Chemical engineering education

a a.
Se ed
VOLUE 30NUMBR2 SRING199

Moto M. De

ACKNOWLEDGEMENT

DEPARTMENTAL SPONSORS

The following 159 departments contribute to the support of CEE with bulk subscriptions.

If your department is not a contributor, write to
CHEMICAL ENGINEERING EDUCATION,
c/o Chemical Engineering Department University of Florida Gainesville, FL 32611-6005
for information on bulk subscriptions

University of Akron
University of Alabama
University of Alberta
University of Arizona
Arizona State University
University of Arkansas
Auburn University
Ben Gurion University of the Negev
Brigham Young University
University of British Columbia
Brown University
Bucknell University
University of Calgary
University of California, Berkeley
University of California, Davis
University of California, Irvine
University of California, San Diego
University of California, Santa Barabara
California Institute of Technology
California State Polytechnic Institute
California State University
Carnegie-Mellon University
Case Western Reserve University
University of Cincinnati
Clarkson University
Clemson University
Cleveland State University
University of Colorado
Colorado School of Mines
Colorado State University
Columbia University
University of Connecticut
Cornell University
Dartmouth College
University of Dayton
University of Delaware
Drexel University
University of Edinburgh
University of Florida
Florida Institute of Technology
Florida State/Florida A&M University
Georgia Institute of Technology
University of Houston
Howard University
University of Idaho
University of Illinois, Chicago
University of Illinois, Urbana
Illinois Institute of Technology
University of Iowa
Iowa State University
Johns Hopkins University
University of Kansas
Kansas State University

University of Kentucky
Lafayette College
Lakehead University
Lamar University
Laval University
Lehigh University
Loughborough University
Louisiana State University
Louisiana Technical University
University of Louisville
University of Maine
Manhattan College
University of Maryland
University of Maryland, Baltimore County
University of Massachusetts
University of Massachusetts, Lowell
Massachusetts Institute of Technology
McGill University
McMaster University
McNeese State University
University of Michigan
Michigan State University
Michigan Technical University
University of Minnesota
University of Minnesota, Duluth
University of Mississippi
Mississippi State University
University of Missouri, Columbia
University of Missouri, Rolla
Montana State University
University of Nebraska
University of New Hampshire
University of New Haven
New Jersey Institute of Technology
University of New Mexico
New Mexico State University
North Carolina A & T University
North Carolina State University
University of North Dakota
Northeastern University
Northwestern University
University of Notre Dame
Technical University of Nova Scotia
Ohio State University
Ohio University
University of Oklahoma
Oklahoma State University
Oregon State University
University of Ottawa
University of Pennsylvania
Pennsylvania State University
University of Pittsburgh
Polytechnic Institute of New York

Princeton University
Purdue University
Queen's University
Rensselaer Polytechnic Institute
University of Rhode Island
Rice University
University of Rochester
Rose-Hulman Institute of Technology
Rowan College
Rutgers, The State University
San Jose State University
University of Saskatchewan
University of Sherbrooke
University of South Alabama
University of South Carolina
South Dakota School of Mines
University of South Florida
University of Southern California
University of Southwestern Louisiana
State University of New York, Buffalo
Stevens Institute of Technology
University of Sydney
Syracuse University
Technion-Israel Institute of Technology
University of Tennessee
Tennessee Technological University
University of Texas
Texas A & M University, College Station
Texas Tech University
University of Toledo
Tri-State University
Tufts University
University of Tulsa
Tuskegee Institute
University of Utah
Vanderbilt University
Villanova University
University of Virginia
Virginia Polytechnic Institute
University of Washington
Washington State University
Washington University
University of Waterloo
Wayne State University
West Virginia Graduate College
West Virginia Institute of Technology
West Virginia University
Widener University
University of Wisconsin
Worcester Polytechnic Institute
University of Wyoming
Yale University
Youngstown State University

EDITORIAL AND BUSINESS ADDRESS:
Chemical Engineering Education
Department of Chemical Engineering
University of Florida Gainesville, FL 32611
PHONE and FAX : 352-392-0861
e-mail: cee@che.ufl.edu

EDITOR
T. J. Anderson

ASSOCIATE EDITOR
Phillip C. Wankat

CONSULTING EDITOR
Mack Tyner

MANAGING EDITOR
Carole Yocum

PROBLEM EDITORS
James 0. Wilkes and Mark A. Burns
University of Michigan
LEARNING IN INDUSTRY EDITOR
William J. Koros
University of Texas, Austin

PUBLICATIONS BOARD -
CHAIRMAN *
E. Dendy Sloan, Jr.
Colorado School of Mines

PAST CHAIRMEN *
Gary Poehlein
Georgia Institute of Technology
Klaus Timmerhaus
University of Colorado

MEMBERS *
Anthony T. DiBenedetto
University of Connecticut
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
Angelo J. Perna
New Jersey Institute of Technology
Stanley I Sandier
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

Spring 1996

Chemical Engineering Education

Volume 30

Number 2

Spring 1996

> DEPARTMENT
82 University of Washington, Eric M. Stuve

> EDUCATOR
88 Morton M. Denn, Arup K. Chakraborty,
Arthur B. Metzner, T.W. Fraser Russell

> CLASSROOM
94 The Mass Transfer Boundary Layer with Finite Transfer Rate,
Morton M. Denn
102 Design Competition for Second-Year Students, W. A. Davies
108 Demonstrations to Complement a Course in General Engineering
Thermodynamics, Douglas J. Dudgeon, J. W. Rogers, Jr.

> LABORATORY
98 A Laboratory Experiment that Enhances Environmental Awareness,
Ken K. Robinson, Joshua S. Dranoff
142 Low-Cost Experiments in Mass Transfer: Part 2,
I. Nirdosh, M.H.I. Baird

> COMPUTERS IN EDUCATION
114 CESL: The Chemical Engineering Simulation Laboratory,
David A. Kofke, Marc R. Grosso, Sreenivas Gollapudi, Carl R.F. Lund

> SURVEY
122 Applied Statistics: Are ChE Educators Meeting the Challenge?
Roger E. Eckert
146 Current Trends in Chemical Reaction Engineering Education,
Mazen Shalabi, Muhammad Al-Saleh, Jorge Beltramini, Dulaihan Al-
Harbi

> LEARNING IN INDUSTRY
126 International Engineering Internship Program,
John M. Grandin, Kristen L. Verduchi

> RANDOM THOUGHTS
130 Speaking of Everything, Richard M. Felder

> CLASS AND HOME PROBLEMS
132 Dynamic and Steady-State Behavior of a CSTR, Aziz M. Abu-Khalaf

> CURRICULUM
136 Mathematica in the ChE Curriculum,
John R. Dorgan, J. Thomas McKinnon
150 Chemical Engineering Education in Turkey and the United States, J.
Richard Elliott, Jr.

> ADVISING
156 Undergraduate Academic Advising, Michael L. Mavrovouniotis

> 145 ERRATUM

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 1996 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.

I

r department

UNIVERSITY OF

WASHING TO,

ERIC M. STUVE
University of Washington Seattle, WA 98195-1750

It was too early to exchange pleasantries; the ringing
phone could only mean that something was wrong-at best a
wrong number, at worst, well, who knows? Thus awakened
on Saturday morning in February, 1964, Albert "Les" Babb
quickly ruled out the first possibility. The easily recogniz-
able voice belonged to Dr. Belding Scribner, inventor of the
arteriovenous shunt, and he was obviously distressed.
Scribner related that the "Who-Shall-Live Committee" had
just denied treatment to a 16-year-old high school honor
student and "[her] only hope for survival would be to get
intermittent dialysis therapy within the next four months!"
Babb, a chemical engineering professor at the University
of Washington, knew full well the binding and irrevocable
nature of the committee's decision and its consequences
for the patient, the daughter of a friend of his. He had
just overseen development of the "monster, a multi-
patient dialysis machine in use at the University of Wash-
ington hospital.
After Scribner's call, Babb's next course of action was
clear; he must develop an in-home dialysis machine that
could be used without medical supervision and he must do so
within the next four months. Scribner and Babb quickly
assembled a team of physicians and engineers, and the "Mini-
I, the first in-home portable dialysis machine, made its
lifesaving debut in June of that year. Commercial produc-
tion of dialysis machines based on the Mini-I and its novel
continuous proportioning dialysate system began the follow-
ing year, and by 1969 the system was the predominant
method of dialysis. Today it is the exclusive method.
he portable dialysis machine is one of the many
treasured accomplishments of our Department of
Chemical Engineering. Accredited by the AIChE in
1926 (only the second year in which accreditation was of-
fered) the department has a long history leading up to its
current position as a flourishing institution of research and

teaching with 14 full-time faculty,
10 staff members, 60 graduate
students, and 150 undergraduate
students. I

THE UNIVERSITY OF
WASHINGTON AND
SEATTLE I
The 680-acre University of
Washington (UW) campus blends woodlands, wetlands, land-
scaped verandas, and urban architecture into a setting of
uncommon beauty. The main campus sits on a predomi-
nantly southward-facing, and sometimes steep, slope etched
out by glaciers during the ice age. Two of the many visual
highlights of the campus include the "Quad," with its breath-
taking display of blooming Japanese Cherry trees each spring,
and the Rainier Vista, the main promenade that makes a
virtual connection between the campus and the 14,400-ft
peak of Mt. Rainier a hundred miles to the southeast. Within
this setting, 25,000 undergraduate students, 9,000 graduate
and professional students, and 2,800 faculty carry out the
pursuit of education and scholarship.
The campus is located just four miles northeast of down-
town Seattle, where the University began as the Territorial
University in 1861. It remained there until 1893, when it
moved "as far away from the city as is reasonable" to its
present location.
Seattle has prospered as a center of transportation-water,
rail, air, and even bicycle. In Cry of the Wolf Jack London
describes Seattle's role as the starting point in the Alaskan-
Yukon gold rush of 1898-99. It also became the western
terminus of the Great Northern railroad from Minneapolis-
St. Paul in 1893-under the ownership and control of the
"empire builder," James J. Hill, it was the first transconti-
nental railroad built without government subsidy. Three stat-
ues stand on the UW campus today: one, naturally, is of

Copyright ChE Division ofASEE 1996

Chemical Engineering Education

George Washington; another is of the composer Edward
Grieg; and the third, J.J. Hill.
In 1916, lumberman William E. Boeing began building
seaplanes to improve transportation throughout the vast Pa-
cific Coastal waterways of the Northwest, and late in the
1920s, he formed a regional airline, both as a market for his
planes and to deliver mail across the Cascade Mountains. In
the early 50s it took
"betting the company" _
to build the Boeing 707,
the first successful com- -:-" -.
mercial jet, establishing
Boeing's place as the
preeminent builder of
commercial airliners.
What about the airline?
In 1934 the Roosevelt
administration forced
Boeing to divest itself of
the airline-it is now
United Airlines.
The newest chapter in
transportation involves
the bicycle. A vast net-
work of bicycle trails
threads through Seattle
and neighboring areas,
with the most prominent
being the 17-mile
Burke-Gilman trail,
which runs through the
campus along the right-
of-way of an abandoned Rainier Vista, the main pro
tower of the Administration B
railroad. Chemical engi- Mt. Rainier in the back
neering students, staff,
and faculty all use the
trail for commuting, not just by bicycle but also by skate-
board, roller blades, roller skis, and, of course, by foot!

THE CHEMICAL ENGINEERING DEPARTMENT
Chemical engineering began in 1904 as a discipline within
the chemistry department. Its first BS degree was granted in
1907, and one of its first PhDs went to Waldo L. Semon in
1924 -Semon, now installed in the Inventor's Hall of
Fame, first synthesized polyvinyl chloride; among his
many other inventions, bubble gum remains the favorite
of his grandchildren.
Henry K. Benson began teaching in 1905, specializing in
industrial and physical chemistry and placing special em-
phasis on continuous (as opposed to batch) processing. At
the end of World War I, he became Executive Officer of the
new Department of Chemistry and Chemical Engineering.
Subsequently, Prof. Beuschlein was hired to take on chemi-
Spring 1996

mel
uild
rou

cal engineering matters as Benson's responsibilities were set
mostly by the much larger chemistry program.
Throughout the next two decades, chemical engineer-
ing continued in the hands of Benson and Beuschlein,
and then in 1947 two relative newcomers, R.
Wells Moulton, a PhD graduate of UW, and Joseph L.
McCarthy, a former UW student with a PhD from
McGill, took the

S over Beuschlein's
responsibilities of
running the chem-
ical engineering
program,while
McCarthy worked on
cleaning up the pol-
luted effluents of the
local pulp mills.
In 1953, the Depart-
ment of Chemical En-
gineering became es-
tablished as a separate
department within the
College of Engineer-
ing, but remained
physically with the
Department of Chem-
istry in Bagley Hall.
Moulton became
chairman of the new
department and over-
saw a period of dy-
ade of the campus, with the namic growth. He en-
ling, Drumheller Fountain, and g
nd. (Photo by Davis Freeman) courage Les Babb to
pursue the new field
of nuclear engineering
and brought Babb and Scribner together for their collabora-
tion leading to the Monster and Mini-I dialysis machines.
Chemistry, chemical engineering, and the nuclear engineer-
ing group all expanded rapidly within Bagley Hall, and the
time eventually came when additional space was needed. In
1967 chemical and nuclear engineering (with Babb as its
chair) moved into their new home, Benson Hall-named
after the founder of chemical engineering at UW.
Charles Sleicher became the next chairman in 1977 and
continued the tradition of growth started by Moulton, estab-
lishing a center for surface science in 1983. The department's
investment in new faculty also returned high dividends in
the form of four PYI awards among a faculty of twelve.
Sleicher remained chairman until 1989 when Bruce
Finlayson became only the fourth chairman in the
department's long history.
Benson Hall sits amid a diverse collection of natural and

Above: Flanked by twin 20,000 volt coupling devices,
graduate student Tim Pinkerton adjusts tip alignment in the
field ion microscope. At Right: One of the twenty-five
Japanese Cherry trees in full bloom in the Quad.

architectural beauty. It has 20,000 square feet of research
laboratory space and 5,300 square feet of teaching laborato-
ries, including the laboratories of unit operations, computer
and process control, colloid and surface science, and electro-
chemistry. Supporting facilities include a well-equipped ma-
chine shop, electronics shop, and a computer network of 75
Macintosh-based machines and three Vaxes.

UNDERGRADUATE INSTRUCTION
Over the past twenty years, an average of 62 BS degrees
have been granted yearly by the department. Recent enroll-
ment has been quite stable, with 57 BS degrees granted last
year, of which 40% went to women. Participation of women
in the program has increased from an average of 30% over
the last five years to 48% for the most recently admitted
class. The department maintains an active scholarship pro-
gram, and approximately 20% of each undergraduate class
receives either full or partial support of their education through
chemical engineering scholarships.
Undergraduate students from UW enter the department at
the end of their sophomore year, while students transferring
from one of the community colleges or state-wide four-year
universities enter at the beginning of their junior year. Typi-
cally, about 40% of each class is made up of transfer stu-
dents. The average entering grade-point average is 3.43,
with 30% of students having a GPA of 3.7 or higher.
Once in the department, students take the canonical mass
and energy balance course. Process simulation with Aspen is
an integral part of all required undergraduate courses, and
Babb has the formidable task of introducing sophomores and
juniors to Aspen. Many students have had little experience
with spreadsheet programs and computer communications pro-
tocols, so Babb also undertakes to redress these deficiencies.

From there, the students take traditional courses in trans-
port processes (fluid mechanics, heat transfer, and mass
transfer), reactor design, and process control. Homework
problems and class projects include work in design and
statistics. Students also take two quarters of unit operations
laboratory where they receive their first exposure to inde-
pendent problem formulation, working in groups, taking and
statistically analyzing data, and preparing convincing writ-
ten and oral reports. All experiments are open-ended, with
only minimal guidance from instructors; students are al-
lowed, and even encouraged, to learn from each other.
Process design is covered in two courses in the last two
quarters of the senior year. The first course introduces stu-
dents to engineering economics and gives them a first crack
at a "conceptual design" of an entire process in about three
weeks. This design is done on Aspen and reawakens stu-
dents to the systems approach to engineering design-for
many a rude awakening, indeed. In the capstone course,
students work in groups of four to design an entire chemical
process both at the conceptual stage, where equipment speci-
fications float, and at the application stage, where optimi-
zation proceeds with fixed equipment parameters. Stu-
dents then submit phone-book size reports detailing their
designs and rationale.
A number of elective options are available to students
along the way. The department offers "specialty options" in
which students focus at least 9 credits of their advanced
chemistry and engineering science electives into a particular
Chemical Engineering Education

Over the past twenty years, an average of 62 BS degrees have been granted yearly by the department.
Recent enrollment has been quite stable, with 57 BS degrees granted last year, of which 40% went
to women. Participation of women in the program has increased from an average of 30%
over the last five years to 48% for the most recently admitted class.

Graduate student Theresa Jurgens-Kowal examines
sample alignment in the X-ray photoelectron
spectrometer.
area: biotechnology; polymers, composites, colloids, and
interfaces; electronic materials; computers applied to chemi-
cal engineering; environmental engineering; and nuclear
chemical engineering. About 40% of students perform un-
dergraduate research projects both to prepare for entry into
the job market and for graduate school.
At graduation time, the most important problem looming
for students is finding a job. To help increase their employ-
ment prospects, the department encourages students to take
co-ops during their studies and offers a two-track curriculum
so that, with careful planning, students can take a six month
co-op without lengthening their time to degree. Currently,
about 19% of students take co-ops.
The current employment situation looks good, especially
because of the increase in semiconductor manufacturing in the
Northwest. No single type of job is preeminent; students find
employment in small local companies, waste remediation at
the Hanford Nuclear Reservation, semiconductor manufactur-
ing, and of course, the oil and chemical process industries.

GRADUATE INSTRUCTION AND RESEARCH
The department has about sixty graduate students, all en-
gaged in research activities. During the past five years the
department has graduated an average of 10-11 PhDs each
year, making it fourteenth in annual production of PhD
chemical engineers. Entering graduate students take
courses in applied mathematics, thermodynamics, fluid
mechanics, heat transfer, and reaction kinetics during
their first two quarters.
Students earn entry into the PhD program upon successful
Spring 1996

completion of a Preliminary Exam that embodies the goal of
teaching students to "learn how to learn" by assessing their
abilities in three areas: a knowledge-base of chemical engi-
neering fundamentals, critical and analytical thinking skills,
and research performance itself. Fundamentals of chemical
engineering are probed in an oral examination, and critical
and analytical skills are assessed through a written and
oral critique of a current research paper assigned to the
student by the faculty. Once these two parts have been
successfully completed (at the end of the second quar-
ter), students then choose a research advisor and begin
their thesis research in earnest. About six months later,
students take the last part of the Preliminary Exam, in
which they present and defend their research to date and
field related questions from the faculty.

CHEMICAL ENGINEERING RESEARCH
The nature of research spans the full spectrum from mo-
lecular-level fundamental research to applied research for
end-use products. Some major research successes of the
department include the home kidney dialysis machine (Babb)
mentioned at the beginning of this article, textbooks on
numerical mathematics for chemical engineering (Finlayson),
polymer composites for airplane manufacture (Seferis), de-
sign of a Seattle Metro sewer project (1993) that has saved
local taxpayers about $10 million (Ricker), and development
of a biocompatible, plasma deposited polymer (Ratner).
Research in the department has grown according to a
strategy of developing strength in a number of critical re-
search areas. Today, the department offers a variety of re-
search programs in surface science, biotechnology, environ-
mental studies, computers and process control, and materials
(polymers and thin films). These five topics represent the
subject matter of eighteen research groups, yet research meth-
odologies can be broadly categorized (with a few excep-
tions) into three groups. Seven research groups engage in
some form of surface science, six in biotechnology, and four
in computer methods. Perhaps the most distinctive exception
is the work on industrial teaming, which has been imple-
mented in the Boeing 777 program.
The relatively large proportion of surface science work is
perhaps surprising for a chemical engineering department.
John Berg began the department's modern surface sci-
ence era by studying liquid interfacial and colloidal phe-
nomena. Today, Berg offers an industrial short course on
surface and colloidal science with about 45 participants
each year. Berg's current research focuses on improved

85

absorbents and de-inking of papers.
In 1983 a $500,000 grant from the Shell Com-
panies Foundation established a state-of-the-art
surface analysis center based on X-ray photo-
electron spectroscopy. Buddy Ratner combined
this center with operating funds from the NIH
and began to study surfaces of biological inter-
est, especially the biocompatibility of polymers
used in implants. Shortly thereafter, Eric Stuve
started his program of electrochemical surface
science that incorporates both solid/liquid and
solid/gas surface science, including a system
that does both. This work focuses on funda-
mentals of electrochemistry and mechanisms
of fuel cell reactions.
The growth of both "wet" and "dry" surface
science has been steady throughout the last ten
years. In the former category, Dan Schwartz
has established an electrochemical engineering
program that, among other projects, has devel- Biker
oped a method of Raman spectroscopy that im- Barbo
ages multicomponent electrodeposition of mag- Br
netic thin films in situ and in real time. Ratner
has added scanning probe techniques (scanning tunneling
microscopy and atomic force microscopy) for imaging bio-
logical surfaces under water. Tom Horbett studies the re-
sponse of cells to polymer surfaces with respect to platelet
activation and surface thrombogenicity by thermodynamic
means (adsorption isotherms) and a new imaging system
incorporating a mega-pixel, cooled CCD camera.
In the latter (dry) category, Dave Castner studies poly-
mer-metal interactions with X-ray photoelectron spectros-
copy and X-ray absorption methods, the latter at the Na-
tional Synchrotron Light Source (Brookhaven). Bill Rogers
uses a wide range of surface science techniques to study
novel organometallic precursors for growth of boron nitride
and aluminum nitride thin films.
With all of this surface science comes a thirst for instru-
mentation. For its major surface science equipment, the de-
partment has four X-ray photoelectron spectrometers, two
secondary ion mass spectrometers (one of which is time-of-
flight), three atomic resolution microscopes, and a field ion
microscope. Nine ultrahigh vacuum chambers support this
equipment, as well as other (minor) instruments such as
mass spectrometers, electron diffraction optics, and Au-
ger spectrometers.
The department also maintains strong research programs
in biotechnology. Those with surface science components
(Castner, Horbett, and Ratner) were mentioned above. Two
programs involve research on bacteria. Mechanisms of pro-
tein production in the bacterium E. Coli and regulation of
protein folding by chaperonins are studied by Franqois
Baneyx. Another class of bacteria, the methylotrophs, which
86

professors at a local rally. From left to right: Larry Ricker,
Ta Krieger-Brockett, Dan Schwartz, Brad Holt, Eric Stuve,
ce Finlayson, and Lew Wedgewood. (Photo by Eric Stuve.)

feed on methane or chlorinated C, hydrocarbons, are exam-
ined with respect to their potential in bioremediation and
other biocatalytic reactions by Mary Lidstrom.
Materials related research includes the surface science
work of Schwartz and Rogers, mentioned above, as well as a
strong emphasis on polymers. The Polymer Composites Labo-
ratory of Jim Seferis examines the influence of processing
methods on polymer performance. This work has been in-
corporated into the new Boeing 777, which is the first com-
mercial airliner to incorporate polymers in structural compo-
nents. Graham Allan works with natural polymers (wood
fibers) to precipitate inorganic pigments inside the fiber.
This increases the fiber's opacity to allow reduced fiber
content in paper. The flow properties of macromolecules are
studied both experimentally and numerically by Lew
Wedgewood. This work combines traditional rheological
measurements with numerical analysis and has led to a new
vorticity theory for non-Newtonian fluids.
The department's program on environmental studies in-
cludes studies of the behavior of aerosols and solid particu-
lates in the atmosphere, marine bioremediation, and reaction
engineering. Jim Davis studies gas/solid and gas/droplet
interactions and reactions with the aid of the dynamic
electrobalance, a device capable of isolating and suspending
single particles or droplets of micrometer size. Absorption
(or release) of pollutant gases can be followed by light
scattering, mass spectroscopy, and Raman spectroscopy.
Barbara Krieger-Brockett examines thermal reactions
(combustion) of solids and then uses statistical models to
predict gas evolution rates during these reactions. She also
Chemical Engineering Education

TABLE 1
Chemical Engineering Faculty, University of Washington

G. Graham Allan (Professor of Forest Resources): fiber and polymer science
Albert L. Babb (Professor Emeritus): reactor engineering, bioengineering
Past chair, Committee on Future Nuclear Power Development (National Academy of Sciences)
National Academy of Engineers
Franqois Baneyx (Asst. Professor): biotechnology, protein technology, biochemical engineering
NSF-CAREER award
John C. Berg (Rehnberg Professor): interfacial phenomena, surface and colloid science;
Co-Editor in Chief, Adv. Colloid & Interf. Sci.; Editorial Board, Langmuir, J. Colloid & Interf. Sci.,
Colloids & Surfaces, J. Adhesion Sci. & Tech.
UW Distinguished Teaching Award; Guggenheim Fellowship; Alpha Chi Sigma Award
David G. Castner (Research Assoc. Professor): polymer and metal-organic interfaces, catalytic materials
Society for Biomaterials (chair, surf. charact. & modification group); American Vacuum Society
(past local chapter chair, applied surface science division); Assoc. Editor. Plasmas & Polymers
E. James Davis (Professor): colloid and aerosol physics and chemistry, electrokinetics, light scattering
Regional Editor, Colloid & Polymer Sci.; Assoc. Editor, Aerosol Sci. & Tech.; Editorial Board, J.
Aerosol Sci., J. Colloid & Interf. Sci.; American Association for Aerosol Research (past treasurer,
vice president)
Sinclair Award
Bruce A. Finlayson (Rehnberg Professor and Chair): modeling of chemical engineering problems
CACHE (past trustee); AIChE (past director. CAST division chair): National Research Council;
Editorial Board, Intl. J. Num. Methods in Fluids, Num. Heat Transfer, Num. Methods Part. Diff.
Eqns., Chem. Engr. Educ.
Walker Award; AIChE Fellow; National Academy of Engineers
Bradley R. Holt (Assoc. Professor): process design and control
NSF-PYI Award
Thomas A. Horbett (Professor of Bioengineering): proteins at interfaces, biomaterials, drug delivery
Clemson Award
Barbara B. Krieger-Brockett (Assoc. Professor): reaction engineering
Mary E. Lidstrom (Jungers Professor): biomolecular engineering, metabolic engineering
American Academy of Microbiology; Editorial Board, J. Bacterial., FEMS Microbiol. Rev., FEMS
Microbiol. Ecol.
NSF Faculty Award for Women
Buddy D. Ratner (Professor of Bioengineering): biomedical polymer surfaces and interfaces
Editor, Plasmas & Polymers; Assoc. Editor, J. Biomed. Marls. Res.; Editorial Board, Surface Sci.
Spectra, J. Biomatls. Sci., Biomaterials, Biomed. Matls.
Clemson Award; Perkin Elmer Award
N. Lawrence Ricker (Professor): process control and optimization
J.W. Rogers, Jr. (Professor): surface chemistry and engineering, applications to thin film deposition
American Vacuum Society (past national program chair, publications committee chair); Editorial
Board, J. Vacuum Sci. & Tech. B
Battelle Professorship
Daniel T. Schwartz (Asst. Professor): electrochemical engineering
Electrochemical Society (chapter chair)
DOE Jr. Faculty Award; NSF-PYI Award; UW Outstanding Faculty Achievement
James C. Seferis (Boeing/Steiner Professor): polymers and their composites, manufacturing, teaming
Past chair, Gordon Conference
NSF-PYI Award; Humboldt Prize; National Academy of Athens; Metier Award
Eric M. Stuve (Professor): electrochemical surface science
American Vacuum Society (past chapter chair, past director, trustee); Co-chair, Gordon Conference
Humboldt Fellowship; NSF-PYI Award
Lewis E. Wedgewood (Asst. Professor): computational fluid mechanics, macromolecular fluid flow,

Teaching Faculty
Kermit Garlid (Professor Emeritus): past Chair of Nuclear Engineering
William J. Heideger (Professor Emeritus): Director of Engineering Advising Center
Gene L. Woodruff (Professor): Dean Emeritus of the Graduate School

Spring 1996

uses statistical methods to chart and pre-
dict the effects of marine bioremediation
in Puget Sound in collaboration with the
Department of Oceanography.
The computer-intensive research pro-
grams include those of Finlayson, Holt,
and Ricker. Bruce Finlayson continues to
apply computer methods to solution of
chemical engineering problems, work that
has grown out of his three textbooks on the
subject. Current work includes a model for
blood solutes passing through the heart and
skeletal muscles and a model for the swal-
lowing process with the ultimate aim of
improving the condition of the millions of
people with swallowing problems. Brad
Holt studies the use of nonlinear control
algorithms and neural networks to imple-
ment robust controllers, and Larry Ricker
uses model predictive control to develop
algorithms for complex continuous and
batch processes. The Metro sewer project
previously mentioned consisted of a pro-
gram to manage twenty-three flow con-
trollers in the existing sewer system to
handle storm water runoff without pollut-
ing local bodies of water.

FACULTY
The department is fortunate to have a
faculty deeply committed to all aspects of
academic life: teaching, research, and ad-
ministration. A goodly number of major
awards have been bestowed on the faculty,
including two positions in the National
Academy of Engineering (Babb, Finlayson)
and five NSF young investigator awards
(Seferis, Holt, Stuve, Schwartz, Baneyx).
The department also has a strong commit-
ment to service; many faculty hold major
positions on editorial boards or within pro-
fessional societies. Table 1 lists the de-
partmental faculty along with their re-
search interests, major awards, and ser-
vice appointments.

ACKNOWLEDGMENTS
Our special thanks go to J. Ray Bowen,
Dean of UW's College of Engineering,
Joseph L. McCarthy, Dean Emeritus of the
Graduate School, and to R. Wells Moulton
and Charles A. Sleicher, both of whom
served as chair of the Chemical Engineer-
ing Department. I

educator

MORTON M. DENN

ARUP K. CHAKRABORTY,
ARTHUR B. METZNER,*
T.W. FRASER RUSSELL*
University of California
Berkeley, CA 94720
M orton Denn is a household
name in modern chemical en-
gineering circles, not only be-
cause of his many research accomplish-
ments and service to our profession, but
also because of the influence his books
and classroom teaching have had on the
education of young chemical engineers. In
this paper, we, three of his former and
current colleagues who span roughly three
generations of chemical engineering aca-
demics, will highlight some of Mort's re-
search accomplishments and provide per- and...
spective on some of his enduring contri-
butions to chemical engineering education.
Mort was born and raised in Paterson then
(New Jersey), close to the city of New
York. He got his first taste of "chemical (circa
engineering" in 1957 during a six-month
stint as a laboratory assistant making
foamed plastic boat bumpers and fishnet sans
floats for the Linen Thread Company. This
taste of chemical engineering served to beard.
whet Mort's appetite for the subject, and
he later chose to major in chemical engi-
neering at Princeton University.
An incident during his junior year at
Princeton may have helped shape Mort's concise style of
writing (it is difficult to find an extraneous phrase in articles
and books written solely by Morton M. Denn). A laboratory
report, graded by Michel Boudart, was returned to Mort
liberally covered with red ink. He was humiliated that a
native French speaker could find so much fault with his
English and vowed that he would henceforth write in such a

way that it would never happen again.
Reading his textbooks and knowing of
his exemplary work as editor of AIChE
Journal and the Journal of Rheology,
most would agree that he has succeeded.
Mort worked on his senior thesis with
Bill Schowalter, a newly arrived Assis-
tant Professor at the time. They aimed
to carry out an experimental study of
normal stresses in polymer solutions
(people were still not convinced nor-
mal stresses were real in those days).
Mort often remarks that after that expe-
rience it is a wonder that either of them
ever did experiments again, despite the
fact that both the data points were good
ones. Meanwhile, he read tensor analy-
V A sis and basic rheology papers with
Schowalter and his students and devel-
oped a strong interest in the mathemati-
cal analysis of physical problems rel-
evant to chemical engineers. This would
be the underlying theme in much of
Mort's early research and the pedagogy
in his books.
Mort decided to pursue graduate stud-
ies in Minnesota. He was interested in
mathematics in chemical engineering,
and Richard Wilhelm had told him to
go there. Also, in the summer of 1961,
SMort had worked at the Dupont Engi-
neering Research Laboratory in
Wilmington with Alan Foss (later a fac-
ulty colleague at Berkeley), Jon Olson
(later a colleague at Delaware), Steve Whitaker, and the late
Forest Mixon. These "experienced" coworkers told him that
Rutherford Aris was the hot item in applied mathematics in
chemical engineering, so Mort arrived in Minnesota with the
goal of becoming an Aris student. As a result, Aris agreed to
become the thesis advisor of yet another student who would
later become one of the leaders of our profession.

* Address: University of Delaware, Newark, DE 19716
Copyright ChE Division ofASEE 1996

Chemical Engineering Education

TO

Mort's thesis fo-
cused on optimiza-
tion of systems with
interconnecting
structures, a problem
that had been caus-
ing real difficulties
but which proved
to be "relatively
straightforward"
(Mort's view) when
put into the proper format

Just prior to completing his thesis, Mort learned from
Schowalter (who was spending a sabbatical leave at Minne-
sota) that Metzner at Delaware was looking for a postdoctoral
coworker in the area of non-Newtonian fluid mechanics. He
applied for, and was offered, the job. At Delaware he contin-
ued to spend time working on optimization problems and
explored a newfound interest in optimal control.
After one year as a postdoctoral scholar, Mort joined the
faculty at the University of Delaware. Interestingly, he had a
joint appointment: 75% in computer science and 25% in
chemical engineering. Three years later he moved full time
into chemical engineering since he found his interests and
those of the computer science department were not moving
in the same direction. Mort was also promoted to tenure in
1968 (in just three years!). Once, when he and Henry
Weinberg were asked "How did you fellows get tenure so
quickly in those days? Were you just so much better, or was
it easier?" even before the irrepressible Weinberg could say
anything, Mort replied without hesitation, "Both!"
What do we know about the young Mort Denn, and what
fostered his ability to make incisive decisions with such
apparent ease? Some insight may be gained through a story
his father related at the annual AIChE meeting in 1977 when
Mort received the Professional Progress Award. Mort had
invited his parents to attend the ceremony, and when he was
absent from the table for a few minutes, the conversation
turned to the clarity with which Mort is able to analyze
societal as well as professional issues and at what age these
attributes may have developed. His father responded just as
definitively as Mort usually does with the following story:
At a very early age. I remember an occasion when Morton
had just passed his thirteenth birthday and General
Eisenhower, running for election, scheduled a campaign
appearance in our home town. By pressing the local Re-
publican officials I was able to arrange for the General's
motorcade to change its itinerary sufficiently to enable a
brief stop at our store. I wanted our son to meet this
famous hero of World War II-perhaps a once-in-a-life-
time opportunity. All was arranged; Morton came to the
store at the appointed time, but was less than enthralled
when I explained to him, for the first time, what I had in
mind. The motorcade arrived and I stepped outside to meet
Spring 1996

the general, but when he
and I entered the store,
Morton was nowhere to
be found!. Clearly, he
had taken full control of
his own destiny by age
thirteen and was ready
to proceed incisively
with developing it as he
saw fit and with mini-
mum outside interfer-
ence.

t.

Morton Denn is a household name in modern
chemical engineering circles, not only because of
his many research accomplishments and service
to our profession, but also because of the
influence his books and classroom teaching
have had on the education of young
chemical engineers.

The period from 1965 to 1977 was a remarkably produc-
tive period in Mort's career. He continued working on opti-
mization of distributed parameter systems, and many of the
papers he wrote during this period were among the first in
that area. But his most insightful and lasting contributions
were made in the field of rheology and non-Newtonian fluid
mechanics. The following examples, taken from the broad
range of problems that Mort worked on, make clear the
impact of his research in the first dozen years of his career.
In 1970, Mort and one of his students, James Ultman
addressed an issue that has since turned out to have deep
implications for polymer flow and transport behavior."1I They
were studying heat transfer from cylinders submerged in
flowing polymer solutions. The experimental data of David
James showed some rather unusual behavior, which they
ascribed to the fact that the relevant transport equations
changed from elliptic to hyperbolic (thus allowing
discontinuities) when certain pertinent variables acquired
values that exceeded a certain threshold. While the imple-
mentation was not done rigorously by Ultman and Denn, this
idea has proved to be wide-ranging in its applicability to
transport phenomena in polymeric liquids, and Dan Joseph
and coworkers[21 applied the idea rigorously to a variety of
situations in the 1980s.
In 1971, Pino Marrucci and Mort published a paper on
stretching of viscoelastic fluids that is characteristic of much
of Mort's work in the first decade of his career- incisive,
with profound implications, and yet mathematically simple.
They analyzed the stretching of a rod of viscoelastic mate-
rial.31 Marrucci was a former student of Gianni Astarita in
Naples, and as we have noted earlier, Mort had started work
on non-Newtonian fluids as Art Metzner's postdoc. Astarita
and Metzner[41 had published an analysis of some experi-
ments based on the notion that the stress saturated as the
stretching rate increased. Marrucci and Denn showed that
this analysis was flawed, as it considered only the slow
stretching rate asymptote; consideration of the fast stretch-
ing rate asymptote quickly showed the real physics. One can
only imagine the pleasure the two youngsters derived from
correcting their mentors' analysis. This paper with Marrucci,
together with Mort's earlier work on boundary layers for
elastic fluids, had an impact far beyond the difficulty of the

v

analysis be-
cause it led to
deep physical
insights. Both
papers are
widely cited
and used as ex-
amples in text-
books.
In the late
1960s and
early 1970s,
Mort wrote a
series of theo-
retical and experimental papers on Taylor-Couette instabilities in dilute poly-
mer solutionsth5 that set the stage for a lot of subsequent work on flow stability.
These papers emphasized, for perhaps the first time, the extreme sensitivity of
certain flows to second-order rheological effects. Mort still regrets the fact that
he missed the interesting case of the zero Reynolds number oscillatory insta-
bility that has been exploited extensively by Muller (now a Berkeley col-
league), Larson, and Shaqfeh in the early 1990s.6o As he ruefully points out,
the analysis for this case is there in his student Sun's thesis, but in a form that
did not reveal the interesting physics.
The mid-70s marked the beginning of Mort's research on the simulation of
polymer processing operations, especially melt spinning, where his work with
students and collaborators has defined the state of the art. The spinning work
spanned twenty years, starting with an unpublished MS thesis of Glen Zeichner
(co-advised with Art Metzner and Byron Anshus), followed by a seminal 1975
paper with Petrie and Avenas,1 through a comprehensive book chapter8" to a
very recent paper on dynamics and disturbance propagation.[91 This period also
marked the beginning of a productive collaborations with Jim Wei on the
modeling of coal gasification reactors.

In addition to making seminal contributions to non-Newtonian fluid me-
chanics and stability, during this period Mort also wrote three of his five
books. They cover a rather wide range of subjects: two are concerned with his
stability and optimization interests, and one is a textbook for the first course in
chemical engineering.l"Io Perhaps the most popular textbook that Mort has
written is Process Fluid Mechanics,[1[ used in many chemical engineering
departments around the world for undergraduate (and graduate) instruction.
Starting in 1965, the Delaware chemical engineering department initiated a
project to redesign the six-credit introductory sophomore courses in industrial
stoichiometry. Mort and TWF Russell devoted six years of effort to this
task, and the final product was a totally new approach to the traditional
first course in chemical engineering. The history of this course's develop-
ment (and the resulting textbook) was outlined in an issue of this journal
over twenty-two years ago.[121
Some conclusions drawn by the authors that have stood the test of time and
many years of classroom teaching by others are the "Denn-Russell rules":
Since chemical engineering departments require students to take freshman
courses in mathematics, physics, and chemistry, it's incumbent upon them to
use this material effectively in the introductory major course in chemical
engineering.

4 Mort and David Boger in the bush, late
1970s.
Mort, circa 1972, at the University of Dela-
ware: How do you operate a Weissenberg
rheogonimometer with the plates six
inches apart? V

* A crucial part of chemical engineering
is the design, planning, and interpreta-
tion of experiment, and this must be
introduced early in the curriculum.
* To analyze experiments properly and to
use the results for effective process or
product design, it is essential that one
be proficient in chemical engineering
analysis. This is distinctly different
from applied mathematics.
* There are two critical issues in
chemical engineering analysis that
must be effectively learned. The first is
the ability to look at any physical
situation and to develop the mathemati-
cal description; the second is the ability
to compare model behavior with
experiment and to draw conclusions.
* Students at the freshman or sophomore
level have great difficulty obtaining the
mathematical descriptions or model
equations, and a key component of
analysis is the presentation of logical
procedures to obtain the correct
equation.
Chemical Engineering Education

A
Mort with Bob Bird and Bruce
Finlayson on Ben Lomond
during a break between
morning and afternoon
sessions at a 1981 conference.

A
Mort and Jan Mewis, wine
tasting at one of the area
vineyards, 1981,
and

in 1986, Mort accepts the
Bingham Medal from the
Society of Rheology.

To teach chemical engineering analysis effectively, one should begin
with studies of liquid-phase reacting systems. With such systems, one
can show students a direct relationship between laboratory experiment,
mathematical modeling, and simple reactor and process design.
In the context of teaching chemical engineering analysis, students must
be introduced to rate of reaction and rate of mass transfer, the two
cornerstones of chemical engineering.
Analysis is a necessary, but not sufficient, skill for chemical engineers.

Mort left a rich legacy of discussions with faculty, administrative
officials, the Board of Trustees, and perhaps others, in Delaware. One
concerns Mort's appointment as chairman of the newly instituted univer-
sity-wide senate Committee on Promotions and Tenure. The new provost
had decided that the University of Delaware could not really qualify for
admission to the circle of respected research universities as long as
faculty who were of indeterminate quality and motivation were promoted
Spring 1996

to advanced status. Some departments had excel-
lent criteria for promotion at that time, but others
had rather slipshod criteria and procedures. The
provost was anxious to see this ambivalence cor-
rected, and the senate wisely chose Mort to carry
the flag for this purpose. Mort denies the accuracy
of the following recollection, but we do believe
the salient points to be correct.
Evidently, in prior years (before the establish-
ment of the senate committee) some department
chairmen submitted rather informal and incom-
plete files directly to the administration in support
of promotion requests, confident that in those days
of rapid enrollment growth every warm body was
sorely needed to meet teaching needs. One chair-
man, late to submit his document, had it carried to
Mort's department where it was simply popped
into the appropriate mailbox. When Mort found
the shoddy file, he hastily scribbled a note to the
chairman who had tendered the offending dos-
sier-a note to the effect that it was apparent that
this chairman did not have his office functioning
at the level he would wish since "some junior
clerk must have sent me this mess before you had
a chance to consider it." Subsequent submissions,
from all departments, were carefully prepared!

THE MOVE TO BERKELEY
In the late 1970s, Mort decided to leave Dela-
ware. This decision was not based on any profes-
sional dissatisfaction with the University of Dela-
ware. He simply wanted to live in an urban setting.
While several distinguished universities tried to
recruit him, he chose the University of California
at Berkeley, joining the faculty in 1981. The diver-
sity and academic excellence that characterizes
Berkeley, together with the cultural opulence and
natural beauty of the Bay Area, proved to be the
determining factors.
Upon arriving at Berkeley, Mort established a
polymers and composites research program in con-
junction with the Lawrence Berkeley National
Laboratory. This is now a vibrant program with a
principal focus on anisotropic polymers and poly-
mer-surface interactions that has many principal
investigators working on diverse problems rang-
ing from field-theoretic descriptions of the statisti-
cal physics of polymers to issues pertinent to large-
scale processing. Mort runs the program like a
benevolent dictator.
The emphasis of Mort's research at Berkeley
has shifted toward more experimental work. In the
1980s, there were two parallel efforts in his group:

large-scale computations of viscoelastic fluid flows and
complementary experimental work. The computational work
was done largely in collaboration with Roland Keunings
(now Professor at The Catholic University of Louvain). Many
interesting things emerged from the Denn laboratory in these
years, but perhaps the most comprehensive and exciting
studies concerned the flow of fiber suspen-
sions~'31 and viscoelastic jet breakup.1141 Theo-
retical and computational work on the rheol- In his e
ogy of fiber suspensions involved a syner- Mort w
gistic collaboration with David Boger in sma
Melbourne, whose group did the comple-
mentary experiments. Computational and and so
analytical work done in Mort's group, single it
combined with pre-existing experiments, various
solved many aspects of the viscoelastic
jet-breakup problem. world
collabora
Mort developed a very strong interest in
the rheology and fluid mechanics of liquid joint si
crystalline polymers, a subject that is still a student
central focus of study in his laboratory. He emphasis
made the decision in the late 1980s to aban-
don, for the moment, computational and theo-
retical work in this area and to focus exclu- perhaps
sively on experimental work, because he the Ber)
strongly believed that there were many is- have led
sues in this field that were not understood colla
even phenomenologically. As such, he de-
cided that complex algorithm development rec
was not fruitful until some of the relevant

questions were understood at the simplest physical and phe-
nomenological level. Mort also decided that such an under-
standing could only be obtained via well-designed experi-
ments. Over the past few years, Mort's lab and those of
others have produced experimental data that (in Mort's view)
have begun to provide the phenomenological understanding
of liquid crystalline polymer rheology that he believed was
missing five years ago.
The study of extrusion instabilities has been a major focus
of Mort's research for years.1'51 During the past decade he
has emphasized the role of the material of construction of the
extrusion die on melt flow, particularly phenomena associ-
ated with failure of the no-slip boundary condition. Mort's
innovative work in this area has combined adhesion theory,
rheology, and surface spectroscopes ranging from XPS to
ATR/FTIR in an effort to understand how surface chemistry
and physics affect flow. He has had the satisfaction in recent
years of seeing enough activity generated by his contribu-
tions (e.g., a paper on adhesion and slip in polymer melts,[161
that he views as his most significant work of the decade) that
his own papers have become secondary sources.
In his early years at the University of Delaware, Mort
worked with a small number of students and sometimes with

single investigators in various parts of the world; there were
few collaborations involving joint supervision of students.
The greater emphasis on state-of-the-art experiments, and
perhaps the character of the Berkeley campus, have led to
much more collaborative work in recent years. He collabo-
rates extensively with Jeff Reimer in using NMR spectros-
copy to probe polymer structure. He has
also collaborated with Clay Radke, Alex
years..., Bell, David Soane, Miquel Salmeron, and
S..-.AL Susan Muller.

TEACHING AND
PROFESSIONAL ACTIVITIES
The Berkeley years have also seen Mort
giving much of his effort and time to pro-
fessional service and mentoring junior fac-
ulty in the department. In the late 1980s
and early 90s, the Berkeley department
had an eclectic group of young faculty
(mostly assistant professors) doing excit-
ing research as well as conceiving and
teaching new courses. They were also a
feisty bunch and respectfully challenged
some of the existing ways of doing things.
Mort, at least fifteen years senior to the
oldest member of this group of renegades,
was the leader of the pack. He took time
out to mentor many of these young fac-
ulty, even if their research areas were quite
different from his own. It is perhaps fair to

say that the success this group of now not-so-young faculty
has realized can be partly attributed to the supportive and
intellectually dynamic environment created by Mort and a
few other senior faculty at Berkeley. Today, Mort continues
to play a supportive role to the next generation of untenured
faculty at Berkeley.
Over the years, Mort has served the chemical engineering
profession in a variety of ways. For six years (1985-91) he
was the sole editor of the AIChE Journal, and he worked
hard to streamline the journal's operation. It is fair to say
that when he handed the baton to Matt Tirrell, the Jour-
nal had regained its stature as one of the flagship jour-
nals of our profession.
Mort has served on innumerable review panels, editorial
boards, and national committees of the AIChE, National
Research Council, and other organizations. He is not just a
warm body when it comes to these activities. He spends time
to consider the various issues, he does his homework, and he
contributes actively to the task at hand. His thoughts on the
evolution of the profession of chemical engineering'[17 reveal
that he worries about deep issues pertaining to the future of our
profession. His article, "The Identity of Our Profession," may
be controversial, but no one would deny that it is thoughtful.
Chemical Engineering Education

early

UreITU wiUI U
iber of students
metimes with
ivestigators in
s parts of the
here were few
rtions involving
supervision of
s. The greater
on state-of-the-
eriments, and
the character of
keley campus,
to much more
rative work in
ent years.

Mort served as department chairman at Berkeley for three
years (1991-94). He is very interested in curriculum devel-
opment and modification and continues to serve as the con-
science of the department in this
regard. After stepping down from
the chairmanship, we all expected
Mort to take a breather and con-
centrate purely on teaching and
research; however, this plan
seems to have gone awry, as he
has now taken over as editor of
the Journal of Rheology.
Mort's former students and
postdocs are profitably employed
in industry, academia, and national
laboratories. Perhaps a sign of the
fact that he is "maturing" is the Mort and Viv
many former students and
postdocs who are now in academia. These include Benny
Freeman (NC State), Alejandro Rey (McGill), Jim Ultman
(Penn State), Rakesh Jain (Harvard), Roland Keunings (C.U.
Louvain, Belgium), Davide Hill (Notre Dame), Wen-Ching
Yu (former dean of engineering at Tunghai University,
Republic of China), Bob Fisher (UConn), Doug Kalika
(Kentucky), Glenn Lipscomb (Toledo), Doug Bousfield
(Maine), Rakesh Jain (Harvard), and S.J. Lee (Seoul Na-
tional University, Korea).
Mort's remarkable achievements as a teacher, a researcher,
and a responsible member of the profession have been rec-
ognized through a multitude of awards and honors, includ-
ing a Guggenheim Fellowship, the Professional Progress
and William H. Walker awards of the AIChE, the Bingham
medal of the Society of Rheology, a Fullbright Lecture, and
the Chemical Engineering Lectureship Award of the Ameri-
can Society for Engineering Education. In 1986 Mort was
elected to the National Academy of Engineering.
Mort has a policy of introducing his seminars with a brief
exposition of how his work complements that of his col-
leagues and how it was intended to contribute to wider
departmental goals of teaching and research. This act of
unselfish and unusual devotion to his colleagues is not
one that he discusses, but it has been brought to our
attention by faculty from other institutions who have
heard his presentations. This devotion-to his profes-
sion, to his students, to his department-is a thread run-
ning through all his activities.
Mort continues to serve the profession and the chemical
engineering department at Berkeley by his exemplary stan-
dards in research and teaching, by generously giving of his
time to mentor young faculty, and by looking after the Jour-
nal of Rheology. Mort is happily married to Vivienne
Roumani-Denn, head of the Berkeley Earth Sciences and
Map Library. Mort and Vivienne enjoy living in the Bay
Spring 1996

enne

area and often welcome young faculty and their families to
their home for a pleasant evening and panoramic bay views.
They have six children between them: Matthew (practicing
law in Wilmington), Susannah
(works for the District Court in
San Francisco), Rebekah (writes
for Gannett papers in
Westchester County, NY), Aryeh
(a financial analyst for Smith-
Barney in New York), Dania (a
student at UCLA), and Natan (a
student at Berkeley High
School). The chemical engineer-
ing faculty at Berkeley are glad
that Mort's thoughful voice will
be around in the coming years to
at home, 1995. help face and overcome the chal-
lenges that await us.

ACKNOWLEDGMENTS
We thank colleagues at Berkeley and Delaware for sharing
some of their recollections and thoughts regarding Morton
M. Denn.

REFERENCES
1. Ultman, J.S., and M. M. Denn, Trans. Soc. Rheol., 14, 307
(1970)
2. Joseph, D.D., M. Renardy, and J.C. Saut, Arch. Rat. Mech.
Anal., 87, 213 (1985)
3. Denn, M.M., and G. Marrucci, AIChE J., 17, 101 (1971)
4. Metzner, A.B., and G. Astarita, AIChE J., 13, 316 (1967)
5. For example, see Roisman, J.J., and M.M. Denn, AIChE J.,
15, 454 (1969); Sun, Z.S., and M.M. Denn, AIChE J., 18,
1010 (1972)
6. Larson, R.G., E.S.G. Shaqfeh, and S.J. Muller, J. Fluid
Mech., 218, 573 (1990)
7. Denn, M.M., C.J.S. Petrie, and P. Avenas,AIChE J., 21, 791
(1975)
8. Denn, M.M., "Fiber Spinning," in Computational Anaysis of
Polymer Processing, J.R.A. Pearson, S. Richardson, eds.,
Elsevier Applied Science Publications, 179 (1983)
9. Devereux, B.M., and M.M. Denn, Ind. Eng. Chem. Res., 33
2384(1994)
10. Denn, M.M., and T.W.F. Russell, Introduction to Chemical
Engineering Analysis, Wiley (1972)
11. Denn, M.M., Process Fluid Mechanics, Prentice-Hall (1980)
12. Denn, M.M., and T.W.F. Russell, Chem. Eng. Ed., 7, 117
(1973)
13. Lipscomb, G.G., D.U. Hur, D.V. Boger, and M.M. Denn, J.
Non-Newtonian Fluid Mech., 26, 297 (1988)
14. Bousfield, D.W., R. Keunings, G. Marrucci, M.M. Denn, J.
Non-Newtonian Fluid Mech., 21, 79 (1986)
15. For example, see Denn, M.M., Ann. Rev. in Fluid Mech., 22
13 (1990); Denn, M.M., ASEE Chemical Engineering Divi-
sion Award Lecture, Chem. Eng. Ed., 28, 162 (1994)
16. Hill, D.A., T. Hasegawa, and M.M. Denn, J. Rheology, 34,
891(1990)
17. Denn, M.M., in C.W. Colton, Ed., Perspective in Chemical
Engineering (Advances in Chemical Engineering, 16), Aca-
demic Press, 565 (1991) p

Mj classroom

THE MASS TRANSFER

BOUNDARY LAYER

WITH FINITE TRANSFER RATE

MORTON M. DENN
University of California Berkeley CA 94720-1462

Mass transfer with finite rates is usually passed over
with only a brief mention in undergraduate trans-
port courses because of the complexity of the
coupled problem in mass and momentum transport. Solu-
tions to classical problems are available; Bird, Stewart, and
Lightfoot (BSL)," for example, present an elegant similarity
solution for simultaneous heat, mass, and momentum trans-
fer in a laminar boundary layer. The similarity solution is
of necessity numerical in the final stages, and the form is
sufficiently complex in structure that it is difficult to
obtain insight to the underlying physical process in a
straightforward manner.
I have found that it is much more effective to introduce
students, undergraduate and postgraduate alike, to the con-
cept of finite transfer rates through the von Karman-Polhausen
(vK-P) approximation. Undergraduates have often seen the
method previously in their study of fluid mechanics (e.g.,
Denn[21). With graduate students I precede this material with
a scaling analysis to obtain the basic structure of the relation
between the Sherwood, Schmidt, and Reynolds numbers
(Denn[31). Standard texts often use the vK-P method for mass
transfer boundary layers, but they fail to take advantage of
one of its most significant pedagogical features: the finite-
rate problem is no more complex than the uncoupled one.
While there is no question but that the exact solution, as

Morton M. Denn is Professor of Chemical En-
gineering at the University of California, Berke-
ley, and Head of Materials Chemistry and Pro-
gram Leader for Polymers in the Materials Sci-
ences Division of the Lawrence Berkeley Na-
tional Laboratory. He received a BSE from
Princeton University in 1961 and a PhD from
the University of Minnesota in 1964, followed
by a postdoctoral year at the University of Dela-
ware. A profile of Professor Denn can be found
elsewhere in this issue.

Copyright ChE Division ofASEE 1996

given in BSL, is to be preferred for accuracy, the important
structural features of the coupled solution are more clearly
revealed through the simple analytical expression afforded
by the vK-P approach.
The solution given here is not completely new; the prob-
lem was considered within a broader class, for example, by
Spalding.t4'5' The analogous transpiration cooling problem in
heat transfer was analyzed in part by Yuan and Ness.[61 I
believe it is useful, however, to present the specific case of
finite mass transfer in laminar flow over a flat plate in a form
that is easily accessible to students, who would find the
original literature on vK-P solutions as difficult as the exact
solutions. In that regard, this paper might be thought of as a
lesson plan and a possible supplement to a course text.

PROBLEM FORMULATION
Consider flow of an incompressible Newtonian fluid with
constant physical properties over a flat plate at zero angle of
incidence. The flow direction is x and the transverse direc-
tion y. The surface contains a soluble species A, while the
fluid phase is an ideal mixture of A and B. It is implicit in the
standard problem formulation (cf BSL, p. 608 ff) that the
molecular weights of A and B are equal (MA = MB). The
overall continuity, momentum, and species continuity equa-
tions are, respectively,

-+ = (10
3x dy

vx Ox Vy2= (2Y

X A + V XA = D a A (3
vx a y AB y2

where

Chemical Engineering Education

I believe it is useful... to present the specific case of finite mass transfer in laminar flow
over a flat plate in a form that is easily accessible to students, who would find the original literature on
vK-P solutions as difficult as the exact solutions. In that regard, this paper might be thought
of as a lesson plan and a possible supplement to a course text.

v, Vy = x- and y-components of velocity, respectively
xA = mole fraction of A
1,DAB = kinematic viscosity and binary diffusivity, respec-
tively
Boundary conditions are
y=0: Vx=0, XA=XAO, NB=0 (4a,b,c
x<0andy->-o: vx=V-, XA=XA. (4d,e
where
v, X A. = "free-stream" values of velocity and mole fraction,
respectively ( XA, will often be zero)
XAO = equilibrium or saturation level of A in the fluid
phase at the surface y=0

NB = molar flux of B
It is also useful to record the molar flux of the transferring
species A at y=0 as

PVyO CDAB aXA (5;
NA MA 1-XA0 y y=0

where v is the transverse velocity at y=0 and c is the total
molar concentration. Because MA = M5, p = MAC. Equation
(5) establishes the coupling between the mass and momen-
tum equations.
It is convenient to use the dimensionless variables

= Vx V W XA -XA
U=-, V=-, w A -
V, V XAO XA

(6a, b,c)

Note that u and w range from zero to unity, but this is not
true of v, so these are not appropriate variables for a scaling
analysis (cf Denn131). The dimensionless form of Eq. (5) is

vo =-DAB A XA0 aw (7)
V_ I I-XAO )ay

Equations (1) through (4) then become

ax ay

au au ( ) a2u
u-+v- =-- (9)
Bx ay v_)3y2

dw dw ( 2w (10)
u -+v -= v 2 (10)
ax ay v-Sc ay 2

y=0 : u=w=0

(lla,b)

x 0 andy- -o : u=w=l (llc,d)
where Sc = U/DAB is the Schmidt number. When Sc = 1,
Spring 1996

Eqs. (9) and (10) for u and w are identical with identical
boundary conditions for any v, so it follows immediately
that u = w and the dimensionless velocity and concentration
profiles are the same.

vK-P FORMULATION

) The vK-P approach converts the differential equations to
an integral form. Firstly, the continuity equation is inte-
grated to obtain

Vy =vo-f a dy (12)

Equations (9) and (10) are then integrated from y=0 to y= oc;
with some manipulation and the use of Eqs. (8) and (12), we
obtain the starting point:

0

vo+f [ u(W-

o
0

-l)dy=--
v. ay
y=o

1)]dy vSc

y=0

We have assumed continuity of au / ay and aw / ay through-
out 0 < y < oo. This is a subtlety that should be addressed in
the classroom because it can become a problem for some
students subsequently.

The vK-P approximation is based on two fundamental
hypotheses. Firstly, one assumes the asymptotic approach to
free-stream conditions can be replaced by well-defined bound-
ary layers 8(x) and 8c(x) for velocity and concentration,
respectively. For y- 8(x)[6c] the velocity (concentration)
varies continuously from the surface condition to the value
at y = o; for greater values of y the velocity (concentration)
is constant. Secondly, one assumes that u and w are unique
functions of y / 8(x) and y /5 (x) respectively; this is a simi-
larity assumption that is in fact rigorous for the problem at
hand, but generally not for other problems in which the vK-P
approach is employed. Students usually need some discus-
sion of the physical meanings of both these (independent)
assumptions. We therefore write u and w as the functions

u =(p[y/8(x)], y ; u =1, y>8

w=V[y/8c(x)],y<8,c; w=l, y>8c

(15a)

(15b)

:)

)

The functions (p and y must be continuously differentiable
and satisfy boundary conditions
(p(0)= (0)= 0, p(1)= i(1)= 1 (16)
They are otherwise arbitrary. Equations (13) and (14) then
become

J ) d = I p'(0) (17)
XAOXA _I y'(o)

v Sc 1-XA0 8c

+ 8 [ )- dy=- v) -I(0) (18)
J 0x 5 C) ] V-Sc Sc

The upper limits of the integrals are 8 and 8c, respectively,
because the integrands are identically zero beyond these
points. It is important to remember that the argument of 9p is
y / 8(x) in both equations, whereas that of y is y / 8c (x).
There is a very convenient variable change here, which is
not necessary but simplifies the manipulations greatly. Stu-
dents often have a problem with the details of the calculus.
Define = y/8(x) in Eq. (17) and = y/8c(x) in Eq. (18).
The range of both integrals is then from = 0 to = 1; terms
and operations involving only x are independent of 4 and
can be taken outside the definite integrals. We thus obtain

) (XAO-XA- _1 '0)
v-Sc 1-XAO c

+ d8 IjdJ = 1 0(0) (19)
dxo v
0
v1)c C XA0OXA-J I,,Q
vSc ( [XAO 8c

I

Now define A = 8c / 8, where we assume A is a constant.
This assumption requires a consistency check later. With a
bit of rearrangement, Eqs. (19) and (20) then become

2 dx {(() [p(i )

0- )[(o)+ 1 C XAO-)XA '() (21)
v_ ASc 1- XA0

,1+ x0 [ XA-" '(0
v-Sc I ( 1- XA0 )

The right-hand sides and the coefficients in braces on the left
are all constants in these equations for specified (p and y, so
the square-root dependence of the boundary layer develop-
ment is established. For (XA0 XA)/(1- XA) <<1 the prob-
lems uncouple, in that Eq. (21) involves only fluid-mechani-
cal variables. We can write the solution to the coupled prob-
lem as

2 I '(0) + (XA 0) x
v. ASc I-XA0 )
1

0

,2 VSc XAO x J )j
c c-
S Jp(A) [v( 1)-1]d
0

The equation for A is obtained by setting A2 =2 / 82; this
ratio is clearly independent of x, establishing that A is in-
deed a constant.

CLOSED-FORM SOLUTION

The functions (p and xy are typically taken to be polyno-
mials, usually cubics. The structure of the solution is best
revealed by taking (p() = V(4) = since in that case A sim-
ply factors out of the integral involving (p(A4). There is a
problem at 4 = i, where the derivatives are not continuous,
which will be a concern to some students, but it is a technical
detail; the function can be taken as linear arbitrarily close to
S= 1 and then rounded suitably to provide continuity of the
derivative without changing the integrals in Eqs. (13) and
(14) by more than an infinitesimal amount. With
I
=-V=~=4, 4'= w'= 1, and f( 1-l)d =-6
0
we have

2 12 1+ 1 AO XA
v_ ASc I-XA0 ]

Chemical Engineering Education

I2 d6,
0

2 12 1+ XAO -XA. x (26
v,ASc 1-xA0 jj

and

A3Sc+ XAO ]A 2 +XAO A'- (27.
( 1-XA0 ) 1-XAO )

To first order in (xA0 XA,)/(1 XAO) the solution to Eq. (27)
is

A~Sc 3 1+1 XAO XA l-Sc 3 (28)
3l I-XA0 I

Defining Rex=xv' /v we obtain, to the same order in
(XAO -XA)/(1 XA ),

= 3.46 Rex 2 1+ 3 AO XA Sc'3 (29)
x 1 -XA0

c 3.46 Rex2 Sc 3 1 + XA 2+Sc (30)

MASS-TRANSFER COEFFICIENT

There are a number of ways in which the mass-transfer
coefficient can be defined. In my opinion the transport coef-
ficients need to be viewed as quantities defined by experi-
ment; for mass transfer the interfacial flux NAO and driving
force XAO XAA are the quantities that can be measured in
principle, so the proper operational definition for the mass
transfer coefficient kx is

kx A(3a)
C(XAO- XA m)
BSL choose to define the transport coefficient relative to the
molar average velocity, so their definition would be

S NAO (- XAO)
c(XAO XA-0 )

(31b)

Failure to distinguish between these definitions can cause a
great deal of confusion. (There is a dimensionality differ-
ence between both these definitions and those used by BSL;
k, as defined here, has dimensions of velocity.) The Sherwood
number, Shx = xkx /DAB, then follows from Eqs. (7) and
(30) to first order as

S1 1 2 )
Shx =0.289 Re2 Sc3 1- |6 2+Sc 3 XAO XA0 (32)
(The coefficient for the exact solution is 0.332. The form J

(The coefficient for the exact solution is 0.332. The form

Re"'Sc"3 follows directly from scaling arguments; cf Denn.m31)
) The correction relative to the low-flux limit is therefore

Shx 1 1- 12+Sc3 XAO -XA- (33
Shox 1-XAO 6 1-XA0

There are two contributions to the correction term. The
term in brackets, which reflects the contribution of the driv-
ing force, is less than unity; this term corresponds to the
effect of thickening the boundary layer with a constant driv-
ing force, hence reducing the gradient for diffusion. The
demoninator term 1- XA0 reflects the convective contribu-
tion to the total flux. The net effect of the two terms is to
increase the Sherwood number relative to the zero-flux limit.
(This is most easily seen by setting XA~= 0, XA0 << in

which case Sh-x -+ -Sc 3 xAO >l.) A Sherwood
Shox 3 6

number based on kx would decrease, however, since the
thickened boundary layer decreases the diffusive flux.

CONCLUDING REMARKS

I have found this example to be effective in introducing
students to the concept of corrections for finite mass-transfer
rates because it uses a methodology with which they are (or
should be) familiar and leads to a reasonably accurate result
in closed form. I like to emphasize the limitations inherent in
the usual assumption of no surface flux and the breakdown
of transport analogies at high rates of mass transfer. The
boundary layer, approached in the manner outlined here, is
an excellent vehicle for doing so, and at no "cost" since the
same methodology is likely to be used for the solution of the
uncoupled problem in order to establish the Sc"3 dependence
in the Sherwood number relation.

ACKNOWLEDGMENT
Warren Stewart called my attention to the papers by
Spalding and Yuan.

REFERENCES
1. Bird, R.B., W.E. Stewart, and E.N. Lightfoot, Transport
Phenomena, Wiley, New York, NY (1960)
2. Denn, M.M., Process Fluid Mechanics, Prentice Hall,
Englewood Cliffs, NJ (1980)
3. Denn. M.M., Process Modeling, Wiley, New York, NY (1986)
4. Spalding, D.B., "Mass Transfer in Laminar Flow," Proc.
Roy. Soc. (London), A221, 78 (1954)
5. Spalding, D.B., "Mass Transfer from a Laminar Stream to a
Flat Plate," Proc. Roy. Soc. (London), A221, 100 (1954)
6. Yuan, S.W., and N. Ness, "Heat Transfer in a Laminar
Compressible Boundary Layer on a Porous Flat Plate with
Variable Fluid Injection," Tech Memorandum P1B-15, Project
Squid, Polytechnic Institute of Brooklyn, Sept. 1, 1950 0

Spring 1996

laboratory

A LABORATORY EXPERIMENT

THAT ENHANCES

ENVIRONMENTAL AWARENESS

KEN K. ROBINSON, JOSHUA S. DRANOFF
Northwestern University Evanston IL 60208-3120

he goals of the senior-year chemical engineering

laboratory course at Northwestern University are to
nurture critical thinking skills so that students can
analyze open-ended problems, to develop and improve the
student's technical communication skills, and to provide
experience with typical equipment and instrumentation. We
try to accomplish these goals in one academic quarter by
requiring student teams of 3-4 students each to run five
different experiments (from a current total of eight, listed in
Table 1), to prepare a detailed written technical report for
each experiment, and to present an oral report as the final
exam. Furthermore, we attempt to stimulate their thinking
skills by purposely giving students very brief instructions for
most of the more standard experiments so they cannot sim-
ply "follow the cookbook" in running the experiment.
A laboratory course essentially similar to ours, frequently
designated as a "Unit Operations Laboratory," has been a
core course in chemical engineering curricula for many years.
Thirty years ago, most chemical engineering instructional
laboratory equipment was large and multi-storied, while to-

Ken Robinson has been a Lecturer at North-
western University since 1993, with primary re-
sponsibility for the undergraduate chemical en-
gineering laboratory. He received his BS and
MS from the University of Michigan and his DSc
from Washington University. He has worked in
industry for both Amoco (1973-1992) and
Monsanto (1965-1973).

Joshua Dranoff is Professor of Chemical En-
gineering at Northwestern University, where he
has been on the faculty since 1958. He re-
ceived his BE degree from Yale University and
his PhD from Princeton University. His research
interests are in chemical reaction engineering
and chromatographic separations.
Copyright ChE Division ofASEE 1996

We continually try to keep the
experiments relevant by phasing out older
ones and introducing newer ones that expose
students to problems of the day....
Environmental awareness is an important
theme and provides the motivation for the
experiment discussed in this article.

day the scale is usually smaller and more compact due to
safety and space considerations. The unit operations labora-
tory has normally been the first opportunity for students,
particularly those not in a cooperative education program, to
observe and operate larger-scale equipment and to begin to
appreciate some of the more realistic situations they might
face in industry. Some of the concepts we hope they learn in
the course are an appreciation of the difference between
steady-state and transient operation and how long it can take
to reach steady state in particular equipment, knowing when
a computerized data acquisition system is giving realistic
and reliable numbers, and how to use and draw reasonable
conclusions from data that are limited and far from perfect.
In this connection, Dahlstrom, in a recent article on the
history of chemical engineering education,m11 described the
important role played by Olaf Hougen through his practical
approach to education. Hougen's Principles12'3] have particu-
lar relevance to the way we prepare chemical engineering
students for industrial careers with such laboratory courses.
Although Hougen listed twelve principles, the three most
applicable to this course are
If you can findd relevant problems to give the student, then you
shouldn't be teaching the material to the students.
Well-founded and well-tested empiricisms are to be preferred
over theories that have only a limited range of applicability.
It is vital for engineers to know how to solve problems with
limited or incomplete data.
We continually try to keep the experiments relevant by
phasing out older ones and introducing newer ones that
Chemical Engineering Education

expose students to problems of the day. For example, com-
puterized data acquisition is standard in most petroleum
refineries and chemical plants, so we have incorporated per-
sonal computers with data acquisition hardware/software for
all of the more sophisticated experiments. Environmental
awareness is another important theme and provides the moti-
vation for the experiment discussed in this article.

FLUID-BED INCINERATION
We have developed a new fluid-bed incineration experi-
ment for the chemical engineering laboratory that we believe
gives students useful exposure to solid and liquid waste
treatment. A recent article by J. Mullen[41 described the use
of fluid-bed incinerators for the destruction of hazardous
wastes. Such an incinerator basically consists of a shallow
fluid bed in which air, fuel, and combustible waste are fed
into the bottom where combustion of fuel and waste takes
place in the fluid bed medium, typically sand. The sus-
pended solids/gas mixture has a vigorous boiling action and
high heat transfer, which results in rapid and thorough mix-
ing of the air, fuel, waste combustibles, and fluid-bed media.
Some of the advantages of the fluidized bed combustor
compared to other types of incinerator include efficient com-
bustion, easy control, the ability to handle variable feeds,
and much lower emissions of NOx and metals; the fluid bed
typically operates at temperatures
between 1400-1650'F. Combus-
tibles are exposed to full com- Northwestern
bustion temperature for 5-8 sec-
onds or more. _
onds or more. Pressure drop in fixed and
We decided to build a bench- 0 Heattransferindouble-pi
scale fluid-bed incinerator for in- i0- Mass transfer in a wetted
corporation into our laboratory >0 Sucrose inversion in aplu
course. The basis for the design 0 Propanoldehydrationina
was provided by Ecova Inc., a Fractional distillation of
subsidiary of Amoco Oil located column
in Denver, Colorado. They had 0 AMixing andesidencetim
developed a bench-scale unit for 0 Unsteady-stateat condu
establishing design parameters
on a commercial facility in
Kimball, Nebraska, rated at an
annual capacity of 45,000 tons
of hazardous waste. This com-
mercial unit was started up in
the spring of 1994 but has since 3.4 kwheater
been shut down due to unfavor-
able economics and potential li- 24*
ability from its operation.
The laboratory unit was con-
structed in the Northwestern Uni- fluidizer cap.
versity shop over a three-month n n
period. The entire unit, complete
with instrumentation, was built
for slightly less than $10,000. Figure 1. Flu
Spring 1996

lid b

(An additional $5,000 would be required for analytical equip-
ment, specifically a dual-column gas chromatograph and
electronic integrator.) The unit is simple to run and seems to
fit well into the framework of the chemical engineering
laboratory.
Experimental Unit Design The main elements of the ex-
periment unit are the fluid-bed combustor, the feed systems
for air, methane fuel, and waste material, and the analytical
equipment. The feed gas monitoring and mixing chamber
and the combustion unit are located in a fume hood.
The fluid-bed incinerator is a vertically-mounted 3-inch
stainless tube, which is 24 inches long. It is enclosed with
two semi-cylindrical electric heaters (3.5-inch ID), with a
total heating capacity of 3.4 kilowatts, and 2 inches of insu-
lation. Bed temperature is controlled with a solid-state Omega
temperature controller that cycles current through a 220-volt
relay and into the electrical heating jacket around the incin-
erator. The chamber is surmounted by a particle disengaging
zone in which the diameter increases to 8 inches.
As shown in Figure 1, there are four ports along the
combustor where temperature and pressure drop can be mea-
sured. Type K (chromel-alumel) thermocouples have been
inserted into the ports, and a pressure tap, connected with a
Swagelok tee, is at the same location. Pressure drops are
measured using three Omega
LE 1 Engineering differential pres-
ratoryExperiments sure (DP) cells with a full
range of 20 inches water. The
zedxxis unit as described has a high
shell-and-tube exchangers height-to-diameter ratio that
olunn leads to bed slugging and a
catalytic reactor small freeboard height. For im-
ential catalytic reactor provements in the design, we
ol-waterinamulti-trayglass would suggest that the diam-
eter be somewhat enlarged (by
bution foratank-in-seriessystem about 4 inches) and the free-
in solids board increased to prevent flu-
idizing media from being
blown out of the bed.
8. disen.ingzone A side injection port is pro-
vided near the base of the in-
-saiess sel pi cinerator to introduce wastes
such as newsprint, plastic, rub-
ber, and liquid hydrocarbons.
It consists of a horizontal stain-
hermocouples(typeK) less-steel pipe with two 1/2-
Sinch ball valves, one to seal
the injection chamber from the
hot fluid bed and the other to
allow access to the injection
air chamber for loading with solid
-mehane wastes. A piston is mounted
ed incinerator on the end of a 1/4-inch tube
99

AB
Laboi

fluid
~eand
wallo
g-flow
differ
etban

distri
action

that slides through an end cap. For solid-waste injection, the
piston is advanced forward through the body of the ball
valve and stops with the piston flush with the inside wall of
the combustion chamber. If liquid wastes are to be pumped
into the incinerator, the piston is left advanced and a small
syringe pump feeds liquid into a tee located at the handle end
and through the center of the tube.
The bed is filled with Mulcoa 47-20X-50S, an alumino-
silicate medium (25x35 mesh) that can withstand higher
temperatures than regular silica sand, but sea or river sand
can also be used if desired since many commercial incinera-
tors use it.
Air and methane from cylinders are metered through rota-
meters, premixed in-line in a mixing chamber, and then fed
cold into the bottom of the fluid-bed unit through four jets
consisting of sintered porous fluidizer caps. The caps are
arranged with three in an equilateral triangle pattern and the
fourth in the center. Excess air is used to fluidize the bed and
lower the combustion temperature. Additional air purges are
located on either side of the side injection port to clear the
region of fluid bed media when a waste injection is being
performed. A layout for the gas distributor is given in Figure 2.
Flue gas from the unit is analyzed with a dual column gas
chromatograph equipped with two different columns and
thermal conductivity detectors. One column contains mo-
lecular sieve 5A and measures oxygen, nitrogen, methane,
and carbon monoxide, while the other contains PorapakQ
and primarily determines the amount of carbon dioxide and
water in the flue gas. A Teflon gas bag is used to collect an
average sample of the flue gas and is similar to the bag
sample concept used for testing automobile emissions. Typi-
cal chromatograms for these two columns are shown in
Figure 3. (The most important flue gas components for this
experiment are carbon monoxide and carbon dioxide.)

OPERATING PROCEDURES
Start-up of Unit:

Airflow is initiated at a moderate rate, somewhat less than the
fluidization velocity.
The temperature controller is then turned on and set to 8007F.
In about thirty minutes that temperature is reached, and the
temperature set point is increased to 1300F, which is slightly
higher than the auto ignition temperature of methane
(1200 F).
Once the temperature reaches 1300F, the airflow is
increased to start fluidization of the bed and methane is slowly
introduced into the mixing chamber outside the bed. Ignition of
the methane starts immediately. For the 3-inch diameter bed,
we use 1 SCFM of air and 0.07 SCFM of methane (air/fuel
ratio of about 16).
A small mirror is positioned above the top of the unit so
that the bed can be observed during operation. "Light-off" of
the incinerator is usually accompanied by a soft pop, after
100

which the bed begins to glow red. The set point to the
temperature controller can now be shut down since methane
combustion will sustain the temperature in the bed with no
external heating. The adiabatic flame temperature of meth-
ane with a stoichiometric amount of air (9/1 air-to-fuel ratio)
is 35000F, so it is important to feed the fluidization air in
excess or the incinerator will get too hot and severely dam-
age the steel combustion chamber as well as fuse the bed
medium into large aggregates. We suggest that a tempera-
ture controller with a high-limit switch be used to shut off
fuel gas flow with a normally closed solenoid valve.
Once the bed has reached steady-state operation, wastes
may be injected by means of the injection port at the bottom
of the fluid bed. As noted earlier, the injection port has two
stainless-steel ball valves (1/2-inch). The valves are mounted
so that one serves as a vertical loading port, while the other
is used to keep fluid bed media out of the injection chamber.
To load the port with waste, the vertically mounted ball valve
is opened and about 10 ml of waste material is dropped in.
The valve is then closed and an air purge is started on both
sides of the horizontal ball valve to keep the valve free of
granular solids.
The horizontal ball valve is then opened and a push rod is
manually advanced toward the bed to pass through the ball
valve and inject the solid waste into the fluid bed.
Injection of solid wastes is clearly an unsteady-state pro-
cess; consequently, the observations made with solid wastes

Top View

Sintered Metal Disperser Caps

Side View

Figure 2. Gas distributor layout.

1. oxygen
2. nitrogen
3. methane
4. carbon monoxide

4

S10mmin
6' x 1/8 mol sieve 5A
35 C

1. air, carbon monoxide
2. carbon dioxide
3. water
4. carbonyt sulfide
5. sulfur dioxide

0 6min
6'x 1/8" Porapak 0
75 C

Figure 3. Gas chromatograph analysis offlue gas.
Chemical Engineering Education

will be mostly qualitative. But steady-state conditions can be
achieved with liquid wastes. The push rod consists of a
hollow tube connected to a disc with a small hole drilled in
it; this can be used with liquid wastes that are pumped
continuously into the incinerator. It should be noted that the
combustion reactions of the fuel as well as the waste materi-
als produce large heat releases; dealing with these is consid-
erably more stimulating and challenging for the student than
the more typical use of air and water as process media in the
laboratory setting.

THEORY AND DATA ANALYSIS
This experiment provides opportunities for application of
several different types of theory and data analysis.
Minimum Fluidization Velocity and Pressure Drop The
student needs to estimate the minimum fluidization velocity
before starting operation of the incinerator. We find that you
cannot exceed this value too much or granular material is
entrained and blown out of the bed. Once the fluidization air
rate is established, this controls how much methane fuel can
be added for heating the unit.
Combustion and Adiabatic Flame Temperature

d (ps -Pg)g
Um= 1650p Rep<20

2 dp(Ps -Pg)g
S 24.5 g Rep > 1000 (1)

where Umf = minimum fluidization velocity
dp = particle diameter
pg = gas density
p, = particle density
g = fluid viscosity
g = gravitational constant
Rep = particle Reynolds number, (pgUmfdp)/ p
Stoichiometry We typically use methane as the fuel gas,
but propane will also function well. But the heating value of
propane is significantly higher than methane, so this must be
taken into consideration.

CH4 + 2 02 -> C2 + 2 H20

AH = -345,661 BTU/lb mole

Adiabatic Temperature Rise The target temperature for
the unit is 1400-18000F, so the student must calculate the
adiabatic temperature rise for some typical fuel gas feed
rates. The approximate equation for temperature rise with an
energy balance yielding the more exact expression is
-AH
AT = r (2)
Cp

where AT = temperature rise
AHr = heat of reaction
Cp = specific heat of feed stream
Spring 1996

GENERAL PROCEDURE
FOR TESTING AND ANALYSIS
The fluid-bed incinerator experiment is reasonably chal-
lenging for even the best students. Consequently, we recom-
mend that it be placed midway or later in the laboratory
course so that the students will be more experienced with
operating procedures and not require such close supervision.
We usually schedule the laboratory experiments so that the
students have operated the fixed/fluid bed experiment first
and have some appreciation of what fluidization is by ob-
serving it in a clear tube with small glass packing. They are
then better prepared to run the fluid-bed incinerator unit and
to address the aspects of fluidization, combustion, and incin-
eration of industrial wastes.
The experiment is performed over two laboratory periods
of eight hours each. During the first period, the students start
up the unit, measure temperature and pressure drop across
the bed and analyze the flue gas by filling the Teflon bags,
and then inject samples into both columns of the gas chro-
matograph with a gas syringe. During the second period,
wastes are fed into the fluid bed through the injection port.
Small pieces of polystyrene, plastic eating utensils, and rub-
ber stopper are first dropped into the top of the bed to
observe incomplete combustion at short residence times (typi-
cally yielding black particulates or soot). Then the same
materials are injected into the bottom port of the fluid bed
incinerator for essentially complete combustion under con-
ditions of longer residence time and better heat transfer.
Also, a bag sample of the exhaust is taken while the waste
injection is performed to compare to the operation with no
waste incineration. The following questions are proposed for
the students to think about and answer in their reports:
Does an energy balance on the unit predict an outlet
temperature close to what was observed?
How do the measured pressure drops and temperature
distribution relate to theory?
Does the minimum fluidization velocity calculation agree
with what you actually observed?
Discuss the flue gas composition and the distribution of
carbon monoxide/carbon dioxide.
Do you think significant nitrogen oxides (NOx) formed in the
fluid bed incinerator?
What about dioxin formation ? Besides hydrogen and carbon,
what is the key element needed to make dioxin in the flue gas
and what temperatures favor its formation?

CONCLUSIONS
The new fluid-bed incinerator experiment has lived up to
our expectations and has given a new spark to the array of
teaching experiments used for the unit operations laboratory.
It seems to satisfy a number of the criteria in Hougen's
Principles. More specifically, it is a relevant industrial prob-
Continued on page 160.
101

_ classroom

DESIGN COMPETITION

FOR SECOND-YEAR STUDENTS*

A Retrospective

W. A. DAVIES
University of Sydney Sydney, New South Wales, Australia 2006

After spending their high-school years solving theo-
retical problems for which there are always some
sort of defined "right answers," students tend to
arrive at the university with the belief that there is a right
answer to everything. In the practical world, however, there
is an inestimable number of ways of going wrong while the
number of ways of going right are preciously few. This is
one reason why our department includes formal practical
training throughout all stages of the curriculum. A less for-
mal exercise, but also serving the same purpose, is our
annual "Second-Year Design Competition."
The Competition provides students with a genuine chal-
lenge in process engineering, allowing them to show off
their ingenuity in the face of a strict set of constraints. It also
offers an entertaining spectacle for the onlookers, good cash
prizes for the winners, public humiliation for failure, and
triumph for the victors. It has become a keenly anticipated
highlight of the academic year.
In this paper we will look at some of the more successful
Competitions from the past dozen years, the organization
needed to mount them, and how they have benefitted both
the department and the faculty of engineering as a whole.
Organization Normally, the class of about eighty stu-
dents is asked to form into groups of two students each. Solo
entries are permitted, but the complexity of demonstrating

Wayne Davies took his degrees in the same de-
partment where he now combines the roles of
part-time honorary lecturer and full-time operator
of a consulting business. He has interests in
biologicals, minerals, and petroleum, and has ex-
perience in everything from antibodies to zinc and
from LPG to LPS. He would like to hear from any
readers who intend to mount a design competi-
tion. He can be reached by e-mail at
"davies@chem.eng.usyd.edu.au" or by regular
mail.

Published in an earlier form in "Chemical Engineering in
Australia," September 1995
102

[The Competition] offers an entertaining
spectacle for the onlookers, good cash prizes for
the winners, public humiliation for failure, and
triumph for the victors. It has become a keenly
anticipated highlight of the academic year.

the design usually requires two people. About a month be-
fore the Competition date and just before a mid-semester
break, each student is given a handout describing the Com-
petition rules and constraints. They then have a break of one
or two weeks in which to design their entry. In most cases
there is some equipment used to test the entries and the
essential features of this are put on display in a special
cabinet in public view. This cabinet also contains Competi-
tion memorabilia such as photographs of the past events and
actual parts of winning entries from previous years.
Prizes and Sponsorship We make a point of inviting an
industrial sponsor to the Competition. The sponsors, whose
products are in some way relevant to the Challenge, donate
prize money in return for a novel form of lunchtime enter-
tainment and the opportunity to put their company name in
front of potential clients. Sponsors have always been eager
to attend and to make some observations of their own at
prize-giving time.
Designing the Competitions "Using familiar materials
for unsuitable purposes in an impossible time frame in front
of a noisy crowd" is the basis of the design Competitions.
Many entries have been taken from the domestic environ-
ment and given a twist such as the "Egg Separator," the
"Cut-the-Soap," and "Transporting the Beer" Competitions.
Conventional process engineering has inspired three tasks in
the form of water pumping, pneumatic conveying, and heat
exchanger design. Risk engineering gave us the "Bursting
Disk" and "Extinguish the Flame." Nearly any familiar con-
@ Copyright ChE Division ofASEE 1996
Chemical Engineering Education

cept can be made into a competition topic by imparting an
awkward constraint. We never seem to be short of ideas.
We choose second-year students deliberately. These stu-
dents are still at the beginning of the course and have not yet
studied subjects such as unit operations, thermodynamics,
and control. Lacking the formal benefit of these skills, sec-
ond-year students offer refreshing novelty because they are
still essentially free from the
constraints of conventional
wisdom and are not so so- lid
phisticated in outlook that /
their designs are conservative
and predictable. This in-
creases the likelihood of the
unexpected. air space
Judging To be successful
in all respects, as a spectacle
and as an assessable effort,
the Competition must have a
definite criterion for winning
or losing, and for best visual
impact, this criterion should heated rocks
be immediately obvious to the Figure 1. Schematic dia
spectators. Devices or pro- the Bursting Disk
cesses that fail or succeed
spectacularly are best. Crite-
ria which involve lengthy calculations of some performance
index are less so. For this reason, we have endeavored to
move away from measuring an optimum such as "perfor-
mance per unit weight of entry," preferring a design to
simply work well despite, or because of, its size.
Prizes The winning team, the runner-up, and the most
deserving team receive prizes made up from the entry fee,
which is usually $5 per team, plus $300 to $400 donated by
the sponsor. The winners also receive certificates that docu-
ment their accomplishment in suitable style. Students tend to
value these certificates as much as the cash.
Academic Assessment A potential 5% bonus marks in
Chemical Engineering II are offered for participating in the
Competition. Students do not have to participate, but we find
that offering such a bonus gives them the necessary incen-
tive, which they may not have otherwise. As a result, the
majority of the class participates.

MOST MEMORABLE COMPETITIONS

The Explosion Vent Bursting Disk (1992)
The task was to design a vessel and equip it with a bursting
disk such that it would vent safely, an explosion occurring
inside the vessel. The conditions were:
The vessel would receive by injection through a port, a
precise amount of liquid fuel (e.g., 2 ml of acetone). The
explosion would then be vented safely by blowing off the
Spring 1996

integral bursting disk. The loudest report as measured by
a pair of sound pressure meters would win.
To reduce risk to the participants, all operations were
handled by a departmental technician suitably equipped with
full-face mask and gloves. To guard against the vessel al-
ready containing fuel (which might augment the sound level
achieved), the technician also fired the spark before fuel was
introduced into the vessel. A
premature explosion would dis-
can qualify the entry. To guard the
spectators, a circular plexiglass
panel open at the top was placed
around the entry.

gram of the winning entry in
Competition of 1992.

In their design, the students
had to consider several things:
a means of vaporizing the fuel
so as to achieve a good fuel/air
mixture; the volume of the ves-
sel required to achieve a good
fuel/air ratio; the area of the
bursting disk; and its firmness
of attachment to the vessel.
The winning entry was a 4-
liter paint can containing a
handful of rocks that had been

previously warmed in an oven
(see Figure 1). Foam insulation around the can helped keep
the rocks warm until the time of the test. The lid of the can
acted as the bursting disk. When tested, this entry gave a
maximum sound pressure of over 100 dB at 3 meters, with
the lid being blown vertically to a height of about 5 m.
An entry that did not win, however, was more spectacular.
It consisted of a paint-solvent can of about 4 liters and not
much else. The bursting disk was the cap of the can, which
was comparatively small at about 40-mm diameter. When
tested, the entry gave out a tremendous bang and simply
disappeared. At this point the spectators looked left and right
for the missing can, then realizing that there was only one
direction it could possibly have gone, they simultaneously
looked up to the see the can still heading upwards and
getting smaller all the while. The can naturally fell to earth
again, heading for a space that spectators hurriedly cleared.
The base of the can was found on the test table. As a result,
the entry was disqualified because the bursting disk failed to
vent the explosion safely. The judges suspected that the can
had been filled with oxygen.

The Egg Separator (1988)
A quieter competition required students to design a device
that would automatically separate the white from the yolk of
an egg. A whole unbroken egg would be inserted into the
device, and all operations after that would occur automati-
cally, with the result that the white and the yolk would be

insulation

- spark plug

delivered into two waiting recepticles, e.g., two petri dishes. The
students had to design a device that would break the egg and allow
its contents to flow and to separate as they did so. Understanding
and exploiting the properties of unusual fluids was the major
challenge of this exercise.
Several entire failed because their operation required manual
intervention, most often to break the egg. Some failed because
the egg would not flow across the separating device, or if it
did, the yolk broke.

Figure 2. Schematic diagram of the winning entry in the
Separate the Egg Competition of 1988.

There was a clear winner, however, who surpassed all expecta-
tions. It satisfied the design conditions in every detail and ex-
ploited the properties of the materials beautifully. In this design
(see Figure 2), the egg was dropped via a chute onto a metal
breaking edge pointing upward. Contact broke the egg, which then
flowed down the sides of the edge onto a sloping table. Two
barriers were positioned on the sloping table such that the flow of
yolk and egg would continue by gravity. The first barrier was an
underflowing weir that held back the yolk while the white flowed
underneath, while the second barrier captured the white. Both
barriers allowed the yolk and white to continue their flow to
the edge of the sloping table where they were collected into
separate petri dishes. When tested in front of an enthusiastic
audience, the final plop of the yolk into its dish was greeted
with thunderous applause.

The Bead Elevator (1991)
Planning, timing, and teamwork were especially required in this
competition. The aim was to design a mechanical means of con-
104

veying foam beads from a ground-level feed bin to a
receiver bin located on a platform 2 m above the
ground. The fastest time to convey more than 95%
of the beads, or failing this, the greatest quantity of
beads moved in a 30-second period would be the
winner. Students would need to understand the na-
ture of pneumatic transport as well as to optimize
the efficiency and operation of their design.
Many designs based on vacuum cleaners appeared on
the day of competition, but none was as well calculated
to succeed as the eventual winner. In most entries, an
operator applied suction to the receiver bin via a hose
inserted into a removable lid. Another hose was let
down to the feed bin where the other operators could
manipulate it. Ideally, the resultant air movement was
supposed to suck up the beads and deposit them into
the receiver bin. Several problems emerged with these
designs, however. Frequently the hoses, being fairly
narrow, clogged up with beads. In some cases the
vacuum applied to the receiver bin was too great and

Figure 3. "They should have won!" Students
face defeat graciously in the Bead Transport
Competition (1991).
Chemical Engineering Education

collapsed it by sucking it in. Hoses would fall out of their
intended mountings, or the weight of the hose would pull the
receiver bin over and spill the beads. Some entrants found
that the beads did not remain in the receiver bin but contin-
ued on into the vacuum cleaner. This did no good for the
flow of air, especially if the cleaner's dust bag had been
removed. The most valiant attempt that did not work was

Figure 4. Encouragement Prize winners of
the Extinguish the Gas Flame Competition
(1994) brave the heat with the elegant nozzle-
extension-on-a-pole.

A~ _
Figure 5. Winners of the same competition show their style w.
and gas bottle. The flame was put out in under five s
Spring 1996

based on an industrial blower. In it, the feed bin was pressur-
ized and a large-bore hose conveyed beads to the receiver.
Unfortunately, the pressure applied was too great for the
flimsy plastic bin and its lid blew off in a white blizzard (see
Figure 3). The entrants, who showed great courage in the
face of defeat, received the Encouragement Award.
The winners had a different approach altogether. Their
design was based on a rotary-motor mower in which a light-
weight fan was substituted for the normal blade. A card-
board ducting system was attached beneath the mower so
that the beads could be neatly sucked up. The outlet of the
mower (where the grass is normally ejected into the catcher)
was fitted with a duct that led to a lid fitted over the receiver
bin (which was firmly held down by an operator). This lid
was fitted with a gauze mesh to allow air out of the bin while
retaining beads within. After starting the mower, the whole
operation was over in eighteen seconds, and the recovery of
beads was nearly perfect. This winning entry demonstrated
the advantage of establishing clear design parameters and
good teamwork, as well as thorough planning and testing.
A joke entry came in the form of a firework that was
placed in the feed bin. The resulting explosion blew beads
out of the bin and in a total duration of twelve seconds, half a
handful was transported to the receiver bin. This entry did
not come in last.

Extinguish the Gas Flame (1994)
This was possibly the most entertaining spectacle of all.
The aim was to extinguish a luminous flame of LPG emerg-
ing from a 25-mm nozzle at full cylinder pressure. This
competition was advertised as providing "a serious flame"
later estimated at about 40 kwatts of heat. Competitors were
allowed within a minimum radius of 3.5 m of the flame and
were given thirty seconds. The fastest
Time to extinguish the flame would be
the winner.
B Many entrants tried to smother the
flame with a variety of devices held by
poles over the nozzle. Many devices
1 caught fire, and others did not succeed
when the flame found a way around them
and ignited elsewhere. Some attempted
V S to pour water down the nozzle, but the
So .,. :. gas pressure simply blew it back out. An
ingenious entry (winner of the Encour-
agement Award) used a piece of pipe
that fitted neatly over the nozzle and ex-
i tended it some 400 mm (see Figure 4).
Manipulating this pipe attached to the
S end of a pole, the entrants placed it over
S" "' t'-*" *"" the nozzle, causing the flame to jump to
ith baby stroller the end of the extension. The entrants
seconds. then quickly jerked the extension upward
105

and off the nozzle, lifting the flame away from
the gas source. The flame went out accompanied
by loud acclamation from the audience. Unfortu-
nately, the unwieldy apparatus was difficult to
manipulate at the distance involved and the en-
trants took too much time.
The winning team carried out their task with in-
souciant ease. Their entry consisted of a bottle of
carbon dioxide gas resting in a baby stroller that
could be pushed by a long pole (see Figure 5). One
member opened the valve, allowing the gas to flow,
while the other poled the stroller out to the flame.
In less than five seconds the flame was out. Al-
though the crowd was greatly impressed, there was
some resentment from the other contestants since
the competition conditions stipulated that no com-
mercial fire extinguishers were to be used. Indeed,
this entry used a commercial principle, but it was
not itself a commercial extinguisher-so it won.

Antigravity Water Transfer (1995)
The Competition for 1995 was based on moving
water from one vessel to another using a pump.
Normally, this is a routine task, but there was a
catch. Entrants could not put any external power to
the pump, so all of the energy for pumping the
water had to be derived from the head of water
itself. Water was supplied in an 800-liter tank, filled
to the brim about 1.6 m above ground level. A
receiver was arranged so that its entry point was at
the same height as the water in the tank. Obviously,
siphons would not work.
Many entrants designed variations of the hydrau-
lic ram. This well-established invention inspired by
the phenomenon of "water hammer"'" converts the
momentum of a falling water column into pressure
energy when the flow is suddenly stopped by a
valve. The increase in pressure then elevates a frac-
tion of the water to a higher level. Because the
hydraulic ram is readily found in textbooks, the
organizers expected most entries would be based
on this principle, and, in fact, the winner was. One
of the delights of the Competition, however, is be-
ing amazed by the unexpected and, as hoped for,
several entries exploited entirely different principles.
One entry (which won second place) caused fall-
ing water to pressurize the air in a vessel of about
20-liter capacity situated at ground level. This in-
creased air pressure was transmitted via a tube to a
second vessel at a higher level that had been previ-
ously filled with water from the tank. The air pres-
sure then pushed water out of the second vessel into
the receiver. Other less successful but amusing en-
106

tries included a flimsy cardboard-and-rubber-band turbine that turned
half a dozen times, got waterlogged, and stopped, meanwhile giving the
operators a bath (see Figure 6), a great slow-working piston pump,
reminiscent of a Boulton and Watt steam engine, which did not even fill
up in the thirty seconds, and an unofficial entry by graduate students
that used solid carbon dioxide pellets and a lid on the tank to pressurize
the water, forcing it directly to the receiver (shown in Figure 7). Al-

Figure 6. Typical of the more whimsical entries in this cardboard and
rubber-band turbine pump seen in the Water Transport Competition
(1995). Despite the team's well-drilled display, no water entered the
receiver.

Figure 7. Unofficial entries often appear with the intention of getting
around the rules. In the same competition and amid clouds of CO,
vapor, graduate students successfully put a lid on the tank and
transport 8 liters of water in 30 seconds.

Chemical Engineering Education

though the resultant flow was spectacular, using the heat
energy of the water was a too-clever interpretation of the
rules and it was disqualified.
The winning entry, based on the hydraulic ram principle,
was a well-researched and well-designed device that trans-
ported 2.6 liters of water in thirty seconds. The winners had
consulted an engineer (actually, one of the team's parents),
and having gotten the general idea, made their entry from
PVC pipe and fittings, a 2-liter plastic soft-drink bottle, a
squash ball, a marble, and a piece of garden hose. Unlike
many of the others, this team hardly got wet at all.

A COMPETITION FOR VISITORS
A related competition is held from time to time for visitors
to the department on University Open Days. This competi-
tion requires a more spontaneous approach, and because
there is no time for the contestants to design and prepare an
entry, construction materials are provided and the assigned
task is simple.
Our favorite example involves a hot-air balloon for which
contestants must design and build a burner of greatest effi-
ciency. This burner contains a small quantity (say, 25 ml) of
fuel, such as ethanol. The burner is attached by metal wires
to the inside of the hot-air balloon and the fuel is ignited. The
longest duration aloft wins. The balloon, made from plastic
foil, is only 2 m tall and slides up and down a taut wire inside
an atrium within the building. Typical materials provided
are aluminum foil, wire, cotton wool, scissors, pliers, etc.
Obviously, the lighter the burner, the less is the effort
required to lift it, but the more flimsy it is. The shape of
the burner is important in determining the time aloft. An
initial burst of heat is generally required, and heat should
then taper off for maximum duration. Several minutes
aloft is not uncommon.
Typically, the response to this competition overwhelms
the organizers. On our first attempt, there was a line of
contestants waiting for thirty or more minutes to have a go at
it. In subsequent years we had four balloons going at once
and still had no time to relax. Entries were widely variable
and imaginative, but typically the simplest designs did best.

DISCUSSION
Students take the challenge seriously, using imagination
and intuition together with some formal engineering to de-
vise a wonderful range of exotic devices. The Competition
not only broadens their scope but also gives them an excel-
lent excuse to have some fun. It also performs a service in
socializing students. For most, this is their first time in a
public exhibition in front of their peers. Although there may
be some degree of humiliation in defeat, this soon passes as
the rewards, as ever, are in participating.
Other Competitions not described in this paper have been
based on various examples of process unit operations. They
Spring 1996

have included optimizing a distillation rig for producing drink-
ing water (1987), operation of a precision soap-cutting device
(1989), transporting beer using only the pressure in the can
(1990), and optimizing a simple heat exchanger (1993).
Originally, most Competitions involved a means of trans-
porting matter or energy by some method that could be
optimized. More recently, two of the Competitions have
been based on the area of risk in the process industries (e.g.,
the Bursting Disk and Extinguish the LPG Flame). There is
an obvious extra level of excitement in this type of competi-
tion and we will continue to include them, mindful always of
the necessary safety precautions.
Despite the general levity of the occasion, there is usually
some scholarly relevance. In the bursting disk competition
of 1992, the solvent can that exploded and rocketed was
noted to have relevance to a major industrial fire in which a
solvent storage tank ruptured at its base, similarly exploding
and rocketing.[21 The 1994 Competition to extinguish the
LPG flame had relevance to the oil fires in Kuwait after the
Gulf War. Interestingly, the winning entry used a similar
principle to extinguish the fire that was used by the team of
experts on the real thing.
One lesson might be, "Do your research carefully, espe-
cially if deception is the aim." One of the entries in the soap-
cutting Competition of 1989, a mysteriously modified card-
board box produced two perfect halves from a whole cake of
soap in record time and looked like winning. But it was not
to be. A spirited audience observed, and the judges con-
firmed, that the cake inserted into the device was pale pink
while the two halves that emerged were white!
The Competition's real message is, I like to think, that
Experience is the best teacher. Until they reach the univer-
sity, nearly all of the students' academic experiences have
been theoretical. The concept of actual catastrophe never
seems to emerge. In the real world, however, catastrophe is
always ready to exploit the unready. Perhaps most students
do not realize the Competition's lessons at the time, but
some will in the future-we hope, to their benefit.

ACKNOWLEDGMENTS
I would like to especially thank the workshop and techni-
cal staff of the department who have given many hours of
their time to build equipment, supervise the Competition,
and enforce safety requirements on the day. In addition,
learned colleagues Dr. T.A.G. Langrish, Prof. D. Reible, and
Dr. B. Walsh, acting as masters of ceremonies, have given
invaluable support. The Competition flourished under the
championing of Prof. R.G.H. Prince.

REFERENCES
1. Smith, B.E., "Building a Model Hydraulic Ram," Modeltec,
pg. 8, April (1995)
2. Thomas, F., "Coode Island: Vapour Recovery to Blame?"
The Chemical Engineer, 506, pg. 17, October (1991) 0
107

classroom

DEMONSTRATIONS TO COMPLEMENT

A COURSE IN

GENERAL ENGINEERING

THERMODYNAMICS

DOUGLAS J. DUDGEON,* J.W. ROGERS, JR.
University of Washington Seattle WA 98195-1750

At the University of Washington, the College of En-
gineering offers a lower-division course in general
engineering thermodynamics. The course is part of
the core engineering curriculum that students are supposed
to take prior to applying for admission to a department. As
such, it serves many departments in the College, most of
which provide instructors for the course. Chemical and me-
chanical engineering students take the course in their sopho-
more year to satisfy departmental admission requirements
and to serve as a foundation for further study of advanced
thermodynamics in their respective programs. Students from
other departments take the course to satisfy specific gradua-
tion requirements or to fulfill their elective engineering sci-
ence credits. These students often do not enroll in the course
until late in their senior year; thus the sections usually con-
tain a very broad cross-section of students in various stages

Doug Dudgeon received his BS in chemical
engineering from the University of California, Ber-
keley (1987), and his PhD from the University of
Washington (1995). At the UW he was twice
awarded the McCarthy Prize for Excellence as a
Teaching Assistant and was involved with the
ECSEL project. His research interests include
applications of non-Newtonian fluids and deter-
mination of these fluids' rheological properties.

J.W. (Bill) Rogers, Jr., is Professor of Chemical
Engineering at the University of Washington. He
received his BS in Chemistry (1975) and PhD in
Physical Chemistry (1979) from the University of
Texas. He spent eleven years at Sandia Na-
tional Laboratories doing basic and applied re-
search on energetic materials, thin film deposi-
tion and modification, and finally as Supervisor
of the Ceramic Development Division, before
joining the ChE Department at UW in 1991.

Address: Kraft Foods Technology Center, 801 Waukegan Road,
Glenview, Illinois 60025-4312
108

Engineering Thermodynamics Schedule and Layout

WuA Ta

1 Bic concepts -Congtat-volumne thermomeny
-2 --1 pdua.. a- af~ bon-- dioxide

-t

5 2dhl h Airpomna 4 ltcattflpaoit o foair
(closE sd donm m}
6-74Ennapyctwtffiaamzy Emo ffi ciecaof -compressor
_t-9 Pow6rad-rcfufigmatncyda t iatp

of their undergraduate careers.
Engineering thermodynamics is a four-credit course that
meets for three 50-minute lectures and one 110-minute quiz
section per week. Prerequisites are two quarters of freshman
chemistry (general), three quarters of freshman calculus with
analytical geometry, and one quarter of freshman physics
(mechanics) with a laboratory. The course content, outlined
in Table 1, includes
Concepts of units and dimensions, pressure, tempera-
ture, heat, and work
Macroscopic properties of substances
Principles offirst-law analysis for closed systems
Principles of energy analysis for open systems,
including flow and shaft work
Concepts of the second law of thermodynamics in its

@ Copyright ChE Division ofASEE 1996
Chemical Engineering Education

DaUonslburtn-

.. this class ideally should have a laboratory to accompany the lecture material.
This would give students hands-on experience with the concepts presented in the lectures
and would also expose them to engineering devices. Budget, space, and time considerations prohibit
this, but we have found a reasonable substitute in a set of classroom demonstrations
that can easily be integrated into the course sequence.

macroscopic form for open and closed systems and
engineering devices
Power and refrigeration cycles

The circumstances under which the course is offered make
it pedagogically challenging. First, the various departments
wish to emphasize topics that are relevant to their own
program. For example, one department wants heavy empha-
sis on thermodynamic cycles, while another wishes to em-
phasize psychrometrics. As a result, it is difficult to obtain a
consensus between the departments on course content, and it
is difficult to coordinate among the various instructors to
ensure a uniform coverage of the material. Coordination
with local community colleges, which also offer the course
and from which we receive transfer students, further compli-
cates course administration. Second, there is a universal fear
of the course material across the spectrum of students who
take the course. Indeed, there are many new and unfamiliar
concepts introduced in the course. Terms such as enthalpy
and entropy may have been encountered before, but they are
usually not well understood and it is difficult for the students
to grasp such intangible concepts. This is compounded by
the third problem, which is that few
of the students have any mechanical
skills and most have had very little
laboratory experience. Because the
material has a strong emphasis on
engineering devices, many students T
find it difficult to comprehend how
some of these devices function.
Fourth, the course is very crowded
and has little room for new material. H

To improve comprehension, this
class ideally should have a labora-
tory to accompany the lecture mate-
rial. This would give students hands-
on experience with the concepts pre-
sented in the lectures and would also
expose them to engineering devices.
Budget, space, and time consider-
ations prohibit this, but we have
found a reasonable substitute in a
set of classroom demonstrations
that can easily be integrated into
the course sequence. We present
below the set of demonstrations that
were developed under the auspices
Spring 1996

Figure 1. Schematic di
constant-volume the

of the National Science Foundation's "Engineering Coa-
lition of Schools for Excellence in Education and Lead-
ership" (ECSEL).
The set consists of demonstration hardware, notes for in-
structors, and student handouts where necessary. The dem-
onstrations were developed with the concept that "seeing is
believing" and that some familiarity with engineering de-
vices will improve understanding and comprehension of the
lecture material. Our criteria for developing the demonstra-
tions were that they had to be inexpensive, easy to construct,
portable, and completable in less than fifteen minutes of
class time. In addition, each had to convey several key
concepts from the lecture material. In the two years the
demonstrations have been used, they have received excellent
reviews from students and faculty alike.

> Constant-Volume Thermometry 4
This classic apparatus may be used to demonstrate the
absolute temperature scale and the ideal gas law. The con-
stant-volume thermometer is based on Gay-Lussac's law
(1802) that the temperature of a gas is proportional to its
pressure at constant volume. Gay-
Lussac' s law is, of course, a special case
of the ideal gas law. The goals of this
demonstration are to illustrate Gay-
Lussac's law and to estimate the value
of the temperature at zero pressure, i.e.,
the absolute zero of temperature.
The apparatus (see Figure 1) consists
of a thin-tube water manometer, a stop-
pered 125-ml Erlenmeyer flask, a 1000-
ml beaker, a digital thermometer, and a
supporting stand. One end of the ma-
nometer is open to atmosphere and the
other is connected to the flask through a
short piece of Tygon tube. For best
results, it is important to use a thin-tube
manometer (diameter less than 5 mm)
and a short piece of tube to connect the
manometer to the flask to minimize the
volume of air that is not in the water
bath. In addition to the apparatus, the
instructor must supply approximately
2000 g of ice.
agram of the In order to establish the P-T propor-
rmometer. tionality, at least three data points are
109

required, although it is easy enough to acquire more points.
For the first point, the beaker is filled with room-temperature
water and the flask is immersed in the beaker. After allowing
the system to come to equilibrium, the pressure on the ma-
nometer and the temperature of the water bath are measured.
For additional points, ice is added to cool the water to a
lower temperature (it is usually necessary to pour off excess
water). The final data point is obtained using a saturated ice/
water solution (273 K, approximately 0C). Of course, the
system is not at constant temperature except at the end points
of 273 K (00C) and room temperature (approximately 295
K). But the change in temperature is sufficiently slow that its
contribution to experimental error is negligible.
Gay-Lussac's law may be expressed as T = a + bP, where a
and b are constants. Thus, a plot of the P-T data should give
a linear result. Extrapolation to P = 0 gives a measure of
absolute zero. With our system, using the procedure outlined
above, consistently we have been able to come within 20 K
of the true value, which is remarkable due to the simplicity
of the apparatus and the large extrapolation involved. The
demonstration also illustrates the use of a manometer to
measure pressure and the concepts of differential, absolute,
and gauge pressures.

D Critical State of Carbon Dioxide A4

This demonstration allows a pure, two-phase mixture to be
observed as it passes through the critical point. The appara-
tus was developed in 1959 by an undergraduate student at
the University of Washington's Department of Mechanical
Engineering;[" we merely added it to the current package of
demonstrations. It consists of a sealed quartz tube containing
a two-phase mixture of carbon dioxide (CO2). The mixture is
sealed into the tube at the critical molar volume (0.0943 m3/
kmol), and when the contents are heated from room tem-
perature (state 1) to a higher temperature (state 2), the mix-
ture passes through the critical point (304.2 K, 7.39 MPa) as
shown on the property diagram depicted in Figure 2. The
meniscus is observed to pass into critical opalescence, then
to disappear.
The tube is conveniently heated by placing it in a fixture
that fits into the focal point of a surplus 35-mm film strip
projector. Heat from the 150-W projection bulb raises the
temperature of the mixture such that it reaches the critical
state in about ten minutes. The optics of the projector are
used to focus the meniscus onto a screen for viewing by the
entire class. Similar designs have been reported recently in
this journal.121 This demonstration illustrates the critical prop-
erties of a pure substance, the structure of property and
phase diagrams, and the concept of quality for a two-phase
mixture.
Because the sealed quartz tube contains a fluid at very

high pressure, this demonstration poses a significant explo-
sion hazard, so appropriate safety procedures should be ob-
served. As a minimum, the instructor should wear safety
glasses, transport the tube in a shielded case, and quickly
mount the tube into the projector (preferably before students
are present). Once the tube is mounted, the geometry of the
projector provides a blast shield against the effects of an
accidental explosion. We suggest that before preparing a
tube for this demonstration, interested readers should con-
sult relevant literature[12 and select a fluid with a lower
critical pressure.

> First Law for an Open System 4

This demonstration is the realization of a problem from
the _engel and Boles' textbook, Thermodynamics: An Engi-
neering Approach, which is currently used for the course.13,'41
The first law is used to determine the mass flow rate of air
exiting a hair dryer. The apparatus consists of a hand-held
hair dryer mounted on a stand and a Chromel-Alumel (type-
K) thermocouple with an attached digital thermometer. The
thermocouple tip should be located approximately 7-8 cm
from the dryer exit and in the middle of the flow (radially).
The tip should be no closer to the dryer exit since the tem-
perature distribution of the air nearer the exit is not uniform
and radiative heating of the thermocouple by the dryer's hot
filament becomes a problem. The hair dryer is operated for
two or three minutes to reach steady state, and the outlet (T2)
and inlet (T1) temperatures are measured.
This is a simple application of the first law
Q W = ri(Ah + Ake + Ape) (1)
where
Q W rate of heat transfer and work, respectively

I T(K)

Figure 2. The T-v property diagram for CO2.
Chemical Engineering Education

304

295

0.0943

v(m3/kmol)

m mass flow rate
Ah, Ake, Ape change in specific enthalpy, kinetic energy, and
potential energy, respectively
We make a number of assumptions for the analysis: negli-
gible heat loss through the dryer's walls, negligible changes
in kinetic and potential energy, steady state and uniform
flow, that air is an ideal gas under these conditions, and that
its heat capacity is constant over the measured temperature
range. This gives

-W = mAh = rhC (T2 T ) (2)

where Cp is the heat capacity of air.
We further assume that negligible electric energy is
consumed by the dryer's fan so the power delivered to
the air is the electric power into the dryer, which was
measured as 1176 W (+5%) out of class using a voltme-
ter and a clip-on ammeter. Typical temperatures mea-
sured are
T, = 220C (295K) and T, = 720C (345K)
so that the mass flow rate works out to
rm 0.023 kg / s

or, on rearranging,

-w
m=-
Cp(T2-T )

Figure 3. Apparatus for measuring the
heat capacity ratio of air.

p

3

2

V

Figure 4. Pressure-volume diagram for
the heat capacity ratio experiment.
Spring 1996

This value is within 4% of the mass flow rate measured with
a velometer. Because the cross-sectional area of the exit can
(3) be measured, the mass flow rate can easily be manipulated to
give the average gas velocity and the volumetric flow rate to
illustrate the relationship between these quantities, in addi-
tion to illustrating the first law. The principles and opera-
tion of a thermocouple are also shown, and if the thermo-
couple at the exit is moved radially, the assumption of
uniform flow used in Eq. (1) can be relaxed and nonuni-
form flow can be demonstrated.

> Heat Capacity Ratio for Air 4

This demonstration is a simplified version of an experi-
ment given by Shoemaker, et al.'51 The goal is to determine
the heat capacity ratio

for air near standard temperature and pressure. The demon-
stration gives students experience with properties of an
ideal gas, adiabatic processes, and the first law. It also
illustrates how P-V-T data are used to measure other
thermodynamic properties and how to measure differen-
tial pressure using a manometer.
The apparatus is illustrated in Figure 3 and consists of a
stoppered, 19-liter (five-gallon) carboy and an open-tube,
water-filled manometer. The carboy stopper is punctured by
two pieces of glass tubing for connection to Tygon tubing.
One piece of tubing connects the carboy to the manometer,
while the other is used to pressurize the carboy through a
hose clamp. The water in the manometer is tinted with food
coloring for better viewing.
The experiment involves a two-step process (see Figure
4):

Process 1-2: Expand the gas in the carboy adiabatically and
reversibly from P, to P,.

Process 2-3: At constant volume, allow the gas to return to
thermal equilibrium as its pressure changes to P .

Iff

The system is initialized by opening the vent
clamp, blowing into the carboy, and then closing
the vent. An initial pressure of 16-18 inches H20
(6.4 to 7.2 x 10-2 Pa) is generally sufficient. The
system is then allowed to reach thermal equilibrium
(about fifteen minutes). This preparation may be
done outside of the actual class time since it is not a
part of the process outlined above.
For the demonstration, the initial pressure P, is
read on the manometer. The carboy is momentarily
unstoppered and then the stopper is replaced. This
allows the gas in the carboy to expand as the pres-
sure drops to atmospheric (P2). Note that this ex-
pansion is approximately adiabatic and reversible.151
After 10 to 15 minutes, the system returns to ther-
mal equilibrium once again and the final pressure
P3 is read on the manometer.
The analysis, which comes directly from Shoe-
maker, et al.,151 is lengthy but straightforward, and a
handout is provided for the students. It will not be
repeated here, but starting with the first law and the
assumption of ideal gas behavior, the heat capacity
ratio may be related to the measured pressures
Cp fn PI -_ n P2
7 -=- (4)
C, enP n P3

Although our apparatus is crude in comparison to
that detailed by Shoemaker, consistently we are
able to determine y for air to within 5% of the
reported value of 1.40. 31

> Efficiency of a Compressor 4

The goal of this demonstration is to determine the
second-law efficiency of a small compressor. Dis-
cussion may also center around the operation of a
compressor (actually a pump, in our case), a Bourdon
pressure gauge, and/or a gas flow meter.
The apparatus, shown schematically in Figure 5,
consists of a 40-W reciprocating-piston pump that
compresses air into a 3.5-liter tank. (The type of
compressor is not important; any suitable compres-
sor available from the university salvage pool will
work.) The gas flow rate of the air exiting the tank
is measured by a rotameter and controlled by a
valve on the rotameter; this valve also controls the
pressure in the tank. The tank pressure is measured
with a Bourdon gauge, and a digital thermometer is
used to measure the temperature both inside the
tank and in the external environment.
In our apparatus, the pump generates pressures
from 0 to 6 psig (101 to 142 kPa) and mass flow
rates of air from 0 to 0.33 g/s, drawing approxi-

mately 40 W electrical power for any setting. The power consumption
was measured (outside of class) to 5% using a wattmeter. Operat-
ing at steady state, we observe no significant change in tempera-
ture between the inlet and the tank, indicating that this system is
essentially isothermal.

Figure 5. Schematic diagram of the apparatus for
measuring the efficiency of a compressor (pump).

Figure 6. Photograph of the heat pump system showing (0) the
compressor, (1,2,3) the condenser, (4) the dryer and expansion
valve, and (6,7) the evaporator.
Chemical Engineering Education

For purposes of analysis, the compressor comprises the
thermodynamic system (control volume) and the working
fluid is air. The second-law efficiency for a compressor is
the ratio of the reversible work to the actual work

Wrev
w= (5)

The actual work, W, we know to be -40 W. To obtain the
reversible work, we start with the first law for a steady-flow
system

Q Wrev = m(Ah + Ake + Ape) (6)
and the second law for a reversible steady-flow system
(Sgen =0)

rinAs=- (7)
To

These expressions are combined to eliminate Q. Neglect-
ing changes in kinetic and potential energy, we arrive at the
following expression relating the reversible work to changes
in enthalpy and entropy:

WreN = m (ToAs Ah) (8)
If we assume an ideal gas, then the changes in enthalpy
and entropy are easily calculated from the heat capacity of
air and the measured temperature and pressure. Thus, the
reversible work and the second-law efficiency of the com-
pressor may be calculated. Typical values for our setup are
Wrev =-6W giving Tn = 0.15. This is one of the few demon-
strations that conveniently allows students to observe and
quantify a change in entropy.

> Heat Pump 41

Power and refrigeration cycles are an important topic of
this course, and we have a working heat-pump cycle demon-
stration to illustrate the main principles of thermodynamic
cycles. The heat pump was fabricated from a surplus refrig-
erator compressor, two cross-flow, air heat exchangers, and
an expansion valve (see Figure 6). The heat-exchanger com-
partments of the condenser and evaporator are covered with
Plexiglass and several of the heat-exchanger tubes are con-
structed from glass to observe the changes in phase of the
working fluid, refrigerant-12 (R-12) as it transits the cycle.
Pressure gauges and thermocouples are included between
processes in the cycle so that students can observe pressure
drop and temperature changes associated with each process
and compare it with property changes predicted from the
property diagram of the refrigerant.
During operation, the evaporator operates at a pressure of
586 kPa (85 psia), the condenser pressure is 974 kPa (141
psia), and the temperature exiting the compressor is 357 K
Spring 1996

(1830F). From these values and the R-12 property tables or
diagrams, the students can calculate the coefficient of per-
formance for the heat pump, the work supplied to the com-
pressor assuming isentropic operation, the irreversibility as-
sociated with the expansion valve, etc. Discussion can also
center around changes in expected performance if R-12 is
replaced by the more environmentally friendly refrigerant-
134a (R-134a), assuming the same amount of work is sup-
plied to the compressor.
This demonstration does not fit in the same category as
those discussed above; it is portable (but not small) and
inexpensive, but relatively complicated to construct. It was
constructed several years ago in the Department of Mechani-
cal Engineering and was available for use in this class, so we
have included it to round out the demonstration set.

EVALUATION
The demonstrations were designed to supplement each of
the major sections of the course as shown in Table 1. The
addition of the demonstration set to the lecture material was
evaluated by two sections of students who took our offering
of the course. The evaluation took the form of a set of
questions prepared by the UW Office of Educational Assess-
ment and the ECSEL assessment team. The students were
asked to respond to four or five questions concerning the
demonstrations and how helpful they had been in en-
hancing their understanding of the underlying thermody-
namic concepts. On a scale of 0 to 5, the pooled average
and standard deviation were 3.76 and 1.01, respectively,
with 53 students responding. Verbal feedback from fac-
ulty who used the demonstrations has been uniformly
positive.

ACKNOWLEDGMENTS
This work was conducted under the auspices of the Na-
tional Science Foundation's ECSEL (Engineering Coalition
of Schools for Excellence in Education and Leadership)
Program, and the authors gratefully acknowledge their sup-
port. We also thank Professors Creighton Depew and Ashley
Emery of the UW Department of Mechanical Engineering
for development of the heat-pump system and the critical-
point apparatus, respectively.

REFERENCES
1. Leichester, J.R., University of Washington (1959)
2. Marcotte, R.E., L.C. Zepeda, and D.L. Schruben, Chem.
Eng. Ed., 28(1), 44 (1994)
3. Qengel, Y.A., and M.A. Boles, Thermodynamics: An Engi-
neering Approach, McGraw-Hill Publishing Co., New York,
NY (1994)
4. Shakerin, S., Am. Soc. of Mech. Engs., 20, 63 (1990)
5. Shoemaker, D.P., C.W. Garland, J.I. Steinfeld, and J.W.
Nibler, Experiments in Physical Chemistry, McGraw-Hill
Publishing Co., New York, NY (1981) 1

" computers in education

-CESL-

The Chemical Engineering Simulation Laboratory

DAVID A. KOFKE, MARC R. GROSSO, SREENIVAS GOLLAPUDI, CARL R.F. LUND
State University of New York at Buffalo Buffalo, NY 14260-4200

Engineering and science research today are conducted
within an emerging paradigm in which theory, ex-
periment, and computer simulation play distinct but
equally vital roles. Progress is often made in leapfrog fash-
ion as each leg surmounts hurdles that have stalled the other
two. Thus application of one technique never diminishes the
role of the others, but rather enhances them.
The situation in the realm of science and engineering
education is somewhat less advanced. Undergraduate in-
struction for decades has relied on a two-pronged approach
of classroom and laboratory experiences. Classroom lectures
convey concepts, while laboratory provides the students with
physical experience-it exposes them to valves, gauges, flow-
ing fluids, and generally, real-life operating equipment. Labo-
ratory also teaches the students how to perform and analyze
experiments, and well-designed laboratory exercises teach
them how to plan experiments as well. Laboratories teach
the limits of experiments, analysis of error, the importance
of significant figures, and application of the models pre-
sented in the classroom.
Classroom and laboratory experiences are each irreplaceable

David A Kofke is Associate Professor of Chemical Engineering at SUNY
Buffalo. He earned his PhD in chemical engineering from the University of
Pennsylvania and his BSChE from Carnegie-Mellon University. His re-
search interests are in molecular thermodynamics.
Marc R. Grosso served as manager of the CESL project. He earned his
PhD in Learning and Instruction from SUNY Buffalo in 1994. He also
holds an MSE in Computer and Information Science from the University of
Pennsylvania, a BS in Information Systems Management from Buffalo
State College, and BSEd and MA degrees in Secondary Education. His
professional interests are in the application of computing in instruction.
Sreenivas Gollapudi holds a MS in chemical engineering from SUNY
Buffalo and is presently pursuing a MS in Computer Science. His BS
degree in chemical engineering is from lIT Bombay. His research inter-
ests are parallel systems and multimedia.
Carl R.F. Lund is Associate Professor of Chemical Engineering at SUNY
Buffalo. He earned his PhD in chemical engineering from the University of
Wisconsin and his BSChE from Purdue University. His research interests
are in catalysis and reaction engineering.

Copyright ChE Division ofASEE 1996

components of undergraduate engineering education. Nev-
ertheless, they have shortcomings:
Space, safety, cost, and time considerations restrict the choice
of laboratories
Class sizes often preclude direct participation by students
Laboratories must be maintained
Class demonstrations are operated by the instructor
It is difficult to make laboratory experiences substantially
different for each student
Lecture examples and homework often sample only a "small
corner" of parameter space
Laboratories are difficult to disseminate; the mere description
of a well-designed lab does not suffice for someone else to
implement it.

A more fundamental drawback of the classroom and labo-
ratory is their uncertain ability to instill physical intuition (as
opposed to physical experience, which laboratory does well).
Rarely is the laboratory a truly interactive exercise. The
student conducts a series of pre-planned experiments and
heads home to perform the analysis. This experience does
not leave the students with an intuitive feel for the nature
of the process. Likewise, classroom instruction is inter-
active in the sense of instructor-student, but it is not in
the sense of student-process; classroom instruction is
akin to teaching bicycling through the use of force and
torque balances.
Shortcomings of the classroom and the laboratory can be
alleviated through proper use of computer-based instruction.
In chemical engineering, substantial progress has recently
been made in this direction. A group at the University of
Michigan"' has produced a set of tutorial modules that ad-
dresses topics across the chemical engineering curriculum.
This we view as a valuable tool directed at the shortcomings
of the classroom. Software that focuses on the laboratory
also exists. In particular, a group at Purdue University'21 has
created a suite of modules that lets students perform pilot-
Chemical Engineering Education

scale laboratories on the computer. Additionally, a host of
packages has been developed for more specialized topics
in chemical engineering, such as process control. Indus-
trial simulation packages (e.g., Hysim, Aspen) are used
routinely and effectively, although this software has not
been developed with an eye toward pedagogy. We feel
that the potential of simulation as a tool for education is
largely unfulfilled.
Indeed, the recent literature in engineering and science
education journals has highlighted the tremendous potential
of computers as a pedagogical tool, while at the same time
lamenting the degree to which this potential is not being met.
Seiderl31 (prior to the efforts at Michigan and Purdue) noted
that in chemical engineering, instructional computing has
kept pace with the profession only in the areas of process
design and control. Many authors13-61 have noted two ob-
stacles to the complete integration of computers within
the engineering curriculum: the absence of powerful but
inexpensive computers with strong graphics capabilities
and the high cost (in terms of faculty time) of software
development. The steady improvement of computing hard-
ware has made the former a problem no longer. The latter
obstacle is our target.

WHAT IS CESL?
At SUNY Buffalo we have developed a detailed plan and
completed early development of software that enables edu-
cation via simulation; we call our package CESL (pronounced
"Cecil"). Seven department faculty have been active on this
project (these include, in addition to two of the authors of
this report, Scott Diamond, Johannes Nitsche, T.J.
Mountziaris, Tom Weber, and Mike Ryan). CESL is de-
signed to perform three functions:
It is an authoring tool-CESL provides instructors with
the ability to construct simulations with relative ease.
It is an environment for conducting simulations-CESL
permits the student to explore a process with a minimum
of unnecessary effort.
> It is an instruction and class management tool-CESL
allows the instructor to monitor, guide, record, and
sometimes restrict the student's actions.
While all these features are inherent in its design, to date
CESL has been developed only to a level that has permitted
three prototype simulations to be implemented. In particular,
many of the capabilities related to classroom use are de-
signed but not yet coded.

SIMULATION
Once one considers simulation as part of the educational
paradigm, one begins to realize how naturally and substan-
tially it complements the laboratory and classroom experi-
ences. As a simulator, CESL is much more than a simulated
laboratory; it does more than just port the traditional lab to
Spring 1996

Many authorst13-6 have noted two obstacles to the
complete integration of computers within the
engineering curriculum: the absence of powerful
but inexpensive computers with strong graphics
capabilities and the high cost (in terms of faculty
time) of software development. The steady
improvement of computing hardware has
made the former a problem no longer.
The latter obstacle is our target.

the computer (although it can be used in that way too). There
are obvious features such as time compression or expansion,
a unique laboratory for each student, etc., enabled by simula-
tion. But beyond this are more the pedagogical features of

Random events can be programmed to occur, to which the
student must respond appropriately; less dramatic but just as
useful, equipment can be programmed to "age" as it is used.
Students can be quizzed in a number of ways, and their re-
sponses can be recorded; this can occur before, during, or
after the simulation, and subsequent operation of the simula-
tion can be based on their performance (e.g., the student
cannot begin until satisfactorily completing a pretest).
Simulations can be conducted intermittently over several days,
with the student being given a fixed period of time to return to
the simulation; this time can be used by the student to reflect
on subsequent actions.
Access to a simulation can be restricted; likewise, the student
may be given a fixed number of practice runs before conduct-
ing one or more runs for grading.

No doubt many more novel features can be conceived. Our
goal in developing CESL is to enable the simulation author
to program and implement these features and the instructor
to use them.
Clearly, there is great potential for diversity in design of
simulations. It is helpful then to have an organizing principle
when considering the options. We have identified the fol-
lowing four categories of simulation:

Laboratory The student is presented with a piece of
(virtual) equipment, or an entire process, and he or she
conducts "experiments" on it to characterize its opera-
tion. This simulation is meant to mimic as closely as
possible an actual laboratory experience; it is the mere
simulated laboratory.
Steady-state simulation The student must choose
conditions that optimize the operation of some equip-
ment when run at steady state. The parameters under
which the simulation proceeds vary through the course
of the experience, building an intuitive sense of cause-
and-effect. The student is given a period of time
(ranging from minutes to days, per simulation design) to
reflect on each action.

Unsteady-state simulation The students operate equip-
ment in "real" time (which may be compressed or
expanded time, if needed). They must respond to regular
or random changes in process operating conditions,
relying mainly on an acquired intuitive "feel" for the
equipment's operation. Many actions are demanded of
the student, each by itself being of small consequence,
but together adding up to success or failure (perhaps
catastrophic) in operating the equipment.
Design The student is given equipment, or a budget with
which to "purchase" equipment, and must assemble,
test, and operate a process that is in some sense optimal.
Presently, only the first three categories of simulation can
be constructed with CESL, and we have developed a proto-
type module for each: a simple tank-draining laboratory; a
pump-sizing steady-state module; and an unsteady-state con-
tinuous stirred-tank reactor (CSTR) module. We will de-
scribe the last of these prototypes.

THE PROTOTYPE CSTR MODULE
The display for the prototype CSTR module is presented
in Figure 1 (the actual display is in color). Series-parallel
reactions take place within the reactor:
A+X-X B+Y

B+X--C+Y
The reactions are exothermic
and the reactor is not isother-
mal. Reactant X is presumed to
be a gas that dissolves very rap-
idly in the liquid-phase reaction
medium, so that the dissolved
concentration of X is always
proportional to its feed partial
pressure (i.e., Henry's law is
obeyed). Normally, the supply
pressure of X is essentially con-
stant (though there may be small
fluctuations). The other reactant,
A, is supplied from three tanks:
one containing A in relatively
high concentration, one contain-
ing A in "medium" concentra-
tion, and one containing A in
"low" concentration. Deliveries
are made at random times to
replenish the three tanks.
The students must operate the
system attempting to maximize
the yield of the intermediate
product, B. At the same time,
they must prevent any of the Figure 1. Display
tanks from overflowing and m
116

must keep the reactor temperature under control. The stu-
dents can manipulate the flow rate leaving each tank, the
feed pressure of X (which must be less than or equal to the
supply pressure), the flow of steam to the reactant pre-
heater, and the flow of coolant to a coil within the reactor. A
reactor quench can be used in an emergency if the student
needs it. As already noted, the simulation will provide ran-
dom deliveries to the feed tanks. The concentration and
temperature in the tanks may fluctuate slightly due to these
deliveries and seasonal conditions. Other potential problems
that the simulation may invoke include a reduction in or loss
of steam pressure for the pre-heater, a reduction in or loss of
supply pressure of reactant X, a loss of cooling water flow
or increase in cooling water temperature, and a gradual
decrease in the heat transfer coefficients for the two
exchangers. Because the reactions are exothermic, the
student will need to exercise care whenever the feed
concentration of reactant is increased or a thermal run-
away may occur. Similarly, a large decrease in feed con-
centration may result in a significant temperature drop
and thereby a loss of conversion.
The module is designed to develop within the students an
intuitive feel for how conversion, yield, selectivity, and out-
let temperature (call them response variables) are affected
by changes in operating variables (feed composition, feed
temperature, feed flow rate, and heat exchange) for series-
parallel reaction networks. There are several stages or levels

presented to student while operating CSTR prototype
odule. The actual display is in color.
Chemical Engineering Education

of attaining such an intuition. The module allows the student
to progress through these levels.
At the most elementary level, the student intuitively knows
which operating variables to change and whether to increase
or decrease them, in order to effect a specified change in one
of the response variables. Often, a desired change can be
brought about by manipulation of more than one variable. At
the next level of intuition, the student knows which of the
operating variables will be most effective in bringing about
the desired response (i.e., he or she knows which operating
variable will cause the least change in the other response
variables). At a yet higher level, the student knows how all
the responses will change (at least direction and qualitative
magnitude) when a given operating variable is changed.
At a still higher level of understanding, the student can
explain why the system responds as it does to a given change
in operating variables. Here the student should be able to
formulate the explanation lucidly without the use of equa-
tions and mathematics. Finally, the ultimate objective of the
module is that the student knows how all the above would
differ if other parameters of the reactor were changed (e.g.,
if the reaction was endothermic instead of exothermic, if the
kinetic order of one or the other of the reactions increased or
decreased, etc.).

OVERVIEW OF CESL DESIGN
A schematic of CESL and its role in the development and
implementation of simulation laboratories is presented in
Figure 2. In the upper-left corner of the figure is the process
to be simulated; perhaps it is too large, dangerous, expen-
sive, etc., to expose to the student. CESL comprises the
elements within the gray-shaded region. The white-on-black
components have not been implemented (or even designed)
in the present version of CESL, but they will be developed as
part of future work.
The simulation author is responsible for identifying the
appropriate model for the physical system and for program-
ming it using an established language (let us say that this is
done in FORTRAN). We delegate this task to the module
author for several reasons. First, quantitative modeling and
programming form part of the undergraduate and graduate
training of chemical engineers, so an instructor should have
some competence here, at least for sufficiently simple ex-
periments. Second, general and robust process simulators
already exist, so any efforts expended by us in this direction
would be inefficiently placed and thus detract from the im-
portant task of developing the novel features of CESL. Third,
by making the model and simulation code separate from the
core of CESL, we introduce a large element of flexibility in a

simulation
author

Figure 2. Schematic of simulation laboratory and CESL's role in implementing it.
Spring 1996 11

desired laboratory

operating system:
unix, windowing,
file system
--- TI --I

Student

module, and indeed in CESL itself. Instructors may modify
or even replace the modeling code (presumably to im-
prove it) while retaining the general simulation design
and interface. More significant, extension of CESL to
disciplines other than chemical engineering is well fa-
cilitated by this design.
The simulation author is also responsible for creating a
graphical display through which the student interacts with
the modeling code. Our goal has been to make this task as
simple as possible. To this end, we have devised the Simula-
tion Script Language (SSL), a declarative language through
which the module author "equips the simulation" and speci-
fies rules concerning how the equipment may be used and
how it interacts with the numerical modeling code. The
module author prepares the script using any text editor. The
script is parsed by CESL when the student calls for the
laboratory to be loaded into the system, and the "Simulation
Control" element of CESL is thereby programmed with the
laboratory. The Simulation Control element interfaces with
the FORTRAN modeling program (to gather data and make
sure that solution of the numerical model is proceeding
synchronously with the wall clock), the operating system
(e.g., to record data to file), and the CESL interface (through
which the student conducts the simulation).
There are two elements presented in Figure 2 that were not
needed to implement the prototypes but which are important
to the ultimate success of CESL. First is a "pedagogy mod-
ule," which monitors the activity of the student and reports
to the Simulation Control the actions needed to improve the
student's understanding of the lesson. Second is a script-
writing interface. In simplest form, this interface will enable
the author to prepare the SSL script using mouse-oriented
actions; it will also guide the author in creating modules that
are pedagogically sound. We plan to incorporate these fea-
tures over the coming years.

MODULE WRITING
The SSL is a declarative language comprising a set of
keywords and qualifiers that the module author uses to con-
struct a simulation. Declarations may be categorized into the
following three types:
Object statements declare "variables" and place correspond-
ing graphical elements on the screen; these graphics can
display or allow user-specified changes to the value of the
variable. Simple examples include a temperature gauge or a
valve that may be opened and shut.
Procedure statements declare the numerical routines that
model the system's behavior. Included in these statements is a
specification of the object-declared variables that are passed
to the routine, and when or how often the routine is called.
Routines may be called at fixed points in the experiment (e.g.,
immediately after the student initializes the laboratory), at
regular intervals (of 0.1 sec, for example, if the routine is
integrating unsteady equations in time), at random intervals
118

(to cause random events to which the student must respond),
or at the behest of the student (by clicking on an appropriate
graphic button). As programming the routines is completely up
to the module author, they can make anything happen (e.g., a
pressure loss is programmed by having the routine simply set
the appropriate pressure variable to the newly desired value).
Controls declare restrictions and monitors of student's
actions. The design and implementation of these features is in
an early stage.
There are only two basic conceptual matters that a module
author must grasp to construct a laboratory. The first deals
with how CESL, the student, and the modeling routines
change and communicate values of the laboratory param-
eters (e.g., temperatures, flow rates, status of valves).
This is done using the "shared memory" concept. The
idea is simple: there is one "official" repository of all
parameter values, and they may be accessed or changed
at any time by CESL, the student, or the modeling rou-
tines. Thus, once the modeling routine has computed a
set of updated values (perhaps by completing a time-step
calculation), it makes a simple call to a library routine
that updates the shared-memory values.
The second conceptual matter concerns how CESL keeps
in sync with the wall clock (an issue only with Laboratory
and Unsteady-state simulations). The SSL script specifies
how often a procedure is to update process variables. After
computing its values, the routine suspends itself (again using
a simple library call), until restarted by CESL (after a period
of, say, 500 msec). When restarted, the values in shared
memory may have been altered (e.g., a valve may have been
shut off). When the routine next uses such values, it will
produce results that reflect the changes. In particular, while
the routine is suspended, CESL can update the "time vari-
able" using the system clock. Thus the routine can be pro-
grammed to blindly update its variables to whatever time it
reads from shared memory, without any concern about
whether or how that time matches the wall clock.
The simple calls to routines that read and write shared
memory, or suspend subroutine execution, are the only addi-
tions that the FORTRAN-routine author must include to
interact with CESL. Everything else is familiar and standard.

PLANNED FEATURES OF CESL
CESL is a work in progress, and the following features
have been designed in some detail but not yet incorporated
in the software:
I Experimental error may be introduced to an arbitrary extent
and in two ways: the first is what we call "gauge error," and it
describes the simple addition of normally distributed stochas-
tic noise to the values reported to the student; the second is
what we call "fluctuation," and it involves random perturba-
tions to the process variables themselves. In contrast to gauge
error, fluctuations are propagated through the system. They
may in fact be viewed as part of the model that describes the
Chemical Engineering Education

physical behavior.
> Any process variable may be alarmed, with setpoints specified
by the module author or the student, and with notification
made audibly or via a visual indicator. It may prove interest-
ing to observe which variables the students decide to alarm.
O Any process variable may be subject to automatic control
using a PID 'device' that is tuned by the author or the student.
> The variability of the simulations is easily controlled; each
student may be provided with a unique piece of equipment, or
all students may be presented with the same equipment, or
either of a pair of pieces of equipment, etc. Equipment may
also be programmed to 'age,' with its operating characteris-
tics changing in an appropriate way as it get older.
0 The instructor may schedule the "availability" of the (virtual)
equipment, restricting its use to, say, a particular one-week
period. Also, the number of practice and grading runs may be
specified, along with separate time periods for each.
> Data output by CESL for analysis by the student may be
presented in any of several pre- or student-defined formats.
The predefined formats are chosen to make them suitable for
immediate input to popular graphing and analysis programs.
This specification reflects a general design principle of CESL
to exploit pre-existing software to the fullest extent possible.
We do not wish to re-invent software that already exists and
functions well.

Phillips,"51 Koper,171 and Wankat and Oreovicz81" each em-
phasize the importance of the team approach to educational
software development. Phillips notes the need for both cur-
riculum and computer specialists on such a team, and Koper
stresses the additional role of the educational technologist.
In addition to the expertise offered by computer science
majors and over half of our department's faculty, we have
recently recruited to the project experts in education technol-
ogy (Prof. Thomas Shuell of our Graduate School of Educa-
tion) and human-computer interfaces (Prof. Valerie Shalin
of our Department of Industrial Engineering). Their impact
will be felt particularly in our subsequent efforts.
An interesting application of CESL concerns the develop-
ment of new modules. We plan to offer to our students, in
the form of an elective Projects course, the opportunity to
develop new modules that could be used for instruction of
subsequent classes. As part of this project, the student will
be given the task of creating a working module. This will
entail the concept for the module, considering carefully the
instructional goal (provided by a faculty advisor), design of
the module, writing of the script, programming the model,
and testing the product. In this manner CESL will provide to
the student a unique experience in pedagogy and design that
simply could not be offered by other means.

DEVELOPMENT PLATFORM
We have chosen to develop our software on a Unix plat-
form. We have been careful to employ development tools for
which there exist industry standards. Thus all of our code is
Spring 1996

written in ANSI-standard FORTRAN and (predominantly)
C. We use the X-windowing system because it is widely
portable and freely available. Because CESL itself interacts
very well with the Unix operating system, we can readily
introduce file-handling and classroom-management features
that will underlie many of CESL's capabilities. This capac-
ity also will facilitate the introduction of pedagogical func-
tions that contribute to the realization of a complete com-
puter-based instructional environment.J91 For example, a
record of student achievement and errors can be designed
and maintained, allowing CESL's activities to be tailored to
the student's progress.
Alternatives to our choice include the use of the C++
programming language and the traditional personal com-
puter platforms. C++ is object-oriented and thus very well
suited to our needs, so we are giving serious consideration to
its eventual use. There is, however, no present ANSI stan-
dard for this language, and it is not as widely available or
portable as C. The Macintosh and PC platforms are appeal-
ing because of their wide availability. These platforms are
capable of running Unix and X-windows, so our present
approach does not preclude porting to them.

ACKNOWLEDGMENTS
CESL was developed with the support of a Leadership in
Laboratory Development grant from the National Science
Foundation (DUE-9352500) and from the SUNY Buffalo
School of Engineering. We wish to thank both Mr. Rich
Alberth for very important contributions during CESL's for-
mative stages and Dr. Nitin Ingle, who provided program-
ming assistance to the project. Finally, we thank Sun
Microsystems, Inc., for substantial equipment discounts and
other support, and Mr. Corky Brunskill and his staff for their
many contributions to our efforts.

REFERENCES
1. Fogler, H.S., and S. Montgomery, "Interactive Computer
Modules for Chemical Engineering Instruction," CACHE
News, 37, 1 (1993)
2. Squires, R.G., G.V. Reklaitis, N.C. Yeh, J.F. Mosby, I.A.
Karimi, and P.K. Andersen, "Purdue-Industry Computer
Simulation Modules: The Amoco Resid Hydrotreater Pro-
cess," Chem. Eng. Ed., 25, 98 (1991)
3. Seider, W., "Chemical Engineering Instruction and Com-
puting: Are They in Step?" Chem. Eng. Ed., 27, 134 (1988)
4. Shacham, M., and M.B. Cutlip, "Authoring Systems for
Laboratory Experiment Simulators," Computers Educ., 12,
277 (1988)
5. Phillips, W.A., "Individual Author Prototyping: Desktop De-
velopment of Courseware," Computers Educ., 1, 9 (1990)
6. Carnahan, B., "Computing in Engineering Education: From
There, To Here, To Where?" Chem. Eng. Ed., 25, 218 (1991)
7. Koper, R., "Inscript: A Courseware Specification Language,"
Computers Educ., 16, 185 (1991)
8. Wankat, P.C., and F.S. Oreovicz, Teaching Engineering,
McGraw-Hill, New York, NY (1993)
9. Wenger, E.L., Artificial Intelligence and Tutoring Systems,
Morgan Kaufmann, Los Altos, CA (1987) C

ASEE

ANNUAL MEETING

Washington, D.C.
June 23-26, 1996

Chemical Engineering Division Program

FEATURE SESSION
The Chemical Engineering Curriculum
#1213 Is Anyone Out There Doing Anything Different?
Panelists:
Harold Knickle Mahbub Uddin
Barrie Jackson James Watters
David DiBiasio Andrew Wilson

A panel discussion of current attempts to make sea changes in the total chemical engineer-
ing curriculum. Panelists will represent schools offering an alternative approach to deliver-
ing chemical engineering education. The discussion will focus on the rationale for each
approach, good and bad experiences, student acceptance, and measures of success.

1,4^_ _^ ^- _^------- -^ -

Chemical Engineering Education

-- ADDITIONAL SESSIONS
#1413 Chemical Engineering Chairperson's Lunch
#3413 Chemical Engineering Division Business Meeting and Lunch
#2613 Chemical Engineering Division Award Lecture
#2713 Chemical Engineering Division Dinner
=moo

-- REGULAR SESSIONS

#1613 Innovative Uses for Educational Software Available Through CACHE
D Employing the CACHE CD-ROM as an Educational Resource
> Facilitating Numerical Problem Solving With POLYMATH
0 PiclesTM for Bridging the Gap Between Laboratory and Textbook Learning
Course Implementation of the Michigan PC Modules for Chemical Engineering

#2213 Experimental Experience in the Undergraduate Curriculum
> Integration of a Manufacturing Experience into the Undergraduate Curriculum in
Polymer Engineering
Implementation of Peer Feedback and Improvement Planning in the Unit Operations
Laboratory
1 Supercritical Fluid Extraction in the Undergraduate Laboratory
> Development of Multifunctional Laboratories in a New Engineering School
0- Development of a Multidisciplinary Biochemical Engineering Laboratory
Teaching Senior Unit Operations Laboratory Experiments to Engineering Freshmen

#2313 Assessment of Learning Outcomes
O- Student Learning Assessment and the ABET Student Outcomes Criteria: Good News/
Bad News
Project Gelegenheit: Skills Certifications Curricula for Engineering and Computer
Science Disciplines
A Portfolio-Based Assessment Program
Outcomes Assessment Measures
Assessment Recent Regional Accreditation Experience

#2513 Mentoring Graduate Students-Panel Session
Panelists: Richard C. Seagrave, Janice A. Phillips, Timothy J. Anderson,
John P. O'Connell, and Jennifer Sinclair

#3213 Homework Problem and Lecture Exchange
> Exercises in Process Design for a Freshman Course in Chemical Engineering
> A Process Troubleshooting Program
1 Entropy: Esoteric or Utility Infielder?
Design of a Propylene Storage Facility
0- Using Statistical Experimental Design to Optimize GC Operation
Statistical Exercises in Chemical Engineering
Teaching Data Analysis Techniques Using Practical Polymer Examples

#3513 Technology Enhanced Instruction
Use of Computational Tools in Engineering Education
Problem-Centered Course Objectives Leading to Multimedia Lessons
1 Recent Developments in Virtual-Reality Based Education
D Controls Laboratory Teaching Via the World Wide Web
Incorporating Bioengineering Examples into the Core ChE Courses

Spring 1996

survey

APPLIED STATISTICS

Are ChE Educators Meeting the Challenge?
(A Survey of Statistics in the Chemical Engineering Curricula)

ROGER E. ECKERT
Purdue University West Lafayette, IN 47907

W.A. Shewhart had a dream of a "statistically
minded generation" (W. E. Deming, 1939'"). This
dream exists today, says R.G. Batson,121 "but, as
we are all keenly aware, not in the U.S."
For years, industry has recognized the need for chemical
engineers to understand and use applied statistics. Increas-
ingly stringent quality requirements and emphasis on TQM
(Total Quality Management) have intensified this need in
recent years. Probably more in-house courses on statis-
tics-related topics are given to engineers in industry than
on any other technical subject. Since at least the 1960s,
surveys of engineers and managers in industry have re-
peatedly shown a shortfall in this area.[31 What are chemi-
cal engineering educators doing to correct this deficiency?
Will we relinquish our job of educating chemical engi-
neers in such an important area?
ABET has added the accreditation criterion, "Students
must demonstrate knowledge of the application of probabil-
ity and statistics to engineering problems" (Sec. IV.C.3.h).
What evidence will educators present to a visiting team
during ABET accreditation that graduates meet this crite-
rion? Chemical engineering students take mathematics
courses through differential equations, but do they have
even an introduction to the random variable and stochastic
processes? Even more important, are they sufficiently pre-
pared to use statistics correctly and efficiently in their indus-
trial careers? Or will the industrial emphasis on TQM be a
surprise and a stumbling block for the new graduate?

Roger E. Eckert is a professor of chemical
engineering at Purdue University. He received
his BS in chemical engineering at Princeton Uni-
versity and his MS and PhD at the University of
Illinois, joining DuPont after graduation. After
over a decade of industrial research and devel-
opment, he moved to Purdue, where he has
taught and researched applied statistics and ex-
perimental design, multiphase reactions, and
polymers.

Copyright ChE Division ofASEE 1996

The survey that forms the basis of this article was sent to
172 chemical engineering departments in the United States
and Canada and was devised to develop information on the
status of applied statistics as background for educators and
industry to view our progress toward the above goals. The
purpose of the survey was not to establish the "ideal course"
or sequence, but to stimulate thought and discussion in this
direction. It is, to the best of our knowledge, the first survey
on the extent to which statistics is included in the chemical
engineering curriculum.
The questionnaire was sent as a single-page attachment to
the annual course survey by the Chemical Engineering Edu-
cation Projects Committee of AIChE in April of 1993 and
was conducted by the late Edwin Eisen (McNeese State
University). A follow-up request from John Griffith was
sent to non-respondents in August, and forms continued to
be returned in 1994. Inquiries were made about the status
of inclusion of the usual topics of applied statistics. The
important related subjects of experimental design and
quality control were added.

THE SURVEY
The purpose of this survey was to present a summary of
the status of statistics-related education in chemical engi-
neering. The questions asked were
ABET has added the requirement "Students must demon-
strate knowledge of the applications of probability and
statistics to engineering problems." Are these topics in-
cluded in your required curriculum? Are you planning
curriculum changes on this topic? Are elective courses
including these topics taken by some of the chemical engi-
neering students?
In addition to responses to the above questions, information
about the courses (e.g., is the course required, the depart-
ment that offers the course, coverage of the field, and the
textbooks used) was requested, and this information is sum-
marized along with responses to the above questions.

Chemical Engineering Education

RESPONSES
Of the 112 departments that replied to the survey (65%
returned), all but five provided usable information for analy-
sis. Eighty-six (80%) reported that topics pertaining to the
ABET criterion are included in their required curriculum; 21
(20%) said no such topics are required. Therefore, most
chemical engineering curricula require some statistics either
as a separate course or as part of a course.
Note that the survey was sent to the entire population of
chemical engineering departments and not simply to a sample
selection. Therefore, there are no confidence intervals or
"margins of error." The 60 non-responding departments
(35%), however, are not expected to parallel the results
presented here. Without direct substantiation, the author be-
lieves departments that do not have applied statistics in their
curricula were less likely to respond to the survey. An abso-
lute lower limit on the percentage of schools that include
such topics is obtained from the positive replies out of the
questionnaires sent, which is 50%, in comparison with 80%
of the actual responses.
The question "Are you planning curriculum changes on
this topic?" resulted in 31% "yes" responses from the par-
ticipating schools. Percentages of the 107 useful responses
are presented in tabular form for accuracy in Figure 1. These
results, along with the 80% including statistics in the cur-
riculum noted above are shown on the 3-D bar graph in
Figure 1. The rectangle heights represent percentages of the
useful 107 responses. Please refer to the "Total" columns

PIanning Chin ge? C". I, a. I "P

Is Statistics in the Curriculum and Do You Plan Changes?

10
70
60
g 50
40

Figure 1. Percentages of the total 107 survey responses to the two
questions "Are statistical topics included in your required cur-
riculum?" and "Are you planning curriculum changes on this
topic?"
Spring 1996

first on each question; these are the tallest bars at the rear for
each axis. The 100% grand total that would appear in the
rear corner is purposely omitted to better display the impor-
tant results. You will first see the 80-20% response to the "In
Curriculum?" question. On the question of changes to the
curriculum, 18% of the responses were blank, but all these
respondents did answer the requirement question. Then, as
lower limits, 31% plan to make changes and 51% do not. As
expected, these two responses interact. For those with a
requirement, only one-third are planning changes (21% yes;
42% no), while about half of those with no requirement plan
changes (10% yes; 9% no).
Of the 80% that require statistics in the curriculum, 42%
have a major course on this topic (see Figure 2). Major
course coverage is defined as greater than one credit hour of
statistics that includes four or more of the seven topics listed
in Table 1. Slightly more, 45%, include statistics in a course
and do not have the coverage defined for major coverage.

Figure 2. For those schools that require statistical topics, the
percentage that require a major amount of a course (greater than 1
credit hour and more than 3 topics from Table 1) or a minor
amount are compared. If it was not possible to make this decision
based on the information supplied, the school was counted in the
(?) bar.

TABLE 1
Statistics Related Topics Listed on the Questionnaire
(Submitter noted if topic is in each course.)

( Distributions 3 Analysis of Variance
(3 Statistics, t, ,'. F .1 Design of Experiments
( Regression Modeling 4 Quality Control
J Other (as needed)

1_ ____^_____n

I

I

Some (11%) report a course but did not provide sufficient
detail to make this determination.
Figure 3 is a bar chart and table of detailed breakdown
percentages for those departments that require a major course.
Sixty percent place it in the junior year, and the rest are
equally divided between the sophomore and senior years.
No "major" course coverage, as defined above, was reported
for the freshman year. The majority of these courses (57%)
are offered by the chemical engineering departments. About
a third of the courses are from mathematics or statistics
departments, and the remaining 9% are offered by other
engineering departments, mostly industrial engineering. The
math/stat courses are concentrated even more heavily in the
junior year; the number is close to that given by chemical
engineering departments in the junior year.
When only a smaller part of a required course is statistics
(see Figure 4), the courses reported are all taught within
chemical engineering but are divided between class (58%)
and laboratory (42%). More of the classes are in the sopho-
more and junior years. Few are during the freshman year,
and an intermediate number are in the senior year. The
laboratory courses are, of course, heavily concentrated in the
senior year, with less than one-third of the labs in the junior
year. Overall, there is a progression of increasing numbers
of courses toward the end of the curriculum. Application of
the content to other courses is thus limited.

STATISTICS ELECTIVES
Suitable applied statistics elective courses) that some
chemical engineering students take were reported by 86% of
the departments responding to the questionnaire. Figure 5
includes the "blank" responses, so 78% is the corresponding
figure. Surprisingly, whether or not statistics is required in
the curriculum, about the same percentage of departments
report some chemical engineering students take statistics
electives. Perhaps the students' need for statistics when none
is required is balanced by further interest in electives when
the subject is required. Furthermore, whether or not some
chemical engineering students take a statistics elective does
not affect the school's decision to change the curriculum.
The regularity of the response bars indicating no interaction
to these two questions can be noted in Figure 5. The elective
response is totally independent of the other two "yes and no"
questions of Figure 1. For this reason, electives were ana-
lyzed separately.
The number of chemical engineering students who elect
such a course was not asked on this survey since a detailed
study of student records would have been required to deter-
mine the percentage of students who take such an elective
course. Not even a single remark to the effect that many or
most of the chemical engineers take these electives was
offered on the survey sheets. We might infer that the
number of chemical engineers introduced to statistics

TeaA
-'-' ------ ~-----Te ---- --;
T -.r,...... ..l '...r '"' .
Dept. r. IEI -

A Major Course in Applied Statistics is Required

Figure 3. The percentage of schools requiring a course with major
amount of statistics is displayed by the combination of the depart-
ment offering the course and the year in which students normally
take the course.

--- Tear" .--
sr : Fr... ;r, .;-.,,r '- .-fe J r- :.r -f *-*..
20 32 13
S 21' 2-. 3
. .1' ': .

12

Part of a Course in Applied Statistics is Required

Figure 4. When the schools require less statistics than that de-
fined as major in Figure 2, statistics is often included with a
chemical engineering laboratory course. Alternatively, statistics is
included with a "class," which is a lecture and/or recitation. The
percentage of each of these and the year in which students nor-
mally take the course is displayed.
Chemical Engineering Education

so

140

10
0

N

I

. ,
-hU ,:):

-o "y

through an elective course is low.
Elective courses that have a major emphasis on statistics
are offered in sixteen chemical engineering departments.
Twelve of these courses include six or all seven of the topics
listed in Table 1. The topics most often not included from
this list are the applied topics of experimental design and
quality control. Although these are useful topics, they usu-
ally follow more basic statistics in a course. Other topics
of a statistical nature that were noted by the respondents
are propagation of error, time series, nonparametric analy-
sis, maximum likelihood, uncertainty analysis, and SPC.
Also noted were several software packages such as SAS,
Minitab, and Matlab.

TEXTBOOKS
The response to the request that the text be listed was both
disappointing and revealing. It had been expected that texts
for courses outside chemical engineering would often not be
known by the supplier of this information, but even within
chemical engineering only a few texts were listed, and there
was no predominant one. The five mentioned more than one
time are
Box, G.E.P., W.G. Hunter, and J.S. Hunter, Statistics for
Experimenters, Wiley, New York, NY (1978)
Devore, Jay L., Probability and Statistics for Engineering
and the Sciences, 3rd ed., Brooks/Cole, Monterey, CA (1991)
Hogg, R. V., and J. Ledolter, Engineering Statistics, 2nd ed.,
Macmillan (1992)
Mason, R.L., R.F. Gunst, and J.L. Hess, Statistical Design

SomeElective?

Change Currictulum?
Si ll- '.. Total
7c.,"ij ,, 1 1" 2 I'
If,. '. 78
1. 2 5 to T13
T-1J rz, I 2' 9
Statistics Electives

Figure 5. Whether or not some chemical engineering engineering
students elect statistics-related courses does not affect the
department's decision to make curriculum changes regarding sta-
tistics.
Spring 1996

and Analysis of Experiments: With Applications to Engineer-
ing and Science, Wiley, New York, NY (1989)
Walpole, R.E., and R.H. Myers, Probability and Statistics for
Engineers and Scientists, 3rd ed., Macmillan (1985)

It should be noted that several new texts with titles that
included "statistics for engineers" have been published since
the start of this survey.

CONCLUSIONS
Over half of the chemical engineering curricula has at
least some applied statistics in required courses. Many de-
partments are adding statistics to required courses and are
currently changing their curriculum in this regard. It will be
necessary for others to include this general topic to meet
the new ABET criteria. The required extent of coverage
will undoubtedly depend much on the specific
accreditor(s)' interpretation as this requirement is en-
forced through the coming years.
A preferred approach, whether the curriculum now in-
cludes applied statistics or not, is to look to the needs of the
graduate chemical engineer. With computers in universal
use, the extensive calculations to apply statistics can be
readily performed. Good experimental design is essential for
the efficiency and productivity demanded in today's mar-
kets. There is an elusive optimum range for the amount of
applied statistics in the chemical engineering curriculum. In
achieving these goals, decisions will be necessary on what
portion of the current curricula will be supplemented or
supplanted. A single course, with material carefully selected
for maximum application both in the curriculum and in
industry, will likely be the minimum need.
An early introduction to stochastic variables in mathemat-
ics courses would help to "pave the way" toward engineer-
ing applications courses. Elective courses will continue to
increase the availability of these topics to the interested
student, but they are not a substitute for a minimum require-
ment for understanding of statistics of the stochastic variable
in parallel with mathematics of the deterministic variable.

ACKNOWLEDGMENT
Edwin Eisen kindly offered to include the questionnaire
with his annual survey of courses. After his death, D. John
Griffith continued the work and I am indebted to them for
such help. I also thank the 107 professors who completed the
questionnaires that provided the data for this paper.

REFERENCES
1. Deming, W.E., Shewhart, W.A., Statistical Method from the
Viewpoint of Quality Control, Dept. of Agriculture, Washing-
ton, DC (1939)
2. Batson, R.G., "Statistical Training: A National Necessity,"
Eng. Ed., Sept/Oct, 598 (1989)
3. "Two Surveys Show: What Engineers Would Study," Engi-
neer (Engineers Joint Council), 6(2), Summer/Fall, 4 (1965) J

r 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 interns and
coop assignments typify such experiences; however, reports of more unusual cases are also welcome.
Description of 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.

INTERNATIONAL ENGINEERING

INTERNSHIP PROGRAM

At The University of Rhode Island

JOHN M. GRANDIN, KRISTEN L. VERDUCHI*
University of Rhode Island Kingston, RI 02881-0805

Through the cooperative efforts of engineering and
foreign language faculty, the University of Rhode
Island (URI) offers a five-year International Engi-
neering Program (IEP) that leads simultaneously to both the
Bachelor of Arts degree with a major in German and the
Bachelor of Science degree in one of the engineering disci-
plines. Key features of the IEP include separate sections of
German language courses specially designed for engineer-
ing students, a six-month internship with an engineering
firm in a German-speaking country, and a capstone engi-
neering course taught in German by bilingual engineering
faculty. In many cases, IEP students also complete intern-
ships with IEP partner companies in the Rhode Island area

John M. Grandin is Professor of German, Chair of the Department of
Languages, and Director of the International Engineering Program at
the University of Rhode Island. He holds a PhD from the University of
Michigan. In recent years he has spoken and published widely on the
internationalization of engineering education.
Kristen L. Verduchi is a Process Engineer at the Hoechst Celanese
Corporation in Coventry, Rhode Island. She graduated in 1994 from the
International Engineering Program at the University of Rhode Island
with a Bachelor of Science in chemical engineering and the Bachelor of
Arts in German.

* Address: Hoechst Celanese Corporation Coventry, Rhode
Island

before commencing the European internship experience in
the fourth year of the program.
The program was begun eight years ago with the help of a
grant from the Fund for the Improvement of Post Secondary
Education (U.S. Department of Education). Since that time,
over fifty students have completed their six-month intern-
ships abroad, and approximately thirty-five students have
graduated from the University with both degrees, several of
whom have gone to work for the companies participating in
the internship program. The successes of the IEP have led to
a growing popularity of the program, which now boasts an
enrollment of over eighty students.
The internship program in Germany is a key element to the
success of the IEP. The concept serves first of all as a
motivator for students to study Germanic language and cul-
ture along with their engineering studies. Above all, the
internship is the experience that puts study into practice and
provides concrete experience for students who plan to prac-
tice their careers in a cross-cultural environment. Although
arrangements are made in advance for the students by the
program's director, expectations on-site are the same as they
would be for German students. All work is intended to draw
on their engineering background and is carried out exclu-
sively in the German language. The goals are to experience
engineering as it is practiced in Germany and to refine

@ Copyright ChE Division ofASEE 1996

Chemical Engineering Education

German language and intercultural communication skills, i.e., to put
their seven semesters of engineering and language study into practice
in a very real situation.
The program was developed through a variety of contacts, many of
which have been facilitated through the efforts of the IEP Advisory Board
members, a group of Rhode Islanders active in various aspects of interna-
tional business and technology. Internship opportunities in Germany are
maintained with approximately twenty companies through annual visits
abroad and through various forms of communication, including mail,
telephone, fax, and e-mail. Some companies are subsidiaries or partners of
Rhode Island-based American firms; some are the home bases of German
subsidiaries in Rhode Island; others are firms that have come to our
attention through our earliest contacts in Germany. In each case, the
companies agree to provide an organized engineering work experience and
both housing and a subsistence-level monthly stipend for the IEP interns.
In general, arrangements are made approximately six months in advance in
order to ensure enough time for proper arrangements, including the neces-
sary visa and work permit.
As of this year, the program also includes an optional study abroad
component. With the support of a grant from FIPSE, URI business, engi-
neering, and language faculty have worked closely with their counterparts
at the Technische Universitat Braunschweig to develop a one-to-one stu-
dent exchange through which students may study in their field of expertise
at the partner school with full accreditation at the home institution. Five
IEP students are in Braunschweig this academic year, some of whom are
planning a semester of study and a six-month internship at one of our
partner companies. At the same time, five Braunschweig students are
currently in Rhode Island pursuing their engineering studies at the ad-
vanced undergraduate and graduate levels.
It is important to stress that the program is a full double-degree program
that satisfies all requirements for both the Bachelor of Arts and the Bach-
elor of Science degrees. Several attempts have been made at other institu-
tions to internationalize the engineering curriculum through the addition of
a few relevant courses or an intensive summer seminar. URI takes the
position, however, that this challenge can only be met through a commit-
ment of additional time at the undergraduate level. Students in the program
are expected to take a German language or culture course each semester
throughout the five years. They are not recommended for an internship
abroad until they have completed at least six semesters of German along
with at least three years of their technical studies. The companies in
Germany are eager to have young American engineering students partici-
pate in their internship programs, but only when they are able to communi-
cate in German and contribute to their technical needs.
URI views its program with a German orientation as a successfully
developed model that is equally appropriate for other languages and cul-
tures. German was chosen for the pilot program because of the level of
commitment on the German faculty and the presence of several German-
speaking faculty in the College of Engineering. URI is currently working
on the development of a parallel program in French and is also studying the
feasibility of a similar program in Spanish that would serve the needs of
Spanish-speaking immigrants in Rhode Island.
Chemical engineering majors in the URI program have an excellent
Spring 1996

... [the program] leads
simultaneously to both the
Bachelor of Arts degree
with a major in German
and the Bachelor of
Science degree in one of
the engineering
disciplines. Key
features... include
separate sections of
German language courses
specially designed for
engineering students, a
six-month internship with
an engineering firm in a
German-speaking
country, and a capstone
engineering course taught
in German by bilingual
engineering faculty. . .
Above all, the internship
is the experience that puts
study into practice and
provides concrete
experience for students
who plan to practice their
careers in a cross-cultural
environment.

opportunity for both U.S. and German internships since the
program is strongly endorsed by a local division of the
Hoechst Celanese Corporation. Seven IEP students have com-
pleted internships with the Hoechst headquarters in Ger-
many, most of whom had the opportunity to work with
Hoechst in Rhode Island in advance of the German experi-
ence and five of whom subsequently joined Hoechst Celanese
on a permanent basis after graduation. Ms. Kristen Verduchi,
one of these five students and now a process engineer with
this company, offers the following overview of her IEP
education at URI:

In May of 1994, I completed the International Engineer-
ing Program at the University of Rhode Island following my
return from a six-month internship at the headquarters of
Hoechst AG in Frankfurt, Germany. The five-year IEP
consists of nine semesters with an average of 17 to 19
credits/semester of university study and six months of an
internship in Germany. A semester during the fourth or fifth
year is spent abroad gaining practical experience at an
engineering firm in Germany. The university's program of
study provides the theoretical knowledge in the classroom
and the internship provides the opportunity to put the
engineering knowledge into practice. In this way, the
International Engineering Program is not only a novel way
to imbue an engineering curriculum with international
flavor, but also a way for students to put the classroom
theory into practice.

The undergraduates who plan to undertake the dual
degree IEP must decide to do so early in their college
careers, for one must complete approximately four years of
the IEP curriculum prior to traveling to Germany for an
internship at an engineering firm. Declaring participation
in the program early also has the advantage of making the
student eligible to intern at a Rhode Island engineering firm
in his/her field prior to going to Germany. It is this oppor-
tunity and subsequent ones that I will describe here.

My summer assignment following completion of the first
year of the undergraduate IEP was an internship at
Hoechst Celanese Corporation in Coventry, Rhode Island. I
worked during the summers of 1990-92 as an engineering
intern in the Process Engineering Department, carrying out
projects such as those listed below, each of which provided
me with practical skills and hands-on training not taught in
the classroom:

Investigation of the composition of a waste-air stack
from pharmaceuticals' processes using a Fourier
Transform Infrared instrument and development of a
user manual.
Survey ofplantwide lighting for the Narragansett
Electric Efficiency Program.

Development of piping and instrumentation diagrams
of pharmaceuticals' processes and of plantwide steam
and condensate lines.
Study of pressure relief devices.

Following completion of the fourth year of study, I
applied through the URI program to Hoechst AG in
Frankfurt am Main, Germany, for a professional intern-
ship. Between June and December of 1993, I interned in the
Process Engineering Department in Environmental
Engineering at Hoechst AG. In this extraordinarily
diverse work environment and culture, I experienced a
new lifestyle and adapted to new customs and to a new
style of communication.
New friends from various parts of the world as well as
professional relations with the employees from Hoechst AG
were among the many benefits of the internship. Under-
standing and respect for the German culture (or any
foreign culture) is the key to successful relations in
professional and personal German life. One must not be
overcome by culture shock, but must accept the foreign
culture with an open mind, must be willing to attempt new
things, and must modify one's thinking patterns. Success in
a foreign work environment is representative of how well-
suited one is to change and therefore to success in the
constantly changing world. It was with this mode of
thinking that I was able to learn about environmental
engineering from the chemists and technical assistants at
Hoechst AG in Frankfurt.
For the six months between June and December of 1993,

TABLE 1
Source Material on the
International Engineering Program
John M. Grandin, Author

"German and Engineering. An Overdue Alliance," Die
Unterrichtspraxis, 22, 146-152 (1989)
"Deutsch Fir Lngenieure- Das Rhode Island Programm." in Das
Jahrbuch DeuLsch als Fremdsprache, 15, 297-306 (1989)
"Developing Internships in Germany for International Engineering
Students." Die Unterrichtspraxis, 2, 209-214 (19911
"The Changing Goals of Language Instruction," .with Kandace
Einbeck and Walter %on Reinhart) in Languages fora Multiculrural
world in Transition, ed Heidi Byrnes (Lincolnwood, Ilhnois- Na-
inonal Textbook Company and Northeasi Conference) 123-.1b3
11992)
"international Experience for Engineers," (with H. Viets), in The
International Journal of Engineerng Education, 9( 1I, 93-94 1 1993'1
"The University of Rhode Island's Iniernational Engineering Pro-
gram," in Language and Content: Discipline and Content-Based
Approaches to Language Study, ed. Merle Krueger and Frank Ryan
(Lexington, Massachusetts: D.C. Heath and Company), 130-137
(1992)

Chemical Engineering Education

e CALL FOR PAPERS

I conducted study, design, and testing on an experimental
absorption facility in a laboratory. There, I designed a
laboratory-scale test facility and investigated the absorp-
tion capabilities of several packing used to purify a waste-
air stream using infrared spectroscopy. The experimental
and theoretical results of the design, testing, and analysis
were documented and distributed to Hoechst AG for review
and to URIfor academic credit. The report submitted to the
language and chemical engineering departments (in both
English and German) earned me two additional credits
toward the language degree, in addition to the six credits
awarded for the internship abroad. It was this report that
was later submitted to and presented at the 1994 American
Institute of Chemical Engineers'paper contest during my
last year at URI.
I am grateful to URI and to the creators of the IEP for
establishing such a fantastic undergraduate curriculum
that affords engineering students the opportunity to study
and work in an engineering field in Germany. The industry
needs engineering students who are bilingual, for we are
entering a time when global competition in business
demands international ties, relationships, and communi-
cation. The IEP trains engineering students for the
global marketplace.
Subsequent to the internships at Hoechst Celanese
Corporation in Coventry and following the six-month
internship in Frankfurt, I applied to Hoechst Celanese
Corporation in Coventry for a permanent position as a
process engineer/process safety engineer-I joined the
company in that capacity in June of 1994. Ifelt that my
educational background and internship experience in
chemical engineering with both the home and the parent
companies corresponded quite well with the job require-
ments. Since I joined the company, I have greatly enjoyed
Spring 1996

the challenging work in the process engineering/process
safety position and can report that the dual degree has
provided me with important tools. I find myself conversing
in German quite regularly with native Germans assigned to
our facility and with visitors from Germany. In addition, I
provide the service of translating technical documents from
Germany for process engineers.

Although Hoechst Celanese plays a major role in the prac-
tical education of chemical engineering students in the IEP,
it is not the only regional firm cooperating with the URI
program. Several students have also worked in a parallel
manner, for example, with a division of TRW in nearby
Massachusetts. To date, TRW has employed several chemi-
cal engineering students from the program in local summer
internship situations, three of whom have completed six-
month internships with TRW subsidiaries in Germany.
Among the four IEP grads who have gone to work full-time
for TRW, two are chemical engineers with assignments in
materials engineering and airbag deployment technology.
URI and its partners in the private sector take pride in the
development of the IEP as a model for the global education
of young engineers. Engineering educators are challenged to
prepare students for the international nature of their fields
today and also for the contemporary needs of research, de-
sign, and manufacturing. URI believes that the best response
to these challenges is through genuine interdisciplinary co-
operation within the structure of higher education, e.g., engi-
neering and language, as well as through carefully coordi-
nated partnerships between higher education and those com-
panies who will be employing our future graduates.
Additional source material on the International Engineer-
ing Program is presented in Table 1 for those readers who
are interested in further information. I

[

Fall 1996 Graduate Education Issue of
Chemical Engineering Education

Each year CEE publishes a special fall issue devoted to graduate education.
It includes articles on graduate courses and research as well as ads describing university
graduate programs.

Anyone interested in contributing to the editorial content of the 1996 fall issue
should write to CEE, indicating the subject of the contribution and the
tentative date it will be submitted.

Deadline is July 1, 1996.

Random Thoughts...

SPEAKING OF EVERYTHING

RICHARD M. FIELDER
North Carolina State University Raleigh, NC 27695

3 A rock pile ceases to be a rock pile the moment a single man contemplates it, bearing within him the
image of a cathedral.
Antoine de Sant Exup6ry

3 Whenever things get really bad, there is always some one to assure us amid great applause that
nothing has happened and everything is in order.
Carl Jung

3 Children have never been very good at listening to their elders, but they have never failed to imitate
them.
James Baldwin

3 We get the best results in education and research if we leave their management to people who know
something about them.
Robert Hutchins

3 Every man is a damn fool for at least five minutes every day. Wisdom consists in not exceeding the
limit.
Elbert Hubbard

3 Law of Probable Dispersal: Whatever hits the fan will not be evenly distributed.
Anonymous

3 A man's hatred is always concentrated on the thing that makes him conscious of his bad qualities.
Carl Jung

3 There is something that is much more scarce, something rarer than ability. It is the ability to recognize
ability.
Robert Half

Richard M. Felder i H.echsi Celanese Professor of
Crm.rr.:ai En,.r-onr ii Nonrm Carolina State Univer-
#t .-Me rcet7 e a r, BCrE trom.: City College of CUNY
anr r. s PnrO ro-m Princrtoin He- nas presented courses
on crremcat enonee-n rngprinc, pie reactor design, pro-
e C.e oopim,zaron an-. enecje teaching to various
4merican and torear, ,n,juii'r,e and institutions. He is
coaurnor ol tre text Elementary Principles of Chemical
Processes (Wiley, 1986).
Copyright ChE Division ofASEE 1996
130 Chemical Engineering Education

3 Never attribute to malice that which is adequately explained by stupidity.
Fred Vorhis

3 Expertise in one field does not carry over into other fields. But experts often think so. The narrower
their field of knowledge the more likely they are to think so.
Robert Heinlein

3 Many men stumble over discoveries, but most of them pick themselves up and walk away.
Winston Churchill

( Experience is a hard teacher because she gives the test first, the lesson afterwards.
Vernon Law

3 The only reason I would take up jogging is to hear heavy breathing again.
Erma Bombeck

3 Among those whom I like or admire, I can find no common denominator, but among those I love I can:
all of them make me laugh.
W. H. Auden

3 La causa de lo que hacemos es lo que creemos y tambiMn lo que buscamos.
Armando Rugarcia

3 Creativity always dies a quick death in rooms that house conference tables.
Bruce Herschensoln

3 I write when I'm inspired, and I see to it that I'm inspired at nine o'clock every morning.
Peter DeVries

3 I believe in miracles in every area of life except writing. Experience has shown me that there are no
miracles in writing. The only thing that produces good writing is hard work.
Isaac Bashevis Singer

3 We are so constituted that we believe the most incredible things, and once they are engraved upon the
memory, woe to him that would endeavor to erase them.
Goethe

3 Good judgment comes from experience. Experience comes from bad judgment.
Anonymous

3 Our achievements speak for themselves. What we have to keep track of are our failures, discourage-
ments, and doubts. We tend to forget the past difficulties, the many false starts, and the painful groping.
We see our past achievements as the end result of a clean forward thrust, and our present difficulties as
signs of decline and decay.
Eric Hoffer

3 More than any other time in history, mankind faces the crossroads. One path leads to despair and utter
hopelessness, the other to total extinction. I pray we have the wisdom to choose wisely.
Woody Allen

(3 never forget a face, but in your case I'll make an exception.
Groucho Marx

Spring 1996

BHclass 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 which
elucidate difficult concepts. Please submit them to Professor James 0. Wilkes (e-mail:
wilkes@engin.umich.edu) or Mark A. Bums (e-mail: maburns@engin.umich.edu), Chemical
Engineering Department, University of Michigan, Ann Arbor, MI 48109-2136.

DYNAMIC AND STEADY-STATE

BEHAVIOR OF A CSTR

Aziz M. ABU-KHALAF
King Saud University Riyadh 11421, Saudi Arabia

Mathematical models describe real systems in terms
of a set of mathematical equations (differential or
algebraic). This representation of the physical and
chemical phenomena governing the system, along with the
analytical and/or numerical solutions developed, enables us
to predict the dynamic and/or steady-state behavior of this
system. Analytical solutions are usually the most satisfac-
tory, but they become difficult with increasingly complex
systems. Numerical solutions are useful when analytical so-
lutions cannot be obtained or when comparison with the
available analytical solution is required to confirm the cor-
rectness of the latter.
Abu-KhalafP1 described a reactor setup (see Figure 1)
where a second-order reaction was studied with equimolar
feed concentrations and equal flow rates of the reactants
under isothermal conditions. Three stages of the CSTR were
modeled, namely from beginning to overflow (e.g., while
the reactor is filling up to a constant volume), from overflow
to the approach to steady state, and the final steady-state
r1

Copyright ChE Division ofASEE 1996

Figure 1. Reactor setup.
operation. Mathematical models with both analytical and
numerical solutions were developed. The mathematics in-
volved, however, is rather advanced, and students at this stage
may not be familiar with Bessel functions and other advanced
techniques. The mathematics can be simplified if afirst-order
reaction is studied instead of a second-order reaction.
A great many reactions follow first-order kinetics or pseudo
first-order kinetics over certain ranges of experimental con-
ditions.[21 Examples are the gas-phase decomposition of sul-
furyl chloride, the radioactive disintegration of unstable nu-
clei, the hydrolysis of methyl chloride, CH3Cl, the isomer-
ization of cyclopropane to propenes, and the decomposition
of dimethyl ether. A representative liquid phase chemical
reaction, which can easily be followed, is the reaction be-
tween acetic anhydride and water. This is a first-order reac-
tion with rate constant of 0.16 min' at room temperature.
Using this as a model, the following problem is suggested.
Chemical Engineering Education

Aziz M. Abu-Khalaf is a member of the chemi-
cal engineering teaching staff at King Saud Uni-
versity. His main interests are in mathematical
modeling, corrosion, and controlled-release sys-
tems.

PROBLEM)

Given the same setup as described above (Figure 1) and
considering a first-order reaction under the same conditions,
perform the following:
1. Develop mathematical models to describe the three
stages mentioned above.
2. Solve the models both analytically and numerically.
3. Considering the differential equation and the final
solution describing stage one, show how to circum-
vent the difficulty of defining the initial condition.

SOLUTIONN

Stage 1
During this stage, the reactor is filling up and the contents
are still below the overflow level. This means that both the
concentration and the volume are changing with time. A
component mass balance gives
rate of accumulation = rate of input rate of consumption
Therefore

d (VC) = FCo VkC (1)

or

V +CV =FCo-VkC (2)
dt dt
but V = Ft (by total mass balance, assuming constant density
and flow rate and noting that at t = 0, V = 0). Equation (2)
becomes, after some manipulation

dC I- +k)C -co 0 (3)

Equation (3) is a linear first-order differential equation, which
can be solved by the integrating factor method. The integrat-
ing factor in this case is t exp(kt), and the final solution is

C = C [ exp(-kt)] (4)
kt
Note here that application of the initial condition has to
follow physical sense, e.g., at t = 0', C = Co. This difficulty
will be obvious in the numerical solution.
Regarding the final solution, the exponential term can be
written in a series form as exp(-kt) = 1 kt, and upon substi-
tuting this into Eq. (4), it is easily shown that as time ap-
proaches zero, C = C0.
Another approach to model this stage is working with the
number of moles, N, of reactant in the reactor at any time. A
molal balance gives
kN + FC0 (5)
dt

The final solution (note that at t=0, N=0) is

N FC [- exp(-kt)] (6)

Stage 2
In this stage the volume is constant, but the concentration
is still changing with time because the process, although
continuous, is not yet steady. A component material balance
gives

rate of = rate of rate of
accumulation input output

rate of
consumption

dC
V_-= FC0 -FC- kVC
dT

and therefore

d C- kC (8)
dT T t
where
T = t = time in minutes
S= V / F = time constant
Again, Eq. (8) is a linear first-order differential equation,
which can be solved by the integrating factor method; the

integrating factor in this case is exp + k T. The final solu-

tion is

C C 1+ L =A 1 exp(-AT)] (9)

where A = (1 / T) + k, and CI is the concentration at the begin-
ning of stage two (at t = T).
Stage 3
In this stage, steady state prevails. Modeling of this stage
can be approached either by simplifying Eq. (9) as T -> o or
by a component mass balance. Both will give, after some
manipulation

Cs =- (10)
1+kT

NUMERICAL SOLUTION)

The initial value problems (Eqs. 3 and 8) can be solved by a
suitable Runge-Kutta subroutine (e.g., IVPRK or DIVPRK
from IMSL). Note that the initial condition for stage two is
that C = C, at T = 0 (or t = t), where C, is the concentration
at the end of stage one. The application of the initial condi-
tion to Eq. (3), when solving it numerically, may cause some
confusion. Here we have to give a specific value at zero
time, but at zero time Eq. (3) is not defined mathematically.
This should not cause any difficulty if we consider the ap-

Spring 1996

proach to t=0 rather than at t=0 itself. Thus, by consider
the definition of a derivative, the two terms in Eq. (3) w
time in the demonimator can be rewritten as time approach
zero as

lim C CO (dC
t-40 t dt )t=

Thus, the limiting form of Eq. (3) as time approaches zerc

(dC) 1 (1
Sdt Jt=O+ 2

which can then be used to start the numerical solution.

DISCUSSION

Startup is one of the interesting points of the system sho'
in Figure 1. The way we start up the system will affect 1
derivation and behavior of the model. In our case, for
ample, the reactor is initially empty, and thus we have
care about the definition of the initial conditions. Thus,
say that at t = 0, V = 0, but as t 0, C = Co. This
important when one has to deal with Eq. (4) and its differ
tial form, Eq. (3), but not with Eq. (6) because
initially the number of moles is zero. But we mea-
sure the concentration and not the moles, so if we
express the number of moles in terms of concentra-
tion, we will arrive at the same difficulty as above.
It will be an interesting point if one considers the
startup with a known volume and concentration of
the reactant. I strongly encourage the students to 3
try this and note the difference.
The analytical and numerical profiles, which de-
scribe the dynamic and steady-state behavior of the
CSTR, are shown in Figure 2 for kV/F = 2.5. The
two profiles agree, confirming the correctness of
tht analytical 'snlitions In Figure 3 the effect of -

the parameter kV/F is shown. This parameter can
be affected by a change in flow rate and/or a change
in temperature since, usually, the volume of the
reactor is constant. In our case, temperature is con-
stant at 25 'C (where k = 0.16 min- ), the molar
feed concentration = 10.5 M, and V = 3.8 1; thus
the change in kV/F is affected by the change in F
only. As can be seen, the approach to steady state
is faster for lower values of kV/F. Shown in the
figure are the limits of the three stages for the case
of kV/F = 3.0. This makes the comparison easier
for the other two cases where the time constant can
be calculated from the data given above.
The mathematics involved here is simpler than it
would be with higher orders of reaction, and be-
cause of this students will enjoy the process of
modeling and will, hopefully, gain the courage and

experience to tackle more complicated problems.

ACKNOWLEDGMENT
The author wishes to acknowledge helpful discussion and
suggestions by the editors of the "Class and Home Prob-
lems" section and the reviewers of this article.

is NOMENCLATURE
C concentration of the reactant at any time in mols/liter
1) Co concentration of the reactant in feed, mols/liter
C, concentration at the end of stage one, mols/liter
F total flow rate, liter/min
k reaction rate constant, min-'
N number of moles of reactant at any time, moles
t time, min
T time for stage two, min
wn V volume of the reacting system, liter
the T time constant, min
ex- REFERENCES
to
1. Abu-Khalaf, A.M., Chem. Eng. Ed., 28(1), 48 (1994)
we 2. Hill, G.C., An Introduction to Chemical Engineering Kinet-
is ics and Reactor Design, 1st ed., John Wiley, New York, NY
en- (1977) n

10

10 ..
00 40 4

10 5 20 25 30
Time, min

Figure 2. Analytical and numerical profiles.

| 1v/F=10

i

ae 1sages 2 and 3

0 10 20 30 40 50 60
Tim, min

Figure 3. Analytical solution as compared to different values of
kV/F.

Chemical Engineering Education

s0

I A N-i

5 40

ASEE
1996Annual Conference and Exposition
Sheraton Washington Hotel
Washington, DC
June 23-26, 1996

Plan now to attend the Annual Conference!

Conference highlights include:
* stimulating conference program featuring more than 350 technical sessions
* only forum specifically designed for all disciplines of engineering education
* bustling exposition with the latest in products and services for the engineering educator
* Awards Banquet, featuring the renowned political satire group "The Capital Steps"
* special Wednesday "Expo-Open House" for engineering students
* ideal networking environment for educators, researchers, administrators and related industry
professionals
* self-contained conference all sessions, meal events and exposition in one place!
* close to the nation's most historic monuments and museums - great place to bring the family!

ASEE PRISM will update you with ongoing conference
details. Don't miss out on this exciting annual event!
For more information, feel free to contact ASEE Meetings and
Conferences Department at (202) 331-3530.

curriculum

MATHEMATICS IN THE

ChE CURRICULUM

JOHN R. DORGAN, J. THOMAS McKINNON
Colorado School of Mines Golden, CO 80401-1887

We have been active in incorporating Mathematica,
a multifunctional general computer programming
environment, into the chemical engineering se-
nior year curriculum at the Colorado School of Mines (CSM).
This integrated platform for symbolic, numeric, and graphi-
cal analyses has been used in the process control and reac-
tion engineering courses. Students in these classes have
had mixed but generally positive reactions toward its
use. In this article we will describe our experiences and
provide information that allows access to example prob-
lems posted on the Internet.
The use of Mathematica provides several advantages in
teaching chemical engineering concepts-graphical accu-
racy in problem solutions presented in class, the ability to do
more involved problems, and most importantly, discovery
learning on the part of the student. Furthermore, the multi-
media capabilities of Mathematica can be used to stimulate
student interest in subjects that are inherently heavily math-
ematical in nature. Finally, we believe that the incorpora-
tion of this program into the curriculum provides a gen-
eral engineering tool that can be useful to the students
throughout their careers.

John R. Dorgan is Assistant Professor of chemi-
cal engineering and petroleum refining at Colo-
rado School of Mines. He received his BS from
the University of Massachusetts at Amherst and
his PhD from the University of California at Ber-
keley. His research interests include phase trans-
formations in liquid crystalline polymers, self-as-
sembly of block copolymers at surfaces, polymer -
encapsulation of mixed wastes, and pedagogical
methods in chemical engineering.

J. Thomas McKinnon is Assistant Professor of
chemical engineering and petroleum refining at
the Colorado School of Mines. He received his
BS in chemical engineering from Comell Uni-
versity in 1979 and his PhD from MIT in 1989.
He teaches kinetics and thermodynamics at the
undergraduate and graduate levels. His research
interests are in combustion chemistry and ap-
plied industrial pyrolysis.
Copyright ChE Diviston ofASEE 1996

COMPUTER INTEGRATION AT CSM
Our department's A. Bernard Coady Computing Labora-
tory is an excellent facility for such activities. It consists of
25 IBM RISC-6000 Model 220 computer workstations, a
Model 350 file server, a Model 560 computational server,
and a three-gun overhead projection system for the instructor's
workstation. The ability to integrate powerful computer pack-
ages into classroom use rests on the availability of appropri-
ate computer equipment, and we are truly fortunate to have
such facilities readily available.
Our interest in incorporating Mathematica was sparked by
Professor Stanley Sandler's essay, "Technological and Soci-
etal Change and Chemical Engineering Education."'" It
pointed out that while computers have revolutionized the
workplace, the biggest innovation in most college class-
rooms has been the introduction of the overhead projector.
In the essay, Professor Stanley predicted the coming of a
revolution in the delivery of education and educational meth-
ods due to changing technologies. We are in wholehearted
agreement with this assessment.
It should be noted at this point that the hardware require-
ments for Mathematica are not severe. The program can be
run on either an IBM compatible PC or on a Macintosh,
provided that there is 8MB of RAM. We have used the
program running on such machines coupled to a liquid crys-
tal display device that sits atop an overhead projector. This
allows quick in-class demonstrations, and the hardware nec-
essary for this type of implementation is modest in cost.
Such an arrangement works well and represents a viable
alternative to dedicated computer classrooms.

MATHEMATICS AT CSM
CSM, like many other colleges, has a site license for
Mathematica that allows an unlimited number of campus
machines to run the program. All students at CSM are intro-
duced to the capabilities of Mathematica in their freshman
calculus sequence.[2] Additionally, the program is used in
our elementary stoichiometry class for finding roots of alge-
Chemical Engineering Education

braic equations. This early integration into the curriculum
provides an excellent background for use at the senior level.
Mathematica was introduced to the process control class
in the spring of 1994 and into the reaction kinetics class the
following semester. Thus, the program has been in class-
room use for five consecutive semesters. During this
time, several example programs have been developed
that are of general interest to both the academician and
the practicing engineer wishing to explore the use of the
Mathematica platform.

WHAT IS MATHEMATICS?
Mathematica, in our opinion, is best described as a compu-
tational environment that allows symbolic manipulation, nu-
merical analysis, and powerful graphical representation. Ex-
amples of the symbolic manipulation capabilities include
differentiation and indefinite integration of functions, solv-
ing systems of linear algebraic or ordinary differential equa-
tions, and manipulation of Laplace transforms. Numerical
capabilities include fitting data with polynomials or splines,
root finding of algebraic equations, and solution of systems
of nonlinear ordinary differential equations through numeri-
cal integration. Graphical capabilities include 2-D and 3-D
representation and contour and parametric plots, as well as
animation of a series of sequential graphs.
The syntax of Mathematica is logical, but can be cumber-
some; we have found this to be the biggest stumbling block
for students in our classes. For example, the following com-
mand analytically solves a linear differential equation:

DSolve [{T' [t] a T[t] + b = = 0, T[0] = = 0}, T[t], t]

This command, translated, says to solve the differential equa-
tion

-I -aT+b= 0
ddt
subject to the following initial condition:
T(t = 0)= 0
Mathematica returns with the analytical solution in the form

{{T[t] b/a-(b*E^A(a*t))/a}

The mixed use of parenthesis, square brackets, and curled
brackets can be confusing for engineers used to program-
ming in FORTRAN.
Fortunately, the built-in Help Menu found in Mathematica
can be used to great advantage in overcoming syntactical
errors. This is due to the fact that within the Help Menu a
"Function Browser" can be invoked. Searching through this
Browser by subject allows the user to find the command of
interest along with a short description of its function. In
addition, the syntax of the command is given and may be
pasted directly into the program being written. This obviates
Spring 1996

... Mathematica provides several advantages
in teaching chemical engineering concepts-
graphical accuracy in problem solutions presented
in class, the ability to do more involved problems,
and most importantly, discovery learning
on the part of the student.

the need for directly typing in command syntax.
The resources associated with developing Mathematica
applications are considerable and can be used to great ad-
vantage when incorporating the program into the chemical
engineering curriculum. Most significant is the existence of
MathSource, an electronic bulletin board for public domain
programs and example problems. The bulletin board can be
accessed through the Mathematica home page using any
web browser. The URL for MathSource is
http://www.wri.com.mathsource
All of the usual web searching capabilities are supported
at this location.
In addition, the bulletin board may be accessed by e-mail
at mathsource@wri.com. Searches for topics of interest can
be performed by simply sending the e-mail message "Find
subject" where subject is the field of interest (chemical engi-
neering, perhaps). For general help in using MathSource,
send the above one-line message with intro as the subject. In
addition to the applications bulletin board, the program is
supported by a very competent technical staff that can be
contacted regarding questions and possible bugs at the e-
mail address support@wri.com.
MATHEMATICS IN PROCESS CONTROL
Our use of Mathematica in the process control class cen-
ters on the subjects introduced in the early part of the semes-
ter. Roughly the first one-third of the course deals only with
process dynamics; formulation of state models and calcu-
lation of uncontrolled response to input changes are ad-
dressed during this time period. Experience shows that
the program can be used to great advantage in demon-
strating the difference between nonlinear process models
and the corresponding linear models that form the basis
of modern control theory.
A relevant example of this type of use can be taken from a
homework assignment during the spring 1995 semester in
which the students were asked to compare the behavior of
draining liquid tanks of varying geometry (cylindrical, coni-
cal, and spherical) for the case in which the outflow is
proportional to the square root of the liquid level. The pro-
cess model for the cylindrical tank appears in most text-
books, the conical tank model is developed in lecture, and
the spherical tank model is assigned as an earlier, indepen-
dent homework assignment. A simple representation of the
tank geometries and the resulting process models is given in
137

Figure 1 (next page). In these models, Fi is the
volumetric inlet flow, Fo is the outlet flow, p3 rep-
resents the discharge coefficient, h is the liquid
level height in the tank, and h, is the initial steady-
state level.
The objective of the Mathematica assignment is
to compare the full nonlinear process behavior to
the linearized approximation and to draw conclu-
sions about the severity of the linearization ap-
proximations.
For this particular problem, an example solution
for the cylindrical tank is provided as part of the
problem statement. The heart of the assignment
consists of using only three Mathematica com-
mands: DSolve[] for solving the linearized equa-
tions, NDSolve[] for numerically solving the non-
linear equations, and Plot[] for comparing the so-
lutions. Figure 2 is part of an actual program list-
ing for a solution submitted by one of the students.

I. Cylindrical Tank
Fi A. Nonlinear process model
| dh + 1l/2 Fi
dt A A
B. Linearized process model
dh P h /2 + (h-h)=F
dt A 2Ah h/2 ) A
h F= 04h

11. Conical Tank

III. Spherical Tank

From this straightforward application, students
learn that linear approximations may either under-
or overestimate the actual system response (in this
case, the time required to empty the given tank). In
addition, they experience the phenomena that some
systems deviate more quickly from their linear
approximations than do others and that this de-
pends on the strength of the nonlinearities present
in the process. The type of discovery learning which the
program allows is thought to be advantageous toward stu-
dent learning and content retention.131
Another example from the process control class involved
the use of the symbolic capabilities for manipulation of
Laplace transforms. The assignment in this case was to write
a general purpose simulator for the response of any linear
second-order system and to test the response of the system
toward different types of forcing functions. This assignment
was given in the fall 1994 and spring 1995 semesters.
As a preface to this assignment, the students are given an
example program that constitutes a general purpose simula-
tor for a first-order system. Recall that the standard form for
a linear first-order system is

Tp +y=K f(t)

where up is the process time constant, Kp is the static gain,
and f(t) is the forcing function. The resulting transfer func-
tion is

G(s)- p
TpS + 1

where s represents the independent variable in Laplace space.
The demonstration program allows the user to input his or
her own function for f(t). This user-defined forcing function
138

A. Nonlinear process model
dh +____ -3/2 Fi
dt .tan 20 h2 tan2 0
B. Linearized process model
dh + h-3/2 (3 P
dt itan20 s -2 h/2tan20

A. Nonlinear process model
dh 1P F,
dt r(2R-h)h1/2 ut(2R-h)h112
B. Linearized process model

2F i,, Fi
h atanB29 h lan26

dh, 3 (,Rh I/2 -1.5hl/2 2 F,,(R-h,) (h-
dt n(2R-h-)h/2 h 2Rhl/2 h /2)2 2Rh- 2-h2
F2
1l(2R-h,)hs

Figure 1

is then transformed, multiplied by the transfer function, and
inverted to give the time domain response of the system to
the given forcing function (which is defined as a plot versus
time). The students are asked to run this program for a
variety of different forcing functions: step inputs, ramps, and
sine waves. In addition, the students run the program for
different values of the system parameters Kp and rp, thus
experiencing the meaning of these quantities directly. This
type of exploration learning can be highly effective.
Based on the example given, the students are asked to
rewrite the program to be applicable for second-order sys-
tems. The standard form of the transfer function is this case
is
Kp
G2(s)=2 p
Ss2 + 2ts+1
where is the damping coefficient, while Tp and Kp and f(t)
have the same meanings as above. For second-order sys-
tems, when the damping coefficient is less than unity, oscil-
latory behavior is observed.
While the students' solutions generally mimic the ex-
ample program, we wish to emphasize that we are not trying
to teach Mathematica programming, per se, but rather are
using the capabilities of the program in order to demonstrate
concepts in chemical engineering. Again, the students are
asked to test their simulator with a variety of forcing func-
Chemical Engineering Education

Figure 2

tions for different parameter values. The responses of a
second-order system to a unit step input for both overdamped
( > 1) and underdamped ( < 1) cases are explored.
The ability of students to explore system dynamics using
these approaches is important in making the connection be-
tween process models and real-world behavior. Later in the
semester, the excellent process simulation capabilities of
PICLES are used as part of the class.141 Thorough prepara-
tion in the fundamentals means the students are better pre-
pared for logically connecting the mathematics of process
modeling with observable system dynamics.

MATHEMATICS IN REACTION KINETICS

We introduced Mathematica in our senior-level kinetics
and reactor design class in the fall 1994 semester. Solving
differential equations that result from the reactor design
equations is a large part of this class, along with root finding
and plotting of results. In previous semesters, we used
Polymath for these tasks.'51 This program has the advantage
of being very easy to use; we found that the students needed
Spring 1996

(* PART 2 THE CONICAL TANK*)
Clear [Fi, A, he, beta, theta, Fis]

(* Define a function which is the linear O.D.E.*)
conf[t_] := conh' It] Fis hs)-2 + beta/(N[Pi] N(Tan [theta])2]) hs^(-3/2)
-hs^-2 (Fi-Fis)-(3 beta/(2 N[Pi] N[(Tan [theta])^2]) hsA(-5/2) -
2 Fis hs^-3) (conh[t]-hs)
(* Solve the linear O.D.E with the initial tank height. *)
sol3 = DSolve[(conft[t]==o0, conh[0]==hs},conh[t],t];
(* Define a function which can be plotted. *)
conhsolve[t] := conh[t] /. sol3[[l]]
(* Define a function which is the non-linear O.D.E. *)
connlf[t_] := connlh'[t] + beta/(N[Pi] N[(Tan [theta])^2]) connlh[t]^(-3/2) -
Fi connlh[t](-2)
(* Numerical solution requires parameter values. *)
Fi = 0;
Fis = 0;
beta =1;
hs = 5;
theta = N[Pi]/6;
(* Solve the non-linear O.D.E. with the initial tank height set to hs m. *)
sol4 = NDSolve [{connlf[t]==0,connlh[0]==hs),connlh[t],{t,0,23)];
(* Convert the numerical solution to a function which may be plotted. *)
connlhsolve[t_] := connlh[t] /. sol4
(* Plot the two functions for comparitive purposes. *)
Plot[{(conhsolve[tl,connlhsolve[t]},(t,0,23}];
5|k

3 1. Chemical equilibrium The first week of our
class is a review of chemical reaction equilibrium. The
FindRoot[] function in Mathematica is used to evaluate
the roots of high-order polynomials to solve for equilib-
rium conversion. The programming environment in
Mathematica allows us to easily explore the effect of
temperature changes on equilibrium composition. One
difficulty for new users is in discovering which, appar-
ently equivalent, Mathematica function is appropriate
for a given problem. In the case of reaction equilibrium,
both NSolve[] and FindRoot[] should work, as described
by Wolfram.'71 The former function finds all the roots to
a polynomial, while the later finds a root to any function
over a specified interval. It is only by trial and error that
the user discovers that NSolve[] rarely works for these
problems and that FindRoot[] is the more useful approach. This
highlights the importance of the instructor directing the students in
lecture before they start out on a problem.

3 2. Comparison of ODE solution methods Since solving
ordinary differential equations (ODEs) is such a large part of the
class, we spend one entire homework assignment exploring differ-
ent solution methods. We start with a relatively simple first-order
batch reactor problem with the reaction steps
A -> B -- C
The solution for [B] as function of time can be found analytically
by hand using the integrating factor method. Next, we move on to a
more complicated reaction network with parallel reactions and
reversibility:
B
A
C
This system of ODEs has an analytical solution, but it would be
rather difficult to solve by hand (and solving difficult ODEs is not
the objective of our course). For this job, we turn to the NSolve[]
function in Mathematica. The raw output of NSolve[] for this
139

little more than a ten- or fifteen-minute introduc-
tion to the program in lecture to use it effectively.
But we found that the ODE solving routines in
Polymath are not as robust as those in Mathematica,
particularly with regard to the stiff differential
equations that routinely arise in chemical kinetics
problems. Furthermore, the programming envi-
ronment allows more complex problems to be con-
structed; the results of one calculation (e.g., a
numerical integration) can be fed directly into the
subsequent one (e.g., finding a root). The price to
pay for this increased usefulness is, of course, a
much steeper learning curve.
We describe below some of the homework prob-
lems for which Mathematica was used. It should
also be noted that our textbook, Elements of Chemi-
cal Reaction Engineering[61 also lists a few
Mathematica examples.

system is exceedingly cumbersome, but with the use of the Sim-
plify[] function, it becomes at least somewhat manageable.
We then use this same reaction network to explore numerical
solutions to ODEs. The existence of the analytical solution allows
the students to examine the errors that can arise in numerics. They
first do a solution by implementing Euler integration using a spread-
sheet. In the lecture discussion of Euler integration, we also show
how Runge-Kutta methods improve the Euler integration results.
The integration stepsize can be varied and the resulting time pro-
files are compared to the analytical solution. Finally, we solve the
same system numerically, using the NDSolve[] function. Here there
is no discernible difference between the analytical and the numeri-
cal result.

3 3. Multiple reaction with phase change Our most ambitious
Mathematica homework problem was to explore the effect of tem-
perature on a system that includes both multiple reactions (both
series and parallel) and phase change. We postulated that a new
methanol synthesis catalyst had been discovered that has reason-
able activity around 400K and 50atm. The reaction considered are

CO+22 H2 CH3OH

CO + H20 <=> CO2 + H2

CH3OH = CH20 + H2

Rxn.l

Rxn.2

Rxn. 3

Arrhenius parameters are provided for all the reactions. The stu-
dents find the reverse rates of Rxns. 1 and 2 from the reaction
thermodynamics. The feed is specified to be a mixture of CO, CO2,
H2, steam, and inert (N2). The volume of the reactor and the total
molar feed rate are specified. The vapor pressure of methanol is
computed from the Harlacher equation. Although the deviations
from ideality are high at these conditions, we still use the ideal gas
law for the purpose of simplification. The objective is to find the
temperature at which the methanol production rate is maximized.
Increasing the temperature obviously favors faster kinetics, but this
advantage is rapidly diminished by the approach to equilibrium of
Rxn. 1 and the more rapid product decomposition by Rxn. 3.
Furthermore, the condensation of methanol at low temperatures
pulls the equilibrium conversion of Rxn. 1 to the right.
A multistep solution process is required for this problem and can be
performed in the Mathematica environment. First, the students
must numerically solve the six coupled ODEs (one for each spe-
cies) for the gas-phase problem over the whole reactor volume.
Next, the volume Vc,, at which methanol first begins to condense,
must be identified from the methanol mole fraction and the solution
to the Harlacher equation. After condensation (V > Vct,), the rela-
tionships between moles and mole fractions are different, and a
second numerical integration must be carried out from Vcrit to Vfnl.
The process is repeated at different temperatures until the optimum
temperature is identified.

3 4. Unsteady flow problems Unsteady flow problems are ideal
as examples where all the terms of the mole balances must be
considered. One of our favorite problems of this nature is the
unsteady-state startup of three CSTRs in series (found in Fogler,161
Problem 4-32). The students write three coupled time-dependent
ODEs for the conversions in each stirred tank reactor and solve
them using NDSolve[]. Figure 3 gives a listing of the Mathematica
140

commands needed for this problem.

IMPLEMENTATION PROBLEMS

While we are in general very pleased with Mathematica
and plan to continue using it, we would be remiss in an
article of this nature if we failed to point out some of the
difficulties we encountered. Probably the greatest source of
frustration on the part of the students is the inscrutable error
messages that are spawned by Mathematica. The incorrect
definition of a variable within NDSolve[] generates many
lines of error messages giving no real hint of the problem,
while at the same time (and more troublesome) holding out
several red herrings leading to misplaced debugging efforts.
The second greatest source of frustration is that Mathematica
holds values of variables until they are explicitly cleared.
One solution to this problem is to include the function Clear[]
at the beginning of every Mathematica problem where every
variable used is included in the square brackets. When a
change is made to the notebook, the problem is rerun from
the top (menu commands: Action, Evaluate, Evaluate Note-
book). This action removes all previous values from the
variables and resets them to the desired value. The value
of carrying out this step, although somewhat cumber-
some, cannot be stressed enough.
The solution to the first problem is more difficult. We
generally assign it to a lack of practice using debugging
skills brought on by the increased use of software pack-
ages by our students. Although all students do take a
programming class (FORTRAN is taught at CSM), they
practice very little in subsequent classes. In many ways,
Mathematica can be considered to be a programming
language, thus the same debugging strategies can be used.
We find that it is necessary to include a discussion of
debugging methods in lectures when discussing problem
solutions using Mathematica.

CONCLUSIONS

Mathematica is a very powerful computational tool that
can be used successfully in the undergraduate chemical en-
gineering curriculum. We have used the program for several
semesters and in different classes at the senior level. Based
on our experiences and recent trends in the development of
new media for the delivery of education, we expect the use
of programs that can combine symbolic, numeric, and graphi-
cal capabilities to increase.
In the process control class, Mathematica has been
implemented in order to elucidate the difference between
nonlinear system response and the linearized approxima-
tion of this behavior. Additionally, the symbolic manipu-
lation capabilities associated with the Laplace transform
have been exploited in order to allow discovery learning
on the part of the students.
Chemical Engineering Education

The use of Mathematica in the reaction engineering class
has greatly expanded our horizons regarding the types of
problems we can address. The robust ODE solver in
Mathematica frees us from the constraint of artificially se-
lecting reaction parameters. The programming environment
within Mathematica allows results of one calculation to be
automatically cascaded into the next. Furthermore, these
assignments are somewhat open ended, and there are few

(* CR418 HW 7.1 rogler Problem 4-32 *)
(* Unsteady Flow Problem Startup of 3 Series CSTRs *)
(* Rxn: A + B --> C, elementary. Equimolar feed rate *3
(* Find the time required for the exit of the third *)
(* reactor to reach 99% of its steady-state value *)

(* Importantly Start every problem by clearing *)
(* previous values *)
Clear[dnaldt, dna2dt, dna3dt, fao, vdot, cal, ca2, ca3,
nal, na2, na3, nair, na2r, na3r, v, fao, vdot,
final, k]
(* Mole balances for species A in each reactor. *)
(* Writing these equations in the form: *)
(* -accum. in out + generation- aids debugging *)
dnaldt = fao vdot cal k cal^2 v;
dna2dt = vdot cal vdot ca2 k ca2A2 v;
dna3dt = vdot ca2 vdot ca3 k ca3"2 v;
(* Auxiliary equations. Relate all the variables in *)
(* problem to the dependent variables of the ODEs, and *)
(* assign values to the constants. "ca3ss" is the *)
(* steady-state value solved in a separate notebook. *)
cal = nal[tl/v; ca2 = na2[t]/v; ca3 = na3St]/v;
fao = 20; v = 200; vdot = 20; final = 65; k = 0.025;
ca3ss = 0.609;
(* NDSolve numerically evaluates the coupled ODEs *)
(* given the initial conditions. *)
sol NDSolve[ Inal' [t] dnaldt,
na2'[t] -l dna2dt,
na3'[t] -- dna3dt,
nal[0] =- 0,
na2[0] =- 0,
na3[0] == 0),
(nal, na2, na3),
(* The direct output of KDSolve is a List. The *)
(* Replace operator (/.) extracts the desired function*)
(* from the List. *)
nalr nal /. sol[[1]]; na2r na2 /. sol[[l]];
na3r na3 /. sol[[l]];
Plot[{nalr[t]/v, na2r[t]/v, na3r[t]/v), {t, 0, finall,
GridLines -> Automatic, Frame -> True,
FrameLabel -> ("Time (min)-, "Cone. (mol/L)-}];

Time (min)
(* Find the point at which the exit cone. of Reactor 3 is *
(* 99% of its steady-state value. *
FindRoot[na3r[t]/v == 0.99 ca3ss, {t, 0}]
(t -> 59.6826}

Figure 3.

software limits placed on the more ambitious students who
want to develop elegant solutions. Finally, the advanced
plotting capabilities of the program allow visualization of
the results for different cases.

Beyond the benefits of enhanced education in chemical
engineering, we believe that the type of computer environ-
ment that Mathematica represents is here to stay. As a result,
the exposure that students receive when using the program
introduces them to a computational tool that should prove
valuable to them in the future regardless of the specifics of
their career paths.

REFERENCES
1. Sandler, S.I., "Technological and Societal Change and Chemi-
cal Engineering Education," Phillips Petroleum Company
Lecture (1993)
2. Hagin, F.G., and J.K. Cohen, Calculus Explorations with
Mathematica, Prentice-Hall, Englewood Cliffs, NJ (1995)
3. Wankat, P.C., and F.S. Oreovicz, Teaching Engineering,
McGraw-Hill, New York, NY (1993)
4. Cooper, D.J., "PICLES," Chem. Eng. Ed., 27(4), 176 (1993)
and Cooper, D.G., "PICLES: The Process Identification and
Control Laboratory Experiment Simulator," CACHE News,
6-12 (1993)
5. Polymath, developed by M. Cutlip and M. Sacham. Avail-
able from CACHE Corporation, PO Box 7939, Austin, TX
78713
6. Fogler, H.S., Elements of Chemical Reaction Engineering,
2nd ed., Prentice Hall (1992)
7. Wolfram, S., Mathematica: A System for Doing Mathemat-
ics by Computer, 2nd ed, Addison Wesley (1991)

- APPENDIX -

Obtaining Examples Through e-mail

As stated in the body of this article, the MathSource bulletin
board may be accessed through the Internet at the e-mail address
mathsource@wri.com

To access the process control examples described in this article as
well as other control examples, send the e-mail message

Find Control
The MathSource server will send a listing of all examples contain-
ing these keywords. In the returned file, each of the listings will
have an identifying number (as of this writing, the specific numbers
for our own examples are not known).
One process control application that is already available is called
control.m. This public domain package allows students to easily
generate Bode and Nyquist diagrams, root locus plots, and most
other types of plots used in a control course.
To obtain copies of the examples of interest, simply send
MathSource another e-mail message asking it to send the appropri-
ate identifier. As an example, if the message

Send 0202-554-0011
is sent to the MathSource address, an ASCI text file of the
MathSource Technical Report will be returned to the requester
through e-mail. Further details of the downloading procedure may
be found in this document. 0

Spring 1996

laboratory

LOW-COST EXPERIMENTS IN MASS

TRANSFER

Part 2'

I. NIRDOSH, M.H.I. BAIRD2
Lakehead University Thunder Bay, Ontario, Canada P7B 5E1
Liquid-liquid (or solvent) extraction is an important
unit operation in mineral processing, in nuclear and 'Vibromixer'
non-nuclear waste treatment, and in the chemical and
pharmaceutical industries. It usually involves the transfer of -a 60 Hz
a solute from an aqueous to an organic phase, or vice versa. variable voltage
The two phases are generally immiscible with each other,
and one phase is dispersed in the other, which is known as
the continuous phase. This provides a large interfacial area
for mass transfer and improves the process kinetics. Conducive y Meter
Prediction of mass transfer coefficients for a solvent ex-
traction process becomes difficult because of the absence of
any accurate knowledge of the interfacial area and is compli-
cated further by the influence of ionic strength and surface
Phase 1 ....
contaminants on the area.
The study of extraction rates at the bench scale, with a
known interfacial area, has been carried out using batch cells Phase 2
) sample tap
Inder Nirdosh received his BSc and MSc in
chemical engineering from Panjab University
(India) and his PhD from Birmingham Univer- Figure 1. The apparatus.
sity (United Kingdom). He joined Lakehead Uni-
versity in 1986, and his research interests are
in the fields of mineral processing and electro-
chemical engineering. with rotary agitation. Special baffles are provided to ensure
that the liquid-liquid interface remains flat. The original
Lewis celltl was modified by Bulicka and Prochazka,12] who
Malcolm Baird received his PhD in chemical incorporated vertical baffles and a cylindrical perforated
engineering from Cambridge University in 1960. grid in the design in order to achieve greater turbulence
After some industrial experience and a post-doc- together with increased stability of the interface. A brief
toral fellowship at the University of Edinburgh, he
joined the McMaster University faculty in 1967. review of various stirred cells is given by Lo, et alp
His research interests are liquid-liquid extraction, The objective of this experiment is to introduce chemical
oscillatory fluid flows, and hydrodynamic model-
ing of metallurgical processes. engineering students to a simple technique of predicting
mass transfer coefficients using equipment with a well-de-
'Part 1 appeared in the last issue of CEE, Vol. 30, No. 1 (1996) fined interfacial area.
2 Address: ChE Department, McMaster University, Hamilton,
Ontario, Canada L8S 4L7 Copyright ChE Division of ASEE 1996
142 Chemical Engineering Education

THEORY
Solvent extraction of acetic acid from an organic phase
into an aqueous phase is taken as an example because it is a
non-reacting system. The concentration increase of acetic
acid in water is followed with time. The rate of acetic acid
transfer is assumed to be first order with respect to the
difference between the final (Ca. ) and the actual (Ca) solute
concentration. For a volume Va of the aqueous phase, the
rate of change of solute concentration (dCa / dt) can be ex-
pressed by

Va = kaA (Ca -Ca) (1)

which can be integrated as

c dC kaA dt
( ca va dt (2)
C, -C) Va

yielding

(n(Ca -Ca)= en(Ca -Ca- kaAt (3)
Va
where
Ca, initial concentration of acetic acid in water (generally
zero)
Ca acid concentration in water at time t
ka local aqueous phase mass transfer coefficient
A interfacial area that is kept constant during the experi-
ment
Ca aqueous phase acetic acid concentration at equilibrium

Equation (3) indicates that by plotting in(Ca Ca) versus
t, a linear line of negative slope (-kaA / Va) is obtained, from

Figure 2. A typical calibration curve.

The objective of this experiment
is to introduce chemical engineering students
to a simple technique of predicting mass transfer
coefficients using equipment with a
well-defined interfacial area.

which ka can be calculated.

APPARATUS
A sketch of the apparatus is shown in Figure 1. It is a
modification of the Lewis cell.41] A 1-liter glass beaker is
used as the extraction cell. The cell has a sample draw-off
tap at the bottom that is installed by glass blowing. Agitation
is achieved by a single-phase Chemapec Inc. Vibro Mixer
Model El-11043 operating at standard 60-Hz frequency and
powered through a Variac to change the amplitude of vibra-
tion (degree of agitation). Mixing is provided by two 4.6-cm
diameter stainless steel agitator discs mounted on the shaft
of the Vibro Mixer. The liquid streams outward from the
edges of the vibrating discs and circulates as shown in the
figure. Mixing within the phases can also be achieved by a
small impeller agitator provided the interface stays flat.

EXPERIMENTAL PROCEDURE
The organic phase may be either lighter (such as kerosene
as Phase 1) or denser (such as carbon tetrachloride as Phase
2) than water. The concentration of acetic acid in the aque-
ous phase at any time may be determined by either titrating a
small sample of solution or by monitoring the conductivity
of the aqueous phase, which would eliminate the need for
removal of aqueous samples for titration. Sample removal
has some effect on the value of Va, which is assumed to
remain constant during the integra-
tion of Eq. (2). Although the present
data have been obtained with carbon
S- tetrachloride, the use of a relatively
non-toxic organic phase denser than
water (such as methylene chloride,
CH2Cl2) is recommended. When us-
ing a lighter organic phase, the con-
ductivity probe should be submerged
in the aqueous phase before adding
the organic phase in the cell.
On the other hand, when a denser
organic phase is used, it should be
I- added to the cell first, then the water,
1' 1.o and then the conductivity probe
should be placed in the top aqueous
layer. A calibration curve can be
drawn for [CH3COOH] versus con-

Spring 1996

ductivity to facilitate analysis of results. A typical calibra-
tion curve is plotted in Figure 2 that indicates excellent
reproducibility and linearity in the [H*] range of 10-4 to 1 M.
The data plotted in Figure 3 indicate that the results obtained
with titrations versus conductivity measurements are within
+5% of each other.
Based on the experimental investigations followed for
this study, the following procedure is recommended for the
setup using CCl4 and conductivity measurements:

1. Take 3 L of CC14 and add about 26 mL of glacial
acetic acid. Mix well and use it as a stock solu-
tion. Titrate a 10-mL sample to determine the
exact acid concentration. Adding 10 mL of water
before titration is recommended.
2. Take 450-500 mL of the stock solution in the
extraction cell.
3. Adjust the Vibro Mixer so that the agitator discs
would be in the middle of
each phase.

4. Take the same volume
(450-500 mL) of water and
add it quickly to the cell,
pouring it over the top disc
rather than directly into the
organic phase. This will
keep the transfer of acetic
acid during start-up to a
minimum.
5. Place the conductivity
probe in the aqueous phase
and start the agitator at a
fixed voltage setting on the
Variac (fixed amplitude).
6. Start conductivity measure-
ments immediately, initially
using one-minute intervals
and later at longer intervals.
Discontinue measurements
when conductivity changes
are insignificant over a ten-
minute period.
7. Mix the two phases vigor-
ously, preferably in a
separatory funnel, to
establish equilibrium.
Measure the conductivity of
the aqueous phase at
equilibrium to determine
Ca- .
8. Using the predrawn calibra-

5.0
4.8
4.6
4.4
4.2
U 4.0
' 3.8
. 3.6
3.4
3.2
3.0

tion curve (Figure 2), obtain the concentration
values corresponding to the recorded conductivity
values.
9. Take a sample of each phase and titrate to deter-
mine the acid content. Use these values to verify
the mass balance and the concentration values
obtained from the calibration curve.
10. Repeat twice more, each time with a fresh batch of
CCl4 stock solution, but a different Variac setting.

11. Plot -in(Ca. -Ca) versus t for each of the three
cases and determine ka values from the slopes.

TYPICAL RESULTS AND DISCUSSION
Figure 4 is a plot of -in(Ca -Ca) versus t for three differ-
ent agitation rates. The plots are linear, conforming to Eq.
(3), and indicate that the slope (and hence the mass transfer
coefficient) in each case increases with the increase in rate of

U I I I I I I I I I I
10 15 20 25 30 35 40 45 50 55 60
t, min

Figure 3. Comparison of conductivity versus titration measurements for
obtaining acid concentrations.

t, mmin

Figure 4. Typical -In(Ca -Ca) vs. t plots.

Chemical Engineering Education

lTtration Method
Conductivity Method
__dqt ~to

L J

agitation. This is also depicted by the data (ka versus Variac
setting) plotted in Figure 5.
Figure 6 is a plot of ka versus the aqueous phase viscosity.
The changes in the phase viscosity were achieved by dis-
solving varying amounts of sucrose in water. The plot clearly
indicates the adverse effect an increase in the viscosity has
on the mass transfer coefficient.

CONCLUSIONS
The class should be divided into various groups.
> Each group should follow steps 1 through 11 above.
1- Various groups should use aqueous or organic phases
of varying viscosities (the organic phase viscosity may
be changed by adding various amounts of heavy

10 15 20
\Vriac Setting, V

Figure 5. Dependence of k. on degree of agitation.

10'

10-

10-'

A, Pa.s x 10

Figure 6. Dependence of k. on viscosity of
aqueous phase.

mineral oil to the organic phase) and submit a formal
report containing discussion on:
Comparison of n (Ca Ca) versus t curves for
various Variac settings
Dependence of k. on liquid viscosity

POSSIBLE FUTURE PROJECTS
In future years the experiments may include investigation
of a) the dependence of ka on temperature, and b) the analy-
sis based on the changes in acid concentration in the organic
phase rather than the aqueous phase, e.g., from the following
equations:

Vo = koA(Co -C) (4)

NOMENCLATURE
A interfacial area (cross-sectional area of the beaker)
C bulk phase concentration of acetic acid at any time
k local mass transfer coefficient
t time
V phase volume
Subscripts
a aqueous phase
i initial
o organic phase
equilibrium, final

ACKNOWLEDGMENTS
Financial support for this work was provided by the Natu-
ral Sciences and Engineering Research Council of Canada.
Thanks are due to Mr. J.S. Chatha for obtaining some of the
experimental data.

REFERENCES
1. Lewis, J.B., "The Mechanism of Mass Transfer of Solutes
Across Liquid-Liquid Interfaces. I. The Determination of
Individual Transfer Coefficients for Binary Systems," Chem.
Eng. Sci., 3, 248 (1954)
2. Bulicka, J., and J. Prochazka, "Mass Transfer Between Two
Turbulent Liquid Phases," Chem. Eng. Sci., 31, 137 (1976)
3. Lo, T.C., M.H.I. Baird, and C. Hanson, Handbook of Solvent
Extraction, John Wiley, New York, NY, pp. 113-115 (1983)
4. Nirdosh, I., and M.H.I. Baird, "Copper Extraction in a Vi-
brating Plate Cell," AChE Symp. Ser., No. 173, Vol. 74, 107
(1978) 0

ERRATUM
In the book review, Chemical Thermodynamics: Basic
Theory and Methods, 5th ed., by Irving M. Lkotz and
Robert M. Rosenberg, reviewed by Pablo G. Debenedetti
in Chemical Engineering Education, 30(1), 69 (1996),
the sixth sentence of the fifth paragraph should read
Similarly, the definition of an ideal gas as one
satisfying PV = RT and, in addition, having a
volume-independent energy is redundant.

Spring 1996

Survey

CURRENT TRENDS IN

CHEMICAL REACTION ENGINEERING

EDUCATION

MAZEN SHALABI, MUHAMMAD AL-SALEH, JORGE BELTRAMINI, DULAIHAN AL-HARBI
King Fahd University of Petroleum and Minerals Dhahran 31261, Saudi Arabia

C chemical engineering is a practical specialization com-
bining several fundamental scientific disciplines within
the field of engineering in order to solve the problems
and challenges facing mankind. It evolved around the begin-
ning of this century in response to the world's increasing
industrialization. Required courses for chemical engineering
majors almost universally include a sequence of
multidisciplinary courses. Chemical reaction engineering (CRE),
the study of the basic knowledge of chemical kinetics and
reactor design, is the most important area of study and is the
main area that distinctly characterizes chemical engineering
from other engineering disciplines.
CRE also uses areas (such as thermodynamics, transport
phenomena, and other related disciplines) in analyzing small-
and large-scale reaction systems. In a recent article,
Doraiswamy'" addressed the fact that students majoring in
CRE not only must have a proper background in the funda-
mentals (such as ideal reactor design), but they should also
tackle new and challenging areas, such as biochemical, micro-
electronic, polymer, and electrochemical reaction engineering.
He also stressed the importance of CRE education as a possible
"interdisciplinary single umbrella" to cover all these different
disciplines. If one basic course is not enough, Savage and
Blaine"2' have proposed that additional elective courses should

Mazen A. Shalabi received his BS and MS degrees in Chemical Engi-
neering from Cornell University and his PhD in Chemical and Petroleum
Refining from Colorado School of Mines. His major research interests
are heterogeneous catalysis, kinetic modeling, and chemical reaction
engineering.
Muhammad AI-Saleh received his BS and MS in Chemical Engineering
from King Fahd University of Petroleum & Minerals, and his PhD from
Colorado School of Mines. His currents interests include catalysis, reac-
tion engineering, and electrochemical reaction engineering.
Jorge N. Beltramini has taught undergraduate and graduate courses in
thermodynamics, reaction engineering, catalysis, and mass transfer for
more than fifteen years. His areas of interest cover petroleum refining,
petrochemical processes, and environmental catalysis.
Dulaihan K. AI-Harbi received his BS and MS degrees in Chemical
Engineering from the Pennsylvania State University and his PhD in
Chemical Engineering from Oklahoma State University. His current inter-
ests include thermodynamics, transport properties, and fluid flow.
Copyright ChE Division of ASEE 1996

be introduced to cover these areas.
Another important CRE area is catalysis, especially hetero-
geneous catalysis-it has evolved as a consequence of the
widespread use of the many catalytic reaction systems in in-
dustry and is not usually covered in depth in engineering
courses. Heterogeneous catalysis courses and their importance
in engineering education have been reviewed by Vannice131
and Miranda.[4] They combined the traditional material of ca-
talysis with the more advanced knowledge of solid state, sur-
face chemistry, and material processing, such as sol-gel tech-
nology and chemical vapor deposition.
A more thorough review of CRE education can be found in
an article by Dudukovic151 comparing the results of a 1982
AIChE survey of CRE courses in US and Canadian chemical
engineering departments with a similar study completed by
Eisen'61 in the early 1970s.
This article reports the results obtained from a similar sur-
vey, conducted during the first part of 1993, that involved more
than a hundred chemical engineering departments worldwide.
Recent trends in CRE related to type of courses offered, teach-
ing material, and textbooks used on both undergraduate and
graduate levels, will be compared with the results of
Dudukovic's previous survey.

QUESTIONNAIRE DESIGN AND ANALYSIS
The main objective of the survey was to determine what is
taught as CRE at both the undergraduate and the graduate level
in chemical engineering departments in North and South
America, Europe, the Middle East, Asia, and Australia. A
questionnaire was designed that included two main streams of
questions: the first part consisted of eight questions and was
primarily concerned with teaching and organization of the
undergraduate and graduate courses; the second part dealt
principally with the main and reference textbooks and the level
of satisfaction with current teaching material, including the
type of PC software packages. At the end of the questionnaire
the departments were asked to give their perception of the
Chemical Engineering Education

future of CRE education for the next ten years.
A total of 137 questionnaires were sent out: 58 to U.S.
chemical engineering departments and 79 to the other world
areas mentioned above. The response rate was roughly 69%,
with returns from 40 US departments and 55 from the other
countries.

COURSE CONTENT
Figure la shows the number of CRE courses available for
undergraduate students. The majority of schools have only one
compulsory course (66%) while only 35% have at least one
elective undergraduate course. This finding is similar to
Dudukovic's survey of thirteen years ago showing that 69% of
the schools had at least one CRE course available. On the other
hand, Figure lb shows that
42% of the schools have only A.Undergraduate cou
one compulsory course for
graduate students, compared one
to 64% thirteen years ago. None-

Figure 2 compares the dis-
tribution of undergraduate
CRE courses (compulsory
and elective) with the geo-
graphical distribution of the
chemical engineering depart-
ments around the world. A
very different distribution can
be seen in the case of com-
pulsory courses (Figure 2a).
In the US, nearly 90% of the
schools have one compulsory
course, while in the United
Kingdom practically all the

Two-26 /.
Compulsory
B. Graduate courses

schools have two compulsory courses. A similar trend is ob-
served in Asian countries, with 77% of the schools having two
compulsory courses. On the other hand, nearly 20% of the
continental European schools do not have compulsory courses
at all. For the other regions (Middle East, Canada, Australia,
New Zealand, and South America), the course distribution is
between one and two compulsory courses, ranging from a
minimum of 30% (Middle East) to a maximum of 75% (Canada)
for one compulsory course. With reference to elective courses,
the figure shows that with the exception of South American
countries, there is an average tendency of 50% for not offering
an elective course in CRE, with a minimum of 25% for Asian
countries and a maximum of 80% for the United Kingdom.
With regard to the average class size for undergraduate
courses, most of the de-

irses

Three or more
5 /.
Two 4%.

One-33%/
Electives

Three or more
13..

One-42 /. T
Two- 23/.
Compulsory Electives

Figure la,b. Number of CRE courses available for under-
graduate and graduate students.

apartments (51%) have
classes of forty or more
students (see Figure 3a),
while only 5% have
classes of less than
twenty students. The
majority of departments
(75%) offer two to three
hours of lecture per week
(see Figure 3b), but it is
important to note that
51% have less than two
hours tutorial/problem
solving per week, as
shown in Figure 3c. Fig-
ure 3d summarizes the
answers to a question re-
garding the number of
laboratory experiments:

() COMPULSORY

75i 5 /

... ,... ......

0* - -lj~

( ELECTIVES

F G H

- NONE 3ONE LCMTWO MS THREEORMORE

B C D E F G H
PLAC ES
1 NONE I"ONE E"TWO ES THREE OR MORE

A:Middle East
E* Europe

B: Asia

C: Australia /New Zealand

F South America G United Kingdom

D :Canada
H : USA

Figure 2a,b. Worldwide distribution of undergraduate CRE courses.

Spring 1996

A B C D E
PLACES

it can be seen that only 12% of the departments require at
least one experiment during the teaching of the CRE
course, and that 46% of the departments have no labora-
tory experiment.
Figure 4 compares class structure as a function of
geographical distribution of the schools. Regarding class
size, two well-defined groups were observed. One group,
the Middle East and South American countries, have
small classes (up to thirty), while the rest of the world
accomodates larger classes (forty or more). Concerning
time spent on lectures during the week, Figure 4b shows
equal distribution for all locales, with two to three hours
a week devoted to lectures. But there is ample disparity
in the time distribution for tutorials per week as well as in
the number of experiments performed, as can be seen in
Figures 4c and 4d, respectively. Analysis of the data
concerning the number of experiments performed (Fig-
ure 4d) shows that 70% of all the Middle East and US
schools do not have experiments, but that 25% of the
South American schools have at least one experiment.
Table 1 summarizes the basic concepts covered during
the undergraduate course. Kinetics and mechanism, along
with interpretation of kinetic data, are regarded as the
most important concepts by 54% of the schools, fol-
lowed by ideal reactor design and catalytic reactor de-
sign, by 46% and 43%, respectively. Only 23% of the
schools ranked the importance of non-ideal reactors. In
general, the different departments expressed that homo-
geneous systems receive more attention than heteroge-
neous systems, but as many as 18% of the departments

Fogler's textbook was his structured approach to teaching CRE that
involves problem solving and decision making techniques. Some of the
departments also cited the material covering the emerging areas (micro-
electronics, biotechnology, and polymerization reactions) as a reason for
changing. Still another reason for the change was the use of computer
software in solving the prescribed textbook exercises.
For graduate courses, Froment and Bischoff's textbook is the most
popular (33.7%), while nearly 27.3% of the schools actually produce
their own teaching material. Most of the instructors (92% undergraduate,

(a ) ( b )
LESS THAN 20 LESS THAN 2
20- 2 9 8 / 6/*
255. OVER 3
2 T03 20%/.
76%'.

30-39
'/* 40ANDHIGHER

UNDERGRADUATE CLASS SIZE UNDERGRADUATE LECTURETIME(h)
( C ) ( d)
NONE 12%/ FOUR OR MORE
OVER 3 NON E 2 O /.

2 TO 3 THREE
34 10%/.
LESS THAN
12%/. 12
UNDERGRADUATE TUTORIAL TIME (h) UNDERGRADUATE EXPERIMENTS

Figure 3. Class structure on undergraduate courses.

do not cover heterogeneous systems at all in
the compulsory undergraduate CRE. As far
as industrial input is concerned, 32% of the
departments claimed it as a very important
part of the course.

TEACHING MATERIAL
AND RECOMMENDED TEXTBOOKS

Table 2 shows the textbooks used for both
undergraduate and graduate CRE courses.
Eleven known textbooks were listed in the
survey.[7-17] For undergraduate courses,
Fogler's textbook is the most popular, pres-
ently used by 41% of the schools. This re-
placed Levenspiel's textbook, written 23 years
ago, that once enjoyed a popularity of 58%
but which is presently used by only 25.3% of
the departments. It is interesting to note that
nearly 10% of the departments use their own
textbook. They claimed that this allows them
to properly match the CRE undergraduate
curriculum with their time and place needs.
The main reason cited for the change to
148

CLASS SIZE

ii ILL--Y

A B C 0 E F G H
PLACES
0 LESS THAN 20 L 20-29
E 30-39 1 40AND HIGHER
(a)

TUTORIAL TIME A WEEK ( h)

L -
.,I i

A 8 C D E F G H
PLACES
NONE ELESSTHAN2EZ 2 TO3 Eas

LECTURE TIME A WEEK(h)

MLESSTHAN2 M2 2TO3 E- OVER 3
(b)

NUMBER OF EXPERIMENTS

_ I i. 4 1
A B C D E F G H
PLACES
I NONE OM ONE C TWO
EM THREE U2 FOUR OR MORE
(d)

Figure 4. Class structure as a function of geographical distribution
of the schools.
Chemical Engineering Education

83% graduate) view their present textbook is satisfactory. Forty-
four percent changed their textbook in the last five years, but
only 25% of undergraduate and 18% of graduate courses are
considering a change in the next two years.
Concerning development of software material for problem
solving, 60% of the departments have developed or purchased
computer software made available through main frame and
personal computer facilities. The distribution by regions shows
very clearly that in the case of US and Canada, 66% of the
schools adopted this approach, while only 15% of the schools
in Australia, New Zealand, South America, the Middle East,
and Asia adopted it. In Europe and the United Kingdom, the
percentage of positive answers was nearly 50%. The types of
software ranged from commercial mathematical programs (such
as Polymath and Mathematica) to more sophisticated software.

CURRENT AND FUTURE TRENDS
In answer to a question relating to the future of CRE courses
in the next ten years, the following were the main points
addressed by the departments:

TABLE 1
Course Dedication to Various Key Concepts in CRE

Topics
Kinetics and mechanisms
Interpretation of kinetic data
Reactor design
Non-ideal reactors
Kinetics of catalytic systems
Diffusion and reaction in
heterogeneous systems
Catalytic reactor design
Industrial oriented examples

Very Not
Important Important Average Inportant
54 28 15 3
54 33 13 -
46 21 30 3
23 44 32 1
41 43 15 1

TABLE 2
Undergraduate and Graduate Textbook Distribution

Textbook

1. Smith, J.M.t71
2. Levelspiel, O.'"
3. Fogler, S.H.191
4. Hill, Sr., C.G.1101
5. Froment, G.F./Bischoff, K.B.t1"
6. Carberry, J.J.[121
7. Wallas, S.M.t131
8. Holland, C.DJ Anthony, R.G."141
9. Cooper, A.R./Jeffrey, G.V.151
10. Denbigh, K.GfTumer, J.C.R."61
11. Nauman, E.B."J'
Own Text

Mai Tet Rdferace Text
Undergrad Grad Under Gra

- 3.0
1.3 1.5 3.5
- 1
1.5 8
1.5 1.5 2
10.0 27.3 5.4

Of increasing importance would be computer applications and
software packages, with the introduction of more computer-as-
sisted problem solving and experimentation as well as modeling
real reactor operations.
More emphasis should be given to non-ideal reactors and to
heterogeneous reactor design.
Some aspects of heterogeneous catalysis should be covered in
depth.
Newer applications and technologies such as biochemical engi-
neering, pollution control, and electrochemical reactors, should be
introduced.
More industrial examples with realistic data should be used.

SUMMARY AND RECOMMENDATIONS
In a field that covers such a large mix of possibilities, it
would be presumptuous to list areas for continued or future
attention. Even so, there are certain areas that have the poten-
tial for significant impact on the current and future chemical
industry: catalysis and catalytic reaction engineering, solid
state reaction engineering, mineral processing, etc.

ACKNOWLEDGMENT
The authors acknowledge the financial support of the King
Fahd University of Petroleum & Minerals for the performance
of this study.

REFERENCES
1. Doraiswamy, L.K., "Chemical Reaction Engineering: A Story of
Continuing Fascination," Chem. Eng. Ed., 26(4), 184 (1992)
2. Savage, P.E., and S. Blaine, "Chemical Reaction Engineering Appli-
cations in Non-Traditional Technologies: A Textbook Supplement,"
Chem. Eng. Ed., 25(3), 150 (1991)
3. Vannice, M.A., "A Course on Heterogeneous Catalysis," Chem. Eng.
Ed., 13(4), 164 (1979)
4. Miranda, R., "Heterogeneous Catalysis," Chem. Eng. Ed., 23(2), 166
(1989)
5. Dudukovic, M.P., "Survey of Chemical Reaction Engineering
Courses," Chem. Eng. Prog., p. 25, Feb. (1982)
6. Eisen, E.O., AIChE Report, "Teaching of Undergraduate Kinetics,"
67th AIChE An. Meet., Washington DC (1974)
7. Smith, J.M., Chemical Engineering Kinetics, 3rd ed., McGraw-Hill
(1981)
8. Levenspiel, 0., Chemical Reaction Engineering, 2nd ed., Wiley and
Sons (1972)
9. Fogler, S., Elements of Chemical Reaction Engineering, 2nd ed.,
Prentice Hall (1991)
10. Hill, Sr., C.G., An Introduction to Chemical Engineering Kinetics
and Reactor Design, Wiley & Sons (1977)
11. Froment, G.F., and K.B. Bischoff, Chemical Reactor Analysis and
Design, 2nd ed., Wiley & Sons (1990)
12. Carberry, J.J., Chemical and Catalytic Reaction Engineering,
McGraw-Hill (1976)
13. Walas, Reaction Kinetics for Chemical Engineers, McGraw-Hill (1959)
14. Holland, C.D., and R.G. Anthony, Fundamentals of Chemical Reac-
tion Engineering, 2nd ed., Prentice Hall (1978)
15. Cooper, A.R., and G.V. Jeffreys, Chemical Kinetics and Reactor
Design, Prentice Hall (1973)
16. Denbigh, K.G., and J. Turner, Chemical Reactor Theory: An Intro-
duction, 3rd ed., Cambridge University Press (1984)
17. Nauman, E.B., Chemical Reactor Design, Wiley and Sons (1987) 0

Spring 1996

Curriculum

CHEMICAL ENGINEERING EDUCATION

IN TURKEY

AND THE UNITED STATES

J. RICHARD ELLIOTT, JR.
The University of Akron Akron, OH 44325-3906

A few years ago, Floyd"13] wrote an interesting series
of articles comparing the chemical engineering edu-
cational systems of the US and Japan. As an under-
graduate at Tokyo Institute of Technology and a graduate
student at the University of Wisconsin, he was able to iden-
tify a number of striking differences between the two sys-
tems. In this article, I would like to reconsider a number of
Floyd's observations in relation to another country-Tur-
key. This presentation goes beyond Floyd's presentation in
providing more systematic comparisons between the stu-
dents' performances and backgrounds. The expectation is
that seeing a number of educational systems juxtaposed in
this way can lend some insight into the strengths of each
system and suggest improvements for all.
According to Floyd, the most notable differences between
the American and Japanese systems were related to the for-
midable entrance exams in Japan and the practice of attend-
ing intensive preparatory schools ("junku" in Japanese) dur-
ing the period of secondary education. This intensive prepa-
ration evidently led to two significant outcomes: 1) greater
preparation of the entering students made it possible to place
much of the technical content earlier in the curriculum, and
2) a person's performance in college was less important than
the college attended, so most of the college experience was
considered as a time of rest (a kind of "relaxation" phenom-
enon). Since the Turkish system has a similar pre-college
program ("dershane" in Turkish), one point of interest
was to determine whether a number of Floyd's observa-
tions regarding the Japanese students might also be rec-
ognized in Turkish students.
There is a more broadly global justification for consider-
ing a country such as Turkey in relation to countries such as
the US and Japan. If you made a list of all the countries with
chemical engineering departments and sorted them accord-
ing to gross domestic product per capital, the US and Japan

J. Richard Elliott is Associate Professor of
Chemical Engineering at The University ofAk-
ron, where he has taught since 1986. He holds
chemical engineering degrees from Penn State
(PhD) and Va Tech (MS) and a BS degree in
math/chemistry from Newport College. His
interests are primarily in molecular thermo-
dynamics and related applications. For the
1994-95 academic year, he served as a
Fulbright Lecturer at Bogazici University in
Istanbul, Turkey.

might not be considered as being globally representative.141
Countries such as India, China, and Argentina share mea-
sures like gross domestic product per capital more closely
with Turkey than with the US or Japan. Thus, one might
hope that the insights gained with respect to Turkey should
reflect similar insights that might come from studying any
one of many countries around the world.
This paper compares the curricula and, perhaps of most
interest, characterizes the differences between the students
by means of quantitative comparisons. Specifically, I have
conducted the same course in thermodynamics at Bogazici
University in Istanbul, Turkey, that I have taught for eight
years at the University of Akron. A detailed description of
my personal emphasis in this course was previously pre-
sentedV5I but, for the most part, this course represents a stan-
dard course in the chemical engineering curricula world-
wide. Both sets of students used the same primary text and
syllabus, had access to the same computational facilities,
and faced identical examinations. By comparing perfor-
mance on identical examinations simultaneously with
other indicators of performance and background, a con-
nection can be drawn between the local system and its
overseas counterpart.
The essential computational resources were made avail-
able through the generosity of the Fulbright Program in that

Copyright ChE Diuvsion ofASEE 1996

Chemical Engineering Education

they funded programmable cal-
culators for every student in the
Turkish course. Normally, Akron
students are expected to purchase
their own calculators with suffi-
cient RAM (32 KB) to support
programs for compressibility fac-
tor and departure function calcu-
lations as well as vapor-liquid
K-ratios and bubble point pres-
sure by the Peng-Robinson equa-
tion of state. The necessary pro-
grams are made available if the
students purchase either a Sharp
EL9300 or HP48G. Questions
that are greatly facilitated by
these programs are included on
the tests and final exam. Through
the Fulbright grant, all the
Bogazici University students
were provided with pre-pro-
grammed calculators that they
could keep and carry to the ex-
ams. In the Turkish system, it is
not typical for computational re-
sources to be so integrated into
the coursework at the under-
graduate level, but, in order to
give identical examinations, it
was necessary that the computa-
tional resources be equalized to
this extent.

BACKGROUND
ON THE CURRICULA
AND STUDENTS
The curricula of the schools
are compared in Figure 1. The
Bogazici University (BU) cur-
riculum resembles the University
of Akron (UA) to a higher de-
gree than the Tokyo Institute of
Technology (TI) curriculum. The
TI curriculum reflects a consid-
erably greater emphasis on gen-
eral subjects and foreign lan-
guages. The major difference be-
tween BU and UA is that the BU

This paper compares the
curricula and, perhaps of most
interest, characterizes the differences
between the students [of Bogazici
University and the University
of Akron] by means of
quantitative comparisons.

Sem. Bogaznici University, Turkey (151 sem. credits not including English)
0 English
0 (if preliminary test indicates need)
1 General Eng. Phys Chemistry, Physics, and Math
2 Subjects Core Ed.
3

6 1
7 Specalization/
8 Research Project

Figure 1. Outlines of undergraduate currucula.

curriculum places more emphasis on the specialization/
research option than the UA curriculum, with a corre-
sponding decrease in the general science emphasis. Both
the TI and BU curricula place significant emphasis on the
specialization/research option, whereas UA's research em-
phasis is relatively light. The most striking feature of the
Spring 1996

UA curriculum is the high pro-
portion of emphasis on general
science courses. The emphasis
on general science at UA limits
the time available for special-
ization/research.
A detailed comparison of the
courses shows that the BU stu-
dents are required to repeat the
freshman calculus, chemistry,
and physics at the same level as
their UA counter parts. This sug-
gests that not every country takes
advantage of the higher degree
of preparedness of students ad-
mitted by a competitive national
exam. One might suspect that
calculus and physics are not cov-
ered on the BU national exam,
but they are.
Another significant difference
between the three curricula is the
requirement of more total credit
hours to graduate from an over-
seas university-151 at BU vs.
136 at UA. These extra credit
hours are largely dedicated to
industrial chemistry courses. The
emphasis on industrial chemis-
try is similar to that in the TI
curriculum, although it is not
quite as intensive in Turkey.
Concerning the BU and UA
curricula, a couple of slight cur-
ricular deviations are relevant to
the thermodynamics course. The
BU students take chemical engi-
neering thermodynamics in the
fall of the junior year vs. the
spring of the sophomore year at
UA. Furthermore, the BU stu-
dents have had a full year of
physical chemistry prior to the
thermodynamics course. This de-
viation makes for a slight dif-
ference in the degree of pre-
paredness of the students, be-

yond their pre-college backgrounds. A small allowance
was made for this difference by adjusting the grade scales,
which will be discussed later.
It should be noted that BU has historical ties with the US
educational system, especially in that all courses included in
the four-year degree program are conducted in English. This
151

is also true of Middle East Technical University, another
engineering school in Turkey with a substantial number of
chemical engineering graduates. With this in mind, it is not
too surprising that the BU and UA curricula are so similar.
As for the student backgrounds, entrance into BU is ex-
tremely competitive. Students indicate the schools and de-
partments into which they would like to matriculate on their
test papers. Students are matched with departments accord-
ing to their performance on the exam and the availability of
positions in each department. In a recent instance, of roughly
1,200,000 applicants taking the national entrance exams, the
lowest score admitted to the BU chemical engineering pro-
gram belonged to the 1400th student from the top (approxi-
mately the top 0.1% on average).
On the other hand, once admitted, all courses are practi-
cally free of charge, and it is somewhat difficult to fail a
student from the curriculum. Furthermore, meals are subsi-
dized, making housing the only significant expense after
admission. To reduce housing costs, many stu-
dents commute long distances, living with fam-
ily members. It was suggested to me that the
underlying student educational capacities should
be roughly equivalent. The reasoning was that
the UA students should be motivated by their
more substantial tuition costs and fear of fail- I. The ase
ure. BU students, on the other hand, should be to scho
count t
predisposed to be successful based on the
selection process for admission, but the ad- 2. Al the b
vantage is somewhat nullified by the "relax- einee
ation" effect. It should be noted that most for pay,
UA students pay the bulk of their tuition
themselves, either through cooperative edu- 3. I am ia
cation or through part-time work. semester
Entrance to UA, like many US institutions, is 4. My cun
not based on a competitive exam. America is courseA,
known as the land of opportunity, and UA sub-
scribes heartily to this proud tradition. Our ad-
5 I found
missions procedure is to admit virtually anyone sigmnfic
who applies into the general program. Students matena
may take courses in any curriculum until the For exa
junior level, by which time they are expected to ,ocabu
achieve sufficient success to be admitted into a such th
degree program or to continue sophomore Agree'S
courses until they can be admitted. A natural 6 Mst.
consequence of this admission policy is a rela- 6 while a
tively high attrition rate. To illustrate, our foot-
7. Chemic
ball coach was once challenged because only field of
60% of his athletes were graduating. After about wanted
a week of his fumbling around, someone in- Agree,'
formed him that 60% was nearly double the 8. My car
University average! engine
busnes
While the admissions policy and attrition rate .lgee *
at UA are subjects of some concern, it is not

entirely obvious how to devise an alternative policy that
serves our mission of widely available public education. As
for typical results in the chemical engineering department at
UA, of 90 UA students initially registered for the sophomore
course in material and energy balances in 1992, 40 eventu-
ally graduated. Approximately 120 students expressed an
interest in chemical engineering at the freshman level and 60
students began the thermodynamics course (about 50% of
those expressing interest at the freshman level). During a
comparable period at BU, 50 students graduated while 55
students were admitted at the freshman level.
There are a number of other potential differences between
the students and their backgrounds. To address these, the
questionnaire in Table 1 was developed. The questions ad-
dress issues such as commute times, part-time work hours,
and course loads. The results from Table 1 suggest that UA
students live away from their families and have shorter com-
mute times and lighter course loads. They work part-time to
about the same extent. According to Floyd, TI students tended

TABLE 1
Student Background Information Questionnaire

'rage time I spend commuting from home
l is __ hours per day Please do not
he return mip. I ill do that-

0.7 aig (hrs)

51% 7 > 0 75

UA
0.3

egmning of the semester, I was working II jig hrs among 9
ee hours per week m addition to my those working
ring studies it g. pnvate lessons/tutoring
tour guide. McDonald's... 1 54% c working 50

ang total course credits this 19 avg 15
:r.
nularne grade point average including all 3.4 max 4
completed to date is- /4.0 2.4 mean 3.2
1.3 nnm 2

the instructor's use of English to be a 20% '. agree or 13%
ant impediment to my learning this somewhatt agree
1 relative to professors in my other classes
mple. he either spoke too fast or used
larn I could not understand, or in some other
oke differently from my other professors
at my performance was impeded.
omewhat.4gree e Somewhat Disagree Disagree
I have been hI mg wvth family members 51% %cI true 25%
rending school this semester True*Falhe
al engineering was not my first choice as a 56% % agree or 13%
study, but it was the best I could do if I somewhat agree
to come to this school.
somewhat Agree SomewhatDisagree Disagree
ser goal is to work my way out of 70% % agree or 31%

ring in the next file years and into some
s or management posiuon
omewhai .gree Somewhat Disagree Disagree

somewhat agree

Chemical Engineering Education

~

to share similar backgrounds with the Turkish stu-
dents. The higher grade point averages among the UA
students probably reflect a difference between the two
educational systems. For example US students with
lower averages would have sought another major by
this stage in the curriculum. Even so, when considered
in conjunction with the test results presented below,
the difference in grade point averages suggests a sig-
nificant degree of either grade inflation at UA or "de-

TABLE 2
Grade Scales Applied at the Respective Universities

Bogazici University
82 AA
76 BA
70 BB

University of Akron
A 80
A- 75
B+ 70

20

10 \
S' / Testt -

10 20 30 40 50 60 70 80 90 100

Figure 2. Frequency distribution plot based on test
scores at the University of Akron.

20
18 S Test2

Te0 I---4--45--85 ---47--- 18------0--53--89-
16 Test 1 ,
14 \4 Final
S/ 1
12 -
10 \
8 N --- \
6 / /
4 *
2 -
10 20 30 40 50 60 70 80 90 100
Figure 3. Frequency distribution plot based on test
scores at Bogazici University.

TABLE 3
Summary Statistics for Tests at UA and BU
University of Akron Bogazici 1
Min Med Max Mean Std Dev Min Med Max
Test1 4 45 85 47 18 0 53 89
Test2 12 64 96 63 17 50 78 97
Final 13 48 81 47 16 20 55 85

Spring 1996

flation" at BU. Until recently, the math and science courses at BU
have been the exclusive domain of engineering students. It seems
possible that the teaching of these courses may have been designed
to obtain a normal distribution in the grade scales that would be very
different if the courses were offered to more diverse groups of
students, as they are at UA.
It was surprising to learn that one of the BU students was working
42 hours per week (not counting the commute) and taking 21 course
credits. The most similar UA student was working 30 hours per
week while taking 13 course credits. BU faculty were generally
surprised at the extent to which the students were working part-time.
Another surprise was the number of BU students for whom chemical
engineering was not a first choice-56% at BU vs. 13% at UA. The
observation that 70% of BU students hope to work their way out of
engineering in the next five years (31% for UA students) is perhaps
related. Altogether, these results lend some insight into the manner
by which a homogeneous selection of students still leads to a broad
distribution in performance. The systematic placement of students
into curricula that do not represent their first choice would seem to
indicate a broad distribution in levels of motivation, especially con-
sidering that prospective employers rarely ask about the cumulative
grade point averages of BU graduates.

GRADE SCALES AND PERCEPTIONS
One difference in the manner of conducting the two courses was
the grade scale. Although the difference could influence the conclu-
sions of this study, some differences were unavoidable and others
were judged to be the best compromise between maintaining compa-
rability between the two courses vs. adaptation to local influences.
The most unavoidable difference was that the grade scale at BU was
composed of fewer gradations. As shown in Table 2, there was only
one intermediate grade between each whole grade level, instead of
the A, A-, B+, . that exist at UA. Comparing the numerical values,
it should be apparent that the whole letter grades were roughly
matched at the A level (82 vs. 80), but there was a more significant
deviation at the lower end of the scale. A 58 was a C at BU vs. a 50
for a C at UA. This difference might be expected to bias the BU
grades to higher averages. On the other hand, the BU students had
already taken physical chemistry and were juniors instead of sopho-
mores. The minimum standards for such students should naturally
be slightly higher. There was one other justification for this slight
upward shift that had to do with what I refer to as a local influence.
A sample of students revealed that 74% of students from BU gradu-
ated with two or more Ds on their transcript, vs. 19% at UA.
Evidently, students at BU perceived a D less negatively than UA
students. This last observation provides sup-
port for the suggestion that BU students be-
come complacent about their grades once they
University have been admitted.
Mean Std Dev
54 18 RESULTS OF EXAMINATIONS
77 11 Figures 2 and 3 present the distributions of test
55 15 scores at BU and UA. Table 3 presents summary
statistics. The means at BU were significantly
153

higher than the means at UA at the 95% confidence level on
all examinations. This seems to support the expectation that
the more highly selected students should perform better. On
the other hand, some influences apparently act to broaden
the distribution of the highly selected group relative to the
narrowness of the initial selection.
There are some differences between the groups of students
that go beyond the summary statistics. The
most striking difference is that the BU
students went from a high-low bimodal
distribution on the first test to a high
unimodal distribution on the second test.
As a result, the BU mean went from course r
being seven points higher than the UA standard
mean to fourteen points higher. Means
at both schools were higher on the sec- chemical
ond test, indicating that it was a rela- curricula
lively easy test. Both sets
One interpretation of these observations
would be that BU students needed to adapt
to the new professor. On the other hand, I text and s
provided both groups with sample tests access t
from the previous five years on both tests.
Informal interviews with BU students who compi
had made the switch indicated that the facilities
switching students had not studied the ide
sample tests on the first test, but studied
them seriously on the second test. The examine
distribution on the second test and final com
were closer to what one would expect
based on the differences in admissions eror
policies. The UA test distribution was identical
broader, reflecting a broader selection of simultan
admitted students. other i
It is interesting to note that a number of
students achieved high scores at both perform
schools, indicating that US schools are back
capable of producing top students despite connect
the relatively low intensity during second-
ary education. Another observation about drawn L
the differences relates to a qualitative local sy!
observation about the way the students
answered the questions. Over the years, overseas
I have become accustomed to writing

the tests such that the more mathematical questions come
at the end, because many of the UA students tend to have
difficulty with these questions, but I noticed that the BU
students tended to solve the test questions in reverse
order (the first page was often left blank). It seems that
the difference in the means on the second test is largely
attributable to the difference in performance on the last
question (worth 20 points on each test). This would con-
cur with the other anecdotal observations about the stron-

. th
dyn
epr
cou
eng
wo
of
ama
ylla
o th
utat
a, a
nti<
ati(
pai
11a
exa
eou
dic
man
rou
tion
etwi
ten
cou

ger mathematical backgrounds at foreign schools. Copies
of the test questions can be obtained by contacting the
author.

CONCLUSIONS AND RECOMMENDATIONS
The transition from making observations to making rec-
ommendations regarding such evolved educational systems
is very delicate. The observations are never
complete, and the recommendations often
is require personal judgements that may con-
flict with those of others. Noting these limi-
iamics] stations, I have attempted to present the
events a data completely in the preceding sections,
rse in the such that alternative interpretations are
not precluded. On the other hand, engi-
neering neering estimates occasionally require
rldwide. making the best recommendation based
on the limited data at hand-this is the
spirit of the recommendations below.
e primary My most significant impression was
ibus, had that a substantial fraction of the BU stu-
e same dents were underachieving. This impres-
sion derives primarily from talking with
ional students who were having difficulty and
d faced learning that their problems related sim-
cal ply to a lack of study. The reasons why
this might happen are reflected by the
ons. By fact that most of them are not really in-
ring terested in chemical engineering and be-
cause administrative practices and em-
nce on ployment prospects are less motivating
minations than those experienced by UA students.
sly with The BU administrative system makes it
very difficult for them to fail and dis-
ators of courage transfer to other majors. Pro-
ice and spective employers show little interest in
d, a what grades the students have obtained;
their primary interest concerns from
can be which school the students have gradu-
veen the ated. Tuition is free, so there is little
penalty for taking extra time to graduate.
n and uis Entirely different attitudes prevail for the
nterpart. UA students on each of these scores.
Many of these issues are beyond the con-
trol of the faculty, but faculty can influence the curriculum
in a way that might help to balance their negative impact by
integrating the students' motivations into the overall plan.
Regarding the curriculum at BU, it seems that the nature of
the students' backgrounds and interests are not taken into
account with optimal efficiency. Calculus, physics, and chem-
istry are required elements of the national entrance exam.
Students scoring in the top 0.1% on this exam can be as-
sumed to know something about these subjects, but the
Chemical Engineering Education

curriculum begins with these subjects in direct emulation of
its US counterpart. Anecdotal reports indicate that these
courses tend not to enhance the students' attitudes towards
learning.
Shortcomings at such an early stage in the college
curriculum reinforce attitudes deriving from the "relax-
ation" effect occurring after the intense preparation to
pass the entrance exam. My recommendation is that the
BU curriculum be revised to reduce the required credit
hours in freshman chemistry, physics, and mathematics.
Students should begin these subjects at the sophomore
level, similar to the Japanese students. Such a step would
bring the total number of required courses much closer to
the 15 credit/semester level.
At the same time, a strict limit should be applied on the
number of credits that a student is allowed to take if his
cumulative grade point average drops below 2.25. A com-
puter program should be implemented to enforce this
limit since human nature is not always reliably strict.
Such a simultaneous give-and-take should have the ef-
fect of emphasizing quality over quantity in a way that
would be purely beneficial.
As for the organization of the curriculum, the large
number of BU students who expect to be out of engineer-
ing in five years would seem to indicate a need for some
innovation. It is easy to proclaim that engineering in-
struction is the only proper domain of engineering educa-
tors, but my experience has been that motivated students
are more effective learners. Furthermore, the interests of
the students may reflect a practical perception of the
opportunities available in the local job market. Such prac-
tical considerations should be of interest to the engineer-
ing educator.
Is it possible that the same engineering content could
be covered while recognizing the motivations of the stu-
dents? I believe it is possible, and I would like to outline
one example of how it might be achieved. Most of the
investment economics, costing, optimization, and safety
aspects of the traditional senior design course require
little knowledge that is limited to senior status. There is
no reason why the bulk of this coursework could not be
moved to the sophomore year at the latest. Since the
students are primarily motivated by the business aspects
of engineering, such a move would bolster their interest
levels at a time when they might still be positively influ-
enced. Given such a background at an early stage, incor-
poration of business-oriented projects into the remaining
curriculum would be greatly facilitated. This is one ex-
ample of how adaptation to the local educational envi-
ronment can be achieved with little practical penalty, and
I expect that many more could be conceived.
As for the UA students, it is encouraging that the best

UA students were not significantly disadvantaged rela-
tive to the BU students. This means that their pre-college
preparation was not entirely disabling. On the other hand,
there were a number of students in the UA thermody-
namics class who were not competitive at the interna-
tional level. This observation reaffirms the need for main-
taining significant minimal performance standards in the
UA curriculum and, perhaps, indicates a need to raise
them slightly. Furthermore, the emphasis on pre-college
preparation often cited by engineering and science edu-
cators, especially with regard to math skills, should be
reiterated. The self-determination of the UA system of-
fers the advantage of more motivated engineering stu-
dents, but the reduction in the level of technical prepara-
tion at the pre-college level should not be ignored.
More generally, US faculty should recognize some dif-
ferences in chemical engineering education as practiced
in other parts of the world. One benefit of such a global
perspective is in helping to understand the educational
developments of many of the students who come to the
US for graduate school. About 10% of BU graduates
pursue graduate school abroad. By understanding the en-
vironments in which students were brought up and how
they differ from the US environment, it should be pos-
sible to develop favorable interactions more quickly and
easily.
At the undergraduate level, the greater emphases on
industrial chemistry and undergraduate specialization/re-
search are common to Japan and Turkey at least. We
should ask ourselves whether there might be some valid-
ity in emphasizing these topics to a greater extent in US
curricula. ABET's emphasis on enhancing the "design
content" of chemical engineering curricula is somewhat
similar to the manner in which schools overseas are al-
ready practicing. We should also question the signifi-
cance and implications of grade inflation in the US.
Finally, US professors must exhort themselves to pro-
duce globally competitive graduates in spite of all ob-
stacles. The evidence shows that fairly average but highly
motivated undergraduates are capable of nearly catching
up with others who are highly selected. Apparently, there
are advantages to a fairly broad admissions policy that
should not be discounted.

REFERENCES
1. Floyd, S., Chem. Eng. Ed., 22, 144 (1988)
2. Floyd, S., Chem. Eng. Ed., 22, 218 (1988)
3. Floyd, S., ChemTech, 19, 422 (1989)
4. The gross domestic product per capital of Turkey is roughly
$3400, whereas the comparable amount in the US is roughly
$22,000. These are 1991 figures taken from Grolier's Ency-
clopedia on CD ROM (1994)
5. Elliott, J.R., Chem. Eng. Ed., 27, 44 (1993) C

Spring 1996

Advising

UNDERGRADUATE

ACADEMIC ADVISING

MICHAEL L. MAVROVOUNIOTIS
Northwestern University Evanston, IL 60208-3120

he balance between research and teaching has been
the subject of considerable analysis in the past few
years, but less attention has been paid to another
important component of academic activity: academic advis-
ing. It is easy to overlook the significance of advising and
classify it as a support activity with only a minor role in the
educational effort compared to the central role of classroom
teaching. But if a poorly advised student is in the wrong
major or the wrong class, given his or her talents and
desires, then even the best classroom teachers are sowing
their seeds on poor soil.
Advising also has an impact under less dramatic circum-
stances: If students do not understand the role of a class for
their needs and goals, they would be taking the right class for
the wrong reasons and would likely lack the motivation to do
well. In short, advising can be an invisible hero allowing
classroom teaching to bear the most fruit, or a villain acting
as an obstruction and inhibitor to classroom learning.
There is a variety of competing theories and strategies for
middle-school or high-school advising,'l but literature on
the advising role of faculty in undergraduate education (let
alone engineering education) is rather limited, and some of it
addresses only specific subareas, such as women's issues,121
undecided students,'3' morality, prejudice, and mental health,141
and it tends to target counseling or student-affairs offices
rather than university faculty advisors. There is, neverthe-
less, some useful (and essentially unanimous) guidance in
the literature. Winston and his coworkers have presented a
comprehensive treatment of undergraduate academic advis-
ing[51 as well as a succinct practical guide.'61 Other sources
include an early report published by the National Education
Association[71 and more recent handbooks by Gordon181 and
by Kuh.19" This brief article presents some of the observa-
tions and recommendations made in these sources along
with the author's own views and experiences.

Michael Mavrovouniotis is Associate Professor
of Chemical Engineering at Northwestern Univer-
sity. He received his Diploma of Engineering from
the National Technical University of Athens
(Greece) in 1984 and his PhD from the Massachu-
setts Institute of Technology in 1989. Prior to join-
ing Northwestern in 1993, he served on the faculty
of the Institute for Systems Research and the chemi-
cal engineering department at the University of
Maryland, College Park.

IMPORTANCE OF ACADEMIC ADVISING
Academic advising is an integral part of the educational
process, not just a support service. Effective advising pro-
grams offer students an opportunity to realize their full po-
tential.'5 The educational process and the goals it serves are
complex and confusing to most undergraduate students. The
tremendous opportunities offered by a university education,
with all their ramifications for the rest of the student's life
and career, can easily be missed or under-used. As the stu-
dent is coping with the complex university environment it is
easy for him to focus on minor details and daily duties rather
than on his personal and educational development. The
advisor's office, with its many systematic student contacts,
is a powerful mechanism for implementing intentional and
deliberate student development.161
Academic advising has an impact on students' retention,
academic success, and the career-choice process.'61 Its influ-
ence and significance are much greater than its duration
would suggest because it occurs at critical points when the
advisee is faced with crucial decisions-decisions which
can be facilitated by the professional experience and per-
sonal maturity of the advisor.
Unfortunately, many present-day advising programs oper-
ate as bureaucratic, clerical activities on the periphery of
effective educational services.161 With the growth in univer-

Copyright ChE Division ofASEE 1996

Chemical Engineering Education

sity enrollments in recent years, academic advising seems to
have become a victim of the rush to admit greater numbers
of students. As a result, the nature of students' educational
experience has been profoundly affected.151 Surveys of stu-
dents and college administrators alike reveal general dissat-
isfaction with both the quality and the effec-
tiveness of academic advising on most col-
lege campuses.51 Itis easy

ADVISING ROLES AND MODES the sign
advisil
There can be considerable variation in ad- dii
vising modes. The prescriptive advisor is viewed a
one who focuses on student limitations.'81 activity
Students are assumed to be naturally imma- minor r
ture and irresponsible, and thus the educa- educati
tional pace, mode, and direction are con- compact
trolled by the advisor. Developmental advi- central
sors, on the other hand, concentrate on stu- classroom
dents' potentialities and view students as
striving, responsible, and capable of self- can be
direction. The advisor's role in the prescrip- cn
tive approach is one of authority and judge. hero a
The developmental advisor's role is one of classroom
helper and interactive teacher.181 to bear
There are limits to an advisor's responsi- fruit, 01
ability to the students, and students must learn actin
what these limits are. How intrusive should obstruct
an advisor be? At one extreme, one might inhij
declare that students are adults who can read classroom
the catalog and therefore should make their
own academic decisions; this reduces the
advisor to a rubber-stamping role. The other extreme is to be
in constant contact with students about every detail and
technicality of every decision they make. There is an appro-
priate middle path in developmental advising. Advisors may
try to motivate the students through encouragement and
support, but the responsibility for taking action is the
student's.81 Kramer and Gardner171 cite the advisee's "right
to fail." Ultimately, the advisee makes decisions and as-
sumes responsibility; the advisor should alert the student to
potential consequences, but should accept the student's deci-
sions even when they appear unwise and may lead to failure.
Kramer and Gardner offer an expanded classification of
advisor roles, emphasizing that the roles shift depending on
the particulars of the advising question and on symmetric
roles assumed by the student. The advisor's role might be
one of
An adult (signifying age, experience, and maturity)
An expert (possessing mastery of subject matter area
through training and achievement)
A teacher (charged with transmission of skills and
information)
A researcher (investigator, explorer)
Spring 1996

to
ific
ng,
s a
witi
,ole
ona
red
l rC
m tt
] ad
n in
lloi
m te
the
a v
g a
.tioi
ditoj
ile

A friend (offering emotional and personal attachment;
confidant)
A judge (carrying out evaluation, assessment,
arbitration, criticism)
An authority (possessing prestige and
power, giving orders and directions)
A rubber stamp (confirming and
overlook agreeing with any position presented)
dance of A lecturer (offering systematic and
often formal instruction to a body of knowl-
support edge)
only a The advisor's response to a simple and com-
in the mon encounter can come from any of these
l effort roles. For example, when the student comes in
to the with a drop slip and the request, "I want to
le of drop a course," the advisor may check the
aching rules on what courses may be dropped and
Fvising when, or simply sign the slip and let the regis-
visible trar figure out the legalities."7' Or, starting from
wing the roles of teacher and friend, the advisor can
wing set the administrative aspects of the request
.aching aside and try to find out what has prompted
most the student to make the request and whether a
/llain change in the student's long-term plans is un-
s an der way (or may be needed). Misconception
n and of roles between student and advisor is very
r to common. To avoid miscommunication, it is
warning. important to clarify what role the student ex-
pects you to assume[71 and either assume that
role or adjust the student's expectation.

CHANGING THE SYSTEM:
SEPARATION OF TASKS

In an epilogue to their excellent book, Winston and co-
workers offer a model for effective, efficient, and attentive
academic advising."0' One of their recommendations is the
separation of the clerical, record-keeping side of advising
from interpersonal and intellectual guidance.
Advising programs that emphasize registration and record
keeping, while neglecting attention to the student's educa-
tional and personal experiences in the institution, are miss-
ing an excellent opportunity to directly and immediately
influence the quality of a student's education. Such pro-
grams are also highly inefficient' 1 since they are most likely
employing highly educated (expensive) personnel who are
performing what are essentially clerical tasks. It is easier to
set the right priorities for the faculty's involvement in advis-
ing if the process of academic advising is separated from
class scheduling and registration. In many instances, class
scheduling masquerades as academic advising, which ex-
plains why many in higher education view academic advis-
ing as simply an administrative chore.

In Winston's model, class scheduling and registration are
to be accomplished by student paraprofessionals who are
closer to the process than most advisors can or desire to be.
Student paraprofessionals and support staff are more likely
to master the mechanics and nuances of the registration
process than are advisors who have other priorities to attend
to, such as class preparation, research, and strategic adminis-
trative tasks. This neatly separates the developmental side
of advising from the naturally prescriptive administra-
tive side and allows the faculty advisor to more fully play
the role of teacher rather than authority.'8' An alternative
to this systemic change is for a faculty advisor to cluster
the clerical side of advising to a single meeting with all
of his or her advisees.I" I
The obvious risk in the separation of tasks may be less
frequent contact between students and advisors. Students
may consider contact with faculty unnecessary if the bureau-
cratic side of advising is handled elsewhere. Even faculty
might come to regard exclusively developmental advising
sessions as low-priority items, easily squeezed out of the
faculty's schedules and attention. Thus, an important pre-
requisite for beneficial separation of tasks is that both
faculty and students must come to regard regular advis-
ing contact as essential.

WHAT THE INDIVIDUAL ADVISOR CAN DO
The most important action on the side of the individual
advisor is to adopt the developmental approach. Unfortu-
nately, without the systemic change cited above, there is a
certain incongruity between the developmental role and uni-
versity regulations, since the advisor usually has the author-
ity to approve or deny student requests and is responsible for
ensuring that degree requirements are met. But great strides
can be made by encouraging students themselves to master
some of the nuances in the regulations, by explaining broader
options and consequences, and by respecting the student's
decisions. Naturally, lower-division students are less likely
to be independent and responsible decision makers, and
the advisor may need to be more proactive and leave
somewhat less freedom to the students."'I This is under-
standable, and the advisor should provide the necessary
guidance while gradually pushing the students toward
more control and responsibility.
There is often incongruity between the student's and the
advisor's perceptions of the process,181 so it is important to
instill the developmental approach in the student. For ex-
ample, students are often extremely reluctant to "take up
your valuable time" if they perceive advising only in its
prescriptive and clerical side. Such a consideration is not
to be downgraded, but if you are interested in offering
your assistance in developmental advising, it is impor-
tant to convey to advises your willingness to engage in
conversation and consultation. 71
158

LEARN THE CULTURE OF YOUR STUDENTS
The influence of the culture of institutions and student
groups on the educational experience is substantial.191 Each
educational institution has its own culture, but there is often
little intersection between the faculty subculture and the
(undergraduate) student subculture. An advisor's entire in-
teraction with students improves as the advisor gains famil-
iarity with the culture the students are immersed in. It is,
admittedly, not easy to learn the student side of the institu-
tional culture, but even a general awareness of student life,
customs, and habits is helpful;191 simple steps, such as read-
ing the student newspapers (few faculty do!), are a good
start. The process has an obvious autocatalytic character
since, as you learn more, your interactions with students
become easier and give you a clearer view of their culture.
Depending on how (in)homogeneous your institution is
and how unusual your advises, advising effectiveness might
even require familiarity with specific student subcultures,
groups, or activities. It is not uncommon for advises' aca-
demic problems to be related to the study habits of living
groups.J' If one of your advises is on the swimming team,
make it a point to follow the results of the swimming
competitions and find out a little about the students travel
and training schedule (you will discover that swimmers
train every day for about four hours, or until they are
completely exhausted).
On the subject of culture, Kuh191 offers several pieces of
advice directed to student-affairs professionals, but clearly
relevant for faculty: know thyself-your values and assump-
tions; discover the various student cultures on campus; use
cultural perspectives for diagnosis and analysis; recog-
nize the importance of living areas and affinity groups to
student culture; be wary of attempts to systematically
change student culture, but recognize that student cul-
tures can be changed; use cultural perspectives when
working with marginal groups.

STRUCTURE OF THE ADVISING SESSION
In routine advising sessions, whether associated with reg-
istration or not, my own strategy is to simply review progress-
to-date and the future plans of the student, trying to learn a
bit more about the student's talents and goals with every
session. The following simple structure is adequate-pro-
vided that the advisor follows up on student answers with
genuine interest, useful observations, and information that
helps the student place the modest academic issues in the
greater context of her education and career.
0- First, take care of the administrative details rather
briskly, making sure they don't consume the entire
meeting. Some of the administrative matters can be
used an entry points to probe the student's aca-
demic and career interests during the rest of the
meeting.
Chemical Engineering Education

The second step is to ask some questions on the
advisee's recent work: Which classes did the
student like or dislike last quarter and why? Were
there any unexpected academic difficulties or
achievements in recent classes?
1 After assessing the past and present, ask for the
student's thoughts on educational plans such as
specialization areas and career paths. Offer
guidance on feasibility and good means for
achieving the student's goals-such as course
choices or internships. Mention other opportunities
the student may have overlooked or may be
misinformed about, such as graduate or profes-
sional education.
Encourage the student to get second opinions on
major issues. Suggest other faculty, university
offices, or other information sources.
In those cases where the student comes for a special con-
sultation, the following generic structure, adapted from an
outline given by Gordon,181 can be used for the advising
session.

1. Opening the Interview
Show openness, interest, and concentrated attention.
If possible, obtain the student's folder or record so
that relevant information is available during the
interview.

2. Identifying the Problem
Ask the student to state the problem, helping the
student articulate, if needed. Gather as much
information as possible by prompting the student to
provide all relevant facts.
Is the problem presented by the student actually
covering a different real problem? Ask open-ended
probing questions.
Is the student presenting several problems? Ask the
student to isolate the primary foremost concern.
Multiple problems may have a single root cause that
should be identified, or the student may be so
troubled by one issue that he or she takes a gloomy
view that makes lots of secondary issues appear as
equal obstacles.
State your interpretation of the problem and give the
student a chance to clarify, elaborate, or correct your
interpretation.

3. Identifying Possible Solutions

Ask the student for his or her ideas on solving the
problem. Help the student generate additional
Spring 1996

alternative solutions.
Discuss the mechanics of each solution (what, how,
when, who). Discuss implications of each solution.
Will a solution create conflicts with other plans?

4. Taking Action on the Solution

Plan a specific order and time frame for action steps,
including procurement of additional information and
referrals to other university resources or offices. The
advisor should have handy referral information on a
variety of campus offices and resources.
Plan follow-up by the student and/or the advisor.

5. Summarizing the Transaction

Review what has transpired and restate action steps.
Encourage future contact; make a definite appoint-
ment for review if necessary.
Once the student departs, summarize the transaction
for your own notes or for the student's file.

While the above structure is appropriate for most advising
sessions, the duration of the session can vary a great deal. In
my own experience, ten minutes is the bare minimum for
any session, no matter how trivial the problem. A simple
answer to a student dropping in with a simple question only
takes a minute or two, but there are two reasons for taking
longer. First, in the interest of cultivating the relationship
and ensuring that the student will not hesitate to seek future
contact, the advisor should take the time to make the student
feel at ease, inquiring about other advising issues or the
student's interests; it takes ten minutes to demonstrate your
interest in the advisee. Second, it is important for the advisor
to explore the origin of the question and whether it is related
to other obstacles the student is faced with but has not
brought up. Naturally, advising sessions dealing with com-
plex issues, such as specialization and career paths, will take
much longer than this minimum and may have to be broken
down into a series of meetings; in such cases, at the end of
each meeting the student should always be left with a set of
questions to ponder for the next meeting.
The commonplace use of electronic mail by faculty and
students provides an alternative efficient way to answer simple
questions without appearing hurried or damaging the advi-
sor-student relationship. If more discussion is nevertheless
warranted, the advisor can respond with an offer for an
appointment, which is easily arranged by e-mail.
In the opposite extreme, a note of caution is in order;
serious problems of a nonacademic nature (even if brought
about by academic events) occasionally arise (i.e., severe
depression, thoughts of suicide, substance abuse, acute inter-

personal problems among students). The engineering faculty
advisor is unlikely to possess the requisite special training
and skills to deal with such issues effectively, and a well-
intentioned attempt to help might even worsen the problem.
It is best to refer such problems to trained counselors avail-
able at any university. To make sure that the student receives
help, the advisor may insist that an appointment be made
right then and there. Afterward, the advisor should follow up
with the counseling office to make sure that the student kept
the appointment, as well as check with the student periodi-
cally so the student knows someone is concerned. In order to
deal with such crises, the advisor should be aware of all the
relevant campus resources, along with contact names and
phone numbers-before any crisis occurs.

CONCLUSION
The central premise of this article is that the advising
process is an integral part of the educational process. Unfor-
tunately, it is too often misinterpreted as a purely clerical
task and receives only limited attention by the faculty, stu-
dents, and administration. A valuable systemic change would
be the separation of the clerical and developmental sides of
advising; the former can be handled by staff, allowing the
faculty's full attention to be devoted to the intellectual growth
of the students.
Faculty advisors should strive to improve the strategies
they follow in encouraging student contact, acting in a teach-
ing and supportive role, allowing the students ultimate deci-
sion-making and responsibility, and helping students to fo-
cus on the greater educational and professional decisions
and objectives and the means for accomplishing them.

ACKNOWLEDGMENTS
The author wishes to thank Professors Richard Felder,
John O'Connell, Christopher Bowman, William Miller, and
Julio Ottino for their comments, which guided a substantial
revision of this paper.

REFERENCES
1. Myrick, R.D., L.S. Myrick, and contributors, The Teacher
Advisor Program, ERIC Counseling and Personnel Services
Clearinghouse, Ann Arbor, MI (1990)
2. Gelwick, B., ed., Up the Ladder: Women, Professionals, and
Clients in College Student Personnel, American College Per-
sonnel Association, Cincinnati, OH (1979)
3. Gordon, V., The Undecided College Student: An Academic
and Career Advising Challenge, Charles C. Thomas, Spring-
field, IL (1984)
4. Farnsworth, D.L., and G.B.Blaine, Jr., eds., Counseling and
the College Student, Little, Brown, and Co., Boston, MA
(1970)
5. Winston, R.B., Jr., T.K. Miller, S.C. Ender, T.J. Grites, and
associates, Developmental Academic Advising: Addressing
Students' Educational, Career, and Personal Needs, Jossey-
Bass Publishers, San Francisco, CA (1984)
6. Winston, R.B., Jr., S.C. Ender, and T.K. Miller, eds., Devel-
opmental Approaches to Academic Advising, Jossey-Bass

Publishers, San Francisco (1982)
7. Kramer, H.C., and R.E. Gardner, Advising by Faculty, Na-
tional Education Association, Washington, DC (1977)
8. Gordon, V.N., Handbook of Academic Advising, Greenwood
Press, Westport, CT (1992)
9. Kuh, G.D., ed., Cultural Perspectives in Student Affairs
Work, American College Personnel Association, Cincinnati,
OH (1993)
10. Winston, R.B., Jr., T.J. Grites, T.K. Miller, and S.C. Ender,
"Epilogue: Improving Academic Advising," in Developmen-
tal Approaches to Academic Advising, R.B. Winston, Jr.,
S.C. Ender, and T.K. Miller, eds., Jossey-Bass Publishers,
San Francisco, CA (1982)
11. Wankat, P.C., and F.S. Oreovicz, Teaching Engineering,
McGraw-Hill, New York, NY, pp. 201-205 (1993) J

LABORATORY EXPERIMENT
Continued from page 101.
lem, particularly for the 1990s, it is based on well-tested
combustion phenomenon and not limited to a special limited
situation, and the data from the unit are not complete, forc-
ing the soon-to-be practicing engineers to solve problems
and perform an analysis based on their best judgment.
The experiment is best performed over an entire day, so
trying to carry it out in a half-day session is not recom-
mended. We strongly advise that the entire experiment be
located in a fume hood so that the flue gas is swept out of the
unit and no dangerous or noxious odors are emitted into the
laboratory. Finally, the potential to overheat the fluid-bed
unit from feeding too much fuel gas means that the stu-
dents need to be monitored periodically to be sure they
are operating the unit in a controlled and safe manner.
The use of a high-temperature limit switch will eliminate
this potential problem.

ACKNOWLEDGMENTS
KKR wishes to express his extreme appreciation to James
Jaeger in Northwestern's machine shop for his help in ma-
chining and fabricating the fluid-bed unit. He also appreci-
ates the technical information provided by George Rasmussen
at Ecova for the incinerator design. Finally, the Northwest-
ern Chemical Engineering Department wishes to express its
thanks to Procter and Gamble Company for generously pro-
viding the funds for the development of this "hands-on"
engineering experiment.

REFERENCES
1. Dahlstrom, D.A., "Chemical Engineering: Notes on Its Past
and Its Future," Chem. Eng. Ed., 28(4), 226 (1994)
2. Hougan, Olaf A., "Seven Decades of Chemical Engineering,"
Bicentennial Lecture of Chemical Engineering History,
AIChE 82nd National Meeting, Atlantic City, NJ
3. Bird, R. Byron, "Hougen's Principles," Chem. Eng. Ed., 20(4),
161 (1986)
4. Mullen, J.F., "Consider Fluid-Bed Incineration for Hazard-
ous Waste Destruction," Chem. Eng. Prog., 88(6), 50 (1992)

Chemical Engineering Education

AUTHOR GUIDELINES

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.

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
consecutively.

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

F

CALFO APR

	Front Cover
	Table of Contents
	University of Washington
	Morton M. Denn
	The mass transfer boundary layer...
	A laboratory experiment that enhances...
	Design competition for second-year...
	Demonstrations to complement a...
	CESL: The chemical engineering...
	Applied statistics: Are ChE educators...
	International engineering internship...
	Speaking of everything
	Dynamic and steady-state behavior...
	Mathematica in the ChE curricu...
	Low-cost experiments in mass transfer:...
	Errata
	Current trends in chemical reaction...
	Chemical engineering education...
	Undergraduate academic advisin...
	Back Cover

MILLISECOND	CLASS.METHOD	MESSAGE
0	sobekcm_page_globals.constructor
0	sobekcm_page_globals.constructor	Application State validated or built
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.constructor	Navigation Object created from URI query string
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.display_item	Retrieving item or group information
0	sobekcm_page_globals.get_entire_collection_hierarchy	Retrieving hierarchy information
0	sobekcm_assistant.get_entire_collection_hierarchy
0	cached_data_manager.retrieve_item_aggregation
0	cached_data_manager.retrieve_item_aggregation	Found item aggregation on local cache
0	item_aggregation_builder.get_item_aggregation	Found 'all' item aggregation in cache
0	system.web.ui.page.page_load (ufdc.page_load)
0	sobekcm_page_globals.constructor.on_page_load
0	html_echo_mainwriter.add_style_references	Adding style references to HTML
0	html_echo_mainwriter.add_text_to_page	Reading the text from the file and echoing back to the output stream
82	html_echo_mainwriter.add_text_to_page	Finished reading and writing the file


UFDC Home	<%MYSOBEK%> \| Help \| RSS