Chemical engineering education ( Journal Site )

Material Information

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


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


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

Record Information

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

Full Text



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

T. J. Anderson

Phillip C. Wankat

Carole Yocum

James 0. Wilkes, U. Michigan

William J. Koros, University of Texas, Austin


E. Dendy Sloan, Jr.
Colorado School of Mines

Pablo Debenedetti
Princeton University
Dianne Dorland
University of Minnesota, Duluth
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
William J. Koros
University of Texas at Austin
David F. Ollis
North Carolina State University
Angelo J. Perna
New Jersey Institute of Technology
Ronald W. Rousseau
Georgia Institute of Technology
Stanley I. Sander
University of Delaware
Richard C. Seagrave
Iowa State University
M. Sami Selim
Colorado School of Mines
Stewart Slater
Rowan University
James E. Stice
University of Texas at Austin
Donald R. Woods
McMaster University

Chemical Engineering Education

Volume 34 Number 3 Summer 2000

186 Worcester Polytechnic Institute, Alvin H. Weiss

192 Frank Doyle III

198 Part 5. Assessing Teaching Effectiveness and Educational Scholarship,
Richard M. Felder, Armando Rugarcia, James E. Stice
208 Part 6. Making Reform Happen, Richard M. Felder, James E. Stice, Annando Rugarcia

216 An Alternate Method for Teaching and Implementing Dimensional Analysis,
William B. Krantz
258 ASTutE: Computer-Aided Teaching of Materials Balancing,
David W. Edwards, Fiona M. Lamb, Vian S. Ahmed, Steve J. Rothberg
278 A Real-Time Approach to Process Control Education,
Brent R. Young, Donald P. Mahoney, William Y Svrcek

222 A Project-Based, Spiral Curriculum for Introductory Courses in ChE:
Part 1. Curriculum Design, William M. Clark, David DiBiasio, Anthony G. Dixon
234 Effective Communication for Professional Engineering: Beyond Problem Sets and Lab
Reports, Mark R. Prausnitz, Melissa J. Bradley
272 A Pollution Prevention Course that Helps Meet EC 2000 Objectives, Martin A. Abraham

228 Spreading the Word (About Chemical Engineering), Gilbert Shama, Klaus Hellgardt

238 The Alumni Speak, Richard M. Felder

240 Automotive Catalytic Reaction Engineering Experiment, R.P.Hesketh, D. Bosak, L. Kline
252 A Training Simulator for Computer-Aided Process Control Education,
Doug Cooper, Danielle Doughertv
264 Simple Mass Transfer Experiment Using Nanofiltration Membranes,
A Wahab Mohammad

246 Tuning and Activation of a PI Controller During Startup of a Non-Isothermal CSTR,
Aziz M. Abu-Khalaf, Emad M. Ali

268 Issues in Developing and Implementing an Assessment Plan in ChE Departments,
James A. Newell, Heidi L. Newell. Thomas C. Owens, John J. Erjavec, Rasid A. Hasan,
Steven P.K. Stermberg

284 Courses in fluid Mechanics and Chemical Reaction Engineering in Europe,
Juan A. Conesa, Ignacio Martin-Gull6n

245, 251, 282 Letters to the Editor

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 2000 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 CEE, Chemical Engineering Department., University of Florida,
Gainesville, FL 32611-6005. Periodicals Postage Paid at Gainesville, Florida.

Summer 2000

eM ia department

Chemical Engineering at...

Worcester Polytechnic Institute

Worcester Polytechnic Institute Worcester, MA 01609
Worcester (pronounced "Wustah"), Massachusetts,
is situated on seven of the gently rolling hills of
the Massachusetts Piedmont. It is about forty-
five miles due west of Boston, all uphill. It is considered to
be in the Great Midwest, far too distant for most proper
Bostonians to venture, except on those weekends when many
pass it by on their way to the West (i.e., the Berkshires).
When a Bostonian is asked, "What is the best route to San
Francisco?" the invariable answer is "Through Worcester."
Given this perception, which has always been the situa-
tion, Worcester has developed very independently from Bos-
ton. There is surprisingly little interaction, e.g., between
WPI and MIT or Harvard.
In addition to WPI, Worcester boasts of having the Uni-

versity of Massachusetts Medical School, Clark University,
College of the Holy Cross, Assumption College, Anna Maria
College, Becker Junior College, Worcester State College,
and Quinsigamond Community College. In short, education
is the major industry in this city of less than two hundred
thousand. Worcester has a sizeable BioTechnology Park and
was home to the Worcester Foundation for Experimental
Biology (where the "Pill" was invented). These, together
with the research proceeding at the medical school and the
other colleges, combine to make Worcester's bio research a
major industry. There are also major computer companies in
Worcester and the nearby areas (e.g., DEC, EMC, Allegro),
as well as a myriad of smaller high-tech companies whose
manpower needs draw on the local colleges and universities.
Copyright ChE Division of ASEE 2000

Chemical Engineering Education

The Blackstone River Valley, in and around Worcester, is
considered to be the birthplace of the Industrial Revolution.
Worcester still has some of the heavy industry it was once
known for, including abrasives (Norton) and heavy forgings
(Wyman-Gordon). But much of this low-tech industry has
moved away, mainly to the South. At one time there were
major wire mills (including US Steel), textile operations,
and plastic processors.

Worcester's isolation from Boston has
resulted in cultural facilities that developed
far beyond the level one might expect for
the city's size. These include both the spec-
tacular Worcester Art Museum and
Mechanic's Hall (opened in 1857), one of
the most acoustically perfect and beautiful
concert halls in the nation (the name re-
flects Worcester's industrial beginnings).
The theatre and music seasons provide the
best of internationally and nationally known
performers and orchestras for concerts, re-
citals, operas, and theatre. Worcester is quite
culturally independent of Boston.

Going back in time to the nineteenth cen-
tury, it was well-recognized that local in-
dustry needed a local source of trained en-
gineers to work in the burgeoning light and
heavy manufacturing industry that was suc-
cessfully operating in Worcester. A local
philanthropist, John Boynton, established
an endowment of one hundred thousand
dollars on May 1, 1865, to establish "The
Worcester Free Institute." Thus was estab-
lished the third oldest technological college
in the USA (following RPI and MIT).
"The aim of this school shall ever be the

instruction of

instruction of

youth (boys of Worcester County) in those branches of edu-
cation best adapted to train the young for practical life; and
especially that such as are intending to be mechanics, or
manufacturers, or farmers, may attain the principles of sci-
ence applicable to their pursuits."
Mr. Boynton's gift was quickly supplemented by a gift of
one hundred and ten thousand dollars by Stephen Salisbury
of Worcester. The new Institution was chartered by the
legislature of Massachusetts and opened to receive its first
students on May 12, 1868, as the "Free Institute of Industrial
Science of Worcester, Massachusetts." Farmers seem to have
been neglected in the list of courses offered: Mechanical
Engineering, Civil Engineering and Topography, Architec-
ture, Drawing and Design, Chemistry, and English, French,
and German.
The first catalog stated that "Certain [of the above] studies

are common to all of these Departments, for it is the aim of
the school to give as complete a general education as pos-
sible, and to point out the true relation of theory and prac-
tice." This statement is, in effect, from the very beginning,
the essence of the WPI education. The motto of WPI -Lehr
und Kunst (Science and Technology)-states the school's
intent to provide engineers and chemists to industry who
have not only practical capabilities but also an understand-
ing of the fundamental principles they
will need to advance knowledge. It is
noteworthy that such a philosophy was
.s must new to technical education, which here-
ht not tofore had been almost universally theo-
t retical and which left the new graduate
St short on practical skills. To accomplish
te these goals, the first catalog pointed out
gy, but the need for practicums. Drafting and
design, shop training, and laboratories
ISSeSS were required, so that the graduate
age the would be able to enter his profession
and directly, not as an apprentice who had
nd to learn to apply his theoretical back-
an ground to practice.

be taug
only ht
also to
and main
That api
is knoW
"The 1

rCes of Speeches made at the time (1869) also
t fixed the goals of the school. Chester S.
Lyman of Yale stated, "I see not simply a
ogy." new institution, but a new class of institu-
roach tions." Charles O. Thompson, the first
Professor of Chemistry and Principal,
m I as noted, "It is not the boy we are training,
/TPI but the giant he is to become." Mr.
Thompson also said at that time, "We
cannot receive any women without un-
dertaking to instruct all competent women
who apply. This we have not room for
now. It is our purpose to throw the school open to youth of
both sexes as soon as we can." (This finally happened, but it
took one hundred years until the first female student graced
the halls of WPI.)

There was, of course, no such thing as "Chemical Engi-
neering" in 1868. Industrial chemists and mechanical engi-
neers met industry's needs in the chemical technology area.
The founders not only felt that it was important for engineers
to have an understanding of chemistry, but also that it was
important for chemists to have engineering skills. Besides a
spectrum of courses in theoretical and applied chemistry,
chemistry majors had extensive laboratories in chemistry,
focusing heavily on analysis and physical chemical mea-
surements. But they also took the drafting workshops, ma-
chine shops, and metallurgy and forging workshops required
of the mechanical engineering students. A new building was

Summer 2000





opened in 1887 (Salisbury Laboratories) to house facilities
for chemistry that ultimately evolved into elaborate unit
operations experimentation. It was purported to provide the
chemistry students "with even more air to contaminate."
Proceeding to the catalog of 1898, laboratory and work-
shop capabilities and facilities had advanced considerably,
allowing a philosophy of education in which engineering
students produced the parts for, and actually built, complex
machines. The next step was for the students to actually
produce commercially valuable products, and to understand
the costs involved in their manufacture and sale. (The rev-
enues obtained from the sale of student-produced products,
such as drafting tables, were applied to the costs of education!)
The Chemistry Department had not only General, Analyti-
cal, and Organic laboratories, but also very practical Indus-
trial, Sanitary, and Gas Laboratories. These were comple-
mented by chemistry-major courses in General Chemistry,
Qualitative Analysis, Quantitative Analysis, Advanced Inor-
ganic and Theoretical Chemistry, Organic Chemistry, Indus-
trial Chemistry, and Thesis. The Chemistry Department also
offered Mineralogy (for all students except electrical engi-
neers), Metallurgy (for all students), Sanitary Chemistry (for
chemists and civil engineers), and Gas Analysis (for all
students except civil engineers).
Seniors received sixty lectures describing the most impor-
tant chemical manufacturing processes. Visits were made to
manufacturing enterprises, and large equipment for grind-
ing, mixing, pressing, filtering, drying, etc., was studied,
both at the company's and in Salisbury's laboratory "for

study of chemical processes on a large scale."
The catalog states "The principal work in the laboratory is
not the preparation of pure chemicals, but rather, the study of
practical methods of working up the waste products of vari-
ous manufacturing industries, which are in reality the stu-
dent research projects."
The very practical chemistry degree (without, of course,
actually bearing the name "chemical engineering") offered
in 1898 corresponded very much to the chemical engineer-
ing programs at most schools even sixty years later.

On January 6, 1917, Dr. Robert H. Goddard (WPI class of
1908, and Assistant Professor of Physics at the local Clark
College) was provided a $5,000 grant through the Smithsonian
Institution to further his pioneering work on rockets. He
arranged to work together with WPI staff, and the Magnetic
Laboratory at WPI was rewired to enable testing of the
devices he had made in Clark's Machine Shop. The military
potential of his research was immediately recognized, and
by January of 1918, he was also working with funds pro-
vided by the US Signal Corps. Because of spy paranoia
during World War I, it was necessary to hire watchmen; the
staff of the Magnetic Laboratory working on the Testing
Project grew to about ten. Shots were heard from the small
building for about four months, frightening the neighbors on
a regular basis. A good deal of public curiosity resulted, and

A Dr. Robert Goddard's early experiments on rockets were
done during World War I in the WPI Magnetic Laboratory,
pictured above, which was isolated from other campus

Dr. Robert Goddard is shown at his liquid-fueled rocket-
launch site in Auburn, Massachusetts (1926) N

Chemical Engineering Education

the Signal Corps decided to move the project to Mount
Wilson Observatory shops at Pasadena, California, and sub-
sequently to Aberdeen Proving Ground, Maryland. After the
armistice was signed, funding for the Project ended abruptly,
but the efforts had produced a prototype of the "bazooka"
World War II anti-tank weapon.
Dr. Goddard returned to the Worcester area to forge his
place in history. On March 26, 1926, the first flight of a
rocket using liquid fuel was recorded as having been made
from Aunt Effie's Farm in neighboring Auburn. Testing of
larger rockets followed in short order, making the neighbors
and fire marshal nervous enough to have Goddard exiled to a
far corer of Fort Devens, Massachusetts. By 1930, he was
in Roswell, New Mexico, continuing his research.
There is no record of formal close collaboration between
Robert Goddard and the WPI Chemistry Department, even
though his work on liquid-oxygen-fueled rockets encom-
passed both chemical and chemical engineering principles,
not the least of which were design and construction of pumps
and cooling systems. WPI felt that it was important to recog-
nize Dr. Goddard's great contributions by naming a new
building after him. A grant from the Olin Foundation funded
the construction of this new building for the combined Chemi-
cal Engineering and Chemistry Department (united under one
Department Head in 1940). The new building, named Goddard
Hall of Chemical Engineering and Chemistry, was dedicated
in 1965. It replaced the old Salisbury Laboratories building.

WPI's first course in "Chemical Engineering" was offered
in the Chemistry Department in 1922.* It was a senior course
taught by non-resident lecturer Barnett F. Dodge. In 1928,
Thomas K. Sherwood was appointed as Assistant Professor
of Chemical Engineering. He joined MIT three years later.
By 1936, chemical engineering was established so firmly
that the department changed its name to Chemical Engineer-
ing and Chemistry. Internal jousting resulted in splitting the
combined department into two separate departments in 1938;
the Department of Chemistry and the Department of Chemi-
cal Engineering. The departments merged again in 1940-
and separated again, to go their separate ways (possibly
forever) in 1967.
Once Goddard Hall was opened, the chemical engineering
faculty, with Yankee frugality, moved the antique chemical
engineering equipment from Salisbury to the state-of-the-art
three-story Goddard Hall unit operations laboratory. An
equipment grant allowed the department to purchase sophis-
ticated new pilot-scale experimental units to supplement the
old. A decision was also made to increase the chemical

* WPI had initiated an advanced one-year professional engineer-
ing program in Chemical Engineering in 1917. The first graduate
that year was William Bartlett Jones (BS in Chemistry, 1916).
The first BS in Chemical Engineering was granted in 1935.

engineering faculty in the new building by fifty percent.
Eventually, the new faculty, not burdened by the nostalgia of
the senior faculty, threw out the old equipment.
The reason for increasing the size of the chemical engi-
neering faculty was to implement research activities in the
department, now that facilities were available. This was
quite in accord with new policies that were developing across
all of the departments of the Institute. Under the guidance of
its gifted Dean of Faculty, M. Lawrence Price, WPI's courses
had been completely revised in 1959 to keep up with the
new technologies of the era, such as ultrasonics, aero-
space, nuclear power, magneto hydrodynamics, cryogen-
ics, etc. The problem became what to stop teaching to
make way for the new technologies.
Along with existing technologies, both educational meth-
ods and the mechanical equipment in the laboratories were
becoming obsolete. Concepts, rather than courses, were now
important: how to learn, not what was learned, was the key
point of departure from the past. The need for the type of
practical training exemplified by the foundry and manufac-
turing in Washburn Laboratory had long passed. For ex-
ample, Washburn was now converted to a I-KW pool reac-
tor nuclear facility, the only such facility in the Northeast
dedicated to teaching (and, of course, research). The Physics
Department had installed a Van der Graaf accelerator in its
new building for research in nuclear magnetic resonance and
X-ray diffraction. WPI's Alden Research Laboratories, lo-
cated in nearby Holden, had become both nationally and
internationally renowned as a center for hydraulic research
and development. Forty truckloads of antiquated power equip-
ment were removed from electrical engineering to make way
for the new activities in microwaves, electronics, and comput-
ers. Civil engineering redirected itself to human problems of
environment, safety, health, sanitation, and transportation.
The Chemical Engineering and Chemistry departments
shared the new Goddard Hall, and the new facilities allowed
research that had been started years before to blossom in
both departments. Most notable in chemical engineering
were the catalysis research of Professor Wilmer Kranich (the
Department Head) and the combustion research of Professor
C. William Shipman. Starting in 1965, new faculty were
brought in with a departmental goal of establishing a critical
research and teaching mass in the related and burgeoning
areas of reaction study-catalysis, kinetics, transport, and
zeolites. These included, in order of arrival over a three-year
period: Imre Zwiebel (adsorption), Alvin Weiss (kinetics
and catalysis), Yi Hua Ma (transport and diffusion), Leonard
Sand zeolitee synthesis). They quickly aggrandized all avail-
able laboratory space and as much funding as could be
found, and began to make WPI a well-known name in these
areas of chemical engineering research. The Catalysis Soci-
ety of New England was organized at WPI in 1967, and the
Second International Conference on Molecular Sieve Zeo-

Summer 2000

lites was held at WPI in 1971. The department quickly
became recognized as a center for research excellence in the
above listed fields, particularly for synthesis of zeolites and
for catalytic studies that showed the relationship of acidity to
silica-alumina ratio in zeolites. The 1973 energy crisis brought
major funding in both energy and
environmental areas. Professor
Robert Thompson, hired in 1976,
became Editor of the Zeolites I
Journal (now Non Porous and
Mesoporous Materials Journal).
International research collabora-
tions on food synthesis in space
for NASA established space re-
search as a key activity of the
department. Professor Albert
Sacco, hired in 1977, became an
astronaut and did microgravity
zeolite synthesis research during
space flight. Professors William
Clark and David DiBiasio estab-
lished biochemical engineering
research in the department. Pro-
fessors Anthony Dixon and Wil-
liam Moser were hired for their
respective contributions to trans-
port and catalysis studies.
Endless debates were held in
1968 faculty meetings on the
question of admitting female stu- The Three-story Unit
dents. There were heated discus- has large-sized mi
student exp
sions by a small vocal group at student ex
these meetings, both about inad-
equate lavatory facilities for women and about corruption of
WPI's young men. Quite unilaterally, the Chemistry Depart-
ment admitted a nun to the Master's program. The lady had
the wit to locate lavatories (and laboratories); and her morals
could not be questioned, so the ice was broken. Females now
represent about fifty percent of the students in chemical
engineering, and three of the current ten chemical engineer-
ing professors are women.

An important change in the school's administration took
place when Dean Price was joined by the new President of
the Institute, Lt. General Harry P. Storke (Ret.). These two
men recognized the general state of engineering education at
the time. Strict lists of required courses with inflexible pre-
requisites were the norm in engineering-degree programs
across the nation, the goal being the "weeding out" rather
than the nurturing of students. But, simultaneously, technol-
ogy was changing at an unheard-of pace, and there was no
way for engineering students to learn everything. What was
needed was a new approach to engineering education where


the student learned to learn, to acquire skills, and to function,
even if he or she faced technologies that were unheard of at the
time of their undergraduate education. According to the WPI
catalog, "Students must be taught not only how to create
technology, but also to assess and manage the social and
human consequences of that tech-
nology." The approach is known
as "The WPI Plan." The first "Plan"
0 students graduated in 1973.
The WPI-Plan undergraduate
chemical engineering program
now averages fifty graduates per
year and is centered on three key
projects that distill classroom ex-
perience into real-world projects.
The vision of the WPI Plan is that
the project experiences prepare
students to manage team efforts
and to communicate exceptionally
well, both in oral and written re-
porting. The Major Qualifying
Project is a senior-year chemical
engineering study, either labora-
tory research or a design task (usu-
ally as a subset of an ongoing
graduate research project). The
Interactive Qualifying Project em-
phasizes the need to understand
and to relate how chemical tech-
rationms Laboratory nology affects society and its in-
nentation stitutions. There is now a great
emphasis on global issues. The
project work can be done either at
WPI or at any of the seventeen global program sites located
both in the United States and in foreign countries. All students
also do a project called a "Sufficiency," based on a five-course
self-selected coherent series of courses in the Humanities and
Arts. The goal is to develop an in-depth understanding and
appreciation of one of the cultural aspects of our society.
It seems amazing that this WPI chemical engineering cur-
riculum has existed for three decades. ABET (Accreditation
Board for Engineering and Technology) requirements had
to shift to accommodate the innovations. The new ABET
2000 requirements actually contain many of the WPI
educational elements.
Undergraduates develop their own programs of technical
courses (with, of course, careful faculty guidance). There are
no prerequisites for any courses. Failed courses disappear
from the student's transcript; only successes are recorded.
The student can go on to the next course, repeat the course,
or forget about it. Apart from their projects, chemical engi-
neering students are expected to complete twelve courses in
mathematics and basic science, fifteen engineering science

Chemical Engineering Education

Faculty and Their Research

T.A. Camesano
Assistant Professor, PhD, Pennsylvania State University
Bioremediation; bacterial adhesion; atomic force microscopy; colloidal

W.M. Clark
Associate Professor, PhD, Rice University
Bioseparations; two-phase electrophoresis; aqueous two-phase
extraction. membrane filtration, teaching methodologies
R. Datta
Professor and Department Head, PhD, California, Santa Barbara
Catalyst and reaction engineering; supported molten metal catalysts;
catalytic microkinetics; fuels and chemicals from renewable resources;
fuel cells and reformers; transport in porous media and membranes

D. DiBiasio
Associate Professor, PhD, Purdue University
Bioreactor engineering, magneuc resonance imaging of bioreactors,
mammalian cell culture. hollow fiber reactors, immobilized cell
reactors; teaching and learning methodologies

A.G. Dixon
Professor, PhD, University of Edinburgh
Reaction engineering; computational fluid dynamics for gas-solid
catalytic reactors; dense and porous inorganic membrane reactors;
zeolite membrane reactors; heat-transfer problems in fixed-bed
membrane and microchannel reactors

Y.H. Ma
Professor, ScD, Massachusetts Institute of Technology
Inorganic membranes; palladium membranes for hydrogen separa-
tions; perovskite and perovskite-like membranes for air separation;
zeolite membranes, membrane reactors; adsorbent development,
adsorption and diffusion
K.M. McNamara
Assistant Professor, PhD, Massachusetts Institute of Technology
Chemical vapor deposition; CVD growth processes for semiconduc-
tors; impurity and defect incorporation; optical and electrical
properties; materials for space applications; art and historical objects
W.R. Moser
Professor Emeritus, PhD, Massachusetts Institute of Technology
F.H. Ribeiro
Assistant Professor, PhD, Stanford University
Catalysis and surface science; heterogeneous catalysis; kinetics; model
R.W. Thompson
Professor, PhD, Iowa State University
Applied reactor design and particulate systems; zeolite crystallization;
polymer degradation; water purification; film formation
R.E. Wagner
Professor Emeritus, PhD, Princeton University
A.HL Weiss
Professor Emeritus, PhD, University of Pennsylvania
B.E. Wyslouzil
A ociate Profe sor. PhD, California Institute of Technology
Aerosol science; small-angle neutron scattering from aerosols;
multicomponent aerosol formation; condensation in supersonic
nozzles; aerosol transport in plant tissue reactors

and design courses, and six advanced chemistry courses.
The well-rounded graduates of the Plan are qualified and
sought after for industrial positions of responsibility, gradu-
ate schools, and professional schools. Many WPI graduates
occupy key positions throughout the chemical industry.

The WPI Chemical Engineering Department's graduate
program averages about thirty-five graduate students, as well
as about ten post-doctoral scientists and visiting professors.
The department's ten professors and their fields of research
are listed in Table 1. The department plans to grow to a total
of thirteen faculty, with increasing emphasis on graduate
study and research. Catalysis, zeolites, and reaction engi-
neering continue to be the cornerstones of the department's
research recognition, but other active research areas include
fuel cells, materials science and technology, environmental
studies, life sciences and bioengineering, space sciences and
engineering, and computational modeling. The flavor of
graduate research is highly international, and papers are
presented at conferences throughout the world. Graduate
study also involves a great stress on the student's research
and great independence in course selection. Almost every
project incorporates undergraduate participation, giving the
graduate student experience in leadership.

There is no denying the dramatic changes that are occur-
ring in chemical engineering technology. New instrumenta-
tion and concepts develop constantly. Most basic chemical
and petroleum industry products are now commodities, re-
sulting in limited growth in traditional industries. Although
there is much debate on the future of the discipline, WPI's
approach to chemical engineering of "Learning to Learn"
accommodates present and future contributions of chemical
engineers in new, as well as existing, technologies. The
independence given the student in structuring his or her own
program of courses, in managing and doing teamwork on
projects, in oral and written reporting, guarantees future
leaders in our discipline. The substantial exposure to hu-
manities, other cultures, and international activities and com-
merce, produces well-rounded chemical engineers ready to
function in today's world. Many schools have followed or are
considering anew WPI's example in chemical engineering
education. It is a future path of education for the discipline.

The author wishes to express his thanks for the help given by
Laura Brueck, Archivist of WPI's Goddard Library, in provid-
ing background information for this article. Considerable his-
torical information has also been taken from the book Two
Towers by Mildred McClary Tymeson (Library of Congress
Catalog Card 64-66162). The author appreciates the support
and information provided by Professor Ravindra Datta. O

Summer 2000

ff] educator

Frank Doyle III

Reflections from his friends and colleagues

E dgar Allan Poe is rumored
to have placed a curse on
the city of Newark, Dela-
ware, after a bad experience in a
local watering hole. The curse:
"Anyone who is in Newark can-
not wait to leave, and when you
are gone, you are homesick for
Newark." This legend has spe-
cial significance for Frank
Doyle, who grew up in Newark,
left for college and graduate
school, and after sixteen years
has returned to teach at the Uni-
versity of Delaware.
The origins of Frank's interest
in chemical engineering can be
traced to Newark. First, his father
is a chemical engineer (he is the
Information Technology Manager
at the nearby Motiva Refinery),
which seems to have influenced
both Frank and his younger brother
Patrick. Second, he was drawn to
chemistry and mathematics from as early in his education as
the eighth grade. He recalls attending classes on Saturday
mornings to learn algebra under the watchful eye of Sister
Mary Alice. Frank recalls, "The nuns at my high school
almost talked me out of chemical engineering-however,
the alternative they had in mind was the priesthood."
Although Frank grew up in the shadow of the top-ranked
chemical engineering program at the University of Dela-
ware, he chose to go out of state to Princeton University for
his undergraduate training. His decision was partly based on

the proximity of his parents'
house to the campus, and partly
on his desire to pursue a well-
rounded Ivy League education.
In fact, English was his second
choice for a major.
To balance the challenging
workload at Princeton, Frank be-
came involved in the varsity sail-
ing team and served as its co-
captain in both his junior and se-
nior years. A love of sailing was
also a result of his father's influ-
ence; his father had built his own
sailboat when Frank was only
five. In his senior year, Frank was
awarded the "Class of 1916 Cup"
for the varsity athlete graduating
with the highest GPA.
On the academic front, the big-
gest influence on Frank while at
Princeton was Roy Jackson, with
whom he worked on his senior
thesis. Frank's thesis won the
Xerox Prize for the most outstanding thesis in the chemical
engineering department. Professor Jackson's command of
the English language and his ability to communicate ideas in
the classroom greatly influenced Frank's decision to pur-
sue an academic career, and it set a standard for Frank in
effective classroom teaching. Roy was also a role model
as a mentor, and he taught Frank the valuable lesson of
impartial advising.
In 1985, as graduation neared, Frank kept all his post-
graduate options open as he applied for business school,

Copyright ChE Division of ASEE 2000

Chemical Engineering Education

... he was drawn to
chemistry and mathematics
from as early in his education as the
eighth grade. .. "The nuns at my high
school almost talked me out of chemical
engineering-however, the
alternative they had in
mind was the

engineering graduate school (chemical and nuclear), industrial
positions, and study-abroad programs. "It's good to have op-
tions," he observes. He also weighed a decision to join the
nuclear navy when he was a junior at Princeton. He soon real-
ized, however, that submarines were a poor approximation for
sailing on the surface, and he quickly came to his senses and
decided to go to graduate school instead.
In the end, Frank was drawn to the field of process control as
his graduate research topic. There were a number of events that
steered Frank toward the field of dynamics and control. They
included summer internships in 1981 and 1982 with a process
control instrumentation company (Process Control Inc.), as well
as a senior-year course in process dynamics and control.
After talking to Professor Jackson about his experiences as a
student at Princeton, Frank decided to spend a year at Cam-
bridge University and was awarded a Winston Churchill Fellow-
ship in 1985. He worked with professor Allan Hayhurst on the
development of an ionization sensor for a spark ignition engine.

This thesis project was in some ways a prelude to his
PhD work as his project involved a novel sensor for a
feedback loop in an engine control system.
In Frank's effort to immerse himself in the local Brit-
ish lifestyle, he took up rowing, cricket, and pub hop-
ping. His rowing activities eventually led to participa-
tion on the chemical engineering boat to race in the
Cambridge town bumps. His team went on to "bump"
each of the four nights of the regatta, and each member
was awarded his own oar as a trophy. That oar is Frank's
prized souvenir from his Cambridge days. At the end of
his year there, Frank was awarded the W. Averell
Harriman Scholar prize for being "the most outstanding
Churchill Scholar" in the program in 1986.
Leaving the dreary weather of England for sunny Cali-
fornia, in the fall of 1986 Frank began work on his PhD
at the California Institute of Technology. He chose that
program because of its intimate class size and because of
the international reputation of Manfred Morari's research
program in process control there. Morari taught Frank
about the importance of close interaction with students
as well as how to think critically. Frank practices what
he learned with his own research group by having weekly
one-on-one meetings and weekly student seminars, and
by setting a high standard for the group.
In spite of the grueling schedule of graduate school,
Frank found time to keep his love of sailboat racing
alive. He was fortunate enough to become part of a
three-person team and qualified for the World Champi-
onship in the Echells class while at CalTech.
But, sailing had an even more important consequence
in Frank's life-he met his future wife, Diana, during a
regatta weekend in Newport Beach.

Summer 2000

Members of the Churchill College rowing Team at Cambridge University compet-
ing in a regatta on the river Cam in the fall of 1985. Frank is rowing in the stroke
position facing the coxswain.

In 1991, as Frank was putting
the finishing touches on his PhD,
he accepted an offer to teach at
Purdue University. Recognizing
the practical nature of Purdue's
program, however, he felt he
would be a better teacher if he
had some industrial experience
under his belt before standing in
the front of a classroom. A
chance meeting with W. David
Smith at the CPC IV meeting
led to working for DuPont in
1991, prior to commencing his
career at Purdue. Dr. Tunde
Ogunnaike, a collaborator from
the group, recalls:
My first professional -
contact with Frank began
as a result of a program
instituted in 1990 at
DuPont. In an attempt to
promote more meaningful Princeton Sailing Te
interaction between and Rob Schoelkopf
academia and industry, in the Woodrow Wi
with the objective of them on a 1984 covei
encouraging research
efforts focused on prob-
lems of industrial significance, our group at DuPont
instituted an academic visiting-scientist program.
The first ever "Young Faculty" visiting position
was offered to Frank. By the end of his year in this
program, Frank had established what would
become the gold standard by which all subsequent
program participants would be measured. I had
the privilege of working very closely with Frank
during this one-year period; it was to be the
beginning of a long and fruitful collaboration
that continues to this day.
Nearly ten years after, it is now possible to see
the importance of Frank's year at DuPont in
shaping his career; from his teaching style and his
selection of research problems to the formation and
operation of the University of Delaware Process
Monitoring and Control Consortium. Each activity
bears the unique trademark of strong theoretical
and analytical fundamentals appropriately tem-
pered with practical considerations. Frank has also
evolved into one of only a few academicians who
are well-regarded both in industrial as well as

am Co-(
pull a s
Ison PI
of Prin

Chemical Engineering Education

academic circles; in Frank's
case, he is not only a
productive researcher, he is
also an all-around good
citizen of the chemical
engineering community at
While at DuPont, Frank learned
about practical problems and the
challenges in implementing con-
trol in the real world. He collabo-
rated with Tunde and Ron
Pearson on Volterra series model-
based control; they have contin-
ued those interactions through the
years and are presently writing a
book on the subject. Frank also
kindled an interest in biosystems
control through a collaborative
project with Tunde and Jim
Schwaber, and that collaboration
S has likewise been maintained
through the years despite several
employment changes by both Jim
Captains, Frank Doyle and Frank. They studied the way
stunt with windsurfers that nature regulates blood pres-
aza pool that landed sure on a beat-by-beat basis; that
ceton Alumni Weekly. project led to a major thrust of
biosystems analysis and control
in Frank's research-several
years before the DuPont Life Science revolution took place.
In the fall of 1992, Frank headed to Purdue to start teach-
ing and building his own research group. His research pro-
gram took root quickly, and he was awarded the National
Science Foundation's National Young Investigator Award
for his process control work in 1992. This was followed
in 1996 by the Office of Naval Research Young Investi-
gator Award for his research on biosystems analysis and
Purdue introduced a new teaching challenge for Frank-a
large classroom environment. But, he took it as an opportu-
nity to establish his lectures with multifaceted ways of inter-
acting with his students through a combination of lectures,
labs, help sessions, and office hours. He organized a tutorial
session in the evenings with the dual purpose of going over
some complex problems that would capture the attention of
the accelerated students and solving them in complete detail
for the benefit of students who need a little extra help.
Frank was greatly influenced in his teaching style by Phillip
Wankat, who taught a course at Purdue on teaching engi-
neering. In that class, Frank first learned about the "learning
types" that helped him in approaching diversity in the class-

room. His hard work and dedication paid off. Frank was awarded many teaching
awards in the next few years, including ASEE Section Outstanding Teacher Award
(Illinois/Indiana, 1996); Shreve Prize, Chemical Engineering Department teaching
award, 1995, 1997; A.A. Potter Award, School of Engineering teaching award,
1995; Purdue's Teacher's for Tomorrow Award, 1996; and Tau Beta Pi's Dean
Marion B. Scott Exemplary Character Award, 1996. In 1996 he was also inducted
as a member of the first group of elite teaching faculty in the Teachers-for-
Tomorrow program.
Frank also began development of a set of educational software modules while at
Purdue. With the encouragement of his departmental head, Rex Reklaitis, he
created a virtual laboratory for process control-constrained by the large class sizes
at Purdue (as high as 200 graduating seniors). After a number of years of hard work
by his graduate and undergraduate students, the modules were published as a book
in 1999 by Prentice Hall, under the title of Process Control Modules-A Software
Laboratory for Process Control.
In addition, Frank established a significant program on both traditional process
control and in the area of biosystems control that he had studied at DuPont. The
traditional process control research was expanded by interactions with the other
process systems engineering faculty at Purdue (Reklaitis, Pekny,
Venkatasubramanian). The four faculty co-founded the very successful Computer
Integratred Process Operations Center (CIPAC). One of the themes that was cham-
pioned by Frank was the application of advanced systems engineering methods to
the pulp and paper industry. He would later place two of his Purdue PhD students at
top pulp and paper companies, and has established himself as a leading researcher
in that industry. Dr. Philip Wisnewski, one of Frank's first PhD students (presently
at Weyerhaeuser) recalls:
Frank has influenced me, as he has many of his students, by the example
of professionalism that he exudes. He sets very high standards for himself;
and through his dedication to teaching and research; through his sincer-
ity, honesty, and loyalty in dealing with others; and through his support
and encouragement which he freely offers to his students, he inspires those
around him to set high standards and expectations for themselves. Frank
is truly a gentleman, a person of quality and integrity.
Frank also expanded his biosystems research activities to include biomedical-
engineering control problems and began a collaboration with Nicholas Peppas in
1994 with two jointly advised graduate students in the area of controlled drug
delivery for diabetic patients. One of those students, Robert Parker, is now an
Assistant Professor in the Chemical Engineering Department at the University of
Pittsburgh. Professor Peppas offers these words to describe his colleague:
Frank is a truly unique individual, an outstanding researcher, and a most
gifted teacher and educator. He brings a great deal of creativity and
ingenuity to the classroom. By bringing his real chemical engineering
experiences to the students, instead of just textbook examples, he not only
helps them to learn but also to want to learn even more. For Frank, teaching
is not a job, but a mission.....Our research interaction over the past six years
has taught me that he is one of the most imaginative and innovative chemical
engineers of our times. He has an uncanny ability to grasp difficult biomedical
ideas and transfer them to the level of systems theory.
On a personal note, although Frank was far from the coast while he was a Purdue,
he still sailed as often as he could on Lake Michigan, often accompanied by his wife
Diana and his daughters Sara and Brianna.

is not


a productive


but he

is also






of the



Summer 2000

Family Album

In 1997, Frank took advantage of an opportunity to move
closer to home; he accepted a faculty position at the Univer-
sity of Delaware. It was a rare opportunity to advance both
his professional and personal pursuits. Here, Frank contin-
ues to make a major impact on innovation in teaching. In the
spring of 1998, he was one of several instructors to beta-test
a new program for a web-based class organizer called SERF
(Server-side Educational Records Facilitator). SERF allows
students to view curriculum, homework questions, send e-
mail to the class and to the professor, check on their progress
in the course, and review lecture notes. Frank has published
conference papers and a book chapter about his experiences
using SERF for control education. He was also inducted as a
Fellow in the NSF-sponsored Institute for Transforming Un-
dergraduate Education at the University of Delaware.
Frank was also the first instructor at Delaware to employ
live video streaming via the Worldwide Web in his fall 1998
control class. He worked with the staff of Media Services to

create an on-line version of his course to facilitate the sched-
ules of students who were on industrial internships as well as
to engage a more geographically distributed continuing edu-
cation audience. The experience formed the basis of an
invited book chapter on the subject. A student taking the
class off-site said
It has given me the knowledge and insight into the
subject that would have taken me several years to
acquire on my own. Prof Doyle was very receptive
to student questions both in and out of class and
was very professional.
As mentioned earlier, Frank has also developed an interac-
tive software package for undergraduate dynamics and con-
trol education known as PCM (Process Control Modules).
The modules were published in the fall of 1999 as a textbook
with the CD-ROM. The software is developed in the
MATLAB and Simulink environment and incorporates sev-
eral realistic simulation models of industrial unit operations
(furnace, distillation column, bioreactor, pulp digester, chemi-

Chemical Engineering Education

Frank, at one year of age
(top) and all grown up,
the centerpiece of three
generations of Doyles
(bottom). V

Frank has
to his
Here with
two-and-a-half year old daughter Sara (top)
and one-year-old Brianna (bottom) aboard
their sailboat "Son of a Sailor" in 1998.

cal plant). The accompanying exercises demonstrate dy-
namic analysis, PID controller design, frequency response
analysis, and controller tuning. There are numerous chemi-
cal engineering departments in the world using this software
for process control education.
Another of Frank's teaching innovations at the University
of Delaware was the development of a college-wide interdis-
ciplinary, experimental control-engineering course. The
course is centered around a state-of-the-art laboratory devel-
oped using actual industrial control equipment and software
from Aspen Technology, ABB/Bailey, and Honeywell. The
experiments include a distillation column, a gyroscope,
an inverted pendulum, a servomotor, a level control
system, and a spring-mass damper. The course has drawn
students from chemical, electrical, and mechanical engi-
neering as well as from operations research, which is
outside the college.
Frank's educational activities were recognized in 2000 by
the Ray Fahien Award from the Chemical Engineering Divi-
sion of the ASEE. He is also serving on the Provost's Teach-
ing and Learning Technology Roundtable, which has a mis-
sion to define the university's vision for effective teaching
technology tools.
The work of Frank's research group at Delaware is charac-
terized by two dominant themes: 1) the application of non-
linear model-based control techniques to multivariable, non-
linear, constrained industrial processes, and 2) the use of
systems-analysis tools as a bridge between chemical engi-
neering and biology through neuromimetic and therapeutic
approaches. At Delaware, Frank has also taken on a new
research challenge-an experimental project. His group built
a pilot-scale emulsion polymerization reactor shortly after
their arrival that is interfaced to state-of-the-art industry
sensors and control hardware. With an objective of particle
size distribution control, it will serve as a test bed for a
number of control projects in his research program. Frank
feels there is no simulation substitute for the complex non-
linear behavior that his research team will face in trying to
optimize the operation of this system.
The practical impact of Frank's research is also evidenced
by establishment of the University of Delaware Process Con-
trol and Monitoring Consortium in 1998, shortly after his
arrival. Frank is the director of this Consortium that has the
support of twelve industrial companies and involves col-
laborations across campus, including chemical engineering,
mechanical engineering, computer science, and computer
engineering. The Consortium structure enables participation
of both industrial companies and control vendors with the
university in the pursuit of applied research.

Frank maintains a healthy perspective about service work

in the academic profession: " is the responsibility of the
most active members of the teaching and research commu-
nity to take a leadership position in service activities." As
such, he has served on a number of important committees,
including the graduating recruiting activities at both Purdue
and Delaware. He has served on many search committees,
including three concurrently in early 2000. He holds two
editorial posts with the Journal of Process Control (Special
Papers Editor) and IEEE Transaction on Control Systems
Technology (Associate Editor). He has organized or chaired
over thirty sessions at meetings of the AIChE or ACC and is
on the International Programming Committee of many of the
conferences in the process control area. His teaching service
has led him to give lectures to high school teachers on
technology topics as well as to the members of the Academy
for LifeLong Learning (a continuing education group for
retirees) on diabetes therapy research.
Although Frank is a young faculty member, he has already
had a large influence on the students in the profession. He
estimates that approximately 900 seniors have taken his
process control course, and of those, approximately 40 have
done independent research projects under his supervision.
He is presently advising, or has supervised, over 20 MS and
PhD students in his relatively short career.

Frank was born in Philadelphia in 1963, but moved to
Newark, Delaware, at the age of 2. He was the oldest of five
children and has three sisters and a brother (Patrick-also a
chemical engineer). Frank married his wife Diana in 1992
and they have three beautiful children: Sara (age 4), Brianna
(age 3), and newborn Francis Joseph IVth. Now that Frank is
"back home" in Delaware, he enjoys sailing the Chesapeake
with his parents and hopes to introduce his children to the
excitement of sailboat racing. Another perk of returning
home is the renewal of hope for ending a 25-year losing
streak on the tennis court to his father. He is involved
with the Knights of Columbus in the Parish where he
went to grammar school.

Frank sees many changes ahead for the field of chemical
engineering, both in teaching and research. There is no ques-
tion in his mind that future research successes will occur at
the boundaries between disciplines, and he feels confident
that biosystems and chemical engineering will be a fruitful
path. On the educational front, he sees similar challenges as
the new ABET requirements put an increased emphasis on
interdisciplinary coursework. He is presently working with
his Dean and the department heads in the college to create a
common control engineering course for all the engineering
majors. He also has plans to tap the power of the World
Wide Web to create tools for process control education. 0

Summer 2000

Special Feature Section


Part 5. Assessing Teaching Effectiveness

and Educational Scholarship

Richard M. Felder North Carolina State University Raleigh, NC 27695-7905
Armando Rugarcia Iberoamericana University Puebla, Mexico
James E. Stice University of Texas Austin, TX 78712-1062

T he first four papers in this series[' 4] offered a number
of ideas for effective teaching and preparing faculty
members to teach. An inevitable question is, how
does one determine whether or not a faculty member's teach-
ing is effective? Another important question is, how does
one determine whether or not an instructional program-
such as that of an engineering department-is effective?
The instructional component of the mission of every edu-
cational institution is to produce graduates with satisfactory
levels of knowledge, skills, and attitudes.'" The specific
knowledge, skills, and attitudes may differ from one depart-
ment to another and the definition of satisfactory may differ
from one institution to another, but the instructional mission
is invariant. In engineering, the basis of a department's ac-
creditation is the extent to which the department is fulfilling
this mission. An instructor may be a brilliant lecturer with
student ratings at the top of the charts, but if his or her
teaching is not furthering the instructional mission of the
department, that teaching cannot be considered effective.
To appraise programmatic teaching effectiveness, we must
answer the following questions:'5'6'
1. Educational goals. What are the published goals of the
instructional program? Does the faculty know what they are?
Does the faculty generally agree with them?
2. Performance criteria. Are the criteria that will be used to
evaluate faculty performance measurable and clearly tied to
the goals? Does the faculty know what they are? Does the
faculty generally agree with them?
3. Assessment process. What assessment data will be col-
lected? How and when and by whom will they be collected
and analyzed? Are available resources (including faculty
time) adequate to permit their collection and analysis?
4. Evaluation process. How will conclusions about teaching
effectiveness be inferred from the data, and by whom? What

type of feedback will be provided to the faculty, and when and
by whom will it be provided?

The answers to these questions should be based on the
university mission statement and program accreditation re-
quirements, with additional criteria and procedures contrib-
uted by the program administration and faculty.
An additional factor enters into the appraisal of an indi-
vidual faculty member's teaching performance-namely, the
extent to which he or she is contributing to the improvement
of education. We refer to this performance factor as educa-
tional scholarship. It encompasses developing or systemati-
cally improving teaching methods and methods of assessing
learning outcomes, writing textbooks and courseware, and
publishing scholarly papers and monographs and giving work-

Richard M. Felder is Hoechst Celanese Professor (Emeritus) of Chemi-
cal Engineering at North Carolina State University. He received his BChE
from City College of New York and his PhD from Princeton. He has
presented courses on chemical engineering principles, reactor design,
process optimization, and effective teaching to various American and
foreign industries and institutions. He is coauthor of the text Elementary
Principles of Chemical Processes (Wiley, 2000).
Armando Rugarcia graduated from the Universidad Iberoamericana
(UIA) in 1970 and went on to earn his MS in chemical engineering from
the University of Wisconsin in 1973 and his Doctorate in Education from
West Virginia University in 1985. He has been a full-time professor of
engineering at UIA since 1974 and was Chair of the Chemical Engineer-
ing Department there from 1975 to 1980. He was also Director of the
Center for Teaching Effectiveness at UIA from 1980 until 1986. He has
written four books on education, one on process engineering, and more
than 130 articles.
James Stice is Bob R. Dorsey Professor of Engineering (Emeritus) at the
University of Texas at Austin. He received his BS degree from the
University of Arkansas and his MS and PhD degrees from Illinois Institute
of Technology, all in chemical engineering. He has taught chemical
engineering for 44 years at the University of Arkansas, Illinois Tech, the
University of Texas, and the University of Wyoming. At UT he was
Director of the Bureau of Engineering Teaching Center and initiated the
campus-wide Center for Teaching Effectiveness, which he directed for 16
Copyright ChE Division of ASEE 2000

Chemical Engineering Education

Future of Engineering Education

shops and seminars on education-related topics. For individual faculty performance
evaluation (as opposed to instructional program evaluation), the questions listed above
should therefore be augmented by these:
5. Educational scholarship. What evidence of scholarly contributions to education will be
collected? How and by whom will the evidence be evaluated?
In this paper we suggest options for answering most of these questions. We first
propose principles of instructional assessment and summarize common violations of
these principles. Then we elaborate on how to assess the effectiveness of both teaching
and educational scholarship, leaving the evaluation process (determining what qualifies
as satisfactory performance) to be determined by institutional norms and values.


In the educational literature, the two terms assessment and evaluation are constantly
encountered. They are sometimes used interchangeably as synonyms for appraisal of
instructional effectiveness; sometimes assessment denotes the appraisal of individual
teaching and evaluation the appraisal of teaching programs;1'61 and sometimes assess-
ment denotes collecting and analyzing data that reflect on teaching quality and evalua-
tion denotes interpreting the assessment outcomes and drawing conclusions about teach-
ing quality.17 Unless otherwise noted, we will use the latter definitions in our discus-
An important distinction is that between formative assessment, which has improve-
ment of teaching as its objective, and summative assessment, which produces informa-
tion that can be used to make decisions about instructional personnel or programs.
Formative assessment is (or should be) an important part of institutional programs to
help faculty members become more effective as teachers, a topic discussed in the
preceding paper in this series.141 This paper concerns summative assessment.


Evaluation of either programmatic teaching effectiveness or individual faculty mem-
ber performance involves assessing the quality of instruction in individual courses.
Extensive research supports the use of the following criteria as a basis for the assess-
ment:1 '-51
1. The course contributes toward published program goals.
2. The course has clearly stated measurable learning objectives."2
3. The assignments and tests are tied to the learning objectives and are fair, valid, and reli-
4. Appropriate methods have been devised to monitor the effectiveness of the instruction.
5. The learning environment is appropriate.1 [1
6. The instructor has appropriate expertise in the course subject.
7. The instructor communicates high expectations of students and a belief that they can meet
those expectations, interacts extensively with them inside and outside class, conveys a strong
desire for them to learn and motivates them to do so.
8. The instructor seeks to provide an education in the broadest sense of the word, not just
knowledge of technical content.1'1
9. The instructor integrates teaching with research.
10. The instructor continually attempts to improve the course by updating the content and/or
making use of new instructional materials and methods (including applications of instruc-
tional technology).
11. The students achieve the learning objectives.
More details are given by Woods.[151

In this paper

we .propose

principles of


assessment and


common violations

of these

principles. Then

we elaborate

on how to assess

the effectiveness of

both teaching

and educational


leaving the


process... to be

determined by


norms and


Summer 2000

Special Feature Section


An assessment plan should involve assembling several
types of evidence to determine the degree to which the
foregoing criteria are being met. Among the possibilities are
the following:
SLearning outcomes assessments: student performance on stan-
dardized tests, comparisons of student performance with per-
formance of control groups, evaluations of student products by
external reviewers.
SStudent end-of-course ratings.
SStudent surveys, focus groups, or interviews directed at speci-
fied criteria.
SRetrospective student ratings of courses and instructors (e.g.,
pre-graduation ratings by seniors).
Alumni ratings of courses and instructors.
Peer ratings of classroom instruction, learning objectives, as-
signments and tests.
Evaluations submitted by external referees.
Self-evaluations by instructors.

The assessment data may be collected for individual fac-
ulty members in teaching portfolios (or teaching dossiers),
which may be evaluated by a review team to provide an
effective assessment of instructional effectiveness. The port-
folios assembled by all members of a department collec-
tively provide a partial basis for evaluating the effectiveness
of the department's instructional program. More would have
to be done to demonstrate that the program graduates meet
specified criteria related to their knowledge, skills, and atti-
tudes (such as those specified as Outcomes 3a-3k of ABET
Engineering Criteria 2000).

Assessment of Learning
The ultimate assessment of teaching is assessment of learn-
ing. Teaching that does not satisfy institutional, departmen-
tal, and individual instructors' learning objectives cannot be
considered effective, regardless of what other assessment
measures may indicate.
The past decade has seen a growing realization that the
traditional assessment tool used in undergraduate engineer-
ing education for most of the past century-the written ex-
amination on material covered in lectures and readings-
provides an inadequate measure of the knowledge, skills,
and attitudes that engineering schools wish to impart to their
students. Driven in large part by the impending adoption of
Engineering Criteria 2000 as the accreditation system for all
U.S. engineering departments, a large and constantly grow-
ing body of work on the systematic assessment of specified
learning outcomes has arisen. A full review of this literature
is well beyond the scope of this paper; what follows is a brief
summary of the principal ideas.

Assessment-whether of learning or teaching, whether
for individual courses or entire instructional programs-can
only be done meaningfully in the light of clearly stated goals
and measurable objectives. In the case of assessment of
learning, the requirements are explicit statements of the
knowledge, skills, and attitudes that the students are sup-
posed to acquire (the goals) and of what the students must do
to demonstrate that the goals have been met (the objec-
tives). The following assessment tools may be used as
part of that demonstration. The terms in parentheses indi-
cate the categories of objectives that the specified tools
may be used to assess, including outcomes specified by
Engineering Criteria 2000.

El Complete tests and individual test items (knowledge,
conceptual understanding, engineering problem-solving
skills). Tests given in engineering courses may provide good
measures of relative learning among students in a particular
class, but they are frequently unsuitable for assessment of
true conceptual understanding and problem-solving skills.
They may also provide misleading results. For example, if
tests are too long for most students to finish (a situation that
unfortunately characterizes many engineering tests), students
who work sloppily but quickly may earn much higher grades
than students who work accurately but slowly. The most
meaningful assessment is provided when the test results may
be compared with established norms or with results from
comparison groups, such as another class taught in parallel
to the one in question by a different instructor and/or using a
different instructional method. The nationally normed Fun-
damentals of Engineering (FE) examination has the poten-
tial to provide a basis for assessment.1'61

El Laboratory reports, design project reports, live or
videotaped oral presentations, research proposals (knowl-
edge, conceptual understanding, analysis, creative thinking,
critical thinking, experimental design, identification of engi-
neering problems, teamwork, written and oral communica-
tion skills, professional or social awareness, lifelong learn-
ing skills). The usual drawback of reports as assessment
instruments is subjectivity in their evaluation. One way to
improve their effectiveness is to use detailed checklists in
evaluating the reports, tying the checklist items to specific
learning objectives. Even greater assessment validity is pro-
vided by using several independent raters who reconcile
their ratings after completing their checklists.

E Resumes, letters, memos (written communication skills,
professional or ethical awareness). An effective way to
prepare students to function as professionals is to ask them
to engage in common professional activities and provide
them with feedback on their efforts. For example, periodi-
cally ask engineering students to prepare resumes and to
write letters and memos dealing with common hypothetical

Chemical Engineering Education


Future of Engineering Education

situations, such as reporting a result to a supervisor, asking
for an interview with a prospective employer, persuading a
client or a prospective client to purchase a product or ser-
vice, or recommending an action to a superior or a subordi-
nate in a situation that has ethical implications.
E Critiques of technical reports, papers, letters, and
memos (analysis, critical thinking, written communication
skills). It is often easier to see weaknesses in someone else's
work than in one's own. Having students critique one
another's first drafts of written documents and revise their
own documents based on the feedback they get helps them
develop critical thinking skills, especially if the critiques are
collected and graded. The papers handed in to the instructor
are generally much better than they would have been with-
out the preliminary feedback, and the grading job of the
instructor is consequently much less burdensome.
LI Self-evaluations, learning logs, journals (any skills or
attitudes). Surveying or interviewing students is a direct way
to obtain their impressions of how much their skills have
improved as a consequence of their education. The validity
of such data is greatest if the data are consistent with results
obtained by other means, or if the same data are available for
comparison groups subjected to different forms of instruc-
tion. Student learning logs or journals can be rich indicators
of the degree of acquisition of selected skills and attitudes,
but trained evaluators are needed to make such inferences
and the process can be extremely time- and labor-intensive.
E Other classroom assessment techniques (any skills or
attitudes). The classic reference on classroom research by
Angelo and Cross"'7] suggests a large variety of techniques
for assessing knowledge, recall, understanding, and ability
to apply learned information; skills in analysis and critical
thinking, synthesis and creative thinking, and problem solv-
ing; and self-awareness as learners. While the usual applica-
tions of these techniques are formative, any of them may
also be used for summative assessment.
A comprehensive picture of student learning is provided
by assembling student portfolios-longitudinal records of
student learning assessment results. Panitz1181 describes uses
of portfolios for both formative and summative purposes at
different schools. Some instructors allow students to deter-
mine how much weight should be assigned to different course
components, assemble the portfolios themselves, indicate
the grade they think they have earned, and write a statement
indicating how the portfolio contents justify the grade. Oth-
ers set up competency matrices of one type or another. One
format consists of rows for different student products in the
portfolio and columns for specific learning outcomes or
objectives, with marks to show which products demonstrate
which outcomes or the levels (A, B, C,...) at which the
objectives are satisfied. Rogers and Williams1191 describe a

web-based portfolio system created at the Rose-Hulman In-
stitute of Technology. Students enter work that they believe
demonstrates their progress toward meeting specific perfor-
mance criteria and state justifications for their claims, and
faculty raters evaluate the entries.
Student Ratings of Instruction
The most common method-and in many programs, the
only method-of assessing instructional quality is to collect
student ratings at the end of each course. The rating form is
often haphazardly designed, and the results may be difficult
to interpret with any degree of objectivity. In part because of
these defects, many faculty members discount the validity
and value of student ratings. Commonly heard criticisms are
that ratings do not correlate with quality of learning and the
easiest teachers get the highest ratings.
In fact, more than a thousand research studies of student
ratings have been performed, and the results collectively
show that ratings are reliable, stable over time, and posi-
tively correlated with results obtained using other forms of
teaching assessment, including assessment of learning out-
comes.12"-23] Contrary to popular assertions, they are not af-
fected appreciably by the instructor's personality or gender
or the time of day a class is offered.121 Difficult courses that
do not require unreasonable expenditures of time and effort
are rated somewhat more favorably than courses that lack
challenge."22 Some studies show positive correlations be-
tween ratings and grades, but it is not clear whether the
higher grades in the more highly rated courses reflect inap-
propriately easy grading or superior learning. The positive
correlations observed between ratings and learning outcomes
suggest that the latter may be a strong contributing factor.
Their validity notwithstanding, student ratings should not
be the only method used to assess instructional quality.
There are several important aspects of teaching that students
lack the knowledge and perspective to judge fairly, includ-
ing the currency and importance of the course content, the
instructor's understanding of the subject, and the appropri-
ateness of the assignments, tests, and grading policies.[22241
Many institutions use non-standardized assessment instru-
ments and fail to take into account extraneous factors such as
class size, course level, and whether courses are required or
elective, making the results for different faculty members
difficult or impossible to compare.
Nevertheless, course-end student ratings are an essential
component of instructional quality assessment. As long as
they are to be collected, certain steps should be taken to
maximize their effectiveness. 251
E[ Collect ratings of the effectiveness of the course and
the instructor in a few critical aspects. The most com-
monly used format is probably the five-point Likert scale
(e.g., l=strongly disagree, 2=disagree, 3=neutral, 4=agree,

Summer 2000

Special Feature Section

5=strongly agree) applied to items related to the quality of
teaching and learning in the course. The following items
have been shown to be related to teacher effectiveness as
measured by mean student performance on examinations:[261
Each class period was carefully planned in advance.
The instructor presented the material clearly.
The professor made students feel free to ask questions,
disagree, express their ideas, etc.
The professor used examples from his/her own research or
This course has increased my knowledge and competence.

Other questions might be asked related to acquisition of
specific skills included in the course goals (e.g., critical or
creative thinking, writing, teamwork, etc.). Since a standard-
ized form is desirable for summative assessment, however,
the items chosen should be small in number and general
enough to apply to different courses and instructors within a
single discipline and across disciplines. (For formative as-
sessment, items may be included on any aspect of the
instruction on which the instructor wishes feedback.) The
form should not contain questions about things the stu-
dents are not equipped to evaluate, such as the instructor's
knowledge of the subject.
El Collect overall course-end ratings of instruction. "Rate
the instruction you received in this course on a scale from 1
to 5, with 1 being the highest response." Ratings of this sort
are most effective when the numbers on the response scale
are clearly defined. Definitions like "excellent," "above av-
erage," "fair," etc., are subjective and ambiguous, and when
they are used a very broad performance range tends to be
lumped into "above average." Greater discrimination is ob-
tained by giving descriptions of the characteristics of in-
structors in each category, making it clear that very few
instructors are likely to fall into the extreme categories.
One approach is to use a norm-referenced system, wherein
5 means that the instructor is one of the three best teachers
the student has ever had (or is in the top 1% or the top 5%), 1
signifies one of the three worst teachers (or the bottom 1% or
5%), and 2, 3, and 4 represent different percentile ranges
(e.g., bottom 20%, middle 60%, and top 20%). The problem
with this system is that it penalizes faculty members in
departments with a large number of excellent instructors. A
better approach calls on students to base their overall rating
on the average of their ratings of individual characteristics of
the course and instructor (previous bullet). For example, the
students could be asked to total their ratings of the individual
items, and ranges for the total could be given on the form
that translate to overall ratings of 5, 4, 3, 2, and 1. The ranges
corresponding to the highest and lowest overall ratings should
be relatively narrow (e.g., a total that would yield an average
rating in the range 4.75 to 5.0 might correspond to an overall

rating of 5; 3.75 4.75 to a rating of 4; 2.25 3.75 to a 3; 1.25
- 2.25 to a 2; and 1 1.25 to a 1). If this system were used,
instructors who get 5 would clearly be worthy of nomi-
nation for an outstanding teacher award and instructors
who get 1 would clearly have very serious problems with
their teaching.
E Administer and collect course-end ratings in a single
class session rather than counting on students to return
them later. Results of evaluations for which the return rate
is less than a minimal percentage should be regarded with
deep suspicion: the recommended minimum is 50%
(classes of 100 or more), 66% (50-100), 75% (20-50),
and 80% (<20).127] The environment used for gathering
the data should include student anonymity and absence
of the instructor from the room.
E Interpret ratings collected over a period of at least two
years. One semester of low ratings (or high ratings, for that
matter) does not provide a valid measure of an instructor's
teaching effectiveness.
EL Periodically collect retrospective student evaluations
in addition to course-end ratings. Ratings from seniors
and alumni of how well individual instructors helped them
acquire knowledge and develop skills are powerful indica-
tors of teaching effectiveness. These retrospective ratings
help identify the relatively small percentage of instructors
whose students only appreciate their effectiveness as teach-
ers years after taking their courses. For faculty members at
research universities, ratings from former research advises
attesting to the degree to which professors promoted their
intellectual curiosity and research skills should also be sought.

Peer Ratings
Peer ratings can contribute significantly to the evaluation
of teaching if they are well designed and conducted, but the
common practice of having untrained faculty members sit in
on a lecture and make notes on whatever happens to catch
their attention yields results that are neither reliable nor
valid.128 To be effective, summative peer ratings should
include the features described below.[29,301
E Who should do the reviewing? Reviewers should be
good teachers who have received training on what to look
for in a classroom and who recognize that different styles of
teaching can be equally effective. Training dramatically in-
creases the likelihood that evaluations from different re-
viewers will be consistent with one another (reliability) and
with accepted standards for good teaching (validity).
a How should classroom observations be performed? At
least two reviewers should conduct at least two class visits
during a semester, preceding each visit with a brief meeting
at which the instructor provides information about the class
to be observed. The reviewers independently complete stan-

Chemical Engineering Education

Future of Engineering Education

dardized rating checklists after each observation and soon
afterwards visit with the instructor to discuss their observa-
tions and invite responses. After all individual observations
and reviews have been completed, the reviewers compare
and reconcile their checklists to the greatest extent possible
and write a summary report to be placed in the instructor's
teaching portfolio or personnel file.
[ What should the lecture observation checklist con-
tain? The checklist is a collection of statements about the
observed classroom instruction with which the reviewers
indicate their levels of agreement or disagreement, adding
explanatory comments where appropriate. Statements such
as the following might be included:1311
Organization. The instructor (a) begins class on time, (b)
reviews prior material, (c) previews the lecture content, (d)
presents material in a logical sequence, (e) summarizes main
points at the end of the period, (f) ends class on time.
Knowledge. The instructor (a) has a good understanding of
the course material, (b) integrates ideas from current research
and engineering practice into the lectures, (c) answers
questions clearly and accurately.
Presentation. The instructor (a) speaks clearly, (b) holds the
students' attention throughout the period, (c) highlights
important points, (d) presents appropriate examples, (e)
encourages questions, (f) seeks active student involvement
beyond simple questioning, (g) attains active student
involvement, (h) explains assignments clearly and thor-
Rapport. The instructor (a) listens carefully to student
comments, questions, and answers and responds construc-
tively, (b) checks periodically for students' understanding, (c)
treats all students in a courteous and equitable manner.
Many other statements could be included, some of which
might be particularly applicable to laboratory or clinic set-
tings. Examples of validated observation instruments are
given in a recent book edited by Seldin.131'
E[ How should instructional materials be rated? Exami-
nation of instructional objectives, lecture notes, assignments,
tests, and representative student products may provide a
better picture of teaching effectiveness than classroom ob-
servation. Trained observers can judge whether (a) the
objectives cover a suitable range of knowledge and skills,
(b) the course content is sufficiently comprehensive and
current, (c) the assignments and tests are appropriately rigor-
ous, fair, and consistent with the stated objectives. As with
classroom observation, the ratings should be done by two or
more independent observers using a validated checklist and
reconciled to arrive at a consensus rating.
The Teaching Portfolio
The teaching portfolio (or teaching dossier) is a device
used for assessing the teaching effectiveness of an individual
faculty member, as opposed to effectiveness of instruction in

a single course or of an instructional program. The portfolio
is a summary of teaching assessment data, including self-
assessment. Most authors who discuss portfolios132-381 do so
in the context of formative assessment and recommend cus-
tomizing the portfolio to fit the strengths and objectives of
the individual faculty member. In keeping with the theme of
this paper, we will confine our discussion to summative
assessment, which requires using a standard format to pro-
vide evaluative consistency.
A recommended format for a summative portfolio consists
of several parts:
E Preamble. Context of the portfolio, time period covered,
and outline of the contents.
[L Reflective statement of teaching philosophy, goals,
and practices. The instructor's answers to such questions
as: "What is my mission as a teacher?" "What skills and
attitudes should I be helping my students develop?" "What
methods am I using in and out of class to fulfill my mission
and enable my students to develop the desired skills and
attitudes?" "What am I doing to motivate and equip them to
succeed, academically, professionally, and personally?"
E Summary of teaching and advising responsibilities.
Titles, levels, contact hours, and class sizes for all courses
taught over the past five years, annotated with brief com-
ments about the way each course is taught. Number of stu-
dents advised and comments about the nature of the advis-
ing. Comments should relate explicitly to the reflective state-
ment and to published institutional and departmental goals.
E Representative instructional materials and student
products. Illustrative assignment statements and tests with
grade distributions. Copies of outstanding and typical graded
assignments, tests, and project reports. Discussion of the
materials in the context of the reflective statement.
E Evidence of teaching effectiveness. Results of student
ratings in the context of average departmental ratings for the
same courses over the past six years. Results of retrospective
senior and alumni ratings and peer ratings. Results of learn-
ing assessments, including student performance on standard-
ized tests. Data from instruments that assess approaches to
and attitudes toward learning such as the Lancaster Ap-
proaches to Studying Questionnaire and the Course Percep-
tions Questionnaire13'15'391 and the Perry or King/Kitchener
Inventory.[36'151 Reference letters from students and
alumni. Implications of the evidence in the context of
the reflective statement.
a Efforts to improve teaching effectiveness. Steps taken
to keep knowledge of course content and effective instruc-
tional methods up-to-date: workshops, seminars, and confer-
ences attended, papers read, networking done. Steps taken to
obtain student feedback and to monitor and improve the

Summer 2000

Special Feature Section

learning environment and quality of classroom instruction.
EL Teaching innovations. New courses developed and
changes made to existing courses. New instructional materi-
als generated, teaching strategies adopted, and methods used
to motivate and empower students. Copies of publications or
presentation abstracts describing innovations. Discussion of
the innovations in the context of the reflective statement.
El Evidence of effectiveness of advising and mentoring.
Successes of and recognition received by advises. Refer-
ence letters from advises. Implications of the evidence in
the context of the reflective statement.
E Awards and recognition. Nominations for awards and
awards received (include award criteria). Other recognition.
When the portfolio is used as part of the basis for person-
nel decisions (e.g. awarding of promotion or tenure or deter-
mining merit raises), it should be independently reviewed by
at least two raters who have been trained in portfolio evalua-
tion. Following a predetermined scheme, the raters should
assign values to the quality of reflection and documentation,
the instructor's commitment to high quality teaching and
learning, and the instructor's teaching and advising effec-
tiveness and (if appropriate) educational scholarship. The
raters should compare and discuss their ratings, make any
changes they believe to be appropriate, and arrive at a
consensus rating. The individual and consensus ratings
should be included in the portfolio to be used in the
decision-making process.
Eventually, the department head must make a determina-
tion of teaching effectiveness based on his or her review of
the assessment data. A form for guiding this review is avail-
able from the Kansas State University IDEA Center.1241


In his landmark work Scholarship Reconsidered,[401 Ernest
Boyer proposed that academics can pursue scholarly activi-
ties in four different arenas: discovery (advancement of the
frontier of knowledge in a discipline), integration (making
connections across disciplines, putting research discoveries
in broader contexts and larger intellectual patterns), applica-
tion (applying the outcomes of discovery and integration to
socially consequential problems), and teaching (helping stu-
dents acquire knowledge and develop skills). Boyer argued
that these four areas are all equally vital to the mission of the
research university and that universities should therefore
recognize and reward them all equally.
The publication of Scholarship Reconsidered intensified
an ongoing discussion about the role of teaching in the
evaluation of faculty performance at research universities.

Among the focal questions of the discussion are "What is
educational scholarship?" and "How can you assess its qual-
ity?" The following discussion is taken largely from a recent
article that addresses these questions.1411
What is educational scholarship?
Boyer lists the elements that make teaching a scholarly
1. Subject knowledge. The scholarly instructor has a deep
conceptual understanding and a broad awareness of the
current state of knowledge of the subject being taught.
2. Pedagogical knowledge. The scholarly instructor can
formulate analogies, metaphors, and images that build bridges
between his or her understanding of the subject and the
knowledge and level of experience of the students. The
instructor is also familiar with a variety of effective instruc-
tional methods and the research base that confirms their
3. Commitment to continuing growth as an educator. The
scholarly instructor is committed to continuous improvement
of his or her disciplinary and pedagogical knowledge. Indi-
cations of such a commitment are books read, journals
subscribed to, and seminars, workshops, and conferences
A fourth element might be added to this list:[411
4. Involvement in development, assessment, and dissemina-
tion of innovative instructional methods and materials.
Instructors who keep their subject knowledge current, learn
about and implement effective teaching methods, and con-
tinue to work on improving their teaching may be said to be
effective teachers, worthy of being nominated for whatever
rewards the institution offers for teaching effectiveness, but
they are not necessarily educational scholars. To qualify for
that title, we propose that they must also undertake the
activities associated with traditional disciplinary research:
innovation and rigorous assessment and evaluation of the
innovations. In educational scholarship as in disciplinary
scholarship, the fruits of the labor might be products (e.g.,
textbooks or instructional software) or processes (e.g., new
or improved methods for motivating students, promoting
their intellectual development, or assessing their learning).
Also as in disciplinary scholarship, making results available to
the professional community for evaluation, replication, and
adoption is a necessary component of educational scholarship.

The improving climate
for educational scholarship
In the past, even if engineering professors were inclined to
do scholarly work in education there were barriers to their
doing so successfully. Grants for engineering education re-
search were in short supply and provided minimal funding.
Engineering education journals did not require rigorous as-
sessment as a condition for publication, and journals in

Chemical Engineering Education

Future of Engineering Education

education and educational psychology that did so were not
receptive to contributions of an applied nature from other
disciplines. Engineering administrators and faculty peers
called on to evaluate faculty performance reports were unfa-
miliar with the education literature and generally discounted
all education-related papers, including those that adhered to
good assessment practices and were published in journals
with high standards. Campus awards for outstanding schol-
arship in teaching did not exist.
The climate for scholarship in engineering education has
become considerably warmer in recent years. The National
Science Foundation has provided millions of dollars of fund-
ing through its Division of Undergraduate Education and the
Engineering Education Coalition program, and corporate
foundations have also provided significant support to efforts
to improve engineering education. The Journal of Engineer-
ing Education has become a first-rate vehicle for scholarly
publications, and other high-quality refereed journals now
accept papers on engineering education research. 151 National,
regional, and-on some campuses-local awards for out-
standing scholarship in engineering education are given.
Unfortunately, many who rate faculty performance in engi-
neering are still inclined to discount education-related activi-
ties as not worthy of being counted toward promotion, ten-
ure, and merit raises, funded and published though they may
be. Hopefully, this situation will also improve before too
long as more and more professors are motivated to under-
take serious efforts to study and improve engineering educa-
tion-rigorously setting goals, developing measurable out-
comes, gathering data about the effectiveness of their inter-
ventions in the classroom, and subjecting the data to rigor-
ous analysis and interpretation.
How can educational scholarship be
assessed and evaluated?
Earlier in this paper, we proposed that for teaching to
qualify as a scholarly activity, the instructor should demon-
strate a command of both subject and pedagogical knowl-
edge, a commitment to continuing growth as an educator,
and an involvement in innovation in teaching and dissemina-
tion of results. We further propose that assessment of an
instructor's educational scholarship should consist of an-
swering the following three questions:1411
1. Did the teaching qualify as a scholarly activity?
2. Was the teaching effective?
3. Were the innovative products and processes developed by the
instructor well conceived, implemented, assessed and evalu-
ated, and disseminated?
The data obtained using the assessment tools described in
the preceding sections of this paper and summarized in the
section on the teaching portfolio should be adequate to as-
sess the first two questions. To answer the third question, the

same forms of evidence traditionally used in the assessment
of disciplinary research may be gathered. Acceptable evi-
dence includes the number and quality of conference presen-
tations, invited seminars, books, monographs, and refereed
publications; number of grants and contracts; citations of
publications; referee comments on submitted manuscripts
and grant proposals; internal and external reference letters
and comments, and recognition and awards.
The following standards proposed by Glassick, et al.,1421
provide a good basis for evaluating the quality of educa-
tional innovations:
1. Clear goals. Is the basis of the work clearly stated, the
questions addressed important in the field, and the objectives
realistic and achievable?
2. Adequate preparation. Does the scholar show an under-
standing of existing scholarship in the field, the necessary
skills to do the work, and the ability to assemble the necessary
3. Appropriate methods. Were the methods used appropriate to
the goals, applied effectively, and appropriately modified
when necessary?
4. Significant results. Were the goals achieved? Did the work
contribute significantly to the field? Did it open areas for
further exploration?
5. Effective presentation. Was the work presented effectively
and with integrity in appropriate forums?
6. Reflective critique. Does the scholar critically evaluate his
or her own work, bringing an appropriate breadth of evidence
to the critique and using the critique to improve the quality of
future work?

Faculty members who meet these standards are clearly
vital to both the educational and scholarly missions of the
university. They merit advancement up the faculty lad-
der-tenure, promotion, and merit raises-no less than
faculty members who meet institutional standards for
disciplinary research.


The assessment of teaching should done for a clearly
defined purpose-to evaluate teaching effectiveness
summativee assessment) or to improve it (formative assess-
ment). It should be done in the context of published goals,
measurable performance criteria, and agreed-upon forms of
evidence. The evidence should come from a variety of
sources, including learning outcomes assessments, student
end-of-course ratings, student surveys, focus groups, or in-
terviews, retrospective student evaluations of courses and
instructors, alumni and peer evaluations, and self-assess-
The ultimate measure of the effectiveness of teaching is

Summner 2000

Special Feature Section

the quality of the resulting learning. As with any other area
of assessment, meaningful assessment of learning requires
prior formulation of learning goals and measurable objec-
tives that address all desired knowledge, skills, and attitudes.
Tools for assessing learning include tests and test items,
written reports and proposals, oral presentations and inter-
views, student-generated critiques of work produced by oth-
ers, student self-evaluations, learning logs and journals. The
validity of inferences drawn from the data is increased if
norms or control group responses are available for objective
tests and test items and if multiple independent evaluations
are submitted and reconciled for subjective judgments such
as ratings of written project reports and oral presentations.
Student ratings of teaching are a valid and important
source of evidence for teaching effectiveness, especially if
they are averaged over at least a two-year period. Extensive
research shows that student ratings correlate positively with
both learning outcomes and ratings submitted by alumni and
peers. They should not be the sole instrument used to evalu-
ate teaching, however, since students are generally not quali-
fied to judge aspects of instruction like the currency and
importance of the course content, the depth of the instructor's
knowledge, and the appropriateness of the assignments, tests,
and grading policies. Peer ratings are the most appropriate
source of such judgments.
The common approach to peer rating is for untrained
faculty members to observe lectures and write about what-
ever catches their attention, an approach that yields informa-
tion of doubtful value. For peer ratings of instruction in a
course to be reliable and valid, the ratings should be ob-
tained from at least two good teachers who have received
training on what to look for in a classroom. The raters should
use a checklist of items regarding specific aspects of the
instruction and associated instructional materials (syllabi,
handouts, assignments, and tests), and the independent rat-
ings should be reconciled to arrive at a consensus rating.
A summative teaching portfolio may be assembled to
evaluate the teaching effectiveness of an individual faculty
member (as opposed to the effectiveness of teaching in a
single course or an instructional program). The portfolio
should contain a reflective statement of the faculty member's
teaching and advising philosophy, goals, and practices; a
summary of teaching and advising responsibilities; repre-
sentative instructional materials and student products; as-
sessment data that reflect on teaching and advising effective-
ness; documentation of efforts to improve effectiveness; a
summary of teaching innovations (new courses, instructional
materials, and teaching methods developed, and education-
related papers and presentations); and a list of teaching awards
and award nominations. When the portfolio is used as part
of the basis for personnel decisions, at least two indepen-
dent evaluations of the portfolio should be performed by

trained raters and reconciled.
Since the publication of Scholarship Reconsidered,140]
recognition has been growing that teaching can be a schol-
arly activity no less than disciplinary research, and that
scholarship in teaching should play the same role in deter-
mining faculty advancement that disciplinary research has
played for the past four decades. Following Boyer, we pro-
pose that the defining elements of scholarly teaching are
mastery of subject knowledge, familiarity with both general
and subject-specific pedagogy, and commitment to continu-
ing personal growth as an educator, and we propose the
additional element of involvement in development, assess-
ment, and dissemination of innovative instructional materi-
als and methods. The innovations should reflect an aware-
ness of the current state of the art of engineering education,
and analysis and evaluation of the results should adhere to
the same standards of rigor customarily applied to traditional
disciplinary research.
Assessment of the quality of a faculty member's educa-
tional scholarship should be based on the answers to three
questions: (1) Did the faculty member's teaching qualify as
a scholarly activity? (2) Was his/her teaching effective? (3)
Were his/her innovations well conceived, implemented, as-
sessed and evaluated, and disseminated? The faculty
member's subject knowledge, pedagogical knowledge, com-
mitment to continuing personal growth, and involvement in
innovation (the elements of scholarly teaching) and the ef-
fectiveness of the teaching can be judged from the material
assembled in a teaching portfolio. The quality and impact of
educational innovations can be inferred from the same forms
of evidence used to evaluate disciplinary research (num-
ber and quality of books, papers, and presentations; lit-
erature citations; number of research grants and con-
tracts; reference letters; and recognition and awards).
Faculty members who meet or exceed institutional stan-
dards for educational research merit the same recogni-
tion and opportunities for advancement as faculty mem-
bers who excel in disciplinary research.


Thanks for valuable suggestions during the preparation of
this paper to Chris Knapper of Queen's University, Kingston,
and Paola Borin, Dale Roy, Erika Kustra, Barb Love, Heather
Sheardown, and Phil Wood of McMaster University. Thanks
also for helpful reviews of the paper to Brian Baetz of
McMaster University, Suzanne Kresta of the University of
Alberta, Rich Noble of the University of Colorado, and
Dendy Sloan of the Colorado School of Mines. Special
thanks go to Don Woods of McMaster, whose ideas had a
profound influence on the content of this paper and the
thinking of its authors.

Chemical Engineering Education

Future of Engineering Education

1. Rugarcia, A., R.M. Felder, D.R. Woods, and J.E. Stice, "The
Future of Engineering Education: Part 1. A Vision for a New
Century," Chem. Eng. Ed., 34(1), 16 (2000)
2. Felder, R.M., D.R. Woods, J.E. Stice, and A. Rugarcia, "The
Future of Engineering Education: Part 2. Teaching Methods
that Work," Chem. Eng. Ed., 34(1), 26 (2000)
3. Woods, D.R., R.M. Felder, A. Rugarcia, and J.E. Stice, "The
Future of Engineering Education: Part 3. Developing Criti-
cal Skills," Chem. Eng. Ed., 34(2), 108 (2000)
4. Stice, J.E., R.M. Felder, D.R. Woods, and A. Rugarcia, "The
Future of Engineering Education: Part 4. Learning How to
Teach," Chem. Eng. Ed., 34(2), 118 (2000)
5. Alverno College, Assessment at Alverno College. Alverno Col-
lege Publications, Milwaukee (1985)
6. Woods, D.R., Problem-Based Leaning: How to Gain the Most
from PBL. Woods, Waterdown, Ontario, Canada (1994). Dis-
tributed by McMaster University Bookstore, Hamilton
Ontario, Canada
7. Rogers, G.M., and J.K. Sando, Stepping Ahead: An Assess-
ment Plan Development Guide, Rose Hulman Institute of
Technology, Terre Haute, IN (1996)
8. Feldman, K.A., "The Superior College Teacher's from the
Students' View," Res. in Higher Ed., 5, 43 (1976)
9. Cross, K.P., "On College Teaching," J. Eng. Ed., 82(1), 9
10. Chickering, A.W., and Z.F. Gamson, "Seven Principles for
Good Practice in Undergraduate Education," AAHE Bulle-
tin, 3 (March 1987)
11. Eble, K.E., "Project to Improve Teaching," Academe, 4, 3
12. Woods,D.R., Ideas for Teachers to Improve Learning, 3rd ed.,
Chemical Engineering Department, McMaster University,
Hamilton, Ontario, Canada (1999)
13. Ramsden, P., and N.J. Entwistle, "Effects of Academic De-
partments on Students' Approaches to Studying," British J.
of Ed. Psychology, 51, 368 (1981)
14. Terenzini, P.T., and E.T. Pascarelli, "Living with Myths:
Undergraduate Education in America," Change, 28 (Jan-
Feb 1994)
15. Woods, D.R., Motivating Teachers to Improve Learning. De-
partment of Chemical Engineering, McMaster University,
Hamilton Ontario, Canada, in press.
16. Watson, J.L., "An Analysis of the Value of the FE Examina-
tion for the Assessment of Student Learning in Engineering
and Science Topics," J. Eng. Ed., 87(3), 305 (1998)
17. Angelo, T.A., and K.P. Cross, Classroom Assessment Tech-
niques: A Handbook for College Teachers, 2nd ed., Jossey-
Bass, San Francisco, CA (1993)
18. Panitz, B., "The Student Portfolio: A Powerful Assessment
Tool," ASEE Prism, 5(7), 24 (1996)
19. Rogers, G.M., and J. Williams, "Building a Better Portfolio,"
ASEE Prism, 8(5), 30 (1999).
20. Felder, R.M., "What Do They Know, Anyway?" Chem. Eng.
Ed., 26(3), 134 (1992). Available on-line at < http:l /
www2. /unity/ lockers / users /f/felder/public / Col-
umns /Eval.html >
21. Cashin, W.E., "Student Ratings of Teaching: The Research
Revisited," IDEA Paper No. 32, IDEA Center, Kansas State
University (1995). Available on-line at .
22. McKeachie, W., "Student Ratings: The Validity ofUse,"Ameri-
can Psychologist, 52(11), 1218 (1997)
23. Marsh, H.W., and L.A. Roche, "Making Students' Evalua-
tions of Teaching Effectiveness Effective: The Critical Issues
of Validity, Bias, and Utility," American Psychologist, 52(11),

24. Hoyt, D.P., and W.H. Pallett, "Appraising Teaching Effec-
tiveness: Beyond Student Ratings," IDEA Paper No. 36, IDEA
Center, Kansas State University (1999). Available on-line at
. Forms for peer rating of instructional
materials and for department head review of teaching as-
sessment data are Appendices B and D of this paper.
25. Felder, R.M., "What Do They Know, Anyway? II. Making
Evaluations Effective," Chem. Eng. Ed., 27(1), 28 (1993).
Available on-line at < http:/ / /unity/lockers
users/f/felder/public /Columns/Eval2.html >
26. McKeachie, W.J., Teaching Tips: Strategies, Research, and
Theory for College and University Professors, 10th ed.,
Houghton Miflin, Boston, MA (1999)
27. Theall, M., and J. Franklin, eds., Effective Practices for Im-
proving Teaching., New Directions for Teaching and Learn-
ing No. 48, Jossey-Bass, San Francisco, CA, p. 89 (1991)
28. Felder, R.M., and R. Brent, "It Takes One to Know One,"
Chem. Eng. Ed., 31(1), 32 (1997). Available on-line at <
http: / / /unity/lockers/ users /f/felder/pub-
lic /Columns /Peer-rev.html >
29. Keig, L., and M.D. Waggoner, Collaborative Peer Review:
The Role of Faculty in Improving College Teaching. ASHE-
ERIC Higher Education Report No. 2, Washington, DC,
George Washington University (1994)
30. Peer Observation of Classroom Teaching. Report #CTL-15,
Center for Teaching and Learning, University of North Caro-
lina at Chapel Hill, Chapel Hill, NC (1994)
31. Seldin, P., Changing Practices in Evaluating Teaching, Anker
Publishing Co., Bolton, MA (1999)
32. Woods, D.R., Ideas for Teachers to Improve Learning, 3rd
ed., Chemical Engineering Department, McMaster Univer-
sity, Hamilton, Ontario, Canada (1999)
33. Felder, R.M., and R. Brent, "If You've Got It, Flaunt It: Uses
and Abuses of Teaching Portfolios," Chemical Eng. Ed., 30(3),
188 (1996). Available on-line at <
unity / lockers / users / f fielder public / Columns /
Portfolios.html >
34. Seldin, P., The Teaching Portfolio: A Practical Guide to Im-
proved Performance and Promotion/Tenure Decisions, 2nd
ed., Anker Publishing Co., Bolton, MA (1997)
35. Seldin, P., Successful Use of Teaching Portfolios, Anker Pub-
lishing Co., Bolton, MA (1993)
36. Shore, B.M., S.F. Foster, C.K. Knapper, G.G. Nadeau, N.
Neill, and V. Sims, Guide to the Teaching Dossier, Its Prepa-
ration and Use, 2nd ed., Canadian Association of University
Teachers, Ottawa, Ontario, Canada (1986)
37. Diamond, R.M., Preparing for Promotion and Tenure Re-
view: A Faculty Guide, Anker Publishing Co., Bolton, MA,
38. Weeks, P., "Developing a Teaching Portfolio." Available on-
line at portfolio.html>
39. Ramsden, P., The Lancaster Approaches to Studying and
Course Perceptions Questionnaires: Lecturer's Handbook, Edu-
cational Methods Unit, Oxford Polytechnic, Oxford (1983)
40. Boyer, E., Scholarship Reconsidered: Priorities of the Profes-
soriate, Carnegie Foundation for the Advancement of Teach-
ing, Princeton, NJ (1990)
41. Felder, R.M., "The Scholarship of Teaching," Chem. Eng.
Ed., 34(2), 144 (2000)
42. Glassick, C.E., M.T. Huber, and G.I. Maeroff, Scholarship
Assessed: Evaluation of the Professoriate, Jossey-Bass, San
Francisco, CA (1997)

Summer 2000

Special Feature Section



Part 6. Making Reform Happen

Richard M. Felder North Carolina State University, Raleigh, NC 27695
James E. Stice University of Texas, Austin, TX 78712
Armando Rugarcia Iberoamericana University, Puebla, Mexico

We have dealt in this series with changing condi-
tions in technology and society that will require
major reforms in engineering education,"1 instruc-
tional techniques that have been shown by theoretical and
empirical research to produce learning outcomes consistent
with these reforms,[2.3 ways to prepare faculty members to
implement the techniques,141 and effective techniques for
assessing both teaching and educational scholarship. 51
Those were the easy matters. The real challenge is to
create a favorable climate for these changes at research
universities-a climate that motivates faculty members to
improve their teaching and the quality of instruction in their
departments, supports their efforts to do so, and rewards
their successes. In this paper we suggest steps that might be
taken to create such a climate.


Evidence of the low status of undergraduate education at
research universities for the past half century is easy to find.
Every campus has its stories of outstanding teachers being
denied tenure because their record of research funding and
publications was judged inadequate. Tenured professors com-
monly warn their non-tenured junior colleagues not to spend
too much time on their teaching, telling them that a teaching
award before tenure is the "kiss of death."
Growing numbers of administrators and faculty members
acknowledge that a problem exists. Gray, et al.,[6' surveyed
over 23,000 academics in 47 United States universities and
concluded that many faculty, chairs, deans and academic
administrators at research universities believe that an appro-
priate balance between research and undergraduate teaching
does not now exist at their institutions. Knapper and Rogers17'

report a similar result of a recent extended survey that in-
cluded Canadian faculty and administrators.
Faculty members of the eight campuses comprising the
SUCCEED Engineering Education Coalition were recently
asked to rate the importance of teaching effectiveness to
them, to their colleagues and administrators, and in their
institution's faculty reward system, with 0 being not at all
important and 10 being extremely important.1 On average,
the 504 respondents rated the importance of teaching to
them personally at 9.3, the importance to their colleagues
and administrators in the range 7.0-7.3, and the importance
in the faculty reward system at 4.7. In the respondents' open
comments about administrative support for teaching, the
term "lip service" came up with remarkable frequency.
The teaching and assessment methods described in the
first five papers can be used by any instructor at any institu-
Richard M. Felder is Hoechst Celanese Professor (Emeritus) of Chemi-
cal Engineering at North Carolina State University. He received his BChE
from City College of New York and his PhD from Princeton. He has
presented courses on chemical engineering principles, reactor design,
process optimization, and effective teaching to various American and
foreign industries and institutions. He is coauthor of the text Elementary
Principles of Chemical Processes (Wiley, 2000).
Armando Rugarcia graduated from the Universidad Iberoamericana
(UIA) in 1970 and went on to earn his MS in chemical engineering from
the University of Wisconsin in 1973 and his Doctorate in Education from
West Virginia University in 1985. He has been a full-time professor of
engineering at UIA since 1974 and was Chair of the Chemical Engineer-
ing Department there from 1975 to 1980. He was also Director of the
Center for Teaching Effectiveness at UIA from 1980 until 1986. He has
written four books on education, one on process engineering, and more
than 130 articles.
James Stice is Bob R. Dorsey Professor of Engineering (Emeritus) at the
University of Texas at Austin. He received his BS degree from the
University of Arkansas and his MS and PhD degrees from Illinois Institute
of Technology, all in chemical engineering. He has taught chemical
engineering for 44 years at the University of Arkansas, Illinois Tech, the
University of Texas, and the University of Wyoming. At UT he was
Director of the Bureau of Engineering Teaching Center and initiated the
campus-wide Center for Teaching Effectiveness, which he directed for 16

Copyright ChE Division of ASEE 2000

Chemical Engineering Education

Future of Engineering Education

tion, but while most of them do not require a heavy alloca-
tion of resources, they all require time and effort to learn,
implement, and refine. As long as faculty members feel that
their efforts to improve teaching will be largely unappreci-
ated and unrewarded-and in fact could jeopardize their
chances for advancement up the faculty ladder-educational
reform will be difficult to achieve. In
such a climate, the temptation is great to
simply dust off and recycle the old lec-
ture notes once more. As loni
On the brighter side, a positive shift in member
campus attitudes toward teaching began their
in the early 1980s and has grown steadily
since then. There were several catalysts impro
for this shift, including reports by the will b
National Research Council, the National unapJ
Academy of Engineering, and the Ameri- and urn
can Society for Engineering Education and in
that called for reforms in engineering edu-
cation.191 Another driver for change was a jeo
growing chorus of complaints from em- chAL
players of engineers about deficiencies adva
among recent graduates in the skills u
needed for success in modern engineer- faculty
ing practice.19' "1 Responding to these calls, educate
the National Science Foundation began
to provide significant funding for educa- will b
tion-related research and to consider the to a
impact of proposed research on educa- In such
tion in its proposal review process, and the te
ABET adopted the outcome-based Engi-
neering Criteria 2000 as its accreditation i
standard beginning in 2001. r1 Adminis- the c
trators and faculty members are begin- and t
ning to recognize that new instructional way t
methods and materials will be required to
to receive continued accreditation if that
standard is rigorously enforced.
Evidence for the growing importance
of effective undergraduate instruction at research universi-
ties is abundant. Presentations at ASEE conferences and
submissions to the Journal of Engineering Education have
grown dramatically in both number and quality. Tested in-
novations in teaching methods, technology-based instruc-
tion and distance education, assessment and evaluation of
teaching and learning, design across the curriculum, and
multidisciplinary curriculum integration are described with
increasing frequency in the literature."[2 Teaching seminars
and workshops are regularly presented to engineering fac-
ulty and graduate students on campuses where nothing of the
sort had ever been done. In its annual offerings from 1991

e tb
e I

Summer 2000

through 1999, the National Effective Teaching Institute has
reached 472 professors of engineering and engineering tech-
nology from 157 institutions, many of whom have gone back
to present programs on their home campuses. These are
indeed exciting times for engineering education.
It is not yet time to break out the champagne, however.
The dramatic progress made in recent
years notwithstanding, most engineer-
ing classes still consist of professors
faculty talking and writing on the board and
feel that students sitting and listening (or not
brts to listening); rigorous assessment of learn-
caching ing and teaching is still not part of the
ig culture of most institutions; faculty
largely members are still not routinely given
ciated any preparation for teaching; and se-
rarded- nior faculty are still advising junior fac-
ct could ulty (often correctly) that if they spend
e their too much time on their teaching they
could be jeopardizing their future aca-
s for demic careers.
Most faculty members are reluc-
tant to move away from the familiar
dder- and comfortable teaching methods with
11 reform which they were taught, especially if
difficult they believe that changing methods will
e. require substantial expenditures of time
ev. and could hinder their chances for ten-
climate, ure and promotion. They will only con-
tation sider doing so if they are first made
st dust off aware of the need for change, presented
notes with alternative methods, given con-
ch the vincing evidence of the effectiveness
of the alternatives, and assured that
Were adopting the methods does not neces-
ht. sarily require sacrificing syllabus cov-
erage or spending less time on research.
A prerequisite for significant educa-
tional reform is therefore the establish-
ment of instructional development programs that provide
this information and these assurances. Another necessary
condition for reform is for faculty members to be convinced
that their efforts to improve teaching will not work against
their career advancement, and that if successful, the efforts
can in fact work in their favor.
Descriptions of the need for change, alternative teaching
methods, and evidence of the effectiveness of those methods
are given in the first five papers in this series. The next
section suggests how faculty members can make meaningful
improvements in teaching without spending excessive

Special Feature Section

amounts of time or shortening their syllabi, and how they
can deal effectively with student resistance to student-cen-
tered teaching methods. The section after that proposes tan-
gible steps that administrators might take to make their cam-
pus culture supportive of educational reform.


Finding Time to Plan and Implement Changes

In their teaching workshops, the authors
...they caution the participants that if they try to
are implement every new technique they hear
advised about, they will probably be overwhelmed
to select by the time they find themselves spending
only and the student resistance they encounter,
one or get discouraged, and go back to old ways of
two doing things.
ideas Instead, they are advised to select only
at a one or two ideas at a time and try them long
time enough for the students to acclimate to the
and try new methods. If a method seems to be work-
them ing well, they should keep on using it; if it
long seems to be ineffective, they should first
enough check back in the literature to make sure
for that the recommended guidelines have been
the followed, and if they have, discontinue it.
students Some time later (perhaps in the following
to semester or quarter), they might try another
acclimate one or two new ideas. There is no hurry.
to the Many effective instructional strategies
new take relatively little time to plan and imple-
methods. ment, several of which are listed below.
Instructors who have been teaching tradi-
tionally and wish to make changes might consider trying one
or two of these strategies first, consulting earlier papers in
the series for details.

> Motivate the presentation of each new topic by
relating it to previously learned material and
familiar applications, perhaps starting with a
realistic problem or illustrative case study.
> Write clear instructional objectives for course
topics and give them to the students as study
guides for tests.
> Assign brief small-group activities in class (have
the students respond to questions, formulate
questions, begin problem solutions, carry out
steps in problem solutions and derivations,
brainstorm ideas,...)

Have students complete one or two out-of-class
assignments in teams (rather than moving
immediately to full-scale formal cooperative
Periodically ask students to monitor and reflect
on their learning, either in the form of minute
papers (What was the main point of the lecture?
The muddiest point?) or using feedback forms
collected at the end of a lecture period. The
instructor need only skim the responses to look
for patterns and begin the next class by respond-
ing to common points of confusion; it is not
necessary to provide individual feedback on
every response.
Collect midterm evaluations of the class. (What
am I doing in this class that is helping you learn
and you would like me to continue to do? What
am I doing that is hindering your learning and
you would like me to discontinue?) Respond to
any reasonable suggestions made by more than
two students, accepting those you wish to accept
and explaining why you will not go along with
the others.
Once comfortable with these strategies, instructors
should gradually move on to methods that involve greater
departures from usual teaching practice and take more
time to implement.2'3'

Covering the Syllabus

At teaching workshops that ad-
vocate active and cooperative learn-
ing and other student-centered in-
structional methods, the first ques-
tion is almost invariably a version
of "Can I do all that without sacri-
ficing coverage of important course
Our initial response to this ques-
tion is that the goal of teaching is
not to cover material but to uncover
it. Virtually all cognitive scientists
agree that people learn by doing
and reflecting, not by watching and
listening to someone else tell them
what they are supposed to know.1131
Instructors can present almost
any amount of material in any
amount of time by using transpar-
encies or presentation graphics and

...the goal
is not to
but to
uncover it.
agree that
learn by
doing and
not by

Chemical Engineering Education


Future of Engineering Education

talking quickly enough, but there is not much point in doing
so if hardly any of the material is being learned.
Having said that, we add that using active learning does
not require reducing the syllabus. It is true that in-class
activities take time (although not nearly as much as instruc-
tors who have not used them fear), and if these activities are
included in classes and no compensatory changes are made,
content coverage might indeed decrease.
Compensatory changes can be made, however. Felder and
Brent[r31 propose that instructors take large chunks of the
lecture material they usually present explicitly in class-the
complete step-by-step derivations, sentences, flowcharts,
schematics, and plots-and give them to the students in
handouts or a coursepack. The handouts should include
gaps-missing steps in derivations, axes with no curves
shown, questions and problems with spaces for answers
to be filled in.
The students can skim through short sections of the notes
during class, and the instructor can either lecture on the gaps,
get the students to work in small groups during class to fill
them in, or leave the gaps to be filled in outside class. The
instructor should caution the students that he/she will put
variants of the gaps on the tests and then do it.
If this recommendation is followed, most students will
read the notes and make sure the gaps get filled one way or
another-at least after the first test, when they discover that
the instructor was serious about including them.
Class sessions can now be devoted primarily to the most
important and/or conceptually difficult material in the lec-
ture notes, and the students will have opportunities during
those sessions for the action and reflection that lead to true
learning. The class time the instructor saves by not having to
spell out every word and formula in the lecture notes is
sufficient to allow for all the active learning exercises he/she
might wish to include, and the syllabus may actually be
expanded to cover more material rather than less.

Defusing Student Resistance

Students do not always welcome unfamiliar teaching meth-
ods with open arms, especially if the new methods push
them out of the comfort zone in which the instructor tells
them everything they need to know and then asks them to
repeat it on the test. Students introduced to active learning,
for example, sometimes accuse their instructors of not doing
their jobs when they require the students to learn some
things on their own. This hostile reaction is extremely dis-
turbing to instructors who are not expecting it and don't
know how to deal with it, and many who encounter it
become discouraged and revert to the less effective but

Students do not always welcome unfamiliar
teaching methods with open arms....
This hostile reaction is extremely
disturbing to instructors who are
not expecting it and don't know
how to deal with it,

safer lecture-based approach.
The occurrence of student resistance is familiar to anyone
who has attempted a student-centered instructional approach
like cooperative or problem-based learning.[15-18] Woods1151
suggests that students who find themselves deprived of the
support they are used to getting from their instructors go
through some or all of the stages that psychologists associate
with the grieving process: shock, denial, strong emotion,
resistance and withdrawal, surrender and acceptance, struggle
and exploration, return of confidence, and integration and
success. Some students quickly pass through or even skip
some of the stages and others get stuck and never emerge
from one of them for the entire semester, but the process
is a natural one and instructors should not be surprised
when they see its sometimes unpleasant manifestations
in their students.
Felder and Brent1l71 suggest several things instructors might
do to hold down the resistance to student-centered instruc-
tion long enough for the students to start seeing the benefits
of the approach for themselves.

E[ Start early, start small, and build.
Instructors planning to use a student-centered
approach should start using it near the beginning of
the course, making their goals and expectations clear
from the outset. If they have not used the approach
before, however, they should not make it the primary
mode of instruction in the course the first time they
try it. It is better to take small steps and gradually
increase the level of commitment to the approach,
never venturing too far beyond the zone of personal
comfort and confidence.

El Explain what you are doing and why.
When students are plunged into a new instructional
approach-especially one that shifts some of the
burden of learning away from the instructor and onto
them-they may assume that they are guinea pigs in
some sort of experiment and react negatively. Before
any of the new methods are implemented, they
should be outlined to the class and some of the
reasons for using them given, and periodic reminders
might be beneficial at several points during the
semester. Felder and Brent"'7-81 outline several

Summer 2000

Special Feature Section

arguments that can be offered, including research
results that demonstrate the learning benefits of the
methods and statements about the relevance of the
methods to the engineering workplace.

EI Be flexible when implementing new instructional
Some students in every class have unique needs and
constraints for which allowances should be made. For
example, if students are supposed to work in teams
outside class on homework assignments or projects
and a student has a full-time job or commutes to
campus from a considerable distance, he or she might
be permitted to work individually. Several students in
a class in the same position could be organized into a
virtual team that works together on an Internet chat
facility or via periodic conference calls. At a com-
muting campus where many students would find it
difficult to meet outside class hours, the instructor
might set up a number of virtual teams or make
provision for teams to use some class time to work on
their projects.

EU When all else fails, consult the manual.
If student hostility to an instructional method is
excessive or if it seems to be growing rather than
diminishing with time, check back in the literature on
the method to see if any recommendations (including
those just given) have been neglected. If any have,
take remedial measures.


Excellent teaching has generally enjoyed vigorous rhetori-
cal support from university administrators but limited tan-
gible reward or public recognition. Excellent research, on
the other hand, yields summer salaries, funds for national
and international travel, release from teaching and service
responsibilities, merit raises, and most significantly, tenure
and promotion.
The components of academic research-fundraising, plan-
ning, carrying out and assessing research projects, supervis-
ing graduate students, giving seminars, and writing papers,
among other tasks-all take a great deal of time. The com-
ponents of college teaching-learning and implementing ef-
fective teaching and assessment methods, designing and up-
dating courses to reflect the current state of the art and meet
accreditation standards, and advising and mentoring stu-
dents-take an equally great deal of time.

For most faculty members, time is in severely short sup-
ply. If research offers the promise of substantial rewards and
career advancement and teaching offers little more than in-
ternal self-satisfaction, the only faculty members likely to
engage in educational reform are those already deeply com-
mitted to teaching excellence. The number of such faculty
members has grown rapidly in the past twenty years, but it is
still too small to achieve reform of the magnitude that will be
required to meet the demands on engineering education ex-
pected in the coming decades.'"
We believe that most university faculty genuinely want to
be good teachers. Their desire is not motivated by the pros-
pect of external rewards but by intrinsic motivators such as
the sense of accomplishment that comes from equipping
students with new skills and self-confidence.
For all but the most dedicated, however, intrinsic motiva-
tion to teach as well as one can eventually diminishes if the
campus culture offers little more than empty rhetoric and a
few awards to demonstrate its commitment to excellence in
teaching. Sloan201l suggests-and we agree-that external
recognition and rewards for effective teaching are needed to
support and reinforce intrinsic motivation to teach well.
Gmelch, et al.,"2 and Boyer[221 support this idea, identifying
inadequate recognition and reward as a major contributor to
faculty stress and burnout.
The need to improve the campus climate for teaching is
emphasized in a 1999 report of the National Research Coun-
cil on transforming education in Science, Mathematics, En-
gineering, and Technology (SME&T).123' The fifth "vision"
articulated in this report speaks directly to the point:
Vision 5: All postsecondary institutions would provide
the rewards and recognition, resources, tools, and infra-
structure necessary to promote innovative and effective
undergraduate SME&T teaching and learning.

The report notes that if Vision 5 were to be realized, univer-
sities "would recognize and appropriately reward faculty
leaders and departments or program units that have intro-
duced new teaching and learning methods into their courses
and curricular programs," recognition and rewards that the
Council clearly believes are not currently in place.
Motivated by pressures from respected organizations like
the NRC and from industry, governing bodies, and accredit-
ing agencies, many academic programs at research universi-
ties have instituted measures to improve teaching. Brent and
Felderl241 have assembled a list of such measures, which we
summarize below. The more of these measures in place on a
campus, the more likely the faculty will be to play an active
role in efforts to reform engineering education.
To avoid excessive repetition, in the remainder of this

Chemical Engineering Education

Future of Engineering Education

section we will use "teaching" to be a catch-all term cover-
ing classroom instruction, advising and mentoring under-
graduate students, mentoring faculty colleagues and gradu-
ate students in teaching, and educational scholarship.

3 To encourage and help faculty to improve their teaching
Provide funds for travel to education-related work-
shops and conferences.
Provide internal grant support-summer salary and/
or materials/supply money-for faculty who propose
to carry out a specific project related to their teaching
When giving new faculty start-up money, designate
some of it for teaching-related activities
Purchase good books on teaching-e.g., McKeachie[251
and Boice1261-and give them to new faculty mem-
bers, perhaps in conjunction with an orientation work-
Establish and support a Engineering Center for Teach-
ing and Learning that sponsors a variety of teaching
improvements for new faculty, experienced faculty,
and graduate students.[271 Alternatively, if a campus-
wide center already exists, establish a half-time or
full-time engineering faculty development coordina-
tor to work with Center personnel on programs spe-
cifically for engineers and to help involve engineers
in suitable campus-wide programs.
Institute a Teaching Leaders program in which out-
standing engineering teachers are identified and com-
pensated for jointly facilitating teaching courses, semi-
nars and workshops with faculty development per-
sonnell41 and serving as mentors to new faculty mem-
Establish an Engineering Teaching Fellows program
in which faculty members in their first few years of
teaching receive observation and individual consult-
ing by teaching center personnel and regularly attend
seminars or learning communities devoted to good
teaching and educational scholarship. Provide sti-
pends, travel funds, or some other tangible form of
support to the teaching fellows.

E To encourage faculty to redesign curricula and courses
(e.g. for distance education or Web-based instruction):
Offer summer salary, release time during the semes-
ter, and/or the assistance of a graduate student or
work-study student for planning the redesign and

preparing new instructional materials.
Offer and support faculty participation in intensive,
personalized summer programs with a specific focus
(i.e., teaching and learning via the Web).
Provide funds for a retreat to a team of faculty work-
ing on course or program redesign.

E[ To encourage departments to improve teaching or un-
dertake curriculum renewal:
Devote some of the regular departmental or college
seminars to topics related to teaching.
Identify graduate TA's to help faculty members to
incorporate technology (e.g., design a course Web
site and put class materials on it, design tutorials to
help students with a new piece of software) or to
update their courses specifically to address accredita-
tion criteria.
Establish a college of engineering reserve fund to
support multi-faculty departmental initiatives to im-
prove instruction, courses, and curricula and to re-
ward departments that demonstrate the success of the
initiatives using systematic assessment.

[J To reward faculty members for excellence in teaching,
advising, mentoring, and educational scholarship:
Require faculty members seeking tenure and/or pro-
motion to prepare a teaching portfolio containing
evidence of teaching effectiveness and educational
scholarship.'11 Have multiple evaluators rate the port-
folio using a standardized rating form and reconcile
their ratings. Include the results in a meaningful way
when making tenure and promotion decisions.
Hold workshops or seminars for senior faculty in-
volved in making tenure and promotion decisions to
teach them how to evaluate teaching documentation.
In preparation for evaluation, early in the school year
have faculty determine a percentage for each aspect
of their jobs (teaching, research, service, extension)
and goals related to each part. Base evaluations and
salary recommendations on the predetermined per-
centage. Percentages might change from year to year
as faculty members explore new research and teach-
ing projects and move into new phases of their ca-
Allocate a portion of merit raise funds for outstand-
ing teaching or mentoring.
Establish numerous small awards and several large

Summer 2000

Special Feature Section

awards ($5000 or more) to reward excellence in teach
ing, advising, mentoring, and educational scholar-
SRecognize teaching achievements at faculty and ad-
visory board meetings and in departmental and uni-
versity publicity releases.
Adopting some of these suggestions can enable an institu-
tion to improve the quality of its instructional program sub-
stantially, especially if the suggestion about taking teaching
into account in tenure and promotion decisions is one of
those adopted.

The devaluation of teaching in the faculty incentive and
reward structure of most research universities that began
four decades ago has begun to reverse; however, much re-
mains to be done before the educational reforms suggested
in this series of papers can become part of the mainstream of
engineering education. The key is to provide instructional
development that informs faculty members about alterna-
tives to traditional teaching and assessment methods-what
they are, what the evidence is for their effectiveness, and
how they can be implemented without taking excessive prepa-
ration time or having to sacrifice important course content.
The time demands imposed by the adoption of student-
centered instructional approaches like active, cooperative,
and problem-based learning can be minimal as long as new
methods are introduced gradually, starting with techniques
that do not require much preparation or class time. For
example, instructors might motivate the presentation of
new course topics with short industrially relevant case
studies, hand out instructional objectives in the form of
study guides for one or more tests, and include some
brief active exercises in class.
To compensate for the additional class time taken by these
instructional techniques, the instructors can put portions of
the lecture notes in handouts or a coursepack, including gaps
and questions to be addressed in or out of class. The time
saved by not having to say and write everything in the
lecture notes should be sufficient to allow as many active
learning exercises as the instructor wishes to assign.
Another concern that makes faculty members reluctant to
move to more student-centered instructional approaches is
the fear (often based on experience) of student aversion to
these methods. Starting slowly and gradually increasing the
use of such approaches serves to minimize student resis-
tance. Instructors should also explain to the students what
they plan on doing and their reasons for doing it, including
some published research results attesting to the learning
benefits of the approach to be used. Instructors should also

avoid rigidity in the application of the methods, recognizing
that some students have unique time constraints and other
problems that should be dealt with on an individual basis.
Convincing faculty members that alternative teaching and
assessment approaches lead to effective learning and ad-
dressing their concerns about implementation of the ap-
proaches are necessary but not sufficient conditions for edu-
cational reform. Before most engineering faculty members
will be willing to invest much time and energy to improve
teaching, they must be convinced that teaching improvement
is truly valued by their institution and that their efforts will
not limit their prospects for tenure and promotion.
A list of possible steps that institutions can take to com-
municate that message is presented in this paper. The steps
include establishing and supporting workshops and
mentorships for new faculty members and graduate students,
providing grants and release time for efforts to update and
improve the effectiveness of curricula and courses, recog-
nizing and meaningfully rewarding excellence in teaching,
advising, mentoring, and educational scholarship, instituting
formal procedures for assessing teaching performance and
educational scholarship, and taking the results into account
when making decisions on tenure, promotion, and merit
raises. The latter step alone could be sufficient to raise the
quality of an institution's instructional program to a level
that exceeds the expectations of the most idealistic propo-
nents of educational reform.

We thank Heather Sheardown, Erika Kustra, Paola Borin
(McMaster University), Antonio Rocha (Instituto
Technol6gico de Celaya), John O'Connell (University of
Virginia), Dendy Sloan (Colorado School of Mines), and
Suzanne Kresta (University of Alberta) for their sugges-
tions during the writing of this paper. Special thanks go
to Don Woods for his major contributions to all of the
papers in the series.

1. Rugarcia, A., R.M. Felder, D.R. Woods, and J.E. Stice, "The
Future of Engineering Education: Part 1. A Vision for a
New Century," Chem. Eng. Ed., 34(1), 16 (2000)
2. Felder, R.M., D.R. Woods, J.E. Stice, and A. Rugarcia, "The
Future of Engineering Education: Part 2. Teaching Meth-
ods that Work," Chem. Eng. Ed., 34(1), 26 (2000)
3. Woods, D.R., R.M. Felder, A. Rugarcia, and J.E. Stice, "The
Future of Engineering Education: Part 3. Developing Criti-
cal Skills," Chem. Eng. Ed., 34(2), 108 (2000)
4. Stice, J.E., R.M. Felder, D.R. Woods, and A. Rugarcia, "The
Future of Engineering Education: Part 4. Learning How to
Teach," Chem. Eng. Ed., 34(2), 118 (2000)
5. Felder, R.M., A. Rugarcia, and J.E. Stice, "The Future of
Engineering Education: Part 5. Assessing Teaching Effec-

Chemical Engineering Education


Future of Engineering Education

tiveness and Educational Scholarship," Chem. Eng. Ed.,
34(3), 198 (2000)
6. Gray, P.J., R.C. Froh, and R.M. Diamond, "A National Study
of Research Universities on the Balance between Research
and Undergraduate Teaching," Center for Instructional De-
velopment, Syracuse University, May (1992). A summary of
this report by the same authors appears as "Myths and
Realities," AAHE Bulletin, 44(4), 4 (1991)
7. Knapper, C., and P. Rogers, Increasing the Emphasis on
Teaching in Ontario Universities, Ontario Council on Uni-
versity Affairs, 700 Bloor St., Toronto ON, Canada (1994)
8. Felder, R.M., R. Brent, T.K. Miller, C. Brawner, and R.H.
Allen, "Faculty Teaching Practices and Perceptions of Insti-
tutional Attitudes toward Teaching at Eight Engineering
Schools," Proceedings, 1999 Frontiers in Education Confer-
ence, ASEE/IEEE, Phoenix, AZ, November (1998)
9. Prados, J.W., and S.I. Proctor, "What Will It Take to Reform
Engineering Education?" Chem. Eng. Prog., 96(3), 91, (2000)
10. Todd, R.H., C.D. Sorensen, and S.P. Magleby, "Designing a
Capstone Senior Course to Satisfy Industrial Customers,"
J. Eng. Ed., 82(2), 92 (1993), and J. H. McMasters and J.D.
Lang, "Enhancing Engineering and Manufacturing Educa-
tion: Industry Needs, Industry Roles," Proceedings, 1995
ASEE Annual Conference, ASEE, Washington, DC, (1995)
are two of many reports that make the same point.
11. Felder, R.M., "ABET Criteria 2000: An Exercise in Engi-
neering Problem Solving," Chem. Eng. Ed., 32(2), 126 (1998).
Available on-line at f/felder/public Columns /ABET.html>
12. Felder, R.M., "The Warm Winds of Change," Chem. Eng.
Ed.,, 30(1), 34 (1996). Available on-line at unity / lockers / users If/felder Ipublic / Columns /
13. Haile, J.M., "Toward Technical Understanding." (i) "Part 1.
Brain Structure and Function," Chem. Eng. Ed., 31(3), 152
(1997); (ii) "Part 2. Elementary Levels," Chem. Eng. Ed.,
31(4), 214 (1997); (iii) "Part 3. Advanced Levels," Chem.
Eng. Ed., 32(1), 30 (1998)
14. Felder, R.M., and R. Brent, "FAQs II," Chem. Eng. Ed.,
33(4), 276 (1999). Available on-line at unity/ lockers/ users Iffelder Ipublic / Columns FAQs-
15. Woods, D.R., Problem-Based Learning: How to Gain the
Most from PBL, Woods Publishing, Waterdown, (1994). Dis-
tributed by McMaster University Bookstore, Hamilton,
Ontario, Canada
16. Felder, R.M., "We Never Said It Would Be Easy," Chem.
Eng. Ed., 29(1), 32 (1995). Available on-line at / unity /lockers / users /f/felder/public / Columns /
17. Felder, R.M., and R. Brent, "Navigating the Bumpy Road to
Student-Centered Instruction," College Teaching, 44(2), 43
(1996). Available on-line at ers /users /f/felder/public /Papers /Resist.html>
18. Felder R.M., and R. Brent, Cooperative Learning in Techni-
cal Courses: Procedures, Pitfalls, and Payoffs, ERIC Docu-
ment Reproduction Service, ED 377038 (1994). Available
on-line at public/Papers / Coopreport.html>
19. Felder, R.M., "Reaching the Second Tier: Learning and
Teaching Styles in College Science Education,"
J. College Sci. Teach., 23(5), 286 (1993). Available on-line at
pers / Secondtier.html>
20. Sloan, Jr., E.D., "Extrinsic versus Intrinsic Motivation in

Faculty Development," Chem. Eng. Ed., 23(3), 134, (1989)
21. Gmelch, W.H., P.K. Wilke, and N.P. Lovrich, "Dimensions
of Stress Among University Faculty: Factor-Analytic Re-
sults from a National Study," Res. in Higher Ed., 24(3), 266-
22. Boyer, E., Scholarship Reconsidered: Priorities of the Pro-
fessoriate, Carnegie Foundation for the Advancement of
Teaching, Princeton, NJ (1990)
23. National Research Council, Transforming Undergraduate
Education in Science, Mathematics, Engineering, and Tech-
nology, Washington, DC, National Academy Press (1999)
24. Brent, R., and R.M. Felder, "Mentoring and Supporting
New Faculty Members: A Workshop," North Carolina State
University, (1999)
25. McKeachie, W.J., Teaching Tips: Strategies, Research, and
Theory for College and University Teachers, 10th ed.
Houghton Mifflin, Boston, MA (1999)
26. Boice, R., Advice for New Faculty Members, Allyn and Ba-
con, Needham Heights, MA (2000)
27. Felder, R.M., R. Brent, D. Hirt, D. Switzer, and S. Holzer,
"A Model Program for Promoting Effective Teaching in Col-
leges of Engineering," Proceedings of the 1999 Annual Con-
ference of the ASEE, Charlotte, NC, June (1999)
28. Felder, R.M., "Teaching Teachers to Teach: The Case for
Mentoring," Chem. Eng. Ed., 27(3), 176 (1993). Available
on-line at public / Columns /Mentoring.html> O

The editorship of CEE would like to ex-
press its appreciation to Professors Richard
M. Felder, Armando Rugarcia, James E.
Stice, and Donald R. Woods, for their ef-
forts in producing this series of articles on
the Future of Engineering Education. They
are to be complimented on their scholarly
approach to the subject, the generality of
their conclusions, and the timeliness of the
subject. They have captured a set of best
practices for promoting student learning and
expressed it in a concise and understand-
able fashion. Their vision, the result of many
years of combined teaching experience,
paints an exciting and optimistic picture
for the future of engineering education.

We are pleased to have been chosen as
the vehicle of information and discussion
for this important series. As we enter this
new millennium, we hope that it will in-
spire chemical engineering educators to
seek ways to enhance the learning experi-
ence of all their students.


Summer 2000

. M classroom




University of Cincinnati Cincinnati, OH 45221-0171

When my daughter was a little girl, her mother
asked, "Whom do you love most, mommy or
daddy?" My daughter gave the politically correct
answer "I love you both equally." But then her mother (who
is a mathematician!) asked, "But who is more equal?" My
daughter tactfully responded, "You are, mommy!" Some
might contend that dimensional analysis is elementary and
that all approaches are equal. If so, my response is that there
is a "more equal" approach if one can write down the de-
scribing equations. This approach is an alternative to the Pi
Theorem method that involves the following three steps
1. List all qualities on which the phenomenon depends.
2. Write the dimensional formula for each quantity.
3. Demand that these quantities be combined into a functional
relation that remains true independently of the size of the
In Step 3, one invokes the Pi Theorem, which states that n -
m dimensionless groups are formed from n quantities ex-
pressed in terms of m units. A proof of the Pi Theorem and
discussion of the special case n = m are given in Bridgman.Jf1
Unfortunately, using the Pi Theorem approach is not al-
ways straightforward. For example, how do we select the
quantities? When do we include dimensional constants such
as g, (Newton's Law constant), R (gas constant), etc? How
are dimensionless quantities such as angles involved? How
many units must be considered? For example, force can be
considered a primary quantity expressed in units of its own

William B. Krantzreceived a BS in chemistry (1961) from Saint Joseph's
College (Indiana) and his BS (1962) and PhD (1968) degrees in chemical
engineering from the University of Illinois-Urbana and the University of
California-Berkeley, respectively. From 1968-1999 he was Professor of
Chemical Engineering, Research Fellow in the Institute of Arctic and
Alpine Research, and President's Teaching Scholar at the University of
Colorado-Boulder, and in 1999 he accepted the Rieveschl Ohio Eminent
Scholar chair at the University of Cincinnati where he is Co-Director of the
NSF I/U CRC for Membrane Applied Science and Technology.
Copyright ChE Division of ASEE 2000

kind (e.g., Newton's), or a secondary quantity expressed in
terms of mass, length, and time (e.g., kg.m/s2). This problem
also arises with quantities involving energy or temperature
units, since both can be considered either as primary or
secondary quantities. If one can write the describing equa-
tions, the approach proffered here can be used to avoid the
aforementioned difficulties.

The procedure for this method is as follows:
1. Write the algebraic and/or differential equations
needed to solve the problem.
2. Write any required initial, boundary, and auxiliary
3. Use all available information in order to simplify the
equations in steps 1 and 2.
4. Define dimensionless dependent and independent
variables using arbitrary scale and reference factors.
5. Nondimensionalize the equations, initial, boundary,
and auxiliary conditions in steps 1 and 2.
6. Determine the scale and reference factors by setting
dimensionless groups equal to one (for scale factors) or
zero (for reference factors); this yields the minimum
parametric representation in the form
f(1,n,2 ,...,nk)= (1)
where Hi denotes a dimensionless group. These Hi s
include dimensionless groups formed from combina-
tions of the physical and geometric quantities and any
dimensionless independent variables; the latter will not
appear if they are integrated out or evaluated at fixed
spatial or temporal conditions.
7. The dimensionless groups in step 6 are not unique; it

Chemical Engineering Education

Some might contend that dimensional analysis is elementary and that all approaches are
equal. If so, my response is that there is a "more equal" approach if one can write down
the describing equations. This approach is an alternative to the Pi Theorem method...

may be advantageous to isolate two dimensional
quantities into one group (if possible) in order to
determine their interdependence. This is done by
forming a new group from the k dimensionless groups
via the operation
Hp,n b,..., (2)
p 12 k (2)
where 3 is a constant, and a,b,...,j are constants chosen
to isolate the desired quantities into the new dimen-
sionless group ip; one can use this new group along
with any k 1 of the original groups; however, this
operation cannot result in eliminating a dimensional
quantity from the analysis.
8. The number of groups can be reduced further when a
Hi is either very large or very small by expanding Eq.
(1) in a Taylor series in the small (or reciprocal of a

f(ni,n2,., nk)=f + Of ri +O(ni ) (3)
2i-n ..0.i in=0

If Eq. (3) can be truncated at the first term, the
correlation will be a function of k 1 Hi s.

This method is closely related to scaling analysis.[2'3 It
differs, however, in that no attempt is made to insure that the
dimensionless quantities are of order one.
This approach is not new-indeed, Bird, et al., have out-
lined the technique.14' Hellums and Churchillts' also used it to
achieve the minimum parametric representation and to iden-
tify similarity transformations. This approach has also been
suggested in two articles in Chemical Engineering Educa-
tion. Andrewsl61 commented, "The subject is best taught by
writing down known equations as relations between dimen-
sionless groups," and Churchill17' stated "Dimensional analy-
sis is most powerful when it is applied to a complete math-
ematical model in algebraic (differential and/or integral)
Despite this recognition that dimensional analysis is "best
taught" and "most powerful" when applied to a complete
model, no article has appeared in Chemical Engineering
Education describing the approach in detail. The latter is the
principal goal of this paper. It will also indicate how this
approach, when combined with asymptotic analysis, can
lead to useful limiting forms. In addition, this paper will
indicate how dimensional analysis can be combined with
empirical results to obtain information about the functional
form of a dimensionless correlation. Finally, it will dispel

the notion that dimensional analysis is somehow limited to
fluid dynamics and will provide examples drawn from heat
and mass transfer!
This alternative method will be illustrated via four ex-
amples. The first will be shown in detail while the others will
be outlined. The Pi Theorem approach will also be applied to
each example, but in doing this, a less general approach
typical of that often used by students who have limited
experience will be invoked. The author has found that the Pi
Theorem can be used to obtain the same result as that of the
alternative method proffered here. Far more physical insight
is required, however, to achieve the most general result
using the Pi Theorem. There are two principal problems in
using the Pi Theorem. The first involves Step 3 and the
second relates to choosing the proper units. Some believe
that the choice of fundamental units is arbitrary. This mis-
conception is the source of much confusion concerning the
Pi Theorem and is the cause of many of its alleged viola-
tions. For example, in the system of statics, one must use the
units of force, mass, length, and time, and should not intro-
duce the dimensional constant go. In contrast, in the system
of dynamics, if one introduces force, mass, length, and time
as the units, one must introduce the dimensional constant go,
since these units are interrelated by Newton's law of motion.
For example, one must introduce g, if one uses SI units for
the quantities involved in a dynamical system, since this
system considers force, mass, length, and time as primary
These subtle aspects of dimensional analysis are discussed
in Bridgman,"11 which should be required reading for anyone
interested in dimensional analysis. But the need to know
these subtleties can be avoided by using the alternative ap-
proach suggested here. As such, this alternative method is
ideally suited as a teaching tool to give students a working
knowledge of dimensional analysis, Its implementation, as
well as other judicious operations useful in dimensional
analysis, will be illustrated.

We seek to correlate the terminal velocity, V,, of a spheri-
cal particle of radius R and density ps falling owing to
gravitational acceleration, g, through an incompressible
Newtonian liquid having density p and viscosity 1, as
shown in Figure 1. We obtain V, from a force balance on the
sphere given by

Summer 2000

-ff8r Pp-P VV+VV 1 dS+(p -p)g R3 =0 (4)
S I ( r=R
where 8i is the unit vector in the i-direction, S is the surface
area, 8 is the identity tensor, P is the dynamic pressure, V is
the fluid velocity, and denotes the transpose. In order to
carry out the integration in Eq. (4), one would have to solve
the equations of motion in spherical coordinates with bound-
ary conditions consisting of no-slip at the sphere surface and
a far-field velocity condition. In a coordinate system at-
tached to the sphere, these are given by

pV-VV=-VP+tV2 V
V=0 at r=R

V.r = -Vt cos
V60 =-Vt sinj

as r--oo

where r and 0 denote the radial and circumferential coordi-
nates, respectively.
Define the following dimensionless variables:

4 P V-LsV S~S (8)
Vs Ps Ls
where ^ denotes a dimensionless variable, and V,, Ps, and Ls
denote velocity, pressure, and length scales that will be
chosen to obtain the minimum parametric representation.
Introducing these into Eqs. (4-7) and dividing through by the
dimensional coefficient of one term in each equation yields

dS(Ps-p)g 4 R
+ VsLs

pVsLs. VVp V+v2V

V=0 at ^=r

V-ir = cos9
Vsn- as r--oo
V.e= V's sin 0
One possible set of scale factors is obtained by setting
following dimensionless groups equal to one:

R 1=>L= R


PsLs 51-sVt
PS 1P E-
u'V- R

Hence, the solution to Eq. (10) will depend on r, 0, and the
dimensionless group pVtR/t. When this solution is substi-
tuted into Eq. (9), evaluated at i = 1, and integrated over the
surface area, the resulting solution for the dimensionless


r R


g viscous liquid

Figure 1.
Spherical particle falling through a viscous liquid.

terminal velocity can be correlated in terms of the following
two dimensionless groups:

I (sp)gR2 and 12 (a Reynoldsnumber) (14)

Hence, either data or a numerical solution for V, can be
correlated in terms of HF and F12. These are not optimal,
however, if one is seeking a correlation for V,, since it
appears in both groups. By invoking the transformation in
Step 7 with a=l and b=l in Eq. (2), a new dimensionless
group, 113, not containing V,, can be obtained:

S(ps -p)pgR3 (15)
n1-3= (15)

Hence, data or numerical results for V, can be correlated in
terms of 13 and either Fli or f2.

- 0 A naive application of the Pi Theorem with n=6 and m=3
(or n=7 and m=4 if F is also used as a unit and gc included as
a quantity) indicates that the correlation for Vt requires three
(9) rather than two dimensionless groups. In order to obtain the
most general result from the Pi Theorem, one must recog-
(10) nize that g appears as the product g(ps-p). The alternative
approach suggested here avoids these subtleties associated
with the Pi Theorem method.
Standard references14J suggest that V, can be correlated in
terms of just Il; that is


g the

t2g( R -P) = (pPs-p)gR2_ 9 (16)
9g gVt 2
This is for the special case of creeping flow for which the
intertia terms can be neglected, however. Hence, 12 (or
equivalently 13) no longer appears in the minimum para-
metric representation. Note, a naive application of the Pi
Theorem would suggest three dimensionless groups (n=6
and m=3). But the Pi Theorem will predict one group if one
recognizes that g appears as the product g(ps -p) and that F
must be introduced as a unit (without including ge), since
creeping flow is a problem in statics. Clearly, the alternative
approach proffered here obviates the need to be aware of

Chemical Engineering Education


-n PsLsp V+- V Vv J
s~ I r=-;

these special considerations required to get the most general
result using the Pi Theorem.
A simpler way to obtain this result is to use Step 8. Since
creeping flow implies Re = H2 -+ 0, the expansion must use
groups 1, and F12. In the limit of F2 -> 0 one obtains that
V, can be correlated solely in terms of 11l.

The falling-head method is used to determine permeability
of soils. This test, shown in Figure 2, involves driving a pipe
of radius R into the soil until it penetrates the water table,
which is shown here at the exit of the tube. The pipe is filled
with water to a height Li and the time tD required to drain it to
a height L, is measured. The draining time tD is related to the
axial velocity Vz and Darcy's law via the equation
tD tD k
(Li L)=- Vz dt = =+pgj dt 0 r
where k is the permeability, g- is the viscosity, p is the
density, and g is the gravitational acceleration. The incom-
pressible continuity equation implies that the pressure, P, is
obtained from a solution to the axisymmetric Laplace's equa-
tion in cylindrical coordinates:
1 a( aP a2
a ara 0 (18)

This is subject to the boundary conditions

P=Patm+pgL(t) at z=0, Lf P=Patm at z=0, R k aP
Vz- =0 as z --, 0 u z
k aP
Vr- k -0 at r=0, -o- S a rr
P=Patm- pgz as r->-, -co

Applying Steps 4, 5, and 6 leads to the following dimen-
sionless variables:
P-Patm r ;=z L kpgt (24)
pgLi R Li Li i(Li-Lf)
Note that the solution to the dimensionless form of Eq. (18)
will give the pressure as a function of t, ^, and 2. But when
the pressure is substituted into the dimensionless form of Eq.
(17), evaluated at z=0, and the integration carried out, the
resulting solution will contain only three dimensionless pa-
rameters. This is the minimum parametric representation
and implies that k is correlated in terms of tD as follows:

S kpgtD ( R L =f(F2, 3) (25)
p(Li -Lf) Li' L- f( ) (25

A naive application of the Pi Theorem would give five
dimensionless groups (i.e., n=8 and m=3)! Equation (25) can
be obtained from the Pi Theorem if one recognizes that
kpg/[t can be considered as a single quantity, thereby giv-
ing two groups (i.e., m=5 and n=2).
Equation (25) can be cast into a more useful form by using
an empirical correlation obtained using water and a 5-cm
radius pipe for a soil having a k=5.9xl0-6cm2:181

tD =4.94F i ) (26)

It is convenient to recast Eq. (25) in terms of a new dimen-
sionless group that does not contain L, and L,:

n14 tpkpg =f(2,3 tkpg 5 f2,3)
IT=-,- /1 2f p.---5 =f( I, H3
R( 1Li)

(22) If R<
(23) tDkpgn5=f(F)
= 5 3


round surface
around surface

water table I

Figure 2. Falling-head apparatus for measuring the
permeability of a soil.

Comparing Eq. (28) with Eq. (26) then implies that

tD =I5 ( p )=--4.94 fn 13 (29)

Hence, if H2 <<1, the generalized correlation relating k and
to is obtained by substituting values for the quantities in Eq.
(29) and is given by

4.94(5.9x10-6cm2)( -g Y980cmj
F15 s= cm ) s nH, =-0.572enFH3
0.01-g s(5cm)


In this case, an enlightened approach to dimensional analy-
sis in combination with data for a specific falling-head test
gives the functional form of the generalized tD correlation in

Summer 2000

terms of the relevant parameters; i.e., Eq. (30) applies for
any falling-head test for which I2 <<1, irrespective of the
fluid, pipe size, and soil.

Mom and Dad are planning to roast turkey for the entire
clan and need to know the cooking time, tc, for the 28-lb bird
shown in Figure 3.
The cookbook pro-
vides the information
shown in Table 1.[91
This is your opportu-
nity to impress them
with what you have I
learned by using di-
mensional analysis to
generalize Table 1.
Assume that it takes p
geometric parameters
to characterize the
shape of a turkey and Figure 3. A very large turkey
that all turkeys are having a characteristic length L.
geometrically similar.
Hence, the p-1 dimen- TABLE 1
sionless geometric ra- Timetable for Roasting Turkey'19
tios characterizing tur-
keys will be the same. Ready-to-Cook Approximate Total
One need only include Weight Cooking Time
one geometric quan- W (b) tc (hr)
tity such as some char- 6 to 8 3 to 3
acteristic body dimen-
8to 12 3 2 to4 1
sion, L, along with the
quantities that charac- 12 to 16 42 to 5 2
terize the heat trans- 16 to 20 5 to 6 1
fer in the dimensional
20 to 24 6 2 to 7
analysis. Roasting tur-
key involves a con-
stant oven tempera-
ture Ts (3250F[91). The bird is done when the center of the
stuffing reaches a temperature To (165"F[91). The heat trans-
fer is limited by heat conduction through the bird and stuff-
ing, whose thermal conductivities and diffu-sivities are kB,
ks, and a, and as, respectively, and are assumed constant
for all turkeys. Hence, tc is obtained from a solution to the
three-dimensional unsteady-state conduction equation in the
turkey and the stuffing:
=aBV2T (31)
SasV2T (32)
The initial and boundary conditions are given by
T=TI at t=0 (33)
T=Ts at the surface of the turkey (34)

T+ =T- atthe interfacebetweenturkeyandstuffing (35)
kgVT + =ksVT at the interfacebetweenturkeyandstuffing (36)
VT=0 along the plane of symmetry in the turkey (37)
Applying Steps 4, 5, and 6 leads to the following dimen-
sionless variables:
_i T-T t tB 2=L2V2 (38)
Ts-Ti L2
Introducing these variables into Eqs. (31-37) implies the
following minimum parametric representation:

tcCU=fI (T0-T, ( a. k.
Ltc [( k,...,geometricratios (39)
L 2 (s-T Ti Js ks

Hence, for geometrically similar turkeys and constant physi-
cal properties, tc L2. For a spherical turkey body, LocW1/3.
Hence, tc = KW213, where K is a proportionality constant
determined from the data in Table 1. The following correla-
tion fits these data with an R2 = 0.994:
tc=0.864W2/3 (40)
Hence, Mom and Dad's 28-lb bird will require a tc = 8 hours.
Note that a naive application of the Pi Theorem would imply
five dimensionless groups (i.e., n=9 and m=4) in addition to
the geometric ratios.

A recent patent describes a hollow fiber membrane blood
oxygenator that offers a 300% increase in 02 and CO, mass
transfer.[l' This is achieved by oscillating the hollow fiber
membrane module relative to the blood flow. This enhances
the mass transfer on the blood side where O diffusion is
limiting. We seek to correlate the mass-transfer coefficient
for this device by considering a single oscillating hollow
fiber as shown in Figure 4.
An analytical solution has been developed for the hydro-
dynamics when the membrane tube bundle is pulsated har-
monically at a frequency w and amplitude A; the velocity
profile, Vz, is of the form[I'

z_ Vz=r, rt, v A(41)
fV IR _TOR2' V )
where V is the volume-average velocity, R is the hollow-
fiber radius, v is the kinematic viscosity of the Newtonian
fluid, and t is the time. We seek a correlation for the mass-
transfer coefficient defined in terms of the log-mean driving
force, ACim, and the overall length of the tube, L, as follows:

27h/o L
Nw CD C (42)
kL-ACI 2tLACm dzdt (42)
0 0 r=R

Chemical Engineering Education

Figure 4. Schematic of a single hollow fiber in a
membrane lung oxygenator.

ACim[(Cw-C&-CW-C0) ((Cw (CL)
S (Cw-C) (43)
ACam-=[(Cw-CL)-CW-CO, n (Cw-Co) 4)

in which Nw and C, are the mass-transfer flux and concen-
tration, respectively, at the blood side of the membrane, Co
and CL are the concentrations at z=0 and z=L, respectively,
where L is the length of the hollow fiber, and D is the binary
diffusion coefficient. The concentration in Eq. (42) is ob-
tained from a solution to the axisymmetric form of the ad-
vective diffusion equation in cylindrical coordinates:

ac aC Da(ac C
at +Vz = -- r r 3 (44)
dt T- z-r Tr( -5-r )

Axial diffusion is neglected based on scaling arguments.[2'3
The boundary and periodic solution conditions are

C=Cw at r=R (45)
c=0 at r=0 (46)
C=Co at z=0 (47)
Ct.rz =Clt+21/o,r,z (48)

Applying Steps 4,5, and 6 leads to the following dimen-
sionless variables:
C-Co r -z
C--C r- t-t (49)

Introducing these into Eqs. (42) and (44) leads to

kR (n 1 -CL) 21 I
kLR -I fl f ) dzdt (50)
D 2rpC J J r
0L 0 r=l

coR2 ac 2Gz ac aC a (51)
D +t zrr -^ (51)

where Gz- VR2/2DL=xRPe/2L is the Graetz number and
Pe is the Peclet number. Equation (50) implies that Sh, the
Sherwood number, is a function of the dimensionless groups
involved in determining CL and adC/b at the fiber wall and
hence will be functions of only the dimensionless groups
involved in solving Eq. (51); therefore

Sh=f(Gz,0oR2 R2 GA) or Sh=f Gz,Sc, R2 AI) (52)
D v V D V


V Ar

V, = Acocoswot

Summer 2000

where Sc-v/D is the Schmidt number, introduced by using
the transformation given by Eq. (2). For large Sc (i.e., for
blood) we can use the expansion (in Sc') suggested by Eq.
(3) to conclude that the oxygenator performance can be
correlated in terms of only four dimensionless groups. Note
that a naive application of the Pi Theorem would imply that
eight dimensionless groups would be required (i.e., n=ll
and m=3).

The Pi Theorem will yield the minimum number of di-
mensionless groups, if one can determine the proper quanti-
ties and units to use. But this requires considerable physical
insight, often well beyond the experience of many students.
The alternative method proposed here directly yields the
minimum parametric representation without having to in-
voke any sophisticated reasoning concerning how variables
appear in certain combinations or the proper units for the
particular physical system. Hence, in the words of my young
daughter, it is "more equal!"

The author acknowledges the two reviewers who provided
considerable constructive criticism. He also gratefully ac-
knowledges his daughter Brigette (who now has a degree in
mathematics!) for her thought-provoking observations on
"equal" and for drawing Figure 3, which she asserts is not a
caricature of her father!

1. Bridgman, P.W., Dimensional Analysis, Yale University
Press, New Haven and London (1922)
2. Krantz, W.B., "Scaling Initial and Boundary Value Prob-
lems as a Teaching Tool for a Course in Transport Phenom-
ena," Chem. Eng. Ed., 4, 145 (1970)
3. Krantz, W.B., and J.G.. Sczechowski, "Scaling Initial and
Boundary Value Problems: A Tool in Engineering Teaching
and Practice," Chem. Eng. Ed., 28, 236 (1994)
4. Bird, R.B., W.E. Stewart, and E.N. Lightfoot, Transport
Phenomena, John Wiley & Sons, New York, NY, 60 (1960)
5. Hellums, J.D., and S.W. Churchill, "Simplification of the
Mathematical Description of Boundary and Initial Value
Problems," AIChE J., 10, 110 (1964)
6. Andrews, G.F., "Dimensionless Education," Chem. Eng. Ed.,
17, 112 (1984)
7. Churchill, S.W., "A New Approach to Teaching Dimensional
Analysis," Chem. Eng. Ed., 30, 158 (1997)
8. Ray, R.J., "A Rayleigh Free Convection Compliant Ice Front
Model for Sorted Patterned Ground," MS Thesis, University
of Colorado, Boulder (1981)
9. Crocker, B., Betty Crocker's Cookbook, General Mills, Inc.,
Golden Press, New York, NY (1972)
10. Krantz, W.B., R.R. Bilodeau, R.J. Elgas, and M.E. Voorhees,
"Blood Treatment Device with Moving Membrane for En-
hancing Mass Transfer," U.S. Patent No. 5,626,759, issued
May 6, 1997
11. Krantz, W.B., R.R. Bilodeau, M.E. Voorhees, and R.J. Elgas,
"Use of Axial Membrane Vibrations to Enhance Mass Trans-
fer in a Hollow Tube Oxygenator," J. Membrane Science,
124, 283 (1997) O

1=1 curriculum




Part 1. Curriculum Design

Worcester Polytechnic Institute Worcester, MA 01609

Engineering education in the United States today faces
many challenges, including: 1) attracting students
with a diversity of backgrounds, learning styles, and
pre-college preparations for engineering careers; 2) main-
taining interest and motivation during a four-year under-
graduate education, while at the same time assuring quality
and relevance to engineering practice; 3) preparing students
for demanding careers that not only require technical com-
petence in an engineering discipline but also require com-

William M. Clark is Associate Professor of
Chemical Engineering at WPI. He holds BS and
PhD degrees in chemical engineering from
Clemson University and Rice University, respec-
tively, and has 13 years of experience teaching
thermodynamics, unit operations, and separa-
tion processes. His educational research focuses
on developing and evaluating computer-aided
learning tools.

David DiBiasio is Associate Professor of Chemi-
cal Engineering at WPI. He received his BS, MS,
and PhD degrees in chemical engineering from
Purdue University. His educational work focuses
on active and cooperative learning and educa-
tional assessment. His other research interests
are in biochemical engineering, specifically bio-
logical reactor analysis.

Anthony G. Dixon is Professor of Chemical
Engineering at WPI He holds a BSc degree in
mathematics and a PhD degree in chemical en-
gineering from the University of Edinburgh. His
research has included development of interac-
tive graphics software to aid in teaching process
design and mathematics to engineers.

munication, teamwork, and life-long learning skills; and 4)
maintaining or enhancing quality programs in the face of
increasing financial pressure.['2] Since the traditional ap-
proach to chemical engineering education was designed for
a somewhat different set of challenges, we question whether
it is well suited to meet today's needs.
In the traditional approach, the chemical engineering cur-
riculum provides a compartmentalized sequence of courses
that aims to build a solid, fundamental foundation before
providing integrated, capstone and/or engineering practice
experiences in the senior year. Problems that arise from this
educational structure include
Lack of motivation for learning fundamental material
Poor retention of sophomore- and junior-level material that
is needed for the senior-year integrated experiences
Segmented learning resulting in a lack of ability to integrate
material presented in several different courses
Lack of ability to extrapolate knowledge and skills gained
in one context (e.g., thermodynamics) to a different context
(e.g., thermodynamic limitations in reactor design)
While cognitive science indicates that repetition is central to
learning, 31 all too often important material is presented once
and assumed to be "learned." Moreover, the traditional lec-
ture format has not been conducive to accommodating dif-
ferent learning styles or to a desirable shift away from pas-
sive learning to active learning.[4,'5
To address the challenges and deficiencies noted above,
we have developed a project-based, spiral curriculum for our
chemical engineering sophomore year. The new curriculum
is "spiral" because the understanding of basic concepts and
their interrelations is reinforced by revisiting them in differ-
ent contexts with increasing sophistication each time. It is

Copyright ChE Division of ASEE 2000

Chemical Engineering Education


The new curriculum is "spiral" because the understanding of basic concepts and their
interrelations is reinforced by revisiting them in different contexts
with increasing sophistication each time

"project-based" because students learn and apply chemical
engineering principles by actively completing a series of
projects (including open-ended design projects and labora-
tory experiments) throughout their first year of study, rather
than by simply passing a series of tests on related but com-
partmentalized subjects in a lecture-based course sequence.
In this series of papers we will describe the design, imple-

mentation, and evalua-
tion of the new curricu-
lum. This paper pre-
sents the philosophy,
objectives, and curricu-
lum design. Subsequent
papers will describe the
details of the projects
and curriculum, our
implementation experi-
ences, and our extensive
assessment efforts. Al-
though some features
of the new curriculum
are unique to Worces-
ter Polytechnic Insti-
tute (WPI), we antici-
pate that much of it
will be transferable to
that our approach can
neering disciplines.

Typical Schedule of WI

Term A Term B
(Fall Semester)
Sophomore Material and Classical
Energy Balances Thermodynamic
Junior Fluid Mechanics Heat Transfer

Senior Unit Operations Unit Operation
Laboratory I Laboratory II
Process Design Chemical Plan
and Economics Design Project

other settings and timetables and
serve as a model for other engi-

The problems noted above are neither newly discovered
nor limited to chemical engineering. There are ongoing ef-
forts aimed at addressing these same problems in engineer-
ing programs across the country. These efforts can be placed
into three main categories that differ in approach from the
one described here. First, there are programs aimed at inte-
grating math, science, English, and engineering subjects at
the freshman and sophomore level before beginning disci-
pline-specific studies. Drexel's E4 program, some of the
National-Science-Foundation-supported Engineering Educa-
tion Coalitions, and several other programs have focused on
providing an active learning integrated curriculum that in-
troduces engineering practice to freshman and sophomore
students.[6-131 All of these programs focus on interdiscipli-
nary or general engineering principles at the earliest level of
engineering education, whereas our new curriculum is di-
rected toward more in-depth study of core chemical engi-

neering courses. The project-based, spiral curriculum could
thus follow one of the newly developed interdisciplinary
introductory programs, or it could follow a more traditional
basic math-and-science introductory curriculum, as is cur-
rently the case at WPI.
The second type of related-but-different approach to re-
form is aimed at bringing the excitement of engineering
design to the fresh-
man level as a moti-
vational introduction
rE 1 to engineering with-
I ChE Core Courses out necessarily reor-

Term C Term D ganizing the entire
(Spring Semester) freshman experience.
Mixture Staged Separation In some cases there
:s Thermodynamics Processes are cross-disciplinary
Mass Transfer Kinetics and "introduction to engi-
Reactor Design neering" courses[14-151
and in others there are
s Process Control Applied Math and in others there are
Laboratory forChem.Eng. discipline-specific
Scourses.[16-17] The
third type of reform
effort aims to pro-
vide design across
the curriculum by integrating design into existing courses
throughout the curriculum.1t8-211
Virtually all of the recent reform efforts have incorporated
proven learning-enhancement strategies of active, coopera-
tive learning, 2223' and problem-based or project-based learn-
ing.124251 We have also used these strategies, but what distin-
guishes our program is that we have completely reformed
the first set of core chemical engineering courses to em-
phasize these features and to integrate material that is
normally taught in a compartmentalized sequence of fun-
damental courses.

WPI has an atypical academic calendar consisting of five
seven-week terms; four during the regular academic year
and an optional, fifth one during the summer. Normally
students take three courses or activities during each of the
four academic-year terms denoted terms A, B, C, and D, and
complete their studies in four years. Our A and B terms
correspond to Fall semester, and C and D correspond to
Spring semester in other programs. The typical sequence of
core chemical engineering courses encountered by our stu-

Summer 2000

dents is presented in Table 1. During their freshman year,
they study chemistry, physics, calculus, and humanities or
social science electives. Their first exposure to chemical
engineering begins in their sophomore year. In addition to
the core material balance and thermodynamics courses shown
in Table 1, they normally take physical chemistry, organic
chemistry, differential equations, and more humanities and
social science courses in their sophomore year. During their
junior year, students usually take engineering electives and
complete a three-course-equivalent "interactive qualifying
project" relating science and technology to society in addi-
tion to completing the transport and reactor design courses
shown in the table. During their senior year, all of our
students complete a three-course-equivalent "major qualify-
ing project," similar to a senior thesis, in addition to the unit
operations, design, and control courses shown.
Although the format is different, the core content of our
curriculum is similar to that of most other chemical engi-
neering departments. We teach the fundamental subjects
underlying chemical engineering process analysis and de-
sign in a compartmentalized sequence of courses during the
sophomore and junior years. Then, in the senior year we ask
the students to work in teams on integrated laboratory and
design problems using those fundamentals. In addition to
assigning complex, open-ended problems for the first time,
we also emphasize teamwork and oral and written communi-
cation skills for the first time in the senior year.
This process has been likened to the following hypotheti-
cal method of training a baseball team. Suppose you take
nine people who don't know the game of baseball and train
them individually in all the fundamentals for two years; two
months on throwing, three months on catching, five months
on hitting, etc. Then, without ever having them practice, or
even watch a game, you suddenly ask them to play the game
properly as a team. Many would likely quit after the first few
months because they didn't like throwing the ball over and
over when they didn't know why they were doing it.
Those that survived the program would probably play
well at the end, but they'd have bruises from those first
few games when they knew the fundamentals but not
how to put them together.
Our students often complain that the first half of their
senior year was the hardest thing they have ever done, but at
the same time acknowledge that it was their best educational
experience. They recognize that solving practical laboratory
and design problems and communicating their results forced
them to relearn and better understand the fundamentals as
well as prepared them for the role of a practicing engineer.
Part of our motivation for the project-based, spiral curricu-
lum was to bring some of these rich senior-year experiences
into the earlier years.
Although we have recently begun incorporating active and
cooperative learning exercises within some of our courses,

Goals of the New Curriculum

Integrate material from our first four core courses
Reinforce key concepts by repetition with increasing complexity
Provide semirealistic applications of fundamentals
Provide laboratory and design experiences
Emphasize active learning
Integrate computer use throughout the curriculum
Introduce AspenPlus to sophomores
Improve student motivation for learning fundamentals
Improve problem-solving abilities
Improve mastery of fundamentals
Improve communication skills
Improve teamwork skills
Maintain individual accountability
Promote lifelong learning
Improve attitudes and satisfaction with chemical engineering
Use computer-aided instruction and peer learning assistants to
maintain costs

the format of lecture/followed by homework/followed by
test, dominated the learning process. Our students use com-
puters for word processing, spreadsheets, math packages,
programming, and the process modeling and design program
AspenPlus, but there is no emphasis on computer use and no
specific computer skills development strategy. AspenPlus is
not used until the senior-year design course.

The goals of our new curriculum are listed in Table 2 and
can be seen to be consistent with the goals of ABET's
Engineering Criteria 2000.[26J These goals should result in
students who can work effectively in teams to combine
material and energy balances with thermodynamics, trans-
port phenomena, chemical kinetics, and reaction engineer-
ing to analyze and design chemical processes.
In considering how best to achieve these goals, we used
our baseball analogy and wanted students to play some prac-
tice games and enjoy what they were doing as they devel-
oped their fundamental skills. We wanted them to encounter
some semirealistic chemical process analysis and design
problems throughout their chemical engineering studies,
rather than only at the end. We hypothesized that a series of
well-structured projects could provide motivation for learn-
ing fundamentals as well as provide practical applications of
those fundamentals. Integrating material throughout the cur-

Chemical Engineering Education

riculum would reinforce interrelationships between subjects and help
develop abilities to solve realistic problems.
To completely fulfill our plans, we realized, however, that the
entire first two years of chemical engineering courses would have to
be reorganized. Beginning knowledge and skills from several tradi-
tional courses needed early introduction to accommodate meaning-
ful, but carefully structured, early projects. Knowledge and skills
could then be added on a "just-in-time" basis to help students progress
through a series of projects with increasing complexity. There was
no need to change our senior year, because it already had integrated,
project-based, team-oriented laboratory and design experiences.
Although we realized that integration of all material from the
sophomore and junior years was important, we decided to focus only
on the sophomore year. We thus sought the more modest goal of
integrating material and energy balances with thermodynamics and
stage-wise separation processes, hoping to produce rising juniors
who could work in teams to combine these subjects to analyze and
design processes, albeit those without regard to rate behavior. Rea-
sons for neglecting to integrate the transport and reactor courses into
our new curriculum included: 1) complete reform of two year's
curriculum seemed unmanageable; 2) meaningful projects could be
done that did not require rate information; 3) some of our transport
courses are taken by non-chemical engineering students and/or are
taught by non-chemical engineering professors; 4) many of our stu-
dents study off-campus for one or more terms during their junior
year, creating scheduling problems with a year-long integrated jun-
ior-year course; and 5) senior-year courses provide an opportunity to
integrate the rate material with other topics.
Since a major revision of the sophomore year was required, we
took the opportunity to introduce other desirable features into the
new curriculum. As shown in Table 2, most (but not all) of the goals
followed directly from our desire to produce students with the ability
to "play the game" and not just those who could "hit well in batting
practice." Although not a specific goal of our new sophomore-year
curriculum, one additional positive outcome might be that the student's
senior year becomes more enjoyable as well as more productive. This
might happen because students who go through the new curriculum

Material and Classical Mixture Staged Separation
(a) Energy Balances Thermodynamics Thermodynamics Processes



SaImpl Coepex

Figure 1. Rescheduling the traditional curriculum from simple-to-
complex. (a) traditional four courses; (b) within each course mate-
rial flows from simple (blue) to complex (red); (c) a new year-long
course with material from each of the four traditional courses
rearranged from simple to complex.

as sophomores will have experience with team-based
integrated projects before their senior year.

To develop our new curriculum, we itemized and
prioritized detailed learning objectives for the four
traditional sophomore-year courses (see Figure la).
We noted that in each course the material progressed
from simple to complex, as illustrated with the color-
coding in Figure lb. In one sense, what we sought
was a year-long course that integrated topics from the
four traditional courses by teaching the beginning
material from each course in a new first course, fol-
lowed by the intermediate material from each tradi-
tional course in a new second course, and so on, as
illustrated in Figure Ic. We also wanted to revisit key
concepts throughout the year and to emphasize the
connection of ideas normally presented separately in
separate courses. We therefore developed the "spiral"
curriculum concept, shown schematically in Figure 2.
The sophomore year was divided into four levels
shown in the vertical direction of the diagram and
corresponding to our four terms. At WPI the four
levels correspond to discrete courses, but that need
not be the case. For example, in a semester system,
the material from levels 1 and 2 could be taught in a
single 5-6 credit semester-long course.
Our four traditional courses are shown at the base
of the diagram to provide a reference frame for com-

Level 4

Level 3

Level 2

Level 1

Figure 2. Schematic diagram of the spiral

Summer 2000

prison. Students begin the new curriculum at Level 1, where
they are introduced to the basic skills and concepts from all
four traditional courses. In Level 2, in addition to introduc-
ing new material we build on the previously acquired skills
and concepts by requiring them to be re-used and extended
to more complex tasks. The succeeding two levels follow
similarly, with the students revisiting topics met before at
lower levels, extending them to more sophisticated uses and
ideas, as well as acquiring new knowledge and concepts
needed to address more challenging problems.
Table 3 presents the results of prioritizing and rearranging
important topics from our sophomore year into a spiral cur-
riculum with four levels. At Level 1, we introduced simple

material and energy balances with no recycle, the thermody-
namics of pure components, first-law energy balances, ideal-
phase equilibria, and simple flash separations. At Level 2,
students were exposed to ideas of recycle, staged separation
systems, and the applications of energy balances to flow
systems. Non-ideal gas-phase behavior was introduced
through entropy concepts in the analysis of flow processes
and the use of real gas relations, including residual proper-
ties. The students' experience with separation equipment
was broadened first by extensive coverage of distillation at
the start of the level, followed by a short look at isothermal
gas absorption toward the end. In Level 3 the focus was on
the properties of mixtures, especially non-ideal solutions

Curriculum Material by Level

Level Topics

Level Topics

1 Material balances and stoichiometry
unit conversions, temperature and pressure scales, mass, mole fractions
stoichiometry, conversion, yield, limiting and excess reactants
density; ideal gas law and partial pressure
non-reactive material balances; choice of basis
material balances on reactive systems; tie components
Energy balances-first law
properties of pure fluids
gas-liquid systems, relative saturation
phase rule, vapor pressure, Raoult's law
W, Q, U, P1 law for closed systems; reversible systems
enthalpy, 1" law for steady-flow processes
thermodynamic data: steam tables, AH, cp
heat effects for phase changes
energy balance applications
Introduction to staged separations
binary VLE, y-x diagram, bubble and dew-point calculations
multicomponent VLE; K-factors
single-stage binary flash
single-stage multicomponent flash
multistage distillation and external column balances
introduction to McCabe-Thiele methods

2 McCabe-Thiele methods for binary distillation
stage-to-stage calculations
enriching and stripping sections, feed line
effect of reflux ratio, plate efficiency
introduction to non-CMO methods
2nd law, thermodynamics of steadyflows
material balances on reactions, excess air
heat effects of industrial reactions
entropy, 2"d law of thermodynamics
Carnot heat engine, thermal efficiency
combined law of thermodynamics, fundamental property relations
steady-flow processes, efficiencies of flow devices
power cycles
thermodynamic analysis, thermodynamic efficiencies
Material balances with recycle; real gases; absorption
material balances with recycle, purge, and by-pass
real gases and compressibility; cubic equations of state
real gas mixtures, Kay's rule
critical properties, acentric factor and principles of corresponding states

residual properties and compressor, turbine analysis
isothermal absorption of gases in staged equipment
Kremser equations for dilute gas absorption; plate efficiencies

3 Property changes on mixing
partial molar properties, ideal solution, and excess properties
heat effects of mixing-heats of solution, formation
enthalpy-concentration charts; adiabatic mixing
Solution thermodynamics and VLE
phase rule, vapor-liquid equilibrium in ideal mixtures
corrections to ideal-solution behavior, activity coefficient models
excess Gibbs energy and activity coefficient models
chemical potential and equilibrium criterion
fugacities and fugacity coefficients
activity coefficients and standard states
low-pressure VLE calculations
solubility of a gas in a liquid; VLE from EOS
calculation of fugacity for pure components, mixtures
azeotropes and distillation
Liquid-liquid extraction
liquid-liquid equilibria
immiscible extraction: LLE, single stage
immiscible extraction: multistage methods
miscible extraction: LLE, lever rule
miscible extraction: single-stage and cross-flow
miscible extraction: multistage cross-flow

4 Chemical reaction equilibria
reaction coordinate
standard heat of reaction, standard Gibbs energy of reaction
evaluating equilibrium constants
relating equilibrium constants to composition
equilibrium conversion for single reactions
Le Chatelier's principle
multireaction equilibria
Unsteady-state balances
transient material and energy balances; filling tanks and cylinders
staged batch distillation
Combined material and energy balances
non-CMO distillation, Ponchon-Savarit method
psychrometry; psychrometric charts; adiabatic humidification
computer-aided material and energy balances; degrees of freedom
simultaneous material and energy balances on reactive processes

Chemical Engineering Education

and phase equilibria. Property changes
on mixing were followed by vapor- Impor
liquid equilibria and, finally, liquid- from lo
liquid equilibria. In the latter two cases,
the thermodynamic material was were rei
coupled strongly to applications involv- more soj
ing distillation of azeotropes and liquid- at higl
liquid extraction, respectively. Finally, and e
in Level 4, chemical reaction equilib-
rium was covered, followed by advanced contain
process calculations, including unsteady from eac
material balances and simultaneous ma- tradition
trial and energy balances. We also pro- We att
vided brief exposure to the process simu-
lator, AspenPlus, in Level 4. di
It should be noted that Table 3 indi- tradition
cates the level at which a topic is first evenly
introduced. Important topics from lower allfoj
levels were revisited with more sophis- but
tication at higher levels, and each level
contained material from each of the four not
traditional courses. We attempted to dis- po0
tribute the traditional material evenly
throughout all four levels, but this was
not always possible. Material balances, for example, were
introduced early in Level 1 for acyclic systems, including
stoichiometry and reactive systems. Material balances on
reactions were revisited in Level 2 for heat effects associated
with combustion, then were more formally extended to in-
clude recycle systems. Little formal instruction on material
balances took place at Level 3, but at Level 4 the topics of
unsteady material balances and combined material and en-
ergy balances, with reaction, were taken up. Topics in staged
separations were distributed quite successfully throughout
the curriculum, coupling somewhat with the student's in-
creasing sophistication in the use of phase equilibria. Distil-
lation, in particular, appeared in some form in all four levels,
moving from simple flash distillation to staged binary distil-
lation to distillation of azeotropic mixtures to unsteady staged-
batch columns and non-constant molal overflow operations.
The hardest material to fit into the spiral form was solution
thermodynamics, since it is conceptually more advanced for
most students. Level 1 used Raoult's law for vapor-liquid
equilibria, but we did not find it advisable to develop this
theme further until Level 3, when the usual topics in
VLE were covered. Nevertheless, we found that spiraling
of separation processes eased the introduction of solution
This new curriculum forced repetition of high-priority
learning objectives throughout the entire year and empha-
sized their connection to ideas usually presented entirely
separately in a later course. Low-priority learning objectives
were de-emphasized and some were omitted, subscribing to


Summer 2000

the "less is more" philosophy that pre-
t topics fers a clear understanding of key con-
r levels cepts over superficial exposure to al-
most everything. Thus, by the end of
ted with the year every student should realize
stication that chemical engineers are called upon
levels, to combine material and energy bal-
level ances with thermodynamic information
to analyze or design processes.
ateri The spiral curriculum was structured
f the four around a series of industrially relevant
courses. cooperative-group projects that served
pted to as a framework for achieving the learn-
e the ing objectives for each level. Within
each level in Table 3, topics are grouped
material together under headings that describe
ughout projects designed for each level. In
levels, Level 1, for example, the initial project
focused on material balances and sto-
ichiometry, the second focused on en-
rays ergy balances, the third introduced
le. staged separation processes. Some
projects were design oriented, some
were mostly analysis, and others in-
cluded laboratory experiments. Project deliverables included
written reports and sometimes included oral reports. The
projects themselves are described in detail in the second
paper of this series.

We taught the spiral curriculum to one-third of the 1997-
98 sophomore class and to one-half of the sophomore class
in 1998-99. The other sophomores were taught by the tradi-
tional curriculum each year and were used as a comparison
for assessment of our curriculum reform. The spiral curricu-
lum was delivered through a variety of channels, including
cooperative-group projects, traditional lectures, homework
problems, in-class active learning sessions, interactive mul-
timedia learning tools, and laboratory experiments. To as-
sure individual accountability, individual homework grades
were recorded and an individual test was given at the end of
each project period. A thorough understanding of the projects
prepared students for most of the material on the tests, but
some material was covered only in supplemental lectures
and homework problems. Details of our delivery meth-
ods and our implementation experiences will be given in
Part 2 of this series.
Our overall project assessment goals were to evaluate how
the project-based, spiral curriculum affected students' abil-
ity to: solve problems at several levels of cognition, work in
teams, work independently, master the fundamentals of
chemical engineering, and integrate material from several
Continued on page 233.

, outreach


(About Chemical Engineering)

Loughborough University Loughborough, Leicestershire, United Kingdom LE11 3TU

here is concern in the UK academic community over
a decline in university applicants to chemical engi-
neering (see Table 1). There is an almost instinctive
reaction to attempt to account for the trend, but recalling
Sherlock Holmes' stern injunction that to theorize in the
absence of data is a "capital mistake," we shall refrain. What
would constitute hard data in this case is difficult to con-
ceive. One is dealing with opinions and views formed over
long periods of time that were subject to a host of influences.
Even when a clear question can be formed, such as "Why do
females account for only approximately 25% of applicants?"
it rapidly becomes apparent that there are no simple answers.
Fretting over the causes of the decline will, in any case,
not result in any useful outcome. Our principal concern here
is with describing a proposal for halting (and perhaps even
reversing) the trend. It will not be an easy task, and in order
to make a real impact on the situation, it will require imple-
mentation on a large scale.
Our basic premise is that opportunities exist within the
teaching of chemistry at schools to introduce information
about disciplines allied to chemistry, i.e., chemical engineer-
ing. We describe below a scheme intended for integration
into the teaching of practical organic chemistry. Writing in
the UK, we felt it logical to set out our proposal in the
context of secondary-school (i.e., pre-university) education
in the UK. We hope that in this way the non-UK reader
would be able to compare and contrast the situation existing
in his or her own country with the UK problem. In addi-
tion, it should enable the reader to better determine the
most appropriate age for applying the proposed project in
his or her own country.
This last point is very important; we are firmly of the
belief that the project work we describe (and equally impor-
tant, its implementation) should be of universal applicabil-
ity. One of us (KH) has recently returned from conducting,

with other representatives of UK chemical engineering, a
survey of academic research at the top US departments,
and the opportunities to discuss other, related matters-
including student recruitment-proved too great to re-
sist. The impression gained from those informal discus-
sions was that the US faces a broadly similar situation in
declining enrollments.

Before embarking on a description of the project, we feel
it is necessary to offer a brief explanation of the entry pro-
cess to UK universities. The current matriculation route
(excluding Scottish universities) is via the Advanced Level
of the General Certificate of Education, commonly referred
to as 'A' levels. At 16 years of age, prospective 'A'-level
school or college candidates will have elected to sit exami-
nations in, typically, three subjects, which they study for a
period of two years. An offer of a university place is made to
individual students in the form of a cumulated 'A'-level

Gilbert Shama has been in the Department of
Chemical Engineering at Loughborough for
just over ten years. He has long been a mem-
ber of the Departmental Admissions Team and
has special responsibilities for new initiatives
aimed at boosting undergraduate recruitment.

Klaus Hellgardt joined the Loughborough
Chemical Engineering Department in 1995. He
is a relatively recent recruit to the Admissions
Team. He is responsible for overseas recruit-

Copyright ChE Division of ASEE 2000

Chemical Engineering Education

Applications to UK's
ChE Degree Courses*

Year Male Female Total
1994 1017 324 1341
1995 913 353 1266
1996 897 328 1225
1997 873 319 1192
1998 787 324 1111

*The figures shown here were
derived from data supplied by
the Universities and Colleges
Admissions Service based in
Gloucester. Under existing
conditions, students may
apply for entry to up to six
university departments in any
academic discipline. The vast
majority of students apply to
the maximum number of
departments to study a single
discipline. We have assumed
this to be the case in arriving
at the figures presented here.

Students Sitting 'A'-Level
Examinations by Subject*

Year Mathematics Physics Chemistry
1994 67,984 36,600 41,805
1995 65,892 35,234 42,836
1996 68,709 33,361 40,917
1997 70,414 33,657 42,841
1998 71,615 34,518 43,385
*Source: Qualifications and Curriculum
Authority, London, 1999

Students Sitting 'A'-Level
Examinations in the
Combination Mathematics,
Physics, and Chemistry*

Year # of Students
1994 8,507
1995 8,762
1996 8,069
1997 8,794
1998 9,754
*Source: Qualifications and Curriculum
Authority, London, 1999

score, which may or may not be accompanied by other
constraints, such as minimum grades in one or more sub-
jects. The Scottish system is different in that students there
sit the Higher Grade of the Scottish Certificate of Edu-
cation ('Highers'). Students take a broader range of
subjects somewhat at the expense of depth of coverage,
with the consequence that courses at Scottish universi-
ties are generally correspondingly longer than those
elsewhere in the UK.
'A' levels have long been criticized as requiring young
people to specialize at far too early an age. This is particu-
larly the case when compared to the majority of other
European states. (There is now a real prospect of change
being introduced, with the aim of maintaining educational
breadth beyond the age of 16 without compromising depth.
Whether or not this is achievable remains to be seen, but it
is in any case outside the scope of this paper.)
The preferred combination of 'A' levels for entry into
most UK chemical engineering departments is mathemat-
ics, physics, and chemistry. The number of students taking
any one of these particular 'A' levels has remained rela-
tively steady over the last five years (see Table 2). The
crucially important figure, however, is the number of stu-
dents presenting the combination of mathematics, physics,
and chemistry (see Table 3). The encouraging feature of
this data is that it shows an increase in students offering
this 'A'-level combination; less encouraging is the fact
that the decline in applications to chemical engineering
has been occurring against the background of this increas-
ing pool of suitably qualified young people.


Our purpose in the following proposed program of inte-
gration is to raise awareness of the discipline of chemical
engineering among our young people. There are probably
many ways of achieving this end, but whatever approach
is taken, we feel that there are certain principles that must
be upheld. The task must be seen as one of informing
young people, not as one that attempts to entice them away
from other disciplines. It has been our experience that one
gains the respect of audiences of young people if one
states, and adheres to, this principle.
We do not wish to be seen as claiming that the situation
described here has previously gone unrecognized. It hasn't.
Both professional and industrial bodies have expended
considerable time and money in producing educational
materials in a variety of formats, including leaflets, post-
ers, videos, and CD-ROMs aimed at exciting interest about
chemical engineering in the minds of young people.
Whether or not the producers of such materials realize
how fierce the competition is for those young minds is
another matter.

... .the

decline in


to chemical


has been


against [a]


of [an]


pool of





Summer 2000

Those most immediately affected by falling admis-
sions (the universities) have responded to the situation
by offer ing short residential courses aimed at giving
young people "hands-on" experience in the different
branches of engineering. Most of these schemes are
aimed at 16-year-olds, those about to embark on their
'A' levels. The majority of these students will have
already made up their minds whether they will choose
Arts or Science 'A' levels. This latter group is the one
from which engineering departments recruit students.
While such recruitment schemes are admirable in
achieving their rather limited purpose, they can actu-
ally serve only to attract students away from the pure
sciences or to redistribute students among the various
engineering disciplines. This is not intended as a cyni-
cal criticism of these so-called "taster courses," since
in many cases they have helped individuals form opin-
ions about their future careers. Nor do we argue that
these courses should no longer be offered. But it is
evident that from a recruitment point of view, they
have only limited impact, and applicants for chemi-
cal engineering continue to drop despite the exist-
ence of such courses.
It should be equally clear that the time to promulgate
the message about the virtues of a career in engineer-
ing is before the age of 16. At 14, for instance, fewer
children will have strongly held ideas as to their
future careers. Even more important, they will not
yet have committed themselves to some branch of
intellectual endeavor (e.g., Arts or Sciences) as the
British system requires.
Those who have organized engineering taster courses
appreciate the resources, and the time, that must be
expended to operate them successfully. Insofar as en-
gineering is about making things happen, about doing,
it is difficult to convey this sense of achievement in
leaflets or videos. It is relatively simple for a univer-
sity department to produce 20,000 leaflets extolling
the virtues of chemical engineering, but quite a differ-
ent thing to actually enable 20,000 individuals to ex-
perience chemical engineering first hand.
No single department could hope to bring about
significant changes on its own; instead, a concerted
effort is required. In one sense, approaches to younger
children should help eliminate some of the rivalry that
exists between departments; recruitment by universi-
ties of students who attend their existing taster courses
aimed at 16-year-olds is not insignificant. At 14, it
could be argued, the objective becomes more one of
gaining a convert to science and engineering as a
whole. Also, at that early age it is unlikely that an
individual would commit him or herself to any one

You are to carry out the synthesis of methyl salicylate. This is one of the
active ingredients in the traditional remedy known as "Oil of Wintergreen."
Methyl salicylate has a pattern of action similar to that of aspirin, i.e., cure
for headaches and other general low-level aches and pains. Take I mol of
salicylic acid (2-hydroxybenzoic acid) and react with 5 mol of methanol,
using 0.2 mol concentrated sulfuric acid as a catalyst. The reaction should
be carried out under reflux conditions (boiling of the mixture) for 5 hours.
Extract the product into an equal amount ofcyclohexane. Wash the organic
phase with aqueous sodium carbonate solution and dry subsequently over
anhydrous sodium sulfate. Distil to remove the cyclohexane and analyze
your product.
The reaction scheme is given below:
o1 + CH3-OH 0 o + HO

What does the sulfuric acid do? Why do you use excess methanol: Why do
you wash with sodium carbonate solution?
You will have to assemble your own apparatus from the glassware provided.
This is a simple and safe synthesis to carry out provided that you adhere to
normal safety procedures.

Figure 1. Instructions for the laboratory synthesis of
methyl salicylate.

You are required to prepare an outline design and provide preliminary
costing for an industrial facility to produce 300 tonnes per year of methyl
salicylate. Methyl salicylate is produced by the reaction of salicylic acid and
methanol in the presence ofH,SO as follows:
H2SO4 (MW=98)
C6H4.0H.COOH + CH3.OH + C6H4.OH.COOO.H3 + HO2
Salicylic acid Methanol Methyl salicylate Water
MW=138 MW=32 MW=152 MW=18
The reaction takes 5 hours at the boiling point of the mixture. The methyl
salicylate product separates out as an immiscible liquid layer, which subse-
quently has to be washed with dilute sodium hydroxide and then with water.
To be cost effective, the industrial process operates somewhat differently
from the laboratory scale experiment.
The facility will be operated as a batch process, but you are free to choose
any operating pattern you consider suitable (e.g., 8 or 24 hours per day, 5 or
7 days per week, all year round campaigns, etc.). You may assume that sales
demand is spread reasonably evenly over the year.
Your task
You will be split up into design teams for the purpose of this exercise and
you are expected to elect one of your team members as team leader and s/he
will be responsible for co-ordinating the teams' activities and for ensuring
that a design is produced in time to make a brief presentation. For this, you
will be issued with only two overhead transparency sheets. You should
consider the following:
* How you are going to make the reaction happen.
* How the plant is to be operated.
* The size and cost of your reactor (in n').
* How you will get the raw materials into the reactor.
* How you are going to avoid environmental pollution and what opportu-
nities exist for recovery of raw materials.
What other considerations would have to be taken into account in
determining whether your plant will operate at a profit.
Will this process yield a saleable product? If not, what additional steps
need to be taken ?

Figure 2. Design specification for a plant to produce
methyl salicylate.

Chemical Engineering Education

particular department. We believe that the proposal described
in this paper could form the basis of an initiative aimed at
increasing awareness of engineering in these young people.

We decided to try our recruitment scheme on 14-year-old
children from a local school. We felt that this age group
would have a sufficient knowledge of chemistry for the
project and would be able to tackle what would be some
quite novel concepts. Both we and their teachers felt the
project should begin in an environment familiar to them-
their own school laboratories. The proximity of the school to
the university also meant that the children could continue to
make use of school resources such as the libraries, internet
access, etc. Involvement of a
local school had other ben-
efits; in particular, it was un- T
necessary to arrange accom- Progr
modation and catering. I TAoo,

Involving the school's
teachers was a key factor in
the success and smooth run-
ning of the project. They were
invited to the university about
a week before the program
was implemented and were
fully briefed on logistic,
safety, and other matters.
They were given a brief over-
view of our laboratories (we
intended that the school chil-
dren would make use of them,

as explained below). Overall supervisory and other duties
were shared between two members of the academic staff of
the department.
The students were asked to undertake laboratory synthesis
of methyl salicylate. The instructions given to them are
shown in Figure 1. We found it convenient to split the
students up into teams of three or four, with each team being
given one set of the necessary apparatus. The exercise proved
to be well within their capabilities and was readily executed
by the students over a two-day period.
The next stage of the exercise was for the students to
design a facility for producing 300 tonnes per annum of
methyl salicylate. This task was performed at the university.
Before starting this part of the program, the students were
reminded of the operations they had already conducted and
which they would now have to "translate" to a larger scale.
We achieved this by mimicking the operations using water
in the place of methanol and sulphuric acid and sugar in the
place of the salicylic acid. By removing the chemistry, the
students were able to concentrate on the actual process of

introducing the reactants into a vessel, heating them, mixing,
cooling, etc. Although simple in concept, it proved highly
effective. The design brief given to them is shown in Figure 2.
For this exercise we found it preferable to divide the
students into teams of not more than six. Groups of this size
helped break down the reserve felt by some individuals and
to lead to lively discussions. It is essential to let the students
generate ideas themselves, but it is equally important to have
someone on hand to help them eliminate ideas that are evi-
dently impractical, as well as to keep them on track gener-
ally. Both authors have personally carried out this role, and
we have also made use of postgraduate demonstrators, who
were carefully primed for the role with emphasis on the need
to gently guide the students toward a solution.

The exercise can, to quite
a large extent, be tailored to
fit time constraints. We have
operated it comfortably over
a period of five days; two
days in the chemistry labo-
ratory and three at the uni-
versity (see Table 4). In or-
der to maintain high levels
of enthusiasm among the
students, we found it useful
to intersperse this period of
time with occasional forays
into our teaching laborato-
ries, where the students were
first shown, and then al-
lowed to operate, pumps,
pneumatic solids conveying

equipment, valves, heat exchangers, and other process plant
equipment, Naturally, this was done under very closely su-
pervised conditions. For this the children were divided into
two groups, A and B, as shown in Table 4.
We feel it is essential that students obtain an overall pro-
cess flowsheet relatively early in the exercise. It is also
important to illustrate the depth of reasoning and the rigor
that is required in the design of individual items of a
chemical plant. We have tended to focus on the reactor as
it is central to the process and therefore consideration of
what happens both up and downstream of it needs to be
taken into account.
Rather than overburdening the students with large collec-
tions of data sheets and the like, we have found it preferable
to provide them with information as the need for it arose. For
example, the students were required to make use of the
library to determine what materials were suitable for fabri-
cating the reactor. Another task, that of finding the bulk
selling price of methyl salicylate, required them to make use
of the internet. An advantage of this approach is that it does

Summer 2000

am Timetable

W d nar^d Th..rc ndr Fridr

nl u ay ues uay e ullues ay urs au y I ua
Introduction to Introduction to
the program the strategy
Laboratory of design Group A labs Design
-synthesis Group B design (both
Introduction to Group A labs groups)
chemical Group B design
Lunch Lunch Lunch Lunch Lunch
The chemistry Visit to Group A design Group A design Design
of salicylates industrial Group B labs Group B labs presentations

not appear to the students as being over-prescriptive: as long
as progress toward the final presentation is made, there is
scope for exploring at least some of the "alleys and byways."
In the current generation of young people, it is perhaps not
surprising to find the extent to which concern for the envi-
ronment crept into a design exercise such as this. This actu-
ally presents an opportunity, which if seized can help the
participants form a more
balanced view about the
chemical industry. It has Salic lic Acid
been our experience that
opinions held by most Metl Sm
of the students on this
subject tend to be nega- Sulphuric A I
tive ones; such views
may not be well-in- steam
formed, but they never-
,Reactor :; Wash
theless represent ones' W
that have to be con- interface inteface
fronted. The students detector detect
themselves raised envi-
ronmental concerns and A nueous Aqueous
were encouraged to see
that solutions were
available to meet those
concerns. More impor- Figure 3. Flow shee
tantly, they discovered produce
this for themselves. We
found that they acquired a rather proprietorial attitude-this
was their plant and they were going to operate it in such a
way as to cause minimum impact on the environment. Inter-
estingly, no student has ever questioned the need for the
plant; they seemed to recognize that such pharmaceutical
products are of general benefit to humanity.
We stated earlier in this paper that the collaboration of
industry was helpful in operating a scheme such as we have
proposed here. We should note that this needs to be done
with sensitivity. We have often been surprised at how alert
young people are to what they consider to be overt forms of
company propaganda. We wanted to ensure that the students
primarily associated the whole exercise as a collaboration
between their school and the university. We wanted to en-
sure that industrial involvement was less obtrusive, but was
nonetheless significant to the success of the scheme.
In our case, this consideration took the form of a link with
a local pharmaceutical company that undertook to conduct
chemical analyses of the reaction products that the students
had synthesized. The students were invited to the company
laboratories and given a brief tour. They were then given
a brief survey of potentially suitable analytical techniques
followed by a fuller explanation of the process (High
Pressure Liquid Chromatography, HPLC) actually used
for their samples.

t for

Chemical Engineering Education

In order to make the design exercise as realistic as pos-
sible, a number of our industrial contacts agreed to receive
telephone inquiries from the students concerning the costs of
shift labor. There are a number of forms that industrial
involvement could take, and those cited above represent
typical examples only. The students' final task was to make
a short presentation to an audience that included their peers.
This in itself was a new
experience for many of
them, and it provided an
opportunity to explain
de Water their processes and to
present their principal
We decided to pre-
pare a process flowsheet
(see Figure 3) to enable
SWthe students to see what
one looked like and also
dccto to allow them to com-
pare their designs to it.
Aqueous Methyl We wished also to in-
troduce the concept that
in engineering it is pos-
sible to generate more
a methyl salicylate than one perfectly ac-
facility, ceptable solution to the
same problem.

We have operated this exercise for two successive years,
in each case to groups of about twenty students. Feedback
from both the students and their teachers has been extremely
encouraging. It is justifiable to ask how the success of this
scheme is to be assessed, but it is still too early to say if the
initial enthusiasm expressed by the students will translate
itself into more of them selecting the right combination of
'A' levels and ultimately electing to study chemical engi-
neering, or indeed chemistry, at university. We view the
operation of our scheme very much as a pilot exercise.
In any event, we are dealing with the statistics of small
numbers, and as we have already stated, real benefits will
only become apparent if the scheme is carried out on a
significantly larger scale. The fact remains that all those who
were involved in the program were unanimous in finding it a
useful and enjoyable exercise. If it served no other purpose
than to put the chemistry that these children receive at school
into a context that is not normally available to them, then we
have performed a useful service.

We wish to acknowledge the work of our colleague, Dr.
Robin Wilcockson, in devising the design exercise that we
adapted as part of our initiative. 1


Project-Based, Spiral Curriculum
Continued from page 227.

courses. We were also interested in how it affects student
attitudes and satisfaction about chemical engineering and
their professional development within the discipline. Exter-
nal consultants were used to provide objective assessment
through a variety of qualitative and quantitative measures.
These included surveys, interviews, videotaping of class and
project work, end-of-term course evaluations, a novel sopho-
more process-design competition, and an end-of-year com-
prehensive exam. The details and results of these assessment
efforts will be described in Part 3 of this series of papers.
The following quotes from students on what they like
most about the class after our first offering of the new
curriculum support our belief that it was well received by
students and that at least some of our objectives were met:
"...this cooperative learning thing through group
projects has made this class one of the most thorough
learning experineces of my life. Found it much easier
to do assigned homework and do well on tests because
of the thought processes established while working on
a project."
"...the ability to work in groups to solve problems. I
really wasn't a big fan of group work because I could
usually do just as well on my own. I've come to realize
that groups can do so much more than an individual."
"...without step-by-step procedures we were really
forced to think and comprehend exactly what we were
doing and why we were doing it."
" taught all of us to use our heads first, then use the
book. For the first time since coming to college, it felt
as if I was learning to do something that would be very
valuable to me in the future."

This project was funded by the U.S. Department of
Education's Fund for the Improvement of Postsecondary
Education under grant number P 16B60511.

1. Guskin, A.E., "Reducing Student Cost and Enhancing Stu-
dents Learning: The University Challenge of the 1990s.
Part I. Restructuring the Administration," Change, (July/
August), p. 23 (1994)
2. Parrish, E.A., "A Work in Progress: WPI and the future of
Technological Higher Education," WPI J., 3, Fall (1995)
3. Haile, J.M., "Toward Technical Understanding. Part 3. Ad-
vanced Levels," Chem. Eng. Ed., 32, 30 (1998)
4. NSF Publication, "Report from the Presidential Young In-
vestigator Colloquium on U.S. Engineering, Mathematics,
and Science Education for the Year 2010 and Beyond,"
5. Felder, R.M., and L.K. Silverman, "Learning and Teaching
Styles in Engineering Education," J. Eng. Ed., 78,674 (1988)

6. Quinn, R.G., "Drexel's E4 Program: A Different Professional
Experience for Engineering Students and Faculty," J. Eng.
Ed., 82, 196 (1993)
7. Parker, J., et. al., "Curriculum Integration in the Freshman
Year at the University of Alabama: Foundation Coalition
Program," paper 4all, Proc. 25th Frontiers in Ed. Conf.,
Atlanta, November (1995)
8. Richards, D.E., "Developing a Sophomore Engineering Cur-
riculum: The Rose-Hulman Experience," paper 2b61, Proc.
25th Frontiers in Ed. Conf, Atlanta, November (1995)
9. Coleman, R.J., "Studio for Engineering Practice, 'STEP',
Lessons Learned About Engineering Practice," paper 2c42,
Proc. 25th Frontiers in Ed. Conf, Atlanta, November (1995)
10. Felder, R., L. Bernhold, E. Burniston, J. Gastineau, and J.
O'Neal, "An Integrated First-Year Engineering Curriculum
at North Carolina State University," paper 4d42, Proc. 25th
Frontiers in Ed. Conf., Atlanta, November (1995)
11. Froyd, J.E., "Integrated, First-Year Curriculum in Science,
Engineering, and Mathematics: A Ten-Year Process," paper
4d44, Proc. 25th Frontiers in Ed. Conf., Atlanta, November
12. Miller, R.L., and B.M. Olds, "Connections: Integrated First
Year Engineering Education at the Colorado School of
Mines," paper 4a12, Proc. 25th Frontiers in Ed. Conf., At-
lanta, November (1995)
13. Panitz, B., "The Integrated Curriculum," ASEE Prism, 7(1),
24 (1997)
14. Richards, L.G., and S. Carlson-Skalak, "Faculty Reactions
to Teaching Engineering Design to First Year Students," J.
Eng. Ed., 86, 233 (1997)
15. Yokomoto, C.F., et al., "Developing a Motivational Fresh-
man Course Using the Principle of Attached Learning," J.
Eng. Ed., 88, 99 (1999)
16. McConica, C.M., "Freshman Design Course for Chemical
Engineers," paper 174c, 1998 Annual AIChE Meeting, Mi-
ami, FL, November (1998)
17. Solen, K.A., and J.N. Harb, "An Introductory ChE Course
for First-Year Students," Chem. Eng. Ed., 32(1), 52 (1998)
18. Mitchell, B.S., "Early Introduction of Design Fundamentals
into the Chemical Engineering Curriculum," Proc. 1997Ann.
ASEE Conf., Milwaukee, WI, June (1997)
19. Woods, D.R., "Introducing Design/Synthesis in Thermody-
namics and Transport Courses," paper 236e, Annual AIChE
Meeting, Miami Beach, FL, November (1995)
20. Gatehouse, R.J., J.R. McWhirter, and G.I. Selembo, "The
Vertical Integration of Design in Chemical Engineering,"
paper 174e, 1998 Annual AIChE Meeting, Miami, Fl, No-
vember (1998)
21. Bauer, L.G., D.J. Dixon, and J.A. Puszynski, "Introduction
of Design and AspenPlus across Chemical Engineering Cur-
riculum," paper 174d, 1998 Annual AIChE Meeting, Miami,
FL, November (1998)
22. Johnson, D.W., R.T. Johnson, and K.A. Smith, "Maximizing
Instruction Through Cooperative Learning," ASEE Prism,
7(6), 20 (1998)
23. Johnson, D.W., R.T. Johnson, and K.A. Smith, Active Learn-
ing: Cooperation in the College Classroom, Interaction Book
Company, Edina, MN (1991)
24. ASEE Prism, "Let Problems Drive the Learning in Your
Classroom," ASEE Prism, 6(2), 30 (1996)
25. Woods, D.R., Problem-Based Learning: How to Gain the
Most in PBL, Waterdown, Ontario, Canada (1994)
26. Criteria for Accrediting Engineering Programs, Accredita-
tion Board for Engineering and Technology, Inc., Baltimore,
MD (1999) O

Summer 2000

IT curriculum



Beyond Problem Sets and Lab Reports

Georgia Institute of Technology Atlanta, GA 30332-0100

or the past three years, Georgia Tech's School of
Chemical Engineering has offered a novel course for
undergraduate students. It addresses topics in oral
and written communication in the context of a bioengineer-
ing case study that simulates the variety of communications
encountered in the modern workplace. "Effective Communi-
cation for Professional Engineering" features weekly guest
speakers from a variety of professional disciplines, ranging
from a practicing chemical engineer to an FDA regulator to a
patent lawyer. This course is unlike traditional English
courses, which are designed to generally broaden literary
and composition horizons, and unlike traditional technical
communication courses, which usually focus on lab reports
and memos outside any "real-world" context.
The innovation of this particular course, which is subtitled
"Beyond Problem Sets and Lab Reports," lies in its place-
ment of student assignments in context with realistic profes-
sional settings. By bringing in a wealth of outside speakers
and information, instructors encourage students to think in
more creative ways to solve communication problems be-
yond the creation of a technically correct report. Students
find themselves writing to a variety of audiences in more
thoughtful ways, whether they are allaying the fears of a
hypothetical public or persuading a corporate boardroom to
adopt new technologies. They are required to draft a broad
range of written and oral communication, including press
releases, abstracts, patent disclosures, and speeches to a
Board of Directors. Each new audience builds on a core of
technical information common to the previous writing as-
signments, while lectures on audience analysis focus stu-
dents on the thought processes involved in tailoring these
various communications to different levels of technical un-
derstanding and informational need.
The pedagogical approach of this course also differs greatly
from the relative anonymity of the larger lab or design class

where technical communication is often addressed in chemi-
cal engineering curricula."'2] The students who sign up for
this elective course continually experience a high level of
peer and student-to-faculty interaction. Peer critiques of both
written and oral presentations allow students to comment on
each other's strengths and weaknesses. During the quarter,
students are placed in shifting teams of two or three for
activities that reinforce both the talks given by outside speak-
ers and the instructors' lectures; these classroom interactions
are then incorporated into the writing or speaking assign-
ment due in the following class.

Universities generally do well at teaching science and
engineering students the fundamentals of their field. Our
industrial colleagues tell us, however, that academia needs
to do better at teaching students how to talk and write
about technical topics to both fellow engineers and non-

Mark Prausnitz is Assistant Professor of Chemi-
cal Engineering at Georgia Tech. He was edu-
cated at Stanford University (BS, '88) and M. I. T
(PhD, '94). He currently teaches mass and en-
ergy balances to chemical engineering sopho-
mores. His research addresses novel mecha-
nisms for improved drug delivery using electric
fields, ultrasound, and microfabricated devices.

Melissa Bradley is the Writing Program Spe-
cialist and Publications Coordinator of the School
of Chemical Engineering at Georgia Tech. She
has a BA in English from the University of Geor-
gia and is currently pursuing her Master's de-
gree in Information Design and Technology. She
has freelanced as a technical and feature writer
for a number of companies and magazines.
Copyright ChE Division of ASEE 2000

Chemical Engineering Education

engineers.3'4 This course addresses that need in a real
and practical way.
Although self-contained, this course also conforms to the
broader curricular goals of the department, which include a
writing and speaking program within the required unit-op-
erations lab course. The writing aspect of the lab course,
however, usually focuses on the technical and written skills
necessary to produce just one form of communication-the
technical report. And although these technical reporting skills
are critical to an engineer's education, by no means will his
or her future career as a professional be limited to communi-
cating in only one format or to only one audience type.-[571
Clearly, a range of communication issues, audience types,
and writing and speaking formats should be presented in the
undergraduate curriculum to prepare students for their roles as
active participants at all levels of the engineering community.
Other communication courses offered outside of engineer-
ing are available, but generally have a weak connection with
science as practiced in industry.'8] Moreover, we feel that the
textbook-driven nature of most
technical writing courses, with
an emphasis on writing memos T
and lab reports, does not fully C
describe the diversity of form
and content prevalent in inter- Week I To set the stage for ou
and intra-industrial communi- transdermal drug deliv
cation. In contrast, our course administer nicotine ac
provides a broad scope of com- Week 2 To learn first-hand abi
munication issues and audi- bioengineer from ind
transdermal nicotine p
ences, is based on a case study transdermal nicotine
in context, and is linked to the Week 3 Congratulations! You
delivery system. We n
real experiences of working an invention disclosure
professionals in the field. Week 4 Patent protection allo
work. We have a disc
COURSE GOALS journal, prepare an at
To give students widely ap- Week 5 Your presentation was
plicable tools for written and of attention. We talk
popular press, write
oral communication, we have presentation to conce
emphasized audience analysis W 6
[9,1 Week 6 Now that you have fig
and critical thinking9'"' rather according to federal g
than conforming students to a regulator and prepare
series of prescribed formats. Week 7 Manufacturing your n
The goals of the course are to that the plant workers
manager and prepare
1. Provide students with the
Week 8 To bolster public supp
opportunity to write and saving invention, we t
speak to a diversity of an advertising campai
audiences on at least a Week 9 You develop a second
weekly basis. effective, but costs me
to invest in this techno
2. Bring in outside profes- and prepare a talk for
sionals who work either
signals who work either Week 10 Your success has led t
in or with the engineering career counselor and
industry to discuss what resume, and have a mi
they do and how they

3. Simulate the world of professional engineering by
relating oral and written assignments to a common
case study.
4. Focus on audience analysis as the basic building block
of communication.
5. Integrate group projects with individual projects to
acclimate students to working with others as a team.[]1

The integrated approach of this course, which seeks to
simulate the "life of a professional engineer," is highlighted
by extensive in-class discussion with outside professionals,
as summarized in the ten-week course outline shown in
Table 1.
The course starts by introducing students to technical as-
pects of our bioengineering case study, based on the Nicoderm
transdermal nicotine patch developed by the Alza Corpora-
tion. "1231 During the first two
weeks, while digesting this
technical information, stu-
,E 1 dents write a series of short
'erview reports on a current techni-
cal topic of broad interest,
study, you are hired to develop a e
stem to safely and effectively using information easily
ie skin. found on the Internet (e.g.,
r case study, we will meet a we used Viagra this year and
who actually developed the first the ValuJet crash last year).
These reports are not for-
nvented a new transdermal mally graded, but are cri-
ith a patent lawyer and prepare tiqued to help students start
thinking about the effective
to talk publicly about your communication of technical
with the editor of a scientific
Sand present a scientific talk. information to a variety of

Once the bioengineering
case study begins in earnest,
students are required to pre-
pare a written and/or oral as-
signment every week (see
syllabus excerpts in Table 2
and a sample assignment in
Figure 1). Each written as-
signment is turned in to the
instructors as well as to an
anonymous classmate for
peer critique." 4 The oral as-
signments are followed by
immediate feedback from
the instructor and from stu-
dents (additional student
self-assessment using a
video tape of the presenta-

Summer 2000

se Ov

r case
very sy
ross th
out ou
have i
neet w
vs you

a great success and has attracted a lot
vith a science journalist from the
a press release, and give a short
ned citizens.
:ured out how to deliver the drug
guidelines, we speak with a government
Sa brief for the FDA.
ew device involves difficult procedures
dislike. We meet with a business
a memo for disgruntled employees.
>ort for the company based on your life-
alk with a graphic designer and create
-generation technology that is more
ore to make. To convince your superiors
ology, we speak with a corporate CEO
the Board of Directors.
o new job possibilities. We meet with a
a professional interviewer, prepare a
ock job interview.

tion would also be helpful, but is not
something we have yet implemented).
By getting feedback from their peers
as well as their instructors, students
can simulate the roundtable discus-
sions of teams in the workplace and
implement suggestions for the next

While lectures by the instructors
provide general lessons on commu-
nication relevant to the topic at hand,
each guest speaker gives a detailed
look at the requirements of his or her
job and the communication issues
arising from that job. Follow-up as-
signments permit students to put les-
sons from both lectures into practice.


A number of features of this course
distinguish it from other engineering
and writing courses and have been
critical to the course's success. Many
of these features are not by them-
selves new, but their combination pro-
vides a novel approach to integrating
concepts often missed in a conven-
tional engineering curriculum.

* Academic, Industry, and Community Involvement Lectures
are given not only by an engineer (Prausnitz) and a writing
specialist (Bradley), but also by industry professionals who
visit the class on a weekly basis. Moreover, we have
involved newspaper journalism students from a local high
school to attend some lectures and critique our students'
press releases and oral presentations.
* Case-Study Format By following a single case study
through the whole course, students have a sense of continuity
and can focus on communication issues without having to
learn new technical information each week. This approach
also simulates the long-term development of projects found
in industry.
* "Real-World" Context This course is as much about
introducing students to the broad scope of life as a profes-
sional engineer as it is about communication. This helps
students understand why good communication needs to be an
integral part of their professional careers.
* Frequent, Short Writing and Speaking Assignments To
build student confidence in communicating effectively,
written or oral assignments are due in almost every class.
Most assignments are short: 1000 words written or 4 minutes
* Emphasis on Audience Analysis Assignments and
classroom discussion emphasize selection of content and
format tailored to the intended audience to achieve the

Bradley, Canatella and Prausnitz

778 Atlantic Drive, Suite 307
Atlanta, GA 30332
telephone (404) 894-8471
facsimile (404) 894-2866

April 15, 1999

Dr. Ronald W. Rousseau
Drug Delivery Research and Development
Acme Corporation
7000 Industrial Blvd.
Atlanta, GA 30322
Dear Dr. Rousseau,
I understand that Acme Corporation has developed a proprietary technology for
transdennal drug delivery using a rate controlling membrane. I have been asked to
prepare and file a patent application for this invention. To initiate the process, please
prepare a confidential invention disclosure, file it with Acme's intellectual property office
and send a copy to me. The disclosure should include a brief description of the invention,
date of conception, description and date of its reduction to practice, and discussion of its
potential applications. The disclosure should be signed and dated by yourself. It should
also be signed and dated by two other Acme employees to whom you disclose and
explain the invention. Do not discuss any information relating to the invention with
anyone outside Acme Corporation through any form of written or oral communication
until a patent has been issued.
I look forward to receiving your invention disclosure at your earliest convenience.

1 o( \ a


Melissa Bradley, Esq.
Senior Partner

Figure 1. Sample assignment for invention disclosure.

Chemical Engineering Education

Syllabus Excerpts

Week 3: Communicating with Lawyers
Guest Speakers Assignment
Patrea Pabst, Arnall, Gregory, and Golden Invention disclosure (see Figure 1)
Stephen Dorvee, Arnall, Gregory, and Golden
Reading Materials
* "Patents: What, Why, and How," M.H. Heines, Chem. Eng. Prog., pp. 79-85, July (1992)
* Do's and Don'tsfor Keeping Lab Notebooks, Fish & Richardson, P.C., Boston, MA
* "Copyright and Permissions," B.F. Polansky, in The ACS Style Guide, J.S. Dodd, ed., American
Chemical Society, Washington, DC, pp. 137-143 (1986)
"Subsaturated Nicotine Transdermal Therapeutic System," J.L. Osborne, et al., U.S. Patent No.
Material Evaluation Agreement, Georgia Tech Research Corporation, Atlanta, GA
Proprietary Information Agreement, Georgia Tech Research Corporation, Atlanta, GA
Product Development and Commercialization Agreement, ALZA Corporation, Palo Alto, CA

Week 5: Communicating with the Public
Guest Speakers (Only one speaker per course offering) Assignments
Ann Kellan, CNN Press release
James Pilcher, Associated Press Presentation at community meeting
David Jarmul, Howard Hughes Medical Institute
Reading Materials
* Communicating Science News: A Guide for Public Information Officers, Scientists, and Physicians,
The National Association of Science Writers, Greenlawn, NY
Headline News, Science Views II, David Jarmul, ed., National Research Council, Washington, DC
"Marion Merrell Dow Inc. Introduces First Nicotine Patch for Smoking Cessation," press release
from Marion Merrell Dow, Kansas City, MO, Nov. 8 (1991)
"Heart Attacks Reported in Patch Users Still Smoking," S.L. Hwang and M. Waldholz, The Wall
Street Journal, p. B1, June 19 (1992)

desired effect rather than reiterating grammatical and generic
stylistic rules already covered in freshman English courses.
SInstructor and Peer Critique of All Assignments All
students receive written and oral feedback from both their
instructors and their classroom peers on every written and
oral assignment. Peer critiques are educational to the
recipient as well as to the person offering the judgment." 14

The course's impact has been assessed by students in the
course, guest speakers who vis-
ited the course, and the instruc-
tors. Results from Georgia Quality of outside spe
Quality of outside spea
Tech's standard anonymous Quality of reading mat
student evaluation form, supple- Helpfulness of peer criti
mented by a course-specific Improvement in writing s
written evaluation, showed that Improvement in speaking s
Improvement in understand
students were very supportive of communication is
of the course and strongly rec- Overall lear
Overall enjoy
ommended it to others. Figure Recommendation to offer class
2 shows responses to questions
about overall effectiveness of Figure 2. Student assess
the course. These responses 14 students was sur
suggest that the speakers, read- different aspects o,
ing materials, and emphasis on of 1 (terrible)
written assignments were all well received. As indicated in
the figure and in other comments, students felt a stronger
emphasis on oral communication would be helpful, and fol-
lowing this suggestion, we will replace some of the written
assignments with oral assignments in future course offer-
ings. Also, the peer critiques were not perceived to be as
useful as instructor feedback. To address this, we now require
that peer critiques be at least a half-page long and that they
identify specific problems and suggest concrete solutions.
Guest speakers have been uniformly supportive of the
course and frequently commented that they wish they
could have taken a similar course when they were stu-
dents. All of them liked the guest-lecture format and
found the approach relevant to their careers and their
interactions with engineers.
As instructors, our assessment of the course is that it met
the objectives we set out to achieve and has been beneficial
to students, guests, and instructors alike. We believe that the
guest speakers, who are coached in advance, have been
critical to the course's success because they broaden the
scope of the course and ensure its relevance to "real-world"
issues. To maintain the sense of continuity necessary to the
case-study course format, it was also important for us to
continually clarify the connections between speakers, read-
ing materials, and overall course objectives by providing
follow-up presentations.
Despite the strongly positive reviews from students and
guest speakers, it is difficult to get many students to sign up
for the course in view of Georgia Tech curriculum require-

Summer 2000

ments. Because this course cannot count toward any gradua-
tion requirement other than free elective credits (which most
students do not need), the only students who have taken the
course are those with enough interest and who are far enough
ahead of the curriculum to fit it into their schedules. Thus,
each course offering has attracted a small group (8 to 14) of
strong, motivated students out of a chemical engineering
graduating class of 120 to 150. In an attempt to expose a
larger fraction of students to communication issues, we plan

to offer the course on a

1 2 3 4 5

nent of the course. A class of
veyed and asked to rate
fthe course on a scale
Sto 5 (excellent).

1-unit pass/fail basis (guest lectures
only) in addition to the regular
3-unit graded option. We are also
permitting graduate students to
take the class and advertising the
course more intensively outside
of chemical engineering.


This work was supported in
part by a CAREER Young In-
vestigator Award from the Na-
tional Science Foundation (BES-

1. Newell, J.A., D.K. Ludlow, and S.P.K. Sternberg, "Develop-
ment of Oral and Written Communication Skills Across an
Integrated Laboratory Sequence," Chem. Eng. Ed., 31, 116
2. Hanzevack, E.L., and R.A. McKean, "Teaching Effective
Oral Presentation as Part of the Senior Design Course,"
Chem. Eng. Ed., 24, 28 (1990)
3. Kranzber, M., "Educating the Whole Engineer,"ASEE Prism,
p. 28 (Nov. 1993)
4. McConica, C., "A Course in Communication Skills for the
Corporate Environment of the 1990s," Chem. Eng. Ed., 29,
158 (1995)
5. Odell, L., and D. Goswami, eds., Writing in Nonacademic
Settings, Guilford Press, New York, NY (1985)
6. Evered, D., and M. O'Connor, eds., Communicating Science
to the Public, Wiley, New York, NY (1987)
7. Ludlow, D.K., and K.H. Schulz, "Writing Across the Chemi-
cal Engineering Curriculum at the University of North Da-
kota," J. Engineering Ed., 83, 161 (1994)
8. Ovitt, G., "Technical Writing and the Two Cultures," J.
Tech. Writ. Comm., 12, 89 (1982)
9. Halpern, D.F. Critical Thinking Across the Curriculum: A
Brief Edition of Thought and Knowledge, Erlbaum Assoc.,
Mahwah, NJ (1997)
10. Moriarty, M.F., Writing Science Through Critical Thinking,
Jones & Bartlett, Sudbury, MA (1997)
11. Schulz, K.H., and D.K. Ludlow, "Group Writing Assign-
ments in Engineering Education," J. Eng. Ed., 85, 227 (1996)
12. Gora, M.L., "Nicotine Transdermal Systems," Ann. Phar-
macotherapy, 27, 742 (1993)
13. Marion Merrell Dow, "Nicoderm," in Physicians'Desk Refer-
ence, S.B. Greenberg, ed., Medical Economics Data, Montvale,
NJ, p. 1306 (1994)
14. Newell, J.A., "The Use of Peer Review in the Undergradu-
ate Laboratory," Chem. Eng. Ed., 32, 194 (1998) 0


Random Thoughts...


North Carolina State University Raleigh, NC 27695

Several years ago I taught five chemical engineering
courses in successive semesters to a cohort of stu-
dents, beginning with the introductory course on ma-
terial and energy balances (the stoichiometry course, CHE
205). I consistently used a variety of nontraditional instruc-
tional methods, most notably cooperative learning (assign-
ments carried out by teams of three or four students with
various measures being taken to assure individual account-
ability), and compared various learning outcomes for the
students in these classes with the same outcomes for a group
of traditionally-taught students. A description of the instruc-
tional methods and a summary of the results may be found in
the Journal of Engineering Education.*
In the fall of 1999, I sent a questionnaire to the 72 stu-
dents in the study who graduated in chemical engineering,
inviting them to reflect on their undergraduate education-
what they liked and disliked, what helped prepare them for
their current careers, and what advice they would have for
today's beginning chemical engineering students. I eventu-
ally heard back from 50 of them, a respectable 69% return.
Of the respondents-most of whom graduated in 1994 or
1995-33 (66%) were still involved in engineering and the
remaining 17 were in different fields. Eleven (22%) had
earned advanced professional degrees-four PhDs in chemi-
cal engineering, four medical degrees, and three law de-
grees. Those still in engineering included four process engi-
neers, four environmental engineers, three each in engineer-
ing management, product development, production engineer-

*(1) R.M. Felder, "A Longitudinal Study of Engineering Student
Performance and Retention. IV. Instructional Methods and
Student Responses to Them," J. Engr. Education, 84(4),
361-367 (1995);
(2) R.M. Felder, G.N. Felder, and E.J. Dietz, "A Longitudinal
Study of Engineering Student Performance and Retention.
V. Comparisons with Traditionally-Taught Students," J.
Engr. Education, 87(4), 469-480 (1998)

ing, research and development, and quality assurance, nine
in other engineering jobs, and one in graduate school. Those
who left engineering included four computer systems man-
agers or programmers, four physicians, three attorneys, two
full-time homemakers, one executive recruiter, one human-
resources manager, one machine operator, and one doctoral
candidate in science and technology.
The respondents were asked to list the features of their
undergraduate education that had proved to be most valu-
able in their career development. Items mentioned and the
number of respondents citing them included

El The problem-solving and time-management skills they
acquired by working on so many long and difficult
assignments (25)
E A variety of benefits gained from working in teams on
homework (23)
E What they learned in the stoichiometry course (8)
El The broad knowledge base they acquired in the
curriculum (6)
El Troubleshooting skills (3)
El Knowledge of statistics (3)
No other item was mentioned by more than two individuals.

Copyright ChE Division of ASEE 2000

Chemical Engineering Education

Richard M. Felder is Hoechst Celanese Pro-
fessor 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 Insti-


Specific courses besides stoichiometry that were cited more
than once included thermodynamics (2), mass transfer and
separation processes (2), freshman chemistry (2), and math-
ematics (2).
In their open comments, almost every respondent spoke
positively about group work, mentioning its learning ben-
efits and/or the interactions with classmates that it fostered.
For example, "I formed very close relationships with my
group members that remain today. I realized that I wasn't
alone in struggling with new concepts and could garner
support and help from teammates." and "Being forced to
meet other students through required groupwork. .kept me
in the course long enough to develop the skills and self-
confidence necessary to continue on in the CHE curricu-
lum." No one said anything negative about group work,
although two respondents indicated that they disliked it ini-
tially and only later came to see its benefits.
Other features of the curriculum that got favorable cita-
tions from several students included
E[ The laboratory courses ("I always enjoyed the labs
because you put to use all those hours of class
time. ")
E In-class exercises ("The structure of the classes
helped me to learn more by having active involve-
ment in the class instead of the typical 'I lecture,
you take notes and shut up' approach. ")
EL Connections with chemical engineering practice
("Not only did Prof ... try to provide real life
examples, but we also had visitors from industry
come in and explain how they used their college
backgrounds in their fields. This information helped
me to decide which industry was most appealing
and best suited to my interests. ")

Common recommendations for beginning chemical engi-
neering students were
E[ Pay attention to the stoichiometry course (10)
("CHE 205 is the most important course you can
take-the first step in any engineering calculation is
a material/energy balance. ")
LI Study and work hard (9) ("Prepare yourself for a
new way of thinking, 'cause this ain't high school,
and you're not going to be able to coast. Work
hard early and you won't have to play catch-up.")
E[ Stick with it (8) ("Don't get discouraged if you
don't do so well at first. People do get better as the

curriculum progresses. (I did.)")
E[ Take teamwork seriously (7) ("Get to know as
many people in the class as soon as you can-this
will get you through the homework and the tests.
Teamwork is a way of life out in the real world. It
willfrequently be a major factor in how you are
'tested' at work. ")

Two students suggested that students struggling to make it
through most of their chemical engineering courses might
reconsider their choice of a career path. One put it this way:
"Any time you feel stubbornness getting you through some
trial, you should consider why you need it. Ifully believe that
anybody will make passing marks in any subject area that
truly interests him or her. If on the other hand, you find the
problems and concepts difficult, do not take this as a sign of
intellectual failing, but rather as a sign of disinterest."
Several points about the survey responses submitted by
these alumni are particularly noteworthy. I was struck by the
fact that only four respondents were involved in process
engineering and three in engineering research and develop-
ment, which is to say that fewer than one out of six were
working in the areas addressed by essentially all of the core
chemical engineering courses beyond the stoichiometry
course. Many cited the value of the stoichiometry course in
their academic and/or professional careers, stressing its im-
portance in the advice they gave to beginning chemical
engineering students, while no other core chemical engi-
neering course was cited by more than two respondents. In
contrast, almost every respondent noted the benefits of the
problem-solving and teamwork skills they had acquired in
the curriculum and many mentioned the value of their expo-
sures to engineering practice. (The term "real world" came
up fairly often.)
These observations suggest to me that the specific content
of our core courses beyond stoichiometry may be less im-
portant than we tend to believe-much less important than
the industrial relevance of what we teach and the extent to
which we help our students develop problem-solving, com-
munication, and teamwork skills. When we review and re-
vise our curricula, we might do well to concentrate on ad-
dressing modern engineering practice beyond process de-
sign and analysis and on explicitly facilitating critical skill
development, and worry less about how many advanced unit
operations and differential equation solution techniques we
can shoehorn into the courses. Besides helping our stu-
dents, this change in focus won't do us a bit of harm at the
next ABET visit. O

Summer 2000

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

r laboratory




Rowan University Glassboro, NJ 08028-1701

his experiment explores the area of heterogeneous
catalysis using the automotive catalytic converter,
which is the largest market for heterogeneous cata-
lytic reactors. Since catalysts have been placed in approxi-
mately 225 million automobiles, nearly everyone with a
car owns a catalytic converter. Students' interest in this
experiment is piqued when they realize that their cars use
this device every day. This immediate familiarity with
the automobile allows students to approach the experi-
ment with a confidence that helps them master the
experiment's objectives.
A smaller, but growing, market for oxidation catalysts is
in the destruction of volatile organic compounds from manu-
facturing sources. These catalytic reactors are designed us-
ing similar principles to the automotive catalysts. Base met-
als and platinum-group metals catalyze the CO oxidation
and unburned hydrocarbons as well as reduce NO,. Large
installations have been in place on stationary internal com-
bustion engines and gas turbines. For example, Johnson-
Matthey has developed other products, such as CONCAT, for
halogenated hydrocarbon destruction and Honeycat for standby
generators and diesel engines working in confined spaces.
The automotive catalytic converter was originally intro-
duced to reduce the photochemical smog problems in large
cities such as Los Angeles and Tokyo. The automobile was
identified as the major producer of smog precursors and a
catalytic converter was required to rectify the problem. The
standard catalytic converter consists of a honeycomb monolith
support with a washcoat of metals on the surface of the sup-
port. A typical monolith is either ceramic or metal and consists
of approximately 1-mm square channels 6 inches in length.
The current state-of-the-art catalyst is a three-way catalyst
in which unburned hydrocarbons and CO are oxidized to
CO, and H20, and NO is reduced to N2. A brief review of

This experiment is an
excellent combination of
hands-on experiments, advanced
analytical instrumentation, reactor
modeling, and successful application of
a numerical technique.

these reactions and reactors has been presented by Schmidt.']
Typical metals used in catalysis are platinum (Pt) and/or
palladium (Pd) to oxidize CO and hydrocarbons, and rhodium
(Rh) to reduce NO,. Jacoby[21 reports that these catalysts are
continually being engineered to reduce emissions from cars
with cold engines and that they meet California's air stan-
dards of low and ultra-low emission vehicles. In all new
catalysts, the reaction rate must be determined.
The temperatures that are used in the following experi-
ment are above the 468'C autoignition temperature of pro-
pane. This presents an excellent vehicle for introducing safety
concepts such as flammability limits and autoignition tem-
peratures. At room temperature and atmospheric pressure,
the flammability limits for propane are between 2.3 and 9.5
vol% in air. The concentration of propane we are using in
this experiment is 2625 ppm, or 0.3% propane, and at these
low concentrations it is difficult to bur a hydrocarbon.

Robert Hesketh is Associate Professor of Chemical Engineering at Rowan
University. He received his BS in 1982 from the University of Illinois and
his PhD from the University of Delaware in 1987. After his PhD he
conducted research at the University of Cambridge, England. His teach-
ing and research interests are in reaction engineering, freshman engi-
neering, and separations.
Dan Bosak and Luke Kline are junior chemical engineering students at
Rowan University. They have worked over the past two years as research
assistants for the department and a National Science Foundation Under-
graduate Faculty Enhancement Workshop. In addition to full-time studies,
both are chemical operators on the weekends for the precious metal
division of Johnson Matthey.
Copyright ChE Division of ASEE 2000

Chemical Engineering Education

Furthermore, the autoignition temperature is defined as the
temperature at which spontaneous reaction occurs at stoichio-
metric propane and air concentrations (4% propane in air).

Construction of this experiment is relatively simple; a
small monolith is sealed inside a stainless steel tube, as
shown in Figure 1. This yields a method for measuring gas
flowrates, mixing the gases, and measuring the inlet and
outlet concentrations.
The experiment can be performed
with any catalyst that can oxidize a
hydrocarbon, such as Pt or Pd. In
this experiment, an actual automo-
tive catalyst, donated by Johnson
Matthey, was used. This catalyst can
be fabricated using an aqueous Pt T
solution and buying the corderite
monolith from Coring.[31 Alterna-
tively, a used catalytic converter can Furnace
be purchased from your local
junkyard. If the junkyard catalyst
has lost its activity, it can be regen-
erated using an acetic acid leaching
process following the method of
Angelidis and Papadakis.141 A typi-
cal automotive catalyst has an el-
lipsoid cross section with axes of 5
5/8 and 3 inches. There are 400
square channels per in2 having a
length of 6 inches. To keep the re-
actor size small, you will need to Pressure
cut a cylindrical section out of this Propane Regulator
monolith with a hacksaw and sand (C3H8)
the section into a cylinder. A typi-
cal geometry for the cylindrical sec- Figure 1. Experim
tion has a diameter of 25 mm and a
length of 30 mm.
To prevent gas from bypassing the catalyst, it is wrapped
in a 3-mm-thick ceramic fiber blanket or felt. The monolith/
felt combination should be carefully inserted into a 1 1/4"
Schedule 80 (1.278" ID) 316 stainless steel pipe and gently
pushed to 8 inches from the top of the 24" tube. The catalyst
is held in place using the friction fit of the ceramic fiber
gasket. A flow distributor and gas preheat zone is con-
structed from blank monoliths or a 6-in bed of washed sand
is situated at the bottom of the reactor to both preheat and
distribute the gas. The reactor temperature is maintained
using a model F79345 Thermolyne split-tube furnace. The
furnace has a 12-in heated zone and is rated for 2880 Watts
and a maximum temperature of 12000C. If your budget
allows, you can purchase a 3-zone furnace to control the heat
loss at the ends of the reactor tube.

The stainless steel tube can be sealed using machined end
caps or the more expensive welded flange set. Thermo-
couples are placed through the top and bottom of the reactor.
If threaded end caps are used, the thermocouple fitting must
be placed exactly in the center of the tube so that the thermo-
couple can be inserted approximately 2 mm inside the mono-
lith catalyst channel. The thermocouple end should be touch-
ing the channel walls to give the catalyst temperature. The
second thermocouple inserted through the opposite end can
either be placed in the upstream end of the monolith channel
or be shielded and used to mea-
sure the entering gas tempera-
ture. The external diameter of
these thermocouples is depen-
dent on the width of an indi-
-bypass to exhaust-- vidual channel and is typically
-- less than 1.16 in.

mental catalytic reactor.

If a new catalyst is used it
should be aged to give a nearly
steady-state performance. The
catalyst used in this experiment
had palladium active sites and was
aged at 9000C for 12 hours in
10% H20 and balance nitrogen.
The reaction rates of catalysts are
typically reported on a mass, sur-
face area, or active site basis. For
commercial catalysts, either the
mass or the external surface is
known. For example, the catalyst
used in this experiment has an av-
erage weight (monolith + metal
washcoat) of 8.434 g.
Either propane or methane
should be used as the hydrocar-
bon in this study. Propane, for

this experiment, was obtained
from MG Industries, has a pu-
rity of 99% propane, and is rated chemical pure (CP grade).
Care must be taken to avoid any catalyst poisons and the
standard barbecue propane tank would not be suitable since
it contains sulfur compounds to warn the user of gas leaks.
Air was obtained through the house compressor and regu-
lated from 120 psig to 14 psig.
Since this experiment uses low concentrations of propane
in air, the flowrates must be precisely controlled. The stan-
dard method is to use 2-stage regulators and rotameters. The
air pressure was controlled using a ControlAir Inc High
Precision 100-BA regulator. Propane can be controlled
using a MG industries 2-stage regulator. A more user-
friendly control scheme would be to use mass flow con-
trollers at a cost of approximately $3,000 for two control-
lers and a control station.

Summer 2000

For this experiment, gas concentrations can be analyzed
using a number of analytical instruments, including gas chro-
matography, online flame ionization detectors, NDIR ana-
lyzers, or FTIR spectrometers. At Rowan, we are using a
Nicolet Magna-IR E.S.P. spectrometer. This spectrometer
uses a 2-m gas cell path length with a KBr substrate
beamsplitter. In this analysis the spectrometer can detect
compounds that have a net dipole moment, such as CO2 or
NO, but cannot detect compounds such as 02 and N2, as
shown in Figure 2. In the basic experiment presented in this
paper, only the detection and quantification of propane is
required. The complete Fourier Transform of a signal from a

mixture of 1000 ppm of propane passed
through a catalyst sample at a furnace
temperature of 5000C is given in Figure
3. Propane is detected primarily from the
C-H stretch in the range of 3000 to 2850
cm1. Figure 3 can be compared with stan-
dard spectra shown in Figure 2 to deter-
mine if other product gases are present.
The cost of the experimental apparatus,
excluding analytical instrumentation, is
approximately $5,000. If new analytical
instrumentation is purchased, then the
costs of an NDIR gas analyzer, gas chro-
matograph, or a Nicolet FTIR would cost
approximately $7,000, $20,000, or

There are many experiments that can
be performed with this reactor configura-
tion, including
Examination of light-off curves
Determination of reaction rates using
integral-reactor method
Determination of reaction rates using
differential-reactor method

In this paper the results of the light-off
experiment will be presented since it is
unique to the automotive catalyst indus-
try. The experiment can be integrated eas-
ily in the first month of a reaction engi-
neering course in which the basic mole
balances for a plug flow reactor have been
introduced to the student. It uses the con-
cepts of conversion and the trade-off be-
tween the reaction rate and residence time
in the reactor. In addition, this experi-
ment can be conducted in approximately
1.5 hours and does not require high-preci-
sion quantification.

In cars with catalytic converters, the majority of pollutants
are emitted from the car during the start-up period when the
catalyst is cold and the required reactions are too slow. This
is known in industry as the cold-start problem. Various
strategies are being employed to eliminate this problem,
such as electrically preheating the catalyst and adsorbing
and storing the pollutants on a separate bed. The temperature
required to activate the car catalyst is commonly referred to
as the light-off temperature. At this temperature, the net heat
released from the reactions is sufficient to maintain the cata-
lyst at temperatures required to obtain high conversions. It is
commonly thought that at temperatures below the light-off

Figure 2. Standard spectra of possible absorbing compounds from
reactor outlet.


,.i ... i i .
3500 3000 2500 2000 1500 1000
Wavenumbes (cml)

Figure 3. FTIR spectrum from the reactor outlet stream using an inlet
concentration of 1000 ppm propane in air at a
furnace temperature of 500 C.

Chemical Engineering Education

6 j ter
Abs 4

C02 991ppm


1 0 -CO 1001ppm i
:Ethene 300 ppm
Abs 1

.CH4 1010ppm
Abs 2-

1500 1000

3500 3000 2500 2000
Wavenumbers (cm-1)

Al -1


temperature the catalyst is not removing pollutants, and at
temperatures above the light-off temperature the catalyst is
working. There are many definitions of the light-off tempera-
ture, but the most common is the temperature for which the
conversion of a reactant reaches 50%.
Hayes and Kolaczkowski131 show that this light-off tem-
perature is a function of reactor size and reactor flowrate.
This result should be obvious if you have just finished a
reactor design course! Assuming a plug-flow model, the
mole balance on propane is
dFc3H8 r (1)
where Fc38, is the molar flowrate of propane, W is the
weight of catalyst, and r is the reaction rates with units mole
Assuming first-order kinetics and a constant flowrate, Q,
through the reactor, Eq. (1) can be integrated to give

XC3H8- P 1- exp (2)

Assuming an Arrhenius reaction rate constant and adjusting
for the difference in rotameter flowrate and actual flowrate
through the reactor gives

A exp(-E / RT)1
XC3H8 = -exp-W A e(- T ) (3)
Q 0

As the reactor temperature increases, both the reaction rate
and conversion increase. At a given temperature, the conver-
sion at high flowrates is less than the conversion at low
flowrates. This clearly shows that the light-off temperature
is dependent on the volume or geometry of the reactor and
the flowrate of gases through the reactor.
The procedure for this experiment is relatively simple.
The two flowrates chosen are of 14.5 and 7.76 L/min. The

E- -flow

400 450 500 550
Temperature (C)

600 650 700

Figure 4. Results of light-off determination experiment.

Using this experiment, students
are able to see the catalyst, to measure
gas-phase concentrations and flowrates, and to
model the reactor and find the reaction-rate
parameters using a nonlinear
regression of the data.

inlet concentration of propane in air is held constant at 2625
ppm. The furnace is initially set to a temperature of 550C
and the gas is analyzed at each flowrate. Next, the furnace
temperature is increased by 250C and samples are again
taken at the above two flowrates. If the furnace tempera-
ture controller is programmable and an online analyzer is
available, then the experiment can be automated follow-
ing industry practice.
An example of the experimental results is shown in Figure
4. From these results the light-off temperatures at 50% con-
version of propane are 620 and 5380C at 14.5 and 7.76 L/
min, respectively. These distinctly different curves show
that the outlet conversion of propane is a function of the
flowrate through the monolith. This confirms that the light-
off temperatures quoted in the literature can only be used to
compare similar catalysts of equivalent geometry and gas
flowrate. Students conducting these experiments will imme-
diately see how they can use a simple reactor model to show
the effect of flowrate on outlet conversion. The use of this
experiment at an early stage in reactor design courses will
help students learn basic concepts in reaction engineering.

A range of reaction rate expressions have been reported in
the literature. Morooka, et al.,15' reported the reaction rate for
a palladium catalyst in an atmospheric flow reactor as nearly
first order in propane:

1.3 1.6
r= kpropaneC oxygen
6.89( mo13 ) l 151.8kJ/mol
k=10 exp -- (4)
k= m89 TsJ ex 14x10-3 kJ/(molK)T

Recent reaction rate expressions for propane oxidation using
commercial monolithic catalysts were reported as first order
by Wanke161 and Bennett, et al.171 The pre-exponential and
activation energy values reported for these first order rates
are A = 3.15x104 mV/(kg s) and E, = 89,126 J/mol by Wanke
and A = 2.40x105 m/s and E, = 89,791 J/mol by Bennett.
In order for the students to determine the reaction rate
parameters, the reaction is assumed to be first order in pro-

Summer 2000


pane concentration, which is
in agreement with the results 1
of the previous two investiga- ) 0.9 7.76 Umin
tors. Next, the students are 0 0.8 model
asked to perform a nonlinear g 0.7
regression of the data using Eq. 0.6
(3). With most packages, a 0 0.5
good initial guess of the con- .2 0.4
stants is required and an esti- o 0.3
mate is obtained using Eq. (2). 0 0.2
Using a conversion of 0.52 and 0 0.1
temperature of 813 K, the value 0
of k is 0.0312. Next, a second 650 750
value of k is determined and Ter
the values of A and Ea are cal-
culated. An alternative method Figure 5. Comparison
is to use the rate parameters light-ofn
given above as initial starting
points. Finally, these values are used as the initial values for
the regression of Eq. (3) with the data using the nonlinear
regression package in POLYMATH.
The results of a regression of the low flowrate data are
shown in Figure 5. The reaction rate parameters from this fit
are A = 4.81E+04 m3/(kg s) and E, = 99,120 J/mol. This
value of activation energy is very close to that of both
Wanke and Bennett and is a reasonable activation energy. A
fit using all of the data results in the reaction rate parameters
of A = 2.39E+06 m3/(kg s) and E, = 124,500 J/mol. This
value of activation energy is higher than the values reported
by Wanke and Bennett and below the values reported by
Mooroka. This discrepancy in the activation energy could be
related to a poor flowrate measurement using a rotameter
(2% of full scale). Another possibility is that the catalyst
reaction rate deviates from first-order kinetics.
This fit of the data with a reactor model enhances the
student's connection between the concrete and the abstract.
Students perform this simple experiment and observe the
peak heights decrease as a function of furnace temperature
and increasing flowrate. The peak area is related to the
concentration of propane and a conversion is calculated.
Finally, the students apply a reactor model to successfully
describe the data. This experiment is an excellent combina-
tion of hands-on experiments, advanced analytical instru-
mentation, reactor modeling, and successful application of a
numerical technique.

There are many other experiments that can be performed
using this equipment. The most basic in reaction engineering
is the integral and the differential reaction-rate determina-
tions. In the differential reaction-rate determination, the total



of m

conversion must be kept
lower than 15% to obtain
S* good results for an automo-
tive catalyst.171 In addition,
the analytical technique for
gas measurement must
achieve a resolution of 1%.
For the integral reactor, ad-
ditional catalyst monoliths
may need to be added. Using
this apparatus, new catalysts
can be tested and the experi-
850 950 ment can be varied from year
nature (K) to year to minimize copying.
Alternatively, a new hydro-
odel with experimental carbon could be examined.
ve data.
e data. The major drawback of the

above two reaction rate de-
terminations is that they are time intensive; they require a
large number of trials to obtain a reaction rate as a function
of temperature. We have found the students quickly grow
disinterested in this experiment when they conduct a large
number of trials. The light-off experiments are simple since
only one inlet concentration is used and the temperature can
be ramped automatically.

This experiment must be operated in a safe and environ-
mentally responsible manner. All vessels must be rated for
pressures greater than the release pressure of the liquefied
propane tank. The concentrations of propane in the air stream
are representative of hydrocarbons present in the exhaust
gases and are well below the flammability limit of 2 mol%.
The products of this oxidation are primarily CO2 and water,
which are not harmful and can be vented to an exhaust
system. The furnace outlet is hot and students must be pre-
vented from touching it.

These experiments have been run by Rowan engineering
students and chemical engineering faculty at a unique hands-
on industrially integrated NSF workshop on Novel Process
Science and Engineering conducted at Rowan University.
We believe that reaction engineering comes alive when stu-
dents conduct innovative experiments in a laboratory set-
ting. In addition, these experiments catch the students' inter-
est because they are related to a commercially important
process-the automotive catalytic converter. Using this ex-
periment, students are able to see the catalyst, to measure
gas-phase concentrations and flowrates, and to model the
reactor and find the reaction-rate parameters using a nonlin-
ear regression of the data.

Chemical Engineering Education

Special thanks to Chris Bennett, George Quinlan, and
Wendy Manning from Johnson Matthey Catalytic Systems
Division for donating pretreated catalyst samples and giving
advice on reactor fabrication and catalyst operating condi-
tions. Support for the development of this experiment was
provided by a grant (DUE-9752789) from the National Sci-
ence Foundation through the Division for Undergraduate
Education and Rowan University.

1. Schmidt, L.D., The Engineering of Chemical Reactions, Ox-
ford University Press, New York, NY (1998)
2. Jacoby, Mitch, "Getting Auto Exhausts to Pristine," Chem. &

Eng. News, 25 Jan., p. 36 (1999)
3. Hayes, R.E., and S.T. Kolaczkowski, Introduction to Catalytic
Combustion, Gordon and Breach Science Publishers,
Amsterdam (1997)
4. Angelidis, T.N., and V.G. Papadakis, "Partial Regeneration of
an Aged Commercial Automotive Catalyst," Appl. Catalysis B:
Environmental, 12, 193 (1997)
5. Morooka, Y.,Y. Morikawa, and A. Ozaki, J. Catal., 7, 23 (1967).
Rates summarized in Mezaki, R., and H. Inoue, Rate Equa-
tions of Solid-Catalyzed Reactions, University of Tokyo Press
6. Wanke, S.E., "Oxidation of Propane Over a Diesel Exhaust
Catalyst," Can. J. Chem. Eng., 51, 454 (1973)
7. Bennett, C.J., S.T. Kolaczkowski, and W. J. Thomas, "Deter-
mination of Heterogeneous Reaction Kinetics and Reaction Rates
Under Mass Transfer Controlled Conditions for a Monolith
Reactor," Trans. IChemE., 69, B November (1991) J

rM. letter to the editor

Dear Sirs,
We welcome the comments of Baird and Rama Rao['l on
our paper concerning a simple experiment on two-phase
film flow[21 and hope that this discussion attracts the
readers to this somewhat neglected area in lab courses of
fluid mechanics.
Baird and Rama Rao point out that the experiment we
described must be performed in tubes with internal diameter
greater than 15 mm, for otherwise the bubble velocity will
not be given by the simple equation
U = 0.345(gD)05 (1)
It is true that for smaller diameters the effect of surface
tension becomes important and Eq. (1) ceases to be valid (if
the tube is small enough, the slug will not move as pointed

0.20 -



s o.:o

0.00 -

0.05 0.10 0.15 0.20
5/D (Nusselt's analysis for plane surface)

out), but in our paper we also present a general analysis for
laminar film flow that makes no use of Eq. (1) (see Eq. (7) in
ref. 2). However, if smaller tubes are used, the analysis
presented is only approximate since the curvature of the film
can no longer be neglected, and Nusselt's analysis is no
longer applicable. Figure 1 shows the correct film thickness
(calculated for cylindrical film flowt31) as a function of the
approximate film thickness given by Nusselt's analysis (ne-
glecting the film curvature). It can be seen that if the dimen-
sionless film thickness is greater than about 0.2, the errors in
film thickness become larger than 10%.
In order to minimize possible sources of error, it is sug-
gested that columns with internal diameters in the range of
15-35 mm be used. With larger tube diameters it may be
difficult to obtain laminar film flow, unless very viscous
solutions are used. Also, one has to use longer columns due
to greater velocities of the bubble, and the complexity of the
installation increases.
M.A. Alves, Teaching Assistant
A.M. Pinto, Associate Professor
J.R. Guedes de Carvalho, Professor
University of Porto, Portugal

1. Baird, M.H.I., and N.V. Rama Rao, "Letter to the Editor,"
Chem. Eng. Ed., 34, 65 (2000)
2. Alves, M.A., A.M. Pinto, and J.R. Guedes de Carvalho, "Two
Simple Experiments for the Fluid-Mechanics and Heat-
Transfer Laboratory Class," Chem. Eng. Ed., 33, 226 (1999)
3. Brown, R. A. S., "The Mechanism of Large Gas Bubbles in
Tubes. I- Bubble Velocities in Stagnant Liquids," Can. J.
Chem. Eng., 43, 217 (1965) 0

Summer 2000

Figure 1- Correct versus approximate film thickness.

Class and home problems

The object of this column is to enhance our readers' collections of interesting and novel
problems in chemical engineering. Problems of the type that can be used to motivate the student
by presenting a particular principle in class, or in a new light, or that can be assigned as a novel
home problem, are requested, as well as those that are more traditional in nature and that
elucidate difficult concepts. Manuscripts should not exceed ten double-spaced pages if possible
and should be accompanied by the originals of any figures or photographs. Please submit them to
Professor James O. Wilkes (e-mail:, Chemical Engineering Depart-
ment, University of Michigan, Ann Arbor, MI 48109-2136.




King Saud University Riyadh 11421, Saudi Arabia

Continuous processes show dynamic behavior during
start-up, shutdown, and when upsets occur during
steady-state operation. Mathematical modeling, simu-
lation, and control of these processes is relatively difficult
because of the nonlinear nature of these processes and the
activation and tuning difficulties of the controllers.
This paper applies proportional plus integral (PI) control
to start up a non-isothermal CSTR. PI eliminates offsets and
maintains an acceptable speed of response. Simple and
straightforward schemes of activation are tried to start up the
CSTR smoothly and to get the maximum attainable con-
version. The importance of this control problem lies in
the difficulty of triggering the controller and the returning
of the PI settings.

We will consider the start-up of a non-isothermal CSTR,
which has been studied in detail."'2] A reaction of the form
A + B -- C + D and of known kinetics has been considered.
The CSTR has an overflow and two feed streams, one for
pure A and the other for pure B. Mathematical models along
with analytical and numerical solutions have been devel-
oped. Various types of start-up have been modeled and

simulated, each type being represented by a different model.
The models treated A and B as if they were in a total feed
flow. In the present study, however, the models are modified
to account for separate feed flows, because each feed flow is
used here as a manipulated variable. Also, instead of us-
ing different models, the equations are grouped here in
one general form:
dV dV
dt = F +F2 for V < Vr, otherwise = 0 (1)

Aziz M. Abu-Khalaf has educational inter-
ests that include developing new objectives
and improving the performance of laborato-
ries at the Chemical Engineering Department
at King Saud University. Research interests
include controlled release systems and cor-
rosion. He can be reached by e-mail at
A 0

Emad M. All is an assistant professor at King Saud University. He
received his PhD ion Chemical Engineering from the University of
Maryland in 1996. His research interests are in process control. (Photo
not available.

Copyright ChE Division of ASEE 2000

Chemical Engineering Education

dt A FICAf rV

dVCB = F2CBf rV

dVCC -(FI +F2)(Ccf)+ rV (4)

dVCD (F +F2)(CDf)+ rV (5)

(4 )dT
i=I P' dt

FCAfCpA(Tf -T)+ F2CBfCpB(Tf -T)-rVAH Q (6)


r=kCACB k=10.ex (-48.3) Q =4r hair (T Tamb)

The values of the parameters and the initial conditions are
listed in Tables 1 and 2, respectively.
1. Operating conditions during start-up have been
found to affect the product quality. For example, changes in
pressure drop in the head tanks produce changes in feed
flows. Perform an open-loop simulation of the model and
find the effect of changing the flow rate on the yield.
2. Operate the system starting from an empty tank
with no inlet flows up to a fully filled tank with the maxi-
mum product concentration. Perform start-up once without
disturbances and then with disturbances in the feed flows
and in the input reactant concentration. Use a standard feed-
back PI control system, with the controlled variable being
the concentration of product C.

Process Parameters

Parameter Value Parameter Value
Tf 24'C Cp^ 75.25 J/molC
Tamb 290C Cp, 175.3 J/molC
Vr 2.8 1 Cpc 78.2 J/molC
h 2.5 J/mC min CpD 103.8 J/molC
AH -1.5 kJ/mol CA, 0.1 mol/1
D 15 cm CB 0.1 mol1
Tr, 240C Ccf 0.0 mol/
CD, 0.0 mol/

Initial and Starting Conditions

F1 F, CA CB Cc C, V T
0.0 1/min 0.0 1/min 0.1 mol/1 0.1 mol/1 0.0 mol/1 0.0 mol/1 0.0 1 24C

3. To perform start-up successfully, we need to tune
(2) and activate the controller. Use the modified Ziegler-Nichols131
(Z-N) tuning method to find out the best settings, and then
(3) suggest different schemes to activate the controller.


Simulation of the model equations (Eqs. 1-6) was per-
formed using the package DASSL. This is a differential/
algebraic system solver that uses the backward differentia-
tion formulas of orders one through five. Figure 1 shows an
open-loop simulation of the models for two different values
of F, and F,. As shown, a lower yield is obtained at a higher
flow rate. Thus, the CSTR needs to be operated at lower
flow rates. Specifically, F, = F, = 0.1 1/min, which produces
Cc = CD = 0.0327 mole/1, is considered as the desired operat-
ing condition (set point) in this work.
The manipulated variables are the positions of the valves,
where the feedback control law141 is given by

v(t)= vo +kc1C'F -CC(t)] + J[Cs -CC(t)]dt (7)
v2(t)= V2o +kc2[CP -Cc(t)]+k2 J[c- Cc(t)]dt (8)
___12 R c c 0

These laws will be implemented in a discrete time fashion
with sampling time of 1 min. Note that these control loops
form a split-range control scheme, because there are two
manipulated variables to control one variable, with the con-
trol signal being split into equal parts, each affecting one
valve. Although this is not the common split-range configu-
ration, it simplifies the problem. Alternatively, v2 can be
driven by the error signal of CD, which will result in the same

r I

0.10 -


0 04




0 2

0 20 40 60
Time (min)

Figure 1. Open-loop simulation:
solid lines, F, = F,= 0.1 1/min;
dashed lines, F, = F2 = 0.5 1/min.


0 40
ime (min)







Summer 2000



control performance because the reaction has a constant
stoichiometric ratio of one and identical valve dynamics.
Note that feed flows are taken as linearly proportional to the
valve positions of

F, =Cvli (9)
F2 =C2V2 (10)
Here, C, = Cv2 =1.0 mole/min.
Controller Tuning
Tuning determines the best settings for the adjustable pa-
rameters of a feedback controller. Closed-loop testing that
produces constant output cycling is used in selecting these
values. The desired values for ki and T1 are determined
based on the modified Z-N tuning criterion. The original Z-
NI51 method is based on the quarter decay ratio, which might
result in oscillatory feedback response. The modified Z-N
method gives more conservative settings. The purpose of
the controller tuning is to obtain an initial value for the
PI settings, which will be adapted on-line in a gain-
scheduling formulation.
Tuning of the feedback control using an ultimate gain
methods, i.e., Z-N, is based on continuous incrementation of
k, and observation of the resulting closed-loop response.
The proportional gain that produces sustained oscillation is
known as the ultimate gain from which the PI settings can be
inferred.161 Since the zero steady state is an unstable one, this
method cannot be applied at this operating point. Thus, the
PI settings are obtained by applying the Z-N as

kcl = kc2 =-76.0
TI1 = 12 = 1.0 min
The negative controller gain is an indication of the reverse
action mode, because the process has a negative static gain,
k, (see Figure 1).
Controller Activation
Consider Eqs. (7) and (8). In order to start up the reactor at
t=0, the following condition must be satisfied:

vi(0)= vio +kciCc -C(0)]>0 (11)

Obviously, this cannot be satisfied for a negative controller
gain and a zero value of initial valve position. In order to
overcome this problem, we examine four strategies of start-
Strategy I Perturb the inflows manually and trigger the PI
algorithm simultaneously.
Strategy II Perturb the inflows manually and trigger the PI
algorithm after a specific time interval.
Strategy III Trigger the PI algorithm with gain-scheduling
according to kc,k = constant.
Strategy IV Trigger the PI algorithm with gain-scheduling

using the IMC-type controller, kcikp =T / X
Notice that for Strategy 1, a large value of k, might cause
vi to remain zero for any value of vio in its allowable range
of [0,1]. The maximum allowable magnitude of k,, in this
case is Ikci for k are used in Strategy I. As for Strategy II, the larger
value of kc, obtained by the Z-N method can be used with
Vio =0.1 and the controller can be triggered one sampling
time later.
In Strategy III, the controller gain will be adapted on-line
according to ki(t) = kokpo/kp(t), where k1 and kpo are con-
stant reference values. In this investigation, k1o is taken equal
to -76, which is found by the Z-N method, and kpo as the
static gain corresponding to F, = F2 =0.1 1/min. Values of
kp(t) can be estimated from Table 3, which lists different
values for the static gain at various operating conditions.
Values of kp for F,>0 were computed using the exact linear-
ization of the process model and using the reaction-curve
method. Both methods gave almost identical results. Ini-
tially, with F,=0, the operation behaves like an integrator
process; hence kp(0) was found by the pulse testing.'71
For Strategy IV, the controller gain will also be adapted
on-line with a changing process gain (k,) and time constant
(X) according to kci =(t)/[kkp(t)] where X is the IMC
filter parameter or closed-loop time constant used to adjust
the speed of the closed-loop response. Usually, an IMC-type
tuning is used to determine fixed PI settings using identified
process parameters. In addition, robustly tuned PI settings
can be obtained by conducting an adequate robustness analy-
sis.11 Here, we allow the IMC-type controller gain to vary in
order to adapt to the process changing gain.
The gain-scheduling approaches (Strategies III and IV)
are conducted as follows:
At t = 0
set kp(0) = kp(F,=0) and T(0)=Tav
At t > 0
set kp(t) = kpav or interpolate kp from the various values
of kp in the range F, e [0.01,1.0]

set T(t)= Tav or interpolate c from the various values
of r in the range F1 e[0.01,1.0]

Start-Up Without Upsets

Closed-loop simulation for C-P=0.0327 using the pro-
posed activation strategies is depicted in Figure 2, which
shows the time response of the product concentration, Cc,
and the inlet flow of pure A, F,. In all cases the controller
was able to bring the product concentration to the desired

Chemical Engineering Education

value. The feed flows varied initially and then settled at their
expected values of 0.1 1/min; however, the smooth re-
sponse for Strategy II was only achieved by de-tuning
the PI settings to ke, = k,2 = -9.1, whereas the feedback
response using the original values of the PI settings was
found to be very aggressive.

It is clear from Figure 2 that Strategy I has the slowest
response due to a small controller gain. A larger value of kci
and consequently a faster response of Strategy I can be
achieved using a larger initial perturbation value. With re-
spect to the responses, Strategy III outperformed other strat-
egies where the closed-loop response using a constant aver-
age value and a variable interpolated value of kp are almost
the same. Initially, F, varies for a few samples, then settles
down to a constant value, giving a constant process gain. For
strategy IV, the use of average values for the process param-





Strategy 1
Strategy II
Strategy III
Strategy IV

0.10 .
- 0.05

.0.65 !


-0.05 -

20 40 60
Time (min)

Figure 2. Closed-loop response for CP =0.0327 mole/1.

S-trategy I
-- Strategy It

Strategy III
Strategy IV

80 100

Identified Process Gain and Time Constant
at Various Operating Conditions

Step Change in Operating
F, (1/min) (min) Condition kp

0.01 8 V=0 0.0111
0.05 6.5 V= 2.8, F,= 0.01 -0.1382
0.1 5 V =2.8, F = 0.1 -0.0338
0.2 3.5 V = 2.8, F, =0.2 -0.02
0.4 2.5 V = 2.8, F, =0.4 -0.0108
0.6 2.0 V = 2.8, F = 0.6 -0.0071
0.8 1.5 V = 2.8, F = 0.8 -0.0051
1.0 1.35 V = 2.8, F, =1.0 -0.0039


o 0.03 Strategy I
-- Strategy 1I
0.02 Strategy I
.... Strategy IV

0.5 Strategy I
Strategy I

1.00 -


-0 05
0 20 40 60 80 100
Time (min)

Figure 3. Closed-loop response for Cc =0.0327 mole/i
with -0.05 step change in both feeds starting at t=5 min.

0.15 --

S0.05 -- Strategy I

-~ I .Strategy IIf
;'-0.05 ^ --

S.... Strategy IV


60 80

20 40
Time (min)

Figure 4. Time response of valve position for Cs =0.0327
mole/i with -0.05 step change in both feeds starting at t=5

Summer 2000

eters gave a smoother closed-loop. A value of X = 2 pro-
duced a less aggressive performance, which took Strategy
IV to a slightly slower response than that of Strategy III.

Start-Up with Upset in the Feed Flow

We next examine the change in the set point with a step of
-0.05 in the feed flows at t=5 min during start-up (see Figure
3). The associated response of the valve openings is demon-
strated in Figure 4. Obviously, the upset in the feed flows
has marginal effect on feedback response of the product
concentration for all cases except for Strategy I with vi. = 0.1,
where a larger overshoot is observed. Unlike the previous
case, the valve response differs from that of the feed flow
since the disturbance affects the latter only. In this case, the

valve position settles at a steady-state point higher than
that in Figure 2. This increase in the valve opening was
made by the controller to balance reduction in the feed
flows produced by the disturbances.
Start-Up with Upset in the Feed Concentration
In order to illustrate the efficiency of the feedback
control scheme, all of the control activation strategies
were tested for the same set point as above, but with a
disturbance in CBf. A step change of -0.02 mole/1 start-
ing at t=10 min was considered (see Figure 5). Obvi-
ously, the proposed feedback schemes maintained
the product quality at the desired value despite the
sudden reduction in the inlet concentration, CB,. Ul-
timately, the inlet flows reached a value lower than
that without upsets.
Start-Up with Different Set Point
Another advantage of the feedback scheme is its
ability to maintain desired specifications without the
knowledge of the optimal operating conditions before-
hand. For example, in order to maximize the product
yield, a larger set point for the product concentration
can be specified for the controller. Figure 6(a,b), for
example, illustrates the feedback response of the pro-
cess for Cs =0.04 mole/1. Although simulation indi-
cates that such a yield is achievable, it operates the
process at a very low throughput of 0.03 1/min. Simi-
larly, Figure 6(c,d) demonstrates the process dynamic
behavior for Cs = 0.05 mole/I. Obviously, the reaction
can be brought to such a high yield, but this would be at
the expense of operating the process in a semi-batch
mode as the feed flows approached zero at steady state.


Automatic start-up of a non-isothermal CSTR using
a conventional PI control algorithm was considered.
Four controller activation/adaptation schemes were
tested and compared. Overall, Strategy III presented
superior performance, full automation, and ease of
implementation. Strategy I had the most sluggish re-
sponse since the maximum allowable controller gain is
restricted by the value of the initial valve opening.
Strategy II lacks full automation and requires re-tuning
for stability. On the other hand, Strategy IV requires
proper adjustment of the IMC filter for good perfor-
mance. Nevertheless, the performance of gain-schedul-
ing approaches (Strategies III and IV) depends on the
identified process parameters.
A theoretical model should be developed or identifi-
cation methods be used along with these approaches.
Another practical operation of the process is to maxi-
mize the yield and throughput. This issue can be ad-
dressed through implementation of optimal control

t,0.0 /^ ^.--~-~--mL;;~;-c------- ^---
S0.03 Strategy
0.02 -- Strategy II
E 0.01 Strategy 1l
0 o 0 Strategy III

.- Strategy 1
-0.Strategy 11

0 20 40 60 80
Time (min)

Figure 5. Closed-loop response for CP =0.0327 mole/1
with -0.02 step change in CBf starting at t= 10 min.

(a) Strategy I
. 003 / 1
| 0.65 --- Strategy II
S- strategy IIIl
S ......... Strategy IV

0. 00- -
1.00Time (min(b)


with -0.02 step change in C__ starting at t=10 min.


0Strategy I

,Strategy IV~



E 0.65

-0.05 -- -- -- -

0 20 40 60 80
Time (min)

Figure 6. Closed-loop response for CsP =0.04 mole/1 (a,b)
and C = 0.05 mole/I (c,d).

Chemical Engineering Education

Questions for Further Study
1 Question #1: Derive the model equations considering one
mode of start-up, e.g., adding both reactants simulta-
neously, until the reactor overflows.
[L Question #2: Consider an emergency shutdown in which
the feed flows are suddenly stopped and the reactor is
to be drained. Would the equations for this case be
different from those representing start-up? How?
El Question #3: Repeat the above calculations using a first-
order reaction.191 Is it going to affect the controller
settings and activation?
El Question #4: What would be the effect of adding a de-
rivative action to the controller (i.e., using a PID) on
the start-up of the process?

C concentration of species i, mole/1
Cf feed concentration of species i, mole/l


concentration set point for species i, mole/l
heat capacity of species i, J/mole C
characteristic constant for valve i
reactor diameter, m
feed flow rate of pure component A, 1/min
feed flow rate of pure component B, 1/min
heat transfer coefficient for air, kJ/m2C min
standard heat of reaction, kJ/mole
reaction rate constant, I/mole min
controller gain for loop i
process gain
average process gain

Q rate of heat loss to the surrounding, kJ/min
r reaction rate, mole/1 min
R gas constant, 0.008314 kJ/mole K
T reactor temperature, C
Tamb ambient temperature, C
T, feed temperature, C
T rereference temperature, C
t time, min
V fluid volume, 1
vi valve i position
vo initial position for valve i
Vr reactor volume, 1
? IMC filter (closed-loop time constant)
T process time constant
,av average time constant
T1 integral time for PI controller

1. Abu-Khalaf, A.M., "Start-Up of a Non-Isothermal CSTR:
Mathematical Modeling," Chem. Eng. Ed., 31(4), 250 (1997)
2. Abu-Khalaf, A.M., "Mathematical Modeling of an Experi-
mental Reaction System," Chem. Eng. Ed., 28(1), 48 (1994)
3. Seborg, D.E., T.F. Edgar, and D.A. Mellichamp, Process
Dynamics and Control, John Wiley, New York, NY (1989)
4. Stephanopoulos, G., Chemical Process Control, Prentice-
Hall, Englewood Cliffs, NJ (1984)
5. Ziegler, J.G., and N.B. Nichols, "Optimum Settings for Au-
tomatic Controllers," Trans. ASME, 64, 759 (1942)
6. Ogunnaike, B., and W. Ray, Process Dynamics, Modeling,
and Control, Oxford University Press, New York, NY (1994)
7. Luyben, W., Process Modeling, Simulation, and Control for
Chemical Engineers, McGraw-Hill, New York, NY (1990)
8. Morari, M., and E. Zafiriou, Robust Process Control, Prentice-
Hall, Englewood Cliffs, NJ (1989)
9. Abu-Khalaf, A.M., "Dynamic and Steady-State Behavior of
a CSTR," Chem. Eng. Ed., 30(2), 132 (1996) 3

s letter to the editor

To the Editor:
In the Winter 2000 issue of Chemical Engineering Educa-
tion there was an interesting paper by S.H. Munson-McGee 11
that presented a laboratory sequence with the objective of
developing abilities in chemical engineering students ac-
cording to EC 2000 criteria.[21 The author describes a four-
course sequence, beginning with the study of the theoretical
aspects of experimental design and data analysis and finish-
ing with a unit operations laboratory.
Table 1 of that paper shows a short description of each of
the nine experiments that can be carried out by the students
with the Process Instrumentation Laboratory course. Unfor-
tunately, the mentioned Table 1 contains a typographical
mistake and the simple change of a "d" for a "b" causes a
considerable conceptual effect: effectively, the experiment,
titled "Absorption by activated carbon. Blue food coloring
was absorbed from aqueous solutions..." is actually an ad-

sorption experiment. (Table 2 refers, correctly, to this ex-
periment as an adsorption process.)
From my point of view, it is important to correct this type
of typographical error where two very similar words refer to
two very different processes, in order to prevent confusion
and conceptual mistakes among students. This is especially
important in journals such as Chemical Engineering Educa-
tion because of its content, which is very readable by chemi-
cal engineering undergraduates.
Amparo G6mez Siurana
Universidad de Alicante

1. Munson-McGee, S.H., "An Introductory ChE Laboratory In-
corporating EC 2000 Criteria," Chem. Eng. Ed., 34(1), 80
2. "Engineering Criteria 2000," Accreditation Board for Engi-
neering and Technology, Inc., 111 Market Place, Suite 1050,
Baltimore, MD (1998) O

Summer 2000


e= laboratory




University of Connecticut Storrs, CT 06269-3222

Hands-on challenges that demonstrate and reinforce
important concepts benefit the learning process-
this is especially true for the often abstract subject
of process dynamics and control. Hands-on challenges can
be motivating, can promote critical thinking, facilitate un-
derstanding in the use and limitations of the theory, and help
prepare students for the challenges of the professional world.
Too often the application of textbook theory is limited to
solving questions listed at the end of the chapter. One typical
question is to have the student expand or extend a math-
ematical development presented in the book. Another is to
provide bits of data and challenge the student to select and
employ a combination of formulas to obtain a desired result.
Unfortunately, even when cleverly crafted, these textbook
problems fall short of providing students with the depth or
breadth of practice required for comprehension and mas-
tery.t 2] Thus, the Chemical Engineering Department at the
University of Connecticut supplements the textbook with
laboratory exercises. Hands-on laboratory exercises are ex-
tremely important to learning because they help students
make the intellectual transition from theory to practice.t3]
The abstractions presented in textbooks are literally brought
to life through the tactile nature of a lab experience.
Unfortunately, the reality of the laboratory at the Univer-
sity of Connecticut is that each study can take many hours,
and even days, to perform. Also, equipment failures and
other problems teach the important lesson that the real world
can be uncertain (this lesson is not usually intended to be
the objective of a particular assignment, however). Thus,
students rarely explore more than a very few central
concepts in the lab.
A training simulator offers an alluring method for provid-

ing students with the significant hands-on practice critical to
learning process control. The proper tool can provide virtual
experience much the way airplane and power-plant simula-
tors do in those fields. It can give students a broad range of
focused engineering applications of theory in an efficient,
safe, and economical fashion. And it can work as an instruc-
tional companion as it provides interactive challenges that
track along with classroom lectures.
Process control is a subject area well suited to exploit the
benefits of a training simulator.[4] Modern control installa-
tions are computer based, so a video display is the natural
window through which the subject is practiced. With color
graphic animation and interactive challenges, a training simu-
lator can offer experiences that literally rival those of the real
world." These experiences can be obtained risk free and at
minimal cost, enabling students to feel comfortable explor-
ing nonstandard solutions at their desks. If properly de-
signed as a pedagogical tool with case studies organized to

Doug Cooper is Professor of Chemical Engi-
neering at the University of Connecticut. His re-
search focuses on developing control methods
that are both reliable and easy for practitioners
to use. He has studied the control of fluidized
bed incineration, heat exchange, distillation, in-
jection molding, surge tanks, and catalytic reac-

Danielle Dougherty received a BS in Chemical
Engineering from Widener University in 1997 and
is currently working toward her PhD in Chemical
Engineering under the direction of Doug Cooper
at the University of Connecticut Her current re-
search interests include model predictive control
and nonlinear adaptive control.

Copyright ChE Division of ASEE 2000

Chemical Engineering Education

present incremental challenges, we believe learning can be
enormously enhanced for process control with such a train-
ing simulator.


Each discipline views process control from a different
perspective. To help orient the reader, consider these "typi-
cal" examples drawn from chemical process control:

Process Variables: temperature, pressure, pressure
drop, level, flow, density, concentration

Final Control Elements: solenoid, valve, variable speed
pump or compressor, heater or cooler

Control Algorithms: on/off PID, cascade, ratio, feed
forward, multivariable decoupler, model predictive

Process Applications: reactors, separators, distillation
columns, heat exchangers, furnaces

Many chemical engineering processes are literally one-of-
a-kind. Consequently, their associated control system will
be unique in design and implementation.

Additionally, chemical processes can be nonlinear and
nonstationary, and can have long time constants, significant
dead time, and/or noisy measurement signals. Disturbances
occur from numerous sources, including loop interaction

I e Fl (Jnl-1
I I 178

Cor am.l
Output (.I

manipulated varable

process variable
La. Ta
I 402

controller out

level sensor
S& controller
4^_ ^ gtotm


disturbance PupedFio, ou r F" "
variable 158

Figure Gavity-drained tanks graphic.
Figure 1. Gravity-drained tanks graphic.

Model: First Order Plus Dead Time (FOPDT) File Name: tanks.txt
35 --- --- ----
. t:: Process Data
S3.0 -
2.5 -- ---
i "":::::I Model Fit of Data

65 ------------;----.: --------- 5--,-------.--.---
0 60
55 ---I- ---- -- ----_ 1 .._ .. .-...--- .. _;-

0 5 10 15 20
Gain (K)= 0 1084, Time Constant (TI) = 1.49, Dead Time (TD) 0.592
SSE: 0.0697

Figure 2. FOPDT model fit of text data.

from other controllers in the plant.


The following lessons have been drawn from the Control
Station process control training simulator to illustrate the
value such software provides the curriculum. We note that
training simulators are distinguished here from tools such as
Matlab,15' which have a primary function of design, analysis,
and simulation. The reader can download a free control-
station demo at

P-Only Controller Performance

The computer graphic display for the gravity-drained tanks
process, shown in Figure 1, is two vessels stacked one above
the other. Liquid drains freely through a hole in the bottom
of each tank. The controller output signal manipulates the
flow rate of liquid entering the top tank. The measured
process variable is liquid level in the lower tank. The distur-
bance variable is a secondary flow out of the lower tank
from a positive displacement pump, so it is independent of
liquid level except when the tank is empty.

Students begin their studies with this process because its
dynamic behavior is reasonably intuitive. If they increase
the liquid flow rate into the top tank, the liquid levels rise in
the tanks; if they decrease the flow rate, the levels fall.

The traditional place to begin a course is with the study of
process dynamics. Students generate a step-test plot and
compute by hand the first-order-plus-dead-time (FOPDT)
model parameters: steady-state process gain, Kp, overall time
constant, rp, and apparent dead time, 6p. After they have
gained mastery with hand calculations, they use tools that
automate the model-fitting task so they can explore more
practical tests. A Control Station fit of test data is shown in
Figure 2 for the gravity-drained tanks.

Students use their FOPDT model parameters in tuning
correlations to compute a P-Only controller gain, Kc. Figure
3 displays a Control Station strip chart showing set-point

Process: Gravity Drained Tank Controller: PID ( P- RA, I off, D= off)
36 ---- -
34 --- -- ; --------
32 ------ --

S28 - ------------- Offset ----------
Design Set Point
64 ---- ------------- -

S58 -----. ------------------- ----------
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
Time (mins)
Tuning: Bias = 60.0, Gain = 5.75, Sample Time =I 100

Figure 3. P-Only set-point tracking results in offset.

Summer 2000

I, .


... we do not believe a training simulator is better than,

or a replacement for, real lab experiences. In fact, we believe that

hands-on studies with actual equipment are fundamental to the learning process.

tracking performance for the gravity-drained tanks under P-
Only control. The Kc for the controller is computed from the
integral time-weighted absolute error (ITAE) correlation,6"71
using the FOPDT model parameters from Figure 2.

With this as a starting point, the students now turn to what-
if studies. The investigation of Figure 4 explores how Kc
impacts offset and oscillatory behavior for set-point tracking
under P-Only control. Students also explore disturbance re-
jection under P-Only control. Is the best tuning for set-point
tracking the same as for disturbance rejection? And, how is
"best" tuning defined?

For this and all Control Station processes, the student can
change the level of random noise in the measured process
variable. They can also manipulate the controller output
signal, set point, and disturbance variable using a step, oscil-
lating, ramp, or pseudo-random binary-sequence (PRBS)
signal sequence. The current version of Control Station of-

fers only one disturbance variable for each process, and this
disturbance can be changed at will by the student. We note
that this is not realistic in that a real plant can have many
disturbances from a variety of sources that will affect the
process, and as disturbances, they are generally not available
for manipulation by the engineer. The students are made
aware of this during class.

PI Control and Nonlinear Behavior

The computer graphic for the countercurrent, shell and
tube, lube oil cooler (a kind of heat exchanger) is shown in
Figure 5. The controller-output signal manipulates the flow
rate of cooling liquid on the shell side. The measured pro-
cess variable is lube oil temperature exiting on the tube side.

Students learn an important lesson about process dynam-
ics by studying the nonlinear character of this process as
shown in Figure 6. The steady-state gain of the process

I 50

0 10 20 30 40
Time (mins)
Tuning: Bias = 60.0, Gain = 30.0, Sample Time = 1.00

Figure 4. P-Only performance changes as Kc changes.

Figure 5. Heat exchanger graphic.





- 50
6 30

10 15 20 25
Time (mins)

30 35 40

Figure 6. Heat exchanger displays nonlinear behavior.

5 10 15 20 25 30 35 40 45
Time (mins)
Tuning: Gain = -1.97, Reset Time = 1.04, Deriv Time = 0.0, Sample Time = 1.00

Figure 7. Nonlinear behavior impacts performance.

Chemical Engineering Education

Controller: PID ( P= RA, I= off, D= off)

Gravity Drained Tank

- -- I

--O ---- -e --- ---------
----- --'- O ffset- - -

-c--------[-- KC 30.0 ] __ -

Controller: Manual Mode

:Heat Exchanger

Measured Response Changes
With Operating Level
----- -- : :-:---- -- -- ---

Even Though Steps in Controller
-- ---- ---- Output Are All the Same

--- -- - -

Med Se
Temp (c)
1 2200

cooling flow exit

controller output I
Outpt [m)
F93 ---
temperature sensor .
& controller sa Pr tO
| 1400

process variable

clearly changes as operating level changes. Less obvious is that the
time constant of the process also changes.
For processes that have such a nonlinear character, the perfor-
mance of a controller will change as the process moves across
operating levels. Figure 7 illustrates this point. The exchanger is
under PI control, and as the set point is stepped to different
operating levels, the nonlinear behavior of the process clearly
impacts set-point tracking performance. Thus, students learn
that a controller is designed for a specific or design level of
operation. Best practice is to collect dynamic test data as near


"desired" L
_____ performance


0.5 r, r 2r,

Figure 8. PI controller tuning impacts performance.

--- :-- ITAE tuned -
------- -- -- / -- PID

0.5r Tio 2 r

Figure 9. Derivative action impacts oscillatory behavior.

10 20 30 40 50 60
Time (mins)
Tuning: Gain = -1.70, Reset Time = 1.40, Deriv Time = 0.3051, Sample Time = 1.00

Figure 10. Measurement noise is amplified and reflected
in controller output signal.

as practical to this design.
Figure 6 also shows that the heat exchanger has a
negative steady-state gain. Students learn that a com-
plete design includes specifying the action of the con-
troller (reverse vs. direct acting).[6-8] They learn this
concept because if they enter it wrong, the controller
output will quickly drive the valve to either full open
or full closed and it will remain there until the cor-
rect controller action is entered.
For what-if studies, students explore how PI control-
ler tuning parameters interact and affect set-point track-
ing performance. Figure 8 shows a tuning map that
they develop from an orderly tuning investigation
using an ideal linear transfer function process avail-
able in Control Station.

PID Control and Measurement Noise

Derivative action can decrease the process settling
time because it resists rapid movement in the measured
process variable.161 In Control Station, the PID control-
ler algorithm is currently implemented using the ideal
(noninteracting) form[6-9] with a choice of derivative ac-
tion either on controller error or process measure-
ment. Students learn how derivative action impacts
controller performance with studies similar to that
shown in Figure 9, which focuses on the derivative
time tuning parameter.
The center plot of Figure 9 shows the set-point track-
ing performance of a PID controller tuned using the
ITAE16.7' for set-point tracking correlation. For all plots
in Figure 9, Kc and T, remain constant and the measure-
ment noise has been set to zero. The plot to the left in
Figure 9 shows how the oscillating nature of the
response increases as derivative action is cut in half.
The plot to the right shows that when derivative
action is too large, it inhibits rapid movement in the
measure process variable, causing the rise time and
settling time to lengthen.
When noise is added to the measured process vari-
able, students learn that derivative action amplifies it
and reflects it in the controller output signal. Figure 10
shows this quite clearly with a side-by-side comparison
of a PI and PID controller. For this comparison, the
same amount of measurement noise was used through-
out the experiment. This study helps students visualize
that a PI controller is not impacted by noise while the
derivative action of the PID controller reflects and am-
plifies it in the controller output signal.
Students also compare derivative on controller error
to derivative on process measurement. Watching the
derivative on error "kick" after a set-point step is a more
memorable experience than simply hearing about it.

Summer 2000

Figure 11. Jacketed reactor graphic.

Cascade, Feed Forward, and Disturbance Rejection

The jacketed reactor graphic, shown in Figure 11, is a
continuously stirred tank reactor in which an irreversible
exothermic reaction occurs. Residence time is constant in
this perfectly mixed reactor, so the steady-state conversion
from the reactor can be directly inferred from the tempera-
ture of the reactor product stream. To control reactor tem-
perature, the vessel is enclosed with a jacket through which a
coolant passes.

The controller output manipulates the coolant flow rate
through the jacket. The measured process variable is product
exit-stream temperature. If the exit-stream temperature is
too high, the controller increases the coolant jacket flow to
cool down the reactor. The disturbance variable is the inlet
temperature of coolant entering the cooling jacket.

The jacketed reactor can be run in three configurations:
feedback control, as shown in Figure 11, feed forward with
feedback trim, and cascade control. When the cooling jacket
inlet temperature changes, the ability to remove heat changes
and the control system must compensate for this distur-
bance. Cascade and feed forward are control strategies used
for improved disturbance rejection.

t 93
a 92
> 90
( 50
8 45


s: Single Loop Jacketed Reactor

Controller: PID with Feed Forward

0 10 20 30 40 50
Time (mins)
Tuning: Gain = -3.78, Reset Time = 1.89, Deriv Time = 0.0. Sample Time = 1.00
Process Model: Gain(Kp)= -0.34, TI = 1.90, T2 = 0.0, TD = 0.81, TL = 0.0
Disturbance Model: Gain(Kd) = 0.72, T1 = 2.70, T2 = 0.0, TD = 1.00, TL = 0.0

Figure 12. Benefits of feed-forward control.

Cascade design involves the tuning of two controllers.
Feed forward requires identification of an appropriate pro-
cess and disturbance model.

The rejection of a step change in the disturbance variable
(jacket inlet temperature) for a single loop PI controller is
compared in Figure 12 to a PI with feed forward control-
ler. The benefit of feed forward is clear for this process
because for the same disturbance, the measured process
variable has a much smaller maximum deviation and a
faster settling time.

Students compare single-loop, feed-forward, and cascade
control in a series of exercises. They investigate tuning
issues, which PID modes to use in a cascade, the order and
accuracy of the models needed for feed-forward design,
plant-model mismatch, dead-time issues, and a host of other
interesting challenges.

Control Loop Interaction and Decoupling

The graphic shown in Figure 13 is a binary distillation
column based on the model of McCune and Gallier.'lo0 The
column has two measured process variables and two ma-
nipulated variables. The reflux rate is used to control distil-

disturbance Fmie.
variable .

manipulated variable

Sl..m] y Dilit
S840 cmp~aO ) measured
process variable
con -. 0 ~_Sn
op.mpo8 samr
l 5r I 946 composition sensor
& controller

kseel c ap, 0 manipulated variable
I 216 I t0t

r- I 2.
S(cc)l---- composition sensor
1 -J & controller
c 2y.} -- measured
process variable

Chemical Engineering Education

S PI Only PI w/ Feed Forward

.. .. ... . .. ..: :

---- -

Figure 13.





Process: Distillation Column Top: PID /w Decoupler / Bot: PID /w Decoupler

S95 .
9: : .* *

S No Decoupling I----With Decoupling ..
-- -- ---- -- -

0 10 20 30 40 50 60 70 80 90 100
Time (mins)
Tuning: Gain 1.20, Reset Time = 0.8913, Deriv Time = 0.0, Sample Time = 1.00
Process Model: Gain(Kp) = 1.06. TI = 0.8913, T2 = 0.0, TD = 0.336, TL = 0.0
Disturbance Model Gain(Kd)= -0.8944, TI = 1.07, T2 = 0.0, TD = 0.3366, TL = 0.0
Tuning: Gain = -9.72, Reset Time = 1.00, Deriv Time = 0.0, Sample Time = 1.00
Process Model: Gain(Kp)= -0.2993, TI = 1.00, T2= 0.0, TD = 0.1657, TL = 0.0
Disturbance Model: Gain(Kd)= 0.2556, TI = 1.17, T2 = 0.0, TD = 0.1657, TL = 0.0

Figure 14. Distillation column shows loop interaction.

late purity and the steam rate is used to control purity of the
bottoms stream.
Students use this process to explore the interactions that
can occur in such multicontroller applications. Control-loop
interaction occurs because when the distillate purity out of
the top of the column is too low, the top controller compen-
sates by increasing the flow of cold reflux into the column.
This increased reflux flow will indeed cause an increase in
the distillate purity. The additional cold reflux will work its
way down the column trays, however, and eventually begin
to cool the bottom of the column. This cooling causes the
purity of the bottoms stream to move off set point and
produce a controller error.
The bottom controller compensates by increasing the flow
of steam into the reboiler. This produces an increase in hot
vapors traveling up the column, which eventually causes the
top of the column to begin to heat up. The result is that
distillate purity again becomes too low. In response, the top
controller compensates by again increasing the flow of cold
reflux into the column.
This controller "fight" is shown on the left side of Figure
14. The upper trace shows the distillate composition re-
sponding to a step set-point change. Controller interaction
causes the set point response to be quite slow since both
controllers are working at cross purposes.
Decouplers are feed-forward elements where the mea-
sured disturbance is the controller output signal of another
loop on the process. Two decouplers are required to com-
pensate for loop interaction, one for each controller.71 Like a
feed-forward element, each decoupler requires identification
of a process and disturbance model. The right side of Figure
14 shows that with decouplers in place, this loop interaction
is dramatically reduced.
Students explore different controller modes, loop tunings,
model structures, and many other design issues. With two
controllers and four models for complete decoupling, stu-
dents also learn how important bookkeeping is to the

control designer.

Presented here are some examples of the lessons and chal-
lenges that a training simulator can provide. Space prohibits
presentation of other studies available in Control Station,
including the control of integrating processes, the use of the
Smith predictor controller that is a simplest form of a model
predictive controller, and a host of process identification
methods and procedures.
We stress that we do not believe a training simulator is
better than or a replacement for real lab experiences. In fact,
we believe that hands-on studies with actual equipment are
fundamental to the learning process.
We are of the opinion, however, that a training simulator
like Control Station can provide students with a broad range
of meaningful experiences in a safe and efficient fashion.
These experiences can be obtained risk free and at minimal
cost, enabling students to feel comfortable exploring non-
standard solutions at their desk. We believe if a training
simulator is properly designed, it can bridge the gap between
textbook and laboratory, enabling significantly enhanced
learning for process control theory and practice. If the read-
ers would like to learn more, they are encouraged to contact
Doug Cooper at
or visit

1. VanDoren, V.J., "Simulation Simplifies 'What-If Analysis,"
Control Eng., 45, 68 (1998)
2. Cooper, C.M., "DIST: A Computer Supplement Used in the
Teaching of a Graduate Distillation Course," Proc. 1991
Annual Conf. American Society for Engineering Education,
2, 1449 (1991)
3. Rabins, M.J., and C.E. Harriss, Jr., "Controls, Risk and
Educational Responsibility: The Ethical/Professional Links,"
Control Eng. Prac., 3, 711 (1995)
4. Marlin, T.E., "Software Laboratory for Undergraduate Pro-
cess Control Education," Computers & Chem. Eng., 20, S1371
5. Hanselman, D., and B. Littlefield, The Student Edition of
Matlab, Prentice-Hall, Englewood Cliffs, NJ (1995)
6. Seborg, D.E., T.F. Edgar, and D.A. Mellichamp, Process
Dynamics and Control, Wiley, New York, NY (1989)
7. Smith, C.A., and A.B. Corripio, Principles and Practice of
Automated Process Control, Wiley, New York, NY (1985)
8. Luyben, M.L., and W.L. Luyben, Essentials of Process Con-
trol, McGraw-Hill, New York, NY (1997)
9. Ogunnaike, B.A., and W.H. Ray, Process Dynamics, Model-
ing, and Control, Oxford, New York, NY (1994)
10. McCune, L.C., and P.W. Gallier, "Digital Simulation: A Tool
for Analysis and Design of Distillation Controls," ISA Trans-
actions, 12, 193 (1973) 0

Summer 2000

= classroom



Loughborough University Loughborough, LE11 3TU, United Kingdom

he ASTutE (Automated Student Tutorial Environ-
ment) project"l'2 has created a computer-based tuto-
rial resource that has been successfully used to teach
material balancing. This ASTutE software works alongside
tutors to help meet the challenge of teaching students with
increasingly diverse backgrounds while student-to-staff ra-
tios continue to rise at the same time. The project is managed
and staffed by the faculty of the Engineering Teaching &
Learning Support Centre[31 at Loughborough University,'4]
which is one of the leading institutes in the United Kingdom
that encourages implementation of new technologies in teach-
ing. ASTutE is being developed in conjunction with the
engineering and science faculties at Loughborough Univer-
sity, with initial trials in the chemical engineering, math-
ematical sciences, and mechanical engineering departments.
Students use ASTutE interactively to check their answers
to problems or as a detailed help system, but access only as
much help as they require. Chemical engineering problems
are often complex and best solved by following a prescribed
method; ASTutE uses a variety of interactive information
displays and question types to break such problems down
into stages, guiding students to the solution. This approach
encourages students to develop a methodological approach
to problem solving. There are usually many ways to solve
chemical engineering problems, and ASTutE provides feed-
back tailored to the responses of an individual student fol-
lowing one of many possible routes through a tutorial prob-
lem, consistent with their understanding of the topic. ASTutE
solves simple, recurring student misunderstandings and dif-
ficulties, freeing academic tutors to deal with the more com-
plex issues arising from the problems. A key feature of
ASTutE is the ease with which tutors can quickly write new,
or modify existing, problems by editing a simple template.
This paper describes the use of ASTutE at Loughborough
to teach first-year chemical engineering students material
balancing. This computer-aided learning (CAL) approach

proved highly satisfactory for both staff and students and
has a great potential for enhancing student learning.

Chemical engineering, unlike many subjects in United
Kingdom (UK) universities, has not been subject to an in-
crease in student numbers. There has been an increase in the
administrative load, however, and there is intense pressure
for the professors to pursue and publish research. These two
factors have reduced the available time for teaching, and
coupled with increasingly diverse student backgrounds has
left those professors who use traditional teaching methods
struggling to cope. In order to maintain quality, which is
now subject to rigorous external assessment, professors have
needed to develop new teaching strategies.
Tutorials have been particularly affected by these pres-
sures-most UK chemical engineering departments can no

David W. Edwards is a Senior Lecturer in the Department of Chemi-
cal Engineering at Loughborough University. He teaches material
and energy balances, process economics, design, and optimization.
His research interests are in feasibility assessment, safety and loss
prevention, and focus on inherent safety, environmental protection,
and cost estimation.
Fiona M. Lamb is Centre Manager of the Faculty of Engineering
Teaching and Learning Support Centre at Loughborough University.
The Centre supports staff wishing to integrate innovative learning
technologies into the curriculum.
Vian S. Ahmed has a BEng in Civil Engineering from the University
of Hertfordshire and a MSc in Construction from Loughborough Uni-
versity. She is currently preparing her doctoral thesis on "Adopting
Strategic Approaches for the Development, Implementation, and
Evaluation of Computer-Aided Learning Tools in Construction."
Steve J. Rothberg is a Senior Lecturer in the Department of Me-
chanical Engineering at Loughborough University, teaching and re-
searching in the areas of engineering mechanics, noise, and vibra-
tion. His active interest in CAL has included supervision of the ASTutE
project and a key role in the creation of Loughborough's Engineering
Teaching and Learning Support Centre.

Copyright ChE Division of ASEE 2000

Chemical Engineering Education

longer offer small-group tutorials. Yet tutorial w
always been an essential component in the consolid
students' understanding of lecture material, particu
subjects that are new to them. Traditionally, sheets
lems were handed out and then small groups of
discussed and solved any difficulties with the pr
guided by an academic tutor. The material balances i
described in this paper, was once taught entirely by
tutorial problems. But because of the increase in group
"discussing and solving difficulties" on an individu
has become more and more difficult; students are le:
vated to attempt problems, and tutorials degener;
problems classes where the tutor solves the problem
little student interaction. This is unsatisfactory fc
students of lower ability and those with nontradition
grounds-those who might hold vocational qualifi
or have graduated from a conversion course from
engineering, or have returned to academe after a lon
outside of full-time education. These students requi
more individual help than was available in the past, at
the critical early stages of chemical engineering educa
The first author thought that CAL might offer a
enhancing his teaching and overcoming these diff
He collaborated with the ASTutE developers to proc
material balances tutorial resource, with the aim of
these tutorials more effective for all involved.
To achieve this aim, several criteria for the tutc
source were identified:
It must be integrated into the curriculum.
It should be incorporated into existing
courses without major modification,
retaining the use of existing problem
Tutors should be able to quickly and easily
write or modify problems without recourse
to special software and with a minimal
"learning curve."
It should be available at all times for use
both in and outside of time-tabled ses-
The student interface must mirror the
pencil-and-paper approach as much as
Students mustfind it easy to use and be
able to work at their own pace, taking the
learning route they are best suited to.
It should solve the majority of difficulties
for the majority of students, freeing tutors
to deal with more complex issues.
It was found that these criteria could be best
met by programming a bespoke system in the

multimedia authoring language, Authorware.]51

Af prob- ASTutE encourages students to systematically tackle a
students problem in much the same way as they would on paper. This
oblems, generally consists of reading the text and then working
nodule, through the following steps: 1) create a diagram, 2) discover
solving the problem data and add it to the diagram, 3) set a basis for
ip sizes, calculations, 4) formulate equations to use, and 5) solve the
al basis equations. ASTutE replicates this method by displaying the
ss moti- problem text and then offering help, which is provided in
ate into five key stages corresponding to the steps above.
ms with ASTutE was designed to be used in different ways accord-
r those ing to student needs and abilities. Three scenarios are de-
al back- scribed below to show this flexibility.
arts to Scenario 1 The initial screen of an ASTutE problem,
g period shown in Figure 1, presents students with an answer box in
re even which to enter their answers. In this simplest scenario,
least in ASTutE is used as an answer-checking system. Strong stu-
tion. dents who have already successfully completed a problem
may log on simply to check their answer or perhaps to look
way of through the model solution before moving on to the next
iculties. problem. This is the quickest route through the problems
luce the written in ASTutE.
Scenario 2 When students experience difficulty at a
certain point in a problem or are uncertain about their chosen
ial re- method, pressing the "I'm stuck" button on the initial screen
brings up a help menu of key stages, shown in Figure 2.
Students choose any key stages) of help they require-for
example, choosing a basis. ASTutE provides specific help,

T2Q1: Ina ditsay.2Uk, lnabr ofia .tamBnF,
Stem I) coliah 0.04 mletfaclam ao alcmib mind
wfh3 aonomlbac.07 inoltaccil ioholairemn
VP b S ktm 2 2)
a) Whati thdowa oftapoduct mPa, Stm) ?
b)w Vrtl t al momlemi 01 onath produ ct?

)'B. Elk. ynrauaw inv~~hrirSI.4e ,IaTn
Jf-gu Ir fubjaV*~r*'i~u ~ -?

Figure 1. Initial ASTutE screen displaying the answer box. The student
can either input answers directly or click on "I'm stuck!" to access the
detailed help system.

Summer 2000

Students use ASTutE interactively to check their answers to problems or as a detailed help system, but
access only as much help as they require. Chemical engineering problems are often complex and
best solved by following a prescribed method; ASTutE uses a variety of interactive
information displays and question types to break such problems down into
stages, guiding students to the solution.

solving the student's particular difficulty with-
out solving the whole problem.
Scenario 3 Weak students are led interac-
tively, step by step, through a problem by ac-
cessing the help system from start to finish.
They learn how to tackle a problem with logi-
cal methods, which will help them solve fu-
ture problems.
ASTutE is a tutorial system and not an assess-
ment system, and is therefore designed to be as
flexible as possible. For example, the student is
not required to create the diagram in order to be
able to gain help on setting a basis for calcula-
tions, because these stages are independent. But
the student will need to set the basis for calcula-
tions before solving the equations. Also, stu-
dents are always able to input the final answer
and can exit the help system at any point. This
encourages the students to work on paper and
try to solve the problems themselves, referring
to ASTutE only as needed.
The feedback given to a student at any point is
tailored to his or her individual responses. Stu-
dents typically make one of a limited number of
possible mistakes at any stage in a problem, and
so the necessary feedback options can be iden-
tified relatively easily by a tutor. If an
unpredicted but common difficulty emerges,
the tutor modifies the existing problem tem-
plate to incorporate it.
By catering to different student needs, strong
students are not held back and the weaker ones
are not too embarrassed to obtain the intensive
help they need. It would be too ambitious and
inefficient, however, to attempt to solve every
conceivable difficulty for every student, so the
tutor is always available to address the more
complex or unusual misunderstandings.

ASTutE supports an extensive range of basic
question types, including multiple choice/re-
sponse, text/number entry, hot spot, and drag-
and-drop. Some of the complex questions types
that have been created from these basic ques-
tions types are

Figure 2. The five key stages of a materials balance problem. A
student has partially completed the "Choose Basis" stage, which
then displays some summary information.

T3Q4: Inapra-_ce sm -M- -\ilk"
bflronvarafltinlhct:tcawtertr.IMOLR W ar,
fresh mnoment lannleaPtibte~q thoaetlr. bdfymsw -.
removed asm prodtoto tsiaW ar. -
Whitt eqdenc( *mngtaste*aS omer
feediftho over' convoenroflmonmertopolyaneris _
What s itifhe overaD omwvason is 97?

OK- Pows
,^ !

Set sp1he Flowsheet Drag conpoei pan to ft gpd to crMa sd
i(C rk Check whey. you hree bjmuheSj KBY
i Ce e you he uI FFFresh Feed 5 Sepatm Im
[11~i ri F? F1 (1 P1 IFPFwr.WiiodiPu PPumr
.- sWar RFWacWiR

F21 EN Elt --t o II RII

4P I


Figure 3. The "Set up the Flowsheet" stage of a materials balance
problem. A student is just about to complete one of the many possible
correct flowsheet configurations.

Chemical Engineering Education



Number-Unit Entry Students must provide suitable
units with a numeric answer. The Dynamic Teaching
Solutions project6'7] developed the implementation.
An example is shown in Figure 1.
Diagram Creation Drag-and-drop allows flow sheets to
be built up and labeled. This question type is illus-
trated in Figure 3.
Math Entry Complex mathematical expressions can be
entered in an easy and familiar way, as developed by
Multiple Response Although this is a very straightfor-
ward question type, the diagnosis is complex. ASTutE
allows the tutor to give specific feedback to any
multiple response combination offered by the student.

ASTutE runs as an executable file, reading in problem
data from text files. Tutors do not need to learn to use any
special software. The text files are created or modified
using any spreadsheet package. The tutor opens a prob-
lem template and follows the instructions in this file to
write a problem.

We have used ASTutE in the Department of Chemical
Engineering to provide additional tutorial assistance for ma-
terial-balances teaching. Problem sheets were handed out in
the normal way to a first-year class of 55 students who

attended weekly 1.5-hour tutorials in a computer lab with 40
networked PCs running ASTutE and other CAL software.
The first set of simple problems were presented in the first
two weeks, using proprietary CAL software called Question
Mark Designer."9' Then, for the rest of the semester of 11
weeks, ASTutE was used. During the sessions, tutors circu-
lated around the room and were available to help with the
more intractable and/or uncommon problems that ASTutE
could not solve. The tutors would also bring interesting
issues that arose to the attention of the whole class.
Although we recommend Elementary Principles of Chemi-
cal Processes[110 as a text that students might like to consult
as part of their learning process, ASTutE is self-contained
and not tied to any particular book.
We believe that drawing and labeling diagrams is the key
to understanding and then solving material balance prob-
lems, so these are the first two of the five help key stages
shown in Figure 2 -"Set up the Flowsheet" and "Label

Set up the Flowsheet (Figure 3) Students create
flowsheets by selecting individual elements from a library in
the bottom half of the screen and placing them anywhere on
the grid in the top right-hand corner. They receive instant
feedback as components are placed-for example, if flow is
discontinuous. Elements can be repositioned or removed at
any stage. When students are satisfied with their flowsheet,
they click the "Check" button and more feedback is given for
flowsheets that have the wrong process units or are topologi-
cally incorrect. The tutor can provide specific feed-
back for specific errors.
When a student has constructed one of the
ieid,. (usually many) topologically equivalent correct
Sflowsheets, ASTutE presents its standard ver-
7' sion, which will be used from this point on. An
Example of a standard version is shown in Fig-
ure 4. Students must satisfy themselves that the
standard diagram is equivalent to theirs. This
S encourages understanding of flowsheets and
AM(n 1 good drafting practice.
Label Flowsheet Students first identify the
chemical components in each stream of the stan-
dard flowsheet. Then they indicate where data
specified in the problem text (which might refer
to a stream, a unit, or the whole process) should
be labeled on this flowsheet. On successful
completion, a new annotated standard diagram
is provided. In the example shown in Figure 4,
the three pieces of data are: the input flow of "9"
problem. (gal/min), labeled above stream 1; the "21" (gal/
using the min) flow through the pump; and the "3 equal
am. Each demands" legend above the three splitters.
Choose Basis First, students decide whether

Summer 2000

Figure 4. The "Unit Balances" stage of a materials balance
Students can write as many equations as they feel necessary
stream numbering system displayed on the annotated diagr
equation is checked for validity against the chosen basis.

they will solve the problem using mass, moles, or volume
flows. ASTutE gives appropriate feedback for each choice.
Then students must select a stream, a componentss, and a
flow value to use as the basis for their calculations. ASTutE
comments on the desirability of their specified basis. All feed-
back has been written by the tutor in the problem text file.

for the tutor to present solutions. We feel that this is the most
important benefit of using CAL-the students attempt the
problems. It is unclear if this is because they like using
computers, or because CAL presentation is more interesting
than problem sheets, or they dislike writing attempts at solu-
tions that might be seen to be incorrect in class, or the help

Unit Balances Students formu-
late balance equations, over
flowsheet units or any part of the
flowsheet, in terms of component
or total flows. First, the site of the
balance is specified-for example,
over the entire flowsheet ("overall")
or a unit ("mixer," "splitter," etc.).
Then, as shown in Figure 4, the
balance is constructed from the rel-
evant in-flows and out-flows. In the
equations, Water3, for example, rep-
resents the flow of water in stream
3. Each equation is checked for va-
lidity and against the chosen basis
by ASTutE. Students may enter an
unlimited number of valid equations
and they are all added to the "Valid
Balances" list.

Student Scoring o

Interaction with courseware

Solve Units Students see a grid Navigating through ASTutE
with rows for streams and columns Man g te
Manipulating the interface (
corresponding to components and
Presentation of diagrams an
with their selected basis value in
the appropriate cell. They must fill Presentation colors (uncoml
in the component flows in each Clarity of information prese
stream in order to complete the ma-
terial balance. ASTutE checks that TA]
all flows entered in the grid corre- Student
Interest and Confid
spond to the chosen basis. At any Interest and nf
point, students may answer the prob- Score 1
lem or display a summary of help
Interest 0% 0
stages completed so far. When the
Confidence 2% 5
student either solves the problem or onfdence 2%
"Gives up," a model answer con-
sisting of text and diagrams written by the tutor is displayed.


Informal Observations
Student attendance at the CAL sessions was better than
that at the conventional tutorials held in previous years, and
attendance did not deteriorate through the semester. Most
students said that they liked the sessions. A conventional
tutorial that was offered each week as an alternative to the
CAL sessions was usually attended by only a few, if any,
students. The students were much keener to attempt the
problems rather than waiting, as they had done previously,

if ASTutE's Usability
of 1 to 5)

(difficult to easy) 4
(difficult to easy) 3
difficult to easy) 4
d pictures (poor to good) 4
portable to comfortable) 4
nted (vague to clear) 4

t Scores for
ence in Subject Matter

2 3 4 5
)% 70% 25% 5%
% 49% 42% 2%

Student Performance on
Materials Balancing Coursework

Statistic Year
1997-98 1998-99
Number of students 57 53
Number that Achieved Full Marks 13 12
Number that Achieved Greater than 70% 48 47
Average Mark, % 87 84
Minimum Mark, % 32 56

discover its meaning, and translate it into a diagram.

Assessment Feedback
We switched from conventional problem classes to prob-
lems presented in CAL sessions in one semester. We took
the view that freshman students would not know there was
an alternative approach, so a clean break with tradition would
be possible. In order to check our impression that the stu-
dents were learning as well as they had in previous years, we
gave them a conventional paper-based test after four weeks
of CAL-based material balancing teaching. The test con-
sisted of two difficult questions to be completed in a short

Chemical Engineering Education

system is immediately accessible, or
computer interaction is less threat-
ening, or for some other reason.
An unexpected benefit of the CAL
approach was that many of the stu-
dents naturally formed into working
groups. They helped each other
learn, using the computer as a shared
The software met all the develop-
ment criteria-in particular, the most
important factor of handling frequent
or simple questions and thus freeing
tutors to give individual help and to
answer difficult and peculiar ques-
tions. The attitude of the staff was
therefore extremely positive.
We were surprised that the CAL
software was rarely used outside of
the timetabled sessions. This may
be because the students have a lot to
do already and have no time for ex-
tra work, or that access to comput-
ers was difficult, or that they ob-
tained the full benefit in the
timetabled sessions.
A common student complaint, ex-
pressed many times over the years
of teaching material balancing, is "I
cannot start difficult problems." Stu-
dents still had this complaint.
ASTutE development will further
address this issue by helping stu-
dents deconstruct the problem text,

time. Most students complained that they were not given
enough time, but most of them got good marks, and some
turned in excellent papers.
The students attempted and handed in written coursework
in the eighth week. This was an old exam question that had
been personalized for each student by changing the data in
the question. Table 1 summarizes the results for the trial year
(98/99) and compares them to those from the previous year
when CAL was not used. Both the proportion of the class
achieving full marks and the proportion achieving first-class
performance were almost identical for the two years. The
average mark was down slightly in 98/99, but the lowest
mark was significantly up, whereas there were four students
in 97/98 who scored less than the lowest 98/99 mark. This
indicates that the CAL approach particularly helped weaker
students. We were most encouraged by these results.

Student Session Questionnaire
A class of 43 students was surveyed by questionnaire
during and after using ASTutE. The survey data showed that
most of the students were confident in using computers, and
80% of the students had used CAL courseware before;
ASTutE was the first CAL experience for the rest of the
The students gave scores (lowest, 1, to highest, 5) to
different usability aspects of ASTutE. Table 2 presents the
distribution of the scores; all aspects of its utility were at
least in the satisfactory range.
Students scored ASTutE's effect on developing their in-
terest and confidence in the subject matter. Table 3 shows
that ASTutE increased interest and confidence in the subject
for 30% and 44% of the students, respectively, while 7% of
the them found it reduced their confidence in the subject-
these were students who had very little or no confidence in
using the computer.
The students' opinion about how ASTutE would be most
useful was as
a) Revision materialfor self-access in your own time and space
b) Additional tutorial resource, used with the tutor present
c) A lab session when you are first introduced to the subject
The percentages for multiple selection were: (b)(c), 8%;
(a)(c), 6%; (a)(b), 2%; (a)(b)(c), 2%.

Both informal observation and questionnaire data indicate
that the first group of students exposed to ASTutE were
happy to use it as a tool in learning material balances. It was
interesting to use and thus encouraged active engagement
and a deeper approach to learning.

Staff found the CAL sessions more satisfying than the
previous conventional problem classes because they did not
have to "replay" their solutions to an unresponsive audience
a number of times and could focus on the more interesting
and demanding questions posed by the students. While set-
ting up the CAL materials is extremely time-consuming, it
only needs to be done once and requires minimal ongoing
maintenance work. It is not yet possible to judge if this effort
is cost-effective because at this stage it was not practical to
split the time spent on software development from that on
problem entry. Although CAL increases choice and diver-
sity in the method of learning, it decreases flexibility in
content. The possible questions will always be constrained
by the capabilities of the software, whereas the combination
of the English language, numbers, and freehand sketches are
practically unconstrained for setting questions and describ-
ing solutions.
Although we have only limited data so far, the learning
outcomes as measured by formal coursework, tests, and
examinations, all indicate that the students learned at least as
much using CAL as they did when taught by conventional
methods. The CAL approach seems to be particularly help-
ful for weaker students.
Overall, ASTutE has proved to be a useful tool for teach-
ing material balances-one that we will continue to use and
enhance. Most importantly, the students attempted the prob-
lems presented in ASTutE, and when all is said and done,
engineering is a subject best learned by doing.
ASTutE is still under development, but it could in prin-
ciple be made available to other institutions. ASTutE runs
under Windows 95/98 or NT. Implementation of the soft-
ware in these environments should be simple. Extra tutor
effort would be in proportion to the number and complexity
of new problems. More information can be obtained by
contacting the author at .

1. Austin, F., and S. Rothberg, "The ASTutE Tutorial Assis-
tant: Efficient, Accessible, and Interactive," Global J. Eng.
Ed., 2(3), 271 (1998)
2. Austin, Fiona, and S.J. Rothberg, "ASTutE: Automated Stu-
dent Tutorial Environment," CIT Eng., 12, Winter (1997/
6. Stone, B.J., and N. Scott, Innovation in Teaching Engineer-
ing Dynamics: UWA-CPCS,
7. Dynamic Teaching Solutions Home Page: http://
8. Beilby, M., "Making a Feature of Assessment," Maths&Stats,
2(5), 6 (1994)
9. Question Mark:
10. Felder, R.M., and R.W. Rousseau, Elementary Principles of
Chemical Processes, 3rd ed., John Wiley & Sons (2000) O

Summer 2000

r laboratory



University Kebangsaan Malaysia 43600 UKM Bangi, Selangor, Malaysia

Membrane technology is increasingly recognized as
an important separation process with a wide range
of applications in biotechnology, food process-
ing, water and wastewater treatment, and gas separations.
But at the undergraduate level, laboratory experiments in-
volving membrane technology have been lacking. This is
probably due to the fact that the preparations involved in
setting up an experiment using a membrane are quite diffi-
cult. For example, in reverse osmosis the applied pressure
needed is very high, whereas for ultrafiltration, the sample
preparation and analysis is quite laborious.
This paper will describe a simple experiment using a
nanofiltration (NF) membrane for determination of mass
transfer correlations at the feed side of the membrane cell.
Conceptually, this experiment will introduce the students to
the concept of concentration polarization, separation due to
charge of ions, and overall performance of an NF mem-
brane. Experimentally, the students will be able to determine
the mass-transfer correlations, which will involve dimen-
sionless numbers such as Reynolds, Sherwood, and
Schmidt-the "classical" chemical engineering numbers.

A nanofiltration membrane is a type of membrane that has
properties in between ultrafiltration membranes and reverse
osmosis membranes. It is usually identified as having a
negative charge and pore size of approximately 1 nm. The
charge and small pore size mean it can provide separation
based on the Donnan effect for charged molecules and ions
and sieving effect for neutral molecules. As such, it offers a
wide range of separation capabilities in many areas of inter-
One of the major problems with any membrane operation
is the occurrence of concentration polarization at the mem-
brane surface. The solutes that are rejected are held back in a
layer next to the membrane surface. This solute buildup is

called concentration polarization.
Observed rejection is defined as
Robs= (1)
where C, is the permeate concentration and Cb is the bulk
solution concentration. It is an experimentally obtained mea-
surement of the degree of rejection of the solute by the
membrane. Due to concentration polarization, however, this
definition of rejection is not accurate. The real concentration
at the membrane interface is higher than the bulk concentra-
tion. Thus, the real rejection is defined as
Rreal -- (2)
The problem now is in determining the value of C,, which is
the concentration at the membrane wall. Figure 1 shows a
schematic diagram of the interface between the bulk solution
and the membrane surface for a three-component system.
Concentration polarization close to the membrane surface is
assumed to occur within a boundary film layer of thickness,
8. For a system containing charged ions, a mass balance for
the film layer yields
dCi ziF _dyf
ji =-Di,_ d CiDid x +Ci, (3)
dx dx( d)
where Di,- is the bulk diffusivity of ion i in the solution.
The equation can be solved using the boundary conditions

A. Wahab Mohammad received his BSChE
from Lehigh University in 1989 and worked
with Professor Phillip C. Wankat for his master's
thesis, receiving his MSChE from Purdue Uni-
versity in 1991. He worked for several years in
Malaysia before obtaining his PhD from the
University of Wales Swansea, United King-
dom, in 1998. His research interests are in
S membranes and separation technology. He is
currently a lecturer in the Chemical and Pro-
cess Engineering Department, Universiti
Kebangsaan Malaysia.
Copyright ChE Division of ASEE 2000

Chemical Engineering Education

at x= -8, c = Ci,b
at x = 0, c = Ci,
For a binary salt, the cation and anion will move together
due to the requirement of electroneutrality. Equation (3) can
be solved for both ions and the flux expressed as

j+ = j_ = -Deff dC + C+Jv (4)
where Deff,_ is the effective diffusivity of the salt defined

Deff, D~,-(z z_) (5)
zD+,- -z D.
Using the boundary condition defined above, the wall con-
centration, C,, can be correlated to other measurable param-
eters as

n= (n fn (6)
k (Cb -Cp) 1-Robs
I Robs
where k is the mass-transfer coefficient in the polarized
boundary layer and is defined as

k -De(7)
This result is equally applicable to a system of neutral sol-
utes. The mass-transfer coefficient is often characterized by
a Sherwood number (N,,) correlation that is expressed as a
function of Reynolds number (NRe) and Schmidt number
(Nse). For a laminar flow (NRe<32000) in a stirred cell, the
correlation is given as[3'41

NSh = (NRe n (Nsc)033 (8)

kr or2
Nsh -= NRe -
Deff,= V

Ns -De


where r is the radius of the stirred cell, m is the angular
velocity of the stirrer in radians per second, v is the kine-
matic viscosity defined as rl/ p, where rl and p are the
viscosity and the density of the fluid, respectively.
Another method of obtaining k is by extrapolating Eq.
(6).15s When written in linear form, Eq. (6) becomes

film membrane
Component 1 bl
Component 3 Cb3 C.

Component 2 Cb2 Cp2=C p
-,- -- -- A x ---

Figure 1. Schematic of the film layer and membrane
for three-component system.

In I-Robs n/1-Rreal J v 10)
/ ={L + (10)
I Robs = I Rreal k
k was found to be a function of the stirring speed, and thus
from Eq. (8), k can be written as a function of o as
k = k' "n (11)

S r2 )n( v )0.33 Deff (12)
k'= ( -I (12)

Thus, a plot of n(l Robs / Robs) vs Jv / on will give a slope
of 1/k' and intercept of fn(l Rreal \ Rreal). The value of Rrea
obtained is the real rejection at infinite stirring speed. The
most suitable values for n and 0 have been determined by
other workers using extensive sets of data and were found to
be 0.567 and 0.23, respectively.'341 These values were used
in this work.

The experiment can be carried out either in a stirred cell or
a cross-flow cell. In this work, the experiments were carried
out using a stirred cell with an effective membrane area of
28.7 cm2. The setup of the whole experiment is shown in
Figure 2. A simpler setup involving only the stirred cell, the
magnetic stirrer, the nitrogen flask, and a pressure gauge is
also feasible. The pressure is varied from 100 to 500 kN m2.
The membrane used was NF-PES5, obtained from Hoechst.
The membrane has a pore radius of about 1.2 nm.[6' The
concentration of NaC1 used for the feed is 1 mM. The
diffusivity values for Na+(D. = 1.33 x 10-9m2s-) and
CI-(D = 2.03 x 10-9m2s-1) were obtained from published
The membrane should be soaked overnight in the pure
water solution. For the experiments, the following procedure
was used:

1. In each run, 180 ml of fresh solution was used as the feed
solution. The stirring speed was set initially at 20 rpm and the
solution was left for 2 to 3 minutes to equilibrate in the cell.
The operating pressure was started at 100 kN m-2.
2. After 20 grams (ml) of permeate was collected, the experi-
ment was stopped. Then the conductivities of the fresh


4 5
6 78

Figure 2. Schematic diagram of dead-end filtration sys-
tem. (1) nitrogen flask; (2) pressure transducer; (3) 250 mL
reservoir; (4) water jacket; (5) filtration cell; (6) magnetic
stirrer; (7) electronic balance; (8) personal computer

Summer 2000

solution, the retentate, and the permeate were recorded.
3. The conductivities of the feed solution, the retentate, and the
permeate were converted into concentration using the
conductivity calibration curve. The average bulk concentra-
tion in the feed was calculated as

C Cfeed +Cretentate
where Cree and Cret,t,,a are the concentrations in the cell
before and after the experiments, respectively. The observed
rejection can be determined from Eq. (1).
4. The experiment was then repeated at the same operating
pressure for the stirring speed of between 30, 50, 100, 200,
300, and 350 rpm.
5. The experiments were then repeated for operating pressures
of 300 and 500 kN m-2.
6. The weight of permeate collected against time was plotted
and the data analyzed using linear regression. The slope of
the plot represented the permeate flux in g/s or cm3/s.

For a loose membrane such as NS-PES5, the flux is rela-
tively large, and thus the experiments can be completed with
three hours. But some membranes can be very "tight," and
the fluxes are very low. Thus, careful consideration should
be given in choosing which membrane to use.

The experimental data were analyzed using the calculated
k' obtained from the slope of Eq. (10). The data were then
compared to the theoretical values obtained by using the
calculated values of k' from Eq. (12).
Figure 3a shows the observed rejection of NaCI at three
different pressures as a function of stirring speed for NF-
PES5. It can be seen that Rob changed considerably with
stirring speed. The large changes were caused by the con-
centration polarization effect.
The data were then used to obtain k' using Eq. (10). Figure
3b shows the plot of (n[(l -Robs)/ Robs] vs J / 0567. The
lines are very linear, the calculated values of k' obtained by
using Eqs. (10) and (12) are tabulated in Table 1 together
with the values for exp(Jv / k' w0o657). This value is essen-
tially the measure of the degree of concentration polariza-
tion. It can be seen that the error obtained when k' calculated
by using Eqs. (10) and (12) is less than 5%.
The k' values were then used to calculate the experimental
real rejection of NaCI. Figure 4 shows the real rejection as a
function of the stirring speed. The filled points are those
calculated using k' from Eq. (10), while the blank points are
those calculated theoretically using Eq. (12). It can be seen
that the effect of stirring speed has diminished, especially at
larger stirring speeds. This means that the concentration
polarization effect has been corrected. Agreement in the
calculated real rejection using k' from Eqs. (10) and (12) are
reasonably good, which means that the calculated k' from

Eq. (12) can be used to estimate the mass transfer coefficient
of the feed side.
An interesting observation is that the real rejection, R,,ai,
was found to be dependent on the pressure drop. For
nanofiltration membranes, this dependence is due to the fact
that the rejection mechanism is not only determined by its
intrinsic pore size, but also by the charge of the membrane
(through the Donnan effect).[',] Furthermore, the transmem-
brane pressures used in this work produced fluxes in the
range where the transport mechanisms of diffusion convec-
tion and electromigration are equally important, and thus the
real rejection varied as a function of the applied pressure.19]
This behavior is different from the "standard" ultrafiltration
or microfiltration membranes that show no dependence of

0 AP=100kN m-2
0 AP=300 kN m'2
30 A AP=500kN m2
o 0
S20 -

o a o
10- on

D (a)
0 ---
0 100 200 300 400
(o / rpm
0 AP=100 kN m2
3 AP=500 kN m-2

2 -
3 AP=500


0 3 6 9 12 15 18
(jo0 567 )X 106

Figure 3. Result for NF-PES5. a) Observed rejection as a
function of stirring speed; b) linearization of the data to
obtain k'.

Calculated Values for k'
and the Exponential of the Peclet Number
Pressure k' exp(J /k'co0.567)
Eq. (10) Eq. (12) Eq. (10) Eq. (12) % error
100 kN m 5.02 x 106 4.859 x 106 1.26 1.27 0.8
300 kN m2 5.10x 10- 4.859 x 106 1.98 2.05 3.4
500 kN m 5.02 x 106 4.859 x 106 2.58 2.66 3.0

Chemical Engineering Education

Rreal on transmembrane pressure at high fluxes where the
transport of solutes is only through convection.

In conclusion, the experiments described here are very
suitable for showing the concept of concentration polariza-
tion in membrane operations and how the "classical" mass
transfer correlations are used to describe it. The equipment
needed to run the experiments is quite easy to set up, and the
experiment can be completed within the three-hour period
normally reserved for the junior/senior laboratory.
The students should be able to make the following infer-
Concentration polarization is an important phenomenon in
membrane filtration.
A simple mass balance of the system allows one to deduce the
mass transfer coefficient, which relates the solvent flux to the
boundary concentrations.
The mass transfer coefficient, k, can be calculated through an
equation involving dimensionless numbers such as the
Sherwood number, Reynolds number, and Schmidt number.
The dimensionless equation can be confirmed through a
simple experiment involving the measurement of rejection of a
binary salt.
For additional experiments, the correlations can be used to
determine the diffusivity (by reversing the calculation method
used here) value for another salt, such as KCl or LiCI (the
student should not be told what salt they are using). The
calculated diffusivity value can then be compared to the
published data. This will be a good check on whether or not
the students did the experiments correctly.






i i

i i


0 100 200 300 400
o / rpm
0 100 300 A 500

Figure 4. Real rejection as a function of stirring speed. The
darkened points show those calculated using k' from Eq.
(10), while the clear points are those calculated using Eq.
(12) The legend shows pressure drop in units of kN m2.

Summer 2000

Cb bulk concentration on the feed side of membrane (mol m3)
C local concentration on the feed side of membrane (mol m-')
C concentration in permeate (mol m3)
Cw wall concentration on the feed side of membrane (mol m3)
Di,, bulk diffusivity (m2s 1)
Deff,- effective bulk diffusivity (m2s-1)
F Faraday constant (C mol')
j, ion flux (based on membrane area)(mol m 'sl)
J, volume flux (based on membrane area)(m s')
k mass transfer constant (m s-')
k' mass transfer constant defined by Eq. (8)
n constant defined in Eq. (8)
AP applied pressure drop (kN m 2)
r radius of stirrer (m)
R gas constant (J mol-'K-')
Rrea real rejection
Rob, observed rejection
T absolute temperature (K)
x distance normal to membrane (m)
z valence of ion
6 thickness of film layer (m)
( constant defined in Eq. (8)
ri viscosity of solution (Pa s)
v kinematic viscosity (m's )
o) stirring speed (rad s ')
yf potential in bulk solution (V)
+ referring to cation
-referring to anion

1. Raman, L.P., M Cheryan, and N. Rajagopalan, "Consider
Nanofiltration for Membrane Separations," Chem. Eng.
Prog., 90, 68 (1994)
2. Krishna, R., and J.A. Wesselingh, "The Maxwell-Stefan Ap-
proach to Mass Transfer," Chem. Eng. Sci., 52, 861 (1997)
3. Opong, W.S., and A.L. Zydney, "Diffusive and Convective
Protein Transport Through Asymmetric Membranes," AIChE
J., 37,1497(1991)
4. Malone, D.M., and J.L. Anderson, "Diffusional Boundary-
Layer Resistance for Membranes with Low Porosity," AIChE
J., 23, 177 (1977)
5. Nakao, S., and S. Kimura, "Analysis of Solutes Rejection in
Ultrafiltration," J. Chem. Eng. Japan, 14, 32 (1981)
6. Bowen, W.R., A.W. Mohammad, and H. Hilal,
"Characterisation of Nanofiltration Membranes for Predic-
tive Purposes: Use of Salts, Uncharged Solutes, and Atomic
Force Microscopy," J. Membrane Sci., 126, 91 (1997)
7. Atkins, P.W., Physical Chemistry, 4th ed., Oxford Univer-
sity Press, Oxford, UK (1990)
8. Schaep, J., C. Vandecasteele, A.W. Mohammad, and W.R.
Bowen, "Analysis of the Salt Retention of Nanofiltration
Membranes Using the Donnan-Steric Partitioning Pore
Model," Sep. Sci. & Technol., 34(15), 3009 (1999)
9. Bowen, W.R., and A.W. Mohammad, "Characterisation and
Prediction of Nanofiltration Membrane Performance: A Gen-
eral Assessment," Transaction IChemE, 76A, 885, (1998) 0






In ChE Departments

Rowan University Glassboro, NJ 08028-1701

initially, the assessment requirements imposed by the
new ABET Engineering Criteria 2000'" appear daunt-
ing. Even the terminology is confusing. Compounding
the challenge is the fact that engineering faculty typically
lack experience in conducting outcomes assessment. Several
authors have made analogies between the outcomes process
of assessment and chemical process control loops.[2,3' Al-
though these may be useful analogies for defining the pur-
pose, they do not provide specific ideas on how to approach
such a large and ill-defined problem as program assessment.
No matter how hard we try, we cannot use Laplace trans-
forms and transfer functions to make our problems go away.
Instead, we must recognize that we will have to face these
new challenges head-on.
The University of North Dakota was slated to be a pilot
program for reaccreditation review under EC 2000 in the fall
of 1997. Unfortunately, the massive flooding of the nearby
Red River of the North in the spring of 1997 caused the
accreditation visit to be postponed for one year. Although
the flood was devastating to the city, the university, and
faculty homes, it did save us from going up for accreditation
prematurely. We were not ready! Like many programs that
had been reviewed under the previous system, we did not
realize how much lead time an organized and documentable
assessment plan would require.
We had spent time rewriting mission statements and ask-
ing ourselves how we could determine if our students were
really learning. Like most programs do, we saved every-
thing: tests, final exams, lab reports, homework assignments,
journal entries, etc. But we had no real plan as to what we
should do with them. With the extra time afforded us by the
flood, we began a series of discussions, planning sessions,
* Address: University of North Dakota, Grand Forks, ND 58202


and activities that helped us, finally, to address the pivotal
issues. We were able to involve our constituencies by in-
cluding students directly in the writing and planning and by
meeting with our industrial advisory board. In the fall of
1998, the chemical engineering program at the University of
North Dakota was visited under EC 2000. This site visit was
the culmination of a two-year-long process (which really
should have been longer) of preparing and implementing an
assessment plan.
We wanted to write a paper that provided practical sugges-

James Newell is Associate Professor of Chemical Engineering at Rowan
University. He serves as a Director of the Chemical Engineering Division
of ASEE and has received the Dow Outstanding New Faculty Award. His
areas of interest include high-performance polymers, integrating commu-
nications across the curriculum, and undergraduate research.
Heidi Newell is currently the assessment/accreditation consultant for the
College of Engineering at Rowan University. She previously served as
assessment consultant for the University of North Dakota. She holds a
PhD in Educational Leadership from the University of North Dakota, an
MS in Industrial and Organizational Psychology from Clemson, and a BA
in Sociology from Bloomsburg University.
Tom Owens has been Chairman of the Chemical Engineering Depart-
ment at the University of North Dakota since 1975 (with two brief interrup-
tions to serve as acting Dean). He holds a PhD from Iowa State and is
actively involved in self-paced learning, distance learning, and technical
John Erjavec is Associate Professor of Chemical Engineering at the
University of North Dakota. He holds a PhD from Princeton University
and specializes in experimental design and statistical analysis of experi-
mental data. He is actively involved with the Energy and Environmental
Research Center.
Rashid Hasan is Professor of Chemical Engineering at the University of
North Dakota, where he has been since 1979. Originally from Bangladesh,
he holds a PhD in chemical engineering from the University of Waterloo,
and his research focuses on multiphase flow modeling.
Steven P.K. Sternberg is Assistant Professor of Chemical Engineering
at the University of North Dakota. His PhD is from Purdue and his
research focuses on flow through porous media and the use of native
plants to remove metal contaminants from water. He is actively involved
with novel teaching techniques and undergraduate research.
Copyright ChE Division of ASEE 2000

Chemical Engineering Education

tions for reassessment that may not appear in a manual. The
remainder of this paper is devoted to providing answers to
questions that we struggled with and to providing advice for
other departments.


How do we get started?
Schedule a relaxed meeting that does not occur during
normal school hours or take place in your usual, more stress-
ful surroundings. Use this meeting to discuss
the steps and develop a timeline. Much of the
accreditation and assessment preparation is e
sequential. Therefore, you will create your
timeline for activities by noting your ultimate r
deadline for submitting your self-study to Lik
ABET (e.g., June 1st prior to your accredita- progau
tion visit) and working your way backward to been
the present. The major phases are under tJ
Selecting and writing about: vision, mission, system,
goals, objectives, outcomes, indicators, realize
practices, assessment methods, and lead
assessment criteria
Discussing and writing the self-study report do
Designing, pilot testing, and administering
your assessment tools or collecting other aSSessI
data for assessment purposes would.
Collecting materialsfor the various We beg
appendices to the self-study report of dis
Analyzing collected data planning
Making changes to the educational experi- and
ence based on your findings that h
Assessing your improvements
We cannot provide a timeline for you since to ad
all programs are different. Remember that you plivot
will need at least one complete cycle before
your accreditation visit, so you must be done
planning at least one year before the visit.

Should we get help?
Yes. Although assessment is worth it, it does add to al-
ready overburdened faculty workloads. Therefore, we hired
an accreditation/assessment consultant who kept us on pace
and helped translate the assessment-speak into ideas that we
could understand. The consultant should not make decisions
for you, but rather should serve as a facilitator in your
efforts. One of the coauthors of this paper holds a PhD in
higher education administration and had several years of
assessment experience, so finding a consultant was easy for
us. But almost all universities have a potential consultant in
place. Some university personnel have been doing assess-
ment for years. If you cannot afford (or do not prefer) an
external consultant, try talking to individuals within your

institutional research department or your college of educa-
tion who may be able to recommend cost-effective assis-
tance. Remember, a consultant should be just that-a con-
sultant. He or she can help you greatly, but he or she cannot
plan or do the assessment for you.

What do all these terms mean?
We recommend that early in the process you develop a set
of common terms and definitions so that each of you will

were not
e many
ms that had
he previous
we did not
how much
time an
sized and
ment plan
an a series
ig sessions,
helped us,
dress the
Ul issues.

know what everyone else is talking about.
Regrettably, there is no consensus in the as-
sessment community. The important thing is
that you all use the same terms and define
them for the evaluators. We used the follow-
ing definitions:
Vision statements outline your mission of
the future.
Mission statements outline the purpose of
your program.
Goals are the lofty aims. Things such as
"We want our graduates to be effective com-
municators" are goals. You may wish to in-
clude university and college goals with your
program goals.
Objectives are more specific. Perhaps
things such as "When giving an oral presen-
tation, our students will a) provide an intro-
duction appropriate for a given audience, b)
speak clearly, c) present facts in a logical
manner d) support their arguments with facts
and data, and e) clearly summarize key points.
Outcomes tell us what specific results)
will occur, such as "Students will write effec-
tive documents."

Indicators are the specific items to which a "yes" or "no"
answer to the outcomes questions can be applied, such as "Is
the document formatted correctly?"
Practices are opportunities in your educational experi-
ence for student learning, such as a class or an activity.
Assessment Methods are the actual tools or other data-
collection techniques you use to assess student learning,
such as portfolios, alumni surveys, the Fundamentals of
Engineering Exam, etc.
Assessment Criteria are the stated levels of performance
for each assessment method that will be used to guide deci-
sions and set priorities for improvement. You will want to
develop those ideas that are unique to your program and
highlight your strengths in addition to ideas required by
outside bodies.

Summer 2000

How can we make sure all of us are
addressing criteria that we need to
Use various matrices to give you
visual pictures of how your outcomes
map to your curriculum and also to
your assessment methods. Two sample
matrices are shown in Tables 1 and 2.
You should try to make sure there are
at least three "hits" for every item in
the rows and columns for each matrix.
On the other hand, if there are too
many "hits" in a row or column, you
may be able to eliminate some in fa-
vor of addressing other desired areas.

How much data
should we accumulate?
If you save every exam and home-
work assignment, you will be buried
with so much data that you will be
unable to figure out what is meaning-
ful. By planning carefully in mapping
instruments to your objectives, you can
reduce the data collection consider-
ably. Remember, assessment is not just
"do it once and forget about it." Sam-
pling is the key concept in data collec-
tion. In general, you should gather the
least amount of data that will give you
the most information. In other words,
some assessment methods may require
input from all sources, other may only
require strategically selected samples.
Whenever possible, use or modify ex-
isting data collection opportunities to
reduce the burden of data collection.
For example, your university might
already be collecting information you
need. You will want to do a project
cost analysis (i.e., in terms of materi-
als and time) in conjunction with data
collection and, in reality, this may im-
pact how much data you can feasibly

How do we keep track of things?
First, set up a data warehouse. You
might want to include the following
electronic folders for each program:
self-study, syllabi, curriculum vita,
tables, policies, references, (assess-

Senior Design Reports
Senior Design Orals
Unit Ops Lab Reports
Clinic Reports
Clinic Presentations
FE Exam
Alumni Surveys
Employer/Recruit Surveys
Exit Interviews
Peer Reviews

List of Practices Mapped to Educational Objectives

A = knowledge of math, science, and engineering
B = design and conduct experiments; analyze and interpret data
C = design a system, process, or component
D = multidisciplinary teams
E = identify, formulate, and solve engineering problems
F = ethics
G = communicate effectively
H = broad education to see societal impact
I = lifelong learning
J = knowledge of contemporary issues
K = use modern tools
L = working knowledge of chemistry
M = working knowledge of ChE principles
N = department-specific objective
O = department-specific objective

ChE Courses
Senior Design
Unit Ops Lab
Oral Presentations
General Eds
AIChE, SWE, etc.

_____II iiIi iIIIii 1 IIIll

Chemical Engineering Education

List of Assessment Methods Mapped to Objectives
A = knowledge of math, science, and engineering
B = design and conduct experiments; analyze and interpret data
C = design a system, process, or component
D = multidisciplinary teams
E = identify, formulate, and solve engineering problems
F = ethics
G = communicate effectively
H = broad education to see societal impact
I = lifelong learning
J = knowledge of contemporary issues
K = use modem tools
L = working knowledge of chemistry
M = working knowledge of ChE principles
N = department-specific objective
O = department-specific objective


ment) tools, data, and reports. Second, set up a data storage
system for paper copies. Third, develop a timeline for the
fall, spring, and summer semesters, with suggestions of what
should occur early in the semester, in the middle, or at its
end. Obviously, assessment activities should be distributed
in such a way that allows for moderate activity throughout
each semester instead of periods with too much or too little
collection. Finally, make each faculty member responsible
for at least one assessment activity. This way, all instru-
ments will be used without any one faculty member being
overburdened. Each coordinator should maintain a set of
written responsibilities as a reference to facilitate adminis-
tering tools.

How do we plan improvements?
There must be a formalized system in place. We recom-
mend having a retreat during the summer each year, which
we called the "assessment marathon." Over two days, we
discussed all aspects of our program, including the data from
each tool, in turn. We identified strengths as well as areas for
improvement and made decisions affecting our curriculum
and policies. These discussions were wonderfully produc-
tive, and we left with a better feel for the program as a

How involved should we be with other departments?
We were not nearly as involved with the other depart-
ments in the college as we should have been. Although we
shared information, an alumni survey was the only common
instrument used. A significant aspect of the assessment pro-
cess is discussions about improvement, and all departments
can benefit from each other's experiences. Most university-
level information is useful in assessing the general educa-
tional experiences that are likely to be common across the
programs. It is important, however, to remember that ac-
creditation occurs at the departmental level, so each depart-
ment is ultimately responsible for itself.

Why won't anyone provide specific
answers instead ofjust general advice?
Outcomes assessment is a highly personal activity. The
whole point of moving away from bean counting and into
outcomes was to enable programs to set their own goals,
defend their importance, and prove that they are being
achieved. Even a reviewer of an early draft of this paper
asked "What should we collect-finals exams but not home-
work, materials from every student or every tenth student?"
There are no single answers to these kinds of questions. A
program that graduates 15 students a year will keep different
information than one that graduates 100 or more. If final
exams are one of five assessment instruments you are using
to demonstrate that you have achieved an objective, you may

not need homework as well. You own the process and must
make your own decisions. Below is one education objective,
as an example-but your department may have very differ-
ent ideas.

Goal-Develop students who communicate their ideas effec-
tively in various formats to both technical and non-tech-
nical audiences.
Objective The Chemical Engineering Program at Rowan
University will produce graduates who demonstrate ef-
fective oral and written communication skills (ABET-G).
Outcome A Students in the Chemical Engineering Program
will write effective documents, including memos, e-mails,
business letters, technical reports, operations manuals,
and descriptions of systems, processes, or components.
1. Written at the appropriate levelfor the intended reader
2. Presents correct technical information
3. Contains few, if any, typographical or grammatical er-
4. Formatted correctly
5. Contains an introduction that interests and orients a reader
6. Contains a body that is relevant and covers important
7. Contains a conclusion with summary and recommenda-
tions, when appropriate
1. Chemical engineering courses
2. Unit operations lab
3. Internships
4. Senior plant design
Assessment Instruments
1. Senior plant design reports
2. Portfolios
3. Alumni surveys
4. Recruiter/employer surveys
5. Exit interviews
6. Peer reviews

The process of developing a workable assessment plan
that is useful in preparing for accreditation under EC 2000 is
long and filled with challenges. Departments must begin to
analyze their program goals early and recognize the size of
the task they face. Through progressive discussions and a
systematic approach to planning, the task can be accom-
plished. Key points to remember include: identify your goals
first, involve students and other constituents, minimize the
data that you are required to collect and analyze, have mul-
tiple indicators for each objective (ideally involving mul-
tiple sources), and get started yesterday!

1. ABET Engineering Criteria 2000
2. Felder, R., Chem. Eng. Ed., 32(2), 126 (1998)
3. Shaeiwitz, J., Chem. Eng. Ed., 32(2), 128 (1998) 0

Summer 2000

eff curriculum


That Helps Meet EC 2000 Objectives

University of Toledo Toledo, OH 43606

oday's engineer must be more aware of the environ-
mental implications of technology than ever before.
The last ten years have heralded a paradigm shift in
the way engineers view environmental propriety; waste treat-
ment is no longer the most acceptable means of dealing with
process wastes. The Pollution Prevention Act of 1990 states
that the option of first choice is to prevent formation of
waste at the source.1' Since it is unlikely that people will
give up the products that improve the quality of their lives, it
is imperative that engineers learn how to evaluate and mini-
mize its environmental impact during the concept and design
stage of a product or process.
Engineers are always faced with the challenge of imple-
menting new technologies at minimum cost. With increased
awareness of the impact of technology on the environment,
engineers must also deal with the added constraint of opti-
mizing processes for minimum environmental impact. But,
today's engineering students get little, if any, training on
analyzing a manufacturing process for its environmental
impact. A required course on pollution prevention (P2) can
remedy this shortcoming in the engineering curriculum.
In many respects, the P2 course serves as a second design
course--one with the objective of minimizing the environ-
mental impact of the process. This clearly complements the
standard chemical engineering design course in which the
objective is to minimize the cost of a process. Thus, the P2
course supplements the curriculum by providing an alterna-

Martin Abraham is Professor of Chemical and
Environmental Engineering and Director of the
Environmental Reaction Engineering Laboratory
at The Unviersity of Toledo. He received his BS
from Rensselaer Polytechnic Institute in 1982 P
and his PhD from the University of Delaware in
1987. His teaching and research interests are in
chemical reaction engineering applied to envi-
ronmental issues, reactions in benign solvents,
and the application of pollution prevention in
chemical engineering.

Copyright ChE Division of ASEE 2000

tive context in which to view process design.
The ABET EC 2000 guidelines dictate several new chal-
lenges for engineering curricula, including the use of inter-
disciplinary teams, an awareness of the impact of technol-
ogy on the world, and an appreciation for lifelong learning.
We believe that all three of these items represent an essential
element of environmental analysis. Thus, the pollution pre-
vention requirement is an ideal environment in which to
stress these elements of chemical engineering education.

Recent chemical engineering literature has contained nu-
merous references to the use of pollution prevention meth-
ods in industrial practice. Among the most frequent indus-
trial proponents of pollution prevention have been Jim Dyer
and Ken Mulholland of DuPont, the authors of a newly
published book on pollution prevention.'2] Among their ex-
amples, they use the DuPont Chamber Works facility to
illustrate that chemical engineers are taught the necessary
tools to propose and test modifications in plant design that
will lead to cleaner manufacturing processes. Old (and envi-
ronmentally negligent) methods become entrenched within
the system, however, making industrial leaders resistant to
new technologies. Certainly, new engineers must be trained
to seek the best environmental solution along with the usual
economic design solution.[4]
In order to make engineers more aware of environmental
problems, universities have recently begun instituting courses
in waste treatment or pollution prevention, or discussing the
issues of pollution prevention in their capstone design course.
Some of these efforts are beginning to appear in archival
publications. For example, Grant, et al.,'51 reported on a
graduate course in pollution prevention that emphasizes waste
audits and life-cycle assessment. More recently, El-Halwagi
and Spriggs161 described their use of a process simulator and
mass integration as a basis for pollution prevention for se-
nior/graduate chemical engineering students.

Chemical Engineering Education

In order to gain more information about teaching pollution
prevention, we conducted a survey of chemical engineering
departments throughout the United States. Thirty-five pro-
grams provided information about their course offerings in
pollution prevention. Including our course at the University
of Toledo, we identified eleven programs that offer a stand-
alone course in pollution prevention. One department teaches
a course in industrial ecology as part of an environmental
management program for engineers. Several additional de-
partments (14) provide elective courses in some form of
waste treatment and include pollution prevention as a com-
ponent of the course. Six respondents indicated that they had
no pollution prevention course, but taught the principles of
pollution prevention in other courses, usually the senior
design course. Only two respondents indicated that they
offer no material in pollution prevention-these programs
referred interested students to courses taught in other depart-
ments. To our knowledge, only our program required a pol-
lution prevention course for chemical engineering students.
Additional information on pollution prevention course
material is available through the University of Michigan
National Pollution Prevention Center (as listed on the web-
site at , the Michigan
State University Pollution Prevention Workshop at>, and the Canadian Centre for Pollu-
tion Prevention at .
Without exception, each course described on the returned
survey form is a one-term course (quarter or semester) and
provides 3 credit hours toward the major. Class size aver-

cently published textl"I may also serve as a valuable re-
source. In nearly all cases, supplementary material obtained
through NPPC, EPA, or other industrial and environmental
organizations, was used to augment the textbooks.
Because most of the courses use the Allen and Rosselot
text, it is not surprising that they follow a similar structure.
The typical course begins with an introduction to pollution
prevention concepts, moves on to a discussion of pollution
in a broad sense (often including an introduction to life cycle
analysis), and concludes with an extensive analysis of chemi-
cal processes and the development of process synthesis tools.
As an indication of the typical content, the course outline
used for our course in the fall of 1999 is given in Table 1.
Our survey also revealed that many chemical engineering
departments offer elective courses in waste treatment or
pollution control, and most of these contain a pollution pre-
vention component. In addition, several programs provide
pollution prevention education within the capstone design
courses. For these programs, pollution prevention accounts
for approximately two weeks of a fifteen-week semester.

We have recently implemented a required course, "Pollu-
tion Prevention," describing methods that can be used to
minimize the impact of chemical processes on the environ-
ment. The course contains three components: an introduc-
tion to environmental pollutants, a discussion of life cycle
analysis, and an environmental analysis of a chemical pro-
cess. As a component of the course, we included a case study
provided by a local chemical manufacturer.

ages between 15 and
20 students, and the
typical course is of-
fered once a year.
One goal of the sur-
vey was to identify the
source material used
in the development
and delivery of a pol-
lution prevention
course. The majority
of instructors listed
Allen and Rosselot[7'
as an assigned text for
this course. Other
books that were men-
tioned, either as a
required text or
as supplementary ma-
terial, included those
of Higgins, 8' El-
Hawagi,[9] and Graedl
and Allenby.[10' A re-

Example of Course Schedule (Fall, 1999)

Beginning Topic
Aug. 30 Pollution in context
Sept. 6 Pollution control: legal context
Sept. 13 Energy production from fossil fuel
Sept. 20 Alternate energy sources
Sept. 27 Life cycle analysis

Oct. 4
Oct. 11
Oct. 18
Oct 25
Nov 1.
Nov. 8
Nov. 15
Nov. 22
Nov. 29
Dec. 6

Life cycle analysis-the mechanics
Life cycle analysis case study-food packaging
Pollution prevention in unit operations
Environmental risk analysis and solvent selection
Risk modeling

Rossiter. Chapter 1
Chapter 2

Chapter 4

McDonalds/EDF case study141
Chapter 5
Allen, 2nd ed., Ch. 61"1

Emissions from unit operation Chapter 6. 1
Pollution prevent, in reaction eng.-green chemistry Chapter 6.4
Flowsheet analysis for pollution prevention Chapter 8.1

Rigorous methods: design for the environment
Environmental cost accounting

-3: Chapter 7

Chapter 8.2
Chapter 9

After analyzing
available resource ma-
terials, development of
the course followed the
outline in Table 1. It
was designed as a
stand-alone course on
environmental issues in
chemical engineering,
focusing on the meth-
ods that can be used to
minimize the environ-
mental impact of ChE
processes. The course
can be loosely divided
into three sections: 1)
defining the problem,
2) analyzing processes
and products through
their life cycle, and 3)
designing environmen-
tally responsible chem-
ical processes.

Summer 2000

Each of the class units is accompanied by a project that
must be completed by a team of students. Project descrip-
tions are purposely left vague so that the teams can proceed
in as creative a manner as possible. Each project requires
that students obtain background information, ensuring that
they will gain some experience in using reference materials.
Finally, a written report is required for each project, enhanc-
ing the students' writing skills.
What is Pollution?
As an initial class exercise, the students work together in
small groups to define a pollutant. They begin with very
different ideas about pollutants and generally present vague
notions of things that are either bad for the environment,
hazardous, or unsightly. Throughout the class, we hone the
definition until we arrive at an acceptable compromise, e.g.:
"An undesirable substance or disturbance that causes harm
to the ecosystem or serves no beneficial purpose."
Given this definition of pollution, we next move on to a
discussion of pollution prevention, focusing on the Pollution
Prevention Act of 1990. Specific examples are normally
presented at this point. During the current year, we are
focusing our discussion on energy production from coal
combustion. To illustrate the difference between pollution
prevention and pollution control, we consider the differ-
ences between flue gas desulfurization (in which SO2 is
captured for disposal) and coal gasification (in which coal is
converted to CO and H2, and sulfur is recovered). This topic
also provides an opportunity to discuss alternative energy,
sustainability and the use of biomass, and low-energy den-
sity systems such as wind power. Our discussion of biomass
utilization provides an introduction into life cycle analysis,
which whets the student's appetite for later material.
It is important that students gain an appreciation for the
range of laws and regulations that affect wastes evolving
from a chemical process. This course is not intended to
provide a definitive discussion of all the regulations, how-
ever; a cursory overview of only the most significant laws is
provided, with more detailed information included about the
Clean Air Act because of its relevance to energy production.
This also provides an entry for discussing a range of environ-
mental problems, from global warming (planetary, decades) to
acid rain (regional, years) to urban smog (local, days).
Life Cycle Assessment (LCA)
When evaluating a process or product, it is important to
consider its entire life cycle to adequately assess its impact
on the environment. For example, the development of an
electric car shifts the emissions from the tailpipe to the
power plant. If one only considers the vehicle, then one
neglects the emissions from the energy production needed to
charge the batteries and the impact of energy storage (e.g.,
lead acid batteries), thus incorrectly evaluating the real ef-
fect of this change on the environment.

The Allen and Rosselot textbook17] introduces LCA in
Chapter 4 and provides a few specific examples for illustra-
tive purposes. Numerous life-cycle studies have also been
completed within the past few years, and rigorous quantita-
tive methods for LCA have been developed. The Graedl and
Allenby texts"" '12 are an excellent resource on LCA. Addi-
tional examples are contained in a set of homework prob-
lems compiled by Allen, Bakshini, and Rosselot.'"3 In addi-
tion, the EPA provides life cycle assessments of several
products in the form of EPA reports, many of which can be
accessed through its web site. A particularly appropriate
study is one completed by McDonalds and the Environmen-
tal Defense Fund on the comparison of polystyrene clamshell
and paper containers for hamburger delivery."[4]
At the heart of the LCA is completion of a material-and-
energy-balance problem, information that is well known to
chemical engineering students. They have less experience in
defining the system boundaries or in allocating streams to
individual products, but this can quickly be addressed through
the material-and-energy-balance analogy. The students are
accustomed to the necessary information being included in
the problem statement, however, and are not aware of the
level of difficulty encountered in obtaining the data neces-
sary to complete an LCA or the problems associated with the
quality of data. This is best illustrated using published stud-
ies wherein large volumes of data are compiled and evalu-
ated (see the large list of information available through NPPC
or the EPA Office of Pollution Prevention and Toxics).
Another component of the LCA that is new to the students
is the assessment stage, where process effluents are com-
pared to evaluate the environmental impact of a particular
product. Although several techniques have been published
in the literature, we have chosen to address this issue using
the streamlined life cycle assessment (SLCA, described in
detail by Graedl'51). In this case, the raw material and energy
uses and the process emissions are evaluated and ranked on a
scale of 1 through 4. The total score of the product is then
calculated, with the highest score representing the more
environmentally friendly product. Using SLCA provides a
means of converting the data obtained by the students into a
quantitative measure of environmental friendliness.
In order to assess the student's ability to complete an
LCA, we asked them to evaluate two different products that
serve the same purpose. For example, students compared
paper cups to styrofoam cups, or disposable diapers to cloth
diapers, using the methodology described in class. Students
were grouped into teams of two, with the idea that each of
them could evaluate one product. They compiled the data
using resources available at the library or through the internet,
and then assigned a quantitative score to each of the compo-
nents in the SLCA matrix. They then completed the analysis
by identifying any important environmental issues that were
not captured through this ranking system. The students com-

Chemical Engineering Education

pleted the assignment by preparing a short technical report
describing the results of their analysis.
Environmental Analysis of Chemical Processes
Because our course is designed for chemical engineering
students, we put significant emphasis on the design of envi-
ronmentally friendly chemical processes. Two components
comprise this analysis: first, students must be able to evalu-
ate the emissions from a particular unit operation or to de-
sign the unit operation to minimize its environmental im-
pact; second, they must evaluate entire processes to determine
where new technologies or new flow patterns could be imple-
mented so that higher yields or selectivities can be achieved.
In evaluating specific unit operations, students must imple-
ment material they learned in the undergraduate curriculum
to achieve their goal of minimal environmental impact. For
example, we use equilibrium thermodynamics and mass trans-
fer to estimate the VOC emissions from a vapor degreaser.
Students complete the mass balance to determine the amount
of solvent evaporated and then watch the emissions decrease
when the solvent is changed to one with lower volatility.
Similar calculations using fundamental methods describe
how to estimate emissions from other vessels. Alternatively,
the EPA and the American Petroleum Institute have com-
piled software tools to perform unit-specific emissions esti-
mates, accessed through the EPA or
P2workshop websites.
One can also consider optimizing an individual unit opera-
tion for maximum performance. Consider the design of a
chemical reactor: suppose that two parallel pathways can
occur, one producing the desired product and the other pro-
ducing an undesired product. The reactor design textbook116
teaches us that when the activation energy for the reaction
leading to the desired product is greater than that of the other
reaction, raising the temperature will improve the process
selectivity. Thus, students must recall previously learned
material or expand upon material that may have only been
superficially discussed in previous classes.
A second component of this section is analysis of the
overall operation of the plant to minimize the environmental
impact of the process. Here, we use the concepts of the
process-design course, along with the process simulator, but
our objective is the minimization of waste streams. When we
incorporate a technique for calculating the environmental
impact of a process stream, we obtain a quantitative measure
of the environmental impact of the proposed process.
We have used the Waste Reduction (WAR) algorithm
developed by the EPA Risk Reduction Research Labora-
tory"71 as our basis for estimating potential environmental
impact. Although several algorithms exist that could be used,
WAR provides a consistent measure of estimating the envi-
ronmental impact of each process stream. For us, WAR has
two primary advantages: first, it is designed to work with the

ChemCad simulator, the process simulator used in all of our
undergraduate classes; second, WAR is being developed
locally, so the developers are accessible to our students. We
asked one of the researchers developing the algorithm to
present the methodology to the class in a guest presentation.
This researcher then served as a consultant to our class as
they used the WAR algorithm to work on the project.
A process analysis project served as the capstone assign-
ment for the class. Working in concert with a production
facility in Toledo, we developed a process based on their
commercial plant for the manufacture of pentaerythritol from
formaldehyde and acetaldehyde. The use of a local industrial
process had both advantages and disadvantages. On the plus
side, the students were able to visit the plant and observe the
process in operation, the process presentation was made by
an engineer involved in the operation of the plant, and they
were working on a real problem. But because of proprietary
needs, the company was reluctant to share all of the informa-
tion about their process with the students. In addition, the
process contained elements that were not easily simulated
using ChemCad, obscuring some of the pollution prevention
issues within the simulation difficulties. Of course, any pro-
cess can be used for this analysis as long as data can be
found in the literature to support the simulation effort.
Despite the difficulties that we had in accessing propri-
etary information and implementing the pentaerythritol pro-
cess on the simulator, the students were quite successful in
evaluating pollution prevention opportunities for the pro-
cess. For illustrative purposes, the results of two different
groups' efforts are summarized here. In the original process,
formic acid is used to neutralize the product stream leaving
the reactor (the reaction must be carried out in basic media to
facilitate the reaction). One group proposed using sulfuric
acid, which converts by-product sodium format into formic
acid and sodium sulfate; the formic acid is converted to
methyl format (a salable product) while the latter is recov-
ered as sodium sulfate (also a salable product). Analysis of
the economics of the modified process showed a potential
17% cost savings relative to the cost of the original process.
Use of the WAR algorithm, which reports potential environ-
mental impact in terms of impact units of non-product per
kilogram of product, showed a reduction in environmental
impact of nearly 75%. Other groups focused more exten-
sively on the inclusion of recycle streams. Nearly complete
recycle could be achieved, decreasing the potential environ-
mental impact by almost 99%. But, completing the recovery
and recycle operations required installation of two distilla-
tion towers, a heat exchanger, and several pumps. Thus, the
economics of this operation were not entirely favorable.

The ABET EC 2000 guidelines dictate several new chal-

Summer 2000

lenges for engineering curricula that can be met, in part,
through the P2 class. In particular, this class impacts five of
the nine expected program outcomes, including
f an understanding of professional and ethical responsibility
g. an ability to communicate effectively
h. the broad education necessary to understand the impact of
engineering solution in a global and societal context
i. a recognition of the need for, and an ability to engage in,
life-long learning
j. a knowledge of contemporary issues
In addition, the AIChE Program Criteria specified that
students must achieve a "working knowledge, including safety
and environmental aspects" of specific elements of chemical
engineering. The pollution prevention requirement repre-
sents an ideal environment in which to stress these elements
of chemical engineering education.
0 Outcome f: Environmental considerations are an important com-
ponent of a chemical engineer's professional responsibility. The
pollution prevention course provides an excellent opportunity for
addressing the ABET criterion on developing an understanding of
those responsibilities. It places a major emphasis on environmental
awareness, and in particular it focuses the students' attention on
their role in minimizing the environmental impact of chemical
processes. Course assignments require evaluation of the environ-
mental impact of a chemical process and the engineer's responsi-
bility in reducing that impact. We discuss alternative ways of
meeting the environmental challenge, and even raise the issue of
meeting the emissions standard by diluting the stream. Evidence
that students understand their professional and ethical responsibili-
ties is demonstrated through their evaluation of these issues within
their project presentations.
C Outcome g: It is difficult to develop standard assessment tools
(i.e., exams) for complex environmental problems; thus, we assess
the student's understanding through a series of written projects.
These assignments require analysis of a problem and a discussion
of potential solutions. A written report is normally the result of the
student team's analysis. The reports are graded on both the techni-
cal content and the written communication. Because engineers
must learn to communicate environmental issues in non-technical
terms, this element is stressed with the report. Students demon-
strate enhanced written communication skills by the improvement
of their written reports over the course of the term.
3 Outcome h: The framework of the P2 course includes a discus-
sion of pollution, but pollution includes more than just evaluating
the emissions from a chemical process. We also discuss the global
impact of emissions in the context of global warming, ozone deple-
tion, acid rain, etc. Perhaps the area in which this recognition is
most clearly addressed is through life cycle analysis, which pro-
vides the class with a quantitative means of assessing the impact of
a product or process on the environment. Within this evaluation,
students must assign an importance to each environmental impact
from a global standpoint. Their ability to describe CO, emissions in
terms of its impact on global warming, and the different emissions
levels achieved in process alternatives, demonstrates their under-
standing of the global impact of their decisions.
3 Outcome i: Because environmental problems are constantly
changing, assignments in this class generally require that students

seek out data through which an evaluation can be made. We en-
courage students to look outside of the classroom to access data
needed to solve the problem. Moreover, the assignments are framed
in a general context. For example, consider this problem statement
for an assignment on energy and the environment:
There is a wealth of information describing the electric utility industry, including
detailed descriptions of numerous coal-fired power plants. You should identify a
particular power plant and calculate (as best as possible) the overall efficiency of
the plant. This estimate can be based on the amount of coal consumed in the plant
and the amount of electricity generated. Also, estimate the emissions of CO,, SO,,
PM, and NO, in the plant, per unit of energy. Compare these values against
industry standards (averages) to determine the plant efficiency.
The students have been given guidance on where to find the re-
quired information, but the data are not provided. Moreover, the
specific information required to solve this problem is not stated,
although several different sources of data were identified. By re-
quiring students to confront open-ended technological issues for
which information must be sought outside the usual classroom/
textbook sources, they experience the challenge and excitement of
tackling problems as a professional does. Completion of the assign-
ment demonstrates their ability to gain information and learn on their
3 Outcome j: One of the primary technical issues in society today
is the impact of technology on the global environment. Recent
discussions in the popular press on global warming, loss of
biodiversity, farmland, and habitat, on acid rain and ozone deple-
tion, among other things, point to the importance of the environ-
mental challenge we are currently facing. This is also the focus of
the P2 course. Environmentally conscious design of all processes
and products can minimize the environmental impact of the techni-
cal products that modem society demands. A completed LCA
compares different products in a rational way and provides a tool
for evaluating different alternatives and choosing the one that is
most environmentally benign. Analysis of chemical processes, ei-
ther individual unit operations or the entire process, teaches the
students some simple techniques that can be used to reduce the
environmental impact of these activities.
In direct response to this issue, a new component of the class has
been introduced this year. Students are now asked to maintain an
environmental journal, described in the current syllabus as:
Environmental Journal: Virtually every day there is an article in the
popular or technical press regarding an environmental issue. Each
student should collect articles throughout the semester and place them
into a journal. After reading the article, the student should write a
short (approximately 3-5 sentences) technical evaluation of the science
and engineering basis behind the written article. Towards the end of
the semester, the student should look for a series of articles from his/
her journal on a common subject and prepare an extended evaluation
(approximately 2-3 pages) describing the issue. Additional technical
information may be required and should be cited, as necessary. The
journal will be submitted on the last day of class.
Clearly, the journal requires that the students remain aware of
contemporary environmental issues within their own community
and throughout the world.

The primary objective of the P2 course is to give students
a different perspective on the design process and to demon-

Chemical Engineering Education

state the opportunities for using chemical engineering prin-
ciples to minimize environmental impacts. For this reason,
there is not a lot of new material presented within the course.
Rather, it repackages many of the fundamentals learned in
earlier courses and provides new insight into the use of these
calculations. This was noted in student comments from an
end-of-course survey completed in the fall of 1998. In a mid-
course survey conducted in the fall of 1999, two questions
explicitly requested information on this subject. Students
agreed that the course shows that chemical engineering cal-
culations can be applied in a new context and that pollution
prevention is relevant to chemical engineering. Students also
recognized that this course has limited technical content.
One of the primary issues raised by the students in the fall
1998 survey was the lack of exams and the reliance on
projects. In general, students believed that the projects pro-
vided a good extension of the materials discussed in the
class, but because grading a project tends to be more subjec-
tive than grading an exam, they felt that the grades did not
reflect their learning. In addition, because the projects were
group efforts, the students recognized that the project grade
could reflect the abilities of a single member of the group
and might not represent the capabilities of the others.
Based on the recommendations of these students, a series
of quizzes was added for the following semester; the mate-
rial for the quizzes came directly from lecture notes and
provided an objective evaluation of each student's perfor-
mance. In the fall 1999 survey, however, the students were
neutral about the ability of the quizzes to provide a quantita-
tive measure of classroom learning. Rather, they agreed that
the projects were relevant to the lecture material. This conflict-
ing information, obtained from different groups of students,
points out the necessity of having multiple evaluation tools.
At the University of Toledo we are specifically interested
in understanding how pollution prevention can be used to
achieve ABET goals. In the spring of 1999, graduating se-
niors were asked to complete a survey matching the ABET
goals with the course in which those skills were learned. Of
32 graduating seniors, 9 (28%) indicated that P2 was most
helpful with regards to Criterion h, the broad education
necessary to understand the impact of engineering solutions
in a global and societal context. Additionally, 6 out of 32
(19%) listed Criterion j, a knowledge of contemporary is-
sues. Similar information was requested from the mid-course
survey conducted in the fall of 1999. Students agreed strongly
that P2 teaches a knowledge of contemporary issues (Crite-
rion j). The need for and ability to participate in life-long
learning (Criterion i) and the ability to function on
multidisciplinary teams (Criterion d) were also skills ob-
tained within the P2 course.

The pollution prevention course required of chemical en-

gineering students at the University of Toledo provides a
solid introduction to the role of the chemical engineer in
controlling the future environmental impacts of chemical
processes. The course describes the environment in a global
context, forcing students to consider the far-ranging conse-
quences of their engineering decisions. It reinforces many
concepts taught throughout the chemical engineering cur-
riculum and demonstrates new ways to apply the fundamen-
tal calculations that have already been learned. Because the
concept of pollution prevention is still evolving and protec-
tion of the environment is a topical issue, this course pro-
vides an ideal environment in which to educate our students
about the impact of engineering decisions on everyday life.

1. Anastas, P.T., and J.C. Warner, Green Chemistry Theory
and Practice, Oxford University Press, New York, NY, p. 7
2. Mulholland, K.L., and J.A. Dyer, Pollution Prevention: Meth-
odology, Technology, and Practices, AIChE, New York, NY
3. Dyer, J.A., and K.L. Mulholland, "Follow this Path to Pollu-
tion Prevention," Chem. Eng. Prog., p. 34, January (1998)
4. Mulholland, K.L., "Pollution Prevention-Business Re-
sponse: Past, Present, and Future," presented at the Uni-
versity of Toledo, March 4, 1999
5. Grant, C.S., M. Overcash, and S.P. Beaudoin, "A Graduate
Course on Pollution Prevention in Chemical Engineering,"
Chem. Eng. Ed., 30(4), 246 (1996)
6. El-Hawagi, M.M., and H.D. Spriggs, "Educational Tools for
Pollution Prevention Through Process Integration," Chem.
Eng. Ed., 33(4), 246 (1999)
7. Alien, D.T., and K.S. Rosselot, Pollution Prevention for
Chemical Processes, John Wiley & Sons, Inc., New York, NY
8. Higgins, T.E., Pollution Prevention Handbook, CRC Lewis,
Boca Raton, FL (1995)
9. El-Halwagi, M.M., Pollution Prevention Through Process
Integration, Academic Press, New York, NY (1997)
10. Graedl, T.E., and B.R. Allenby, Industrial Ecology, Prentice-
Hall, Englewood Cliffs, NJ (1995)
11. Bishop, P.L., Pollution Prevention: Fundamentals and Prac-
tice, McGraw-Hill, New York, NY (2000)
12. Graedl, T.E., and B.R. Allenby, Industrial Ecology and the
Automobile, Prentice-Hall, Englewood Cliffs, NJ (1998)
13. Allen, D.T., N. Bakshini, and K.S. Rosselot, Pollution Pre-
vention: Homework and Design Problems for Engineering
Curricula, AIChE CWRT, New York, NY (1992)
14. McDonald's-EDF Environmental Task Force, videotape avail-
able from University of Michigan National Pollution Pre-
vention Center (1993)
15. Graedl, T.E., Streamlined Life-Cycle Assessment, Prentice-
Hall, Englewood Cliffs, NJ (1998)
16. Fogler, H.S., Elements of Chemical Reaction Engineering,
3rd ed., Prentice-Hall, Upper Saddle River, NJ (1999)
17. Cabezas, H., J.C. Bare, and S.K. Mallick, "Pollution Preven-
tion with Chemical Process Simulators, The Generalized
Waste Reduction (WAR) Algorithm-Full Version," Comput.
Chem. Eng., 23(4-5) 623 (1999)
18. Prothero, Scott, and Fred Arnold, "Evaluating Environmen-
tal Releases and Exposures," in Green Engineering, D.A.
Allen and D.R. Shonnard, in preparation. Presented at ASEE
Green Engineering Educators Workshop, Charlotte, NC,
June 20, 1999. 0

Summer 2000

3 1 classroom



University of Calgary Calgary, Alberta, Canada T2N 1N4

he classical approach to process control education of
chemical engineering['-4] has been to employ the fre-
quency response methods of process control that were
originally developed as pen-and-paper methods for the mod-
eling of process systems. It has been evident for some time
that the way process control is taught to chemical engineers
needs to be updated."[
There is an academic requirement that the fundamentals of
process control be taught in a more practical and concrete
way than afforded by the traditional classical approaches.
The increasingly overloaded degree syllabus provides addi-
tional impetus to reorganize subjects and reduce superfluous
detail. Brisk and Newell51 recommended training students
"in how to utilize process control systems with just enough
theory that they can understand what they are using and
maintaining." They went on to lament that "unfortunately
most of our institutions are teaching too much theory, very
little on utilization and maintenance." Doss[6' comments in
Edgar's round-table discussion on process control education
in the year 2000['2] that "students tend not to retain the
mathematical theory but to remember the experiences from
control laboratory experiments and simulations." Ramaker,
et al.,["1 point out that "an undergraduate in a chemical
engineering curriculum [studying] process control should be
taught using concepts that fit with the rest of chemical engi-
neering education maintaining the undergraduate cur-
riculum as closely tied as possible to the time domain."
There is also an industrial imperative to teach material that
is of use to the practicing engineer. Downs and Doss171 noted
that "what the [graduating engineer] needs is a base level
understanding of differential equations, process dynamics,
dynamic modeling of basic unit operations, basic control
algorithms (such as PID), cascade structures, and feed for-
ward structures. With these basic tools and an understanding
* Address: AEA Technology Engineering Software, Hyprotech Ltd.,
Suite 8000, 708-8th Ave SW, Calgary, Alb., Canada T2P 3V3

of how to apply them, he can solve most of his control
problems himself. What he does not need is the theory and
mathematics that usually surround process control." The
industrial imperative is further reinforced by comments such
as the following that come from practicing chemical engi-
neers working in process control or process operations:[8]
"I never made use of Bode plots or root locus when I was
designing a control loop."
"There are no transfer functions out there in the real plant."
"The material I had been taught was of no use in com-
missioning a control loop.
Control education clearly needs to do better.

Brent Young is Associate Professor of Chemi-
cal and Petroleum Engineering at the University
of Calgary. He received his BE (1986) and PhD
(1993) degrees in chemical and process engi-
neering from the University of Canterbury, New
Zealand. His teaching and research interests
Center on process control and design.

Donald Mahoney is Vice President of AEA Tech-
nology Engineering Software, Hyprotech Ltd. He
earned a BSc in mechanical engineering from
Penn State, a Masters in control from Purdue,
and an MBA from Delaware. He has lectured
extensively on the topic of process control, pub-
lishing a number of papers in refereed journals.

William Svrcek is Professor of Chemical and
Petroleum Engineering at the University of
Calgary. He received his BSc (1962) and PhD
(1967) degrees in chemical engineering from
the University of Alberta. His teaching and re-
search interests center on process control and
Copyright ChE Division of ASEE 2000

Chemical Engineering Education

Classical control methods were developed between the
1940s and the 1960s in the mechanical and electromechani-
cal engineering disciplines. Given the limitation of computer
hardware and software at that time, it was impractical to
solve large numbers of higher-order differential equations.
Furthermore, since mechanical and electromechanical sys-
tems are typically linear and possess little dead time, they
lend themselves to analytical and graphical techniques. Hence,
there was the development and popularization of analytical
and graphical techniques such as
Transform methods (Laplace and Fourier Transforms)
Graphical frequency domain methods (Bode, Nichols and
Root locus analysis
Given the fit to their purpose, classical control techniques
still prevail and remain relevant in these engineering disci-
plines today.
Although these methods make up almost half the content'41
of standard control texts,"113] they all share a number of
deleterious characteristics. They all require a substantial
amount of applied mathematics. In spite of the high level of
mathematics required, in order to apply the analysis the
system must first be made linear; the methods also have a
transfer function basis, focus on individual units, and are
generally good only for single loops and PID control. Lim-
ited multivariable and no plant-wide controls are possible.
Beyond the engineering deficiencies of classical techniques,
there are also implications from a teaching and learning
perspective. The abstraction of classical methods makes a
difficult subject more difficult, and the methods lack physi-
cal meaning, obscuring the central problem of how to modify
the system in order to achieve control.'" These methods are
also not suited to "what-if' studies, such as determining loop
performance with parameter variation.
Today's ready availability of hardware and software has
called into question the relevance of these classical methods
for a primary course on process control. A number of previ-
ous workers have also identified this need for change. Many
workers in the past decade have incorporated simulation
software into the syllabus and deleted previous graphical
procedures while retaining the classical methods. However,
Brauner, et al.,[91 and then Stillman,181 Bissell,[101 and Ramaker,
et al.,"11 almost simultaneously proposed the more radical
solution of complete replacement of classical methods with
computer simulation, i.e., not as an add-on, but as an integral
part of the teaching and learning of process control. Ramaker,
et al.,1"' possibly said it best when they said "this doesn't
mean that the Laplace transform cannot be used as a tool to
solve differential equations in the undergraduate course. Nei-
ther does it mean that frequency domain analysis and design
are not useful in chemical engineering. It only means that we

feel that frequency domain analysis and design should be
taught at a graduate level, maintaining the undergraduate
curriculum as closely tied as possible to the time domain."
In this paper we will outline and evaluate the actual imple-
mentation of such a complete real-time approach to process

Unlike mechanical and electromechanical systems, chemi-
cal processes are characterized by high degrees of non-
linearity, process interactions, and substantial dead time.
Additionally, due to these non-idealities, chemical process
control demands to be addressed with a multivariable and
plant-wide view. As such, applying classical techniques to
chemical process control is a bit like using a wrench to do a
hammer's work. In an ideal world, the chemical engineer
would have a "virtual plant" on which to experiment. It
would capture most of the important non-idealities the real
world imposes and would allow the engineer to readily test
even the most outlandish of control structures with impu-
Early attempts to realize this "ideal world" date back to the
seventies and eighties114' when dynamic simulators such as
DYFLO, DYNSYS, or SPEEDUP first became available for
the solution of nonlinear differential equations describing
process dynamics. The hardware was slow at that time,
however, and the software was impractical for students to
learn and implement within a reasonable time frame. There
was effectively no user interface in that the graphics were
poor and the programs were run batch-wise.
In today's "simulation-rich" environment, however, the
right combination of hardware and software is available for
implementing a "real-time" approach to process control edu-
cation. l41 The hardware and software, such as HYSYS, As-
pen Dynamics, or MATLAB, is now fast and easy to use.
Simple, complex, and/or user-defined process modules are
available, and it is now easy to do "what-if' studies, multi-
loop, and plant-wide control simulations. The software user
interface is now graphical and interactive, and the software
can be painlessly run on a PC.
In short, the "virtual plant" has arrived.
This real-time approach also quite naturally lends itself to
active, hands-on or resource-based learning."15-16] In our
course, we use a small number of lectures at the beginning to
motivate students and to provide a fundamental understand-
ing rather than simply transmitting information; we also use
hands-on simulation tutorial sessions on case-study projects
facilitated by the instructors, which we call workshops.[131
The syllabus covers the development of mathematical mod-
els to describe the transient real-time response characteris-
tics of basic process elements, capacity, and dead time;
fundamentals of single-input, single-output systems; use of a

Summer 2000

dynamic process simulator; block-flow diagram of a feed-
back-control loop; process-control hardware; basic control
modes; tuning feedback controllers; cascade control;
feedforward control; common control loops; distillation-col-
umn control; design of multiple single-loop controllers; and
plant-wide modeling and control.
We also note that while computer simulations provide
generally favorable experiences, real experiments are still
necessary and desirable.11" Therefore, we employ in our
course a cascade of tanks and a heat exchanger in a pilot
plant laboratory that allows students to perform process iden-
tification exercises on real plants and to tune real controllers.
So that the student understands the underlying "physics" of
process control, modeling exercises that require the student
to write the describing differential equations and solve them
numerically using MATLAB are associated with these labo-
ratory plant experiences.

The real-time methodology will now be illustrated and
compared with the classical approach by application to the
feedback control of liquid level in a separator (see Figure 1).
The unit of Figure 1 is usually represented by a system of
transfer functions as shown in the block diagram for the
liquid level loop of the separator, shown in Figure 2.
It is obvious that, from a learning perspective, the transfer
function block diagram of Figure 2 bears no obvious rela-
tionship to the real plant in Figure 1, i.e., the representation
lacks physical meaning. Many assumptions and empirical
determinations are necessary in order to relate the two. It
bears repeating that the abstract nature of these sorts of
classical methods makes the subject unnecessarily difficult,18
obscuring the key issue of real process control, i.e., how to
modify the system of Figure 1 in order to achieve control.
In pursuit of the real-time approach, we need to find a
better, more intuitive representation of the real plant. A
better start is the word-block diagram of the separator liquid
level loop shown in Figure 3. Although no underlying math-
ematics has been introduced, the word-block diagram illus-
trates the real process control situation of Figure 1 in a more
physically meaningful way. The underlying mathematical
representation of the process is the set of non-linear differen-
tial equations that can be written for each block and solved
numerically or simulated by current process simulators such
In the simulation approach, the student can now easily
construct a real-time simulation given the input flow, tank
volume, temperature, and pressure. Figure 4 is the plant
process-flow diagram simulated in HYSYS, which shows a
one-for-one match with the real plant.
The student can then easily indulge in "what-if' studies to
find an optimal control structure and set of control param-

Figure 1. Schematic of vapor-liquid separator with
standard feedback controls.

Figure 2. Classical transfer function block diagram of the
liquid level loop of the separator.

Figure 3. Word-block diagram of the liquid level loop of
the separator plant.

Figure 4. Plant process-flow diagram simulated in
HYSYS process.

Chemical Engineering Education

eters for the controllers-the fundamental air
of process control. Figure 5 shows a screen
shot of the simulated response of the separa-
tor to a step change in the set point of the
liquid-level controller.
It bears mentioning again here that both the Z
real-time and "what-if' studies described here
are both difficult and extremely time con-
suming to perform when employing classical
methods of process control instruction.
This real-time approach to process educa-
tion was first developed in 1996 as a text and
an associated set of workshops. This version
was used at the University of Calgary during
the 1997 academic year as a pilot course for
nine students as their senior-year controls Fit
course. Their comments motivated a revised
second version of the notes and workshops.
This second version was used as the basis for the cla
1998, 1999, and 2000, totaling forty-five, sixty, and
students, respectively. A further revision has just bee

60B O 0---------"-------- ...-- ---
1.479e+004 (kg/h)
5900 (kPa)

499 [kgmole/h]
24.45 [%)

2988e+004 2.991e+004 2993e+004 2.996e+004 2999e*004

fure 5. Real-time simulator response to a set point change in the level

sses of ated the new "hands-on" real-time simulation workshop ap-
eighty proach as effective, useful, and applicable.
n pub-

As a means of generating feedback, the students were
asked to complete a questionnaire. Overall, the overwhelm-
ing majority of students preferred the "hands-on real-time
approach" to learning process control. More than 80% of the
students said the approach was clear, concise, useful, and
applicable. The major complaints, but only from a minority
of students, were that they did not like "hands-on" self-
directed learning, found the workshops too involved and
time-consuming, and would have preferred a standard course
consisting of standard lectures, assignments, quizzes, and a
written final exam. Our anecdotal feedback from former
students in industry is also overwhelmingly positive.

There is a need for change to conventional process control
education-a change from a classical frequency domain meth-
odology to instruction using concepts that fit with the rest of
chemical engineering education, i.e., a real-time approach.
A real-time simulation approach to undergraduate process
control education in chemical engineering with the aid of
realistic "hands-on" workshops involving real-time simula-
tion of chemical processes was presented. The workshops
are based on fundamental process models of industrial unit
operations using educationally affordable and readily avail-
able commercial process simulation software. The real-time
simulation approach to process control education was pre-
sented with the aid of a case study and compared with the
traditional classical approach.
Student feedback from four years of implementation evalu-

b measured variable
LIC level indicating controller
e controller input error signal, r-b
K controller gain
mn manipulated variable, control effort
PC personal comp. or pressure controller
s Laplace transform variable
t time constant
c controller
f flow
L liquid
v valve

c controlled variable, proc. variable
e natural logarithm
F flow
L dead time
P pressure
r set point
T temperature
V volume

d derivative or disturbance
i integral
m measurement
V vapor

1. Coughanowr, D.R., and L.B. Koppel, Process Systems Analy-
sis and Control, McGraw-Hill (1965)
2. Ogunnaike, B., and W.H. Ray, Process Dynamics, Modeling,
and Control, Oxford University Press (1994)
3. Marlin, T.E., Process Control: Designing Processes and Con-
trol Systems for Dynamic Performance, 2nd ed., McGraw-
Hill (2000)
4. Rawlings, J.B., Review of Undergraduate Process Control
Textbooks, Amer. Contr. Conf., Seattle, WA, June 23 (1995)
5. Brisk, M., and R.B. Newell, "Current Issues and Future
Directions in Process Control," Chem. Eng. in Australia,
14(3), 8 (1989)
6. Doss, J.E., "Commentary on Process Control Education in
the Year 2000," Chem. Eng. Ed., 24(2), 76 (1990)
7. Downs, J.J., and J.E. Doss, "Present Status and Future
Needs: A View from North American Industry," in Proceed-
ings of Chemical Process Control IV, Y. Arkun and W.H.
Ray, eds., AIChE, New York, NY (1991)

Summer 2000

8. Stillman, K.A., "The Place of Classical Control in Control
Education," Proc. 4th IFAC Symp. on Advs. in Cont. Ed.,
Istanbul, Turkey, 235 (1997)
9. Brauner, N., M. Shacham, and M.B. Cutlip, "Application of
an Interactive ODE Simulation Program in Process Control
Education," Chem. Eng. Ed., 28(2), 130 (1994)
10. Bissell, C.C., "Control Education: An Iconoclast's View,"
Proc. 4th IFAC Symp. on Advs. in Cont. Ed., Istanbul,
Turkey, 229 (1997)
11. Ramaker, B.L., H.K. Lau, and E. Hernandez, "Control Tech-
nology Challenges for the Future," in Proc. Chem. Proc.
Cont. V Conf., Tahoe City, CA (1996); AIChE Symposium
Series No. 316, 93, 1 (1997)
12. Edgar, T.F., "Process Control Education in the Year 2000,"
Chem. Eng. Ed., 24(2), 72 (1990)
13. Svrcek, W.Y., D. Mahoney, and B.R. Young, A Real-Time
Approach to Process Control, John Wiley & Sons, Ltd.,
Chichester, England, (2000)
14. Svrcek, W ., W.D. Sim, and M.A. Satyro, "From Large
Computers and Small Solutions to Small Computers and
Large Solutions," Proc. CHEMECA '96, 24th Australian
and New Zealand Chem. Eng. Conf., Sydney, Australia, 2,
15. Litster, J.D., R.B. Newell, and P.L. Lee, "Resource Based
Education for Training Chemical Engineering," Chem. Eng.
in Australia, 15(2), 15 (1990)
16. Clough, D.E., "Bringing Active Learning into the Tradi-
tional Classroom: Teaching Process Control the Right Way,
ASEE Annual Conf., Seattle, WA (1998)
17. White, S.R., and G.R. Bodner, "Evaluation of Computer-
Simulation Experiments in a Senior-Level Capstone ChE
Course," Chem. Eng. Ed., 33(1), 34 (1999) J

Ba letters to the editor

To the Editor:
Professor Grossmann correctly points out errors that can
occur when using citation statistics to compare graduate
programs.1' However, the differences between the results of
the two studies that Professor Grossman considered (the
National Research Council report'2] and Science Watch[3])
should not be used as a reason for discounting the value of
citation statistics. The major difference in the results likely
arises from a difference in what the two studies were de-
signed to measure, rather than from errors. The NRC study
attempted to measure quality of departments or programs;
the Science Watch study compared institutions. Therefore,
the NRC study reported citations arising from a single pro-
gram or department within a university while Science Watch
reported citations from the entire university. Furthermore,
while the NRC study attempted to be inclusive and cover all
journals, the Science Watch study covered a very narrow
range of journals. For example, the Science Watch list in-
cluded no electrochemical journals, no materials journals
other than polymers (and only three of those), and only one
biotechnology journal.
As a consequence, even without errors of the types noted
by Professor Grossmann, the citation counts will vary greatly

between the two studies. These differences could be in either
direction. A university's chemical engineering activities
would appear relatively weaker in the Science Watch study
if it had major efforts in fields not included in the Science
Watch journal list. Conversely, the chemical engineering
activities would appear relatively stronger in Science Watch
if the university had efforts in areas such a catalysis, surface
chemistry, and combustion outside of the chemical engi-
neering department. The Science Watch study is appropriate
for comparing universities in the particular fields of applied
chemistry and chemical engineering covered in the Science
Watch database; it is not appropriate for comparing chemi-
cal engineering departments and should not be used for that
purpose. The NRC study, which referred to programs rather
than universities, has a more comprehensive database of
publications and is appropriate for comparing chemical en-
gineering programs.
Professor Grossmann is correct when he says we should
use great care in interpreting countable indices such as cita-
tions and publications. However, it is possible to devise
multiple, countable criteria that can give an alternative mea-
sure of graduate program quality.[41 Engineers, in particular,
should not be reluctant to use countable indices rather than
reputationall rankings." The reputationall rankings" give
little more than historical perspective and cannot accurately
portray a dynamic field such as modern chemical engineer-
John C. Angus
Case Western Reserve University
1. Grossmann, I.E., "Some Pitfalls with Citation Statistics,"
Chem. Eng. Ed., 34(1), 62 (2000)
2. Goldberger, M.L., B.A. Maher, and P.E. Flattau, Eds., Re-
search Doctorate Programs in the United States: Continuity
and Change, National Academy Press, Washington, DC
3. Science Watch, 3(2), 1 (1992)
4. Angus, J.C., R.V. Edwards, and B.D. Schultz, Chem. Eng.
Ed., 33(1), 72 (1999) p

To The Editor:
At the risk of faning the flames of controversy concerning
use of citation statistics in rankings of chemical engineering
programs, I would like to add some comments engendered
by the recent article by Ignacio Grossmann.L" I do so from
the point of view of a department that has admittedly fared
reasonably well by current measures, as indicated below.
Professor Grossmann has pointed out some real and poten-
tial flaws in the citation statistics compiled by ISI and fre-
quently used by one group or another to establish relative
rankings of research programs in many fields, including
chemical engineering. Assuming that errors arising from
misspellings will tend to be randomly distributed, I would
like to focus on some pitfalls that are far more serious.

Chemical Engineering Education

Full Text

xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EB18SM4TX_GOH14K INGEST_TIME 2012-02-17T16:50:45Z PACKAGE AA00000383_00147