinforce process dynamics education
in a laboratory exercise. The wider incorporation of pro-
cess dynamics into the curriculum is considered to be a
key component in process control education of chemical
engineering students.t1M
There are a number of simulation-based chemical process
dynamics experiments presented in the engineering educa-
tion literature. They range from modules incorporated into a
commercial process control computer system,t2] case studies
illustrating various process control concepts programmed
using MATLAB/SIMULINK,'351 and workshops based on
real-time simulation of industrial unit operations.161 Although
there are benefits of simulation-based experiments, a major
disadvantage is the lack of an actual physical process that the
students can watch, hear, and touch while it is operating.
Understanding the dynamic behavior of a process is greatly
enhanced by observing the physical process operation. Visu-
alization provides a significant benefit to many students as
they attempt to apply the theoretical concepts taught in the
classroom.17,81 This aspect was one of the main motivations
for developing the experience documented in this work.
A review of the equipment-based chemical process dynam-
ics experiments presented in the engineering education lit-

erature reveals a wide range of complexity in the processes
considered. They range from relatively simple liquid-levelt91
and stirred-tankt10l systems, multiple tank systems,[1" quite
complex reaction[121 and distillationt"31 systems, and combi-
nations of simple, more complex, and simulated systems.1141
Because this experience is intended to be an introductory
exposure to process dynamics, simulation, and control, us-
ing an easily modeled, simple, physical process that incorpo-
rates the introductory concepts from the process control and
simulation course is appropriate. For this reason, a single-
tank liquid-level system was chosen.
Feedback control is performed using a proportional-only
controller. Proportional control provides two benefits for this
introductory experience. The first is that a proportional con-
troller is easily simulated. The additional complexity required
in the simulation of integral action in the controller provides
little, if any, benefit to the understanding of process dynam-
ics and dynamic simulation in an introductory experience.
The second benefit is that proportional control results in
steady-state offset of the tank level. This concept is often dif-
ficult for some students to initially grasp in the classroom.
The ability to observe this phenomenon on a real physical
system can be very helpful for these students.

Kenneth Muske is Associate Professor of Chemical Engineering at
Villanova University, where he has taught since 1997. He received his
BSChE and MS from Northwestern (1980) and his PhD from The Univer-
sity of Texas (1990), all in chemical engineering. Prior to teaching at
Villanova, he was a technical staff member at Los Alamos National Labo-
ratory and worked as a process control consultant for Setpoint, Inc. His
research and teaching interests are in the areas of process modeling,
control, and optimization.

Copyright ChE Division of ASEE 2003

Chemical Engineering Education

For many students, understanding concepts taught in the classroom improves significantly
when they have the opportunity to gain hands-on experience in a laboratory.
A laboratory exercise also provides an opportunity to apply the
theory they have learned in the classroom to an
actual engineering problem.

EQUIPMENT Manual Inlet
The experiment is carried
out using a 50-gallon gravity-
drained tank equipped with F
liquid level and outlet flow in
sensors. Liquid level is con-
trolled by a valve on the inlet
water pipe. There is an addi- h
tional inlet water pipe with a
manual valve. The outlet flow
rate can be adjusted by a
manual valve on the outlet pipe (
of the tank. A steam heater is
connected to the tank but not
used in this experiment. The Figure 1. Expe
tank is 2.5 feet in height and
1.8 feet in diameter. The tank
system is shown in Figure 1.
Liquid level control is provided by a single-loop electronic
controller. There is also a distributed computer control sys-
tem connected to the tank. Because we believe there is value
in exposing the students to both the single-loop electronic
and computer control systems, we have the ability to switch
between the two systems on this tank. The single-loop con-
troller is used for this introductory experiment, while the com-
puter control system is used for temperature-control experi-
ments and a model predictive control experiment in the se-
nior laboratory course.1151

The exercise comprises two three-hour laboratory sessions.
During the first session, the student groups become familiar
with the tank system and perform the experimental work. In
the second session, they develop their dynamic simulation
model, compare their simulated results with those obtained
experimentally, and document their findings in a short memo
report to the instructor.
The laboratory exercise begins with the tank operating at
steady state under proportional-only feedback control with a
water level setpoint of 50%. The instructor reviews the physi-
cal operation of the tank, goes over each component com-
prising the feedback control loop, and leads a short discus-
sion concerning the options to remove steady-state offset with
the student group. The students are then instructed to adjust


the controller bias to remove
the steady-state offset in the
Control Valve tank level. They may either put
the controller in manual and
adjust the control valve posi-
tion or adjust the bias directly
@P to eliminate the offset. The
S value of this exercise is gain-
A ing an appreciation for the re-
sponse time of a real physical
LT system. The students are
prompted to estimate both the
Manual Outlet time constant of the system,
Valve which is on the order of five
Fout minutes, and the open-loop re-
sponse time of the tank in or-
tal tank system. der to determine how long it
should take for the tank level
to reach steady state after a
change to the inlet water valve position is made. Although
process simulators provide valuable training experience for
the students, a major drawback is that those experiences are
in "simulation" time. The first part of this laboratory exer-
cise demonstrates that real process dynamics are not on this
same simulation time scale.
After the students have adjusted the bias to eliminate the
steady-state offset in the tank level, the system is returned to
closed-loop control and they are allowed to choose one dis-
turbance from a list in the laboratory instructions. This list
contains the following disturbances:
Simultaneously dump two small buckets of water into
the tank
Dump one large bucket of water into the tank
Change the inlet flow rate by opening the manual
disturbance flow valve
Change the outlet flow rate by opening or closing the
manual outlet valve
Change the level setpoint
Change one of the level controller tuning parameters
where the two small buckets are each two gallons, resulting
in an impulse disturbance that is approximately 20% of the
liquid volume; the large bucket is 25 gallons, resulting in an
impulse disturbance when full that is about the same as the
liquid volume; and the disturbance flow results in a step

Fall 2003

disturbance that is about the same as the initial steady-
state inlet flow rate.
From the instructor's perspective, it is desirable to have
as much variation in the selected disturbances between
groups as possible to make the students' semester-end oral
reports on this experiment more interesting. In practice,
other than discouraging the one large bucket, prompting by
the instructor in order to provide this variation has seldom
been necessary.
Prior to implementing their chosen disturbance, the tasks
of time keeper, data logger, and disturbance initiator are
distributed by the group members among themselves. Their
selected disturbance is then implemented on the tank sys-
tem under closed-loop level control. The initial data point
is collected after the disturbance has been completed. In
the case of the buckets, this point is the time when all of the
water has been emptied into the tank. Because two students
(and sometimes the instructor) are required to empty the
bucket contents into the tank, the time-keeping and data-
logging tasks are performed by one student at the begin-
ning of the experiment. For the other disturbances, the ini-
tial data point is taken immediately after the valve position
or controller tuning parameter has been changed. Data is
collected at intervals of ten to twenty seconds until the tank
level reaches steady state. The experimental phase of this
exercise is typically completed well within the three-hour
laboratory period.

The second phase of this exercise involves the dynamic
simulation of the closed-loop tank system with the distur-
bance chosen by the group. This phase is carried out during
the laboratory period immediately following the experimen-
tal session. Process simulation begins with an unsteady-
state material balance over the tank. Assuming a constant
cross-sectional area of the tank, Ac, and the same constant
density for all water streams, a macroscopic mass balance
results in
Ac = Fin Fout (
where h is the height of water in the tank, Fin is the inlet
volumetric flow rate, and Fo.t is the outlet volumetric flow
The inlet volumetric flow rate of water is determined by
the position of the control valve. Although this control valve
is linear, the inlet flow rate is not a linear function of valve
position over the entire valve position range due to varia-
tion in the water supply pressure as the valve position
changes. The students are given a calibration curve, shown
in Figure 2, that is used to relate the inlet flow rate to the
control valve position. Over the linear operating range of
the valve, the following correlation can be used to deter-

mine the inlet flow rate

Fin =0.171 (Vp)-1.03


where Fn is the flow rate in units of gpm and V is the control
valve position in units of % open. If the disturbance flow was
selected, there is a second constant inlet flow rate that must be
added to this relationship.
The outlet volumetric flow rate is assumed to be propor-
tional to the square root of the pressure drop across the manual
outlet valve due to the static head of fluid in the tank

Fout=Kvh+1 (3)

where Fout is the flow rate in units of gpm, Kv is the proportion-
ality constant, h is the height of the water in the tank in units of
feet, and the bottom of the tank is 19 inches above the outlet
valve. The proportionality constant Kv is determined from the
measured outlet flow rate and water height when the tank level
is at steady state.
The control valve position is determined by the level con-
troller on the tank. For the proportional-only controller, the
valve position is determined from the controller equation

Vp =B+Kc(Sp L) (4)

where V is the valve position in units of % open, B is the
controller bias in units of % open, Kc is the proportional gain
in units of % open/% level, S is the water level setpoint in
units of % level, and L is the level of the water in the tank in
units of % level. In practice, the controller gain is kept at a
value around 1 %/% to prevent the control valve position from
moving out of its linear operating range during the transient
response due to the disturbance.
Simulation of the process is carried out by numerical solu-
tion of Eqs. (1) through (4). Before they can be solved, how-
ever, the units must be made consistent throughout all of the

Figure 2. Inlet water control valve calibration curve.

Chemical Engineering Education

relationships. The controlled variable for the controller is con-
figured to be in units of % while the tank dimensions are given
to the students in units of feet and the flow rate calibrations are
given in units of gpm. This variation in the units given to the
students is intentional. Numerical solution is typically carried
out by the student groups using MathCad, which is used by the
department in the introductory material balance and the nu-
merical methods prerequisite courses, although they are free
to use any of the other mathematical software packages such
as MATLAB, EXCEL, and Maple that are available on the
engineering college server.

Example experimental and simulation results are shown in
Figures 3 and 4. Figure 3 presents the results for the large bucket
impulse disturbance. In this example, the large bucket was only
about half full. Figure 4 presents the results for a reduction in
the outlet flow rate from closing the manual outlet valve. In

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time (minutes)

Figure 3. Experimental and simulated closed-loop tank
level for the large bucket impulse disturbance.

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Time (minutes)

Figure 4. Experimental and simulated closed-loop tank
level for a change in the outlet valve position.

both cases, the experimental and simulated responses
are very similar. These results are typical for most of
the student lab groups.
In addition to presenting their experimental and simula-
tion results, the student groups are asked to discuss the
sources of error in this experiment in their group memo
report. Examination of the experimental and simulated dy-
namic responses reveals that the simulation leads the ex-
perimental response. Because there are dynamics associ-
ated with the level sensor and control valve that are not
included in the simulation model, this result would not be
unexpected. The effects of valve friction, sensor noise, and
the precision of the liquid level value displayed by the con-
troller can also contribute to error as well as the assump-
tion of a perfect square root relationship and a constant Kv
value for the outlet flow rate that may not be valid over the
liquid level ranges encountered in the experiments. Experi-
mental error in the timing of the collected level data samples
is also present. Almost every student group mentions the
valve, sensor noise, and sampling error as sources of error
in their report. Some groups also mention the outlet flow
relationship used in the simulation model. Few groups dis-
cuss the dynamic effect of the valve and sensor.

As part of the student evaluation of the process simula-
tion and control course, a number of supplemental ques-
tions concerning the value of the text and controller simu-
lation software used in the course, the laboratory experi-
ence documented here, and the preparation received in the
required prerequisite courses are included. The evaluation
scores ranged from 5=Very Effective to l=Very Ineffec-
tive. The average scores from the last four years are: pre-
sentation and explanation of concepts in the textbook,
3.04; use of CStation for class examples, 3.25; use of
CStation for homework problems, 2.96; process control
experiment in Lab II, 4.02.
CStation1161 is the process control simulation software
package used in the course, Essentials of Process ControltI7'
was the course text at the time of these evaluations, and the
process control experiment in Lab II is the experience docu-
mented in this work.
The average score given by the students for this labora-
tory experience is considerably higher than for the text and
process control simulation package and is essentially the
same as the average score of 4.10 for the value of the pro-
cess control and simulation course over the same period. It
should be noted that a number of students have provided
somewhat negative comments concerning the length of the
loop tuning homework assignments requiring the use of
CStation. These feelings may have had some influence on
the CStation scores. It should also be noted that only one
Continued on page 315.

Fall 2003

learning in industry

This column provides examples of cases in which students have gained knowledge, insight, and experience in the
practice of chemical engineering while in an industrial setting. Summer internships and co-op assignments typify such
experiences; however, reports of more unusual cases are also welcome. Description of the analytical tools used and the
skills developed during the project should be emphasized. These examples should stimulate innovative approaches to
bring real world tools and experiences back to campus for integration into the curriculum. Please submit manuscripts to
Professor W. J. Koros, Chemical Engineering Department, Georgia Institute of Technology, Atlanta, GA 30332-0100.


Purdue University West Lafayette, IN 47907-1283

Chemical engineering departments should stay in con-
tact with industry so their students can be taught
material that is useful after graduation and so research
will be relevant to the needs of industry. Unfortunately, very
few of the professors being hired from graduate schools or
after post-doctoral experiences have significant industrial ex-
perience. In this article we will discuss one way to maintain
this contact-by bringing back experienced engineers from
industry as post-early retirement Industrial Professors or En-
gineers in Residence. These individuals should be integrated
into all aspects of teaching and research within the depart-
ment. This article is based on the experiences of one of the
authors (GB) who joined the School of Chemical Engineer-
ing at Purdue University during the spring semester of 1998
and who has served in the Industrial Professor role since that
time. The results are generalized for other "wannabe profes-
sors" from industry who are contemplating a career shift, as
well as other chemical engineering departments considering
initiating such an infusion of industrial talent.

Dr. Gary Blau had a successful industrial career with the
Dow Chemical Company, primarily on the technical ladder.
A Canadian, Gary graduated in 1964 from the University of
Waterloo with a BASc in chemical engineering and went on
to get his PhD at Stanford University in 1968. He then went
to work with Dow Chemical. In 1991, he accepted an offer to
work with DowElanco (now DowAgrociences), a joint ven-
ture between the agrochemical business interests of Dow
Chemical and Eli Lilly.
At DowElanco, his final assignment was leading a group

of six engineers in the development and application of math-
ematical modeling tools to optimize work processes within
the company. He remained professionally active by writing
over fifty journal articles, coauthoring a book on mathemati-
cal modeling, editing another on Environmental Exposure to
Chemicals, organizing meetings, and serving in various lead-
ership roles in the Computing and Systems Technology
(CAST) division of AIChE. In 1998 he won the Computing
Practice Award from AIChE.
Why, then, did a highly successful, mid-career engineer
decide to take early retirement for a lower-paying, temporary
position? The answer embraces timing, location, opportunity
to develop a business, fulfillment of a dream, and idealism.
Gary's sponsor had just retired, and the resultant lack of man-
agement support made fighting for proper recognition for his
group increasingly stressful. In other words, the job was
not much fun any more. After thirty years of experience,
he was eligible for early retirement-so he began to look
for new challenges.
Gary, like many PhD engineers in industry, was a "closet
academic." Going to academe would fulfill the academic
dream that all Stanford graduate students harbor. A model for
this ambition was Dr. Park M. Reilly, who had worked in

Gary Blau received his BASc from the University of Waterloo and his MS
and PhD from Stanford University, all in chemical engineering. Following a
successful career with Dow and DowAgrociences, he joined the faculty at
Purdue University as a Visiting Industrial Professor. His research is in risk
analysis and decision-making.
Phillip Wankat received his BSChE from Purdue, his PhD from Princeton,
and a MSED from Purdue. He is the Clifton L. Lovell Distinguished Profes-
sor of Chemical Engineering and the Head of Interdisciplinary Engineering
at Purdue University. His research is in separation processes.
Copyright ChE Division of ASEE 2003

Chemical Engineering Education

His primary motivation, however, was his desire to be the faculty "industrial guy" and to
share his real-world experiences with future engineers. He felt he could prepare
the chemical engineering students for what industry is really like.

industry for 25 years before pursuing a successful academic
career at Waterloo. Gary also had a desire to do independent
consulting in the modeling area, but his networks and paper
trail were too thin to support this activity.
His primary motivation, however, was his desire to be the
faculty "industrial guy" and to share his real-world experi-
ences with future engineers. He felt he could prepare the
chemical engineering students for what industry is really like.
He planned to develop "ill-defined, open-ended" problems
and to teach the students how to use their engineering skills
to solve them. He would train them in the proper use of sta-
tistical modeling, quality control, and risk management tech-
niques so they could have an immediate impact in industry.
He would show by examples and "war stories" that "soft
skills" are really important in industry. Since Gary had de-
veloped and taught process optimization short courses to lit-
erally hundreds of Dow engineers and chemists around the
globe, he felt his teaching skills were sufficiently honed to
motivate students.

While at Dow Agrociences, Gary became involved in sup-
ply-chain optimization issues, resulting in joint research col-
laboration with Professor Joe Pekny at Purdue. Gary helped
support some of Joe's research, worked with him on organiz-
ing the Foundations in Computer-Aided Process Operations
conference, and presented some lectures. He also knew Pro-
fessors Rex Reklaitis and Nick Delgass (Head and Associate
Head of Chemical Engineering, respectively) from their days
as graduate students at Stanford. Since Purdue was a short
50-minute commute on the interstate, Gary decided to accept
Rex's offer to be a Visiting Industrial Professor.
The courses Gary would teach and his time commitment to
Purdue were finalized during initial discussion with Rex. Gary
felt that 2/3 time during the academic year, with summers
off, would provide time for him to pursue both consulting
and leisure activities. He identified a required core course he
was well-qualified to teach and a new elective course he would
like to develop. He would also become involved in the sys-
tems research program and help jointly direct the research
program of some of Joe Pekny's graduate students.
Once a mutually agreeable salary was determined, Gary
became a Purdue professor. Since tenure was not an impor-
tant consideration, the position was not on the tenure track-
making it easier for the university to develop the position.
The process by which the new industrial professor and the
institution found each other is probably transferable, so a list

of the steps that took place is appropriate. First, the potential
professor should generate a record of accomplishment in in-
dustry that is respected by the academic community. If re-
search is to be part of the academic position, this industrial
record must include an adequate publication record. There
should be personal contact (networking) with some of the
professors in the department. The applicant should become
involved with the research program-sponsoring research is
particularly helpful. He/she needs to give some lectures to
undergraduate classes and, if appropriate, to the graduate re-
search seminar.
After determining core undergraduate courses and one or
more electives that he/she is qualified for and wants to teach,
the applicant can make a proposal to the department head
concerning time commitment, course assignments, research
involvement, salary, etc. Since the question of tenure can be
a major stumbling block, formal ties are easier to forge if a
yearly contract is acceptable.
Of course, "making it happen" is not the sole responsibil-
ity of the industrial practitioner. The academic institution can
also be proactive by identifying individuals who might be
good candidates for the industrial professor program. A de-
partmental representative can talk with the prospect about
the program and determine if there is any interest.


Like many other senior engineers, Gary frequently taught
short courses, and since there had never been complaints, he
believed he was prepared for teaching in a university. Al-
though he hadn't graded in these short courses, he expected
that the grading scheme would be unchanged from what he
had observed as a student or that he would be given clear
instructions on how to grade. He expected the students would
be courteous and neatly groomed. After all, he reasoned, they
were in school because they wanted to learn and thus should
be highly motivated to learn from someone who could show
them the value of their education.
Gary assumed that the students would have the same strong
technical background in mathematics and science as he and
his classmates had thirty-five years ago. He expected to share
his knowledge and thus prepare the students for careers in
industry. He looked forward to building personal relation-
ships with some of the students, and he viewed teaching as
the most important role he would fill, with research being a
secondary consideration. Since he was experienced, teach-
ing just one class (even if it was a large class) three times a
week would be a "piece of cake," allowing him plenty of

Fall 2003

time for other pursuits. Finally, it never occurred to him that
the students might challenge him.
Those were the expectations. The reality had a different
face. A book by Peter Sacks[" gives an accurate portrayal of
the trials of returning from the working world to become a
professor, and although he had read Sacks' book, Gary thought
that engineering students would be different. That did not
turn out to be true. Unfortunately, many students believe that
if they do poorly, it's the professor's fault, not theirs. As a
new professor, Gary unwittingly played into this rationaliza-
tion for poor performance by giving tests that were difficult
and too long. With test averages in the 30s and 40s, the stu-
dents blamed Gary for their lack of preparation.
Gary also learned that many students seemed to be more
interested in the final course grade than they were in learning
the material. Since grading procedures were never well de-
fined for him, he tried to enforce what he thought were rea-
sonable grading standards, but some students argued con-
stantly with him in an effort to raise their grades.
Teaching evaluations were a new concept for Gary since
they had not been in use when he was a student, and he found
that some students used them to "dump" on a professor. He
worried that firm grading would be penalized when students
filled out the mandatory teaching evaluations. Also, being
connected on-line to the class allowed students to provide
immediate, often unflattering, feedback on Gary's classroom
performance. Because of his position in industry, Gary had
always been insulated from direct criticism from subordinates,
and he found the harsh criticism from 19- and 20-year-olds
hard to accept.
Gary found that today's students are different from those
of thirty-five years ago in other ways. For one thing, they
wanted to be entertained. Fortunately, he found that his ex-
perience was useful in this regard since the students enjoyed
listening to his industrial stories. He also found that the stu-
dents' work ethic was low and that they were less comfort-
able with ambiguity. When he introduced his industrially in-
spired "open-ended, ill-defined" problems, many students
were frustrated with them, wanting instead problems that were
crystal clear and did not require assumptions.
Talking during class was also rampant and required special
control techniques that had been unnecessary in his small classes
in industry. To his surprise, some students were actually rude
and disrespectful. Some staff members also let Gary know that
as a visiting professor, he was at the bottom of the heap.
Gary found that teaching was much more difficult in aca-
deme than in industry and that it was difficult to learn
everyone's name, let alone build a personal relationship with
them. Students in industry were more polite, more motivated,
and better prepared-they wanted to be in the educational
session, and any errors in presentation were quietly corrected.
But at the university the students were vocal about errors and

were unwilling to think their way through a derivation and
develop all the details by themselves.
While students in industry never complained about too
much material (apparently they sorted out for themselves what
was useful and what wasn't), university students were over-
whelmed by the amount of material, and trying to introduce
new methods such as new software only made matters worse.
The resistance to change was much greater at the university
than it was in industry.
Testing was a challenge that had never arisen in industry.
Some students would become angry if they were unable to
finish a test, so creating one that was not too long became an
enormous challenge. Gary found that students often seemed
incapable of doing very simple problems, particularly if there
were multiple steps. Grading the exams took much more time
than he had anticipated, and Gary soon came to see the beauty
of short-answer problems. Student cheating led to Gary's
general distrust of students, and he learned to place students
in alternate seats and to roam the classroom during exams.
Gary also found that teaching was more time consuming
than he expected. For the initial course offering, each 50-
minute lecture took over four hours to prepare, and the sec-
ond offering took almost as long because of the many revi-
sions that had to be made. An unexpected responsibility was
supervising two graduate TAs and eight undergraduate
TAs-Gary felt that he had traded six highly motivated
self-starters in industry for ten students who were just
"doing a job" for the money.
Since it was essential to be present for every class, teach-
ing meant a much less flexible schedule than Gary had had in
industry. It came as a surprise that in some ways, time pres-
sures were greater in academe than they had been in industry.
Finally, Gary found that at a research university, teaching
was not the top priority. It had to be adequate, but research
(specifically, research funding) was most important. Most pro-
fessors wanted to talk about research, not teaching.

The most important step to surviving is adaptation: reflect-
ing on what happens, sorting through the criticism, talking to
students and professors alike, revising, and doing it again.
Talking to students revealed that their pressures are much
different than they used to be-they are busier and many of
them work part-time. Fewer have strong backgrounds in
mathematics, chemistry, physics, or other areas of engineer-
ing, and fewer still are dedicated to becoming chemical engi-
neers. In talking to the students and professors, Gary also
found that he had not been singled out as a new professor for
criticism-it was more of an equal-opportunity endeavor, and
more important, there were methods to prevent it. There were
also approaches that would improve the course while at the
same time reduce the time required to conduct it.

Chemical Engineering Education

Initially, Gary's self-confidence was so badly shaken that
he asked the associate head to team him with one of the more
successful teachers in the department. Although this presented
a scheduling nightmare, Gary was allowed to share his teach-
ing responsibilities with Joe Pekny, a younger but more ex-
perienced professor. Gary was a good student and absorbed
the lessons rapidly.
Discussions with students and experienced professors con-
vinced Gary that a course outline with firm
dates for exams and major assignments was .-
absolutely necessary since students could not
adjust their complicated schedules to accom- -
modate last-minute changes. He was also
advised to reduce the amount of material cov-
ered, which was easy initially but proved to
be more difficult as additional cuts were
needed. The process is ongoing.
Gary tried several presentation styles. Stu-
dents complained that his handwriting was
unreadable, so he tried class notes-but in
addition to the construction being time con-
suming, they were boring and students either
did not attend or went to sleep. He then tried
notes with spaces for one-word answers, but
the students viewed those as childish. He -
found that notes with large spaces for ex- --
amples or derivations worked best, and they
were easy to construct. He already had com-
plete notes for lectures and created student
notes by simply removing parts from them.
This procedure allowed for more time for questions and
stories, kept the students interested, and reduced the num-
ber of board errors.
As he became more experienced, Gary was able to write
exams and projects that would challenge, but not overwhelm,
the students. He scheduled exams at times that gave the stu-
dents extra time, and he learned to recognize those problems
that would be difficult to grade and used them only when it
was educationally necessary. With more time available to him,
Gary was able to get to know the TAs and the students better
and found they were more like the students he remembered
than he initially thought.
By taking a section of the computer laboratory in the sta-
tistics class, Gary learned what was giving the students trouble
and was able to improve descriptions for the laboratory exer-
cises. He also got to know the students in this small section
very well and was able to recruit competent undergraduate
TAs for the next offering of the course.
Today, Gary is teaching a large sophomore-junior class by
himself, along with his cadre of selected graduate and under-
graduate TAs. Although his grade distribution has not
changed, his teaching evaluations have risen significantly and


Fall 2003

he has been able to put to rest his worry that firm grading
would by punished by low evaluations. Both he and the stu-
dents now agree that it is fun to go to class.
Gary has learned to be an "edutainer," and the students are
fascinated by the "war stories" he has built into his projects
and exams. Since he better understands the pressures facing
students, he is now better able to counsel them and give ad-
vice. Many of them realize that a letter of recommendation
from someone with thirty years of industrial
experience is an advantage and they make the
7 effort to establish the one--one relationship
Gary was looking for in the beginning. Al-
though the large undergraduate course was
Gary's major challenge, he built such enthu-
siasm for his Risk Management elective course
that enrollment had to be capped.

Gary's transition to the research environ-
ment was smoother than it was to the teaching
component of his new career, although expec-
tations and reality were, again, frequently in
conflict. Professors were always willing to talk
about their research interests provided Gary
could track them down, but they were not ac-
tively looking for an "industrial perspective."
Gary soon learned that researchers were driven
Sby the primeval need for survival-they
needed funding for everything: graduate stu-
dents, laboratory facilities, computers, release time, tele-
phones, and even copying services. They needed to write pro-
posals, give presentations, serve on committees, teach courses,
supervise graduate students, etc. If they were successful, up-
per-level management in the university often rewarded them
with administrative responsibilities.
Gary had no desire to compete in this environment. He sim-
ply wanted to work with some of the graduate students to
pursue some ideas he felt compelled to develop. Professor
Joe Pekny helped by involving Gary with his research group.
In addition, he managed to bring in sufficient annual funding
for a graduate student or two. This was ideal. He could con-
duct research, which resulted in publications and presenta-
tions while providing the necessary new material for his up-
per-level course in risk management. He also accepted a po-
sition as director for an industrial consortium that would chan-
nel funding into the department's systems research efforts.
Gary developed an increasing appreciation for the stimu-
lating nature of the academic environment. Working with pro-
fessors from other universities and engineers from industry
was stimulating, and he was able to develop lasting relation-
ships with some of the students (easier with graduate stu-
dents, however, than with undergraduates).

Gary had success in bringing an industrial perspective to
the department and found it extremely rewarding to see former
students who said they used his "stuff' in their jobs and found
it valuable. Also, having the summer off for consulting and/
or "down time" at the lake has been a bonus no industry is
willing or able to match.
The challenges of academe are such that Gary, who was
slightly bored in industry, now looks forward to every day.
He has found that doing research and working with graduate
students is exciting and has led to even more ideas for re-
search. Since there is no textbook for the graduate-level course
on risk analysis, Gary would like to write one...and there is
always the challenge of getting everything right in teaching
an undergraduate class!

There are a number of advantages for both the department
and the students in having industrial professors on staff. For
one thing, it is useful to maintain contact with industry to
ensure that curriculum content is relevant. Other professional
schools, such as medical and law schools, regularly have prac-
titioners teach, and engineering colleges could benefit by
following that lead and developing positions for professors
with industrial experience. Industrial professors can help with
the teaching load, particularly in design classes (although this
was not the case with Gary). They tend to be more commit-
ted to teaching'21 and are more likely to assign open-ended
problems and projects that include teamwork and writing.
Their presence can also help the department prepare students
for industrial careers and thus satisfy ABET criterion 4. Pro-
fessors who had active industrial research programs and con-
tacts can help the department bring in more research con-
tracts and ensure that the research is relevant.
The process can be improved, however. First, a critical mass
of professors with extensive industrial experience is needed.
One on staff is not enough-it is more appropriate if 20-30%
of the faculty has extensive industrial experience.
Although tenure is probably not an issue for most return-
ees, planning for the next year is. A rolling contract that would
allow the visiting professor to know at least a semester in
advance whether or not he/she would still have a job would
be helpful. (Although Gary's position was originally viewed
as a visiting position, the "visit" is now in its fifth year.) A
title that implies a longer-term commitment (but without ten-
ure) would be appropriate. In addition, a formal program
would ease the red tape involved when the next engineer-in-
residence is hired.
The department should ease the teaching transition of in-
dustrial professors and realize that they need to be taught how
to teach-that they want to be taught how to teach. If there
are no local workshops on teaching, the returnee should be
encouraged to go to a national workshop such as the ASEE
NETI (National Effective Teaching Workshop) at the annual

meeting of ASEE. The department should provide a teaching
mentor to discuss teaching with the returnee on a regular ba-
sis where both general pedagogical principles and the
university's specific rules can be explored. This mentor should
invite the returnee to visit his/her classes and volunteer to
visit the new professor's classes. The mentor can be particu-
larly helpful with grading and can provide resources such as
a guide to teaching at that particular university. The returnee
should be made aware of pedagogical journals such as Chemi-
cal Engineering Education, ASEE PRISM, the Journal of En-
gineering Education and appropriate books such as Teaching
Engineering (available free at https://Engineering.Purdue. edu/
ChE/Newsand_Events/publications/teaching _engineering).
These suggestions are also true for new assistant profes-
sors, most of whom want or are at least willing to be taught
how to teach. Such teaching resources should also be made
available to the experienced professors in the department who
may want to "tune up" their teaching or learn some new tricks.
If the industrial professor is to be involved in research, the
department should ease that transition. The returnee will usu-
ally be experienced in research, but the differences between
academic and industrial approaches to research are likely to
be surprising. Providing a research mentor who is in the same
research area, but who is not the same person as the teaching
mentor, is advisable. The mentor should carefully explain the
need, and the mechanisms, for funding.
The returnee should collaborate with experienced profes-
sors on research and on proposals and should be encouraged
to write proposals on his/her own-but the research mentor
should review the proposal before it is sent out. The mentor
should discuss the role of graduate students and undergradu-
ates in research. The returnee should be told of the resources
available at the university and of the formal and informal
procedures for sharing those resources.
The department should prepare its faculty and staff for the
arrival of the new professor, and the faculty should agree be-
forehand that hiring an industrial professor is a good idea. If
the professors treat the individual with respect, the staff will
also. The returnee's office should be ready from the start and
should be equivalent to the offices of other professors-
"ready" means the office is clean, has furniture, the com-
puter is hooked up to the network, the telephone is working,
the nametag is on the door, secretarial assignments have been
made, etc. Every new professor, whether industrial or not,
should be introduced to everyone else in the department and
should be invited to faculty meetings and other gatherings.
All new faculty, not just those from industry, can benefit
from formal courses or workshops on pedagogy and from
informal discussions with experienced teachers. They need
mentoring in both research and teaching. Also, some indus-
trial perspectives could be useful for universities-resource
planning, for one, is a major push in industry but does not
receive the same effort in academe. Risk analysis has proved

Chemical Engineering Education

valuable in industry and could be useful in academe (e.g.,
which centers should the university compete for?). Asking
questions is valuable-how long do professors need for dif-
ferent tasks, what can be done to improve the process, if stu-
dents aren't the customers, who is, etc. Industrial faculty
members can help ask the questions and help search for an-

Many chemical engineering departments have been criti-
cized for a lack of industrial experience in their faculty. One

Process Control Laboratory Experience
Continued from page 309.

student provided written comments about the text and that
no student provided written comments about the laboratory
experience over the four-year period. Because there are no
formal course evaluations for laboratory courses, student re-
sponse data from the second-semester junior laboratory course
concerning the laboratory course and this experiment is not
available. Qualitative assessment of this experience based on
comments received from the students during and after the
experiment indicate that this experience has been generally
well received by the students.

An introductory laboratory experience in process dynam-
ics and control that is conducted concurrently with the pro-
cess simulation and control course at Villanova University
has been presented here. The experience is intended to rein-
force the introductory concepts of dynamic simulation and
feedback control presented in the classroom by using a simple
liquid-level process. Based on quantitative and qualitative
student responses in the laboratory and process simula-
tion and control courses, the students found the experi-
ence a valuable addition to their process simulation and
control education.

A curriculum revision grant to the Villanova University
Chemical Engineering Department from Air Products and
Chemical Co. that supported development of this laboratory
experience, and the contributions of Professor Robert
Sweeney in the design and construction of the experimental
system are gratefully acknowledged. I would also like to thank
the Villanova University chemical engineering students over
the past four years for their active participation in the con-
tinuing development and improvement of this experience, and
Ami Badami and Jenny Papatolis of the class of 2003 for
supplying their respective group's experimental and simula-
tion results that are presented in this paper.

approach to partially solving this problem is to hire early re-
tirees from industry. As shown by the experiences related in
this paper, these returning professors will probably experi-
ence some degree of cultural shock. Their transition to be-
coming productive contributors can be eased by providing
both formal training in teaching and informal mentoring.

1. Sacks, P., Generation X Goes to College, Open Court, Chicago, IL
2. Fairweather, J., and K. Paulson, "Industrial Experience: Its Role in
Faculty Commitment to Teaching," J. Eng. Ed., 85, 209 (1996) O

1. Luyben, W., "The Integration of Research and Undergraduate Process
Control Education," AIChE Annual Meeting, paper 204d (1999)
2. Rivera, D., K. Jun, V. Sater, and M. Shetty, "Teaching Process Dy-
namics and Control Using an Industrial-Scale Real-Time Computing
Environment," Comput. Appl. Eng. Ed., 4(3), 191 (1996)
3. Bequette, B.W., "Computer Applications in Process Dynamics and
Control Courses," Comput. Appl. Eng. Ed., 6(3), 193 (1998)
4. Bequette, B.W., K. Schott, V. Prasad, V. Natarajan, and R. Rao, "Case
Study Projects in an Undergraduate Process Control Course, Chem.
Eg. Ed., 32(3), 214 (1998)
5. Doyle, F., E. Gatzke, and R. Parker, "Practical Case Studies for Un-
dergraduate Process Dynamics and Control Using Process Control
Modules," Comput. Appl. Eng. Ed., 6(3), 181 (1998)
6. Young, B., D. Mahoney, and W. Svrcek, "Real-Time Computer Simu-
lation Workshops for the Process Control Education of Undergraduate
Chemical Engineering," Comput. Appl. Eng. Ed., 9(1), 57 (2001)
7. Felder, R., D. Woods, J. Stice, and A. Rugarcia, "The Future of Engi-
neering Education: Part 2. Teaching Methods that Work," Chem. Eng.
Ed., 34(1), 26 (2000)
8. Vivaldo-Lima, E., "Student Motivation, Attitude, and Approach to
Learning," Chem. Eng. Ed., 35(4), 62 (2001)
9. Palanki, S., and V. Sampath, "A Simple Process Dynamics Experi-
ment," Chem. Eng. Ed., 31(1), 64 (1997)
10. Romagnoli, J., A. Palazoglu, and S. Whitaker, "Dynamics of a Stirred-
Tank Heater," Chem. Eng. Ed., 35(1), 46 (2001)
11. Johansson, K.H., and J.L. Rocha Nunes, "A Multivariable Laboratory
Process With an Adjustable Zero," in Proceedings of the 1998Ameri-
can Control Conference, 2045 (1998)
12. Luyben, W., "A Feed-Effluent Heat Exchanger/Reactor Dynamic Con-
trol Laboratory Experiment," Chem. Eng. Ed., 34(1), 56 (2000)
13. Pintar, A., D. Caspary, T. Co, E. Fisher, and N. Kim, "Process Simula-
tion and Control Center: An Automated Pilot Plant Laboratory,"
Comput. Appl. Eng. Ed., 6(3), 145 (1998)
14. Joseph, B., C.-M. Ying, and D. Srinivasagupta, "A Laboratory to
Supplement Courses in Process Control," Chem. Eng. Ed., 36(1), 20
15. Muske, K., "A Model-Based Control Laboratory Experiment," in Pro-
ceedings of the 2003 American Control Conference, 700 (2003)
16. Cooper, D., and D. Dougherty, "Enhancing Process Control Educa-
tion with the Control Station Training Simulator, Comput. Appl. Eng.
Ed., 7(4), 203 (1999)
17. Luyben, W., and M. Luyben, Essentials of Process Control, McGraw-
Hill, New York, NY (1996) 0

Fall 2003

re classroom



As Teaching Aids for a Unit Operations Course

Tulane University New Orleans, LA 70118

he design of separation processes frequently uses in-
tensive trial-and-error procedures as well as graphi-
cal methods such as McCabe-Thiele, Ponchon-
Savarit, and Triangular diagrams. Using process simulation
design packages such as ChemCAD, HYSYS, or AspenPLUS
facilitates the design of complex processes, but students of-
ten treat simulators as black-boxes and tend to accept the re-
sults they obtain without further analysis.1' Additionally simu-
lators may not provide the user with knowledge of all calcu-
lations that are performed or the respective algorithms. On
the other hand, manual step-by-step calculations and graphi-
cal methods, while allowing students to understand the fun-
damentals of the design process, do not equip them with the
ability to adapt software tools to the solution of chemical
engineering problems or to critically use existing simulation
packages. Tools such as MS Excel Macros and small Visual
Basic Applications (VBA) bridge the gap between the previ-
ous alternatives.
At Tulane University, spreadsheets have been intensively
used as teaching aids in undergraduate courses.12'31 Using
spreadsheets and VBA to solve chemical engineering prob-
lems requires a deep understanding of the concepts behind
the calculations, while the extensive and time-consuming trial-
and-error procedures are left to the computer. The interactive
nature of the spreadsheets and VBA programs allows "what
if" analyses in which the parameter values are changed and
the results are immediately displayed.121 One advantage of
using spreadsheets as teaching tools is that the instructor can
spend significantly more time discussing the fundamentals
of mass transfer and the conceptual and quantitative descrip-
tion of processes, as well as the engineering insight that is
needed in designing distillation, absorption, and other sepa-
rations, by spending less class time on the details of solving
problems graphically and by trial-and-error."41
* Currently an Assistant Professor at North Carolina State University.

During this course, we initially lectured on the fundamen-
tals of the calculation methods and presented illustrative ex-
amples. These first examples were solved using the trial-and-
error and graphical procedures. Then we presented a solu-
tion to the same problem using spreadsheets and VBA. We
discussed the details on how to program the spreadsheets and
elaborate the macros. Once the students had learned how to
use these tools, we asked them to develop their own Excel
Macros and VBA programs to solve problems for homework.
We used a process simulator (ChemCAD) during some of
the lectures and compared results obtained from both ap-
proaches. As a final project, we asked the students to create a
more complex algorithm for the design of an absorber. We
were very pleased to see that the students had developed very
creative, user-friendly computer programs. By the end of the
course, 76% of the students used some kind of Excel spread-
sheets and Macros to solve their homework, compared to
11.5% at the beginning of the semester.

Juan P. Hinestroza is Assistant Professorat North
Carolina State University, Department of Textile
Engineering, Chemistry and Science. He received
his PhD from Tulane University in 2002. His re-
search interests are in the development, testing,
and modeling of novel protective clothing and bar-
rier materials.

Kyrlakos D. Papadopoulos is Professor of
Chemical Engineering at Tulane Universil hav-
ing joined its faculty in 1981 and served as De-
partment Chair from 1998 to 2001. He obtained
his DEngSc from Columbia University in 1982.
His research focuses on some of the phenom-
ena that are important in the separation, trans-
port, and reaction processes of particulate sys-
tems, with emphasis on drug delivery, lubricant-
technology, and environmental applications.
Copyright ChE Division of ASEE 2003

Chemical Engineering Education

It should be noted that although we have chosen Excel,
Version 2000, for all examples in this paper, other spread-
sheet programs such as Quattro Pro and Lotus will perform
equally well.


The design of absorption using the McCabe-Thiele diagram
can be considered as a graphical solution to a series of se-
quential nonlinear equations.41] Spreadsheets have been used
in solving simultaneous nonlinear equations due to their in-
corporation of a variety of mathematical functions and the
ease of interactive programming, modification, and rapid
graph generation."1'

Example 10.3 from Geankopolis[6] is used here to illustrate
the use of spreadsheets in the design of absorption units. The
problem requires removal of acetone from an acetone-air gas
stream using water in a countercurrent stage tower. The pro-
cess schematic and spreadsheet used for solving this prob-




lem are shown in Figure 1. The initial data provided in the
problem, such as the percentage of recovery and the flows
and composition of the entering gas and liquid streams, are
shown in the upper portion of the spreadsheet under design
parameters. Assumptions include a constant molar overflow
in the tower, negligible solubility of air in the water, and a
phase equilibrium relationship that could be represented by
Henry's Law. The compositions of acetone in the liquid and
vapor outlets, xN and yN+, can be obtained from a mass
balance as shown in cells D8 and D9 of the spreadsheet;
the equations have been added to the respective comment
bubbles on the graph.

The equilibrium and operating lines are plotted using
Henry's Law and Equation 10.3-13 from Geankoplis, as
shown in the D15-E25 cell range of the spreadsheet and the
respective comment bubbles. Using Excel's "chart wizard,"
an X-Y plot can be readily constructed showing the equilib-
rium and operating lines.

Absorption of Acetone in a Countercurrent Stage Tower

% Recovery 80 0.0300
L [kg-mol/h], V[kg mol-h] 90 30 pe
x0, yN+1 0 0.01 0.0250 -
xN,yl 0.00267 0.00200
Henry's Constant 0"0 2.53 0 0200 -

[-(D6*D7+E6*E7- 5 1 1



Y Operating
0 0260



Equilibrium 0.0100 Equilibrium
0.0000 Line
0.0034 00050
0.0135 00000
00169 a 0000 0.0020 0.0040 0.0060 0.0080 0.0100
0.0236 X
0 0304
37 Calculating the # stages using Function
=FORECAST(E7,G30:G42,F30:F42) of Stages


Figure 1. Spreadsheet used for ilustrating absorption of acetone in a countercurrent stage tower.

Fall 2003

IB [ C D E F G H J K L M I N

Concentration in Concentration in the Number
x the Liquid Phase Y Vapor Phase of Stages
0.000000 0.000000
x0 0.000000 0.000000
0.000000 yl 0.002000 1
X1 0.000791 0.002000
0.000791 y2 0 004372 2
x2 0.001728 0.004372
0.001728 y3 0007184 3
X3 0.002839 0.007184
0 002839 y4 0.010518 4
X4 0004157 0010518
0004157 y5 0.014472 5
X5 0.005720 0.014472
0.005720 y6 0.019161 6
x6 0 007573 .==.:-0.019161
Q-FS0S6J 'C~F2


The next step involves the plate-by-plate calculation of the
concentration of acetone in the liquid and vapor phases. In
hand calculations, this step is made graphically,[4] but in this
case advantage is taken of the fact that both the operating and
equilibrium lines are expressed as mathematical functions,
so the concentration in the liquid and vapor phases can be
easily determined numerically, as shown in the D29-F42 cell
range of the spreadsheet. The (x,y) data series is added to the
existing graph, thus completing a numerical McCabe-Thiele
diagram, also shown in Figure 1.

Finally, the number of ideal stages required for the desired
percentage of recovery is determined using the built-in
"FORECAST' function, which operates as a linear interpola-
tor. The series in the interpolation represents the number of
stages and the concentration of the vapor phase (columns start-
ing at F29 and G29), while the value to be interpolated is the
calculated concentration at the exit of the tower (E7).

Once the spreadsheet is built, several "what-if' scenarios

can be analyzed. For instance, in this example an increase in
the recovery requirement as well as a moderate increase in
the gas flow will readily show that the number of stages re-
quired will increase significantly. Also, a large decrease in
the liquid flow rate or an increase in the gas flow rate will
demonstrate that the separation is impossible to achieve as
the operating line and the equilibrium line cross each other.


Trial-and-error iterative procedures for determining the
composition of the interface between immiscible phases are
frequently required in mass-transfer-based separation pro-
cesses. To demonstrate the versatility of spreadsheets in ac-
complishing this task, we use Example 10.4-1 from
Geankoplis,61] as shown in Figure 2. The objective of this prob-
lem is to determine the interfacial concentrations of the va-
por and liquid phases YA, and xAP, respectively, in a wetted-

I I c I D I E [ F G H I J K

2 Iterative calculation of interface compositions in interphase mass transfer

4 Gas phase film mass transfer coefficient ky= 0.001465 Kg mol A/ m^2
5 Liquid phase mass transfer coefficient kx= 0.001967 Kg mol A/s a m^3

a XAL 0.1 Ne
7 YAG 0.38 1re

10 Formula First Iteration Secon,
11 xi Initial Guess 0.4 V
12 yi Initial Guess 0.9
13 (1-xasi) =((1-$DS6)-(1-E11))/LN((1-$D$6)/(1-E11)) 0739891
14 (1-yal) =((1-E12)-(1-$D$7))/LN((1-E12)/(1-$D$7)) 0.285002
Calculated Slope from
is Equation 10-4-8 =-(F$5/E13)/(SF$4/E14) -0.517186
16 x equiliblum 0.302346
y equi (From slope 5. 0
17 calculations) =+E15*(E16-$D$6)+$D$7 0. 50
y equi (from Equilibrium =8.1414*E16^3-1.4766*E16^2+O.6184*E16-
ie correlation) 0.0022 0.27480
19 Error =+(E17-E18)^2 0 000

w guess from the
vious iteration J
7/'- ---

ration Third Iteration Fourth Iteration
0302346 0.258376 0.257171
0.275350 0200427 0.197928
0.794537 0.818259 0.818902
0.670966 0.705984 0.707134
1 133842 -1 158433 -1 159408
0.258376 0.257171 0.257145
0200427 0197928 0197805
0.199593 0 197805 0 197767
.000001 0 000000 0.00000C

Value obtained from
solving numerically the
Equilibrium equation and
Equation 10-4-8

0.090 0.140 0.190 0.240 0.290 0.340

Final Value

Equilibrium Data
xa ya
0.000 0.000
0.050 0.022
0.100 0.052
0.150 0.087
0.200 0.131
0.250 0.187
0.300 0.265
0.360 0.385


Error between the value of y
from Equation 10-4-8 and the
value predicted by the
equilibrium correlation
This is the target cell for the
SOLVER routine and the
objective is to minimize it.

Figure 2. Spreadsheet and Macro used for interactive calculation of interface compositions in interphase mass transfer.

Chemical Engineering Education


' Macro recorded 9/22/2002 by JPH

Range("G2T7) Select
SolverOk SetCell:="$E$19", MaxMinVal:=3, ValueOf:="0", ByChange:="$E$16"
SolverOk SetCell.="$F$19". MaxMinVal:=3. ValueOf:="0". ByChange:="$F$16"
SolverOk SetCell:="$G$19", MaxMinVal:=3, ValueOf:="0", ByChange'="$G$16"
SolverOk SetCell:="$H$19", MaxMinVal -3, ValueOf:="0", ByChange:="$H$16"
End Sub

wall tower. Experimental equilibrium data are provided as
well as the gas and liquid phase film mass-transfer coeffi-
cients. In this problem, the solute A diffuses through stag-
nant B in the gas phase and then through a liquid film.

The first step in solving this problem involves initial guesses
for xAi and YAi- In solving the problem by hand, these guesses
are crucial to the rapid convergence of the iterative process.
Spreadsheets are less sensitive to the initial guesses as a large
number of iterations can be processed and visualized in frac-
tions of seconds.

In Figure 2, cells D13-D19 display the equations and col-
umns E H display results from four iterations. Once the
initial guesses are selected (cells E 11 and E12), the slope for
the line connecting the bulk concentration and the assumed
interfacial concentrations is calculated, as shown in cell E15.
With the slope from El 15 and point P (the bulk concentration
in cells D6 and D7) on the x-y plot in the lower-right comer
of Figure 2, an equation for a straight line is deduced as shown
in cell D 17. A third-order polynomial was used to fit the equi-
librium data (cell D18).

The Excel function "SOLVER' is used to solve simulta-

neously the equations in cells D17 and D18 by minimizing
the error between the values of cells E16 and E17. SOLVER
can use a Newton or a conjugate numerical procedure to find
the answer; the default Newton procedure was chosen for
this example. When comparison between the values for x,
and yA. from this procedure (cells E15 and E16) and the
initial guesses (cells El2 and E13) shows a discrepancy,
an additional iteration is required. The latest calculated
values for xAi and yAi (cells E15 and E16) are used as the
new initial guesses.

Due to the ease of modification of spreadsheets, the cells
containing the equations can be copied and pasted into the
next columns as many times as necessary. In this example,
four iterations provide a reliable answer (less than 0.1% be-
tween the latest and penultimate calculated values).

What-if" scenarios in this example include how an increase
in the liquid-film mass-transfer coefficient will readily show
that the value for the interfacial concentrations xAi and YAi
increase and how a large decrease in the bulk concentration
will produce a significant decrease in xAi and YAi- In order to
automate the iteration process, a MACRO was created using


D _^ _Calculatin the DItillhat and Bo9tsm FFowrai
D 70. 401 Overall M Bala 560098E-05 =D5-C15-C16
1241173Component Mass Balance 2 3696E-05 =D5D6-C15"D7-C16'D8
i E Mrr 369859EA. =E15'2-E16^2
0 1141 =(D10+D11"(D13-D12))/D010
Slope q Line 6 163 =9/(C1-1)
Intercept -2 065 +D6-C20'D6

1 300431=IOO E2411(D8-E23) Num8b880W Plare8
41033"0433 -E24-C31.E23

7.373345262 ,135143)

32 1 V 400deposes hfl E2r~n '1e 8505 Un L;SC2
-10 "

0 578
0 22

-0 033040433
0 612205667
1 017978988

-2 065273689
-1 572218952




0.7 Ennching



0.3 Snppin



0.0 0.2 0.4 0.6 0.8 1.0


98I ag 1 29 139 39 04400979109 i* IMacr0osfillabof Mact

40 X Y Equilibdum X StrippIng X Enriching Sta8 e
=IF(K35>L35.K35.L35) =4 3392'130*5-12 539130*4 -(J30-$CS33)/SC$32 =(C30-$J$309)/$C29
*14.226'130&3 -
8 4962'130'2 3 4699'130
42 0 1 0 275043492 0 231571649 0 106304365 1
43 0231571649 0491409759 0394203612 0 37676219 2
44 0394203612 0657535069 0519071890 0584418837 3
48 0 54418837 0 798732794 0.625203337 0 760915993 4
48 0760915993 0.891914543 0695243509 0877393178 5
47 0877393178 0 93799581 0729880549 0934994783 6
4 0 934994763 0 962684833 0 748438082 0 965856041 7
49 0965856041 0978595855 0760397621 0 985744819 8
0 0 9085744819 0 9900439752 0 7693001 1 00054969 9

SMacro recorded 9292002 by JPH

SolverOk SetCelL="SF$18", MaxMinVa.=2, ValueOf ="0, ByChange.="SC$16 SCS17"
Solve0Ok SetCell ="$F$27, MaxMinVal =2, ValueOf ="". ByChange ="$F24"
End Sub

Figure 3. Spreadsheet and Macro used for distillation of a benzene-toluene mixture.

Fall 2003

Rectification of a Benzene-Toluene Mixture
F 200 molnh
xF 04
"D 0.90
.W 0.1
R 4
Latent Hea 32090 KJKg
C,. 1859 /Kg-ol1
T, 327.6 K
T. 366.700 K

NiT EtelogpLi

29 Slope
30 lid1e-p

IY firnll, enh~el On9e 0 264816 IC8E3
Emor 930E-M (.9393A


VBA; its text is also shown in Figure 2.
To create a user-friendly interface, a button is inserted into
the spreadsheet using the 'FORM TOOLBAR' menu from
Excel and assigning the Macro to it. The button allows the
user to run several "what-if' scenarios by changing the de-
sign parameters.

While interfacial composition calculations used a VBA
program and the absorption example was based on cell and
formula manipulation of the spreadsheet, in this example a
combination of both approaches is used for the design of a
distillation unit. Such design is made using the McCabe-Thiele
diagram with special considerations for the location of the
feed and the types of condenser and reboiler.161
Example 11.4-2 from Geankoplis' book is chosen to illus-
trate use of spreadsheets in the design of distillation towers.
The problem requires the rectification of a benzene-toluene
mixture. Initial data of the problem include the flow and con-
dition of the feed stream as well as its composition. The re-
flux ratio and the compositions of the distillate and bottoms
are also specified. These design parameters are located in the
upper portion of Figure 3 under design parameters. It is as-
sumed that a constant molar overflow is present in the tower.
Solving the overall mass balance (cell F15) and a benzene
mass balance (cell F16) simultaneously with SOLVER pro-
vides the values for the distillate and bottom-stream flow-
rates (cells C15 and C16). In this example, we take advan-
tage of the capabilities of SOLVER for multivariable calcu-
lations. The error cell (cell F17) is set as the target cell, and
the SOLVER should change the values of cells C15 and C16
until the value of F16 becomes negligible. The multivariable
optimization capabilities of SOLVER are implicit, which is
very useful since no additional programming is required.
After calculating all flowrates, the next step is to build the
equilibrium and operating lines. The equilibrium line is con-
structed using experimental equilibrium data and fitted to a
fifth-degree polynomial using the TRENDLINE option of
Excel. The "q line" is calculated by using a boiling point dia-
gram and the physical properties of the feed stream. Cells
B 18 to D21 show the calculations performed to obtain the
value of q and hence the slope and intercept of the "q line."
The enriching line is constructed using Eq. 11.4-8 from
Geankoplis,[6] as shown in cell range B26 to D27. Once the
slope and intercept of the q and enriching lines are deter-
mined, a numerical method is used to calculate the intercept
between these two lines. SOLVER is again used as shown in
cell range E21 to G25. Since this problem requires the use of
SOLVER twice, a VBA program is built and assigned to a
button so these calculations are automated with a single click
by the user. The stripping line is constructed using the initial
conditions of the problem and the intercept between the q
and the enriching lines as shown in cell range B28 to D 30.

The table containing the data as well as the formulas used
to determine the equilibrium, enriching, stripping, and q lines
is shown on cell range B32 to G38. To calculate the number
of plates required for the rectification, the following proce-
dure is followed, as shown in cell range B40 to F50. The
initial point (cell B42) corresponds to the bottoms concentra-
tion, xw,, y is calculated using the equilibrium equation (cell
C42), and the equations for the enriching and stripping lines
are used for cells D42 to E50. For every iteration an IF state-
ment is used to select the larger value for x. This IF statement
initially selects the stripping line as the operating line, but
once the "q line" is reached, the enriching line becomes the
operating line. The number of plates is calculated using the
FORECAST function as shown in cells E29 to G29. Based
on the spreadsheet, "what-if' scenarios can be considered and
the student is able to visualize the effect of changes in the
design parameters such as concentrations, flowrates, reflux
ratios, etc., on the number of plates required for a desired
separation. Concepts such as the pinch point and the mini-
mum reflux ratio can also be analyzed.

MS Excel Macros and Visual Basic for Applications en-
hanced the educational experience of students in a junior-
level separation processes course, teaching them to develop
simple software and providing them with an intermediate step
between doing hand calculations and using commercially
available packages. Distillation, absorption, and interfacial
mass-transfer problems were solved using spreadsheets and
were incorporated into a web-based learning platform.
In addition to analyzing several "what-if" scenarios, these
teaching tools can also be slightly modified to solve the in-
verse problems. For example, in the absorption case, the num-
ber of stages as well as the inlet flowrates and concentrations
can be given as design parameters, and then the students can
be asked to determine the concentration of the outlet streams.
Also, in the distillation case, the number of plates in the en-
riching and stripping sections can be fixed, and the students
can be asked to determine the appropriate reflux ratio and
inlet flowrates to achieve a certain degree of purity in the top
or bottom streams.

1. Wankat, P., "Teaching Separations: Why, What, When, and How?,
Chem. Eng. Ed., 35, 168 (2001)
2. Rives, C., and D. Lacks, "Teaching Process Control with a Numerical
Approach Based on Spreadsheets, Chem. Eng. Ed., 36, 242 (2002)
3. Mitchell, B.S., "Use of Spreadsheets in Introductory Statistics and
Probability," Chem. Eng. Ed., 31, 194 (1997)
4. Bums, M., and J. Sung, "Design of Separation Units Using Spread-
sheets," Chem. Eng. Ed., 29, (1995)
5. Mackenzie, J., and M. Allen, "Mathematical Power Tools, Chem. Eng.
Ed., 32, (1998)
6. Geankoplis, C., Transport Processes and Unit Operations, 3rd ed.,
Prentice Hall PTR (1993) 1

Chemical Engineering Education

Graduate Education in Chemical Engineering

Teaching and
research assistantships
as well as
industrially sponsored
up to

In addition to
tuition and fees
are waived.

PhD students
may get
some incentive

The deadline for
April 15th.

Multiphase Processes,
Fluid Flow, Interfacial
Phenomena, Filtration,

Nanocomposite Materials,
Sonochemical Processing,
Polymerization in
Nanostructured Fluids,
Supercritical Fluid

Catalysis, Reaction
Engineering, Environ-
mentally Benign

Molecular Simulation,
Phase Behavior, Physical
Properties, Process

Materials Processing and
CVD Modeling
Plasma Enhanced Deposition
and Crystal Growth

L. K. JU
Biochemical Engineering,

BioMaterial Engineering
and Polymer Engineering

Surface Modification,
Polymer Thin film

Nonlinear Control,
Chaotic Processes
Product Development

Biocatalysis and

For Additional Information, Write
Chairman, Graduate Committee
Department of Chemical Engineering The University of Akron Akron, OH 44325-3906
Phone (330) 972-7250 Fax (330) 972-5856

Fall 2003





A dedicated faculty with state of the art
facilities offer research programs leading to
Doctor of Philosophy and Master of Science
Research Areas:

Biomaterials, Catalysis and Reactor Design,
Drug Delivery Materials and Systems,
Electrohydrodynamics, Electronic Materials,
Environmental Studies, Fuel Cells, Interfacial
Transport, Magnetic Materials, Membrance
Separations and Reactors, Microelectro-
Mechanical Systems, Molecular Simulations,
Nanoscale Modeling, Polymer Processing and
Rheology, Process Dynamics, Self-Assembled
Materials, Suspension and Slurry Rheology,
Transport Process Modeling
For Information Contact:
Director of Graduate Studies
Department of Chemical Engineering
The University of Alabama
Box 870203
Tuscaloosa, AL 35487-0203
Phone: (205) 348-6450

G. C. April, Ph.D. (Louisiana State)
D. W. Arnold, Ph.D. (Purdue)
C. S. Brazel, Ph.D. (Purdue)
E. S. Carlson, Ph.D. (Wyoming)
P. E. Clark, Ph.D. (Oklahoma State)
W. C. Clements, Jr., Ph.D. (Vanderbilt)
R. A. Griffin, Ph.D. (Utah State)
D. T. Johnson, Ph.D. (Florida)
T. M. Klein, Ph.D. (NC State)
A. M. Lane, Ph.D. (Massachusetts)
M. D. McKinley, Ph.D. (Florida)
S. M. C. Ritchie, Ph.D. (Kentucky)
C. H. Turner, Ph.D. (NC State)
J. M. Wiest, Ph.D. (Wisconsin)
M. L. Weaver, Ph.D. (Florida)

An equal employment / equal
educational opportunity institution

Chemical Engineering Education

Chemical &

Materials Engineering


he Department of Chemical and Materi-
als Engineering at the University of Ala-
bama in Huntsville offers you the oppor-
tunity for a solid and rewarding graduate career
that will lead to further success at the forefront
of academia and industry.
We will provide graduate programs that educate
and train students in advanced areas of chemical
engineering, materials science and engineering,
and biotechnology. Options for an MS and PhD
degree in Engineering or Materials Science are
Our faculty are dedicated to international lead-
ership in research. Projects are ongoing in Mass
Transfer, Fluid Mechanics, Combustion,
Biosparations, Biomaterials, Microgravity Mate-
rials Processing, and Adhesion. Collaborations
have been established with nearby NASA/
Marshall Space Flight Center as well as leading
edge biotechnology and engineering companies.
We are also dedicated to innovation in teaching.
Our classes incorporate advances in computational
methods and multi-media presentations.

Department of Chemical Engineering
The University of Alabama in Huntsville
130 Engineering Building
Huntsville, AL 35899

Michael R. Banish- Ph.D. (University of Utah)
Thermo physical property measurements
(256) 824-6969,
Ram6n L. Cero Ph.D. (UC-Davis)
Professor and Chair
Capillary hydrodynamics, multiphase flows, enhanced heat transfer
(256) 824-7313,
Chien P. Chen Ph.D. (Michigan State)
Multiphase flows, spray combustion, turbulence modeling,
numerical methods in fluids and heat transfer.
(256) 824-6194,
Krishnan K. Chittur Ph.D. (Rice)
Protein adsorption to biomaterials, FTR/ATR at solid-liquid
interfaces, biosensing.
(256) 824-6850,
Douglas G. Hayes Ph.D. (Michigan)
Associate Professor
Enzyme reactions in nonaqueous media, separations involving
biomolecules, lipids and surfactants, surfactant-based colloidal
(256) 824-6874,
James E. Smith Jr. Ph.D. (South Carolina)
Kinetics and catalysis, powdered materials processing, combustion
diagnostics and fluids visualization using optical methods.
(256) 824-6439,
Jeffrey J. Weimer Ph.D. (MIT)
Associate Professor, Joint Appointment in Chemistry
Adhesion, biomaterials surface properties, thin film growth, surface
spectroscopies, scanning prode microscopies.
(256) 824-6954,

The University of Alabama in Huntsville
An Affirmative Action/Equal Opportunity Institution
Web page:
Ph: 256*824*6810 FAX: 256-824.6839

Fall 2003

The University of Alberta is well known
for its commitment to excellence in teach-
ing and research. The Department of
Chemical and Materials Engineering has
37 professors and over 140 graduate
students. Degrees are offered at the M.Sc.
and Ph.D. levels in Chemical Engineer-
ing, Materials Engineering, and Process
Control. All full-time graduate students in
the research programs receive a stipend
to cover living expenses and tuition.

For further information, contact
Graduate Program Officer
Department of Chemical and Materials Engineering
University of Alberta
Edmonton, Alberta, Canada T6G 2G6
PHONE (780) 492-1823 FAX (780) 492-2881
e-mail: chemical. engineering@ ualberta. ca

M. BHUSHAN, Ph.D. (I.I.T. Bombay)
Sensor Location Fault Diagnosis Process Safety
R.E. BURRELL, Ph.D. (University of Waterloo)
Nanostructured Biomaterials Drug Delivery Biofilms Tissue Integration with Materials
P. CHOI, Ph.D. (University of Waterloo)
Molecular Modeling of Polymers Thermodynamics of Polymer Solutions and Blends
K. T. CHUANG, Ph.D. (University of Alberta)
Fuel Cell Catalysis Separation Processes Pollution Control
I. G. DALLA LANA, Ph.D. (Univ. of Minnesota) EMERITUS
Chemical Reaction Engineering Heterogeneous Catalysis
J. A. W. ELLIOTT, Ph.D. (University of Toronto)
Thermodynamics Statistical Thermodynamics Interfacial Phenomena
D. G. FISHER, Ph.D. (University of Michigan) EMERITUS
Process Dynamics and Control Real-Time Computer Applications
J.F. FORBES, Ph.D. (McMaster University) CHAIR
Real-Time Optimization Scheduling and Planning
M. R. GRAY, Ph.D. (California Inst. of Tech.)
Bioreactors Chemical Kinetics Bitumen Processing
R. E. HAYES, Ph.D. (University of Bath)
Numerical Analysis Reactor Modeling Computational Fluid Dynamics
B. HUANG, Ph.D. (University of Alberta)
Controller Performance Assessment Multivariable Control Statistics
S. M. KRESTA, Ph.D. (McMaster University)
Turbulent & Transitional Flows Multiphase Flows CFD
S. LIU, Ph.D. (University of Alberta)
Fluid-Particle Dynamics Transport Phenomena Kinetics
D. T. LYNCH, Ph.D. (University of Alberta) DEAN OF ENGINEERING
Catalysis Kinetic Modeling Numerical Methods Polymerization
J. H. MASLIYAH, Ph.D. (University of British Columbia)
Transport Phenomena Colloids Particle-Fluid Dynamics Oil Sands
A. E. MATHER, Ph.D. (University of Michigan)
Phase Equilibria Fluid Properties at High Pressures Thermodynamics
E. S. MEADOWS, Ph.D. (University of Texas)
Process Control Fuel Cell Modeling and Control Optimization
W. C. MCCAFFREY, Ph.D. (McGill University)
Reaction Kinetics Heavy Oil Upgrading Polymer Recycling Biotechnology
K. NANDAKUMAR, Ph.D. (Princeton University)
Transport Phenomena Distillation Computational Fluid Dynamics
A.E. NELSON, Ph.D. (Michigan Technological University)
Heterogeneous Catalysis UHV Surface Science Chemical Kinetics
M. RAO, Ph.D. (Rutgers University)
AI Intelligent Control Process Control
S. L. SHAH, Ph.D. (University of Alberta)
Computer Process Control System Identification Process and Performance Monitoring
J.M. SHAW, Ph.D. (University of British Columbia)
Petroleum Thermodynamics Multiphase Mixing Process Modeling
U. SUNDARARAJ, Ph.D. (University of Minnesota)
Polymer Processing Polymer Blends Interfacial Phenomena
H. ULUDAG, Ph.D. (University of Toronto)
Biomaterials Tissue Engineering Drug Delivery
S. E. WANKE, Ph.D. (University of California, Davis)
Heterogeneous Catalysis Kinetics Polymerization
M. C. WILLIAMS, Ph.D. (University of Wisconsin) EMERITUS
Rheology Polymer Characterization Polymer Processing
Z. XU, Ph.D. (Virginia Polytechnic Institute and State University)
Surface Science & Engineering Mineral Processing Waste Management
T. YEUNG, Ph.D. (University of British Columbia)
Emulsions Interfacial Phenomena Micromechanics

Chemical Engineering Education

ROBERT G. ARNOLD, Professor (CalTech)
Microbiological Hazardous Waste Treatment, Metals Speciation and Toxicity
PAUL BLOWERS, Assistant Professor (Illinois, Urbana-Champaign)
Chemical Kinetics, Catalysis, Surface Phenomena

JAMES C. BAYGENTS, Associate Professor (Princeton)
Fluid Mechanics, Transport and Colloidal Phenomena, Bioseparations

WENDELL ELA, Assistant Professor (Stanford)
Particle-Particle Interactions, Environmental Chemistry

JAMES FARRELL, Associate Professor (Stanford)
Sorption/desorption of Organics in Soils

JAMES A. FIELD, Associate Professor (Wagenigen Agricultural Univ.)
Bioremediation, Microbiology, White Rot Fungi, Hazardous Waste

ROBERTO GUZMAN, Associate Professor (North Carolina State)
Affinity Protein Separations, Polymeric Surface Science

ANTHONY MUSCAT, Associate Professor (Stanford)
Kinetics, Surface Chemistry, Surface Engineering, Semiconductor
Processing, Microcontamination

KIMBERLY OGDEN, Professor (Colorado)
Bioreactors, Bioremediation, Organics Removal from Soils

THOMAS W. PETERSON, Professor and Dean (CalTech)
Aerosols, Hazardous Waste Incineration, Microcontamination

ARA PHILIPOSSIAN, Associate Professor (Tufts)
Chemical/Mechanical Polishing, Semiconductor Processing

EDUARDO SAEZ, Associate Professor (UC, Davis)
Polymer Flows, Multiphase Reactors, Colloids

FARHANG SHADMAN, Professor (Berkeley)
Reaction Engineering, Kinetics, Catalysis, Reactive Membranes,

JOST 0. L. WENDT, Professor and Head (Johns Hopkins)
Combustion-Generated Air Pollution, Incineration, Waste

For further information, write to

Chemical and Environmental





The Department of Chemical and Enrivonmental Engineering
at the University of Arizona offers a wide range of research
opportunities in all major areas of chemical engineering and
environmental engineering. The department offers a fully accredited
undergraduate degree in chemical engineering, as well as MS and PhD
degrees in both chemical and environmental engineering. A signifi-
cant portion of research efforts is devoted to areas at the boundary
between chemical and environmental engineering, including environ-
mentally benign semiconductor manufacturing, environmental
remediation, environmental biotechnology, and novel
water treatment technologies.
Financial support is available through fellowships, government
and industrial grants and contracts, teaching and
research assistantships.

Tucson has an excellent climate and many
recreational opportunities. It is a growing modern city that
retains much of the old Southwestern atmosphere.

or write

Chairman, Graduate Study Committee
Department of Chemical and
Environmental Engineering
P.O. BOX 210011
The University of Arizona
Tucson, AZ 85721

The University of Arizona is an equal
opportunity educational institution/equal opportunity employer.
Women and minorities are encouraged to apply.

Fall 2003



Department of Chemical and Materials Engineering

A Distinguished and Diverse Faculty A multi-disciplinary research
Chemical Engneering environment with opportunities
in electronic materials
Jonathan Allen, Ph.D., MIT. Atmospheric aerosol chemistry, single-particle measurement processing biotechnology *
techniques, environmental fate of organic pollutants processing, characterization,
James Beckman, Ph.D., Arizona. Unit operations, applied mathematics, energy-efficient water and simulation of materials *
purification, fractionation, CMP reclamation ceramics air and water
Veronica Burrows, Ph.D., Princeton. Surface science, environmental sensors, semiconductor purification atmospheric
processing, interfacial chemical and physical processes in sensor processing chemistry process control
Ann Dillner, Ph.D., Illinois, Urbana-Champaign. Atmospheric particulate matter (aerosols)
chemistry and physics, ultra fine aerosols, light scattering, climate and health effects of
Chan Beum Park, Ph.D., POSTTECH, South Korea. Bioprocess in extremist, novel cell-free
protein synthesis, biolab-on-a-chip technology
Gregory Raupp, Ph.D., Wisconsin. Gas-solid surface reactions mechanisms and kinetics,
interactions between surface reactions and simultaneous transport processes, semiconductor
materials processing, thermal and plasma-enhanced chemical vapor deposition (CVD)
Anneta Razatos, Ph.D., Texas at Austin. Bacterial adhesion, colloid interactions, AFM, biofilms,
genetic engineering
Daniel Rivera, Ph.D., Caltech. Control systems engineering, dynamic modeling via system
identification, robust control, computer-aided control system design
Michael Sierks, Ph.D., Iowa State. Protein engineering, biomedical engineering, enzyme
kinetics, antibody engineering

Materials Science and Engineering
James Adams, Ph.D., Atomistic stimulation of metallic surfaces, adhesion, wear, and automotive
catalysts, heavy metal toxicity
Terry Alford, Ph.D., Comrnell. Electronic materials, physical metallurgy, electronic thin films
Nikhilesh Chawla, Ph.D., Michigan. Lead-free solders, composite materials, powder metallurgy
Sandwip Dey, Ph.D., Alfred. Electro-ceramics, MOCVD and ALCVD, dielectrics: leakage, loss
mechanisms and modeling
Stephen Krause, Ph.D., Michigan. Characterization of structural changes in processing of semiconductors
Subhash Mahajan (Chair), Ph.D., Berkeley. Semiconductor defects, high temperature semiconductors, structural materials deformation
James Mayer, Ph.D., Purdue. Thin film processing, ion beam modification of materials
Nathan Newman, Ph.D., Stanford. Growth, characterization, and modeling of solid-state materials
S. Tom Picraux, Ph.D. Caltech. Nanostructured materials, epitaxy, and thin-film electronic materials
Karl Sieradzki, Ph.D. Syracuse. Fracture of solids, thin-film deposition and growth, corrosion
Mark van Schilfgaarde, Ph.D. Stanford. Methods and applications of electronic structure theory, dilute magnetic semiconductors, GW approximation

For details concerning graduate opportunities in Chemical and Materials Engineering atASU, please call Marlene Bolf
at (480) 965-3313, or write to Subhash Mahajan, Chair, Chemical and Materials Engineering, Arizona State University,
Tempe, Arizona 85287-6006 (

Chemical Engineering Education

Graduate Program in the Department of Chemical Engineering

University of Arkansas

cZ& The Department of Chemical Engineering at the University of Arkansas
S. offers graduate programs leading to M.S. and Ph.D. Degrees.
I Qualified applicants are eligible for financial aid. Annual Departmental
stipends provide up to $15,000, Doctoral Academy Fellowships provide
up to $20,000, and Distinguished Doctoral Fellowships provide $30,000.
For stipend and fellowship recipients, all tuition is waived. Applications
'ars z00 received before April 1st will be given first consideration.

Areas of Research

[1 Biochemical engineering
EN Biological and food systems
EN Biomaterials
EN Chemical process safety
EN Consequence analysis of hazardous chemical releases
EI Electronic materials processing
E[ Fate of pollutants in the environment
El Fluid phase equilibria and process
EU Integrated passive electronic Faculty
El Membrane separations M.D. Ackerson
E[ Mixing in chemical processes R.E. Babcock
R.R. Beitle
E.C. Clausen
R.A. Cross
J.A. Havens
W.A. Myers
W.R. Penney
T.O. Spicer
G.J. Thoma
J.L. Turpin
R.K. Ulrich

For more information contact
Dr. Richard Ulrich or 479-575-5645
Chemical Engineering Graduate Program Information:

Fall 2003


Chemical Engineering

Fd -
Mark E. Byrne Purdue University
Robert P. Chambers University of California, Berkeley
Harry T. Cullinan Carnegie Mellon University
Christine W. Curtis Florida State University
Steve R. Duke University of Illinois
Mark R. Eden Technical University of Denmark
Said S.E.H. Elnashaie University of Edinburgh
James A. Guin University of Texas, Austin
Ram B. Gupta University of Texas at Austin
Gopal A. Krishnagopalan University of Maine
Yoon Y. Lee Iowa State University
Glennon Maples Oklahoma State University
Ronald D. Neuman The Institute of Paper Chemistry
Timothy D. Placek University of Kentucky
Christopher B. Roberts University of Notre Dame
Arthur R. Tarrer Purdue University
Bruce J. Tatarchuk University of Wisconsin
4wi-, is- W i

Research Areas
" Fuel Cell* Hydrogen
" Biochemical Engineering Drug Delivery
" Pulp and Paper Microfibrous Materials
" Process Systems Engineering
" Integrated Process Design
" Environmental Chemical Engineering
" Catalysis and Reaction Engineering
. Materials. Polymers. Nanotechnology
" Surface and Interfacial Science
" Thermodynamics. Supercritical Fluids
" Electrochemical Engineering
" Transport Phenomena

Chemical Engineering Education



R. G. Moore, Head (Alberta)
J. Azaiez (Stanford)
L. A. Behie (Western Ontario)
C. Bellehumeur (McMaster)
P. R. Bishnoi (Alberta)
J.M. Hill (Wisconsin)
A. A. Jeje (MIT)
M. S. Kallos (Calgary)
A. Kantzas (Waterloo)
B. B. Maini (Univ. Washington)
A. K. Mehrotra (Calgary)
S. A. Mehta (Calgary)
P. Pereira (France)
M. Pooladi-Darvish (Alberta)
A. Sen (Calgary)
A. Settari (Calgary)
W. Y. Svrcek (Alberta)
M. A. Trebble (Calgary)
H. W. Yarranton (Alberta)
B. Young (Canterbury, NZ)
L. Zanzotto (Slovak Tech. Univ., Czechoslovakia)

The Department offers graduate programs leading to the M.Sc. and Ph.D.
degrees in Chemical Engineering (full-time) and the M.Eng. degree in Chemical
Engineering, Petroleum Reservoir Engineering or Engineering for the
Environment (part-time) in the following areas:
Biochemical Engineering & Biotechnology
Biomedical Engineering
Upgrading, Catalysis and Fuel Cells
Environmental Engineering
Modeling, Simulation & Control
Petroleum Recovery & Reservoir Engineering
Polymer Processing & Rheology
Process Development
Reaction Engineering/Kinetics
Transport Phenomena
Fellowships and Research Assistantships are available to all qualified applicants.

For Additional Information Write *
Dr. W.Y. Svrcek Associate Head, Graduate Studies
Department of Chemical and Petroleum Engineering
University of Calgary Calgary, Alberta, Canada T2N 1 N4

The University is located in the City of Calgary, the Oil capital of Canada, the home of the world famous Calgary Stampede and the 1988
Winter Olympics. The City combines the traditions of the Old West with the sophistication of a modern urban center Beautiful Banff
National Park is 110 km west of the City and the ski resorts of Banff, Lake Louise,and Kananaskis areas are readily accessible. In the
above photo the University Campus is shown in the foreground. The Engineering complex is on the left of the picture, and the Olympic
Oval is on the right of the picture. U




Fall 2003


The Chemical Engineering Department at the
University of California, Berkeley, one of the pre-
eminent departments in the field, offers graduate pro-
grams leading to the Master of Science and Doctor
of Philosophy. Students also have the opportunity
to take part in the many cultural offerings of the San
Francisco Bay Area and the recreational activities
of California's northern coast and mountains.


Nitash P. Balsara
Harvey W. Blanch
Arup K. Chakraborty
David B. Graves
Alexander Katz
C. Judson King
Susan J. Muller
John M. Prausnitz
Jeffrey A. Reimer
Alexis T. Bell

Elton J. Cairns
Douglas S. Clark
Enrique Iglesia
Jay D. Keasling
Roya Maboudlan
John S. Newman
Clayton J. Radke
David V. Schaffer
Rachel A. Segalman

Univrsiy o CaiforiaBerele

Blanch, Clark,
Keasling, Schaffer,
Chakraborty, Muller,
Prausnitz & Radke


Balsara, Chakraborty,
Muller, Prausnitz, Radke,
Reimer & Segalman

Chairman: Arup K. Chakraborty


Chemical Engineering Education


Bell, Chakraborty,
Iglesia, Katz & Reimer

Cairns, Newman &


Graves, Maboudian,
Reimer & Segalman

-*k.10 b~ilyof Wmfot 092%

E(W dQmr.b.D, IaI*


vx~~aancvu~~ns~e~aujeIi zzm ooR eai

Department of
Chemical Engineering & Materials Science


The multifaceted graduate study experience in the Department
of Chemical Engineering and Materials Science allows students
to choose research projects and thesis advisers from any of our
faculty with expertise in chemical engineering, biochemical engi-
neering, and/or materials science and engineering.
Our goal is to provide the financial and academic support for
students to complete a substantive research project within 2 years
for the M.S. and 4 years for the Ph.D.


Sacramento: 17 miles
San Francisco: 72 miles
Lake Tahoe: 90 miles

Davis is a small, bike-friendly
university town located 17
miles west of Sacramento
INTO and 72 miles northeast of
San Francisco, within
driving distance of a
"I multitude of recreational
activities. We also enjoy
WNnES close collaborations
with national
SAN EOo including
and Sandia.

For information about our program,
look up our web site at

or contact us via e-mail at


Fall 2003



Graduate Studies in
Chemical Engineering IR VINE
and Materials Science and Engineering
for Chemical Engineering, Engineering, and Materials Science Majors
Offering degrees at the M.S. and Ph.D. levels. Research in frontier areas
in chemical engineering, biochemical engineering, biomedical engineering, and materials
science and engineering. Strong physical and life science and engineering groups on campus.
Nancy A. Da Silva (California Institute of Technology)
James C. Earthman (Stanford University)
Steven C. George (University of Washington)
Stanley B. Grant (California Institute of Technology)
Juan Hong (Purdue University)
Enrique J. Lavernia (Massachusetts Institute of Technology)
Henry C. Lim (Northwestern University)
Jia Grace Lu (Harvard University)
Martha L. Mecartney (Stanford University)
Farghalli A. Mohamed (University of California, Berkeley)
Daniel R. Mumm (Northwestern University)
Andrew J. utnam (University of Michigan)
Frank G. Shi (California Institute of Technology)
Vasan Venugopalan (Massachusetts Institute of Technology)
Joint Appointments:
G. Wesley Hatfield (Purdue University)
Noo Li Jeon (University of Illinois)
Sunny Jiang (University of South Florida)
Roger H. Rangel (University of California, Berkeley)
William A. Sirignano (Princeton University)
Adjunct Professors
Russell Chou (Carnegie Mellon University)
Andrew Shapiro (University of Califoria, Irvine)
Victoria Tellkamp (University of Califoria, Irvine)

The 1,510-acre UC Irvine campus is in Orange County, five miles from the Pacific Ocean and 40 miles south of
Los Angeles. Irvine is one of the nation's fastest growing residential, industrial, and business areas. Nearby
beaches, mountain and desert area recreational activities, and local cultural activities make Irvine a pleasant
city in which to live and study.
For further information and application forms, please visit
or contact
Department of Chemical Engineering and Materials Science
School of Engineering University of California Irvine, CA 92697-2575

Chemical Engineering Education


C w1-..dA


0 Molecular and Cellular

a Process Systems Engi-
neering (Design,
Optimization, Dynam-
ics, and Control)



0 Energy and the

g Nanoengineering i


UCLA's Chemical
Engineering Department
offers a program of teaching
and research linking
fundamental engineering
science and industrial practice. Our Department has strong graduate research programs in Bioengineer-
ing, Energy and Environment, Semiconductor Manufacturing, Engineering of Materials, and Process
and Control Systems Engineering.
Fellowships are available for outstanding applicants interested in Ph.D. degree programs. A
fellowship includes a waiver of tuition and fees plus a stipend.
Located five miles from the Pacific Coast, UCLA's attractive 417-acre campus extends
from Bel Air to Westwood Village. Students have access to the highly regarded science pro-
grams and to a variety of experiences in theatre, music, art, and sports on campus.


J. P. Chang
(William F Seyer Chair in
Materials Electrochemistry)
P. D. Christofides
Y. Cohen
J. Davis
(Vice Chancellor for
Information Technology)
S. K. Friedlander
(Parsons Professor of
Chemical Engineering)
R. F. Hicks
L. Ignarro
(Nobel Laureate)
E. L. Knuth
(Professor Emeritus)
J. C. Liao
V. Manousiouthakis
H. G. Monbouquette
K. Nobe
G. Orkoulas
L. B. Robinson
(Professor Emeritus)
S. M. Senkan
Y. Tang
W. D. Van Vorst
(Professor Emeritus)
V. L. Vilker
(Professor Emeritus)
A.R. Wazzan
(Dean Emeritus)

AdmsinsOfc Chmia Engineerin Deparmen
551 -erHl UCL o Lo Angeles CA 9009159
Teehn at (30 8296 or vii us. at wwwhemng ca .ed

Fall 2003

Offering degrees at the M.S. and Ph.D. levels in frontier areas of Chemical, Biochemical
and Biomedical, Advanced Materials, and Environmental Engineering. We welcome you
interest and would be delighted to discuss with you the details of our graduate program,
your admission into our graduate program, or your interest in our research.

* Bio- and Chemical Sensors
* Structural Bioinformatics
* Biomolecular Engineering
* Environmental Biotechnology
* Catalysis and Biocatalysis
* Nanostructured Materials
* Carbon Nanotubes
* Complex Fluids & Colloids
* Electrochemistry
* Zeolites & Fuel Cells
* Membrane Processes
* Aerosol Physics
* Atmospheric Chemistry
* Renewable Fuels
* Advanced Vehicle Technology
* Water/Wastewater Treatment
* Advanced Water Reclamation
* Site Remediation Processes

* Wilfred Chen, Caltech
* David R. Cocker, Caltech
* Marc A. Deshusses, ETH, Zurich
* Robert C. Haddon, Penn State
* Eric M.V. Hoek, Yale
* Mark R. Matsumoto, UC Davis
* Dimitrios Morikis, Northeastern
* Ashok Mulchandani, McGill
* Nosang V. Myung, UCLA
* Joseph M. Norbeck, Nebraska
* Mihri Ozkan, UC San Diego
* Jianzhong Wu, UC Berkeley
* Yushan Yan, Caltech

The University of California, Riverside (UCR) is the fastest growing and most ethnically diverse of the 10
campuses of the University of California. UCR is located on over 1,100 acres at the foot of the Box Springs
Mountains, about 50 miles east of Los Angeles. Our picturesque campus provides convenient access to the
vibrant and growing Inland Empire, and is within easy driving distance to most of the major cultural and
recreational offerings in Southern California. In addition, it is virtually equidistant from the desert, the mountains,
and the ocean. This is an ideal setting for students, faculty and staff seeking to study, work, and live in a
community steeped in rich heritage, offering a dynamic mix of arts and entertainment and an opportunity for
affordable living.

For more information and application materials,
please visit:
or contact:
Graduate Advisor
Department of Chemical/ Environmental
Engineering, University of California
Riverside, CA 92521

Chemical Engineering Education

Chemical Engineering at the





"At the Leading Edge"

Frances H. Arnold
Anand R. Asthagiri
John F Brady
Mark E. Davis
Richard C. Flagan

John H. Seinfeld
Christina D. Smolke
David A. Tirrell
Nicholas W Tschoegl (Emeritus)
Zhen-Gang Wang

Colloid Physics
Fluid Mechanics
Materials Processing
Microelectronics Processing
Microstructured Fluids
Polymer Science
Protein Engineering
Statistical Mechanics

For further information, write
Director of Graduate Studies
Chemical Engineering 210-41 California Institute of Technology Pasadena, California 91125-4100
Also, visit us on the World Wide Web for an on-line brochure:

George R. Gavalas (Emeritus)
Konstantinos P Giapis
Sossina M. Haile
Julia A. Kornfield



Aerosol Science
Applied Mathematics
Atmospheric Chemistry and Physics
Biocatalysis and Bioreactor Engineering
Biomedical Engineering
Chemical Vapor Deposition


Fall 2003

wee......C.....C.**** ...e C C CC COOS
,eee.Ceseees eeS*.C 98 *** ***
eg ** * * * *

SCUgesgeeUees.eeS*C ** U S i
ee eeC 0 *a a 0

-0eAe *t *** '
* e **SC C *
* ** -e.C~ u r

-:1 .*.-t b e M

* A
** At

C 5' 5'
CC. 5'

* C C

* Ce.. -
* e e
,g Ce



a .


tr- ins V sup eroer

become, a ri. |
Syour ia

,Ore (rt~ t .o~ OW l

Crm tiorl oft ...
Vor or siipe'herO-
C.-. ee.C sC Cr CCC
4 SOe6

omili e.
C,, P &
II hco rnU .'."@@

i o on o *.o*,o*,o
'.. '.;;.. r..; t ... .:.'. .

Case Wetr Reserv Universit

Research Opportunities

Advanced Energy Systems
Fuel Cells and Batteries
Micro Fuel Cells
Hydrogen Infrastructure
Energy Storage
Membrane Transport
Membrane Fabrication
Biomedical Engineering
Transport in Biological Systems
Biomedical Sensors and Actuators
Wound Healing
Inflammation and Cancer Metastasis
Neural Prosthetic Devices
Advanced Materials and Devices
Diamond and Nitride Synthesis
Coatings, Thin Films, and Surfaces
In-Situ Diagnostics and Sensors
Fine Particle Science and Processing
Polymer Nanocomposites
Electrochemical Microfabrication
Self Assembly Chemistry

For more information on
Graduate Research, Admission, and Financial Aid, contact:

Graduate Coordinator
Department of Chemical Engineering
8 E-mail:


John Angus
Harihara Baskaran
Robert Edwards
Donald Feke
Daniel Lacks
Uziel Landau
Chung-Chiun Liu
J. Adin Mann
Heidi Martin
Peter Pintauro
Syed Qutubuddin
Robert Savinell
Thomas Zawodzinski

Case Western Reserve University
10900 Euclid Avenue
Cleveland, Ohio 44106-7217

Fall 2003

Opportunities for Graduate Study in Chemical Engineering at the

M.S. and Ph.D. Degrees in Chemical Engineering

= Faculty

Carlos Co

Joel Fried

Rakesh Govind

Vadim Guliants

Daniel Hershey

Chia-chi Ho

Sun-Tak Hwang

Yuen-Koh Kao

Soon-Jai Khang

William Krantz

Jerry Y. S. Lin

Neville Pinto

Peter Smirniotis

Financial Aid Available

The University of Cincinnati is
committed to a policy of
non-discrimination in
awarding financial aid.

For Admission Information
Director, Graduate Studies
Department Chemical and
Materials Engineering
PO Box 210171
University of Cincinnati
Cincinnati, Ohio 45221-0171

The faculty and students in the Department of Chemical Engineering are engaged in a diverse range
of exciting research topics. Assistantships and tuition scholarships are available to highly qualified
applicants to the MS and PhD degree programs.

El Advanced Materials
Inorganic membranes, nanostructured materials, microporous and mesoporous materials,
thin film technology, fuel cell and sensor materials
El Biotechnology
Nano/microbiotechnology, novel bioseparation techniques, affinity separation, biodegrada-
tion of toxic wastes, controlled drug delivery, two-phase flow
El Catalysis and Chemical Reaction Engineering
Heterogeneous catalysis, environmental catalysis, zeolite catalysis, novel chemical reactors,
modeling and design of chemical reactors, polymerization processes in interfaces, membrane
0 Environmental Research
Desulfurization and denitrication of flue gas, new technologies for coal combustion power
plant, wastewater treatment, removal of volatile organic vapors

D Membrane Technology
Membrane synthesis and characterization, membrane gas separation, membrane filtration
processes, pervaporation, biomedical, food and environmental applications of membranes,
high-temperature membrane technology, natural gas processing by membranes

El Polymers
Thermodynamics, polymer blends and composites, high-temperature polymers, hydrogels,
polymer rheology, computational polymer science, molecular engineering and synthesis of
surfactants, surfactants and interfacial phenomena
El Separation Technologies
Membrane separation, adsorption, chromatography, separation system synthesis, chemical
reaction-based separation processes

Chemical Engineering Education

Full Text