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
University of Michigan
William J. Koros
University of Texas, Austin

E. Dendy Sloan, Jr.
Colorado School of Mines

Gary Poehlein
Georgia Institute of Technology
Klaus Timmerhaus
University of Colorado

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
David F. Ollis
North Carolina State University
Angelo J. Perna
New Jersey Institute of Technology
Ronald W. Rousseau
Georgia Institute of Technology
Stanley I. Sandler
University of Delaware
Richard C. Seagrave
Iowa State University
M. Sami Selim
Colorado School of Mines
James E. Stice
University of Texas at Austin
Donald R. Woods
McMaster University

Chemical Engineering Education

Volume 33

Number 2

Spring 1999

90 Art Westerberg, of Carnegie Mellon University,
Lorenz Biegler, Ignacio Grossmann, Storge Stephanopoulos

96 Rose-Hulman Institute of Technology, Jerry Caskey, Hossein Harrir

102 Outcomes Assessment: Its Time Has Come, Joseph A. Shaeiwitz
104 Re-Engineering Engineering Education: Chemical Engineering and ABET EC
2000, Stan Proctor, Dick Seagrave
106 Outcomes Assessment: Opportunity on the Wings of Danger, Gloria Rogers
108 Using Portfolios to Assess a ChE Program, Barbara M. Olds, Ronald L. Miller
116 Outcomes Assessment: An Unstable Process? David DiBiasio
122 The Articulation Matrix: A Tool for Defining and Assessing a Course, Barry
McNeill, Lynn Bellamy
128 Building the EC 2000 Environment, Daina Briedis

136 Memo, R. M. Felder

138 Important Concepts in Undertraduate Kinetics and Reactor Design Courses,
John L. Falconer, Gary S. Huvard
142 A New Approach to Teaching Turbulent Flow, Stuart W. Churchill
150 Teaching Creative Problem-Solving Skills in Engineering Design,
J.G. Mackenzie, R.M. Allen, W.B. Earl, I.A. Gilmour
158 Experience with Teaching Design: Do We Blend the Old With the New?
Lawrance Flach
166 The Green Square Manufacturing Game: Demonstrating Environmentally
Sound Manufacturing Principles, Suzanne S. Fenton, James M. Fenton

154 Icing the Rink: A Problem for the Stoichiometry Course, David A. Dudek

162 Internationalizing Practical ChE Education: The M.I.T. Practice School in
Japan, Andrea J. O'Connor, Angelo W. Kandas, Yukikazu Natori, T. Alan

172 Removal of Heavy Metals in Wastewater by Electrochemical Treatment,
Eduardo Exp6sito, Marina Ingles, Jestls Iniesta, Jose Gonzdles-Garcia,
Pedro Bonete, Vicente Garcia-Garcia, Vicente Montiel

> 121 Book Review
> 134 Calendar: ASEE Annual Conference & Exposition
> 141 Letter to the Editor
> 161 Books Received

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 1999 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.

Spring 1999


Art Westerberg

... of




Carnegie Mellon University
Pittsburgh, PA 15213

It is refreshing to find a professor who, having had a
profound impact in his field, is also a truly nice, honest
individual, well liked and highly respected by all. It is
also rare that a single person's career can be used as a
measure in describing an entire research area. This is true in
Arthur Westerberg's case. He has built a foundation that is
used as a roadmap for process systems engineering, a dy-
namic and highly successful area of chemical engineering.
Art approaches both life and research with a great sense of
humor and optimism.
Arthur W. Westerberg was born in St. Paul, Minnesota, in
1938, and spent his childhood in the farm country of that
area. He inherited a practical bent as a problem solver from
his father, who provided engineering services for farm equip-
ment and construction. He also inherited a knack for apply-
ing general concepts and principles to a variety of different
areas, and he later found that those interests were well suited
to chemical engineering. He thus found himself in the young
and dynamic atmosphere at the University of Minnesota in
the late '50s in the company of such trendsetters as
Amundson and Aris. After completing his BS degree in

Address: Massachusetts Institute of Technology, Cambridge,
MA 02139
Copyright ChE Division of ASEE 1999
90 Chemical Engineering Education

[Art] inherited a practical bent
as a problem solver from
his father ....
He also inherited a knack
for applying
general concepts and
principles to a variety of
different areas ...
he later found that those
interests were
well suited to chemical

1960, Art was advised by Amundson to seek out the best
graduate education he could find, and his choice was
Britain's Imperial College.
American postgraduate students were unusual in Britain in
1960, especially since there were few mechanisms to pro-
vide funding for them. Consequently, Art pursued a master's
degree at Princeton before accepting the position of Assis-
tant Lecturer at Imperial to study for his PhD. He worked
with Professor Roger Sargent, a young faculty member at
Imperial who spearheaded the field of process systems engi-
neering (PSE) and who spawned a legacy that includes virtu-
ally all PSE researchers. Working together on the develop-
ment of SPEEDUP (a process simulator widely used today
for steady state and dynamic modeling, analysis and optimi-
zation), they laid the foundation for computer-aided process
analysis. For instance, the Sargent-Westerberg algorithm"l
for partitioning flowsheets and systems of equations is still
widely used and quoted in the literature.
After receiving his PhD in 1964, Art joined Control Data
Corporation in San Diego as a systems analyst, where he
subsequently developed a number of novel codes for nu-
merical analysis. In particular, his work on fast Fourier trans-
forms led to efficient and powerful strategies for resolving
chromatographic peaks and spurred the application of this
separation technique in analytic chemistry.
The opportunity to teach and do research beckoned, how-
ever, and in 1967, Art went to the University of Florida and
began his academic career in process systems engineering.
PSE was in the embryonic stage in the late 60s, with only a
handful of researchers and a host of unsolved and poorly
defined problems in process design and analysis. Art's early
research pioneered the equation-oriented approach to pro-

Art Westerberg showed high proficiency in
handling cool issues at an early stage.

cess flowsheeting. In particular, he focused on systematic
methods for developing and solving process simulation prob-
lems and extending them so the processes could be opti-
mized as well. In addition, Art, along with other young
researchers, banded together to form the CACHE Corpora-
tion for education, and he also organized a number of sym-
posia to discuss open research problems and to share experi-
ences. Moreover, this emerging community forged impor-
tant links to other engineering disciplines where design prob-
lems still required definition and solution strategies.
The efforts of the young engineering design community
attracted quite a bit of attention in the 70s, especially at (the
newly renamed) Carnegie Mellon University (CMU). At
CMU, Dean Herbert Toor (with prudent advice from his
neighbor, the Nobel laureate Herbert Simon) initiated a De-
sign Research Center in 1975 and strongly influenced the
hiring of a number of faculty in the computer-aided design
area. They included Gary Powers in chemical engineering,
Steven Fenves in civil engineering, and Stephen Director in
electrical engineering. It was then only a matter of time
before the fertile environment at CMU and Art's own re-
search directions coincided. While Art was on sabbatical
with Rudy Motard at the Computer Aided Design Centre in
Cambridge, UK, the connection was made, and Art subse-
quently moved to Pittsburgh and to CMU.

Spring 1999

The following summary does little justice
to Art's research contributions. His
trailblazing research has strongly influenced
the current practice of process simulation,
process modeling environments, process op-
timization strategies, synthesis of chemical
processes, and the modeling of engineer-
ing activities.
Art's early work on process simulation con-
centrated on exploiting the structure of pro-
cess equations. This led to efficient strategies
for partitioning equation sets into subsets, tear-
ing strategies for process equations and
streams to ensure faster convergence, and se-
lection of decision variables to avoid struc-
tural singularities. This work led to the evolu-
tion of equation-based strategies for process
simulation that employ rapidly converging
techniques for simultaneous convergence,
which was also enhanced by Art and his col-
leagues. Developed in the early- to mid-70s,
these tools were almost two decades ahead of
their time and are now regarded as the basic
elements for large-scale modeling and opti-
mization of chemical processes.
With this background, Art's work evolved
to handle larger-scale problems and to con-
sider dynamic simulation. Coupled with this
task is the importance of handling differen-
tial-algebraic equations (and the treatment of
high-index systems, where simulations are
doomed to failure in a novice's hands) and
the extension to structuring calculations for
partial differential equations. The thread run-
ning through all of these novel contributions
was that the simultaneous equation-based
paradigm allowed tremendous generality in
dealing with a large set of simulation prob-
lems in process engineering. This principle
was especially powerful when extended to
the optimization of these systems. As a
result of the simultaneous approach, opti-
mization tasks were raised from a set of
tedious case studies to an efficiently inte-
grated part of the simulation platform. The
approaches developed by Art and his co-
workers paved the way for modern real-
time optimization of petrochemical plants
and refineries, which has now become the
industrial standard.


Art Westerberg's
PhD Students

F.C. Edie
J.A. de Souza Neto
R.L. McGalliard
C.J. De Brosse
J. R. Cunningham
S. Nayak
G. Stephanopoulos
G.L. Allen
J.V. Shah
T.J. Berna
S. Kuru
J. Cerda
M.H. Locke
M.H. Chao
P.A. Clark
M.J. Andercovich
J.B. Hillerbrand, Jr.
S.S. Kim
A.N. Hrymak
J. Vaselenak
R. Banares-Alcantara
N.A. Carlberg
F.D. Carvallo
P.C. Piela
A.K. Modi
O.J. Smith, IV
Y. Chung
R.S. Huss
O.M. Wahnschafft
M.E. Thomas
K.A. Abbott
B. Safrit
Joseph J. Zaher
D. Cunningham
B.A. Allan
V. Rico-Raminez

Coupled with his advances in simulation,
Art recognized the importance of developing
modeling systems that capture the physical
phenomena and topology of processing sys-
tems. Here, the challenge is also to embed
within modeling systems efficient simulation
tools that are easy to use, helpful with diag-
nostics, and extendable to building very large-
scale models. The vehicle for these ideas is
the environment that evolved into ASCEND
(Advanced System for Computation in Engi-
neering Design). Moreover, development of
ASCEND spawned a complementary research
effort into the creation of modeling strate-
gies and languages that, like the simulation
tools, were concise and efficient and would
support the construction of extremely large
models. ASCEND has evolved over four
generations and continues to be used by a
variety of researchers for modeling, simu-
lation, and optimization of steady state and
dynamic processes.
Art did not stop with the modeling of pro-
cessing systems, however. With the use of
ASCEND and other design tools in engineer-
ing project teams, Art recognized the impor-
tance of managing project information among
design teams and in developing platforms that
support the entire design process. Heading a
diverse team of researchers (engineers, com-
puter scientists, and even artists), Art has spent
the last decade molding these ideas into the n-
dim system in order to support the 'design' of
the design process for a project team. Off-
shoots of this project include products like the
LIving REpository (LIRE') that supports a
life cycle of information (authoring, search-
ing, editing, publishing, etc.) for a design team.
In tandem with modeling, simulation, and
optimization lie Art's contributions to the de-
sign of processing systems and the synthesis
of chemical processes. Adopting a systems
approach to discover and apply underlying
concepts for putting processes together, Art
attacked a wide variety of problems in process
synthesis from a wide variety of approaches.
His perspective and grasp of the synthesis prob-
lem were envisioned in 1980 in a beautifully
written review paper.121
Art's contributions in process synthesis can
be organized in threads along process lines,
starting with the design of heat-recovery net-
works and evolving to a broad set of separa-
Chemical Engineering Education


The Westerbergs: Art, Barbara, Ken, and Karl.

tion systems that include distillation, heat-integrated column
systems, multi-effect evaporation, and most recently, syn-
thesis of nonideal, azeotropic distillation. In all of these
areas, Art and his coworkers sought underlying guiding prin-
ciples that exploit the nature of the problem and provide an
understanding of 'why' the best design had its essential
characteristics. Art's approaches to synthesis have been novel
and diverse; they include strategies at the cutting edge of
development, evolutionary and heuristic strategies, optimi-
zation, and the use of artificial intelligence and expert sys-
tems. In all of these cases, Art was careful to choose the
'right tool' for the right problem and to develop a synergy
between both, in order to develop a deep understanding of
the designed system.
It goes without saying that Art's efforts have been recog-
nized by the chemical engineering community through nu-
merous honors and awards. They include membership in the
National Academy of Engineering, early recognition in the
CAST Computing Award, the AIChE Walker and Founder's
Awards, and the E. V. Murphree Award from ACS. At
Carnegie Mellon, he received the Swearingen Chair in 1982
and was named University Professor in 1992.

The list of Art's many research contributions does not
begin to present a complete picture of his contributions to
the Chemical Engineering Department at CMU. Shortly af-

ter arriving, Art assumed the position of Director of the
Design Research Center (DRC) and proceeded to build up
interaction among departments across the campus. Under his
leadership, the DRC provided a forum for like-minded fac-
ulty to collaborate and learn about design problems and
solution strategies in other fields. Tangible results were the
creation of a seminar series as well as a widely distributed
technical-report series. Art's influence also led to the hiring
of several design faculty on campus, including Ignacio
Grossmann in Chemical Engineering.
As department head, from 1980-83, Art faced some turbu-
lent times due to transitions in research funding and the
departure of several CMU faculty. During his term, Art
spearheaded rebuilding the department by hiring almost half
of its faculty, including Myung Jhon, Larry Biegler, Mike
Domach, Gary Blau (now at Purdue), and Greg McRae
(now at MIT). Art enjoyed a brief sabbatical rest in 1983-
4 as the Hougen Visiting Professor at the University of
Wisconsin. His return to CMU led to a number of impor-
tant achievements.
From 1985 to 1986, Art led the competition for a new NSF
Engineering Research Center in the area of interdisciplinary
engineering design. The resulting Engineering Design Re-
search Center (EDRC) was founded on the concept that
basic principles and tools for the design process could be
generalized across all engineering disciplines, which in turn
would lead to a more fundamental understanding of how to
improve the cost, quality, and time for developing designs.

Spring 1999

Art served as the first EDRC director from 1986-1989, and
his leadership fostered a culture of interdisciplinary research
and showed how design research cuts across domains.
Art's presence in the EDRC (and its successor, the Insti-
tute for Complex Engineered Systems (ICES)) remains strong
through his guidance of the n-dim group and his advice and
service on the ICES board. Within the chemical engineering
department, many of us look to Art for his advice, wise
counsel, and leadership. In the process-systems area, he has
further contributed to its growth by influenc-
ing the hiring of Erik Ydstie and Steinar
Hauan. Moreover, the department contin- Hec
ues to remain young and active largely divers
through Art's example and leadership. r
Art has creatively integrated the discovery COI
of design concepts, modeling strategies, and scilen
process synthesis approaches into both un- even
dergraduate and graduate teaching. His ap-
proach has been to motivate and to teach has spe
through concrete examples. This presents a decade
clear need for new methods and exposes the these i
important features of the problem as well as
open research questions that need to be ad- the n-d
dressed. Art's teaching incorporates funda- in .
mental concepts for design and synthesis, with SUp]
less emphasis on specific computer tools than designn
on a general understanding of what needs to
be done. Nevertheless, novel modeling fea- design
tures of his research (including prototypes of for a
ASCEND) have been incorporated into both ft
undergraduate and graduate courses.
Moreover, Art's teaching legacy is evi-
denced in two texts: the widely distributed work on process
simulation (Westerberg, A.W., H.P. Hutchison, R.L. Motard
and P. Winter, Process Flowsheeting, Cambridge University
Press, Cambridge, England, 1979) and the recent design text
(Biegler, L.T., I.E. Grossmann, and A.W. Westerberg, Sys-
tematic Methods of Chemical Process Design, Prentice-Hall,
Englewood Cliffs, NJ, 1997)
Finally, Art's legacy as a graduate mentor can be seen in
the education of thirty-seven PhD students; six of them have
pursued academic careers (see Table 1). Needless to say, he
has strong links with all of his former students and they view
him as an example of outstanding scholarship.

Aside from being an intellectual leader of process systems
engineering and contributing to research and education, Art
has been influential in the chemical engineering profession
itself. He has been active in the AIChE, serving on a number

e te



Chemical Engineering Education

of committees (e.g. CAST Division, Awards) and teaching
short courses. He was one of the founders of CACHE, and
he co-chaired the second conference on Foundations of Com-
puter-Aided Process Design, a meeting that now takes place
every five years. He was also a member of the National
Research Council Committee on Chemical Engineering Fron-
tiers, heading the panel on Process and Control Engineering.
Art has given a large number of seminars in chemical
engineering departments, many of them named lectureships.
He also serves as member of the editorial board
on several journals (I&EC Research, Comput-
Sa ers and Chem.Eng., Chem.Eng. Reviews, AI-
Oam of Edam, Research in Engineering Design, JOTA)
as well as on the visiting committees at Florida,
ies Princeton, and Wisconsin. His interactions with
ers, industry have also been extensive, both in con-
ter suiting and research projects.
;, and
ts), Art
Art's pioneering efforts in research and edu-
the last cation have not left him isolated from the finer
holding things of life. He and his wife of thirty-five
as into years, Barbara, share a love of music and are
frequent visitors to the Pittsburgh Symphony.
system Moreover, Barbara, an accomplished oboist
r to and pianist (trained at Oberlin College and the
the University of Florida) organizes annual musi-
of the cal soirees that have enlivened and enriched
the lives of many of their colleagues at CMU.
oces Art and Barbara's sons, Ken and Karl, are
oject continuing in Art's footsteps, with a PhD in
I. chemical engineering (University of Wash-
ington) for Ken and a PhD in Physics
(Princeton) for Karl. Both maintain strong con-
nections to chemical engineering and are pursuing careers in
mathematical modeling and design.
Art remains active in athletics. He is an avid skier and a
competitive racketball player. Moreover, his interest in per-
sonal electronic devices is legendary among his friends. This
can be traced back to his mastery of sophisticated, multiscale
slide rules as a teenager, as well as to the purchase of a
mechanical calculator for his fraternity house when he was
an undergraduate. More recently, Art has awed his col-
leagues with his expertise on palmtop computers as well as
software and sophisticated operating systems for truly indi-
vidualized computing.

When Art turned sixty last fall, it dawned on many of his
colleagues that the birthday marked a milestone not only in
Art's career but also in the evolution of process systems
engineering as a major research area in chemical engineer-


Mini-Symposium Program
Art Westerberg's 60th Birthday
Carnegie Mellon, November 24, 1998

9:00 AM An Overview of Art Westerberg's Contributions
George Stephanopoulos, MIT
9:40 AM A Further Contribution from Art Westerberg
Karl Westerberg, Princeton
9:55 AM Art Westerberg's Work in Process Flowsheeting
Rodolphe L. Motard, Washington University
10:15 AM Art's Contributions in DRC and EDRC
Steven Fenves, Carnegie Mellon University
10:35 AM BREAK
Benjamin Allan, Sandia National Lab
11:15 AM SPLIT
Oliver Wahnschafft, ASPEN Technology
11:35 AM n-dim
Eswaran Subrahmanian, Carnegie Mellon University
1:00 PM Polymer Flow
Andrew N. Hrymak, McMaster University
1:30 PM Remarks from a Former Colleague
Fritz Prinz, Stanford University
1:45 PM Remarks from Former Dean
Herbert Toor, Carnegie Mellon University
2:00 PM Closing Remarks
Ignacio E. Grossmann, John L. Anderson

Colleagues, friends, andformer students who joined Art for his surprise 60th birthday celebration
and symposium at Carnegie Mellon.

ing. The event was marked with a memorable celebration at
CMU-a complete surprise to Art. Along with a gathering
of his colleagues, friends, and former students from far and
wide, the event included a symposium sponsored by the
Chemical Engineering Department. As shown in Table 2,
participants included former students, collaborating authors,
and researchers as well as colleagues at CMU.
This milestone represents not only Art's legacy but also a
continuation of an exciting research area. Our hope is that
Art will actively participate in the continued evolution of the
area for many years to come. Therefore for all of Art's
contributions in
Defining the core of process systems engineering
Testing the boundaries of the definition
Enriching our approaches
Expanding the scope with a multi-disciplinary context,
and in serving as a

Profound Analyst
Creative Synthesist
Teacher par Excellence
Challenging and Inspiring Advisor and Mentor
Innovative Founder
Excitable Hacker
And Valued Colleague

we acknowledge a debt of gratitude. For those that have
come know him, Process Systems Engineering will always
be at the state of the Art.

1. Sargent, R.W.H., and A.W. Westerberg, "'SPEED-UP' in
Chemical Engineering Design," Trans. Instn. Chem. Engrs,
Vol 42, T190-T197 (1964).
2. Nishida, N., G. Stephanopoulos, and A.W. Westerberg, "A
Review of Process Synthesis," AIChE J (Journal Review),
27(3), 321-351 (1981) 0

Spring 1999

M department

The Chemical Engineering Department at Rose-Hulman Institute of Technology.


Institute of Technology

Rose-Hulman Institute of Technology Terre Haute, Indiana 47803

Rose-Hulman Institute of Technology is one of the
few private colleges for undergraduate engineering,
mathematics, and science in the United States. It has
earned its reputation as one of the nation's leading indepen-
dent colleges because of its educational philosophy focusing
on small classes, dedicated faculty, and an innovative cur-
riculum, all supported by modern educational facilities. The
campus is located in a suburban area about five miles east of
Terre Haute in west-central Indiana.
The college's 1998 freshman class had a combined SAT
average of 1350, with half of the students having achieved
700 or better on the math portion of the standardized test.
Copyright ChE Division ofASEE 1999

More than 90% of the students at Rose-Hulman graduated in
the top fifth of their high school class. Fall enrollment in
1998 was 1,749 students.
Undergraduate degrees are awarded in applied optics,
chemical engineering, chemistry, civil engineering, com-
puter engineering, computer science, electrical engineering,
economics, mathematics, mechanical engineering, and phys-
ics. Master's degrees can be earned in biomedical, chemi-
cal, civil, electrical, environmental, and mechanical en-
gineering as well as in applied optics and engineering
There is also an engineering management graduate pro-
gram designed to help engineers who want to enhance
Chemical Engineering Education

their management skills for use in technol-
ogy-based businesses.
Rose-Hulman prides itself on offering outstand-
ing personal attention to the needs of its stu-
dents, which is illustrated by our 12-to-1 stu-
dent-to-faculty ratio. Rose-Hulman has been hon-
ored in the prestigious Hesburgh Award compe-
tition that recognizes a select group of colleges
for exceptional efforts to improve undergradu-
ate education.
Faculty have been innovators in the use of
computer-aided instruction and in developing
ways to improve the freshman curriculum. In
1995, the Institute was among the first colleges
to require all new students to purchase laptop
Special programs offer Rose-Hulman faculty
and students an opportunity to work as teams
and to use the latest technology to help business
and industry create new products, processes, and
services. The Technology and Entrepreneurial Development pro-
gram is creating a model for project-based engineering and science
education. The program increases the number of students and
faculty involved in industry-sponsored, projects-based programs
and creates new laboratories for product and process development.
Rose-Hulman offers a unique commitment to the humanities
within an engineering, mathematics, and science curriculum. Stu-
dents can earn a minor in East Asian Studies, and they are offered
language courses in Spanish, German, and Japanese.
During the 1998-99 academic year, Rose-Hulman is celebrating
the 125th anniversary of its founding. The college was known as
Rose Polytechnic Institute from the time it was founded in 1874
until 1971, when the name was changed to Rose-Hulman Institute
of Technology in recognition of more than 100 years of support by
the Hulman family. In the fall of 1995, Rose-Hulman became a
coeducational campus, ending its 121-year history as an all-
male institution.

Top Photograph: A view of the Rose-Hulman
Center Photograph: Tubular flow reactor in the
unit ops lab high-bay area (senior student,
Thu Vu Pham).
Bottom Photograph: Classroom scene showing
students using laptop computers.

Spring 1999

The college's
1998 freshman
Class had a
combined SAT
-average -f
1350, with half
of the students-
achieved 700 or
better on the
math portion of
test. More than_
90%/ of the
students at
graduated Mi
the t-p fth Of-
their high
school l ass.
_^ __ __ ^ _

It has been reported that the nation's first four-year cur-
riculum in chemical engineering was announced by M.I.T.
in 1888.1[ But, "Professor Hammond pre-
sented a paper on 'Promotion of Engineer-
ing Education in the Past Forty Years' at
the fortieth anniversary meeting of the So- Rose
city for the Promotion of Engineering Edu-
cation. In this paper, Professor Hammond pride
stated that after searching the early records Oj
and catalogues, it did seem that Rose Poly- outs
technic had actually had the first chemi-
cal engineering graduate in the United pe
States."12] Walter Brown Wiley entered attenI
Rose Polytechnic in September 1885 and nee
graduated from the Chemical Depart- studle
ment in 1889.131 "Mr. Wiley is the first s
graduate in the chemical course from iS illU
the Rose Polytechnic Institute and has OUr
been engaged in a special line of work stu
in connection with fuel engineering, es-
pecially to improve the quality of coke facUlt3
and the investigation of coking coals."'3 W
The Chemical Engineering Department c. GO
is the third largest department at Rose- /cl i
Hulman, with approximately 250 students
at the present time. StrO g
According to Dr. Warren Bowden, there COf
were sixteen to eighteen undergraduates and
per class and three graduate students when po
he joined the department in 1956. The labo- at
ratory provided basic experiments in the
areas of filtration, evaporation, distillation, stu
and heat transfer. The equipment was old,
rusty, and in marginal working order, which
made it difficult for students to obtain
meaningful data from their experiments. The courses were
demanding. Textbooks such as Brown's Unit Operations
and Weber and Meisner's Thermodynamics for Chemical
Engineers were used. The students included some extremely
talented individuals. For example, the class of 1957 included
Ernest Davidson, Glen Miles, and Toby Eubank: Davidson
has had a very successful career as a professor of chemistry;
Glen Miles obtained a ScD at MIT and had a successful
career in industry; and Toby Eubank received a PhD at
Northwestern and has been a professor of chemical engi-
neering at Texas A&M University for many years. These
students did not get sheepskin diplomas since Rose Poly
switched to paper diplomas in the late 1920s.
The department went through a period of low student
enrollment during the early and mid '60s. Then, in 1966, Dr.
Sam C. Hite, Chairman of Chemical Engineering at the
University of Kentucky, went to lunch with a recruiter for

s its

ds (
e w


Commercial Solvents, located in Terre Haute, and was told
that chemical engineering was about to be discontinued at
Rose. He learned that the chairman had already left, and only
two professors (Warren Bowden and Tony
Blake) remained. Sam was interested, so
he applied for and was made chairman of
the Department. He immediately began a
lian drive to increase the faculty from two to
ielf on eight, the BS ChE degrees from 16 to over
!ng 70 per year, and to aggressively find money
ding for new equipment and facilities. The plan
began with the recruiting of Dr. Noel E.
HaI Moore from Kentucky. Noel later followed
to.the Sam as chairman at Rose-Hulman and
of ifs served as chair until 1997, when he stepped
down to prepare for his retirement in 1998.
which Hossein Hariri is the current chairman of
rted by the Department.
fio.. ..
There have been evolutionary changes
at Rose-Hulman and in its Chemical Engi-
1e to neering Department. Historically, the In-
fin a stitute has perceived a constant need to
-i. restructure and renew curricula and has
idng sought the necessary equipment and en-
ment courage faculty to develop new curricu-
give lar materials based on the availability of
ial additional resources. This cycle of discov-
ery, initiation, acquisition, dissemination,
I- tO and integration is one of the Institute's
itS. greatest strengths. The relatively small size
of the Institute, the dedication of its fac-
ulty, and its outstanding student body make
Rose-Hulman a recognized leader in cur-
ricular innovation in undergraduate engineering, science,
and mathematics education.
Approximately one-quarter of the incoming freshmen par-
ticipate in the Integrated First-Year Curriculum in Science
Engineering and Mathematics. In a 12-credit "super course"
during each quarter of the freshman year, students receive
instruction in calculus, physics, chemistry, computer sci-
ence, engineering graphics, and engineering design in a
block-scheduled sequence of carefully coordinated ac-
tivities that emphasize the interrelationships between the
disciplines. Students receive a single grade for the course
each quarter.
A central component of this course is an array of quarterly
projects developed by teams comprised of three or four
students. The following is a partial list of projects chosen by
students in the spring quarter of the 1997-98 academic year:
writing a program for Windows-based scheduling of final
Chemical Engineering Education

4 Senior students
Joe Lathey and
Bill Morphew
shown with the

exams; devising a method for determining the speed of a
fast-moving object whose trajectory is not known; devising
and carrying out an original experiment using a wind tunnel
of the group's own design; investigating how variations of
reactant ratio affect the properties of the plastic you create;
synthesizing a ferrofluid and purifying it to improve perfor-
mance. At the end of each quarter, students present their
projects in a large poster session open to all faculty and
students at the Institute.
One notable characteristic of Rose-Hulman is an interest
in laboratory experiences for the students and a teaching
emphasis leaning toward the practical side of the practical-
theoretical spectrum. "If football teams were coached the
way engineering students are educated, the players would all
sit on the bench reading the play book," says Dr. Sam Hulbert,
Rose-Hulman President.
Academic programs that implement our vision of student
research and engineering design and discovery activities
occur both in departmental courses and interdisciplinary ac-
tivities. In the case of the engineering disciplines, such ac-
tivities are generically known as "project work."
Project-based education was a key recommendation of the
Task Force on Design and Research of the Commission on
the Future of Rose-Hulman. The Commission is a national
410-member group of volunteers from business and indus-
try. It developed 105 recommendations to help Rose-Hulman
Spring 1999

A Seniors Amy Gainey and Jeremy Conner hard
at work on a process control experiment in the
Unit Operations Laboratory low-bay area.

4 The Unit Operations
Laboratory occupies a high-
and a low-bay area as well
as three separate rooms.
Junior Craig Clark works
with the Othmer Still Unit in
one of the separate labs.

maintain its engineering and science educational leadership
into the 21st century. In order to implement the recommen-
dations of the Commission on the Future, the number of
industry-based projects must be significantly increased, spe-
cifically those that benefit from multidisciplinary expertise
applied to industrial projects. Some of these projects could
be used as freshman design projects or as senior engineer-
ing-design projects, or as graduate thesis projects. It is our
intent that each and every Rose-Hulman student should have
a project-based experience.
Until recently, the only type of project work involving
undergraduate students in a significant way has been course-
work projects. They are projects done by a student or a
student team, with the team managed solely by the students,
as an academic exercise for academic credit. The faculty
member is not part of the team. On the other hand, R&D
projects are done by a team composed of undergraduate
students in addition to faculty members.
The U.S. Department of Energy has awarded Rose-Hulman
a $6.7 million grant to construct a 35,000 sq. ft. John T.
Myers Center for Technological Research with Industry.
This two-story facility will provide 7,500 sq. ft. of floor
space dedicated to the W.M. Keck Foundation Laboratories
for Research with Industry. An additional 10,000 sq. ft. will
be dedicated to flexible laboratory space to support student
research projects. Electronics and mechanical shops, a pre-

sentation room, conference room, and adminis-
trative offices will complete the facility.
As part of engineering design in the Chemical
Engineering Department, undergraduate student
project teams have worked on a number of in-
dustrially sponsored projects. One example is
working in conjunction with Siemens Automo-
tive to develop a non-pyrotechnic test to au-
thenticate simulations and steady-state tests in
plastic manifolds.

The Unit Operations Laboratory has a long
history of being an integral part of the under-
graduate chemical engineering program. This is
in keeping with the conviction that students
learn best by doing. In 1984, the Department
moved into a new building that was constructed
with funds donated by the Olin Foundation.
The faculty designed the new facilities around
the Unit Operations Laboratory, and some ex-
isting equipment was moved from the old labo-
ratory, but for the most part new pilot-plant-
size projects were built in the new laboratory.
One piece of equipment that was brought over
from the old laboratory (affectionately referred
to as the "Dungeon") was the Sperry filter press.
A forklift was rented for the move, but when it
was time to connect the piping, it was discov-
ered it had been set with the wrong end facing
the supply tank. Not to be deterred, Professor
Caskey promptly went to the football practice
field (this was in August) and commandeered
several hefty linemen and some steel bars. They
hoisted the filter press, turned it 180 degrees,
and returned to the football field. This type of
family atmosphere continues to be one of the
strengths of Rose-Hulman.
The laboratory has been continuously updated
and now boasts over twenty different experi-
mental modules for unit operations lab projects.
These modules include distillation, gas absorp-
tion, liquid extraction, drying, filtration,
microfiltration, membrane separations, mixing,
heat exchangers of several types (including boil-
ing and condensation), vapor-liquid equilibria,
gas and liquid fluid flow, pumps, cooling tow-
ers, kinetics (including fermentation), process
control, and other miscellaneous modules.
The laboratory has been operated as a project
lab as opposed to a "cookbook" lab. In some
cases, projects are assigned that require data to
be taken for scale up. This has worked well

The mission

of the


is to

provide a



that will

enable our

students to

practice as



the dynamic



to appreciate



to and

respect for



and to




with boiling heat transfer, cooling towers, fil-
tration, drying, and membrane separations.
Operating the laboratory in this fashion re-
quires a large commitment from the faculty,
and in any given quarter five of the eight full-
time faculty are involved in the laboratory.
Oral reports have also been an integral part of
the laboratory. Each lab group of three gives
three oral reports that are critiqued by other
groups, and faculty are present at all oral re-
ports to "grill" the group. This again requires a
hefty commitment of time from the faculty.
The reward comes when students call back
after graduation and report they are able to
hold their own when asked to report orally on a
job assignment.
The Department also recognizes process con-
trol as an industrially important area that our
students need to understand. The Camille Com-
puter company has provided three laboratory/
pilot-scale PC-based control systems that have
been integrated with the lab projects. Our main
Corning glass distillation column, ceramic
cross-flow microfiltration unit, and process-
control pilot plant unit are fully instrumented
and are linked to the Camille systems. The
systems have been designed to also operate in
manual mode in the event of an unexpected
sensor/transmitter fault or a "student-mediated
event" that makes automatic control impos-
sible. Foxboro 761 controllers are also a part of
the process control pilot plant, allowing stu-
dents to gain experience with remote local con-
trollers as well as the PC-based systems.

In addition to the lab projects, most elective
courses also feature projects. For example, the
environmental unit operations course has a
project where the students make drinking wa-
ter from raw sewage, using the operations of
sedimentation, granular filtration, activated car-
bon adsorption, deionization, and
microfiltration. The students conduct tests to
determine the water's purity after each unit
operation. This hands-on project reinforces the
subjects studied in class. The students are in-
vited to make coffee or hot chocolate from the
final water-the ultimate test if they believe
the process really works. In the polymer engi-
neering course, students choose a plastic prod-
uct and analyze it by one or more of the tech-
niques studied in class. During the part of the
course where polymer processing is studied,
Chemical Engineering Education

groups experimentally measure the amount of a charac-
terized polyethylene from an extruder and compare the
amount extruded to the amount predicted by the equa-
tions studied in class.
A number of required courses also have projects associ-
ated with them. Our first contact with students enrolled in
chemical engineering is in the freshman year in a course
titled "Introduction to Design." The objectives of this course
are twofold: to give the students a better understanding of
chemical engineering and what chemical engineers do, and
to give the students insight into the reason for, and the
importance of, subsequent courses in the curriculum. To
accomplish these objectives, the students are given a process
patent along with relevant design data and are asked to do a
preliminary design and economic analysis for a plant using
the process. The students work in groups of three, with the
professor serving as their supervisor/consultant. The stu-
dents do the necessary material and energy balances, size
selected items of equipment, determine the equipment cost
and the total capital investment required, and determine the
total product cost and the return on investment. Obviously,
the process must not be complex, and close supervision and
guidance is required. While the students feel that a lot is
demanded of them in the course, they also feel that it is
worthwhile and accomplishes its objectives.
In the sophomore year, students take a two-quarter se-
quence in material and energy balances. The capstone of this
sequence is a case-study project done in teams of three
completing a material and energy balance over a process
supplied to each team. Students are given assistance
through information on the course web page. They learn
engineering methods in solving a case-study problem
and are able to improve their computer skills as well as
their skill in writing reports.141
Our materials engineering course has for some years re-
quired student teams to participate in a poster presentation,
and the teams now have the option to develop a web page on
some aspect of materials. Projects are also a requirement in
the air pollution control course. The most ambitious project
in this course was the work of a group that exposed pregnant
rats to varying doses of sulfur dioxide and then examined the
offspring for evidence of damage to internal organs.

Rose-Hulman has placed increasing emphasis on curricu-
lar integration. As mentioned previously, about 25% of the
freshmen take a program that emphasizes interrelationships
between disciplines. The mechanical and electrical engi-
neering departments have continued this trend into the sopho-
more year, emphasizing the common principles of conserva-
tion-conservation of charge, energy, mass, and momentum.
The chemical engineering department restructured the
sophomore material and energy balance sequence to include
Spring 1999

this same emphasis. The restructured courses will be offered
in the fall of 1999 and are titled "Conservation Principles
and Balances" and "Basic Chemical Process Calculations."
The first course includes an introduction to engineering cal-
culations, application of numerical techniques, concepts of
systems, conservation, and accounting of extensive proper-
ties (mass, energy, charge, linear, and angular momentum)
as a common framework for engineering analysis and mod-
eling. The second course offers the application of conserva-
tion of mass and energy in analysis of chemical engineering
processes including recycle, bypass, and multi-stream pro-
cesses as well as methodologies used by practicing chemical
engineers. The use of computer software, especially spread-
sheets, is highly integrated into the course.
The Department has also developed another tool to em-
phasize the interrelationships of the sophomore-, junior-,
and senior-required chemical engineering courses. This is a
CD-ROM developed by Professor Caskey using a saturate
gas plant from Marathon's refinery in Robinson, Illinois.
Modules have been made for material and energy balances,
fluid mechanics, heat transfer, thermodynamics, mass trans-
fer, and process design. This CD-ROM provides a resource
for linking subjects regardless of the textbook or teaching
method used in any particular course. A student can perform a
material balance of a multicomponent absorption column in
material balances (in the sophomore year), complete a vapor/
liquid calculation on the same column in thermodynamics (in
the junior year), find heat loss from the column in heat transfer
(in the junior year), and calculate the number of stages required
in mass transfer (in the senior year). Students can use this tool
to get a sense of how courses are interconnected.

The Department faculty roster went through a change last
year when three new faces replaced faculty members who
were retiring. Mentoring and passing on the experience and
tradition of the Department to the new faculty members are
now important tasks that lie ahead. We will continue to
maintain a strong teaching commitment and to give personal
attention to students. The mission of the Department is to
provide a balanced education that will enable our students to
practice as professionals in the dynamic industrial environ-
ment, to appreciate their responsibility to and respect for
their colleagues, and to become life-long learners.

1. Mattill, John, "M.I.T.'s School of Chemical Engineering Prac-
tice," Chem. Eng. Ed., 27(3) (1997)
2. Bloxsome, John, Rose: The First One Hundred Years, Rose-
Hulman Institute of Technology, p. 119 (1973)
3. Rose Polytechnic Institute 1874-1909 Memorial Volume,
Monford & Co., Typographers, Cincinnati, OH; p. 140 (1909)
4. Hariri, Hossein, "A Case Study in Stoichiometry Course
Using Excel and Power Point Presentation," ASEE Annual
Meeting, Milwaukee, WI (1997) 0

We extend our appreciation to Joseph Shaeiwitz (West Virginia University) for acting
as Guest Editor in compiling, reviewing, and editing the following papers that com-
prise a special-feature section on outcomes assessment in this issue.

Commentary ...


Its Time Has Come

Teaching is no longer sufficient;
student learning must be demonstrated.

West Virginia University Morgantown, WV26506-6102

When faced with the necessity of developing an
assessment plan for regional accreditation about
seven years ago, my initial response was directly
from the "textbook of faculty responses when confronted
with doing something new." I was not enthusiastic about
having another task to fit into an already-full schedule.
If you have the same attitude, or if you are faced with a
colleague reciting all of the "textbook responses" as excuses
for not doing outcomes assessment, consider this: all faculty
members reading this article have been doing outcomes as-
sessment for their entire careers. When mentoring graduate
students, the goal is to develop a set of skills to permit the
MS or PhD to do research on any topic; the mentor (con-
sciously or unconsciously) constantly compares the student's
acquired skills to the desired skills and focuses on making
certain that all of the desired skills are acquired before a
degree is granted. Assessing undergraduates involves the
same process-plus the additional challenge of dealing with
a larger number of students.
The concept of outcomes assessment has been around for
about thirty years. It is becoming more prevalent in higher

Joseph A. Shaeiwitz received his degrees in
Chemical Engineering from the University of
Delaware (BS, 1974) and Carnegie Mellon Uni-
versity (MS, 1976; PhD, 1978). He is currently
Associate Professor of Chemical Engineering at
West Virginia University. His research interests
i are in design, design education, and outcomes
assessment. He is coauthor of the text Analysis,
Synthesis, and Design of Chemical Processes,
published in 1998 by Prentice Hall.

education because of its acceptance by regional accredita-
tion agencies and by professional accreditation agencies such
as ABET, and because of the demand for more accountabil-
ity by boards of trustees and state legislatures.
With the adoption of EC 2000, engineering programs are
struggling with how to develop meaningful assessment plans.
The following collection of papers is designed to assist with
the development of those plans. The papers contain opinions
and observations from those involved on both sides of the
first few EC 2000 pilot visits in chemical engineering, those
involved in educating engineers on outcomes assessment,
and those who have experience in developing and imple-
menting assessment strategies for courses and curricula that
can serve as models for other assessment plans.
Some may wonder why we should be doing outcomes
assessment. After all, if our students graduate and get good
jobs or attend graduate or professional school, we are doing
fine. Right? But are employers really satisfied with the skills
of the graduates they hire? Are students really satisfied with
the education they received? Do graduates really possess the
knowledge and skills faculty believe they possess?
How do we know the answers to these questions?
Outcomes assessment allows faculty to begin to answer
them. With outcomes assessment, teaching is no longer just
the act of showing up in class and simply giving lectures,
assignments, and exams. Teaching now includes setting goals
for student learning in a course and/or curriculum, and tak-
ing responsibility for students achieving those goals. Teach-
ing now involves determining those goals in consultation
with constituencies such as employers and students. The
curriculum being taught should now have opportunities for

Copyright ChE Division of ASEE 1999

Chemical Engineering Education

ABET Engineering Criteria 2000)

development of desired skills over time. Teaching now in-
volves evaluating whether students have achieved the de-
sired skills and knowledge and modifying courses and cur-
ricula to ensure that the goals are achieved (continuous qual-
ity improvement). Teaching is now based
on an output model, i.e., measuring the
knowledge and skills of the graduate and
the student moving through the curriculum, In rec
not on an input model where the output is
assumed based solely on curriculum con- We hi
tent. Teaching is no longer sufficient; stu- seven
dent learning must be demonstrated, most
Among the results of implementation of re
an assessment plan are faculty who know
their students-who know their students' Uni
individual strengths and weaknesses and who procla.
are in a position to help individual students that td
exploit those strengths and overcome those
weaknesses. In recent years, we have heard dvt
several of the most prestigious research uni- to UInd
versities proclaim publicly that they plan to edi
devote more time to undergraduate educa- O
tion. Outcomes assessment provides them
with the perfect opportunity to make good aSS
on their promise. proVl
If we agree that the skills and knowledge With t
that students obtain in the curriculum are opp
what is being assessed, the next question is P
how to do such an assessment. This requires malk
an assessment plan with three main elements. their
First, goals must be set, preferably in con-
sultation with external constituencies. A sub-
set of enumerating goals is to determine
performance, i.e., what a student must do to achieve a goal.
Second, multiple measures of achievement of these goals are
necessary. Among the most common measures are inter-
views with students, alumni and employer questionnaires,
portfolios, and capstone projects. Finally, the results ob-
tained from these measures must be used to improve the
educational process.
The papers that follow describe assessment plans that all
use the methodology described above, but each illustrates a
different set of measures. A valid assessment plan need
not and should not contain all of the elements of all of the
plans described in these papers. Over-assessing is as bad
as under-assessing.
Another key question is: When should assessment be done
in the curriculum? The answer is: At multiple points within
and beyond the curriculum. Graduates should be tracked to
obtain a self-evaluation of their education. Employers should
be surveyed to obtain an evaluation of their employees'
Spring 1999

education. Students about to graduate should be assessed
(self-evaluation and faculty evaluation) to determine whether
they have achieved the desired outcomes.


in the curriculum should also be assessed to
determine if mid-course corrections are
needed. Assessment can be done in a single
course, as described in one of the papers
that follows, or classroom assessment can
be done to determine the outcomes of a
single class.111

Finally, who should do the assessment? It
is tempting to say "all faculty." While that
would be an ideal situation, it may be unre-
alistic-although all faculty teaching under-
graduate classes really should be involved
with the undergraduate assessment plan. It
is important, however, to avoid the opposite
extreme! Assessment cannot be the job of
one faculty member, or even worse, solely
the job of an external consultant. There must
be a critical mass of faculty involved in
the assessment process to ensure its con-
tinuity and to ensure that all parts of the
curriculum are covered.
So, now you are enthusiastic about as-
sessment! And, after reading the following
papers, you will want to learn even more
about it. Where do you go? One place is to
the Best Assessment Processes in Engineer-
ing Education conference held at Rose-
Hulman Institute of Technology. Another is
to a web page that I have created (found at

come/index.html) that includes the assessment references I
have found most enlightening.
It should be noted that many faculty are uncomfortable
with outcomes assessment, especially using it for accredita-
tion purposes. We have become comfortable with the much-
maligned "bean counting" process. As noted in several of
the following papers, creating and implementing an assess-
ment plan involves finding one good solution to an ill-
defined problem. At the design level, that is what we as
engineers are trained to do. So, by implementing an assess-
ment plan, we are simply demonstrating to our students that
we can solve the type of problem we expect them to solve in
the capstone course.

1. Angelo, T.A., and K.P. Cross, Classroom Assessment Tech-
niques: A Handbook for College Teachers, 2nd ed., Jossey-
Bass, San Francisco, CA (1993) O

'ent years,
ove heard
ral of the
im publicly
ey plan to
More time
ides them
*he perfect
rtunity to
e good on

ABET Engineering Criteria 2000




Iowa State University Ames, IA 50011-2230

Adoption of ABET Engineering Criteria 2000 by the
Accreditation Board for Engineering and Technol-
ogy, along with a revised set of Program Criteria for
Chemical Engineering by AIChE, will certainly have a sig-
nificant effect on the education of chemical engineering
graduates as we move into the 21st century. There is a
profound shift in emphasis, from requiring a specified cur-
ricular content to evaluation of programs for the success
they demonstrate in meeting their own goals for how and
how well they prepare their graduates. This, along with how
well programs continuously assess and improve their pro-
cesses to achieve these goals, is a major "sea change" for
engineering accreditation. A concomitant increase in flex-
ibility, which should allow chemical engineering programs
significant latitude in meeting their objectives, is perceived
by many, particularly those in industry and in higher aca-
demic administration, as being long overdue. There is, how-
ever, a significant price to be paid.
For the first time in three decades, many chemical engi-
neering programs are now thinking carefully about the goals
and execution of their undergraduate curricula. The typical

Stanley I. Proctor received his BSChE, MS,
and DSc in chemical engineering from Wash-
ington University. He is a past president of
ABET, a past president (and Fellow) of AIChE,
and has recently been named chair of AIChE's
new Career & Education Operating council.
Since his retirement in 1993 he has been in
private consulting.

Richard C. Seagrave is currently a Distinguished Professor of Chemical
Engineering and Interim Provost at Iowa State University. He is a past
chair of the Engineering Accreditation Commission of ABET and is vice-
chair of AIChEs Career and Operating council. He is also an AIChE
Fellow. (Photograph not available.)

Copyright ChE Division of ASEE 1999

curriculum that has been in place since the 1960s, an amal-
gam of our petroleum- and industrial-commodity-chemistry
based past and the transport-phenomena revolution of the
1960s, now faces new challenges. An increasing number of
our graduates find employment in the biologically based
industries (food, textiles, agricultural byproducts, pharma-
ceuticals, biomedical industries) and in the information-pro-
cessing industries (microchip manufacturing, solid-state pro-
cessing, software development). Many graduates find em-
ployment as financial analysts, seek careers in law and medi-
cine, and embark, as engineers, on a wide variety of career
paths. The faculty in charge of undergraduate curricula, who
have put their trust in a prescribed and somewhat narrow set
of courses (usually organic and physical chemistry, fluid
mechanics, heat and mass transfer, chemical reactor design,
thermodynamics, process control, and design) are now be-
ginning to think seriously about the role of subjects such as
biology, solid-state chemistry and physics, new materials,
nanotechnology and mega-systems, etc., as they relate to
their goals for their graduates. They are becoming more
concerned about how elements of the curriculum fit together
and support each other in the educational process.
ABET Engineering Criteria 2000 attempts to provide a
framework by which chemical engineering faculty members
can develop their programs to achieve these desirable evolu-
tionary changes without jeopardizing their accreditation
standing, while at the same time requiring them to change
their emphasis from what is taught to what is learned. This is
no small feat. The new criteria essentially requires that chemi-
cal engineering programs address three basic questions:
0 Within the context of chemical engineering, what are your
objectives for your graduates, and how did you and your
constituencies set them?
1 How do you determine if your objectives are being met?
0 What are you doing to fix things if your objectives are not
being met, or improve things even if they are?
Chemical Engineering Education

ABET Engineering Criteria 2000

After a considerable amount of discussion and public
input, the Engineering Accreditation Commission agreed on

a basic set of attributes (outcomes) which should
be required of all engineering graduates. This list
of 11 attributes (a through k from Criteria 3)
forms the minimum "experience base" that the
profession accepts as necessary attributes (or out-
comes) for all engineering graduates. It is ex-
pected that individual programs will supplement
this list, which is not intended to be exhaustive.

The chall
Criteria 20

The "knowledge base" that is required by the evc
general criteria for all engineering graduates has meet
been considerably modified in the new criteria. goals
For example, neither courses in physics nor chem-
istry are specifically required. Specification of a oppo
minimum amount of social sciences and humani- Those
ties courses is no longer stated. The appropriate wh
coursework will still need to be prescribed by the
faculty, with an eye towards fulfilling the at- constr
tributes stated in EC 2000 and the program crite- ABET i
ria, consistent with the overall goals of their pro-
gram and the nature of their discipline.
AIChE, through its representatives on the Edu- OPPO
cation and Accreditation Committee, has taken a expe
conservative approach in proposing its new out-
comes-based program criteria. The current state of these
criteria, as well as other items of interest regarding the new
process, may be monitored on the World Wide Web at http:/
/ Note that the language addresses the re-
quirements placed on the capabilities of the graduates. No
courses are specifically required. Perhaps the most signifi-
cant change is the discontinuance of the requirement of one-
half year of advanced chemistry, which has been replaced by
a more demanding requirement of a thorough grounding in
advanced chemistry, in a list of areas of chemistry which
may be specified more precisely by the faculty itself in any
program. For many years, there has been the conventional
and wide-spread belief that either ABET or AIChE "re-
quired" both organic and physical chemistry, although this
has never been the case. Chemical engineering faculties
throughout the country have locked themselves into this
box. It will be interesting to see, now that the box has been
unlocked, what choices will be made.
Unquestionably, the greatest area of concern that has
been expressed among chemical engineering faculty and
department chairs has been an uncertainty with respect to
"what does ABET expect of us in the areas of outcomes
assessment and continuous improvement?" The short an-
swer, we believe, is that programs will need to set their own
expectations in this area, as well as in other areas. Actually,
they always have. In a profession that has placed the prin-
ciples of process control in a central place in its curricula, it
Spring 1999



should not be difficult to adapt the basic principles of mea-
surement, feedback, set-points, and load changes, to deter-
mining the degree to which their graduates are
S meeting their objectives. The public, of course,
S expects this.
d by To aid faculty and administrators in this area, as
ring well as to gain experience in the new accredita-
00 for tion process, a series of pilot visits has been com-
S to pleted. In the 1996-97 academic year, two institu-
tions, the University of Arkansas and Worcester
in Polytechnic Institute, were visited using EC 2000.
their In the 1997-98 academic year, three more institu-
S tions, Harvey Mudd College, The Georgia Insti-
tute of Technology, and Union College were evalu-
Uity. ated using EC 2000. One result of these pilot
ulties studies will be a set of case studies that should
S be useful in helping to set goals, in establishing
mechanisms of outcomes assessment, and in
ed by preparing for and participating in accreditation
e past visits. These studies are not intended to be a
"how to" set of instructions, but rather a set of
examples that have been used successfully. The
ity to first case study, for a fictitious institution,
ent. Coastal State University, is now available on
the ABET Web site at It
represents an amalgamation of experiences from
the pilot studies.
In addition, the Engineering Accreditation Commission,
in concert with the educational elements of the various tech-
nical societies, including AIChE, is developing a standard.
set of training materials for engineering program evaluators.
This course will be useful for faculty and administrators in
getting a better understanding of the accreditation process
that will accompany the new criteria. Also, with the sponsor-
ship of NSF and with the cooperation of industry, a series of
twelve regional NSF-sponsored industry-hosted workshops
for training faculty from every engineering program in the
U. S. began in late 1998 and will continue for the next three
years. Watch the ABET Web site for more information.
The challenge provided by Engineering Criteria 2000 for
programs to evolve in meeting their goals is also an oppor-
tunity. Those faculties who felt constrained by ABET in the
past now have an opportunity to experiment. Those institu-
tions whose general accreditation review was scheduled in
1998-99 academic year and is scheduled to occur in the
1999-2000 or 2000-2001 academic years have the choice to
seek re-accreditation under either the existing criteria or
under EC 2000. All engineering programs at the institution
must make the same choice. Beginning in the year 2001-
2002, all institutions will come under the new Engineering
Criteria 2000. What the future holds will be determined by
the experiences we will share during that period. 0

ABET Engineering Criteria 2000

Outcomes Assessment



Rose-Hulman Institute of Technology Terre Haute, IN 47803-3999

Why would 370 engineering faculty from 150 insti-
tutions and four countries travel to Terre Haute,
Indiana, in April? No, ski season was over. Actu-
ally, it was to attend the first "Best Assessment Processes in
Engineering Education" symposium, held on the Rose-Hulman
Institute of Technology campus. Response to the symposium
is indicative of the degree of interest in learning about assess-
ment techniques that can be applied to engineering education.
It would be wonderful to report that the primary motiva-
tion for the interest in assessment is because we are all
wildly interested in learning how we can implement continu-
ous quality improvement in our educational programs. Al-
though we are interested, we also need to answer to a multi-
tude of demands on our time and resources. In reality, the
changes in the accreditation requirements embodied in EC
2000~11 represent a new approach to validation of quality in
engineering education and are driving the interest in out-
comes assessment. Many agree that EC 2000 is the right and
appropriate approach to accreditation. But it also presents
several major challenges for each of us.
I have had opportunities to interact with faculty and ad-
ministrators from various campuses, engineering societies,
and ABET. The purpose of this article is to share my obser-
vations from the field of assessment and my experience from
interacting with those who are working to align their educa-

Gloria Rogers received her BS and MA in So-
ciology and her PhD in Educational Administra-
tion from Indiana State University. She is cur-
rently Vice President for Institutional Resources
and Assessment at Rose-Hulman Institute of
Technology. She is currently working with the
Accreditation Board for Engineering and Tech-
nology on implementation of EC 2000, and is
one of the organizers of the national sympo-
sium "Best Assessment Processes in Engineer-
ing Education."

tional processes to be consistent with both the letter and the
spirit of EC 2000.

There are three Chinese characters that make up the English
word "challenge": 1) opportunity, 2) on the wings, 3) of dan-
ger. I would like to provide what I believe to be the major
challenges facing each of us as we move to outcomes assess-
ment. These challenges will highlight both the opportunities
and the dangers associated with our transition to EC 2000.

* Understanding Assessment and the Continuous
Quality Improvement (CQI) Process
Engineering faculty recognize the importance of the use of
models in solving engineering problems. The value that a
CQI model contributes is that it gives faculty a common
language and a conceptual framework to guide the process.
Opportunity There are many models that have been
developed that depict the CQI process-including the "Two-
Loops of EC 2000"'2] and "Assessment for Continuous Qual-
ity Improvement"13 models. I have had engineering faculty
share with me copies of CQI models they have developed
that represent everything from a chemical process to an
electrical circuit. The important thing is that you develop/
adapt/adopt a model that is meaningful to you and your
program that includes all the elements of the CQI cycle.
Development of this framework will provide a common
understanding of what the process entails and will guide you
as you structure your activities.
Danger There are really two dangers in development of
a model. The first is that all the elements and relationships
that are crucial to the CQI process are not included. The
minimal elements that need to be illustrated in a model are

Copyright ChE Division of ASEE 1999

Chemical Engineering Education

The relationship between your program outcomes and your
program and institutional mission
Student learning outcome goals for your engineering
program (broadly stated, not measurable, e.g., communica-
tions skills)
Involvement of constituencies, i.e., where are your constitu-
ents involved and what is the nature of their involvement
Specific performance specifications for each learning
outcome (measurable, e.g., demonstrated ability to use
correct grammar)
Educational practices and strategies employed to provide
students with the opportunity to gain knowledge, skill, and
experience to achieve the desired outcomes
Collection of evidence (assessment) to determine whether or
not the learning outcome has been met
Evaluation of the evidence, i.e., interpretation of evidence
and recommendations for improvement
Feedback loops, i.e., what is the nature of feedback loops
and how is assessment and evaluation information used to
improve programs
The second challenge in the development of a model ap-
propriate for your program is that you spend all your time
debating the complexity and validity of the model and do not
get on with it. Like the CQI processes you will develop, the
model itself can be improved as you go through the cycle and
learn more about the process and your institutional culture.

* Use of Assessment Terminology
I call this the "Tower of Babel" effect. I recently read an
article that described the importance of language and mean-
ing in systems engineering.141 It described the problems faced
when workers with the same goal cannot communicate well
enough to accomplish the task before them.
Opportunity As in the case of scientific notation, the use
of terminology in assessment is not standardized. The terms
(goals, objectives, criteria, metrics, etc.) are often used dif-
ferently or interchangeably. This can create confusion and
alienation from the CQI process. Development of "Stepping

ABET Engineering Criteria 2000

Ahead: An Assessment Plan Development Guide"151 was
created, in part, to address this issue. But when EC 2000 was
crafted, the term "objective" was used in the way that the
guide uses the term "goal." It is important to note that there
is no one right way to define these terms. Each engineering
program/college should agree upon term definition and use
terms consistently. This provides an opportunity to focus on
the meaning of assessment terms and will, in the long run,
clarify the process and serve the program well. Table 1
demonstrates assessment term definitions'51 with examples.
It is not meant to be exhaustive of all possible combinations
or examples.
Danger The most significant danger is that there will be
no attempt to develop a common assessment language among
the key players and the process will fall apart due to confu-
sion and frustration. Different members of the community
may have strong preferences for term definition because of
their experiences. Listen carefully and bring closure on this
issue early in the process. There will be lots of battles to be
fought during the process-this is not one of them. Agree to
agree and move on. Then use the terminology consistently-
and often.

* Development of Performance Criteria
Development of specific, measurable performance criteria
is probably the most challenging-and important-step in
this process. Most of us can begin with EC 2000, Criterion 3
(the eleven desirable attributes of the engineering graduate),
to develop our student outcome objectives (goals?). These
are broadly stated, however, and cannot be measured. The
challenge of each engineering program is to define what is
meant by each of the objectives. We think we know when
students demonstrate the ability to communicate effectively,
but when faculty begin to spell out what they mean, they find
there is not always a clear consensus. In addition, if we value
"effective" communication skills, we need to tell students
what characteristics should be present in order for them to

Definitions of Assessment Terms

Term Definition Other Terms Used Example
Goal A statement describing a broad outcome; not measurable Objective Graduating students will be effective team members.

Objective Statement(s) derived from the goal that define the
circumstances by which it will be known if the desired
change has occurred; not measurable
Performance Specific, measurable statement identifying performance
Criterion required to meet the objective. The performance
criteria must be confirmable through evidence.
Objectives may have multiple criteria.

Goal When engaged in a dialogue with team members, or as part of a small
Outcome group project, students will perform effectively as team members.


1. Initiate and maintain task-oriented dialog.
2. Work for constructive conflict resolution.
3. Strive for meaningful group consensus.
4. Support other team members in the effective performance of their roles.
5. Initiate and participate in group maintenance activities

Spring 1999

(ABET Engineering Criteria 2000

demonstrate such skills. We also need to provide students
with opportunities to learn, develop, and demonstrate the
skills, and give them feedback on their progress. For this to
happen, we need to develop measurable performance criteria
that give precision to the objective.
Opportunity The exercise of developing measurable
performance criteria will provide faculty with a shared un-
derstanding of the desired outcome. It will also promote
discussion about strategies that can be implemented to give
students the experiences they need to be able to demonstrate
the outcome. The criteria that are developed will also shape the
assessment method and enable faculty to develop assessment
processes that are clearly linked to the desired outcome.
Danger This is a component of the CQI process that is
often left out. It is common for the assessment planners to
move from listing objectives to choosing assessment meth-
ods. This is understandable because the development of
measurable criteria is painstaking-critical, but painstaking.
This is where common sense must prevail. Continuing with
the "effective communications skills" example, it would be
possible to develop fifteen or more very well-defined perfor-
mance criteria for effective communications skill. If you
look at all of your learning objectives (fifteen or more?) and
each of them has ten or more performance criteria, the over-
all assessment task becomes overwhelming. Start with as
many performance criteria as you can think of for each
learning objective and prioritize them in order of impor-
tance. The final number chosen should include those criteria
that are considered to be critical to the objective and still
make the assessment task manageable.

* Use of Local Resources
Recently, I heard an engineer say, "We engineers find it
hard to believe that we can learn anything from someone
who is not an engineer!" Although this was said in jest (I
hope), there seems to be a reluctance to go outside engineer-
ing circles to ask for help in designing and/or implementing
the CQI process as it relates to education.
Opportunity It is important to capitalize on your local
resources. Many regional accreditation agencies have moved
to an outcomes-assessment-based accreditation process for
the institution. The likelihood of there already being some-
one on your campus who is charged with the responsibility
to do outcomes assessment is very great. Find them and
begin a dialog about how what they are doing at the institu-
tional level can inform and assist your program-assessment
efforts. It would also be very unusual if you did not already
have resources on your campus that could provide assistance
in areas of educational assessment design (College of Edu-
cation, Educational Psychology, etc.), data collection (Insti-
tutional Research, Registrar, Admissions, Student Affairs,

etc.), and statistical analysis of social science data (Social
Sciences, Business, etc.). Identifying local resources and
engaging them in the planning and implementation process
will provide both an economy of effort and a perspective
external to the engineering program. This is bound to
strengthen the overall quality of your assessment efforts.
Danger There is a real danger that engineering faculty
and administrators will adopt the attitude that no one outside
of engineering can possibly understand the complexity and
demanding curricula that embody the engineering discipline.
It is important to remember that what you are looking for
here are "worker" bees, not the queen. There are others
outside of engineering who can help you think through the
design of your assessment plan, ask the right questions, and
collect and analyze the data. It is the primary purpose of the
engineering faculty and administrators to give the plan sub-
stance, evaluate the results, make recommendations based
on the evaluation, and implement the improvement. All the
other steps can be done in consultation with others.

* Hiring An "Expert" To Do It For You
There are many resources available to you from within
higher education. People who are knowledgeable and expe-
rienced in assessment and evaluation processes are available
to support you in your efforts.
Opportunity A critical element in satisfying the require-
ments set forth in EC 2000 is to educate yourself in the
assessment and CQI processes as they relate to educational
programs. Having "experts" provide professional develop-
ment activities for faculty is a good way to get the process
started with a common language and understanding. If you
have no local resources, seeking consultation from outside
the college could be very beneficial.
Danger There is a temptation to hire someone with
expertise in assessment to do assessment for you (or, "to"
you). Although having someone on the staff to assist in the
process would be advantageous, there is a danger that others
would expect him or her to develop and implement the plan.
The appropriate role of an assessment specialist on the staff
would be to guide the process and work with faculty to
develop and validate their assessment plan. Determining
responsibility for collection and analysis of data should be
done in consultation with engineering faculty and adminis-
tration. Evaluation of assessment results is more appropri-
ately done by the faculty. Faculty should then recommend
changes for improvement in the engineering program based
on their evaluation of the assessment results.

* Student Involvement
We must never forget that all of this is about improving
the quality of student outcomes. It is designed to prepare
Chemical Engineering Education

ABET Engineering Criteria 2000

students for careers and lifelong learning. As the focus of
this effort, we need to find ways to involve students in the
assessment of their own learning outcomes.
Opportunity As learning objectives are moved from the
abstract to the concrete through devel-
opment of specific, measurable perfor-- _
mance criteria, students will have op- It is ti
portunities to assess their own skills in avrni
ways that are meaningful to them. For
example, the use of peer assessment lessons Ie
when students are asked to give oral thoseiWhc
reports will not only provide opportuni- engaged i
ties for them to assess each other, but
will also reinforce the characteristics that
are important for oral communication to COntextS Qn
the student who is making the assess- tof-ig.
ment. The feedback being given to stu- educati
dents making the reports will help them takef a
know where they need to make improve-
ments. This can be done within the con- the opp
text of an engineering class. provided
Danger The process of assessment -pprl
for continuous quality improvement is accreditat
designed to help us improve our engi-
neering programs. We cannot forget that
we can only improve our programs if The dMan
we improve the educational outcomes --
for individual students. There is a dan- Can be
ger that they will be left out of the pro- if We ant
cess. Statistics are the impersonal repre- learn ft
sentation of a collection of personal ex- who hi
periences of individual students. Let us
accept the challenge of getting them in-
volved in the assessment of themselves
and their peers. Who knows? This act

alone may be the most significant improvement in our pro-
grams and have the greatest impact on student outcomes.

* "One Size Fits All" Mentality
The assessment process is like the engineering design
process in many ways. One of the most significant ways is
that it is a process that is ambiguous-there is no one right
answer. Some answers are better than others, and some
answers are definitely wrong.
Opportunity Although CQI models can provide a good
starting point, development of an assessment plan should
reflect the uniqueness of your institution, your student body,
and your program. The move to outcomes assessment re-
quires conversations about who you are and what outcomes
you want for your students-not someone else's students,
institution, or program. These conversations should contribute
Spring 1999


- -

n o




to shared definitions and understandings that will enhance the
overall educational experience for students and faculty alike.
Danger Because of the sense of urgency that we all feel
to get moving on the development of our assessment plans
and data collection, there is a danger that
we will try to impose someone else's
o take framework or methods to our own pro-
. ff gram. There is a real risk in this approach
f because of the lack of personal buy-in
Oedff from the people who are going to be re-
ive been sponsible for implementation, evaluation,
UtcOBmes recommendations, and improvements-
--di' I -your faculty. Again, reviewing the work
of others is very positive. There will also
pjfy them be things that you will be able to adapt/
Baing adopt for use on your campus, but those
... we decisions need to be made after you have
.nto o of developed a clear understanding of your
t ge learning outcomes objectives and perfor-
nuties mance criteria for your program. Not un-
the new til you reach these understandings will
h to you be able to determine what methods
to -lssess will best fit your program.
S0a whole. WILL EC 2000 SURVIVE?
.L. o The long-term impact of EC 2000 will
SAdoig;SO depend on several factors. As I have talked
oided to engineering faculty from around the
lling to country, I have found mixed emotions about
others whether or not the changes will bring about
n real, significant improvement in the way
een engineering education is delivered and the
quality of learning outcomes for students. I
believe there are four elements critical to
the successful transition to EC 2000.

1. Faculty must believe that EC 2000 will promote student
learning and not be adverse to their own academic agenda.
Many faculty agree that EC 2000 is the "right" thing
to do. They are in general agreement that the
previous criteria were too restrictive and irrelevant to
the changing nature of the engineering profession.
But EC 2000 represents a radical, untested departure
from what was a familiar and "comfortable" pro-
cess-although unpleasant. The new approach to
accreditation will take time and energy before any
"payoff" will be seen. Even where EC 2000 is
embraced, faculty want to know what they will have
to give up in order to comply with the requirements.
Unless they can see the long-term, beneficial results
of their efforts, it will be difficult to get their buy-in.
Continued on page 115.

(ABET Engineering Criteria 2000



Colorado School of Mines Golden, CO 80401

At the Colorado School of Mines (CSM), we have
been assessing students outcomes (both in the core
and in the major) for over a decade. From the begin-
ning, we chose to use portfolios as the major component of
our assessment plan. In this paper, we will briefly describe
the history of our assessment program, sketch the assess-
ment process that we use, list the advantages and disadvan-
tages of portfolios, and then focus on the use of portfolios as
a major instrument of chemical engineering program assess-
ment and evaluation.

During the late 1980s, the Colorado legislature (like those
in many other states) became interested in higher-educa-
tional accountability and assessment and passed legislation
requiring the Colorado Commission on Higher Education to
"develop an accountability policy and report annually on its
implementation." Colorado allowed each institution to de-
velop an individual assessment plan appropriate for its size,
student body, mission, and goals. CSM chose to develop a
portfolio assessment plan, for which we have now been
collecting and evaluating data since 1988.
Barbara M. Olds is Principal Tutor of the McBride
Honors Program in Public Affairs for Engineers
and Professor of Liberal Arts and International
Studies at the Colorado School of Mines, where
she has taught for the past fifteen years. She is
chair of CSM's assessment committee and has
given numerous workshops and presentations
on assessment in engineering education.

Ronald L. Miller is Associate Professor of
Chemical Engineering and Petroleum Refining
at the Colorado School of Mines, where he has
taught chemical engineering and interdisciplinary
courses and conducted research in educational
methods for thirteen years. He is chair of the
Chemical Engineering Department Assessment
Committee and acting chair of the CSM Assess-
ment Committee.
Copyright ChE Division of ASEE 1999

Of course, over the past decade there have been a variety
of external drivers for assessment in addition to legislatures,
including the total quality management model in industry,
regional accrediting agencies (North Central in our case),
and most recently, ABET Engineering Criteria 2000 (EC
2000). Colorado, like many other states, has recently begun
to emphasize performance indicators rather than assessment
programs for accountability, so our assessment focus at CSM
has shifted to satisfying our own internal needs for continu-
ous improvement and to providing evaluation information to
external constituencies, including ABET. Since ABET ac-
credits programs rather than institutions, our assessment fo-
cus has become more diffuse, shifting to departments-but
we continue to use portfolio assessment to measure CSM
core curriculum outcomes, and several of our programs
(including chemical engineering) are building on the ex-
isting portfolio plan in developing their individual as-
sessment processes.
For the institution-wide portion of our assessment plan, a
random sample of incoming students is selected each year
(approximately ten percent of the first-year class) for whom
we develop portfolios. We collect typical quantitative data
for each student, suchlas SAT and ACT scores and GPAs. In
addition, we include in the portfolios samples of classroom
work from a variety of courses as well as surveys and other
feedback on the students' satisfaction with the institution.
Each spring, the portfolios are evaluated by a campus-wide
assessment committee. Results from the assessment/evalua-
tion process are fed back to the campus community as a
whole as well as to other constituent groups.['21

We have learned over the years that it is extremely impor-
tant to develop and improve an assessment process with
clearly delineated steps. Several helpful guides to develop-
ing an assessment plan exist (most notably those by Rogers
and Sando[31 and the National Science Foundation,141), but we

Chemical Engineering Education

ABET Engineering Criteria 2000)

have found that a process based
on answering the questions sum-
marized in Table 1 has been most
helpful for our needs. 51 By answer-
ing the questions iteratively, we
can be assured that we have not
overlooked any important compo-
nents of our assessment plan.
Such a process does not dictate
that a particular assessment
method be used, but it does help
faculty decide which methods are
most appropriate for measuring
certain outcomes. While we are
focusing on the use of portfo-
lios in this paper, we recognize
that a variety of other assess-
ment methods will also yield
valuable results.

Pat Hutchings defines a portfo-
lio as a collection of student work
over time (e.g., a semester, a col-
lege career) for purposes ranging
from student advising to program
evaluation.161 Many types of ma-
terials can be collected, including
papers, reports, projects, oral pre-
sentation tapes, homework, ex-
ams, and self-assessments. Port-

...a portfolio [is] a collection of student
work over time ... Many types of materials
can be collected, including papers, reports,
projects, oral presentation tapes, homework,
exams, and self-assessments.

Program Assessment Matrix

Goals What are the overall goals of the program?
How do they complement institutional and
accreditation expectations?
Educational Objectives What are the program's education
objectives? What should your students
know and be able to do?
Performance Criteria How will you know the objectives have
been met? What level of performance meets
each objective?
Implementation Strategies How will the objectives be met? What
program activities (curricular and co-
curricular) help you to meet each objective?
Evaluation Methods What assessment methods will you use to
collect data? How will you interpret and
evaluate the data?
Timeline When will you measure?
Feedback Who needs to know the results? How can
you convince them the objectives were met?
How can you improve your program and
your assessment process?

folios can be kept by students, department advisors, or insti-
tutional assessment offices. As electronic portfolios (a web-
based system of student work products allowing for on-line
access and evaluation) become more widely used, the possi-
bilities will continue to increase.171
Portfolios, like all assessment instruments, have strengths
and weaknesses. Some of the key advantages of portfolios
cited by Prus and Johnsont81 include:
They can be used to view learning and development
longitudinally (e.g., samples of student writing over
time can be collected), which is a most valid and
useful perspective.
Multiple components of a curriculum can be
measured (e.g., writing, critical thinking, research
skills) at the same time.
The process of reviewing and grading portfolios
provides an excellent opportunity for faculty
exchange and development, discussion of curricu-
lum goals and objectives, review of grading
criteria, and program feedback.
Spring 1999

*They can increase
student participation
(e.g., selection,
revision, evaluation)
in the assessment
At the same time, portfolio
assessment raises several sig-
nificant issues, including: 18
Portfolio assessment
is costly in terms of
evaluator time and
Management of the
collection and
grading process,
including establish-
ment of reliable and
valid grading
criteria, is likely to
be challenging.
Security concerns
about storage of
student work must be
addressed and
To help alleviate these con-
cerns, Prus and Johnson make
the following recommenda-

Consider portfolio submission as part of a course
requirement, especially a "capstone course at the
end of a program.
Use portfolios from representative samples of
students rather than having all students participate.
Provide training for raters.
Prus and Johnson's "bottom line" on portfolios is that they
are potentially very valuable for adding important "longi-
tudinal and 'quantitative' data, in a more natural way,"
and that they are "especially good for multiple-objective
assessment," such as the kind required for ABET EC
2000 compliance.

The Chemical Engineering Department at CSM is the
second largest in the school, with over 400 majors. It has
relied on portfolio assessment for many years, but has re-
cently undertaken extensive revision of its assessment plan
based on careful application of the steps in the assessment
process listed in Table 1 and lessons learned from the legis-

ABET Engineering Criteria 2000

latively mandated program of the past decade. The plan
discussed in this paper is still in the early stages of develop-
ment, testing, and implementation, but it represents the cur-
rent thinking of the chemical engineering faculty. By focus-
ing on ABET EC 2000, especially the student outcomes
description in Criterion 3, by obtaining input from industry,
alumni, students, and faculty, and by participating in a num-
ber of institution-wide and departmental discussions and
workshops, the faculty developed three primary goals for the
chemical engineering program:

Instill in chemical engineering students a high-
quality basic education in chemical engineering
Develop in chemical engineering students the skills
required to apply chemical engineering fundamen-
tals to the analysis, synthesis, and evaluation of
chemical engineering processes and systems.
Foster personal development in chemical engineer-
ing students to ensure a lifetime of professional
success and an appreciation for the ethical and
societal responsibilities of a chemical engineer.

The faculty then developed from three to six measurable
educational objectives (what students should know and be
able to do) for each goal. Table 2 lists a small sampling of
the objectives that were developed and indicates how each
objective maps onto ABET student outcomes.
Once the faculty agreed on program goals and educational
objectives, they began to identify curricular and co-curricu-
lar opportunities for students to achieve each objective. As
part of this work, an implementation matrix was developed
to indicate which educational objectives were addressed in
each course in the ChE curriculum. Table 3 includes a small
portion of the matrix as an example. Working together to
complete the entire matrix, the faculty were able to identify
overlaps and areas of little or no coverage and to begin a
discussion on how to enhance weak areas. In the table, an
"X" denotes that one or more of the learning objectives in a
particular course addresses the indicated departmental
educational objective. A completed matrix for the entire
curriculum will also indicate that achieving many of the
educational objectives requires work over several courses
and semesters.
Once goals and objectives were drafted and the implemen-
tation matrix indicated that all of the objectives were ad-
dressed within the ChE curriculum, the faculty began to
consider evaluation methods that would be appropriate for
measuring each objective. While they ultimately settled on a
number of methods, including senior exit interviews and
alumni surveys (triangulation, or the use of multiple mea-
sures, is usually considered the best approach to meaningful

Do not forget to assess and improve the
portfolio assessment process itself. Few of
us will get it right the first time, so revision
and refinement is essential.

Example Educational Objective for
Each Departmental Goal

Instill in students a high-quality
basic education in chemical
engineering fundamentals.

Develop in students the skills
required to apply chemical
engineering fundamentals to the
analysis, synthesis, and evalua-
tion of chemical engineering
processes and systems.

Foster in students personal
development to ensure a lifetime
of professional success and an
appreciation for the ethical and
societal responsibilities of a
chemical engineer.

Educational Objective
Graduates will be able to apply knowledge
of unit operations to the identification, for-
mulation, and solution of chemical engi-
neering problems (ABET Criteria 3a and 3e).

Graduates will be able to design and con-
duct experiments of chemical engineering
processes or systems and they will be able
to analyze and interpret data from chemical
engineering experiments (ABET Criterion

Graduates will demonstrate an ability to
communicate effectively in writing (ABET
Criterion 3g).

Chemical Engineering Education

Portion of Implementation Matrix


Apply knowledge of rate and X X X X
equilibrium processes
Apply knowledge of unit operations X X X
Design a process or system X X
Function on a team X X
>- e

Effectively communicate orally and X X
equilibrium processes

in writing
Use engineering tools X X X X

ABET Engineering Criteria 2000
\ ^ _______ __________________---------------------------- -

assessment), they agreed to continue to collect and assess
portfolios for a representative sample (approximately 20%)
of their majors. They also decided that it was best to be
selective in deciding what to collect in the portfolio, making
a single item serve multiple needs whenever possible. As a
result, for example, all three of the objectives listed in Table
2 can be assessed by evaluating written reports collected
from the student's unit operations laboratory course. Other
student work products that will be collected include final
exams from junior- and senior-level courses (e.g., fluid me-
chanics, heat transfer, mass transfer, and kinetics), and writ-
ten reports from senior design courses.

Collecting material for the portfolio is only the beginning
of the assessment process, however. Next, faculty had to
decide what constitutes evidence that students met each ob-
jective and how the evidence would be evaluated. They
decided on a set of performance criteria, e.g., 100% of stu-
dents in unit operations lab will be rated at 2 (apprentice), 3
(proficient), or 4 (exemplary) on their ability to apply knowl-
edge of unit operations for each open-ended laboratory re-
port included in the portfolio. The percentage is an estimate
at this point and will have to be refined as the assessment
plan is put into place and tested.

Faculty next needed concrete, articulated levels of perfor-
mance against which to measure student achievement. Thus,
they developed scoring rubrics for each objective, using a
process recommended by Pickett:191

1. Determine learning objectives
2. Keep it short and simple
3. Each rubric item should focus on a different skill
4. Focus on how students develop and express their
5. Evaluate only measurable criteria
6. The entire rubric should fit on one sheet of paper
7. Reevaluate the rubric (Did it work? Was it suffi-
ciently detailed?)
Table 4 shows a sample rubric for assessing unit opera-
tions laboratory reports based on the objectives listed in
Table 2. Note that the department chose to include only four
levels of student performance (exemplary, proficient, ap-
prentice, and novice) that were deemed adequate for the
purpose of program assessment. Rubrics using the same four
performance levels have been developed for each item of
student work collected in the portfolio.
Once each academic year's portfolio materials have been

Scoring Rubric for Assessing Unit Operations Laboratory Reports

Objective 4- Exemplary 3 Proficient 2 Apprentice 1 Novice Score

ChE graduates will be able to Student groups apply knowl- Students groups apply knowl- Student groups apply knowl- Student groups make sig-
apply knowledge of unit edge with virtually no con- edge with no significant con- edge with occasional concep- nificant conceptual and/or
operations to the identifica- ceptual or procedural errors ceptual errors and only minor tual errors and only minor procedural errors affecting
tion formulation, and solution affecting the quality of the procedural errors, procedural errors, the quality of the experi-
of ChE problems experimental results, mental results.

ChE graduates will be able to Student groups design and con- Student groups design and Student groups design and Students groups design and
design and conduct experi- duct unit operations experi- conduct experiments with conduct experiment with no conduct experiments with
ments of ChE processes or ments with virtually no errors; virtually no errors; analysis significant errors; results major conceptual and/or
systems and they will be analysis and interpretation of and interpretation of results are analyzed but not inter- procedural errors; no evi-
able to analyze and inter- results exceed requirements meet requirements of exper- preted; very limited evi- dence of significant analy-
pret data from chemical of experiment and demonstrate iment and demonstrate some dence of higher-order sis and interpretation of
engineering experiments, significant higher-order higher-order thinking ability, thinking ability, results; fail to meet require-
thinking ability, ments of the experiment;
demonstrate only lower-
level thinking ability.

ChE graduates will demon- Written report is virtually Written report presents re- Written report is generally Written report does not pre-
strate an ability to communi- error-free, presents results and suits and analysis logically, is well-written but contains sent results clearly, is
cate effectively in writing, analysis logically, is well or- well organized and easy to some grammatical, rhetorical, poorly organized, and/or
ganized and easy to read, con- read, contains high-quality and/or organizational errors; contains major grammatical
tains high-quality graphics, graphics, contains few minor analysis of results is men- and rhetorical errors; fails to
and articulates interpretation grammatical and rhetorical tioned but not fully articulate analysis of re-
of results beyond requirements errors, and articulates interpre- developed. suits meeting requirements
of the experiment, station of results that meet of the experiment.
requirements of the

Spring 1999 113

ABET Engineering Criteria 2000

collected, the departmental assessment committee meets to
review and refine the language of each rubric. Sample stu-
dent work is used to test the updated rubrics and to ensure
that committee members' scores are normedd" prior to scor-
ing the collected portfolio materials. This procedure ensures
that valid assessment data can be obtained using each rubric
and that inter-rater reliability is assured.
Finally, the information obtained from the assessment and
evaluation process is fed back to the chemical engineering
department students and faculty as well as to other interested
stakeholders, including employers and the departmental ex-
ternal advisory committee. The faculty considers assessment
results from each year in planning revisions to the curricu-
lum. In addition, the process itself is critiqued annually and
changes are made based on stakeholder feedback. In brief,
the process and the product will both undergo continuous
review and improvement.

Based on a decade of portfolio assessment at CSM, we
have learned many lessons; the most important and relevant
ones are:

Avoid the temptation to start collecting portfolio
materials before developing clear goals, objec-
tives, and an assessment process. Before decisions
are made about which materials to collect and
assess, be sure to answer questions about what is
being assessed, how the data will be analyzed,
when materials will be collected, and who will
receive the results.

Be sure to promote stakeholder buy-in by involving
as many constituencies as possible in the portfolio
development and implementation process. If one
lone faculty or staff member is assigned the
assessment task, the plan will almost assuredly

0 Lookfor campus resources to help faculty get
started with portfolio assessment and provide
faculty development opportunities. Most schools
have some level of assessment expertise on
campus-do not be afraid to search for help in
training engineering faculty to become good
portfolio assessors.

Remember that quality of results is more important
than quantity. Portfolio assessment does not have
to measure every learning objective in every
course in the curriculum. Collect results that will
be of most value in improving the learning and
teaching process and use sampling techniques to

collect a longitudinal snapshot of student achieve-
Allay fears in colleagues who view portfolio
assessment as a "touchy-feely" process by
including them in the development of well-
designed scoring rubrics and inter-rater reliability
0 Do not forget to assess and improve the portfolio
assessment process itself Few of us will get it
right the first time, so revision and refinement is

Portfolio assessment is a potentially robust method for
evaluating chemical engineering programs and meeting
ABET EC 2000 requirements. As with any assessment in-
strument, however, successful use of portfolios requires care-
ful implementation of an assessment process based on devel-
oping measurable objectives with articulated performance
criteria, identifying curricular and co-curricular activities
designed to help students meet each objective, collecting
and evaluating assessment data using well-developed scor-
ing rubrics, and providing assessment results to all interested
stakeholders in the process.

1. Olds, B.M., and M.J. Pavelich, "A Portfolio-Based Assess-
ment Program," Proceedings, 1996 ASEE Annual Confer-
ence, ASEE, Session 2313 (1996)
2. Olds, B.M., and R.L. Miller, "Portfolio Assessment: Measur-
ing Moving Targets at an Engineering School," NCA Quar-
terly, 71(4), 462 (1997)
3. Rogers, G.M., and J.K. Sando, Stepping Ahead: An Assess-
ment Plan Development Guide, Rose-Hulman Institute of
Technology, Terre Haute, IN (1996)
4. Stevens, F., F. Lawrence, and L. Sharp, User-Friendly Hand-
book for Program Evaluation: Science, Mathematics, Engi-
neering, and Technology Education, J. Frechtling, ed., NSF
93-152, National Science Foundation (1996)
5. Olds, B.M., and R.L. Miller, "An Assessment Matrix for
Evaluating Engineering Programs," J. of Eng. Ed., 87(2),
6. Hutchings, P., "Learning Over Time: Portfolio Assessment,
AAHE Bulletin, 42(8), 6 (1990)
7. Doering, E., "Electronic Portfolios: What to Do With the
Information," Proceedings, 1998 ASEE Annual Conference,
Session 2613 (1998)
8. Prus, J., and R. Johnson, "A Critical Review of Student
Assessment Options," in Assessment & Testing, Myths &
Realities, T.H. Bers and M.L. Mittler, eds., New Directions
for Community Colleges #88, Jossey-Bass Publishers, San
Francisco, CA (1994)
9. Picket, N. "Guidelines for Rubric Development," http://
(accessed 7/27/98) O

Chemical Engineering Education

Opportunity on the Wings of Danger
Continued from page 109.
2. Administrators will need to expend resources to meet the
requirements and have confidence that the resources ex-
pended will have sufficient benefit to the overall program.
Many engineering programs are already hampered by
the demands for limited resources-money, equip-
ment, space, time. etc. Deans and department heads
need to make decisions about how to balance limited
resources. The need to develop a culture of assess-
ment for continuous quality improvement processes
does not come without costs. If the move to EC 2000
is not see as having a defensible cost/benefit ratio in
relation to other critical needs, there will be resis-
tance to support the effort at a level that will be
sufficient for success.

3. There needs to be an attitude of trust and cooperation
among faculty and between faculty and administrative staff
In the mind of a faculty member, there is a thin line
between evaluating the program and evaluating
"me." Administrators need to be trusted not to use
program-assessment information to evaluate indi-
vidual faculty members. The issue of evaluating a
faculty member must be done outside the context of
program evaluation to maintain faculty confidence in
the process. In addition, faculty need to be trusted to
use the assessment information to enhance student
4. EC 2000 evaluators must understand assessment and CQI
and know adequate processes when they see them.
There are at least three things that can happen when
the EC 2000 evaluators come to your campus. 1)
They will be well versed in EC 2000 and have a clear
understanding of the requirements, limitations, and
possibilities for developing and implementing CQI
and assessment processes in engineering education.
As a result, they will be able to assist you in evaluat-
ing your program and make recommendations for
improving your educational processes based on
sound assessment information. 2) They will not have
a very good understanding of the requirements,
limitations, and possibilities for developing and
implementing CQI and assessments processes in
engineering education. As a result, they will not be
able to assist you in improving your processes, and
because you have done your homework, you will end
up educating them. 3) They will not have a very
good understanding of the requirements, limita-
tions, and possibilities for developing and imple-
menting CQI and assessment processes in engi-
Spring 1999

ABET Engineering Criteria 2000)

neering education-but they do not know it. As a
result, they will apply inappropriate standards to
the processes you have developed.
Scenarios 2 and 3 are outcomes that will erode the good
that EC 2000 can bring to engineering education. Of course,
the horror stories will travel much faster (and be more exag-
gerated) than the success stories.
The good news is that the Engineering Accreditation Com-
mission is working very hard at providing training sessions for
all EC 2000 evaluators. During this process, they are involving
evaluators in discussion of the elements of assessment plan-
ning and CQI processes. Information about the availability of
training sessions can be found on the ABET web site.'61

While there are many dangers that lie ahead for engineer-
ing programs that are implementing outcomes assessment,
there are none that cannot be overcome by careful prepara-
tion and planning. The industry and education representa-
tives of ABET's Engineering Accreditation Commission
(EAC) are taking the lead by providing programs to better
inform evaluators and engineering faculty of the core con-
cepts and processes embedded in EC 2000. It is important to
remember that few of us have more to lose if EC 2000 fails
than those who have had the courage to step forward to
develop, propose, and champion the new accreditation crite-
ria-the EAC. This activity has been in response to the
deafening outcry of engineering faculty to do away with the
rigid, "bean-counting" criteria that previously existed.
It is time to take advantage of the lessons learned from
those who have been engaged in outcomes assessment in
different contexts and apply them to engineering education.
Faculty are already doing outcomes assessment in the class-
room, and we can take advantage of the opportunities pro-
vided by the new approach to accreditation to assess our
programs as a whole. The dangers in doing so can be avoided
if we are willing to learn from others who have been there.

1. Engineering Accreditation Commission, Engineering Crite-
ria 2000, Accreditation Board for Engineering and Technol-
ogy, Inc., Baltimore, MD, (1998)
2. Accreditation Board for Engineering and Technology website
found at
3. Rogers, Gloria, "Assessment for Continuous Quality Im-
provement Model," Rose-Hulman Institute of Technology,
Terre Haute, IN
4. Schindel, William D., "The Tower of Babel: Language and
Meaning in Systems Engineering," SAE Technical Paper
Series, SAE International (1997)
5. Rogers, Gloria M., and Jean K. Sando, "Stepping Ahead: An
Assessment Plan Development Guide," Rose-Hulman Insti-
tute of Technology, Terre Haute, IN (1996)
6. O

(ABET Engineering Criteria 2000



An Unstable Process?

Worcester Polytechnic Institute Worcester, MA 01609

he new outcomes-based world of higher education
and accreditation may be scary to some, annoying to
others, fun to a few, but certainly challenging to all
of us. We are faced with examining our educational process
in new ways, optimizing a complex ill-defined process, and
presenting it clearly and coherently to our accreditation
agency. The goals of this paper are to outline the basics of
outcomes assessment, describe a format useful for assessment
preparation, summarize some lessons learned from the 1996
pilot visit conducted at Worcester Polytechnic Institute (WPI),
and discuss some consequences of outcomes assessment.
Let's start with the question, "Why do assessment?" A
number of responses come to mind, such as "I do it to assign
grades," " make accreditation agencies happy," "
make grant funders happy," " know that what I do as an
instructor is meaningful," " know that students are learn-
ing what I set out to teach them."'m If we care at all about our
teaching, we each embrace one or more of these during any
course we teach. But how many of us examine the complete
curriculum or spend significant time with colleagues (and
not just our chemical engineering colleagues) discussing
and doing something constructive about these issues?
Until recently, I suspect, such faculty and departments
were in the small minority.
So, is there a problem? We have been successfully educat-
ing competent engineers for decades. Why change? I like the
simple answer-we can always do better and are ethically

David DiBiasio has taught in the Chemical En-
gineering Department at Worcester Polytechnic
Institute since 1980, and had industrial experi-
ence at the DuPont Company prior to that. His
current interests are in biological reactor engi-
neering, engineering education, and educational

obligated to strive for the best. There are many ways to
improve teaching and learning, and outcomes assessment is
one of them. Others may prefer the more involved response
that links rapidly changing technological market forces to
needed changes in our graduates' abilities.121 Either way,
future graduates must function effectively in multidisciplinary
teams, communicate well, understand global and societal
issues related to engineering, and of course, master engi-
neering and scientific fundamentals. A rigorous, well-de-
signed assessment process can make that happen, allow new
flexibility in the curriculum, and result in continuous im-
provement. I am hard pressed to find a reason why we
should not do it. We should be aware there are costs associ-
ated with doing it right, however.

Assessment basics are relatively simple: define objectives,
determine if students are meeting them, and improve the
educational process if they are not. An excellent primer is
provided by Rogers and Sando,[31 and other articles in this
issue expand on these principles. The assessment process
usually includes
EI Setting educational objectives
0E Determining performance criteria
E[ Defining practices
[E Defining assessment methods
EN Evaluating the assessment data
E Feeding back the results to improve the curriculum
Measurable outcomes are linked to objectives, and the
whole process drives continuous improvement of the educa-
tional system. Sounds straightforward, right? Well, maybe
not. Goal setting and determining performance levels for
chemical engineering topics may not sound too bad, but
what about these "assessment methods"? Experts tell us we
need methodologies that are both formative and summative.

@ Copyright ChE Division ofASEE 1999

Chemical Engineering Education

ABET Engineering Criteria 2000

Formative methods are those that take place periodically
during a course or curriculum and answer the question, "Is it
working?" For example, a mid-course survey might tell a
professor some on-line course adjustments are needed.
Summative methods take place at the end and answer the
question, "Did it work?" A comprehensive final exam tells
the professor what fraction of the students mastered the
material at specified comprehension levels.
Either method might involve qualitative or quantitative
tools. Simply put, quantitative tools involve numbers such as
exam scores, survey results, and database analysis. Qualita-
tive tools involve textual or verbal information, including
open-ended survey responses, videotape data, and interview
transcripts. The same experts tell us that we should use both
types for formative and summative evaluations, and that
triangulation, redundant measurements with multiple inde-
pendent tools, is very important.
Most of us are comfortable with quantitative data, but
many engineers are uncomfortable with qualitative data. We
may not know how to collect it or analyze it properly, and
too often it is regarded with disdain. Any mention of "dis-
course analysis" may set off the touchy-feely alarm in many
engineers. But some of the richest and most meaningful data
from an educational experience are sometimes obtained only
through qualitative analysis. Several of the items in ABET
EC 2000 Criterion 3 are quite well suited to measurement
using qualitative techniques.
Process Control Analogy
The best assessment
processes include a mix Ideal Students
of methods and tools. Ideal
The process is also ----- -d
closed-loop since it con-
tains an essential feed-
back step that forces us
to correct the curriculum
when we detect prob-
lems in the outcomes. It Perf. Criteria
is hard to ignore the 4,
analogy to process con-
trol, and other authors Assessment
have used block dia- An i
grams to help simplify Ana
the description.[41 I be-
lieve the analogy and I
diagrams are also use-Practices an
ful to make a different
points5 - Employers
Figure 1 presents one Figure 1. The assess
general view of assess- feedback

ment. The primary loop is shown in bold. The "process" is
the curriculum into which students enter. They exit possess-
ing desired abilities or outcomes. The output is measured by
having students demonstrate these abilities through defined
practices. Feedback is achieved by comparing measured out-
comes against the "set point" or performance criteria for
each outcome. The controller is the assessment analysis that
dictates changes in the curriculum when outcomes don't
meet performance criteria. Unfortunately, this is a multivari-
able, multiloop system with difficult measurements. If we
wanted all students to graduate with red hair, then measure-
ment, feedback and correction would be easy. Our goals in-
clude some tough-to-measure qualities, however, such as life-
long learning and understanding of ethics and social issues.[41
We also need to consider some measurements taken well
after graduation. They tell us something about the connec-
tion between our curriculum and job performance. Such
measurements must also enter our feedback loop, even though
there is a significant time lag. Our constituencies include our
students' employers. Since most students take industrial
positions, industry involvement in determining objec-
tives and performance criteria is important. This results
in set-point disturbances.
The characteristics of students entering college change
with time. Why wait until students are well into the curricu-
lum (when it may be too late) to make corrections? Any
good control-system designer would try solving this prob-
lem with a feedforward loop. Such a loop might include
adapting the process
and the controller.


adapt Students

Curriculum C
Measurement E
-- Post Graduation

zent process as a multiloop
control system.

Finally, a major goal
is "continuous im-
provement," or opti-
mization. Model-ref-
erence adaptive con-
trol is one possible
scheme. Changing set
criteria are input to an
ideal educational
model. The theoretical
output is compared to
our actual outcomes
and an adaptation al-
gorithm modifies both
the controller and the
curriculum appropri-
ately so that the sys-
tem moves continu-
ously toward the opti-

Spring 1999

(ABET Engineering Criteria 2000

mum. Maybe such an analogy is a bit un-
realistic, but some time-optimal control
strategy is needed.
At this point, we have a multivariable, feed- serio
back-feedforward, time-optimal control sys-
tem to design and operate. It is potentially asses
unstable. Variable measurement is difficult, their
and we have ignored the sampling problem:
how often and how many students will be OJ
sampled? If we are serious about assessment,
then the only conclusion is that this is not an con
easy problem.51 It is complex, with many that
possible solutions-none of which are simple.
Recall that multiple types of evaluation tools nO
should be applied at several levels of the
curriculum and across appropriate time peri- easy p
ods. Assuming none of us have achieved the It is cC
ideal educational system, we must realize
that even the simplest design will include With
changes in the curriculum. Force fitting of
an existing static and inflexible education pos
system to EC 2000 will probably not work. solut
But chemical engineers are good at attack-
ing and solving difficult, ill-defined, compli- nOI
cated problems. After all, isn't that what we whiC
want our students to do? This problem will
require some effort to do right, but it can silT
be done, and the results should be well
worth our efforts.

The Chemical Engineering Department at WPI was ac-
credited under the new EC 2000 during the first pilot evalua-
tion in 1996. We faced the problems described above and are
still working on solutions. Our specific preparation and time-
table were unique to that pilot experience, but some elements
of preparation and presentation might be useful to others.

Visit Preparation
Visit preparation cannot begin early enough, and two years
is an absolute minimum for the first visit. Four or five years
would be better. It should be clear that departments can no
longer wait until the year before a visit to begin preparations.
Outcomes assessment is continuous, and data collection and
analysis must occur constantly. The educational process has
a four-year time constant, so early formative data collection
is highly recommended. Also consider that alumni and em-
ployer survey data might have little meaning unless col-
lected at the proper time intervals.
Educational objectives must be defined first. Presumably
there will be institutional ones and discipline-specific de-

e c

s a





f a







partmental ones. A common mistake is to launch
into discussions about assessment methods with-
out clear objectives. It is best that faculty and
bout staff avoid many hours in committee meetings
discussing the content and logistics of student
elt, portfolios before they understand exactly how
S portfolios link to objectives. When clear, mea-
surable objectives are in place, the specification
of performance criteria and the choice of as-
sessment tools derive logically. This may seem
On IS obvious, but experience shows that we tend to
s i digress quickly into discussion of assessment
S methodologies prior to understanding how we
n really want to use them.
A committed (compensated) department co-
blem. ordinator can help facilitate the process, but all
plex, faculty must be involved. A time commitment
level of at least 25% is needed for the coordina-
iny tor.[6] There is no secret formula for engaging
all faculty in the process, but one potentially
le useful argument is that outcomes-based assess-
1S- ment is being required in more and more re-
search proposals. Proposals with decent assess-
Of ment plans have the edge over others; hence,
research-oriented faculty might gain useful
ire knowledge from participation in the process.
e. Consultant use is highly recommended, but
consultants cannot and should not write the as-
sessment plan. They cannot substitute for the
faculty. Faculty must define objectives, perfor-
mance criteria, and feedback mechanisms. Consultants can
help recommend methodologies and assist with data-
evaluation strategies.

Documentation Using an Assessment Matrix
Presentation of a complex assessment process to a visiting
team is problematic. One must clearly show how the depart-
ment plan addresses the major evaluation criteria. If student
portfolios are used, you cannot expect your evaluator to read
through several of them looking for evidence of items under
Criterion 3. The portfolio itself needs a guide, probably
written by the student, and it needs evaluation, probably by
faculty. Our experience, and that of others,[71 showed that the
assessment matrix format is quite useful for presenting the
department's plan, but much detailed additional documenta-
tion must accompany the matrix.
The assessment matrix is one way to help organize the
plan. It is concise, and it serves as a guide to additional
documentation. I will show how we used it to outline our
assessment plan, how two years later it portrays some poten-
tial problems, and how it illustrates some consequences of
Chemical Engineering Education

ABET Engineering Criteria 2000 )

outcomes assessment that are important outside of WPI.

Table 1 shows a portion of the WPI Chemical Engineering
Department matrix. The column headings are the general
assessment process steps. The row headings are the indi-
vidual educational objectives. Our department adopted ABET
EC 2000 Criterion 3 (a-k) as objectives and we show two of
them for example purposes. Some definitions are necessary
to follow the matrix:

IQP The Interactive Qualifying Project. This project is a
significant open-ended, non-classroom experience that equals
three courses worth of credit. It is usually done during the
junior year in teams with students from different majors. It
must address a problem that considers the interaction of
technology with society and culture. Faculty advisors may be
from any discipline, and the project topics are interdiscipli-
nary. Since we believe that the global nature of technology is
important, many of our students leave campus to conduct
these projects at our international project sites. The project is
a degree requirement for all students.
CDR Completion of Degree Requirement Form. The form is signed
by the faculty advisor when the final project report is
completed and graded. It is proof that the student has satisfied
the degree requirement and is filed with the Registrar.
PRC Program Review Committee. This is the department
undergraduate committee that annually reviews all senior
transcripts to ensure all degree requirements are met for

So, let us go through the matrix using "an ability to func-
tion on multidisciplinary teams" for our first example. The
performance criteria is that students complete a team-based
IQP, and the practice is that we require all students to do
these projects. Project assessment is done by the faculty
advisor and is documented by a grade appearing on the CDR
form. This is accomplished for nearly all students by the end
of the junior year. The feedback process involves the PRC-

if a student does not complete the project, then the PRC
issues paperwork informing the student and the Registrar of
that fact. Superficially, this might look okay. These projects
are truly multidisciplinary (you will have to take my word for
that), so completing one with a passing grade, or being duly
informed if you did not, may seem like a reasonable assess-
ment loop. Until recently it appeared to be so, but a critical
reexamination of the matrix makes me now think otherwise.
I believe that a real assessment plan must go deeper.
Completing such a project is not always a guarantee that
students function effectively in teams since a dysfunctional
team could still pass this degree requirement. Unless an
evaluation of effective teamwork is a documented part of the
advisor's grading policy, we cannot be sure about the stu-
dents' abilities relative to the objective. If we believe that
argument, then the grade alone is not the proper assessment
method. A tool that measures teamwork effectiveness must
replace it, and measurable standards of effective teaming
must be defined. It logically follows that the PRC review is
not adequate feedback. If our teamwork-effectiveness tool
indicates that significant numbers of students do not meet
our standards, then somehow we must find a way to include
team-building activities into the process and formally docu-
ment the procedure. Some faculty may claim that such a
move threatens their academic freedom as project advisors.
This issue may arise any time faculty are asked to include
new course activities for outcomes assessment purposes.
We, and other universities, must deal with this issue as
assessment plans are developed.
Here is another example (objective "h"): "...the broad
education necessary to understand the impact of engineering
solutions in a global/societal context." Compare this objec-
tive to the goals of the IQP and you will see they are quite

Portion of the Department Assessment Matrix

Performance Criteria I Practices




d) an ability to -complete a -IQP -CDR form -Jr. year -PRC audit,
function on multistudent IQP opportunities are academic
multidisciplinary available for advisor
teams every student

h) the broad education
necessary to
understand the impact
of engineering
solutions in a
global/societal context

Demonstrate an
understanding or
interest in the global
or societal
implications of
engineering by:
-completing an IQP

WPI has extensive
overseas IQP

-CDR forms

-Jr. year
for IQP



Spring 1999

CABET Engineering Criteria 2000

similar. What better way to satisfy this objective than to
complete one of these projects outside the United States?
Such an experience includes a multidisciplinary team work-
ing in a government agency, a company, or a non-profit
organization in another country on a topic interfacing tech-
nology and society. Our assessment plan for this item has
some of the same problems described above, but we focus
here on a different aspect.
WPI has an outstanding and extensive global-projects pro-
gram. We send more engineering students overseas for such
projects than any other university in the country81-quite
surprising considering our relatively small enrollment. But
last year, only one-third of our students went off campus for
their IQP experience. This means that two-thirds of our
students did not satisfy objective (h) unless they completed
some other appropriate, but as yet unknown, academic expe-
rience. Should we try to send all our students outside the
U.S., or do we modify the on-campus curriculum to provide
alternate paths? Both are viable options, but neither is simple.
What about other schools? Should a large university initiate
such an extensive global-projects program? The expertise,
resources, and organization needed to run such a program
are not trivial. Can this objective be equivalently addressed
in a course about global engineering? Does that dilute the
academic impact so much that the original intent of this
objective is lost? We are currently exploring possible
answers to these questions. The process is part of what
EC 2000 is all about.
Clearly and efficiently linking measurable outcomes to
objectives is key to preventing instability in the assessment
process. Good objectives with poor evaluation tools means
we have little idea if educational goals are met. Good objec-
tives and tools with no feedback means no improvement will
occur. Vague objectives, poor evaluation strategies, and ex-
cessive assessment will choke the life out of an academic
system-an instability we must avoid. Chemical engineers
have the skills to design and control complicated chemi-
cal processes. Although some adapting is required, we
are in a good position to apply those skills creatively to
good assessment design.

The consequences of EC 2000 will be different for each
school. The two examples from WPI's assessment plan de-
scribed above illustrate two major points:
Designing and conducting a rigorous outcomes-
based assessment process for engineering education
is a complex task.
We will all have to change the way we do business.
This includes the way we educate and interact
with students, the way learning is measured, and

the way we use the data.
If universities and ABET take this approach seriously,
then we must do it right to make it meaningful."5 It is a
challenging problem that is well worth our efforts. This
holistic approach to education frees us from the rigidity of
past accreditation philosophies. New curriculum flexibil-
ity is possible, so long as we document its successes and
use its failures for improvement. Yes, our learning curve
may be steep, but if we maintain high performance stan-
dards, our students will ultimately be the main beneficia-
ries of these efforts.
An answer to our earlier question about why we should
assess included something about putting meaning into our
instruction and knowing that students learn what we want
them to learn. Typically, we focus only on the technical
content. The new accreditation criteria add a human element
into the process. Perhaps this element coupled with good
assessment plans will ensure that students go beyond our
earlier answers and learn "how to learn." This certainly gets
at the heart of what teaching is all about and may help
guarantee that our students become lifelong learners.

The opinions expressed in this paper are solely those of the
author and do not necessarily represent those of Worcester
Polytechnic Institute, the Chemical Engineering Department
of WPI, or the Accreditation Board for Engineering and
Technology. The author gratefully acknowledges Natalie
A. Mello (Interdisciplinary and Global Studies Division,
WPI) for reviewing the manuscript and suggesting sev-
eral improvements.

1. Mello, N.A., "Outcomes Assessment: How to Design a Plan
that Works," NSF Regional Conference: Educational Re-
form-Issues and Obstacles for the 21st Century, Boston,
MA (1998)
2. Kenny, S.S., Chair of the Boyer Commission Report "Rein-
venting Undergraduate Education: A Blueprint for America's
Research Universities," available at http://notes.notes. (1998)
3. Rogers, G.M., and J. Sando, "Stepping Ahead: An Assess-
ment Plan Development Guide," Rose-Hulman Institute of
Technology (1996)
4. Shaeiwitz, J.A., "Outcomes Assessment Methods," Chem.
Eng. Ed., 32(2), 128 (1998)
5. Felder, R.M., "ABET Criteria 2000: An Exercise in Engi-
neering Problem Solving," Chem. Eng. Ed., 32(2), 126 (1998)
6. Breidis, D.M., "An EC2000 Visit: Perspectives from Both
Sides of the Fence," Session 2613, ASEE Meeting, Seattle
7. Olds, B.M., and R.L. Miller, "An Assessment Matrix for
Evaluating Engineering Programs, J. Eng. Ed., 87(2), 173
8. "Science Students Abroad," Chronicle of Higher Education,
A35 (1995) 0
Chemical Engineering Education

book review

Engineering Flow and Heat Exchange
Revised Edition
by Octave Levenspiel
Plenum Press, New York and London (1998)

Reviewed by
Gabriel I. Tardos

This is the first revised edition of this book, first published
in 1984. Professor Levenspiel should be commended for
producing such an excellent text, written specifically for
engineering students. The book is a pleasure to read and
offers several amusing problems, all stated in the language
of students, with explanations and examples they can easily
understand. Very few texts in engineering can make such a
claim. I have used this text exclusively since 1992 in my
teaching of unit operations to chemical engineering stu-
dents. The material is broad enough, however, to also be
used in mechanical engineering, and perhaps in civil engi-
neering courses as well, to teach flow and heat transfer.
Students (especially undergraduates) tend to sell used text-
books once they finish a subject and pass their final exami-
nation. I found, with great pleasure, that Engineering Flow
and Heat Exchange was not one of those books; seniors use
it in their design courses and many graduates keep the book
as a reference. This is obviously due to the wealth of
information in the book and the ease with which the infor-
mation can be retrieved and used. Inclusion of compressible
and non-Newtonian fluid flow in the fluid-mechanics sec-
tion and direct-contact heat exchangers in the heat-exchang-
ers section is a substantial achievement and significantly
adds to the usefulness of the text.
One example of the book's unique approach to explaining
a complex concept through humor and straightforward, easy-
to-understand language is illustrated by how Professor
Levenspiel explains the concept of equivalent average slurry
density in the problem "Counting Canaries Italian Style."
The "slurry" consists of canaries flying in the air inside a
closed container. Measuring the pressure before and after
the canaries are airborne, and using the Bernoulli equation,
gives the change in density and therefore the number of
"particles" (birds). Ingenious!
As already mentioned, the book is divided into a section
on fluid mechanics and a section on heat transfer. The first
part includes basic equations for isothermal flowing systems
in Chapter 1, and as an example, flow of incompressible
Newtonian fluids in pipes and around solid immersed ob-
jects in Chapters 2 and 8, respectively. Unlike other similar
Spring 1999

texts, the theory is kept short and the assumption is that the
student has taken a prior course in fluid mechanics. It is
assumed, for example, that the student is familiar with the
concept of the Fanning friction factor.
Chapters 3 and 4 address compressible flow of gases
(through material taken mostly from thermodynamics) and
low pressure, "molecular" flows. The concept of "molecular
slip" is introduced here.
Chapter 5 contains, as mentioned above, concepts and
problems of non-Newtonian flow explained in a direct and
simple-to-understand fashion. The student is reminded that,
in general, this complex fluid can be treated as Newtonian
with an additional term and all that is required is to find the
correction due to the non-Newtonian behavior. Since most
fluids in industrial practice are non-Newtonian, the intro-
duction of this material is, I think, crucial. Furthermore,
rheometry to measure non-Newtonian behavior is also
presented in detail.
Part one of the book, also contains chapters of flow in
porous media and in fluidized beds. They are also well
written, with many examples and actual industrial applica-
tions both solved and presented as homework problems.
The second part of the book, on heat transfer and heat
exchanger design, is also enlightening, crisp, and well con-
structed. Chapters 9, 10, 12, and 13 contain the usual mate-
rial on different forms of heat transfer, combined heat trans-
fer, and two-fluid heat exchanger design. Here again, it is
assumed that the student has taken a previous introductory
course in heat transfer since familiarity with, for example,
the Nuselt number is required. The material in Chapters 11,
14, and 15 contains unsteady heating and cooling and design
of direct-contact exchangers and regenerators-material usu-
ally not covered in standard texts. The second part ends
(Chapter 16) with a set of recommended problems involving
material contained in the book, keeping in mind practical,
industrially relevant applications.
There is an extended Appendix with very useful informa-
tion such as transformation of units, some material proper-
ties, dimensionless groups, and values of more important
parameters such as heat transfer coefficients in different
geometries. The text also comes with a set of solutions
(available to the instructor) to the problems in each chap-
ter, with every second problem being solved. The prob-
lems in the last chapter (16) all have solutions. The illus-
trations in the book are inspired and clear, while the
nomograms, mostly for heat transfer calculations, are up-
to-date and easy to use.
Over all, this is an excellent book, written with the heart.
The reader can visibly appreciate this. It should be a perma-
nent fixture on the bookshelf of any engineer who studied or
uses fluid flow and heat transfer in his work. 7

ABET Engineering Criteria 2000


A Tool for Defining and Assessing a Course

Arizona State University Tempe, AZ 85287-6006

s ever-increasing numbers of students initially at-
tend community colleges, articulation is a concern
of public universities. Articulation issues are particu-
larly difficult for engineering design classes, which tend to be
institutionally dependent. In Arizona, a task force of university
and community college engineering faculty addressed this is-
sue for the first-year engineering design class, and a process,
based on the educational research work of Tyler" l and Bloom,121
was developed. It involved creating and analyzing an Articula-
tion Matrix-a matrix that shows the educational relationship
between a course's learning activities and learning objectives.
One strength of the developed process was the creation of an
explicit assessment process to determine if a proposed course
was acceptable.

In Arizona, the first-year engineering (design) course is a
cornerstone in each of the three state university's BS engineer-
ing curriculum. Since introductory design courses do not gen-
erally have the type of defined learning objectives found in a
statics or dynamics course, these introductory courses tend to
be unique at each of the three universities. With the large number
of students who want to take the first-year engineering course at
a community college, the three unique courses have made it very
difficult for the community colleges to offer a course that could
transfer to all three universities. The community colleges have
been forced to select the university most of their students are
likely to attend and then develop a 'course consistent with it.
Articulation problems extend beyond community colleges to
include all course transfers between schools of engineering.
In the fall of 1996, a task force* of faculty from the three
universities and several community colleges started work on
this articulation problem. They were faced with the standard
articulation issues of
What topics, skills, etc., to include in the first-year design course

* Vern Johnson (U. Arizona), Spencer Brinkerhoffand Pamela Eibeck
(Northern Arizona U.), Dan Jankowski, Lynn Bellamy, and Barry
McNeill (Arizona State U.), Mel Heaps (Cent. Arizona College), Dave
May (Pima Com. College), and Don Yee (Mesa Com. College).

How to ensure (establish) that a proposed course was, in fact,
and, a third issue to be considered
How to address the first two issues in a manner such that a
school still had the flexibility to develop its own unique
character for the course, using the school's interests and
The task force developed a process that satisfactorily ad-
dressed all three of the issues. The process requires the
creation and use of an Articulation Matrix, so called because
it helps resolve the articulation problem.
This paper will first present the educational theory upon
which the Articulation Matrix is based, followed by a gen-
eral discussion of the Articulation Matrix and how to create
and analyze it. It will conclude with two examples, one
showing how the matrix was used in the articulation process
and one showing how the matrix could be used as part of an
ABET EC 2000 accreditation effort.

The educational basis for the Articulation Matrix comes
from the published work of two School of Education faculty
members at the University of Chicago, Ralph Tyler["1 and
Benjamin Bloom.[21 Tyler formulated a basis for defining a
course or curriculum, while Bloom worked to clarify the
Barry McNeill received his BS in chemical engi-
neering from Stanford University in 1962 and his
MS and PhD in mechanical engineering from
Stanford in 1976. His current interests include
the design process, how to teach design, and

Lynn Bellamy obtained a BS in chemical engi-
neering from Texas A&M University in 1962, and
MS and PhD degrees in Engineering Science
from Tulane University in 1965. His interests are
in asynchronous and distance learning.
Copyright ChE Division of ASEE 1999
Chemical Engineering Education

ABET Engineering Criteria 2000)

terms used in describing how well a subject has been mastered.
Defining a Curriculum (Course) Criterion 2 of ABET EC
200013] requires a school to have and use a defined process for
the development and continuous improvement of its curricu-
lum. How to satisfy Criterion 2 has created a rash of interest
and a plethora of papers and workshops. But the issues in-
volved in Criterion 2 are not new and were addressed in the late
40s by Tyler. In 1949, he published a short treatise on the basic
principles involved in curriculum and instructional development.
His work is neither a textbook nor a manual on curriculum
development, but rather the "rationale for viewing, analyzing,
and interpreting the curriculum."
Tyler's approach involves answering four basic questions:
1. What educational purposes should the school seek to attain?
2. What educational experiences can be provided that are likely
to attain these purposes?
3. How can these educational experiences be effectively orga-
4. How can we determine whether these purposes are being
Criterion 2 requires schools to: define a set of learning objec-
tives (i.e., answers to question 1); define a strategy to accom-
plish the learning objectives (i.e., answers to questions 2 and 3);
and define an assessment process to measure achievement of the
learning objectives (i.e., answers to question 4).
Learning Objectives-Two-Dimensional Vectors The
first step in defining a course or curriculum is development of a
set of learning objectives, sometimes called learning outcomes.
At first glance, the exemplar learning objectives published in
the literature appear to be one-dimensional, i.e., only a subject,
topic, or skill to be learned.[4'51 But upon closer observation, the
objectives also define some level of performance associated with
the competency (e.g., "graduates will be able to identify, formu-
late, and solve..."141 or "will exhibit good listening skills"[51).
Learning objectives are two-dimensional vectors consisting
of a competency and a degree to which the competency is
learned or mastered. Of these two parts, the first is the easiest
to define precisely. The competency is the subject, topic, or
skill to be learned (e.g., integration by parts). Precisely defin-
ing the degree to which a competency is mastered is more
difficult. For example, what does "really understands integra-
tion by parts" mean? While this example could be improved
and made more specific, the effort is probably not worthwhile,
especially if every competency needs its own special wording.
Rather, what is needed is a concise, precise, agreed-upon set of
terms that can be used to define the degree to which any
competency is mastered.
In the mid-50s, Benjamin Bloom, David Krathwohl, and
others addressed this problem and developed two taxonomies
of educational objectives, one for the Affective Domain171 and
one for the Cognitive Domain.121 In the foreword to the Cogni-
Spring 1999

tive Domain book, Bloom states that the book was "espe-
cially intended to help them [those involved in the develop-
ment of curriculum and courses] discuss these problems
[defining how much is learned] with greater precision," which
is exactly what is needed. (While this paper only addresses the
cognitive domain, understanding the affective domain is a
precursor to appreciation of the cognitive issues.)
The cognitive taxonomy was developed assuming that
There are different degrees of learning to which someone can
know and use information
The different degrees of learning are observable and
The degrees of learning are reasonably hierarchical
The first and second assumptions reflect the observation that
there are noticeable, measurable differences between a nov-
ice and an expert in how they use information. The third
assumption is based on the observation that successful dem-
onstrations of the higher degrees of learning are generally
not possible before successful demonstrations of mastery of
the lower levels.
In the early 1990s, David Langfordt8' updated Bloom's
taxonomy, renaming the learning objectives to Levels of
Learning (LoL). The six Levels of Learning, from lowest to
highest, are: Knowledge (K), Comprehension (C), Applica-
tion (Ap), Analysis (An), Synthesis (S), and Evaluation (E).
Bellamy and McNeill9'o01 modified Langford's definitions
to reflect the type of activities found in engineering educa-
tion. An example of how the Knowledge Level of Learning is
described can be seen in Table 1 (following page). The
description includes information from both the student's and
the teacher's point of view. The list of process verbs at the
end of the description is very helpful in distinguishing be-
tween the various levels of learning.

The Articulation Matrix is a concise way of presenting the
answers to Tyler's first two questions. Further, it presents the
data in a manner that makes it possible to also partially
answer Tyler's third and fourth questions. The matrix con-
sists of a set of rows (the learning objectives), a set of
columns (the class activities) and a set of letters indicating
the LoL impact, if any, each activity has on each learning
objective (see Figure 1, which will be discussed later). Tyler
included an early version of the matrix on page 50 of his
book. To better understand the matrix, consider how it is
used to help answer each of Tyler's four questions.
Question 1: Defining the Learning Objectives Answer-
ing the first of Tyler's questions requires stating the course's
learning objectives (competencies and associated changes in
LoL). The processes used to generate the learning objectives
are many and varied (e.g., 4, 5, and 6) and will not be

SABET Engineering Criteria 2000

discussed here. Once the learning objectives have been developed,
they can be entered into the Articulation Matrix. First, the competen-
cies are entered. Since there is often a hierarchy associated with the
competencies, the matrix allows for this by having Competency Cat-
egories as well as Competencies under each of the categories. Thus, in
Figure 1, there are two major competency categories (Engineering
Design Process and Working in Teams) and eight competencies (e.g.,
formulating the problem, team communication) shown.
Next, to complete the entry of learning objectives, the change in
LoL for the Competency Categories and Competencies must be en-
tered. The change in LoL is indicated by showing the required input
level and the desired output level in the second and third matarix
columns. Thus, in the matrix shown in Figure 1, the change in LoL for
"solving a problem" (competency 1.2) is from Unaware (a "U" in the
second column) to Application (an "A" in the third column).
Question 2: Defining Course Activities and Their Impact
Once the learning objectives have been entered, it is possible to
answer Tyler's second question. This is generally an iterative pro-
cess with the completed matrix showing the results of the final
iteration. The general process involves adding all the class activi-
ties, one at a time, to the matrix, indicating in the body of the matrix
which learning objectives are impacted by the activity, and finally
indicating the degree of learning possible using the activity.
Consider the fourth activity, "orally report to peers and class,"
shown in Figure 1. When this activity was entered into the matrix, it
was felt that it impacted all the shown competencies except for
competency 1.5. Further, the impacts were all judged to be at the
Comprehension LoL; that is, the "C" in the "solving a problem"
competency row indicates that when the activity is completed, the
students could have demonstrated mastery of "solving a problem"
at the Comprehension LoL. This example matrix is rather dense,
i.e., many of the activities impact on many of the learning objec-
tives. It is not uncommon to have less dense matrices.
There is an alternative, somewhat easier but less educationally
rigorous, method for completing the matrix. In this alternative
method, the competencies are loosely viewed as the needs and the
activities as the hows in a House of Quality.[ Thus, in filling out
the matrix, instead of indicating the LoL, a symbol indicating the
degree of impact (high, medium, low) that the activity has on the
competency is entered into the matrix. This method does give a
good picture of which activities have the biggest impact on the
learning, but a matrix completed using this method is much harder
to use when attempting to address Tyler's last two questions.
Question 3: Organizing the Course Activities Since the
previous step focused only on entering all the course activities into
the matrix, the columns in the matrix are generally not in the
desired order, i.e., the actual sequence followed in the course. This
can be seen in Figure 1 where there are "Out-of-Class Activities,"
shown late in the matrix, that would actually occur early in the course
(e.g., "read and summarize textbooks"). While the matrix may not
have the activities organized, it does contain information that can be

Activities of Students and Teachers at the
Knowledge Level of Learning"0l'

Knowledge (Information) Level of Learning
SHow do I know I have reached this level?
I can recall information about the subject, topic, competency,
or competency area; I can recall the appropriate material at
the appropriate time. I have been exposed to and have
received the information about the subject; thus, I can
respond to questions, perform relevant tasks, etc.

* What do I do at this level?
I read material, listen to lectures, watch videos, take notes; I
pass "true/false," "yes/no," "multiple choice," or "fill in the
blank" tests that demonstrate my general knowledge of the
subject. I learn the vocabulary or terminology as well as the
conventions or rules associated with the subject.

* How will the teacher know I am at this level?
The teacher will provide verbal or written tests on the subject
that can be answered by simply recalling the material I have
learned about this subject.
* What does the teacher do at this level?
The teacher directs, tells, shows, identifies, examines the
subject or competency area at this level.

* What are typical ways I can demonstrate my knowledge?
1. Answer "true/false, "yes/no, "fill in the blank, or
"multiple choice" questions correctly.
2. Define technical terms associated with the subject by
stating their attributes, properties, or relations.
3. Recall the major facts about the subject.
4. Name the classes, set, divisions, or arrangements that are
fundamental to the subject.
5. List the criteria used to evaluate facts, data, principles, or
ideas associated with the subject.
6. List the relevant principles and generalizations associated
with the subject.
7. List the characteristic methods of approaching and
presenting ideas associated with the subject (e.g., list the
conventions or rules associated with the subject).
8. Describe the generalproblem-solving method (i.e., the
techniques and procedures) or the methods) of inquiry
commonly used in the subject area.

* What are typical work products?
1. Answers to Knowledge-level quizzes ("true/false," "yes/
no," "fill in the blank," or "multiple choice").
2. Lists of definitions or relevant principles and generaliza-
tions associated with the subject.
3. Modifications of example problems presented in the
textbook; for example, modest changes in numerical
values or units; i.e., solutions to problems that were solved
using "pattern recognition."
* What are descriptive "process" verbs?
define label listen list memorize name
read recall record relate repea view

Chemical Engineering Education

r ABET Engineering Criteria 2000

used to help establish some of the desired organization.
Tyler suggests that the two major course organizational consid-
erations are 1) how to organize for the continual growth in LoL for
a competency, and 2) how to organize for cross-competency re-
quirements (i.e., pre- and/or co-requisite competencies). The Ar-
ticulation Matrix can help with the first, but not the second, of
these considerations. Assuming Bloom's taxonomy is hierarchi-
cal, the activities for a competency need to be scheduled to begin
with the lowest LoL and proceed sequentially to the highest LoL.
Thus, the matrix shown in Figure 1 suggests that for the "formulat-
ing the problem" competency, the Knowledge activities should
occur early, the Application activities should occur late, and the
Comprehension activities should occur in between.
Question 4: Assessing the Course Tyler's fourth question
concerns assessment. While the Articulation Matrix is not directly
concerned with student assessment, it can help in two assessment
areas. First, the matrix can be used to pre-assess the course to deter-
mine if it has the potential of delivering the desired objectives.
Second, the matrix can help select assessment instruments for the
various course activities.
Pre-Assessment of the Course. After all the course activities
have been entered and their impact entered in the matrix, the
matrix can be evaluated to confirm that the proposed course is
complete and has the potential to allow students to achieve the
predefined learning objectives. In pre-assessing the course, there
are four considerations:
1. Is there at least one course activity that impacts each of the
competencies (i.e., no empty rows)? If there are empty rows, one or
more course activities must be added to the matrix or an existing
activity must be modified so it impacts the competency.
2. Is there at least one competency impacted by each course activity

E 0
i I -I ,J I

Competencies li g 0 !
1. Engineering Des n Process U A
1.1 formulating the problem U KJ C C K CAA
1.2 solvng problem U A K C C C K K C CAA
1.3 implementing a solution U A KICI C K K C CAA
1.4 docmentr theprocess U A K C C C K K CCAA
1.5 using engineering/physical U K K
1.6 using uali les U 'A K C C C K K C CAA
SWorkI inTeeams U C
2.1 team dynamics U C K C C K K C
2.2 team communication U C K C C 1K K c
Level of Learning Legend U K C A
Unaware Knowledge Comprehension Application
Figure 1. A portion of ASU's first-year design course
Articulation Matrix.
Spring 1999

(i.e., no empty columns)? If there are empty columns,
the course activity does not impact any of the course
learning objectives and should be eliminated.
3. Does each row have an adequate number of appropriate
course activities? If the competency has an expected
multilevel change in LoL (e.g., from Knowledge to
Analysis), are there activities at the intermediate LoL's
(e.g., Comprehension and Application) as well as the
final expected level (e.g., Analysis)? Are there too many
or too few activities at any given level? Any "No's" must
be addressed by adding or removing activities, modifying
other activities so they impact the problem competency,
or changing the LoL associated with the competency to
match what is actually possible.
4. Do at least 75% of the competencies for a competency
category have course activities at the LoL stipulated for
the competency category? If the answer is "no," then
either more activities at higher LoL must be added or the
competency category LoL must be reduced to match the
LoL of the activities shown in the matrix.
The first two are easy checks and help ensure that all
the course learning objectives are addressed in one or
more of the activities and that there are no extraneous
activities (i.e., activities that have no impact on the de-
sired learning objectives). The answers to the third and
fourth questions are a bit more subjective. The third
assessment question focuses on each competency, to en-
sure that there are enough activities at the appropriate
LoL's so a student could reasonably be expected to achieve
the desired LoL by the end of the course. The fourth
question focuses on whether there are enough course ac-
tivities at a high enough LoL to ensure that the entire
competency category LoL is achieved. The use of 75% is
somewhat arbitrary and may be modified with experience.
Assessing the matrix shown in Figure 1 leads to the
following conclusions. First, each row has at least one
activity that impacts on the competency. Second, there
are two activities ("peer assess design notebooks," "watch
manufacturing videos") that appear to impact no learn-
ing objectives and should be considered for removal
(they actually impact several competencies not shown in
this partial view). Third, the mix of activities for each
competency is good. For example, there are three Knowl-
edge, five Comprehension, and two Application activi-
ties for the first set of competencies. It is possible that
there are actually too many Comprehension activities.
Finally, the competency LoL's support their competency
category LoL's. For example, five of the six competencies
(83%) under the "Engineering Design Process" compe-
tency category have activities at Application LoL, which is
the desired LoL for the competency category. It would
appear that this course is acceptable and should articulate.
Assessment Instruments. Once the expected LoL is

SABET Engineering Criteria 2000

known for an activity, the method of assessing whether the
students have achieved the LoL needs to be determined. As
with the learning objectives, the requirements of EC 2000 have
spawned many articles and workshops on assessment. Since
the LoL of the activity has been defined, there are several
places to find appropriate assessment instruments. First, the
work by Angelo and Cross1121 on classroom assessment can be
reviewed. Next, Bloom121 can be reviewed; it contains a num-
ber of typical testing methods that can be used for each LoL.
Third, the definitions of the various LoL's[g910] provide a vari-
ety of different ways of looking at each LoL, allowing the
generation of appropriate assessment instruments. For example,
material for Comprehension LoL[9'10 states that students should
be able to explain (orally or written) their solution process. This
suggests that for Comprehension LoL activities, a discussion of
the process should be required.

While the Articulation Matrix was developed to resolve the
first-year engineering design course articulation problem, it has
become clear that it has a wider application. Two applications
will be discussed: one that uses the matrix in course articulation
and one that uses it as part of the EC 2000 accreditation process.
Course Articulation Within a State The starting point for
this work was the fact that design courses did not articulate at
the three state universities. The task force developed a two-
step process to resolve this problem. In the first step, the task
force defined the desired learning objectives (six competency
categories and twenty-two competencies) and entered them
into a blank matrix, creating a "skeleton" Assessment Matrix
(Figure 1 shows part of this matrix; see Reference 13 for the
complete skeleton matrix). Much of the task force's work
involved explicitly defining the learning objectives and devel-
oping a complete glossary of operational definitions'131 ("op-
erational definition" is the agreed-upon meaning of the term)
for each Competency Category and Competency in the matrix.
Finally, the task force added several topic and activity con-
straints to ensure the course included the desired type of expe-
rience (e.g., at least two extensive, 3-to-6-week projects). Once
the skeleton matrix was completed, the task force was done;
the various schools then completed the matrix during step two
of the process.
In the second step of the articulation process, each school
(university, community college) that wanted to offer a course
that would articulate started with the skeleton matrix and con-
straints and then completed and assessed an Articulation Ma-
trix for their proposed course. The task force developed an
assessment checklist to aid in the assessment step. Any course
that passed the assessment step would articulate at all of the
three state universities.
The strengths of this process are twofold. First, having each
school start with the skeleton matrix allows considerable flex-

ibility in defining how the learning objectives are met. Each
school can use activities that suit its nature and strengths. The
only constraint is to have enough activities at the appropriate
LoL. Second, having a defined assessment step takes the un-
certainty out of the articulation process. A community col-
lege need no longer wonder if its course is satisfactory. Any
questions that do arise (e.g., Does that activity actually allow
Comprehension LoL?) can be easily resolved by supplying
samples of student work for the activity in question.
EC 2000 Accreditation Process While experience to
date has been primarily limited to using the matrix to resolve
articulation problems, the process of developing the matrix
is general and can be easily extended to defining a curricu-
lum (e.g., the Mechanical and Aerospace Engineering Depart-
ment at Arizona State University has developed the curriculum
matrix for its two undergraduate degrees). When using the
matrix for a curriculum, the following changes are made:
1. The learning objectives, i.e., the rows, are the objectives
related to the entire curriculum and not just a course.
2. The columns become the courses in the curriculum instead of
class activities.
3. The LoL impact indicates the maximum LoL expected to be
achieved in the course.
Part of the matrix for a chemical engineering curriculum is
shown in Figure 2. Looking at the rows, the curriculum
shown in the figure shows that the students are expected to
enter the curriculum at Unaware and to leave at Synthesis
LoL for "Modeling." How this transformation is accom-
plished is partially shown by looking at the two modeling
sub-competencies. The students are expected to achieve
Knowledge LoL about "conservation and accounting" in their
first-year chemistry courses and Analysis LoL in ECE 201.

1S c A C,

1.3 or anicchemist K A CA

2.2 conservation & accounting U S K K An
U N C Ap

Unaware Knowledge Comprehension Application
Level of Learning Legend

chemical engineering curriculum.
Chemical Engineering Education
Competencies W0 ) J

1.3 organic chemistry K C -
2. Modeling m I I I I I 11111111
2.1 principles of mode5i5ng U5 4-CA A A

Unaware Knowledge Comprehension Application
Level of Learning Legend
An S E
Analysis Synthesis Evaluation
Figure 2. Part of an Articulation Matrixfor a
chemical engineering curriculum.
Chemical Engineering Education

ABET Engineering Criteria 2000

The "principles of modeling" are developed to the Compre-
hension LoL in the first-year design class and are then demon-
strated at the Application LoL in the upper-division classes.
Looking at the columns of the matrix is also instructive.
Figure 2 shows that the conservation principles course (ECE
201) is expected to offer the students a chance to demon-
strate Application LoL for "calculus" and "principles of
modeling," and Analysis LoL for "conservation and account-
ing" modeling. How the course might achieve these LoL
goals is not shown in the matrix; this information would be
shown on the course matrix.
As with the course matrix, the curriculum matrix can be
used to sequence courses. The matrix in Figure 2 shows
there are six courses that have an impact on the calculus
competency. Based on the LoL shown in the matrix, it ap-
pears that Mathematics 270, 271, 272, and 274 should come
before ECE 201 (i.e., Comprehension before Application).
The "calculus" row in the matrix highlights the expectation
that students are not entering ECE 201 with the ability to
recognize when to use the calculus skills they have learned
(Application LoL); rather, Application LoL for "calculus"
will be achieved in ECE 201 and other upper-division courses.
Finally, the potential success of the curriculum can be
assessed much as a course is assessed. For the curriculum,
the third assessment question concerns whether there are,
realistically, enough courses to move the students through
the desired change in LoL. It is reasonable, at the lower
LoL's, to expect to be able to move a student through three
(and possibly four) levels in one course. But for the higher
LoL's, it is difficult to move through more than one or two
levels per course. It is not reasonable to expect to take a
student from Unaware through Synthesis or Evaluation LoL
in a single course. The matrix in Figure 2 is clearly not
complete; there are no courses shown at the Synthesis LoL
for any of the "modeling" competencies.
The use of a matrix to define a curriculum is not new. Olds
and Miller'[4 defined just such a matrix. The mapping be-
tween our matrix and that of Olds and Miller is simple. Their
"Program Objectives" become the Curriculum Competen-
cies, the "Implementation Strategies" become the courses in
the curriculum, and the "Performance Criteria" and "Assess-
ment Methods" become the LoL designations. An advantage
of the articulation matrix is that it facilitates assessment of
the curriculum.
One final note: it should be possible, using a set of Articu-
lation Matrices, to create a highly compact integrated picture
of a curriculum. The first Articulation Matrix in the package
would be the curriculum matrix. Then, using the curriculum
matrix's Competency Categories and Competencies as the
skeleton matrix, the Articulation Matrices for each course in
the curriculum would be created. The desired LoL changes
Spring 1999

for the course Competencies would come from the LoL changes
shown in the body of the curriculum matrix. For example, if
the curriculum matrix is that shown in Figure 2, then the
course matrix for ECE 201 would show "An" (Analysis) for the
LoL (out) column and "K" (Knowledge) for the LoL (in) column
for "conservation & accounting" competency. This package of
matrices documents the integrated nature of a curriculum, some-
thing required by ABET EC 2000.

A process that allows Arizona's universities and commu-
nity colleges to independently develop a first-year engineer-
ing design course that will articulate at all of the state's three
universities was the focus of this paper. The process uses an
Articulation Matrix that shows the educational relationship
(Level of Learning achieved) between a course's learning
activities and its learning objectives. The matrix was devel-
oped using the educational research of Tyler and Bloom. A
strong point of the process was development of the assessment
method used to determine if a course is acceptable (i.e., allows
the students to achieve the course learning objectives). The
matrix can be used for any course and is a good way to evaluate
a course syllabus. A similar matrix can be used to show how
curriculum competencies can be defined, an EC 2000 task.

1. Tyler, Ralph, Basic Principles of Curriculum and Instruction, The
University of Chicago Press (1949)
2. Bloom, Benjamin S., editor, Taxonomy of Educational Objectives:
Book 1. Cognitive Domain, ISBN: 0-582-280109-9, Longman (1956)
3. ABET, ABET Engineering Criteria 2000,
eac2000.htm, visited 7/15/98
4. Olds, Barbara M., and Ronald Miller, "An Assessment Matrix for
Evaluating Engineering Programs, J. of Eng. Ed., 87(2), (1998)
5. Rogers, Gloria M., and Jean K. Sando, Stepping Ahead: An Assess-
ment Plan Development Guide, Rose-Hulman Institute of Technol-
ogy, Terre Haute, IN (1996)
6. Martin, Flora, Eric Van Duzer, and Alice Agogino, "Bridging Diverse
Institutions, Multiple Engineering Departments, and Industry: A
Case Study in Assessment Planning," J. ofEng. Ed., 87(2) (1998)
7. Krathwohl, David R., Benjamin S. Bloom, and Bertram B. Masia,
Taxonomy of Educational Objectives: Book 2. Affective Domain, ISBN:
0-582-28239-X, Longman (1964)
8. Langford, David, Total Quality Learning Handbook, Langford Qual-
ity Education (1992) (see, visited 7/14/
9. McNeill, Barry W., and Lynn Bellamy, Introduction to Engineering
Design: The Workbook, 6th ed., McGraw-Hill Companies Inc., Primis
Custom Publishing (1998)
10. McNeill, Barry W., and Lynn Bellamy, LoL_definitions.doc, http://, visited 7/
11. Hauser, John, and Don Clausing, "The House of Quality," Harvard
Business Rev., 66(3), 63 (1988)
12. Angelo, Thomas A., and Patricia K. Cross, Classroom Assessment
Techniques: A Handbook for College Teachers, 2nd ed., Jossey-Bass
Inc. (1993)
13. "Articulation Matrix for First Year Engineering Course," http://, visited 7/22/98 O

ABET Engineering Criteria 2000



Michigan State University East Lansing, MI 48824

By now, most engineering faculty have accepted the
fact that accreditation of engineering programs ac-
cording to ABET EC 2000 is inevitable. No further
introduction or justification of the new criteria is required; it
is simply time to "just do it."
Over the past several years, EC 2000 has been the topic of
discussion at ASEE and FIE conferences, at assessment
workshops, and among engineering faculty nationwide. It
has been quite common to hear comments similar to those
we hear from students who are reluctant to begin a tough
assignment: "When is this due?" "Will this material be on
the final?" "What do I do to get a 'C'?" As Dr. Gloria
Rogers, Dean for Institutional Research and Assessment at
Rose-Hulman Institute of Technology, stated at the 1998
Annual ASEE conference,1 the hope is for engineering
programs to get more than a "C" as we proceed into imple-
mentation of successful assessment and improvement pro-
cesses. But as George Peterson, Executive Director of ABET,
confirms, no one expects this to be easy.1[2

There has been a noticeable turning of the tide. Among
these same reluctant faculty, there can be seen, at a mini-
mum, resignation to the fact that accreditation according to
EC 2000 will happen. Even more commonly observed is an
approach to assessment and program improvement as a schol-
arly activity that will yield positive outcomes; engineering
faculty across the country are rolling up their collective

Daina Briedis is Associate Professor in the De-
partment of Chemical Engineering at Michigan
State University. She received her PhD in chemi-
cal engineering at Iowa State University and her
BS in engineering science at the University of
Wisconsin-Milwaukee. She has served as a
chemical engineering program evaluatorof ABET
for over ten years and is a member of both the
Education andAccreditation Committee of AIChE
and the Engineering Accreditation Commission
Copyright ChE Division of ASEE 1999

sleeves to begin the task set before them.
I consider myself fortunate in having been able to serve as
a program evaluator on an EC 2000 visit while in the midst
of my own department's preparations for an EC 2000 accredi-
tation visit (in the fall of 1998), and thus observing from both
sides of the fence. There has been no better EC 2000 "crash
course" than these combined experiences.
In writing this paper, I do not represent ABET's Engineer-
ing and Accreditation Commission (EAC), since ABET is
deliberately not prescriptive about the nature of quality im-
provement processes adapted by individual programs. The
spirit of EC 2000 is, in fact, to encourage programs to
establish their own customized objectives and improvement
processes that are tailored to that particular institution's and
program's mission and are responsive to the needs of their
constituencies. Rather than proposing a set of instructions, this
article simply relates experiences and lessons learned. Two
topics that frequently surface in discussions about EC 2000 are
examined: constituency "buy-in" and closing of the improve-
ment loop. How these issues affect evaluation and institution-
alization of a program-improvement process will be addressed.

Several institutions already have well-established program
improvement processes in place. These institutions have
been motivated by various factors, including the desire to
achieve a vision, improvement of teaching, competition with
other institutions, state mandates, industrial linkages, or other
factors. 134] In most of these institutions, assessment and pro-
gram improvement are the modus operandi. For most of the
rest of us, this goal is yet to be achieved.
This is not to say that prior to EC 2000 engineering institu-
tions have been operating in an improvement vacuum; for
many years, program improvement has been integral to course
evaluations, curricular revisions, training and mentoring new
faculty, and interactions with employers and industrial advi-
sors. But we have been anecdotal about these methods. ABET
Chemical Engineering Education

ABET Engineering Criteria 2000)

criteria now ask us to become more structured, more fo-
cused, and much more quantitative regarding program im-
provement. Furthermore, we are directed to improve in the
direction of measurable goals-our program educational ob-
jectives-and our student (graduate) outcomes must demon-
strate how well we are doing in this endeavor. This addi-
tional formalism and documentation is what most faculty
find intrusive, in that such up-front planning, careful docu-
mentation, measurement against performance standards, and
analysis of improvement trajectories all take time and repre-
sent a departure from old habits.

Time is a factor. Preparation for an EC 2000 visit, much
less the design and implementation of a sustainable pro-
gram-improvement process, cannot be done overnight. More-
over, not all faculty and students can be expected to contrib-
ute willingly or to be 100% committed to the effort. A
foundational principle of EC 2000 is that program improve-
ment must be permanently integrated into how engineering
programs conduct business. Therefore, as Covey says, we
should plan "with the end in mind"'51 in order to develop a
sustainable process that the academic staff, faculty, and stu-
dents will be comfortable with for the long haul.
One way in which the level of sustained commitment to
these processes can be significantly impacted is by involving
program constituencies in the early planning and prepara-
tions. Leonard, et al.,161 describe two such approaches. Hopes
for permanent implementation and constituency "buy-in"
appear to be maximized if we draw upon current assessment
activities, leverage what has already been done, and involve
as broad a constituency support base as possible.

Since in most institutions the faculty have ultimate respon-
sibility for evolution of academic programs, development of
an improvement process may work best and impel faculty
most if the effort proceeds from faculty. Rather than a pro-
cess being dictated from outside or from above, faculty must
assume some ownership of the planning and implementation
steps. Many institutions have set forth in this mode.
Review the Old: Share the New At Michigan State Uni-
versity (MSU), a college-level ABET task force was estab-
lished in early 1997 to determine the feasibility of an EC
2000 accreditation visit for the 1998-99 cycle. No one as-
sumed a priori that a request would be made to ABET for
evaluation under the new criteria (this will no longer be an
option for accreditation beginning in 2001-02).
Comprised of a faculty representative from each program
(some of whom are ABET evaluators) and selected adminis-
trators, the group first endeavored to understand EC 2000
Spring 1999

and how improvement processes might support the mission
of our institution. The task force was also careful to accept
and use common definitions for EC 2000 terminology (see
Sando and Rogers171 and the NSF User-Friendly Handbook
to Project Evaluation'l8). While the ABET two-loop model
is useful in understanding steps in the processes for setting
objectives and for assessing outcomes, the task force worked
with a more traditional feedback model to visualize how EC
2000 fit into academic programs.9] The model provided
reference points on which to peg the focus of our discussions
and the results of our efforts.
The task force next began a thorough analysis of the
assessment status quo in the college. We inventoried the
existing assessment practices, both at the college level and
within programs. A complete review of the program self-
studies (Volume II) from the previous ABET visit was con-
ducted in order to identify items that overlapped with mate-
rial being requested for the new criteria. (Since then, Sarin
has published an inventory of this type of information.['01)
Because of our lack of expertise in assessment, we some-
times called on industrial and academic experts in this field
for advice. In addition, several task-force members attended
meetings and workshops to learn as much as possible about
best practices in assessment and program improvement. Cur-
rent literature on these topics was reviewed regularly. Most
importantly, information was freely shared among programs,
and reports of task-force progress were regularly transmitted
to the departmental faculty. Requests for input from depart-
mental faculty were equally frequent. Thus, while faculty
had not yet "bought into" the ideas, they were kept apprised
of the process from its inception.
Retrofitting With input and support from the college
faculty, in April of 1997, the task force voted to recommend
a request for evaluation according to EC 2000. Most of the
work from this point forward was carried out in the pro-
grams, but the task force maintained its role of facilitation
and oversight. The task force continued to edit existing
college-level assessment instruments for EC 2000 compat-
ibility by fine-tuning for assessment of the skills and at-
tributes represented in the Criterion 3 outcomes. The indi-
vidual programs were free to choose whether or not to in-
clude these college-level assessments in their own toolbox of
methods. We did not suggest the adoption of a single assess-
ment and evaluation model for the entire college (as proposed
by Aldridge and BenefieldE"1). With full knowledge of what
was available at the college level, however, the individual
programs could streamline their own assessment efforts.
Self-Evaluation The task force members took advantage
of two additional ABET documents relevant to the visit
preparations, both found in the Manual of Evaluation Pro-
cess. We regularly scored our own programs on the "Level
of Implementation" (Manual of Evaluation Process, Appen-

ABET Engineering Criteria 2000

dix A[121), which is completed by the program evaluator to assess the extent to
which programs have implemented several aspects of EC 2000. Another cali-
bration exercise was to perform the Program Deficiency Audit (PDA). It is
used by the visit team as a "roadmap" to criteria deficiencies and their resolu-
tion through the entire accreditation process. In our planning efforts, the PDA
helped several programs focus their process development efforts in the areas
perceived to be weakest.
Lessons Learned This type of college-level approach clearly demonstrated
four important points:
Sharing and review of information is a valuable practice. It is not necessary
that each program reinvent assessment instruments already proven to be
effective. In fact, a somewhat unified approach for the entire college is easier
to manage and may present a stronger case for sustainability to the ABET
program-evaluation team.
It proved time-efficient to retrofit current, in-house assessment and evaluation
practices for EC 2000 compatibility. More important than the savings in time
and effort was the fact that these were already part of the existing environ-
It was helpful to view our efforts through the eyes of an ABET evaluator.
Using the same documents as those used by program evaluators was useful in
focusing our planning and implementation efforts.
Even though buy-in from the entire faculty is desired, it was critical to have
one individual in each program serve as champion and coordinator of that
program's improvement efforts. In fact, as institutions look beyond EC 2000
visits, it is clear that someone or some group must assume responsibility for
maintaining the assessment and evaluation processes. Evaluators will
undoubtedly be looking for this confirmation.
At MSU, development of the EC 2000 environment was accorded enough
importance that task-force members were compensated in various ways for
their efforts. This typically amounted to a fraction of the academic year's
release time, a portion of a summer's salary, payment for student help, or some
combination of these. In the EC 2000 pilot visits that have occurred, many
programs have had "EC 2000 coordinators" who are individuals other than the
program administrator. Preparation for an EC 2000 visit and the institutional-
ization of continuous program improvement processes are significant respon-
sibilities that, if done well, consume more time than any program administrator
is able to provide. But support from the program administrator, the college


Figure 1. Program Improvement Process Model.

administration, and ideally, the institutional
administration is vital to the long-term suc-
cess of these efforts. While an ABET evalu-
ator may not worry too much about whether
or not the EC 2000 coordinator was adequately
compensated, financial support does attest to
administrative commitment to this effort.

Task-force efforts paved the way for work
that needed to be done at the program level,
where curriculum committees frequently as-
sumed responsibility for the bulk of the work.
Students also became more intimately in-
volved in the processes by virtue of their
membership on these committees and in re-
lated assessment-development activities.
Modeling We found the use of a process
model (see Figures 1, 2, and 3) to be an
effective framework for our planning strate-
gies. The model (Figure 1), first used by the
task force, was expanded to include a more
detailed representation of the relationships
among the assessment instruments, imple-
mentation strategies, constituencies, and the
academic program (Figures 2 and 3). Cur-
rent assessment literature addresses the vari-
ous types and hierarchal levels of assess-
ment."3'14] The process diagram helped us vi-
sualize how various levels of assessment would

Figure 2. Assessment and Improvement
Process at the curricular level. Detail of the
"Academic Programs" block in Figure 1.
Chemical Engineering Education

Performance Goals
Related to Program

* Knowledge
* Skills
* Attitudes


ABET Engineering Criteria 2000

integrate into the overall program improvement process.
Objectives and Outcomes Our department's educational objec-
tives and program outcomes had already been developed through
routine departmental and advisory meetings involving faculty,
students, industrial advisors, and alumni. Even though we had not
cemented a formal process at this point, we had involved the major
constituencies of our program and had a starting point in hand. Our
focus turned to implementation.
As discussed by Ewell,31 the major point of contact through
which any program achieves its educational objectives is the cur-
riculum, and therefore, we identified how each individual course
contributes to achieving our program objectives. A discretized
approach was not intended. No one course contributes in achieving
all objectives, and some contribute more strongly for some objec-
tives than others. Interestingly, even this preliminary analysis helped
us identify program weaknesses where objectives were not sup-
ported and outcomes were not realized.
Assessment-Just Do It Early on in our planning, we realized
that we could never become assessment experts. Reaching some-
what beyond the "comfort zone" of the faculty, we plunged into
"doing" the assessment without having read all of the literature
and with the knowledge that the assessment tools we had devel-
oped were not "perfect" or even tested. We borrowed some ideas
from colleagues and developed strategies of our own. This strategy
resembled the typical approach to open-ended design problems-
an initial design is completed, the preliminary results are evalu-
ated, and the process is repeated for an improved design.
Our curriculum committee determined that, to assess all pro-
gram outcomes and to give validating evidence (triangulation)
whenever possible, the chemical engineering program would supple-
ment the college-level surveys with several program-level instru-
ments. After initial trials of these surveys, several problems be-
came obvious. First, we had over-assessed. We therefore reduced
the scope of some of the surveys and decided to use others less
frequently. Second, it took little more than almost useless re-
sponses from the first version of a survey to result very quickly in a
second, more streamlined and effective instrument. Third, these
initial trials quickly established that surveys alone are not enough
to demonstrate student outcomes, as required by Criterion 3.
A better testimony of outcomes-the knowledge, skills, and

Figure 3. Assessment and Analysis.
Detail of the "Assessment" block in Figure 1.
Spring 1999

attributes acquired by our students-is student work.
Our department faculty chose student portfolios as the
major means to demonstrate and assess course and pro-
gram outcomes. Initially, a student task force was estab-
lished to assist in development of the portfolio approach.
It established a reasonable set of guidelines for the con-
tents of portfolios, basing its decisions on group discus-
sions and information from the literature.1'1 These stu-
dents gained an understanding of the philosophy of qual-
ity improvement and became familiar with ABET EC
2000, thereby becoming a supportive constituency.

Performance Goals An important element of assess-
ment analysis is establishment of performance goals, or
performance criteria-specific measures by which to de-
termine if objectives have been met. Programs should
have evidence confirming that students and program
graduates have achieved the desired level of performance.
Performance goals may include such measures as 1) a
certain percentage of satisfactory responses on a survey,
2) a target hiring rate for new graduates, 3) specific skills
or attributes demonstrated by students, 4) a minimum
"score" on student portfolios, or 5) a minimum grade
point average. Such performance goals are not only mea-
sures of acceptable achievement of objectives, but are
also an indication of the relative importance of the objec-
tives to the constituencies-the higher the achievement
standard, the higher the implied priority.

Closing the Loop In all of the preparation for the new
criteria, it seems that more attention has been paid to
assessment rather than what is done with the results of the
assessment evaluation. "Closing the loop" is possibly the
key to EC 2000; many evaluators have found this to be the
weakest link in the implementation of program improve-
ment processes. This step can be facilitated if programs use
the mechanisms already in place to complete this step.
The academic governance and accountability systems
in most engineering colleges are fairly traditional. All
academic programs have regular meetings of the entire
faculty and of specific subcommittees of the whole. Fac-
ulty performance is typically reviewed annually by the
head or chairperson. Faculty and staff retreats are com-
mon, and advisory board meetings occur periodically.
These regular deliberations provide a venue for discus-
sion, review, and action on items related to EC 2000.
Using the existing structure enhances the sustainability
of the processes and demonstrates to an ABET evaluator
that they are "ongoing." If a program has a person or
subcommittee responsible for the continued oversight of
program-improvement efforts, it is not an onerous task
to include regularly in these meetings discussion or ac-
tion items on program improvement.

SABET Engineering Criteria 2000

A flowchart of our program
improvement process is shown in
Figure 4. Included is a list of typi-
cal departmental activities as well
as a timeline for administration
of assessment tools. The only new
element is the Program Review
Meeting. The objectives of this
meeting are to review the results
of the assessment analyses, to rec-
ommend improvement strategies
based on the results, and to pri-
oritize the recommendations. The
Program Review Meeting in-
volves at least one representative
from each of our major constitu-
ent groups. The outcomes of the
meeting are forwarded to depart-
ment faculty and to the industrial
advisory board for recommenda-
tions and implementation.
Increasing Participation Dur-
ing our program's planning pro-
cess, the sphere of constituency
involvement gradually expanded.
Individual faculty were given the
responsibility for describing the
strategies by which program ob-

Figure 4. Program improvement process and
assessment timeline.

jectives were achieved and out-
comes demonstrated in his or her course. This naturally led
to the development of course learning objectives as a set of
benchmarks toward the achievement of program objectives.
Later in the process, the chairperson and faculty contrib-
uted to writing the self-study report; several faculty mem-
bers were directly involved in the design and implementa-
tion of survey instruments. This involvement encouraged
faculty to become more knowledgeable not only about the
contents of the self-study, but also about the practical as-
pects of executing the processes required by EC 2000.
Members of the industrial advisory board (employers,
alumni, and advisors) were involved in development of the
program-improvement process through the regularly sched-
uled meetings of this body where the program's educational
objectives were discussed and approved. Board members
gave recommendations on best practices for surveying and
assessment. Regular reports to the board from the chairper-
son and the ABET coordinator kept the group apprised of
EC 2000 activities in the department.
Students were familiarized with our program's educational
objectives and with course learning objectives and expected
outcomes. More than just being mentioned at the beginning
of the course, learning objectives and expected outcomes

ag Acii Assessment
Beginning of Fall
Semester Employer
Advisory Board surveys at
Meeting Employer Fair
End of Fall Faculty Meetings evaluations
Semester Porfolios
Annual Co-op Reviews
Beginning of Faculty
Spring Semester Iew Alumni phone
Faculty Meetinogslete
End of Spring -evaluations
Semester Portfolios
Co-op Reviews
PR06RAM Senior Exit
REVIEW Interviews
COMMITTEE End-of-year
innin o MEETING web survey
FB..irmninq of

tics course taken by most engineering students-this served to
benefit the entire college. On the other hand, a few chemical
engineering faculty collaborated with their counterparts in chem-
istry to discuss the restructuring of a physical chemistry se-
quence. Both approaches can be effective.
Who To Tell? Olds and Miller[161 emphasize the impor-
tance of reporting back to constituencies. Not only should
constituencies be involved in the program-improvement pro-
cesses, but they should also be made aware of the results of
which they have been a part. Positive results catalyze "buy-in."
As is evident above, students are one of the major con-
stituencies in our department. Without positive feedback, all
that most of them see of the assessment process is the portfo-
lios they must organize, the surveys that must be completed,
and an occasional reference to something called an "abet"! It
is gratifying to be able to come to students and say, "We are
emphasizing this material in class this year because last
year's graduates felt that it was a weakness in our curricu-
lum," or "This course is being offered to help you develop
more of the skills that your future employers think are vital."
Positive results of improvement efforts are also a good
motivator for faculty commitment. Our initial use of student
portfolios yielded good feedback to faculty. Although ini-
Chemical Engineering Education

were integrated into the classroom
culture. Learning objectives were
used to chart progression of the
course material; student portfo-
lios required student self-assess-
ment in achievement of outcomes.
We also involved our students in
administration of the phone sur-
vey and in analysis of the survey
results. Other institutions have in-
volved students directly in sur-
vey design. ABET program evalu-
ators will undoubtedly find that in-
terviews with students will give a
strong indication of their involve-
ment in program improvement.
At some point, the assessment
results may suggest changes that
require involvement of support-
ing departments (e.g., chemistry,
physics, mathematics). Whether
this interaction occurs at the col-
lege or program level may be de-
cided by the extent of the needed
change and whether one or more
engineering programs are in-
volved. At MSU, an inter-college
committee was established to ad-
dress the relevance of the statis-

' Repeat cycle

ABET Engineering Criteria 2000

tially viewed as burdensome, portfolios were adopted by the
faculty as a major means of outcomes assessment.
Off-campus constituencies should also be informed of the
results of their feedback through existing channels such as
reports to alumni through newsletters, meetings with advi-
sory board members, and regular communication with em-
ployers. Keeping alumni and industrial representatives in-
formed as to how their feedback is being used for program
improvement can help encourage continued involvement and
can engender a sense of "connectedness" to the program.

Having presented an example scenario for developing the
EC 2000 environment and preparing for a visit, let's briefly
look at the other side of the fence.
Questions and Answers Even though EC 2000 evaluators
are all trained with similar materials, they will approach a
site visit with different predispositions. This is one thing that
has not changed from evaluation under the present ("old")
criteria. But for EC 2000, all evaluators and team chairs will
be looking for answers to several questions:
What are the program objectives?
Are program objectives linked to appropriate outcomes?
Are the program outcomes (and therefore objectives) being
Are the ABET EC 2000 defined outcomes (Criterion 3, a-k)
being achieved within the context of program outcomes? Is
there evidence to support this?
What processes are in place for enhancing the program? Is
the process improvement loop working and ongoing?
Are the constituencies involved? Is there evidence to
support this?
How much do members of the constituency groups know
about these topics? Faculty should be familiar with these
elements and should have taken some part in their realiza-
tion. Students should also be familiar with objectives both at
the program and the course levels, and they should know
that certain outcomes are expected of a graduate of the
program. Both faculty and students should be able to de-
scribe their participation in the processes and actions that
have been taken to improve the program. The evaluator will
most likely conduct interviews with faculty and students (and
possibly with other constituencies as well) that will provide a
clear indication of the level of implementation and the level of
commitment to the program improvement processes.
Self-Study "Must-Have's" The self-study report is still
the first contact that an evaluator has with a program. Based
on the experiences of the five EC 2000 pilot schools and
their evaluation teams, a better perspective has been gained
on how self-studies can be most informative. The Self-Study
Instructions are now considerably more prescriptive to allow
for more consistent evaluation among programs. The topics
Spring 1999

they delineate are also a useful guideline for preparation for
an EC 2000 visit.
It is no longer the responsibility of the evaluator to pore
over course material to sift out evidence in support of a
program's claims. Evidence of ongoing processes and docu-
mentation of outcomes should be clearly laid out in the self-
study and in materials presented at the time of visit. It is the
responsibility of the program to provide documentation of the
capabilities of their students and graduates. Programs must be
able to identify both the strategies used to achieve outcomes
and the evidence that substantiates the success of these efforts.
In conclusion, effective use of a combination of existing
assessment practices and involvement of a broad base of
constituencies are the key elements in building an effective
EC 2000 environment in engineering colleges. While imple-
mentation of program-improvement processes requires sig-
nificant resources, the resulting program improvements are
evident in a surprisingly short term and, in the long term,
hold promise for keeping pace with the demands of the
engineering profession.

1. Rogers, G., "Assessment for Improvement: Coming Full Cycle,"
presented at the 1998 Annual Meeting, ASEE, Seattle, WA (1998)
2. Peterson, G., "A Bold New Change Agent," in How Do You Mea-
sure Success?, ASEE Professional Books, Washington, DC (1998)
3. Ewell, P.T., "National Trends in Assessing Student Learning," J.
Eng. Ed., 87, 107 (1998)
4. Houshmand, A.A., C.N. Papadakis, and S. Ghoshal, "Benchmarking
Total Quality Management Programs in Engineering College,"
QMJ, Summer (1995)
5. Covey, S.R., The Seven Habits of Highly Effective People, Fireside,
NY (1990)
6. Leonard, M.S., D.E. Beasley, K.E. Scales, and D.J. Elzinga, "Plan-
ning for Curriculum Renewal and Accreditation Under ABET En-
gineering Criteria 2000," 1998 ASEE Annual Conference Proceed-
ings, Seattle, WA (1998)
7. Rogers, G.M., and J.K. Sando, "Stepping Ahead: An Assessment
Plan Development Guide," Rose-Hulman Institute of Technology
8. Stevens, F., F. Lawrenz, L. Sharp, "User-Friendly Handbook of
Project Evaluation," J. Frechtling, ed., NSF
9. Fisher, P.D., "Assessment Process at a Large Institution," 1998
ASEE Annual Conference Proceedings, Seattle, WA (1998)
10. Sarin, S., "A Plan for Addressing ABET Criteria 2000 Require-
ments," 1998 ASEE Annual Conference Proceedings, Seattle, WA
11. Aldridge, M.D., and L.D. Benefield, "Assessing a Specific Pro-
gram," in How Do You Measure Success?, ASEE Professional Books,
Washington, DC (1998)
13. Shaeiwitz, J.A., "Classroom Assessment," J. Eng. Ed., 87, 179
14. Ressler, S.J., and T.A. Lenox, "Implementing an Integrated Sys-
tem for Program Assessment and Improvement," 1998 ASEE An-
nual Conference Proceedings, Seattle, WA (1998)
15. Panitz, B., "The Student Portfolio: A Powerful Assessment Tool,"
ASEE PRISM, 5(7) (1996)
16. Olds, B.M., and R.L. Miller, "A Measure of Success,"ASEE PRISM,
7(4) (1997) D

ASEE Annual Conference & Exposition

Charlotte, N.C.

June 20 23, 1999

0213 LabVIEW for Chemical and Mechanical Labs Sunday
Jim Henry, Charles Knight
University of Tennessee-Chattanooga
0214 Green Engineering Curriculum Development Sunday
David Allen, David Shonnard
University of Texas, Michigan Technological University

1313 Promoting & Rewarding Effective Teaching Monday
U Promoting and Rewarding Effective Teaching
Richard Felder, Rebecca Brent, Douglas Hirt, Debi
Switzer, Siegfried Holzer
North Carolina State University, Clemson University,
Virginia Tech
1613 Innovative ChE Experiments and Demos Monday
U A Heat Exchanger as a Student Design Project
M. Mavromihales
University of Huddesfield
U Experiments to Accompany a First Engineering
Thermodynamics Course
T. Scott, John O'Connell
University of Virginia
C Cost Effective Experiments in Chemical Engineer-
ing Core Courses
Stewart Slater, Robert Hesketh
Rowan University
I Classroom Demonstrations of Separation Process
Keith Schimmel
North Carolina A&T State University
C Introduction of Process Dissection into the Under-
graduate Laboratory
Robert Ybarra
University of Missouri-Rolla
2213 Revitalizing Traditional ChE Courses I Tuesday
C ChE Fundamentals Course Better Learning
Through Computer-Based Delivery
Billy Crynes
University of Oklahoma
B Using Your Unit Operations Lab
Valerie Young Ohio University

U The Vertical Integration of Design in Chemical
Ronald Gatehouse, George Selembo Jr., John McWhirter
Pennsylvania State University
I Group Projects-Based Final Exams: A Novel
Approach to Integrate Fundamentals and Practical
Pedro Arce Florida State University
U Raising the Level of Questioning in the
Undergraduate ChE Curriculum
Anthony Muscat
University of Arizona
2313 Revitalizing Traditional ChE Courses II Tuesday
C The Evolution of Engineering Incorporating
Biology into Traditional Engineering Curriculum
Jennifer Maynard, Anneta Razatos
University of Texas
] Catalytic Oxidation Experiment for Chemical
Reaction Engineering
Robert Hesketh, Stewart Slater
Rowan University
C A Real-Time Approach to Process Control Education
William Svrcek, Donald Mahoney, Brent Young
University of Calgary and Hyprotech
U Structured Trouble-Shooting in Process Design
Anthony Vigil, Dendy Sloan, Ron Miller
Colorado School of Mines

2513 Getting the Best Students to Enter ChE Tuesday
C Role of the Honor Program in the Attraction and
Retention of the Brightest and the Best ChemE
Pedro Arce Florida State University
C The Promise of Silver and Gold: Not the Only Way
to Attract and Retain a Devoted Miner
Robert Ybarra, Douglas Ludlow
University of Missouri-Rolla
B How to Attract Top Students to a Chemical
Engineering Program The Experience at NJIT
Dana Knox, Reginald Tomkins
New Jersey Institute of Technology
Chemical Engineering Education

U Outreach and Recruitment to Attract Students to
Chemical Engineering
Robert Hesketh, Stephanie Farrell, Zenaida Keil,
James Newell
Rowan University
3213 Process Safety in the ChE Curriculum Wednesday
B Teaching Chemical Process Safety: A Separate
Course Versus Integration into Existing Courses
Anton Pintar
Michigan Technological Universityv
Panel Discussion
3513 ABET 2000: Improving ChE Education? Wednesday
Preparing for the First ABET Accreditation Visit
under Criteria 2000
Gary Patterson
University of Missouri-Rolla
A Process for Developing and Implementing an
Assessment Plan in ChE Departments
James Newell, Heidi Newell, Thomas Owens, John
Erjavec, Rashid Hasan, Steven Sternberg
Rowan University and University of North Dakota
Performance Assessment of EC-2000 Student
Outcomes in the Unit Operations Laboratory
Ron Miller, Barbara Olds
Colorado School of Mines
B Development of a Dynamic Curriculum Assessment
John Wagner, David Finley
Tri-State University
Developing an Assessment Plan to Meet ABET EC
Anton Pintar, Besty Aller, Tony Rogers, Kirk Schulz,
David Shonnard
Michigan Technological University

] Round 1: The Curricular Aftermath
Dennis Miller, Daina Briedis
Michigan State University

3613 Innovative Uses of Computers in ChE Wednesday
I Implementing Computational Methods into Classes
Throughout the Undergraduate Chemical
Engineering Curriculum
William Perry, Victor Barocas, David Clough
University of Colorado
I Laptop Computers and Curricula Integration
Jerry Caskey
Rose-Huhnan Institute of Technology
Integrating Research Into The Undergraduate
Curriculum NASA's Microgravity Bioreactor
Shani Francis, Keith Schimmel, Neal Pellis
North Carolina A&T State University and NASA
A Phenomena-Oriented Environmentfor Teaching
Process Modeling: Novel Modeling Software and
Its Use in Problem Solving
Alan Foss, Kevin Geurts, Peter Goodeve, Kevin Dahm,
George Stephanopoulos, Jerry Bieszczad,
Alexandros Koulouris
University of California, Berkeley and Massachusetts
Institute of Technology
Teaching Material and Energy Balances on the
Alec Scranton, Randy Russell, Nicholas Basker, Lisa
Michigan State University
Virtual Laboratory Accidents Designed to Increase
Safety Awareness
John Bell, Scott Fogler
University of Michigan


1113 ChE Div. Executive Committee Meeting/Breakfast Monday Morning

1413 ChE Chairpersons Luncheon Monday
This luncheon meeting will provide an opportunity for department chairpersons to exchange ideas and information
about issues relevant to chemical engineering. The meeting will open with a discussion of the following issues:
enrollment, placement, accreditation, and new curricular trends.
1713 ChE Division Reception/Mixer Sponsored by the CACHE Corporation- Monday Evening

2613 Union Carbide Lectureship Award Presentation Tuesday Afternoon
] Particle Dynamics in Fluidization and Fluid-Particle Systems
Dr. L. S. Fan The Ohio State University
2713 Chemical Engineering Division Dinner Tuesday Evening

3413 ChE Div. Business Meeting/Luncheon Wednesday
More information about the 1999 ASEE Annual Conference & Exposition can be found at
Spring 1999

Random Thoughts...


Students Who Are Disappointed

With Their Last Test Grade

FROM: R.M. Felder

North Carolina State University

Raleigh, NC 27695

Dear Students:
Many of you have told your instructor that you un-
derstood the course material much better than your last
test grade showed, and some of you asked what you
should do to keep the same thing from happening on
the next test.
Let me ask you some questions about how you pre-
pared for the test. Answer them as honestly as you can.
If you answer "No" to many of them, your disappoint-
ing test grade should not be too surprising. If there are
still a lot of "No"s after the next test, your disappoint-
ing grade on that test should be even less surprising. If
your answer to most of these questions is "Yes" and
you still got a poor grade, something else must be
going on. It might be a good idea for you to meet with

Richard M. Felder is Hoechst Celanese Pro-
fessor of Chemical Engineering at North Caro-
lina State University. He received his BChE from
City College of CUNY and his PhD from
Princeton. He has presented courses on chemi-
cal engineering principles, reactor design, pro-
cess optimization, and effective teaching to vari-
ous American and foreign industries and institu-
tions. He is coauthor of the text Elementary
Principles of chemical Processes (Wiley, 1986).

your instructor or a counselor to see if you can figure
out what it is.
You'll notice that several of the questions presume
that you're working with classmates on the homework-
either comparing solutions you first obtained individu-
ally or actually getting together to work out the solu-
tions. Either approach is fine. In fact, if you've been
working entirely by yourself and your test grades are
unsatisfactory, I would strongly encourage you to find
one or two homework and study partners to work with
before the next test. (Be careful about the second ap-
proach, however; if what you're doing is mainly watch-
ing others work out problems you're probably doing
yourself more harm than good.)
The question "How should I prepare for the next
test" becomes easy once you've filled out the checklist.
The answer is...
"Do whatever it takes to be able to answer 'Yes'
to most of the questions."

Good luck,
Richard Felder

Copyright ChE Division of ASEE 1999

Chemical Engineering Education


Test Preparation Checklist

Answer yes if you did these things regularly, not just occasionally.

1. Did you make a serious effort to read and understand the text? (Just
hunting for worked-out examples exactly like the homework problems
doesn't count.) Yes No

2. Did you work with classmates on homework problems, or at least
check your solutions with others? Yes No

3. Did you attempt to outline every homework problem solution before
working with classmates? Yes No
4. Did you participate actively in homework group discussions
(contributing ideas, asking questions)? Yes No
5. Did you consult with the instructor or teaching assistants when you
were having trouble with something? Yes No
6. Did you understand ALL of your homework problem solutions when
they were handed in? Yes No
7. Did you ask in class for explanations of homework problem solutions
that weren't clear to you? Yes No

Test preparation

8. Before the test, did you carefully go through the study guide and
convince yourself that you could do everything on it? Yes No

9. Did you attempt to outline lots of problem solutions quickly, without
spending time on the algebra and calculations? Yes No

10. Did you go over the study guide and problems with classmates and
quiz one another? Yes No

11. Did you attend the review session before the test and ask questions
about anything you weren't sure about? Yes No

12. Did you get a reasonable night's sleep before the test? (If your
answer is no, your answers to 1-11 may not matter.) Yes No


The more "Yes" responses you recorded, the better your preparation for the test. If you
recorded two or more "No" responses, think seriously about making some
changes in how you prepare for the next test.

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

Spring 1999

, classroom

Important Concepts in Undergraduate



University of Colorado Boulder, CO 80309-0424

Most chemical engineers will never design a reac-
tor; they will, however, often be in a position to
specify a reactor type, size, material, and design.
Or, they will be asked to analyze an existing reactor to 1) fit
a new reaction at a required production rate into the reactor,
or 2) improve product quality, or 3) determine how to
squeeze more production from the reactor without spend-
ing much money, adversely affecting the product, or blow-
ing anything up.
Most industrial chemical reactions are scaled up and put
into production without detailed knowledge of the chemical
kinetics and physical chemistries that affect the reactions.
Quite often, reaction engineers must design using instinct,
an understanding of how other, similar systems behave, and
a proper application of the important concepts related to
reactor design. We present a concise list of important kinet-
ics, thermodynamics, reactor design concepts, rules-of-thumb,
and applications that chemical engineering undergraduates
need for entry-level industrial positions or to start graduate
studies. Most are from texts widely used in undergraduate
courses.1'I Although many other aspects of kinetics and reac-
tor design are arguably as important for particular problems,
our list should serve as a solid foundation for attacking the
types of problems typically encountered by recent graduates.
Successfully teaching undergraduates to remember the
points below is only half the battle. We want them to be able
to do things with the information-analyze, design, specify,
simulate, estimate, explain, etc. Felder and Brent161 discussed
the use of instructional objectives as a route to incorporating
higher-level thinking skills into undergraduate courses. They
differentiate between simply listing course topics in a sylla-
bus and writing proactive objectives that teach students to
apply the factual information to problems. In a forthcoming
* Address: Huvard Research and Consulting, Inc., 12218 Prince
Philip Lane, Chesterfield, VA 23838

paper, we will use the list as a starting point for a set of
instructional objectives for an undergraduate kinetics and
reaction engineering course.

1. Thermodynamics does not predict kinetics. A more
negative, free-energy change (i.e., a larger equilibrium
constant) does not imply a faster reaction rate.
2. Catalysts can only increase the rate of processes that
are thermodynamically favored; they cannot initiate
reactions that are not thermodynamically feasible. A
catalyst does not change AG, AH, or the equilibrium
3. Three of the most important calculations for a reactor
Adiabatic temperature: if the heat released for an
exothermic reaction is not removed, this temperature
will be attained at complete conversion
Equilibrium composition: no reactor can produce

John L. Falconer is Professor of Chemical Engi-
neering at the University of Colorado. He re-
ceived a BES from the Johns Hopkins University
and his PhD from Stanford University. His re-
search interests are in heterogeneous catalysis,
photocatalysis, and zeolite membranes for sepa-

Gary S. Huvard earned his BS in chemistry
from Campbell College and his PhD in chemical
engineering from North Carolina State Univer-
sity. He spent eight years with the Corporate
Research Group at BFGoodrich and three years
with du Pont's Tyvek Technical organization be-
fore establishing a private practice in 1989.

Copyright ChE Division of ASEE 1999

Chemical Engineering Education

catalyst surface area before deactivation).

predicted by equilibrium, but we can list of
often choose which reactions to list of
consider in the equilibrium calcula- n
tions (see below). thermos
Isothermal heat load: heat must be react
removed (added) at the same rate at concepts
which it is generated (consumed) by thumb
reaction to keep a reactor isothermal. applica
4. As temperature increases for an chemical
exothermic reaction, equilibrium undergra
conversion decreases. For an for eni
endothermic reaction, equilibrium industrial
conversion increases, to start
KINETICS: our list sl
1. As long as a reaction is not limited for attack
by equilibrium or mass transfer, then of problem
longer reaction times, higher encounter
temperature, and more catalyst all gra
increase conversion. A reaction that
takes place in one hour at 2000C
could take place in less than one second at 4000C.
There are exceptions for certain ionic polymerization
reactions (negative apparent activation energies).
2. The rate of reaction is often the product of a rate
constant, which usually increases exponentially with
temperature (relatively few reaction rates decrease
with temperature) and reactant concentrations raised
to some power. The activation energy is the term in
the exponential that determines how fast the rate
increases with temperature.
3. Most chemical processes involve multiple reactions.
Higher temperatures increase selectivity for reactions
with higher activation energies. Higher reactant
concentrations increase selectivity for reactions with
higher reaction orders.
4. Local concentrations determine reaction rates (e.g., if
an insoluble solid product forms from a pure liquid
reactant, the concentration of reactant does not
5. The steady-state approximation can be applied to a
series of reaction steps by assuming that all reaction
steps proceed at the same rate.
6. The rate-determining step in a series of reaction steps
is the step furthest from equilibrium, and all the other
steps are assumed to be in quasi-equilibrium.
7. For homogeneous reactions, rates are proportional to
volume. In contrast, for heterogeneous reactions, rates
are proportional to surface area (interphase area or
Spring 1999

yields of products beyond those



nt a
r de,
{, ra
b, a
es. .
ng t
ms ty

mics, 1. A kinetic rate expression cannot
sign reliably be extrapolated outside the
les-of- concentration, conversion, or tempera-
nd ture regime where the data were
s that obtained.
sneering 2. For an uncatalyzed reaction, the rate
es need law (rate constant, activation energy,
evel order of reaction) is an intrinsic
itions or property of the reaction and not the
late reactor. For a catalyzed reaction, the
rate law changes when the catalyst is
serve changed.
nation 3. For a first-order, isothermal reaction,
he types fractional conversion (X) depends on
time but not the initial concentration:
icay X=l-e Thus, half life (t2: the time
y recent for X=0.5) contains the same informa-
s. tion as the rate constant (k). t,,2=(ln 2)/k
4. A reaction mechanism can be
suggested, but not proven, by fitting data to a rate
expression derived from a reaction sequence.
5. Rate parameters can be determined from experimental
kinetic data (conversion versus time) in batch or plug
flow reactors by the methods of integration and
differentiation. The method of integration, which
makes use of data in integrated form, is preferred over
the method of differentiation (which exacerbates the
impact of experimental error). If rates are determined,
regression can be used to determine the rate param-

1. Material balances on individual components are most
useful for reactor design:
Accumulation() = In(+) Out(+) + Generation by reaction(+)
2. For systems with multiple reactions, at least as many
material balances should be solved as the number of
independent reactions, but a material balance can be
solved for every component in the system.
3. The first law of thermodynamics applies to reactors,
both closed and open systems, and is used to deter-
mine reactor temperature:

dU / dt = (FiH,)in (FiHi)out +Q- W

where U is the total internal energy of the system, Fi is
the molar flow rate into or out of the reactor of a given
component, Hi is the enthalpy per mole of a given

component at inlet or outlet conditions (the heat of
reaction is contained in these terms), Q is the heat
added per time, and W is the work done by the system
per time.
4. For reactions where the heat of reaction is indepen-
dent of temperature, the temperature change in an
adiabatic reactor is proportional to the conversion.
5. In viscous reaction systems, heat developed by
agitator work is often an important term in the energy

1. Except for an ideal plug flow reactor (PFR), not all
molecules spend the same amount of time in a flow
reactor, and the residence time distribution can affect
both rate and selectivity.
2. The material balances for batch reactors (BR) and
ideal PFRs are mathematically equivalent; time in a
BR is equivalent to residence time in an ideal PFR.
3. Material entering an ideal continuous stirred tank
reactor (CSTR) undergoes a step change in concentra-
tion and temperature. An ideal CSTR operates at the
exit temperature and concentrations.
4. Although all reaction occurs in an ideal CSTR at
constant concentration and temperature, molecules
flowing through both ideal and real CSTRs have a
broad distribution of residence times. The residence
time distribution of an ideal CSTR is exactly known.
5. CSTRs are often used in series to decrease the average
residence time required for a given conversion
(relative to a single, large CSTR) or to narrow the
residence time distribution to one closer to that of a
6. Reactor temperature and concentrations can be
sensitive to feed conditions. Reactor behavior is
nonlinear because of the exponential Arrhenius rate
constant, and reactors are the most likely equipment in
a plant to explode.
7. For an exothermic reaction in a nonisothermal CSTR
(tfeedftreactor), multiple steady states can exist (i.e., the
material and energy balances have multiple solutions).
Multiple steady states are the result of energy feed-
back and the nonlinear behavior of the rate constant.
This can result in an unstable operating condition
leading to a quench (the reaction stops) or a runaway
(the reactor overheats). Either situation can be
dangerous and is to be avoided.
8. Reactor tank volume, and thus heat generated for a
homogeneous exothermic reaction, increases as the
cube of the reactor dimension, but heat transfer
through the external surface increases only as the

square, so temperature control is much more difficult
for larger reactors without internal cooling coils. For
exothermic reactions in jacketed reactors, an upper
limit on reactor volume exists. If the reaction is
carried out in a reactor larger than this, the heat cannot
be removed as fast as it is generated without other
means for cooling.
9. For gas-phase reactions, when the number of moles
changes due to reaction, the concentration of reactants
changes as a result, and flow rates and reaction rates
also change.
10. For series reaction, the more important variable is
space time or reaction time, and for positive-order
kinetics, higher selectivity to an intermediate is
obtained in a PFR than in a CSTR.

1. A catalyst usually lowers the activation energy for
2. The three most important attributes of a catalyst are
selectivity, activity, and stability. Often, selectivity is
the most important attribute.
3. All catalysts eventually deactivate, usually due to a
loss of catalytic sites.
4. A catalyst does more than allow a system to achieve
its most thermodynamically stable state; it can
selectively accelerate a desired reaction. In the
majority of industrial processes, the products are not
those expected from equilibrium conversion for all
5. Many industrial reactions are limited by diffusion
(mass- transfer limited). Concentration gradients
external to a catalyst particle are determined from
mass-transfer coefficient correlations. Concentration
gradients within a porous catalyst particle are ac-
counted for by an effectiveness factor.
6. If a catalyst increases the rate constant of a forward
reaction, it also increases the rate constant of the
reverse reaction (microscopic reversibility).

1. Whenever reaction rates are of the same magnitude as,
or faster than, the mixing rate in a stirred reactor,
mixing will have a serious impact on results. Poor
mixing is a primary source of variability in products
made in batch reactors. The results for a reaction run
in a poorly mixed CSTR may deviate strongly from
those expected.
2. There is no single "correct" agitator type. Different
agitator designs may perform equally well (or equally
poorly) for a given application. Although some
Chemical Engineering Education

detailed design calculations can be carried out,
workable designs are often developed by trial and Fid M
3. Many reactions involve shear-sensitive materials,
which severely limit the maximum mixing rate and
make impeller and reactor design important. Mixing The Liquid-Vapor Transition of
becomes the limiting factor (see item #1). Metals

THE REAL WORLD Friedrich Hensel
1. Real processes involve multiple reactions with and William W.Warren, Jr.
multiple heat effects.
This is a long-needed general introduction
2. Most industrial chemical reactions are exothermic and to the physics and chemistry of the liquid-
heat transfer is often the most important design vapor phase transition of metals. Friedrich
criteria. Hensel and William Warren draw on cutting-
3. Most bioreactions can only be carried out within a edge research and data from carefully selected
narrow temperature range. Generally, these reactions fluid-metal systems as they strive to develop a
are relatively non-energetic and temperature control is rigorous theoretical approach to predict the
easily achieved. Like other heterogeneous reactions, thermodynamic behavior of fluid metals over
mass transfer is usually the most important design the entire liquid-vapor range.
criteria. Physical Chemistry: Science and Engineering
John Prausnitz and Leo Brewer, Editors
4. The largest number of different chemical reactions 6 halftones. 99 line drawings.
(but not the largest quantity of material) are run in Cloth $69.50 ISBN 0-691-05830-X Due une
batch reactors, which are especially common in the
pharmaceutical, biotech, polymer, and cosmetics PrincetOn University Press
industries. Sizes vary from a few liters to over AT FINE BOOKSTORES OR CALL 800-777-4726 HTTP: //PUP.PRINCETON.EDU
200,000 liters.
5. CSTRs are the next most common reactors, followed
by PFRs and then by hybrid reactor types (fluidized Lll letter to the editor
beds, transport beds, trickle beds).
6. Continuous catalytic reactors are common in the Dear Editor:
petrochemical industries and, by far, the largest In the recent paper titled "Permeation fo Gases in
quantities of materials are produced in these types of Asymmetric Ceramic Membranes" [CEE, 33(1),
reactors. p. 58 (1999)], by C. Finol and J. Coronas, there

ACKNOWLEDGMENTS was an error in the equation used to calculate the
Knudsen number. The correct equation for this
We gratefully acknowledge valuable comments and dis- c e f
cussions with Dr. Ed Wolfrum of the National Renewable
Energy Laboratory, Professor H. Scott Fogler of the Univer- X 16 I- RT
sity of Michigan, and Jonathan N. Webb of the University of Kn = = tR
where all the parameters employed in the equation
REFERENCES were already defined in the mentioned article. We
1. Levenspiel, O., Chemical Reaction Engineering, 2nd ed., apologize for any trouble that this mistake may
John Wiley and Sons, New York, NY (1972) have caused.
2. Hill, Jr., C.G., An Introduction to Chemical Engineering
Kinetics and Reactor Design, John Wiley and Sons, New Thank you for your consideration.
York, NY (1977)
3. Smith, J.M., Chemical Engineering Kinetics, 3rd ed., C. Finol
McGraw-Hill, New York, NY (1981) J. Coronas
4. Fogler, H.S., Elements of Chemical Reaction Engineering, University of Zaragoza
2nd ed., Prentice Hall, New Jersey (1992)
5. Schmidt, L.D., The Engineering of Chemical Reactions, Ox- 50009 Zaragoza, Spain
ford University Press (1998)
6. Felder, R.M., and R. Brent, Chem. Eng. Ed., 31, 178 (1997)

Spring 1999 14

pf classroom

A New Approach To


University of Pennsylvania Philadelphia, PA 19104-6393

Turbulent flow, although the most important regime in
practice, is often given short shrift in textbooks on
fluid mechanics, and thereby in the classroom, at
both the undergraduate and graduate levels, because its theo-
retical structure is less developed than that for potential,
creeping, and laminar flows. Furthermore, recent analyses
have shown that much of that limited theoretical structure is
unsound, and in addition recent computational and experi-
mental advances have identified significant errors in the
predictions of the generally accepted correlating equations
in current textbooks and handbooks.
The objective of this article is to provide a supplement for
the teacher that summarizes some of the more important
recent work in turbulent flow and describes the present state
of the art in a form suitable for the classroom. In the interests
of brevity, the actual derivations are limited to fully devel-
oped flow in a round tube, but the methodologies are directly
applicable for parallel-plate channels and circular annuli and
are readily adapted for unconfined flow along a flat plate.
The details of some of the derivations are purposely omitted
in order to provide relevant and constructive material for
homework assignments.
In recognition of the highly variable allotments of time for
this subject and the different inclinations of individual teach-
ers, a discussion of each of the following topics is presented
separately after the main development:
The implicit as well as the explicit idealizations and
postulates that are inherent in the new "exact"

Stuart W. Churchill is the Carl V.S. Patterson
Professor Emeritus at the University of Pennsyl-
vania, where he has been since 1967. His BSE
degrees (in ChE and Math), MSE, and PhD were
all obtained at the University of Michigan, where
he also taught from 1950-1967 Since his formal
retirement in 1990, he has continued to teach
and carry out research on heat transfer and com-
bustion. He is also currently completing a book
on turbulent flow.
Copyright ChE Division of ASEE 1999

The sources and reliability of the individual terms,
coefficients, and exponents in the several new cor-
relating equations
The concepts and expressions in current textbooks
and handbooks that are to be eliminated or at least
identified as false or obsolete

The advances in the description of turbulent flow under
consideration herein are:
1. The proposal by Churchill and Chan'" to use the
dimensionless turbulent shear stress rather than the
eddy viscosity or the mixing length as a correlating
2. The precise determination of the turbulent shear stress
near the walls of a parallel-plate channel by Kim, et
al.,121 Lyons, et al.,t1 Rutledge and Sleicher,"' and
others by means ofDNS (direct numerical simula-
3. The development by Churchill and Chanisl of a
generalized correlating equation for the turbulent
shear stress for the complete cross-section of a
parallel-plate channel or a round tube.
4. The computation by Churchill and Chan1561 of im-
proved numerical values for the velocity distribution
and the friction factor using this correlation, and their
development of generalized correlating equations for
these values.
5. The experimental determination by Zagarola'71 of
improved and extended values for the velocity distribu-
tion and the pressure gradient in a round tube.
6. The slight but significant modification by Churchilll81
of the correlating equations of Churchill and Chan15'6'
for the turbulent shear stress, the velocity distribution,
and the friction factorfor a round tube on the basis of
these new experimental values.

Chemical Engineering Education

Development of a
New Integral Structure for Turbulent Flow

Time-averaging the differential equations for the con
vation of momentum in polar cylindrical coordinates f
Newtonian fluid with invariant physical properties (t(
found, for example, in Bird, et al.,[91) and then special;
that result for steady, fully developed flow in a round 1
leads to

aP I d (-p ru, PU 0

r dr r
rldr^ri^ -o- '

aP Id ( du '
--+-- ir- pruruz =0 (3)
az r dr dr
where here P is the dynamic pressure that arises from changes
in velocity only. Analytical integration of these three expres-
sions with respect to r results in
P P Urr p j ue --Urur -(4)

urue = o (5)
r i P du (6)
---) Z-r + p Urz (6)

Equation 4 provides an expression for the radial variation of
pressure, and Eq. 5 indicates that the Coriolis force is zero
at all radii. Equation 6 serves as the starting point for all
subsequent derivations. (The details of the derivations of
Eqs. 1 through 6 as well as the calculation of the pressure
drop across the tube (using the experimental data of Laufer1o01)
are suggested as homework.)
For convenience in subsequent manipulations, Eq. 6 may
be re-expressed as follows in terms of y=a-r, u=uz, v=Uy=-ur,
and r, [- (y / a)]}= = (r / 2)[-(dP / dz)]:

sionless form for simplicity and convenience:

y -, ] du+
a)L -'v dy+ (8)

where here y+ = y(wp)12 /u, a+ = a(Twp)l /,
u+ = u(p / ) /2, and (u'v')++ = -pu'v' / t. The first three of
these four dimensionless quantities were defined by Prandtl11"
in 1926, while the fourth, which has physical significance as
the fraction of the local shear stress due to turbulence, was
first defined by Churchill" in 1997.

(1) Equation 8 may be integrated formally to obtain

Y [ ,- \,++]f y+ ] +
u J j- (u'v' I dy

0a 2

a+ a- +2+ a

where R=r/a=l-(y+/a+). The form of Eq. 11 reveals that the
contribution of the turbulent fluctuations to the velocity dis-
tribution may be expressed as a simple deduction from the
expression for purely laminar flow at the same value of a+.
The initial form of Eq. 9 in terms of y*, however, proves to
be more convenient for numerical calculations.
The integration of a' from Eq. 10 over the cross-section to
obtain the mixed-mean value may be expressed as

u = u'dR2 =- 1[ ( )++R2dR2

which may be integrated by parts to obtain

U+ = -(v 4 a

. f (u'v')+dR4 (13)

Sy) du
1-- = puv
P a) dy

Suggested Exercises: Show that
Y) -T ap
-a) -2 az
and prove that
aP dP
az dz

Equation 7 may in turn be expressed in the following dimen-
Spring 1999

(7) The right-most form of Eq. 13 reveals that the contribution
of the turbulent fluctuations to the mixed-mean velocity is
also a deduction from the expression for purely laminar flow
at the same value of a+. The formulation in terms of y',
namely Eq. 14, again proves to be the most convenient for
numerical calculations.

(The detailed derivation of Eq. 13 from Eq. 10 is
suggested as an exercise.)

Since the Fanning friction factor is defined by


= [- 1 a-) dy+

f 2 2 (15)
pum ()2
Eqs. 13 and 14 may be interpreted as expressions for the
evaluation of (2/f)"2 and hence of f.
Churchilll8 recently proposed the following generalized
correlating equation for (uv') for use in Eqs. 9 and 14 for
a' > 300:

u'v,) =+

S1 y (1+ 6.95 y+ -
S0.7 I + exp- 8 (16
S10 1 0 .436y 0.436a+ a+

The construction of Eq. 16 is discussed subsequently.
Values of u+ calculated from Eq. 9 using Eq. 16 for (u 1v)+
have been correlated by Churchill8s' with the expression

u- =[(u)-3 + )-3]' (17)

u + -{ ( Y)2
47(y 140

u 1 [ + 1448y 1 y2 3
u = I1n + .48y+ +6.824- -5.314 (19)
0.436 1+ 0.301(e a+ a a

and those for u+ using Eqs. 14 and 16 by

S(22 3301612 47.62 1 a
um,= =3.30- + + fn (20)
a+ a + 0.436 1+0.301 e )a+

The term [l+0.301(e/a)a'] was arbitrarily incorporated in
Eqs. 19 and 20 to extend their coverage to tubes with natural
roughness e. Values of e for various types of commercial
piping are tabulated in all handbooks and most textbooks on
fluid flow. Of course, Eqs. 16 and 18 are inapplicable for
rough pipe for values of y' of the order of magnitude or less
of e+. The construction of Eqs. 17 through 20 is also dis-
cussed subsequently.
The predictions of (uv)++ by the near-equivalent of Eq.
16 are compared with the computed values of Kim, et al.,121
Rutledge and Sleicher,131 and the experimental values of
Eckelmann1'31 for small values of a* in Figure 1 and with the
experimental values of Wei and Willmarth[141 for moderate

values of a+ in Figure 2. The agreement appears to be within
the scatter of the individual values. The waviness of the
curves representing Eq. 16 in Figure 1 is an artifact of that
expression, not an error in graphing.
Homework exercise: Can such apparently anoma-
lous behavior be rejected on physical grounds?
The slight uncertainty in (uv')+ as predicted by Eq. 16 is
reduced in u' and um by the integration by means of which
the latter are evaluated numerically.
Homework exercise: Explain the grounds, if any,
for this assertion.
The predictions of u' by Eqs. 17 through 19 differ no more
than 0.3%, and those of u+ by Eq. 20 no more than 0.1%
from the very precise experimental values of Zagarola (see
Churchillt8'). The use of Eqs. 17 through 19 and Eq. 20 to
predict u and um respectively is more convenient than the
use of Eqs. 9 and 11 by virtue of the avoidance of numerical
integration and has the advantage of including an expression
for the effects of roughness. However, slightly greater accu-
racy for smooth tubes is to be expected from Eqs. 9 and 11
with 16.
Homework exercise: Why?
This completes the direct presentation of the new structure
and correlating equations for fully developed turbulent flow

1 (i7tv)++

Figure 1. Representation of experimental and numerically
computed values of the dimensionless/turbulent shear stress
for small y' by Eq. 16 (from Churchillt12).
Chemical Engineering Education

in a round tube. It is important insofar as time allows, how-
ever, to discuss with the students the postulates and idealiza-
tions inherent in the above structure, to examine the sources
and reliability of the terms, coefficients, and exponents of
the correlating equations, and to describe the obsolete con-
cepts and correlations that are still to be encountered in the
literature of the past and perhaps even some of that of the
future. The background for such discussions follows.

Equations 1 through 14 are exact insofar as steady, fully
developed turbulent flow is attainable, time-averaging is a
valid procedure, and the effects of viscous dissipation and
the variation of physical properties are negligible. The valid-
ity of these idealizations and postulates will now be briefly
Although the existence of a state of fully developed turbu-
lent flow has been questioned on theoretical grounds,'1" the
close approach to such an invariant condition at a moderate
distance from the inlet of the tube is well supported experi-
The propriety of time-averaging has also been questioned
by some on theoretical grounds, but no significant and prac-
tical discrepancy in the expressions obtained by this proce-
dure has been documented experimentally.
The fluctuations in pressure with time as well as the radial
and axial variations in the time-mean pressure result in pro-
portional variations in the density of a gas, but such effects
are ordinarily negligible on the mean. Viscous dissipation
generates radial and axial variations in the temperature of
the fluid and thereby corresponding changes in viscosity as
well as in density. Such effects have, however, been found to


Figure 2. Representation of experimental values of the
dimensionless turbulent shear stress for all y' by Eq. 16
(from Churchill and Chan'5').
Spring 1999

be completely negligible for ordinary fluids and conditions.
Although this discussion implies that Eqs. 1 through 14
may be considered to be essentially exact, it also properly
suggests to students that extreme conditions may be encoun-
tered for which they are not, for example, at very low pres-
sure (hard vacuums), at very high temperatures (ionized
gases), with very viscous fluids (polymers), and in fluids
being heated at the wall. These expressions are also inappli-
cable for developing flow. The quantitative evaluation of
such effects poses problems of higher order than would be
appropriate in most elementary courses in fluid mechanics.

Since students may be expected to be confronted at vari-
ous times in their career by alternatives to Eqs. 16-20, it is
appropriate that they understand their source and reliability
as a basis for making the correct choices and invoking ap-
propriate safety factors.
Equations 16 and 17 both have the canonical form pro-
posed by Churchill and Usagi,"71 namely an arbitrary power-
mean of asymptotic expressions for small and large values
of the independent variable. The asymptote for (u'v') as
y -> 0 was taken to be

(u'v') =0.0007(y)3 (21)
Equation 21 is based on the relationship
u'' oy3 (22)
first derived by Murphree"I1 by means of asymptote expan-
sions. The dedimensionalization of the turbulent shear stress
by t rather than r, in Eq. 21 is arbitrary, but is much more
convenient and has no subsequent adverse consequences.
Equation 21 has experimental support, but even more im-
pressive confirmation by the previously cited direct numeri-
cal simulations, which are also the primary source of the
coefficient of 0.0007. This value is nevertheless somewhat
uncertain and therefore subject to possible future improve-
ment. However, any resulting consequences with respect to
u* and um are certain to be small.
Integration of Eq. 7 using Eq. 21 for uv') results in the
corresponding asymptote for u* as y+ ->0, namely

+ ,4 (y"2 0.00014(y)5
u+ = y-0.000175(y+) (y2a+ 0.00014(y) (23)
S2a+ a+
For conditions such that the second term on the right-hand
side is significant, as well as for large a', the third and fourth
terms may be dropped. Equation 18 constitutes an arbitrary
approximation for this reduced form of Eq. 23 that avoids
the negative values of u* for large y' that would not be
acceptable in the combined form of Eq. 17.
The justification of these two simplifications may
be assigned as exercises.

The starting point for asymptotes for (u'v'+ and u+ for
y+ --a+ is

u+ =A+B n(y+)+C (y+2 +D(y+ (24)

The first two terms on the right-hand side of Eq. 24 were
first derived, as discussed subsequently, by Prandtl on the
basis of his mixing-length model for the "turbulent core near
the wall." The third and fourth terms were proposed by
Churchill191 to represent the "wake," the increased velocity
with respect to the first two terms near the centerline.
Churchill81 chose A=6.13, B=1/0.436, and C-D=1.51 to
fit the experimental measurements of Zagarolav71, and
D=-(B+2C)/3 to force du'/dy' to zero at y+=a0. This latter
choice also results in th e necessary asymptotic approach of
u -u+ E1- y/a+] as y+ a. The net result is

I _y+2 (-3
u = 6.13 + en(y+)+6824(Y 5.14 j) (25)
0.436 a+
Incorporating the constant 6.13 in the argument of the
logarithm, adding unity to that argument to avoid negative
values of u' as y 0, and dividing that result by
1 + 0.301(e / a)a+] to incorporate the effect of roughness, then
results in Eq. 19. The exponent of-3 in Eq. 17 was chosen to
represent experimental velocities for intermediate values of
y+. This representation is relatively insensitive to that value.
The term within the absolute values signs of Eq. 16 was
derived by differentiating Eq. 16 with respect to y+, substi-
tuting the result in Eq. 9, dividing through by [l-(y*/a*)], and
approximating [l-(1/0.436y+)] by exp(-1/0.436y+). The lat-
ter step and the absolute value sign are simply mathematical
devices to avoid negative values of (u') as y 0. The
arbitrary exponent n was evaluated on the basis of the ex-
perimental data of Wei and Willmarth'41 and others for
moderate and large values of a+ and intermediate values of
y+ as well as on the basis of velocity distributions derived
from Eq. 16.15
Equation 20 was constructed by integrating u+ from Eq. 19
over the cross-section and adding arbitrary terms in (a')
and (a+)-2 to account for the decreased velocity near the wall
where u+ = y. The coefficients -161.2 and 47.6 were evalu-
ated on the basis of values of u+ calculated from Eq. 11 with
(uv') from Eq. 16.

Suggested homework: Justify the presence of the
terms in (a')-' and (a')-2 by substituting u+ in Eq. 12
and integrating up to y+ 12.

The term representing the effect of roughness in Eqs. 19
and 20 is based on the experimental data and a correlating
equation of Colebrook.120' Also, see Churchill1211 for a ratio-
nalization for this term.
A further discussion of some of the terms and coefficients
of Eqs. 16 though 19 is provided in the following section.

The methodology described above is applicable to all one-
dimensional fully developed turbulent flows, but most alter-
native geometries and conditions introduce complexities not
encountered in a round tube.
According to the analogy of MacLeod,221 expressions for
the velocity distribution in fully developed flow, either lami-
nar or turbulent, when expressed in terms of u', y* and a', are
directly applicable for flow between identical parallel plates
if a' is simply replaced by b+, the half-spacing of the chan-

Exercise: What is the analog of R?

It follows that the various expressions for (u'v') are also
directly applicable with this substitution, but those for um
are not, owing to the different area of integration.
Homework exercise: Derive the analogs of Eqs.
12 through 14 and 20.

The analogy of MacLeod appears to be beautifully con-
firmed by the plot of the centerline and central-plane veloci-
ties in Figure 3 from Whan and Rothfus.[231 Although this
remarkable relationship may be shown to be exact for lami-
nar flow, it must be considered speculative and possibly only
a very good approximation for turbulent flow.
Exercise: Justify the MacLeod analogy for lami-
nar flow.

This analogy has already been applied implicitly in the de-
termination of the coefficient of Eq. 21 from DNS for paral-
lel plates and in the comparisons of the predictions of Eq. 16
with experimental data in Figures 1 and 2.
Integral formulations for the velocity distribution in terms
of the equivalent of (uv') are possible for planar induced
(Couette) flow between parallel plates, unconfined forced

a* or b*

Figure 3. Confirmation of analogy of MacLeod for
centerline velocity.1361
Chemical Engineering Education

flow along a flat plate (even though it is developing and only
quasi one-dimensional), forced flow between parallel plates
of unequal surface roughness, forced flow through a circular
annulus, longitudinal induced (Couette) flow in a circular
annulus, rotary induced (Couette) flow in a circular annulus,
and combined forced and induced flow between parallel
plates and through a circular annulus. Generalized correlat-
ing equations equivalent to Eq. 16 have yet to be constructed
for any of these flows. Also, except for the first one of these
flows, the total shear stress distribution within the fluid is
not known a priori and must be determined by iterative
numerical calculations. This concept fails completely for all
two-dimensional channels because of the existence of a sec-
ondary motion.

Boussinesq'241 in 1877 proposed perhaps the first quantita-
tive model for a turbulent shear flow, namely

r =(+ t) du (26)
The eddy viscosity defined by Eq. 26 has been determined,
usually with considerable uncertainty, by differentiating ex-
perimental velocity distributions. Generalized correlations
of such values have in turn frequently been used to predict
velocity distributions and rates of heat and mass transfer. At
the same time, this concept has often been scorned by both
analysts and practitioners because of its lack of a mechanis-
tic rationale. But re-expressing Eq. 26 in terms of u' and y',
substituting tw[1 (y+/ a)] for t and then eliminating du'/
dy+ between the resulting expression and Eq. 8, reveals that

Pt (t--(UTV)+ (27)
I- ( -1

Equation 27 is a remarkable relationship in that it indicates
that the eddy viscosity ratio it / p in a round tube may be
interpreted physically as simply the ratio of the rate of trans-
port of momentum by the turbulent fluctuations to that by
viscous shear, and is thereby independent of its heuristic
diffusional origin. (Boussinesq was either very intuitive or
very lucky.) In view of the one-to-one relationship between
p, / p and (u'v') all of the above exact relationships in
terms of the latter may be re-expressed in terms of pt / p
with no loss of generality, albeit with some loss of simplic-
ity. Unfortunately, the validity of the eddy viscosity concept
is limited to forced flow in round tubes, forced or induced
flow between parallel plates of equal roughness, and in
confined flow along a flat plate. For all other geometries and
conditions, the eddy viscosity is unbounded at some point
within the fluid and is negative over a finite adjacent region
even though (u'v')+ remains well-behaved. This anomaly
occurs because the velocity gradient becomes zero and then
Spring 1999

changes sign at a different location than does the total shear
stress. (See Churchill and Chan111 for a fuller explanation.)
Exercise: Derive an expression for (u'v') at the
centerline of a round tube.

Prandtl1251 in 1925 proposed an alternative model for the
turbulent shear stress on the basis of a questionable analogy
with the kinetic theory of gases, namely
T = P-+p(2 du (28)
dy .dy)
where c is a mixing-length. Because of its mechanistic basis,
and thereby the possibility of predicting the value of C, the
mixing-length model has generally been accorded greater
prestige among analysts than the eddy-viscosity model. But
it is actually inferior in every respect. Re-expressing Eq. 28
in dimensionless form and eliminating du'/dy+ between the
resulting expression and Eq. 8 leads to

t+ = ( (29)
a-+)[I-Huv) J

Equation 29 reveals that the mixing length as well as the
eddy viscosity is independent of its heuristic origin and
bears a direct relationship to (u'v) but it also reveals that
the mixing length is unbounded at the centerline of round
tubes and at the central plane of a parallel-plate channel
where j, / g and (u'v') are both finite. In addition, the
mixing length is unbounded and negative within the fluid at
the same locations in other geometries as the eddy viscosity.
How has this false concept survived and found repeated
application in the literature of fluid mechanics and heat and
mass transfer for over seventy years? Apparently as a result
of insufficiently precise experimental data for u'v' and u,
uncritical or biased processing of these data (see, for ex-
ample, Lynn'261 and Churchill1271), and the acceptance of mod-
erate inaccuracy in the resulting predictions of u', f, and Nu
by practitioners.
Most other models for the prediction of turbulent transport
are similarly unsound. For example, the K e models, which
function by predicting the eddy viscosity or the mixing length,
necessarily share their failures and in addition incorporate
considerable empiricism. The LES (large eddy simulation)
models and the turbulent shear stress models have promise,
but their current implementations necessarily incorporate
empirical terms for the region near the wall that may in turn
introduce inaccuracy or shortcomings similar to those of the
eddy viscosity model. The DNS models alone appear to be
free of these sources of error and/or failure, but in their
current state of development, they are limited in numerical
application to the lowest range of fully developed turbulent
flow in very simple geometries. Some aspects of this assess-
ment may be expected to become dated due to new develop-

ments in modeling and computation.
Finally, the origins, credentials, and shortcomings of sev-
eral of the empirical correlations in the literature should be
mentioned. The most commonly used correlating equation
for the velocity distribution in smooth tubes consists of the
first two terms on the right-hand side of Eq. 24, namely

u+ = A + Beny} (30)

This expression is often cited as a major contribution of
mixing-length theory in that it was first derived in 1931 by
Prandtl (see Nikuradsel281) on the basis of the postulate of
Von Kirmin1291 that
= y/B (31)
and the idealizations that y/a is negligibly small relative to
unity, and that the viscous stress is negligibly small relative
to the contribution due to turbulence. But Eq. 30 was subse-
quently derived by Millikan[30] on the basis of dimensionless
considerations alone. He merely postulated a region of over-
lap between the "law of the wall," u' = f{ y'}, which follows
from the postulate of a negligible dependence on a', and the
"law of the center," which follows from the postulate of a
negligible dependence of uc u' on the viscosity.

Exercise: Carry out the details of this derivation.

Nikuradse determined values of A=5.5 and B=2.5 from his
experimental measurements. The resulting expression fails
for y+ < 30 and for y' >' as would be expected from its
derivation. Equation 30 may be considered to be displaced
by Eq. 17 (with Eqs. 18 and 19), which applies for all y' and
is presumably more accurate even for intermediate values of
y* for large a' because of the improved accuracy of the data
of Zagarola[71 relative to that of Nikuradse.1281 Equations 16,
19, and 20 are inaccurate for a' < 300 because the exponen-
tial and logarithmic terms that arise from Eq. 30 are inappli-
cable due to the coincidence of its indicated upper and lower
bounds at that value.
Integration of Eq. 30 over the cross-section gives

3 (32)
uT =A -B + Bna+ (32)
2 L
This expression with separate empirical coefficients rather
than 1.75=5.5-(3/2)(2.5) and 2.5 has been widely used for
correlation of experimental values of (2/f)/2, and together
with the correlating equation of Colebrook1201 for rough pipe,
is the source of the various plots in the literature for the
friction factor. These plots may all now be considered to be
displaced by Eq. 20 in terms of both convenience and accu-
racy. A whole literature exists concerning the approximation
of expressions such as Eqs. 30 and 32 by one explicit in Re
rather than a+ = Re(f/2)1/2, but iterative solution of the origi-
nal expressions for a specified value of Re is a trivial task
even with a hand-held calculator.

One other approach to correlation is worth mentioning
because of a recent attempt at resuscitation. Blasius1311 in
1913 correlated the then-available data for the friction fac-
tor, which only extended to Re=105, with the purely empiri-
cal expression
f= 14Re (33)

Prandtll321 recognized that Eq. 33 implies

u = 8.562(y+) (34)

Nikuradse[331 found that Eq. 34 closely agreed with his own
experimental data for smooth pipe and Re < 105, but that
different coefficients and exponents were required for
Re > 105 and roughened pipe. Nunner[341 subsequently found
that such values could be represented by

u+ = p(y) (35)


(l+ )(1+2oa)u+ (l+fl/2( 1+2fl /2)(2 1/2
= 2(a+/a (2f2 f (37)

Barenblatt1351 has very recently proposed alternative
semitheoretical expressions for a and 3, but his representa-
tion is inferior numerically to that of Nunner for smooth pipe
and does not encompass roughened pipe. In any event, Eq.
35 fails totally near the wall where it predicts a negative
unbounded velocity gradient and near the centerline where it
predicts a finite velocity gradient.
Bird, et al.,19' p.1751 noted that expressions for the turbulent
shear stress as a function of distance from the wall, such as
Eq. 16, lead to simpler integration for the velocity distribu-
tion and the mixed-mean velocity than do expressions in
terms of the velocity gradient, such as those involving the
eddy viscosity or the mixing length. But in applying this
concept, they used an erroneous boundary condition and a
purely empirical expression of Pai1361 that does not conform to
the known asymptotic behavior and thereby leads to final
expressions that are in error both functionally and numerically.

Simple but exact integral expressions are formulated herein
for the velocity distribution and the mixed-mean velocity,
and thereby the friction factor for fully developed flow in a
smooth round tube in terms of a particular dimensionless
turbulent shear stress, namely the local fraction of the shear
stress due to turbulence. The proposed correlating equation
for this latter quantity is not exact, but it has a theoretically
based structure and the small error in its predictions is re-
Chemical Engineering Education

duced by the numerical integration used to evaluate the
local and mixed-mean velocity. The new, theoretically based
correlating equations presented herein for these latter two
quantities are presumed, on the basis of comparisons with
recent, improved experimental data, to be more exact than any
other expressions or graphs in the literature. These formula-
tions are readily extended to other one-dimensional flows.
The eddy viscosity, mixing-length, and K E models are
shown to be inapplicable or inferior in every respect to the
direct use of correlating equations for the turbulent shear
stress itself. Nonetheless, they may produce numerical re-
sults of fair accuracy even in applications where they are not
strictly valid. The utility of the shear-stress and LES models
is currently handicapped by the necessity of incorporating
heuristic differential terms for the region near the surface.
The DNS models are free of this shortcoming, but their use
is limited to simple geometries and a narrow range of flow
just above the minimum for fully developed turbulence. All of
these models share the inapplicability of the expressions herein
for two-dimensional flows such as that in a square duct or an
open channel due to the ubiquitous secondary motion.
The material presented herein can be used by teachers of
both undergraduate and graduate classes in fluid mechanics
and transport as a supplement or replacement for the obso-
lete section on turbulent flow in all current books.

1. Churchill, S.W., and C. Chan, "Turbulent Flow in Channels in
Terms of Local Turbulent Shear and Normal Stresses," AIChE
J., 41,2513(1995)
2. Kim, J., P. Man, and R. Moser, "Turbulence Statistics in Fully
Developed Channel Flow at Low Reynolds Numbers," J. Fluid
Mech., 177, 133 (1987)
3. Lyons, S.L., T.J. Hanratty, and J.B. McLaughlin, "Large Scale
Computer Simulation of Fully Developed Channel Flow with
Heat Transfer," Int. J. Num. Methods Fluids, 13, 999 (1991)
4. Rutledge, J., and C.A. Sleicher, "Direct Simulation of Turbulent
Flow and Heat Transfer in a Channel: I. Smooth Walls," Int. J.
Num. Methods Fluids, 16, 1051 (1993)
5. Churchill, S.W., and C. Chan, "Theoretically Based Correlating
Equations for the Local Characteristics of Fully Turbulent Flow
in Round Tubes and Between Parallel Plates," Ind. Eng. Chem.
Res., 34, 1332 (1995)
6. Churchill, S.W., and C. Chan, "Improved Correlating Equations
for the Friction Factor for Fully Turbulent Flow in Round Tubes
and Between Identical Parallel Plates, Both Smooth and Natu-
rally Rough," Ind. Eng. Chem. Res., 33, 2016 (1994)
7. Zagarola, M.V., "Mean-Flow Scaling of Turbulent Pipe Flow,"
PhD Thesis, Princeton University, Princeton, NJ (1996)
8. Churchill, S.W., "An Appraisal of Experimental Data and Predic-
tive Equations for Fully Developed Turbulent Flow in a Round
Tube," in review (1998)
9. Bird, R.B., W. Stewart, and E.N. Lightfoot, Transport Phenom-
ena, John Wiley & Sons, p. 85 (1960)
10. Laufer, J., The Structure of Turbulence in Fully Developed Pipe
Flow, Nat. Adv. Comm. Aeronaut., Report 1174, Washington, DC
11. Prandtl, L., "Uber die ausgebildete Turbulenz," Verhdl. 2nd In-
tern. Techn. Mech., Zurich, p. 62 (1926)

Spring 1999

12. Churchill, S.W., "New Simplified Models and Formulations for
Turbulent Flow and Convection," AIChE J., 43, 1125 (1997)
13. Eckelmann, H. "The Structure of the Viscous Sublayer and Adja-
cent Wall Region in Turbulent Channel Flow," J. Fluid Mech.,
65, 439 (1974)
14. Wei, T., and W.W. Willmarth, "Reynolds-Number Effects on the
Structure of a Turbulent Channel Flow," J. Fluid Mech., 204, 57
15. Barenblatt, G.I., and N. Goldenfeld, "Does Fully Developed Tur-
bulence Exist? Reynolds Number Dependence versus Asymptotic
Convariance," Phys. Fluids, 7, 3078 (1995)
16. Abbrecht, P.H., and S.W. Churchill, "The Thermal Entrance Re-
gion in Fully Developed Turbulent Flow," AIChE J., 6, 268 (1960)
17. Churchill, S.W., and R. Usagi, "A General Expression for the
Correlation of Rates of Transfer and Other Phenomena," AIChE
J., 18, 1121 (1972)
18. Murphree, E.V., "Relation Between Heat Transfer and Fluid
Friction," Ind. Eng. Chem., 24, 726 (1932)
19. Churchill, S.W., Turbulent Flows. The Practical Use of Theory,
Notes, The University of Pennsylvania, Chapter 5 (1994)
20. Colebrook, C.F., "Turbulent Flow in Pipes with Particular Refer-
ence to the Transition Region Between Smooth and Rough Pipe
Laws," J. Inst. Civ. Eng., 81, 133 (1938-39)
21. Churchill, S.W., "Empirical Expressions for the Shear Stress in
Turbulent Flow in Commercial Pipe," AIChE J., 19, 325 (1973)
22. MacLeod, A.L. "Liquid Turbulence in a Gas-Liquid Absorption
System," PhD Thesis, Carnegie Institute of Technology, Pitts-
burgh, PA (1951)
23. Whan, G.A., and R.R. Rothfus, "Characteristics of Transition
Flow Between Parallel Plates," AIChE J., 5, 204 (1959)
24. Boussinesq, J., "Essai sur la Theorie des Eaux Courantes," Mem.
Pre divers savants Acad. Sci. Inst. Fr., 23, 1 (1877)
25. Prandtl, L., "Bericht uiber Untersuchungen zur ausgebildeten
Turbulenz," Z. angew. Math. Mech., 5, 136 (1925); Engl. Trans.
"Report on Investigation of Developed Turbulence," Nat. Adv.
Comm. Aeronaut., TM123; Washington, DC (1949)
26. Lynn, S. "Centerline Value of the Eddy Viscosity," AIChE J., 5,
27. Churchill, S.W., "A Critical Examination of Turbulent Flow and
Heat Transfer in Circular Annuli," Thermal Sci. Eng., 5, 1 (1997)
28. Nikuradse, J., "Gesetzmassigkeiten der Turbulenten Stromung
in glatten Rohren," Ver.Deutsch. Ing. Forschungsheft, 356 (1932):
English Translk "Laws of Flow in Smooth Tubes," Nat. Adv.
Comm. Aeronaut., TM62, Washington, DC (1950)
29. von Karman, Th., "Mechanische Ahnlichten und Turbulenz," Proc.
Third Intern. Congr. Appl. Mech., Stockholm, Part I, 85 (1930)
30. Millikan, C.B., "A Critical Discussion of Turbulent Flows in
Channels and Circular Tubes," Proc. Fifth Intern. Congr. Appl.
Mech., Cambridge, MA, p. 386 (1938)
31. Blasius, H., "Das Ahnlichkeitsgesetz bei Reibungsvorglngen in
Fliissigkeiten," Ver. Deutsch. Ing. Forsch.-Arb. Ing.-Wes., 131,
Berlin (1913)
32. Prandtl, L., "Neuere Ergebnisse der Turbulenzforschung," Z.
Ver. Deutsch. Ing., 77, 105 (1933); English Trans: "Recent Re-
sults of Turbulence Research," Nat Adv. Comm. Aeronaut., TM720,
Washington, DC (1933)
33. Nikuradse, J., "Widerstandgesetz und Geschwindigkeitsverteilung
von Turbulenten Wasserstr6mungen in Glatten und rauhen
Rohren," Proc. Third Intern. Congr. Appl. Mech., Stockholm,
Part 1, 239 (1930)
34. Nunner, W., "Wirmetibertragung und Druckabfall in rauhen
Rohren," Ver. Deutsch. Ing., Forschungsheft, 455 (1956)
35. Barenblatt, G.I., "Scaling Laws for Fully Developed Turbulent
Shear Flows. Part 1. Basic Hypotheses and Analysis," J. Fluid
Mech., 248, 513 (1993)
36. Pai, S.I., "On the Turbulent Flow in a Circular Pipe," J. Franklin
Inst., 256, 337 (1953) O

ME, classroom




University of Canterbury Christchurch, New Zealand

Problem solving is an activity that engineers are en-
gaged in every day, but not many professional engi-
neers are taught problem-solving strategies as part of
their undergraduate education. In the literature, some writ-
ers'1-3] have been particularly concerned with problems based
on mathematics and logic, whereas others14-8] have been more
concerned with open-ended problems and lateral thinking.
Some research on problem solving has been based on
computer-aided modeling of mental processes known as ar-
tificial intelligence (AI).19 Other work has been based on
problem-solving experiments with human subjects"12 and
studies of thinking."3 Much of the research, however, has
been concerned with logical confusion and errors; little re-
search to date has been done on the teaching of problem-
solving skills in higher education.
The teaching of problem-solving skills to undergraduate
engineering students has been described.[4-'71 From an in-
dustrial point of view,"81 problem solving is a crucial skill
for engineers in manufacturing. The time to teach this
skill is critical in the professional development of an
engineer and may be best immediately following gradua-
tion when the engineer is confronted with what seems to
be an unsolvable problem.
Our view is that there is a balance between teaching prob-
lem-solving skills early in the undergraduate degree course
so the skills can be used in the educational process, and
teaching these skills later when students have the maturity to
appreciate their benefits and the experience to apply the
techniques. We have chosen to present problem-solving tech-
niques in the third year of a four-year degree program.

In traditional design teaching, the design process is often
broken into a number of incremental steps: defining the task,

* Address: Instrumentation and Control Engineering, Murdoch
University, Western Australia 6168

goal setting, establishing a concept, defining the constraints,
setting the specifications, listing the alternatives, evaluating
and selecting the best alternatives, formulating an appropri-
ate mathematical model, calculating, modifying, costing,
drawing, constructing, testing, and finally, commissioning.
A general structure of the design process, once recognized
and defined, is then adapted as a design strategy for future
projects. While the approach of retrospectively studying suc-
cessful design projects, recognizing the various areas of
activities and their logical sequence, and applying it to
new projects works well, it relies heavily on experience
for a successful outcome. It is exactly this experience in
application, however, that our undergraduate students
lack. A practical design strategy is more appropriate for
novice engineers.
Design has been taught to our chemical engineering un-
dergraduates for many years by a traditional case-studies
approach that involved dissecting the design process into its
various elements, imparting relevant knowledge by formal

Judith Mackenzie is Senior Tutor in the School of Engineering, Univer-
sity of Canterbury. She has a Master's degree in Education and has
recently completed a PhD in Chemical and Process Engineering. Her
teaching and research focuses on the application of computers as a tool
for innovative teaching in chemical engineering education.
Maurice Allen is Associate Professor in the Department of Instrumen-
tation and Control Engineering at Murdoch University. His teaching and
research centers on process control, the modeling and simulation of
industrial processes, and the application of computing to process engi-
neering and teaching.
Brian Earl is Associate Professor of Chemical and Process Engineer-
ing, University of Canterbury. His teaching interests are in process
design, corrosion, and electrochemical engineering, and his research
interests are in expert systems in process design, chemical engineering
education, corrosion, and alternative transport fuels. Dr. Earl is cur-
rently on leave at Cornell University
Ian Gilmour is Senior Lecturer in the Department of Chemical and
Process Engineering, University of Canterbury. His teaching interests
are the design of process equipment, heat exchangers, combustion of
fuels and energy, and material balances for chemical processes. His
research interests are the efficient uses of fuels and energy in the
process industry, extraction of nutraceuticals from agricultural and for-
est residuals, and computer modeling for the pulp and paper industry.

Copyright ChE Division of ASEE 1999
Chemical Engineering Education

lectures, and demonstrating how experienced engineers have
designed successful systems. It was hoped that this approach
would imbue the students with sufficient knowledge and
skills to become novice designers.
These efforts to teach design led to the realization that
competence in design seemed to be achieved by a handful of
students who rose to the chal-
lenge and were able to apply ... there is a balance b
skills, knowledge, and other solving skills early in I
personal attributes, often with course so the skil
outstanding results. The educationalprocei
recipe for success seemed to skills later when stud
combine such ingredients as appreciate
organization, lateral thinking,
computation, practical experience in workshop skills, and
an ability to think in abstract terms. These special tal-
ents, which every student possesses, need to be devel-
oped and honed to a sharper edge.
In recent years, we have adopted a problem-solving ap-
proach similar to that of Woods119] for teaching third-year
engineering design. This design course embraces a wide
spectrum of engineering topics: mixing and pumping of
liquids, flowsheeting, column design, pinch technology, pro-
cess reliability, separation processes, and properties of engi-
neering materials. A problem-solving foundation to engi-
neering design provides students with the necessary skills
and confidence to be able to tackle any problem, design or
otherwise, without feeling hindered by lack of direct experi-
ence in the particular topic.

Students learn best when they directly experiment with
subject matter and are actively involved with the material.2"1
Interactive computer instruction provides such active learn-
ing and allows students to "review and demonstrate mastery
of the material at his/her own pace, [and] provides them with
immediate feedback to their responses."121
Interactive computer-based learning depends on software
that is easy to use, maintains a focus on the concepts, has
minimal tediousness, promotes learning, and gives individual
guidance. Strategies for Creative Problem Solving"5 is a
collection of interactive computer modules and is used to
supplement problem-solving lectures. Additional features in
some modules, such as the use of graphical animation and
entertaining motivators, were included to increase student
interest in, and motivation for, the module content.1221
The content of each module is
1. An Introduction to Problem Solving
This module provides the user with the motivation to use
creative problem-solving strategies. Topics include the
characteristics of effective problem solvers, fear of fail-
ure, the need for risk-taking, paradigm shifts, having a
vision, a problem-solving heuristic, creative thinking, and
Spring 1999



working in teams. The introduction presents the topics as
well as their application to a contamination problem in a
municipal water-supply system.
2. Problem Statement Definition Techniques
The goal of this module is to help the user properly define
the problem. Several techniques are used to better define
the problem statement: for
'tween teaching problem- example, the Dunker Dia-
e undergraduate degree gram,"19 the McMaster Five-
can be used in the Point Strategy,"91 the
, and teaching these Present-State Desired-State
nts have the maturity to technique, 12 and the State-
heir benefits. ment-Restatement technique.
The user reviews the meth-
ods of problem definition in two examples: problems at a
flashlight manufacturing plant are analyzed with the
McMaster Five-Point Strategy and a second example
involves a grocery store freezer door fogging up and
blocking the customer's view of the contents.
3. Brainstorming: Methods of Solution Generation
This module helps the user generate original yet appli-
cable solutions to a specific problem through brainstorm-
ing. The review section introduces the basic techniques
and ideas for improvement, including Osborn's check-
list,1241 random-word stimulation, futuring, conceptual
blockbusting, and using other people's views.
These methods are illustrated through specific examples.
To test the techniques, the user is first asked to brain-
storm a list of synonyms for the word "money." Once the
user is finished, the user's list is compared to one gener-
ated by a group of college students. Second, the user
selects at least two brainstorming topics chosen from a
list of five possible scenarios, ranging from encouraging
recycling in a community to preventing zebra mussel
infestations on power-plant water-intake pipes. For each
scenario, a detailed problem statement is given as well as
a few example solutions to get the user started.
4. Potential-Problem Analysis: Avoiding
Future Problems
Potential problems should be anticipated and analyzed
before they happen. Three parts of potential problem
analysis (possible causes, preventative action, contingent
actions) are explained in the introduction. The user then
has a choice of scenarios (either a cross-country road
trip or preparation for an interview) that are used to
review the techniques. The main scenario is based on the
1993 world solar-car race, Sunrayce'93.251 The back-
ground of the race is presented with additional explana-
tion of relevant technology, including the solar-cell mecha-
nism and the importance of gear ratios in power-train
design. A potential-problem analysis chart for the event
is prepared by the user to determine problems that might
occur during a race and their prevention.

5. Planning: Implementation of Solutions
Gantt charts,261 critical-path analysis,[271 deployment charts, and
budget proposals are introduced as tools that aid planning. These
four techniques are illustrated in two introductory scenarios: plan-
ning the ergonomic design of an office and planning a student
conference. In the interactive section of the module, the user is
part of a team participating in a student competition to build a
one-tenth-scale model of a steel bridge. Each of the planning
techniques is then applied to generate a Gantt chart, a critical-
path chart, a deployment chart, and a budget for the project.
6. Evaluation: Solution Evaluation Techniques
The importance of continually re-evaluating a solution throughout
the course of a project is emphasized. The technique presented is
the evaluation checklist, illustrated by the near disaster of market-
ing the new Coca Cola. The example demonstrates the use of an
evaluation checklist to prevent millions of dollars from being
wasted. In the interactive scenario, the user is presented with the
problem of a paper mill that plans to expand its production capac-
ity. The user is given the opportunity to talk to other virtual
employees in the company and to gather the necessary information
to evaluate the proposed expansion. Findings are submitted to the
project supervisor for immediate feedback.

The problem-solving section of the design course consisted of nine
one-hour lectures, six hours of laboratory sessions (with about the same
amount of time devoted to working through set problems), and the
computer problem-solving modules. This allocation of time was just
sufficient to introduce the forty third-year students to the basic concepts,
give them the experience of applying these new skills, and expand their
confidence in analyzing and solving new problems independently.
We emphasized the importance of communication and working in
teams in the process of problem solving. We also used the technique
of attacking problems,1281 with one of a pair playing the role of
problem solver and the other the listener, and then alternating roles.
The first problem-solving assignment, worth 10%, gave randomly
selected student pairs the opportunity to apply this technique to a set
of problems taken from the McMaster problem-solving program.[1l]
The second assignment was based on the Fogler interactive computer
modules and was also worth 10%. Each pair of students was assigned
two of the computer modules to complete each week for three weeks-
six modules in total. At the completion of each module, a computer-
generated performance score was recorded by students and handed in
as part of their assessment. Students were given the option of repeat-
ing the modules as often as they wished to improve their score-and
some did, with their best score being credited. A questionnaire com-
pleted by students at the end of the design course provided an evalua-
tion of Whimbey pairs[281 and the computer problem-solving modules.

The distribution of marks in the first assignment was high, with a
skew toward a possible score of ten. When working through the

Assignment 1 and Grade Point
E 8
M1 7
a 5
5 6 7 8 9
Grade Point Average (Adjusted)

Figure 1. Marks on Assignment 1 compared with
adjusted grade point average.

Assignment 2 and Grade Point
E. 8
r 7
a M 7
eg 6
< 5
5 6 7 8 9 10
Grade Point Average (Adjusted)

Figure 2. Marks on Assignment 2 compared with
adjusted grade point average.

Student Comments on Whimbey's Method

Problem Solver
As the problem solver, I found that when solving
problems, I tend to like to put things into mathematical
For the basic problem, I did write out more than I usually
would. It was fun and I would like to do more of it.
I think I work too much out in my head and tend to rush
to give an answer.
I tend to attack problems head on, noting down all the
information supplied as I read it through.
I enjoyed solving these problems.
The help of a listener was very useful; their ideas and
reasoning are often very different and it's good to
compare and see their point of view.

In problem solving you must read the question carefully,
jot down any conditions, and then determine what the
problem is asking you to solve.
The hardest thing was not to get carried away and tell the
problem solver the answer when I knew it.
It was a much easier task to be the listener than the solver.
It's good to try to show the other person a different way
of thinking.
Being the listener is not an easy task!
Listening is generally not too hard-often the solver
doesn't vocalize everything.

Chemical Engineering Education

Computer Modules are a Useful
Supplement to Lectures


0 m1997

No Indifferent Yes

Figure 3. Student response to the usefulness of the
interactive computer modules.

Figure 4. Students opinion of the value of the
interactive problem-solving modules.

Computer-Based Assessment

30 -
20 0-1996

No Reply Very Poor Average Good Very
Poor Good

Figure 5. Student response to computer-based

Spring 1999

McMaster problem set, a number of students showed en-
thusiasm for the Whimbey pair concept, obtaining full
marks. Table 1 has a sample of student comments taken
from the questionnaire.
The assessment mark for each pair for two problem-
solving assignments has been compared with the group-
adjusted grade point average for the 1996 end-of-year ex-
aminations in Figures 1 and 2.
The grade point average (GPA) for each student is the
average grade for all papers in the 1996 examinations and
reflects the students' final academic achievement. The GPA
is assigned a numerical equivalent ranging from 9 for an
A+ to 2 for a C. The adjusted grade point average (GPAJ)
was recalculated to align means and standard deviation
with marks out of 10.
Statistical analysis showed no significant relationship
between problem-solving performance and previous aca-
demic results for the two problem-solving assignments. A
reason for the results is that problem solving is a different
skill from conventional academic performance. Addition-
ally, the results can be explained by differences in testing
procedures. The problem-solving assignments were power
assessments, without direct time constraints, whereas ex-
aminations were speed tests, with stringent time restric-
tions to the examination time.
Student responses to the questionnaire on the usefulness
of the computer modules, their rating of the interactive
problem solving modules, their opinion of computer-based
assessment and working in pairs are shown in Figures 3
through 6.
Figure 3 shows that 78% of students in 1996 and 1997
found the computer modules to be a useful supplement to
lectures, while 10% were indifferent and 2% did not find
them useful.
In Figure 4, student rating of the interactive problem-
solving modules indicated a positive response, with fewer
than 5% (1996) and fewer than 10% (1997) rating them as
worse than average. Students responded well to the practi-
cal problems in the computer modules, helping to under-
stand different problem-solving techniques.
Computer-based assessment was introduced to the stu-
dents in 1996. Figure 5 shows that 61% (1996) and 63%
(1997) of the students found this form of assessment very
good, good, or average; 17% of the students (1996) did not
respond to this question.
The interactive computer modules provided a new and
different environment for learning that students found to be
a useful supplement to lectures. Working in pairs for As-
signments 1 and 2 enabled students to help each other with
problem solving. Most of the students supported working
in pairs for Assignment 2, as shown in Figure 6; 78% of the
Continued on page 157.

Figure 6. Student response to working in pairs.

e, M 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.


A Problem for the Stoichiometry Course

North Carolina State University Raleigh, NC 27695-7905

During the fall 1997 semester at N.C. State Univer-
sity, I was both captain of the ice hockey club and
the head teaching assistant for the problem session
of the stoichiometry course. It occurred to me that the pro-
cess used to freeze an ice rink illustrates several stoichiom-
etry course topics, so I constructed an exercise that called for
the students to go to a home hockey game, observe the ice
resurfacing operation, and estimate the power rating of the
rink compressor. I believe the exercise served a number of
useful functions, including reinforcing the students' under-
standing of several thermodynamic principles and methods,
showing them that those principles and methods have real-
world relevance, and introducing many of them (notably,
those born and raised in the South) to the fun and excitement
of ice hockey. The fact that it increased attendance at our
home games didn't hurt either.

David Dudek earned a BS in Chemical Engi-
neering and a BA in Economics in 1993 from
the University of Illinois. He received an MS in
Management in 1997 and a PhD in Chemical
Engineering in 1998 from NC State University.
He completed his PhD in chemical engineering
in 1998. His dissertation research concerns mod-
eling and experimental aspects of copper elec-
trodeposition from cuprous cyanide electrolyte.
He also has nineteen years of ice hockey expe-

About a month before the end of the semester I handed out
an outline of the ice resurfacing process and a series of
problems to be solved. I told the students that if they at-
tended a game and made a reasonable effort to solve the
problems, they could replace their lowest problem-session
grade with a 100 (if their estimated compressor rating was
within a factor of three of the actual rating) or an 80 (if their
estimate was outside that range). Although I required game
attendance to receive credit on the problem, attendance is
not necessary to complete the problem presented here.
This paper presents the process description and problem
statement, outlines the problem solutions, and summarizes
the students' performance on the exercise.

The process used to freeze an ice rink is shown schemati-
cally in Figure 1. At the lower left of the figure, saturated
Freon (R-22) vapor (low T, low P) enters a compressor in
which its pressure is raised. Essentially all of the shaft work
transmitted to the gas in the compressor is converted to
internal energy and the gas temperature and pressure accord-
ingly increase. The Freon gas leaving the compressor (high
T, high P) comes into thermal contact with much cooler air
in a heat exchanger (the condenser) and is completely con-
densed. The saturated liquid Freon (low T, high P) goes
through an adiabatic expansion valve in which its pressure

Copyright ChE Division of ASEE 1999

Chemical Engineering Education

drops substantially.
At the lower pressure, the Freon is above its boiling point and some of it vaporizes, which in
turn leads to a dramatic drop in the Freon outlet temperature. (Consider: When you are wet and
water evaporates from your body, you get cold.) The liquid/gas mixture (low T, low P) enters a
second heat exchanger (the boiler) where it comes into thermal contact with cool liquid
propylene glycol (a liquid with a high boiling point similar to ethylene glycol, the coolant used
in automobile radiators). Heat is transferred from the cold propylene glycol to the much colder
Freon, chilling the glycol and vaporizing the remaining Freon liquid. The Freon gas (low T,
low P) then circulates back to the compressor to begin another cycle. The cold propylene
glycol passes through coils under the ice. Heat flows from the ice to the glycol, causing any
liquid water on the ice to freeze and warming the glycol, which circulates back to the boiler to
be chilled again.
The rink freezing process has two distinct periods: one when liquid water placed on the rink
is frozen and the power load on the compressor is high, and the other when the ice is simply
being kept from melting and the power load is low. The required compressor power rating
(size) is based on the load during the first of these periods.

(a) You have just seen that a major component of a process to freeze water is a compression
that heats a gas to a temperature well above 0C. Most people would find heating a gas in
order to freeze a liquid strange. Explain in your own words how the work done by the
compressor eventually leads to freezing the rink.
(b) Consider the Freon and propylene glycol collectively as a system. Where in the process is
energy added to the system and where is energy transferred from the system? (Neglect
heat losses from the lines connecting the different process units.) Assuming that the
system operates at steady state, write an expression relating the various energy inputs and
(c) The purpose of resurfacing the ice is to turn rough, snow-covered ice into a smooth flat
surface. To do this effectively, the Zamboni ( collects the
snow, shaves a thin layer of ice from the surface, and uses hot water to melt the remaining
grooves and leave a smooth wet surface. Thus, the Zamboni is filled with water from the

Figure 1. Cooling system for a typical ice rink.

It occurred
to me that
the process
used to
freeze an
ice rink
topics, so I
that called
for the
students to
go to a
the ice
the power
rating of
the rink

Spring 1999

hot water tap at the rink. At most rinks, this water is
warmer than the water in the average household water
heater. Estimate the temperature (C) of the water com-
ing out of the Zamboni.
(d) Make a reasonable guess at the temperature (C) of the
ice on the frozen rink.
(e) How much energy (kJ/mol) must be transferred to
(i) Cool water from the temperature you estimated in
Part (c) to its freezing point
(ii) Freeze water at 1 atm
(iii) Cool ice from the freezing point to the temperature
estimated in Part (d).
(f) How much energy (kJ/mol) must be transferred to con-
vert water from the Zamboni to ice at the rink tempera-
ture? If your temperature estimates in Parts (c) and (d)
are each incorrect by 100C, what is the maximum per-
centage error in your calculated energy? Why is the
error so small?
(g) If the Zamboni puts 200 gallons of hot water on the ice
and it takes 15 minutes for the entire surface to freeze,
at what rate (kJ/h) must heat be transferred to the cool-
ing coils under the rink?
(h) Compressors are often rated in tons of refrigeration,
where one ton is equivalent to 12,000 BTU/h. Gener-
ally speaking, the required compressor shaft work is
about 30% of the heat load from the ice, and compres-
sors at ice rinks are about 40% mechanically efficient
(meaning that 40% of the total power generated by the
compressor is delivered as shaft work to the process
fluid). Using these rules of thumb, estimate the rating
of the compressor in tons of refrigeration.
(i) Apply the energy balance derived in Part (b) to estimate
-Qc, the rate at which energy is removed from the
system by the compressor.

(a) Compressing the gaseous Freon increases its pressure
and temperature. At the higher pressure, the boiling
point of the Freon is well above room temperature.
Thus, a large amount of energy can be removed from
the Freon by condensing it with room-temperature air.
When the Freon passes through the expansion valve,
its pressure returns to the pre-compression pressure,
but its temperature falls well below the pre-compres-
sion temperature. At the lower pressure, the boiling
point of Freon is below the freezing point of ice. Thus,
a large amount of energy can be removed from the
coolant under the ice by boiling the low-pressure
(b) Energy is added to the system by the ice (Qie) and by
the compressor (Ws). Energy is removed from the

system by the condenser (Qc). Thus, the energy
balance is Qice + Ws + Qc = 0.
(c) 120 to 1600F, or 50 to 70C
(d) 18 to 24'F, or -4 to -80C
(e) Cp,water = 75.4 J/mol 'C; AHmeting= 6,009 J/mol;
and C,,ce = 36.7 J/mol C.
(i) 75.4 J/mol C x (0-60) C = 4,524 J/mol
(ii) -6,009 J/mol
(iii) 36.7 J/mol C x (-6-0) C = -220 J/mol
(f) -4,524 J/mol + -6,009 J/mol + -220 J/mol =
10,753 J/mol.
If both temperature estimates are off by 100C, the
error is only about 10%. This number is so small
because a large portion of the energy requirement in
Part (e) is the latent heat of freezing.
(g) 200 gal x (1 m3/264.17 gal) x 1000 kg/m3 = 757 kg
10,753 kJ/mol x 757 kg x (1 kmol/18 kg) 0.25 hr =
1.81 x 106 kJ/hr
(h) 1.81 x 106 kJ/hr x 1000 J/kJ x 9.486 x 10-4 BTU/J x (1
ton/12,000 BTU/hr) = 143 ton
0.3 x 143 ton + 0.4 = 107 ton
(i) -Q = Qice + W
-Q = 143 + 107 = 250 ton

Of 144 students taking the course, 43 (30%) elected to do
the problem. Of the 43, roughly one-third did the calcula-
tions correctly and came within a factor of three of the actual
compressor rating of 130 tons, one-third made poor esti-
mates of intermediate quantities and estimated ratings out-
side the allowed range despite doing the calculations cor-
rectly, and the remaining one-third did the calculations in-
correctly. The problem given to my students required esti-
mation of the volume of hot water placed on the ice; because
this was the most difficult number to estimate accurately, the
number (200 gallons) is provided in the above problem
statement. Common mistakes included using the heat capac-
ity of water vapor instead of liquid, making mistakes in signs
of sensible and latent heats, and unit conversion errors.

I am indebted to Professors Richard Felder for carefully
reading the problem and providing numerous useful sugges-
tions and Peter Kilpatrick for helpful discussions regarding
the thermodynamics of refrigeration cycles. Thanks also to
Randy Lee for an educational tour of the boiler room at The
Ice House in Cary, North Carolina, and for providing accu-
rate parameter values. O
Chemical Engineering Education

Problem-Solving Skills
Continued from page 153.

students found it to be very useful, useful, or average. Shar-

ing ideas and discussing them were
understood to be valuable problem-
solving techniques. The main disad-
vantage for students was having to
plan time to work together. In 1997,
student opinion of the problem-solv-
ing modules improved significantly,
as shown in Figures 3 through 6.

Each module was assessed and marks
recorded on-line. Tables 2 and 3 show
the mean and standard deviation of stu-
dent assessment for each computer
module. In 1996, all students scored
full marks on the introduction module,
whereas this module ranked fourth in
1997. Planning and brainstorming
ranked the lowest for both 1996 and
1997. We believe that students lack the
necessary experience in these skills that
are developed later in the workforce.

The development of problem-solv-
ing skills is an integrated part of the
teaching of design at the third-year
level of our chemical and process en-
gineering degree course. Students ap-
preciated the problem-solving approach
to assignments. Working in pairs for
problem solving was found to be ben-

Student Assessment
Each Computer Module,

*Introduction 1.00
*Potential Problems 0.98
*Evaluation 0.89
*Define 0.86
Planning 0.68
Brainstorming 0.41

Student Assessment
Each Computer Modul4

* Define 0.92
* Evaluation 0.87
* Potential Problems 0.86
* Introduction 0.83
* Brainstorming 0.65
SPlanning 0.64

eficial by most of the students, although arranging a suitable
time to work together was a disadvantage.
Problem solving is a skill that can be learned. It is impera-
tive that our graduates have the necessary skills and strate-
gies to deal confidently with new situations and problems
encountered in their professional careers.

1. Wickelgren, W.A., How to Solve Problems, Freeman, San
Francisco, CA (1974)
2. Polya, G., How To Solve It, Doubleday, New York, NY (1957)
3. Polya, G., Mathematical Discovery, Vol. 1: On Understand-
ing, Learning, and Teaching Problem Solving, John Wiley,
New York, NY (1962)
4. De Bono, E., The Use of Lateral Thinking, Penguin,
Harmondsworth (1967)
5. De Bono, E., The Five-Day Course in Thinking, Penguin,
Harmondsworth (1968)
6. De Bono, E., The Mechanism of Mind, Penguin,
Spring 1999

Harmondsworth (1969)
7. De Bono, E., PO: Beyond Yes and No, Revised Edition,
Penguin, Harmondsworth (1973)
8. Wertheimer, M., Productive Thinking, Tavistock, London

9. Johnson-Laird, P.N., and P.C. Watson,
eds., Thinking: Readings in Cognitive
Science, Cambridge University Press,
Cambridge (1977)
10. Newell, A., and H.A. Simon, Human
of Problem Solving, Prentice-Hall,
1996 Englewood Cliffs, NJ (1972)
11. Kahney, H., Problem Solving: A Cogni-
tive Approach, Open University Press,
SMilton Keynes (1986)
Demln 12. Johnson-Laird, P.N., Mental Models,
0.01 Cambridge University Press, Cambridge
0.06 (1983)
13. Wittrock, M.C., "Students' Thought Pro-
cesses," in Handbook of Research in
0.15 Teaching, M.C. Wittrock, ed.,
0.21 Macmillan, New York, NY; (1986)
0.25 14. Woods, D.R., The McMaster Problem
Solving (MPS) Program, Chemical En-
gineering Department, McMaster Uni-
versity, Ontario, Canada (1985)
15. Fogler, H.S.L., S.E. LeBlanc, and S.M.
Montgomery, Strategies for Creative
Problem Solving, University of Michi-
of gan, Ann Arbor, MI (1995)
e, 1997 16. Ko, E.I., and J.R. Hayes, "Teaching
Awareness of Problem-Solving Skills to
Staunid Engineering Freshmen," J. of Eng. Ed.,
Deaition October (1994)
0,10 17. Allen, R.M., I.A. Gilmour, and J.G.
Mackenzie, "Creative Problem Solving,"
0.21 Proceedings of 24th Australian and New
0.09 Zealand Chemical Engineering Confer-
0:11 ence and Exhibition, 4, Sydney, Austra-
0.27 lia, (1996)
18. Maul, G.P., and J.S. Gillard, "Teaching
7 Problem Solving Skills," Computers and
Ind. Eng., 31(1/2) (1996)
19. Woods, D.R., "A Strategy for Problem
Solving," Course Notes, McMaster Univer-
sity, Ontario, Canada (1983)
Felder, R.M., "Teaching and Learning Styles in Engineer-
ing Education," Eng. Ed., April (1988)
Fogler, H.S., S.M. Montgomery, and R.P. Zipp, Computer
Applications in Engineering Education, 1(1), September/
October (1992)
Snow, R., and M. Farr, Aptitude, Learning and Instruction,
Vol. 3: Cognitive and Affective Process Analysis, Erlbaum,
Hillsdale, NJ (1987)
Higgins, J.S., et al., "Identifying and Solving Problems in Engi-
neering Design," Studies in Higher Education, 14(2) (1967)
Adams, J.L., Conceptual Blockbusting: A Guide to Better
Ideas, W.H. Freeman and Co., San Francisco, CA (1974)
Morrison, D., University of Minnesota Solar Vehicle Project,
June 11 (1993)
Gantt, H.L., Work, Wages and Profits, New York, NY (1910)
Lockyer, K.G., "Critical Path Analysis," Accountants Digest,
No. 4, Institute of Chartered Accountants in England and
Wales, London (1991)
Whimbey, A., and J. Lochhead, Problem Solving and Com-
prehension:A Short Course in Analytical Reasoning, Franklin
Press, Philadelphia, PA (1980) 0

e, -classroom



Do We Blend the Old With the New?

University of Dayton Dayton, OH 45469-0246

ne of the most challenging tasks facing a chemical
engineering instructor today is presenting a capstone
design experience that comprises an appropriate bal-
ance between the "old" and the "new." The old would be the
classical design experience, heavy on the fundamentals, hand-
calculation intensive, and rigorous in approach, whereas the
new corresponds to a more team-oriented, computer-usage
intensive approach incorporating the development of written
and oral communication skills. Based on the proliferation of
articles in the literature on the subject of teaching design and
how design should be integrated into the curriculum,11-81 it is
obvious that this is an issue that continues to be debated and
for which there is no simple solution. There also appears to
be some disagreement between industry and academia as to
what skills a student should possess after completing a four-
year engineering program.19101
This article addresses some of the issues associated with
teaching design and in particular looks at the capstone de-
sign sequence developed at the University of Dayton and the
experience gained developing and teaching these courses.
The pros and cons of chemical-process flowsheet-simulator
use, together with feedback from students who have taken
the design courses, will also be discussed.

The design course sequence, as it has evolved over the last
seven years, is outlined in Tables 1 and 2. As can be seen,
the first course (Design I) is offered during the fall semester
of each year and the second course (Design II) during the
following semester. Design I is a fairly standard introduc-
tory design course covering topics such as the basic concepts
of design, safety issues, costing and economic analysis, ma-
terials of construction, and some mechanical design. Addi-
tional details can be found in the table. Evaluation of student
performance for this course is based on two midterm exami-

nations as well as a comprehensive final examination. Per-
formance on homework assignments is also weighted into
the final course grade.
Design II, the second course in the sequence, has two main
components: the design problem and special topics. The
design problem is a fairly comprehensive plant design (dis-
cussed later), and the special topics include diagrams and
layout, optimization, shortcut design procedures, and other
topics specifically related to the design project being under-
taken, e.g., if the plant design requires a bioreactor, some
details of microbial growth kinetics and fermenter design
would be discussed. Evaluation of student performance
for this course is based on a single examination (toward
the end of the semester) that covers the special topics as
well as some aspects of the design problem, and grades
from the design-project assignments. The assignment
grades would typically constitute approximately two-
thirds of the final course grade. Some additional details
can be found in Table 2.

As mentioned above, the major component of Design II is
a fairly detailed plant design. Details of some of the projects
undertaken are available in Table 2. Selection of appropriate
projects is important. They need to be feasible, yet challeng-

Copyright ChE Division of ASEE 1999

Chemical Engineering Education

Lawrance Flach, Associate Professor of Chemi-
cal Engineering at the University of Dayton, has
bachelors and masters degrees in chemical en-
gineering from the University of Cape Town,
South Africa, and a PhD from the University of
Colorado, Boulder. He conducts research in
the areas of process modeling, dynamics, and

Design I Course Outline
Fall Semester 3 Credit Hours
1. Introduction (0.5 week)
Basic concepts
Steps in the design procedure

2. Safety, Loss Prevention and Health Issues (2 weeks)
MSDS (Material Safety Data Sheet) terminology
Health, safety hazards: toxicity, flammability, explosions, etc.
Loss prevention, HAZOP studies

3. Capital Cost Estimation (2 weeks)
Equipment costing, cost charts, cost indices, etc.
Overall plant cost, Lang Factor method

4. Manufacturing Cost Estimation (1 week)
Fixed capital, working capital
Direct, indirect, general manufacturing expenses

5. Economics and Profitability Analysis (2 weeks)
Present value of money
Profitability analysis, discounted cashflow
Break-even analysis

6. Materials of Construction (3.5 weeks)
Mechanical properties of materials
Phase transformations and heat treatment
Special properties required of materials
Materials selection
Commonly used materials of construction

7. Mechanical Design of Process Equipment (1 week)
Design pressure, temperature, stress
Joint efficiency, shells and heads
Design equations
Vessel supports

Design II Course Outline
Spring Semester 3 Credit Hours

1. Design Problem (different problem each year) [class size]
1991 Revamp NGL processing unit (based on 1991 AIChE
Student Design Contest Problem) [27 students]
1992 Methanol from coal plant [27 students]
1993 MEK from 2-butanol plant [29 students]
1994 Ethanol production from molasses by fermentation [23
1995 Methanol plant (based on 1995 AIChE Student Design
Contest Problem) [25 students]
1996 Hydrogen from methane plant [37 students]
1997 Ammonia from natural gas plant [23 students]

2. Special Topics
Flowsheeting, P&I diagrams, symbols
Short-cut design procedures, "rules-of-thumb"
Process optimization
Plant layout
Computer-aided design programs, ChemCAD III
Other topics, as required for the design problem

Spring 1999

ing for the students, and preferably should incorporate a
variety of basic unit operations and some form of material
conversion. Exotic separation techniques or reaction sys-
tems are generally avoided due to their complexity, but may
be included to illustrate the power and capability of com-
puter simulation packages.
Students are guided through the plant design via a number
of assignments that are distributed throughout the semester
and are typically 10-14 days duration. Details of a typical
assignment sequence can be found in Table 3.
Each assignment is comprehensive and in some cases
requires the students to write computer programs to perform
the design, e.g., incremental design of a packed-bed catalytic
reactor. Some aspects of the mechanical design of vessels
are included where appropriate.
A report is submitted upon completion of each assign-
ment, and currently each student is expected to submit a
report, i.e., report writing is performed individually. Al-
though some level of collaboration between students does
occur while working on these assignments, excessive col-
laboration is discouraged because overall course grades are
individual and to a large extent are based on the assignment
reports. Performing these assignments and writing the re-
ports in teams is something that has been considered because
of the "teamwork" experience that would be gained by the
students and also for the reduced work load in evaluating the
assignment reports. To date, however, we have continued to
require individual assignment reports, mainly because we
view student performance in this course as somewhat of an
overall indicator of the student's ability, and as such we have
preferred an individual assessment of performance.

Using chemical process flowsheet simulators, such as
ChemCAD,"11 has been gradually introduced during Design
II over the last five years. Initially, use was limited to certain
assignments, and in some cases hand calculations were re-
quired prior to using the simulation package. More recently,
use of simulation packages has not been limited and students

Typical Sequence of Design Project Assignments

1. Process background and literature, overall material balance
2. Material and energy balances
3. Reactor design
4. Heat exchanger design
5. Distillation or absorption design
6. Miscellaneous equipment design and preliminary costing
7. Economic analysis, final report


This article addresses some of the issues associated with teaching design and
in particular looks at the capstone design sequence developed
at the University of Dayton and the experience gained
developing and teaching these courses.

have been free to use them as much as they like. This
approach, however, has had mixed results. Some students
revel in the situation, can use the packages well, and get
good results quickly. Others appear to become obsessed
with getting a computer simulation to work for a situation
where a few hand calculations would be perfectly satis-
factory. As a result, a tremendous amount of time and
effort is wasted on getting the packages to work rather
than learning anything of engineering significance. This
is the situation that we had tried to avoid by limiting
package use previously.
One cannot question the ability of simulation packages to
efficiently and conveniently perform many chemical engi-
neering calculations, but one should never forget that using
these packages is sometimes more difficult than doing calcu-
lations by hand, and that sometimes the results obtained may
just not be realistic or appropriate for the situation being
considered. Many students tend to present results obtained
from a simulation without really considering their applica-
bility. Sometimes a "back-of-the-envelope" hand calcula-
tion or "common sense" can reveal where problems lie.
One should thus try to avoid situations where simulation
packages are overused and thus become a substitute for a
basic understanding of the chemical and physical phe-
nomena occurring.

Student feedback concerning these two courses has been
generally positive. On the university-administered course
evaluation, when asked to give the courses overall ratings,
Design I received an average of 2.8 and Design II an average
of 3.0 for the last six years (4=excellent, 3=above aver-
age, 2=average). When asked if they had learned a great
deal from the courses, Design I received an average of
3.1 and Design II an average of 3.3 (4=strongly agree,
3=agree, 2=neutral).
On a departmentally administered comment sheet used for
Design II, students were asked about the course's strengths,
its weaknesses, and suggestions for improving it. As is typi-
cal for such a survey, there was a wide variety of responses.
The general nature of the student comments was

Course strengths: Good capstone course, incorpo-
rating many of the skills and techniques learned
throughout the curriculum; linked many of the topics
studied separately prior to this course.

Course weaknesses: Excessive work load and time-
consuming; lack of guidance on some assignments.
Suggestions for improvement: More guidance and
details of what is expected on assignments; reduce
workload (possibly by working in teams).
The workload during Design II is high, and we are ad-
dressing the issue by making some curriculum changes that
will allow us to distribute more of the workload between
Design I and Design II. Workload during Design I is typi-
cally a lot lower than that required during Design II, and we
believe that the changes that we will introduce in the near
future will alleviate this problem and result in a more even
distribution of effort. The teamwork issue is one that contin-
ues to be debated.

The senior-year chemical engineering design sequence
at the University of Dayton has evolved over the last
decade. Some of the "new" innovations (use of process
flowsheet simulation packages) have been introduced with
some success, but for the most part we have retained a
more traditional approach to teaching these courses. This
was intentional.
We are living in a time when industry expects students to
graduate not only with a good fundamental understanding of
chemical engineering principles, but also with the skills
traditionally acquired after graduation via industrial or work
experience. The latter are skills best acquired in an actual
work environment, and as such can be acquired by a student
through internship or co-op programs. If a student chooses
not to participate in these programs, then prospective em-
ployers need to appreciate that the graduate will be lacking
that practical experience and associated skills. If the gradu-
ate possesses a good understanding of the fundamentals,
however, then these experience-based skills can be acquired
fairly rapidly in the work environment.
Universities, and engineering programs in particular, are
under tremendous pressure to recruit students and then to
retain them at all costs. As engineering educators, we must
resist the impulse to accomplish these goals by "watering
down" the programs. The design sequence described in this
paper demonstrates our resistance to the trend and indicates
that we are trying to provide our students with a good grasp
of the fundamentals. Feedback that we have received from
industry-employed graduates indicates that they have ben-

Chemical Engineering Education

efited from this approach and that they have been more
than capable of competing and succeeding in today's
work environment.

1. Dutson, A.J., R.H. Todd, S.P. Magleby, and C.D. Sorensen,
"A Review of Literature on Teaching Engineering Design
Through Project-Oriented Capstone Courses," J. Eng. Ed.,
86(1), (1997)
2. Fentiman, A.W., and J.T. Demel, "Teaching Students to
Document a Design Project and Present the Results," J.
Eng. Ed., 84(4), (1995)
3. Rockstraw, D.A., J. Eackman, N. Nabours, and S. Bellner,
"An Integrated Course and Design Project in Chemical Pro-
cess Design," Chem. Eng. Ed., 31(2), (1997)
4. Bell, J.T., "Implementation of Multiple Interrelated Projects
Within a Senior Design Course," Chem. Eng. Ed., 30(3),
5. Shaeiwitz, J.A., W.B. Whiting, and D. Velegol, "A Large-
Group Senior Design Experience," Chem. Eng. Ed., 30(1),
6. Cameron, I.T., P.L. Douglas, and P.L. Lee, "Process Sys-
tems Engineering: The Cornerstone of a Modern Chemical
Engineering Curriculum," Chem. Eng. Ed., 28(3) (1994)
7. Seider, W.D., and A. Kivnick, "Process Design Curriculum
at Penn: Adapting to the 1990s," Chem. Eng. Ed., 28(2)
8. Bailie, R.C., J.A. Shaeiwitz, and W.B. Whiting, "An Inte-
grated Design Sequence: Sophomore and Junior Years,"
Chem. Eng. Ed., 28(1) (1994)
9. Lee, W.E., and R.R. Rhinehart, "Do We Really Want 'Aca-
demic Excellence?'" CEP 93(10) (1997)
10. Horwitz, B.A., and L.G. Nault, "Relate to the Real World,"
CEP, 92(10) (1996)
11. ChemCAD III Process Flowsheet Simulator, Chemstations,
Inc., Houston, Texas. 0
Spring 1999

Babooks received )

Capillary Electrophoresis, by Baker; Wiley, 605 Third Avenue, New York,
NY 10158; 244 pages (1995)
Encyclopedia of Chemical Technology: Helium Group to Hypnotics; 4th
ed, Kirk-Othmer; Wiley, 605 Third Avenue, New York, NY 10158; $295
Intermediate Classical Dynamics with Applications to Beam Physics, by
Michelotti; Wiley, 605 Third Avenue, New York, NY 10158; 340 pages,
$59.95 (1994)
Heterocyclic Compounds, 2nd ed., by Coppola and Schuster; Wiley, 605
Third Avenue, New York, NY 10158; 552 pages, $225 (1995)
Handbook of Laboratory Health and Safety, 2nd ed., by Stricoff and Walters;
Wiley, 605 Third Avenue, New York, NY 10158; 462 pages, $69.95 (1995)
Organic Reactors, Vo. 47, various editors; Wiley, 605 Third Avenue, New
York, NY 10158; 574 pages, $95 (1995)
High-Speed Countercurrent Chromatography, edited by Ito and Conway;
Wiley, 605 Third Avenue, New York, NY 10158; 454 pages, $79.95 (1995)
TProgress in Inorganic Chemistry, Vol. 43, edited by Karlin; Wiley, 605
Third Avenue, New York, NY 10158; 621 pages, $125 (1995)
Chemical Equilibria Bases for Oxide and Organic Superconductors, (in-
cludes disk), by Thorn; Wiley, 605 Third Avenue, New York, NY 10158;
200 pages, $69.95 (1996)
FORTRAN Programs for Chemical Process Design, Analysis, and Simula-
tion, by Coker; Gulf Publishing, 9284 Baythore Road, Houston, TX 77040;
854 pages, $95.00 (1996)
Electronic Structure and Properties of Transition Metal compounds: Intro-
duction to the Theory, by Bersuker; Wiley, 605 Third Avenue, New York,
NY 10158; 668 pages, $95.00 (1996)
Special Trends in Thermal Analysis, by Paulik; Wiley, 605 Third Avenue,
New York, NY 10158; 459 pages (1995)
Symmetry: A Basis for Synthesis Design, by Ho; Wiley, 605 Third Avenue,
New York, NY 10158; 561 pages, $69.95 (1995)


Fall 1999
Graduate Education Issue of

Each year, Chemical Engineering Education publishes a special fall issue devoted to graduate education. It
includes articles on graduate courses and research as well as ads for university graduate programs.
Anyone interested in contributing to the editorial content of the 1999 fall issue should
contact CEE, indicating the subject of the contribution and the tentative date of submission.

Deadline is June 1, 1999

Chemical Engineering Education c/o Chemical Engineering Department
University of Florida Gainesville FL 32611

Phone and Fax: 352-392-0861 e-mail: cee

pM learning in industry

This column provides examples of cases in which students have gained knowledge, insight, and experience in the
practice of chemical engineering while in an industrial setting. Summer internships and co-op assignments typify such
experiences; however, reports of more unusual cases are also welcome. Description of the analytical tools used and the
skills developed during the project should be emphasized. These examples should stimulate innovative approaches to
bring real world tools and experiences back to campus for integration into the curriculum. Please submit manuscripts to
Professor W. J. Koros, Chemical Engineering Department, University of Texas, Austin, Texas 78712.



The M.I.T. Practice School in Japan


Massachusetts Institute of Technology
Cambridge, MA 02139
Mitsubishi Chemical Corporation, 3-10 Ushiodori
Kurashiki Okayama 712, Japan

Graduates entering today's increasingly global chemi-
cal industry require not only strong technical skills,
but also the ability to apply those skills successfully
to solve practical problems along with an appreciation of the
diverse features of industry around the world. Although
classroom education is not the optimal way to develop all of
these skills, it can be well complemented by practical experi-
ence gained outside the traditional university environment.
Many internship and cooperative education programs exist
to provide students with industrial experience during their
undergraduate and graduate engineering courses, but expo-
sure to the international world of industry and business is
generally not available to students until after they commence
graduate employment.
Recognizing the importance of developing an understand-
ing of international industrial practices in its graduates, the
David H. Koch School of Chemical Engineering Practice
(the "Practice School") opened a new chapter in its history
two years ago by initiating its inaugural overseas station.
The Practice School, administered by the Department of
Chemical Engineering at M.I.T., has educated chemical en-
gineers in both the science of chemical engineering and the
art of chemical engineering practice since 1916, with only
Copyright ChE Division of ASEE 1999

Chemical Engineering Education

Andrea J. O'Connor was the Station Director
of the Mitsubishi Chemical Corporation Station
in the summer of 1997 and the Molten Metal
Technology Station in the summer of 1996.
She holds a PhD in chemical engineering from
the University of Melbourne and was a Fulbright
Postdoctoral Fellow at M.I. T. in 1995-96. She is
now a lecturer at the University of Melbourne
and conducts research in separation processes
and surfactant systems.

Angelo W. Kandas was the Assistant Director of the Mitsubishi Chemi-
cal Corporation Station in the summer of 1997. A Practice School gradu-
ate, he holds a PhD in chemical engineering from M.I. T., where his thesis
was on 'The Evolution of Carbon Structure During Oxidation." He is now
a Senior Process Engineer at Intel Corporation, conducting research on
oxide etching. (Photo not available.)
Yukikazu Natori is Deputy Director of the De-
velopment and Engineering Research Center
at the Mitsubishi Chemical Corporation's
Mizushima Plant. He joined Mitsubishi Chemi-
cal Corporation after graduating from Tokyo
Institute of Technology with a Master of Sci-
ence in chemistry and has diverse experience
in the development and design of petrochemi-
cal plants at the Mizushima Plant.

T. Alan Hatton is the Ralph Landau Professor
of Chemical Engineering Practice and has been
Director of the Practice School program since
1989. A native of South Africa, he graduated
from the University of Wisconsin, Madison,
with a PhD in chemical engineering and has
research interests in novel separation pro-
cesses and interfacial phenomena.

brief interruptions during the two World Wars.'" The Mitsubishi Chemical Corporation (M.C.C.)
Mizushima Plant in Kurashiki, Japan, was selected as the site of the program's first industrial
station outside the U.S., and it hosted seven M.I.T. student for two months in the summer of 1997.
The key issue in establishing an overseas Practice School station was to ensure that the quality
of the educational experiences gained by the students was not diminished by the cultural or
language differences, but rather was enhanced by exposure to these differences. With careful
preparation by M.I.T. and M.C.C. staff and strong support of the students' endeavors at the station, the
station operated very successfully. This article describes how the station was established and the
benefits and difficulties of the addition of an overseas station to the Practice School program.

Students enrolled in the Practice School undertake two semesters of graduate-level courses in
chemical engineering at M.I.T., followed by three to four months of intensive project work at two
remote industrial stations. Successful completion of these tasks leads to graduation with a Master
of Science in Chemical Engineering Practice. The industrial project work takes the place of the
thesis component of some other Masters programs, and a high standard of achievement is therefore
expected. Industrial stations have operated at a number of different company sites-recently Dow
Chemical (Freeport, Texas), Merck Pharmaceutical (West Point, Pennsylvania), GE Plastics (Mt.
Vernon, Indiana), and Cargill (Minnespolis, Minnesota) have hosted Practice School groups.
The students work in groups of two or three, and each group works on one project for four
weeks. In this short time, the students must assimilate the problem they are assigned, develop a
method of approach, carry out the project work, and present both written and oral reports. Their
efforts are supervised and assessed by the Station Director and the Assistant Station Director, both
M.I.T. staff members who reside at the station full-time. The program is structured with regular
meetings and reporting deadlines to ensure the students' efforts are appropriately focused and
that good communication is maintained between the students, the project sponsors, and the
Station Directors for the duration of the project. Details of the program structure have been
described previously.'21
One student from each project group is designated as the group leader and is responsible for
management of the project and effective communication throughout the course of the project. All
students contribute to oral and written reports each month, and the groups and group leaders are
changed for each new project. Company engineers act as project sponsors, providing problems for
the students to work on and acting as consultants to the student groups. The M.I.T. staff and
students are bound by confidentiality agreements with the host companies, so they are granted
broad access to in-house data and know-how.
The projects must meet a number of criteria in order to be suitable for the Practice School: they
must be of educational value to the students; they must require in-depth technical work, original
thinking, initiative, and engineering judgment in their execution; and they must be of high priority
to the host company. Furthermore, the personnel, plant, and other resources necessary for the
project to progress must all be available during the four-week period that the project is assigned to
the students. Routine delays such as those required to purchase a new equipment item cannot be
accommodated after projects commence due to their short duration. It has been estimated by plant
project engineers at previous stations that, as a result of these students' single-project focus,
diligence, and aptitude, a typical pair of Practice School students can achieve in one month of
work on a project what a company engineer might require four to six months to complete.

The opportunity to experience from within the day-to-day operations of M.C.C.'s Mizushima
Plant was understandably inspiring to the students selected to attend this station. M.C.C. is the
largest chemical manufacturer in Japan, and the Mizushima Plant is one of their major operations,
with around 1800 employees and $1.1 billion worth of product shipped annually (1995 data). The
plant is located in a petrochemical industry complex with port facilities on the Seto Inland Sea,
and it manufactures a broad range of petrochemicals and memory media for computers. In
Spring 1999




exist to



during their

and graduate

but exposure
to the

world of
and business

is generally
not available

to students
until after



addition to the manufacturing facili-
ties, the Mizushima Plant has the De-
velopment and Engineering Research
Center (DERC), which conducts a
broad range of process development,
modeling, and optimization projects
to develop leading-edge technology for
the future operations of M.C.C. This
center was the host to the Practice
School, and projects were offered by
several specialist groups within the cen-
ter, led by M.C.C. engineers with strong
technical and English language skills.

Seven students were selected to attend
the station based on their expressed in-
terest to work in Japan. They spent one
month at the Dow Coming station (Mid-
land, Michigan) prior to arriving at
M.C.C., where they worked for two
months. It was important for the stu-
dents to gain experience in a U.S. Prac-
tice School station before tackling the
program in Japan, to familiarize them
with the program expectations and to
give them confidence in their ability to
meet these expectations in a more famil-
iar environment. The students all held
undergraduate engineering degrees and
had completed the graduate course re-
quirements of the Practice School pro-
gram at Cambridge. Only one of the se-
lected students spoke Japanese, so the
others also enrolled in an introductory
Japanese course at the local Adult Edu-
cation Center, and all attended two half-
day workshops run by the M.I.T. Japan
Program to prepare them for living and
working in Japan. Visas and work permits
were arranged for the students, the only
difficulty being the added documentation
and guarantees of support required for
some non-U.S. students.

The students arrived in Japan several
days before commencing work, enabling
them to recover from jet lag and accli-
matizing them to living in Japan. Apart-
ments were provided in the M.C.C. hous-
ing complex alongside Japanese employ-
ees and their families. The costs for hous-
ing, as well as for travel to and from
Japan, were covered by M.C.C. Initial
challenges facing the students included
learning to manage shopping, banking,

Case Study 1. Optimization of Polymerization Catalyst Properties

A polymerization catalyst is synthesized in a batch process by precipitation from a reaction
mixture following a complex sequence of reagent additions and temperature adjustments. In-
creased demand for certain polymer product grades necessitated improved control of the catalyst
particle size distribution (PSD) in order to maximize the production capacity of the polymeriza-
tion train.
The students initially performed a statistical analysis on the existing laboratory and pilot-plant
data for the catalyst synthesis under a range of conditions. Using techniques learned in a Practice
School course on statistical analysis and experimental design, they were able to determine which
variables have a dominant effect on the catalyst PSD, but did not have sufficient data to quantify
these effects. So they used the statistical analysis software SAS to design an experimental program
to investigate the effects of the dominant parameters. Appropriate experimental design was
important to maximize the information obtained from the limited number of laboratory experi-
ments possible in the short time frame, as each experiment took about fourteen hours to complete.
The students then conducted the experiments in collaboration with their project sponsors.
They planned their time carefully so that two students were in the laboratory at all times during
an experiment; the third group member was also in the laboratory for the more labor-intensive
parts of the experiments, and in the office the rest of the time. This was necessary to enable the
group to read sufficient background material, to develop data-analysis techniques, and to meet
the weekly reporting deadlines. They arranged a roster to share the duties most efficiently.
On-line measurements of the PSD using a laser-reflectance probe provided extra information
from the students' experiments, enabling them to identify further important variables that had
not been previously studied. It also enabled them to postulate the mechanisms governing the
PSD, providing key insights into the physics of the process. The experimental results were added
to the existing data set and analyzed using SAS to determine the dependence of the catalyst PSD
on the manipulated variables. The results confirmed the trends previously observed in pilot-plant
trials, but also identified some important variables that were previously not considered signifi-
cant. The students recommended changes to the catalyst synthesis procedure based on their
results, and they were implemented with distinct improvements in the catalyst properties.

Case Study 2. Process Flowsheet Development for a Byproduct Purification

A process was being developed to implement a new catalyst in a synthesis process. The new
catalyst produces the desired product (A) plus a significant quantity of a valuable byproduct (B),
which needs to be purified in order to be marketable. The byproduct can be separated from the
major product stream relatively easily, but is contaminated with several other compounds,
including an undesirable byproduct (C). Byproducts B and C are difficult to separate because
they differ only in their degree of saturation. The clients wanted to determine the optimum plant
configuration and operating conditions to purify byproduct B.
The students studied previously conducted experiments and process simulations on azeotropic
distillation as a potential separation method for B and C and found that the physical property data
initially used in the simulations were inadequate, leading to gross overestimation of the separa-
tion efficiencies. They used refined physical property estimates to conduct Aspen Plus simula-
tions, obtaining results matching experimental data, which showed this technique to be infeasible.
They conducted an in-depth literature review of other possible separation techniques and
discovered several possibilities that had not yet been considered. They investigated the feasibil-
ity of a range of techniques, with the assistance of information from the literature, Aspen Plus
simulations, and consultations with experts at M.C.C. One new technique involving solvent
extraction with a pH swing was found to be favorable. As a result of this discovery, measure-
ments of the distribution coefficients of B and C in an appropriate extractant were rapidly
commissioned to provide the necessary data for the students to simulate this process. They used
these data in Aspen Plus simulations to design and optimize the flowsheet for the byproduct
purification section of the plant. The new process required many fewer unit operations than
previous proposals to meet the product specifications and also simultaneously removed other
contaminants from byproduct B.
The students conducted preliminary economic analyses on the proposals investigated and
made recommendations of further refinements needed in the simulations of the most favorable
proposal. As a result of the students' findings, the clients switched their attention from the
unfavorable azeotropic distillation initially proposed to a simpler and more cost-effective extrac-
tion process.

Chemical Engineering Education

After their final oral project reports, the M.I.T. students (Celia Huey,
1; Karen Zee, 2; Justin Zhuang, 3; Alejandro Cano-Ruiz, 4; Thomas
Gubiotti, 5; Susan Dusenbery, 6; Tanya Moy, 7) with the Practice
School Directors (Alan Hatton, 8; Andrea O'Connor, 9; Angelo Kandas,
10) and M.C.C. staff members (including Plant General Manager,
Mitsuyoshi Mitsuoka, 11; D.E.R.C. Director, Hiroyuki Kobayashi, 12;
D.E.R.C. Deputy Director, Yukikazu Natori, 14).

and traveling around the local area where English was rarely
spoken. Bus transport, provided by M.C.C. to all employees,
was used to travel to the plant on weekdays, and bicycles
were used on the weekends.
During the first week in Japan, prior to the start of the
station operations, the students were taken on a three-day
excursion to Kyoto and Nara by bullet train. This trip served
as an excellent way for the students and directors to get to
know each other, and to learn a little about the culture and
history of Japan. Escorted by a member of the Personnel
Section of M.C.C., the students learned about some of the
significant sites in these two old capitals of Japan and gained
confidence in the day-to-day living skills they would need
throughout their stay in Japan.
The station facilities provided by M.C.C. included desks
in a large, open-plan office with other M.C.C. employees,
and company uniforms (a symbol of belonging to the organi-
zation). The students worked within a firmer daily schedule
than at U.S. stations, as they used the company bus for
commuting between the plant and the apartment complex, and
required an English-speaking staff member to be on site with
them when working on weekends, in case of emergencies.

At this station, the students met regularly with their project
teams, comprising several M.C.C. engineers and scientists
Spring 1999

for each project. They also benefited from interaction with
DERC senior engineers and technical specialists in areas
such as modeling and process optimization. M.C.C. staff
members from other sections, and even other sites, attended
the students' oral presentations, and the students had an
opportunity to make poster presentations in an annual M.C.C.
poster session, further broadening their exposure to com-
pany personnel. It was not generally possible, however, to
establish meaningful interactions between the students and
operators because of language barriers. Thus, all plant infor-
mation was gained via the project sponsors, and instructions
for experiments or plant trials run by the students had to be
communicated to the operators via the sponsors. While this
is a drawback of the language barrier, it is not atypical for
consultants working in foreign countries, and it was good
experience for the students, giving them an appreciation of the
difficulties this can generate.
Working at a Practice School station is often a time of
stress for students, and being in a foreign country can exac-
erbate this. Hence, a special effort was made to ensure that
some time for relaxation away from the workplace was set
aside each week. In order to avoid any feelings of isolation,
as well as to make the most of the opportunities to experi-
ence Japanese life and culture, activities or excursions were
Continued on page 171.




Demonstrating Environmentally Sound

Manufacturing Principles

University of Connecticut Storrs, CT 06269-3222

he Green Square Manufacturing Game addresses sev-
eral issues of engineering education. The first is the
need to instill future engineers with "environmen-
tally sound" thinking. The concept of pollution prevention
should be taught across all disciplines of engineering, and an
ideal place to introduce this concept is in the freshman
engineering curricula. The Green Square Game is perfectly
suited for such an application and can be "played" in a
single class period or spread over two class sessions,
with a historical perspective of US environmental laws
included throughout the game.
Another issue the game addresses concerns hands-on learn-
ing versus passive learning. It provides hands-on learning
and group design with minimal fuss or preparation time,
followed by an open discussion of the results. This approach
increases enthusiasm for problem solving, is easy to imple-
ment, and uses inexpensive, readily available supplies.

Suzanne S. Fenton earned her PhD in chemi-
cal engineering at the University of Illinois and
her BS in environmental engineering from North-
western University. In addition to her responsi-
bilities as Assistant Head of the ChE Depart-
ment at Connecticut, she provides innovative
hands-on classroom instruction for undergradu-
ate engineering and secondary school students.

James M. Fenton is ProfessorandActing Head
of Chemical Engineering at the University of f
Connecticut. He earned his PhD from the Uni-
versity of Illinois and his BS from the University -.
of California, Los Angeles, both in chemical en-
gineering. He is Director of the Pollution Pre-
vention Centerat Connecticut and does research
in electrochemical engineering.

The Green Square Game can also be easily adapted for use
at the secondary-school level as a tool to raise awareness
(and the status) of the engineering profession among high
school students. In this case, a faculty member or a second-
ary school teacher can lead the game, providing high school
students with a challenging and thought-provoking glimpse
of engineering relevance in today's world.
The game is played in two sections. During the first round,
students are told that the year is 1953. They are split into
different "companies" (of three, four, or five students each)
that manufacture green squares. Each of the companies is
competing to win YOU (the instructor) as their client. YOU
want to purchase green squares of a particular shade and size
(5 cm x 5 cm white paper squares painted green on one side).
The companies are put to work trying to replicate the sample
green square by using powdered blue and yellow tempura
paint. No mention is made of environmental impact or waste
disposal. Then, during the second round, the year is changed
to 1998. Again, YOU are the customer requesting green
squares, but this time the constraint of waste minimization is
placed on the "companies" and students must try to produce
the green square with as little waste generation as possible.
The game is an ideal hands-on classroom project for fresh-
man engineering and high school students; it highlights waste
minimization, pollution prevention, and industrial ecology
and offers valuable experience in group problem solving.
Faculty members at the University of Connecticut and sec-
ondary school teachers throughout the state have used the
game with great success. Secondary school teachers have
been introduced to the game as "participants" during
university outreach activities. The game was originally
developed by WRITAR11u as a role-playing exercise to
train state regulatory staff.

Copyright ChE Division of ASEE 1999

Chemical Engineering Education

The purpose of the game is to give students a taste of how
waste minimization influences a manufacturing process (the
production of green squares). This mock process is also
intended to help participants become more aware of sources
of waste, options for waste minimization, resistance to change,
and the importance of communication and cooperation in
waste-reduction efforts.
Underlying objectives of the exercise are to demonstrate
the technical challenges (and frustrations) of reducing waste
in an industrial setting and to demonstrate the non-technical
issues that influence waste reduction efforts, including cus-
tomer demands and competition.

1. Assemble participants into "companies" of three,
four, or five students each and seat them around the
"production floor" (flip chart paper spread on a table
is the best production floor since it lies flat and
"waste" is easily seen; a sheet of brown wrapping
paper, a cut-open paper grocery bag, or a sheet of
newspaper could also be used).
2. Each group is a "company" that manufactures green
3. Tell participants that you are a potential client who
would like to purchase green squares that look like
the model shown (a prepared sample). You will be
distributing the green squares overseas and are look-
ing for a new supplier for a $2-million contract next
year. You are asking the "companies" to compete to
win your business.
4. Ask the groups to invent their own company name
and write the team names on a flip
chart, blackboard, or viewgraph.

1. The years 1953.
2. The criterion for the competition is that
the company's product should exactly
match the model shown (including dry-
ness) in the allotted amount of time.
3. Companies use the materials provided
(powdered blue and yellow tempura
paints, paint brushes, mixing cups, wa-
ter, white paper, etc.) to produce a per-
fect green square. They are given 10 to
20 minutes (depending on the time
available) to complete the project.

At the end of the "production time," in-
struct the teams to stop. Inspect each team's
product and evaluate its efforts using the
Spring 1999

criteria suggested in Table 1 (or substitute your own crite-
ria). As the customer, you must assign value to criteria as
you see fit. If time permits, students may also be given an
opportunity to evaluate their own and fellow companies'
performances. Rate or rank-order company performance
based on the evaluations and instruct the teams to clean up.
Discuss the results of the first round. Discussion questions
for it might include:
Was it quiet while the exercise was going on?
Was the project time consuming?
What approaches did groups use to produce the square?
Was a lot of planning involved?
Did students work as a team?
Was waste a concern?
How much waste did the groups generate (production
floor cleanliness; number of contaminated brushes, cups,
and spoons; cleanliness of the back of the green square;
left-over green paint; contaminated hands, clothing, etc.)?
Were raw materials wasted?

1. The year is 1998.
2. Again, the goal is to manufacture green squares-but
with an additional constraint. In addition to product
quality and time, companies must consider environmen-
tal impact.

3. Explain that any surface or object that becomes con-
taminated with paint (of any color) becomes "hazard-
ous." This includes all materials, hands, clothing, table
surface, and floor. Explain that teams will be evaluated
on their ability to paint the square green while generat-
ing the least amount of "hazardous" waste.
.. 1 I 4. Let the teams begin production.

Again, allow 10 to 20 minutes to
complete the exercise.
At the end of the "production time,"
instruct the teams to stop. Again, inspect
each team's product and evaluate its ef-
forts. Table 1 contains a list of possible
evaluation criteria for this round. It is easi-
est to give each criteria equal weight. Rate
or rank-order company performance based
on the round-two criteria and instruct
teams to clean up.
As an alternative to this scheme, more
"realism" can be injected into the project
by assigning dollar values (based on prod-
uct quality) and costs to the items listed in
Table 1. Company performance can then
be judged by profits (i.e., value less cost).


ration Criteria

First-Round Evaluation Criteria
Color and ,ize match model
Color consistency
Drnness of sample
Cleanliness of back of sample

Second-Round Evaluation Criteria
Color and size match model
Color consistency
Dryness of sample
Cleanliness of back of sample
Amount of raw material used
Production floor cleanliness
Number of contaminated brushes
Numberof contaminated cups
Number of contaminated spoons
Left-over paint
Contaminated hands, clothing, etc.

For example, values of $1, $0.50, and $0.00 can be given for
perfect color match, acceptable color match, and unaccept-
able color match, respectively. Equal valuations can be as-
signed for color match, consistency, dryness, and cleanliness
(a perfect sample would be worth $4.00). Similarly, costs
can be assigned to raw materials (paint), equipment (brushes,
cups), labor, and environmental decontamination ($/area of
contamination?). Time permitting, different "value and
cost" schemes can be used in judging performance to
illustrate the interactions between environmental con-
cerns and engineering design.
Second-round discussion could include:
Was it quiet while the exercise was going on? Which
production round was more time consuming?
Did the second round require more planning? Did the
students work more as a team?
How did the focus change?
What techniques were used to minimize waste genera-
tion? Which ones were the most successful?
How much waste minimization was accomplished? How
is it quantified? Is zero discharge possible?
How was the product quality affected during the second
Solicit ideas on how to dispose of waste generated by
each company.
If incineration is recommended, what might happen to
the hazardous material? Solicit ideas on how to dispose
of toxic ash and air pollution.
If landfill is recommended, again ask what possible
environmental impact may result. Solicit ideas on how
to clean up contaminated water and soil, and on what
should be done with the hazardous remains.
Is there a compromise between product quality, cost
(time), and environmental concerns?
Where does the garbage that the students generate at
home go?
Encourage students to research industries in their own
communities that generate air, water, and soil pollution
and where that waste is disposed.


Materials (for each round)
Small paper cups (about 3-ounce size), two for the pow-
dered paints, one for water, and three for mixing
1 tablespoon of powdered blue tempura paint in a paper
cup is provided to each company
1 tablespoon of powdered yellow tempura paint in a
paper cup is provided to each company

Plastic spoons (popsicle sticks or plastic spatulas)
Inexpensive watercolor brushes
White construction paper
"Production Floor" (a sheet of brown wrapping paper, a
cut-open paper grocery bag, or a sheet of newspaper
spread on a flat table).

Short Version (less than 50 minutes)
To play the game in a short time period, it is best for the
instructor to pre-cut the squares and to have two production
floors for each company set up (one for each round) before
the students arrive. The students should be pre-assigned to a
company and the company name should be selected in ad-
vance of the class period. Restrict the production runs of
each round to 10 minutes and have prepared score sheets
available for judging performance. The first- and second-
round discussions can be combined if necessary, and any
issues not discussed in class can be given as homework.

Long Version (less than 120 minutes)
To play the game in a longer period, a lecture (Historical
Perspective: U.S. Industry and the Environment) is given so
the students are brought from 1850 up to 1953 before round
one is played. Then the lecture is continued from 1960 up to
today before round two is played. A brief description of this
historical perspective is given below. It is focused on indus-
trial pollution, but keep in mind that a large percentage of
waste was (and still is) generated by non-industrial
sources. Historical information can be found in refer-
ences 2 through 6.

1850-1900 U.S. industrial waste regulation between
1850 and 1900 was minimal. Industrial expansion and
population growth resulted in severe pollution in urban
areas. The major industrial chemicals manufactured during
the period from 1850 to 1900 were caustic soda (NaOH),
chlorine (Cl2), soda ash (Na2CO3), fertilizers, ammonia,
acids (H2SO4, HCl, HNO), refined petroleum products,
soaps, steel and refined metals, paint, pulp and paper, and
coal gas. Typical waste streams from these activities
included heavy metals (including mercury and lead),
polycyclic aromatic compounds, waste acids, pulp and
paper liquors, solvents from petroleum and wood distilla-
tion, and aquatic nutrients such as nitrates and phosphates.
Waste treatment techniques were rarely used during this
period except for the recovery of valuable by-products, and
waste streams were disposed of in the most convenient
manner available. The first environmental law enacted in
the U.S. was the Refuse Act of 1899 for the purpose of
preventing impediments to navigation!
Chemical Engineering Education

1900-1930 From 1900 to 1930, industry focused on
maximizing production. Factories became centralized in
locations best suited for the particular type of product
being made (e.g., Pittsburgh became the "Steel City").
Emerging technologies during this period included
synthetic rubber, polymers and plastics, and metal finishing
(electroplating). Typical waste streams from these new
endeavors included aqueous heavy metals and metallic
sludge, cyanide compounds, "off-spec" materials and
products, and petrochemical by-products. Waste "treat-
ment" during this period consisted of dumping solid waste
into landfills and liquid waste
into on-site ponds or lagoons.
Urban population and pollu- TA
tion continued to grow at Current Environmental
exponential rates, primarily
due to spectacular increases in Toxic Substances Control
immigration and rural-to- Clean Water Act
urban migration. Pollution Hazardous Materials Tran
problems were regarded as Emergency Planning and
nuisances, but no comprehen- Federal Food, Drug, and C
sive environmental reforms Pollution Prevention Act
were made concerning Poison Prevention Packagi
industrial discharge. Manne Protection. Resear

1930-1950 The period from CleanAir Act
1930 to 1950 saw many Resource Consen nation an
advances in chemical technol- Comprehensive Environm
ogy. Rapid-growth industries Liabilities Act
included pharmaceuticals, Federal Insecticide, Fungi
synthetic organic petrochemi- Consumer Product Safety
cals (pesticides, chlorinated Federal Hazardous Substai
compounds, polymers), Ports and Waterways Safe
detergents, cosmetics, and
coatings. Breakthroughs in one

area often provided abundant and inexpensive raw-
material feedstock for many other operations. For these
reasons, the variety, volumes, and toxicity of industrial
waste streams were rapidly on the rise. Disposal was
preferred to treatment because of economics and a lack of
federal regulations. Disposal methods included the use of
sludge ponds, dilution ("the solution to pollution "), deep-
well injection, and ocean dumping. Most government
regulations at the time were concerned with product quality
and safety rather than environmental issues (Federal Food,
Drug and Cosmetic Act of 1938, Federal Insecticide,
Fungicide, and Rodenticide Act (FIFRA) of 1947).

The first federal law dealing with conventionalforms of
water pollution (the Federal Water Pollution Control Act)
was passed in 1948. It authorized funding for municipal
sewage treatment plants and established broad authority to
regulate industrial and municipal discharges through the
National Pollution Discharge Elimination System
(NPDES). States were given the job of issuing permits and
Spring 1999

enforcing compliance of this law, which dealt mainly with
blatant and acutely toxic industrial discharges.
In the 1950s, an exponential increase in manufacturing
occurred as a result of the healthy economy, a rapidly
growing population, and product-hungry consumers.
During this time, the prevailing government attitude was
"If it's good for business, it's good for America," the
prevailing industrial attitude was "Profits are high,
regulations are few, and life is good!", and the prevailing
consumer attitude was "Buy it, use it once, throw it away."

1950-Present In the
1960s, U.S. industry contin-
ued to grow and prosper.
Crude oil was practically
free, so Americans were
combusting it and polymeriz-
ing it as fast as they could.
Environmental laws were
few, and most discharges
were unregulated. But the
effect of pollution on the
environment was becoming
obvious-major rivers were
catching fire, smog was
causing significant health
problems in many urban
areas, fish were floating
instead of swimming, and
wildlife was disappearing.

These and other problems
resulted in public outrage

Sand the demandforfar-
reaching environmental
legislation. The United States Environmental Protection
Agency (USEPA) was created by presidential order in
1970. It became the lead agency for the control of pollution
of the nation's air, water, and land resources (a job
previously relegated to the U.S. Department of Health
Services and the Federal Water Pollution Control Adminis-
tration). The USEPA was given authority to regulate the
industrial community.
During the late 60s and throughout the 70s, many new
environmental laws were enacted, including the Clean Air
Act, Clean Water Act, Toxic Substances Control Act, and
Resource Conservation and Recovery Act. Significant
national policy changes such as removing lead from
gasoline, removing phosphates from detergents, and
banning a number of dangerous pesticides were also made
during this time. Current laws affecting U.S. Industry are
listed in Table 2.
In 1980, a major piece of legislation, called the "Com-
prehensive Environmental Response, Compensation, and

Laws Affecting U.S. Industry


portation Act
community Right-To-Know Act
cosmetic Act

ng Act
ch. and Sanctuaries Act

d Recovery Act
entalResponse, Compensation, and

:ide, and Rodenticide Act
ices Act

ty L Act

Liabilities Act, was passed. It authorized the federal
government to respond to spills and other releases of
hazardous substances, established a fund for cleanup, and
established industrial cost liability.
Additional environmental issues
that reached public awareness in the
80s include airborne acids, heavy
metals, oxidants, acid rain, strato-
spheric ozone depletion, global
warming, the disappearance of
municipal and hazardous waste land
fills, ocean dumping, and other
illegally dumped hazardous waste.
As a result of these and other
problems, industries found themselves .
faced with lawsuits, protests, opposi- ,
tion to new plant sites, disputes over Students at wo
product approval, and more environ-
mental laws.
Today, every aspect of doing business
is affected by environmental concerns.
Industrial expenses associated with
environmental protection include the
purchase and transport of regulated
materials, consulting fees, air and water
discharge permit fees, and administrative
costs for permitting, reporting, monitor-
ing, sampling, manifesting, and labeling.
Additional costs to industry are incurred
in the purchase of pollution-control
equipment, taxes and insurance, penalties
and fines, lawsuits, off-site transporta-
tion, treatment and disposal, on-site
treatment and control, contaminated site
cleanup, customer dissatisfaction with :- .
poor environmental policy, and long- Students dur
term liability for waste disposal.

During the "long version" it is still best to have two
production floors for each company (one for each round) set
up before the students arrive. There is sufficient time to let
the companies cut their own squares. The production runs of
each round are now limited to 15 to 20 minutes.
Examples of Green Square "Innovations"
Typically in round one, the companies will mix paints in
the extra cups provided, paint several squares, and paint the
production floor while painting the edges of the square.
Often, the production floor is used as a mixing area. Some
groups may assign functions and work as teams, while oth-
ers may produce squares as individuals.
In round two, expect more up-front discussion and more

teamwork. Most of the innovation occurs in round two.
Mixing typically occurs directly on the square, and the paint-
brush may be the only tool. Holding the square so the edges
can be painted without contaminat-
ing other surfaces or hands presents a
challenge. Curiosity as to how the
other companies manufacture their
squares is also greater in round two
("industrial espionage").

rk during round one.

ing r

Often the students will use other
pieces of equipment. The fan on the
viewgraph projector has been used to
dry the green squares. Small contain-
ers such as lipstick lids have been
used for mixing, and tissues are often
used to wipe hands and spills. Occa-
sionally these "outside" items have
been used and then hidden from the
instructor (illegally dumped hazard-
ous waste!").

The Green Squares Manufacturing game
is an excellent hands-on way to introduce
the concept of pollution prevention to under-
graduate engineering and secondary school
students. The game demonstrates both tech-
nical and non-technical challenges of reduc-
ing waste in an industrial setting and makes
participants more aware of sources of waste,
S options for waste minimization, resistance
to change, and the importance of commu-
nication and cooperation in waste-reduc-
tion efforts.

found two. 1. Waste Reduction Institute for Training and
Applications Research, Inc. (WRITAR). WRITAR
is a private, nonprofit organization designed to
identify waste reduction problems, help find their
solutions, and facilitate the dissemination of this informa-
tion to a variety of public and private organizations. Terry
Foecke or Al Innes, Waste Reduction Institute for Training
and Applications Research, Inc., 1313 5th Street, SE, Min-
neapolis, MN 55414-4502
2. Graedel, T.E., and B.R. Allenby, Industrial Ecology, Prentice
Hall, Englewood Cliffs, New Jersey (1995)
3. Markham, A., A Brief History of Pollution, St. Martin's
Press, New York (1994)
4. Melosi, M., editor, Pollution and Reform in American Cities,
1870-1930, University of Texas Press, Texas (1980)
5. Douglas, D., and J. Stewart, eds., The Vanishing Land-
scape: A Collection of Critical Essays on Pollution and Envi-
ronmental Control, National Textbook Company, Illinois
6. Degler, S., Federal Pollution Control Programs: Water, Air,
and Solid Wastes, The Bureau of National Affairs, Inc.,
Washington, DC (1971) 0
Chemical Engineering Education

Internationalizing Practical Chemical Engineering Education

Continued from page 165.
organized for the students each weekend. A number of M.C.C.
staff members often joined these activities, providing further
opportunities for cultural interactions. Examples include a day
trip to Hiroshima to tour the Peace Park and A-Bomb Museum,
plus Miyajima Island, location of one of the "three best views
in Japan"; trips to the beach; playing in an M.C.C. badminton
tournament; and joining the traditional Bon Dance Festival.

The projects selected at the inaugural M.C.C. station were
all ambitious and important to developments underway at
the time. On the first day the students attended the plant,
they were introduced to M.C.C. and the Mizushima Plant,
including the personnel with whom they would interact, and
to company regulations. They also received training in plant
safety. Each group was then presented with a several-page
problem statement, prepared by the Station Directors, de-
scribing the background of their allocated project. The spe-
cific project aims were stated and a suggested method of
approach was provided. There were elements in each project,
however, that were quite open-ended and required the stu-
dents to develop their own plan of attack in consultation with
their sponsors and the Station Directors.
Practice School projects are generally diverse and may
include optimization of an operating plant, research, or de-
sign, or a combination of these. Two examples of projects
undertaken and the work executed by the students are high-
lighted in the sidebar box, within the bounds of company
confidentiality. One student observed that the "projects
were very technical in nature, and our sponsors have
given us a great deal of trust in executing them. I feel that
we actually made a contributiony" Another student noted
that "we were given real problems and were expected to
give real solutions."

The benefits of the experience of successfully completing
a semester at Practice School stations were as strong as ever
for the students who attended the M.C.C. station. The skills
they gained included: problem solving in an industrial con-
text; project planning and management under tight dead-
lines; application of integrated chemical engineering skills
and engineering judgment; strength in written and oral com-
munication; and enhanced teamwork and leadership abilities.
In addition, they gained an intangible but important in-
sight into the operation of a major Japanese corporation
from within. For both the students and the M.C.C. staff with
whom they worked, the opportunity to form international net-
works and build understanding of their differences and simi-
larities in culture, language, and business practices will be
Spring 1999

extremely valuable in their future careers.
There were some disadvantages that we were not able to
overcome. In particular, the language barrier limited the
M.C.C. staff members with whom the students could com-
municate. The project sponsors were very capable and en-
thusiastic to communicate in English, but many of the opera-
tors and technical staff were not able to interact with the
students. This closed off a potential source of process infor-
mation and kept the students from attempting the some-
times-difficult task of forging good working relationships
with operators.
Factors such as this make the combination of experience at
one U.S. station and one overseas station ideal. While these
problems can be alleviated to some extent by considering
language skills during the student selection process, it is
unlikely that all students assigned to an international sta-
tion will have a working knowledge of the language
spoken at that facility. We will continue to provide such
students with basic instruction in the language and cul-
ture of the host country.

The Practice School experience at M.C.C. in Japan dem-
onstrated that operating an international industrial training
station as part of a practical engineering education program
can be highly successful. In spite of initial concerns over
language and cultural barriers, the students' performance in
the overseas environment was of the high standard expected
in the Practice School program, and they also benefited
greatly from exposure to and interactions within a major
Japanese company. Despite the high intensity of the pro-
gram and the stress of living in a new environment, the first
students at the M.C.C. Practice School station all recognized
the advantages of their overseas experience and were pleased
to have had the double opportunity of Practice School train-
ing plus experience in industry overseas.
The Practice School intends to maintain a presence over-
seas; last year international industrial stations operated at Rhone
Poulenc (France), Bayer (Germany), and again at M.C.C.'s
Mizushima Plant. While it is important that the international
stations never replace the U.S. stations completely, they make
an excellent complement to students' experiences in the U.S.

1. Mattill, J.I., The Flagship: The M.I.T. School of Chemical
Engineering Practice, 1916-1991, David H. Koch School of
Chemical Engineering Practice, Massachusetts Institute of
Technology, Cambridge, MA (1991)
2. Johnston, B.S., T.A. Meadowcroft, A.J. Franz, and T.A.
Hatton, "The M.I.T. Practice School: Intensive Practical
Education in Chemical Engineering," Chem. Eng. Ed., 28(1),
38(1994) 0

SM= laboratory




Department of Physical Chemistry Universidad de Alicante Alicante, Spain

During the last decade, there has been increasing
social preoccupation in industrialized countries with
respect to environmental protection, resulting in pro-
gressively tougher legislation regarding waste deposits. The
presence of heavy metals in wastewater constitutes one of
the most important problems in environmental engineering
today,"' fundamentally due to their high toxicity and cumu-
lative character. Contamination by heavy metals is princi-
pally a problem characteristic of industrial effluents, and the
activities that generate dumps of this kind of contaminated
material are both numerous and diverse: metallurgical pro-
cesses, industries involving metal plating, pigmentation and
dyes, and producers of cellulose acetate, accumulators and
batteries, printed circuits, etc. Given that the great majority
of metallic ions can be electrodeposited in a metallic form on
a cathode, electrochemistry offers a way of treating almost
all of these types of wastewater.'1-4
On the other hand, when treating effluents it is normal to
work with low concentrations of heavy metals in solution
(less that 1000 ppm). When two-dimensional electrodes are
used as cathodes, the low concentration originates transport
problems of these ions to the cathode at high current densi-
ties. This fact makes it necessary to design electrochemical
reactors capable of treating these types of effluents in an
efficient way; that is to say, obtaining an almost total recu-
peration of the metal.
One very interesting option concerns the use of three-
dimensional electrodes.'57]1 The principal characteristic of
this type of electrode is that when it extends to three dimen-
sions, it has a high active area on which the electrochemical
reaction can take place; in our case, depositing of the metal-
lic ion. The direct consequence of the high active area that
these electrodes have is a decrease in the real current density
when the deposit reaction takes place, even when working at

high current intensities. This minimizes the problem of trans-
port of the reagent to the electrode and permits almost total
elimination of the metallic ions in the effluent to be treated.
The laboratory experiment described in this paper was
developed for advanced students in chemical engineering or
chemistry. It was designed for groups of three students each
to perform during two periods of four hours each. The stu-
dents must present a full report on the experiment at its end,
including a description of the experiment's objective, the
experimental plan, a description of the experimental system,
a brief summary of the theory behind the experiment, pre-
sentation and treatment of experimental data, discussion of
the results, and finally, any suggestions that might improve
the experiment and a discussion of the sources of error.
The main objective of the experiment is to demonstrate an
Eduardo Exposito received his BS in 1993 and his MS in 1994 from
Cdrdoba University and is working on his PhD at the University of Alicante.
His academic research involves wastewater treatment and organic
Marina Ingles received his BS in 1996 and his MS in 1998 from Alicante
University. His academic research involves organic electrosynthesis.
Jesus Iniesta is a doctoral student at the University of Alicante. He
received his MS in electrochemistry from the same university and has
done work in the electrochemical treatment of hazardous organic com-
Jose Gonzalez-Garcia received his degree in chemistry in 1990 and his
PhD in 1998 from the University of Alicante, where he is currently a faculty
member. His academic interests are in electrochemical engineering and
applied electrochemistry.
Pedro Bonete received his degree in chemistry in 1990 and his PhD in
1995 from the University of Alicante, where he is currently a faculty
member. His academic research involves organic and electroorganic
Vicente Garcia-Garcia received his PhD in chemistry in 1991. His re-
search interest is in applied electrochemistry. He has worked with labora-
tory and pilot plant experiments.
Vicente Montiel Leguey is Associate Professor in the Department of
Physical Chemistry at the University of Alicante. His research interests are
in electroorganic synthesis and wastewater treatment by electrochemical
Copyright ChE Division of ASEE 1999
Chemical Engineering Education

electrochemical application for solution of the very real prob-
lem of removing heavy metals in effluents. In addition, and
following a predominantly applied method, the economic
parameters of the experiment are calculated. At the end of
the experiment, the students must understand and be familiar

Figure 1. (A) Filter-press reactor scheme: 1. Back plates; 2.
Insulator sheet; 3. DSA-O2; 4. Sealing sheet; 5. Solution
frame; 6. Anionic-exchange membrane; 7. Carbon felt; 8.
Graphite plate.
(B) Scheme of the filter-press reactor dimensions.

Figure 2. Scheme of the experimental system: 1. Anolyte
tank; 2. Catholyte tank; 3. Heat exchanger; 4. Pumps; 5.
Flowmeter; 6. Bypass; 7. System for gas measuring; 8.
Filterpress reactor; 9. Coulombimeter; 10. Voltmeter; 11.
Ammeter; 12. Current feeder.
Spring 1999

with the different electrochemical processes (oxidation and
reduction reactions, Faraday's laws, etc.), parameters, and
magnitudes (current density, cell voltage, etc.).
The experiment concerns elimination via electrochemical
treatment of the cation Cu' in a synthetic effluent, using a
three-dimensional electrode as the cathode. Later, treatment
of the effluent was carried out using a two-dimensional
electrode and comparing the results obtained in both ex-
periments. Finally, the economic parameters of the pro-
cess (current efficiency, energy consumption, etc.) were
calculated for the experiment carried out with the three-
dimensional cathode.

Description of the Experimental Assembly
The electrochemical reactor used was a filter-press-type
reactor (see Figure 1) with separate anodic and cathodic
compartments. Reactor dimensions were: length (L), 9 cm;
width (B), 7 cm; height (S), 1 cm. As can be seen in the
figure, this kind of reactor has a sandwich-type structure
where the electrodes are placed at the reactor extremes. Each
compartment consists of a flow distributor (made of polypro-
pylene) where the solution flows parallel to the electronic
surfaces. The anodic and cathodic compartments are sepa-
rated by a membrane. The figure shows how all the de-
scribed elements are separated by sealing sheets to prevent
the escape of solution.
The cathodes are a carbon felt (supplied by Carbone
Lorraine) with an active electrode area per unit volume of
221 7 cm2/cm3 as the three-dimensional cathode and a
graphite plate as the two-dimensional cathode. In the experi-
ment with the three-dimensional electrode, the graphite plate
is used as a current collector. As the anode, a DSA-O2
electrode (Dimensionally Stable Anode for oxygen evolu-
tion) supplied by METAKEM (Usingen, Germany) was used
in both experiments. The separator was a SIBRON 3457
anionic exchange membrane supplied by lonac Chemical
Company (New Jersey). The whole structure is placed be-
tween two steel plates where it is pressed to avoid solution
escape. As shown in the figure, the only difference between
the configuration of the reactor employed in the two experi-
ments is the introduction of the three-dimensional electrode.
Figure 2 shows a simplified diagram of the experimental
system. It includes a filter-press cell, electrolyte tanks, and
magnetically driven pumps. The system permits control and
measurement of the temperature and the catholyte flow by
means of two heat exchangers and a flowmeter, respectively.
The gases generated over the cathode were collected in an
inverted burette to measure their volume. To prevent gas
escape, the cathodic branch of the system was hermetically
sealed. This branch also had a bypass to secure complete
homogenization of the catholyte solution. The necessary
electrical instrumentation consisted of a 3A-30V DC power



supply, two multimeters to measure the intensity flowing
and the cell voltage, and a coulombimeter with a 0.1-1A
shunt to measure the charge passed.
Analysis of the Cu2+ concentration in the samples was
made using the ICP (Inductive Coupled Plasma) with an ICP
Perkin-Elmer Optima 3000. Since most undergraduates do
not have an ICP readily available, they could also analyze
the Cu2+ concentration with the colorimetric method of the

Carrying out the Experiments
The experimental conditions under which the two experi-
ments were carried out are shown in Table 1. Prior to each
practice session, the professors must prepare the anolyte and
catholyte solutions and register the calibration curve of the
ICP analyzer in order to facilitate analysis of the samples
that will be taken by the students. The Cu2" calibration curve
is linear over the entire concentration range of Cu2* (0 1000
ppm Cu2+), and it is not necessary to dilute the samples taken
during the experiments.
First Session In this session, the students must carry out
the elimination using a three-dimensional cathode. Before
the experiment starts, it is helpful for the professor to
show the students a disabled filter-press reactor so they
can better comprehend the structure and method of op-
eration of a filter- press reactor.
First, a sample of 1 ml of catholyte must be taken before
its introduction into the system in order to know the exact
initial concentration of the Cu2*; this sample is labeled
"sample 0." After that, the anodic and cathodic branches
must be filled and washed with distilled water. Then the
system is emptied. The catholyte and anolyte solutions can
then be introduced into the corresponding deposits and the
pumps connected; the catholyte flow is adjusted to the re-
quired value, and after a few minutes the system reaches the
working temperature of 300C. At that moment, a 1-ml vol-
ume sample of the catholyte solution is extracted and labeled
"sample 1." After that, current (0.63A) is made to flow
through the system.
The experiment is carried out for approximately one hour.
Every five minutes, 1-ml-volume samples are
taken until the end of the experiment at 2500 C of
passed charge (this charge is rather more than the
150% charge necessary to deposit the 0.5g Cu2 E
initially present in the catholyte, thus assuming a Operat
100% current efficiency in the copper-deposit
reaction). For each sample, note is taken of the
values of time, cell voltage, quantity of charge Anolyt
circulated, and the volume of gas originated onto Temr
the cathode. At the end of the experiment, the Curren
volumes of the anolyte and catholyte are mea-
sured and a sample of the anolyte solution is
taken to check the presence of Cu2+ that may Total Elec

have passed through the membrane separator. After that,
the system is washed several times with distilled water.
The electrochemical reactor must be filled with water
until the next session.
Hydrogen is a flammable gas. Although the volume of
H2 generated during the experiments is small, care is
necessary and the burette where the H2 is collected must
not be exposed to heat.
During the session, the Cu2+ concentration of the samples
must be measured. The catholyte solution initially has a
light-blue color due to the presence of ions [Cu(H20)]2+] in
solution. As copper is deposited on the cathode, the solution
gradually loses its blue color until it is completely colorless
at the end of the experiment.
Second Session Prior to the second session the profes-
sors must remove the three-dimensional electrode from the
reactor and prepare the system. During this session, the
students must carry out the elimination of Cu2" using a two-
dimensional cathode. The experiment is carried out by using
the same procedure as was used in the first session.

1. The most important charge-transfer processes that take
place inside the electrochemical reactor can be seen in
Figure 3.
2. At this point, it is possible to observe the differences
found between using a three-dimensional and a two-
dimensional cathode to remove Cu2+. Figure 4 shows
Cu2+ concentrations in the catholyte vs time. When the
three-dimensional carbon felt cathode was used, the final
Cu2+ concentration in the catholyte was less than 1 ppm.
On the other hand, when the two-dimensional graphite
cathode was used, the final Cu2+ concentration in the
catholyte was approximately 50% of the initial concen-
Table 2 shows the values of time, concentration of Cu2+ in
the catholyte, volume of H2 generated, charge passed, and
the average cell voltage at different times of electrolysis.
The results up to now show how well a three-dimensional
electrode behaves in the elimination of heavy metal ions

experimental Conditions of the Cu2+ Removal Experiments

ion Mode Galvanostatic
e (V:0.51) 8 x 10'M CuSO, (1g/1 Cu*) + 0.5M NaSO4 + 0.05M H SO0
e(V:0.51) 0.5M NaZSO
perature 30C
tDensity 10 mA/cm2 (current: 0.63 A)
yte Flow 50 1/h
trical Charge 2500C
Chemical Engineering Education

in solution versus the use of conventional two-dimen-
sional electrodes in which the reaction of the formation
of H2 in the cathode is very important and the Cu2,
concentration decreases very slowly, as explained at the
beginning of this article.
Table 3 shows the charge balance with respect to the
electrodeposited Cu and generated H2 at different times of
electrolysis. Moreover, it is interesting in the experiment
with the three-dimensional cathode to do the calculations
at the point where the copper is eliminated, at approxi-
mately 45 minutes.
From the difference in Cu2" concentration between sample
0 and sample 1, we may calculate the real volume of catholyte.
It is interesting to note that the initial Cu2C concentration in
the experiment with the two-dimensional cathode is higher
than the initial Cu2C concentration measured in the experi-
ment with the three-dimensional cathode. This fact can be
explained because the carbon felt used as a three-dimen-
sional electrode has a high porosity and retains a high vol-
ume of distilled water in the preliminary washing stage.
The principal error during the calculation of the charge
balance is caused by the charge employed in the reduction of
the 02 present in the catholyte, which is not experimentally
measured by the students. But this error is not too high due

2H~o 21O 4-

I- so 4
0o, + 4 + 4e c + -2e

Anode Exchange Cathode

Figure 3. Charge transport processes inside the
filter-press reactor.

1000 ----------

S750 7 c

: 0
+ Soo I C I

S 250 0 0
i *
0 I
0 ---------------- ---
0 10 20 30 40 50 60
Time (min)

Figure 4. Representation of Cu2+ concentration (ppm) vs.
time (min) during the experiments.
carbon felt; 0 graphite plate
Spring 1999

to the low concentration of O, in solution, and therefore it
can be disregarded. Other optional methods are: give this
value to the students, or before the experiment eliminate the
02 in solution by bubbling the catholyte with N2 for approxi-
mately 45 minutes, increasing the duration of the practice
session in this way.
One of the most important experimental errors appears in
the charge balance if the cathodic system is not hermetically
closed-the measure of volume of generated H2 will be
incorrect. Close attention must be paid to this, especially
during the sampling, to avoid any type of gas escape.
On the other hand, in the analyses of the anolyte samples
taken at the end of the experiments, the presence of Cu2" ions
was not detected. This fact indicates that during the experi-
ments, these ions did not pass through the membrane separa-
tor. This is logical due to the short duration of the experi-
ments and the use of an anionic-exchange membrane.
Finally, the calculations corresponding to the economic
parameters of the process for the experiment using the three-

Parameters Measured in the Cu2* Removal Experiments

Time (min) Charge passed Cu2* Volume of VM,(V)
(C) concentration (ppm) H, (ml)

Three-Dimensional Cathode
sample 0 -995
sample 1 -873
20 760 485 0 2.06
40 1520 81 11 2.16
45 1750 5.2 24 2.39
60 2280 1.3 78 2.42

Two-Dimensional Cathode
sample 0 -1010
sample 1 934
20 760 735 40 3.22
40 1520 511 87 3.27
60 2280 405 158 3.32

Charge Balances of the Cu2+ Removal Experiments

Time (min) Charge passed Charge Used in Charge Used in
(C) Cu2 reduction (C) H2 generation (C)

Three-Dimensional Cathode
20 760 683
40 1520 1380 87
45 1750 1510 190
60 2280 1515 620

Two-Dimensional Cathode
20 760 343 317
40 1520 716 690
60 2280 892 1254

dimensional electrode are shown in Table
4. The following expressions are used to Current Effi
calculate the characteristic economic pa- sumption of th
rameters of this electrochemical process. ment with a Th
Current Efficiency This parameter re-
lates the total charge passed with the charge Time (min) CI
used in depositing copper.
Current Efficiency Cu(CE%) =(Charge used 20
in depositing copper/Charge passed) x 100 45
Energy Consumption This is the energy 60
necessary to deposit a certain amount of
copper. It is normally expressed in kilowatt-hour (kWh) per
kilogram (kg) of product obtained.

1kW lh
kWh = Vcell(V) x I(A) x T(s)x IW x =
103W 3600s
IkW 1 h
Vcell(V) x Q(C)x x 3600 x

Vel, changes along the experiment. The correct expression
of this parameter is
Vcell = V(t)dt
Nevertheless, the variation of V during the experiment is not
very important, and to simplify the calculation of kWh, an
average value can be used.

S 1 kg
kgcu = number of deposited moles of Cu x AtwtCu(g)x 10k
10 g
b CE% x 10-2 1 kg
Q(C)x x x AtwtCu(g)x 103g
n F 103g

The expression for energy consumption is
kWh 2680.55 x Vcell
kgcu AtwtCu x CE% x b
where Q is the charge passed, I is the current, T is the time of
electrolysis, Vcen is the average value of the electrochemical
reactor voltage, AtwtCu is the atomic weight of copper, b is
the stoichiometric coefficient of metallic copper in the de-
posit reaction (in our case 1, Figure 3), and n is the number
of electrons exchanged in the reaction (in our case 2).
It can be seen in Table 4 that the two calculated parameters
are quite constant until 45 minutes. By this time, all the
copper is deposited over the cathode. At that time, the for-
mation of H, becomes important and the economics param-
eters become worse.

1. First, the students must give a description of the
experimental system and a brief summary of the theory
behind the practice.

e Cu



2. The students must have a clear
LE 4
and Energy Con- comprehension of the charge-transfer
f and Energy Con-
2+ Removal Experi- processes that take place inside the
dimensional Cathode filter-press reactor. In this way they
must sketch a scheme similar to
Energy consumption Figure 3. At this point, the professor
% (kWh/kgc.) can ask them some additional
2.00 questions such as "Why do we use
2.03 an anionic-type exchange membrane
2.34 as the separator?" and "What will
3.00 happen if we use a cationic-exchange
3. The next step is to compare and explain the differences
found between using a three-dimensional cathode and a
two-dimensional cathode with respect to the recovery
of copper. The students must represent Figure 4 and do
the charge balance using the measured experimental
data in the same way as mentioned previously. At this
point, the question "What will happen if we bubble N2
inside the catholyte solution?" can help the students
detect the influence of the dissolved O,.
4. Then the students should calculate the economic
parameters of the experiment.
5. The last point is a critical evaluation of the practice and
discussion of the sources of error.

Student reaction to the experiment has been very satisfac-
tory. The main importance of the practice session is that it
tackles the real problem of treating effluents containing heavy
metals. It also achieves the goals mentioned in the first
paragraphs of this article. The students assimilated the basic
theoretical concepts of electrochemistry (charge balances,
charge and mass transport processes, Faraday's laws, etc.)
and they familiarized themselves with the use of the instru-
ments used in applied electrochemistry (power supplies,
multimeters, coulombimeters, etc.).

1. Miiller, K.J., and G. Kreysa, Dechema Monographien, 98,
2. GonzAlez-Garcia, J., J.R. Perez, G. Codina, V. Montiel, and
A. Aldaz, Ingenieria Quimica, 159, 322 (1997)
3. Electrochemistry for a Cleaner Environment, Electrosynthesis
Co., East Amherst, NY (1992)
4. Environmental Oriented Electrochemistry, C.A.C. Sequeira,
ed., Elsevier, Amsterdam (1995)
5. DeLevie, R., in Advances in Electrochemistry and Electro-
chemical Engineering, Vol. 6, P. Delahay, ed., J. Wiley &
Sons (1967)
6. Newman, J.S., and W. Tiedeman, in Advances in Electro-
chemistry and Electrochemical Engineering, Vol. 11, H.
Gerisher, ed., J. Wiley & Sons (1978)
7. Marchiano, S.L., and A.J. Arvia, Chap 7, pg. 145 in
Electrochemicals Reactors: Their Science and Technology.
Part A., M.J. Ismail, ed, Elsevier (1989)
8. Nakano, S., Yakugaku Zasshi, 82, 486 (1962) 0
Chemical Engineering Education


This guide is offered to aid authors in preparing manuscripts for Chemical Engineering Education (CEE), a quarterly journal
published by the Chemical Engineering Division of the American Society for Engineering Education (ASEE).
CEE publishes papers in the broad field of chemical engineering education. Papers generally describe a course, a laboratory, a
ChE department, a ChE educator, a ChE curriculum, research program, machine computation, special instructional programs, or
give views and opinions on various topics of interest to the profession.

Specific suggestions on preparing papers *
TITLE Use specific and informative titles. They should be as brief as possible, consistent with the need for defining the
subject area covered by the paper.

AUTHORSHIP Be consistent in authorship designation. Use first name, second initial, and surname. Give complete mailing
address of place where work was conducted. If current address is different, include it in a footnote on title page.

ABSTRACT: KEY WORDS Include an abstract of less than seventy-five words and a list (5 or less) of keywords

TEXT We request that manuscripts not exceed twelve double-spaced typewritten pages in length. Longer manuscripts may
be returned to the authors) for revision/shortening before being reviewed. Assume your reader is not a novice in the field.
Include only as much history as is needed to provide background for the particular material covered in your paper. Sectionalize
the article and insert brief appropriate headings.

TABLES Avoid tables and graphs which involve duplication or superfluous data. If you can use a graph, do not include a
table. If the reader needs the table, omit the graph. Substitute a few typical results for lengthy tables when practical. Avoid
computer printouts.

NOMENCLATURE Follow nomenclature style of Chemical Abstracts; avoid trivial names. If trade names are used, define
at point of first use. Trade names should carry an initial capital only, with no accompanying footnote. Use consistent units of
measurement and give dimensions for all terms. Write all equations and formulas clearly, and number important equations

ACKNOWLEDGMENT Include in acknowledgment only such credits as are essential.

LITERATURE CITED References should be numbered and listed on a separate sheet in the order occurring in the text.

COPY REQUIREMENTS Send two legible copies of the typed (double-spaced) manuscript on standard letter-size paper.
Submit original drawings (or clear prints) of graphs and diagrams on separate sheets of paper, and include clear glossy prints of
any photographs that will be used. Choose graph papers with blue cross-sectional lines; other colors interfere with good
reproduction. Label ordinates and abscissas of graphs along the axes and outside the graph proper. Figure captions and legends
will be set in type and need not be lettered on the drawings. Number all illustrations consecutively. Supply all captions and
legends typed on a separate page. State in cover letter if drawings or photographs are to be returned. Authors should also include
brief biographical sketches and recent photographs with the manuscript.

Send your manuscript to
Chemical Engineering Education, c/o Chemical Engineering Department
University of Florida, Gainesville, FL 32611-6005


FALL 1999


Deadline is June 1, 1999

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 E2JL1Z78I_5KHNDF INGEST_TIME 2012-02-17T16:27:47Z PACKAGE AA00000383_00142